904 files changed, 78187 insertions, 41432 deletions

diff --git a/fs/9p/Kconfig b/fs/9p/Kconfig
index ac2ec4543fe1..09fd4a185fd2 100644
--- a/fs/9p/Kconfig
+++ b/fs/9p/Kconfig
@@ -32,13 +32,13 @@ endif
 
 
 config 9P_FS_SECURITY
-        bool "9P Security Labels"
-        depends on 9P_FS
-        help
-          Security labels support alternative access control models
-          implemented by security modules like SELinux.  This option
-          enables an extended attribute handler for file security
-          labels in the 9P filesystem.
-
-          If you are not using a security module that requires using
-          extended attributes for file security labels, say N.
+	bool "9P Security Labels"
+	depends on 9P_FS
+	help
+	  Security labels support alternative access control models
+	  implemented by security modules like SELinux.  This option
+	  enables an extended attribute handler for file security
+	  labels in the 9P filesystem.
+
+	  If you are not using a security module that requires using
+	  extended attributes for file security labels, say N.
diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c
index fe7f0bd2048e..92cd1d80218d 100644
--- a/fs/9p/vfs_file.c
+++ b/fs/9p/vfs_file.c
@@ -388,7 +388,10 @@ v9fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	p9_debug(P9_DEBUG_VFS, "count %zu offset %lld\n",
 		 iov_iter_count(to), iocb->ki_pos);
 
-	ret = p9_client_read(fid, iocb->ki_pos, to, &err);
+	if (iocb->ki_filp->f_flags & O_NONBLOCK)
+		ret = p9_client_read_once(fid, iocb->ki_pos, to, &err);
+	else
+		ret = p9_client_read(fid, iocb->ki_pos, to, &err);
 	if (!ret)
 		return err;
 
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index b82423a72f68..c9255d399917 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -143,7 +143,7 @@ static umode_t p9mode2unixmode(struct v9fs_session_info *v9ses,
 		default:
 			p9_debug(P9_DEBUG_ERROR, "Unknown special type %c %s\n",
 				 type, stat->extension);
-		};
+		}
 		*rdev = MKDEV(major, minor);
 	} else
 		res |= S_IFREG;
diff --git a/fs/Kconfig b/fs/Kconfig
index 2501e6f1f965..d1ad3935fb85 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -40,6 +40,7 @@ source "fs/ocfs2/Kconfig"
 source "fs/btrfs/Kconfig"
 source "fs/nilfs2/Kconfig"
 source "fs/f2fs/Kconfig"
+source "fs/zonefs/Kconfig"
 
 config FS_DAX
 	bool "Direct Access (DAX) support"
@@ -139,9 +140,10 @@ endmenu
 endif # BLOCK
 
 if BLOCK
-menu "DOS/FAT/NT Filesystems"
+menu "DOS/FAT/EXFAT/NT Filesystems"
 
 source "fs/fat/Kconfig"
+source "fs/exfat/Kconfig"
 source "fs/ntfs/Kconfig"
 
 endmenu
@@ -164,7 +166,7 @@ config TMPFS
 	  space. If you unmount a tmpfs instance, everything stored therein is
 	  lost.
 
-	  See <file:Documentation/filesystems/tmpfs.txt> for details.
+	  See <file:Documentation/filesystems/tmpfs.rst> for details.
 
 config TMPFS_POSIX_ACL
 	bool "Tmpfs POSIX Access Control Lists"
@@ -264,6 +266,7 @@ source "fs/pstore/Kconfig"
 source "fs/sysv/Kconfig"
 source "fs/ufs/Kconfig"
 source "fs/erofs/Kconfig"
+source "fs/vboxsf/Kconfig"
 
 endif # MISC_FILESYSTEMS
 
@@ -322,4 +325,7 @@ source "fs/nls/Kconfig"
 source "fs/dlm/Kconfig"
 source "fs/unicode/Kconfig"
 
+config IO_WQ
+	bool
+
 endmenu
diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt
index 62dc4f577ba1..04f86b8c100e 100644
--- a/fs/Kconfig.binfmt
+++ b/fs/Kconfig.binfmt
@@ -36,6 +36,12 @@ config COMPAT_BINFMT_ELF
 config ARCH_BINFMT_ELF_STATE
 	bool
 
+config ARCH_HAVE_ELF_PROT
+	bool
+
+config ARCH_USE_GNU_PROPERTY
+	bool
+
 config BINFMT_ELF_FDPIC
 	bool "Kernel support for FDPIC ELF binaries"
 	default y if !BINFMT_ELF
@@ -72,7 +78,7 @@ config CORE_DUMP_DEFAULT_ELF_HEADERS
 
 	  The core dump behavior can be controlled per process using
 	  the /proc/PID/coredump_filter pseudo-file; this setting is
-	  inherited.  See Documentation/filesystems/proc.txt for details.
+	  inherited.  See Documentation/filesystems/proc.rst for details.
 
 	  This config option changes the default setting of coredump_filter
 	  seen at boot time.  If unsure, say Y.
diff --git a/fs/Makefile b/fs/Makefile
index 14231b4cf383..2ce5112b02c8 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -32,11 +32,12 @@ obj-$(CONFIG_EVENTFD)		+= eventfd.o
 obj-$(CONFIG_USERFAULTFD)	+= userfaultfd.o
 obj-$(CONFIG_AIO)               += aio.o
 obj-$(CONFIG_IO_URING)		+= io_uring.o
+obj-$(CONFIG_IO_WQ)		+= io-wq.o
 obj-$(CONFIG_FS_DAX)		+= dax.o
 obj-$(CONFIG_FS_ENCRYPTION)	+= crypto/
 obj-$(CONFIG_FS_VERITY)		+= verity/
 obj-$(CONFIG_FILE_LOCKING)      += locks.o
-obj-$(CONFIG_COMPAT)		+= compat.o compat_ioctl.o
+obj-$(CONFIG_COMPAT)		+= compat.o
 obj-$(CONFIG_BINFMT_AOUT)	+= binfmt_aout.o
 obj-$(CONFIG_BINFMT_EM86)	+= binfmt_em86.o
 obj-$(CONFIG_BINFMT_MISC)	+= binfmt_misc.o
@@ -82,6 +83,7 @@ obj-$(CONFIG_HUGETLBFS)		+= hugetlbfs/
 obj-$(CONFIG_CODA_FS)		+= coda/
 obj-$(CONFIG_MINIX_FS)		+= minix/
 obj-$(CONFIG_FAT_FS)		+= fat/
+obj-$(CONFIG_EXFAT_FS)		+= exfat/
 obj-$(CONFIG_BFS_FS)		+= bfs/
 obj-$(CONFIG_ISO9660_FS)	+= isofs/
 obj-$(CONFIG_HFSPLUS_FS)	+= hfsplus/ # Before hfs to find wrapped HFS+
@@ -132,3 +134,5 @@ obj-$(CONFIG_CEPH_FS)		+= ceph/
 obj-$(CONFIG_PSTORE)		+= pstore/
 obj-$(CONFIG_EFIVAR_FS)		+= efivarfs/
 obj-$(CONFIG_EROFS_FS)		+= erofs/
+obj-$(CONFIG_VBOXSF_FS)		+= vboxsf/
+obj-$(CONFIG_ZONEFS_FS)		+= zonefs/
diff --git a/fs/adfs/Kconfig b/fs/adfs/Kconfig
index df4650dccf68..44738fed6625 100644
--- a/fs/adfs/Kconfig
+++ b/fs/adfs/Kconfig
@@ -12,7 +12,7 @@ config ADFS_FS
 
 	  The ADFS partition should be the first partition (i.e.,
 	  /dev/[hs]d?1) on each of your drives. Please read the file
-	  <file:Documentation/filesystems/adfs.txt> for further details.
+	  <file:Documentation/filesystems/adfs.rst> for further details.
 
 	  To compile this code as a module, choose M here: the module will be
 	  called adfs.
diff --git a/fs/adfs/adfs.h b/fs/adfs/adfs.h
index b7e844d2f321..699c4fa8b78b 100644
--- a/fs/adfs/adfs.h
+++ b/fs/adfs/adfs.h
@@ -26,14 +26,13 @@ static inline u16 adfs_filetype(u32 loadaddr)
 #define ADFS_NDA_PUBLIC_READ	(1 << 5)
 #define ADFS_NDA_PUBLIC_WRITE	(1 << 6)
 
-#include "dir_f.h"
-
 /*
  * adfs file system inode data in memory
  */
 struct adfs_inode_info {
 	loff_t		mmu_private;
 	__u32		parent_id;	/* parent indirect disc address	*/
+	__u32		indaddr;	/* object indirect disc address	*/
 	__u32		loadaddr;	/* RISC OS load address		*/
 	__u32		execaddr;	/* RISC OS exec address		*/
 	unsigned int	attr;		/* RISC OS permissions		*/
@@ -93,15 +92,19 @@ struct adfs_dir {
 
 	int			nr_buffers;
 	struct buffer_head	*bh[4];
-
-	/* big directories need allocated buffers */
-	struct buffer_head	**bh_fplus;
+	struct buffer_head	**bhs;
 
 	unsigned int		pos;
 	__u32			parent_id;
 
-	struct adfs_dirheader	dirhead;
-	union  adfs_dirtail	dirtail;
+	union {
+		struct adfs_dirheader	*dirhead;
+		struct adfs_bigdirheader *bighead;
+	};
+	union {
+		struct adfs_newdirtail	*newtail;
+		struct adfs_bigdirtail	*bigtail;
+	};
 };
 
 /*
@@ -122,13 +125,13 @@ struct object_info {
 struct adfs_dir_ops {
 	int	(*read)(struct super_block *sb, unsigned int indaddr,
 			unsigned int size, struct adfs_dir *dir);
+	int	(*iterate)(struct adfs_dir *dir, struct dir_context *ctx);
 	int	(*setpos)(struct adfs_dir *dir, unsigned int fpos);
 	int	(*getnext)(struct adfs_dir *dir, struct object_info *obj);
 	int	(*update)(struct adfs_dir *dir, struct object_info *obj);
 	int	(*create)(struct adfs_dir *dir, struct object_info *obj);
 	int	(*remove)(struct adfs_dir *dir, struct object_info *obj);
-	int	(*sync)(struct adfs_dir *dir);
-	void	(*free)(struct adfs_dir *dir);
+	int	(*commit)(struct adfs_dir *dir);
 };
 
 struct adfs_discmap {
@@ -145,7 +148,9 @@ int adfs_notify_change(struct dentry *dentry, struct iattr *attr);
 
 /* map.c */
 int adfs_map_lookup(struct super_block *sb, u32 frag_id, unsigned int offset);
-extern unsigned int adfs_map_free(struct super_block *sb);
+void adfs_map_statfs(struct super_block *sb, struct kstatfs *buf);
+struct adfs_discmap *adfs_read_map(struct super_block *sb, struct adfs_discrecord *dr);
+void adfs_free_map(struct super_block *sb);
 
 /* Misc */
 __printf(3, 4)
@@ -167,6 +172,13 @@ extern const struct dentry_operations adfs_dentry_operations;
 extern const struct adfs_dir_ops adfs_f_dir_ops;
 extern const struct adfs_dir_ops adfs_fplus_dir_ops;
 
+int adfs_dir_copyfrom(void *dst, struct adfs_dir *dir, unsigned int offset,
+		      size_t len);
+int adfs_dir_copyto(struct adfs_dir *dir, unsigned int offset, const void *src,
+		    size_t len);
+void adfs_dir_relse(struct adfs_dir *dir);
+int adfs_dir_read_buffers(struct super_block *sb, u32 indaddr,
+			  unsigned int size, struct adfs_dir *dir);
 void adfs_object_fixup(struct adfs_dir *dir, struct object_info *obj);
 extern int adfs_dir_update(struct super_block *sb, struct object_info *obj,
 			   int wait);
diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c
index a54c53244992..77fbd196008f 100644
--- a/fs/adfs/dir.c
+++ b/fs/adfs/dir.c
@@ -6,12 +6,196 @@
  *
  *  Common directory handling for ADFS
  */
+#include <linux/slab.h>
 #include "adfs.h"
 
 /*
  * For future.  This should probably be per-directory.
  */
-static DEFINE_RWLOCK(adfs_dir_lock);
+static DECLARE_RWSEM(adfs_dir_rwsem);
+
+int adfs_dir_copyfrom(void *dst, struct adfs_dir *dir, unsigned int offset,
+		      size_t len)
+{
+	struct super_block *sb = dir->sb;
+	unsigned int index, remain;
+
+	index = offset >> sb->s_blocksize_bits;
+	offset &= sb->s_blocksize - 1;
+	remain = sb->s_blocksize - offset;
+	if (index + (remain < len) >= dir->nr_buffers)
+		return -EINVAL;
+
+	if (remain < len) {
+		memcpy(dst, dir->bhs[index]->b_data + offset, remain);
+		dst += remain;
+		len -= remain;
+		index += 1;
+		offset = 0;
+	}
+
+	memcpy(dst, dir->bhs[index]->b_data + offset, len);
+
+	return 0;
+}
+
+int adfs_dir_copyto(struct adfs_dir *dir, unsigned int offset, const void *src,
+		    size_t len)
+{
+	struct super_block *sb = dir->sb;
+	unsigned int index, remain;
+
+	index = offset >> sb->s_blocksize_bits;
+	offset &= sb->s_blocksize - 1;
+	remain = sb->s_blocksize - offset;
+	if (index + (remain < len) >= dir->nr_buffers)
+		return -EINVAL;
+
+	if (remain < len) {
+		memcpy(dir->bhs[index]->b_data + offset, src, remain);
+		src += remain;
+		len -= remain;
+		index += 1;
+		offset = 0;
+	}
+
+	memcpy(dir->bhs[index]->b_data + offset, src, len);
+
+	return 0;
+}
+
+static void __adfs_dir_cleanup(struct adfs_dir *dir)
+{
+	dir->nr_buffers = 0;
+
+	if (dir->bhs != dir->bh)
+		kfree(dir->bhs);
+	dir->bhs = NULL;
+	dir->sb = NULL;
+}
+
+void adfs_dir_relse(struct adfs_dir *dir)
+{
+	unsigned int i;
+
+	for (i = 0; i < dir->nr_buffers; i++)
+		brelse(dir->bhs[i]);
+
+	__adfs_dir_cleanup(dir);
+}
+
+static void adfs_dir_forget(struct adfs_dir *dir)
+{
+	unsigned int i;
+
+	for (i = 0; i < dir->nr_buffers; i++)
+		bforget(dir->bhs[i]);
+
+	__adfs_dir_cleanup(dir);
+}
+
+int adfs_dir_read_buffers(struct super_block *sb, u32 indaddr,
+			  unsigned int size, struct adfs_dir *dir)
+{
+	struct buffer_head **bhs;
+	unsigned int i, num;
+	int block;
+
+	num = ALIGN(size, sb->s_blocksize) >> sb->s_blocksize_bits;
+	if (num > ARRAY_SIZE(dir->bh)) {
+		/* We only allow one extension */
+		if (dir->bhs != dir->bh)
+			return -EINVAL;
+
+		bhs = kcalloc(num, sizeof(*bhs), GFP_KERNEL);
+		if (!bhs)
+			return -ENOMEM;
+
+		if (dir->nr_buffers)
+			memcpy(bhs, dir->bhs, dir->nr_buffers * sizeof(*bhs));
+
+		dir->bhs = bhs;
+	}
+
+	for (i = dir->nr_buffers; i < num; i++) {
+		block = __adfs_block_map(sb, indaddr, i);
+		if (!block) {
+			adfs_error(sb, "dir %06x has a hole at offset %u",
+				   indaddr, i);
+			goto error;
+		}
+
+		dir->bhs[i] = sb_bread(sb, block);
+		if (!dir->bhs[i]) {
+			adfs_error(sb,
+				   "dir %06x failed read at offset %u, mapped block 0x%08x",
+				   indaddr, i, block);
+			goto error;
+		}
+
+		dir->nr_buffers++;
+	}
+	return 0;
+
+error:
+	adfs_dir_relse(dir);
+
+	return -EIO;
+}
+
+static int adfs_dir_read(struct super_block *sb, u32 indaddr,
+			 unsigned int size, struct adfs_dir *dir)
+{
+	dir->sb = sb;
+	dir->bhs = dir->bh;
+	dir->nr_buffers = 0;
+
+	return ADFS_SB(sb)->s_dir->read(sb, indaddr, size, dir);
+}
+
+static int adfs_dir_read_inode(struct super_block *sb, struct inode *inode,
+			       struct adfs_dir *dir)
+{
+	int ret;
+
+	ret = adfs_dir_read(sb, ADFS_I(inode)->indaddr, inode->i_size, dir);
+	if (ret)
+		return ret;
+
+	if (ADFS_I(inode)->parent_id != dir->parent_id) {
+		adfs_error(sb,
+			   "parent directory id changed under me! (%06x but got %06x)\n",
+			   ADFS_I(inode)->parent_id, dir->parent_id);
+		adfs_dir_relse(dir);
+		ret = -EIO;
+	}
+
+	return ret;
+}
+
+static void adfs_dir_mark_dirty(struct adfs_dir *dir)
+{
+	unsigned int i;
+
+	/* Mark the buffers dirty */
+	for (i = 0; i < dir->nr_buffers; i++)
+		mark_buffer_dirty(dir->bhs[i]);
+}
+
+static int adfs_dir_sync(struct adfs_dir *dir)
+{
+	int err = 0;
+	int i;
+
+	for (i = dir->nr_buffers - 1; i >= 0; i--) {
+		struct buffer_head *bh = dir->bhs[i];
+		sync_dirty_buffer(bh);
+		if (buffer_req(bh) && !buffer_uptodate(bh))
+			err = -EIO;
+	}
+
+	return err;
+}
 
 void adfs_object_fixup(struct adfs_dir *dir, struct object_info *obj)
 {
@@ -51,87 +235,90 @@ void adfs_object_fixup(struct adfs_dir *dir, struct object_info *obj)
 	}
 }
 
-static int
-adfs_readdir(struct file *file, struct dir_context *ctx)
+static int adfs_iterate(struct file *file, struct dir_context *ctx)
 {
 	struct inode *inode = file_inode(file);
 	struct super_block *sb = inode->i_sb;
 	const struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir;
-	struct object_info obj;
 	struct adfs_dir dir;
-	int ret = 0;
-
-	if (ctx->pos >> 32)
-		return 0;
+	int ret;
 
-	ret = ops->read(sb, inode->i_ino, inode->i_size, &dir);
+	down_read(&adfs_dir_rwsem);
+	ret = adfs_dir_read_inode(sb, inode, &dir);
 	if (ret)
-		return ret;
+		goto unlock;
 
 	if (ctx->pos == 0) {
 		if (!dir_emit_dot(file, ctx))
-			goto free_out;
+			goto unlock_relse;
 		ctx->pos = 1;
 	}
 	if (ctx->pos == 1) {
 		if (!dir_emit(ctx, "..", 2, dir.parent_id, DT_DIR))
-			goto free_out;
+			goto unlock_relse;
 		ctx->pos = 2;
 	}
 
-	read_lock(&adfs_dir_lock);
+	ret = ops->iterate(&dir, ctx);
 
-	ret = ops->setpos(&dir, ctx->pos - 2);
-	if (ret)
-		goto unlock_out;
-	while (ops->getnext(&dir, &obj) == 0) {
-		if (!dir_emit(ctx, obj.name, obj.name_len,
-			      obj.indaddr, DT_UNKNOWN))
-			break;
-		ctx->pos++;
-	}
-
-unlock_out:
-	read_unlock(&adfs_dir_lock);
+unlock_relse:
+	up_read(&adfs_dir_rwsem);
+	adfs_dir_relse(&dir);
+	return ret;
 
-free_out:
-	ops->free(&dir);
+unlock:
+	up_read(&adfs_dir_rwsem);
 	return ret;
 }
 
 int
 adfs_dir_update(struct super_block *sb, struct object_info *obj, int wait)
 {
-	int ret = -EINVAL;
-#ifdef CONFIG_ADFS_FS_RW
 	const struct adfs_dir_ops *ops = ADFS_SB(sb)->s_dir;
 	struct adfs_dir dir;
+	int ret;
 
-	printk(KERN_INFO "adfs_dir_update: object %06x in dir %06x\n",
-		 obj->indaddr, obj->parent_id);
+	if (!IS_ENABLED(CONFIG_ADFS_FS_RW))
+		return -EINVAL;
 
-	if (!ops->update) {
-		ret = -EINVAL;
-		goto out;
-	}
+	if (!ops->update)
+		return -EINVAL;
 
-	ret = ops->read(sb, obj->parent_id, 0, &dir);
+	down_write(&adfs_dir_rwsem);
+	ret = adfs_dir_read(sb, obj->parent_id, 0, &dir);
 	if (ret)
-		goto out;
+		goto unlock;
 
-	write_lock(&adfs_dir_lock);
 	ret = ops->update(&dir, obj);
-	write_unlock(&adfs_dir_lock);
+	if (ret)
+		goto forget;
 
-	if (wait) {
-		int err = ops->sync(&dir);
-		if (!ret)
-			ret = err;
-	}
+	ret = ops->commit(&dir);
+	if (ret)
+		goto forget;
+	up_write(&adfs_dir_rwsem);
+
+	adfs_dir_mark_dirty(&dir);
+
+	if (wait)
+		ret = adfs_dir_sync(&dir);
+
+	adfs_dir_relse(&dir);
+	return ret;
+
+	/*
+	 * If the updated failed because the entry wasn't found, we can
+	 * just release the buffers. If it was any other error, forget
+	 * the dirtied buffers so they aren't written back to the media.
+	 */
+forget:
+	if (ret == -ENOENT)
+		adfs_dir_relse(&dir);
+	else
+		adfs_dir_forget(&dir);
+unlock:
+	up_write(&adfs_dir_rwsem);
 
-	ops->free(&dir);
-out:
-#endif
 	return ret;
 }
 
@@ -167,25 +354,14 @@ static int adfs_dir_lookup_byname(struct inode *inode, const struct qstr *qstr,
 	u32 name_len;
 	int ret;
 
-	ret = ops->read(sb, inode->i_ino, inode->i_size, &dir);
+	down_read(&adfs_dir_rwsem);
+	ret = adfs_dir_read_inode(sb, inode, &dir);
 	if (ret)
-		goto out;
-
-	if (ADFS_I(inode)->parent_id != dir.parent_id) {
-		adfs_error(sb,
-			   "parent directory changed under me! (%06x but got %06x)\n",
-			   ADFS_I(inode)->parent_id, dir.parent_id);
-		ret = -EIO;
-		goto free_out;
-	}
-
-	obj->parent_id = inode->i_ino;
-
-	read_lock(&adfs_dir_lock);
+		goto unlock;
 
 	ret = ops->setpos(&dir, 0);
 	if (ret)
-		goto unlock_out;
+		goto unlock_relse;
 
 	ret = -ENOENT;
 	name = qstr->name;
@@ -196,20 +372,22 @@ static int adfs_dir_lookup_byname(struct inode *inode, const struct qstr *qstr,
 			break;
 		}
 	}
+	obj->parent_id = ADFS_I(inode)->indaddr;
 
-unlock_out:
-	read_unlock(&adfs_dir_lock);
+unlock_relse:
+	up_read(&adfs_dir_rwsem);
+	adfs_dir_relse(&dir);
+	return ret;
 
-free_out:
-	ops->free(&dir);
-out:
+unlock:
+	up_read(&adfs_dir_rwsem);
 	return ret;
 }
 
 const struct file_operations adfs_dir_operations = {
 	.read		= generic_read_dir,
 	.llseek		= generic_file_llseek,
-	.iterate	= adfs_readdir,
+	.iterate_shared	= adfs_iterate,
 	.fsync		= generic_file_fsync,
 };
 
diff --git a/fs/adfs/dir_f.c b/fs/adfs/dir_f.c
index c1a950c7400a..30d526fecc3f 100644
--- a/fs/adfs/dir_f.c
+++ b/fs/adfs/dir_f.c
@@ -9,8 +9,6 @@
 #include "adfs.h"
 #include "dir_f.h"
 
-static void adfs_f_free(struct adfs_dir *dir);
-
 /*
  * Read an (unaligned) value of length 1..4 bytes
  */
@@ -60,7 +58,7 @@ static inline void adfs_writeval(unsigned char *p, int len, unsigned int val)
 #define bufoff(_bh,_idx)			\
 	({ int _buf = _idx >> blocksize_bits;	\
 	   int _off = _idx - (_buf << blocksize_bits);\
-	  (u8 *)(_bh[_buf]->b_data + _off);	\
+	  (void *)(_bh[_buf]->b_data + _off);	\
 	})
 
 /*
@@ -123,65 +121,49 @@ adfs_dir_checkbyte(const struct adfs_dir *dir)
 	return (dircheck ^ (dircheck >> 8) ^ (dircheck >> 16) ^ (dircheck >> 24)) & 0xff;
 }
 
-/* Read and check that a directory is valid */
-static int adfs_dir_read(struct super_block *sb, u32 indaddr,
-			 unsigned int size, struct adfs_dir *dir)
+static int adfs_f_validate(struct adfs_dir *dir)
 {
-	const unsigned int blocksize_bits = sb->s_blocksize_bits;
-	int blk = 0;
-
-	/*
-	 * Directories which are not a multiple of 2048 bytes
-	 * are considered bad v2 [3.6]
-	 */
-	if (size & 2047)
-		goto bad_dir;
-
-	size >>= blocksize_bits;
-
-	dir->nr_buffers = 0;
-	dir->sb = sb;
-
-	for (blk = 0; blk < size; blk++) {
-		int phys;
+	struct adfs_dirheader *head = dir->dirhead;
+	struct adfs_newdirtail *tail = dir->newtail;
+
+	if (head->startmasseq != tail->endmasseq ||
+	    tail->dirlastmask || tail->reserved[0] || tail->reserved[1] ||
+	    (memcmp(&head->startname, "Nick", 4) &&
+	     memcmp(&head->startname, "Hugo", 4)) ||
+	    memcmp(&head->startname, &tail->endname, 4) ||
+	    adfs_dir_checkbyte(dir) != tail->dircheckbyte)
+		return -EIO;
 
-		phys = __adfs_block_map(sb, indaddr, blk);
-		if (!phys) {
-			adfs_error(sb, "dir %06x has a hole at offset %d",
-				   indaddr, blk);
-			goto release_buffers;
-		}
+	return 0;
+}
 
-		dir->bh[blk] = sb_bread(sb, phys);
-		if (!dir->bh[blk])
-			goto release_buffers;
-	}
+/* Read and check that a directory is valid */
+static int adfs_f_read(struct super_block *sb, u32 indaddr, unsigned int size,
+		       struct adfs_dir *dir)
+{
+	const unsigned int blocksize_bits = sb->s_blocksize_bits;
+	int ret;
 
-	memcpy(&dir->dirhead, bufoff(dir->bh, 0), sizeof(dir->dirhead));
-	memcpy(&dir->dirtail, bufoff(dir->bh, 2007), sizeof(dir->dirtail));
+	if (size && size != ADFS_NEWDIR_SIZE)
+		return -EIO;
 
-	if (dir->dirhead.startmasseq != dir->dirtail.new.endmasseq ||
-	    memcmp(&dir->dirhead.startname, &dir->dirtail.new.endname, 4))
-		goto bad_dir;
+	ret = adfs_dir_read_buffers(sb, indaddr, ADFS_NEWDIR_SIZE, dir);
+	if (ret)
+		return ret;
 
-	if (memcmp(&dir->dirhead.startname, "Nick", 4) &&
-	    memcmp(&dir->dirhead.startname, "Hugo", 4))
-		goto bad_dir;
+	dir->dirhead = bufoff(dir->bh, 0);
+	dir->newtail = bufoff(dir->bh, 2007);
 
-	if (adfs_dir_checkbyte(dir) != dir->dirtail.new.dircheckbyte)
+	if (adfs_f_validate(dir))
 		goto bad_dir;
 
-	dir->nr_buffers = blk;
+	dir->parent_id = adfs_readval(dir->newtail->dirparent, 3);
 
 	return 0;
 
 bad_dir:
 	adfs_error(sb, "dir %06x is corrupted", indaddr);
-release_buffers:
-	for (blk -= 1; blk >= 0; blk -= 1)
-		brelse(dir->bh[blk]);
-
-	dir->sb = NULL;
+	adfs_dir_relse(dir);
 
 	return -EIO;
 }
@@ -232,24 +214,12 @@ adfs_obj2dir(struct adfs_direntry *de, struct object_info *obj)
 static int
 __adfs_dir_get(struct adfs_dir *dir, int pos, struct object_info *obj)
 {
-	struct super_block *sb = dir->sb;
 	struct adfs_direntry de;
-	int thissize, buffer, offset;
-
-	buffer = pos >> sb->s_blocksize_bits;
-
-	if (buffer > dir->nr_buffers)
-		return -EINVAL;
-
-	offset = pos & (sb->s_blocksize - 1);
-	thissize = sb->s_blocksize - offset;
-	if (thissize > 26)
-		thissize = 26;
+	int ret;
 
-	memcpy(&de, dir->bh[buffer]->b_data + offset, thissize);
-	if (thissize != 26)
-		memcpy(((char *)&de) + thissize, dir->bh[buffer + 1]->b_data,
-		       26 - thissize);
+	ret = adfs_dir_copyfrom(&de, dir, pos, 26);
+	if (ret)
+		return ret;
 
 	if (!de.dirobname[0])
 		return -ENOENT;
@@ -260,89 +230,6 @@ __adfs_dir_get(struct adfs_dir *dir, int pos, struct object_info *obj)
 }
 
 static int
-__adfs_dir_put(struct adfs_dir *dir, int pos, struct object_info *obj)
-{
-	struct super_block *sb = dir->sb;
-	struct adfs_direntry de;
-	int thissize, buffer, offset;
-
-	buffer = pos >> sb->s_blocksize_bits;
-
-	if (buffer > dir->nr_buffers)
-		return -EINVAL;
-
-	offset = pos & (sb->s_blocksize - 1);
-	thissize = sb->s_blocksize - offset;
-	if (thissize > 26)
-		thissize = 26;
-
-	/*
-	 * Get the entry in total
-	 */
-	memcpy(&de, dir->bh[buffer]->b_data + offset, thissize);
-	if (thissize != 26)
-		memcpy(((char *)&de) + thissize, dir->bh[buffer + 1]->b_data,
-		       26 - thissize);
-
-	/*
-	 * update it
-	 */
-	adfs_obj2dir(&de, obj);
-
-	/*
-	 * Put the new entry back
-	 */
-	memcpy(dir->bh[buffer]->b_data + offset, &de, thissize);
-	if (thissize != 26)
-		memcpy(dir->bh[buffer + 1]->b_data, ((char *)&de) + thissize,
-		       26 - thissize);
-
-	return 0;
-}
-
-/*
- * the caller is responsible for holding the necessary
- * locks.
- */
-static int adfs_dir_find_entry(struct adfs_dir *dir, u32 indaddr)
-{
-	int pos, ret;
-
-	ret = -ENOENT;
-
-	for (pos = 5; pos < ADFS_NUM_DIR_ENTRIES * 26 + 5; pos += 26) {
-		struct object_info obj;
-
-		if (!__adfs_dir_get(dir, pos, &obj))
-			break;
-
-		if (obj.indaddr == indaddr) {
-			ret = pos;
-			break;
-		}
-	}
-
-	return ret;
-}
-
-static int adfs_f_read(struct super_block *sb, u32 indaddr, unsigned int size,
-		       struct adfs_dir *dir)
-{
-	int ret;
-
-	if (size != ADFS_NEWDIR_SIZE)
-		return -EIO;
-
-	ret = adfs_dir_read(sb, indaddr, size, dir);
-	if (ret)
-		adfs_error(sb, "unable to read directory");
-	else
-		dir->parent_id = adfs_readval(dir->dirtail.new.dirparent, 3);
-
-	return ret;
-}
-
-static int
 adfs_f_setpos(struct adfs_dir *dir, unsigned int fpos)
 {
 	if (fpos >= ADFS_NUM_DIR_ENTRIES)
@@ -364,99 +251,74 @@ adfs_f_getnext(struct adfs_dir *dir, struct object_info *obj)
 	return ret;
 }
 
-static int
-adfs_f_update(struct adfs_dir *dir, struct object_info *obj)
+static int adfs_f_iterate(struct adfs_dir *dir, struct dir_context *ctx)
 {
-	struct super_block *sb = dir->sb;
-	int ret, i;
+	struct object_info obj;
+	int pos = 5 + (ctx->pos - 2) * 26;
 
-	ret = adfs_dir_find_entry(dir, obj->indaddr);
-	if (ret < 0) {
-		adfs_error(dir->sb, "unable to locate entry to update");
-		goto out;
+	while (ctx->pos < 2 + ADFS_NUM_DIR_ENTRIES) {
+		if (__adfs_dir_get(dir, pos, &obj))
+			break;
+		if (!dir_emit(ctx, obj.name, obj.name_len,
+			      obj.indaddr, DT_UNKNOWN))
+			break;
+		pos += 26;
+		ctx->pos++;
 	}
+	return 0;
+}
 
-	__adfs_dir_put(dir, ret, obj);
- 
-	/*
-	 * Increment directory sequence number
-	 */
-	dir->bh[0]->b_data[0] += 1;
-	dir->bh[dir->nr_buffers - 1]->b_data[sb->s_blocksize - 6] += 1;
-
-	ret = adfs_dir_checkbyte(dir);
-	/*
-	 * Update directory check byte
-	 */
-	dir->bh[dir->nr_buffers - 1]->b_data[sb->s_blocksize - 1] = ret;
-
-#if 1
-	{
-	const unsigned int blocksize_bits = sb->s_blocksize_bits;
-
-	memcpy(&dir->dirhead, bufoff(dir->bh, 0), sizeof(dir->dirhead));
-	memcpy(&dir->dirtail, bufoff(dir->bh, 2007), sizeof(dir->dirtail));
+static int adfs_f_update(struct adfs_dir *dir, struct object_info *obj)
+{
+	struct adfs_direntry de;
+	int offset, ret;
 
-	if (dir->dirhead.startmasseq != dir->dirtail.new.endmasseq ||
-	    memcmp(&dir->dirhead.startname, &dir->dirtail.new.endname, 4))
-		goto bad_dir;
+	offset = 5 - (int)sizeof(de);
 
-	if (memcmp(&dir->dirhead.startname, "Nick", 4) &&
-	    memcmp(&dir->dirhead.startname, "Hugo", 4))
-		goto bad_dir;
+	do {
+		offset += sizeof(de);
+		ret = adfs_dir_copyfrom(&de, dir, offset, sizeof(de));
+		if (ret) {
+			adfs_error(dir->sb, "error reading directory entry");
+			return -ENOENT;
+		}
+		if (!de.dirobname[0]) {
+			adfs_error(dir->sb, "unable to locate entry to update");
+			return -ENOENT;
+		}
+	} while (adfs_readval(de.dirinddiscadd, 3) != obj->indaddr);
 
-	if (adfs_dir_checkbyte(dir) != dir->dirtail.new.dircheckbyte)
-		goto bad_dir;
-	}
-#endif
-	for (i = dir->nr_buffers - 1; i >= 0; i--)
-		mark_buffer_dirty(dir->bh[i]);
+	/* Update the directory entry with the new object state */
+	adfs_obj2dir(&de, obj);
 
-	ret = 0;
-out:
-	return ret;
-#if 1
-bad_dir:
-	adfs_error(dir->sb, "whoops!  I broke a directory!");
-	return -EIO;
-#endif
+	/* Write the directory entry back to the directory */
+	return adfs_dir_copyto(dir, offset, &de, 26);
 }
 
-static int
-adfs_f_sync(struct adfs_dir *dir)
+static int adfs_f_commit(struct adfs_dir *dir)
 {
-	int err = 0;
-	int i;
-
-	for (i = dir->nr_buffers - 1; i >= 0; i--) {
-		struct buffer_head *bh = dir->bh[i];
-		sync_dirty_buffer(bh);
-		if (buffer_req(bh) && !buffer_uptodate(bh))
-			err = -EIO;
-	}
+	int ret;
 
-	return err;
-}
+	/* Increment directory sequence number */
+	dir->dirhead->startmasseq += 1;
+	dir->newtail->endmasseq += 1;
 
-static void
-adfs_f_free(struct adfs_dir *dir)
-{
-	int i;
+	/* Update directory check byte */
+	dir->newtail->dircheckbyte = adfs_dir_checkbyte(dir);
 
-	for (i = dir->nr_buffers - 1; i >= 0; i--) {
-		brelse(dir->bh[i]);
-		dir->bh[i] = NULL;
-	}
+	/* Make sure the directory still validates correctly */
+	ret = adfs_f_validate(dir);
+	if (ret)
+		adfs_msg(dir->sb, KERN_ERR, "error: update broke directory");
 
-	dir->nr_buffers = 0;
-	dir->sb = NULL;
+	return ret;
 }
 
 const struct adfs_dir_ops adfs_f_dir_ops = {
 	.read		= adfs_f_read,
+	.iterate	= adfs_f_iterate,
 	.setpos		= adfs_f_setpos,
 	.getnext	= adfs_f_getnext,
 	.update		= adfs_f_update,
-	.sync		= adfs_f_sync,
-	.free		= adfs_f_free
+	.commit		= adfs_f_commit,
 };
diff --git a/fs/adfs/dir_f.h b/fs/adfs/dir_f.h
index 5aec332b90f5..a5393e6cf9f4 100644
--- a/fs/adfs/dir_f.h
+++ b/fs/adfs/dir_f.h
@@ -13,9 +13,9 @@
  * Directory header
  */
 struct adfs_dirheader {
-	unsigned char startmasseq;
-	unsigned char startname[4];
-};
+	__u8 startmasseq;
+	__u8 startname[4];
+} __attribute__((packed));
 
 #define ADFS_NEWDIR_SIZE	2048
 #define ADFS_NUM_DIR_ENTRIES	77
@@ -31,32 +31,36 @@ struct adfs_direntry {
 	__u8 dirlen[4];
 	__u8 dirinddiscadd[3];
 	__u8 newdiratts;
-};
+} __attribute__((packed));
 
 /*
  * Directory tail
  */
+struct adfs_olddirtail {
+	__u8 dirlastmask;
+	char dirname[10];
+	__u8 dirparent[3];
+	char dirtitle[19];
+	__u8 reserved[14];
+	__u8 endmasseq;
+	__u8 endname[4];
+	__u8 dircheckbyte;
+} __attribute__((packed));
+
+struct adfs_newdirtail {
+	__u8 dirlastmask;
+	__u8 reserved[2];
+	__u8 dirparent[3];
+	char dirtitle[19];
+	char dirname[10];
+	__u8 endmasseq;
+	__u8 endname[4];
+	__u8 dircheckbyte;
+} __attribute__((packed));
+
 union adfs_dirtail {
-	struct {
-		unsigned char dirlastmask;
-		char dirname[10];
-		unsigned char dirparent[3];
-		char dirtitle[19];
-		unsigned char reserved[14];
-		unsigned char endmasseq;
-		unsigned char endname[4];
-		unsigned char dircheckbyte;
-	} old;
-	struct {
-		unsigned char dirlastmask;
-		unsigned char reserved[2];
-		unsigned char dirparent[3];
-		char dirtitle[19];
-		char dirname[10];
-		unsigned char endmasseq;
-		unsigned char endname[4];
-		unsigned char dircheckbyte;
-	} new;
+	struct adfs_olddirtail old;
+	struct adfs_newdirtail new;
 };
 
 #endif
diff --git a/fs/adfs/dir_fplus.c b/fs/adfs/dir_fplus.c
index d56924c11b17..4a15924014da 100644
--- a/fs/adfs/dir_fplus.c
+++ b/fs/adfs/dir_fplus.c
@@ -4,123 +4,163 @@
  *
  *  Copyright (C) 1997-1999 Russell King
  */
-#include <linux/slab.h>
 #include "adfs.h"
 #include "dir_fplus.h"
 
-static int
-adfs_fplus_read(struct super_block *sb, unsigned int id, unsigned int sz, struct adfs_dir *dir)
+/* Return the byte offset to directory entry pos */
+static unsigned int adfs_fplus_offset(const struct adfs_bigdirheader *h,
+				      unsigned int pos)
 {
-	struct adfs_bigdirheader *h;
-	struct adfs_bigdirtail *t;
-	unsigned long block;
-	unsigned int blk, size;
-	int i, ret = -EIO;
+	return offsetof(struct adfs_bigdirheader, bigdirname) +
+	       ALIGN(le32_to_cpu(h->bigdirnamelen), 4) +
+	       pos * sizeof(struct adfs_bigdirentry);
+}
 
-	dir->nr_buffers = 0;
+static int adfs_fplus_validate_header(const struct adfs_bigdirheader *h)
+{
+	unsigned int size = le32_to_cpu(h->bigdirsize);
+	unsigned int len;
 
-	/* start off using fixed bh set - only alloc for big dirs */
-	dir->bh_fplus = &dir->bh[0];
+	if (h->bigdirversion[0] != 0 || h->bigdirversion[1] != 0 ||
+	    h->bigdirversion[2] != 0 ||
+	    h->bigdirstartname != cpu_to_le32(BIGDIRSTARTNAME) ||
+	    !size || size & 2047 || size > SZ_4M)
+		return -EIO;
 
-	block = __adfs_block_map(sb, id, 0);
-	if (!block) {
-		adfs_error(sb, "dir object %X has a hole at offset 0", id);
-		goto out;
-	}
+	size -= sizeof(struct adfs_bigdirtail) +
+		offsetof(struct adfs_bigdirheader, bigdirname);
 
-	dir->bh_fplus[0] = sb_bread(sb, block);
-	if (!dir->bh_fplus[0])
-		goto out;
-	dir->nr_buffers += 1;
+	/* Check that bigdirnamelen fits within the directory */
+	len = ALIGN(le32_to_cpu(h->bigdirnamelen), 4);
+	if (len > size)
+		return -EIO;
 
-	h = (struct adfs_bigdirheader *)dir->bh_fplus[0]->b_data;
-	size = le32_to_cpu(h->bigdirsize);
-	if (size != sz) {
-		adfs_msg(sb, KERN_WARNING,
-			 "directory header size %X does not match directory size %X",
-			 size, sz);
+	size -= len;
+
+	/* Check that bigdirnamesize fits within the directory */
+	len = le32_to_cpu(h->bigdirnamesize);
+	if (len > size)
+		return -EIO;
+
+	size -= len;
+
+	/*
+	 * Avoid division, we know that absolute maximum number of entries
+	 * can not be so large to cause overflow of the multiplication below.
+	 */
+	len = le32_to_cpu(h->bigdirentries);
+	if (len > SZ_4M / sizeof(struct adfs_bigdirentry) ||
+	    len * sizeof(struct adfs_bigdirentry) > size)
+		return -EIO;
+
+	return 0;
+}
+
+static int adfs_fplus_validate_tail(const struct adfs_bigdirheader *h,
+				    const struct adfs_bigdirtail *t)
+{
+	if (t->bigdirendname != cpu_to_le32(BIGDIRENDNAME) ||
+	    t->bigdirendmasseq != h->startmasseq ||
+	    t->reserved[0] != 0 || t->reserved[1] != 0)
+		return -EIO;
+
+	return 0;
+}
+
+static u8 adfs_fplus_checkbyte(struct adfs_dir *dir)
+{
+	struct adfs_bigdirheader *h = dir->bighead;
+	struct adfs_bigdirtail *t = dir->bigtail;
+	unsigned int end, bs, bi, i;
+	__le32 *bp;
+	u32 dircheck;
+
+	end = adfs_fplus_offset(h, le32_to_cpu(h->bigdirentries)) +
+		le32_to_cpu(h->bigdirnamesize);
+
+	/* Accumulate the contents of the header, entries and names */
+	for (dircheck = 0, bi = 0; end; bi++) {
+		bp = (void *)dir->bhs[bi]->b_data;
+		bs = dir->bhs[bi]->b_size;
+		if (bs > end)
+			bs = end;
+
+		for (i = 0; i < bs; i += sizeof(u32))
+			dircheck = ror32(dircheck, 13) ^ le32_to_cpup(bp++);
+
+		end -= bs;
 	}
 
-	if (h->bigdirversion[0] != 0 || h->bigdirversion[1] != 0 ||
-	    h->bigdirversion[2] != 0 || size & 2047 ||
-	    h->bigdirstartname != cpu_to_le32(BIGDIRSTARTNAME)) {
-		adfs_error(sb, "dir %06x has malformed header", id);
+	/* Accumulate the contents of the tail except for the check byte */
+	dircheck = ror32(dircheck, 13) ^ le32_to_cpu(t->bigdirendname);
+	dircheck = ror32(dircheck, 13) ^ t->bigdirendmasseq;
+	dircheck = ror32(dircheck, 13) ^ t->reserved[0];
+	dircheck = ror32(dircheck, 13) ^ t->reserved[1];
+
+	return dircheck ^ dircheck >> 8 ^ dircheck >> 16 ^ dircheck >> 24;
+}
+
+static int adfs_fplus_read(struct super_block *sb, u32 indaddr,
+			   unsigned int size, struct adfs_dir *dir)
+{
+	struct adfs_bigdirheader *h;
+	struct adfs_bigdirtail *t;
+	unsigned int dirsize;
+	int ret;
+
+	/* Read first buffer */
+	ret = adfs_dir_read_buffers(sb, indaddr, sb->s_blocksize, dir);
+	if (ret)
+		return ret;
+
+	dir->bighead = h = (void *)dir->bhs[0]->b_data;
+	ret = adfs_fplus_validate_header(h);
+	if (ret) {
+		adfs_error(sb, "dir %06x has malformed header", indaddr);
 		goto out;
 	}
 
-	size >>= sb->s_blocksize_bits;
-	if (size > ARRAY_SIZE(dir->bh)) {
-		/* this directory is too big for fixed bh set, must allocate */
-		struct buffer_head **bh_fplus =
-			kcalloc(size, sizeof(struct buffer_head *),
-				GFP_KERNEL);
-		if (!bh_fplus) {
-			adfs_msg(sb, KERN_ERR,
-				 "not enough memory for dir object %X (%d blocks)",
-				 id, size);
-			ret = -ENOMEM;
-			goto out;
-		}
-		dir->bh_fplus = bh_fplus;
-		/* copy over the pointer to the block that we've already read */
-		dir->bh_fplus[0] = dir->bh[0];
+	dirsize = le32_to_cpu(h->bigdirsize);
+	if (size && dirsize != size) {
+		adfs_msg(sb, KERN_WARNING,
+			 "dir %06x header size %X does not match directory size %X",
+			 indaddr, dirsize, size);
 	}
 
-	for (blk = 1; blk < size; blk++) {
-		block = __adfs_block_map(sb, id, blk);
-		if (!block) {
-			adfs_error(sb, "dir object %X has a hole at offset %d", id, blk);
-			goto out;
-		}
+	/* Read remaining buffers */
+	ret = adfs_dir_read_buffers(sb, indaddr, dirsize, dir);
+	if (ret)
+		return ret;
 
-		dir->bh_fplus[blk] = sb_bread(sb, block);
-		if (!dir->bh_fplus[blk]) {
-			adfs_error(sb,	"dir object %x failed read for offset %d, mapped block %lX",
-				   id, blk, block);
-			goto out;
-		}
+	dir->bigtail = t = (struct adfs_bigdirtail *)
+		(dir->bhs[dir->nr_buffers - 1]->b_data + (sb->s_blocksize - 8));
 
-		dir->nr_buffers += 1;
+	ret = adfs_fplus_validate_tail(h, t);
+	if (ret) {
+		adfs_error(sb, "dir %06x has malformed tail", indaddr);
+		goto out;
 	}
 
-	t = (struct adfs_bigdirtail *)
-		(dir->bh_fplus[size - 1]->b_data + (sb->s_blocksize - 8));
-
-	if (t->bigdirendname != cpu_to_le32(BIGDIRENDNAME) ||
-	    t->bigdirendmasseq != h->startmasseq ||
-	    t->reserved[0] != 0 || t->reserved[1] != 0) {
-		adfs_error(sb, "dir %06x has malformed tail", id);
+	if (adfs_fplus_checkbyte(dir) != t->bigdircheckbyte) {
+		adfs_error(sb, "dir %06x checkbyte mismatch\n", indaddr);
 		goto out;
 	}
 
 	dir->parent_id = le32_to_cpu(h->bigdirparent);
-	dir->sb = sb;
 	return 0;
 
 out:
-	if (dir->bh_fplus) {
-		for (i = 0; i < dir->nr_buffers; i++)
-			brelse(dir->bh_fplus[i]);
-
-		if (&dir->bh[0] != dir->bh_fplus)
-			kfree(dir->bh_fplus);
+	adfs_dir_relse(dir);
 
-		dir->bh_fplus = NULL;
-	}
-
-	dir->nr_buffers = 0;
-	dir->sb = NULL;
 	return ret;
 }
 
 static int
 adfs_fplus_setpos(struct adfs_dir *dir, unsigned int fpos)
 {
-	struct adfs_bigdirheader *h =
-		(struct adfs_bigdirheader *) dir->bh_fplus[0]->b_data;
 	int ret = -ENOENT;
 
-	if (fpos <= le32_to_cpu(h->bigdirentries)) {
+	if (fpos <= le32_to_cpu(dir->bighead->bigdirentries)) {
 		dir->pos = fpos;
 		ret = 0;
 	}
@@ -128,51 +168,23 @@ adfs_fplus_setpos(struct adfs_dir *dir, unsigned int fpos)
 	return ret;
 }
 
-static void
-dir_memcpy(struct adfs_dir *dir, unsigned int offset, void *to, int len)
-{
-	struct super_block *sb = dir->sb;
-	unsigned int buffer, partial, remainder;
-
-	buffer = offset >> sb->s_blocksize_bits;
-	offset &= sb->s_blocksize - 1;
-
-	partial = sb->s_blocksize - offset;
-
-	if (partial >= len)
-		memcpy(to, dir->bh_fplus[buffer]->b_data + offset, len);
-	else {
-		char *c = (char *)to;
-
-		remainder = len - partial;
-
-		memcpy(c,
-			dir->bh_fplus[buffer]->b_data + offset,
-			partial);
-
-		memcpy(c + partial,
-			dir->bh_fplus[buffer + 1]->b_data,
-			remainder);
-	}
-}
-
 static int
 adfs_fplus_getnext(struct adfs_dir *dir, struct object_info *obj)
 {
-	struct adfs_bigdirheader *h =
-		(struct adfs_bigdirheader *) dir->bh_fplus[0]->b_data;
+	struct adfs_bigdirheader *h = dir->bighead;
 	struct adfs_bigdirentry bde;
 	unsigned int offset;
-	int ret = -ENOENT;
+	int ret;
 
 	if (dir->pos >= le32_to_cpu(h->bigdirentries))
-		goto out;
+		return -ENOENT;
 
-	offset = offsetof(struct adfs_bigdirheader, bigdirname);
-	offset += ((le32_to_cpu(h->bigdirnamelen) + 4) & ~3);
-	offset += dir->pos * sizeof(struct adfs_bigdirentry);
+	offset = adfs_fplus_offset(h, dir->pos);
 
-	dir_memcpy(dir, offset, &bde, sizeof(struct adfs_bigdirentry));
+	ret = adfs_dir_copyfrom(&bde, dir, offset,
+				sizeof(struct adfs_bigdirentry));
+	if (ret)
+		return ret;
 
 	obj->loadaddr = le32_to_cpu(bde.bigdirload);
 	obj->execaddr = le32_to_cpu(bde.bigdirexec);
@@ -181,59 +193,95 @@ adfs_fplus_getnext(struct adfs_dir *dir, struct object_info *obj)
 	obj->attr     = le32_to_cpu(bde.bigdirattr);
 	obj->name_len = le32_to_cpu(bde.bigdirobnamelen);
 
-	offset = offsetof(struct adfs_bigdirheader, bigdirname);
-	offset += ((le32_to_cpu(h->bigdirnamelen) + 4) & ~3);
-	offset += le32_to_cpu(h->bigdirentries) * sizeof(struct adfs_bigdirentry);
+	offset = adfs_fplus_offset(h, le32_to_cpu(h->bigdirentries));
 	offset += le32_to_cpu(bde.bigdirobnameptr);
 
-	dir_memcpy(dir, offset, obj->name, obj->name_len);
+	ret = adfs_dir_copyfrom(obj->name, dir, offset, obj->name_len);
+	if (ret)
+		return ret;
+
 	adfs_object_fixup(dir, obj);
 
 	dir->pos += 1;
-	ret = 0;
-out:
-	return ret;
+
+	return 0;
 }
 
-static int
-adfs_fplus_sync(struct adfs_dir *dir)
+static int adfs_fplus_iterate(struct adfs_dir *dir, struct dir_context *ctx)
 {
-	int err = 0;
-	int i;
-
-	for (i = dir->nr_buffers - 1; i >= 0; i--) {
-		struct buffer_head *bh = dir->bh_fplus[i];
-		sync_dirty_buffer(bh);
-		if (buffer_req(bh) && !buffer_uptodate(bh))
-			err = -EIO;
+	struct object_info obj;
+
+	if ((ctx->pos - 2) >> 32)
+		return 0;
+
+	if (adfs_fplus_setpos(dir, ctx->pos - 2))
+		return 0;
+
+	while (!adfs_fplus_getnext(dir, &obj)) {
+		if (!dir_emit(ctx, obj.name, obj.name_len,
+			      obj.indaddr, DT_UNKNOWN))
+			break;
+		ctx->pos++;
 	}
 
-	return err;
+	return 0;
 }
 
-static void
-adfs_fplus_free(struct adfs_dir *dir)
+static int adfs_fplus_update(struct adfs_dir *dir, struct object_info *obj)
 {
-	int i;
+	struct adfs_bigdirheader *h = dir->bighead;
+	struct adfs_bigdirentry bde;
+	int offset, end, ret;
 
-	if (dir->bh_fplus) {
-		for (i = 0; i < dir->nr_buffers; i++)
-			brelse(dir->bh_fplus[i]);
+	offset = adfs_fplus_offset(h, 0) - sizeof(bde);
+	end = adfs_fplus_offset(h, le32_to_cpu(h->bigdirentries));
 
-		if (&dir->bh[0] != dir->bh_fplus)
-			kfree(dir->bh_fplus);
+	do {
+		offset += sizeof(bde);
+		if (offset >= end) {
+			adfs_error(dir->sb, "unable to locate entry to update");
+			return -ENOENT;
+		}
+		ret = adfs_dir_copyfrom(&bde, dir, offset, sizeof(bde));
+		if (ret) {
+			adfs_error(dir->sb, "error reading directory entry");
+			return -ENOENT;
+		}
+	} while (le32_to_cpu(bde.bigdirindaddr) != obj->indaddr);
 
-		dir->bh_fplus = NULL;
-	}
+	bde.bigdirload    = cpu_to_le32(obj->loadaddr);
+	bde.bigdirexec    = cpu_to_le32(obj->execaddr);
+	bde.bigdirlen     = cpu_to_le32(obj->size);
+	bde.bigdirindaddr = cpu_to_le32(obj->indaddr);
+	bde.bigdirattr    = cpu_to_le32(obj->attr);
+
+	return adfs_dir_copyto(dir, offset, &bde, sizeof(bde));
+}
+
+static int adfs_fplus_commit(struct adfs_dir *dir)
+{
+	int ret;
 
-	dir->nr_buffers = 0;
-	dir->sb = NULL;
+	/* Increment directory sequence number */
+	dir->bighead->startmasseq += 1;
+	dir->bigtail->bigdirendmasseq += 1;
+
+	/* Update directory check byte */
+	dir->bigtail->bigdircheckbyte = adfs_fplus_checkbyte(dir);
+
+	/* Make sure the directory still validates correctly */
+	ret = adfs_fplus_validate_header(dir->bighead);
+	if (ret == 0)
+		ret = adfs_fplus_validate_tail(dir->bighead, dir->bigtail);
+
+	return ret;
 }
 
 const struct adfs_dir_ops adfs_fplus_dir_ops = {
 	.read		= adfs_fplus_read,
+	.iterate	= adfs_fplus_iterate,
 	.setpos		= adfs_fplus_setpos,
 	.getnext	= adfs_fplus_getnext,
-	.sync		= adfs_fplus_sync,
-	.free		= adfs_fplus_free
+	.update		= adfs_fplus_update,
+	.commit		= adfs_fplus_commit,
 };
diff --git a/fs/adfs/dir_fplus.h b/fs/adfs/dir_fplus.h
index 4ec0931e36ad..d729b1591e5e 100644
--- a/fs/adfs/dir_fplus.h
+++ b/fs/adfs/dir_fplus.h
@@ -22,7 +22,7 @@ struct adfs_bigdirheader {
 	__le32	bigdirnamesize;
 	__le32	bigdirparent;
 	char	bigdirname[1];
-};
+} __attribute__((packed, aligned(4)));
 
 struct adfs_bigdirentry {
 	__le32	bigdirload;
@@ -32,11 +32,11 @@ struct adfs_bigdirentry {
 	__le32	bigdirattr;
 	__le32	bigdirobnamelen;
 	__le32	bigdirobnameptr;
-};
+} __attribute__((packed, aligned(4)));
 
 struct adfs_bigdirtail {
 	__le32	bigdirendname;
 	__u8	bigdirendmasseq;
 	__u8	reserved[2];
 	__u8	bigdircheckbyte;
-};
+} __attribute__((packed, aligned(4)));
diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c
index 124de75413a5..32620f4a7623 100644
--- a/fs/adfs/inode.c
+++ b/fs/adfs/inode.c
@@ -20,7 +20,8 @@ adfs_get_block(struct inode *inode, sector_t block, struct buffer_head *bh,
 		if (block >= inode->i_blocks)
 			goto abort_toobig;
 
-		block = __adfs_block_map(inode->i_sb, inode->i_ino, block);
+		block = __adfs_block_map(inode->i_sb, ADFS_I(inode)->indaddr,
+					 block);
 		if (block)
 			map_bh(bh, inode->i_sb, block);
 		return 0;
@@ -126,29 +127,29 @@ adfs_atts2mode(struct super_block *sb, struct inode *inode)
  * Convert Linux permission to ADFS attribute.  We try to do the reverse
  * of atts2mode, but there is not a 1:1 translation.
  */
-static int
-adfs_mode2atts(struct super_block *sb, struct inode *inode)
+static int adfs_mode2atts(struct super_block *sb, struct inode *inode,
+			  umode_t ia_mode)
 {
+	struct adfs_sb_info *asb = ADFS_SB(sb);
 	umode_t mode;
 	int attr;
-	struct adfs_sb_info *asb = ADFS_SB(sb);
 
 	/* FIXME: should we be able to alter a link? */
 	if (S_ISLNK(inode->i_mode))
 		return ADFS_I(inode)->attr;
 
+	/* Directories do not have read/write permissions on the media */
 	if (S_ISDIR(inode->i_mode))
-		attr = ADFS_NDA_DIRECTORY;
-	else
-		attr = 0;
+		return ADFS_NDA_DIRECTORY;
 
-	mode = inode->i_mode & asb->s_owner_mask;
+	attr = 0;
+	mode = ia_mode & asb->s_owner_mask;
 	if (mode & S_IRUGO)
 		attr |= ADFS_NDA_OWNER_READ;
 	if (mode & S_IWUGO)
 		attr |= ADFS_NDA_OWNER_WRITE;
 
-	mode = inode->i_mode & asb->s_other_mask;
+	mode = ia_mode & asb->s_other_mask;
 	mode &= ~asb->s_owner_mask;
 	if (mode & S_IRUGO)
 		attr |= ADFS_NDA_PUBLIC_READ;
@@ -158,6 +159,8 @@ adfs_mode2atts(struct super_block *sb, struct inode *inode)
 	return attr;
 }
 
+static const s64 nsec_unix_epoch_diff_risc_os_epoch = 2208988800000000000LL;
+
 /*
  * Convert an ADFS time to Unix time.  ADFS has a 40-bit centi-second time
  * referenced to 1 Jan 1900 (til 2248) so we need to discard 2208988800 seconds
@@ -170,8 +173,6 @@ adfs_adfs2unix_time(struct timespec64 *tv, struct inode *inode)
 	/* 01 Jan 1970 00:00:00 (Unix epoch) as nanoseconds since
 	 * 01 Jan 1900 00:00:00 (RISC OS epoch)
 	 */
-	static const s64 nsec_unix_epoch_diff_risc_os_epoch =
-							2208988800000000000LL;
 	s64 nsec;
 
 	if (!adfs_inode_is_stamped(inode))
@@ -204,24 +205,23 @@ adfs_adfs2unix_time(struct timespec64 *tv, struct inode *inode)
 	return;
 }
 
-/*
- * Convert an Unix time to ADFS time.  We only do this if the entry has a
- * time/date stamp already.
- */
-static void
-adfs_unix2adfs_time(struct inode *inode, unsigned int secs)
+/* Convert an Unix time to ADFS time for an entry that is already stamped. */
+static void adfs_unix2adfs_time(struct inode *inode,
+				const struct timespec64 *ts)
 {
-	unsigned int high, low;
+	s64 cs, nsec = timespec64_to_ns(ts);
 
-	if (adfs_inode_is_stamped(inode)) {
-		/* convert 32-bit seconds to 40-bit centi-seconds */
-		low  = (secs & 255) * 100;
-		high = (secs / 256) * 100 + (low >> 8) + 0x336e996a;
+	/* convert from Unix to RISC OS epoch */
+	nsec += nsec_unix_epoch_diff_risc_os_epoch;
 
-		ADFS_I(inode)->loadaddr = (high >> 24) |
-				(ADFS_I(inode)->loadaddr & ~0xff);
-		ADFS_I(inode)->execaddr = (low & 255) | (high << 8);
-	}
+	/* convert from nanoseconds to centiseconds */
+	cs = div_s64(nsec, 10000000);
+
+	cs = clamp_t(s64, cs, 0, 0xffffffffff);
+
+	ADFS_I(inode)->loadaddr &= ~0xff;
+	ADFS_I(inode)->loadaddr |= (cs >> 32) & 0xff;
+	ADFS_I(inode)->execaddr = cs;
 }
 
 /*
@@ -260,6 +260,7 @@ adfs_iget(struct super_block *sb, struct object_info *obj)
 	 * for cross-directory renames.
 	 */
 	ADFS_I(inode)->parent_id = obj->parent_id;
+	ADFS_I(inode)->indaddr   = obj->indaddr;
 	ADFS_I(inode)->loadaddr  = obj->loadaddr;
 	ADFS_I(inode)->execaddr  = obj->execaddr;
 	ADFS_I(inode)->attr      = obj->attr;
@@ -315,10 +316,11 @@ adfs_notify_change(struct dentry *dentry, struct iattr *attr)
 	if (ia_valid & ATTR_SIZE)
 		truncate_setsize(inode, attr->ia_size);
 
-	if (ia_valid & ATTR_MTIME) {
-		inode->i_mtime = attr->ia_mtime;
-		adfs_unix2adfs_time(inode, attr->ia_mtime.tv_sec);
+	if (ia_valid & ATTR_MTIME && adfs_inode_is_stamped(inode)) {
+		adfs_unix2adfs_time(inode, &attr->ia_mtime);
+		adfs_adfs2unix_time(&inode->i_mtime, inode);
 	}
+
 	/*
 	 * FIXME: should we make these == to i_mtime since we don't
 	 * have the ability to represent them in our filesystem?
@@ -328,7 +330,7 @@ adfs_notify_change(struct dentry *dentry, struct iattr *attr)
 	if (ia_valid & ATTR_CTIME)
 		inode->i_ctime = attr->ia_ctime;
 	if (ia_valid & ATTR_MODE) {
-		ADFS_I(inode)->attr = adfs_mode2atts(sb, inode);
+		ADFS_I(inode)->attr = adfs_mode2atts(sb, inode, attr->ia_mode);
 		inode->i_mode = adfs_atts2mode(sb, inode);
 	}
 
@@ -353,7 +355,7 @@ int adfs_write_inode(struct inode *inode, struct writeback_control *wbc)
 	struct object_info obj;
 	int ret;
 
-	obj.indaddr	= inode->i_ino;
+	obj.indaddr	= ADFS_I(inode)->indaddr;
 	obj.name_len	= 0;
 	obj.parent_id	= ADFS_I(inode)->parent_id;
 	obj.loadaddr	= ADFS_I(inode)->loadaddr;
diff --git a/fs/adfs/map.c b/fs/adfs/map.c
index f44d12cef5be..a81de80c45c1 100644
--- a/fs/adfs/map.c
+++ b/fs/adfs/map.c
@@ -4,6 +4,8 @@
  *
  *  Copyright (C) 1997-2002 Russell King
  */
+#include <linux/slab.h>
+#include <linux/statfs.h>
 #include <asm/unaligned.h>
 #include "adfs.h"
 
@@ -66,54 +68,41 @@ static DEFINE_RWLOCK(adfs_map_lock);
 static int lookup_zone(const struct adfs_discmap *dm, const unsigned int idlen,
 		       const u32 frag_id, unsigned int *offset)
 {
-	const unsigned int mapsize = dm->dm_endbit;
+	const unsigned int endbit = dm->dm_endbit;
 	const u32 idmask = (1 << idlen) - 1;
-	unsigned char *map = dm->dm_bh->b_data + 4;
+	unsigned char *map = dm->dm_bh->b_data;
 	unsigned int start = dm->dm_startbit;
-	unsigned int mapptr;
+	unsigned int freelink, fragend;
 	u32 frag;
 
+	frag = GET_FRAG_ID(map, 8, idmask & 0x7fff);
+	freelink = frag ? 8 + frag : 0;
+
 	do {
 		frag = GET_FRAG_ID(map, start, idmask);
-		mapptr = start + idlen;
-
-		/*
-		 * find end of fragment
-		 */
-		{
-			__le32 *_map = (__le32 *)map;
-			u32 v = le32_to_cpu(_map[mapptr >> 5]) >> (mapptr & 31);
-			while (v == 0) {
-				mapptr = (mapptr & ~31) + 32;
-				if (mapptr >= mapsize)
-					goto error;
-				v = le32_to_cpu(_map[mapptr >> 5]);
-			}
-
-			mapptr += 1 + ffz(~v);
+
+		fragend = find_next_bit_le(map, endbit, start + idlen);
+		if (fragend >= endbit)
+			goto error;
+
+		if (start == freelink) {
+			freelink += frag & 0x7fff;
+		} else if (frag == frag_id) {
+			unsigned int length = fragend + 1 - start;
+
+			if (*offset < length)
+				return start + *offset;
+			*offset -= length;
 		}
 
-		if (frag == frag_id)
-			goto found;
-again:
-		start = mapptr;
-	} while (mapptr < mapsize);
+		start = fragend + 1;
+	} while (start < endbit);
 	return -1;
 
 error:
 	printk(KERN_ERR "adfs: oversized fragment 0x%x at 0x%x-0x%x\n",
-		frag, start, mapptr);
+		frag, start, fragend);
 	return -1;
-
-found:
-	{
-		int length = mapptr - start;
-		if (*offset >= length) {
-			*offset -= length;
-			goto again;
-		}
-	}
-	return start + *offset;
 }
 
 /*
@@ -125,12 +114,12 @@ found:
 static unsigned int
 scan_free_map(struct adfs_sb_info *asb, struct adfs_discmap *dm)
 {
-	const unsigned int mapsize = dm->dm_endbit + 32;
+	const unsigned int endbit = dm->dm_endbit;
 	const unsigned int idlen  = asb->s_idlen;
 	const unsigned int frag_idlen = idlen <= 15 ? idlen : 15;
 	const u32 idmask = (1 << frag_idlen) - 1;
 	unsigned char *map = dm->dm_bh->b_data;
-	unsigned int start = 8, mapptr;
+	unsigned int start = 8, fragend;
 	u32 frag;
 	unsigned long total = 0;
 
@@ -149,29 +138,13 @@ scan_free_map(struct adfs_sb_info *asb, struct adfs_discmap *dm)
 	do {
 		start += frag;
 
-		/*
-		 * get fragment id
-		 */
 		frag = GET_FRAG_ID(map, start, idmask);
-		mapptr = start + idlen;
-
-		/*
-		 * find end of fragment
-		 */
-		{
-			__le32 *_map = (__le32 *)map;
-			u32 v = le32_to_cpu(_map[mapptr >> 5]) >> (mapptr & 31);
-			while (v == 0) {
-				mapptr = (mapptr & ~31) + 32;
-				if (mapptr >= mapsize)
-					goto error;
-				v = le32_to_cpu(_map[mapptr >> 5]);
-			}
-
-			mapptr += 1 + ffz(~v);
-		}
 
-		total += mapptr - start;
+		fragend = find_next_bit_le(map, endbit, start + idlen);
+		if (fragend >= endbit)
+			goto error;
+
+		total += fragend + 1 - start;
 	} while (frag >= idlen + 1);
 
 	if (frag != 0)
@@ -220,10 +193,10 @@ found:
  *  total_free = E(free_in_zone_n)
  *              nzones
  */
-unsigned int
-adfs_map_free(struct super_block *sb)
+void adfs_map_statfs(struct super_block *sb, struct kstatfs *buf)
 {
 	struct adfs_sb_info *asb = ADFS_SB(sb);
+	struct adfs_discrecord *dr = adfs_map_discrecord(asb->s_map);
 	struct adfs_discmap *dm;
 	unsigned int total = 0;
 	unsigned int zone;
@@ -235,7 +208,10 @@ adfs_map_free(struct super_block *sb)
 		total += scan_free_map(asb, dm++);
 	} while (--zone > 0);
 
-	return signed_asl(total, asb->s_map2blk);
+	buf->f_blocks  = adfs_disc_size(dr) >> sb->s_blocksize_bits;
+	buf->f_files   = asb->s_ids_per_zone * asb->s_map_size;
+	buf->f_bavail  =
+	buf->f_bfree   = signed_asl(total, asb->s_map2blk);
 }
 
 int adfs_map_lookup(struct super_block *sb, u32 frag_id, unsigned int offset)
@@ -280,3 +256,152 @@ bad_fragment:
 		   frag_id, zone, asb->s_map_size);
 	return 0;
 }
+
+static unsigned char adfs_calczonecheck(struct super_block *sb, unsigned char *map)
+{
+	unsigned int v0, v1, v2, v3;
+	int i;
+
+	v0 = v1 = v2 = v3 = 0;
+	for (i = sb->s_blocksize - 4; i; i -= 4) {
+		v0 += map[i]     + (v3 >> 8);
+		v3 &= 0xff;
+		v1 += map[i + 1] + (v0 >> 8);
+		v0 &= 0xff;
+		v2 += map[i + 2] + (v1 >> 8);
+		v1 &= 0xff;
+		v3 += map[i + 3] + (v2 >> 8);
+		v2 &= 0xff;
+	}
+	v0 +=           v3 >> 8;
+	v1 += map[1] + (v0 >> 8);
+	v2 += map[2] + (v1 >> 8);
+	v3 += map[3] + (v2 >> 8);
+
+	return v0 ^ v1 ^ v2 ^ v3;
+}
+
+static int adfs_checkmap(struct super_block *sb, struct adfs_discmap *dm)
+{
+	unsigned char crosscheck = 0, zonecheck = 1;
+	int i;
+
+	for (i = 0; i < ADFS_SB(sb)->s_map_size; i++) {
+		unsigned char *map;
+
+		map = dm[i].dm_bh->b_data;
+
+		if (adfs_calczonecheck(sb, map) != map[0]) {
+			adfs_error(sb, "zone %d fails zonecheck", i);
+			zonecheck = 0;
+		}
+		crosscheck ^= map[3];
+	}
+	if (crosscheck != 0xff)
+		adfs_error(sb, "crosscheck != 0xff");
+	return crosscheck == 0xff && zonecheck;
+}
+
+/*
+ * Layout the map - the first zone contains a copy of the disc record,
+ * and the last zone must be limited to the size of the filesystem.
+ */
+static void adfs_map_layout(struct adfs_discmap *dm, unsigned int nzones,
+			    struct adfs_discrecord *dr)
+{
+	unsigned int zone, zone_size;
+	u64 size;
+
+	zone_size = (8 << dr->log2secsize) - le16_to_cpu(dr->zone_spare);
+
+	dm[0].dm_bh       = NULL;
+	dm[0].dm_startblk = 0;
+	dm[0].dm_startbit = 32 + ADFS_DR_SIZE_BITS;
+	dm[0].dm_endbit   = 32 + zone_size;
+
+	for (zone = 1; zone < nzones; zone++) {
+		dm[zone].dm_bh       = NULL;
+		dm[zone].dm_startblk = zone * zone_size - ADFS_DR_SIZE_BITS;
+		dm[zone].dm_startbit = 32;
+		dm[zone].dm_endbit   = 32 + zone_size;
+	}
+
+	size = adfs_disc_size(dr) >> dr->log2bpmb;
+	size -= (nzones - 1) * zone_size - ADFS_DR_SIZE_BITS;
+	dm[nzones - 1].dm_endbit = 32 + size;
+}
+
+static int adfs_map_read(struct adfs_discmap *dm, struct super_block *sb,
+			 unsigned int map_addr, unsigned int nzones)
+{
+	unsigned int zone;
+
+	for (zone = 0; zone < nzones; zone++) {
+		dm[zone].dm_bh = sb_bread(sb, map_addr + zone);
+		if (!dm[zone].dm_bh)
+			return -EIO;
+	}
+
+	return 0;
+}
+
+static void adfs_map_relse(struct adfs_discmap *dm, unsigned int nzones)
+{
+	unsigned int zone;
+
+	for (zone = 0; zone < nzones; zone++)
+		brelse(dm[zone].dm_bh);
+}
+
+struct adfs_discmap *adfs_read_map(struct super_block *sb, struct adfs_discrecord *dr)
+{
+	struct adfs_sb_info *asb = ADFS_SB(sb);
+	struct adfs_discmap *dm;
+	unsigned int map_addr, zone_size, nzones;
+	int ret;
+
+	nzones    = dr->nzones | dr->nzones_high << 8;
+	zone_size = (8 << dr->log2secsize) - le16_to_cpu(dr->zone_spare);
+
+	asb->s_idlen = dr->idlen;
+	asb->s_map_size = nzones;
+	asb->s_map2blk = dr->log2bpmb - dr->log2secsize;
+	asb->s_log2sharesize = dr->log2sharesize;
+	asb->s_ids_per_zone = zone_size / (asb->s_idlen + 1);
+
+	map_addr = (nzones >> 1) * zone_size -
+		     ((nzones > 1) ? ADFS_DR_SIZE_BITS : 0);
+	map_addr = signed_asl(map_addr, asb->s_map2blk);
+
+	dm = kmalloc_array(nzones, sizeof(*dm), GFP_KERNEL);
+	if (dm == NULL) {
+		adfs_error(sb, "not enough memory");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	adfs_map_layout(dm, nzones, dr);
+
+	ret = adfs_map_read(dm, sb, map_addr, nzones);
+	if (ret) {
+		adfs_error(sb, "unable to read map");
+		goto error_free;
+	}
+
+	if (adfs_checkmap(sb, dm))
+		return dm;
+
+	adfs_error(sb, "map corrupted");
+
+error_free:
+	adfs_map_relse(dm, nzones);
+	kfree(dm);
+	return ERR_PTR(-EIO);
+}
+
+void adfs_free_map(struct super_block *sb)
+{
+	struct adfs_sb_info *asb = ADFS_SB(sb);
+
+	adfs_map_relse(asb->s_map, asb->s_map_size);
+	kfree(asb->s_map);
+}
diff --git a/fs/adfs/super.c b/fs/adfs/super.c
index 65b04ebb51c3..a3cc8ecb50da 100644
--- a/fs/adfs/super.c
+++ b/fs/adfs/super.c
@@ -88,59 +88,11 @@ static int adfs_checkdiscrecord(struct adfs_discrecord *dr)
 	return 0;
 }
 
-static unsigned char adfs_calczonecheck(struct super_block *sb, unsigned char *map)
-{
-	unsigned int v0, v1, v2, v3;
-	int i;
-
-	v0 = v1 = v2 = v3 = 0;
-	for (i = sb->s_blocksize - 4; i; i -= 4) {
-		v0 += map[i]     + (v3 >> 8);
-		v3 &= 0xff;
-		v1 += map[i + 1] + (v0 >> 8);
-		v0 &= 0xff;
-		v2 += map[i + 2] + (v1 >> 8);
-		v1 &= 0xff;
-		v3 += map[i + 3] + (v2 >> 8);
-		v2 &= 0xff;
-	}
-	v0 +=           v3 >> 8;
-	v1 += map[1] + (v0 >> 8);
-	v2 += map[2] + (v1 >> 8);
-	v3 += map[3] + (v2 >> 8);
-
-	return v0 ^ v1 ^ v2 ^ v3;
-}
-
-static int adfs_checkmap(struct super_block *sb, struct adfs_discmap *dm)
-{
-	unsigned char crosscheck = 0, zonecheck = 1;
-	int i;
-
-	for (i = 0; i < ADFS_SB(sb)->s_map_size; i++) {
-		unsigned char *map;
-
-		map = dm[i].dm_bh->b_data;
-
-		if (adfs_calczonecheck(sb, map) != map[0]) {
-			adfs_error(sb, "zone %d fails zonecheck", i);
-			zonecheck = 0;
-		}
-		crosscheck ^= map[3];
-	}
-	if (crosscheck != 0xff)
-		adfs_error(sb, "crosscheck != 0xff");
-	return crosscheck == 0xff && zonecheck;
-}
-
 static void adfs_put_super(struct super_block *sb)
 {
-	int i;
 	struct adfs_sb_info *asb = ADFS_SB(sb);
 
-	for (i = 0; i < asb->s_map_size; i++)
-		brelse(asb->s_map[i].dm_bh);
-	kfree(asb->s_map);
+	adfs_free_map(sb);
 	kfree_rcu(asb, rcu);
 }
 
@@ -249,16 +201,13 @@ static int adfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct super_block *sb = dentry->d_sb;
 	struct adfs_sb_info *sbi = ADFS_SB(sb);
-	struct adfs_discrecord *dr = adfs_map_discrecord(sbi->s_map);
 	u64 id = huge_encode_dev(sb->s_bdev->bd_dev);
 
+	adfs_map_statfs(sb, buf);
+
 	buf->f_type    = ADFS_SUPER_MAGIC;
 	buf->f_namelen = sbi->s_namelen;
 	buf->f_bsize   = sb->s_blocksize;
-	buf->f_blocks  = adfs_disc_size(dr) >> sb->s_blocksize_bits;
-	buf->f_files   = sbi->s_ids_per_zone * sbi->s_map_size;
-	buf->f_bavail  =
-	buf->f_bfree   = adfs_map_free(sb);
 	buf->f_ffree   = (long)(buf->f_bfree * buf->f_files) / (long)buf->f_blocks;
 	buf->f_fsid.val[0] = (u32)id;
 	buf->f_fsid.val[1] = (u32)(id >> 32);
@@ -282,6 +231,12 @@ static void adfs_free_inode(struct inode *inode)
 	kmem_cache_free(adfs_inode_cachep, ADFS_I(inode));
 }
 
+static int adfs_drop_inode(struct inode *inode)
+{
+	/* always drop inodes if we are read-only */
+	return !IS_ENABLED(CONFIG_ADFS_FS_RW) || IS_RDONLY(inode);
+}
+
 static void init_once(void *foo)
 {
 	struct adfs_inode_info *ei = (struct adfs_inode_info *) foo;
@@ -314,7 +269,7 @@ static void destroy_inodecache(void)
 static const struct super_operations adfs_sops = {
 	.alloc_inode	= adfs_alloc_inode,
 	.free_inode	= adfs_free_inode,
-	.drop_inode	= generic_delete_inode,
+	.drop_inode	= adfs_drop_inode,
 	.write_inode	= adfs_write_inode,
 	.put_super	= adfs_put_super,
 	.statfs		= adfs_statfs,
@@ -322,66 +277,94 @@ static const struct super_operations adfs_sops = {
 	.show_options	= adfs_show_options,
 };
 
-static struct adfs_discmap *adfs_read_map(struct super_block *sb, struct adfs_discrecord *dr)
+static int adfs_probe(struct super_block *sb, unsigned int offset, int silent,
+		      int (*validate)(struct super_block *sb,
+				      struct buffer_head *bh,
+				      struct adfs_discrecord **bhp))
 {
-	struct adfs_discmap *dm;
-	unsigned int map_addr, zone_size, nzones;
-	int i, zone;
 	struct adfs_sb_info *asb = ADFS_SB(sb);
+	struct adfs_discrecord *dr;
+	struct buffer_head *bh;
+	unsigned int blocksize = BLOCK_SIZE;
+	int ret, try;
+
+	for (try = 0; try < 2; try++) {
+		/* try to set the requested block size */
+		if (sb->s_blocksize != blocksize &&
+		    !sb_set_blocksize(sb, blocksize)) {
+			if (!silent)
+				adfs_msg(sb, KERN_ERR,
+					 "error: unsupported blocksize");
+			return -EINVAL;
+		}
 
-	nzones    = asb->s_map_size;
-	zone_size = (8 << dr->log2secsize) - le16_to_cpu(dr->zone_spare);
-	map_addr  = (nzones >> 1) * zone_size -
-		     ((nzones > 1) ? ADFS_DR_SIZE_BITS : 0);
-	map_addr  = signed_asl(map_addr, asb->s_map2blk);
+		/* read the buffer */
+		bh = sb_bread(sb, offset >> sb->s_blocksize_bits);
+		if (!bh) {
+			adfs_msg(sb, KERN_ERR,
+				 "error: unable to read block %u, try %d",
+				 offset >> sb->s_blocksize_bits, try);
+			return -EIO;
+		}
+
+		/* validate it */
+		ret = validate(sb, bh, &dr);
+		if (ret) {
+			brelse(bh);
+			return ret;
+		}
 
-	asb->s_ids_per_zone = zone_size / (asb->s_idlen + 1);
+		/* does the block size match the filesystem block size? */
+		blocksize = 1 << dr->log2secsize;
+		if (sb->s_blocksize == blocksize) {
+			asb->s_map = adfs_read_map(sb, dr);
+			brelse(bh);
+			return PTR_ERR_OR_ZERO(asb->s_map);
+		}
 
-	dm = kmalloc_array(nzones, sizeof(*dm), GFP_KERNEL);
-	if (dm == NULL) {
-		adfs_error(sb, "not enough memory");
-		return ERR_PTR(-ENOMEM);
+		brelse(bh);
 	}
 
-	for (zone = 0; zone < nzones; zone++, map_addr++) {
-		dm[zone].dm_startbit = 0;
-		dm[zone].dm_endbit   = zone_size;
-		dm[zone].dm_startblk = zone * zone_size - ADFS_DR_SIZE_BITS;
-		dm[zone].dm_bh       = sb_bread(sb, map_addr);
+	return -EIO;
+}
 
-		if (!dm[zone].dm_bh) {
-			adfs_error(sb, "unable to read map");
-			goto error_free;
-		}
-	}
+static int adfs_validate_bblk(struct super_block *sb, struct buffer_head *bh,
+			      struct adfs_discrecord **drp)
+{
+	struct adfs_discrecord *dr;
+	unsigned char *b_data;
 
-	/* adjust the limits for the first and last map zones */
-	i = zone - 1;
-	dm[0].dm_startblk = 0;
-	dm[0].dm_startbit = ADFS_DR_SIZE_BITS;
-	dm[i].dm_endbit   = (adfs_disc_size(dr) >> dr->log2bpmb) +
-			    (ADFS_DR_SIZE_BITS - i * zone_size);
+	b_data = bh->b_data + (ADFS_DISCRECORD % sb->s_blocksize);
+	if (adfs_checkbblk(b_data))
+		return -EILSEQ;
 
-	if (adfs_checkmap(sb, dm))
-		return dm;
+	/* Do some sanity checks on the ADFS disc record */
+	dr = (struct adfs_discrecord *)(b_data + ADFS_DR_OFFSET);
+	if (adfs_checkdiscrecord(dr))
+		return -EILSEQ;
+
+	*drp = dr;
+	return 0;
+}
 
-	adfs_error(sb, "map corrupted");
+static int adfs_validate_dr0(struct super_block *sb, struct buffer_head *bh,
+			      struct adfs_discrecord **drp)
+{
+	struct adfs_discrecord *dr;
 
-error_free:
-	while (--zone >= 0)
-		brelse(dm[zone].dm_bh);
+	/* Do some sanity checks on the ADFS disc record */
+	dr = (struct adfs_discrecord *)(bh->b_data + 4);
+	if (adfs_checkdiscrecord(dr) || dr->nzones_high || dr->nzones != 1)
+		return -EILSEQ;
 
-	kfree(dm);
-	return ERR_PTR(-EIO);
+	*drp = dr;
+	return 0;
 }
 
 static int adfs_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct adfs_discrecord *dr;
-	struct buffer_head *bh;
 	struct object_info root_obj;
-	unsigned char *b_data;
-	unsigned int blocksize;
 	struct adfs_sb_info *asb;
 	struct inode *root;
 	int ret = -EINVAL;
@@ -391,7 +374,10 @@ static int adfs_fill_super(struct super_block *sb, void *data, int silent)
 	asb = kzalloc(sizeof(*asb), GFP_KERNEL);
 	if (!asb)
 		return -ENOMEM;
+
 	sb->s_fs_info = asb;
+	sb->s_magic = ADFS_SUPER_MAGIC;
+	sb->s_time_gran = 10000000;
 
 	/* set default options */
 	asb->s_uid = GLOBAL_ROOT_UID;
@@ -403,78 +389,21 @@ static int adfs_fill_super(struct super_block *sb, void *data, int silent)
 	if (parse_options(sb, asb, data))
 		goto error;
 
-	sb_set_blocksize(sb, BLOCK_SIZE);
-	if (!(bh = sb_bread(sb, ADFS_DISCRECORD / BLOCK_SIZE))) {
-		adfs_msg(sb, KERN_ERR, "error: unable to read superblock");
-		ret = -EIO;
-		goto error;
-	}
-
-	b_data = bh->b_data + (ADFS_DISCRECORD % BLOCK_SIZE);
-
-	if (adfs_checkbblk(b_data)) {
-		ret = -EINVAL;
-		goto error_badfs;
-	}
-
-	dr = (struct adfs_discrecord *)(b_data + ADFS_DR_OFFSET);
-
-	/*
-	 * Do some sanity checks on the ADFS disc record
-	 */
-	if (adfs_checkdiscrecord(dr)) {
-		ret = -EINVAL;
-		goto error_badfs;
-	}
-
-	blocksize = 1 << dr->log2secsize;
-	brelse(bh);
-
-	if (sb_set_blocksize(sb, blocksize)) {
-		bh = sb_bread(sb, ADFS_DISCRECORD / sb->s_blocksize);
-		if (!bh) {
-			adfs_msg(sb, KERN_ERR,
-				 "error: couldn't read superblock on 2nd try.");
-			ret = -EIO;
-			goto error;
-		}
-		b_data = bh->b_data + (ADFS_DISCRECORD % sb->s_blocksize);
-		if (adfs_checkbblk(b_data)) {
-			adfs_msg(sb, KERN_ERR,
-				 "error: disc record mismatch, very weird!");
-			ret = -EINVAL;
-			goto error_free_bh;
-		}
-		dr = (struct adfs_discrecord *)(b_data + ADFS_DR_OFFSET);
-	} else {
+	/* Try to probe the filesystem boot block */
+	ret = adfs_probe(sb, ADFS_DISCRECORD, 1, adfs_validate_bblk);
+	if (ret == -EILSEQ)
+		ret = adfs_probe(sb, 0, silent, adfs_validate_dr0);
+	if (ret == -EILSEQ) {
 		if (!silent)
 			adfs_msg(sb, KERN_ERR,
-				 "error: unsupported blocksize");
+				 "error: can't find an ADFS filesystem on dev %s.",
+				 sb->s_id);
 		ret = -EINVAL;
-		goto error;
 	}
+	if (ret)
+		goto error;
 
-	/*
-	 * blocksize on this device should now be set to the ADFS log2secsize
-	 */
-
-	sb->s_magic		= ADFS_SUPER_MAGIC;
-	asb->s_idlen		= dr->idlen;
-	asb->s_map_size		= dr->nzones | (dr->nzones_high << 8);
-	asb->s_map2blk		= dr->log2bpmb - dr->log2secsize;
-	asb->s_log2sharesize	= dr->log2sharesize;
-
-	asb->s_map = adfs_read_map(sb, dr);
-	if (IS_ERR(asb->s_map)) {
-		ret =  PTR_ERR(asb->s_map);
-		goto error_free_bh;
-	}
-
-	brelse(bh);
-
-	/*
-	 * set up enough so that we can read an inode
-	 */
+	/* set up enough so that we can read an inode */
 	sb->s_op = &adfs_sops;
 
 	dr = adfs_map_discrecord(asb->s_map);
@@ -511,23 +440,13 @@ static int adfs_fill_super(struct super_block *sb, void *data, int silent)
 	root = adfs_iget(sb, &root_obj);
 	sb->s_root = d_make_root(root);
 	if (!sb->s_root) {
-		int i;
-		for (i = 0; i < asb->s_map_size; i++)
-			brelse(asb->s_map[i].dm_bh);
-		kfree(asb->s_map);
+		adfs_free_map(sb);
 		adfs_error(sb, "get root inode failed\n");
 		ret = -EIO;
 		goto error;
 	}
 	return 0;
 
-error_badfs:
-	if (!silent)
-		adfs_msg(sb, KERN_ERR,
-			 "error: can't find an ADFS filesystem on dev %s.",
-			 sb->s_id);
-error_free_bh:
-	brelse(bh);
 error:
 	sb->s_fs_info = NULL;
 	kfree(asb);
diff --git a/fs/affs/Kconfig b/fs/affs/Kconfig
index 84c46b9025c5..eb9d0ab850cb 100644
--- a/fs/affs/Kconfig
+++ b/fs/affs/Kconfig
@@ -9,7 +9,7 @@ config AFFS_FS
 	  FFS partition on your hard drive.  Amiga floppies however cannot be
 	  read with this driver due to an incompatibility of the floppy
 	  controller used in an Amiga and the standard floppy controller in
-	  PCs and workstations. Read <file:Documentation/filesystems/affs.txt>
+	  PCs and workstations. Read <file:Documentation/filesystems/affs.rst>
 	  and <file:fs/affs/Changes>.
 
 	  With this driver you can also mount disk files used by Bernd
diff --git a/fs/affs/affs.h b/fs/affs/affs.h
index a92eb6ae2ae2..a755bef7c4c7 100644
--- a/fs/affs/affs.h
+++ b/fs/affs/affs.h
@@ -43,8 +43,8 @@ struct affs_ext_key {
  */
 struct affs_inode_info {
 	atomic_t i_opencnt;
-	struct semaphore i_link_lock;		/* Protects internal inode access. */
-	struct semaphore i_ext_lock;		/* Protects internal inode access. */
+	struct mutex i_link_lock;		/* Protects internal inode access. */
+	struct mutex i_ext_lock;		/* Protects internal inode access. */
 #define i_hash_lock i_ext_lock
 	u32	 i_blkcnt;			/* block count */
 	u32	 i_extcnt;			/* extended block count */
@@ -293,30 +293,30 @@ affs_adjust_bitmapchecksum(struct buffer_head *bh, u32 val)
 static inline void
 affs_lock_link(struct inode *inode)
 {
-	down(&AFFS_I(inode)->i_link_lock);
+	mutex_lock(&AFFS_I(inode)->i_link_lock);
 }
 static inline void
 affs_unlock_link(struct inode *inode)
 {
-	up(&AFFS_I(inode)->i_link_lock);
+	mutex_unlock(&AFFS_I(inode)->i_link_lock);
 }
 static inline void
 affs_lock_dir(struct inode *inode)
 {
-	down(&AFFS_I(inode)->i_hash_lock);
+	mutex_lock_nested(&AFFS_I(inode)->i_hash_lock, SINGLE_DEPTH_NESTING);
 }
 static inline void
 affs_unlock_dir(struct inode *inode)
 {
-	up(&AFFS_I(inode)->i_hash_lock);
+	mutex_unlock(&AFFS_I(inode)->i_hash_lock);
 }
 static inline void
 affs_lock_ext(struct inode *inode)
 {
-	down(&AFFS_I(inode)->i_ext_lock);
+	mutex_lock(&AFFS_I(inode)->i_ext_lock);
 }
 static inline void
 affs_unlock_ext(struct inode *inode)
 {
-	up(&AFFS_I(inode)->i_ext_lock);
+	mutex_unlock(&AFFS_I(inode)->i_ext_lock);
 }
diff --git a/fs/affs/super.c b/fs/affs/super.c
index cc463ae47c12..47107c6712a6 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -121,8 +121,8 @@ static void init_once(void *foo)
 {
 	struct affs_inode_info *ei = (struct affs_inode_info *) foo;
 
-	sema_init(&ei->i_link_lock, 1);
-	sema_init(&ei->i_ext_lock, 1);
+	mutex_init(&ei->i_link_lock);
+	mutex_init(&ei->i_ext_lock);
 	inode_init_once(&ei->vfs_inode);
 }
 
@@ -561,14 +561,9 @@ affs_remount(struct super_block *sb, int *flags, char *data)
 	int			 root_block;
 	unsigned long		 mount_flags;
 	int			 res = 0;
-	char			*new_opts;
 	char			 volume[32];
 	char			*prefix = NULL;
 
-	new_opts = kstrdup(data, GFP_KERNEL);
-	if (data && !new_opts)
-		return -ENOMEM;
-
 	pr_debug("%s(flags=0x%x,opts=\"%s\")\n", __func__, *flags, data);
 
 	sync_filesystem(sb);
@@ -579,7 +574,6 @@ affs_remount(struct super_block *sb, int *flags, char *data)
 			   &blocksize, &prefix, volume,
 			   &mount_flags)) {
 		kfree(prefix);
-		kfree(new_opts);
 		return -EINVAL;
 	}
 
diff --git a/fs/afs/Kconfig b/fs/afs/Kconfig
index 3fb1f559e317..1ad211d72b3b 100644
--- a/fs/afs/Kconfig
+++ b/fs/afs/Kconfig
@@ -8,7 +8,7 @@ config AFS_FS
 	  If you say Y here, you will get an experimental Andrew File System
 	  driver. It currently only supports unsecured read-only AFS access.
 
-	  See <file:Documentation/filesystems/afs.txt> for more information.
+	  See <file:Documentation/filesystems/afs.rst> for more information.
 
 	  If unsure, say N.
 
@@ -18,7 +18,7 @@ config AFS_DEBUG
 	help
 	  Say Y here to make runtime controllable debugging messages appear.
 
-	  See <file:Documentation/filesystems/afs.txt> for more information.
+	  See <file:Documentation/filesystems/afs.rst> for more information.
 
 	  If unsure, say N.
 
@@ -37,6 +37,6 @@ config AFS_DEBUG_CURSOR
 	  the dmesg log if the server rotation algorithm fails to successfully
 	  contact a server.
 
-	  See <file:Documentation/filesystems/afs.txt> for more information.
+	  See <file:Documentation/filesystems/afs.rst> for more information.
 
 	  If unsure, say N.
diff --git a/fs/afs/addr_list.c b/fs/afs/addr_list.c
index df415c05939e..de1ae0bead3b 100644
--- a/fs/afs/addr_list.c
+++ b/fs/afs/addr_list.c
@@ -19,7 +19,7 @@
 void afs_put_addrlist(struct afs_addr_list *alist)
 {
 	if (alist && refcount_dec_and_test(&alist->usage))
-		call_rcu(&alist->rcu, (rcu_callback_t)kfree);
+		kfree_rcu(alist, rcu);
 }
 
 /*
diff --git a/fs/afs/cell.c b/fs/afs/cell.c
index fd5133e26a38..78ba5f932287 100644
--- a/fs/afs/cell.c
+++ b/fs/afs/cell.c
@@ -134,8 +134,17 @@ static struct afs_cell *afs_alloc_cell(struct afs_net *net,
 		_leave(" = -ENAMETOOLONG");
 		return ERR_PTR(-ENAMETOOLONG);
 	}
-	if (namelen == 5 && memcmp(name, "@cell", 5) == 0)
+
+	/* Prohibit cell names that contain unprintable chars, '/' and '@' or
+	 * that begin with a dot.  This also precludes "@cell".
+	 */
+	if (name[0] == '.')
 		return ERR_PTR(-EINVAL);
+	for (i = 0; i < namelen; i++) {
+		char ch = name[i];
+		if (!isprint(ch) || ch == '/' || ch == '@')
+			return ERR_PTR(-EINVAL);
+	}
 
 	_enter("%*.*s,%s", namelen, namelen, name, addresses);
 
diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c
index b86195e4dc6c..380ad5ace7cf 100644
--- a/fs/afs/cmservice.c
+++ b/fs/afs/cmservice.c
@@ -169,7 +169,7 @@ static int afs_record_cm_probe(struct afs_call *call, struct afs_server *server)
 
 	spin_lock(&server->probe_lock);
 
-	if (!test_bit(AFS_SERVER_FL_HAVE_EPOCH, &server->flags)) {
+	if (!test_and_set_bit(AFS_SERVER_FL_HAVE_EPOCH, &server->flags)) {
 		server->cm_epoch = call->epoch;
 		server->probe.cm_epoch = call->epoch;
 		goto out;
@@ -244,6 +244,17 @@ static void afs_cm_destructor(struct afs_call *call)
 }
 
 /*
+ * Abort a service call from within an action function.
+ */
+static void afs_abort_service_call(struct afs_call *call, u32 abort_code, int error,
+				   const char *why)
+{
+	rxrpc_kernel_abort_call(call->net->socket, call->rxcall,
+				abort_code, error, why);
+	afs_set_call_complete(call, error, 0);
+}
+
+/*
  * The server supplied a list of callbacks that it wanted to break.
  */
 static void SRXAFSCB_CallBack(struct work_struct *work)
@@ -342,14 +353,14 @@ static int afs_deliver_cb_callback(struct afs_call *call)
 		if (call->count2 != call->count && call->count2 != 0)
 			return afs_protocol_error(call, -EBADMSG,
 						  afs_eproto_cb_count);
-		call->_iter = &call->iter;
-		iov_iter_discard(&call->iter, READ, call->count2 * 3 * 4);
+		call->iter = &call->def_iter;
+		iov_iter_discard(&call->def_iter, READ, call->count2 * 3 * 4);
 		call->unmarshall++;
 
 		/* Fall through */
 	case 4:
 		_debug("extract discard %zu/%u",
-		       iov_iter_count(&call->iter), call->count2 * 3 * 4);
+		       iov_iter_count(call->iter), call->count2 * 3 * 4);
 
 		ret = afs_extract_data(call, false);
 		if (ret < 0)
@@ -510,8 +521,7 @@ static void SRXAFSCB_ProbeUuid(struct work_struct *work)
 	if (memcmp(r, &call->net->uuid, sizeof(call->net->uuid)) == 0)
 		afs_send_empty_reply(call);
 	else
-		rxrpc_kernel_abort_call(call->net->socket, call->rxcall,
-					1, 1, "K-1");
+		afs_abort_service_call(call, 1, 1, "K-1");
 
 	afs_put_call(call);
 	_leave("");
diff --git a/fs/afs/dir.c b/fs/afs/dir.c
index 497f979018c2..d1e1caa23c8b 100644
--- a/fs/afs/dir.c
+++ b/fs/afs/dir.c
@@ -908,6 +908,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry,
 				 unsigned int flags)
 {
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
+	struct afs_fid fid = {};
 	struct inode *inode;
 	struct dentry *d;
 	struct key *key;
@@ -951,21 +952,18 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry,
 	afs_stat_v(dvnode, n_lookup);
 	inode = afs_do_lookup(dir, dentry, key);
 	key_put(key);
-	if (inode == ERR_PTR(-ENOENT)) {
+	if (inode == ERR_PTR(-ENOENT))
 		inode = afs_try_auto_mntpt(dentry, dir);
-	} else {
-		dentry->d_fsdata =
-			(void *)(unsigned long)dvnode->status.data_version;
-	}
+
+	if (!IS_ERR_OR_NULL(inode))
+		fid = AFS_FS_I(inode)->fid;
+
 	d = d_splice_alias(inode, dentry);
 	if (!IS_ERR_OR_NULL(d)) {
 		d->d_fsdata = dentry->d_fsdata;
-		trace_afs_lookup(dvnode, &d->d_name,
-				 inode ? AFS_FS_I(inode) : NULL);
+		trace_afs_lookup(dvnode, &d->d_name, &fid);
 	} else {
-		trace_afs_lookup(dvnode, &dentry->d_name,
-				 IS_ERR_OR_NULL(inode) ? NULL
-				 : AFS_FS_I(inode));
+		trace_afs_lookup(dvnode, &dentry->d_name, &fid);
 	}
 	return d;
 }
@@ -1034,7 +1032,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
 	struct dentry *parent;
 	struct inode *inode;
 	struct key *key;
-	afs_dataversion_t dir_version;
+	afs_dataversion_t dir_version, invalid_before;
 	long de_version;
 	int ret;
 
@@ -1086,8 +1084,8 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
 	if (de_version == (long)dir_version)
 		goto out_valid_noupdate;
 
-	dir_version = dir->invalid_before;
-	if (de_version - (long)dir_version >= 0)
+	invalid_before = dir->invalid_before;
+	if (de_version - (long)invalid_before >= 0)
 		goto out_valid;
 
 	_debug("dir modified");
@@ -1277,6 +1275,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	struct afs_fs_cursor fc;
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	mode |= S_IFDIR;
@@ -1297,7 +1296,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1318,10 +1317,14 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 		goto error_key;
 	}
 
-	if (ret == 0 &&
-	    test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
-		afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
-				 afs_edit_dir_for_create);
+	if (ret == 0) {
+		down_write(&dvnode->validate_lock);
+		if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+		    dvnode->status.data_version == data_version)
+			afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
+					 afs_edit_dir_for_create);
+		up_write(&dvnode->validate_lock);
+	}
 
 	key_put(key);
 	kfree(scb);
@@ -1362,6 +1365,7 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
 	struct afs_fs_cursor fc;
 	struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode = NULL;
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	_enter("{%llx:%llu},{%pd}",
@@ -1393,7 +1397,7 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1406,9 +1410,12 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
 		ret = afs_end_vnode_operation(&fc);
 		if (ret == 0) {
 			afs_dir_remove_subdir(dentry);
-			if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+			down_write(&dvnode->validate_lock);
+			if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+			    dvnode->status.data_version == data_version)
 				afs_edit_dir_remove(dvnode, &dentry->d_name,
 						    afs_edit_dir_for_rmdir);
+			up_write(&dvnode->validate_lock);
 		}
 	}
 
@@ -1546,10 +1553,15 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry)
 		ret = afs_end_vnode_operation(&fc);
 		if (ret == 0 && !(scb[1].have_status || scb[1].have_error))
 			ret = afs_dir_remove_link(dvnode, dentry, key);
-		if (ret == 0 &&
-		    test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
-			afs_edit_dir_remove(dvnode, &dentry->d_name,
-					    afs_edit_dir_for_unlink);
+
+		if (ret == 0) {
+			down_write(&dvnode->validate_lock);
+			if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+			    dvnode->status.data_version == data_version)
+				afs_edit_dir_remove(dvnode, &dentry->d_name,
+						    afs_edit_dir_for_unlink);
+			up_write(&dvnode->validate_lock);
+		}
 	}
 
 	if (need_rehash && ret < 0 && ret != -ENOENT)
@@ -1575,6 +1587,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	struct afs_status_cb *scb;
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	mode |= S_IFREG;
@@ -1599,7 +1612,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1620,9 +1633,12 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 		goto error_key;
 	}
 
-	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+	down_write(&dvnode->validate_lock);
+	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+	    dvnode->status.data_version == data_version)
 		afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
 				 afs_edit_dir_for_create);
+	up_write(&dvnode->validate_lock);
 
 	kfree(scb);
 	key_put(key);
@@ -1650,6 +1666,7 @@ static int afs_link(struct dentry *from, struct inode *dir,
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
 	struct afs_vnode *vnode = AFS_FS_I(d_inode(from));
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	_enter("{%llx:%llu},{%llx:%llu},{%pd}",
@@ -1674,7 +1691,7 @@ static int afs_link(struct dentry *from, struct inode *dir,
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		if (mutex_lock_interruptible_nested(&vnode->io_lock, 1) < 0) {
 			afs_end_vnode_operation(&fc);
@@ -1704,9 +1721,12 @@ static int afs_link(struct dentry *from, struct inode *dir,
 		goto error_key;
 	}
 
-	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+	down_write(&dvnode->validate_lock);
+	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+	    dvnode->status.data_version == data_version)
 		afs_edit_dir_add(dvnode, &dentry->d_name, &vnode->fid,
 				 afs_edit_dir_for_link);
+	up_write(&dvnode->validate_lock);
 
 	key_put(key);
 	kfree(scb);
@@ -1734,6 +1754,7 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
 	struct afs_status_cb *scb;
 	struct afs_vnode *dvnode = AFS_FS_I(dir);
 	struct key *key;
+	afs_dataversion_t data_version;
 	int ret;
 
 	_enter("{%llx:%llu},{%pd},%s",
@@ -1761,7 +1782,7 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t data_version = dvnode->status.data_version + 1;
+		data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1782,9 +1803,12 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
 		goto error_key;
 	}
 
-	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+	down_write(&dvnode->validate_lock);
+	if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+	    dvnode->status.data_version == data_version)
 		afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
 				 afs_edit_dir_for_symlink);
+	up_write(&dvnode->validate_lock);
 
 	key_put(key);
 	kfree(scb);
@@ -1814,6 +1838,8 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct dentry *tmp = NULL, *rehash = NULL;
 	struct inode *new_inode;
 	struct key *key;
+	afs_dataversion_t orig_data_version;
+	afs_dataversion_t new_data_version;
 	bool new_negative = d_is_negative(new_dentry);
 	int ret;
 
@@ -1892,10 +1918,6 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	ret = -ERESTARTSYS;
 	if (afs_begin_vnode_operation(&fc, orig_dvnode, key, true)) {
-		afs_dataversion_t orig_data_version;
-		afs_dataversion_t new_data_version;
-		struct afs_status_cb *new_scb = &scb[1];
-
 		orig_data_version = orig_dvnode->status.data_version + 1;
 
 		if (orig_dvnode != new_dvnode) {
@@ -1906,7 +1928,6 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			new_data_version = new_dvnode->status.data_version + 1;
 		} else {
 			new_data_version = orig_data_version;
-			new_scb = &scb[0];
 		}
 
 		while (afs_select_fileserver(&fc)) {
@@ -1914,7 +1935,7 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			fc.cb_break_2 = afs_calc_vnode_cb_break(new_dvnode);
 			afs_fs_rename(&fc, old_dentry->d_name.name,
 				      new_dvnode, new_dentry->d_name.name,
-				      &scb[0], new_scb);
+				      &scb[0], &scb[1]);
 		}
 
 		afs_vnode_commit_status(&fc, orig_dvnode, fc.cb_break,
@@ -1932,18 +1953,25 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (ret == 0) {
 		if (rehash)
 			d_rehash(rehash);
-		if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags))
-		    afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name,
-					afs_edit_dir_for_rename_0);
+		down_write(&orig_dvnode->validate_lock);
+		if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) &&
+		    orig_dvnode->status.data_version == orig_data_version)
+			afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name,
+					    afs_edit_dir_for_rename_0);
+		if (orig_dvnode != new_dvnode) {
+			up_write(&orig_dvnode->validate_lock);
 
-		if (!new_negative &&
-		    test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags))
-			afs_edit_dir_remove(new_dvnode, &new_dentry->d_name,
-					    afs_edit_dir_for_rename_1);
+			down_write(&new_dvnode->validate_lock);
+		}
+		if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags) &&
+		    orig_dvnode->status.data_version == new_data_version) {
+			if (!new_negative)
+				afs_edit_dir_remove(new_dvnode, &new_dentry->d_name,
+						    afs_edit_dir_for_rename_1);
 
-		if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags))
 			afs_edit_dir_add(new_dvnode, &new_dentry->d_name,
 					 &vnode->fid, afs_edit_dir_for_rename_2);
+		}
 
 		new_inode = d_inode(new_dentry);
 		if (new_inode) {
@@ -1959,14 +1987,10 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		 * Note that if we ever implement RENAME_EXCHANGE, we'll have
 		 * to update both dentries with opposing dir versions.
 		 */
-		if (new_dvnode != orig_dvnode) {
-			afs_update_dentry_version(&fc, old_dentry, &scb[1]);
-			afs_update_dentry_version(&fc, new_dentry, &scb[1]);
-		} else {
-			afs_update_dentry_version(&fc, old_dentry, &scb[0]);
-			afs_update_dentry_version(&fc, new_dentry, &scb[0]);
-		}
+		afs_update_dentry_version(&fc, old_dentry, &scb[1]);
+		afs_update_dentry_version(&fc, new_dentry, &scb[1]);
 		d_move(old_dentry, new_dentry);
+		up_write(&new_dvnode->validate_lock);
 		goto error_tmp;
 	}
 
diff --git a/fs/afs/dir_edit.c b/fs/afs/dir_edit.c
index d4fbe5f85f1b..b108528bf010 100644
--- a/fs/afs/dir_edit.c
+++ b/fs/afs/dir_edit.c
@@ -68,13 +68,11 @@ static int afs_find_contig_bits(union afs_xdr_dir_block *block, unsigned int nr_
 static void afs_set_contig_bits(union afs_xdr_dir_block *block,
 				int bit, unsigned int nr_slots)
 {
-	u64 mask, before, after;
+	u64 mask;
 
 	mask = (1 << nr_slots) - 1;
 	mask <<= bit;
 
-	before = *(u64 *)block->hdr.bitmap;
-
 	block->hdr.bitmap[0] |= (u8)(mask >> 0 * 8);
 	block->hdr.bitmap[1] |= (u8)(mask >> 1 * 8);
 	block->hdr.bitmap[2] |= (u8)(mask >> 2 * 8);
@@ -83,8 +81,6 @@ static void afs_set_contig_bits(union afs_xdr_dir_block *block,
 	block->hdr.bitmap[5] |= (u8)(mask >> 5 * 8);
 	block->hdr.bitmap[6] |= (u8)(mask >> 6 * 8);
 	block->hdr.bitmap[7] |= (u8)(mask >> 7 * 8);
-
-	after = *(u64 *)block->hdr.bitmap;
 }
 
 /*
@@ -93,13 +89,11 @@ static void afs_set_contig_bits(union afs_xdr_dir_block *block,
 static void afs_clear_contig_bits(union afs_xdr_dir_block *block,
 				  int bit, unsigned int nr_slots)
 {
-	u64 mask, before, after;
+	u64 mask;
 
 	mask = (1 << nr_slots) - 1;
 	mask <<= bit;
 
-	before = *(u64 *)block->hdr.bitmap;
-
 	block->hdr.bitmap[0] &= ~(u8)(mask >> 0 * 8);
 	block->hdr.bitmap[1] &= ~(u8)(mask >> 1 * 8);
 	block->hdr.bitmap[2] &= ~(u8)(mask >> 2 * 8);
@@ -108,8 +102,6 @@ static void afs_clear_contig_bits(union afs_xdr_dir_block *block,
 	block->hdr.bitmap[5] &= ~(u8)(mask >> 5 * 8);
 	block->hdr.bitmap[6] &= ~(u8)(mask >> 6 * 8);
 	block->hdr.bitmap[7] &= ~(u8)(mask >> 7 * 8);
-
-	after = *(u64 *)block->hdr.bitmap;
 }
 
 /*
diff --git a/fs/afs/dir_silly.c b/fs/afs/dir_silly.c
index 361088a5edb9..d94e2b7cddff 100644
--- a/fs/afs/dir_silly.c
+++ b/fs/afs/dir_silly.c
@@ -21,6 +21,7 @@ static int afs_do_silly_rename(struct afs_vnode *dvnode, struct afs_vnode *vnode
 {
 	struct afs_fs_cursor fc;
 	struct afs_status_cb *scb;
+	afs_dataversion_t dir_data_version;
 	int ret = -ERESTARTSYS;
 
 	_enter("%pd,%pd", old, new);
@@ -31,7 +32,7 @@ static int afs_do_silly_rename(struct afs_vnode *dvnode, struct afs_vnode *vnode
 
 	trace_afs_silly_rename(vnode, false);
 	if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
-		afs_dataversion_t dir_data_version = dvnode->status.data_version + 1;
+		dir_data_version = dvnode->status.data_version + 1;
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -54,12 +55,15 @@ static int afs_do_silly_rename(struct afs_vnode *dvnode, struct afs_vnode *vnode
 			dvnode->silly_key = key_get(key);
 		}
 
-		if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+		down_write(&dvnode->validate_lock);
+		if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+		    dvnode->status.data_version == dir_data_version) {
 			afs_edit_dir_remove(dvnode, &old->d_name,
 					    afs_edit_dir_for_silly_0);
-		if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
 			afs_edit_dir_add(dvnode, &new->d_name,
 					 &vnode->fid, afs_edit_dir_for_silly_1);
+		}
+		up_write(&dvnode->validate_lock);
 	}
 
 	kfree(scb);
@@ -181,10 +185,14 @@ static int afs_do_silly_unlink(struct afs_vnode *dvnode, struct afs_vnode *vnode
 				clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
 			}
 		}
-		if (ret == 0 &&
-		    test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
-			afs_edit_dir_remove(dvnode, &dentry->d_name,
-					    afs_edit_dir_for_unlink);
+		if (ret == 0) {
+			down_write(&dvnode->validate_lock);
+			if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+			    dvnode->status.data_version == dir_data_version)
+				afs_edit_dir_remove(dvnode, &dentry->d_name,
+						    afs_edit_dir_for_unlink);
+			up_write(&dvnode->validate_lock);
+		}
 	}
 
 	kfree(scb);
diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c
index 4150280509ff..7503899c0a1b 100644
--- a/fs/afs/dynroot.c
+++ b/fs/afs/dynroot.c
@@ -136,6 +136,9 @@ static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentr
 
 	ASSERTCMP(d_inode(dentry), ==, NULL);
 
+	if (flags & LOOKUP_CREATE)
+		return ERR_PTR(-EOPNOTSUPP);
+
 	if (dentry->d_name.len >= AFSNAMEMAX) {
 		_leave(" = -ENAMETOOLONG");
 		return ERR_PTR(-ENAMETOOLONG);
diff --git a/fs/afs/file.c b/fs/afs/file.c
index dd3c55c9101c..8415733f7bc1 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -223,7 +223,7 @@ static void afs_file_readpage_read_complete(struct page *page,
 /*
  * Fetch file data from the volume.
  */
-int afs_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_read *desc)
+int afs_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_read *req)
 {
 	struct afs_fs_cursor fc;
 	struct afs_status_cb *scb;
@@ -246,7 +246,7 @@ int afs_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_read *de
 
 		while (afs_select_fileserver(&fc)) {
 			fc.cb_break = afs_calc_vnode_cb_break(vnode);
-			afs_fs_fetch_data(&fc, scb, desc);
+			afs_fs_fetch_data(&fc, scb, req);
 		}
 
 		afs_check_for_remote_deletion(&fc, vnode);
@@ -257,7 +257,7 @@ int afs_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_read *de
 
 	if (ret == 0) {
 		afs_stat_v(vnode, n_fetches);
-		atomic_long_add(desc->actual_len,
+		atomic_long_add(req->actual_len,
 				&afs_v2net(vnode)->n_fetch_bytes);
 	}
 
diff --git a/fs/afs/flock.c b/fs/afs/flock.c
index d5e5a6ddc847..0f2a94ba73cb 100644
--- a/fs/afs/flock.c
+++ b/fs/afs/flock.c
@@ -346,8 +346,8 @@ again:
 		if (ret < 0) {
 			trace_afs_flock_ev(vnode, NULL, afs_flock_extend_fail,
 					   ret);
-			pr_warning("AFS: Failed to extend lock on {%llx:%llx} error %d\n",
-				   vnode->fid.vid, vnode->fid.vnode, ret);
+			pr_warn("AFS: Failed to extend lock on {%llx:%llx} error %d\n",
+				vnode->fid.vid, vnode->fid.vnode, ret);
 		}
 
 		spin_lock(&vnode->lock);
diff --git a/fs/afs/fs_probe.c b/fs/afs/fs_probe.c
index cfe62b154f68..37d1bba57b00 100644
--- a/fs/afs/fs_probe.c
+++ b/fs/afs/fs_probe.c
@@ -32,9 +32,8 @@ void afs_fileserver_probe_result(struct afs_call *call)
 	struct afs_server *server = call->server;
 	unsigned int server_index = call->server_index;
 	unsigned int index = call->addr_ix;
-	unsigned int rtt = UINT_MAX;
+	unsigned int rtt_us = 0;
 	bool have_result = false;
-	u64 _rtt;
 	int ret = call->error;
 
 	_enter("%pU,%u", &server->uuid, index);
@@ -93,15 +92,9 @@ responded:
 		}
 	}
 
-	/* Get the RTT and scale it to fit into a 32-bit value that represents
-	 * over a minute of time so that we can access it with one instruction
-	 * on a 32-bit system.
-	 */
-	_rtt = rxrpc_kernel_get_rtt(call->net->socket, call->rxcall);
-	_rtt /= 64;
-	rtt = (_rtt > UINT_MAX) ? UINT_MAX : _rtt;
-	if (rtt < server->probe.rtt) {
-		server->probe.rtt = rtt;
+	rtt_us = rxrpc_kernel_get_srtt(call->net->socket, call->rxcall);
+	if (rtt_us < server->probe.rtt) {
+		server->probe.rtt = rtt_us;
 		alist->preferred = index;
 		have_result = true;
 	}
@@ -113,15 +106,11 @@ out:
 	spin_unlock(&server->probe_lock);
 
 	_debug("probe [%u][%u] %pISpc rtt=%u ret=%d",
-	       server_index, index, &alist->addrs[index].transport,
-	       (unsigned int)rtt, ret);
+	       server_index, index, &alist->addrs[index].transport, rtt_us, ret);
 
 	have_result |= afs_fs_probe_done(server);
-	if (have_result) {
-		server->probe.have_result = true;
-		wake_up_var(&server->probe.have_result);
+	if (have_result)
 		wake_up_all(&server->probe_wq);
-	}
 }
 
 /*
@@ -145,6 +134,7 @@ static int afs_do_probe_fileserver(struct afs_net *net,
 	read_lock(&server->fs_lock);
 	ac.alist = rcu_dereference_protected(server->addresses,
 					     lockdep_is_held(&server->fs_lock));
+	afs_get_addrlist(ac.alist);
 	read_unlock(&server->fs_lock);
 
 	atomic_set(&server->probe_outstanding, ac.alist->nr_addrs);
@@ -163,6 +153,7 @@ static int afs_do_probe_fileserver(struct afs_net *net,
 
 	if (!in_progress)
 		afs_fs_probe_done(server);
+	afs_put_addrlist(ac.alist);
 	return in_progress;
 }
 
diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c
index 6f84231f11a5..d2b3798c1932 100644
--- a/fs/afs/fsclient.c
+++ b/fs/afs/fsclient.c
@@ -65,6 +65,7 @@ static int xdr_decode_AFSFetchStatus(const __be32 **_bp,
 	bool inline_error = (call->operation_ID == afs_FS_InlineBulkStatus);
 	u64 data_version, size;
 	u32 type, abort_code;
+	int ret;
 
 	abort_code = ntohl(xdr->abort_code);
 
@@ -78,7 +79,7 @@ static int xdr_decode_AFSFetchStatus(const __be32 **_bp,
 			 */
 			status->abort_code = abort_code;
 			scb->have_error = true;
-			return 0;
+			goto good;
 		}
 
 		pr_warn("Unknown AFSFetchStatus version %u\n", ntohl(xdr->if_version));
@@ -87,7 +88,8 @@ static int xdr_decode_AFSFetchStatus(const __be32 **_bp,
 
 	if (abort_code != 0 && inline_error) {
 		status->abort_code = abort_code;
-		return 0;
+		scb->have_error = true;
+		goto good;
 	}
 
 	type = ntohl(xdr->type);
@@ -123,13 +125,16 @@ static int xdr_decode_AFSFetchStatus(const __be32 **_bp,
 	data_version |= (u64)ntohl(xdr->data_version_hi) << 32;
 	status->data_version = data_version;
 	scb->have_status = true;
-
+good:
+	ret = 0;
+advance:
 	*_bp = (const void *)*_bp + sizeof(*xdr);
-	return 0;
+	return ret;
 
 bad:
 	xdr_dump_bad(*_bp);
-	return afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status);
+	ret = afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status);
+	goto advance;
 }
 
 static time64_t xdr_decode_expiry(struct afs_call *call, u32 expiry)
@@ -323,7 +328,7 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call)
 	int ret;
 
 	_enter("{%u,%zu/%llu}",
-	       call->unmarshall, iov_iter_count(&call->iter), req->actual_len);
+	       call->unmarshall, iov_iter_count(call->iter), req->actual_len);
 
 	switch (call->unmarshall) {
 	case 0:
@@ -363,14 +368,14 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call)
 		call->bvec[0].bv_len = size;
 		call->bvec[0].bv_offset = req->offset;
 		call->bvec[0].bv_page = req->pages[req->index];
-		iov_iter_bvec(&call->iter, READ, call->bvec, 1, size);
+		iov_iter_bvec(&call->def_iter, READ, call->bvec, 1, size);
 		ASSERTCMP(size, <=, PAGE_SIZE);
 		/* Fall through */
 
 		/* extract the returned data */
 	case 2:
 		_debug("extract data %zu/%llu",
-		       iov_iter_count(&call->iter), req->remain);
+		       iov_iter_count(call->iter), req->remain);
 
 		ret = afs_extract_data(call, true);
 		if (ret < 0)
@@ -380,8 +385,6 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call)
 		ASSERTCMP(req->offset, <=, PAGE_SIZE);
 		if (req->offset == PAGE_SIZE) {
 			req->offset = 0;
-			if (req->page_done)
-				req->page_done(req);
 			req->index++;
 			if (req->remain > 0)
 				goto begin_page;
@@ -398,7 +401,7 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call)
 
 	case 3:
 		_debug("extract discard %zu/%llu",
-		       iov_iter_count(&call->iter), req->actual_len - req->len);
+		       iov_iter_count(call->iter), req->actual_len - req->len);
 
 		ret = afs_extract_data(call, true);
 		if (ret < 0)
@@ -435,11 +438,13 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call)
 		if (req->offset < PAGE_SIZE)
 			zero_user_segment(req->pages[req->index],
 					  req->offset, PAGE_SIZE);
-		if (req->page_done)
-			req->page_done(req);
 		req->offset = 0;
 	}
 
+	if (req->page_done)
+		for (req->index = 0; req->index < req->nr_pages; req->index++)
+			req->page_done(req);
+
 	_leave(" = 0 [done]");
 	return 0;
 }
@@ -490,7 +495,7 @@ static int afs_fs_fetch_data64(struct afs_fs_cursor *fc,
 	call->key = fc->key;
 	call->out_scb = scb;
 	call->out_volsync = NULL;
-	call->read_request = req;
+	call->read_request = afs_get_read(req);
 
 	/* marshall the parameters */
 	bp = call->request;
@@ -503,7 +508,6 @@ static int afs_fs_fetch_data64(struct afs_fs_cursor *fc,
 	bp[6] = 0;
 	bp[7] = htonl(lower_32_bits(req->len));
 
-	refcount_inc(&req->usage);
 	afs_use_fs_server(call, fc->cbi);
 	trace_afs_make_fs_call(call, &vnode->fid);
 	afs_set_fc_call(call, fc);
@@ -540,7 +544,7 @@ int afs_fs_fetch_data(struct afs_fs_cursor *fc,
 	call->key = fc->key;
 	call->out_scb = scb;
 	call->out_volsync = NULL;
-	call->read_request = req;
+	call->read_request = afs_get_read(req);
 
 	/* marshall the parameters */
 	bp = call->request;
@@ -551,7 +555,6 @@ int afs_fs_fetch_data(struct afs_fs_cursor *fc,
 	bp[4] = htonl(lower_32_bits(req->pos));
 	bp[5] = htonl(lower_32_bits(req->len));
 
-	refcount_inc(&req->usage);
 	afs_use_fs_server(call, fc->cbi);
 	trace_afs_make_fs_call(call, &vnode->fid);
 	afs_set_fc_call(call, fc);
@@ -983,16 +986,16 @@ static int afs_deliver_fs_rename(struct afs_call *call)
 	if (ret < 0)
 		return ret;
 
-	/* unmarshall the reply once we've received all of it */
+	/* If the two dirs are the same, we have two copies of the same status
+	 * report, so we just decode it twice.
+	 */
 	bp = call->buffer;
 	ret = xdr_decode_AFSFetchStatus(&bp, call, call->out_dir_scb);
 	if (ret < 0)
 		return ret;
-	if (call->out_dir_scb != call->out_scb) {
-		ret = xdr_decode_AFSFetchStatus(&bp, call, call->out_scb);
-		if (ret < 0)
-			return ret;
-	}
+	ret = xdr_decode_AFSFetchStatus(&bp, call, call->out_scb);
+	if (ret < 0)
+		return ret;
 	xdr_decode_AFSVolSync(&bp, call->out_volsync);
 
 	_leave(" = 0 [done]");
@@ -1852,7 +1855,7 @@ static int afs_deliver_fs_get_capabilities(struct afs_call *call)
 	u32 count;
 	int ret;
 
-	_enter("{%u,%zu}", call->unmarshall, iov_iter_count(&call->iter));
+	_enter("{%u,%zu}", call->unmarshall, iov_iter_count(call->iter));
 
 	switch (call->unmarshall) {
 	case 0:
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index 46d2d7cb461d..281470fe1183 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -34,8 +34,7 @@ static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *paren
 {
 	static unsigned long once_only;
 
-	pr_warn("kAFS: AFS vnode with undefined type %u\n",
-		vnode->status.type);
+	pr_warn("kAFS: AFS vnode with undefined type %u\n", vnode->status.type);
 	pr_warn("kAFS: A=%d m=%o s=%llx v=%llx\n",
 		vnode->status.abort_code,
 		vnode->status.mode,
@@ -175,11 +174,11 @@ static void afs_apply_status(struct afs_fs_cursor *fc,
 	BUG_ON(test_bit(AFS_VNODE_UNSET, &vnode->flags));
 
 	if (status->type != vnode->status.type) {
-		pr_warning("Vnode %llx:%llx:%x changed type %u to %u\n",
-			   vnode->fid.vid,
-			   vnode->fid.vnode,
-			   vnode->fid.unique,
-			   status->type, vnode->status.type);
+		pr_warn("Vnode %llx:%llx:%x changed type %u to %u\n",
+			vnode->fid.vid,
+			vnode->fid.vnode,
+			vnode->fid.unique,
+			status->type, vnode->status.type);
 		afs_protocol_error(NULL, -EBADMSG, afs_eproto_bad_status);
 		return;
 	}
diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index 759e0578012c..80255513e230 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -81,7 +81,7 @@ enum afs_call_state {
  * List of server addresses.
  */
 struct afs_addr_list {
-	struct rcu_head		rcu;		/* Must be first */
+	struct rcu_head		rcu;
 	refcount_t		usage;
 	u32			version;	/* Version */
 	unsigned char		max_addrs;
@@ -115,9 +115,9 @@ struct afs_call {
 	struct afs_vnode	*lvnode;	/* vnode being locked */
 	void			*request;	/* request data (first part) */
 	struct address_space	*mapping;	/* Pages being written from */
-	struct iov_iter		iter;		/* Buffer iterator */
-	struct iov_iter		*_iter;		/* Iterator currently in use */
-	union {	/* Convenience for ->iter */
+	struct iov_iter		def_iter;	/* Default buffer/data iterator */
+	struct iov_iter		*iter;		/* Iterator currently in use */
+	union {	/* Convenience for ->def_iter */
 		struct kvec	kvec[1];
 		struct bio_vec	bvec[1];
 	};
@@ -154,7 +154,7 @@ struct afs_call {
 	};
 	unsigned char		unmarshall;	/* unmarshalling phase */
 	unsigned char		addr_ix;	/* Address in ->alist */
-	bool			incoming;	/* T if incoming call */
+	bool			drop_ref;	/* T if need to drop ref for incoming call */
 	bool			send_pages;	/* T if data from mapping should be sent */
 	bool			need_attention;	/* T if RxRPC poked us */
 	bool			async;		/* T if asynchronous */
@@ -533,12 +533,10 @@ struct afs_server {
 		u32		abort_code;
 		u32		cm_epoch;
 		short		error;
-		bool		have_result;
 		bool		responded:1;
 		bool		is_yfs:1;
 		bool		not_yfs:1;
 		bool		local_failure:1;
-		bool		no_epoch:1;
 		bool		cm_probed:1;
 		bool		said_rebooted:1;
 		bool		said_inconsistent:1;
@@ -934,6 +932,12 @@ extern int afs_fetch_data(struct afs_vnode *, struct key *, struct afs_read *);
 extern int afs_page_filler(void *, struct page *);
 extern void afs_put_read(struct afs_read *);
 
+static inline struct afs_read *afs_get_read(struct afs_read *req)
+{
+	refcount_inc(&req->usage);
+	return req;
+}
+
 /*
  * flock.c
  */
@@ -1136,7 +1140,7 @@ static inline void afs_extract_begin(struct afs_call *call, void *buf, size_t si
 {
 	call->kvec[0].iov_base = buf;
 	call->kvec[0].iov_len = size;
-	iov_iter_kvec(&call->iter, READ, call->kvec, 1, size);
+	iov_iter_kvec(&call->def_iter, READ, call->kvec, 1, size);
 }
 
 static inline void afs_extract_to_tmp(struct afs_call *call)
@@ -1151,7 +1155,7 @@ static inline void afs_extract_to_tmp64(struct afs_call *call)
 
 static inline void afs_extract_discard(struct afs_call *call, size_t size)
 {
-	iov_iter_discard(&call->iter, READ, size);
+	iov_iter_discard(&call->def_iter, READ, size);
 }
 
 static inline void afs_extract_to_buf(struct afs_call *call, size_t size)
@@ -1203,8 +1207,16 @@ static inline void afs_set_call_complete(struct afs_call *call,
 		ok = true;
 	}
 	spin_unlock_bh(&call->state_lock);
-	if (ok)
+	if (ok) {
 		trace_afs_call_done(call);
+
+		/* Asynchronous calls have two refs to release - one from the alloc and
+		 * one queued with the work item - and we can't just deallocate the
+		 * call because the work item may be queued again.
+		 */
+		if (call->drop_ref)
+			afs_put_call(call);
+	}
 }
 
 /*
@@ -1321,7 +1333,7 @@ extern struct afs_volume *afs_create_volume(struct afs_fs_context *);
 extern void afs_activate_volume(struct afs_volume *);
 extern void afs_deactivate_volume(struct afs_volume *);
 extern void afs_put_volume(struct afs_cell *, struct afs_volume *);
-extern int afs_check_volume_status(struct afs_volume *, struct key *);
+extern int afs_check_volume_status(struct afs_volume *, struct afs_fs_cursor *);
 
 /*
  * write.c
diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c
index f532d6d3bd28..79bc5f1338ed 100644
--- a/fs/afs/mntpt.c
+++ b/fs/afs/mntpt.c
@@ -126,7 +126,7 @@ static int afs_mntpt_set_params(struct fs_context *fc, struct dentry *mntpt)
 		if (src_as->cell)
 			ctx->cell = afs_get_cell(src_as->cell);
 
-		if (size > PAGE_SIZE - 1)
+		if (size < 2 || size > PAGE_SIZE - 1)
 			return -EINVAL;
 
 		page = read_mapping_page(d_inode(mntpt)->i_mapping, 0, NULL);
@@ -140,7 +140,9 @@ static int afs_mntpt_set_params(struct fs_context *fc, struct dentry *mntpt)
 		}
 
 		buf = kmap(page);
-		ret = vfs_parse_fs_string(fc, "source", buf, size);
+		ret = -EINVAL;
+		if (buf[size - 1] == '.')
+			ret = vfs_parse_fs_string(fc, "source", buf, size - 1);
 		kunmap(page);
 		put_page(page);
 		if (ret < 0)
diff --git a/fs/afs/proc.c b/fs/afs/proc.c
index fba2ec3a3a9c..468e1713bce1 100644
--- a/fs/afs/proc.c
+++ b/fs/afs/proc.c
@@ -213,13 +213,14 @@ static int afs_proc_cell_volumes_show(struct seq_file *m, void *v)
 
 	/* Display header on line 1 */
 	if (v == &cell->proc_volumes) {
-		seq_puts(m, "USE VID      TY\n");
+		seq_puts(m, "USE VID      TY NAME\n");
 		return 0;
 	}
 
-	seq_printf(m, "%3d %08llx %s\n",
+	seq_printf(m, "%3d %08llx %s %s\n",
 		   atomic_read(&vol->usage), vol->vid,
-		   afs_vol_types[vol->type]);
+		   afs_vol_types[vol->type],
+		   vol->name);
 
 	return 0;
 }
diff --git a/fs/afs/rotate.c b/fs/afs/rotate.c
index 172ba569cd60..2a3305e42b14 100644
--- a/fs/afs/rotate.c
+++ b/fs/afs/rotate.c
@@ -192,7 +192,7 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc)
 			write_unlock(&vnode->volume->servers_lock);
 
 			set_bit(AFS_VOLUME_NEEDS_UPDATE, &vnode->volume->flags);
-			error = afs_check_volume_status(vnode->volume, fc->key);
+			error = afs_check_volume_status(vnode->volume, fc);
 			if (error < 0)
 				goto failed_set_error;
 
@@ -281,7 +281,7 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc)
 
 			set_bit(AFS_VOLUME_WAIT, &vnode->volume->flags);
 			set_bit(AFS_VOLUME_NEEDS_UPDATE, &vnode->volume->flags);
-			error = afs_check_volume_status(vnode->volume, fc->key);
+			error = afs_check_volume_status(vnode->volume, fc);
 			if (error < 0)
 				goto failed_set_error;
 
@@ -341,7 +341,7 @@ start:
 	/* See if we need to do an update of the volume record.  Note that the
 	 * volume may have moved or even have been deleted.
 	 */
-	error = afs_check_volume_status(vnode->volume, fc->key);
+	error = afs_check_volume_status(vnode->volume, fc);
 	if (error < 0)
 		goto failed_set_error;
 
diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c
index 61498d9f06ef..1ecc67da6c1a 100644
--- a/fs/afs/rxrpc.c
+++ b/fs/afs/rxrpc.c
@@ -18,7 +18,6 @@ struct workqueue_struct *afs_async_calls;
 
 static void afs_wake_up_call_waiter(struct sock *, struct rxrpc_call *, unsigned long);
 static void afs_wake_up_async_call(struct sock *, struct rxrpc_call *, unsigned long);
-static void afs_delete_async_call(struct work_struct *);
 static void afs_process_async_call(struct work_struct *);
 static void afs_rx_new_call(struct sock *, struct rxrpc_call *, unsigned long);
 static void afs_rx_discard_new_call(struct rxrpc_call *, unsigned long);
@@ -152,7 +151,7 @@ static struct afs_call *afs_alloc_call(struct afs_net *net,
 	INIT_WORK(&call->async_work, afs_process_async_call);
 	init_waitqueue_head(&call->waitq);
 	spin_lock_init(&call->state_lock);
-	call->_iter = &call->iter;
+	call->iter = &call->def_iter;
 
 	o = atomic_inc_return(&net->nr_outstanding_calls);
 	trace_afs_call(call, afs_call_trace_alloc, 1, o,
@@ -169,7 +168,7 @@ void afs_put_call(struct afs_call *call)
 	int n = atomic_dec_return(&call->usage);
 	int o = atomic_read(&net->nr_outstanding_calls);
 
-	trace_afs_call(call, afs_call_trace_put, n + 1, o,
+	trace_afs_call(call, afs_call_trace_put, n, o,
 		       __builtin_return_address(0));
 
 	ASSERTCMP(n, >=, 0);
@@ -402,8 +401,10 @@ void afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, gfp_t gfp)
 	/* If the call is going to be asynchronous, we need an extra ref for
 	 * the call to hold itself so the caller need not hang on to its ref.
 	 */
-	if (call->async)
+	if (call->async) {
 		afs_get_call(call, afs_call_trace_get);
+		call->drop_ref = true;
+	}
 
 	/* create a call */
 	rxcall = rxrpc_kernel_begin_call(call->net->socket, srx, call->key,
@@ -413,7 +414,8 @@ void afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, gfp_t gfp)
 					  afs_wake_up_async_call :
 					  afs_wake_up_call_waiter),
 					 call->upgrade,
-					 call->intr,
+					 (call->intr ? RXRPC_PREINTERRUPTIBLE :
+					  RXRPC_UNINTERRUPTIBLE),
 					 call->debug_id);
 	if (IS_ERR(rxcall)) {
 		ret = PTR_ERR(rxcall);
@@ -513,12 +515,12 @@ static void afs_deliver_to_call(struct afs_call *call)
 	       state == AFS_CALL_SV_AWAIT_ACK
 	       ) {
 		if (state == AFS_CALL_SV_AWAIT_ACK) {
-			iov_iter_kvec(&call->iter, READ, NULL, 0, 0);
+			iov_iter_kvec(&call->def_iter, READ, NULL, 0, 0);
 			ret = rxrpc_kernel_recv_data(call->net->socket,
-						     call->rxcall, &call->iter,
+						     call->rxcall, &call->def_iter,
 						     false, &remote_abort,
 						     &call->service_id);
-			trace_afs_receive_data(call, &call->iter, false, ret);
+			trace_afs_receive_data(call, &call->def_iter, false, ret);
 
 			if (ret == -EINPROGRESS || ret == -EAGAIN)
 				return;
@@ -584,8 +586,6 @@ static void afs_deliver_to_call(struct afs_call *call)
 done:
 	if (call->type->done)
 		call->type->done(call);
-	if (state == AFS_CALL_COMPLETE && call->incoming)
-		afs_put_call(call);
 out:
 	_leave("");
 	return;
@@ -604,11 +604,7 @@ call_complete:
 long afs_wait_for_call_to_complete(struct afs_call *call,
 				   struct afs_addr_cursor *ac)
 {
-	signed long rtt2, timeout;
 	long ret;
-	bool stalled = false;
-	u64 rtt;
-	u32 life, last_life;
 	bool rxrpc_complete = false;
 
 	DECLARE_WAITQUEUE(myself, current);
@@ -619,14 +615,6 @@ long afs_wait_for_call_to_complete(struct afs_call *call,
 	if (ret < 0)
 		goto out;
 
-	rtt = rxrpc_kernel_get_rtt(call->net->socket, call->rxcall);
-	rtt2 = nsecs_to_jiffies64(rtt) * 2;
-	if (rtt2 < 2)
-		rtt2 = 2;
-
-	timeout = rtt2;
-	rxrpc_kernel_check_life(call->net->socket, call->rxcall, &last_life);
-
 	add_wait_queue(&call->waitq, &myself);
 	for (;;) {
 		set_current_state(TASK_UNINTERRUPTIBLE);
@@ -637,37 +625,19 @@ long afs_wait_for_call_to_complete(struct afs_call *call,
 			call->need_attention = false;
 			__set_current_state(TASK_RUNNING);
 			afs_deliver_to_call(call);
-			timeout = rtt2;
 			continue;
 		}
 
 		if (afs_check_call_state(call, AFS_CALL_COMPLETE))
 			break;
 
-		if (!rxrpc_kernel_check_life(call->net->socket, call->rxcall, &life)) {
+		if (!rxrpc_kernel_check_life(call->net->socket, call->rxcall)) {
 			/* rxrpc terminated the call. */
 			rxrpc_complete = true;
 			break;
 		}
 
-		if (call->intr && timeout == 0 &&
-		    life == last_life && signal_pending(current)) {
-			if (stalled)
-				break;
-			__set_current_state(TASK_RUNNING);
-			rxrpc_kernel_probe_life(call->net->socket, call->rxcall);
-			timeout = rtt2;
-			stalled = true;
-			continue;
-		}
-
-		if (life != last_life) {
-			timeout = rtt2;
-			last_life = life;
-			stalled = false;
-		}
-
-		timeout = schedule_timeout(timeout);
+		schedule();
 	}
 
 	remove_wait_queue(&call->waitq, &myself);
@@ -735,7 +705,7 @@ static void afs_wake_up_async_call(struct sock *sk, struct rxrpc_call *rxcall,
 
 	u = atomic_fetch_add_unless(&call->usage, 1, 0);
 	if (u != 0) {
-		trace_afs_call(call, afs_call_trace_wake, u,
+		trace_afs_call(call, afs_call_trace_wake, u + 1,
 			       atomic_read(&call->net->nr_outstanding_calls),
 			       __builtin_return_address(0));
 
@@ -745,21 +715,6 @@ static void afs_wake_up_async_call(struct sock *sk, struct rxrpc_call *rxcall,
 }
 
 /*
- * Delete an asynchronous call.  The work item carries a ref to the call struct
- * that we need to release.
- */
-static void afs_delete_async_call(struct work_struct *work)
-{
-	struct afs_call *call = container_of(work, struct afs_call, async_work);
-
-	_enter("");
-
-	afs_put_call(call);
-
-	_leave("");
-}
-
-/*
  * Perform I/O processing on an asynchronous call.  The work item carries a ref
  * to the call struct that we either need to release or to pass on.
  */
@@ -774,16 +729,6 @@ static void afs_process_async_call(struct work_struct *work)
 		afs_deliver_to_call(call);
 	}
 
-	if (call->state == AFS_CALL_COMPLETE) {
-		/* We have two refs to release - one from the alloc and one
-		 * queued with the work item - and we can't just deallocate the
-		 * call because the work item may be queued again.
-		 */
-		call->async_work.func = afs_delete_async_call;
-		if (!queue_work(afs_async_calls, &call->async_work))
-			afs_put_call(call);
-	}
-
 	afs_put_call(call);
 	_leave("");
 }
@@ -810,6 +755,7 @@ void afs_charge_preallocation(struct work_struct *work)
 			if (!call)
 				break;
 
+			call->drop_ref = true;
 			call->async = true;
 			call->state = AFS_CALL_SV_AWAIT_OP_ID;
 			init_waitqueue_head(&call->waitq);
@@ -859,7 +805,7 @@ static int afs_deliver_cm_op_id(struct afs_call *call)
 {
 	int ret;
 
-	_enter("{%zu}", iov_iter_count(call->_iter));
+	_enter("{%zu}", iov_iter_count(call->iter));
 
 	/* the operation ID forms the first four bytes of the request data */
 	ret = afs_extract_data(call, true);
@@ -975,7 +921,7 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len)
 int afs_extract_data(struct afs_call *call, bool want_more)
 {
 	struct afs_net *net = call->net;
-	struct iov_iter *iter = call->_iter;
+	struct iov_iter *iter = call->iter;
 	enum afs_call_state state;
 	u32 remote_abort = 0;
 	int ret;
diff --git a/fs/afs/server.c b/fs/afs/server.c
index 64d440aaabc0..11b90ac7ea30 100644
--- a/fs/afs/server.c
+++ b/fs/afs/server.c
@@ -32,18 +32,11 @@ static void afs_dec_servers_outstanding(struct afs_net *net)
 struct afs_server *afs_find_server(struct afs_net *net,
 				   const struct sockaddr_rxrpc *srx)
 {
-	const struct sockaddr_in6 *a = &srx->transport.sin6, *b;
 	const struct afs_addr_list *alist;
 	struct afs_server *server = NULL;
 	unsigned int i;
-	bool ipv6 = true;
 	int seq = 0, diff;
 
-	if (srx->transport.sin6.sin6_addr.s6_addr32[0] == 0 ||
-	    srx->transport.sin6.sin6_addr.s6_addr32[1] == 0 ||
-	    srx->transport.sin6.sin6_addr.s6_addr32[2] == htonl(0xffff))
-		ipv6 = false;
-
 	rcu_read_lock();
 
 	do {
@@ -52,7 +45,8 @@ struct afs_server *afs_find_server(struct afs_net *net,
 		server = NULL;
 		read_seqbegin_or_lock(&net->fs_addr_lock, &seq);
 
-		if (ipv6) {
+		if (srx->transport.family == AF_INET6) {
+			const struct sockaddr_in6 *a = &srx->transport.sin6, *b;
 			hlist_for_each_entry_rcu(server, &net->fs_addresses6, addr6_link) {
 				alist = rcu_dereference(server->addresses);
 				for (i = alist->nr_ipv4; i < alist->nr_addrs; i++) {
@@ -68,15 +62,16 @@ struct afs_server *afs_find_server(struct afs_net *net,
 				}
 			}
 		} else {
+			const struct sockaddr_in *a = &srx->transport.sin, *b;
 			hlist_for_each_entry_rcu(server, &net->fs_addresses4, addr4_link) {
 				alist = rcu_dereference(server->addresses);
 				for (i = 0; i < alist->nr_ipv4; i++) {
-					b = &alist->addrs[i].transport.sin6;
-					diff = ((u16 __force)a->sin6_port -
-						(u16 __force)b->sin6_port);
+					b = &alist->addrs[i].transport.sin;
+					diff = ((u16 __force)a->sin_port -
+						(u16 __force)b->sin_port);
 					if (diff == 0)
-						diff = ((u32 __force)a->sin6_addr.s6_addr32[3] -
-							(u32 __force)b->sin6_addr.s6_addr32[3]);
+						diff = ((u32 __force)a->sin_addr.s_addr -
+							(u32 __force)b->sin_addr.s_addr);
 					if (diff == 0)
 						goto found;
 				}
@@ -151,7 +146,7 @@ static struct afs_server *afs_install_server(struct afs_net *net,
 	const struct afs_addr_list *alist;
 	struct afs_server *server;
 	struct rb_node **pp, *p;
-	int ret = -EEXIST, diff;
+	int diff;
 
 	_enter("%p", candidate);
 
@@ -196,7 +191,6 @@ static struct afs_server *afs_install_server(struct afs_net *net,
 		hlist_add_head_rcu(&server->addr6_link, &net->fs_addresses6);
 
 	write_sequnlock(&net->fs_addr_lock);
-	ret = 0;
 
 exists:
 	afs_get_server(server, afs_server_trace_get_install);
@@ -600,12 +594,9 @@ retry:
 	}
 
 	ret = wait_on_bit(&server->flags, AFS_SERVER_FL_UPDATING,
-			  TASK_INTERRUPTIBLE);
+			  (fc->flags & AFS_FS_CURSOR_INTR) ?
+			  TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE);
 	if (ret == -ERESTARTSYS) {
-		if (!(fc->flags & AFS_FS_CURSOR_INTR) && server->addresses) {
-			_leave(" = t [intr]");
-			return true;
-		}
 		fc->error = ret;
 		_leave(" = f [intr]");
 		return false;
diff --git a/fs/afs/super.c b/fs/afs/super.c
index 488641b1a418..dda7a9a66848 100644
--- a/fs/afs/super.c
+++ b/fs/afs/super.c
@@ -38,13 +38,13 @@ static int afs_statfs(struct dentry *dentry, struct kstatfs *buf);
 static int afs_show_devname(struct seq_file *m, struct dentry *root);
 static int afs_show_options(struct seq_file *m, struct dentry *root);
 static int afs_init_fs_context(struct fs_context *fc);
-static const struct fs_parameter_description afs_fs_parameters;
+static const struct fs_parameter_spec afs_fs_parameters[];
 
 struct file_system_type afs_fs_type = {
 	.owner			= THIS_MODULE,
 	.name			= "afs",
 	.init_fs_context	= afs_init_fs_context,
-	.parameters		= &afs_fs_parameters,
+	.parameters		= afs_fs_parameters,
 	.kill_sb		= afs_kill_super,
 	.fs_flags		= FS_RENAME_DOES_D_MOVE,
 };
@@ -73,28 +73,22 @@ enum afs_param {
 	Opt_source,
 };
 
-static const struct fs_parameter_spec afs_param_specs[] = {
-	fsparam_flag  ("autocell",	Opt_autocell),
-	fsparam_flag  ("dyn",		Opt_dyn),
-	fsparam_enum  ("flock",		Opt_flock),
-	fsparam_string("source",	Opt_source),
+static const struct constant_table afs_param_flock[] = {
+	{"local",	afs_flock_mode_local },
+	{"openafs",	afs_flock_mode_openafs },
+	{"strict",	afs_flock_mode_strict },
+	{"write",	afs_flock_mode_write },
 	{}
 };
 
-static const struct fs_parameter_enum afs_param_enums[] = {
-	{ Opt_flock,	"local",	afs_flock_mode_local },
-	{ Opt_flock,	"openafs",	afs_flock_mode_openafs },
-	{ Opt_flock,	"strict",	afs_flock_mode_strict },
-	{ Opt_flock,	"write",	afs_flock_mode_write },
+static const struct fs_parameter_spec afs_fs_parameters[] = {
+	fsparam_flag  ("autocell",	Opt_autocell),
+	fsparam_flag  ("dyn",		Opt_dyn),
+	fsparam_enum  ("flock",		Opt_flock, afs_param_flock),
+	fsparam_string("source",	Opt_source),
 	{}
 };
 
-static const struct fs_parameter_description afs_fs_parameters = {
-	.name		= "kAFS",
-	.specs		= afs_param_specs,
-	.enums		= afs_param_enums,
-};
-
 /*
  * initialise the filesystem
  */
@@ -323,7 +317,7 @@ static int afs_parse_param(struct fs_context *fc, struct fs_parameter *param)
 	struct afs_fs_context *ctx = fc->fs_private;
 	int opt;
 
-	opt = fs_parse(fc, &afs_fs_parameters, param, &result);
+	opt = fs_parse(fc, afs_fs_parameters, param, &result);
 	if (opt < 0)
 		return opt;
 
@@ -404,6 +398,7 @@ static int afs_test_super(struct super_block *sb, struct fs_context *fc)
 	return (as->net_ns == fc->net_ns &&
 		as->volume &&
 		as->volume->vid == ctx->volume->vid &&
+		as->cell == ctx->cell &&
 		!as->dyn_root);
 }
 
@@ -448,7 +443,6 @@ static int afs_fill_super(struct super_block *sb, struct afs_fs_context *ctx)
 	/* allocate the root inode and dentry */
 	if (as->dyn_root) {
 		inode = afs_iget_pseudo_dir(sb, true);
-		sb->s_flags	|= SB_RDONLY;
 	} else {
 		sprintf(sb->s_id, "%llu", as->volume->vid);
 		afs_activate_volume(as->volume);
diff --git a/fs/afs/vl_list.c b/fs/afs/vl_list.c
index 21eb0c0be912..8fea54eba0c2 100644
--- a/fs/afs/vl_list.c
+++ b/fs/afs/vl_list.c
@@ -279,8 +279,8 @@ struct afs_vlserver_list *afs_extract_vlserver_list(struct afs_cell *cell,
 			struct afs_addr_list *old = addrs;
 
 			write_lock(&server->lock);
-			rcu_swap_protected(server->addresses, old,
-					   lockdep_is_held(&server->lock));
+			old = rcu_replace_pointer(server->addresses, old,
+						  lockdep_is_held(&server->lock));
 			write_unlock(&server->lock);
 			afs_put_addrlist(old);
 		}
diff --git a/fs/afs/vl_probe.c b/fs/afs/vl_probe.c
index 858498cc1b05..e3aa013c2177 100644
--- a/fs/afs/vl_probe.c
+++ b/fs/afs/vl_probe.c
@@ -31,10 +31,9 @@ void afs_vlserver_probe_result(struct afs_call *call)
 	struct afs_addr_list *alist = call->alist;
 	struct afs_vlserver *server = call->vlserver;
 	unsigned int server_index = call->server_index;
+	unsigned int rtt_us = 0;
 	unsigned int index = call->addr_ix;
-	unsigned int rtt = UINT_MAX;
 	bool have_result = false;
-	u64 _rtt;
 	int ret = call->error;
 
 	_enter("%s,%u,%u,%d,%d", server->name, server_index, index, ret, call->abort_code);
@@ -93,15 +92,9 @@ responded:
 		}
 	}
 
-	/* Get the RTT and scale it to fit into a 32-bit value that represents
-	 * over a minute of time so that we can access it with one instruction
-	 * on a 32-bit system.
-	 */
-	_rtt = rxrpc_kernel_get_rtt(call->net->socket, call->rxcall);
-	_rtt /= 64;
-	rtt = (_rtt > UINT_MAX) ? UINT_MAX : _rtt;
-	if (rtt < server->probe.rtt) {
-		server->probe.rtt = rtt;
+	rtt_us = rxrpc_kernel_get_srtt(call->net->socket, call->rxcall);
+	if (rtt_us < server->probe.rtt) {
+		server->probe.rtt = rtt_us;
 		alist->preferred = index;
 		have_result = true;
 	}
@@ -113,8 +106,7 @@ out:
 	spin_unlock(&server->probe_lock);
 
 	_debug("probe [%u][%u] %pISpc rtt=%u ret=%d",
-	       server_index, index, &alist->addrs[index].transport,
-	       (unsigned int)rtt, ret);
+	       server_index, index, &alist->addrs[index].transport, rtt_us, ret);
 
 	have_result |= afs_vl_probe_done(server);
 	if (have_result) {
diff --git a/fs/afs/vl_rotate.c b/fs/afs/vl_rotate.c
index 9a5ce9687779..72eacc14e6e1 100644
--- a/fs/afs/vl_rotate.c
+++ b/fs/afs/vl_rotate.c
@@ -302,8 +302,8 @@ static void afs_vl_dump_edestaddrreq(const struct afs_vl_cursor *vc)
 				pr_notice("VC:  - nr=%u/%u/%u pf=%u\n",
 					  a->nr_ipv4, a->nr_addrs, a->max_addrs,
 					  a->preferred);
-				pr_notice("VC:  - pr=%lx R=%lx F=%lx\n",
-					  a->probed, a->responded, a->failed);
+				pr_notice("VC:  - R=%lx F=%lx\n",
+					  a->responded, a->failed);
 				if (a == vc->ac.alist)
 					pr_notice("VC:  - current\n");
 			}
diff --git a/fs/afs/vlclient.c b/fs/afs/vlclient.c
index cfb0ac4bd039..516e9a3bb5b4 100644
--- a/fs/afs/vlclient.c
+++ b/fs/afs/vlclient.c
@@ -185,7 +185,7 @@ static int afs_deliver_vl_get_addrs_u(struct afs_call *call)
 	int i, ret;
 
 	_enter("{%u,%zu/%u}",
-	       call->unmarshall, iov_iter_count(call->_iter), call->count);
+	       call->unmarshall, iov_iter_count(call->iter), call->count);
 
 	switch (call->unmarshall) {
 	case 0:
@@ -316,7 +316,7 @@ static int afs_deliver_vl_get_capabilities(struct afs_call *call)
 	int ret;
 
 	_enter("{%u,%zu/%u}",
-	       call->unmarshall, iov_iter_count(call->_iter), call->count);
+	       call->unmarshall, iov_iter_count(call->iter), call->count);
 
 	switch (call->unmarshall) {
 	case 0:
@@ -425,7 +425,7 @@ static int afs_deliver_yfsvl_get_endpoints(struct afs_call *call)
 	int ret;
 
 	_enter("{%u,%zu,%u}",
-	       call->unmarshall, iov_iter_count(call->_iter), call->count2);
+	       call->unmarshall, iov_iter_count(call->iter), call->count2);
 
 	switch (call->unmarshall) {
 	case 0:
diff --git a/fs/afs/volume.c b/fs/afs/volume.c
index 92ca5e27573b..4310336b9bb8 100644
--- a/fs/afs/volume.c
+++ b/fs/afs/volume.c
@@ -281,7 +281,7 @@ error:
 /*
  * Make sure the volume record is up to date.
  */
-int afs_check_volume_status(struct afs_volume *volume, struct key *key)
+int afs_check_volume_status(struct afs_volume *volume, struct afs_fs_cursor *fc)
 {
 	time64_t now = ktime_get_real_seconds();
 	int ret, retries = 0;
@@ -299,7 +299,7 @@ retry:
 	}
 
 	if (!test_and_set_bit_lock(AFS_VOLUME_UPDATING, &volume->flags)) {
-		ret = afs_update_volume_status(volume, key);
+		ret = afs_update_volume_status(volume, fc->key);
 		clear_bit_unlock(AFS_VOLUME_WAIT, &volume->flags);
 		clear_bit_unlock(AFS_VOLUME_UPDATING, &volume->flags);
 		wake_up_bit(&volume->flags, AFS_VOLUME_WAIT);
@@ -312,7 +312,9 @@ retry:
 		return 0;
 	}
 
-	ret = wait_on_bit(&volume->flags, AFS_VOLUME_WAIT, TASK_INTERRUPTIBLE);
+	ret = wait_on_bit(&volume->flags, AFS_VOLUME_WAIT,
+			  (fc->flags & AFS_FS_CURSOR_INTR) ?
+			  TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE);
 	if (ret == -ERESTARTSYS) {
 		_leave(" = %d", ret);
 		return ret;
diff --git a/fs/afs/xattr.c b/fs/afs/xattr.c
index 5552d034090a..7af41fd5f3ee 100644
--- a/fs/afs/xattr.c
+++ b/fs/afs/xattr.c
@@ -228,11 +228,11 @@ static int afs_xattr_get_yfs(const struct xattr_handler *handler,
 		break;
 	case 1:
 		data = buf;
-		dsize = snprintf(buf, sizeof(buf), "%u", yacl->inherit_flag);
+		dsize = scnprintf(buf, sizeof(buf), "%u", yacl->inherit_flag);
 		break;
 	case 2:
 		data = buf;
-		dsize = snprintf(buf, sizeof(buf), "%u", yacl->num_cleaned);
+		dsize = scnprintf(buf, sizeof(buf), "%u", yacl->num_cleaned);
 		break;
 	case 3:
 		data = yacl->vol_acl->data;
@@ -370,13 +370,15 @@ static int afs_xattr_get_fid(const struct xattr_handler *handler,
 	/* The volume ID is 64-bit, the vnode ID is 96-bit and the
 	 * uniquifier is 32-bit.
 	 */
-	len = sprintf(text, "%llx:", vnode->fid.vid);
+	len = scnprintf(text, sizeof(text), "%llx:", vnode->fid.vid);
 	if (vnode->fid.vnode_hi)
-		len += sprintf(text + len, "%x%016llx",
-			       vnode->fid.vnode_hi, vnode->fid.vnode);
+		len += scnprintf(text + len, sizeof(text) - len, "%x%016llx",
+				vnode->fid.vnode_hi, vnode->fid.vnode);
 	else
-		len += sprintf(text + len, "%llx", vnode->fid.vnode);
-	len += sprintf(text + len, ":%x", vnode->fid.unique);
+		len += scnprintf(text + len, sizeof(text) - len, "%llx",
+				 vnode->fid.vnode);
+	len += scnprintf(text + len, sizeof(text) - len, ":%x",
+			 vnode->fid.unique);
 
 	if (size == 0)
 		return len;
diff --git a/fs/afs/yfsclient.c b/fs/afs/yfsclient.c
index 3ee7abf4b2d0..fe413e7a5cf4 100644
--- a/fs/afs/yfsclient.c
+++ b/fs/afs/yfsclient.c
@@ -152,8 +152,8 @@ static void yfs_check_req(struct afs_call *call, __be32 *bp)
 		pr_err("kAFS: %s: Request buffer overflow (%zu>%u)\n",
 		       call->type->name, len, call->request_size);
 	else if (len < call->request_size)
-		pr_warning("kAFS: %s: Request buffer underflow (%zu<%u)\n",
-			   call->type->name, len, call->request_size);
+		pr_warn("kAFS: %s: Request buffer underflow (%zu<%u)\n",
+			call->type->name, len, call->request_size);
 }
 
 /*
@@ -165,15 +165,15 @@ static void xdr_dump_bad(const __be32 *bp)
 	int i;
 
 	pr_notice("YFS XDR: Bad status record\n");
-	for (i = 0; i < 5 * 4 * 4; i += 16) {
+	for (i = 0; i < 6 * 4 * 4; i += 16) {
 		memcpy(x, bp, 16);
 		bp += 4;
 		pr_notice("%03x: %08x %08x %08x %08x\n",
 			  i, ntohl(x[0]), ntohl(x[1]), ntohl(x[2]), ntohl(x[3]));
 	}
 
-	memcpy(x, bp, 4);
-	pr_notice("0x50: %08x\n", ntohl(x[0]));
+	memcpy(x, bp, 8);
+	pr_notice("0x60: %08x %08x\n", ntohl(x[0]), ntohl(x[1]));
 }
 
 /*
@@ -186,13 +186,14 @@ static int xdr_decode_YFSFetchStatus(const __be32 **_bp,
 	const struct yfs_xdr_YFSFetchStatus *xdr = (const void *)*_bp;
 	struct afs_file_status *status = &scb->status;
 	u32 type;
+	int ret;
 
 	status->abort_code = ntohl(xdr->abort_code);
 	if (status->abort_code != 0) {
 		if (status->abort_code == VNOVNODE)
 			status->nlink = 0;
 		scb->have_error = true;
-		return 0;
+		goto good;
 	}
 
 	type = ntohl(xdr->type);
@@ -220,13 +221,16 @@ static int xdr_decode_YFSFetchStatus(const __be32 **_bp,
 	status->size		= xdr_to_u64(xdr->size);
 	status->data_version	= xdr_to_u64(xdr->data_version);
 	scb->have_status	= true;
-
+good:
+	ret = 0;
+advance:
 	*_bp += xdr_size(xdr);
-	return 0;
+	return ret;
 
 bad:
 	xdr_dump_bad(*_bp);
-	return afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status);
+	ret = afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status);
+	goto advance;
 }
 
 /*
@@ -441,7 +445,7 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call)
 	int ret;
 
 	_enter("{%u,%zu/%llu}",
-	       call->unmarshall, iov_iter_count(&call->iter), req->actual_len);
+	       call->unmarshall, iov_iter_count(call->iter), req->actual_len);
 
 	switch (call->unmarshall) {
 	case 0:
@@ -476,14 +480,14 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call)
 		call->bvec[0].bv_len = size;
 		call->bvec[0].bv_offset = req->offset;
 		call->bvec[0].bv_page = req->pages[req->index];
-		iov_iter_bvec(&call->iter, READ, call->bvec, 1, size);
+		iov_iter_bvec(&call->def_iter, READ, call->bvec, 1, size);
 		ASSERTCMP(size, <=, PAGE_SIZE);
 		/* Fall through */
 
 		/* extract the returned data */
 	case 2:
 		_debug("extract data %zu/%llu",
-		       iov_iter_count(&call->iter), req->remain);
+		       iov_iter_count(call->iter), req->remain);
 
 		ret = afs_extract_data(call, true);
 		if (ret < 0)
@@ -493,8 +497,6 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call)
 		ASSERTCMP(req->offset, <=, PAGE_SIZE);
 		if (req->offset == PAGE_SIZE) {
 			req->offset = 0;
-			if (req->page_done)
-				req->page_done(req);
 			req->index++;
 			if (req->remain > 0)
 				goto begin_page;
@@ -511,7 +513,7 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call)
 
 	case 3:
 		_debug("extract discard %zu/%llu",
-		       iov_iter_count(&call->iter), req->actual_len - req->len);
+		       iov_iter_count(call->iter), req->actual_len - req->len);
 
 		ret = afs_extract_data(call, true);
 		if (ret < 0)
@@ -552,11 +554,13 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call)
 		if (req->offset < PAGE_SIZE)
 			zero_user_segment(req->pages[req->index],
 					  req->offset, PAGE_SIZE);
-		if (req->page_done)
-			req->page_done(req);
 		req->offset = 0;
 	}
 
+	if (req->page_done)
+		for (req->index = 0; req->index < req->nr_pages; req->index++)
+			req->page_done(req);
+
 	_leave(" = 0 [done]");
 	return 0;
 }
@@ -605,7 +609,7 @@ int yfs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_status_cb *scb,
 	call->key = fc->key;
 	call->out_scb = scb;
 	call->out_volsync = NULL;
-	call->read_request = req;
+	call->read_request = afs_get_read(req);
 
 	/* marshall the parameters */
 	bp = call->request;
@@ -616,7 +620,6 @@ int yfs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_status_cb *scb,
 	bp = xdr_encode_u64(bp, req->len);
 	yfs_check_req(call, bp);
 
-	refcount_inc(&req->usage);
 	afs_use_fs_server(call, fc->cbi);
 	trace_afs_make_fs_call(call, &vnode->fid);
 	afs_set_fc_call(call, fc);
@@ -1154,11 +1157,9 @@ static int yfs_deliver_fs_rename(struct afs_call *call)
 	ret = xdr_decode_YFSFetchStatus(&bp, call, call->out_dir_scb);
 	if (ret < 0)
 		return ret;
-	if (call->out_dir_scb != call->out_scb) {
-		ret = xdr_decode_YFSFetchStatus(&bp, call, call->out_scb);
-		if (ret < 0)
-			return ret;
-	}
+	ret = xdr_decode_YFSFetchStatus(&bp, call, call->out_scb);
+	if (ret < 0)
+		return ret;
 
 	xdr_decode_YFSVolSync(&bp, call->out_volsync);
 	_leave(" = 0 [done]");
diff --git a/fs/aio.c b/fs/aio.c
index 0d9a559d488c..6483f9274d5e 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -176,6 +176,7 @@ struct fsync_iocb {
 	struct file		*file;
 	struct work_struct	work;
 	bool			datasync;
+	struct cred		*creds;
 };
 
 struct poll_iocb {
@@ -1589,8 +1590,11 @@ static int aio_write(struct kiocb *req, const struct iocb *iocb,
 static void aio_fsync_work(struct work_struct *work)
 {
 	struct aio_kiocb *iocb = container_of(work, struct aio_kiocb, fsync.work);
+	const struct cred *old_cred = override_creds(iocb->fsync.creds);
 
 	iocb->ki_res.res = vfs_fsync(iocb->fsync.file, iocb->fsync.datasync);
+	revert_creds(old_cred);
+	put_cred(iocb->fsync.creds);
 	iocb_put(iocb);
 }
 
@@ -1604,12 +1608,24 @@ static int aio_fsync(struct fsync_iocb *req, const struct iocb *iocb,
 	if (unlikely(!req->file->f_op->fsync))
 		return -EINVAL;
 
+	req->creds = prepare_creds();
+	if (!req->creds)
+		return -ENOMEM;
+
 	req->datasync = datasync;
 	INIT_WORK(&req->work, aio_fsync_work);
 	schedule_work(&req->work);
 	return 0;
 }
 
+static void aio_poll_put_work(struct work_struct *work)
+{
+	struct poll_iocb *req = container_of(work, struct poll_iocb, work);
+	struct aio_kiocb *iocb = container_of(req, struct aio_kiocb, poll);
+
+	iocb_put(iocb);
+}
+
 static void aio_poll_complete_work(struct work_struct *work)
 {
 	struct poll_iocb *req = container_of(work, struct poll_iocb, work);
@@ -1674,6 +1690,8 @@ static int aio_poll_wake(struct wait_queue_entry *wait, unsigned mode, int sync,
 	list_del_init(&req->wait.entry);
 
 	if (mask && spin_trylock_irqsave(&iocb->ki_ctx->ctx_lock, flags)) {
+		struct kioctx *ctx = iocb->ki_ctx;
+
 		/*
 		 * Try to complete the iocb inline if we can. Use
 		 * irqsave/irqrestore because not all filesystems (e.g. fuse)
@@ -1683,8 +1701,14 @@ static int aio_poll_wake(struct wait_queue_entry *wait, unsigned mode, int sync,
 		list_del(&iocb->ki_list);
 		iocb->ki_res.res = mangle_poll(mask);
 		req->done = true;
-		spin_unlock_irqrestore(&iocb->ki_ctx->ctx_lock, flags);
-		iocb_put(iocb);
+		if (iocb->ki_eventfd && eventfd_signal_count()) {
+			iocb = NULL;
+			INIT_WORK(&req->work, aio_poll_put_work);
+			schedule_work(&req->work);
+		}
+		spin_unlock_irqrestore(&ctx->ctx_lock, flags);
+		if (iocb)
+			iocb_put(iocb);
 	} else {
 		schedule_work(&req->work);
 	}
@@ -2056,7 +2080,7 @@ static long do_io_getevents(aio_context_t ctx_id,
  *	specifies an infinite timeout. Note that the timeout pointed to by
  *	timeout is relative.  Will fail with -ENOSYS if not implemented.
  */
-#if !defined(CONFIG_64BIT_TIME) || defined(CONFIG_64BIT)
+#ifdef CONFIG_64BIT
 
 SYSCALL_DEFINE5(io_getevents, aio_context_t, ctx_id,
 		long, min_nr,
diff --git a/fs/attr.c b/fs/attr.c
index df28035aa23e..b4bbdbd4c8ca 100644
--- a/fs/attr.c
+++ b/fs/attr.c
@@ -183,18 +183,12 @@ void setattr_copy(struct inode *inode, const struct iattr *attr)
 		inode->i_uid = attr->ia_uid;
 	if (ia_valid & ATTR_GID)
 		inode->i_gid = attr->ia_gid;
-	if (ia_valid & ATTR_ATIME) {
-		inode->i_atime = timestamp_truncate(attr->ia_atime,
-						  inode);
-	}
-	if (ia_valid & ATTR_MTIME) {
-		inode->i_mtime = timestamp_truncate(attr->ia_mtime,
-						  inode);
-	}
-	if (ia_valid & ATTR_CTIME) {
-		inode->i_ctime = timestamp_truncate(attr->ia_ctime,
-						  inode);
-	}
+	if (ia_valid & ATTR_ATIME)
+		inode->i_atime = attr->ia_atime;
+	if (ia_valid & ATTR_MTIME)
+		inode->i_mtime = attr->ia_mtime;
+	if (ia_valid & ATTR_CTIME)
+		inode->i_ctime = attr->ia_ctime;
 	if (ia_valid & ATTR_MODE) {
 		umode_t mode = attr->ia_mode;
 
@@ -268,8 +262,13 @@ int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **de
 	attr->ia_ctime = now;
 	if (!(ia_valid & ATTR_ATIME_SET))
 		attr->ia_atime = now;
+	else
+		attr->ia_atime = timestamp_truncate(attr->ia_atime, inode);
 	if (!(ia_valid & ATTR_MTIME_SET))
 		attr->ia_mtime = now;
+	else
+		attr->ia_mtime = timestamp_truncate(attr->ia_mtime, inode);
+
 	if (ia_valid & ATTR_KILL_PRIV) {
 		error = security_inode_need_killpriv(dentry);
 		if (error < 0)
diff --git a/fs/autofs/autofs_i.h b/fs/autofs/autofs_i.h
index 8bcec8dcabb6..054f97b07754 100644
--- a/fs/autofs/autofs_i.h
+++ b/fs/autofs/autofs_i.h
@@ -63,7 +63,7 @@ struct autofs_info {
 
 	struct autofs_sb_info *sbi;
 	unsigned long last_used;
-	atomic_t count;
+	int count;
 
 	kuid_t uid;
 	kgid_t gid;
diff --git a/fs/autofs/dev-ioctl.c b/fs/autofs/dev-ioctl.c
index a3cdb0036c5d..f3a0f412b43b 100644
--- a/fs/autofs/dev-ioctl.c
+++ b/fs/autofs/dev-ioctl.c
@@ -186,7 +186,7 @@ static int find_autofs_mount(const char *pathname,
 	struct path path;
 	int err;
 
-	err = kern_path_mountpoint(AT_FDCWD, pathname, &path, 0);
+	err = kern_path(pathname, LOOKUP_MOUNTPOINT, &path);
 	if (err)
 		return err;
 	err = -ENOENT;
@@ -519,8 +519,8 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 
 	if (!fp || param->ioctlfd == -1) {
 		if (autofs_type_any(type))
-			err = kern_path_mountpoint(AT_FDCWD,
-						   name, &path, LOOKUP_FOLLOW);
+			err = kern_path(name, LOOKUP_FOLLOW | LOOKUP_MOUNTPOINT,
+					&path);
 		else
 			err = find_autofs_mount(name, &path,
 						test_by_type, &type);
diff --git a/fs/autofs/expire.c b/fs/autofs/expire.c
index 91f5787dae7c..a1c7701007e7 100644
--- a/fs/autofs/expire.c
+++ b/fs/autofs/expire.c
@@ -211,7 +211,7 @@ static int autofs_tree_busy(struct vfsmount *mnt,
 			}
 		} else {
 			struct autofs_info *ino = autofs_dentry_ino(p);
-			unsigned int ino_count = atomic_read(&ino->count);
+			unsigned int ino_count = READ_ONCE(ino->count);
 
 			/* allow for dget above and top is already dgot */
 			if (p == top)
@@ -379,7 +379,7 @@ static struct dentry *should_expire(struct dentry *dentry,
 		/* Not a forced expire? */
 		if (!(how & AUTOFS_EXP_FORCED)) {
 			/* ref-walk currently on this dentry? */
-			ino_count = atomic_read(&ino->count) + 1;
+			ino_count = READ_ONCE(ino->count) + 1;
 			if (d_count(dentry) > ino_count)
 				return NULL;
 		}
@@ -396,7 +396,7 @@ static struct dentry *should_expire(struct dentry *dentry,
 		/* Not a forced expire? */
 		if (!(how & AUTOFS_EXP_FORCED)) {
 			/* ref-walk currently on this dentry? */
-			ino_count = atomic_read(&ino->count) + 1;
+			ino_count = READ_ONCE(ino->count) + 1;
 			if (d_count(dentry) > ino_count)
 				return NULL;
 		}
diff --git a/fs/autofs/root.c b/fs/autofs/root.c
index 29abafc0ce31..5aaa1732bf1e 100644
--- a/fs/autofs/root.c
+++ b/fs/autofs/root.c
@@ -569,10 +569,9 @@ static int autofs_dir_symlink(struct inode *dir,
 	d_add(dentry, inode);
 
 	dget(dentry);
-	atomic_inc(&ino->count);
+	ino->count++;
 	p_ino = autofs_dentry_ino(dentry->d_parent);
-	if (p_ino && !IS_ROOT(dentry))
-		atomic_inc(&p_ino->count);
+	p_ino->count++;
 
 	dir->i_mtime = current_time(dir);
 
@@ -610,11 +609,9 @@ static int autofs_dir_unlink(struct inode *dir, struct dentry *dentry)
 	if (sbi->flags & AUTOFS_SBI_CATATONIC)
 		return -EACCES;
 
-	if (atomic_dec_and_test(&ino->count)) {
-		p_ino = autofs_dentry_ino(dentry->d_parent);
-		if (p_ino && !IS_ROOT(dentry))
-			atomic_dec(&p_ino->count);
-	}
+	ino->count--;
+	p_ino = autofs_dentry_ino(dentry->d_parent);
+	p_ino->count--;
 	dput(ino->dentry);
 
 	d_inode(dentry)->i_size = 0;
@@ -660,7 +657,6 @@ static void autofs_set_leaf_automount_flags(struct dentry *dentry)
 
 static void autofs_clear_leaf_automount_flags(struct dentry *dentry)
 {
-	struct list_head *d_child;
 	struct dentry *parent;
 
 	/* flags for dentrys in the root are handled elsewhere */
@@ -673,10 +669,7 @@ static void autofs_clear_leaf_automount_flags(struct dentry *dentry)
 	/* only consider parents below dentrys in the root */
 	if (IS_ROOT(parent->d_parent))
 		return;
-	d_child = &dentry->d_child;
-	/* Set parent managed if it's becoming empty */
-	if (d_child->next == &parent->d_subdirs &&
-	    d_child->prev == &parent->d_subdirs)
+	if (autofs_dentry_ino(parent)->count == 2)
 		managed_dentry_set_managed(parent);
 }
 
@@ -698,11 +691,10 @@ static int autofs_dir_rmdir(struct inode *dir, struct dentry *dentry)
 	if (sbi->flags & AUTOFS_SBI_CATATONIC)
 		return -EACCES;
 
-	spin_lock(&sbi->lookup_lock);
-	if (!simple_empty(dentry)) {
-		spin_unlock(&sbi->lookup_lock);
+	if (ino->count != 1)
 		return -ENOTEMPTY;
-	}
+
+	spin_lock(&sbi->lookup_lock);
 	__autofs_add_expiring(dentry);
 	d_drop(dentry);
 	spin_unlock(&sbi->lookup_lock);
@@ -710,11 +702,9 @@ static int autofs_dir_rmdir(struct inode *dir, struct dentry *dentry)
 	if (sbi->version < 5)
 		autofs_clear_leaf_automount_flags(dentry);
 
-	if (atomic_dec_and_test(&ino->count)) {
-		p_ino = autofs_dentry_ino(dentry->d_parent);
-		if (p_ino && dentry->d_parent != dentry)
-			atomic_dec(&p_ino->count);
-	}
+	ino->count--;
+	p_ino = autofs_dentry_ino(dentry->d_parent);
+	p_ino->count--;
 	dput(ino->dentry);
 	d_inode(dentry)->i_size = 0;
 	clear_nlink(d_inode(dentry));
@@ -760,10 +750,9 @@ static int autofs_dir_mkdir(struct inode *dir,
 		autofs_set_leaf_automount_flags(dentry);
 
 	dget(dentry);
-	atomic_inc(&ino->count);
+	ino->count++;
 	p_ino = autofs_dentry_ino(dentry->d_parent);
-	if (p_ino && !IS_ROOT(dentry))
-		atomic_inc(&p_ino->count);
+	p_ino->count++;
 	inc_nlink(dir);
 	dir->i_mtime = current_time(dir);
 
diff --git a/fs/bfs/Kconfig b/fs/bfs/Kconfig
index 3e1247f07913..3a757805b585 100644
--- a/fs/bfs/Kconfig
+++ b/fs/bfs/Kconfig
@@ -11,7 +11,7 @@ config BFS_FS
 	  on your /stand slice from within Linux.  You then also need to say Y
 	  to "UnixWare slices support", below.  More information about the BFS
 	  file system is contained in the file
-	  <file:Documentation/filesystems/bfs.txt>.
+	  <file:Documentation/filesystems/bfs.rst>.
 
 	  If you don't know what this is about, say N.
 
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index c5642bcb6b46..8945671fe0e5 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -27,6 +27,7 @@
 #include <linux/highuid.h>
 #include <linux/compiler.h>
 #include <linux/highmem.h>
+#include <linux/hugetlb.h>
 #include <linux/pagemap.h>
 #include <linux/vmalloc.h>
 #include <linux/security.h>
@@ -39,12 +40,18 @@
 #include <linux/sched/coredump.h>
 #include <linux/sched/task_stack.h>
 #include <linux/sched/cputime.h>
+#include <linux/sizes.h>
+#include <linux/types.h>
 #include <linux/cred.h>
 #include <linux/dax.h>
 #include <linux/uaccess.h>
 #include <asm/param.h>
 #include <asm/page.h>
 
+#ifndef ELF_COMPAT
+#define ELF_COMPAT 0
+#endif
+
 #ifndef user_long_t
 #define user_long_t long
 #endif
@@ -97,7 +104,7 @@ static struct linux_binfmt elf_format = {
 	.min_coredump	= ELF_EXEC_PAGESIZE,
 };
 
-#define BAD_ADDR(x) ((unsigned long)(x) >= TASK_SIZE)
+#define BAD_ADDR(x) (unlikely((unsigned long)(x) >= TASK_SIZE))
 
 static int set_brk(unsigned long start, unsigned long end, int prot)
 {
@@ -161,9 +168,11 @@ static int padzero(unsigned long elf_bss)
 #endif
 
 static int
-create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
-		unsigned long load_addr, unsigned long interp_load_addr)
+create_elf_tables(struct linux_binprm *bprm, const struct elfhdr *exec,
+		unsigned long load_addr, unsigned long interp_load_addr,
+		unsigned long e_entry)
 {
+	struct mm_struct *mm = current->mm;
 	unsigned long p = bprm->p;
 	int argc = bprm->argc;
 	int envc = bprm->envc;
@@ -176,7 +185,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	unsigned char k_rand_bytes[16];
 	int items;
 	elf_addr_t *elf_info;
-	int ei_index = 0;
+	int ei_index;
 	const struct cred *cred = current_cred();
 	struct vm_area_struct *vma;
 
@@ -226,12 +235,12 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 		return -EFAULT;
 
 	/* Create the ELF interpreter info */
-	elf_info = (elf_addr_t *)current->mm->saved_auxv;
+	elf_info = (elf_addr_t *)mm->saved_auxv;
 	/* update AT_VECTOR_SIZE_BASE if the number of NEW_AUX_ENT() changes */
 #define NEW_AUX_ENT(id, val) \
 	do { \
-		elf_info[ei_index++] = id; \
-		elf_info[ei_index++] = val; \
+		*elf_info++ = id; \
+		*elf_info++ = val; \
 	} while (0)
 
 #ifdef ARCH_DLINFO
@@ -251,7 +260,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	NEW_AUX_ENT(AT_PHNUM, exec->e_phnum);
 	NEW_AUX_ENT(AT_BASE, interp_load_addr);
 	NEW_AUX_ENT(AT_FLAGS, 0);
-	NEW_AUX_ENT(AT_ENTRY, exec->e_entry);
+	NEW_AUX_ENT(AT_ENTRY, e_entry);
 	NEW_AUX_ENT(AT_UID, from_kuid_munged(cred->user_ns, cred->uid));
 	NEW_AUX_ENT(AT_EUID, from_kuid_munged(cred->user_ns, cred->euid));
 	NEW_AUX_ENT(AT_GID, from_kgid_munged(cred->user_ns, cred->gid));
@@ -275,12 +284,13 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	}
 #undef NEW_AUX_ENT
 	/* AT_NULL is zero; clear the rest too */
-	memset(&elf_info[ei_index], 0,
-	       sizeof current->mm->saved_auxv - ei_index * sizeof elf_info[0]);
+	memset(elf_info, 0, (char *)mm->saved_auxv +
+			sizeof(mm->saved_auxv) - (char *)elf_info);
 
 	/* And advance past the AT_NULL entry.  */
-	ei_index += 2;
+	elf_info += 2;
 
+	ei_index = elf_info - (elf_addr_t *)mm->saved_auxv;
 	sp = STACK_ADD(p, ei_index);
 
 	items = (argc + 1) + (envc + 1) + 1;
@@ -299,7 +309,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	 * Grow the stack manually; some architectures have a limit on how
 	 * far ahead a user-space access may be in order to grow the stack.
 	 */
-	vma = find_extend_vma(current->mm, bprm->p);
+	vma = find_extend_vma(mm, bprm->p);
 	if (!vma)
 		return -EFAULT;
 
@@ -308,7 +318,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 		return -EFAULT;
 
 	/* Populate list of argv pointers back to argv strings. */
-	p = current->mm->arg_end = current->mm->arg_start;
+	p = mm->arg_end = mm->arg_start;
 	while (argc-- > 0) {
 		size_t len;
 		if (__put_user((elf_addr_t)p, sp++))
@@ -320,10 +330,10 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	}
 	if (__put_user(0, sp++))
 		return -EFAULT;
-	current->mm->arg_end = p;
+	mm->arg_end = p;
 
 	/* Populate list of envp pointers back to envp strings. */
-	current->mm->env_end = current->mm->env_start = p;
+	mm->env_end = mm->env_start = p;
 	while (envc-- > 0) {
 		size_t len;
 		if (__put_user((elf_addr_t)p, sp++))
@@ -335,10 +345,10 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	}
 	if (__put_user(0, sp++))
 		return -EFAULT;
-	current->mm->env_end = p;
+	mm->env_end = p;
 
 	/* Put the elf_info on the stack in the right place.  */
-	if (copy_to_user(sp, elf_info, ei_index * sizeof(elf_addr_t)))
+	if (copy_to_user(sp, mm->saved_auxv, ei_index * sizeof(elf_addr_t)))
 		return -EFAULT;
 	return 0;
 }
@@ -404,6 +414,17 @@ static unsigned long total_mapping_size(const struct elf_phdr *cmds, int nr)
 				ELF_PAGESTART(cmds[first_idx].p_vaddr);
 }
 
+static int elf_read(struct file *file, void *buf, size_t len, loff_t pos)
+{
+	ssize_t rv;
+
+	rv = kernel_read(file, buf, len, &pos);
+	if (unlikely(rv != len)) {
+		return (rv < 0) ? rv : -EIO;
+	}
+	return 0;
+}
+
 /**
  * load_elf_phdrs() - load ELF program headers
  * @elf_ex:   ELF header of the binary whose program headers should be loaded
@@ -418,7 +439,6 @@ static struct elf_phdr *load_elf_phdrs(const struct elfhdr *elf_ex,
 {
 	struct elf_phdr *elf_phdata = NULL;
 	int retval, err = -1;
-	loff_t pos = elf_ex->e_phoff;
 	unsigned int size;
 
 	/*
@@ -439,9 +459,9 @@ static struct elf_phdr *load_elf_phdrs(const struct elfhdr *elf_ex,
 		goto out;
 
 	/* Read in the program headers */
-	retval = kernel_read(elf_file, elf_phdata, size, &pos);
-	if (retval != size) {
-		err = (retval < 0) ? retval : -EIO;
+	retval = elf_read(elf_file, elf_phdata, size, elf_ex->e_phoff);
+	if (retval < 0) {
+		err = retval;
 		goto out;
 	}
 
@@ -525,7 +545,8 @@ static inline int arch_check_elf(struct elfhdr *ehdr, bool has_interp,
 
 #endif /* !CONFIG_ARCH_BINFMT_ELF_STATE */
 
-static inline int make_prot(u32 p_flags)
+static inline int make_prot(u32 p_flags, struct arch_elf_state *arch_state,
+			    bool has_interp, bool is_interp)
 {
 	int prot = 0;
 
@@ -535,7 +556,8 @@ static inline int make_prot(u32 p_flags)
 		prot |= PROT_WRITE;
 	if (p_flags & PF_X)
 		prot |= PROT_EXEC;
-	return prot;
+
+	return arch_elf_adjust_prot(prot, arch_state, has_interp, is_interp);
 }
 
 /* This is much more generalized than the library routine read function,
@@ -544,8 +566,9 @@ static inline int make_prot(u32 p_flags)
    an ELF header */
 
 static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
-		struct file *interpreter, unsigned long *interp_map_addr,
-		unsigned long no_base, struct elf_phdr *interp_elf_phdata)
+		struct file *interpreter,
+		unsigned long no_base, struct elf_phdr *interp_elf_phdata,
+		struct arch_elf_state *arch_state)
 {
 	struct elf_phdr *eppnt;
 	unsigned long load_addr = 0;
@@ -577,7 +600,8 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
 	for (i = 0; i < interp_elf_ex->e_phnum; i++, eppnt++) {
 		if (eppnt->p_type == PT_LOAD) {
 			int elf_type = MAP_PRIVATE | MAP_DENYWRITE;
-			int elf_prot = make_prot(eppnt->p_flags);
+			int elf_prot = make_prot(eppnt->p_flags, arch_state,
+						 true, true);
 			unsigned long vaddr = 0;
 			unsigned long k, map_addr;
 
@@ -590,8 +614,6 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex,
 			map_addr = elf_map(interpreter, load_addr + vaddr,
 					eppnt, elf_prot, elf_type, total_size);
 			total_size = 0;
-			if (!*interp_map_addr)
-				*interp_map_addr = map_addr;
 			error = map_addr;
 			if (BAD_ADDR(map_addr))
 				goto out;
@@ -670,6 +692,111 @@ out:
  * libraries.  There is no binary dependent code anywhere else.
  */
 
+static int parse_elf_property(const char *data, size_t *off, size_t datasz,
+			      struct arch_elf_state *arch,
+			      bool have_prev_type, u32 *prev_type)
+{
+	size_t o, step;
+	const struct gnu_property *pr;
+	int ret;
+
+	if (*off == datasz)
+		return -ENOENT;
+
+	if (WARN_ON_ONCE(*off > datasz || *off % ELF_GNU_PROPERTY_ALIGN))
+		return -EIO;
+	o = *off;
+	datasz -= *off;
+
+	if (datasz < sizeof(*pr))
+		return -ENOEXEC;
+	pr = (const struct gnu_property *)(data + o);
+	o += sizeof(*pr);
+	datasz -= sizeof(*pr);
+
+	if (pr->pr_datasz > datasz)
+		return -ENOEXEC;
+
+	WARN_ON_ONCE(o % ELF_GNU_PROPERTY_ALIGN);
+	step = round_up(pr->pr_datasz, ELF_GNU_PROPERTY_ALIGN);
+	if (step > datasz)
+		return -ENOEXEC;
+
+	/* Properties are supposed to be unique and sorted on pr_type: */
+	if (have_prev_type && pr->pr_type <= *prev_type)
+		return -ENOEXEC;
+	*prev_type = pr->pr_type;
+
+	ret = arch_parse_elf_property(pr->pr_type, data + o,
+				      pr->pr_datasz, ELF_COMPAT, arch);
+	if (ret)
+		return ret;
+
+	*off = o + step;
+	return 0;
+}
+
+#define NOTE_DATA_SZ SZ_1K
+#define GNU_PROPERTY_TYPE_0_NAME "GNU"
+#define NOTE_NAME_SZ (sizeof(GNU_PROPERTY_TYPE_0_NAME))
+
+static int parse_elf_properties(struct file *f, const struct elf_phdr *phdr,
+				struct arch_elf_state *arch)
+{
+	union {
+		struct elf_note nhdr;
+		char data[NOTE_DATA_SZ];
+	} note;
+	loff_t pos;
+	ssize_t n;
+	size_t off, datasz;
+	int ret;
+	bool have_prev_type;
+	u32 prev_type;
+
+	if (!IS_ENABLED(CONFIG_ARCH_USE_GNU_PROPERTY) || !phdr)
+		return 0;
+
+	/* load_elf_binary() shouldn't call us unless this is true... */
+	if (WARN_ON_ONCE(phdr->p_type != PT_GNU_PROPERTY))
+		return -ENOEXEC;
+
+	/* If the properties are crazy large, that's too bad (for now): */
+	if (phdr->p_filesz > sizeof(note))
+		return -ENOEXEC;
+
+	pos = phdr->p_offset;
+	n = kernel_read(f, &note, phdr->p_filesz, &pos);
+
+	BUILD_BUG_ON(sizeof(note) < sizeof(note.nhdr) + NOTE_NAME_SZ);
+	if (n < 0 || n < sizeof(note.nhdr) + NOTE_NAME_SZ)
+		return -EIO;
+
+	if (note.nhdr.n_type != NT_GNU_PROPERTY_TYPE_0 ||
+	    note.nhdr.n_namesz != NOTE_NAME_SZ ||
+	    strncmp(note.data + sizeof(note.nhdr),
+		    GNU_PROPERTY_TYPE_0_NAME, n - sizeof(note.nhdr)))
+		return -ENOEXEC;
+
+	off = round_up(sizeof(note.nhdr) + NOTE_NAME_SZ,
+		       ELF_GNU_PROPERTY_ALIGN);
+	if (off > n)
+		return -ENOEXEC;
+
+	if (note.nhdr.n_descsz > n - off)
+		return -ENOEXEC;
+	datasz = off + note.nhdr.n_descsz;
+
+	have_prev_type = false;
+	do {
+		ret = parse_elf_property(note.data, &off, datasz, arch,
+					 have_prev_type, &prev_type);
+		have_prev_type = true;
+	} while (!ret);
+
+	return ret == -ENOENT ? 0 : ret;
+}
+
 static int load_elf_binary(struct linux_binprm *bprm)
 {
 	struct file *interpreter = NULL; /* to shut gcc up */
@@ -677,52 +804,48 @@ static int load_elf_binary(struct linux_binprm *bprm)
 	int load_addr_set = 0;
 	unsigned long error;
 	struct elf_phdr *elf_ppnt, *elf_phdata, *interp_elf_phdata = NULL;
+	struct elf_phdr *elf_property_phdata = NULL;
 	unsigned long elf_bss, elf_brk;
 	int bss_prot = 0;
 	int retval, i;
 	unsigned long elf_entry;
+	unsigned long e_entry;
 	unsigned long interp_load_addr = 0;
 	unsigned long start_code, end_code, start_data, end_data;
 	unsigned long reloc_func_desc __maybe_unused = 0;
 	int executable_stack = EXSTACK_DEFAULT;
-	struct {
-		struct elfhdr elf_ex;
-		struct elfhdr interp_elf_ex;
-	} *loc;
+	struct elfhdr *elf_ex = (struct elfhdr *)bprm->buf;
+	struct elfhdr *interp_elf_ex = NULL;
 	struct arch_elf_state arch_state = INIT_ARCH_ELF_STATE;
+	struct mm_struct *mm;
 	struct pt_regs *regs;
 
-	loc = kmalloc(sizeof(*loc), GFP_KERNEL);
-	if (!loc) {
-		retval = -ENOMEM;
-		goto out_ret;
-	}
-	
-	/* Get the exec-header */
-	loc->elf_ex = *((struct elfhdr *)bprm->buf);
-
 	retval = -ENOEXEC;
 	/* First of all, some simple consistency checks */
-	if (memcmp(loc->elf_ex.e_ident, ELFMAG, SELFMAG) != 0)
+	if (memcmp(elf_ex->e_ident, ELFMAG, SELFMAG) != 0)
 		goto out;
 
-	if (loc->elf_ex.e_type != ET_EXEC && loc->elf_ex.e_type != ET_DYN)
+	if (elf_ex->e_type != ET_EXEC && elf_ex->e_type != ET_DYN)
 		goto out;
-	if (!elf_check_arch(&loc->elf_ex))
+	if (!elf_check_arch(elf_ex))
 		goto out;
-	if (elf_check_fdpic(&loc->elf_ex))
+	if (elf_check_fdpic(elf_ex))
 		goto out;
 	if (!bprm->file->f_op->mmap)
 		goto out;
 
-	elf_phdata = load_elf_phdrs(&loc->elf_ex, bprm->file);
+	elf_phdata = load_elf_phdrs(elf_ex, bprm->file);
 	if (!elf_phdata)
 		goto out;
 
 	elf_ppnt = elf_phdata;
-	for (i = 0; i < loc->elf_ex.e_phnum; i++, elf_ppnt++) {
+	for (i = 0; i < elf_ex->e_phnum; i++, elf_ppnt++) {
 		char *elf_interpreter;
-		loff_t pos;
+
+		if (elf_ppnt->p_type == PT_GNU_PROPERTY) {
+			elf_property_phdata = elf_ppnt;
+			continue;
+		}
 
 		if (elf_ppnt->p_type != PT_INTERP)
 			continue;
@@ -740,14 +863,10 @@ static int load_elf_binary(struct linux_binprm *bprm)
 		if (!elf_interpreter)
 			goto out_free_ph;
 
-		pos = elf_ppnt->p_offset;
-		retval = kernel_read(bprm->file, elf_interpreter,
-				     elf_ppnt->p_filesz, &pos);
-		if (retval != elf_ppnt->p_filesz) {
-			if (retval >= 0)
-				retval = -EIO;
+		retval = elf_read(bprm->file, elf_interpreter, elf_ppnt->p_filesz,
+				  elf_ppnt->p_offset);
+		if (retval < 0)
 			goto out_free_interp;
-		}
 		/* make sure path is NULL terminated */
 		retval = -ENOEXEC;
 		if (elf_interpreter[elf_ppnt->p_filesz - 1] != '\0')
@@ -765,15 +884,17 @@ static int load_elf_binary(struct linux_binprm *bprm)
 		 */
 		would_dump(bprm, interpreter);
 
+		interp_elf_ex = kmalloc(sizeof(*interp_elf_ex), GFP_KERNEL);
+		if (!interp_elf_ex) {
+			retval = -ENOMEM;
+			goto out_free_ph;
+		}
+
 		/* Get the exec headers */
-		pos = 0;
-		retval = kernel_read(interpreter, &loc->interp_elf_ex,
-				     sizeof(loc->interp_elf_ex), &pos);
-		if (retval != sizeof(loc->interp_elf_ex)) {
-			if (retval >= 0)
-				retval = -EIO;
+		retval = elf_read(interpreter, interp_elf_ex,
+				  sizeof(*interp_elf_ex), 0);
+		if (retval < 0)
 			goto out_free_dentry;
-		}
 
 		break;
 
@@ -783,7 +904,7 @@ out_free_interp:
 	}
 
 	elf_ppnt = elf_phdata;
-	for (i = 0; i < loc->elf_ex.e_phnum; i++, elf_ppnt++)
+	for (i = 0; i < elf_ex->e_phnum; i++, elf_ppnt++)
 		switch (elf_ppnt->p_type) {
 		case PT_GNU_STACK:
 			if (elf_ppnt->p_flags & PF_X)
@@ -793,7 +914,7 @@ out_free_interp:
 			break;
 
 		case PT_LOPROC ... PT_HIPROC:
-			retval = arch_elf_pt_proc(&loc->elf_ex, elf_ppnt,
+			retval = arch_elf_pt_proc(elf_ex, elf_ppnt,
 						  bprm->file, false,
 						  &arch_state);
 			if (retval)
@@ -805,25 +926,30 @@ out_free_interp:
 	if (interpreter) {
 		retval = -ELIBBAD;
 		/* Not an ELF interpreter */
-		if (memcmp(loc->interp_elf_ex.e_ident, ELFMAG, SELFMAG) != 0)
+		if (memcmp(interp_elf_ex->e_ident, ELFMAG, SELFMAG) != 0)
 			goto out_free_dentry;
 		/* Verify the interpreter has a valid arch */
-		if (!elf_check_arch(&loc->interp_elf_ex) ||
-		    elf_check_fdpic(&loc->interp_elf_ex))
+		if (!elf_check_arch(interp_elf_ex) ||
+		    elf_check_fdpic(interp_elf_ex))
 			goto out_free_dentry;
 
 		/* Load the interpreter program headers */
-		interp_elf_phdata = load_elf_phdrs(&loc->interp_elf_ex,
+		interp_elf_phdata = load_elf_phdrs(interp_elf_ex,
 						   interpreter);
 		if (!interp_elf_phdata)
 			goto out_free_dentry;
 
 		/* Pass PT_LOPROC..PT_HIPROC headers to arch code */
+		elf_property_phdata = NULL;
 		elf_ppnt = interp_elf_phdata;
-		for (i = 0; i < loc->interp_elf_ex.e_phnum; i++, elf_ppnt++)
+		for (i = 0; i < interp_elf_ex->e_phnum; i++, elf_ppnt++)
 			switch (elf_ppnt->p_type) {
+			case PT_GNU_PROPERTY:
+				elf_property_phdata = elf_ppnt;
+				break;
+
 			case PT_LOPROC ... PT_HIPROC:
-				retval = arch_elf_pt_proc(&loc->interp_elf_ex,
+				retval = arch_elf_pt_proc(interp_elf_ex,
 							  elf_ppnt, interpreter,
 							  true, &arch_state);
 				if (retval)
@@ -832,13 +958,18 @@ out_free_interp:
 			}
 	}
 
+	retval = parse_elf_properties(interpreter ?: bprm->file,
+				      elf_property_phdata, &arch_state);
+	if (retval)
+		goto out_free_dentry;
+
 	/*
 	 * Allow arch code to reject the ELF at this point, whilst it's
 	 * still possible to return an error to the code that invoked
 	 * the exec syscall.
 	 */
-	retval = arch_check_elf(&loc->elf_ex,
-				!!interpreter, &loc->interp_elf_ex,
+	retval = arch_check_elf(elf_ex,
+				!!interpreter, interp_elf_ex,
 				&arch_state);
 	if (retval)
 		goto out_free_dentry;
@@ -850,8 +981,8 @@ out_free_interp:
 
 	/* Do this immediately, since STACK_TOP as used in setup_arg_pages
 	   may depend on the personality.  */
-	SET_PERSONALITY2(loc->elf_ex, &arch_state);
-	if (elf_read_implies_exec(loc->elf_ex, executable_stack))
+	SET_PERSONALITY2(*elf_ex, &arch_state);
+	if (elf_read_implies_exec(*elf_ex, executable_stack))
 		current->personality |= READ_IMPLIES_EXEC;
 
 	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
@@ -878,7 +1009,7 @@ out_free_interp:
 	/* Now we do a little grungy work by mmapping the ELF image into
 	   the correct location in memory. */
 	for(i = 0, elf_ppnt = elf_phdata;
-	    i < loc->elf_ex.e_phnum; i++, elf_ppnt++) {
+	    i < elf_ex->e_phnum; i++, elf_ppnt++) {
 		int elf_prot, elf_flags;
 		unsigned long k, vaddr;
 		unsigned long total_size = 0;
@@ -913,7 +1044,8 @@ out_free_interp:
 			}
 		}
 
-		elf_prot = make_prot(elf_ppnt->p_flags);
+		elf_prot = make_prot(elf_ppnt->p_flags, &arch_state,
+				     !!interpreter, false);
 
 		elf_flags = MAP_PRIVATE | MAP_DENYWRITE | MAP_EXECUTABLE;
 
@@ -922,9 +1054,9 @@ out_free_interp:
 		 * If we are loading ET_EXEC or we have already performed
 		 * the ET_DYN load_addr calculations, proceed normally.
 		 */
-		if (loc->elf_ex.e_type == ET_EXEC || load_addr_set) {
+		if (elf_ex->e_type == ET_EXEC || load_addr_set) {
 			elf_flags |= MAP_FIXED;
-		} else if (loc->elf_ex.e_type == ET_DYN) {
+		} else if (elf_ex->e_type == ET_DYN) {
 			/*
 			 * This logic is run once for the first LOAD Program
 			 * Header for ET_DYN binaries to calculate the
@@ -973,7 +1105,7 @@ out_free_interp:
 			load_bias = ELF_PAGESTART(load_bias - vaddr);
 
 			total_size = total_mapping_size(elf_phdata,
-							loc->elf_ex.e_phnum);
+							elf_ex->e_phnum);
 			if (!total_size) {
 				retval = -EINVAL;
 				goto out_free_dentry;
@@ -991,7 +1123,7 @@ out_free_interp:
 		if (!load_addr_set) {
 			load_addr_set = 1;
 			load_addr = (elf_ppnt->p_vaddr - elf_ppnt->p_offset);
-			if (loc->elf_ex.e_type == ET_DYN) {
+			if (elf_ex->e_type == ET_DYN) {
 				load_bias += error -
 				             ELF_PAGESTART(load_bias + vaddr);
 				load_addr += load_bias;
@@ -999,7 +1131,7 @@ out_free_interp:
 			}
 		}
 		k = elf_ppnt->p_vaddr;
-		if (k < start_code)
+		if ((elf_ppnt->p_flags & PF_X) && k < start_code)
 			start_code = k;
 		if (start_data < k)
 			start_data = k;
@@ -1032,7 +1164,7 @@ out_free_interp:
 		}
 	}
 
-	loc->elf_ex.e_entry += load_bias;
+	e_entry = elf_ex->e_entry + load_bias;
 	elf_bss += load_bias;
 	elf_brk += load_bias;
 	start_code += load_bias;
@@ -1054,19 +1186,17 @@ out_free_interp:
 	}
 
 	if (interpreter) {
-		unsigned long interp_map_addr = 0;
-
-		elf_entry = load_elf_interp(&loc->interp_elf_ex,
+		elf_entry = load_elf_interp(interp_elf_ex,
 					    interpreter,
-					    &interp_map_addr,
-					    load_bias, interp_elf_phdata);
+					    load_bias, interp_elf_phdata,
+					    &arch_state);
 		if (!IS_ERR((void *)elf_entry)) {
 			/*
 			 * load_elf_interp() returns relocation
 			 * adjustment
 			 */
 			interp_load_addr = elf_entry;
-			elf_entry += loc->interp_elf_ex.e_entry;
+			elf_entry += interp_elf_ex->e_entry;
 		}
 		if (BAD_ADDR(elf_entry)) {
 			retval = IS_ERR((void *)elf_entry) ?
@@ -1077,15 +1207,17 @@ out_free_interp:
 
 		allow_write_access(interpreter);
 		fput(interpreter);
+
+		kfree(interp_elf_ex);
+		kfree(interp_elf_phdata);
 	} else {
-		elf_entry = loc->elf_ex.e_entry;
+		elf_entry = e_entry;
 		if (BAD_ADDR(elf_entry)) {
 			retval = -EINVAL;
 			goto out_free_dentry;
 		}
 	}
 
-	kfree(interp_elf_phdata);
 	kfree(elf_phdata);
 
 	set_binfmt(&elf_format);
@@ -1096,15 +1228,17 @@ out_free_interp:
 		goto out;
 #endif /* ARCH_HAS_SETUP_ADDITIONAL_PAGES */
 
-	retval = create_elf_tables(bprm, &loc->elf_ex,
-			  load_addr, interp_load_addr);
+	retval = create_elf_tables(bprm, elf_ex,
+			  load_addr, interp_load_addr, e_entry);
 	if (retval < 0)
 		goto out;
-	current->mm->end_code = end_code;
-	current->mm->start_code = start_code;
-	current->mm->start_data = start_data;
-	current->mm->end_data = end_data;
-	current->mm->start_stack = bprm->p;
+
+	mm = current->mm;
+	mm->end_code = end_code;
+	mm->start_code = start_code;
+	mm->start_data = start_data;
+	mm->end_data = end_data;
+	mm->start_stack = bprm->p;
 
 	if ((current->flags & PF_RANDOMIZE) && (randomize_va_space > 1)) {
 		/*
@@ -1115,12 +1249,11 @@ out_free_interp:
 		 * growing down), and into the unused ELF_ET_DYN_BASE region.
 		 */
 		if (IS_ENABLED(CONFIG_ARCH_HAS_ELF_RANDOMIZE) &&
-		    loc->elf_ex.e_type == ET_DYN && !interpreter)
-			current->mm->brk = current->mm->start_brk =
-				ELF_ET_DYN_BASE;
+		    elf_ex->e_type == ET_DYN && !interpreter) {
+			mm->brk = mm->start_brk = ELF_ET_DYN_BASE;
+		}
 
-		current->mm->brk = current->mm->start_brk =
-			arch_randomize_brk(current->mm);
+		mm->brk = mm->start_brk = arch_randomize_brk(mm);
 #ifdef compat_brk_randomized
 		current->brk_randomized = 1;
 #endif
@@ -1154,12 +1287,11 @@ out_free_interp:
 	start_thread(regs, elf_entry, bprm->p);
 	retval = 0;
 out:
-	kfree(loc);
-out_ret:
 	return retval;
 
 	/* error cleanup */
 out_free_dentry:
+	kfree(interp_elf_ex);
 	kfree(interp_elf_phdata);
 	allow_write_access(interpreter);
 	if (interpreter)
@@ -1179,11 +1311,10 @@ static int load_elf_library(struct file *file)
 	unsigned long elf_bss, bss, len;
 	int retval, error, i, j;
 	struct elfhdr elf_ex;
-	loff_t pos = 0;
 
 	error = -ENOEXEC;
-	retval = kernel_read(file, &elf_ex, sizeof(elf_ex), &pos);
-	if (retval != sizeof(elf_ex))
+	retval = elf_read(file, &elf_ex, sizeof(elf_ex), 0);
+	if (retval < 0)
 		goto out;
 
 	if (memcmp(elf_ex.e_ident, ELFMAG, SELFMAG) != 0)
@@ -1208,9 +1339,8 @@ static int load_elf_library(struct file *file)
 
 	eppnt = elf_phdata;
 	error = -ENOEXEC;
-	pos =  elf_ex.e_phoff;
-	retval = kernel_read(file, eppnt, j, &pos);
-	if (retval != j)
+	retval = elf_read(file, eppnt, j, elf_ex.e_phoff);
+	if (retval < 0)
 		goto out_free_ph;
 
 	for (j = 0, i = 0; i<elf_ex.e_phnum; i++)
@@ -1320,7 +1450,7 @@ static unsigned long vma_dump_size(struct vm_area_struct *vma,
 	}
 
 	/* Hugetlb memory check */
-	if (vma->vm_flags & VM_HUGETLB) {
+	if (is_vm_hugetlb_page(vma)) {
 		if ((vma->vm_flags & VM_SHARED) && FILTER(HUGETLB_SHARED))
 			goto whole;
 		if (!(vma->vm_flags & VM_SHARED) && FILTER(HUGETLB_PRIVATE))
@@ -1358,7 +1488,6 @@ static unsigned long vma_dump_size(struct vm_area_struct *vma,
 	    vma->vm_pgoff == 0 && (vma->vm_flags & VM_READ)) {
 		u32 __user *header = (u32 __user *) vma->vm_start;
 		u32 word;
-		mm_segment_t fs = get_fs();
 		/*
 		 * Doing it this way gets the constant folded by GCC.
 		 */
@@ -1371,14 +1500,8 @@ static unsigned long vma_dump_size(struct vm_area_struct *vma,
 		magic.elfmag[EI_MAG1] = ELFMAG1;
 		magic.elfmag[EI_MAG2] = ELFMAG2;
 		magic.elfmag[EI_MAG3] = ELFMAG3;
-		/*
-		 * Switch to the user "segment" for get_user(),
-		 * then put back what elf_core_dump() had in place.
-		 */
-		set_fs(USER_DS);
 		if (unlikely(get_user(word, header)))
 			word = 0;
-		set_fs(fs);
 		if (word == magic.cmp)
 			return PAGE_SIZE;
 	}
@@ -1489,18 +1612,18 @@ static void fill_prstatus(struct elf_prstatus *prstatus,
 		 * group-wide total, not its individual thread total.
 		 */
 		thread_group_cputime(p, &cputime);
-		prstatus->pr_utime = ns_to_timeval(cputime.utime);
-		prstatus->pr_stime = ns_to_timeval(cputime.stime);
+		prstatus->pr_utime = ns_to_kernel_old_timeval(cputime.utime);
+		prstatus->pr_stime = ns_to_kernel_old_timeval(cputime.stime);
 	} else {
 		u64 utime, stime;
 
 		task_cputime(p, &utime, &stime);
-		prstatus->pr_utime = ns_to_timeval(utime);
-		prstatus->pr_stime = ns_to_timeval(stime);
+		prstatus->pr_utime = ns_to_kernel_old_timeval(utime);
+		prstatus->pr_stime = ns_to_kernel_old_timeval(stime);
 	}
 
-	prstatus->pr_cutime = ns_to_timeval(p->signal->cutime);
-	prstatus->pr_cstime = ns_to_timeval(p->signal->cstime);
+	prstatus->pr_cutime = ns_to_kernel_old_timeval(p->signal->cutime);
+	prstatus->pr_cstime = ns_to_kernel_old_timeval(p->signal->cstime);
 }
 
 static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p,
@@ -1559,10 +1682,7 @@ static void fill_auxv_note(struct memelfnote *note, struct mm_struct *mm)
 static void fill_siginfo_note(struct memelfnote *note, user_siginfo_t *csigdata,
 		const kernel_siginfo_t *siginfo)
 {
-	mm_segment_t old_fs = get_fs();
-	set_fs(KERNEL_DS);
-	copy_siginfo_to_user((user_siginfo_t __user *) csigdata, siginfo);
-	set_fs(old_fs);
+	copy_siginfo_to_external(csigdata, siginfo);
 	fill_note(note, "CORE", NT_SIGINFO, sizeof(*csigdata), csigdata);
 }
 
@@ -1580,6 +1700,7 @@ static void fill_siginfo_note(struct memelfnote *note, user_siginfo_t *csigdata,
  */
 static int fill_files_note(struct memelfnote *note)
 {
+	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned count, size, names_ofs, remaining, n;
 	user_long_t *data;
@@ -1587,7 +1708,7 @@ static int fill_files_note(struct memelfnote *note)
 	char *name_base, *name_curpos;
 
 	/* *Estimated* file count and total data size needed */
-	count = current->mm->map_count;
+	count = mm->map_count;
 	if (count > UINT_MAX / 64)
 		return -EINVAL;
 	size = count * 64;
@@ -1597,6 +1718,10 @@ static int fill_files_note(struct memelfnote *note)
 	if (size >= MAX_FILE_NOTE_SIZE) /* paranoia check */
 		return -EINVAL;
 	size = round_up(size, PAGE_SIZE);
+	/*
+	 * "size" can be 0 here legitimately.
+	 * Let it ENOMEM and omit NT_FILE section which will be empty anyway.
+	 */
 	data = kvmalloc(size, GFP_KERNEL);
 	if (ZERO_OR_NULL_PTR(data))
 		return -ENOMEM;
@@ -1605,7 +1730,7 @@ static int fill_files_note(struct memelfnote *note)
 	name_base = name_curpos = ((char *)data) + names_ofs;
 	remaining = size - names_ofs;
 	count = 0;
-	for (vma = current->mm->mmap; vma != NULL; vma = vma->vm_next) {
+	for (vma = mm->mmap; vma != NULL; vma = vma->vm_next) {
 		struct file *file;
 		const char *filename;
 
@@ -1639,10 +1764,10 @@ static int fill_files_note(struct memelfnote *note)
 	data[0] = count;
 	data[1] = PAGE_SIZE;
 	/*
-	 * Count usually is less than current->mm->map_count,
+	 * Count usually is less than mm->map_count,
 	 * we need to move filenames down.
 	 */
-	n = current->mm->map_count - count;
+	n = mm->map_count - count;
 	if (n != 0) {
 		unsigned shift_bytes = n * 3 * sizeof(data[0]);
 		memmove(name_base - shift_bytes, name_base,
@@ -1731,7 +1856,7 @@ static int fill_thread_core_info(struct elf_thread_core_info *t,
 		    (!regset->active || regset->active(t->task, regset) > 0)) {
 			int ret;
 			size_t size = regset_size(t->task, regset);
-			void *data = kmalloc(size, GFP_KERNEL);
+			void *data = kzalloc(size, GFP_KERNEL);
 			if (unlikely(!data))
 				return 0;
 			ret = regset->get(t->task, regset,
@@ -2184,11 +2309,10 @@ static void fill_extnum_info(struct elfhdr *elf, struct elf_shdr *shdr4extnum,
 static int elf_core_dump(struct coredump_params *cprm)
 {
 	int has_dumped = 0;
-	mm_segment_t fs;
 	int segs, i;
 	size_t vma_data_size = 0;
 	struct vm_area_struct *vma, *gate_vma;
-	struct elfhdr *elf = NULL;
+	struct elfhdr elf;
 	loff_t offset = 0, dataoff;
 	struct elf_note_info info = { };
 	struct elf_phdr *phdr4note = NULL;
@@ -2209,10 +2333,6 @@ static int elf_core_dump(struct coredump_params *cprm)
 	 * exists while dumping the mm->vm_next areas to the core file.
 	 */
   
-	/* alloc memory for large data structures: too large to be on stack */
-	elf = kmalloc(sizeof(*elf), GFP_KERNEL);
-	if (!elf)
-		goto out;
 	/*
 	 * The number of segs are recored into ELF header as 16bit value.
 	 * Please check DEFAULT_MAX_MAP_COUNT definition when you modify here.
@@ -2236,15 +2356,12 @@ static int elf_core_dump(struct coredump_params *cprm)
 	 * Collect all the non-memory information about the process for the
 	 * notes.  This also sets up the file header.
 	 */
-	if (!fill_note_info(elf, e_phnum, &info, cprm->siginfo, cprm->regs))
-		goto cleanup;
+	if (!fill_note_info(&elf, e_phnum, &info, cprm->siginfo, cprm->regs))
+		goto end_coredump;
 
 	has_dumped = 1;
 
-	fs = get_fs();
-	set_fs(KERNEL_DS);
-
-	offset += sizeof(*elf);				/* Elf header */
+	offset += sizeof(elf);				/* Elf header */
 	offset += segs * sizeof(struct elf_phdr);	/* Program headers */
 
 	/* Write notes phdr entry */
@@ -2263,11 +2380,13 @@ static int elf_core_dump(struct coredump_params *cprm)
 
 	dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE);
 
-	if (segs - 1 > ULONG_MAX / sizeof(*vma_filesz))
-		goto end_coredump;
+	/*
+	 * Zero vma process will get ZERO_SIZE_PTR here.
+	 * Let coredump continue for register state at least.
+	 */
 	vma_filesz = kvmalloc(array_size(sizeof(*vma_filesz), (segs - 1)),
 			      GFP_KERNEL);
-	if (ZERO_OR_NULL_PTR(vma_filesz))
+	if (!vma_filesz)
 		goto end_coredump;
 
 	for (i = 0, vma = first_vma(current, gate_vma); vma != NULL;
@@ -2287,12 +2406,12 @@ static int elf_core_dump(struct coredump_params *cprm)
 		shdr4extnum = kmalloc(sizeof(*shdr4extnum), GFP_KERNEL);
 		if (!shdr4extnum)
 			goto end_coredump;
-		fill_extnum_info(elf, shdr4extnum, e_shoff, segs);
+		fill_extnum_info(&elf, shdr4extnum, e_shoff, segs);
 	}
 
 	offset = dataoff;
 
-	if (!dump_emit(cprm, elf, sizeof(*elf)))
+	if (!dump_emit(cprm, &elf, sizeof(elf)))
 		goto end_coredump;
 
 	if (!dump_emit(cprm, phdr4note, sizeof(*phdr4note)))
@@ -2369,15 +2488,10 @@ static int elf_core_dump(struct coredump_params *cprm)
 	}
 
 end_coredump:
-	set_fs(fs);
-
-cleanup:
 	free_note_info(&info);
 	kfree(shdr4extnum);
 	kvfree(vma_filesz);
 	kfree(phdr4note);
-	kfree(elf);
-out:
 	return has_dumped;
 }
 
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index d86ebd0dcc3d..d9501a86cec9 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -1359,17 +1359,17 @@ static void fill_prstatus(struct elf_prstatus *prstatus,
 		 * group-wide total, not its individual thread total.
 		 */
 		thread_group_cputime(p, &cputime);
-		prstatus->pr_utime = ns_to_timeval(cputime.utime);
-		prstatus->pr_stime = ns_to_timeval(cputime.stime);
+		prstatus->pr_utime = ns_to_kernel_old_timeval(cputime.utime);
+		prstatus->pr_stime = ns_to_kernel_old_timeval(cputime.stime);
 	} else {
 		u64 utime, stime;
 
 		task_cputime(p, &utime, &stime);
-		prstatus->pr_utime = ns_to_timeval(utime);
-		prstatus->pr_stime = ns_to_timeval(stime);
+		prstatus->pr_utime = ns_to_kernel_old_timeval(utime);
+		prstatus->pr_stime = ns_to_kernel_old_timeval(stime);
 	}
-	prstatus->pr_cutime = ns_to_timeval(p->signal->cutime);
-	prstatus->pr_cstime = ns_to_timeval(p->signal->cstime);
+	prstatus->pr_cutime = ns_to_kernel_old_timeval(p->signal->cutime);
+	prstatus->pr_cstime = ns_to_kernel_old_timeval(p->signal->cstime);
 
 	prstatus->pr_exec_fdpic_loadmap = p->mm->context.exec_fdpic_loadmap;
 	prstatus->pr_interp_fdpic_loadmap = p->mm->context.interp_fdpic_loadmap;
@@ -1549,7 +1549,6 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
 {
 #define	NUM_NOTES	6
 	int has_dumped = 0;
-	mm_segment_t fs;
 	int segs;
 	int i;
 	struct vm_area_struct *vma;
@@ -1589,31 +1588,31 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
 	/* alloc memory for large data structures: too large to be on stack */
 	elf = kmalloc(sizeof(*elf), GFP_KERNEL);
 	if (!elf)
-		goto cleanup;
+		goto end_coredump;
 	prstatus = kzalloc(sizeof(*prstatus), GFP_KERNEL);
 	if (!prstatus)
-		goto cleanup;
+		goto end_coredump;
 	psinfo = kmalloc(sizeof(*psinfo), GFP_KERNEL);
 	if (!psinfo)
-		goto cleanup;
+		goto end_coredump;
 	notes = kmalloc_array(NUM_NOTES, sizeof(struct memelfnote),
 			      GFP_KERNEL);
 	if (!notes)
-		goto cleanup;
+		goto end_coredump;
 	fpu = kmalloc(sizeof(*fpu), GFP_KERNEL);
 	if (!fpu)
-		goto cleanup;
+		goto end_coredump;
 #ifdef ELF_CORE_COPY_XFPREGS
 	xfpu = kmalloc(sizeof(*xfpu), GFP_KERNEL);
 	if (!xfpu)
-		goto cleanup;
+		goto end_coredump;
 #endif
 
 	for (ct = current->mm->core_state->dumper.next;
 					ct; ct = ct->next) {
 		tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
 		if (!tmp)
-			goto cleanup;
+			goto end_coredump;
 
 		tmp->thread = ct->task;
 		list_add(&tmp->list, &thread_list);
@@ -1678,9 +1677,6 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
 			  "LINUX", ELF_CORE_XFPREG_TYPE, sizeof(*xfpu), xfpu);
 #endif
 
-	fs = get_fs();
-	set_fs(KERNEL_DS);
-
 	offset += sizeof(*elf);				/* Elf header */
 	offset += segs * sizeof(struct elf_phdr);	/* Program headers */
 
@@ -1788,9 +1784,6 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
 	}
 
 end_coredump:
-	set_fs(fs);
-
-cleanup:
 	while (!list_empty(&thread_list)) {
 		struct list_head *tmp = thread_list.next;
 		list_del(tmp);
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 9c073dbdc1b0..47860e589388 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -19,7 +19,6 @@
 #include <linux/module.h>
 #include <linux/blkpg.h>
 #include <linux/magic.h>
-#include <linux/dax.h>
 #include <linux/buffer_head.h>
 #include <linux/swap.h>
 #include <linux/pagevec.h>
@@ -34,6 +33,7 @@
 #include <linux/task_io_accounting_ops.h>
 #include <linux/falloc.h>
 #include <linux/uaccess.h>
+#include <linux/suspend.h>
 #include "internal.h"
 
 struct bdev_inode {
@@ -255,7 +255,7 @@ __blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter,
 			break;
 		if (!(iocb->ki_flags & IOCB_HIPRI) ||
 		    !blk_poll(bdev_get_queue(bdev), qc, true))
-			io_schedule();
+			blk_io_schedule();
 	}
 	__set_current_state(TASK_RUNNING);
 
@@ -449,7 +449,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
 
 		if (!(iocb->ki_flags & IOCB_HIPRI) ||
 		    !blk_poll(bdev_get_queue(bdev), qc, true))
-			io_schedule();
+			blk_io_schedule();
 	}
 	__set_current_state(TASK_RUNNING);
 
@@ -614,10 +614,9 @@ static int blkdev_readpage(struct file * file, struct page * page)
 	return block_read_full_page(page, blkdev_get_block);
 }
 
-static int blkdev_readpages(struct file *file, struct address_space *mapping,
-			struct list_head *pages, unsigned nr_pages)
+static void blkdev_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, blkdev_get_block);
+	mpage_readahead(rac, blkdev_get_block);
 }
 
 static int blkdev_write_begin(struct file *file, struct address_space *mapping,
@@ -672,7 +671,7 @@ int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 	 * i_mutex and doing so causes performance issues with concurrent
 	 * O_SYNC writers to a block device.
 	 */
-	error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL);
+	error = blkdev_issue_flush(bdev, GFP_KERNEL);
 	if (error == -EOPNOTSUPP)
 		error = 0;
 
@@ -713,7 +712,6 @@ int bdev_read_page(struct block_device *bdev, sector_t sector,
 	blk_queue_exit(bdev->bd_queue);
 	return result;
 }
-EXPORT_SYMBOL_GPL(bdev_read_page);
 
 /**
  * bdev_write_page() - Start writing a page to a block device
@@ -758,7 +756,6 @@ int bdev_write_page(struct block_device *bdev, sector_t sector,
 	blk_queue_exit(bdev->bd_queue);
 	return result;
 }
-EXPORT_SYMBOL_GPL(bdev_write_page);
 
 /*
  * pseudo-fs
@@ -882,21 +879,6 @@ static int bdev_set(struct inode *inode, void *data)
 
 static LIST_HEAD(all_bdevs);
 
-/*
- * If there is a bdev inode for this device, unhash it so that it gets evicted
- * as soon as last inode reference is dropped.
- */
-void bdev_unhash_inode(dev_t dev)
-{
-	struct inode *inode;
-
-	inode = ilookup5(blockdev_superblock, hash(dev), bdev_test, &dev);
-	if (inode) {
-		remove_inode_hash(inode);
-		iput(inode);
-	}
-}
-
 struct block_device *bdget(dev_t dev)
 {
 	struct block_device *bdev;
@@ -1403,11 +1385,7 @@ static void flush_disk(struct block_device *bdev, bool kill_dirty)
 		       "resized disk %s\n",
 		       bdev->bd_disk ? bdev->bd_disk->disk_name : "");
 	}
-
-	if (!bdev->bd_disk)
-		return;
-	if (disk_part_scan_enabled(bdev->bd_disk))
-		bdev->bd_invalidated = 1;
+	bdev->bd_invalidated = 1;
 }
 
 /**
@@ -1420,8 +1398,8 @@ static void flush_disk(struct block_device *bdev, bool kill_dirty)
  * and adjusts it if it differs. When shrinking the bdev size, its all caches
  * are freed.
  */
-void check_disk_size_change(struct gendisk *disk, struct block_device *bdev,
-		bool verbose)
+static void check_disk_size_change(struct gendisk *disk,
+		struct block_device *bdev, bool verbose)
 {
 	loff_t disk_size, bdev_size;
 
@@ -1437,6 +1415,7 @@ void check_disk_size_change(struct gendisk *disk, struct block_device *bdev,
 		if (bdev_size > disk_size)
 			flush_disk(bdev, false);
 	}
+	bdev->bd_invalidated = 0;
 }
 
 /**
@@ -1466,7 +1445,6 @@ int revalidate_disk(struct gendisk *disk)
 
 		mutex_lock(&bdev->bd_mutex);
 		check_disk_size_change(disk, bdev, ret == 0);
-		bdev->bd_invalidated = 0;
 		mutex_unlock(&bdev->bd_mutex);
 		bdput(bdev);
 	}
@@ -1512,6 +1490,57 @@ EXPORT_SYMBOL(bd_set_size);
 
 static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part);
 
+int bdev_disk_changed(struct block_device *bdev, bool invalidate)
+{
+	struct gendisk *disk = bdev->bd_disk;
+	int ret;
+
+	lockdep_assert_held(&bdev->bd_mutex);
+
+rescan:
+	ret = blk_drop_partitions(bdev);
+	if (ret)
+		return ret;
+
+	/*
+	 * Historically we only set the capacity to zero for devices that
+	 * support partitions (independ of actually having partitions created).
+	 * Doing that is rather inconsistent, but changing it broke legacy
+	 * udisks polling for legacy ide-cdrom devices.  Use the crude check
+	 * below to get the sane behavior for most device while not breaking
+	 * userspace for this particular setup.
+	 */
+	if (invalidate) {
+		if (disk_part_scan_enabled(disk) ||
+		    !(disk->flags & GENHD_FL_REMOVABLE))
+			set_capacity(disk, 0);
+	} else {
+		if (disk->fops->revalidate_disk)
+			disk->fops->revalidate_disk(disk);
+	}
+
+	check_disk_size_change(disk, bdev, !invalidate);
+
+	if (get_capacity(disk)) {
+		ret = blk_add_partitions(disk, bdev);
+		if (ret == -EAGAIN)
+			goto rescan;
+	} else if (invalidate) {
+		/*
+		 * Tell userspace that the media / partition table may have
+		 * changed.
+		 */
+		kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE);
+	}
+
+	return ret;
+}
+/*
+ * Only exported for for loop and dasd for historic reasons.  Don't use in new
+ * code!
+ */
+EXPORT_SYMBOL_GPL(bdev_disk_changed);
+
 /*
  * bd_mutex locking:
  *
@@ -1594,12 +1623,9 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 			 * The latter is necessary to prevent ghost
 			 * partitions on a removed medium.
 			 */
-			if (bdev->bd_invalidated) {
-				if (!ret)
-					rescan_partitions(disk, bdev);
-				else if (ret == -ENOMEDIUM)
-					invalidate_partitions(disk, bdev);
-			}
+			if (bdev->bd_invalidated &&
+			    (!ret || ret == -ENOMEDIUM))
+				bdev_disk_changed(bdev, ret == -ENOMEDIUM);
 
 			if (ret)
 				goto out_clear;
@@ -1632,12 +1658,9 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 			if (bdev->bd_disk->fops->open)
 				ret = bdev->bd_disk->fops->open(bdev, mode);
 			/* the same as first opener case, read comment there */
-			if (bdev->bd_invalidated) {
-				if (!ret)
-					rescan_partitions(bdev->bd_disk, bdev);
-				else if (ret == -ENOMEDIUM)
-					invalidate_partitions(bdev->bd_disk, bdev);
-			}
+			if (bdev->bd_invalidated &&
+			    (!ret || ret == -ENOMEDIUM))
+				bdev_disk_changed(bdev, ret == -ENOMEDIUM);
 			if (ret)
 				goto out_unlock_bdev;
 		}
@@ -1851,6 +1874,16 @@ static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part)
 	struct gendisk *disk = bdev->bd_disk;
 	struct block_device *victim = NULL;
 
+	/*
+	 * Sync early if it looks like we're the last one.  If someone else
+	 * opens the block device between now and the decrement of bd_openers
+	 * then we did a sync that we didn't need to, but that's not the end
+	 * of the world and we want to avoid long (could be several minute)
+	 * syncs while holding the mutex.
+	 */
+	if (bdev->bd_openers == 1)
+		sync_blockdev(bdev);
+
 	mutex_lock_nested(&bdev->bd_mutex, for_part);
 	if (for_part)
 		bdev->bd_part_count--;
@@ -1972,7 +2005,7 @@ ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	if (bdev_read_only(I_BDEV(bd_inode)))
 		return -EPERM;
 
-	if (IS_SWAPFILE(bd_inode))
+	if (IS_SWAPFILE(bd_inode) && !is_hibernate_resume_dev(bd_inode))
 		return -ETXTBSY;
 
 	if (!iov_iter_count(from))
@@ -2033,7 +2066,7 @@ static int blkdev_writepages(struct address_space *mapping,
 
 static const struct address_space_operations def_blk_aops = {
 	.readpage	= blkdev_readpage,
-	.readpages	= blkdev_readpages,
+	.readahead	= blkdev_readahead,
 	.writepage	= blkdev_writepage,
 	.write_begin	= blkdev_write_begin,
 	.write_end	= blkdev_write_end,
@@ -2131,18 +2164,6 @@ const struct file_operations def_blk_fops = {
 	.fallocate	= blkdev_fallocate,
 };
 
-int ioctl_by_bdev(struct block_device *bdev, unsigned cmd, unsigned long arg)
-{
-	int res;
-	mm_segment_t old_fs = get_fs();
-	set_fs(KERNEL_DS);
-	res = blkdev_ioctl(bdev, 0, cmd, arg);
-	set_fs(old_fs);
-	return res;
-}
-
-EXPORT_SYMBOL(ioctl_by_bdev);
-
 /**
  * lookup_bdev  - lookup a struct block_device by name
  * @pathname:	special file representing the block device
diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig
index 38651fae7f21..575636f6491e 100644
--- a/fs/btrfs/Kconfig
+++ b/fs/btrfs/Kconfig
@@ -5,6 +5,9 @@ config BTRFS_FS
 	select CRYPTO
 	select CRYPTO_CRC32C
 	select LIBCRC32C
+	select CRYPTO_XXHASH
+	select CRYPTO_SHA256
+	select CRYPTO_BLAKE2B
 	select ZLIB_INFLATE
 	select ZLIB_DEFLATE
 	select LZO_COMPRESS
diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile
index 82200dbca5ac..e738f6206ea5 100644
--- a/fs/btrfs/Makefile
+++ b/fs/btrfs/Makefile
@@ -11,7 +11,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
 	   compression.o delayed-ref.o relocation.o delayed-inode.o scrub.o \
 	   reada.o backref.o ulist.o qgroup.o send.o dev-replace.o raid56.o \
 	   uuid-tree.o props.o free-space-tree.o tree-checker.o space-info.o \
-	   block-rsv.o delalloc-space.o block-group.o
+	   block-rsv.o delalloc-space.o block-group.o discard.o reflink.o
 
 btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o
 btrfs-$(CONFIG_BTRFS_FS_CHECK_INTEGRITY) += check-integrity.o
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 2e9e13ffbd08..309516e6a968 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -53,24 +53,12 @@ struct btrfs_workqueue {
 	struct __btrfs_workqueue *high;
 };
 
-static void normal_work_helper(struct btrfs_work *work);
-
-#define BTRFS_WORK_HELPER(name)					\
-noinline_for_stack void btrfs_##name(struct work_struct *arg)		\
-{									\
-	struct btrfs_work *work = container_of(arg, struct btrfs_work,	\
-					       normal_work);		\
-	normal_work_helper(work);					\
-}
-
-struct btrfs_fs_info *
-btrfs_workqueue_owner(const struct __btrfs_workqueue *wq)
+struct btrfs_fs_info * __pure btrfs_workqueue_owner(const struct __btrfs_workqueue *wq)
 {
 	return wq->fs_info;
 }
 
-struct btrfs_fs_info *
-btrfs_work_owner(const struct btrfs_work *work)
+struct btrfs_fs_info * __pure btrfs_work_owner(const struct btrfs_work *work)
 {
 	return work->wq->fs_info;
 }
@@ -89,29 +77,6 @@ bool btrfs_workqueue_normal_congested(const struct btrfs_workqueue *wq)
 	return atomic_read(&wq->normal->pending) > wq->normal->thresh * 2;
 }
 
-BTRFS_WORK_HELPER(worker_helper);
-BTRFS_WORK_HELPER(delalloc_helper);
-BTRFS_WORK_HELPER(flush_delalloc_helper);
-BTRFS_WORK_HELPER(cache_helper);
-BTRFS_WORK_HELPER(submit_helper);
-BTRFS_WORK_HELPER(fixup_helper);
-BTRFS_WORK_HELPER(endio_helper);
-BTRFS_WORK_HELPER(endio_meta_helper);
-BTRFS_WORK_HELPER(endio_meta_write_helper);
-BTRFS_WORK_HELPER(endio_raid56_helper);
-BTRFS_WORK_HELPER(endio_repair_helper);
-BTRFS_WORK_HELPER(rmw_helper);
-BTRFS_WORK_HELPER(endio_write_helper);
-BTRFS_WORK_HELPER(freespace_write_helper);
-BTRFS_WORK_HELPER(delayed_meta_helper);
-BTRFS_WORK_HELPER(readahead_helper);
-BTRFS_WORK_HELPER(qgroup_rescan_helper);
-BTRFS_WORK_HELPER(extent_refs_helper);
-BTRFS_WORK_HELPER(scrub_helper);
-BTRFS_WORK_HELPER(scrubwrc_helper);
-BTRFS_WORK_HELPER(scrubnc_helper);
-BTRFS_WORK_HELPER(scrubparity_helper);
-
 static struct __btrfs_workqueue *
 __btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info, const char *name,
 			unsigned int flags, int limit_active, int thresh)
@@ -252,16 +217,16 @@ out:
 	}
 }
 
-static void run_ordered_work(struct __btrfs_workqueue *wq)
+static void run_ordered_work(struct __btrfs_workqueue *wq,
+			     struct btrfs_work *self)
 {
 	struct list_head *list = &wq->ordered_list;
 	struct btrfs_work *work;
 	spinlock_t *lock = &wq->list_lock;
 	unsigned long flags;
+	bool free_self = false;
 
 	while (1) {
-		void *wtag;
-
 		spin_lock_irqsave(lock, flags);
 		if (list_empty(list))
 			break;
@@ -287,22 +252,53 @@ static void run_ordered_work(struct __btrfs_workqueue *wq)
 		list_del(&work->ordered_list);
 		spin_unlock_irqrestore(lock, flags);
 
-		/*
-		 * We don't want to call the ordered free functions with the
-		 * lock held though. Save the work as tag for the trace event,
-		 * because the callback could free the structure.
-		 */
-		wtag = work;
-		work->ordered_free(work);
-		trace_btrfs_all_work_done(wq->fs_info, wtag);
+		if (work == self) {
+			/*
+			 * This is the work item that the worker is currently
+			 * executing.
+			 *
+			 * The kernel workqueue code guarantees non-reentrancy
+			 * of work items. I.e., if a work item with the same
+			 * address and work function is queued twice, the second
+			 * execution is blocked until the first one finishes. A
+			 * work item may be freed and recycled with the same
+			 * work function; the workqueue code assumes that the
+			 * original work item cannot depend on the recycled work
+			 * item in that case (see find_worker_executing_work()).
+			 *
+			 * Note that different types of Btrfs work can depend on
+			 * each other, and one type of work on one Btrfs
+			 * filesystem may even depend on the same type of work
+			 * on another Btrfs filesystem via, e.g., a loop device.
+			 * Therefore, we must not allow the current work item to
+			 * be recycled until we are really done, otherwise we
+			 * break the above assumption and can deadlock.
+			 */
+			free_self = true;
+		} else {
+			/*
+			 * We don't want to call the ordered free functions with
+			 * the lock held.
+			 */
+			work->ordered_free(work);
+			/* NB: work must not be dereferenced past this point. */
+			trace_btrfs_all_work_done(wq->fs_info, work);
+		}
 	}
 	spin_unlock_irqrestore(lock, flags);
+
+	if (free_self) {
+		self->ordered_free(self);
+		/* NB: self must not be dereferenced past this point. */
+		trace_btrfs_all_work_done(wq->fs_info, self);
+	}
 }
 
-static void normal_work_helper(struct btrfs_work *work)
+static void btrfs_work_helper(struct work_struct *normal_work)
 {
+	struct btrfs_work *work = container_of(normal_work, struct btrfs_work,
+					       normal_work);
 	struct __btrfs_workqueue *wq;
-	void *wtag;
 	int need_order = 0;
 
 	/*
@@ -316,29 +312,26 @@ static void normal_work_helper(struct btrfs_work *work)
 	if (work->ordered_func)
 		need_order = 1;
 	wq = work->wq;
-	/* Safe for tracepoints in case work gets freed by the callback */
-	wtag = work;
 
 	trace_btrfs_work_sched(work);
 	thresh_exec_hook(wq);
 	work->func(work);
 	if (need_order) {
 		set_bit(WORK_DONE_BIT, &work->flags);
-		run_ordered_work(wq);
+		run_ordered_work(wq, work);
+	} else {
+		/* NB: work must not be dereferenced past this point. */
+		trace_btrfs_all_work_done(wq->fs_info, work);
 	}
-	if (!need_order)
-		trace_btrfs_all_work_done(wq->fs_info, wtag);
 }
 
-void btrfs_init_work(struct btrfs_work *work, btrfs_work_func_t uniq_func,
-		     btrfs_func_t func,
-		     btrfs_func_t ordered_func,
-		     btrfs_func_t ordered_free)
+void btrfs_init_work(struct btrfs_work *work, btrfs_func_t func,
+		     btrfs_func_t ordered_func, btrfs_func_t ordered_free)
 {
 	work->func = func;
 	work->ordered_func = ordered_func;
 	work->ordered_free = ordered_free;
-	INIT_WORK(&work->normal_work, uniq_func);
+	INIT_WORK(&work->normal_work, btrfs_work_helper);
 	INIT_LIST_HEAD(&work->ordered_list);
 	work->flags = 0;
 }
@@ -402,3 +395,11 @@ void btrfs_set_work_high_priority(struct btrfs_work *work)
 {
 	set_bit(WORK_HIGH_PRIO_BIT, &work->flags);
 }
+
+void btrfs_flush_workqueue(struct btrfs_workqueue *wq)
+{
+	if (wq->high)
+		flush_workqueue(wq->high->normal_wq);
+
+	flush_workqueue(wq->normal->normal_wq);
+}
diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index 7861c9feba5f..3204daa51b95 100644
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -29,49 +29,21 @@ struct btrfs_work {
 	unsigned long flags;
 };
 
-#define BTRFS_WORK_HELPER_PROTO(name)					\
-void btrfs_##name(struct work_struct *arg)
-
-BTRFS_WORK_HELPER_PROTO(worker_helper);
-BTRFS_WORK_HELPER_PROTO(delalloc_helper);
-BTRFS_WORK_HELPER_PROTO(flush_delalloc_helper);
-BTRFS_WORK_HELPER_PROTO(cache_helper);
-BTRFS_WORK_HELPER_PROTO(submit_helper);
-BTRFS_WORK_HELPER_PROTO(fixup_helper);
-BTRFS_WORK_HELPER_PROTO(endio_helper);
-BTRFS_WORK_HELPER_PROTO(endio_meta_helper);
-BTRFS_WORK_HELPER_PROTO(endio_meta_write_helper);
-BTRFS_WORK_HELPER_PROTO(endio_raid56_helper);
-BTRFS_WORK_HELPER_PROTO(endio_repair_helper);
-BTRFS_WORK_HELPER_PROTO(rmw_helper);
-BTRFS_WORK_HELPER_PROTO(endio_write_helper);
-BTRFS_WORK_HELPER_PROTO(freespace_write_helper);
-BTRFS_WORK_HELPER_PROTO(delayed_meta_helper);
-BTRFS_WORK_HELPER_PROTO(readahead_helper);
-BTRFS_WORK_HELPER_PROTO(qgroup_rescan_helper);
-BTRFS_WORK_HELPER_PROTO(extent_refs_helper);
-BTRFS_WORK_HELPER_PROTO(scrub_helper);
-BTRFS_WORK_HELPER_PROTO(scrubwrc_helper);
-BTRFS_WORK_HELPER_PROTO(scrubnc_helper);
-BTRFS_WORK_HELPER_PROTO(scrubparity_helper);
-
-
 struct btrfs_workqueue *btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info,
 					      const char *name,
 					      unsigned int flags,
 					      int limit_active,
 					      int thresh);
-void btrfs_init_work(struct btrfs_work *work, btrfs_work_func_t helper,
-		     btrfs_func_t func,
-		     btrfs_func_t ordered_func,
-		     btrfs_func_t ordered_free);
+void btrfs_init_work(struct btrfs_work *work, btrfs_func_t func,
+		     btrfs_func_t ordered_func, btrfs_func_t ordered_free);
 void btrfs_queue_work(struct btrfs_workqueue *wq,
 		      struct btrfs_work *work);
 void btrfs_destroy_workqueue(struct btrfs_workqueue *wq);
 void btrfs_workqueue_set_max(struct btrfs_workqueue *wq, int max);
 void btrfs_set_work_high_priority(struct btrfs_work *work);
-struct btrfs_fs_info *btrfs_work_owner(const struct btrfs_work *work);
-struct btrfs_fs_info *btrfs_workqueue_owner(const struct __btrfs_workqueue *wq);
+struct btrfs_fs_info * __pure btrfs_work_owner(const struct btrfs_work *work);
+struct btrfs_fs_info * __pure btrfs_workqueue_owner(const struct __btrfs_workqueue *wq);
 bool btrfs_workqueue_normal_congested(const struct btrfs_workqueue *wq);
+void btrfs_flush_workqueue(struct btrfs_workqueue *wq);
 
 #endif
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index e5d85311d5d5..0cc02577577b 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -347,33 +347,10 @@ static int add_prelim_ref(const struct btrfs_fs_info *fs_info,
 		return -ENOMEM;
 
 	ref->root_id = root_id;
-	if (key) {
+	if (key)
 		ref->key_for_search = *key;
-		/*
-		 * We can often find data backrefs with an offset that is too
-		 * large (>= LLONG_MAX, maximum allowed file offset) due to
-		 * underflows when subtracting a file's offset with the data
-		 * offset of its corresponding extent data item. This can
-		 * happen for example in the clone ioctl.
-		 * So if we detect such case we set the search key's offset to
-		 * zero to make sure we will find the matching file extent item
-		 * at add_all_parents(), otherwise we will miss it because the
-		 * offset taken form the backref is much larger then the offset
-		 * of the file extent item. This can make us scan a very large
-		 * number of file extent items, but at least it will not make
-		 * us miss any.
-		 * This is an ugly workaround for a behaviour that should have
-		 * never existed, but it does and a fix for the clone ioctl
-		 * would touch a lot of places, cause backwards incompatibility
-		 * and would not fix the problem for extents cloned with older
-		 * kernels.
-		 */
-		if (ref->key_for_search.type == BTRFS_EXTENT_DATA_KEY &&
-		    ref->key_for_search.offset >= LLONG_MAX)
-			ref->key_for_search.offset = 0;
-	} else {
+	else
 		memset(&ref->key_for_search, 0, sizeof(ref->key_for_search));
-	}
 
 	ref->inode_list = NULL;
 	ref->level = level;
@@ -409,10 +386,36 @@ static int add_indirect_ref(const struct btrfs_fs_info *fs_info,
 			      wanted_disk_byte, count, sc, gfp_mask);
 }
 
+static int is_shared_data_backref(struct preftrees *preftrees, u64 bytenr)
+{
+	struct rb_node **p = &preftrees->direct.root.rb_root.rb_node;
+	struct rb_node *parent = NULL;
+	struct prelim_ref *ref = NULL;
+	struct prelim_ref target = {};
+	int result;
+
+	target.parent = bytenr;
+
+	while (*p) {
+		parent = *p;
+		ref = rb_entry(parent, struct prelim_ref, rbnode);
+		result = prelim_ref_compare(ref, &target);
+
+		if (result < 0)
+			p = &(*p)->rb_left;
+		else if (result > 0)
+			p = &(*p)->rb_right;
+		else
+			return 1;
+	}
+	return 0;
+}
+
 static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
-			   struct ulist *parents, struct prelim_ref *ref,
+			   struct ulist *parents,
+			   struct preftrees *preftrees, struct prelim_ref *ref,
 			   int level, u64 time_seq, const u64 *extent_item_pos,
-			   u64 total_refs, bool ignore_offset)
+			   bool ignore_offset)
 {
 	int ret = 0;
 	int slot;
@@ -424,6 +427,7 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
 	u64 disk_byte;
 	u64 wanted_disk_byte = ref->wanted_disk_byte;
 	u64 count = 0;
+	u64 data_offset;
 
 	if (level != 0) {
 		eb = path->nodes[level];
@@ -434,18 +438,26 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
 	}
 
 	/*
-	 * We normally enter this function with the path already pointing to
-	 * the first item to check. But sometimes, we may enter it with
-	 * slot==nritems. In that case, go to the next leaf before we continue.
+	 * 1. We normally enter this function with the path already pointing to
+	 *    the first item to check. But sometimes, we may enter it with
+	 *    slot == nritems.
+	 * 2. We are searching for normal backref but bytenr of this leaf
+	 *    matches shared data backref
+	 * 3. The leaf owner is not equal to the root we are searching
+	 *
+	 * For these cases, go to the next leaf before we continue.
 	 */
-	if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
+	eb = path->nodes[0];
+	if (path->slots[0] >= btrfs_header_nritems(eb) ||
+	    is_shared_data_backref(preftrees, eb->start) ||
+	    ref->root_id != btrfs_header_owner(eb)) {
 		if (time_seq == SEQ_LAST)
 			ret = btrfs_next_leaf(root, path);
 		else
 			ret = btrfs_next_old_leaf(root, path, time_seq);
 	}
 
-	while (!ret && count < total_refs) {
+	while (!ret && count < ref->count) {
 		eb = path->nodes[0];
 		slot = path->slots[0];
 
@@ -455,13 +467,31 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
 		    key.type != BTRFS_EXTENT_DATA_KEY)
 			break;
 
+		/*
+		 * We are searching for normal backref but bytenr of this leaf
+		 * matches shared data backref, OR
+		 * the leaf owner is not equal to the root we are searching for
+		 */
+		if (slot == 0 &&
+		    (is_shared_data_backref(preftrees, eb->start) ||
+		     ref->root_id != btrfs_header_owner(eb))) {
+			if (time_seq == SEQ_LAST)
+				ret = btrfs_next_leaf(root, path);
+			else
+				ret = btrfs_next_old_leaf(root, path, time_seq);
+			continue;
+		}
 		fi = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
 		disk_byte = btrfs_file_extent_disk_bytenr(eb, fi);
+		data_offset = btrfs_file_extent_offset(eb, fi);
 
 		if (disk_byte == wanted_disk_byte) {
 			eie = NULL;
 			old = NULL;
-			count++;
+			if (ref->key_for_search.offset == key.offset - data_offset)
+				count++;
+			else
+				goto next;
 			if (extent_item_pos) {
 				ret = check_extent_in_eb(&key, eb, fi,
 						*extent_item_pos,
@@ -502,9 +532,9 @@ next:
  */
 static int resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 				struct btrfs_path *path, u64 time_seq,
+				struct preftrees *preftrees,
 				struct prelim_ref *ref, struct ulist *parents,
-				const u64 *extent_item_pos, u64 total_refs,
-				bool ignore_offset)
+				const u64 *extent_item_pos, bool ignore_offset)
 {
 	struct btrfs_root *root;
 	struct btrfs_key root_key;
@@ -512,23 +542,25 @@ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 	int ret = 0;
 	int root_level;
 	int level = ref->level;
-	int index;
+	struct btrfs_key search_key = ref->key_for_search;
 
 	root_key.objectid = ref->root_id;
 	root_key.type = BTRFS_ROOT_ITEM_KEY;
 	root_key.offset = (u64)-1;
 
-	index = srcu_read_lock(&fs_info->subvol_srcu);
-
 	root = btrfs_get_fs_root(fs_info, &root_key, false);
 	if (IS_ERR(root)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
 		ret = PTR_ERR(root);
+		goto out_free;
+	}
+
+	if (!path->search_commit_root &&
+	    test_bit(BTRFS_ROOT_DELETING, &root->state)) {
+		ret = -ENOENT;
 		goto out;
 	}
 
 	if (btrfs_is_testing(fs_info)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
 		ret = -ENOENT;
 		goto out;
 	}
@@ -540,21 +572,36 @@ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 	else
 		root_level = btrfs_old_root_level(root, time_seq);
 
-	if (root_level + 1 == level) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
+	if (root_level + 1 == level)
 		goto out;
-	}
 
+	/*
+	 * We can often find data backrefs with an offset that is too large
+	 * (>= LLONG_MAX, maximum allowed file offset) due to underflows when
+	 * subtracting a file's offset with the data offset of its
+	 * corresponding extent data item. This can happen for example in the
+	 * clone ioctl.
+	 *
+	 * So if we detect such case we set the search key's offset to zero to
+	 * make sure we will find the matching file extent item at
+	 * add_all_parents(), otherwise we will miss it because the offset
+	 * taken form the backref is much larger then the offset of the file
+	 * extent item. This can make us scan a very large number of file
+	 * extent items, but at least it will not make us miss any.
+	 *
+	 * This is an ugly workaround for a behaviour that should have never
+	 * existed, but it does and a fix for the clone ioctl would touch a lot
+	 * of places, cause backwards incompatibility and would not fix the
+	 * problem for extents cloned with older kernels.
+	 */
+	if (search_key.type == BTRFS_EXTENT_DATA_KEY &&
+	    search_key.offset >= LLONG_MAX)
+		search_key.offset = 0;
 	path->lowest_level = level;
 	if (time_seq == SEQ_LAST)
-		ret = btrfs_search_slot(NULL, root, &ref->key_for_search, path,
-					0, 0);
+		ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0);
 	else
-		ret = btrfs_search_old_slot(root, &ref->key_for_search, path,
-					    time_seq);
-
-	/* root node has been locked, we can release @subvol_srcu safely here */
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
+		ret = btrfs_search_old_slot(root, &search_key, path, time_seq);
 
 	btrfs_debug(fs_info,
 		"search slot in root %llu (level %d, ref count %d) returned %d for key (%llu %u %llu)",
@@ -574,9 +621,11 @@ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 		eb = path->nodes[level];
 	}
 
-	ret = add_all_parents(root, path, parents, ref, level, time_seq,
-			      extent_item_pos, total_refs, ignore_offset);
+	ret = add_all_parents(root, path, parents, preftrees, ref, level,
+			      time_seq, extent_item_pos, ignore_offset);
 out:
+	btrfs_put_root(root);
+out_free:
 	path->lowest_level = 0;
 	btrfs_release_path(path);
 	return ret;
@@ -609,7 +658,7 @@ unode_aux_to_inode_list(struct ulist_node *node)
 static int resolve_indirect_refs(struct btrfs_fs_info *fs_info,
 				 struct btrfs_path *path, u64 time_seq,
 				 struct preftrees *preftrees,
-				 const u64 *extent_item_pos, u64 total_refs,
+				 const u64 *extent_item_pos,
 				 struct share_check *sc, bool ignore_offset)
 {
 	int err;
@@ -653,9 +702,9 @@ static int resolve_indirect_refs(struct btrfs_fs_info *fs_info,
 			ret = BACKREF_FOUND_SHARED;
 			goto out;
 		}
-		err = resolve_indirect_ref(fs_info, path, time_seq, ref,
-					   parents, extent_item_pos,
-					   total_refs, ignore_offset);
+		err = resolve_indirect_ref(fs_info, path, time_seq, preftrees,
+					   ref, parents, extent_item_pos,
+					   ignore_offset);
 		/*
 		 * we can only tolerate ENOENT,otherwise,we should catch error
 		 * and return directly.
@@ -758,8 +807,7 @@ static int add_missing_keys(struct btrfs_fs_info *fs_info,
  */
 static int add_delayed_refs(const struct btrfs_fs_info *fs_info,
 			    struct btrfs_delayed_ref_head *head, u64 seq,
-			    struct preftrees *preftrees, u64 *total_refs,
-			    struct share_check *sc)
+			    struct preftrees *preftrees, struct share_check *sc)
 {
 	struct btrfs_delayed_ref_node *node;
 	struct btrfs_delayed_extent_op *extent_op = head->extent_op;
@@ -793,7 +841,6 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info,
 		default:
 			BUG();
 		}
-		*total_refs += count;
 		switch (node->type) {
 		case BTRFS_TREE_BLOCK_REF_KEY: {
 			/* NORMAL INDIRECT METADATA backref */
@@ -876,7 +923,7 @@ out:
 static int add_inline_refs(const struct btrfs_fs_info *fs_info,
 			   struct btrfs_path *path, u64 bytenr,
 			   int *info_level, struct preftrees *preftrees,
-			   u64 *total_refs, struct share_check *sc)
+			   struct share_check *sc)
 {
 	int ret = 0;
 	int slot;
@@ -900,7 +947,6 @@ static int add_inline_refs(const struct btrfs_fs_info *fs_info,
 
 	ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
 	flags = btrfs_extent_flags(leaf, ei);
-	*total_refs += btrfs_extent_refs(leaf, ei);
 	btrfs_item_key_to_cpu(leaf, &found_key, slot);
 
 	ptr = (unsigned long)(ei + 1);
@@ -1125,8 +1171,6 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans,
 	struct prelim_ref *ref;
 	struct rb_node *node;
 	struct extent_inode_elem *eie = NULL;
-	/* total of both direct AND indirect refs! */
-	u64 total_refs = 0;
 	struct preftrees preftrees = {
 		.direct = PREFTREE_INIT,
 		.indirect = PREFTREE_INIT,
@@ -1195,7 +1239,7 @@ again:
 			}
 			spin_unlock(&delayed_refs->lock);
 			ret = add_delayed_refs(fs_info, head, time_seq,
-					       &preftrees, &total_refs, sc);
+					       &preftrees, sc);
 			mutex_unlock(&head->mutex);
 			if (ret)
 				goto out;
@@ -1216,8 +1260,7 @@ again:
 		    (key.type == BTRFS_EXTENT_ITEM_KEY ||
 		     key.type == BTRFS_METADATA_ITEM_KEY)) {
 			ret = add_inline_refs(fs_info, path, bytenr,
-					      &info_level, &preftrees,
-					      &total_refs, sc);
+					      &info_level, &preftrees, sc);
 			if (ret)
 				goto out;
 			ret = add_keyed_refs(fs_info, path, bytenr, info_level,
@@ -1236,7 +1279,7 @@ again:
 	WARN_ON(!RB_EMPTY_ROOT(&preftrees.indirect_missing_keys.root.rb_root));
 
 	ret = resolve_indirect_refs(fs_info, path, time_seq, &preftrees,
-				    extent_item_pos, total_refs, sc, ignore_offset);
+				    extent_item_pos, sc, ignore_offset);
 	if (ret)
 		goto out;
 
@@ -1362,10 +1405,10 @@ static void free_leaf_list(struct ulist *blocks)
  *
  * returns 0 on success, <0 on error
  */
-static int btrfs_find_all_leafs(struct btrfs_trans_handle *trans,
-				struct btrfs_fs_info *fs_info, u64 bytenr,
-				u64 time_seq, struct ulist **leafs,
-				const u64 *extent_item_pos, bool ignore_offset)
+int btrfs_find_all_leafs(struct btrfs_trans_handle *trans,
+			 struct btrfs_fs_info *fs_info, u64 bytenr,
+			 u64 time_seq, struct ulist **leafs,
+			 const u64 *extent_item_pos, bool ignore_offset)
 {
 	int ret;
 
diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h
index 777f61dc081e..723d6da99114 100644
--- a/fs/btrfs/backref.h
+++ b/fs/btrfs/backref.h
@@ -40,6 +40,10 @@ int iterate_inodes_from_logical(u64 logical, struct btrfs_fs_info *fs_info,
 
 int paths_from_inode(u64 inum, struct inode_fs_paths *ipath);
 
+int btrfs_find_all_leafs(struct btrfs_trans_handle *trans,
+			 struct btrfs_fs_info *fs_info, u64 bytenr,
+			 u64 time_seq, struct ulist **leafs,
+			 const u64 *extent_item_pos, bool ignore_offset);
 int btrfs_find_all_roots(struct btrfs_trans_handle *trans,
 			 struct btrfs_fs_info *fs_info, u64 bytenr,
 			 u64 time_seq, struct ulist **roots, bool ignore_offset);
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 670700cb1110..696f47103cfc 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -14,6 +14,8 @@
 #include "sysfs.h"
 #include "tree-log.h"
 #include "delalloc-space.h"
+#include "discard.h"
+#include "raid56.h"
 
 /*
  * Return target flags in extended format or 0 if restripe for this chunk_type
@@ -95,7 +97,7 @@ static u64 btrfs_reduce_alloc_profile(struct btrfs_fs_info *fs_info, u64 flags)
 	return extended_to_chunk(flags | allowed);
 }
 
-static u64 get_alloc_profile(struct btrfs_fs_info *fs_info, u64 orig_flags)
+u64 btrfs_get_alloc_profile(struct btrfs_fs_info *fs_info, u64 orig_flags)
 {
 	unsigned seq;
 	u64 flags;
@@ -115,23 +117,27 @@ static u64 get_alloc_profile(struct btrfs_fs_info *fs_info, u64 orig_flags)
 	return btrfs_reduce_alloc_profile(fs_info, flags);
 }
 
-u64 btrfs_get_alloc_profile(struct btrfs_fs_info *fs_info, u64 orig_flags)
-{
-	return get_alloc_profile(fs_info, orig_flags);
-}
-
-void btrfs_get_block_group(struct btrfs_block_group_cache *cache)
+void btrfs_get_block_group(struct btrfs_block_group *cache)
 {
 	atomic_inc(&cache->count);
 }
 
-void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
+void btrfs_put_block_group(struct btrfs_block_group *cache)
 {
 	if (atomic_dec_and_test(&cache->count)) {
 		WARN_ON(cache->pinned > 0);
 		WARN_ON(cache->reserved > 0);
 
 		/*
+		 * A block_group shouldn't be on the discard_list anymore.
+		 * Remove the block_group from the discard_list to prevent us
+		 * from causing a panic due to NULL pointer dereference.
+		 */
+		if (WARN_ON(!list_empty(&cache->discard_list)))
+			btrfs_discard_cancel_work(&cache->fs_info->discard_ctl,
+						  cache);
+
+		/*
 		 * If not empty, someone is still holding mutex of
 		 * full_stripe_lock, which can only be released by caller.
 		 * And it will definitely cause use-after-free when caller
@@ -149,22 +155,21 @@ void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
  * This adds the block group to the fs_info rb tree for the block group cache
  */
 static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
-				struct btrfs_block_group_cache *block_group)
+				       struct btrfs_block_group *block_group)
 {
 	struct rb_node **p;
 	struct rb_node *parent = NULL;
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 
 	spin_lock(&info->block_group_cache_lock);
 	p = &info->block_group_cache_tree.rb_node;
 
 	while (*p) {
 		parent = *p;
-		cache = rb_entry(parent, struct btrfs_block_group_cache,
-				 cache_node);
-		if (block_group->key.objectid < cache->key.objectid) {
+		cache = rb_entry(parent, struct btrfs_block_group, cache_node);
+		if (block_group->start < cache->start) {
 			p = &(*p)->rb_left;
-		} else if (block_group->key.objectid > cache->key.objectid) {
+		} else if (block_group->start > cache->start) {
 			p = &(*p)->rb_right;
 		} else {
 			spin_unlock(&info->block_group_cache_lock);
@@ -176,8 +181,8 @@ static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
 	rb_insert_color(&block_group->cache_node,
 			&info->block_group_cache_tree);
 
-	if (info->first_logical_byte > block_group->key.objectid)
-		info->first_logical_byte = block_group->key.objectid;
+	if (info->first_logical_byte > block_group->start)
+		info->first_logical_byte = block_group->start;
 
 	spin_unlock(&info->block_group_cache_lock);
 
@@ -188,10 +193,10 @@ static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
  * This will return the block group at or after bytenr if contains is 0, else
  * it will return the block group that contains the bytenr
  */
-static struct btrfs_block_group_cache *block_group_cache_tree_search(
+static struct btrfs_block_group *block_group_cache_tree_search(
 		struct btrfs_fs_info *info, u64 bytenr, int contains)
 {
-	struct btrfs_block_group_cache *cache, *ret = NULL;
+	struct btrfs_block_group *cache, *ret = NULL;
 	struct rb_node *n;
 	u64 end, start;
 
@@ -199,13 +204,12 @@ static struct btrfs_block_group_cache *block_group_cache_tree_search(
 	n = info->block_group_cache_tree.rb_node;
 
 	while (n) {
-		cache = rb_entry(n, struct btrfs_block_group_cache,
-				 cache_node);
-		end = cache->key.objectid + cache->key.offset - 1;
-		start = cache->key.objectid;
+		cache = rb_entry(n, struct btrfs_block_group, cache_node);
+		end = cache->start + cache->length - 1;
+		start = cache->start;
 
 		if (bytenr < start) {
-			if (!contains && (!ret || start < ret->key.objectid))
+			if (!contains && (!ret || start < ret->start))
 				ret = cache;
 			n = n->rb_left;
 		} else if (bytenr > start) {
@@ -221,8 +225,8 @@ static struct btrfs_block_group_cache *block_group_cache_tree_search(
 	}
 	if (ret) {
 		btrfs_get_block_group(ret);
-		if (bytenr == 0 && info->first_logical_byte > ret->key.objectid)
-			info->first_logical_byte = ret->key.objectid;
+		if (bytenr == 0 && info->first_logical_byte > ret->start)
+			info->first_logical_byte = ret->start;
 	}
 	spin_unlock(&info->block_group_cache_lock);
 
@@ -232,7 +236,7 @@ static struct btrfs_block_group_cache *block_group_cache_tree_search(
 /*
  * Return the block group that starts at or after bytenr
  */
-struct btrfs_block_group_cache *btrfs_lookup_first_block_group(
+struct btrfs_block_group *btrfs_lookup_first_block_group(
 		struct btrfs_fs_info *info, u64 bytenr)
 {
 	return block_group_cache_tree_search(info, bytenr, 0);
@@ -241,14 +245,14 @@ struct btrfs_block_group_cache *btrfs_lookup_first_block_group(
 /*
  * Return the block group that contains the given bytenr
  */
-struct btrfs_block_group_cache *btrfs_lookup_block_group(
+struct btrfs_block_group *btrfs_lookup_block_group(
 		struct btrfs_fs_info *info, u64 bytenr)
 {
 	return block_group_cache_tree_search(info, bytenr, 1);
 }
 
-struct btrfs_block_group_cache *btrfs_next_block_group(
-		struct btrfs_block_group_cache *cache)
+struct btrfs_block_group *btrfs_next_block_group(
+		struct btrfs_block_group *cache)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	struct rb_node *node;
@@ -257,7 +261,7 @@ struct btrfs_block_group_cache *btrfs_next_block_group(
 
 	/* If our block group was removed, we need a full search. */
 	if (RB_EMPTY_NODE(&cache->cache_node)) {
-		const u64 next_bytenr = cache->key.objectid + cache->key.offset;
+		const u64 next_bytenr = cache->start + cache->length;
 
 		spin_unlock(&fs_info->block_group_cache_lock);
 		btrfs_put_block_group(cache);
@@ -266,8 +270,7 @@ struct btrfs_block_group_cache *btrfs_next_block_group(
 	node = rb_next(&cache->cache_node);
 	btrfs_put_block_group(cache);
 	if (node) {
-		cache = rb_entry(node, struct btrfs_block_group_cache,
-				 cache_node);
+		cache = rb_entry(node, struct btrfs_block_group, cache_node);
 		btrfs_get_block_group(cache);
 	} else
 		cache = NULL;
@@ -277,7 +280,7 @@ struct btrfs_block_group_cache *btrfs_next_block_group(
 
 bool btrfs_inc_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
 {
-	struct btrfs_block_group_cache *bg;
+	struct btrfs_block_group *bg;
 	bool ret = true;
 
 	bg = btrfs_lookup_block_group(fs_info, bytenr);
@@ -300,7 +303,7 @@ bool btrfs_inc_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
 
 void btrfs_dec_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
 {
-	struct btrfs_block_group_cache *bg;
+	struct btrfs_block_group *bg;
 
 	bg = btrfs_lookup_block_group(fs_info, bytenr);
 	ASSERT(bg);
@@ -314,7 +317,7 @@ void btrfs_dec_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
 	btrfs_put_block_group(bg);
 }
 
-void btrfs_wait_nocow_writers(struct btrfs_block_group_cache *bg)
+void btrfs_wait_nocow_writers(struct btrfs_block_group *bg)
 {
 	wait_var_event(&bg->nocow_writers, !atomic_read(&bg->nocow_writers));
 }
@@ -322,7 +325,7 @@ void btrfs_wait_nocow_writers(struct btrfs_block_group_cache *bg)
 void btrfs_dec_block_group_reservations(struct btrfs_fs_info *fs_info,
 					const u64 start)
 {
-	struct btrfs_block_group_cache *bg;
+	struct btrfs_block_group *bg;
 
 	bg = btrfs_lookup_block_group(fs_info, start);
 	ASSERT(bg);
@@ -331,7 +334,7 @@ void btrfs_dec_block_group_reservations(struct btrfs_fs_info *fs_info,
 	btrfs_put_block_group(bg);
 }
 
-void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg)
+void btrfs_wait_block_group_reservations(struct btrfs_block_group *bg)
 {
 	struct btrfs_space_info *space_info = bg->space_info;
 
@@ -357,7 +360,7 @@ void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg)
 }
 
 struct btrfs_caching_control *btrfs_get_caching_control(
-		struct btrfs_block_group_cache *cache)
+		struct btrfs_block_group *cache)
 {
 	struct btrfs_caching_control *ctl;
 
@@ -392,7 +395,7 @@ void btrfs_put_caching_control(struct btrfs_caching_control *ctl)
  * Callers of this must check if cache->cached == BTRFS_CACHE_ERROR before using
  * any of the information in this block group.
  */
-void btrfs_wait_block_group_cache_progress(struct btrfs_block_group_cache *cache,
+void btrfs_wait_block_group_cache_progress(struct btrfs_block_group *cache,
 					   u64 num_bytes)
 {
 	struct btrfs_caching_control *caching_ctl;
@@ -401,13 +404,13 @@ void btrfs_wait_block_group_cache_progress(struct btrfs_block_group_cache *cache
 	if (!caching_ctl)
 		return;
 
-	wait_event(caching_ctl->wait, btrfs_block_group_cache_done(cache) ||
+	wait_event(caching_ctl->wait, btrfs_block_group_done(cache) ||
 		   (cache->free_space_ctl->free_space >= num_bytes));
 
 	btrfs_put_caching_control(caching_ctl);
 }
 
-int btrfs_wait_block_group_cache_done(struct btrfs_block_group_cache *cache)
+int btrfs_wait_block_group_cache_done(struct btrfs_block_group *cache)
 {
 	struct btrfs_caching_control *caching_ctl;
 	int ret = 0;
@@ -416,7 +419,7 @@ int btrfs_wait_block_group_cache_done(struct btrfs_block_group_cache *cache)
 	if (!caching_ctl)
 		return (cache->cached == BTRFS_CACHE_ERROR) ? -EIO : 0;
 
-	wait_event(caching_ctl->wait, btrfs_block_group_cache_done(cache));
+	wait_event(caching_ctl->wait, btrfs_block_group_done(cache));
 	if (cache->cached == BTRFS_CACHE_ERROR)
 		ret = -EIO;
 	btrfs_put_caching_control(caching_ctl);
@@ -424,11 +427,11 @@ int btrfs_wait_block_group_cache_done(struct btrfs_block_group_cache *cache)
 }
 
 #ifdef CONFIG_BTRFS_DEBUG
-static void fragment_free_space(struct btrfs_block_group_cache *block_group)
+static void fragment_free_space(struct btrfs_block_group *block_group)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
-	u64 start = block_group->key.objectid;
-	u64 len = block_group->key.offset;
+	u64 start = block_group->start;
+	u64 len = block_group->length;
 	u64 chunk = block_group->flags & BTRFS_BLOCK_GROUP_METADATA ?
 		fs_info->nodesize : fs_info->sectorsize;
 	u64 step = chunk << 1;
@@ -450,15 +453,14 @@ static void fragment_free_space(struct btrfs_block_group_cache *block_group)
  * used yet since their free space will be released as soon as the transaction
  * commits.
  */
-u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
-		       u64 start, u64 end)
+u64 add_new_free_space(struct btrfs_block_group *block_group, u64 start, u64 end)
 {
 	struct btrfs_fs_info *info = block_group->fs_info;
 	u64 extent_start, extent_end, size, total_added = 0;
 	int ret;
 
 	while (start < end) {
-		ret = find_first_extent_bit(info->pinned_extents, start,
+		ret = find_first_extent_bit(&info->excluded_extents, start,
 					    &extent_start, &extent_end,
 					    EXTENT_DIRTY | EXTENT_UPTODATE,
 					    NULL);
@@ -470,8 +472,8 @@ u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
 		} else if (extent_start > start && extent_start < end) {
 			size = extent_start - start;
 			total_added += size;
-			ret = btrfs_add_free_space(block_group, start,
-						   size);
+			ret = btrfs_add_free_space_async_trimmed(block_group,
+								 start, size);
 			BUG_ON(ret); /* -ENOMEM or logic error */
 			start = extent_end + 1;
 		} else {
@@ -482,7 +484,8 @@ u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
 	if (start < end) {
 		size = end - start;
 		total_added += size;
-		ret = btrfs_add_free_space(block_group, start, size);
+		ret = btrfs_add_free_space_async_trimmed(block_group, start,
+							 size);
 		BUG_ON(ret); /* -ENOMEM or logic error */
 	}
 
@@ -491,7 +494,7 @@ u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
 
 static int load_extent_tree_free(struct btrfs_caching_control *caching_ctl)
 {
-	struct btrfs_block_group_cache *block_group = caching_ctl->block_group;
+	struct btrfs_block_group *block_group = caching_ctl->block_group;
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
 	struct btrfs_root *extent_root = fs_info->extent_root;
 	struct btrfs_path *path;
@@ -507,7 +510,7 @@ static int load_extent_tree_free(struct btrfs_caching_control *caching_ctl)
 	if (!path)
 		return -ENOMEM;
 
-	last = max_t(u64, block_group->key.objectid, BTRFS_SUPER_INFO_OFFSET);
+	last = max_t(u64, block_group->start, BTRFS_SUPER_INFO_OFFSET);
 
 #ifdef CONFIG_BTRFS_DEBUG
 	/*
@@ -587,13 +590,12 @@ next:
 			goto next;
 		}
 
-		if (key.objectid < block_group->key.objectid) {
+		if (key.objectid < block_group->start) {
 			path->slots[0]++;
 			continue;
 		}
 
-		if (key.objectid >= block_group->key.objectid +
-		    block_group->key.offset)
+		if (key.objectid >= block_group->start + block_group->length)
 			break;
 
 		if (key.type == BTRFS_EXTENT_ITEM_KEY ||
@@ -617,8 +619,7 @@ next:
 	ret = 0;
 
 	total_found += add_new_free_space(block_group, last,
-					  block_group->key.objectid +
-					  block_group->key.offset);
+				block_group->start + block_group->length);
 	caching_ctl->progress = (u64)-1;
 
 out:
@@ -628,7 +629,7 @@ out:
 
 static noinline void caching_thread(struct btrfs_work *work)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_caching_control *caching_ctl;
 	int ret;
@@ -656,8 +657,7 @@ static noinline void caching_thread(struct btrfs_work *work)
 
 		spin_lock(&block_group->space_info->lock);
 		spin_lock(&block_group->lock);
-		bytes_used = block_group->key.offset -
-			btrfs_block_group_used(&block_group->item);
+		bytes_used = block_group->length - block_group->used;
 		block_group->space_info->bytes_used += bytes_used >> 1;
 		spin_unlock(&block_group->lock);
 		spin_unlock(&block_group->space_info->lock);
@@ -677,8 +677,7 @@ static noinline void caching_thread(struct btrfs_work *work)
 	btrfs_put_block_group(block_group);
 }
 
-int btrfs_cache_block_group(struct btrfs_block_group_cache *cache,
-			    int load_cache_only)
+int btrfs_cache_block_group(struct btrfs_block_group *cache, int load_cache_only)
 {
 	DEFINE_WAIT(wait);
 	struct btrfs_fs_info *fs_info = cache->fs_info;
@@ -693,10 +692,9 @@ int btrfs_cache_block_group(struct btrfs_block_group_cache *cache,
 	mutex_init(&caching_ctl->mutex);
 	init_waitqueue_head(&caching_ctl->wait);
 	caching_ctl->block_group = cache;
-	caching_ctl->progress = cache->key.objectid;
+	caching_ctl->progress = cache->start;
 	refcount_set(&caching_ctl->count, 1);
-	btrfs_init_work(&caching_ctl->work, btrfs_cache_helper,
-			caching_thread, NULL, NULL);
+	btrfs_init_work(&caching_ctl->work, caching_thread, NULL, NULL);
 
 	spin_lock(&cache->lock);
 	/*
@@ -763,8 +761,7 @@ int btrfs_cache_block_group(struct btrfs_block_group_cache *cache,
 
 			spin_lock(&cache->space_info->lock);
 			spin_lock(&cache->lock);
-			bytes_used = cache->key.offset -
-				btrfs_block_group_used(&cache->item);
+			bytes_used = cache->length - cache->used;
 			cache->space_info->bytes_used += bytes_used >> 1;
 			spin_unlock(&cache->lock);
 			spin_unlock(&cache->space_info->lock);
@@ -833,27 +830,36 @@ static void clear_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
  *
  * - RAID56 - in case there's neither RAID5 nor RAID6 profile block group
  *            in the whole filesystem
+ *
+ * - RAID1C34 - same as above for RAID1C3 and RAID1C4 block groups
  */
 static void clear_incompat_bg_bits(struct btrfs_fs_info *fs_info, u64 flags)
 {
-	if (flags & BTRFS_BLOCK_GROUP_RAID56_MASK) {
+	bool found_raid56 = false;
+	bool found_raid1c34 = false;
+
+	if ((flags & BTRFS_BLOCK_GROUP_RAID56_MASK) ||
+	    (flags & BTRFS_BLOCK_GROUP_RAID1C3) ||
+	    (flags & BTRFS_BLOCK_GROUP_RAID1C4)) {
 		struct list_head *head = &fs_info->space_info;
 		struct btrfs_space_info *sinfo;
 
 		list_for_each_entry_rcu(sinfo, head, list) {
-			bool found = false;
-
 			down_read(&sinfo->groups_sem);
 			if (!list_empty(&sinfo->block_groups[BTRFS_RAID_RAID5]))
-				found = true;
+				found_raid56 = true;
 			if (!list_empty(&sinfo->block_groups[BTRFS_RAID_RAID6]))
-				found = true;
+				found_raid56 = true;
+			if (!list_empty(&sinfo->block_groups[BTRFS_RAID_RAID1C3]))
+				found_raid1c34 = true;
+			if (!list_empty(&sinfo->block_groups[BTRFS_RAID_RAID1C4]))
+				found_raid1c34 = true;
 			up_read(&sinfo->groups_sem);
-
-			if (found)
-				return;
 		}
-		btrfs_clear_fs_incompat(fs_info, RAID56);
+		if (!found_raid56)
+			btrfs_clear_fs_incompat(fs_info, RAID56);
+		if (!found_raid1c34)
+			btrfs_clear_fs_incompat(fs_info, RAID1C34);
 	}
 }
 
@@ -863,7 +869,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root = fs_info->extent_root;
 	struct btrfs_path *path;
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_free_cluster *cluster;
 	struct btrfs_root *tree_root = fs_info->tree_root;
 	struct btrfs_key key;
@@ -886,10 +892,9 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	 * remove it.
 	 */
 	btrfs_free_excluded_extents(block_group);
-	btrfs_free_ref_tree_range(fs_info, block_group->key.objectid,
-				  block_group->key.offset);
+	btrfs_free_ref_tree_range(fs_info, block_group->start,
+				  block_group->length);
 
-	memcpy(&key, &block_group->key, sizeof(key));
 	index = btrfs_bg_flags_to_raid_index(block_group->flags);
 	factor = btrfs_bg_type_to_factor(block_group->flags);
 
@@ -911,7 +916,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	path = btrfs_alloc_path();
 	if (!path) {
 		ret = -ENOMEM;
-		goto out;
+		goto out_put_group;
 	}
 
 	/*
@@ -949,7 +954,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 		ret = btrfs_orphan_add(trans, BTRFS_I(inode));
 		if (ret) {
 			btrfs_add_delayed_iput(inode);
-			goto out;
+			goto out_put_group;
 		}
 		clear_nlink(inode);
 		/* One for the block groups ref */
@@ -967,18 +972,18 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	}
 
 	key.objectid = BTRFS_FREE_SPACE_OBJECTID;
-	key.offset = block_group->key.objectid;
 	key.type = 0;
+	key.offset = block_group->start;
 
 	ret = btrfs_search_slot(trans, tree_root, &key, path, -1, 1);
 	if (ret < 0)
-		goto out;
+		goto out_put_group;
 	if (ret > 0)
 		btrfs_release_path(path);
 	if (ret == 0) {
 		ret = btrfs_del_item(trans, tree_root, path);
 		if (ret)
-			goto out;
+			goto out_put_group;
 		btrfs_release_path(path);
 	}
 
@@ -987,7 +992,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 		 &fs_info->block_group_cache_tree);
 	RB_CLEAR_NODE(&block_group->cache_node);
 
-	if (fs_info->first_logical_byte == block_group->key.objectid)
+	if (fs_info->first_logical_byte == block_group->start)
 		fs_info->first_logical_byte = (u64)-1;
 	spin_unlock(&fs_info->block_group_cache_lock);
 
@@ -1048,19 +1053,21 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 
 	if (btrfs_test_opt(fs_info, ENOSPC_DEBUG)) {
 		WARN_ON(block_group->space_info->total_bytes
-			< block_group->key.offset);
+			< block_group->length);
 		WARN_ON(block_group->space_info->bytes_readonly
-			< block_group->key.offset);
+			< block_group->length);
 		WARN_ON(block_group->space_info->disk_total
-			< block_group->key.offset * factor);
+			< block_group->length * factor);
 	}
-	block_group->space_info->total_bytes -= block_group->key.offset;
-	block_group->space_info->bytes_readonly -= block_group->key.offset;
-	block_group->space_info->disk_total -= block_group->key.offset * factor;
+	block_group->space_info->total_bytes -= block_group->length;
+	block_group->space_info->bytes_readonly -= block_group->length;
+	block_group->space_info->disk_total -= block_group->length * factor;
 
 	spin_unlock(&block_group->space_info->lock);
 
-	memcpy(&key, &block_group->key, sizeof(key));
+	key.objectid = block_group->start;
+	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+	key.offset = block_group->length;
 
 	mutex_lock(&fs_info->chunk_mutex);
 	spin_lock(&block_group->lock);
@@ -1095,9 +1102,9 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 
 	ret = remove_block_group_free_space(trans, block_group);
 	if (ret)
-		goto out;
+		goto out_put_group;
 
-	btrfs_put_block_group(block_group);
+	/* Once for the block groups rbtree */
 	btrfs_put_block_group(block_group);
 
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
@@ -1120,6 +1127,10 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 		/* once for the tree */
 		free_extent_map(em);
 	}
+
+out_put_group:
+	/* Once for the lookup reference */
+	btrfs_put_block_group(block_group);
 out:
 	if (remove_rsv)
 		btrfs_delayed_refs_rsv_release(fs_info, 1);
@@ -1180,26 +1191,12 @@ struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
  * data in this block group. That check should be done by relocation routine,
  * not this function.
  */
-static int inc_block_group_ro(struct btrfs_block_group_cache *cache, int force)
+static int inc_block_group_ro(struct btrfs_block_group *cache, int force)
 {
 	struct btrfs_space_info *sinfo = cache->space_info;
 	u64 num_bytes;
-	u64 sinfo_used;
-	u64 min_allocable_bytes;
 	int ret = -ENOSPC;
 
-	/*
-	 * We need some metadata space and system metadata space for
-	 * allocating chunks in some corner cases until we force to set
-	 * it to be readonly.
-	 */
-	if ((sinfo->flags &
-	     (BTRFS_BLOCK_GROUP_SYSTEM | BTRFS_BLOCK_GROUP_METADATA)) &&
-	    !force)
-		min_allocable_bytes = SZ_1M;
-	else
-		min_allocable_bytes = 0;
-
 	spin_lock(&sinfo->lock);
 	spin_lock(&cache->lock);
 
@@ -1209,47 +1206,115 @@ static int inc_block_group_ro(struct btrfs_block_group_cache *cache, int force)
 		goto out;
 	}
 
-	num_bytes = cache->key.offset - cache->reserved - cache->pinned -
-		    cache->bytes_super - btrfs_block_group_used(&cache->item);
-	sinfo_used = btrfs_space_info_used(sinfo, true);
+	num_bytes = cache->length - cache->reserved - cache->pinned -
+		    cache->bytes_super - cache->used;
 
 	/*
-	 * sinfo_used + num_bytes should always <= sinfo->total_bytes.
-	 *
-	 * Here we make sure if we mark this bg RO, we still have enough
-	 * free space as buffer (if min_allocable_bytes is not 0).
+	 * Data never overcommits, even in mixed mode, so do just the straight
+	 * check of left over space in how much we have allocated.
 	 */
-	if (sinfo_used + num_bytes + min_allocable_bytes <=
-	    sinfo->total_bytes) {
+	if (force) {
+		ret = 0;
+	} else if (sinfo->flags & BTRFS_BLOCK_GROUP_DATA) {
+		u64 sinfo_used = btrfs_space_info_used(sinfo, true);
+
+		/*
+		 * Here we make sure if we mark this bg RO, we still have enough
+		 * free space as buffer.
+		 */
+		if (sinfo_used + num_bytes <= sinfo->total_bytes)
+			ret = 0;
+	} else {
+		/*
+		 * We overcommit metadata, so we need to do the
+		 * btrfs_can_overcommit check here, and we need to pass in
+		 * BTRFS_RESERVE_NO_FLUSH to give ourselves the most amount of
+		 * leeway to allow us to mark this block group as read only.
+		 */
+		if (btrfs_can_overcommit(cache->fs_info, sinfo, num_bytes,
+					 BTRFS_RESERVE_NO_FLUSH))
+			ret = 0;
+	}
+
+	if (!ret) {
 		sinfo->bytes_readonly += num_bytes;
 		cache->ro++;
 		list_add_tail(&cache->ro_list, &sinfo->ro_bgs);
-		ret = 0;
 	}
 out:
 	spin_unlock(&cache->lock);
 	spin_unlock(&sinfo->lock);
 	if (ret == -ENOSPC && btrfs_test_opt(cache->fs_info, ENOSPC_DEBUG)) {
 		btrfs_info(cache->fs_info,
-			"unable to make block group %llu ro",
-			cache->key.objectid);
-		btrfs_info(cache->fs_info,
-			"sinfo_used=%llu bg_num_bytes=%llu min_allocable=%llu",
-			sinfo_used, num_bytes, min_allocable_bytes);
+			"unable to make block group %llu ro", cache->start);
 		btrfs_dump_space_info(cache->fs_info, cache->space_info, 0, 0);
 	}
 	return ret;
 }
 
+static bool clean_pinned_extents(struct btrfs_trans_handle *trans,
+				 struct btrfs_block_group *bg)
+{
+	struct btrfs_fs_info *fs_info = bg->fs_info;
+	struct btrfs_transaction *prev_trans = NULL;
+	const u64 start = bg->start;
+	const u64 end = start + bg->length - 1;
+	int ret;
+
+	spin_lock(&fs_info->trans_lock);
+	if (trans->transaction->list.prev != &fs_info->trans_list) {
+		prev_trans = list_last_entry(&trans->transaction->list,
+					     struct btrfs_transaction, list);
+		refcount_inc(&prev_trans->use_count);
+	}
+	spin_unlock(&fs_info->trans_lock);
+
+	/*
+	 * Hold the unused_bg_unpin_mutex lock to avoid racing with
+	 * btrfs_finish_extent_commit(). If we are at transaction N, another
+	 * task might be running finish_extent_commit() for the previous
+	 * transaction N - 1, and have seen a range belonging to the block
+	 * group in pinned_extents before we were able to clear the whole block
+	 * group range from pinned_extents. This means that task can lookup for
+	 * the block group after we unpinned it from pinned_extents and removed
+	 * it, leading to a BUG_ON() at unpin_extent_range().
+	 */
+	mutex_lock(&fs_info->unused_bg_unpin_mutex);
+	if (prev_trans) {
+		ret = clear_extent_bits(&prev_trans->pinned_extents, start, end,
+					EXTENT_DIRTY);
+		if (ret)
+			goto err;
+	}
+
+	ret = clear_extent_bits(&trans->transaction->pinned_extents, start, end,
+				EXTENT_DIRTY);
+	if (ret)
+		goto err;
+	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+	if (prev_trans)
+		btrfs_put_transaction(prev_trans);
+
+	return true;
+
+err:
+	mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+	if (prev_trans)
+		btrfs_put_transaction(prev_trans);
+	btrfs_dec_block_group_ro(bg);
+	return false;
+}
+
 /*
  * Process the unused_bgs list and remove any that don't have any allocated
  * space inside of them.
  */
 void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_space_info *space_info;
 	struct btrfs_trans_handle *trans;
+	const bool async_trim_enabled = btrfs_test_opt(fs_info, DISCARD_ASYNC);
 	int ret = 0;
 
 	if (!test_bit(BTRFS_FS_OPEN, &fs_info->flags))
@@ -1257,11 +1322,10 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 	spin_lock(&fs_info->unused_bgs_lock);
 	while (!list_empty(&fs_info->unused_bgs)) {
-		u64 start, end;
 		int trimming;
 
 		block_group = list_first_entry(&fs_info->unused_bgs,
-					       struct btrfs_block_group_cache,
+					       struct btrfs_block_group,
 					       bg_list);
 		list_del_init(&block_group->bg_list);
 
@@ -1273,14 +1337,31 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		}
 		spin_unlock(&fs_info->unused_bgs_lock);
 
+		btrfs_discard_cancel_work(&fs_info->discard_ctl, block_group);
+
 		mutex_lock(&fs_info->delete_unused_bgs_mutex);
 
 		/* Don't want to race with allocators so take the groups_sem */
 		down_write(&space_info->groups_sem);
+
+		/*
+		 * Async discard moves the final block group discard to be prior
+		 * to the unused_bgs code path.  Therefore, if it's not fully
+		 * trimmed, punt it back to the async discard lists.
+		 */
+		if (btrfs_test_opt(fs_info, DISCARD_ASYNC) &&
+		    !btrfs_is_free_space_trimmed(block_group)) {
+			trace_btrfs_skip_unused_block_group(block_group);
+			up_write(&space_info->groups_sem);
+			/* Requeue if we failed because of async discard */
+			btrfs_discard_queue_work(&fs_info->discard_ctl,
+						 block_group);
+			goto next;
+		}
+
 		spin_lock(&block_group->lock);
 		if (block_group->reserved || block_group->pinned ||
-		    btrfs_block_group_used(&block_group->item) ||
-		    block_group->ro ||
+		    block_group->used || block_group->ro ||
 		    list_is_singular(&block_group->list)) {
 			/*
 			 * We want to bail if we made new allocations or have
@@ -1308,7 +1389,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		 * properly if we fail to join the transaction.
 		 */
 		trans = btrfs_start_trans_remove_block_group(fs_info,
-						     block_group->key.objectid);
+						     block_group->start);
 		if (IS_ERR(trans)) {
 			btrfs_dec_block_group_ro(block_group);
 			ret = PTR_ERR(trans);
@@ -1319,35 +1400,25 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		 * We could have pending pinned extents for this block group,
 		 * just delete them, we don't care about them anymore.
 		 */
-		start = block_group->key.objectid;
-		end = start + block_group->key.offset - 1;
+		if (!clean_pinned_extents(trans, block_group))
+			goto end_trans;
+
 		/*
-		 * Hold the unused_bg_unpin_mutex lock to avoid racing with
-		 * btrfs_finish_extent_commit(). If we are at transaction N,
-		 * another task might be running finish_extent_commit() for the
-		 * previous transaction N - 1, and have seen a range belonging
-		 * to the block group in freed_extents[] before we were able to
-		 * clear the whole block group range from freed_extents[]. This
-		 * means that task can lookup for the block group after we
-		 * unpinned it from freed_extents[] and removed it, leading to
-		 * a BUG_ON() at btrfs_unpin_extent_range().
+		 * At this point, the block_group is read only and should fail
+		 * new allocations.  However, btrfs_finish_extent_commit() can
+		 * cause this block_group to be placed back on the discard
+		 * lists because now the block_group isn't fully discarded.
+		 * Bail here and try again later after discarding everything.
 		 */
-		mutex_lock(&fs_info->unused_bg_unpin_mutex);
-		ret = clear_extent_bits(&fs_info->freed_extents[0], start, end,
-				  EXTENT_DIRTY);
-		if (ret) {
-			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+		spin_lock(&fs_info->discard_ctl.lock);
+		if (!list_empty(&block_group->discard_list)) {
+			spin_unlock(&fs_info->discard_ctl.lock);
 			btrfs_dec_block_group_ro(block_group);
+			btrfs_discard_queue_work(&fs_info->discard_ctl,
+						 block_group);
 			goto end_trans;
 		}
-		ret = clear_extent_bits(&fs_info->freed_extents[1], start, end,
-				  EXTENT_DIRTY);
-		if (ret) {
-			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
-			btrfs_dec_block_group_ro(block_group);
-			goto end_trans;
-		}
-		mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+		spin_unlock(&fs_info->discard_ctl.lock);
 
 		/* Reset pinned so btrfs_put_block_group doesn't complain */
 		spin_lock(&space_info->lock);
@@ -1364,8 +1435,18 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		spin_unlock(&block_group->lock);
 		spin_unlock(&space_info->lock);
 
+		/*
+		 * The normal path here is an unused block group is passed here,
+		 * then trimming is handled in the transaction commit path.
+		 * Async discard interposes before this to do the trimming
+		 * before coming down the unused block group path as trimming
+		 * will no longer be done later in the transaction commit path.
+		 */
+		if (!async_trim_enabled && btrfs_test_opt(fs_info, DISCARD_ASYNC))
+			goto flip_async;
+
 		/* DISCARD can flip during remount */
-		trimming = btrfs_test_opt(fs_info, DISCARD);
+		trimming = btrfs_test_opt(fs_info, DISCARD_SYNC);
 
 		/* Implicit trim during transaction commit. */
 		if (trimming)
@@ -1375,7 +1456,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		 * Btrfs_remove_chunk will abort the transaction if things go
 		 * horribly wrong.
 		 */
-		ret = btrfs_remove_chunk(trans, block_group->key.objectid);
+		ret = btrfs_remove_chunk(trans, block_group->start);
 
 		if (ret) {
 			if (trimming)
@@ -1408,9 +1489,16 @@ next:
 		spin_lock(&fs_info->unused_bgs_lock);
 	}
 	spin_unlock(&fs_info->unused_bgs_lock);
+	return;
+
+flip_async:
+	btrfs_end_transaction(trans);
+	mutex_unlock(&fs_info->delete_unused_bgs_mutex);
+	btrfs_put_block_group(block_group);
+	btrfs_discard_punt_unused_bgs_list(fs_info);
 }
 
-void btrfs_mark_bg_unused(struct btrfs_block_group_cache *bg)
+void btrfs_mark_bg_unused(struct btrfs_block_group *bg)
 {
 	struct btrfs_fs_info *fs_info = bg->fs_info;
 
@@ -1478,7 +1566,7 @@ static int find_first_block_group(struct btrfs_fs_info *fs_info,
 				read_extent_buffer(leaf, &bg,
 					btrfs_item_ptr_offset(leaf, slot),
 					sizeof(bg));
-				flags = btrfs_block_group_flags(&bg) &
+				flags = btrfs_stack_block_group_flags(&bg) &
 					BTRFS_BLOCK_GROUP_TYPE_MASK;
 
 				if (flags != (em->map_lookup->type &
@@ -1518,7 +1606,103 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
 	write_sequnlock(&fs_info->profiles_lock);
 }
 
-static int exclude_super_stripes(struct btrfs_block_group_cache *cache)
+/**
+ * btrfs_rmap_block - Map a physical disk address to a list of logical addresses
+ * @chunk_start:   logical address of block group
+ * @physical:	   physical address to map to logical addresses
+ * @logical:	   return array of logical addresses which map to @physical
+ * @naddrs:	   length of @logical
+ * @stripe_len:    size of IO stripe for the given block group
+ *
+ * Maps a particular @physical disk address to a list of @logical addresses.
+ * Used primarily to exclude those portions of a block group that contain super
+ * block copies.
+ */
+EXPORT_FOR_TESTS
+int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
+		     u64 physical, u64 **logical, int *naddrs, int *stripe_len)
+{
+	struct extent_map *em;
+	struct map_lookup *map;
+	u64 *buf;
+	u64 bytenr;
+	u64 data_stripe_length;
+	u64 io_stripe_size;
+	int i, nr = 0;
+	int ret = 0;
+
+	em = btrfs_get_chunk_map(fs_info, chunk_start, 1);
+	if (IS_ERR(em))
+		return -EIO;
+
+	map = em->map_lookup;
+	data_stripe_length = em->len;
+	io_stripe_size = map->stripe_len;
+
+	if (map->type & BTRFS_BLOCK_GROUP_RAID10)
+		data_stripe_length = div_u64(data_stripe_length,
+					     map->num_stripes / map->sub_stripes);
+	else if (map->type & BTRFS_BLOCK_GROUP_RAID0)
+		data_stripe_length = div_u64(data_stripe_length, map->num_stripes);
+	else if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
+		data_stripe_length = div_u64(data_stripe_length,
+					     nr_data_stripes(map));
+		io_stripe_size = map->stripe_len * nr_data_stripes(map);
+	}
+
+	buf = kcalloc(map->num_stripes, sizeof(u64), GFP_NOFS);
+	if (!buf) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	for (i = 0; i < map->num_stripes; i++) {
+		bool already_inserted = false;
+		u64 stripe_nr;
+		int j;
+
+		if (!in_range(physical, map->stripes[i].physical,
+			      data_stripe_length))
+			continue;
+
+		stripe_nr = physical - map->stripes[i].physical;
+		stripe_nr = div64_u64(stripe_nr, map->stripe_len);
+
+		if (map->type & BTRFS_BLOCK_GROUP_RAID10) {
+			stripe_nr = stripe_nr * map->num_stripes + i;
+			stripe_nr = div_u64(stripe_nr, map->sub_stripes);
+		} else if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
+			stripe_nr = stripe_nr * map->num_stripes + i;
+		}
+		/*
+		 * The remaining case would be for RAID56, multiply by
+		 * nr_data_stripes().  Alternatively, just use rmap_len below
+		 * instead of map->stripe_len
+		 */
+
+		bytenr = chunk_start + stripe_nr * io_stripe_size;
+
+		/* Ensure we don't add duplicate addresses */
+		for (j = 0; j < nr; j++) {
+			if (buf[j] == bytenr) {
+				already_inserted = true;
+				break;
+			}
+		}
+
+		if (!already_inserted)
+			buf[nr++] = bytenr;
+	}
+
+	*logical = buf;
+	*naddrs = nr;
+	*stripe_len = io_stripe_size;
+out:
+	free_extent_map(em);
+	return ret;
+}
+
+static int exclude_super_stripes(struct btrfs_block_group *cache)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	u64 bytenr;
@@ -1526,10 +1710,10 @@ static int exclude_super_stripes(struct btrfs_block_group_cache *cache)
 	int stripe_len;
 	int i, nr, ret;
 
-	if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
-		stripe_len = BTRFS_SUPER_INFO_OFFSET - cache->key.objectid;
+	if (cache->start < BTRFS_SUPER_INFO_OFFSET) {
+		stripe_len = BTRFS_SUPER_INFO_OFFSET - cache->start;
 		cache->bytes_super += stripe_len;
-		ret = btrfs_add_excluded_extent(fs_info, cache->key.objectid,
+		ret = btrfs_add_excluded_extent(fs_info, cache->start,
 						stripe_len);
 		if (ret)
 			return ret;
@@ -1537,7 +1721,7 @@ static int exclude_super_stripes(struct btrfs_block_group_cache *cache)
 
 	for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
 		bytenr = btrfs_sb_offset(i);
-		ret = btrfs_rmap_block(fs_info, cache->key.objectid,
+		ret = btrfs_rmap_block(fs_info, cache->start,
 				       bytenr, &logical, &nr, &stripe_len);
 		if (ret)
 			return ret;
@@ -1545,21 +1729,19 @@ static int exclude_super_stripes(struct btrfs_block_group_cache *cache)
 		while (nr--) {
 			u64 start, len;
 
-			if (logical[nr] > cache->key.objectid +
-			    cache->key.offset)
+			if (logical[nr] > cache->start + cache->length)
 				continue;
 
-			if (logical[nr] + stripe_len <= cache->key.objectid)
+			if (logical[nr] + stripe_len <= cache->start)
 				continue;
 
 			start = logical[nr];
-			if (start < cache->key.objectid) {
-				start = cache->key.objectid;
+			if (start < cache->start) {
+				start = cache->start;
 				len = (logical[nr] + stripe_len) - start;
 			} else {
 				len = min_t(u64, stripe_len,
-					    cache->key.objectid +
-					    cache->key.offset - start);
+					    cache->start + cache->length - start);
 			}
 
 			cache->bytes_super += len;
@@ -1575,7 +1757,7 @@ static int exclude_super_stripes(struct btrfs_block_group_cache *cache)
 	return 0;
 }
 
-static void link_block_group(struct btrfs_block_group_cache *cache)
+static void link_block_group(struct btrfs_block_group *cache)
 {
 	struct btrfs_space_info *space_info = cache->space_info;
 	int index = btrfs_bg_flags_to_raid_index(cache->flags);
@@ -1591,10 +1773,10 @@ static void link_block_group(struct btrfs_block_group_cache *cache)
 		btrfs_sysfs_add_block_group_type(cache);
 }
 
-static struct btrfs_block_group_cache *btrfs_create_block_group_cache(
+static struct btrfs_block_group *btrfs_create_block_group_cache(
 		struct btrfs_fs_info *fs_info, u64 start, u64 size)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 
 	cache = kzalloc(sizeof(*cache), GFP_NOFS);
 	if (!cache)
@@ -1607,14 +1789,15 @@ static struct btrfs_block_group_cache *btrfs_create_block_group_cache(
 		return NULL;
 	}
 
-	cache->key.objectid = start;
-	cache->key.offset = size;
-	cache->key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+	cache->start = start;
+	cache->length = size;
 
 	cache->fs_info = fs_info;
 	cache->full_stripe_len = btrfs_full_stripe_len(fs_info, start);
 	set_free_space_tree_thresholds(cache);
 
+	cache->discard_index = BTRFS_DISCARD_INDEX_UNUSED;
+
 	atomic_set(&cache->count, 1);
 	spin_lock_init(&cache->lock);
 	init_rwsem(&cache->data_rwsem);
@@ -1622,6 +1805,7 @@ static struct btrfs_block_group_cache *btrfs_create_block_group_cache(
 	INIT_LIST_HEAD(&cache->cluster_list);
 	INIT_LIST_HEAD(&cache->bg_list);
 	INIT_LIST_HEAD(&cache->ro_list);
+	INIT_LIST_HEAD(&cache->discard_list);
 	INIT_LIST_HEAD(&cache->dirty_list);
 	INIT_LIST_HEAD(&cache->io_list);
 	btrfs_init_free_space_ctl(cache);
@@ -1640,7 +1824,7 @@ static int check_chunk_block_group_mappings(struct btrfs_fs_info *fs_info)
 {
 	struct extent_map_tree *map_tree = &fs_info->mapping_tree;
 	struct extent_map *em;
-	struct btrfs_block_group_cache *bg;
+	struct btrfs_block_group *bg;
 	u64 start = 0;
 	int ret = 0;
 
@@ -1665,15 +1849,14 @@ static int check_chunk_block_group_mappings(struct btrfs_fs_info *fs_info)
 			free_extent_map(em);
 			break;
 		}
-		if (bg->key.objectid != em->start ||
-		    bg->key.offset != em->len ||
+		if (bg->start != em->start || bg->length != em->len ||
 		    (bg->flags & BTRFS_BLOCK_GROUP_TYPE_MASK) !=
 		    (em->map_lookup->type & BTRFS_BLOCK_GROUP_TYPE_MASK)) {
 			btrfs_err(fs_info,
 "chunk start=%llu len=%llu flags=0x%llx doesn't match block group start=%llu len=%llu flags=0x%llx",
 				em->start, em->len,
 				em->map_lookup->type & BTRFS_BLOCK_GROUP_TYPE_MASK,
-				bg->key.objectid, bg->key.offset,
+				bg->start, bg->length,
 				bg->flags & BTRFS_BLOCK_GROUP_TYPE_MASK);
 			ret = -EUCLEAN;
 			free_extent_map(em);
@@ -1687,22 +1870,120 @@ static int check_chunk_block_group_mappings(struct btrfs_fs_info *fs_info)
 	return ret;
 }
 
+static int read_one_block_group(struct btrfs_fs_info *info,
+				struct btrfs_path *path,
+				const struct btrfs_key *key,
+				int need_clear)
+{
+	struct extent_buffer *leaf = path->nodes[0];
+	struct btrfs_block_group *cache;
+	struct btrfs_space_info *space_info;
+	struct btrfs_block_group_item bgi;
+	const bool mixed = btrfs_fs_incompat(info, MIXED_GROUPS);
+	int slot = path->slots[0];
+	int ret;
+
+	ASSERT(key->type == BTRFS_BLOCK_GROUP_ITEM_KEY);
+
+	cache = btrfs_create_block_group_cache(info, key->objectid, key->offset);
+	if (!cache)
+		return -ENOMEM;
+
+	if (need_clear) {
+		/*
+		 * When we mount with old space cache, we need to
+		 * set BTRFS_DC_CLEAR and set dirty flag.
+		 *
+		 * a) Setting 'BTRFS_DC_CLEAR' makes sure that we
+		 *    truncate the old free space cache inode and
+		 *    setup a new one.
+		 * b) Setting 'dirty flag' makes sure that we flush
+		 *    the new space cache info onto disk.
+		 */
+		if (btrfs_test_opt(info, SPACE_CACHE))
+			cache->disk_cache_state = BTRFS_DC_CLEAR;
+	}
+	read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot),
+			   sizeof(bgi));
+	cache->used = btrfs_stack_block_group_used(&bgi);
+	cache->flags = btrfs_stack_block_group_flags(&bgi);
+	if (!mixed && ((cache->flags & BTRFS_BLOCK_GROUP_METADATA) &&
+	    (cache->flags & BTRFS_BLOCK_GROUP_DATA))) {
+			btrfs_err(info,
+"bg %llu is a mixed block group but filesystem hasn't enabled mixed block groups",
+				  cache->start);
+			ret = -EINVAL;
+			goto error;
+	}
+
+	/*
+	 * We need to exclude the super stripes now so that the space info has
+	 * super bytes accounted for, otherwise we'll think we have more space
+	 * than we actually do.
+	 */
+	ret = exclude_super_stripes(cache);
+	if (ret) {
+		/* We may have excluded something, so call this just in case. */
+		btrfs_free_excluded_extents(cache);
+		goto error;
+	}
+
+	/*
+	 * Check for two cases, either we are full, and therefore don't need
+	 * to bother with the caching work since we won't find any space, or we
+	 * are empty, and we can just add all the space in and be done with it.
+	 * This saves us _a_lot_ of time, particularly in the full case.
+	 */
+	if (key->offset == cache->used) {
+		cache->last_byte_to_unpin = (u64)-1;
+		cache->cached = BTRFS_CACHE_FINISHED;
+		btrfs_free_excluded_extents(cache);
+	} else if (cache->used == 0) {
+		cache->last_byte_to_unpin = (u64)-1;
+		cache->cached = BTRFS_CACHE_FINISHED;
+		add_new_free_space(cache, key->objectid,
+				   key->objectid + key->offset);
+		btrfs_free_excluded_extents(cache);
+	}
+
+	ret = btrfs_add_block_group_cache(info, cache);
+	if (ret) {
+		btrfs_remove_free_space_cache(cache);
+		goto error;
+	}
+	trace_btrfs_add_block_group(info, cache, 0);
+	btrfs_update_space_info(info, cache->flags, key->offset,
+				cache->used, cache->bytes_super, &space_info);
+
+	cache->space_info = space_info;
+
+	link_block_group(cache);
+
+	set_avail_alloc_bits(info, cache->flags);
+	if (btrfs_chunk_readonly(info, cache->start)) {
+		inc_block_group_ro(cache, 1);
+	} else if (cache->used == 0) {
+		ASSERT(list_empty(&cache->bg_list));
+		if (btrfs_test_opt(info, DISCARD_ASYNC))
+			btrfs_discard_queue_work(&info->discard_ctl, cache);
+		else
+			btrfs_mark_bg_unused(cache);
+	}
+	return 0;
+error:
+	btrfs_put_block_group(cache);
+	return ret;
+}
+
 int btrfs_read_block_groups(struct btrfs_fs_info *info)
 {
 	struct btrfs_path *path;
 	int ret;
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	struct btrfs_space_info *space_info;
 	struct btrfs_key key;
-	struct btrfs_key found_key;
-	struct extent_buffer *leaf;
 	int need_clear = 0;
 	u64 cache_gen;
-	u64 feature;
-	int mixed;
-
-	feature = btrfs_super_incompat_flags(info->super_copy);
-	mixed = !!(feature & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS);
 
 	key.objectid = 0;
 	key.offset = 0;
@@ -1726,110 +2007,16 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 		if (ret != 0)
 			goto error;
 
-		leaf = path->nodes[0];
-		btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
-
-		cache = btrfs_create_block_group_cache(info, found_key.objectid,
-						       found_key.offset);
-		if (!cache) {
-			ret = -ENOMEM;
-			goto error;
-		}
-
-		if (need_clear) {
-			/*
-			 * When we mount with old space cache, we need to
-			 * set BTRFS_DC_CLEAR and set dirty flag.
-			 *
-			 * a) Setting 'BTRFS_DC_CLEAR' makes sure that we
-			 *    truncate the old free space cache inode and
-			 *    setup a new one.
-			 * b) Setting 'dirty flag' makes sure that we flush
-			 *    the new space cache info onto disk.
-			 */
-			if (btrfs_test_opt(info, SPACE_CACHE))
-				cache->disk_cache_state = BTRFS_DC_CLEAR;
-		}
-
-		read_extent_buffer(leaf, &cache->item,
-				   btrfs_item_ptr_offset(leaf, path->slots[0]),
-				   sizeof(cache->item));
-		cache->flags = btrfs_block_group_flags(&cache->item);
-		if (!mixed &&
-		    ((cache->flags & BTRFS_BLOCK_GROUP_METADATA) &&
-		    (cache->flags & BTRFS_BLOCK_GROUP_DATA))) {
-			btrfs_err(info,
-"bg %llu is a mixed block group but filesystem hasn't enabled mixed block groups",
-				  cache->key.objectid);
-			btrfs_put_block_group(cache);
-			ret = -EINVAL;
+		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+		ret = read_one_block_group(info, path, &key, need_clear);
+		if (ret < 0)
 			goto error;
-		}
-
-		key.objectid = found_key.objectid + found_key.offset;
+		key.objectid += key.offset;
+		key.offset = 0;
 		btrfs_release_path(path);
-
-		/*
-		 * We need to exclude the super stripes now so that the space
-		 * info has super bytes accounted for, otherwise we'll think
-		 * we have more space than we actually do.
-		 */
-		ret = exclude_super_stripes(cache);
-		if (ret) {
-			/*
-			 * We may have excluded something, so call this just in
-			 * case.
-			 */
-			btrfs_free_excluded_extents(cache);
-			btrfs_put_block_group(cache);
-			goto error;
-		}
-
-		/*
-		 * Check for two cases, either we are full, and therefore
-		 * don't need to bother with the caching work since we won't
-		 * find any space, or we are empty, and we can just add all
-		 * the space in and be done with it.  This saves us _a_lot_ of
-		 * time, particularly in the full case.
-		 */
-		if (found_key.offset == btrfs_block_group_used(&cache->item)) {
-			cache->last_byte_to_unpin = (u64)-1;
-			cache->cached = BTRFS_CACHE_FINISHED;
-			btrfs_free_excluded_extents(cache);
-		} else if (btrfs_block_group_used(&cache->item) == 0) {
-			cache->last_byte_to_unpin = (u64)-1;
-			cache->cached = BTRFS_CACHE_FINISHED;
-			add_new_free_space(cache, found_key.objectid,
-					   found_key.objectid +
-					   found_key.offset);
-			btrfs_free_excluded_extents(cache);
-		}
-
-		ret = btrfs_add_block_group_cache(info, cache);
-		if (ret) {
-			btrfs_remove_free_space_cache(cache);
-			btrfs_put_block_group(cache);
-			goto error;
-		}
-
-		trace_btrfs_add_block_group(info, cache, 0);
-		btrfs_update_space_info(info, cache->flags, found_key.offset,
-					btrfs_block_group_used(&cache->item),
-					cache->bytes_super, &space_info);
-
-		cache->space_info = space_info;
-
-		link_block_group(cache);
-
-		set_avail_alloc_bits(info, cache->flags);
-		if (btrfs_chunk_readonly(info, cache->key.objectid)) {
-			inc_block_group_ro(cache, 1);
-		} else if (btrfs_block_group_used(&cache->item) == 0) {
-			ASSERT(list_empty(&cache->bg_list));
-			btrfs_mark_bg_unused(cache);
-		}
 	}
 
+	rcu_read_lock();
 	list_for_each_entry_rcu(space_info, &info->space_info, list) {
 		if (!(btrfs_get_alloc_profile(info, space_info->flags) &
 		      (BTRFS_BLOCK_GROUP_RAID10 |
@@ -1850,6 +2037,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 				list)
 			inc_block_group_ro(cache, 1);
 	}
+	rcu_read_unlock();
 
 	btrfs_init_global_block_rsv(info);
 	ret = check_chunk_block_group_mappings(info);
@@ -1861,7 +2049,7 @@ error:
 void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_root *extent_root = fs_info->extent_root;
 	struct btrfs_block_group_item item;
 	struct btrfs_key key;
@@ -1872,14 +2060,19 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 
 	while (!list_empty(&trans->new_bgs)) {
 		block_group = list_first_entry(&trans->new_bgs,
-					       struct btrfs_block_group_cache,
+					       struct btrfs_block_group,
 					       bg_list);
 		if (ret)
 			goto next;
 
 		spin_lock(&block_group->lock);
-		memcpy(&item, &block_group->item, sizeof(item));
-		memcpy(&key, &block_group->key, sizeof(key));
+		btrfs_set_stack_block_group_used(&item, block_group->used);
+		btrfs_set_stack_block_group_chunk_objectid(&item,
+				BTRFS_FIRST_CHUNK_TREE_OBJECTID);
+		btrfs_set_stack_block_group_flags(&item, block_group->flags);
+		key.objectid = block_group->start;
+		key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+		key.offset = block_group->length;
 		spin_unlock(&block_group->lock);
 
 		ret = btrfs_insert_item(trans, extent_root, &key, &item,
@@ -1902,7 +2095,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used,
 			   u64 type, u64 chunk_offset, u64 size)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	int ret;
 
 	btrfs_set_log_full_commit(trans);
@@ -1911,11 +2104,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used,
 	if (!cache)
 		return -ENOMEM;
 
-	btrfs_set_block_group_used(&cache->item, bytes_used);
-	btrfs_set_block_group_chunk_objectid(&cache->item,
-					     BTRFS_FIRST_CHUNK_TREE_OBJECTID);
-	btrfs_set_block_group_flags(&cache->item, type);
-
+	cache->used = bytes_used;
 	cache->flags = type;
 	cache->last_byte_to_unpin = (u64)-1;
 	cache->cached = BTRFS_CACHE_FINISHED;
@@ -2022,8 +2211,17 @@ static u64 update_block_group_flags(struct btrfs_fs_info *fs_info, u64 flags)
 	return flags;
 }
 
-int btrfs_inc_block_group_ro(struct btrfs_block_group_cache *cache)
-
+/*
+ * Mark one block group RO, can be called several times for the same block
+ * group.
+ *
+ * @cache:		the destination block group
+ * @do_chunk_alloc:	whether need to do chunk pre-allocation, this is to
+ * 			ensure we still have some free space after marking this
+ * 			block group RO.
+ */
+int btrfs_inc_block_group_ro(struct btrfs_block_group *cache,
+			     bool do_chunk_alloc)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	struct btrfs_trans_handle *trans;
@@ -2053,25 +2251,29 @@ again:
 		goto again;
 	}
 
-	/*
-	 * if we are changing raid levels, try to allocate a corresponding
-	 * block group with the new raid level.
-	 */
-	alloc_flags = update_block_group_flags(fs_info, cache->flags);
-	if (alloc_flags != cache->flags) {
-		ret = btrfs_chunk_alloc(trans, alloc_flags, CHUNK_ALLOC_FORCE);
+	if (do_chunk_alloc) {
 		/*
-		 * ENOSPC is allowed here, we may have enough space
-		 * already allocated at the new raid level to
-		 * carry on
+		 * If we are changing raid levels, try to allocate a
+		 * corresponding block group with the new raid level.
 		 */
-		if (ret == -ENOSPC)
-			ret = 0;
-		if (ret < 0)
-			goto out;
+		alloc_flags = update_block_group_flags(fs_info, cache->flags);
+		if (alloc_flags != cache->flags) {
+			ret = btrfs_chunk_alloc(trans, alloc_flags,
+						CHUNK_ALLOC_FORCE);
+			/*
+			 * ENOSPC is allowed here, we may have enough space
+			 * already allocated at the new raid level to carry on
+			 */
+			if (ret == -ENOSPC)
+				ret = 0;
+			if (ret < 0)
+				goto out;
+		}
 	}
 
 	ret = inc_block_group_ro(cache, 0);
+	if (!do_chunk_alloc)
+		goto unlock_out;
 	if (!ret)
 		goto out;
 	alloc_flags = btrfs_get_alloc_profile(fs_info, cache->space_info->flags);
@@ -2086,13 +2288,14 @@ out:
 		check_system_chunk(trans, alloc_flags);
 		mutex_unlock(&fs_info->chunk_mutex);
 	}
+unlock_out:
 	mutex_unlock(&fs_info->ro_block_group_mutex);
 
 	btrfs_end_transaction(trans);
 	return ret;
 }
 
-void btrfs_dec_block_group_ro(struct btrfs_block_group_cache *cache)
+void btrfs_dec_block_group_ro(struct btrfs_block_group *cache)
 {
 	struct btrfs_space_info *sinfo = cache->space_info;
 	u64 num_bytes;
@@ -2102,9 +2305,8 @@ void btrfs_dec_block_group_ro(struct btrfs_block_group_cache *cache)
 	spin_lock(&sinfo->lock);
 	spin_lock(&cache->lock);
 	if (!--cache->ro) {
-		num_bytes = cache->key.offset - cache->reserved -
-			    cache->pinned - cache->bytes_super -
-			    btrfs_block_group_used(&cache->item);
+		num_bytes = cache->length - cache->reserved -
+			    cache->pinned - cache->bytes_super - cache->used;
 		sinfo->bytes_readonly -= num_bytes;
 		list_del_init(&cache->ro_list);
 	}
@@ -2114,15 +2316,21 @@ void btrfs_dec_block_group_ro(struct btrfs_block_group_cache *cache)
 
 static int write_one_cache_group(struct btrfs_trans_handle *trans,
 				 struct btrfs_path *path,
-				 struct btrfs_block_group_cache *cache)
+				 struct btrfs_block_group *cache)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret;
 	struct btrfs_root *extent_root = fs_info->extent_root;
 	unsigned long bi;
 	struct extent_buffer *leaf;
+	struct btrfs_block_group_item bgi;
+	struct btrfs_key key;
 
-	ret = btrfs_search_slot(trans, extent_root, &cache->key, path, 0, 1);
+	key.objectid = cache->start;
+	key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+	key.offset = cache->length;
+
+	ret = btrfs_search_slot(trans, extent_root, &key, path, 0, 1);
 	if (ret) {
 		if (ret > 0)
 			ret = -ENOENT;
@@ -2131,7 +2339,11 @@ static int write_one_cache_group(struct btrfs_trans_handle *trans,
 
 	leaf = path->nodes[0];
 	bi = btrfs_item_ptr_offset(leaf, path->slots[0]);
-	write_extent_buffer(leaf, &cache->item, bi, sizeof(cache->item));
+	btrfs_set_stack_block_group_used(&bgi, cache->used);
+	btrfs_set_stack_block_group_chunk_objectid(&bgi,
+			BTRFS_FIRST_CHUNK_TREE_OBJECTID);
+	btrfs_set_stack_block_group_flags(&bgi, cache->flags);
+	write_extent_buffer(leaf, &bgi, bi, sizeof(bgi));
 	btrfs_mark_buffer_dirty(leaf);
 fail:
 	btrfs_release_path(path);
@@ -2139,7 +2351,7 @@ fail:
 
 }
 
-static int cache_save_setup(struct btrfs_block_group_cache *block_group,
+static int cache_save_setup(struct btrfs_block_group *block_group,
 			    struct btrfs_trans_handle *trans,
 			    struct btrfs_path *path)
 {
@@ -2157,14 +2369,14 @@ static int cache_save_setup(struct btrfs_block_group_cache *block_group,
 	 * If this block group is smaller than 100 megs don't bother caching the
 	 * block group.
 	 */
-	if (block_group->key.offset < (100 * SZ_1M)) {
+	if (block_group->length < (100 * SZ_1M)) {
 		spin_lock(&block_group->lock);
 		block_group->disk_cache_state = BTRFS_DC_WRITTEN;
 		spin_unlock(&block_group->lock);
 		return 0;
 	}
 
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return 0;
 again:
 	inode = lookup_free_space_inode(block_group, path);
@@ -2258,7 +2470,7 @@ again:
 	 * taking up quite a bit since it's not folded into the other space
 	 * cache.
 	 */
-	num_pages = div_u64(block_group->key.offset, SZ_256M);
+	num_pages = div_u64(block_group->length, SZ_256M);
 	if (!num_pages)
 		num_pages = 1;
 
@@ -2303,7 +2515,7 @@ out:
 int btrfs_setup_space_cache(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_block_group_cache *cache, *tmp;
+	struct btrfs_block_group *cache, *tmp;
 	struct btrfs_transaction *cur_trans = trans->transaction;
 	struct btrfs_path *path;
 
@@ -2341,7 +2553,7 @@ int btrfs_setup_space_cache(struct btrfs_trans_handle *trans)
 int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	struct btrfs_transaction *cur_trans = trans->transaction;
 	int ret = 0;
 	int should_put;
@@ -2378,8 +2590,7 @@ again:
 	while (!list_empty(&dirty)) {
 		bool drop_reserve = true;
 
-		cache = list_first_entry(&dirty,
-					 struct btrfs_block_group_cache,
+		cache = list_first_entry(&dirty, struct btrfs_block_group,
 					 dirty_list);
 		/*
 		 * This can happen if something re-dirties a block group that
@@ -2504,7 +2715,7 @@ again:
 int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	struct btrfs_transaction *cur_trans = trans->transaction;
 	int ret = 0;
 	int should_put;
@@ -2534,7 +2745,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 	spin_lock(&cur_trans->dirty_bgs_lock);
 	while (!list_empty(&cur_trans->dirty_bgs)) {
 		cache = list_first_entry(&cur_trans->dirty_bgs,
-					 struct btrfs_block_group_cache,
+					 struct btrfs_block_group,
 					 dirty_list);
 
 		/*
@@ -2616,7 +2827,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 	 * to use it without any locking
 	 */
 	while (!list_empty(io)) {
-		cache = list_first_entry(io, struct btrfs_block_group_cache,
+		cache = list_first_entry(io, struct btrfs_block_group,
 					 io_list);
 		list_del_init(&cache->io_list);
 		btrfs_wait_cache_io(trans, cache, path);
@@ -2631,7 +2842,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 			     u64 bytenr, u64 num_bytes, int alloc)
 {
 	struct btrfs_fs_info *info = trans->fs_info;
-	struct btrfs_block_group_cache *cache = NULL;
+	struct btrfs_block_group *cache = NULL;
 	u64 total = num_bytes;
 	u64 old_val;
 	u64 byte_in_group;
@@ -2662,11 +2873,11 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 		 * is because we need the unpinning stage to actually add the
 		 * space back to the block group, otherwise we will leak space.
 		 */
-		if (!alloc && cache->cached == BTRFS_CACHE_NO)
+		if (!alloc && !btrfs_block_group_done(cache))
 			btrfs_cache_block_group(cache, 1);
 
-		byte_in_group = bytenr - cache->key.objectid;
-		WARN_ON(byte_in_group > cache->key.offset);
+		byte_in_group = bytenr - cache->start;
+		WARN_ON(byte_in_group > cache->length);
 
 		spin_lock(&cache->space_info->lock);
 		spin_lock(&cache->lock);
@@ -2675,11 +2886,11 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 		    cache->disk_cache_state < BTRFS_DC_CLEAR)
 			cache->disk_cache_state = BTRFS_DC_CLEAR;
 
-		old_val = btrfs_block_group_used(&cache->item);
-		num_bytes = min(total, cache->key.offset - byte_in_group);
+		old_val = cache->used;
+		num_bytes = min(total, cache->length - byte_in_group);
 		if (alloc) {
 			old_val += num_bytes;
-			btrfs_set_block_group_used(&cache->item, old_val);
+			cache->used = old_val;
 			cache->reserved -= num_bytes;
 			cache->space_info->bytes_reserved -= num_bytes;
 			cache->space_info->bytes_used += num_bytes;
@@ -2688,7 +2899,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 			spin_unlock(&cache->space_info->lock);
 		} else {
 			old_val -= num_bytes;
-			btrfs_set_block_group_used(&cache->item, old_val);
+			cache->used = old_val;
 			cache->pinned += num_bytes;
 			btrfs_space_info_update_bytes_pinned(info,
 					cache->space_info, num_bytes);
@@ -2701,7 +2912,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 					&cache->space_info->total_bytes_pinned,
 					num_bytes,
 					BTRFS_TOTAL_BYTES_PINNED_BATCH);
-			set_extent_dirty(info->pinned_extents,
+			set_extent_dirty(&trans->transaction->pinned_extents,
 					 bytenr, bytenr + num_bytes - 1,
 					 GFP_NOFS | __GFP_NOFAIL);
 		}
@@ -2721,8 +2932,10 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 		 * dirty list to avoid races between cleaner kthread and space
 		 * cache writeout.
 		 */
-		if (!alloc && old_val == 0)
-			btrfs_mark_bg_unused(cache);
+		if (!alloc && old_val == 0) {
+			if (!btrfs_test_opt(info, DISCARD_ASYNC))
+				btrfs_mark_bg_unused(cache);
+		}
 
 		btrfs_put_block_group(cache);
 		total -= num_bytes;
@@ -2746,7 +2959,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
  * reservation and the block group has become read only we cannot make the
  * reservation and return -EAGAIN, otherwise this function always succeeds.
  */
-int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache,
+int btrfs_add_reserved_bytes(struct btrfs_block_group *cache,
 			     u64 ram_bytes, u64 num_bytes, int delalloc)
 {
 	struct btrfs_space_info *space_info = cache->space_info;
@@ -2782,7 +2995,7 @@ int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache,
  * A and before transaction A commits you free that leaf, you call this with
  * reserve set to 0 in order to clear the reservation.
  */
-void btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
+void btrfs_free_reserved_bytes(struct btrfs_block_group *cache,
 			       u64 num_bytes, int delalloc)
 {
 	struct btrfs_space_info *space_info = cache->space_info;
@@ -2988,9 +3201,7 @@ static u64 get_profile_num_devs(struct btrfs_fs_info *fs_info, u64 type)
 }
 
 /*
- * If @is_allocation is true, reserve space in the system space info necessary
- * for allocating a chunk, otherwise if it's false, reserve space necessary for
- * removing a chunk.
+ * Reserve space in the system space for allocating or removing a chunk
  */
 void check_system_chunk(struct btrfs_trans_handle *trans, u64 type)
 {
@@ -3047,7 +3258,7 @@ void check_system_chunk(struct btrfs_trans_handle *trans, u64 type)
 
 void btrfs_put_block_group_cache(struct btrfs_fs_info *info)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	u64 last = 0;
 
 	while (1) {
@@ -3075,7 +3286,7 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info)
 		spin_unlock(&block_group->lock);
 		ASSERT(block_group->io_ctl.inode == NULL);
 		iput(inode);
-		last = block_group->key.objectid + block_group->key.offset;
+		last = block_group->start + block_group->length;
 		btrfs_put_block_group(block_group);
 	}
 }
@@ -3087,7 +3298,7 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info)
  */
 int btrfs_free_block_groups(struct btrfs_fs_info *info)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_space_info *space_info;
 	struct btrfs_caching_control *caching_ctl;
 	struct rb_node *n;
@@ -3104,7 +3315,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 	spin_lock(&info->unused_bgs_lock);
 	while (!list_empty(&info->unused_bgs)) {
 		block_group = list_first_entry(&info->unused_bgs,
-					       struct btrfs_block_group_cache,
+					       struct btrfs_block_group,
 					       bg_list);
 		list_del_init(&block_group->bg_list);
 		btrfs_put_block_group(block_group);
@@ -3113,7 +3324,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 
 	spin_lock(&info->block_group_cache_lock);
 	while ((n = rb_last(&info->block_group_cache_tree)) != NULL) {
-		block_group = rb_entry(n, struct btrfs_block_group_cache,
+		block_group = rb_entry(n, struct btrfs_block_group,
 				       cache_node);
 		rb_erase(&block_group->cache_node,
 			 &info->block_group_cache_tree);
@@ -3167,6 +3378,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 			    space_info->bytes_reserved > 0 ||
 			    space_info->bytes_may_use > 0))
 			btrfs_dump_space_info(info, space_info, 0, 0);
+		WARN_ON(space_info->reclaim_size > 0);
 		list_del(&space_info->list);
 		btrfs_sysfs_remove_space_info(space_info);
 	}
diff --git a/fs/btrfs/block-group.h b/fs/btrfs/block-group.h
index c391800388dd..107bb557ca8d 100644
--- a/fs/btrfs/block-group.h
+++ b/fs/btrfs/block-group.h
@@ -13,6 +13,19 @@ enum btrfs_disk_cache_state {
 };
 
 /*
+ * This describes the state of the block_group for async discard.  This is due
+ * to the two pass nature of it where extent discarding is prioritized over
+ * bitmap discarding.  BTRFS_DISCARD_RESET_CURSOR is set when we are resetting
+ * between lists to prevent contention for discard state variables
+ * (eg. discard_cursor).
+ */
+enum btrfs_discard_state {
+	BTRFS_DISCARD_EXTENTS,
+	BTRFS_DISCARD_BITMAPS,
+	BTRFS_DISCARD_RESET_CURSOR,
+};
+
+/*
  * Control flags for do_chunk_alloc's force field CHUNK_ALLOC_NO_FORCE means to
  * only allocate a chunk if we really need one.
  *
@@ -34,7 +47,7 @@ struct btrfs_caching_control {
 	struct mutex mutex;
 	wait_queue_head_t wait;
 	struct btrfs_work work;
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	u64 progress;
 	refcount_t count;
 };
@@ -42,14 +55,15 @@ struct btrfs_caching_control {
 /* Once caching_thread() finds this much free space, it will wake up waiters. */
 #define CACHING_CTL_WAKE_UP SZ_2M
 
-struct btrfs_block_group_cache {
-	struct btrfs_key key;
-	struct btrfs_block_group_item item;
+struct btrfs_block_group {
 	struct btrfs_fs_info *fs_info;
 	struct inode *inode;
 	spinlock_t lock;
+	u64 start;
+	u64 length;
 	u64 pinned;
 	u64 reserved;
+	u64 used;
 	u64 delalloc_bytes;
 	u64 bytes_super;
 	u64 flags;
@@ -115,7 +129,13 @@ struct btrfs_block_group_cache {
 	/* For read-only block groups */
 	struct list_head ro_list;
 
+	/* For discard operations */
 	atomic_t trimming;
+	struct list_head discard_list;
+	int discard_index;
+	u64 discard_eligible_time;
+	u64 discard_cursor;
+	enum btrfs_discard_state discard_state;
 
 	/* For dirty block groups */
 	struct list_head dirty_list;
@@ -157,9 +177,25 @@ struct btrfs_block_group_cache {
 	struct btrfs_full_stripe_locks_tree full_stripe_locks_root;
 };
 
+static inline u64 btrfs_block_group_end(struct btrfs_block_group *block_group)
+{
+	return (block_group->start + block_group->length);
+}
+
+static inline bool btrfs_is_block_group_data_only(
+					struct btrfs_block_group *block_group)
+{
+	/*
+	 * In mixed mode the fragmentation is expected to be high, lowering the
+	 * efficiency, so only proper data block groups are considered.
+	 */
+	return (block_group->flags & BTRFS_BLOCK_GROUP_DATA) &&
+	       !(block_group->flags & BTRFS_BLOCK_GROUP_METADATA);
+}
+
 #ifdef CONFIG_BTRFS_DEBUG
 static inline int btrfs_should_fragment_free_space(
-		struct btrfs_block_group_cache *block_group)
+		struct btrfs_block_group *block_group)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
 
@@ -170,29 +206,29 @@ static inline int btrfs_should_fragment_free_space(
 }
 #endif
 
-struct btrfs_block_group_cache *btrfs_lookup_first_block_group(
+struct btrfs_block_group *btrfs_lookup_first_block_group(
 		struct btrfs_fs_info *info, u64 bytenr);
-struct btrfs_block_group_cache *btrfs_lookup_block_group(
+struct btrfs_block_group *btrfs_lookup_block_group(
 		struct btrfs_fs_info *info, u64 bytenr);
-struct btrfs_block_group_cache *btrfs_next_block_group(
-		struct btrfs_block_group_cache *cache);
-void btrfs_get_block_group(struct btrfs_block_group_cache *cache);
-void btrfs_put_block_group(struct btrfs_block_group_cache *cache);
+struct btrfs_block_group *btrfs_next_block_group(
+		struct btrfs_block_group *cache);
+void btrfs_get_block_group(struct btrfs_block_group *cache);
+void btrfs_put_block_group(struct btrfs_block_group *cache);
 void btrfs_dec_block_group_reservations(struct btrfs_fs_info *fs_info,
 					const u64 start);
-void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg);
+void btrfs_wait_block_group_reservations(struct btrfs_block_group *bg);
 bool btrfs_inc_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr);
 void btrfs_dec_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr);
-void btrfs_wait_nocow_writers(struct btrfs_block_group_cache *bg);
-void btrfs_wait_block_group_cache_progress(struct btrfs_block_group_cache *cache,
+void btrfs_wait_nocow_writers(struct btrfs_block_group *bg);
+void btrfs_wait_block_group_cache_progress(struct btrfs_block_group *cache,
 				           u64 num_bytes);
-int btrfs_wait_block_group_cache_done(struct btrfs_block_group_cache *cache);
-int btrfs_cache_block_group(struct btrfs_block_group_cache *cache,
+int btrfs_wait_block_group_cache_done(struct btrfs_block_group *cache);
+int btrfs_cache_block_group(struct btrfs_block_group *cache,
 			    int load_cache_only);
 void btrfs_put_caching_control(struct btrfs_caching_control *ctl);
 struct btrfs_caching_control *btrfs_get_caching_control(
-		struct btrfs_block_group_cache *cache);
-u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
+		struct btrfs_block_group *cache);
+u64 add_new_free_space(struct btrfs_block_group *block_group,
 		       u64 start, u64 end);
 struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
 				struct btrfs_fs_info *fs_info,
@@ -200,21 +236,22 @@ struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 			     u64 group_start, struct extent_map *em);
 void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info);
-void btrfs_mark_bg_unused(struct btrfs_block_group_cache *bg);
+void btrfs_mark_bg_unused(struct btrfs_block_group *bg);
 int btrfs_read_block_groups(struct btrfs_fs_info *info);
 int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used,
 			   u64 type, u64 chunk_offset, u64 size);
 void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans);
-int btrfs_inc_block_group_ro(struct btrfs_block_group_cache *cache);
-void btrfs_dec_block_group_ro(struct btrfs_block_group_cache *cache);
+int btrfs_inc_block_group_ro(struct btrfs_block_group *cache,
+			     bool do_chunk_alloc);
+void btrfs_dec_block_group_ro(struct btrfs_block_group *cache);
 int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans);
 int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans);
 int btrfs_setup_space_cache(struct btrfs_trans_handle *trans);
 int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 			     u64 bytenr, u64 num_bytes, int alloc);
-int btrfs_add_reserved_bytes(struct btrfs_block_group_cache *cache,
+int btrfs_add_reserved_bytes(struct btrfs_block_group *cache,
 			     u64 ram_bytes, u64 num_bytes, int delalloc);
-void btrfs_free_reserved_bytes(struct btrfs_block_group_cache *cache,
+void btrfs_free_reserved_bytes(struct btrfs_block_group *cache,
 			       u64 num_bytes, int delalloc);
 int btrfs_chunk_alloc(struct btrfs_trans_handle *trans, u64 flags,
 		      enum btrfs_chunk_alloc_enum force);
@@ -239,12 +276,16 @@ static inline u64 btrfs_system_alloc_profile(struct btrfs_fs_info *fs_info)
 	return btrfs_get_alloc_profile(fs_info, BTRFS_BLOCK_GROUP_SYSTEM);
 }
 
-static inline int btrfs_block_group_cache_done(
-		struct btrfs_block_group_cache *cache)
+static inline int btrfs_block_group_done(struct btrfs_block_group *cache)
 {
 	smp_mb();
 	return cache->cached == BTRFS_CACHE_FINISHED ||
 		cache->cached == BTRFS_CACHE_ERROR;
 }
 
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
+		     u64 physical, u64 **logical, int *naddrs, int *stripe_len);
+#endif
+
 #endif /* BTRFS_BLOCK_GROUP_H */
diff --git a/fs/btrfs/block-rsv.c b/fs/btrfs/block-rsv.c
index d07bd41a7c1e..27efec8f7c5b 100644
--- a/fs/btrfs/block-rsv.c
+++ b/fs/btrfs/block-rsv.c
@@ -6,6 +6,98 @@
 #include "space-info.h"
 #include "transaction.h"
 
+/*
+ * HOW DO BLOCK RESERVES WORK
+ *
+ *   Think of block_rsv's as buckets for logically grouped metadata
+ *   reservations.  Each block_rsv has a ->size and a ->reserved.  ->size is
+ *   how large we want our block rsv to be, ->reserved is how much space is
+ *   currently reserved for this block reserve.
+ *
+ *   ->failfast exists for the truncate case, and is described below.
+ *
+ * NORMAL OPERATION
+ *
+ *   -> Reserve
+ *     Entrance: btrfs_block_rsv_add, btrfs_block_rsv_refill
+ *
+ *     We call into btrfs_reserve_metadata_bytes() with our bytes, which is
+ *     accounted for in space_info->bytes_may_use, and then add the bytes to
+ *     ->reserved, and ->size in the case of btrfs_block_rsv_add.
+ *
+ *     ->size is an over-estimation of how much we may use for a particular
+ *     operation.
+ *
+ *   -> Use
+ *     Entrance: btrfs_use_block_rsv
+ *
+ *     When we do a btrfs_alloc_tree_block() we call into btrfs_use_block_rsv()
+ *     to determine the appropriate block_rsv to use, and then verify that
+ *     ->reserved has enough space for our tree block allocation.  Once
+ *     successful we subtract fs_info->nodesize from ->reserved.
+ *
+ *   -> Finish
+ *     Entrance: btrfs_block_rsv_release
+ *
+ *     We are finished with our operation, subtract our individual reservation
+ *     from ->size, and then subtract ->size from ->reserved and free up the
+ *     excess if there is any.
+ *
+ *     There is some logic here to refill the delayed refs rsv or the global rsv
+ *     as needed, otherwise the excess is subtracted from
+ *     space_info->bytes_may_use.
+ *
+ * TYPES OF BLOCK RESERVES
+ *
+ * BLOCK_RSV_TRANS, BLOCK_RSV_DELOPS, BLOCK_RSV_CHUNK
+ *   These behave normally, as described above, just within the confines of the
+ *   lifetime of their particular operation (transaction for the whole trans
+ *   handle lifetime, for example).
+ *
+ * BLOCK_RSV_GLOBAL
+ *   It is impossible to properly account for all the space that may be required
+ *   to make our extent tree updates.  This block reserve acts as an overflow
+ *   buffer in case our delayed refs reserve does not reserve enough space to
+ *   update the extent tree.
+ *
+ *   We can steal from this in some cases as well, notably on evict() or
+ *   truncate() in order to help users recover from ENOSPC conditions.
+ *
+ * BLOCK_RSV_DELALLOC
+ *   The individual item sizes are determined by the per-inode size
+ *   calculations, which are described with the delalloc code.  This is pretty
+ *   straightforward, it's just the calculation of ->size encodes a lot of
+ *   different items, and thus it gets used when updating inodes, inserting file
+ *   extents, and inserting checksums.
+ *
+ * BLOCK_RSV_DELREFS
+ *   We keep a running tally of how many delayed refs we have on the system.
+ *   We assume each one of these delayed refs are going to use a full
+ *   reservation.  We use the transaction items and pre-reserve space for every
+ *   operation, and use this reservation to refill any gap between ->size and
+ *   ->reserved that may exist.
+ *
+ *   From there it's straightforward, removing a delayed ref means we remove its
+ *   count from ->size and free up reservations as necessary.  Since this is
+ *   the most dynamic block reserve in the system, we will try to refill this
+ *   block reserve first with any excess returned by any other block reserve.
+ *
+ * BLOCK_RSV_EMPTY
+ *   This is the fallback block reserve to make us try to reserve space if we
+ *   don't have a specific bucket for this allocation.  It is mostly used for
+ *   updating the device tree and such, since that is a separate pool we're
+ *   content to just reserve space from the space_info on demand.
+ *
+ * BLOCK_RSV_TEMP
+ *   This is used by things like truncate and iput.  We will temporarily
+ *   allocate a block reserve, set it to some size, and then truncate bytes
+ *   until we have no space left.  With ->failfast set we'll simply return
+ *   ENOSPC from btrfs_use_block_rsv() to signal that we need to unwind and try
+ *   to make a new reservation.  This is because these operations are
+ *   unbounded, so we want to do as much work as we can, and then back off and
+ *   re-reserve.
+ */
+
 static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info,
 				    struct btrfs_block_rsv *block_rsv,
 				    struct btrfs_block_rsv *dest, u64 num_bytes,
@@ -111,7 +203,7 @@ void btrfs_free_block_rsv(struct btrfs_fs_info *fs_info,
 {
 	if (!rsv)
 		return;
-	btrfs_block_rsv_release(fs_info, rsv, (u64)-1);
+	btrfs_block_rsv_release(fs_info, rsv, (u64)-1, NULL);
 	kfree(rsv);
 }
 
@@ -178,9 +270,9 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
 	return ret;
 }
 
-u64 __btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
-			      struct btrfs_block_rsv *block_rsv,
-			      u64 num_bytes, u64 *qgroup_to_release)
+u64 btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
+			    struct btrfs_block_rsv *block_rsv, u64 num_bytes,
+			    u64 *qgroup_to_release)
 {
 	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
 	struct btrfs_block_rsv *delayed_rsv = &fs_info->delayed_refs_rsv;
@@ -297,9 +389,9 @@ void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info)
 
 	if (block_rsv->reserved < block_rsv->size) {
 		num_bytes = block_rsv->size - block_rsv->reserved;
-		block_rsv->reserved += num_bytes;
 		btrfs_space_info_update_bytes_may_use(fs_info, sinfo,
 						      num_bytes);
+		block_rsv->reserved = block_rsv->size;
 	} else if (block_rsv->reserved > block_rsv->size) {
 		num_bytes = block_rsv->reserved - block_rsv->size;
 		btrfs_space_info_update_bytes_may_use(fs_info, sinfo,
@@ -344,7 +436,8 @@ void btrfs_init_global_block_rsv(struct btrfs_fs_info *fs_info)
 
 void btrfs_release_global_block_rsv(struct btrfs_fs_info *fs_info)
 {
-	btrfs_block_rsv_release(fs_info, &fs_info->global_block_rsv, (u64)-1);
+	btrfs_block_rsv_release(fs_info, &fs_info->global_block_rsv, (u64)-1,
+				NULL);
 	WARN_ON(fs_info->trans_block_rsv.size > 0);
 	WARN_ON(fs_info->trans_block_rsv.reserved > 0);
 	WARN_ON(fs_info->chunk_block_rsv.size > 0);
diff --git a/fs/btrfs/block-rsv.h b/fs/btrfs/block-rsv.h
index d1428bb73fc5..0b6ae5302837 100644
--- a/fs/btrfs/block-rsv.h
+++ b/fs/btrfs/block-rsv.h
@@ -73,7 +73,7 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 			     int min_factor);
 void btrfs_block_rsv_add_bytes(struct btrfs_block_rsv *block_rsv,
 			       u64 num_bytes, bool update_size);
-u64 __btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
+u64 btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
 			      struct btrfs_block_rsv *block_rsv,
 			      u64 num_bytes, u64 *qgroup_to_release);
 void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info);
@@ -82,20 +82,12 @@ void btrfs_release_global_block_rsv(struct btrfs_fs_info *fs_info);
 struct btrfs_block_rsv *btrfs_use_block_rsv(struct btrfs_trans_handle *trans,
 					    struct btrfs_root *root,
 					    u32 blocksize);
-
-static inline void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info,
-					   struct btrfs_block_rsv *block_rsv,
-					   u64 num_bytes)
-{
-	__btrfs_block_rsv_release(fs_info, block_rsv, num_bytes, NULL);
-}
-
 static inline void btrfs_unuse_block_rsv(struct btrfs_fs_info *fs_info,
 					 struct btrfs_block_rsv *block_rsv,
 					 u32 blocksize)
 {
 	btrfs_block_rsv_add_bytes(block_rsv, blocksize, false);
-	btrfs_block_rsv_release(fs_info, block_rsv, 0);
+	btrfs_block_rsv_release(fs_info, block_rsv, 0, NULL);
 }
 
 #endif /* BTRFS_BLOCK_RSV_H */
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index f853835c409c..27a1fefce508 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -60,12 +60,15 @@ struct btrfs_inode {
 	 */
 	struct extent_io_tree io_failure_tree;
 
+	/*
+	 * Keep track of where the inode has extent items mapped in order to
+	 * make sure the i_size adjustments are accurate
+	 */
+	struct extent_io_tree file_extent_tree;
+
 	/* held while logging the inode in tree-log.c */
 	struct mutex log_mutex;
 
-	/* held while doing delalloc reservations */
-	struct mutex delalloc_mutex;
-
 	/* used to order data wrt metadata */
 	struct btrfs_ordered_inode_tree ordered_tree;
 
diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
index 0b52ab4cb964..32e11a23b47f 100644
--- a/fs/btrfs/check-integrity.c
+++ b/fs/btrfs/check-integrity.c
@@ -77,7 +77,6 @@
 
 #include <linux/sched.h>
 #include <linux/slab.h>
-#include <linux/buffer_head.h>
 #include <linux/mutex.h>
 #include <linux/genhd.h>
 #include <linux/blkdev.h>
@@ -152,11 +151,8 @@ struct btrfsic_block {
 	struct list_head ref_to_list;	/* list */
 	struct list_head ref_from_list;	/* list */
 	struct btrfsic_block *next_in_same_bio;
-	void *orig_bio_bh_private;
-	union {
-		bio_end_io_t *bio;
-		bh_end_io_t *bh;
-	} orig_bio_bh_end_io;
+	void *orig_bio_private;
+	bio_end_io_t *orig_bio_end_io;
 	int submit_bio_bh_rw;
 	u64 flush_gen; /* only valid if !never_written */
 };
@@ -325,14 +321,12 @@ static void btrfsic_process_written_block(struct btrfsic_dev_state *dev_state,
 					  u64 dev_bytenr, char **mapped_datav,
 					  unsigned int num_pages,
 					  struct bio *bio, int *bio_is_patched,
-					  struct buffer_head *bh,
 					  int submit_bio_bh_rw);
 static int btrfsic_process_written_superblock(
 		struct btrfsic_state *state,
 		struct btrfsic_block *const block,
 		struct btrfs_super_block *const super_hdr);
 static void btrfsic_bio_end_io(struct bio *bp);
-static void btrfsic_bh_end_io(struct buffer_head *bh, int uptodate);
 static int btrfsic_is_block_ref_by_superblock(const struct btrfsic_state *state,
 					      const struct btrfsic_block *block,
 					      int recursion_level);
@@ -399,8 +393,8 @@ static void btrfsic_block_init(struct btrfsic_block *b)
 	b->never_written = 0;
 	b->mirror_num = 0;
 	b->next_in_same_bio = NULL;
-	b->orig_bio_bh_private = NULL;
-	b->orig_bio_bh_end_io.bio = NULL;
+	b->orig_bio_private = NULL;
+	b->orig_bio_end_io = NULL;
 	INIT_LIST_HEAD(&b->collision_resolving_node);
 	INIT_LIST_HEAD(&b->all_blocks_node);
 	INIT_LIST_HEAD(&b->ref_to_list);
@@ -629,7 +623,6 @@ static struct btrfsic_dev_state *btrfsic_dev_state_hashtable_lookup(dev_t dev,
 static int btrfsic_process_superblock(struct btrfsic_state *state,
 				      struct btrfs_fs_devices *fs_devices)
 {
-	struct btrfs_fs_info *fs_info = state->fs_info;
 	struct btrfs_super_block *selected_super;
 	struct list_head *dev_head = &fs_devices->devices;
 	struct btrfs_device *device;
@@ -637,7 +630,6 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 	int ret = 0;
 	int pass;
 
-	BUG_ON(NULL == state);
 	selected_super = kzalloc(sizeof(*selected_super), GFP_NOFS);
 	if (NULL == selected_super) {
 		pr_info("btrfsic: error, kmalloc failed!\n");
@@ -700,7 +692,7 @@ static int btrfsic_process_superblock(struct btrfsic_state *state,
 			break;
 		}
 
-		num_copies = btrfs_num_copies(fs_info, next_bytenr,
+		num_copies = btrfs_num_copies(state->fs_info, next_bytenr,
 					      state->metablock_size);
 		if (state->print_mask & BTRFSIC_PRINT_MASK_NUM_COPIES)
 			pr_info("num_copies(log_bytenr=%llu) = %d\n",
@@ -769,29 +761,31 @@ static int btrfsic_process_superblock_dev_mirror(
 	struct btrfs_fs_info *fs_info = state->fs_info;
 	struct btrfs_super_block *super_tmp;
 	u64 dev_bytenr;
-	struct buffer_head *bh;
 	struct btrfsic_block *superblock_tmp;
 	int pass;
 	struct block_device *const superblock_bdev = device->bdev;
+	struct page *page;
+	struct address_space *mapping = superblock_bdev->bd_inode->i_mapping;
+	int ret = 0;
 
 	/* super block bytenr is always the unmapped device bytenr */
 	dev_bytenr = btrfs_sb_offset(superblock_mirror_num);
 	if (dev_bytenr + BTRFS_SUPER_INFO_SIZE > device->commit_total_bytes)
 		return -1;
-	bh = __bread(superblock_bdev, dev_bytenr / BTRFS_BDEV_BLOCKSIZE,
-		     BTRFS_SUPER_INFO_SIZE);
-	if (NULL == bh)
+
+	page = read_cache_page_gfp(mapping, dev_bytenr >> PAGE_SHIFT, GFP_NOFS);
+	if (IS_ERR(page))
 		return -1;
-	super_tmp = (struct btrfs_super_block *)
-	    (bh->b_data + (dev_bytenr & (BTRFS_BDEV_BLOCKSIZE - 1)));
+
+	super_tmp = page_address(page);
 
 	if (btrfs_super_bytenr(super_tmp) != dev_bytenr ||
 	    btrfs_super_magic(super_tmp) != BTRFS_MAGIC ||
 	    memcmp(device->uuid, super_tmp->dev_item.uuid, BTRFS_UUID_SIZE) ||
 	    btrfs_super_nodesize(super_tmp) != state->metablock_size ||
 	    btrfs_super_sectorsize(super_tmp) != state->datablock_size) {
-		brelse(bh);
-		return 0;
+		ret = 0;
+		goto out;
 	}
 
 	superblock_tmp =
@@ -802,8 +796,8 @@ static int btrfsic_process_superblock_dev_mirror(
 		superblock_tmp = btrfsic_block_alloc();
 		if (NULL == superblock_tmp) {
 			pr_info("btrfsic: error, kmalloc failed!\n");
-			brelse(bh);
-			return -1;
+			ret = -1;
+			goto out;
 		}
 		/* for superblock, only the dev_bytenr makes sense */
 		superblock_tmp->dev_bytenr = dev_bytenr;
@@ -887,8 +881,8 @@ static int btrfsic_process_superblock_dev_mirror(
 					      mirror_num)) {
 				pr_info("btrfsic: btrfsic_map_block(bytenr @%llu, mirror %d) failed!\n",
 				       next_bytenr, mirror_num);
-				brelse(bh);
-				return -1;
+				ret = -1;
+				goto out;
 			}
 
 			next_block = btrfsic_block_lookup_or_add(
@@ -897,8 +891,8 @@ static int btrfsic_process_superblock_dev_mirror(
 					mirror_num, NULL);
 			if (NULL == next_block) {
 				btrfsic_release_block_ctx(&tmp_next_block_ctx);
-				brelse(bh);
-				return -1;
+				ret = -1;
+				goto out;
 			}
 
 			next_block->disk_key = tmp_disk_key;
@@ -909,16 +903,17 @@ static int btrfsic_process_superblock_dev_mirror(
 					BTRFSIC_GENERATION_UNKNOWN);
 			btrfsic_release_block_ctx(&tmp_next_block_ctx);
 			if (NULL == l) {
-				brelse(bh);
-				return -1;
+				ret = -1;
+				goto out;
 			}
 		}
 	}
 	if (state->print_mask & BTRFSIC_PRINT_MASK_INITIAL_ALL_TREES)
 		btrfsic_dump_tree_sub(state, superblock_tmp, 0);
 
-	brelse(bh);
-	return 0;
+out:
+	put_page(page);
+	return ret;
 }
 
 static struct btrfsic_stack_frame *btrfsic_stack_frame_alloc(void)
@@ -1745,7 +1740,6 @@ static void btrfsic_process_written_block(struct btrfsic_dev_state *dev_state,
 					  u64 dev_bytenr, char **mapped_datav,
 					  unsigned int num_pages,
 					  struct bio *bio, int *bio_is_patched,
-					  struct buffer_head *bh,
 					  int submit_bio_bh_rw)
 {
 	int is_metadata;
@@ -1904,9 +1898,9 @@ again:
 				block->is_iodone = 0;
 				BUG_ON(NULL == bio_is_patched);
 				if (!*bio_is_patched) {
-					block->orig_bio_bh_private =
+					block->orig_bio_private =
 					    bio->bi_private;
-					block->orig_bio_bh_end_io.bio =
+					block->orig_bio_end_io =
 					    bio->bi_end_io;
 					block->next_in_same_bio = NULL;
 					bio->bi_private = block;
@@ -1918,25 +1912,17 @@ again:
 					    bio->bi_private;
 
 					BUG_ON(NULL == chained_block);
-					block->orig_bio_bh_private =
-					    chained_block->orig_bio_bh_private;
-					block->orig_bio_bh_end_io.bio =
-					    chained_block->orig_bio_bh_end_io.
-					    bio;
+					block->orig_bio_private =
+					    chained_block->orig_bio_private;
+					block->orig_bio_end_io =
+					    chained_block->orig_bio_end_io;
 					block->next_in_same_bio = chained_block;
 					bio->bi_private = block;
 				}
-			} else if (NULL != bh) {
-				block->is_iodone = 0;
-				block->orig_bio_bh_private = bh->b_private;
-				block->orig_bio_bh_end_io.bh = bh->b_end_io;
-				block->next_in_same_bio = NULL;
-				bh->b_private = block;
-				bh->b_end_io = btrfsic_bh_end_io;
 			} else {
 				block->is_iodone = 1;
-				block->orig_bio_bh_private = NULL;
-				block->orig_bio_bh_end_io.bio = NULL;
+				block->orig_bio_private = NULL;
+				block->orig_bio_end_io = NULL;
 				block->next_in_same_bio = NULL;
 			}
 		}
@@ -2044,8 +2030,8 @@ again:
 			block->is_iodone = 0;
 			BUG_ON(NULL == bio_is_patched);
 			if (!*bio_is_patched) {
-				block->orig_bio_bh_private = bio->bi_private;
-				block->orig_bio_bh_end_io.bio = bio->bi_end_io;
+				block->orig_bio_private = bio->bi_private;
+				block->orig_bio_end_io = bio->bi_end_io;
 				block->next_in_same_bio = NULL;
 				bio->bi_private = block;
 				bio->bi_end_io = btrfsic_bio_end_io;
@@ -2056,24 +2042,17 @@ again:
 				    bio->bi_private;
 
 				BUG_ON(NULL == chained_block);
-				block->orig_bio_bh_private =
-				    chained_block->orig_bio_bh_private;
-				block->orig_bio_bh_end_io.bio =
-				    chained_block->orig_bio_bh_end_io.bio;
+				block->orig_bio_private =
+				    chained_block->orig_bio_private;
+				block->orig_bio_end_io =
+				    chained_block->orig_bio_end_io;
 				block->next_in_same_bio = chained_block;
 				bio->bi_private = block;
 			}
-		} else if (NULL != bh) {
-			block->is_iodone = 0;
-			block->orig_bio_bh_private = bh->b_private;
-			block->orig_bio_bh_end_io.bh = bh->b_end_io;
-			block->next_in_same_bio = NULL;
-			bh->b_private = block;
-			bh->b_end_io = btrfsic_bh_end_io;
 		} else {
 			block->is_iodone = 1;
-			block->orig_bio_bh_private = NULL;
-			block->orig_bio_bh_end_io.bio = NULL;
+			block->orig_bio_private = NULL;
+			block->orig_bio_end_io = NULL;
 			block->next_in_same_bio = NULL;
 		}
 		if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
@@ -2114,8 +2093,8 @@ static void btrfsic_bio_end_io(struct bio *bp)
 		iodone_w_error = 1;
 
 	BUG_ON(NULL == block);
-	bp->bi_private = block->orig_bio_bh_private;
-	bp->bi_end_io = block->orig_bio_bh_end_io.bio;
+	bp->bi_private = block->orig_bio_private;
+	bp->bi_end_io = block->orig_bio_end_io;
 
 	do {
 		struct btrfsic_block *next_block;
@@ -2148,38 +2127,6 @@ static void btrfsic_bio_end_io(struct bio *bp)
 	bp->bi_end_io(bp);
 }
 
-static void btrfsic_bh_end_io(struct buffer_head *bh, int uptodate)
-{
-	struct btrfsic_block *block = (struct btrfsic_block *)bh->b_private;
-	int iodone_w_error = !uptodate;
-	struct btrfsic_dev_state *dev_state;
-
-	BUG_ON(NULL == block);
-	dev_state = block->dev_state;
-	if ((dev_state->state->print_mask & BTRFSIC_PRINT_MASK_END_IO_BIO_BH))
-		pr_info("bh_end_io(error=%d) for %c @%llu (%s/%llu/%d)\n",
-		       iodone_w_error,
-		       btrfsic_get_block_type(dev_state->state, block),
-		       block->logical_bytenr, block->dev_state->name,
-		       block->dev_bytenr, block->mirror_num);
-
-	block->iodone_w_error = iodone_w_error;
-	if (block->submit_bio_bh_rw & REQ_PREFLUSH) {
-		dev_state->last_flush_gen++;
-		if ((dev_state->state->print_mask &
-		     BTRFSIC_PRINT_MASK_END_IO_BIO_BH))
-			pr_info("bh_end_io() new %s flush_gen=%llu\n",
-			       dev_state->name, dev_state->last_flush_gen);
-	}
-	if (block->submit_bio_bh_rw & REQ_FUA)
-		block->flush_gen = 0; /* FUA completed means block is on disk */
-
-	bh->b_private = block->orig_bio_bh_private;
-	bh->b_end_io = block->orig_bio_bh_end_io.bh;
-	block->is_iodone = 1; /* for FLUSH, this releases the block */
-	bh->b_end_io(bh, uptodate);
-}
-
 static int btrfsic_process_written_superblock(
 		struct btrfsic_state *state,
 		struct btrfsic_block *const superblock,
@@ -2732,63 +2679,6 @@ static struct btrfsic_dev_state *btrfsic_dev_state_lookup(dev_t dev)
 						  &btrfsic_dev_state_hashtable);
 }
 
-int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh)
-{
-	struct btrfsic_dev_state *dev_state;
-
-	if (!btrfsic_is_initialized)
-		return submit_bh(op, op_flags, bh);
-
-	mutex_lock(&btrfsic_mutex);
-	/* since btrfsic_submit_bh() might also be called before
-	 * btrfsic_mount(), this might return NULL */
-	dev_state = btrfsic_dev_state_lookup(bh->b_bdev->bd_dev);
-
-	/* Only called to write the superblock (incl. FLUSH/FUA) */
-	if (NULL != dev_state &&
-	    (op == REQ_OP_WRITE) && bh->b_size > 0) {
-		u64 dev_bytenr;
-
-		dev_bytenr = BTRFS_BDEV_BLOCKSIZE * bh->b_blocknr;
-		if (dev_state->state->print_mask &
-		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
-			pr_info("submit_bh(op=0x%x,0x%x, blocknr=%llu (bytenr %llu), size=%zu, data=%p, bdev=%p)\n",
-			       op, op_flags, (unsigned long long)bh->b_blocknr,
-			       dev_bytenr, bh->b_size, bh->b_data, bh->b_bdev);
-		btrfsic_process_written_block(dev_state, dev_bytenr,
-					      &bh->b_data, 1, NULL,
-					      NULL, bh, op_flags);
-	} else if (NULL != dev_state && (op_flags & REQ_PREFLUSH)) {
-		if (dev_state->state->print_mask &
-		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
-			pr_info("submit_bh(op=0x%x,0x%x FLUSH, bdev=%p)\n",
-			       op, op_flags, bh->b_bdev);
-		if (!dev_state->dummy_block_for_bio_bh_flush.is_iodone) {
-			if ((dev_state->state->print_mask &
-			     (BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH |
-			      BTRFSIC_PRINT_MASK_VERBOSE)))
-				pr_info("btrfsic_submit_bh(%s) with FLUSH but dummy block already in use (ignored)!\n",
-				       dev_state->name);
-		} else {
-			struct btrfsic_block *const block =
-				&dev_state->dummy_block_for_bio_bh_flush;
-
-			block->is_iodone = 0;
-			block->never_written = 0;
-			block->iodone_w_error = 0;
-			block->flush_gen = dev_state->last_flush_gen + 1;
-			block->submit_bio_bh_rw = op_flags;
-			block->orig_bio_bh_private = bh->b_private;
-			block->orig_bio_bh_end_io.bh = bh->b_end_io;
-			block->next_in_same_bio = NULL;
-			bh->b_private = block;
-			bh->b_end_io = btrfsic_bh_end_io;
-		}
-	}
-	mutex_unlock(&btrfsic_mutex);
-	return submit_bh(op, op_flags, bh);
-}
-
 static void __btrfsic_submit_bio(struct bio *bio)
 {
 	struct btrfsic_dev_state *dev_state;
@@ -2840,7 +2730,7 @@ static void __btrfsic_submit_bio(struct bio *bio)
 		btrfsic_process_written_block(dev_state, dev_bytenr,
 					      mapped_datav, segs,
 					      bio, &bio_is_patched,
-					      NULL, bio->bi_opf);
+					      bio->bi_opf);
 		bio_for_each_segment(bvec, bio, iter)
 			kunmap(bvec.bv_page);
 		kfree(mapped_datav);
@@ -2864,8 +2754,8 @@ static void __btrfsic_submit_bio(struct bio *bio)
 			block->iodone_w_error = 0;
 			block->flush_gen = dev_state->last_flush_gen + 1;
 			block->submit_bio_bh_rw = bio->bi_opf;
-			block->orig_bio_bh_private = bio->bi_private;
-			block->orig_bio_bh_end_io.bio = bio->bi_end_io;
+			block->orig_bio_private = bio->bi_private;
+			block->orig_bio_end_io = bio->bi_end_io;
 			block->next_in_same_bio = NULL;
 			bio->bi_private = block;
 			bio->bi_end_io = btrfsic_bio_end_io;
diff --git a/fs/btrfs/check-integrity.h b/fs/btrfs/check-integrity.h
index 9bf4359cc44c..bcc730a06cb5 100644
--- a/fs/btrfs/check-integrity.h
+++ b/fs/btrfs/check-integrity.h
@@ -7,11 +7,9 @@
 #define BTRFS_CHECK_INTEGRITY_H
 
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
-int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh);
 void btrfsic_submit_bio(struct bio *bio);
 int btrfsic_submit_bio_wait(struct bio *bio);
 #else
-#define btrfsic_submit_bh submit_bh
 #define btrfsic_submit_bio submit_bio
 #define btrfsic_submit_bio_wait submit_bio_wait
 #endif
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index b05b361e2062..9ab610cc9114 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -29,6 +29,41 @@
 #include "extent_io.h"
 #include "extent_map.h"
 
+int zlib_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out);
+int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
+int zlib_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen);
+struct list_head *zlib_alloc_workspace(unsigned int level);
+void zlib_free_workspace(struct list_head *ws);
+struct list_head *zlib_get_workspace(unsigned int level);
+
+int lzo_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out);
+int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
+int lzo_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen);
+struct list_head *lzo_alloc_workspace(unsigned int level);
+void lzo_free_workspace(struct list_head *ws);
+
+int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out);
+int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb);
+int zstd_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen);
+void zstd_init_workspace_manager(void);
+void zstd_cleanup_workspace_manager(void);
+struct list_head *zstd_alloc_workspace(unsigned int level);
+void zstd_free_workspace(struct list_head *ws);
+struct list_head *zstd_get_workspace(unsigned int level);
+void zstd_put_workspace(struct list_head *ws);
+
 static const char* const btrfs_compress_types[] = { "", "zlib", "lzo", "zstd" };
 
 const char* btrfs_compress_type2str(enum btrfs_compression_type type)
@@ -39,6 +74,8 @@ const char* btrfs_compress_type2str(enum btrfs_compression_type type)
 	case BTRFS_COMPRESS_ZSTD:
 	case BTRFS_COMPRESS_NONE:
 		return btrfs_compress_types[type];
+	default:
+		break;
 	}
 
 	return NULL;
@@ -60,6 +97,70 @@ bool btrfs_compress_is_valid_type(const char *str, size_t len)
 	return false;
 }
 
+static int compression_compress_pages(int type, struct list_head *ws,
+               struct address_space *mapping, u64 start, struct page **pages,
+               unsigned long *out_pages, unsigned long *total_in,
+               unsigned long *total_out)
+{
+	switch (type) {
+	case BTRFS_COMPRESS_ZLIB:
+		return zlib_compress_pages(ws, mapping, start, pages,
+				out_pages, total_in, total_out);
+	case BTRFS_COMPRESS_LZO:
+		return lzo_compress_pages(ws, mapping, start, pages,
+				out_pages, total_in, total_out);
+	case BTRFS_COMPRESS_ZSTD:
+		return zstd_compress_pages(ws, mapping, start, pages,
+				out_pages, total_in, total_out);
+	case BTRFS_COMPRESS_NONE:
+	default:
+		/*
+		 * This can't happen, the type is validated several times
+		 * before we get here. As a sane fallback, return what the
+		 * callers will understand as 'no compression happened'.
+		 */
+		return -E2BIG;
+	}
+}
+
+static int compression_decompress_bio(int type, struct list_head *ws,
+		struct compressed_bio *cb)
+{
+	switch (type) {
+	case BTRFS_COMPRESS_ZLIB: return zlib_decompress_bio(ws, cb);
+	case BTRFS_COMPRESS_LZO:  return lzo_decompress_bio(ws, cb);
+	case BTRFS_COMPRESS_ZSTD: return zstd_decompress_bio(ws, cb);
+	case BTRFS_COMPRESS_NONE:
+	default:
+		/*
+		 * This can't happen, the type is validated several times
+		 * before we get here.
+		 */
+		BUG();
+	}
+}
+
+static int compression_decompress(int type, struct list_head *ws,
+               unsigned char *data_in, struct page *dest_page,
+               unsigned long start_byte, size_t srclen, size_t destlen)
+{
+	switch (type) {
+	case BTRFS_COMPRESS_ZLIB: return zlib_decompress(ws, data_in, dest_page,
+						start_byte, srclen, destlen);
+	case BTRFS_COMPRESS_LZO:  return lzo_decompress(ws, data_in, dest_page,
+						start_byte, srclen, destlen);
+	case BTRFS_COMPRESS_ZSTD: return zstd_decompress(ws, data_in, dest_page,
+						start_byte, srclen, destlen);
+	case BTRFS_COMPRESS_NONE:
+	default:
+		/*
+		 * This can't happen, the type is validated several times
+		 * before we get here.
+		 */
+		BUG();
+	}
+}
+
 static int btrfs_decompress_bio(struct compressed_bio *cb);
 
 static inline int compressed_bio_size(struct btrfs_fs_info *fs_info,
@@ -311,7 +412,8 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 				 unsigned long compressed_len,
 				 struct page **compressed_pages,
 				 unsigned long nr_pages,
-				 unsigned int write_flags)
+				 unsigned int write_flags,
+				 struct cgroup_subsys_state *blkcg_css)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct bio *bio = NULL;
@@ -320,7 +422,6 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 	int pg_index = 0;
 	struct page *page;
 	u64 first_byte = disk_start;
-	struct block_device *bdev;
 	blk_status_t ret;
 	int skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;
 
@@ -339,13 +440,15 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 	cb->orig_bio = NULL;
 	cb->nr_pages = nr_pages;
 
-	bdev = fs_info->fs_devices->latest_bdev;
-
 	bio = btrfs_bio_alloc(first_byte);
-	bio_set_dev(bio, bdev);
 	bio->bi_opf = REQ_OP_WRITE | write_flags;
 	bio->bi_private = cb;
 	bio->bi_end_io = end_compressed_bio_write;
+
+	if (blkcg_css) {
+		bio->bi_opf |= REQ_CGROUP_PUNT;
+		kthread_associate_blkcg(blkcg_css);
+	}
 	refcount_set(&cb->pending_bios, 1);
 
 	/* create and submit bios for the compressed pages */
@@ -378,17 +481,18 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 				BUG_ON(ret); /* -ENOMEM */
 			}
 
-			ret = btrfs_map_bio(fs_info, bio, 0, 1);
+			ret = btrfs_map_bio(fs_info, bio, 0);
 			if (ret) {
 				bio->bi_status = ret;
 				bio_endio(bio);
 			}
 
 			bio = btrfs_bio_alloc(first_byte);
-			bio_set_dev(bio, bdev);
 			bio->bi_opf = REQ_OP_WRITE | write_flags;
 			bio->bi_private = cb;
 			bio->bi_end_io = end_compressed_bio_write;
+			if (blkcg_css)
+				bio->bi_opf |= REQ_CGROUP_PUNT;
 			bio_add_page(bio, page, PAGE_SIZE, 0);
 		}
 		if (bytes_left < PAGE_SIZE) {
@@ -409,12 +513,15 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 		BUG_ON(ret); /* -ENOMEM */
 	}
 
-	ret = btrfs_map_bio(fs_info, bio, 0, 1);
+	ret = btrfs_map_bio(fs_info, bio, 0);
 	if (ret) {
 		bio->bi_status = ret;
 		bio_endio(bio);
 	}
 
+	if (blkcg_css)
+		kthread_associate_blkcg(NULL);
+
 	return 0;
 }
 
@@ -553,7 +660,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	unsigned long nr_pages;
 	unsigned long pg_index;
 	struct page *page;
-	struct block_device *bdev;
 	struct bio *comp_bio;
 	u64 cur_disk_byte = (u64)bio->bi_iter.bi_sector << 9;
 	u64 em_len;
@@ -604,8 +710,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	if (!cb->compressed_pages)
 		goto fail1;
 
-	bdev = fs_info->fs_devices->latest_bdev;
-
 	for (pg_index = 0; pg_index < nr_pages; pg_index++) {
 		cb->compressed_pages[pg_index] = alloc_page(GFP_NOFS |
 							      __GFP_HIGHMEM);
@@ -624,7 +728,6 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	cb->len = bio->bi_iter.bi_size;
 
 	comp_bio = btrfs_bio_alloc(cur_disk_byte);
-	bio_set_dev(comp_bio, bdev);
 	comp_bio->bi_opf = REQ_OP_READ;
 	comp_bio->bi_private = cb;
 	comp_bio->bi_end_io = end_compressed_bio_read;
@@ -660,7 +763,7 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 
 			if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
 				ret = btrfs_lookup_bio_sums(inode, comp_bio,
-							    sums);
+							    (u64)-1, sums);
 				BUG_ON(ret); /* -ENOMEM */
 			}
 
@@ -668,14 +771,13 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 						  fs_info->sectorsize);
 			sums += csum_size * nr_sectors;
 
-			ret = btrfs_map_bio(fs_info, comp_bio, mirror_num, 0);
+			ret = btrfs_map_bio(fs_info, comp_bio, mirror_num);
 			if (ret) {
 				comp_bio->bi_status = ret;
 				bio_endio(comp_bio);
 			}
 
 			comp_bio = btrfs_bio_alloc(cur_disk_byte);
-			bio_set_dev(comp_bio, bdev);
 			comp_bio->bi_opf = REQ_OP_READ;
 			comp_bio->bi_private = cb;
 			comp_bio->bi_end_io = end_compressed_bio_read;
@@ -689,11 +791,11 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	BUG_ON(ret); /* -ENOMEM */
 
 	if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
-		ret = btrfs_lookup_bio_sums(inode, comp_bio, sums);
+		ret = btrfs_lookup_bio_sums(inode, comp_bio, (u64)-1, sums);
 		BUG_ON(ret); /* -ENOMEM */
 	}
 
-	ret = btrfs_map_bio(fs_info, comp_bio, mirror_num, 0);
+	ret = btrfs_map_bio(fs_info, comp_bio, mirror_num);
 	if (ret) {
 		comp_bio->bi_status = ret;
 		bio_endio(comp_bio);
@@ -764,26 +866,6 @@ struct heuristic_ws {
 
 static struct workspace_manager heuristic_wsm;
 
-static void heuristic_init_workspace_manager(void)
-{
-	btrfs_init_workspace_manager(&heuristic_wsm, &btrfs_heuristic_compress);
-}
-
-static void heuristic_cleanup_workspace_manager(void)
-{
-	btrfs_cleanup_workspace_manager(&heuristic_wsm);
-}
-
-static struct list_head *heuristic_get_workspace(unsigned int level)
-{
-	return btrfs_get_workspace(&heuristic_wsm, level);
-}
-
-static void heuristic_put_workspace(struct list_head *ws)
-{
-	btrfs_put_workspace(&heuristic_wsm, ws);
-}
-
 static void free_heuristic_ws(struct list_head *ws)
 {
 	struct heuristic_ws *workspace;
@@ -824,12 +906,7 @@ fail:
 }
 
 const struct btrfs_compress_op btrfs_heuristic_compress = {
-	.init_workspace_manager = heuristic_init_workspace_manager,
-	.cleanup_workspace_manager = heuristic_cleanup_workspace_manager,
-	.get_workspace = heuristic_get_workspace,
-	.put_workspace = heuristic_put_workspace,
-	.alloc_workspace = alloc_heuristic_ws,
-	.free_workspace = free_heuristic_ws,
+	.workspace_manager = &heuristic_wsm,
 };
 
 static const struct btrfs_compress_op * const btrfs_compress_op[] = {
@@ -840,13 +917,44 @@ static const struct btrfs_compress_op * const btrfs_compress_op[] = {
 	&btrfs_zstd_compress,
 };
 
-void btrfs_init_workspace_manager(struct workspace_manager *wsm,
-				  const struct btrfs_compress_op *ops)
+static struct list_head *alloc_workspace(int type, unsigned int level)
 {
-	struct list_head *workspace;
+	switch (type) {
+	case BTRFS_COMPRESS_NONE: return alloc_heuristic_ws(level);
+	case BTRFS_COMPRESS_ZLIB: return zlib_alloc_workspace(level);
+	case BTRFS_COMPRESS_LZO:  return lzo_alloc_workspace(level);
+	case BTRFS_COMPRESS_ZSTD: return zstd_alloc_workspace(level);
+	default:
+		/*
+		 * This can't happen, the type is validated several times
+		 * before we get here.
+		 */
+		BUG();
+	}
+}
+
+static void free_workspace(int type, struct list_head *ws)
+{
+	switch (type) {
+	case BTRFS_COMPRESS_NONE: return free_heuristic_ws(ws);
+	case BTRFS_COMPRESS_ZLIB: return zlib_free_workspace(ws);
+	case BTRFS_COMPRESS_LZO:  return lzo_free_workspace(ws);
+	case BTRFS_COMPRESS_ZSTD: return zstd_free_workspace(ws);
+	default:
+		/*
+		 * This can't happen, the type is validated several times
+		 * before we get here.
+		 */
+		BUG();
+	}
+}
 
-	wsm->ops = ops;
+static void btrfs_init_workspace_manager(int type)
+{
+	struct workspace_manager *wsm;
+	struct list_head *workspace;
 
+	wsm = btrfs_compress_op[type]->workspace_manager;
 	INIT_LIST_HEAD(&wsm->idle_ws);
 	spin_lock_init(&wsm->ws_lock);
 	atomic_set(&wsm->total_ws, 0);
@@ -856,7 +964,7 @@ void btrfs_init_workspace_manager(struct workspace_manager *wsm,
 	 * Preallocate one workspace for each compression type so we can
 	 * guarantee forward progress in the worst case
 	 */
-	workspace = wsm->ops->alloc_workspace(0);
+	workspace = alloc_workspace(type, 0);
 	if (IS_ERR(workspace)) {
 		pr_warn(
 	"BTRFS: cannot preallocate compression workspace, will try later\n");
@@ -867,14 +975,16 @@ void btrfs_init_workspace_manager(struct workspace_manager *wsm,
 	}
 }
 
-void btrfs_cleanup_workspace_manager(struct workspace_manager *wsman)
+static void btrfs_cleanup_workspace_manager(int type)
 {
+	struct workspace_manager *wsman;
 	struct list_head *ws;
 
+	wsman = btrfs_compress_op[type]->workspace_manager;
 	while (!list_empty(&wsman->idle_ws)) {
 		ws = wsman->idle_ws.next;
 		list_del(ws);
-		wsman->ops->free_workspace(ws);
+		free_workspace(type, ws);
 		atomic_dec(&wsman->total_ws);
 	}
 }
@@ -885,9 +995,9 @@ void btrfs_cleanup_workspace_manager(struct workspace_manager *wsman)
  * Preallocation makes a forward progress guarantees and we do not return
  * errors.
  */
-struct list_head *btrfs_get_workspace(struct workspace_manager *wsm,
-				      unsigned int level)
+struct list_head *btrfs_get_workspace(int type, unsigned int level)
 {
+	struct workspace_manager *wsm;
 	struct list_head *workspace;
 	int cpus = num_online_cpus();
 	unsigned nofs_flag;
@@ -897,6 +1007,7 @@ struct list_head *btrfs_get_workspace(struct workspace_manager *wsm,
 	wait_queue_head_t *ws_wait;
 	int *free_ws;
 
+	wsm = btrfs_compress_op[type]->workspace_manager;
 	idle_ws	 = &wsm->idle_ws;
 	ws_lock	 = &wsm->ws_lock;
 	total_ws = &wsm->total_ws;
@@ -932,7 +1043,7 @@ again:
 	 * context of btrfs_compress_bio/btrfs_compress_pages
 	 */
 	nofs_flag = memalloc_nofs_save();
-	workspace = wsm->ops->alloc_workspace(level);
+	workspace = alloc_workspace(type, level);
 	memalloc_nofs_restore(nofs_flag);
 
 	if (IS_ERR(workspace)) {
@@ -965,21 +1076,34 @@ again:
 
 static struct list_head *get_workspace(int type, int level)
 {
-	return btrfs_compress_op[type]->get_workspace(level);
+	switch (type) {
+	case BTRFS_COMPRESS_NONE: return btrfs_get_workspace(type, level);
+	case BTRFS_COMPRESS_ZLIB: return zlib_get_workspace(level);
+	case BTRFS_COMPRESS_LZO:  return btrfs_get_workspace(type, level);
+	case BTRFS_COMPRESS_ZSTD: return zstd_get_workspace(level);
+	default:
+		/*
+		 * This can't happen, the type is validated several times
+		 * before we get here.
+		 */
+		BUG();
+	}
 }
 
 /*
  * put a workspace struct back on the list or free it if we have enough
  * idle ones sitting around
  */
-void btrfs_put_workspace(struct workspace_manager *wsm, struct list_head *ws)
+void btrfs_put_workspace(int type, struct list_head *ws)
 {
+	struct workspace_manager *wsm;
 	struct list_head *idle_ws;
 	spinlock_t *ws_lock;
 	atomic_t *total_ws;
 	wait_queue_head_t *ws_wait;
 	int *free_ws;
 
+	wsm = btrfs_compress_op[type]->workspace_manager;
 	idle_ws	 = &wsm->idle_ws;
 	ws_lock	 = &wsm->ws_lock;
 	total_ws = &wsm->total_ws;
@@ -995,7 +1119,7 @@ void btrfs_put_workspace(struct workspace_manager *wsm, struct list_head *ws)
 	}
 	spin_unlock(ws_lock);
 
-	wsm->ops->free_workspace(ws);
+	free_workspace(type, ws);
 	atomic_dec(total_ws);
 wake:
 	cond_wake_up(ws_wait);
@@ -1003,7 +1127,18 @@ wake:
 
 static void put_workspace(int type, struct list_head *ws)
 {
-	return btrfs_compress_op[type]->put_workspace(ws);
+	switch (type) {
+	case BTRFS_COMPRESS_NONE: return btrfs_put_workspace(type, ws);
+	case BTRFS_COMPRESS_ZLIB: return btrfs_put_workspace(type, ws);
+	case BTRFS_COMPRESS_LZO:  return btrfs_put_workspace(type, ws);
+	case BTRFS_COMPRESS_ZSTD: return zstd_put_workspace(ws);
+	default:
+		/*
+		 * This can't happen, the type is validated several times
+		 * before we get here.
+		 */
+		BUG();
+	}
 }
 
 /*
@@ -1042,10 +1177,8 @@ int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping,
 
 	level = btrfs_compress_set_level(type, level);
 	workspace = get_workspace(type, level);
-	ret = btrfs_compress_op[type]->compress_pages(workspace, mapping,
-						      start, pages,
-						      out_pages,
-						      total_in, total_out);
+	ret = compression_compress_pages(type, workspace, mapping, start, pages,
+					 out_pages, total_in, total_out);
 	put_workspace(type, workspace);
 	return ret;
 }
@@ -1071,7 +1204,7 @@ static int btrfs_decompress_bio(struct compressed_bio *cb)
 	int type = cb->compress_type;
 
 	workspace = get_workspace(type, 0);
-	ret = btrfs_compress_op[type]->decompress_bio(workspace, cb);
+	ret = compression_decompress_bio(type, workspace, cb);
 	put_workspace(type, workspace);
 
 	return ret;
@@ -1089,9 +1222,8 @@ int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
 	int ret;
 
 	workspace = get_workspace(type, 0);
-	ret = btrfs_compress_op[type]->decompress(workspace, data_in,
-						  dest_page, start_byte,
-						  srclen, destlen);
+	ret = compression_decompress(type, workspace, data_in, dest_page,
+				     start_byte, srclen, destlen);
 	put_workspace(type, workspace);
 
 	return ret;
@@ -1099,18 +1231,18 @@ int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
 
 void __init btrfs_init_compress(void)
 {
-	int i;
-
-	for (i = 0; i < BTRFS_NR_WORKSPACE_MANAGERS; i++)
-		btrfs_compress_op[i]->init_workspace_manager();
+	btrfs_init_workspace_manager(BTRFS_COMPRESS_NONE);
+	btrfs_init_workspace_manager(BTRFS_COMPRESS_ZLIB);
+	btrfs_init_workspace_manager(BTRFS_COMPRESS_LZO);
+	zstd_init_workspace_manager();
 }
 
 void __cold btrfs_exit_compress(void)
 {
-	int i;
-
-	for (i = 0; i < BTRFS_NR_WORKSPACE_MANAGERS; i++)
-		btrfs_compress_op[i]->cleanup_workspace_manager();
+	btrfs_cleanup_workspace_manager(BTRFS_COMPRESS_NONE);
+	btrfs_cleanup_workspace_manager(BTRFS_COMPRESS_ZLIB);
+	btrfs_cleanup_workspace_manager(BTRFS_COMPRESS_LZO);
+	zstd_cleanup_workspace_manager();
 }
 
 /*
@@ -1158,7 +1290,7 @@ int btrfs_decompress_buf2page(const char *buf, unsigned long buf_start,
 	/* copy bytes from the working buffer into the pages */
 	while (working_bytes > 0) {
 		bytes = min_t(unsigned long, bvec.bv_len,
-				PAGE_SIZE - buf_offset);
+				PAGE_SIZE - (buf_offset % PAGE_SIZE));
 		bytes = min(bytes, working_bytes);
 
 		kaddr = kmap_atomic(bvec.bv_page);
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index 4cb8be9ff88b..d253f7aa8ed5 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -93,7 +93,8 @@ blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
 				  unsigned long compressed_len,
 				  struct page **compressed_pages,
 				  unsigned long nr_pages,
-				  unsigned int write_flags);
+				  unsigned int write_flags,
+				  struct cgroup_subsys_state *blkcg_css);
 blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 				 int mirror_num, unsigned long bio_flags);
 
@@ -104,11 +105,10 @@ enum btrfs_compression_type {
 	BTRFS_COMPRESS_ZLIB  = 1,
 	BTRFS_COMPRESS_LZO   = 2,
 	BTRFS_COMPRESS_ZSTD  = 3,
-	BTRFS_COMPRESS_TYPES = 3,
+	BTRFS_NR_COMPRESS_TYPES = 4,
 };
 
 struct workspace_manager {
-	const struct btrfs_compress_op *ops;
 	struct list_head idle_ws;
 	spinlock_t ws_lock;
 	/* Number of free workspaces */
@@ -119,50 +119,18 @@ struct workspace_manager {
 	wait_queue_head_t ws_wait;
 };
 
-void btrfs_init_workspace_manager(struct workspace_manager *wsm,
-				  const struct btrfs_compress_op *ops);
-struct list_head *btrfs_get_workspace(struct workspace_manager *wsm,
-				      unsigned int level);
-void btrfs_put_workspace(struct workspace_manager *wsm, struct list_head *ws);
-void btrfs_cleanup_workspace_manager(struct workspace_manager *wsm);
+struct list_head *btrfs_get_workspace(int type, unsigned int level);
+void btrfs_put_workspace(int type, struct list_head *ws);
 
 struct btrfs_compress_op {
-	void (*init_workspace_manager)(void);
-
-	void (*cleanup_workspace_manager)(void);
-
-	struct list_head *(*get_workspace)(unsigned int level);
-
-	void (*put_workspace)(struct list_head *ws);
-
-	struct list_head *(*alloc_workspace)(unsigned int level);
-
-	void (*free_workspace)(struct list_head *workspace);
-
-	int (*compress_pages)(struct list_head *workspace,
-			      struct address_space *mapping,
-			      u64 start,
-			      struct page **pages,
-			      unsigned long *out_pages,
-			      unsigned long *total_in,
-			      unsigned long *total_out);
-
-	int (*decompress_bio)(struct list_head *workspace,
-				struct compressed_bio *cb);
-
-	int (*decompress)(struct list_head *workspace,
-			  unsigned char *data_in,
-			  struct page *dest_page,
-			  unsigned long start_byte,
-			  size_t srclen, size_t destlen);
-
+	struct workspace_manager *workspace_manager;
 	/* Maximum level supported by the compression algorithm */
 	unsigned int max_level;
 	unsigned int default_level;
 };
 
 /* The heuristic workspaces are managed via the 0th workspace manager */
-#define BTRFS_NR_WORKSPACE_MANAGERS	(BTRFS_COMPRESS_TYPES + 1)
+#define BTRFS_NR_WORKSPACE_MANAGERS	BTRFS_NR_COMPRESS_TYPES
 
 extern const struct btrfs_compress_op btrfs_heuristic_compress;
 extern const struct btrfs_compress_op btrfs_zlib_compress;
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index e59cde204b2f..bfedbbe2311f 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -31,9 +31,14 @@ static void del_ptr(struct btrfs_root *root, struct btrfs_path *path,
 
 static const struct btrfs_csums {
 	u16		size;
-	const char	*name;
+	const char	name[10];
+	const char	driver[12];
 } btrfs_csums[] = {
 	[BTRFS_CSUM_TYPE_CRC32] = { .size = 4, .name = "crc32c" },
+	[BTRFS_CSUM_TYPE_XXHASH] = { .size = 8, .name = "xxhash64" },
+	[BTRFS_CSUM_TYPE_SHA256] = { .size = 32, .name = "sha256" },
+	[BTRFS_CSUM_TYPE_BLAKE2] = { .size = 32, .name = "blake2b",
+				     .driver = "blake2b-256" },
 };
 
 int btrfs_super_csum_size(const struct btrfs_super_block *s)
@@ -51,34 +56,26 @@ const char *btrfs_super_csum_name(u16 csum_type)
 	return btrfs_csums[csum_type].name;
 }
 
-struct btrfs_path *btrfs_alloc_path(void)
+/*
+ * Return driver name if defined, otherwise the name that's also a valid driver
+ * name
+ */
+const char *btrfs_super_csum_driver(u16 csum_type)
 {
-	return kmem_cache_zalloc(btrfs_path_cachep, GFP_NOFS);
+	/* csum type is validated at mount time */
+	return btrfs_csums[csum_type].driver[0] ?
+		btrfs_csums[csum_type].driver :
+		btrfs_csums[csum_type].name;
 }
 
-/*
- * set all locked nodes in the path to blocking locks.  This should
- * be done before scheduling
- */
-noinline void btrfs_set_path_blocking(struct btrfs_path *p)
+size_t __const btrfs_get_num_csums(void)
 {
-	int i;
-	for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
-		if (!p->nodes[i] || !p->locks[i])
-			continue;
-		/*
-		 * If we currently have a spinning reader or writer lock this
-		 * will bump the count of blocking holders and drop the
-		 * spinlock.
-		 */
-		if (p->locks[i] == BTRFS_READ_LOCK) {
-			btrfs_set_lock_blocking_read(p->nodes[i]);
-			p->locks[i] = BTRFS_READ_LOCK_BLOCKING;
-		} else if (p->locks[i] == BTRFS_WRITE_LOCK) {
-			btrfs_set_lock_blocking_write(p->nodes[i]);
-			p->locks[i] = BTRFS_WRITE_LOCK_BLOCKING;
-		}
-	}
+	return ARRAY_SIZE(btrfs_csums);
+}
+
+struct btrfs_path *btrfs_alloc_path(void)
+{
+	return kmem_cache_zalloc(btrfs_path_cachep, GFP_NOFS);
 }
 
 /* this also releases the path */
@@ -147,44 +144,6 @@ struct extent_buffer *btrfs_root_node(struct btrfs_root *root)
 	return eb;
 }
 
-/* loop around taking references on and locking the root node of the
- * tree until you end up with a lock on the root.  A locked buffer
- * is returned, with a reference held.
- */
-struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root)
-{
-	struct extent_buffer *eb;
-
-	while (1) {
-		eb = btrfs_root_node(root);
-		btrfs_tree_lock(eb);
-		if (eb == root->node)
-			break;
-		btrfs_tree_unlock(eb);
-		free_extent_buffer(eb);
-	}
-	return eb;
-}
-
-/* loop around taking references on and locking the root node of the
- * tree until you end up with a lock on the root.  A locked buffer
- * is returned, with a reference held.
- */
-struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
-{
-	struct extent_buffer *eb;
-
-	while (1) {
-		eb = btrfs_root_node(root);
-		btrfs_tree_read_lock(eb);
-		if (eb == root->node)
-			break;
-		btrfs_tree_read_unlock(eb);
-		free_extent_buffer(eb);
-	}
-	return eb;
-}
-
 /* cowonly root (everything not a reference counted cow subvolume), just get
  * put onto a simple dirty list.  transaction.c walks this to make sure they
  * get properly updated on disk.
@@ -330,12 +289,10 @@ u64 btrfs_get_tree_mod_seq(struct btrfs_fs_info *fs_info,
 			   struct seq_list *elem)
 {
 	write_lock(&fs_info->tree_mod_log_lock);
-	spin_lock(&fs_info->tree_mod_seq_lock);
 	if (!elem->seq) {
 		elem->seq = btrfs_inc_tree_mod_seq(fs_info);
 		list_add_tail(&elem->list, &fs_info->tree_mod_seq_list);
 	}
-	spin_unlock(&fs_info->tree_mod_seq_lock);
 	write_unlock(&fs_info->tree_mod_log_lock);
 
 	return elem->seq;
@@ -347,7 +304,6 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 	struct rb_root *tm_root;
 	struct rb_node *node;
 	struct rb_node *next;
-	struct seq_list *cur_elem;
 	struct tree_mod_elem *tm;
 	u64 min_seq = (u64)-1;
 	u64 seq_putting = elem->seq;
@@ -355,35 +311,35 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 	if (!seq_putting)
 		return;
 
-	spin_lock(&fs_info->tree_mod_seq_lock);
+	write_lock(&fs_info->tree_mod_log_lock);
 	list_del(&elem->list);
 	elem->seq = 0;
 
-	list_for_each_entry(cur_elem, &fs_info->tree_mod_seq_list, list) {
-		if (cur_elem->seq < min_seq) {
-			if (seq_putting > cur_elem->seq) {
-				/*
-				 * blocker with lower sequence number exists, we
-				 * cannot remove anything from the log
-				 */
-				spin_unlock(&fs_info->tree_mod_seq_lock);
-				return;
-			}
-			min_seq = cur_elem->seq;
+	if (!list_empty(&fs_info->tree_mod_seq_list)) {
+		struct seq_list *first;
+
+		first = list_first_entry(&fs_info->tree_mod_seq_list,
+					 struct seq_list, list);
+		if (seq_putting > first->seq) {
+			/*
+			 * Blocker with lower sequence number exists, we
+			 * cannot remove anything from the log.
+			 */
+			write_unlock(&fs_info->tree_mod_log_lock);
+			return;
 		}
+		min_seq = first->seq;
 	}
-	spin_unlock(&fs_info->tree_mod_seq_lock);
 
 	/*
 	 * anything that's lower than the lowest existing (read: blocked)
 	 * sequence number can be removed from the tree.
 	 */
-	write_lock(&fs_info->tree_mod_log_lock);
 	tm_root = &fs_info->tree_mod_log;
 	for (node = rb_first(tm_root); node; node = next) {
 		next = rb_next(node);
 		tm = rb_entry(node, struct tree_mod_elem, node);
-		if (tm->seq > min_seq)
+		if (tm->seq >= min_seq)
 			continue;
 		rb_erase(node, tm_root);
 		kfree(tm);
@@ -970,9 +926,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 		if (new_flags != 0) {
 			int level = btrfs_header_level(buf);
 
-			ret = btrfs_set_disk_extent_flags(trans,
-							  buf->start,
-							  buf->len,
+			ret = btrfs_set_disk_extent_flags(trans, buf,
 							  new_flags, level, 0);
 			if (ret)
 				return ret;
@@ -1125,7 +1079,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		    btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV)
 			parent_start = buf->start;
 
-		extent_buffer_get(cow);
+		atomic_inc(&cow->refs);
 		ret = tree_mod_log_insert_root(root->node, cow, 1);
 		BUG_ON(ret < 0);
 		rcu_assign_pointer(root->node, cow);
@@ -1563,7 +1517,7 @@ static int comp_keys(const struct btrfs_disk_key *disk,
 /*
  * same as comp_keys only with two btrfs_key's
  */
-int btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2)
+int __pure btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2)
 {
 	if (k1->objectid > k2->objectid)
 		return 1;
@@ -2036,7 +1990,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 	/* update the path */
 	if (left) {
 		if (btrfs_header_nritems(left) > orig_slot) {
-			extent_buffer_get(left);
+			atomic_inc(&left->refs);
 			/* left was locked after cow */
 			path->nodes[level] = left;
 			path->slots[level + 1] -= 1;
@@ -2379,32 +2333,6 @@ static noinline void unlock_up(struct btrfs_path *path, int level,
 }
 
 /*
- * This releases any locks held in the path starting at level and
- * going all the way up to the root.
- *
- * btrfs_search_slot will keep the lock held on higher nodes in a few
- * corner cases, such as COW of the block at slot zero in the node.  This
- * ignores those rules, and it should only be called when there are no
- * more updates to be done higher up in the tree.
- */
-noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level)
-{
-	int i;
-
-	if (path->keep_locks)
-		return;
-
-	for (i = level; i < BTRFS_MAX_LEVEL; i++) {
-		if (!path->nodes[i])
-			continue;
-		if (!path->locks[i])
-			continue;
-		btrfs_tree_unlock_rw(path->nodes[i], path->locks[i]);
-		path->locks[i] = 0;
-	}
-}
-
-/*
  * helper function for btrfs_search_slot.  The goal is to find a block
  * in cache without setting the path to blocking.  If we find the block
  * we return zero and the path is unchanged.
@@ -2652,7 +2580,7 @@ static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root,
 
 		} else {
 			b = root->commit_root;
-			extent_buffer_get(b);
+			atomic_inc(&b->refs);
 		}
 		level = btrfs_header_level(b);
 		/*
@@ -2785,12 +2713,10 @@ again:
 	}
 
 	while (b) {
+		int dec = 0;
+
 		level = btrfs_header_level(b);
 
-		/*
-		 * setup the path here so we can release it under lock
-		 * contention with the cow code
-		 */
 		if (cow) {
 			bool last_level = (level == (BTRFS_MAX_LEVEL - 1));
 
@@ -2861,73 +2787,7 @@ cow_done:
 		if (ret < 0)
 			goto done;
 
-		if (level != 0) {
-			int dec = 0;
-			if (ret && slot > 0) {
-				dec = 1;
-				slot -= 1;
-			}
-			p->slots[level] = slot;
-			err = setup_nodes_for_search(trans, root, p, b, level,
-					     ins_len, &write_lock_level);
-			if (err == -EAGAIN)
-				goto again;
-			if (err) {
-				ret = err;
-				goto done;
-			}
-			b = p->nodes[level];
-			slot = p->slots[level];
-
-			/*
-			 * slot 0 is special, if we change the key
-			 * we have to update the parent pointer
-			 * which means we must have a write lock
-			 * on the parent
-			 */
-			if (slot == 0 && ins_len &&
-			    write_lock_level < level + 1) {
-				write_lock_level = level + 1;
-				btrfs_release_path(p);
-				goto again;
-			}
-
-			unlock_up(p, level, lowest_unlock,
-				  min_write_lock_level, &write_lock_level);
-
-			if (level == lowest_level) {
-				if (dec)
-					p->slots[level]++;
-				goto done;
-			}
-
-			err = read_block_for_search(root, p, &b, level,
-						    slot, key);
-			if (err == -EAGAIN)
-				goto again;
-			if (err) {
-				ret = err;
-				goto done;
-			}
-
-			if (!p->skip_locking) {
-				level = btrfs_header_level(b);
-				if (level <= write_lock_level) {
-					if (!btrfs_try_tree_write_lock(b)) {
-						btrfs_set_path_blocking(p);
-						btrfs_tree_lock(b);
-					}
-					p->locks[level] = BTRFS_WRITE_LOCK;
-				} else {
-					if (!btrfs_tree_read_lock_atomic(b)) {
-						btrfs_set_path_blocking(p);
-						btrfs_tree_read_lock(b);
-					}
-					p->locks[level] = BTRFS_READ_LOCK;
-				}
-				p->nodes[level] = b;
-			}
-		} else {
+		if (level == 0) {
 			p->slots[level] = slot;
 			if (ins_len > 0 &&
 			    btrfs_leaf_free_space(b) < ins_len) {
@@ -2952,6 +2812,67 @@ cow_done:
 					  min_write_lock_level, NULL);
 			goto done;
 		}
+		if (ret && slot > 0) {
+			dec = 1;
+			slot--;
+		}
+		p->slots[level] = slot;
+		err = setup_nodes_for_search(trans, root, p, b, level, ins_len,
+					     &write_lock_level);
+		if (err == -EAGAIN)
+			goto again;
+		if (err) {
+			ret = err;
+			goto done;
+		}
+		b = p->nodes[level];
+		slot = p->slots[level];
+
+		/*
+		 * Slot 0 is special, if we change the key we have to update
+		 * the parent pointer which means we must have a write lock on
+		 * the parent
+		 */
+		if (slot == 0 && ins_len && write_lock_level < level + 1) {
+			write_lock_level = level + 1;
+			btrfs_release_path(p);
+			goto again;
+		}
+
+		unlock_up(p, level, lowest_unlock, min_write_lock_level,
+			  &write_lock_level);
+
+		if (level == lowest_level) {
+			if (dec)
+				p->slots[level]++;
+			goto done;
+		}
+
+		err = read_block_for_search(root, p, &b, level, slot, key);
+		if (err == -EAGAIN)
+			goto again;
+		if (err) {
+			ret = err;
+			goto done;
+		}
+
+		if (!p->skip_locking) {
+			level = btrfs_header_level(b);
+			if (level <= write_lock_level) {
+				if (!btrfs_try_tree_write_lock(b)) {
+					btrfs_set_path_blocking(p);
+					btrfs_tree_lock(b);
+				}
+				p->locks[level] = BTRFS_WRITE_LOCK;
+			} else {
+				if (!btrfs_tree_read_lock_atomic(b)) {
+					btrfs_set_path_blocking(p);
+					btrfs_tree_read_lock(b);
+				}
+				p->locks[level] = BTRFS_READ_LOCK;
+			}
+			p->nodes[level] = b;
+		}
 	}
 	ret = 1;
 done:
@@ -3008,6 +2929,8 @@ again:
 	p->locks[level] = BTRFS_READ_LOCK;
 
 	while (b) {
+		int dec = 0;
+
 		level = btrfs_header_level(b);
 		p->nodes[level] = b;
 
@@ -3028,47 +2951,45 @@ again:
 		if (ret < 0)
 			goto done;
 
-		if (level != 0) {
-			int dec = 0;
-			if (ret && slot > 0) {
-				dec = 1;
-				slot -= 1;
-			}
+		if (level == 0) {
 			p->slots[level] = slot;
 			unlock_up(p, level, lowest_unlock, 0, NULL);
+			goto done;
+		}
 
-			if (level == lowest_level) {
-				if (dec)
-					p->slots[level]++;
-				goto done;
-			}
+		if (ret && slot > 0) {
+			dec = 1;
+			slot--;
+		}
+		p->slots[level] = slot;
+		unlock_up(p, level, lowest_unlock, 0, NULL);
 
-			err = read_block_for_search(root, p, &b, level,
-						    slot, key);
-			if (err == -EAGAIN)
-				goto again;
-			if (err) {
-				ret = err;
-				goto done;
-			}
+		if (level == lowest_level) {
+			if (dec)
+				p->slots[level]++;
+			goto done;
+		}
 
-			level = btrfs_header_level(b);
-			if (!btrfs_tree_read_lock_atomic(b)) {
-				btrfs_set_path_blocking(p);
-				btrfs_tree_read_lock(b);
-			}
-			b = tree_mod_log_rewind(fs_info, p, b, time_seq);
-			if (!b) {
-				ret = -ENOMEM;
-				goto done;
-			}
-			p->locks[level] = BTRFS_READ_LOCK;
-			p->nodes[level] = b;
-		} else {
-			p->slots[level] = slot;
-			unlock_up(p, level, lowest_unlock, 0, NULL);
+		err = read_block_for_search(root, p, &b, level, slot, key);
+		if (err == -EAGAIN)
+			goto again;
+		if (err) {
+			ret = err;
+			goto done;
+		}
+
+		level = btrfs_header_level(b);
+		if (!btrfs_tree_read_lock_atomic(b)) {
+			btrfs_set_path_blocking(p);
+			btrfs_tree_read_lock(b);
+		}
+		b = tree_mod_log_rewind(fs_info, p, b, time_seq);
+		if (!b) {
+			ret = -ENOMEM;
 			goto done;
 		}
+		p->locks[level] = BTRFS_READ_LOCK;
+		p->nodes[level] = b;
 	}
 	ret = 1;
 done:
@@ -3433,7 +3354,7 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans,
 	free_extent_buffer(old);
 
 	add_root_to_dirty_list(root);
-	extent_buffer_get(c);
+	atomic_inc(&c->refs);
 	path->nodes[level] = c;
 	path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING;
 	path->slots[level] = 0;
@@ -4966,7 +4887,7 @@ static noinline void btrfs_del_leaf(struct btrfs_trans_handle *trans,
 
 	root_sub_used(root, leaf->len);
 
-	extent_buffer_get(leaf);
+	atomic_inc(&leaf->refs);
 	btrfs_free_tree_block(trans, root, leaf, 0, 1);
 	free_extent_buffer_stale(leaf);
 }
@@ -5047,7 +4968,7 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 			 * for possible call to del_ptr below
 			 */
 			slot = path->slots[1];
-			extent_buffer_get(leaf);
+			atomic_inc(&leaf->refs);
 
 			btrfs_set_path_blocking(path);
 			wret = push_leaf_left(trans, root, path, 1, 1,
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index fe2b8765d9e6..8aa7b9dac405 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -28,17 +28,19 @@
 #include <linux/dynamic_debug.h>
 #include <linux/refcount.h>
 #include <linux/crc32c.h>
+#include "extent-io-tree.h"
 #include "extent_io.h"
 #include "extent_map.h"
 #include "async-thread.h"
 #include "block-rsv.h"
+#include "locking.h"
 
 struct btrfs_trans_handle;
 struct btrfs_transaction;
 struct btrfs_pending_snapshot;
 struct btrfs_delayed_ref_root;
 struct btrfs_space_info;
-struct btrfs_block_group_cache;
+struct btrfs_block_group;
 extern struct kmem_cache *btrfs_trans_handle_cachep;
 extern struct kmem_cache *btrfs_bit_radix_cachep;
 extern struct kmem_cache *btrfs_path_cachep;
@@ -56,9 +58,9 @@ struct btrfs_ref;
  * filesystem data as well that can be used to read data in order to repair
  * read errors on other disks.
  *
- * Current value is derived from RAID1 with 2 copies.
+ * Current value is derived from RAID1C4 with 4 copies.
  */
-#define BTRFS_MAX_MIRRORS (2 + 1)
+#define BTRFS_MAX_MIRRORS (4 + 1)
 
 #define BTRFS_MAX_LEVEL 8
 
@@ -100,6 +102,14 @@ struct btrfs_ref;
 
 #define BTRFS_MAX_EXTENT_SIZE SZ_128M
 
+/*
+ * Deltas are an effective way to populate global statistics.  Give macro names
+ * to make it clear what we're doing.  An example is discard_extents in
+ * btrfs_free_space_ctl.
+ */
+#define BTRFS_STAT_NR_ENTRIES	2
+#define BTRFS_STAT_CURR		0
+#define BTRFS_STAT_PREV		1
 
 /*
  * Count how many BTRFS_MAX_EXTENT_SIZE cover the @size
@@ -291,7 +301,8 @@ struct btrfs_super_block {
 	 BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF |		\
 	 BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA |	\
 	 BTRFS_FEATURE_INCOMPAT_NO_HOLES	|	\
-	 BTRFS_FEATURE_INCOMPAT_METADATA_UUID)
+	 BTRFS_FEATURE_INCOMPAT_METADATA_UUID	|	\
+	 BTRFS_FEATURE_INCOMPAT_RAID1C34)
 
 #define BTRFS_FEATURE_INCOMPAT_SAFE_SET			\
 	(BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF)
@@ -413,7 +424,7 @@ struct btrfs_free_cluster {
 	/* We did a full search and couldn't create a cluster */
 	bool fragmented;
 
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	/*
 	 * when a cluster is allocated from a block group, we put the
 	 * cluster onto a list in the block group so that it can
@@ -438,6 +449,36 @@ struct btrfs_full_stripe_locks_tree {
 	struct mutex lock;
 };
 
+/* Discard control. */
+/*
+ * Async discard uses multiple lists to differentiate the discard filter
+ * parameters.  Index 0 is for completely free block groups where we need to
+ * ensure the entire block group is trimmed without being lossy.  Indices
+ * afterwards represent monotonically decreasing discard filter sizes to
+ * prioritize what should be discarded next.
+ */
+#define BTRFS_NR_DISCARD_LISTS		3
+#define BTRFS_DISCARD_INDEX_UNUSED	0
+#define BTRFS_DISCARD_INDEX_START	1
+
+struct btrfs_discard_ctl {
+	struct workqueue_struct *discard_workers;
+	struct delayed_work work;
+	spinlock_t lock;
+	struct btrfs_block_group *block_group;
+	struct list_head discard_list[BTRFS_NR_DISCARD_LISTS];
+	u64 prev_discard;
+	atomic_t discardable_extents;
+	atomic64_t discardable_bytes;
+	u64 max_discard_size;
+	unsigned long delay;
+	u32 iops_limit;
+	u32 kbps_limit;
+	u64 discard_extent_bytes;
+	u64 discard_bitmap_bytes;
+	atomic64_t discard_bytes_saved;
+};
+
 /* delayed seq elem */
 struct seq_list {
 	struct list_head list;
@@ -476,8 +517,8 @@ struct btrfs_swapfile_pin {
 	void *ptr;
 	struct inode *inode;
 	/*
-	 * If true, ptr points to a struct btrfs_block_group_cache. Otherwise,
-	 * ptr points to a struct btrfs_device.
+	 * If true, ptr points to a struct btrfs_block_group. Otherwise, ptr
+	 * points to a struct btrfs_device.
 	 */
 	bool is_block_group;
 };
@@ -524,6 +565,9 @@ enum {
 	 * so we don't need to offload checksums to workqueues.
 	 */
 	BTRFS_FS_CSUM_IMPL_FAST,
+
+	/* Indicate that the discard workqueue can service discards. */
+	BTRFS_FS_DISCARD_RUNNING,
 };
 
 struct btrfs_fs_info {
@@ -553,8 +597,8 @@ struct btrfs_fs_info {
 	/* keep track of unallocated space */
 	atomic64_t free_chunk_space;
 
-	struct extent_io_tree freed_extents[2];
-	struct extent_io_tree *pinned_extents;
+	/* Track ranges which are used by log trees blocks/logged data extents */
+	struct extent_io_tree excluded_extents;
 
 	/* logical->physical extent mapping */
 	struct extent_map_tree mapping_tree;
@@ -653,7 +697,6 @@ struct btrfs_fs_info {
 	struct rw_semaphore cleanup_work_sem;
 
 	struct rw_semaphore subvol_sem;
-	struct srcu_struct subvol_srcu;
 
 	spinlock_t trans_lock;
 	/*
@@ -671,14 +714,12 @@ struct btrfs_fs_info {
 	atomic_t nr_delayed_iputs;
 	wait_queue_head_t delayed_iputs_wait;
 
-	/* this protects tree_mod_seq_list */
-	spinlock_t tree_mod_seq_lock;
 	atomic64_t tree_mod_seq;
-	struct list_head tree_mod_seq_list;
 
-	/* this protects tree_mod_log */
+	/* this protects tree_mod_log and tree_mod_seq_list */
 	rwlock_t tree_mod_log_lock;
 	struct rb_root tree_mod_log;
+	struct list_head tree_mod_seq_list;
 
 	atomic_t async_delalloc_pages;
 
@@ -722,7 +763,6 @@ struct btrfs_fs_info {
 	struct btrfs_workqueue *endio_meta_write_workers;
 	struct btrfs_workqueue *endio_write_workers;
 	struct btrfs_workqueue *endio_freespace_worker;
-	struct btrfs_workqueue *submit_workers;
 	struct btrfs_workqueue *caching_workers;
 	struct btrfs_workqueue *readahead_workers;
 
@@ -815,6 +855,8 @@ struct btrfs_fs_info {
 	struct btrfs_workqueue *scrub_wr_completion_workers;
 	struct btrfs_workqueue *scrub_parity_workers;
 
+	struct btrfs_discard_ctl discard_ctl;
+
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
 	u32 check_integrity_print_mask;
 #endif
@@ -901,6 +943,15 @@ struct btrfs_fs_info {
 	spinlock_t ref_verify_lock;
 	struct rb_root block_tree;
 #endif
+
+#ifdef CONFIG_BTRFS_DEBUG
+	struct kobject *debug_kobj;
+	struct kobject *discard_debug_kobj;
+	struct list_head allocated_roots;
+
+	spinlock_t eb_leak_lock;
+	struct list_head allocated_ebs;
+#endif
 };
 
 static inline struct btrfs_fs_info *btrfs_sb(struct super_block *sb)
@@ -908,11 +959,6 @@ static inline struct btrfs_fs_info *btrfs_sb(struct super_block *sb)
 	return sb->s_fs_info;
 }
 
-struct btrfs_subvolume_writers {
-	struct percpu_counter	counter;
-	wait_queue_head_t	wait;
-};
-
 /*
  * The state of btrfs root
  */
@@ -1084,8 +1130,9 @@ struct btrfs_root {
 	 * root_item_lock.
 	 */
 	int dedupe_in_progress;
-	struct btrfs_subvolume_writers *subv_writers;
-	atomic_t will_be_snapshotted;
+	/* For exclusion of snapshot creation and nocow writes */
+	struct btrfs_drew_lock snapshot_lock;
+
 	atomic_t snapshot_force_cow;
 
 	/* For qgroup metadata reserved space */
@@ -1102,6 +1149,10 @@ struct btrfs_root {
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 	u64 alloc_bytenr;
 #endif
+
+#ifdef CONFIG_BTRFS_DEBUG
+	struct list_head leak_list;
+#endif
 };
 
 struct btrfs_clone_extent_info {
@@ -1169,7 +1220,7 @@ static inline u32 BTRFS_MAX_XATTR_SIZE(const struct btrfs_fs_info *info)
 #define BTRFS_MOUNT_FLUSHONCOMMIT       (1 << 7)
 #define BTRFS_MOUNT_SSD_SPREAD		(1 << 8)
 #define BTRFS_MOUNT_NOSSD		(1 << 9)
-#define BTRFS_MOUNT_DISCARD		(1 << 10)
+#define BTRFS_MOUNT_DISCARD_SYNC	(1 << 10)
 #define BTRFS_MOUNT_FORCE_COMPRESS      (1 << 11)
 #define BTRFS_MOUNT_SPACE_CACHE		(1 << 12)
 #define BTRFS_MOUNT_CLEAR_CACHE		(1 << 13)
@@ -1188,6 +1239,7 @@ static inline u32 BTRFS_MAX_XATTR_SIZE(const struct btrfs_fs_info *info)
 #define BTRFS_MOUNT_FREE_SPACE_TREE	(1 << 26)
 #define BTRFS_MOUNT_NOLOGREPLAY		(1 << 27)
 #define BTRFS_MOUNT_REF_VERIFY		(1 << 28)
+#define BTRFS_MOUNT_DISCARD_ASYNC	(1 << 29)
 
 #define BTRFS_DEFAULT_COMMIT_INTERVAL	(30)
 #define BTRFS_DEFAULT_MAX_INLINE	(2048)
@@ -1519,18 +1571,18 @@ static inline u64 btrfs_stripe_devid_nr(struct extent_buffer *eb,
 }
 
 /* struct btrfs_block_group_item */
-BTRFS_SETGET_STACK_FUNCS(block_group_used, struct btrfs_block_group_item,
+BTRFS_SETGET_STACK_FUNCS(stack_block_group_used, struct btrfs_block_group_item,
 			 used, 64);
-BTRFS_SETGET_FUNCS(disk_block_group_used, struct btrfs_block_group_item,
+BTRFS_SETGET_FUNCS(block_group_used, struct btrfs_block_group_item,
 			 used, 64);
-BTRFS_SETGET_STACK_FUNCS(block_group_chunk_objectid,
+BTRFS_SETGET_STACK_FUNCS(stack_block_group_chunk_objectid,
 			struct btrfs_block_group_item, chunk_objectid, 64);
 
-BTRFS_SETGET_FUNCS(disk_block_group_chunk_objectid,
+BTRFS_SETGET_FUNCS(block_group_chunk_objectid,
 		   struct btrfs_block_group_item, chunk_objectid, 64);
-BTRFS_SETGET_FUNCS(disk_block_group_flags,
+BTRFS_SETGET_FUNCS(block_group_flags,
 		   struct btrfs_block_group_item, flags, 64);
-BTRFS_SETGET_STACK_FUNCS(block_group_flags,
+BTRFS_SETGET_STACK_FUNCS(stack_block_group_flags,
 			struct btrfs_block_group_item, flags, 64);
 
 /* struct btrfs_free_space_info */
@@ -1923,16 +1975,6 @@ static inline void btrfs_set_header_backref_rev(struct extent_buffer *eb,
 	btrfs_set_header_flags(eb, flags);
 }
 
-static inline unsigned long btrfs_header_fsid(void)
-{
-	return offsetof(struct btrfs_header, fsid);
-}
-
-static inline unsigned long btrfs_header_chunk_tree_uuid(const struct extent_buffer *eb)
-{
-	return offsetof(struct btrfs_header, chunk_tree_uuid);
-}
-
 static inline int btrfs_is_leaf(const struct extent_buffer *eb)
 {
 	return btrfs_header_level(eb) == 0;
@@ -2163,6 +2205,9 @@ BTRFS_SETGET_STACK_FUNCS(super_uuid_tree_generation, struct btrfs_super_block,
 
 int btrfs_super_csum_size(const struct btrfs_super_block *s);
 const char *btrfs_super_csum_name(u16 csum_type);
+const char *btrfs_super_csum_driver(u16 csum_type);
+size_t __const btrfs_get_num_csums(void);
+
 
 /*
  * The leaf data grows from end-to-front in the node.
@@ -2397,7 +2442,7 @@ static inline u64 btrfs_calc_metadata_size(struct btrfs_fs_info *fs_info,
 
 int btrfs_add_excluded_extent(struct btrfs_fs_info *fs_info,
 			      u64 start, u64 num_bytes);
-void btrfs_free_excluded_extents(struct btrfs_block_group_cache *cache);
+void btrfs_free_excluded_extents(struct btrfs_block_group *cache);
 int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 			   unsigned long count);
 void btrfs_cleanup_ref_head_accounting(struct btrfs_fs_info *fs_info,
@@ -2407,9 +2452,9 @@ int btrfs_lookup_data_extent(struct btrfs_fs_info *fs_info, u64 start, u64 len);
 int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans,
 			     struct btrfs_fs_info *fs_info, u64 bytenr,
 			     u64 offset, int metadata, u64 *refs, u64 *flags);
-int btrfs_pin_extent(struct btrfs_fs_info *fs_info,
-		     u64 bytenr, u64 num, int reserved);
-int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info,
+int btrfs_pin_extent(struct btrfs_trans_handle *trans, u64 bytenr, u64 num,
+		     int reserved);
+int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans,
 				    u64 bytenr, u64 num_bytes);
 int btrfs_exclude_logged_extents(struct extent_buffer *eb);
 int btrfs_cross_ref_exist(struct btrfs_root *root,
@@ -2439,22 +2484,22 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		  struct extent_buffer *buf, int full_backref);
 int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
-				u64 bytenr, u64 num_bytes, u64 flags,
+				struct extent_buffer *eb, u64 flags,
 				int level, int is_data);
 int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref);
 
 int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
 			       u64 start, u64 len, int delalloc);
-int btrfs_free_and_pin_reserved_extent(struct btrfs_fs_info *fs_info,
-				       u64 start, u64 len);
+int btrfs_pin_reserved_extent(struct btrfs_trans_handle *trans, u64 start,
+			      u64 len);
 void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info);
 int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans);
 int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 			 struct btrfs_ref *generic_ref);
 
 int btrfs_extent_readonly(struct btrfs_fs_info *fs_info, u64 bytenr);
-void btrfs_get_block_group_trimming(struct btrfs_block_group_cache *cache);
-void btrfs_put_block_group_trimming(struct btrfs_block_group_cache *cache);
+void btrfs_get_block_group_trimming(struct btrfs_block_group *cache);
+void btrfs_put_block_group_trimming(struct btrfs_block_group *cache);
 void btrfs_clear_space_info_full(struct btrfs_fs_info *info);
 
 enum btrfs_reserve_flush_enum {
@@ -2507,7 +2552,7 @@ void btrfs_wait_for_snapshot_creation(struct btrfs_root *root);
 /* ctree.c */
 int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key,
 		     int level, int *slot);
-int btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2);
+int __pure btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2);
 int btrfs_previous_item(struct btrfs_root *root,
 			struct btrfs_path *path, u64 min_objectid,
 			int type);
@@ -2567,8 +2612,6 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
 void btrfs_release_path(struct btrfs_path *p);
 struct btrfs_path *btrfs_alloc_path(void);
 void btrfs_free_path(struct btrfs_path *p);
-void btrfs_set_path_blocking(struct btrfs_path *p);
-void btrfs_unlock_up_safe(struct btrfs_path *p, int level);
 
 int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		   struct btrfs_path *path, int slot, int nr);
@@ -2616,9 +2659,8 @@ static inline int btrfs_next_item(struct btrfs_root *root, struct btrfs_path *p)
 	return btrfs_next_old_item(root, p, 0);
 }
 int btrfs_leaf_free_space(struct extent_buffer *leaf);
-int __must_check btrfs_drop_snapshot(struct btrfs_root *root,
-				     struct btrfs_block_rsv *block_rsv,
-				     int update_ref, int for_reloc);
+int __must_check btrfs_drop_snapshot(struct btrfs_root *root, int update_ref,
+				     int for_reloc);
 int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
 			struct btrfs_root *root,
 			struct extent_buffer *node,
@@ -2646,23 +2688,6 @@ static inline int btrfs_need_cleaner_sleep(struct btrfs_fs_info *fs_info)
 	return fs_info->sb->s_flags & SB_RDONLY || btrfs_fs_closing(fs_info);
 }
 
-static inline void free_fs_info(struct btrfs_fs_info *fs_info)
-{
-	kfree(fs_info->balance_ctl);
-	kfree(fs_info->delayed_root);
-	kfree(fs_info->extent_root);
-	kfree(fs_info->tree_root);
-	kfree(fs_info->chunk_root);
-	kfree(fs_info->dev_root);
-	kfree(fs_info->csum_root);
-	kfree(fs_info->quota_root);
-	kfree(fs_info->uuid_root);
-	kfree(fs_info->free_space_root);
-	kfree(fs_info->super_copy);
-	kfree(fs_info->super_for_commit);
-	kvfree(fs_info);
-}
-
 /* tree mod log functions from ctree.c */
 u64 btrfs_get_tree_mod_seq(struct btrfs_fs_info *fs_info,
 			   struct seq_list *elem);
@@ -2701,9 +2726,7 @@ int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 			u64 subid);
 int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 			u64 subid);
-int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info,
-			    int (*check_func)(struct btrfs_fs_info *, u8 *, u8,
-					      u64));
+int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info);
 
 /* dir-item.c */
 int btrfs_check_dir_item_collision(struct btrfs_root *root, u64 dir,
@@ -2785,11 +2808,9 @@ struct btrfs_inode_extref *btrfs_find_name_in_ext_backref(
 /* file-item.c */
 struct btrfs_dio_private;
 int btrfs_del_csums(struct btrfs_trans_handle *trans,
-		    struct btrfs_fs_info *fs_info, u64 bytenr, u64 len);
+		    struct btrfs_root *root, u64 bytenr, u64 len);
 blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
-				   u8 *dst);
-blk_status_t btrfs_lookup_bio_sums_dio(struct inode *inode, struct bio *bio,
-			      u64 logical_offset);
+				   u64 offset, u8 *dst);
 int btrfs_insert_file_extent(struct btrfs_trans_handle *trans,
 			     struct btrfs_root *root,
 			     u64 objectid, u64 pos,
@@ -2812,6 +2833,12 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 				     struct btrfs_file_extent_item *fi,
 				     const bool new_inline,
 				     struct extent_map *em);
+int btrfs_inode_clear_file_extent_range(struct btrfs_inode *inode, u64 start,
+					u64 len);
+int btrfs_inode_set_file_extent_range(struct btrfs_inode *inode, u64 start,
+				      u64 len);
+void btrfs_inode_safe_disk_i_size_write(struct inode *inode, u64 new_i_size);
+u64 btrfs_file_extent_end(const struct btrfs_path *path);
 
 /* inode.c */
 struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
@@ -2870,13 +2897,12 @@ int btrfs_drop_inode(struct inode *inode);
 int __init btrfs_init_cachep(void);
 void __cold btrfs_destroy_cachep(void);
 struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
-			      struct btrfs_root *root, int *new,
-			      struct btrfs_path *path);
+			      struct btrfs_root *root, struct btrfs_path *path);
 struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
-			 struct btrfs_root *root, int *was_new);
+			 struct btrfs_root *root);
 struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 				    struct page *page, size_t pg_offset,
-				    u64 start, u64 end, int create);
+				    u64 start, u64 end);
 int btrfs_update_inode(struct btrfs_trans_handle *trans,
 			      struct btrfs_root *root,
 			      struct inode *inode);
@@ -2909,7 +2935,7 @@ long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 int btrfs_ioctl_get_supported_features(void __user *arg);
 void btrfs_sync_inode_flags_to_i_flags(struct inode *inode);
-int btrfs_is_empty_uuid(u8 *uuid);
+int __pure btrfs_is_empty_uuid(u8 *uuid);
 int btrfs_defrag_file(struct inode *inode, struct file *file,
 		      struct btrfs_ioctl_defrag_range_args *range,
 		      u64 newer_than, unsigned long max_pages);
@@ -2950,9 +2976,6 @@ int btrfs_dirty_pages(struct inode *inode, struct page **pages,
 		      size_t num_pages, loff_t pos, size_t write_bytes,
 		      struct extent_state **cached);
 int btrfs_fdatawrite_range(struct inode *inode, loff_t start, loff_t end);
-loff_t btrfs_remap_file_range(struct file *file_in, loff_t pos_in,
-			      struct file *file_out, loff_t pos_out,
-			      loff_t len, unsigned int remap_flags);
 
 /* tree-defrag.c */
 int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
@@ -2962,6 +2985,8 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
 int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 			unsigned long new_flags);
 int btrfs_sync_fs(struct super_block *sb, int wait);
+char *btrfs_get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
+					  u64 subvol_objectid);
 
 static inline __printf(2, 3) __cold
 void btrfs_no_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
@@ -3109,17 +3134,21 @@ do {								\
 	rcu_read_unlock();					\
 } while (0)
 
-__cold
-static inline void assfail(const char *expr, const char *file, int line)
+#ifdef CONFIG_BTRFS_ASSERT
+__cold __noreturn
+static inline void assertfail(const char *expr, const char *file, int line)
 {
-	if (IS_ENABLED(CONFIG_BTRFS_ASSERT)) {
-		pr_err("assertion failed: %s, in %s:%d\n", expr, file, line);
-		BUG();
-	}
+	pr_err("assertion failed: %s, in %s:%d\n", expr, file, line);
+	BUG();
 }
 
-#define ASSERT(expr)	\
-	(likely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__))
+#define ASSERT(expr)						\
+	(likely(expr) ? (void)0 : assertfail(#expr, __FILE__, __LINE__))
+
+#else
+static inline void assertfail(const char *expr, const char* file, int line) { }
+#define ASSERT(expr)	(void)(expr)
+#endif
 
 /*
  * Use that for functions that are conditionally exported for sanity tests but
@@ -3143,7 +3172,7 @@ __cold
 void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function,
 		     unsigned int line, int errno, const char *fmt, ...);
 
-const char *btrfs_decode_error(int errno);
+const char * __attribute_const__ btrfs_decode_error(int errno);
 
 __cold
 void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
@@ -3351,6 +3380,7 @@ void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending,
 			      u64 *bytes_to_reserve);
 int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 			      struct btrfs_pending_snapshot *pending);
+int btrfs_should_cancel_balance(struct btrfs_fs_info *fs_info);
 
 /* scrub.c */
 int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
diff --git a/fs/btrfs/delalloc-space.c b/fs/btrfs/delalloc-space.c
index db9f2c58eb4a..1245739a3a6e 100644
--- a/fs/btrfs/delalloc-space.c
+++ b/fs/btrfs/delalloc-space.c
@@ -9,6 +9,108 @@
 #include "qgroup.h"
 #include "block-group.h"
 
+/*
+ * HOW DOES THIS WORK
+ *
+ * There are two stages to data reservations, one for data and one for metadata
+ * to handle the new extents and checksums generated by writing data.
+ *
+ *
+ * DATA RESERVATION
+ *   The general flow of the data reservation is as follows
+ *
+ *   -> Reserve
+ *     We call into btrfs_reserve_data_bytes() for the user request bytes that
+ *     they wish to write.  We make this reservation and add it to
+ *     space_info->bytes_may_use.  We set EXTENT_DELALLOC on the inode io_tree
+ *     for the range and carry on if this is buffered, or follow up trying to
+ *     make a real allocation if we are pre-allocating or doing O_DIRECT.
+ *
+ *   -> Use
+ *     At writepages()/prealloc/O_DIRECT time we will call into
+ *     btrfs_reserve_extent() for some part or all of this range of bytes.  We
+ *     will make the allocation and subtract space_info->bytes_may_use by the
+ *     original requested length and increase the space_info->bytes_reserved by
+ *     the allocated length.  This distinction is important because compression
+ *     may allocate a smaller on disk extent than we previously reserved.
+ *
+ *   -> Allocation
+ *     finish_ordered_io() will insert the new file extent item for this range,
+ *     and then add a delayed ref update for the extent tree.  Once that delayed
+ *     ref is written the extent size is subtracted from
+ *     space_info->bytes_reserved and added to space_info->bytes_used.
+ *
+ *   Error handling
+ *
+ *   -> By the reservation maker
+ *     This is the simplest case, we haven't completed our operation and we know
+ *     how much we reserved, we can simply call
+ *     btrfs_free_reserved_data_space*() and it will be removed from
+ *     space_info->bytes_may_use.
+ *
+ *   -> After the reservation has been made, but before cow_file_range()
+ *     This is specifically for the delalloc case.  You must clear
+ *     EXTENT_DELALLOC with the EXTENT_CLEAR_DATA_RESV bit, and the range will
+ *     be subtracted from space_info->bytes_may_use.
+ *
+ * METADATA RESERVATION
+ *   The general metadata reservation lifetimes are discussed elsewhere, this
+ *   will just focus on how it is used for delalloc space.
+ *
+ *   We keep track of two things on a per inode bases
+ *
+ *   ->outstanding_extents
+ *     This is the number of file extent items we'll need to handle all of the
+ *     outstanding DELALLOC space we have in this inode.  We limit the maximum
+ *     size of an extent, so a large contiguous dirty area may require more than
+ *     one outstanding_extent, which is why count_max_extents() is used to
+ *     determine how many outstanding_extents get added.
+ *
+ *   ->csum_bytes
+ *     This is essentially how many dirty bytes we have for this inode, so we
+ *     can calculate the number of checksum items we would have to add in order
+ *     to checksum our outstanding data.
+ *
+ *   We keep a per-inode block_rsv in order to make it easier to keep track of
+ *   our reservation.  We use btrfs_calculate_inode_block_rsv_size() to
+ *   calculate the current theoretical maximum reservation we would need for the
+ *   metadata for this inode.  We call this and then adjust our reservation as
+ *   necessary, either by attempting to reserve more space, or freeing up excess
+ *   space.
+ *
+ * OUTSTANDING_EXTENTS HANDLING
+ *
+ *  ->outstanding_extents is used for keeping track of how many extents we will
+ *  need to use for this inode, and it will fluctuate depending on where you are
+ *  in the life cycle of the dirty data.  Consider the following normal case for
+ *  a completely clean inode, with a num_bytes < our maximum allowed extent size
+ *
+ *  -> reserve
+ *    ->outstanding_extents += 1 (current value is 1)
+ *
+ *  -> set_delalloc
+ *    ->outstanding_extents += 1 (currrent value is 2)
+ *
+ *  -> btrfs_delalloc_release_extents()
+ *    ->outstanding_extents -= 1 (current value is 1)
+ *
+ *    We must call this once we are done, as we hold our reservation for the
+ *    duration of our operation, and then assume set_delalloc will update the
+ *    counter appropriately.
+ *
+ *  -> add ordered extent
+ *    ->outstanding_extents += 1 (current value is 2)
+ *
+ *  -> btrfs_clear_delalloc_extent
+ *    ->outstanding_extents -= 1 (current value is 1)
+ *
+ *  -> finish_ordered_io/btrfs_remove_ordered_extent
+ *    ->outstanding_extents -= 1 (current value is 0)
+ *
+ *  Each stage is responsible for their own accounting of the extent, thus
+ *  making error handling and cleanup easier.
+ */
+
 int btrfs_alloc_data_chunk_ondemand(struct btrfs_inode *inode, u64 bytes)
 {
 	struct btrfs_root *root = inode->root;
@@ -228,8 +330,8 @@ static void btrfs_inode_rsv_release(struct btrfs_inode *inode, bool qgroup_free)
 	 * are releasing 0 bytes, and then we'll just get the reservation over
 	 * the size free'd.
 	 */
-	released = __btrfs_block_rsv_release(fs_info, block_rsv, 0,
-					     &qgroup_to_release);
+	released = btrfs_block_rsv_release(fs_info, block_rsv, 0,
+					   &qgroup_to_release);
 	if (released > 0)
 		trace_btrfs_space_reservation(fs_info, "delalloc",
 					      btrfs_ino(inode), released, 0);
@@ -307,7 +409,6 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes)
 	unsigned nr_extents;
 	enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_FLUSH_ALL;
 	int ret = 0;
-	bool delalloc_lock = true;
 
 	/*
 	 * If we are a free space inode we need to not flush since we will be in
@@ -320,7 +421,6 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes)
 	 */
 	if (btrfs_is_free_space_inode(inode)) {
 		flush = BTRFS_RESERVE_NO_FLUSH;
-		delalloc_lock = false;
 	} else {
 		if (current->journal_info)
 			flush = BTRFS_RESERVE_FLUSH_LIMIT;
@@ -329,9 +429,6 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes)
 			schedule_timeout(1);
 	}
 
-	if (delalloc_lock)
-		mutex_lock(&inode->delalloc_mutex);
-
 	num_bytes = ALIGN(num_bytes, fs_info->sectorsize);
 
 	/*
@@ -348,10 +445,12 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes)
 				&qgroup_reserve);
 	ret = btrfs_qgroup_reserve_meta_prealloc(root, qgroup_reserve, true);
 	if (ret)
-		goto out_fail;
+		return ret;
 	ret = btrfs_reserve_metadata_bytes(root, block_rsv, meta_reserve, flush);
-	if (ret)
-		goto out_qgroup;
+	if (ret) {
+		btrfs_qgroup_free_meta_prealloc(root, qgroup_reserve);
+		return ret;
+	}
 
 	/*
 	 * Now we need to update our outstanding extents and csum bytes _first_
@@ -375,15 +474,7 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes)
 	block_rsv->qgroup_rsv_reserved += qgroup_reserve;
 	spin_unlock(&block_rsv->lock);
 
-	if (delalloc_lock)
-		mutex_unlock(&inode->delalloc_mutex);
 	return 0;
-out_qgroup:
-	btrfs_qgroup_free_meta_prealloc(root, qgroup_reserve);
-out_fail:
-	if (delalloc_lock)
-		mutex_unlock(&inode->delalloc_mutex);
-	return ret;
 }
 
 /**
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 1f7f39b10bd0..bf1595a42a98 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -6,12 +6,14 @@
 
 #include <linux/slab.h>
 #include <linux/iversion.h>
+#include <linux/sched/mm.h>
 #include "misc.h"
 #include "delayed-inode.h"
 #include "disk-io.h"
 #include "transaction.h"
 #include "ctree.h"
 #include "qgroup.h"
+#include "locking.h"
 
 #define BTRFS_DELAYED_WRITEBACK		512
 #define BTRFS_DELAYED_BACKGROUND	128
@@ -594,8 +596,7 @@ static void btrfs_delayed_item_release_metadata(struct btrfs_root *root,
 	trace_btrfs_space_reservation(fs_info, "delayed_item",
 				      item->key.objectid, item->bytes_reserved,
 				      0);
-	btrfs_block_rsv_release(fs_info, rsv,
-				item->bytes_reserved);
+	btrfs_block_rsv_release(fs_info, rsv, item->bytes_reserved, NULL);
 }
 
 static int btrfs_delayed_inode_reserve_metadata(
@@ -676,8 +677,7 @@ static void btrfs_delayed_inode_release_metadata(struct btrfs_fs_info *fs_info,
 	rsv = &fs_info->delayed_block_rsv;
 	trace_btrfs_space_reservation(fs_info, "delayed_inode",
 				      node->inode_id, node->bytes_reserved, 0);
-	btrfs_block_rsv_release(fs_info, rsv,
-				node->bytes_reserved);
+	btrfs_block_rsv_release(fs_info, rsv, node->bytes_reserved, NULL);
 	if (qgroup_free)
 		btrfs_qgroup_free_meta_prealloc(node->root,
 				node->bytes_reserved);
@@ -804,11 +804,14 @@ static int btrfs_insert_delayed_item(struct btrfs_trans_handle *trans,
 				     struct btrfs_delayed_item *delayed_item)
 {
 	struct extent_buffer *leaf;
+	unsigned int nofs_flag;
 	char *ptr;
 	int ret;
 
+	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_insert_empty_item(trans, root, path, &delayed_item->key,
 				      delayed_item->data_len);
+	memalloc_nofs_restore(nofs_flag);
 	if (ret < 0 && ret != -EEXIST)
 		return ret;
 
@@ -936,6 +939,7 @@ static int btrfs_delete_delayed_items(struct btrfs_trans_handle *trans,
 				      struct btrfs_delayed_node *node)
 {
 	struct btrfs_delayed_item *curr, *prev;
+	unsigned int nofs_flag;
 	int ret = 0;
 
 do_again:
@@ -944,7 +948,9 @@ do_again:
 	if (!curr)
 		goto delete_fail;
 
+	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_search_slot(trans, root, &curr->key, path, -1, 1);
+	memalloc_nofs_restore(nofs_flag);
 	if (ret < 0)
 		goto delete_fail;
 	else if (ret > 0) {
@@ -1011,6 +1017,7 @@ static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
 	struct btrfs_key key;
 	struct btrfs_inode_item *inode_item;
 	struct extent_buffer *leaf;
+	unsigned int nofs_flag;
 	int mod;
 	int ret;
 
@@ -1023,7 +1030,9 @@ static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
 	else
 		mod = 1;
 
+	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_lookup_inode(trans, root, path, &key, mod);
+	memalloc_nofs_restore(nofs_flag);
 	if (ret > 0) {
 		btrfs_release_path(path);
 		return -ENOENT;
@@ -1074,7 +1083,10 @@ search:
 
 	key.type = BTRFS_INODE_EXTREF_KEY;
 	key.offset = -1;
+
+	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
+	memalloc_nofs_restore(nofs_flag);
 	if (ret < 0)
 		goto err_out;
 	ASSERT(ret);
@@ -1138,7 +1150,7 @@ static int __btrfs_run_delayed_items(struct btrfs_trans_handle *trans, int nr)
 	int ret = 0;
 	bool count = (nr > 0);
 
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return -EIO;
 
 	path = btrfs_alloc_path();
@@ -1367,8 +1379,8 @@ static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
 		return -ENOMEM;
 
 	async_work->delayed_root = delayed_root;
-	btrfs_init_work(&async_work->work, btrfs_delayed_meta_helper,
-			btrfs_async_run_delayed_root, NULL, NULL);
+	btrfs_init_work(&async_work->work, btrfs_async_run_delayed_root, NULL,
+			NULL);
 	async_work->nr = nr;
 
 	btrfs_queue_work(fs_info->delayed_workers, &async_work->work);
@@ -1759,6 +1771,7 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans,
 
 int btrfs_fill_inode(struct inode *inode, u32 *rdev)
 {
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
 	struct btrfs_delayed_node *delayed_node;
 	struct btrfs_inode_item *inode_item;
 
@@ -1778,6 +1791,8 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
 	i_uid_write(inode, btrfs_stack_inode_uid(inode_item));
 	i_gid_write(inode, btrfs_stack_inode_gid(inode_item));
 	btrfs_i_size_write(BTRFS_I(inode), btrfs_stack_inode_size(inode_item));
+	btrfs_inode_set_file_extent_range(BTRFS_I(inode), 0,
+			round_up(i_size_read(inode), fs_info->sectorsize));
 	inode->i_mode = btrfs_stack_inode_mode(inode_item);
 	set_nlink(inode, btrfs_stack_inode_nlink(inode_item));
 	inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item));
@@ -1949,12 +1964,19 @@ void btrfs_kill_all_delayed_nodes(struct btrfs_root *root)
 		}
 
 		inode_id = delayed_nodes[n - 1]->inode_id + 1;
-
-		for (i = 0; i < n; i++)
-			refcount_inc(&delayed_nodes[i]->refs);
+		for (i = 0; i < n; i++) {
+			/*
+			 * Don't increase refs in case the node is dead and
+			 * about to be removed from the tree in the loop below
+			 */
+			if (!refcount_inc_not_zero(&delayed_nodes[i]->refs))
+				delayed_nodes[i] = NULL;
+		}
 		spin_unlock(&root->inode_lock);
 
 		for (i = 0; i < n; i++) {
+			if (!delayed_nodes[i])
+				continue;
 			__btrfs_kill_delayed_node(delayed_nodes[i]);
 			btrfs_release_delayed_node(delayed_nodes[i]);
 		}
diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h
index 74ae226ffaf0..ca96ef007d8f 100644
--- a/fs/btrfs/delayed-inode.h
+++ b/fs/btrfs/delayed-inode.h
@@ -70,7 +70,7 @@ struct btrfs_delayed_item {
 	refcount_t refs;
 	int ins_or_del;
 	u32 data_len;
-	char data[0];
+	char data[];
 };
 
 static inline void btrfs_init_delayed_root(
diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c
index df3bd880061d..353cc2994d10 100644
--- a/fs/btrfs/delayed-ref.c
+++ b/fs/btrfs/delayed-ref.c
@@ -82,8 +82,7 @@ void btrfs_delayed_refs_rsv_release(struct btrfs_fs_info *fs_info, int nr)
 	u64 num_bytes = btrfs_calc_insert_metadata_size(fs_info, nr);
 	u64 released = 0;
 
-	released = __btrfs_block_rsv_release(fs_info, block_rsv, num_bytes,
-					     NULL);
+	released = btrfs_block_rsv_release(fs_info, block_rsv, num_bytes, NULL);
 	if (released)
 		trace_btrfs_space_reservation(fs_info, "delayed_refs_rsv",
 					      0, released, 0);
@@ -492,7 +491,7 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
 	if (head->is_data)
 		return;
 
-	spin_lock(&fs_info->tree_mod_seq_lock);
+	read_lock(&fs_info->tree_mod_log_lock);
 	if (!list_empty(&fs_info->tree_mod_seq_list)) {
 		struct seq_list *elem;
 
@@ -500,7 +499,7 @@ void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
 					struct seq_list, list);
 		seq = elem->seq;
 	}
-	spin_unlock(&fs_info->tree_mod_seq_lock);
+	read_unlock(&fs_info->tree_mod_log_lock);
 
 again:
 	for (node = rb_first_cached(&head->ref_tree); node;
@@ -518,7 +517,7 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq)
 	struct seq_list *elem;
 	int ret = 0;
 
-	spin_lock(&fs_info->tree_mod_seq_lock);
+	read_lock(&fs_info->tree_mod_log_lock);
 	if (!list_empty(&fs_info->tree_mod_seq_list)) {
 		elem = list_first_entry(&fs_info->tree_mod_seq_list,
 					struct seq_list, list);
@@ -531,7 +530,7 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq)
 		}
 	}
 
-	spin_unlock(&fs_info->tree_mod_seq_lock);
+	read_unlock(&fs_info->tree_mod_log_lock);
 	return ret;
 }
 
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index 48890826b5e6..db93909b25e0 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -22,6 +22,46 @@
 #include "dev-replace.h"
 #include "sysfs.h"
 
+/*
+ * Device replace overview
+ *
+ * [Objective]
+ * To copy all extents (both new and on-disk) from source device to target
+ * device, while still keeping the filesystem read-write.
+ *
+ * [Method]
+ * There are two main methods involved:
+ *
+ * - Write duplication
+ *
+ *   All new writes will be written to both target and source devices, so even
+ *   if replace gets canceled, sources device still contans up-to-date data.
+ *
+ *   Location:		handle_ops_on_dev_replace() from __btrfs_map_block()
+ *   Start:		btrfs_dev_replace_start()
+ *   End:		btrfs_dev_replace_finishing()
+ *   Content:		Latest data/metadata
+ *
+ * - Copy existing extents
+ *
+ *   This happens by re-using scrub facility, as scrub also iterates through
+ *   existing extents from commit root.
+ *
+ *   Location:		scrub_write_block_to_dev_replace() from
+ *   			scrub_block_complete()
+ *   Content:		Data/meta from commit root.
+ *
+ * Due to the content difference, we need to avoid nocow write when dev-replace
+ * is happening.  This is done by marking the block group read-only and waiting
+ * for NOCOW writes.
+ *
+ * After replace is done, the finishing part is done by swapping the target and
+ * source devices.
+ *
+ *   Location:		btrfs_dev_replace_update_device_in_mapping_tree() from
+ *   			btrfs_dev_replace_finishing()
+ */
+
 static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 				       int scrub_ret);
 static void btrfs_dev_replace_update_device_in_mapping_tree(
@@ -472,7 +512,7 @@ static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 	atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
 	up_write(&dev_replace->rwsem);
 
-	ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device);
+	ret = btrfs_sysfs_add_devices_dir(tgt_device->fs_devices, tgt_device);
 	if (ret)
 		btrfs_err(fs_info, "kobj add dev failed %d", ret);
 
@@ -500,11 +540,8 @@ static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
 			      &dev_replace->scrub_progress, 0, 1);
 
 	ret = btrfs_dev_replace_finishing(fs_info, ret);
-	if (ret == -EINPROGRESS) {
+	if (ret == -EINPROGRESS)
 		ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS;
-	} else if (ret != -ECANCELED) {
-		WARN_ON(ret);
-	}
 
 	return ret;
 
@@ -706,7 +743,8 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
 	/* replace the sysfs entry */
-	btrfs_sysfs_rm_device_link(fs_info->fs_devices, src_device);
+	btrfs_sysfs_remove_devices_dir(fs_info->fs_devices, src_device);
+	btrfs_sysfs_update_devid(tgt_device);
 	btrfs_rm_dev_replace_free_srcdev(src_device);
 
 	/* write back the superblocks */
@@ -986,7 +1024,7 @@ static int btrfs_dev_replace_kthread(void *data)
 	return 0;
 }
 
-int btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace)
+int __pure btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace)
 {
 	if (!dev_replace->is_valid)
 		return 0;
diff --git a/fs/btrfs/dev-replace.h b/fs/btrfs/dev-replace.h
index 78c5d8f1adda..60b70dacc299 100644
--- a/fs/btrfs/dev-replace.h
+++ b/fs/btrfs/dev-replace.h
@@ -17,6 +17,6 @@ void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info,
 int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info);
 void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info);
 int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info);
-int btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace);
+int __pure btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace);
 
 #endif
diff --git a/fs/btrfs/discard.c b/fs/btrfs/discard.c
new file mode 100644
index 000000000000..5615320fa659
--- /dev/null
+++ b/fs/btrfs/discard.c
@@ -0,0 +1,702 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/jiffies.h>
+#include <linux/kernel.h>
+#include <linux/ktime.h>
+#include <linux/list.h>
+#include <linux/math64.h>
+#include <linux/sizes.h>
+#include <linux/workqueue.h>
+#include "ctree.h"
+#include "block-group.h"
+#include "discard.h"
+#include "free-space-cache.h"
+
+/*
+ * This contains the logic to handle async discard.
+ *
+ * Async discard manages trimming of free space outside of transaction commit.
+ * Discarding is done by managing the block_groups on a LRU list based on free
+ * space recency.  Two passes are used to first prioritize discarding extents
+ * and then allow for trimming in the bitmap the best opportunity to coalesce.
+ * The block_groups are maintained on multiple lists to allow for multiple
+ * passes with different discard filter requirements.  A delayed work item is
+ * used to manage discarding with timeout determined by a max of the delay
+ * incurred by the iops rate limit, the byte rate limit, and the max delay of
+ * BTRFS_DISCARD_MAX_DELAY.
+ *
+ * Note, this only keeps track of block_groups that are explicitly for data.
+ * Mixed block_groups are not supported.
+ *
+ * The first list is special to manage discarding of fully free block groups.
+ * This is necessary because we issue a final trim for a full free block group
+ * after forgetting it.  When a block group becomes unused, instead of directly
+ * being added to the unused_bgs list, we add it to this first list.  Then
+ * from there, if it becomes fully discarded, we place it onto the unused_bgs
+ * list.
+ *
+ * The in-memory free space cache serves as the backing state for discard.
+ * Consequently this means there is no persistence.  We opt to load all the
+ * block groups in as not discarded, so the mount case degenerates to the
+ * crashing case.
+ *
+ * As the free space cache uses bitmaps, there exists a tradeoff between
+ * ease/efficiency for find_free_extent() and the accuracy of discard state.
+ * Here we opt to let untrimmed regions merge with everything while only letting
+ * trimmed regions merge with other trimmed regions.  This can cause
+ * overtrimming, but the coalescing benefit seems to be worth it.  Additionally,
+ * bitmap state is tracked as a whole.  If we're able to fully trim a bitmap,
+ * the trimmed flag is set on the bitmap.  Otherwise, if an allocation comes in,
+ * this resets the state and we will retry trimming the whole bitmap.  This is a
+ * tradeoff between discard state accuracy and the cost of accounting.
+ */
+
+/* This is an initial delay to give some chance for block reuse */
+#define BTRFS_DISCARD_DELAY		(120ULL * NSEC_PER_SEC)
+#define BTRFS_DISCARD_UNUSED_DELAY	(10ULL * NSEC_PER_SEC)
+
+/* Target completion latency of discarding all discardable extents */
+#define BTRFS_DISCARD_TARGET_MSEC	(6 * 60 * 60UL * MSEC_PER_SEC)
+#define BTRFS_DISCARD_MIN_DELAY_MSEC	(1UL)
+#define BTRFS_DISCARD_MAX_DELAY_MSEC	(1000UL)
+#define BTRFS_DISCARD_MAX_IOPS		(10U)
+
+/* Montonically decreasing minimum length filters after index 0 */
+static int discard_minlen[BTRFS_NR_DISCARD_LISTS] = {
+	0,
+	BTRFS_ASYNC_DISCARD_MAX_FILTER,
+	BTRFS_ASYNC_DISCARD_MIN_FILTER
+};
+
+static struct list_head *get_discard_list(struct btrfs_discard_ctl *discard_ctl,
+					  struct btrfs_block_group *block_group)
+{
+	return &discard_ctl->discard_list[block_group->discard_index];
+}
+
+static void __add_to_discard_list(struct btrfs_discard_ctl *discard_ctl,
+				  struct btrfs_block_group *block_group)
+{
+	if (!btrfs_run_discard_work(discard_ctl))
+		return;
+
+	if (list_empty(&block_group->discard_list) ||
+	    block_group->discard_index == BTRFS_DISCARD_INDEX_UNUSED) {
+		if (block_group->discard_index == BTRFS_DISCARD_INDEX_UNUSED)
+			block_group->discard_index = BTRFS_DISCARD_INDEX_START;
+		block_group->discard_eligible_time = (ktime_get_ns() +
+						      BTRFS_DISCARD_DELAY);
+		block_group->discard_state = BTRFS_DISCARD_RESET_CURSOR;
+	}
+
+	list_move_tail(&block_group->discard_list,
+		       get_discard_list(discard_ctl, block_group));
+}
+
+static void add_to_discard_list(struct btrfs_discard_ctl *discard_ctl,
+				struct btrfs_block_group *block_group)
+{
+	if (!btrfs_is_block_group_data_only(block_group))
+		return;
+
+	spin_lock(&discard_ctl->lock);
+	__add_to_discard_list(discard_ctl, block_group);
+	spin_unlock(&discard_ctl->lock);
+}
+
+static void add_to_discard_unused_list(struct btrfs_discard_ctl *discard_ctl,
+				       struct btrfs_block_group *block_group)
+{
+	spin_lock(&discard_ctl->lock);
+
+	if (!btrfs_run_discard_work(discard_ctl)) {
+		spin_unlock(&discard_ctl->lock);
+		return;
+	}
+
+	list_del_init(&block_group->discard_list);
+
+	block_group->discard_index = BTRFS_DISCARD_INDEX_UNUSED;
+	block_group->discard_eligible_time = (ktime_get_ns() +
+					      BTRFS_DISCARD_UNUSED_DELAY);
+	block_group->discard_state = BTRFS_DISCARD_RESET_CURSOR;
+	list_add_tail(&block_group->discard_list,
+		      &discard_ctl->discard_list[BTRFS_DISCARD_INDEX_UNUSED]);
+
+	spin_unlock(&discard_ctl->lock);
+}
+
+static bool remove_from_discard_list(struct btrfs_discard_ctl *discard_ctl,
+				     struct btrfs_block_group *block_group)
+{
+	bool running = false;
+
+	spin_lock(&discard_ctl->lock);
+
+	if (block_group == discard_ctl->block_group) {
+		running = true;
+		discard_ctl->block_group = NULL;
+	}
+
+	block_group->discard_eligible_time = 0;
+	list_del_init(&block_group->discard_list);
+
+	spin_unlock(&discard_ctl->lock);
+
+	return running;
+}
+
+/**
+ * find_next_block_group - find block_group that's up next for discarding
+ * @discard_ctl: discard control
+ * @now: current time
+ *
+ * Iterate over the discard lists to find the next block_group up for
+ * discarding checking the discard_eligible_time of block_group.
+ */
+static struct btrfs_block_group *find_next_block_group(
+					struct btrfs_discard_ctl *discard_ctl,
+					u64 now)
+{
+	struct btrfs_block_group *ret_block_group = NULL, *block_group;
+	int i;
+
+	for (i = 0; i < BTRFS_NR_DISCARD_LISTS; i++) {
+		struct list_head *discard_list = &discard_ctl->discard_list[i];
+
+		if (!list_empty(discard_list)) {
+			block_group = list_first_entry(discard_list,
+						       struct btrfs_block_group,
+						       discard_list);
+
+			if (!ret_block_group)
+				ret_block_group = block_group;
+
+			if (ret_block_group->discard_eligible_time < now)
+				break;
+
+			if (ret_block_group->discard_eligible_time >
+			    block_group->discard_eligible_time)
+				ret_block_group = block_group;
+		}
+	}
+
+	return ret_block_group;
+}
+
+/**
+ * peek_discard_list - wrap find_next_block_group()
+ * @discard_ctl: discard control
+ * @discard_state: the discard_state of the block_group after state management
+ * @discard_index: the discard_index of the block_group after state management
+ *
+ * This wraps find_next_block_group() and sets the block_group to be in use.
+ * discard_state's control flow is managed here.  Variables related to
+ * discard_state are reset here as needed (eg discard_cursor).  @discard_state
+ * and @discard_index are remembered as it may change while we're discarding,
+ * but we want the discard to execute in the context determined here.
+ */
+static struct btrfs_block_group *peek_discard_list(
+					struct btrfs_discard_ctl *discard_ctl,
+					enum btrfs_discard_state *discard_state,
+					int *discard_index)
+{
+	struct btrfs_block_group *block_group;
+	const u64 now = ktime_get_ns();
+
+	spin_lock(&discard_ctl->lock);
+again:
+	block_group = find_next_block_group(discard_ctl, now);
+
+	if (block_group && now > block_group->discard_eligible_time) {
+		if (block_group->discard_index == BTRFS_DISCARD_INDEX_UNUSED &&
+		    block_group->used != 0) {
+			if (btrfs_is_block_group_data_only(block_group))
+				__add_to_discard_list(discard_ctl, block_group);
+			else
+				list_del_init(&block_group->discard_list);
+			goto again;
+		}
+		if (block_group->discard_state == BTRFS_DISCARD_RESET_CURSOR) {
+			block_group->discard_cursor = block_group->start;
+			block_group->discard_state = BTRFS_DISCARD_EXTENTS;
+		}
+		discard_ctl->block_group = block_group;
+		*discard_state = block_group->discard_state;
+		*discard_index = block_group->discard_index;
+	} else {
+		block_group = NULL;
+	}
+
+	spin_unlock(&discard_ctl->lock);
+
+	return block_group;
+}
+
+/**
+ * btrfs_discard_check_filter - updates a block groups filters
+ * @block_group: block group of interest
+ * @bytes: recently freed region size after coalescing
+ *
+ * Async discard maintains multiple lists with progressively smaller filters
+ * to prioritize discarding based on size.  Should a free space that matches
+ * a larger filter be returned to the free_space_cache, prioritize that discard
+ * by moving @block_group to the proper filter.
+ */
+void btrfs_discard_check_filter(struct btrfs_block_group *block_group,
+				u64 bytes)
+{
+	struct btrfs_discard_ctl *discard_ctl;
+
+	if (!block_group ||
+	    !btrfs_test_opt(block_group->fs_info, DISCARD_ASYNC))
+		return;
+
+	discard_ctl = &block_group->fs_info->discard_ctl;
+
+	if (block_group->discard_index > BTRFS_DISCARD_INDEX_START &&
+	    bytes >= discard_minlen[block_group->discard_index - 1]) {
+		int i;
+
+		remove_from_discard_list(discard_ctl, block_group);
+
+		for (i = BTRFS_DISCARD_INDEX_START; i < BTRFS_NR_DISCARD_LISTS;
+		     i++) {
+			if (bytes >= discard_minlen[i]) {
+				block_group->discard_index = i;
+				add_to_discard_list(discard_ctl, block_group);
+				break;
+			}
+		}
+	}
+}
+
+/**
+ * btrfs_update_discard_index - moves a block group along the discard lists
+ * @discard_ctl: discard control
+ * @block_group: block_group of interest
+ *
+ * Increment @block_group's discard_index.  If it falls of the list, let it be.
+ * Otherwise add it back to the appropriate list.
+ */
+static void btrfs_update_discard_index(struct btrfs_discard_ctl *discard_ctl,
+				       struct btrfs_block_group *block_group)
+{
+	block_group->discard_index++;
+	if (block_group->discard_index == BTRFS_NR_DISCARD_LISTS) {
+		block_group->discard_index = 1;
+		return;
+	}
+
+	add_to_discard_list(discard_ctl, block_group);
+}
+
+/**
+ * btrfs_discard_cancel_work - remove a block_group from the discard lists
+ * @discard_ctl: discard control
+ * @block_group: block_group of interest
+ *
+ * This removes @block_group from the discard lists.  If necessary, it waits on
+ * the current work and then reschedules the delayed work.
+ */
+void btrfs_discard_cancel_work(struct btrfs_discard_ctl *discard_ctl,
+			       struct btrfs_block_group *block_group)
+{
+	if (remove_from_discard_list(discard_ctl, block_group)) {
+		cancel_delayed_work_sync(&discard_ctl->work);
+		btrfs_discard_schedule_work(discard_ctl, true);
+	}
+}
+
+/**
+ * btrfs_discard_queue_work - handles queuing the block_groups
+ * @discard_ctl: discard control
+ * @block_group: block_group of interest
+ *
+ * This maintains the LRU order of the discard lists.
+ */
+void btrfs_discard_queue_work(struct btrfs_discard_ctl *discard_ctl,
+			      struct btrfs_block_group *block_group)
+{
+	if (!block_group || !btrfs_test_opt(block_group->fs_info, DISCARD_ASYNC))
+		return;
+
+	if (block_group->used == 0)
+		add_to_discard_unused_list(discard_ctl, block_group);
+	else
+		add_to_discard_list(discard_ctl, block_group);
+
+	if (!delayed_work_pending(&discard_ctl->work))
+		btrfs_discard_schedule_work(discard_ctl, false);
+}
+
+/**
+ * btrfs_discard_schedule_work - responsible for scheduling the discard work
+ * @discard_ctl: discard control
+ * @override: override the current timer
+ *
+ * Discards are issued by a delayed workqueue item.  @override is used to
+ * update the current delay as the baseline delay interval is reevaluated on
+ * transaction commit.  This is also maxed with any other rate limit.
+ */
+void btrfs_discard_schedule_work(struct btrfs_discard_ctl *discard_ctl,
+				 bool override)
+{
+	struct btrfs_block_group *block_group;
+	const u64 now = ktime_get_ns();
+
+	spin_lock(&discard_ctl->lock);
+
+	if (!btrfs_run_discard_work(discard_ctl))
+		goto out;
+
+	if (!override && delayed_work_pending(&discard_ctl->work))
+		goto out;
+
+	block_group = find_next_block_group(discard_ctl, now);
+	if (block_group) {
+		unsigned long delay = discard_ctl->delay;
+		u32 kbps_limit = READ_ONCE(discard_ctl->kbps_limit);
+
+		/*
+		 * A single delayed workqueue item is responsible for
+		 * discarding, so we can manage the bytes rate limit by keeping
+		 * track of the previous discard.
+		 */
+		if (kbps_limit && discard_ctl->prev_discard) {
+			u64 bps_limit = ((u64)kbps_limit) * SZ_1K;
+			u64 bps_delay = div64_u64(discard_ctl->prev_discard *
+						  MSEC_PER_SEC, bps_limit);
+
+			delay = max(delay, msecs_to_jiffies(bps_delay));
+		}
+
+		/*
+		 * This timeout is to hopefully prevent immediate discarding
+		 * in a recently allocated block group.
+		 */
+		if (now < block_group->discard_eligible_time) {
+			u64 bg_timeout = block_group->discard_eligible_time - now;
+
+			delay = max(delay, nsecs_to_jiffies(bg_timeout));
+		}
+
+		mod_delayed_work(discard_ctl->discard_workers,
+				 &discard_ctl->work, delay);
+	}
+out:
+	spin_unlock(&discard_ctl->lock);
+}
+
+/**
+ * btrfs_finish_discard_pass - determine next step of a block_group
+ * @discard_ctl: discard control
+ * @block_group: block_group of interest
+ *
+ * This determines the next step for a block group after it's finished going
+ * through a pass on a discard list.  If it is unused and fully trimmed, we can
+ * mark it unused and send it to the unused_bgs path.  Otherwise, pass it onto
+ * the appropriate filter list or let it fall off.
+ */
+static void btrfs_finish_discard_pass(struct btrfs_discard_ctl *discard_ctl,
+				      struct btrfs_block_group *block_group)
+{
+	remove_from_discard_list(discard_ctl, block_group);
+
+	if (block_group->used == 0) {
+		if (btrfs_is_free_space_trimmed(block_group))
+			btrfs_mark_bg_unused(block_group);
+		else
+			add_to_discard_unused_list(discard_ctl, block_group);
+	} else {
+		btrfs_update_discard_index(discard_ctl, block_group);
+	}
+}
+
+/**
+ * btrfs_discard_workfn - discard work function
+ * @work: work
+ *
+ * This finds the next block_group to start discarding and then discards a
+ * single region.  It does this in a two-pass fashion: first extents and second
+ * bitmaps.  Completely discarded block groups are sent to the unused_bgs path.
+ */
+static void btrfs_discard_workfn(struct work_struct *work)
+{
+	struct btrfs_discard_ctl *discard_ctl;
+	struct btrfs_block_group *block_group;
+	enum btrfs_discard_state discard_state;
+	int discard_index = 0;
+	u64 trimmed = 0;
+	u64 minlen = 0;
+
+	discard_ctl = container_of(work, struct btrfs_discard_ctl, work.work);
+
+	block_group = peek_discard_list(discard_ctl, &discard_state,
+					&discard_index);
+	if (!block_group || !btrfs_run_discard_work(discard_ctl))
+		return;
+
+	/* Perform discarding */
+	minlen = discard_minlen[discard_index];
+
+	if (discard_state == BTRFS_DISCARD_BITMAPS) {
+		u64 maxlen = 0;
+
+		/*
+		 * Use the previous levels minimum discard length as the max
+		 * length filter.  In the case something is added to make a
+		 * region go beyond the max filter, the entire bitmap is set
+		 * back to BTRFS_TRIM_STATE_UNTRIMMED.
+		 */
+		if (discard_index != BTRFS_DISCARD_INDEX_UNUSED)
+			maxlen = discard_minlen[discard_index - 1];
+
+		btrfs_trim_block_group_bitmaps(block_group, &trimmed,
+				       block_group->discard_cursor,
+				       btrfs_block_group_end(block_group),
+				       minlen, maxlen, true);
+		discard_ctl->discard_bitmap_bytes += trimmed;
+	} else {
+		btrfs_trim_block_group_extents(block_group, &trimmed,
+				       block_group->discard_cursor,
+				       btrfs_block_group_end(block_group),
+				       minlen, true);
+		discard_ctl->discard_extent_bytes += trimmed;
+	}
+
+	discard_ctl->prev_discard = trimmed;
+
+	/* Determine next steps for a block_group */
+	if (block_group->discard_cursor >= btrfs_block_group_end(block_group)) {
+		if (discard_state == BTRFS_DISCARD_BITMAPS) {
+			btrfs_finish_discard_pass(discard_ctl, block_group);
+		} else {
+			block_group->discard_cursor = block_group->start;
+			spin_lock(&discard_ctl->lock);
+			if (block_group->discard_state !=
+			    BTRFS_DISCARD_RESET_CURSOR)
+				block_group->discard_state =
+							BTRFS_DISCARD_BITMAPS;
+			spin_unlock(&discard_ctl->lock);
+		}
+	}
+
+	spin_lock(&discard_ctl->lock);
+	discard_ctl->block_group = NULL;
+	spin_unlock(&discard_ctl->lock);
+
+	btrfs_discard_schedule_work(discard_ctl, false);
+}
+
+/**
+ * btrfs_run_discard_work - determines if async discard should be running
+ * @discard_ctl: discard control
+ *
+ * Checks if the file system is writeable and BTRFS_FS_DISCARD_RUNNING is set.
+ */
+bool btrfs_run_discard_work(struct btrfs_discard_ctl *discard_ctl)
+{
+	struct btrfs_fs_info *fs_info = container_of(discard_ctl,
+						     struct btrfs_fs_info,
+						     discard_ctl);
+
+	return (!(fs_info->sb->s_flags & SB_RDONLY) &&
+		test_bit(BTRFS_FS_DISCARD_RUNNING, &fs_info->flags));
+}
+
+/**
+ * btrfs_discard_calc_delay - recalculate the base delay
+ * @discard_ctl: discard control
+ *
+ * Recalculate the base delay which is based off the total number of
+ * discardable_extents.  Clamp this between the lower_limit (iops_limit or 1ms)
+ * and the upper_limit (BTRFS_DISCARD_MAX_DELAY_MSEC).
+ */
+void btrfs_discard_calc_delay(struct btrfs_discard_ctl *discard_ctl)
+{
+	s32 discardable_extents;
+	s64 discardable_bytes;
+	u32 iops_limit;
+	unsigned long delay;
+	unsigned long lower_limit = BTRFS_DISCARD_MIN_DELAY_MSEC;
+
+	discardable_extents = atomic_read(&discard_ctl->discardable_extents);
+	if (!discardable_extents)
+		return;
+
+	spin_lock(&discard_ctl->lock);
+
+	/*
+	 * The following is to fix a potential -1 discrepenancy that we're not
+	 * sure how to reproduce. But given that this is the only place that
+	 * utilizes these numbers and this is only called by from
+	 * btrfs_finish_extent_commit() which is synchronized, we can correct
+	 * here.
+	 */
+	if (discardable_extents < 0)
+		atomic_add(-discardable_extents,
+			   &discard_ctl->discardable_extents);
+
+	discardable_bytes = atomic64_read(&discard_ctl->discardable_bytes);
+	if (discardable_bytes < 0)
+		atomic64_add(-discardable_bytes,
+			     &discard_ctl->discardable_bytes);
+
+	if (discardable_extents <= 0) {
+		spin_unlock(&discard_ctl->lock);
+		return;
+	}
+
+	iops_limit = READ_ONCE(discard_ctl->iops_limit);
+	if (iops_limit)
+		lower_limit = max_t(unsigned long, lower_limit,
+				    MSEC_PER_SEC / iops_limit);
+
+	delay = BTRFS_DISCARD_TARGET_MSEC / discardable_extents;
+	delay = clamp(delay, lower_limit, BTRFS_DISCARD_MAX_DELAY_MSEC);
+	discard_ctl->delay = msecs_to_jiffies(delay);
+
+	spin_unlock(&discard_ctl->lock);
+}
+
+/**
+ * btrfs_discard_update_discardable - propagate discard counters
+ * @block_group: block_group of interest
+ * @ctl: free_space_ctl of @block_group
+ *
+ * This propagates deltas of counters up to the discard_ctl.  It maintains a
+ * current counter and a previous counter passing the delta up to the global
+ * stat.  Then the current counter value becomes the previous counter value.
+ */
+void btrfs_discard_update_discardable(struct btrfs_block_group *block_group,
+				      struct btrfs_free_space_ctl *ctl)
+{
+	struct btrfs_discard_ctl *discard_ctl;
+	s32 extents_delta;
+	s64 bytes_delta;
+
+	if (!block_group ||
+	    !btrfs_test_opt(block_group->fs_info, DISCARD_ASYNC) ||
+	    !btrfs_is_block_group_data_only(block_group))
+		return;
+
+	discard_ctl = &block_group->fs_info->discard_ctl;
+
+	extents_delta = ctl->discardable_extents[BTRFS_STAT_CURR] -
+			ctl->discardable_extents[BTRFS_STAT_PREV];
+	if (extents_delta) {
+		atomic_add(extents_delta, &discard_ctl->discardable_extents);
+		ctl->discardable_extents[BTRFS_STAT_PREV] =
+			ctl->discardable_extents[BTRFS_STAT_CURR];
+	}
+
+	bytes_delta = ctl->discardable_bytes[BTRFS_STAT_CURR] -
+		      ctl->discardable_bytes[BTRFS_STAT_PREV];
+	if (bytes_delta) {
+		atomic64_add(bytes_delta, &discard_ctl->discardable_bytes);
+		ctl->discardable_bytes[BTRFS_STAT_PREV] =
+			ctl->discardable_bytes[BTRFS_STAT_CURR];
+	}
+}
+
+/**
+ * btrfs_discard_punt_unused_bgs_list - punt unused_bgs list to discard lists
+ * @fs_info: fs_info of interest
+ *
+ * The unused_bgs list needs to be punted to the discard lists because the
+ * order of operations is changed.  In the normal sychronous discard path, the
+ * block groups are trimmed via a single large trim in transaction commit.  This
+ * is ultimately what we are trying to avoid with asynchronous discard.  Thus,
+ * it must be done before going down the unused_bgs path.
+ */
+void btrfs_discard_punt_unused_bgs_list(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_block_group *block_group, *next;
+
+	spin_lock(&fs_info->unused_bgs_lock);
+	/* We enabled async discard, so punt all to the queue */
+	list_for_each_entry_safe(block_group, next, &fs_info->unused_bgs,
+				 bg_list) {
+		list_del_init(&block_group->bg_list);
+		btrfs_discard_queue_work(&fs_info->discard_ctl, block_group);
+	}
+	spin_unlock(&fs_info->unused_bgs_lock);
+}
+
+/**
+ * btrfs_discard_purge_list - purge discard lists
+ * @discard_ctl: discard control
+ *
+ * If we are disabling async discard, we may have intercepted block groups that
+ * are completely free and ready for the unused_bgs path.  As discarding will
+ * now happen in transaction commit or not at all, we can safely mark the
+ * corresponding block groups as unused and they will be sent on their merry
+ * way to the unused_bgs list.
+ */
+static void btrfs_discard_purge_list(struct btrfs_discard_ctl *discard_ctl)
+{
+	struct btrfs_block_group *block_group, *next;
+	int i;
+
+	spin_lock(&discard_ctl->lock);
+	for (i = 0; i < BTRFS_NR_DISCARD_LISTS; i++) {
+		list_for_each_entry_safe(block_group, next,
+					 &discard_ctl->discard_list[i],
+					 discard_list) {
+			list_del_init(&block_group->discard_list);
+			spin_unlock(&discard_ctl->lock);
+			if (block_group->used == 0)
+				btrfs_mark_bg_unused(block_group);
+			spin_lock(&discard_ctl->lock);
+		}
+	}
+	spin_unlock(&discard_ctl->lock);
+}
+
+void btrfs_discard_resume(struct btrfs_fs_info *fs_info)
+{
+	if (!btrfs_test_opt(fs_info, DISCARD_ASYNC)) {
+		btrfs_discard_cleanup(fs_info);
+		return;
+	}
+
+	btrfs_discard_punt_unused_bgs_list(fs_info);
+
+	set_bit(BTRFS_FS_DISCARD_RUNNING, &fs_info->flags);
+}
+
+void btrfs_discard_stop(struct btrfs_fs_info *fs_info)
+{
+	clear_bit(BTRFS_FS_DISCARD_RUNNING, &fs_info->flags);
+}
+
+void btrfs_discard_init(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_discard_ctl *discard_ctl = &fs_info->discard_ctl;
+	int i;
+
+	spin_lock_init(&discard_ctl->lock);
+	INIT_DELAYED_WORK(&discard_ctl->work, btrfs_discard_workfn);
+
+	for (i = 0; i < BTRFS_NR_DISCARD_LISTS; i++)
+		INIT_LIST_HEAD(&discard_ctl->discard_list[i]);
+
+	discard_ctl->prev_discard = 0;
+	atomic_set(&discard_ctl->discardable_extents, 0);
+	atomic64_set(&discard_ctl->discardable_bytes, 0);
+	discard_ctl->max_discard_size = BTRFS_ASYNC_DISCARD_DEFAULT_MAX_SIZE;
+	discard_ctl->delay = BTRFS_DISCARD_MAX_DELAY_MSEC;
+	discard_ctl->iops_limit = BTRFS_DISCARD_MAX_IOPS;
+	discard_ctl->kbps_limit = 0;
+	discard_ctl->discard_extent_bytes = 0;
+	discard_ctl->discard_bitmap_bytes = 0;
+	atomic64_set(&discard_ctl->discard_bytes_saved, 0);
+}
+
+void btrfs_discard_cleanup(struct btrfs_fs_info *fs_info)
+{
+	btrfs_discard_stop(fs_info);
+	cancel_delayed_work_sync(&fs_info->discard_ctl.work);
+	btrfs_discard_purge_list(&fs_info->discard_ctl);
+}
diff --git a/fs/btrfs/discard.h b/fs/btrfs/discard.h
new file mode 100644
index 000000000000..353228d62f5a
--- /dev/null
+++ b/fs/btrfs/discard.h
@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef BTRFS_DISCARD_H
+#define BTRFS_DISCARD_H
+
+#include <linux/sizes.h>
+
+struct btrfs_fs_info;
+struct btrfs_discard_ctl;
+struct btrfs_block_group;
+
+/* Discard size limits */
+#define BTRFS_ASYNC_DISCARD_DEFAULT_MAX_SIZE		(SZ_64M)
+#define BTRFS_ASYNC_DISCARD_MAX_FILTER			(SZ_1M)
+#define BTRFS_ASYNC_DISCARD_MIN_FILTER			(SZ_32K)
+
+/* List operations */
+void btrfs_discard_check_filter(struct btrfs_block_group *block_group, u64 bytes);
+
+/* Work operations */
+void btrfs_discard_cancel_work(struct btrfs_discard_ctl *discard_ctl,
+			       struct btrfs_block_group *block_group);
+void btrfs_discard_queue_work(struct btrfs_discard_ctl *discard_ctl,
+			      struct btrfs_block_group *block_group);
+void btrfs_discard_schedule_work(struct btrfs_discard_ctl *discard_ctl,
+				 bool override);
+bool btrfs_run_discard_work(struct btrfs_discard_ctl *discard_ctl);
+
+/* Update operations */
+void btrfs_discard_calc_delay(struct btrfs_discard_ctl *discard_ctl);
+void btrfs_discard_update_discardable(struct btrfs_block_group *block_group,
+				      struct btrfs_free_space_ctl *ctl);
+
+/* Setup/cleanup operations */
+void btrfs_discard_punt_unused_bgs_list(struct btrfs_fs_info *fs_info);
+void btrfs_discard_resume(struct btrfs_fs_info *fs_info);
+void btrfs_discard_stop(struct btrfs_fs_info *fs_info);
+void btrfs_discard_init(struct btrfs_fs_info *fs_info);
+void btrfs_discard_cleanup(struct btrfs_fs_info *fs_info);
+
+#endif
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 402b61bf345c..7278789ff8a7 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -7,7 +7,6 @@
 #include <linux/blkdev.h>
 #include <linux/radix-tree.h>
 #include <linux/writeback.h>
-#include <linux/buffer_head.h>
 #include <linux/workqueue.h>
 #include <linux/kthread.h>
 #include <linux/slab.h>
@@ -41,6 +40,8 @@
 #include "tree-checker.h"
 #include "ref-verify.h"
 #include "block-group.h"
+#include "discard.h"
+#include "space-info.h"
 
 #define BTRFS_SUPER_FLAG_SUPP	(BTRFS_HEADER_FLAG_WRITTEN |\
 				 BTRFS_HEADER_FLAG_RELOC |\
@@ -97,6 +98,12 @@ void __cold btrfs_end_io_wq_exit(void)
 	kmem_cache_destroy(btrfs_end_io_wq_cache);
 }
 
+static void btrfs_free_csum_hash(struct btrfs_fs_info *fs_info)
+{
+	if (fs_info->csum_shash)
+		crypto_free_shash(fs_info->csum_shash);
+}
+
 /*
  * async submit bios are used to offload expensive checksumming
  * onto the worker threads.  They checksum file and metadata bios
@@ -202,10 +209,9 @@ void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb,
  * that covers the entire device
  */
 struct extent_map *btree_get_extent(struct btrfs_inode *inode,
-		struct page *page, size_t pg_offset, u64 start, u64 len,
-		int create)
+				    struct page *page, size_t pg_offset,
+				    u64 start, u64 len)
 {
-	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct extent_map_tree *em_tree = &inode->extent_tree;
 	struct extent_map *em;
 	int ret;
@@ -213,7 +219,6 @@ struct extent_map *btree_get_extent(struct btrfs_inode *inode,
 	read_lock(&em_tree->lock);
 	em = lookup_extent_mapping(em_tree, start, len);
 	if (em) {
-		em->bdev = fs_info->fs_devices->latest_bdev;
 		read_unlock(&em_tree->lock);
 		goto out;
 	}
@@ -228,7 +233,6 @@ struct extent_map *btree_get_extent(struct btrfs_inode *inode,
 	em->len = (u64)-1;
 	em->block_len = (u64)-1;
 	em->block_start = 0;
-	em->bdev = fs_info->fs_devices->latest_bdev;
 
 	write_lock(&em_tree->lock);
 	ret = add_extent_mapping(em_tree, em, 0);
@@ -249,47 +253,27 @@ out:
 
 /*
  * Compute the csum of a btree block and store the result to provided buffer.
- *
- * Returns error if the extent buffer cannot be mapped.
  */
-static int csum_tree_block(struct extent_buffer *buf, u8 *result)
+static void csum_tree_block(struct extent_buffer *buf, u8 *result)
 {
 	struct btrfs_fs_info *fs_info = buf->fs_info;
+	const int num_pages = fs_info->nodesize >> PAGE_SHIFT;
 	SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
-	unsigned long len;
-	unsigned long cur_len;
-	unsigned long offset = BTRFS_CSUM_SIZE;
 	char *kaddr;
-	unsigned long map_start;
-	unsigned long map_len;
-	int err;
+	int i;
 
 	shash->tfm = fs_info->csum_shash;
 	crypto_shash_init(shash);
+	kaddr = page_address(buf->pages[0]);
+	crypto_shash_update(shash, kaddr + BTRFS_CSUM_SIZE,
+			    PAGE_SIZE - BTRFS_CSUM_SIZE);
 
-	len = buf->len - offset;
-
-	while (len > 0) {
-		/*
-		 * Note: we don't need to check for the err == 1 case here, as
-		 * with the given combination of 'start = BTRFS_CSUM_SIZE (32)'
-		 * and 'min_len = 32' and the currently implemented mapping
-		 * algorithm we cannot cross a page boundary.
-		 */
-		err = map_private_extent_buffer(buf, offset, 32,
-					&kaddr, &map_start, &map_len);
-		if (WARN_ON(err))
-			return err;
-		cur_len = min(len, map_len - (offset - map_start));
-		crypto_shash_update(shash, kaddr + offset - map_start, cur_len);
-		len -= cur_len;
-		offset += cur_len;
+	for (i = 1; i < num_pages; i++) {
+		kaddr = page_address(buf->pages[i]);
+		crypto_shash_update(shash, kaddr, PAGE_SIZE);
 	}
 	memset(result, 0, BTRFS_CSUM_SIZE);
-
 	crypto_shash_final(shash, result);
-
-	return 0;
 }
 
 /*
@@ -352,6 +336,9 @@ static bool btrfs_supported_super_csum(u16 csum_type)
 {
 	switch (csum_type) {
 	case BTRFS_CSUM_TYPE_CRC32:
+	case BTRFS_CSUM_TYPE_XXHASH:
+	case BTRFS_CSUM_TYPE_SHA256:
+	case BTRFS_CSUM_TYPE_BLAKE2:
 		return true;
 	default:
 		return false;
@@ -534,10 +521,10 @@ static int csum_dirty_buffer(struct btrfs_fs_info *fs_info, struct page *page)
 		return -EUCLEAN;
 
 	ASSERT(memcmp_extent_buffer(eb, fs_info->fs_devices->metadata_uuid,
-			btrfs_header_fsid(), BTRFS_FSID_SIZE) == 0);
+				    offsetof(struct btrfs_header, fsid),
+				    BTRFS_FSID_SIZE) == 0);
 
-	if (csum_tree_block(eb, result))
-		return -EINVAL;
+	csum_tree_block(eb, result);
 
 	if (btrfs_header_level(eb))
 		ret = btrfs_check_node(eb);
@@ -545,9 +532,11 @@ static int csum_dirty_buffer(struct btrfs_fs_info *fs_info, struct page *page)
 		ret = btrfs_check_leaf_full(eb);
 
 	if (ret < 0) {
+		btrfs_print_tree(eb, 0);
 		btrfs_err(fs_info,
 		"block=%llu write time tree block corruption detected",
 			  eb->start);
+		WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG));
 		return ret;
 	}
 	write_extent_buffer(eb, result, 0, csum_size);
@@ -562,7 +551,8 @@ static int check_tree_block_fsid(struct extent_buffer *eb)
 	u8 fsid[BTRFS_FSID_SIZE];
 	int ret = 1;
 
-	read_extent_buffer(eb, fsid, btrfs_header_fsid(), BTRFS_FSID_SIZE);
+	read_extent_buffer(eb, fsid, offsetof(struct btrfs_header, fsid),
+			   BTRFS_FSID_SIZE);
 	while (fs_devices) {
 		u8 *metadata_uuid;
 
@@ -593,9 +583,8 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
 	u64 found_start;
 	int found_level;
 	struct extent_buffer *eb;
-	struct btrfs_root *root = BTRFS_I(page->mapping->host)->root;
-	struct btrfs_fs_info *fs_info = root->fs_info;
-	u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
+	struct btrfs_fs_info *fs_info;
+	u16 csum_size;
 	int ret = 0;
 	u8 result[BTRFS_CSUM_SIZE];
 	int reads_done;
@@ -604,11 +593,13 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
 		goto out;
 
 	eb = (struct extent_buffer *)page->private;
+	fs_info = eb->fs_info;
+	csum_size = btrfs_super_csum_size(fs_info->super_copy);
 
 	/* the pending IO might have been the only thing that kept this buffer
 	 * in memory.  Make sure we have a ref for all this other checks
 	 */
-	extent_buffer_get(eb);
+	atomic_inc(&eb->refs);
 
 	reads_done = atomic_dec_and_test(&eb->io_pages);
 	if (!reads_done)
@@ -644,9 +635,7 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio,
 	btrfs_set_buffer_lockdep_class(btrfs_header_owner(eb),
 				       eb, found_level);
 
-	ret = csum_tree_block(eb, result);
-	if (ret)
-		goto err;
+	csum_tree_block(eb, result);
 
 	if (memcmp_extent_buffer(eb, result, 0, csum_size)) {
 		u32 val;
@@ -706,43 +695,31 @@ static void end_workqueue_bio(struct bio *bio)
 	struct btrfs_end_io_wq *end_io_wq = bio->bi_private;
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_workqueue *wq;
-	btrfs_work_func_t func;
 
 	fs_info = end_io_wq->info;
 	end_io_wq->status = bio->bi_status;
 
 	if (bio_op(bio) == REQ_OP_WRITE) {
-		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA) {
+		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA)
 			wq = fs_info->endio_meta_write_workers;
-			func = btrfs_endio_meta_write_helper;
-		} else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE) {
+		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE)
 			wq = fs_info->endio_freespace_worker;
-			func = btrfs_freespace_write_helper;
-		} else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) {
+		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
 			wq = fs_info->endio_raid56_workers;
-			func = btrfs_endio_raid56_helper;
-		} else {
+		else
 			wq = fs_info->endio_write_workers;
-			func = btrfs_endio_write_helper;
-		}
 	} else {
-		if (unlikely(end_io_wq->metadata ==
-			     BTRFS_WQ_ENDIO_DIO_REPAIR)) {
+		if (unlikely(end_io_wq->metadata == BTRFS_WQ_ENDIO_DIO_REPAIR))
 			wq = fs_info->endio_repair_workers;
-			func = btrfs_endio_repair_helper;
-		} else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56) {
+		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
 			wq = fs_info->endio_raid56_workers;
-			func = btrfs_endio_raid56_helper;
-		} else if (end_io_wq->metadata) {
+		else if (end_io_wq->metadata)
 			wq = fs_info->endio_meta_workers;
-			func = btrfs_endio_meta_helper;
-		} else {
+		else
 			wq = fs_info->endio_workers;
-			func = btrfs_endio_helper;
-		}
 	}
 
-	btrfs_init_work(&end_io_wq->work, func, end_workqueue_fn, NULL, NULL);
+	btrfs_init_work(&end_io_wq->work, end_workqueue_fn, NULL, NULL);
 	btrfs_queue_work(wq, &end_io_wq->work);
 }
 
@@ -803,8 +780,13 @@ static void run_one_async_done(struct btrfs_work *work)
 		return;
 	}
 
-	ret = btrfs_map_bio(btrfs_sb(inode->i_sb), async->bio,
-			async->mirror_num, 1);
+	/*
+	 * All of the bios that pass through here are from async helpers.
+	 * Use REQ_CGROUP_PUNT to issue them from the owning cgroup's context.
+	 * This changes nothing when cgroups aren't in use.
+	 */
+	async->bio->bi_opf |= REQ_CGROUP_PUNT;
+	ret = btrfs_map_bio(btrfs_sb(inode->i_sb), async->bio, async->mirror_num);
 	if (ret) {
 		async->bio->bi_status = ret;
 		bio_endio(async->bio);
@@ -835,8 +817,8 @@ blk_status_t btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 	async->mirror_num = mirror_num;
 	async->submit_bio_start = submit_bio_start;
 
-	btrfs_init_work(&async->work, btrfs_worker_helper, run_one_async_start,
-			run_one_async_done, run_one_async_free);
+	btrfs_init_work(&async->work, run_one_async_start, run_one_async_done,
+			run_one_async_free);
 
 	async->bio_offset = bio_offset;
 
@@ -904,12 +886,12 @@ static blk_status_t btree_submit_bio_hook(struct inode *inode, struct bio *bio,
 					  BTRFS_WQ_ENDIO_METADATA);
 		if (ret)
 			goto out_w_error;
-		ret = btrfs_map_bio(fs_info, bio, mirror_num, 0);
+		ret = btrfs_map_bio(fs_info, bio, mirror_num);
 	} else if (!async) {
 		ret = btree_csum_one_bio(bio);
 		if (ret)
 			goto out_w_error;
-		ret = btrfs_map_bio(fs_info, bio, mirror_num, 0);
+		ret = btrfs_map_bio(fs_info, bio, mirror_num);
 	} else {
 		/*
 		 * kthread helpers are used to submit writes so that
@@ -976,9 +958,7 @@ static int btree_writepages(struct address_space *mapping,
 
 static int btree_readpage(struct file *file, struct page *page)
 {
-	struct extent_io_tree *tree;
-	tree = &BTRFS_I(page->mapping->host)->io_tree;
-	return extent_read_full_page(tree, page, btree_get_extent, 0);
+	return extent_read_full_page(page, btree_get_extent, 0);
 }
 
 static int btree_releasepage(struct page *page, gfp_t gfp_flags)
@@ -1000,9 +980,7 @@ static void btree_invalidatepage(struct page *page, unsigned int offset,
 		btrfs_warn(BTRFS_I(page->mapping->host)->root->fs_info,
 			   "page private not zero on page %llu",
 			   (unsigned long long)page_offset(page));
-		ClearPagePrivate(page);
-		set_page_private(page, 0);
-		put_page(page);
+		detach_page_private(page);
 	}
 }
 
@@ -1104,36 +1082,11 @@ void btrfs_clean_tree_block(struct extent_buffer *buf)
 	}
 }
 
-static struct btrfs_subvolume_writers *btrfs_alloc_subvolume_writers(void)
-{
-	struct btrfs_subvolume_writers *writers;
-	int ret;
-
-	writers = kmalloc(sizeof(*writers), GFP_NOFS);
-	if (!writers)
-		return ERR_PTR(-ENOMEM);
-
-	ret = percpu_counter_init(&writers->counter, 0, GFP_NOFS);
-	if (ret < 0) {
-		kfree(writers);
-		return ERR_PTR(ret);
-	}
-
-	init_waitqueue_head(&writers->wait);
-	return writers;
-}
-
-static void
-btrfs_free_subvolume_writers(struct btrfs_subvolume_writers *writers)
-{
-	percpu_counter_destroy(&writers->counter);
-	kfree(writers);
-}
-
 static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 			 u64 objectid)
 {
 	bool dummy = test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state);
+	root->fs_info = fs_info;
 	root->node = NULL;
 	root->commit_root = NULL;
 	root->state = 0;
@@ -1177,7 +1130,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	atomic_set(&root->log_writers, 0);
 	atomic_set(&root->log_batch, 0);
 	refcount_set(&root->refs, 1);
-	atomic_set(&root->will_be_snapshotted, 0);
 	atomic_set(&root->snapshot_force_cow, 0);
 	atomic_set(&root->nr_swapfiles, 0);
 	root->log_transid = 0;
@@ -1199,14 +1151,20 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 
 	spin_lock_init(&root->root_item_lock);
 	btrfs_qgroup_init_swapped_blocks(&root->swapped_blocks);
+#ifdef CONFIG_BTRFS_DEBUG
+	INIT_LIST_HEAD(&root->leak_list);
+	spin_lock(&fs_info->fs_roots_radix_lock);
+	list_add_tail(&root->leak_list, &fs_info->allocated_roots);
+	spin_unlock(&fs_info->fs_roots_radix_lock);
+#endif
 }
 
 static struct btrfs_root *btrfs_alloc_root(struct btrfs_fs_info *fs_info,
-		gfp_t flags)
+					   u64 objectid, gfp_t flags)
 {
 	struct btrfs_root *root = kzalloc(sizeof(*root), flags);
 	if (root)
-		root->fs_info = fs_info;
+		__setup_root(root, fs_info, objectid);
 	return root;
 }
 
@@ -1219,12 +1177,11 @@ struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info)
 	if (!fs_info)
 		return ERR_PTR(-EINVAL);
 
-	root = btrfs_alloc_root(fs_info, GFP_KERNEL);
+	root = btrfs_alloc_root(fs_info, BTRFS_ROOT_TREE_OBJECTID, GFP_KERNEL);
 	if (!root)
 		return ERR_PTR(-ENOMEM);
 
 	/* We don't use the stripesize in selftest, set it as sectorsize */
-	__setup_root(root, fs_info, BTRFS_ROOT_TREE_OBJECTID);
 	root->alloc_bytenr = 0;
 
 	return root;
@@ -1241,19 +1198,17 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 	struct btrfs_key key;
 	unsigned int nofs_flag;
 	int ret = 0;
-	uuid_le uuid = NULL_UUID_LE;
 
 	/*
 	 * We're holding a transaction handle, so use a NOFS memory allocation
 	 * context to avoid deadlock if reclaim happens.
 	 */
 	nofs_flag = memalloc_nofs_save();
-	root = btrfs_alloc_root(fs_info, GFP_KERNEL);
+	root = btrfs_alloc_root(fs_info, objectid, GFP_KERNEL);
 	memalloc_nofs_restore(nofs_flag);
 	if (!root)
 		return ERR_PTR(-ENOMEM);
 
-	__setup_root(root, fs_info, objectid);
 	root->root_key.objectid = objectid;
 	root->root_key.type = BTRFS_ROOT_ITEM_KEY;
 	root->root_key.offset = 0;
@@ -1281,8 +1236,9 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 	btrfs_set_root_last_snapshot(&root->root_item, 0);
 	btrfs_set_root_dirid(&root->root_item, 0);
 	if (is_fstree(objectid))
-		uuid_le_gen(&uuid);
-	memcpy(root->root_item.uuid, uuid.b, BTRFS_UUID_SIZE);
+		generate_random_guid(root->root_item.uuid);
+	else
+		export_guid(root->root_item.uuid, &guid_null);
 	root->root_item.drop_level = 0;
 
 	key.objectid = objectid;
@@ -1297,12 +1253,9 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 	return root;
 
 fail:
-	if (leaf) {
+	if (leaf)
 		btrfs_tree_unlock(leaf);
-		free_extent_buffer(root->commit_root);
-		free_extent_buffer(leaf);
-	}
-	kfree(root);
+	btrfs_put_root(root);
 
 	return ERR_PTR(ret);
 }
@@ -1313,12 +1266,10 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
 	struct btrfs_root *root;
 	struct extent_buffer *leaf;
 
-	root = btrfs_alloc_root(fs_info, GFP_NOFS);
+	root = btrfs_alloc_root(fs_info, BTRFS_TREE_LOG_OBJECTID, GFP_NOFS);
 	if (!root)
 		return ERR_PTR(-ENOMEM);
 
-	__setup_root(root, fs_info, BTRFS_TREE_LOG_OBJECTID);
-
 	root->root_key.objectid = BTRFS_TREE_LOG_OBJECTID;
 	root->root_key.type = BTRFS_ROOT_ITEM_KEY;
 	root->root_key.offset = BTRFS_TREE_LOG_OBJECTID;
@@ -1335,7 +1286,7 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
 	leaf = btrfs_alloc_tree_block(trans, root, 0, BTRFS_TREE_LOG_OBJECTID,
 			NULL, 0, 0, 0);
 	if (IS_ERR(leaf)) {
-		kfree(root);
+		btrfs_put_root(root);
 		return ERR_CAST(leaf);
 	}
 
@@ -1391,8 +1342,8 @@ int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
-static struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root,
-					       struct btrfs_key *key)
+struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root,
+					struct btrfs_key *key)
 {
 	struct btrfs_root *root;
 	struct btrfs_fs_info *fs_info = tree_root->fs_info;
@@ -1405,14 +1356,12 @@ static struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root,
 	if (!path)
 		return ERR_PTR(-ENOMEM);
 
-	root = btrfs_alloc_root(fs_info, GFP_NOFS);
+	root = btrfs_alloc_root(fs_info, key->objectid, GFP_NOFS);
 	if (!root) {
 		ret = -ENOMEM;
 		goto alloc_fail;
 	}
 
-	__setup_root(root, fs_info, key->objectid);
-
 	ret = btrfs_find_root(tree_root, key, path,
 			      &root->root_item, &root->root_key);
 	if (ret) {
@@ -1428,10 +1377,10 @@ static struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root,
 				     generation, level, NULL);
 	if (IS_ERR(root->node)) {
 		ret = PTR_ERR(root->node);
+		root->node = NULL;
 		goto find_fail;
 	} else if (!btrfs_buffer_uptodate(root->node, generation, 0)) {
 		ret = -EIO;
-		free_extent_buffer(root->node);
 		goto find_fail;
 	}
 	root->commit_root = btrfs_root_node(root);
@@ -1440,33 +1389,16 @@ out:
 	return root;
 
 find_fail:
-	kfree(root);
+	btrfs_put_root(root);
 alloc_fail:
 	root = ERR_PTR(ret);
 	goto out;
 }
 
-struct btrfs_root *btrfs_read_fs_root(struct btrfs_root *tree_root,
-				      struct btrfs_key *location)
-{
-	struct btrfs_root *root;
-
-	root = btrfs_read_tree_root(tree_root, location);
-	if (IS_ERR(root))
-		return root;
-
-	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
-		set_bit(BTRFS_ROOT_REF_COWS, &root->state);
-		btrfs_check_and_init_root_item(&root->root_item);
-	}
-
-	return root;
-}
-
-int btrfs_init_fs_root(struct btrfs_root *root)
+static int btrfs_init_fs_root(struct btrfs_root *root)
 {
 	int ret;
-	struct btrfs_subvolume_writers *writers;
+	unsigned int nofs_flag;
 
 	root->free_ino_ctl = kzalloc(sizeof(*root->free_ino_ctl), GFP_NOFS);
 	root->free_ino_pinned = kzalloc(sizeof(*root->free_ino_pinned),
@@ -1476,12 +1408,20 @@ int btrfs_init_fs_root(struct btrfs_root *root)
 		goto fail;
 	}
 
-	writers = btrfs_alloc_subvolume_writers();
-	if (IS_ERR(writers)) {
-		ret = PTR_ERR(writers);
+	/*
+	 * We might be called under a transaction (e.g. indirect backref
+	 * resolution) which could deadlock if it triggers memory reclaim
+	 */
+	nofs_flag = memalloc_nofs_save();
+	ret = btrfs_drew_lock_init(&root->snapshot_lock);
+	memalloc_nofs_restore(nofs_flag);
+	if (ret)
 		goto fail;
+
+	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
+		set_bit(BTRFS_ROOT_REF_COWS, &root->state);
+		btrfs_check_and_init_root_item(&root->root_item);
 	}
-	root->subv_writers = writers;
 
 	btrfs_init_free_ino_ctl(root);
 	spin_lock_init(&root->ino_cache_lock);
@@ -1509,14 +1449,16 @@ fail:
 	return ret;
 }
 
-struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
-					u64 root_id)
+static struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
+					       u64 root_id)
 {
 	struct btrfs_root *root;
 
 	spin_lock(&fs_info->fs_roots_radix_lock);
 	root = radix_tree_lookup(&fs_info->fs_roots_radix,
 				 (unsigned long)root_id);
+	if (root)
+		root = btrfs_grab_root(root);
 	spin_unlock(&fs_info->fs_roots_radix_lock);
 	return root;
 }
@@ -1534,14 +1476,62 @@ int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info,
 	ret = radix_tree_insert(&fs_info->fs_roots_radix,
 				(unsigned long)root->root_key.objectid,
 				root);
-	if (ret == 0)
+	if (ret == 0) {
+		btrfs_grab_root(root);
 		set_bit(BTRFS_ROOT_IN_RADIX, &root->state);
+	}
 	spin_unlock(&fs_info->fs_roots_radix_lock);
 	radix_tree_preload_end();
 
 	return ret;
 }
 
+void btrfs_check_leaked_roots(struct btrfs_fs_info *fs_info)
+{
+#ifdef CONFIG_BTRFS_DEBUG
+	struct btrfs_root *root;
+
+	while (!list_empty(&fs_info->allocated_roots)) {
+		root = list_first_entry(&fs_info->allocated_roots,
+					struct btrfs_root, leak_list);
+		btrfs_err(fs_info, "leaked root %llu-%llu refcount %d",
+			  root->root_key.objectid, root->root_key.offset,
+			  refcount_read(&root->refs));
+		while (refcount_read(&root->refs) > 1)
+			btrfs_put_root(root);
+		btrfs_put_root(root);
+	}
+#endif
+}
+
+void btrfs_free_fs_info(struct btrfs_fs_info *fs_info)
+{
+	percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
+	percpu_counter_destroy(&fs_info->delalloc_bytes);
+	percpu_counter_destroy(&fs_info->dio_bytes);
+	percpu_counter_destroy(&fs_info->dev_replace.bio_counter);
+	btrfs_free_csum_hash(fs_info);
+	btrfs_free_stripe_hash_table(fs_info);
+	btrfs_free_ref_cache(fs_info);
+	kfree(fs_info->balance_ctl);
+	kfree(fs_info->delayed_root);
+	btrfs_put_root(fs_info->extent_root);
+	btrfs_put_root(fs_info->tree_root);
+	btrfs_put_root(fs_info->chunk_root);
+	btrfs_put_root(fs_info->dev_root);
+	btrfs_put_root(fs_info->csum_root);
+	btrfs_put_root(fs_info->quota_root);
+	btrfs_put_root(fs_info->uuid_root);
+	btrfs_put_root(fs_info->free_space_root);
+	btrfs_put_root(fs_info->fs_root);
+	btrfs_check_leaked_roots(fs_info);
+	btrfs_extent_buffer_leak_debug_check(fs_info);
+	kfree(fs_info->super_copy);
+	kfree(fs_info->super_for_commit);
+	kvfree(fs_info);
+}
+
+
 struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info,
 				     struct btrfs_key *location,
 				     bool check_ref)
@@ -1552,33 +1542,35 @@ struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info,
 	int ret;
 
 	if (location->objectid == BTRFS_ROOT_TREE_OBJECTID)
-		return fs_info->tree_root;
+		return btrfs_grab_root(fs_info->tree_root);
 	if (location->objectid == BTRFS_EXTENT_TREE_OBJECTID)
-		return fs_info->extent_root;
+		return btrfs_grab_root(fs_info->extent_root);
 	if (location->objectid == BTRFS_CHUNK_TREE_OBJECTID)
-		return fs_info->chunk_root;
+		return btrfs_grab_root(fs_info->chunk_root);
 	if (location->objectid == BTRFS_DEV_TREE_OBJECTID)
-		return fs_info->dev_root;
+		return btrfs_grab_root(fs_info->dev_root);
 	if (location->objectid == BTRFS_CSUM_TREE_OBJECTID)
-		return fs_info->csum_root;
+		return btrfs_grab_root(fs_info->csum_root);
 	if (location->objectid == BTRFS_QUOTA_TREE_OBJECTID)
-		return fs_info->quota_root ? fs_info->quota_root :
-					     ERR_PTR(-ENOENT);
+		return btrfs_grab_root(fs_info->quota_root) ?
+			fs_info->quota_root : ERR_PTR(-ENOENT);
 	if (location->objectid == BTRFS_UUID_TREE_OBJECTID)
-		return fs_info->uuid_root ? fs_info->uuid_root :
-					    ERR_PTR(-ENOENT);
+		return btrfs_grab_root(fs_info->uuid_root) ?
+			fs_info->uuid_root : ERR_PTR(-ENOENT);
 	if (location->objectid == BTRFS_FREE_SPACE_TREE_OBJECTID)
-		return fs_info->free_space_root ? fs_info->free_space_root :
-						  ERR_PTR(-ENOENT);
+		return btrfs_grab_root(fs_info->free_space_root) ?
+			fs_info->free_space_root : ERR_PTR(-ENOENT);
 again:
 	root = btrfs_lookup_fs_root(fs_info, location->objectid);
 	if (root) {
-		if (check_ref && btrfs_root_refs(&root->root_item) == 0)
+		if (check_ref && btrfs_root_refs(&root->root_item) == 0) {
+			btrfs_put_root(root);
 			return ERR_PTR(-ENOENT);
+		}
 		return root;
 	}
 
-	root = btrfs_read_fs_root(fs_info->tree_root, location);
+	root = btrfs_read_tree_root(fs_info->tree_root, location);
 	if (IS_ERR(root))
 		return root;
 
@@ -1609,15 +1601,14 @@ again:
 
 	ret = btrfs_insert_fs_root(fs_info, root);
 	if (ret) {
-		if (ret == -EEXIST) {
-			btrfs_free_fs_root(root);
+		btrfs_put_root(root);
+		if (ret == -EEXIST)
 			goto again;
-		}
 		goto fail;
 	}
 	return root;
 fail:
-	btrfs_free_fs_root(root);
+	btrfs_put_root(root);
 	return ERR_PTR(ret);
 }
 
@@ -1657,8 +1648,8 @@ static void end_workqueue_fn(struct btrfs_work *work)
 	bio->bi_status = end_io_wq->status;
 	bio->bi_private = end_io_wq->private;
 	bio->bi_end_io = end_io_wq->end_io;
-	kmem_cache_free(btrfs_end_io_wq_cache, end_io_wq);
 	bio_endio(bio);
+	kmem_cache_free(btrfs_end_io_wq_cache, end_io_wq);
 }
 
 static int cleaner_kthread(void *arg)
@@ -1753,7 +1744,7 @@ static int transaction_kthread(void *arg)
 		}
 
 		now = ktime_get_seconds();
-		if (cur->state < TRANS_STATE_BLOCKED &&
+		if (cur->state < TRANS_STATE_COMMIT_START &&
 		    !test_bit(BTRFS_FS_NEED_ASYNC_COMMIT, &fs_info->flags) &&
 		    (now < cur->start_time ||
 		     now - cur->start_time < fs_info->commit_interval)) {
@@ -1792,18 +1783,18 @@ sleep:
 }
 
 /*
- * this will find the highest generation in the array of
- * root backups.  The index of the highest array is returned,
- * or -1 if we can't find anything.
+ * This will find the highest generation in the array of root backups.  The
+ * index of the highest array is returned, or -EINVAL if we can't find
+ * anything.
  *
  * We check to make sure the array is valid by comparing the
  * generation of the latest  root in the array with the generation
  * in the super block.  If they don't match we pitch it.
  */
-static int find_newest_super_backup(struct btrfs_fs_info *info, u64 newest_gen)
+static int find_newest_super_backup(struct btrfs_fs_info *info)
 {
+	const u64 newest_gen = btrfs_super_generation(info->super_copy);
 	u64 cur;
-	int newest_index = -1;
 	struct btrfs_root_backup *root_backup;
 	int i;
 
@@ -1811,37 +1802,10 @@ static int find_newest_super_backup(struct btrfs_fs_info *info, u64 newest_gen)
 		root_backup = info->super_copy->super_roots + i;
 		cur = btrfs_backup_tree_root_gen(root_backup);
 		if (cur == newest_gen)
-			newest_index = i;
+			return i;
 	}
 
-	/* check to see if we actually wrapped around */
-	if (newest_index == BTRFS_NUM_BACKUP_ROOTS - 1) {
-		root_backup = info->super_copy->super_roots;
-		cur = btrfs_backup_tree_root_gen(root_backup);
-		if (cur == newest_gen)
-			newest_index = 0;
-	}
-	return newest_index;
-}
-
-
-/*
- * find the oldest backup so we know where to store new entries
- * in the backup array.  This will set the backup_root_index
- * field in the fs_info struct
- */
-static void find_oldest_super_backup(struct btrfs_fs_info *info,
-				     u64 newest_gen)
-{
-	int newest_index = -1;
-
-	newest_index = find_newest_super_backup(info, newest_gen);
-	/* if there was garbage in there, just move along */
-	if (newest_index == -1) {
-		info->backup_root_index = 0;
-	} else {
-		info->backup_root_index = (newest_index + 1) % BTRFS_NUM_BACKUP_ROOTS;
-	}
+	return -EINVAL;
 }
 
 /*
@@ -1851,22 +1815,8 @@ static void find_oldest_super_backup(struct btrfs_fs_info *info,
  */
 static void backup_super_roots(struct btrfs_fs_info *info)
 {
-	int next_backup;
+	const int next_backup = info->backup_root_index;
 	struct btrfs_root_backup *root_backup;
-	int last_backup;
-
-	next_backup = info->backup_root_index;
-	last_backup = (next_backup + BTRFS_NUM_BACKUP_ROOTS - 1) %
-		BTRFS_NUM_BACKUP_ROOTS;
-
-	/*
-	 * just overwrite the last backup if we're at the same generation
-	 * this happens only at umount
-	 */
-	root_backup = info->super_for_commit->super_roots + last_backup;
-	if (btrfs_backup_tree_root_gen(root_backup) ==
-	    btrfs_header_generation(info->tree_root->node))
-		next_backup = last_backup;
 
 	root_backup = info->super_for_commit->super_roots + next_backup;
 
@@ -1939,40 +1889,31 @@ static void backup_super_roots(struct btrfs_fs_info *info)
 }
 
 /*
- * this copies info out of the root backup array and back into
- * the in-memory super block.  It is meant to help iterate through
- * the array, so you send it the number of backups you've already
- * tried and the last backup index you used.
+ * read_backup_root - Reads a backup root based on the passed priority. Prio 0
+ * is the newest, prio 1/2/3 are 2nd newest/3rd newest/4th (oldest) backup roots
  *
- * this returns -1 when it has tried all the backups
+ * fs_info - filesystem whose backup roots need to be read
+ * priority - priority of backup root required
+ *
+ * Returns backup root index on success and -EINVAL otherwise.
  */
-static noinline int next_root_backup(struct btrfs_fs_info *info,
-				     struct btrfs_super_block *super,
-				     int *num_backups_tried, int *backup_index)
+static int read_backup_root(struct btrfs_fs_info *fs_info, u8 priority)
 {
+	int backup_index = find_newest_super_backup(fs_info);
+	struct btrfs_super_block *super = fs_info->super_copy;
 	struct btrfs_root_backup *root_backup;
-	int newest = *backup_index;
-
-	if (*num_backups_tried == 0) {
-		u64 gen = btrfs_super_generation(super);
 
-		newest = find_newest_super_backup(info, gen);
-		if (newest == -1)
-			return -1;
+	if (priority < BTRFS_NUM_BACKUP_ROOTS && backup_index >= 0) {
+		if (priority == 0)
+			return backup_index;
 
-		*backup_index = newest;
-		*num_backups_tried = 1;
-	} else if (*num_backups_tried == BTRFS_NUM_BACKUP_ROOTS) {
-		/* we've tried all the backups, all done */
-		return -1;
+		backup_index = backup_index + BTRFS_NUM_BACKUP_ROOTS - priority;
+		backup_index %= BTRFS_NUM_BACKUP_ROOTS;
 	} else {
-		/* jump to the next oldest backup */
-		newest = (*backup_index + BTRFS_NUM_BACKUP_ROOTS - 1) %
-			BTRFS_NUM_BACKUP_ROOTS;
-		*backup_index = newest;
-		*num_backups_tried += 1;
+		return -EINVAL;
 	}
-	root_backup = super->super_roots + newest;
+
+	root_backup = super->super_roots + backup_index;
 
 	btrfs_set_super_generation(super,
 				   btrfs_backup_tree_root_gen(root_backup));
@@ -1982,12 +1923,13 @@ static noinline int next_root_backup(struct btrfs_fs_info *info,
 	btrfs_set_super_bytes_used(super, btrfs_backup_bytes_used(root_backup));
 
 	/*
-	 * fixme: the total bytes and num_devices need to match or we should
+	 * Fixme: the total bytes and num_devices need to match or we should
 	 * need a fsck
 	 */
 	btrfs_set_super_total_bytes(super, btrfs_backup_total_bytes(root_backup));
 	btrfs_set_super_num_devices(super, btrfs_backup_num_devices(root_backup));
-	return 0;
+
+	return backup_index;
 }
 
 /* helper to cleanup workers */
@@ -2002,12 +1944,13 @@ static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
 	btrfs_destroy_workqueue(fs_info->rmw_workers);
 	btrfs_destroy_workqueue(fs_info->endio_write_workers);
 	btrfs_destroy_workqueue(fs_info->endio_freespace_worker);
-	btrfs_destroy_workqueue(fs_info->submit_workers);
 	btrfs_destroy_workqueue(fs_info->delayed_workers);
 	btrfs_destroy_workqueue(fs_info->caching_workers);
 	btrfs_destroy_workqueue(fs_info->readahead_workers);
 	btrfs_destroy_workqueue(fs_info->flush_workers);
 	btrfs_destroy_workqueue(fs_info->qgroup_rescan_workers);
+	if (fs_info->discard_ctl.discard_workers)
+		destroy_workqueue(fs_info->discard_ctl.discard_workers);
 	/*
 	 * Now that all other work queues are destroyed, we can safely destroy
 	 * the queues used for metadata I/O, since tasks from those other work
@@ -2028,7 +1971,7 @@ static void free_root_extent_buffers(struct btrfs_root *root)
 }
 
 /* helper to cleanup tree roots */
-static void free_root_pointers(struct btrfs_fs_info *info, int chunk_root)
+static void free_root_pointers(struct btrfs_fs_info *info, bool free_chunk_root)
 {
 	free_root_extent_buffers(info->tree_root);
 
@@ -2037,11 +1980,35 @@ static void free_root_pointers(struct btrfs_fs_info *info, int chunk_root)
 	free_root_extent_buffers(info->csum_root);
 	free_root_extent_buffers(info->quota_root);
 	free_root_extent_buffers(info->uuid_root);
-	if (chunk_root)
+	free_root_extent_buffers(info->fs_root);
+	if (free_chunk_root)
 		free_root_extent_buffers(info->chunk_root);
 	free_root_extent_buffers(info->free_space_root);
 }
 
+void btrfs_put_root(struct btrfs_root *root)
+{
+	if (!root)
+		return;
+
+	if (refcount_dec_and_test(&root->refs)) {
+		WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
+		if (root->anon_dev)
+			free_anon_bdev(root->anon_dev);
+		btrfs_drew_lock_destroy(&root->snapshot_lock);
+		free_extent_buffer(root->node);
+		free_extent_buffer(root->commit_root);
+		kfree(root->free_ino_ctl);
+		kfree(root->free_ino_pinned);
+#ifdef CONFIG_BTRFS_DEBUG
+		spin_lock(&root->fs_info->fs_roots_radix_lock);
+		list_del_init(&root->leak_list);
+		spin_unlock(&root->fs_info->fs_roots_radix_lock);
+#endif
+		kfree(root);
+	}
+}
+
 void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info)
 {
 	int ret;
@@ -2053,13 +2020,9 @@ void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info)
 				     struct btrfs_root, root_list);
 		list_del(&gang[0]->root_list);
 
-		if (test_bit(BTRFS_ROOT_IN_RADIX, &gang[0]->state)) {
+		if (test_bit(BTRFS_ROOT_IN_RADIX, &gang[0]->state))
 			btrfs_drop_and_free_fs_root(fs_info, gang[0]);
-		} else {
-			free_extent_buffer(gang[0]->node);
-			free_extent_buffer(gang[0]->commit_root);
-			btrfs_put_fs_root(gang[0]);
-		}
+		btrfs_put_root(gang[0]);
 	}
 
 	while (1) {
@@ -2071,11 +2034,6 @@ void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info)
 		for (i = 0; i < ret; i++)
 			btrfs_drop_and_free_fs_root(fs_info, gang[i]);
 	}
-
-	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
-		btrfs_free_log_root_tree(NULL, fs_info);
-		btrfs_destroy_pinned_extent(fs_info, fs_info->pinned_extents);
-	}
 }
 
 static void btrfs_init_scrub(struct btrfs_fs_info *fs_info)
@@ -2121,7 +2079,7 @@ static void btrfs_init_btree_inode(struct btrfs_fs_info *fs_info)
 
 	BTRFS_I(inode)->io_tree.ops = &btree_extent_io_ops;
 
-	BTRFS_I(inode)->root = fs_info->tree_root;
+	BTRFS_I(inode)->root = btrfs_grab_root(fs_info->tree_root);
 	memset(&BTRFS_I(inode)->location, 0, sizeof(struct btrfs_key));
 	set_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags);
 	btrfs_insert_inode_hash(inode);
@@ -2167,16 +2125,6 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 	fs_info->caching_workers =
 		btrfs_alloc_workqueue(fs_info, "cache", flags, max_active, 0);
 
-	/*
-	 * a higher idle thresh on the submit workers makes it much more
-	 * likely that bios will be send down in a sane order to the
-	 * devices
-	 */
-	fs_info->submit_workers =
-		btrfs_alloc_workqueue(fs_info, "submit", flags,
-				      min_t(u64, fs_devices->num_devices,
-					    max_active), 64);
-
 	fs_info->fixup_workers =
 		btrfs_alloc_workqueue(fs_info, "fixup", flags, 1, 0);
 
@@ -2213,9 +2161,11 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 				      max_active, 2);
 	fs_info->qgroup_rescan_workers =
 		btrfs_alloc_workqueue(fs_info, "qgroup-rescan", flags, 1, 0);
+	fs_info->discard_ctl.discard_workers =
+		alloc_workqueue("btrfs_discard", WQ_UNBOUND | WQ_FREEZABLE, 1);
 
 	if (!(fs_info->workers && fs_info->delalloc_workers &&
-	      fs_info->submit_workers && fs_info->flush_workers &&
+	      fs_info->flush_workers &&
 	      fs_info->endio_workers && fs_info->endio_meta_workers &&
 	      fs_info->endio_meta_write_workers &&
 	      fs_info->endio_repair_workers &&
@@ -2223,7 +2173,8 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 	      fs_info->endio_freespace_worker && fs_info->rmw_workers &&
 	      fs_info->caching_workers && fs_info->readahead_workers &&
 	      fs_info->fixup_workers && fs_info->delayed_workers &&
-	      fs_info->qgroup_rescan_workers)) {
+	      fs_info->qgroup_rescan_workers &&
+	      fs_info->discard_ctl.discard_workers)) {
 		return -ENOMEM;
 	}
 
@@ -2233,13 +2184,13 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 static int btrfs_init_csum_hash(struct btrfs_fs_info *fs_info, u16 csum_type)
 {
 	struct crypto_shash *csum_shash;
-	const char *csum_name = btrfs_super_csum_name(csum_type);
+	const char *csum_driver = btrfs_super_csum_driver(csum_type);
 
-	csum_shash = crypto_alloc_shash(csum_name, 0, 0);
+	csum_shash = crypto_alloc_shash(csum_driver, 0, 0);
 
 	if (IS_ERR(csum_shash)) {
 		btrfs_err(fs_info, "error allocating %s hash for checksum",
-			  csum_name);
+			  csum_driver);
 		return PTR_ERR(csum_shash);
 	}
 
@@ -2248,11 +2199,6 @@ static int btrfs_init_csum_hash(struct btrfs_fs_info *fs_info, u16 csum_type)
 	return 0;
 }
 
-static void btrfs_free_csum_hash(struct btrfs_fs_info *fs_info)
-{
-	crypto_free_shash(fs_info->csum_shash);
-}
-
 static int btrfs_replay_log(struct btrfs_fs_info *fs_info,
 			    struct btrfs_fs_devices *fs_devices)
 {
@@ -2267,24 +2213,23 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info,
 		return -EIO;
 	}
 
-	log_tree_root = btrfs_alloc_root(fs_info, GFP_KERNEL);
+	log_tree_root = btrfs_alloc_root(fs_info, BTRFS_TREE_LOG_OBJECTID,
+					 GFP_KERNEL);
 	if (!log_tree_root)
 		return -ENOMEM;
 
-	__setup_root(log_tree_root, fs_info, BTRFS_TREE_LOG_OBJECTID);
-
 	log_tree_root->node = read_tree_block(fs_info, bytenr,
 					      fs_info->generation + 1,
 					      level, NULL);
 	if (IS_ERR(log_tree_root->node)) {
 		btrfs_warn(fs_info, "failed to read log tree");
 		ret = PTR_ERR(log_tree_root->node);
-		kfree(log_tree_root);
+		log_tree_root->node = NULL;
+		btrfs_put_root(log_tree_root);
 		return ret;
 	} else if (!extent_buffer_uptodate(log_tree_root->node)) {
 		btrfs_err(fs_info, "failed to read log tree");
-		free_extent_buffer(log_tree_root->node);
-		kfree(log_tree_root);
+		btrfs_put_root(log_tree_root);
 		return -EIO;
 	}
 	/* returns with log_tree_root freed on success */
@@ -2292,8 +2237,7 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info,
 	if (ret) {
 		btrfs_handle_fs_error(fs_info, ret,
 				      "Failed to recover log tree");
-		free_extent_buffer(log_tree_root->node);
-		kfree(log_tree_root);
+		btrfs_put_root(log_tree_root);
 		return ret;
 	}
 
@@ -2589,69 +2533,102 @@ out:
 	return ret;
 }
 
-int open_ctree(struct super_block *sb,
-	       struct btrfs_fs_devices *fs_devices,
-	       char *options)
+static int __cold init_tree_roots(struct btrfs_fs_info *fs_info)
 {
-	u32 sectorsize;
-	u32 nodesize;
-	u32 stripesize;
-	u64 generation;
-	u64 features;
-	u16 csum_type;
-	struct btrfs_key location;
-	struct buffer_head *bh;
-	struct btrfs_super_block *disk_super;
-	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
-	struct btrfs_root *tree_root;
-	struct btrfs_root *chunk_root;
-	int ret;
-	int err = -EINVAL;
-	int num_backups_tried = 0;
-	int backup_index = 0;
-	int clear_free_space_tree = 0;
-	int level;
+	int backup_index = find_newest_super_backup(fs_info);
+	struct btrfs_super_block *sb = fs_info->super_copy;
+	struct btrfs_root *tree_root = fs_info->tree_root;
+	bool handle_error = false;
+	int ret = 0;
+	int i;
 
-	tree_root = fs_info->tree_root = btrfs_alloc_root(fs_info, GFP_KERNEL);
-	chunk_root = fs_info->chunk_root = btrfs_alloc_root(fs_info, GFP_KERNEL);
-	if (!tree_root || !chunk_root) {
-		err = -ENOMEM;
-		goto fail;
-	}
+	for (i = 0; i < BTRFS_NUM_BACKUP_ROOTS; i++) {
+		u64 generation;
+		int level;
 
-	ret = init_srcu_struct(&fs_info->subvol_srcu);
-	if (ret) {
-		err = ret;
-		goto fail;
-	}
+		if (handle_error) {
+			if (!IS_ERR(tree_root->node))
+				free_extent_buffer(tree_root->node);
+			tree_root->node = NULL;
 
-	ret = percpu_counter_init(&fs_info->dio_bytes, 0, GFP_KERNEL);
-	if (ret) {
-		err = ret;
-		goto fail_srcu;
-	}
+			if (!btrfs_test_opt(fs_info, USEBACKUPROOT))
+				break;
 
-	ret = percpu_counter_init(&fs_info->dirty_metadata_bytes, 0, GFP_KERNEL);
-	if (ret) {
-		err = ret;
-		goto fail_dio_bytes;
-	}
-	fs_info->dirty_metadata_batch = PAGE_SIZE *
-					(1 + ilog2(nr_cpu_ids));
+			free_root_pointers(fs_info, 0);
 
-	ret = percpu_counter_init(&fs_info->delalloc_bytes, 0, GFP_KERNEL);
-	if (ret) {
-		err = ret;
-		goto fail_dirty_metadata_bytes;
-	}
+			/*
+			 * Don't use the log in recovery mode, it won't be
+			 * valid
+			 */
+			btrfs_set_super_log_root(sb, 0);
 
-	ret = percpu_counter_init(&fs_info->dev_replace.bio_counter, 0,
-			GFP_KERNEL);
-	if (ret) {
-		err = ret;
-		goto fail_delalloc_bytes;
+			/* We can't trust the free space cache either */
+			btrfs_set_opt(fs_info->mount_opt, CLEAR_CACHE);
+
+			ret = read_backup_root(fs_info, i);
+			backup_index = ret;
+			if (ret < 0)
+				return ret;
+		}
+		generation = btrfs_super_generation(sb);
+		level = btrfs_super_root_level(sb);
+		tree_root->node = read_tree_block(fs_info, btrfs_super_root(sb),
+						  generation, level, NULL);
+		if (IS_ERR(tree_root->node) ||
+		    !extent_buffer_uptodate(tree_root->node)) {
+			handle_error = true;
+
+			if (IS_ERR(tree_root->node))
+				ret = PTR_ERR(tree_root->node);
+			else if (!extent_buffer_uptodate(tree_root->node))
+				ret = -EUCLEAN;
+
+			btrfs_warn(fs_info, "failed to read tree root");
+			continue;
+		}
+
+		btrfs_set_root_node(&tree_root->root_item, tree_root->node);
+		tree_root->commit_root = btrfs_root_node(tree_root);
+		btrfs_set_root_refs(&tree_root->root_item, 1);
+
+		/*
+		 * No need to hold btrfs_root::objectid_mutex since the fs
+		 * hasn't been fully initialised and we are the only user
+		 */
+		ret = btrfs_find_highest_objectid(tree_root,
+						&tree_root->highest_objectid);
+		if (ret < 0) {
+			handle_error = true;
+			continue;
+		}
+
+		ASSERT(tree_root->highest_objectid <= BTRFS_LAST_FREE_OBJECTID);
+
+		ret = btrfs_read_roots(fs_info);
+		if (ret < 0) {
+			handle_error = true;
+			continue;
+		}
+
+		/* All successful */
+		fs_info->generation = generation;
+		fs_info->last_trans_committed = generation;
+
+		/* Always begin writing backup roots after the one being used */
+		if (backup_index < 0) {
+			fs_info->backup_root_index = 0;
+		} else {
+			fs_info->backup_root_index = backup_index + 1;
+			fs_info->backup_root_index %= BTRFS_NUM_BACKUP_ROOTS;
+		}
+		break;
 	}
 
+	return ret;
+}
+
+void btrfs_init_fs_info(struct btrfs_fs_info *fs_info)
+{
 	INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
 	INIT_RADIX_TREE(&fs_info->buffer_radix, GFP_ATOMIC);
 	INIT_LIST_HEAD(&fs_info->trans_list);
@@ -2664,7 +2641,6 @@ int open_ctree(struct super_block *sb,
 	spin_lock_init(&fs_info->fs_roots_radix_lock);
 	spin_lock_init(&fs_info->delayed_iput_lock);
 	spin_lock_init(&fs_info->defrag_inodes_lock);
-	spin_lock_init(&fs_info->tree_mod_seq_lock);
 	spin_lock_init(&fs_info->super_lock);
 	spin_lock_init(&fs_info->buffer_lock);
 	spin_lock_init(&fs_info->unused_bgs_lock);
@@ -2679,6 +2655,11 @@ int open_ctree(struct super_block *sb,
 	INIT_LIST_HEAD(&fs_info->space_info);
 	INIT_LIST_HEAD(&fs_info->tree_mod_seq_list);
 	INIT_LIST_HEAD(&fs_info->unused_bgs);
+#ifdef CONFIG_BTRFS_DEBUG
+	INIT_LIST_HEAD(&fs_info->allocated_roots);
+	INIT_LIST_HEAD(&fs_info->allocated_ebs);
+	spin_lock_init(&fs_info->eb_leak_lock);
+#endif
 	extent_map_tree_init(&fs_info->mapping_tree);
 	btrfs_init_block_rsv(&fs_info->global_block_rsv,
 			     BTRFS_BLOCK_RSV_GLOBAL);
@@ -2695,7 +2676,6 @@ int open_ctree(struct super_block *sb,
 	atomic_set(&fs_info->reada_works_cnt, 0);
 	atomic_set(&fs_info->nr_delayed_iputs, 0);
 	atomic64_set(&fs_info->tree_mod_seq, 0);
-	fs_info->sb = sb;
 	fs_info->max_inline = BTRFS_DEFAULT_MAX_INLINE;
 	fs_info->metadata_ratio = 0;
 	fs_info->defrag_inodes = RB_ROOT;
@@ -2714,21 +2694,6 @@ int open_ctree(struct super_block *sb,
 	INIT_LIST_HEAD(&fs_info->ordered_roots);
 	spin_lock_init(&fs_info->ordered_root_lock);
 
-	fs_info->btree_inode = new_inode(sb);
-	if (!fs_info->btree_inode) {
-		err = -ENOMEM;
-		goto fail_bio_counter;
-	}
-	mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS);
-
-	fs_info->delayed_root = kmalloc(sizeof(struct btrfs_delayed_root),
-					GFP_KERNEL);
-	if (!fs_info->delayed_root) {
-		err = -ENOMEM;
-		goto fail_iput;
-	}
-	btrfs_init_delayed_root(fs_info->delayed_root);
-
 	btrfs_init_scrub(fs_info);
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
 	fs_info->check_integrity_print_mask = 0;
@@ -2736,20 +2701,12 @@ int open_ctree(struct super_block *sb,
 	btrfs_init_balance(fs_info);
 	btrfs_init_async_reclaim_work(&fs_info->async_reclaim_work);
 
-	sb->s_blocksize = BTRFS_BDEV_BLOCKSIZE;
-	sb->s_blocksize_bits = blksize_bits(BTRFS_BDEV_BLOCKSIZE);
-
-	btrfs_init_btree_inode(fs_info);
-
 	spin_lock_init(&fs_info->block_group_cache_lock);
 	fs_info->block_group_cache_tree = RB_ROOT;
 	fs_info->first_logical_byte = (u64)-1;
 
-	extent_io_tree_init(fs_info, &fs_info->freed_extents[0],
-			    IO_TREE_FS_INFO_FREED_EXTENTS0, NULL);
-	extent_io_tree_init(fs_info, &fs_info->freed_extents[1],
-			    IO_TREE_FS_INFO_FREED_EXTENTS1, NULL);
-	fs_info->pinned_extents = &fs_info->freed_extents[0];
+	extent_io_tree_init(fs_info, &fs_info->excluded_extents,
+			    IO_TREE_FS_EXCLUDED_EXTENTS, NULL);
 	set_bit(BTRFS_FS_BARRIER, &fs_info->flags);
 
 	mutex_init(&fs_info->ordered_operations_mutex);
@@ -2765,6 +2722,7 @@ int open_ctree(struct super_block *sb,
 
 	btrfs_init_dev_replace_locks(fs_info);
 	btrfs_init_qgroup(fs_info);
+	btrfs_discard_init(fs_info);
 
 	btrfs_init_free_cluster(&fs_info->meta_alloc_cluster);
 	btrfs_init_free_cluster(&fs_info->data_alloc_cluster);
@@ -2784,23 +2742,135 @@ int open_ctree(struct super_block *sb,
 	fs_info->swapfile_pins = RB_ROOT;
 
 	fs_info->send_in_progress = 0;
+}
+
+static int init_mount_fs_info(struct btrfs_fs_info *fs_info, struct super_block *sb)
+{
+	int ret;
+
+	fs_info->sb = sb;
+	sb->s_blocksize = BTRFS_BDEV_BLOCKSIZE;
+	sb->s_blocksize_bits = blksize_bits(BTRFS_BDEV_BLOCKSIZE);
+
+	ret = percpu_counter_init(&fs_info->dio_bytes, 0, GFP_KERNEL);
+	if (ret)
+		return ret;
+
+	ret = percpu_counter_init(&fs_info->dirty_metadata_bytes, 0, GFP_KERNEL);
+	if (ret)
+		return ret;
+
+	fs_info->dirty_metadata_batch = PAGE_SIZE *
+					(1 + ilog2(nr_cpu_ids));
+
+	ret = percpu_counter_init(&fs_info->delalloc_bytes, 0, GFP_KERNEL);
+	if (ret)
+		return ret;
+
+	ret = percpu_counter_init(&fs_info->dev_replace.bio_counter, 0,
+			GFP_KERNEL);
+	if (ret)
+		return ret;
+
+	fs_info->delayed_root = kmalloc(sizeof(struct btrfs_delayed_root),
+					GFP_KERNEL);
+	if (!fs_info->delayed_root)
+		return -ENOMEM;
+	btrfs_init_delayed_root(fs_info->delayed_root);
+
+	return btrfs_alloc_stripe_hash_table(fs_info);
+}
+
+static int btrfs_uuid_rescan_kthread(void *data)
+{
+	struct btrfs_fs_info *fs_info = (struct btrfs_fs_info *)data;
+	int ret;
+
+	/*
+	 * 1st step is to iterate through the existing UUID tree and
+	 * to delete all entries that contain outdated data.
+	 * 2nd step is to add all missing entries to the UUID tree.
+	 */
+	ret = btrfs_uuid_tree_iterate(fs_info);
+	if (ret < 0) {
+		if (ret != -EINTR)
+			btrfs_warn(fs_info, "iterating uuid_tree failed %d",
+				   ret);
+		up(&fs_info->uuid_tree_rescan_sem);
+		return ret;
+	}
+	return btrfs_uuid_scan_kthread(data);
+}
 
-	ret = btrfs_alloc_stripe_hash_table(fs_info);
+static int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info)
+{
+	struct task_struct *task;
+
+	down(&fs_info->uuid_tree_rescan_sem);
+	task = kthread_run(btrfs_uuid_rescan_kthread, fs_info, "btrfs-uuid");
+	if (IS_ERR(task)) {
+		/* fs_info->update_uuid_tree_gen remains 0 in all error case */
+		btrfs_warn(fs_info, "failed to start uuid_rescan task");
+		up(&fs_info->uuid_tree_rescan_sem);
+		return PTR_ERR(task);
+	}
+
+	return 0;
+}
+
+int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_devices,
+		      char *options)
+{
+	u32 sectorsize;
+	u32 nodesize;
+	u32 stripesize;
+	u64 generation;
+	u64 features;
+	u16 csum_type;
+	struct btrfs_key location;
+	struct btrfs_super_block *disk_super;
+	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
+	struct btrfs_root *tree_root;
+	struct btrfs_root *chunk_root;
+	int ret;
+	int err = -EINVAL;
+	int clear_free_space_tree = 0;
+	int level;
+
+	ret = init_mount_fs_info(fs_info, sb);
 	if (ret) {
 		err = ret;
-		goto fail_alloc;
+		goto fail;
+	}
+
+	/* These need to be init'ed before we start creating inodes and such. */
+	tree_root = btrfs_alloc_root(fs_info, BTRFS_ROOT_TREE_OBJECTID,
+				     GFP_KERNEL);
+	fs_info->tree_root = tree_root;
+	chunk_root = btrfs_alloc_root(fs_info, BTRFS_CHUNK_TREE_OBJECTID,
+				      GFP_KERNEL);
+	fs_info->chunk_root = chunk_root;
+	if (!tree_root || !chunk_root) {
+		err = -ENOMEM;
+		goto fail;
 	}
 
-	__setup_root(tree_root, fs_info, BTRFS_ROOT_TREE_OBJECTID);
+	fs_info->btree_inode = new_inode(sb);
+	if (!fs_info->btree_inode) {
+		err = -ENOMEM;
+		goto fail;
+	}
+	mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS);
+	btrfs_init_btree_inode(fs_info);
 
 	invalidate_bdev(fs_devices->latest_bdev);
 
 	/*
 	 * Read super block and check the signature bytes only
 	 */
-	bh = btrfs_read_dev_super(fs_devices->latest_bdev);
-	if (IS_ERR(bh)) {
-		err = PTR_ERR(bh);
+	disk_super = btrfs_read_dev_super(fs_devices->latest_bdev);
+	if (IS_ERR(disk_super)) {
+		err = PTR_ERR(disk_super);
 		goto fail_alloc;
 	}
 
@@ -2808,18 +2878,19 @@ int open_ctree(struct super_block *sb,
 	 * Verify the type first, if that or the the checksum value are
 	 * corrupted, we'll find out
 	 */
-	csum_type = btrfs_super_csum_type((struct btrfs_super_block *)bh->b_data);
+	csum_type = btrfs_super_csum_type(disk_super);
 	if (!btrfs_supported_super_csum(csum_type)) {
 		btrfs_err(fs_info, "unsupported checksum algorithm: %u",
 			  csum_type);
 		err = -EINVAL;
-		brelse(bh);
+		btrfs_release_disk_super(disk_super);
 		goto fail_alloc;
 	}
 
 	ret = btrfs_init_csum_hash(fs_info, csum_type);
 	if (ret) {
 		err = ret;
+		btrfs_release_disk_super(disk_super);
 		goto fail_alloc;
 	}
 
@@ -2827,11 +2898,11 @@ int open_ctree(struct super_block *sb,
 	 * We want to check superblock checksum, the type is stored inside.
 	 * Pass the whole disk block of size BTRFS_SUPER_INFO_SIZE (4k).
 	 */
-	if (btrfs_check_super_csum(fs_info, bh->b_data)) {
+	if (btrfs_check_super_csum(fs_info, (u8 *)disk_super)) {
 		btrfs_err(fs_info, "superblock checksum mismatch");
 		err = -EINVAL;
-		brelse(bh);
-		goto fail_csum;
+		btrfs_release_disk_super(disk_super);
+		goto fail_alloc;
 	}
 
 	/*
@@ -2839,8 +2910,8 @@ int open_ctree(struct super_block *sb,
 	 * following bytes up to INFO_SIZE, the checksum is calculated from
 	 * the whole block of INFO_SIZE
 	 */
-	memcpy(fs_info->super_copy, bh->b_data, sizeof(*fs_info->super_copy));
-	brelse(bh);
+	memcpy(fs_info->super_copy, disk_super, sizeof(*fs_info->super_copy));
+	btrfs_release_disk_super(disk_super);
 
 	disk_super = fs_info->super_copy;
 
@@ -2868,24 +2939,17 @@ int open_ctree(struct super_block *sb,
 	if (ret) {
 		btrfs_err(fs_info, "superblock contains fatal errors");
 		err = -EINVAL;
-		goto fail_csum;
+		goto fail_alloc;
 	}
 
 	if (!btrfs_super_root(disk_super))
-		goto fail_csum;
+		goto fail_alloc;
 
 	/* check FS state, whether FS is broken. */
 	if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_ERROR)
 		set_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state);
 
 	/*
-	 * run through our array of backup supers and setup
-	 * our ring pointer to the oldest one
-	 */
-	generation = btrfs_super_generation(disk_super);
-	find_oldest_super_backup(fs_info, generation);
-
-	/*
 	 * In the long term, we'll store the compression type in the super
 	 * block, and it'll be used for per file compression control.
 	 */
@@ -2894,7 +2958,7 @@ int open_ctree(struct super_block *sb,
 	ret = btrfs_parse_options(fs_info, options, sb->s_flags);
 	if (ret) {
 		err = ret;
-		goto fail_csum;
+		goto fail_alloc;
 	}
 
 	features = btrfs_super_incompat_flags(disk_super) &
@@ -2904,7 +2968,7 @@ int open_ctree(struct super_block *sb,
 		    "cannot mount because of unsupported optional features (%llx)",
 		    features);
 		err = -EINVAL;
-		goto fail_csum;
+		goto fail_alloc;
 	}
 
 	features = btrfs_super_incompat_flags(disk_super);
@@ -2948,7 +3012,7 @@ int open_ctree(struct super_block *sb,
 		btrfs_err(fs_info,
 "unequal nodesize/sectorsize (%u != %u) are not allowed for mixed block groups",
 			nodesize, sectorsize);
-		goto fail_csum;
+		goto fail_alloc;
 	}
 
 	/*
@@ -2964,7 +3028,7 @@ int open_ctree(struct super_block *sb,
 	"cannot mount read-write because of unsupported optional features (%llx)",
 		       features);
 		err = -EINVAL;
-		goto fail_csum;
+		goto fail_alloc;
 	}
 
 	ret = btrfs_init_workqueues(fs_info, fs_devices);
@@ -2995,8 +3059,6 @@ int open_ctree(struct super_block *sb,
 	generation = btrfs_super_chunk_root_generation(disk_super);
 	level = btrfs_super_chunk_root_level(disk_super);
 
-	__setup_root(chunk_root, fs_info, BTRFS_CHUNK_TREE_OBJECTID);
-
 	chunk_root->node = read_tree_block(fs_info,
 					   btrfs_super_chunk_root(disk_super),
 					   generation, level, NULL);
@@ -3012,7 +3074,8 @@ int open_ctree(struct super_block *sb,
 	chunk_root->commit_root = btrfs_root_node(chunk_root);
 
 	read_extent_buffer(chunk_root->node, fs_info->chunk_tree_uuid,
-	   btrfs_header_chunk_tree_uuid(chunk_root->node), BTRFS_UUID_SIZE);
+			   offsetof(struct btrfs_header, chunk_tree_uuid),
+			   BTRFS_UUID_SIZE);
 
 	ret = btrfs_read_chunk_tree(fs_info);
 	if (ret) {
@@ -3031,44 +3094,21 @@ int open_ctree(struct super_block *sb,
 		goto fail_tree_roots;
 	}
 
-retry_root_backup:
-	generation = btrfs_super_generation(disk_super);
-	level = btrfs_super_root_level(disk_super);
-
-	tree_root->node = read_tree_block(fs_info,
-					  btrfs_super_root(disk_super),
-					  generation, level, NULL);
-	if (IS_ERR(tree_root->node) ||
-	    !extent_buffer_uptodate(tree_root->node)) {
-		btrfs_warn(fs_info, "failed to read tree root");
-		if (!IS_ERR(tree_root->node))
-			free_extent_buffer(tree_root->node);
-		tree_root->node = NULL;
-		goto recovery_tree_root;
-	}
-
-	btrfs_set_root_node(&tree_root->root_item, tree_root->node);
-	tree_root->commit_root = btrfs_root_node(tree_root);
-	btrfs_set_root_refs(&tree_root->root_item, 1);
-
-	mutex_lock(&tree_root->objectid_mutex);
-	ret = btrfs_find_highest_objectid(tree_root,
-					&tree_root->highest_objectid);
-	if (ret) {
-		mutex_unlock(&tree_root->objectid_mutex);
-		goto recovery_tree_root;
-	}
-
-	ASSERT(tree_root->highest_objectid <= BTRFS_LAST_FREE_OBJECTID);
-
-	mutex_unlock(&tree_root->objectid_mutex);
-
-	ret = btrfs_read_roots(fs_info);
+	ret = init_tree_roots(fs_info);
 	if (ret)
-		goto recovery_tree_root;
+		goto fail_tree_roots;
 
-	fs_info->generation = generation;
-	fs_info->last_trans_committed = generation;
+	/*
+	 * If we have a uuid root and we're not being told to rescan we need to
+	 * check the generation here so we can set the
+	 * BTRFS_FS_UPDATE_UUID_TREE_GEN bit.  Otherwise we could commit the
+	 * transaction during a balance or the log replay without updating the
+	 * uuid generation, and then if we crash we would rescan the uuid tree,
+	 * even though it was perfectly fine.
+	 */
+	if (fs_info->uuid_root && !btrfs_test_opt(fs_info, RESCAN_UUID_TREE) &&
+	    fs_info->generation == btrfs_super_uuid_tree_generation(disk_super))
+		set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags);
 
 	ret = btrfs_verify_dev_extents(fs_info);
 	if (ret) {
@@ -3097,20 +3137,13 @@ retry_root_backup:
 
 	btrfs_free_extra_devids(fs_devices, 1);
 
-	ret = btrfs_sysfs_add_fsid(fs_devices, NULL);
+	ret = btrfs_sysfs_add_fsid(fs_devices);
 	if (ret) {
 		btrfs_err(fs_info, "failed to init sysfs fsid interface: %d",
 				ret);
 		goto fail_block_groups;
 	}
 
-	ret = btrfs_sysfs_add_device(fs_devices);
-	if (ret) {
-		btrfs_err(fs_info, "failed to init sysfs device interface: %d",
-				ret);
-		goto fail_fsdev_sysfs;
-	}
-
 	ret = btrfs_sysfs_add_mounted(fs_info);
 	if (ret) {
 		btrfs_err(fs_info, "failed to init sysfs interface: %d", ret);
@@ -3180,6 +3213,7 @@ retry_root_backup:
 	/* do not make disk changes in broken FS or nologreplay is given */
 	if (btrfs_super_log_root(disk_super) != 0 &&
 	    !btrfs_test_opt(fs_info, NOLOGREPLAY)) {
+		btrfs_info(fs_info, "start tree-log replay");
 		ret = btrfs_replay_log(fs_info, fs_devices);
 		if (ret) {
 			err = ret;
@@ -3211,10 +3245,11 @@ retry_root_backup:
 	location.type = BTRFS_ROOT_ITEM_KEY;
 	location.offset = 0;
 
-	fs_info->fs_root = btrfs_read_fs_root_no_name(fs_info, &location);
+	fs_info->fs_root = btrfs_get_fs_root(fs_info, &location, true);
 	if (IS_ERR(fs_info->fs_root)) {
 		err = PTR_ERR(fs_info->fs_root);
 		btrfs_warn(fs_info, "failed to read fs tree: %d", err);
+		fs_info->fs_root = NULL;
 		goto fail_qgroup;
 	}
 
@@ -3277,6 +3312,7 @@ retry_root_backup:
 	}
 
 	btrfs_qgroup_rescan_resume(fs_info);
+	btrfs_discard_resume(fs_info);
 
 	if (!fs_info->uuid_root) {
 		btrfs_info(fs_info, "creating UUID tree");
@@ -3298,8 +3334,6 @@ retry_root_backup:
 			close_ctree(fs_info);
 			return ret;
 		}
-	} else {
-		set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags);
 	}
 	set_bit(BTRFS_FS_OPEN, &fs_info->flags);
 
@@ -3336,114 +3370,84 @@ fail_block_groups:
 	btrfs_put_block_group_cache(fs_info);
 
 fail_tree_roots:
-	free_root_pointers(fs_info, 1);
+	free_root_pointers(fs_info, true);
 	invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 
 fail_sb_buffer:
 	btrfs_stop_all_workers(fs_info);
 	btrfs_free_block_groups(fs_info);
-fail_csum:
-	btrfs_free_csum_hash(fs_info);
 fail_alloc:
-fail_iput:
 	btrfs_mapping_tree_free(&fs_info->mapping_tree);
 
 	iput(fs_info->btree_inode);
-fail_bio_counter:
-	percpu_counter_destroy(&fs_info->dev_replace.bio_counter);
-fail_delalloc_bytes:
-	percpu_counter_destroy(&fs_info->delalloc_bytes);
-fail_dirty_metadata_bytes:
-	percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
-fail_dio_bytes:
-	percpu_counter_destroy(&fs_info->dio_bytes);
-fail_srcu:
-	cleanup_srcu_struct(&fs_info->subvol_srcu);
 fail:
-	btrfs_free_stripe_hash_table(fs_info);
 	btrfs_close_devices(fs_info->fs_devices);
 	return err;
-
-recovery_tree_root:
-	if (!btrfs_test_opt(fs_info, USEBACKUPROOT))
-		goto fail_tree_roots;
-
-	free_root_pointers(fs_info, 0);
-
-	/* don't use the log in recovery mode, it won't be valid */
-	btrfs_set_super_log_root(disk_super, 0);
-
-	/* we can't trust the free space cache either */
-	btrfs_set_opt(fs_info->mount_opt, CLEAR_CACHE);
-
-	ret = next_root_backup(fs_info, fs_info->super_copy,
-			       &num_backups_tried, &backup_index);
-	if (ret == -1)
-		goto fail_block_groups;
-	goto retry_root_backup;
 }
 ALLOW_ERROR_INJECTION(open_ctree, ERRNO);
 
-static void btrfs_end_buffer_write_sync(struct buffer_head *bh, int uptodate)
+static void btrfs_end_super_write(struct bio *bio)
 {
-	if (uptodate) {
-		set_buffer_uptodate(bh);
-	} else {
-		struct btrfs_device *device = (struct btrfs_device *)
-			bh->b_private;
-
-		btrfs_warn_rl_in_rcu(device->fs_info,
-				"lost page write due to IO error on %s",
-					  rcu_str_deref(device->name));
-		/* note, we don't set_buffer_write_io_error because we have
-		 * our own ways of dealing with the IO errors
-		 */
-		clear_buffer_uptodate(bh);
-		btrfs_dev_stat_inc_and_print(device, BTRFS_DEV_STAT_WRITE_ERRS);
+	struct btrfs_device *device = bio->bi_private;
+	struct bio_vec *bvec;
+	struct bvec_iter_all iter_all;
+	struct page *page;
+
+	bio_for_each_segment_all(bvec, bio, iter_all) {
+		page = bvec->bv_page;
+
+		if (bio->bi_status) {
+			btrfs_warn_rl_in_rcu(device->fs_info,
+				"lost page write due to IO error on %s (%d)",
+				rcu_str_deref(device->name),
+				blk_status_to_errno(bio->bi_status));
+			ClearPageUptodate(page);
+			SetPageError(page);
+			btrfs_dev_stat_inc_and_print(device,
+						     BTRFS_DEV_STAT_WRITE_ERRS);
+		} else {
+			SetPageUptodate(page);
+		}
+
+		put_page(page);
+		unlock_page(page);
 	}
-	unlock_buffer(bh);
-	put_bh(bh);
+
+	bio_put(bio);
 }
 
-int btrfs_read_dev_one_super(struct block_device *bdev, int copy_num,
-			struct buffer_head **bh_ret)
+struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev,
+						   int copy_num)
 {
-	struct buffer_head *bh;
 	struct btrfs_super_block *super;
+	struct page *page;
 	u64 bytenr;
+	struct address_space *mapping = bdev->bd_inode->i_mapping;
 
 	bytenr = btrfs_sb_offset(copy_num);
 	if (bytenr + BTRFS_SUPER_INFO_SIZE >= i_size_read(bdev->bd_inode))
-		return -EINVAL;
+		return ERR_PTR(-EINVAL);
 
-	bh = __bread(bdev, bytenr / BTRFS_BDEV_BLOCKSIZE, BTRFS_SUPER_INFO_SIZE);
-	/*
-	 * If we fail to read from the underlying devices, as of now
-	 * the best option we have is to mark it EIO.
-	 */
-	if (!bh)
-		return -EIO;
+	page = read_cache_page_gfp(mapping, bytenr >> PAGE_SHIFT, GFP_NOFS);
+	if (IS_ERR(page))
+		return ERR_CAST(page);
 
-	super = (struct btrfs_super_block *)bh->b_data;
+	super = page_address(page);
 	if (btrfs_super_bytenr(super) != bytenr ||
 		    btrfs_super_magic(super) != BTRFS_MAGIC) {
-		brelse(bh);
-		return -EINVAL;
+		btrfs_release_disk_super(super);
+		return ERR_PTR(-EINVAL);
 	}
 
-	*bh_ret = bh;
-	return 0;
+	return super;
 }
 
 
-struct buffer_head *btrfs_read_dev_super(struct block_device *bdev)
+struct btrfs_super_block *btrfs_read_dev_super(struct block_device *bdev)
 {
-	struct buffer_head *bh;
-	struct buffer_head *latest = NULL;
-	struct btrfs_super_block *super;
+	struct btrfs_super_block *super, *latest = NULL;
 	int i;
 	u64 transid = 0;
-	int ret = -EINVAL;
 
 	/* we would like to check all the supers, but that would make
 	 * a btrfs mount succeed after a mkfs from a different FS.
@@ -3451,48 +3455,41 @@ struct buffer_head *btrfs_read_dev_super(struct block_device *bdev)
 	 * later supers, using BTRFS_SUPER_MIRROR_MAX instead
 	 */
 	for (i = 0; i < 1; i++) {
-		ret = btrfs_read_dev_one_super(bdev, i, &bh);
-		if (ret)
+		super = btrfs_read_dev_one_super(bdev, i);
+		if (IS_ERR(super))
 			continue;
 
-		super = (struct btrfs_super_block *)bh->b_data;
-
 		if (!latest || btrfs_super_generation(super) > transid) {
-			brelse(latest);
-			latest = bh;
+			if (latest)
+				btrfs_release_disk_super(super);
+
+			latest = super;
 			transid = btrfs_super_generation(super);
-		} else {
-			brelse(bh);
 		}
 	}
 
-	if (!latest)
-		return ERR_PTR(ret);
-
-	return latest;
+	return super;
 }
 
 /*
  * Write superblock @sb to the @device. Do not wait for completion, all the
- * buffer heads we write are pinned.
+ * pages we use for writing are locked.
  *
  * Write @max_mirrors copies of the superblock, where 0 means default that fit
  * the expected device size at commit time. Note that max_mirrors must be
  * same for write and wait phases.
  *
- * Return number of errors when buffer head is not found or submission fails.
+ * Return number of errors when page is not found or submission fails.
  */
 static int write_dev_supers(struct btrfs_device *device,
 			    struct btrfs_super_block *sb, int max_mirrors)
 {
 	struct btrfs_fs_info *fs_info = device->fs_info;
+	struct address_space *mapping = device->bdev->bd_inode->i_mapping;
 	SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
-	struct buffer_head *bh;
 	int i;
-	int ret;
 	int errors = 0;
 	u64 bytenr;
-	int op_flags;
 
 	if (max_mirrors == 0)
 		max_mirrors = BTRFS_SUPER_MIRROR_MAX;
@@ -3500,6 +3497,10 @@ static int write_dev_supers(struct btrfs_device *device,
 	shash->tfm = fs_info->csum_shash;
 
 	for (i = 0; i < max_mirrors; i++) {
+		struct page *page;
+		struct bio *bio;
+		struct btrfs_super_block *disk_super;
+
 		bytenr = btrfs_sb_offset(i);
 		if (bytenr + BTRFS_SUPER_INFO_SIZE >=
 		    device->commit_total_bytes)
@@ -3512,37 +3513,45 @@ static int write_dev_supers(struct btrfs_device *device,
 				    BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE);
 		crypto_shash_final(shash, sb->csum);
 
-		/* One reference for us, and we leave it for the caller */
-		bh = __getblk(device->bdev, bytenr / BTRFS_BDEV_BLOCKSIZE,
-			      BTRFS_SUPER_INFO_SIZE);
-		if (!bh) {
+		page = find_or_create_page(mapping, bytenr >> PAGE_SHIFT,
+					   GFP_NOFS);
+		if (!page) {
 			btrfs_err(device->fs_info,
-			    "couldn't get super buffer head for bytenr %llu",
+			    "couldn't get super block page for bytenr %llu",
 			    bytenr);
 			errors++;
 			continue;
 		}
 
-		memcpy(bh->b_data, sb, BTRFS_SUPER_INFO_SIZE);
+		/* Bump the refcount for wait_dev_supers() */
+		get_page(page);
 
-		/* one reference for submit_bh */
-		get_bh(bh);
+		disk_super = page_address(page);
+		memcpy(disk_super, sb, BTRFS_SUPER_INFO_SIZE);
 
-		set_buffer_uptodate(bh);
-		lock_buffer(bh);
-		bh->b_end_io = btrfs_end_buffer_write_sync;
-		bh->b_private = device;
+		/*
+		 * Directly use bios here instead of relying on the page cache
+		 * to do I/O, so we don't lose the ability to do integrity
+		 * checking.
+		 */
+		bio = bio_alloc(GFP_NOFS, 1);
+		bio_set_dev(bio, device->bdev);
+		bio->bi_iter.bi_sector = bytenr >> SECTOR_SHIFT;
+		bio->bi_private = device;
+		bio->bi_end_io = btrfs_end_super_write;
+		__bio_add_page(bio, page, BTRFS_SUPER_INFO_SIZE,
+			       offset_in_page(bytenr));
 
 		/*
-		 * we fua the first super.  The others we allow
-		 * to go down lazy.
+		 * We FUA only the first super block.  The others we allow to
+		 * go down lazy and there's a short window where the on-disk
+		 * copies might still contain the older version.
 		 */
-		op_flags = REQ_SYNC | REQ_META | REQ_PRIO;
+		bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_META | REQ_PRIO;
 		if (i == 0 && !btrfs_test_opt(device->fs_info, NOBARRIER))
-			op_flags |= REQ_FUA;
-		ret = btrfsic_submit_bh(REQ_OP_WRITE, op_flags, bh);
-		if (ret)
-			errors++;
+			bio->bi_opf |= REQ_FUA;
+
+		btrfsic_submit_bio(bio);
 	}
 	return errors < i ? 0 : -1;
 }
@@ -3551,12 +3560,11 @@ static int write_dev_supers(struct btrfs_device *device,
  * Wait for write completion of superblocks done by write_dev_supers,
  * @max_mirrors same for write and wait phases.
  *
- * Return number of errors when buffer head is not found or not marked up to
+ * Return number of errors when page is not found or not marked up to
  * date.
  */
 static int wait_dev_supers(struct btrfs_device *device, int max_mirrors)
 {
-	struct buffer_head *bh;
 	int i;
 	int errors = 0;
 	bool primary_failed = false;
@@ -3566,32 +3574,34 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors)
 		max_mirrors = BTRFS_SUPER_MIRROR_MAX;
 
 	for (i = 0; i < max_mirrors; i++) {
+		struct page *page;
+
 		bytenr = btrfs_sb_offset(i);
 		if (bytenr + BTRFS_SUPER_INFO_SIZE >=
 		    device->commit_total_bytes)
 			break;
 
-		bh = __find_get_block(device->bdev,
-				      bytenr / BTRFS_BDEV_BLOCKSIZE,
-				      BTRFS_SUPER_INFO_SIZE);
-		if (!bh) {
+		page = find_get_page(device->bdev->bd_inode->i_mapping,
+				     bytenr >> PAGE_SHIFT);
+		if (!page) {
 			errors++;
 			if (i == 0)
 				primary_failed = true;
 			continue;
 		}
-		wait_on_buffer(bh);
-		if (!buffer_uptodate(bh)) {
+		/* Page is submitted locked and unlocked once the IO completes */
+		wait_on_page_locked(page);
+		if (PageError(page)) {
 			errors++;
 			if (i == 0)
 				primary_failed = true;
 		}
 
-		/* drop our reference */
-		brelse(bh);
+		/* Drop our reference */
+		put_page(page);
 
-		/* drop the reference from the writing run */
-		brelse(bh);
+		/* Drop the reference from the writing run */
+		put_page(page);
 	}
 
 	/* log error, force error return */
@@ -3863,20 +3873,19 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
 void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 				  struct btrfs_root *root)
 {
+	bool drop_ref = false;
+
 	spin_lock(&fs_info->fs_roots_radix_lock);
 	radix_tree_delete(&fs_info->fs_roots_radix,
 			  (unsigned long)root->root_key.objectid);
+	if (test_and_clear_bit(BTRFS_ROOT_IN_RADIX, &root->state))
+		drop_ref = true;
 	spin_unlock(&fs_info->fs_roots_radix_lock);
 
-	if (btrfs_root_refs(&root->root_item) == 0)
-		synchronize_srcu(&fs_info->subvol_srcu);
-
 	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
-		btrfs_free_log(NULL, root);
+		ASSERT(root->log_root == NULL);
 		if (root->reloc_root) {
-			free_extent_buffer(root->reloc_root->node);
-			free_extent_buffer(root->reloc_root->commit_root);
-			btrfs_put_fs_root(root->reloc_root);
+			btrfs_put_root(root->reloc_root);
 			root->reloc_root = NULL;
 		}
 	}
@@ -3885,22 +3894,12 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 		__btrfs_remove_free_space_cache(root->free_ino_pinned);
 	if (root->free_ino_ctl)
 		__btrfs_remove_free_space_cache(root->free_ino_ctl);
-	btrfs_free_fs_root(root);
-}
-
-void btrfs_free_fs_root(struct btrfs_root *root)
-{
-	iput(root->ino_cache_inode);
-	WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
-	if (root->anon_dev)
-		free_anon_bdev(root->anon_dev);
-	if (root->subv_writers)
-		btrfs_free_subvolume_writers(root->subv_writers);
-	free_extent_buffer(root->node);
-	free_extent_buffer(root->commit_root);
-	kfree(root->free_ino_ctl);
-	kfree(root->free_ino_pinned);
-	btrfs_put_fs_root(root);
+	if (root->ino_cache_inode) {
+		iput(root->ino_cache_inode);
+		root->ino_cache_inode = NULL;
+	}
+	if (drop_ref)
+		btrfs_put_root(root);
 }
 
 int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
@@ -3910,15 +3909,14 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
 	int i = 0;
 	int err = 0;
 	unsigned int ret = 0;
-	int index;
 
 	while (1) {
-		index = srcu_read_lock(&fs_info->subvol_srcu);
+		spin_lock(&fs_info->fs_roots_radix_lock);
 		ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
 					     (void **)gang, root_objectid,
 					     ARRAY_SIZE(gang));
 		if (!ret) {
-			srcu_read_unlock(&fs_info->subvol_srcu, index);
+			spin_unlock(&fs_info->fs_roots_radix_lock);
 			break;
 		}
 		root_objectid = gang[ret - 1]->root_key.objectid + 1;
@@ -3930,9 +3928,9 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
 				continue;
 			}
 			/* grab all the search result for later use */
-			gang[i] = btrfs_grab_fs_root(gang[i]);
+			gang[i] = btrfs_grab_root(gang[i]);
 		}
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
+		spin_unlock(&fs_info->fs_roots_radix_lock);
 
 		for (i = 0; i < ret; i++) {
 			if (!gang[i])
@@ -3941,7 +3939,7 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
 			err = btrfs_orphan_cleanup(gang[i]);
 			if (err)
 				break;
-			btrfs_put_fs_root(gang[i]);
+			btrfs_put_root(gang[i]);
 		}
 		root_objectid++;
 	}
@@ -3949,7 +3947,7 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
 	/* release the uncleaned roots due to error */
 	for (; i < ret; i++) {
 		if (gang[i])
-			btrfs_put_fs_root(gang[i]);
+			btrfs_put_root(gang[i]);
 	}
 	return err;
 }
@@ -3974,7 +3972,7 @@ int btrfs_commit_super(struct btrfs_fs_info *fs_info)
 	return btrfs_commit_transaction(trans);
 }
 
-void close_ctree(struct btrfs_fs_info *fs_info)
+void __cold close_ctree(struct btrfs_fs_info *fs_info)
 {
 	int ret;
 
@@ -4011,6 +4009,9 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 
 	cancel_work_sync(&fs_info->async_reclaim_work);
 
+	/* Cancel or finish ongoing discard work */
+	btrfs_discard_cleanup(fs_info);
+
 	if (!sb_rdonly(fs_info->sb)) {
 		/*
 		 * The cleaner kthread is stopped, so do one final pass over
@@ -4018,6 +4019,19 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 		 */
 		btrfs_delete_unused_bgs(fs_info);
 
+		/*
+		 * There might be existing delayed inode workers still running
+		 * and holding an empty delayed inode item. We must wait for
+		 * them to complete first because they can create a transaction.
+		 * This happens when someone calls btrfs_balance_delayed_items()
+		 * and then a transaction commit runs the same delayed nodes
+		 * before any delayed worker has done something with the nodes.
+		 * We must wait for any worker here and not at transaction
+		 * commit time since that could cause a deadlock.
+		 * This is a very rare case.
+		 */
+		btrfs_flush_workqueue(fs_info->delayed_workers);
+
 		ret = btrfs_commit_super(fs_info);
 		if (ret)
 			btrfs_err(fs_info, "commit super ret %d", ret);
@@ -4048,8 +4062,6 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 	btrfs_sysfs_remove_mounted(fs_info);
 	btrfs_sysfs_remove_fsid(fs_info->fs_devices);
 
-	btrfs_free_fs_roots(fs_info);
-
 	btrfs_put_block_group_cache(fs_info);
 
 	/*
@@ -4059,10 +4071,18 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 	invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 	btrfs_stop_all_workers(fs_info);
 
-	btrfs_free_block_groups(fs_info);
-
 	clear_bit(BTRFS_FS_OPEN, &fs_info->flags);
-	free_root_pointers(fs_info, 1);
+	free_root_pointers(fs_info, true);
+	btrfs_free_fs_roots(fs_info);
+
+	/*
+	 * We must free the block groups after dropping the fs_roots as we could
+	 * have had an IO error and have left over tree log blocks that aren't
+	 * cleaned up until the fs roots are freed.  This makes the block group
+	 * accounting appear to be wrong because there's pending reserved bytes,
+	 * so make sure we do the block group cleanup afterwards.
+	 */
+	btrfs_free_block_groups(fs_info);
 
 	iput(fs_info->btree_inode);
 
@@ -4073,16 +4093,6 @@ void close_ctree(struct btrfs_fs_info *fs_info)
 
 	btrfs_mapping_tree_free(&fs_info->mapping_tree);
 	btrfs_close_devices(fs_info->fs_devices);
-
-	percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
-	percpu_counter_destroy(&fs_info->delalloc_bytes);
-	percpu_counter_destroy(&fs_info->dio_bytes);
-	percpu_counter_destroy(&fs_info->dev_replace.bio_counter);
-	cleanup_srcu_struct(&fs_info->subvol_srcu);
-
-	btrfs_free_csum_hash(fs_info);
-	btrfs_free_stripe_hash_table(fs_info);
-	btrfs_free_ref_cache(fs_info);
 }
 
 int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid,
@@ -4196,6 +4206,36 @@ static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info)
 	up_write(&fs_info->cleanup_work_sem);
 }
 
+static void btrfs_drop_all_logs(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_root *gang[8];
+	u64 root_objectid = 0;
+	int ret;
+
+	spin_lock(&fs_info->fs_roots_radix_lock);
+	while ((ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
+					     (void **)gang, root_objectid,
+					     ARRAY_SIZE(gang))) != 0) {
+		int i;
+
+		for (i = 0; i < ret; i++)
+			gang[i] = btrfs_grab_root(gang[i]);
+		spin_unlock(&fs_info->fs_roots_radix_lock);
+
+		for (i = 0; i < ret; i++) {
+			if (!gang[i])
+				continue;
+			root_objectid = gang[i]->root_key.objectid;
+			btrfs_free_log(NULL, gang[i]);
+			btrfs_put_root(gang[i]);
+		}
+		root_objectid++;
+		spin_lock(&fs_info->fs_roots_radix_lock);
+	}
+	spin_unlock(&fs_info->fs_roots_radix_lock);
+	btrfs_free_log_root_tree(NULL, fs_info);
+}
+
 static void btrfs_destroy_ordered_extents(struct btrfs_root *root)
 {
 	struct btrfs_ordered_extent *ordered;
@@ -4256,7 +4296,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 	spin_lock(&delayed_refs->lock);
 	if (atomic_read(&delayed_refs->num_entries) == 0) {
 		spin_unlock(&delayed_refs->lock);
-		btrfs_info(fs_info, "delayed_refs has NO entry");
+		btrfs_debug(fs_info, "delayed_refs has NO entry");
 		return ret;
 	}
 
@@ -4290,14 +4330,36 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 		spin_unlock(&delayed_refs->lock);
 		mutex_unlock(&head->mutex);
 
-		if (pin_bytes)
-			btrfs_pin_extent(fs_info, head->bytenr,
-					 head->num_bytes, 1);
+		if (pin_bytes) {
+			struct btrfs_block_group *cache;
+
+			cache = btrfs_lookup_block_group(fs_info, head->bytenr);
+			BUG_ON(!cache);
+
+			spin_lock(&cache->space_info->lock);
+			spin_lock(&cache->lock);
+			cache->pinned += head->num_bytes;
+			btrfs_space_info_update_bytes_pinned(fs_info,
+				cache->space_info, head->num_bytes);
+			cache->reserved -= head->num_bytes;
+			cache->space_info->bytes_reserved -= head->num_bytes;
+			spin_unlock(&cache->lock);
+			spin_unlock(&cache->space_info->lock);
+			percpu_counter_add_batch(
+				&cache->space_info->total_bytes_pinned,
+				head->num_bytes, BTRFS_TOTAL_BYTES_PINNED_BATCH);
+
+			btrfs_put_block_group(cache);
+
+			btrfs_error_unpin_extent_range(fs_info, head->bytenr,
+				head->bytenr + head->num_bytes - 1);
+		}
 		btrfs_cleanup_ref_head_accounting(fs_info, delayed_refs, head);
 		btrfs_put_delayed_ref_head(head);
 		cond_resched();
 		spin_lock(&delayed_refs->lock);
 	}
+	btrfs_qgroup_destroy_extent_records(trans);
 
 	spin_unlock(&delayed_refs->lock);
 
@@ -4347,12 +4409,12 @@ static void btrfs_destroy_all_delalloc_inodes(struct btrfs_fs_info *fs_info)
 	while (!list_empty(&splice)) {
 		root = list_first_entry(&splice, struct btrfs_root,
 					 delalloc_root);
-		root = btrfs_grab_fs_root(root);
+		root = btrfs_grab_root(root);
 		BUG_ON(!root);
 		spin_unlock(&fs_info->delalloc_root_lock);
 
 		btrfs_destroy_delalloc_inodes(root);
-		btrfs_put_fs_root(root);
+		btrfs_put_root(root);
 
 		spin_lock(&fs_info->delalloc_root_lock);
 	}
@@ -4393,16 +4455,12 @@ static int btrfs_destroy_marked_extents(struct btrfs_fs_info *fs_info,
 }
 
 static int btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info,
-				       struct extent_io_tree *pinned_extents)
+				       struct extent_io_tree *unpin)
 {
-	struct extent_io_tree *unpin;
 	u64 start;
 	u64 end;
 	int ret;
-	bool loop = true;
 
-	unpin = pinned_extents;
-again:
 	while (1) {
 		struct extent_state *cached_state = NULL;
 
@@ -4427,19 +4485,10 @@ again:
 		cond_resched();
 	}
 
-	if (loop) {
-		if (unpin == &fs_info->freed_extents[0])
-			unpin = &fs_info->freed_extents[1];
-		else
-			unpin = &fs_info->freed_extents[0];
-		loop = false;
-		goto again;
-	}
-
 	return 0;
 }
 
-static void btrfs_cleanup_bg_io(struct btrfs_block_group_cache *cache)
+static void btrfs_cleanup_bg_io(struct btrfs_block_group *cache)
 {
 	struct inode *inode;
 
@@ -4456,12 +4505,12 @@ static void btrfs_cleanup_bg_io(struct btrfs_block_group_cache *cache)
 void btrfs_cleanup_dirty_bgs(struct btrfs_transaction *cur_trans,
 			     struct btrfs_fs_info *fs_info)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 
 	spin_lock(&cur_trans->dirty_bgs_lock);
 	while (!list_empty(&cur_trans->dirty_bgs)) {
 		cache = list_first_entry(&cur_trans->dirty_bgs,
-					 struct btrfs_block_group_cache,
+					 struct btrfs_block_group,
 					 dirty_list);
 
 		if (!list_empty(&cache->io_list)) {
@@ -4489,7 +4538,7 @@ void btrfs_cleanup_dirty_bgs(struct btrfs_transaction *cur_trans,
 	 */
 	while (!list_empty(&cur_trans->io_bgs)) {
 		cache = list_first_entry(&cur_trans->io_bgs,
-					 struct btrfs_block_group_cache,
+					 struct btrfs_block_group,
 					 io_list);
 
 		list_del_init(&cache->io_list);
@@ -4523,12 +4572,10 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
 	wake_up(&fs_info->transaction_wait);
 
 	btrfs_destroy_delayed_inodes(fs_info);
-	btrfs_assert_delayed_root_empty(fs_info);
 
 	btrfs_destroy_marked_extents(fs_info, &cur_trans->dirty_pages,
 				     EXTENT_DIRTY);
-	btrfs_destroy_pinned_extent(fs_info,
-				    fs_info->pinned_extents);
+	btrfs_destroy_pinned_extent(fs_info, &cur_trans->pinned_extents);
 
 	cur_trans->state =TRANS_STATE_COMPLETED;
 	wake_up(&cur_trans->commit_wait);
@@ -4580,8 +4627,8 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info)
 	btrfs_destroy_all_ordered_extents(fs_info);
 	btrfs_destroy_delayed_inodes(fs_info);
 	btrfs_assert_delayed_root_empty(fs_info);
-	btrfs_destroy_pinned_extent(fs_info, fs_info->pinned_extents);
 	btrfs_destroy_all_delalloc_inodes(fs_info);
+	btrfs_drop_all_logs(fs_info);
 	mutex_unlock(&fs_info->transaction_kthread_mutex);
 
 	return 0;
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index a6958103d87e..cd629113f61c 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -39,6 +39,8 @@ static inline u64 btrfs_sb_offset(int mirror)
 struct btrfs_device;
 struct btrfs_fs_devices;
 
+void btrfs_check_leaked_roots(struct btrfs_fs_info *fs_info);
+void btrfs_init_fs_info(struct btrfs_fs_info *fs_info);
 int btrfs_verify_level_key(struct extent_buffer *eb, int level,
 			   struct btrfs_key *first_key, u64 parent_transid);
 struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr,
@@ -49,20 +51,17 @@ struct extent_buffer *btrfs_find_create_tree_block(
 						struct btrfs_fs_info *fs_info,
 						u64 bytenr);
 void btrfs_clean_tree_block(struct extent_buffer *buf);
-int open_ctree(struct super_block *sb,
+int __cold open_ctree(struct super_block *sb,
 	       struct btrfs_fs_devices *fs_devices,
 	       char *options);
-void close_ctree(struct btrfs_fs_info *fs_info);
+void __cold close_ctree(struct btrfs_fs_info *fs_info);
 int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors);
-struct buffer_head *btrfs_read_dev_super(struct block_device *bdev);
-int btrfs_read_dev_one_super(struct block_device *bdev, int copy_num,
-			struct buffer_head **bh_ret);
+struct btrfs_super_block *btrfs_read_dev_super(struct block_device *bdev);
+struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev,
+						   int copy_num);
 int btrfs_commit_super(struct btrfs_fs_info *fs_info);
-struct btrfs_root *btrfs_read_fs_root(struct btrfs_root *tree_root,
-				      struct btrfs_key *location);
-int btrfs_init_fs_root(struct btrfs_root *root);
-struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
-					u64 root_id);
+struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root,
+					struct btrfs_key *key);
 int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info,
 			 struct btrfs_root *root);
 void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info);
@@ -70,19 +69,13 @@ void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info);
 struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info,
 				     struct btrfs_key *key,
 				     bool check_ref);
-static inline struct btrfs_root *
-btrfs_read_fs_root_no_name(struct btrfs_fs_info *fs_info,
-			   struct btrfs_key *location)
-{
-	return btrfs_get_fs_root(fs_info, location, true);
-}
 
+void btrfs_free_fs_info(struct btrfs_fs_info *fs_info);
 int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info);
 void btrfs_btree_balance_dirty(struct btrfs_fs_info *fs_info);
 void btrfs_btree_balance_dirty_nodelay(struct btrfs_fs_info *fs_info);
 void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 				 struct btrfs_root *root);
-void btrfs_free_fs_root(struct btrfs_root *root);
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info);
@@ -95,19 +88,16 @@ struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info);
  * If you want to ensure the whole tree is safe, you should use
  * 	fs_info->subvol_srcu
  */
-static inline struct btrfs_root *btrfs_grab_fs_root(struct btrfs_root *root)
+static inline struct btrfs_root *btrfs_grab_root(struct btrfs_root *root)
 {
+	if (!root)
+		return NULL;
 	if (refcount_inc_not_zero(&root->refs))
 		return root;
 	return NULL;
 }
 
-static inline void btrfs_put_fs_root(struct btrfs_root *root)
-{
-	if (refcount_dec_and_test(&root->refs))
-		kfree(root);
-}
-
+void btrfs_put_root(struct btrfs_root *root);
 void btrfs_mark_buffer_dirty(struct extent_buffer *buf);
 int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid,
 			  int atomic);
@@ -134,8 +124,8 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 int btree_lock_page_hook(struct page *page, void *data,
 				void (*flush_fn)(void *));
 struct extent_map *btree_get_extent(struct btrfs_inode *inode,
-		struct page *page, size_t pg_offset, u64 start, u64 len,
-		int create);
+				    struct page *page, size_t pg_offset,
+				    u64 start, u64 len);
 int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags);
 int __init btrfs_end_io_wq_init(void);
 void __cold btrfs_end_io_wq_exit(void);
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index ddf28ecf17f9..2bb25d2dc44b 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -57,16 +57,14 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len,
 	return type;
 }
 
-static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
-				       u64 root_objectid, u32 generation,
-				       int check_generation)
+struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
+				u64 root_objectid, u32 generation,
+				int check_generation)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
 	struct btrfs_root *root;
 	struct inode *inode;
 	struct btrfs_key key;
-	int index;
-	int err = 0;
 
 	if (objectid < BTRFS_FIRST_FREE_OBJECTID)
 		return ERR_PTR(-ESTALE);
@@ -75,25 +73,18 @@ static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
 	key.type = BTRFS_ROOT_ITEM_KEY;
 	key.offset = (u64)-1;
 
-	index = srcu_read_lock(&fs_info->subvol_srcu);
-
-	root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(root)) {
-		err = PTR_ERR(root);
-		goto fail;
-	}
+	root = btrfs_get_fs_root(fs_info, &key, true);
+	if (IS_ERR(root))
+		return ERR_CAST(root);
 
 	key.objectid = objectid;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 
-	inode = btrfs_iget(sb, &key, root, NULL);
-	if (IS_ERR(inode)) {
-		err = PTR_ERR(inode);
-		goto fail;
-	}
-
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
+	inode = btrfs_iget(sb, &key, root);
+	btrfs_put_root(root);
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
 
 	if (check_generation && generation != inode->i_generation) {
 		iput(inode);
@@ -101,9 +92,6 @@ static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
 	}
 
 	return d_obtain_alias(inode);
-fail:
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
-	return ERR_PTR(err);
 }
 
 static struct dentry *btrfs_fh_to_parent(struct super_block *sb, struct fid *fh,
@@ -152,7 +140,7 @@ static struct dentry *btrfs_fh_to_dentry(struct super_block *sb, struct fid *fh,
 	return btrfs_get_dentry(sb, objectid, root_objectid, generation, 1);
 }
 
-static struct dentry *btrfs_get_parent(struct dentry *child)
+struct dentry *btrfs_get_parent(struct dentry *child)
 {
 	struct inode *dir = d_inode(child);
 	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
@@ -214,7 +202,7 @@ static struct dentry *btrfs_get_parent(struct dentry *child)
 
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	return d_obtain_alias(btrfs_iget(fs_info->sb, &key, root, NULL));
+	return d_obtain_alias(btrfs_iget(fs_info->sb, &key, root));
 fail:
 	btrfs_free_path(path);
 	return ERR_PTR(ret);
diff --git a/fs/btrfs/export.h b/fs/btrfs/export.h
index 57488ecd7d4e..f32f4113c976 100644
--- a/fs/btrfs/export.h
+++ b/fs/btrfs/export.h
@@ -18,4 +18,9 @@ struct btrfs_fid {
 	u64 parent_root_objectid;
 } __attribute__ ((packed));
 
+struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
+				u64 root_objectid, u32 generation,
+				int check_generation);
+struct dentry *btrfs_get_parent(struct dentry *child);
+
 #endif
diff --git a/fs/btrfs/extent-io-tree.h b/fs/btrfs/extent-io-tree.h
new file mode 100644
index 000000000000..b4a7bad3e82e
--- /dev/null
+++ b/fs/btrfs/extent-io-tree.h
@@ -0,0 +1,251 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef BTRFS_EXTENT_IO_TREE_H
+#define BTRFS_EXTENT_IO_TREE_H
+
+struct extent_changeset;
+struct io_failure_record;
+
+/* Bits for the extent state */
+#define EXTENT_DIRTY		(1U << 0)
+#define EXTENT_UPTODATE		(1U << 1)
+#define EXTENT_LOCKED		(1U << 2)
+#define EXTENT_NEW		(1U << 3)
+#define EXTENT_DELALLOC		(1U << 4)
+#define EXTENT_DEFRAG		(1U << 5)
+#define EXTENT_BOUNDARY		(1U << 6)
+#define EXTENT_NODATASUM	(1U << 7)
+#define EXTENT_CLEAR_META_RESV	(1U << 8)
+#define EXTENT_NEED_WAIT	(1U << 9)
+#define EXTENT_DAMAGED		(1U << 10)
+#define EXTENT_NORESERVE	(1U << 11)
+#define EXTENT_QGROUP_RESERVED	(1U << 12)
+#define EXTENT_CLEAR_DATA_RESV	(1U << 13)
+#define EXTENT_DELALLOC_NEW	(1U << 14)
+#define EXTENT_DO_ACCOUNTING    (EXTENT_CLEAR_META_RESV | \
+				 EXTENT_CLEAR_DATA_RESV)
+#define EXTENT_CTLBITS		(EXTENT_DO_ACCOUNTING)
+
+/*
+ * Redefined bits above which are used only in the device allocation tree,
+ * shouldn't be using EXTENT_LOCKED / EXTENT_BOUNDARY / EXTENT_CLEAR_META_RESV
+ * / EXTENT_CLEAR_DATA_RESV because they have special meaning to the bit
+ * manipulation functions
+ */
+#define CHUNK_ALLOCATED				EXTENT_DIRTY
+#define CHUNK_TRIMMED				EXTENT_DEFRAG
+
+enum {
+	IO_TREE_FS_PINNED_EXTENTS,
+	IO_TREE_FS_EXCLUDED_EXTENTS,
+	IO_TREE_INODE_IO,
+	IO_TREE_INODE_IO_FAILURE,
+	IO_TREE_RELOC_BLOCKS,
+	IO_TREE_TRANS_DIRTY_PAGES,
+	IO_TREE_ROOT_DIRTY_LOG_PAGES,
+	IO_TREE_INODE_FILE_EXTENT,
+	IO_TREE_SELFTEST,
+};
+
+struct extent_io_tree {
+	struct rb_root state;
+	struct btrfs_fs_info *fs_info;
+	void *private_data;
+	u64 dirty_bytes;
+	bool track_uptodate;
+
+	/* Who owns this io tree, should be one of IO_TREE_* */
+	u8 owner;
+
+	spinlock_t lock;
+	const struct extent_io_ops *ops;
+};
+
+struct extent_state {
+	u64 start;
+	u64 end; /* inclusive */
+	struct rb_node rb_node;
+
+	/* ADD NEW ELEMENTS AFTER THIS */
+	wait_queue_head_t wq;
+	refcount_t refs;
+	unsigned state;
+
+	struct io_failure_record *failrec;
+
+#ifdef CONFIG_BTRFS_DEBUG
+	struct list_head leak_list;
+#endif
+};
+
+int __init extent_state_cache_init(void);
+void __cold extent_state_cache_exit(void);
+
+void extent_io_tree_init(struct btrfs_fs_info *fs_info,
+			 struct extent_io_tree *tree, unsigned int owner,
+			 void *private_data);
+void extent_io_tree_release(struct extent_io_tree *tree);
+
+int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
+		     struct extent_state **cached);
+
+static inline int lock_extent(struct extent_io_tree *tree, u64 start, u64 end)
+{
+	return lock_extent_bits(tree, start, end, NULL);
+}
+
+int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
+
+int __init extent_io_init(void);
+void __cold extent_io_exit(void);
+
+u64 count_range_bits(struct extent_io_tree *tree,
+		     u64 *start, u64 search_end,
+		     u64 max_bytes, unsigned bits, int contig);
+
+void free_extent_state(struct extent_state *state);
+int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
+		   unsigned bits, int filled,
+		   struct extent_state *cached_state);
+int clear_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
+		unsigned bits, struct extent_changeset *changeset);
+int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
+		     unsigned bits, int wake, int delete,
+		     struct extent_state **cached);
+int __clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
+		     unsigned bits, int wake, int delete,
+		     struct extent_state **cached, gfp_t mask,
+		     struct extent_changeset *changeset);
+
+static inline int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end)
+{
+	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, NULL);
+}
+
+static inline int unlock_extent_cached(struct extent_io_tree *tree, u64 start,
+		u64 end, struct extent_state **cached)
+{
+	return __clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
+				GFP_NOFS, NULL);
+}
+
+static inline int unlock_extent_cached_atomic(struct extent_io_tree *tree,
+		u64 start, u64 end, struct extent_state **cached)
+{
+	return __clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
+				GFP_ATOMIC, NULL);
+}
+
+static inline int clear_extent_bits(struct extent_io_tree *tree, u64 start,
+		u64 end, unsigned bits)
+{
+	int wake = 0;
+
+	if (bits & EXTENT_LOCKED)
+		wake = 1;
+
+	return clear_extent_bit(tree, start, end, bits, wake, 0, NULL);
+}
+
+int set_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
+			   unsigned bits, struct extent_changeset *changeset);
+int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
+		   unsigned bits, u64 *failed_start,
+		   struct extent_state **cached_state, gfp_t mask);
+int set_extent_bits_nowait(struct extent_io_tree *tree, u64 start, u64 end,
+			   unsigned bits);
+
+static inline int set_extent_bits(struct extent_io_tree *tree, u64 start,
+		u64 end, unsigned bits)
+{
+	return set_extent_bit(tree, start, end, bits, NULL, NULL, GFP_NOFS);
+}
+
+static inline int clear_extent_uptodate(struct extent_io_tree *tree, u64 start,
+		u64 end, struct extent_state **cached_state)
+{
+	return __clear_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, 0,
+				cached_state, GFP_NOFS, NULL);
+}
+
+static inline int set_extent_dirty(struct extent_io_tree *tree, u64 start,
+		u64 end, gfp_t mask)
+{
+	return set_extent_bit(tree, start, end, EXTENT_DIRTY, NULL,
+			      NULL, mask);
+}
+
+static inline int clear_extent_dirty(struct extent_io_tree *tree, u64 start,
+				     u64 end, struct extent_state **cached)
+{
+	return clear_extent_bit(tree, start, end,
+				EXTENT_DIRTY | EXTENT_DELALLOC |
+				EXTENT_DO_ACCOUNTING, 0, 0, cached);
+}
+
+int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
+		       unsigned bits, unsigned clear_bits,
+		       struct extent_state **cached_state);
+
+static inline int set_extent_delalloc(struct extent_io_tree *tree, u64 start,
+				      u64 end, unsigned int extra_bits,
+				      struct extent_state **cached_state)
+{
+	return set_extent_bit(tree, start, end,
+			      EXTENT_DELALLOC | EXTENT_UPTODATE | extra_bits,
+			      NULL, cached_state, GFP_NOFS);
+}
+
+static inline int set_extent_defrag(struct extent_io_tree *tree, u64 start,
+		u64 end, struct extent_state **cached_state)
+{
+	return set_extent_bit(tree, start, end,
+			      EXTENT_DELALLOC | EXTENT_UPTODATE | EXTENT_DEFRAG,
+			      NULL, cached_state, GFP_NOFS);
+}
+
+static inline int set_extent_new(struct extent_io_tree *tree, u64 start,
+		u64 end)
+{
+	return set_extent_bit(tree, start, end, EXTENT_NEW, NULL, NULL,
+			GFP_NOFS);
+}
+
+static inline int set_extent_uptodate(struct extent_io_tree *tree, u64 start,
+		u64 end, struct extent_state **cached_state, gfp_t mask)
+{
+	return set_extent_bit(tree, start, end, EXTENT_UPTODATE, NULL,
+			      cached_state, mask);
+}
+
+int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
+			  u64 *start_ret, u64 *end_ret, unsigned bits,
+			  struct extent_state **cached_state);
+void find_first_clear_extent_bit(struct extent_io_tree *tree, u64 start,
+				 u64 *start_ret, u64 *end_ret, unsigned bits);
+int find_contiguous_extent_bit(struct extent_io_tree *tree, u64 start,
+			       u64 *start_ret, u64 *end_ret, unsigned bits);
+int extent_invalidatepage(struct extent_io_tree *tree,
+			  struct page *page, unsigned long offset);
+bool btrfs_find_delalloc_range(struct extent_io_tree *tree, u64 *start,
+			       u64 *end, u64 max_bytes,
+			       struct extent_state **cached_state);
+
+/* This should be reworked in the future and put elsewhere. */
+int get_state_failrec(struct extent_io_tree *tree, u64 start,
+		      struct io_failure_record **failrec);
+int set_state_failrec(struct extent_io_tree *tree, u64 start,
+		      struct io_failure_record *failrec);
+void btrfs_free_io_failure_record(struct btrfs_inode *inode, u64 start,
+		u64 end);
+int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
+				struct io_failure_record **failrec_ret);
+int free_io_failure(struct extent_io_tree *failure_tree,
+		    struct extent_io_tree *io_tree,
+		    struct io_failure_record *rec);
+int clean_io_failure(struct btrfs_fs_info *fs_info,
+		     struct extent_io_tree *failure_tree,
+		     struct extent_io_tree *io_tree, u64 start,
+		     struct page *page, u64 ino, unsigned int pg_offset);
+
+#endif /* BTRFS_EXTENT_IO_TREE_H */
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 49cb26fa7c63..54a64d1e18c6 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -32,6 +32,7 @@
 #include "block-rsv.h"
 #include "delalloc-space.h"
 #include "block-group.h"
+#include "discard.h"
 
 #undef SCRAMBLE_DELAYED_REFS
 
@@ -54,7 +55,7 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
 static int find_next_key(struct btrfs_path *path, int level,
 			 struct btrfs_key *key);
 
-static int block_group_bits(struct btrfs_block_group_cache *cache, u64 bits)
+static int block_group_bits(struct btrfs_block_group *cache, u64 bits)
 {
 	return (cache->flags & bits) == bits;
 }
@@ -63,25 +64,21 @@ int btrfs_add_excluded_extent(struct btrfs_fs_info *fs_info,
 			      u64 start, u64 num_bytes)
 {
 	u64 end = start + num_bytes - 1;
-	set_extent_bits(&fs_info->freed_extents[0],
-			start, end, EXTENT_UPTODATE);
-	set_extent_bits(&fs_info->freed_extents[1],
-			start, end, EXTENT_UPTODATE);
+	set_extent_bits(&fs_info->excluded_extents, start, end,
+			EXTENT_UPTODATE);
 	return 0;
 }
 
-void btrfs_free_excluded_extents(struct btrfs_block_group_cache *cache)
+void btrfs_free_excluded_extents(struct btrfs_block_group *cache)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	u64 start, end;
 
-	start = cache->key.objectid;
-	end = start + cache->key.offset - 1;
+	start = cache->start;
+	end = start + cache->length - 1;
 
-	clear_extent_bits(&fs_info->freed_extents[0],
-			  start, end, EXTENT_UPTODATE);
-	clear_extent_bits(&fs_info->freed_extents[1],
-			  start, end, EXTENT_UPTODATE);
+	clear_extent_bits(&fs_info->excluded_extents, start, end,
+			  EXTENT_UPTODATE);
 }
 
 static u64 generic_ref_to_space_flags(struct btrfs_ref *ref)
@@ -1192,24 +1189,6 @@ int insert_inline_extent_backref(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-static int insert_extent_backref(struct btrfs_trans_handle *trans,
-				 struct btrfs_path *path,
-				 u64 bytenr, u64 parent, u64 root_objectid,
-				 u64 owner, u64 offset, int refs_to_add)
-{
-	int ret;
-	if (owner < BTRFS_FIRST_FREE_OBJECTID) {
-		BUG_ON(refs_to_add != 1);
-		ret = insert_tree_block_ref(trans, path, bytenr, parent,
-					    root_objectid);
-	} else {
-		ret = insert_extent_data_ref(trans, path, bytenr, parent,
-					     root_objectid, owner, offset,
-					     refs_to_add);
-	}
-	return ret;
-}
-
 static int remove_extent_backref(struct btrfs_trans_handle *trans,
 				 struct btrfs_path *path,
 				 struct btrfs_extent_inline_ref *iref,
@@ -1306,8 +1285,10 @@ static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len,
 int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr,
 			 u64 num_bytes, u64 *actual_bytes)
 {
-	int ret;
+	int ret = 0;
 	u64 discarded_bytes = 0;
+	u64 end = bytenr + num_bytes;
+	u64 cur = bytenr;
 	struct btrfs_bio *bbio = NULL;
 
 
@@ -1316,15 +1297,23 @@ int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr,
 	 * associated to its stripes that don't go away while we are discarding.
 	 */
 	btrfs_bio_counter_inc_blocked(fs_info);
-	/* Tell the block device(s) that the sectors can be discarded */
-	ret = btrfs_map_block(fs_info, BTRFS_MAP_DISCARD, bytenr, &num_bytes,
-			      &bbio, 0);
-	/* Error condition is -ENOMEM */
-	if (!ret) {
-		struct btrfs_bio_stripe *stripe = bbio->stripes;
+	while (cur < end) {
+		struct btrfs_bio_stripe *stripe;
 		int i;
 
+		num_bytes = end - cur;
+		/* Tell the block device(s) that the sectors can be discarded */
+		ret = btrfs_map_block(fs_info, BTRFS_MAP_DISCARD, cur,
+				      &num_bytes, &bbio, 0);
+		/*
+		 * Error can be -ENOMEM, -ENOENT (no such chunk mapping) or
+		 * -EOPNOTSUPP. For any such error, @num_bytes is not updated,
+		 * thus we can't continue anyway.
+		 */
+		if (ret < 0)
+			goto out;
 
+		stripe = bbio->stripes;
 		for (i = 0; i < bbio->num_stripes; i++, stripe++) {
 			u64 bytes;
 			struct request_queue *req_q;
@@ -1341,10 +1330,19 @@ int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr,
 						  stripe->physical,
 						  stripe->length,
 						  &bytes);
-			if (!ret)
+			if (!ret) {
 				discarded_bytes += bytes;
-			else if (ret != -EOPNOTSUPP)
-				break; /* Logic errors or -ENOMEM, or -EIO but I don't know how that could happen JDM */
+			} else if (ret != -EOPNOTSUPP) {
+				/*
+				 * Logic errors or -ENOMEM, or -EIO, but
+				 * unlikely to happen.
+				 *
+				 * And since there are two loops, explicitly
+				 * go to out to avoid confusion.
+				 */
+				btrfs_put_bbio(bbio);
+				goto out;
+			}
 
 			/*
 			 * Just in case we get back EOPNOTSUPP for some reason,
@@ -1354,7 +1352,9 @@ int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr,
 			ret = 0;
 		}
 		btrfs_put_bbio(bbio);
+		cur += num_bytes;
 	}
+out:
 	btrfs_bio_counter_dec(fs_info);
 
 	if (actual_bytes)
@@ -1447,7 +1447,6 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 	if (!path)
 		return -ENOMEM;
 
-	path->reada = READA_FORWARD;
 	path->leave_spinning = 1;
 	/* this will setup the path even if it fails to insert the back ref */
 	ret = insert_inline_extent_backref(trans, path, bytenr, num_bytes,
@@ -1472,11 +1471,17 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_release_path(path);
 
-	path->reada = READA_FORWARD;
 	path->leave_spinning = 1;
 	/* now insert the actual backref */
-	ret = insert_extent_backref(trans, path, bytenr, parent, root_objectid,
-				    owner, offset, refs_to_add);
+	if (owner < BTRFS_FIRST_FREE_OBJECTID) {
+		BUG_ON(refs_to_add != 1);
+		ret = insert_tree_block_ref(trans, path, bytenr, parent,
+					    root_objectid);
+	} else {
+		ret = insert_extent_data_ref(trans, path, bytenr, parent,
+					     root_objectid, owner, offset,
+					     refs_to_add);
+	}
 	if (ret)
 		btrfs_abort_transaction(trans, ret);
 out:
@@ -1561,7 +1566,7 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans,
 	int err = 0;
 	int metadata = !extent_op->is_data;
 
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return 0;
 
 	if (metadata && !btrfs_fs_incompat(fs_info, SKINNY_METADATA))
@@ -1582,7 +1587,6 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans,
 	}
 
 again:
-	path->reada = READA_FORWARD;
 	path->leave_spinning = 1;
 	ret = btrfs_search_slot(trans, fs_info->extent_root, &key, path, 0, 1);
 	if (ret < 0) {
@@ -1681,10 +1685,9 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
 {
 	int ret = 0;
 
-	if (trans->aborted) {
+	if (TRANS_ABORTED(trans)) {
 		if (insert_reserved)
-			btrfs_pin_extent(trans->fs_info, node->bytenr,
-					 node->num_bytes, 1);
+			btrfs_pin_extent(trans, node->bytenr, node->num_bytes, 1);
 		return 0;
 	}
 
@@ -1699,8 +1702,7 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
 	else
 		BUG();
 	if (ret && insert_reserved)
-		btrfs_pin_extent(trans->fs_info, node->bytenr,
-				 node->num_bytes, 1);
+		btrfs_pin_extent(trans, node->bytenr, node->num_bytes, 1);
 	return ret;
 }
 
@@ -1845,11 +1847,10 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans,
 	spin_unlock(&delayed_refs->lock);
 
 	if (head->must_insert_reserved) {
-		btrfs_pin_extent(fs_info, head->bytenr,
-				 head->num_bytes, 1);
+		btrfs_pin_extent(trans, head->bytenr, head->num_bytes, 1);
 		if (head->is_data) {
-			ret = btrfs_del_csums(trans, fs_info, head->bytenr,
-					      head->num_bytes);
+			ret = btrfs_del_csums(trans, fs_info->csum_root,
+					      head->bytenr, head->num_bytes);
 		}
 	}
 
@@ -2169,7 +2170,7 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans,
 	int run_all = count == (unsigned long)-1;
 
 	/* We'll clean this up in btrfs_cleanup_transaction */
-	if (trans->aborted)
+	if (TRANS_ABORTED(trans))
 		return 0;
 
 	if (test_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &fs_info->flags))
@@ -2216,7 +2217,7 @@ out:
 }
 
 int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
-				u64 bytenr, u64 num_bytes, u64 flags,
+				struct extent_buffer *eb, u64 flags,
 				int level, int is_data)
 {
 	struct btrfs_delayed_extent_op *extent_op;
@@ -2232,7 +2233,7 @@ int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
 	extent_op->is_data = is_data ? true : false;
 	extent_op->level = level;
 
-	ret = btrfs_add_delayed_extent_op(trans, bytenr, num_bytes, extent_op);
+	ret = btrfs_add_delayed_extent_op(trans, eb->start, eb->len, extent_op);
 	if (ret)
 		btrfs_free_delayed_extent_op(extent_op);
 	return ret;
@@ -2516,7 +2517,7 @@ int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 
 int btrfs_extent_readonly(struct btrfs_fs_info *fs_info, u64 bytenr)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	int readonly = 0;
 
 	block_group = btrfs_lookup_block_group(fs_info, bytenr);
@@ -2546,7 +2547,7 @@ static u64 get_alloc_profile_by_root(struct btrfs_root *root, int data)
 
 static u64 first_logical_byte(struct btrfs_fs_info *fs_info, u64 search_start)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	u64 bytenr;
 
 	spin_lock(&fs_info->block_group_cache_lock);
@@ -2560,13 +2561,14 @@ static u64 first_logical_byte(struct btrfs_fs_info *fs_info, u64 search_start)
 	if (!cache)
 		return 0;
 
-	bytenr = cache->key.objectid;
+	bytenr = cache->start;
 	btrfs_put_block_group(cache);
 
 	return bytenr;
 }
 
-static int pin_down_extent(struct btrfs_block_group_cache *cache,
+static int pin_down_extent(struct btrfs_trans_handle *trans,
+			   struct btrfs_block_group *cache,
 			   u64 bytenr, u64 num_bytes, int reserved)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
@@ -2585,23 +2587,20 @@ static int pin_down_extent(struct btrfs_block_group_cache *cache,
 
 	percpu_counter_add_batch(&cache->space_info->total_bytes_pinned,
 		    num_bytes, BTRFS_TOTAL_BYTES_PINNED_BATCH);
-	set_extent_dirty(fs_info->pinned_extents, bytenr,
+	set_extent_dirty(&trans->transaction->pinned_extents, bytenr,
 			 bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL);
 	return 0;
 }
 
-/*
- * this function must be called within transaction
- */
-int btrfs_pin_extent(struct btrfs_fs_info *fs_info,
+int btrfs_pin_extent(struct btrfs_trans_handle *trans,
 		     u64 bytenr, u64 num_bytes, int reserved)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 
-	cache = btrfs_lookup_block_group(fs_info, bytenr);
+	cache = btrfs_lookup_block_group(trans->fs_info, bytenr);
 	BUG_ON(!cache); /* Logic error */
 
-	pin_down_extent(cache, bytenr, num_bytes, reserved);
+	pin_down_extent(trans, cache, bytenr, num_bytes, reserved);
 
 	btrfs_put_block_group(cache);
 	return 0;
@@ -2610,13 +2609,15 @@ int btrfs_pin_extent(struct btrfs_fs_info *fs_info,
 /*
  * this function must be called within transaction
  */
-int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info,
+int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans,
 				    u64 bytenr, u64 num_bytes)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	int ret;
 
-	cache = btrfs_lookup_block_group(fs_info, bytenr);
+	btrfs_add_excluded_extent(trans->fs_info, bytenr, num_bytes);
+
+	cache = btrfs_lookup_block_group(trans->fs_info, bytenr);
 	if (!cache)
 		return -EINVAL;
 
@@ -2628,7 +2629,7 @@ int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info,
 	 */
 	btrfs_cache_block_group(cache, 1);
 
-	pin_down_extent(cache, bytenr, num_bytes, 0);
+	pin_down_extent(trans, cache, bytenr, num_bytes, 0);
 
 	/* remove us from the free space cache (if we're there at all) */
 	ret = btrfs_remove_free_space(cache, bytenr, num_bytes);
@@ -2640,7 +2641,7 @@ static int __exclude_logged_extent(struct btrfs_fs_info *fs_info,
 				   u64 start, u64 num_bytes)
 {
 	int ret;
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_caching_control *caching_ctl;
 
 	block_group = btrfs_lookup_block_group(fs_info, start);
@@ -2652,7 +2653,7 @@ static int __exclude_logged_extent(struct btrfs_fs_info *fs_info,
 
 	if (!caching_ctl) {
 		/* Logic error */
-		BUG_ON(!btrfs_block_group_cache_done(block_group));
+		BUG_ON(!btrfs_block_group_done(block_group));
 		ret = btrfs_remove_free_space(block_group, start, num_bytes);
 	} else {
 		mutex_lock(&caching_ctl->mutex);
@@ -2717,7 +2718,7 @@ int btrfs_exclude_logged_extents(struct extent_buffer *eb)
 }
 
 static void
-btrfs_inc_block_group_reservations(struct btrfs_block_group_cache *bg)
+btrfs_inc_block_group_reservations(struct btrfs_block_group *bg)
 {
 	atomic_inc(&bg->reservations);
 }
@@ -2726,14 +2727,14 @@ void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_caching_control *next;
 	struct btrfs_caching_control *caching_ctl;
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 
 	down_write(&fs_info->commit_root_sem);
 
 	list_for_each_entry_safe(caching_ctl, next,
 				 &fs_info->caching_block_groups, list) {
 		cache = caching_ctl->block_group;
-		if (btrfs_block_group_cache_done(cache)) {
+		if (btrfs_block_group_done(cache)) {
 			cache->last_byte_to_unpin = (u64)-1;
 			list_del_init(&caching_ctl->list);
 			btrfs_put_caching_control(caching_ctl);
@@ -2742,11 +2743,6 @@ void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info)
 		}
 	}
 
-	if (fs_info->pinned_extents == &fs_info->freed_extents[0])
-		fs_info->pinned_extents = &fs_info->freed_extents[1];
-	else
-		fs_info->pinned_extents = &fs_info->freed_extents[0];
-
 	up_write(&fs_info->commit_root_sem);
 
 	btrfs_update_global_block_rsv(fs_info);
@@ -2785,7 +2781,7 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info,
 			      u64 start, u64 end,
 			      const bool return_free_space)
 {
-	struct btrfs_block_group_cache *cache = NULL;
+	struct btrfs_block_group *cache = NULL;
 	struct btrfs_space_info *space_info;
 	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
 	struct btrfs_free_cluster *cluster = NULL;
@@ -2797,7 +2793,7 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info,
 	while (start <= end) {
 		readonly = false;
 		if (!cache ||
-		    start >= cache->key.objectid + cache->key.offset) {
+		    start >= cache->start + cache->length) {
 			if (cache)
 				btrfs_put_block_group(cache);
 			total_unpinned = 0;
@@ -2810,7 +2806,7 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info,
 			empty_cluster <<= 1;
 		}
 
-		len = cache->key.objectid + cache->key.offset - start;
+		len = cache->start + cache->length - start;
 		len = min(len, end + 1 - start);
 
 		if (start < cache->last_byte_to_unpin) {
@@ -2880,19 +2876,16 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info,
 int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_block_group_cache *block_group, *tmp;
+	struct btrfs_block_group *block_group, *tmp;
 	struct list_head *deleted_bgs;
 	struct extent_io_tree *unpin;
 	u64 start;
 	u64 end;
 	int ret;
 
-	if (fs_info->pinned_extents == &fs_info->freed_extents[0])
-		unpin = &fs_info->freed_extents[1];
-	else
-		unpin = &fs_info->freed_extents[0];
+	unpin = &trans->transaction->pinned_extents;
 
-	while (!trans->aborted) {
+	while (!TRANS_ABORTED(trans)) {
 		struct extent_state *cached_state = NULL;
 
 		mutex_lock(&fs_info->unused_bg_unpin_mutex);
@@ -2902,8 +2895,11 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 			break;
 		}
+		if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags))
+			clear_extent_bits(&fs_info->excluded_extents, start,
+					  end, EXTENT_UPTODATE);
 
-		if (btrfs_test_opt(fs_info, DISCARD))
+		if (btrfs_test_opt(fs_info, DISCARD_SYNC))
 			ret = btrfs_discard_extent(fs_info, start,
 						   end + 1 - start, NULL);
 
@@ -2914,6 +2910,11 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 		cond_resched();
 	}
 
+	if (btrfs_test_opt(fs_info, DISCARD_ASYNC)) {
+		btrfs_discard_calc_delay(&fs_info->discard_ctl);
+		btrfs_discard_schedule_work(&fs_info->discard_ctl, true);
+	}
+
 	/*
 	 * Transaction is finished.  We don't need the lock anymore.  We
 	 * do need to clean up the block groups in case of a transaction
@@ -2924,10 +2925,10 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 		u64 trimmed = 0;
 
 		ret = -EROFS;
-		if (!trans->aborted)
+		if (!TRANS_ABORTED(trans))
 			ret = btrfs_discard_extent(fs_info,
-						   block_group->key.objectid,
-						   block_group->key.offset,
+						   block_group->start,
+						   block_group->length,
 						   &trimmed);
 
 		list_del_init(&block_group->bg_list);
@@ -2974,7 +2975,6 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 	if (!path)
 		return -ENOMEM;
 
-	path->reada = READA_FORWARD;
 	path->leave_spinning = 1;
 
 	is_data = owner_objectid >= BTRFS_FIRST_FREE_OBJECTID;
@@ -3155,7 +3155,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		btrfs_release_path(path);
 
 		if (is_data) {
-			ret = btrfs_del_csums(trans, info, bytenr, num_bytes);
+			ret = btrfs_del_csums(trans, info->csum_root, bytenr,
+					      num_bytes);
 			if (ret) {
 				btrfs_abort_transaction(trans, ret);
 				goto out;
@@ -3262,7 +3263,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 	}
 
 	if (last_ref && btrfs_header_generation(buf) == trans->transid) {
-		struct btrfs_block_group_cache *cache;
+		struct btrfs_block_group *cache;
 
 		if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
 			ret = check_ref_cleanup(trans, buf->start);
@@ -3274,7 +3275,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 		cache = btrfs_lookup_block_group(fs_info, buf->start);
 
 		if (btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) {
-			pin_down_extent(cache, buf->start, buf->len, 1);
+			pin_down_extent(trans, cache, buf->start, buf->len, 1);
 			btrfs_put_block_group(cache);
 			goto out;
 		}
@@ -3318,7 +3319,7 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref)
 	    (ref->type == BTRFS_REF_DATA &&
 	     ref->data_ref.ref_root == BTRFS_TREE_LOG_OBJECTID)) {
 		/* unlocks the pinned mutex */
-		btrfs_pin_extent(fs_info, ref->bytenr, ref->len, 1);
+		btrfs_pin_extent(trans, ref->bytenr, ref->len, 1);
 		old_ref_mod = new_ref_mod = 0;
 		ret = 0;
 	} else if (ref->type == BTRFS_REF_METADATA) {
@@ -3349,15 +3350,14 @@ enum btrfs_loop_type {
 };
 
 static inline void
-btrfs_lock_block_group(struct btrfs_block_group_cache *cache,
+btrfs_lock_block_group(struct btrfs_block_group *cache,
 		       int delalloc)
 {
 	if (delalloc)
 		down_read(&cache->data_rwsem);
 }
 
-static inline void
-btrfs_grab_block_group(struct btrfs_block_group_cache *cache,
+static inline void btrfs_grab_block_group(struct btrfs_block_group *cache,
 		       int delalloc)
 {
 	btrfs_get_block_group(cache);
@@ -3365,12 +3365,12 @@ btrfs_grab_block_group(struct btrfs_block_group_cache *cache,
 		down_read(&cache->data_rwsem);
 }
 
-static struct btrfs_block_group_cache *
-btrfs_lock_cluster(struct btrfs_block_group_cache *block_group,
+static struct btrfs_block_group *btrfs_lock_cluster(
+		   struct btrfs_block_group *block_group,
 		   struct btrfs_free_cluster *cluster,
 		   int delalloc)
 {
-	struct btrfs_block_group_cache *used_bg = NULL;
+	struct btrfs_block_group *used_bg = NULL;
 
 	spin_lock(&cluster->refill_lock);
 	while (1) {
@@ -3404,7 +3404,7 @@ btrfs_lock_cluster(struct btrfs_block_group_cache *block_group,
 }
 
 static inline void
-btrfs_release_block_group(struct btrfs_block_group_cache *cache,
+btrfs_release_block_group(struct btrfs_block_group *cache,
 			 int delalloc)
 {
 	if (delalloc)
@@ -3412,13 +3412,16 @@ btrfs_release_block_group(struct btrfs_block_group_cache *cache,
 	btrfs_put_block_group(cache);
 }
 
+enum btrfs_extent_allocation_policy {
+	BTRFS_EXTENT_ALLOC_CLUSTERED,
+};
+
 /*
  * Structure used internally for find_free_extent() function.  Wraps needed
  * parameters.
  */
 struct find_free_extent_ctl {
 	/* Basic allocation info */
-	u64 ram_bytes;
 	u64 num_bytes;
 	u64 empty_size;
 	u64 flags;
@@ -3429,6 +3432,8 @@ struct find_free_extent_ctl {
 
 	/* For clustered allocation */
 	u64 empty_cluster;
+	struct btrfs_free_cluster *last_ptr;
+	bool use_cluster;
 
 	bool have_caching_bg;
 	bool orig_have_caching_bg;
@@ -3464,6 +3469,12 @@ struct find_free_extent_ctl {
 
 	/* Found result */
 	u64 found_offset;
+
+	/* Hint where to start looking for an empty space */
+	u64 hint_byte;
+
+	/* Allocation policy */
+	enum btrfs_extent_allocation_policy policy;
 };
 
 
@@ -3475,12 +3486,12 @@ struct find_free_extent_ctl {
  * Return >0 to inform caller that we find nothing
  * Return 0 means we have found a location and set ffe_ctl->found_offset.
  */
-static int find_free_extent_clustered(struct btrfs_block_group_cache *bg,
-		struct btrfs_free_cluster *last_ptr,
-		struct find_free_extent_ctl *ffe_ctl,
-		struct btrfs_block_group_cache **cluster_bg_ret)
+static int find_free_extent_clustered(struct btrfs_block_group *bg,
+				      struct find_free_extent_ctl *ffe_ctl,
+				      struct btrfs_block_group **cluster_bg_ret)
 {
-	struct btrfs_block_group_cache *cluster_bg;
+	struct btrfs_block_group *cluster_bg;
+	struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr;
 	u64 aligned_cluster;
 	u64 offset;
 	int ret;
@@ -3493,7 +3504,7 @@ static int find_free_extent_clustered(struct btrfs_block_group_cache *bg,
 		goto release_cluster;
 
 	offset = btrfs_alloc_from_cluster(cluster_bg, last_ptr,
-			ffe_ctl->num_bytes, cluster_bg->key.objectid,
+			ffe_ctl->num_bytes, cluster_bg->start,
 			&ffe_ctl->max_extent_size);
 	if (offset) {
 		/* We have a block, we're done */
@@ -3579,10 +3590,10 @@ refill_cluster:
  * Return 0 when we found an free extent and set ffe_ctrl->found_offset
  * Return -EAGAIN to inform caller that we need to re-search this block group
  */
-static int find_free_extent_unclustered(struct btrfs_block_group_cache *bg,
-		struct btrfs_free_cluster *last_ptr,
-		struct find_free_extent_ctl *ffe_ctl)
+static int find_free_extent_unclustered(struct btrfs_block_group *bg,
+					struct find_free_extent_ctl *ffe_ctl)
 {
+	struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr;
 	u64 offset;
 
 	/*
@@ -3638,16 +3649,101 @@ static int find_free_extent_unclustered(struct btrfs_block_group_cache *bg,
 	return 0;
 }
 
+static int do_allocation_clustered(struct btrfs_block_group *block_group,
+				   struct find_free_extent_ctl *ffe_ctl,
+				   struct btrfs_block_group **bg_ret)
+{
+	int ret;
+
+	/* We want to try and use the cluster allocator, so lets look there */
+	if (ffe_ctl->last_ptr && ffe_ctl->use_cluster) {
+		ret = find_free_extent_clustered(block_group, ffe_ctl, bg_ret);
+		if (ret >= 0 || ret == -EAGAIN)
+			return ret;
+		/* ret == -ENOENT case falls through */
+	}
+
+	return find_free_extent_unclustered(block_group, ffe_ctl);
+}
+
+static int do_allocation(struct btrfs_block_group *block_group,
+			 struct find_free_extent_ctl *ffe_ctl,
+			 struct btrfs_block_group **bg_ret)
+{
+	switch (ffe_ctl->policy) {
+	case BTRFS_EXTENT_ALLOC_CLUSTERED:
+		return do_allocation_clustered(block_group, ffe_ctl, bg_ret);
+	default:
+		BUG();
+	}
+}
+
+static void release_block_group(struct btrfs_block_group *block_group,
+				struct find_free_extent_ctl *ffe_ctl,
+				int delalloc)
+{
+	switch (ffe_ctl->policy) {
+	case BTRFS_EXTENT_ALLOC_CLUSTERED:
+		ffe_ctl->retry_clustered = false;
+		ffe_ctl->retry_unclustered = false;
+		break;
+	default:
+		BUG();
+	}
+
+	BUG_ON(btrfs_bg_flags_to_raid_index(block_group->flags) !=
+	       ffe_ctl->index);
+	btrfs_release_block_group(block_group, delalloc);
+}
+
+static void found_extent_clustered(struct find_free_extent_ctl *ffe_ctl,
+				   struct btrfs_key *ins)
+{
+	struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr;
+
+	if (!ffe_ctl->use_cluster && last_ptr) {
+		spin_lock(&last_ptr->lock);
+		last_ptr->window_start = ins->objectid;
+		spin_unlock(&last_ptr->lock);
+	}
+}
+
+static void found_extent(struct find_free_extent_ctl *ffe_ctl,
+			 struct btrfs_key *ins)
+{
+	switch (ffe_ctl->policy) {
+	case BTRFS_EXTENT_ALLOC_CLUSTERED:
+		found_extent_clustered(ffe_ctl, ins);
+		break;
+	default:
+		BUG();
+	}
+}
+
+static int chunk_allocation_failed(struct find_free_extent_ctl *ffe_ctl)
+{
+	switch (ffe_ctl->policy) {
+	case BTRFS_EXTENT_ALLOC_CLUSTERED:
+		/*
+		 * If we can't allocate a new chunk we've already looped through
+		 * at least once, move on to the NO_EMPTY_SIZE case.
+		 */
+		ffe_ctl->loop = LOOP_NO_EMPTY_SIZE;
+		return 0;
+	default:
+		BUG();
+	}
+}
+
 /*
  * Return >0 means caller needs to re-search for free extent
  * Return 0 means we have the needed free extent.
  * Return <0 means we failed to locate any free extent.
  */
 static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info,
-					struct btrfs_free_cluster *last_ptr,
 					struct btrfs_key *ins,
 					struct find_free_extent_ctl *ffe_ctl,
-					int full_search, bool use_cluster)
+					bool full_search)
 {
 	struct btrfs_root *root = fs_info->extent_root;
 	int ret;
@@ -3664,11 +3760,7 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info,
 		return 1;
 
 	if (ins->objectid) {
-		if (!use_cluster && last_ptr) {
-			spin_lock(&last_ptr->lock);
-			last_ptr->window_start = ins->objectid;
-			spin_unlock(&last_ptr->lock);
-		}
+		found_extent(ffe_ctl, ins);
 		return 0;
 	}
 
@@ -3714,16 +3806,10 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info,
 			ret = btrfs_chunk_alloc(trans, ffe_ctl->flags,
 						CHUNK_ALLOC_FORCE);
 
-			/*
-			 * If we can't allocate a new chunk we've already looped
-			 * through at least once, move on to the NO_EMPTY_SIZE
-			 * case.
-			 */
-			if (ret == -ENOSPC)
-				ffe_ctl->loop = LOOP_NO_EMPTY_SIZE;
-
 			/* Do not bail out on ENOSPC since we can do more. */
-			if (ret < 0 && ret != -ENOSPC)
+			if (ret == -ENOSPC)
+				ret = chunk_allocation_failed(ffe_ctl);
+			else if (ret < 0)
 				btrfs_abort_transaction(trans, ret);
 			else
 				ret = 0;
@@ -3734,6 +3820,9 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info,
 		}
 
 		if (ffe_ctl->loop == LOOP_NO_EMPTY_SIZE) {
+			if (ffe_ctl->policy != BTRFS_EXTENT_ALLOC_CLUSTERED)
+				return -ENOSPC;
+
 			/*
 			 * Don't loop again if we already have no empty_size and
 			 * no empty_cluster.
@@ -3749,6 +3838,71 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info,
 	return -ENOSPC;
 }
 
+static int prepare_allocation_clustered(struct btrfs_fs_info *fs_info,
+					struct find_free_extent_ctl *ffe_ctl,
+					struct btrfs_space_info *space_info,
+					struct btrfs_key *ins)
+{
+	/*
+	 * If our free space is heavily fragmented we may not be able to make
+	 * big contiguous allocations, so instead of doing the expensive search
+	 * for free space, simply return ENOSPC with our max_extent_size so we
+	 * can go ahead and search for a more manageable chunk.
+	 *
+	 * If our max_extent_size is large enough for our allocation simply
+	 * disable clustering since we will likely not be able to find enough
+	 * space to create a cluster and induce latency trying.
+	 */
+	if (space_info->max_extent_size) {
+		spin_lock(&space_info->lock);
+		if (space_info->max_extent_size &&
+		    ffe_ctl->num_bytes > space_info->max_extent_size) {
+			ins->offset = space_info->max_extent_size;
+			spin_unlock(&space_info->lock);
+			return -ENOSPC;
+		} else if (space_info->max_extent_size) {
+			ffe_ctl->use_cluster = false;
+		}
+		spin_unlock(&space_info->lock);
+	}
+
+	ffe_ctl->last_ptr = fetch_cluster_info(fs_info, space_info,
+					       &ffe_ctl->empty_cluster);
+	if (ffe_ctl->last_ptr) {
+		struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr;
+
+		spin_lock(&last_ptr->lock);
+		if (last_ptr->block_group)
+			ffe_ctl->hint_byte = last_ptr->window_start;
+		if (last_ptr->fragmented) {
+			/*
+			 * We still set window_start so we can keep track of the
+			 * last place we found an allocation to try and save
+			 * some time.
+			 */
+			ffe_ctl->hint_byte = last_ptr->window_start;
+			ffe_ctl->use_cluster = false;
+		}
+		spin_unlock(&last_ptr->lock);
+	}
+
+	return 0;
+}
+
+static int prepare_allocation(struct btrfs_fs_info *fs_info,
+			      struct find_free_extent_ctl *ffe_ctl,
+			      struct btrfs_space_info *space_info,
+			      struct btrfs_key *ins)
+{
+	switch (ffe_ctl->policy) {
+	case BTRFS_EXTENT_ALLOC_CLUSTERED:
+		return prepare_allocation_clustered(fs_info, ffe_ctl,
+						    space_info, ins);
+	default:
+		BUG();
+	}
+}
+
 /*
  * walks the btree of allocated extents and find a hole of a given size.
  * The key ins is changed to record the hole:
@@ -3776,31 +3930,35 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info,
  */
 static noinline int find_free_extent(struct btrfs_fs_info *fs_info,
 				u64 ram_bytes, u64 num_bytes, u64 empty_size,
-				u64 hint_byte, struct btrfs_key *ins,
+				u64 hint_byte_orig, struct btrfs_key *ins,
 				u64 flags, int delalloc)
 {
 	int ret = 0;
-	struct btrfs_free_cluster *last_ptr = NULL;
-	struct btrfs_block_group_cache *block_group = NULL;
+	int cache_block_group_error = 0;
+	struct btrfs_block_group *block_group = NULL;
 	struct find_free_extent_ctl ffe_ctl = {0};
 	struct btrfs_space_info *space_info;
-	bool use_cluster = true;
 	bool full_search = false;
 
 	WARN_ON(num_bytes < fs_info->sectorsize);
 
-	ffe_ctl.ram_bytes = ram_bytes;
 	ffe_ctl.num_bytes = num_bytes;
 	ffe_ctl.empty_size = empty_size;
 	ffe_ctl.flags = flags;
 	ffe_ctl.search_start = 0;
-	ffe_ctl.retry_clustered = false;
-	ffe_ctl.retry_unclustered = false;
 	ffe_ctl.delalloc = delalloc;
 	ffe_ctl.index = btrfs_bg_flags_to_raid_index(flags);
 	ffe_ctl.have_caching_bg = false;
 	ffe_ctl.orig_have_caching_bg = false;
 	ffe_ctl.found_offset = 0;
+	ffe_ctl.hint_byte = hint_byte_orig;
+	ffe_ctl.policy = BTRFS_EXTENT_ALLOC_CLUSTERED;
+
+	/* For clustered allocation */
+	ffe_ctl.retry_clustered = false;
+	ffe_ctl.retry_unclustered = false;
+	ffe_ctl.last_ptr = NULL;
+	ffe_ctl.use_cluster = true;
 
 	ins->type = BTRFS_EXTENT_ITEM_KEY;
 	ins->objectid = 0;
@@ -3814,51 +3972,14 @@ static noinline int find_free_extent(struct btrfs_fs_info *fs_info,
 		return -ENOSPC;
 	}
 
-	/*
-	 * If our free space is heavily fragmented we may not be able to make
-	 * big contiguous allocations, so instead of doing the expensive search
-	 * for free space, simply return ENOSPC with our max_extent_size so we
-	 * can go ahead and search for a more manageable chunk.
-	 *
-	 * If our max_extent_size is large enough for our allocation simply
-	 * disable clustering since we will likely not be able to find enough
-	 * space to create a cluster and induce latency trying.
-	 */
-	if (unlikely(space_info->max_extent_size)) {
-		spin_lock(&space_info->lock);
-		if (space_info->max_extent_size &&
-		    num_bytes > space_info->max_extent_size) {
-			ins->offset = space_info->max_extent_size;
-			spin_unlock(&space_info->lock);
-			return -ENOSPC;
-		} else if (space_info->max_extent_size) {
-			use_cluster = false;
-		}
-		spin_unlock(&space_info->lock);
-	}
-
-	last_ptr = fetch_cluster_info(fs_info, space_info,
-				      &ffe_ctl.empty_cluster);
-	if (last_ptr) {
-		spin_lock(&last_ptr->lock);
-		if (last_ptr->block_group)
-			hint_byte = last_ptr->window_start;
-		if (last_ptr->fragmented) {
-			/*
-			 * We still set window_start so we can keep track of the
-			 * last place we found an allocation to try and save
-			 * some time.
-			 */
-			hint_byte = last_ptr->window_start;
-			use_cluster = false;
-		}
-		spin_unlock(&last_ptr->lock);
-	}
+	ret = prepare_allocation(fs_info, &ffe_ctl, space_info, ins);
+	if (ret < 0)
+		return ret;
 
 	ffe_ctl.search_start = max(ffe_ctl.search_start,
 				   first_logical_byte(fs_info, 0));
-	ffe_ctl.search_start = max(ffe_ctl.search_start, hint_byte);
-	if (ffe_ctl.search_start == hint_byte) {
+	ffe_ctl.search_start = max(ffe_ctl.search_start, ffe_ctl.hint_byte);
+	if (ffe_ctl.search_start == ffe_ctl.hint_byte) {
 		block_group = btrfs_lookup_block_group(fs_info,
 						       ffe_ctl.search_start);
 		/*
@@ -3899,12 +4020,14 @@ search:
 	down_read(&space_info->groups_sem);
 	list_for_each_entry(block_group,
 			    &space_info->block_groups[ffe_ctl.index], list) {
+		struct btrfs_block_group *bg_ret;
+
 		/* If the block group is read-only, we can skip it entirely. */
 		if (unlikely(block_group->ro))
 			continue;
 
 		btrfs_grab_block_group(block_group, delalloc);
-		ffe_ctl.search_start = block_group->key.objectid;
+		ffe_ctl.search_start = block_group->start;
 
 		/*
 		 * this can happen if we end up cycling through all the
@@ -3935,56 +4058,50 @@ search:
 		}
 
 have_block_group:
-		ffe_ctl.cached = btrfs_block_group_cache_done(block_group);
+		ffe_ctl.cached = btrfs_block_group_done(block_group);
 		if (unlikely(!ffe_ctl.cached)) {
 			ffe_ctl.have_caching_bg = true;
 			ret = btrfs_cache_block_group(block_group, 0);
-			BUG_ON(ret < 0);
+
+			/*
+			 * If we get ENOMEM here or something else we want to
+			 * try other block groups, because it may not be fatal.
+			 * However if we can't find anything else we need to
+			 * save our return here so that we return the actual
+			 * error that caused problems, not ENOSPC.
+			 */
+			if (ret < 0) {
+				if (!cache_block_group_error)
+					cache_block_group_error = ret;
+				ret = 0;
+				goto loop;
+			}
 			ret = 0;
 		}
 
 		if (unlikely(block_group->cached == BTRFS_CACHE_ERROR))
 			goto loop;
 
-		/*
-		 * Ok we want to try and use the cluster allocator, so
-		 * lets look there
-		 */
-		if (last_ptr && use_cluster) {
-			struct btrfs_block_group_cache *cluster_bg = NULL;
-
-			ret = find_free_extent_clustered(block_group, last_ptr,
-							 &ffe_ctl, &cluster_bg);
-
-			if (ret == 0) {
-				if (cluster_bg && cluster_bg != block_group) {
-					btrfs_release_block_group(block_group,
-								  delalloc);
-					block_group = cluster_bg;
-				}
-				goto checks;
-			} else if (ret == -EAGAIN) {
-				goto have_block_group;
-			} else if (ret > 0) {
-				goto loop;
+		bg_ret = NULL;
+		ret = do_allocation(block_group, &ffe_ctl, &bg_ret);
+		if (ret == 0) {
+			if (bg_ret && bg_ret != block_group) {
+				btrfs_release_block_group(block_group, delalloc);
+				block_group = bg_ret;
 			}
-			/* ret == -ENOENT case falls through */
-		}
-
-		ret = find_free_extent_unclustered(block_group, last_ptr,
-						   &ffe_ctl);
-		if (ret == -EAGAIN)
+		} else if (ret == -EAGAIN) {
 			goto have_block_group;
-		else if (ret > 0)
+		} else if (ret > 0) {
 			goto loop;
-		/* ret == 0 case falls through */
-checks:
+		}
+
+		/* Checks */
 		ffe_ctl.search_start = round_up(ffe_ctl.found_offset,
 					     fs_info->stripesize);
 
 		/* move on to the next group */
 		if (ffe_ctl.search_start + num_bytes >
-		    block_group->key.objectid + block_group->key.offset) {
+		    block_group->start + block_group->length) {
 			btrfs_add_free_space(block_group, ffe_ctl.found_offset,
 					     num_bytes);
 			goto loop;
@@ -4012,21 +4129,16 @@ checks:
 		btrfs_release_block_group(block_group, delalloc);
 		break;
 loop:
-		ffe_ctl.retry_clustered = false;
-		ffe_ctl.retry_unclustered = false;
-		BUG_ON(btrfs_bg_flags_to_raid_index(block_group->flags) !=
-		       ffe_ctl.index);
-		btrfs_release_block_group(block_group, delalloc);
+		release_block_group(block_group, &ffe_ctl, delalloc);
 		cond_resched();
 	}
 	up_read(&space_info->groups_sem);
 
-	ret = find_free_extent_update_loop(fs_info, last_ptr, ins, &ffe_ctl,
-					   full_search, use_cluster);
+	ret = find_free_extent_update_loop(fs_info, ins, &ffe_ctl, full_search);
 	if (ret > 0)
 		goto search;
 
-	if (ret == -ENOSPC) {
+	if (ret == -ENOSPC && !cache_block_group_error) {
 		/*
 		 * Use ffe_ctl->total_free_space as fallback if we can't find
 		 * any contiguous hole.
@@ -4037,6 +4149,8 @@ loop:
 		space_info->max_extent_size = ffe_ctl.max_extent_size;
 		spin_unlock(&space_info->lock);
 		ins->offset = ffe_ctl.max_extent_size;
+	} else if (ret == -ENOSPC) {
+		ret = cache_block_group_error;
 	}
 	return ret;
 }
@@ -4129,12 +4243,10 @@ again:
 	return ret;
 }
 
-static int __btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
-					u64 start, u64 len,
-					int pin, int delalloc)
+int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
+			       u64 start, u64 len, int delalloc)
 {
-	struct btrfs_block_group_cache *cache;
-	int ret = 0;
+	struct btrfs_block_group *cache;
 
 	cache = btrfs_lookup_block_group(fs_info, start);
 	if (!cache) {
@@ -4143,30 +4255,30 @@ static int __btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
 		return -ENOSPC;
 	}
 
-	if (pin)
-		pin_down_extent(cache, start, len, 1);
-	else {
-		if (btrfs_test_opt(fs_info, DISCARD))
-			ret = btrfs_discard_extent(fs_info, start, len, NULL);
-		btrfs_add_free_space(cache, start, len);
-		btrfs_free_reserved_bytes(cache, len, delalloc);
-		trace_btrfs_reserved_extent_free(fs_info, start, len);
-	}
+	btrfs_add_free_space(cache, start, len);
+	btrfs_free_reserved_bytes(cache, len, delalloc);
+	trace_btrfs_reserved_extent_free(fs_info, start, len);
 
 	btrfs_put_block_group(cache);
-	return ret;
+	return 0;
 }
 
-int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
-			       u64 start, u64 len, int delalloc)
+int btrfs_pin_reserved_extent(struct btrfs_trans_handle *trans, u64 start,
+			      u64 len)
 {
-	return __btrfs_free_reserved_extent(fs_info, start, len, 0, delalloc);
-}
+	struct btrfs_block_group *cache;
+	int ret = 0;
 
-int btrfs_free_and_pin_reserved_extent(struct btrfs_fs_info *fs_info,
-				       u64 start, u64 len)
-{
-	return __btrfs_free_reserved_extent(fs_info, start, len, 1, 0);
+	cache = btrfs_lookup_block_group(trans->fs_info, start);
+	if (!cache) {
+		btrfs_err(trans->fs_info, "unable to find block group for %llu",
+			  start);
+		return -ENOSPC;
+	}
+
+	ret = pin_down_extent(trans, cache, start, len, 1);
+	btrfs_put_block_group(cache);
+	return ret;
 }
 
 static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
@@ -4366,7 +4478,7 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret;
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_space_info *space_info;
 
 	/*
@@ -4394,6 +4506,8 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans,
 
 	ret = alloc_reserved_file_extent(trans, 0, root_objectid, 0, owner,
 					 offset, ins, 1);
+	if (ret)
+		btrfs_pin_extent(trans, ins->objectid, ins->offset, 1);
 	btrfs_put_block_group(block_group);
 	return ret;
 }
@@ -4712,8 +4826,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
 		BUG_ON(ret); /* -ENOMEM */
 		ret = btrfs_dec_ref(trans, root, eb, 0);
 		BUG_ON(ret); /* -ENOMEM */
-		ret = btrfs_set_disk_extent_flags(trans, eb->start,
-						  eb->len, flag,
+		ret = btrfs_set_disk_extent_flags(trans, eb, flag,
 						  btrfs_header_level(eb), 0);
 		BUG_ON(ret); /* -ENOMEM */
 		wc->flags[level] |= flag;
@@ -5171,9 +5284,7 @@ static noinline int walk_up_tree(struct btrfs_trans_handle *trans,
  *
  * If called with for_reloc == 0, may exit early with -EAGAIN
  */
-int btrfs_drop_snapshot(struct btrfs_root *root,
-			 struct btrfs_block_rsv *block_rsv, int update_ref,
-			 int for_reloc)
+int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref, int for_reloc)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_path *path;
@@ -5212,9 +5323,6 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	if (err)
 		goto out_end_trans;
 
-	if (block_rsv)
-		trans->block_rsv = block_rsv;
-
 	/*
 	 * This will help us catch people modifying the fs tree while we're
 	 * dropping it.  It is unsafe to mess with the fs tree while it's being
@@ -5342,8 +5450,6 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 				err = PTR_ERR(trans);
 				goto out_free;
 			}
-			if (block_rsv)
-				trans->block_rsv = block_rsv;
 		}
 	}
 	btrfs_release_path(path);
@@ -5375,13 +5481,10 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 		}
 	}
 
-	if (test_bit(BTRFS_ROOT_IN_RADIX, &root->state)) {
+	if (test_bit(BTRFS_ROOT_IN_RADIX, &root->state))
 		btrfs_add_dropped_root(trans, root);
-	} else {
-		free_extent_buffer(root->node);
-		free_extent_buffer(root->commit_root);
-		btrfs_put_fs_root(root);
-	}
+	else
+		btrfs_put_root(root);
 	root_dropped = true;
 out_end_trans:
 	btrfs_end_transaction_throttle(trans);
@@ -5436,7 +5539,7 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
 
 	btrfs_assert_tree_locked(parent);
 	parent_level = btrfs_header_level(parent);
-	extent_buffer_get(parent);
+	atomic_inc(&parent->refs);
 	path->nodes[parent_level] = parent;
 	path->slots[parent_level] = btrfs_header_nritems(parent);
 
@@ -5480,7 +5583,7 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
  */
 u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	u64 free_bytes = 0;
 	int factor;
 
@@ -5498,9 +5601,8 @@ u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo)
 		}
 
 		factor = btrfs_bg_type_to_factor(block_group->flags);
-		free_bytes += (block_group->key.offset -
-			       btrfs_block_group_used(&block_group->item)) *
-			       factor;
+		free_bytes += (block_group->length -
+			       block_group->used) * factor;
 
 		spin_unlock(&block_group->lock);
 	}
@@ -5623,7 +5725,7 @@ static int btrfs_trim_free_extents(struct btrfs_device *device, u64 *trimmed)
  */
 int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
 {
-	struct btrfs_block_group_cache *cache = NULL;
+	struct btrfs_block_group *cache = NULL;
 	struct btrfs_device *device;
 	struct list_head *devices;
 	u64 group_trimmed;
@@ -5647,16 +5749,16 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
 
 	cache = btrfs_lookup_first_block_group(fs_info, range->start);
 	for (; cache; cache = btrfs_next_block_group(cache)) {
-		if (cache->key.objectid >= range_end) {
+		if (cache->start >= range_end) {
 			btrfs_put_block_group(cache);
 			break;
 		}
 
-		start = max(range->start, cache->key.objectid);
-		end = min(range_end, cache->key.objectid + cache->key.offset);
+		start = max(range->start, cache->start);
+		end = min(range_end, cache->start + cache->length);
 
 		if (end - start >= range->minlen) {
-			if (!btrfs_block_group_cache_done(cache)) {
+			if (!btrfs_block_group_done(cache)) {
 				ret = btrfs_cache_block_group(cache, 0);
 				if (ret) {
 					bg_failed++;
@@ -5712,47 +5814,3 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
 		return bg_ret;
 	return dev_ret;
 }
-
-/*
- * btrfs_{start,end}_write_no_snapshotting() are similar to
- * mnt_{want,drop}_write(), they are used to prevent some tasks from writing
- * data into the page cache through nocow before the subvolume is snapshoted,
- * but flush the data into disk after the snapshot creation, or to prevent
- * operations while snapshotting is ongoing and that cause the snapshot to be
- * inconsistent (writes followed by expanding truncates for example).
- */
-void btrfs_end_write_no_snapshotting(struct btrfs_root *root)
-{
-	percpu_counter_dec(&root->subv_writers->counter);
-	cond_wake_up(&root->subv_writers->wait);
-}
-
-int btrfs_start_write_no_snapshotting(struct btrfs_root *root)
-{
-	if (atomic_read(&root->will_be_snapshotted))
-		return 0;
-
-	percpu_counter_inc(&root->subv_writers->counter);
-	/*
-	 * Make sure counter is updated before we check for snapshot creation.
-	 */
-	smp_mb();
-	if (atomic_read(&root->will_be_snapshotted)) {
-		btrfs_end_write_no_snapshotting(root);
-		return 0;
-	}
-	return 1;
-}
-
-void btrfs_wait_for_snapshot_creation(struct btrfs_root *root)
-{
-	while (true) {
-		int ret;
-
-		ret = btrfs_start_write_no_snapshotting(root);
-		if (ret)
-			break;
-		wait_var_event(&root->will_be_snapshotted,
-			       !atomic_read(&root->will_be_snapshotted));
-	}
-}
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index cceaf05aada2..e12eb32d9e17 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -14,6 +14,7 @@
 #include <linux/prefetch.h>
 #include <linux/cleancache.h>
 #include "extent_io.h"
+#include "extent-io-tree.h"
 #include "extent_map.h"
 #include "ctree.h"
 #include "btrfs_inode.h"
@@ -34,36 +35,59 @@ static inline bool extent_state_in_tree(const struct extent_state *state)
 }
 
 #ifdef CONFIG_BTRFS_DEBUG
-static LIST_HEAD(buffers);
 static LIST_HEAD(states);
-
 static DEFINE_SPINLOCK(leak_lock);
 
-static inline
-void btrfs_leak_debug_add(struct list_head *new, struct list_head *head)
+static inline void btrfs_leak_debug_add(spinlock_t *lock,
+					struct list_head *new,
+					struct list_head *head)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&leak_lock, flags);
+	spin_lock_irqsave(lock, flags);
 	list_add(new, head);
-	spin_unlock_irqrestore(&leak_lock, flags);
+	spin_unlock_irqrestore(lock, flags);
 }
 
-static inline
-void btrfs_leak_debug_del(struct list_head *entry)
+static inline void btrfs_leak_debug_del(spinlock_t *lock,
+					struct list_head *entry)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&leak_lock, flags);
+	spin_lock_irqsave(lock, flags);
 	list_del(entry);
-	spin_unlock_irqrestore(&leak_lock, flags);
+	spin_unlock_irqrestore(lock, flags);
 }
 
-static inline
-void btrfs_leak_debug_check(void)
+void btrfs_extent_buffer_leak_debug_check(struct btrfs_fs_info *fs_info)
 {
-	struct extent_state *state;
 	struct extent_buffer *eb;
+	unsigned long flags;
+
+	/*
+	 * If we didn't get into open_ctree our allocated_ebs will not be
+	 * initialized, so just skip this.
+	 */
+	if (!fs_info->allocated_ebs.next)
+		return;
+
+	spin_lock_irqsave(&fs_info->eb_leak_lock, flags);
+	while (!list_empty(&fs_info->allocated_ebs)) {
+		eb = list_first_entry(&fs_info->allocated_ebs,
+				      struct extent_buffer, leak_list);
+		pr_err(
+	"BTRFS: buffer leak start %llu len %lu refs %d bflags %lu owner %llu\n",
+		       eb->start, eb->len, atomic_read(&eb->refs), eb->bflags,
+		       btrfs_header_owner(eb));
+		list_del(&eb->leak_list);
+		kmem_cache_free(extent_buffer_cache, eb);
+	}
+	spin_unlock_irqrestore(&fs_info->eb_leak_lock, flags);
+}
+
+static inline void btrfs_extent_state_leak_debug_check(void)
+{
+	struct extent_state *state;
 
 	while (!list_empty(&states)) {
 		state = list_entry(states.next, struct extent_state, leak_list);
@@ -74,14 +98,6 @@ void btrfs_leak_debug_check(void)
 		list_del(&state->leak_list);
 		kmem_cache_free(extent_state_cache, state);
 	}
-
-	while (!list_empty(&buffers)) {
-		eb = list_entry(buffers.next, struct extent_buffer, leak_list);
-		pr_err("BTRFS: buffer leak start %llu len %lu refs %d bflags %lu\n",
-		       eb->start, eb->len, atomic_read(&eb->refs), eb->bflags);
-		list_del(&eb->leak_list);
-		kmem_cache_free(extent_buffer_cache, eb);
-	}
 }
 
 #define btrfs_debug_check_extent_io_range(tree, start, end)		\
@@ -103,9 +119,9 @@ static inline void __btrfs_debug_check_extent_io_range(const char *caller,
 	}
 }
 #else
-#define btrfs_leak_debug_add(new, head)	do {} while (0)
-#define btrfs_leak_debug_del(entry)	do {} while (0)
-#define btrfs_leak_debug_check()	do {} while (0)
+#define btrfs_leak_debug_add(lock, new, head)	do {} while (0)
+#define btrfs_leak_debug_del(lock, entry)	do {} while (0)
+#define btrfs_extent_state_leak_debug_check()	do {} while (0)
 #define btrfs_debug_check_extent_io_range(c, s, e)	do {} while (0)
 #endif
 
@@ -117,7 +133,6 @@ struct tree_entry {
 
 struct extent_page_data {
 	struct bio *bio;
-	struct extent_io_tree *tree;
 	/* tells writepage not to lock the state bits for this range
 	 * it still does the unlocking
 	 */
@@ -196,19 +211,23 @@ static int __must_check flush_write_bio(struct extent_page_data *epd)
 	return ret;
 }
 
-int __init extent_io_init(void)
+int __init extent_state_cache_init(void)
 {
 	extent_state_cache = kmem_cache_create("btrfs_extent_state",
 			sizeof(struct extent_state), 0,
 			SLAB_MEM_SPREAD, NULL);
 	if (!extent_state_cache)
 		return -ENOMEM;
+	return 0;
+}
 
+int __init extent_io_init(void)
+{
 	extent_buffer_cache = kmem_cache_create("btrfs_extent_buffer",
 			sizeof(struct extent_buffer), 0,
 			SLAB_MEM_SPREAD, NULL);
 	if (!extent_buffer_cache)
-		goto free_state_cache;
+		return -ENOMEM;
 
 	if (bioset_init(&btrfs_bioset, BIO_POOL_SIZE,
 			offsetof(struct btrfs_io_bio, bio),
@@ -226,27 +245,35 @@ free_bioset:
 free_buffer_cache:
 	kmem_cache_destroy(extent_buffer_cache);
 	extent_buffer_cache = NULL;
+	return -ENOMEM;
+}
 
-free_state_cache:
+void __cold extent_state_cache_exit(void)
+{
+	btrfs_extent_state_leak_debug_check();
 	kmem_cache_destroy(extent_state_cache);
-	extent_state_cache = NULL;
-	return -ENOMEM;
 }
 
 void __cold extent_io_exit(void)
 {
-	btrfs_leak_debug_check();
-
 	/*
 	 * Make sure all delayed rcu free are flushed before we
 	 * destroy caches.
 	 */
 	rcu_barrier();
-	kmem_cache_destroy(extent_state_cache);
 	kmem_cache_destroy(extent_buffer_cache);
 	bioset_exit(&btrfs_bioset);
 }
 
+/*
+ * For the file_extent_tree, we want to hold the inode lock when we lookup and
+ * update the disk_i_size, but lockdep will complain because our io_tree we hold
+ * the tree lock and get the inode lock when setting delalloc.  These two things
+ * are unrelated, so make a class for the file_extent_tree so we don't get the
+ * two locking patterns mixed up.
+ */
+static struct lock_class_key file_extent_tree_class;
+
 void extent_io_tree_init(struct btrfs_fs_info *fs_info,
 			 struct extent_io_tree *tree, unsigned int owner,
 			 void *private_data)
@@ -258,6 +285,8 @@ void extent_io_tree_init(struct btrfs_fs_info *fs_info,
 	spin_lock_init(&tree->lock);
 	tree->private_data = private_data;
 	tree->owner = owner;
+	if (owner == IO_TREE_INODE_FILE_EXTENT)
+		lockdep_set_class(&tree->lock, &file_extent_tree_class);
 }
 
 void extent_io_tree_release(struct extent_io_tree *tree)
@@ -304,7 +333,7 @@ static struct extent_state *alloc_extent_state(gfp_t mask)
 	state->state = 0;
 	state->failrec = NULL;
 	RB_CLEAR_NODE(&state->rb_node);
-	btrfs_leak_debug_add(&state->leak_list, &states);
+	btrfs_leak_debug_add(&leak_lock, &state->leak_list, &states);
 	refcount_set(&state->refs, 1);
 	init_waitqueue_head(&state->wq);
 	trace_alloc_extent_state(state, mask, _RET_IP_);
@@ -317,7 +346,7 @@ void free_extent_state(struct extent_state *state)
 		return;
 	if (refcount_dec_and_test(&state->refs)) {
 		WARN_ON(extent_state_in_tree(state));
-		btrfs_leak_debug_del(&state->leak_list);
+		btrfs_leak_debug_del(&leak_lock, &state->leak_list);
 		trace_free_extent_state(state, _RET_IP_);
 		kmem_cache_free(extent_state_cache, state);
 	}
@@ -1043,6 +1072,16 @@ hit_next:
 			goto out;
 		}
 
+		/*
+		 * If this extent already has all the bits we want set, then
+		 * skip it, not necessary to split it or do anything with it.
+		 */
+		if ((state->state & bits) == bits) {
+			start = state->end + 1;
+			cache_state(state, cached_state);
+			goto search_again;
+		}
+
 		prealloc = alloc_extent_state_atomic(prealloc);
 		BUG_ON(!prealloc);
 		err = split_state(tree, state, prealloc, start);
@@ -1558,6 +1597,43 @@ out:
 }
 
 /**
+ * find_contiguous_extent_bit: find a contiguous area of bits
+ * @tree - io tree to check
+ * @start - offset to start the search from
+ * @start_ret - the first offset we found with the bits set
+ * @end_ret - the final contiguous range of the bits that were set
+ * @bits - bits to look for
+ *
+ * set_extent_bit and clear_extent_bit can temporarily split contiguous ranges
+ * to set bits appropriately, and then merge them again.  During this time it
+ * will drop the tree->lock, so use this helper if you want to find the actual
+ * contiguous area for given bits.  We will search to the first bit we find, and
+ * then walk down the tree until we find a non-contiguous area.  The area
+ * returned will be the full contiguous area with the bits set.
+ */
+int find_contiguous_extent_bit(struct extent_io_tree *tree, u64 start,
+			       u64 *start_ret, u64 *end_ret, unsigned bits)
+{
+	struct extent_state *state;
+	int ret = 1;
+
+	spin_lock(&tree->lock);
+	state = find_first_extent_bit_state(tree, start, bits);
+	if (state) {
+		*start_ret = state->start;
+		*end_ret = state->end;
+		while ((state = next_state(state)) != NULL) {
+			if (state->start > (*end_ret + 1))
+				break;
+			*end_ret = state->end;
+		}
+		ret = 0;
+	}
+	spin_unlock(&tree->lock);
+	return ret;
+}
+
+/**
  * find_first_clear_extent_bit - find the first range that has @bits not set.
  * This range could start before @start.
  *
@@ -1583,21 +1659,25 @@ void find_first_clear_extent_bit(struct extent_io_tree *tree, u64 start,
 	/* Find first extent with bits cleared */
 	while (1) {
 		node = __etree_search(tree, start, &next, &prev, NULL, NULL);
-		if (!node) {
+		if (!node && !next && !prev) {
+			/*
+			 * Tree is completely empty, send full range and let
+			 * caller deal with it
+			 */
+			*start_ret = 0;
+			*end_ret = -1;
+			goto out;
+		} else if (!node && !next) {
+			/*
+			 * We are past the last allocated chunk, set start at
+			 * the end of the last extent.
+			 */
+			state = rb_entry(prev, struct extent_state, rb_node);
+			*start_ret = state->end + 1;
+			*end_ret = -1;
+			goto out;
+		} else if (!node) {
 			node = next;
-			if (!node) {
-				/*
-				 * We are past the last allocated chunk,
-				 * set start at the end of the last extent. The
-				 * device alloc tree should never be empty so
-				 * prev is always set.
-				 */
-				ASSERT(prev);
-				state = rb_entry(prev, struct extent_state, rb_node);
-				*start_ret = state->end + 1;
-				*end_ret = -1;
-				goto out;
-			}
 		}
 		/*
 		 * At this point 'node' either contains 'start' or start is
@@ -1676,9 +1756,9 @@ out:
  *
  * true is returned if we find something, false if nothing was in the tree
  */
-static noinline bool find_delalloc_range(struct extent_io_tree *tree,
-					u64 *start, u64 *end, u64 max_bytes,
-					struct extent_state **cached_state)
+bool btrfs_find_delalloc_range(struct extent_io_tree *tree, u64 *start,
+			       u64 *end, u64 max_bytes,
+			       struct extent_state **cached_state)
 {
 	struct rb_node *node;
 	struct extent_state *state;
@@ -1796,8 +1876,8 @@ again:
 	/* step one, find a bunch of delalloc bytes starting at start */
 	delalloc_start = *start;
 	delalloc_end = 0;
-	found = find_delalloc_range(tree, &delalloc_start, &delalloc_end,
-				    max_bytes, &cached_state);
+	found = btrfs_find_delalloc_range(tree, &delalloc_start, &delalloc_end,
+					  max_bytes, &cached_state);
 	if (!found || delalloc_end <= *start) {
 		*start = delalloc_start;
 		*end = delalloc_end;
@@ -1899,7 +1979,7 @@ static int __process_pages_contig(struct address_space *mapping,
 			if (page_ops & PAGE_SET_PRIVATE2)
 				SetPagePrivate2(pages[i]);
 
-			if (pages[i] == locked_page) {
+			if (locked_page && pages[i] == locked_page) {
 				put_page(pages[i]);
 				pages_locked++;
 				continue;
@@ -2014,8 +2094,8 @@ out:
  * set the private field for a given byte offset in the tree.  If there isn't
  * an extent_state there already, this does nothing.
  */
-static noinline int set_state_failrec(struct extent_io_tree *tree, u64 start,
-		struct io_failure_record *failrec)
+int set_state_failrec(struct extent_io_tree *tree, u64 start,
+		      struct io_failure_record *failrec)
 {
 	struct rb_node *node;
 	struct extent_state *state;
@@ -2042,8 +2122,8 @@ out:
 	return ret;
 }
 
-static noinline int get_state_failrec(struct extent_io_tree *tree, u64 start,
-		struct io_failure_record **failrec)
+int get_state_failrec(struct extent_io_tree *tree, u64 start,
+		      struct io_failure_record **failrec)
 {
 	struct rb_node *node;
 	struct extent_state *state;
@@ -2534,7 +2614,6 @@ struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
 	bio = btrfs_io_bio_alloc(1);
 	bio->bi_end_io = endio_func;
 	bio->bi_iter.bi_sector = failrec->logical >> 9;
-	bio_set_dev(bio, fs_info->fs_devices->latest_bdev);
 	bio->bi_iter.bi_size = 0;
 	bio->bi_private = data;
 
@@ -2913,25 +2992,22 @@ struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size)
 
 /*
  * @opf:	bio REQ_OP_* and REQ_* flags as one value
- * @tree:	tree so we can call our merge_bio hook
  * @wbc:	optional writeback control for io accounting
  * @page:	page to add to the bio
  * @pg_offset:	offset of the new bio or to check whether we are adding
  *              a contiguous page to the previous one
  * @size:	portion of page that we want to write
  * @offset:	starting offset in the page
- * @bdev:	attach newly created bios to this bdev
  * @bio_ret:	must be valid pointer, newly allocated bio will be stored there
  * @end_io_func:     end_io callback for new bio
  * @mirror_num:	     desired mirror to read/write
  * @prev_bio_flags:  flags of previous bio to see if we can merge the current one
  * @bio_flags:	flags of the current bio to see if we can merge them
  */
-static int submit_extent_page(unsigned int opf, struct extent_io_tree *tree,
+static int submit_extent_page(unsigned int opf,
 			      struct writeback_control *wbc,
 			      struct page *page, u64 offset,
 			      size_t size, unsigned long pg_offset,
-			      struct block_device *bdev,
 			      struct bio **bio_ret,
 			      bio_end_io_t end_io_func,
 			      int mirror_num,
@@ -2943,6 +3019,7 @@ static int submit_extent_page(unsigned int opf, struct extent_io_tree *tree,
 	struct bio *bio;
 	size_t page_size = min_t(size_t, size, PAGE_SIZE);
 	sector_t sector = offset >> 9;
+	struct extent_io_tree *tree = &BTRFS_I(page->mapping->host)->io_tree;
 
 	ASSERT(bio_ret);
 
@@ -2977,13 +3054,16 @@ static int submit_extent_page(unsigned int opf, struct extent_io_tree *tree,
 	}
 
 	bio = btrfs_bio_alloc(offset);
-	bio_set_dev(bio, bdev);
 	bio_add_page(bio, page, page_size, pg_offset);
 	bio->bi_end_io = end_io_func;
 	bio->bi_private = tree;
 	bio->bi_write_hint = page->mapping->host->i_write_hint;
 	bio->bi_opf = opf;
 	if (wbc) {
+		struct block_device *bdev;
+
+		bdev = BTRFS_I(page->mapping->host)->root->fs_info->fs_devices->latest_bdev;
+		bio_set_dev(bio, bdev);
 		wbc_init_bio(wbc, bio);
 		wbc_account_cgroup_owner(wbc, page, page_size);
 	}
@@ -2996,22 +3076,16 @@ static int submit_extent_page(unsigned int opf, struct extent_io_tree *tree,
 static void attach_extent_buffer_page(struct extent_buffer *eb,
 				      struct page *page)
 {
-	if (!PagePrivate(page)) {
-		SetPagePrivate(page);
-		get_page(page);
-		set_page_private(page, (unsigned long)eb);
-	} else {
+	if (!PagePrivate(page))
+		attach_page_private(page, eb);
+	else
 		WARN_ON(page->private != (unsigned long)eb);
-	}
 }
 
 void set_page_extent_mapped(struct page *page)
 {
-	if (!PagePrivate(page)) {
-		SetPagePrivate(page);
-		get_page(page);
-		set_page_private(page, EXTENT_PAGE_PRIVATE);
-	}
+	if (!PagePrivate(page))
+		attach_page_private(page, (void *)EXTENT_PAGE_PRIVATE);
 }
 
 static struct extent_map *
@@ -3033,7 +3107,7 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
 		*em_cached = NULL;
 	}
 
-	em = get_extent(BTRFS_I(inode), page, pg_offset, start, len, 0);
+	em = get_extent(BTRFS_I(inode), page, pg_offset, start, len);
 	if (em_cached && !IS_ERR_OR_NULL(em)) {
 		BUG_ON(*em_cached);
 		refcount_inc(&em->refs);
@@ -3048,8 +3122,7 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
  * XXX JDM: This needs looking at to ensure proper page locking
  * return 0 on success, otherwise return error
  */
-static int __do_readpage(struct extent_io_tree *tree,
-			 struct page *page,
+static int __do_readpage(struct page *page,
 			 get_extent_t *get_extent,
 			 struct extent_map **em_cached,
 			 struct bio **bio, int mirror_num,
@@ -3065,7 +3138,6 @@ static int __do_readpage(struct extent_io_tree *tree,
 	u64 block_start;
 	u64 cur_end;
 	struct extent_map *em;
-	struct block_device *bdev;
 	int ret = 0;
 	int nr = 0;
 	size_t pg_offset = 0;
@@ -3073,6 +3145,7 @@ static int __do_readpage(struct extent_io_tree *tree,
 	size_t disk_io_size;
 	size_t blocksize = inode->i_sb->s_blocksize;
 	unsigned long this_bio_flag = 0;
+	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
 
 	set_page_extent_mapped(page);
 
@@ -3142,7 +3215,6 @@ static int __do_readpage(struct extent_io_tree *tree,
 			offset = em->block_start + extent_offset;
 			disk_io_size = iosize;
 		}
-		bdev = em->bdev;
 		block_start = em->block_start;
 		if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
 			block_start = EXTENT_MAP_HOLE;
@@ -3230,9 +3302,9 @@ static int __do_readpage(struct extent_io_tree *tree,
 			continue;
 		}
 
-		ret = submit_extent_page(REQ_OP_READ | read_flags, tree, NULL,
+		ret = submit_extent_page(REQ_OP_READ | read_flags, NULL,
 					 page, offset, disk_io_size,
-					 pg_offset, bdev, bio,
+					 pg_offset, bio,
 					 end_bio_extent_readpage, mirror_num,
 					 *bio_flags,
 					 this_bio_flag,
@@ -3257,8 +3329,7 @@ out:
 	return ret;
 }
 
-static inline void contiguous_readpages(struct extent_io_tree *tree,
-					     struct page *pages[], int nr_pages,
+static inline void contiguous_readpages(struct page *pages[], int nr_pages,
 					     u64 start, u64 end,
 					     struct extent_map **em_cached,
 					     struct bio **bio,
@@ -3268,17 +3339,16 @@ static inline void contiguous_readpages(struct extent_io_tree *tree,
 	struct btrfs_inode *inode = BTRFS_I(pages[0]->mapping->host);
 	int index;
 
-	btrfs_lock_and_flush_ordered_range(tree, inode, start, end, NULL);
+	btrfs_lock_and_flush_ordered_range(inode, start, end, NULL);
 
 	for (index = 0; index < nr_pages; index++) {
-		__do_readpage(tree, pages[index], btrfs_get_extent, em_cached,
+		__do_readpage(pages[index], btrfs_get_extent, em_cached,
 				bio, 0, bio_flags, REQ_RAHEAD, prev_em_start);
 		put_page(pages[index]);
 	}
 }
 
-static int __extent_read_full_page(struct extent_io_tree *tree,
-				   struct page *page,
+static int __extent_read_full_page(struct page *page,
 				   get_extent_t *get_extent,
 				   struct bio **bio, int mirror_num,
 				   unsigned long *bio_flags,
@@ -3289,21 +3359,21 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
 	u64 end = start + PAGE_SIZE - 1;
 	int ret;
 
-	btrfs_lock_and_flush_ordered_range(tree, inode, start, end, NULL);
+	btrfs_lock_and_flush_ordered_range(inode, start, end, NULL);
 
-	ret = __do_readpage(tree, page, get_extent, NULL, bio, mirror_num,
+	ret = __do_readpage(page, get_extent, NULL, bio, mirror_num,
 			    bio_flags, read_flags, NULL);
 	return ret;
 }
 
-int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
-			    get_extent_t *get_extent, int mirror_num)
+int extent_read_full_page(struct page *page, get_extent_t *get_extent,
+			  int mirror_num)
 {
 	struct bio *bio = NULL;
 	unsigned long bio_flags = 0;
 	int ret;
 
-	ret = __extent_read_full_page(tree, page, get_extent, &bio, mirror_num,
+	ret = __extent_read_full_page(page, get_extent, &bio, mirror_num,
 				      &bio_flags, 0);
 	if (bio)
 		ret = submit_one_bio(bio, mirror_num, bio_flags);
@@ -3409,9 +3479,9 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 				 struct extent_page_data *epd,
 				 loff_t i_size,
 				 unsigned long nr_written,
-				 unsigned int write_flags, int *nr_ret)
+				 int *nr_ret)
 {
-	struct extent_io_tree *tree = epd->tree;
+	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
 	u64 start = page_offset(page);
 	u64 page_end = start + PAGE_SIZE - 1;
 	u64 end;
@@ -3420,21 +3490,17 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 	u64 block_start;
 	u64 iosize;
 	struct extent_map *em;
-	struct block_device *bdev;
 	size_t pg_offset = 0;
 	size_t blocksize;
 	int ret = 0;
 	int nr = 0;
+	const unsigned int write_flags = wbc_to_write_flags(wbc);
 	bool compressed;
 
 	ret = btrfs_writepage_cow_fixup(page, start, page_end);
 	if (ret) {
 		/* Fixup worker will requeue */
-		if (ret == -EBUSY)
-			wbc->pages_skipped++;
-		else
-			redirty_page_for_writepage(wbc, page);
-
+		redirty_page_for_writepage(wbc, page);
 		update_nr_written(wbc, nr_written);
 		unlock_page(page);
 		return 1;
@@ -3447,11 +3513,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 	update_nr_written(wbc, nr_written + 1);
 
 	end = page_end;
-	if (i_size <= start) {
-		btrfs_writepage_endio_finish_ordered(page, start, page_end, 1);
-		goto done;
-	}
-
 	blocksize = inode->i_sb->s_blocksize;
 
 	while (cur <= end) {
@@ -3463,8 +3524,8 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 							     page_end, 1);
 			break;
 		}
-		em = btrfs_get_extent(BTRFS_I(inode), page, pg_offset, cur,
-				     end - cur + 1, 1);
+		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur,
+				      end - cur + 1);
 		if (IS_ERR_OR_NULL(em)) {
 			SetPageError(page);
 			ret = PTR_ERR_OR_ZERO(em);
@@ -3478,7 +3539,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 		iosize = min(em_end - cur, end - cur + 1);
 		iosize = ALIGN(iosize, blocksize);
 		offset = em->block_start + extent_offset;
-		bdev = em->bdev;
 		block_start = em->block_start;
 		compressed = test_bit(EXTENT_FLAG_COMPRESSED, &em->flags);
 		free_extent_map(em);
@@ -3490,22 +3550,11 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 		 */
 		if (compressed || block_start == EXTENT_MAP_HOLE ||
 		    block_start == EXTENT_MAP_INLINE) {
-			/*
-			 * end_io notification does not happen here for
-			 * compressed extents
-			 */
-			if (!compressed)
-				btrfs_writepage_endio_finish_ordered(page, cur,
-							    cur + iosize - 1,
-							    1);
-			else if (compressed) {
-				/* we don't want to end_page_writeback on
-				 * a compressed extent.  this happens
-				 * elsewhere
-				 */
+			if (compressed)
 				nr++;
-			}
-
+			else
+				btrfs_writepage_endio_finish_ordered(page, cur,
+							cur + iosize - 1, 1);
 			cur += iosize;
 			pg_offset += iosize;
 			continue;
@@ -3518,9 +3567,9 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 			       page->index, cur, end);
 		}
 
-		ret = submit_extent_page(REQ_OP_WRITE | write_flags, tree, wbc,
+		ret = submit_extent_page(REQ_OP_WRITE | write_flags, wbc,
 					 page, offset, iosize, pg_offset,
-					 bdev, &epd->bio,
+					 &epd->bio,
 					 end_bio_extent_writepage,
 					 0, 0, 0, false);
 		if (ret) {
@@ -3533,7 +3582,6 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 		pg_offset += iosize;
 		nr++;
 	}
-done:
 	*nr_ret = nr;
 	return ret;
 }
@@ -3555,14 +3603,11 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc,
 	u64 page_end = start + PAGE_SIZE - 1;
 	int ret;
 	int nr = 0;
-	size_t pg_offset = 0;
+	size_t pg_offset;
 	loff_t i_size = i_size_read(inode);
 	unsigned long end_index = i_size >> PAGE_SHIFT;
-	unsigned int write_flags = 0;
 	unsigned long nr_written = 0;
 
-	write_flags = wbc_to_write_flags(wbc);
-
 	trace___extent_writepage(page, inode, wbc);
 
 	WARN_ON(!PageLocked(page));
@@ -3587,22 +3632,20 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc,
 		flush_dcache_page(page);
 	}
 
-	pg_offset = 0;
-
 	set_page_extent_mapped(page);
 
 	if (!epd->extent_locked) {
 		ret = writepage_delalloc(inode, page, wbc, start, &nr_written);
 		if (ret == 1)
-			goto done_unlocked;
+			return 0;
 		if (ret)
 			goto done;
 	}
 
 	ret = __extent_writepage_io(inode, page, wbc, epd,
-				    i_size, nr_written, write_flags, &nr);
+				    i_size, nr_written, &nr);
 	if (ret == 1)
-		goto done_unlocked;
+		return 0;
 
 done:
 	if (nr == 0) {
@@ -3617,9 +3660,6 @@ done:
 	unlock_page(page);
 	ASSERT(ret <= 0);
 	return ret;
-
-done_unlocked:
-	return 0;
 }
 
 void wait_on_extent_buffer_writeback(struct extent_buffer *eb)
@@ -3848,9 +3888,6 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
 			struct writeback_control *wbc,
 			struct extent_page_data *epd)
 {
-	struct btrfs_fs_info *fs_info = eb->fs_info;
-	struct block_device *bdev = fs_info->fs_devices->latest_bdev;
-	struct extent_io_tree *tree = &BTRFS_I(fs_info->btree_inode)->io_tree;
 	u64 offset = eb->start;
 	u32 nritems;
 	int i, num_pages;
@@ -3883,8 +3920,8 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
 
 		clear_page_dirty_for_io(p);
 		set_page_writeback(p);
-		ret = submit_extent_page(REQ_OP_WRITE | write_flags, tree, wbc,
-					 p, offset, PAGE_SIZE, 0, bdev,
+		ret = submit_extent_page(REQ_OP_WRITE | write_flags, wbc,
+					 p, offset, PAGE_SIZE, 0,
 					 &epd->bio,
 					 end_bio_extent_buffer_writepage,
 					 0, 0, 0, false);
@@ -3916,14 +3953,13 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
 int btree_write_cache_pages(struct address_space *mapping,
 				   struct writeback_control *wbc)
 {
-	struct extent_io_tree *tree = &BTRFS_I(mapping->host)->io_tree;
 	struct extent_buffer *eb, *prev_eb = NULL;
 	struct extent_page_data epd = {
 		.bio = NULL,
-		.tree = tree,
 		.extent_locked = 0,
 		.sync_io = wbc->sync_mode == WB_SYNC_ALL,
 	};
+	struct btrfs_fs_info *fs_info = BTRFS_I(mapping->host)->root->fs_info;
 	int ret = 0;
 	int done = 0;
 	int nr_to_write_done = 0;
@@ -3938,6 +3974,11 @@ int btree_write_cache_pages(struct address_space *mapping,
 	if (wbc->range_cyclic) {
 		index = mapping->writeback_index; /* Start from prev offset */
 		end = -1;
+		/*
+		 * Start from the beginning does not need to cycle over the
+		 * range, mark it as scanned.
+		 */
+		scanned = (index == 0);
 	} else {
 		index = wbc->range_start >> PAGE_SHIFT;
 		end = wbc->range_end >> PAGE_SHIFT;
@@ -3955,7 +3996,6 @@ retry:
 			tag))) {
 		unsigned i;
 
-		scanned = 1;
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
 
@@ -4033,7 +4073,39 @@ retry:
 		end_write_bio(&epd, ret);
 		return ret;
 	}
-	ret = flush_write_bio(&epd);
+	/*
+	 * If something went wrong, don't allow any metadata write bio to be
+	 * submitted.
+	 *
+	 * This would prevent use-after-free if we had dirty pages not
+	 * cleaned up, which can still happen by fuzzed images.
+	 *
+	 * - Bad extent tree
+	 *   Allowing existing tree block to be allocated for other trees.
+	 *
+	 * - Log tree operations
+	 *   Exiting tree blocks get allocated to log tree, bumps its
+	 *   generation, then get cleaned in tree re-balance.
+	 *   Such tree block will not be written back, since it's clean,
+	 *   thus no WRITTEN flag set.
+	 *   And after log writes back, this tree block is not traced by
+	 *   any dirty extent_io_tree.
+	 *
+	 * - Offending tree block gets re-dirtied from its original owner
+	 *   Since it has bumped generation, no WRITTEN flag, it can be
+	 *   reused without COWing. This tree block will not be traced
+	 *   by btrfs_transaction::dirty_pages.
+	 *
+	 *   Now such dirty tree block will not be cleaned by any dirty
+	 *   extent io tree. Thus we don't want to submit such wild eb
+	 *   if the fs already has error.
+	 */
+	if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
+		ret = flush_write_bio(&epd);
+	} else {
+		ret = -EUCLEAN;
+		end_write_bio(&epd, ret);
+	}
 	return ret;
 }
 
@@ -4084,6 +4156,11 @@ static int extent_write_cache_pages(struct address_space *mapping,
 	if (wbc->range_cyclic) {
 		index = mapping->writeback_index; /* Start from prev offset */
 		end = -1;
+		/*
+		 * Start from the beginning does not need to cycle over the
+		 * range, mark it as scanned.
+		 */
+		scanned = (index == 0);
 	} else {
 		index = wbc->range_start >> PAGE_SHIFT;
 		end = wbc->range_end >> PAGE_SHIFT;
@@ -4117,11 +4194,10 @@ retry:
 						&index, end, tag))) {
 		unsigned i;
 
-		scanned = 1;
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
 
-			done_index = page->index;
+			done_index = page->index + 1;
 			/*
 			 * At this point we hold neither the i_pages lock nor
 			 * the page lock: the page may be truncated or
@@ -4156,16 +4232,6 @@ retry:
 
 			ret = __extent_writepage(page, wbc, epd);
 			if (ret < 0) {
-				/*
-				 * done_index is set past this page,
-				 * so media errors will not choke
-				 * background writeout for the entire
-				 * file. This has consequences for
-				 * range_cyclic semantics (ie. it may
-				 * not be suitable for data integrity
-				 * writeout).
-				 */
-				done_index = page->index + 1;
 				done = 1;
 				break;
 			}
@@ -4187,7 +4253,16 @@ retry:
 		 */
 		scanned = 1;
 		index = 0;
-		goto retry;
+
+		/*
+		 * If we're looping we could run into a page that is locked by a
+		 * writer and that writer could be waiting on writeback for a
+		 * page in our current bio, and thus deadlock, so flush the
+		 * write bio here.
+		 */
+		ret = flush_write_bio(epd);
+		if (!ret)
+			goto retry;
 	}
 
 	if (wbc->range_cyclic || (wbc->nr_to_write > 0 && range_whole))
@@ -4202,7 +4277,6 @@ int extent_write_full_page(struct page *page, struct writeback_control *wbc)
 	int ret;
 	struct extent_page_data epd = {
 		.bio = NULL,
-		.tree = &BTRFS_I(page->mapping->host)->io_tree,
 		.extent_locked = 0,
 		.sync_io = wbc->sync_mode == WB_SYNC_ALL,
 	};
@@ -4224,14 +4298,12 @@ int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 {
 	int ret = 0;
 	struct address_space *mapping = inode->i_mapping;
-	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
 	struct page *page;
 	unsigned long nr_pages = (end - start + PAGE_SIZE) >>
 		PAGE_SHIFT;
 
 	struct extent_page_data epd = {
 		.bio = NULL,
-		.tree = tree,
 		.extent_locked = 1,
 		.sync_io = mode == WB_SYNC_ALL,
 	};
@@ -4240,8 +4312,12 @@ int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 		.nr_to_write	= nr_pages * 2,
 		.range_start	= start,
 		.range_end	= end + 1,
+		/* We're called from an async helper function */
+		.punt_to_cgroup	= 1,
+		.no_cgroup_owner = 1,
 	};
 
+	wbc_attach_fdatawrite_inode(&wbc_writepages, inode);
 	while (start <= end) {
 		page = find_get_page(mapping, start >> PAGE_SHIFT);
 		if (clear_page_dirty_for_io(page))
@@ -4256,11 +4332,12 @@ int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 	}
 
 	ASSERT(ret <= 0);
-	if (ret < 0) {
+	if (ret == 0)
+		ret = flush_write_bio(&epd);
+	else
 		end_write_bio(&epd, ret);
-		return ret;
-	}
-	ret = flush_write_bio(&epd);
+
+	wbc_detach_inode(&wbc_writepages);
 	return ret;
 }
 
@@ -4270,7 +4347,6 @@ int extent_writepages(struct address_space *mapping,
 	int ret = 0;
 	struct extent_page_data epd = {
 		.bio = NULL,
-		.tree = &BTRFS_I(mapping->host)->io_tree,
 		.extent_locked = 0,
 		.sync_io = wbc->sync_mode == WB_SYNC_ALL,
 	};
@@ -4285,52 +4361,32 @@ int extent_writepages(struct address_space *mapping,
 	return ret;
 }
 
-int extent_readpages(struct address_space *mapping, struct list_head *pages,
-		     unsigned nr_pages)
+void extent_readahead(struct readahead_control *rac)
 {
 	struct bio *bio = NULL;
 	unsigned long bio_flags = 0;
 	struct page *pagepool[16];
 	struct extent_map *em_cached = NULL;
-	struct extent_io_tree *tree = &BTRFS_I(mapping->host)->io_tree;
-	int nr = 0;
 	u64 prev_em_start = (u64)-1;
+	int nr;
 
-	while (!list_empty(pages)) {
-		u64 contig_end = 0;
+	while ((nr = readahead_page_batch(rac, pagepool))) {
+		u64 contig_start = page_offset(pagepool[0]);
+		u64 contig_end = page_offset(pagepool[nr - 1]) + PAGE_SIZE - 1;
 
-		for (nr = 0; nr < ARRAY_SIZE(pagepool) && !list_empty(pages);) {
-			struct page *page = lru_to_page(pages);
+		ASSERT(contig_start + nr * PAGE_SIZE - 1 == contig_end);
 
-			prefetchw(&page->flags);
-			list_del(&page->lru);
-			if (add_to_page_cache_lru(page, mapping, page->index,
-						readahead_gfp_mask(mapping))) {
-				put_page(page);
-				break;
-			}
-
-			pagepool[nr++] = page;
-			contig_end = page_offset(page) + PAGE_SIZE - 1;
-		}
-
-		if (nr) {
-			u64 contig_start = page_offset(pagepool[0]);
-
-			ASSERT(contig_start + nr * PAGE_SIZE - 1 == contig_end);
-
-			contiguous_readpages(tree, pagepool, nr, contig_start,
-				     contig_end, &em_cached, &bio, &bio_flags,
-				     &prev_em_start);
-		}
+		contiguous_readpages(pagepool, nr, contig_start, contig_end,
+				&em_cached, &bio, &bio_flags, &prev_em_start);
 	}
 
 	if (em_cached)
 		free_extent_map(em_cached);
 
-	if (bio)
-		return submit_one_bio(bio, 0, bio_flags);
-	return 0;
+	if (bio) {
+		if (submit_one_bio(bio, 0, bio_flags))
+			return;
+	}
 }
 
 /*
@@ -4803,7 +4859,6 @@ out_free_ulist:
 
 static void __free_extent_buffer(struct extent_buffer *eb)
 {
-	btrfs_leak_debug_del(&eb->leak_list);
 	kmem_cache_free(extent_buffer_cache, eb);
 }
 
@@ -4849,10 +4904,7 @@ static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb)
 			 * We need to make sure we haven't be attached
 			 * to a new eb.
 			 */
-			ClearPagePrivate(page);
-			set_page_private(page, 0);
-			/* One for the page private */
-			put_page(page);
+			detach_page_private(page);
 		}
 
 		if (mapped)
@@ -4869,6 +4921,7 @@ static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb)
 static inline void btrfs_release_extent_buffer(struct extent_buffer *eb)
 {
 	btrfs_release_extent_buffer_pages(eb);
+	btrfs_leak_debug_del(&eb->fs_info->eb_leak_lock, &eb->leak_list);
 	__free_extent_buffer(eb);
 }
 
@@ -4890,7 +4943,8 @@ __alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start,
 	init_waitqueue_head(&eb->write_lock_wq);
 	init_waitqueue_head(&eb->read_lock_wq);
 
-	btrfs_leak_debug_add(&eb->leak_list, &buffers);
+	btrfs_leak_debug_add(&fs_info->eb_leak_lock, &eb->leak_list,
+			     &fs_info->allocated_ebs);
 
 	spin_lock_init(&eb->refs_lock);
 	atomic_set(&eb->refs, 1);
@@ -5076,12 +5130,14 @@ struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
 		return eb;
 	eb = alloc_dummy_extent_buffer(fs_info, start);
 	if (!eb)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 	eb->fs_info = fs_info;
 again:
 	ret = radix_tree_preload(GFP_NOFS);
-	if (ret)
+	if (ret) {
+		exists = ERR_PTR(ret);
 		goto free_eb;
+	}
 	spin_lock(&fs_info->buffer_lock);
 	ret = radix_tree_insert(&fs_info->buffer_radix,
 				start >> PAGE_SHIFT, eb);
@@ -5235,6 +5291,7 @@ static inline void btrfs_release_extent_buffer_rcu(struct rcu_head *head)
 }
 
 static int release_extent_buffer(struct extent_buffer *eb)
+	__releases(&eb->refs_lock)
 {
 	lockdep_assert_held(&eb->refs_lock);
 
@@ -5253,6 +5310,7 @@ static int release_extent_buffer(struct extent_buffer *eb)
 			spin_unlock(&eb->refs_lock);
 		}
 
+		btrfs_leak_debug_del(&eb->fs_info->eb_leak_lock, &eb->leak_list);
 		/* Should be safe to release our pages at this point */
 		btrfs_release_extent_buffer_pages(eb);
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
@@ -5410,7 +5468,6 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num)
 	unsigned long num_reads = 0;
 	struct bio *bio = NULL;
 	unsigned long bio_flags = 0;
-	struct extent_io_tree *tree = &BTRFS_I(eb->fs_info->btree_inode)->io_tree;
 
 	if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags))
 		return 0;
@@ -5458,7 +5515,7 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num)
 			}
 
 			ClearPageError(page);
-			err = __extent_read_full_page(tree, page,
+			err = __extent_read_full_page(page,
 						      btree_get_extent, &bio,
 						      mirror_num, &bio_flags,
 						      REQ_META);
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index cf3424d58fec..25594e09fdcd 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -7,35 +7,6 @@
 #include <linux/refcount.h>
 #include "ulist.h"
 
-/* bits for the extent state */
-#define EXTENT_DIRTY		(1U << 0)
-#define EXTENT_UPTODATE		(1U << 1)
-#define EXTENT_LOCKED		(1U << 2)
-#define EXTENT_NEW		(1U << 3)
-#define EXTENT_DELALLOC		(1U << 4)
-#define EXTENT_DEFRAG		(1U << 5)
-#define EXTENT_BOUNDARY		(1U << 6)
-#define EXTENT_NODATASUM	(1U << 7)
-#define EXTENT_CLEAR_META_RESV	(1U << 8)
-#define EXTENT_NEED_WAIT	(1U << 9)
-#define EXTENT_DAMAGED		(1U << 10)
-#define EXTENT_NORESERVE	(1U << 11)
-#define EXTENT_QGROUP_RESERVED	(1U << 12)
-#define EXTENT_CLEAR_DATA_RESV	(1U << 13)
-#define EXTENT_DELALLOC_NEW	(1U << 14)
-#define EXTENT_DO_ACCOUNTING    (EXTENT_CLEAR_META_RESV | \
-				 EXTENT_CLEAR_DATA_RESV)
-#define EXTENT_CTLBITS		(EXTENT_DO_ACCOUNTING)
-
-/*
- * Redefined bits above which are used only in the device allocation tree,
- * shouldn't be using EXTENT_LOCKED / EXTENT_BOUNDARY / EXTENT_CLEAR_META_RESV
- * / EXTENT_CLEAR_DATA_RESV because they have special meaning to the bit
- * manipulation functions
- */
-#define CHUNK_ALLOCATED EXTENT_DIRTY
-#define CHUNK_TRIMMED   EXTENT_DEFRAG
-
 /*
  * flags for bio submission. The high bits indicate the compression
  * type for this bio
@@ -89,12 +60,11 @@ enum {
 #define BITMAP_LAST_BYTE_MASK(nbits) \
 	(BYTE_MASK >> (-(nbits) & (BITS_PER_BYTE - 1)))
 
-struct extent_state;
 struct btrfs_root;
 struct btrfs_inode;
 struct btrfs_io_bio;
 struct io_failure_record;
-
+struct extent_io_tree;
 
 typedef blk_status_t (extent_submit_bio_start_t)(void *private_data,
 		struct bio *bio, u64 bio_offset);
@@ -111,47 +81,6 @@ struct extent_io_ops {
 				    int mirror);
 };
 
-enum {
-	IO_TREE_FS_INFO_FREED_EXTENTS0,
-	IO_TREE_FS_INFO_FREED_EXTENTS1,
-	IO_TREE_INODE_IO,
-	IO_TREE_INODE_IO_FAILURE,
-	IO_TREE_RELOC_BLOCKS,
-	IO_TREE_TRANS_DIRTY_PAGES,
-	IO_TREE_ROOT_DIRTY_LOG_PAGES,
-	IO_TREE_SELFTEST,
-};
-
-struct extent_io_tree {
-	struct rb_root state;
-	struct btrfs_fs_info *fs_info;
-	void *private_data;
-	u64 dirty_bytes;
-	bool track_uptodate;
-
-	/* Who owns this io tree, should be one of IO_TREE_* */
-	u8 owner;
-
-	spinlock_t lock;
-	const struct extent_io_ops *ops;
-};
-
-struct extent_state {
-	u64 start;
-	u64 end; /* inclusive */
-	struct rb_node rb_node;
-
-	/* ADD NEW ELEMENTS AFTER THIS */
-	wait_queue_head_t wq;
-	refcount_t refs;
-	unsigned state;
-
-	struct io_failure_record *failrec;
-
-#ifdef CONFIG_BTRFS_DEBUG
-	struct list_head leak_list;
-#endif
-};
 
 #define INLINE_EXTENT_BUFFER_PAGES 16
 #define MAX_INLINE_EXTENT_BUFFER_SIZE (INLINE_EXTENT_BUFFER_PAGES * PAGE_SIZE)
@@ -254,157 +183,14 @@ static inline int extent_compress_type(unsigned long bio_flags)
 struct extent_map_tree;
 
 typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode,
-					  struct page *page,
-					  size_t pg_offset,
-					  u64 start, u64 len,
-					  int create);
-
-void extent_io_tree_init(struct btrfs_fs_info *fs_info,
-			 struct extent_io_tree *tree, unsigned int owner,
-			 void *private_data);
-void extent_io_tree_release(struct extent_io_tree *tree);
+					  struct page *page, size_t pg_offset,
+					  u64 start, u64 len);
+
 int try_release_extent_mapping(struct page *page, gfp_t mask);
 int try_release_extent_buffer(struct page *page);
-int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
-		     struct extent_state **cached);
-
-static inline int lock_extent(struct extent_io_tree *tree, u64 start, u64 end)
-{
-	return lock_extent_bits(tree, start, end, NULL);
-}
-
-int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
-int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
-			  get_extent_t *get_extent, int mirror_num);
-int __init extent_io_init(void);
-void __cold extent_io_exit(void);
-
-u64 count_range_bits(struct extent_io_tree *tree,
-		     u64 *start, u64 search_end,
-		     u64 max_bytes, unsigned bits, int contig);
-
-void free_extent_state(struct extent_state *state);
-int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		   unsigned bits, int filled,
-		   struct extent_state *cached_state);
-int clear_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
-		unsigned bits, struct extent_changeset *changeset);
-int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		     unsigned bits, int wake, int delete,
-		     struct extent_state **cached);
-int __clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		     unsigned bits, int wake, int delete,
-		     struct extent_state **cached, gfp_t mask,
-		     struct extent_changeset *changeset);
-
-static inline int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end)
-{
-	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, NULL);
-}
-
-static inline int unlock_extent_cached(struct extent_io_tree *tree, u64 start,
-		u64 end, struct extent_state **cached)
-{
-	return __clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
-				GFP_NOFS, NULL);
-}
-
-static inline int unlock_extent_cached_atomic(struct extent_io_tree *tree,
-		u64 start, u64 end, struct extent_state **cached)
-{
-	return __clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
-				GFP_ATOMIC, NULL);
-}
-
-static inline int clear_extent_bits(struct extent_io_tree *tree, u64 start,
-		u64 end, unsigned bits)
-{
-	int wake = 0;
 
-	if (bits & EXTENT_LOCKED)
-		wake = 1;
-
-	return clear_extent_bit(tree, start, end, bits, wake, 0, NULL);
-}
-
-int set_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
-			   unsigned bits, struct extent_changeset *changeset);
-int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		   unsigned bits, u64 *failed_start,
-		   struct extent_state **cached_state, gfp_t mask);
-int set_extent_bits_nowait(struct extent_io_tree *tree, u64 start, u64 end,
-			   unsigned bits);
-
-static inline int set_extent_bits(struct extent_io_tree *tree, u64 start,
-		u64 end, unsigned bits)
-{
-	return set_extent_bit(tree, start, end, bits, NULL, NULL, GFP_NOFS);
-}
-
-static inline int clear_extent_uptodate(struct extent_io_tree *tree, u64 start,
-		u64 end, struct extent_state **cached_state)
-{
-	return __clear_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, 0,
-				cached_state, GFP_NOFS, NULL);
-}
-
-static inline int set_extent_dirty(struct extent_io_tree *tree, u64 start,
-		u64 end, gfp_t mask)
-{
-	return set_extent_bit(tree, start, end, EXTENT_DIRTY, NULL,
-			      NULL, mask);
-}
-
-static inline int clear_extent_dirty(struct extent_io_tree *tree, u64 start,
-				     u64 end, struct extent_state **cached)
-{
-	return clear_extent_bit(tree, start, end,
-				EXTENT_DIRTY | EXTENT_DELALLOC |
-				EXTENT_DO_ACCOUNTING, 0, 0, cached);
-}
-
-int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
-		       unsigned bits, unsigned clear_bits,
-		       struct extent_state **cached_state);
-
-static inline int set_extent_delalloc(struct extent_io_tree *tree, u64 start,
-				      u64 end, unsigned int extra_bits,
-				      struct extent_state **cached_state)
-{
-	return set_extent_bit(tree, start, end,
-			      EXTENT_DELALLOC | EXTENT_UPTODATE | extra_bits,
-			      NULL, cached_state, GFP_NOFS);
-}
-
-static inline int set_extent_defrag(struct extent_io_tree *tree, u64 start,
-		u64 end, struct extent_state **cached_state)
-{
-	return set_extent_bit(tree, start, end,
-			      EXTENT_DELALLOC | EXTENT_UPTODATE | EXTENT_DEFRAG,
-			      NULL, cached_state, GFP_NOFS);
-}
-
-static inline int set_extent_new(struct extent_io_tree *tree, u64 start,
-		u64 end)
-{
-	return set_extent_bit(tree, start, end, EXTENT_NEW, NULL, NULL,
-			GFP_NOFS);
-}
-
-static inline int set_extent_uptodate(struct extent_io_tree *tree, u64 start,
-		u64 end, struct extent_state **cached_state, gfp_t mask)
-{
-	return set_extent_bit(tree, start, end, EXTENT_UPTODATE, NULL,
-			      cached_state, mask);
-}
-
-int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
-			  u64 *start_ret, u64 *end_ret, unsigned bits,
-			  struct extent_state **cached_state);
-void find_first_clear_extent_bit(struct extent_io_tree *tree, u64 start,
-				 u64 *start_ret, u64 *end_ret, unsigned bits);
-int extent_invalidatepage(struct extent_io_tree *tree,
-			  struct page *page, unsigned long offset);
+int extent_read_full_page(struct page *page, get_extent_t *get_extent,
+			  int mirror_num);
 int extent_write_full_page(struct page *page, struct writeback_control *wbc);
 int extent_write_locked_range(struct inode *inode, u64 start, u64 end,
 			      int mode);
@@ -412,8 +198,7 @@ int extent_writepages(struct address_space *mapping,
 		      struct writeback_control *wbc);
 int btree_write_cache_pages(struct address_space *mapping,
 			    struct writeback_control *wbc);
-int extent_readpages(struct address_space *mapping, struct list_head *pages,
-		     unsigned nr_pages);
+void extent_readahead(struct readahead_control *rac);
 int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		__u64 start, __u64 len);
 void set_page_extent_mapped(struct page *page);
@@ -442,11 +227,6 @@ static inline int num_extent_pages(const struct extent_buffer *eb)
 	       (eb->start >> PAGE_SHIFT);
 }
 
-static inline void extent_buffer_get(struct extent_buffer *eb)
-{
-	atomic_inc(&eb->refs);
-}
-
 static inline int extent_buffer_uptodate(struct extent_buffer *eb)
 {
 	return test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
@@ -508,10 +288,6 @@ struct btrfs_inode;
 int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
 		      u64 length, u64 logical, struct page *page,
 		      unsigned int pg_offset, int mirror_num);
-int clean_io_failure(struct btrfs_fs_info *fs_info,
-		     struct extent_io_tree *failure_tree,
-		     struct extent_io_tree *io_tree, u64 start,
-		     struct page *page, u64 ino, unsigned int pg_offset);
 void end_extent_writepage(struct page *page, int err, u64 start, u64 end);
 int btrfs_repair_eb_io_failure(struct extent_buffer *eb, int mirror_num);
 
@@ -535,19 +311,12 @@ struct io_failure_record {
 };
 
 
-void btrfs_free_io_failure_record(struct btrfs_inode *inode, u64 start,
-		u64 end);
-int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
-				struct io_failure_record **failrec_ret);
 bool btrfs_check_repairable(struct inode *inode, unsigned failed_bio_pages,
 			    struct io_failure_record *failrec, int fail_mirror);
 struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
 				    struct io_failure_record *failrec,
 				    struct page *page, int pg_offset, int icsum,
 				    bio_end_io_t *endio_func, void *data);
-int free_io_failure(struct extent_io_tree *failure_tree,
-		    struct extent_io_tree *io_tree,
-		    struct io_failure_record *rec);
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 bool find_lock_delalloc_range(struct inode *inode,
 			     struct page *locked_page, u64 *start,
@@ -556,4 +325,10 @@ bool find_lock_delalloc_range(struct inode *inode,
 struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
 					       u64 start);
 
+#ifdef CONFIG_BTRFS_DEBUG
+void btrfs_extent_buffer_leak_debug_check(struct btrfs_fs_info *fs_info);
+#else
+#define btrfs_extent_buffer_leak_debug_check(fs_info)	do {} while (0)
+#endif
+
 #endif
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 9d30acca55e1..bd6229fb2b6f 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -214,9 +214,13 @@ static int mergable_maps(struct extent_map *prev, struct extent_map *next)
 	ASSERT(next->block_start != EXTENT_MAP_DELALLOC &&
 	       prev->block_start != EXTENT_MAP_DELALLOC);
 
+	if (prev->map_lookup || next->map_lookup)
+		ASSERT(test_bit(EXTENT_FLAG_FS_MAPPING, &prev->flags) &&
+		       test_bit(EXTENT_FLAG_FS_MAPPING, &next->flags));
+
 	if (extent_map_end(prev) == next->start &&
 	    prev->flags == next->flags &&
-	    prev->bdev == next->bdev &&
+	    prev->map_lookup == next->map_lookup &&
 	    ((next->block_start == EXTENT_MAP_HOLE &&
 	      prev->block_start == EXTENT_MAP_HOLE) ||
 	     (next->block_start == EXTENT_MAP_INLINE &&
@@ -233,6 +237,17 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em)
 	struct extent_map *merge = NULL;
 	struct rb_node *rb;
 
+	/*
+	 * We can't modify an extent map that is in the tree and that is being
+	 * used by another task, as it can cause that other task to see it in
+	 * inconsistent state during the merging. We always have 1 reference for
+	 * the tree and 1 for this task (which is unpinning the extent map or
+	 * clearing the logging flag), so anything > 2 means it's being used by
+	 * other tasks too.
+	 */
+	if (refcount_read(&em->refs) > 2)
+		return;
+
 	if (em->start != 0) {
 		rb = rb_prev(&em->rb_node);
 		if (rb)
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index 473f039fcd7c..8e217337dff9 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -42,15 +42,8 @@ struct extent_map {
 	u64 block_len;
 	u64 generation;
 	unsigned long flags;
-	union {
-		struct block_device *bdev;
-
-		/*
-		 * used for chunk mappings
-		 * flags & EXTENT_FLAG_FS_MAPPING must be set
-		 */
-		struct map_lookup *map_lookup;
-	};
+	/* Used for chunk mappings, flag EXTENT_FLAG_FS_MAPPING must be set */
+	struct map_lookup *map_lookup;
 	refcount_t refs;
 	unsigned int compress_type;
 	struct list_head list;
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 1a599f50837b..b618ad5339ba 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -23,6 +23,97 @@
 #define MAX_CSUM_ITEMS(r, size) (min_t(u32, __MAX_CSUM_ITEMS(r, size), \
 				       PAGE_SIZE))
 
+/**
+ * @inode - the inode we want to update the disk_i_size for
+ * @new_i_size - the i_size we want to set to, 0 if we use i_size
+ *
+ * With NO_HOLES set this simply sets the disk_is_size to whatever i_size_read()
+ * returns as it is perfectly fine with a file that has holes without hole file
+ * extent items.
+ *
+ * However without NO_HOLES we need to only return the area that is contiguous
+ * from the 0 offset of the file.  Otherwise we could end up adjust i_size up
+ * to an extent that has a gap in between.
+ *
+ * Finally new_i_size should only be set in the case of truncate where we're not
+ * ready to use i_size_read() as the limiter yet.
+ */
+void btrfs_inode_safe_disk_i_size_write(struct inode *inode, u64 new_i_size)
+{
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+	u64 start, end, i_size;
+	int ret;
+
+	i_size = new_i_size ?: i_size_read(inode);
+	if (btrfs_fs_incompat(fs_info, NO_HOLES)) {
+		BTRFS_I(inode)->disk_i_size = i_size;
+		return;
+	}
+
+	spin_lock(&BTRFS_I(inode)->lock);
+	ret = find_contiguous_extent_bit(&BTRFS_I(inode)->file_extent_tree, 0,
+					 &start, &end, EXTENT_DIRTY);
+	if (!ret && start == 0)
+		i_size = min(i_size, end + 1);
+	else
+		i_size = 0;
+	BTRFS_I(inode)->disk_i_size = i_size;
+	spin_unlock(&BTRFS_I(inode)->lock);
+}
+
+/**
+ * @inode - the inode we're modifying
+ * @start - the start file offset of the file extent we've inserted
+ * @len - the logical length of the file extent item
+ *
+ * Call when we are inserting a new file extent where there was none before.
+ * Does not need to call this in the case where we're replacing an existing file
+ * extent, however if not sure it's fine to call this multiple times.
+ *
+ * The start and len must match the file extent item, so thus must be sectorsize
+ * aligned.
+ */
+int btrfs_inode_set_file_extent_range(struct btrfs_inode *inode, u64 start,
+				      u64 len)
+{
+	if (len == 0)
+		return 0;
+
+	ASSERT(IS_ALIGNED(start + len, inode->root->fs_info->sectorsize));
+
+	if (btrfs_fs_incompat(inode->root->fs_info, NO_HOLES))
+		return 0;
+	return set_extent_bits(&inode->file_extent_tree, start, start + len - 1,
+			       EXTENT_DIRTY);
+}
+
+/**
+ * @inode - the inode we're modifying
+ * @start - the start file offset of the file extent we've inserted
+ * @len - the logical length of the file extent item
+ *
+ * Called when we drop a file extent, for example when we truncate.  Doesn't
+ * need to be called for cases where we're replacing a file extent, like when
+ * we've COWed a file extent.
+ *
+ * The start and len must match the file extent item, so thus must be sectorsize
+ * aligned.
+ */
+int btrfs_inode_clear_file_extent_range(struct btrfs_inode *inode, u64 start,
+					u64 len)
+{
+	if (len == 0)
+		return 0;
+
+	ASSERT(IS_ALIGNED(start + len, inode->root->fs_info->sectorsize) ||
+	       len == (u64)-1);
+
+	if (btrfs_fs_incompat(inode->root->fs_info, NO_HOLES))
+		return 0;
+	return clear_extent_bit(&inode->file_extent_tree, start,
+				start + len - 1, EXTENT_DIRTY, 0, 0, NULL);
+}
+
 static inline u32 max_ordered_sum_bytes(struct btrfs_fs_info *fs_info,
 					u16 csum_size)
 {
@@ -148,8 +239,19 @@ int btrfs_lookup_file_extent(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
-static blk_status_t __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
-				   u64 logical_offset, u8 *dst, int dio)
+/**
+ * btrfs_lookup_bio_sums - Look up checksums for a bio.
+ * @inode: inode that the bio is for.
+ * @bio: bio embedded in btrfs_io_bio.
+ * @offset: Unless (u64)-1, look up checksums for this offset in the file.
+ *          If (u64)-1, use the page offsets from the bio instead.
+ * @dst: Buffer of size btrfs_super_csum_size() used to return checksum. If
+ *       NULL, the checksum is returned in btrfs_io_bio(bio)->csum instead.
+ *
+ * Return: BLK_STS_RESOURCE if allocating memory fails, BLK_STS_OK otherwise.
+ */
+blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
+				   u64 offset, u8 *dst)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct bio_vec bvec;
@@ -158,8 +260,8 @@ static blk_status_t __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio
 	struct btrfs_csum_item *item = NULL;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct btrfs_path *path;
+	const bool page_offsets = (offset == (u64)-1);
 	u8 *csum;
-	u64 offset = 0;
 	u64 item_start_offset = 0;
 	u64 item_last_offset = 0;
 	u64 disk_bytenr;
@@ -205,15 +307,13 @@ static blk_status_t __btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio
 	}
 
 	disk_bytenr = (u64)bio->bi_iter.bi_sector << 9;
-	if (dio)
-		offset = logical_offset;
 
 	bio_for_each_segment(bvec, bio, iter) {
 		page_bytes_left = bvec.bv_len;
 		if (count)
 			goto next;
 
-		if (!dio)
+		if (page_offsets)
 			offset = page_offset(bvec.bv_page) + bvec.bv_offset;
 		count = btrfs_find_ordered_sum(inode, offset, disk_bytenr,
 					       csum, nblocks);
@@ -274,7 +374,8 @@ found:
 		csum += count * csum_size;
 		nblocks -= count;
 next:
-		while (count--) {
+		while (count > 0) {
+			count--;
 			disk_bytenr += fs_info->sectorsize;
 			offset += fs_info->sectorsize;
 			page_bytes_left -= fs_info->sectorsize;
@@ -285,18 +386,7 @@ next:
 
 	WARN_ON_ONCE(count);
 	btrfs_free_path(path);
-	return 0;
-}
-
-blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio,
-				   u8 *dst)
-{
-	return __btrfs_lookup_bio_sums(inode, bio, 0, dst, 0);
-}
-
-blk_status_t btrfs_lookup_bio_sums_dio(struct inode *inode, struct bio *bio, u64 offset)
-{
-	return __btrfs_lookup_bio_sums(inode, bio, offset, NULL, 1);
+	return BLK_STS_OK;
 }
 
 int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end,
@@ -483,8 +573,8 @@ blk_status_t btrfs_csum_one_bio(struct inode *inode, struct bio *bio,
 						 - 1);
 
 		for (i = 0; i < nr_sectors; i++) {
-			if (offset >= ordered->file_offset + ordered->len ||
-				offset < ordered->file_offset) {
+			if (offset >= ordered->file_offset + ordered->num_bytes ||
+			    offset < ordered->file_offset) {
 				unsigned long bytes_left;
 
 				sums->len = this_sum_bytes;
@@ -590,9 +680,9 @@ static noinline void truncate_one_csum(struct btrfs_fs_info *fs_info,
  * range of bytes.
  */
 int btrfs_del_csums(struct btrfs_trans_handle *trans,
-		    struct btrfs_fs_info *fs_info, u64 bytenr, u64 len)
+		    struct btrfs_root *root, u64 bytenr, u64 len)
 {
-	struct btrfs_root *root = fs_info->csum_root;
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	u64 end_byte = bytenr + len;
@@ -602,6 +692,9 @@ int btrfs_del_csums(struct btrfs_trans_handle *trans,
 	u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
 	int blocksize_bits = fs_info->sb->s_blocksize_bits;
 
+	ASSERT(root == fs_info->csum_root ||
+	       root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID);
+
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
@@ -945,21 +1038,9 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 	u8 type = btrfs_file_extent_type(leaf, fi);
 	int compress_type = btrfs_file_extent_compression(leaf, fi);
 
-	em->bdev = fs_info->fs_devices->latest_bdev;
 	btrfs_item_key_to_cpu(leaf, &key, slot);
 	extent_start = key.offset;
-
-	if (type == BTRFS_FILE_EXTENT_REG ||
-	    type == BTRFS_FILE_EXTENT_PREALLOC) {
-		extent_end = extent_start +
-			btrfs_file_extent_num_bytes(leaf, fi);
-	} else if (type == BTRFS_FILE_EXTENT_INLINE) {
-		size_t size;
-		size = btrfs_file_extent_ram_bytes(leaf, fi);
-		extent_end = ALIGN(extent_start + size,
-				   fs_info->sectorsize);
-	}
-
+	extent_end = btrfs_file_extent_end(path);
 	em->ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
 	if (type == BTRFS_FILE_EXTENT_REG ||
 	    type == BTRFS_FILE_EXTENT_PREALLOC) {
@@ -1006,3 +1087,30 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 			  root->root_key.objectid);
 	}
 }
+
+/*
+ * Returns the end offset (non inclusive) of the file extent item the given path
+ * points to. If it points to an inline extent, the returned offset is rounded
+ * up to the sector size.
+ */
+u64 btrfs_file_extent_end(const struct btrfs_path *path)
+{
+	const struct extent_buffer *leaf = path->nodes[0];
+	const int slot = path->slots[0];
+	struct btrfs_file_extent_item *fi;
+	struct btrfs_key key;
+	u64 end;
+
+	btrfs_item_key_to_cpu(leaf, &key, slot);
+	ASSERT(key.type == BTRFS_EXTENT_DATA_KEY);
+	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
+
+	if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) {
+		end = btrfs_file_extent_ram_bytes(leaf, fi);
+		end = ALIGN(key.offset + end, leaf->fs_info->sectorsize);
+	} else {
+		end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
+	}
+
+	return end;
+}
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 435a502a3226..719e68ab552c 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -27,6 +27,7 @@
 #include "qgroup.h"
 #include "compression.h"
 #include "delalloc-space.h"
+#include "reflink.h"
 
 static struct kmem_cache *btrfs_inode_defrag_cachep;
 /*
@@ -277,7 +278,6 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 	struct btrfs_key key;
 	struct btrfs_ioctl_defrag_range_args range;
 	int num_defrag;
-	int index;
 	int ret;
 
 	/* get the inode */
@@ -285,9 +285,7 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 	key.type = BTRFS_ROOT_ITEM_KEY;
 	key.offset = (u64)-1;
 
-	index = srcu_read_lock(&fs_info->subvol_srcu);
-
-	inode_root = btrfs_read_fs_root_no_name(fs_info, &key);
+	inode_root = btrfs_get_fs_root(fs_info, &key, true);
 	if (IS_ERR(inode_root)) {
 		ret = PTR_ERR(inode_root);
 		goto cleanup;
@@ -296,12 +294,12 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 	key.objectid = defrag->ino;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	inode = btrfs_iget(fs_info->sb, &key, inode_root, NULL);
+	inode = btrfs_iget(fs_info->sb, &key, inode_root);
+	btrfs_put_root(inode_root);
 	if (IS_ERR(inode)) {
 		ret = PTR_ERR(inode);
 		goto cleanup;
 	}
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
 
 	/* do a chunk of defrag */
 	clear_bit(BTRFS_INODE_IN_DEFRAG, &BTRFS_I(inode)->runtime_flags);
@@ -337,7 +335,6 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 	iput(inode);
 	return 0;
 cleanup:
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
 	kmem_cache_free(btrfs_inode_defrag_cachep, defrag);
 	return ret;
 }
@@ -477,8 +474,7 @@ static int btrfs_find_new_delalloc_bytes(struct btrfs_inode *inode,
 		u64 em_len;
 		int ret = 0;
 
-		em = btrfs_get_extent(inode, NULL, 0, search_start,
-				      search_len, 0);
+		em = btrfs_get_extent(inode, NULL, 0, search_start, search_len);
 		if (IS_ERR(em))
 			return PTR_ERR(em);
 
@@ -667,7 +663,6 @@ void btrfs_drop_extent_cache(struct btrfs_inode *inode, u64 start, u64 end,
 			}
 
 			split->generation = gen;
-			split->bdev = em->bdev;
 			split->flags = flags;
 			split->compress_type = em->compress_type;
 			replace_extent_mapping(em_tree, em, split, modified);
@@ -680,7 +675,6 @@ void btrfs_drop_extent_cache(struct btrfs_inode *inode, u64 start, u64 end,
 
 			split->start = start + len;
 			split->len = em->start + em->len - (start + len);
-			split->bdev = em->bdev;
 			split->flags = flags;
 			split->compress_type = em->compress_type;
 			split->generation = gen;
@@ -1503,7 +1497,7 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 		ordered = btrfs_lookup_ordered_range(inode, start_pos,
 						     last_pos - start_pos + 1);
 		if (ordered &&
-		    ordered->file_offset + ordered->len > start_pos &&
+		    ordered->file_offset + ordered->num_bytes > start_pos &&
 		    ordered->file_offset <= last_pos) {
 			unlock_extent_cached(&inode->io_tree, start_pos,
 					last_pos, cached_state);
@@ -1555,15 +1549,14 @@ static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos,
 	u64 num_bytes;
 	int ret;
 
-	ret = btrfs_start_write_no_snapshotting(root);
-	if (!ret)
+	if (!btrfs_drew_try_write_lock(&root->snapshot_lock))
 		return -EAGAIN;
 
 	lockstart = round_down(pos, fs_info->sectorsize);
 	lockend = round_up(pos + *write_bytes,
 			   fs_info->sectorsize) - 1;
 
-	btrfs_lock_and_flush_ordered_range(&inode->io_tree, inode, lockstart,
+	btrfs_lock_and_flush_ordered_range(inode, lockstart,
 					   lockend, NULL);
 
 	num_bytes = lockend - lockstart + 1;
@@ -1571,7 +1564,7 @@ static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos,
 			NULL, NULL, NULL);
 	if (ret <= 0) {
 		ret = 0;
-		btrfs_end_write_no_snapshotting(root);
+		btrfs_drew_write_unlock(&root->snapshot_lock);
 	} else {
 		*write_bytes = min_t(size_t, *write_bytes ,
 				     num_bytes - pos + lockstart);
@@ -1636,6 +1629,7 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 			break;
 		}
 
+		only_release_metadata = false;
 		sector_offset = pos & (fs_info->sectorsize - 1);
 		reserve_bytes = round_up(write_bytes + sector_offset,
 				fs_info->sectorsize);
@@ -1676,7 +1670,7 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 						data_reserved, pos,
 						write_bytes);
 			else
-				btrfs_end_write_no_snapshotting(root);
+				btrfs_drew_write_unlock(&root->snapshot_lock);
 			break;
 		}
 
@@ -1780,7 +1774,7 @@ again:
 
 		release_bytes = 0;
 		if (only_release_metadata)
-			btrfs_end_write_no_snapshotting(root);
+			btrfs_drew_write_unlock(&root->snapshot_lock);
 
 		if (only_release_metadata && copied > 0) {
 			lockstart = round_down(pos,
@@ -1791,7 +1785,6 @@ again:
 			set_extent_bit(&BTRFS_I(inode)->io_tree, lockstart,
 				       lockend, EXTENT_NORESERVE, NULL,
 				       NULL, GFP_NOFS);
-			only_release_metadata = false;
 		}
 
 		btrfs_drop_pages(pages, num_pages);
@@ -1810,7 +1803,7 @@ again:
 
 	if (release_bytes) {
 		if (only_release_metadata) {
-			btrfs_end_write_no_snapshotting(root);
+			btrfs_drew_write_unlock(&root->snapshot_lock);
 			btrfs_delalloc_release_metadata(BTRFS_I(inode),
 					release_bytes, true);
 		} else {
@@ -1903,9 +1896,10 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb,
 	    (iocb->ki_flags & IOCB_NOWAIT))
 		return -EOPNOTSUPP;
 
-	if (!inode_trylock(inode)) {
-		if (iocb->ki_flags & IOCB_NOWAIT)
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock(inode))
 			return -EAGAIN;
+	} else {
 		inode_lock(inode);
 	}
 
@@ -2073,6 +2067,16 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	btrfs_init_log_ctx(&ctx, inode);
 
 	/*
+	 * Set the range to full if the NO_HOLES feature is not enabled.
+	 * This is to avoid missing file extent items representing holes after
+	 * replaying the log.
+	 */
+	if (!btrfs_fs_incompat(fs_info, NO_HOLES)) {
+		start = 0;
+		end = LLONG_MAX;
+	}
+
+	/*
 	 * We write the dirty pages in the range and wait until they complete
 	 * out of the ->i_mutex. If so, we can flush the dirty pages by
 	 * multi-task, and make the performance up.  See
@@ -2096,9 +2100,11 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	/*
 	 * If the inode needs a full sync, make sure we use a full range to
 	 * avoid log tree corruption, due to hole detection racing with ordered
-	 * extent completion for adjacent ranges, and assertion failures during
-	 * hole detection. Do this while holding the inode lock, to avoid races
-	 * with other tasks.
+	 * extent completion for adjacent ranges and races between logging and
+	 * completion of ordered extents for adjancent ranges - both races
+	 * could lead to file extent items in the log with overlapping ranges.
+	 * Do this while holding the inode lock, to avoid races with other
+	 * tasks.
 	 */
 	if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
 		     &BTRFS_I(inode)->runtime_flags)) {
@@ -2126,6 +2132,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 */
 	ret = start_ordered_ops(inode, start, end);
 	if (ret) {
+		up_write(&BTRFS_I(inode)->dio_sem);
 		inode_unlock(inode);
 		goto out;
 	}
@@ -2359,7 +2366,6 @@ out:
 		hole_em->block_start = EXTENT_MAP_HOLE;
 		hole_em->block_len = 0;
 		hole_em->orig_block_len = 0;
-		hole_em->bdev = fs_info->fs_devices->latest_bdev;
 		hole_em->compress_type = BTRFS_COMPRESS_NONE;
 		hole_em->generation = trans->transid;
 
@@ -2392,7 +2398,7 @@ static int find_first_non_hole(struct inode *inode, u64 *start, u64 *len)
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0,
 			      round_down(*start, fs_info->sectorsize),
-			      round_up(*len, fs_info->sectorsize), 0);
+			      round_up(*len, fs_info->sectorsize));
 	if (IS_ERR(em))
 		return PTR_ERR(em);
 
@@ -2428,7 +2434,7 @@ static int btrfs_punch_hole_lock_range(struct inode *inode,
 		 * we need to try again.
 		 */
 		if ((!ordered ||
-		    (ordered->file_offset + ordered->len <= lockstart ||
+		    (ordered->file_offset + ordered->num_bytes <= lockstart ||
 		     ordered->file_offset > lockend)) &&
 		     !filemap_range_has_page(inode->i_mapping,
 					     lockstart, lockend)) {
@@ -2489,6 +2495,11 @@ static int btrfs_insert_clone_extent(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_release_path(path);
 
+	ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode),
+			clone_info->file_offset, clone_len);
+	if (ret)
+		return ret;
+
 	/* If it's a hole, nothing more needs to be done. */
 	if (clone_info->disk_offset == 0)
 		return 0;
@@ -2599,10 +2610,28 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 				btrfs_abort_transaction(trans, ret);
 				break;
 			}
+		} else if (!clone_info && cur_offset < drop_end) {
+			/*
+			 * We are past the i_size here, but since we didn't
+			 * insert holes we need to clear the mapped area so we
+			 * know to not set disk_i_size in this area until a new
+			 * file extent is inserted here.
+			 */
+			ret = btrfs_inode_clear_file_extent_range(BTRFS_I(inode),
+					cur_offset, drop_end - cur_offset);
+			if (ret) {
+				/*
+				 * We couldn't clear our area, so we could
+				 * presumably adjust up and corrupt the fs, so
+				 * we need to abort.
+				 */
+				btrfs_abort_transaction(trans, ret);
+				break;
+			}
 		}
 
-		if (clone_info) {
-			u64 clone_len = drop_end - cur_offset;
+		if (clone_info && drop_end > clone_info->file_offset) {
+			u64 clone_len = drop_end - clone_info->file_offset;
 
 			ret = btrfs_insert_clone_extent(trans, inode, path,
 							clone_info, clone_len);
@@ -2689,6 +2718,15 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path,
 			btrfs_abort_transaction(trans, ret);
 			goto out_trans;
 		}
+	} else if (!clone_info && cur_offset < drop_end) {
+		/* See the comment in the loop above for the reasoning here. */
+		ret = btrfs_inode_clear_file_extent_range(BTRFS_I(inode),
+					cur_offset, drop_end - cur_offset);
+		if (ret) {
+			btrfs_abort_transaction(trans, ret);
+			goto out_trans;
+		}
+
 	}
 	if (clone_info) {
 		ret = btrfs_insert_clone_extent(trans, inode, path, clone_info,
@@ -2938,7 +2976,7 @@ static int btrfs_fallocate_update_isize(struct inode *inode,
 
 	inode->i_ctime = current_time(inode);
 	i_size_write(inode, end);
-	btrfs_ordered_update_i_size(inode, end, NULL);
+	btrfs_inode_safe_disk_i_size_write(inode, 0);
 	ret = btrfs_update_inode(trans, root, inode);
 	ret2 = btrfs_end_transaction(trans);
 
@@ -2959,7 +2997,7 @@ static int btrfs_zero_range_check_range_boundary(struct inode *inode,
 	int ret;
 
 	offset = round_down(offset, sectorsize);
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em))
 		return PTR_ERR(em);
 
@@ -2992,8 +3030,8 @@ static int btrfs_zero_range(struct inode *inode,
 
 	inode_dio_wait(inode);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0,
-			      alloc_start, alloc_end - alloc_start, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, alloc_start,
+			      alloc_end - alloc_start);
 	if (IS_ERR(em)) {
 		ret = PTR_ERR(em);
 		goto out;
@@ -3036,8 +3074,8 @@ static int btrfs_zero_range(struct inode *inode,
 
 	if (BTRFS_BYTES_TO_BLKS(fs_info, offset) ==
 	    BTRFS_BYTES_TO_BLKS(fs_info, offset + len - 1)) {
-		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0,
-				      alloc_start, sectorsize, 0);
+		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, alloc_start,
+				      sectorsize);
 		if (IS_ERR(em)) {
 			ret = PTR_ERR(em);
 			goto out;
@@ -3250,7 +3288,7 @@ static long btrfs_fallocate(struct file *file, int mode,
 		ordered = btrfs_lookup_first_ordered_extent(inode, locked_end);
 
 		if (ordered &&
-		    ordered->file_offset + ordered->len > alloc_start &&
+		    ordered->file_offset + ordered->num_bytes > alloc_start &&
 		    ordered->file_offset < alloc_end) {
 			btrfs_put_ordered_extent(ordered);
 			unlock_extent_cached(&BTRFS_I(inode)->io_tree,
@@ -3275,7 +3313,7 @@ static long btrfs_fallocate(struct file *file, int mode,
 	INIT_LIST_HEAD(&reserve_list);
 	while (cur_offset < alloc_end) {
 		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset,
-				      alloc_end - cur_offset, 0);
+				      alloc_end - cur_offset);
 		if (IS_ERR(em)) {
 			ret = PTR_ERR(em);
 			break;
@@ -3350,29 +3388,30 @@ out:
 	return ret;
 }
 
-static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
+static loff_t find_desired_extent(struct inode *inode, loff_t offset,
+				  int whence)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_map *em = NULL;
 	struct extent_state *cached_state = NULL;
+	loff_t i_size = inode->i_size;
 	u64 lockstart;
 	u64 lockend;
 	u64 start;
 	u64 len;
 	int ret = 0;
 
-	if (inode->i_size == 0)
+	if (i_size == 0 || offset >= i_size)
 		return -ENXIO;
 
 	/*
-	 * *offset can be negative, in this case we start finding DATA/HOLE from
+	 * offset can be negative, in this case we start finding DATA/HOLE from
 	 * the very start of the file.
 	 */
-	start = max_t(loff_t, 0, *offset);
+	start = max_t(loff_t, 0, offset);
 
 	lockstart = round_down(start, fs_info->sectorsize);
-	lockend = round_up(i_size_read(inode),
-			   fs_info->sectorsize);
+	lockend = round_up(i_size, fs_info->sectorsize);
 	if (lockend <= lockstart)
 		lockend = lockstart + fs_info->sectorsize;
 	lockend--;
@@ -3381,7 +3420,7 @@ static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
 	lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 			 &cached_state);
 
-	while (start < inode->i_size) {
+	while (start < i_size) {
 		em = btrfs_get_extent_fiemap(BTRFS_I(inode), start, len);
 		if (IS_ERR(em)) {
 			ret = PTR_ERR(em);
@@ -3404,46 +3443,39 @@ static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
 		cond_resched();
 	}
 	free_extent_map(em);
-	if (!ret) {
-		if (whence == SEEK_DATA && start >= inode->i_size)
-			ret = -ENXIO;
-		else
-			*offset = min_t(loff_t, start, inode->i_size);
-	}
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 			     &cached_state);
-	return ret;
+	if (ret) {
+		offset = ret;
+	} else {
+		if (whence == SEEK_DATA && start >= i_size)
+			offset = -ENXIO;
+		else
+			offset = min_t(loff_t, start, i_size);
+	}
+
+	return offset;
 }
 
 static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence)
 {
 	struct inode *inode = file->f_mapping->host;
-	int ret;
 
-	inode_lock(inode);
 	switch (whence) {
-	case SEEK_END:
-	case SEEK_CUR:
-		offset = generic_file_llseek(file, offset, whence);
-		goto out;
+	default:
+		return generic_file_llseek(file, offset, whence);
 	case SEEK_DATA:
 	case SEEK_HOLE:
-		if (offset >= i_size_read(inode)) {
-			inode_unlock(inode);
-			return -ENXIO;
-		}
-
-		ret = find_desired_extent(inode, &offset, whence);
-		if (ret) {
-			inode_unlock(inode);
-			return ret;
-		}
+		inode_lock_shared(inode);
+		offset = find_desired_extent(inode, offset, whence);
+		inode_unlock_shared(inode);
+		break;
 	}
 
-	offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
-out:
-	inode_unlock(inode);
-	return offset;
+	if (offset < 0)
+		return offset;
+
+	return vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
 }
 
 static int btrfs_file_open(struct inode *inode, struct file *filp)
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index d54dcd0ab230..3613da065a73 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -21,9 +21,11 @@
 #include "space-info.h"
 #include "delalloc-space.h"
 #include "block-group.h"
+#include "discard.h"
 
 #define BITS_PER_BITMAP		(PAGE_SIZE * 8UL)
-#define MAX_CACHE_BYTES_PER_GIG	SZ_32K
+#define MAX_CACHE_BYTES_PER_GIG	SZ_64K
+#define FORCE_EXTENT_THRESHOLD	SZ_1M
 
 struct btrfs_trim_range {
 	u64 start;
@@ -31,6 +33,8 @@ struct btrfs_trim_range {
 	struct list_head list;
 };
 
+static int count_bitmap_extents(struct btrfs_free_space_ctl *ctl,
+				struct btrfs_free_space *bitmap_info);
 static int link_free_space(struct btrfs_free_space_ctl *ctl,
 			   struct btrfs_free_space *info);
 static void unlink_free_space(struct btrfs_free_space_ctl *ctl,
@@ -78,7 +82,7 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
 	 * sure NOFS is set to keep us from deadlocking.
 	 */
 	nofs_flag = memalloc_nofs_save();
-	inode = btrfs_iget_path(fs_info->sb, &location, root, NULL, path);
+	inode = btrfs_iget_path(fs_info->sb, &location, root, path);
 	btrfs_release_path(path);
 	memalloc_nofs_restore(nofs_flag);
 	if (IS_ERR(inode))
@@ -91,8 +95,7 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
 	return inode;
 }
 
-struct inode *lookup_free_space_inode(
-		struct btrfs_block_group_cache *block_group,
+struct inode *lookup_free_space_inode(struct btrfs_block_group *block_group,
 		struct btrfs_path *path)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
@@ -107,7 +110,7 @@ struct inode *lookup_free_space_inode(
 		return inode;
 
 	inode = __lookup_free_space_inode(fs_info->tree_root, path,
-					  block_group->key.objectid);
+					  block_group->start);
 	if (IS_ERR(inode))
 		return inode;
 
@@ -190,7 +193,7 @@ static int __create_free_space_inode(struct btrfs_root *root,
 }
 
 int create_free_space_inode(struct btrfs_trans_handle *trans,
-			    struct btrfs_block_group_cache *block_group,
+			    struct btrfs_block_group *block_group,
 			    struct btrfs_path *path)
 {
 	int ret;
@@ -201,7 +204,7 @@ int create_free_space_inode(struct btrfs_trans_handle *trans,
 		return ret;
 
 	return __create_free_space_inode(trans->fs_info->tree_root, trans, path,
-					 ino, block_group->key.objectid);
+					 ino, block_group->start);
 }
 
 int btrfs_check_trunc_cache_free_space(struct btrfs_fs_info *fs_info,
@@ -224,7 +227,7 @@ int btrfs_check_trunc_cache_free_space(struct btrfs_fs_info *fs_info,
 }
 
 int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans,
-				    struct btrfs_block_group_cache *block_group,
+				    struct btrfs_block_group *block_group,
 				    struct inode *inode)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -368,10 +371,10 @@ static void io_ctl_drop_pages(struct btrfs_io_ctl *io_ctl)
 	}
 }
 
-static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, struct inode *inode,
-				int uptodate)
+static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, bool uptodate)
 {
 	struct page *page;
+	struct inode *inode = io_ctl->inode;
 	gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping);
 	int i;
 
@@ -385,6 +388,12 @@ static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, struct inode *inode
 		if (uptodate && !PageUptodate(page)) {
 			btrfs_readpage(NULL, page);
 			lock_page(page);
+			if (page->mapping != inode->i_mapping) {
+				btrfs_err(BTRFS_I(inode)->root->fs_info,
+					  "free space cache page truncated");
+				io_ctl_drop_pages(io_ctl);
+				return -EIO;
+			}
 			if (!PageUptodate(page)) {
 				btrfs_err(BTRFS_I(inode)->root->fs_info,
 					   "error reading free space cache");
@@ -723,7 +732,7 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
 
 	readahead_cache(inode);
 
-	ret = io_ctl_prepare_pages(&io_ctl, inode, 1);
+	ret = io_ctl_prepare_pages(&io_ctl, true);
 	if (ret)
 		goto out;
 
@@ -747,6 +756,16 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
 			goto free_cache;
 		}
 
+		/*
+		 * Sync discard ensures that the free space cache is always
+		 * trimmed.  So when reading this in, the state should reflect
+		 * that.  We also do this for async as a stop gap for lack of
+		 * persistence.
+		 */
+		if (btrfs_test_opt(fs_info, DISCARD_SYNC) ||
+		    btrfs_test_opt(fs_info, DISCARD_ASYNC))
+			e->trim_state = BTRFS_TRIM_STATE_TRIMMED;
+
 		if (!e->bytes) {
 			kmem_cache_free(btrfs_free_space_cachep, e);
 			goto free_cache;
@@ -800,12 +819,19 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
 		ret = io_ctl_read_bitmap(&io_ctl, e);
 		if (ret)
 			goto free_cache;
+		e->bitmap_extents = count_bitmap_extents(ctl, e);
+		if (!btrfs_free_space_trimmed(e)) {
+			ctl->discardable_extents[BTRFS_STAT_CURR] +=
+				e->bitmap_extents;
+			ctl->discardable_bytes[BTRFS_STAT_CURR] += e->bytes;
+		}
 	}
 
 	io_ctl_drop_pages(&io_ctl);
 	merge_space_tree(ctl);
 	ret = 1;
 out:
+	btrfs_discard_update_discardable(ctl->private, ctl);
 	io_ctl_free(&io_ctl);
 	return ret;
 free_cache:
@@ -814,7 +840,7 @@ free_cache:
 	goto out;
 }
 
-int load_free_space_cache(struct btrfs_block_group_cache *block_group)
+int load_free_space_cache(struct btrfs_block_group *block_group)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
@@ -822,7 +848,7 @@ int load_free_space_cache(struct btrfs_block_group_cache *block_group)
 	struct btrfs_path *path;
 	int ret = 0;
 	bool matched;
-	u64 used = btrfs_block_group_used(&block_group->item);
+	u64 used = block_group->used;
 
 	/*
 	 * If this block group has been marked to be cleared for one reason or
@@ -876,13 +902,13 @@ int load_free_space_cache(struct btrfs_block_group_cache *block_group)
 	spin_unlock(&block_group->lock);
 
 	ret = __load_free_space_cache(fs_info->tree_root, inode, ctl,
-				      path, block_group->key.objectid);
+				      path, block_group->start);
 	btrfs_free_path(path);
 	if (ret <= 0)
 		goto out;
 
 	spin_lock(&ctl->tree_lock);
-	matched = (ctl->free_space == (block_group->key.offset - used -
+	matched = (ctl->free_space == (block_group->length - used -
 				       block_group->bytes_super));
 	spin_unlock(&ctl->tree_lock);
 
@@ -890,7 +916,7 @@ int load_free_space_cache(struct btrfs_block_group_cache *block_group)
 		__btrfs_remove_free_space_cache(ctl);
 		btrfs_warn(fs_info,
 			   "block group %llu has wrong amount of free space",
-			   block_group->key.objectid);
+			   block_group->start);
 		ret = -1;
 	}
 out:
@@ -903,7 +929,7 @@ out:
 
 		btrfs_warn(fs_info,
 			   "failed to load free space cache for block group %llu, rebuilding it now",
-			   block_group->key.objectid);
+			   block_group->start);
 	}
 
 	iput(inode);
@@ -913,7 +939,7 @@ out:
 static noinline_for_stack
 int write_cache_extent_entries(struct btrfs_io_ctl *io_ctl,
 			      struct btrfs_free_space_ctl *ctl,
-			      struct btrfs_block_group_cache *block_group,
+			      struct btrfs_block_group *block_group,
 			      int *entries, int *bitmaps,
 			      struct list_head *bitmap_list)
 {
@@ -1041,7 +1067,8 @@ fail:
 }
 
 static noinline_for_stack int write_pinned_extent_entries(
-			    struct btrfs_block_group_cache *block_group,
+			    struct btrfs_trans_handle *trans,
+			    struct btrfs_block_group *block_group,
 			    struct btrfs_io_ctl *io_ctl,
 			    int *entries)
 {
@@ -1059,11 +1086,11 @@ static noinline_for_stack int write_pinned_extent_entries(
 	 * We shouldn't have switched the pinned extents yet so this is the
 	 * right one
 	 */
-	unpin = block_group->fs_info->pinned_extents;
+	unpin = &trans->transaction->pinned_extents;
 
-	start = block_group->key.objectid;
+	start = block_group->start;
 
-	while (start < block_group->key.objectid + block_group->key.offset) {
+	while (start < block_group->start + block_group->length) {
 		ret = find_first_extent_bit(unpin, start,
 					    &extent_start, &extent_end,
 					    EXTENT_DIRTY, NULL);
@@ -1071,13 +1098,12 @@ static noinline_for_stack int write_pinned_extent_entries(
 			return 0;
 
 		/* This pinned extent is out of our range */
-		if (extent_start >= block_group->key.objectid +
-		    block_group->key.offset)
+		if (extent_start >= block_group->start + block_group->length)
 			return 0;
 
 		extent_start = max(extent_start, start);
-		extent_end = min(block_group->key.objectid +
-				 block_group->key.offset, extent_end + 1);
+		extent_end = min(block_group->start + block_group->length,
+				 extent_end + 1);
 		len = extent_end - extent_start;
 
 		*entries += 1;
@@ -1141,7 +1167,7 @@ cleanup_write_cache_enospc(struct inode *inode,
 
 static int __btrfs_wait_cache_io(struct btrfs_root *root,
 				 struct btrfs_trans_handle *trans,
-				 struct btrfs_block_group_cache *block_group,
+				 struct btrfs_block_group *block_group,
 				 struct btrfs_io_ctl *io_ctl,
 				 struct btrfs_path *path, u64 offset)
 {
@@ -1165,10 +1191,10 @@ out:
 		invalidate_inode_pages2(inode->i_mapping);
 		BTRFS_I(inode)->generation = 0;
 		if (block_group) {
-#ifdef DEBUG
+#ifdef CONFIG_BTRFS_DEBUG
 			btrfs_err(root->fs_info,
 				  "failed to write free space cache for block group %llu",
-				  block_group->key.objectid);
+				  block_group->start);
 #endif
 		}
 	}
@@ -1210,12 +1236,12 @@ static int btrfs_wait_cache_io_root(struct btrfs_root *root,
 }
 
 int btrfs_wait_cache_io(struct btrfs_trans_handle *trans,
-			struct btrfs_block_group_cache *block_group,
+			struct btrfs_block_group *block_group,
 			struct btrfs_path *path)
 {
 	return __btrfs_wait_cache_io(block_group->fs_info->tree_root, trans,
 				     block_group, &block_group->io_ctl,
-				     path, block_group->key.objectid);
+				     path, block_group->start);
 }
 
 /**
@@ -1231,7 +1257,7 @@ int btrfs_wait_cache_io(struct btrfs_trans_handle *trans,
  */
 static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 				   struct btrfs_free_space_ctl *ctl,
-				   struct btrfs_block_group_cache *block_group,
+				   struct btrfs_block_group *block_group,
 				   struct btrfs_io_ctl *io_ctl,
 				   struct btrfs_trans_handle *trans)
 {
@@ -1266,7 +1292,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 	}
 
 	/* Lock all pages first so we can lock the extent safely. */
-	ret = io_ctl_prepare_pages(io_ctl, inode, 0);
+	ret = io_ctl_prepare_pages(io_ctl, false);
 	if (ret)
 		goto out_unlock;
 
@@ -1292,7 +1318,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 	 * If this changes while we are working we'll get added back to
 	 * the dirty list and redo it.  No locking needed
 	 */
-	ret = write_pinned_extent_entries(block_group, io_ctl, &entries);
+	ret = write_pinned_extent_entries(trans, block_group, io_ctl, &entries);
 	if (ret)
 		goto out_nospc_locked;
 
@@ -1341,18 +1367,6 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 
 	return 0;
 
-out:
-	io_ctl->inode = NULL;
-	io_ctl_free(io_ctl);
-	if (ret) {
-		invalidate_inode_pages2(inode->i_mapping);
-		BTRFS_I(inode)->generation = 0;
-	}
-	btrfs_update_inode(trans, root, inode);
-	if (must_iput)
-		iput(inode);
-	return ret;
-
 out_nospc_locked:
 	cleanup_bitmap_list(&bitmap_list);
 	spin_unlock(&ctl->tree_lock);
@@ -1365,11 +1379,21 @@ out_unlock:
 	if (block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA))
 		up_write(&block_group->data_rwsem);
 
-	goto out;
+out:
+	io_ctl->inode = NULL;
+	io_ctl_free(io_ctl);
+	if (ret) {
+		invalidate_inode_pages2(inode->i_mapping);
+		BTRFS_I(inode)->generation = 0;
+	}
+	btrfs_update_inode(trans, root, inode);
+	if (must_iput)
+		iput(inode);
+	return ret;
 }
 
 int btrfs_write_out_cache(struct btrfs_trans_handle *trans,
-			  struct btrfs_block_group_cache *block_group,
+			  struct btrfs_block_group *block_group,
 			  struct btrfs_path *path)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
@@ -1391,10 +1415,10 @@ int btrfs_write_out_cache(struct btrfs_trans_handle *trans,
 	ret = __btrfs_write_out_cache(fs_info->tree_root, inode, ctl,
 				block_group, &block_group->io_ctl, trans);
 	if (ret) {
-#ifdef DEBUG
+#ifdef CONFIG_BTRFS_DEBUG
 		btrfs_err(fs_info,
 			  "failed to write free space cache for block group %llu",
-			  block_group->key.objectid);
+			  block_group->start);
 #endif
 		spin_lock(&block_group->lock);
 		block_group->disk_cache_state = BTRFS_DC_ERROR;
@@ -1620,6 +1644,11 @@ __unlink_free_space(struct btrfs_free_space_ctl *ctl,
 {
 	rb_erase(&info->offset_index, &ctl->free_space_offset);
 	ctl->free_extents--;
+
+	if (!info->bitmap && !btrfs_free_space_trimmed(info)) {
+		ctl->discardable_extents[BTRFS_STAT_CURR]--;
+		ctl->discardable_bytes[BTRFS_STAT_CURR] -= info->bytes;
+	}
 }
 
 static void unlink_free_space(struct btrfs_free_space_ctl *ctl,
@@ -1640,6 +1669,11 @@ static int link_free_space(struct btrfs_free_space_ctl *ctl,
 	if (ret)
 		return ret;
 
+	if (!info->bitmap && !btrfs_free_space_trimmed(info)) {
+		ctl->discardable_extents[BTRFS_STAT_CURR]++;
+		ctl->discardable_bytes[BTRFS_STAT_CURR] += info->bytes;
+	}
+
 	ctl->free_space += info->bytes;
 	ctl->free_extents++;
 	return ret;
@@ -1647,11 +1681,11 @@ static int link_free_space(struct btrfs_free_space_ctl *ctl,
 
 static void recalculate_thresholds(struct btrfs_free_space_ctl *ctl)
 {
-	struct btrfs_block_group_cache *block_group = ctl->private;
+	struct btrfs_block_group *block_group = ctl->private;
 	u64 max_bytes;
 	u64 bitmap_bytes;
 	u64 extent_bytes;
-	u64 size = block_group->key.offset;
+	u64 size = block_group->length;
 	u64 bytes_per_bg = BITS_PER_BITMAP * ctl->unit;
 	u64 max_bitmaps = div64_u64(size + bytes_per_bg - 1, bytes_per_bg);
 
@@ -1660,26 +1694,17 @@ static void recalculate_thresholds(struct btrfs_free_space_ctl *ctl)
 	ASSERT(ctl->total_bitmaps <= max_bitmaps);
 
 	/*
-	 * The goal is to keep the total amount of memory used per 1gb of space
-	 * at or below 32k, so we need to adjust how much memory we allow to be
-	 * used by extent based free space tracking
+	 * We are trying to keep the total amount of memory used per 1GiB of
+	 * space to be MAX_CACHE_BYTES_PER_GIG.  However, with a reclamation
+	 * mechanism of pulling extents >= FORCE_EXTENT_THRESHOLD out of
+	 * bitmaps, we may end up using more memory than this.
 	 */
 	if (size < SZ_1G)
 		max_bytes = MAX_CACHE_BYTES_PER_GIG;
 	else
 		max_bytes = MAX_CACHE_BYTES_PER_GIG * div_u64(size, SZ_1G);
 
-	/*
-	 * we want to account for 1 more bitmap than what we have so we can make
-	 * sure we don't go over our overall goal of MAX_CACHE_BYTES_PER_GIG as
-	 * we add more bitmaps.
-	 */
-	bitmap_bytes = (ctl->total_bitmaps + 1) * ctl->unit;
-
-	if (bitmap_bytes >= max_bytes) {
-		ctl->extents_thresh = 0;
-		return;
-	}
+	bitmap_bytes = ctl->total_bitmaps * ctl->unit;
 
 	/*
 	 * we want the extent entry threshold to always be at most 1/2 the max
@@ -1696,17 +1721,31 @@ static inline void __bitmap_clear_bits(struct btrfs_free_space_ctl *ctl,
 				       struct btrfs_free_space *info,
 				       u64 offset, u64 bytes)
 {
-	unsigned long start, count;
+	unsigned long start, count, end;
+	int extent_delta = -1;
 
 	start = offset_to_bit(info->offset, ctl->unit, offset);
 	count = bytes_to_bits(bytes, ctl->unit);
-	ASSERT(start + count <= BITS_PER_BITMAP);
+	end = start + count;
+	ASSERT(end <= BITS_PER_BITMAP);
 
 	bitmap_clear(info->bitmap, start, count);
 
 	info->bytes -= bytes;
 	if (info->max_extent_size > ctl->unit)
 		info->max_extent_size = 0;
+
+	if (start && test_bit(start - 1, info->bitmap))
+		extent_delta++;
+
+	if (end < BITS_PER_BITMAP && test_bit(end, info->bitmap))
+		extent_delta++;
+
+	info->bitmap_extents += extent_delta;
+	if (!btrfs_free_space_trimmed(info)) {
+		ctl->discardable_extents[BTRFS_STAT_CURR] += extent_delta;
+		ctl->discardable_bytes[BTRFS_STAT_CURR] -= bytes;
+	}
 }
 
 static void bitmap_clear_bits(struct btrfs_free_space_ctl *ctl,
@@ -1721,16 +1760,30 @@ static void bitmap_set_bits(struct btrfs_free_space_ctl *ctl,
 			    struct btrfs_free_space *info, u64 offset,
 			    u64 bytes)
 {
-	unsigned long start, count;
+	unsigned long start, count, end;
+	int extent_delta = 1;
 
 	start = offset_to_bit(info->offset, ctl->unit, offset);
 	count = bytes_to_bits(bytes, ctl->unit);
-	ASSERT(start + count <= BITS_PER_BITMAP);
+	end = start + count;
+	ASSERT(end <= BITS_PER_BITMAP);
 
 	bitmap_set(info->bitmap, start, count);
 
 	info->bytes += bytes;
 	ctl->free_space += bytes;
+
+	if (start && test_bit(start - 1, info->bitmap))
+		extent_delta--;
+
+	if (end < BITS_PER_BITMAP && test_bit(end, info->bitmap))
+		extent_delta--;
+
+	info->bitmap_extents += extent_delta;
+	if (!btrfs_free_space_trimmed(info)) {
+		ctl->discardable_extents[BTRFS_STAT_CURR] += extent_delta;
+		ctl->discardable_bytes[BTRFS_STAT_CURR] += bytes;
+	}
 }
 
 /*
@@ -1866,11 +1919,35 @@ out:
 	return NULL;
 }
 
+static int count_bitmap_extents(struct btrfs_free_space_ctl *ctl,
+				struct btrfs_free_space *bitmap_info)
+{
+	struct btrfs_block_group *block_group = ctl->private;
+	u64 bytes = bitmap_info->bytes;
+	unsigned int rs, re;
+	int count = 0;
+
+	if (!block_group || !bytes)
+		return count;
+
+	bitmap_for_each_set_region(bitmap_info->bitmap, rs, re, 0,
+				   BITS_PER_BITMAP) {
+		bytes -= (rs - re) * ctl->unit;
+		count++;
+
+		if (!bytes)
+			break;
+	}
+
+	return count;
+}
+
 static void add_new_bitmap(struct btrfs_free_space_ctl *ctl,
 			   struct btrfs_free_space *info, u64 offset)
 {
 	info->offset = offset_to_bitmap(ctl, offset);
 	info->bytes = 0;
+	info->bitmap_extents = 0;
 	INIT_LIST_HEAD(&info->list);
 	link_free_space(ctl, info);
 	ctl->total_bitmaps++;
@@ -1881,6 +1958,18 @@ static void add_new_bitmap(struct btrfs_free_space_ctl *ctl,
 static void free_bitmap(struct btrfs_free_space_ctl *ctl,
 			struct btrfs_free_space *bitmap_info)
 {
+	/*
+	 * Normally when this is called, the bitmap is completely empty. However,
+	 * if we are blowing up the free space cache for one reason or another
+	 * via __btrfs_remove_free_space_cache(), then it may not be freed and
+	 * we may leave stats on the table.
+	 */
+	if (bitmap_info->bytes && !btrfs_free_space_trimmed(bitmap_info)) {
+		ctl->discardable_extents[BTRFS_STAT_CURR] -=
+			bitmap_info->bitmap_extents;
+		ctl->discardable_bytes[BTRFS_STAT_CURR] -= bitmap_info->bytes;
+
+	}
 	unlink_free_space(ctl, bitmap_info);
 	kmem_cache_free(btrfs_free_space_bitmap_cachep, bitmap_info->bitmap);
 	kmem_cache_free(btrfs_free_space_cachep, bitmap_info);
@@ -1967,11 +2056,24 @@ again:
 
 static u64 add_bytes_to_bitmap(struct btrfs_free_space_ctl *ctl,
 			       struct btrfs_free_space *info, u64 offset,
-			       u64 bytes)
+			       u64 bytes, enum btrfs_trim_state trim_state)
 {
 	u64 bytes_to_set = 0;
 	u64 end;
 
+	/*
+	 * This is a tradeoff to make bitmap trim state minimal.  We mark the
+	 * whole bitmap untrimmed if at any point we add untrimmed regions.
+	 */
+	if (trim_state == BTRFS_TRIM_STATE_UNTRIMMED) {
+		if (btrfs_free_space_trimmed(info)) {
+			ctl->discardable_extents[BTRFS_STAT_CURR] +=
+				info->bitmap_extents;
+			ctl->discardable_bytes[BTRFS_STAT_CURR] += info->bytes;
+		}
+		info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
+	}
+
 	end = info->offset + (u64)(BITS_PER_BITMAP * ctl->unit);
 
 	bytes_to_set = min(end - offset, bytes);
@@ -1991,7 +2093,7 @@ static u64 add_bytes_to_bitmap(struct btrfs_free_space_ctl *ctl,
 static bool use_bitmap(struct btrfs_free_space_ctl *ctl,
 		      struct btrfs_free_space *info)
 {
-	struct btrfs_block_group_cache *block_group = ctl->private;
+	struct btrfs_block_group *block_group = ctl->private;
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
 	bool forced = false;
 
@@ -2000,6 +2102,10 @@ static bool use_bitmap(struct btrfs_free_space_ctl *ctl,
 		forced = true;
 #endif
 
+	/* This is a way to reclaim large regions from the bitmaps. */
+	if (!forced && info->bytes >= FORCE_EXTENT_THRESHOLD)
+		return false;
+
 	/*
 	 * If we are below the extents threshold then we can add this as an
 	 * extent, and don't have to deal with the bitmap
@@ -2012,8 +2118,8 @@ static bool use_bitmap(struct btrfs_free_space_ctl *ctl,
 		 * of cache left then go ahead an dadd them, no sense in adding
 		 * the overhead of a bitmap if we don't have to.
 		 */
-		if (info->bytes <= fs_info->sectorsize * 4) {
-			if (ctl->free_extents * 2 <= ctl->extents_thresh)
+		if (info->bytes <= fs_info->sectorsize * 8) {
+			if (ctl->free_extents * 3 <= ctl->extents_thresh)
 				return false;
 		} else {
 			return false;
@@ -2028,7 +2134,7 @@ static bool use_bitmap(struct btrfs_free_space_ctl *ctl,
 	 * so allow those block groups to still be allowed to have a bitmap
 	 * entry.
 	 */
-	if (((BITS_PER_BITMAP * ctl->unit) >> 1) > block_group->key.offset)
+	if (((BITS_PER_BITMAP * ctl->unit) >> 1) > block_group->length)
 		return false;
 
 	return true;
@@ -2043,13 +2149,15 @@ static int insert_into_bitmap(struct btrfs_free_space_ctl *ctl,
 			      struct btrfs_free_space *info)
 {
 	struct btrfs_free_space *bitmap_info;
-	struct btrfs_block_group_cache *block_group = NULL;
+	struct btrfs_block_group *block_group = NULL;
 	int added = 0;
 	u64 bytes, offset, bytes_added;
+	enum btrfs_trim_state trim_state;
 	int ret;
 
 	bytes = info->bytes;
 	offset = info->offset;
+	trim_state = info->trim_state;
 
 	if (!ctl->op->use_bitmap(ctl, info))
 		return 0;
@@ -2084,8 +2192,8 @@ again:
 		}
 
 		if (entry->offset == offset_to_bitmap(ctl, offset)) {
-			bytes_added = add_bytes_to_bitmap(ctl, entry,
-							  offset, bytes);
+			bytes_added = add_bytes_to_bitmap(ctl, entry, offset,
+							  bytes, trim_state);
 			bytes -= bytes_added;
 			offset += bytes_added;
 		}
@@ -2104,7 +2212,8 @@ no_cluster_bitmap:
 		goto new_bitmap;
 	}
 
-	bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes);
+	bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes,
+					  trim_state);
 	bytes -= bytes_added;
 	offset += bytes_added;
 	added = 0;
@@ -2138,6 +2247,7 @@ new_bitmap:
 		/* allocate the bitmap */
 		info->bitmap = kmem_cache_zalloc(btrfs_free_space_bitmap_cachep,
 						 GFP_NOFS);
+		info->trim_state = BTRFS_TRIM_STATE_TRIMMED;
 		spin_lock(&ctl->tree_lock);
 		if (!info->bitmap) {
 			ret = -ENOMEM;
@@ -2157,6 +2267,22 @@ out:
 	return ret;
 }
 
+/*
+ * Free space merging rules:
+ *  1) Merge trimmed areas together
+ *  2) Let untrimmed areas coalesce with trimmed areas
+ *  3) Always pull neighboring regions from bitmaps
+ *
+ * The above rules are for when we merge free space based on btrfs_trim_state.
+ * Rules 2 and 3 are subtle because they are suboptimal, but are done for the
+ * same reason: to promote larger extent regions which makes life easier for
+ * find_free_extent().  Rule 2 enables coalescing based on the common path
+ * being returning free space from btrfs_finish_extent_commit().  So when free
+ * space is trimmed, it will prevent aggregating trimmed new region and
+ * untrimmed regions in the rb_tree.  Rule 3 is purely to obtain larger extents
+ * and provide find_free_extent() with the largest extents possible hoping for
+ * the reuse path.
+ */
 static bool try_merge_free_space(struct btrfs_free_space_ctl *ctl,
 			  struct btrfs_free_space *info, bool update_stat)
 {
@@ -2165,6 +2291,7 @@ static bool try_merge_free_space(struct btrfs_free_space_ctl *ctl,
 	bool merged = false;
 	u64 offset = info->offset;
 	u64 bytes = info->bytes;
+	const bool is_trimmed = btrfs_free_space_trimmed(info);
 
 	/*
 	 * first we want to see if there is free space adjacent to the range we
@@ -2178,7 +2305,9 @@ static bool try_merge_free_space(struct btrfs_free_space_ctl *ctl,
 	else
 		left_info = tree_search_offset(ctl, offset - 1, 0, 0);
 
-	if (right_info && !right_info->bitmap) {
+	/* See try_merge_free_space() comment. */
+	if (right_info && !right_info->bitmap &&
+	    (!is_trimmed || btrfs_free_space_trimmed(right_info))) {
 		if (update_stat)
 			unlink_free_space(ctl, right_info);
 		else
@@ -2188,8 +2317,10 @@ static bool try_merge_free_space(struct btrfs_free_space_ctl *ctl,
 		merged = true;
 	}
 
+	/* See try_merge_free_space() comment. */
 	if (left_info && !left_info->bitmap &&
-	    left_info->offset + left_info->bytes == offset) {
+	    left_info->offset + left_info->bytes == offset &&
+	    (!is_trimmed || btrfs_free_space_trimmed(left_info))) {
 		if (update_stat)
 			unlink_free_space(ctl, left_info);
 		else
@@ -2225,6 +2356,10 @@ static bool steal_from_bitmap_to_end(struct btrfs_free_space_ctl *ctl,
 	bytes = (j - i) * ctl->unit;
 	info->bytes += bytes;
 
+	/* See try_merge_free_space() comment. */
+	if (!btrfs_free_space_trimmed(bitmap))
+		info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
+
 	if (update_stat)
 		bitmap_clear_bits(ctl, bitmap, end, bytes);
 	else
@@ -2278,6 +2413,10 @@ static bool steal_from_bitmap_to_front(struct btrfs_free_space_ctl *ctl,
 	info->offset -= bytes;
 	info->bytes += bytes;
 
+	/* See try_merge_free_space() comment. */
+	if (!btrfs_free_space_trimmed(bitmap))
+		info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
+
 	if (update_stat)
 		bitmap_clear_bits(ctl, bitmap, info->offset, bytes);
 	else
@@ -2327,10 +2466,13 @@ static void steal_from_bitmap(struct btrfs_free_space_ctl *ctl,
 
 int __btrfs_add_free_space(struct btrfs_fs_info *fs_info,
 			   struct btrfs_free_space_ctl *ctl,
-			   u64 offset, u64 bytes)
+			   u64 offset, u64 bytes,
+			   enum btrfs_trim_state trim_state)
 {
+	struct btrfs_block_group *block_group = ctl->private;
 	struct btrfs_free_space *info;
 	int ret = 0;
+	u64 filter_bytes = bytes;
 
 	info = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS);
 	if (!info)
@@ -2338,6 +2480,7 @@ int __btrfs_add_free_space(struct btrfs_fs_info *fs_info,
 
 	info->offset = offset;
 	info->bytes = bytes;
+	info->trim_state = trim_state;
 	RB_CLEAR_NODE(&info->offset_index);
 
 	spin_lock(&ctl->tree_lock);
@@ -2366,10 +2509,13 @@ link:
 	 */
 	steal_from_bitmap(ctl, info, true);
 
+	filter_bytes = max(filter_bytes, info->bytes);
+
 	ret = link_free_space(ctl, info);
 	if (ret)
 		kmem_cache_free(btrfs_free_space_cachep, info);
 out:
+	btrfs_discard_update_discardable(block_group, ctl);
 	spin_unlock(&ctl->tree_lock);
 
 	if (ret) {
@@ -2377,18 +2523,47 @@ out:
 		ASSERT(ret != -EEXIST);
 	}
 
+	if (trim_state != BTRFS_TRIM_STATE_TRIMMED) {
+		btrfs_discard_check_filter(block_group, filter_bytes);
+		btrfs_discard_queue_work(&fs_info->discard_ctl, block_group);
+	}
+
 	return ret;
 }
 
-int btrfs_add_free_space(struct btrfs_block_group_cache *block_group,
+int btrfs_add_free_space(struct btrfs_block_group *block_group,
 			 u64 bytenr, u64 size)
 {
+	enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
+
+	if (btrfs_test_opt(block_group->fs_info, DISCARD_SYNC))
+		trim_state = BTRFS_TRIM_STATE_TRIMMED;
+
 	return __btrfs_add_free_space(block_group->fs_info,
 				      block_group->free_space_ctl,
-				      bytenr, size);
+				      bytenr, size, trim_state);
 }
 
-int btrfs_remove_free_space(struct btrfs_block_group_cache *block_group,
+/*
+ * This is a subtle distinction because when adding free space back in general,
+ * we want it to be added as untrimmed for async. But in the case where we add
+ * it on loading of a block group, we want to consider it trimmed.
+ */
+int btrfs_add_free_space_async_trimmed(struct btrfs_block_group *block_group,
+				       u64 bytenr, u64 size)
+{
+	enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
+
+	if (btrfs_test_opt(block_group->fs_info, DISCARD_SYNC) ||
+	    btrfs_test_opt(block_group->fs_info, DISCARD_ASYNC))
+		trim_state = BTRFS_TRIM_STATE_TRIMMED;
+
+	return __btrfs_add_free_space(block_group->fs_info,
+				      block_group->free_space_ctl,
+				      bytenr, size, trim_state);
+}
+
+int btrfs_remove_free_space(struct btrfs_block_group *block_group,
 			    u64 offset, u64 bytes)
 {
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
@@ -2460,8 +2635,10 @@ again:
 			}
 			spin_unlock(&ctl->tree_lock);
 
-			ret = btrfs_add_free_space(block_group, offset + bytes,
-						   old_end - (offset + bytes));
+			ret = __btrfs_add_free_space(block_group->fs_info, ctl,
+						     offset + bytes,
+						     old_end - (offset + bytes),
+						     info->trim_state);
 			WARN_ON(ret);
 			goto out;
 		}
@@ -2473,12 +2650,13 @@ again:
 		goto again;
 	}
 out_lock:
+	btrfs_discard_update_discardable(block_group, ctl);
 	spin_unlock(&ctl->tree_lock);
 out:
 	return ret;
 }
 
-void btrfs_dump_free_space(struct btrfs_block_group_cache *block_group,
+void btrfs_dump_free_space(struct btrfs_block_group *block_group,
 			   u64 bytes)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
@@ -2503,14 +2681,14 @@ void btrfs_dump_free_space(struct btrfs_block_group_cache *block_group,
 		   "%d blocks of free space at or bigger than bytes is", count);
 }
 
-void btrfs_init_free_space_ctl(struct btrfs_block_group_cache *block_group)
+void btrfs_init_free_space_ctl(struct btrfs_block_group *block_group)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 
 	spin_lock_init(&ctl->tree_lock);
 	ctl->unit = fs_info->sectorsize;
-	ctl->start = block_group->key.objectid;
+	ctl->start = block_group->start;
 	ctl->private = block_group;
 	ctl->op = &free_space_op;
 	INIT_LIST_HEAD(&ctl->trimming_ranges);
@@ -2532,7 +2710,7 @@ void btrfs_init_free_space_ctl(struct btrfs_block_group_cache *block_group)
  */
 static int
 __btrfs_return_cluster_to_free_space(
-			     struct btrfs_block_group_cache *block_group,
+			     struct btrfs_block_group *block_group,
 			     struct btrfs_free_cluster *cluster)
 {
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
@@ -2558,8 +2736,22 @@ __btrfs_return_cluster_to_free_space(
 
 		bitmap = (entry->bitmap != NULL);
 		if (!bitmap) {
+			/* Merging treats extents as if they were new */
+			if (!btrfs_free_space_trimmed(entry)) {
+				ctl->discardable_extents[BTRFS_STAT_CURR]--;
+				ctl->discardable_bytes[BTRFS_STAT_CURR] -=
+					entry->bytes;
+			}
+
 			try_merge_free_space(ctl, entry, false);
 			steal_from_bitmap(ctl, entry, false);
+
+			/* As we insert directly, update these statistics */
+			if (!btrfs_free_space_trimmed(entry)) {
+				ctl->discardable_extents[BTRFS_STAT_CURR]++;
+				ctl->discardable_bytes[BTRFS_STAT_CURR] +=
+					entry->bytes;
+			}
 		}
 		tree_insert_offset(&ctl->free_space_offset,
 				   entry->offset, &entry->offset_index, bitmap);
@@ -2595,10 +2787,12 @@ void __btrfs_remove_free_space_cache(struct btrfs_free_space_ctl *ctl)
 {
 	spin_lock(&ctl->tree_lock);
 	__btrfs_remove_free_space_cache_locked(ctl);
+	if (ctl->private)
+		btrfs_discard_update_discardable(ctl->private, ctl);
 	spin_unlock(&ctl->tree_lock);
 }
 
-void btrfs_remove_free_space_cache(struct btrfs_block_group_cache *block_group)
+void btrfs_remove_free_space_cache(struct btrfs_block_group *block_group)
 {
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	struct btrfs_free_cluster *cluster;
@@ -2616,20 +2810,55 @@ void btrfs_remove_free_space_cache(struct btrfs_block_group_cache *block_group)
 		cond_resched_lock(&ctl->tree_lock);
 	}
 	__btrfs_remove_free_space_cache_locked(ctl);
+	btrfs_discard_update_discardable(block_group, ctl);
 	spin_unlock(&ctl->tree_lock);
 
 }
 
-u64 btrfs_find_space_for_alloc(struct btrfs_block_group_cache *block_group,
+/**
+ * btrfs_is_free_space_trimmed - see if everything is trimmed
+ * @block_group: block_group of interest
+ *
+ * Walk @block_group's free space rb_tree to determine if everything is trimmed.
+ */
+bool btrfs_is_free_space_trimmed(struct btrfs_block_group *block_group)
+{
+	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
+	struct btrfs_free_space *info;
+	struct rb_node *node;
+	bool ret = true;
+
+	spin_lock(&ctl->tree_lock);
+	node = rb_first(&ctl->free_space_offset);
+
+	while (node) {
+		info = rb_entry(node, struct btrfs_free_space, offset_index);
+
+		if (!btrfs_free_space_trimmed(info)) {
+			ret = false;
+			break;
+		}
+
+		node = rb_next(node);
+	}
+
+	spin_unlock(&ctl->tree_lock);
+	return ret;
+}
+
+u64 btrfs_find_space_for_alloc(struct btrfs_block_group *block_group,
 			       u64 offset, u64 bytes, u64 empty_size,
 			       u64 *max_extent_size)
 {
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
+	struct btrfs_discard_ctl *discard_ctl =
+					&block_group->fs_info->discard_ctl;
 	struct btrfs_free_space *entry = NULL;
 	u64 bytes_search = bytes + empty_size;
 	u64 ret = 0;
 	u64 align_gap = 0;
 	u64 align_gap_len = 0;
+	enum btrfs_trim_state align_gap_trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
 
 	spin_lock(&ctl->tree_lock);
 	entry = find_free_space(ctl, &offset, &bytes_search,
@@ -2640,12 +2869,20 @@ u64 btrfs_find_space_for_alloc(struct btrfs_block_group_cache *block_group,
 	ret = offset;
 	if (entry->bitmap) {
 		bitmap_clear_bits(ctl, entry, offset, bytes);
+
+		if (!btrfs_free_space_trimmed(entry))
+			atomic64_add(bytes, &discard_ctl->discard_bytes_saved);
+
 		if (!entry->bytes)
 			free_bitmap(ctl, entry);
 	} else {
 		unlink_free_space(ctl, entry);
 		align_gap_len = offset - entry->offset;
 		align_gap = entry->offset;
+		align_gap_trim_state = entry->trim_state;
+
+		if (!btrfs_free_space_trimmed(entry))
+			atomic64_add(bytes, &discard_ctl->discard_bytes_saved);
 
 		entry->offset = offset + bytes;
 		WARN_ON(entry->bytes < bytes + align_gap_len);
@@ -2657,11 +2894,13 @@ u64 btrfs_find_space_for_alloc(struct btrfs_block_group_cache *block_group,
 			link_free_space(ctl, entry);
 	}
 out:
+	btrfs_discard_update_discardable(block_group, ctl);
 	spin_unlock(&ctl->tree_lock);
 
 	if (align_gap_len)
 		__btrfs_add_free_space(block_group->fs_info, ctl,
-				       align_gap, align_gap_len);
+				       align_gap, align_gap_len,
+				       align_gap_trim_state);
 	return ret;
 }
 
@@ -2674,7 +2913,7 @@ out:
  * cluster and remove the cluster from it.
  */
 int btrfs_return_cluster_to_free_space(
-			       struct btrfs_block_group_cache *block_group,
+			       struct btrfs_block_group *block_group,
 			       struct btrfs_free_cluster *cluster)
 {
 	struct btrfs_free_space_ctl *ctl;
@@ -2703,12 +2942,14 @@ int btrfs_return_cluster_to_free_space(
 	ret = __btrfs_return_cluster_to_free_space(block_group, cluster);
 	spin_unlock(&ctl->tree_lock);
 
+	btrfs_discard_queue_work(&block_group->fs_info->discard_ctl, block_group);
+
 	/* finally drop our ref */
 	btrfs_put_block_group(block_group);
 	return ret;
 }
 
-static u64 btrfs_alloc_from_bitmap(struct btrfs_block_group_cache *block_group,
+static u64 btrfs_alloc_from_bitmap(struct btrfs_block_group *block_group,
 				   struct btrfs_free_cluster *cluster,
 				   struct btrfs_free_space *entry,
 				   u64 bytes, u64 min_start,
@@ -2741,11 +2982,13 @@ static u64 btrfs_alloc_from_bitmap(struct btrfs_block_group_cache *block_group,
  * if it couldn't find anything suitably large, or a logical disk offset
  * if things worked out
  */
-u64 btrfs_alloc_from_cluster(struct btrfs_block_group_cache *block_group,
+u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group,
 			     struct btrfs_free_cluster *cluster, u64 bytes,
 			     u64 min_start, u64 *max_extent_size)
 {
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
+	struct btrfs_discard_ctl *discard_ctl =
+					&block_group->fs_info->discard_ctl;
 	struct btrfs_free_space *entry = NULL;
 	struct rb_node *node;
 	u64 ret = 0;
@@ -2810,7 +3053,12 @@ out:
 
 	spin_lock(&ctl->tree_lock);
 
+	if (!btrfs_free_space_trimmed(entry))
+		atomic64_add(bytes, &discard_ctl->discard_bytes_saved);
+
 	ctl->free_space -= bytes;
+	if (!entry->bitmap && !btrfs_free_space_trimmed(entry))
+		ctl->discardable_bytes[BTRFS_STAT_CURR] -= bytes;
 	if (entry->bytes == 0) {
 		ctl->free_extents--;
 		if (entry->bitmap) {
@@ -2818,6 +3066,8 @@ out:
 					entry->bitmap);
 			ctl->total_bitmaps--;
 			ctl->op->recalc_thresholds(ctl);
+		} else if (!btrfs_free_space_trimmed(entry)) {
+			ctl->discardable_extents[BTRFS_STAT_CURR]--;
 		}
 		kmem_cache_free(btrfs_free_space_cachep, entry);
 	}
@@ -2827,7 +3077,7 @@ out:
 	return ret;
 }
 
-static int btrfs_bitmap_cluster(struct btrfs_block_group_cache *block_group,
+static int btrfs_bitmap_cluster(struct btrfs_block_group *block_group,
 				struct btrfs_free_space *entry,
 				struct btrfs_free_cluster *cluster,
 				u64 offset, u64 bytes,
@@ -2909,7 +3159,7 @@ again:
  * extent of cont1_bytes, and other clusters of at least min_bytes.
  */
 static noinline int
-setup_cluster_no_bitmap(struct btrfs_block_group_cache *block_group,
+setup_cluster_no_bitmap(struct btrfs_block_group *block_group,
 			struct btrfs_free_cluster *cluster,
 			struct list_head *bitmaps, u64 offset, u64 bytes,
 			u64 cont1_bytes, u64 min_bytes)
@@ -3000,7 +3250,7 @@ setup_cluster_no_bitmap(struct btrfs_block_group_cache *block_group,
  * that we have already failed to find extents that will work.
  */
 static noinline int
-setup_cluster_bitmap(struct btrfs_block_group_cache *block_group,
+setup_cluster_bitmap(struct btrfs_block_group *block_group,
 		     struct btrfs_free_cluster *cluster,
 		     struct list_head *bitmaps, u64 offset, u64 bytes,
 		     u64 cont1_bytes, u64 min_bytes)
@@ -3050,7 +3300,7 @@ setup_cluster_bitmap(struct btrfs_block_group_cache *block_group,
  * returns zero and sets up cluster if things worked out, otherwise
  * it returns -enospc
  */
-int btrfs_find_space_cluster(struct btrfs_block_group_cache *block_group,
+int btrfs_find_space_cluster(struct btrfs_block_group *block_group,
 			     struct btrfs_free_cluster *cluster,
 			     u64 offset, u64 bytes, u64 empty_size)
 {
@@ -3141,9 +3391,10 @@ void btrfs_init_free_cluster(struct btrfs_free_cluster *cluster)
 	cluster->block_group = NULL;
 }
 
-static int do_trimming(struct btrfs_block_group_cache *block_group,
+static int do_trimming(struct btrfs_block_group *block_group,
 		       u64 *total_trimmed, u64 start, u64 bytes,
 		       u64 reserved_start, u64 reserved_bytes,
+		       enum btrfs_trim_state reserved_trim_state,
 		       struct btrfs_trim_range *trim_entry)
 {
 	struct btrfs_space_info *space_info = block_group->space_info;
@@ -3151,6 +3402,9 @@ static int do_trimming(struct btrfs_block_group_cache *block_group,
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	int ret;
 	int update = 0;
+	const u64 end = start + bytes;
+	const u64 reserved_end = reserved_start + reserved_bytes;
+	enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
 	u64 trimmed = 0;
 
 	spin_lock(&space_info->lock);
@@ -3164,11 +3418,20 @@ static int do_trimming(struct btrfs_block_group_cache *block_group,
 	spin_unlock(&space_info->lock);
 
 	ret = btrfs_discard_extent(fs_info, start, bytes, &trimmed);
-	if (!ret)
+	if (!ret) {
 		*total_trimmed += trimmed;
+		trim_state = BTRFS_TRIM_STATE_TRIMMED;
+	}
 
 	mutex_lock(&ctl->cache_writeout_mutex);
-	btrfs_add_free_space(block_group, reserved_start, reserved_bytes);
+	if (reserved_start < start)
+		__btrfs_add_free_space(fs_info, ctl, reserved_start,
+				       start - reserved_start,
+				       reserved_trim_state);
+	if (start + bytes < reserved_start + reserved_bytes)
+		__btrfs_add_free_space(fs_info, ctl, end, reserved_end - end,
+				       reserved_trim_state);
+	__btrfs_add_free_space(fs_info, ctl, start, bytes, trim_state);
 	list_del(&trim_entry->list);
 	mutex_unlock(&ctl->cache_writeout_mutex);
 
@@ -3186,16 +3449,24 @@ static int do_trimming(struct btrfs_block_group_cache *block_group,
 	return ret;
 }
 
-static int trim_no_bitmap(struct btrfs_block_group_cache *block_group,
-			  u64 *total_trimmed, u64 start, u64 end, u64 minlen)
+/*
+ * If @async is set, then we will trim 1 region and return.
+ */
+static int trim_no_bitmap(struct btrfs_block_group *block_group,
+			  u64 *total_trimmed, u64 start, u64 end, u64 minlen,
+			  bool async)
 {
+	struct btrfs_discard_ctl *discard_ctl =
+					&block_group->fs_info->discard_ctl;
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	struct btrfs_free_space *entry;
 	struct rb_node *node;
 	int ret = 0;
 	u64 extent_start;
 	u64 extent_bytes;
+	enum btrfs_trim_state extent_trim_state;
 	u64 bytes;
+	const u64 max_discard_size = READ_ONCE(discard_ctl->max_discard_size);
 
 	while (start < end) {
 		struct btrfs_trim_range trim_entry;
@@ -3203,49 +3474,66 @@ static int trim_no_bitmap(struct btrfs_block_group_cache *block_group,
 		mutex_lock(&ctl->cache_writeout_mutex);
 		spin_lock(&ctl->tree_lock);
 
-		if (ctl->free_space < minlen) {
-			spin_unlock(&ctl->tree_lock);
-			mutex_unlock(&ctl->cache_writeout_mutex);
-			break;
-		}
+		if (ctl->free_space < minlen)
+			goto out_unlock;
 
 		entry = tree_search_offset(ctl, start, 0, 1);
-		if (!entry) {
-			spin_unlock(&ctl->tree_lock);
-			mutex_unlock(&ctl->cache_writeout_mutex);
-			break;
-		}
+		if (!entry)
+			goto out_unlock;
 
-		/* skip bitmaps */
-		while (entry->bitmap) {
+		/* Skip bitmaps and if async, already trimmed entries */
+		while (entry->bitmap ||
+		       (async && btrfs_free_space_trimmed(entry))) {
 			node = rb_next(&entry->offset_index);
-			if (!node) {
-				spin_unlock(&ctl->tree_lock);
-				mutex_unlock(&ctl->cache_writeout_mutex);
-				goto out;
-			}
+			if (!node)
+				goto out_unlock;
 			entry = rb_entry(node, struct btrfs_free_space,
 					 offset_index);
 		}
 
-		if (entry->offset >= end) {
-			spin_unlock(&ctl->tree_lock);
-			mutex_unlock(&ctl->cache_writeout_mutex);
-			break;
-		}
+		if (entry->offset >= end)
+			goto out_unlock;
 
 		extent_start = entry->offset;
 		extent_bytes = entry->bytes;
-		start = max(start, extent_start);
-		bytes = min(extent_start + extent_bytes, end) - start;
-		if (bytes < minlen) {
-			spin_unlock(&ctl->tree_lock);
-			mutex_unlock(&ctl->cache_writeout_mutex);
-			goto next;
-		}
+		extent_trim_state = entry->trim_state;
+		if (async) {
+			start = entry->offset;
+			bytes = entry->bytes;
+			if (bytes < minlen) {
+				spin_unlock(&ctl->tree_lock);
+				mutex_unlock(&ctl->cache_writeout_mutex);
+				goto next;
+			}
+			unlink_free_space(ctl, entry);
+			/*
+			 * Let bytes = BTRFS_MAX_DISCARD_SIZE + X.
+			 * If X < BTRFS_ASYNC_DISCARD_MIN_FILTER, we won't trim
+			 * X when we come back around.  So trim it now.
+			 */
+			if (max_discard_size &&
+			    bytes >= (max_discard_size +
+				      BTRFS_ASYNC_DISCARD_MIN_FILTER)) {
+				bytes = max_discard_size;
+				extent_bytes = max_discard_size;
+				entry->offset += max_discard_size;
+				entry->bytes -= max_discard_size;
+				link_free_space(ctl, entry);
+			} else {
+				kmem_cache_free(btrfs_free_space_cachep, entry);
+			}
+		} else {
+			start = max(start, extent_start);
+			bytes = min(extent_start + extent_bytes, end) - start;
+			if (bytes < minlen) {
+				spin_unlock(&ctl->tree_lock);
+				mutex_unlock(&ctl->cache_writeout_mutex);
+				goto next;
+			}
 
-		unlink_free_space(ctl, entry);
-		kmem_cache_free(btrfs_free_space_cachep, entry);
+			unlink_free_space(ctl, entry);
+			kmem_cache_free(btrfs_free_space_cachep, entry);
+		}
 
 		spin_unlock(&ctl->tree_lock);
 		trim_entry.start = extent_start;
@@ -3254,11 +3542,17 @@ static int trim_no_bitmap(struct btrfs_block_group_cache *block_group,
 		mutex_unlock(&ctl->cache_writeout_mutex);
 
 		ret = do_trimming(block_group, total_trimmed, start, bytes,
-				  extent_start, extent_bytes, &trim_entry);
-		if (ret)
+				  extent_start, extent_bytes, extent_trim_state,
+				  &trim_entry);
+		if (ret) {
+			block_group->discard_cursor = start + bytes;
 			break;
+		}
 next:
 		start += bytes;
+		block_group->discard_cursor = start;
+		if (async && *total_trimmed)
+			break;
 
 		if (fatal_signal_pending(current)) {
 			ret = -ERESTARTSYS;
@@ -3267,19 +3561,76 @@ next:
 
 		cond_resched();
 	}
-out:
+
+	return ret;
+
+out_unlock:
+	block_group->discard_cursor = btrfs_block_group_end(block_group);
+	spin_unlock(&ctl->tree_lock);
+	mutex_unlock(&ctl->cache_writeout_mutex);
+
 	return ret;
 }
 
-static int trim_bitmaps(struct btrfs_block_group_cache *block_group,
-			u64 *total_trimmed, u64 start, u64 end, u64 minlen)
+/*
+ * If we break out of trimming a bitmap prematurely, we should reset the
+ * trimming bit.  In a rather contrieved case, it's possible to race here so
+ * reset the state to BTRFS_TRIM_STATE_UNTRIMMED.
+ *
+ * start = start of bitmap
+ * end = near end of bitmap
+ *
+ * Thread 1:			Thread 2:
+ * trim_bitmaps(start)
+ *				trim_bitmaps(end)
+ *				end_trimming_bitmap()
+ * reset_trimming_bitmap()
+ */
+static void reset_trimming_bitmap(struct btrfs_free_space_ctl *ctl, u64 offset)
+{
+	struct btrfs_free_space *entry;
+
+	spin_lock(&ctl->tree_lock);
+	entry = tree_search_offset(ctl, offset, 1, 0);
+	if (entry) {
+		if (btrfs_free_space_trimmed(entry)) {
+			ctl->discardable_extents[BTRFS_STAT_CURR] +=
+				entry->bitmap_extents;
+			ctl->discardable_bytes[BTRFS_STAT_CURR] += entry->bytes;
+		}
+		entry->trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
+	}
+
+	spin_unlock(&ctl->tree_lock);
+}
+
+static void end_trimming_bitmap(struct btrfs_free_space_ctl *ctl,
+				struct btrfs_free_space *entry)
+{
+	if (btrfs_free_space_trimming_bitmap(entry)) {
+		entry->trim_state = BTRFS_TRIM_STATE_TRIMMED;
+		ctl->discardable_extents[BTRFS_STAT_CURR] -=
+			entry->bitmap_extents;
+		ctl->discardable_bytes[BTRFS_STAT_CURR] -= entry->bytes;
+	}
+}
+
+/*
+ * If @async is set, then we will trim 1 region and return.
+ */
+static int trim_bitmaps(struct btrfs_block_group *block_group,
+			u64 *total_trimmed, u64 start, u64 end, u64 minlen,
+			u64 maxlen, bool async)
 {
+	struct btrfs_discard_ctl *discard_ctl =
+					&block_group->fs_info->discard_ctl;
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	struct btrfs_free_space *entry;
 	int ret = 0;
 	int ret2;
 	u64 bytes;
 	u64 offset = offset_to_bitmap(ctl, start);
+	const u64 max_discard_size = READ_ONCE(discard_ctl->max_discard_size);
 
 	while (offset < end) {
 		bool next_bitmap = false;
@@ -3289,35 +3640,84 @@ static int trim_bitmaps(struct btrfs_block_group_cache *block_group,
 		spin_lock(&ctl->tree_lock);
 
 		if (ctl->free_space < minlen) {
+			block_group->discard_cursor =
+				btrfs_block_group_end(block_group);
 			spin_unlock(&ctl->tree_lock);
 			mutex_unlock(&ctl->cache_writeout_mutex);
 			break;
 		}
 
 		entry = tree_search_offset(ctl, offset, 1, 0);
-		if (!entry) {
+		/*
+		 * Bitmaps are marked trimmed lossily now to prevent constant
+		 * discarding of the same bitmap (the reason why we are bound
+		 * by the filters).  So, retrim the block group bitmaps when we
+		 * are preparing to punt to the unused_bgs list.  This uses
+		 * @minlen to determine if we are in BTRFS_DISCARD_INDEX_UNUSED
+		 * which is the only discard index which sets minlen to 0.
+		 */
+		if (!entry || (async && minlen && start == offset &&
+			       btrfs_free_space_trimmed(entry))) {
 			spin_unlock(&ctl->tree_lock);
 			mutex_unlock(&ctl->cache_writeout_mutex);
 			next_bitmap = true;
 			goto next;
 		}
 
+		/*
+		 * Async discard bitmap trimming begins at by setting the start
+		 * to be key.objectid and the offset_to_bitmap() aligns to the
+		 * start of the bitmap.  This lets us know we are fully
+		 * scanning the bitmap rather than only some portion of it.
+		 */
+		if (start == offset)
+			entry->trim_state = BTRFS_TRIM_STATE_TRIMMING;
+
 		bytes = minlen;
 		ret2 = search_bitmap(ctl, entry, &start, &bytes, false);
 		if (ret2 || start >= end) {
+			/*
+			 * We lossily consider a bitmap trimmed if we only skip
+			 * over regions <= BTRFS_ASYNC_DISCARD_MIN_FILTER.
+			 */
+			if (ret2 && minlen <= BTRFS_ASYNC_DISCARD_MIN_FILTER)
+				end_trimming_bitmap(ctl, entry);
+			else
+				entry->trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
 			spin_unlock(&ctl->tree_lock);
 			mutex_unlock(&ctl->cache_writeout_mutex);
 			next_bitmap = true;
 			goto next;
 		}
 
+		/*
+		 * We already trimmed a region, but are using the locking above
+		 * to reset the trim_state.
+		 */
+		if (async && *total_trimmed) {
+			spin_unlock(&ctl->tree_lock);
+			mutex_unlock(&ctl->cache_writeout_mutex);
+			goto out;
+		}
+
 		bytes = min(bytes, end - start);
-		if (bytes < minlen) {
+		if (bytes < minlen || (async && maxlen && bytes > maxlen)) {
 			spin_unlock(&ctl->tree_lock);
 			mutex_unlock(&ctl->cache_writeout_mutex);
 			goto next;
 		}
 
+		/*
+		 * Let bytes = BTRFS_MAX_DISCARD_SIZE + X.
+		 * If X < @minlen, we won't trim X when we come back around.
+		 * So trim it now.  We differ here from trimming extents as we
+		 * don't keep individual state per bit.
+		 */
+		if (async &&
+		    max_discard_size &&
+		    bytes > (max_discard_size + minlen))
+			bytes = max_discard_size;
+
 		bitmap_clear_bits(ctl, entry, start, bytes);
 		if (entry->bytes == 0)
 			free_bitmap(ctl, entry);
@@ -3329,19 +3729,25 @@ static int trim_bitmaps(struct btrfs_block_group_cache *block_group,
 		mutex_unlock(&ctl->cache_writeout_mutex);
 
 		ret = do_trimming(block_group, total_trimmed, start, bytes,
-				  start, bytes, &trim_entry);
-		if (ret)
+				  start, bytes, 0, &trim_entry);
+		if (ret) {
+			reset_trimming_bitmap(ctl, offset);
+			block_group->discard_cursor =
+				btrfs_block_group_end(block_group);
 			break;
+		}
 next:
 		if (next_bitmap) {
 			offset += BITS_PER_BITMAP * ctl->unit;
+			start = offset;
 		} else {
 			start += bytes;
-			if (start >= offset + BITS_PER_BITMAP * ctl->unit)
-				offset += BITS_PER_BITMAP * ctl->unit;
 		}
+		block_group->discard_cursor = start;
 
 		if (fatal_signal_pending(current)) {
+			if (start != offset)
+				reset_trimming_bitmap(ctl, offset);
 			ret = -ERESTARTSYS;
 			break;
 		}
@@ -3349,15 +3755,19 @@ next:
 		cond_resched();
 	}
 
+	if (offset >= end)
+		block_group->discard_cursor = end;
+
+out:
 	return ret;
 }
 
-void btrfs_get_block_group_trimming(struct btrfs_block_group_cache *cache)
+void btrfs_get_block_group_trimming(struct btrfs_block_group *cache)
 {
 	atomic_inc(&cache->trimming);
 }
 
-void btrfs_put_block_group_trimming(struct btrfs_block_group_cache *block_group)
+void btrfs_put_block_group_trimming(struct btrfs_block_group *block_group)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
 	struct extent_map_tree *em_tree;
@@ -3373,7 +3783,7 @@ void btrfs_put_block_group_trimming(struct btrfs_block_group_cache *block_group)
 		mutex_lock(&fs_info->chunk_mutex);
 		em_tree = &fs_info->mapping_tree;
 		write_lock(&em_tree->lock);
-		em = lookup_extent_mapping(em_tree, block_group->key.objectid,
+		em = lookup_extent_mapping(em_tree, block_group->start,
 					   1);
 		BUG_ON(!em); /* logic error, can't happen */
 		remove_extent_mapping(em_tree, em);
@@ -3392,10 +3802,12 @@ void btrfs_put_block_group_trimming(struct btrfs_block_group_cache *block_group)
 	}
 }
 
-int btrfs_trim_block_group(struct btrfs_block_group_cache *block_group,
+int btrfs_trim_block_group(struct btrfs_block_group *block_group,
 			   u64 *trimmed, u64 start, u64 end, u64 minlen)
 {
+	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	int ret;
+	u64 rem = 0;
 
 	*trimmed = 0;
 
@@ -3407,16 +3819,66 @@ int btrfs_trim_block_group(struct btrfs_block_group_cache *block_group,
 	btrfs_get_block_group_trimming(block_group);
 	spin_unlock(&block_group->lock);
 
-	ret = trim_no_bitmap(block_group, trimmed, start, end, minlen);
+	ret = trim_no_bitmap(block_group, trimmed, start, end, minlen, false);
 	if (ret)
 		goto out;
 
-	ret = trim_bitmaps(block_group, trimmed, start, end, minlen);
+	ret = trim_bitmaps(block_group, trimmed, start, end, minlen, 0, false);
+	div64_u64_rem(end, BITS_PER_BITMAP * ctl->unit, &rem);
+	/* If we ended in the middle of a bitmap, reset the trimming flag */
+	if (rem)
+		reset_trimming_bitmap(ctl, offset_to_bitmap(ctl, end));
 out:
 	btrfs_put_block_group_trimming(block_group);
 	return ret;
 }
 
+int btrfs_trim_block_group_extents(struct btrfs_block_group *block_group,
+				   u64 *trimmed, u64 start, u64 end, u64 minlen,
+				   bool async)
+{
+	int ret;
+
+	*trimmed = 0;
+
+	spin_lock(&block_group->lock);
+	if (block_group->removed) {
+		spin_unlock(&block_group->lock);
+		return 0;
+	}
+	btrfs_get_block_group_trimming(block_group);
+	spin_unlock(&block_group->lock);
+
+	ret = trim_no_bitmap(block_group, trimmed, start, end, minlen, async);
+	btrfs_put_block_group_trimming(block_group);
+
+	return ret;
+}
+
+int btrfs_trim_block_group_bitmaps(struct btrfs_block_group *block_group,
+				   u64 *trimmed, u64 start, u64 end, u64 minlen,
+				   u64 maxlen, bool async)
+{
+	int ret;
+
+	*trimmed = 0;
+
+	spin_lock(&block_group->lock);
+	if (block_group->removed) {
+		spin_unlock(&block_group->lock);
+		return 0;
+	}
+	btrfs_get_block_group_trimming(block_group);
+	spin_unlock(&block_group->lock);
+
+	ret = trim_bitmaps(block_group, trimmed, start, end, minlen, maxlen,
+			   async);
+
+	btrfs_put_block_group_trimming(block_group);
+
+	return ret;
+}
+
 /*
  * Find the left-most item in the cache tree, and then return the
  * smallest inode number in the item.
@@ -3573,7 +4035,7 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
 		if (release_metadata)
 			btrfs_delalloc_release_metadata(BTRFS_I(inode),
 					inode->i_size, true);
-#ifdef DEBUG
+#ifdef CONFIG_BTRFS_DEBUG
 		btrfs_err(fs_info,
 			  "failed to write free ino cache for root %llu",
 			  root->root_key.objectid);
@@ -3590,12 +4052,13 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
  * how the free space cache loading stuff works, so you can get really weird
  * configurations.
  */
-int test_add_free_space_entry(struct btrfs_block_group_cache *cache,
+int test_add_free_space_entry(struct btrfs_block_group *cache,
 			      u64 offset, u64 bytes, bool bitmap)
 {
 	struct btrfs_free_space_ctl *ctl = cache->free_space_ctl;
 	struct btrfs_free_space *info = NULL, *bitmap_info;
 	void *map = NULL;
+	enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_TRIMMED;
 	u64 bytes_added;
 	int ret;
 
@@ -3637,7 +4100,8 @@ again:
 		info = NULL;
 	}
 
-	bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes);
+	bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes,
+					  trim_state);
 
 	bytes -= bytes_added;
 	offset += bytes_added;
@@ -3658,7 +4122,7 @@ again:
  * just used to check the absence of space, so if there is free space in the
  * range at all we will return 1.
  */
-int test_check_exists(struct btrfs_block_group_cache *cache,
+int test_check_exists(struct btrfs_block_group *cache,
 		      u64 offset, u64 bytes)
 {
 	struct btrfs_free_space_ctl *ctl = cache->free_space_ctl;
diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h
index 39c32c8fc24f..2e0a8077aa74 100644
--- a/fs/btrfs/free-space-cache.h
+++ b/fs/btrfs/free-space-cache.h
@@ -6,6 +6,20 @@
 #ifndef BTRFS_FREE_SPACE_CACHE_H
 #define BTRFS_FREE_SPACE_CACHE_H
 
+/*
+ * This is the trim state of an extent or bitmap.
+ *
+ * BTRFS_TRIM_STATE_TRIMMING is special and used to maintain the state of a
+ * bitmap as we may need several trims to fully trim a single bitmap entry.
+ * This is reset should any free space other than trimmed space be added to the
+ * bitmap.
+ */
+enum btrfs_trim_state {
+	BTRFS_TRIM_STATE_UNTRIMMED,
+	BTRFS_TRIM_STATE_TRIMMED,
+	BTRFS_TRIM_STATE_TRIMMING,
+};
+
 struct btrfs_free_space {
 	struct rb_node offset_index;
 	u64 offset;
@@ -13,8 +27,21 @@ struct btrfs_free_space {
 	u64 max_extent_size;
 	unsigned long *bitmap;
 	struct list_head list;
+	enum btrfs_trim_state trim_state;
+	s32 bitmap_extents;
 };
 
+static inline bool btrfs_free_space_trimmed(struct btrfs_free_space *info)
+{
+	return (info->trim_state == BTRFS_TRIM_STATE_TRIMMED);
+}
+
+static inline bool btrfs_free_space_trimming_bitmap(
+					    struct btrfs_free_space *info)
+{
+	return (info->trim_state == BTRFS_TRIM_STATE_TRIMMING);
+}
+
 struct btrfs_free_space_ctl {
 	spinlock_t tree_lock;
 	struct rb_root free_space_offset;
@@ -24,6 +51,8 @@ struct btrfs_free_space_ctl {
 	int total_bitmaps;
 	int unit;
 	u64 start;
+	s32 discardable_extents[BTRFS_STAT_NR_ENTRIES];
+	s64 discardable_bytes[BTRFS_STAT_NR_ENTRIES];
 	const struct btrfs_free_space_op *op;
 	void *private;
 	struct mutex cache_writeout_mutex;
@@ -50,24 +79,23 @@ struct btrfs_io_ctl {
 	unsigned check_crcs:1;
 };
 
-struct inode *lookup_free_space_inode(
-		struct btrfs_block_group_cache *block_group,
+struct inode *lookup_free_space_inode(struct btrfs_block_group *block_group,
 		struct btrfs_path *path);
 int create_free_space_inode(struct btrfs_trans_handle *trans,
-			    struct btrfs_block_group_cache *block_group,
+			    struct btrfs_block_group *block_group,
 			    struct btrfs_path *path);
 
 int btrfs_check_trunc_cache_free_space(struct btrfs_fs_info *fs_info,
 				       struct btrfs_block_rsv *rsv);
 int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans,
-				    struct btrfs_block_group_cache *block_group,
+				    struct btrfs_block_group *block_group,
 				    struct inode *inode);
-int load_free_space_cache(struct btrfs_block_group_cache *block_group);
+int load_free_space_cache(struct btrfs_block_group *block_group);
 int btrfs_wait_cache_io(struct btrfs_trans_handle *trans,
-			struct btrfs_block_group_cache *block_group,
+			struct btrfs_block_group *block_group,
 			struct btrfs_path *path);
 int btrfs_write_out_cache(struct btrfs_trans_handle *trans,
-			  struct btrfs_block_group_cache *block_group,
+			  struct btrfs_block_group *block_group,
 			  struct btrfs_path *path);
 struct inode *lookup_free_ino_inode(struct btrfs_root *root,
 				    struct btrfs_path *path);
@@ -81,42 +109,50 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
 			      struct btrfs_path *path,
 			      struct inode *inode);
 
-void btrfs_init_free_space_ctl(struct btrfs_block_group_cache *block_group);
+void btrfs_init_free_space_ctl(struct btrfs_block_group *block_group);
 int __btrfs_add_free_space(struct btrfs_fs_info *fs_info,
 			   struct btrfs_free_space_ctl *ctl,
-			   u64 bytenr, u64 size);
-int btrfs_add_free_space(struct btrfs_block_group_cache *block_group,
+			   u64 bytenr, u64 size,
+			   enum btrfs_trim_state trim_state);
+int btrfs_add_free_space(struct btrfs_block_group *block_group,
 			 u64 bytenr, u64 size);
-int btrfs_remove_free_space(struct btrfs_block_group_cache *block_group,
+int btrfs_add_free_space_async_trimmed(struct btrfs_block_group *block_group,
+				       u64 bytenr, u64 size);
+int btrfs_remove_free_space(struct btrfs_block_group *block_group,
 			    u64 bytenr, u64 size);
 void __btrfs_remove_free_space_cache(struct btrfs_free_space_ctl *ctl);
-void btrfs_remove_free_space_cache(struct btrfs_block_group_cache
-				     *block_group);
-u64 btrfs_find_space_for_alloc(struct btrfs_block_group_cache *block_group,
+void btrfs_remove_free_space_cache(struct btrfs_block_group *block_group);
+bool btrfs_is_free_space_trimmed(struct btrfs_block_group *block_group);
+u64 btrfs_find_space_for_alloc(struct btrfs_block_group *block_group,
 			       u64 offset, u64 bytes, u64 empty_size,
 			       u64 *max_extent_size);
 u64 btrfs_find_ino_for_alloc(struct btrfs_root *fs_root);
-void btrfs_dump_free_space(struct btrfs_block_group_cache *block_group,
+void btrfs_dump_free_space(struct btrfs_block_group *block_group,
 			   u64 bytes);
-int btrfs_find_space_cluster(struct btrfs_block_group_cache *block_group,
+int btrfs_find_space_cluster(struct btrfs_block_group *block_group,
 			     struct btrfs_free_cluster *cluster,
 			     u64 offset, u64 bytes, u64 empty_size);
 void btrfs_init_free_cluster(struct btrfs_free_cluster *cluster);
-u64 btrfs_alloc_from_cluster(struct btrfs_block_group_cache *block_group,
+u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group,
 			     struct btrfs_free_cluster *cluster, u64 bytes,
 			     u64 min_start, u64 *max_extent_size);
 int btrfs_return_cluster_to_free_space(
-			       struct btrfs_block_group_cache *block_group,
+			       struct btrfs_block_group *block_group,
 			       struct btrfs_free_cluster *cluster);
-int btrfs_trim_block_group(struct btrfs_block_group_cache *block_group,
+int btrfs_trim_block_group(struct btrfs_block_group *block_group,
 			   u64 *trimmed, u64 start, u64 end, u64 minlen);
+int btrfs_trim_block_group_extents(struct btrfs_block_group *block_group,
+				   u64 *trimmed, u64 start, u64 end, u64 minlen,
+				   bool async);
+int btrfs_trim_block_group_bitmaps(struct btrfs_block_group *block_group,
+				   u64 *trimmed, u64 start, u64 end, u64 minlen,
+				   u64 maxlen, bool async);
 
 /* Support functions for running our sanity tests */
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
-int test_add_free_space_entry(struct btrfs_block_group_cache *cache,
+int test_add_free_space_entry(struct btrfs_block_group *cache,
 			      u64 offset, u64 bytes, bool bitmap);
-int test_check_exists(struct btrfs_block_group_cache *cache,
-		      u64 offset, u64 bytes);
+int test_check_exists(struct btrfs_block_group *cache, u64 offset, u64 bytes);
 #endif
 
 #endif
diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c
index 48a03f5240f5..8b1f5c8897b7 100644
--- a/fs/btrfs/free-space-tree.c
+++ b/fs/btrfs/free-space-tree.c
@@ -13,10 +13,10 @@
 #include "block-group.h"
 
 static int __add_block_group_free_space(struct btrfs_trans_handle *trans,
-					struct btrfs_block_group_cache *block_group,
+					struct btrfs_block_group *block_group,
 					struct btrfs_path *path);
 
-void set_free_space_tree_thresholds(struct btrfs_block_group_cache *cache)
+void set_free_space_tree_thresholds(struct btrfs_block_group *cache)
 {
 	u32 bitmap_range;
 	size_t bitmap_size;
@@ -27,8 +27,7 @@ void set_free_space_tree_thresholds(struct btrfs_block_group_cache *cache)
 	 * exceeds that required for using bitmaps.
 	 */
 	bitmap_range = cache->fs_info->sectorsize * BTRFS_FREE_SPACE_BITMAP_BITS;
-	num_bitmaps = div_u64(cache->key.offset + bitmap_range - 1,
-			      bitmap_range);
+	num_bitmaps = div_u64(cache->length + bitmap_range - 1, bitmap_range);
 	bitmap_size = sizeof(struct btrfs_item) + BTRFS_FREE_SPACE_BITMAP_SIZE;
 	total_bitmap_size = num_bitmaps * bitmap_size;
 	cache->bitmap_high_thresh = div_u64(total_bitmap_size,
@@ -45,7 +44,7 @@ void set_free_space_tree_thresholds(struct btrfs_block_group_cache *cache)
 }
 
 static int add_new_free_space_info(struct btrfs_trans_handle *trans,
-				   struct btrfs_block_group_cache *block_group,
+				   struct btrfs_block_group *block_group,
 				   struct btrfs_path *path)
 {
 	struct btrfs_root *root = trans->fs_info->free_space_root;
@@ -54,9 +53,9 @@ static int add_new_free_space_info(struct btrfs_trans_handle *trans,
 	struct extent_buffer *leaf;
 	int ret;
 
-	key.objectid = block_group->key.objectid;
+	key.objectid = block_group->start;
 	key.type = BTRFS_FREE_SPACE_INFO_KEY;
-	key.offset = block_group->key.offset;
+	key.offset = block_group->length;
 
 	ret = btrfs_insert_empty_item(trans, root, path, &key, sizeof(*info));
 	if (ret)
@@ -78,7 +77,7 @@ out:
 EXPORT_FOR_TESTS
 struct btrfs_free_space_info *search_free_space_info(
 		struct btrfs_trans_handle *trans,
-		struct btrfs_block_group_cache *block_group,
+		struct btrfs_block_group *block_group,
 		struct btrfs_path *path, int cow)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
@@ -86,16 +85,16 @@ struct btrfs_free_space_info *search_free_space_info(
 	struct btrfs_key key;
 	int ret;
 
-	key.objectid = block_group->key.objectid;
+	key.objectid = block_group->start;
 	key.type = BTRFS_FREE_SPACE_INFO_KEY;
-	key.offset = block_group->key.offset;
+	key.offset = block_group->length;
 
 	ret = btrfs_search_slot(trans, root, &key, path, 0, cow);
 	if (ret < 0)
 		return ERR_PTR(ret);
 	if (ret != 0) {
 		btrfs_warn(fs_info, "missing free space info for %llu",
-			   block_group->key.objectid);
+			   block_group->start);
 		ASSERT(0);
 		return ERR_PTR(-ENOENT);
 	}
@@ -180,7 +179,7 @@ static void le_bitmap_set(unsigned long *map, unsigned int start, int len)
 
 EXPORT_FOR_TESTS
 int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
-				  struct btrfs_block_group_cache *block_group,
+				  struct btrfs_block_group *block_group,
 				  struct btrfs_path *path)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
@@ -197,7 +196,7 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 	int done = 0, nr;
 	int ret;
 
-	bitmap_size = free_space_bitmap_size(block_group->key.offset,
+	bitmap_size = free_space_bitmap_size(block_group->length,
 					     fs_info->sectorsize);
 	bitmap = alloc_bitmap(bitmap_size);
 	if (!bitmap) {
@@ -205,8 +204,8 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	start = block_group->key.objectid;
-	end = block_group->key.objectid + block_group->key.offset;
+	start = block_group->start;
+	end = block_group->start + block_group->length;
 
 	key.objectid = end - 1;
 	key.type = (u8)-1;
@@ -224,8 +223,8 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 			btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0] - 1);
 
 			if (found_key.type == BTRFS_FREE_SPACE_INFO_KEY) {
-				ASSERT(found_key.objectid == block_group->key.objectid);
-				ASSERT(found_key.offset == block_group->key.offset);
+				ASSERT(found_key.objectid == block_group->start);
+				ASSERT(found_key.offset == block_group->length);
 				done = 1;
 				break;
 			} else if (found_key.type == BTRFS_FREE_SPACE_EXTENT_KEY) {
@@ -271,7 +270,7 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
 	if (extent_count != expected_extent_count) {
 		btrfs_err(fs_info,
 			  "incorrect extent count for %llu; counted %u, expected %u",
-			  block_group->key.objectid, extent_count,
+			  block_group->start, extent_count,
 			  expected_extent_count);
 		ASSERT(0);
 		ret = -EIO;
@@ -320,7 +319,7 @@ out:
 
 EXPORT_FOR_TESTS
 int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
-				  struct btrfs_block_group_cache *block_group,
+				  struct btrfs_block_group *block_group,
 				  struct btrfs_path *path)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
@@ -336,7 +335,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 	int done = 0, nr;
 	int ret;
 
-	bitmap_size = free_space_bitmap_size(block_group->key.offset,
+	bitmap_size = free_space_bitmap_size(block_group->length,
 					     fs_info->sectorsize);
 	bitmap = alloc_bitmap(bitmap_size);
 	if (!bitmap) {
@@ -344,8 +343,8 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	start = block_group->key.objectid;
-	end = block_group->key.objectid + block_group->key.offset;
+	start = block_group->start;
+	end = block_group->start + block_group->length;
 
 	key.objectid = end - 1;
 	key.type = (u8)-1;
@@ -363,8 +362,8 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 			btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0] - 1);
 
 			if (found_key.type == BTRFS_FREE_SPACE_INFO_KEY) {
-				ASSERT(found_key.objectid == block_group->key.objectid);
-				ASSERT(found_key.offset == block_group->key.offset);
+				ASSERT(found_key.objectid == block_group->start);
+				ASSERT(found_key.offset == block_group->length);
 				done = 1;
 				break;
 			} else if (found_key.type == BTRFS_FREE_SPACE_BITMAP_KEY) {
@@ -413,7 +412,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_release_path(path);
 
-	nrbits = div_u64(block_group->key.offset, block_group->fs_info->sectorsize);
+	nrbits = div_u64(block_group->length, block_group->fs_info->sectorsize);
 	start_bit = find_next_bit_le(bitmap, nrbits, 0);
 
 	while (start_bit < nrbits) {
@@ -437,7 +436,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
 	if (extent_count != expected_extent_count) {
 		btrfs_err(fs_info,
 			  "incorrect extent count for %llu; counted %u, expected %u",
-			  block_group->key.objectid, extent_count,
+			  block_group->start, extent_count,
 			  expected_extent_count);
 		ASSERT(0);
 		ret = -EIO;
@@ -453,7 +452,7 @@ out:
 }
 
 static int update_free_space_extent_count(struct btrfs_trans_handle *trans,
-					  struct btrfs_block_group_cache *block_group,
+					  struct btrfs_block_group *block_group,
 					  struct btrfs_path *path,
 					  int new_extents)
 {
@@ -491,7 +490,7 @@ out:
 }
 
 EXPORT_FOR_TESTS
-int free_space_test_bit(struct btrfs_block_group_cache *block_group,
+int free_space_test_bit(struct btrfs_block_group *block_group,
 			struct btrfs_path *path, u64 offset)
 {
 	struct extent_buffer *leaf;
@@ -513,7 +512,7 @@ int free_space_test_bit(struct btrfs_block_group_cache *block_group,
 	return !!extent_buffer_test_bit(leaf, ptr, i);
 }
 
-static void free_space_set_bits(struct btrfs_block_group_cache *block_group,
+static void free_space_set_bits(struct btrfs_block_group *block_group,
 				struct btrfs_path *path, u64 *start, u64 *size,
 				int bit)
 {
@@ -581,7 +580,7 @@ static int free_space_next_bitmap(struct btrfs_trans_handle *trans,
  * the bitmap.
  */
 static int modify_free_space_bitmap(struct btrfs_trans_handle *trans,
-				    struct btrfs_block_group_cache *block_group,
+				    struct btrfs_block_group *block_group,
 				    struct btrfs_path *path,
 				    u64 start, u64 size, int remove)
 {
@@ -597,7 +596,7 @@ static int modify_free_space_bitmap(struct btrfs_trans_handle *trans,
 	 * Read the bit for the block immediately before the extent of space if
 	 * that block is within the block group.
 	 */
-	if (start > block_group->key.objectid) {
+	if (start > block_group->start) {
 		u64 prev_block = start - block_group->fs_info->sectorsize;
 
 		key.objectid = prev_block;
@@ -649,7 +648,7 @@ static int modify_free_space_bitmap(struct btrfs_trans_handle *trans,
 	 * Read the bit for the block immediately after the extent of space if
 	 * that block is within the block group.
 	 */
-	if (end < block_group->key.objectid + block_group->key.offset) {
+	if (end < block_group->start + block_group->length) {
 		/* The next block may be in the next bitmap. */
 		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
 		if (end >= key.objectid + key.offset) {
@@ -694,7 +693,7 @@ out:
 }
 
 static int remove_free_space_extent(struct btrfs_trans_handle *trans,
-				    struct btrfs_block_group_cache *block_group,
+				    struct btrfs_block_group *block_group,
 				    struct btrfs_path *path,
 				    u64 start, u64 size)
 {
@@ -781,7 +780,7 @@ out:
 
 EXPORT_FOR_TESTS
 int __remove_from_free_space_tree(struct btrfs_trans_handle *trans,
-				  struct btrfs_block_group_cache *block_group,
+				  struct btrfs_block_group *block_group,
 				  struct btrfs_path *path, u64 start, u64 size)
 {
 	struct btrfs_free_space_info *info;
@@ -812,7 +811,7 @@ int __remove_from_free_space_tree(struct btrfs_trans_handle *trans,
 int remove_from_free_space_tree(struct btrfs_trans_handle *trans,
 				u64 start, u64 size)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_path *path;
 	int ret;
 
@@ -846,7 +845,7 @@ out:
 }
 
 static int add_free_space_extent(struct btrfs_trans_handle *trans,
-				 struct btrfs_block_group_cache *block_group,
+				 struct btrfs_block_group *block_group,
 				 struct btrfs_path *path,
 				 u64 start, u64 size)
 {
@@ -880,7 +879,7 @@ static int add_free_space_extent(struct btrfs_trans_handle *trans,
 	new_key.offset = size;
 
 	/* Search for a neighbor on the left. */
-	if (start == block_group->key.objectid)
+	if (start == block_group->start)
 		goto right;
 	key.objectid = start - 1;
 	key.type = (u8)-1;
@@ -900,8 +899,8 @@ static int add_free_space_extent(struct btrfs_trans_handle *trans,
 
 	found_start = key.objectid;
 	found_end = key.objectid + key.offset;
-	ASSERT(found_start >= block_group->key.objectid &&
-	       found_end > block_group->key.objectid);
+	ASSERT(found_start >= block_group->start &&
+	       found_end > block_group->start);
 	ASSERT(found_start < start && found_end <= start);
 
 	/*
@@ -920,7 +919,7 @@ static int add_free_space_extent(struct btrfs_trans_handle *trans,
 
 right:
 	/* Search for a neighbor on the right. */
-	if (end == block_group->key.objectid + block_group->key.offset)
+	if (end == block_group->start + block_group->length)
 		goto insert;
 	key.objectid = end;
 	key.type = (u8)-1;
@@ -940,8 +939,8 @@ right:
 
 	found_start = key.objectid;
 	found_end = key.objectid + key.offset;
-	ASSERT(found_start >= block_group->key.objectid &&
-	       found_end > block_group->key.objectid);
+	ASSERT(found_start >= block_group->start &&
+	       found_end > block_group->start);
 	ASSERT((found_start < start && found_end <= start) ||
 	       (found_start >= end && found_end > end));
 
@@ -974,7 +973,7 @@ out:
 
 EXPORT_FOR_TESTS
 int __add_to_free_space_tree(struct btrfs_trans_handle *trans,
-			     struct btrfs_block_group_cache *block_group,
+			     struct btrfs_block_group *block_group,
 			     struct btrfs_path *path, u64 start, u64 size)
 {
 	struct btrfs_free_space_info *info;
@@ -1005,7 +1004,7 @@ int __add_to_free_space_tree(struct btrfs_trans_handle *trans,
 int add_to_free_space_tree(struct btrfs_trans_handle *trans,
 			   u64 start, u64 size)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_path *path;
 	int ret;
 
@@ -1043,7 +1042,7 @@ out:
  * through the normal add/remove hooks.
  */
 static int populate_free_space_tree(struct btrfs_trans_handle *trans,
-				    struct btrfs_block_group_cache *block_group)
+				    struct btrfs_block_group *block_group)
 {
 	struct btrfs_root *extent_root = trans->fs_info->extent_root;
 	struct btrfs_path *path, *path2;
@@ -1075,7 +1074,7 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans,
 	 * BLOCK_GROUP_ITEM, so an extent may precede the block group that it's
 	 * contained in.
 	 */
-	key.objectid = block_group->key.objectid;
+	key.objectid = block_group->start;
 	key.type = BTRFS_EXTENT_ITEM_KEY;
 	key.offset = 0;
 
@@ -1084,8 +1083,8 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans,
 		goto out_locked;
 	ASSERT(ret == 0);
 
-	start = block_group->key.objectid;
-	end = block_group->key.objectid + block_group->key.offset;
+	start = block_group->start;
+	end = block_group->start + block_group->length;
 	while (1) {
 		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
 
@@ -1109,7 +1108,7 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans,
 			else
 				start += key.offset;
 		} else if (key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) {
-			if (key.objectid != block_group->key.objectid)
+			if (key.objectid != block_group->start)
 				break;
 		}
 
@@ -1140,7 +1139,7 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *tree_root = fs_info->tree_root;
 	struct btrfs_root *free_space_root;
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct rb_node *node;
 	int ret;
 
@@ -1159,7 +1158,7 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
 
 	node = rb_first(&fs_info->block_group_cache_tree);
 	while (node) {
-		block_group = rb_entry(node, struct btrfs_block_group_cache,
+		block_group = rb_entry(node, struct btrfs_block_group,
 				       cache_node);
 		ret = populate_free_space_tree(trans, block_group);
 		if (ret)
@@ -1252,9 +1251,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
 	btrfs_free_tree_block(trans, free_space_root, free_space_root->node,
 			      0, 1);
 
-	free_extent_buffer(free_space_root->node);
-	free_extent_buffer(free_space_root->commit_root);
-	kfree(free_space_root);
+	btrfs_put_root(free_space_root);
 
 	return btrfs_commit_transaction(trans);
 
@@ -1265,7 +1262,7 @@ abort:
 }
 
 static int __add_block_group_free_space(struct btrfs_trans_handle *trans,
-					struct btrfs_block_group_cache *block_group,
+					struct btrfs_block_group *block_group,
 					struct btrfs_path *path)
 {
 	int ret;
@@ -1277,12 +1274,12 @@ static int __add_block_group_free_space(struct btrfs_trans_handle *trans,
 		return ret;
 
 	return __add_to_free_space_tree(trans, block_group, path,
-					block_group->key.objectid,
-					block_group->key.offset);
+					block_group->start,
+					block_group->length);
 }
 
 int add_block_group_free_space(struct btrfs_trans_handle *trans,
-			       struct btrfs_block_group_cache *block_group)
+			       struct btrfs_block_group *block_group)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_path *path = NULL;
@@ -1312,7 +1309,7 @@ out:
 }
 
 int remove_block_group_free_space(struct btrfs_trans_handle *trans,
-				  struct btrfs_block_group_cache *block_group)
+				  struct btrfs_block_group *block_group)
 {
 	struct btrfs_root *root = trans->fs_info->free_space_root;
 	struct btrfs_path *path;
@@ -1336,8 +1333,8 @@ int remove_block_group_free_space(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	start = block_group->key.objectid;
-	end = block_group->key.objectid + block_group->key.offset;
+	start = block_group->start;
+	end = block_group->start + block_group->length;
 
 	key.objectid = end - 1;
 	key.type = (u8)-1;
@@ -1355,8 +1352,8 @@ int remove_block_group_free_space(struct btrfs_trans_handle *trans,
 			btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0] - 1);
 
 			if (found_key.type == BTRFS_FREE_SPACE_INFO_KEY) {
-				ASSERT(found_key.objectid == block_group->key.objectid);
-				ASSERT(found_key.offset == block_group->key.offset);
+				ASSERT(found_key.objectid == block_group->start);
+				ASSERT(found_key.offset == block_group->length);
 				done = 1;
 				nr++;
 				path->slots[0]--;
@@ -1391,7 +1388,7 @@ static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl,
 				   struct btrfs_path *path,
 				   u32 expected_extent_count)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_root *root;
 	struct btrfs_key key;
@@ -1407,7 +1404,7 @@ static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl,
 	fs_info = block_group->fs_info;
 	root = fs_info->free_space_root;
 
-	end = block_group->key.objectid + block_group->key.offset;
+	end = block_group->start + block_group->length;
 
 	while (1) {
 		ret = btrfs_next_item(root, path);
@@ -1454,7 +1451,7 @@ static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl,
 	if (extent_count != expected_extent_count) {
 		btrfs_err(fs_info,
 			  "incorrect extent count for %llu; counted %u, expected %u",
-			  block_group->key.objectid, extent_count,
+			  block_group->start, extent_count,
 			  expected_extent_count);
 		ASSERT(0);
 		ret = -EIO;
@@ -1472,7 +1469,7 @@ static int load_free_space_extents(struct btrfs_caching_control *caching_ctl,
 				   struct btrfs_path *path,
 				   u32 expected_extent_count)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_root *root;
 	struct btrfs_key key;
@@ -1485,7 +1482,7 @@ static int load_free_space_extents(struct btrfs_caching_control *caching_ctl,
 	fs_info = block_group->fs_info;
 	root = fs_info->free_space_root;
 
-	end = block_group->key.objectid + block_group->key.offset;
+	end = block_group->start + block_group->length;
 
 	while (1) {
 		ret = btrfs_next_item(root, path);
@@ -1516,7 +1513,7 @@ static int load_free_space_extents(struct btrfs_caching_control *caching_ctl,
 	if (extent_count != expected_extent_count) {
 		btrfs_err(fs_info,
 			  "incorrect extent count for %llu; counted %u, expected %u",
-			  block_group->key.objectid, extent_count,
+			  block_group->start, extent_count,
 			  expected_extent_count);
 		ASSERT(0);
 		ret = -EIO;
@@ -1532,7 +1529,7 @@ out:
 
 int load_free_space_tree(struct btrfs_caching_control *caching_ctl)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	struct btrfs_free_space_info *info;
 	struct btrfs_path *path;
 	u32 extent_count, flags;
diff --git a/fs/btrfs/free-space-tree.h b/fs/btrfs/free-space-tree.h
index 360d50e1cdea..dc2463e4cfe3 100644
--- a/fs/btrfs/free-space-tree.h
+++ b/fs/btrfs/free-space-tree.h
@@ -16,14 +16,14 @@ struct btrfs_caching_control;
 #define BTRFS_FREE_SPACE_BITMAP_SIZE 256
 #define BTRFS_FREE_SPACE_BITMAP_BITS (BTRFS_FREE_SPACE_BITMAP_SIZE * BITS_PER_BYTE)
 
-void set_free_space_tree_thresholds(struct btrfs_block_group_cache *block_group);
+void set_free_space_tree_thresholds(struct btrfs_block_group *block_group);
 int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info);
 int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info);
 int load_free_space_tree(struct btrfs_caching_control *caching_ctl);
 int add_block_group_free_space(struct btrfs_trans_handle *trans,
-			       struct btrfs_block_group_cache *block_group);
+			       struct btrfs_block_group *block_group);
 int remove_block_group_free_space(struct btrfs_trans_handle *trans,
-				  struct btrfs_block_group_cache *block_group);
+				  struct btrfs_block_group *block_group);
 int add_to_free_space_tree(struct btrfs_trans_handle *trans,
 			   u64 start, u64 size);
 int remove_from_free_space_tree(struct btrfs_trans_handle *trans,
@@ -32,21 +32,21 @@ int remove_from_free_space_tree(struct btrfs_trans_handle *trans,
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 struct btrfs_free_space_info *
 search_free_space_info(struct btrfs_trans_handle *trans,
-		       struct btrfs_block_group_cache *block_group,
+		       struct btrfs_block_group *block_group,
 		       struct btrfs_path *path, int cow);
 int __add_to_free_space_tree(struct btrfs_trans_handle *trans,
-			     struct btrfs_block_group_cache *block_group,
+			     struct btrfs_block_group *block_group,
 			     struct btrfs_path *path, u64 start, u64 size);
 int __remove_from_free_space_tree(struct btrfs_trans_handle *trans,
-				  struct btrfs_block_group_cache *block_group,
+				  struct btrfs_block_group *block_group,
 				  struct btrfs_path *path, u64 start, u64 size);
 int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
-				  struct btrfs_block_group_cache *block_group,
+				  struct btrfs_block_group *block_group,
 				  struct btrfs_path *path);
 int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
-				  struct btrfs_block_group_cache *block_group,
+				  struct btrfs_block_group *block_group,
 				  struct btrfs_path *path);
-int free_space_test_bit(struct btrfs_block_group_cache *block_group,
+int free_space_test_bit(struct btrfs_block_group *block_group,
 			struct btrfs_path *path, u64 offset);
 #endif
 
diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c
index 37345fb6191d..6009e0e939b5 100644
--- a/fs/btrfs/inode-map.c
+++ b/fs/btrfs/inode-map.c
@@ -107,7 +107,7 @@ again:
 
 		if (last != (u64)-1 && last + 1 != key.objectid) {
 			__btrfs_add_free_space(fs_info, ctl, last + 1,
-					       key.objectid - last - 1);
+					       key.objectid - last - 1, 0);
 			wake_up(&root->ino_cache_wait);
 		}
 
@@ -118,7 +118,7 @@ next:
 
 	if (last < root->highest_objectid - 1) {
 		__btrfs_add_free_space(fs_info, ctl, last + 1,
-				       root->highest_objectid - last - 1);
+				       root->highest_objectid - last - 1, 0);
 	}
 
 	spin_lock(&root->ino_cache_lock);
@@ -175,7 +175,8 @@ static void start_caching(struct btrfs_root *root)
 	ret = btrfs_find_free_objectid(root, &objectid);
 	if (!ret && objectid <= BTRFS_LAST_FREE_OBJECTID) {
 		__btrfs_add_free_space(fs_info, ctl, objectid,
-				       BTRFS_LAST_FREE_OBJECTID - objectid + 1);
+				       BTRFS_LAST_FREE_OBJECTID - objectid + 1,
+				       0);
 		wake_up(&root->ino_cache_wait);
 	}
 
@@ -221,7 +222,7 @@ void btrfs_return_ino(struct btrfs_root *root, u64 objectid)
 		return;
 again:
 	if (root->ino_cache_state == BTRFS_CACHE_FINISHED) {
-		__btrfs_add_free_space(fs_info, pinned, objectid, 1);
+		__btrfs_add_free_space(fs_info, pinned, objectid, 1, 0);
 	} else {
 		down_write(&fs_info->commit_root_sem);
 		spin_lock(&root->ino_cache_lock);
@@ -234,7 +235,7 @@ again:
 
 		start_caching(root);
 
-		__btrfs_add_free_space(fs_info, pinned, objectid, 1);
+		__btrfs_add_free_space(fs_info, pinned, objectid, 1, 0);
 
 		up_write(&fs_info->commit_root_sem);
 	}
@@ -281,7 +282,7 @@ void btrfs_unpin_free_ino(struct btrfs_root *root)
 		spin_unlock(rbroot_lock);
 		if (count)
 			__btrfs_add_free_space(root->fs_info, ctl,
-					       info->offset, count);
+					       info->offset, count, 0);
 		kmem_cache_free(btrfs_free_space_cachep, info);
 	}
 }
@@ -514,7 +515,7 @@ out_release:
 	trace_btrfs_space_reservation(fs_info, "ino_cache", trans->transid,
 				      trans->bytes_reserved, 0);
 	btrfs_block_rsv_release(fs_info, trans->block_rsv,
-				trans->bytes_reserved);
+				trans->bytes_reserved, NULL);
 out:
 	trans->block_rsv = rsv;
 	trans->bytes_reserved = num_bytes;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 015910079e73..8b3489f229c7 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -28,6 +28,7 @@
 #include <linux/magic.h>
 #include <linux/iversion.h>
 #include <linux/swap.h>
+#include <linux/migrate.h>
 #include <linux/sched/mm.h>
 #include <asm/unaligned.h>
 #include "misc.h"
@@ -44,7 +45,6 @@
 #include "locking.h"
 #include "free-space-cache.h"
 #include "inode-map.h"
-#include "backref.h"
 #include "props.h"
 #include "qgroup.h"
 #include "delalloc-space.h"
@@ -64,7 +64,6 @@ struct btrfs_dio_data {
 
 static const struct inode_operations btrfs_dir_inode_operations;
 static const struct inode_operations btrfs_symlink_inode_operations;
-static const struct inode_operations btrfs_dir_ro_inode_operations;
 static const struct inode_operations btrfs_special_inode_operations;
 static const struct inode_operations btrfs_file_inode_operations;
 static const struct address_space_operations btrfs_aops;
@@ -244,6 +243,15 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans,
 	btrfs_release_path(path);
 
 	/*
+	 * We align size to sectorsize for inline extents just for simplicity
+	 * sake.
+	 */
+	size = ALIGN(size, root->fs_info->sectorsize);
+	ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode), start, size);
+	if (ret)
+		goto fail;
+
+	/*
 	 * we're an inline extent, so nobody can
 	 * extend the file past i_size without locking
 	 * a page we already have locked.
@@ -368,6 +376,7 @@ struct async_chunk {
 	u64 end;
 	unsigned int write_flags;
 	struct list_head extents;
+	struct cgroup_subsys_state *blkcg_css;
 	struct btrfs_work work;
 	atomic_t *pending;
 };
@@ -712,10 +721,12 @@ cleanup_and_bail_uncompressed:
 	 * to our extent and set things up for the async work queue to run
 	 * cow_file_range to do the normal delalloc dance.
 	 */
-	if (page_offset(async_chunk->locked_page) >= start &&
-	    page_offset(async_chunk->locked_page) <= end)
+	if (async_chunk->locked_page &&
+	    (page_offset(async_chunk->locked_page) >= start &&
+	     page_offset(async_chunk->locked_page)) <= end) {
 		__set_page_dirty_nobuffers(async_chunk->locked_page);
 		/* unlocked later on in the async handlers */
+	}
 
 	if (redirty)
 		extent_range_redirty_for_io(inode, start, end);
@@ -795,7 +806,7 @@ retry:
 						  async_extent->start +
 						  async_extent->ram_size - 1,
 						  WB_SYNC_ALL);
-			else if (ret)
+			else if (ret && async_chunk->locked_page)
 				unlock_page(async_chunk->locked_page);
 			kfree(async_extent);
 			cond_resched();
@@ -878,7 +889,8 @@ retry:
 				    ins.objectid,
 				    ins.offset, async_extent->pages,
 				    async_extent->nr_pages,
-				    async_chunk->write_flags)) {
+				    async_chunk->write_flags,
+				    async_chunk->blkcg_css)) {
 			struct page *p = async_extent->pages[0];
 			const u64 start = async_extent->start;
 			const u64 end = start + async_extent->ram_size - 1;
@@ -1196,6 +1208,8 @@ static noinline void async_cow_free(struct btrfs_work *work)
 	async_chunk = container_of(work, struct async_chunk, work);
 	if (async_chunk->inode)
 		btrfs_add_delayed_iput(async_chunk->inode);
+	if (async_chunk->blkcg_css)
+		css_put(async_chunk->blkcg_css);
 	/*
 	 * Since the pointer to 'pending' is at the beginning of the array of
 	 * async_chunk's, freeing it ensures the whole array has been freed.
@@ -1204,12 +1218,14 @@ static noinline void async_cow_free(struct btrfs_work *work)
 		kvfree(async_chunk->pending);
 }
 
-static int cow_file_range_async(struct inode *inode, struct page *locked_page,
+static int cow_file_range_async(struct inode *inode,
+				struct writeback_control *wbc,
+				struct page *locked_page,
 				u64 start, u64 end, int *page_started,
-				unsigned long *nr_written,
-				unsigned int write_flags)
+				unsigned long *nr_written)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct cgroup_subsys_state *blkcg_css = wbc_blkcg_css(wbc);
 	struct async_cow *ctx;
 	struct async_chunk *async_chunk;
 	unsigned long nr_pages;
@@ -1218,6 +1234,7 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
 	int i;
 	bool should_compress;
 	unsigned nofs_flag;
+	const unsigned int write_flags = wbc_to_write_flags(wbc);
 
 	unlock_extent(&BTRFS_I(inode)->io_tree, start, end);
 
@@ -1264,14 +1281,45 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
 		async_chunk[i].inode = inode;
 		async_chunk[i].start = start;
 		async_chunk[i].end = cur_end;
-		async_chunk[i].locked_page = locked_page;
 		async_chunk[i].write_flags = write_flags;
 		INIT_LIST_HEAD(&async_chunk[i].extents);
 
-		btrfs_init_work(&async_chunk[i].work,
-				btrfs_delalloc_helper,
-				async_cow_start, async_cow_submit,
-				async_cow_free);
+		/*
+		 * The locked_page comes all the way from writepage and its
+		 * the original page we were actually given.  As we spread
+		 * this large delalloc region across multiple async_chunk
+		 * structs, only the first struct needs a pointer to locked_page
+		 *
+		 * This way we don't need racey decisions about who is supposed
+		 * to unlock it.
+		 */
+		if (locked_page) {
+			/*
+			 * Depending on the compressibility, the pages might or
+			 * might not go through async.  We want all of them to
+			 * be accounted against wbc once.  Let's do it here
+			 * before the paths diverge.  wbc accounting is used
+			 * only for foreign writeback detection and doesn't
+			 * need full accuracy.  Just account the whole thing
+			 * against the first page.
+			 */
+			wbc_account_cgroup_owner(wbc, locked_page,
+						 cur_end - start);
+			async_chunk[i].locked_page = locked_page;
+			locked_page = NULL;
+		} else {
+			async_chunk[i].locked_page = NULL;
+		}
+
+		if (blkcg_css != blkcg_root_css) {
+			css_get(blkcg_css);
+			async_chunk[i].blkcg_css = blkcg_css;
+		} else {
+			async_chunk[i].blkcg_css = NULL;
+		}
+
+		btrfs_init_work(&async_chunk[i].work, async_cow_start,
+				async_cow_submit, async_cow_free);
 
 		nr_pages = DIV_ROUND_UP(cur_end - start, PAGE_SIZE);
 		atomic_add(nr_pages, &fs_info->async_delalloc_pages);
@@ -1439,10 +1487,10 @@ next_slot:
 			disk_num_bytes =
 				btrfs_file_extent_disk_num_bytes(leaf, fi);
 			/*
-			 * If extent we got ends before our range starts, skip
-			 * to next extent
+			 * If the extent we got ends before our current offset,
+			 * skip to the next extent.
 			 */
-			if (extent_end <= start) {
+			if (extent_end <= cur_offset) {
 				path->slots[0]++;
 				goto next_slot;
 			}
@@ -1697,7 +1745,6 @@ int btrfs_run_delalloc_range(struct inode *inode, struct page *locked_page,
 {
 	int ret;
 	int force_cow = need_force_cow(inode, start, end);
-	unsigned int write_flags = wbc_to_write_flags(wbc);
 
 	if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW && !force_cow) {
 		ret = run_delalloc_nocow(inode, locked_page, start, end,
@@ -1712,9 +1759,8 @@ int btrfs_run_delalloc_range(struct inode *inode, struct page *locked_page,
 	} else {
 		set_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
 			&BTRFS_I(inode)->runtime_flags);
-		ret = cow_file_range_async(inode, locked_page, start, end,
-					   page_started, nr_written,
-					   write_flags);
+		ret = cow_file_range_async(inode, wbc, locked_page, start, end,
+					   page_started, nr_written);
 	}
 	if (ret)
 		btrfs_cleanup_ordered_extents(inode, locked_page, start,
@@ -2090,7 +2136,7 @@ static blk_status_t btrfs_submit_bio_hook(struct inode *inode, struct bio *bio,
 							   bio_flags);
 			goto out;
 		} else if (!skip_sum) {
-			ret = btrfs_lookup_bio_sums(inode, bio, NULL);
+			ret = btrfs_lookup_bio_sums(inode, bio, (u64)-1, NULL);
 			if (ret)
 				goto out;
 		}
@@ -2110,7 +2156,7 @@ static blk_status_t btrfs_submit_bio_hook(struct inode *inode, struct bio *bio,
 	}
 
 mapit:
-	ret = btrfs_map_bio(fs_info, bio, mirror_num, 0);
+	ret = btrfs_map_bio(fs_info, bio, mirror_num);
 
 out:
 	if (ret) {
@@ -2153,6 +2199,7 @@ int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
 /* see btrfs_writepage_start_hook for details on why this is required */
 struct btrfs_writepage_fixup {
 	struct page *page;
+	struct inode *inode;
 	struct btrfs_work work;
 };
 
@@ -2166,27 +2213,71 @@ static void btrfs_writepage_fixup_worker(struct btrfs_work *work)
 	struct inode *inode;
 	u64 page_start;
 	u64 page_end;
-	int ret;
+	int ret = 0;
+	bool free_delalloc_space = true;
 
 	fixup = container_of(work, struct btrfs_writepage_fixup, work);
 	page = fixup->page;
+	inode = fixup->inode;
+	page_start = page_offset(page);
+	page_end = page_offset(page) + PAGE_SIZE - 1;
+
+	/*
+	 * This is similar to page_mkwrite, we need to reserve the space before
+	 * we take the page lock.
+	 */
+	ret = btrfs_delalloc_reserve_space(inode, &data_reserved, page_start,
+					   PAGE_SIZE);
 again:
 	lock_page(page);
+
+	/*
+	 * Before we queued this fixup, we took a reference on the page.
+	 * page->mapping may go NULL, but it shouldn't be moved to a different
+	 * address space.
+	 */
 	if (!page->mapping || !PageDirty(page) || !PageChecked(page)) {
-		ClearPageChecked(page);
+		/*
+		 * Unfortunately this is a little tricky, either
+		 *
+		 * 1) We got here and our page had already been dealt with and
+		 *    we reserved our space, thus ret == 0, so we need to just
+		 *    drop our space reservation and bail.  This can happen the
+		 *    first time we come into the fixup worker, or could happen
+		 *    while waiting for the ordered extent.
+		 * 2) Our page was already dealt with, but we happened to get an
+		 *    ENOSPC above from the btrfs_delalloc_reserve_space.  In
+		 *    this case we obviously don't have anything to release, but
+		 *    because the page was already dealt with we don't want to
+		 *    mark the page with an error, so make sure we're resetting
+		 *    ret to 0.  This is why we have this check _before_ the ret
+		 *    check, because we do not want to have a surprise ENOSPC
+		 *    when the page was already properly dealt with.
+		 */
+		if (!ret) {
+			btrfs_delalloc_release_extents(BTRFS_I(inode),
+						       PAGE_SIZE);
+			btrfs_delalloc_release_space(inode, data_reserved,
+						     page_start, PAGE_SIZE,
+						     true);
+		}
+		ret = 0;
 		goto out_page;
 	}
 
-	inode = page->mapping->host;
-	page_start = page_offset(page);
-	page_end = page_offset(page) + PAGE_SIZE - 1;
+	/*
+	 * We can't mess with the page state unless it is locked, so now that
+	 * it is locked bail if we failed to make our space reservation.
+	 */
+	if (ret)
+		goto out_page;
 
 	lock_extent_bits(&BTRFS_I(inode)->io_tree, page_start, page_end,
 			 &cached_state);
 
 	/* already ordered? We're done */
 	if (PagePrivate2(page))
-		goto out;
+		goto out_reserved;
 
 	ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), page_start,
 					PAGE_SIZE);
@@ -2199,35 +2290,49 @@ again:
 		goto again;
 	}
 
-	ret = btrfs_delalloc_reserve_space(inode, &data_reserved, page_start,
-					   PAGE_SIZE);
-	if (ret) {
-		mapping_set_error(page->mapping, ret);
-		end_extent_writepage(page, ret, page_start, page_end);
-		ClearPageChecked(page);
-		goto out;
-	 }
-
 	ret = btrfs_set_extent_delalloc(inode, page_start, page_end, 0,
 					&cached_state);
-	if (ret) {
-		mapping_set_error(page->mapping, ret);
-		end_extent_writepage(page, ret, page_start, page_end);
-		ClearPageChecked(page);
-		goto out;
-	}
+	if (ret)
+		goto out_reserved;
 
-	ClearPageChecked(page);
-	set_page_dirty(page);
+	/*
+	 * Everything went as planned, we're now the owner of a dirty page with
+	 * delayed allocation bits set and space reserved for our COW
+	 * destination.
+	 *
+	 * The page was dirty when we started, nothing should have cleaned it.
+	 */
+	BUG_ON(!PageDirty(page));
+	free_delalloc_space = false;
+out_reserved:
 	btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE);
-out:
+	if (free_delalloc_space)
+		btrfs_delalloc_release_space(inode, data_reserved, page_start,
+					     PAGE_SIZE, true);
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, page_end,
 			     &cached_state);
 out_page:
+	if (ret) {
+		/*
+		 * We hit ENOSPC or other errors.  Update the mapping and page
+		 * to reflect the errors and clean the page.
+		 */
+		mapping_set_error(page->mapping, ret);
+		end_extent_writepage(page, ret, page_start, page_end);
+		clear_page_dirty_for_io(page);
+		SetPageError(page);
+	}
+	ClearPageChecked(page);
 	unlock_page(page);
 	put_page(page);
 	kfree(fixup);
 	extent_changeset_free(data_reserved);
+	/*
+	 * As a precaution, do a delayed iput in case it would be the last iput
+	 * that could need flushing space. Recursing back to fixup worker would
+	 * deadlock.
+	 */
+	btrfs_add_delayed_iput(inode);
 }
 
 /*
@@ -2251,6 +2356,13 @@ int btrfs_writepage_cow_fixup(struct page *page, u64 start, u64 end)
 	if (TestClearPagePrivate2(page))
 		return 0;
 
+	/*
+	 * PageChecked is set below when we create a fixup worker for this page,
+	 * don't try to create another one if we're already PageChecked()
+	 *
+	 * The extent_io writepage code will redirty the page if we send back
+	 * EAGAIN.
+	 */
 	if (PageChecked(page))
 		return -EAGAIN;
 
@@ -2258,13 +2370,21 @@ int btrfs_writepage_cow_fixup(struct page *page, u64 start, u64 end)
 	if (!fixup)
 		return -EAGAIN;
 
+	/*
+	 * We are already holding a reference to this inode from
+	 * write_cache_pages.  We need to hold it because the space reservation
+	 * takes place outside of the page lock, and we can't trust
+	 * page->mapping outside of the page lock.
+	 */
+	ihold(inode);
 	SetPageChecked(page);
 	get_page(page);
-	btrfs_init_work(&fixup->work, btrfs_fixup_helper,
-			btrfs_writepage_fixup_worker, NULL, NULL);
+	btrfs_init_work(&fixup->work, btrfs_writepage_fixup_worker, NULL, NULL);
 	fixup->page = page;
+	fixup->inode = inode;
 	btrfs_queue_work(fs_info->fixup_workers, &fixup->work);
-	return -EBUSY;
+
+	return -EAGAIN;
 }
 
 static int insert_reserved_file_extent(struct btrfs_trans_handle *trans,
@@ -2336,6 +2456,11 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans,
 	ins.offset = disk_num_bytes;
 	ins.type = BTRFS_EXTENT_ITEM_KEY;
 
+	ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode), file_pos,
+						ram_bytes);
+	if (ret)
+		goto out;
+
 	/*
 	 * Release the reserved range from inode dirty range map, as it is
 	 * already moved into delayed_ref_head
@@ -2353,653 +2478,10 @@ out:
 	return ret;
 }
 
-/* snapshot-aware defrag */
-struct sa_defrag_extent_backref {
-	struct rb_node node;
-	struct old_sa_defrag_extent *old;
-	u64 root_id;
-	u64 inum;
-	u64 file_pos;
-	u64 extent_offset;
-	u64 num_bytes;
-	u64 generation;
-};
-
-struct old_sa_defrag_extent {
-	struct list_head list;
-	struct new_sa_defrag_extent *new;
-
-	u64 extent_offset;
-	u64 bytenr;
-	u64 offset;
-	u64 len;
-	int count;
-};
-
-struct new_sa_defrag_extent {
-	struct rb_root root;
-	struct list_head head;
-	struct btrfs_path *path;
-	struct inode *inode;
-	u64 file_pos;
-	u64 len;
-	u64 bytenr;
-	u64 disk_len;
-	u8 compress_type;
-};
-
-static int backref_comp(struct sa_defrag_extent_backref *b1,
-			struct sa_defrag_extent_backref *b2)
-{
-	if (b1->root_id < b2->root_id)
-		return -1;
-	else if (b1->root_id > b2->root_id)
-		return 1;
-
-	if (b1->inum < b2->inum)
-		return -1;
-	else if (b1->inum > b2->inum)
-		return 1;
-
-	if (b1->file_pos < b2->file_pos)
-		return -1;
-	else if (b1->file_pos > b2->file_pos)
-		return 1;
-
-	/*
-	 * [------------------------------] ===> (a range of space)
-	 *     |<--->|   |<---->| =============> (fs/file tree A)
-	 * |<---------------------------->| ===> (fs/file tree B)
-	 *
-	 * A range of space can refer to two file extents in one tree while
-	 * refer to only one file extent in another tree.
-	 *
-	 * So we may process a disk offset more than one time(two extents in A)
-	 * and locate at the same extent(one extent in B), then insert two same
-	 * backrefs(both refer to the extent in B).
-	 */
-	return 0;
-}
-
-static void backref_insert(struct rb_root *root,
-			   struct sa_defrag_extent_backref *backref)
-{
-	struct rb_node **p = &root->rb_node;
-	struct rb_node *parent = NULL;
-	struct sa_defrag_extent_backref *entry;
-	int ret;
-
-	while (*p) {
-		parent = *p;
-		entry = rb_entry(parent, struct sa_defrag_extent_backref, node);
-
-		ret = backref_comp(backref, entry);
-		if (ret < 0)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-
-	rb_link_node(&backref->node, parent, p);
-	rb_insert_color(&backref->node, root);
-}
-
-/*
- * Note the backref might has changed, and in this case we just return 0.
- */
-static noinline int record_one_backref(u64 inum, u64 offset, u64 root_id,
-				       void *ctx)
-{
-	struct btrfs_file_extent_item *extent;
-	struct old_sa_defrag_extent *old = ctx;
-	struct new_sa_defrag_extent *new = old->new;
-	struct btrfs_path *path = new->path;
-	struct btrfs_key key;
-	struct btrfs_root *root;
-	struct sa_defrag_extent_backref *backref;
-	struct extent_buffer *leaf;
-	struct inode *inode = new->inode;
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	int slot;
-	int ret;
-	u64 extent_offset;
-	u64 num_bytes;
-
-	if (BTRFS_I(inode)->root->root_key.objectid == root_id &&
-	    inum == btrfs_ino(BTRFS_I(inode)))
-		return 0;
-
-	key.objectid = root_id;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(root)) {
-		if (PTR_ERR(root) == -ENOENT)
-			return 0;
-		WARN_ON(1);
-		btrfs_debug(fs_info, "inum=%llu, offset=%llu, root_id=%llu",
-			 inum, offset, root_id);
-		return PTR_ERR(root);
-	}
-
-	key.objectid = inum;
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	if (offset > (u64)-1 << 32)
-		key.offset = 0;
-	else
-		key.offset = offset;
-
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (WARN_ON(ret < 0))
-		return ret;
-	ret = 0;
-
-	while (1) {
-		cond_resched();
-
-		leaf = path->nodes[0];
-		slot = path->slots[0];
-
-		if (slot >= btrfs_header_nritems(leaf)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0) {
-				goto out;
-			} else if (ret > 0) {
-				ret = 0;
-				goto out;
-			}
-			continue;
-		}
-
-		path->slots[0]++;
-
-		btrfs_item_key_to_cpu(leaf, &key, slot);
-
-		if (key.objectid > inum)
-			goto out;
-
-		if (key.objectid < inum || key.type != BTRFS_EXTENT_DATA_KEY)
-			continue;
-
-		extent = btrfs_item_ptr(leaf, slot,
-					struct btrfs_file_extent_item);
-
-		if (btrfs_file_extent_disk_bytenr(leaf, extent) != old->bytenr)
-			continue;
-
-		/*
-		 * 'offset' refers to the exact key.offset,
-		 * NOT the 'offset' field in btrfs_extent_data_ref, ie.
-		 * (key.offset - extent_offset).
-		 */
-		if (key.offset != offset)
-			continue;
-
-		extent_offset = btrfs_file_extent_offset(leaf, extent);
-		num_bytes = btrfs_file_extent_num_bytes(leaf, extent);
-
-		if (extent_offset >= old->extent_offset + old->offset +
-		    old->len || extent_offset + num_bytes <=
-		    old->extent_offset + old->offset)
-			continue;
-		break;
-	}
-
-	backref = kmalloc(sizeof(*backref), GFP_NOFS);
-	if (!backref) {
-		ret = -ENOENT;
-		goto out;
-	}
-
-	backref->root_id = root_id;
-	backref->inum = inum;
-	backref->file_pos = offset;
-	backref->num_bytes = num_bytes;
-	backref->extent_offset = extent_offset;
-	backref->generation = btrfs_file_extent_generation(leaf, extent);
-	backref->old = old;
-	backref_insert(&new->root, backref);
-	old->count++;
-out:
-	btrfs_release_path(path);
-	WARN_ON(ret);
-	return ret;
-}
-
-static noinline bool record_extent_backrefs(struct btrfs_path *path,
-				   struct new_sa_defrag_extent *new)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb);
-	struct old_sa_defrag_extent *old, *tmp;
-	int ret;
-
-	new->path = path;
-
-	list_for_each_entry_safe(old, tmp, &new->head, list) {
-		ret = iterate_inodes_from_logical(old->bytenr +
-						  old->extent_offset, fs_info,
-						  path, record_one_backref,
-						  old, false);
-		if (ret < 0 && ret != -ENOENT)
-			return false;
-
-		/* no backref to be processed for this extent */
-		if (!old->count) {
-			list_del(&old->list);
-			kfree(old);
-		}
-	}
-
-	if (list_empty(&new->head))
-		return false;
-
-	return true;
-}
-
-static int relink_is_mergable(struct extent_buffer *leaf,
-			      struct btrfs_file_extent_item *fi,
-			      struct new_sa_defrag_extent *new)
-{
-	if (btrfs_file_extent_disk_bytenr(leaf, fi) != new->bytenr)
-		return 0;
-
-	if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG)
-		return 0;
-
-	if (btrfs_file_extent_compression(leaf, fi) != new->compress_type)
-		return 0;
-
-	if (btrfs_file_extent_encryption(leaf, fi) ||
-	    btrfs_file_extent_other_encoding(leaf, fi))
-		return 0;
-
-	return 1;
-}
-
-/*
- * Note the backref might has changed, and in this case we just return 0.
- */
-static noinline int relink_extent_backref(struct btrfs_path *path,
-				 struct sa_defrag_extent_backref *prev,
-				 struct sa_defrag_extent_backref *backref)
-{
-	struct btrfs_file_extent_item *extent;
-	struct btrfs_file_extent_item *item;
-	struct btrfs_ordered_extent *ordered;
-	struct btrfs_trans_handle *trans;
-	struct btrfs_ref ref = { 0 };
-	struct btrfs_root *root;
-	struct btrfs_key key;
-	struct extent_buffer *leaf;
-	struct old_sa_defrag_extent *old = backref->old;
-	struct new_sa_defrag_extent *new = old->new;
-	struct btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb);
-	struct inode *inode;
-	struct extent_state *cached = NULL;
-	int ret = 0;
-	u64 start;
-	u64 len;
-	u64 lock_start;
-	u64 lock_end;
-	bool merge = false;
-	int index;
-
-	if (prev && prev->root_id == backref->root_id &&
-	    prev->inum == backref->inum &&
-	    prev->file_pos + prev->num_bytes == backref->file_pos)
-		merge = true;
-
-	/* step 1: get root */
-	key.objectid = backref->root_id;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-
-	index = srcu_read_lock(&fs_info->subvol_srcu);
-
-	root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(root)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
-		if (PTR_ERR(root) == -ENOENT)
-			return 0;
-		return PTR_ERR(root);
-	}
-
-	if (btrfs_root_readonly(root)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
-		return 0;
-	}
-
-	/* step 2: get inode */
-	key.objectid = backref->inum;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-
-	inode = btrfs_iget(fs_info->sb, &key, root, NULL);
-	if (IS_ERR(inode)) {
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
-		return 0;
-	}
-
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
-
-	/* step 3: relink backref */
-	lock_start = backref->file_pos;
-	lock_end = backref->file_pos + backref->num_bytes - 1;
-	lock_extent_bits(&BTRFS_I(inode)->io_tree, lock_start, lock_end,
-			 &cached);
-
-	ordered = btrfs_lookup_first_ordered_extent(inode, lock_end);
-	if (ordered) {
-		btrfs_put_ordered_extent(ordered);
-		goto out_unlock;
-	}
-
-	trans = btrfs_join_transaction(root);
-	if (IS_ERR(trans)) {
-		ret = PTR_ERR(trans);
-		goto out_unlock;
-	}
-
-	key.objectid = backref->inum;
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = backref->file_pos;
-
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0) {
-		goto out_free_path;
-	} else if (ret > 0) {
-		ret = 0;
-		goto out_free_path;
-	}
-
-	extent = btrfs_item_ptr(path->nodes[0], path->slots[0],
-				struct btrfs_file_extent_item);
-
-	if (btrfs_file_extent_generation(path->nodes[0], extent) !=
-	    backref->generation)
-		goto out_free_path;
-
-	btrfs_release_path(path);
-
-	start = backref->file_pos;
-	if (backref->extent_offset < old->extent_offset + old->offset)
-		start += old->extent_offset + old->offset -
-			 backref->extent_offset;
-
-	len = min(backref->extent_offset + backref->num_bytes,
-		  old->extent_offset + old->offset + old->len);
-	len -= max(backref->extent_offset, old->extent_offset + old->offset);
-
-	ret = btrfs_drop_extents(trans, root, inode, start,
-				 start + len, 1);
-	if (ret)
-		goto out_free_path;
-again:
-	key.objectid = btrfs_ino(BTRFS_I(inode));
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = start;
-
-	path->leave_spinning = 1;
-	if (merge) {
-		struct btrfs_file_extent_item *fi;
-		u64 extent_len;
-		struct btrfs_key found_key;
-
-		ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
-		if (ret < 0)
-			goto out_free_path;
-
-		path->slots[0]--;
-		leaf = path->nodes[0];
-		btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
-
-		fi = btrfs_item_ptr(leaf, path->slots[0],
-				    struct btrfs_file_extent_item);
-		extent_len = btrfs_file_extent_num_bytes(leaf, fi);
-
-		if (extent_len + found_key.offset == start &&
-		    relink_is_mergable(leaf, fi, new)) {
-			btrfs_set_file_extent_num_bytes(leaf, fi,
-							extent_len + len);
-			btrfs_mark_buffer_dirty(leaf);
-			inode_add_bytes(inode, len);
-
-			ret = 1;
-			goto out_free_path;
-		} else {
-			merge = false;
-			btrfs_release_path(path);
-			goto again;
-		}
-	}
-
-	ret = btrfs_insert_empty_item(trans, root, path, &key,
-					sizeof(*extent));
-	if (ret) {
-		btrfs_abort_transaction(trans, ret);
-		goto out_free_path;
-	}
-
-	leaf = path->nodes[0];
-	item = btrfs_item_ptr(leaf, path->slots[0],
-				struct btrfs_file_extent_item);
-	btrfs_set_file_extent_disk_bytenr(leaf, item, new->bytenr);
-	btrfs_set_file_extent_disk_num_bytes(leaf, item, new->disk_len);
-	btrfs_set_file_extent_offset(leaf, item, start - new->file_pos);
-	btrfs_set_file_extent_num_bytes(leaf, item, len);
-	btrfs_set_file_extent_ram_bytes(leaf, item, new->len);
-	btrfs_set_file_extent_generation(leaf, item, trans->transid);
-	btrfs_set_file_extent_type(leaf, item, BTRFS_FILE_EXTENT_REG);
-	btrfs_set_file_extent_compression(leaf, item, new->compress_type);
-	btrfs_set_file_extent_encryption(leaf, item, 0);
-	btrfs_set_file_extent_other_encoding(leaf, item, 0);
-
-	btrfs_mark_buffer_dirty(leaf);
-	inode_add_bytes(inode, len);
-	btrfs_release_path(path);
-
-	btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, new->bytenr,
-			       new->disk_len, 0);
-	btrfs_init_data_ref(&ref, backref->root_id, backref->inum,
-			    new->file_pos);  /* start - extent_offset */
-	ret = btrfs_inc_extent_ref(trans, &ref);
-	if (ret) {
-		btrfs_abort_transaction(trans, ret);
-		goto out_free_path;
-	}
-
-	ret = 1;
-out_free_path:
-	btrfs_release_path(path);
-	path->leave_spinning = 0;
-	btrfs_end_transaction(trans);
-out_unlock:
-	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start, lock_end,
-			     &cached);
-	iput(inode);
-	return ret;
-}
-
-static void free_sa_defrag_extent(struct new_sa_defrag_extent *new)
-{
-	struct old_sa_defrag_extent *old, *tmp;
-
-	if (!new)
-		return;
-
-	list_for_each_entry_safe(old, tmp, &new->head, list) {
-		kfree(old);
-	}
-	kfree(new);
-}
-
-static void relink_file_extents(struct new_sa_defrag_extent *new)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(new->inode->i_sb);
-	struct btrfs_path *path;
-	struct sa_defrag_extent_backref *backref;
-	struct sa_defrag_extent_backref *prev = NULL;
-	struct rb_node *node;
-	int ret;
-
-	path = btrfs_alloc_path();
-	if (!path)
-		return;
-
-	if (!record_extent_backrefs(path, new)) {
-		btrfs_free_path(path);
-		goto out;
-	}
-	btrfs_release_path(path);
-
-	while (1) {
-		node = rb_first(&new->root);
-		if (!node)
-			break;
-		rb_erase(node, &new->root);
-
-		backref = rb_entry(node, struct sa_defrag_extent_backref, node);
-
-		ret = relink_extent_backref(path, prev, backref);
-		WARN_ON(ret < 0);
-
-		kfree(prev);
-
-		if (ret == 1)
-			prev = backref;
-		else
-			prev = NULL;
-		cond_resched();
-	}
-	kfree(prev);
-
-	btrfs_free_path(path);
-out:
-	free_sa_defrag_extent(new);
-
-	atomic_dec(&fs_info->defrag_running);
-	wake_up(&fs_info->transaction_wait);
-}
-
-static struct new_sa_defrag_extent *
-record_old_file_extents(struct inode *inode,
-			struct btrfs_ordered_extent *ordered)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_path *path;
-	struct btrfs_key key;
-	struct old_sa_defrag_extent *old;
-	struct new_sa_defrag_extent *new;
-	int ret;
-
-	new = kmalloc(sizeof(*new), GFP_NOFS);
-	if (!new)
-		return NULL;
-
-	new->inode = inode;
-	new->file_pos = ordered->file_offset;
-	new->len = ordered->len;
-	new->bytenr = ordered->start;
-	new->disk_len = ordered->disk_len;
-	new->compress_type = ordered->compress_type;
-	new->root = RB_ROOT;
-	INIT_LIST_HEAD(&new->head);
-
-	path = btrfs_alloc_path();
-	if (!path)
-		goto out_kfree;
-
-	key.objectid = btrfs_ino(BTRFS_I(inode));
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = new->file_pos;
-
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out_free_path;
-	if (ret > 0 && path->slots[0] > 0)
-		path->slots[0]--;
-
-	/* find out all the old extents for the file range */
-	while (1) {
-		struct btrfs_file_extent_item *extent;
-		struct extent_buffer *l;
-		int slot;
-		u64 num_bytes;
-		u64 offset;
-		u64 end;
-		u64 disk_bytenr;
-		u64 extent_offset;
-
-		l = path->nodes[0];
-		slot = path->slots[0];
-
-		if (slot >= btrfs_header_nritems(l)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				goto out_free_path;
-			else if (ret > 0)
-				break;
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(l, &key, slot);
-
-		if (key.objectid != btrfs_ino(BTRFS_I(inode)))
-			break;
-		if (key.type != BTRFS_EXTENT_DATA_KEY)
-			break;
-		if (key.offset >= new->file_pos + new->len)
-			break;
-
-		extent = btrfs_item_ptr(l, slot, struct btrfs_file_extent_item);
-
-		num_bytes = btrfs_file_extent_num_bytes(l, extent);
-		if (key.offset + num_bytes < new->file_pos)
-			goto next;
-
-		disk_bytenr = btrfs_file_extent_disk_bytenr(l, extent);
-		if (!disk_bytenr)
-			goto next;
-
-		extent_offset = btrfs_file_extent_offset(l, extent);
-
-		old = kmalloc(sizeof(*old), GFP_NOFS);
-		if (!old)
-			goto out_free_path;
-
-		offset = max(new->file_pos, key.offset);
-		end = min(new->file_pos + new->len, key.offset + num_bytes);
-
-		old->bytenr = disk_bytenr;
-		old->extent_offset = extent_offset;
-		old->offset = offset - key.offset;
-		old->len = end - offset;
-		old->new = new;
-		old->count = 0;
-		list_add_tail(&old->list, &new->head);
-next:
-		path->slots[0]++;
-		cond_resched();
-	}
-
-	btrfs_free_path(path);
-	atomic_inc(&fs_info->defrag_running);
-
-	return new;
-
-out_free_path:
-	btrfs_free_path(path);
-out_kfree:
-	free_sa_defrag_extent(new);
-	return NULL;
-}
-
 static void btrfs_release_delalloc_bytes(struct btrfs_fs_info *fs_info,
 					 u64 start, u64 len)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 
 	cache = btrfs_lookup_block_group(fs_info, start);
 	ASSERT(cache);
@@ -3023,32 +2505,33 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 	struct btrfs_trans_handle *trans = NULL;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct extent_state *cached_state = NULL;
-	struct new_sa_defrag_extent *new = NULL;
+	u64 start, end;
 	int compress_type = 0;
 	int ret = 0;
-	u64 logical_len = ordered_extent->len;
-	bool nolock;
+	u64 logical_len = ordered_extent->num_bytes;
+	bool freespace_inode;
 	bool truncated = false;
 	bool range_locked = false;
 	bool clear_new_delalloc_bytes = false;
 	bool clear_reserved_extent = true;
+	unsigned int clear_bits;
+
+	start = ordered_extent->file_offset;
+	end = start + ordered_extent->num_bytes - 1;
 
 	if (!test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) &&
 	    !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags) &&
 	    !test_bit(BTRFS_ORDERED_DIRECT, &ordered_extent->flags))
 		clear_new_delalloc_bytes = true;
 
-	nolock = btrfs_is_free_space_inode(BTRFS_I(inode));
+	freespace_inode = btrfs_is_free_space_inode(BTRFS_I(inode));
 
 	if (test_bit(BTRFS_ORDERED_IOERR, &ordered_extent->flags)) {
 		ret = -EIO;
 		goto out;
 	}
 
-	btrfs_free_io_failure_record(BTRFS_I(inode),
-			ordered_extent->file_offset,
-			ordered_extent->file_offset +
-			ordered_extent->len - 1);
+	btrfs_free_io_failure_record(BTRFS_I(inode), start, end);
 
 	if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered_extent->flags)) {
 		truncated = true;
@@ -3066,11 +2549,11 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		 * space for NOCOW range.
 		 * As NOCOW won't cause a new delayed ref, just free the space
 		 */
-		btrfs_qgroup_free_data(inode, NULL, ordered_extent->file_offset,
-				       ordered_extent->len);
-		btrfs_ordered_update_i_size(inode, 0, ordered_extent);
-		if (nolock)
-			trans = btrfs_join_transaction_nolock(root);
+		btrfs_qgroup_free_data(inode, NULL, start,
+				       ordered_extent->num_bytes);
+		btrfs_inode_safe_disk_i_size_write(inode, 0);
+		if (freespace_inode)
+			trans = btrfs_join_transaction_spacecache(root);
 		else
 			trans = btrfs_join_transaction(root);
 		if (IS_ERR(trans)) {
@@ -3086,26 +2569,10 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 	}
 
 	range_locked = true;
-	lock_extent_bits(io_tree, ordered_extent->file_offset,
-			 ordered_extent->file_offset + ordered_extent->len - 1,
-			 &cached_state);
-
-	ret = test_range_bit(io_tree, ordered_extent->file_offset,
-			ordered_extent->file_offset + ordered_extent->len - 1,
-			EXTENT_DEFRAG, 0, cached_state);
-	if (ret) {
-		u64 last_snapshot = btrfs_root_last_snapshot(&root->root_item);
-		if (0 && last_snapshot >= BTRFS_I(inode)->generation)
-			/* the inode is shared */
-			new = record_old_file_extents(inode, ordered_extent);
+	lock_extent_bits(io_tree, start, end, &cached_state);
 
-		clear_extent_bit(io_tree, ordered_extent->file_offset,
-			ordered_extent->file_offset + ordered_extent->len - 1,
-			EXTENT_DEFRAG, 0, 0, &cached_state);
-	}
-
-	if (nolock)
-		trans = btrfs_join_transaction_nolock(root);
+	if (freespace_inode)
+		trans = btrfs_join_transaction_spacecache(root);
 	else
 		trans = btrfs_join_transaction(root);
 	if (IS_ERR(trans)) {
@@ -3120,31 +2587,30 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		compress_type = ordered_extent->compress_type;
 	if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
 		BUG_ON(compress_type);
-		btrfs_qgroup_free_data(inode, NULL, ordered_extent->file_offset,
-				       ordered_extent->len);
+		btrfs_qgroup_free_data(inode, NULL, start,
+				       ordered_extent->num_bytes);
 		ret = btrfs_mark_extent_written(trans, BTRFS_I(inode),
 						ordered_extent->file_offset,
 						ordered_extent->file_offset +
 						logical_len);
 	} else {
 		BUG_ON(root == fs_info->tree_root);
-		ret = insert_reserved_file_extent(trans, inode,
-						ordered_extent->file_offset,
-						ordered_extent->start,
-						ordered_extent->disk_len,
+		ret = insert_reserved_file_extent(trans, inode, start,
+						ordered_extent->disk_bytenr,
+						ordered_extent->disk_num_bytes,
 						logical_len, logical_len,
 						compress_type, 0, 0,
 						BTRFS_FILE_EXTENT_REG);
 		if (!ret) {
 			clear_reserved_extent = false;
 			btrfs_release_delalloc_bytes(fs_info,
-						     ordered_extent->start,
-						     ordered_extent->disk_len);
+						ordered_extent->disk_bytenr,
+						ordered_extent->disk_num_bytes);
 		}
 	}
 	unpin_extent_cache(&BTRFS_I(inode)->extent_tree,
-			   ordered_extent->file_offset, ordered_extent->len,
-			   trans->transid);
+			   ordered_extent->file_offset,
+			   ordered_extent->num_bytes, trans->transid);
 	if (ret < 0) {
 		btrfs_abort_transaction(trans, ret);
 		goto out;
@@ -3156,7 +2622,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 		goto out;
 	}
 
-	btrfs_ordered_update_i_size(inode, 0, ordered_extent);
+	btrfs_inode_safe_disk_i_size_write(inode, 0);
 	ret = btrfs_update_inode_fallback(trans, root, inode);
 	if (ret) { /* -ENOMEM or corruption */
 		btrfs_abort_transaction(trans, ret);
@@ -3164,37 +2630,27 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent)
 	}
 	ret = 0;
 out:
-	if (range_locked || clear_new_delalloc_bytes) {
-		unsigned int clear_bits = 0;
-
-		if (range_locked)
-			clear_bits |= EXTENT_LOCKED;
-		if (clear_new_delalloc_bytes)
-			clear_bits |= EXTENT_DELALLOC_NEW;
-		clear_extent_bit(&BTRFS_I(inode)->io_tree,
-				 ordered_extent->file_offset,
-				 ordered_extent->file_offset +
-				 ordered_extent->len - 1,
-				 clear_bits,
-				 (clear_bits & EXTENT_LOCKED) ? 1 : 0,
-				 0, &cached_state);
-	}
+	clear_bits = EXTENT_DEFRAG;
+	if (range_locked)
+		clear_bits |= EXTENT_LOCKED;
+	if (clear_new_delalloc_bytes)
+		clear_bits |= EXTENT_DELALLOC_NEW;
+	clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, clear_bits,
+			 (clear_bits & EXTENT_LOCKED) ? 1 : 0, 0,
+			 &cached_state);
 
 	if (trans)
 		btrfs_end_transaction(trans);
 
 	if (ret || truncated) {
-		u64 start, end;
+		u64 unwritten_start = start;
 
 		if (truncated)
-			start = ordered_extent->file_offset + logical_len;
-		else
-			start = ordered_extent->file_offset;
-		end = ordered_extent->file_offset + ordered_extent->len - 1;
-		clear_extent_uptodate(io_tree, start, end, NULL);
+			unwritten_start += logical_len;
+		clear_extent_uptodate(io_tree, unwritten_start, end, NULL);
 
 		/* Drop the cache for the part of the extent we didn't write. */
-		btrfs_drop_extent_cache(BTRFS_I(inode), start, end, 0);
+		btrfs_drop_extent_cache(BTRFS_I(inode), unwritten_start, end, 0);
 
 		/*
 		 * If the ordered extent had an IOERR or something else went
@@ -3209,29 +2665,28 @@ out:
 		if ((ret || !logical_len) &&
 		    clear_reserved_extent &&
 		    !test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags) &&
-		    !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags))
+		    !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
+			/*
+			 * Discard the range before returning it back to the
+			 * free space pool
+			 */
+			if (ret && btrfs_test_opt(fs_info, DISCARD_SYNC))
+				btrfs_discard_extent(fs_info,
+						ordered_extent->disk_bytenr,
+						ordered_extent->disk_num_bytes,
+						NULL);
 			btrfs_free_reserved_extent(fs_info,
-						   ordered_extent->start,
-						   ordered_extent->disk_len, 1);
+					ordered_extent->disk_bytenr,
+					ordered_extent->disk_num_bytes, 1);
+		}
 	}
 
-
 	/*
 	 * This needs to be done to make sure anybody waiting knows we are done
 	 * updating everything for this ordered extent.
 	 */
 	btrfs_remove_ordered_extent(inode, ordered_extent);
 
-	/* for snapshot-aware defrag */
-	if (new) {
-		if (ret) {
-			free_sa_defrag_extent(new);
-			atomic_dec(&fs_info->defrag_running);
-		} else {
-			relink_file_extents(new);
-		}
-	}
-
 	/* once for us */
 	btrfs_put_ordered_extent(ordered_extent);
 	/* once for the tree */
@@ -3254,7 +2709,6 @@ void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start,
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ordered_extent *ordered_extent = NULL;
 	struct btrfs_workqueue *wq;
-	btrfs_work_func_t func;
 
 	trace_btrfs_writepage_end_io_hook(page, start, end, uptodate);
 
@@ -3263,16 +2717,12 @@ void btrfs_writepage_endio_finish_ordered(struct page *page, u64 start,
 					    end - start + 1, uptodate))
 		return;
 
-	if (btrfs_is_free_space_inode(BTRFS_I(inode))) {
+	if (btrfs_is_free_space_inode(BTRFS_I(inode)))
 		wq = fs_info->endio_freespace_worker;
-		func = btrfs_freespace_write_helper;
-	} else {
+	else
 		wq = fs_info->endio_write_workers;
-		func = btrfs_endio_write_helper;
-	}
 
-	btrfs_init_work(&ordered_extent->work, func, finish_ordered_fn, NULL,
-			NULL);
+	btrfs_init_work(&ordered_extent->work, finish_ordered_fn, NULL, NULL);
 	btrfs_queue_work(wq, &ordered_extent->work);
 }
 
@@ -3531,7 +2981,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 		found_key.objectid = found_key.offset;
 		found_key.type = BTRFS_INODE_ITEM_KEY;
 		found_key.offset = 0;
-		inode = btrfs_iget(fs_info->sb, &found_key, root, NULL);
+		inode = btrfs_iget(fs_info->sb, &found_key, root);
 		ret = PTR_ERR_OR_ZERO(inode);
 		if (ret && ret != -ENOENT)
 			goto out;
@@ -3752,6 +3202,8 @@ static int btrfs_read_locked_inode(struct inode *inode,
 	i_uid_write(inode, btrfs_inode_uid(leaf, inode_item));
 	i_gid_write(inode, btrfs_inode_gid(leaf, inode_item));
 	btrfs_i_size_write(BTRFS_I(inode), btrfs_inode_size(leaf, inode_item));
+	btrfs_inode_set_file_extent_range(BTRFS_I(inode), 0,
+			round_up(i_size_read(inode), fs_info->sectorsize));
 
 	inode->i_atime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->atime);
 	inode->i_atime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->atime);
@@ -4202,18 +3654,30 @@ out:
 }
 
 static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
-			       struct inode *dir, u64 objectid,
-			       const char *name, int name_len)
+			       struct inode *dir, struct dentry *dentry)
 {
 	struct btrfs_root *root = BTRFS_I(dir)->root;
+	struct btrfs_inode *inode = BTRFS_I(d_inode(dentry));
 	struct btrfs_path *path;
 	struct extent_buffer *leaf;
 	struct btrfs_dir_item *di;
 	struct btrfs_key key;
+	const char *name = dentry->d_name.name;
+	int name_len = dentry->d_name.len;
 	u64 index;
 	int ret;
+	u64 objectid;
 	u64 dir_ino = btrfs_ino(BTRFS_I(dir));
 
+	if (btrfs_ino(inode) == BTRFS_FIRST_FREE_OBJECTID) {
+		objectid = inode->root->root_key.objectid;
+	} else if (btrfs_ino(inode) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) {
+		objectid = inode->location.objectid;
+	} else {
+		WARN_ON(1);
+		return -EINVAL;
+	}
+
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
@@ -4235,13 +3699,16 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 	}
 	btrfs_release_path(path);
 
-	ret = btrfs_del_root_ref(trans, objectid, root->root_key.objectid,
-				 dir_ino, &index, name, name_len);
-	if (ret < 0) {
-		if (ret != -ENOENT) {
-			btrfs_abort_transaction(trans, ret);
-			goto out;
-		}
+	/*
+	 * This is a placeholder inode for a subvolume we didn't have a
+	 * reference to at the time of the snapshot creation.  In the meantime
+	 * we could have renamed the real subvol link into our snapshot, so
+	 * depending on btrfs_del_root_ref to return -ENOENT here is incorret.
+	 * Instead simply lookup the dir_index_item for this entry so we can
+	 * remove it.  Otherwise we know we have a ref to the root and we can
+	 * call btrfs_del_root_ref, and it _shouldn't_ fail.
+	 */
+	if (btrfs_ino(inode) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) {
 		di = btrfs_search_dir_index_item(root, path, dir_ino,
 						 name, name_len);
 		if (IS_ERR_OR_NULL(di)) {
@@ -4256,8 +3723,16 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 		leaf = path->nodes[0];
 		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 		index = key.offset;
+		btrfs_release_path(path);
+	} else {
+		ret = btrfs_del_root_ref(trans, objectid,
+					 root->root_key.objectid, dir_ino,
+					 &index, name, name_len);
+		if (ret) {
+			btrfs_abort_transaction(trans, ret);
+			goto out;
+		}
 	}
-	btrfs_release_path(path);
 
 	ret = btrfs_delete_delayed_dir_index(trans, BTRFS_I(dir), index);
 	if (ret) {
@@ -4451,8 +3926,7 @@ int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry)
 
 	btrfs_record_snapshot_destroy(trans, BTRFS_I(dir));
 
-	ret = btrfs_unlink_subvol(trans, dir, dest->root_key.objectid,
-				  dentry->d_name.name, dentry->d_name.len);
+	ret = btrfs_unlink_subvol(trans, dir, dentry);
 	if (ret) {
 		err = ret;
 		btrfs_abort_transaction(trans, ret);
@@ -4547,10 +4021,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
 		return PTR_ERR(trans);
 
 	if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
-		err = btrfs_unlink_subvol(trans, dir,
-					  BTRFS_I(inode)->location.objectid,
-					  dentry->d_name.name,
-					  dentry->d_name.len);
+		err = btrfs_unlink_subvol(trans, dir, dentry);
 		goto out;
 	}
 
@@ -4631,6 +4102,8 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 	u64 bytes_deleted = 0;
 	bool be_nice = false;
 	bool should_throttle = false;
+	const u64 lock_start = ALIGN_DOWN(new_size, fs_info->sectorsize);
+	struct extent_state *cached_state = NULL;
 
 	BUG_ON(new_size > 0 && min_type != BTRFS_EXTENT_DATA_KEY);
 
@@ -4647,6 +4120,10 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 		return -ENOMEM;
 	path->reada = READA_BACK;
 
+	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
+		lock_extent_bits(&BTRFS_I(inode)->io_tree, lock_start, (u64)-1,
+				 &cached_state);
+
 	/*
 	 * We want to drop from the next block forward in case this new size is
 	 * not block aligned since we will be keeping the last block of the
@@ -4683,7 +4160,6 @@ search_again:
 		goto out;
 	}
 
-	path->leave_spinning = 1;
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 	if (ret < 0)
 		goto out;
@@ -4699,6 +4175,8 @@ search_again:
 	}
 
 	while (1) {
+		u64 clear_start = 0, clear_len = 0;
+
 		fi = NULL;
 		leaf = path->nodes[0];
 		btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
@@ -4749,6 +4227,8 @@ search_again:
 
 		if (extent_type != BTRFS_FILE_EXTENT_INLINE) {
 			u64 num_dec;
+
+			clear_start = found_key.offset;
 			extent_start = btrfs_file_extent_disk_bytenr(leaf, fi);
 			if (!del_item) {
 				u64 orig_num_bytes =
@@ -4756,6 +4236,7 @@ search_again:
 				extent_num_bytes = ALIGN(new_size -
 						found_key.offset,
 						fs_info->sectorsize);
+				clear_start = ALIGN(new_size, fs_info->sectorsize);
 				btrfs_set_file_extent_num_bytes(leaf, fi,
 							 extent_num_bytes);
 				num_dec = (orig_num_bytes -
@@ -4781,6 +4262,7 @@ search_again:
 						inode_sub_bytes(inode, num_dec);
 				}
 			}
+			clear_len = num_dec;
 		} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
 			/*
 			 * we can't truncate inline items that have had
@@ -4802,12 +4284,33 @@ search_again:
 				 */
 				ret = NEED_TRUNCATE_BLOCK;
 				break;
+			} else {
+				/*
+				 * Inline extents are special, we just treat
+				 * them as a full sector worth in the file
+				 * extent tree just for simplicity sake.
+				 */
+				clear_len = fs_info->sectorsize;
 			}
 
 			if (test_bit(BTRFS_ROOT_REF_COWS, &root->state))
 				inode_sub_bytes(inode, item_end + 1 - new_size);
 		}
 delete:
+		/*
+		 * We use btrfs_truncate_inode_items() to clean up log trees for
+		 * multiple fsyncs, and in this case we don't want to clear the
+		 * file extent range because it's just the log.
+		 */
+		if (root == BTRFS_I(inode)->root) {
+			ret = btrfs_inode_clear_file_extent_range(BTRFS_I(inode),
+						  clear_start, clear_len);
+			if (ret) {
+				btrfs_abort_transaction(trans, ret);
+				break;
+			}
+		}
+
 		if (del_item)
 			last_size = found_key.offset;
 		else
@@ -4835,7 +4338,6 @@ delete:
 		     root == fs_info->tree_root)) {
 			struct btrfs_ref ref = { 0 };
 
-			btrfs_set_path_blocking(path);
 			bytes_deleted += extent_num_bytes;
 
 			btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF,
@@ -4910,7 +4412,9 @@ out:
 		ASSERT(last_size >= new_size);
 		if (!ret && last_size > new_size)
 			last_size = new_size;
-		btrfs_ordered_update_i_size(inode, last_size, NULL);
+		btrfs_inode_safe_disk_i_size_write(inode, last_size);
+		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start,
+				     (u64)-1, &cached_state);
 	}
 
 	btrfs_free_path(path);
@@ -5116,12 +4620,12 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
 	if (size <= hole_start)
 		return 0;
 
-	btrfs_lock_and_flush_ordered_range(io_tree, BTRFS_I(inode), hole_start,
+	btrfs_lock_and_flush_ordered_range(BTRFS_I(inode), hole_start,
 					   block_end - 1, &cached_state);
 	cur_offset = hole_start;
 	while (1) {
 		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset,
-				block_end - cur_offset, 0);
+				      block_end - cur_offset);
 		if (IS_ERR(em)) {
 			err = PTR_ERR(em);
 			em = NULL;
@@ -5129,14 +4633,21 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
 		}
 		last_byte = min(extent_map_end(em), block_end);
 		last_byte = ALIGN(last_byte, fs_info->sectorsize);
+		hole_size = last_byte - cur_offset;
+
 		if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) {
 			struct extent_map *hole_em;
-			hole_size = last_byte - cur_offset;
 
 			err = maybe_insert_hole(root, inode, cur_offset,
 						hole_size);
 			if (err)
 				break;
+
+			err = btrfs_inode_set_file_extent_range(BTRFS_I(inode),
+							cur_offset, hole_size);
+			if (err)
+				break;
+
 			btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset,
 						cur_offset + hole_size - 1, 0);
 			hole_em = alloc_extent_map();
@@ -5153,7 +4664,6 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
 			hole_em->block_len = 0;
 			hole_em->orig_block_len = 0;
 			hole_em->ram_bytes = hole_size;
-			hole_em->bdev = fs_info->fs_devices->latest_bdev;
 			hole_em->compress_type = BTRFS_COMPRESS_NONE;
 			hole_em->generation = fs_info->generation;
 
@@ -5169,6 +4679,11 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
 							hole_size - 1, 0);
 			}
 			free_extent_map(hole_em);
+		} else {
+			err = btrfs_inode_set_file_extent_range(BTRFS_I(inode),
+							cur_offset, hole_size);
+			if (err)
+				break;
 		}
 next:
 		free_extent_map(em);
@@ -5212,24 +4727,24 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
 		 * truncation, it must capture all writes that happened before
 		 * this truncation.
 		 */
-		btrfs_wait_for_snapshot_creation(root);
+		btrfs_drew_write_lock(&root->snapshot_lock);
 		ret = btrfs_cont_expand(inode, oldsize, newsize);
 		if (ret) {
-			btrfs_end_write_no_snapshotting(root);
+			btrfs_drew_write_unlock(&root->snapshot_lock);
 			return ret;
 		}
 
 		trans = btrfs_start_transaction(root, 1);
 		if (IS_ERR(trans)) {
-			btrfs_end_write_no_snapshotting(root);
+			btrfs_drew_write_unlock(&root->snapshot_lock);
 			return PTR_ERR(trans);
 		}
 
 		i_size_write(inode, newsize);
-		btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL);
+		btrfs_inode_safe_disk_i_size_write(inode, 0);
 		pagecache_isize_extended(inode, oldsize, newsize);
 		ret = btrfs_update_inode(trans, root, inode);
-		btrfs_end_write_no_snapshotting(root);
+		btrfs_drew_write_unlock(&root->snapshot_lock);
 		btrfs_end_transaction(trans);
 	} else {
 
@@ -5341,8 +4856,8 @@ static void evict_inode_truncate_pages(struct inode *inode)
 
 	/*
 	 * Keep looping until we have no more ranges in the io tree.
-	 * We can have ongoing bios started by readpages (called from readahead)
-	 * that have their endio callback (extent_io.c:end_bio_extent_readpage)
+	 * We can have ongoing bios started by readahead that have
+	 * their endio callback (extent_io.c:end_bio_extent_readpage)
 	 * still in progress (unlocked the pages in the bio but did not yet
 	 * unlocked the ranges in the io tree). Therefore this means some
 	 * ranges can still be locked and eviction started because before
@@ -5639,7 +5154,7 @@ static int fixup_tree_root_location(struct btrfs_fs_info *fs_info,
 
 	btrfs_release_path(path);
 
-	new_root = btrfs_read_fs_root_no_name(fs_info, location);
+	new_root = btrfs_get_fs_root(fs_info, location, true);
 	if (IS_ERR(new_root)) {
 		err = PTR_ERR(new_root);
 		goto out;
@@ -5693,7 +5208,6 @@ static void inode_tree_add(struct inode *inode)
 
 static void inode_tree_del(struct inode *inode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int empty = 0;
 
@@ -5706,7 +5220,6 @@ static void inode_tree_del(struct inode *inode)
 	spin_unlock(&root->inode_lock);
 
 	if (empty && btrfs_root_refs(&root->root_item) == 0) {
-		synchronize_srcu(&fs_info->subvol_srcu);
 		spin_lock(&root->inode_lock);
 		empty = RB_EMPTY_ROOT(&root->inode_tree);
 		spin_unlock(&root->inode_lock);
@@ -5722,7 +5235,8 @@ static int btrfs_init_locked_inode(struct inode *inode, void *p)
 	inode->i_ino = args->location->objectid;
 	memcpy(&BTRFS_I(inode)->location, args->location,
 	       sizeof(*args->location));
-	BTRFS_I(inode)->root = args->root;
+	BTRFS_I(inode)->root = btrfs_grab_root(args->root);
+	BUG_ON(args->root && !BTRFS_I(inode)->root);
 	return 0;
 }
 
@@ -5750,12 +5264,14 @@ static struct inode *btrfs_iget_locked(struct super_block *s,
 	return inode;
 }
 
-/* Get an inode object given its location and corresponding root.
- * Returns in *is_new if the inode was read from disk
+/*
+ * Get an inode object given its location and corresponding root.
+ * Path can be preallocated to prevent recursing back to iget through
+ * allocator. NULL is also valid but may require an additional allocation
+ * later.
  */
 struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
-			      struct btrfs_root *root, int *new,
-			      struct btrfs_path *path)
+			      struct btrfs_root *root, struct btrfs_path *path)
 {
 	struct inode *inode;
 
@@ -5770,8 +5286,6 @@ struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
 		if (!ret) {
 			inode_tree_add(inode);
 			unlock_new_inode(inode);
-			if (new)
-				*new = 1;
 		} else {
 			iget_failed(inode);
 			/*
@@ -5789,9 +5303,9 @@ struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
 }
 
 struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
-			 struct btrfs_root *root, int *new)
+			 struct btrfs_root *root)
 {
-	return btrfs_iget_path(s, location, root, new, NULL);
+	return btrfs_iget_path(s, location, root, NULL);
 }
 
 static struct inode *new_simple_dir(struct super_block *s,
@@ -5803,12 +5317,16 @@ static struct inode *new_simple_dir(struct super_block *s,
 	if (!inode)
 		return ERR_PTR(-ENOMEM);
 
-	BTRFS_I(inode)->root = root;
+	BTRFS_I(inode)->root = btrfs_grab_root(root);
 	memcpy(&BTRFS_I(inode)->location, key, sizeof(*key));
 	set_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags);
 
 	inode->i_ino = BTRFS_EMPTY_SUBVOL_DIR_OBJECTID;
-	inode->i_op = &btrfs_dir_ro_inode_operations;
+	/*
+	 * We only need lookup, the rest is read-only and there's no inode
+	 * associated with the dentry
+	 */
+	inode->i_op = &simple_dir_inode_operations;
 	inode->i_opflags &= ~IOP_XATTR;
 	inode->i_fop = &simple_dir_operations;
 	inode->i_mode = S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO;
@@ -5846,7 +5364,6 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 	struct btrfs_root *sub_root = root;
 	struct btrfs_key location;
 	u8 di_type = 0;
-	int index;
 	int ret = 0;
 
 	if (dentry->d_name.len > BTRFS_NAME_LEN)
@@ -5857,7 +5374,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 		return ERR_PTR(ret);
 
 	if (location.type == BTRFS_INODE_ITEM_KEY) {
-		inode = btrfs_iget(dir->i_sb, &location, root, NULL);
+		inode = btrfs_iget(dir->i_sb, &location, root);
 		if (IS_ERR(inode))
 			return inode;
 
@@ -5873,7 +5390,6 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 		return inode;
 	}
 
-	index = srcu_read_lock(&fs_info->subvol_srcu);
 	ret = fixup_tree_root_location(fs_info, dir, dentry,
 				       &location, &sub_root);
 	if (ret < 0) {
@@ -5882,9 +5398,10 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 		else
 			inode = new_simple_dir(dir->i_sb, &location, sub_root);
 	} else {
-		inode = btrfs_iget(dir->i_sb, &location, sub_root, NULL);
+		inode = btrfs_iget(dir->i_sb, &location, sub_root);
 	}
-	srcu_read_unlock(&fs_info->subvol_srcu, index);
+	if (root != sub_root)
+		btrfs_put_root(sub_root);
 
 	if (!IS_ERR(inode) && root != sub_root) {
 		down_read(&fs_info->cleanup_work_sem);
@@ -6365,7 +5882,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	 */
 	BTRFS_I(inode)->index_cnt = 2;
 	BTRFS_I(inode)->dir_index = *index;
-	BTRFS_I(inode)->root = root;
+	BTRFS_I(inode)->root = btrfs_grab_root(root);
 	BTRFS_I(inode)->generation = trans->transid;
 	inode->i_generation = BTRFS_I(inode)->generation;
 
@@ -6899,18 +6416,27 @@ static noinline int uncompress_inline(struct btrfs_path *path,
 	return ret;
 }
 
-/*
- * a bit scary, this does extent mapping from logical file offset to the disk.
- * the ugly parts come from merging extents from the disk with the in-ram
- * representation.  This gets more complex because of the data=ordered code,
- * where the in-ram extents might be locked pending data=ordered completion.
+/**
+ * btrfs_get_extent - Lookup the first extent overlapping a range in a file.
+ * @inode:	file to search in
+ * @page:	page to read extent data into if the extent is inline
+ * @pg_offset:	offset into @page to copy to
+ * @start:	file offset
+ * @len:	length of range starting at @start
+ *
+ * This returns the first &struct extent_map which overlaps with the given
+ * range, reading it from the B-tree and caching it if necessary. Note that
+ * there may be more extents which overlap the given range after the returned
+ * extent_map.
+ *
+ * If @page is not NULL and the extent is inline, this also reads the extent
+ * data directly into the page and marks the extent up to date in the io_tree.
  *
- * This also copies inline extents directly into the page.
+ * Return: ERR_PTR on error, non-NULL extent_map on success.
  */
 struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
-				    struct page *page,
-				    size_t pg_offset, u64 start, u64 len,
-				    int create)
+				    struct page *page, size_t pg_offset,
+				    u64 start, u64 len)
 {
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	int ret;
@@ -6927,12 +6453,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	struct extent_map *em = NULL;
 	struct extent_map_tree *em_tree = &inode->extent_tree;
 	struct extent_io_tree *io_tree = &inode->io_tree;
-	const bool new_inline = !page || create;
 
 	read_lock(&em_tree->lock);
 	em = lookup_extent_mapping(em_tree, start, len);
-	if (em)
-		em->bdev = fs_info->fs_devices->latest_bdev;
 	read_unlock(&em_tree->lock);
 
 	if (em) {
@@ -6948,7 +6471,6 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 		err = -ENOMEM;
 		goto out;
 	}
-	em->bdev = fs_info->fs_devices->latest_bdev;
 	em->start = EXTENT_MAP_HOLE;
 	em->orig_start = EXTENT_MAP_HOLE;
 	em->len = (u64)-1;
@@ -6997,6 +6519,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 
 	extent_type = btrfs_file_extent_type(leaf, item);
 	extent_start = found_key.offset;
+	extent_end = btrfs_file_extent_end(path);
 	if (extent_type == BTRFS_FILE_EXTENT_REG ||
 	    extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
 		/* Only regular file could have regular/prealloc extent */
@@ -7007,18 +6530,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 				   btrfs_ino(inode));
 			goto out;
 		}
-		extent_end = extent_start +
-		       btrfs_file_extent_num_bytes(leaf, item);
-
 		trace_btrfs_get_extent_show_fi_regular(inode, leaf, item,
 						       extent_start);
 	} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
-		size_t size;
-
-		size = btrfs_file_extent_ram_bytes(leaf, item);
-		extent_end = ALIGN(extent_start + size,
-				   fs_info->sectorsize);
-
 		trace_btrfs_get_extent_show_fi_inline(inode, leaf, item,
 						      path->slots[0],
 						      extent_start);
@@ -7053,8 +6567,7 @@ next:
 		goto insert;
 	}
 
-	btrfs_extent_item_to_extent_map(inode, path, item,
-			new_inline, em);
+	btrfs_extent_item_to_extent_map(inode, path, item, !page, em);
 
 	if (extent_type == BTRFS_FILE_EXTENT_REG ||
 	    extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
@@ -7066,7 +6579,7 @@ next:
 		size_t extent_offset;
 		size_t copy_size;
 
-		if (new_inline)
+		if (!page)
 			goto out;
 
 		size = btrfs_file_extent_ram_bytes(leaf, item);
@@ -7149,7 +6662,7 @@ struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
 	u64 delalloc_end;
 	int err = 0;
 
-	em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
+	em = btrfs_get_extent(inode, NULL, 0, start, len);
 	if (IS_ERR(em))
 		return em;
 	/*
@@ -7207,7 +6720,6 @@ struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode,
 			err = -ENOMEM;
 			goto out;
 		}
-		em->bdev = NULL;
 
 		ASSERT(hole_em);
 		/*
@@ -7538,11 +7050,11 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
 			 * for it to complete) and then invalidate the pages for
 			 * this range (through invalidate_inode_pages2_range()),
 			 * but that can lead us to a deadlock with a concurrent
-			 * call to readpages() (a buffered read or a defrag call
+			 * call to readahead (a buffered read or a defrag call
 			 * triggered a readahead) on a page lock due to an
 			 * ordered dio extent we created before but did not have
 			 * yet a corresponding bio submitted (whence it can not
-			 * complete), which makes readpages() wait for that
+			 * complete), which makes readahead wait for that
 			 * ordered extent to complete while holding a lock on
 			 * that page.
 			 */
@@ -7567,7 +7079,6 @@ static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len,
 {
 	struct extent_map_tree *em_tree;
 	struct extent_map *em;
-	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int ret;
 
 	ASSERT(type == BTRFS_ORDERED_PREALLOC ||
@@ -7585,7 +7096,6 @@ static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len,
 	em->len = len;
 	em->block_len = block_len;
 	em->block_start = block_start;
-	em->bdev = root->fs_info->fs_devices->latest_bdev;
 	em->orig_block_len = orig_block_len;
 	em->ram_bytes = ram_bytes;
 	em->generation = -1;
@@ -7624,6 +7134,8 @@ static int btrfs_get_blocks_direct_read(struct extent_map *em,
 					struct inode *inode,
 					u64 start, u64 len)
 {
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+
 	if (em->block_start == EXTENT_MAP_HOLE ||
 			test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
 		return -ENOENT;
@@ -7633,7 +7145,7 @@ static int btrfs_get_blocks_direct_read(struct extent_map *em,
 	bh_result->b_blocknr = (em->block_start + (start - em->start)) >>
 		inode->i_blkbits;
 	bh_result->b_size = len;
-	bh_result->b_bdev = em->bdev;
+	bh_result->b_bdev = fs_info->fs_devices->latest_bdev;
 	set_buffer_mapped(bh_result);
 
 	return 0;
@@ -7716,7 +7228,7 @@ skip_cow:
 	bh_result->b_blocknr = (em->block_start + (start - em->start)) >>
 		inode->i_blkbits;
 	bh_result->b_size = len;
-	bh_result->b_bdev = em->bdev;
+	bh_result->b_bdev = fs_info->fs_devices->latest_bdev;
 	set_buffer_mapped(bh_result);
 
 	if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
@@ -7775,7 +7287,7 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock,
 		goto err;
 	}
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len);
 	if (IS_ERR(em)) {
 		ret = PTR_ERR(em);
 		goto unlock_err;
@@ -7858,7 +7370,7 @@ static inline blk_status_t submit_dio_repair_bio(struct inode *inode,
 	if (ret)
 		return ret;
 
-	ret = btrfs_map_bio(fs_info, bio, mirror_num, 0);
+	ret = btrfs_map_bio(fs_info, bio, mirror_num);
 
 	return ret;
 }
@@ -8211,18 +7723,14 @@ static void __endio_write_update_ordered(struct inode *inode,
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ordered_extent *ordered = NULL;
 	struct btrfs_workqueue *wq;
-	btrfs_work_func_t func;
 	u64 ordered_offset = offset;
 	u64 ordered_bytes = bytes;
 	u64 last_offset;
 
-	if (btrfs_is_free_space_inode(BTRFS_I(inode))) {
+	if (btrfs_is_free_space_inode(BTRFS_I(inode)))
 		wq = fs_info->endio_freespace_worker;
-		func = btrfs_freespace_write_helper;
-	} else {
+	else
 		wq = fs_info->endio_write_workers;
-		func = btrfs_endio_write_helper;
-	}
 
 	while (ordered_offset < offset + bytes) {
 		last_offset = ordered_offset;
@@ -8230,9 +7738,8 @@ static void __endio_write_update_ordered(struct inode *inode,
 							   &ordered_offset,
 							   ordered_bytes,
 							   uptodate)) {
-			btrfs_init_work(&ordered->work, func,
-					finish_ordered_fn,
-					NULL, NULL);
+			btrfs_init_work(&ordered->work, finish_ordered_fn, NULL,
+					NULL);
 			btrfs_queue_work(wq, &ordered->work);
 		}
 		/*
@@ -8324,6 +7831,7 @@ static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode,
 {
 	struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
 	struct btrfs_io_bio *orig_io_bio = btrfs_io_bio(dip->orig_bio);
+	u16 csum_size;
 	blk_status_t ret;
 
 	/*
@@ -8332,8 +7840,8 @@ static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode,
 	 * contention.
 	 */
 	if (dip->logical_offset == file_offset) {
-		ret = btrfs_lookup_bio_sums_dio(inode, dip->orig_bio,
-						file_offset);
+		ret = btrfs_lookup_bio_sums(inode, dip->orig_bio, file_offset,
+					    NULL);
 		if (ret)
 			return ret;
 	}
@@ -8343,7 +7851,8 @@ static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode,
 
 	file_offset -= dip->logical_offset;
 	file_offset >>= inode->i_sb->s_blocksize_bits;
-	io_bio->csum = (u8 *)(((u32 *)orig_io_bio->csum) + file_offset);
+	csum_size = btrfs_super_csum_size(btrfs_sb(inode->i_sb)->super_copy);
+	io_bio->csum = orig_io_bio->csum + csum_size * file_offset;
 
 	return 0;
 }
@@ -8389,7 +7898,7 @@ static inline blk_status_t btrfs_submit_dio_bio(struct bio *bio,
 			goto err;
 	}
 map:
-	ret = btrfs_map_bio(fs_info, bio, 0, 0);
+	ret = btrfs_map_bio(fs_info, bio, 0);
 err:
 	return ret;
 }
@@ -8750,9 +8259,7 @@ static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 
 int btrfs_readpage(struct file *file, struct page *page)
 {
-	struct extent_io_tree *tree;
-	tree = &BTRFS_I(page->mapping->host)->io_tree;
-	return extent_read_full_page(tree, page, btrfs_get_extent, 0);
+	return extent_read_full_page(page, btrfs_get_extent, 0);
 }
 
 static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
@@ -8786,21 +8293,16 @@ static int btrfs_writepages(struct address_space *mapping,
 	return extent_writepages(mapping, wbc);
 }
 
-static int
-btrfs_readpages(struct file *file, struct address_space *mapping,
-		struct list_head *pages, unsigned nr_pages)
+static void btrfs_readahead(struct readahead_control *rac)
 {
-	return extent_readpages(mapping, pages, nr_pages);
+	extent_readahead(rac);
 }
 
 static int __btrfs_releasepage(struct page *page, gfp_t gfp_flags)
 {
 	int ret = try_release_extent_mapping(page, gfp_flags);
-	if (ret == 1) {
-		ClearPagePrivate(page);
-		set_page_private(page, 0);
-		put_page(page);
-	}
+	if (ret == 1)
+		detach_page_private(page);
 	return ret;
 }
 
@@ -8811,6 +8313,33 @@ static int btrfs_releasepage(struct page *page, gfp_t gfp_flags)
 	return __btrfs_releasepage(page, gfp_flags);
 }
 
+#ifdef CONFIG_MIGRATION
+static int btrfs_migratepage(struct address_space *mapping,
+			     struct page *newpage, struct page *page,
+			     enum migrate_mode mode)
+{
+	int ret;
+
+	ret = migrate_page_move_mapping(mapping, newpage, page, 0);
+	if (ret != MIGRATEPAGE_SUCCESS)
+		return ret;
+
+	if (page_has_private(page))
+		attach_page_private(newpage, detach_page_private(page));
+
+	if (PagePrivate2(page)) {
+		ClearPagePrivate2(page);
+		SetPagePrivate2(newpage);
+	}
+
+	if (mode != MIGRATE_SYNC_NO_COPY)
+		migrate_page_copy(newpage, page);
+	else
+		migrate_page_states(newpage, page);
+	return MIGRATEPAGE_SUCCESS;
+}
+#endif
+
 static void btrfs_invalidatepage(struct page *page, unsigned int offset,
 				 unsigned int length)
 {
@@ -8846,7 +8375,8 @@ again:
 	ordered = btrfs_lookup_ordered_range(BTRFS_I(inode), start,
 					page_end - start + 1);
 	if (ordered) {
-		end = min(page_end, ordered->file_offset + ordered->len - 1);
+		end = min(page_end,
+			  ordered->file_offset + ordered->num_bytes - 1);
 		/*
 		 * IO on this page will never be started, so we need
 		 * to account for any ordered extents now
@@ -8917,11 +8447,7 @@ again:
 	}
 
 	ClearPageChecked(page);
-	if (PagePrivate(page)) {
-		ClearPagePrivate(page);
-		set_page_private(page, 0);
-		put_page(page);
-	}
+	detach_page_private(page);
 }
 
 /*
@@ -9047,7 +8573,6 @@ again:
 		ret = VM_FAULT_SIGBUS;
 		goto out_unlock;
 	}
-	ret2 = 0;
 
 	/* page is wholly or partially inside EOF */
 	if (page_start + PAGE_SIZE > size)
@@ -9071,12 +8596,10 @@ again:
 
 	unlock_extent_cached(io_tree, page_start, page_end, &cached_state);
 
-	if (!ret2) {
-		btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE);
-		sb_end_pagefault(inode->i_sb);
-		extent_changeset_free(data_reserved);
-		return VM_FAULT_LOCKED;
-	}
+	btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE);
+	sb_end_pagefault(inode->i_sb);
+	extent_changeset_free(data_reserved);
+	return VM_FAULT_LOCKED;
 
 out_unlock:
 	unlock_page(page);
@@ -9188,7 +8711,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 			break;
 		}
 
-		btrfs_block_rsv_release(fs_info, rsv, -1);
+		btrfs_block_rsv_release(fs_info, rsv, -1, NULL);
 		ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv,
 					      rsv, min_size, false);
 		BUG_ON(ret);	/* shouldn't happen */
@@ -9213,7 +8736,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback)
 			ret = PTR_ERR(trans);
 			goto out;
 		}
-		btrfs_ordered_update_i_size(inode, inode->i_size, NULL);
+		btrfs_inode_safe_disk_i_size_write(inode, 0);
 	}
 
 	if (trans) {
@@ -9317,11 +8840,12 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
 	extent_io_tree_init(fs_info, &ei->io_tree, IO_TREE_INODE_IO, inode);
 	extent_io_tree_init(fs_info, &ei->io_failure_tree,
 			    IO_TREE_INODE_IO_FAILURE, inode);
+	extent_io_tree_init(fs_info, &ei->file_extent_tree,
+			    IO_TREE_INODE_FILE_EXTENT, inode);
 	ei->io_tree.track_uptodate = true;
 	ei->io_failure_tree.track_uptodate = true;
 	atomic_set(&ei->sync_writers, 0);
 	mutex_init(&ei->log_mutex);
-	mutex_init(&ei->delalloc_mutex);
 	btrfs_ordered_inode_tree_init(&ei->ordered_tree);
 	INIT_LIST_HEAD(&ei->delalloc_inodes);
 	INIT_LIST_HEAD(&ei->delayed_iput);
@@ -9375,7 +8899,7 @@ void btrfs_destroy_inode(struct inode *inode)
 		else {
 			btrfs_err(fs_info,
 				  "found ordered extent %llu %llu on inode cleanup",
-				  ordered->file_offset, ordered->len);
+				  ordered->file_offset, ordered->num_bytes);
 			btrfs_remove_ordered_extent(inode, ordered);
 			btrfs_put_ordered_extent(ordered);
 			btrfs_put_ordered_extent(ordered);
@@ -9384,6 +8908,8 @@ void btrfs_destroy_inode(struct inode *inode)
 	btrfs_qgroup_check_reserved_leak(inode);
 	inode_tree_del(inode);
 	btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0);
+	btrfs_inode_clear_file_extent_range(BTRFS_I(inode), 0, (u64)-1);
+	btrfs_put_root(BTRFS_I(inode)->root);
 }
 
 int btrfs_drop_inode(struct inode *inode)
@@ -9513,7 +9039,6 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 	u64 new_ino = btrfs_ino(BTRFS_I(new_inode));
 	u64 old_idx = 0;
 	u64 new_idx = 0;
-	u64 root_objectid;
 	int ret;
 	bool root_log_pinned = false;
 	bool dest_log_pinned = false;
@@ -9531,9 +9056,8 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 	btrfs_init_log_ctx(&ctx_dest, new_inode);
 
 	/* close the race window with snapshot create/destroy ioctl */
-	if (old_ino == BTRFS_FIRST_FREE_OBJECTID)
-		down_read(&fs_info->subvol_sem);
-	if (new_ino == BTRFS_FIRST_FREE_OBJECTID)
+	if (old_ino == BTRFS_FIRST_FREE_OBJECTID ||
+	    new_ino == BTRFS_FIRST_FREE_OBJECTID)
 		down_read(&fs_info->subvol_sem);
 
 	/*
@@ -9550,6 +9074,9 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 		goto out_notrans;
 	}
 
+	if (dest != root)
+		btrfs_record_root_in_trans(trans, dest);
+
 	/*
 	 * We need to find a free sequence number both in the source and
 	 * in the destination directory for the exchange.
@@ -9617,10 +9144,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 
 	/* src is a subvolume */
 	if (old_ino == BTRFS_FIRST_FREE_OBJECTID) {
-		root_objectid = BTRFS_I(old_inode)->root->root_key.objectid;
-		ret = btrfs_unlink_subvol(trans, old_dir, root_objectid,
-					  old_dentry->d_name.name,
-					  old_dentry->d_name.len);
+		ret = btrfs_unlink_subvol(trans, old_dir, old_dentry);
 	} else { /* src is an inode */
 		ret = __btrfs_unlink_inode(trans, root, BTRFS_I(old_dir),
 					   BTRFS_I(old_dentry->d_inode),
@@ -9636,10 +9160,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 
 	/* dest is a subvolume */
 	if (new_ino == BTRFS_FIRST_FREE_OBJECTID) {
-		root_objectid = BTRFS_I(new_inode)->root->root_key.objectid;
-		ret = btrfs_unlink_subvol(trans, new_dir, root_objectid,
-					  new_dentry->d_name.name,
-					  new_dentry->d_name.len);
+		ret = btrfs_unlink_subvol(trans, new_dir, new_dentry);
 	} else { /* dest is an inode */
 		ret = __btrfs_unlink_inode(trans, dest, BTRFS_I(new_dir),
 					   BTRFS_I(new_dentry->d_inode),
@@ -9764,9 +9285,8 @@ out_fail:
 		ret = ret ? ret : ret2;
 	}
 out_notrans:
-	if (new_ino == BTRFS_FIRST_FREE_OBJECTID)
-		up_read(&fs_info->subvol_sem);
-	if (old_ino == BTRFS_FIRST_FREE_OBJECTID)
+	if (new_ino == BTRFS_FIRST_FREE_OBJECTID ||
+	    old_ino == BTRFS_FIRST_FREE_OBJECTID)
 		up_read(&fs_info->subvol_sem);
 
 	ASSERT(list_empty(&ctx_root.list));
@@ -9838,7 +9358,6 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct inode *new_inode = d_inode(new_dentry);
 	struct inode *old_inode = d_inode(old_dentry);
 	u64 index = 0;
-	u64 root_objectid;
 	int ret;
 	u64 old_ino = btrfs_ino(BTRFS_I(old_inode));
 	bool log_pinned = false;
@@ -9946,10 +9465,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 				BTRFS_I(old_inode), 1);
 
 	if (unlikely(old_ino == BTRFS_FIRST_FREE_OBJECTID)) {
-		root_objectid = BTRFS_I(old_inode)->root->root_key.objectid;
-		ret = btrfs_unlink_subvol(trans, old_dir, root_objectid,
-					old_dentry->d_name.name,
-					old_dentry->d_name.len);
+		ret = btrfs_unlink_subvol(trans, old_dir, old_dentry);
 	} else {
 		ret = __btrfs_unlink_inode(trans, root, BTRFS_I(old_dir),
 					BTRFS_I(d_inode(old_dentry)),
@@ -9968,10 +9484,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		new_inode->i_ctime = current_time(new_inode);
 		if (unlikely(btrfs_ino(BTRFS_I(new_inode)) ==
 			     BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
-			root_objectid = BTRFS_I(new_inode)->location.objectid;
-			ret = btrfs_unlink_subvol(trans, new_dir, root_objectid,
-						new_dentry->d_name.name,
-						new_dentry->d_name.len);
+			ret = btrfs_unlink_subvol(trans, new_dir, new_dentry);
 			BUG_ON(new_inode->i_nlink == 0);
 		} else {
 			ret = btrfs_unlink_inode(trans, dest, BTRFS_I(new_dir),
@@ -10051,6 +9564,10 @@ out_fail:
 		ret = btrfs_sync_log(trans, BTRFS_I(old_inode)->root, &ctx);
 		if (ret)
 			commit_transaction = true;
+	} else if (sync_log) {
+		mutex_lock(&root->log_mutex);
+		list_del(&ctx.list);
+		mutex_unlock(&root->log_mutex);
 	}
 	if (commit_transaction) {
 		ret = btrfs_commit_transaction(trans);
@@ -10116,8 +9633,7 @@ static struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode
 	init_completion(&work->completion);
 	INIT_LIST_HEAD(&work->list);
 	work->inode = inode;
-	btrfs_init_work(&work->work, btrfs_flush_delalloc_helper,
-			btrfs_run_delalloc_work, NULL, NULL);
+	btrfs_init_work(&work->work, btrfs_run_delalloc_work, NULL, NULL);
 
 	return work;
 }
@@ -10221,14 +9737,14 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr)
 	while (!list_empty(&splice) && nr) {
 		root = list_first_entry(&splice, struct btrfs_root,
 					delalloc_root);
-		root = btrfs_grab_fs_root(root);
+		root = btrfs_grab_root(root);
 		BUG_ON(!root);
 		list_move_tail(&root->delalloc_root,
 			       &fs_info->delalloc_roots);
 		spin_unlock(&fs_info->delalloc_root_lock);
 
 		ret = start_delalloc_inodes(root, nr, false);
-		btrfs_put_fs_root(root);
+		btrfs_put_root(root);
 		if (ret < 0)
 			goto out;
 
@@ -10382,6 +9898,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_key ins;
 	u64 cur_offset = start;
+	u64 clear_offset = start;
 	u64 i_size;
 	u64 cur_bytes;
 	u64 last_alloc = (u64)-1;
@@ -10416,6 +9933,15 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 				btrfs_end_transaction(trans);
 			break;
 		}
+
+		/*
+		 * We've reserved this space, and thus converted it from
+		 * ->bytes_may_use to ->bytes_reserved.  Any error that happens
+		 * from here on out we will only need to clear our reservation
+		 * for the remaining unreserved area, so advance our
+		 * clear_offset by our extent size.
+		 */
+		clear_offset += ins.offset;
 		btrfs_dec_block_group_reservations(fs_info, ins.objectid);
 
 		last_alloc = ins.offset;
@@ -10450,7 +9976,6 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 		em->block_len = ins.offset;
 		em->orig_block_len = ins.offset;
 		em->ram_bytes = ins.offset;
-		em->bdev = fs_info->fs_devices->latest_bdev;
 		set_bit(EXTENT_FLAG_PREALLOC, &em->flags);
 		em->generation = trans->transid;
 
@@ -10481,7 +10006,7 @@ next:
 			else
 				i_size = cur_offset;
 			i_size_write(inode, i_size);
-			btrfs_ordered_update_i_size(inode, i_size, NULL);
+			btrfs_inode_safe_disk_i_size_write(inode, 0);
 		}
 
 		ret = btrfs_update_inode(trans, root, inode);
@@ -10496,9 +10021,9 @@ next:
 		if (own_trans)
 			btrfs_end_transaction(trans);
 	}
-	if (cur_offset < end)
-		btrfs_free_reserved_data_space(inode, NULL, cur_offset,
-			end - cur_offset + 1);
+	if (clear_offset < end)
+		btrfs_free_reserved_data_space(inode, NULL, clear_offset,
+			end - clear_offset + 1);
 	return ret;
 }
 
@@ -10806,10 +10331,10 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 	start = 0;
 	while (start < isize) {
 		u64 logical_block_start, physical_block_start;
-		struct btrfs_block_group_cache *bg;
+		struct btrfs_block_group *bg;
 		u64 len = isize - start;
 
-		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0);
+		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len);
 		if (IS_ERR(em)) {
 			ret = PTR_ERR(em);
 			goto out;
@@ -10977,11 +10502,6 @@ static const struct inode_operations btrfs_dir_inode_operations = {
 	.update_time	= btrfs_update_time,
 	.tmpfile        = btrfs_tmpfile,
 };
-static const struct inode_operations btrfs_dir_ro_inode_operations = {
-	.lookup		= btrfs_lookup,
-	.permission	= btrfs_permission,
-	.update_time	= btrfs_update_time,
-};
 
 static const struct file_operations btrfs_dir_file_operations = {
 	.llseek		= generic_file_llseek,
@@ -11018,10 +10538,13 @@ static const struct address_space_operations btrfs_aops = {
 	.readpage	= btrfs_readpage,
 	.writepage	= btrfs_writepage,
 	.writepages	= btrfs_writepages,
-	.readpages	= btrfs_readpages,
+	.readahead	= btrfs_readahead,
 	.direct_IO	= btrfs_direct_IO,
 	.invalidatepage = btrfs_invalidatepage,
 	.releasepage	= btrfs_releasepage,
+#ifdef CONFIG_MIGRATION
+	.migratepage	= btrfs_migratepage,
+#endif
 	.set_page_dirty	= btrfs_set_page_dirty,
 	.error_remove_page = generic_error_remove_page,
 	.swap_activate	= btrfs_swap_activate,
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 23272d9154f3..40b729dce91c 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -28,6 +28,7 @@
 #include <linux/iversion.h>
 #include "ctree.h"
 #include "disk-io.h"
+#include "export.h"
 #include "transaction.h"
 #include "btrfs_inode.h"
 #include "print-tree.h"
@@ -86,10 +87,6 @@ struct btrfs_ioctl_send_args_32 {
 			       struct btrfs_ioctl_send_args_32)
 #endif
 
-static int btrfs_clone(struct inode *src, struct inode *inode,
-		       u64 off, u64 olen, u64 olen_aligned, u64 destoff,
-		       int no_time_update);
-
 /* Mask out flags that are inappropriate for the given type of inode. */
 static unsigned int btrfs_mask_fsflags_for_type(struct inode *inode,
 		unsigned int flags)
@@ -479,10 +476,9 @@ static int btrfs_ioctl_getversion(struct file *file, int __user *arg)
 	return put_user(inode->i_generation, arg);
 }
 
-static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
+static noinline int btrfs_ioctl_fitrim(struct btrfs_fs_info *fs_info,
+					void __user *arg)
 {
-	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_device *device;
 	struct request_queue *q;
 	struct fstrim_range range;
@@ -541,7 +537,7 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
 	return 0;
 }
 
-int btrfs_is_empty_uuid(u8 *uuid)
+int __pure btrfs_is_empty_uuid(u8 *uuid)
 {
 	int i;
 
@@ -555,7 +551,6 @@ int btrfs_is_empty_uuid(u8 *uuid)
 static noinline int create_subvol(struct inode *dir,
 				  struct dentry *dentry,
 				  const char *name, int namelen,
-				  u64 *async_transid,
 				  struct btrfs_qgroup_inherit *inherit)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
@@ -574,7 +569,6 @@ static noinline int create_subvol(struct inode *dir,
 	u64 objectid;
 	u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
 	u64 index = 0;
-	uuid_le new_uuid;
 
 	root_item = kzalloc(sizeof(*root_item), GFP_KERNEL);
 	if (!root_item)
@@ -644,8 +638,7 @@ static noinline int create_subvol(struct inode *dir,
 
 	btrfs_set_root_generation_v2(root_item,
 			btrfs_root_generation(root_item));
-	uuid_le_gen(&new_uuid);
-	memcpy(root_item->uuid, new_uuid.b, BTRFS_UUID_SIZE);
+	generate_random_guid(root_item->uuid);
 	btrfs_set_stack_timespec_sec(&root_item->otime, cur_time.tv_sec);
 	btrfs_set_stack_timespec_nsec(&root_item->otime, cur_time.tv_nsec);
 	root_item->ctime = root_item->otime;
@@ -667,7 +660,7 @@ static noinline int create_subvol(struct inode *dir,
 		goto fail;
 
 	key.offset = (u64)-1;
-	new_root = btrfs_read_fs_root_no_name(fs_info, &key);
+	new_root = btrfs_get_fs_root(fs_info, &key, true);
 	if (IS_ERR(new_root)) {
 		ret = PTR_ERR(new_root);
 		btrfs_abort_transaction(trans, ret);
@@ -677,6 +670,7 @@ static noinline int create_subvol(struct inode *dir,
 	btrfs_record_root_in_trans(trans, new_root);
 
 	ret = btrfs_create_subvol_root(trans, new_root, root, new_dirid);
+	btrfs_put_root(new_root);
 	if (ret) {
 		/* We potentially lose an unused inode item here */
 		btrfs_abort_transaction(trans, ret);
@@ -705,11 +699,17 @@ static noinline int create_subvol(struct inode *dir,
 
 	btrfs_i_size_write(BTRFS_I(dir), dir->i_size + namelen * 2);
 	ret = btrfs_update_inode(trans, root, dir);
-	BUG_ON(ret);
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
+		goto fail;
+	}
 
 	ret = btrfs_add_root_ref(trans, objectid, root->root_key.objectid,
 				 btrfs_ino(BTRFS_I(dir)), index, name, namelen);
-	BUG_ON(ret);
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
+		goto fail;
+	}
 
 	ret = btrfs_uuid_tree_add(trans, root_item->uuid,
 				  BTRFS_UUID_KEY_SUBVOL, objectid);
@@ -722,14 +722,7 @@ fail:
 	trans->bytes_reserved = 0;
 	btrfs_subvolume_release_metadata(fs_info, &block_rsv);
 
-	if (async_transid) {
-		*async_transid = trans->transid;
-		err = btrfs_commit_transaction_async(trans, 1);
-		if (err)
-			err = btrfs_commit_transaction(trans);
-	} else {
-		err = btrfs_commit_transaction(trans);
-	}
+	err = btrfs_commit_transaction(trans);
 	if (err && !ret)
 		ret = err;
 
@@ -747,8 +740,7 @@ fail_free:
 }
 
 static int create_snapshot(struct btrfs_root *root, struct inode *dir,
-			   struct dentry *dentry,
-			   u64 *async_transid, bool readonly,
+			   struct dentry *dentry, bool readonly,
 			   struct btrfs_qgroup_inherit *inherit)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
@@ -784,11 +776,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	 * possible. This is to avoid later writeback (running dealloc) to
 	 * fallback to COW mode and unexpectedly fail with ENOSPC.
 	 */
-	atomic_inc(&root->will_be_snapshotted);
-	smp_mb__after_atomic();
-	/* wait for no snapshot writes */
-	wait_event(root->subv_writers->wait,
-		   percpu_counter_sum(&root->subv_writers->counter) == 0);
+	btrfs_drew_read_lock(&root->snapshot_lock);
 
 	ret = btrfs_start_delalloc_snapshot(root);
 	if (ret)
@@ -836,14 +824,8 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	list_add(&pending_snapshot->list,
 		 &trans->transaction->pending_snapshots);
 	spin_unlock(&fs_info->trans_lock);
-	if (async_transid) {
-		*async_transid = trans->transid;
-		ret = btrfs_commit_transaction_async(trans, 1);
-		if (ret)
-			ret = btrfs_commit_transaction(trans);
-	} else {
-		ret = btrfs_commit_transaction(trans);
-	}
+
+	ret = btrfs_commit_transaction(trans);
 	if (ret)
 		goto fail;
 
@@ -864,12 +846,13 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	d_instantiate(dentry, inode);
 	ret = 0;
 fail:
+	btrfs_put_root(pending_snapshot->snap);
 	btrfs_subvolume_release_metadata(fs_info, &pending_snapshot->block_rsv);
 dec_and_free:
 	if (snapshot_force_cow)
 		atomic_dec(&root->snapshot_force_cow);
-	if (atomic_dec_and_test(&root->will_be_snapshotted))
-		wake_up_var(&root->will_be_snapshotted);
+	btrfs_drew_read_unlock(&root->snapshot_lock);
+
 free_pending:
 	kfree(pending_snapshot->root_item);
 	btrfs_free_path(pending_snapshot->path);
@@ -948,7 +931,7 @@ static inline int btrfs_may_create(struct inode *dir, struct dentry *child)
 static noinline int btrfs_mksubvol(const struct path *parent,
 				   const char *name, int namelen,
 				   struct btrfs_root *snap_src,
-				   u64 *async_transid, bool readonly,
+				   bool readonly,
 				   struct btrfs_qgroup_inherit *inherit)
 {
 	struct inode *dir = d_inode(parent->dentry);
@@ -984,13 +967,11 @@ static noinline int btrfs_mksubvol(const struct path *parent,
 	if (btrfs_root_refs(&BTRFS_I(dir)->root->root_item) == 0)
 		goto out_up_read;
 
-	if (snap_src) {
-		error = create_snapshot(snap_src, dir, dentry,
-					async_transid, readonly, inherit);
-	} else {
-		error = create_subvol(dir, dentry, name, namelen,
-				      async_transid, inherit);
-	}
+	if (snap_src)
+		error = create_snapshot(snap_src, dir, dentry, readonly, inherit);
+	else
+		error = create_subvol(dir, dentry, name, namelen, inherit);
+
 	if (!error)
 		fsnotify_mkdir(dir, dentry);
 out_up_read:
@@ -1123,7 +1104,7 @@ static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start)
 
 		/* get the big lock and read metadata off disk */
 		lock_extent_bits(io_tree, start, end, &cached);
-		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len, 0);
+		em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len);
 		unlock_extent_cached(io_tree, start, end, &cached);
 
 		if (IS_ERR(em))
@@ -1409,7 +1390,7 @@ int btrfs_defrag_file(struct inode *inode, struct file *file,
 		return -EINVAL;
 
 	if (do_compress) {
-		if (range->compress_type > BTRFS_COMPRESS_TYPES)
+		if (range->compress_type >= BTRFS_NR_COMPRESS_TYPES)
 			return -EINVAL;
 		if (range->compress_type)
 			compress_type = range->compress_type;
@@ -1706,9 +1687,6 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 
 	new_size = round_down(new_size, fs_info->sectorsize);
 
-	btrfs_info_in_rcu(fs_info, "new size for %s is %llu",
-			  rcu_str_deref(device->name), new_size);
-
 	if (new_size > old_size) {
 		trans = btrfs_start_transaction(root, 0);
 		if (IS_ERR(trans)) {
@@ -1721,6 +1699,11 @@ static noinline int btrfs_ioctl_resize(struct file *file,
 		ret = btrfs_shrink_device(device, new_size);
 	} /* equal, nothing need to do */
 
+	if (ret == 0 && new_size != old_size)
+		btrfs_info_in_rcu(fs_info,
+			"resize device %s (devid %llu) from %llu to %llu",
+			rcu_str_deref(device->name), device->devid,
+			old_size, new_size);
 out_free:
 	kfree(vol_args);
 out:
@@ -1729,9 +1712,9 @@ out:
 	return ret;
 }
 
-static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
+static noinline int __btrfs_ioctl_snap_create(struct file *file,
 				const char *name, unsigned long fd, int subvol,
-				u64 *transid, bool readonly,
+				bool readonly,
 				struct btrfs_qgroup_inherit *inherit)
 {
 	int namelen;
@@ -1758,7 +1741,7 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
 
 	if (subvol) {
 		ret = btrfs_mksubvol(&file->f_path, name, namelen,
-				     NULL, transid, readonly, inherit);
+				     NULL, readonly, inherit);
 	} else {
 		struct fd src = fdget(fd);
 		struct inode *src_inode;
@@ -1781,7 +1764,7 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
 		} else {
 			ret = btrfs_mksubvol(&file->f_path, name, namelen,
 					     BTRFS_I(src_inode)->root,
-					     transid, readonly, inherit);
+					     readonly, inherit);
 		}
 		fdput(src);
 	}
@@ -1805,9 +1788,8 @@ static noinline int btrfs_ioctl_snap_create(struct file *file,
 		return PTR_ERR(vol_args);
 	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
 
-	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
-					      vol_args->fd, subvol,
-					      NULL, false, NULL);
+	ret = __btrfs_ioctl_snap_create(file, vol_args->name, vol_args->fd,
+					subvol, false, NULL);
 
 	kfree(vol_args);
 	return ret;
@@ -1818,8 +1800,6 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 {
 	struct btrfs_ioctl_vol_args_v2 *vol_args;
 	int ret;
-	u64 transid = 0;
-	u64 *ptr = NULL;
 	bool readonly = false;
 	struct btrfs_qgroup_inherit *inherit = NULL;
 
@@ -1831,22 +1811,11 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 		return PTR_ERR(vol_args);
 	vol_args->name[BTRFS_SUBVOL_NAME_MAX] = '\0';
 
-	if (vol_args->flags &
-	    ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY |
-	      BTRFS_SUBVOL_QGROUP_INHERIT)) {
+	if (vol_args->flags & ~BTRFS_SUBVOL_CREATE_ARGS_MASK) {
 		ret = -EOPNOTSUPP;
 		goto free_args;
 	}
 
-	if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC) {
-		struct inode *inode = file_inode(file);
-		struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-
-		btrfs_warn(fs_info,
-"SNAP_CREATE_V2 ioctl with CREATE_ASYNC is deprecated and will be removed in kernel 5.7");
-
-		ptr = &transid;
-	}
 	if (vol_args->flags & BTRFS_SUBVOL_RDONLY)
 		readonly = true;
 	if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) {
@@ -1861,18 +1830,10 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file,
 		}
 	}
 
-	ret = btrfs_ioctl_snap_create_transid(file, vol_args->name,
-					      vol_args->fd, subvol, ptr,
-					      readonly, inherit);
+	ret = __btrfs_ioctl_snap_create(file, vol_args->name, vol_args->fd,
+					subvol, readonly, inherit);
 	if (ret)
 		goto free_inherit;
-
-	if (ptr && copy_to_user(arg +
-				offsetof(struct btrfs_ioctl_vol_args_v2,
-					transid),
-				ptr, sizeof(*ptr)))
-		ret = -EFAULT;
-
 free_inherit:
 	kfree(inherit);
 free_args:
@@ -1931,11 +1892,6 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file,
 		goto out_drop_write;
 	}
 
-	if (flags & BTRFS_SUBVOL_CREATE_ASYNC) {
-		ret = -EINVAL;
-		goto out_drop_write;
-	}
-
 	if (flags & ~BTRFS_SUBVOL_RDONLY) {
 		ret = -EOPNOTSUPP;
 		goto out_drop_write;
@@ -2169,12 +2125,12 @@ static noinline int search_ioctl(struct inode *inode,
 
 	if (sk->tree_id == 0) {
 		/* search the root of the inode that was passed */
-		root = BTRFS_I(inode)->root;
+		root = btrfs_grab_root(BTRFS_I(inode)->root);
 	} else {
 		key.objectid = sk->tree_id;
 		key.type = BTRFS_ROOT_ITEM_KEY;
 		key.offset = (u64)-1;
-		root = btrfs_read_fs_root_no_name(info, &key);
+		root = btrfs_get_fs_root(info, &key, true);
 		if (IS_ERR(root)) {
 			btrfs_free_path(path);
 			return PTR_ERR(root);
@@ -2203,6 +2159,7 @@ static noinline int search_ioctl(struct inode *inode,
 		ret = 0;
 err:
 	sk->nr_items = num_found;
+	btrfs_put_root(root);
 	btrfs_free_path(path);
 	return ret;
 }
@@ -2309,9 +2266,10 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 	key.objectid = tree_id;
 	key.type = BTRFS_ROOT_ITEM_KEY;
 	key.offset = (u64)-1;
-	root = btrfs_read_fs_root_no_name(info, &key);
+	root = btrfs_get_fs_root(info, &key, true);
 	if (IS_ERR(root)) {
 		ret = PTR_ERR(root);
+		root = NULL;
 		goto out;
 	}
 
@@ -2362,6 +2320,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 	name[total_len] = '\0';
 	ret = 0;
 out:
+	btrfs_put_root(root);
 	btrfs_free_path(path);
 	return ret;
 }
@@ -2378,7 +2337,7 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 	unsigned long item_len;
 	struct btrfs_inode_ref *iref;
 	struct btrfs_root_ref *rref;
-	struct btrfs_root *root;
+	struct btrfs_root *root = NULL;
 	struct btrfs_path *path;
 	struct btrfs_key key, key2;
 	struct extent_buffer *leaf;
@@ -2403,7 +2362,7 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 		key.objectid = treeid;
 		key.type = BTRFS_ROOT_ITEM_KEY;
 		key.offset = (u64)-1;
-		root = btrfs_read_fs_root_no_name(fs_info, &key);
+		root = btrfs_get_fs_root(fs_info, &key, true);
 		if (IS_ERR(root)) {
 			ret = PTR_ERR(root);
 			goto out;
@@ -2415,15 +2374,15 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 		while (1) {
 			ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 			if (ret < 0) {
-				goto out;
+				goto out_put;
 			} else if (ret > 0) {
 				ret = btrfs_previous_item(root, path, dirid,
 							  BTRFS_INODE_REF_KEY);
 				if (ret < 0) {
-					goto out;
+					goto out_put;
 				} else if (ret > 0) {
 					ret = -ENOENT;
-					goto out;
+					goto out_put;
 				}
 			}
 
@@ -2437,7 +2396,7 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 			total_len += len + 1;
 			if (ptr < args->path) {
 				ret = -ENAMETOOLONG;
-				goto out;
+				goto out_put;
 			}
 
 			*(ptr + len) = '/';
@@ -2448,10 +2407,10 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 			ret = btrfs_previous_item(root, path, dirid,
 						  BTRFS_INODE_ITEM_KEY);
 			if (ret < 0) {
-				goto out;
+				goto out_put;
 			} else if (ret > 0) {
 				ret = -ENOENT;
-				goto out;
+				goto out_put;
 			}
 
 			leaf = path->nodes[0];
@@ -2459,26 +2418,26 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 			btrfs_item_key_to_cpu(leaf, &key2, slot);
 			if (key2.objectid != dirid) {
 				ret = -ENOENT;
-				goto out;
+				goto out_put;
 			}
 
-			temp_inode = btrfs_iget(sb, &key2, root, NULL);
+			temp_inode = btrfs_iget(sb, &key2, root);
 			if (IS_ERR(temp_inode)) {
 				ret = PTR_ERR(temp_inode);
-				goto out;
+				goto out_put;
 			}
 			ret = inode_permission(temp_inode, MAY_READ | MAY_EXEC);
 			iput(temp_inode);
 			if (ret) {
 				ret = -EACCES;
-				goto out;
+				goto out_put;
 			}
 
 			if (key.offset == upper_limit.objectid)
 				break;
 			if (key.objectid == BTRFS_FIRST_FREE_OBJECTID) {
 				ret = -EACCES;
-				goto out;
+				goto out_put;
 			}
 
 			btrfs_release_path(path);
@@ -2489,15 +2448,16 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 
 		memmove(args->path, ptr, total_len);
 		args->path[total_len] = '\0';
+		btrfs_put_root(root);
+		root = NULL;
 		btrfs_release_path(path);
 	}
 
 	/* Get the bottom subvolume's name from ROOT_REF */
-	root = fs_info->tree_root;
 	key.objectid = treeid;
 	key.type = BTRFS_ROOT_REF_KEY;
 	key.offset = args->treeid;
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0);
 	if (ret < 0) {
 		goto out;
 	} else if (ret > 0) {
@@ -2524,6 +2484,8 @@ static int btrfs_search_path_in_tree_user(struct inode *inode,
 	read_extent_buffer(leaf, args->name, item_off, item_len);
 	args->name[item_len] = 0;
 
+out_put:
+	btrfs_put_root(root);
 out:
 	btrfs_free_path(path);
 	return ret;
@@ -2648,10 +2610,10 @@ static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp)
 	key.objectid = BTRFS_I(inode)->root->root_key.objectid;
 	key.type = BTRFS_ROOT_ITEM_KEY;
 	key.offset = (u64)-1;
-	root = btrfs_read_fs_root_no_name(fs_info, &key);
+	root = btrfs_get_fs_root(fs_info, &key, true);
 	if (IS_ERR(root)) {
 		ret = PTR_ERR(root);
-		goto out;
+		goto out_free;
 	}
 	root_item = &root->root_item;
 
@@ -2684,16 +2646,14 @@ static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp)
 
 	if (key.objectid != BTRFS_FS_TREE_OBJECTID) {
 		/* Search root tree for ROOT_BACKREF of this subvolume */
-		root = fs_info->tree_root;
-
 		key.type = BTRFS_ROOT_BACKREF_KEY;
 		key.offset = 0;
-		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+		ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0);
 		if (ret < 0) {
 			goto out;
 		} else if (path->slots[0] >=
 			   btrfs_header_nritems(path->nodes[0])) {
-			ret = btrfs_next_leaf(root, path);
+			ret = btrfs_next_leaf(fs_info->tree_root, path);
 			if (ret < 0) {
 				goto out;
 			} else if (ret > 0) {
@@ -2728,6 +2688,8 @@ static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp)
 		ret = -EFAULT;
 
 out:
+	btrfs_put_root(root);
+out_free:
 	btrfs_free_path(path);
 	kzfree(subvol_info);
 	return ret;
@@ -2831,7 +2793,8 @@ out:
 }
 
 static noinline int btrfs_ioctl_snap_destroy(struct file *file,
-					     void __user *arg)
+					     void __user *arg,
+					     bool destroy_v2)
 {
 	struct dentry *parent = file->f_path.dentry;
 	struct btrfs_fs_info *fs_info = btrfs_sb(parent->d_sb);
@@ -2840,34 +2803,120 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 	struct inode *inode;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_root *dest = NULL;
-	struct btrfs_ioctl_vol_args *vol_args;
-	int namelen;
+	struct btrfs_ioctl_vol_args *vol_args = NULL;
+	struct btrfs_ioctl_vol_args_v2 *vol_args2 = NULL;
+	char *subvol_name, *subvol_name_ptr = NULL;
+	int subvol_namelen;
 	int err = 0;
+	bool destroy_parent = false;
 
-	if (!S_ISDIR(dir->i_mode))
-		return -ENOTDIR;
+	if (destroy_v2) {
+		vol_args2 = memdup_user(arg, sizeof(*vol_args2));
+		if (IS_ERR(vol_args2))
+			return PTR_ERR(vol_args2);
 
-	vol_args = memdup_user(arg, sizeof(*vol_args));
-	if (IS_ERR(vol_args))
-		return PTR_ERR(vol_args);
+		if (vol_args2->flags & ~BTRFS_SUBVOL_DELETE_ARGS_MASK) {
+			err = -EOPNOTSUPP;
+			goto out;
+		}
 
-	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
-	namelen = strlen(vol_args->name);
-	if (strchr(vol_args->name, '/') ||
-	    strncmp(vol_args->name, "..", namelen) == 0) {
-		err = -EINVAL;
-		goto out;
+		/*
+		 * If SPEC_BY_ID is not set, we are looking for the subvolume by
+		 * name, same as v1 currently does.
+		 */
+		if (!(vol_args2->flags & BTRFS_SUBVOL_SPEC_BY_ID)) {
+			vol_args2->name[BTRFS_SUBVOL_NAME_MAX] = 0;
+			subvol_name = vol_args2->name;
+
+			err = mnt_want_write_file(file);
+			if (err)
+				goto out;
+		} else {
+			if (vol_args2->subvolid < BTRFS_FIRST_FREE_OBJECTID) {
+				err = -EINVAL;
+				goto out;
+			}
+
+			err = mnt_want_write_file(file);
+			if (err)
+				goto out;
+
+			dentry = btrfs_get_dentry(fs_info->sb,
+					BTRFS_FIRST_FREE_OBJECTID,
+					vol_args2->subvolid, 0, 0);
+			if (IS_ERR(dentry)) {
+				err = PTR_ERR(dentry);
+				goto out_drop_write;
+			}
+
+			/*
+			 * Change the default parent since the subvolume being
+			 * deleted can be outside of the current mount point.
+			 */
+			parent = btrfs_get_parent(dentry);
+
+			/*
+			 * At this point dentry->d_name can point to '/' if the
+			 * subvolume we want to destroy is outsite of the
+			 * current mount point, so we need to release the
+			 * current dentry and execute the lookup to return a new
+			 * one with ->d_name pointing to the
+			 * <mount point>/subvol_name.
+			 */
+			dput(dentry);
+			if (IS_ERR(parent)) {
+				err = PTR_ERR(parent);
+				goto out_drop_write;
+			}
+			dir = d_inode(parent);
+
+			/*
+			 * If v2 was used with SPEC_BY_ID, a new parent was
+			 * allocated since the subvolume can be outside of the
+			 * current mount point. Later on we need to release this
+			 * new parent dentry.
+			 */
+			destroy_parent = true;
+
+			subvol_name_ptr = btrfs_get_subvol_name_from_objectid(
+						fs_info, vol_args2->subvolid);
+			if (IS_ERR(subvol_name_ptr)) {
+				err = PTR_ERR(subvol_name_ptr);
+				goto free_parent;
+			}
+			/* subvol_name_ptr is already NULL termined */
+			subvol_name = (char *)kbasename(subvol_name_ptr);
+		}
+	} else {
+		vol_args = memdup_user(arg, sizeof(*vol_args));
+		if (IS_ERR(vol_args))
+			return PTR_ERR(vol_args);
+
+		vol_args->name[BTRFS_PATH_NAME_MAX] = 0;
+		subvol_name = vol_args->name;
+
+		err = mnt_want_write_file(file);
+		if (err)
+			goto out;
 	}
 
-	err = mnt_want_write_file(file);
-	if (err)
-		goto out;
+	subvol_namelen = strlen(subvol_name);
 
+	if (strchr(subvol_name, '/') ||
+	    strncmp(subvol_name, "..", subvol_namelen) == 0) {
+		err = -EINVAL;
+		goto free_subvol_name;
+	}
+
+	if (!S_ISDIR(dir->i_mode)) {
+		err = -ENOTDIR;
+		goto free_subvol_name;
+	}
 
 	err = down_write_killable_nested(&dir->i_rwsem, I_MUTEX_PARENT);
 	if (err == -EINTR)
-		goto out_drop_write;
-	dentry = lookup_one_len(vol_args->name, parent, namelen);
+		goto free_subvol_name;
+	dentry = lookup_one_len(subvol_name, parent, subvol_namelen);
 	if (IS_ERR(dentry)) {
 		err = PTR_ERR(dentry);
 		goto out_unlock_dir;
@@ -2936,9 +2985,15 @@ out_dput:
 	dput(dentry);
 out_unlock_dir:
 	inode_unlock(dir);
+free_subvol_name:
+	kfree(subvol_name_ptr);
+free_parent:
+	if (destroy_parent)
+		dput(parent);
 out_drop_write:
 	mnt_drop_write_file(file);
 out:
+	kfree(vol_args2);
 	kfree(vol_args);
 	return err;
 }
@@ -3064,8 +3119,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 		goto err_drop;
 	}
 
-	/* Check for compatibility reject unknown flags */
-	if (vol_args->flags & ~BTRFS_VOL_ARG_V2_FLAGS_SUPPORTED) {
+	if (vol_args->flags & ~BTRFS_DEVICE_REMOVE_ARGS_MASK) {
 		ret = -EOPNOTSUPP;
 		goto out;
 	}
@@ -3215,738 +3269,6 @@ out:
 	return ret;
 }
 
-static void btrfs_double_extent_unlock(struct inode *inode1, u64 loff1,
-				       struct inode *inode2, u64 loff2, u64 len)
-{
-	unlock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1);
-	unlock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1);
-}
-
-static void btrfs_double_extent_lock(struct inode *inode1, u64 loff1,
-				     struct inode *inode2, u64 loff2, u64 len)
-{
-	if (inode1 < inode2) {
-		swap(inode1, inode2);
-		swap(loff1, loff2);
-	} else if (inode1 == inode2 && loff2 < loff1) {
-		swap(loff1, loff2);
-	}
-	lock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1);
-	lock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1);
-}
-
-static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 len,
-				   struct inode *dst, u64 dst_loff)
-{
-	int ret;
-
-	/*
-	 * Lock destination range to serialize with concurrent readpages() and
-	 * source range to serialize with relocation.
-	 */
-	btrfs_double_extent_lock(src, loff, dst, dst_loff, len);
-	ret = btrfs_clone(src, dst, loff, len, len, dst_loff, 1);
-	btrfs_double_extent_unlock(src, loff, dst, dst_loff, len);
-
-	return ret;
-}
-
-#define BTRFS_MAX_DEDUPE_LEN	SZ_16M
-
-static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
-			     struct inode *dst, u64 dst_loff)
-{
-	int ret;
-	u64 i, tail_len, chunk_count;
-	struct btrfs_root *root_dst = BTRFS_I(dst)->root;
-
-	spin_lock(&root_dst->root_item_lock);
-	if (root_dst->send_in_progress) {
-		btrfs_warn_rl(root_dst->fs_info,
-"cannot deduplicate to root %llu while send operations are using it (%d in progress)",
-			      root_dst->root_key.objectid,
-			      root_dst->send_in_progress);
-		spin_unlock(&root_dst->root_item_lock);
-		return -EAGAIN;
-	}
-	root_dst->dedupe_in_progress++;
-	spin_unlock(&root_dst->root_item_lock);
-
-	tail_len = olen % BTRFS_MAX_DEDUPE_LEN;
-	chunk_count = div_u64(olen, BTRFS_MAX_DEDUPE_LEN);
-
-	for (i = 0; i < chunk_count; i++) {
-		ret = btrfs_extent_same_range(src, loff, BTRFS_MAX_DEDUPE_LEN,
-					      dst, dst_loff);
-		if (ret)
-			goto out;
-
-		loff += BTRFS_MAX_DEDUPE_LEN;
-		dst_loff += BTRFS_MAX_DEDUPE_LEN;
-	}
-
-	if (tail_len > 0)
-		ret = btrfs_extent_same_range(src, loff, tail_len, dst,
-					      dst_loff);
-out:
-	spin_lock(&root_dst->root_item_lock);
-	root_dst->dedupe_in_progress--;
-	spin_unlock(&root_dst->root_item_lock);
-
-	return ret;
-}
-
-static int clone_finish_inode_update(struct btrfs_trans_handle *trans,
-				     struct inode *inode,
-				     u64 endoff,
-				     const u64 destoff,
-				     const u64 olen,
-				     int no_time_update)
-{
-	struct btrfs_root *root = BTRFS_I(inode)->root;
-	int ret;
-
-	inode_inc_iversion(inode);
-	if (!no_time_update)
-		inode->i_mtime = inode->i_ctime = current_time(inode);
-	/*
-	 * We round up to the block size at eof when determining which
-	 * extents to clone above, but shouldn't round up the file size.
-	 */
-	if (endoff > destoff + olen)
-		endoff = destoff + olen;
-	if (endoff > inode->i_size)
-		btrfs_i_size_write(BTRFS_I(inode), endoff);
-
-	ret = btrfs_update_inode(trans, root, inode);
-	if (ret) {
-		btrfs_abort_transaction(trans, ret);
-		btrfs_end_transaction(trans);
-		goto out;
-	}
-	ret = btrfs_end_transaction(trans);
-out:
-	return ret;
-}
-
-/*
- * Make sure we do not end up inserting an inline extent into a file that has
- * already other (non-inline) extents. If a file has an inline extent it can
- * not have any other extents and the (single) inline extent must start at the
- * file offset 0. Failing to respect these rules will lead to file corruption,
- * resulting in EIO errors on read/write operations, hitting BUG_ON's in mm, etc
- *
- * We can have extents that have been already written to disk or we can have
- * dirty ranges still in delalloc, in which case the extent maps and items are
- * created only when we run delalloc, and the delalloc ranges might fall outside
- * the range we are currently locking in the inode's io tree. So we check the
- * inode's i_size because of that (i_size updates are done while holding the
- * i_mutex, which we are holding here).
- * We also check to see if the inode has a size not greater than "datal" but has
- * extents beyond it, due to an fallocate with FALLOC_FL_KEEP_SIZE (and we are
- * protected against such concurrent fallocate calls by the i_mutex).
- *
- * If the file has no extents but a size greater than datal, do not allow the
- * copy because we would need turn the inline extent into a non-inline one (even
- * with NO_HOLES enabled). If we find our destination inode only has one inline
- * extent, just overwrite it with the source inline extent if its size is less
- * than the source extent's size, or we could copy the source inline extent's
- * data into the destination inode's inline extent if the later is greater then
- * the former.
- */
-static int clone_copy_inline_extent(struct inode *dst,
-				    struct btrfs_trans_handle *trans,
-				    struct btrfs_path *path,
-				    struct btrfs_key *new_key,
-				    const u64 drop_start,
-				    const u64 datal,
-				    const u64 skip,
-				    const u64 size,
-				    char *inline_data)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(dst->i_sb);
-	struct btrfs_root *root = BTRFS_I(dst)->root;
-	const u64 aligned_end = ALIGN(new_key->offset + datal,
-				      fs_info->sectorsize);
-	int ret;
-	struct btrfs_key key;
-
-	if (new_key->offset > 0)
-		return -EOPNOTSUPP;
-
-	key.objectid = btrfs_ino(BTRFS_I(dst));
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = 0;
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0) {
-		return ret;
-	} else if (ret > 0) {
-		if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				return ret;
-			else if (ret > 0)
-				goto copy_inline_extent;
-		}
-		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
-		if (key.objectid == btrfs_ino(BTRFS_I(dst)) &&
-		    key.type == BTRFS_EXTENT_DATA_KEY) {
-			ASSERT(key.offset > 0);
-			return -EOPNOTSUPP;
-		}
-	} else if (i_size_read(dst) <= datal) {
-		struct btrfs_file_extent_item *ei;
-		u64 ext_len;
-
-		/*
-		 * If the file size is <= datal, make sure there are no other
-		 * extents following (can happen do to an fallocate call with
-		 * the flag FALLOC_FL_KEEP_SIZE).
-		 */
-		ei = btrfs_item_ptr(path->nodes[0], path->slots[0],
-				    struct btrfs_file_extent_item);
-		/*
-		 * If it's an inline extent, it can not have other extents
-		 * following it.
-		 */
-		if (btrfs_file_extent_type(path->nodes[0], ei) ==
-		    BTRFS_FILE_EXTENT_INLINE)
-			goto copy_inline_extent;
-
-		ext_len = btrfs_file_extent_num_bytes(path->nodes[0], ei);
-		if (ext_len > aligned_end)
-			return -EOPNOTSUPP;
-
-		ret = btrfs_next_item(root, path);
-		if (ret < 0) {
-			return ret;
-		} else if (ret == 0) {
-			btrfs_item_key_to_cpu(path->nodes[0], &key,
-					      path->slots[0]);
-			if (key.objectid == btrfs_ino(BTRFS_I(dst)) &&
-			    key.type == BTRFS_EXTENT_DATA_KEY)
-				return -EOPNOTSUPP;
-		}
-	}
-
-copy_inline_extent:
-	/*
-	 * We have no extent items, or we have an extent at offset 0 which may
-	 * or may not be inlined. All these cases are dealt the same way.
-	 */
-	if (i_size_read(dst) > datal) {
-		/*
-		 * If the destination inode has an inline extent...
-		 * This would require copying the data from the source inline
-		 * extent into the beginning of the destination's inline extent.
-		 * But this is really complex, both extents can be compressed
-		 * or just one of them, which would require decompressing and
-		 * re-compressing data (which could increase the new compressed
-		 * size, not allowing the compressed data to fit anymore in an
-		 * inline extent).
-		 * So just don't support this case for now (it should be rare,
-		 * we are not really saving space when cloning inline extents).
-		 */
-		return -EOPNOTSUPP;
-	}
-
-	btrfs_release_path(path);
-	ret = btrfs_drop_extents(trans, root, dst, drop_start, aligned_end, 1);
-	if (ret)
-		return ret;
-	ret = btrfs_insert_empty_item(trans, root, path, new_key, size);
-	if (ret)
-		return ret;
-
-	if (skip) {
-		const u32 start = btrfs_file_extent_calc_inline_size(0);
-
-		memmove(inline_data + start, inline_data + start + skip, datal);
-	}
-
-	write_extent_buffer(path->nodes[0], inline_data,
-			    btrfs_item_ptr_offset(path->nodes[0],
-						  path->slots[0]),
-			    size);
-	inode_add_bytes(dst, datal);
-	set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(dst)->runtime_flags);
-
-	return 0;
-}
-
-/**
- * btrfs_clone() - clone a range from inode file to another
- *
- * @src: Inode to clone from
- * @inode: Inode to clone to
- * @off: Offset within source to start clone from
- * @olen: Original length, passed by user, of range to clone
- * @olen_aligned: Block-aligned value of olen
- * @destoff: Offset within @inode to start clone
- * @no_time_update: Whether to update mtime/ctime on the target inode
- */
-static int btrfs_clone(struct inode *src, struct inode *inode,
-		       const u64 off, const u64 olen, const u64 olen_aligned,
-		       const u64 destoff, int no_time_update)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_path *path = NULL;
-	struct extent_buffer *leaf;
-	struct btrfs_trans_handle *trans;
-	char *buf = NULL;
-	struct btrfs_key key;
-	u32 nritems;
-	int slot;
-	int ret;
-	const u64 len = olen_aligned;
-	u64 last_dest_end = destoff;
-
-	ret = -ENOMEM;
-	buf = kvmalloc(fs_info->nodesize, GFP_KERNEL);
-	if (!buf)
-		return ret;
-
-	path = btrfs_alloc_path();
-	if (!path) {
-		kvfree(buf);
-		return ret;
-	}
-
-	path->reada = READA_FORWARD;
-	/* clone data */
-	key.objectid = btrfs_ino(BTRFS_I(src));
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = off;
-
-	while (1) {
-		u64 next_key_min_offset = key.offset + 1;
-		struct btrfs_file_extent_item *extent;
-		int type;
-		u32 size;
-		struct btrfs_key new_key;
-		u64 disko = 0, diskl = 0;
-		u64 datao = 0, datal = 0;
-		u8 comp;
-		u64 drop_start;
-
-		/*
-		 * note the key will change type as we walk through the
-		 * tree.
-		 */
-		path->leave_spinning = 1;
-		ret = btrfs_search_slot(NULL, BTRFS_I(src)->root, &key, path,
-				0, 0);
-		if (ret < 0)
-			goto out;
-		/*
-		 * First search, if no extent item that starts at offset off was
-		 * found but the previous item is an extent item, it's possible
-		 * it might overlap our target range, therefore process it.
-		 */
-		if (key.offset == off && ret > 0 && path->slots[0] > 0) {
-			btrfs_item_key_to_cpu(path->nodes[0], &key,
-					      path->slots[0] - 1);
-			if (key.type == BTRFS_EXTENT_DATA_KEY)
-				path->slots[0]--;
-		}
-
-		nritems = btrfs_header_nritems(path->nodes[0]);
-process_slot:
-		if (path->slots[0] >= nritems) {
-			ret = btrfs_next_leaf(BTRFS_I(src)->root, path);
-			if (ret < 0)
-				goto out;
-			if (ret > 0)
-				break;
-			nritems = btrfs_header_nritems(path->nodes[0]);
-		}
-		leaf = path->nodes[0];
-		slot = path->slots[0];
-
-		btrfs_item_key_to_cpu(leaf, &key, slot);
-		if (key.type > BTRFS_EXTENT_DATA_KEY ||
-		    key.objectid != btrfs_ino(BTRFS_I(src)))
-			break;
-
-		ASSERT(key.type == BTRFS_EXTENT_DATA_KEY);
-
-		extent = btrfs_item_ptr(leaf, slot,
-					struct btrfs_file_extent_item);
-		comp = btrfs_file_extent_compression(leaf, extent);
-		type = btrfs_file_extent_type(leaf, extent);
-		if (type == BTRFS_FILE_EXTENT_REG ||
-		    type == BTRFS_FILE_EXTENT_PREALLOC) {
-			disko = btrfs_file_extent_disk_bytenr(leaf, extent);
-			diskl = btrfs_file_extent_disk_num_bytes(leaf, extent);
-			datao = btrfs_file_extent_offset(leaf, extent);
-			datal = btrfs_file_extent_num_bytes(leaf, extent);
-		} else if (type == BTRFS_FILE_EXTENT_INLINE) {
-			/* Take upper bound, may be compressed */
-			datal = btrfs_file_extent_ram_bytes(leaf, extent);
-		}
-
-		/*
-		 * The first search might have left us at an extent item that
-		 * ends before our target range's start, can happen if we have
-		 * holes and NO_HOLES feature enabled.
-		 */
-		if (key.offset + datal <= off) {
-			path->slots[0]++;
-			goto process_slot;
-		} else if (key.offset >= off + len) {
-			break;
-		}
-		next_key_min_offset = key.offset + datal;
-		size = btrfs_item_size_nr(leaf, slot);
-		read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf, slot),
-				   size);
-
-		btrfs_release_path(path);
-		path->leave_spinning = 0;
-
-		memcpy(&new_key, &key, sizeof(new_key));
-		new_key.objectid = btrfs_ino(BTRFS_I(inode));
-		if (off <= key.offset)
-			new_key.offset = key.offset + destoff - off;
-		else
-			new_key.offset = destoff;
-
-		/*
-		 * Deal with a hole that doesn't have an extent item that
-		 * represents it (NO_HOLES feature enabled).
-		 * This hole is either in the middle of the cloning range or at
-		 * the beginning (fully overlaps it or partially overlaps it).
-		 */
-		if (new_key.offset != last_dest_end)
-			drop_start = last_dest_end;
-		else
-			drop_start = new_key.offset;
-
-		if (type == BTRFS_FILE_EXTENT_REG ||
-		    type == BTRFS_FILE_EXTENT_PREALLOC) {
-			struct btrfs_clone_extent_info clone_info;
-
-			/*
-			 *    a  | --- range to clone ---|  b
-			 * | ------------- extent ------------- |
-			 */
-
-			/* Subtract range b */
-			if (key.offset + datal > off + len)
-				datal = off + len - key.offset;
-
-			/* Subtract range a */
-			if (off > key.offset) {
-				datao += off - key.offset;
-				datal -= off - key.offset;
-			}
-
-			clone_info.disk_offset = disko;
-			clone_info.disk_len = diskl;
-			clone_info.data_offset = datao;
-			clone_info.data_len = datal;
-			clone_info.file_offset = new_key.offset;
-			clone_info.extent_buf = buf;
-			clone_info.item_size = size;
-			ret = btrfs_punch_hole_range(inode, path,
-						     drop_start,
-						     new_key.offset + datal - 1,
-						     &clone_info, &trans);
-			if (ret)
-				goto out;
-		} else if (type == BTRFS_FILE_EXTENT_INLINE) {
-			u64 skip = 0;
-			u64 trim = 0;
-
-			if (off > key.offset) {
-				skip = off - key.offset;
-				new_key.offset += skip;
-			}
-
-			if (key.offset + datal > off + len)
-				trim = key.offset + datal - (off + len);
-
-			if (comp && (skip || trim)) {
-				ret = -EINVAL;
-				goto out;
-			}
-			size -= skip + trim;
-			datal -= skip + trim;
-
-			/*
-			 * If our extent is inline, we know we will drop or
-			 * adjust at most 1 extent item in the destination root.
-			 *
-			 * 1 - adjusting old extent (we may have to split it)
-			 * 1 - add new extent
-			 * 1 - inode update
-			 */
-			trans = btrfs_start_transaction(root, 3);
-			if (IS_ERR(trans)) {
-				ret = PTR_ERR(trans);
-				goto out;
-			}
-
-			ret = clone_copy_inline_extent(inode, trans, path,
-						       &new_key, drop_start,
-						       datal, skip, size, buf);
-			if (ret) {
-				if (ret != -EOPNOTSUPP)
-					btrfs_abort_transaction(trans, ret);
-				btrfs_end_transaction(trans);
-				goto out;
-			}
-		}
-
-		btrfs_release_path(path);
-
-		last_dest_end = ALIGN(new_key.offset + datal,
-				      fs_info->sectorsize);
-		ret = clone_finish_inode_update(trans, inode, last_dest_end,
-						destoff, olen, no_time_update);
-		if (ret)
-			goto out;
-		if (new_key.offset + datal >= destoff + len)
-			break;
-
-		btrfs_release_path(path);
-		key.offset = next_key_min_offset;
-
-		if (fatal_signal_pending(current)) {
-			ret = -EINTR;
-			goto out;
-		}
-	}
-	ret = 0;
-
-	if (last_dest_end < destoff + len) {
-		struct btrfs_clone_extent_info clone_info = { 0 };
-		/*
-		 * We have an implicit hole (NO_HOLES feature is enabled) that
-		 * fully or partially overlaps our cloning range at its end.
-		 */
-		btrfs_release_path(path);
-		path->leave_spinning = 0;
-
-		/*
-		 * We are dealing with a hole and our clone_info already has a
-		 * disk_offset of 0, we only need to fill the data length and
-		 * file offset.
-		 */
-		clone_info.data_len = destoff + len - last_dest_end;
-		clone_info.file_offset = last_dest_end;
-		ret = btrfs_punch_hole_range(inode, path,
-					     last_dest_end, destoff + len - 1,
-					     &clone_info, &trans);
-		if (ret)
-			goto out;
-
-		ret = clone_finish_inode_update(trans, inode, destoff + len,
-						destoff, olen, no_time_update);
-	}
-
-out:
-	btrfs_free_path(path);
-	kvfree(buf);
-	return ret;
-}
-
-static noinline int btrfs_clone_files(struct file *file, struct file *file_src,
-					u64 off, u64 olen, u64 destoff)
-{
-	struct inode *inode = file_inode(file);
-	struct inode *src = file_inode(file_src);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	int ret;
-	u64 len = olen;
-	u64 bs = fs_info->sb->s_blocksize;
-
-	/*
-	 * TODO:
-	 * - split compressed inline extents.  annoying: we need to
-	 *   decompress into destination's address_space (the file offset
-	 *   may change, so source mapping won't do), then recompress (or
-	 *   otherwise reinsert) a subrange.
-	 *
-	 * - split destination inode's inline extents.  The inline extents can
-	 *   be either compressed or non-compressed.
-	 */
-
-	/*
-	 * VFS's generic_remap_file_range_prep() protects us from cloning the
-	 * eof block into the middle of a file, which would result in corruption
-	 * if the file size is not blocksize aligned. So we don't need to check
-	 * for that case here.
-	 */
-	if (off + len == src->i_size)
-		len = ALIGN(src->i_size, bs) - off;
-
-	if (destoff > inode->i_size) {
-		const u64 wb_start = ALIGN_DOWN(inode->i_size, bs);
-
-		ret = btrfs_cont_expand(inode, inode->i_size, destoff);
-		if (ret)
-			return ret;
-		/*
-		 * We may have truncated the last block if the inode's size is
-		 * not sector size aligned, so we need to wait for writeback to
-		 * complete before proceeding further, otherwise we can race
-		 * with cloning and attempt to increment a reference to an
-		 * extent that no longer exists (writeback completed right after
-		 * we found the previous extent covering eof and before we
-		 * attempted to increment its reference count).
-		 */
-		ret = btrfs_wait_ordered_range(inode, wb_start,
-					       destoff - wb_start);
-		if (ret)
-			return ret;
-	}
-
-	/*
-	 * Lock destination range to serialize with concurrent readpages() and
-	 * source range to serialize with relocation.
-	 */
-	btrfs_double_extent_lock(src, off, inode, destoff, len);
-	ret = btrfs_clone(src, inode, off, olen, len, destoff, 0);
-	btrfs_double_extent_unlock(src, off, inode, destoff, len);
-	/*
-	 * Truncate page cache pages so that future reads will see the cloned
-	 * data immediately and not the previous data.
-	 */
-	truncate_inode_pages_range(&inode->i_data,
-				round_down(destoff, PAGE_SIZE),
-				round_up(destoff + len, PAGE_SIZE) - 1);
-
-	return ret;
-}
-
-static int btrfs_remap_file_range_prep(struct file *file_in, loff_t pos_in,
-				       struct file *file_out, loff_t pos_out,
-				       loff_t *len, unsigned int remap_flags)
-{
-	struct inode *inode_in = file_inode(file_in);
-	struct inode *inode_out = file_inode(file_out);
-	u64 bs = BTRFS_I(inode_out)->root->fs_info->sb->s_blocksize;
-	bool same_inode = inode_out == inode_in;
-	u64 wb_len;
-	int ret;
-
-	if (!(remap_flags & REMAP_FILE_DEDUP)) {
-		struct btrfs_root *root_out = BTRFS_I(inode_out)->root;
-
-		if (btrfs_root_readonly(root_out))
-			return -EROFS;
-
-		if (file_in->f_path.mnt != file_out->f_path.mnt ||
-		    inode_in->i_sb != inode_out->i_sb)
-			return -EXDEV;
-	}
-
-	/* don't make the dst file partly checksummed */
-	if ((BTRFS_I(inode_in)->flags & BTRFS_INODE_NODATASUM) !=
-	    (BTRFS_I(inode_out)->flags & BTRFS_INODE_NODATASUM)) {
-		return -EINVAL;
-	}
-
-	/*
-	 * Now that the inodes are locked, we need to start writeback ourselves
-	 * and can not rely on the writeback from the VFS's generic helper
-	 * generic_remap_file_range_prep() because:
-	 *
-	 * 1) For compression we must call filemap_fdatawrite_range() range
-	 *    twice (btrfs_fdatawrite_range() does it for us), and the generic
-	 *    helper only calls it once;
-	 *
-	 * 2) filemap_fdatawrite_range(), called by the generic helper only
-	 *    waits for the writeback to complete, i.e. for IO to be done, and
-	 *    not for the ordered extents to complete. We need to wait for them
-	 *    to complete so that new file extent items are in the fs tree.
-	 */
-	if (*len == 0 && !(remap_flags & REMAP_FILE_DEDUP))
-		wb_len = ALIGN(inode_in->i_size, bs) - ALIGN_DOWN(pos_in, bs);
-	else
-		wb_len = ALIGN(*len, bs);
-
-	/*
-	 * Since we don't lock ranges, wait for ongoing lockless dio writes (as
-	 * any in progress could create its ordered extents after we wait for
-	 * existing ordered extents below).
-	 */
-	inode_dio_wait(inode_in);
-	if (!same_inode)
-		inode_dio_wait(inode_out);
-
-	/*
-	 * Workaround to make sure NOCOW buffered write reach disk as NOCOW.
-	 *
-	 * Btrfs' back references do not have a block level granularity, they
-	 * work at the whole extent level.
-	 * NOCOW buffered write without data space reserved may not be able
-	 * to fall back to CoW due to lack of data space, thus could cause
-	 * data loss.
-	 *
-	 * Here we take a shortcut by flushing the whole inode, so that all
-	 * nocow write should reach disk as nocow before we increase the
-	 * reference of the extent. We could do better by only flushing NOCOW
-	 * data, but that needs extra accounting.
-	 *
-	 * Also we don't need to check ASYNC_EXTENT, as async extent will be
-	 * CoWed anyway, not affecting nocow part.
-	 */
-	ret = filemap_flush(inode_in->i_mapping);
-	if (ret < 0)
-		return ret;
-
-	ret = btrfs_wait_ordered_range(inode_in, ALIGN_DOWN(pos_in, bs),
-				       wb_len);
-	if (ret < 0)
-		return ret;
-	ret = btrfs_wait_ordered_range(inode_out, ALIGN_DOWN(pos_out, bs),
-				       wb_len);
-	if (ret < 0)
-		return ret;
-
-	return generic_remap_file_range_prep(file_in, pos_in, file_out, pos_out,
-					    len, remap_flags);
-}
-
-loff_t btrfs_remap_file_range(struct file *src_file, loff_t off,
-		struct file *dst_file, loff_t destoff, loff_t len,
-		unsigned int remap_flags)
-{
-	struct inode *src_inode = file_inode(src_file);
-	struct inode *dst_inode = file_inode(dst_file);
-	bool same_inode = dst_inode == src_inode;
-	int ret;
-
-	if (remap_flags & ~(REMAP_FILE_DEDUP | REMAP_FILE_ADVISORY))
-		return -EINVAL;
-
-	if (same_inode)
-		inode_lock(src_inode);
-	else
-		lock_two_nondirectories(src_inode, dst_inode);
-
-	ret = btrfs_remap_file_range_prep(src_file, off, dst_file, destoff,
-					  &len, remap_flags);
-	if (ret < 0 || len == 0)
-		goto out_unlock;
-
-	if (remap_flags & REMAP_FILE_DEDUP)
-		ret = btrfs_extent_same(src_inode, off, len, dst_inode, destoff);
-	else
-		ret = btrfs_clone_files(dst_file, src_file, off, len, destoff);
-
-out_unlock:
-	if (same_inode)
-		inode_unlock(src_inode);
-	else
-		unlock_two_nondirectories(src_inode, dst_inode);
-
-	return ret < 0 ? ret : len;
-}
-
 static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
 {
 	struct inode *inode = file_inode(file);
@@ -3955,7 +3277,7 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
 	struct btrfs_root *new_root;
 	struct btrfs_dir_item *di;
 	struct btrfs_trans_handle *trans;
-	struct btrfs_path *path;
+	struct btrfs_path *path = NULL;
 	struct btrfs_key location;
 	struct btrfs_disk_key disk_key;
 	u64 objectid = 0;
@@ -3981,49 +3303,51 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp)
 	location.type = BTRFS_ROOT_ITEM_KEY;
 	location.offset = (u64)-1;
 
-	new_root = btrfs_read_fs_root_no_name(fs_info, &location);
+	new_root = btrfs_get_fs_root(fs_info, &location, true);
 	if (IS_ERR(new_root)) {
 		ret = PTR_ERR(new_root);
 		goto out;
 	}
 	if (!is_fstree(new_root->root_key.objectid)) {
 		ret = -ENOENT;
-		goto out;
+		goto out_free;
 	}
 
 	path = btrfs_alloc_path();
 	if (!path) {
 		ret = -ENOMEM;
-		goto out;
+		goto out_free;
 	}
 	path->leave_spinning = 1;
 
 	trans = btrfs_start_transaction(root, 1);
 	if (IS_ERR(trans)) {
-		btrfs_free_path(path);
 		ret = PTR_ERR(trans);
-		goto out;
+		goto out_free;
 	}
 
 	dir_id = btrfs_super_root_dir(fs_info->super_copy);
 	di = btrfs_lookup_dir_item(trans, fs_info->tree_root, path,
 				   dir_id, "default", 7, 1);
 	if (IS_ERR_OR_NULL(di)) {
-		btrfs_free_path(path);
+		btrfs_release_path(path);
 		btrfs_end_transaction(trans);
 		btrfs_err(fs_info,
 			  "Umm, you don't have the default diritem, this isn't going to work");
 		ret = -ENOENT;
-		goto out;
+		goto out_free;
 	}
 
 	btrfs_cpu_key_to_disk(&disk_key, &new_root->root_key);
 	btrfs_set_dir_item_key(path->nodes[0], di, &disk_key);
 	btrfs_mark_buffer_dirty(path->nodes[0]);
-	btrfs_free_path(path);
+	btrfs_release_path(path);
 
 	btrfs_set_fs_incompat(fs_info, DEFAULT_SUBVOL);
 	btrfs_end_transaction(trans);
+out_free:
+	btrfs_put_root(new_root);
+	btrfs_free_path(path);
 out:
 	mnt_drop_write_file(file);
 	return ret;
@@ -4032,16 +3356,15 @@ out:
 static void get_block_group_info(struct list_head *groups_list,
 				 struct btrfs_ioctl_space_info *space)
 {
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 
 	space->total_bytes = 0;
 	space->used_bytes = 0;
 	space->flags = 0;
 	list_for_each_entry(block_group, groups_list, list) {
 		space->flags = block_group->flags;
-		space->total_bytes += block_group->key.offset;
-		space->used_bytes +=
-			btrfs_block_group_used(&block_group->item);
+		space->total_bytes += block_group->length;
+		space->used_bytes += block_group->used;
 	}
 }
 
@@ -4254,7 +3577,19 @@ static long btrfs_ioctl_scrub(struct file *file, void __user *arg)
 			      &sa->progress, sa->flags & BTRFS_SCRUB_READONLY,
 			      0);
 
-	if (ret == 0 && copy_to_user(arg, sa, sizeof(*sa)))
+	/*
+	 * Copy scrub args to user space even if btrfs_scrub_dev() returned an
+	 * error. This is important as it allows user space to know how much
+	 * progress scrub has done. For example, if scrub is canceled we get
+	 * -ECANCELED from btrfs_scrub_dev() and return that error back to user
+	 * space. Later user space can inspect the progress from the structure
+	 * btrfs_ioctl_scrub_args and resume scrub from where it left off
+	 * previously (btrfs-progs does this).
+	 * If we fail to copy the btrfs_ioctl_scrub_args structure to user space
+	 * then return -EFAULT to signal the structure was not copied or it may
+	 * be corrupt and unreliable due to a partial copy.
+	 */
+	if (copy_to_user(arg, sa, sizeof(*sa)))
 		ret = -EFAULT;
 
 	if (!(sa->flags & BTRFS_SCRUB_READONLY))
@@ -4952,10 +4287,9 @@ drop_write:
 	return ret;
 }
 
-static long btrfs_ioctl_quota_rescan_status(struct file *file, void __user *arg)
+static long btrfs_ioctl_quota_rescan_status(struct btrfs_fs_info *fs_info,
+						void __user *arg)
 {
-	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ioctl_quota_rescan_args *qsa;
 	int ret = 0;
 
@@ -4978,11 +4312,9 @@ static long btrfs_ioctl_quota_rescan_status(struct file *file, void __user *arg)
 	return ret;
 }
 
-static long btrfs_ioctl_quota_rescan_wait(struct file *file, void __user *arg)
+static long btrfs_ioctl_quota_rescan_wait(struct btrfs_fs_info *fs_info,
+						void __user *arg)
 {
-	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
@@ -5154,10 +4486,9 @@ out:
 	return ret;
 }
 
-static int btrfs_ioctl_get_fslabel(struct file *file, void __user *arg)
+static int btrfs_ioctl_get_fslabel(struct btrfs_fs_info *fs_info,
+					void __user *arg)
 {
-	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	size_t len;
 	int ret;
 	char label[BTRFS_LABEL_SIZE];
@@ -5241,10 +4572,9 @@ int btrfs_ioctl_get_supported_features(void __user *arg)
 	return 0;
 }
 
-static int btrfs_ioctl_get_features(struct file *file, void __user *arg)
+static int btrfs_ioctl_get_features(struct btrfs_fs_info *fs_info,
+					void __user *arg)
 {
-	struct inode *inode = file_inode(file);
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_super_block *super_block = fs_info->super_copy;
 	struct btrfs_ioctl_feature_flags features;
 
@@ -5445,11 +4775,11 @@ long btrfs_ioctl(struct file *file, unsigned int
 	case FS_IOC_GETVERSION:
 		return btrfs_ioctl_getversion(file, argp);
 	case FS_IOC_GETFSLABEL:
-		return btrfs_ioctl_get_fslabel(file, argp);
+		return btrfs_ioctl_get_fslabel(fs_info, argp);
 	case FS_IOC_SETFSLABEL:
 		return btrfs_ioctl_set_fslabel(file, argp);
 	case FITRIM:
-		return btrfs_ioctl_fitrim(file, argp);
+		return btrfs_ioctl_fitrim(fs_info, argp);
 	case BTRFS_IOC_SNAP_CREATE:
 		return btrfs_ioctl_snap_create(file, argp, 0);
 	case BTRFS_IOC_SNAP_CREATE_V2:
@@ -5459,7 +4789,9 @@ long btrfs_ioctl(struct file *file, unsigned int
 	case BTRFS_IOC_SUBVOL_CREATE_V2:
 		return btrfs_ioctl_snap_create_v2(file, argp, 1);
 	case BTRFS_IOC_SNAP_DESTROY:
-		return btrfs_ioctl_snap_destroy(file, argp);
+		return btrfs_ioctl_snap_destroy(file, argp, false);
+	case BTRFS_IOC_SNAP_DESTROY_V2:
+		return btrfs_ioctl_snap_destroy(file, argp, true);
 	case BTRFS_IOC_SUBVOL_GETFLAGS:
 		return btrfs_ioctl_subvol_getflags(file, argp);
 	case BTRFS_IOC_SUBVOL_SETFLAGS:
@@ -5554,15 +4886,15 @@ long btrfs_ioctl(struct file *file, unsigned int
 	case BTRFS_IOC_QUOTA_RESCAN:
 		return btrfs_ioctl_quota_rescan(file, argp);
 	case BTRFS_IOC_QUOTA_RESCAN_STATUS:
-		return btrfs_ioctl_quota_rescan_status(file, argp);
+		return btrfs_ioctl_quota_rescan_status(fs_info, argp);
 	case BTRFS_IOC_QUOTA_RESCAN_WAIT:
-		return btrfs_ioctl_quota_rescan_wait(file, argp);
+		return btrfs_ioctl_quota_rescan_wait(fs_info, argp);
 	case BTRFS_IOC_DEV_REPLACE:
 		return btrfs_ioctl_dev_replace(fs_info, argp);
 	case BTRFS_IOC_GET_SUPPORTED_FEATURES:
 		return btrfs_ioctl_get_supported_features(argp);
 	case BTRFS_IOC_GET_FEATURES:
-		return btrfs_ioctl_get_features(file, argp);
+		return btrfs_ioctl_get_features(fs_info, argp);
 	case BTRFS_IOC_SET_FEATURES:
 		return btrfs_ioctl_set_features(file, argp);
 	case FS_IOC_FSGETXATTR:
diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c
index 7f9a578a1a20..fb647d8cf527 100644
--- a/fs/btrfs/locking.c
+++ b/fs/btrfs/locking.c
@@ -13,65 +13,164 @@
 #include "extent_io.h"
 #include "locking.h"
 
+/*
+ * Extent buffer locking
+ * =====================
+ *
+ * The locks use a custom scheme that allows to do more operations than are
+ * available fromt current locking primitives. The building blocks are still
+ * rwlock and wait queues.
+ *
+ * Required semantics:
+ *
+ * - reader/writer exclusion
+ * - writer/writer exclusion
+ * - reader/reader sharing
+ * - spinning lock semantics
+ * - blocking lock semantics
+ * - try-lock semantics for readers and writers
+ * - one level nesting, allowing read lock to be taken by the same thread that
+ *   already has write lock
+ *
+ * The extent buffer locks (also called tree locks) manage access to eb data
+ * related to the storage in the b-tree (keys, items, but not the individual
+ * members of eb).
+ * We want concurrency of many readers and safe updates. The underlying locking
+ * is done by read-write spinlock and the blocking part is implemented using
+ * counters and wait queues.
+ *
+ * spinning semantics - the low-level rwlock is held so all other threads that
+ *                      want to take it are spinning on it.
+ *
+ * blocking semantics - the low-level rwlock is not held but the counter
+ *                      denotes how many times the blocking lock was held;
+ *                      sleeping is possible
+ *
+ * Write lock always allows only one thread to access the data.
+ *
+ *
+ * Debugging
+ * ---------
+ *
+ * There are additional state counters that are asserted in various contexts,
+ * removed from non-debug build to reduce extent_buffer size and for
+ * performance reasons.
+ *
+ *
+ * Lock nesting
+ * ------------
+ *
+ * A write operation on a tree might indirectly start a look up on the same
+ * tree.  This can happen when btrfs_cow_block locks the tree and needs to
+ * lookup free extents.
+ *
+ * btrfs_cow_block
+ *   ..
+ *   alloc_tree_block_no_bg_flush
+ *     btrfs_alloc_tree_block
+ *       btrfs_reserve_extent
+ *         ..
+ *         load_free_space_cache
+ *           ..
+ *           btrfs_lookup_file_extent
+ *             btrfs_search_slot
+ *
+ *
+ * Locking pattern - spinning
+ * --------------------------
+ *
+ * The simple locking scenario, the +--+ denotes the spinning section.
+ *
+ * +- btrfs_tree_lock
+ * | - extent_buffer::rwlock is held
+ * | - no heavy operations should happen, eg. IO, memory allocations, large
+ * |   structure traversals
+ * +- btrfs_tree_unock
+*
+*
+ * Locking pattern - blocking
+ * --------------------------
+ *
+ * The blocking write uses the following scheme.  The +--+ denotes the spinning
+ * section.
+ *
+ * +- btrfs_tree_lock
+ * |
+ * +- btrfs_set_lock_blocking_write
+ *
+ *   - allowed: IO, memory allocations, etc.
+ *
+ * -- btrfs_tree_unlock - note, no explicit unblocking necessary
+ *
+ *
+ * Blocking read is similar.
+ *
+ * +- btrfs_tree_read_lock
+ * |
+ * +- btrfs_set_lock_blocking_read
+ *
+ *  - heavy operations allowed
+ *
+ * +- btrfs_tree_read_unlock_blocking
+ * |
+ * +- btrfs_tree_read_unlock
+ *
+ */
+
 #ifdef CONFIG_BTRFS_DEBUG
-static void btrfs_assert_spinning_writers_get(struct extent_buffer *eb)
+static inline void btrfs_assert_spinning_writers_get(struct extent_buffer *eb)
 {
 	WARN_ON(eb->spinning_writers);
 	eb->spinning_writers++;
 }
 
-static void btrfs_assert_spinning_writers_put(struct extent_buffer *eb)
+static inline void btrfs_assert_spinning_writers_put(struct extent_buffer *eb)
 {
 	WARN_ON(eb->spinning_writers != 1);
 	eb->spinning_writers--;
 }
 
-static void btrfs_assert_no_spinning_writers(struct extent_buffer *eb)
+static inline void btrfs_assert_no_spinning_writers(struct extent_buffer *eb)
 {
 	WARN_ON(eb->spinning_writers);
 }
 
-static void btrfs_assert_spinning_readers_get(struct extent_buffer *eb)
+static inline void btrfs_assert_spinning_readers_get(struct extent_buffer *eb)
 {
 	atomic_inc(&eb->spinning_readers);
 }
 
-static void btrfs_assert_spinning_readers_put(struct extent_buffer *eb)
+static inline void btrfs_assert_spinning_readers_put(struct extent_buffer *eb)
 {
 	WARN_ON(atomic_read(&eb->spinning_readers) == 0);
 	atomic_dec(&eb->spinning_readers);
 }
 
-static void btrfs_assert_tree_read_locks_get(struct extent_buffer *eb)
+static inline void btrfs_assert_tree_read_locks_get(struct extent_buffer *eb)
 {
 	atomic_inc(&eb->read_locks);
 }
 
-static void btrfs_assert_tree_read_locks_put(struct extent_buffer *eb)
+static inline void btrfs_assert_tree_read_locks_put(struct extent_buffer *eb)
 {
 	atomic_dec(&eb->read_locks);
 }
 
-static void btrfs_assert_tree_read_locked(struct extent_buffer *eb)
+static inline void btrfs_assert_tree_read_locked(struct extent_buffer *eb)
 {
 	BUG_ON(!atomic_read(&eb->read_locks));
 }
 
-static void btrfs_assert_tree_write_locks_get(struct extent_buffer *eb)
+static inline void btrfs_assert_tree_write_locks_get(struct extent_buffer *eb)
 {
 	eb->write_locks++;
 }
 
-static void btrfs_assert_tree_write_locks_put(struct extent_buffer *eb)
+static inline void btrfs_assert_tree_write_locks_put(struct extent_buffer *eb)
 {
 	eb->write_locks--;
 }
 
-void btrfs_assert_tree_locked(struct extent_buffer *eb)
-{
-	BUG_ON(!eb->write_locks);
-}
-
 #else
 static void btrfs_assert_spinning_writers_get(struct extent_buffer *eb) { }
 static void btrfs_assert_spinning_writers_put(struct extent_buffer *eb) { }
@@ -81,11 +180,19 @@ static void btrfs_assert_spinning_readers_get(struct extent_buffer *eb) { }
 static void btrfs_assert_tree_read_locked(struct extent_buffer *eb) { }
 static void btrfs_assert_tree_read_locks_get(struct extent_buffer *eb) { }
 static void btrfs_assert_tree_read_locks_put(struct extent_buffer *eb) { }
-void btrfs_assert_tree_locked(struct extent_buffer *eb) { }
 static void btrfs_assert_tree_write_locks_get(struct extent_buffer *eb) { }
 static void btrfs_assert_tree_write_locks_put(struct extent_buffer *eb) { }
 #endif
 
+/*
+ * Mark already held read lock as blocking. Can be nested in write lock by the
+ * same thread.
+ *
+ * Use when there are potentially long operations ahead so other thread waiting
+ * on the lock will not actively spin but sleep instead.
+ *
+ * The rwlock is released and blocking reader counter is increased.
+ */
 void btrfs_set_lock_blocking_read(struct extent_buffer *eb)
 {
 	trace_btrfs_set_lock_blocking_read(eb);
@@ -102,6 +209,14 @@ void btrfs_set_lock_blocking_read(struct extent_buffer *eb)
 	read_unlock(&eb->lock);
 }
 
+/*
+ * Mark already held write lock as blocking.
+ *
+ * Use when there are potentially long operations ahead so other threads
+ * waiting on the lock will not actively spin but sleep instead.
+ *
+ * The rwlock is released and blocking writers is set.
+ */
 void btrfs_set_lock_blocking_write(struct extent_buffer *eb)
 {
 	trace_btrfs_set_lock_blocking_write(eb);
@@ -115,14 +230,19 @@ void btrfs_set_lock_blocking_write(struct extent_buffer *eb)
 	if (eb->blocking_writers == 0) {
 		btrfs_assert_spinning_writers_put(eb);
 		btrfs_assert_tree_locked(eb);
-		eb->blocking_writers++;
+		WRITE_ONCE(eb->blocking_writers, 1);
 		write_unlock(&eb->lock);
 	}
 }
 
 /*
- * take a spinning read lock.  This will wait for any blocking
- * writers
+ * Lock the extent buffer for read. Wait for any writers (spinning or blocking).
+ * Can be nested in write lock by the same thread.
+ *
+ * Use when the locked section does only lightweight actions and busy waiting
+ * would be cheaper than making other threads do the wait/wake loop.
+ *
+ * The rwlock is held upon exit.
  */
 void btrfs_tree_read_lock(struct extent_buffer *eb)
 {
@@ -134,23 +254,24 @@ again:
 	read_lock(&eb->lock);
 	BUG_ON(eb->blocking_writers == 0 &&
 	       current->pid == eb->lock_owner);
-	if (eb->blocking_writers && current->pid == eb->lock_owner) {
-		/*
-		 * This extent is already write-locked by our thread. We allow
-		 * an additional read lock to be added because it's for the same
-		 * thread. btrfs_find_all_roots() depends on this as it may be
-		 * called on a partly (write-)locked tree.
-		 */
-		BUG_ON(eb->lock_nested);
-		eb->lock_nested = true;
-		read_unlock(&eb->lock);
-		trace_btrfs_tree_read_lock(eb, start_ns);
-		return;
-	}
 	if (eb->blocking_writers) {
+		if (current->pid == eb->lock_owner) {
+			/*
+			 * This extent is already write-locked by our thread.
+			 * We allow an additional read lock to be added because
+			 * it's for the same thread. btrfs_find_all_roots()
+			 * depends on this as it may be called on a partly
+			 * (write-)locked tree.
+			 */
+			BUG_ON(eb->lock_nested);
+			eb->lock_nested = true;
+			read_unlock(&eb->lock);
+			trace_btrfs_tree_read_lock(eb, start_ns);
+			return;
+		}
 		read_unlock(&eb->lock);
 		wait_event(eb->write_lock_wq,
-			   eb->blocking_writers == 0);
+			   READ_ONCE(eb->blocking_writers) == 0);
 		goto again;
 	}
 	btrfs_assert_tree_read_locks_get(eb);
@@ -159,17 +280,19 @@ again:
 }
 
 /*
- * take a spinning read lock.
- * returns 1 if we get the read lock and 0 if we don't
- * this won't wait for blocking writers
+ * Lock extent buffer for read, optimistically expecting that there are no
+ * contending blocking writers. If there are, don't wait.
+ *
+ * Return 1 if the rwlock has been taken, 0 otherwise
  */
 int btrfs_tree_read_lock_atomic(struct extent_buffer *eb)
 {
-	if (eb->blocking_writers)
+	if (READ_ONCE(eb->blocking_writers))
 		return 0;
 
 	read_lock(&eb->lock);
-	if (eb->blocking_writers) {
+	/* Refetch value after lock */
+	if (READ_ONCE(eb->blocking_writers)) {
 		read_unlock(&eb->lock);
 		return 0;
 	}
@@ -180,18 +303,20 @@ int btrfs_tree_read_lock_atomic(struct extent_buffer *eb)
 }
 
 /*
- * returns 1 if we get the read lock and 0 if we don't
- * this won't wait for blocking writers
+ * Try-lock for read. Don't block or wait for contending writers.
+ *
+ * Retrun 1 if the rwlock has been taken, 0 otherwise
  */
 int btrfs_try_tree_read_lock(struct extent_buffer *eb)
 {
-	if (eb->blocking_writers)
+	if (READ_ONCE(eb->blocking_writers))
 		return 0;
 
 	if (!read_trylock(&eb->lock))
 		return 0;
 
-	if (eb->blocking_writers) {
+	/* Refetch value after lock */
+	if (READ_ONCE(eb->blocking_writers)) {
 		read_unlock(&eb->lock);
 		return 0;
 	}
@@ -202,16 +327,19 @@ int btrfs_try_tree_read_lock(struct extent_buffer *eb)
 }
 
 /*
- * returns 1 if we get the read lock and 0 if we don't
- * this won't wait for blocking writers or readers
+ * Try-lock for write. May block until the lock is uncontended, but does not
+ * wait until it is free.
+ *
+ * Retrun 1 if the rwlock has been taken, 0 otherwise
  */
 int btrfs_try_tree_write_lock(struct extent_buffer *eb)
 {
-	if (eb->blocking_writers || atomic_read(&eb->blocking_readers))
+	if (READ_ONCE(eb->blocking_writers) || atomic_read(&eb->blocking_readers))
 		return 0;
 
 	write_lock(&eb->lock);
-	if (eb->blocking_writers || atomic_read(&eb->blocking_readers)) {
+	/* Refetch value after lock */
+	if (READ_ONCE(eb->blocking_writers) || atomic_read(&eb->blocking_readers)) {
 		write_unlock(&eb->lock);
 		return 0;
 	}
@@ -223,7 +351,10 @@ int btrfs_try_tree_write_lock(struct extent_buffer *eb)
 }
 
 /*
- * drop a spinning read lock
+ * Release read lock. Must be used only if the lock is in spinning mode.  If
+ * the read lock is nested, must pair with read lock before the write unlock.
+ *
+ * The rwlock is not held upon exit.
  */
 void btrfs_tree_read_unlock(struct extent_buffer *eb)
 {
@@ -245,7 +376,11 @@ void btrfs_tree_read_unlock(struct extent_buffer *eb)
 }
 
 /*
- * drop a blocking read lock
+ * Release read lock, previously set to blocking by a pairing call to
+ * btrfs_set_lock_blocking_read(). Can be nested in write lock by the same
+ * thread.
+ *
+ * State of rwlock is unchanged, last reader wakes waiting threads.
  */
 void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb)
 {
@@ -269,8 +404,10 @@ void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb)
 }
 
 /*
- * take a spinning write lock.  This will wait for both
- * blocking readers or writers
+ * Lock for write. Wait for all blocking and spinning readers and writers. This
+ * starts context where reader lock could be nested by the same thread.
+ *
+ * The rwlock is held for write upon exit.
  */
 void btrfs_tree_lock(struct extent_buffer *eb)
 {
@@ -282,9 +419,11 @@ void btrfs_tree_lock(struct extent_buffer *eb)
 	WARN_ON(eb->lock_owner == current->pid);
 again:
 	wait_event(eb->read_lock_wq, atomic_read(&eb->blocking_readers) == 0);
-	wait_event(eb->write_lock_wq, eb->blocking_writers == 0);
+	wait_event(eb->write_lock_wq, READ_ONCE(eb->blocking_writers) == 0);
 	write_lock(&eb->lock);
-	if (atomic_read(&eb->blocking_readers) || eb->blocking_writers) {
+	/* Refetch value after lock */
+	if (atomic_read(&eb->blocking_readers) ||
+	    READ_ONCE(eb->blocking_writers)) {
 		write_unlock(&eb->lock);
 		goto again;
 	}
@@ -295,10 +434,19 @@ again:
 }
 
 /*
- * drop a spinning or a blocking write lock.
+ * Release the write lock, either blocking or spinning (ie. there's no need
+ * for an explicit blocking unlock, like btrfs_tree_read_unlock_blocking).
+ * This also ends the context for nesting, the read lock must have been
+ * released already.
+ *
+ * Tasks blocked and waiting are woken, rwlock is not held upon exit.
  */
 void btrfs_tree_unlock(struct extent_buffer *eb)
 {
+	/*
+	 * This is read both locked and unlocked but always by the same thread
+	 * that already owns the lock so we don't need to use READ_ONCE
+	 */
 	int blockers = eb->blocking_writers;
 
 	BUG_ON(blockers > 1);
@@ -310,7 +458,8 @@ void btrfs_tree_unlock(struct extent_buffer *eb)
 
 	if (blockers) {
 		btrfs_assert_no_spinning_writers(eb);
-		eb->blocking_writers--;
+		/* Unlocked write */
+		WRITE_ONCE(eb->blocking_writers, 0);
 		/*
 		 * We need to order modifying blocking_writers above with
 		 * actually waking up the sleepers to ensure they see the
@@ -322,3 +471,190 @@ void btrfs_tree_unlock(struct extent_buffer *eb)
 		write_unlock(&eb->lock);
 	}
 }
+
+/*
+ * Set all locked nodes in the path to blocking locks.  This should be done
+ * before scheduling
+ */
+void btrfs_set_path_blocking(struct btrfs_path *p)
+{
+	int i;
+
+	for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
+		if (!p->nodes[i] || !p->locks[i])
+			continue;
+		/*
+		 * If we currently have a spinning reader or writer lock this
+		 * will bump the count of blocking holders and drop the
+		 * spinlock.
+		 */
+		if (p->locks[i] == BTRFS_READ_LOCK) {
+			btrfs_set_lock_blocking_read(p->nodes[i]);
+			p->locks[i] = BTRFS_READ_LOCK_BLOCKING;
+		} else if (p->locks[i] == BTRFS_WRITE_LOCK) {
+			btrfs_set_lock_blocking_write(p->nodes[i]);
+			p->locks[i] = BTRFS_WRITE_LOCK_BLOCKING;
+		}
+	}
+}
+
+/*
+ * This releases any locks held in the path starting at level and going all the
+ * way up to the root.
+ *
+ * btrfs_search_slot will keep the lock held on higher nodes in a few corner
+ * cases, such as COW of the block at slot zero in the node.  This ignores
+ * those rules, and it should only be called when there are no more updates to
+ * be done higher up in the tree.
+ */
+void btrfs_unlock_up_safe(struct btrfs_path *path, int level)
+{
+	int i;
+
+	if (path->keep_locks)
+		return;
+
+	for (i = level; i < BTRFS_MAX_LEVEL; i++) {
+		if (!path->nodes[i])
+			continue;
+		if (!path->locks[i])
+			continue;
+		btrfs_tree_unlock_rw(path->nodes[i], path->locks[i]);
+		path->locks[i] = 0;
+	}
+}
+
+/*
+ * Loop around taking references on and locking the root node of the tree until
+ * we end up with a lock on the root node.
+ *
+ * Return: root extent buffer with write lock held
+ */
+struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root)
+{
+	struct extent_buffer *eb;
+
+	while (1) {
+		eb = btrfs_root_node(root);
+		btrfs_tree_lock(eb);
+		if (eb == root->node)
+			break;
+		btrfs_tree_unlock(eb);
+		free_extent_buffer(eb);
+	}
+	return eb;
+}
+
+/*
+ * Loop around taking references on and locking the root node of the tree until
+ * we end up with a lock on the root node.
+ *
+ * Return: root extent buffer with read lock held
+ */
+struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
+{
+	struct extent_buffer *eb;
+
+	while (1) {
+		eb = btrfs_root_node(root);
+		btrfs_tree_read_lock(eb);
+		if (eb == root->node)
+			break;
+		btrfs_tree_read_unlock(eb);
+		free_extent_buffer(eb);
+	}
+	return eb;
+}
+
+/*
+ * DREW locks
+ * ==========
+ *
+ * DREW stands for double-reader-writer-exclusion lock. It's used in situation
+ * where you want to provide A-B exclusion but not AA or BB.
+ *
+ * Currently implementation gives more priority to reader. If a reader and a
+ * writer both race to acquire their respective sides of the lock the writer
+ * would yield its lock as soon as it detects a concurrent reader. Additionally
+ * if there are pending readers no new writers would be allowed to come in and
+ * acquire the lock.
+ */
+
+int btrfs_drew_lock_init(struct btrfs_drew_lock *lock)
+{
+	int ret;
+
+	ret = percpu_counter_init(&lock->writers, 0, GFP_KERNEL);
+	if (ret)
+		return ret;
+
+	atomic_set(&lock->readers, 0);
+	init_waitqueue_head(&lock->pending_readers);
+	init_waitqueue_head(&lock->pending_writers);
+
+	return 0;
+}
+
+void btrfs_drew_lock_destroy(struct btrfs_drew_lock *lock)
+{
+	percpu_counter_destroy(&lock->writers);
+}
+
+/* Return true if acquisition is successful, false otherwise */
+bool btrfs_drew_try_write_lock(struct btrfs_drew_lock *lock)
+{
+	if (atomic_read(&lock->readers))
+		return false;
+
+	percpu_counter_inc(&lock->writers);
+
+	/* Ensure writers count is updated before we check for pending readers */
+	smp_mb();
+	if (atomic_read(&lock->readers)) {
+		btrfs_drew_write_unlock(lock);
+		return false;
+	}
+
+	return true;
+}
+
+void btrfs_drew_write_lock(struct btrfs_drew_lock *lock)
+{
+	while (true) {
+		if (btrfs_drew_try_write_lock(lock))
+			return;
+		wait_event(lock->pending_writers, !atomic_read(&lock->readers));
+	}
+}
+
+void btrfs_drew_write_unlock(struct btrfs_drew_lock *lock)
+{
+	percpu_counter_dec(&lock->writers);
+	cond_wake_up(&lock->pending_readers);
+}
+
+void btrfs_drew_read_lock(struct btrfs_drew_lock *lock)
+{
+	atomic_inc(&lock->readers);
+
+	/*
+	 * Ensure the pending reader count is perceieved BEFORE this reader
+	 * goes to sleep in case of active writers. This guarantees new writers
+	 * won't be allowed and that the current reader will be woken up when
+	 * the last active writer finishes its jobs.
+	 */
+	smp_mb__after_atomic();
+
+	wait_event(lock->pending_readers,
+		   percpu_counter_sum(&lock->writers) == 0);
+}
+
+void btrfs_drew_read_unlock(struct btrfs_drew_lock *lock)
+{
+	/*
+	 * atomic_dec_and_test implies a full barrier, so woken up writers
+	 * are guaranteed to see the decrement
+	 */
+	if (atomic_dec_and_test(&lock->readers))
+		wake_up(&lock->pending_writers);
+}
diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h
index b775a4207ed9..d715846c10b8 100644
--- a/fs/btrfs/locking.h
+++ b/fs/btrfs/locking.h
@@ -6,11 +6,18 @@
 #ifndef BTRFS_LOCKING_H
 #define BTRFS_LOCKING_H
 
+#include <linux/atomic.h>
+#include <linux/wait.h>
+#include <linux/percpu_counter.h>
+#include "extent_io.h"
+
 #define BTRFS_WRITE_LOCK 1
 #define BTRFS_READ_LOCK 2
 #define BTRFS_WRITE_LOCK_BLOCKING 3
 #define BTRFS_READ_LOCK_BLOCKING 4
 
+struct btrfs_path;
+
 void btrfs_tree_lock(struct extent_buffer *eb);
 void btrfs_tree_unlock(struct extent_buffer *eb);
 
@@ -19,11 +26,20 @@ void btrfs_tree_read_unlock(struct extent_buffer *eb);
 void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb);
 void btrfs_set_lock_blocking_read(struct extent_buffer *eb);
 void btrfs_set_lock_blocking_write(struct extent_buffer *eb);
-void btrfs_assert_tree_locked(struct extent_buffer *eb);
 int btrfs_try_tree_read_lock(struct extent_buffer *eb);
 int btrfs_try_tree_write_lock(struct extent_buffer *eb);
 int btrfs_tree_read_lock_atomic(struct extent_buffer *eb);
 
+#ifdef CONFIG_BTRFS_DEBUG
+static inline void btrfs_assert_tree_locked(struct extent_buffer *eb) {
+	BUG_ON(!eb->write_locks);
+}
+#else
+static inline void btrfs_assert_tree_locked(struct extent_buffer *eb) { }
+#endif
+
+void btrfs_set_path_blocking(struct btrfs_path *p);
+void btrfs_unlock_up_safe(struct btrfs_path *path, int level);
 
 static inline void btrfs_tree_unlock_rw(struct extent_buffer *eb, int rw)
 {
@@ -37,4 +53,19 @@ static inline void btrfs_tree_unlock_rw(struct extent_buffer *eb, int rw)
 		BUG();
 }
 
+struct btrfs_drew_lock {
+	atomic_t readers;
+	struct percpu_counter writers;
+	wait_queue_head_t pending_writers;
+	wait_queue_head_t pending_readers;
+};
+
+int btrfs_drew_lock_init(struct btrfs_drew_lock *lock);
+void btrfs_drew_lock_destroy(struct btrfs_drew_lock *lock);
+void btrfs_drew_write_lock(struct btrfs_drew_lock *lock);
+bool btrfs_drew_try_write_lock(struct btrfs_drew_lock *lock);
+void btrfs_drew_write_unlock(struct btrfs_drew_lock *lock);
+void btrfs_drew_read_lock(struct btrfs_drew_lock *lock);
+void btrfs_drew_read_unlock(struct btrfs_drew_lock *lock);
+
 #endif
diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c
index acad4174f68d..aa9cd11f4b78 100644
--- a/fs/btrfs/lzo.c
+++ b/fs/btrfs/lzo.c
@@ -63,27 +63,7 @@ struct workspace {
 
 static struct workspace_manager wsm;
 
-static void lzo_init_workspace_manager(void)
-{
-	btrfs_init_workspace_manager(&wsm, &btrfs_lzo_compress);
-}
-
-static void lzo_cleanup_workspace_manager(void)
-{
-	btrfs_cleanup_workspace_manager(&wsm);
-}
-
-static struct list_head *lzo_get_workspace(unsigned int level)
-{
-	return btrfs_get_workspace(&wsm, level);
-}
-
-static void lzo_put_workspace(struct list_head *ws)
-{
-	btrfs_put_workspace(&wsm, ws);
-}
-
-static void lzo_free_workspace(struct list_head *ws)
+void lzo_free_workspace(struct list_head *ws)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 
@@ -93,7 +73,7 @@ static void lzo_free_workspace(struct list_head *ws)
 	kfree(workspace);
 }
 
-static struct list_head *lzo_alloc_workspace(unsigned int level)
+struct list_head *lzo_alloc_workspace(unsigned int level)
 {
 	struct workspace *workspace;
 
@@ -131,13 +111,9 @@ static inline size_t read_compress_length(const char *buf)
 	return le32_to_cpu(dlen);
 }
 
-static int lzo_compress_pages(struct list_head *ws,
-			      struct address_space *mapping,
-			      u64 start,
-			      struct page **pages,
-			      unsigned long *out_pages,
-			      unsigned long *total_in,
-			      unsigned long *total_out)
+int lzo_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	int ret = 0;
@@ -303,7 +279,7 @@ out:
 	return ret;
 }
 
-static int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
+int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	int ret = 0, ret2;
@@ -444,10 +420,9 @@ done:
 	return ret;
 }
 
-static int lzo_decompress(struct list_head *ws, unsigned char *data_in,
-			  struct page *dest_page,
-			  unsigned long start_byte,
-			  size_t srclen, size_t destlen)
+int lzo_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	size_t in_len;
@@ -508,15 +483,7 @@ out:
 }
 
 const struct btrfs_compress_op btrfs_lzo_compress = {
-	.init_workspace_manager	= lzo_init_workspace_manager,
-	.cleanup_workspace_manager = lzo_cleanup_workspace_manager,
-	.get_workspace		= lzo_get_workspace,
-	.put_workspace		= lzo_put_workspace,
-	.alloc_workspace	= lzo_alloc_workspace,
-	.free_workspace		= lzo_free_workspace,
-	.compress_pages		= lzo_compress_pages,
-	.decompress_bio		= lzo_decompress_bio,
-	.decompress		= lzo_decompress,
+	.workspace_manager	= &wsm,
 	.max_level		= 1,
 	.default_level		= 1,
 };
diff --git a/fs/btrfs/misc.h b/fs/btrfs/misc.h
index 7d564924dfeb..72bab64ecf60 100644
--- a/fs/btrfs/misc.h
+++ b/fs/btrfs/misc.h
@@ -47,4 +47,15 @@ static inline u64 div_factor_fine(u64 num, int factor)
 	return div_u64(num, 100);
 }
 
+/* Copy of is_power_of_two that is 64bit safe */
+static inline bool is_power_of_two_u64(u64 n)
+{
+	return n != 0 && (n & (n - 1)) == 0;
+}
+
+static inline bool has_single_bit_set(u64 n)
+{
+	return is_power_of_two_u64(n);
+}
+
 #endif
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 24b6c72b9a59..e13b3d28c063 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -20,9 +20,9 @@ static struct kmem_cache *btrfs_ordered_extent_cache;
 
 static u64 entry_end(struct btrfs_ordered_extent *entry)
 {
-	if (entry->file_offset + entry->len < entry->file_offset)
+	if (entry->file_offset + entry->num_bytes < entry->file_offset)
 		return (u64)-1;
-	return entry->file_offset + entry->len;
+	return entry->file_offset + entry->num_bytes;
 }
 
 /* returns NULL if the insertion worked, or it returns the node it did find
@@ -52,14 +52,6 @@ static struct rb_node *tree_insert(struct rb_root *root, u64 file_offset,
 	return NULL;
 }
 
-static void ordered_data_tree_panic(struct inode *inode, int errno,
-					       u64 offset)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	btrfs_panic(fs_info, errno,
-		    "Inconsistency in ordered tree at offset %llu", offset);
-}
-
 /*
  * look for a given offset in the tree, and if it can't be found return the
  * first lesser offset
@@ -120,7 +112,7 @@ static struct rb_node *__tree_search(struct rb_root *root, u64 file_offset,
 static int offset_in_entry(struct btrfs_ordered_extent *entry, u64 file_offset)
 {
 	if (file_offset < entry->file_offset ||
-	    entry->file_offset + entry->len <= file_offset)
+	    entry->file_offset + entry->num_bytes <= file_offset)
 		return 0;
 	return 1;
 }
@@ -129,7 +121,7 @@ static int range_overlaps(struct btrfs_ordered_extent *entry, u64 file_offset,
 			  u64 len)
 {
 	if (file_offset + len <= entry->file_offset ||
-	    entry->file_offset + entry->len <= file_offset)
+	    entry->file_offset + entry->num_bytes <= file_offset)
 		return 0;
 	return 1;
 }
@@ -161,19 +153,14 @@ static inline struct rb_node *tree_search(struct btrfs_ordered_inode_tree *tree,
 }
 
 /* allocate and add a new ordered_extent into the per-inode tree.
- * file_offset is the logical offset in the file
- *
- * start is the disk block number of an extent already reserved in the
- * extent allocation tree
- *
- * len is the length of the extent
  *
  * The tree is given a single reference on the ordered extent that was
  * inserted.
  */
 static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
-				      u64 start, u64 len, u64 disk_len,
-				      int type, int dio, int compress_type)
+				      u64 disk_bytenr, u64 num_bytes,
+				      u64 disk_num_bytes, int type, int dio,
+				      int compress_type)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -187,10 +174,10 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 		return -ENOMEM;
 
 	entry->file_offset = file_offset;
-	entry->start = start;
-	entry->len = len;
-	entry->disk_len = disk_len;
-	entry->bytes_left = len;
+	entry->disk_bytenr = disk_bytenr;
+	entry->num_bytes = num_bytes;
+	entry->disk_num_bytes = disk_num_bytes;
+	entry->bytes_left = num_bytes;
 	entry->inode = igrab(inode);
 	entry->compress_type = compress_type;
 	entry->truncated_len = (u64)-1;
@@ -198,7 +185,7 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 		set_bit(type, &entry->flags);
 
 	if (dio) {
-		percpu_counter_add_batch(&fs_info->dio_bytes, len,
+		percpu_counter_add_batch(&fs_info->dio_bytes, num_bytes,
 					 fs_info->delalloc_batch);
 		set_bit(BTRFS_ORDERED_DIRECT, &entry->flags);
 	}
@@ -219,7 +206,9 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 	node = tree_insert(&tree->tree, file_offset,
 			   &entry->rb_node);
 	if (node)
-		ordered_data_tree_panic(inode, -EEXIST, file_offset);
+		btrfs_panic(fs_info, -EEXIST,
+				"inconsistency in ordered tree at offset %llu",
+				file_offset);
 	spin_unlock_irq(&tree->lock);
 
 	spin_lock(&root->ordered_extent_lock);
@@ -247,27 +236,30 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
 }
 
 int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
-			     u64 start, u64 len, u64 disk_len, int type)
+			     u64 disk_bytenr, u64 num_bytes, u64 disk_num_bytes,
+			     int type)
 {
-	return __btrfs_add_ordered_extent(inode, file_offset, start, len,
-					  disk_len, type, 0,
+	return __btrfs_add_ordered_extent(inode, file_offset, disk_bytenr,
+					  num_bytes, disk_num_bytes, type, 0,
 					  BTRFS_COMPRESS_NONE);
 }
 
 int btrfs_add_ordered_extent_dio(struct inode *inode, u64 file_offset,
-				 u64 start, u64 len, u64 disk_len, int type)
+				 u64 disk_bytenr, u64 num_bytes,
+				 u64 disk_num_bytes, int type)
 {
-	return __btrfs_add_ordered_extent(inode, file_offset, start, len,
-					  disk_len, type, 1,
+	return __btrfs_add_ordered_extent(inode, file_offset, disk_bytenr,
+					  num_bytes, disk_num_bytes, type, 1,
 					  BTRFS_COMPRESS_NONE);
 }
 
 int btrfs_add_ordered_extent_compress(struct inode *inode, u64 file_offset,
-				      u64 start, u64 len, u64 disk_len,
-				      int type, int compress_type)
+				      u64 disk_bytenr, u64 num_bytes,
+				      u64 disk_num_bytes, int type,
+				      int compress_type)
 {
-	return __btrfs_add_ordered_extent(inode, file_offset, start, len,
-					  disk_len, type, 0,
+	return __btrfs_add_ordered_extent(inode, file_offset, disk_bytenr,
+					  num_bytes, disk_num_bytes, type, 0,
 					  compress_type);
 }
 
@@ -328,8 +320,8 @@ int btrfs_dec_test_first_ordered_pending(struct inode *inode,
 	}
 
 	dec_start = max(*file_offset, entry->file_offset);
-	dec_end = min(*file_offset + io_size, entry->file_offset +
-		      entry->len);
+	dec_end = min(*file_offset + io_size,
+		      entry->file_offset + entry->num_bytes);
 	*file_offset = dec_end;
 	if (dec_start > dec_end) {
 		btrfs_crit(fs_info, "bad ordering dec_start %llu end %llu",
@@ -471,10 +463,11 @@ void btrfs_remove_ordered_extent(struct inode *inode,
 	btrfs_mod_outstanding_extents(btrfs_inode, -1);
 	spin_unlock(&btrfs_inode->lock);
 	if (root != fs_info->tree_root)
-		btrfs_delalloc_release_metadata(btrfs_inode, entry->len, false);
+		btrfs_delalloc_release_metadata(btrfs_inode, entry->num_bytes,
+						false);
 
 	if (test_bit(BTRFS_ORDERED_DIRECT, &entry->flags))
-		percpu_counter_add_batch(&fs_info->dio_bytes, -entry->len,
+		percpu_counter_add_batch(&fs_info->dio_bytes, -entry->num_bytes,
 					 fs_info->delalloc_batch);
 
 	tree = &btrfs_inode->ordered_tree;
@@ -534,8 +527,8 @@ u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
 		ordered = list_first_entry(&splice, struct btrfs_ordered_extent,
 					   root_extent_list);
 
-		if (range_end <= ordered->start ||
-		    ordered->start + ordered->disk_len <= range_start) {
+		if (range_end <= ordered->disk_bytenr ||
+		    ordered->disk_bytenr + ordered->disk_num_bytes <= range_start) {
 			list_move_tail(&ordered->root_extent_list, &skipped);
 			cond_resched_lock(&root->ordered_extent_lock);
 			continue;
@@ -547,7 +540,6 @@ u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
 		spin_unlock(&root->ordered_extent_lock);
 
 		btrfs_init_work(&ordered->flush_work,
-				btrfs_flush_delalloc_helper,
 				btrfs_run_ordered_extent_work, NULL, NULL);
 		list_add_tail(&ordered->work_list, &works);
 		btrfs_queue_work(fs_info->flush_workers, &ordered->flush_work);
@@ -573,12 +565,11 @@ u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
 	return count;
 }
 
-u64 btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
+void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
 			     const u64 range_start, const u64 range_len)
 {
 	struct btrfs_root *root;
 	struct list_head splice;
-	u64 total_done = 0;
 	u64 done;
 
 	INIT_LIST_HEAD(&splice);
@@ -589,7 +580,7 @@ u64 btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
 	while (!list_empty(&splice) && nr) {
 		root = list_first_entry(&splice, struct btrfs_root,
 					ordered_root);
-		root = btrfs_grab_fs_root(root);
+		root = btrfs_grab_root(root);
 		BUG_ON(!root);
 		list_move_tail(&root->ordered_root,
 			       &fs_info->ordered_roots);
@@ -597,8 +588,7 @@ u64 btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
 
 		done = btrfs_wait_ordered_extents(root, nr,
 						  range_start, range_len);
-		btrfs_put_fs_root(root);
-		total_done += done;
+		btrfs_put_root(root);
 
 		spin_lock(&fs_info->ordered_root_lock);
 		if (nr != U64_MAX) {
@@ -608,8 +598,6 @@ u64 btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
 	list_splice_tail(&splice, &fs_info->ordered_roots);
 	spin_unlock(&fs_info->ordered_root_lock);
 	mutex_unlock(&fs_info->ordered_operations_mutex);
-
-	return total_done;
 }
 
 /*
@@ -624,7 +612,7 @@ void btrfs_start_ordered_extent(struct inode *inode,
 				       int wait)
 {
 	u64 start = entry->file_offset;
-	u64 end = start + entry->len - 1;
+	u64 end = start + entry->num_bytes - 1;
 
 	trace_btrfs_ordered_extent_start(inode, entry);
 
@@ -685,16 +673,21 @@ int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
 			btrfs_put_ordered_extent(ordered);
 			break;
 		}
-		if (ordered->file_offset + ordered->len <= start) {
+		if (ordered->file_offset + ordered->num_bytes <= start) {
 			btrfs_put_ordered_extent(ordered);
 			break;
 		}
 		btrfs_start_ordered_extent(inode, ordered, 1);
 		end = ordered->file_offset;
+		/*
+		 * If the ordered extent had an error save the error but don't
+		 * exit without waiting first for all other ordered extents in
+		 * the range to complete.
+		 */
 		if (test_bit(BTRFS_ORDERED_IOERR, &ordered->flags))
 			ret = -EIO;
 		btrfs_put_ordered_extent(ordered);
-		if (ret || end == 0 || end == start)
+		if (end == 0 || end == start)
 			break;
 		end--;
 	}
@@ -793,134 +786,6 @@ out:
 }
 
 /*
- * After an extent is done, call this to conditionally update the on disk
- * i_size.  i_size is updated to cover any fully written part of the file.
- */
-int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
-				struct btrfs_ordered_extent *ordered)
-{
-	struct btrfs_ordered_inode_tree *tree = &BTRFS_I(inode)->ordered_tree;
-	u64 disk_i_size;
-	u64 new_i_size;
-	u64 i_size = i_size_read(inode);
-	struct rb_node *node;
-	struct rb_node *prev = NULL;
-	struct btrfs_ordered_extent *test;
-	int ret = 1;
-	u64 orig_offset = offset;
-
-	spin_lock_irq(&tree->lock);
-	if (ordered) {
-		offset = entry_end(ordered);
-		if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered->flags))
-			offset = min(offset,
-				     ordered->file_offset +
-				     ordered->truncated_len);
-	} else {
-		offset = ALIGN(offset, btrfs_inode_sectorsize(inode));
-	}
-	disk_i_size = BTRFS_I(inode)->disk_i_size;
-
-	/*
-	 * truncate file.
-	 * If ordered is not NULL, then this is called from endio and
-	 * disk_i_size will be updated by either truncate itself or any
-	 * in-flight IOs which are inside the disk_i_size.
-	 *
-	 * Because btrfs_setsize() may set i_size with disk_i_size if truncate
-	 * fails somehow, we need to make sure we have a precise disk_i_size by
-	 * updating it as usual.
-	 *
-	 */
-	if (!ordered && disk_i_size > i_size) {
-		BTRFS_I(inode)->disk_i_size = orig_offset;
-		ret = 0;
-		goto out;
-	}
-
-	/*
-	 * if the disk i_size is already at the inode->i_size, or
-	 * this ordered extent is inside the disk i_size, we're done
-	 */
-	if (disk_i_size == i_size)
-		goto out;
-
-	/*
-	 * We still need to update disk_i_size if outstanding_isize is greater
-	 * than disk_i_size.
-	 */
-	if (offset <= disk_i_size &&
-	    (!ordered || ordered->outstanding_isize <= disk_i_size))
-		goto out;
-
-	/*
-	 * walk backward from this ordered extent to disk_i_size.
-	 * if we find an ordered extent then we can't update disk i_size
-	 * yet
-	 */
-	if (ordered) {
-		node = rb_prev(&ordered->rb_node);
-	} else {
-		prev = tree_search(tree, offset);
-		/*
-		 * we insert file extents without involving ordered struct,
-		 * so there should be no ordered struct cover this offset
-		 */
-		if (prev) {
-			test = rb_entry(prev, struct btrfs_ordered_extent,
-					rb_node);
-			BUG_ON(offset_in_entry(test, offset));
-		}
-		node = prev;
-	}
-	for (; node; node = rb_prev(node)) {
-		test = rb_entry(node, struct btrfs_ordered_extent, rb_node);
-
-		/* We treat this entry as if it doesn't exist */
-		if (test_bit(BTRFS_ORDERED_UPDATED_ISIZE, &test->flags))
-			continue;
-
-		if (entry_end(test) <= disk_i_size)
-			break;
-		if (test->file_offset >= i_size)
-			break;
-
-		/*
-		 * We don't update disk_i_size now, so record this undealt
-		 * i_size. Or we will not know the real i_size.
-		 */
-		if (test->outstanding_isize < offset)
-			test->outstanding_isize = offset;
-		if (ordered &&
-		    ordered->outstanding_isize > test->outstanding_isize)
-			test->outstanding_isize = ordered->outstanding_isize;
-		goto out;
-	}
-	new_i_size = min_t(u64, offset, i_size);
-
-	/*
-	 * Some ordered extents may completed before the current one, and
-	 * we hold the real i_size in ->outstanding_isize.
-	 */
-	if (ordered && ordered->outstanding_isize > new_i_size)
-		new_i_size = min_t(u64, ordered->outstanding_isize, i_size);
-	BTRFS_I(inode)->disk_i_size = new_i_size;
-	ret = 0;
-out:
-	/*
-	 * We need to do this because we can't remove ordered extents until
-	 * after the i_disk_size has been updated and then the inode has been
-	 * updated to reflect the change, so we need to tell anybody who finds
-	 * this ordered extent that we've already done all the real work, we
-	 * just haven't completed all the other work.
-	 */
-	if (ordered)
-		set_bit(BTRFS_ORDERED_UPDATED_ISIZE, &ordered->flags);
-	spin_unlock_irq(&tree->lock);
-	return ret;
-}
-
-/*
  * search the ordered extents for one corresponding to 'offset' and
  * try to find a checksum.  This is used because we allow pages to
  * be reclaimed before their checksum is actually put into the btree
@@ -970,7 +835,6 @@ out:
  * btrfs_flush_ordered_range - Lock the passed range and ensures all pending
  * ordered extents in it are run to completion.
  *
- * @tree:         IO tree used for locking out other users of the range
  * @inode:        Inode whose ordered tree is to be searched
  * @start:        Beginning of range to flush
  * @end:          Last byte of range to lock
@@ -980,8 +844,7 @@ out:
  * This function always returns with the given range locked, ensuring after it's
  * called no order extent can be pending.
  */
-void btrfs_lock_and_flush_ordered_range(struct extent_io_tree *tree,
-					struct btrfs_inode *inode, u64 start,
+void btrfs_lock_and_flush_ordered_range(struct btrfs_inode *inode, u64 start,
 					u64 end,
 					struct extent_state **cached_state)
 {
@@ -993,7 +856,7 @@ void btrfs_lock_and_flush_ordered_range(struct extent_io_tree *tree,
 		cachedp = cached_state;
 
 	while (1) {
-		lock_extent_bits(tree, start, end, cachedp);
+		lock_extent_bits(&inode->io_tree, start, end, cachedp);
 		ordered = btrfs_lookup_ordered_range(inode, start,
 						     end - start + 1);
 		if (!ordered) {
@@ -1006,7 +869,7 @@ void btrfs_lock_and_flush_ordered_range(struct extent_io_tree *tree,
 				refcount_dec(&cache->refs);
 			break;
 		}
-		unlock_extent_cached(tree, start, end, cachedp);
+		unlock_extent_cached(&inode->io_tree, start, end, cachedp);
 		btrfs_start_ordered_extent(&inode->vfs_inode, ordered, 1);
 		btrfs_put_ordered_extent(ordered);
 	}
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 5204171ea962..c01c9698250b 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -52,11 +52,6 @@ enum {
 	BTRFS_ORDERED_DIRECT,
 	/* We had an io error when writing this out */
 	BTRFS_ORDERED_IOERR,
-	/*
-	 * indicates whether this ordered extent has done its due diligence in
-	 * updating the isize
-	 */
-	BTRFS_ORDERED_UPDATED_ISIZE,
 	/* Set when we have to truncate an extent */
 	BTRFS_ORDERED_TRUNCATED,
 	/* Regular IO for COW */
@@ -67,14 +62,13 @@ struct btrfs_ordered_extent {
 	/* logical offset in the file */
 	u64 file_offset;
 
-	/* disk byte number */
-	u64 start;
-
-	/* ram length of the extent in bytes */
-	u64 len;
-
-	/* extent length on disk */
-	u64 disk_len;
+	/*
+	 * These fields directly correspond to the same fields in
+	 * btrfs_file_extent_item.
+	 */
+	u64 disk_bytenr;
+	u64 num_bytes;
+	u64 disk_num_bytes;
 
 	/* number of bytes that still need writing */
 	u64 bytes_left;
@@ -161,12 +155,15 @@ int btrfs_dec_test_first_ordered_pending(struct inode *inode,
 				   u64 *file_offset, u64 io_size,
 				   int uptodate);
 int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
-			     u64 start, u64 len, u64 disk_len, int type);
+			     u64 disk_bytenr, u64 num_bytes, u64 disk_num_bytes,
+			     int type);
 int btrfs_add_ordered_extent_dio(struct inode *inode, u64 file_offset,
-				 u64 start, u64 len, u64 disk_len, int type);
+				 u64 disk_bytenr, u64 num_bytes,
+				 u64 disk_num_bytes, int type);
 int btrfs_add_ordered_extent_compress(struct inode *inode, u64 file_offset,
-				      u64 start, u64 len, u64 disk_len,
-				      int type, int compress_type);
+				      u64 disk_bytenr, u64 num_bytes,
+				      u64 disk_num_bytes, int type,
+				      int compress_type);
 void btrfs_add_ordered_sum(struct btrfs_ordered_extent *entry,
 			   struct btrfs_ordered_sum *sum);
 struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct inode *inode,
@@ -180,16 +177,13 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_range(
 		struct btrfs_inode *inode,
 		u64 file_offset,
 		u64 len);
-int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
-				struct btrfs_ordered_extent *ordered);
 int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
 			   u8 *sum, int len);
 u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr,
 			       const u64 range_start, const u64 range_len);
-u64 btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
+void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr,
 			      const u64 range_start, const u64 range_len);
-void btrfs_lock_and_flush_ordered_range(struct extent_io_tree *tree,
-					struct btrfs_inode *inode, u64 start,
+void btrfs_lock_and_flush_ordered_range(struct btrfs_inode *inode, u64 start,
 					u64 end,
 					struct extent_state **cached_state);
 int __init ordered_data_init(void);
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index 9cb50577d982..61f44e78e3c9 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -266,9 +266,9 @@ void btrfs_print_leaf(struct extent_buffer *l)
 					    struct btrfs_block_group_item);
 			pr_info(
 		   "\t\tblock group used %llu chunk_objectid %llu flags %llu\n",
-				btrfs_disk_block_group_used(l, bi),
-				btrfs_disk_block_group_chunk_objectid(l, bi),
-				btrfs_disk_block_group_flags(l, bi));
+				btrfs_block_group_used(l, bi),
+				btrfs_block_group_chunk_objectid(l, bi),
+				btrfs_block_group_flags(l, bi));
 			break;
 		case BTRFS_CHUNK_ITEM_KEY:
 			print_chunk(l, btrfs_item_ptr(l, i,
@@ -317,7 +317,7 @@ void btrfs_print_leaf(struct extent_buffer *l)
 			print_uuid_item(l, btrfs_item_ptr_offset(l, i),
 					btrfs_item_size_nr(l, i));
 			break;
-		};
+		}
 	}
 }
 
diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
index 1e664e0b59b8..ff1ff90e48b1 100644
--- a/fs/btrfs/props.c
+++ b/fs/btrfs/props.c
@@ -383,7 +383,7 @@ static int inherit_props(struct btrfs_trans_handle *trans,
 
 		if (need_reserve) {
 			btrfs_block_rsv_release(fs_info, trans->block_rsv,
-					num_bytes);
+					num_bytes, NULL);
 			if (ret)
 				return ret;
 		}
@@ -416,11 +416,11 @@ int btrfs_subvol_inherit_props(struct btrfs_trans_handle *trans,
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 
-	parent_inode = btrfs_iget(sb, &key, parent_root, NULL);
+	parent_inode = btrfs_iget(sb, &key, parent_root);
 	if (IS_ERR(parent_inode))
 		return PTR_ERR(parent_inode);
 
-	child_inode = btrfs_iget(sb, &key, root, NULL);
+	child_inode = btrfs_iget(sb, &key, root);
 	if (IS_ERR(child_inode)) {
 		iput(parent_inode);
 		return PTR_ERR(child_inode);
@@ -437,8 +437,6 @@ void __init btrfs_props_init(void)
 {
 	int i;
 
-	hash_init(prop_handlers_ht);
-
 	for (i = 0; i < ARRAY_SIZE(prop_handlers); i++) {
 		struct prop_handler *p = &prop_handlers[i];
 		u64 h = btrfs_name_hash(p->xattr_name, strlen(p->xattr_name));
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 3ad151655eb8..c3888fb367e7 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -1030,6 +1030,7 @@ out_add_root:
 	ret = qgroup_rescan_init(fs_info, 0, 1);
 	if (!ret) {
 	        qgroup_rescan_zero_tracking(fs_info);
+		fs_info->qgroup_rescan_running = true;
 	        btrfs_queue_work(fs_info->qgroup_rescan_workers,
 	                         &fs_info->qgroup_rescan_work);
 	}
@@ -1037,11 +1038,8 @@ out_add_root:
 out_free_path:
 	btrfs_free_path(path);
 out_free_root:
-	if (ret) {
-		free_extent_buffer(quota_root->node);
-		free_extent_buffer(quota_root->commit_root);
-		kfree(quota_root);
-	}
+	if (ret)
+		btrfs_put_root(quota_root);
 out:
 	if (ret) {
 		ulist_free(fs_info->qgroup_ulist);
@@ -1104,9 +1102,7 @@ int btrfs_quota_disable(struct btrfs_fs_info *fs_info)
 	btrfs_tree_unlock(quota_root->node);
 	btrfs_free_tree_block(trans, quota_root, quota_root->node, 0, 1);
 
-	free_extent_buffer(quota_root->node);
-	free_extent_buffer(quota_root->commit_root);
-	kfree(quota_root);
+	btrfs_put_root(quota_root);
 
 end_trans:
 	ret = btrfs_end_transaction(trans);
@@ -1243,7 +1239,6 @@ int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
 			      u64 dst)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *parent;
 	struct btrfs_qgroup *member;
 	struct btrfs_qgroup_list *list;
@@ -1259,9 +1254,8 @@ int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
 		return -ENOMEM;
 
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
-	quota_root = fs_info->quota_root;
-	if (!quota_root) {
-		ret = -EINVAL;
+	if (!fs_info->quota_root) {
+		ret = -ENOTCONN;
 		goto out;
 	}
 	member = find_qgroup_rb(fs_info, src);
@@ -1307,7 +1301,6 @@ static int __del_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
 				 u64 dst)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *parent;
 	struct btrfs_qgroup *member;
 	struct btrfs_qgroup_list *list;
@@ -1320,9 +1313,8 @@ static int __del_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
 	if (!tmp)
 		return -ENOMEM;
 
-	quota_root = fs_info->quota_root;
-	if (!quota_root) {
-		ret = -EINVAL;
+	if (!fs_info->quota_root) {
+		ret = -ENOTCONN;
 		goto out;
 	}
 
@@ -1387,11 +1379,11 @@ int btrfs_create_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid)
 	int ret = 0;
 
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
-	quota_root = fs_info->quota_root;
-	if (!quota_root) {
-		ret = -EINVAL;
+	if (!fs_info->quota_root) {
+		ret = -ENOTCONN;
 		goto out;
 	}
+	quota_root = fs_info->quota_root;
 	qgroup = find_qgroup_rb(fs_info, qgroupid);
 	if (qgroup) {
 		ret = -EEXIST;
@@ -1416,15 +1408,13 @@ out:
 int btrfs_remove_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *qgroup;
 	struct btrfs_qgroup_list *list;
 	int ret = 0;
 
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
-	quota_root = fs_info->quota_root;
-	if (!quota_root) {
-		ret = -EINVAL;
+	if (!fs_info->quota_root) {
+		ret = -ENOTCONN;
 		goto out;
 	}
 
@@ -1465,7 +1455,6 @@ int btrfs_limit_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid,
 		       struct btrfs_qgroup_limit *limit)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *qgroup;
 	int ret = 0;
 	/* Sometimes we would want to clear the limit on this qgroup.
@@ -1475,9 +1464,8 @@ int btrfs_limit_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid,
 	const u64 CLEAR_VALUE = -1;
 
 	mutex_lock(&fs_info->qgroup_ioctl_lock);
-	quota_root = fs_info->quota_root;
-	if (!quota_root) {
-		ret = -EINVAL;
+	if (!fs_info->quota_root) {
+		ret = -ENOTCONN;
 		goto out;
 	}
 
@@ -1811,7 +1799,7 @@ static int qgroup_trace_extent_swap(struct btrfs_trans_handle* trans,
 		btrfs_item_key_to_cpu(dst_path->nodes[dst_level], &key, 0);
 
 	/* For src_path */
-	extent_buffer_get(src_eb);
+	atomic_inc(&src_eb->refs);
 	src_path->nodes[root_level] = src_eb;
 	src_path->slots[root_level] = dst_path->slots[root_level];
 	src_path->locks[root_level] = 0;
@@ -2067,7 +2055,7 @@ static int qgroup_trace_subtree_swap(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 	/* For dst_path */
-	extent_buffer_get(dst_eb);
+	atomic_inc(&dst_eb->refs);
 	dst_path->nodes[level] = dst_eb;
 	dst_path->slots[level] = 0;
 	dst_path->locks[level] = 0;
@@ -2126,7 +2114,7 @@ int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 	 * walk back up the tree (adjusting slot pointers as we go)
 	 * and restart the search process.
 	 */
-	extent_buffer_get(root_eb); /* For path */
+	atomic_inc(&root_eb->refs);	/* For path */
 	path->nodes[root_level] = root_eb;
 	path->slots[root_level] = 0;
 	path->locks[root_level] = 0; /* so release_path doesn't try to unlock */
@@ -2423,8 +2411,12 @@ int btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans, u64 bytenr,
 	u64 nr_old_roots = 0;
 	int ret = 0;
 
+	/*
+	 * If quotas get disabled meanwhile, the resouces need to be freed and
+	 * we can't just exit here.
+	 */
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
-		return 0;
+		goto out_free;
 
 	if (new_roots) {
 		if (!maybe_fs_roots(new_roots))
@@ -2578,10 +2570,9 @@ cleanup:
 int btrfs_run_qgroups(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_root *quota_root = fs_info->quota_root;
 	int ret = 0;
 
-	if (!quota_root)
+	if (!fs_info->quota_root)
 		return ret;
 
 	spin_lock(&fs_info->qgroup_lock);
@@ -2875,7 +2866,6 @@ static bool qgroup_check_limits(struct btrfs_fs_info *fs_info,
 static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes, bool enforce,
 			  enum btrfs_qgroup_rsv_type type)
 {
-	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *qgroup;
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	u64 ref_root = root->root_key.objectid;
@@ -2894,8 +2884,7 @@ static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes, bool enforce,
 		enforce = false;
 
 	spin_lock(&fs_info->qgroup_lock);
-	quota_root = fs_info->quota_root;
-	if (!quota_root)
+	if (!fs_info->quota_root)
 		goto out;
 
 	qgroup = find_qgroup_rb(fs_info, ref_root);
@@ -2962,7 +2951,6 @@ void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
 			       u64 ref_root, u64 num_bytes,
 			       enum btrfs_qgroup_rsv_type type)
 {
-	struct btrfs_root *quota_root;
 	struct btrfs_qgroup *qgroup;
 	struct ulist_node *unode;
 	struct ulist_iterator uiter;
@@ -2980,8 +2968,7 @@ void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
 	}
 	spin_lock(&fs_info->qgroup_lock);
 
-	quota_root = fs_info->quota_root;
-	if (!quota_root)
+	if (!fs_info->quota_root)
 		goto out;
 
 	qgroup = find_qgroup_rb(fs_info, ref_root);
@@ -3232,12 +3219,12 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 		if (!(fs_info->qgroup_flags &
 		      BTRFS_QGROUP_STATUS_FLAG_RESCAN)) {
 			btrfs_warn(fs_info,
-			"qgroup rescan init failed, qgroup is not enabled");
+			"qgroup rescan init failed, qgroup rescan is not queued");
 			ret = -EINVAL;
 		} else if (!(fs_info->qgroup_flags &
 			     BTRFS_QGROUP_STATUS_FLAG_ON)) {
 			btrfs_warn(fs_info,
-			"qgroup rescan init failed, qgroup rescan is not queued");
+			"qgroup rescan init failed, qgroup is not enabled");
 			ret = -EINVAL;
 		}
 
@@ -3246,7 +3233,6 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 	}
 
 	mutex_lock(&fs_info->qgroup_rescan_lock);
-	spin_lock(&fs_info->qgroup_lock);
 
 	if (init_flags) {
 		if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) {
@@ -3261,7 +3247,6 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 		}
 
 		if (ret) {
-			spin_unlock(&fs_info->qgroup_lock);
 			mutex_unlock(&fs_info->qgroup_rescan_lock);
 			return ret;
 		}
@@ -3272,15 +3257,9 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 		sizeof(fs_info->qgroup_rescan_progress));
 	fs_info->qgroup_rescan_progress.objectid = progress_objectid;
 	init_completion(&fs_info->qgroup_rescan_completion);
-	fs_info->qgroup_rescan_running = true;
-
-	spin_unlock(&fs_info->qgroup_lock);
 	mutex_unlock(&fs_info->qgroup_rescan_lock);
 
-	memset(&fs_info->qgroup_rescan_work, 0,
-	       sizeof(fs_info->qgroup_rescan_work));
 	btrfs_init_work(&fs_info->qgroup_rescan_work,
-			btrfs_qgroup_rescan_helper,
 			btrfs_qgroup_rescan_worker, NULL, NULL);
 	return 0;
 }
@@ -3338,8 +3317,11 @@ btrfs_qgroup_rescan(struct btrfs_fs_info *fs_info)
 
 	qgroup_rescan_zero_tracking(fs_info);
 
+	mutex_lock(&fs_info->qgroup_rescan_lock);
+	fs_info->qgroup_rescan_running = true;
 	btrfs_queue_work(fs_info->qgroup_rescan_workers,
 			 &fs_info->qgroup_rescan_work);
+	mutex_unlock(&fs_info->qgroup_rescan_lock);
 
 	return 0;
 }
@@ -3351,9 +3333,7 @@ int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info,
 	int ret = 0;
 
 	mutex_lock(&fs_info->qgroup_rescan_lock);
-	spin_lock(&fs_info->qgroup_lock);
 	running = fs_info->qgroup_rescan_running;
-	spin_unlock(&fs_info->qgroup_lock);
 	mutex_unlock(&fs_info->qgroup_rescan_lock);
 
 	if (!running)
@@ -3375,9 +3355,13 @@ int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info,
 void
 btrfs_qgroup_rescan_resume(struct btrfs_fs_info *fs_info)
 {
-	if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN)
+	if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) {
+		mutex_lock(&fs_info->qgroup_rescan_lock);
+		fs_info->qgroup_rescan_running = true;
 		btrfs_queue_work(fs_info->qgroup_rescan_workers,
 				 &fs_info->qgroup_rescan_work);
+		mutex_unlock(&fs_info->qgroup_rescan_lock);
+	}
 }
 
 /*
@@ -3684,7 +3668,6 @@ void __btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes,
 static void qgroup_convert_meta(struct btrfs_fs_info *fs_info, u64 ref_root,
 				int num_bytes)
 {
-	struct btrfs_root *quota_root = fs_info->quota_root;
 	struct btrfs_qgroup *qgroup;
 	struct ulist_node *unode;
 	struct ulist_iterator uiter;
@@ -3692,7 +3675,7 @@ static void qgroup_convert_meta(struct btrfs_fs_info *fs_info, u64 ref_root,
 
 	if (num_bytes == 0)
 		return;
-	if (!quota_root)
+	if (!fs_info->quota_root)
 		return;
 
 	spin_lock(&fs_info->qgroup_lock);
@@ -3826,7 +3809,7 @@ out:
  */
 int btrfs_qgroup_add_swapped_blocks(struct btrfs_trans_handle *trans,
 		struct btrfs_root *subvol_root,
-		struct btrfs_block_group_cache *bg,
+		struct btrfs_block_group *bg,
 		struct extent_buffer *subvol_parent, int subvol_slot,
 		struct extent_buffer *reloc_parent, int reloc_slot,
 		u64 last_snapshot)
@@ -4015,3 +3998,16 @@ out:
 	}
 	return ret;
 }
+
+void btrfs_qgroup_destroy_extent_records(struct btrfs_transaction *trans)
+{
+	struct btrfs_qgroup_extent_record *entry;
+	struct btrfs_qgroup_extent_record *next;
+	struct rb_root *root;
+
+	root = &trans->delayed_refs.dirty_extent_root;
+	rbtree_postorder_for_each_entry_safe(entry, next, root, node) {
+		ulist_free(entry->old_roots);
+		kfree(entry);
+	}
+}
diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h
index 46ba7bd2961c..1bc654459469 100644
--- a/fs/btrfs/qgroup.h
+++ b/fs/btrfs/qgroup.h
@@ -408,11 +408,12 @@ void btrfs_qgroup_init_swapped_blocks(
 void btrfs_qgroup_clean_swapped_blocks(struct btrfs_root *root);
 int btrfs_qgroup_add_swapped_blocks(struct btrfs_trans_handle *trans,
 		struct btrfs_root *subvol_root,
-		struct btrfs_block_group_cache *bg,
+		struct btrfs_block_group *bg,
 		struct extent_buffer *subvol_parent, int subvol_slot,
 		struct extent_buffer *reloc_parent, int reloc_slot,
 		u64 last_snapshot);
 int btrfs_qgroup_trace_subtree_after_cow(struct btrfs_trans_handle *trans,
 		struct btrfs_root *root, struct extent_buffer *eb);
+void btrfs_qgroup_destroy_extent_records(struct btrfs_transaction *trans);
 
 #endif
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index 57a2ac721985..c870ef70f817 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -190,7 +190,7 @@ static void scrub_parity_work(struct btrfs_work *work);
 
 static void start_async_work(struct btrfs_raid_bio *rbio, btrfs_func_t work_func)
 {
-	btrfs_init_work(&rbio->work, btrfs_rmw_helper, work_func, NULL, NULL);
+	btrfs_init_work(&rbio->work, work_func, NULL, NULL);
 	btrfs_queue_work(rbio->fs_info->rmw_workers, &rbio->work);
 }
 
@@ -206,7 +206,6 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info)
 	struct btrfs_stripe_hash *h;
 	int num_entries = 1 << BTRFS_STRIPE_HASH_TABLE_BITS;
 	int i;
-	int table_size;
 
 	if (info->stripe_hash_table)
 		return 0;
@@ -218,8 +217,7 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info)
 	 * Try harder to allocate and fallback to vmalloc to lower the chance
 	 * of a failing mount.
 	 */
-	table_size = sizeof(*table) + sizeof(*h) * num_entries;
-	table = kvzalloc(table_size, GFP_KERNEL);
+	table = kvzalloc(struct_size(table, table, num_entries), GFP_KERNEL);
 	if (!table)
 		return -ENOMEM;
 
@@ -671,8 +669,7 @@ static struct page *rbio_qstripe_page(struct btrfs_raid_bio *rbio, int index)
  */
 static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
 {
-	int bucket = rbio_bucket(rbio);
-	struct btrfs_stripe_hash *h = rbio->fs_info->stripe_hash_table->table + bucket;
+	struct btrfs_stripe_hash *h;
 	struct btrfs_raid_bio *cur;
 	struct btrfs_raid_bio *pending;
 	unsigned long flags;
@@ -680,64 +677,63 @@ static noinline int lock_stripe_add(struct btrfs_raid_bio *rbio)
 	struct btrfs_raid_bio *cache_drop = NULL;
 	int ret = 0;
 
+	h = rbio->fs_info->stripe_hash_table->table + rbio_bucket(rbio);
+
 	spin_lock_irqsave(&h->lock, flags);
 	list_for_each_entry(cur, &h->hash_list, hash_list) {
-		if (cur->bbio->raid_map[0] == rbio->bbio->raid_map[0]) {
-			spin_lock(&cur->bio_list_lock);
-
-			/* can we steal this cached rbio's pages? */
-			if (bio_list_empty(&cur->bio_list) &&
-			    list_empty(&cur->plug_list) &&
-			    test_bit(RBIO_CACHE_BIT, &cur->flags) &&
-			    !test_bit(RBIO_RMW_LOCKED_BIT, &cur->flags)) {
-				list_del_init(&cur->hash_list);
-				refcount_dec(&cur->refs);
-
-				steal_rbio(cur, rbio);
-				cache_drop = cur;
-				spin_unlock(&cur->bio_list_lock);
+		if (cur->bbio->raid_map[0] != rbio->bbio->raid_map[0])
+			continue;
 
-				goto lockit;
-			}
+		spin_lock(&cur->bio_list_lock);
 
-			/* can we merge into the lock owner? */
-			if (rbio_can_merge(cur, rbio)) {
-				merge_rbio(cur, rbio);
-				spin_unlock(&cur->bio_list_lock);
-				freeit = rbio;
-				ret = 1;
-				goto out;
-			}
+		/* Can we steal this cached rbio's pages? */
+		if (bio_list_empty(&cur->bio_list) &&
+		    list_empty(&cur->plug_list) &&
+		    test_bit(RBIO_CACHE_BIT, &cur->flags) &&
+		    !test_bit(RBIO_RMW_LOCKED_BIT, &cur->flags)) {
+			list_del_init(&cur->hash_list);
+			refcount_dec(&cur->refs);
 
+			steal_rbio(cur, rbio);
+			cache_drop = cur;
+			spin_unlock(&cur->bio_list_lock);
 
-			/*
-			 * we couldn't merge with the running
-			 * rbio, see if we can merge with the
-			 * pending ones.  We don't have to
-			 * check for rmw_locked because there
-			 * is no way they are inside finish_rmw
-			 * right now
-			 */
-			list_for_each_entry(pending, &cur->plug_list,
-					    plug_list) {
-				if (rbio_can_merge(pending, rbio)) {
-					merge_rbio(pending, rbio);
-					spin_unlock(&cur->bio_list_lock);
-					freeit = rbio;
-					ret = 1;
-					goto out;
-				}
-			}
+			goto lockit;
+		}
 
-			/* no merging, put us on the tail of the plug list,
-			 * our rbio will be started with the currently
-			 * running rbio unlocks
-			 */
-			list_add_tail(&rbio->plug_list, &cur->plug_list);
+		/* Can we merge into the lock owner? */
+		if (rbio_can_merge(cur, rbio)) {
+			merge_rbio(cur, rbio);
 			spin_unlock(&cur->bio_list_lock);
+			freeit = rbio;
 			ret = 1;
 			goto out;
 		}
+
+
+		/*
+		 * We couldn't merge with the running rbio, see if we can merge
+		 * with the pending ones.  We don't have to check for rmw_locked
+		 * because there is no way they are inside finish_rmw right now
+		 */
+		list_for_each_entry(pending, &cur->plug_list, plug_list) {
+			if (rbio_can_merge(pending, rbio)) {
+				merge_rbio(pending, rbio);
+				spin_unlock(&cur->bio_list_lock);
+				freeit = rbio;
+				ret = 1;
+				goto out;
+			}
+		}
+
+		/*
+		 * No merging, put us on the tail of the plug list, our rbio
+		 * will be started with the currently running rbio unlocks
+		 */
+		list_add_tail(&rbio->plug_list, &cur->plug_list);
+		spin_unlock(&cur->bio_list_lock);
+		ret = 1;
+		goto out;
 	}
 lockit:
 	refcount_inc(&rbio->refs);
@@ -1198,22 +1194,19 @@ static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
 	int nr_data = rbio->nr_data;
 	int stripe;
 	int pagenr;
-	int p_stripe = -1;
-	int q_stripe = -1;
+	bool has_qstripe;
 	struct bio_list bio_list;
 	struct bio *bio;
 	int ret;
 
 	bio_list_init(&bio_list);
 
-	if (rbio->real_stripes - rbio->nr_data == 1) {
-		p_stripe = rbio->real_stripes - 1;
-	} else if (rbio->real_stripes - rbio->nr_data == 2) {
-		p_stripe = rbio->real_stripes - 2;
-		q_stripe = rbio->real_stripes - 1;
-	} else {
+	if (rbio->real_stripes - rbio->nr_data == 1)
+		has_qstripe = false;
+	else if (rbio->real_stripes - rbio->nr_data == 2)
+		has_qstripe = true;
+	else
 		BUG();
-	}
 
 	/* at this point we either have a full stripe,
 	 * or we've read the full stripe from the drive.
@@ -1257,7 +1250,7 @@ static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
 		SetPageUptodate(p);
 		pointers[stripe++] = kmap(p);
 
-		if (q_stripe != -1) {
+		if (has_qstripe) {
 
 			/*
 			 * raid6, add the qstripe and call the
@@ -1743,8 +1736,7 @@ static void btrfs_raid_unplug(struct blk_plug_cb *cb, bool from_schedule)
 	plug = container_of(cb, struct btrfs_plug_cb, cb);
 
 	if (from_schedule) {
-		btrfs_init_work(&plug->work, btrfs_rmw_helper,
-				unplug_work, NULL, NULL);
+		btrfs_init_work(&plug->work, unplug_work, NULL, NULL);
 		btrfs_queue_work(plug->info->rmw_workers,
 				 &plug->work);
 		return;
@@ -2356,8 +2348,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 	int nr_data = rbio->nr_data;
 	int stripe;
 	int pagenr;
-	int p_stripe = -1;
-	int q_stripe = -1;
+	bool has_qstripe;
 	struct page *p_page = NULL;
 	struct page *q_page = NULL;
 	struct bio_list bio_list;
@@ -2367,14 +2358,12 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 
 	bio_list_init(&bio_list);
 
-	if (rbio->real_stripes - rbio->nr_data == 1) {
-		p_stripe = rbio->real_stripes - 1;
-	} else if (rbio->real_stripes - rbio->nr_data == 2) {
-		p_stripe = rbio->real_stripes - 2;
-		q_stripe = rbio->real_stripes - 1;
-	} else {
+	if (rbio->real_stripes - rbio->nr_data == 1)
+		has_qstripe = false;
+	else if (rbio->real_stripes - rbio->nr_data == 2)
+		has_qstripe = true;
+	else
 		BUG();
-	}
 
 	if (bbio->num_tgtdevs && bbio->tgtdev_map[rbio->scrubp]) {
 		is_replace = 1;
@@ -2396,7 +2385,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 		goto cleanup;
 	SetPageUptodate(p_page);
 
-	if (q_stripe != -1) {
+	if (has_qstripe) {
 		q_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
 		if (!q_page) {
 			__free_page(p_page);
@@ -2419,8 +2408,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 		/* then add the parity stripe */
 		pointers[stripe++] = kmap(p_page);
 
-		if (q_stripe != -1) {
-
+		if (has_qstripe) {
 			/*
 			 * raid6, add the qstripe and call the
 			 * library function to fill in our p/q
diff --git a/fs/btrfs/rcu-string.h b/fs/btrfs/rcu-string.h
index a97dc74a4d3d..5c1a617eb25d 100644
--- a/fs/btrfs/rcu-string.h
+++ b/fs/btrfs/rcu-string.h
@@ -8,7 +8,7 @@
 
 struct rcu_string {
 	struct rcu_head rcu;
-	char str[0];
+	char str[];
 };
 
 static inline struct rcu_string *rcu_string_strdup(const char *src, gfp_t mask)
diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c
index ee6f60547a8d..243a2e44526e 100644
--- a/fs/btrfs/reada.c
+++ b/fs/btrfs/reada.c
@@ -227,7 +227,7 @@ static struct reada_zone *reada_find_zone(struct btrfs_device *dev, u64 logical,
 	struct btrfs_fs_info *fs_info = dev->fs_info;
 	int ret;
 	struct reada_zone *zone;
-	struct btrfs_block_group_cache *cache = NULL;
+	struct btrfs_block_group *cache = NULL;
 	u64 start;
 	u64 end;
 	int i;
@@ -248,8 +248,8 @@ static struct reada_zone *reada_find_zone(struct btrfs_device *dev, u64 logical,
 	if (!cache)
 		return NULL;
 
-	start = cache->key.objectid;
-	end = start + cache->key.offset - 1;
+	start = cache->start;
+	end = start + cache->length - 1;
 	btrfs_put_block_group(cache);
 
 	zone = kzalloc(sizeof(*zone), GFP_KERNEL);
@@ -752,21 +752,19 @@ static int reada_start_machine_dev(struct btrfs_device *dev)
 static void reada_start_machine_worker(struct btrfs_work *work)
 {
 	struct reada_machine_work *rmw;
-	struct btrfs_fs_info *fs_info;
 	int old_ioprio;
 
 	rmw = container_of(work, struct reada_machine_work, work);
-	fs_info = rmw->fs_info;
-
-	kfree(rmw);
 
 	old_ioprio = IOPRIO_PRIO_VALUE(task_nice_ioclass(current),
 				       task_nice_ioprio(current));
 	set_task_ioprio(current, BTRFS_IOPRIO_READA);
-	__reada_start_machine(fs_info);
+	__reada_start_machine(rmw->fs_info);
 	set_task_ioprio(current, old_ioprio);
 
-	atomic_dec(&fs_info->reada_works_cnt);
+	atomic_dec(&rmw->fs_info->reada_works_cnt);
+
+	kfree(rmw);
 }
 
 static void __reada_start_machine(struct btrfs_fs_info *fs_info)
@@ -821,8 +819,7 @@ static void reada_start_machine(struct btrfs_fs_info *fs_info)
 		/* FIXME we cannot handle this properly right now */
 		BUG();
 	}
-	btrfs_init_work(&rmw->work, btrfs_readahead_helper,
-			reada_start_machine_worker, NULL, NULL);
+	btrfs_init_work(&rmw->work, reada_start_machine_worker, NULL, NULL);
 	rmw->fs_info = fs_info;
 
 	btrfs_queue_work(fs_info->readahead_workers, &rmw->work);
diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c
index b57f3618e58e..7887317033c9 100644
--- a/fs/btrfs/ref-verify.c
+++ b/fs/btrfs/ref-verify.c
@@ -744,6 +744,7 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info,
 		 */
 		be = add_block_entry(fs_info, bytenr, num_bytes, ref_root);
 		if (IS_ERR(be)) {
+			kfree(ref);
 			kfree(ra);
 			ret = PTR_ERR(be);
 			goto out;
@@ -757,6 +758,8 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info,
 			"re-allocated a block that still has references to it!");
 			dump_block_entry(fs_info, be);
 			dump_ref_action(fs_info, ra);
+			kfree(ref);
+			kfree(ra);
 			goto out_unlock;
 		}
 
@@ -800,6 +803,15 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info,
 			kfree(ref);
 			kfree(ra);
 			goto out_unlock;
+		} else if (be->num_refs == 0) {
+			btrfs_err(fs_info,
+		"trying to do action %d for a bytenr that has 0 total references",
+				action);
+			dump_block_entry(fs_info, be);
+			dump_ref_action(fs_info, ra);
+			kfree(ref);
+			kfree(ra);
+			goto out_unlock;
 		}
 
 		if (!parent) {
@@ -819,6 +831,7 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info,
 "dropping a ref for a existing root that doesn't have a ref on the block");
 				dump_block_entry(fs_info, be);
 				dump_ref_action(fs_info, ra);
+				kfree(ref);
 				kfree(ra);
 				goto out_unlock;
 			}
@@ -834,6 +847,7 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info,
 "attempting to add another ref for an existing ref on a tree block");
 			dump_block_entry(fs_info, be);
 			dump_ref_action(fs_info, ra);
+			kfree(ref);
 			kfree(ra);
 			goto out_unlock;
 		}
diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c
new file mode 100644
index 000000000000..040009d1cc31
--- /dev/null
+++ b/fs/btrfs/reflink.c
@@ -0,0 +1,805 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/blkdev.h>
+#include <linux/iversion.h>
+#include "compression.h"
+#include "ctree.h"
+#include "delalloc-space.h"
+#include "reflink.h"
+#include "transaction.h"
+
+#define BTRFS_MAX_DEDUPE_LEN	SZ_16M
+
+static int clone_finish_inode_update(struct btrfs_trans_handle *trans,
+				     struct inode *inode,
+				     u64 endoff,
+				     const u64 destoff,
+				     const u64 olen,
+				     int no_time_update)
+{
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	int ret;
+
+	inode_inc_iversion(inode);
+	if (!no_time_update)
+		inode->i_mtime = inode->i_ctime = current_time(inode);
+	/*
+	 * We round up to the block size at eof when determining which
+	 * extents to clone above, but shouldn't round up the file size.
+	 */
+	if (endoff > destoff + olen)
+		endoff = destoff + olen;
+	if (endoff > inode->i_size) {
+		i_size_write(inode, endoff);
+		btrfs_inode_safe_disk_i_size_write(inode, 0);
+	}
+
+	ret = btrfs_update_inode(trans, root, inode);
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
+		btrfs_end_transaction(trans);
+		goto out;
+	}
+	ret = btrfs_end_transaction(trans);
+out:
+	return ret;
+}
+
+static int copy_inline_to_page(struct inode *inode,
+			       const u64 file_offset,
+			       char *inline_data,
+			       const u64 size,
+			       const u64 datal,
+			       const u8 comp_type)
+{
+	const u64 block_size = btrfs_inode_sectorsize(inode);
+	const u64 range_end = file_offset + block_size - 1;
+	const size_t inline_size = size - btrfs_file_extent_calc_inline_size(0);
+	char *data_start = inline_data + btrfs_file_extent_calc_inline_size(0);
+	struct extent_changeset *data_reserved = NULL;
+	struct page *page = NULL;
+	int ret;
+
+	ASSERT(IS_ALIGNED(file_offset, block_size));
+
+	/*
+	 * We have flushed and locked the ranges of the source and destination
+	 * inodes, we also have locked the inodes, so we are safe to do a
+	 * reservation here. Also we must not do the reservation while holding
+	 * a transaction open, otherwise we would deadlock.
+	 */
+	ret = btrfs_delalloc_reserve_space(inode, &data_reserved, file_offset,
+					   block_size);
+	if (ret)
+		goto out;
+
+	page = find_or_create_page(inode->i_mapping, file_offset >> PAGE_SHIFT,
+				   btrfs_alloc_write_mask(inode->i_mapping));
+	if (!page) {
+		ret = -ENOMEM;
+		goto out_unlock;
+	}
+
+	set_page_extent_mapped(page);
+	clear_extent_bit(&BTRFS_I(inode)->io_tree, file_offset, range_end,
+			 EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
+			 0, 0, NULL);
+	ret = btrfs_set_extent_delalloc(inode, file_offset, range_end, 0, NULL);
+	if (ret)
+		goto out_unlock;
+
+	if (comp_type == BTRFS_COMPRESS_NONE) {
+		char *map;
+
+		map = kmap(page);
+		memcpy(map, data_start, datal);
+		flush_dcache_page(page);
+		kunmap(page);
+	} else {
+		ret = btrfs_decompress(comp_type, data_start, page, 0,
+				       inline_size, datal);
+		if (ret)
+			goto out_unlock;
+		flush_dcache_page(page);
+	}
+
+	/*
+	 * If our inline data is smaller then the block/page size, then the
+	 * remaining of the block/page is equivalent to zeroes. We had something
+	 * like the following done:
+	 *
+	 * $ xfs_io -f -c "pwrite -S 0xab 0 500" file
+	 * $ sync  # (or fsync)
+	 * $ xfs_io -c "falloc 0 4K" file
+	 * $ xfs_io -c "pwrite -S 0xcd 4K 4K"
+	 *
+	 * So what's in the range [500, 4095] corresponds to zeroes.
+	 */
+	if (datal < block_size) {
+		char *map;
+
+		map = kmap(page);
+		memset(map + datal, 0, block_size - datal);
+		flush_dcache_page(page);
+		kunmap(page);
+	}
+
+	SetPageUptodate(page);
+	ClearPageChecked(page);
+	set_page_dirty(page);
+out_unlock:
+	if (page) {
+		unlock_page(page);
+		put_page(page);
+	}
+	if (ret)
+		btrfs_delalloc_release_space(inode, data_reserved, file_offset,
+					     block_size, true);
+	btrfs_delalloc_release_extents(BTRFS_I(inode), block_size);
+out:
+	extent_changeset_free(data_reserved);
+
+	return ret;
+}
+
+/*
+ * Deal with cloning of inline extents. We try to copy the inline extent from
+ * the source inode to destination inode when possible. When not possible we
+ * copy the inline extent's data into the respective page of the inode.
+ */
+static int clone_copy_inline_extent(struct inode *dst,
+				    struct btrfs_path *path,
+				    struct btrfs_key *new_key,
+				    const u64 drop_start,
+				    const u64 datal,
+				    const u64 size,
+				    const u8 comp_type,
+				    char *inline_data,
+				    struct btrfs_trans_handle **trans_out)
+{
+	struct btrfs_fs_info *fs_info = btrfs_sb(dst->i_sb);
+	struct btrfs_root *root = BTRFS_I(dst)->root;
+	const u64 aligned_end = ALIGN(new_key->offset + datal,
+				      fs_info->sectorsize);
+	struct btrfs_trans_handle *trans = NULL;
+	int ret;
+	struct btrfs_key key;
+
+	if (new_key->offset > 0) {
+		ret = copy_inline_to_page(dst, new_key->offset, inline_data,
+					  size, datal, comp_type);
+		goto out;
+	}
+
+	key.objectid = btrfs_ino(BTRFS_I(dst));
+	key.type = BTRFS_EXTENT_DATA_KEY;
+	key.offset = 0;
+	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	if (ret < 0) {
+		return ret;
+	} else if (ret > 0) {
+		if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
+			ret = btrfs_next_leaf(root, path);
+			if (ret < 0)
+				return ret;
+			else if (ret > 0)
+				goto copy_inline_extent;
+		}
+		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+		if (key.objectid == btrfs_ino(BTRFS_I(dst)) &&
+		    key.type == BTRFS_EXTENT_DATA_KEY) {
+			/*
+			 * There's an implicit hole at file offset 0, copy the
+			 * inline extent's data to the page.
+			 */
+			ASSERT(key.offset > 0);
+			ret = copy_inline_to_page(dst, new_key->offset,
+						  inline_data, size, datal,
+						  comp_type);
+			goto out;
+		}
+	} else if (i_size_read(dst) <= datal) {
+		struct btrfs_file_extent_item *ei;
+
+		ei = btrfs_item_ptr(path->nodes[0], path->slots[0],
+				    struct btrfs_file_extent_item);
+		/*
+		 * If it's an inline extent replace it with the source inline
+		 * extent, otherwise copy the source inline extent data into
+		 * the respective page at the destination inode.
+		 */
+		if (btrfs_file_extent_type(path->nodes[0], ei) ==
+		    BTRFS_FILE_EXTENT_INLINE)
+			goto copy_inline_extent;
+
+		ret = copy_inline_to_page(dst, new_key->offset, inline_data,
+					  size, datal, comp_type);
+		goto out;
+	}
+
+copy_inline_extent:
+	ret = 0;
+	/*
+	 * We have no extent items, or we have an extent at offset 0 which may
+	 * or may not be inlined. All these cases are dealt the same way.
+	 */
+	if (i_size_read(dst) > datal) {
+		/*
+		 * At the destination offset 0 we have either a hole, a regular
+		 * extent or an inline extent larger then the one we want to
+		 * clone. Deal with all these cases by copying the inline extent
+		 * data into the respective page at the destination inode.
+		 */
+		ret = copy_inline_to_page(dst, new_key->offset, inline_data,
+					   size, datal, comp_type);
+		goto out;
+	}
+
+	btrfs_release_path(path);
+	/*
+	 * If we end up here it means were copy the inline extent into a leaf
+	 * of the destination inode. We know we will drop or adjust at most one
+	 * extent item in the destination root.
+	 *
+	 * 1 unit - adjusting old extent (we may have to split it)
+	 * 1 unit - add new extent
+	 * 1 unit - inode update
+	 */
+	trans = btrfs_start_transaction(root, 3);
+	if (IS_ERR(trans)) {
+		ret = PTR_ERR(trans);
+		trans = NULL;
+		goto out;
+	}
+	ret = btrfs_drop_extents(trans, root, dst, drop_start, aligned_end, 1);
+	if (ret)
+		goto out;
+	ret = btrfs_insert_empty_item(trans, root, path, new_key, size);
+	if (ret)
+		goto out;
+
+	write_extent_buffer(path->nodes[0], inline_data,
+			    btrfs_item_ptr_offset(path->nodes[0],
+						  path->slots[0]),
+			    size);
+	inode_add_bytes(dst, datal);
+	set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(dst)->runtime_flags);
+	ret = btrfs_inode_set_file_extent_range(BTRFS_I(dst), 0, aligned_end);
+out:
+	if (!ret && !trans) {
+		/*
+		 * No transaction here means we copied the inline extent into a
+		 * page of the destination inode.
+		 *
+		 * 1 unit to update inode item
+		 */
+		trans = btrfs_start_transaction(root, 1);
+		if (IS_ERR(trans)) {
+			ret = PTR_ERR(trans);
+			trans = NULL;
+		}
+	}
+	if (ret && trans) {
+		btrfs_abort_transaction(trans, ret);
+		btrfs_end_transaction(trans);
+	}
+	if (!ret)
+		*trans_out = trans;
+
+	return ret;
+}
+
+/**
+ * btrfs_clone() - clone a range from inode file to another
+ *
+ * @src: Inode to clone from
+ * @inode: Inode to clone to
+ * @off: Offset within source to start clone from
+ * @olen: Original length, passed by user, of range to clone
+ * @olen_aligned: Block-aligned value of olen
+ * @destoff: Offset within @inode to start clone
+ * @no_time_update: Whether to update mtime/ctime on the target inode
+ */
+static int btrfs_clone(struct inode *src, struct inode *inode,
+		       const u64 off, const u64 olen, const u64 olen_aligned,
+		       const u64 destoff, int no_time_update)
+{
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct btrfs_path *path = NULL;
+	struct extent_buffer *leaf;
+	struct btrfs_trans_handle *trans;
+	char *buf = NULL;
+	struct btrfs_key key;
+	u32 nritems;
+	int slot;
+	int ret;
+	const u64 len = olen_aligned;
+	u64 last_dest_end = destoff;
+
+	ret = -ENOMEM;
+	buf = kvmalloc(fs_info->nodesize, GFP_KERNEL);
+	if (!buf)
+		return ret;
+
+	path = btrfs_alloc_path();
+	if (!path) {
+		kvfree(buf);
+		return ret;
+	}
+
+	path->reada = READA_FORWARD;
+	/* Clone data */
+	key.objectid = btrfs_ino(BTRFS_I(src));
+	key.type = BTRFS_EXTENT_DATA_KEY;
+	key.offset = off;
+
+	while (1) {
+		u64 next_key_min_offset = key.offset + 1;
+		struct btrfs_file_extent_item *extent;
+		int type;
+		u32 size;
+		struct btrfs_key new_key;
+		u64 disko = 0, diskl = 0;
+		u64 datao = 0, datal = 0;
+		u8 comp;
+		u64 drop_start;
+
+		/* Note the key will change type as we walk through the tree */
+		path->leave_spinning = 1;
+		ret = btrfs_search_slot(NULL, BTRFS_I(src)->root, &key, path,
+				0, 0);
+		if (ret < 0)
+			goto out;
+		/*
+		 * First search, if no extent item that starts at offset off was
+		 * found but the previous item is an extent item, it's possible
+		 * it might overlap our target range, therefore process it.
+		 */
+		if (key.offset == off && ret > 0 && path->slots[0] > 0) {
+			btrfs_item_key_to_cpu(path->nodes[0], &key,
+					      path->slots[0] - 1);
+			if (key.type == BTRFS_EXTENT_DATA_KEY)
+				path->slots[0]--;
+		}
+
+		nritems = btrfs_header_nritems(path->nodes[0]);
+process_slot:
+		if (path->slots[0] >= nritems) {
+			ret = btrfs_next_leaf(BTRFS_I(src)->root, path);
+			if (ret < 0)
+				goto out;
+			if (ret > 0)
+				break;
+			nritems = btrfs_header_nritems(path->nodes[0]);
+		}
+		leaf = path->nodes[0];
+		slot = path->slots[0];
+
+		btrfs_item_key_to_cpu(leaf, &key, slot);
+		if (key.type > BTRFS_EXTENT_DATA_KEY ||
+		    key.objectid != btrfs_ino(BTRFS_I(src)))
+			break;
+
+		ASSERT(key.type == BTRFS_EXTENT_DATA_KEY);
+
+		extent = btrfs_item_ptr(leaf, slot,
+					struct btrfs_file_extent_item);
+		comp = btrfs_file_extent_compression(leaf, extent);
+		type = btrfs_file_extent_type(leaf, extent);
+		if (type == BTRFS_FILE_EXTENT_REG ||
+		    type == BTRFS_FILE_EXTENT_PREALLOC) {
+			disko = btrfs_file_extent_disk_bytenr(leaf, extent);
+			diskl = btrfs_file_extent_disk_num_bytes(leaf, extent);
+			datao = btrfs_file_extent_offset(leaf, extent);
+			datal = btrfs_file_extent_num_bytes(leaf, extent);
+		} else if (type == BTRFS_FILE_EXTENT_INLINE) {
+			/* Take upper bound, may be compressed */
+			datal = btrfs_file_extent_ram_bytes(leaf, extent);
+		}
+
+		/*
+		 * The first search might have left us at an extent item that
+		 * ends before our target range's start, can happen if we have
+		 * holes and NO_HOLES feature enabled.
+		 */
+		if (key.offset + datal <= off) {
+			path->slots[0]++;
+			goto process_slot;
+		} else if (key.offset >= off + len) {
+			break;
+		}
+		next_key_min_offset = key.offset + datal;
+		size = btrfs_item_size_nr(leaf, slot);
+		read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf, slot),
+				   size);
+
+		btrfs_release_path(path);
+		path->leave_spinning = 0;
+
+		memcpy(&new_key, &key, sizeof(new_key));
+		new_key.objectid = btrfs_ino(BTRFS_I(inode));
+		if (off <= key.offset)
+			new_key.offset = key.offset + destoff - off;
+		else
+			new_key.offset = destoff;
+
+		/*
+		 * Deal with a hole that doesn't have an extent item that
+		 * represents it (NO_HOLES feature enabled).
+		 * This hole is either in the middle of the cloning range or at
+		 * the beginning (fully overlaps it or partially overlaps it).
+		 */
+		if (new_key.offset != last_dest_end)
+			drop_start = last_dest_end;
+		else
+			drop_start = new_key.offset;
+
+		if (type == BTRFS_FILE_EXTENT_REG ||
+		    type == BTRFS_FILE_EXTENT_PREALLOC) {
+			struct btrfs_clone_extent_info clone_info;
+
+			/*
+			 *    a  | --- range to clone ---|  b
+			 * | ------------- extent ------------- |
+			 */
+
+			/* Subtract range b */
+			if (key.offset + datal > off + len)
+				datal = off + len - key.offset;
+
+			/* Subtract range a */
+			if (off > key.offset) {
+				datao += off - key.offset;
+				datal -= off - key.offset;
+			}
+
+			clone_info.disk_offset = disko;
+			clone_info.disk_len = diskl;
+			clone_info.data_offset = datao;
+			clone_info.data_len = datal;
+			clone_info.file_offset = new_key.offset;
+			clone_info.extent_buf = buf;
+			clone_info.item_size = size;
+			ret = btrfs_punch_hole_range(inode, path, drop_start,
+					new_key.offset + datal - 1, &clone_info,
+					&trans);
+			if (ret)
+				goto out;
+		} else if (type == BTRFS_FILE_EXTENT_INLINE) {
+			/*
+			 * Inline extents always have to start at file offset 0
+			 * and can never be bigger then the sector size. We can
+			 * never clone only parts of an inline extent, since all
+			 * reflink operations must start at a sector size aligned
+			 * offset, and the length must be aligned too or end at
+			 * the i_size (which implies the whole inlined data).
+			 */
+			ASSERT(key.offset == 0);
+			ASSERT(datal <= fs_info->sectorsize);
+			if (key.offset != 0 || datal > fs_info->sectorsize)
+				return -EUCLEAN;
+
+			ret = clone_copy_inline_extent(inode, path, &new_key,
+						       drop_start, datal, size,
+						       comp, buf, &trans);
+			if (ret)
+				goto out;
+		}
+
+		btrfs_release_path(path);
+
+		last_dest_end = ALIGN(new_key.offset + datal,
+				      fs_info->sectorsize);
+		ret = clone_finish_inode_update(trans, inode, last_dest_end,
+						destoff, olen, no_time_update);
+		if (ret)
+			goto out;
+		if (new_key.offset + datal >= destoff + len)
+			break;
+
+		btrfs_release_path(path);
+		key.offset = next_key_min_offset;
+
+		if (fatal_signal_pending(current)) {
+			ret = -EINTR;
+			goto out;
+		}
+	}
+	ret = 0;
+
+	if (last_dest_end < destoff + len) {
+		/*
+		 * We have an implicit hole that fully or partially overlaps our
+		 * cloning range at its end. This means that we either have the
+		 * NO_HOLES feature enabled or the implicit hole happened due to
+		 * mixing buffered and direct IO writes against this file.
+		 */
+		btrfs_release_path(path);
+		path->leave_spinning = 0;
+
+		ret = btrfs_punch_hole_range(inode, path, last_dest_end,
+				destoff + len - 1, NULL, &trans);
+		if (ret)
+			goto out;
+
+		ret = clone_finish_inode_update(trans, inode, destoff + len,
+						destoff, olen, no_time_update);
+	}
+
+out:
+	btrfs_free_path(path);
+	kvfree(buf);
+	return ret;
+}
+
+static void btrfs_double_extent_unlock(struct inode *inode1, u64 loff1,
+				       struct inode *inode2, u64 loff2, u64 len)
+{
+	unlock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1);
+	unlock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1);
+}
+
+static void btrfs_double_extent_lock(struct inode *inode1, u64 loff1,
+				     struct inode *inode2, u64 loff2, u64 len)
+{
+	if (inode1 < inode2) {
+		swap(inode1, inode2);
+		swap(loff1, loff2);
+	} else if (inode1 == inode2 && loff2 < loff1) {
+		swap(loff1, loff2);
+	}
+	lock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1);
+	lock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1);
+}
+
+static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 len,
+				   struct inode *dst, u64 dst_loff)
+{
+	const u64 bs = BTRFS_I(src)->root->fs_info->sb->s_blocksize;
+	int ret;
+
+	/*
+	 * Lock destination range to serialize with concurrent readpages() and
+	 * source range to serialize with relocation.
+	 */
+	btrfs_double_extent_lock(src, loff, dst, dst_loff, len);
+	ret = btrfs_clone(src, dst, loff, len, ALIGN(len, bs), dst_loff, 1);
+	btrfs_double_extent_unlock(src, loff, dst, dst_loff, len);
+
+	return ret;
+}
+
+static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
+			     struct inode *dst, u64 dst_loff)
+{
+	int ret;
+	u64 i, tail_len, chunk_count;
+	struct btrfs_root *root_dst = BTRFS_I(dst)->root;
+
+	spin_lock(&root_dst->root_item_lock);
+	if (root_dst->send_in_progress) {
+		btrfs_warn_rl(root_dst->fs_info,
+"cannot deduplicate to root %llu while send operations are using it (%d in progress)",
+			      root_dst->root_key.objectid,
+			      root_dst->send_in_progress);
+		spin_unlock(&root_dst->root_item_lock);
+		return -EAGAIN;
+	}
+	root_dst->dedupe_in_progress++;
+	spin_unlock(&root_dst->root_item_lock);
+
+	tail_len = olen % BTRFS_MAX_DEDUPE_LEN;
+	chunk_count = div_u64(olen, BTRFS_MAX_DEDUPE_LEN);
+
+	for (i = 0; i < chunk_count; i++) {
+		ret = btrfs_extent_same_range(src, loff, BTRFS_MAX_DEDUPE_LEN,
+					      dst, dst_loff);
+		if (ret)
+			goto out;
+
+		loff += BTRFS_MAX_DEDUPE_LEN;
+		dst_loff += BTRFS_MAX_DEDUPE_LEN;
+	}
+
+	if (tail_len > 0)
+		ret = btrfs_extent_same_range(src, loff, tail_len, dst, dst_loff);
+out:
+	spin_lock(&root_dst->root_item_lock);
+	root_dst->dedupe_in_progress--;
+	spin_unlock(&root_dst->root_item_lock);
+
+	return ret;
+}
+
+static noinline int btrfs_clone_files(struct file *file, struct file *file_src,
+					u64 off, u64 olen, u64 destoff)
+{
+	struct inode *inode = file_inode(file);
+	struct inode *src = file_inode(file_src);
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	int ret;
+	int wb_ret;
+	u64 len = olen;
+	u64 bs = fs_info->sb->s_blocksize;
+
+	/*
+	 * VFS's generic_remap_file_range_prep() protects us from cloning the
+	 * eof block into the middle of a file, which would result in corruption
+	 * if the file size is not blocksize aligned. So we don't need to check
+	 * for that case here.
+	 */
+	if (off + len == src->i_size)
+		len = ALIGN(src->i_size, bs) - off;
+
+	if (destoff > inode->i_size) {
+		const u64 wb_start = ALIGN_DOWN(inode->i_size, bs);
+
+		ret = btrfs_cont_expand(inode, inode->i_size, destoff);
+		if (ret)
+			return ret;
+		/*
+		 * We may have truncated the last block if the inode's size is
+		 * not sector size aligned, so we need to wait for writeback to
+		 * complete before proceeding further, otherwise we can race
+		 * with cloning and attempt to increment a reference to an
+		 * extent that no longer exists (writeback completed right after
+		 * we found the previous extent covering eof and before we
+		 * attempted to increment its reference count).
+		 */
+		ret = btrfs_wait_ordered_range(inode, wb_start,
+					       destoff - wb_start);
+		if (ret)
+			return ret;
+	}
+
+	/*
+	 * Lock destination range to serialize with concurrent readpages() and
+	 * source range to serialize with relocation.
+	 */
+	btrfs_double_extent_lock(src, off, inode, destoff, len);
+	ret = btrfs_clone(src, inode, off, olen, len, destoff, 0);
+	btrfs_double_extent_unlock(src, off, inode, destoff, len);
+
+	/*
+	 * We may have copied an inline extent into a page of the destination
+	 * range, so wait for writeback to complete before truncating pages
+	 * from the page cache. This is a rare case.
+	 */
+	wb_ret = btrfs_wait_ordered_range(inode, destoff, len);
+	ret = ret ? ret : wb_ret;
+	/*
+	 * Truncate page cache pages so that future reads will see the cloned
+	 * data immediately and not the previous data.
+	 */
+	truncate_inode_pages_range(&inode->i_data,
+				round_down(destoff, PAGE_SIZE),
+				round_up(destoff + len, PAGE_SIZE) - 1);
+
+	return ret;
+}
+
+static int btrfs_remap_file_range_prep(struct file *file_in, loff_t pos_in,
+				       struct file *file_out, loff_t pos_out,
+				       loff_t *len, unsigned int remap_flags)
+{
+	struct inode *inode_in = file_inode(file_in);
+	struct inode *inode_out = file_inode(file_out);
+	u64 bs = BTRFS_I(inode_out)->root->fs_info->sb->s_blocksize;
+	bool same_inode = inode_out == inode_in;
+	u64 wb_len;
+	int ret;
+
+	if (!(remap_flags & REMAP_FILE_DEDUP)) {
+		struct btrfs_root *root_out = BTRFS_I(inode_out)->root;
+
+		if (btrfs_root_readonly(root_out))
+			return -EROFS;
+
+		if (file_in->f_path.mnt != file_out->f_path.mnt ||
+		    inode_in->i_sb != inode_out->i_sb)
+			return -EXDEV;
+	}
+
+	/* Don't make the dst file partly checksummed */
+	if ((BTRFS_I(inode_in)->flags & BTRFS_INODE_NODATASUM) !=
+	    (BTRFS_I(inode_out)->flags & BTRFS_INODE_NODATASUM)) {
+		return -EINVAL;
+	}
+
+	/*
+	 * Now that the inodes are locked, we need to start writeback ourselves
+	 * and can not rely on the writeback from the VFS's generic helper
+	 * generic_remap_file_range_prep() because:
+	 *
+	 * 1) For compression we must call filemap_fdatawrite_range() range
+	 *    twice (btrfs_fdatawrite_range() does it for us), and the generic
+	 *    helper only calls it once;
+	 *
+	 * 2) filemap_fdatawrite_range(), called by the generic helper only
+	 *    waits for the writeback to complete, i.e. for IO to be done, and
+	 *    not for the ordered extents to complete. We need to wait for them
+	 *    to complete so that new file extent items are in the fs tree.
+	 */
+	if (*len == 0 && !(remap_flags & REMAP_FILE_DEDUP))
+		wb_len = ALIGN(inode_in->i_size, bs) - ALIGN_DOWN(pos_in, bs);
+	else
+		wb_len = ALIGN(*len, bs);
+
+	/*
+	 * Since we don't lock ranges, wait for ongoing lockless dio writes (as
+	 * any in progress could create its ordered extents after we wait for
+	 * existing ordered extents below).
+	 */
+	inode_dio_wait(inode_in);
+	if (!same_inode)
+		inode_dio_wait(inode_out);
+
+	/*
+	 * Workaround to make sure NOCOW buffered write reach disk as NOCOW.
+	 *
+	 * Btrfs' back references do not have a block level granularity, they
+	 * work at the whole extent level.
+	 * NOCOW buffered write without data space reserved may not be able
+	 * to fall back to CoW due to lack of data space, thus could cause
+	 * data loss.
+	 *
+	 * Here we take a shortcut by flushing the whole inode, so that all
+	 * nocow write should reach disk as nocow before we increase the
+	 * reference of the extent. We could do better by only flushing NOCOW
+	 * data, but that needs extra accounting.
+	 *
+	 * Also we don't need to check ASYNC_EXTENT, as async extent will be
+	 * CoWed anyway, not affecting nocow part.
+	 */
+	ret = filemap_flush(inode_in->i_mapping);
+	if (ret < 0)
+		return ret;
+
+	ret = btrfs_wait_ordered_range(inode_in, ALIGN_DOWN(pos_in, bs),
+				       wb_len);
+	if (ret < 0)
+		return ret;
+	ret = btrfs_wait_ordered_range(inode_out, ALIGN_DOWN(pos_out, bs),
+				       wb_len);
+	if (ret < 0)
+		return ret;
+
+	return generic_remap_file_range_prep(file_in, pos_in, file_out, pos_out,
+					    len, remap_flags);
+}
+
+loff_t btrfs_remap_file_range(struct file *src_file, loff_t off,
+		struct file *dst_file, loff_t destoff, loff_t len,
+		unsigned int remap_flags)
+{
+	struct inode *src_inode = file_inode(src_file);
+	struct inode *dst_inode = file_inode(dst_file);
+	bool same_inode = dst_inode == src_inode;
+	int ret;
+
+	if (remap_flags & ~(REMAP_FILE_DEDUP | REMAP_FILE_ADVISORY))
+		return -EINVAL;
+
+	if (same_inode)
+		inode_lock(src_inode);
+	else
+		lock_two_nondirectories(src_inode, dst_inode);
+
+	ret = btrfs_remap_file_range_prep(src_file, off, dst_file, destoff,
+					  &len, remap_flags);
+	if (ret < 0 || len == 0)
+		goto out_unlock;
+
+	if (remap_flags & REMAP_FILE_DEDUP)
+		ret = btrfs_extent_same(src_inode, off, len, dst_inode, destoff);
+	else
+		ret = btrfs_clone_files(dst_file, src_file, off, len, destoff);
+
+out_unlock:
+	if (same_inode)
+		inode_unlock(src_inode);
+	else
+		unlock_two_nondirectories(src_inode, dst_inode);
+
+	return ret < 0 ? ret : len;
+}
diff --git a/fs/btrfs/reflink.h b/fs/btrfs/reflink.h
new file mode 100644
index 000000000000..ecb309b4dad0
--- /dev/null
+++ b/fs/btrfs/reflink.h
@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef BTRFS_REFLINK_H
+#define BTRFS_REFLINK_H
+
+#include <linux/fs.h>
+
+loff_t btrfs_remap_file_range(struct file *file_in, loff_t pos_in,
+			      struct file *file_out, loff_t pos_out,
+			      loff_t len, unsigned int remap_flags);
+
+#endif /* BTRFS_REFLINK_H */
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 5cd42b66818c..03bc7134e8cb 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -9,6 +9,7 @@
 #include <linux/blkdev.h>
 #include <linux/rbtree.h>
 #include <linux/slab.h>
+#include <linux/error-injection.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -22,6 +23,54 @@
 #include "print-tree.h"
 #include "delalloc-space.h"
 #include "block-group.h"
+#include "backref.h"
+
+/*
+ * Relocation overview
+ *
+ * [What does relocation do]
+ *
+ * The objective of relocation is to relocate all extents of the target block
+ * group to other block groups.
+ * This is utilized by resize (shrink only), profile converting, compacting
+ * space, or balance routine to spread chunks over devices.
+ *
+ * 		Before		|		After
+ * ------------------------------------------------------------------
+ *  BG A: 10 data extents	| BG A: deleted
+ *  BG B:  2 data extents	| BG B: 10 data extents (2 old + 8 relocated)
+ *  BG C:  1 extents		| BG C:  3 data extents (1 old + 2 relocated)
+ *
+ * [How does relocation work]
+ *
+ * 1.   Mark the target block group read-only
+ *      New extents won't be allocated from the target block group.
+ *
+ * 2.1  Record each extent in the target block group
+ *      To build a proper map of extents to be relocated.
+ *
+ * 2.2  Build data reloc tree and reloc trees
+ *      Data reloc tree will contain an inode, recording all newly relocated
+ *      data extents.
+ *      There will be only one data reloc tree for one data block group.
+ *
+ *      Reloc tree will be a special snapshot of its source tree, containing
+ *      relocated tree blocks.
+ *      Each tree referring to a tree block in target block group will get its
+ *      reloc tree built.
+ *
+ * 2.3  Swap source tree with its corresponding reloc tree
+ *      Each involved tree only refers to new extents after swap.
+ *
+ * 3.   Cleanup reloc trees and data reloc tree.
+ *      As old extents in the target block group are still referenced by reloc
+ *      trees, we need to clean them up before really freeing the target block
+ *      group.
+ *
+ * The main complexity is in steps 2.2 and 2.3.
+ *
+ * The entry point of relocation is relocate_block_group() function.
+ */
 
 /*
  * backref_node, mapping_node and tree_block start with this
@@ -147,7 +196,7 @@ struct file_extent_cluster {
 
 struct reloc_control {
 	/* block group to relocate */
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	/* extent tree */
 	struct btrfs_root *extent_root;
 	/* inode for moving data */
@@ -256,6 +305,7 @@ static void free_backref_node(struct backref_cache *cache,
 {
 	if (node) {
 		cache->nr_nodes--;
+		btrfs_put_root(node->root);
 		kfree(node);
 	}
 }
@@ -517,6 +567,34 @@ static int update_backref_cache(struct btrfs_trans_handle *trans,
 	return 1;
 }
 
+static bool reloc_root_is_dead(struct btrfs_root *root)
+{
+	/*
+	 * Pair with set_bit/clear_bit in clean_dirty_subvols and
+	 * btrfs_update_reloc_root. We need to see the updated bit before
+	 * trying to access reloc_root
+	 */
+	smp_rmb();
+	if (test_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state))
+		return true;
+	return false;
+}
+
+/*
+ * Check if this subvolume tree has valid reloc tree.
+ *
+ * Reloc tree after swap is considered dead, thus not considered as valid.
+ * This is enough for most callers, as they don't distinguish dead reloc root
+ * from no reloc root.  But should_ignore_root() below is a special case.
+ */
+static bool have_reloc_root(struct btrfs_root *root)
+{
+	if (reloc_root_is_dead(root))
+		return false;
+	if (!root->reloc_root)
+		return false;
+	return true;
+}
 
 static int should_ignore_root(struct btrfs_root *root)
 {
@@ -525,12 +603,16 @@ static int should_ignore_root(struct btrfs_root *root)
 	if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
 		return 0;
 
+	/* This root has been merged with its reloc tree, we can ignore it */
+	if (reloc_root_is_dead(root))
+		return 1;
+
 	reloc_root = root->reloc_root;
 	if (!reloc_root)
 		return 0;
 
-	if (btrfs_root_last_snapshot(&reloc_root->root_item) ==
-	    root->fs_info->running_transaction->transid - 1)
+	if (btrfs_header_generation(reloc_root->commit_root) ==
+	    root->fs_info->running_transaction->transid)
 		return 0;
 	/*
 	 * if there is reloc tree and it was created in previous
@@ -557,22 +639,7 @@ static struct btrfs_root *find_reloc_root(struct reloc_control *rc,
 		root = (struct btrfs_root *)node->data;
 	}
 	spin_unlock(&rc->reloc_root_tree.lock);
-	return root;
-}
-
-static int is_cowonly_root(u64 root_objectid)
-{
-	if (root_objectid == BTRFS_ROOT_TREE_OBJECTID ||
-	    root_objectid == BTRFS_EXTENT_TREE_OBJECTID ||
-	    root_objectid == BTRFS_CHUNK_TREE_OBJECTID ||
-	    root_objectid == BTRFS_DEV_TREE_OBJECTID ||
-	    root_objectid == BTRFS_TREE_LOG_OBJECTID ||
-	    root_objectid == BTRFS_CSUM_TREE_OBJECTID ||
-	    root_objectid == BTRFS_UUID_TREE_OBJECTID ||
-	    root_objectid == BTRFS_QUOTA_TREE_OBJECTID ||
-	    root_objectid == BTRFS_FREE_SPACE_TREE_OBJECTID)
-		return 1;
-	return 0;
+	return btrfs_grab_root(root);
 }
 
 static struct btrfs_root *read_fs_root(struct btrfs_fs_info *fs_info,
@@ -582,10 +649,7 @@ static struct btrfs_root *read_fs_root(struct btrfs_fs_info *fs_info,
 
 	key.objectid = root_objectid;
 	key.type = BTRFS_ROOT_ITEM_KEY;
-	if (is_cowonly_root(root_objectid))
-		key.offset = 0;
-	else
-		key.offset = (u64)-1;
+	key.offset = (u64)-1;
 
 	return btrfs_get_fs_root(fs_info, &key, false);
 }
@@ -679,8 +743,6 @@ struct backref_node *build_backref_tree(struct reloc_control *rc,
 		err = -ENOMEM;
 		goto out;
 	}
-	path1->reada = READA_FORWARD;
-	path2->reada = READA_FORWARD;
 
 	node = alloc_backref_node(cache);
 	if (!node) {
@@ -867,10 +929,12 @@ again:
 			/* tree root */
 			ASSERT(btrfs_root_bytenr(&root->root_item) ==
 			       cur->bytenr);
-			if (should_ignore_root(root))
+			if (should_ignore_root(root)) {
+				btrfs_put_root(root);
 				list_add(&cur->list, &useless);
-			else
+			} else {
 				cur->root = root;
+			}
 			break;
 		}
 
@@ -883,6 +947,7 @@ again:
 		ret = btrfs_search_slot(NULL, root, node_key, path2, 0, 0);
 		path2->lowest_level = 0;
 		if (ret < 0) {
+			btrfs_put_root(root);
 			err = ret;
 			goto out;
 		}
@@ -898,6 +963,7 @@ again:
 				  root->root_key.objectid,
 				  node_key->objectid, node_key->type,
 				  node_key->offset);
+			btrfs_put_root(root);
 			err = -ENOENT;
 			goto out;
 		}
@@ -909,15 +975,18 @@ again:
 			if (!path2->nodes[level]) {
 				ASSERT(btrfs_root_bytenr(&root->root_item) ==
 				       lower->bytenr);
-				if (should_ignore_root(root))
+				if (should_ignore_root(root)) {
+					btrfs_put_root(root);
 					list_add(&lower->list, &useless);
-				else
+				} else {
 					lower->root = root;
+				}
 				break;
 			}
 
 			edge = alloc_backref_edge(cache);
 			if (!edge) {
+				btrfs_put_root(root);
 				err = -ENOMEM;
 				goto out;
 			}
@@ -927,6 +996,7 @@ again:
 			if (!rb_node) {
 				upper = alloc_backref_node(cache);
 				if (!upper) {
+					btrfs_put_root(root);
 					free_backref_edge(cache, edge);
 					err = -ENOMEM;
 					goto out;
@@ -974,8 +1044,10 @@ again:
 			edge->node[LOWER] = lower;
 			edge->node[UPPER] = upper;
 
-			if (rb_node)
+			if (rb_node) {
+				btrfs_put_root(root);
 				break;
+			}
 			lower = upper;
 			upper = NULL;
 		}
@@ -1154,7 +1226,7 @@ out:
 			free_backref_node(cache, lower);
 		}
 
-		free_backref_node(cache, node);
+		remove_backref_node(cache, node);
 		return ERR_PTR(err);
 	}
 	ASSERT(!node || !node->detached);
@@ -1212,7 +1284,8 @@ static int clone_backref_node(struct btrfs_trans_handle *trans,
 	new_node->level = node->level;
 	new_node->lowest = node->lowest;
 	new_node->checked = 1;
-	new_node->root = dest;
+	new_node->root = btrfs_grab_root(dest);
+	ASSERT(new_node->root);
 
 	if (!node->lowest) {
 		list_for_each_entry(edge, &node->lower, list[UPPER]) {
@@ -1266,7 +1339,7 @@ static int __must_check __add_reloc_root(struct btrfs_root *root)
 	if (!node)
 		return -ENOMEM;
 
-	node->bytenr = root->node->start;
+	node->bytenr = root->commit_root->start;
 	node->data = root;
 
 	spin_lock(&rc->reloc_root_tree.lock);
@@ -1293,14 +1366,16 @@ static void __del_reloc_root(struct btrfs_root *root)
 	struct rb_node *rb_node;
 	struct mapping_node *node = NULL;
 	struct reloc_control *rc = fs_info->reloc_ctl;
+	bool put_ref = false;
 
 	if (rc && root->node) {
 		spin_lock(&rc->reloc_root_tree.lock);
 		rb_node = tree_search(&rc->reloc_root_tree.rb_root,
-				      root->node->start);
+				      root->commit_root->start);
 		if (rb_node) {
 			node = rb_entry(rb_node, struct mapping_node, rb_node);
 			rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root);
+			RB_CLEAR_NODE(&node->rb_node);
 		}
 		spin_unlock(&rc->reloc_root_tree.lock);
 		if (!node)
@@ -1308,9 +1383,22 @@ static void __del_reloc_root(struct btrfs_root *root)
 		BUG_ON((struct btrfs_root *)node->data != root);
 	}
 
+	/*
+	 * We only put the reloc root here if it's on the list.  There's a lot
+	 * of places where the pattern is to splice the rc->reloc_roots, process
+	 * the reloc roots, and then add the reloc root back onto
+	 * rc->reloc_roots.  If we call __del_reloc_root while it's off of the
+	 * list we don't want the reference being dropped, because the guy
+	 * messing with the list is in charge of the reference.
+	 */
 	spin_lock(&fs_info->trans_lock);
-	list_del_init(&root->root_list);
+	if (!list_empty(&root->root_list)) {
+		put_ref = true;
+		list_del_init(&root->root_list);
+	}
 	spin_unlock(&fs_info->trans_lock);
+	if (put_ref)
+		btrfs_put_root(root);
 	kfree(node);
 }
 
@@ -1318,7 +1406,7 @@ static void __del_reloc_root(struct btrfs_root *root)
  * helper to update the 'address of tree root -> reloc tree'
  * mapping
  */
-static int __update_reloc_root(struct btrfs_root *root, u64 new_bytenr)
+static int __update_reloc_root(struct btrfs_root *root)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct rb_node *rb_node;
@@ -1327,7 +1415,7 @@ static int __update_reloc_root(struct btrfs_root *root, u64 new_bytenr)
 
 	spin_lock(&rc->reloc_root_tree.lock);
 	rb_node = tree_search(&rc->reloc_root_tree.rb_root,
-			      root->node->start);
+			      root->commit_root->start);
 	if (rb_node) {
 		node = rb_entry(rb_node, struct mapping_node, rb_node);
 		rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root);
@@ -1339,7 +1427,7 @@ static int __update_reloc_root(struct btrfs_root *root, u64 new_bytenr)
 	BUG_ON((struct btrfs_root *)node->data != root);
 
 	spin_lock(&rc->reloc_root_tree.lock);
-	node->bytenr = new_bytenr;
+	node->bytenr = root->node->start;
 	rb_node = tree_insert(&rc->reloc_root_tree.rb_root,
 			      node->bytenr, &node->rb_node);
 	spin_unlock(&rc->reloc_root_tree.lock);
@@ -1415,8 +1503,9 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans,
 	BUG_ON(ret);
 	kfree(root_item);
 
-	reloc_root = btrfs_read_fs_root(fs_info->tree_root, &root_key);
+	reloc_root = btrfs_read_tree_root(fs_info->tree_root, &root_key);
 	BUG_ON(IS_ERR(reloc_root));
+	set_bit(BTRFS_ROOT_REF_COWS, &reloc_root->state);
 	reloc_root->last_trans = trans->transid;
 	return reloc_root;
 }
@@ -1424,6 +1513,9 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans,
 /*
  * create reloc tree for a given fs tree. reloc tree is just a
  * snapshot of the fs tree with special root objectid.
+ *
+ * The reloc_root comes out of here with two references, one for
+ * root->reloc_root, and another for being on the rc->reloc_roots list.
  */
 int btrfs_init_reloc_root(struct btrfs_trans_handle *trans,
 			  struct btrfs_root *root)
@@ -1435,20 +1527,35 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans,
 	int clear_rsv = 0;
 	int ret;
 
+	if (!rc)
+		return 0;
+
 	/*
 	 * The subvolume has reloc tree but the swap is finished, no need to
 	 * create/update the dead reloc tree
 	 */
-	if (test_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state))
+	if (reloc_root_is_dead(root))
 		return 0;
 
+	/*
+	 * This is subtle but important.  We do not do
+	 * record_root_in_transaction for reloc roots, instead we record their
+	 * corresponding fs root, and then here we update the last trans for the
+	 * reloc root.  This means that we have to do this for the entire life
+	 * of the reloc root, regardless of which stage of the relocation we are
+	 * in.
+	 */
 	if (root->reloc_root) {
 		reloc_root = root->reloc_root;
 		reloc_root->last_trans = trans->transid;
 		return 0;
 	}
 
-	if (!rc || !rc->create_reloc_tree ||
+	/*
+	 * We are merging reloc roots, we do not need new reloc trees.  Also
+	 * reloc trees never need their own reloc tree.
+	 */
+	if (!rc->create_reloc_tree ||
 	    root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
 		return 0;
 
@@ -1463,7 +1570,7 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans,
 
 	ret = __add_reloc_root(reloc_root);
 	BUG_ON(ret < 0);
-	root->reloc_root = reloc_root;
+	root->reloc_root = btrfs_grab_root(reloc_root);
 	return 0;
 }
 
@@ -1478,21 +1585,33 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans,
 	struct btrfs_root_item *root_item;
 	int ret;
 
-	if (test_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state) ||
-	    !root->reloc_root)
+	if (!have_reloc_root(root))
 		goto out;
 
 	reloc_root = root->reloc_root;
 	root_item = &reloc_root->root_item;
 
+	/*
+	 * We are probably ok here, but __del_reloc_root() will drop its ref of
+	 * the root.  We have the ref for root->reloc_root, but just in case
+	 * hold it while we update the reloc root.
+	 */
+	btrfs_grab_root(reloc_root);
+
 	/* root->reloc_root will stay until current relocation finished */
 	if (fs_info->reloc_ctl->merge_reloc_tree &&
 	    btrfs_root_refs(root_item) == 0) {
 		set_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
+		/*
+		 * Mark the tree as dead before we change reloc_root so
+		 * have_reloc_root will not touch it from now on.
+		 */
+		smp_wmb();
 		__del_reloc_root(reloc_root);
 	}
 
 	if (reloc_root->commit_root != reloc_root->node) {
+		__update_reloc_root(reloc_root);
 		btrfs_set_root_node(root_item, reloc_root->node);
 		free_extent_buffer(reloc_root->commit_root);
 		reloc_root->commit_root = btrfs_root_node(reloc_root);
@@ -1501,7 +1620,7 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans,
 	ret = btrfs_update_root(trans, fs_info->tree_root,
 				&reloc_root->root_key, root_item);
 	BUG_ON(ret);
-
+	btrfs_put_root(reloc_root);
 out:
 	return 0;
 }
@@ -1560,11 +1679,10 @@ again:
 	return NULL;
 }
 
-static int in_block_group(u64 bytenr,
-			  struct btrfs_block_group_cache *block_group)
+static int in_block_group(u64 bytenr, struct btrfs_block_group *block_group)
 {
-	if (bytenr >= block_group->key.objectid &&
-	    bytenr < block_group->key.objectid + block_group->key.offset)
+	if (bytenr >= block_group->start &&
+	    bytenr < block_group->start + block_group->length)
 		return 1;
 	return 0;
 }
@@ -2176,7 +2294,7 @@ static void insert_dirty_subvol(struct btrfs_trans_handle *trans,
 	btrfs_update_reloc_root(trans, root);
 
 	if (list_empty(&root->reloc_dirty_list)) {
-		btrfs_grab_fs_root(root);
+		btrfs_grab_root(root);
 		list_add_tail(&root->reloc_dirty_list, &rc->dirty_subvol_roots);
 	}
 }
@@ -2196,19 +2314,34 @@ static int clean_dirty_subvols(struct reloc_control *rc)
 
 			list_del_init(&root->reloc_dirty_list);
 			root->reloc_root = NULL;
+			/*
+			 * Need barrier to ensure clear_bit() only happens after
+			 * root->reloc_root = NULL. Pairs with have_reloc_root.
+			 */
+			smp_wmb();
+			clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
 			if (reloc_root) {
-
-				ret2 = btrfs_drop_snapshot(reloc_root, NULL, 0, 1);
-				if (ret2 < 0 && !ret)
-					ret = ret2;
+				/*
+				 * btrfs_drop_snapshot drops our ref we hold for
+				 * ->reloc_root.  If it fails however we must
+				 * drop the ref ourselves.
+				 */
+				ret2 = btrfs_drop_snapshot(reloc_root, 0, 1);
+				if (ret2 < 0) {
+					btrfs_put_root(reloc_root);
+					if (!ret)
+						ret = ret2;
+				}
 			}
-			clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
-			btrfs_put_fs_root(root);
+			btrfs_put_root(root);
 		} else {
 			/* Orphan reloc tree, just clean it up */
-			ret2 = btrfs_drop_snapshot(root, NULL, 0, 1);
-			if (ret2 < 0 && !ret)
-				ret = ret2;
+			ret2 = btrfs_drop_snapshot(root, 0, 1);
+			if (ret2 < 0) {
+				btrfs_put_root(root);
+				if (!ret)
+					ret = ret2;
+			}
 		}
 	}
 	return ret;
@@ -2246,7 +2379,7 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
 
 	if (btrfs_disk_key_objectid(&root_item->drop_progress) == 0) {
 		level = btrfs_root_level(root_item);
-		extent_buffer_get(reloc_root->node);
+		atomic_inc(&reloc_root->node->refs);
 		path->nodes[level] = reloc_root->node;
 		path->slots[level] = 0;
 	} else {
@@ -2285,6 +2418,18 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
 			trans = NULL;
 			goto out;
 		}
+
+		/*
+		 * At this point we no longer have a reloc_control, so we can't
+		 * depend on btrfs_init_reloc_root to update our last_trans.
+		 *
+		 * But that's ok, we started the trans handle on our
+		 * corresponding fs_root, which means it's been added to the
+		 * dirty list.  At commit time we'll still call
+		 * btrfs_update_reloc_root() and update our root item
+		 * appropriately.
+		 */
+		reloc_root->last_trans = trans->transid;
 		trans->block_rsv = rc->block_rsv;
 
 		replaced = 0;
@@ -2395,7 +2540,7 @@ again:
 	if (IS_ERR(trans)) {
 		if (!err)
 			btrfs_block_rsv_release(fs_info, rc->block_rsv,
-						num_bytes);
+						num_bytes, NULL);
 		return PTR_ERR(trans);
 	}
 
@@ -2403,7 +2548,7 @@ again:
 		if (num_bytes != rc->merging_rsv_size) {
 			btrfs_end_transaction(trans);
 			btrfs_block_rsv_release(fs_info, rc->block_rsv,
-						num_bytes);
+						num_bytes, NULL);
 			goto again;
 		}
 	}
@@ -2428,6 +2573,7 @@ again:
 		btrfs_update_reloc_root(trans, root);
 
 		list_add(&reloc_root->root_list, &reloc_roots);
+		btrfs_put_root(root);
 	}
 
 	list_splice(&reloc_roots, &rc->reloc_roots);
@@ -2448,10 +2594,6 @@ void free_reloc_roots(struct list_head *list)
 		reloc_root = list_entry(list->next, struct btrfs_root,
 					root_list);
 		__del_reloc_root(reloc_root);
-		free_extent_buffer(reloc_root->node);
-		free_extent_buffer(reloc_root->commit_root);
-		reloc_root->node = NULL;
-		reloc_root->commit_root = NULL;
 	}
 }
 
@@ -2489,6 +2631,7 @@ again:
 			BUG_ON(root->reloc_root != reloc_root);
 
 			ret = merge_reloc_root(rc, root);
+			btrfs_put_root(root);
 			if (ret) {
 				if (list_empty(&reloc_root->root_list))
 					list_add_tail(&reloc_root->root_list,
@@ -2521,7 +2664,21 @@ out:
 			free_reloc_roots(&reloc_roots);
 	}
 
-	BUG_ON(!RB_EMPTY_ROOT(&rc->reloc_root_tree.rb_root));
+	/*
+	 * We used to have
+	 *
+	 * BUG_ON(!RB_EMPTY_ROOT(&rc->reloc_root_tree.rb_root));
+	 *
+	 * here, but it's wrong.  If we fail to start the transaction in
+	 * prepare_to_merge() we will have only 0 ref reloc roots, none of which
+	 * have actually been removed from the reloc_root_tree rb tree.  This is
+	 * fine because we're bailing here, and we hold a reference on the root
+	 * for the list that holds it, so these roots will be cleaned up when we
+	 * do the reloc_dirty_list afterwards.  Meanwhile the root->reloc_root
+	 * will be cleaned up on unmount.
+	 *
+	 * The remaining nodes will be cleaned up by free_reloc_control.
+	 */
 }
 
 static void free_block_list(struct rb_root *blocks)
@@ -2540,6 +2697,7 @@ static int record_reloc_root_in_trans(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_fs_info *fs_info = reloc_root->fs_info;
 	struct btrfs_root *root;
+	int ret;
 
 	if (reloc_root->last_trans == trans->transid)
 		return 0;
@@ -2547,8 +2705,10 @@ static int record_reloc_root_in_trans(struct btrfs_trans_handle *trans,
 	root = read_fs_root(fs_info, reloc_root->root_key.offset);
 	BUG_ON(IS_ERR(root));
 	BUG_ON(root->reloc_root != reloc_root);
+	ret = btrfs_record_root_in_trans(trans, root);
+	btrfs_put_root(root);
 
-	return btrfs_record_root_in_trans(trans, root);
+	return ret;
 }
 
 static noinline_for_stack
@@ -2581,7 +2741,9 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,
 			BUG_ON(next->new_bytenr);
 			BUG_ON(!list_empty(&next->list));
 			next->new_bytenr = root->node->start;
-			next->root = root;
+			btrfs_put_root(next->root);
+			next->root = btrfs_grab_root(root);
+			ASSERT(next->root);
 			list_add_tail(&next->list,
 				      &rc->backref_cache.changed);
 			__mark_block_processed(rc, next);
@@ -3000,7 +3162,6 @@ static int get_tree_block_key(struct btrfs_fs_info *fs_info,
 {
 	struct extent_buffer *eb;
 
-	BUG_ON(block->key_ready);
 	eb = read_tree_block(fs_info, block->bytenr, block->key.offset,
 			     block->level, NULL);
 	if (IS_ERR(eb)) {
@@ -3033,6 +3194,14 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans,
 	if (!node)
 		return 0;
 
+	/*
+	 * If we fail here we want to drop our backref_node because we are going
+	 * to start over and regenerate the tree for it.
+	 */
+	ret = reserve_metadata_space(trans, rc, node);
+	if (ret)
+		goto out;
+
 	BUG_ON(node->processed);
 	root = select_one_root(node);
 	if (root == ERR_PTR(-ENOENT)) {
@@ -3040,12 +3209,6 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	if (!root || test_bit(BTRFS_ROOT_REF_COWS, &root->state)) {
-		ret = reserve_metadata_space(trans, rc, node);
-		if (ret)
-			goto out;
-	}
-
 	if (root) {
 		if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) {
 			BUG_ON(node->new_bytenr);
@@ -3053,7 +3216,9 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans,
 			btrfs_record_root_in_trans(trans, root);
 			root = root->reloc_root;
 			node->new_bytenr = root->node->start;
-			node->root = root;
+			btrfs_put_root(node->root);
+			node->root = btrfs_grab_root(root);
+			ASSERT(node->root);
 			list_add_tail(&node->list, &rc->backref_cache.changed);
 		} else {
 			path->lowest_level = node->level;
@@ -3121,9 +3286,8 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans,
 		ret = relocate_tree_block(trans, rc, node, &block->key,
 					  path);
 		if (ret < 0) {
-			if (ret != -EAGAIN || &block->rb_node == rb_first(blocks))
-				err = ret;
-			goto out;
+			err = ret;
+			break;
 		}
 	}
 out:
@@ -3195,7 +3359,6 @@ static noinline_for_stack
 int setup_extent_mapping(struct inode *inode, u64 start, u64 end,
 			 u64 block_start)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
 	struct extent_map *em;
 	int ret = 0;
@@ -3208,7 +3371,6 @@ int setup_extent_mapping(struct inode *inode, u64 start, u64 end,
 	em->len = end + 1 - start;
 	em->block_len = em->len;
 	em->block_start = block_start;
-	em->bdev = fs_info->fs_devices->latest_bdev;
 	set_bit(EXTENT_FLAG_PINNED, &em->flags);
 
 	lock_extent(&BTRFS_I(inode)->io_tree, start, end);
@@ -3226,6 +3388,15 @@ int setup_extent_mapping(struct inode *inode, u64 start, u64 end,
 	return ret;
 }
 
+/*
+ * Allow error injection to test balance cancellation
+ */
+int btrfs_should_cancel_balance(struct btrfs_fs_info *fs_info)
+{
+	return atomic_read(&fs_info->balance_cancel_req);
+}
+ALLOW_ERROR_INJECTION(btrfs_should_cancel_balance, TRUE);
+
 static int relocate_file_extent_cluster(struct inode *inode,
 					struct file_extent_cluster *cluster)
 {
@@ -3347,6 +3518,10 @@ static int relocate_file_extent_cluster(struct inode *inode,
 		btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE);
 		balance_dirty_pages_ratelimited(inode->i_mapping);
 		btrfs_throttle(fs_info);
+		if (btrfs_should_cancel_balance(fs_info)) {
+			ret = -ECANCELED;
+			goto out;
+		}
 	}
 	WARN_ON(nr != cluster->nr);
 out:
@@ -3518,33 +3693,8 @@ out:
 	return ret;
 }
 
-/*
- * helper to check if the block use full backrefs for pointers in it
- */
-static int block_use_full_backref(struct reloc_control *rc,
-				  struct extent_buffer *eb)
-{
-	u64 flags;
-	int ret;
-
-	if (btrfs_header_flag(eb, BTRFS_HEADER_FLAG_RELOC) ||
-	    btrfs_header_backref_rev(eb) < BTRFS_MIXED_BACKREF_REV)
-		return 1;
-
-	ret = btrfs_lookup_extent_info(NULL, rc->extent_root->fs_info,
-				       eb->start, btrfs_header_level(eb), 1,
-				       NULL, &flags);
-	BUG_ON(ret);
-
-	if (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)
-		ret = 1;
-	else
-		ret = 0;
-	return ret;
-}
-
 static int delete_block_group_cache(struct btrfs_fs_info *fs_info,
-				    struct btrfs_block_group_cache *block_group,
+				    struct btrfs_block_group *block_group,
 				    struct inode *inode,
 				    u64 ino)
 {
@@ -3560,7 +3710,7 @@ static int delete_block_group_cache(struct btrfs_fs_info *fs_info,
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 
-	inode = btrfs_iget(fs_info->sb, &key, root, NULL);
+	inode = btrfs_iget(fs_info->sb, &key, root);
 	if (IS_ERR(inode))
 		return -ENOENT;
 
@@ -3586,172 +3736,40 @@ out:
 }
 
 /*
- * helper to add tree blocks for backref of type BTRFS_EXTENT_DATA_REF_KEY
- * this function scans fs tree to find blocks reference the data extent
+ * Locate the free space cache EXTENT_DATA in root tree leaf and delete the
+ * cache inode, to avoid free space cache data extent blocking data relocation.
  */
-static int find_data_references(struct reloc_control *rc,
-				struct btrfs_key *extent_key,
-				struct extent_buffer *leaf,
-				struct btrfs_extent_data_ref *ref,
-				struct rb_root *blocks)
+static int delete_v1_space_cache(struct extent_buffer *leaf,
+				 struct btrfs_block_group *block_group,
+				 u64 data_bytenr)
 {
-	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	struct btrfs_path *path;
-	struct tree_block *block;
-	struct btrfs_root *root;
-	struct btrfs_file_extent_item *fi;
-	struct rb_node *rb_node;
+	u64 space_cache_ino;
+	struct btrfs_file_extent_item *ei;
 	struct btrfs_key key;
-	u64 ref_root;
-	u64 ref_objectid;
-	u64 ref_offset;
-	u32 ref_count;
-	u32 nritems;
-	int err = 0;
-	int added = 0;
-	int counted;
+	bool found = false;
+	int i;
 	int ret;
 
-	ref_root = btrfs_extent_data_ref_root(leaf, ref);
-	ref_objectid = btrfs_extent_data_ref_objectid(leaf, ref);
-	ref_offset = btrfs_extent_data_ref_offset(leaf, ref);
-	ref_count = btrfs_extent_data_ref_count(leaf, ref);
-
-	/*
-	 * This is an extent belonging to the free space cache, lets just delete
-	 * it and redo the search.
-	 */
-	if (ref_root == BTRFS_ROOT_TREE_OBJECTID) {
-		ret = delete_block_group_cache(fs_info, rc->block_group,
-					       NULL, ref_objectid);
-		if (ret != -ENOENT)
-			return ret;
-		ret = 0;
-	}
-
-	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
-	path->reada = READA_FORWARD;
-
-	root = read_fs_root(fs_info, ref_root);
-	if (IS_ERR(root)) {
-		err = PTR_ERR(root);
-		goto out;
-	}
-
-	key.objectid = ref_objectid;
-	key.type = BTRFS_EXTENT_DATA_KEY;
-	if (ref_offset > ((u64)-1 << 32))
-		key.offset = 0;
-	else
-		key.offset = ref_offset;
-
-	path->search_commit_root = 1;
-	path->skip_locking = 1;
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0) {
-		err = ret;
-		goto out;
-	}
-
-	leaf = path->nodes[0];
-	nritems = btrfs_header_nritems(leaf);
-	/*
-	 * the references in tree blocks that use full backrefs
-	 * are not counted in
-	 */
-	if (block_use_full_backref(rc, leaf))
-		counted = 0;
-	else
-		counted = 1;
-	rb_node = tree_search(blocks, leaf->start);
-	if (rb_node) {
-		if (counted)
-			added = 1;
-		else
-			path->slots[0] = nritems;
-	}
-
-	while (ref_count > 0) {
-		while (path->slots[0] >= nritems) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0) {
-				err = ret;
-				goto out;
-			}
-			if (WARN_ON(ret > 0))
-				goto out;
-
-			leaf = path->nodes[0];
-			nritems = btrfs_header_nritems(leaf);
-			added = 0;
-
-			if (block_use_full_backref(rc, leaf))
-				counted = 0;
-			else
-				counted = 1;
-			rb_node = tree_search(blocks, leaf->start);
-			if (rb_node) {
-				if (counted)
-					added = 1;
-				else
-					path->slots[0] = nritems;
-			}
-		}
+	if (btrfs_header_owner(leaf) != BTRFS_ROOT_TREE_OBJECTID)
+		return 0;
 
-		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-		if (WARN_ON(key.objectid != ref_objectid ||
-		    key.type != BTRFS_EXTENT_DATA_KEY))
+	for (i = 0; i < btrfs_header_nritems(leaf); i++) {
+		btrfs_item_key_to_cpu(leaf, &key, i);
+		if (key.type != BTRFS_EXTENT_DATA_KEY)
+			continue;
+		ei = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
+		if (btrfs_file_extent_type(leaf, ei) == BTRFS_FILE_EXTENT_REG &&
+		    btrfs_file_extent_disk_bytenr(leaf, ei) == data_bytenr) {
+			found = true;
+			space_cache_ino = key.objectid;
 			break;
-
-		fi = btrfs_item_ptr(leaf, path->slots[0],
-				    struct btrfs_file_extent_item);
-
-		if (btrfs_file_extent_type(leaf, fi) ==
-		    BTRFS_FILE_EXTENT_INLINE)
-			goto next;
-
-		if (btrfs_file_extent_disk_bytenr(leaf, fi) !=
-		    extent_key->objectid)
-			goto next;
-
-		key.offset -= btrfs_file_extent_offset(leaf, fi);
-		if (key.offset != ref_offset)
-			goto next;
-
-		if (counted)
-			ref_count--;
-		if (added)
-			goto next;
-
-		if (!tree_block_processed(leaf->start, rc)) {
-			block = kmalloc(sizeof(*block), GFP_NOFS);
-			if (!block) {
-				err = -ENOMEM;
-				break;
-			}
-			block->bytenr = leaf->start;
-			btrfs_item_key_to_cpu(leaf, &block->key, 0);
-			block->level = 0;
-			block->key_ready = 1;
-			rb_node = tree_insert(blocks, block->bytenr,
-					      &block->rb_node);
-			if (rb_node)
-				backref_tree_panic(rb_node, -EEXIST,
-						   block->bytenr);
 		}
-		if (counted)
-			added = 1;
-		else
-			path->slots[0] = nritems;
-next:
-		path->slots[0]++;
-
 	}
-out:
-	btrfs_free_path(path);
-	return err;
+	if (!found)
+		return -ENOENT;
+	ret = delete_block_group_cache(leaf->fs_info, block_group, NULL,
+					space_cache_ino);
+	return ret;
 }
 
 /*
@@ -3763,91 +3781,41 @@ int add_data_references(struct reloc_control *rc,
 			struct btrfs_path *path,
 			struct rb_root *blocks)
 {
-	struct btrfs_key key;
-	struct extent_buffer *eb;
-	struct btrfs_extent_data_ref *dref;
-	struct btrfs_extent_inline_ref *iref;
-	unsigned long ptr;
-	unsigned long end;
-	u32 blocksize = rc->extent_root->fs_info->nodesize;
+	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
+	struct ulist *leaves = NULL;
+	struct ulist_iterator leaf_uiter;
+	struct ulist_node *ref_node = NULL;
+	const u32 blocksize = fs_info->nodesize;
 	int ret = 0;
-	int err = 0;
-
-	eb = path->nodes[0];
-	ptr = btrfs_item_ptr_offset(eb, path->slots[0]);
-	end = ptr + btrfs_item_size_nr(eb, path->slots[0]);
-	ptr += sizeof(struct btrfs_extent_item);
 
-	while (ptr < end) {
-		iref = (struct btrfs_extent_inline_ref *)ptr;
-		key.type = btrfs_get_extent_inline_ref_type(eb, iref,
-							BTRFS_REF_TYPE_DATA);
-		if (key.type == BTRFS_SHARED_DATA_REF_KEY) {
-			key.offset = btrfs_extent_inline_ref_offset(eb, iref);
-			ret = __add_tree_block(rc, key.offset, blocksize,
-					       blocks);
-		} else if (key.type == BTRFS_EXTENT_DATA_REF_KEY) {
-			dref = (struct btrfs_extent_data_ref *)(&iref->offset);
-			ret = find_data_references(rc, extent_key,
-						   eb, dref, blocks);
-		} else {
-			ret = -EUCLEAN;
-			btrfs_err(rc->extent_root->fs_info,
-		     "extent %llu slot %d has an invalid inline ref type",
-			     eb->start, path->slots[0]);
-		}
-		if (ret) {
-			err = ret;
-			goto out;
-		}
-		ptr += btrfs_extent_inline_ref_size(key.type);
-	}
-	WARN_ON(ptr > end);
+	btrfs_release_path(path);
+	ret = btrfs_find_all_leafs(NULL, fs_info, extent_key->objectid,
+				   0, &leaves, NULL, true);
+	if (ret < 0)
+		return ret;
 
-	while (1) {
-		cond_resched();
-		eb = path->nodes[0];
-		if (path->slots[0] >= btrfs_header_nritems(eb)) {
-			ret = btrfs_next_leaf(rc->extent_root, path);
-			if (ret < 0) {
-				err = ret;
-				break;
-			}
-			if (ret > 0)
-				break;
-			eb = path->nodes[0];
-		}
+	ULIST_ITER_INIT(&leaf_uiter);
+	while ((ref_node = ulist_next(leaves, &leaf_uiter))) {
+		struct extent_buffer *eb;
 
-		btrfs_item_key_to_cpu(eb, &key, path->slots[0]);
-		if (key.objectid != extent_key->objectid)
+		eb = read_tree_block(fs_info, ref_node->val, 0, 0, NULL);
+		if (IS_ERR(eb)) {
+			ret = PTR_ERR(eb);
 			break;
-
-		if (key.type == BTRFS_SHARED_DATA_REF_KEY) {
-			ret = __add_tree_block(rc, key.offset, blocksize,
-					       blocks);
-		} else if (key.type == BTRFS_EXTENT_DATA_REF_KEY) {
-			dref = btrfs_item_ptr(eb, path->slots[0],
-					      struct btrfs_extent_data_ref);
-			ret = find_data_references(rc, extent_key,
-						   eb, dref, blocks);
-		} else if (unlikely(key.type == BTRFS_EXTENT_REF_V0_KEY)) {
-			btrfs_print_v0_err(eb->fs_info);
-			btrfs_handle_fs_error(eb->fs_info, -EINVAL, NULL);
-			ret = -EINVAL;
-		} else {
-			ret = 0;
 		}
-		if (ret) {
-			err = ret;
+		ret = delete_v1_space_cache(eb, rc->block_group,
+					    extent_key->objectid);
+		free_extent_buffer(eb);
+		if (ret < 0)
+			break;
+		ret = __add_tree_block(rc, ref_node->val, blocksize, blocks);
+		if (ret < 0)
 			break;
-		}
-		path->slots[0]++;
 	}
-out:
-	btrfs_release_path(path);
-	if (err)
+	if (ret < 0)
 		free_block_list(blocks);
-	return err;
+	ulist_free(leaves);
+	return ret;
 }
 
 /*
@@ -3863,7 +3831,7 @@ int find_next_extent(struct reloc_control *rc, struct btrfs_path *path,
 	u64 start, end, last;
 	int ret;
 
-	last = rc->block_group->key.objectid + rc->block_group->key.offset;
+	last = rc->block_group->start + rc->block_group->length;
 	while (1) {
 		cond_resched();
 		if (rc->search_start >= last) {
@@ -3980,7 +3948,7 @@ int prepare_to_relocate(struct reloc_control *rc)
 		return -ENOMEM;
 
 	memset(&rc->cluster, 0, sizeof(rc->cluster));
-	rc->search_start = rc->block_group->key.objectid;
+	rc->search_start = rc->block_group->start;
 	rc->extents_found = 0;
 	rc->nodes_relocated = 0;
 	rc->merging_rsv_size = 0;
@@ -4099,12 +4067,6 @@ restart:
 		if (!RB_EMPTY_ROOT(&blocks)) {
 			ret = relocate_tree_blocks(trans, rc, &blocks);
 			if (ret < 0) {
-				/*
-				 * if we fail to relocate tree blocks, force to update
-				 * backref cache when committing transaction.
-				 */
-				rc->backref_cache.last_trans = trans->transid - 1;
-
 				if (ret != -EAGAIN) {
 					err = ret;
 					break;
@@ -4128,6 +4090,10 @@ restart:
 				break;
 			}
 		}
+		if (btrfs_should_cancel_balance(fs_info)) {
+			err = -ECANCELED;
+			break;
+		}
 	}
 	if (trans && progress && err == -ENOSPC) {
 		ret = btrfs_force_chunk_alloc(trans, rc->block_group->flags);
@@ -4157,15 +4123,23 @@ restart:
 	set_reloc_control(rc);
 
 	backref_cache_cleanup(&rc->backref_cache);
-	btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1);
+	btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1, NULL);
 
+	/*
+	 * Even in the case when the relocation is cancelled, we should all go
+	 * through prepare_to_merge() and merge_reloc_roots().
+	 *
+	 * For error (including cancelled balance), prepare_to_merge() will
+	 * mark all reloc trees orphan, then queue them for cleanup in
+	 * merge_reloc_roots()
+	 */
 	err = prepare_to_merge(rc, err);
 
 	merge_reloc_roots(rc);
 
 	rc->merge_reloc_tree = 0;
 	unset_reloc_control(rc);
-	btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1);
+	btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1, NULL);
 
 	/* get rid of pinned extents */
 	trans = btrfs_join_transaction(rc->extent_root);
@@ -4174,10 +4148,10 @@ restart:
 		goto out_free;
 	}
 	btrfs_commit_transaction(trans);
+out_free:
 	ret = clean_dirty_subvols(rc);
 	if (ret < 0 && !err)
 		err = ret;
-out_free:
 	btrfs_free_block_rsv(fs_info, rc->block_rsv);
 	btrfs_free_path(path);
 	return err;
@@ -4219,7 +4193,7 @@ out:
  */
 static noinline_for_stack
 struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info,
-				 struct btrfs_block_group_cache *group)
+				 struct btrfs_block_group *group)
 {
 	struct inode *inode = NULL;
 	struct btrfs_trans_handle *trans;
@@ -4233,8 +4207,10 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info,
 		return ERR_CAST(root);
 
 	trans = btrfs_start_transaction(root, 6);
-	if (IS_ERR(trans))
+	if (IS_ERR(trans)) {
+		btrfs_put_root(root);
 		return ERR_CAST(trans);
+	}
 
 	err = btrfs_find_free_objectid(root, &objectid);
 	if (err)
@@ -4246,12 +4222,13 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info,
 	key.objectid = objectid;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	inode = btrfs_iget(fs_info->sb, &key, root, NULL);
+	inode = btrfs_iget(fs_info->sb, &key, root);
 	BUG_ON(IS_ERR(inode));
-	BTRFS_I(inode)->index_cnt = group->key.objectid;
+	BTRFS_I(inode)->index_cnt = group->start;
 
 	err = btrfs_orphan_add(trans, BTRFS_I(inode));
 out:
+	btrfs_put_root(root);
 	btrfs_end_transaction(trans);
 	btrfs_btree_balance_dirty(fs_info);
 	if (err) {
@@ -4279,11 +4256,23 @@ static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info)
 	return rc;
 }
 
+static void free_reloc_control(struct reloc_control *rc)
+{
+	struct mapping_node *node, *tmp;
+
+	free_reloc_roots(&rc->reloc_roots);
+	rbtree_postorder_for_each_entry_safe(node, tmp,
+			&rc->reloc_root_tree.rb_root, rb_node)
+		kfree(node);
+
+	kfree(rc);
+}
+
 /*
  * Print the block group being relocated
  */
 static void describe_relocation(struct btrfs_fs_info *fs_info,
-				struct btrfs_block_group_cache *block_group)
+				struct btrfs_block_group *block_group)
 {
 	char buf[128] = {'\0'};
 
@@ -4291,7 +4280,16 @@ static void describe_relocation(struct btrfs_fs_info *fs_info,
 
 	btrfs_info(fs_info,
 		   "relocating block group %llu flags %s",
-		   block_group->key.objectid, buf);
+		   block_group->start, buf);
+}
+
+static const char *stage_to_string(int stage)
+{
+	if (stage == MOVE_DATA_EXTENTS)
+		return "move data extents";
+	if (stage == UPDATE_DATA_PTRS)
+		return "update data pointers";
+	return "unknown";
 }
 
 /*
@@ -4299,7 +4297,7 @@ static void describe_relocation(struct btrfs_fs_info *fs_info,
  */
 int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 {
-	struct btrfs_block_group_cache *bg;
+	struct btrfs_block_group *bg;
 	struct btrfs_root *extent_root = fs_info->extent_root;
 	struct reloc_control *rc;
 	struct inode *inode;
@@ -4326,7 +4324,7 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 	rc->extent_root = extent_root;
 	rc->block_group = bg;
 
-	ret = btrfs_inc_block_group_ro(rc->block_group);
+	ret = btrfs_inc_block_group_ro(rc->block_group, true);
 	if (ret) {
 		err = ret;
 		goto out;
@@ -4364,16 +4362,19 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 	btrfs_wait_block_group_reservations(rc->block_group);
 	btrfs_wait_nocow_writers(rc->block_group);
 	btrfs_wait_ordered_roots(fs_info, U64_MAX,
-				 rc->block_group->key.objectid,
-				 rc->block_group->key.offset);
+				 rc->block_group->start,
+				 rc->block_group->length);
 
 	while (1) {
+		int finishes_stage;
+
 		mutex_lock(&fs_info->cleaner_mutex);
 		ret = relocate_block_group(rc);
 		mutex_unlock(&fs_info->cleaner_mutex);
 		if (ret < 0)
 			err = ret;
 
+		finishes_stage = rc->stage;
 		/*
 		 * We may have gotten ENOSPC after we already dirtied some
 		 * extents.  If writeout happens while we're relocating a
@@ -4399,19 +4400,19 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 		if (rc->extents_found == 0)
 			break;
 
-		btrfs_info(fs_info, "found %llu extents", rc->extents_found);
-
+		btrfs_info(fs_info, "found %llu extents, stage: %s",
+			   rc->extents_found, stage_to_string(finishes_stage));
 	}
 
 	WARN_ON(rc->block_group->pinned > 0);
 	WARN_ON(rc->block_group->reserved > 0);
-	WARN_ON(btrfs_block_group_used(&rc->block_group->item) > 0);
+	WARN_ON(rc->block_group->used > 0);
 out:
 	if (err && rw)
 		btrfs_dec_block_group_ro(rc->block_group);
 	iput(rc->data_inode);
 	btrfs_put_block_group(rc->block_group);
-	kfree(rc);
+	free_reloc_control(rc);
 	return err;
 }
 
@@ -4487,12 +4488,13 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 		    key.type != BTRFS_ROOT_ITEM_KEY)
 			break;
 
-		reloc_root = btrfs_read_fs_root(root, &key);
+		reloc_root = btrfs_read_tree_root(root, &key);
 		if (IS_ERR(reloc_root)) {
 			err = PTR_ERR(reloc_root);
 			goto out;
 		}
 
+		set_bit(BTRFS_ROOT_REF_COWS, &reloc_root->state);
 		list_add(&reloc_root->root_list, &reloc_roots);
 
 		if (btrfs_root_refs(&reloc_root->root_item) > 0) {
@@ -4509,6 +4511,8 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 					err = ret;
 					goto out;
 				}
+			} else {
+				btrfs_put_root(fs_root);
 			}
 		}
 
@@ -4534,9 +4538,8 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 
 	trans = btrfs_join_transaction(rc->extent_root);
 	if (IS_ERR(trans)) {
-		unset_reloc_control(rc);
 		err = PTR_ERR(trans);
-		goto out_free;
+		goto out_unset;
 	}
 
 	rc->merge_reloc_tree = 1;
@@ -4555,17 +4558,20 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 		fs_root = read_fs_root(fs_info, reloc_root->root_key.offset);
 		if (IS_ERR(fs_root)) {
 			err = PTR_ERR(fs_root);
-			goto out_free;
+			list_add_tail(&reloc_root->root_list, &reloc_roots);
+			btrfs_end_transaction(trans);
+			goto out_unset;
 		}
 
 		err = __add_reloc_root(reloc_root);
 		BUG_ON(err < 0); /* -ENOMEM or logic error */
-		fs_root->reloc_root = reloc_root;
+		fs_root->reloc_root = btrfs_grab_root(reloc_root);
+		btrfs_put_root(fs_root);
 	}
 
 	err = btrfs_commit_transaction(trans);
 	if (err)
-		goto out_free;
+		goto out_unset;
 
 	merge_reloc_roots(rc);
 
@@ -4574,15 +4580,16 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 	trans = btrfs_join_transaction(rc->extent_root);
 	if (IS_ERR(trans)) {
 		err = PTR_ERR(trans);
-		goto out_free;
+		goto out_clean;
 	}
 	err = btrfs_commit_transaction(trans);
-
+out_clean:
 	ret = clean_dirty_subvols(rc);
 	if (ret < 0 && !err)
 		err = ret;
-out_free:
-	kfree(rc);
+out_unset:
+	unset_reloc_control(rc);
+	free_reloc_control(rc);
 out:
 	if (!list_empty(&reloc_roots))
 		free_reloc_roots(&reloc_roots);
@@ -4592,10 +4599,12 @@ out:
 	if (err == 0) {
 		/* cleanup orphan inode in data relocation tree */
 		fs_root = read_fs_root(fs_info, BTRFS_DATA_RELOC_TREE_OBJECTID);
-		if (IS_ERR(fs_root))
+		if (IS_ERR(fs_root)) {
 			err = PTR_ERR(fs_root);
-		else
+		} else {
 			err = btrfs_orphan_cleanup(fs_root);
+			btrfs_put_root(fs_root);
+		}
 	}
 	return err;
 }
@@ -4617,7 +4626,7 @@ int btrfs_reloc_clone_csums(struct inode *inode, u64 file_pos, u64 len)
 	LIST_HEAD(list);
 
 	ordered = btrfs_lookup_ordered_extent(inode, file_pos);
-	BUG_ON(ordered->file_offset != file_pos || ordered->len != len);
+	BUG_ON(ordered->file_offset != file_pos || ordered->num_bytes != len);
 
 	disk_bytenr = file_pos + BTRFS_I(inode)->index_cnt;
 	ret = btrfs_lookup_csums_range(fs_info->csum_root, disk_bytenr,
@@ -4641,7 +4650,7 @@ int btrfs_reloc_clone_csums(struct inode *inode, u64 file_pos, u64 len)
 		 * disk_len vs real len like with real inodes since it's all
 		 * disk length.
 		 */
-		new_bytenr = ordered->start + (sums->bytenr - disk_bytenr);
+		new_bytenr = ordered->disk_bytenr + sums->bytenr - disk_bytenr;
 		sums->bytenr = new_bytenr;
 
 		btrfs_add_ordered_sum(ordered, sums);
@@ -4669,11 +4678,6 @@ int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans,
 	BUG_ON(rc->stage == UPDATE_DATA_PTRS &&
 	       root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID);
 
-	if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) {
-		if (buf == root->node)
-			__update_reloc_root(root, cow->start);
-	}
-
 	level = btrfs_header_level(buf);
 	if (btrfs_header_generation(buf) <=
 	    btrfs_root_last_snapshot(&root->root_item))
@@ -4688,7 +4692,7 @@ int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans,
 		       node->new_bytenr != buf->start);
 
 		drop_node_buffer(node);
-		extent_buffer_get(cow);
+		atomic_inc(&cow->refs);
 		node->eb = cow;
 		node->new_bytenr = cow->start;
 
@@ -4720,7 +4724,7 @@ void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending,
 	struct btrfs_root *root = pending->root;
 	struct reloc_control *rc = root->fs_info->reloc_ctl;
 
-	if (!root->reloc_root || !rc)
+	if (!rc || !have_reloc_root(root))
 		return;
 
 	if (!rc->merge_reloc_tree)
@@ -4744,6 +4748,10 @@ void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending,
 /*
  * called after snapshot is created. migrate block reservation
  * and create reloc root for the newly created snapshot
+ *
+ * This is similar to btrfs_init_reloc_root(), we come out of here with two
+ * references held on the reloc_root, one for root->reloc_root and one for
+ * rc->reloc_roots.
  */
 int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 			       struct btrfs_pending_snapshot *pending)
@@ -4754,7 +4762,7 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 	struct reloc_control *rc = root->fs_info->reloc_ctl;
 	int ret;
 
-	if (!root->reloc_root || !rc)
+	if (!rc || !have_reloc_root(root))
 		return 0;
 
 	rc = root->fs_info->reloc_ctl;
@@ -4776,7 +4784,7 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 
 	ret = __add_reloc_root(reloc_root);
 	BUG_ON(ret < 0);
-	new_root->reloc_root = reloc_root;
+	new_root->reloc_root = btrfs_grab_root(reloc_root);
 
 	if (rc->create_reloc_tree)
 		ret = clone_backref_node(trans, rc, root, reloc_root);
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 3b17b647d002..668f22844017 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -22,7 +22,6 @@
 static void btrfs_read_root_item(struct extent_buffer *eb, int slot,
 				struct btrfs_root_item *item)
 {
-	uuid_le uuid;
 	u32 len;
 	int need_reset = 0;
 
@@ -44,8 +43,7 @@ static void btrfs_read_root_item(struct extent_buffer *eb, int slot,
 			sizeof(*item) - offsetof(struct btrfs_root_item,
 					generation_v2));
 
-		uuid_le_gen(&uuid);
-		memcpy(item->uuid, uuid.b, BTRFS_UUID_SIZE);
+		generate_random_guid(item->uuid);
 	}
 }
 
@@ -255,25 +253,7 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 		root_key.objectid = key.offset;
 		key.offset++;
 
-		/*
-		 * The root might have been inserted already, as before we look
-		 * for orphan roots, log replay might have happened, which
-		 * triggers a transaction commit and qgroup accounting, which
-		 * in turn reads and inserts fs roots while doing backref
-		 * walking.
-		 */
-		root = btrfs_lookup_fs_root(fs_info, root_key.objectid);
-		if (root) {
-			WARN_ON(!test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED,
-					  &root->state));
-			if (btrfs_root_refs(&root->root_item) == 0) {
-				set_bit(BTRFS_ROOT_DEAD_TREE, &root->state);
-				btrfs_add_dead_root(root);
-			}
-			continue;
-		}
-
-		root = btrfs_read_fs_root(tree_root, &root_key);
+		root = btrfs_get_fs_root(fs_info, &root_key, false);
 		err = PTR_ERR_OR_ZERO(root);
 		if (err && err != -ENOENT) {
 			break;
@@ -300,25 +280,12 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info)
 			continue;
 		}
 
-		err = btrfs_init_fs_root(root);
-		if (err) {
-			btrfs_free_fs_root(root);
-			break;
-		}
-
-		set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state);
-
-		err = btrfs_insert_fs_root(fs_info, root);
-		if (err) {
-			BUG_ON(err == -EEXIST);
-			btrfs_free_fs_root(root);
-			break;
-		}
-
+		WARN_ON(!test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state));
 		if (btrfs_root_refs(&root->root_item) == 0) {
 			set_bit(BTRFS_ROOT_DEAD_TREE, &root->state);
 			btrfs_add_dead_root(root);
 		}
+		btrfs_put_root(root);
 	}
 
 	btrfs_free_path(path);
@@ -376,11 +343,13 @@ again:
 		leaf = path->nodes[0];
 		ref = btrfs_item_ptr(leaf, path->slots[0],
 				     struct btrfs_root_ref);
-
-		WARN_ON(btrfs_root_ref_dirid(leaf, ref) != dirid);
-		WARN_ON(btrfs_root_ref_name_len(leaf, ref) != name_len);
 		ptr = (unsigned long)(ref + 1);
-		WARN_ON(memcmp_extent_buffer(leaf, name, ptr, name_len));
+		if ((btrfs_root_ref_dirid(leaf, ref) != dirid) ||
+		    (btrfs_root_ref_name_len(leaf, ref) != name_len) ||
+		    memcmp_extent_buffer(leaf, name, ptr, name_len)) {
+			err = -ENOENT;
+			goto out;
+		}
 		*sequence = btrfs_root_ref_sequence(leaf, ref);
 
 		ret = btrfs_del_item(trans, tree_root, path);
@@ -551,5 +520,5 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info,
 				      struct btrfs_block_rsv *rsv)
 {
-	btrfs_block_rsv_release(fs_info, rsv, (u64)-1);
+	btrfs_block_rsv_release(fs_info, rsv, (u64)-1, NULL);
 }
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index f7d4e03f4c5d..adaf8ab694d5 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -8,6 +8,7 @@
 #include <linux/sched/mm.h>
 #include <crypto/hash.h>
 #include "ctree.h"
+#include "discard.h"
 #include "volumes.h"
 #include "disk-io.h"
 #include "ordered-data.h"
@@ -148,7 +149,7 @@ struct scrub_parity {
 	 */
 	unsigned long		*ebitmap;
 
-	unsigned long		bitmap[0];
+	unsigned long		bitmap[];
 };
 
 struct scrub_ctx {
@@ -389,8 +390,7 @@ static struct full_stripe_lock *search_full_stripe_lock(
  *
  * Caller must ensure @cache is a RAID56 block group.
  */
-static u64 get_full_stripe_logical(struct btrfs_block_group_cache *cache,
-				   u64 bytenr)
+static u64 get_full_stripe_logical(struct btrfs_block_group *cache, u64 bytenr)
 {
 	u64 ret;
 
@@ -404,8 +404,8 @@ static u64 get_full_stripe_logical(struct btrfs_block_group_cache *cache,
 	 * round_down() can only handle power of 2, while RAID56 full
 	 * stripe length can be 64KiB * n, so we need to manually round down.
 	 */
-	ret = div64_u64(bytenr - cache->key.objectid, cache->full_stripe_len) *
-		cache->full_stripe_len + cache->key.objectid;
+	ret = div64_u64(bytenr - cache->start, cache->full_stripe_len) *
+			cache->full_stripe_len + cache->start;
 	return ret;
 }
 
@@ -423,7 +423,7 @@ static u64 get_full_stripe_logical(struct btrfs_block_group_cache *cache,
 static int lock_full_stripe(struct btrfs_fs_info *fs_info, u64 bytenr,
 			    bool *locked_ret)
 {
-	struct btrfs_block_group_cache *bg_cache;
+	struct btrfs_block_group *bg_cache;
 	struct btrfs_full_stripe_locks_tree *locks_root;
 	struct full_stripe_lock *existing;
 	u64 fstripe_start;
@@ -470,7 +470,7 @@ out:
 static int unlock_full_stripe(struct btrfs_fs_info *fs_info, u64 bytenr,
 			      bool locked)
 {
-	struct btrfs_block_group_cache *bg_cache;
+	struct btrfs_block_group *bg_cache;
 	struct btrfs_full_stripe_locks_tree *locks_root;
 	struct full_stripe_lock *fstripe_lock;
 	u64 fstripe_start;
@@ -598,8 +598,8 @@ static noinline_for_stack struct scrub_ctx *scrub_setup_ctx(
 		sbio->index = i;
 		sbio->sctx = sctx;
 		sbio->page_count = 0;
-		btrfs_init_work(&sbio->work, btrfs_scrub_helper,
-				scrub_bio_end_io_worker, NULL, NULL);
+		btrfs_init_work(&sbio->work, scrub_bio_end_io_worker, NULL,
+				NULL);
 
 		if (i != SCRUB_BIOS_PER_SCTX - 1)
 			sctx->bios[i]->next_free = i + 1;
@@ -653,7 +653,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
 	root_key.objectid = root;
 	root_key.type = BTRFS_ROOT_ITEM_KEY;
 	root_key.offset = (u64)-1;
-	local_root = btrfs_read_fs_root_no_name(fs_info, &root_key);
+	local_root = btrfs_get_fs_root(fs_info, &root_key, true);
 	if (IS_ERR(local_root)) {
 		ret = PTR_ERR(local_root);
 		goto err;
@@ -668,6 +668,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
 
 	ret = btrfs_search_slot(NULL, local_root, &key, swarn->path, 0, 0);
 	if (ret) {
+		btrfs_put_root(local_root);
 		btrfs_release_path(swarn->path);
 		goto err;
 	}
@@ -688,6 +689,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
 	ipath = init_ipath(4096, local_root, swarn->path);
 	memalloc_nofs_restore(nofs_flag);
 	if (IS_ERR(ipath)) {
+		btrfs_put_root(local_root);
 		ret = PTR_ERR(ipath);
 		ipath = NULL;
 		goto err;
@@ -711,6 +713,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
 				  min(isize - offset, (u64)PAGE_SIZE), nlink,
 				  (char *)(unsigned long)ipath->fspath->val[i]);
 
+	btrfs_put_root(local_root);
 	free_ipath(ipath);
 	return 0;
 
@@ -1720,8 +1723,7 @@ static void scrub_wr_bio_end_io(struct bio *bio)
 	sbio->status = bio->bi_status;
 	sbio->bio = bio;
 
-	btrfs_init_work(&sbio->work, btrfs_scrubwrc_helper,
-			 scrub_wr_bio_end_io_worker, NULL, NULL);
+	btrfs_init_work(&sbio->work, scrub_wr_bio_end_io_worker, NULL, NULL);
 	btrfs_queue_work(fs_info->scrub_wr_completion_workers, &sbio->work);
 }
 
@@ -2149,14 +2151,13 @@ static void scrub_missing_raid56_worker(struct btrfs_work *work)
 		scrub_write_block_to_dev_replace(sblock);
 	}
 
-	scrub_block_put(sblock);
-
 	if (sctx->is_dev_replace && sctx->flush_all_writes) {
 		mutex_lock(&sctx->wr_lock);
 		scrub_wr_submit(sctx);
 		mutex_unlock(&sctx->wr_lock);
 	}
 
+	scrub_block_put(sblock);
 	scrub_pending_bio_dec(sctx);
 }
 
@@ -2204,8 +2205,7 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock)
 		raid56_add_scrub_pages(rbio, spage->page, spage->logical);
 	}
 
-	btrfs_init_work(&sblock->work, btrfs_scrub_helper,
-			scrub_missing_raid56_worker, NULL, NULL);
+	btrfs_init_work(&sblock->work, scrub_missing_raid56_worker, NULL, NULL);
 	scrub_block_get(sblock);
 	scrub_pending_bio_inc(sctx);
 	raid56_submit_missing_rbio(rbio);
@@ -2743,8 +2743,8 @@ static void scrub_parity_bio_endio(struct bio *bio)
 
 	bio_put(bio);
 
-	btrfs_init_work(&sparity->work, btrfs_scrubparity_helper,
-			scrub_parity_bio_endio_worker, NULL, NULL);
+	btrfs_init_work(&sparity->work, scrub_parity_bio_endio_worker, NULL,
+			NULL);
 	btrfs_queue_work(fs_info->scrub_parity_workers, &sparity->work);
 }
 
@@ -3420,7 +3420,7 @@ static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx,
 					  struct btrfs_device *scrub_dev,
 					  u64 chunk_offset, u64 length,
 					  u64 dev_offset,
-					  struct btrfs_block_group_cache *cache)
+					  struct btrfs_block_group *cache)
 {
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
 	struct extent_map_tree *map_tree = &fs_info->mapping_tree;
@@ -3484,7 +3484,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 	struct extent_buffer *l;
 	struct btrfs_key key;
 	struct btrfs_key found_key;
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
 
 	path = btrfs_alloc_path();
@@ -3563,55 +3563,45 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		 * -> btrfs_scrub_pause()
 		 */
 		scrub_pause_on(fs_info);
-		ret = btrfs_inc_block_group_ro(cache);
-		if (!ret && sctx->is_dev_replace) {
-			/*
-			 * If we are doing a device replace wait for any tasks
-			 * that started delalloc right before we set the block
-			 * group to RO mode, as they might have just allocated
-			 * an extent from it or decided they could do a nocow
-			 * write. And if any such tasks did that, wait for their
-			 * ordered extents to complete and then commit the
-			 * current transaction, so that we can later see the new
-			 * extent items in the extent tree - the ordered extents
-			 * create delayed data references (for cow writes) when
-			 * they complete, which will be run and insert the
-			 * corresponding extent items into the extent tree when
-			 * we commit the transaction they used when running
-			 * inode.c:btrfs_finish_ordered_io(). We later use
-			 * the commit root of the extent tree to find extents
-			 * to copy from the srcdev into the tgtdev, and we don't
-			 * want to miss any new extents.
-			 */
-			btrfs_wait_block_group_reservations(cache);
-			btrfs_wait_nocow_writers(cache);
-			ret = btrfs_wait_ordered_roots(fs_info, U64_MAX,
-						       cache->key.objectid,
-						       cache->key.offset);
-			if (ret > 0) {
-				struct btrfs_trans_handle *trans;
-
-				trans = btrfs_join_transaction(root);
-				if (IS_ERR(trans))
-					ret = PTR_ERR(trans);
-				else
-					ret = btrfs_commit_transaction(trans);
-				if (ret) {
-					scrub_pause_off(fs_info);
-					btrfs_put_block_group(cache);
-					break;
-				}
-			}
-		}
-		scrub_pause_off(fs_info);
 
+		/*
+		 * Don't do chunk preallocation for scrub.
+		 *
+		 * This is especially important for SYSTEM bgs, or we can hit
+		 * -EFBIG from btrfs_finish_chunk_alloc() like:
+		 * 1. The only SYSTEM bg is marked RO.
+		 *    Since SYSTEM bg is small, that's pretty common.
+		 * 2. New SYSTEM bg will be allocated
+		 *    Due to regular version will allocate new chunk.
+		 * 3. New SYSTEM bg is empty and will get cleaned up
+		 *    Before cleanup really happens, it's marked RO again.
+		 * 4. Empty SYSTEM bg get scrubbed
+		 *    We go back to 2.
+		 *
+		 * This can easily boost the amount of SYSTEM chunks if cleaner
+		 * thread can't be triggered fast enough, and use up all space
+		 * of btrfs_super_block::sys_chunk_array
+		 *
+		 * While for dev replace, we need to try our best to mark block
+		 * group RO, to prevent race between:
+		 * - Write duplication
+		 *   Contains latest data
+		 * - Scrub copy
+		 *   Contains data from commit tree
+		 *
+		 * If target block group is not marked RO, nocow writes can
+		 * be overwritten by scrub copy, causing data corruption.
+		 * So for dev-replace, it's not allowed to continue if a block
+		 * group is not RO.
+		 */
+		ret = btrfs_inc_block_group_ro(cache, sctx->is_dev_replace);
 		if (ret == 0) {
 			ro_set = 1;
-		} else if (ret == -ENOSPC) {
+		} else if (ret == -ENOSPC && !sctx->is_dev_replace) {
 			/*
 			 * btrfs_inc_block_group_ro return -ENOSPC when it
 			 * failed in creating new chunk for metadata.
-			 * It is not a problem for scrub/replace, because
+			 * It is not a problem for scrub, because
 			 * metadata are always cowed, and our scrub paused
 			 * commit_transactions.
 			 */
@@ -3620,10 +3610,23 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			btrfs_warn(fs_info,
 				   "failed setting block group ro: %d", ret);
 			btrfs_put_block_group(cache);
+			scrub_pause_off(fs_info);
 			break;
 		}
 
-		down_write(&fs_info->dev_replace.rwsem);
+		/*
+		 * Now the target block is marked RO, wait for nocow writes to
+		 * finish before dev-replace.
+		 * COW is fine, as COW never overwrites extents in commit tree.
+		 */
+		if (sctx->is_dev_replace) {
+			btrfs_wait_nocow_writers(cache);
+			btrfs_wait_ordered_roots(fs_info, U64_MAX, cache->start,
+					cache->length);
+		}
+
+		scrub_pause_off(fs_info);
+		down_write(&dev_replace->rwsem);
 		dev_replace->cursor_right = found_key.offset + length;
 		dev_replace->cursor_left = found_key.offset;
 		dev_replace->item_needs_writeback = 1;
@@ -3664,10 +3667,10 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 
 		scrub_pause_off(fs_info);
 
-		down_write(&fs_info->dev_replace.rwsem);
+		down_write(&dev_replace->rwsem);
 		dev_replace->cursor_left = dev_replace->cursor_right;
 		dev_replace->item_needs_writeback = 1;
-		up_write(&fs_info->dev_replace.rwsem);
+		up_write(&dev_replace->rwsem);
 
 		if (ro_set)
 			btrfs_dec_block_group_ro(cache);
@@ -3681,9 +3684,13 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		 */
 		spin_lock(&cache->lock);
 		if (!cache->removed && !cache->ro && cache->reserved == 0 &&
-		    btrfs_block_group_used(&cache->item) == 0) {
+		    cache->used == 0) {
 			spin_unlock(&cache->lock);
-			btrfs_mark_bg_unused(cache);
+			if (btrfs_test_opt(fs_info, DISCARD_ASYNC))
+				btrfs_discard_queue_work(&fs_info->discard_ctl,
+							 cache);
+			else
+				btrfs_mark_bg_unused(cache);
 		} else {
 			spin_unlock(&cache->lock);
 		}
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 123ac54af071..6a92ecf9eaa2 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -25,6 +25,14 @@
 #include "compression.h"
 
 /*
+ * Maximum number of references an extent can have in order for us to attempt to
+ * issue clone operations instead of write operations. This currently exists to
+ * avoid hitting limitations of the backreference walking code (taking a lot of
+ * time and using too much memory for extents with large number of references).
+ */
+#define SEND_MAX_EXTENT_REFS	64
+
+/*
  * A fs_path is a helper to dynamically build path names with unknown size.
  * It reallocates the internal buffer on demand.
  * It allows fast adding of path elements on the right side (normal path) and
@@ -1248,12 +1256,21 @@ static int __iterate_backrefs(u64 ino, u64 offset, u64 root, void *ctx_)
 	 */
 	if (found->root == bctx->sctx->send_root) {
 		/*
-		 * TODO for the moment we don't accept clones from the inode
-		 * that is currently send. We may change this when
-		 * BTRFS_IOC_CLONE_RANGE supports cloning from and to the same
-		 * file.
+		 * If the source inode was not yet processed we can't issue a
+		 * clone operation, as the source extent does not exist yet at
+		 * the destination of the stream.
+		 */
+		if (ino > bctx->cur_objectid)
+			return 0;
+		/*
+		 * We clone from the inode currently being sent as long as the
+		 * source extent is already processed, otherwise we could try
+		 * to clone from an extent that does not exist yet at the
+		 * destination of the stream.
 		 */
-		if (ino >= bctx->cur_objectid)
+		if (ino == bctx->cur_objectid &&
+		    offset + bctx->extent_len >
+		    bctx->sctx->cur_inode_next_write_offset)
 			return 0;
 	}
 
@@ -1302,6 +1319,7 @@ static int find_extent_clone(struct send_ctx *sctx,
 	struct clone_root *cur_clone_root;
 	struct btrfs_key found_key;
 	struct btrfs_path *tmp_path;
+	struct btrfs_extent_item *ei;
 	int compressed;
 	u32 i;
 
@@ -1349,7 +1367,6 @@ static int find_extent_clone(struct send_ctx *sctx,
 	ret = extent_from_logical(fs_info, disk_byte, tmp_path,
 				  &found_key, &flags);
 	up_read(&fs_info->commit_root_sem);
-	btrfs_release_path(tmp_path);
 
 	if (ret < 0)
 		goto out;
@@ -1358,6 +1375,21 @@ static int find_extent_clone(struct send_ctx *sctx,
 		goto out;
 	}
 
+	ei = btrfs_item_ptr(tmp_path->nodes[0], tmp_path->slots[0],
+			    struct btrfs_extent_item);
+	/*
+	 * Backreference walking (iterate_extent_inodes() below) is currently
+	 * too expensive when an extent has a large number of references, both
+	 * in time spent and used memory. So for now just fallback to write
+	 * operations instead of clone operations when an extent has more than
+	 * a certain amount of references.
+	 */
+	if (btrfs_extent_refs(tmp_path->nodes[0], ei) > SEND_MAX_EXTENT_REFS) {
+		ret = -ENOENT;
+		goto out;
+	}
+	btrfs_release_path(tmp_path);
+
 	/*
 	 * Setup the clone roots.
 	 */
@@ -4779,7 +4811,7 @@ static ssize_t fill_read_buf(struct send_ctx *sctx, u64 offset, u32 len)
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
 
-	inode = btrfs_iget(fs_info->sb, &key, root, NULL);
+	inode = btrfs_iget(fs_info->sb, &key, root);
 	if (IS_ERR(inode))
 		return PTR_ERR(inode);
 
@@ -5554,10 +5586,7 @@ static int get_last_extent(struct send_ctx *sctx, u64 offset)
 {
 	struct btrfs_path *path;
 	struct btrfs_root *root = sctx->send_root;
-	struct btrfs_file_extent_item *fi;
 	struct btrfs_key key;
-	u64 extent_end;
-	u8 type;
 	int ret;
 
 	path = alloc_path_for_send();
@@ -5577,18 +5606,7 @@ static int get_last_extent(struct send_ctx *sctx, u64 offset)
 	if (key.objectid != sctx->cur_ino || key.type != BTRFS_EXTENT_DATA_KEY)
 		goto out;
 
-	fi = btrfs_item_ptr(path->nodes[0], path->slots[0],
-			    struct btrfs_file_extent_item);
-	type = btrfs_file_extent_type(path->nodes[0], fi);
-	if (type == BTRFS_FILE_EXTENT_INLINE) {
-		u64 size = btrfs_file_extent_ram_bytes(path->nodes[0], fi);
-		extent_end = ALIGN(key.offset + size,
-				   sctx->send_root->fs_info->sectorsize);
-	} else {
-		extent_end = key.offset +
-			btrfs_file_extent_num_bytes(path->nodes[0], fi);
-	}
-	sctx->cur_inode_last_extent = extent_end;
+	sctx->cur_inode_last_extent = btrfs_file_extent_end(path);
 out:
 	btrfs_free_path(path);
 	return ret;
@@ -5642,16 +5660,7 @@ static int range_is_hole_in_parent(struct send_ctx *sctx,
 			break;
 
 		fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
-		if (btrfs_file_extent_type(leaf, fi) ==
-		    BTRFS_FILE_EXTENT_INLINE) {
-			u64 size = btrfs_file_extent_ram_bytes(leaf, fi);
-
-			extent_end = ALIGN(key.offset + size,
-					   root->fs_info->sectorsize);
-		} else {
-			extent_end = key.offset +
-				btrfs_file_extent_num_bytes(leaf, fi);
-		}
+		extent_end = btrfs_file_extent_end(path);
 		if (extent_end <= start)
 			goto next;
 		if (btrfs_file_extent_disk_bytenr(leaf, fi) == 0) {
@@ -5672,9 +5681,6 @@ out:
 static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path,
 			   struct btrfs_key *key)
 {
-	struct btrfs_file_extent_item *fi;
-	u64 extent_end;
-	u8 type;
 	int ret = 0;
 
 	if (sctx->cur_ino != key->objectid || !need_send_hole(sctx))
@@ -5686,18 +5692,6 @@ static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path,
 			return ret;
 	}
 
-	fi = btrfs_item_ptr(path->nodes[0], path->slots[0],
-			    struct btrfs_file_extent_item);
-	type = btrfs_file_extent_type(path->nodes[0], fi);
-	if (type == BTRFS_FILE_EXTENT_INLINE) {
-		u64 size = btrfs_file_extent_ram_bytes(path->nodes[0], fi);
-		extent_end = ALIGN(key->offset + size,
-				   sctx->send_root->fs_info->sectorsize);
-	} else {
-		extent_end = key->offset +
-			btrfs_file_extent_num_bytes(path->nodes[0], fi);
-	}
-
 	if (path->slots[0] == 0 &&
 	    sctx->cur_inode_last_extent < key->offset) {
 		/*
@@ -5723,7 +5717,7 @@ static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path,
 		else
 			ret = 0;
 	}
-	sctx->cur_inode_last_extent = extent_end;
+	sctx->cur_inode_last_extent = btrfs_file_extent_end(path);
 	return ret;
 }
 
@@ -7034,7 +7028,6 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	int clone_sources_to_rollback = 0;
 	unsigned alloc_size;
 	int sort_clone_roots = 0;
-	int index;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -7053,12 +7046,6 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	spin_unlock(&send_root->root_item_lock);
 
 	/*
-	 * This is done when we lookup the root, it should already be complete
-	 * by the time we get here.
-	 */
-	WARN_ON(send_root->orphan_cleanup_state != ORPHAN_CLEANUP_DONE);
-
-	/*
 	 * Userspace tools do the checks and warn the user if it's
 	 * not RO.
 	 */
@@ -7078,13 +7065,6 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 		goto out;
 	}
 
-	if (!access_ok(arg->clone_sources,
-			sizeof(*arg->clone_sources) *
-			arg->clone_sources_count)) {
-		ret = -EFAULT;
-		goto out;
-	}
-
 	if (arg->flags & ~BTRFS_SEND_FLAG_MASK) {
 		ret = -EINVAL;
 		goto out;
@@ -7167,11 +7147,8 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 			key.type = BTRFS_ROOT_ITEM_KEY;
 			key.offset = (u64)-1;
 
-			index = srcu_read_lock(&fs_info->subvol_srcu);
-
-			clone_root = btrfs_read_fs_root_no_name(fs_info, &key);
+			clone_root = btrfs_get_fs_root(fs_info, &key, true);
 			if (IS_ERR(clone_root)) {
-				srcu_read_unlock(&fs_info->subvol_srcu, index);
 				ret = PTR_ERR(clone_root);
 				goto out;
 			}
@@ -7179,20 +7156,19 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 			if (!btrfs_root_readonly(clone_root) ||
 			    btrfs_root_dead(clone_root)) {
 				spin_unlock(&clone_root->root_item_lock);
-				srcu_read_unlock(&fs_info->subvol_srcu, index);
+				btrfs_put_root(clone_root);
 				ret = -EPERM;
 				goto out;
 			}
 			if (clone_root->dedupe_in_progress) {
 				dedupe_in_progress_warn(clone_root);
 				spin_unlock(&clone_root->root_item_lock);
-				srcu_read_unlock(&fs_info->subvol_srcu, index);
+				btrfs_put_root(clone_root);
 				ret = -EAGAIN;
 				goto out;
 			}
 			clone_root->send_in_progress++;
 			spin_unlock(&clone_root->root_item_lock);
-			srcu_read_unlock(&fs_info->subvol_srcu, index);
 
 			sctx->clone_roots[i].root = clone_root;
 			clone_sources_to_rollback = i + 1;
@@ -7206,11 +7182,8 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 		key.type = BTRFS_ROOT_ITEM_KEY;
 		key.offset = (u64)-1;
 
-		index = srcu_read_lock(&fs_info->subvol_srcu);
-
-		sctx->parent_root = btrfs_read_fs_root_no_name(fs_info, &key);
+		sctx->parent_root = btrfs_get_fs_root(fs_info, &key, true);
 		if (IS_ERR(sctx->parent_root)) {
-			srcu_read_unlock(&fs_info->subvol_srcu, index);
 			ret = PTR_ERR(sctx->parent_root);
 			goto out;
 		}
@@ -7220,20 +7193,16 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 		if (!btrfs_root_readonly(sctx->parent_root) ||
 				btrfs_root_dead(sctx->parent_root)) {
 			spin_unlock(&sctx->parent_root->root_item_lock);
-			srcu_read_unlock(&fs_info->subvol_srcu, index);
 			ret = -EPERM;
 			goto out;
 		}
 		if (sctx->parent_root->dedupe_in_progress) {
 			dedupe_in_progress_warn(sctx->parent_root);
 			spin_unlock(&sctx->parent_root->root_item_lock);
-			srcu_read_unlock(&fs_info->subvol_srcu, index);
 			ret = -EAGAIN;
 			goto out;
 		}
 		spin_unlock(&sctx->parent_root->root_item_lock);
-
-		srcu_read_unlock(&fs_info->subvol_srcu, index);
 	}
 
 	/*
@@ -7241,7 +7210,8 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg)
 	 * is behind the current send position. This is checked while searching
 	 * for possible clone sources.
 	 */
-	sctx->clone_roots[sctx->clone_roots_cnt++].root = sctx->send_root;
+	sctx->clone_roots[sctx->clone_roots_cnt++].root =
+		btrfs_grab_root(sctx->send_root);
 
 	/* We do a bsearch later */
 	sort(sctx->clone_roots, sctx->clone_roots_cnt,
@@ -7326,18 +7296,24 @@ out:
 	}
 
 	if (sort_clone_roots) {
-		for (i = 0; i < sctx->clone_roots_cnt; i++)
+		for (i = 0; i < sctx->clone_roots_cnt; i++) {
 			btrfs_root_dec_send_in_progress(
 					sctx->clone_roots[i].root);
+			btrfs_put_root(sctx->clone_roots[i].root);
+		}
 	} else {
-		for (i = 0; sctx && i < clone_sources_to_rollback; i++)
+		for (i = 0; sctx && i < clone_sources_to_rollback; i++) {
 			btrfs_root_dec_send_in_progress(
 					sctx->clone_roots[i].root);
+			btrfs_put_root(sctx->clone_roots[i].root);
+		}
 
 		btrfs_root_dec_send_in_progress(send_root);
 	}
-	if (sctx && !IS_ERR_OR_NULL(sctx->parent_root))
+	if (sctx && !IS_ERR_OR_NULL(sctx->parent_root)) {
 		btrfs_root_dec_send_in_progress(sctx->parent_root);
+		btrfs_put_root(sctx->parent_root);
+	}
 
 	kvfree(clone_sources_tmp);
 
diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c
index e8a4b0ebe97f..ff17a4420358 100644
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -10,7 +10,154 @@
 #include "transaction.h"
 #include "block-group.h"
 
-u64 btrfs_space_info_used(struct btrfs_space_info *s_info,
+/*
+ * HOW DOES SPACE RESERVATION WORK
+ *
+ * If you want to know about delalloc specifically, there is a separate comment
+ * for that with the delalloc code.  This comment is about how the whole system
+ * works generally.
+ *
+ * BASIC CONCEPTS
+ *
+ *   1) space_info.  This is the ultimate arbiter of how much space we can use.
+ *   There's a description of the bytes_ fields with the struct declaration,
+ *   refer to that for specifics on each field.  Suffice it to say that for
+ *   reservations we care about total_bytes - SUM(space_info->bytes_) when
+ *   determining if there is space to make an allocation.  There is a space_info
+ *   for METADATA, SYSTEM, and DATA areas.
+ *
+ *   2) block_rsv's.  These are basically buckets for every different type of
+ *   metadata reservation we have.  You can see the comment in the block_rsv
+ *   code on the rules for each type, but generally block_rsv->reserved is how
+ *   much space is accounted for in space_info->bytes_may_use.
+ *
+ *   3) btrfs_calc*_size.  These are the worst case calculations we used based
+ *   on the number of items we will want to modify.  We have one for changing
+ *   items, and one for inserting new items.  Generally we use these helpers to
+ *   determine the size of the block reserves, and then use the actual bytes
+ *   values to adjust the space_info counters.
+ *
+ * MAKING RESERVATIONS, THE NORMAL CASE
+ *
+ *   We call into either btrfs_reserve_data_bytes() or
+ *   btrfs_reserve_metadata_bytes(), depending on which we're looking for, with
+ *   num_bytes we want to reserve.
+ *
+ *   ->reserve
+ *     space_info->bytes_may_reserve += num_bytes
+ *
+ *   ->extent allocation
+ *     Call btrfs_add_reserved_bytes() which does
+ *     space_info->bytes_may_reserve -= num_bytes
+ *     space_info->bytes_reserved += extent_bytes
+ *
+ *   ->insert reference
+ *     Call btrfs_update_block_group() which does
+ *     space_info->bytes_reserved -= extent_bytes
+ *     space_info->bytes_used += extent_bytes
+ *
+ * MAKING RESERVATIONS, FLUSHING NORMALLY (non-priority)
+ *
+ *   Assume we are unable to simply make the reservation because we do not have
+ *   enough space
+ *
+ *   -> __reserve_bytes
+ *     create a reserve_ticket with ->bytes set to our reservation, add it to
+ *     the tail of space_info->tickets, kick async flush thread
+ *
+ *   ->handle_reserve_ticket
+ *     wait on ticket->wait for ->bytes to be reduced to 0, or ->error to be set
+ *     on the ticket.
+ *
+ *   -> btrfs_async_reclaim_metadata_space/btrfs_async_reclaim_data_space
+ *     Flushes various things attempting to free up space.
+ *
+ *   -> btrfs_try_granting_tickets()
+ *     This is called by anything that either subtracts space from
+ *     space_info->bytes_may_use, ->bytes_pinned, etc, or adds to the
+ *     space_info->total_bytes.  This loops through the ->priority_tickets and
+ *     then the ->tickets list checking to see if the reservation can be
+ *     completed.  If it can the space is added to space_info->bytes_may_use and
+ *     the ticket is woken up.
+ *
+ *   -> ticket wakeup
+ *     Check if ->bytes == 0, if it does we got our reservation and we can carry
+ *     on, if not return the appropriate error (ENOSPC, but can be EINTR if we
+ *     were interrupted.)
+ *
+ * MAKING RESERVATIONS, FLUSHING HIGH PRIORITY
+ *
+ *   Same as the above, except we add ourselves to the
+ *   space_info->priority_tickets, and we do not use ticket->wait, we simply
+ *   call flush_space() ourselves for the states that are safe for us to call
+ *   without deadlocking and hope for the best.
+ *
+ * THE FLUSHING STATES
+ *
+ *   Generally speaking we will have two cases for each state, a "nice" state
+ *   and a "ALL THE THINGS" state.  In btrfs we delay a lot of work in order to
+ *   reduce the locking over head on the various trees, and even to keep from
+ *   doing any work at all in the case of delayed refs.  Each of these delayed
+ *   things however hold reservations, and so letting them run allows us to
+ *   reclaim space so we can make new reservations.
+ *
+ *   FLUSH_DELAYED_ITEMS
+ *     Every inode has a delayed item to update the inode.  Take a simple write
+ *     for example, we would update the inode item at write time to update the
+ *     mtime, and then again at finish_ordered_io() time in order to update the
+ *     isize or bytes.  We keep these delayed items to coalesce these operations
+ *     into a single operation done on demand.  These are an easy way to reclaim
+ *     metadata space.
+ *
+ *   FLUSH_DELALLOC
+ *     Look at the delalloc comment to get an idea of how much space is reserved
+ *     for delayed allocation.  We can reclaim some of this space simply by
+ *     running delalloc, but usually we need to wait for ordered extents to
+ *     reclaim the bulk of this space.
+ *
+ *   FLUSH_DELAYED_REFS
+ *     We have a block reserve for the outstanding delayed refs space, and every
+ *     delayed ref operation holds a reservation.  Running these is a quick way
+ *     to reclaim space, but we want to hold this until the end because COW can
+ *     churn a lot and we can avoid making some extent tree modifications if we
+ *     are able to delay for as long as possible.
+ *
+ *   ALLOC_CHUNK
+ *     We will skip this the first time through space reservation, because of
+ *     overcommit and we don't want to have a lot of useless metadata space when
+ *     our worst case reservations will likely never come true.
+ *
+ *   RUN_DELAYED_IPUTS
+ *     If we're freeing inodes we're likely freeing checksums, file extent
+ *     items, and extent tree items.  Loads of space could be freed up by these
+ *     operations, however they won't be usable until the transaction commits.
+ *
+ *   COMMIT_TRANS
+ *     may_commit_transaction() is the ultimate arbiter on whether we commit the
+ *     transaction or not.  In order to avoid constantly churning we do all the
+ *     above flushing first and then commit the transaction as the last resort.
+ *     However we need to take into account things like pinned space that would
+ *     be freed, plus any delayed work we may not have gotten rid of in the case
+ *     of metadata.
+ *
+ * OVERCOMMIT
+ *
+ *   Because we hold so many reservations for metadata we will allow you to
+ *   reserve more space than is currently free in the currently allocate
+ *   metadata space.  This only happens with metadata, data does not allow
+ *   overcommitting.
+ *
+ *   You can see the current logic for when we allow overcommit in
+ *   btrfs_can_overcommit(), but it only applies to unallocated space.  If there
+ *   is no unallocated space to be had, all reservations are kept within the
+ *   free space in the allocated metadata chunks.
+ *
+ *   Because of overcommitting, you generally want to use the
+ *   btrfs_can_overcommit() logic for metadata allocations, as it does the right
+ *   thing with or without extra unallocated space.
+ */
+
+u64 __pure btrfs_space_info_used(struct btrfs_space_info *s_info,
 			  bool may_use_included)
 {
 	ASSERT(s_info);
@@ -58,7 +205,6 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags)
 	spin_lock_init(&space_info->lock);
 	space_info->flags = flags & BTRFS_BLOCK_GROUP_TYPE_MASK;
 	space_info->force_alloc = CHUNK_ALLOC_NO_FORCE;
-	init_waitqueue_head(&space_info->wait);
 	INIT_LIST_HEAD(&space_info->ro_bgs);
 	INIT_LIST_HEAD(&space_info->tickets);
 	INIT_LIST_HEAD(&space_info->priority_tickets);
@@ -160,26 +306,19 @@ static inline u64 calc_global_rsv_need_space(struct btrfs_block_rsv *global)
 	return (global->size << 1);
 }
 
-static int can_overcommit(struct btrfs_fs_info *fs_info,
-			  struct btrfs_space_info *space_info, u64 bytes,
-			  enum btrfs_reserve_flush_enum flush,
-			  bool system_chunk)
+static u64 calc_available_free_space(struct btrfs_fs_info *fs_info,
+			  struct btrfs_space_info *space_info,
+			  enum btrfs_reserve_flush_enum flush)
 {
 	u64 profile;
 	u64 avail;
-	u64 used;
 	int factor;
 
-	/* Don't overcommit when in mixed mode. */
-	if (space_info->flags & BTRFS_BLOCK_GROUP_DATA)
-		return 0;
-
-	if (system_chunk)
+	if (space_info->flags & BTRFS_BLOCK_GROUP_SYSTEM)
 		profile = btrfs_system_alloc_profile(fs_info);
 	else
 		profile = btrfs_metadata_alloc_profile(fs_info);
 
-	used = btrfs_space_info_used(space_info, true);
 	avail = atomic64_read(&fs_info->free_chunk_space);
 
 	/*
@@ -200,12 +339,38 @@ static int can_overcommit(struct btrfs_fs_info *fs_info,
 		avail >>= 3;
 	else
 		avail >>= 1;
+	return avail;
+}
+
+int btrfs_can_overcommit(struct btrfs_fs_info *fs_info,
+			 struct btrfs_space_info *space_info, u64 bytes,
+			 enum btrfs_reserve_flush_enum flush)
+{
+	u64 avail;
+	u64 used;
+
+	/* Don't overcommit when in mixed mode */
+	if (space_info->flags & BTRFS_BLOCK_GROUP_DATA)
+		return 0;
+
+	used = btrfs_space_info_used(space_info, true);
+	avail = calc_available_free_space(fs_info, space_info, flush);
 
 	if (used + bytes < space_info->total_bytes + avail)
 		return 1;
 	return 0;
 }
 
+static void remove_ticket(struct btrfs_space_info *space_info,
+			  struct reserve_ticket *ticket)
+{
+	if (!list_empty(&ticket->list)) {
+		list_del_init(&ticket->list);
+		ASSERT(space_info->reclaim_size >= ticket->bytes);
+		space_info->reclaim_size -= ticket->bytes;
+	}
+}
+
 /*
  * This is for space we already have accounted in space_info->bytes_may_use, so
  * basically when we're returning space from block_rsv's.
@@ -228,12 +393,12 @@ again:
 
 		/* Check and see if our ticket can be satisified now. */
 		if ((used + ticket->bytes <= space_info->total_bytes) ||
-		    can_overcommit(fs_info, space_info, ticket->bytes, flush,
-				   false)) {
+		    btrfs_can_overcommit(fs_info, space_info, ticket->bytes,
+					 flush)) {
 			btrfs_space_info_update_bytes_may_use(fs_info,
 							      space_info,
 							      ticket->bytes);
-			list_del_init(&ticket->list);
+			remove_ticket(space_info, ticket);
 			ticket->bytes = 0;
 			space_info->tickets_id++;
 			wake_up(&ticket->wait);
@@ -285,7 +450,7 @@ void btrfs_dump_space_info(struct btrfs_fs_info *fs_info,
 			   struct btrfs_space_info *info, u64 bytes,
 			   int dump_block_groups)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	int index = 0;
 
 	spin_lock(&info->lock);
@@ -301,8 +466,7 @@ again:
 		spin_lock(&cache->lock);
 		btrfs_info(fs_info,
 			"block group %llu has %llu bytes, %llu used %llu pinned %llu reserved %s",
-			cache->key.objectid, cache->key.offset,
-			btrfs_block_group_used(&cache->item), cache->pinned,
+			cache->start, cache->length, cache->used, cache->pinned,
 			cache->reserved, cache->ro ? "[readonly]" : "");
 		btrfs_dump_free_space(cache, bytes);
 		spin_unlock(&cache->lock);
@@ -628,30 +792,40 @@ static void flush_space(struct btrfs_fs_info *fs_info,
 
 static inline u64
 btrfs_calc_reclaim_metadata_size(struct btrfs_fs_info *fs_info,
-				 struct btrfs_space_info *space_info,
-				 bool system_chunk)
+				 struct btrfs_space_info *space_info)
 {
-	struct reserve_ticket *ticket;
 	u64 used;
+	u64 avail;
 	u64 expected;
-	u64 to_reclaim = 0;
+	u64 to_reclaim = space_info->reclaim_size;
+
+	lockdep_assert_held(&space_info->lock);
+
+	avail = calc_available_free_space(fs_info, space_info,
+					  BTRFS_RESERVE_FLUSH_ALL);
+	used = btrfs_space_info_used(space_info, true);
+
+	/*
+	 * We may be flushing because suddenly we have less space than we had
+	 * before, and now we're well over-committed based on our current free
+	 * space.  If that's the case add in our overage so we make sure to put
+	 * appropriate pressure on the flushing state machine.
+	 */
+	if (space_info->total_bytes + avail < used)
+		to_reclaim += used - (space_info->total_bytes + avail);
 
-	list_for_each_entry(ticket, &space_info->tickets, list)
-		to_reclaim += ticket->bytes;
-	list_for_each_entry(ticket, &space_info->priority_tickets, list)
-		to_reclaim += ticket->bytes;
 	if (to_reclaim)
 		return to_reclaim;
 
 	to_reclaim = min_t(u64, num_online_cpus() * SZ_1M, SZ_16M);
-	if (can_overcommit(fs_info, space_info, to_reclaim,
-			   BTRFS_RESERVE_FLUSH_ALL, system_chunk))
+	if (btrfs_can_overcommit(fs_info, space_info, to_reclaim,
+				 BTRFS_RESERVE_FLUSH_ALL))
 		return 0;
 
 	used = btrfs_space_info_used(space_info, true);
 
-	if (can_overcommit(fs_info, space_info, SZ_1M,
-			   BTRFS_RESERVE_FLUSH_ALL, system_chunk))
+	if (btrfs_can_overcommit(fs_info, space_info, SZ_1M,
+				 BTRFS_RESERVE_FLUSH_ALL))
 		expected = div_factor_fine(space_info->total_bytes, 95);
 	else
 		expected = div_factor_fine(space_info->total_bytes, 90);
@@ -667,7 +841,7 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_fs_info *fs_info,
 
 static inline int need_do_async_reclaim(struct btrfs_fs_info *fs_info,
 					struct btrfs_space_info *space_info,
-					u64 used, bool system_chunk)
+					u64 used)
 {
 	u64 thresh = div_factor_fine(space_info->total_bytes, 98);
 
@@ -675,8 +849,7 @@ static inline int need_do_async_reclaim(struct btrfs_fs_info *fs_info,
 	if ((space_info->bytes_used + space_info->bytes_reserved) >= thresh)
 		return 0;
 
-	if (!btrfs_calc_reclaim_metadata_size(fs_info, space_info,
-					      system_chunk))
+	if (!btrfs_calc_reclaim_metadata_size(fs_info, space_info))
 		return 0;
 
 	return (used >= thresh && !btrfs_fs_closing(fs_info) &&
@@ -734,7 +907,7 @@ static bool maybe_fail_all_tickets(struct btrfs_fs_info *fs_info,
 			btrfs_info(fs_info, "failing ticket with %llu bytes",
 				   ticket->bytes);
 
-		list_del_init(&ticket->list);
+		remove_ticket(space_info, ticket);
 		ticket->error = -ENOSPC;
 		wake_up(&ticket->wait);
 
@@ -767,8 +940,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 	space_info = btrfs_find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
 
 	spin_lock(&space_info->lock);
-	to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info, space_info,
-						      false);
+	to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info, space_info);
 	if (!to_reclaim) {
 		space_info->flush = 0;
 		spin_unlock(&space_info->lock);
@@ -787,8 +959,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 			return;
 		}
 		to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info,
-							      space_info,
-							      false);
+							      space_info);
 		if (last_tickets_id == space_info->tickets_id) {
 			flush_state++;
 		} else {
@@ -860,8 +1031,7 @@ static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info,
 	int flush_state;
 
 	spin_lock(&space_info->lock);
-	to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info, space_info,
-						      false);
+	to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info, space_info);
 	if (!to_reclaim) {
 		spin_unlock(&space_info->lock);
 		return;
@@ -901,7 +1071,7 @@ static void wait_reserve_ticket(struct btrfs_fs_info *fs_info,
 			 * despite getting an error, resulting in a space leak
 			 * (bytes_may_use counter of our space_info).
 			 */
-			list_del_init(&ticket->list);
+			remove_ticket(space_info, ticket);
 			ticket->error = -EINTR;
 			break;
 		}
@@ -959,7 +1129,7 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 		 * either the async reclaim job deletes the ticket from the list
 		 * or we delete it ourselves at wait_reserve_ticket().
 		 */
-		list_del_init(&ticket->list);
+		remove_ticket(space_info, ticket);
 		if (!ret)
 			ret = -ENOSPC;
 	}
@@ -992,8 +1162,7 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 				    struct btrfs_space_info *space_info,
 				    u64 orig_bytes,
-				    enum btrfs_reserve_flush_enum flush,
-				    bool system_chunk)
+				    enum btrfs_reserve_flush_enum flush)
 {
 	struct reserve_ticket ticket;
 	u64 used;
@@ -1015,8 +1184,7 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 	 */
 	if (!pending_tickets &&
 	    ((used + orig_bytes <= space_info->total_bytes) ||
-	     can_overcommit(fs_info, space_info, orig_bytes, flush,
-			   system_chunk))) {
+	     btrfs_can_overcommit(fs_info, space_info, orig_bytes, flush))) {
 		btrfs_space_info_update_bytes_may_use(fs_info, space_info,
 						      orig_bytes);
 		ret = 0;
@@ -1030,8 +1198,10 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 	 * the list and we will do our own flushing further down.
 	 */
 	if (ret && flush != BTRFS_RESERVE_NO_FLUSH) {
+		ASSERT(space_info->reclaim_size >= 0);
 		ticket.bytes = orig_bytes;
 		ticket.error = 0;
+		space_info->reclaim_size += ticket.bytes;
 		init_waitqueue_head(&ticket.wait);
 		if (flush == BTRFS_RESERVE_FLUSH_ALL) {
 			list_add_tail(&ticket.list, &space_info->tickets);
@@ -1056,8 +1226,7 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info,
 		 * the async reclaim as we will panic.
 		 */
 		if (!test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags) &&
-		    need_do_async_reclaim(fs_info, space_info,
-					  used, system_chunk) &&
+		    need_do_async_reclaim(fs_info, space_info, used) &&
 		    !work_busy(&fs_info->async_reclaim_work)) {
 			trace_btrfs_trigger_flush(fs_info, space_info->flags,
 						  orig_bytes, flush, "preempt");
@@ -1094,10 +1263,9 @@ int btrfs_reserve_metadata_bytes(struct btrfs_root *root,
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv;
 	int ret;
-	bool system_chunk = (root == fs_info->chunk_root);
 
 	ret = __reserve_metadata_bytes(fs_info, block_rsv->space_info,
-				       orig_bytes, flush, system_chunk);
+				       orig_bytes, flush);
 	if (ret == -ENOSPC &&
 	    unlikely(root->orphan_cleanup_state == ORPHAN_CLEANUP_STARTED)) {
 		if (block_rsv != global_rsv &&
diff --git a/fs/btrfs/space-info.h b/fs/btrfs/space-info.h
index 8867e84aa33d..0a5001ef1481 100644
--- a/fs/btrfs/space-info.h
+++ b/fs/btrfs/space-info.h
@@ -54,6 +54,13 @@ struct btrfs_space_info {
 	struct list_head ro_bgs;
 	struct list_head priority_tickets;
 	struct list_head tickets;
+
+	/*
+	 * Size of space that needs to be reclaimed in order to satisfy pending
+	 * tickets
+	 */
+	u64 reclaim_size;
+
 	/*
 	 * tickets_id just indicates the next ticket will be handled, so note
 	 * it's not stored per ticket.
@@ -63,7 +70,6 @@ struct btrfs_space_info {
 	struct rw_semaphore groups_sem;
 	/* for block groups in our same type */
 	struct list_head block_groups[BTRFS_NR_RAID_TYPES];
-	wait_queue_head_t wait;
 
 	struct kobject kobj;
 	struct kobject *block_group_kobjs[BTRFS_NR_RAID_TYPES];
@@ -116,7 +122,7 @@ void btrfs_update_space_info(struct btrfs_fs_info *info, u64 flags,
 			     struct btrfs_space_info **space_info);
 struct btrfs_space_info *btrfs_find_space_info(struct btrfs_fs_info *info,
 					       u64 flags);
-u64 btrfs_space_info_used(struct btrfs_space_info *s_info,
+u64 __pure btrfs_space_info_used(struct btrfs_space_info *s_info,
 			  bool may_use_included);
 void btrfs_clear_space_info_full(struct btrfs_fs_info *info);
 void btrfs_dump_space_info(struct btrfs_fs_info *fs_info,
@@ -128,6 +134,9 @@ int btrfs_reserve_metadata_bytes(struct btrfs_root *root,
 				 enum btrfs_reserve_flush_enum flush);
 void btrfs_try_granting_tickets(struct btrfs_fs_info *fs_info,
 				struct btrfs_space_info *space_info);
+int btrfs_can_overcommit(struct btrfs_fs_info *fs_info,
+			 struct btrfs_space_info *space_info, u64 bytes,
+			 enum btrfs_reserve_flush_enum flush);
 
 static inline void btrfs_space_info_free_bytes_may_use(
 				struct btrfs_fs_info *fs_info,
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 1b151af25772..7932d8d07cff 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -46,6 +46,7 @@
 #include "sysfs.h"
 #include "tests/btrfs-tests.h"
 #include "block-group.h"
+#include "discard.h"
 
 #include "qgroup.h"
 #define CREATE_TRACE_POINTS
@@ -66,7 +67,7 @@ static struct file_system_type btrfs_root_fs_type;
 
 static int btrfs_remount(struct super_block *sb, int *flags, char *data);
 
-const char *btrfs_decode_error(int errno)
+const char * __attribute_const__ btrfs_decode_error(int errno)
 {
 	char *errstr = "unknown";
 
@@ -146,6 +147,8 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function
 	if (sb_rdonly(sb))
 		return;
 
+	btrfs_discard_stop(fs_info);
+
 	/* btrfs handle error by forcing the filesystem readonly */
 	sb->s_flags |= SB_RDONLY;
 	btrfs_info(fs_info, "forced readonly");
@@ -187,7 +190,7 @@ static struct ratelimit_state printk_limits[] = {
 	RATELIMIT_STATE_INIT(printk_limits[7], DEFAULT_RATELIMIT_INTERVAL, 100),
 };
 
-void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
+void __cold btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
 {
 	char lvl[PRINTK_MAX_SINGLE_HEADER_LEN + 1] = "\0";
 	struct va_format vaf;
@@ -241,7 +244,7 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 
-	trans->aborted = errno;
+	WRITE_ONCE(trans->aborted, errno);
 	/* Nothing used. The other threads that have joined this
 	 * transaction may be able to continue. */
 	if (!trans->dirty && list_empty(&trans->new_bgs)) {
@@ -313,6 +316,7 @@ enum {
 	Opt_datasum, Opt_nodatasum,
 	Opt_defrag, Opt_nodefrag,
 	Opt_discard, Opt_nodiscard,
+	Opt_discard_mode,
 	Opt_nologreplay,
 	Opt_norecovery,
 	Opt_ratio,
@@ -375,6 +379,7 @@ static const match_table_t tokens = {
 	{Opt_defrag, "autodefrag"},
 	{Opt_nodefrag, "noautodefrag"},
 	{Opt_discard, "discard"},
+	{Opt_discard_mode, "discard=%s"},
 	{Opt_nodiscard, "nodiscard"},
 	{Opt_nologreplay, "nologreplay"},
 	{Opt_norecovery, "norecovery"},
@@ -695,12 +700,26 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 				   info->metadata_ratio);
 			break;
 		case Opt_discard:
-			btrfs_set_and_info(info, DISCARD,
-					   "turning on discard");
+		case Opt_discard_mode:
+			if (token == Opt_discard ||
+			    strcmp(args[0].from, "sync") == 0) {
+				btrfs_clear_opt(info->mount_opt, DISCARD_ASYNC);
+				btrfs_set_and_info(info, DISCARD_SYNC,
+						   "turning on sync discard");
+			} else if (strcmp(args[0].from, "async") == 0) {
+				btrfs_clear_opt(info->mount_opt, DISCARD_SYNC);
+				btrfs_set_and_info(info, DISCARD_ASYNC,
+						   "turning on async discard");
+			} else {
+				ret = -EINVAL;
+				goto out;
+			}
 			break;
 		case Opt_nodiscard:
-			btrfs_clear_and_info(info, DISCARD,
+			btrfs_clear_and_info(info, DISCARD_SYNC,
 					     "turning off discard");
+			btrfs_clear_and_info(info, DISCARD_ASYNC,
+					     "turning off async discard");
 			break;
 		case Opt_space_cache:
 		case Opt_space_cache_version:
@@ -854,7 +873,7 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 			break;
 #endif
 		case Opt_err:
-			btrfs_info(info, "unrecognized mount option '%s'", p);
+			btrfs_err(info, "unrecognized mount option '%s'", p);
 			ret = -EINVAL;
 			goto out;
 		default:
@@ -1005,11 +1024,11 @@ out:
 	return error;
 }
 
-static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
-					   u64 subvol_objectid)
+char *btrfs_get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
+					  u64 subvol_objectid)
 {
 	struct btrfs_root *root = fs_info->tree_root;
-	struct btrfs_root *fs_root;
+	struct btrfs_root *fs_root = NULL;
 	struct btrfs_root_ref *root_ref;
 	struct btrfs_inode_ref *inode_ref;
 	struct btrfs_key key;
@@ -1077,9 +1096,10 @@ static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
 		key.objectid = subvol_objectid;
 		key.type = BTRFS_ROOT_ITEM_KEY;
 		key.offset = (u64)-1;
-		fs_root = btrfs_read_fs_root_no_name(fs_info, &key);
+		fs_root = btrfs_get_fs_root(fs_info, &key, true);
 		if (IS_ERR(fs_root)) {
 			ret = PTR_ERR(fs_root);
+			fs_root = NULL;
 			goto err;
 		}
 
@@ -1124,6 +1144,8 @@ static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
 			ptr[0] = '/';
 			btrfs_release_path(path);
 		}
+		btrfs_put_root(fs_root);
+		fs_root = NULL;
 	}
 
 	btrfs_free_path(path);
@@ -1136,6 +1158,7 @@ static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info,
 	return name;
 
 err:
+	btrfs_put_root(fs_root);
 	btrfs_free_path(path);
 	kfree(name);
 	return ERR_PTR(ret);
@@ -1219,7 +1242,7 @@ static int btrfs_fill_super(struct super_block *sb,
 	key.objectid = BTRFS_FIRST_FREE_OBJECTID;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	inode = btrfs_iget(sb, &key, fs_info->fs_root, NULL);
+	inode = btrfs_iget(sb, &key, fs_info->fs_root);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
 		goto fail_close;
@@ -1322,8 +1345,10 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 		seq_puts(seq, ",nologreplay");
 	if (btrfs_test_opt(info, FLUSHONCOMMIT))
 		seq_puts(seq, ",flushoncommit");
-	if (btrfs_test_opt(info, DISCARD))
+	if (btrfs_test_opt(info, DISCARD_SYNC))
 		seq_puts(seq, ",discard");
+	if (btrfs_test_opt(info, DISCARD_ASYNC))
+		seq_puts(seq, ",discard=async");
 	if (!(info->sb->s_flags & SB_POSIXACL))
 		seq_puts(seq, ",noacl");
 	if (btrfs_test_opt(info, SPACE_CACHE))
@@ -1417,8 +1442,8 @@ static struct dentry *mount_subvol(const char *subvol_name, u64 subvol_objectid,
 				goto out;
 			}
 		}
-		subvol_name = get_subvol_name_from_objectid(btrfs_sb(mnt->mnt_sb),
-							    subvol_objectid);
+		subvol_name = btrfs_get_subvol_name_from_objectid(
+					btrfs_sb(mnt->mnt_sb), subvol_objectid);
 		if (IS_ERR(subvol_name)) {
 			root = ERR_CAST(subvol_name);
 			subvol_name = NULL;
@@ -1497,14 +1522,17 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
 	/*
 	 * Setup a dummy root and fs_info for test/set super.  This is because
 	 * we don't actually fill this stuff out until open_ctree, but we need
-	 * it for searching for existing supers, so this lets us do that and
-	 * then open_ctree will properly initialize everything later.
+	 * then open_ctree will properly initialize the file system specific
+	 * settings later.  btrfs_init_fs_info initializes the static elements
+	 * of the fs_info (locks and such) to make cleanup easier if we find a
+	 * superblock with our given fs_devices later on at sget() time.
 	 */
 	fs_info = kvzalloc(sizeof(struct btrfs_fs_info), GFP_KERNEL);
 	if (!fs_info) {
 		error = -ENOMEM;
 		goto error_sec_opts;
 	}
+	btrfs_init_fs_info(fs_info);
 
 	fs_info->super_copy = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_KERNEL);
 	fs_info->super_for_commit = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_KERNEL);
@@ -1550,7 +1578,7 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
 
 	if (s->s_root) {
 		btrfs_close_devices(fs_devices);
-		free_fs_info(fs_info);
+		btrfs_free_fs_info(fs_info);
 		if ((flags ^ s->s_flags) & SB_RDONLY)
 			error = -EBUSY;
 	} else {
@@ -1573,7 +1601,7 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type,
 error_close_devices:
 	btrfs_close_devices(fs_devices);
 error_fs_info:
-	free_fs_info(fs_info);
+	btrfs_free_fs_info(fs_info);
 error_sec_opts:
 	security_free_mnt_opts(&new_sec_opts);
 	return ERR_PTR(error);
@@ -1669,7 +1697,6 @@ static void btrfs_resize_thread_pool(struct btrfs_fs_info *fs_info,
 
 	btrfs_workqueue_set_max(fs_info->workers, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->delalloc_workers, new_pool_size);
-	btrfs_workqueue_set_max(fs_info->submit_workers, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->caching_workers, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->endio_workers, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->endio_meta_workers, new_pool_size);
@@ -1714,6 +1741,14 @@ static inline void btrfs_remount_cleanup(struct btrfs_fs_info *fs_info,
 		btrfs_cleanup_defrag_inodes(fs_info);
 	}
 
+	/* If we toggled discard async */
+	if (!btrfs_raw_test_opt(old_opts, DISCARD_ASYNC) &&
+	    btrfs_test_opt(fs_info, DISCARD_ASYNC))
+		btrfs_discard_resume(fs_info);
+	else if (btrfs_raw_test_opt(old_opts, DISCARD_ASYNC) &&
+		 !btrfs_test_opt(fs_info, DISCARD_ASYNC))
+		btrfs_discard_cleanup(fs_info);
+
 	clear_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state);
 }
 
@@ -1761,6 +1796,8 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 		 */
 		cancel_work_sync(&fs_info->async_reclaim_work);
 
+		btrfs_discard_cleanup(fs_info);
+
 		/* wait for the uuid_scan task to finish */
 		down(&fs_info->uuid_tree_rescan_sem);
 		/* avoid complains from lockdep et al. */
@@ -1804,6 +1841,8 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 		}
 
 		if (btrfs_super_log_root(fs_info->super_copy) != 0) {
+			btrfs_warn(fs_info,
+		"mount required to replay tree-log, cannot remount read-write");
 			ret = -EINVAL;
 			goto restore;
 		}
@@ -1936,6 +1975,10 @@ static inline int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
 		num_stripes = nr_devices;
 	else if (type & BTRFS_BLOCK_GROUP_RAID1)
 		num_stripes = 2;
+	else if (type & BTRFS_BLOCK_GROUP_RAID1C3)
+		num_stripes = 3;
+	else if (type & BTRFS_BLOCK_GROUP_RAID1C4)
+		num_stripes = 4;
 	else if (type & BTRFS_BLOCK_GROUP_RAID10)
 		num_stripes = 4;
 
@@ -2022,7 +2065,6 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(dentry->d_sb);
 	struct btrfs_super_block *disk_super = fs_info->super_copy;
-	struct list_head *head = &fs_info->space_info;
 	struct btrfs_space_info *found;
 	u64 total_used = 0;
 	u64 total_free_data = 0;
@@ -2036,7 +2078,7 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	int mixed = 0;
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(found, head, list) {
+	list_for_each_entry_rcu(found, &fs_info->space_info, list) {
 		if (found->flags & BTRFS_BLOCK_GROUP_DATA) {
 			int i;
 
@@ -2102,7 +2144,15 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	 */
 	thresh = SZ_4M;
 
-	if (!mixed && total_free_meta - thresh < block_rsv->size)
+	/*
+	 * We only want to claim there's no available space if we can no longer
+	 * allocate chunks for our metadata profile and our global reserve will
+	 * not fit in the free metadata space.  If we aren't ->full then we
+	 * still can allocate chunks and thus are fine using the currently
+	 * calculated f_bavail.
+	 */
+	if (!mixed && block_rsv->space_info->full &&
+	    total_free_meta - thresh < block_rsv->size)
 		buf->f_bavail = 0;
 
 	buf->f_type = BTRFS_SUPER_MAGIC;
@@ -2127,7 +2177,7 @@ static void btrfs_kill_super(struct super_block *sb)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(sb);
 	kill_anon_super(sb);
-	free_fs_info(fs_info);
+	btrfs_free_fs_info(fs_info);
 }
 
 static struct file_system_type btrfs_fs_type = {
@@ -2160,7 +2210,7 @@ static int btrfs_control_open(struct inode *inode, struct file *file)
 }
 
 /*
- * used by btrfsctl to scan devices when no FS is mounted
+ * Used by /dev/btrfs-control for devices ioctls.
  */
 static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 				unsigned long arg)
@@ -2297,7 +2347,7 @@ static const struct super_operations btrfs_super_ops = {
 static const struct file_operations btrfs_ctl_fops = {
 	.open = btrfs_control_open,
 	.unlocked_ioctl	 = btrfs_control_ioctl,
-	.compat_ioctl = btrfs_control_ioctl,
+	.compat_ioctl = compat_ptr_ioctl,
 	.owner	 = THIS_MODULE,
 	.llseek = noop_llseek,
 };
@@ -2360,10 +2410,14 @@ static int __init init_btrfs_fs(void)
 	if (err)
 		goto free_cachep;
 
-	err = extent_map_init();
+	err = extent_state_cache_init();
 	if (err)
 		goto free_extent_io;
 
+	err = extent_map_init();
+	if (err)
+		goto free_extent_state_cache;
+
 	err = ordered_data_init();
 	if (err)
 		goto free_extent_map;
@@ -2422,6 +2476,8 @@ free_ordered_data:
 	ordered_data_exit();
 free_extent_map:
 	extent_map_exit();
+free_extent_state_cache:
+	extent_state_cache_exit();
 free_extent_io:
 	extent_io_exit();
 free_cachep:
@@ -2442,6 +2498,7 @@ static void __exit exit_btrfs_fs(void)
 	btrfs_prelim_ref_exit();
 	ordered_data_exit();
 	extent_map_exit();
+	extent_state_cache_exit();
 	extent_io_exit();
 	btrfs_interface_exit();
 	btrfs_end_io_wq_exit();
@@ -2456,3 +2513,6 @@ module_exit(exit_btrfs_fs)
 
 MODULE_LICENSE("GPL");
 MODULE_SOFTDEP("pre: crc32c");
+MODULE_SOFTDEP("pre: xxhash64");
+MODULE_SOFTDEP("pre: sha256");
+MODULE_SOFTDEP("pre: blake2b-256");
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index f6d3c80f2e28..a39bff64ff24 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -9,8 +9,10 @@
 #include <linux/spinlock.h>
 #include <linux/completion.h>
 #include <linux/bug.h>
+#include <crypto/hash.h>
 
 #include "ctree.h"
+#include "discard.h"
 #include "disk-io.h"
 #include "transaction.h"
 #include "sysfs.h"
@@ -153,7 +155,7 @@ static ssize_t btrfs_feature_attr_show(struct kobject *kobj,
 	} else
 		val = can_modify_feature(fa);
 
-	return snprintf(buf, PAGE_SIZE, "%d\n", val);
+	return scnprintf(buf, PAGE_SIZE, "%d\n", val);
 }
 
 static ssize_t btrfs_feature_attr_store(struct kobject *kobj,
@@ -258,6 +260,7 @@ BTRFS_FEAT_ATTR_INCOMPAT(skinny_metadata, SKINNY_METADATA);
 BTRFS_FEAT_ATTR_INCOMPAT(no_holes, NO_HOLES);
 BTRFS_FEAT_ATTR_INCOMPAT(metadata_uuid, METADATA_UUID);
 BTRFS_FEAT_ATTR_COMPAT_RO(free_space_tree, FREE_SPACE_TREE);
+BTRFS_FEAT_ATTR_INCOMPAT(raid1c34, RAID1C34);
 
 static struct attribute *btrfs_supported_feature_attrs[] = {
 	BTRFS_FEAT_ATTR_PTR(mixed_backref),
@@ -272,6 +275,7 @@ static struct attribute *btrfs_supported_feature_attrs[] = {
 	BTRFS_FEAT_ATTR_PTR(no_holes),
 	BTRFS_FEAT_ATTR_PTR(metadata_uuid),
 	BTRFS_FEAT_ATTR_PTR(free_space_tree),
+	BTRFS_FEAT_ATTR_PTR(raid1c34),
 	NULL
 };
 
@@ -291,12 +295,34 @@ static const struct attribute_group btrfs_feature_attr_group = {
 static ssize_t rmdir_subvol_show(struct kobject *kobj,
 				 struct kobj_attribute *ka, char *buf)
 {
-	return snprintf(buf, PAGE_SIZE, "0\n");
+	return scnprintf(buf, PAGE_SIZE, "0\n");
 }
 BTRFS_ATTR(static_feature, rmdir_subvol, rmdir_subvol_show);
 
+static ssize_t supported_checksums_show(struct kobject *kobj,
+					struct kobj_attribute *a, char *buf)
+{
+	ssize_t ret = 0;
+	int i;
+
+	for (i = 0; i < btrfs_get_num_csums(); i++) {
+		/*
+		 * This "trick" only works as long as 'enum btrfs_csum_type' has
+		 * no holes in it
+		 */
+		ret += scnprintf(buf + ret, PAGE_SIZE - ret, "%s%s",
+				(i == 0 ? "" : " "), btrfs_super_csum_name(i));
+
+	}
+
+	ret += scnprintf(buf + ret, PAGE_SIZE - ret, "\n");
+	return ret;
+}
+BTRFS_ATTR(static_feature, supported_checksums, supported_checksums_show);
+
 static struct attribute *btrfs_supported_static_feature_attrs[] = {
 	BTRFS_ATTR_PTR(static_feature, rmdir_subvol),
+	BTRFS_ATTR_PTR(static_feature, supported_checksums),
 	NULL
 };
 
@@ -314,11 +340,177 @@ static const struct attribute_group btrfs_static_feature_attr_group = {
 #ifdef CONFIG_BTRFS_DEBUG
 
 /*
+ * Discard statistics and tunables
+ */
+#define discard_to_fs_info(_kobj)	to_fs_info((_kobj)->parent->parent)
+
+static ssize_t btrfs_discardable_bytes_show(struct kobject *kobj,
+					    struct kobj_attribute *a,
+					    char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return scnprintf(buf, PAGE_SIZE, "%lld\n",
+			atomic64_read(&fs_info->discard_ctl.discardable_bytes));
+}
+BTRFS_ATTR(discard, discardable_bytes, btrfs_discardable_bytes_show);
+
+static ssize_t btrfs_discardable_extents_show(struct kobject *kobj,
+					      struct kobj_attribute *a,
+					      char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return scnprintf(buf, PAGE_SIZE, "%d\n",
+			atomic_read(&fs_info->discard_ctl.discardable_extents));
+}
+BTRFS_ATTR(discard, discardable_extents, btrfs_discardable_extents_show);
+
+static ssize_t btrfs_discard_bitmap_bytes_show(struct kobject *kobj,
+					       struct kobj_attribute *a,
+					       char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return scnprintf(buf, PAGE_SIZE, "%lld\n",
+			fs_info->discard_ctl.discard_bitmap_bytes);
+}
+BTRFS_ATTR(discard, discard_bitmap_bytes, btrfs_discard_bitmap_bytes_show);
+
+static ssize_t btrfs_discard_bytes_saved_show(struct kobject *kobj,
+					      struct kobj_attribute *a,
+					      char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return scnprintf(buf, PAGE_SIZE, "%lld\n",
+		atomic64_read(&fs_info->discard_ctl.discard_bytes_saved));
+}
+BTRFS_ATTR(discard, discard_bytes_saved, btrfs_discard_bytes_saved_show);
+
+static ssize_t btrfs_discard_extent_bytes_show(struct kobject *kobj,
+					       struct kobj_attribute *a,
+					       char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return scnprintf(buf, PAGE_SIZE, "%lld\n",
+			fs_info->discard_ctl.discard_extent_bytes);
+}
+BTRFS_ATTR(discard, discard_extent_bytes, btrfs_discard_extent_bytes_show);
+
+static ssize_t btrfs_discard_iops_limit_show(struct kobject *kobj,
+					     struct kobj_attribute *a,
+					     char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return scnprintf(buf, PAGE_SIZE, "%u\n",
+			READ_ONCE(fs_info->discard_ctl.iops_limit));
+}
+
+static ssize_t btrfs_discard_iops_limit_store(struct kobject *kobj,
+					      struct kobj_attribute *a,
+					      const char *buf, size_t len)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+	struct btrfs_discard_ctl *discard_ctl = &fs_info->discard_ctl;
+	u32 iops_limit;
+	int ret;
+
+	ret = kstrtou32(buf, 10, &iops_limit);
+	if (ret)
+		return -EINVAL;
+
+	WRITE_ONCE(discard_ctl->iops_limit, iops_limit);
+
+	return len;
+}
+BTRFS_ATTR_RW(discard, iops_limit, btrfs_discard_iops_limit_show,
+	      btrfs_discard_iops_limit_store);
+
+static ssize_t btrfs_discard_kbps_limit_show(struct kobject *kobj,
+					     struct kobj_attribute *a,
+					     char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return scnprintf(buf, PAGE_SIZE, "%u\n",
+			READ_ONCE(fs_info->discard_ctl.kbps_limit));
+}
+
+static ssize_t btrfs_discard_kbps_limit_store(struct kobject *kobj,
+					      struct kobj_attribute *a,
+					      const char *buf, size_t len)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+	struct btrfs_discard_ctl *discard_ctl = &fs_info->discard_ctl;
+	u32 kbps_limit;
+	int ret;
+
+	ret = kstrtou32(buf, 10, &kbps_limit);
+	if (ret)
+		return -EINVAL;
+
+	WRITE_ONCE(discard_ctl->kbps_limit, kbps_limit);
+
+	return len;
+}
+BTRFS_ATTR_RW(discard, kbps_limit, btrfs_discard_kbps_limit_show,
+	      btrfs_discard_kbps_limit_store);
+
+static ssize_t btrfs_discard_max_discard_size_show(struct kobject *kobj,
+						   struct kobj_attribute *a,
+						   char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return scnprintf(buf, PAGE_SIZE, "%llu\n",
+			READ_ONCE(fs_info->discard_ctl.max_discard_size));
+}
+
+static ssize_t btrfs_discard_max_discard_size_store(struct kobject *kobj,
+						    struct kobj_attribute *a,
+						    const char *buf, size_t len)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+	struct btrfs_discard_ctl *discard_ctl = &fs_info->discard_ctl;
+	u64 max_discard_size;
+	int ret;
+
+	ret = kstrtou64(buf, 10, &max_discard_size);
+	if (ret)
+		return -EINVAL;
+
+	WRITE_ONCE(discard_ctl->max_discard_size, max_discard_size);
+
+	return len;
+}
+BTRFS_ATTR_RW(discard, max_discard_size, btrfs_discard_max_discard_size_show,
+	      btrfs_discard_max_discard_size_store);
+
+static const struct attribute *discard_debug_attrs[] = {
+	BTRFS_ATTR_PTR(discard, discardable_bytes),
+	BTRFS_ATTR_PTR(discard, discardable_extents),
+	BTRFS_ATTR_PTR(discard, discard_bitmap_bytes),
+	BTRFS_ATTR_PTR(discard, discard_bytes_saved),
+	BTRFS_ATTR_PTR(discard, discard_extent_bytes),
+	BTRFS_ATTR_PTR(discard, iops_limit),
+	BTRFS_ATTR_PTR(discard, kbps_limit),
+	BTRFS_ATTR_PTR(discard, max_discard_size),
+	NULL,
+};
+
+/*
  * Runtime debugging exported via sysfs
  *
  * /sys/fs/btrfs/debug - applies to module or all filesystems
  * /sys/fs/btrfs/UUID  - applies only to the given filesystem
  */
+static const struct attribute *btrfs_debug_mount_attrs[] = {
+	NULL,
+};
+
 static struct attribute *btrfs_debug_feature_attrs[] = {
 	NULL
 };
@@ -338,7 +530,7 @@ static ssize_t btrfs_show_u64(u64 *value_ptr, spinlock_t *lock, char *buf)
 	val = *value_ptr;
 	if (lock)
 		spin_unlock(lock);
-	return snprintf(buf, PAGE_SIZE, "%llu\n", val);
+	return scnprintf(buf, PAGE_SIZE, "%llu\n", val);
 }
 
 static ssize_t global_rsv_size_show(struct kobject *kobj,
@@ -372,19 +564,19 @@ static ssize_t raid_bytes_show(struct kobject *kobj,
 
 {
 	struct btrfs_space_info *sinfo = to_space_info(kobj->parent);
-	struct btrfs_block_group_cache *block_group;
+	struct btrfs_block_group *block_group;
 	int index = btrfs_bg_flags_to_raid_index(to_raid_kobj(kobj)->flags);
 	u64 val = 0;
 
 	down_read(&sinfo->groups_sem);
 	list_for_each_entry(block_group, &sinfo->block_groups[index], list) {
 		if (&attr->attr == BTRFS_ATTR_PTR(raid, total_bytes))
-			val += block_group->key.offset;
+			val += block_group->length;
 		else
-			val += btrfs_block_group_used(&block_group->item);
+			val += block_group->used;
 	}
 	up_read(&sinfo->groups_sem);
-	return snprintf(buf, PAGE_SIZE, "%llu\n", val);
+	return scnprintf(buf, PAGE_SIZE, "%llu\n", val);
 }
 
 static struct attribute *raid_attrs[] = {
@@ -421,7 +613,7 @@ static ssize_t btrfs_space_info_show_total_bytes_pinned(struct kobject *kobj,
 {
 	struct btrfs_space_info *sinfo = to_space_info(kobj);
 	s64 val = percpu_counter_sum(&sinfo->total_bytes_pinned);
-	return snprintf(buf, PAGE_SIZE, "%lld\n", val);
+	return scnprintf(buf, PAGE_SIZE, "%lld\n", val);
 }
 
 SPACE_INFO_ATTR(flags);
@@ -478,7 +670,7 @@ static ssize_t btrfs_label_show(struct kobject *kobj,
 	ssize_t ret;
 
 	spin_lock(&fs_info->super_lock);
-	ret = snprintf(buf, PAGE_SIZE, label[0] ? "%s\n" : "%s", label);
+	ret = scnprintf(buf, PAGE_SIZE, label[0] ? "%s\n" : "%s", label);
 	spin_unlock(&fs_info->super_lock);
 
 	return ret;
@@ -526,7 +718,7 @@ static ssize_t btrfs_nodesize_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->nodesize);
+	return scnprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->nodesize);
 }
 
 BTRFS_ATTR(, nodesize, btrfs_nodesize_show);
@@ -536,8 +728,8 @@ static ssize_t btrfs_sectorsize_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n",
-			fs_info->super_copy->sectorsize);
+	return scnprintf(buf, PAGE_SIZE, "%u\n",
+			 fs_info->super_copy->sectorsize);
 }
 
 BTRFS_ATTR(, sectorsize, btrfs_sectorsize_show);
@@ -547,8 +739,7 @@ static ssize_t btrfs_clone_alignment_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n",
-			fs_info->super_copy->sectorsize);
+	return scnprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->sectorsize);
 }
 
 BTRFS_ATTR(, clone_alignment, btrfs_clone_alignment_show);
@@ -560,7 +751,7 @@ static ssize_t quota_override_show(struct kobject *kobj,
 	int quota_override;
 
 	quota_override = test_bit(BTRFS_FS_QUOTA_OVERRIDE, &fs_info->flags);
-	return snprintf(buf, PAGE_SIZE, "%d\n", quota_override);
+	return scnprintf(buf, PAGE_SIZE, "%d\n", quota_override);
 }
 
 static ssize_t quota_override_store(struct kobject *kobj,
@@ -598,12 +789,25 @@ static ssize_t btrfs_metadata_uuid_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 
-	return snprintf(buf, PAGE_SIZE, "%pU\n",
+	return scnprintf(buf, PAGE_SIZE, "%pU\n",
 			fs_info->fs_devices->metadata_uuid);
 }
 
 BTRFS_ATTR(, metadata_uuid, btrfs_metadata_uuid_show);
 
+static ssize_t btrfs_checksum_show(struct kobject *kobj,
+				   struct kobj_attribute *a, char *buf)
+{
+	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
+	u16 csum_type = btrfs_super_csum_type(fs_info->super_copy);
+
+	return scnprintf(buf, PAGE_SIZE, "%s (%s)\n",
+			btrfs_super_csum_name(csum_type),
+			crypto_shash_driver_name(fs_info->csum_shash));
+}
+
+BTRFS_ATTR(, checksum, btrfs_checksum_show);
+
 static const struct attribute *btrfs_attrs[] = {
 	BTRFS_ATTR_PTR(, label),
 	BTRFS_ATTR_PTR(, nodesize),
@@ -611,6 +815,7 @@ static const struct attribute *btrfs_attrs[] = {
 	BTRFS_ATTR_PTR(, clone_alignment),
 	BTRFS_ATTR_PTR(, quota_override),
 	BTRFS_ATTR_PTR(, metadata_uuid),
+	BTRFS_ATTR_PTR(, checksum),
 	NULL,
 };
 
@@ -695,10 +900,16 @@ static int addrm_unknown_feature_attrs(struct btrfs_fs_info *fs_info, bool add)
 
 static void __btrfs_sysfs_remove_fsid(struct btrfs_fs_devices *fs_devs)
 {
-	if (fs_devs->device_dir_kobj) {
-		kobject_del(fs_devs->device_dir_kobj);
-		kobject_put(fs_devs->device_dir_kobj);
-		fs_devs->device_dir_kobj = NULL;
+	if (fs_devs->devinfo_kobj) {
+		kobject_del(fs_devs->devinfo_kobj);
+		kobject_put(fs_devs->devinfo_kobj);
+		fs_devs->devinfo_kobj = NULL;
+	}
+
+	if (fs_devs->devices_kobj) {
+		kobject_del(fs_devs->devices_kobj);
+		kobject_put(fs_devs->devices_kobj);
+		fs_devs->devices_kobj = NULL;
 	}
 
 	if (fs_devs->fsid_kobj.state_initialized) {
@@ -732,10 +943,23 @@ void btrfs_sysfs_remove_mounted(struct btrfs_fs_info *fs_info)
 		kobject_del(fs_info->space_info_kobj);
 		kobject_put(fs_info->space_info_kobj);
 	}
+#ifdef CONFIG_BTRFS_DEBUG
+	if (fs_info->discard_debug_kobj) {
+		sysfs_remove_files(fs_info->discard_debug_kobj,
+				   discard_debug_attrs);
+		kobject_del(fs_info->discard_debug_kobj);
+		kobject_put(fs_info->discard_debug_kobj);
+	}
+	if (fs_info->debug_kobj) {
+		sysfs_remove_files(fs_info->debug_kobj, btrfs_debug_mount_attrs);
+		kobject_del(fs_info->debug_kobj);
+		kobject_put(fs_info->debug_kobj);
+	}
+#endif
 	addrm_unknown_feature_attrs(fs_info, false);
 	sysfs_remove_group(&fs_info->fs_devices->fsid_kobj, &btrfs_feature_attr_group);
 	sysfs_remove_files(&fs_info->fs_devices->fsid_kobj, btrfs_attrs);
-	btrfs_sysfs_rm_device_link(fs_info->fs_devices, NULL);
+	btrfs_sysfs_remove_devices_dir(fs_info->fs_devices, NULL);
 }
 
 static const char * const btrfs_feature_set_names[FEAT_MAX] = {
@@ -767,7 +991,7 @@ char *btrfs_printable_features(enum btrfs_feature_set set, u64 flags)
 			continue;
 
 		name = btrfs_feature_attrs[set][i].kobj_attr.attr.name;
-		len += snprintf(str + len, bufsize - len, "%s%s",
+		len += scnprintf(str + len, bufsize - len, "%s%s",
 				len ? "," : "", name);
 	}
 
@@ -822,7 +1046,7 @@ static void init_feature_attrs(void)
  * Create a sysfs entry for a given block group type at path
  * /sys/fs/btrfs/UUID/allocation/data/TYPE
  */
-void btrfs_sysfs_add_block_group_type(struct btrfs_block_group_cache *cache)
+void btrfs_sysfs_add_block_group_type(struct btrfs_block_group *cache)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	struct btrfs_space_info *space_info = cache->space_info;
@@ -924,75 +1148,158 @@ int btrfs_sysfs_add_space_info_type(struct btrfs_fs_info *fs_info,
 
 /* when one_device is NULL, it removes all device links */
 
-int btrfs_sysfs_rm_device_link(struct btrfs_fs_devices *fs_devices,
+int btrfs_sysfs_remove_devices_dir(struct btrfs_fs_devices *fs_devices,
 		struct btrfs_device *one_device)
 {
 	struct hd_struct *disk;
 	struct kobject *disk_kobj;
 
-	if (!fs_devices->device_dir_kobj)
+	if (!fs_devices->devices_kobj)
 		return -EINVAL;
 
-	if (one_device && one_device->bdev) {
-		disk = one_device->bdev->bd_part;
-		disk_kobj = &part_to_dev(disk)->kobj;
+	if (one_device) {
+		if (one_device->bdev) {
+			disk = one_device->bdev->bd_part;
+			disk_kobj = &part_to_dev(disk)->kobj;
+			sysfs_remove_link(fs_devices->devices_kobj,
+					  disk_kobj->name);
+		}
 
-		sysfs_remove_link(fs_devices->device_dir_kobj,
-						disk_kobj->name);
-	}
+		kobject_del(&one_device->devid_kobj);
+		kobject_put(&one_device->devid_kobj);
+
+		wait_for_completion(&one_device->kobj_unregister);
 
-	if (one_device)
 		return 0;
+	}
 
-	list_for_each_entry(one_device,
-			&fs_devices->devices, dev_list) {
-		if (!one_device->bdev)
-			continue;
-		disk = one_device->bdev->bd_part;
-		disk_kobj = &part_to_dev(disk)->kobj;
+	list_for_each_entry(one_device, &fs_devices->devices, dev_list) {
+
+		if (one_device->bdev) {
+			disk = one_device->bdev->bd_part;
+			disk_kobj = &part_to_dev(disk)->kobj;
+			sysfs_remove_link(fs_devices->devices_kobj,
+					  disk_kobj->name);
+		}
+		kobject_del(&one_device->devid_kobj);
+		kobject_put(&one_device->devid_kobj);
 
-		sysfs_remove_link(fs_devices->device_dir_kobj,
-						disk_kobj->name);
+		wait_for_completion(&one_device->kobj_unregister);
 	}
 
 	return 0;
 }
 
-int btrfs_sysfs_add_device(struct btrfs_fs_devices *fs_devs)
+static ssize_t btrfs_devinfo_in_fs_metadata_show(struct kobject *kobj,
+					         struct kobj_attribute *a,
+					         char *buf)
 {
-	if (!fs_devs->device_dir_kobj)
-		fs_devs->device_dir_kobj = kobject_create_and_add("devices",
-						&fs_devs->fsid_kobj);
+	int val;
+	struct btrfs_device *device = container_of(kobj, struct btrfs_device,
+						   devid_kobj);
 
-	if (!fs_devs->device_dir_kobj)
-		return -ENOMEM;
+	val = !!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
 
-	return 0;
+	return scnprintf(buf, PAGE_SIZE, "%d\n", val);
+}
+BTRFS_ATTR(devid, in_fs_metadata, btrfs_devinfo_in_fs_metadata_show);
+
+static ssize_t btrfs_devinfo_missing_show(struct kobject *kobj,
+					struct kobj_attribute *a, char *buf)
+{
+	int val;
+	struct btrfs_device *device = container_of(kobj, struct btrfs_device,
+						   devid_kobj);
+
+	val = !!test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state);
+
+	return scnprintf(buf, PAGE_SIZE, "%d\n", val);
+}
+BTRFS_ATTR(devid, missing, btrfs_devinfo_missing_show);
+
+static ssize_t btrfs_devinfo_replace_target_show(struct kobject *kobj,
+					         struct kobj_attribute *a,
+					         char *buf)
+{
+	int val;
+	struct btrfs_device *device = container_of(kobj, struct btrfs_device,
+						   devid_kobj);
+
+	val = !!test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
+
+	return scnprintf(buf, PAGE_SIZE, "%d\n", val);
+}
+BTRFS_ATTR(devid, replace_target, btrfs_devinfo_replace_target_show);
+
+static ssize_t btrfs_devinfo_writeable_show(struct kobject *kobj,
+					    struct kobj_attribute *a, char *buf)
+{
+	int val;
+	struct btrfs_device *device = container_of(kobj, struct btrfs_device,
+						   devid_kobj);
+
+	val = !!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
+
+	return scnprintf(buf, PAGE_SIZE, "%d\n", val);
+}
+BTRFS_ATTR(devid, writeable, btrfs_devinfo_writeable_show);
+
+static struct attribute *devid_attrs[] = {
+	BTRFS_ATTR_PTR(devid, in_fs_metadata),
+	BTRFS_ATTR_PTR(devid, missing),
+	BTRFS_ATTR_PTR(devid, replace_target),
+	BTRFS_ATTR_PTR(devid, writeable),
+	NULL
+};
+ATTRIBUTE_GROUPS(devid);
+
+static void btrfs_release_devid_kobj(struct kobject *kobj)
+{
+	struct btrfs_device *device = container_of(kobj, struct btrfs_device,
+						   devid_kobj);
+
+	memset(&device->devid_kobj, 0, sizeof(struct kobject));
+	complete(&device->kobj_unregister);
 }
 
-int btrfs_sysfs_add_device_link(struct btrfs_fs_devices *fs_devices,
+static struct kobj_type devid_ktype = {
+	.sysfs_ops	= &kobj_sysfs_ops,
+	.default_groups = devid_groups,
+	.release	= btrfs_release_devid_kobj,
+};
+
+int btrfs_sysfs_add_devices_dir(struct btrfs_fs_devices *fs_devices,
 				struct btrfs_device *one_device)
 {
 	int error = 0;
 	struct btrfs_device *dev;
 
 	list_for_each_entry(dev, &fs_devices->devices, dev_list) {
-		struct hd_struct *disk;
-		struct kobject *disk_kobj;
-
-		if (!dev->bdev)
-			continue;
 
 		if (one_device && one_device != dev)
 			continue;
 
-		disk = dev->bdev->bd_part;
-		disk_kobj = &part_to_dev(disk)->kobj;
+		if (dev->bdev) {
+			struct hd_struct *disk;
+			struct kobject *disk_kobj;
 
-		error = sysfs_create_link(fs_devices->device_dir_kobj,
-					  disk_kobj, disk_kobj->name);
-		if (error)
+			disk = dev->bdev->bd_part;
+			disk_kobj = &part_to_dev(disk)->kobj;
+
+			error = sysfs_create_link(fs_devices->devices_kobj,
+						  disk_kobj, disk_kobj->name);
+			if (error)
+				break;
+		}
+
+		init_completion(&dev->kobj_unregister);
+		error = kobject_init_and_add(&dev->devid_kobj, &devid_ktype,
+					     fs_devices->devinfo_kobj, "%llu",
+					     dev->devid);
+		if (error) {
+			kobject_put(&dev->devid_kobj);
 			break;
+		}
 	}
 
 	return error;
@@ -1024,27 +1331,58 @@ void btrfs_sysfs_update_sprout_fsid(struct btrfs_fs_devices *fs_devices,
 				"sysfs: failed to create fsid for sprout");
 }
 
+void btrfs_sysfs_update_devid(struct btrfs_device *device)
+{
+	char tmp[24];
+
+	snprintf(tmp, sizeof(tmp), "%llu", device->devid);
+
+	if (kobject_rename(&device->devid_kobj, tmp))
+		btrfs_warn(device->fs_devices->fs_info,
+			   "sysfs: failed to update devid for %llu",
+			   device->devid);
+}
+
 /* /sys/fs/btrfs/ entry */
 static struct kset *btrfs_kset;
 
 /*
+ * Creates:
+ *		/sys/fs/btrfs/UUID
+ *
  * Can be called by the device discovery thread.
- * And parent can be specified for seed device
  */
-int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs,
-				struct kobject *parent)
+int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs)
 {
 	int error;
 
 	init_completion(&fs_devs->kobj_unregister);
 	fs_devs->fsid_kobj.kset = btrfs_kset;
-	error = kobject_init_and_add(&fs_devs->fsid_kobj,
-				&btrfs_ktype, parent, "%pU", fs_devs->fsid);
+	error = kobject_init_and_add(&fs_devs->fsid_kobj, &btrfs_ktype, NULL,
+				     "%pU", fs_devs->fsid);
 	if (error) {
 		kobject_put(&fs_devs->fsid_kobj);
 		return error;
 	}
 
+	fs_devs->devices_kobj = kobject_create_and_add("devices",
+						       &fs_devs->fsid_kobj);
+	if (!fs_devs->devices_kobj) {
+		btrfs_err(fs_devs->fs_info,
+			  "failed to init sysfs device interface");
+		btrfs_sysfs_remove_fsid(fs_devs);
+		return -ENOMEM;
+	}
+
+	fs_devs->devinfo_kobj = kobject_create_and_add("devinfo",
+						       &fs_devs->fsid_kobj);
+	if (!fs_devs->devinfo_kobj) {
+		btrfs_err(fs_devs->fs_info,
+			  "failed to init sysfs devinfo kobject");
+		btrfs_sysfs_remove_fsid(fs_devs);
+		return -ENOMEM;
+	}
+
 	return 0;
 }
 
@@ -1056,13 +1394,13 @@ int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info)
 
 	btrfs_set_fs_info_ptr(fs_info);
 
-	error = btrfs_sysfs_add_device_link(fs_devs, NULL);
+	error = btrfs_sysfs_add_devices_dir(fs_devs, NULL);
 	if (error)
 		return error;
 
 	error = sysfs_create_files(fsid_kobj, btrfs_attrs);
 	if (error) {
-		btrfs_sysfs_rm_device_link(fs_devs, NULL);
+		btrfs_sysfs_remove_devices_dir(fs_devs, NULL);
 		return error;
 	}
 
@@ -1072,8 +1410,26 @@ int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info)
 		goto failure;
 
 #ifdef CONFIG_BTRFS_DEBUG
-	error = sysfs_create_group(fsid_kobj,
-				   &btrfs_debug_feature_attr_group);
+	fs_info->debug_kobj = kobject_create_and_add("debug", fsid_kobj);
+	if (!fs_info->debug_kobj) {
+		error = -ENOMEM;
+		goto failure;
+	}
+
+	error = sysfs_create_files(fs_info->debug_kobj, btrfs_debug_mount_attrs);
+	if (error)
+		goto failure;
+
+	/* Discard directory */
+	fs_info->discard_debug_kobj = kobject_create_and_add("discard",
+						     fs_info->debug_kobj);
+	if (!fs_info->discard_debug_kobj) {
+		error = -ENOMEM;
+		goto failure;
+	}
+
+	error = sysfs_create_files(fs_info->discard_debug_kobj,
+				   discard_debug_attrs);
 	if (error)
 		goto failure;
 #endif
@@ -1170,6 +1526,9 @@ void __cold btrfs_exit_sysfs(void)
 	sysfs_unmerge_group(&btrfs_kset->kobj,
 			    &btrfs_static_feature_attr_group);
 	sysfs_remove_group(&btrfs_kset->kobj, &btrfs_feature_attr_group);
+#ifdef CONFIG_BTRFS_DEBUG
+	sysfs_remove_group(&btrfs_kset->kobj, &btrfs_debug_feature_attr_group);
+#endif
 	kset_unregister(btrfs_kset);
 }
 
diff --git a/fs/btrfs/sysfs.h b/fs/btrfs/sysfs.h
index 610e9c36a94c..718a26c97833 100644
--- a/fs/btrfs/sysfs.h
+++ b/fs/btrfs/sysfs.h
@@ -14,13 +14,11 @@ enum btrfs_feature_set {
 
 char *btrfs_printable_features(enum btrfs_feature_set set, u64 flags);
 const char * const btrfs_feature_set_name(enum btrfs_feature_set set);
-int btrfs_sysfs_add_device_link(struct btrfs_fs_devices *fs_devices,
+int btrfs_sysfs_add_devices_dir(struct btrfs_fs_devices *fs_devices,
 		struct btrfs_device *one_device);
-int btrfs_sysfs_rm_device_link(struct btrfs_fs_devices *fs_devices,
+int btrfs_sysfs_remove_devices_dir(struct btrfs_fs_devices *fs_devices,
                 struct btrfs_device *one_device);
-int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs,
-				struct kobject *parent);
-int btrfs_sysfs_add_device(struct btrfs_fs_devices *fs_devs);
+int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs);
 void btrfs_sysfs_remove_fsid(struct btrfs_fs_devices *fs_devs);
 void btrfs_sysfs_update_sprout_fsid(struct btrfs_fs_devices *fs_devices,
 				    const u8 *fsid);
@@ -32,9 +30,10 @@ int __init btrfs_init_sysfs(void);
 void __cold btrfs_exit_sysfs(void);
 int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info);
 void btrfs_sysfs_remove_mounted(struct btrfs_fs_info *fs_info);
-void btrfs_sysfs_add_block_group_type(struct btrfs_block_group_cache *cache);
+void btrfs_sysfs_add_block_group_type(struct btrfs_block_group *cache);
 int btrfs_sysfs_add_space_info_type(struct btrfs_fs_info *fs_info,
 				    struct btrfs_space_info *space_info);
 void btrfs_sysfs_remove_space_info(struct btrfs_space_info *space_info);
+void btrfs_sysfs_update_devid(struct btrfs_device *device);
 
 #endif
diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c
index 99fe9bf3fdac..999c14e5d0bd 100644
--- a/fs/btrfs/tests/btrfs-tests.c
+++ b/fs/btrfs/tests/btrfs-tests.c
@@ -86,6 +86,27 @@ static void btrfs_destroy_test_fs(void)
 	unregister_filesystem(&test_type);
 }
 
+struct btrfs_device *btrfs_alloc_dummy_device(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_device *dev;
+
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		return ERR_PTR(-ENOMEM);
+
+	extent_io_tree_init(NULL, &dev->alloc_state, 0, NULL);
+	INIT_LIST_HEAD(&dev->dev_list);
+	list_add(&dev->dev_list, &fs_info->fs_devices->devices);
+
+	return dev;
+}
+
+static void btrfs_free_dummy_device(struct btrfs_device *dev)
+{
+	extent_io_tree_release(&dev->alloc_state);
+	kfree(dev);
+}
+
 struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize)
 {
 	struct btrfs_fs_info *fs_info = kzalloc(sizeof(struct btrfs_fs_info),
@@ -99,6 +120,8 @@ struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize)
 		kfree(fs_info);
 		return NULL;
 	}
+	INIT_LIST_HEAD(&fs_info->fs_devices->devices);
+
 	fs_info->super_copy = kzalloc(sizeof(struct btrfs_super_block),
 				      GFP_KERNEL);
 	if (!fs_info->super_copy) {
@@ -107,38 +130,10 @@ struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize)
 		return NULL;
 	}
 
+	btrfs_init_fs_info(fs_info);
+
 	fs_info->nodesize = nodesize;
 	fs_info->sectorsize = sectorsize;
-
-	if (init_srcu_struct(&fs_info->subvol_srcu)) {
-		kfree(fs_info->fs_devices);
-		kfree(fs_info->super_copy);
-		kfree(fs_info);
-		return NULL;
-	}
-
-	spin_lock_init(&fs_info->buffer_lock);
-	spin_lock_init(&fs_info->qgroup_lock);
-	spin_lock_init(&fs_info->super_lock);
-	spin_lock_init(&fs_info->fs_roots_radix_lock);
-	spin_lock_init(&fs_info->tree_mod_seq_lock);
-	mutex_init(&fs_info->qgroup_ioctl_lock);
-	mutex_init(&fs_info->qgroup_rescan_lock);
-	rwlock_init(&fs_info->tree_mod_log_lock);
-	fs_info->running_transaction = NULL;
-	fs_info->qgroup_tree = RB_ROOT;
-	fs_info->qgroup_ulist = NULL;
-	atomic64_set(&fs_info->tree_mod_seq, 0);
-	INIT_LIST_HEAD(&fs_info->dirty_qgroups);
-	INIT_LIST_HEAD(&fs_info->dead_roots);
-	INIT_LIST_HEAD(&fs_info->tree_mod_seq_list);
-	INIT_RADIX_TREE(&fs_info->buffer_radix, GFP_ATOMIC);
-	INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
-	extent_io_tree_init(fs_info, &fs_info->freed_extents[0],
-			    IO_TREE_FS_INFO_FREED_EXTENTS0, NULL);
-	extent_io_tree_init(fs_info, &fs_info->freed_extents[1],
-			    IO_TREE_FS_INFO_FREED_EXTENTS1, NULL);
-	fs_info->pinned_extents = &fs_info->freed_extents[0];
 	set_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state);
 
 	test_mnt->mnt_sb->s_fs_info = fs_info;
@@ -150,6 +145,7 @@ void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info)
 {
 	struct radix_tree_iter iter;
 	void **slot;
+	struct btrfs_device *dev, *tmp;
 
 	if (!fs_info)
 		return;
@@ -180,10 +176,16 @@ void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info)
 	}
 	spin_unlock(&fs_info->buffer_lock);
 
+	btrfs_mapping_tree_free(&fs_info->mapping_tree);
+	list_for_each_entry_safe(dev, tmp, &fs_info->fs_devices->devices,
+				 dev_list) {
+		btrfs_free_dummy_device(dev);
+	}
 	btrfs_free_qgroup_config(fs_info);
 	btrfs_free_fs_roots(fs_info);
-	cleanup_srcu_struct(&fs_info->subvol_srcu);
 	kfree(fs_info->super_copy);
+	btrfs_check_leaked_roots(fs_info);
+	btrfs_extent_buffer_leak_debug_check(fs_info);
 	kfree(fs_info->fs_devices);
 	kfree(fs_info);
 }
@@ -195,18 +197,14 @@ void btrfs_free_dummy_root(struct btrfs_root *root)
 	/* Will be freed by btrfs_free_fs_roots */
 	if (WARN_ON(test_bit(BTRFS_ROOT_IN_RADIX, &root->state)))
 		return;
-	if (root->node) {
-		/* One for allocate_extent_buffer */
-		free_extent_buffer(root->node);
-	}
-	kfree(root);
+	btrfs_put_root(root);
 }
 
-struct btrfs_block_group_cache *
+struct btrfs_block_group *
 btrfs_alloc_dummy_block_group(struct btrfs_fs_info *fs_info,
 			      unsigned long length)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 
 	cache = kzalloc(sizeof(*cache), GFP_KERNEL);
 	if (!cache)
@@ -218,9 +216,8 @@ btrfs_alloc_dummy_block_group(struct btrfs_fs_info *fs_info,
 		return NULL;
 	}
 
-	cache->key.objectid = 0;
-	cache->key.offset = length;
-	cache->key.type = BTRFS_BLOCK_GROUP_ITEM_KEY;
+	cache->start = 0;
+	cache->length = length;
 	cache->full_stripe_len = fs_info->sectorsize;
 	cache->fs_info = fs_info;
 
@@ -233,7 +230,7 @@ btrfs_alloc_dummy_block_group(struct btrfs_fs_info *fs_info,
 	return cache;
 }
 
-void btrfs_free_dummy_block_group(struct btrfs_block_group_cache *cache)
+void btrfs_free_dummy_block_group(struct btrfs_block_group *cache)
 {
 	if (!cache)
 		return;
diff --git a/fs/btrfs/tests/btrfs-tests.h b/fs/btrfs/tests/btrfs-tests.h
index ee277bbd939b..7a2d7ffbe30e 100644
--- a/fs/btrfs/tests/btrfs-tests.h
+++ b/fs/btrfs/tests/btrfs-tests.h
@@ -41,11 +41,12 @@ struct inode *btrfs_new_test_inode(void);
 struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize);
 void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info);
 void btrfs_free_dummy_root(struct btrfs_root *root);
-struct btrfs_block_group_cache *
+struct btrfs_block_group *
 btrfs_alloc_dummy_block_group(struct btrfs_fs_info *fs_info, unsigned long length);
-void btrfs_free_dummy_block_group(struct btrfs_block_group_cache *cache);
+void btrfs_free_dummy_block_group(struct btrfs_block_group *cache);
 void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans,
 			    struct btrfs_fs_info *fs_info);
+struct btrfs_device *btrfs_alloc_dummy_device(struct btrfs_fs_info *fs_info);
 #else
 static inline int btrfs_run_sanity_tests(void)
 {
diff --git a/fs/btrfs/tests/extent-io-tests.c b/fs/btrfs/tests/extent-io-tests.c
index 123d9a614357..df7ce874a74b 100644
--- a/fs/btrfs/tests/extent-io-tests.c
+++ b/fs/btrfs/tests/extent-io-tests.c
@@ -441,8 +441,17 @@ static int test_find_first_clear_extent_bit(void)
 	int ret = -EINVAL;
 
 	test_msg("running find_first_clear_extent_bit test");
+
 	extent_io_tree_init(NULL, &tree, IO_TREE_SELFTEST, NULL);
 
+	/* Test correct handling of empty tree */
+	find_first_clear_extent_bit(&tree, 0, &start, &end, CHUNK_TRIMMED);
+	if (start != 0 || end != -1) {
+		test_err(
+	"error getting a range from completely empty tree: start %llu end %llu",
+			 start, end);
+		goto out;
+	}
 	/*
 	 * Set 1M-4M alloc/discard and 32M-64M thus leaving a hole between
 	 * 4M-32M
diff --git a/fs/btrfs/tests/extent-map-tests.c b/fs/btrfs/tests/extent-map-tests.c
index 4a7f796c9900..57379e96ccc9 100644
--- a/fs/btrfs/tests/extent-map-tests.c
+++ b/fs/btrfs/tests/extent-map-tests.c
@@ -6,6 +6,9 @@
 #include <linux/types.h>
 #include "btrfs-tests.h"
 #include "../ctree.h"
+#include "../volumes.h"
+#include "../disk-io.h"
+#include "../block-group.h"
 
 static void free_extent_map_tree(struct extent_map_tree *em_tree)
 {
@@ -437,11 +440,153 @@ static int test_case_4(struct btrfs_fs_info *fs_info,
 	return ret;
 }
 
+struct rmap_test_vector {
+	u64 raid_type;
+	u64 physical_start;
+	u64 data_stripe_size;
+	u64 num_data_stripes;
+	u64 num_stripes;
+	/* Assume we won't have more than 5 physical stripes */
+	u64 data_stripe_phys_start[5];
+	bool expected_mapped_addr;
+	/* Physical to logical addresses */
+	u64 mapped_logical[5];
+};
+
+static int test_rmap_block(struct btrfs_fs_info *fs_info,
+			   struct rmap_test_vector *test)
+{
+	struct extent_map *em;
+	struct map_lookup *map = NULL;
+	u64 *logical = NULL;
+	int i, out_ndaddrs, out_stripe_len;
+	int ret;
+
+	em = alloc_extent_map();
+	if (!em) {
+		test_std_err(TEST_ALLOC_EXTENT_MAP);
+		return -ENOMEM;
+	}
+
+	map = kmalloc(map_lookup_size(test->num_stripes), GFP_KERNEL);
+	if (!map) {
+		kfree(em);
+		test_std_err(TEST_ALLOC_EXTENT_MAP);
+		return -ENOMEM;
+	}
+
+	set_bit(EXTENT_FLAG_FS_MAPPING, &em->flags);
+	/* Start at 4GiB logical address */
+	em->start = SZ_4G;
+	em->len = test->data_stripe_size * test->num_data_stripes;
+	em->block_len = em->len;
+	em->orig_block_len = test->data_stripe_size;
+	em->map_lookup = map;
+
+	map->num_stripes = test->num_stripes;
+	map->stripe_len = BTRFS_STRIPE_LEN;
+	map->type = test->raid_type;
+
+	for (i = 0; i < map->num_stripes; i++) {
+		struct btrfs_device *dev = btrfs_alloc_dummy_device(fs_info);
+
+		if (IS_ERR(dev)) {
+			test_err("cannot allocate device");
+			ret = PTR_ERR(dev);
+			goto out;
+		}
+		map->stripes[i].dev = dev;
+		map->stripes[i].physical = test->data_stripe_phys_start[i];
+	}
+
+	write_lock(&fs_info->mapping_tree.lock);
+	ret = add_extent_mapping(&fs_info->mapping_tree, em, 0);
+	write_unlock(&fs_info->mapping_tree.lock);
+	if (ret) {
+		test_err("error adding block group mapping to mapping tree");
+		goto out_free;
+	}
+
+	ret = btrfs_rmap_block(fs_info, em->start, btrfs_sb_offset(1),
+			       &logical, &out_ndaddrs, &out_stripe_len);
+	if (ret || (out_ndaddrs == 0 && test->expected_mapped_addr)) {
+		test_err("didn't rmap anything but expected %d",
+			 test->expected_mapped_addr);
+		goto out;
+	}
+
+	if (out_stripe_len != BTRFS_STRIPE_LEN) {
+		test_err("calculated stripe length doesn't match");
+		goto out;
+	}
+
+	if (out_ndaddrs != test->expected_mapped_addr) {
+		for (i = 0; i < out_ndaddrs; i++)
+			test_msg("mapped %llu", logical[i]);
+		test_err("unexpected number of mapped addresses: %d", out_ndaddrs);
+		goto out;
+	}
+
+	for (i = 0; i < out_ndaddrs; i++) {
+		if (logical[i] != test->mapped_logical[i]) {
+			test_err("unexpected logical address mapped");
+			goto out;
+		}
+	}
+
+	ret = 0;
+out:
+	write_lock(&fs_info->mapping_tree.lock);
+	remove_extent_mapping(&fs_info->mapping_tree, em);
+	write_unlock(&fs_info->mapping_tree.lock);
+	/* For us */
+	free_extent_map(em);
+out_free:
+	/* For the tree */
+	free_extent_map(em);
+	kfree(logical);
+	return ret;
+}
+
 int btrfs_test_extent_map(void)
 {
 	struct btrfs_fs_info *fs_info = NULL;
 	struct extent_map_tree *em_tree;
-	int ret = 0;
+	int ret = 0, i;
+	struct rmap_test_vector rmap_tests[] = {
+		{
+			/*
+			 * Test a chunk with 2 data stripes one of which
+			 * interesects the physical address of the super block
+			 * is correctly recognised.
+			 */
+			.raid_type = BTRFS_BLOCK_GROUP_RAID1,
+			.physical_start = SZ_64M - SZ_4M,
+			.data_stripe_size = SZ_256M,
+			.num_data_stripes = 2,
+			.num_stripes = 2,
+			.data_stripe_phys_start =
+				{SZ_64M - SZ_4M, SZ_64M - SZ_4M + SZ_256M},
+			.expected_mapped_addr = true,
+			.mapped_logical= {SZ_4G + SZ_4M}
+		},
+		{
+			/*
+			 * Test that out-of-range physical addresses are
+			 * ignored
+			 */
+
+			 /* SINGLE chunk type */
+			.raid_type = 0,
+			.physical_start = SZ_4G,
+			.data_stripe_size = SZ_256M,
+			.num_data_stripes = 1,
+			.num_stripes = 1,
+			.data_stripe_phys_start = {SZ_256M},
+			.expected_mapped_addr = false,
+			.mapped_logical = {0}
+		}
+	};
 
 	test_msg("running extent_map tests");
 
@@ -474,6 +619,13 @@ int btrfs_test_extent_map(void)
 		goto out;
 	ret = test_case_4(fs_info, em_tree);
 
+	test_msg("running rmap tests");
+	for (i = 0; i < ARRAY_SIZE(rmap_tests); i++) {
+		ret = test_rmap_block(fs_info, &rmap_tests[i]);
+		if (ret)
+			goto out;
+	}
+
 out:
 	kfree(em_tree);
 	btrfs_free_dummy_fs_info(fs_info);
diff --git a/fs/btrfs/tests/free-space-tests.c b/fs/btrfs/tests/free-space-tests.c
index 43ec7060fcd2..aebdf23f0cdd 100644
--- a/fs/btrfs/tests/free-space-tests.c
+++ b/fs/btrfs/tests/free-space-tests.c
@@ -17,7 +17,7 @@
  * entry and remove space from either end and the middle, and make sure we can
  * remove space that covers adjacent extent entries.
  */
-static int test_extents(struct btrfs_block_group_cache *cache)
+static int test_extents(struct btrfs_block_group *cache)
 {
 	int ret = 0;
 
@@ -87,8 +87,7 @@ static int test_extents(struct btrfs_block_group_cache *cache)
 	return 0;
 }
 
-static int test_bitmaps(struct btrfs_block_group_cache *cache,
-			u32 sectorsize)
+static int test_bitmaps(struct btrfs_block_group *cache, u32 sectorsize)
 {
 	u64 next_bitmap_offset;
 	int ret;
@@ -156,7 +155,7 @@ static int test_bitmaps(struct btrfs_block_group_cache *cache,
 }
 
 /* This is the high grade jackassery */
-static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache,
+static int test_bitmaps_and_extents(struct btrfs_block_group *cache,
 				    u32 sectorsize)
 {
 	u64 bitmap_offset = (u64)(BITS_PER_BITMAP * sectorsize);
@@ -331,7 +330,7 @@ static bool test_use_bitmap(struct btrfs_free_space_ctl *ctl,
 
 /* Used by test_steal_space_from_bitmap_to_extent(). */
 static int
-check_num_extents_and_bitmaps(const struct btrfs_block_group_cache *cache,
+check_num_extents_and_bitmaps(const struct btrfs_block_group *cache,
 			      const int num_extents,
 			      const int num_bitmaps)
 {
@@ -351,7 +350,7 @@ check_num_extents_and_bitmaps(const struct btrfs_block_group_cache *cache,
 }
 
 /* Used by test_steal_space_from_bitmap_to_extent(). */
-static int check_cache_empty(struct btrfs_block_group_cache *cache)
+static int check_cache_empty(struct btrfs_block_group *cache)
 {
 	u64 offset;
 	u64 max_extent_size;
@@ -393,7 +392,7 @@ static int check_cache_empty(struct btrfs_block_group_cache *cache)
  * requests.
  */
 static int
-test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
+test_steal_space_from_bitmap_to_extent(struct btrfs_block_group *cache,
 				       u32 sectorsize)
 {
 	int ret;
@@ -829,7 +828,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache,
 int btrfs_test_free_space_cache(u32 sectorsize, u32 nodesize)
 {
 	struct btrfs_fs_info *fs_info;
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	struct btrfs_root *root = NULL;
 	int ret = -ENOMEM;
 
diff --git a/fs/btrfs/tests/free-space-tree-tests.c b/fs/btrfs/tests/free-space-tree-tests.c
index bc92df977630..914eea5ba6a7 100644
--- a/fs/btrfs/tests/free-space-tree-tests.c
+++ b/fs/btrfs/tests/free-space-tree-tests.c
@@ -18,7 +18,7 @@ struct free_space_extent {
 
 static int __check_free_space_extents(struct btrfs_trans_handle *trans,
 				      struct btrfs_fs_info *fs_info,
-				      struct btrfs_block_group_cache *cache,
+				      struct btrfs_block_group *cache,
 				      struct btrfs_path *path,
 				      const struct free_space_extent * const extents,
 				      unsigned int num_extents)
@@ -48,7 +48,7 @@ static int __check_free_space_extents(struct btrfs_trans_handle *trans,
 	if (flags & BTRFS_FREE_SPACE_USING_BITMAPS) {
 		if (path->slots[0] != 0)
 			goto invalid;
-		end = cache->key.objectid + cache->key.offset;
+		end = cache->start + cache->length;
 		i = 0;
 		while (++path->slots[0] < btrfs_header_nritems(path->nodes[0])) {
 			btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
@@ -107,7 +107,7 @@ invalid:
 
 static int check_free_space_extents(struct btrfs_trans_handle *trans,
 				    struct btrfs_fs_info *fs_info,
-				    struct btrfs_block_group_cache *cache,
+				    struct btrfs_block_group *cache,
 				    struct btrfs_path *path,
 				    const struct free_space_extent * const extents,
 				    unsigned int num_extents)
@@ -150,12 +150,12 @@ static int check_free_space_extents(struct btrfs_trans_handle *trans,
 
 static int test_empty_block_group(struct btrfs_trans_handle *trans,
 				  struct btrfs_fs_info *fs_info,
-				  struct btrfs_block_group_cache *cache,
+				  struct btrfs_block_group *cache,
 				  struct btrfs_path *path,
 				  u32 alignment)
 {
 	const struct free_space_extent extents[] = {
-		{cache->key.objectid, cache->key.offset},
+		{cache->start, cache->length},
 	};
 
 	return check_free_space_extents(trans, fs_info, cache, path,
@@ -164,7 +164,7 @@ static int test_empty_block_group(struct btrfs_trans_handle *trans,
 
 static int test_remove_all(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
-			   struct btrfs_block_group_cache *cache,
+			   struct btrfs_block_group *cache,
 			   struct btrfs_path *path,
 			   u32 alignment)
 {
@@ -172,8 +172,8 @@ static int test_remove_all(struct btrfs_trans_handle *trans,
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, cache, path,
-					    cache->key.objectid,
-					    cache->key.offset);
+					    cache->start,
+					    cache->length);
 	if (ret) {
 		test_err("could not remove free space");
 		return ret;
@@ -185,18 +185,17 @@ static int test_remove_all(struct btrfs_trans_handle *trans,
 
 static int test_remove_beginning(struct btrfs_trans_handle *trans,
 				 struct btrfs_fs_info *fs_info,
-				 struct btrfs_block_group_cache *cache,
+				 struct btrfs_block_group *cache,
 				 struct btrfs_path *path,
 				 u32 alignment)
 {
 	const struct free_space_extent extents[] = {
-		{cache->key.objectid + alignment,
-			cache->key.offset - alignment},
+		{cache->start + alignment, cache->length - alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, cache, path,
-					    cache->key.objectid, alignment);
+					    cache->start, alignment);
 	if (ret) {
 		test_err("could not remove free space");
 		return ret;
@@ -209,19 +208,18 @@ static int test_remove_beginning(struct btrfs_trans_handle *trans,
 
 static int test_remove_end(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
-			   struct btrfs_block_group_cache *cache,
+			   struct btrfs_block_group *cache,
 			   struct btrfs_path *path,
 			   u32 alignment)
 {
 	const struct free_space_extent extents[] = {
-		{cache->key.objectid, cache->key.offset - alignment},
+		{cache->start, cache->length - alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, cache, path,
-					    cache->key.objectid +
-					    cache->key.offset - alignment,
-					    alignment);
+				    cache->start + cache->length - alignment,
+				    alignment);
 	if (ret) {
 		test_err("could not remove free space");
 		return ret;
@@ -233,19 +231,18 @@ static int test_remove_end(struct btrfs_trans_handle *trans,
 
 static int test_remove_middle(struct btrfs_trans_handle *trans,
 			      struct btrfs_fs_info *fs_info,
-			      struct btrfs_block_group_cache *cache,
+			      struct btrfs_block_group *cache,
 			      struct btrfs_path *path,
 			      u32 alignment)
 {
 	const struct free_space_extent extents[] = {
-		{cache->key.objectid, alignment},
-		{cache->key.objectid + 2 * alignment,
-			cache->key.offset - 2 * alignment},
+		{cache->start, alignment},
+		{cache->start + 2 * alignment, cache->length - 2 * alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, cache, path,
-					    cache->key.objectid + alignment,
+					    cache->start + alignment,
 					    alignment);
 	if (ret) {
 		test_err("could not remove free space");
@@ -258,24 +255,23 @@ static int test_remove_middle(struct btrfs_trans_handle *trans,
 
 static int test_merge_left(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
-			   struct btrfs_block_group_cache *cache,
+			   struct btrfs_block_group *cache,
 			   struct btrfs_path *path,
 			   u32 alignment)
 {
 	const struct free_space_extent extents[] = {
-		{cache->key.objectid, 2 * alignment},
+		{cache->start, 2 * alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, cache, path,
-					    cache->key.objectid,
-					    cache->key.offset);
+					    cache->start, cache->length);
 	if (ret) {
 		test_err("could not remove free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid,
+	ret = __add_to_free_space_tree(trans, cache, path, cache->start,
 				       alignment);
 	if (ret) {
 		test_err("could not add free space");
@@ -283,7 +279,7 @@ static int test_merge_left(struct btrfs_trans_handle *trans,
 	}
 
 	ret = __add_to_free_space_tree(trans, cache, path,
-				       cache->key.objectid + alignment,
+				       cache->start + alignment,
 				       alignment);
 	if (ret) {
 		test_err("could not add free space");
@@ -296,25 +292,24 @@ static int test_merge_left(struct btrfs_trans_handle *trans,
 
 static int test_merge_right(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
-			   struct btrfs_block_group_cache *cache,
+			   struct btrfs_block_group *cache,
 			   struct btrfs_path *path,
 			   u32 alignment)
 {
 	const struct free_space_extent extents[] = {
-		{cache->key.objectid + alignment, 2 * alignment},
+		{cache->start + alignment, 2 * alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, cache, path,
-					    cache->key.objectid,
-					    cache->key.offset);
+					    cache->start, cache->length);
 	if (ret) {
 		test_err("could not remove free space");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, cache, path,
-				       cache->key.objectid + 2 * alignment,
+				       cache->start + 2 * alignment,
 				       alignment);
 	if (ret) {
 		test_err("could not add free space");
@@ -322,7 +317,7 @@ static int test_merge_right(struct btrfs_trans_handle *trans,
 	}
 
 	ret = __add_to_free_space_tree(trans, cache, path,
-				       cache->key.objectid + alignment,
+				       cache->start + alignment,
 				       alignment);
 	if (ret) {
 		test_err("could not add free space");
@@ -335,24 +330,23 @@ static int test_merge_right(struct btrfs_trans_handle *trans,
 
 static int test_merge_both(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
-			   struct btrfs_block_group_cache *cache,
+			   struct btrfs_block_group *cache,
 			   struct btrfs_path *path,
 			   u32 alignment)
 {
 	const struct free_space_extent extents[] = {
-		{cache->key.objectid, 3 * alignment},
+		{cache->start, 3 * alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, cache, path,
-					    cache->key.objectid,
-					    cache->key.offset);
+					    cache->start, cache->length);
 	if (ret) {
 		test_err("could not remove free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid,
+	ret = __add_to_free_space_tree(trans, cache, path, cache->start,
 				       alignment);
 	if (ret) {
 		test_err("could not add free space");
@@ -360,16 +354,14 @@ static int test_merge_both(struct btrfs_trans_handle *trans,
 	}
 
 	ret = __add_to_free_space_tree(trans, cache, path,
-				       cache->key.objectid + 2 * alignment,
-				       alignment);
+				       cache->start + 2 * alignment, alignment);
 	if (ret) {
 		test_err("could not add free space");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, cache, path,
-				       cache->key.objectid + alignment,
-				       alignment);
+				       cache->start + alignment, alignment);
 	if (ret) {
 		test_err("could not add free space");
 		return ret;
@@ -381,26 +373,25 @@ static int test_merge_both(struct btrfs_trans_handle *trans,
 
 static int test_merge_none(struct btrfs_trans_handle *trans,
 			   struct btrfs_fs_info *fs_info,
-			   struct btrfs_block_group_cache *cache,
+			   struct btrfs_block_group *cache,
 			   struct btrfs_path *path,
 			   u32 alignment)
 {
 	const struct free_space_extent extents[] = {
-		{cache->key.objectid, alignment},
-		{cache->key.objectid + 2 * alignment, alignment},
-		{cache->key.objectid + 4 * alignment, alignment},
+		{cache->start, alignment},
+		{cache->start + 2 * alignment, alignment},
+		{cache->start + 4 * alignment, alignment},
 	};
 	int ret;
 
 	ret = __remove_from_free_space_tree(trans, cache, path,
-					    cache->key.objectid,
-					    cache->key.offset);
+					    cache->start, cache->length);
 	if (ret) {
 		test_err("could not remove free space");
 		return ret;
 	}
 
-	ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid,
+	ret = __add_to_free_space_tree(trans, cache, path, cache->start,
 				       alignment);
 	if (ret) {
 		test_err("could not add free space");
@@ -408,16 +399,14 @@ static int test_merge_none(struct btrfs_trans_handle *trans,
 	}
 
 	ret = __add_to_free_space_tree(trans, cache, path,
-				       cache->key.objectid + 4 * alignment,
-				       alignment);
+				       cache->start + 4 * alignment, alignment);
 	if (ret) {
 		test_err("could not add free space");
 		return ret;
 	}
 
 	ret = __add_to_free_space_tree(trans, cache, path,
-				       cache->key.objectid + 2 * alignment,
-				       alignment);
+				       cache->start + 2 * alignment, alignment);
 	if (ret) {
 		test_err("could not add free space");
 		return ret;
@@ -429,7 +418,7 @@ static int test_merge_none(struct btrfs_trans_handle *trans,
 
 typedef int (*test_func_t)(struct btrfs_trans_handle *,
 			   struct btrfs_fs_info *,
-			   struct btrfs_block_group_cache *,
+			   struct btrfs_block_group *,
 			   struct btrfs_path *,
 			   u32 alignment);
 
@@ -438,7 +427,7 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
 {
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_root *root = NULL;
-	struct btrfs_block_group_cache *cache = NULL;
+	struct btrfs_block_group *cache = NULL;
 	struct btrfs_trans_handle trans;
 	struct btrfs_path *path = NULL;
 	int ret;
@@ -463,9 +452,9 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize,
 	root->fs_info->tree_root = root;
 
 	root->node = alloc_test_extent_buffer(root->fs_info, nodesize);
-	if (!root->node) {
+	if (IS_ERR(root->node)) {
 		test_std_err(TEST_ALLOC_EXTENT_BUFFER);
-		ret = -ENOMEM;
+		ret = PTR_ERR(root->node);
 		goto out;
 	}
 	btrfs_set_header_level(root->node, 0);
diff --git a/fs/btrfs/tests/inode-tests.c b/fs/btrfs/tests/inode-tests.c
index 09ecf7dc7b08..24a8c714f56c 100644
--- a/fs/btrfs/tests/inode-tests.c
+++ b/fs/btrfs/tests/inode-tests.c
@@ -263,7 +263,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 
 	/* First with no extents */
 	BTRFS_I(inode)->root = root;
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, sectorsize);
 	if (IS_ERR(em)) {
 		em = NULL;
 		test_err("got an error when we shouldn't have");
@@ -283,7 +283,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	 */
 	setup_file_extents(root, sectorsize);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, (u64)-1, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, (u64)-1);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -305,7 +305,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -333,7 +333,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -356,7 +356,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	free_extent_map(em);
 
 	/* Regular extent */
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -384,7 +384,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	free_extent_map(em);
 
 	/* The next 3 are split extents */
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -413,7 +413,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -435,7 +435,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -469,7 +469,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	free_extent_map(em);
 
 	/* Prealloc extent */
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -498,7 +498,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	free_extent_map(em);
 
 	/* The next 3 are a half written prealloc extent */
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -528,7 +528,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -561,7 +561,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -596,7 +596,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	free_extent_map(em);
 
 	/* Now for the compressed extent */
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -630,7 +630,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	free_extent_map(em);
 
 	/* Split compressed extent */
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -665,7 +665,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -692,7 +692,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -727,8 +727,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	free_extent_map(em);
 
 	/* A hole between regular extents but no hole extent */
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset + 6,
-			sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset + 6, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -755,7 +754,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, SZ_4M, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, SZ_4M);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -788,7 +787,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize)
 	offset = em->start + em->len;
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -872,7 +871,7 @@ static int test_hole_first(u32 sectorsize, u32 nodesize)
 	insert_inode_item_key(root);
 	insert_extent(root, sectorsize, sectorsize, sectorsize, 0, sectorsize,
 		      sectorsize, BTRFS_FILE_EXTENT_REG, 0, 1);
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, 2 * sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, 2 * sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
@@ -894,8 +893,7 @@ static int test_hole_first(u32 sectorsize, u32 nodesize)
 	}
 	free_extent_map(em);
 
-	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, sectorsize,
-			2 * sectorsize, 0);
+	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, sectorsize, 2 * sectorsize);
 	if (IS_ERR(em)) {
 		test_err("got an error when we shouldn't have");
 		goto out;
diff --git a/fs/btrfs/tests/qgroup-tests.c b/fs/btrfs/tests/qgroup-tests.c
index 09aaca1efd62..ce1ca8e73c2d 100644
--- a/fs/btrfs/tests/qgroup-tests.c
+++ b/fs/btrfs/tests/qgroup-tests.c
@@ -484,9 +484,9 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 	 * *cough*backref walking code*cough*
 	 */
 	root->node = alloc_test_extent_buffer(root->fs_info, nodesize);
-	if (!root->node) {
+	if (IS_ERR(root->node)) {
 		test_err("couldn't allocate dummy buffer");
-		ret = -ENOMEM;
+		ret = PTR_ERR(root->node);
 		goto out;
 	}
 	btrfs_set_header_level(root->node, 0);
@@ -507,6 +507,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 		test_err("couldn't insert fs root %d", ret);
 		goto out;
 	}
+	btrfs_put_root(tmp_root);
 
 	tmp_root = btrfs_alloc_dummy_root(fs_info);
 	if (IS_ERR(tmp_root)) {
@@ -521,6 +522,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize)
 		test_err("couldn't insert fs root %d", ret);
 		goto out;
 	}
+	btrfs_put_root(tmp_root);
 
 	test_msg("running qgroup tests");
 	ret = test_no_shared_qgroup(root, sectorsize, nodesize);
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 8624bdee8c5b..2d5498136e5e 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -24,9 +24,79 @@
 
 #define BTRFS_ROOT_TRANS_TAG 0
 
+/*
+ * Transaction states and transitions
+ *
+ * No running transaction (fs tree blocks are not modified)
+ * |
+ * | To next stage:
+ * |  Call start_transaction() variants. Except btrfs_join_transaction_nostart().
+ * V
+ * Transaction N [[TRANS_STATE_RUNNING]]
+ * |
+ * | New trans handles can be attached to transaction N by calling all
+ * | start_transaction() variants.
+ * |
+ * | To next stage:
+ * |  Call btrfs_commit_transaction() on any trans handle attached to
+ * |  transaction N
+ * V
+ * Transaction N [[TRANS_STATE_COMMIT_START]]
+ * |
+ * | Will wait for previous running transaction to completely finish if there
+ * | is one
+ * |
+ * | Then one of the following happes:
+ * | - Wait for all other trans handle holders to release.
+ * |   The btrfs_commit_transaction() caller will do the commit work.
+ * | - Wait for current transaction to be committed by others.
+ * |   Other btrfs_commit_transaction() caller will do the commit work.
+ * |
+ * | At this stage, only btrfs_join_transaction*() variants can attach
+ * | to this running transaction.
+ * | All other variants will wait for current one to finish and attach to
+ * | transaction N+1.
+ * |
+ * | To next stage:
+ * |  Caller is chosen to commit transaction N, and all other trans handle
+ * |  haven been released.
+ * V
+ * Transaction N [[TRANS_STATE_COMMIT_DOING]]
+ * |
+ * | The heavy lifting transaction work is started.
+ * | From running delayed refs (modifying extent tree) to creating pending
+ * | snapshots, running qgroups.
+ * | In short, modify supporting trees to reflect modifications of subvolume
+ * | trees.
+ * |
+ * | At this stage, all start_transaction() calls will wait for this
+ * | transaction to finish and attach to transaction N+1.
+ * |
+ * | To next stage:
+ * |  Until all supporting trees are updated.
+ * V
+ * Transaction N [[TRANS_STATE_UNBLOCKED]]
+ * |						    Transaction N+1
+ * | All needed trees are modified, thus we only    [[TRANS_STATE_RUNNING]]
+ * | need to write them back to disk and update	    |
+ * | super blocks.				    |
+ * |						    |
+ * | At this stage, new transaction is allowed to   |
+ * | start.					    |
+ * | All new start_transaction() calls will be	    |
+ * | attached to transid N+1.			    |
+ * |						    |
+ * | To next stage:				    |
+ * |  Until all tree blocks are super blocks are    |
+ * |  written to block devices			    |
+ * V						    |
+ * Transaction N [[TRANS_STATE_COMPLETED]]	    V
+ *   All tree blocks and super blocks are written.  Transaction N+1
+ *   This transaction is finished and all its	    [[TRANS_STATE_COMMIT_START]]
+ *   data structures will be cleaned up.	    | Life goes on
+ */
 static const unsigned int btrfs_blocked_trans_types[TRANS_STATE_MAX] = {
 	[TRANS_STATE_RUNNING]		= 0U,
-	[TRANS_STATE_BLOCKED]		=  __TRANS_START,
 	[TRANS_STATE_COMMIT_START]	= (__TRANS_START | __TRANS_ATTACH),
 	[TRANS_STATE_COMMIT_DOING]	= (__TRANS_START |
 					   __TRANS_ATTACH |
@@ -51,6 +121,8 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction)
 		BUG_ON(!list_empty(&transaction->list));
 		WARN_ON(!RB_EMPTY_ROOT(
 				&transaction->delayed_refs.href_root.rb_root));
+		WARN_ON(!RB_EMPTY_ROOT(
+				&transaction->delayed_refs.dirty_extent_root));
 		if (transaction->delayed_refs.pending_csums)
 			btrfs_err(transaction->fs_info,
 				  "pending csums is %llu",
@@ -63,10 +135,10 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction)
 		 * discard the physical locations of the block groups.
 		 */
 		while (!list_empty(&transaction->deleted_bgs)) {
-			struct btrfs_block_group_cache *cache;
+			struct btrfs_block_group *cache;
 
 			cache = list_first_entry(&transaction->deleted_bgs,
-						 struct btrfs_block_group_cache,
+						 struct btrfs_block_group,
 						 bg_list);
 			list_del_init(&cache->bg_list);
 			btrfs_put_block_group_trimming(cache);
@@ -77,13 +149,14 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction)
 	}
 }
 
-static noinline void switch_commit_roots(struct btrfs_transaction *trans)
+static noinline void switch_commit_roots(struct btrfs_trans_handle *trans)
 {
+	struct btrfs_transaction *cur_trans = trans->transaction;
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_root *root, *tmp;
 
 	down_write(&fs_info->commit_root_sem);
-	list_for_each_entry_safe(root, tmp, &trans->switch_commits,
+	list_for_each_entry_safe(root, tmp, &cur_trans->switch_commits,
 				 dirty_list) {
 		list_del_init(&root->dirty_list);
 		free_extent_buffer(root->commit_root);
@@ -95,16 +168,17 @@ static noinline void switch_commit_roots(struct btrfs_transaction *trans)
 	}
 
 	/* We can free old roots now. */
-	spin_lock(&trans->dropped_roots_lock);
-	while (!list_empty(&trans->dropped_roots)) {
-		root = list_first_entry(&trans->dropped_roots,
+	spin_lock(&cur_trans->dropped_roots_lock);
+	while (!list_empty(&cur_trans->dropped_roots)) {
+		root = list_first_entry(&cur_trans->dropped_roots,
 					struct btrfs_root, root_list);
 		list_del_init(&root->root_list);
-		spin_unlock(&trans->dropped_roots_lock);
+		spin_unlock(&cur_trans->dropped_roots_lock);
+		btrfs_free_log(trans, root);
 		btrfs_drop_and_free_fs_root(fs_info, root);
-		spin_lock(&trans->dropped_roots_lock);
+		spin_lock(&cur_trans->dropped_roots_lock);
 	}
-	spin_unlock(&trans->dropped_roots_lock);
+	spin_unlock(&cur_trans->dropped_roots_lock);
 	up_write(&fs_info->commit_root_sem);
 }
 
@@ -147,7 +221,7 @@ void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans)
 	WARN_ON_ONCE(!list_empty(&trans->new_bgs));
 
 	btrfs_block_rsv_release(fs_info, &fs_info->chunk_block_rsv,
-				trans->chunk_bytes_reserved);
+				trans->chunk_bytes_reserved, NULL);
 	trans->chunk_bytes_reserved = 0;
 }
 
@@ -169,7 +243,7 @@ loop:
 
 	cur_trans = fs_info->running_transaction;
 	if (cur_trans) {
-		if (cur_trans->aborted) {
+		if (TRANS_ABORTED(cur_trans)) {
 			spin_unlock(&fs_info->trans_lock);
 			return cur_trans->aborted;
 		}
@@ -262,6 +336,8 @@ loop:
 	list_add_tail(&cur_trans->list, &fs_info->trans_list);
 	extent_io_tree_init(fs_info, &cur_trans->dirty_pages,
 			IO_TREE_TRANS_DIRTY_PAGES, fs_info->btree_inode);
+	extent_io_tree_init(fs_info, &cur_trans->pinned_extents,
+			IO_TREE_FS_PINNED_EXTENTS, NULL);
 	fs_info->generation++;
 	cur_trans->transid = fs_info->generation;
 	fs_info->running_transaction = cur_trans;
@@ -383,9 +459,9 @@ int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
 
 static inline int is_transaction_blocked(struct btrfs_transaction *trans)
 {
-	return (trans->state >= TRANS_STATE_BLOCKED &&
+	return (trans->state >= TRANS_STATE_COMMIT_START &&
 		trans->state < TRANS_STATE_UNBLOCKED &&
-		!trans->aborted);
+		!TRANS_ABORTED(trans));
 }
 
 /* wait for commit against the current transaction to become unblocked
@@ -404,7 +480,7 @@ static void wait_current_trans(struct btrfs_fs_info *fs_info)
 
 		wait_event(fs_info->transaction_wait,
 			   cur_trans->state >= TRANS_STATE_UNBLOCKED ||
-			   cur_trans->aborted);
+			   TRANS_ABORTED(cur_trans));
 		btrfs_put_transaction(cur_trans);
 	} else {
 		spin_unlock(&fs_info->trans_lock);
@@ -570,7 +646,7 @@ again:
 	INIT_LIST_HEAD(&h->new_bgs);
 
 	smp_mb();
-	if (cur_trans->state >= TRANS_STATE_BLOCKED &&
+	if (cur_trans->state >= TRANS_STATE_COMMIT_START &&
 	    may_wait_transaction(fs_info, type)) {
 		current->journal_info = h;
 		btrfs_commit_transaction(h);
@@ -586,10 +662,19 @@ again:
 	}
 
 got_it:
-	btrfs_record_root_in_trans(h, root);
-
 	if (!current->journal_info)
 		current->journal_info = h;
+
+	/*
+	 * btrfs_record_root_in_trans() needs to alloc new extents, and may
+	 * call btrfs_join_transaction() while we're also starting a
+	 * transaction.
+	 *
+	 * Thus it need to be called after current->journal_info initialized,
+	 * or we can deadlock.
+	 */
+	btrfs_record_root_in_trans(h, root);
+
 	return h;
 
 join_fail:
@@ -599,7 +684,7 @@ join_fail:
 alloc_fail:
 	if (num_bytes)
 		btrfs_block_rsv_release(fs_info, &fs_info->trans_block_rsv,
-					num_bytes);
+					num_bytes, NULL);
 reserve_fail:
 	btrfs_qgroup_free_meta_pertrans(root, qgroup_reserved);
 	return ERR_PTR(ret);
@@ -659,7 +744,7 @@ struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root)
 				 true);
 }
 
-struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root)
+struct btrfs_trans_handle *btrfs_join_transaction_spacecache(struct btrfs_root *root)
 {
 	return start_transaction(root, 0, TRANS_JOIN_NOLOCK,
 				 BTRFS_RESERVE_NO_FLUSH, true);
@@ -798,7 +883,7 @@ int btrfs_should_end_transaction(struct btrfs_trans_handle *trans)
 	struct btrfs_transaction *cur_trans = trans->transaction;
 
 	smp_mb();
-	if (cur_trans->state >= TRANS_STATE_BLOCKED ||
+	if (cur_trans->state >= TRANS_STATE_COMMIT_START ||
 	    cur_trans->delayed_refs.flushing)
 		return 1;
 
@@ -822,7 +907,7 @@ static void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans)
 	trace_btrfs_space_reservation(fs_info, "transaction",
 				      trans->transid, trans->bytes_reserved, 0);
 	btrfs_block_rsv_release(fs_info, trans->block_rsv,
-				trans->bytes_reserved);
+				trans->bytes_reserved, NULL);
 	trans->bytes_reserved = 0;
 }
 
@@ -831,7 +916,6 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_fs_info *info = trans->fs_info;
 	struct btrfs_transaction *cur_trans = trans->transaction;
-	int lock = (trans->type != TRANS_JOIN_NOLOCK);
 	int err = 0;
 
 	if (refcount_read(&trans->use_count) > 1) {
@@ -847,13 +931,6 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 
 	btrfs_trans_release_chunk_metadata(trans);
 
-	if (lock && READ_ONCE(cur_trans->state) == TRANS_STATE_BLOCKED) {
-		if (throttle)
-			return btrfs_commit_transaction(trans);
-		else
-			wake_up_process(info->transaction_kthread);
-	}
-
 	if (trans->type & __TRANS_FREEZABLE)
 		sb_end_intwrite(info->sb);
 
@@ -871,7 +948,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 	if (throttle)
 		btrfs_run_delayed_iputs(info);
 
-	if (trans->aborted ||
+	if (TRANS_ABORTED(trans) ||
 	    test_bit(BTRFS_FS_STATE_ERROR, &info->fs_state)) {
 		wake_up_process(info->transaction_kthread);
 		err = -EIO;
@@ -990,7 +1067,7 @@ static int __btrfs_wait_marked_extents(struct btrfs_fs_info *fs_info,
 	return werr;
 }
 
-int btrfs_wait_extents(struct btrfs_fs_info *fs_info,
+static int btrfs_wait_extents(struct btrfs_fs_info *fs_info,
 		       struct extent_io_tree *dirty_pages)
 {
 	bool errors = false;
@@ -1196,8 +1273,10 @@ void btrfs_add_dead_root(struct btrfs_root *root)
 	struct btrfs_fs_info *fs_info = root->fs_info;
 
 	spin_lock(&fs_info->trans_lock);
-	if (list_empty(&root->root_list))
+	if (list_empty(&root->root_list)) {
+		btrfs_grab_root(root);
 		list_add_tail(&root->root_list, &fs_info->dead_roots);
+	}
 	spin_unlock(&fs_info->trans_lock);
 }
 
@@ -1359,7 +1438,7 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,
 	ret = commit_cowonly_roots(trans);
 	if (ret)
 		goto out;
-	switch_commit_roots(trans->transaction);
+	switch_commit_roots(trans);
 	ret = btrfs_write_and_wait_transaction(trans);
 	if (ret)
 		btrfs_handle_fs_error(fs_info, ret,
@@ -1411,7 +1490,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	u64 index = 0;
 	u64 objectid;
 	u64 root_flags;
-	uuid_le new_uuid;
 
 	ASSERT(pending->path);
 	path = pending->path;
@@ -1504,8 +1582,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 
 	btrfs_set_root_generation_v2(new_root_item,
 			trans->transid);
-	uuid_le_gen(&new_uuid);
-	memcpy(new_root_item->uuid, new_uuid.b, BTRFS_UUID_SIZE);
+	generate_random_guid(new_root_item->uuid);
 	memcpy(new_root_item->parent_uuid, root->root_item.uuid,
 			BTRFS_UUID_SIZE);
 	if (!(root_flags & BTRFS_ROOT_SUBVOL_RDONLY)) {
@@ -1567,7 +1644,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	}
 
 	key.offset = (u64)-1;
-	pending->snap = btrfs_read_fs_root_no_name(fs_info, &key);
+	pending->snap = btrfs_get_fs_root(fs_info, &key, true);
 	if (IS_ERR(pending->snap)) {
 		ret = PTR_ERR(pending->snap);
 		btrfs_abort_transaction(trans, ret);
@@ -1616,7 +1693,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 		btrfs_abort_transaction(trans, ret);
 		goto fail;
 	}
-	ret = btrfs_uuid_tree_add(trans, new_uuid.b, BTRFS_UUID_KEY_SUBVOL,
+	ret = btrfs_uuid_tree_add(trans, new_root_item->uuid,
+				  BTRFS_UUID_KEY_SUBVOL,
 				  objectid);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
@@ -1728,7 +1806,8 @@ static void wait_current_trans_commit_start(struct btrfs_fs_info *fs_info,
 					    struct btrfs_transaction *trans)
 {
 	wait_event(fs_info->transaction_blocked_wait,
-		   trans->state >= TRANS_STATE_COMMIT_START || trans->aborted);
+		   trans->state >= TRANS_STATE_COMMIT_START ||
+		   TRANS_ABORTED(trans));
 }
 
 /*
@@ -1740,7 +1819,8 @@ static void wait_current_trans_commit_start_and_unblock(
 					struct btrfs_transaction *trans)
 {
 	wait_event(fs_info->transaction_wait,
-		   trans->state >= TRANS_STATE_UNBLOCKED || trans->aborted);
+		   trans->state >= TRANS_STATE_UNBLOCKED ||
+		   TRANS_ABORTED(trans));
 }
 
 /*
@@ -1875,7 +1955,7 @@ static void cleanup_transaction(struct btrfs_trans_handle *trans, int err)
 static void btrfs_cleanup_pending_block_groups(struct btrfs_trans_handle *trans)
 {
        struct btrfs_fs_info *fs_info = trans->fs_info;
-       struct btrfs_block_group_cache *block_group, *tmp;
+       struct btrfs_block_group *block_group, *tmp;
 
        list_for_each_entry_safe(block_group, tmp, &trans->new_bgs, bg_list) {
                btrfs_delayed_refs_rsv_release(fs_info, 1);
@@ -1949,8 +2029,18 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	struct btrfs_transaction *prev_trans = NULL;
 	int ret;
 
+	ASSERT(refcount_read(&trans->use_count) == 1);
+
+	/*
+	 * Some places just start a transaction to commit it.  We need to make
+	 * sure that if this commit fails that the abort code actually marks the
+	 * transaction as failed, so set trans->dirty to make the abort code do
+	 * the right thing.
+	 */
+	trans->dirty = true;
+
 	/* Stop the commit early if ->aborted is set */
-	if (unlikely(READ_ONCE(cur_trans->aborted))) {
+	if (TRANS_ABORTED(cur_trans)) {
 		ret = cur_trans->aborted;
 		btrfs_end_transaction(trans);
 		return ret;
@@ -2024,7 +2114,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 
 		wait_for_commit(cur_trans);
 
-		if (unlikely(cur_trans->aborted))
+		if (TRANS_ABORTED(cur_trans))
 			ret = cur_trans->aborted;
 
 		btrfs_put_transaction(cur_trans);
@@ -2043,7 +2133,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 			spin_unlock(&fs_info->trans_lock);
 
 			wait_for_commit(prev_trans);
-			ret = prev_trans->aborted;
+			ret = READ_ONCE(prev_trans->aborted);
 
 			btrfs_put_transaction(prev_trans);
 			if (ret)
@@ -2097,8 +2187,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	wait_event(cur_trans->writer_wait,
 		   atomic_read(&cur_trans->num_writers) == 1);
 
-	/* ->aborted might be set after the previous check, so check it */
-	if (unlikely(READ_ONCE(cur_trans->aborted))) {
+	if (TRANS_ABORTED(cur_trans)) {
 		ret = cur_trans->aborted;
 		goto scrub_continue;
 	}
@@ -2115,10 +2204,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	 * core function of the snapshot creation.
 	 */
 	ret = create_pending_snapshots(trans);
-	if (ret) {
-		mutex_unlock(&fs_info->reloc_mutex);
-		goto scrub_continue;
-	}
+	if (ret)
+		goto unlock_reloc;
 
 	/*
 	 * We insert the dir indexes of the snapshots and update the inode
@@ -2131,16 +2218,12 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	 * the nodes and leaves.
 	 */
 	ret = btrfs_run_delayed_items(trans);
-	if (ret) {
-		mutex_unlock(&fs_info->reloc_mutex);
-		goto scrub_continue;
-	}
+	if (ret)
+		goto unlock_reloc;
 
 	ret = btrfs_run_delayed_refs(trans, (unsigned long)-1);
-	if (ret) {
-		mutex_unlock(&fs_info->reloc_mutex);
-		goto scrub_continue;
-	}
+	if (ret)
+		goto unlock_reloc;
 
 	/*
 	 * make sure none of the code above managed to slip in a
@@ -2166,11 +2249,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	mutex_lock(&fs_info->tree_log_mutex);
 
 	ret = commit_fs_roots(trans);
-	if (ret) {
-		mutex_unlock(&fs_info->tree_log_mutex);
-		mutex_unlock(&fs_info->reloc_mutex);
-		goto scrub_continue;
-	}
+	if (ret)
+		goto unlock_tree_log;
 
 	/*
 	 * Since the transaction is done, we can apply the pending changes
@@ -2188,39 +2268,28 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	 * new delayed refs. Must handle them or qgroup can be wrong.
 	 */
 	ret = btrfs_run_delayed_refs(trans, (unsigned long)-1);
-	if (ret) {
-		mutex_unlock(&fs_info->tree_log_mutex);
-		mutex_unlock(&fs_info->reloc_mutex);
-		goto scrub_continue;
-	}
+	if (ret)
+		goto unlock_tree_log;
 
 	/*
 	 * Since fs roots are all committed, we can get a quite accurate
 	 * new_roots. So let's do quota accounting.
 	 */
 	ret = btrfs_qgroup_account_extents(trans);
-	if (ret < 0) {
-		mutex_unlock(&fs_info->tree_log_mutex);
-		mutex_unlock(&fs_info->reloc_mutex);
-		goto scrub_continue;
-	}
+	if (ret < 0)
+		goto unlock_tree_log;
 
 	ret = commit_cowonly_roots(trans);
-	if (ret) {
-		mutex_unlock(&fs_info->tree_log_mutex);
-		mutex_unlock(&fs_info->reloc_mutex);
-		goto scrub_continue;
-	}
+	if (ret)
+		goto unlock_tree_log;
 
 	/*
 	 * The tasks which save the space cache and inode cache may also
 	 * update ->aborted, check it.
 	 */
-	if (unlikely(READ_ONCE(cur_trans->aborted))) {
+	if (TRANS_ABORTED(cur_trans)) {
 		ret = cur_trans->aborted;
-		mutex_unlock(&fs_info->tree_log_mutex);
-		mutex_unlock(&fs_info->reloc_mutex);
-		goto scrub_continue;
+		goto unlock_tree_log;
 	}
 
 	btrfs_prepare_extent_commit(fs_info);
@@ -2237,7 +2306,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	list_add_tail(&fs_info->chunk_root->dirty_list,
 		      &cur_trans->switch_commits);
 
-	switch_commit_roots(cur_trans);
+	switch_commit_roots(trans);
 
 	ASSERT(list_empty(&cur_trans->dirty_bgs));
 	ASSERT(list_empty(&cur_trans->io_bgs));
@@ -2267,6 +2336,10 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	if (ret) {
 		btrfs_handle_fs_error(fs_info, ret,
 				      "Error while writing out transaction");
+		/*
+		 * reloc_mutex has been unlocked, tree_log_mutex is still held
+		 * but we can't jump to unlock_tree_log causing double unlock
+		 */
 		mutex_unlock(&fs_info->tree_log_mutex);
 		goto scrub_continue;
 	}
@@ -2315,6 +2388,10 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 
 	return ret;
 
+unlock_tree_log:
+	mutex_unlock(&fs_info->tree_log_mutex);
+unlock_reloc:
+	mutex_unlock(&fs_info->reloc_mutex);
 scrub_continue:
 	btrfs_scrub_continue(fs_info);
 cleanup_transaction:
@@ -2358,13 +2435,18 @@ int btrfs_clean_one_deleted_snapshot(struct btrfs_root *root)
 	btrfs_debug(fs_info, "cleaner removing %llu", root->root_key.objectid);
 
 	btrfs_kill_all_delayed_nodes(root);
+	if (root->ino_cache_inode) {
+		iput(root->ino_cache_inode);
+		root->ino_cache_inode = NULL;
+	}
 
 	if (btrfs_header_backref_rev(root->node) <
 			BTRFS_MIXED_BACKREF_REV)
-		ret = btrfs_drop_snapshot(root, NULL, 0, 0);
+		ret = btrfs_drop_snapshot(root, 0, 0);
 	else
-		ret = btrfs_drop_snapshot(root, NULL, 1, 0);
+		ret = btrfs_drop_snapshot(root, 1, 0);
 
+	btrfs_put_root(root);
 	return (ret < 0) ? 0 : 1;
 }
 
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 2c5a6f6e5bb0..31ae8d273065 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -13,7 +13,6 @@
 
 enum btrfs_trans_state {
 	TRANS_STATE_RUNNING,
-	TRANS_STATE_BLOCKED,
 	TRANS_STATE_COMMIT_START,
 	TRANS_STATE_COMMIT_DOING,
 	TRANS_STATE_UNBLOCKED,
@@ -72,6 +71,7 @@ struct btrfs_transaction {
 	 */
 	struct list_head io_bgs;
 	struct list_head dropped_roots;
+	struct extent_io_tree pinned_extents;
 
 	/*
 	 * we need to make sure block group deletion doesn't race with
@@ -116,6 +116,10 @@ struct btrfs_trans_handle {
 	struct btrfs_block_rsv *orig_rsv;
 	refcount_t use_count;
 	unsigned int type;
+	/*
+	 * Error code of transaction abort, set outside of locks and must use
+	 * the READ_ONCE/WRITE_ONCE access
+	 */
 	short aborted;
 	bool adding_csums;
 	bool allocating_chunk;
@@ -127,6 +131,14 @@ struct btrfs_trans_handle {
 	struct list_head new_bgs;
 };
 
+/*
+ * The abort status can be changed between calls and is not protected by locks.
+ * This accepts btrfs_transaction and btrfs_trans_handle as types. Once it's
+ * set to a non-zero value it does not change, so the macro should be in checks
+ * but is not necessary for further reads of the value.
+ */
+#define TRANS_ABORTED(trans)		(unlikely(READ_ONCE((trans)->aborted)))
+
 struct btrfs_pending_snapshot {
 	struct dentry *dentry;
 	struct inode *dir;
@@ -184,7 +196,7 @@ struct btrfs_trans_handle *btrfs_start_transaction_fallback_global_rsv(
 					unsigned int num_items,
 					int min_factor);
 struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root);
-struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root);
+struct btrfs_trans_handle *btrfs_join_transaction_spacecache(struct btrfs_root *root);
 struct btrfs_trans_handle *btrfs_join_transaction_nostart(struct btrfs_root *root);
 struct btrfs_trans_handle *btrfs_attach_transaction(struct btrfs_root *root);
 struct btrfs_trans_handle *btrfs_attach_transaction_barrier(
@@ -218,8 +230,6 @@ int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root);
 int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info,
 				struct extent_io_tree *dirty_pages, int mark);
-int btrfs_wait_extents(struct btrfs_fs_info *fs_info,
-		       struct extent_io_tree *dirty_pages);
 int btrfs_wait_tree_log_extents(struct btrfs_root *root, int mark);
 int btrfs_transaction_blocked(struct btrfs_fs_info *info);
 int btrfs_transaction_in_commit(struct btrfs_fs_info *info);
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 076d5b8014fb..a92f8a6dd192 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -23,6 +23,7 @@
 #include "disk-io.h"
 #include "compression.h"
 #include "volumes.h"
+#include "misc.h"
 
 /*
  * Error message should follow the following format:
@@ -124,6 +125,74 @@ static u64 file_extent_end(struct extent_buffer *leaf,
 	return end;
 }
 
+/*
+ * Customized report for dir_item, the only new important information is
+ * key->objectid, which represents inode number
+ */
+__printf(3, 4)
+__cold
+static void dir_item_err(const struct extent_buffer *eb, int slot,
+			 const char *fmt, ...)
+{
+	const struct btrfs_fs_info *fs_info = eb->fs_info;
+	struct btrfs_key key;
+	struct va_format vaf;
+	va_list args;
+
+	btrfs_item_key_to_cpu(eb, &key, slot);
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	btrfs_crit(fs_info,
+		"corrupt %s: root=%llu block=%llu slot=%d ino=%llu, %pV",
+		btrfs_header_level(eb) == 0 ? "leaf" : "node",
+		btrfs_header_owner(eb), btrfs_header_bytenr(eb), slot,
+		key.objectid, &vaf);
+	va_end(args);
+}
+
+/*
+ * This functions checks prev_key->objectid, to ensure current key and prev_key
+ * share the same objectid as inode number.
+ *
+ * This is to detect missing INODE_ITEM in subvolume trees.
+ *
+ * Return true if everything is OK or we don't need to check.
+ * Return false if anything is wrong.
+ */
+static bool check_prev_ino(struct extent_buffer *leaf,
+			   struct btrfs_key *key, int slot,
+			   struct btrfs_key *prev_key)
+{
+	/* No prev key, skip check */
+	if (slot == 0)
+		return true;
+
+	/* Only these key->types needs to be checked */
+	ASSERT(key->type == BTRFS_XATTR_ITEM_KEY ||
+	       key->type == BTRFS_INODE_REF_KEY ||
+	       key->type == BTRFS_DIR_INDEX_KEY ||
+	       key->type == BTRFS_DIR_ITEM_KEY ||
+	       key->type == BTRFS_EXTENT_DATA_KEY);
+
+	/*
+	 * Only subvolume trees along with their reloc trees need this check.
+	 * Things like log tree doesn't follow this ino requirement.
+	 */
+	if (!is_fstree(btrfs_header_owner(leaf)))
+		return true;
+
+	if (key->objectid == prev_key->objectid)
+		return true;
+
+	/* Error found */
+	dir_item_err(leaf, slot,
+		"invalid previous key objectid, have %llu expect %llu",
+		prev_key->objectid, key->objectid);
+	return false;
+}
 static int check_extent_data_item(struct extent_buffer *leaf,
 				  struct btrfs_key *key, int slot,
 				  struct btrfs_key *prev_key)
@@ -141,13 +210,33 @@ static int check_extent_data_item(struct extent_buffer *leaf,
 		return -EUCLEAN;
 	}
 
+	/*
+	 * Previous key must have the same key->objectid (ino).
+	 * It can be XATTR_ITEM, INODE_ITEM or just another EXTENT_DATA.
+	 * But if objectids mismatch, it means we have a missing
+	 * INODE_ITEM.
+	 */
+	if (!check_prev_ino(leaf, key, slot, prev_key))
+		return -EUCLEAN;
+
 	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
 
-	if (btrfs_file_extent_type(leaf, fi) > BTRFS_FILE_EXTENT_TYPES) {
+	/*
+	 * Make sure the item contains at least inline header, so the file
+	 * extent type is not some garbage.
+	 */
+	if (item_size < BTRFS_FILE_EXTENT_INLINE_DATA_START) {
+		file_extent_err(leaf, slot,
+				"invalid item size, have %u expect [%zu, %u)",
+				item_size, BTRFS_FILE_EXTENT_INLINE_DATA_START,
+				SZ_4K);
+		return -EUCLEAN;
+	}
+	if (btrfs_file_extent_type(leaf, fi) >= BTRFS_NR_FILE_EXTENT_TYPES) {
 		file_extent_err(leaf, slot,
 		"invalid type for file extent, have %u expect range [0, %u]",
 			btrfs_file_extent_type(leaf, fi),
-			BTRFS_FILE_EXTENT_TYPES);
+			BTRFS_NR_FILE_EXTENT_TYPES - 1);
 		return -EUCLEAN;
 	}
 
@@ -155,11 +244,11 @@ static int check_extent_data_item(struct extent_buffer *leaf,
 	 * Support for new compression/encryption must introduce incompat flag,
 	 * and must be caught in open_ctree().
 	 */
-	if (btrfs_file_extent_compression(leaf, fi) > BTRFS_COMPRESS_TYPES) {
+	if (btrfs_file_extent_compression(leaf, fi) >= BTRFS_NR_COMPRESS_TYPES) {
 		file_extent_err(leaf, slot,
 	"invalid compression for file extent, have %u expect range [0, %u]",
 			btrfs_file_extent_compression(leaf, fi),
-			BTRFS_COMPRESS_TYPES);
+			BTRFS_NR_COMPRESS_TYPES - 1);
 		return -EUCLEAN;
 	}
 	if (btrfs_file_extent_encryption(leaf, fi)) {
@@ -243,7 +332,7 @@ static int check_extent_data_item(struct extent_buffer *leaf,
 }
 
 static int check_csum_item(struct extent_buffer *leaf, struct btrfs_key *key,
-			   int slot)
+			   int slot, struct btrfs_key *prev_key)
 {
 	struct btrfs_fs_info *fs_info = leaf->fs_info;
 	u32 sectorsize = fs_info->sectorsize;
@@ -267,53 +356,142 @@ static int check_csum_item(struct extent_buffer *leaf, struct btrfs_key *key,
 			btrfs_item_size_nr(leaf, slot), csumsize);
 		return -EUCLEAN;
 	}
+	if (slot > 0 && prev_key->type == BTRFS_EXTENT_CSUM_KEY) {
+		u64 prev_csum_end;
+		u32 prev_item_size;
+
+		prev_item_size = btrfs_item_size_nr(leaf, slot - 1);
+		prev_csum_end = (prev_item_size / csumsize) * sectorsize;
+		prev_csum_end += prev_key->offset;
+		if (prev_csum_end > key->offset) {
+			generic_err(leaf, slot - 1,
+"csum end range (%llu) goes beyond the start range (%llu) of the next csum item",
+				    prev_csum_end, key->offset);
+			return -EUCLEAN;
+		}
+	}
 	return 0;
 }
 
-/*
- * Customized reported for dir_item, only important new info is key->objectid,
- * which represents inode number
- */
-__printf(3, 4)
-__cold
-static void dir_item_err(const struct extent_buffer *eb, int slot,
-			 const char *fmt, ...)
+/* Inode item error output has the same format as dir_item_err() */
+#define inode_item_err(eb, slot, fmt, ...)			\
+	dir_item_err(eb, slot, fmt, __VA_ARGS__)
+
+static int check_inode_key(struct extent_buffer *leaf, struct btrfs_key *key,
+			   int slot)
 {
-	const struct btrfs_fs_info *fs_info = eb->fs_info;
-	struct btrfs_key key;
-	struct va_format vaf;
-	va_list args;
+	struct btrfs_key item_key;
+	bool is_inode_item;
 
-	btrfs_item_key_to_cpu(eb, &key, slot);
-	va_start(args, fmt);
+	btrfs_item_key_to_cpu(leaf, &item_key, slot);
+	is_inode_item = (item_key.type == BTRFS_INODE_ITEM_KEY);
 
-	vaf.fmt = fmt;
-	vaf.va = &args;
+	/* For XATTR_ITEM, location key should be all 0 */
+	if (item_key.type == BTRFS_XATTR_ITEM_KEY) {
+		if (key->type != 0 || key->objectid != 0 || key->offset != 0)
+			return -EUCLEAN;
+		return 0;
+	}
 
-	btrfs_crit(fs_info,
-	"corrupt %s: root=%llu block=%llu slot=%d ino=%llu, %pV",
-		btrfs_header_level(eb) == 0 ? "leaf" : "node",
-		btrfs_header_owner(eb), btrfs_header_bytenr(eb), slot,
-		key.objectid, &vaf);
-	va_end(args);
+	if ((key->objectid < BTRFS_FIRST_FREE_OBJECTID ||
+	     key->objectid > BTRFS_LAST_FREE_OBJECTID) &&
+	    key->objectid != BTRFS_ROOT_TREE_DIR_OBJECTID &&
+	    key->objectid != BTRFS_FREE_INO_OBJECTID) {
+		if (is_inode_item) {
+			generic_err(leaf, slot,
+	"invalid key objectid: has %llu expect %llu or [%llu, %llu] or %llu",
+				key->objectid, BTRFS_ROOT_TREE_DIR_OBJECTID,
+				BTRFS_FIRST_FREE_OBJECTID,
+				BTRFS_LAST_FREE_OBJECTID,
+				BTRFS_FREE_INO_OBJECTID);
+		} else {
+			dir_item_err(leaf, slot,
+"invalid location key objectid: has %llu expect %llu or [%llu, %llu] or %llu",
+				key->objectid, BTRFS_ROOT_TREE_DIR_OBJECTID,
+				BTRFS_FIRST_FREE_OBJECTID,
+				BTRFS_LAST_FREE_OBJECTID,
+				BTRFS_FREE_INO_OBJECTID);
+		}
+		return -EUCLEAN;
+	}
+	if (key->offset != 0) {
+		if (is_inode_item)
+			inode_item_err(leaf, slot,
+				       "invalid key offset: has %llu expect 0",
+				       key->offset);
+		else
+			dir_item_err(leaf, slot,
+				"invalid location key offset:has %llu expect 0",
+				key->offset);
+		return -EUCLEAN;
+	}
+	return 0;
+}
+
+static int check_root_key(struct extent_buffer *leaf, struct btrfs_key *key,
+			  int slot)
+{
+	struct btrfs_key item_key;
+	bool is_root_item;
+
+	btrfs_item_key_to_cpu(leaf, &item_key, slot);
+	is_root_item = (item_key.type == BTRFS_ROOT_ITEM_KEY);
+
+	/* No such tree id */
+	if (key->objectid == 0) {
+		if (is_root_item)
+			generic_err(leaf, slot, "invalid root id 0");
+		else
+			dir_item_err(leaf, slot,
+				     "invalid location key root id 0");
+		return -EUCLEAN;
+	}
+
+	/* DIR_ITEM/INDEX/INODE_REF is not allowed to point to non-fs trees */
+	if (!is_fstree(key->objectid) && !is_root_item) {
+		dir_item_err(leaf, slot,
+		"invalid location key objectid, have %llu expect [%llu, %llu]",
+				key->objectid, BTRFS_FIRST_FREE_OBJECTID,
+				BTRFS_LAST_FREE_OBJECTID);
+		return -EUCLEAN;
+	}
+
+	/*
+	 * ROOT_ITEM with non-zero offset means this is a snapshot, created at
+	 * @offset transid.
+	 * Furthermore, for location key in DIR_ITEM, its offset is always -1.
+	 *
+	 * So here we only check offset for reloc tree whose key->offset must
+	 * be a valid tree.
+	 */
+	if (key->objectid == BTRFS_TREE_RELOC_OBJECTID && key->offset == 0) {
+		generic_err(leaf, slot, "invalid root id 0 for reloc tree");
+		return -EUCLEAN;
+	}
+	return 0;
 }
 
 static int check_dir_item(struct extent_buffer *leaf,
-			  struct btrfs_key *key, int slot)
+			  struct btrfs_key *key, struct btrfs_key *prev_key,
+			  int slot)
 {
 	struct btrfs_fs_info *fs_info = leaf->fs_info;
 	struct btrfs_dir_item *di;
 	u32 item_size = btrfs_item_size_nr(leaf, slot);
 	u32 cur = 0;
 
+	if (!check_prev_ino(leaf, key, slot, prev_key))
+		return -EUCLEAN;
 	di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
 	while (cur < item_size) {
+		struct btrfs_key location_key;
 		u32 name_len;
 		u32 data_len;
 		u32 max_name_len;
 		u32 total_size;
 		u32 name_hash;
 		u8 dir_type;
+		int ret;
 
 		/* header itself should not cross item boundary */
 		if (cur + sizeof(*di) > item_size) {
@@ -323,6 +501,25 @@ static int check_dir_item(struct extent_buffer *leaf,
 			return -EUCLEAN;
 		}
 
+		/* Location key check */
+		btrfs_dir_item_key_to_cpu(leaf, di, &location_key);
+		if (location_key.type == BTRFS_ROOT_ITEM_KEY) {
+			ret = check_root_key(leaf, &location_key, slot);
+			if (ret < 0)
+				return ret;
+		} else if (location_key.type == BTRFS_INODE_ITEM_KEY ||
+			   location_key.type == 0) {
+			ret = check_inode_key(leaf, &location_key, slot);
+			if (ret < 0)
+				return ret;
+		} else {
+			dir_item_err(leaf, slot,
+			"invalid location key type, have %u, expect %u or %u",
+				     location_key.type, BTRFS_ROOT_ITEM_KEY,
+				     BTRFS_INODE_ITEM_KEY);
+			return -EUCLEAN;
+		}
+
 		/* dir type check */
 		dir_type = btrfs_dir_type(leaf, di);
 		if (dir_type >= BTRFS_FT_MAX) {
@@ -459,23 +656,23 @@ static int check_block_group_item(struct extent_buffer *leaf,
 
 	read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot),
 			   sizeof(bgi));
-	if (btrfs_block_group_chunk_objectid(&bgi) !=
+	if (btrfs_stack_block_group_chunk_objectid(&bgi) !=
 	    BTRFS_FIRST_CHUNK_TREE_OBJECTID) {
 		block_group_err(leaf, slot,
 		"invalid block group chunk objectid, have %llu expect %llu",
-				btrfs_block_group_chunk_objectid(&bgi),
+				btrfs_stack_block_group_chunk_objectid(&bgi),
 				BTRFS_FIRST_CHUNK_TREE_OBJECTID);
 		return -EUCLEAN;
 	}
 
-	if (btrfs_block_group_used(&bgi) > key->offset) {
+	if (btrfs_stack_block_group_used(&bgi) > key->offset) {
 		block_group_err(leaf, slot,
 			"invalid block group used, have %llu expect [0, %llu)",
-				btrfs_block_group_used(&bgi), key->offset);
+				btrfs_stack_block_group_used(&bgi), key->offset);
 		return -EUCLEAN;
 	}
 
-	flags = btrfs_block_group_flags(&bgi);
+	flags = btrfs_stack_block_group_flags(&bgi);
 	if (hweight64(flags & BTRFS_BLOCK_GROUP_PROFILE_MASK) > 1) {
 		block_group_err(leaf, slot,
 "invalid profile flags, have 0x%llx (%lu bits set) expect no more than 1 bit set",
@@ -609,7 +806,7 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf,
 		return -EUCLEAN;
 	}
 
-	if (!is_power_of_2(type & BTRFS_BLOCK_GROUP_PROFILE_MASK) &&
+	if (!has_single_bit_set(type & BTRFS_BLOCK_GROUP_PROFILE_MASK) &&
 	    (type & BTRFS_BLOCK_GROUP_PROFILE_MASK) != 0) {
 		chunk_err(leaf, chunk, logical,
 		"invalid chunk profile flag: 0x%llx, expect 0 or 1 bit set",
@@ -660,6 +857,44 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf,
 	return 0;
 }
 
+/*
+ * Enhanced version of chunk item checker.
+ *
+ * The common btrfs_check_chunk_valid() doesn't check item size since it needs
+ * to work on super block sys_chunk_array which doesn't have full item ptr.
+ */
+static int check_leaf_chunk_item(struct extent_buffer *leaf,
+				 struct btrfs_chunk *chunk,
+				 struct btrfs_key *key, int slot)
+{
+	int num_stripes;
+
+	if (btrfs_item_size_nr(leaf, slot) < sizeof(struct btrfs_chunk)) {
+		chunk_err(leaf, chunk, key->offset,
+			"invalid chunk item size: have %u expect [%zu, %u)",
+			btrfs_item_size_nr(leaf, slot),
+			sizeof(struct btrfs_chunk),
+			BTRFS_LEAF_DATA_SIZE(leaf->fs_info));
+		return -EUCLEAN;
+	}
+
+	num_stripes = btrfs_chunk_num_stripes(leaf, chunk);
+	/* Let btrfs_check_chunk_valid() handle this error type */
+	if (num_stripes == 0)
+		goto out;
+
+	if (btrfs_chunk_item_size(num_stripes) !=
+	    btrfs_item_size_nr(leaf, slot)) {
+		chunk_err(leaf, chunk, key->offset,
+			"invalid chunk item size: have %u expect %lu",
+			btrfs_item_size_nr(leaf, slot),
+			btrfs_chunk_item_size(num_stripes));
+		return -EUCLEAN;
+	}
+out:
+	return btrfs_check_chunk_valid(leaf, chunk, key->offset);
+}
+
 __printf(3, 4)
 __cold
 static void dev_item_err(const struct extent_buffer *eb, int slot,
@@ -723,7 +958,7 @@ static int check_dev_item(struct extent_buffer *leaf,
 }
 
 /* Inode item error output has the same format as dir_item_err() */
-#define inode_item_err(fs_info, eb, slot, fmt, ...)			\
+#define inode_item_err(eb, slot, fmt, ...)			\
 	dir_item_err(eb, slot, fmt, __VA_ARGS__)
 
 static int check_inode_item(struct extent_buffer *leaf,
@@ -734,30 +969,17 @@ static int check_inode_item(struct extent_buffer *leaf,
 	u64 super_gen = btrfs_super_generation(fs_info->super_copy);
 	u32 valid_mask = (S_IFMT | S_ISUID | S_ISGID | S_ISVTX | 0777);
 	u32 mode;
+	int ret;
+
+	ret = check_inode_key(leaf, key, slot);
+	if (ret < 0)
+		return ret;
 
-	if ((key->objectid < BTRFS_FIRST_FREE_OBJECTID ||
-	     key->objectid > BTRFS_LAST_FREE_OBJECTID) &&
-	    key->objectid != BTRFS_ROOT_TREE_DIR_OBJECTID &&
-	    key->objectid != BTRFS_FREE_INO_OBJECTID) {
-		generic_err(leaf, slot,
-	"invalid key objectid: has %llu expect %llu or [%llu, %llu] or %llu",
-			    key->objectid, BTRFS_ROOT_TREE_DIR_OBJECTID,
-			    BTRFS_FIRST_FREE_OBJECTID,
-			    BTRFS_LAST_FREE_OBJECTID,
-			    BTRFS_FREE_INO_OBJECTID);
-		return -EUCLEAN;
-	}
-	if (key->offset != 0) {
-		inode_item_err(fs_info, leaf, slot,
-			"invalid key offset: has %llu expect 0",
-			key->offset);
-		return -EUCLEAN;
-	}
 	iitem = btrfs_item_ptr(leaf, slot, struct btrfs_inode_item);
 
 	/* Here we use super block generation + 1 to handle log tree */
 	if (btrfs_inode_generation(leaf, iitem) > super_gen + 1) {
-		inode_item_err(fs_info, leaf, slot,
+		inode_item_err(leaf, slot,
 			"invalid inode generation: has %llu expect (0, %llu]",
 			       btrfs_inode_generation(leaf, iitem),
 			       super_gen + 1);
@@ -765,7 +987,7 @@ static int check_inode_item(struct extent_buffer *leaf,
 	}
 	/* Note for ROOT_TREE_DIR_ITEM, mkfs could set its transid 0 */
 	if (btrfs_inode_transid(leaf, iitem) > super_gen + 1) {
-		inode_item_err(fs_info, leaf, slot,
+		inode_item_err(leaf, slot,
 			"invalid inode generation: has %llu expect [0, %llu]",
 			       btrfs_inode_transid(leaf, iitem), super_gen + 1);
 		return -EUCLEAN;
@@ -778,33 +1000,33 @@ static int check_inode_item(struct extent_buffer *leaf,
 	 */
 	mode = btrfs_inode_mode(leaf, iitem);
 	if (mode & ~valid_mask) {
-		inode_item_err(fs_info, leaf, slot,
+		inode_item_err(leaf, slot,
 			       "unknown mode bit detected: 0x%x",
 			       mode & ~valid_mask);
 		return -EUCLEAN;
 	}
 
 	/*
-	 * S_IFMT is not bit mapped so we can't completely rely on is_power_of_2,
-	 * but is_power_of_2() can save us from checking FIFO/CHR/DIR/REG.
-	 * Only needs to check BLK, LNK and SOCKS
+	 * S_IFMT is not bit mapped so we can't completely rely on
+	 * is_power_of_2/has_single_bit_set, but it can save us from checking
+	 * FIFO/CHR/DIR/REG.  Only needs to check BLK, LNK and SOCKS
 	 */
-	if (!is_power_of_2(mode & S_IFMT)) {
+	if (!has_single_bit_set(mode & S_IFMT)) {
 		if (!S_ISLNK(mode) && !S_ISBLK(mode) && !S_ISSOCK(mode)) {
-			inode_item_err(fs_info, leaf, slot,
+			inode_item_err(leaf, slot,
 			"invalid mode: has 0%o expect valid S_IF* bit(s)",
 				       mode & S_IFMT);
 			return -EUCLEAN;
 		}
 	}
 	if (S_ISDIR(mode) && btrfs_inode_nlink(leaf, iitem) > 1) {
-		inode_item_err(fs_info, leaf, slot,
+		inode_item_err(leaf, slot,
 		       "invalid nlink: has %u expect no more than 1 for dir",
 			btrfs_inode_nlink(leaf, iitem));
 		return -EUCLEAN;
 	}
 	if (btrfs_inode_flags(leaf, iitem) & ~BTRFS_INODE_FLAG_MASK) {
-		inode_item_err(fs_info, leaf, slot,
+		inode_item_err(leaf, slot,
 			       "unknown flags detected: 0x%llx",
 			       btrfs_inode_flags(leaf, iitem) &
 			       ~BTRFS_INODE_FLAG_MASK);
@@ -820,22 +1042,11 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key,
 	struct btrfs_root_item ri;
 	const u64 valid_root_flags = BTRFS_ROOT_SUBVOL_RDONLY |
 				     BTRFS_ROOT_SUBVOL_DEAD;
+	int ret;
 
-	/* No such tree id */
-	if (key->objectid == 0) {
-		generic_err(leaf, slot, "invalid root id 0");
-		return -EUCLEAN;
-	}
-
-	/*
-	 * Some older kernel may create ROOT_ITEM with non-zero offset, so here
-	 * we only check offset for reloc tree whose key->offset must be a
-	 * valid tree.
-	 */
-	if (key->objectid == BTRFS_TREE_RELOC_OBJECTID && key->offset == 0) {
-		generic_err(leaf, slot, "invalid root id 0 for reloc tree");
-		return -EUCLEAN;
-	}
+	ret = check_root_key(leaf, key, slot);
+	if (ret < 0)
+		return ret;
 
 	if (btrfs_item_size_nr(leaf, slot) != sizeof(ri)) {
 		generic_err(leaf, slot,
@@ -1010,8 +1221,8 @@ static int check_extent_item(struct extent_buffer *leaf,
 			   btrfs_super_generation(fs_info->super_copy) + 1);
 		return -EUCLEAN;
 	}
-	if (!is_power_of_2(flags & (BTRFS_EXTENT_FLAG_DATA |
-				    BTRFS_EXTENT_FLAG_TREE_BLOCK))) {
+	if (!has_single_bit_set(flags & (BTRFS_EXTENT_FLAG_DATA |
+					 BTRFS_EXTENT_FLAG_TREE_BLOCK))) {
 		extent_err(leaf, slot,
 		"invalid extent flag, have 0x%llx expect 1 bit set in 0x%llx",
 			flags, BTRFS_EXTENT_FLAG_DATA |
@@ -1224,6 +1435,58 @@ static int check_extent_data_ref(struct extent_buffer *leaf,
 	return 0;
 }
 
+#define inode_ref_err(eb, slot, fmt, args...)			\
+	inode_item_err(eb, slot, fmt, ##args)
+static int check_inode_ref(struct extent_buffer *leaf,
+			   struct btrfs_key *key, struct btrfs_key *prev_key,
+			   int slot)
+{
+	struct btrfs_inode_ref *iref;
+	unsigned long ptr;
+	unsigned long end;
+
+	if (!check_prev_ino(leaf, key, slot, prev_key))
+		return -EUCLEAN;
+	/* namelen can't be 0, so item_size == sizeof() is also invalid */
+	if (btrfs_item_size_nr(leaf, slot) <= sizeof(*iref)) {
+		inode_ref_err(leaf, slot,
+			"invalid item size, have %u expect (%zu, %u)",
+			btrfs_item_size_nr(leaf, slot),
+			sizeof(*iref), BTRFS_LEAF_DATA_SIZE(leaf->fs_info));
+		return -EUCLEAN;
+	}
+
+	ptr = btrfs_item_ptr_offset(leaf, slot);
+	end = ptr + btrfs_item_size_nr(leaf, slot);
+	while (ptr < end) {
+		u16 namelen;
+
+		if (ptr + sizeof(iref) > end) {
+			inode_ref_err(leaf, slot,
+			"inode ref overflow, ptr %lu end %lu inode_ref_size %zu",
+				ptr, end, sizeof(iref));
+			return -EUCLEAN;
+		}
+
+		iref = (struct btrfs_inode_ref *)ptr;
+		namelen = btrfs_inode_ref_name_len(leaf, iref);
+		if (ptr + sizeof(*iref) + namelen > end) {
+			inode_ref_err(leaf, slot,
+				"inode ref overflow, ptr %lu end %lu namelen %u",
+				ptr, end, namelen);
+			return -EUCLEAN;
+		}
+
+		/*
+		 * NOTE: In theory we should record all found index numbers
+		 * to find any duplicated indexes, but that will be too time
+		 * consuming for inodes with too many hard links.
+		 */
+		ptr += sizeof(*iref) + namelen;
+	}
+	return 0;
+}
+
 /*
  * Common point to switch the item-specific validation.
  */
@@ -1239,19 +1502,22 @@ static int check_leaf_item(struct extent_buffer *leaf,
 		ret = check_extent_data_item(leaf, key, slot, prev_key);
 		break;
 	case BTRFS_EXTENT_CSUM_KEY:
-		ret = check_csum_item(leaf, key, slot);
+		ret = check_csum_item(leaf, key, slot, prev_key);
 		break;
 	case BTRFS_DIR_ITEM_KEY:
 	case BTRFS_DIR_INDEX_KEY:
 	case BTRFS_XATTR_ITEM_KEY:
-		ret = check_dir_item(leaf, key, slot);
+		ret = check_dir_item(leaf, key, prev_key, slot);
+		break;
+	case BTRFS_INODE_REF_KEY:
+		ret = check_inode_ref(leaf, key, prev_key, slot);
 		break;
 	case BTRFS_BLOCK_GROUP_ITEM_KEY:
 		ret = check_block_group_item(leaf, key, slot);
 		break;
 	case BTRFS_CHUNK_ITEM_KEY:
 		chunk = btrfs_item_ptr(leaf, slot, struct btrfs_chunk);
-		ret = btrfs_check_chunk_valid(leaf, chunk, key->offset);
+		ret = check_leaf_chunk_item(leaf, chunk, key, slot);
 		break;
 	case BTRFS_DEV_ITEM_KEY:
 		ret = check_dev_item(leaf, key, slot);
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 8a6cc600bf18..02ebdd9edc19 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -18,6 +18,8 @@
 #include "compression.h"
 #include "qgroup.h"
 #include "inode-map.h"
+#include "block-group.h"
+#include "space-info.h"
 
 /* magic values for the inode_only field in btrfs_log_inode:
  *
@@ -311,7 +313,7 @@ static int process_one_buffer(struct btrfs_root *log,
 	}
 
 	if (wc->pin)
-		ret = btrfs_pin_extent_for_log_replay(fs_info, eb->start,
+		ret = btrfs_pin_extent_for_log_replay(wc->trans, eb->start,
 						      eb->len);
 
 	if (!ret && btrfs_buffer_uptodate(eb, gen, 0)) {
@@ -559,7 +561,7 @@ static noinline struct inode *read_one_inode(struct btrfs_root *root,
 	key.objectid = objectid;
 	key.type = BTRFS_INODE_ITEM_KEY;
 	key.offset = 0;
-	inode = btrfs_iget(root->fs_info->sb, &key, root, NULL);
+	inode = btrfs_iget(root->fs_info->sb, &key, root);
 	if (IS_ERR(inode))
 		inode = NULL;
 	return inode;
@@ -808,7 +810,8 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 						struct btrfs_ordered_sum,
 						list);
 				if (!ret)
-					ret = btrfs_del_csums(trans, fs_info,
+					ret = btrfs_del_csums(trans,
+							      fs_info->csum_root,
 							      sums->bytenr,
 							      sums->len);
 				if (!ret)
@@ -829,6 +832,11 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 			goto out;
 	}
 
+	ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode), start,
+						extent_end - start);
+	if (ret)
+		goto out;
+
 	inode_add_bytes(inode, nbytes);
 update_inode:
 	ret = btrfs_update_inode(trans, root, inode);
@@ -945,54 +953,32 @@ static noinline int backref_in_log(struct btrfs_root *log,
 				   const char *name, int namelen)
 {
 	struct btrfs_path *path;
-	struct btrfs_inode_ref *ref;
-	unsigned long ptr;
-	unsigned long ptr_end;
-	unsigned long name_ptr;
-	int found_name_len;
-	int item_size;
 	int ret;
-	int match = 0;
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
 
 	ret = btrfs_search_slot(NULL, log, key, path, 0, 0);
-	if (ret != 0)
+	if (ret < 0) {
 		goto out;
-
-	ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]);
-
-	if (key->type == BTRFS_INODE_EXTREF_KEY) {
-		if (btrfs_find_name_in_ext_backref(path->nodes[0],
-						   path->slots[0],
-						   ref_objectid,
-						   name, namelen))
-			match = 1;
-
+	} else if (ret == 1) {
+		ret = 0;
 		goto out;
 	}
 
-	item_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]);
-	ptr_end = ptr + item_size;
-	while (ptr < ptr_end) {
-		ref = (struct btrfs_inode_ref *)ptr;
-		found_name_len = btrfs_inode_ref_name_len(path->nodes[0], ref);
-		if (found_name_len == namelen) {
-			name_ptr = (unsigned long)(ref + 1);
-			ret = memcmp_extent_buffer(path->nodes[0], name,
-						   name_ptr, namelen);
-			if (ret == 0) {
-				match = 1;
-				goto out;
-			}
-		}
-		ptr = (unsigned long)(ref + 1) + found_name_len;
-	}
+	if (key->type == BTRFS_INODE_EXTREF_KEY)
+		ret = !!btrfs_find_name_in_ext_backref(path->nodes[0],
+						       path->slots[0],
+						       ref_objectid,
+						       name, namelen);
+	else
+		ret = !!btrfs_find_name_in_backref(path->nodes[0],
+						   path->slots[0],
+						   name, namelen);
 out:
 	btrfs_free_path(path);
-	return match;
+	return ret;
 }
 
 static inline int __add_inode_ref(struct btrfs_trans_handle *trans,
@@ -1050,10 +1036,13 @@ again:
 					   (unsigned long)(victim_ref + 1),
 					   victim_name_len);
 
-			if (!backref_in_log(log_root, &search_key,
-					    parent_objectid,
-					    victim_name,
-					    victim_name_len)) {
+			ret = backref_in_log(log_root, &search_key,
+					     parent_objectid, victim_name,
+					     victim_name_len);
+			if (ret < 0) {
+				kfree(victim_name);
+				return ret;
+			} else if (!ret) {
 				inc_nlink(&inode->vfs_inode);
 				btrfs_release_path(path);
 
@@ -1115,10 +1104,12 @@ again:
 			search_key.offset = btrfs_extref_hash(parent_objectid,
 							      victim_name,
 							      victim_name_len);
-			ret = 0;
-			if (!backref_in_log(log_root, &search_key,
-					    parent_objectid, victim_name,
-					    victim_name_len)) {
+			ret = backref_in_log(log_root, &search_key,
+					     parent_objectid, victim_name,
+					     victim_name_len);
+			if (ret < 0) {
+				return ret;
+			} else if (!ret) {
 				ret = -ENOENT;
 				victim_parent = read_one_inode(root,
 						parent_objectid);
@@ -1885,30 +1876,6 @@ static noinline int insert_one_name(struct btrfs_trans_handle *trans,
 }
 
 /*
- * Return true if an inode reference exists in the log for the given name,
- * inode and parent inode.
- */
-static bool name_in_log_ref(struct btrfs_root *log_root,
-			    const char *name, const int name_len,
-			    const u64 dirid, const u64 ino)
-{
-	struct btrfs_key search_key;
-
-	search_key.objectid = ino;
-	search_key.type = BTRFS_INODE_REF_KEY;
-	search_key.offset = dirid;
-	if (backref_in_log(log_root, &search_key, dirid, name, name_len))
-		return true;
-
-	search_key.type = BTRFS_INODE_EXTREF_KEY;
-	search_key.offset = btrfs_extref_hash(dirid, name, name_len);
-	if (backref_in_log(log_root, &search_key, dirid, name, name_len))
-		return true;
-
-	return false;
-}
-
-/*
  * take a single entry in a log directory item and replay it into
  * the subvolume.
  *
@@ -2024,8 +1991,31 @@ out:
 	return ret;
 
 insert:
-	if (name_in_log_ref(root->log_root, name, name_len,
-			    key->objectid, log_key.objectid)) {
+	/*
+	 * Check if the inode reference exists in the log for the given name,
+	 * inode and parent inode
+	 */
+	found_key.objectid = log_key.objectid;
+	found_key.type = BTRFS_INODE_REF_KEY;
+	found_key.offset = key->objectid;
+	ret = backref_in_log(root->log_root, &found_key, 0, name, name_len);
+	if (ret < 0) {
+	        goto out;
+	} else if (ret) {
+	        /* The dentry will be added later. */
+	        ret = 0;
+	        update_size = false;
+	        goto out;
+	}
+
+	found_key.objectid = log_key.objectid;
+	found_key.type = BTRFS_INODE_EXTREF_KEY;
+	found_key.offset = key->objectid;
+	ret = backref_in_log(root->log_root, &found_key, key->objectid, name,
+			     name_len);
+	if (ret < 0) {
+		goto out;
+	} else if (ret) {
 		/* The dentry will be added later. */
 		ret = 0;
 		update_size = false;
@@ -2676,29 +2666,45 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 	return ret;
 }
 
+/*
+ * Correctly adjust the reserved bytes occupied by a log tree extent buffer
+ */
+static void unaccount_log_buffer(struct btrfs_fs_info *fs_info, u64 start)
+{
+	struct btrfs_block_group *cache;
+
+	cache = btrfs_lookup_block_group(fs_info, start);
+	if (!cache) {
+		btrfs_err(fs_info, "unable to find block group for %llu", start);
+		return;
+	}
+
+	spin_lock(&cache->space_info->lock);
+	spin_lock(&cache->lock);
+	cache->reserved -= fs_info->nodesize;
+	cache->space_info->bytes_reserved -= fs_info->nodesize;
+	spin_unlock(&cache->lock);
+	spin_unlock(&cache->space_info->lock);
+
+	btrfs_put_block_group(cache);
+}
+
 static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 				   struct btrfs_root *root,
 				   struct btrfs_path *path, int *level,
 				   struct walk_control *wc)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
-	u64 root_owner;
 	u64 bytenr;
 	u64 ptr_gen;
 	struct extent_buffer *next;
 	struct extent_buffer *cur;
-	struct extent_buffer *parent;
 	u32 blocksize;
 	int ret = 0;
 
-	WARN_ON(*level < 0);
-	WARN_ON(*level >= BTRFS_MAX_LEVEL);
-
 	while (*level > 0) {
 		struct btrfs_key first_key;
 
-		WARN_ON(*level < 0);
-		WARN_ON(*level >= BTRFS_MAX_LEVEL);
 		cur = path->nodes[*level];
 
 		WARN_ON(btrfs_header_level(cur) != *level);
@@ -2712,9 +2718,6 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 		btrfs_node_key_to_cpu(cur, &first_key, path->slots[*level]);
 		blocksize = fs_info->nodesize;
 
-		parent = path->nodes[*level];
-		root_owner = btrfs_header_owner(parent);
-
 		next = btrfs_find_create_tree_block(fs_info, bytenr);
 		if (IS_ERR(next))
 			return PTR_ERR(next);
@@ -2742,19 +2745,16 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 					btrfs_clean_tree_block(next);
 					btrfs_wait_tree_block_writeback(next);
 					btrfs_tree_unlock(next);
+					ret = btrfs_pin_reserved_extent(trans,
+							bytenr, blocksize);
+					if (ret) {
+						free_extent_buffer(next);
+						return ret;
+					}
 				} else {
 					if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags))
 						clear_extent_buffer_dirty(next);
-				}
-
-				WARN_ON(root_owner !=
-					BTRFS_TREE_LOG_OBJECTID);
-				ret = btrfs_free_and_pin_reserved_extent(
-							fs_info, bytenr,
-							blocksize);
-				if (ret) {
-					free_extent_buffer(next);
-					return ret;
+					unaccount_log_buffer(fs_info, bytenr);
 				}
 			}
 			free_extent_buffer(next);
@@ -2766,7 +2766,6 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 			return ret;
 		}
 
-		WARN_ON(*level <= 0);
 		if (path->nodes[*level-1])
 			free_extent_buffer(path->nodes[*level-1]);
 		path->nodes[*level-1] = next;
@@ -2774,9 +2773,6 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 		path->slots[*level] = 0;
 		cond_resched();
 	}
-	WARN_ON(*level < 0);
-	WARN_ON(*level >= BTRFS_MAX_LEVEL);
-
 	path->slots[*level] = btrfs_header_nritems(path->nodes[*level]);
 
 	cond_resched();
@@ -2789,7 +2785,6 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans,
 				 struct walk_control *wc)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
-	u64 root_owner;
 	int i;
 	int slot;
 	int ret;
@@ -2802,13 +2797,6 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans,
 			WARN_ON(*level == 0);
 			return 0;
 		} else {
-			struct extent_buffer *parent;
-			if (path->nodes[*level] == root->node)
-				parent = path->nodes[*level];
-			else
-				parent = path->nodes[*level + 1];
-
-			root_owner = btrfs_header_owner(parent);
 			ret = wc->process_func(root, path->nodes[*level], wc,
 				 btrfs_header_generation(path->nodes[*level]),
 				 *level);
@@ -2826,18 +2814,18 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans,
 					btrfs_clean_tree_block(next);
 					btrfs_wait_tree_block_writeback(next);
 					btrfs_tree_unlock(next);
+					ret = btrfs_pin_reserved_extent(trans,
+						     path->nodes[*level]->start,
+						     path->nodes[*level]->len);
+					if (ret)
+						return ret;
 				} else {
 					if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags))
 						clear_extent_buffer_dirty(next);
-				}
 
-				WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID);
-				ret = btrfs_free_and_pin_reserved_extent(
-						fs_info,
-						path->nodes[*level]->start,
-						path->nodes[*level]->len);
-				if (ret)
-					return ret;
+					unaccount_log_buffer(fs_info,
+						path->nodes[*level]->start);
+				}
 			}
 			free_extent_buffer(path->nodes[*level]);
 			path->nodes[*level] = NULL;
@@ -2869,7 +2857,7 @@ static int walk_log_tree(struct btrfs_trans_handle *trans,
 	level = btrfs_header_level(log->node);
 	orig_level = level;
 	path->nodes[level] = log->node;
-	extent_buffer_get(log->node);
+	atomic_inc(&log->node->refs);
 	path->slots[level] = 0;
 
 	while (1) {
@@ -2908,17 +2896,15 @@ static int walk_log_tree(struct btrfs_trans_handle *trans,
 				btrfs_clean_tree_block(next);
 				btrfs_wait_tree_block_writeback(next);
 				btrfs_tree_unlock(next);
+				ret = btrfs_pin_reserved_extent(trans,
+						next->start, next->len);
+				if (ret)
+					goto out;
 			} else {
 				if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags))
 					clear_extent_buffer_dirty(next);
+				unaccount_log_buffer(fs_info, next->start);
 			}
-
-			WARN_ON(log->root_key.objectid !=
-				BTRFS_TREE_LOG_OBJECTID);
-			ret = btrfs_free_and_pin_reserved_extent(fs_info,
-							next->start, next->len);
-			if (ret)
-				goto out;
 		}
 	}
 
@@ -3313,8 +3299,7 @@ static void free_log_tree(struct btrfs_trans_handle *trans,
 
 	clear_extent_bits(&log->dirty_log_pages, 0, (u64)-1,
 			  EXTENT_DIRTY | EXTENT_NEW | EXTENT_NEED_WAIT);
-	free_extent_buffer(log->node);
-	kfree(log);
+	btrfs_put_root(log);
 }
 
 /*
@@ -3927,10 +3912,32 @@ static int log_inode_item(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
+static int log_csums(struct btrfs_trans_handle *trans,
+		     struct btrfs_root *log_root,
+		     struct btrfs_ordered_sum *sums)
+{
+	int ret;
+
+	/*
+	 * Due to extent cloning, we might have logged a csum item that covers a
+	 * subrange of a cloned extent, and later we can end up logging a csum
+	 * item for a larger subrange of the same extent or the entire range.
+	 * This would leave csum items in the log tree that cover the same range
+	 * and break the searches for checksums in the log tree, resulting in
+	 * some checksums missing in the fs/subvolume tree. So just delete (or
+	 * trim and adjust) any existing csum items in the log for this range.
+	 */
+	ret = btrfs_del_csums(trans, log_root, sums->bytenr, sums->len);
+	if (ret)
+		return ret;
+
+	return btrfs_csum_file_blocks(trans, log_root, sums);
+}
+
 static noinline int copy_items(struct btrfs_trans_handle *trans,
 			       struct btrfs_inode *inode,
 			       struct btrfs_path *dst_path,
-			       struct btrfs_path *src_path, u64 *last_extent,
+			       struct btrfs_path *src_path,
 			       int start_slot, int nr, int inode_only,
 			       u64 logged_isize)
 {
@@ -3941,7 +3948,6 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 	struct btrfs_file_extent_item *extent;
 	struct btrfs_inode_item *inode_item;
 	struct extent_buffer *src = src_path->nodes[0];
-	struct btrfs_key first_key, last_key, key;
 	int ret;
 	struct btrfs_key *ins_keys;
 	u32 *ins_sizes;
@@ -3949,9 +3955,6 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 	int i;
 	struct list_head ordered_sums;
 	int skip_csum = inode->flags & BTRFS_INODE_NODATASUM;
-	bool has_extents = false;
-	bool need_find_last_extent = true;
-	bool done = false;
 
 	INIT_LIST_HEAD(&ordered_sums);
 
@@ -3960,8 +3963,6 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 	if (!ins_data)
 		return -ENOMEM;
 
-	first_key.objectid = (u64)-1;
-
 	ins_sizes = (u32 *)ins_data;
 	ins_keys = (struct btrfs_key *)(ins_data + nr * sizeof(u32));
 
@@ -3982,9 +3983,6 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 
 		src_offset = btrfs_item_ptr_offset(src, start_slot + i);
 
-		if (i == nr - 1)
-			last_key = ins_keys[i];
-
 		if (ins_keys[i].type == BTRFS_INODE_ITEM_KEY) {
 			inode_item = btrfs_item_ptr(dst_path->nodes[0],
 						    dst_path->slots[0],
@@ -3998,20 +3996,6 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 					   src_offset, ins_sizes[i]);
 		}
 
-		/*
-		 * We set need_find_last_extent here in case we know we were
-		 * processing other items and then walk into the first extent in
-		 * the inode.  If we don't hit an extent then nothing changes,
-		 * we'll do the last search the next time around.
-		 */
-		if (ins_keys[i].type == BTRFS_EXTENT_DATA_KEY) {
-			has_extents = true;
-			if (first_key.objectid == (u64)-1)
-				first_key = ins_keys[i];
-		} else {
-			need_find_last_extent = false;
-		}
-
 		/* take a reference on file data extents so that truncates
 		 * or deletes of this inode don't have to relog the inode
 		 * again
@@ -4072,172 +4056,11 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 						   struct btrfs_ordered_sum,
 						   list);
 		if (!ret)
-			ret = btrfs_csum_file_blocks(trans, log, sums);
+			ret = log_csums(trans, log, sums);
 		list_del(&sums->list);
 		kfree(sums);
 	}
 
-	if (!has_extents)
-		return ret;
-
-	if (need_find_last_extent && *last_extent == first_key.offset) {
-		/*
-		 * We don't have any leafs between our current one and the one
-		 * we processed before that can have file extent items for our
-		 * inode (and have a generation number smaller than our current
-		 * transaction id).
-		 */
-		need_find_last_extent = false;
-	}
-
-	/*
-	 * Because we use btrfs_search_forward we could skip leaves that were
-	 * not modified and then assume *last_extent is valid when it really
-	 * isn't.  So back up to the previous leaf and read the end of the last
-	 * extent before we go and fill in holes.
-	 */
-	if (need_find_last_extent) {
-		u64 len;
-
-		ret = btrfs_prev_leaf(inode->root, src_path);
-		if (ret < 0)
-			return ret;
-		if (ret)
-			goto fill_holes;
-		if (src_path->slots[0])
-			src_path->slots[0]--;
-		src = src_path->nodes[0];
-		btrfs_item_key_to_cpu(src, &key, src_path->slots[0]);
-		if (key.objectid != btrfs_ino(inode) ||
-		    key.type != BTRFS_EXTENT_DATA_KEY)
-			goto fill_holes;
-		extent = btrfs_item_ptr(src, src_path->slots[0],
-					struct btrfs_file_extent_item);
-		if (btrfs_file_extent_type(src, extent) ==
-		    BTRFS_FILE_EXTENT_INLINE) {
-			len = btrfs_file_extent_ram_bytes(src, extent);
-			*last_extent = ALIGN(key.offset + len,
-					     fs_info->sectorsize);
-		} else {
-			len = btrfs_file_extent_num_bytes(src, extent);
-			*last_extent = key.offset + len;
-		}
-	}
-fill_holes:
-	/* So we did prev_leaf, now we need to move to the next leaf, but a few
-	 * things could have happened
-	 *
-	 * 1) A merge could have happened, so we could currently be on a leaf
-	 * that holds what we were copying in the first place.
-	 * 2) A split could have happened, and now not all of the items we want
-	 * are on the same leaf.
-	 *
-	 * So we need to adjust how we search for holes, we need to drop the
-	 * path and re-search for the first extent key we found, and then walk
-	 * forward until we hit the last one we copied.
-	 */
-	if (need_find_last_extent) {
-		/* btrfs_prev_leaf could return 1 without releasing the path */
-		btrfs_release_path(src_path);
-		ret = btrfs_search_slot(NULL, inode->root, &first_key,
-				src_path, 0, 0);
-		if (ret < 0)
-			return ret;
-		ASSERT(ret == 0);
-		src = src_path->nodes[0];
-		i = src_path->slots[0];
-	} else {
-		i = start_slot;
-	}
-
-	/*
-	 * Ok so here we need to go through and fill in any holes we may have
-	 * to make sure that holes are punched for those areas in case they had
-	 * extents previously.
-	 */
-	while (!done) {
-		u64 offset, len;
-		u64 extent_end;
-
-		if (i >= btrfs_header_nritems(src_path->nodes[0])) {
-			ret = btrfs_next_leaf(inode->root, src_path);
-			if (ret < 0)
-				return ret;
-			ASSERT(ret == 0);
-			src = src_path->nodes[0];
-			i = 0;
-			need_find_last_extent = true;
-		}
-
-		btrfs_item_key_to_cpu(src, &key, i);
-		if (!btrfs_comp_cpu_keys(&key, &last_key))
-			done = true;
-		if (key.objectid != btrfs_ino(inode) ||
-		    key.type != BTRFS_EXTENT_DATA_KEY) {
-			i++;
-			continue;
-		}
-		extent = btrfs_item_ptr(src, i, struct btrfs_file_extent_item);
-		if (btrfs_file_extent_type(src, extent) ==
-		    BTRFS_FILE_EXTENT_INLINE) {
-			len = btrfs_file_extent_ram_bytes(src, extent);
-			extent_end = ALIGN(key.offset + len,
-					   fs_info->sectorsize);
-		} else {
-			len = btrfs_file_extent_num_bytes(src, extent);
-			extent_end = key.offset + len;
-		}
-		i++;
-
-		if (*last_extent == key.offset) {
-			*last_extent = extent_end;
-			continue;
-		}
-		offset = *last_extent;
-		len = key.offset - *last_extent;
-		ret = btrfs_insert_file_extent(trans, log, btrfs_ino(inode),
-				offset, 0, 0, len, 0, len, 0, 0, 0);
-		if (ret)
-			break;
-		*last_extent = extent_end;
-	}
-
-	/*
-	 * Check if there is a hole between the last extent found in our leaf
-	 * and the first extent in the next leaf. If there is one, we need to
-	 * log an explicit hole so that at replay time we can punch the hole.
-	 */
-	if (ret == 0 &&
-	    key.objectid == btrfs_ino(inode) &&
-	    key.type == BTRFS_EXTENT_DATA_KEY &&
-	    i == btrfs_header_nritems(src_path->nodes[0])) {
-		ret = btrfs_next_leaf(inode->root, src_path);
-		need_find_last_extent = true;
-		if (ret > 0) {
-			ret = 0;
-		} else if (ret == 0) {
-			btrfs_item_key_to_cpu(src_path->nodes[0], &key,
-					      src_path->slots[0]);
-			if (key.objectid == btrfs_ino(inode) &&
-			    key.type == BTRFS_EXTENT_DATA_KEY &&
-			    *last_extent < key.offset) {
-				const u64 len = key.offset - *last_extent;
-
-				ret = btrfs_insert_file_extent(trans, log,
-							       btrfs_ino(inode),
-							       *last_extent, 0,
-							       0, len, 0, len,
-							       0, 0, 0);
-				*last_extent += len;
-			}
-		}
-	}
-	/*
-	 * Need to let the callers know we dropped the path so they should
-	 * re-search.
-	 */
-	if (!ret && need_find_last_extent)
-		ret = 1;
 	return ret;
 }
 
@@ -4292,7 +4115,7 @@ static int log_extent_csums(struct btrfs_trans_handle *trans,
 						   struct btrfs_ordered_sum,
 						   list);
 		if (!ret)
-			ret = btrfs_csum_file_blocks(trans, log_root, sums);
+			ret = log_csums(trans, log_root, sums);
 		list_del(&sums->list);
 		kfree(sums);
 	}
@@ -4402,7 +4225,10 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans,
 	const u64 i_size = i_size_read(&inode->vfs_inode);
 	const u64 ino = btrfs_ino(inode);
 	struct btrfs_path *dst_path = NULL;
-	u64 last_extent = (u64)-1;
+	bool dropped_extents = false;
+	u64 truncate_offset = i_size;
+	struct extent_buffer *leaf;
+	int slot;
 	int ins_nr = 0;
 	int start_slot;
 	int ret;
@@ -4417,15 +4243,48 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans,
 	if (ret < 0)
 		goto out;
 
+	/*
+	 * We must check if there is a prealloc extent that starts before the
+	 * i_size and crosses the i_size boundary. This is to ensure later we
+	 * truncate down to the end of that extent and not to the i_size, as
+	 * otherwise we end up losing part of the prealloc extent after a log
+	 * replay and with an implicit hole if there is another prealloc extent
+	 * that starts at an offset beyond i_size.
+	 */
+	ret = btrfs_previous_item(root, path, ino, BTRFS_EXTENT_DATA_KEY);
+	if (ret < 0)
+		goto out;
+
+	if (ret == 0) {
+		struct btrfs_file_extent_item *ei;
+
+		leaf = path->nodes[0];
+		slot = path->slots[0];
+		ei = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
+
+		if (btrfs_file_extent_type(leaf, ei) ==
+		    BTRFS_FILE_EXTENT_PREALLOC) {
+			u64 extent_end;
+
+			btrfs_item_key_to_cpu(leaf, &key, slot);
+			extent_end = key.offset +
+				btrfs_file_extent_num_bytes(leaf, ei);
+
+			if (extent_end > i_size)
+				truncate_offset = extent_end;
+		}
+	} else {
+		ret = 0;
+	}
+
 	while (true) {
-		struct extent_buffer *leaf = path->nodes[0];
-		int slot = path->slots[0];
+		leaf = path->nodes[0];
+		slot = path->slots[0];
 
 		if (slot >= btrfs_header_nritems(leaf)) {
 			if (ins_nr > 0) {
 				ret = copy_items(trans, inode, dst_path, path,
-						 &last_extent, start_slot,
-						 ins_nr, 1, 0);
+						 start_slot, ins_nr, 1, 0);
 				if (ret < 0)
 					goto out;
 				ins_nr = 0;
@@ -4449,8 +4308,7 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans,
 			path->slots[0]++;
 			continue;
 		}
-		if (last_extent == (u64)-1) {
-			last_extent = key.offset;
+		if (!dropped_extents) {
 			/*
 			 * Avoid logging extent items logged in past fsync calls
 			 * and leading to duplicate keys in the log tree.
@@ -4459,11 +4317,12 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans,
 				ret = btrfs_truncate_inode_items(trans,
 							 root->log_root,
 							 &inode->vfs_inode,
-							 i_size,
+							 truncate_offset,
 							 BTRFS_EXTENT_DATA_KEY);
 			} while (ret == -EAGAIN);
 			if (ret)
 				goto out;
+			dropped_extents = true;
 		}
 		if (ins_nr == 0)
 			start_slot = slot;
@@ -4478,7 +4337,7 @@ static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans,
 		}
 	}
 	if (ins_nr > 0) {
-		ret = copy_items(trans, inode, dst_path, path, &last_extent,
+		ret = copy_items(trans, inode, dst_path, path,
 				 start_slot, ins_nr, 1, 0);
 		if (ret > 0)
 			ret = 0;
@@ -4665,13 +4524,8 @@ static int btrfs_log_all_xattrs(struct btrfs_trans_handle *trans,
 
 		if (slot >= nritems) {
 			if (ins_nr > 0) {
-				u64 last_extent = 0;
-
 				ret = copy_items(trans, inode, dst_path, path,
-						 &last_extent, start_slot,
-						 ins_nr, 1, 0);
-				/* can't be 1, extent items aren't processed */
-				ASSERT(ret <= 0);
+						 start_slot, ins_nr, 1, 0);
 				if (ret < 0)
 					return ret;
 				ins_nr = 0;
@@ -4695,13 +4549,8 @@ static int btrfs_log_all_xattrs(struct btrfs_trans_handle *trans,
 		cond_resched();
 	}
 	if (ins_nr > 0) {
-		u64 last_extent = 0;
-
 		ret = copy_items(trans, inode, dst_path, path,
-				 &last_extent, start_slot,
-				 ins_nr, 1, 0);
-		/* can't be 1, extent items aren't processed */
-		ASSERT(ret <= 0);
+				 start_slot, ins_nr, 1, 0);
 		if (ret < 0)
 			return ret;
 	}
@@ -4710,100 +4559,106 @@ static int btrfs_log_all_xattrs(struct btrfs_trans_handle *trans,
 }
 
 /*
- * If the no holes feature is enabled we need to make sure any hole between the
- * last extent and the i_size of our inode is explicitly marked in the log. This
- * is to make sure that doing something like:
- *
- *      1) create file with 128Kb of data
- *      2) truncate file to 64Kb
- *      3) truncate file to 256Kb
- *      4) fsync file
- *      5) <crash/power failure>
- *      6) mount fs and trigger log replay
- *
- * Will give us a file with a size of 256Kb, the first 64Kb of data match what
- * the file had in its first 64Kb of data at step 1 and the last 192Kb of the
- * file correspond to a hole. The presence of explicit holes in a log tree is
- * what guarantees that log replay will remove/adjust file extent items in the
- * fs/subvol tree.
- *
- * Here we do not need to care about holes between extents, that is already done
- * by copy_items(). We also only need to do this in the full sync path, where we
- * lookup for extents from the fs/subvol tree only. In the fast path case, we
- * lookup the list of modified extent maps and if any represents a hole, we
- * insert a corresponding extent representing a hole in the log tree.
+ * When using the NO_HOLES feature if we punched a hole that causes the
+ * deletion of entire leafs or all the extent items of the first leaf (the one
+ * that contains the inode item and references) we may end up not processing
+ * any extents, because there are no leafs with a generation matching the
+ * current transaction that have extent items for our inode. So we need to find
+ * if any holes exist and then log them. We also need to log holes after any
+ * truncate operation that changes the inode's size.
  */
-static int btrfs_log_trailing_hole(struct btrfs_trans_handle *trans,
-				   struct btrfs_root *root,
-				   struct btrfs_inode *inode,
-				   struct btrfs_path *path)
+static int btrfs_log_holes(struct btrfs_trans_handle *trans,
+			   struct btrfs_root *root,
+			   struct btrfs_inode *inode,
+			   struct btrfs_path *path)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
-	int ret;
 	struct btrfs_key key;
-	u64 hole_start;
-	u64 hole_size;
-	struct extent_buffer *leaf;
-	struct btrfs_root *log = root->log_root;
 	const u64 ino = btrfs_ino(inode);
 	const u64 i_size = i_size_read(&inode->vfs_inode);
+	u64 prev_extent_end = 0;
+	int ret;
 
-	if (!btrfs_fs_incompat(fs_info, NO_HOLES))
+	if (!btrfs_fs_incompat(fs_info, NO_HOLES) || i_size == 0)
 		return 0;
 
 	key.objectid = ino;
 	key.type = BTRFS_EXTENT_DATA_KEY;
-	key.offset = (u64)-1;
+	key.offset = 0;
 
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	ASSERT(ret != 0);
 	if (ret < 0)
 		return ret;
 
-	ASSERT(path->slots[0] > 0);
-	path->slots[0]--;
-	leaf = path->nodes[0];
-	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
+	while (true) {
+		struct extent_buffer *leaf = path->nodes[0];
 
-	if (key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY) {
-		/* inode does not have any extents */
-		hole_start = 0;
-		hole_size = i_size;
-	} else {
-		struct btrfs_file_extent_item *extent;
-		u64 len;
+		if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
+			ret = btrfs_next_leaf(root, path);
+			if (ret < 0)
+				return ret;
+			if (ret > 0) {
+				ret = 0;
+				break;
+			}
+			leaf = path->nodes[0];
+		}
 
-		/*
-		 * If there's an extent beyond i_size, an explicit hole was
-		 * already inserted by copy_items().
-		 */
-		if (key.offset >= i_size)
-			return 0;
+		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
+		if (key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY)
+			break;
 
-		extent = btrfs_item_ptr(leaf, path->slots[0],
-					struct btrfs_file_extent_item);
+		/* We have a hole, log it. */
+		if (prev_extent_end < key.offset) {
+			const u64 hole_len = key.offset - prev_extent_end;
 
-		if (btrfs_file_extent_type(leaf, extent) ==
-		    BTRFS_FILE_EXTENT_INLINE)
-			return 0;
+			/*
+			 * Release the path to avoid deadlocks with other code
+			 * paths that search the root while holding locks on
+			 * leafs from the log root.
+			 */
+			btrfs_release_path(path);
+			ret = btrfs_insert_file_extent(trans, root->log_root,
+						       ino, prev_extent_end, 0,
+						       0, hole_len, 0, hole_len,
+						       0, 0, 0);
+			if (ret < 0)
+				return ret;
 
-		len = btrfs_file_extent_num_bytes(leaf, extent);
-		/* Last extent goes beyond i_size, no need to log a hole. */
-		if (key.offset + len > i_size)
-			return 0;
-		hole_start = key.offset + len;
-		hole_size = i_size - hole_start;
+			/*
+			 * Search for the same key again in the root. Since it's
+			 * an extent item and we are holding the inode lock, the
+			 * key must still exist. If it doesn't just emit warning
+			 * and return an error to fall back to a transaction
+			 * commit.
+			 */
+			ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+			if (ret < 0)
+				return ret;
+			if (WARN_ON(ret > 0))
+				return -ENOENT;
+			leaf = path->nodes[0];
+		}
+
+		prev_extent_end = btrfs_file_extent_end(path);
+		path->slots[0]++;
+		cond_resched();
 	}
-	btrfs_release_path(path);
 
-	/* Last extent ends at i_size. */
-	if (hole_size == 0)
-		return 0;
+	if (prev_extent_end < i_size) {
+		u64 hole_len;
 
-	hole_size = ALIGN(hole_size, fs_info->sectorsize);
-	ret = btrfs_insert_file_extent(trans, log, ino, hole_start, 0, 0,
-				       hole_size, 0, hole_size, 0, 0, 0);
-	return ret;
+		btrfs_release_path(path);
+		hole_len = ALIGN(i_size - prev_extent_end, fs_info->sectorsize);
+		ret = btrfs_insert_file_extent(trans, root->log_root,
+					       ino, prev_extent_end, 0, 0,
+					       hole_len, 0, hole_len,
+					       0, 0, 0);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
 }
 
 /*
@@ -4983,7 +4838,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 		key.objectid = ino;
 		key.type = BTRFS_INODE_ITEM_KEY;
 		key.offset = 0;
-		inode = btrfs_iget(fs_info->sb, &key, root, NULL);
+		inode = btrfs_iget(fs_info->sb, &key, root);
 		/*
 		 * If the other inode that had a conflicting dir entry was
 		 * deleted in the current transaction, we need to log its parent
@@ -4993,8 +4848,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 			ret = PTR_ERR(inode);
 			if (ret == -ENOENT) {
 				key.objectid = parent;
-				inode = btrfs_iget(fs_info->sb, &key, root,
-						   NULL);
+				inode = btrfs_iget(fs_info->sb, &key, root);
 				if (IS_ERR(inode)) {
 					ret = PTR_ERR(inode);
 				} else {
@@ -5008,6 +4862,50 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 			continue;
 		}
 		/*
+		 * If the inode was already logged skip it - otherwise we can
+		 * hit an infinite loop. Example:
+		 *
+		 * From the commit root (previous transaction) we have the
+		 * following inodes:
+		 *
+		 * inode 257 a directory
+		 * inode 258 with references "zz" and "zz_link" on inode 257
+		 * inode 259 with reference "a" on inode 257
+		 *
+		 * And in the current (uncommitted) transaction we have:
+		 *
+		 * inode 257 a directory, unchanged
+		 * inode 258 with references "a" and "a2" on inode 257
+		 * inode 259 with reference "zz_link" on inode 257
+		 * inode 261 with reference "zz" on inode 257
+		 *
+		 * When logging inode 261 the following infinite loop could
+		 * happen if we don't skip already logged inodes:
+		 *
+		 * - we detect inode 258 as a conflicting inode, with inode 261
+		 *   on reference "zz", and log it;
+		 *
+		 * - we detect inode 259 as a conflicting inode, with inode 258
+		 *   on reference "a", and log it;
+		 *
+		 * - we detect inode 258 as a conflicting inode, with inode 259
+		 *   on reference "zz_link", and log it - again! After this we
+		 *   repeat the above steps forever.
+		 */
+		spin_lock(&BTRFS_I(inode)->lock);
+		/*
+		 * Check the inode's logged_trans only instead of
+		 * btrfs_inode_in_log(). This is because the last_log_commit of
+		 * the inode is not updated when we only log that it exists and
+		 * and it has the full sync bit set (see btrfs_log_inode()).
+		 */
+		if (BTRFS_I(inode)->logged_trans == trans->transid) {
+			spin_unlock(&BTRFS_I(inode)->lock);
+			btrfs_add_delayed_iput(inode);
+			continue;
+		}
+		spin_unlock(&BTRFS_I(inode)->lock);
+		/*
 		 * We are safe logging the other inode without acquiring its
 		 * lock as long as we log with the LOG_INODE_EXISTS mode. We
 		 * are safe against concurrent renames of the other inode as
@@ -5079,6 +4977,138 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
+static int copy_inode_items_to_log(struct btrfs_trans_handle *trans,
+				   struct btrfs_inode *inode,
+				   struct btrfs_key *min_key,
+				   const struct btrfs_key *max_key,
+				   struct btrfs_path *path,
+				   struct btrfs_path *dst_path,
+				   const u64 logged_isize,
+				   const bool recursive_logging,
+				   const int inode_only,
+				   struct btrfs_log_ctx *ctx,
+				   bool *need_log_inode_item)
+{
+	struct btrfs_root *root = inode->root;
+	int ins_start_slot = 0;
+	int ins_nr = 0;
+	int ret;
+
+	while (1) {
+		ret = btrfs_search_forward(root, min_key, path, trans->transid);
+		if (ret < 0)
+			return ret;
+		if (ret > 0) {
+			ret = 0;
+			break;
+		}
+again:
+		/* Note, ins_nr might be > 0 here, cleanup outside the loop */
+		if (min_key->objectid != max_key->objectid)
+			break;
+		if (min_key->type > max_key->type)
+			break;
+
+		if (min_key->type == BTRFS_INODE_ITEM_KEY)
+			*need_log_inode_item = false;
+
+		if ((min_key->type == BTRFS_INODE_REF_KEY ||
+		     min_key->type == BTRFS_INODE_EXTREF_KEY) &&
+		    inode->generation == trans->transid &&
+		    !recursive_logging) {
+			u64 other_ino = 0;
+			u64 other_parent = 0;
+
+			ret = btrfs_check_ref_name_override(path->nodes[0],
+					path->slots[0], min_key, inode,
+					&other_ino, &other_parent);
+			if (ret < 0) {
+				return ret;
+			} else if (ret > 0 && ctx &&
+				   other_ino != btrfs_ino(BTRFS_I(ctx->inode))) {
+				if (ins_nr > 0) {
+					ins_nr++;
+				} else {
+					ins_nr = 1;
+					ins_start_slot = path->slots[0];
+				}
+				ret = copy_items(trans, inode, dst_path, path,
+						 ins_start_slot, ins_nr,
+						 inode_only, logged_isize);
+				if (ret < 0)
+					return ret;
+				ins_nr = 0;
+
+				ret = log_conflicting_inodes(trans, root, path,
+						ctx, other_ino, other_parent);
+				if (ret)
+					return ret;
+				btrfs_release_path(path);
+				goto next_key;
+			}
+		}
+
+		/* Skip xattrs, we log them later with btrfs_log_all_xattrs() */
+		if (min_key->type == BTRFS_XATTR_ITEM_KEY) {
+			if (ins_nr == 0)
+				goto next_slot;
+			ret = copy_items(trans, inode, dst_path, path,
+					 ins_start_slot,
+					 ins_nr, inode_only, logged_isize);
+			if (ret < 0)
+				return ret;
+			ins_nr = 0;
+			goto next_slot;
+		}
+
+		if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) {
+			ins_nr++;
+			goto next_slot;
+		} else if (!ins_nr) {
+			ins_start_slot = path->slots[0];
+			ins_nr = 1;
+			goto next_slot;
+		}
+
+		ret = copy_items(trans, inode, dst_path, path, ins_start_slot,
+				 ins_nr, inode_only, logged_isize);
+		if (ret < 0)
+			return ret;
+		ins_nr = 1;
+		ins_start_slot = path->slots[0];
+next_slot:
+		path->slots[0]++;
+		if (path->slots[0] < btrfs_header_nritems(path->nodes[0])) {
+			btrfs_item_key_to_cpu(path->nodes[0], min_key,
+					      path->slots[0]);
+			goto again;
+		}
+		if (ins_nr) {
+			ret = copy_items(trans, inode, dst_path, path,
+					 ins_start_slot, ins_nr, inode_only,
+					 logged_isize);
+			if (ret < 0)
+				return ret;
+			ins_nr = 0;
+		}
+		btrfs_release_path(path);
+next_key:
+		if (min_key->offset < (u64)-1) {
+			min_key->offset++;
+		} else if (min_key->type < max_key->type) {
+			min_key->type++;
+			min_key->offset = 0;
+		} else {
+			break;
+		}
+	}
+	if (ins_nr)
+		ret = copy_items(trans, inode, dst_path, path, ins_start_slot,
+				 ins_nr, inode_only, logged_isize);
+
+	return ret;
+}
+
 /* log a single inode in the tree log.
  * At least one parent directory for this inode must exist in the tree
  * or be logged already.
@@ -5106,12 +5136,8 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 	struct btrfs_key min_key;
 	struct btrfs_key max_key;
 	struct btrfs_root *log = root->log_root;
-	u64 last_extent = 0;
 	int err = 0;
 	int ret;
-	int nritems;
-	int ins_start_slot = 0;
-	int ins_nr;
 	bool fast_search = false;
 	u64 ino = btrfs_ino(inode);
 	struct extent_map_tree *em_tree = &inode->extent_tree;
@@ -5242,150 +5268,12 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 		goto out_unlock;
 	}
 
-	while (1) {
-		ins_nr = 0;
-		ret = btrfs_search_forward(root, &min_key,
-					   path, trans->transid);
-		if (ret < 0) {
-			err = ret;
-			goto out_unlock;
-		}
-		if (ret != 0)
-			break;
-again:
-		/* note, ins_nr might be > 0 here, cleanup outside the loop */
-		if (min_key.objectid != ino)
-			break;
-		if (min_key.type > max_key.type)
-			break;
-
-		if (min_key.type == BTRFS_INODE_ITEM_KEY)
-			need_log_inode_item = false;
-
-		if ((min_key.type == BTRFS_INODE_REF_KEY ||
-		     min_key.type == BTRFS_INODE_EXTREF_KEY) &&
-		    inode->generation == trans->transid &&
-		    !recursive_logging) {
-			u64 other_ino = 0;
-			u64 other_parent = 0;
-
-			ret = btrfs_check_ref_name_override(path->nodes[0],
-					path->slots[0], &min_key, inode,
-					&other_ino, &other_parent);
-			if (ret < 0) {
-				err = ret;
-				goto out_unlock;
-			} else if (ret > 0 && ctx &&
-				   other_ino != btrfs_ino(BTRFS_I(ctx->inode))) {
-				if (ins_nr > 0) {
-					ins_nr++;
-				} else {
-					ins_nr = 1;
-					ins_start_slot = path->slots[0];
-				}
-				ret = copy_items(trans, inode, dst_path, path,
-						 &last_extent, ins_start_slot,
-						 ins_nr, inode_only,
-						 logged_isize);
-				if (ret < 0) {
-					err = ret;
-					goto out_unlock;
-				}
-				ins_nr = 0;
-
-				err = log_conflicting_inodes(trans, root, path,
-						ctx, other_ino, other_parent);
-				if (err)
-					goto out_unlock;
-				btrfs_release_path(path);
-				goto next_key;
-			}
-		}
-
-		/* Skip xattrs, we log them later with btrfs_log_all_xattrs() */
-		if (min_key.type == BTRFS_XATTR_ITEM_KEY) {
-			if (ins_nr == 0)
-				goto next_slot;
-			ret = copy_items(trans, inode, dst_path, path,
-					 &last_extent, ins_start_slot,
-					 ins_nr, inode_only, logged_isize);
-			if (ret < 0) {
-				err = ret;
-				goto out_unlock;
-			}
-			ins_nr = 0;
-			if (ret) {
-				btrfs_release_path(path);
-				continue;
-			}
-			goto next_slot;
-		}
-
-		if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) {
-			ins_nr++;
-			goto next_slot;
-		} else if (!ins_nr) {
-			ins_start_slot = path->slots[0];
-			ins_nr = 1;
-			goto next_slot;
-		}
-
-		ret = copy_items(trans, inode, dst_path, path, &last_extent,
-				 ins_start_slot, ins_nr, inode_only,
-				 logged_isize);
-		if (ret < 0) {
-			err = ret;
-			goto out_unlock;
-		}
-		if (ret) {
-			ins_nr = 0;
-			btrfs_release_path(path);
-			continue;
-		}
-		ins_nr = 1;
-		ins_start_slot = path->slots[0];
-next_slot:
-
-		nritems = btrfs_header_nritems(path->nodes[0]);
-		path->slots[0]++;
-		if (path->slots[0] < nritems) {
-			btrfs_item_key_to_cpu(path->nodes[0], &min_key,
-					      path->slots[0]);
-			goto again;
-		}
-		if (ins_nr) {
-			ret = copy_items(trans, inode, dst_path, path,
-					 &last_extent, ins_start_slot,
-					 ins_nr, inode_only, logged_isize);
-			if (ret < 0) {
-				err = ret;
-				goto out_unlock;
-			}
-			ret = 0;
-			ins_nr = 0;
-		}
-		btrfs_release_path(path);
-next_key:
-		if (min_key.offset < (u64)-1) {
-			min_key.offset++;
-		} else if (min_key.type < max_key.type) {
-			min_key.type++;
-			min_key.offset = 0;
-		} else {
-			break;
-		}
-	}
-	if (ins_nr) {
-		ret = copy_items(trans, inode, dst_path, path, &last_extent,
-				 ins_start_slot, ins_nr, inode_only,
-				 logged_isize);
-		if (ret < 0) {
-			err = ret;
-			goto out_unlock;
-		}
-		ret = 0;
-		ins_nr = 0;
-	}
+	err = copy_inode_items_to_log(trans, inode, &min_key, &max_key,
+				      path, dst_path, logged_isize,
+				      recursive_logging, inode_only, ctx,
+				      &need_log_inode_item);
+	if (err)
+		goto out_unlock;
 
 	btrfs_release_path(path);
 	btrfs_release_path(dst_path);
@@ -5396,7 +5284,7 @@ next_key:
 	if (max_key.type >= BTRFS_EXTENT_DATA_KEY && !fast_search) {
 		btrfs_release_path(path);
 		btrfs_release_path(dst_path);
-		err = btrfs_log_trailing_hole(trans, root, inode, path);
+		err = btrfs_log_holes(trans, root, inode, path);
 		if (err)
 			goto out_unlock;
 	}
@@ -5699,7 +5587,7 @@ process_leaf:
 				continue;
 
 			btrfs_release_path(path);
-			di_inode = btrfs_iget(fs_info->sb, &di_key, root, NULL);
+			di_inode = btrfs_iget(fs_info->sb, &di_key, root);
 			if (IS_ERR(di_inode)) {
 				ret = PTR_ERR(di_inode);
 				goto next_dir_inode;
@@ -5825,8 +5713,7 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
 				cur_offset = item_size;
 			}
 
-			dir_inode = btrfs_iget(fs_info->sb, &inode_key,
-					       root, NULL);
+			dir_inode = btrfs_iget(fs_info->sb, &inode_key, root);
 			/*
 			 * If the parent inode was deleted, return an error to
 			 * fallback to a transaction commit. This is to prevent
@@ -5900,7 +5787,7 @@ static int log_new_ancestors(struct btrfs_trans_handle *trans,
 		search_key.objectid = found_key.offset;
 		search_key.type = BTRFS_INODE_ITEM_KEY;
 		search_key.offset = 0;
-		inode = btrfs_iget(fs_info->sb, &search_key, root, NULL);
+		inode = btrfs_iget(fs_info->sb, &search_key, root);
 		if (IS_ERR(inode))
 			return PTR_ERR(inode);
 
@@ -6299,7 +6186,7 @@ again:
 		if (found_key.objectid != BTRFS_TREE_LOG_OBJECTID)
 			break;
 
-		log = btrfs_read_fs_root(log_root_tree, &found_key);
+		log = btrfs_read_tree_root(log_root_tree, &found_key);
 		if (IS_ERR(log)) {
 			ret = PTR_ERR(log);
 			btrfs_handle_fs_error(fs_info, ret,
@@ -6311,12 +6198,29 @@ again:
 		tmp_key.type = BTRFS_ROOT_ITEM_KEY;
 		tmp_key.offset = (u64)-1;
 
-		wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key);
+		wc.replay_dest = btrfs_get_fs_root(fs_info, &tmp_key, true);
 		if (IS_ERR(wc.replay_dest)) {
 			ret = PTR_ERR(wc.replay_dest);
-			free_extent_buffer(log->node);
-			free_extent_buffer(log->commit_root);
-			kfree(log);
+
+			/*
+			 * We didn't find the subvol, likely because it was
+			 * deleted.  This is ok, simply skip this log and go to
+			 * the next one.
+			 *
+			 * We need to exclude the root because we can't have
+			 * other log replays overwriting this log as we'll read
+			 * it back in a few more times.  This will keep our
+			 * block from being modified, and we'll just bail for
+			 * each subsequent pass.
+			 */
+			if (ret == -ENOENT)
+				ret = btrfs_pin_extent_for_log_replay(trans,
+							log->node->start,
+							log->node->len);
+			btrfs_put_root(log);
+
+			if (!ret)
+				goto next;
 			btrfs_handle_fs_error(fs_info, ret,
 				"Couldn't read target root for tree log recovery.");
 			goto error;
@@ -6348,17 +6252,16 @@ again:
 						  &root->highest_objectid);
 		}
 
-		key.offset = found_key.offset - 1;
 		wc.replay_dest->log_root = NULL;
-		free_extent_buffer(log->node);
-		free_extent_buffer(log->commit_root);
-		kfree(log);
+		btrfs_put_root(wc.replay_dest);
+		btrfs_put_root(log);
 
 		if (ret)
 			goto error;
-
+next:
 		if (found_key.offset == 0)
 			break;
+		key.offset = found_key.offset - 1;
 	}
 	btrfs_release_path(path);
 
@@ -6382,10 +6285,9 @@ again:
 	if (ret)
 		return ret;
 
-	free_extent_buffer(log_root_tree->node);
 	log_root_tree->log_root = NULL;
 	clear_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags);
-	kfree(log_root_tree);
+	btrfs_put_root(log_root_tree);
 
 	return 0;
 error:
diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c
index 91caab63bdf5..76671a6bcb61 100644
--- a/fs/btrfs/uuid-tree.c
+++ b/fs/btrfs/uuid-tree.c
@@ -246,9 +246,53 @@ out:
 	return ret;
 }
 
-int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info,
-			    int (*check_func)(struct btrfs_fs_info *, u8 *, u8,
-					      u64))
+/*
+ * Check if there's an matching subvolume for given UUID
+ *
+ * Return:
+ * 0	check succeeded, the entry is not outdated
+ * > 0	if the check failed, the caller should remove the entry
+ * < 0	if an error occurred
+ */
+static int btrfs_check_uuid_tree_entry(struct btrfs_fs_info *fs_info,
+				       u8 *uuid, u8 type, u64 subvolid)
+{
+	struct btrfs_key key;
+	int ret = 0;
+	struct btrfs_root *subvol_root;
+
+	if (type != BTRFS_UUID_KEY_SUBVOL &&
+	    type != BTRFS_UUID_KEY_RECEIVED_SUBVOL)
+		goto out;
+
+	key.objectid = subvolid;
+	key.type = BTRFS_ROOT_ITEM_KEY;
+	key.offset = (u64)-1;
+	subvol_root = btrfs_get_fs_root(fs_info, &key, true);
+	if (IS_ERR(subvol_root)) {
+		ret = PTR_ERR(subvol_root);
+		if (ret == -ENOENT)
+			ret = 1;
+		goto out;
+	}
+
+	switch (type) {
+	case BTRFS_UUID_KEY_SUBVOL:
+		if (memcmp(uuid, subvol_root->root_item.uuid, BTRFS_UUID_SIZE))
+			ret = 1;
+		break;
+	case BTRFS_UUID_KEY_RECEIVED_SUBVOL:
+		if (memcmp(uuid, subvol_root->root_item.received_uuid,
+			   BTRFS_UUID_SIZE))
+			ret = 1;
+		break;
+	}
+	btrfs_put_root(subvol_root);
+out:
+	return ret;
+}
+
+int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_root *root = fs_info->uuid_root;
 	struct btrfs_key key;
@@ -278,6 +322,10 @@ again_search_slot:
 	}
 
 	while (1) {
+		if (btrfs_fs_closing(fs_info)) {
+			ret = -EINTR;
+			goto out;
+		}
 		cond_resched();
 		leaf = path->nodes[0];
 		slot = path->slots[0];
@@ -305,7 +353,8 @@ again_search_slot:
 			read_extent_buffer(leaf, &subid_le, offset,
 					   sizeof(subid_le));
 			subid_cpu = le64_to_cpu(subid_le);
-			ret = check_func(fs_info, uuid, key.type, subid_cpu);
+			ret = btrfs_check_uuid_tree_entry(fs_info, uuid,
+							  key.type, subid_cpu);
 			if (ret < 0)
 				goto out;
 			if (ret > 0) {
@@ -324,6 +373,8 @@ again_search_slot:
 				}
 				if (ret < 0 && ret != -ENOENT)
 					goto out;
+				key.offset++;
+				goto again_search_slot;
 			}
 			item_size -= sizeof(subid_le);
 			offset += sizeof(subid_le);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index e04409f85063..c1909e5f4506 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -6,7 +6,6 @@
 #include <linux/sched.h>
 #include <linux/bio.h>
 #include <linux/slab.h>
-#include <linux/buffer_head.h>
 #include <linux/blkdev.h>
 #include <linux/ratelimit.h>
 #include <linux/kthread.h>
@@ -30,6 +29,7 @@
 #include "tree-checker.h"
 #include "space-info.h"
 #include "block-group.h"
+#include "discard.h"
 
 const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 	[BTRFS_RAID_RAID10] = {
@@ -58,6 +58,32 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.bg_flag	= BTRFS_BLOCK_GROUP_RAID1,
 		.mindev_error	= BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET,
 	},
+	[BTRFS_RAID_RAID1C3] = {
+		.sub_stripes	= 1,
+		.dev_stripes	= 1,
+		.devs_max	= 3,
+		.devs_min	= 3,
+		.tolerated_failures = 2,
+		.devs_increment	= 3,
+		.ncopies	= 3,
+		.nparity        = 0,
+		.raid_name	= "raid1c3",
+		.bg_flag	= BTRFS_BLOCK_GROUP_RAID1C3,
+		.mindev_error	= BTRFS_ERROR_DEV_RAID1C3_MIN_NOT_MET,
+	},
+	[BTRFS_RAID_RAID1C4] = {
+		.sub_stripes	= 1,
+		.dev_stripes	= 1,
+		.devs_max	= 4,
+		.devs_min	= 4,
+		.tolerated_failures = 3,
+		.devs_increment	= 4,
+		.ncopies	= 4,
+		.nparity        = 0,
+		.raid_name	= "raid1c4",
+		.bg_flag	= BTRFS_BLOCK_GROUP_RAID1C4,
+		.mindev_error	= BTRFS_ERROR_DEV_RAID1C4_MIN_NOT_MET,
+	},
 	[BTRFS_RAID_DUP] = {
 		.sub_stripes	= 1,
 		.dev_stripes	= 2,
@@ -297,7 +323,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
 
 DEFINE_MUTEX(uuid_mutex);
 static LIST_HEAD(fs_uuids);
-struct list_head *btrfs_get_fs_uuids(void)
+struct list_head * __attribute_const__ btrfs_get_fs_uuids(void)
 {
 	return &fs_uuids;
 }
@@ -397,8 +423,6 @@ static struct btrfs_device *__alloc_device(void)
 	INIT_LIST_HEAD(&dev->dev_alloc_list);
 	INIT_LIST_HEAD(&dev->post_commit_list);
 
-	spin_lock_init(&dev->io_lock);
-
 	atomic_set(&dev->reada_in_flight, 0);
 	atomic_set(&dev->dev_stats_ccnt, 0);
 	btrfs_device_data_ordered_init(dev);
@@ -416,39 +440,6 @@ static noinline struct btrfs_fs_devices *find_fsid(
 
 	ASSERT(fsid);
 
-	if (metadata_fsid) {
-		/*
-		 * Handle scanned device having completed its fsid change but
-		 * belonging to a fs_devices that was created by first scanning
-		 * a device which didn't have its fsid/metadata_uuid changed
-		 * at all and the CHANGING_FSID_V2 flag set.
-		 */
-		list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
-			if (fs_devices->fsid_change &&
-			    memcmp(metadata_fsid, fs_devices->fsid,
-				   BTRFS_FSID_SIZE) == 0 &&
-			    memcmp(fs_devices->fsid, fs_devices->metadata_uuid,
-				   BTRFS_FSID_SIZE) == 0) {
-				return fs_devices;
-			}
-		}
-		/*
-		 * Handle scanned device having completed its fsid change but
-		 * belonging to a fs_devices that was created by a device that
-		 * has an outdated pair of fsid/metadata_uuid and
-		 * CHANGING_FSID_V2 flag set.
-		 */
-		list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
-			if (fs_devices->fsid_change &&
-			    memcmp(fs_devices->metadata_uuid,
-				   fs_devices->fsid, BTRFS_FSID_SIZE) != 0 &&
-			    memcmp(metadata_fsid, fs_devices->metadata_uuid,
-				   BTRFS_FSID_SIZE) == 0) {
-				return fs_devices;
-			}
-		}
-	}
-
 	/* Handle non-split brain cases */
 	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
 		if (metadata_fsid) {
@@ -464,10 +455,51 @@ static noinline struct btrfs_fs_devices *find_fsid(
 	return NULL;
 }
 
+static struct btrfs_fs_devices *find_fsid_with_metadata_uuid(
+				struct btrfs_super_block *disk_super)
+{
+
+	struct btrfs_fs_devices *fs_devices;
+
+	/*
+	 * Handle scanned device having completed its fsid change but
+	 * belonging to a fs_devices that was created by first scanning
+	 * a device which didn't have its fsid/metadata_uuid changed
+	 * at all and the CHANGING_FSID_V2 flag set.
+	 */
+	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
+		if (fs_devices->fsid_change &&
+		    memcmp(disk_super->metadata_uuid, fs_devices->fsid,
+			   BTRFS_FSID_SIZE) == 0 &&
+		    memcmp(fs_devices->fsid, fs_devices->metadata_uuid,
+			   BTRFS_FSID_SIZE) == 0) {
+			return fs_devices;
+		}
+	}
+	/*
+	 * Handle scanned device having completed its fsid change but
+	 * belonging to a fs_devices that was created by a device that
+	 * has an outdated pair of fsid/metadata_uuid and
+	 * CHANGING_FSID_V2 flag set.
+	 */
+	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
+		if (fs_devices->fsid_change &&
+		    memcmp(fs_devices->metadata_uuid,
+			   fs_devices->fsid, BTRFS_FSID_SIZE) != 0 &&
+		    memcmp(disk_super->metadata_uuid, fs_devices->metadata_uuid,
+			   BTRFS_FSID_SIZE) == 0) {
+			return fs_devices;
+		}
+	}
+
+	return find_fsid(disk_super->fsid, disk_super->metadata_uuid);
+}
+
+
 static int
 btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder,
 		      int flush, struct block_device **bdev,
-		      struct buffer_head **bh)
+		      struct btrfs_super_block **disk_super)
 {
 	int ret;
 
@@ -486,9 +518,9 @@ btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder,
 		goto error;
 	}
 	invalidate_bdev(*bdev);
-	*bh = btrfs_read_dev_super(*bdev);
-	if (IS_ERR(*bh)) {
-		ret = PTR_ERR(*bh);
+	*disk_super = btrfs_read_dev_super(*bdev);
+	if (IS_ERR(*disk_super)) {
+		ret = PTR_ERR(*disk_super);
 		blkdev_put(*bdev, flags);
 		goto error;
 	}
@@ -497,216 +529,9 @@ btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder,
 
 error:
 	*bdev = NULL;
-	*bh = NULL;
 	return ret;
 }
 
-static void requeue_list(struct btrfs_pending_bios *pending_bios,
-			struct bio *head, struct bio *tail)
-{
-
-	struct bio *old_head;
-
-	old_head = pending_bios->head;
-	pending_bios->head = head;
-	if (pending_bios->tail)
-		tail->bi_next = old_head;
-	else
-		pending_bios->tail = tail;
-}
-
-/*
- * we try to collect pending bios for a device so we don't get a large
- * number of procs sending bios down to the same device.  This greatly
- * improves the schedulers ability to collect and merge the bios.
- *
- * But, it also turns into a long list of bios to process and that is sure
- * to eventually make the worker thread block.  The solution here is to
- * make some progress and then put this work struct back at the end of
- * the list if the block device is congested.  This way, multiple devices
- * can make progress from a single worker thread.
- */
-static noinline void run_scheduled_bios(struct btrfs_device *device)
-{
-	struct btrfs_fs_info *fs_info = device->fs_info;
-	struct bio *pending;
-	struct backing_dev_info *bdi;
-	struct btrfs_pending_bios *pending_bios;
-	struct bio *tail;
-	struct bio *cur;
-	int again = 0;
-	unsigned long num_run;
-	unsigned long batch_run = 0;
-	unsigned long last_waited = 0;
-	int force_reg = 0;
-	int sync_pending = 0;
-	struct blk_plug plug;
-
-	/*
-	 * this function runs all the bios we've collected for
-	 * a particular device.  We don't want to wander off to
-	 * another device without first sending all of these down.
-	 * So, setup a plug here and finish it off before we return
-	 */
-	blk_start_plug(&plug);
-
-	bdi = device->bdev->bd_bdi;
-
-loop:
-	spin_lock(&device->io_lock);
-
-loop_lock:
-	num_run = 0;
-
-	/* take all the bios off the list at once and process them
-	 * later on (without the lock held).  But, remember the
-	 * tail and other pointers so the bios can be properly reinserted
-	 * into the list if we hit congestion
-	 */
-	if (!force_reg && device->pending_sync_bios.head) {
-		pending_bios = &device->pending_sync_bios;
-		force_reg = 1;
-	} else {
-		pending_bios = &device->pending_bios;
-		force_reg = 0;
-	}
-
-	pending = pending_bios->head;
-	tail = pending_bios->tail;
-	WARN_ON(pending && !tail);
-
-	/*
-	 * if pending was null this time around, no bios need processing
-	 * at all and we can stop.  Otherwise it'll loop back up again
-	 * and do an additional check so no bios are missed.
-	 *
-	 * device->running_pending is used to synchronize with the
-	 * schedule_bio code.
-	 */
-	if (device->pending_sync_bios.head == NULL &&
-	    device->pending_bios.head == NULL) {
-		again = 0;
-		device->running_pending = 0;
-	} else {
-		again = 1;
-		device->running_pending = 1;
-	}
-
-	pending_bios->head = NULL;
-	pending_bios->tail = NULL;
-
-	spin_unlock(&device->io_lock);
-
-	while (pending) {
-
-		rmb();
-		/* we want to work on both lists, but do more bios on the
-		 * sync list than the regular list
-		 */
-		if ((num_run > 32 &&
-		    pending_bios != &device->pending_sync_bios &&
-		    device->pending_sync_bios.head) ||
-		   (num_run > 64 && pending_bios == &device->pending_sync_bios &&
-		    device->pending_bios.head)) {
-			spin_lock(&device->io_lock);
-			requeue_list(pending_bios, pending, tail);
-			goto loop_lock;
-		}
-
-		cur = pending;
-		pending = pending->bi_next;
-		cur->bi_next = NULL;
-
-		BUG_ON(atomic_read(&cur->__bi_cnt) == 0);
-
-		/*
-		 * if we're doing the sync list, record that our
-		 * plug has some sync requests on it
-		 *
-		 * If we're doing the regular list and there are
-		 * sync requests sitting around, unplug before
-		 * we add more
-		 */
-		if (pending_bios == &device->pending_sync_bios) {
-			sync_pending = 1;
-		} else if (sync_pending) {
-			blk_finish_plug(&plug);
-			blk_start_plug(&plug);
-			sync_pending = 0;
-		}
-
-		btrfsic_submit_bio(cur);
-		num_run++;
-		batch_run++;
-
-		cond_resched();
-
-		/*
-		 * we made progress, there is more work to do and the bdi
-		 * is now congested.  Back off and let other work structs
-		 * run instead
-		 */
-		if (pending && bdi_write_congested(bdi) && batch_run > 8 &&
-		    fs_info->fs_devices->open_devices > 1) {
-			struct io_context *ioc;
-
-			ioc = current->io_context;
-
-			/*
-			 * the main goal here is that we don't want to
-			 * block if we're going to be able to submit
-			 * more requests without blocking.
-			 *
-			 * This code does two great things, it pokes into
-			 * the elevator code from a filesystem _and_
-			 * it makes assumptions about how batching works.
-			 */
-			if (ioc && ioc->nr_batch_requests > 0 &&
-			    time_before(jiffies, ioc->last_waited + HZ/50UL) &&
-			    (last_waited == 0 ||
-			     ioc->last_waited == last_waited)) {
-				/*
-				 * we want to go through our batch of
-				 * requests and stop.  So, we copy out
-				 * the ioc->last_waited time and test
-				 * against it before looping
-				 */
-				last_waited = ioc->last_waited;
-				cond_resched();
-				continue;
-			}
-			spin_lock(&device->io_lock);
-			requeue_list(pending_bios, pending, tail);
-			device->running_pending = 1;
-
-			spin_unlock(&device->io_lock);
-			btrfs_queue_work(fs_info->submit_workers,
-					 &device->work);
-			goto done;
-		}
-	}
-
-	cond_resched();
-	if (again)
-		goto loop;
-
-	spin_lock(&device->io_lock);
-	if (device->pending_bios.head || device->pending_sync_bios.head)
-		goto loop_lock;
-	spin_unlock(&device->io_lock);
-
-done:
-	blk_finish_plug(&plug);
-}
-
-static void pending_bios_fn(struct btrfs_work *work)
-{
-	struct btrfs_device *device;
-
-	device = container_of(work, struct btrfs_device, work);
-	run_scheduled_bios(device);
-}
-
 static bool device_path_matched(const char *path, struct btrfs_device *device)
 {
 	int found;
@@ -784,7 +609,6 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices,
 {
 	struct request_queue *q;
 	struct block_device *bdev;
-	struct buffer_head *bh;
 	struct btrfs_super_block *disk_super;
 	u64 devid;
 	int ret;
@@ -795,17 +619,16 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices,
 		return -EINVAL;
 
 	ret = btrfs_get_bdev_and_sb(device->name->str, flags, holder, 1,
-				    &bdev, &bh);
+				    &bdev, &disk_super);
 	if (ret)
 		return ret;
 
-	disk_super = (struct btrfs_super_block *)bh->b_data;
 	devid = btrfs_stack_device_id(&disk_super->dev_item);
 	if (devid != device->devid)
-		goto error_brelse;
+		goto error_free_page;
 
 	if (memcmp(device->uuid, disk_super->dev_item.uuid, BTRFS_UUID_SIZE))
-		goto error_brelse;
+		goto error_free_page;
 
 	device->generation = btrfs_super_generation(disk_super);
 
@@ -814,11 +637,11 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices,
 		    BTRFS_FEATURE_INCOMPAT_METADATA_UUID) {
 			pr_err(
 		"BTRFS: Invalid seeding and uuid-changed device detected\n");
-			goto error_brelse;
+			goto error_free_page;
 		}
 
 		clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
-		fs_devices->seeding = 1;
+		fs_devices->seeding = true;
 	} else {
 		if (bdev_read_only(bdev))
 			clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
@@ -828,7 +651,7 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices,
 
 	q = bdev_get_queue(bdev);
 	if (!blk_queue_nonrot(q))
-		fs_devices->rotating = 1;
+		fs_devices->rotating = true;
 
 	device->bdev = bdev;
 	clear_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
@@ -840,12 +663,12 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices,
 		fs_devices->rw_devices++;
 		list_add_tail(&device->dev_alloc_list, &fs_devices->alloc_list);
 	}
-	brelse(bh);
+	btrfs_release_disk_super(disk_super);
 
 	return 0;
 
-error_brelse:
-	brelse(bh);
+error_free_page:
+	btrfs_release_disk_super(disk_super);
 	blkdev_put(bdev, flags);
 
 	return -EINVAL;
@@ -853,7 +676,9 @@ error_brelse:
 
 /*
  * Handle scanned device having its CHANGING_FSID_V2 flag set and the fs_devices
- * being created with a disk that has already completed its fsid change.
+ * being created with a disk that has already completed its fsid change. Such
+ * disk can belong to an fs which has its FSID changed or to one which doesn't.
+ * Handle both cases here.
  */
 static struct btrfs_fs_devices *find_fsid_inprogress(
 					struct btrfs_super_block *disk_super)
@@ -869,7 +694,7 @@ static struct btrfs_fs_devices *find_fsid_inprogress(
 		}
 	}
 
-	return NULL;
+	return find_fsid(disk_super->fsid, NULL);
 }
 
 
@@ -881,17 +706,54 @@ static struct btrfs_fs_devices *find_fsid_changed(
 	/*
 	 * Handles the case where scanned device is part of an fs that had
 	 * multiple successful changes of FSID but curently device didn't
-	 * observe it. Meaning our fsid will be different than theirs.
+	 * observe it. Meaning our fsid will be different than theirs. We need
+	 * to handle two subcases :
+	 *  1 - The fs still continues to have different METADATA/FSID uuids.
+	 *  2 - The fs is switched back to its original FSID (METADATA/FSID
+	 *  are equal).
 	 */
 	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
+		/* Changed UUIDs */
 		if (memcmp(fs_devices->metadata_uuid, fs_devices->fsid,
 			   BTRFS_FSID_SIZE) != 0 &&
 		    memcmp(fs_devices->metadata_uuid, disk_super->metadata_uuid,
 			   BTRFS_FSID_SIZE) == 0 &&
 		    memcmp(fs_devices->fsid, disk_super->fsid,
-			   BTRFS_FSID_SIZE) != 0) {
+			   BTRFS_FSID_SIZE) != 0)
+			return fs_devices;
+
+		/* Unchanged UUIDs */
+		if (memcmp(fs_devices->metadata_uuid, fs_devices->fsid,
+			   BTRFS_FSID_SIZE) == 0 &&
+		    memcmp(fs_devices->fsid, disk_super->metadata_uuid,
+			   BTRFS_FSID_SIZE) == 0)
+			return fs_devices;
+	}
+
+	return NULL;
+}
+
+static struct btrfs_fs_devices *find_fsid_reverted_metadata(
+				struct btrfs_super_block *disk_super)
+{
+	struct btrfs_fs_devices *fs_devices;
+
+	/*
+	 * Handle the case where the scanned device is part of an fs whose last
+	 * metadata UUID change reverted it to the original FSID. At the same
+	 * time * fs_devices was first created by another constitutent device
+	 * which didn't fully observe the operation. This results in an
+	 * btrfs_fs_devices created with metadata/fsid different AND
+	 * btrfs_fs_devices::fsid_change set AND the metadata_uuid of the
+	 * fs_devices equal to the FSID of the disk.
+	 */
+	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
+		if (memcmp(fs_devices->fsid, fs_devices->metadata_uuid,
+			   BTRFS_FSID_SIZE) != 0 &&
+		    memcmp(fs_devices->metadata_uuid, disk_super->fsid,
+			   BTRFS_FSID_SIZE) == 0 &&
+		    fs_devices->fsid_change)
 			return fs_devices;
-		}
 	}
 
 	return NULL;
@@ -918,24 +780,16 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 					BTRFS_SUPER_FLAG_CHANGING_FSID_V2);
 
 	if (fsid_change_in_progress) {
-		if (!has_metadata_uuid) {
-			/*
-			 * When we have an image which has CHANGING_FSID_V2 set
-			 * it might belong to either a filesystem which has
-			 * disks with completed fsid change or it might belong
-			 * to fs with no UUID changes in effect, handle both.
-			 */
+		if (!has_metadata_uuid)
 			fs_devices = find_fsid_inprogress(disk_super);
-			if (!fs_devices)
-				fs_devices = find_fsid(disk_super->fsid, NULL);
-		} else {
+		else
 			fs_devices = find_fsid_changed(disk_super);
-		}
 	} else if (has_metadata_uuid) {
-		fs_devices = find_fsid(disk_super->fsid,
-				       disk_super->metadata_uuid);
+		fs_devices = find_fsid_with_metadata_uuid(disk_super);
 	} else {
-		fs_devices = find_fsid(disk_super->fsid, NULL);
+		fs_devices = find_fsid_reverted_metadata(disk_super);
+		if (!fs_devices)
+			fs_devices = find_fsid(disk_super->fsid, NULL);
 	}
 
 
@@ -965,12 +819,18 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 		 * a device which had the CHANGING_FSID_V2 flag then replace the
 		 * metadata_uuid/fsid values of the fs_devices.
 		 */
-		if (has_metadata_uuid && fs_devices->fsid_change &&
+		if (fs_devices->fsid_change &&
 		    found_transid > fs_devices->latest_generation) {
 			memcpy(fs_devices->fsid, disk_super->fsid,
 					BTRFS_FSID_SIZE);
-			memcpy(fs_devices->metadata_uuid,
-					disk_super->metadata_uuid, BTRFS_FSID_SIZE);
+
+			if (has_metadata_uuid)
+				memcpy(fs_devices->metadata_uuid,
+				       disk_super->metadata_uuid,
+				       BTRFS_FSID_SIZE);
+			else
+				memcpy(fs_devices->metadata_uuid,
+				       disk_super->fsid, BTRFS_FSID_SIZE);
 
 			fs_devices->fsid_change = false;
 		}
@@ -1005,11 +865,15 @@ static noinline struct btrfs_device *device_list_add(const char *path,
 		*new_device_added = true;
 
 		if (disk_super->label[0])
-			pr_info("BTRFS: device label %s devid %llu transid %llu %s\n",
-				disk_super->label, devid, found_transid, path);
+			pr_info(
+	"BTRFS: device label %s devid %llu transid %llu %s scanned by %s (%d)\n",
+				disk_super->label, devid, found_transid, path,
+				current->comm, task_pid_nr(current));
 		else
-			pr_info("BTRFS: device fsid %pU devid %llu transid %llu %s\n",
-				disk_super->fsid, devid, found_transid, path);
+			pr_info(
+	"BTRFS: device fsid %pU devid %llu transid %llu %s scanned by %s (%d)\n",
+				disk_super->fsid, devid, found_transid, path,
+				current->comm, task_pid_nr(current));
 
 	} else if (!device->name || strcmp(device->name->str, path)) {
 		/*
@@ -1244,11 +1108,6 @@ static void btrfs_close_bdev(struct btrfs_device *device)
 static void btrfs_close_one_device(struct btrfs_device *device)
 {
 	struct btrfs_fs_devices *fs_devices = device->fs_devices;
-	struct btrfs_device *new_device;
-	struct rcu_string *name;
-
-	if (device->bdev)
-		fs_devices->open_devices--;
 
 	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) &&
 	    device->devid != BTRFS_DEV_REPLACE_DEVID) {
@@ -1260,23 +1119,22 @@ static void btrfs_close_one_device(struct btrfs_device *device)
 		fs_devices->missing_devices--;
 
 	btrfs_close_bdev(device);
-
-	new_device = btrfs_alloc_device(NULL, &device->devid,
-					device->uuid);
-	BUG_ON(IS_ERR(new_device)); /* -ENOMEM */
-
-	/* Safe because we are under uuid_mutex */
-	if (device->name) {
-		name = rcu_string_strdup(device->name->str, GFP_NOFS);
-		BUG_ON(!name); /* -ENOMEM */
-		rcu_assign_pointer(new_device->name, name);
+	if (device->bdev) {
+		fs_devices->open_devices--;
+		device->bdev = NULL;
 	}
+	clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
 
-	list_replace_rcu(&device->dev_list, &new_device->dev_list);
-	new_device->fs_devices = device->fs_devices;
+	device->fs_info = NULL;
+	atomic_set(&device->dev_stats_ccnt, 0);
+	extent_io_tree_release(&device->alloc_state);
 
-	synchronize_rcu();
-	btrfs_free_device(device);
+	/* Verify the device is back in a pristine state  */
+	ASSERT(!test_bit(BTRFS_DEV_STATE_FLUSH_SENT, &device->dev_state));
+	ASSERT(!test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state));
+	ASSERT(list_empty(&device->dev_alloc_list));
+	ASSERT(list_empty(&device->post_commit_list));
+	ASSERT(atomic_read(&device->reada_in_flight) == 0);
 }
 
 static int close_fs_devices(struct btrfs_fs_devices *fs_devices)
@@ -1295,7 +1153,7 @@ static int close_fs_devices(struct btrfs_fs_devices *fs_devices)
 	WARN_ON(fs_devices->open_devices);
 	WARN_ON(fs_devices->rw_devices);
 	fs_devices->opened = 0;
-	fs_devices->seeding = 0;
+	fs_devices->seeding = false;
 
 	return 0;
 }
@@ -1347,6 +1205,7 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices,
 	fs_devices->opened = 1;
 	fs_devices->latest_bdev = latest_dev->bdev;
 	fs_devices->total_rw_bytes = 0;
+	fs_devices->chunk_alloc_policy = BTRFS_CHUNK_ALLOC_REGULAR;
 out:
 	return ret;
 }
@@ -1385,9 +1244,10 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 	return ret;
 }
 
-static void btrfs_release_disk_super(struct page *page)
+void btrfs_release_disk_super(struct btrfs_super_block *super)
 {
-	kunmap(page);
+	struct page *page = virt_to_page(super);
+
 	put_page(page);
 }
 
@@ -1415,17 +1275,17 @@ static int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
 	*page = read_cache_page_gfp(bdev->bd_inode->i_mapping,
 				   index, GFP_KERNEL);
 
-	if (IS_ERR_OR_NULL(*page))
+	if (IS_ERR(*page))
 		return 1;
 
-	p = kmap(*page);
+	p = page_address(*page);
 
 	/* align our pointer to the offset of the super block */
 	*disk_super = p + offset_in_page(bytenr);
 
 	if (btrfs_super_bytenr(*disk_super) != bytenr ||
 	    btrfs_super_magic(*disk_super) != BTRFS_MAGIC) {
-		btrfs_release_disk_super(*page);
+		btrfs_release_disk_super(p);
 		return 1;
 	}
 
@@ -1488,7 +1348,7 @@ struct btrfs_device *btrfs_scan_one_device(const char *path, fmode_t flags,
 			btrfs_free_stale_devices(path, device);
 	}
 
-	btrfs_release_disk_super(page);
+	btrfs_release_disk_super(disk_super);
 
 error_bdev_put:
 	blkdev_put(bdev, flags);
@@ -1521,6 +1381,59 @@ static bool contains_pending_extent(struct btrfs_device *device, u64 *start,
 	return false;
 }
 
+static u64 dev_extent_search_start(struct btrfs_device *device, u64 start)
+{
+	switch (device->fs_devices->chunk_alloc_policy) {
+	case BTRFS_CHUNK_ALLOC_REGULAR:
+		/*
+		 * We don't want to overwrite the superblock on the drive nor
+		 * any area used by the boot loader (grub for example), so we
+		 * make sure to start at an offset of at least 1MB.
+		 */
+		return max_t(u64, start, SZ_1M);
+	default:
+		BUG();
+	}
+}
+
+/**
+ * dev_extent_hole_check - check if specified hole is suitable for allocation
+ * @device:	the device which we have the hole
+ * @hole_start: starting position of the hole
+ * @hole_size:	the size of the hole
+ * @num_bytes:	the size of the free space that we need
+ *
+ * This function may modify @hole_start and @hole_end to reflect the suitable
+ * position for allocation. Returns 1 if hole position is updated, 0 otherwise.
+ */
+static bool dev_extent_hole_check(struct btrfs_device *device, u64 *hole_start,
+				  u64 *hole_size, u64 num_bytes)
+{
+	bool changed = false;
+	u64 hole_end = *hole_start + *hole_size;
+
+	/*
+	 * Check before we set max_hole_start, otherwise we could end up
+	 * sending back this offset anyway.
+	 */
+	if (contains_pending_extent(device, hole_start, *hole_size)) {
+		if (hole_end >= *hole_start)
+			*hole_size = hole_end - *hole_start;
+		else
+			*hole_size = 0;
+		changed = true;
+	}
+
+	switch (device->fs_devices->chunk_alloc_policy) {
+	case BTRFS_CHUNK_ALLOC_REGULAR:
+		/* No extra check */
+		break;
+	default:
+		BUG();
+	}
+
+	return changed;
+}
 
 /*
  * find_free_dev_extent_start - find free space in the specified device
@@ -1567,12 +1480,7 @@ static int find_free_dev_extent_start(struct btrfs_device *device,
 	int slot;
 	struct extent_buffer *l;
 
-	/*
-	 * We don't want to overwrite the superblock on the drive nor any area
-	 * used by the boot loader (grub for example), so we make sure to start
-	 * at an offset of at least 1MB.
-	 */
-	search_start = max_t(u64, search_start, SZ_1M);
+	search_start = dev_extent_search_start(device, search_start);
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -1630,18 +1538,8 @@ again:
 
 		if (key.offset > search_start) {
 			hole_size = key.offset - search_start;
-
-			/*
-			 * Have to check before we set max_hole_start, otherwise
-			 * we could end up sending back this offset anyway.
-			 */
-			if (contains_pending_extent(device, &search_start,
-						    hole_size)) {
-				if (key.offset >= search_start)
-					hole_size = key.offset - search_start;
-				else
-					hole_size = 0;
-			}
+			dev_extent_hole_check(device, &search_start, &hole_size,
+					      num_bytes);
 
 			if (hole_size > max_hole_size) {
 				max_hole_start = search_start;
@@ -1680,8 +1578,8 @@ next:
 	 */
 	if (search_end > search_start) {
 		hole_size = search_end - search_start;
-
-		if (contains_pending_extent(device, &search_start, hole_size)) {
+		if (dev_extent_hole_check(device, &search_start, &hole_size,
+					  num_bytes)) {
 			btrfs_release_path(path);
 			goto again;
 		}
@@ -2048,7 +1946,7 @@ static struct btrfs_device * btrfs_find_next_active_device(
  * where this function called, there should be always be another device (or
  * this_dev) which is active.
  */
-void btrfs_assign_next_active_device(struct btrfs_device *device,
+void __cold btrfs_assign_next_active_device(struct btrfs_device *device,
 				     struct btrfs_device *this_dev)
 {
 	struct btrfs_fs_info *fs_info = device->fs_info;
@@ -2087,6 +1985,46 @@ static u64 btrfs_num_devices(struct btrfs_fs_info *fs_info)
 	return num_devices;
 }
 
+static void btrfs_scratch_superblocks(struct btrfs_fs_info *fs_info,
+				      struct block_device *bdev,
+				      const char *device_path)
+{
+	struct btrfs_super_block *disk_super;
+	int copy_num;
+
+	if (!bdev)
+		return;
+
+	for (copy_num = 0; copy_num < BTRFS_SUPER_MIRROR_MAX; copy_num++) {
+		struct page *page;
+		int ret;
+
+		disk_super = btrfs_read_dev_one_super(bdev, copy_num);
+		if (IS_ERR(disk_super))
+			continue;
+
+		memset(&disk_super->magic, 0, sizeof(disk_super->magic));
+
+		page = virt_to_page(disk_super);
+		set_page_dirty(page);
+		lock_page(page);
+		/* write_on_page() unlocks the page */
+		ret = write_one_page(page);
+		if (ret)
+			btrfs_warn(fs_info,
+				"error clearing superblock number %d (%d)",
+				copy_num, ret);
+		btrfs_release_disk_super(disk_super);
+
+	}
+
+	/* Notify udev that device has changed */
+	btrfs_kobject_uevent(bdev, KOBJ_CHANGE);
+
+	/* Update ctime/mtime for device path for libblkid */
+	update_dev_time(device_path);
+}
+
 int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 		u64 devid)
 {
@@ -2192,7 +2130,7 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	if (device->bdev) {
 		cur_devices->open_devices--;
 		/* remove sysfs entry */
-		btrfs_sysfs_rm_device_link(fs_devices, device);
+		btrfs_sysfs_remove_devices_dir(fs_devices, device);
 	}
 
 	num_devices = btrfs_super_num_devices(fs_info->super_copy) - 1;
@@ -2205,7 +2143,8 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path,
 	 * supers and free the device.
 	 */
 	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state))
-		btrfs_scratch_superblocks(device->bdev, device->name->str);
+		btrfs_scratch_superblocks(fs_info, device->bdev,
+					  device->name->str);
 
 	btrfs_close_bdev(device);
 	synchronize_rcu();
@@ -2273,7 +2212,8 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_device *srcdev)
 
 	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &srcdev->dev_state)) {
 		/* zero out the old super if it is writable */
-		btrfs_scratch_superblocks(srcdev->bdev, srcdev->name->str);
+		btrfs_scratch_superblocks(fs_info, srcdev->bdev,
+					  srcdev->name->str);
 	}
 
 	btrfs_close_bdev(srcdev);
@@ -2310,10 +2250,9 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev)
 {
 	struct btrfs_fs_devices *fs_devices = tgtdev->fs_info->fs_devices;
 
-	WARN_ON(!tgtdev);
 	mutex_lock(&fs_devices->device_list_mutex);
 
-	btrfs_sysfs_rm_device_link(fs_devices, tgtdev);
+	btrfs_sysfs_remove_devices_dir(fs_devices, tgtdev);
 
 	if (tgtdev->bdev)
 		fs_devices->open_devices--;
@@ -2333,7 +2272,8 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev)
 	 * is already out of device list, so we don't have to hold
 	 * the device_list_mutex lock.
 	 */
-	btrfs_scratch_superblocks(tgtdev->bdev, tgtdev->name->str);
+	btrfs_scratch_superblocks(tgtdev->fs_info, tgtdev->bdev,
+				  tgtdev->name->str);
 
 	btrfs_close_bdev(tgtdev);
 	synchronize_rcu();
@@ -2348,14 +2288,13 @@ static struct btrfs_device *btrfs_find_device_by_path(
 	u64 devid;
 	u8 *dev_uuid;
 	struct block_device *bdev;
-	struct buffer_head *bh;
 	struct btrfs_device *device;
 
 	ret = btrfs_get_bdev_and_sb(device_path, FMODE_READ,
-				    fs_info->bdev_holder, 0, &bdev, &bh);
+				    fs_info->bdev_holder, 0, &bdev, &disk_super);
 	if (ret)
 		return ERR_PTR(ret);
-	disk_super = (struct btrfs_super_block *)bh->b_data;
+
 	devid = btrfs_stack_device_id(&disk_super->dev_item);
 	dev_uuid = disk_super->dev_item.uuid;
 	if (btrfs_fs_incompat(fs_info, METADATA_UUID))
@@ -2365,7 +2304,7 @@ static struct btrfs_device *btrfs_find_device_by_path(
 		device = btrfs_find_device(fs_info->fs_devices, devid, dev_uuid,
 					   disk_super->fsid, true);
 
-	brelse(bh);
+	btrfs_release_disk_super(disk_super);
 	if (!device)
 		device = ERR_PTR(-ENOENT);
 	blkdev_put(bdev, FMODE_READ);
@@ -2450,11 +2389,11 @@ static int btrfs_prepare_sprout(struct btrfs_fs_info *fs_info)
 	list_splice_init(&fs_devices->alloc_list, &seed_devices->alloc_list);
 	mutex_unlock(&fs_info->chunk_mutex);
 
-	fs_devices->seeding = 0;
+	fs_devices->seeding = false;
 	fs_devices->num_devices = 0;
 	fs_devices->open_devices = 0;
 	fs_devices->missing_devices = 0;
-	fs_devices->rotating = 0;
+	fs_devices->rotating = false;
 	fs_devices->seed = seed_devices;
 
 	generate_random_uuid(fs_devices->fsid);
@@ -2649,7 +2588,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	atomic64_add(device->total_bytes, &fs_info->free_chunk_space);
 
 	if (!blk_queue_nonrot(q))
-		fs_devices->rotating = 1;
+		fs_devices->rotating = true;
 
 	orig_super_total_bytes = btrfs_super_total_bytes(fs_info->super_copy);
 	btrfs_set_super_total_bytes(fs_info->super_copy,
@@ -2661,7 +2600,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 				    orig_super_num_devices + 1);
 
 	/* add sysfs device entry */
-	btrfs_sysfs_add_device_link(fs_devices, device);
+	btrfs_sysfs_add_devices_dir(fs_devices, device);
 
 	/*
 	 * we've got more storage, clear any full flags on the space
@@ -2729,7 +2668,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path
 	return ret;
 
 error_sysfs:
-	btrfs_sysfs_rm_device_link(fs_devices, device);
+	btrfs_sysfs_remove_devices_dir(fs_devices, device);
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
 	mutex_lock(&fs_info->chunk_mutex);
 	list_del_rcu(&device->dev_list);
@@ -3055,6 +2994,7 @@ static int btrfs_relocate_chunk(struct btrfs_fs_info *fs_info, u64 chunk_offset)
 {
 	struct btrfs_root *root = fs_info->chunk_root;
 	struct btrfs_trans_handle *trans;
+	struct btrfs_block_group *block_group;
 	int ret;
 
 	/*
@@ -3078,6 +3018,12 @@ static int btrfs_relocate_chunk(struct btrfs_fs_info *fs_info, u64 chunk_offset)
 	if (ret)
 		return ret;
 
+	block_group = btrfs_lookup_block_group(fs_info, chunk_offset);
+	if (!block_group)
+		return -ENOENT;
+	btrfs_discard_cancel_work(&fs_info->discard_ctl, block_group);
+	btrfs_put_block_group(block_group);
+
 	trans = btrfs_start_trans_remove_block_group(root->fs_info,
 						     chunk_offset);
 	if (IS_ERR(trans)) {
@@ -3177,7 +3123,7 @@ error:
 static int btrfs_may_alloc_data_chunk(struct btrfs_fs_info *fs_info,
 				      u64 chunk_offset)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	u64 bytes_used;
 	u64 chunk_type;
 
@@ -3186,27 +3132,28 @@ static int btrfs_may_alloc_data_chunk(struct btrfs_fs_info *fs_info,
 	chunk_type = cache->flags;
 	btrfs_put_block_group(cache);
 
-	if (chunk_type & BTRFS_BLOCK_GROUP_DATA) {
-		spin_lock(&fs_info->data_sinfo->lock);
-		bytes_used = fs_info->data_sinfo->bytes_used;
-		spin_unlock(&fs_info->data_sinfo->lock);
+	if (!(chunk_type & BTRFS_BLOCK_GROUP_DATA))
+		return 0;
 
-		if (!bytes_used) {
-			struct btrfs_trans_handle *trans;
-			int ret;
+	spin_lock(&fs_info->data_sinfo->lock);
+	bytes_used = fs_info->data_sinfo->bytes_used;
+	spin_unlock(&fs_info->data_sinfo->lock);
 
-			trans =	btrfs_join_transaction(fs_info->tree_root);
-			if (IS_ERR(trans))
-				return PTR_ERR(trans);
+	if (!bytes_used) {
+		struct btrfs_trans_handle *trans;
+		int ret;
 
-			ret = btrfs_force_chunk_alloc(trans,
-						      BTRFS_BLOCK_GROUP_DATA);
-			btrfs_end_transaction(trans);
-			if (ret < 0)
-				return ret;
-			return 1;
-		}
+		trans =	btrfs_join_transaction(fs_info->tree_root);
+		if (IS_ERR(trans))
+			return PTR_ERR(trans);
+
+		ret = btrfs_force_chunk_alloc(trans, BTRFS_BLOCK_GROUP_DATA);
+		btrfs_end_transaction(trans);
+		if (ret < 0)
+			return ret;
+		return 1;
 	}
+
 	return 0;
 }
 
@@ -3385,28 +3332,28 @@ static int chunk_profiles_filter(u64 chunk_type,
 static int chunk_usage_range_filter(struct btrfs_fs_info *fs_info, u64 chunk_offset,
 			      struct btrfs_balance_args *bargs)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	u64 chunk_used;
 	u64 user_thresh_min;
 	u64 user_thresh_max;
 	int ret = 1;
 
 	cache = btrfs_lookup_block_group(fs_info, chunk_offset);
-	chunk_used = btrfs_block_group_used(&cache->item);
+	chunk_used = cache->used;
 
 	if (bargs->usage_min == 0)
 		user_thresh_min = 0;
 	else
-		user_thresh_min = div_factor_fine(cache->key.offset,
-					bargs->usage_min);
+		user_thresh_min = div_factor_fine(cache->length,
+						  bargs->usage_min);
 
 	if (bargs->usage_max == 0)
 		user_thresh_max = 1;
 	else if (bargs->usage_max > 100)
-		user_thresh_max = cache->key.offset;
+		user_thresh_max = cache->length;
 	else
-		user_thresh_max = div_factor_fine(cache->key.offset,
-					bargs->usage_max);
+		user_thresh_max = div_factor_fine(cache->length,
+						  bargs->usage_max);
 
 	if (user_thresh_min <= chunk_used && chunk_used < user_thresh_max)
 		ret = 0;
@@ -3418,20 +3365,19 @@ static int chunk_usage_range_filter(struct btrfs_fs_info *fs_info, u64 chunk_off
 static int chunk_usage_filter(struct btrfs_fs_info *fs_info,
 		u64 chunk_offset, struct btrfs_balance_args *bargs)
 {
-	struct btrfs_block_group_cache *cache;
+	struct btrfs_block_group *cache;
 	u64 chunk_used, user_thresh;
 	int ret = 1;
 
 	cache = btrfs_lookup_block_group(fs_info, chunk_offset);
-	chunk_used = btrfs_block_group_used(&cache->item);
+	chunk_used = cache->used;
 
 	if (bargs->usage_min == 0)
 		user_thresh = 1;
 	else if (bargs->usage > 100)
-		user_thresh = cache->key.offset;
+		user_thresh = cache->length;
 	else
-		user_thresh = div_factor_fine(cache->key.offset,
-					      bargs->usage);
+		user_thresh = div_factor_fine(cache->length, bargs->usage);
 
 	if (chunk_used < user_thresh)
 		ret = 0;
@@ -3844,12 +3790,7 @@ static int alloc_profile_is_valid(u64 flags, int extended)
 	if (flags == 0)
 		return !extended; /* "0" is valid for usual profiles */
 
-	/* true if exactly one bit set */
-	/*
-	 * Don't use is_power_of_2(unsigned long) because it won't work
-	 * for the single profile (1ULL << 48) on 32-bit CPUs.
-	 */
-	return flags != 0 && (flags & (flags - 1)) == 0;
+	return has_single_bit_set(flags);
 }
 
 static inline int balance_need_close(struct btrfs_fs_info *fs_info)
@@ -3860,13 +3801,25 @@ static inline int balance_need_close(struct btrfs_fs_info *fs_info)
 		 atomic_read(&fs_info->balance_cancel_req) == 0);
 }
 
-/* Non-zero return value signifies invalidity */
-static inline int validate_convert_profile(struct btrfs_balance_args *bctl_arg,
-		u64 allowed)
+/*
+ * Validate target profile against allowed profiles and return true if it's OK.
+ * Otherwise print the error message and return false.
+ */
+static inline int validate_convert_profile(struct btrfs_fs_info *fs_info,
+		const struct btrfs_balance_args *bargs,
+		u64 allowed, const char *type)
 {
-	return ((bctl_arg->flags & BTRFS_BALANCE_ARGS_CONVERT) &&
-		(!alloc_profile_is_valid(bctl_arg->target, 1) ||
-		 (bctl_arg->target & ~allowed)));
+	if (!(bargs->flags & BTRFS_BALANCE_ARGS_CONVERT))
+		return true;
+
+	/* Profile is valid and does not have bits outside of the allowed set */
+	if (alloc_profile_is_valid(bargs->target, 1) &&
+	    (bargs->target & ~allowed) == 0)
+		return true;
+
+	btrfs_err(fs_info, "balance: invalid convert %s profile %s",
+			type, btrfs_bg_type_to_raid_name(bargs->target));
+	return false;
 }
 
 /*
@@ -4036,12 +3989,12 @@ int btrfs_balance(struct btrfs_fs_info *fs_info,
 	int ret;
 	u64 num_devices;
 	unsigned seq;
-	bool reducing_integrity;
+	bool reducing_redundancy;
 	int i;
 
 	if (btrfs_fs_closing(fs_info) ||
 	    atomic_read(&fs_info->balance_pause_req) ||
-	    atomic_read(&fs_info->balance_cancel_req)) {
+	    btrfs_should_cancel_balance(fs_info)) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -4066,7 +4019,11 @@ int btrfs_balance(struct btrfs_fs_info *fs_info,
 		}
 	}
 
-	num_devices = btrfs_num_devices(fs_info);
+	/*
+	 * rw_devices will not change at the moment, device add/delete/replace
+	 * are excluded by EXCL_OP
+	 */
+	num_devices = fs_info->fs_devices->rw_devices;
 
 	/*
 	 * SINGLE profile on-disk has no profile bit, but in-memory we have a
@@ -4078,24 +4035,9 @@ int btrfs_balance(struct btrfs_fs_info *fs_info,
 		if (num_devices >= btrfs_raid_array[i].devs_min)
 			allowed |= btrfs_raid_array[i].bg_flag;
 
-	if (validate_convert_profile(&bctl->data, allowed)) {
-		btrfs_err(fs_info,
-			  "balance: invalid convert data profile %s",
-			  btrfs_bg_type_to_raid_name(bctl->data.target));
-		ret = -EINVAL;
-		goto out;
-	}
-	if (validate_convert_profile(&bctl->meta, allowed)) {
-		btrfs_err(fs_info,
-			  "balance: invalid convert metadata profile %s",
-			  btrfs_bg_type_to_raid_name(bctl->meta.target));
-		ret = -EINVAL;
-		goto out;
-	}
-	if (validate_convert_profile(&bctl->sys, allowed)) {
-		btrfs_err(fs_info,
-			  "balance: invalid convert system profile %s",
-			  btrfs_bg_type_to_raid_name(bctl->sys.target));
+	if (!validate_convert_profile(fs_info, &bctl->data, allowed, "data") ||
+	    !validate_convert_profile(fs_info, &bctl->meta, allowed, "metadata") ||
+	    !validate_convert_profile(fs_info, &bctl->sys,  allowed, "system")) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -4119,9 +4061,9 @@ int btrfs_balance(struct btrfs_fs_info *fs_info,
 		    ((bctl->meta.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
 		     (fs_info->avail_metadata_alloc_bits & allowed) &&
 		     !(bctl->meta.target & allowed)))
-			reducing_integrity = true;
+			reducing_redundancy = true;
 		else
-			reducing_integrity = false;
+			reducing_redundancy = false;
 
 		/* if we're not converting, the target field is uninitialized */
 		meta_target = (bctl->meta.flags & BTRFS_BALANCE_ARGS_CONVERT) ?
@@ -4130,13 +4072,13 @@ int btrfs_balance(struct btrfs_fs_info *fs_info,
 			bctl->data.target : fs_info->avail_data_alloc_bits;
 	} while (read_seqretry(&fs_info->profiles_lock, seq));
 
-	if (reducing_integrity) {
+	if (reducing_redundancy) {
 		if (bctl->flags & BTRFS_BALANCE_FORCE) {
 			btrfs_info(fs_info,
-				   "balance: force reducing metadata integrity");
+			   "balance: force reducing metadata redundancy");
 		} else {
 			btrfs_err(fs_info,
-	  "balance: reduces metadata integrity, use --force if you want this");
+	"balance: reduces metadata redundancy, use --force if you want this");
 			ret = -EINVAL;
 			goto out;
 		}
@@ -4407,7 +4349,7 @@ int btrfs_cancel_balance(struct btrfs_fs_info *fs_info)
 	return 0;
 }
 
-static int btrfs_uuid_scan_kthread(void *data)
+int btrfs_uuid_scan_kthread(void *data)
 {
 	struct btrfs_fs_info *fs_info = data;
 	struct btrfs_root *root = fs_info->tree_root;
@@ -4419,6 +4361,7 @@ static int btrfs_uuid_scan_kthread(void *data)
 	struct btrfs_root_item root_item;
 	u32 item_size;
 	struct btrfs_trans_handle *trans = NULL;
+	bool closing = false;
 
 	path = btrfs_alloc_path();
 	if (!path) {
@@ -4431,6 +4374,10 @@ static int btrfs_uuid_scan_kthread(void *data)
 	key.offset = 0;
 
 	while (1) {
+		if (btrfs_fs_closing(fs_info)) {
+			closing = true;
+			break;
+		}
 		ret = btrfs_search_forward(root, &key, path,
 				BTRFS_OLDEST_GENERATION);
 		if (ret) {
@@ -4530,76 +4477,12 @@ out:
 		btrfs_end_transaction(trans);
 	if (ret)
 		btrfs_warn(fs_info, "btrfs_uuid_scan_kthread failed %d", ret);
-	else
+	else if (!closing)
 		set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags);
 	up(&fs_info->uuid_tree_rescan_sem);
 	return 0;
 }
 
-/*
- * Callback for btrfs_uuid_tree_iterate().
- * returns:
- * 0	check succeeded, the entry is not outdated.
- * < 0	if an error occurred.
- * > 0	if the check failed, which means the caller shall remove the entry.
- */
-static int btrfs_check_uuid_tree_entry(struct btrfs_fs_info *fs_info,
-				       u8 *uuid, u8 type, u64 subid)
-{
-	struct btrfs_key key;
-	int ret = 0;
-	struct btrfs_root *subvol_root;
-
-	if (type != BTRFS_UUID_KEY_SUBVOL &&
-	    type != BTRFS_UUID_KEY_RECEIVED_SUBVOL)
-		goto out;
-
-	key.objectid = subid;
-	key.type = BTRFS_ROOT_ITEM_KEY;
-	key.offset = (u64)-1;
-	subvol_root = btrfs_read_fs_root_no_name(fs_info, &key);
-	if (IS_ERR(subvol_root)) {
-		ret = PTR_ERR(subvol_root);
-		if (ret == -ENOENT)
-			ret = 1;
-		goto out;
-	}
-
-	switch (type) {
-	case BTRFS_UUID_KEY_SUBVOL:
-		if (memcmp(uuid, subvol_root->root_item.uuid, BTRFS_UUID_SIZE))
-			ret = 1;
-		break;
-	case BTRFS_UUID_KEY_RECEIVED_SUBVOL:
-		if (memcmp(uuid, subvol_root->root_item.received_uuid,
-			   BTRFS_UUID_SIZE))
-			ret = 1;
-		break;
-	}
-
-out:
-	return ret;
-}
-
-static int btrfs_uuid_rescan_kthread(void *data)
-{
-	struct btrfs_fs_info *fs_info = (struct btrfs_fs_info *)data;
-	int ret;
-
-	/*
-	 * 1st step is to iterate through the existing UUID tree and
-	 * to delete all entries that contain outdated data.
-	 * 2nd step is to add all missing entries to the UUID tree.
-	 */
-	ret = btrfs_uuid_tree_iterate(fs_info, btrfs_check_uuid_tree_entry);
-	if (ret < 0) {
-		btrfs_warn(fs_info, "iterating uuid_tree failed %d", ret);
-		up(&fs_info->uuid_tree_rescan_sem);
-		return ret;
-	}
-	return btrfs_uuid_scan_kthread(data);
-}
-
 int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_trans_handle *trans;
@@ -4642,22 +4525,6 @@ int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info)
 	return 0;
 }
 
-int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info)
-{
-	struct task_struct *task;
-
-	down(&fs_info->uuid_tree_rescan_sem);
-	task = kthread_run(btrfs_uuid_rescan_kthread, fs_info, "btrfs-uuid");
-	if (IS_ERR(task)) {
-		/* fs_info->update_uuid_tree_gen remains 0 in all error case */
-		btrfs_warn(fs_info, "failed to start uuid_rescan task");
-		up(&fs_info->uuid_tree_rescan_sem);
-		return PTR_ERR(task);
-	}
-
-	return 0;
-}
-
 /*
  * shrinking a device means finding all of the device extents past
  * the new size, and then following the back refs to the chunks.
@@ -4902,96 +4769,119 @@ static void check_raid56_incompat_flag(struct btrfs_fs_info *info, u64 type)
 	btrfs_set_fs_incompat(info, RAID56);
 }
 
-static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
-			       u64 start, u64 type)
+static void check_raid1c34_incompat_flag(struct btrfs_fs_info *info, u64 type)
 {
-	struct btrfs_fs_info *info = trans->fs_info;
-	struct btrfs_fs_devices *fs_devices = info->fs_devices;
-	struct btrfs_device *device;
-	struct map_lookup *map = NULL;
-	struct extent_map_tree *em_tree;
-	struct extent_map *em;
-	struct btrfs_device_info *devices_info = NULL;
-	u64 total_avail;
-	int num_stripes;	/* total number of stripes to allocate */
-	int data_stripes;	/* number of stripes that count for
-				   block group size */
-	int sub_stripes;	/* sub_stripes info for map */
-	int dev_stripes;	/* stripes per dev */
-	int devs_max;		/* max devs to use */
-	int devs_min;		/* min devs needed */
-	int devs_increment;	/* ndevs has to be a multiple of this */
-	int ncopies;		/* how many copies to data has */
-	int nparity;		/* number of stripes worth of bytes to
-				   store parity information */
-	int ret;
+	if (!(type & (BTRFS_BLOCK_GROUP_RAID1C3 | BTRFS_BLOCK_GROUP_RAID1C4)))
+		return;
+
+	btrfs_set_fs_incompat(info, RAID1C34);
+}
+
+/*
+ * Structure used internally for __btrfs_alloc_chunk() function.
+ * Wraps needed parameters.
+ */
+struct alloc_chunk_ctl {
+	u64 start;
+	u64 type;
+	/* Total number of stripes to allocate */
+	int num_stripes;
+	/* sub_stripes info for map */
+	int sub_stripes;
+	/* Stripes per device */
+	int dev_stripes;
+	/* Maximum number of devices to use */
+	int devs_max;
+	/* Minimum number of devices to use */
+	int devs_min;
+	/* ndevs has to be a multiple of this */
+	int devs_increment;
+	/* Number of copies */
+	int ncopies;
+	/* Number of stripes worth of bytes to store parity information */
+	int nparity;
 	u64 max_stripe_size;
 	u64 max_chunk_size;
+	u64 dev_extent_min;
 	u64 stripe_size;
 	u64 chunk_size;
 	int ndevs;
-	int i;
-	int j;
-	int index;
-
-	BUG_ON(!alloc_profile_is_valid(type, 0));
-
-	if (list_empty(&fs_devices->alloc_list)) {
-		if (btrfs_test_opt(info, ENOSPC_DEBUG))
-			btrfs_debug(info, "%s: no writable device", __func__);
-		return -ENOSPC;
-	}
-
-	index = btrfs_bg_flags_to_raid_index(type);
+};
 
-	sub_stripes = btrfs_raid_array[index].sub_stripes;
-	dev_stripes = btrfs_raid_array[index].dev_stripes;
-	devs_max = btrfs_raid_array[index].devs_max;
-	if (!devs_max)
-		devs_max = BTRFS_MAX_DEVS(info);
-	devs_min = btrfs_raid_array[index].devs_min;
-	devs_increment = btrfs_raid_array[index].devs_increment;
-	ncopies = btrfs_raid_array[index].ncopies;
-	nparity = btrfs_raid_array[index].nparity;
+static void init_alloc_chunk_ctl_policy_regular(
+				struct btrfs_fs_devices *fs_devices,
+				struct alloc_chunk_ctl *ctl)
+{
+	u64 type = ctl->type;
 
 	if (type & BTRFS_BLOCK_GROUP_DATA) {
-		max_stripe_size = SZ_1G;
-		max_chunk_size = BTRFS_MAX_DATA_CHUNK_SIZE;
+		ctl->max_stripe_size = SZ_1G;
+		ctl->max_chunk_size = BTRFS_MAX_DATA_CHUNK_SIZE;
 	} else if (type & BTRFS_BLOCK_GROUP_METADATA) {
-		/* for larger filesystems, use larger metadata chunks */
+		/* For larger filesystems, use larger metadata chunks */
 		if (fs_devices->total_rw_bytes > 50ULL * SZ_1G)
-			max_stripe_size = SZ_1G;
+			ctl->max_stripe_size = SZ_1G;
 		else
-			max_stripe_size = SZ_256M;
-		max_chunk_size = max_stripe_size;
+			ctl->max_stripe_size = SZ_256M;
+		ctl->max_chunk_size = ctl->max_stripe_size;
 	} else if (type & BTRFS_BLOCK_GROUP_SYSTEM) {
-		max_stripe_size = SZ_32M;
-		max_chunk_size = 2 * max_stripe_size;
-		devs_max = min_t(int, devs_max, BTRFS_MAX_DEVS_SYS_CHUNK);
+		ctl->max_stripe_size = SZ_32M;
+		ctl->max_chunk_size = 2 * ctl->max_stripe_size;
+		ctl->devs_max = min_t(int, ctl->devs_max,
+				      BTRFS_MAX_DEVS_SYS_CHUNK);
 	} else {
-		btrfs_err(info, "invalid chunk type 0x%llx requested",
-		       type);
 		BUG();
 	}
 
 	/* We don't want a chunk larger than 10% of writable space */
-	max_chunk_size = min(div_factor(fs_devices->total_rw_bytes, 1),
-			     max_chunk_size);
+	ctl->max_chunk_size = min(div_factor(fs_devices->total_rw_bytes, 1),
+				  ctl->max_chunk_size);
+	ctl->dev_extent_min = BTRFS_STRIPE_LEN * ctl->dev_stripes;
+}
+
+static void init_alloc_chunk_ctl(struct btrfs_fs_devices *fs_devices,
+				 struct alloc_chunk_ctl *ctl)
+{
+	int index = btrfs_bg_flags_to_raid_index(ctl->type);
+
+	ctl->sub_stripes = btrfs_raid_array[index].sub_stripes;
+	ctl->dev_stripes = btrfs_raid_array[index].dev_stripes;
+	ctl->devs_max = btrfs_raid_array[index].devs_max;
+	if (!ctl->devs_max)
+		ctl->devs_max = BTRFS_MAX_DEVS(fs_devices->fs_info);
+	ctl->devs_min = btrfs_raid_array[index].devs_min;
+	ctl->devs_increment = btrfs_raid_array[index].devs_increment;
+	ctl->ncopies = btrfs_raid_array[index].ncopies;
+	ctl->nparity = btrfs_raid_array[index].nparity;
+	ctl->ndevs = 0;
+
+	switch (fs_devices->chunk_alloc_policy) {
+	case BTRFS_CHUNK_ALLOC_REGULAR:
+		init_alloc_chunk_ctl_policy_regular(fs_devices, ctl);
+		break;
+	default:
+		BUG();
+	}
+}
 
-	devices_info = kcalloc(fs_devices->rw_devices, sizeof(*devices_info),
-			       GFP_NOFS);
-	if (!devices_info)
-		return -ENOMEM;
+static int gather_device_info(struct btrfs_fs_devices *fs_devices,
+			      struct alloc_chunk_ctl *ctl,
+			      struct btrfs_device_info *devices_info)
+{
+	struct btrfs_fs_info *info = fs_devices->fs_info;
+	struct btrfs_device *device;
+	u64 total_avail;
+	u64 dev_extent_want = ctl->max_stripe_size * ctl->dev_stripes;
+	int ret;
+	int ndevs = 0;
+	u64 max_avail;
+	u64 dev_offset;
 
 	/*
 	 * in the first pass through the devices list, we gather information
 	 * about the available holes on each device.
 	 */
-	ndevs = 0;
 	list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) {
-		u64 max_avail;
-		u64 dev_offset;
-
 		if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) {
 			WARN(1, KERN_ERR
 			       "BTRFS: read-only device in alloc_list\n");
@@ -5009,24 +4899,23 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 			total_avail = 0;
 
 		/* If there is no space on this device, skip it. */
-		if (total_avail == 0)
+		if (total_avail < ctl->dev_extent_min)
 			continue;
 
-		ret = find_free_dev_extent(device,
-					   max_stripe_size * dev_stripes,
-					   &dev_offset, &max_avail);
+		ret = find_free_dev_extent(device, dev_extent_want, &dev_offset,
+					   &max_avail);
 		if (ret && ret != -ENOSPC)
-			goto error;
+			return ret;
 
 		if (ret == 0)
-			max_avail = max_stripe_size * dev_stripes;
+			max_avail = dev_extent_want;
 
-		if (max_avail < BTRFS_STRIPE_LEN * dev_stripes) {
+		if (max_avail < ctl->dev_extent_min) {
 			if (btrfs_test_opt(info, ENOSPC_DEBUG))
 				btrfs_debug(info,
-			"%s: devid %llu has no free space, have=%llu want=%u",
+			"%s: devid %llu has no free space, have=%llu want=%llu",
 					    __func__, device->devid, max_avail,
-					    BTRFS_STRIPE_LEN * dev_stripes);
+					    ctl->dev_extent_min);
 			continue;
 		}
 
@@ -5041,6 +4930,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 		devices_info[ndevs].dev = device;
 		++ndevs;
 	}
+	ctl->ndevs = ndevs;
 
 	/*
 	 * now sort the devices by hole size / available space
@@ -5048,20 +4938,14 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	sort(devices_info, ndevs, sizeof(struct btrfs_device_info),
 	     btrfs_cmp_device_info, NULL);
 
-	/* round down to number of usable stripes */
-	ndevs = round_down(ndevs, devs_increment);
-
-	if (ndevs < devs_min) {
-		ret = -ENOSPC;
-		if (btrfs_test_opt(info, ENOSPC_DEBUG)) {
-			btrfs_debug(info,
-	"%s: not enough devices with free space: have=%d minimum required=%d",
-				    __func__, ndevs, devs_min);
-		}
-		goto error;
-	}
+	return 0;
+}
 
-	ndevs = min(ndevs, devs_max);
+static int decide_stripe_size_regular(struct alloc_chunk_ctl *ctl,
+				      struct btrfs_device_info *devices_info)
+{
+	/* Number of stripes that count for block group size */
+	int data_stripes;
 
 	/*
 	 * The primary goal is to maximize the number of stripes, so use as
@@ -5070,73 +4954,116 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	 * The DUP profile stores more than one stripe per device, the
 	 * max_avail is the total size so we have to adjust.
 	 */
-	stripe_size = div_u64(devices_info[ndevs - 1].max_avail, dev_stripes);
-	num_stripes = ndevs * dev_stripes;
+	ctl->stripe_size = div_u64(devices_info[ctl->ndevs - 1].max_avail,
+				   ctl->dev_stripes);
+	ctl->num_stripes = ctl->ndevs * ctl->dev_stripes;
 
-	/*
-	 * this will have to be fixed for RAID1 and RAID10 over
-	 * more drives
-	 */
-	data_stripes = (num_stripes - nparity) / ncopies;
+	/* This will have to be fixed for RAID1 and RAID10 over more drives */
+	data_stripes = (ctl->num_stripes - ctl->nparity) / ctl->ncopies;
 
 	/*
-	 * Use the number of data stripes to figure out how big this chunk
-	 * is really going to be in terms of logical address space,
-	 * and compare that answer with the max chunk size. If it's higher,
-	 * we try to reduce stripe_size.
+	 * Use the number of data stripes to figure out how big this chunk is
+	 * really going to be in terms of logical address space, and compare
+	 * that answer with the max chunk size. If it's higher, we try to
+	 * reduce stripe_size.
 	 */
-	if (stripe_size * data_stripes > max_chunk_size) {
+	if (ctl->stripe_size * data_stripes > ctl->max_chunk_size) {
 		/*
 		 * Reduce stripe_size, round it up to a 16MB boundary again and
 		 * then use it, unless it ends up being even bigger than the
 		 * previous value we had already.
 		 */
-		stripe_size = min(round_up(div_u64(max_chunk_size,
-						   data_stripes), SZ_16M),
-				  stripe_size);
+		ctl->stripe_size = min(round_up(div_u64(ctl->max_chunk_size,
+							data_stripes), SZ_16M),
+				       ctl->stripe_size);
 	}
 
-	/* align to BTRFS_STRIPE_LEN */
-	stripe_size = round_down(stripe_size, BTRFS_STRIPE_LEN);
+	/* Align to BTRFS_STRIPE_LEN */
+	ctl->stripe_size = round_down(ctl->stripe_size, BTRFS_STRIPE_LEN);
+	ctl->chunk_size = ctl->stripe_size * data_stripes;
 
-	map = kmalloc(map_lookup_size(num_stripes), GFP_NOFS);
-	if (!map) {
-		ret = -ENOMEM;
-		goto error;
+	return 0;
+}
+
+static int decide_stripe_size(struct btrfs_fs_devices *fs_devices,
+			      struct alloc_chunk_ctl *ctl,
+			      struct btrfs_device_info *devices_info)
+{
+	struct btrfs_fs_info *info = fs_devices->fs_info;
+
+	/*
+	 * Round down to number of usable stripes, devs_increment can be any
+	 * number so we can't use round_down() that requires power of 2, while
+	 * rounddown is safe.
+	 */
+	ctl->ndevs = rounddown(ctl->ndevs, ctl->devs_increment);
+
+	if (ctl->ndevs < ctl->devs_min) {
+		if (btrfs_test_opt(info, ENOSPC_DEBUG)) {
+			btrfs_debug(info,
+	"%s: not enough devices with free space: have=%d minimum required=%d",
+				    __func__, ctl->ndevs, ctl->devs_min);
+		}
+		return -ENOSPC;
 	}
-	map->num_stripes = num_stripes;
 
-	for (i = 0; i < ndevs; ++i) {
-		for (j = 0; j < dev_stripes; ++j) {
-			int s = i * dev_stripes + j;
+	ctl->ndevs = min(ctl->ndevs, ctl->devs_max);
+
+	switch (fs_devices->chunk_alloc_policy) {
+	case BTRFS_CHUNK_ALLOC_REGULAR:
+		return decide_stripe_size_regular(ctl, devices_info);
+	default:
+		BUG();
+	}
+}
+
+static int create_chunk(struct btrfs_trans_handle *trans,
+			struct alloc_chunk_ctl *ctl,
+			struct btrfs_device_info *devices_info)
+{
+	struct btrfs_fs_info *info = trans->fs_info;
+	struct map_lookup *map = NULL;
+	struct extent_map_tree *em_tree;
+	struct extent_map *em;
+	u64 start = ctl->start;
+	u64 type = ctl->type;
+	int ret;
+	int i;
+	int j;
+
+	map = kmalloc(map_lookup_size(ctl->num_stripes), GFP_NOFS);
+	if (!map)
+		return -ENOMEM;
+	map->num_stripes = ctl->num_stripes;
+
+	for (i = 0; i < ctl->ndevs; ++i) {
+		for (j = 0; j < ctl->dev_stripes; ++j) {
+			int s = i * ctl->dev_stripes + j;
 			map->stripes[s].dev = devices_info[i].dev;
 			map->stripes[s].physical = devices_info[i].dev_offset +
-						   j * stripe_size;
+						   j * ctl->stripe_size;
 		}
 	}
 	map->stripe_len = BTRFS_STRIPE_LEN;
 	map->io_align = BTRFS_STRIPE_LEN;
 	map->io_width = BTRFS_STRIPE_LEN;
 	map->type = type;
-	map->sub_stripes = sub_stripes;
-
-	chunk_size = stripe_size * data_stripes;
+	map->sub_stripes = ctl->sub_stripes;
 
-	trace_btrfs_chunk_alloc(info, map, start, chunk_size);
+	trace_btrfs_chunk_alloc(info, map, start, ctl->chunk_size);
 
 	em = alloc_extent_map();
 	if (!em) {
 		kfree(map);
-		ret = -ENOMEM;
-		goto error;
+		return -ENOMEM;
 	}
 	set_bit(EXTENT_FLAG_FS_MAPPING, &em->flags);
 	em->map_lookup = map;
 	em->start = start;
-	em->len = chunk_size;
+	em->len = ctl->chunk_size;
 	em->block_start = 0;
 	em->block_len = em->len;
-	em->orig_block_len = stripe_size;
+	em->orig_block_len = ctl->stripe_size;
 
 	em_tree = &info->mapping_tree;
 	write_lock(&em_tree->lock);
@@ -5144,29 +5071,31 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 	if (ret) {
 		write_unlock(&em_tree->lock);
 		free_extent_map(em);
-		goto error;
+		return ret;
 	}
 	write_unlock(&em_tree->lock);
 
-	ret = btrfs_make_block_group(trans, 0, type, start, chunk_size);
+	ret = btrfs_make_block_group(trans, 0, type, start, ctl->chunk_size);
 	if (ret)
 		goto error_del_extent;
 
 	for (i = 0; i < map->num_stripes; i++) {
 		struct btrfs_device *dev = map->stripes[i].dev;
 
-		btrfs_device_set_bytes_used(dev, dev->bytes_used + stripe_size);
+		btrfs_device_set_bytes_used(dev,
+					    dev->bytes_used + ctl->stripe_size);
 		if (list_empty(&dev->post_commit_list))
 			list_add_tail(&dev->post_commit_list,
 				      &trans->transaction->dev_update_list);
 	}
 
-	atomic64_sub(stripe_size * map->num_stripes, &info->free_chunk_space);
+	atomic64_sub(ctl->stripe_size * map->num_stripes,
+		     &info->free_chunk_space);
 
 	free_extent_map(em);
 	check_raid56_incompat_flag(info, type);
+	check_raid1c34_incompat_flag(info, type);
 
-	kfree(devices_info);
 	return 0;
 
 error_del_extent:
@@ -5178,11 +5107,68 @@ error_del_extent:
 	free_extent_map(em);
 	/* One for the tree reference */
 	free_extent_map(em);
-error:
+
+	return ret;
+}
+
+int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type)
+{
+	struct btrfs_fs_info *info = trans->fs_info;
+	struct btrfs_fs_devices *fs_devices = info->fs_devices;
+	struct btrfs_device_info *devices_info = NULL;
+	struct alloc_chunk_ctl ctl;
+	int ret;
+
+	lockdep_assert_held(&info->chunk_mutex);
+
+	if (!alloc_profile_is_valid(type, 0)) {
+		ASSERT(0);
+		return -EINVAL;
+	}
+
+	if (list_empty(&fs_devices->alloc_list)) {
+		if (btrfs_test_opt(info, ENOSPC_DEBUG))
+			btrfs_debug(info, "%s: no writable device", __func__);
+		return -ENOSPC;
+	}
+
+	if (!(type & BTRFS_BLOCK_GROUP_TYPE_MASK)) {
+		btrfs_err(info, "invalid chunk type 0x%llx requested", type);
+		ASSERT(0);
+		return -EINVAL;
+	}
+
+	ctl.start = find_next_chunk(info);
+	ctl.type = type;
+	init_alloc_chunk_ctl(fs_devices, &ctl);
+
+	devices_info = kcalloc(fs_devices->rw_devices, sizeof(*devices_info),
+			       GFP_NOFS);
+	if (!devices_info)
+		return -ENOMEM;
+
+	ret = gather_device_info(fs_devices, &ctl, devices_info);
+	if (ret < 0)
+		goto out;
+
+	ret = decide_stripe_size(fs_devices, &ctl, devices_info);
+	if (ret < 0)
+		goto out;
+
+	ret = create_chunk(trans, &ctl, devices_info);
+
+out:
 	kfree(devices_info);
 	return ret;
 }
 
+/*
+ * Chunk allocation falls into two parts. The first part does work
+ * that makes the new allocated chunk usable, but does not do any operation
+ * that modifies the chunk tree. The second part does the work that
+ * requires modifying the chunk tree. This division is important for the
+ * bootstrap process of adding storage to a seed btrfs.
+ */
 int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
 			     u64 chunk_offset, u64 chunk_size)
 {
@@ -5281,39 +5267,19 @@ out:
 	return ret;
 }
 
-/*
- * Chunk allocation falls into two parts. The first part does work
- * that makes the new allocated chunk usable, but does not do any operation
- * that modifies the chunk tree. The second part does the work that
- * requires modifying the chunk tree. This division is important for the
- * bootstrap process of adding storage to a seed btrfs.
- */
-int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type)
-{
-	u64 chunk_offset;
-
-	lockdep_assert_held(&trans->fs_info->chunk_mutex);
-	chunk_offset = find_next_chunk(trans->fs_info);
-	return __btrfs_alloc_chunk(trans, chunk_offset, type);
-}
-
 static noinline int init_first_rw_device(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	u64 chunk_offset;
-	u64 sys_chunk_offset;
 	u64 alloc_profile;
 	int ret;
 
-	chunk_offset = find_next_chunk(fs_info);
 	alloc_profile = btrfs_metadata_alloc_profile(fs_info);
-	ret = __btrfs_alloc_chunk(trans, chunk_offset, alloc_profile);
+	ret = btrfs_alloc_chunk(trans, alloc_profile);
 	if (ret)
 		return ret;
 
-	sys_chunk_offset = find_next_chunk(fs_info);
 	alloc_profile = btrfs_system_alloc_profile(fs_info);
-	ret = __btrfs_alloc_chunk(trans, sys_chunk_offset, alloc_profile);
+	ret = btrfs_alloc_chunk(trans, alloc_profile);
 	return ret;
 }
 
@@ -5510,31 +5476,19 @@ static int find_live_mirror(struct btrfs_fs_info *fs_info,
 	return preferred_mirror;
 }
 
-static inline int parity_smaller(u64 a, u64 b)
-{
-	return a > b;
-}
-
 /* Bubble-sort the stripe set to put the parity/syndrome stripes last */
 static void sort_parity_stripes(struct btrfs_bio *bbio, int num_stripes)
 {
-	struct btrfs_bio_stripe s;
 	int i;
-	u64 l;
 	int again = 1;
 
 	while (again) {
 		again = 0;
 		for (i = 0; i < num_stripes - 1; i++) {
-			if (parity_smaller(bbio->raid_map[i],
-					   bbio->raid_map[i+1])) {
-				s = bbio->stripes[i];
-				l = bbio->raid_map[i];
-				bbio->stripes[i] = bbio->stripes[i+1];
-				bbio->raid_map[i] = bbio->raid_map[i+1];
-				bbio->stripes[i+1] = s;
-				bbio->raid_map[i+1] = l;
-
+			/* Swap if parity is on a smaller index */
+			if (bbio->raid_map[i] > bbio->raid_map[i + 1]) {
+				swap(bbio->stripes[i], bbio->stripes[i + 1]);
+				swap(bbio->raid_map[i], bbio->raid_map[i + 1]);
 				again = 1;
 			}
 		}
@@ -5583,12 +5537,13 @@ void btrfs_put_bbio(struct btrfs_bio *bbio)
  * replace.
  */
 static int __btrfs_map_block_for_discard(struct btrfs_fs_info *fs_info,
-					 u64 logical, u64 length,
+					 u64 logical, u64 *length_ret,
 					 struct btrfs_bio **bbio_ret)
 {
 	struct extent_map *em;
 	struct map_lookup *map;
 	struct btrfs_bio *bbio;
+	u64 length = *length_ret;
 	u64 offset;
 	u64 stripe_nr;
 	u64 stripe_nr_end;
@@ -5621,7 +5576,8 @@ static int __btrfs_map_block_for_discard(struct btrfs_fs_info *fs_info,
 	}
 
 	offset = logical - em->start;
-	length = min_t(u64, em->len - offset, length);
+	length = min_t(u64, em->start + em->len - logical, length);
+	*length_ret = length;
 
 	stripe_len = map->stripe_len;
 	/*
@@ -6033,10 +5989,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
 	struct btrfs_io_geometry geom;
 
 	ASSERT(bbio_ret);
-
-	if (op == BTRFS_MAP_DISCARD)
-		return __btrfs_map_block_for_discard(fs_info, logical,
-						     *length, bbio_ret);
+	ASSERT(op != BTRFS_MAP_DISCARD);
 
 	ret = btrfs_get_io_geometry(fs_info, op, logical, *length, &geom);
 	if (ret < 0)
@@ -6266,6 +6219,10 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 		      u64 logical, u64 *length,
 		      struct btrfs_bio **bbio_ret, int mirror_num)
 {
+	if (op == BTRFS_MAP_DISCARD)
+		return __btrfs_map_block_for_discard(fs_info, logical,
+						     length, bbio_ret);
+
 	return __btrfs_map_block(fs_info, op, logical, length, bbio_ret,
 				 mirror_num, 0);
 }
@@ -6278,75 +6235,6 @@ int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 	return __btrfs_map_block(fs_info, op, logical, length, bbio_ret, 0, 1);
 }
 
-int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
-		     u64 physical, u64 **logical, int *naddrs, int *stripe_len)
-{
-	struct extent_map *em;
-	struct map_lookup *map;
-	u64 *buf;
-	u64 bytenr;
-	u64 length;
-	u64 stripe_nr;
-	u64 rmap_len;
-	int i, j, nr = 0;
-
-	em = btrfs_get_chunk_map(fs_info, chunk_start, 1);
-	if (IS_ERR(em))
-		return -EIO;
-
-	map = em->map_lookup;
-	length = em->len;
-	rmap_len = map->stripe_len;
-
-	if (map->type & BTRFS_BLOCK_GROUP_RAID10)
-		length = div_u64(length, map->num_stripes / map->sub_stripes);
-	else if (map->type & BTRFS_BLOCK_GROUP_RAID0)
-		length = div_u64(length, map->num_stripes);
-	else if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
-		length = div_u64(length, nr_data_stripes(map));
-		rmap_len = map->stripe_len * nr_data_stripes(map);
-	}
-
-	buf = kcalloc(map->num_stripes, sizeof(u64), GFP_NOFS);
-	BUG_ON(!buf); /* -ENOMEM */
-
-	for (i = 0; i < map->num_stripes; i++) {
-		if (map->stripes[i].physical > physical ||
-		    map->stripes[i].physical + length <= physical)
-			continue;
-
-		stripe_nr = physical - map->stripes[i].physical;
-		stripe_nr = div64_u64(stripe_nr, map->stripe_len);
-
-		if (map->type & BTRFS_BLOCK_GROUP_RAID10) {
-			stripe_nr = stripe_nr * map->num_stripes + i;
-			stripe_nr = div_u64(stripe_nr, map->sub_stripes);
-		} else if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
-			stripe_nr = stripe_nr * map->num_stripes + i;
-		} /* else if RAID[56], multiply by nr_data_stripes().
-		   * Alternatively, just use rmap_len below instead of
-		   * map->stripe_len */
-
-		bytenr = chunk_start + stripe_nr * rmap_len;
-		WARN_ON(nr >= map->num_stripes);
-		for (j = 0; j < nr; j++) {
-			if (buf[j] == bytenr)
-				break;
-		}
-		if (j == nr) {
-			WARN_ON(nr >= map->num_stripes);
-			buf[nr++] = bytenr;
-		}
-	}
-
-	*logical = buf;
-	*naddrs = nr;
-	*stripe_len = rmap_len;
-
-	free_extent_map(em);
-	return 0;
-}
-
 static inline void btrfs_end_bbio(struct btrfs_bio *bbio, struct bio *bio)
 {
 	bio->bi_private = bbio->private;
@@ -6416,52 +6304,8 @@ static void btrfs_end_bio(struct bio *bio)
 	}
 }
 
-/*
- * see run_scheduled_bios for a description of why bios are collected for
- * async submit.
- *
- * This will add one bio to the pending list for a device and make sure
- * the work struct is scheduled.
- */
-static noinline void btrfs_schedule_bio(struct btrfs_device *device,
-					struct bio *bio)
-{
-	struct btrfs_fs_info *fs_info = device->fs_info;
-	int should_queue = 1;
-	struct btrfs_pending_bios *pending_bios;
-
-	/* don't bother with additional async steps for reads, right now */
-	if (bio_op(bio) == REQ_OP_READ) {
-		btrfsic_submit_bio(bio);
-		return;
-	}
-
-	WARN_ON(bio->bi_next);
-	bio->bi_next = NULL;
-
-	spin_lock(&device->io_lock);
-	if (op_is_sync(bio->bi_opf))
-		pending_bios = &device->pending_sync_bios;
-	else
-		pending_bios = &device->pending_bios;
-
-	if (pending_bios->tail)
-		pending_bios->tail->bi_next = bio;
-
-	pending_bios->tail = bio;
-	if (!pending_bios->head)
-		pending_bios->head = bio;
-	if (device->running_pending)
-		should_queue = 0;
-
-	spin_unlock(&device->io_lock);
-
-	if (should_queue)
-		btrfs_queue_work(fs_info->submit_workers, &device->work);
-}
-
 static void submit_stripe_bio(struct btrfs_bio *bbio, struct bio *bio,
-			      u64 physical, int dev_nr, int async)
+			      u64 physical, int dev_nr)
 {
 	struct btrfs_device *dev = bbio->stripes[dev_nr].dev;
 	struct btrfs_fs_info *fs_info = bbio->fs_info;
@@ -6473,16 +6317,13 @@ static void submit_stripe_bio(struct btrfs_bio *bbio, struct bio *bio,
 	btrfs_debug_in_rcu(fs_info,
 	"btrfs_map_bio: rw %d 0x%x, sector=%llu, dev=%lu (%s id %llu), size=%u",
 		bio_op(bio), bio->bi_opf, (u64)bio->bi_iter.bi_sector,
-		(u_long)dev->bdev->bd_dev, rcu_str_deref(dev->name), dev->devid,
-		bio->bi_iter.bi_size);
+		(unsigned long)dev->bdev->bd_dev, rcu_str_deref(dev->name),
+		dev->devid, bio->bi_iter.bi_size);
 	bio_set_dev(bio, dev->bdev);
 
 	btrfs_bio_counter_inc_noblocked(fs_info);
 
-	if (async)
-		btrfs_schedule_bio(dev, bio);
-	else
-		btrfsic_submit_bio(bio);
+	btrfsic_submit_bio(bio);
 }
 
 static void bbio_error(struct btrfs_bio *bbio, struct bio *bio, u64 logical)
@@ -6503,7 +6344,7 @@ static void bbio_error(struct btrfs_bio *bbio, struct bio *bio, u64 logical)
 }
 
 blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
-			   int mirror_num, int async_submit)
+			   int mirror_num)
 {
 	struct btrfs_device *dev;
 	struct bio *first_bio = bio;
@@ -6572,7 +6413,7 @@ blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 			bio = first_bio;
 
 		submit_stripe_bio(bbio, bio, bbio->stripes[dev_nr].physical,
-				  dev_nr, async_submit);
+				  dev_nr);
 	}
 	btrfs_bio_counter_dec(fs_info);
 	return BLK_STS_OK;
@@ -6676,9 +6517,6 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 	else
 		generate_random_uuid(dev->uuid);
 
-	btrfs_init_work(&dev->work, btrfs_submit_helper,
-			pending_bios_fn, NULL, NULL);
-
 	return dev;
 }
 
@@ -6697,19 +6535,14 @@ static u64 calc_stripe_length(u64 type, u64 chunk_len, int num_stripes)
 {
 	int index = btrfs_bg_flags_to_raid_index(type);
 	int ncopies = btrfs_raid_array[index].ncopies;
+	const int nparity = btrfs_raid_array[index].nparity;
 	int data_stripes;
 
-	switch (type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
-	case BTRFS_BLOCK_GROUP_RAID5:
-		data_stripes = num_stripes - 1;
-		break;
-	case BTRFS_BLOCK_GROUP_RAID6:
-		data_stripes = num_stripes - 2;
-		break;
-	default:
+	if (nparity)
+		data_stripes = num_stripes - nparity;
+	else
 		data_stripes = num_stripes / ncopies;
-		break;
-	}
+
 	return div_u64(chunk_len, data_stripes);
 }
 
@@ -6875,7 +6708,7 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info,
 		if (IS_ERR(fs_devices))
 			return fs_devices;
 
-		fs_devices->seeding = 1;
+		fs_devices->seeding = true;
 		fs_devices->opened = 1;
 		return fs_devices;
 	}
@@ -7064,48 +6897,49 @@ int btrfs_read_sys_array(struct btrfs_fs_info *fs_info)
 		sb_array_offset += len;
 		cur_offset += len;
 
-		if (key.type == BTRFS_CHUNK_ITEM_KEY) {
-			chunk = (struct btrfs_chunk *)sb_array_offset;
-			/*
-			 * At least one btrfs_chunk with one stripe must be
-			 * present, exact stripe count check comes afterwards
-			 */
-			len = btrfs_chunk_item_size(1);
-			if (cur_offset + len > array_size)
-				goto out_short_read;
-
-			num_stripes = btrfs_chunk_num_stripes(sb, chunk);
-			if (!num_stripes) {
-				btrfs_err(fs_info,
-					"invalid number of stripes %u in sys_array at offset %u",
-					num_stripes, cur_offset);
-				ret = -EIO;
-				break;
-			}
+		if (key.type != BTRFS_CHUNK_ITEM_KEY) {
+			btrfs_err(fs_info,
+			    "unexpected item type %u in sys_array at offset %u",
+				  (u32)key.type, cur_offset);
+			ret = -EIO;
+			break;
+		}
 
-			type = btrfs_chunk_type(sb, chunk);
-			if ((type & BTRFS_BLOCK_GROUP_SYSTEM) == 0) {
-				btrfs_err(fs_info,
-			    "invalid chunk type %llu in sys_array at offset %u",
-					type, cur_offset);
-				ret = -EIO;
-				break;
-			}
+		chunk = (struct btrfs_chunk *)sb_array_offset;
+		/*
+		 * At least one btrfs_chunk with one stripe must be present,
+		 * exact stripe count check comes afterwards
+		 */
+		len = btrfs_chunk_item_size(1);
+		if (cur_offset + len > array_size)
+			goto out_short_read;
 
-			len = btrfs_chunk_item_size(num_stripes);
-			if (cur_offset + len > array_size)
-				goto out_short_read;
+		num_stripes = btrfs_chunk_num_stripes(sb, chunk);
+		if (!num_stripes) {
+			btrfs_err(fs_info,
+			"invalid number of stripes %u in sys_array at offset %u",
+				  num_stripes, cur_offset);
+			ret = -EIO;
+			break;
+		}
 
-			ret = read_one_chunk(&key, sb, chunk);
-			if (ret)
-				break;
-		} else {
+		type = btrfs_chunk_type(sb, chunk);
+		if ((type & BTRFS_BLOCK_GROUP_SYSTEM) == 0) {
 			btrfs_err(fs_info,
-			    "unexpected item type %u in sys_array at offset %u",
-				  (u32)key.type, cur_offset);
+			"invalid chunk type %llu in sys_array at offset %u",
+				  type, cur_offset);
 			ret = -EIO;
 			break;
 		}
+
+		len = btrfs_chunk_item_size(num_stripes);
+		if (cur_offset + len > array_size)
+			goto out_short_read;
+
+		ret = read_one_chunk(&key, sb, chunk);
+		if (ret)
+			break;
+
 		array_ptr += len;
 		sb_array_offset += len;
 		cur_offset += len;
@@ -7547,6 +7381,8 @@ int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info,
 			else
 				btrfs_dev_stat_set(dev, i, 0);
 		}
+		btrfs_info(fs_info, "device stats zeroed by %s (%d)",
+			   current->comm, task_pid_nr(current));
 	} else {
 		for (i = 0; i < BTRFS_DEV_STAT_VALUES_MAX; i++)
 			if (stats->nr_items > i)
@@ -7557,36 +7393,6 @@ int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
-void btrfs_scratch_superblocks(struct block_device *bdev, const char *device_path)
-{
-	struct buffer_head *bh;
-	struct btrfs_super_block *disk_super;
-	int copy_num;
-
-	if (!bdev)
-		return;
-
-	for (copy_num = 0; copy_num < BTRFS_SUPER_MIRROR_MAX;
-		copy_num++) {
-
-		if (btrfs_read_dev_one_super(bdev, copy_num, &bh))
-			continue;
-
-		disk_super = (struct btrfs_super_block *)bh->b_data;
-
-		memset(&disk_super->magic, 0, sizeof(disk_super->magic));
-		set_buffer_dirty(bh);
-		sync_dirty_buffer(bh);
-		brelse(bh);
-	}
-
-	/* Notify udev that device has changed */
-	btrfs_kobject_uevent(bdev, KOBJ_CHANGE);
-
-	/* Update ctime/mtime for device path for libblkid */
-	update_dev_time(device_path);
-}
-
 /*
  * Update the size and bytes used for each device where it changed.  This is
  * delayed since we would otherwise get errors while writing out the
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index a7da1f3e3627..f067b5934c46 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -17,12 +17,6 @@ extern struct mutex uuid_mutex;
 
 #define BTRFS_STRIPE_LEN	SZ_64K
 
-struct buffer_head;
-struct btrfs_pending_bios {
-	struct bio *head;
-	struct bio *tail;
-};
-
 struct btrfs_io_geometry {
 	/* remaining bytes before crossing a stripe */
 	u64 len;
@@ -68,13 +62,6 @@ struct btrfs_device {
 
 	u64 generation;
 
-	spinlock_t io_lock ____cacheline_aligned;
-	int running_pending;
-	/* regular prio bios */
-	struct btrfs_pending_bios pending_bios;
-	/* sync bios */
-	struct btrfs_pending_bios pending_sync_bios;
-
 	struct block_device *bdev;
 
 	/* the mode sent to blkdev_get */
@@ -131,8 +118,6 @@ struct btrfs_device {
 	/* per-device scrub information */
 	struct scrub_ctx *scrub_ctx;
 
-	struct btrfs_work work;
-
 	/* readahead state */
 	atomic_t reada_in_flight;
 	u64 reada_next;
@@ -149,6 +134,10 @@ struct btrfs_device {
 	atomic_t dev_stat_values[BTRFS_DEV_STAT_VALUES_MAX];
 
 	struct extent_io_tree alloc_state;
+
+	struct completion kobj_unregister;
+	/* For sysfs/FSID/devinfo/devid/ */
+	struct kobject devid_kobj;
 };
 
 /*
@@ -179,7 +168,7 @@ btrfs_device_set_##name(struct btrfs_device *dev, u64 size)		\
 	write_seqcount_end(&dev->data_seqcount);			\
 	preempt_enable();						\
 }
-#elif BITS_PER_LONG==32 && defined(CONFIG_PREEMPT)
+#elif BITS_PER_LONG==32 && defined(CONFIG_PREEMPTION)
 #define BTRFS_DEVICE_GETSET_FUNCS(name)					\
 static inline u64							\
 btrfs_device_get_##name(const struct btrfs_device *dev)			\
@@ -218,6 +207,10 @@ BTRFS_DEVICE_GETSET_FUNCS(total_bytes);
 BTRFS_DEVICE_GETSET_FUNCS(disk_total_bytes);
 BTRFS_DEVICE_GETSET_FUNCS(bytes_used);
 
+enum btrfs_chunk_allocation_policy {
+	BTRFS_CHUNK_ALLOC_REGULAR,
+};
+
 struct btrfs_fs_devices {
 	u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */
 	u8 metadata_uuid[BTRFS_FSID_SIZE];
@@ -254,20 +247,23 @@ struct btrfs_fs_devices {
 	struct list_head alloc_list;
 
 	struct btrfs_fs_devices *seed;
-	int seeding;
+	bool seeding;
 
 	int opened;
 
 	/* set when we find or add a device that doesn't have the
 	 * nonrot flag set
 	 */
-	int rotating;
+	bool rotating;
 
 	struct btrfs_fs_info *fs_info;
 	/* sysfs kobjects */
 	struct kobject fsid_kobj;
-	struct kobject *device_dir_kobj;
+	struct kobject *devices_kobj;
+	struct kobject *devinfo_kobj;
 	struct completion kobj_unregister;
+
+	enum btrfs_chunk_allocation_policy chunk_alloc_policy;
 };
 
 #define BTRFS_BIO_INLINE_CSUM_SIZE	64
@@ -330,7 +326,6 @@ struct btrfs_bio {
 	u64 map_type; /* get from map_lookup->type */
 	bio_end_io_t *end_io;
 	struct bio *orig_bio;
-	unsigned long flags;
 	void *private;
 	atomic_t error;
 	int max_errors;
@@ -429,14 +424,12 @@ int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 		     struct btrfs_bio **bbio_ret);
 int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 		u64 logical, u64 len, struct btrfs_io_geometry *io_geom);
-int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
-		     u64 physical, u64 **logical, int *naddrs, int *stripe_len);
 int btrfs_read_sys_array(struct btrfs_fs_info *fs_info);
 int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info);
 int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type);
 void btrfs_mapping_tree_free(struct extent_map_tree *tree);
 blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
-			   int mirror_num, int async_submit);
+			   int mirror_num);
 int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		       fmode_t flags, void *holder);
 struct btrfs_device *btrfs_scan_one_device(const char *path,
@@ -472,7 +465,7 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info);
 int btrfs_pause_balance(struct btrfs_fs_info *fs_info);
 int btrfs_cancel_balance(struct btrfs_fs_info *fs_info);
 int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info);
-int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info);
+int btrfs_uuid_scan_kthread(void *data);
 int btrfs_chunk_readonly(struct btrfs_fs_info *fs_info, u64 chunk_offset);
 int find_free_dev_extent(struct btrfs_device *device, u64 num_bytes,
 			 u64 *start, u64 *max_avail);
@@ -485,7 +478,6 @@ int btrfs_run_dev_stats(struct btrfs_trans_handle *trans);
 void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_device *srcdev);
 void btrfs_rm_dev_replace_free_srcdev(struct btrfs_device *srcdev);
 void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev);
-void btrfs_scratch_superblocks(struct block_device *bdev, const char *device_path);
 int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info,
 			   u64 logical, u64 len);
 unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
@@ -495,6 +487,7 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
 int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset);
 struct extent_map *btrfs_get_chunk_map(struct btrfs_fs_info *fs_info,
 				       u64 logical, u64 length);
+void btrfs_release_disk_super(struct btrfs_super_block *super);
 
 static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,
 				      int index)
@@ -557,6 +550,10 @@ static inline enum btrfs_raid_types btrfs_bg_flags_to_raid_index(u64 flags)
 		return BTRFS_RAID_RAID10;
 	else if (flags & BTRFS_BLOCK_GROUP_RAID1)
 		return BTRFS_RAID_RAID1;
+	else if (flags & BTRFS_BLOCK_GROUP_RAID1C3)
+		return BTRFS_RAID_RAID1C3;
+	else if (flags & BTRFS_BLOCK_GROUP_RAID1C4)
+		return BTRFS_RAID_RAID1C4;
 	else if (flags & BTRFS_BLOCK_GROUP_DUP)
 		return BTRFS_RAID_DUP;
 	else if (flags & BTRFS_BLOCK_GROUP_RAID0)
@@ -571,7 +568,7 @@ static inline enum btrfs_raid_types btrfs_bg_flags_to_raid_index(u64 flags)
 
 void btrfs_commit_device_sizes(struct btrfs_transaction *trans);
 
-struct list_head *btrfs_get_fs_uuids(void);
+struct list_head * __attribute_const__ btrfs_get_fs_uuids(void);
 void btrfs_set_fs_info_ptr(struct btrfs_fs_info *fs_info);
 void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info);
 bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info,
diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c
index df1aace5df50..05615a1099db 100644
--- a/fs/btrfs/zlib.c
+++ b/fs/btrfs/zlib.c
@@ -20,28 +20,22 @@
 #include <linux/refcount.h>
 #include "compression.h"
 
+/* workspace buffer size for s390 zlib hardware support */
+#define ZLIB_DFLTCC_BUF_SIZE    (4 * PAGE_SIZE)
+
 struct workspace {
 	z_stream strm;
 	char *buf;
+	unsigned int buf_size;
 	struct list_head list;
 	int level;
 };
 
 static struct workspace_manager wsm;
 
-static void zlib_init_workspace_manager(void)
+struct list_head *zlib_get_workspace(unsigned int level)
 {
-	btrfs_init_workspace_manager(&wsm, &btrfs_zlib_compress);
-}
-
-static void zlib_cleanup_workspace_manager(void)
-{
-	btrfs_cleanup_workspace_manager(&wsm);
-}
-
-static struct list_head *zlib_get_workspace(unsigned int level)
-{
-	struct list_head *ws = btrfs_get_workspace(&wsm, level);
+	struct list_head *ws = btrfs_get_workspace(BTRFS_COMPRESS_ZLIB, level);
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 
 	workspace->level = level;
@@ -49,12 +43,7 @@ static struct list_head *zlib_get_workspace(unsigned int level)
 	return ws;
 }
 
-static void zlib_put_workspace(struct list_head *ws)
-{
-	btrfs_put_workspace(&wsm, ws);
-}
-
-static void zlib_free_workspace(struct list_head *ws)
+void zlib_free_workspace(struct list_head *ws)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 
@@ -63,7 +52,7 @@ static void zlib_free_workspace(struct list_head *ws)
 	kfree(workspace);
 }
 
-static struct list_head *zlib_alloc_workspace(unsigned int level)
+struct list_head *zlib_alloc_workspace(unsigned int level)
 {
 	struct workspace *workspace;
 	int workspacesize;
@@ -76,7 +65,21 @@ static struct list_head *zlib_alloc_workspace(unsigned int level)
 			zlib_inflate_workspacesize());
 	workspace->strm.workspace = kvmalloc(workspacesize, GFP_KERNEL);
 	workspace->level = level;
-	workspace->buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	workspace->buf = NULL;
+	/*
+	 * In case of s390 zlib hardware support, allocate lager workspace
+	 * buffer. If allocator fails, fall back to a single page buffer.
+	 */
+	if (zlib_deflate_dfltcc_enabled()) {
+		workspace->buf = kmalloc(ZLIB_DFLTCC_BUF_SIZE,
+					 __GFP_NOMEMALLOC | __GFP_NORETRY |
+					 __GFP_NOWARN | GFP_NOIO);
+		workspace->buf_size = ZLIB_DFLTCC_BUF_SIZE;
+	}
+	if (!workspace->buf) {
+		workspace->buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+		workspace->buf_size = PAGE_SIZE;
+	}
 	if (!workspace->strm.workspace || !workspace->buf)
 		goto fail;
 
@@ -88,13 +91,9 @@ fail:
 	return ERR_PTR(-ENOMEM);
 }
 
-static int zlib_compress_pages(struct list_head *ws,
-			       struct address_space *mapping,
-			       u64 start,
-			       struct page **pages,
-			       unsigned long *out_pages,
-			       unsigned long *total_in,
-			       unsigned long *total_out)
+int zlib_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	int ret;
@@ -104,6 +103,7 @@ static int zlib_compress_pages(struct list_head *ws,
 	struct page *in_page = NULL;
 	struct page *out_page = NULL;
 	unsigned long bytes_left;
+	unsigned int in_buf_pages;
 	unsigned long len = *total_out;
 	unsigned long nr_dest_pages = *out_pages;
 	const unsigned long max_out = nr_dest_pages * PAGE_SIZE;
@@ -121,9 +121,6 @@ static int zlib_compress_pages(struct list_head *ws,
 	workspace->strm.total_in = 0;
 	workspace->strm.total_out = 0;
 
-	in_page = find_get_page(mapping, start >> PAGE_SHIFT);
-	data_in = kmap(in_page);
-
 	out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
 	if (out_page == NULL) {
 		ret = -ENOMEM;
@@ -133,12 +130,51 @@ static int zlib_compress_pages(struct list_head *ws,
 	pages[0] = out_page;
 	nr_pages = 1;
 
-	workspace->strm.next_in = data_in;
+	workspace->strm.next_in = workspace->buf;
+	workspace->strm.avail_in = 0;
 	workspace->strm.next_out = cpage_out;
 	workspace->strm.avail_out = PAGE_SIZE;
-	workspace->strm.avail_in = min(len, PAGE_SIZE);
 
 	while (workspace->strm.total_in < len) {
+		/*
+		 * Get next input pages and copy the contents to
+		 * the workspace buffer if required.
+		 */
+		if (workspace->strm.avail_in == 0) {
+			bytes_left = len - workspace->strm.total_in;
+			in_buf_pages = min(DIV_ROUND_UP(bytes_left, PAGE_SIZE),
+					   workspace->buf_size / PAGE_SIZE);
+			if (in_buf_pages > 1) {
+				int i;
+
+				for (i = 0; i < in_buf_pages; i++) {
+					if (in_page) {
+						kunmap(in_page);
+						put_page(in_page);
+					}
+					in_page = find_get_page(mapping,
+								start >> PAGE_SHIFT);
+					data_in = kmap(in_page);
+					memcpy(workspace->buf + i * PAGE_SIZE,
+					       data_in, PAGE_SIZE);
+					start += PAGE_SIZE;
+				}
+				workspace->strm.next_in = workspace->buf;
+			} else {
+				if (in_page) {
+					kunmap(in_page);
+					put_page(in_page);
+				}
+				in_page = find_get_page(mapping,
+							start >> PAGE_SHIFT);
+				data_in = kmap(in_page);
+				start += PAGE_SIZE;
+				workspace->strm.next_in = data_in;
+			}
+			workspace->strm.avail_in = min(bytes_left,
+						       (unsigned long) workspace->buf_size);
+		}
+
 		ret = zlib_deflate(&workspace->strm, Z_SYNC_FLUSH);
 		if (ret != Z_OK) {
 			pr_debug("BTRFS: deflate in loop returned %d\n",
@@ -180,33 +216,43 @@ static int zlib_compress_pages(struct list_head *ws,
 		/* we're all done */
 		if (workspace->strm.total_in >= len)
 			break;
-
-		/* we've read in a full page, get a new one */
-		if (workspace->strm.avail_in == 0) {
-			if (workspace->strm.total_out > max_out)
-				break;
-
-			bytes_left = len - workspace->strm.total_in;
-			kunmap(in_page);
-			put_page(in_page);
-
-			start += PAGE_SIZE;
-			in_page = find_get_page(mapping,
-						start >> PAGE_SHIFT);
-			data_in = kmap(in_page);
-			workspace->strm.avail_in = min(bytes_left,
-							   PAGE_SIZE);
-			workspace->strm.next_in = data_in;
-		}
+		if (workspace->strm.total_out > max_out)
+			break;
 	}
 	workspace->strm.avail_in = 0;
-	ret = zlib_deflate(&workspace->strm, Z_FINISH);
-	zlib_deflateEnd(&workspace->strm);
-
-	if (ret != Z_STREAM_END) {
-		ret = -EIO;
-		goto out;
+	/*
+	 * Call deflate with Z_FINISH flush parameter providing more output
+	 * space but no more input data, until it returns with Z_STREAM_END.
+	 */
+	while (ret != Z_STREAM_END) {
+		ret = zlib_deflate(&workspace->strm, Z_FINISH);
+		if (ret == Z_STREAM_END)
+			break;
+		if (ret != Z_OK && ret != Z_BUF_ERROR) {
+			zlib_deflateEnd(&workspace->strm);
+			ret = -EIO;
+			goto out;
+		} else if (workspace->strm.avail_out == 0) {
+			/* get another page for the stream end */
+			kunmap(out_page);
+			if (nr_pages == nr_dest_pages) {
+				out_page = NULL;
+				ret = -E2BIG;
+				goto out;
+			}
+			out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+			if (out_page == NULL) {
+				ret = -ENOMEM;
+				goto out;
+			}
+			cpage_out = kmap(out_page);
+			pages[nr_pages] = out_page;
+			nr_pages++;
+			workspace->strm.avail_out = PAGE_SIZE;
+			workspace->strm.next_out = cpage_out;
+		}
 	}
+	zlib_deflateEnd(&workspace->strm);
 
 	if (workspace->strm.total_out >= workspace->strm.total_in) {
 		ret = -E2BIG;
@@ -228,7 +274,7 @@ out:
 	return ret;
 }
 
-static int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
+int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	int ret = 0, ret2;
@@ -250,7 +296,7 @@ static int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 
 	workspace->strm.total_out = 0;
 	workspace->strm.next_out = workspace->buf;
-	workspace->strm.avail_out = PAGE_SIZE;
+	workspace->strm.avail_out = workspace->buf_size;
 
 	/* If it's deflate, and it's got no preset dictionary, then
 	   we can tell zlib to skip the adler32 check. */
@@ -289,7 +335,7 @@ static int zlib_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 		}
 
 		workspace->strm.next_out = workspace->buf;
-		workspace->strm.avail_out = PAGE_SIZE;
+		workspace->strm.avail_out = workspace->buf_size;
 
 		if (workspace->strm.avail_in == 0) {
 			unsigned long tmp;
@@ -319,10 +365,9 @@ done:
 	return ret;
 }
 
-static int zlib_decompress(struct list_head *ws, unsigned char *data_in,
-			   struct page *dest_page,
-			   unsigned long start_byte,
-			   size_t srclen, size_t destlen)
+int zlib_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	int ret = 0;
@@ -340,7 +385,7 @@ static int zlib_decompress(struct list_head *ws, unsigned char *data_in,
 	workspace->strm.total_in = 0;
 
 	workspace->strm.next_out = workspace->buf;
-	workspace->strm.avail_out = PAGE_SIZE;
+	workspace->strm.avail_out = workspace->buf_size;
 	workspace->strm.total_out = 0;
 	/* If it's deflate, and it's got no preset dictionary, then
 	   we can tell zlib to skip the adler32 check. */
@@ -384,7 +429,7 @@ static int zlib_decompress(struct list_head *ws, unsigned char *data_in,
 			buf_offset = 0;
 
 		bytes = min(PAGE_SIZE - pg_offset,
-			    PAGE_SIZE - buf_offset);
+			    PAGE_SIZE - (buf_offset % PAGE_SIZE));
 		bytes = min(bytes, bytes_left);
 
 		kaddr = kmap_atomic(dest_page);
@@ -395,7 +440,7 @@ static int zlib_decompress(struct list_head *ws, unsigned char *data_in,
 		bytes_left -= bytes;
 next:
 		workspace->strm.next_out = workspace->buf;
-		workspace->strm.avail_out = PAGE_SIZE;
+		workspace->strm.avail_out = workspace->buf_size;
 	}
 
 	if (ret != Z_STREAM_END && bytes_left != 0)
@@ -419,15 +464,7 @@ next:
 }
 
 const struct btrfs_compress_op btrfs_zlib_compress = {
-	.init_workspace_manager	= zlib_init_workspace_manager,
-	.cleanup_workspace_manager = zlib_cleanup_workspace_manager,
-	.get_workspace		= zlib_get_workspace,
-	.put_workspace		= zlib_put_workspace,
-	.alloc_workspace	= zlib_alloc_workspace,
-	.free_workspace		= zlib_free_workspace,
-	.compress_pages		= zlib_compress_pages,
-	.decompress_bio		= zlib_decompress_bio,
-	.decompress		= zlib_decompress,
+	.workspace_manager	= &wsm,
 	.max_level		= 9,
 	.default_level		= BTRFS_ZLIB_DEFAULT_LEVEL,
 };
diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c
index 764d47b107e5..9a4871636c6c 100644
--- a/fs/btrfs/zstd.c
+++ b/fs/btrfs/zstd.c
@@ -91,9 +91,8 @@ static inline struct workspace *list_to_workspace(struct list_head *list)
 	return container_of(list, struct workspace, list);
 }
 
-static void zstd_free_workspace(struct list_head *ws);
-static struct list_head *zstd_alloc_workspace(unsigned int level);
-
+void zstd_free_workspace(struct list_head *ws);
+struct list_head *zstd_alloc_workspace(unsigned int level);
 /*
  * zstd_reclaim_timer_fn - reclaim timer
  * @t: timer
@@ -168,7 +167,7 @@ static void zstd_calc_ws_mem_sizes(void)
 	}
 }
 
-static void zstd_init_workspace_manager(void)
+void zstd_init_workspace_manager(void)
 {
 	struct list_head *ws;
 	int i;
@@ -194,7 +193,7 @@ static void zstd_init_workspace_manager(void)
 	}
 }
 
-static void zstd_cleanup_workspace_manager(void)
+void zstd_cleanup_workspace_manager(void)
 {
 	struct workspace *workspace;
 	int i;
@@ -261,7 +260,7 @@ static struct list_head *zstd_find_workspace(unsigned int level)
  * attempt to allocate a new workspace.  If we fail to allocate one due to
  * memory pressure, go to sleep waiting for the max level workspace to free up.
  */
-static struct list_head *zstd_get_workspace(unsigned int level)
+struct list_head *zstd_get_workspace(unsigned int level)
 {
 	struct list_head *ws;
 	unsigned int nofs_flag;
@@ -302,7 +301,7 @@ again:
  * isn't set, it is also set here.  Only the max level workspace tries and wakes
  * up waiting workspaces.
  */
-static void zstd_put_workspace(struct list_head *ws)
+void zstd_put_workspace(struct list_head *ws)
 {
 	struct workspace *workspace = list_to_workspace(ws);
 
@@ -332,7 +331,7 @@ static void zstd_put_workspace(struct list_head *ws)
 		cond_wake_up(&wsm.wait);
 }
 
-static void zstd_free_workspace(struct list_head *ws)
+void zstd_free_workspace(struct list_head *ws)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 
@@ -341,7 +340,7 @@ static void zstd_free_workspace(struct list_head *ws)
 	kfree(workspace);
 }
 
-static struct list_head *zstd_alloc_workspace(unsigned int level)
+struct list_head *zstd_alloc_workspace(unsigned int level)
 {
 	struct workspace *workspace;
 
@@ -367,13 +366,9 @@ fail:
 	return ERR_PTR(-ENOMEM);
 }
 
-static int zstd_compress_pages(struct list_head *ws,
-		struct address_space *mapping,
-		u64 start,
-		struct page **pages,
-		unsigned long *out_pages,
-		unsigned long *total_in,
-		unsigned long *total_out)
+int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
+		u64 start, struct page **pages, unsigned long *out_pages,
+		unsigned long *total_in, unsigned long *total_out)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	ZSTD_CStream *stream;
@@ -548,7 +543,7 @@ out:
 	return ret;
 }
 
-static int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
+int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	struct page **pages_in = cb->compressed_pages;
@@ -626,10 +621,9 @@ done:
 	return ret;
 }
 
-static int zstd_decompress(struct list_head *ws, unsigned char *data_in,
-		struct page *dest_page,
-		unsigned long start_byte,
-		size_t srclen, size_t destlen)
+int zstd_decompress(struct list_head *ws, unsigned char *data_in,
+		struct page *dest_page, unsigned long start_byte, size_t srclen,
+		size_t destlen)
 {
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	ZSTD_DStream *stream;
@@ -712,15 +706,8 @@ finish:
 }
 
 const struct btrfs_compress_op btrfs_zstd_compress = {
-	.init_workspace_manager = zstd_init_workspace_manager,
-	.cleanup_workspace_manager = zstd_cleanup_workspace_manager,
-	.get_workspace = zstd_get_workspace,
-	.put_workspace = zstd_put_workspace,
-	.alloc_workspace = zstd_alloc_workspace,
-	.free_workspace = zstd_free_workspace,
-	.compress_pages = zstd_compress_pages,
-	.decompress_bio = zstd_decompress_bio,
-	.decompress = zstd_decompress,
+	/* ZSTD uses own workspace manager */
+	.workspace_manager = NULL,
 	.max_level	= ZSTD_BTRFS_MAX_LEVEL,
 	.default_level	= ZSTD_BTRFS_DEFAULT_LEVEL,
 };
diff --git a/fs/buffer.c b/fs/buffer.c
index 86a38b979323..64fe82ec65ff 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -47,6 +47,9 @@
 #include <linux/pagevec.h>
 #include <linux/sched/mm.h>
 #include <trace/events/block.h>
+#include <linux/fscrypt.h>
+
+#include "internal.h"
 
 static int fsync_buffers_list(spinlock_t *lock, struct list_head *list);
 static int submit_bh_wbc(int op, int op_flags, struct buffer_head *bh,
@@ -120,14 +123,6 @@ void __wait_on_buffer(struct buffer_head * bh)
 }
 EXPORT_SYMBOL(__wait_on_buffer);
 
-static void
-__clear_page_buffers(struct page *page)
-{
-	ClearPagePrivate(page);
-	set_page_private(page, 0);
-	put_page(page);
-}
-
 static void buffer_io_error(struct buffer_head *bh, char *msg)
 {
 	if (!test_bit(BH_Quiet, &bh->b_state))
@@ -246,10 +241,6 @@ out:
 	return ret;
 }
 
-/*
- * I/O completion handler for block_read_full_page() - pages
- * which come unlocked at the end of I/O.
- */
 static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 {
 	unsigned long flags;
@@ -275,8 +266,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 	 * decide that the page is now completely done.
 	 */
 	first = page_buffers(page);
-	local_irq_save(flags);
-	bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
+	spin_lock_irqsave(&first->b_uptodate_lock, flags);
 	clear_buffer_async_read(bh);
 	unlock_buffer(bh);
 	tmp = bh;
@@ -289,8 +279,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 		}
 		tmp = tmp->b_this_page;
 	} while (tmp != bh);
-	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
 
 	/*
 	 * If none of the buffers had errors and they are all
@@ -302,11 +291,51 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 	return;
 
 still_busy:
-	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
 	return;
 }
 
+struct decrypt_bh_ctx {
+	struct work_struct work;
+	struct buffer_head *bh;
+};
+
+static void decrypt_bh(struct work_struct *work)
+{
+	struct decrypt_bh_ctx *ctx =
+		container_of(work, struct decrypt_bh_ctx, work);
+	struct buffer_head *bh = ctx->bh;
+	int err;
+
+	err = fscrypt_decrypt_pagecache_blocks(bh->b_page, bh->b_size,
+					       bh_offset(bh));
+	end_buffer_async_read(bh, err == 0);
+	kfree(ctx);
+}
+
+/*
+ * I/O completion handler for block_read_full_page() - pages
+ * which come unlocked at the end of I/O.
+ */
+static void end_buffer_async_read_io(struct buffer_head *bh, int uptodate)
+{
+	/* Decrypt if needed */
+	if (uptodate && IS_ENABLED(CONFIG_FS_ENCRYPTION) &&
+	    IS_ENCRYPTED(bh->b_page->mapping->host) &&
+	    S_ISREG(bh->b_page->mapping->host->i_mode)) {
+		struct decrypt_bh_ctx *ctx = kmalloc(sizeof(*ctx), GFP_ATOMIC);
+
+		if (ctx) {
+			INIT_WORK(&ctx->work, decrypt_bh);
+			ctx->bh = bh;
+			fscrypt_enqueue_decrypt_work(&ctx->work);
+			return;
+		}
+		uptodate = 0;
+	}
+	end_buffer_async_read(bh, uptodate);
+}
+
 /*
  * Completion handler for block_write_full_page() - pages which are unlocked
  * during I/O, and which have PageWriteback cleared upon I/O completion.
@@ -331,8 +360,7 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate)
 	}
 
 	first = page_buffers(page);
-	local_irq_save(flags);
-	bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
+	spin_lock_irqsave(&first->b_uptodate_lock, flags);
 
 	clear_buffer_async_write(bh);
 	unlock_buffer(bh);
@@ -344,14 +372,12 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate)
 		}
 		tmp = tmp->b_this_page;
 	}
-	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
 	end_page_writeback(page);
 	return;
 
 still_busy:
-	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
 	return;
 }
 EXPORT_SYMBOL(end_buffer_async_write);
@@ -379,7 +405,7 @@ EXPORT_SYMBOL(end_buffer_async_write);
  */
 static void mark_buffer_async_read(struct buffer_head *bh)
 {
-	bh->b_end_io = end_buffer_async_read;
+	bh->b_end_io = end_buffer_async_read_io;
 	set_buffer_async_read(bh);
 }
 
@@ -872,7 +898,7 @@ link_dev_buffers(struct page *page, struct buffer_head *head)
 		bh = bh->b_this_page;
 	} while (bh);
 	tail->b_this_page = head;
-	attach_page_buffers(page, head);
+	attach_page_private(page, head);
 }
 
 static sector_t blkdev_max_block(struct block_device *bdev, unsigned int size)
@@ -933,7 +959,7 @@ grow_dev_page(struct block_device *bdev, sector_t block,
 	struct page *page;
 	struct buffer_head *bh;
 	sector_t end_block;
-	int ret = 0;		/* Will call free_more_memory() */
+	int ret = 0;
 	gfp_t gfp_mask;
 
 	gfp_mask = mapping_gfp_constraint(inode->i_mapping, ~__GFP_FS) | gfp;
@@ -1120,12 +1146,19 @@ EXPORT_SYMBOL(mark_buffer_dirty);
 
 void mark_buffer_write_io_error(struct buffer_head *bh)
 {
+	struct super_block *sb;
+
 	set_buffer_write_io_error(bh);
 	/* FIXME: do we need to set this in both places? */
 	if (bh->b_page && bh->b_page->mapping)
 		mapping_set_error(bh->b_page->mapping, -EIO);
 	if (bh->b_assoc_map)
 		mapping_set_error(bh->b_assoc_map, -EIO);
+	rcu_read_lock();
+	sb = READ_ONCE(bh->b_bdev->bd_super);
+	if (sb)
+		errseq_set(&sb->s_wb_err, -EIO);
+	rcu_read_unlock();
 }
 EXPORT_SYMBOL(mark_buffer_write_io_error);
 
@@ -1337,6 +1370,17 @@ void __breadahead(struct block_device *bdev, sector_t block, unsigned size)
 }
 EXPORT_SYMBOL(__breadahead);
 
+void __breadahead_gfp(struct block_device *bdev, sector_t block, unsigned size,
+		      gfp_t gfp)
+{
+	struct buffer_head *bh = __getblk_gfp(bdev, block, size, gfp);
+	if (likely(bh)) {
+		ll_rw_block(REQ_OP_READ, REQ_RAHEAD, 1, &bh);
+		brelse(bh);
+	}
+}
+EXPORT_SYMBOL(__breadahead_gfp);
+
 /**
  *  __bread_gfp() - reads a specified block and returns the bh
  *  @bdev: the block_device to read from
@@ -1385,15 +1429,15 @@ static bool has_bh_in_lru(int cpu, void *dummy)
 	
 	for (i = 0; i < BH_LRU_SIZE; i++) {
 		if (b->bhs[i])
-			return 1;
+			return true;
 	}
 
-	return 0;
+	return false;
 }
 
 void invalidate_bh_lrus(void)
 {
-	on_each_cpu_cond(has_bh_in_lru, invalidate_bh_lru, NULL, 1, GFP_KERNEL);
+	on_each_cpu_cond(has_bh_in_lru, invalidate_bh_lru, NULL, 1);
 }
 EXPORT_SYMBOL_GPL(invalidate_bh_lrus);
 
@@ -1535,7 +1579,7 @@ void create_empty_buffers(struct page *page,
 			bh = bh->b_this_page;
 		} while (bh != head);
 	}
-	attach_page_buffers(page, head);
+	attach_page_private(page, head);
 	spin_unlock(&page->mapping->private_lock);
 }
 EXPORT_SYMBOL(create_empty_buffers);
@@ -2522,7 +2566,7 @@ static void attach_nobh_buffers(struct page *page, struct buffer_head *head)
 			bh->b_this_page = head;
 		bh = bh->b_this_page;
 	} while (bh != head);
-	attach_page_buffers(page, head);
+	attach_page_private(page, head);
 	spin_unlock(&page->mapping->private_lock);
 }
 
@@ -2979,72 +3023,6 @@ static void end_bio_bh_io_sync(struct bio *bio)
 	bio_put(bio);
 }
 
-/*
- * This allows us to do IO even on the odd last sectors
- * of a device, even if the block size is some multiple
- * of the physical sector size.
- *
- * We'll just truncate the bio to the size of the device,
- * and clear the end of the buffer head manually.
- *
- * Truly out-of-range accesses will turn into actual IO
- * errors, this only handles the "we need to be able to
- * do IO at the final sector" case.
- */
-void guard_bio_eod(int op, struct bio *bio)
-{
-	sector_t maxsector;
-	struct bio_vec *bvec = bio_last_bvec_all(bio);
-	unsigned truncated_bytes;
-	struct hd_struct *part;
-
-	rcu_read_lock();
-	part = __disk_get_part(bio->bi_disk, bio->bi_partno);
-	if (part)
-		maxsector = part_nr_sects_read(part);
-	else
-		maxsector = get_capacity(bio->bi_disk);
-	rcu_read_unlock();
-
-	if (!maxsector)
-		return;
-
-	/*
-	 * If the *whole* IO is past the end of the device,
-	 * let it through, and the IO layer will turn it into
-	 * an EIO.
-	 */
-	if (unlikely(bio->bi_iter.bi_sector >= maxsector))
-		return;
-
-	maxsector -= bio->bi_iter.bi_sector;
-	if (likely((bio->bi_iter.bi_size >> 9) <= maxsector))
-		return;
-
-	/* Uhhuh. We've got a bio that straddles the device size! */
-	truncated_bytes = bio->bi_iter.bi_size - (maxsector << 9);
-
-	/*
-	 * The bio contains more than one segment which spans EOD, just return
-	 * and let IO layer turn it into an EIO
-	 */
-	if (truncated_bytes > bvec->bv_len)
-		return;
-
-	/* Truncate the bio.. */
-	bio->bi_iter.bi_size -= truncated_bytes;
-	bvec->bv_len -= truncated_bytes;
-
-	/* ..and clear the end of the buffer for reads */
-	if (op == REQ_OP_READ) {
-		struct bio_vec bv;
-
-		mp_bvec_last_segment(bvec, &bv);
-		zero_user(bv.bv_page, bv.bv_offset + bv.bv_len,
-				truncated_bytes);
-	}
-}
-
 static int submit_bh_wbc(int op, int op_flags, struct buffer_head *bh,
 			 enum rw_hint write_hint, struct writeback_control *wbc)
 {
@@ -3078,15 +3056,15 @@ static int submit_bh_wbc(int op, int op_flags, struct buffer_head *bh,
 	bio->bi_end_io = end_bio_bh_io_sync;
 	bio->bi_private = bh;
 
-	/* Take care of bh's that straddle the end of the device */
-	guard_bio_eod(op, bio);
-
 	if (buffer_meta(bh))
 		op_flags |= REQ_META;
 	if (buffer_prio(bh))
 		op_flags |= REQ_PRIO;
 	bio_set_op_attrs(bio, op, op_flags);
 
+	/* Take care of bh's that straddle the end of the device */
+	guard_bio_eod(bio);
+
 	if (wbc) {
 		wbc_init_bio(wbc, bio);
 		wbc_account_cgroup_owner(wbc, bh->b_page, bh->b_size);
@@ -3248,7 +3226,7 @@ drop_buffers(struct page *page, struct buffer_head **buffers_to_free)
 		bh = next;
 	} while (bh != head);
 	*buffers_to_free = head;
-	__clear_page_buffers(page);
+	detach_page_private(page);
 	return 1;
 failed:
 	return 0;
@@ -3368,6 +3346,7 @@ struct buffer_head *alloc_buffer_head(gfp_t gfp_flags)
 	struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, gfp_flags);
 	if (ret) {
 		INIT_LIST_HEAD(&ret->b_assoc_buffers);
+		spin_lock_init(&ret->b_uptodate_lock);
 		preempt_disable();
 		__this_cpu_inc(bh_accounting.nr);
 		recalc_bh_state();
diff --git a/fs/cachefiles/Kconfig b/fs/cachefiles/Kconfig
index ae559ed5b3b3..ff9ca55a9ae9 100644
--- a/fs/cachefiles/Kconfig
+++ b/fs/cachefiles/Kconfig
@@ -8,7 +8,7 @@ config CACHEFILES
 	  filesystems - primarily networking filesystems - thus allowing fast
 	  local disk to enhance the speed of slower devices.
 
-	  See Documentation/filesystems/caching/cachefiles.txt for more
+	  See Documentation/filesystems/caching/cachefiles.rst for more
 	  information.
 
 config CACHEFILES_DEBUG
@@ -36,5 +36,5 @@ config CACHEFILES_HISTOGRAM
 	  bouncing between CPUs.  On the other hand, the histogram may be
 	  useful for debugging purposes.  Saying 'N' here is recommended.
 
-	  See Documentation/filesystems/caching/cachefiles.txt for more
+	  See Documentation/filesystems/caching/cachefiles.rst for more
 	  information.
diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c
index 44a3ce1e4ce4..e7726f5f1241 100644
--- a/fs/cachefiles/rdwr.c
+++ b/fs/cachefiles/rdwr.c
@@ -60,9 +60,9 @@ static int cachefiles_read_waiter(wait_queue_entry_t *wait, unsigned mode,
 	object = container_of(op->op.object, struct cachefiles_object, fscache);
 	spin_lock(&object->work_lock);
 	list_add_tail(&monitor->op_link, &op->to_do);
+	fscache_enqueue_retrieval(op);
 	spin_unlock(&object->work_lock);
 
-	fscache_enqueue_retrieval(op);
 	fscache_put_retrieval(op);
 	return 0;
 }
@@ -396,9 +396,9 @@ int cachefiles_read_or_alloc_page(struct fscache_retrieval *op,
 	struct cachefiles_object *object;
 	struct cachefiles_cache *cache;
 	struct inode *inode;
-	sector_t block0, block;
+	sector_t block;
 	unsigned shift;
-	int ret;
+	int ret, ret2;
 
 	object = container_of(op->op.object,
 			      struct cachefiles_object, fscache);
@@ -412,7 +412,6 @@ int cachefiles_read_or_alloc_page(struct fscache_retrieval *op,
 
 	inode = d_backing_inode(object->backer);
 	ASSERT(S_ISREG(inode->i_mode));
-	ASSERT(inode->i_mapping->a_ops->bmap);
 	ASSERT(inode->i_mapping->a_ops->readpages);
 
 	/* calculate the shift required to use bmap */
@@ -428,12 +427,14 @@ int cachefiles_read_or_alloc_page(struct fscache_retrieval *op,
 	 *   enough for this as it doesn't indicate errors, but it's all we've
 	 *   got for the moment
 	 */
-	block0 = page->index;
-	block0 <<= shift;
+	block = page->index;
+	block <<= shift;
+
+	ret2 = bmap(inode, &block);
+	ASSERT(ret2 == 0);
 
-	block = inode->i_mapping->a_ops->bmap(inode->i_mapping, block0);
 	_debug("%llx -> %llx",
-	       (unsigned long long) block0,
+	       (unsigned long long) (page->index << shift),
 	       (unsigned long long) block);
 
 	if (block) {
@@ -711,7 +712,6 @@ int cachefiles_read_or_alloc_pages(struct fscache_retrieval *op,
 
 	inode = d_backing_inode(object->backer);
 	ASSERT(S_ISREG(inode->i_mode));
-	ASSERT(inode->i_mapping->a_ops->bmap);
 	ASSERT(inode->i_mapping->a_ops->readpages);
 
 	/* calculate the shift required to use bmap */
@@ -728,7 +728,7 @@ int cachefiles_read_or_alloc_pages(struct fscache_retrieval *op,
 
 	ret = space ? -ENODATA : -ENOBUFS;
 	list_for_each_entry_safe(page, _n, pages, lru) {
-		sector_t block0, block;
+		sector_t block;
 
 		/* we assume the absence or presence of the first block is a
 		 * good enough indication for the page as a whole
@@ -736,13 +736,14 @@ int cachefiles_read_or_alloc_pages(struct fscache_retrieval *op,
 		 *   good enough for this as it doesn't indicate errors, but
 		 *   it's all we've got for the moment
 		 */
-		block0 = page->index;
-		block0 <<= shift;
+		block = page->index;
+		block <<= shift;
+
+		ret2 = bmap(inode, &block);
+		ASSERT(ret2 == 0);
 
-		block = inode->i_mapping->a_ops->bmap(inode->i_mapping,
-						      block0);
 		_debug("%llx -> %llx",
-		       (unsigned long long) block0,
+		       (unsigned long long) (page->index << shift),
 		       (unsigned long long) block);
 
 		if (block) {
diff --git a/fs/ceph/Makefile b/fs/ceph/Makefile
index c1da294418d1..0a0823d378db 100644
--- a/fs/ceph/Makefile
+++ b/fs/ceph/Makefile
@@ -8,7 +8,7 @@ obj-$(CONFIG_CEPH_FS) += ceph.o
 ceph-y := super.o inode.o dir.o file.o locks.o addr.o ioctl.o \
 	export.o caps.o snap.o xattr.o quota.o io.o \
 	mds_client.o mdsmap.o strings.o ceph_frag.o \
-	debugfs.o
+	debugfs.o util.o
 
 ceph-$(CONFIG_CEPH_FSCACHE) += cache.o
 ceph-$(CONFIG_CEPH_FS_POSIX_ACL) += acl.o
diff --git a/fs/ceph/acl.c b/fs/ceph/acl.c
index aa55f412a6e3..26be6520d3fb 100644
--- a/fs/ceph/acl.c
+++ b/fs/ceph/acl.c
@@ -222,8 +222,8 @@ int ceph_pre_init_acls(struct inode *dir, umode_t *mode,
 		err = ceph_pagelist_reserve(pagelist, len + val_size2 + 8);
 		if (err)
 			goto out_err;
-		err = ceph_pagelist_encode_string(pagelist,
-						  XATTR_NAME_POSIX_ACL_DEFAULT, len);
+		ceph_pagelist_encode_string(pagelist,
+					  XATTR_NAME_POSIX_ACL_DEFAULT, len);
 		err = posix_acl_to_xattr(&init_user_ns, default_acl,
 					 tmp_buf, val_size2);
 		if (err < 0)
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 7ab616601141..6f4678d98df7 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -159,8 +159,6 @@ static void ceph_invalidatepage(struct page *page, unsigned int offset,
 	if (!PagePrivate(page))
 		return;
 
-	ClearPageChecked(page);
-
 	dout("%p invalidatepage %p idx %lu full dirty page\n",
 	     inode, page, page->index);
 
@@ -183,6 +181,47 @@ static int ceph_releasepage(struct page *page, gfp_t g)
 }
 
 /*
+ * Read some contiguous pages.  If we cross a stripe boundary, shorten
+ * *plen.  Return number of bytes read, or error.
+ */
+static int ceph_sync_readpages(struct ceph_fs_client *fsc,
+			       struct ceph_vino vino,
+			       struct ceph_file_layout *layout,
+			       u64 off, u64 *plen,
+			       u32 truncate_seq, u64 truncate_size,
+			       struct page **pages, int num_pages,
+			       int page_align)
+{
+	struct ceph_osd_client *osdc = &fsc->client->osdc;
+	struct ceph_osd_request *req;
+	int rc = 0;
+
+	dout("readpages on ino %llx.%llx on %llu~%llu\n", vino.ino,
+	     vino.snap, off, *plen);
+	req = ceph_osdc_new_request(osdc, layout, vino, off, plen, 0, 1,
+				    CEPH_OSD_OP_READ, CEPH_OSD_FLAG_READ,
+				    NULL, truncate_seq, truncate_size,
+				    false);
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+
+	/* it may be a short read due to an object boundary */
+	osd_req_op_extent_osd_data_pages(req, 0,
+				pages, *plen, page_align, false, false);
+
+	dout("readpages  final extent is %llu~%llu (%llu bytes align %d)\n",
+	     off, *plen, *plen, page_align);
+
+	rc = ceph_osdc_start_request(osdc, req, false);
+	if (!rc)
+		rc = ceph_osdc_wait_request(osdc, req);
+
+	ceph_osdc_put_request(req);
+	dout("readpages result %d\n", rc);
+	return rc;
+}
+
+/*
  * read a single page, without unlocking it.
  */
 static int ceph_do_readpage(struct file *filp, struct page *page)
@@ -218,7 +257,7 @@ static int ceph_do_readpage(struct file *filp, struct page *page)
 
 	dout("readpage inode %p file %p page %p index %lu\n",
 	     inode, filp, page, page->index);
-	err = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode),
+	err = ceph_sync_readpages(fsc, ceph_vino(inode),
 				  &ci->i_layout, off, &len,
 				  ci->i_truncate_seq, ci->i_truncate_size,
 				  &page, 1, 0);
@@ -571,6 +610,47 @@ static u64 get_writepages_data_length(struct inode *inode,
 }
 
 /*
+ * do a synchronous write on N pages
+ */
+static int ceph_sync_writepages(struct ceph_fs_client *fsc,
+				struct ceph_vino vino,
+				struct ceph_file_layout *layout,
+				struct ceph_snap_context *snapc,
+				u64 off, u64 len,
+				u32 truncate_seq, u64 truncate_size,
+				struct timespec64 *mtime,
+				struct page **pages, int num_pages)
+{
+	struct ceph_osd_client *osdc = &fsc->client->osdc;
+	struct ceph_osd_request *req;
+	int rc = 0;
+	int page_align = off & ~PAGE_MASK;
+
+	req = ceph_osdc_new_request(osdc, layout, vino, off, &len, 0, 1,
+				    CEPH_OSD_OP_WRITE, CEPH_OSD_FLAG_WRITE,
+				    snapc, truncate_seq, truncate_size,
+				    true);
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+
+	/* it may be a short write due to an object boundary */
+	osd_req_op_extent_osd_data_pages(req, 0, pages, len, page_align,
+				false, false);
+	dout("writepages %llu~%llu (%llu bytes)\n", off, len, len);
+
+	req->r_mtime = *mtime;
+	rc = ceph_osdc_start_request(osdc, req, true);
+	if (!rc)
+		rc = ceph_osdc_wait_request(osdc, req);
+
+	ceph_osdc_put_request(req);
+	if (rc == 0)
+		rc = len;
+	dout("writepages result %d\n", rc);
+	return rc;
+}
+
+/*
  * Write a single page, but leave the page locked.
  *
  * If we get a write error, mark the mapping for error, but still adjust the
@@ -628,7 +708,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 		set_bdi_congested(inode_to_bdi(inode), BLK_RW_ASYNC);
 
 	set_page_writeback(page);
-	err = ceph_osdc_writepages(&fsc->client->osdc, ceph_vino(inode),
+	err = ceph_sync_writepages(fsc, ceph_vino(inode),
 				   &ci->i_layout, snapc, page_off, len,
 				   ceph_wbc.truncate_seq,
 				   ceph_wbc.truncate_size,
@@ -1575,7 +1655,7 @@ static vm_fault_t ceph_page_mkwrite(struct vm_fault *vmf)
 	do {
 		lock_page(page);
 
-		if ((off > size) || (page->mapping != inode->i_mapping)) {
+		if (page_mkwrite_check_truncate(page, inode) < 0) {
 			unlock_page(page);
 			ret = VM_FAULT_NOPAGE;
 			break;
diff --git a/fs/ceph/cache.c b/fs/ceph/cache.c
index b2ec29eeb4c4..2f5cb6bc78e1 100644
--- a/fs/ceph/cache.c
+++ b/fs/ceph/cache.c
@@ -8,6 +8,7 @@
 
 #include <linux/ceph/ceph_debug.h>
 
+#include <linux/fs_context.h>
 #include "super.h"
 #include "cache.h"
 
@@ -31,7 +32,7 @@ struct ceph_fscache_entry {
 	size_t uniq_len;
 	/* The following members must be last */
 	struct ceph_fsid fsid;
-	char uniquifier[0];
+	char uniquifier[];
 };
 
 static const struct fscache_cookie_def ceph_fscache_fsid_object_def = {
@@ -49,7 +50,7 @@ void ceph_fscache_unregister(void)
 	fscache_unregister_netfs(&ceph_cache_netfs);
 }
 
-int ceph_fscache_register_fs(struct ceph_fs_client* fsc)
+int ceph_fscache_register_fs(struct ceph_fs_client* fsc, struct fs_context *fc)
 {
 	const struct ceph_fsid *fsid = &fsc->client->fsid;
 	const char *fscache_uniq = fsc->mount_options->fscache_uniq;
@@ -66,8 +67,8 @@ int ceph_fscache_register_fs(struct ceph_fs_client* fsc)
 		if (uniq_len && memcmp(ent->uniquifier, fscache_uniq, uniq_len))
 			continue;
 
-		pr_err("fscache cookie already registered for fsid %pU\n", fsid);
-		pr_err("  use fsc=%%s mount option to specify a uniquifier\n");
+		errorfc(fc, "fscache cookie already registered for fsid %pU, use fsc=<uniquifier> option",
+		       fsid);
 		err = -EBUSY;
 		goto out_unlock;
 	}
@@ -95,7 +96,7 @@ int ceph_fscache_register_fs(struct ceph_fs_client* fsc)
 		list_add_tail(&ent->list, &ceph_fscache_list);
 	} else {
 		kfree(ent);
-		pr_err("unable to register fscache cookie for fsid %pU\n",
+		errorfc(fc, "unable to register fscache cookie for fsid %pU",
 		       fsid);
 		/* all other fs ignore this error */
 	}
diff --git a/fs/ceph/cache.h b/fs/ceph/cache.h
index e486fac3434d..89dbdd1eb14a 100644
--- a/fs/ceph/cache.h
+++ b/fs/ceph/cache.h
@@ -16,7 +16,7 @@ extern struct fscache_netfs ceph_cache_netfs;
 int ceph_fscache_register(void);
 void ceph_fscache_unregister(void);
 
-int ceph_fscache_register_fs(struct ceph_fs_client* fsc);
+int ceph_fscache_register_fs(struct ceph_fs_client* fsc, struct fs_context *fc);
 void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc);
 
 void ceph_fscache_register_inode_cookie(struct inode *inode);
@@ -88,7 +88,8 @@ static inline void ceph_fscache_unregister(void)
 {
 }
 
-static inline int ceph_fscache_register_fs(struct ceph_fs_client* fsc)
+static inline int ceph_fscache_register_fs(struct ceph_fs_client* fsc,
+					   struct fs_context *fc)
 {
 	return 0;
 }
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index f5a38910a82b..f1acde6fb9a6 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -490,13 +490,10 @@ static void __cap_set_timeouts(struct ceph_mds_client *mdsc,
 			       struct ceph_inode_info *ci)
 {
 	struct ceph_mount_options *opt = mdsc->fsc->mount_options;
-
-	ci->i_hold_caps_min = round_jiffies(jiffies +
-					    opt->caps_wanted_delay_min * HZ);
 	ci->i_hold_caps_max = round_jiffies(jiffies +
 					    opt->caps_wanted_delay_max * HZ);
-	dout("__cap_set_timeouts %p min %lu max %lu\n", &ci->vfs_inode,
-	     ci->i_hold_caps_min - jiffies, ci->i_hold_caps_max - jiffies);
+	dout("__cap_set_timeouts %p %lu\n", &ci->vfs_inode,
+	     ci->i_hold_caps_max - jiffies);
 }
 
 /*
@@ -508,10 +505,9 @@ static void __cap_set_timeouts(struct ceph_mds_client *mdsc,
  *    -> we take mdsc->cap_delay_lock
  */
 static void __cap_delay_requeue(struct ceph_mds_client *mdsc,
-				struct ceph_inode_info *ci,
-				bool set_timeout)
+				struct ceph_inode_info *ci)
 {
-	dout("__cap_delay_requeue %p flags %d at %lu\n", &ci->vfs_inode,
+	dout("__cap_delay_requeue %p flags 0x%lx at %lu\n", &ci->vfs_inode,
 	     ci->i_ceph_flags, ci->i_hold_caps_max);
 	if (!mdsc->stopping) {
 		spin_lock(&mdsc->cap_delay_lock);
@@ -520,8 +516,7 @@ static void __cap_delay_requeue(struct ceph_mds_client *mdsc,
 				goto no_change;
 			list_del_init(&ci->i_cap_delay_list);
 		}
-		if (set_timeout)
-			__cap_set_timeouts(mdsc, ci);
+		__cap_set_timeouts(mdsc, ci);
 		list_add_tail(&ci->i_cap_delay_list, &mdsc->cap_delay_list);
 no_change:
 		spin_unlock(&mdsc->cap_delay_lock);
@@ -561,19 +556,20 @@ static void __cap_delay_cancel(struct ceph_mds_client *mdsc,
 	spin_unlock(&mdsc->cap_delay_lock);
 }
 
-/*
- * Common issue checks for add_cap, handle_cap_grant.
- */
+/* Common issue checks for add_cap, handle_cap_grant. */
 static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap,
 			      unsigned issued)
 {
 	unsigned had = __ceph_caps_issued(ci, NULL);
 
+	lockdep_assert_held(&ci->i_ceph_lock);
+
 	/*
 	 * Each time we receive FILE_CACHE anew, we increment
 	 * i_rdcache_gen.
 	 */
-	if ((issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) &&
+	if (S_ISREG(ci->vfs_inode.i_mode) &&
+	    (issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) &&
 	    (had & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0) {
 		ci->i_rdcache_gen++;
 	}
@@ -592,6 +588,13 @@ static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap,
 			__ceph_dir_clear_complete(ci);
 		}
 	}
+
+	/* Wipe saved layout if we're losing DIR_CREATE caps */
+	if (S_ISDIR(ci->vfs_inode.i_mode) && (had & CEPH_CAP_DIR_CREATE) &&
+		!(issued & CEPH_CAP_DIR_CREATE)) {
+	     ceph_put_string(rcu_dereference_raw(ci->i_cached_layout.pool_ns));
+	     memset(&ci->i_cached_layout, 0, sizeof(ci->i_cached_layout));
+	}
 }
 
 /*
@@ -605,7 +608,7 @@ static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap,
  */
 void ceph_add_cap(struct inode *inode,
 		  struct ceph_mds_session *session, u64 cap_id,
-		  int fmode, unsigned issued, unsigned wanted,
+		  unsigned issued, unsigned wanted,
 		  unsigned seq, unsigned mseq, u64 realmino, int flags,
 		  struct ceph_cap **new_cap)
 {
@@ -621,13 +624,6 @@ void ceph_add_cap(struct inode *inode,
 	dout("add_cap %p mds%d cap %llx %s seq %d\n", inode,
 	     session->s_mds, cap_id, ceph_cap_string(issued), seq);
 
-	/*
-	 * If we are opening the file, include file mode wanted bits
-	 * in wanted.
-	 */
-	if (fmode >= 0)
-		wanted |= ceph_caps_for_mode(fmode);
-
 	spin_lock(&session->s_gen_ttl_lock);
 	gen = session->s_cap_gen;
 	spin_unlock(&session->s_gen_ttl_lock);
@@ -725,7 +721,7 @@ void ceph_add_cap(struct inode *inode,
 		dout(" issued %s, mds wanted %s, actual %s, queueing\n",
 		     ceph_cap_string(issued), ceph_cap_string(wanted),
 		     ceph_cap_string(actual_wanted));
-		__cap_delay_requeue(mdsc, ci, true);
+		__cap_delay_requeue(mdsc, ci);
 	}
 
 	if (flags & CEPH_CAP_FLAG_AUTH) {
@@ -752,9 +748,6 @@ void ceph_add_cap(struct inode *inode,
 	cap->issue_seq = seq;
 	cap->mseq = mseq;
 	cap->cap_gen = gen;
-
-	if (fmode >= 0)
-		__ceph_get_fmode(ci, fmode);
 }
 
 /*
@@ -908,7 +901,8 @@ int __ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch)
 						       ci_node);
 					if (!__cap_is_valid(cap))
 						continue;
-					__touch_cap(cap);
+					if (cap->issued & mask)
+						__touch_cap(cap);
 				}
 			}
 			return 1;
@@ -957,29 +951,97 @@ int __ceph_caps_used(struct ceph_inode_info *ci)
 	if (ci->i_rd_ref)
 		used |= CEPH_CAP_FILE_RD;
 	if (ci->i_rdcache_ref ||
-	    (!S_ISDIR(ci->vfs_inode.i_mode) && /* ignore readdir cache */
+	    (S_ISREG(ci->vfs_inode.i_mode) &&
 	     ci->vfs_inode.i_data.nrpages))
 		used |= CEPH_CAP_FILE_CACHE;
 	if (ci->i_wr_ref)
 		used |= CEPH_CAP_FILE_WR;
 	if (ci->i_wb_ref || ci->i_wrbuffer_ref)
 		used |= CEPH_CAP_FILE_BUFFER;
+	if (ci->i_fx_ref)
+		used |= CEPH_CAP_FILE_EXCL;
 	return used;
 }
 
+#define FMODE_WAIT_BIAS 1000
+
 /*
  * wanted, by virtue of open file modes
  */
 int __ceph_caps_file_wanted(struct ceph_inode_info *ci)
 {
-	int i, bits = 0;
-	for (i = 0; i < CEPH_FILE_MODE_BITS; i++) {
-		if (ci->i_nr_by_mode[i])
-			bits |= 1 << i;
+	const int PIN_SHIFT = ffs(CEPH_FILE_MODE_PIN);
+	const int RD_SHIFT = ffs(CEPH_FILE_MODE_RD);
+	const int WR_SHIFT = ffs(CEPH_FILE_MODE_WR);
+	const int LAZY_SHIFT = ffs(CEPH_FILE_MODE_LAZY);
+	struct ceph_mount_options *opt =
+		ceph_inode_to_client(&ci->vfs_inode)->mount_options;
+	unsigned long used_cutoff = jiffies - opt->caps_wanted_delay_max * HZ;
+	unsigned long idle_cutoff = jiffies - opt->caps_wanted_delay_min * HZ;
+
+	if (S_ISDIR(ci->vfs_inode.i_mode)) {
+		int want = 0;
+
+		/* use used_cutoff here, to keep dir's wanted caps longer */
+		if (ci->i_nr_by_mode[RD_SHIFT] > 0 ||
+		    time_after(ci->i_last_rd, used_cutoff))
+			want |= CEPH_CAP_ANY_SHARED;
+
+		if (ci->i_nr_by_mode[WR_SHIFT] > 0 ||
+		    time_after(ci->i_last_wr, used_cutoff)) {
+			want |= CEPH_CAP_ANY_SHARED | CEPH_CAP_FILE_EXCL;
+			if (opt->flags & CEPH_MOUNT_OPT_ASYNC_DIROPS)
+				want |= CEPH_CAP_ANY_DIR_OPS;
+		}
+
+		if (want || ci->i_nr_by_mode[PIN_SHIFT] > 0)
+			want |= CEPH_CAP_PIN;
+
+		return want;
+	} else {
+		int bits = 0;
+
+		if (ci->i_nr_by_mode[RD_SHIFT] > 0) {
+			if (ci->i_nr_by_mode[RD_SHIFT] >= FMODE_WAIT_BIAS ||
+			    time_after(ci->i_last_rd, used_cutoff))
+				bits |= 1 << RD_SHIFT;
+		} else if (time_after(ci->i_last_rd, idle_cutoff)) {
+			bits |= 1 << RD_SHIFT;
+		}
+
+		if (ci->i_nr_by_mode[WR_SHIFT] > 0) {
+			if (ci->i_nr_by_mode[WR_SHIFT] >= FMODE_WAIT_BIAS ||
+			    time_after(ci->i_last_wr, used_cutoff))
+				bits |= 1 << WR_SHIFT;
+		} else if (time_after(ci->i_last_wr, idle_cutoff)) {
+			bits |= 1 << WR_SHIFT;
+		}
+
+		/* check lazyio only when read/write is wanted */
+		if ((bits & (CEPH_FILE_MODE_RDWR << 1)) &&
+		    ci->i_nr_by_mode[LAZY_SHIFT] > 0)
+			bits |= 1 << LAZY_SHIFT;
+
+		return bits ? ceph_caps_for_mode(bits >> 1) : 0;
 	}
-	if (bits == 0)
-		return 0;
-	return ceph_caps_for_mode(bits >> 1);
+}
+
+/*
+ * wanted, by virtue of open file modes AND cap refs (buffered/cached data)
+ */
+int __ceph_caps_wanted(struct ceph_inode_info *ci)
+{
+	int w = __ceph_caps_file_wanted(ci) | __ceph_caps_used(ci);
+	if (S_ISDIR(ci->vfs_inode.i_mode)) {
+		/* we want EXCL if holding caps of dir ops */
+		if (w & CEPH_CAP_ANY_DIR_OPS)
+			w |= CEPH_CAP_FILE_EXCL;
+	} else {
+		/* we want EXCL if dirty data */
+		if (w & CEPH_CAP_FILE_BUFFER)
+			w |= CEPH_CAP_FILE_EXCL;
+	}
+	return w;
 }
 
 /*
@@ -1003,26 +1065,13 @@ int __ceph_caps_mds_wanted(struct ceph_inode_info *ci, bool check)
 	return mds_wanted;
 }
 
-/*
- * called under i_ceph_lock
- */
-static int __ceph_is_single_caps(struct ceph_inode_info *ci)
-{
-	return rb_first(&ci->i_caps) == rb_last(&ci->i_caps);
-}
-
-static int __ceph_is_any_caps(struct ceph_inode_info *ci)
-{
-	return !RB_EMPTY_ROOT(&ci->i_caps);
-}
-
 int ceph_is_any_caps(struct inode *inode)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int ret;
 
 	spin_lock(&ci->i_ceph_lock);
-	ret = __ceph_is_any_caps(ci);
+	ret = __ceph_is_any_real_caps(ci);
 	spin_unlock(&ci->i_ceph_lock);
 
 	return ret;
@@ -1099,15 +1148,16 @@ void __ceph_remove_cap(struct ceph_cap *cap, bool queue_release)
 	if (removed)
 		ceph_put_cap(mdsc, cap);
 
-	/* when reconnect denied, we remove session caps forcibly,
-	 * i_wr_ref can be non-zero. If there are ongoing write,
-	 * keep i_snap_realm.
-	 */
-	if (!__ceph_is_any_caps(ci) && ci->i_wr_ref == 0 && ci->i_snap_realm)
-		drop_inode_snap_realm(ci);
+	if (!__ceph_is_any_real_caps(ci)) {
+		/* when reconnect denied, we remove session caps forcibly,
+		 * i_wr_ref can be non-zero. If there are ongoing write,
+		 * keep i_snap_realm.
+		 */
+		if (ci->i_wr_ref == 0 && ci->i_snap_realm)
+			drop_inode_snap_realm(ci);
 
-	if (!__ceph_is_any_real_caps(ci))
 		__cap_delay_cancel(mdsc, ci);
+	}
 }
 
 struct cap_msg_args {
@@ -1277,9 +1327,15 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 	struct cap_msg_args arg;
 	int held, revoking;
 	int wake = 0;
-	int delayed = 0;
 	int ret;
 
+	/* Don't send anything if it's still being created. Return delayed */
+	if (ci->i_ceph_flags & CEPH_I_ASYNC_CREATE) {
+		spin_unlock(&ci->i_ceph_lock);
+		dout("%s async create in flight for %p\n", __func__, inode);
+		return 1;
+	}
+
 	held = cap->issued | cap->implemented;
 	revoking = cap->implemented & ~cap->issued;
 	retain &= ~revoking;
@@ -1290,28 +1346,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 	     ceph_cap_string(revoking));
 	BUG_ON((retain & CEPH_CAP_PIN) == 0);
 
-	arg.session = cap->session;
-
-	/* don't release wanted unless we've waited a bit. */
-	if ((ci->i_ceph_flags & CEPH_I_NODELAY) == 0 &&
-	    time_before(jiffies, ci->i_hold_caps_min)) {
-		dout(" delaying issued %s -> %s, wanted %s -> %s on send\n",
-		     ceph_cap_string(cap->issued),
-		     ceph_cap_string(cap->issued & retain),
-		     ceph_cap_string(cap->mds_wanted),
-		     ceph_cap_string(want));
-		want |= cap->mds_wanted;
-		retain |= cap->issued;
-		delayed = 1;
-	}
-	ci->i_ceph_flags &= ~(CEPH_I_NODELAY | CEPH_I_FLUSH);
-	if (want & ~cap->mds_wanted) {
-		/* user space may open/close single file frequently.
-		 * This avoids droping mds_wanted immediately after
-		 * requesting new mds_wanted.
-		 */
-		__cap_set_timeouts(mdsc, ci);
-	}
+	ci->i_ceph_flags &= ~CEPH_I_FLUSH;
 
 	cap->issued &= retain;  /* drop bits we don't want */
 	if (cap->implemented & ~cap->issued) {
@@ -1326,6 +1361,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 	cap->implemented &= cap->issued | used;
 	cap->mds_wanted = want;
 
+	arg.session = cap->session;
 	arg.ino = ceph_vino(inode).ino;
 	arg.cid = cap->cap_id;
 	arg.follows = flushing ? ci->i_head_snapc->seq : 0;
@@ -1335,7 +1371,8 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 	arg.size = inode->i_size;
 	ci->i_reported_size = arg.size;
 	arg.max_size = ci->i_wanted_max_size;
-	ci->i_requested_max_size = arg.max_size;
+	if (cap == ci->i_auth_cap)
+		ci->i_requested_max_size = arg.max_size;
 
 	if (flushing & CEPH_CAP_XATTR_EXCL) {
 		old_blob = __ceph_build_xattrs_blob(ci);
@@ -1386,14 +1423,19 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 
 	ret = send_cap_msg(&arg);
 	if (ret < 0) {
-		dout("error sending cap msg, must requeue %p\n", inode);
-		delayed = 1;
+		pr_err("error sending cap msg, ino (%llx.%llx) "
+		       "flushing %s tid %llu, requeue\n",
+		       ceph_vinop(inode), ceph_cap_string(flushing),
+		       flush_tid);
+		spin_lock(&ci->i_ceph_lock);
+		__cap_delay_requeue(mdsc, ci);
+		spin_unlock(&ci->i_ceph_lock);
 	}
 
 	if (wake)
 		wake_up_all(&ci->i_cap_wq);
 
-	return delayed;
+	return ret;
 }
 
 static inline int __send_flush_snap(struct inode *inode,
@@ -1620,6 +1662,8 @@ int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask,
 	int was = ci->i_dirty_caps;
 	int dirty = 0;
 
+	lockdep_assert_held(&ci->i_ceph_lock);
+
 	if (!ci->i_auth_cap) {
 		pr_warn("__mark_dirty_caps %p %llx mask %s, "
 			"but no auth cap (session was closed?)\n",
@@ -1657,7 +1701,7 @@ int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask,
 	if (((was | ci->i_flushing_caps) & CEPH_CAP_FILE_BUFFER) &&
 	    (mask & CEPH_CAP_FILE_BUFFER))
 		dirty |= I_DIRTY_DATASYNC;
-	__cap_delay_requeue(mdsc, ci, true);
+	__cap_delay_requeue(mdsc, ci);
 	return dirty;
 }
 
@@ -1729,6 +1773,7 @@ static u64 __mark_caps_flushing(struct inode *inode,
 	struct ceph_cap_flush *cf = NULL;
 	int flushing;
 
+	lockdep_assert_held(&ci->i_ceph_lock);
 	BUG_ON(ci->i_dirty_caps == 0);
 	BUG_ON(list_empty(&ci->i_dirty_item));
 	BUG_ON(!ci->i_prealloc_cap_flush);
@@ -1808,8 +1853,6 @@ bool __ceph_should_report_size(struct ceph_inode_info *ci)
  * versus held caps.  Release, flush, ack revoked caps to mds as
  * appropriate.
  *
- *  CHECK_CAPS_NODELAY - caller is delayed work and we should not delay
- *    cap release further.
  *  CHECK_CAPS_AUTHONLY - we should only check the auth cap
  *  CHECK_CAPS_FLUSH - we should flush any dirty caps immediately, without
  *    further delay.
@@ -1828,24 +1871,13 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags,
 	int mds = -1;   /* keep track of how far we've gone through i_caps list
 			   to avoid an infinite loop on retry */
 	struct rb_node *p;
-	int delayed = 0, sent = 0;
-	bool no_delay = flags & CHECK_CAPS_NODELAY;
 	bool queue_invalidate = false;
 	bool tried_invalidate = false;
 
-	/* if we are unmounting, flush any unused caps immediately. */
-	if (mdsc->stopping)
-		no_delay = true;
-
 	spin_lock(&ci->i_ceph_lock);
-
 	if (ci->i_ceph_flags & CEPH_I_FLUSH)
 		flags |= CHECK_CAPS_FLUSH;
 
-	if (!(flags & CHECK_CAPS_AUTHONLY) ||
-	    (ci->i_auth_cap && __ceph_is_single_caps(ci)))
-		__cap_delay_cancel(mdsc, ci);
-
 	goto retry_locked;
 retry:
 	spin_lock(&ci->i_ceph_lock);
@@ -1869,10 +1901,11 @@ retry_locked:
 			 * revoking the shared cap on every create/unlink
 			 * operation.
 			 */
-			if (IS_RDONLY(inode))
+			if (IS_RDONLY(inode)) {
 				want = CEPH_CAP_ANY_SHARED;
-			else
-				want = CEPH_CAP_ANY_SHARED | CEPH_CAP_FILE_EXCL;
+			} else {
+				want |= CEPH_CAP_ANY_SHARED | CEPH_CAP_FILE_EXCL;
+			}
 			retain |= want;
 		} else {
 
@@ -1888,14 +1921,13 @@ retry_locked:
 	}
 
 	dout("check_caps %p file_want %s used %s dirty %s flushing %s"
-	     " issued %s revoking %s retain %s %s%s%s\n", inode,
+	     " issued %s revoking %s retain %s %s%s\n", inode,
 	     ceph_cap_string(file_wanted),
 	     ceph_cap_string(used), ceph_cap_string(ci->i_dirty_caps),
 	     ceph_cap_string(ci->i_flushing_caps),
 	     ceph_cap_string(issued), ceph_cap_string(revoking),
 	     ceph_cap_string(retain),
 	     (flags & CHECK_CAPS_AUTHONLY) ? " AUTHONLY" : "",
-	     (flags & CHECK_CAPS_NODELAY) ? " NODELAY" : "",
 	     (flags & CHECK_CAPS_FLUSH) ? " FLUSH" : "");
 
 	/*
@@ -1903,8 +1935,8 @@ retry_locked:
 	 * have cached pages, but don't want them, then try to invalidate.
 	 * If we fail, it's because pages are locked.... try again later.
 	 */
-	if ((!no_delay || mdsc->stopping) &&
-	    !S_ISDIR(inode->i_mode) &&		/* ignore readdir cache */
+	if ((!(flags & CHECK_CAPS_NOINVAL) || mdsc->stopping) &&
+	    S_ISREG(inode->i_mode) &&
 	    !(ci->i_wb_ref || ci->i_wrbuffer_ref) &&   /* no dirty pages... */
 	    inode->i_data.nrpages &&		/* have cached pages */
 	    (revoking & (CEPH_CAP_FILE_CACHE|
@@ -1976,28 +2008,17 @@ retry_locked:
 		}
 
 		/* want more caps from mds? */
-		if (want & ~(cap->mds_wanted | cap->issued))
-			goto ack;
+		if (want & ~cap->mds_wanted) {
+			if (want & ~(cap->mds_wanted | cap->issued))
+				goto ack;
+			if (!__cap_is_valid(cap))
+				goto ack;
+		}
 
 		/* things we might delay */
 		if ((cap->issued & ~retain) == 0)
 			continue;     /* nope, all good */
 
-		if (no_delay)
-			goto ack;
-
-		/* delay? */
-		if ((ci->i_ceph_flags & CEPH_I_NODELAY) == 0 &&
-		    time_before(jiffies, ci->i_hold_caps_max)) {
-			dout(" delaying issued %s -> %s, wanted %s -> %s\n",
-			     ceph_cap_string(cap->issued),
-			     ceph_cap_string(cap->issued & retain),
-			     ceph_cap_string(cap->mds_wanted),
-			     ceph_cap_string(want));
-			delayed++;
-			continue;
-		}
-
 ack:
 		if (session && session != cap->session) {
 			dout("oops, wrong session %p mutex\n", session);
@@ -2058,18 +2079,20 @@ ack:
 		}
 
 		mds = cap->mds;  /* remember mds, so we don't repeat */
-		sent++;
 
 		/* __send_cap drops i_ceph_lock */
-		delayed += __send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, 0,
-				cap_used, want, retain, flushing,
-				flush_tid, oldest_flush_tid);
+		__send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, 0, cap_used, want,
+			   retain, flushing, flush_tid, oldest_flush_tid);
 		goto retry; /* retake i_ceph_lock and restart our cap scan. */
 	}
 
-	/* Reschedule delayed caps release if we delayed anything */
-	if (delayed)
-		__cap_delay_requeue(mdsc, ci, false);
+	/* periodically re-calculate caps wanted by open files */
+	if (__ceph_is_any_real_caps(ci) &&
+	    list_empty(&ci->i_cap_delay_list) &&
+	    (file_wanted & ~CEPH_CAP_PIN) &&
+	    !(used & (CEPH_CAP_FILE_RD | CEPH_CAP_ANY_FILE_WR))) {
+		__cap_delay_requeue(mdsc, ci);
+	}
 
 	spin_unlock(&ci->i_ceph_lock);
 
@@ -2098,7 +2121,6 @@ retry:
 retry_locked:
 	if (ci->i_dirty_caps && ci->i_auth_cap) {
 		struct ceph_cap *cap = ci->i_auth_cap;
-		int delayed;
 
 		if (session != cap->session) {
 			spin_unlock(&ci->i_ceph_lock);
@@ -2127,18 +2149,10 @@ retry_locked:
 						 &oldest_flush_tid);
 
 		/* __send_cap drops i_ceph_lock */
-		delayed = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH,
-				     CEPH_CLIENT_CAPS_SYNC,
-				     __ceph_caps_used(ci),
-				     __ceph_caps_wanted(ci),
-				     (cap->issued | cap->implemented),
-				     flushing, flush_tid, oldest_flush_tid);
-
-		if (delayed) {
-			spin_lock(&ci->i_ceph_lock);
-			__cap_delay_requeue(mdsc, ci, true);
-			spin_unlock(&ci->i_ceph_lock);
-		}
+		__send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH, CEPH_CLIENT_CAPS_SYNC,
+			   __ceph_caps_used(ci), __ceph_caps_wanted(ci),
+			   (cap->issued | cap->implemented),
+			   flushing, flush_tid, oldest_flush_tid);
 	} else {
 		if (!list_empty(&ci->i_cap_flush_list)) {
 			struct ceph_cap_flush *cf =
@@ -2236,6 +2250,10 @@ int ceph_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 	if (datasync)
 		goto out;
 
+	ret = ceph_wait_on_async_create(inode);
+	if (ret)
+		goto out;
+
 	dirty = try_flush_caps(inode, &flush_tid);
 	dout("fsync dirty caps are %s\n", ceph_cap_string(dirty));
 
@@ -2338,22 +2356,13 @@ static void __kick_flushing_caps(struct ceph_mds_client *mdsc,
 		if (cf->caps) {
 			dout("kick_flushing_caps %p cap %p tid %llu %s\n",
 			     inode, cap, cf->tid, ceph_cap_string(cf->caps));
-			ci->i_ceph_flags |= CEPH_I_NODELAY;
-
-			ret = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH,
+			__send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH,
 					 (cf->tid < last_snap_flush ?
 					  CEPH_CLIENT_CAPS_PENDING_CAPSNAP : 0),
 					  __ceph_caps_used(ci),
 					  __ceph_caps_wanted(ci),
 					  (cap->issued | cap->implemented),
 					  cf->caps, cf->tid, oldest_flush_tid);
-			if (ret) {
-				pr_err("kick_flushing_caps: error sending "
-					"cap flush, ino (%llx.%llx) "
-					"tid %llu flushing %s\n",
-					ceph_vinop(inode), cf->tid,
-					ceph_cap_string(cf->caps));
-			}
 		} else {
 			struct ceph_cap_snap *capsnap =
 					container_of(cf, struct ceph_cap_snap,
@@ -2460,16 +2469,15 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc,
 	}
 }
 
-static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc,
-				     struct ceph_mds_session *session,
-				     struct inode *inode)
-	__releases(ci->i_ceph_lock)
+void ceph_kick_flushing_inode_caps(struct ceph_mds_session *session,
+				   struct ceph_inode_info *ci)
 {
-	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_cap *cap;
+	struct ceph_mds_client *mdsc = session->s_mdsc;
+	struct ceph_cap *cap = ci->i_auth_cap;
+
+	lockdep_assert_held(&ci->i_ceph_lock);
 
-	cap = ci->i_auth_cap;
-	dout("kick_flushing_inode_caps %p flushing %s\n", inode,
+	dout("%s %p flushing %s\n", __func__, &ci->vfs_inode,
 	     ceph_cap_string(ci->i_flushing_caps));
 
 	if (!list_empty(&ci->i_cap_flush_list)) {
@@ -2481,9 +2489,6 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc,
 		spin_unlock(&mdsc->cap_dirty_lock);
 
 		__kick_flushing_caps(mdsc, session, ci, oldest_flush_tid);
-		spin_unlock(&ci->i_ceph_lock);
-	} else {
-		spin_unlock(&ci->i_ceph_lock);
 	}
 }
 
@@ -2491,18 +2496,20 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc,
 /*
  * Take references to capabilities we hold, so that we don't release
  * them to the MDS prematurely.
- *
- * Protected by i_ceph_lock.
  */
-static void __take_cap_refs(struct ceph_inode_info *ci, int got,
+void ceph_take_cap_refs(struct ceph_inode_info *ci, int got,
 			    bool snap_rwsem_locked)
 {
+	lockdep_assert_held(&ci->i_ceph_lock);
+
 	if (got & CEPH_CAP_PIN)
 		ci->i_pin_ref++;
 	if (got & CEPH_CAP_FILE_RD)
 		ci->i_rd_ref++;
 	if (got & CEPH_CAP_FILE_CACHE)
 		ci->i_rdcache_ref++;
+	if (got & CEPH_CAP_FILE_EXCL)
+		ci->i_fx_ref++;
 	if (got & CEPH_CAP_FILE_WR) {
 		if (ci->i_wr_ref == 0 && !ci->i_head_snapc) {
 			BUG_ON(!snap_rwsem_locked);
@@ -2515,7 +2522,7 @@ static void __take_cap_refs(struct ceph_inode_info *ci, int got,
 		if (ci->i_wb_ref == 0)
 			ihold(&ci->vfs_inode);
 		ci->i_wb_ref++;
-		dout("__take_cap_refs %p wb %d -> %d (?)\n",
+		dout("%s %p wb %d -> %d (?)\n", __func__,
 		     &ci->vfs_inode, ci->i_wb_ref-1, ci->i_wb_ref);
 	}
 }
@@ -2527,14 +2534,16 @@ static void __take_cap_refs(struct ceph_inode_info *ci, int got,
  * Note that caller is responsible for ensuring max_size increases are
  * requested from the MDS.
  *
- * Returns 0 if caps were not able to be acquired (yet), a 1 if they were,
- * or a negative error code.
- *
- * FIXME: how does a 0 return differ from -EAGAIN?
+ * Returns 0 if caps were not able to be acquired (yet), 1 if succeed,
+ * or a negative error code. There are 3 speical error codes:
+ *  -EAGAIN: need to sleep but non-blocking is specified
+ *  -EFBIG:  ask caller to call check_max_size() and try again.
+ *  -ESTALE: ask caller to call ceph_renew_caps() and try again.
  */
 enum {
-	NON_BLOCKING	= 1,
-	CHECK_FILELOCK	= 2,
+	/* first 8 bits are reserved for CEPH_FILE_MODE_FOO */
+	NON_BLOCKING	= (1 << 8),
+	CHECK_FILELOCK	= (1 << 9),
 };
 
 static int try_get_cap_refs(struct inode *inode, int need, int want,
@@ -2544,7 +2553,6 @@ static int try_get_cap_refs(struct inode *inode, int need, int want,
 	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	int ret = 0;
 	int have, implemented;
-	int file_wanted;
 	bool snap_rwsem_locked = false;
 
 	dout("get_cap_refs %p need %s want %s\n", inode,
@@ -2560,15 +2568,6 @@ again:
 		goto out_unlock;
 	}
 
-	/* make sure file is actually open */
-	file_wanted = __ceph_caps_file_wanted(ci);
-	if ((file_wanted & need) != need) {
-		dout("try_get_cap_refs need %s file_wanted %s, EBADF\n",
-		     ceph_cap_string(need), ceph_cap_string(file_wanted));
-		ret = -EBADF;
-		goto out_unlock;
-	}
-
 	/* finish pending truncate */
 	while (ci->i_truncate_pending) {
 		spin_unlock(&ci->i_ceph_lock);
@@ -2587,7 +2586,7 @@ again:
 			dout("get_cap_refs %p endoff %llu > maxsize %llu\n",
 			     inode, endoff, ci->i_max_size);
 			if (endoff > ci->i_requested_max_size)
-				ret = -EAGAIN;
+				ret = ci->i_auth_cap ? -EFBIG : -ESTALE;
 			goto out_unlock;
 		}
 		/*
@@ -2633,51 +2632,55 @@ again:
 				}
 				snap_rwsem_locked = true;
 			}
-			*got = need | (have & want);
-			if ((need & CEPH_CAP_FILE_RD) &&
+			if ((have & want) == want)
+				*got = need | want;
+			else
+				*got = need;
+			if (S_ISREG(inode->i_mode) &&
+			    (need & CEPH_CAP_FILE_RD) &&
 			    !(*got & CEPH_CAP_FILE_CACHE))
 				ceph_disable_fscache_readpage(ci);
-			__take_cap_refs(ci, *got, true);
+			ceph_take_cap_refs(ci, *got, true);
 			ret = 1;
 		}
 	} else {
 		int session_readonly = false;
-		if ((need & CEPH_CAP_FILE_WR) && ci->i_auth_cap) {
+		int mds_wanted;
+		if (ci->i_auth_cap &&
+		    (need & (CEPH_CAP_FILE_WR | CEPH_CAP_FILE_EXCL))) {
 			struct ceph_mds_session *s = ci->i_auth_cap->session;
 			spin_lock(&s->s_cap_lock);
 			session_readonly = s->s_readonly;
 			spin_unlock(&s->s_cap_lock);
 		}
 		if (session_readonly) {
-			dout("get_cap_refs %p needed %s but mds%d readonly\n",
+			dout("get_cap_refs %p need %s but mds%d readonly\n",
 			     inode, ceph_cap_string(need), ci->i_auth_cap->mds);
 			ret = -EROFS;
 			goto out_unlock;
 		}
 
-		if (ci->i_ceph_flags & CEPH_I_CAP_DROPPED) {
-			int mds_wanted;
-			if (READ_ONCE(mdsc->fsc->mount_state) ==
-			    CEPH_MOUNT_SHUTDOWN) {
-				dout("get_cap_refs %p forced umount\n", inode);
-				ret = -EIO;
-				goto out_unlock;
-			}
-			mds_wanted = __ceph_caps_mds_wanted(ci, false);
-			if (need & ~(mds_wanted & need)) {
-				dout("get_cap_refs %p caps were dropped"
-				     " (session killed?)\n", inode);
-				ret = -ESTALE;
-				goto out_unlock;
-			}
-			if (!(file_wanted & ~mds_wanted))
-				ci->i_ceph_flags &= ~CEPH_I_CAP_DROPPED;
+		if (READ_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_SHUTDOWN) {
+			dout("get_cap_refs %p forced umount\n", inode);
+			ret = -EIO;
+			goto out_unlock;
+		}
+		mds_wanted = __ceph_caps_mds_wanted(ci, false);
+		if (need & ~mds_wanted) {
+			dout("get_cap_refs %p need %s > mds_wanted %s\n",
+			     inode, ceph_cap_string(need),
+			     ceph_cap_string(mds_wanted));
+			ret = -ESTALE;
+			goto out_unlock;
 		}
 
-		dout("get_cap_refs %p have %s needed %s\n", inode,
+		dout("get_cap_refs %p have %s need %s\n", inode,
 		     ceph_cap_string(have), ceph_cap_string(need));
 	}
 out_unlock:
+
+	__ceph_touch_fmode(ci, mdsc, flags);
+
 	spin_unlock(&ci->i_ceph_lock);
 	if (snap_rwsem_locked)
 		up_read(&mdsc->snap_rwsem);
@@ -2715,20 +2718,40 @@ static void check_max_size(struct inode *inode, loff_t endoff)
 		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
 }
 
+static inline int get_used_fmode(int caps)
+{
+	int fmode = 0;
+	if (caps & CEPH_CAP_FILE_RD)
+		fmode |= CEPH_FILE_MODE_RD;
+	if (caps & CEPH_CAP_FILE_WR)
+		fmode |= CEPH_FILE_MODE_WR;
+	return fmode;
+}
+
 int ceph_try_get_caps(struct inode *inode, int need, int want,
 		      bool nonblock, int *got)
 {
-	int ret;
+	int ret, flags;
 
 	BUG_ON(need & ~CEPH_CAP_FILE_RD);
-	BUG_ON(want & ~(CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO|CEPH_CAP_FILE_SHARED));
-	ret = ceph_pool_perm_check(inode, need);
-	if (ret < 0)
-		return ret;
+	BUG_ON(want & ~(CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO |
+			CEPH_CAP_FILE_SHARED | CEPH_CAP_FILE_EXCL |
+			CEPH_CAP_ANY_DIR_OPS));
+	if (need) {
+		ret = ceph_pool_perm_check(inode, need);
+		if (ret < 0)
+			return ret;
+	}
+
+	flags = get_used_fmode(need | want);
+	if (nonblock)
+		flags |= NON_BLOCKING;
 
-	ret = try_get_cap_refs(inode, need, want, 0,
-			       (nonblock ? NON_BLOCKING : 0), got);
-	return ret == -EAGAIN ? 0 : ret;
+	ret = try_get_cap_refs(inode, need, want, 0, flags, got);
+	/* three special error codes */
+	if (ret == -EAGAIN || ret == -EFBIG || ret == -ESTALE)
+		ret = 0;
+	return ret;
 }
 
 /*
@@ -2753,18 +2776,32 @@ int ceph_get_caps(struct file *filp, int need, int want,
 	    fi->filp_gen != READ_ONCE(fsc->filp_gen))
 		return -EBADF;
 
-	while (true) {
-		if (endoff > 0)
-			check_max_size(inode, endoff);
+	flags = get_used_fmode(need | want);
 
-		flags = atomic_read(&fi->num_locks) ? CHECK_FILELOCK : 0;
+	while (true) {
+		flags &= CEPH_FILE_MODE_MASK;
+		if (atomic_read(&fi->num_locks))
+			flags |= CHECK_FILELOCK;
 		_got = 0;
 		ret = try_get_cap_refs(inode, need, want, endoff,
 				       flags, &_got);
-		if (ret == -EAGAIN)
-			continue;
+		WARN_ON_ONCE(ret == -EAGAIN);
 		if (!ret) {
+			struct ceph_mds_client *mdsc = fsc->mdsc;
+			struct cap_wait cw;
 			DEFINE_WAIT_FUNC(wait, woken_wake_function);
+
+			cw.ino = inode->i_ino;
+			cw.tgid = current->tgid;
+			cw.need = need;
+			cw.want = want;
+
+			spin_lock(&mdsc->caps_list_lock);
+			list_add(&cw.list, &mdsc->cap_wait_list);
+			spin_unlock(&mdsc->caps_list_lock);
+
+			/* make sure used fmode not timeout */
+			ceph_get_fmode(ci, flags, FMODE_WAIT_BIAS);
 			add_wait_queue(&ci->i_cap_wq, &wait);
 
 			flags |= NON_BLOCKING;
@@ -2778,6 +2815,12 @@ int ceph_get_caps(struct file *filp, int need, int want,
 			}
 
 			remove_wait_queue(&ci->i_cap_wq, &wait);
+			ceph_put_fmode(ci, flags, FMODE_WAIT_BIAS);
+
+			spin_lock(&mdsc->caps_list_lock);
+			list_del(&cw.list);
+			spin_unlock(&mdsc->caps_list_lock);
+
 			if (ret == -EAGAIN)
 				continue;
 		}
@@ -2790,16 +2833,26 @@ int ceph_get_caps(struct file *filp, int need, int want,
 		}
 
 		if (ret < 0) {
+			if (ret == -EFBIG || ret == -ESTALE) {
+				int ret2 = ceph_wait_on_async_create(inode);
+				if (ret2 < 0)
+					return ret2;
+			}
+			if (ret == -EFBIG) {
+				check_max_size(inode, endoff);
+				continue;
+			}
 			if (ret == -ESTALE) {
 				/* session was killed, try renew caps */
-				ret = ceph_renew_caps(inode);
+				ret = ceph_renew_caps(inode, flags);
 				if (ret == 0)
 					continue;
 			}
 			return ret;
 		}
 
-		if (ci->i_inline_version != CEPH_INLINE_NONE &&
+		if (S_ISREG(ci->vfs_inode.i_mode) &&
+		    ci->i_inline_version != CEPH_INLINE_NONE &&
 		    (_got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) &&
 		    i_size_read(inode) > 0) {
 			struct page *page =
@@ -2832,7 +2885,8 @@ int ceph_get_caps(struct file *filp, int need, int want,
 		break;
 	}
 
-	if ((_got & CEPH_CAP_FILE_RD) && (_got & CEPH_CAP_FILE_CACHE))
+	if (S_ISREG(ci->vfs_inode.i_mode) &&
+	    (_got & CEPH_CAP_FILE_RD) && (_got & CEPH_CAP_FILE_CACHE))
 		ceph_fscache_revalidate_cookie(ci);
 
 	*got = _got;
@@ -2846,7 +2900,7 @@ int ceph_get_caps(struct file *filp, int need, int want,
 void ceph_get_cap_refs(struct ceph_inode_info *ci, int caps)
 {
 	spin_lock(&ci->i_ceph_lock);
-	__take_cap_refs(ci, caps, false);
+	ceph_take_cap_refs(ci, caps, false);
 	spin_unlock(&ci->i_ceph_lock);
 }
 
@@ -2897,6 +2951,9 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
 	if (had & CEPH_CAP_FILE_CACHE)
 		if (--ci->i_rdcache_ref == 0)
 			last++;
+	if (had & CEPH_CAP_FILE_EXCL)
+		if (--ci->i_fx_ref == 0)
+			last++;
 	if (had & CEPH_CAP_FILE_BUFFER) {
 		if (--ci->i_wb_ref == 0) {
 			last++;
@@ -2928,7 +2985,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
 				ci->i_head_snapc = NULL;
 			}
 			/* see comment in __ceph_remove_cap() */
-			if (!__ceph_is_any_caps(ci) && ci->i_snap_realm)
+			if (!__ceph_is_any_real_caps(ci) && ci->i_snap_realm)
 				drop_inode_snap_realm(ci);
 		}
 	spin_unlock(&ci->i_ceph_lock);
@@ -2936,7 +2993,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
 	dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had),
 	     last ? " last" : "", put ? " put" : "");
 
-	if (last && !flushsnaps)
+	if (last)
 		ceph_check_caps(ci, 0, NULL);
 	else if (flushsnaps)
 		ceph_flush_snaps(ci, NULL);
@@ -3018,7 +3075,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
 	spin_unlock(&ci->i_ceph_lock);
 
 	if (last) {
-		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
+		ceph_check_caps(ci, 0, NULL);
 	} else if (flush_snaps) {
 		ceph_flush_snaps(ci, NULL);
 	}
@@ -3119,7 +3176,7 @@ static void handle_cap_grant(struct inode *inode,
 	 * try to invalidate (once).  (If there are dirty buffers, we
 	 * will invalidate _after_ writeback.)
 	 */
-	if (!S_ISDIR(inode->i_mode) && /* don't invalidate readdir cache */
+	if (S_ISREG(inode->i_mode) && /* don't invalidate readdir cache */
 	    ((cap->issued & ~newcaps) & CEPH_CAP_FILE_CACHE) &&
 	    (newcaps & CEPH_CAP_FILE_LAZYIO) == 0 &&
 	    !(ci->i_wrbuffer_ref || ci->i_wb_ref)) {
@@ -3283,11 +3340,12 @@ static void handle_cap_grant(struct inode *inode,
 		     ceph_cap_string(cap->issued),
 		     ceph_cap_string(newcaps),
 		     ceph_cap_string(revoking));
-		if (revoking & used & CEPH_CAP_FILE_BUFFER)
+		if (S_ISREG(inode->i_mode) &&
+		    (revoking & used & CEPH_CAP_FILE_BUFFER))
 			writeback = true;  /* initiate writeback; will delay ack */
-		else if (revoking == CEPH_CAP_FILE_CACHE &&
-			 (newcaps & CEPH_CAP_FILE_LAZYIO) == 0 &&
-			 queue_invalidate)
+		else if (queue_invalidate &&
+			 revoking == CEPH_CAP_FILE_CACHE &&
+			 (newcaps & CEPH_CAP_FILE_LAZYIO) == 0)
 			; /* do nothing yet, invalidation will be queued */
 		else if (cap == ci->i_auth_cap)
 			check_caps = 1; /* check auth cap only */
@@ -3325,7 +3383,8 @@ static void handle_cap_grant(struct inode *inode,
 	if (le32_to_cpu(grant->op) == CEPH_CAP_OP_IMPORT) {
 		if (newcaps & ~extra_info->issued)
 			wake = true;
-		kick_flushing_inode_caps(session->s_mdsc, session, inode);
+		ceph_kick_flushing_inode_caps(session, ci);
+		spin_unlock(&ci->i_ceph_lock);
 		up_read(&session->s_mdsc->snap_rwsem);
 	} else {
 		spin_unlock(&ci->i_ceph_lock);
@@ -3353,10 +3412,10 @@ static void handle_cap_grant(struct inode *inode,
 		wake_up_all(&ci->i_cap_wq);
 
 	if (check_caps == 1)
-		ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_AUTHONLY,
+		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY | CHECK_CAPS_NOINVAL,
 				session);
 	else if (check_caps == 2)
-		ceph_check_caps(ci, CHECK_CAPS_NODELAY, session);
+		ceph_check_caps(ci, CHECK_CAPS_NOINVAL, session);
 	else
 		mutex_unlock(&session->s_mutex);
 }
@@ -3605,8 +3664,6 @@ retry:
 		goto out_unlock;
 
 	if (target < 0) {
-		if (cap->mds_wanted | cap->issued)
-			ci->i_ceph_flags |= CEPH_I_CAP_DROPPED;
 		__ceph_remove_cap(cap, false);
 		goto out_unlock;
 	}
@@ -3654,7 +3711,7 @@ retry:
 		/* add placeholder for the export tagert */
 		int flag = (cap == ci->i_auth_cap) ? CEPH_CAP_FLAG_AUTH : 0;
 		tcap = new_cap;
-		ceph_add_cap(inode, tsession, t_cap_id, -1, issued, 0,
+		ceph_add_cap(inode, tsession, t_cap_id, issued, 0,
 			     t_seq - 1, t_mseq, (u64)-1, flag, &new_cap);
 
 		if (!list_empty(&ci->i_cap_flush_list) &&
@@ -3689,6 +3746,7 @@ retry:
 		WARN_ON(1);
 		tsession = NULL;
 		target = -1;
+		mutex_lock(&session->s_mutex);
 	}
 	goto retry;
 
@@ -3759,7 +3817,7 @@ retry:
 	__ceph_caps_issued(ci, &issued);
 	issued |= __ceph_caps_dirty(ci);
 
-	ceph_add_cap(inode, session, cap_id, -1, caps, wanted, seq, mseq,
+	ceph_add_cap(inode, session, cap_id, caps, wanted, seq, mseq,
 		     realmino, CEPH_CAP_FLAG_AUTH, &new_cap);
 
 	ocap = peer >= 0 ? __get_cap_for_mds(ci, peer) : NULL;
@@ -3933,7 +3991,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 			__ceph_queue_cap_release(session, cap);
 			spin_unlock(&session->s_cap_lock);
 		}
-		goto done;
+		goto flush_cap_releases;
 	}
 
 	/* these will work even if we don't have a cap yet */
@@ -4033,7 +4091,6 @@ void ceph_check_delayed_caps(struct ceph_mds_client *mdsc)
 {
 	struct inode *inode;
 	struct ceph_inode_info *ci;
-	int flags = CHECK_CAPS_NODELAY;
 
 	dout("check_delayed_caps\n");
 	while (1) {
@@ -4053,7 +4110,7 @@ void ceph_check_delayed_caps(struct ceph_mds_client *mdsc)
 
 		if (inode) {
 			dout("check_delayed_caps on %p\n", inode);
-			ceph_check_caps(ci, flags, NULL);
+			ceph_check_caps(ci, 0, NULL);
 			/* avoid calling iput_final() in tick thread */
 			ceph_async_iput(inode);
 		}
@@ -4078,7 +4135,7 @@ void ceph_flush_dirty_caps(struct ceph_mds_client *mdsc)
 		ihold(inode);
 		dout("flush_dirty_caps %p\n", inode);
 		spin_unlock(&mdsc->cap_dirty_lock);
-		ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_FLUSH, NULL);
+		ceph_check_caps(ci, CHECK_CAPS_FLUSH, NULL);
 		iput(inode);
 		spin_lock(&mdsc->cap_dirty_lock);
 	}
@@ -4086,14 +4143,31 @@ void ceph_flush_dirty_caps(struct ceph_mds_client *mdsc)
 	dout("flush_dirty_caps done\n");
 }
 
-void __ceph_get_fmode(struct ceph_inode_info *ci, int fmode)
+void __ceph_touch_fmode(struct ceph_inode_info *ci,
+			struct ceph_mds_client *mdsc, int fmode)
+{
+	unsigned long now = jiffies;
+	if (fmode & CEPH_FILE_MODE_RD)
+		ci->i_last_rd = now;
+	if (fmode & CEPH_FILE_MODE_WR)
+		ci->i_last_wr = now;
+	/* queue periodic check */
+	if (fmode &&
+	    __ceph_is_any_real_caps(ci) &&
+	    list_empty(&ci->i_cap_delay_list))
+		__cap_delay_requeue(mdsc, ci);
+}
+
+void ceph_get_fmode(struct ceph_inode_info *ci, int fmode, int count)
 {
 	int i;
 	int bits = (fmode << 1) | 1;
+	spin_lock(&ci->i_ceph_lock);
 	for (i = 0; i < CEPH_FILE_MODE_BITS; i++) {
 		if (bits & (1 << i))
-			ci->i_nr_by_mode[i]++;
+			ci->i_nr_by_mode[i] += count;
 	}
+	spin_unlock(&ci->i_ceph_lock);
 }
 
 /*
@@ -4101,26 +4175,18 @@ void __ceph_get_fmode(struct ceph_inode_info *ci, int fmode)
  * we may need to release capabilities to the MDS (or schedule
  * their delayed release).
  */
-void ceph_put_fmode(struct ceph_inode_info *ci, int fmode)
+void ceph_put_fmode(struct ceph_inode_info *ci, int fmode, int count)
 {
-	int i, last = 0;
+	int i;
 	int bits = (fmode << 1) | 1;
 	spin_lock(&ci->i_ceph_lock);
 	for (i = 0; i < CEPH_FILE_MODE_BITS; i++) {
 		if (bits & (1 << i)) {
-			BUG_ON(ci->i_nr_by_mode[i] == 0);
-			if (--ci->i_nr_by_mode[i] == 0)
-				last++;
+			BUG_ON(ci->i_nr_by_mode[i] < count);
+			ci->i_nr_by_mode[i] -= count;
 		}
 	}
-	dout("put_fmode %p fmode %d {%d,%d,%d,%d}\n",
-	     &ci->vfs_inode, fmode,
-	     ci->i_nr_by_mode[0], ci->i_nr_by_mode[1],
-	     ci->i_nr_by_mode[2], ci->i_nr_by_mode[3]);
 	spin_unlock(&ci->i_ceph_lock);
-
-	if (last && ci->i_vino.snap == CEPH_NOSNAP)
-		ceph_check_caps(ci, 0, NULL);
 }
 
 /*
@@ -4138,7 +4204,6 @@ int ceph_drop_caps_for_unlink(struct inode *inode)
 	if (inode->i_nlink == 1) {
 		drop |= ~(__ceph_caps_wanted(ci) | CEPH_CAP_PIN);
 
-		ci->i_ceph_flags |= CEPH_I_NODELAY;
 		if (__ceph_caps_dirty(ci)) {
 			struct ceph_mds_client *mdsc =
 				ceph_inode_to_client(inode)->mdsc;
@@ -4194,8 +4259,6 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
 		if (force || (cap->issued & drop)) {
 			if (cap->issued & drop) {
 				int wanted = __ceph_caps_wanted(ci);
-				if ((ci->i_ceph_flags & CEPH_I_NODELAY) == 0)
-					wanted |= cap->mds_wanted;
 				dout("encode_inode_release %p cap %p "
 				     "%s -> %s, wanted %s -> %s\n", inode, cap,
 				     ceph_cap_string(cap->issued),
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index facb387c2735..dcaed75de9e6 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -33,7 +33,7 @@ static int mdsmap_show(struct seq_file *s, void *p)
 	seq_printf(s, "max_mds %d\n", mdsmap->m_max_mds);
 	seq_printf(s, "session_timeout %d\n", mdsmap->m_session_timeout);
 	seq_printf(s, "session_autoclose %d\n", mdsmap->m_session_autoclose);
-	for (i = 0; i < mdsmap->m_num_mds; i++) {
+	for (i = 0; i < mdsmap->possible_max_rank; i++) {
 		struct ceph_entity_addr *addr = &mdsmap->m_info[i].addr;
 		int state = mdsmap->m_info[i].state;
 		seq_printf(s, "\tmds%d\t%s\t(%s)\n", i,
@@ -139,6 +139,7 @@ static int caps_show(struct seq_file *s, void *p)
 	struct ceph_fs_client *fsc = s->private;
 	struct ceph_mds_client *mdsc = fsc->mdsc;
 	int total, avail, used, reserved, min, i;
+	struct cap_wait	*cw;
 
 	ceph_reservation_status(fsc, &total, &avail, &used, &reserved, &min);
 	seq_printf(s, "total\t\t%d\n"
@@ -166,6 +167,18 @@ static int caps_show(struct seq_file *s, void *p)
 	}
 	mutex_unlock(&mdsc->mutex);
 
+	seq_printf(s, "\n\nWaiters:\n--------\n");
+	seq_printf(s, "tgid         ino                need             want\n");
+	seq_printf(s, "-----------------------------------------------------\n");
+
+	spin_lock(&mdsc->caps_list_lock);
+	list_for_each_entry(cw, &mdsc->cap_wait_list, list) {
+		seq_printf(s, "%-13d0x%-17lx%-17s%-17s\n", cw->tgid, cw->ino,
+				ceph_cap_string(cw->need),
+				ceph_cap_string(cw->want));
+	}
+	spin_unlock(&mdsc->caps_list_lock);
+
 	return 0;
 }
 
@@ -205,10 +218,10 @@ static int mds_sessions_show(struct seq_file *s, void *ptr)
 	return 0;
 }
 
-CEPH_DEFINE_SHOW_FUNC(mdsmap_show)
-CEPH_DEFINE_SHOW_FUNC(mdsc_show)
-CEPH_DEFINE_SHOW_FUNC(caps_show)
-CEPH_DEFINE_SHOW_FUNC(mds_sessions_show)
+DEFINE_SHOW_ATTRIBUTE(mdsmap);
+DEFINE_SHOW_ATTRIBUTE(mdsc);
+DEFINE_SHOW_ATTRIBUTE(caps);
+DEFINE_SHOW_ATTRIBUTE(mds_sessions);
 
 
 /*
@@ -258,7 +271,7 @@ void ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 				    &congestion_kb_fops);
 
 	snprintf(name, sizeof(name), "../../bdi/%s",
-		 dev_name(fsc->sb->s_bdi->dev));
+		 bdi_dev_name(fsc->sb->s_bdi));
 	fsc->debugfs_bdi =
 		debugfs_create_symlink("bdi",
 				       fsc->client->debugfs_dir,
@@ -268,25 +281,25 @@ void ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 					0400,
 					fsc->client->debugfs_dir,
 					fsc,
-					&mdsmap_show_fops);
+					&mdsmap_fops);
 
 	fsc->debugfs_mds_sessions = debugfs_create_file("mds_sessions",
 					0400,
 					fsc->client->debugfs_dir,
 					fsc,
-					&mds_sessions_show_fops);
+					&mds_sessions_fops);
 
 	fsc->debugfs_mdsc = debugfs_create_file("mdsc",
 						0400,
 						fsc->client->debugfs_dir,
 						fsc,
-						&mdsc_show_fops);
+						&mdsc_fops);
 
 	fsc->debugfs_caps = debugfs_create_file("caps",
 						   0400,
 						   fsc->client->debugfs_dir,
 						   fsc,
-						   &caps_show_fops);
+						   &caps_fops);
 }
 
 
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index d17a789fd856..4c4202c93b71 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -335,8 +335,11 @@ static int ceph_readdir(struct file *file, struct dir_context *ctx)
 		ctx->pos = 2;
 	}
 
-	/* can we use the dcache? */
 	spin_lock(&ci->i_ceph_lock);
+	/* request Fx cap. if have Fx, we don't need to release Fs cap
+	 * for later create/unlink. */
+	__ceph_touch_fmode(ci, mdsc, CEPH_FILE_MODE_WR);
+	/* can we use the dcache? */
 	if (ceph_test_mount_opt(fsc, DCACHE) &&
 	    !ceph_test_mount_opt(fsc, NOASYNCREADDIR) &&
 	    ceph_snap(inode) != CEPH_SNAPDIR &&
@@ -752,7 +755,7 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
 		struct ceph_dentry_info *di = ceph_dentry(dentry);
 
 		spin_lock(&ci->i_ceph_lock);
-		dout(" dir %p flags are %d\n", dir, ci->i_ceph_flags);
+		dout(" dir %p flags are 0x%lx\n", dir, ci->i_ceph_flags);
 		if (strncmp(dentry->d_name.name,
 			    fsc->mount_options->snapdir_name,
 			    dentry->d_name.len) &&
@@ -760,6 +763,7 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
 		    ceph_test_mount_opt(fsc, DCACHE) &&
 		    __ceph_dir_is_complete(ci) &&
 		    (__ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1))) {
+			__ceph_touch_fmode(ci, mdsc, CEPH_FILE_MODE_RD);
 			spin_unlock(&ci->i_ceph_lock);
 			dout(" dir %p complete, -ENOENT\n", dir);
 			d_add(dentry, NULL);
@@ -1036,6 +1040,78 @@ static int ceph_link(struct dentry *old_dentry, struct inode *dir,
 	return err;
 }
 
+static void ceph_async_unlink_cb(struct ceph_mds_client *mdsc,
+				 struct ceph_mds_request *req)
+{
+	int result = req->r_err ? req->r_err :
+			le32_to_cpu(req->r_reply_info.head->result);
+
+	if (result == -EJUKEBOX)
+		goto out;
+
+	/* If op failed, mark everyone involved for errors */
+	if (result) {
+		int pathlen = 0;
+		u64 base = 0;
+		char *path = ceph_mdsc_build_path(req->r_dentry, &pathlen,
+						  &base, 0);
+
+		/* mark error on parent + clear complete */
+		mapping_set_error(req->r_parent->i_mapping, result);
+		ceph_dir_clear_complete(req->r_parent);
+
+		/* drop the dentry -- we don't know its status */
+		if (!d_unhashed(req->r_dentry))
+			d_drop(req->r_dentry);
+
+		/* mark inode itself for an error (since metadata is bogus) */
+		mapping_set_error(req->r_old_inode->i_mapping, result);
+
+		pr_warn("ceph: async unlink failure path=(%llx)%s result=%d!\n",
+			base, IS_ERR(path) ? "<<bad>>" : path, result);
+		ceph_mdsc_free_path(path, pathlen);
+	}
+out:
+	iput(req->r_old_inode);
+	ceph_mdsc_release_dir_caps(req);
+}
+
+static int get_caps_for_async_unlink(struct inode *dir, struct dentry *dentry)
+{
+	struct ceph_inode_info *ci = ceph_inode(dir);
+	struct ceph_dentry_info *di;
+	int got = 0, want = CEPH_CAP_FILE_EXCL | CEPH_CAP_DIR_UNLINK;
+
+	spin_lock(&ci->i_ceph_lock);
+	if ((__ceph_caps_issued(ci, NULL) & want) == want) {
+		ceph_take_cap_refs(ci, want, false);
+		got = want;
+	}
+	spin_unlock(&ci->i_ceph_lock);
+
+	/* If we didn't get anything, return 0 */
+	if (!got)
+		return 0;
+
+        spin_lock(&dentry->d_lock);
+        di = ceph_dentry(dentry);
+	/*
+	 * - We are holding Fx, which implies Fs caps.
+	 * - Only support async unlink for primary linkage
+	 */
+	if (atomic_read(&ci->i_shared_gen) != di->lease_shared_gen ||
+	    !(di->flags & CEPH_DENTRY_PRIMARY_LINK))
+		want = 0;
+        spin_unlock(&dentry->d_lock);
+
+	/* Do we still want what we've got? */
+	if (want == got)
+		return got;
+
+	ceph_put_cap_refs(ci, got);
+	return 0;
+}
+
 /*
  * rmdir and unlink are differ only by the metadata op code
  */
@@ -1045,6 +1121,7 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry)
 	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = d_inode(dentry);
 	struct ceph_mds_request *req;
+	bool try_async = ceph_test_mount_opt(fsc, ASYNC_DIROPS);
 	int err = -EROFS;
 	int op;
 
@@ -1059,6 +1136,7 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry)
 			CEPH_MDS_OP_RMDIR : CEPH_MDS_OP_UNLINK;
 	} else
 		goto out;
+retry:
 	req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS);
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
@@ -1067,13 +1145,39 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry)
 	req->r_dentry = dget(dentry);
 	req->r_num_caps = 2;
 	req->r_parent = dir;
-	set_bit(CEPH_MDS_R_PARENT_LOCKED, &req->r_req_flags);
 	req->r_dentry_drop = CEPH_CAP_FILE_SHARED;
 	req->r_dentry_unless = CEPH_CAP_FILE_EXCL;
 	req->r_inode_drop = ceph_drop_caps_for_unlink(inode);
-	err = ceph_mdsc_do_request(mdsc, dir, req);
-	if (!err && !req->r_reply_info.head->is_dentry)
-		d_delete(dentry);
+
+	if (try_async && op == CEPH_MDS_OP_UNLINK &&
+	    (req->r_dir_caps = get_caps_for_async_unlink(dir, dentry))) {
+		dout("async unlink on %lu/%.*s caps=%s", dir->i_ino,
+		     dentry->d_name.len, dentry->d_name.name,
+		     ceph_cap_string(req->r_dir_caps));
+		set_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags);
+		req->r_callback = ceph_async_unlink_cb;
+		req->r_old_inode = d_inode(dentry);
+		ihold(req->r_old_inode);
+		err = ceph_mdsc_submit_request(mdsc, dir, req);
+		if (!err) {
+			/*
+			 * We have enough caps, so we assume that the unlink
+			 * will succeed. Fix up the target inode and dcache.
+			 */
+			drop_nlink(inode);
+			d_delete(dentry);
+		} else if (err == -EJUKEBOX) {
+			try_async = false;
+			ceph_mdsc_put_request(req);
+			goto retry;
+		}
+	} else {
+		set_bit(CEPH_MDS_R_PARENT_LOCKED, &req->r_req_flags);
+		err = ceph_mdsc_do_request(mdsc, dir, req);
+		if (!err && !req->r_reply_info.head->is_dentry)
+			d_delete(dentry);
+	}
+
 	ceph_mdsc_put_request(req);
 out:
 	return err;
@@ -1186,7 +1290,7 @@ void __ceph_dentry_dir_lease_touch(struct ceph_dentry_info *di)
 	struct dentry *dn = di->dentry;
 	struct ceph_mds_client *mdsc;
 
-	dout("dentry_dir_lease_touch %p %p '%pd' (offset %lld)\n",
+	dout("dentry_dir_lease_touch %p %p '%pd' (offset 0x%llx)\n",
 	     di, dn, dn, di->offset);
 
 	if (!list_empty(&di->lease_list)) {
@@ -1411,6 +1515,7 @@ void ceph_invalidate_dentry_lease(struct dentry *dentry)
 	spin_lock(&dentry->d_lock);
 	di->time = jiffies;
 	di->lease_shared_gen = 0;
+	di->flags &= ~CEPH_DENTRY_PRIMARY_LINK;
 	__dentry_lease_unlist(di);
 	spin_unlock(&dentry->d_lock);
 }
@@ -1520,7 +1625,8 @@ static int __dir_lease_try_check(const struct dentry *dentry)
 /*
  * Check if directory-wide content lease/cap is valid.
  */
-static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry)
+static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry,
+			      struct ceph_mds_client *mdsc)
 {
 	struct ceph_inode_info *ci = ceph_inode(dir);
 	int valid;
@@ -1528,7 +1634,10 @@ static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry)
 
 	spin_lock(&ci->i_ceph_lock);
 	valid = __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1);
-	shared_gen = atomic_read(&ci->i_shared_gen);
+	if (valid) {
+		__ceph_touch_fmode(ci, mdsc, CEPH_FILE_MODE_RD);
+		shared_gen = atomic_read(&ci->i_shared_gen);
+	}
 	spin_unlock(&ci->i_ceph_lock);
 	if (valid) {
 		struct ceph_dentry_info *di;
@@ -1554,6 +1663,7 @@ static int ceph_d_revalidate(struct dentry *dentry, unsigned int flags)
 	int valid = 0;
 	struct dentry *parent;
 	struct inode *dir, *inode;
+	struct ceph_mds_client *mdsc;
 
 	if (flags & LOOKUP_RCU) {
 		parent = READ_ONCE(dentry->d_parent);
@@ -1567,9 +1677,11 @@ static int ceph_d_revalidate(struct dentry *dentry, unsigned int flags)
 		inode = d_inode(dentry);
 	}
 
-	dout("d_revalidate %p '%pd' inode %p offset %lld\n", dentry,
+	dout("d_revalidate %p '%pd' inode %p offset 0x%llx\n", dentry,
 	     dentry, inode, ceph_dentry(dentry)->offset);
 
+	mdsc = ceph_sb_to_client(dir->i_sb)->mdsc;
+
 	/* always trust cached snapped dentries, snapdir dentry */
 	if (ceph_snap(dir) != CEPH_NOSNAP) {
 		dout("d_revalidate %p '%pd' inode %p is SNAPPED\n", dentry,
@@ -1581,7 +1693,7 @@ static int ceph_d_revalidate(struct dentry *dentry, unsigned int flags)
 		valid = dentry_lease_is_valid(dentry, flags);
 		if (valid == -ECHILD)
 			return valid;
-		if (valid || dir_lease_is_valid(dir, dentry)) {
+		if (valid || dir_lease_is_valid(dir, dentry, mdsc)) {
 			if (inode)
 				valid = ceph_is_any_caps(inode);
 			else
@@ -1590,8 +1702,6 @@ static int ceph_d_revalidate(struct dentry *dentry, unsigned int flags)
 	}
 
 	if (!valid) {
-		struct ceph_mds_client *mdsc =
-			ceph_sb_to_client(dir->i_sb)->mdsc;
 		struct ceph_mds_request *req;
 		int op, err;
 		u32 mask;
@@ -1809,6 +1919,7 @@ const struct file_operations ceph_dir_fops = {
 	.open = ceph_open,
 	.release = ceph_release,
 	.unlocked_ioctl = ceph_ioctl,
+	.compat_ioctl = compat_ptr_ioctl,
 	.fsync = ceph_fsync,
 	.lock = ceph_lock,
 	.flock = ceph_flock,
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index b6bfa94332c3..79dc06881e78 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -315,6 +315,11 @@ static struct dentry *__get_parent(struct super_block *sb,
 
 	req->r_num_caps = 1;
 	err = ceph_mdsc_do_request(mdsc, NULL, req);
+	if (err) {
+		ceph_mdsc_put_request(req);
+		return ERR_PTR(err);
+	}
+
 	inode = req->r_target_inode;
 	if (inode)
 		ihold(inode);
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 8de633964dc3..afdfca965a7f 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -212,10 +212,8 @@ static int ceph_init_file_info(struct inode *inode, struct file *file,
 	if (isdir) {
 		struct ceph_dir_file_info *dfi =
 			kmem_cache_zalloc(ceph_dir_file_cachep, GFP_KERNEL);
-		if (!dfi) {
-			ceph_put_fmode(ci, fmode); /* clean up */
+		if (!dfi)
 			return -ENOMEM;
-		}
 
 		file->private_data = dfi;
 		fi = &dfi->file_info;
@@ -223,15 +221,15 @@ static int ceph_init_file_info(struct inode *inode, struct file *file,
 		dfi->readdir_cache_idx = -1;
 	} else {
 		fi = kmem_cache_zalloc(ceph_file_cachep, GFP_KERNEL);
-		if (!fi) {
-			ceph_put_fmode(ci, fmode); /* clean up */
+		if (!fi)
 			return -ENOMEM;
-		}
 
 		file->private_data = fi;
 	}
 
+	ceph_get_fmode(ci, fmode, 1);
 	fi->fmode = fmode;
+
 	spin_lock_init(&fi->rw_contexts_lock);
 	INIT_LIST_HEAD(&fi->rw_contexts);
 	fi->meta_err = errseq_sample(&ci->i_meta_err);
@@ -263,7 +261,6 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
 	case S_IFLNK:
 		dout("init_file %p %p 0%o (symlink)\n", inode, file,
 		     inode->i_mode);
-		ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */
 		break;
 
 	default:
@@ -273,7 +270,6 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
 		 * we need to drop the open ref now, since we don't
 		 * have .release set to ceph_release.
 		 */
-		ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */
 		BUG_ON(inode->i_fop->release == ceph_release);
 
 		/* call the proper open fop */
@@ -285,14 +281,15 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
 /*
  * try renew caps after session gets killed.
  */
-int ceph_renew_caps(struct inode *inode)
+int ceph_renew_caps(struct inode *inode, int fmode)
 {
-	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_mds_request *req;
 	int err, flags, wanted;
 
 	spin_lock(&ci->i_ceph_lock);
+	__ceph_touch_fmode(ci, mdsc, fmode);
 	wanted = __ceph_caps_file_wanted(ci);
 	if (__ceph_is_any_real_caps(ci) &&
 	    (!(wanted & CEPH_CAP_ANY_WR) || ci->i_auth_cap)) {
@@ -326,7 +323,6 @@ int ceph_renew_caps(struct inode *inode)
 	req->r_inode = inode;
 	ihold(inode);
 	req->r_num_caps = 1;
-	req->r_fmode = -1;
 
 	err = ceph_mdsc_do_request(mdsc, NULL, req);
 	ceph_mdsc_put_request(req);
@@ -372,9 +368,6 @@ int ceph_open(struct inode *inode, struct file *file)
 
 	/* trivially open snapdir */
 	if (ceph_snap(inode) == CEPH_SNAPDIR) {
-		spin_lock(&ci->i_ceph_lock);
-		__ceph_get_fmode(ci, fmode);
-		spin_unlock(&ci->i_ceph_lock);
 		return ceph_init_file(inode, file, fmode);
 	}
 
@@ -392,7 +385,7 @@ int ceph_open(struct inode *inode, struct file *file)
 		dout("open %p fmode %d want %s issued %s using existing\n",
 		     inode, fmode, ceph_cap_string(wanted),
 		     ceph_cap_string(issued));
-		__ceph_get_fmode(ci, fmode);
+		__ceph_touch_fmode(ci, mdsc, fmode);
 		spin_unlock(&ci->i_ceph_lock);
 
 		/* adjust wanted? */
@@ -404,7 +397,7 @@ int ceph_open(struct inode *inode, struct file *file)
 		return ceph_init_file(inode, file, fmode);
 	} else if (ceph_snap(inode) != CEPH_NOSNAP &&
 		   (ci->i_snap_caps & wanted) == wanted) {
-		__ceph_get_fmode(ci, fmode);
+		__ceph_touch_fmode(ci, mdsc, fmode);
 		spin_unlock(&ci->i_ceph_lock);
 		return ceph_init_file(inode, file, fmode);
 	}
@@ -430,6 +423,236 @@ out:
 	return err;
 }
 
+/* Clone the layout from a synchronous create, if the dir now has Dc caps */
+static void
+cache_file_layout(struct inode *dst, struct inode *src)
+{
+	struct ceph_inode_info *cdst = ceph_inode(dst);
+	struct ceph_inode_info *csrc = ceph_inode(src);
+
+	spin_lock(&cdst->i_ceph_lock);
+	if ((__ceph_caps_issued(cdst, NULL) & CEPH_CAP_DIR_CREATE) &&
+	    !ceph_file_layout_is_valid(&cdst->i_cached_layout)) {
+		memcpy(&cdst->i_cached_layout, &csrc->i_layout,
+			sizeof(cdst->i_cached_layout));
+		rcu_assign_pointer(cdst->i_cached_layout.pool_ns,
+				   ceph_try_get_string(csrc->i_layout.pool_ns));
+	}
+	spin_unlock(&cdst->i_ceph_lock);
+}
+
+/*
+ * Try to set up an async create. We need caps, a file layout, and inode number,
+ * and either a lease on the dentry or complete dir info. If any of those
+ * criteria are not satisfied, then return false and the caller can go
+ * synchronous.
+ */
+static int try_prep_async_create(struct inode *dir, struct dentry *dentry,
+				 struct ceph_file_layout *lo, u64 *pino)
+{
+	struct ceph_inode_info *ci = ceph_inode(dir);
+	struct ceph_dentry_info *di = ceph_dentry(dentry);
+	int got = 0, want = CEPH_CAP_FILE_EXCL | CEPH_CAP_DIR_CREATE;
+	u64 ino;
+
+	spin_lock(&ci->i_ceph_lock);
+	/* No auth cap means no chance for Dc caps */
+	if (!ci->i_auth_cap)
+		goto no_async;
+
+	/* Any delegated inos? */
+	if (xa_empty(&ci->i_auth_cap->session->s_delegated_inos))
+		goto no_async;
+
+	if (!ceph_file_layout_is_valid(&ci->i_cached_layout))
+		goto no_async;
+
+	if ((__ceph_caps_issued(ci, NULL) & want) != want)
+		goto no_async;
+
+	if (d_in_lookup(dentry)) {
+		if (!__ceph_dir_is_complete(ci))
+			goto no_async;
+		spin_lock(&dentry->d_lock);
+		di->lease_shared_gen = atomic_read(&ci->i_shared_gen);
+		spin_unlock(&dentry->d_lock);
+	} else if (atomic_read(&ci->i_shared_gen) !=
+		   READ_ONCE(di->lease_shared_gen)) {
+		goto no_async;
+	}
+
+	ino = ceph_get_deleg_ino(ci->i_auth_cap->session);
+	if (!ino)
+		goto no_async;
+
+	*pino = ino;
+	ceph_take_cap_refs(ci, want, false);
+	memcpy(lo, &ci->i_cached_layout, sizeof(*lo));
+	rcu_assign_pointer(lo->pool_ns,
+			   ceph_try_get_string(ci->i_cached_layout.pool_ns));
+	got = want;
+no_async:
+	spin_unlock(&ci->i_ceph_lock);
+	return got;
+}
+
+static void restore_deleg_ino(struct inode *dir, u64 ino)
+{
+	struct ceph_inode_info *ci = ceph_inode(dir);
+	struct ceph_mds_session *s = NULL;
+
+	spin_lock(&ci->i_ceph_lock);
+	if (ci->i_auth_cap)
+		s = ceph_get_mds_session(ci->i_auth_cap->session);
+	spin_unlock(&ci->i_ceph_lock);
+	if (s) {
+		int err = ceph_restore_deleg_ino(s, ino);
+		if (err)
+			pr_warn("ceph: unable to restore delegated ino 0x%llx to session: %d\n",
+				ino, err);
+		ceph_put_mds_session(s);
+	}
+}
+
+static void ceph_async_create_cb(struct ceph_mds_client *mdsc,
+                                 struct ceph_mds_request *req)
+{
+	int result = req->r_err ? req->r_err :
+			le32_to_cpu(req->r_reply_info.head->result);
+
+	if (result == -EJUKEBOX)
+		goto out;
+
+	mapping_set_error(req->r_parent->i_mapping, result);
+
+	if (result) {
+		struct dentry *dentry = req->r_dentry;
+		int pathlen = 0;
+		u64 base = 0;
+		char *path = ceph_mdsc_build_path(req->r_dentry, &pathlen,
+						  &base, 0);
+
+		ceph_dir_clear_complete(req->r_parent);
+		if (!d_unhashed(dentry))
+			d_drop(dentry);
+
+		/* FIXME: start returning I/O errors on all accesses? */
+		pr_warn("ceph: async create failure path=(%llx)%s result=%d!\n",
+			base, IS_ERR(path) ? "<<bad>>" : path, result);
+		ceph_mdsc_free_path(path, pathlen);
+	}
+
+	if (req->r_target_inode) {
+		struct ceph_inode_info *ci = ceph_inode(req->r_target_inode);
+		u64 ino = ceph_vino(req->r_target_inode).ino;
+
+		if (req->r_deleg_ino != ino)
+			pr_warn("%s: inode number mismatch! err=%d deleg_ino=0x%llx target=0x%llx\n",
+				__func__, req->r_err, req->r_deleg_ino, ino);
+		mapping_set_error(req->r_target_inode->i_mapping, result);
+
+		spin_lock(&ci->i_ceph_lock);
+		if (ci->i_ceph_flags & CEPH_I_ASYNC_CREATE) {
+			ci->i_ceph_flags &= ~CEPH_I_ASYNC_CREATE;
+			wake_up_bit(&ci->i_ceph_flags, CEPH_ASYNC_CREATE_BIT);
+		}
+		ceph_kick_flushing_inode_caps(req->r_session, ci);
+		spin_unlock(&ci->i_ceph_lock);
+	} else {
+		pr_warn("%s: no req->r_target_inode for 0x%llx\n", __func__,
+			req->r_deleg_ino);
+	}
+out:
+	ceph_mdsc_release_dir_caps(req);
+}
+
+static int ceph_finish_async_create(struct inode *dir, struct dentry *dentry,
+				    struct file *file, umode_t mode,
+				    struct ceph_mds_request *req,
+				    struct ceph_acl_sec_ctx *as_ctx,
+				    struct ceph_file_layout *lo)
+{
+	int ret;
+	char xattr_buf[4];
+	struct ceph_mds_reply_inode in = { };
+	struct ceph_mds_reply_info_in iinfo = { .in = &in };
+	struct ceph_inode_info *ci = ceph_inode(dir);
+	struct inode *inode;
+	struct timespec64 now;
+	struct ceph_vino vino = { .ino = req->r_deleg_ino,
+				  .snap = CEPH_NOSNAP };
+
+	ktime_get_real_ts64(&now);
+
+	inode = ceph_get_inode(dentry->d_sb, vino);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
+
+	iinfo.inline_version = CEPH_INLINE_NONE;
+	iinfo.change_attr = 1;
+	ceph_encode_timespec64(&iinfo.btime, &now);
+
+	iinfo.xattr_len = ARRAY_SIZE(xattr_buf);
+	iinfo.xattr_data = xattr_buf;
+	memset(iinfo.xattr_data, 0, iinfo.xattr_len);
+
+	in.ino = cpu_to_le64(vino.ino);
+	in.snapid = cpu_to_le64(CEPH_NOSNAP);
+	in.version = cpu_to_le64(1);	// ???
+	in.cap.caps = in.cap.wanted = cpu_to_le32(CEPH_CAP_ALL_FILE);
+	in.cap.cap_id = cpu_to_le64(1);
+	in.cap.realm = cpu_to_le64(ci->i_snap_realm->ino);
+	in.cap.flags = CEPH_CAP_FLAG_AUTH;
+	in.ctime = in.mtime = in.atime = iinfo.btime;
+	in.mode = cpu_to_le32((u32)mode);
+	in.truncate_seq = cpu_to_le32(1);
+	in.truncate_size = cpu_to_le64(-1ULL);
+	in.xattr_version = cpu_to_le64(1);
+	in.uid = cpu_to_le32(from_kuid(&init_user_ns, current_fsuid()));
+	in.gid = cpu_to_le32(from_kgid(&init_user_ns, dir->i_mode & S_ISGID ?
+				dir->i_gid : current_fsgid()));
+	in.nlink = cpu_to_le32(1);
+	in.max_size = cpu_to_le64(lo->stripe_unit);
+
+	ceph_file_layout_to_legacy(lo, &in.layout);
+
+	ret = ceph_fill_inode(inode, NULL, &iinfo, NULL, req->r_session,
+			      req->r_fmode, NULL);
+	if (ret) {
+		dout("%s failed to fill inode: %d\n", __func__, ret);
+		ceph_dir_clear_complete(dir);
+		if (!d_unhashed(dentry))
+			d_drop(dentry);
+		if (inode->i_state & I_NEW)
+			discard_new_inode(inode);
+	} else {
+		struct dentry *dn;
+
+		dout("%s d_adding new inode 0x%llx to 0x%lx/%s\n", __func__,
+			vino.ino, dir->i_ino, dentry->d_name.name);
+		ceph_dir_clear_ordered(dir);
+		ceph_init_inode_acls(inode, as_ctx);
+		if (inode->i_state & I_NEW) {
+			/*
+			 * If it's not I_NEW, then someone created this before
+			 * we got here. Assume the server is aware of it at
+			 * that point and don't worry about setting
+			 * CEPH_I_ASYNC_CREATE.
+			 */
+			ceph_inode(inode)->i_ceph_flags = CEPH_I_ASYNC_CREATE;
+			unlock_new_inode(inode);
+		}
+		if (d_in_lookup(dentry) || d_really_is_negative(dentry)) {
+			if (!d_unhashed(dentry))
+				d_drop(dentry);
+			dn = d_splice_alias(inode, dentry);
+			WARN_ON_ONCE(dn && dn != dentry);
+		}
+		file->f_mode |= FMODE_CREATED;
+		ret = finish_open(file, dentry, ceph_open);
+	}
+	return ret;
+}
 
 /*
  * Do a lookup + open with a single request.  If we get a non-existent
@@ -443,6 +666,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
 	struct ceph_mds_request *req;
 	struct dentry *dn;
 	struct ceph_acl_sec_ctx as_ctx = {};
+	bool try_async = ceph_test_mount_opt(fsc, ASYNC_DIROPS);
 	int mask;
 	int err;
 
@@ -466,7 +690,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
 		/* If it's not being looked up, it's negative */
 		return -ENOENT;
 	}
-
+retry:
 	/* do the open */
 	req = prepare_open_request(dir->i_sb, flags, mode);
 	if (IS_ERR(req)) {
@@ -475,21 +699,43 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
 	}
 	req->r_dentry = dget(dentry);
 	req->r_num_caps = 2;
+	mask = CEPH_STAT_CAP_INODE | CEPH_CAP_AUTH_SHARED;
+	if (ceph_security_xattr_wanted(dir))
+		mask |= CEPH_CAP_XATTR_SHARED;
+	req->r_args.open.mask = cpu_to_le32(mask);
+	req->r_parent = dir;
+
 	if (flags & O_CREAT) {
+		struct ceph_file_layout lo;
+
 		req->r_dentry_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_AUTH_EXCL;
 		req->r_dentry_unless = CEPH_CAP_FILE_EXCL;
 		if (as_ctx.pagelist) {
 			req->r_pagelist = as_ctx.pagelist;
 			as_ctx.pagelist = NULL;
 		}
+		if (try_async &&
+		    (req->r_dir_caps =
+		      try_prep_async_create(dir, dentry, &lo,
+					    &req->r_deleg_ino))) {
+			set_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags);
+			req->r_args.open.flags |= cpu_to_le32(CEPH_O_EXCL);
+			req->r_callback = ceph_async_create_cb;
+			err = ceph_mdsc_submit_request(mdsc, dir, req);
+			if (!err) {
+				err = ceph_finish_async_create(dir, dentry,
+							file, mode, req,
+							&as_ctx, &lo);
+			} else if (err == -EJUKEBOX) {
+				restore_deleg_ino(dir, req->r_deleg_ino);
+				ceph_mdsc_put_request(req);
+				try_async = false;
+				goto retry;
+			}
+			goto out_req;
+		}
 	}
 
-       mask = CEPH_STAT_CAP_INODE | CEPH_CAP_AUTH_SHARED;
-       if (ceph_security_xattr_wanted(dir))
-               mask |= CEPH_CAP_XATTR_SHARED;
-       req->r_args.open.mask = cpu_to_le32(mask);
-
-	req->r_parent = dir;
 	set_bit(CEPH_MDS_R_PARENT_LOCKED, &req->r_req_flags);
 	err = ceph_mdsc_do_request(mdsc,
 				   (flags & (O_CREAT|O_TRUNC)) ? dir : NULL,
@@ -518,14 +764,15 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
 	} else {
 		dout("atomic_open finish_open on dn %p\n", dn);
 		if (req->r_op == CEPH_MDS_OP_CREATE && req->r_reply_info.has_create_ino) {
-			ceph_init_inode_acls(d_inode(dentry), &as_ctx);
+			struct inode *newino = d_inode(dentry);
+
+			cache_file_layout(dir, newino);
+			ceph_init_inode_acls(newino, &as_ctx);
 			file->f_mode |= FMODE_CREATED;
 		}
 		err = finish_open(file, dentry, ceph_open);
 	}
 out_req:
-	if (!req->r_err && req->r_target_inode)
-		ceph_put_fmode(ceph_inode(req->r_target_inode), req->r_fmode);
 	ceph_mdsc_put_request(req);
 out_ctx:
 	ceph_release_acl_sec_ctx(&as_ctx);
@@ -542,7 +789,7 @@ int ceph_release(struct inode *inode, struct file *file)
 		dout("release inode %p dir file %p\n", inode, file);
 		WARN_ON(!list_empty(&dfi->file_info.rw_contexts));
 
-		ceph_put_fmode(ci, dfi->file_info.fmode);
+		ceph_put_fmode(ci, dfi->file_info.fmode, 1);
 
 		if (dfi->last_readdir)
 			ceph_mdsc_put_request(dfi->last_readdir);
@@ -554,7 +801,8 @@ int ceph_release(struct inode *inode, struct file *file)
 		dout("release inode %p regular file %p\n", inode, file);
 		WARN_ON(!list_empty(&fi->rw_contexts));
 
-		ceph_put_fmode(ci, fi->fmode);
+		ceph_put_fmode(ci, fi->fmode, 1);
+
 		kmem_cache_free(ceph_file_cachep, fi);
 	}
 
@@ -1415,9 +1663,13 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	struct inode *inode = file_inode(file);
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+	struct ceph_osd_client *osdc = &fsc->client->osdc;
 	struct ceph_cap_flush *prealloc_cf;
 	ssize_t count, written = 0;
 	int err, want, got;
+	bool direct_lock = false;
+	u32 map_flags;
+	u64 pool_flags;
 	loff_t pos;
 	loff_t limit = max(i_size_read(inode), fsc->max_file_size);
 
@@ -1428,8 +1680,11 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	if (!prealloc_cf)
 		return -ENOMEM;
 
+	if ((iocb->ki_flags & (IOCB_DIRECT | IOCB_APPEND)) == IOCB_DIRECT)
+		direct_lock = true;
+
 retry_snap:
-	if (iocb->ki_flags & IOCB_DIRECT)
+	if (direct_lock)
 		ceph_start_io_direct(inode);
 	else
 		ceph_start_io_write(inode);
@@ -1477,8 +1732,12 @@ retry_snap:
 			goto out;
 	}
 
-	/* FIXME: not complete since it doesn't account for being at quota */
-	if (ceph_osdmap_flag(&fsc->client->osdc, CEPH_OSDMAP_FULL)) {
+	down_read(&osdc->lock);
+	map_flags = osdc->osdmap->flags;
+	pool_flags = ceph_pg_pool_flags(osdc->osdmap, ci->i_layout.pool_id);
+	up_read(&osdc->lock);
+	if ((map_flags & CEPH_OSDMAP_FULL) ||
+	    (pool_flags & CEPH_POOL_FLAG_FULL)) {
 		err = -ENOSPC;
 		goto out;
 	}
@@ -1519,14 +1778,15 @@ retry_snap:
 
 		/* we might need to revert back to that point */
 		data = *from;
-		if (iocb->ki_flags & IOCB_DIRECT) {
+		if (iocb->ki_flags & IOCB_DIRECT)
 			written = ceph_direct_read_write(iocb, &data, snapc,
 							 &prealloc_cf);
-			ceph_end_io_direct(inode);
-		} else {
+		else
 			written = ceph_sync_write(iocb, &data, pos, snapc);
+		if (direct_lock)
+			ceph_end_io_direct(inode);
+		else
 			ceph_end_io_write(inode);
-		}
 		if (written > 0)
 			iov_iter_advance(from, written);
 		ceph_put_snap_context(snapc);
@@ -1555,7 +1815,7 @@ retry_snap:
 		if (dirty)
 			__mark_inode_dirty(inode, dirty);
 		if (ceph_quota_is_max_bytes_approaching(inode, iocb->ki_pos))
-			ceph_check_caps(ci, CHECK_CAPS_NODELAY, NULL);
+			ceph_check_caps(ci, 0, NULL);
 	}
 
 	dout("aio_write %p %llx.%llx %llu~%u  dropping cap refs on %s\n",
@@ -1570,14 +1830,15 @@ retry_snap:
 	}
 
 	if (written >= 0) {
-		if (ceph_osdmap_flag(&fsc->client->osdc, CEPH_OSDMAP_NEARFULL))
+		if ((map_flags & CEPH_OSDMAP_NEARFULL) ||
+		    (pool_flags & CEPH_POOL_FLAG_NEARFULL))
 			iocb->ki_flags |= IOCB_DSYNC;
 		written = generic_write_sync(iocb, written);
 	}
 
 	goto out_unlocked;
 out:
-	if (iocb->ki_flags & IOCB_DIRECT)
+	if (direct_lock)
 		ceph_end_io_direct(inode);
 	else
 		ceph_end_io_write(inode);
@@ -1931,6 +2192,71 @@ static int is_file_size_ok(struct inode *src_inode, struct inode *dst_inode,
 	return 0;
 }
 
+static ssize_t ceph_do_objects_copy(struct ceph_inode_info *src_ci, u64 *src_off,
+				    struct ceph_inode_info *dst_ci, u64 *dst_off,
+				    struct ceph_fs_client *fsc,
+				    size_t len, unsigned int flags)
+{
+	struct ceph_object_locator src_oloc, dst_oloc;
+	struct ceph_object_id src_oid, dst_oid;
+	size_t bytes = 0;
+	u64 src_objnum, src_objoff, dst_objnum, dst_objoff;
+	u32 src_objlen, dst_objlen;
+	u32 object_size = src_ci->i_layout.object_size;
+	int ret;
+
+	src_oloc.pool = src_ci->i_layout.pool_id;
+	src_oloc.pool_ns = ceph_try_get_string(src_ci->i_layout.pool_ns);
+	dst_oloc.pool = dst_ci->i_layout.pool_id;
+	dst_oloc.pool_ns = ceph_try_get_string(dst_ci->i_layout.pool_ns);
+
+	while (len >= object_size) {
+		ceph_calc_file_object_mapping(&src_ci->i_layout, *src_off,
+					      object_size, &src_objnum,
+					      &src_objoff, &src_objlen);
+		ceph_calc_file_object_mapping(&dst_ci->i_layout, *dst_off,
+					      object_size, &dst_objnum,
+					      &dst_objoff, &dst_objlen);
+		ceph_oid_init(&src_oid);
+		ceph_oid_printf(&src_oid, "%llx.%08llx",
+				src_ci->i_vino.ino, src_objnum);
+		ceph_oid_init(&dst_oid);
+		ceph_oid_printf(&dst_oid, "%llx.%08llx",
+				dst_ci->i_vino.ino, dst_objnum);
+		/* Do an object remote copy */
+		ret = ceph_osdc_copy_from(&fsc->client->osdc,
+					  src_ci->i_vino.snap, 0,
+					  &src_oid, &src_oloc,
+					  CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL |
+					  CEPH_OSD_OP_FLAG_FADVISE_NOCACHE,
+					  &dst_oid, &dst_oloc,
+					  CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL |
+					  CEPH_OSD_OP_FLAG_FADVISE_DONTNEED,
+					  dst_ci->i_truncate_seq,
+					  dst_ci->i_truncate_size,
+					  CEPH_OSD_COPY_FROM_FLAG_TRUNCATE_SEQ);
+		if (ret) {
+			if (ret == -EOPNOTSUPP) {
+				fsc->have_copy_from2 = false;
+				pr_notice("OSDs don't support copy-from2; disabling copy offload\n");
+			}
+			dout("ceph_osdc_copy_from returned %d\n", ret);
+			if (!bytes)
+				bytes = ret;
+			goto out;
+		}
+		len -= object_size;
+		bytes += object_size;
+		*src_off += object_size;
+		*dst_off += object_size;
+	}
+
+out:
+	ceph_oloc_destroy(&src_oloc);
+	ceph_oloc_destroy(&dst_oloc);
+	return bytes;
+}
+
 static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off,
 				      struct file *dst_file, loff_t dst_off,
 				      size_t len, unsigned int flags)
@@ -1941,14 +2267,11 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off,
 	struct ceph_inode_info *dst_ci = ceph_inode(dst_inode);
 	struct ceph_cap_flush *prealloc_cf;
 	struct ceph_fs_client *src_fsc = ceph_inode_to_client(src_inode);
-	struct ceph_object_locator src_oloc, dst_oloc;
-	struct ceph_object_id src_oid, dst_oid;
-	loff_t endoff = 0, size;
-	ssize_t ret = -EIO;
+	loff_t size;
+	ssize_t ret = -EIO, bytes;
 	u64 src_objnum, dst_objnum, src_objoff, dst_objoff;
-	u32 src_objlen, dst_objlen, object_size;
+	u32 src_objlen, dst_objlen;
 	int src_got = 0, dst_got = 0, err, dirty;
-	bool do_final_copy = false;
 
 	if (src_inode->i_sb != dst_inode->i_sb) {
 		struct ceph_fs_client *dst_fsc = ceph_inode_to_client(dst_inode);
@@ -1974,6 +2297,9 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off,
 	if (ceph_test_mount_opt(src_fsc, NOCOPYFROM))
 		return -EOPNOTSUPP;
 
+	if (!src_fsc->have_copy_from2)
+		return -EOPNOTSUPP;
+
 	/*
 	 * Striped file layouts require that we copy partial objects, but the
 	 * OSD copy-from operation only supports full-object copies.  Limit
@@ -2023,22 +2349,14 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off,
 	if (ret < 0)
 		goto out_caps;
 
-	size = i_size_read(dst_inode);
-	endoff = dst_off + len;
-
 	/* Drop dst file cached pages */
 	ret = invalidate_inode_pages2_range(dst_inode->i_mapping,
 					    dst_off >> PAGE_SHIFT,
-					    endoff >> PAGE_SHIFT);
+					    (dst_off + len) >> PAGE_SHIFT);
 	if (ret < 0) {
 		dout("Failed to invalidate inode pages (%zd)\n", ret);
 		ret = 0; /* XXX */
 	}
-	src_oloc.pool = src_ci->i_layout.pool_id;
-	src_oloc.pool_ns = ceph_try_get_string(src_ci->i_layout.pool_ns);
-	dst_oloc.pool = dst_ci->i_layout.pool_id;
-	dst_oloc.pool_ns = ceph_try_get_string(dst_ci->i_layout.pool_ns);
-
 	ceph_calc_file_object_mapping(&src_ci->i_layout, src_off,
 				      src_ci->i_layout.object_size,
 				      &src_objnum, &src_objoff, &src_objlen);
@@ -2057,6 +2375,8 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off,
 	 * starting at the src_off
 	 */
 	if (src_objoff) {
+		dout("Initial partial copy of %u bytes\n", src_objlen);
+
 		/*
 		 * we need to temporarily drop all caps as we'll be calling
 		 * {read,write}_iter, which will get caps again.
@@ -2064,8 +2384,9 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off,
 		put_rd_wr_caps(src_ci, src_got, dst_ci, dst_got);
 		ret = do_splice_direct(src_file, &src_off, dst_file,
 				       &dst_off, src_objlen, flags);
-		if (ret < 0) {
-			dout("do_splice_direct returned %d\n", err);
+		/* Abort on short copies or on error */
+		if (ret < src_objlen) {
+			dout("Failed partial copy (%zd)\n", ret);
 			goto out;
 		}
 		len -= ret;
@@ -2078,59 +2399,27 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off,
 		if (err < 0)
 			goto out_caps;
 	}
-	object_size = src_ci->i_layout.object_size;
-	while (len >= object_size) {
-		ceph_calc_file_object_mapping(&src_ci->i_layout, src_off,
-					      object_size, &src_objnum,
-					      &src_objoff, &src_objlen);
-		ceph_calc_file_object_mapping(&dst_ci->i_layout, dst_off,
-					      object_size, &dst_objnum,
-					      &dst_objoff, &dst_objlen);
-		ceph_oid_init(&src_oid);
-		ceph_oid_printf(&src_oid, "%llx.%08llx",
-				src_ci->i_vino.ino, src_objnum);
-		ceph_oid_init(&dst_oid);
-		ceph_oid_printf(&dst_oid, "%llx.%08llx",
-				dst_ci->i_vino.ino, dst_objnum);
-		/* Do an object remote copy */
-		err = ceph_osdc_copy_from(
-			&src_fsc->client->osdc,
-			src_ci->i_vino.snap, 0,
-			&src_oid, &src_oloc,
-			CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL |
-			CEPH_OSD_OP_FLAG_FADVISE_NOCACHE,
-			&dst_oid, &dst_oloc,
-			CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL |
-			CEPH_OSD_OP_FLAG_FADVISE_DONTNEED, 0);
-		if (err) {
-			dout("ceph_osdc_copy_from returned %d\n", err);
-			if (!ret)
-				ret = err;
-			goto out_caps;
-		}
-		len -= object_size;
-		src_off += object_size;
-		dst_off += object_size;
-		ret += object_size;
-	}
 
-	if (len)
-		/* We still need one final local copy */
-		do_final_copy = true;
+	size = i_size_read(dst_inode);
+	bytes = ceph_do_objects_copy(src_ci, &src_off, dst_ci, &dst_off,
+				     src_fsc, len, flags);
+	if (bytes <= 0) {
+		if (!ret)
+			ret = bytes;
+		goto out_caps;
+	}
+	dout("Copied %zu bytes out of %zu\n", bytes, len);
+	len -= bytes;
+	ret += bytes;
 
 	file_update_time(dst_file);
 	inode_inc_iversion_raw(dst_inode);
 
-	if (endoff > size) {
-		int caps_flags = 0;
-
+	if (dst_off > size) {
 		/* Let the MDS know about dst file size change */
-		if (ceph_quota_is_max_bytes_approaching(dst_inode, endoff))
-			caps_flags |= CHECK_CAPS_NODELAY;
-		if (ceph_inode_set_size(dst_inode, endoff))
-			caps_flags |= CHECK_CAPS_AUTHONLY;
-		if (caps_flags)
-			ceph_check_caps(dst_ci, caps_flags, NULL);
+		if (ceph_inode_set_size(dst_inode, dst_off) ||
+		    ceph_quota_is_max_bytes_approaching(dst_inode, dst_off))
+			ceph_check_caps(dst_ci, CHECK_CAPS_AUTHONLY, NULL);
 	}
 	/* Mark Fw dirty */
 	spin_lock(&dst_ci->i_ceph_lock);
@@ -2143,15 +2432,18 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off,
 out_caps:
 	put_rd_wr_caps(src_ci, src_got, dst_ci, dst_got);
 
-	if (do_final_copy) {
-		err = do_splice_direct(src_file, &src_off, dst_file,
-				       &dst_off, len, flags);
-		if (err < 0) {
-			dout("do_splice_direct returned %d\n", err);
-			goto out;
-		}
-		len -= err;
-		ret += err;
+	/*
+	 * Do the final manual copy if we still have some bytes left, unless
+	 * there were errors in remote object copies (len >= object_size).
+	 */
+	if (len && (len < src_ci->i_layout.object_size)) {
+		dout("Final partial copy of %zu bytes\n", len);
+		bytes = do_splice_direct(src_file, &src_off, dst_file,
+					 &dst_off, len, flags);
+		if (bytes > 0)
+			ret += bytes;
+		else
+			dout("Failed partial copy (%zd)\n", bytes);
 	}
 
 out:
@@ -2188,7 +2480,7 @@ const struct file_operations ceph_file_fops = {
 	.splice_read = generic_file_splice_read,
 	.splice_write = iter_file_splice_write,
 	.unlocked_ioctl = ceph_ioctl,
-	.compat_ioctl	= ceph_ioctl,
+	.compat_ioctl = compat_ptr_ioctl,
 	.fallocate	= ceph_fallocate,
 	.copy_file_range = ceph_copy_file_range,
 };
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index c07407586ce8..7fef94fd1e55 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -55,11 +55,9 @@ struct inode *ceph_get_inode(struct super_block *sb, struct ceph_vino vino)
 	inode = iget5_locked(sb, t, ceph_ino_compare, ceph_set_ino_cb, &vino);
 	if (!inode)
 		return ERR_PTR(-ENOMEM);
-	if (inode->i_state & I_NEW) {
+	if (inode->i_state & I_NEW)
 		dout("get_inode created new inode %p %llx.%llx ino %llx\n",
 		     inode, ceph_vinop(inode), (u64)inode->i_ino);
-		unlock_new_inode(inode);
-	}
 
 	dout("get_inode on %lu=%llx.%llx got %p\n", inode->i_ino, vino.ino,
 	     vino.snap, inode);
@@ -84,10 +82,18 @@ struct inode *ceph_get_snapdir(struct inode *parent)
 	inode->i_mode = parent->i_mode;
 	inode->i_uid = parent->i_uid;
 	inode->i_gid = parent->i_gid;
+	inode->i_mtime = parent->i_mtime;
+	inode->i_ctime = parent->i_ctime;
+	inode->i_atime = parent->i_atime;
 	inode->i_op = &ceph_snapdir_iops;
 	inode->i_fop = &ceph_snapdir_fops;
 	ci->i_snap_caps = CEPH_CAP_PIN; /* so we can open */
 	ci->i_rbytes = 0;
+	ci->i_btime = ceph_inode(parent)->i_btime;
+
+	if (inode->i_state & I_NEW)
+		unlock_new_inode(inode);
+
 	return inode;
 }
 
@@ -445,6 +451,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 	ci->i_max_files = 0;
 
 	memset(&ci->i_dir_layout, 0, sizeof(ci->i_dir_layout));
+	memset(&ci->i_cached_layout, 0, sizeof(ci->i_cached_layout));
 	RCU_INIT_POINTER(ci->i_layout.pool_ns, NULL);
 
 	ci->i_fragtree = RB_ROOT;
@@ -469,13 +476,13 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 	ci->i_prealloc_cap_flush = NULL;
 	INIT_LIST_HEAD(&ci->i_cap_flush_list);
 	init_waitqueue_head(&ci->i_cap_wq);
-	ci->i_hold_caps_min = 0;
 	ci->i_hold_caps_max = 0;
 	INIT_LIST_HEAD(&ci->i_cap_delay_list);
 	INIT_LIST_HEAD(&ci->i_cap_snaps);
 	ci->i_head_snapc = NULL;
 	ci->i_snap_caps = 0;
 
+	ci->i_last_rd = ci->i_last_wr = jiffies - 3600 * HZ;
 	for (i = 0; i < CEPH_FILE_MODE_BITS; i++)
 		ci->i_nr_by_mode[i] = 0;
 
@@ -494,6 +501,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 	ci->i_rdcache_ref = 0;
 	ci->i_wr_ref = 0;
 	ci->i_wb_ref = 0;
+	ci->i_fx_ref = 0;
 	ci->i_wrbuffer_ref = 0;
 	ci->i_wrbuffer_ref_head = 0;
 	atomic_set(&ci->i_filelock_ref, 0);
@@ -584,6 +592,7 @@ void ceph_evict_inode(struct inode *inode)
 		ceph_buffer_put(ci->i_xattrs.prealloc_blob);
 
 	ceph_put_string(rcu_dereference_raw(ci->i_layout.pool_ns));
+	ceph_put_string(rcu_dereference_raw(ci->i_cached_layout.pool_ns));
 }
 
 static inline blkcnt_t calc_inode_blocks(u64 size)
@@ -634,7 +643,7 @@ int ceph_fill_file_size(struct inode *inode, int issued,
 			if ((issued & (CEPH_CAP_FILE_CACHE|
 				       CEPH_CAP_FILE_BUFFER)) ||
 			    mapping_mapped(inode->i_mapping) ||
-			    __ceph_caps_file_wanted(ci)) {
+			    __ceph_is_file_opened(ci)) {
 				ci->i_truncate_pending++;
 				queue_trunc = 1;
 			}
@@ -725,12 +734,11 @@ void ceph_fill_file_time(struct inode *inode, int issued,
  * Populate an inode based on info from mds.  May be called on new or
  * existing inodes.
  */
-static int fill_inode(struct inode *inode, struct page *locked_page,
-		      struct ceph_mds_reply_info_in *iinfo,
-		      struct ceph_mds_reply_dirfrag *dirinfo,
-		      struct ceph_mds_session *session,
-		      unsigned long ttl_from, int cap_fmode,
-		      struct ceph_cap_reservation *caps_reservation)
+int ceph_fill_inode(struct inode *inode, struct page *locked_page,
+		    struct ceph_mds_reply_info_in *iinfo,
+		    struct ceph_mds_reply_dirfrag *dirinfo,
+		    struct ceph_mds_session *session, int cap_fmode,
+		    struct ceph_cap_reservation *caps_reservation)
 {
 	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	struct ceph_mds_reply_inode *info = iinfo->in;
@@ -747,15 +755,18 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 	bool new_version = false;
 	bool fill_inline = false;
 
-	dout("fill_inode %p ino %llx.%llx v %llu had %llu\n",
+	dout("%s %p ino %llx.%llx v %llu had %llu\n", __func__,
 	     inode, ceph_vinop(inode), le64_to_cpu(info->version),
 	     ci->i_version);
 
 	info_caps = le32_to_cpu(info->cap.caps);
 
 	/* prealloc new cap struct */
-	if (info_caps && ceph_snap(inode) == CEPH_NOSNAP)
+	if (info_caps && ceph_snap(inode) == CEPH_NOSNAP) {
 		new_cap = ceph_get_cap(mdsc, caps_reservation);
+		if (!new_cap)
+			return -ENOMEM;
+	}
 
 	/*
 	 * prealloc xattr data, if it looks like we'll need it.  only
@@ -765,7 +776,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 	if (iinfo->xattr_len > 4) {
 		xattr_blob = ceph_buffer_new(iinfo->xattr_len, GFP_NOFS);
 		if (!xattr_blob)
-			pr_err("fill_inode ENOMEM xattr blob %d bytes\n",
+			pr_err("%s ENOMEM xattr blob %d bytes\n", __func__,
 			       iinfo->xattr_len);
 	}
 
@@ -928,8 +939,9 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 			spin_unlock(&ci->i_ceph_lock);
 
 			if (symlen != i_size_read(inode)) {
-				pr_err("fill_inode %llx.%llx BAD symlink "
-					"size %lld\n", ceph_vinop(inode),
+				pr_err("%s %llx.%llx BAD symlink "
+					"size %lld\n", __func__,
+					ceph_vinop(inode),
 					i_size_read(inode));
 				i_size_write(inode, symlen);
 				inode->i_blocks = calc_inode_blocks(symlen);
@@ -953,7 +965,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 		inode->i_fop = &ceph_dir_fops;
 		break;
 	default:
-		pr_err("fill_inode %llx.%llx BAD mode 0%o\n",
+		pr_err("%s %llx.%llx BAD mode 0%o\n", __func__,
 		       ceph_vinop(inode), inode->i_mode);
 	}
 
@@ -962,7 +974,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 		if (ceph_snap(inode) == CEPH_NOSNAP) {
 			ceph_add_cap(inode, session,
 				     le64_to_cpu(info->cap.cap_id),
-				     cap_fmode, info_caps,
+				     info_caps,
 				     le32_to_cpu(info->cap.wanted),
 				     le32_to_cpu(info->cap.seq),
 				     le32_to_cpu(info->cap.mseq),
@@ -987,13 +999,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 			dout(" %p got snap_caps %s\n", inode,
 			     ceph_cap_string(info_caps));
 			ci->i_snap_caps |= info_caps;
-			if (cap_fmode >= 0)
-				__ceph_get_fmode(ci, cap_fmode);
 		}
-	} else if (cap_fmode >= 0) {
-		pr_warn("mds issued no caps on %llx.%llx\n",
-			   ceph_vinop(inode));
-		__ceph_get_fmode(ci, cap_fmode);
 	}
 
 	if (iinfo->inline_version > 0 &&
@@ -1005,6 +1011,13 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 			fill_inline = true;
 	}
 
+	if (cap_fmode >= 0) {
+		if (!info_caps)
+			pr_warn("mds issued no caps on %llx.%llx\n",
+				ceph_vinop(inode));
+		__ceph_touch_fmode(ci, mdsc, cap_fmode);
+	}
+
 	spin_unlock(&ci->i_ceph_lock);
 
 	if (fill_inline)
@@ -1046,6 +1059,7 @@ static void __update_dentry_lease(struct inode *dir, struct dentry *dentry,
 				  struct ceph_mds_session **old_lease_session)
 {
 	struct ceph_dentry_info *di = ceph_dentry(dentry);
+	unsigned mask = le16_to_cpu(lease->mask);
 	long unsigned duration = le32_to_cpu(lease->duration_ms);
 	long unsigned ttl = from_time + (duration * HZ) / 1000;
 	long unsigned half_ttl = from_time + (duration * HZ / 2) / 1000;
@@ -1057,8 +1071,13 @@ static void __update_dentry_lease(struct inode *dir, struct dentry *dentry,
 	if (ceph_snap(dir) != CEPH_NOSNAP)
 		return;
 
+	if (mask & CEPH_LEASE_PRIMARY_LINK)
+		di->flags |= CEPH_DENTRY_PRIMARY_LINK;
+	else
+		di->flags &= ~CEPH_DENTRY_PRIMARY_LINK;
+
 	di->lease_shared_gen = atomic_read(&ceph_inode(dir)->i_shared_gen);
-	if (duration == 0) {
+	if (!(mask & CEPH_LEASE_VALID)) {
 		__ceph_dentry_dir_lease_touch(di);
 		return;
 	}
@@ -1235,10 +1254,9 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req)
 		struct inode *dir = req->r_parent;
 
 		if (dir) {
-			err = fill_inode(dir, NULL,
-					 &rinfo->diri, rinfo->dirfrag,
-					 session, req->r_request_started, -1,
-					 &req->r_caps_reservation);
+			err = ceph_fill_inode(dir, NULL, &rinfo->diri,
+					      rinfo->dirfrag, session, -1,
+					      &req->r_caps_reservation);
 			if (err < 0)
 				goto done;
 		} else {
@@ -1302,18 +1320,23 @@ retry_lookup:
 			err = PTR_ERR(in);
 			goto done;
 		}
-		req->r_target_inode = in;
 
-		err = fill_inode(in, req->r_locked_page, &rinfo->targeti, NULL,
-				session, req->r_request_started,
+		err = ceph_fill_inode(in, req->r_locked_page, &rinfo->targeti,
+				NULL, session,
 				(!test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags) &&
-				rinfo->head->result == 0) ?  req->r_fmode : -1,
+				 !test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags) &&
+				 rinfo->head->result == 0) ?  req->r_fmode : -1,
 				&req->r_caps_reservation);
 		if (err < 0) {
-			pr_err("fill_inode badness %p %llx.%llx\n",
+			pr_err("ceph_fill_inode badness %p %llx.%llx\n",
 				in, ceph_vinop(in));
+			if (in->i_state & I_NEW)
+				discard_new_inode(in);
 			goto done;
 		}
+		req->r_target_inode = in;
+		if (in->i_state & I_NEW)
+			unlock_new_inode(in);
 	}
 
 	/*
@@ -1492,13 +1515,20 @@ static int readdir_prepopulate_inodes_only(struct ceph_mds_request *req,
 			dout("new_inode badness got %d\n", err);
 			continue;
 		}
-		rc = fill_inode(in, NULL, &rde->inode, NULL, session,
-				req->r_request_started, -1,
-				&req->r_caps_reservation);
+		rc = ceph_fill_inode(in, NULL, &rde->inode, NULL, session,
+				     -1, &req->r_caps_reservation);
 		if (rc < 0) {
-			pr_err("fill_inode badness on %p got %d\n", in, rc);
+			pr_err("ceph_fill_inode badness on %p got %d\n",
+			       in, rc);
 			err = rc;
+			if (in->i_state & I_NEW) {
+				ihold(in);
+				discard_new_inode(in);
+			}
+		} else if (in->i_state & I_NEW) {
+			unlock_new_inode(in);
 		}
+
 		/* avoid calling iput_final() in mds dispatch threads */
 		ceph_async_iput(in);
 	}
@@ -1693,20 +1723,25 @@ retry_lookup:
 			}
 		}
 
-		ret = fill_inode(in, NULL, &rde->inode, NULL, session,
-				 req->r_request_started, -1,
-				 &req->r_caps_reservation);
+		ret = ceph_fill_inode(in, NULL, &rde->inode, NULL, session,
+				      -1, &req->r_caps_reservation);
 		if (ret < 0) {
-			pr_err("fill_inode badness on %p\n", in);
+			pr_err("ceph_fill_inode badness on %p\n", in);
 			if (d_really_is_negative(dn)) {
 				/* avoid calling iput_final() in mds
 				 * dispatch threads */
+				if (in->i_state & I_NEW) {
+					ihold(in);
+					discard_new_inode(in);
+				}
 				ceph_async_iput(in);
 			}
 			d_drop(dn);
 			err = ret;
 			goto next_item;
 		}
+		if (in->i_state & I_NEW)
+			unlock_new_inode(in);
 
 		if (d_really_is_negative(dn)) {
 			if (ceph_security_xattr_deadlock(in)) {
@@ -1953,7 +1988,7 @@ retry:
 	mutex_unlock(&ci->i_truncate_mutex);
 
 	if (wrbuffer_refs == 0)
-		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
+		ceph_check_caps(ci, 0, NULL);
 
 	wake_up_all(&ci->i_cap_wq);
 }
diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index c90f03beb15d..6e061bf62ad4 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -243,11 +243,13 @@ static long ceph_ioctl_lazyio(struct file *file)
 	struct ceph_file_info *fi = file->private_data;
 	struct inode *inode = file_inode(file);
 	struct ceph_inode_info *ci = ceph_inode(inode);
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 
 	if ((fi->fmode & CEPH_FILE_MODE_LAZY) == 0) {
 		spin_lock(&ci->i_ceph_lock);
 		fi->fmode |= CEPH_FILE_MODE_LAZY;
 		ci->i_nr_by_mode[ffs(CEPH_FILE_MODE_LAZY)]++;
+		__ceph_touch_fmode(ci, mdsc, fi->fmode);
 		spin_unlock(&ci->i_ceph_lock);
 		dout("ioctl_layzio: file %p marked lazy\n", file);
 
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index 544e9e85b120..d6b9166e71e4 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -210,6 +210,21 @@ static int ceph_lock_wait_for_completion(struct ceph_mds_client *mdsc,
 	return 0;
 }
 
+static int try_unlock_file(struct file *file, struct file_lock *fl)
+{
+	int err;
+	unsigned int orig_flags = fl->fl_flags;
+	fl->fl_flags |= FL_EXISTS;
+	err = locks_lock_file_wait(file, fl);
+	fl->fl_flags = orig_flags;
+	if (err == -ENOENT) {
+		if (!(orig_flags & FL_EXISTS))
+			err = 0;
+		return err;
+	}
+	return 1;
+}
+
 /**
  * Attempt to set an fcntl lock.
  * For now, this just goes away to the server. Later it may be more awesome.
@@ -255,9 +270,15 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
 	else
 		lock_cmd = CEPH_LOCK_UNLOCK;
 
+	if (op == CEPH_MDS_OP_SETFILELOCK && F_UNLCK == fl->fl_type) {
+		err = try_unlock_file(file, fl);
+		if (err <= 0)
+			return err;
+	}
+
 	err = ceph_lock_message(CEPH_LOCK_FCNTL, op, inode, lock_cmd, wait, fl);
 	if (!err) {
-		if (op == CEPH_MDS_OP_SETFILELOCK) {
+		if (op == CEPH_MDS_OP_SETFILELOCK && F_UNLCK != fl->fl_type) {
 			dout("mds locked, locking locally\n");
 			err = posix_lock_file(file, fl, NULL);
 			if (err) {
@@ -311,9 +332,15 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
 	else
 		lock_cmd = CEPH_LOCK_UNLOCK;
 
+	if (F_UNLCK == fl->fl_type) {
+		err = try_unlock_file(file, fl);
+		if (err <= 0)
+			return err;
+	}
+
 	err = ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK,
 				inode, lock_cmd, wait, fl);
-	if (!err) {
+	if (!err && F_UNLCK != fl->fl_type) {
 		err = locks_lock_file_wait(file, fl);
 		if (err) {
 			ceph_lock_message(CEPH_LOCK_FLOCK,
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index a5163296d9d9..7c63abf5bea9 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -9,6 +9,7 @@
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
 #include <linux/ratelimit.h>
+#include <linux/bits.h>
 
 #include "super.h"
 #include "mds_client.h"
@@ -414,21 +415,121 @@ bad:
 	return -EIO;
 }
 
+
+#if BITS_PER_LONG == 64
+
+#define DELEGATED_INO_AVAILABLE		xa_mk_value(1)
+
+static int ceph_parse_deleg_inos(void **p, void *end,
+				 struct ceph_mds_session *s)
+{
+	u32 sets;
+
+	ceph_decode_32_safe(p, end, sets, bad);
+	dout("got %u sets of delegated inodes\n", sets);
+	while (sets--) {
+		u64 start, len, ino;
+
+		ceph_decode_64_safe(p, end, start, bad);
+		ceph_decode_64_safe(p, end, len, bad);
+		while (len--) {
+			int err = xa_insert(&s->s_delegated_inos, ino = start++,
+					    DELEGATED_INO_AVAILABLE,
+					    GFP_KERNEL);
+			if (!err) {
+				dout("added delegated inode 0x%llx\n",
+				     start - 1);
+			} else if (err == -EBUSY) {
+				pr_warn("ceph: MDS delegated inode 0x%llx more than once.\n",
+					start - 1);
+			} else {
+				return err;
+			}
+		}
+	}
+	return 0;
+bad:
+	return -EIO;
+}
+
+u64 ceph_get_deleg_ino(struct ceph_mds_session *s)
+{
+	unsigned long ino;
+	void *val;
+
+	xa_for_each(&s->s_delegated_inos, ino, val) {
+		val = xa_erase(&s->s_delegated_inos, ino);
+		if (val == DELEGATED_INO_AVAILABLE)
+			return ino;
+	}
+	return 0;
+}
+
+int ceph_restore_deleg_ino(struct ceph_mds_session *s, u64 ino)
+{
+	return xa_insert(&s->s_delegated_inos, ino, DELEGATED_INO_AVAILABLE,
+			 GFP_KERNEL);
+}
+#else /* BITS_PER_LONG == 64 */
+/*
+ * FIXME: xarrays can't handle 64-bit indexes on a 32-bit arch. For now, just
+ * ignore delegated_inos on 32 bit arch. Maybe eventually add xarrays for top
+ * and bottom words?
+ */
+static int ceph_parse_deleg_inos(void **p, void *end,
+				 struct ceph_mds_session *s)
+{
+	u32 sets;
+
+	ceph_decode_32_safe(p, end, sets, bad);
+	if (sets)
+		ceph_decode_skip_n(p, end, sets * 2 * sizeof(__le64), bad);
+	return 0;
+bad:
+	return -EIO;
+}
+
+u64 ceph_get_deleg_ino(struct ceph_mds_session *s)
+{
+	return 0;
+}
+
+int ceph_restore_deleg_ino(struct ceph_mds_session *s, u64 ino)
+{
+	return 0;
+}
+#endif /* BITS_PER_LONG == 64 */
+
 /*
  * parse create results
  */
 static int parse_reply_info_create(void **p, void *end,
 				  struct ceph_mds_reply_info_parsed *info,
-				  u64 features)
+				  u64 features, struct ceph_mds_session *s)
 {
+	int ret;
+
 	if (features == (u64)-1 ||
 	    (features & CEPH_FEATURE_REPLY_CREATE_INODE)) {
-		/* Malformed reply? */
 		if (*p == end) {
+			/* Malformed reply? */
 			info->has_create_ino = false;
-		} else {
+		} else if (test_bit(CEPHFS_FEATURE_DELEG_INO, &s->s_features)) {
+			u8 struct_v, struct_compat;
+			u32 len;
+
 			info->has_create_ino = true;
+			ceph_decode_8_safe(p, end, struct_v, bad);
+			ceph_decode_8_safe(p, end, struct_compat, bad);
+			ceph_decode_32_safe(p, end, len, bad);
+			ceph_decode_64_safe(p, end, info->ino, bad);
+			ret = ceph_parse_deleg_inos(p, end, s);
+			if (ret)
+				return ret;
+		} else {
+			/* legacy */
 			ceph_decode_64_safe(p, end, info->ino, bad);
+			info->has_create_ino = true;
 		}
 	} else {
 		if (*p != end)
@@ -447,7 +548,7 @@ bad:
  */
 static int parse_reply_info_extra(void **p, void *end,
 				  struct ceph_mds_reply_info_parsed *info,
-				  u64 features)
+				  u64 features, struct ceph_mds_session *s)
 {
 	u32 op = le32_to_cpu(info->head->op);
 
@@ -456,7 +557,7 @@ static int parse_reply_info_extra(void **p, void *end,
 	else if (op == CEPH_MDS_OP_READDIR || op == CEPH_MDS_OP_LSSNAP)
 		return parse_reply_info_readdir(p, end, info, features);
 	else if (op == CEPH_MDS_OP_CREATE)
-		return parse_reply_info_create(p, end, info, features);
+		return parse_reply_info_create(p, end, info, features, s);
 	else
 		return -EIO;
 }
@@ -464,7 +565,7 @@ static int parse_reply_info_extra(void **p, void *end,
 /*
  * parse entire mds reply
  */
-static int parse_reply_info(struct ceph_msg *msg,
+static int parse_reply_info(struct ceph_mds_session *s, struct ceph_msg *msg,
 			    struct ceph_mds_reply_info_parsed *info,
 			    u64 features)
 {
@@ -489,7 +590,7 @@ static int parse_reply_info(struct ceph_msg *msg,
 	ceph_decode_32_safe(&p, end, len, bad);
 	if (len > 0) {
 		ceph_decode_need(&p, end, len, bad);
-		err = parse_reply_info_extra(&p, p+len, info, features);
+		err = parse_reply_info_extra(&p, p+len, info, features, s);
 		if (err < 0)
 			goto out_bad;
 	}
@@ -530,6 +631,7 @@ const char *ceph_session_state_name(int s)
 	case CEPH_MDS_SESSION_OPEN: return "open";
 	case CEPH_MDS_SESSION_HUNG: return "hung";
 	case CEPH_MDS_SESSION_CLOSING: return "closing";
+	case CEPH_MDS_SESSION_CLOSED: return "closed";
 	case CEPH_MDS_SESSION_RESTARTING: return "restarting";
 	case CEPH_MDS_SESSION_RECONNECTING: return "reconnecting";
 	case CEPH_MDS_SESSION_REJECTED: return "rejected";
@@ -537,7 +639,7 @@ const char *ceph_session_state_name(int s)
 	}
 }
 
-static struct ceph_mds_session *get_session(struct ceph_mds_session *s)
+struct ceph_mds_session *ceph_get_mds_session(struct ceph_mds_session *s)
 {
 	if (refcount_inc_not_zero(&s->s_ref)) {
 		dout("mdsc get_session %p %d -> %d\n", s,
@@ -556,6 +658,7 @@ void ceph_put_mds_session(struct ceph_mds_session *s)
 	if (refcount_dec_and_test(&s->s_ref)) {
 		if (s->s_auth.authorizer)
 			ceph_auth_destroy_authorizer(s->s_auth.authorizer);
+		xa_destroy(&s->s_delegated_inos);
 		kfree(s);
 	}
 }
@@ -568,7 +671,7 @@ struct ceph_mds_session *__ceph_lookup_mds_session(struct ceph_mds_client *mdsc,
 {
 	if (mds >= mdsc->max_sessions || !mdsc->sessions[mds])
 		return NULL;
-	return get_session(mdsc->sessions[mds]);
+	return ceph_get_mds_session(mdsc->sessions[mds]);
 }
 
 static bool __have_session(struct ceph_mds_client *mdsc, int mds)
@@ -597,7 +700,7 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
 {
 	struct ceph_mds_session *s;
 
-	if (mds >= mdsc->mdsmap->m_num_mds)
+	if (mds >= mdsc->mdsmap->possible_max_rank)
 		return ERR_PTR(-EINVAL);
 
 	s = kzalloc(sizeof(*s), GFP_NOFS);
@@ -643,6 +746,7 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
 	refcount_set(&s->s_ref, 1);
 	INIT_LIST_HEAD(&s->s_waiting);
 	INIT_LIST_HEAD(&s->s_unsafe);
+	xa_init(&s->s_delegated_inos);
 	s->s_num_cap_releases = 0;
 	s->s_cap_reconnect = 0;
 	s->s_cap_iterator = NULL;
@@ -674,7 +778,6 @@ static void __unregister_session(struct ceph_mds_client *mdsc,
 	dout("__unregister_session mds%d %p\n", s->s_mds, s);
 	BUG_ON(mdsc->sessions[s->s_mds] != s);
 	mdsc->sessions[s->s_mds] = NULL;
-	s->s_state = 0;
 	ceph_con_close(&s->s_con);
 	ceph_put_mds_session(s);
 	atomic_dec(&mdsc->num_sessions);
@@ -698,6 +801,7 @@ void ceph_mdsc_release_request(struct kref *kref)
 	struct ceph_mds_request *req = container_of(kref,
 						    struct ceph_mds_request,
 						    r_kref);
+	ceph_mdsc_release_dir_caps(req);
 	destroy_reply_info(&req->r_reply_info);
 	if (req->r_request)
 		ceph_msg_put(req->r_request);
@@ -708,8 +812,10 @@ void ceph_mdsc_release_request(struct kref *kref)
 		/* avoid calling iput_final() in mds dispatch threads */
 		ceph_async_iput(req->r_inode);
 	}
-	if (req->r_parent)
+	if (req->r_parent) {
 		ceph_put_cap_refs(ceph_inode(req->r_parent), CEPH_CAP_PIN);
+		ceph_async_iput(req->r_parent);
+	}
 	ceph_async_iput(req->r_target_inode);
 	if (req->r_dentry)
 		dput(req->r_dentry);
@@ -733,7 +839,7 @@ void ceph_mdsc_release_request(struct kref *kref)
 	put_request_session(req);
 	ceph_unreserve_caps(req->r_mdsc, &req->r_caps_reservation);
 	WARN_ON_ONCE(!list_empty(&req->r_wait));
-	kfree(req);
+	kmem_cache_free(ceph_mds_request_cachep, req);
 }
 
 DEFINE_RB_FUNCS(request, struct ceph_mds_request, r_tid, r_node)
@@ -790,8 +896,13 @@ static void __register_request(struct ceph_mds_client *mdsc,
 		mdsc->oldest_tid = req->r_tid;
 
 	if (dir) {
+		struct ceph_inode_info *ci = ceph_inode(dir);
+
 		ihold(dir);
 		req->r_unsafe_dir = dir;
+		spin_lock(&ci->i_unsafe_lock);
+		list_add_tail(&req->r_unsafe_dir_item, &ci->i_unsafe_dirops);
+		spin_unlock(&ci->i_unsafe_lock);
 	}
 }
 
@@ -819,8 +930,7 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
 
 	erase_request(&mdsc->request_tree, req);
 
-	if (req->r_unsafe_dir  &&
-	    test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
+	if (req->r_unsafe_dir) {
 		struct ceph_inode_info *ci = ceph_inode(req->r_unsafe_dir);
 		spin_lock(&ci->i_unsafe_lock);
 		list_del_init(&req->r_unsafe_dir_item);
@@ -876,7 +986,8 @@ static struct inode *get_nonsnap_parent(struct dentry *dentry)
  * Called under mdsc->mutex.
  */
 static int __choose_mds(struct ceph_mds_client *mdsc,
-			struct ceph_mds_request *req)
+			struct ceph_mds_request *req,
+			bool *random)
 {
 	struct inode *inode;
 	struct ceph_inode_info *ci;
@@ -886,6 +997,9 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 	u32 hash = req->r_direct_hash;
 	bool is_hash = test_bit(CEPH_MDS_R_DIRECT_IS_HASH, &req->r_req_flags);
 
+	if (random)
+		*random = false;
+
 	/*
 	 * is there a specific mds we should try?  ignore hint if we have
 	 * no session and the mds is not up (active or recovering).
@@ -893,7 +1007,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 	if (req->r_resend_mds >= 0 &&
 	    (__have_session(mdsc, req->r_resend_mds) ||
 	     ceph_mdsmap_get_state(mdsc->mdsmap, req->r_resend_mds) > 0)) {
-		dout("choose_mds using resend_mds mds%d\n",
+		dout("%s using resend_mds mds%d\n", __func__,
 		     req->r_resend_mds);
 		return req->r_resend_mds;
 	}
@@ -911,7 +1025,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 			rcu_read_lock();
 			inode = get_nonsnap_parent(req->r_dentry);
 			rcu_read_unlock();
-			dout("__choose_mds using snapdir's parent %p\n", inode);
+			dout("%s using snapdir's parent %p\n", __func__, inode);
 		}
 	} else if (req->r_dentry) {
 		/* ignore race with rename; old or new d_parent is okay */
@@ -931,7 +1045,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 			/* direct snapped/virtual snapdir requests
 			 * based on parent dir inode */
 			inode = get_nonsnap_parent(parent);
-			dout("__choose_mds using nonsnap parent %p\n", inode);
+			dout("%s using nonsnap parent %p\n", __func__, inode);
 		} else {
 			/* dentry target */
 			inode = d_inode(req->r_dentry);
@@ -947,8 +1061,8 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 		rcu_read_unlock();
 	}
 
-	dout("__choose_mds %p is_hash=%d (%d) mode %d\n", inode, (int)is_hash,
-	     (int)hash, mode);
+	dout("%s %p is_hash=%d (0x%x) mode %d\n", __func__, inode, (int)is_hash,
+	     hash, mode);
 	if (!inode)
 		goto random;
 	ci = ceph_inode(inode);
@@ -966,30 +1080,33 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 				get_random_bytes(&r, 1);
 				r %= frag.ndist;
 				mds = frag.dist[r];
-				dout("choose_mds %p %llx.%llx "
-				     "frag %u mds%d (%d/%d)\n",
-				     inode, ceph_vinop(inode),
-				     frag.frag, mds,
-				     (int)r, frag.ndist);
+				dout("%s %p %llx.%llx frag %u mds%d (%d/%d)\n",
+				     __func__, inode, ceph_vinop(inode),
+				     frag.frag, mds, (int)r, frag.ndist);
 				if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
-				    CEPH_MDS_STATE_ACTIVE)
+				    CEPH_MDS_STATE_ACTIVE &&
+				    !ceph_mdsmap_is_laggy(mdsc->mdsmap, mds))
 					goto out;
 			}
 
 			/* since this file/dir wasn't known to be
 			 * replicated, then we want to look for the
 			 * authoritative mds. */
-			mode = USE_AUTH_MDS;
 			if (frag.mds >= 0) {
 				/* choose auth mds */
 				mds = frag.mds;
-				dout("choose_mds %p %llx.%llx "
-				     "frag %u mds%d (auth)\n",
-				     inode, ceph_vinop(inode), frag.frag, mds);
+				dout("%s %p %llx.%llx frag %u mds%d (auth)\n",
+				     __func__, inode, ceph_vinop(inode),
+				     frag.frag, mds);
 				if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
-				    CEPH_MDS_STATE_ACTIVE)
-					goto out;
+				    CEPH_MDS_STATE_ACTIVE) {
+					if (mode == USE_ANY_MDS &&
+					    !ceph_mdsmap_is_laggy(mdsc->mdsmap,
+								  mds))
+						goto out;
+				}
 			}
+			mode = USE_AUTH_MDS;
 		}
 	}
 
@@ -1005,7 +1122,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 		goto random;
 	}
 	mds = cap->session->s_mds;
-	dout("choose_mds %p %llx.%llx mds%d (%scap %p)\n",
+	dout("%s %p %llx.%llx mds%d (%scap %p)\n", __func__,
 	     inode, ceph_vinop(inode), mds,
 	     cap == ci->i_auth_cap ? "auth " : "", cap);
 	spin_unlock(&ci->i_ceph_lock);
@@ -1016,8 +1133,11 @@ out:
 	return mds;
 
 random:
+	if (random)
+		*random = true;
+
 	mds = ceph_mdsmap_get_random_mds(mdsc->mdsmap);
-	dout("choose_mds chose random mds%d\n", mds);
+	dout("%s chose random mds%d\n", __func__, mds);
 	return mds;
 }
 
@@ -1043,20 +1163,21 @@ static struct ceph_msg *create_session_msg(u32 op, u64 seq)
 	return msg;
 }
 
+static const unsigned char feature_bits[] = CEPHFS_FEATURES_CLIENT_SUPPORTED;
+#define FEATURE_BYTES(c) (DIV_ROUND_UP((size_t)feature_bits[c - 1] + 1, 64) * 8)
 static void encode_supported_features(void **p, void *end)
 {
-	static const unsigned char bits[] = CEPHFS_FEATURES_CLIENT_SUPPORTED;
-	static const size_t count = ARRAY_SIZE(bits);
+	static const size_t count = ARRAY_SIZE(feature_bits);
 
 	if (count > 0) {
 		size_t i;
-		size_t size = ((size_t)bits[count - 1] + 64) / 64 * 8;
+		size_t size = FEATURE_BYTES(count);
 
 		BUG_ON(*p + 4 + size > end);
 		ceph_encode_32(p, size);
 		memset(*p, 0, size);
 		for (i = 0; i < count; i++)
-			((unsigned char*)(*p))[i / 8] |= 1 << (bits[i] % 8);
+			((unsigned char*)(*p))[i / 8] |= BIT(feature_bits[i] % 8);
 		*p += size;
 	} else {
 		BUG_ON(*p + 4 > end);
@@ -1077,6 +1198,7 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 	int metadata_key_count = 0;
 	struct ceph_options *opt = mdsc->fsc->client->options;
 	struct ceph_mount_options *fsopt = mdsc->fsc->mount_options;
+	size_t size, count;
 	void *p, *end;
 
 	const char* metadata[][2] = {
@@ -1094,8 +1216,13 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 			strlen(metadata[i][1]);
 		metadata_key_count++;
 	}
+
 	/* supported feature */
-	extra_bytes += 4 + 8;
+	size = 0;
+	count = ARRAY_SIZE(feature_bits);
+	if (count > 0)
+		size = FEATURE_BYTES(count);
+	extra_bytes += 4 + size;
 
 	/* Allocate the message */
 	msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h) + extra_bytes,
@@ -1115,7 +1242,7 @@ static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u6
 	 * Serialize client metadata into waiting buffer space, using
 	 * the format that userspace expects for map<string, string>
 	 *
-	 * ClientSession messages with metadata are v2
+	 * ClientSession messages with metadata are v3
 	 */
 	msg->hdr.version = cpu_to_le16(3);
 	msg->hdr.compat_version = cpu_to_le16(1);
@@ -1217,7 +1344,7 @@ static void __open_export_target_sessions(struct ceph_mds_client *mdsc,
 	struct ceph_mds_session *ts;
 	int i, mds = session->s_mds;
 
-	if (mds >= mdsc->mdsmap->m_num_mds)
+	if (mds >= mdsc->mdsmap->possible_max_rank)
 		return;
 
 	mi = &mdsc->mdsmap->m_info[mds];
@@ -1387,8 +1514,6 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap,
 	dout("removing cap %p, ci is %p, inode is %p\n",
 	     cap, ci, &ci->vfs_inode);
 	spin_lock(&ci->i_ceph_lock);
-	if (cap->mds_wanted | cap->issued)
-		ci->i_ceph_flags |= CEPH_I_CAP_DROPPED;
 	__ceph_remove_cap(cap, false);
 	if (!ci->i_auth_cap) {
 		struct ceph_cap_flush *cf;
@@ -1554,9 +1679,6 @@ static int wake_up_session_cb(struct inode *inode, struct ceph_cap *cap,
 			/* mds did not re-issue stale cap */
 			spin_lock(&ci->i_ceph_lock);
 			cap->issued = cap->implemented = CEPH_CAP_PIN;
-			/* make sure mds knows what we want */
-			if (__ceph_caps_file_wanted(ci) & ~cap->mds_wanted)
-				ci->i_ceph_flags |= CEPH_I_CAP_DROPPED;
 			spin_unlock(&ci->i_ceph_lock);
 		}
 	} else if (ev == FORCE_RO) {
@@ -1752,7 +1874,8 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg)
 	}
 	/* The inode has cached pages, but it's no longer used.
 	 * we can safely drop it */
-	if (wanted == 0 && used == CEPH_CAP_FILE_CACHE &&
+	if (S_ISREG(inode->i_mode) &&
+	    wanted == 0 && used == CEPH_CAP_FILE_CACHE &&
 	    !(oissued & CEPH_CAP_FILE_CACHE)) {
 	  used = 0;
 	  oissued = 0;
@@ -1965,7 +2088,7 @@ void ceph_flush_cap_releases(struct ceph_mds_client *mdsc,
 	if (mdsc->stopping)
 		return;
 
-	get_session(session);
+	ceph_get_mds_session(session);
 	if (queue_work(mdsc->fsc->cap_wq,
 		       &session->s_cap_release_work)) {
 		dout("cap release work queued\n");
@@ -2015,7 +2138,7 @@ void ceph_reclaim_caps_nr(struct ceph_mds_client *mdsc, int nr)
 	if (!nr)
 		return;
 	val = atomic_add_return(nr, &mdsc->cap_reclaim_pending);
-	if (!(val % CEPH_CAPS_PER_RELEASE)) {
+	if ((val % CEPH_CAPS_PER_RELEASE) < nr) {
 		atomic_set(&mdsc->cap_reclaim_pending, 0);
 		ceph_queue_cap_reclaim_work(mdsc);
 	}
@@ -2032,12 +2155,13 @@ int ceph_alloc_readdir_reply_buffer(struct ceph_mds_request *req,
 	struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
 	struct ceph_mount_options *opt = req->r_mdsc->fsc->mount_options;
 	size_t size = sizeof(struct ceph_mds_reply_dir_entry);
-	int order, num_entries;
+	unsigned int num_entries;
+	int order;
 
 	spin_lock(&ci->i_ceph_lock);
 	num_entries = ci->i_files + ci->i_subdirs;
 	spin_unlock(&ci->i_ceph_lock);
-	num_entries = max(num_entries, 1);
+	num_entries = max(num_entries, 1U);
 	num_entries = min(num_entries, opt->max_readdir);
 
 	order = get_order(size * num_entries);
@@ -2068,9 +2192,9 @@ int ceph_alloc_readdir_reply_buffer(struct ceph_mds_request *req,
 struct ceph_mds_request *
 ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode)
 {
-	struct ceph_mds_request *req = kzalloc(sizeof(*req), GFP_NOFS);
-	struct timespec64 ts;
+	struct ceph_mds_request *req;
 
+	req = kmem_cache_zalloc(ceph_mds_request_cachep, GFP_NOFS);
 	if (!req)
 		return ERR_PTR(-ENOMEM);
 
@@ -2088,8 +2212,7 @@ ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode)
 	init_completion(&req->r_safe_completion);
 	INIT_LIST_HEAD(&req->r_unsafe_item);
 
-	ktime_get_coarse_real_ts64(&ts);
-	req->r_stamp = timespec64_trunc(ts, mdsc->fsc->sb->s_time_gran);
+	ktime_get_coarse_real_ts64(&req->r_stamp);
 
 	req->r_op = op;
 	req->r_direct_mode = mode;
@@ -2182,13 +2305,17 @@ retry:
 	}
 	base = ceph_ino(d_inode(temp));
 	rcu_read_unlock();
-	if (pos < 0 || read_seqretry(&rename_lock, seq)) {
-		pr_err("build_path did not end path lookup where "
-		       "expected, pos is %d\n", pos);
-		/* presumably this is only possible if racing with a
-		   rename of one of the parent directories (we can not
-		   lock the dentries above us to prevent this, but
-		   retrying should be harmless) */
+
+	if (read_seqretry(&rename_lock, seq))
+		goto retry;
+
+	if (pos < 0) {
+		/*
+		 * A rename didn't occur, but somehow we didn't end up where
+		 * we thought we would. Throw a warning and try again.
+		 */
+		pr_warn("build_path did not end path lookup where "
+			"expected, pos is %d\n", pos);
 		goto retry;
 	}
 
@@ -2345,6 +2472,7 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
 	head->op = cpu_to_le32(req->r_op);
 	head->caller_uid = cpu_to_le32(from_kuid(&init_user_ns, req->r_uid));
 	head->caller_gid = cpu_to_le32(from_kgid(&init_user_ns, req->r_gid));
+	head->ino = cpu_to_le64(req->r_deleg_ino);
 	head->args = req->r_args;
 
 	ceph_encode_filepath(&p, end, ino1, path1);
@@ -2358,7 +2486,8 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
 	if (req->r_inode_drop)
 		releases += ceph_encode_inode_release(&p,
 		      req->r_inode ? req->r_inode : d_inode(req->r_dentry),
-		      mds, req->r_inode_drop, req->r_inode_unless, 0);
+		      mds, req->r_inode_drop, req->r_inode_unless,
+		      req->r_op == CEPH_MDS_OP_READDIR);
 	if (req->r_dentry_drop)
 		releases += ceph_encode_dentry_release(&p, req->r_dentry,
 				req->r_parent, mds, req->r_dentry_drop,
@@ -2498,18 +2627,39 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
 	rhead->oldest_client_tid = cpu_to_le64(__get_oldest_tid(mdsc));
 	if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags))
 		flags |= CEPH_MDS_FLAG_REPLAY;
+	if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags))
+		flags |= CEPH_MDS_FLAG_ASYNC;
 	if (req->r_parent)
 		flags |= CEPH_MDS_FLAG_WANT_DENTRY;
 	rhead->flags = cpu_to_le32(flags);
 	rhead->num_fwd = req->r_num_fwd;
 	rhead->num_retry = req->r_attempts - 1;
-	rhead->ino = 0;
 
 	dout(" r_parent = %p\n", req->r_parent);
 	return 0;
 }
 
 /*
+ * called under mdsc->mutex
+ */
+static int __send_request(struct ceph_mds_client *mdsc,
+			  struct ceph_mds_session *session,
+			  struct ceph_mds_request *req,
+			  bool drop_cap_releases)
+{
+	int err;
+
+	err = __prepare_send_request(mdsc, req, session->s_mds,
+				     drop_cap_releases);
+	if (!err) {
+		ceph_msg_get(req->r_request);
+		ceph_con_send(&session->s_con, req->r_request);
+	}
+
+	return err;
+}
+
+/*
  * send request, or put it on the appropriate wait list.
  */
 static void __do_request(struct ceph_mds_client *mdsc,
@@ -2518,6 +2668,7 @@ static void __do_request(struct ceph_mds_client *mdsc,
 	struct ceph_mds_session *session = NULL;
 	int mds = -1;
 	int err = 0;
+	bool random;
 
 	if (req->r_err || test_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags)) {
 		if (test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags))
@@ -2528,7 +2679,7 @@ static void __do_request(struct ceph_mds_client *mdsc,
 	if (req->r_timeout &&
 	    time_after_eq(jiffies, req->r_started + req->r_timeout)) {
 		dout("do_request timed out\n");
-		err = -EIO;
+		err = -ETIMEDOUT;
 		goto finish;
 	}
 	if (READ_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_SHUTDOWN) {
@@ -2550,17 +2701,20 @@ static void __do_request(struct ceph_mds_client *mdsc,
 		if (!(mdsc->fsc->mount_options->flags &
 		      CEPH_MOUNT_OPT_MOUNTWAIT) &&
 		    !ceph_mdsmap_is_cluster_available(mdsc->mdsmap)) {
-			err = -ENOENT;
-			pr_info("probably no mds server is up\n");
+			err = -EHOSTUNREACH;
 			goto finish;
 		}
 	}
 
 	put_request_session(req);
 
-	mds = __choose_mds(mdsc, req);
+	mds = __choose_mds(mdsc, req, &random);
 	if (mds < 0 ||
 	    ceph_mdsmap_get_state(mdsc->mdsmap, mds) < CEPH_MDS_STATE_ACTIVE) {
+		if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags)) {
+			err = -EJUKEBOX;
+			goto finish;
+		}
 		dout("do_request no mds or not active, waiting for map\n");
 		list_add(&req->r_wait, &mdsc->waiting_for_map);
 		return;
@@ -2575,7 +2729,7 @@ static void __do_request(struct ceph_mds_client *mdsc,
 			goto finish;
 		}
 	}
-	req->r_session = get_session(session);
+	req->r_session = ceph_get_mds_session(session);
 
 	dout("do_request mds%d session %p state %s\n", mds, session,
 	     ceph_session_state_name(session->s_state));
@@ -2585,9 +2739,22 @@ static void __do_request(struct ceph_mds_client *mdsc,
 			err = -EACCES;
 			goto out_session;
 		}
+		/*
+		 * We cannot queue async requests since the caps and delegated
+		 * inodes are bound to the session. Just return -EJUKEBOX and
+		 * let the caller retry a sync request in that case.
+		 */
+		if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags)) {
+			err = -EJUKEBOX;
+			goto out_session;
+		}
 		if (session->s_state == CEPH_MDS_SESSION_NEW ||
-		    session->s_state == CEPH_MDS_SESSION_CLOSING)
+		    session->s_state == CEPH_MDS_SESSION_CLOSING) {
 			__open_session(mdsc, session);
+			/* retry the same mds later */
+			if (random)
+				req->r_resend_mds = mds;
+		}
 		list_add(&req->r_wait, &session->s_waiting);
 		goto out_session;
 	}
@@ -2598,11 +2765,7 @@ static void __do_request(struct ceph_mds_client *mdsc,
 	if (req->r_request_started == 0)   /* note request start time */
 		req->r_request_started = jiffies;
 
-	err = __prepare_send_request(mdsc, req, mds, false);
-	if (!err) {
-		ceph_msg_get(req->r_request);
-		ceph_con_send(&session->s_con, req->r_request);
-	}
+	err = __send_request(mdsc, session, req, false);
 
 out_session:
 	ceph_put_mds_session(session);
@@ -2665,17 +2828,43 @@ static void kick_requests(struct ceph_mds_client *mdsc, int mds)
 int ceph_mdsc_submit_request(struct ceph_mds_client *mdsc, struct inode *dir,
 			      struct ceph_mds_request *req)
 {
-	int err;
+	int err = 0;
 
 	/* take CAP_PIN refs for r_inode, r_parent, r_old_dentry */
 	if (req->r_inode)
 		ceph_get_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN);
-	if (req->r_parent)
-		ceph_get_cap_refs(ceph_inode(req->r_parent), CEPH_CAP_PIN);
+	if (req->r_parent) {
+		struct ceph_inode_info *ci = ceph_inode(req->r_parent);
+		int fmode = (req->r_op & CEPH_MDS_OP_WRITE) ?
+			    CEPH_FILE_MODE_WR : CEPH_FILE_MODE_RD;
+		spin_lock(&ci->i_ceph_lock);
+		ceph_take_cap_refs(ci, CEPH_CAP_PIN, false);
+		__ceph_touch_fmode(ci, mdsc, fmode);
+		spin_unlock(&ci->i_ceph_lock);
+		ihold(req->r_parent);
+	}
 	if (req->r_old_dentry_dir)
 		ceph_get_cap_refs(ceph_inode(req->r_old_dentry_dir),
 				  CEPH_CAP_PIN);
 
+	if (req->r_inode) {
+		err = ceph_wait_on_async_create(req->r_inode);
+		if (err) {
+			dout("%s: wait for async create returned: %d\n",
+			     __func__, err);
+			return err;
+		}
+	}
+
+	if (!err && req->r_old_inode) {
+		err = ceph_wait_on_async_create(req->r_old_inode);
+		if (err) {
+			dout("%s: wait for async create returned: %d\n",
+			     __func__, err);
+			return err;
+		}
+	}
+
 	dout("submit_request on %p for inode %p\n", req, dir);
 	mutex_lock(&mdsc->mutex);
 	__register_request(mdsc, req, dir);
@@ -2701,7 +2890,7 @@ static int ceph_mdsc_wait_request(struct ceph_mds_client *mdsc,
 		if (timeleft > 0)
 			err = 0;
 		else if (!timeleft)
-			err = -EIO;  /* timed out */
+			err = -ETIMEDOUT;  /* timed out */
 		else
 			err = timeleft;  /* killed */
 	}
@@ -2853,7 +3042,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 			mutex_unlock(&mdsc->mutex);
 			goto out;
 		} else  {
-			int mds = __choose_mds(mdsc, req);
+			int mds = __choose_mds(mdsc, req, NULL);
 			if (mds >= 0 && mds != req->r_session->s_mds) {
 				dout("but auth changed, so resending\n");
 				__do_request(mdsc, req);
@@ -2869,6 +3058,10 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 		set_bit(CEPH_MDS_R_GOT_SAFE, &req->r_req_flags);
 		__unregister_request(mdsc, req);
 
+		/* last request during umount? */
+		if (mdsc->stopping && !__get_oldest_req(mdsc))
+			complete_all(&mdsc->safe_umount_waiters);
+
 		if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
 			/*
 			 * We already handled the unsafe response, now do the
@@ -2879,31 +3072,20 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 			 */
 			dout("got safe reply %llu, mds%d\n", tid, mds);
 
-			/* last unsafe request during umount? */
-			if (mdsc->stopping && !__get_oldest_req(mdsc))
-				complete_all(&mdsc->safe_umount_waiters);
 			mutex_unlock(&mdsc->mutex);
 			goto out;
 		}
 	} else {
 		set_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags);
 		list_add_tail(&req->r_unsafe_item, &req->r_session->s_unsafe);
-		if (req->r_unsafe_dir) {
-			struct ceph_inode_info *ci =
-					ceph_inode(req->r_unsafe_dir);
-			spin_lock(&ci->i_unsafe_lock);
-			list_add_tail(&req->r_unsafe_dir_item,
-				      &ci->i_unsafe_dirops);
-			spin_unlock(&ci->i_unsafe_lock);
-		}
 	}
 
 	dout("handle_reply tid %lld result %d\n", tid, result);
 	rinfo = &req->r_reply_info;
 	if (test_bit(CEPHFS_FEATURE_REPLY_ENCODING, &session->s_features))
-		err = parse_reply_info(msg, rinfo, (u64)-1);
+		err = parse_reply_info(session, msg, rinfo, (u64)-1);
 	else
-		err = parse_reply_info(msg, rinfo, session->s_con.peer_features);
+		err = parse_reply_info(session, msg, rinfo, session->s_con.peer_features);
 	mutex_unlock(&mdsc->mutex);
 
 	mutex_lock(&session->s_mutex);
@@ -3069,8 +3251,7 @@ static void handle_session(struct ceph_mds_session *session,
 	void *end = p + msg->front.iov_len;
 	struct ceph_mds_session_head *h;
 	u32 op;
-	u64 seq;
-	unsigned long features = 0;
+	u64 seq, features = 0;
 	int wake = 0;
 	bool blacklisted = false;
 
@@ -3089,14 +3270,13 @@ static void handle_session(struct ceph_mds_session *session,
 			goto bad;
 		/* version >= 3, feature bits */
 		ceph_decode_32_safe(&p, end, len, bad);
-		ceph_decode_need(&p, end, len, bad);
-		memcpy(&features, p, min_t(size_t, len, sizeof(features)));
-		p += len;
+		ceph_decode_64_safe(&p, end, features, bad);
+		p += len - sizeof(features);
 	}
 
 	mutex_lock(&mdsc->mutex);
 	if (op == CEPH_SESSION_CLOSE) {
-		get_session(session);
+		ceph_get_mds_session(session);
 		__unregister_session(mdsc, session);
 	}
 	/* FIXME: this ttl calculation is generous */
@@ -3134,6 +3314,7 @@ static void handle_session(struct ceph_mds_session *session,
 	case CEPH_SESSION_CLOSE:
 		if (session->s_state == CEPH_MDS_SESSION_RECONNECTING)
 			pr_info("mds%d reconnect denied\n", session->s_mds);
+		session->s_state = CEPH_MDS_SESSION_CLOSED;
 		cleanup_session_requests(mdsc, session);
 		remove_session_caps(session);
 		wake = 2; /* for good measure */
@@ -3201,6 +3382,16 @@ bad:
 	return;
 }
 
+void ceph_mdsc_release_dir_caps(struct ceph_mds_request *req)
+{
+	int dcaps;
+
+	dcaps = xchg(&req->r_dir_caps, 0);
+	if (dcaps) {
+		dout("releasing r_dir_caps=%s\n", ceph_cap_string(dcaps));
+		ceph_put_cap_refs(ceph_inode(req->r_parent), dcaps);
+	}
+}
 
 /*
  * called under session->mutex.
@@ -3210,18 +3401,12 @@ static void replay_unsafe_requests(struct ceph_mds_client *mdsc,
 {
 	struct ceph_mds_request *req, *nreq;
 	struct rb_node *p;
-	int err;
 
 	dout("replay_unsafe_requests mds%d\n", session->s_mds);
 
 	mutex_lock(&mdsc->mutex);
-	list_for_each_entry_safe(req, nreq, &session->s_unsafe, r_unsafe_item) {
-		err = __prepare_send_request(mdsc, req, session->s_mds, true);
-		if (!err) {
-			ceph_msg_get(req->r_request);
-			ceph_con_send(&session->s_con, req->r_request);
-		}
-	}
+	list_for_each_entry_safe(req, nreq, &session->s_unsafe, r_unsafe_item)
+		__send_request(mdsc, session, req, true);
 
 	/*
 	 * also re-send old requests when MDS enters reconnect stage. So that MDS
@@ -3235,15 +3420,14 @@ static void replay_unsafe_requests(struct ceph_mds_client *mdsc,
 			continue;
 		if (req->r_attempts == 0)
 			continue; /* only old requests */
-		if (req->r_session &&
-		    req->r_session->s_mds == session->s_mds) {
-			err = __prepare_send_request(mdsc, req,
-						     session->s_mds, true);
-			if (!err) {
-				ceph_msg_get(req->r_request);
-				ceph_con_send(&session->s_con, req->r_request);
-			}
-		}
+		if (!req->r_session)
+			continue;
+		if (req->r_session->s_mds != session->s_mds)
+			continue;
+
+		ceph_mdsc_release_dir_caps(req);
+
+		__send_request(mdsc, session, req, true);
 	}
 	mutex_unlock(&mdsc->mutex);
 }
@@ -3327,7 +3511,7 @@ fail_msg:
 /*
  * Encode information about a cap for a reconnect with the MDS.
  */
-static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
+static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap,
 			  void *arg)
 {
 	union {
@@ -3350,6 +3534,15 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
 	cap->mseq = 0;       /* and migrate_seq */
 	cap->cap_gen = cap->session->s_cap_gen;
 
+	/* These are lost when the session goes away */
+	if (S_ISDIR(inode->i_mode)) {
+		if (cap->issued & CEPH_CAP_DIR_CREATE) {
+			ceph_put_string(rcu_dereference_raw(ci->i_cached_layout.pool_ns));
+			memset(&ci->i_cached_layout, 0, sizeof(ci->i_cached_layout));
+		}
+		cap->issued &= ~CEPH_CAP_ANY_DIR_OPS;
+	}
+
 	if (recon_state->msg_version >= 2) {
 		rec.v2.cap_id = cpu_to_le64(cap->cap_id);
 		rec.v2.wanted = cpu_to_le32(__ceph_caps_wanted(ci));
@@ -3591,6 +3784,8 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
 	if (!reply)
 		goto fail_nomsg;
 
+	xa_destroy(&session->s_delegated_inos);
+
 	mutex_lock(&session->s_mutex);
 	session->s_state = CEPH_MDS_SESSION_RECONNECTING;
 	session->s_seq = 0;
@@ -3646,7 +3841,7 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
 		recon_state.msg_version = 2;
 	}
 	/* trsaverse this session's caps */
-	err = ceph_iterate_session_caps(session, encode_caps_cb, &recon_state);
+	err = ceph_iterate_session_caps(session, reconnect_caps_cb, &recon_state);
 
 	spin_lock(&session->s_cap_lock);
 	session->s_cap_reconnect = 0;
@@ -3754,7 +3949,7 @@ static void check_new_map(struct ceph_mds_client *mdsc,
 	dout("check_new_map new %u old %u\n",
 	     newmap->m_epoch, oldmap->m_epoch);
 
-	for (i = 0; i < oldmap->m_num_mds && i < mdsc->max_sessions; i++) {
+	for (i = 0; i < oldmap->possible_max_rank && i < mdsc->max_sessions; i++) {
 		if (!mdsc->sessions[i])
 			continue;
 		s = mdsc->sessions[i];
@@ -3768,9 +3963,9 @@ static void check_new_map(struct ceph_mds_client *mdsc,
 		     ceph_mdsmap_is_laggy(newmap, i) ? " (laggy)" : "",
 		     ceph_session_state_name(s->s_state));
 
-		if (i >= newmap->m_num_mds) {
+		if (i >= newmap->possible_max_rank) {
 			/* force close session for stopped mds */
-			get_session(s);
+			ceph_get_mds_session(s);
 			__unregister_session(mdsc, s);
 			__wake_requests(mdsc, &s->s_waiting);
 			mutex_unlock(&mdsc->mutex);
@@ -3825,7 +4020,7 @@ static void check_new_map(struct ceph_mds_client *mdsc,
 		}
 	}
 
-	for (i = 0; i < newmap->m_num_mds && i < mdsc->max_sessions; i++) {
+	for (i = 0; i < newmap->possible_max_rank && i < mdsc->max_sessions; i++) {
 		s = mdsc->sessions[i];
 		if (!s)
 			continue;
@@ -4163,6 +4358,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc)
 	INIT_DELAYED_WORK(&mdsc->delayed_work, delayed_work);
 	mdsc->last_renew_caps = jiffies;
 	INIT_LIST_HEAD(&mdsc->cap_delay_list);
+	INIT_LIST_HEAD(&mdsc->cap_wait_list);
 	spin_lock_init(&mdsc->cap_delay_lock);
 	INIT_LIST_HEAD(&mdsc->snap_flush_list);
 	spin_lock_init(&mdsc->snap_flush_lock);
@@ -4370,7 +4566,7 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
 	mutex_lock(&mdsc->mutex);
 	for (i = 0; i < mdsc->max_sessions; i++) {
 		if (mdsc->sessions[i]) {
-			session = get_session(mdsc->sessions[i]);
+			session = ceph_get_mds_session(mdsc->sessions[i]);
 			__unregister_session(mdsc, session);
 			mutex_unlock(&mdsc->mutex);
 			mutex_lock(&session->s_mutex);
@@ -4598,11 +4794,8 @@ static struct ceph_connection *con_get(struct ceph_connection *con)
 {
 	struct ceph_mds_session *s = con->private;
 
-	if (get_session(s)) {
-		dout("mdsc con_get %p ok (%d)\n", s, refcount_read(&s->s_ref));
+	if (ceph_get_mds_session(s))
 		return con;
-	}
-	dout("mdsc con_get %p FAIL\n", s);
 	return NULL;
 }
 
@@ -4610,7 +4803,6 @@ static void con_put(struct ceph_connection *con)
 {
 	struct ceph_mds_session *s = con->private;
 
-	dout("mdsc con_put %p (%d)\n", s, refcount_read(&s->s_ref) - 1);
 	ceph_put_mds_session(s);
 }
 
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index 5cd131b41d84..903d9edfd4bf 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -17,22 +17,33 @@
 #include <linux/ceph/auth.h>
 
 /* The first 8 bits are reserved for old ceph releases */
-#define CEPHFS_FEATURE_MIMIC		8
-#define CEPHFS_FEATURE_REPLY_ENCODING	9
-#define CEPHFS_FEATURE_RECLAIM_CLIENT	10
-#define CEPHFS_FEATURE_LAZY_CAP_WANTED	11
-#define CEPHFS_FEATURE_MULTI_RECONNECT  12
+enum ceph_feature_type {
+	CEPHFS_FEATURE_MIMIC = 8,
+	CEPHFS_FEATURE_REPLY_ENCODING,
+	CEPHFS_FEATURE_RECLAIM_CLIENT,
+	CEPHFS_FEATURE_LAZY_CAP_WANTED,
+	CEPHFS_FEATURE_MULTI_RECONNECT,
+	CEPHFS_FEATURE_DELEG_INO,
+
+	CEPHFS_FEATURE_MAX = CEPHFS_FEATURE_DELEG_INO,
+};
 
-#define CEPHFS_FEATURES_CLIENT_SUPPORTED { 	\
+/*
+ * This will always have the highest feature bit value
+ * as the last element of the array.
+ */
+#define CEPHFS_FEATURES_CLIENT_SUPPORTED {	\
 	0, 1, 2, 3, 4, 5, 6, 7,			\
 	CEPHFS_FEATURE_MIMIC,			\
 	CEPHFS_FEATURE_REPLY_ENCODING,		\
 	CEPHFS_FEATURE_LAZY_CAP_WANTED,		\
 	CEPHFS_FEATURE_MULTI_RECONNECT,		\
+	CEPHFS_FEATURE_DELEG_INO,		\
+						\
+	CEPHFS_FEATURE_MAX,			\
 }
 #define CEPHFS_FEATURES_CLIENT_REQUIRED {}
 
-
 /*
  * Some lock dependencies:
  *
@@ -151,7 +162,8 @@ enum {
 	CEPH_MDS_SESSION_RESTARTING = 5,
 	CEPH_MDS_SESSION_RECONNECTING = 6,
 	CEPH_MDS_SESSION_CLOSING = 7,
-	CEPH_MDS_SESSION_REJECTED = 8,
+	CEPH_MDS_SESSION_CLOSED = 8,
+	CEPH_MDS_SESSION_REJECTED = 9,
 };
 
 struct ceph_mds_session {
@@ -174,6 +186,7 @@ struct ceph_mds_session {
 
 	/* protected by s_cap_lock */
 	spinlock_t        s_cap_lock;
+	refcount_t        s_ref;
 	struct list_head  s_caps;     /* all caps issued by this session */
 	struct ceph_cap  *s_cap_iterator;
 	int               s_nr_caps;
@@ -188,9 +201,9 @@ struct ceph_mds_session {
 	unsigned long     s_renew_requested; /* last time we sent a renew req */
 	u64               s_renew_seq;
 
-	refcount_t        s_ref;
 	struct list_head  s_waiting;  /* waiting requests */
 	struct list_head  s_unsafe;   /* unsafe requests */
+	struct xarray	  s_delegated_inos;
 };
 
 /*
@@ -224,6 +237,7 @@ struct ceph_mds_request {
 	struct rb_node r_node;
 	struct ceph_mds_client *r_mdsc;
 
+	struct kref       r_kref;
 	int r_op;                    /* mds op code */
 
 	/* operation on what? */
@@ -244,6 +258,7 @@ struct ceph_mds_request {
 #define CEPH_MDS_R_GOT_RESULT		(5) /* got a result */
 #define CEPH_MDS_R_DID_PREPOPULATE	(6) /* prepopulated readdir */
 #define CEPH_MDS_R_PARENT_LOCKED	(7) /* is r_parent->i_rwsem wlocked? */
+#define CEPH_MDS_R_ASYNC		(8) /* async request */
 	unsigned long	r_req_flags;
 
 	struct mutex r_fill_mutex;
@@ -252,6 +267,7 @@ struct ceph_mds_request {
 	int r_fmode;        /* file mode, if expecting cap */
 	kuid_t r_uid;
 	kgid_t r_gid;
+	int r_request_release_offset;
 	struct timespec64 r_stamp;
 
 	/* for choosing which mds to send this request to */
@@ -269,12 +285,16 @@ struct ceph_mds_request {
 	int r_old_inode_drop, r_old_inode_unless;
 
 	struct ceph_msg  *r_request;  /* original request */
-	int r_request_release_offset;
 	struct ceph_msg  *r_reply;
 	struct ceph_mds_reply_info_parsed r_reply_info;
-	struct page *r_locked_page;
 	int r_err;
 
+
+	struct page *r_locked_page;
+	int r_dir_caps;
+	int r_num_caps;
+	u32               r_readdir_offset;
+
 	unsigned long r_timeout;  /* optional.  jiffies, 0 is "wait forever" */
 	unsigned long r_started;  /* start time to measure timeout against */
 	unsigned long r_request_started; /* start time for mds request only,
@@ -293,8 +313,8 @@ struct ceph_mds_request {
 	int               r_num_fwd;    /* number of forward attempts */
 	int               r_resend_mds; /* mds to resend to next, if any*/
 	u32               r_sent_on_mseq; /* cap mseq request was sent at*/
+	u64		  r_deleg_ino;
 
-	struct kref       r_kref;
 	struct list_head  r_wait;
 	struct completion r_completion;
 	struct completion r_safe_completion;
@@ -305,10 +325,8 @@ struct ceph_mds_request {
 	long long	  r_dir_release_cnt;
 	long long	  r_dir_ordered_cnt;
 	int		  r_readdir_cache_idx;
-	u32               r_readdir_offset;
 
 	struct ceph_cap_reservation r_caps_reservation;
-	int r_num_caps;
 };
 
 struct ceph_pool_perm {
@@ -340,6 +358,14 @@ struct ceph_quotarealm_inode {
 	struct inode *inode;
 };
 
+struct cap_wait {
+	struct list_head	list;
+	unsigned long		ino;
+	pid_t			tgid;
+	int			need;
+	int			want;
+};
+
 /*
  * mds client state
  */
@@ -416,6 +442,7 @@ struct ceph_mds_client {
 	spinlock_t	caps_list_lock;
 	struct		list_head caps_list; /* unused (reserved or
 						unreserved) */
+	struct		list_head cap_wait_list;
 	int		caps_total_count;    /* total caps allocated */
 	int		caps_use_count;      /* in use */
 	int		caps_use_max;	     /* max used caps */
@@ -442,15 +469,10 @@ extern const char *ceph_mds_op_name(int op);
 extern struct ceph_mds_session *
 __ceph_lookup_mds_session(struct ceph_mds_client *, int mds);
 
-static inline struct ceph_mds_session *
-ceph_get_mds_session(struct ceph_mds_session *s)
-{
-	refcount_inc(&s->s_ref);
-	return s;
-}
-
 extern const char *ceph_session_state_name(int s);
 
+extern struct ceph_mds_session *
+ceph_get_mds_session(struct ceph_mds_session *s);
 extern void ceph_put_mds_session(struct ceph_mds_session *s);
 
 extern int ceph_send_msg_mds(struct ceph_mds_client *mdsc,
@@ -474,6 +496,7 @@ extern int ceph_mdsc_submit_request(struct ceph_mds_client *mdsc,
 extern int ceph_mdsc_do_request(struct ceph_mds_client *mdsc,
 				struct inode *dir,
 				struct ceph_mds_request *req);
+extern void ceph_mdsc_release_dir_caps(struct ceph_mds_request *req);
 static inline void ceph_mdsc_get_request(struct ceph_mds_request *req)
 {
 	kref_get(&req->r_kref);
@@ -498,7 +521,7 @@ extern void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc);
 
 static inline void ceph_mdsc_free_path(char *path, int len)
 {
-	if (path)
+	if (!IS_ERR_OR_NULL(path))
 		__putname(path - (PATH_MAX - 1 - len));
 }
 
@@ -523,4 +546,15 @@ extern void ceph_mdsc_open_export_target_sessions(struct ceph_mds_client *mdsc,
 extern int ceph_trim_caps(struct ceph_mds_client *mdsc,
 			  struct ceph_mds_session *session,
 			  int max_caps);
+
+static inline int ceph_wait_on_async_create(struct inode *inode)
+{
+	struct ceph_inode_info *ci = ceph_inode(inode);
+
+	return wait_on_bit(&ci->i_ceph_flags, CEPH_ASYNC_CREATE_BIT,
+			   TASK_INTERRUPTIBLE);
+}
+
+extern u64 ceph_get_deleg_ino(struct ceph_mds_session *session);
+extern int ceph_restore_deleg_ino(struct ceph_mds_session *session, u64 ino);
 #endif
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index ce2d00da5096..889627817e52 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -13,35 +13,47 @@
 
 #include "super.h"
 
+#define CEPH_MDS_IS_READY(i, ignore_laggy) \
+	(m->m_info[i].state > 0 && ignore_laggy ? true : !m->m_info[i].laggy)
 
-/*
- * choose a random mds that is "up" (i.e. has a state > 0), or -1.
- */
-int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m)
+static int __mdsmap_get_random_mds(struct ceph_mdsmap *m, bool ignore_laggy)
 {
 	int n = 0;
-	int i;
-
-	/* special case for one mds */
-	if (1 == m->m_num_mds && m->m_info[0].state > 0)
-		return 0;
+	int i, j;
 
 	/* count */
-	for (i = 0; i < m->m_num_mds; i++)
-		if (m->m_info[i].state > 0)
+	for (i = 0; i < m->possible_max_rank; i++)
+		if (CEPH_MDS_IS_READY(i, ignore_laggy))
 			n++;
 	if (n == 0)
 		return -1;
 
 	/* pick */
 	n = prandom_u32() % n;
-	for (i = 0; n > 0; i++, n--)
-		while (m->m_info[i].state <= 0)
-			i++;
+	for (j = 0, i = 0; i < m->possible_max_rank; i++) {
+		if (CEPH_MDS_IS_READY(i, ignore_laggy))
+			j++;
+		if (j > n)
+			break;
+	}
 
 	return i;
 }
 
+/*
+ * choose a random mds that is "up" (i.e. has a state > 0), or -1.
+ */
+int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m)
+{
+	int mds;
+
+	mds = __mdsmap_get_random_mds(m, false);
+	if (mds == m->possible_max_rank || mds == -1)
+		mds = __mdsmap_get_random_mds(m, true);
+
+	return mds == m->possible_max_rank ? -1 : mds;
+}
+
 #define __decode_and_drop_type(p, end, type, bad)		\
 	do {							\
 		if (*p + sizeof(type) > end)			\
@@ -135,14 +147,29 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 	m->m_session_autoclose = ceph_decode_32(p);
 	m->m_max_file_size = ceph_decode_64(p);
 	m->m_max_mds = ceph_decode_32(p);
-	m->m_num_mds = m->m_max_mds;
 
-	m->m_info = kcalloc(m->m_num_mds, sizeof(*m->m_info), GFP_NOFS);
+	/*
+	 * pick out the active nodes as the m_num_active_mds, the
+	 * m_num_active_mds maybe larger than m_max_mds when decreasing
+	 * the max_mds in cluster side, in other case it should less
+	 * than or equal to m_max_mds.
+	 */
+	m->m_num_active_mds = n = ceph_decode_32(p);
+
+	/*
+	 * the possible max rank, it maybe larger than the m_num_active_mds,
+	 * for example if the mds_max == 2 in the cluster, when the MDS(0)
+	 * was laggy and being replaced by a new MDS, we will temporarily
+	 * receive a new mds map with n_num_mds == 1 and the active MDS(1),
+	 * and the mds rank >= m_num_active_mds.
+	 */
+	m->possible_max_rank = max(m->m_num_active_mds, m->m_max_mds);
+
+	m->m_info = kcalloc(m->possible_max_rank, sizeof(*m->m_info), GFP_NOFS);
 	if (!m->m_info)
 		goto nomem;
 
 	/* pick out active nodes from mds_info (state > 0) */
-	n = ceph_decode_32(p);
 	for (i = 0; i < n; i++) {
 		u64 global_id;
 		u32 namelen;
@@ -155,6 +182,7 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		void *pexport_targets = NULL;
 		struct ceph_timespec laggy_since;
 		struct ceph_mds_info *info;
+		bool laggy;
 
 		ceph_decode_need(p, end, sizeof(u64) + 1, bad);
 		global_id = ceph_decode_64(p);
@@ -187,6 +215,7 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		if (err)
 			goto corrupt;
 		ceph_decode_copy(p, &laggy_since, sizeof(laggy_since));
+		laggy = laggy_since.tv_sec != 0 || laggy_since.tv_nsec != 0;
 		*p += sizeof(u32);
 		ceph_decode_32_safe(p, end, namelen, bad);
 		*p += namelen;
@@ -204,31 +233,28 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 			*p = info_end;
 		}
 
-		dout("mdsmap_decode %d/%d %lld mds%d.%d %s %s\n",
+		dout("mdsmap_decode %d/%d %lld mds%d.%d %s %s%s\n",
 		     i+1, n, global_id, mds, inc,
 		     ceph_pr_addr(&addr),
-		     ceph_mds_state_name(state));
+		     ceph_mds_state_name(state),
+		     laggy ? "(laggy)" : "");
 
-		if (mds < 0 || state <= 0)
+		if (mds < 0 || mds >= m->possible_max_rank) {
+			pr_warn("mdsmap_decode got incorrect mds(%d)\n", mds);
 			continue;
+		}
 
-		if (mds >= m->m_num_mds) {
-			int new_num = max(mds + 1, m->m_num_mds * 2);
-			void *new_m_info = krealloc(m->m_info,
-						new_num * sizeof(*m->m_info),
-						GFP_NOFS | __GFP_ZERO);
-			if (!new_m_info)
-				goto nomem;
-			m->m_info = new_m_info;
-			m->m_num_mds = new_num;
+		if (state <= 0) {
+			pr_warn("mdsmap_decode got incorrect state(%s)\n",
+				ceph_mds_state_name(state));
+			continue;
 		}
 
 		info = &m->m_info[mds];
 		info->global_id = global_id;
 		info->state = state;
 		info->addr = addr;
-		info->laggy = (laggy_since.tv_sec != 0 ||
-			       laggy_since.tv_nsec != 0);
+		info->laggy = laggy;
 		info->num_export_targets = num_export_targets;
 		if (num_export_targets) {
 			info->export_targets = kcalloc(num_export_targets,
@@ -242,14 +268,6 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 			info->export_targets = NULL;
 		}
 	}
-	if (m->m_num_mds > m->m_max_mds) {
-		/* find max up mds */
-		for (i = m->m_num_mds; i >= m->m_max_mds; i--) {
-			if (i == 0 || m->m_info[i-1].state > 0)
-				break;
-		}
-		m->m_num_mds = i;
-	}
 
 	/* pg_pools */
 	ceph_decode_32_safe(p, end, n, bad);
@@ -291,14 +309,14 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 
 		for (i = 0; i < n; i++) {
 			s32 mds = ceph_decode_32(p);
-			if (mds >= 0 && mds < m->m_num_mds) {
+			if (mds >= 0 && mds < m->possible_max_rank) {
 				if (m->m_info[mds].laggy)
 					num_laggy++;
 			}
 		}
 		m->m_num_laggy = num_laggy;
 
-		if (n > m->m_num_mds) {
+		if (n > m->possible_max_rank) {
 			void *new_m_info = krealloc(m->m_info,
 						    n * sizeof(*m->m_info),
 						    GFP_NOFS | __GFP_ZERO);
@@ -306,7 +324,7 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 				goto nomem;
 			m->m_info = new_m_info;
 		}
-		m->m_num_mds = n;
+		m->possible_max_rank = n;
 	}
 
 	/* inc */
@@ -352,6 +370,8 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		m->m_damaged = false;
 	}
 bad_ext:
+	dout("mdsmap_decode m_enabled: %d, m_damaged: %d, m_num_laggy: %d\n",
+	     !!m->m_enabled, !!m->m_damaged, m->m_num_laggy);
 	*p = end;
 	dout("mdsmap_decode success epoch %u\n", m->m_epoch);
 	return m;
@@ -375,7 +395,7 @@ void ceph_mdsmap_destroy(struct ceph_mdsmap *m)
 {
 	int i;
 
-	for (i = 0; i < m->m_num_mds; i++)
+	for (i = 0; i < m->possible_max_rank; i++)
 		kfree(m->m_info[i].export_targets);
 	kfree(m->m_info);
 	kfree(m->m_data_pg_pools);
@@ -389,9 +409,9 @@ bool ceph_mdsmap_is_cluster_available(struct ceph_mdsmap *m)
 		return false;
 	if (m->m_damaged)
 		return false;
-	if (m->m_num_laggy > 0)
+	if (m->m_num_laggy == m->m_num_active_mds)
 		return false;
-	for (i = 0; i < m->m_num_mds; i++) {
+	for (i = 0; i < m->possible_max_rank; i++) {
 		if (m->m_info[i].state == CEPH_MDS_STATE_ACTIVE)
 			nr_active++;
 	}
diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c
index de56dee60540..19507e2fdb57 100644
--- a/fs/ceph/quota.c
+++ b/fs/ceph/quota.c
@@ -159,8 +159,8 @@ static struct inode *lookup_quotarealm_inode(struct ceph_mds_client *mdsc,
 	}
 
 	if (IS_ERR(in)) {
-		pr_warn("Can't lookup inode %llx (err: %ld)\n",
-			realm->ino, PTR_ERR(in));
+		dout("Can't lookup inode %llx (err: %ld)\n",
+		     realm->ino, PTR_ERR(in));
 		qri->timeout = jiffies + msecs_to_jiffies(60 * 1000); /* XXX */
 	} else {
 		qri->timeout = 0;
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index ccfcc66aaf44..923be9399b21 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -1155,5 +1155,6 @@ void ceph_cleanup_snapid_map(struct ceph_mds_client *mdsc)
 			pr_err("snapid map %llx -> %x still in use\n",
 			       sm->snap, sm->dev);
 		}
+		kfree(sm);
 	}
 }
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index b47f43fc2d68..c9784eb1159a 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -9,7 +9,8 @@
 #include <linux/in6.h>
 #include <linux/module.h>
 #include <linux/mount.h>
-#include <linux/parser.h>
+#include <linux/fs_context.h>
+#include <linux/fs_parser.h>
 #include <linux/sched.h>
 #include <linux/seq_file.h>
 #include <linux/slab.h>
@@ -106,7 +107,6 @@ static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf)
 	return 0;
 }
 
-
 static int ceph_sync_fs(struct super_block *sb, int wait)
 {
 	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
@@ -138,280 +138,335 @@ enum {
 	Opt_readdir_max_entries,
 	Opt_readdir_max_bytes,
 	Opt_congestion_kb,
-	Opt_last_int,
 	/* int args above */
 	Opt_snapdirname,
 	Opt_mds_namespace,
-	Opt_fscache_uniq,
 	Opt_recover_session,
-	Opt_last_string,
+	Opt_source,
 	/* string args above */
 	Opt_dirstat,
-	Opt_nodirstat,
 	Opt_rbytes,
-	Opt_norbytes,
 	Opt_asyncreaddir,
-	Opt_noasyncreaddir,
 	Opt_dcache,
-	Opt_nodcache,
 	Opt_ino32,
-	Opt_noino32,
 	Opt_fscache,
-	Opt_nofscache,
 	Opt_poolperm,
-	Opt_nopoolperm,
 	Opt_require_active_mds,
-	Opt_norequire_active_mds,
-#ifdef CONFIG_CEPH_FS_POSIX_ACL
 	Opt_acl,
-#endif
-	Opt_noacl,
 	Opt_quotadf,
-	Opt_noquotadf,
 	Opt_copyfrom,
-	Opt_nocopyfrom,
+	Opt_wsync,
 };
 
-static match_table_t fsopt_tokens = {
-	{Opt_wsize, "wsize=%d"},
-	{Opt_rsize, "rsize=%d"},
-	{Opt_rasize, "rasize=%d"},
-	{Opt_caps_wanted_delay_min, "caps_wanted_delay_min=%d"},
-	{Opt_caps_wanted_delay_max, "caps_wanted_delay_max=%d"},
-	{Opt_caps_max, "caps_max=%d"},
-	{Opt_readdir_max_entries, "readdir_max_entries=%d"},
-	{Opt_readdir_max_bytes, "readdir_max_bytes=%d"},
-	{Opt_congestion_kb, "write_congestion_kb=%d"},
-	/* int args above */
-	{Opt_snapdirname, "snapdirname=%s"},
-	{Opt_mds_namespace, "mds_namespace=%s"},
-	{Opt_recover_session, "recover_session=%s"},
-	{Opt_fscache_uniq, "fsc=%s"},
-	/* string args above */
-	{Opt_dirstat, "dirstat"},
-	{Opt_nodirstat, "nodirstat"},
-	{Opt_rbytes, "rbytes"},
-	{Opt_norbytes, "norbytes"},
-	{Opt_asyncreaddir, "asyncreaddir"},
-	{Opt_noasyncreaddir, "noasyncreaddir"},
-	{Opt_dcache, "dcache"},
-	{Opt_nodcache, "nodcache"},
-	{Opt_ino32, "ino32"},
-	{Opt_noino32, "noino32"},
-	{Opt_fscache, "fsc"},
-	{Opt_nofscache, "nofsc"},
-	{Opt_poolperm, "poolperm"},
-	{Opt_nopoolperm, "nopoolperm"},
-	{Opt_require_active_mds, "require_active_mds"},
-	{Opt_norequire_active_mds, "norequire_active_mds"},
-#ifdef CONFIG_CEPH_FS_POSIX_ACL
-	{Opt_acl, "acl"},
-#endif
-	{Opt_noacl, "noacl"},
-	{Opt_quotadf, "quotadf"},
-	{Opt_noquotadf, "noquotadf"},
-	{Opt_copyfrom, "copyfrom"},
-	{Opt_nocopyfrom, "nocopyfrom"},
-	{-1, NULL}
+enum ceph_recover_session_mode {
+	ceph_recover_session_no,
+	ceph_recover_session_clean
 };
 
-static int parse_fsopt_token(char *c, void *private)
+static const struct constant_table ceph_param_recover[] = {
+	{ "no",		ceph_recover_session_no },
+	{ "clean",	ceph_recover_session_clean },
+	{}
+};
+
+static const struct fs_parameter_spec ceph_mount_parameters[] = {
+	fsparam_flag_no ("acl",				Opt_acl),
+	fsparam_flag_no ("asyncreaddir",		Opt_asyncreaddir),
+	fsparam_s32	("caps_max",			Opt_caps_max),
+	fsparam_u32	("caps_wanted_delay_max",	Opt_caps_wanted_delay_max),
+	fsparam_u32	("caps_wanted_delay_min",	Opt_caps_wanted_delay_min),
+	fsparam_u32	("write_congestion_kb",		Opt_congestion_kb),
+	fsparam_flag_no ("copyfrom",			Opt_copyfrom),
+	fsparam_flag_no ("dcache",			Opt_dcache),
+	fsparam_flag_no ("dirstat",			Opt_dirstat),
+	fsparam_flag_no	("fsc",				Opt_fscache), // fsc|nofsc
+	fsparam_string	("fsc",				Opt_fscache), // fsc=...
+	fsparam_flag_no ("ino32",			Opt_ino32),
+	fsparam_string	("mds_namespace",		Opt_mds_namespace),
+	fsparam_flag_no ("poolperm",			Opt_poolperm),
+	fsparam_flag_no ("quotadf",			Opt_quotadf),
+	fsparam_u32	("rasize",			Opt_rasize),
+	fsparam_flag_no ("rbytes",			Opt_rbytes),
+	fsparam_u32	("readdir_max_bytes",		Opt_readdir_max_bytes),
+	fsparam_u32	("readdir_max_entries",		Opt_readdir_max_entries),
+	fsparam_enum	("recover_session",		Opt_recover_session, ceph_param_recover),
+	fsparam_flag_no ("require_active_mds",		Opt_require_active_mds),
+	fsparam_u32	("rsize",			Opt_rsize),
+	fsparam_string	("snapdirname",			Opt_snapdirname),
+	fsparam_string	("source",			Opt_source),
+	fsparam_u32	("wsize",			Opt_wsize),
+	fsparam_flag_no	("wsync",			Opt_wsync),
+	{}
+};
+
+struct ceph_parse_opts_ctx {
+	struct ceph_options		*copts;
+	struct ceph_mount_options	*opts;
+};
+
+/*
+ * Remove adjacent slashes and then the trailing slash, unless it is
+ * the only remaining character.
+ *
+ * E.g. "//dir1////dir2///" --> "/dir1/dir2", "///" --> "/".
+ */
+static void canonicalize_path(char *path)
 {
-	struct ceph_mount_options *fsopt = private;
-	substring_t argstr[MAX_OPT_ARGS];
-	int token, intval, ret;
+	int i, j = 0;
 
-	token = match_token((char *)c, fsopt_tokens, argstr);
-	if (token < 0)
-		return -EINVAL;
+	for (i = 0; path[i] != '\0'; i++) {
+		if (path[i] != '/' || j < 1 || path[j - 1] != '/')
+			path[j++] = path[i];
+	}
 
-	if (token < Opt_last_int) {
-		ret = match_int(&argstr[0], &intval);
-		if (ret < 0) {
-			pr_err("bad option arg (not int) at '%s'\n", c);
-			return ret;
-		}
-		dout("got int token %d val %d\n", token, intval);
-	} else if (token > Opt_last_int && token < Opt_last_string) {
-		dout("got string token %d val %s\n", token,
-		     argstr[0].from);
+	if (j > 1 && path[j - 1] == '/')
+		j--;
+	path[j] = '\0';
+}
+
+/*
+ * Parse the source parameter.  Distinguish the server list from the path.
+ *
+ * The source will look like:
+ *     <server_spec>[,<server_spec>...]:[<path>]
+ * where
+ *     <server_spec> is <ip>[:<port>]
+ *     <path> is optional, but if present must begin with '/'
+ */
+static int ceph_parse_source(struct fs_parameter *param, struct fs_context *fc)
+{
+	struct ceph_parse_opts_ctx *pctx = fc->fs_private;
+	struct ceph_mount_options *fsopt = pctx->opts;
+	char *dev_name = param->string, *dev_name_end;
+	int ret;
+
+	dout("%s '%s'\n", __func__, dev_name);
+	if (!dev_name || !*dev_name)
+		return invalfc(fc, "Empty source");
+
+	dev_name_end = strchr(dev_name, '/');
+	if (dev_name_end) {
+		/*
+		 * The server_path will include the whole chars from userland
+		 * including the leading '/'.
+		 */
+		kfree(fsopt->server_path);
+		fsopt->server_path = kstrdup(dev_name_end, GFP_KERNEL);
+		if (!fsopt->server_path)
+			return -ENOMEM;
+
+		canonicalize_path(fsopt->server_path);
 	} else {
-		dout("got token %d\n", token);
+		dev_name_end = dev_name + strlen(dev_name);
 	}
 
+	dev_name_end--;		/* back up to ':' separator */
+	if (dev_name_end < dev_name || *dev_name_end != ':')
+		return invalfc(fc, "No path or : separator in source");
+
+	dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
+	if (fsopt->server_path)
+		dout("server path '%s'\n", fsopt->server_path);
+
+	ret = ceph_parse_mon_ips(param->string, dev_name_end - dev_name,
+				 pctx->copts, fc->log.log);
+	if (ret)
+		return ret;
+
+	fc->source = param->string;
+	param->string = NULL;
+	return 0;
+}
+
+static int ceph_parse_mount_param(struct fs_context *fc,
+				  struct fs_parameter *param)
+{
+	struct ceph_parse_opts_ctx *pctx = fc->fs_private;
+	struct ceph_mount_options *fsopt = pctx->opts;
+	struct fs_parse_result result;
+	unsigned int mode;
+	int token, ret;
+
+	ret = ceph_parse_param(param, pctx->copts, fc->log.log);
+	if (ret != -ENOPARAM)
+		return ret;
+
+	token = fs_parse(fc, ceph_mount_parameters, param, &result);
+	dout("%s fs_parse '%s' token %d\n", __func__, param->key, token);
+	if (token < 0)
+		return token;
+
 	switch (token) {
 	case Opt_snapdirname:
 		kfree(fsopt->snapdir_name);
-		fsopt->snapdir_name = kstrndup(argstr[0].from,
-					       argstr[0].to-argstr[0].from,
-					       GFP_KERNEL);
-		if (!fsopt->snapdir_name)
-			return -ENOMEM;
+		fsopt->snapdir_name = param->string;
+		param->string = NULL;
 		break;
 	case Opt_mds_namespace:
 		kfree(fsopt->mds_namespace);
-		fsopt->mds_namespace = kstrndup(argstr[0].from,
-						argstr[0].to-argstr[0].from,
-						GFP_KERNEL);
-		if (!fsopt->mds_namespace)
-			return -ENOMEM;
+		fsopt->mds_namespace = param->string;
+		param->string = NULL;
 		break;
 	case Opt_recover_session:
-		if (!strncmp(argstr[0].from, "no",
-			     argstr[0].to - argstr[0].from)) {
+		mode = result.uint_32;
+		if (mode == ceph_recover_session_no)
 			fsopt->flags &= ~CEPH_MOUNT_OPT_CLEANRECOVER;
-		} else if (!strncmp(argstr[0].from, "clean",
-				    argstr[0].to - argstr[0].from)) {
+		else if (mode == ceph_recover_session_clean)
 			fsopt->flags |= CEPH_MOUNT_OPT_CLEANRECOVER;
-		} else {
-			return -EINVAL;
-		}
+		else
+			BUG();
 		break;
-	case Opt_fscache_uniq:
-#ifdef CONFIG_CEPH_FSCACHE
-		kfree(fsopt->fscache_uniq);
-		fsopt->fscache_uniq = kstrndup(argstr[0].from,
-					       argstr[0].to-argstr[0].from,
-					       GFP_KERNEL);
-		if (!fsopt->fscache_uniq)
-			return -ENOMEM;
-		fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE;
-		break;
-#else
-		pr_err("fscache support is disabled\n");
-		return -EINVAL;
-#endif
+	case Opt_source:
+		if (fc->source)
+			return invalfc(fc, "Multiple sources specified");
+		return ceph_parse_source(param, fc);
 	case Opt_wsize:
-		if (intval < (int)PAGE_SIZE || intval > CEPH_MAX_WRITE_SIZE)
-			return -EINVAL;
-		fsopt->wsize = ALIGN(intval, PAGE_SIZE);
+		if (result.uint_32 < PAGE_SIZE ||
+		    result.uint_32 > CEPH_MAX_WRITE_SIZE)
+			goto out_of_range;
+		fsopt->wsize = ALIGN(result.uint_32, PAGE_SIZE);
 		break;
 	case Opt_rsize:
-		if (intval < (int)PAGE_SIZE || intval > CEPH_MAX_READ_SIZE)
-			return -EINVAL;
-		fsopt->rsize = ALIGN(intval, PAGE_SIZE);
+		if (result.uint_32 < PAGE_SIZE ||
+		    result.uint_32 > CEPH_MAX_READ_SIZE)
+			goto out_of_range;
+		fsopt->rsize = ALIGN(result.uint_32, PAGE_SIZE);
 		break;
 	case Opt_rasize:
-		if (intval < 0)
-			return -EINVAL;
-		fsopt->rasize = ALIGN(intval, PAGE_SIZE);
+		fsopt->rasize = ALIGN(result.uint_32, PAGE_SIZE);
 		break;
 	case Opt_caps_wanted_delay_min:
-		if (intval < 1)
-			return -EINVAL;
-		fsopt->caps_wanted_delay_min = intval;
+		if (result.uint_32 < 1)
+			goto out_of_range;
+		fsopt->caps_wanted_delay_min = result.uint_32;
 		break;
 	case Opt_caps_wanted_delay_max:
-		if (intval < 1)
-			return -EINVAL;
-		fsopt->caps_wanted_delay_max = intval;
+		if (result.uint_32 < 1)
+			goto out_of_range;
+		fsopt->caps_wanted_delay_max = result.uint_32;
 		break;
 	case Opt_caps_max:
-		if (intval < 0)
-			return -EINVAL;
-		fsopt->caps_max = intval;
+		if (result.int_32 < 0)
+			goto out_of_range;
+		fsopt->caps_max = result.int_32;
 		break;
 	case Opt_readdir_max_entries:
-		if (intval < 1)
-			return -EINVAL;
-		fsopt->max_readdir = intval;
+		if (result.uint_32 < 1)
+			goto out_of_range;
+		fsopt->max_readdir = result.uint_32;
 		break;
 	case Opt_readdir_max_bytes:
-		if (intval < (int)PAGE_SIZE && intval != 0)
-			return -EINVAL;
-		fsopt->max_readdir_bytes = intval;
+		if (result.uint_32 < PAGE_SIZE && result.uint_32 != 0)
+			goto out_of_range;
+		fsopt->max_readdir_bytes = result.uint_32;
 		break;
 	case Opt_congestion_kb:
-		if (intval < 1024) /* at least 1M */
-			return -EINVAL;
-		fsopt->congestion_kb = intval;
+		if (result.uint_32 < 1024) /* at least 1M */
+			goto out_of_range;
+		fsopt->congestion_kb = result.uint_32;
 		break;
 	case Opt_dirstat:
-		fsopt->flags |= CEPH_MOUNT_OPT_DIRSTAT;
-		break;
-	case Opt_nodirstat:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_DIRSTAT;
+		if (!result.negated)
+			fsopt->flags |= CEPH_MOUNT_OPT_DIRSTAT;
+		else
+			fsopt->flags &= ~CEPH_MOUNT_OPT_DIRSTAT;
 		break;
 	case Opt_rbytes:
-		fsopt->flags |= CEPH_MOUNT_OPT_RBYTES;
-		break;
-	case Opt_norbytes:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_RBYTES;
+		if (!result.negated)
+			fsopt->flags |= CEPH_MOUNT_OPT_RBYTES;
+		else
+			fsopt->flags &= ~CEPH_MOUNT_OPT_RBYTES;
 		break;
 	case Opt_asyncreaddir:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_NOASYNCREADDIR;
-		break;
-	case Opt_noasyncreaddir:
-		fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR;
+		if (!result.negated)
+			fsopt->flags &= ~CEPH_MOUNT_OPT_NOASYNCREADDIR;
+		else
+			fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR;
 		break;
 	case Opt_dcache:
-		fsopt->flags |= CEPH_MOUNT_OPT_DCACHE;
-		break;
-	case Opt_nodcache:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_DCACHE;
+		if (!result.negated)
+			fsopt->flags |= CEPH_MOUNT_OPT_DCACHE;
+		else
+			fsopt->flags &= ~CEPH_MOUNT_OPT_DCACHE;
 		break;
 	case Opt_ino32:
-		fsopt->flags |= CEPH_MOUNT_OPT_INO32;
-		break;
-	case Opt_noino32:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_INO32;
+		if (!result.negated)
+			fsopt->flags |= CEPH_MOUNT_OPT_INO32;
+		else
+			fsopt->flags &= ~CEPH_MOUNT_OPT_INO32;
 		break;
+
 	case Opt_fscache:
 #ifdef CONFIG_CEPH_FSCACHE
-		fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE;
 		kfree(fsopt->fscache_uniq);
 		fsopt->fscache_uniq = NULL;
+		if (result.negated) {
+			fsopt->flags &= ~CEPH_MOUNT_OPT_FSCACHE;
+		} else {
+			fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE;
+			fsopt->fscache_uniq = param->string;
+			param->string = NULL;
+		}
 		break;
 #else
-		pr_err("fscache support is disabled\n");
-		return -EINVAL;
+		return invalfc(fc, "fscache support is disabled");
 #endif
-	case Opt_nofscache:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_FSCACHE;
-		kfree(fsopt->fscache_uniq);
-		fsopt->fscache_uniq = NULL;
-		break;
 	case Opt_poolperm:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_NOPOOLPERM;
-		break;
-	case Opt_nopoolperm:
-		fsopt->flags |= CEPH_MOUNT_OPT_NOPOOLPERM;
+		if (!result.negated)
+			fsopt->flags &= ~CEPH_MOUNT_OPT_NOPOOLPERM;
+		else
+			fsopt->flags |= CEPH_MOUNT_OPT_NOPOOLPERM;
 		break;
 	case Opt_require_active_mds:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_MOUNTWAIT;
-		break;
-	case Opt_norequire_active_mds:
-		fsopt->flags |= CEPH_MOUNT_OPT_MOUNTWAIT;
+		if (!result.negated)
+			fsopt->flags &= ~CEPH_MOUNT_OPT_MOUNTWAIT;
+		else
+			fsopt->flags |= CEPH_MOUNT_OPT_MOUNTWAIT;
 		break;
 	case Opt_quotadf:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_NOQUOTADF;
-		break;
-	case Opt_noquotadf:
-		fsopt->flags |= CEPH_MOUNT_OPT_NOQUOTADF;
+		if (!result.negated)
+			fsopt->flags &= ~CEPH_MOUNT_OPT_NOQUOTADF;
+		else
+			fsopt->flags |= CEPH_MOUNT_OPT_NOQUOTADF;
 		break;
 	case Opt_copyfrom:
-		fsopt->flags &= ~CEPH_MOUNT_OPT_NOCOPYFROM;
-		break;
-	case Opt_nocopyfrom:
-		fsopt->flags |= CEPH_MOUNT_OPT_NOCOPYFROM;
+		if (!result.negated)
+			fsopt->flags &= ~CEPH_MOUNT_OPT_NOCOPYFROM;
+		else
+			fsopt->flags |= CEPH_MOUNT_OPT_NOCOPYFROM;
 		break;
-#ifdef CONFIG_CEPH_FS_POSIX_ACL
 	case Opt_acl:
-		fsopt->sb_flags |= SB_POSIXACL;
-		break;
+		if (!result.negated) {
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+			fc->sb_flags |= SB_POSIXACL;
+#else
+			return invalfc(fc, "POSIX ACL support is disabled");
 #endif
-	case Opt_noacl:
-		fsopt->sb_flags &= ~SB_POSIXACL;
+		} else {
+			fc->sb_flags &= ~SB_POSIXACL;
+		}
+		break;
+	case Opt_wsync:
+		if (!result.negated)
+			fsopt->flags &= ~CEPH_MOUNT_OPT_ASYNC_DIROPS;
+		else
+			fsopt->flags |= CEPH_MOUNT_OPT_ASYNC_DIROPS;
 		break;
 	default:
-		BUG_ON(token);
+		BUG();
 	}
 	return 0;
+
+out_of_range:
+	return invalfc(fc, "%s out of range", param->key);
 }
 
 static void destroy_mount_options(struct ceph_mount_options *args)
 {
 	dout("destroy_mount_options %p\n", args);
+	if (!args)
+		return;
+
 	kfree(args->snapdir_name);
 	kfree(args->mds_namespace);
 	kfree(args->server_path);
@@ -446,12 +501,15 @@ static int compare_mount_options(struct ceph_mount_options *new_fsopt,
 	ret = strcmp_null(fsopt1->snapdir_name, fsopt2->snapdir_name);
 	if (ret)
 		return ret;
+
 	ret = strcmp_null(fsopt1->mds_namespace, fsopt2->mds_namespace);
 	if (ret)
 		return ret;
+
 	ret = strcmp_null(fsopt1->server_path, fsopt2->server_path);
 	if (ret)
 		return ret;
+
 	ret = strcmp_null(fsopt1->fscache_uniq, fsopt2->fscache_uniq);
 	if (ret)
 		return ret;
@@ -459,91 +517,6 @@ static int compare_mount_options(struct ceph_mount_options *new_fsopt,
 	return ceph_compare_options(new_opt, fsc->client);
 }
 
-static int parse_mount_options(struct ceph_mount_options **pfsopt,
-			       struct ceph_options **popt,
-			       int flags, char *options,
-			       const char *dev_name)
-{
-	struct ceph_mount_options *fsopt;
-	const char *dev_name_end;
-	int err;
-
-	if (!dev_name || !*dev_name)
-		return -EINVAL;
-
-	fsopt = kzalloc(sizeof(*fsopt), GFP_KERNEL);
-	if (!fsopt)
-		return -ENOMEM;
-
-	dout("parse_mount_options %p, dev_name '%s'\n", fsopt, dev_name);
-
-	fsopt->sb_flags = flags;
-	fsopt->flags = CEPH_MOUNT_OPT_DEFAULT;
-
-	fsopt->wsize = CEPH_MAX_WRITE_SIZE;
-	fsopt->rsize = CEPH_MAX_READ_SIZE;
-	fsopt->rasize = CEPH_RASIZE_DEFAULT;
-	fsopt->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
-	if (!fsopt->snapdir_name) {
-		err = -ENOMEM;
-		goto out;
-	}
-
-	fsopt->caps_wanted_delay_min = CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT;
-	fsopt->caps_wanted_delay_max = CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT;
-	fsopt->max_readdir = CEPH_MAX_READDIR_DEFAULT;
-	fsopt->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
-	fsopt->congestion_kb = default_congestion_kb();
-
-	/*
-	 * Distinguish the server list from the path in "dev_name".
-	 * Internally we do not include the leading '/' in the path.
-	 *
-	 * "dev_name" will look like:
-	 *     <server_spec>[,<server_spec>...]:[<path>]
-	 * where
-	 *     <server_spec> is <ip>[:<port>]
-	 *     <path> is optional, but if present must begin with '/'
-	 */
-	dev_name_end = strchr(dev_name, '/');
-	if (dev_name_end) {
-		if (strlen(dev_name_end) > 1) {
-			fsopt->server_path = kstrdup(dev_name_end, GFP_KERNEL);
-			if (!fsopt->server_path) {
-				err = -ENOMEM;
-				goto out;
-			}
-		}
-	} else {
-		dev_name_end = dev_name + strlen(dev_name);
-	}
-	err = -EINVAL;
-	dev_name_end--;		/* back up to ':' separator */
-	if (dev_name_end < dev_name || *dev_name_end != ':') {
-		pr_err("device name is missing path (no : separator in %s)\n",
-				dev_name);
-		goto out;
-	}
-	dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
-	if (fsopt->server_path)
-		dout("server path '%s'\n", fsopt->server_path);
-
-	*popt = ceph_parse_options(options, dev_name, dev_name_end,
-				 parse_fsopt_token, (void *)fsopt);
-	if (IS_ERR(*popt)) {
-		err = PTR_ERR(*popt);
-		goto out;
-	}
-
-	/* success */
-	*pfsopt = fsopt;
-	return 0;
-
-out:
-	destroy_mount_options(fsopt);
-	return err;
-}
-
 /**
  * ceph_show_options - Show mount options in /proc/mounts
  * @m: seq_file to write to
@@ -587,7 +560,7 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root)
 		seq_puts(m, ",noquotadf");
 
 #ifdef CONFIG_CEPH_FS_POSIX_ACL
-	if (fsopt->sb_flags & SB_POSIXACL)
+	if (root->d_sb->s_flags & SB_POSIXACL)
 		seq_puts(m, ",acl");
 	else
 		seq_puts(m, ",noacl");
@@ -602,26 +575,29 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root)
 	if (fsopt->flags & CEPH_MOUNT_OPT_CLEANRECOVER)
 		seq_show_option(m, "recover_session", "clean");
 
+	if (fsopt->flags & CEPH_MOUNT_OPT_ASYNC_DIROPS)
+		seq_puts(m, ",nowsync");
+
 	if (fsopt->wsize != CEPH_MAX_WRITE_SIZE)
-		seq_printf(m, ",wsize=%d", fsopt->wsize);
+		seq_printf(m, ",wsize=%u", fsopt->wsize);
 	if (fsopt->rsize != CEPH_MAX_READ_SIZE)
-		seq_printf(m, ",rsize=%d", fsopt->rsize);
+		seq_printf(m, ",rsize=%u", fsopt->rsize);
 	if (fsopt->rasize != CEPH_RASIZE_DEFAULT)
-		seq_printf(m, ",rasize=%d", fsopt->rasize);
+		seq_printf(m, ",rasize=%u", fsopt->rasize);
 	if (fsopt->congestion_kb != default_congestion_kb())
-		seq_printf(m, ",write_congestion_kb=%d", fsopt->congestion_kb);
+		seq_printf(m, ",write_congestion_kb=%u", fsopt->congestion_kb);
 	if (fsopt->caps_max)
 		seq_printf(m, ",caps_max=%d", fsopt->caps_max);
 	if (fsopt->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT)
-		seq_printf(m, ",caps_wanted_delay_min=%d",
+		seq_printf(m, ",caps_wanted_delay_min=%u",
 			 fsopt->caps_wanted_delay_min);
 	if (fsopt->caps_wanted_delay_max != CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT)
-		seq_printf(m, ",caps_wanted_delay_max=%d",
+		seq_printf(m, ",caps_wanted_delay_max=%u",
 			   fsopt->caps_wanted_delay_max);
 	if (fsopt->max_readdir != CEPH_MAX_READDIR_DEFAULT)
-		seq_printf(m, ",readdir_max_entries=%d", fsopt->max_readdir);
+		seq_printf(m, ",readdir_max_entries=%u", fsopt->max_readdir);
 	if (fsopt->max_readdir_bytes != CEPH_MAX_READDIR_BYTES_DEFAULT)
-		seq_printf(m, ",readdir_max_bytes=%d", fsopt->max_readdir_bytes);
+		seq_printf(m, ",readdir_max_bytes=%u", fsopt->max_readdir_bytes);
 	if (strcmp(fsopt->snapdir_name, CEPH_SNAPDIRNAME_DEFAULT))
 		seq_show_option(m, "snapdirname", fsopt->snapdir_name);
 
@@ -691,6 +667,7 @@ static struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
 	fsc->sb = NULL;
 	fsc->mount_state = CEPH_MOUNT_MOUNTING;
 	fsc->filp_gen = 1;
+	fsc->have_copy_from2 = true;
 
 	atomic_long_set(&fsc->writeback_count, 0);
 
@@ -763,6 +740,7 @@ struct kmem_cache *ceph_cap_flush_cachep;
 struct kmem_cache *ceph_dentry_cachep;
 struct kmem_cache *ceph_file_cachep;
 struct kmem_cache *ceph_dir_file_cachep;
+struct kmem_cache *ceph_mds_request_cachep;
 
 static void ceph_inode_init_once(void *foo)
 {
@@ -803,6 +781,10 @@ static int __init init_caches(void)
 	if (!ceph_dir_file_cachep)
 		goto bad_dir_file;
 
+	ceph_mds_request_cachep = KMEM_CACHE(ceph_mds_request, SLAB_MEM_SPREAD);
+	if (!ceph_mds_request_cachep)
+		goto bad_mds_req;
+
 	error = ceph_fscache_register();
 	if (error)
 		goto bad_fscache;
@@ -810,6 +792,8 @@ static int __init init_caches(void)
 	return 0;
 
 bad_fscache:
+	kmem_cache_destroy(ceph_mds_request_cachep);
+bad_mds_req:
 	kmem_cache_destroy(ceph_dir_file_cachep);
 bad_dir_file:
 	kmem_cache_destroy(ceph_file_cachep);
@@ -838,11 +822,11 @@ static void destroy_caches(void)
 	kmem_cache_destroy(ceph_dentry_cachep);
 	kmem_cache_destroy(ceph_file_cachep);
 	kmem_cache_destroy(ceph_dir_file_cachep);
+	kmem_cache_destroy(ceph_mds_request_cachep);
 
 	ceph_fscache_unregister();
 }
 
-
 /*
  * ceph_umount_begin - initiate forced umount.  Tear down down the
  * mount, skipping steps that may hang while waiting for server(s).
@@ -860,12 +844,6 @@ static void ceph_umount_begin(struct super_block *sb)
 	fsc->filp_gen++; // invalidate open files
 }
 
-static int ceph_remount(struct super_block *sb, int *flags, char *data)
-{
-	sync_filesystem(sb);
-	return 0;
-}
-
 static const struct super_operations ceph_super_ops = {
 	.alloc_inode	= ceph_alloc_inode,
 	.free_inode	= ceph_free_inode,
@@ -874,7 +852,6 @@ static const struct super_operations ceph_super_ops = {
 	.evict_inode	= ceph_evict_inode,
 	.sync_fs        = ceph_sync_fs,
 	.put_super	= ceph_put_super,
-	.remount_fs	= ceph_remount,
 	.show_options   = ceph_show_options,
 	.statfs		= ceph_statfs,
 	.umount_begin   = ceph_umount_begin,
@@ -929,13 +906,11 @@ out:
 	return root;
 }
 
-
-
-
 /*
  * mount: join the ceph cluster, and open root directory.
  */
-static struct dentry *ceph_real_mount(struct ceph_fs_client *fsc)
+static struct dentry *ceph_real_mount(struct ceph_fs_client *fsc,
+				      struct fs_context *fc)
 {
 	int err;
 	unsigned long started = jiffies;  /* note the start time */
@@ -945,25 +920,21 @@ static struct dentry *ceph_real_mount(struct ceph_fs_client *fsc)
 	mutex_lock(&fsc->client->mount_mutex);
 
 	if (!fsc->sb->s_root) {
-		const char *path;
+		const char *path = fsc->mount_options->server_path ?
+				     fsc->mount_options->server_path + 1 : "";
+
 		err = __ceph_open_session(fsc->client, started);
 		if (err < 0)
 			goto out;
 
 		/* setup fscache */
 		if (fsc->mount_options->flags & CEPH_MOUNT_OPT_FSCACHE) {
-			err = ceph_fscache_register_fs(fsc);
+			err = ceph_fscache_register_fs(fsc, fc);
 			if (err < 0)
 				goto out;
 		}
 
-		if (!fsc->mount_options->server_path) {
-			path = "";
-			dout("mount opening path \\t\n");
-		} else {
-			path = fsc->mount_options->server_path + 1;
-			dout("mount opening path %s\n", path);
-		}
+		dout("mount opening path '%s'\n", path);
 
 		ceph_fs_debugfs_init(fsc);
 
@@ -987,18 +958,16 @@ out:
 	return ERR_PTR(err);
 }
 
-static int ceph_set_super(struct super_block *s, void *data)
+static int ceph_set_super(struct super_block *s, struct fs_context *fc)
 {
-	struct ceph_fs_client *fsc = data;
+	struct ceph_fs_client *fsc = s->s_fs_info;
 	int ret;
 
-	dout("set_super %p data %p\n", s, data);
+	dout("set_super %p\n", s);
 
-	s->s_flags = fsc->mount_options->sb_flags;
 	s->s_maxbytes = MAX_LFS_FILESIZE;
 
 	s->s_xattr = ceph_xattr_handlers;
-	s->s_fs_info = fsc;
 	fsc->sb = s;
 	fsc->max_file_size = 1ULL << 40; /* temp value until we get mdsmap */
 
@@ -1010,24 +979,18 @@ static int ceph_set_super(struct super_block *s, void *data)
 	s->s_time_min = 0;
 	s->s_time_max = U32_MAX;
 
-	ret = set_anon_super(s, NULL);  /* what is that second arg for? */
+	ret = set_anon_super_fc(s, fc);
 	if (ret != 0)
-		goto fail;
-
-	return ret;
-
-fail:
-	s->s_fs_info = NULL;
-	fsc->sb = NULL;
+		fsc->sb = NULL;
 	return ret;
 }
 
 /*
  * share superblock if same fs AND options
  */
-static int ceph_compare_super(struct super_block *sb, void *data)
+static int ceph_compare_super(struct super_block *sb, struct fs_context *fc)
 {
-	struct ceph_fs_client *new = data;
+	struct ceph_fs_client *new = fc->s_fs_info;
 	struct ceph_mount_options *fsopt = new->mount_options;
 	struct ceph_options *opt = new->client->options;
 	struct ceph_fs_client *other = ceph_sb_to_client(sb);
@@ -1043,7 +1006,7 @@ static int ceph_compare_super(struct super_block *sb, void *data)
 		dout("fsid doesn't match\n");
 		return 0;
 	}
-	if (fsopt->sb_flags != other->mount_options->sb_flags) {
+	if (fc->sb_flags != (sb->s_flags & ~SB_BORN)) {
 		dout("flags differ\n");
 		return 0;
 	}
@@ -1073,46 +1036,42 @@ static int ceph_setup_bdi(struct super_block *sb, struct ceph_fs_client *fsc)
 	return 0;
 }
 
-static struct dentry *ceph_mount(struct file_system_type *fs_type,
-		       int flags, const char *dev_name, void *data)
+static int ceph_get_tree(struct fs_context *fc)
 {
+	struct ceph_parse_opts_ctx *pctx = fc->fs_private;
 	struct super_block *sb;
 	struct ceph_fs_client *fsc;
 	struct dentry *res;
+	int (*compare_super)(struct super_block *, struct fs_context *) =
+		ceph_compare_super;
 	int err;
-	int (*compare_super)(struct super_block *, void *) = ceph_compare_super;
-	struct ceph_mount_options *fsopt = NULL;
-	struct ceph_options *opt = NULL;
 
-	dout("ceph_mount\n");
+	dout("ceph_get_tree\n");
 
-#ifdef CONFIG_CEPH_FS_POSIX_ACL
-	flags |= SB_POSIXACL;
-#endif
-	err = parse_mount_options(&fsopt, &opt, flags, data, dev_name);
-	if (err < 0) {
-		res = ERR_PTR(err);
-		goto out_final;
-	}
+	if (!fc->source)
+		return invalfc(fc, "No source");
 
 	/* create client (which we may/may not use) */
-	fsc = create_fs_client(fsopt, opt);
+	fsc = create_fs_client(pctx->opts, pctx->copts);
+	pctx->opts = NULL;
+	pctx->copts = NULL;
 	if (IS_ERR(fsc)) {
-		res = ERR_CAST(fsc);
+		err = PTR_ERR(fsc);
 		goto out_final;
 	}
 
 	err = ceph_mdsc_init(fsc);
-	if (err < 0) {
-		res = ERR_PTR(err);
+	if (err < 0)
 		goto out;
-	}
 
 	if (ceph_test_opt(fsc->client, NOSHARE))
 		compare_super = NULL;
-	sb = sget(fs_type, compare_super, ceph_set_super, flags, fsc);
+
+	fc->s_fs_info = fsc;
+	sb = sget_fc(fc, compare_super, ceph_set_super);
+	fc->s_fs_info = NULL;
 	if (IS_ERR(sb)) {
-		res = ERR_CAST(sb);
+		err = PTR_ERR(sb);
 		goto out;
 	}
 
@@ -1123,20 +1082,26 @@ static struct dentry *ceph_mount(struct file_system_type *fs_type,
 	} else {
 		dout("get_sb using new client %p\n", fsc);
 		err = ceph_setup_bdi(sb, fsc);
-		if (err < 0) {
-			res = ERR_PTR(err);
+		if (err < 0)
 			goto out_splat;
-		}
 	}
 
-	res = ceph_real_mount(fsc);
-	if (IS_ERR(res))
+	res = ceph_real_mount(fsc, fc);
+	if (IS_ERR(res)) {
+		err = PTR_ERR(res);
 		goto out_splat;
+	}
 	dout("root %p inode %p ino %llx.%llx\n", res,
 	     d_inode(res), ceph_vinop(d_inode(res)));
-	return res;
+	fc->root = fsc->sb->s_root;
+	return 0;
 
 out_splat:
+	if (!ceph_mdsmap_is_cluster_available(fsc->mdsc->mdsmap)) {
+		pr_info("No mds server is up or the cluster is laggy\n");
+		err = -EHOSTUNREACH;
+	}
+
 	ceph_mdsc_close_sessions(fsc->mdsc);
 	deactivate_locked_super(sb);
 	goto out_final;
@@ -1144,8 +1109,92 @@ out_splat:
 out:
 	destroy_fs_client(fsc);
 out_final:
-	dout("ceph_mount fail %ld\n", PTR_ERR(res));
-	return res;
+	dout("ceph_get_tree fail %d\n", err);
+	return err;
+}
+
+static void ceph_free_fc(struct fs_context *fc)
+{
+	struct ceph_parse_opts_ctx *pctx = fc->fs_private;
+
+	if (pctx) {
+		destroy_mount_options(pctx->opts);
+		ceph_destroy_options(pctx->copts);
+		kfree(pctx);
+	}
+}
+
+static int ceph_reconfigure_fc(struct fs_context *fc)
+{
+	struct ceph_parse_opts_ctx *pctx = fc->fs_private;
+	struct ceph_mount_options *fsopt = pctx->opts;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(fc->root->d_sb);
+
+	if (fsopt->flags & CEPH_MOUNT_OPT_ASYNC_DIROPS)
+		ceph_set_mount_opt(fsc, ASYNC_DIROPS);
+	else
+		ceph_clear_mount_opt(fsc, ASYNC_DIROPS);
+
+	sync_filesystem(fc->root->d_sb);
+	return 0;
+}
+
+static const struct fs_context_operations ceph_context_ops = {
+	.free		= ceph_free_fc,
+	.parse_param	= ceph_parse_mount_param,
+	.get_tree	= ceph_get_tree,
+	.reconfigure	= ceph_reconfigure_fc,
+};
+
+/*
+ * Set up the filesystem mount context.
+ */
+static int ceph_init_fs_context(struct fs_context *fc)
+{
+	struct ceph_parse_opts_ctx *pctx;
+	struct ceph_mount_options *fsopt;
+
+	pctx = kzalloc(sizeof(*pctx), GFP_KERNEL);
+	if (!pctx)
+		return -ENOMEM;
+
+	pctx->copts = ceph_alloc_options();
+	if (!pctx->copts)
+		goto nomem;
+
+	pctx->opts = kzalloc(sizeof(*pctx->opts), GFP_KERNEL);
+	if (!pctx->opts)
+		goto nomem;
+
+	fsopt = pctx->opts;
+	fsopt->flags = CEPH_MOUNT_OPT_DEFAULT;
+
+	fsopt->wsize = CEPH_MAX_WRITE_SIZE;
+	fsopt->rsize = CEPH_MAX_READ_SIZE;
+	fsopt->rasize = CEPH_RASIZE_DEFAULT;
+	fsopt->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
+	if (!fsopt->snapdir_name)
+		goto nomem;
+
+	fsopt->caps_wanted_delay_min = CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT;
+	fsopt->caps_wanted_delay_max = CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT;
+	fsopt->max_readdir = CEPH_MAX_READDIR_DEFAULT;
+	fsopt->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
+	fsopt->congestion_kb = default_congestion_kb();
+
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+	fc->sb_flags |= SB_POSIXACL;
+#endif
+
+	fc->fs_private = pctx;
+	fc->ops = &ceph_context_ops;
+	return 0;
+
+nomem:
+	destroy_mount_options(pctx->opts);
+	ceph_destroy_options(pctx->copts);
+	kfree(pctx);
+	return -ENOMEM;
 }
 
 static void ceph_kill_sb(struct super_block *s)
@@ -1172,7 +1221,7 @@ static void ceph_kill_sb(struct super_block *s)
 static struct file_system_type ceph_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "ceph",
-	.mount		= ceph_mount,
+	.init_fs_context = ceph_init_fs_context,
 	.kill_sb	= ceph_kill_sb,
 	.fs_flags	= FS_RENAME_DOES_D_MOVE,
 };
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index f98d9247f9cb..60aac3aee055 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -43,13 +43,16 @@
 #define CEPH_MOUNT_OPT_MOUNTWAIT       (1<<12) /* mount waits if no mds is up */
 #define CEPH_MOUNT_OPT_NOQUOTADF       (1<<13) /* no root dir quota in statfs */
 #define CEPH_MOUNT_OPT_NOCOPYFROM      (1<<14) /* don't use RADOS 'copy-from' op */
+#define CEPH_MOUNT_OPT_ASYNC_DIROPS    (1<<15) /* allow async directory ops */
 
 #define CEPH_MOUNT_OPT_DEFAULT			\
 	(CEPH_MOUNT_OPT_DCACHE |		\
 	 CEPH_MOUNT_OPT_NOCOPYFROM)
 
 #define ceph_set_mount_opt(fsc, opt) \
-	(fsc)->mount_options->flags |= CEPH_MOUNT_OPT_##opt;
+	(fsc)->mount_options->flags |= CEPH_MOUNT_OPT_##opt
+#define ceph_clear_mount_opt(fsc, opt) \
+	(fsc)->mount_options->flags &= ~CEPH_MOUNT_OPT_##opt
 #define ceph_test_mount_opt(fsc, opt) \
 	(!!((fsc)->mount_options->flags & CEPH_MOUNT_OPT_##opt))
 
@@ -73,17 +76,16 @@
 #define CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT     60  /* cap release delay */
 
 struct ceph_mount_options {
-	int flags;
-	int sb_flags;
-
-	int wsize;            /* max write size */
-	int rsize;            /* max read size */
-	int rasize;           /* max readahead */
-	int congestion_kb;    /* max writeback in flight */
-	int caps_wanted_delay_min, caps_wanted_delay_max;
+	unsigned int flags;
+
+	unsigned int wsize;            /* max write size */
+	unsigned int rsize;            /* max read size */
+	unsigned int rasize;           /* max readahead */
+	unsigned int congestion_kb;    /* max writeback in flight */
+	unsigned int caps_wanted_delay_min, caps_wanted_delay_max;
 	int caps_max;
-	int max_readdir;       /* max readdir result (entires) */
-	int max_readdir_bytes; /* max readdir result (bytes) */
+	unsigned int max_readdir;       /* max readdir result (entries) */
+	unsigned int max_readdir_bytes; /* max readdir result (bytes) */
 
 	/*
 	 * everything above this point can be memcmp'd; everything below
@@ -92,7 +94,7 @@ struct ceph_mount_options {
 
 	char *snapdir_name;   /* default ".snap" */
 	char *mds_namespace;  /* default NULL */
-	char *server_path;    /* default  "/" */
+	char *server_path;    /* default NULL (means "/") */
 	char *fscache_uniq;   /* default NULL */
 };
 
@@ -107,6 +109,8 @@ struct ceph_fs_client {
 	unsigned long last_auto_reconnect;
 	bool blacklisted;
 
+	bool have_copy_from2;
+
 	u32 filp_gen;
 	loff_t max_file_size;
 
@@ -169,9 +173,9 @@ struct ceph_cap {
 	struct list_head caps_item;
 };
 
-#define CHECK_CAPS_NODELAY    1  /* do not delay any further */
-#define CHECK_CAPS_AUTHONLY   2  /* only check auth cap */
-#define CHECK_CAPS_FLUSH      4  /* flush any dirty caps */
+#define CHECK_CAPS_AUTHONLY   1  /* only check auth cap */
+#define CHECK_CAPS_FLUSH      2  /* flush any dirty caps */
+#define CHECK_CAPS_NOINVAL    4  /* don't invalidate pagecache */
 
 struct ceph_cap_flush {
 	u64 tid;
@@ -283,6 +287,7 @@ struct ceph_dentry_info {
 #define CEPH_DENTRY_REFERENCED		1
 #define CEPH_DENTRY_LEASE_LIST		2
 #define CEPH_DENTRY_SHRINK_LIST		4
+#define CEPH_DENTRY_PRIMARY_LINK	8
 
 struct ceph_inode_xattrs_info {
 	/*
@@ -314,13 +319,14 @@ struct ceph_inode_info {
 	u64 i_inline_version;
 	u32 i_time_warp_seq;
 
-	unsigned i_ceph_flags;
+	unsigned long i_ceph_flags;
 	atomic64_t i_release_count;
 	atomic64_t i_ordered_count;
 	atomic64_t i_complete_seq[2];
 
 	struct ceph_dir_layout i_dir_layout;
 	struct ceph_file_layout i_layout;
+	struct ceph_file_layout i_cached_layout;	// for async creates
 	char *i_symlink;
 
 	/* for dirs */
@@ -351,7 +357,6 @@ struct ceph_inode_info {
 	struct ceph_cap_flush *i_prealloc_cap_flush;
 	struct list_head i_cap_flush_list;
 	wait_queue_head_t i_cap_wq;      /* threads waiting on a capability */
-	unsigned long i_hold_caps_min; /* jiffies */
 	unsigned long i_hold_caps_max; /* jiffies */
 	struct list_head i_cap_delay_list;  /* for delayed cap release to mds */
 	struct ceph_cap_reservation i_cap_migration_resv;
@@ -360,6 +365,8 @@ struct ceph_inode_info {
 						    dirty|flushing caps */
 	unsigned i_snap_caps;           /* cap bits for snapped files */
 
+	unsigned long i_last_rd;
+	unsigned long i_last_wr;
 	int i_nr_by_mode[CEPH_FILE_MODE_BITS];  /* open file counts */
 
 	struct mutex i_truncate_mutex;
@@ -374,7 +381,7 @@ struct ceph_inode_info {
 
 	/* held references to caps */
 	int i_pin_ref;
-	int i_rd_ref, i_rdcache_ref, i_wr_ref, i_wb_ref;
+	int i_rd_ref, i_rdcache_ref, i_wr_ref, i_wb_ref, i_fx_ref;
 	int i_wrbuffer_ref, i_wrbuffer_ref_head;
 	atomic_t i_filelock_ref;
 	atomic_t i_shared_gen;       /* increment each time we get FILE_SHARED */
@@ -407,22 +414,26 @@ struct ceph_inode_info {
 	struct inode vfs_inode; /* at end */
 };
 
-static inline struct ceph_inode_info *ceph_inode(struct inode *inode)
+static inline struct ceph_inode_info *
+ceph_inode(const struct inode *inode)
 {
 	return container_of(inode, struct ceph_inode_info, vfs_inode);
 }
 
-static inline struct ceph_fs_client *ceph_inode_to_client(struct inode *inode)
+static inline struct ceph_fs_client *
+ceph_inode_to_client(const struct inode *inode)
 {
 	return (struct ceph_fs_client *)inode->i_sb->s_fs_info;
 }
 
-static inline struct ceph_fs_client *ceph_sb_to_client(struct super_block *sb)
+static inline struct ceph_fs_client *
+ceph_sb_to_client(const struct super_block *sb)
 {
 	return (struct ceph_fs_client *)sb->s_fs_info;
 }
 
-static inline struct ceph_vino ceph_vino(struct inode *inode)
+static inline struct ceph_vino
+ceph_vino(const struct inode *inode)
 {
 	return ceph_inode(inode)->i_vino;
 }
@@ -506,18 +517,18 @@ static inline struct inode *ceph_find_inode(struct super_block *sb,
  * Ceph inode.
  */
 #define CEPH_I_DIR_ORDERED	(1 << 0)  /* dentries in dir are ordered */
-#define CEPH_I_NODELAY		(1 << 1)  /* do not delay cap release */
 #define CEPH_I_FLUSH		(1 << 2)  /* do not delay flush of dirty metadata */
 #define CEPH_I_POOL_PERM	(1 << 3)  /* pool rd/wr bits are valid */
 #define CEPH_I_POOL_RD		(1 << 4)  /* can read from pool */
 #define CEPH_I_POOL_WR		(1 << 5)  /* can write to pool */
 #define CEPH_I_SEC_INITED	(1 << 6)  /* security initialized */
-#define CEPH_I_CAP_DROPPED	(1 << 7)  /* caps were forcibly dropped */
-#define CEPH_I_KICK_FLUSH	(1 << 8)  /* kick flushing caps */
-#define CEPH_I_FLUSH_SNAPS	(1 << 9)  /* need flush snapss */
-#define CEPH_I_ERROR_WRITE	(1 << 10) /* have seen write errors */
-#define CEPH_I_ERROR_FILELOCK	(1 << 11) /* have seen file lock errors */
-#define CEPH_I_ODIRECT		(1 << 12) /* inode in direct I/O mode */
+#define CEPH_I_KICK_FLUSH	(1 << 7)  /* kick flushing caps */
+#define CEPH_I_FLUSH_SNAPS	(1 << 8)  /* need flush snapss */
+#define CEPH_I_ERROR_WRITE	(1 << 9) /* have seen write errors */
+#define CEPH_I_ERROR_FILELOCK	(1 << 10) /* have seen file lock errors */
+#define CEPH_I_ODIRECT		(1 << 11) /* inode in direct I/O mode */
+#define CEPH_ASYNC_CREATE_BIT	(12)	  /* async create in flight for this */
+#define CEPH_I_ASYNC_CREATE	(1 << CEPH_ASYNC_CREATE_BIT)
 
 /*
  * Masks of ceph inode work.
@@ -669,18 +680,12 @@ extern int __ceph_caps_revoking_other(struct ceph_inode_info *ci,
 extern int ceph_caps_revoking(struct ceph_inode_info *ci, int mask);
 extern int __ceph_caps_used(struct ceph_inode_info *ci);
 
-extern int __ceph_caps_file_wanted(struct ceph_inode_info *ci);
-
-/*
- * wanted, by virtue of open file modes AND cap refs (buffered/cached data)
- */
-static inline int __ceph_caps_wanted(struct ceph_inode_info *ci)
+static inline bool __ceph_is_file_opened(struct ceph_inode_info *ci)
 {
-	int w = __ceph_caps_file_wanted(ci) | __ceph_caps_used(ci);
-	if (w & CEPH_CAP_FILE_BUFFER)
-		w |= CEPH_CAP_FILE_EXCL;  /* we want EXCL if dirty data */
-	return w;
+	return ci->i_nr_by_mode[0];
 }
+extern int __ceph_caps_file_wanted(struct ceph_inode_info *ci);
+extern int __ceph_caps_wanted(struct ceph_inode_info *ci);
 
 /* what the mds thinks we want */
 extern int __ceph_caps_mds_wanted(struct ceph_inode_info *ci, bool check);
@@ -894,6 +899,9 @@ static inline bool __ceph_have_pending_cap_snap(struct ceph_inode_info *ci)
 }
 
 /* inode.c */
+struct ceph_mds_reply_info_in;
+struct ceph_mds_reply_dirfrag;
+
 extern const struct inode_operations ceph_file_iops;
 
 extern struct inode *ceph_alloc_inode(struct super_block *sb);
@@ -909,6 +917,11 @@ extern void ceph_fill_file_time(struct inode *inode, int issued,
 				u64 time_warp_seq, struct timespec64 *ctime,
 				struct timespec64 *mtime,
 				struct timespec64 *atime);
+extern int ceph_fill_inode(struct inode *inode, struct page *locked_page,
+		    struct ceph_mds_reply_info_in *iinfo,
+		    struct ceph_mds_reply_dirfrag *dirinfo,
+		    struct ceph_mds_session *session, int cap_fmode,
+		    struct ceph_cap_reservation *caps_reservation);
 extern int ceph_fill_trace(struct super_block *sb,
 			   struct ceph_mds_request *req);
 extern int ceph_readdir_prepopulate(struct ceph_mds_request *req,
@@ -1037,7 +1050,7 @@ extern struct ceph_cap *ceph_get_cap(struct ceph_mds_client *mdsc,
 				     struct ceph_cap_reservation *ctx);
 extern void ceph_add_cap(struct inode *inode,
 			 struct ceph_mds_session *session, u64 cap_id,
-			 int fmode, unsigned issued, unsigned wanted,
+			 unsigned issued, unsigned wanted,
 			 unsigned cap, unsigned seq, u64 realmino, int flags,
 			 struct ceph_cap **new_cap);
 extern void __ceph_remove_cap(struct ceph_cap *cap, bool queue_release);
@@ -1053,8 +1066,12 @@ extern void ceph_early_kick_flushing_caps(struct ceph_mds_client *mdsc,
 					  struct ceph_mds_session *session);
 extern void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc,
 				    struct ceph_mds_session *session);
+void ceph_kick_flushing_inode_caps(struct ceph_mds_session *session,
+				   struct ceph_inode_info *ci);
 extern struct ceph_cap *ceph_get_cap_for_mds(struct ceph_inode_info *ci,
 					     int mds);
+extern void ceph_take_cap_refs(struct ceph_inode_info *ci, int caps,
+				bool snap_rwsem_locked);
 extern void ceph_get_cap_refs(struct ceph_inode_info *ci, int caps);
 extern void ceph_put_cap_refs(struct ceph_inode_info *ci, int had);
 extern void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
@@ -1079,8 +1096,10 @@ extern int ceph_try_get_caps(struct inode *inode,
 			     int need, int want, bool nonblock, int *got);
 
 /* for counting open files by mode */
-extern void __ceph_get_fmode(struct ceph_inode_info *ci, int mode);
-extern void ceph_put_fmode(struct ceph_inode_info *ci, int mode);
+extern void ceph_get_fmode(struct ceph_inode_info *ci, int mode, int count);
+extern void ceph_put_fmode(struct ceph_inode_info *ci, int mode, int count);
+extern void __ceph_touch_fmode(struct ceph_inode_info *ci,
+			       struct ceph_mds_client *mdsc, int fmode);
 
 /* addr.c */
 extern const struct address_space_operations ceph_aops;
@@ -1092,7 +1111,7 @@ extern void ceph_pool_perm_destroy(struct ceph_mds_client* mdsc);
 /* file.c */
 extern const struct file_operations ceph_file_fops;
 
-extern int ceph_renew_caps(struct inode *inode);
+extern int ceph_renew_caps(struct inode *inode, int fmode);
 extern int ceph_open(struct inode *inode, struct file *file);
 extern int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
 			    struct file *file, unsigned flags, umode_t mode);
diff --git a/fs/ceph/util.c b/fs/ceph/util.c
new file mode 100644
index 000000000000..2c34875675bf
--- /dev/null
+++ b/fs/ceph/util.c
@@ -0,0 +1,100 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Some non-inline ceph helpers
+ */
+#include <linux/module.h>
+#include <linux/ceph/types.h>
+
+/*
+ * return true if @layout appears to be valid
+ */
+int ceph_file_layout_is_valid(const struct ceph_file_layout *layout)
+{
+	__u32 su = layout->stripe_unit;
+	__u32 sc = layout->stripe_count;
+	__u32 os = layout->object_size;
+
+	/* stripe unit, object size must be non-zero, 64k increment */
+	if (!su || (su & (CEPH_MIN_STRIPE_UNIT-1)))
+		return 0;
+	if (!os || (os & (CEPH_MIN_STRIPE_UNIT-1)))
+		return 0;
+	/* object size must be a multiple of stripe unit */
+	if (os < su || os % su)
+		return 0;
+	/* stripe count must be non-zero */
+	if (!sc)
+		return 0;
+	return 1;
+}
+
+void ceph_file_layout_from_legacy(struct ceph_file_layout *fl,
+				  struct ceph_file_layout_legacy *legacy)
+{
+	fl->stripe_unit = le32_to_cpu(legacy->fl_stripe_unit);
+	fl->stripe_count = le32_to_cpu(legacy->fl_stripe_count);
+	fl->object_size = le32_to_cpu(legacy->fl_object_size);
+	fl->pool_id = le32_to_cpu(legacy->fl_pg_pool);
+	if (fl->pool_id == 0 && fl->stripe_unit == 0 &&
+	    fl->stripe_count == 0 && fl->object_size == 0)
+		fl->pool_id = -1;
+}
+
+void ceph_file_layout_to_legacy(struct ceph_file_layout *fl,
+				struct ceph_file_layout_legacy *legacy)
+{
+	legacy->fl_stripe_unit = cpu_to_le32(fl->stripe_unit);
+	legacy->fl_stripe_count = cpu_to_le32(fl->stripe_count);
+	legacy->fl_object_size = cpu_to_le32(fl->object_size);
+	if (fl->pool_id >= 0)
+		legacy->fl_pg_pool = cpu_to_le32(fl->pool_id);
+	else
+		legacy->fl_pg_pool = 0;
+}
+
+int ceph_flags_to_mode(int flags)
+{
+	int mode;
+
+#ifdef O_DIRECTORY  /* fixme */
+	if ((flags & O_DIRECTORY) == O_DIRECTORY)
+		return CEPH_FILE_MODE_PIN;
+#endif
+
+	switch (flags & O_ACCMODE) {
+	case O_WRONLY:
+		mode = CEPH_FILE_MODE_WR;
+		break;
+	case O_RDONLY:
+		mode = CEPH_FILE_MODE_RD;
+		break;
+	case O_RDWR:
+	case O_ACCMODE: /* this is what the VFS does */
+		mode = CEPH_FILE_MODE_RDWR;
+		break;
+	}
+#ifdef O_LAZY
+	if (flags & O_LAZY)
+		mode |= CEPH_FILE_MODE_LAZY;
+#endif
+
+	return mode;
+}
+
+int ceph_caps_for_mode(int mode)
+{
+	int caps = CEPH_CAP_PIN;
+
+	if (mode & CEPH_FILE_MODE_RD)
+		caps |= CEPH_CAP_FILE_SHARED |
+			CEPH_CAP_FILE_RD | CEPH_CAP_FILE_CACHE;
+	if (mode & CEPH_FILE_MODE_WR)
+		caps |= CEPH_CAP_FILE_EXCL |
+			CEPH_CAP_FILE_WR | CEPH_CAP_FILE_BUFFER |
+			CEPH_CAP_AUTH_SHARED | CEPH_CAP_AUTH_EXCL |
+			CEPH_CAP_XATTR_SHARED | CEPH_CAP_XATTR_EXCL;
+	if (mode & CEPH_FILE_MODE_LAZY)
+		caps |= CEPH_CAP_FILE_LAZYIO;
+
+	return caps;
+}
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index cb18ee637cb7..7b8a070a782d 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -655,7 +655,7 @@ static int __build_xattrs(struct inode *inode)
 	u32 len;
 	const char *name, *val;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	int xattr_version;
+	u64 xattr_version;
 	struct ceph_inode_xattr **xattrs = NULL;
 	int err = 0;
 	int i;
@@ -851,7 +851,7 @@ ssize_t __ceph_getxattr(struct inode *inode, const char *name, void *value,
 	req_mask = __get_request_mask(inode);
 
 	spin_lock(&ci->i_ceph_lock);
-	dout("getxattr %p ver=%lld index_ver=%lld\n", inode,
+	dout("getxattr %p name '%s' ver=%lld index_ver=%lld\n", inode, name,
 	     ci->i_xattrs.version, ci->i_xattrs.index_version);
 
 	if (ci->i_xattrs.version == 0 ||
@@ -1078,7 +1078,8 @@ retry:
 		}
 	}
 
-	dout("setxattr %p issued %s\n", inode, ceph_cap_string(issued));
+	dout("setxattr %p name '%s' issued %s\n", inode, name,
+	     ceph_cap_string(issued));
 	__build_xattrs(inode);
 
 	required_blob_size = __get_required_blob_size(ci, name_len, val_len);
diff --git a/fs/char_dev.c b/fs/char_dev.c
index 00dfe17871ac..ba0ded7842a7 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -352,7 +352,7 @@ static struct kobject *cdev_get(struct cdev *p)
 
 	if (owner && !try_module_get(owner))
 		return NULL;
-	kobj = kobject_get(&p->kobj);
+	kobj = kobject_get_unless_zero(&p->kobj);
 	if (!kobj)
 		module_put(owner);
 	return kobj;
@@ -483,6 +483,9 @@ int cdev_add(struct cdev *p, dev_t dev, unsigned count)
 	p->dev = dev;
 	p->count = count;
 
+	if (WARN_ON(dev == WHITEOUT_DEV))
+		return -EBUSY;
+
 	error = kobj_map(cdev_map, dev, count, NULL,
 			 exact_match, exact_lock, p);
 	if (error)
diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig
index 22cf04fb32d3..604f65f4b6c5 100644
--- a/fs/cifs/Kconfig
+++ b/fs/cifs/Kconfig
@@ -202,7 +202,7 @@ config CIFS_SMB_DIRECT
 	help
 	  Enables SMB Direct support for SMB 3.0, 3.02 and 3.1.1.
 	  SMB Direct allows transferring SMB packets over RDMA. If unsure,
-	  say N.
+	  say Y.
 
 config CIFS_FSCACHE
 	bool "Provide CIFS client caching support"
diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c
index 0b4eee3bed66..916567d770f5 100644
--- a/fs/cifs/cifs_debug.c
+++ b/fs/cifs/cifs_debug.c
@@ -122,6 +122,27 @@ static void cifs_debug_tcon(struct seq_file *m, struct cifs_tcon *tcon)
 }
 
 static void
+cifs_dump_channel(struct seq_file *m, int i, struct cifs_chan *chan)
+{
+	struct TCP_Server_Info *server = chan->server;
+
+	seq_printf(m, "\t\tChannel %d Number of credits: %d Dialect 0x%x "
+		   "TCP status: %d Instance: %d Local Users To Server: %d "
+		   "SecMode: 0x%x Req On Wire: %d In Send: %d "
+		   "In MaxReq Wait: %d\n",
+		   i+1,
+		   server->credits,
+		   server->dialect,
+		   server->tcpStatus,
+		   server->reconnect_instance,
+		   server->srv_count,
+		   server->sec_mode,
+		   in_flight(server),
+		   atomic_read(&server->in_send),
+		   atomic_read(&server->num_waiters));
+}
+
+static void
 cifs_dump_iface(struct seq_file *m, struct cifs_server_iface *iface)
 {
 	struct sockaddr_in *ipv4 = (struct sockaddr_in *)&iface->sockaddr;
@@ -256,6 +277,11 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 		if (!server->rdma)
 			goto skip_rdma;
 
+		if (!server->smbd_conn) {
+			seq_printf(m, "\nSMBDirect transport not available");
+			goto skip_rdma;
+		}
+
 		seq_printf(m, "\nSMBDirect (in hex) protocol version: %x "
 			"transport status: %x",
 			server->smbd_conn->protocol,
@@ -297,10 +323,8 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 			atomic_read(&server->smbd_conn->send_credits),
 			atomic_read(&server->smbd_conn->receive_credits),
 			server->smbd_conn->receive_credit_target);
-		seq_printf(m, "\nPending send_pending: %x "
-			"send_payload_pending: %x",
-			atomic_read(&server->smbd_conn->send_pending),
-			atomic_read(&server->smbd_conn->send_payload_pending));
+		seq_printf(m, "\nPending send_pending: %x ",
+			atomic_read(&server->smbd_conn->send_pending));
 		seq_printf(m, "\nReceive buffers count_receive_queue: %x "
 			"count_empty_packet_queue: %x",
 			server->smbd_conn->count_receive_queue,
@@ -360,11 +384,10 @@ skip_rdma:
 				   server->srv_count,
 				   server->sec_mode, in_flight(server));
 
-#ifdef CONFIG_CIFS_STATS2
 			seq_printf(m, " In Send: %d In MaxReq Wait: %d",
 				atomic_read(&server->in_send),
 				atomic_read(&server->num_waiters));
-#endif
+
 			/* dump session id helpful for use with network trace */
 			seq_printf(m, " SessionId: 0x%llx", ses->Suid);
 			if (ses->session_flags & SMB2_SESSION_FLAG_ENCRYPT_DATA)
@@ -372,6 +395,13 @@ skip_rdma:
 			if (ses->sign)
 				seq_puts(m, " signed");
 
+			if (ses->chan_count > 1) {
+				seq_printf(m, "\n\n\tExtra Channels: %zu\n",
+					   ses->chan_count-1);
+				for (j = 1; j < ses->chan_count; j++)
+					cifs_dump_channel(m, j, &ses->chans[j]);
+			}
+
 			seq_puts(m, "\n\tShares:");
 			j = 0;
 
@@ -410,8 +440,13 @@ skip_rdma:
 				seq_printf(m, "\n\tServer interfaces: %zu\n",
 					   ses->iface_count);
 			for (j = 0; j < ses->iface_count; j++) {
+				struct cifs_server_iface *iface;
+
+				iface = &ses->iface_list[j];
 				seq_printf(m, "\t%d)", j);
-				cifs_dump_iface(m, &ses->iface_list[j]);
+				cifs_dump_iface(m, iface);
+				if (is_ses_using_iface(ses, iface))
+					seq_puts(m, "\t\t[CONNECTED]\n");
 			}
 			spin_unlock(&ses->iface_lock);
 		}
@@ -574,12 +609,12 @@ static int cifs_stats_proc_open(struct inode *inode, struct file *file)
 	return single_open(file, cifs_stats_proc_show, NULL);
 }
 
-static const struct file_operations cifs_stats_proc_fops = {
-	.open		= cifs_stats_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-	.write		= cifs_stats_proc_write,
+static const struct proc_ops cifs_stats_proc_ops = {
+	.proc_open	= cifs_stats_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= cifs_stats_proc_write,
 };
 
 #ifdef CONFIG_CIFS_SMB_DIRECT
@@ -603,12 +638,12 @@ static int name##_open(struct inode *inode, struct file *file) \
 	return single_open(file, name##_proc_show, NULL); \
 } \
 \
-static const struct file_operations cifs_##name##_proc_fops = { \
-	.open		= name##_open, \
-	.read		= seq_read, \
-	.llseek		= seq_lseek, \
-	.release	= single_release, \
-	.write		= name##_write, \
+static const struct proc_ops cifs_##name##_proc_fops = { \
+	.proc_open	= name##_open, \
+	.proc_read	= seq_read, \
+	.proc_lseek	= seq_lseek, \
+	.proc_release	= single_release, \
+	.proc_write	= name##_write, \
 }
 
 PROC_FILE_DEFINE(rdma_readwrite_threshold);
@@ -622,11 +657,11 @@ PROC_FILE_DEFINE(smbd_receive_credit_max);
 #endif
 
 static struct proc_dir_entry *proc_fs_cifs;
-static const struct file_operations cifsFYI_proc_fops;
-static const struct file_operations cifs_lookup_cache_proc_fops;
-static const struct file_operations traceSMB_proc_fops;
-static const struct file_operations cifs_security_flags_proc_fops;
-static const struct file_operations cifs_linux_ext_proc_fops;
+static const struct proc_ops cifsFYI_proc_ops;
+static const struct proc_ops cifs_lookup_cache_proc_ops;
+static const struct proc_ops traceSMB_proc_ops;
+static const struct proc_ops cifs_security_flags_proc_ops;
+static const struct proc_ops cifs_linux_ext_proc_ops;
 
 void
 cifs_proc_init(void)
@@ -641,18 +676,18 @@ cifs_proc_init(void)
 	proc_create_single("open_files", 0400, proc_fs_cifs,
 			cifs_debug_files_proc_show);
 
-	proc_create("Stats", 0644, proc_fs_cifs, &cifs_stats_proc_fops);
-	proc_create("cifsFYI", 0644, proc_fs_cifs, &cifsFYI_proc_fops);
-	proc_create("traceSMB", 0644, proc_fs_cifs, &traceSMB_proc_fops);
+	proc_create("Stats", 0644, proc_fs_cifs, &cifs_stats_proc_ops);
+	proc_create("cifsFYI", 0644, proc_fs_cifs, &cifsFYI_proc_ops);
+	proc_create("traceSMB", 0644, proc_fs_cifs, &traceSMB_proc_ops);
 	proc_create("LinuxExtensionsEnabled", 0644, proc_fs_cifs,
-		    &cifs_linux_ext_proc_fops);
+		    &cifs_linux_ext_proc_ops);
 	proc_create("SecurityFlags", 0644, proc_fs_cifs,
-		    &cifs_security_flags_proc_fops);
+		    &cifs_security_flags_proc_ops);
 	proc_create("LookupCacheEnabled", 0644, proc_fs_cifs,
-		    &cifs_lookup_cache_proc_fops);
+		    &cifs_lookup_cache_proc_ops);
 
 #ifdef CONFIG_CIFS_DFS_UPCALL
-	proc_create("dfscache", 0644, proc_fs_cifs, &dfscache_proc_fops);
+	proc_create("dfscache", 0644, proc_fs_cifs, &dfscache_proc_ops);
 #endif
 
 #ifdef CONFIG_CIFS_SMB_DIRECT
@@ -737,12 +772,12 @@ static ssize_t cifsFYI_proc_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
-static const struct file_operations cifsFYI_proc_fops = {
-	.open		= cifsFYI_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-	.write		= cifsFYI_proc_write,
+static const struct proc_ops cifsFYI_proc_ops = {
+	.proc_open	= cifsFYI_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= cifsFYI_proc_write,
 };
 
 static int cifs_linux_ext_proc_show(struct seq_file *m, void *v)
@@ -768,12 +803,12 @@ static ssize_t cifs_linux_ext_proc_write(struct file *file,
 	return count;
 }
 
-static const struct file_operations cifs_linux_ext_proc_fops = {
-	.open		= cifs_linux_ext_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-	.write		= cifs_linux_ext_proc_write,
+static const struct proc_ops cifs_linux_ext_proc_ops = {
+	.proc_open	= cifs_linux_ext_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= cifs_linux_ext_proc_write,
 };
 
 static int cifs_lookup_cache_proc_show(struct seq_file *m, void *v)
@@ -799,12 +834,12 @@ static ssize_t cifs_lookup_cache_proc_write(struct file *file,
 	return count;
 }
 
-static const struct file_operations cifs_lookup_cache_proc_fops = {
-	.open		= cifs_lookup_cache_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-	.write		= cifs_lookup_cache_proc_write,
+static const struct proc_ops cifs_lookup_cache_proc_ops = {
+	.proc_open	= cifs_lookup_cache_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= cifs_lookup_cache_proc_write,
 };
 
 static int traceSMB_proc_show(struct seq_file *m, void *v)
@@ -830,12 +865,12 @@ static ssize_t traceSMB_proc_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
-static const struct file_operations traceSMB_proc_fops = {
-	.open		= traceSMB_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-	.write		= traceSMB_proc_write,
+static const struct proc_ops traceSMB_proc_ops = {
+	.proc_open	= traceSMB_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= traceSMB_proc_write,
 };
 
 static int cifs_security_flags_proc_show(struct seq_file *m, void *v)
@@ -941,12 +976,12 @@ static ssize_t cifs_security_flags_proc_write(struct file *file,
 	return count;
 }
 
-static const struct file_operations cifs_security_flags_proc_fops = {
-	.open		= cifs_security_flags_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-	.write		= cifs_security_flags_proc_write,
+static const struct proc_ops cifs_security_flags_proc_ops = {
+	.proc_open	= cifs_security_flags_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= cifs_security_flags_proc_write,
 };
 #else
 inline void cifs_proc_init(void)
diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c
index 41957b82d796..cc3ada12848d 100644
--- a/fs/cifs/cifs_dfs_ref.c
+++ b/fs/cifs/cifs_dfs_ref.c
@@ -120,17 +120,17 @@ cifs_build_devname(char *nodename, const char *prepath)
 
 
 /**
- * cifs_compose_mount_options	-	creates mount options for refferral
+ * cifs_compose_mount_options	-	creates mount options for referral
  * @sb_mountdata:	parent/root DFS mount options (template)
  * @fullpath:		full path in UNC format
- * @ref:		server's referral
+ * @ref:		optional server's referral
  * @devname:		optional pointer for saving device name
  *
  * creates mount options for submount based on template options sb_mountdata
  * and replacing unc,ip,prefixpath options with ones we've got form ref_unc.
  *
  * Returns: pointer to new mount options or ERR_PTR.
- * Caller is responcible for freeing retunrned value if it is not error.
+ * Caller is responsible for freeing returned value if it is not error.
  */
 char *cifs_compose_mount_options(const char *sb_mountdata,
 				   const char *fullpath,
@@ -150,18 +150,27 @@ char *cifs_compose_mount_options(const char *sb_mountdata,
 	if (sb_mountdata == NULL)
 		return ERR_PTR(-EINVAL);
 
-	if (strlen(fullpath) - ref->path_consumed) {
-		prepath = fullpath + ref->path_consumed;
-		/* skip initial delimiter */
-		if (*prepath == '/' || *prepath == '\\')
-			prepath++;
-	}
+	if (ref) {
+		if (strlen(fullpath) - ref->path_consumed) {
+			prepath = fullpath + ref->path_consumed;
+			/* skip initial delimiter */
+			if (*prepath == '/' || *prepath == '\\')
+				prepath++;
+		}
 
-	name = cifs_build_devname(ref->node_name, prepath);
-	if (IS_ERR(name)) {
-		rc = PTR_ERR(name);
-		name = NULL;
-		goto compose_mount_options_err;
+		name = cifs_build_devname(ref->node_name, prepath);
+		if (IS_ERR(name)) {
+			rc = PTR_ERR(name);
+			name = NULL;
+			goto compose_mount_options_err;
+		}
+	} else {
+		name = cifs_build_devname((char *)fullpath, NULL);
+		if (IS_ERR(name)) {
+			rc = PTR_ERR(name);
+			name = NULL;
+			goto compose_mount_options_err;
+		}
 	}
 
 	rc = dns_resolve_server_name_to_ip(name, &srvIP);
@@ -225,6 +234,8 @@ char *cifs_compose_mount_options(const char *sb_mountdata,
 
 	if (devname)
 		*devname = name;
+	else
+		kfree(name);
 
 	/*cifs_dbg(FYI, "%s: parent mountdata: %s\n", __func__, sb_mountdata);*/
 	/*cifs_dbg(FYI, "%s: submount mountdata: %s\n", __func__, mountdata );*/
@@ -241,23 +252,23 @@ compose_mount_options_err:
 }
 
 /**
- * cifs_dfs_do_refmount - mounts specified path using provided refferal
+ * cifs_dfs_do_mount - mounts specified path using DFS full path
+ *
+ * Always pass down @fullpath to smb3_do_mount() so we can use the root server
+ * to perform failover in case we failed to connect to the first target in the
+ * referral.
+ *
  * @cifs_sb:		parent/root superblock
  * @fullpath:		full path in UNC format
- * @ref:		server's referral
  */
-static struct vfsmount *cifs_dfs_do_refmount(struct dentry *mntpt,
-		struct cifs_sb_info *cifs_sb,
-		const char *fullpath, const struct dfs_info3_param *ref)
+static struct vfsmount *cifs_dfs_do_mount(struct dentry *mntpt,
+					  struct cifs_sb_info *cifs_sb,
+					  const char *fullpath)
 {
 	struct vfsmount *mnt;
 	char *mountdata;
 	char *devname;
 
-	/*
-	 * Always pass down the DFS full path to smb3_do_mount() so we
-	 * can use it later for failover.
-	 */
 	devname = kstrndup(fullpath, strlen(fullpath), GFP_KERNEL);
 	if (!devname)
 		return ERR_PTR(-ENOMEM);
@@ -266,7 +277,7 @@ static struct vfsmount *cifs_dfs_do_refmount(struct dentry *mntpt,
 
 	/* strip first '\' from fullpath */
 	mountdata = cifs_compose_mount_options(cifs_sb->mountdata,
-					       fullpath + 1, ref, NULL);
+					       fullpath + 1, NULL, NULL);
 	if (IS_ERR(mountdata)) {
 		kfree(devname);
 		return (struct vfsmount *)mountdata;
@@ -278,28 +289,16 @@ static struct vfsmount *cifs_dfs_do_refmount(struct dentry *mntpt,
 	return mnt;
 }
 
-static void dump_referral(const struct dfs_info3_param *ref)
-{
-	cifs_dbg(FYI, "DFS: ref path: %s\n", ref->path_name);
-	cifs_dbg(FYI, "DFS: node path: %s\n", ref->node_name);
-	cifs_dbg(FYI, "DFS: fl: %d, srv_type: %d\n",
-		 ref->flags, ref->server_type);
-	cifs_dbg(FYI, "DFS: ref_flags: %d, path_consumed: %d\n",
-		 ref->ref_flag, ref->path_consumed);
-}
-
 /*
  * Create a vfsmount that we can automount
  */
 static struct vfsmount *cifs_dfs_do_automount(struct dentry *mntpt)
 {
-	struct dfs_info3_param referral = {0};
 	struct cifs_sb_info *cifs_sb;
 	struct cifs_ses *ses;
 	struct cifs_tcon *tcon;
 	char *full_path, *root_path;
 	unsigned int xid;
-	int len;
 	int rc;
 	struct vfsmount *mnt;
 
@@ -325,6 +324,8 @@ static struct vfsmount *cifs_dfs_do_automount(struct dentry *mntpt)
 	if (full_path == NULL)
 		goto cdda_exit;
 
+	convert_delimiter(full_path, '\\');
+
 	cifs_dbg(FYI, "%s: full_path: %s\n", __func__, full_path);
 
 	if (!cifs_sb_master_tlink(cifs_sb)) {
@@ -357,7 +358,7 @@ static struct vfsmount *cifs_dfs_do_automount(struct dentry *mntpt)
 	if (!rc) {
 		rc = dfs_cache_find(xid, ses, cifs_sb->local_nls,
 				    cifs_remap(cifs_sb), full_path + 1,
-				    &referral, NULL);
+				    NULL, NULL);
 	}
 
 	free_xid(xid);
@@ -366,26 +367,16 @@ static struct vfsmount *cifs_dfs_do_automount(struct dentry *mntpt)
 		mnt = ERR_PTR(rc);
 		goto free_root_path;
 	}
-
-	dump_referral(&referral);
-
-	len = strlen(referral.node_name);
-	if (len < 2) {
-		cifs_dbg(VFS, "%s: Net Address path too short: %s\n",
-			 __func__, referral.node_name);
-		mnt = ERR_PTR(-EINVAL);
-		goto free_dfs_ref;
-	}
 	/*
-	 * cifs_mount() will retry every available node server in case
-	 * of failures.
+	 * OK - we were able to get and cache a referral for @full_path.
+	 *
+	 * Now, pass it down to cifs_mount() and it will retry every available
+	 * node server in case of failures - no need to do it here.
 	 */
-	mnt = cifs_dfs_do_refmount(mntpt, cifs_sb, full_path, &referral);
-	cifs_dbg(FYI, "%s: cifs_dfs_do_refmount:%s , mnt:%p\n", __func__,
-		 referral.node_name, mnt);
+	mnt = cifs_dfs_do_mount(mntpt, cifs_sb, full_path);
+	cifs_dbg(FYI, "%s: cifs_dfs_do_mount:%s , mnt:%p\n", __func__,
+		 full_path + 1, mnt);
 
-free_dfs_ref:
-	free_dfs_info_param(&referral);
 free_root_path:
 	kfree(root_path);
 free_full_path:
diff --git a/fs/cifs/cifs_ioctl.h b/fs/cifs/cifs_ioctl.h
index 0f0dc1c1fe41..153d5c842a9b 100644
--- a/fs/cifs/cifs_ioctl.h
+++ b/fs/cifs/cifs_ioctl.h
@@ -65,6 +65,11 @@ struct smb3_key_debug_info {
 	__u8	smb3decryptionkey[SMB3_SIGN_KEY_SIZE];
 } __packed;
 
+struct smb3_notify {
+	__u32	completion_filter;
+	bool	watch_tree;
+} __packed;
+
 #define CIFS_IOCTL_MAGIC	0xCF
 #define CIFS_IOC_COPYCHUNK_FILE	_IOW(CIFS_IOCTL_MAGIC, 3, int)
 #define CIFS_IOC_SET_INTEGRITY  _IO(CIFS_IOCTL_MAGIC, 4)
@@ -72,3 +77,4 @@ struct smb3_key_debug_info {
 #define CIFS_ENUMERATE_SNAPSHOTS _IOR(CIFS_IOCTL_MAGIC, 6, struct smb_snapshot_array)
 #define CIFS_QUERY_INFO _IOWR(CIFS_IOCTL_MAGIC, 7, struct smb_query_info)
 #define CIFS_DUMP_KEY _IOWR(CIFS_IOCTL_MAGIC, 8, struct smb3_key_debug_info)
+#define CIFS_IOC_NOTIFY _IOW(CIFS_IOCTL_MAGIC, 9, struct smb3_notify)
diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c
index 7f01c6e60791..7b9b876b513b 100644
--- a/fs/cifs/cifs_spnego.c
+++ b/fs/cifs/cifs_spnego.c
@@ -98,7 +98,7 @@ struct key_type cifs_spnego_key_type = {
 struct key *
 cifs_get_spnego_key(struct cifs_ses *sesInfo)
 {
-	struct TCP_Server_Info *server = sesInfo->server;
+	struct TCP_Server_Info *server = cifs_ses_server(sesInfo);
 	struct sockaddr_in *sa = (struct sockaddr_in *) &server->dstaddr;
 	struct sockaddr_in6 *sa6 = (struct sockaddr_in6 *) &server->dstaddr;
 	char *description, *dp;
diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c
index f842944a5c76..ae421634aa42 100644
--- a/fs/cifs/cifsacl.c
+++ b/fs/cifs/cifsacl.c
@@ -39,8 +39,6 @@ static const struct cifs_sid sid_everyone = {
 /* security id for Authenticated Users system group */
 static const struct cifs_sid sid_authusers = {
 	1, 1, {0, 0, 0, 0, 0, 5}, {cpu_to_le32(11)} };
-/* group users */
-static const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {} };
 
 /* S-1-22-1 Unmapped Unix users */
 static const struct cifs_sid sid_unix_users = {1, 1, {0, 0, 0, 0, 0, 22},
@@ -344,7 +342,7 @@ static int
 sid_to_id(struct cifs_sb_info *cifs_sb, struct cifs_sid *psid,
 		struct cifs_fattr *fattr, uint sidtype)
 {
-	int rc;
+	int rc = 0;
 	struct key *sidkey;
 	char *sidstr;
 	const struct cred *saved_cred;
@@ -452,11 +450,12 @@ out_revert_creds:
 	 * fails then we just fall back to using the mnt_uid/mnt_gid.
 	 */
 got_valid_id:
+	rc = 0;
 	if (sidtype == SIDOWNER)
 		fattr->cf_uid = fuid;
 	else
 		fattr->cf_gid = fgid;
-	return 0;
+	return rc;
 }
 
 int
@@ -603,7 +602,7 @@ static void access_flags_to_mode(__le32 ace_flags, int type, umode_t *pmode,
 			((flags & FILE_EXEC_RIGHTS) == FILE_EXEC_RIGHTS))
 		*pmode |= (S_IXUGO & (*pbits_to_set));
 
-	cifs_dbg(NOISY, "access flags 0x%x mode now 0x%x\n", flags, *pmode);
+	cifs_dbg(NOISY, "access flags 0x%x mode now %04o\n", flags, *pmode);
 	return;
 }
 
@@ -632,7 +631,7 @@ static void mode_to_access_flags(umode_t mode, umode_t bits_to_use,
 	if (mode & S_IXUGO)
 		*pace_flags |= SET_FILE_EXEC_RIGHTS;
 
-	cifs_dbg(NOISY, "mode: 0x%x, access flags now 0x%x\n",
+	cifs_dbg(NOISY, "mode: %04o, access flags now 0x%x\n",
 		 mode, *pace_flags);
 	return;
 }
@@ -804,6 +803,51 @@ static void parse_dacl(struct cifs_acl *pdacl, char *end_of_acl,
 	return;
 }
 
+unsigned int setup_authusers_ACE(struct cifs_ace *pntace)
+{
+	int i;
+	unsigned int ace_size = 20;
+
+	pntace->type = ACCESS_ALLOWED_ACE_TYPE;
+	pntace->flags = 0x0;
+	pntace->access_req = cpu_to_le32(GENERIC_ALL);
+	pntace->sid.num_subauth = 1;
+	pntace->sid.revision = 1;
+	for (i = 0; i < NUM_AUTHS; i++)
+		pntace->sid.authority[i] =  sid_authusers.authority[i];
+
+	pntace->sid.sub_auth[0] =  sid_authusers.sub_auth[0];
+
+	/* size = 1 + 1 + 2 + 4 + 1 + 1 + 6 + (psid->num_subauth*4) */
+	pntace->size = cpu_to_le16(ace_size);
+	return ace_size;
+}
+
+/*
+ * Fill in the special SID based on the mode. See
+ * http://technet.microsoft.com/en-us/library/hh509017(v=ws.10).aspx
+ */
+unsigned int setup_special_mode_ACE(struct cifs_ace *pntace, __u64 nmode)
+{
+	int i;
+	unsigned int ace_size = 28;
+
+	pntace->type = ACCESS_DENIED_ACE_TYPE;
+	pntace->flags = 0x0;
+	pntace->access_req = 0;
+	pntace->sid.num_subauth = 3;
+	pntace->sid.revision = 1;
+	for (i = 0; i < NUM_AUTHS; i++)
+		pntace->sid.authority[i] = sid_unix_NFS_mode.authority[i];
+
+	pntace->sid.sub_auth[0] = sid_unix_NFS_mode.sub_auth[0];
+	pntace->sid.sub_auth[1] = sid_unix_NFS_mode.sub_auth[1];
+	pntace->sid.sub_auth[2] = cpu_to_le32(nmode & 07777);
+
+	/* size = 1 + 1 + 2 + 4 + 1 + 1 + 6 + (psid->num_subauth*4) */
+	pntace->size = cpu_to_le16(ace_size);
+	return ace_size;
+}
 
 static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid,
 			struct cifs_sid *pgrpsid, __u64 nmode, bool modefromsid)
@@ -817,23 +861,8 @@ static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid,
 	if (modefromsid) {
 		struct cifs_ace *pntace =
 			(struct cifs_ace *)((char *)pnndacl + size);
-		int i;
 
-		pntace->type = ACCESS_ALLOWED;
-		pntace->flags = 0x0;
-		pntace->access_req = 0;
-		pntace->sid.num_subauth = 3;
-		pntace->sid.revision = 1;
-		for (i = 0; i < NUM_AUTHS; i++)
-			pntace->sid.authority[i] =
-				sid_unix_NFS_mode.authority[i];
-		pntace->sid.sub_auth[0] = sid_unix_NFS_mode.sub_auth[0];
-		pntace->sid.sub_auth[1] = sid_unix_NFS_mode.sub_auth[1];
-		pntace->sid.sub_auth[2] = cpu_to_le32(nmode & 07777);
-
-		/* size = 1 + 1 + 2 + 4 + 1 + 1 + 6 + (psid->num_subauth*4) */
-		pntace->size = cpu_to_le16(28);
-		size += 28;
+		size += setup_special_mode_ACE(pntace, nmode);
 		num_aces++;
 	}
 
@@ -1056,7 +1085,7 @@ static struct cifs_ntsd *get_cifs_acl_by_path(struct cifs_sb_info *cifs_sb,
 	struct cifs_ntsd *pntsd = NULL;
 	int oplock = 0;
 	unsigned int xid;
-	int rc, create_options = 0;
+	int rc;
 	struct cifs_tcon *tcon;
 	struct tcon_link *tlink = cifs_sb_tlink(cifs_sb);
 	struct cifs_fid fid;
@@ -1068,13 +1097,10 @@ static struct cifs_ntsd *get_cifs_acl_by_path(struct cifs_sb_info *cifs_sb,
 	tcon = tlink_tcon(tlink);
 	xid = get_xid();
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
-
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = READ_CONTROL;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = path;
 	oparms.fid = &fid;
@@ -1119,7 +1145,7 @@ int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 {
 	int oplock = 0;
 	unsigned int xid;
-	int rc, access_flags, create_options = 0;
+	int rc, access_flags;
 	struct cifs_tcon *tcon;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
 	struct tcon_link *tlink = cifs_sb_tlink(cifs_sb);
@@ -1132,9 +1158,6 @@ int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 	tcon = tlink_tcon(tlink);
 	xid = get_xid();
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
-
 	if (aclflag == CIFS_ACL_OWNER || aclflag == CIFS_ACL_GROUP)
 		access_flags = WRITE_OWNER;
 	else
@@ -1143,7 +1166,7 @@ int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = access_flags;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = path;
 	oparms.fid = &fid;
diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h
index 439b99cefeb0..21d7dee98d01 100644
--- a/fs/cifs/cifsacl.h
+++ b/fs/cifs/cifsacl.h
@@ -147,22 +147,22 @@ struct smb3_sd {
 } __packed;
 
 /* Meaning of 'Control' field flags */
-#define ACL_CONTROL_SR	0x0001	/* Self relative */
-#define ACL_CONTROL_RM	0x0002	/* Resource manager control bits */
-#define ACL_CONTROL_PS	0x0004	/* SACL protected from inherits */
-#define ACL_CONTROL_PD	0x0008	/* DACL protected from inherits */
-#define ACL_CONTROL_SI	0x0010	/* SACL Auto-Inherited */
-#define ACL_CONTROL_DI	0x0020	/* DACL Auto-Inherited */
-#define ACL_CONTROL_SC	0x0040	/* SACL computed through inheritance */
-#define ACL_CONTROL_DC	0x0080	/* DACL computed through inheritence */
-#define ACL_CONTROL_SS	0x0100	/* Create server ACL */
-#define ACL_CONTROL_DT	0x0200	/* DACL provided by trusteed source */
-#define ACL_CONTROL_SD	0x0400	/* SACL defaulted */
-#define ACL_CONTROL_SP	0x0800	/* SACL is present on object */
-#define ACL_CONTROL_DD	0x1000	/* DACL defaulted */
-#define ACL_CONTROL_DP	0x2000	/* DACL is present on object */
-#define ACL_CONTROL_GD	0x4000	/* Group was defaulted */
-#define ACL_CONTROL_OD	0x8000	/* User was defaulted */
+#define ACL_CONTROL_SR	0x8000	/* Self relative */
+#define ACL_CONTROL_RM	0x4000	/* Resource manager control bits */
+#define ACL_CONTROL_PS	0x2000	/* SACL protected from inherits */
+#define ACL_CONTROL_PD	0x1000	/* DACL protected from inherits */
+#define ACL_CONTROL_SI	0x0800	/* SACL Auto-Inherited */
+#define ACL_CONTROL_DI	0x0400	/* DACL Auto-Inherited */
+#define ACL_CONTROL_SC	0x0200	/* SACL computed through inheritance */
+#define ACL_CONTROL_DC	0x0100	/* DACL computed through inheritence */
+#define ACL_CONTROL_SS	0x0080	/* Create server ACL */
+#define ACL_CONTROL_DT	0x0040	/* DACL provided by trusted source */
+#define ACL_CONTROL_SD	0x0020	/* SACL defaulted */
+#define ACL_CONTROL_SP	0x0010	/* SACL is present on object */
+#define ACL_CONTROL_DD	0x0008	/* DACL defaulted */
+#define ACL_CONTROL_DP	0x0004	/* DACL is present on object */
+#define ACL_CONTROL_GD	0x0002	/* Group was defaulted */
+#define ACL_CONTROL_OD	0x0001	/* User was defaulted */
 
 /* Meaning of AclRevision flags */
 #define ACL_REVISION	0x02 /* See section 2.4.4.1 of MS-DTYP */
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 1a135d1b85bd..c31f362fa098 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -119,6 +119,7 @@ extern mempool_t *cifs_mid_poolp;
 
 struct workqueue_struct	*cifsiod_wq;
 struct workqueue_struct	*decrypt_wq;
+struct workqueue_struct	*fileinfo_put_wq;
 struct workqueue_struct	*cifsoplockd_wq;
 __u32 cifs_lock_secret;
 
@@ -274,7 +275,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_ffree = 0;	/* unlimited */
 
 	if (server->ops->queryfs)
-		rc = server->ops->queryfs(xid, tcon, buf);
+		rc = server->ops->queryfs(xid, tcon, cifs_sb, buf);
 
 	free_xid(xid);
 	return 0;
@@ -413,7 +414,7 @@ cifs_show_security(struct seq_file *s, struct cifs_ses *ses)
 		seq_puts(s, "ntlm");
 		break;
 	case Kerberos:
-		seq_printf(s, "krb5,cruid=%u", from_kuid_munged(&init_user_ns,ses->cred_uid));
+		seq_puts(s, "krb5");
 		break;
 	case RawNTLMSSP:
 		seq_puts(s, "ntlmssp");
@@ -426,6 +427,10 @@ cifs_show_security(struct seq_file *s, struct cifs_ses *ses)
 
 	if (ses->sign)
 		seq_puts(s, "i");
+
+	if (ses->sectype == Kerberos)
+		seq_printf(s, ",cruid=%u",
+			   from_kuid_munged(&init_user_ns, ses->cred_uid));
 }
 
 static void
@@ -525,6 +530,8 @@ cifs_show_options(struct seq_file *s, struct dentry *root)
 
 	if (tcon->seal)
 		seq_puts(s, ",seal");
+	else if (tcon->ses->server->ignore_signature)
+		seq_puts(s, ",signloosely");
 	if (tcon->nocase)
 		seq_puts(s, ",nocase");
 	if (tcon->local_lease)
@@ -613,6 +620,10 @@ cifs_show_options(struct seq_file *s, struct dentry *root)
 	/* convert actimeo and display it in seconds */
 	seq_printf(s, ",actimeo=%lu", cifs_sb->actimeo / HZ);
 
+	if (tcon->ses->chan_max > 1)
+		seq_printf(s, ",multichannel,max_channel=%zu",
+			   tcon->ses->chan_max);
+
 	return 0;
 }
 
@@ -725,11 +736,6 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb)
 		struct inode *dir = d_inode(dentry);
 		struct dentry *child;
 
-		if (!dir) {
-			dput(dentry);
-			dentry = ERR_PTR(-ENOENT);
-			break;
-		}
 		if (!S_ISDIR(dir->i_mode)) {
 			dput(dentry);
 			dentry = ERR_PTR(-ENOTDIR);
@@ -746,7 +752,7 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb)
 		while (*s && *s != sep)
 			s++;
 
-		child = lookup_one_len_unlocked(p, dentry, s - p);
+		child = lookup_positive_unlocked(p, dentry, s - p);
 		dput(dentry);
 		dentry = child;
 	} while (!IS_ERR(dentry));
@@ -1012,7 +1018,7 @@ struct file_system_type cifs_fs_type = {
 	.name = "cifs",
 	.mount = cifs_do_mount,
 	.kill_sb = cifs_kill_sb,
-	/*  .fs_flags */
+	.fs_flags = FS_RENAME_DOES_D_MOVE,
 };
 MODULE_ALIAS_FS("cifs");
 
@@ -1021,7 +1027,7 @@ static struct file_system_type smb3_fs_type = {
 	.name = "smb3",
 	.mount = smb3_do_mount,
 	.kill_sb = cifs_kill_sb,
-	/*  .fs_flags */
+	.fs_flags = FS_RENAME_DOES_D_MOVE,
 };
 MODULE_ALIAS_FS("smb3");
 MODULE_ALIAS("smb3");
@@ -1202,6 +1208,10 @@ static ssize_t cifs_copy_file_range(struct file *src_file, loff_t off,
 {
 	unsigned int xid = get_xid();
 	ssize_t rc;
+	struct cifsFileInfo *cfile = dst_file->private_data;
+
+	if (cfile->swapfile)
+		return -EOPNOTSUPP;
 
 	rc = cifs_file_copychunk_range(xid, src_file, off, dst_file, destoff,
 					len, flags);
@@ -1219,6 +1229,7 @@ const struct file_operations cifs_file_ops = {
 	.open = cifs_open,
 	.release = cifs_close,
 	.lock = cifs_lock,
+	.flock = cifs_flock,
 	.fsync = cifs_fsync,
 	.flush = cifs_flush,
 	.mmap  = cifs_file_mmap,
@@ -1238,6 +1249,7 @@ const struct file_operations cifs_file_strict_ops = {
 	.open = cifs_open,
 	.release = cifs_close,
 	.lock = cifs_lock,
+	.flock = cifs_flock,
 	.fsync = cifs_strict_fsync,
 	.flush = cifs_flush,
 	.mmap = cifs_file_strict_mmap,
@@ -1257,6 +1269,7 @@ const struct file_operations cifs_file_direct_ops = {
 	.open = cifs_open,
 	.release = cifs_close,
 	.lock = cifs_lock,
+	.flock = cifs_flock,
 	.fsync = cifs_fsync,
 	.flush = cifs_flush,
 	.mmap = cifs_file_mmap,
@@ -1543,7 +1556,7 @@ init_cifs(void)
 	/*
 	 * Consider in future setting limit!=0 maybe to min(num_of_cores - 1, 3)
 	 * so that we don't launch too many worker threads but
-	 * Documentation/workqueue.txt recommends setting it to 0
+	 * Documentation/core-api/workqueue.rst recommends setting it to 0
 	 */
 
 	/* WQ_UNBOUND allows decrypt tasks to run on any CPU */
@@ -1554,11 +1567,18 @@ init_cifs(void)
 		goto out_destroy_cifsiod_wq;
 	}
 
+	fileinfo_put_wq = alloc_workqueue("cifsfileinfoput",
+				     WQ_UNBOUND|WQ_FREEZABLE|WQ_MEM_RECLAIM, 0);
+	if (!fileinfo_put_wq) {
+		rc = -ENOMEM;
+		goto out_destroy_decrypt_wq;
+	}
+
 	cifsoplockd_wq = alloc_workqueue("cifsoplockd",
 					 WQ_FREEZABLE|WQ_MEM_RECLAIM, 0);
 	if (!cifsoplockd_wq) {
 		rc = -ENOMEM;
-		goto out_destroy_decrypt_wq;
+		goto out_destroy_fileinfo_put_wq;
 	}
 
 	rc = cifs_fscache_register();
@@ -1624,6 +1644,8 @@ out_unreg_fscache:
 	cifs_fscache_unregister();
 out_destroy_cifsoplockd_wq:
 	destroy_workqueue(cifsoplockd_wq);
+out_destroy_fileinfo_put_wq:
+	destroy_workqueue(fileinfo_put_wq);
 out_destroy_decrypt_wq:
 	destroy_workqueue(decrypt_wq);
 out_destroy_cifsiod_wq:
@@ -1653,6 +1675,7 @@ exit_cifs(void)
 	cifs_fscache_unregister();
 	destroy_workqueue(cifsoplockd_wq);
 	destroy_workqueue(decrypt_wq);
+	destroy_workqueue(fileinfo_put_wq);
 	destroy_workqueue(cifsiod_wq);
 	cifs_proc_clean();
 }
@@ -1663,17 +1686,17 @@ MODULE_DESCRIPTION
 	("VFS to access SMB3 servers e.g. Samba, Macs, Azure and Windows (and "
 	"also older servers complying with the SNIA CIFS Specification)");
 MODULE_VERSION(CIFS_VERSION);
-MODULE_SOFTDEP("pre: ecb");
-MODULE_SOFTDEP("pre: hmac");
-MODULE_SOFTDEP("pre: md4");
-MODULE_SOFTDEP("pre: md5");
-MODULE_SOFTDEP("pre: nls");
-MODULE_SOFTDEP("pre: aes");
-MODULE_SOFTDEP("pre: cmac");
-MODULE_SOFTDEP("pre: sha256");
-MODULE_SOFTDEP("pre: sha512");
-MODULE_SOFTDEP("pre: aead2");
-MODULE_SOFTDEP("pre: ccm");
-MODULE_SOFTDEP("pre: gcm");
+MODULE_SOFTDEP("ecb");
+MODULE_SOFTDEP("hmac");
+MODULE_SOFTDEP("md4");
+MODULE_SOFTDEP("md5");
+MODULE_SOFTDEP("nls");
+MODULE_SOFTDEP("aes");
+MODULE_SOFTDEP("cmac");
+MODULE_SOFTDEP("sha256");
+MODULE_SOFTDEP("sha512");
+MODULE_SOFTDEP("aead2");
+MODULE_SOFTDEP("ccm");
+MODULE_SOFTDEP("gcm");
 module_init(init_cifs)
 module_exit(exit_cifs)
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index bc4ca94137f2..c9e2e6bbca13 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -108,6 +108,7 @@ extern ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to);
 extern ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from);
 extern ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from);
 extern ssize_t cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from);
+extern int cifs_flock(struct file *pfile, int cmd, struct file_lock *plock);
 extern int cifs_lock(struct file *, int, struct file_lock *);
 extern int cifs_fsync(struct file *, loff_t, loff_t, int);
 extern int cifs_strict_fsync(struct file *, loff_t, loff_t, int);
@@ -148,9 +149,12 @@ extern ssize_t cifs_file_copychunk_range(unsigned int xid,
 					size_t len, unsigned int flags);
 
 extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
+extern void cifs_setsize(struct inode *inode, loff_t offset);
+extern int cifs_truncate_page(struct address_space *mapping, loff_t from);
+
 #ifdef CONFIG_CIFS_NFSD_EXPORT
 extern const struct export_operations cifs_export_ops;
 #endif /* CONFIG_CIFS_NFSD_EXPORT */
 
-#define CIFS_VERSION   "2.23"
+#define CIFS_VERSION   "2.26"
 #endif				/* _CIFSFS_H */
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index d78bfcc19156..39b708d9d86d 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -230,7 +230,8 @@ struct smb_version_operations {
 	bool (*compare_fids)(struct cifsFileInfo *, struct cifsFileInfo *);
 	/* setup request: allocate mid, sign message */
 	struct mid_q_entry *(*setup_request)(struct cifs_ses *,
-						struct smb_rqst *);
+					     struct TCP_Server_Info *,
+					     struct smb_rqst *);
 	/* setup async request: allocate mid, sign message */
 	struct mid_q_entry *(*setup_async_request)(struct TCP_Server_Info *,
 						struct smb_rqst *);
@@ -268,8 +269,9 @@ struct smb_version_operations {
 	int (*check_message)(char *, unsigned int, struct TCP_Server_Info *);
 	bool (*is_oplock_break)(char *, struct TCP_Server_Info *);
 	int (*handle_cancelled_mid)(char *, struct TCP_Server_Info *);
-	void (*downgrade_oplock)(struct TCP_Server_Info *,
-					struct cifsInodeInfo *, bool);
+	void (*downgrade_oplock)(struct TCP_Server_Info *server,
+				 struct cifsInodeInfo *cinode, __u32 oplock,
+				 unsigned int epoch, bool *purge_cache);
 	/* process transaction2 response */
 	bool (*check_trans2)(struct mid_q_entry *, struct TCP_Server_Info *,
 			     char *, int);
@@ -296,7 +298,8 @@ struct smb_version_operations {
 			     const char *, struct dfs_info3_param **,
 			     unsigned int *, const struct nls_table *, int);
 	/* informational QFS call */
-	void (*qfs_tcon)(const unsigned int, struct cifs_tcon *);
+	void (*qfs_tcon)(const unsigned int, struct cifs_tcon *,
+			 struct cifs_sb_info *);
 	/* check if a path is accessible or not */
 	int (*is_path_accessible)(const unsigned int, struct cifs_tcon *,
 				  struct cifs_sb_info *, const char *);
@@ -366,6 +369,9 @@ struct smb_version_operations {
 	/* close a file */
 	void (*close)(const unsigned int, struct cifs_tcon *,
 		      struct cifs_fid *);
+	/* close a file, returning file attributes and timestamps */
+	void (*close_getattr)(const unsigned int xid, struct cifs_tcon *tcon,
+		      struct cifsFileInfo *pfile_info);
 	/* send a flush request to the server */
 	int (*flush)(const unsigned int, struct cifs_tcon *, struct cifs_fid *);
 	/* async read from the server */
@@ -404,7 +410,7 @@ struct smb_version_operations {
 			       struct cifsInodeInfo *);
 	/* query remote filesystem */
 	int (*queryfs)(const unsigned int, struct cifs_tcon *,
-		       struct kstatfs *);
+		       struct cifs_sb_info *, struct kstatfs *);
 	/* send mandatory brlock to the server */
 	int (*mand_lock)(const unsigned int, struct cifsFileInfo *, __u64,
 			 __u64, __u32, int, int, bool);
@@ -420,11 +426,14 @@ struct smb_version_operations {
 	/* generate new lease key */
 	void (*new_lease_key)(struct cifs_fid *);
 	int (*generate_signingkey)(struct cifs_ses *);
-	int (*calc_signature)(struct smb_rqst *, struct TCP_Server_Info *);
+	int (*calc_signature)(struct smb_rqst *, struct TCP_Server_Info *,
+				bool allocate_crypto);
 	int (*set_integrity)(const unsigned int, struct cifs_tcon *tcon,
 			     struct cifsFileInfo *src_file);
 	int (*enum_snapshots)(const unsigned int xid, struct cifs_tcon *tcon,
 			     struct cifsFileInfo *src_file, void __user *);
+	int (*notify)(const unsigned int xid, struct file *pfile,
+			     void __user *pbuf);
 	int (*query_mf_symlink)(unsigned int, struct cifs_tcon *,
 				struct cifs_sb_info *, const unsigned char *,
 				char *, unsigned int *);
@@ -485,6 +494,7 @@ struct smb_version_operations {
 	/* ioctl passthrough for query_info */
 	int (*ioctl_query_info)(const unsigned int xid,
 				struct cifs_tcon *tcon,
+				struct cifs_sb_info *cifs_sb,
 				__le16 *path, int is_dir,
 				unsigned long p);
 	/* make unix special files (block, char, fifo, socket) */
@@ -591,6 +601,10 @@ struct smb_vol {
 	bool resilient:1; /* noresilient not required since not fored for CA */
 	bool domainauto:1;
 	bool rdma:1;
+	bool multichannel:1;
+	bool use_client_guid:1;
+	/* reuse existing guid for multichannel */
+	u8 client_guid[SMB2_CLIENT_GUID_SIZE];
 	unsigned int bsize;
 	unsigned int rsize;
 	unsigned int wsize;
@@ -607,6 +621,7 @@ struct smb_vol {
 	__u64 snapshot_time; /* needed for timewarp tokens */
 	__u32 handle_timeout; /* persistent and durable handle timeout in ms */
 	unsigned int max_credits; /* smb3 max_credits 10 < credits < 60000 */
+	unsigned int max_channels;
 	__u16 compression; /* compression algorithm 0xFFFF default 0=disabled */
 	bool rootfs:1; /* if it's a SMB root file system */
 };
@@ -736,12 +751,12 @@ struct TCP_Server_Info {
 	/* Total size of this PDU. Only valid from cifs_demultiplex_thread */
 	unsigned int pdu_size;
 	unsigned int total_read; /* total amount of data read in this pass */
+	atomic_t in_send; /* requests trying to send */
+	atomic_t num_waiters;   /* blocked waiting to get in sendrecv */
 #ifdef CONFIG_CIFS_FSCACHE
 	struct fscache_cookie   *fscache; /* client index cache cookie */
 #endif
 #ifdef CONFIG_CIFS_STATS2
-	atomic_t in_send; /* requests trying to send */
-	atomic_t num_waiters;   /* blocked waiting to get in sendrecv */
 	atomic_t num_cmds[NUMBER_OF_SMB2_COMMANDS]; /* total requests by cmd */
 	atomic_t smb2slowcmd[NUMBER_OF_SMB2_COMMANDS]; /* count resps > 1 sec */
 	__u64 time_per_cmd[NUMBER_OF_SMB2_COMMANDS]; /* total time per cmd */
@@ -767,6 +782,7 @@ struct TCP_Server_Info {
 	 */
 	int nr_targets;
 	bool noblockcnt; /* use non-blocking connect() */
+	bool is_channel; /* if a session channel */
 };
 
 struct cifs_credits {
@@ -953,6 +969,11 @@ struct cifs_server_iface {
 	struct sockaddr_storage sockaddr;
 };
 
+struct cifs_chan {
+	struct TCP_Server_Info *server;
+	__u8 signkey[SMB3_SIGN_KEY_SIZE];
+};
+
 /*
  * Session structure.  One of these for each uid session with a particular host
  */
@@ -983,12 +1004,15 @@ struct cifs_ses {
 	bool sign;		/* is signing required? */
 	bool need_reconnect:1; /* connection reset, uid now invalid */
 	bool domainAuto:1;
+	bool binding:1; /* are we binding the session? */
 	__u16 session_flags;
 	__u8 smb3signingkey[SMB3_SIGN_KEY_SIZE];
 	__u8 smb3encryptionkey[SMB3_SIGN_KEY_SIZE];
 	__u8 smb3decryptionkey[SMB3_SIGN_KEY_SIZE];
 	__u8 preauth_sha_hash[SMB2_PREAUTH_HASH_SIZE];
 
+	__u8 binding_preauth_sha_hash[SMB2_PREAUTH_HASH_SIZE];
+
 	/*
 	 * Network interfaces available on the server this session is
 	 * connected to.
@@ -1002,8 +1026,37 @@ struct cifs_ses {
 	struct cifs_server_iface *iface_list;
 	size_t iface_count;
 	unsigned long iface_last_update; /* jiffies */
+
+#define CIFS_MAX_CHANNELS 16
+	struct cifs_chan chans[CIFS_MAX_CHANNELS];
+	size_t chan_count;
+	size_t chan_max;
+	atomic_t chan_seq; /* round robin state */
 };
 
+/*
+ * When binding a new channel, we need to access the channel which isn't fully
+ * established yet (one past the established count)
+ */
+
+static inline
+struct cifs_chan *cifs_ses_binding_channel(struct cifs_ses *ses)
+{
+	if (ses->binding)
+		return &ses->chans[ses->chan_count];
+	else
+		return NULL;
+}
+
+static inline
+struct TCP_Server_Info *cifs_ses_server(struct cifs_ses *ses)
+{
+	if (ses->binding)
+		return ses->chans[ses->chan_count].server;
+	else
+		return ses->server;
+}
+
 static inline bool
 cap_unix(struct cifs_ses *ses)
 {
@@ -1013,7 +1066,7 @@ cap_unix(struct cifs_ses *ses)
 struct cached_fid {
 	bool is_valid:1;	/* Do we have a useable root fid */
 	bool file_all_info_is_valid:1;
-
+	bool has_lease:1;
 	struct kref refcount;
 	struct cifs_fid *fid;
 	struct mutex fid_mutex;
@@ -1229,6 +1282,7 @@ struct cifs_fid {
 	__u64 volatile_fid;	/* volatile file id for smb2 */
 	__u8 lease_key[SMB2_LEASE_KEY_SIZE];	/* lease key for smb2 */
 	__u8 create_guid[16];
+	__u32 access;
 	struct cifs_pending_open *pending_open;
 	unsigned int epoch;
 #ifdef CONFIG_CIFS_DEBUG2
@@ -1259,12 +1313,16 @@ struct cifsFileInfo {
 	struct tcon_link *tlink;
 	unsigned int f_flags;
 	bool invalidHandle:1;	/* file closed via session abend */
+	bool swapfile:1;
 	bool oplock_break_cancelled:1;
+	unsigned int oplock_epoch; /* epoch from the lease break */
+	__u32 oplock_level; /* oplock/lease level from the lease break */
 	int count;
 	spinlock_t file_info_lock; /* protects four flag/count fields above */
 	struct mutex fh_mutex; /* prevents reopen race after dead ses*/
 	struct cifs_search_info srch_inf;
 	struct work_struct oplock_break; /* work for oplock breaks */
+	struct work_struct put; /* work for the final part of _put */
 };
 
 struct cifs_io_parms {
@@ -1370,7 +1428,8 @@ cifsFileInfo_get_locked(struct cifsFileInfo *cifs_file)
 }
 
 struct cifsFileInfo *cifsFileInfo_get(struct cifsFileInfo *cifs_file);
-void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_hdlr);
+void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_hdlr,
+		       bool offload);
 void cifsFileInfo_put(struct cifsFileInfo *cifs_file);
 
 #define CIFS_CACHE_READ_FLG	1
@@ -1405,7 +1464,7 @@ struct cifsInodeInfo {
 	unsigned int epoch;		/* used to track lease state changes */
 #define CIFS_INODE_PENDING_OPLOCK_BREAK   (0) /* oplock break in progress */
 #define CIFS_INODE_PENDING_WRITERS	  (1) /* Writes in progress */
-#define CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2 (2) /* Downgrade oplock to L2 */
+#define CIFS_INODE_FLAG_UNUSED		  (2) /* Unused flag */
 #define CIFS_INO_DELETE_PENDING		  (3) /* delete pending on server */
 #define CIFS_INO_INVALID_MAPPING	  (4) /* pagecache is invalid */
 #define CIFS_INO_LOCK			  (5) /* lock bit for synchronization */
@@ -1524,6 +1583,7 @@ struct mid_q_entry {
 	struct TCP_Server_Info *server;	/* server corresponding to this mid */
 	__u64 mid;		/* multiplex id */
 	__u16 credits;		/* number of credits consumed by this mid */
+	__u16 credits_received;	/* number of credits from the response */
 	__u32 pid;		/* process id */
 	__u32 sequence_number;  /* for CIFS signing */
 	unsigned long when_alloc;  /* when mid was created */
@@ -1535,6 +1595,7 @@ struct mid_q_entry {
 	mid_callback_t *callback; /* call completion callback */
 	mid_handle_t *handle; /* call handle mid callback */
 	void *callback_data;	  /* general purpose pointer for callback */
+	struct task_struct *creator;
 	void *resp_buf;		/* pointer to received SMB header */
 	unsigned int resp_buf_size;
 	int mid_state;	/* wish this were enum but can not pass to wait_event */
@@ -1551,12 +1612,12 @@ struct close_cancelled_open {
 	struct cifs_fid         fid;
 	struct cifs_tcon        *tcon;
 	struct work_struct      work;
+	__u64 mid;
+	__u16 cmd;
 };
 
 /*	Make code in transport.c a little cleaner by moving
 	update of optional stats into function below */
-#ifdef CONFIG_CIFS_STATS2
-
 static inline void cifs_in_send_inc(struct TCP_Server_Info *server)
 {
 	atomic_inc(&server->in_send);
@@ -1577,26 +1638,12 @@ static inline void cifs_num_waiters_dec(struct TCP_Server_Info *server)
 	atomic_dec(&server->num_waiters);
 }
 
+#ifdef CONFIG_CIFS_STATS2
 static inline void cifs_save_when_sent(struct mid_q_entry *mid)
 {
 	mid->when_sent = jiffies;
 }
 #else
-static inline void cifs_in_send_inc(struct TCP_Server_Info *server)
-{
-}
-static inline void cifs_in_send_dec(struct TCP_Server_Info *server)
-{
-}
-
-static inline void cifs_num_waiters_inc(struct TCP_Server_Info *server)
-{
-}
-
-static inline void cifs_num_waiters_dec(struct TCP_Server_Info *server)
-{
-}
-
 static inline void cifs_save_when_sent(struct mid_q_entry *mid)
 {
 }
@@ -1654,6 +1701,7 @@ struct cifs_fattr {
 	struct timespec64 cf_atime;
 	struct timespec64 cf_mtime;
 	struct timespec64 cf_ctime;
+	u32             cf_cifstag;
 };
 
 static inline void free_dfs_info_param(struct dfs_info3_param *param)
@@ -1696,6 +1744,12 @@ static inline bool is_retryable_error(int error)
 	return false;
 }
 
+
+/* cifs_get_writable_file() flags */
+#define FIND_WR_ANY         0
+#define FIND_WR_FSUID_ONLY  1
+#define FIND_WR_WITH_DELETE 2
+
 #define   MID_FREE 0
 #define   MID_REQUEST_ALLOCATED 1
 #define   MID_REQUEST_SUBMITTED 2
@@ -1837,7 +1891,8 @@ GLOBAL_EXTERN struct list_head		cifs_tcp_ses_list;
 /*
  * This lock protects the cifs_tcp_ses_list, the list of smb sessions per
  * tcp session, and the list of tcon's per smb session. It also protects
- * the reference counters for the server, smb session, and tcon. Finally,
+ * the reference counters for the server, smb session, and tcon. It also
+ * protects some fields in the TCP_Server_Info struct such as dstaddr. Finally,
  * changes to the tcon->tidStatus should be done while holding this lock.
  * generally the locks should be taken in order tcp_ses_lock before
  * tcon->open_file_lock and that before file->file_info_lock since the
@@ -1907,6 +1962,7 @@ void cifs_queue_oplock_break(struct cifsFileInfo *cfile);
 extern const struct slow_work_ops cifs_oplock_break_ops;
 extern struct workqueue_struct *cifsiod_wq;
 extern struct workqueue_struct *decrypt_wq;
+extern struct workqueue_struct *fileinfo_put_wq;
 extern struct workqueue_struct *cifsoplockd_wq;
 extern __u32 cifs_lock_secret;
 
@@ -1937,4 +1993,10 @@ extern struct smb_version_values smb302_values;
 #define ALT_SMB311_VERSION_STRING "3.11"
 extern struct smb_version_operations smb311_operations;
 extern struct smb_version_values smb311_values;
+
+static inline bool is_smb1_server(struct TCP_Server_Info *server)
+{
+	return strcmp(server->vals->version_string, SMB1_VERSION_STRING) == 0;
+}
+
 #endif	/* _CIFS_GLOB_H */
diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h
index 79d842e7240c..593d826820c3 100644
--- a/fs/cifs/cifspdu.h
+++ b/fs/cifs/cifspdu.h
@@ -1021,7 +1021,7 @@ typedef struct smb_com_writex_req {
 	__le16 ByteCount;
 	__u8 Pad;		/* BB check for whether padded to DWORD
 				   boundary and optimum performance here */
-	char Data[0];
+	char Data[];
 } __attribute__((packed)) WRITEX_REQ;
 
 typedef struct smb_com_write_req {
@@ -1041,7 +1041,7 @@ typedef struct smb_com_write_req {
 	__le16 ByteCount;
 	__u8 Pad;		/* BB check for whether padded to DWORD
 				   boundary and optimum performance here */
-	char Data[0];
+	char Data[];
 } __attribute__((packed)) WRITE_REQ;
 
 typedef struct smb_com_write_rsp {
@@ -1306,7 +1306,7 @@ typedef struct smb_com_ntransact_req {
 	/* SetupCount words follow then */
 	__le16 ByteCount;
 	__u8 Pad[3];
-	__u8 Parms[0];
+	__u8 Parms[];
 } __attribute__((packed)) NTRANSACT_REQ;
 
 typedef struct smb_com_ntransact_rsp {
@@ -1523,7 +1523,7 @@ struct file_notify_information {
 	__le32 NextEntryOffset;
 	__le32 Action;
 	__le32 FileNameLength;
-	__u8  FileName[0];
+	__u8  FileName[];
 } __attribute__((packed));
 
 /* For IO_REPARSE_TAG_SYMLINK */
@@ -1536,7 +1536,7 @@ struct reparse_symlink_data {
 	__le16	PrintNameOffset;
 	__le16	PrintNameLength;
 	__le32	Flags;
-	char	PathBuffer[0];
+	char	PathBuffer[];
 } __attribute__((packed));
 
 /* Flag above */
@@ -1553,7 +1553,7 @@ struct reparse_posix_data {
 	__le16	ReparseDataLength;
 	__u16	Reserved;
 	__le64	InodeType; /* LNK, FIFO, CHR etc. */
-	char	PathBuffer[0];
+	char	PathBuffer[];
 } __attribute__((packed));
 
 struct cifs_quota_data {
@@ -1691,6 +1691,7 @@ struct smb_t2_rsp {
 #define SMB_FIND_FILE_ID_FULL_DIR_INFO    0x105
 #define SMB_FIND_FILE_ID_BOTH_DIR_INFO    0x106
 #define SMB_FIND_FILE_UNIX                0x202
+#define SMB_FIND_FILE_POSIX_INFO          0x064
 
 typedef struct smb_com_transaction2_qpi_req {
 	struct smb_hdr hdr;	/* wct = 14+ */
@@ -1761,7 +1762,7 @@ struct set_file_rename {
 	__le32 overwrite;   /* 1 = overwrite dest */
 	__u32 root_fid;   /* zero */
 	__le32 target_name_len;
-	char  target_name[0];  /* Must be unicode */
+	char  target_name[];  /* Must be unicode */
 } __attribute__((packed));
 
 struct smb_com_transaction2_sfi_req {
@@ -2450,7 +2451,7 @@ struct cifs_posix_acl { /* access conrol list  (ACL) */
 	__le16	version;
 	__le16	access_entry_count;  /* access ACL - count of entries */
 	__le16	default_entry_count; /* default ACL - count of entries */
-	struct cifs_posix_ace ace_array[0];
+	struct cifs_posix_ace ace_array[];
 	/* followed by
 	struct cifs_posix_ace default_ace_arraay[] */
 } __attribute__((packed));  /* level 0x204 */
@@ -2756,7 +2757,7 @@ typedef struct file_xattr_info {
 	/* BB do we need another field for flags? BB */
 	__u32 xattr_name_len;
 	__u32 xattr_value_len;
-	char  xattr_name[0];
+	char  xattr_name[];
 	/* followed by xattr_value[xattr_value_len], no pad */
 } __attribute__((packed)) FILE_XATTR_INFO; /* extended attribute info
 					      level 0x205 */
diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h
index fe597d3d5208..12a895e02db4 100644
--- a/fs/cifs/cifsproto.h
+++ b/fs/cifs/cifsproto.h
@@ -109,6 +109,7 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifs_ses *,
 extern int SendReceiveNoRsp(const unsigned int xid, struct cifs_ses *ses,
 			    char *in_buf, int flags);
 extern struct mid_q_entry *cifs_setup_request(struct cifs_ses *,
+				struct TCP_Server_Info *,
 				struct smb_rqst *);
 extern struct mid_q_entry *cifs_setup_async_request(struct TCP_Server_Info *,
 						struct smb_rqst *);
@@ -133,11 +134,12 @@ extern bool backup_cred(struct cifs_sb_info *);
 extern bool is_size_safe_to_change(struct cifsInodeInfo *, __u64 eof);
 extern void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset,
 			    unsigned int bytes_written);
-extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, bool);
+extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, int);
 extern int cifs_get_writable_file(struct cifsInodeInfo *cifs_inode,
-				  bool fsuid_only,
+				  int flags,
 				  struct cifsFileInfo **ret_file);
 extern int cifs_get_writable_path(struct cifs_tcon *tcon, const char *name,
+				  int flags,
 				  struct cifsFileInfo **ret_file);
 extern struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *, bool);
 extern int cifs_get_readable_path(struct cifs_tcon *tcon, const char *name,
@@ -212,6 +214,8 @@ extern struct cifs_ntsd *get_cifs_acl_by_fid(struct cifs_sb_info *,
 						const struct cifs_fid *, u32 *);
 extern int set_cifs_acl(struct cifs_ntsd *, __u32, struct inode *,
 				const char *, int);
+extern unsigned int setup_authusers_ACE(struct cifs_ace *pace);
+extern unsigned int setup_special_mode_ACE(struct cifs_ace *pace, __u64 nmode);
 
 extern void dequeue_mid(struct mid_q_entry *mid, bool malformed);
 extern int cifs_read_from_socket(struct TCP_Server_Info *server, char *buf,
@@ -242,6 +246,7 @@ extern void cifs_add_pending_open_locked(struct cifs_fid *fid,
 					 struct tcon_link *tlink,
 					 struct cifs_pending_open *open);
 extern void cifs_del_pending_open(struct cifs_pending_open *open);
+extern struct TCP_Server_Info *cifs_get_tcp_session(struct smb_vol *vol);
 extern void cifs_put_tcp_session(struct TCP_Server_Info *server,
 				 int from_reconnect);
 extern void cifs_put_tcon(struct cifs_tcon *tcon);
@@ -584,9 +589,23 @@ void cifs_free_hash(struct crypto_shash **shash, struct sdesc **sdesc);
 
 extern void rqst_page_get_length(struct smb_rqst *rqst, unsigned int page,
 				unsigned int *len, unsigned int *offset);
+int cifs_try_adding_channels(struct cifs_ses *ses);
+int cifs_ses_add_channel(struct cifs_ses *ses,
+				struct cifs_server_iface *iface);
+bool is_server_using_iface(struct TCP_Server_Info *server,
+			   struct cifs_server_iface *iface);
+bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface);
 
 void extract_unc_hostname(const char *unc, const char **h, size_t *len);
 int copy_path_name(char *dst, const char *src);
+int smb2_parse_query_directory(struct cifs_tcon *tcon, struct kvec *rsp_iov,
+			       int resp_buftype,
+			       struct cifs_search_info *srch_inf);
+
+struct super_block *cifs_get_tcp_super(struct TCP_Server_Info *server);
+void cifs_put_tcp_super(struct super_block *sb);
+int update_super_prepath(struct cifs_tcon *tcon, const char *prefix,
+			 size_t prefix_len);
 
 #ifdef CONFIG_CIFS_DFS_UPCALL
 static inline int get_dfs_path(const unsigned int xid, struct cifs_ses *ses,
@@ -599,4 +618,12 @@ static inline int get_dfs_path(const unsigned int xid, struct cifs_ses *ses,
 }
 #endif
 
+static inline int cifs_create_options(struct cifs_sb_info *cifs_sb, int options)
+{
+	if (cifs_sb && (backup_cred(cifs_sb)))
+		return options | CREATE_OPEN_BACKUP_INTENT;
+	else
+		return options;
+}
+
 #endif			/* _CIFSPROTO_H */
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index 4f554f019a98..5014a82391ff 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -42,6 +42,7 @@
 #include "cifsproto.h"
 #include "cifs_unicode.h"
 #include "cifs_debug.h"
+#include "smb2proto.h"
 #include "fscache.h"
 #include "smbdirect.h"
 #ifdef CONFIG_CIFS_DFS_UPCALL
@@ -112,6 +113,8 @@ cifs_mark_open_files_invalid(struct cifs_tcon *tcon)
 
 	mutex_lock(&tcon->crfid.fid_mutex);
 	tcon->crfid.is_valid = false;
+	/* cached handle is not valid, so SMB2_CLOSE won't be sent below */
+	close_shroot_lease_locked(&tcon->crfid);
 	memset(tcon->crfid.fid, 0, sizeof(struct cifs_fid));
 	mutex_unlock(&tcon->crfid.fid_mutex);
 
@@ -159,9 +162,18 @@ static int __cifs_reconnect_tcon(const struct nls_table *nlsc,
 
 	for (it = dfs_cache_get_tgt_iterator(&tl); it;
 	     it = dfs_cache_get_next_tgt(&tl, it)) {
-		const char *tgt = dfs_cache_get_tgt_name(it);
+		const char *share, *prefix;
+		size_t share_len, prefix_len;
 
-		extract_unc_hostname(tgt, &dfs_host, &dfs_host_len);
+		rc = dfs_cache_get_tgt_share(it, &share, &share_len, &prefix,
+					     &prefix_len);
+		if (rc) {
+			cifs_dbg(VFS, "%s: failed to parse target share %d\n",
+				 __func__, rc);
+			continue;
+		}
+
+		extract_unc_hostname(share, &dfs_host, &dfs_host_len);
 
 		if (dfs_host_len != tcp_host_len
 		    || strncasecmp(dfs_host, tcp_host, dfs_host_len) != 0) {
@@ -172,11 +184,13 @@ static int __cifs_reconnect_tcon(const struct nls_table *nlsc,
 			continue;
 		}
 
-		scnprintf(tree, MAX_TREE_SIZE, "\\%s", tgt);
+		scnprintf(tree, MAX_TREE_SIZE, "\\%.*s", (int)share_len, share);
 
 		rc = CIFSTCon(0, tcon->ses, tree, tcon, nlsc);
-		if (!rc)
+		if (!rc) {
+			rc = update_super_prepath(tcon, prefix, prefix_len);
 			break;
+		}
 		if (rc == -EREMOTE)
 			break;
 	}
@@ -257,7 +271,7 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command)
 		if (server->tcpStatus != CifsNeedReconnect)
 			break;
 
-		if (--retries)
+		if (retries && --retries)
 			continue;
 
 		/*
@@ -317,7 +331,7 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command)
 	atomic_inc(&tconInfoReconnectCount);
 
 	/* tell server Unix caps we support */
-	if (ses->capabilities & CAP_UNIX)
+	if (cap_unix(ses))
 		reset_cifs_unix_caps(0, tcon, NULL, NULL);
 
 	/*
@@ -580,6 +594,8 @@ decode_lanman_negprot_rsp(struct TCP_Server_Info *server, NEGOTIATE_RSP *pSMBr)
 			       cifs_max_pending);
 	set_credits(server, server->maxReq);
 	server->maxBuf = le16_to_cpu(rsp->MaxBufSize);
+	/* set up max_read for readpages check */
+	server->max_read = server->maxBuf;
 	/* even though we do not use raw we might as well set this
 	accurately, in case we ever find a need for it */
 	if ((le16_to_cpu(rsp->RawMode) & RAW_ENABLE) == RAW_ENABLE) {
@@ -741,6 +757,8 @@ CIFSSMBNegotiate(const unsigned int xid, struct cifs_ses *ses)
 	set_credits(server, server->maxReq);
 	/* probably no need to store and check maxvcs */
 	server->maxBuf = le32_to_cpu(pSMBr->MaxBufferSize);
+	/* set up max_read for readpages check */
+	server->max_read = server->maxBuf;
 	server->max_rw = le32_to_cpu(pSMBr->MaxRawSize);
 	cifs_dbg(NOISY, "Max buf = %d\n", ses->server->maxBuf);
 	server->capabilities = le32_to_cpu(pSMBr->Capabilities);
@@ -1489,6 +1507,7 @@ openRetry:
 	*oplock = rsp->OplockLevel;
 	/* cifs fid stays in le */
 	oparms->fid->netfid = rsp->Fid;
+	oparms->fid->access = desired_access;
 
 	/* Let caller know file was created so we can set the mode. */
 	/* Do we care about the CreateAction in any other cases? */
@@ -1587,7 +1606,6 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 	if (server->ops->is_session_expired &&
 	    server->ops->is_session_expired(buf)) {
 		cifs_reconnect(server);
-		wake_up(&server->response_q);
 		return -1;
 	}
 
@@ -2112,7 +2130,7 @@ cifs_writev_requeue(struct cifs_writedata *wdata)
 		wdata2->tailsz = tailsz;
 		wdata2->bytes = cur_len;
 
-		rc = cifs_get_writable_file(CIFS_I(inode), false,
+		rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY,
 					    &wdata2->cfile);
 		if (!wdata2->cfile) {
 			cifs_dbg(VFS, "No writable handle to retry writepages rc=%d\n",
@@ -2134,8 +2152,8 @@ cifs_writev_requeue(struct cifs_writedata *wdata)
 			}
 		}
 
+		kref_put(&wdata2->refcount, cifs_writedata_release);
 		if (rc) {
-			kref_put(&wdata2->refcount, cifs_writedata_release);
 			if (is_retryable_error(rc))
 				continue;
 			i += nr_pages;
@@ -4616,7 +4634,7 @@ findFirstRetry:
 				psrch_inf->unicode = false;
 
 			psrch_inf->ntwrk_buf_start = (char *)pSMBr;
-			psrch_inf->smallBuf = 0;
+			psrch_inf->smallBuf = false;
 			psrch_inf->srch_entries_start =
 				(char *) &pSMBr->hdr.Protocol +
 					le16_to_cpu(pSMBr->t2.DataOffset);
@@ -4750,7 +4768,7 @@ int CIFSFindNext(const unsigned int xid, struct cifs_tcon *tcon,
 				cifs_buf_release(psrch_inf->ntwrk_buf_start);
 			psrch_inf->srch_entries_start = response_data;
 			psrch_inf->ntwrk_buf_start = (char *)pSMB;
-			psrch_inf->smallBuf = 0;
+			psrch_inf->smallBuf = false;
 			if (parms->EndofSearch)
 				psrch_inf->endOfSearch = true;
 			else
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index ccaa8bad336f..28268ed461b8 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -21,6 +21,7 @@
 #include <linux/fs.h>
 #include <linux/net.h>
 #include <linux/string.h>
+#include <linux/sched/mm.h>
 #include <linux/sched/signal.h>
 #include <linux/list.h>
 #include <linux/wait.h>
@@ -57,7 +58,6 @@
 #include "smb2proto.h"
 #include "smbdirect.h"
 #include "dns_resolve.h"
-#include "cifsfs.h"
 #ifdef CONFIG_CIFS_DFS_UPCALL
 #include "dfs_cache.h"
 #endif
@@ -97,6 +97,7 @@ enum {
 	Opt_persistent, Opt_nopersistent,
 	Opt_resilient, Opt_noresilient,
 	Opt_domainauto, Opt_rdma, Opt_modesid, Opt_rootfs,
+	Opt_multichannel, Opt_nomultichannel,
 	Opt_compress,
 
 	/* Mount options which take numeric value */
@@ -106,7 +107,7 @@ enum {
 	Opt_min_enc_offload,
 	Opt_blocksize, Opt_rsize, Opt_wsize, Opt_actimeo,
 	Opt_echo_interval, Opt_max_credits, Opt_handletimeout,
-	Opt_snapshot,
+	Opt_snapshot, Opt_max_channels,
 
 	/* Mount options which take string value */
 	Opt_user, Opt_pass, Opt_ip,
@@ -199,6 +200,8 @@ static const match_table_t cifs_mount_option_tokens = {
 	{ Opt_noresilient, "noresilienthandles"},
 	{ Opt_domainauto, "domainauto"},
 	{ Opt_rdma, "rdma"},
+	{ Opt_multichannel, "multichannel" },
+	{ Opt_nomultichannel, "nomultichannel" },
 
 	{ Opt_backupuid, "backupuid=%s" },
 	{ Opt_backupgid, "backupgid=%s" },
@@ -218,6 +221,7 @@ static const match_table_t cifs_mount_option_tokens = {
 	{ Opt_echo_interval, "echo_interval=%s" },
 	{ Opt_max_credits, "max_credits=%s" },
 	{ Opt_snapshot, "snapshot=%s" },
+	{ Opt_max_channels, "max_channels=%s" },
 	{ Opt_compress, "compress=%s" },
 
 	{ Opt_blank_user, "user=" },
@@ -371,8 +375,10 @@ static int reconn_set_ipaddr(struct TCP_Server_Info *server)
 		return rc;
 	}
 
+	spin_lock(&cifs_tcp_ses_lock);
 	rc = cifs_convert_address((struct sockaddr *)&server->dstaddr, ipaddr,
 				  strlen(ipaddr));
+	spin_unlock(&cifs_tcp_ses_lock);
 	kfree(ipaddr);
 
 	return !rc ? -1 : 0;
@@ -385,40 +391,7 @@ static inline int reconn_set_ipaddr(struct TCP_Server_Info *server)
 #endif
 
 #ifdef CONFIG_CIFS_DFS_UPCALL
-struct super_cb_data {
-	struct TCP_Server_Info *server;
-	struct cifs_sb_info *cifs_sb;
-};
-
 /* These functions must be called with server->srv_mutex held */
-
-static void super_cb(struct super_block *sb, void *arg)
-{
-	struct super_cb_data *d = arg;
-	struct cifs_sb_info *cifs_sb;
-	struct cifs_tcon *tcon;
-
-	if (d->cifs_sb)
-		return;
-
-	cifs_sb = CIFS_SB(sb);
-	tcon = cifs_sb_master_tcon(cifs_sb);
-	if (tcon->ses->server == d->server)
-		d->cifs_sb = cifs_sb;
-}
-
-static inline struct cifs_sb_info *
-find_super_by_tcp(struct TCP_Server_Info *server)
-{
-	struct super_cb_data d = {
-		.server = server,
-		.cifs_sb = NULL,
-	};
-
-	iterate_supers_type(&cifs_fs_type, super_cb, &d);
-	return d.cifs_sb ? d.cifs_sb : ERR_PTR(-ENOENT);
-}
-
 static void reconn_inval_dfs_target(struct TCP_Server_Info *server,
 				    struct cifs_sb_info *cifs_sb,
 				    struct dfs_cache_tgt_list *tgt_list,
@@ -480,6 +453,7 @@ cifs_reconnect(struct TCP_Server_Info *server)
 	struct mid_q_entry *mid_entry;
 	struct list_head retry_list;
 #ifdef CONFIG_CIFS_DFS_UPCALL
+	struct super_block *sb = NULL;
 	struct cifs_sb_info *cifs_sb = NULL;
 	struct dfs_cache_tgt_list tgt_list = {0};
 	struct dfs_cache_tgt_iterator *tgt_it = NULL;
@@ -489,13 +463,15 @@ cifs_reconnect(struct TCP_Server_Info *server)
 	server->nr_targets = 1;
 #ifdef CONFIG_CIFS_DFS_UPCALL
 	spin_unlock(&GlobalMid_Lock);
-	cifs_sb = find_super_by_tcp(server);
-	if (IS_ERR(cifs_sb)) {
-		rc = PTR_ERR(cifs_sb);
+	sb = cifs_get_tcp_super(server);
+	if (IS_ERR(sb)) {
+		rc = PTR_ERR(sb);
 		cifs_dbg(FYI, "%s: will not do DFS failover: rc = %d\n",
 			 __func__, rc);
-		cifs_sb = NULL;
+		sb = NULL;
 	} else {
+		cifs_sb = CIFS_SB(sb);
+
 		rc = reconn_setup_dfs_targets(cifs_sb, &tgt_list, &tgt_it);
 		if (rc && (rc != -EOPNOTSUPP)) {
 			cifs_server_dbg(VFS, "%s: no target servers for DFS failover\n",
@@ -512,6 +488,11 @@ cifs_reconnect(struct TCP_Server_Info *server)
 		/* the demux thread will exit normally
 		next time through the loop */
 		spin_unlock(&GlobalMid_Lock);
+#ifdef CONFIG_CIFS_DFS_UPCALL
+		dfs_cache_free_tgts(&tgt_list);
+		cifs_put_tcp_super(sb);
+#endif
+		wake_up(&server->response_q);
 		return rc;
 	} else
 		server->tcpStatus = CifsNeedReconnect;
@@ -638,11 +619,15 @@ cifs_reconnect(struct TCP_Server_Info *server)
 				 __func__, rc);
 		}
 		dfs_cache_free_tgts(&tgt_list);
+
 	}
+
+	cifs_put_tcp_super(sb);
 #endif
 	if (server->tcpStatus == CifsNeedNegotiate)
 		mod_delayed_work(cifsiod_wq, &server->echo, 0);
 
+	wake_up(&server->response_q);
 	return rc;
 }
 
@@ -737,7 +722,6 @@ server_unresponsive(struct TCP_Server_Info *server)
 		cifs_server_dbg(VFS, "has not responded in %lu seconds. Reconnecting...\n",
 			 (3 * server->echo_interval) / HZ);
 		cifs_reconnect(server);
-		wake_up(&server->response_q);
 		return true;
 	}
 
@@ -870,7 +854,6 @@ is_smb_response(struct TCP_Server_Info *server, unsigned char type)
 		 */
 		cifs_set_port((struct sockaddr *)&server->dstaddr, CIFS_PORT);
 		cifs_reconnect(server);
-		wake_up(&server->response_q);
 		break;
 	default:
 		cifs_server_dbg(VFS, "RFC 1002 unknown response type 0x%x\n", type);
@@ -905,6 +888,20 @@ dequeue_mid(struct mid_q_entry *mid, bool malformed)
 	spin_unlock(&GlobalMid_Lock);
 }
 
+static unsigned int
+smb2_get_credits_from_hdr(char *buffer, struct TCP_Server_Info *server)
+{
+	struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buffer;
+
+	/*
+	 * SMB1 does not use credits.
+	 */
+	if (server->vals->header_preamble_size)
+		return 0;
+
+	return le16_to_cpu(shdr->CreditRequest);
+}
+
 static void
 handle_mid(struct mid_q_entry *mid, struct TCP_Server_Info *server,
 	   char *buf, int malformed)
@@ -912,6 +909,7 @@ handle_mid(struct mid_q_entry *mid, struct TCP_Server_Info *server,
 	if (server->ops->check_trans2 &&
 	    server->ops->check_trans2(mid, server, buf, malformed))
 		return;
+	mid->credits_received = smb2_get_credits_from_hdr(buf, server);
 	mid->resp_buf = buf;
 	mid->large_buf = server->large_buf;
 	/* Was previous buf put in mpx struct for multi-rsp? */
@@ -1027,7 +1025,6 @@ standard_receive3(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 		server->vals->header_preamble_size) {
 		cifs_server_dbg(VFS, "SMB response too long (%u bytes)\n", pdu_length);
 		cifs_reconnect(server);
-		wake_up(&server->response_q);
 		return -ECONNABORTED;
 	}
 
@@ -1075,7 +1072,6 @@ cifs_handle_standard(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 	if (server->ops->is_session_expired &&
 	    server->ops->is_session_expired(buf)) {
 		cifs_reconnect(server);
-		wake_up(&server->response_q);
 		return -1;
 	}
 
@@ -1121,8 +1117,9 @@ cifs_demultiplex_thread(void *p)
 	struct task_struct *task_to_wake = NULL;
 	struct mid_q_entry *mids[MAX_COMPOUND];
 	char *bufs[MAX_COMPOUND];
+	unsigned int noreclaim_flag;
 
-	current->flags |= PF_MEMALLOC;
+	noreclaim_flag = memalloc_noreclaim_save();
 	cifs_dbg(FYI, "Demultiplex PID: %d\n", task_pid_nr(current));
 
 	length = atomic_inc_return(&tcpSesAllocCount);
@@ -1169,7 +1166,6 @@ next_pdu:
 			cifs_server_dbg(VFS, "SMB response too short (%u bytes)\n",
 				 server->pdu_size);
 			cifs_reconnect(server);
-			wake_up(&server->response_q);
 			continue;
 		}
 
@@ -1222,12 +1218,6 @@ next_pdu:
 		for (i = 0; i < num_mids; i++) {
 			if (mids[i] != NULL) {
 				mids[i]->resp_buf_size = server->pdu_size;
-				if ((mids[i]->mid_flags & MID_WAIT_CANCELLED) &&
-				    mids[i]->mid_state == MID_RESPONSE_RECEIVED &&
-				    server->ops->handle_cancelled_mid)
-					server->ops->handle_cancelled_mid(
-							mids[i]->resp_buf,
-							server);
 
 				if (!mids[i]->multiRsp || mids[i]->multiEnd)
 					mids[i]->callback(mids[i]);
@@ -1283,6 +1273,7 @@ next_pdu:
 		set_current_state(TASK_RUNNING);
 	}
 
+	memalloc_noreclaim_restore(noreclaim_flag);
 	module_put_and_exit(0);
 }
 
@@ -1485,6 +1476,9 @@ cifs_parse_smb_version(char *value, struct smb_vol *vol, bool is_smb3)
 			cifs_dbg(VFS, "vers=1.0 (cifs) not permitted when mounting with smb3\n");
 			return 1;
 		}
+		cifs_dbg(VFS, "Use of the less secure dialect vers=1.0 "
+			   "is not recommended unless required for "
+			   "access to very old servers\n");
 		vol->ops = &smb1_operations;
 		vol->vals = &smb1_values;
 		break;
@@ -1672,6 +1666,10 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 
 	vol->echo_interval = SMB_ECHO_INTERVAL_DEFAULT;
 
+	/* default to no multichannel (single server connection) */
+	vol->multichannel = false;
+	vol->max_channels = 1;
+
 	if (!mountdata)
 		goto cifs_parse_mount_err;
 
@@ -1965,6 +1963,12 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 		case Opt_rdma:
 			vol->rdma = true;
 			break;
+		case Opt_multichannel:
+			vol->multichannel = true;
+			break;
+		case Opt_nomultichannel:
+			vol->multichannel = false;
+			break;
 		case Opt_compress:
 			vol->compression = UNKNOWN_TYPE;
 			cifs_dbg(VFS,
@@ -2128,6 +2132,15 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 			}
 			vol->max_credits = option;
 			break;
+		case Opt_max_channels:
+			if (get_option_ul(args, &option) || option < 1 ||
+				option > CIFS_MAX_CHANNELS) {
+				cifs_dbg(VFS, "%s: Invalid max_channels value, needs to be 1-%d\n",
+					 __func__, CIFS_MAX_CHANNELS);
+				goto cifs_parse_mount_err;
+			}
+			vol->max_channels = option;
+			break;
 
 		/* String Arguments */
 
@@ -2461,11 +2474,12 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
 		pr_notice("CIFS: ignoring forcegid mount option specified with no gid= option.\n");
 
 	if (got_version == false)
-		pr_warn("No dialect specified on mount. Default has changed to "
-			"a more secure dialect, SMB2.1 or later (e.g. SMB3), from CIFS "
-			"(SMB1). To use the less secure SMB1 dialect to access "
-			"old servers which do not support SMB3 (or SMB2.1) specify vers=1.0"
-			" on mount.\n");
+		pr_warn_once("No dialect specified on mount. Default has changed"
+			" to a more secure dialect, SMB2.1 or later (e.g. "
+			"SMB3.1.1), from CIFS (SMB1). To use the less secure "
+			"SMB1 dialect to access old servers which do not "
+			"support SMB3.1.1 (or even SMB3 or SMB2.1) specify "
+			"vers=1.0 on mount.\n");
 
 	kfree(mountdata_copy);
 	return 0;
@@ -2656,7 +2670,11 @@ cifs_find_tcp_session(struct smb_vol *vol)
 
 	spin_lock(&cifs_tcp_ses_lock);
 	list_for_each_entry(server, &cifs_tcp_ses_list, tcp_ses_list) {
-		if (!match_server(server, vol))
+		/*
+		 * Skip ses channels since they're only handled in lower layers
+		 * (e.g. cifs_send_recv).
+		 */
+		if (server->is_channel || !match_server(server, vol))
 			continue;
 
 		++server->srv_count;
@@ -2713,7 +2731,7 @@ cifs_put_tcp_session(struct TCP_Server_Info *server, int from_reconnect)
 		send_sig(SIGKILL, task, 1);
 }
 
-static struct TCP_Server_Info *
+struct TCP_Server_Info *
 cifs_get_tcp_session(struct smb_vol *volume_info)
 {
 	struct TCP_Server_Info *tcp_ses = NULL;
@@ -2772,7 +2790,11 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
 	       sizeof(tcp_ses->srcaddr));
 	memcpy(&tcp_ses->dstaddr, &volume_info->dstaddr,
 		sizeof(tcp_ses->dstaddr));
-	generate_random_uuid(tcp_ses->client_guid);
+	if (volume_info->use_client_guid)
+		memcpy(tcp_ses->client_guid, volume_info->client_guid,
+		       SMB2_CLIENT_GUID_SIZE);
+	else
+		generate_random_uuid(tcp_ses->client_guid);
 	/*
 	 * at this point we are the only ones with the pointer
 	 * to the struct since the kernel thread not created yet
@@ -2861,6 +2883,13 @@ static int match_session(struct cifs_ses *ses, struct smb_vol *vol)
 	    vol->sectype != ses->sectype)
 		return 0;
 
+	/*
+	 * If an existing session is limited to less channels than
+	 * requested, it should not be reused
+	 */
+	if (ses->chan_max < vol->max_channels)
+		return 0;
+
 	switch (ses->sectype) {
 	case Kerberos:
 		if (!uid_eq(vol->cred_uid, ses->cred_uid))
@@ -3031,6 +3060,14 @@ void cifs_put_smb_ses(struct cifs_ses *ses)
 	list_del_init(&ses->smb_ses_list);
 	spin_unlock(&cifs_tcp_ses_lock);
 
+	/* close any extra channels */
+	if (ses->chan_count > 1) {
+		int i;
+
+		for (i = 1; i < ses->chan_count; i++)
+			cifs_put_tcp_session(ses->chans[i].server, 0);
+	}
+
 	sesInfoFree(ses);
 	cifs_put_tcp_session(server, 0);
 }
@@ -3277,14 +3314,25 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info)
 	ses->sectype = volume_info->sectype;
 	ses->sign = volume_info->sign;
 	mutex_lock(&ses->session_mutex);
+
+	/* add server as first channel */
+	ses->chans[0].server = server;
+	ses->chan_count = 1;
+	ses->chan_max = volume_info->multichannel ? volume_info->max_channels:1;
+
 	rc = cifs_negotiate_protocol(xid, ses);
 	if (!rc)
 		rc = cifs_setup_session(xid, ses, volume_info->local_nls);
+
+	/* each channel uses a different signing key */
+	memcpy(ses->chans[0].signkey, ses->smb3signingkey,
+	       sizeof(ses->smb3signingkey));
+
 	mutex_unlock(&ses->session_mutex);
 	if (rc)
 		goto get_ses_fail;
 
-	/* success, put it on the list */
+	/* success, put it on the list and add it as first channel */
 	spin_lock(&cifs_tcp_ses_lock);
 	list_add(&ses->smb_ses_list, &server->smb_ses_list);
 	spin_unlock(&cifs_tcp_ses_lock);
@@ -3327,6 +3375,10 @@ cifs_find_tcon(struct cifs_ses *ses, struct smb_vol *volume_info)
 	spin_lock(&cifs_tcp_ses_lock);
 	list_for_each(tmp, &ses->tcon_list) {
 		tcon = list_entry(tmp, struct cifs_tcon, tcon_list);
+#ifdef CONFIG_CIFS_DFS_UPCALL
+		if (tcon->dfs_path)
+			continue;
+#endif
 		if (!match_tcon(tcon, volume_info))
 			continue;
 		++tcon->tc_count;
@@ -3619,8 +3671,10 @@ match_prepath(struct super_block *sb, struct cifs_mnt_data *mnt_data)
 {
 	struct cifs_sb_info *old = CIFS_SB(sb);
 	struct cifs_sb_info *new = mnt_data->cifs_sb;
-	bool old_set = old->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH;
-	bool new_set = new->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH;
+	bool old_set = (old->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) &&
+		old->prepath;
+	bool new_set = (new->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) &&
+		new->prepath;
 
 	if (old_set && new_set && !strcmp(new->prepath, old->prepath))
 		return 1;
@@ -4059,7 +4113,7 @@ int cifs_setup_cifs_sb(struct smb_vol *pvolume_info,
 	cifs_sb->mnt_gid = pvolume_info->linux_gid;
 	cifs_sb->mnt_file_mode = pvolume_info->file_mode;
 	cifs_sb->mnt_dir_mode = pvolume_info->dir_mode;
-	cifs_dbg(FYI, "file mode: 0x%hx  dir mode: 0x%hx\n",
+	cifs_dbg(FYI, "file mode: %04ho  dir mode: %04ho\n",
 		 cifs_sb->mnt_file_mode, cifs_sb->mnt_dir_mode);
 
 	cifs_sb->actimeo = pvolume_info->actimeo;
@@ -4273,7 +4327,7 @@ static int mount_get_conns(struct smb_vol *vol, struct cifs_sb_info *cifs_sb,
 
 	/* do not care if a following call succeed - informational */
 	if (!tcon->pipe && server->ops->qfs_tcon) {
-		server->ops->qfs_tcon(*xid, tcon);
+		server->ops->qfs_tcon(*xid, tcon, cifs_sb);
 		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RO_CACHE) {
 			if (tcon->fsDevInfo.DeviceCharacteristics &
 			    cpu_to_le32(FILE_READ_ONLY_DEVICE))
@@ -4700,6 +4754,17 @@ static int is_path_remote(struct cifs_sb_info *cifs_sb, struct smb_vol *vol,
 }
 
 #ifdef CONFIG_CIFS_DFS_UPCALL
+static inline void set_root_tcon(struct cifs_sb_info *cifs_sb,
+				 struct cifs_tcon *tcon,
+				 struct cifs_tcon **root)
+{
+	spin_lock(&cifs_tcp_ses_lock);
+	tcon->tc_count++;
+	tcon->remap = cifs_remap(cifs_sb);
+	spin_unlock(&cifs_tcp_ses_lock);
+	*root = tcon;
+}
+
 int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb_vol *vol)
 {
 	int rc = 0;
@@ -4801,18 +4866,10 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb_vol *vol)
 	/* Cache out resolved root server */
 	(void)dfs_cache_find(xid, ses, cifs_sb->local_nls, cifs_remap(cifs_sb),
 			     root_path + 1, NULL, NULL);
-	/*
-	 * Save root tcon for additional DFS requests to update or create a new
-	 * DFS cache entry, or even perform DFS failover.
-	 */
-	spin_lock(&cifs_tcp_ses_lock);
-	tcon->tc_count++;
-	tcon->dfs_path = root_path;
+	kfree(root_path);
 	root_path = NULL;
-	tcon->remap = cifs_remap(cifs_sb);
-	spin_unlock(&cifs_tcp_ses_lock);
 
-	root_tcon = tcon;
+	set_root_tcon(cifs_sb, tcon, &root_tcon);
 
 	for (count = 1; ;) {
 		if (!rc && tcon) {
@@ -4849,6 +4906,15 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb_vol *vol)
 			mount_put_conns(cifs_sb, xid, server, ses, tcon);
 			rc = mount_get_conns(vol, cifs_sb, &xid, &server, &ses,
 					     &tcon);
+			/*
+			 * Ensure that DFS referrals go through new root server.
+			 */
+			if (!rc && tcon &&
+			    (tcon->share_flags & (SHI1005_FLAGS_DFS |
+						  SHI1005_FLAGS_DFS_ROOT))) {
+				cifs_put_tcon(root_tcon);
+				set_root_tcon(cifs_sb, tcon, &root_tcon);
+			}
 		}
 		if (rc) {
 			if (rc == -EACCES || rc == -EOPNOTSUPP)
@@ -4895,8 +4961,18 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb_vol *vol)
 	 * dentry revalidation to think the dentry are stale (ESTALE).
 	 */
 	cifs_autodisable_serverino(cifs_sb);
+	/*
+	 * Force the use of prefix path to support failover on DFS paths that
+	 * resolve to targets that have different prefix paths.
+	 */
+	cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH;
+	kfree(cifs_sb->prepath);
+	cifs_sb->prepath = vol->prepath;
+	vol->prepath = NULL;
+
 out:
 	free_xid(xid);
+	cifs_try_adding_channels(ses);
 	return mount_setup_tlink(cifs_sb, ses, tcon);
 
 error:
@@ -5142,7 +5218,7 @@ int
 cifs_negotiate_protocol(const unsigned int xid, struct cifs_ses *ses)
 {
 	int rc = 0;
-	struct TCP_Server_Info *server = ses->server;
+	struct TCP_Server_Info *server = cifs_ses_server(ses);
 
 	if (!server->ops->need_neg || !server->ops->negotiate)
 		return -ENOSYS;
@@ -5169,23 +5245,25 @@ cifs_setup_session(const unsigned int xid, struct cifs_ses *ses,
 		   struct nls_table *nls_info)
 {
 	int rc = -ENOSYS;
-	struct TCP_Server_Info *server = ses->server;
-
-	ses->capabilities = server->capabilities;
-	if (linuxExtEnabled == 0)
-		ses->capabilities &= (~server->vals->cap_unix);
+	struct TCP_Server_Info *server = cifs_ses_server(ses);
+
+	if (!ses->binding) {
+		ses->capabilities = server->capabilities;
+		if (linuxExtEnabled == 0)
+			ses->capabilities &= (~server->vals->cap_unix);
+
+		if (ses->auth_key.response) {
+			cifs_dbg(FYI, "Free previous auth_key.response = %p\n",
+				 ses->auth_key.response);
+			kfree(ses->auth_key.response);
+			ses->auth_key.response = NULL;
+			ses->auth_key.len = 0;
+		}
+	}
 
 	cifs_dbg(FYI, "Security Mode: 0x%x Capabilities: 0x%x TimeAdjust: %d\n",
 		 server->sec_mode, server->capabilities, server->timeAdj);
 
-	if (ses->auth_key.response) {
-		cifs_dbg(FYI, "Free previous auth_key.response = %p\n",
-			 ses->auth_key.response);
-		kfree(ses->auth_key.response);
-		ses->auth_key.response = NULL;
-		ses->auth_key.len = 0;
-	}
-
 	if (server->ops->sess_setup)
 		rc = server->ops->sess_setup(xid, ses, nls_info);
 
diff --git a/fs/cifs/dfs_cache.c b/fs/cifs/dfs_cache.c
index 1692c0c6c23a..a67f88bf7ae1 100644
--- a/fs/cifs/dfs_cache.c
+++ b/fs/cifs/dfs_cache.c
@@ -5,11 +5,10 @@
  * Copyright (c) 2018-2019 Paulo Alcantara <palcantara@suse.de>
  */
 
-#include <linux/rcupdate.h>
-#include <linux/rculist.h>
 #include <linux/jhash.h>
 #include <linux/ktime.h>
 #include <linux/slab.h>
+#include <linux/proc_fs.h>
 #include <linux/nls.h>
 #include <linux/workqueue.h>
 #include "cifsglob.h"
@@ -22,67 +21,68 @@
 
 #include "dfs_cache.h"
 
-#define DFS_CACHE_HTABLE_SIZE 32
-#define DFS_CACHE_MAX_ENTRIES 64
+#define CACHE_HTABLE_SIZE 32
+#define CACHE_MAX_ENTRIES 64
 
 #define IS_INTERLINK_SET(v) ((v) & (DFSREF_REFERRAL_SERVER | \
 				    DFSREF_STORAGE_SERVER))
 
-struct dfs_cache_tgt {
-	char *t_name;
-	struct list_head t_list;
+struct cache_dfs_tgt {
+	char *name;
+	struct list_head list;
 };
 
-struct dfs_cache_entry {
-	struct hlist_node ce_hlist;
-	const char *ce_path;
-	int ce_ttl;
-	int ce_srvtype;
-	int ce_flags;
-	struct timespec64 ce_etime;
-	int ce_path_consumed;
-	int ce_numtgts;
-	struct list_head ce_tlist;
-	struct dfs_cache_tgt *ce_tgthint;
-	struct rcu_head ce_rcu;
+struct cache_entry {
+	struct hlist_node hlist;
+	const char *path;
+	int ttl;
+	int srvtype;
+	int flags;
+	struct timespec64 etime;
+	int path_consumed;
+	int numtgts;
+	struct list_head tlist;
+	struct cache_dfs_tgt *tgthint;
 };
 
-static struct kmem_cache *dfs_cache_slab __read_mostly;
-
-struct dfs_cache_vol_info {
-	char *vi_fullpath;
-	struct smb_vol vi_vol;
-	char *vi_mntdata;
-	struct list_head vi_list;
+struct vol_info {
+	char *fullpath;
+	spinlock_t smb_vol_lock;
+	struct smb_vol smb_vol;
+	char *mntdata;
+	struct list_head list;
+	struct list_head rlist;
+	struct kref refcnt;
 };
 
-struct dfs_cache {
-	struct mutex dc_lock;
-	struct nls_table *dc_nlsc;
-	struct list_head dc_vol_list;
-	int dc_ttl;
-	struct delayed_work dc_refresh;
-};
+static struct kmem_cache *cache_slab __read_mostly;
+static struct workqueue_struct *dfscache_wq __read_mostly;
 
-static struct dfs_cache dfs_cache;
+static int cache_ttl;
+static DEFINE_SPINLOCK(cache_ttl_lock);
+
+static struct nls_table *cache_nlsc;
 
 /*
  * Number of entries in the cache
  */
-static size_t dfs_cache_count;
+static atomic_t cache_count;
+
+static struct hlist_head cache_htable[CACHE_HTABLE_SIZE];
+static DECLARE_RWSEM(htable_rw_lock);
 
-static DEFINE_MUTEX(dfs_cache_list_lock);
-static struct hlist_head dfs_cache_htable[DFS_CACHE_HTABLE_SIZE];
+static LIST_HEAD(vol_list);
+static DEFINE_SPINLOCK(vol_list_lock);
 
 static void refresh_cache_worker(struct work_struct *work);
 
-static inline bool is_path_valid(const char *path)
-{
-	return path && (strchr(path + 1, '\\') || strchr(path + 1, '/'));
-}
+static DECLARE_DELAYED_WORK(refresh_task, refresh_cache_worker);
 
-static inline int get_normalized_path(const char *path, char **npath)
+static int get_normalized_path(const char *path, char **npath)
 {
+	if (!path || strlen(path) < 3 || (*path != '\\' && *path != '/'))
+		return -EINVAL;
+
 	if (*path == '\\') {
 		*npath = (char *)path;
 	} else {
@@ -100,57 +100,48 @@ static inline void free_normalized_path(const char *path, char *npath)
 		kfree(npath);
 }
 
-static inline bool cache_entry_expired(const struct dfs_cache_entry *ce)
+static inline bool cache_entry_expired(const struct cache_entry *ce)
 {
 	struct timespec64 ts;
 
 	ktime_get_coarse_real_ts64(&ts);
-	return timespec64_compare(&ts, &ce->ce_etime) >= 0;
+	return timespec64_compare(&ts, &ce->etime) >= 0;
 }
 
-static inline void free_tgts(struct dfs_cache_entry *ce)
+static inline void free_tgts(struct cache_entry *ce)
 {
-	struct dfs_cache_tgt *t, *n;
+	struct cache_dfs_tgt *t, *n;
 
-	list_for_each_entry_safe(t, n, &ce->ce_tlist, t_list) {
-		list_del(&t->t_list);
-		kfree(t->t_name);
+	list_for_each_entry_safe(t, n, &ce->tlist, list) {
+		list_del(&t->list);
+		kfree(t->name);
 		kfree(t);
 	}
 }
 
-static void free_cache_entry(struct rcu_head *rcu)
+static inline void flush_cache_ent(struct cache_entry *ce)
 {
-	struct dfs_cache_entry *ce = container_of(rcu, struct dfs_cache_entry,
-						  ce_rcu);
-	kmem_cache_free(dfs_cache_slab, ce);
-}
-
-static inline void flush_cache_ent(struct dfs_cache_entry *ce)
-{
-	if (hlist_unhashed(&ce->ce_hlist))
-		return;
-
-	hlist_del_init_rcu(&ce->ce_hlist);
-	kfree_const(ce->ce_path);
+	hlist_del_init(&ce->hlist);
+	kfree(ce->path);
 	free_tgts(ce);
-	dfs_cache_count--;
-	call_rcu(&ce->ce_rcu, free_cache_entry);
+	atomic_dec(&cache_count);
+	kmem_cache_free(cache_slab, ce);
 }
 
 static void flush_cache_ents(void)
 {
 	int i;
 
-	rcu_read_lock();
-	for (i = 0; i < DFS_CACHE_HTABLE_SIZE; i++) {
-		struct hlist_head *l = &dfs_cache_htable[i];
-		struct dfs_cache_entry *ce;
+	for (i = 0; i < CACHE_HTABLE_SIZE; i++) {
+		struct hlist_head *l = &cache_htable[i];
+		struct hlist_node *n;
+		struct cache_entry *ce;
 
-		hlist_for_each_entry_rcu(ce, l, ce_hlist)
-			flush_cache_ent(ce);
+		hlist_for_each_entry_safe(ce, n, l, hlist) {
+			if (!hlist_unhashed(&ce->hlist))
+				flush_cache_ent(ce);
+		}
 	}
-	rcu_read_unlock();
 }
 
 /*
@@ -158,36 +149,39 @@ static void flush_cache_ents(void)
  */
 static int dfscache_proc_show(struct seq_file *m, void *v)
 {
-	int bucket;
-	struct dfs_cache_entry *ce;
-	struct dfs_cache_tgt *t;
+	int i;
+	struct cache_entry *ce;
+	struct cache_dfs_tgt *t;
 
 	seq_puts(m, "DFS cache\n---------\n");
 
-	mutex_lock(&dfs_cache_list_lock);
-
-	rcu_read_lock();
-	hash_for_each_rcu(dfs_cache_htable, bucket, ce, ce_hlist) {
-		seq_printf(m,
-			   "cache entry: path=%s,type=%s,ttl=%d,etime=%ld,"
-			   "interlink=%s,path_consumed=%d,expired=%s\n",
-			   ce->ce_path,
-			   ce->ce_srvtype == DFS_TYPE_ROOT ? "root" : "link",
-			   ce->ce_ttl, ce->ce_etime.tv_nsec,
-			   IS_INTERLINK_SET(ce->ce_flags) ? "yes" : "no",
-			   ce->ce_path_consumed,
-			   cache_entry_expired(ce) ? "yes" : "no");
-
-		list_for_each_entry(t, &ce->ce_tlist, t_list) {
-			seq_printf(m, "  %s%s\n",
-				   t->t_name,
-				   ce->ce_tgthint == t ? " (target hint)" : "");
+	down_read(&htable_rw_lock);
+	for (i = 0; i < CACHE_HTABLE_SIZE; i++) {
+		struct hlist_head *l = &cache_htable[i];
+
+		hlist_for_each_entry(ce, l, hlist) {
+			if (hlist_unhashed(&ce->hlist))
+				continue;
+
+			seq_printf(m,
+				   "cache entry: path=%s,type=%s,ttl=%d,etime=%ld,"
+				   "interlink=%s,path_consumed=%d,expired=%s\n",
+				   ce->path,
+				   ce->srvtype == DFS_TYPE_ROOT ? "root" : "link",
+				   ce->ttl, ce->etime.tv_nsec,
+				   IS_INTERLINK_SET(ce->flags) ? "yes" : "no",
+				   ce->path_consumed,
+				   cache_entry_expired(ce) ? "yes" : "no");
+
+			list_for_each_entry(t, &ce->tlist, list) {
+				seq_printf(m, "  %s%s\n",
+					   t->name,
+					   ce->tgthint == t ? " (target hint)" : "");
+			}
 		}
-
 	}
-	rcu_read_unlock();
+	up_read(&htable_rw_lock);
 
-	mutex_unlock(&dfs_cache_list_lock);
 	return 0;
 }
 
@@ -205,9 +199,10 @@ static ssize_t dfscache_proc_write(struct file *file, const char __user *buffer,
 		return -EINVAL;
 
 	cifs_dbg(FYI, "clearing dfs cache");
-	mutex_lock(&dfs_cache_list_lock);
+
+	down_write(&htable_rw_lock);
 	flush_cache_ents();
-	mutex_unlock(&dfs_cache_list_lock);
+	up_write(&htable_rw_lock);
 
 	return count;
 }
@@ -217,34 +212,34 @@ static int dfscache_proc_open(struct inode *inode, struct file *file)
 	return single_open(file, dfscache_proc_show, NULL);
 }
 
-const struct file_operations dfscache_proc_fops = {
-	.open		= dfscache_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-	.write		= dfscache_proc_write,
+const struct proc_ops dfscache_proc_ops = {
+	.proc_open	= dfscache_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= dfscache_proc_write,
 };
 
 #ifdef CONFIG_CIFS_DEBUG2
-static inline void dump_tgts(const struct dfs_cache_entry *ce)
+static inline void dump_tgts(const struct cache_entry *ce)
 {
-	struct dfs_cache_tgt *t;
+	struct cache_dfs_tgt *t;
 
 	cifs_dbg(FYI, "target list:\n");
-	list_for_each_entry(t, &ce->ce_tlist, t_list) {
-		cifs_dbg(FYI, "  %s%s\n", t->t_name,
-			 ce->ce_tgthint == t ? " (target hint)" : "");
+	list_for_each_entry(t, &ce->tlist, list) {
+		cifs_dbg(FYI, "  %s%s\n", t->name,
+			 ce->tgthint == t ? " (target hint)" : "");
 	}
 }
 
-static inline void dump_ce(const struct dfs_cache_entry *ce)
+static inline void dump_ce(const struct cache_entry *ce)
 {
 	cifs_dbg(FYI, "cache entry: path=%s,type=%s,ttl=%d,etime=%ld,"
-		 "interlink=%s,path_consumed=%d,expired=%s\n", ce->ce_path,
-		 ce->ce_srvtype == DFS_TYPE_ROOT ? "root" : "link", ce->ce_ttl,
-		 ce->ce_etime.tv_nsec,
-		 IS_INTERLINK_SET(ce->ce_flags) ? "yes" : "no",
-		 ce->ce_path_consumed,
+		 "interlink=%s,path_consumed=%d,expired=%s\n", ce->path,
+		 ce->srvtype == DFS_TYPE_ROOT ? "root" : "link", ce->ttl,
+		 ce->etime.tv_nsec,
+		 IS_INTERLINK_SET(ce->flags) ? "yes" : "no",
+		 ce->path_consumed,
 		 cache_entry_expired(ce) ? "yes" : "no");
 	dump_tgts(ce);
 }
@@ -284,25 +279,34 @@ static inline void dump_refs(const struct dfs_info3_param *refs, int numrefs)
  */
 int dfs_cache_init(void)
 {
+	int rc;
 	int i;
 
-	dfs_cache_slab = kmem_cache_create("cifs_dfs_cache",
-					   sizeof(struct dfs_cache_entry), 0,
-					   SLAB_HWCACHE_ALIGN, NULL);
-	if (!dfs_cache_slab)
+	dfscache_wq = alloc_workqueue("cifs-dfscache",
+				      WQ_FREEZABLE | WQ_MEM_RECLAIM, 1);
+	if (!dfscache_wq)
 		return -ENOMEM;
 
-	for (i = 0; i < DFS_CACHE_HTABLE_SIZE; i++)
-		INIT_HLIST_HEAD(&dfs_cache_htable[i]);
+	cache_slab = kmem_cache_create("cifs_dfs_cache",
+				       sizeof(struct cache_entry), 0,
+				       SLAB_HWCACHE_ALIGN, NULL);
+	if (!cache_slab) {
+		rc = -ENOMEM;
+		goto out_destroy_wq;
+	}
+
+	for (i = 0; i < CACHE_HTABLE_SIZE; i++)
+		INIT_HLIST_HEAD(&cache_htable[i]);
 
-	INIT_LIST_HEAD(&dfs_cache.dc_vol_list);
-	mutex_init(&dfs_cache.dc_lock);
-	INIT_DELAYED_WORK(&dfs_cache.dc_refresh, refresh_cache_worker);
-	dfs_cache.dc_ttl = -1;
-	dfs_cache.dc_nlsc = load_nls_default();
+	atomic_set(&cache_count, 0);
+	cache_nlsc = load_nls_default();
 
 	cifs_dbg(FYI, "%s: initialized DFS referral cache\n", __func__);
 	return 0;
+
+out_destroy_wq:
+	destroy_workqueue(dfscache_wq);
+	return rc;
 }
 
 static inline unsigned int cache_entry_hash(const void *data, int size)
@@ -310,7 +314,7 @@ static inline unsigned int cache_entry_hash(const void *data, int size)
 	unsigned int h;
 
 	h = jhash(data, size, 0);
-	return h & (DFS_CACHE_HTABLE_SIZE - 1);
+	return h & (CACHE_HTABLE_SIZE - 1);
 }
 
 /* Check whether second path component of @path is SYSVOL or NETLOGON */
@@ -325,11 +329,11 @@ static inline bool is_sysvol_or_netlogon(const char *path)
 }
 
 /* Return target hint of a DFS cache entry */
-static inline char *get_tgt_name(const struct dfs_cache_entry *ce)
+static inline char *get_tgt_name(const struct cache_entry *ce)
 {
-	struct dfs_cache_tgt *t = ce->ce_tgthint;
+	struct cache_dfs_tgt *t = ce->tgthint;
 
-	return t ? t->t_name : ERR_PTR(-ENOENT);
+	return t ? t->name : ERR_PTR(-ENOENT);
 }
 
 /* Return expire time out of a new entry's TTL */
@@ -346,19 +350,19 @@ static inline struct timespec64 get_expire_time(int ttl)
 }
 
 /* Allocate a new DFS target */
-static inline struct dfs_cache_tgt *alloc_tgt(const char *name)
+static struct cache_dfs_tgt *alloc_target(const char *name)
 {
-	struct dfs_cache_tgt *t;
+	struct cache_dfs_tgt *t;
 
-	t = kmalloc(sizeof(*t), GFP_KERNEL);
+	t = kmalloc(sizeof(*t), GFP_ATOMIC);
 	if (!t)
 		return ERR_PTR(-ENOMEM);
-	t->t_name = kstrndup(name, strlen(name), GFP_KERNEL);
-	if (!t->t_name) {
+	t->name = kstrndup(name, strlen(name), GFP_ATOMIC);
+	if (!t->name) {
 		kfree(t);
 		return ERR_PTR(-ENOMEM);
 	}
-	INIT_LIST_HEAD(&t->t_list);
+	INIT_LIST_HEAD(&t->list);
 	return t;
 }
 
@@ -367,180 +371,184 @@ static inline struct dfs_cache_tgt *alloc_tgt(const char *name)
  * target hint.
  */
 static int copy_ref_data(const struct dfs_info3_param *refs, int numrefs,
-			 struct dfs_cache_entry *ce, const char *tgthint)
+			 struct cache_entry *ce, const char *tgthint)
 {
 	int i;
 
-	ce->ce_ttl = refs[0].ttl;
-	ce->ce_etime = get_expire_time(ce->ce_ttl);
-	ce->ce_srvtype = refs[0].server_type;
-	ce->ce_flags = refs[0].ref_flag;
-	ce->ce_path_consumed = refs[0].path_consumed;
+	ce->ttl = refs[0].ttl;
+	ce->etime = get_expire_time(ce->ttl);
+	ce->srvtype = refs[0].server_type;
+	ce->flags = refs[0].ref_flag;
+	ce->path_consumed = refs[0].path_consumed;
 
 	for (i = 0; i < numrefs; i++) {
-		struct dfs_cache_tgt *t;
+		struct cache_dfs_tgt *t;
 
-		t = alloc_tgt(refs[i].node_name);
+		t = alloc_target(refs[i].node_name);
 		if (IS_ERR(t)) {
 			free_tgts(ce);
 			return PTR_ERR(t);
 		}
-		if (tgthint && !strcasecmp(t->t_name, tgthint)) {
-			list_add(&t->t_list, &ce->ce_tlist);
+		if (tgthint && !strcasecmp(t->name, tgthint)) {
+			list_add(&t->list, &ce->tlist);
 			tgthint = NULL;
 		} else {
-			list_add_tail(&t->t_list, &ce->ce_tlist);
+			list_add_tail(&t->list, &ce->tlist);
 		}
-		ce->ce_numtgts++;
+		ce->numtgts++;
 	}
 
-	ce->ce_tgthint = list_first_entry_or_null(&ce->ce_tlist,
-						  struct dfs_cache_tgt, t_list);
+	ce->tgthint = list_first_entry_or_null(&ce->tlist,
+					       struct cache_dfs_tgt, list);
 
 	return 0;
 }
 
 /* Allocate a new cache entry */
-static struct dfs_cache_entry *
-alloc_cache_entry(const char *path, const struct dfs_info3_param *refs,
-		  int numrefs)
+static struct cache_entry *alloc_cache_entry(const char *path,
+					     const struct dfs_info3_param *refs,
+					     int numrefs)
 {
-	struct dfs_cache_entry *ce;
+	struct cache_entry *ce;
 	int rc;
 
-	ce = kmem_cache_zalloc(dfs_cache_slab, GFP_KERNEL);
+	ce = kmem_cache_zalloc(cache_slab, GFP_KERNEL);
 	if (!ce)
 		return ERR_PTR(-ENOMEM);
 
-	ce->ce_path = kstrdup_const(path, GFP_KERNEL);
-	if (!ce->ce_path) {
-		kmem_cache_free(dfs_cache_slab, ce);
+	ce->path = kstrndup(path, strlen(path), GFP_KERNEL);
+	if (!ce->path) {
+		kmem_cache_free(cache_slab, ce);
 		return ERR_PTR(-ENOMEM);
 	}
-	INIT_HLIST_NODE(&ce->ce_hlist);
-	INIT_LIST_HEAD(&ce->ce_tlist);
+	INIT_HLIST_NODE(&ce->hlist);
+	INIT_LIST_HEAD(&ce->tlist);
 
 	rc = copy_ref_data(refs, numrefs, ce, NULL);
 	if (rc) {
-		kfree_const(ce->ce_path);
-		kmem_cache_free(dfs_cache_slab, ce);
+		kfree(ce->path);
+		kmem_cache_free(cache_slab, ce);
 		ce = ERR_PTR(rc);
 	}
 	return ce;
 }
 
+/* Must be called with htable_rw_lock held */
 static void remove_oldest_entry(void)
 {
-	int bucket;
-	struct dfs_cache_entry *ce;
-	struct dfs_cache_entry *to_del = NULL;
-
-	rcu_read_lock();
-	hash_for_each_rcu(dfs_cache_htable, bucket, ce, ce_hlist) {
-		if (!to_del || timespec64_compare(&ce->ce_etime,
-						  &to_del->ce_etime) < 0)
-			to_del = ce;
+	int i;
+	struct cache_entry *ce;
+	struct cache_entry *to_del = NULL;
+
+	for (i = 0; i < CACHE_HTABLE_SIZE; i++) {
+		struct hlist_head *l = &cache_htable[i];
+
+		hlist_for_each_entry(ce, l, hlist) {
+			if (hlist_unhashed(&ce->hlist))
+				continue;
+			if (!to_del || timespec64_compare(&ce->etime,
+							  &to_del->etime) < 0)
+				to_del = ce;
+		}
 	}
+
 	if (!to_del) {
 		cifs_dbg(FYI, "%s: no entry to remove", __func__);
-		goto out;
+		return;
 	}
+
 	cifs_dbg(FYI, "%s: removing entry", __func__);
 	dump_ce(to_del);
 	flush_cache_ent(to_del);
-out:
-	rcu_read_unlock();
 }
 
 /* Add a new DFS cache entry */
-static inline struct dfs_cache_entry *
-add_cache_entry(unsigned int hash, const char *path,
-		const struct dfs_info3_param *refs, int numrefs)
+static int add_cache_entry(const char *path, unsigned int hash,
+			   struct dfs_info3_param *refs, int numrefs)
 {
-	struct dfs_cache_entry *ce;
+	struct cache_entry *ce;
 
 	ce = alloc_cache_entry(path, refs, numrefs);
 	if (IS_ERR(ce))
-		return ce;
-
-	hlist_add_head_rcu(&ce->ce_hlist, &dfs_cache_htable[hash]);
+		return PTR_ERR(ce);
 
-	mutex_lock(&dfs_cache.dc_lock);
-	if (dfs_cache.dc_ttl < 0) {
-		dfs_cache.dc_ttl = ce->ce_ttl;
-		queue_delayed_work(cifsiod_wq, &dfs_cache.dc_refresh,
-				   dfs_cache.dc_ttl * HZ);
+	spin_lock(&cache_ttl_lock);
+	if (!cache_ttl) {
+		cache_ttl = ce->ttl;
+		queue_delayed_work(dfscache_wq, &refresh_task, cache_ttl * HZ);
 	} else {
-		dfs_cache.dc_ttl = min_t(int, dfs_cache.dc_ttl, ce->ce_ttl);
-		mod_delayed_work(cifsiod_wq, &dfs_cache.dc_refresh,
-				 dfs_cache.dc_ttl * HZ);
+		cache_ttl = min_t(int, cache_ttl, ce->ttl);
+		mod_delayed_work(dfscache_wq, &refresh_task, cache_ttl * HZ);
 	}
-	mutex_unlock(&dfs_cache.dc_lock);
+	spin_unlock(&cache_ttl_lock);
 
-	return ce;
+	down_write(&htable_rw_lock);
+	hlist_add_head(&ce->hlist, &cache_htable[hash]);
+	dump_ce(ce);
+	up_write(&htable_rw_lock);
+
+	return 0;
 }
 
-static struct dfs_cache_entry *__find_cache_entry(unsigned int hash,
-						  const char *path)
+/*
+ * Find a DFS cache entry in hash table and optionally check prefix path against
+ * @path.
+ * Use whole path components in the match.
+ * Must be called with htable_rw_lock held.
+ *
+ * Return ERR_PTR(-ENOENT) if the entry is not found.
+ */
+static struct cache_entry *lookup_cache_entry(const char *path,
+					      unsigned int *hash)
 {
-	struct dfs_cache_entry *ce;
+	struct cache_entry *ce;
+	unsigned int h;
 	bool found = false;
 
-	rcu_read_lock();
-	hlist_for_each_entry_rcu(ce, &dfs_cache_htable[hash], ce_hlist) {
-		if (!strcasecmp(path, ce->ce_path)) {
-#ifdef CONFIG_CIFS_DEBUG2
-			char *name = get_tgt_name(ce);
+	h = cache_entry_hash(path, strlen(path));
 
-			if (IS_ERR(name)) {
-				rcu_read_unlock();
-				return ERR_CAST(name);
-			}
-			cifs_dbg(FYI, "%s: cache hit\n", __func__);
-			cifs_dbg(FYI, "%s: target hint: %s\n", __func__, name);
-#endif
+	hlist_for_each_entry(ce, &cache_htable[h], hlist) {
+		if (!strcasecmp(path, ce->path)) {
 			found = true;
+			dump_ce(ce);
 			break;
 		}
 	}
-	rcu_read_unlock();
-	return found ? ce : ERR_PTR(-ENOENT);
-}
 
-/*
- * Find a DFS cache entry in hash table and optionally check prefix path against
- * @path.
- * Use whole path components in the match.
- * Return ERR_PTR(-ENOENT) if the entry is not found.
- */
-static inline struct dfs_cache_entry *find_cache_entry(const char *path,
-						       unsigned int *hash)
-{
-	*hash = cache_entry_hash(path, strlen(path));
-	return __find_cache_entry(*hash, path);
+	if (!found)
+		ce = ERR_PTR(-ENOENT);
+	if (hash)
+		*hash = h;
+
+	return ce;
 }
 
-static inline void destroy_slab_cache(void)
+static void __vol_release(struct vol_info *vi)
 {
-	rcu_barrier();
-	kmem_cache_destroy(dfs_cache_slab);
+	kfree(vi->fullpath);
+	kfree(vi->mntdata);
+	cifs_cleanup_volume_info_contents(&vi->smb_vol);
+	kfree(vi);
 }
 
-static inline void free_vol(struct dfs_cache_vol_info *vi)
+static void vol_release(struct kref *kref)
 {
-	list_del(&vi->vi_list);
-	kfree(vi->vi_fullpath);
-	kfree(vi->vi_mntdata);
-	cifs_cleanup_volume_info_contents(&vi->vi_vol);
-	kfree(vi);
+	struct vol_info *vi = container_of(kref, struct vol_info, refcnt);
+
+	spin_lock(&vol_list_lock);
+	list_del(&vi->list);
+	spin_unlock(&vol_list_lock);
+	__vol_release(vi);
 }
 
 static inline void free_vol_list(void)
 {
-	struct dfs_cache_vol_info *vi, *nvi;
+	struct vol_info *vi, *nvi;
 
-	list_for_each_entry_safe(vi, nvi, &dfs_cache.dc_vol_list, vi_list)
-		free_vol(vi);
+	list_for_each_entry_safe(vi, nvi, &vol_list, list) {
+		list_del_init(&vi->list);
+		__vol_release(vi);
+	}
 }
 
 /**
@@ -548,83 +556,78 @@ static inline void free_vol_list(void)
  */
 void dfs_cache_destroy(void)
 {
-	cancel_delayed_work_sync(&dfs_cache.dc_refresh);
-	unload_nls(dfs_cache.dc_nlsc);
+	cancel_delayed_work_sync(&refresh_task);
+	unload_nls(cache_nlsc);
 	free_vol_list();
-	mutex_destroy(&dfs_cache.dc_lock);
-
 	flush_cache_ents();
-	destroy_slab_cache();
-	mutex_destroy(&dfs_cache_list_lock);
+	kmem_cache_destroy(cache_slab);
+	destroy_workqueue(dfscache_wq);
 
 	cifs_dbg(FYI, "%s: destroyed DFS referral cache\n", __func__);
 }
 
-static inline struct dfs_cache_entry *
-__update_cache_entry(const char *path, const struct dfs_info3_param *refs,
-		     int numrefs)
+/* Must be called with htable_rw_lock held */
+static int __update_cache_entry(const char *path,
+				const struct dfs_info3_param *refs,
+				int numrefs)
 {
 	int rc;
-	unsigned int h;
-	struct dfs_cache_entry *ce;
+	struct cache_entry *ce;
 	char *s, *th = NULL;
 
-	ce = find_cache_entry(path, &h);
+	ce = lookup_cache_entry(path, NULL);
 	if (IS_ERR(ce))
-		return ce;
+		return PTR_ERR(ce);
 
-	if (ce->ce_tgthint) {
-		s = ce->ce_tgthint->t_name;
-		th = kstrndup(s, strlen(s), GFP_KERNEL);
+	if (ce->tgthint) {
+		s = ce->tgthint->name;
+		th = kstrndup(s, strlen(s), GFP_ATOMIC);
 		if (!th)
-			return ERR_PTR(-ENOMEM);
+			return -ENOMEM;
 	}
 
 	free_tgts(ce);
-	ce->ce_numtgts = 0;
+	ce->numtgts = 0;
 
 	rc = copy_ref_data(refs, numrefs, ce, th);
-	kfree(th);
 
-	if (rc)
-		ce = ERR_PTR(rc);
+	kfree(th);
 
-	return ce;
+	return rc;
 }
 
-/* Update an expired cache entry by getting a new DFS referral from server */
-static struct dfs_cache_entry *
-update_cache_entry(const unsigned int xid, struct cifs_ses *ses,
-		   const struct nls_table *nls_codepage, int remap,
-		   const char *path, struct dfs_cache_entry *ce)
+static int get_dfs_referral(const unsigned int xid, struct cifs_ses *ses,
+			    const struct nls_table *nls_codepage, int remap,
+			    const char *path,  struct dfs_info3_param **refs,
+			    int *numrefs)
 {
-	int rc;
-	struct dfs_info3_param *refs = NULL;
-	int numrefs = 0;
+	cifs_dbg(FYI, "%s: get an DFS referral for %s\n", __func__, path);
 
-	cifs_dbg(FYI, "%s: update expired cache entry\n", __func__);
-	/*
-	 * Check if caller provided enough parameters to update an expired
-	 * entry.
-	 */
 	if (!ses || !ses->server || !ses->server->ops->get_dfs_refer)
-		return ERR_PTR(-ETIME);
+		return -EOPNOTSUPP;
 	if (unlikely(!nls_codepage))
-		return ERR_PTR(-ETIME);
+		return -EINVAL;
 
-	cifs_dbg(FYI, "%s: DFS referral request for %s\n", __func__, path);
+	*refs = NULL;
+	*numrefs = 0;
 
-	rc = ses->server->ops->get_dfs_refer(xid, ses, path, &refs, &numrefs,
-					     nls_codepage, remap);
-	if (rc)
-		ce = ERR_PTR(rc);
-	else
-		ce = __update_cache_entry(path, refs, numrefs);
+	return ses->server->ops->get_dfs_refer(xid, ses, path, refs, numrefs,
+					       nls_codepage, remap);
+}
 
-	dump_refs(refs, numrefs);
-	free_dfs_info_array(refs, numrefs);
+/* Update an expired cache entry by getting a new DFS referral from server */
+static int update_cache_entry(const char *path,
+			      const struct dfs_info3_param *refs,
+			      int numrefs)
+{
 
-	return ce;
+	int rc;
+
+	down_write(&htable_rw_lock);
+	rc = __update_cache_entry(path, refs, numrefs);
+	up_write(&htable_rw_lock);
+
+	return rc;
 }
 
 /*
@@ -636,95 +639,86 @@ update_cache_entry(const unsigned int xid, struct cifs_ses *ses,
  * For interlinks, __cifs_dfs_mount() and expand_dfs_referral() are supposed to
  * handle them properly.
  */
-static struct dfs_cache_entry *
-do_dfs_cache_find(const unsigned int xid, struct cifs_ses *ses,
-		  const struct nls_table *nls_codepage, int remap,
-		  const char *path, bool noreq)
+static int __dfs_cache_find(const unsigned int xid, struct cifs_ses *ses,
+			    const struct nls_table *nls_codepage, int remap,
+			    const char *path, bool noreq)
 {
 	int rc;
-	unsigned int h;
-	struct dfs_cache_entry *ce;
-	struct dfs_info3_param *nrefs;
-	int numnrefs;
+	unsigned int hash;
+	struct cache_entry *ce;
+	struct dfs_info3_param *refs = NULL;
+	int numrefs = 0;
+	bool newent = false;
 
 	cifs_dbg(FYI, "%s: search path: %s\n", __func__, path);
 
-	ce = find_cache_entry(path, &h);
-	if (IS_ERR(ce)) {
-		cifs_dbg(FYI, "%s: cache miss\n", __func__);
-		/*
-		 * If @noreq is set, no requests will be sent to the server for
-		 * either updating or getting a new DFS referral.
-		 */
-		if (noreq)
-			return ce;
-		/*
-		 * No cache entry was found, so check for valid parameters that
-		 * will be required to get a new DFS referral and then create a
-		 * new cache entry.
-		 */
-		if (!ses || !ses->server || !ses->server->ops->get_dfs_refer) {
-			ce = ERR_PTR(-EOPNOTSUPP);
-			return ce;
-		}
-		if (unlikely(!nls_codepage)) {
-			ce = ERR_PTR(-EINVAL);
-			return ce;
-		}
+	down_read(&htable_rw_lock);
 
-		nrefs = NULL;
-		numnrefs = 0;
+	ce = lookup_cache_entry(path, &hash);
 
-		cifs_dbg(FYI, "%s: DFS referral request for %s\n", __func__,
-			 path);
+	/*
+	 * If @noreq is set, no requests will be sent to the server. Just return
+	 * the cache entry.
+	 */
+	if (noreq) {
+		up_read(&htable_rw_lock);
+		return PTR_ERR_OR_ZERO(ce);
+	}
 
-		rc = ses->server->ops->get_dfs_refer(xid, ses, path, &nrefs,
-						     &numnrefs, nls_codepage,
-						     remap);
-		if (rc) {
-			ce = ERR_PTR(rc);
-			return ce;
+	if (!IS_ERR(ce)) {
+		if (!cache_entry_expired(ce)) {
+			dump_ce(ce);
+			up_read(&htable_rw_lock);
+			return 0;
 		}
+	} else {
+		newent = true;
+	}
 
-		dump_refs(nrefs, numnrefs);
+	up_read(&htable_rw_lock);
 
-		cifs_dbg(FYI, "%s: new cache entry\n", __func__);
+	/*
+	 * No entry was found.
+	 *
+	 * Request a new DFS referral in order to create a new cache entry, or
+	 * updating an existing one.
+	 */
+	rc = get_dfs_referral(xid, ses, nls_codepage, remap, path,
+			      &refs, &numrefs);
+	if (rc)
+		return rc;
 
-		if (dfs_cache_count >= DFS_CACHE_MAX_ENTRIES) {
-			cifs_dbg(FYI, "%s: reached max cache size (%d)",
-				 __func__, DFS_CACHE_MAX_ENTRIES);
-			remove_oldest_entry();
-		}
-		ce = add_cache_entry(h, path, nrefs, numnrefs);
-		free_dfs_info_array(nrefs, numnrefs);
+	dump_refs(refs, numrefs);
 
-		if (IS_ERR(ce))
-			return ce;
+	if (!newent) {
+		rc = update_cache_entry(path, refs, numrefs);
+		goto out_free_refs;
+	}
 
-		dfs_cache_count++;
+	if (atomic_read(&cache_count) >= CACHE_MAX_ENTRIES) {
+		cifs_dbg(FYI, "%s: reached max cache size (%d)", __func__,
+			 CACHE_MAX_ENTRIES);
+		down_write(&htable_rw_lock);
+		remove_oldest_entry();
+		up_write(&htable_rw_lock);
 	}
 
-	dump_ce(ce);
+	rc = add_cache_entry(path, hash, refs, numrefs);
+	if (!rc)
+		atomic_inc(&cache_count);
 
-	/* Just return the found cache entry in case @noreq is set */
-	if (noreq)
-		return ce;
-
-	if (cache_entry_expired(ce)) {
-		cifs_dbg(FYI, "%s: expired cache entry\n", __func__);
-		ce = update_cache_entry(xid, ses, nls_codepage, remap, path,
-					ce);
-		if (IS_ERR(ce)) {
-			cifs_dbg(FYI, "%s: failed to update expired entry\n",
-				 __func__);
-		}
-	}
-	return ce;
+out_free_refs:
+	free_dfs_info_array(refs, numrefs);
+	return rc;
 }
 
-/* Set up a new DFS referral from a given cache entry */
-static int setup_ref(const char *path, const struct dfs_cache_entry *ce,
-		     struct dfs_info3_param *ref, const char *tgt)
+/*
+ * Set up a DFS referral from a given cache entry.
+ *
+ * Must be called with htable_rw_lock held.
+ */
+static int setup_referral(const char *path, struct cache_entry *ce,
+			  struct dfs_info3_param *ref, const char *target)
 {
 	int rc;
 
@@ -732,21 +726,20 @@ static int setup_ref(const char *path, const struct dfs_cache_entry *ce,
 
 	memset(ref, 0, sizeof(*ref));
 
-	ref->path_name = kstrndup(path, strlen(path), GFP_KERNEL);
+	ref->path_name = kstrndup(path, strlen(path), GFP_ATOMIC);
 	if (!ref->path_name)
 		return -ENOMEM;
 
-	ref->path_consumed = ce->ce_path_consumed;
-
-	ref->node_name = kstrndup(tgt, strlen(tgt), GFP_KERNEL);
+	ref->node_name = kstrndup(target, strlen(target), GFP_ATOMIC);
 	if (!ref->node_name) {
 		rc = -ENOMEM;
 		goto err_free_path;
 	}
 
-	ref->ttl = ce->ce_ttl;
-	ref->server_type = ce->ce_srvtype;
-	ref->ref_flag = ce->ce_flags;
+	ref->path_consumed = ce->path_consumed;
+	ref->ttl = ce->ttl;
+	ref->server_type = ce->srvtype;
+	ref->ref_flag = ce->flags;
 
 	return 0;
 
@@ -757,38 +750,37 @@ err_free_path:
 }
 
 /* Return target list of a DFS cache entry */
-static int get_tgt_list(const struct dfs_cache_entry *ce,
-			struct dfs_cache_tgt_list *tl)
+static int get_targets(struct cache_entry *ce, struct dfs_cache_tgt_list *tl)
 {
 	int rc;
 	struct list_head *head = &tl->tl_list;
-	struct dfs_cache_tgt *t;
+	struct cache_dfs_tgt *t;
 	struct dfs_cache_tgt_iterator *it, *nit;
 
 	memset(tl, 0, sizeof(*tl));
 	INIT_LIST_HEAD(head);
 
-	list_for_each_entry(t, &ce->ce_tlist, t_list) {
-		it = kzalloc(sizeof(*it), GFP_KERNEL);
+	list_for_each_entry(t, &ce->tlist, list) {
+		it = kzalloc(sizeof(*it), GFP_ATOMIC);
 		if (!it) {
 			rc = -ENOMEM;
 			goto err_free_it;
 		}
 
-		it->it_name = kstrndup(t->t_name, strlen(t->t_name),
-				       GFP_KERNEL);
+		it->it_name = kstrndup(t->name, strlen(t->name), GFP_ATOMIC);
 		if (!it->it_name) {
 			kfree(it);
 			rc = -ENOMEM;
 			goto err_free_it;
 		}
 
-		if (ce->ce_tgthint == t)
+		if (ce->tgthint == t)
 			list_add(&it->it_list, head);
 		else
 			list_add_tail(&it->it_list, head);
 	}
-	tl->tl_numtgts = ce->ce_numtgts;
+
+	tl->tl_numtgts = ce->numtgts;
 
 	return 0;
 
@@ -829,28 +821,35 @@ int dfs_cache_find(const unsigned int xid, struct cifs_ses *ses,
 {
 	int rc;
 	char *npath;
-	struct dfs_cache_entry *ce;
-
-	if (unlikely(!is_path_valid(path)))
-		return -EINVAL;
+	struct cache_entry *ce;
 
 	rc = get_normalized_path(path, &npath);
 	if (rc)
 		return rc;
 
-	mutex_lock(&dfs_cache_list_lock);
-	ce = do_dfs_cache_find(xid, ses, nls_codepage, remap, npath, false);
-	if (!IS_ERR(ce)) {
-		if (ref)
-			rc = setup_ref(path, ce, ref, get_tgt_name(ce));
-		else
-			rc = 0;
-		if (!rc && tgt_list)
-			rc = get_tgt_list(ce, tgt_list);
-	} else {
+	rc = __dfs_cache_find(xid, ses, nls_codepage, remap, npath, false);
+	if (rc)
+		goto out_free_path;
+
+	down_read(&htable_rw_lock);
+
+	ce = lookup_cache_entry(npath, NULL);
+	if (IS_ERR(ce)) {
+		up_read(&htable_rw_lock);
 		rc = PTR_ERR(ce);
+		goto out_free_path;
 	}
-	mutex_unlock(&dfs_cache_list_lock);
+
+	if (ref)
+		rc = setup_referral(path, ce, ref, get_tgt_name(ce));
+	else
+		rc = 0;
+	if (!rc && tgt_list)
+		rc = get_targets(ce, tgt_list);
+
+	up_read(&htable_rw_lock);
+
+out_free_path:
 	free_normalized_path(path, npath);
 	return rc;
 }
@@ -876,31 +875,33 @@ int dfs_cache_noreq_find(const char *path, struct dfs_info3_param *ref,
 {
 	int rc;
 	char *npath;
-	struct dfs_cache_entry *ce;
-
-	if (unlikely(!is_path_valid(path)))
-		return -EINVAL;
+	struct cache_entry *ce;
 
 	rc = get_normalized_path(path, &npath);
 	if (rc)
 		return rc;
 
-	mutex_lock(&dfs_cache_list_lock);
-	ce = do_dfs_cache_find(0, NULL, NULL, 0, npath, true);
+	cifs_dbg(FYI, "%s: path: %s\n", __func__, npath);
+
+	down_read(&htable_rw_lock);
+
+	ce = lookup_cache_entry(npath, NULL);
 	if (IS_ERR(ce)) {
 		rc = PTR_ERR(ce);
-		goto out;
+		goto out_unlock;
 	}
 
 	if (ref)
-		rc = setup_ref(path, ce, ref, get_tgt_name(ce));
+		rc = setup_referral(path, ce, ref, get_tgt_name(ce));
 	else
 		rc = 0;
 	if (!rc && tgt_list)
-		rc = get_tgt_list(ce, tgt_list);
-out:
-	mutex_unlock(&dfs_cache_list_lock);
+		rc = get_targets(ce, tgt_list);
+
+out_unlock:
+	up_read(&htable_rw_lock);
 	free_normalized_path(path, npath);
+
 	return rc;
 }
 
@@ -929,44 +930,46 @@ int dfs_cache_update_tgthint(const unsigned int xid, struct cifs_ses *ses,
 {
 	int rc;
 	char *npath;
-	struct dfs_cache_entry *ce;
-	struct dfs_cache_tgt *t;
-
-	if (unlikely(!is_path_valid(path)))
-		return -EINVAL;
+	struct cache_entry *ce;
+	struct cache_dfs_tgt *t;
 
 	rc = get_normalized_path(path, &npath);
 	if (rc)
 		return rc;
 
-	cifs_dbg(FYI, "%s: path: %s\n", __func__, npath);
+	cifs_dbg(FYI, "%s: update target hint - path: %s\n", __func__, npath);
 
-	mutex_lock(&dfs_cache_list_lock);
-	ce = do_dfs_cache_find(xid, ses, nls_codepage, remap, npath, false);
+	rc = __dfs_cache_find(xid, ses, nls_codepage, remap, npath, false);
+	if (rc)
+		goto out_free_path;
+
+	down_write(&htable_rw_lock);
+
+	ce = lookup_cache_entry(npath, NULL);
 	if (IS_ERR(ce)) {
 		rc = PTR_ERR(ce);
-		goto out;
+		goto out_unlock;
 	}
 
-	rc = 0;
-
-	t = ce->ce_tgthint;
+	t = ce->tgthint;
 
-	if (likely(!strcasecmp(it->it_name, t->t_name)))
-		goto out;
+	if (likely(!strcasecmp(it->it_name, t->name)))
+		goto out_unlock;
 
-	list_for_each_entry(t, &ce->ce_tlist, t_list) {
-		if (!strcasecmp(t->t_name, it->it_name)) {
-			ce->ce_tgthint = t;
+	list_for_each_entry(t, &ce->tlist, list) {
+		if (!strcasecmp(t->name, it->it_name)) {
+			ce->tgthint = t;
 			cifs_dbg(FYI, "%s: new target hint: %s\n", __func__,
 				 it->it_name);
 			break;
 		}
 	}
 
-out:
-	mutex_unlock(&dfs_cache_list_lock);
+out_unlock:
+	up_write(&htable_rw_lock);
+out_free_path:
 	free_normalized_path(path, npath);
+
 	return rc;
 }
 
@@ -989,10 +992,10 @@ int dfs_cache_noreq_update_tgthint(const char *path,
 {
 	int rc;
 	char *npath;
-	struct dfs_cache_entry *ce;
-	struct dfs_cache_tgt *t;
+	struct cache_entry *ce;
+	struct cache_dfs_tgt *t;
 
-	if (unlikely(!is_path_valid(path)) || !it)
+	if (!it)
 		return -EINVAL;
 
 	rc = get_normalized_path(path, &npath);
@@ -1001,33 +1004,33 @@ int dfs_cache_noreq_update_tgthint(const char *path,
 
 	cifs_dbg(FYI, "%s: path: %s\n", __func__, npath);
 
-	mutex_lock(&dfs_cache_list_lock);
+	down_write(&htable_rw_lock);
 
-	ce = do_dfs_cache_find(0, NULL, NULL, 0, npath, true);
+	ce = lookup_cache_entry(npath, NULL);
 	if (IS_ERR(ce)) {
 		rc = PTR_ERR(ce);
-		goto out;
+		goto out_unlock;
 	}
 
 	rc = 0;
+	t = ce->tgthint;
 
-	t = ce->ce_tgthint;
-
-	if (unlikely(!strcasecmp(it->it_name, t->t_name)))
-		goto out;
+	if (unlikely(!strcasecmp(it->it_name, t->name)))
+		goto out_unlock;
 
-	list_for_each_entry(t, &ce->ce_tlist, t_list) {
-		if (!strcasecmp(t->t_name, it->it_name)) {
-			ce->ce_tgthint = t;
+	list_for_each_entry(t, &ce->tlist, list) {
+		if (!strcasecmp(t->name, it->it_name)) {
+			ce->tgthint = t;
 			cifs_dbg(FYI, "%s: new target hint: %s\n", __func__,
 				 it->it_name);
 			break;
 		}
 	}
 
-out:
-	mutex_unlock(&dfs_cache_list_lock);
+out_unlock:
+	up_write(&htable_rw_lock);
 	free_normalized_path(path, npath);
+
 	return rc;
 }
 
@@ -1047,13 +1050,10 @@ int dfs_cache_get_tgt_referral(const char *path,
 {
 	int rc;
 	char *npath;
-	struct dfs_cache_entry *ce;
-	unsigned int h;
+	struct cache_entry *ce;
 
 	if (!it || !ref)
 		return -EINVAL;
-	if (unlikely(!is_path_valid(path)))
-		return -EINVAL;
 
 	rc = get_normalized_path(path, &npath);
 	if (rc)
@@ -1061,21 +1061,22 @@ int dfs_cache_get_tgt_referral(const char *path,
 
 	cifs_dbg(FYI, "%s: path: %s\n", __func__, npath);
 
-	mutex_lock(&dfs_cache_list_lock);
+	down_read(&htable_rw_lock);
 
-	ce = find_cache_entry(npath, &h);
+	ce = lookup_cache_entry(npath, NULL);
 	if (IS_ERR(ce)) {
 		rc = PTR_ERR(ce);
-		goto out;
+		goto out_unlock;
 	}
 
 	cifs_dbg(FYI, "%s: target name: %s\n", __func__, it->it_name);
 
-	rc = setup_ref(path, ce, ref, it->it_name);
+	rc = setup_referral(path, ce, ref, it->it_name);
 
-out:
-	mutex_unlock(&dfs_cache_list_lock);
+out_unlock:
+	up_read(&htable_rw_lock);
 	free_normalized_path(path, npath);
+
 	return rc;
 }
 
@@ -1085,7 +1086,7 @@ static int dup_vol(struct smb_vol *vol, struct smb_vol *new)
 
 	if (vol->username) {
 		new->username = kstrndup(vol->username, strlen(vol->username),
-					GFP_KERNEL);
+					 GFP_KERNEL);
 		if (!new->username)
 			return -ENOMEM;
 	}
@@ -1103,7 +1104,7 @@ static int dup_vol(struct smb_vol *vol, struct smb_vol *new)
 	}
 	if (vol->domainname) {
 		new->domainname = kstrndup(vol->domainname,
-					  strlen(vol->domainname), GFP_KERNEL);
+					   strlen(vol->domainname), GFP_KERNEL);
 		if (!new->domainname)
 			goto err_free_unc;
 	}
@@ -1150,7 +1151,7 @@ err_free_username:
 int dfs_cache_add_vol(char *mntdata, struct smb_vol *vol, const char *fullpath)
 {
 	int rc;
-	struct dfs_cache_vol_info *vi;
+	struct vol_info *vi;
 
 	if (!vol || !fullpath || !mntdata)
 		return -EINVAL;
@@ -1161,38 +1162,41 @@ int dfs_cache_add_vol(char *mntdata, struct smb_vol *vol, const char *fullpath)
 	if (!vi)
 		return -ENOMEM;
 
-	vi->vi_fullpath = kstrndup(fullpath, strlen(fullpath), GFP_KERNEL);
-	if (!vi->vi_fullpath) {
+	vi->fullpath = kstrndup(fullpath, strlen(fullpath), GFP_KERNEL);
+	if (!vi->fullpath) {
 		rc = -ENOMEM;
 		goto err_free_vi;
 	}
 
-	rc = dup_vol(vol, &vi->vi_vol);
+	rc = dup_vol(vol, &vi->smb_vol);
 	if (rc)
 		goto err_free_fullpath;
 
-	vi->vi_mntdata = mntdata;
+	vi->mntdata = mntdata;
+	spin_lock_init(&vi->smb_vol_lock);
+	kref_init(&vi->refcnt);
+
+	spin_lock(&vol_list_lock);
+	list_add_tail(&vi->list, &vol_list);
+	spin_unlock(&vol_list_lock);
 
-	mutex_lock(&dfs_cache.dc_lock);
-	list_add_tail(&vi->vi_list, &dfs_cache.dc_vol_list);
-	mutex_unlock(&dfs_cache.dc_lock);
 	return 0;
 
 err_free_fullpath:
-	kfree(vi->vi_fullpath);
+	kfree(vi->fullpath);
 err_free_vi:
 	kfree(vi);
 	return rc;
 }
 
-static inline struct dfs_cache_vol_info *find_vol(const char *fullpath)
+/* Must be called with vol_list_lock held */
+static struct vol_info *find_vol(const char *fullpath)
 {
-	struct dfs_cache_vol_info *vi;
+	struct vol_info *vi;
 
-	list_for_each_entry(vi, &dfs_cache.dc_vol_list, vi_list) {
-		cifs_dbg(FYI, "%s: vi->vi_fullpath: %s\n", __func__,
-			 vi->vi_fullpath);
-		if (!strcasecmp(vi->vi_fullpath, fullpath))
+	list_for_each_entry(vi, &vol_list, list) {
+		cifs_dbg(FYI, "%s: vi->fullpath: %s\n", __func__, vi->fullpath);
+		if (!strcasecmp(vi->fullpath, fullpath))
 			return vi;
 	}
 	return ERR_PTR(-ENOENT);
@@ -1208,30 +1212,31 @@ static inline struct dfs_cache_vol_info *find_vol(const char *fullpath)
  */
 int dfs_cache_update_vol(const char *fullpath, struct TCP_Server_Info *server)
 {
-	int rc;
-	struct dfs_cache_vol_info *vi;
+	struct vol_info *vi;
 
 	if (!fullpath || !server)
 		return -EINVAL;
 
 	cifs_dbg(FYI, "%s: fullpath: %s\n", __func__, fullpath);
 
-	mutex_lock(&dfs_cache.dc_lock);
-
+	spin_lock(&vol_list_lock);
 	vi = find_vol(fullpath);
 	if (IS_ERR(vi)) {
-		rc = PTR_ERR(vi);
-		goto out;
+		spin_unlock(&vol_list_lock);
+		return PTR_ERR(vi);
 	}
+	kref_get(&vi->refcnt);
+	spin_unlock(&vol_list_lock);
 
 	cifs_dbg(FYI, "%s: updating volume info\n", __func__);
-	memcpy(&vi->vi_vol.dstaddr, &server->dstaddr,
-	       sizeof(vi->vi_vol.dstaddr));
-	rc = 0;
+	spin_lock(&vi->smb_vol_lock);
+	memcpy(&vi->smb_vol.dstaddr, &server->dstaddr,
+	       sizeof(vi->smb_vol.dstaddr));
+	spin_unlock(&vi->smb_vol_lock);
 
-out:
-	mutex_unlock(&dfs_cache.dc_lock);
-	return rc;
+	kref_put(&vi->refcnt, vol_release);
+
+	return 0;
 }
 
 /**
@@ -1241,18 +1246,56 @@ out:
  */
 void dfs_cache_del_vol(const char *fullpath)
 {
-	struct dfs_cache_vol_info *vi;
+	struct vol_info *vi;
 
 	if (!fullpath || !*fullpath)
 		return;
 
 	cifs_dbg(FYI, "%s: fullpath: %s\n", __func__, fullpath);
 
-	mutex_lock(&dfs_cache.dc_lock);
+	spin_lock(&vol_list_lock);
 	vi = find_vol(fullpath);
-	if (!IS_ERR(vi))
-		free_vol(vi);
-	mutex_unlock(&dfs_cache.dc_lock);
+	spin_unlock(&vol_list_lock);
+
+	kref_put(&vi->refcnt, vol_release);
+}
+
+/**
+ * dfs_cache_get_tgt_share - parse a DFS target
+ *
+ * @it: DFS target iterator.
+ * @share: tree name.
+ * @share_len: length of tree name.
+ * @prefix: prefix path.
+ * @prefix_len: length of prefix path.
+ *
+ * Return zero if target was parsed correctly, otherwise non-zero.
+ */
+int dfs_cache_get_tgt_share(const struct dfs_cache_tgt_iterator *it,
+			    const char **share, size_t *share_len,
+			    const char **prefix, size_t *prefix_len)
+{
+	char *s, sep;
+
+	if (!it || !share || !share_len || !prefix || !prefix_len)
+		return -EINVAL;
+
+	sep = it->it_name[0];
+	if (sep != '\\' && sep != '/')
+		return -EINVAL;
+
+	s = strchr(it->it_name + 1, sep);
+	if (!s)
+		return -EINVAL;
+
+	s = strchrnul(s + 1, sep);
+
+	*share = it->it_name;
+	*share_len = s - it->it_name;
+	*prefix = *s ? s + 1 : s;
+	*prefix_len = &it->it_name[strlen(it->it_name)] - *prefix;
+
+	return 0;
 }
 
 /* Get all tcons that are within a DFS namespace and can be refreshed */
@@ -1280,7 +1323,7 @@ static void get_tcons(struct TCP_Server_Info *server, struct list_head *head)
 	spin_unlock(&cifs_tcp_ses_lock);
 }
 
-static inline bool is_dfs_link(const char *path)
+static bool is_dfs_link(const char *path)
 {
 	char *s;
 
@@ -1290,7 +1333,7 @@ static inline bool is_dfs_link(const char *path)
 	return !!strchr(s + 1, '\\');
 }
 
-static inline char *get_dfs_root(const char *path)
+static char *get_dfs_root(const char *path)
 {
 	char *s, *npath;
 
@@ -1309,32 +1352,67 @@ static inline char *get_dfs_root(const char *path)
 	return npath;
 }
 
+static inline void put_tcp_server(struct TCP_Server_Info *server)
+{
+	cifs_put_tcp_session(server, 0);
+}
+
+static struct TCP_Server_Info *get_tcp_server(struct smb_vol *vol)
+{
+	struct TCP_Server_Info *server;
+
+	server = cifs_find_tcp_session(vol);
+	if (IS_ERR_OR_NULL(server))
+		return NULL;
+
+	spin_lock(&GlobalMid_Lock);
+	if (server->tcpStatus != CifsGood) {
+		spin_unlock(&GlobalMid_Lock);
+		put_tcp_server(server);
+		return NULL;
+	}
+	spin_unlock(&GlobalMid_Lock);
+
+	return server;
+}
+
 /* Find root SMB session out of a DFS link path */
-static struct cifs_ses *find_root_ses(struct dfs_cache_vol_info *vi,
-				      struct cifs_tcon *tcon, const char *path)
+static struct cifs_ses *find_root_ses(struct vol_info *vi,
+				      struct cifs_tcon *tcon,
+				      const char *path)
 {
 	char *rpath;
 	int rc;
+	struct cache_entry *ce;
 	struct dfs_info3_param ref = {0};
 	char *mdata = NULL, *devname = NULL;
-	bool is_smb3 = tcon->ses->server->vals->header_preamble_size == 0;
 	struct TCP_Server_Info *server;
 	struct cifs_ses *ses;
-	struct smb_vol vol;
+	struct smb_vol vol = {NULL};
 
 	rpath = get_dfs_root(path);
 	if (IS_ERR(rpath))
 		return ERR_CAST(rpath);
 
-	memset(&vol, 0, sizeof(vol));
+	down_read(&htable_rw_lock);
 
-	rc = dfs_cache_noreq_find(rpath, &ref, NULL);
+	ce = lookup_cache_entry(rpath, NULL);
+	if (IS_ERR(ce)) {
+		up_read(&htable_rw_lock);
+		ses = ERR_CAST(ce);
+		goto out;
+	}
+
+	rc = setup_referral(path, ce, &ref, get_tgt_name(ce));
 	if (rc) {
+		up_read(&htable_rw_lock);
 		ses = ERR_PTR(rc);
 		goto out;
 	}
 
-	mdata = cifs_compose_mount_options(vi->vi_mntdata, rpath, &ref,
+	up_read(&htable_rw_lock);
+
+	mdata = cifs_compose_mount_options(vi->mntdata, rpath, &ref,
 					   &devname);
 	free_dfs_info_param(&ref);
 
@@ -1344,7 +1422,7 @@ static struct cifs_ses *find_root_ses(struct dfs_cache_vol_info *vi,
 		goto out;
 	}
 
-	rc = cifs_setup_volume_info(&vol, mdata, devname, is_smb3);
+	rc = cifs_setup_volume_info(&vol, mdata, devname, false);
 	kfree(devname);
 
 	if (rc) {
@@ -1352,13 +1430,8 @@ static struct cifs_ses *find_root_ses(struct dfs_cache_vol_info *vi,
 		goto out;
 	}
 
-	server = cifs_find_tcp_session(&vol);
-	if (IS_ERR_OR_NULL(server)) {
-		ses = ERR_PTR(-EHOSTDOWN);
-		goto out;
-	}
-	if (server->tcpStatus != CifsGood) {
-		cifs_put_tcp_session(server, 0);
+	server = get_tcp_server(&vol);
+	if (!server) {
 		ses = ERR_PTR(-EHOSTDOWN);
 		goto out;
 	}
@@ -1374,17 +1447,15 @@ out:
 }
 
 /* Refresh DFS cache entry from a given tcon */
-static void do_refresh_tcon(struct dfs_cache *dc, struct dfs_cache_vol_info *vi,
-			    struct cifs_tcon *tcon)
+static int refresh_tcon(struct vol_info *vi, struct cifs_tcon *tcon)
 {
 	int rc = 0;
 	unsigned int xid;
 	char *path, *npath;
-	unsigned int h;
-	struct dfs_cache_entry *ce;
+	struct cache_entry *ce;
+	struct cifs_ses *root_ses = NULL, *ses;
 	struct dfs_info3_param *refs = NULL;
 	int numrefs = 0;
-	struct cifs_ses *root_ses = NULL, *ses;
 
 	xid = get_xid();
 
@@ -1392,19 +1463,23 @@ static void do_refresh_tcon(struct dfs_cache *dc, struct dfs_cache_vol_info *vi,
 
 	rc = get_normalized_path(path, &npath);
 	if (rc)
-		goto out;
+		goto out_free_xid;
 
-	mutex_lock(&dfs_cache_list_lock);
-	ce = find_cache_entry(npath, &h);
-	mutex_unlock(&dfs_cache_list_lock);
+	down_read(&htable_rw_lock);
 
+	ce = lookup_cache_entry(npath, NULL);
 	if (IS_ERR(ce)) {
 		rc = PTR_ERR(ce);
-		goto out;
+		up_read(&htable_rw_lock);
+		goto out_free_path;
 	}
 
-	if (!cache_entry_expired(ce))
-		goto out;
+	if (!cache_entry_expired(ce)) {
+		up_read(&htable_rw_lock);
+		goto out_free_path;
+	}
+
+	up_read(&htable_rw_lock);
 
 	/* If it's a DFS Link, then use root SMB session for refreshing it */
 	if (is_dfs_link(npath)) {
@@ -1412,35 +1487,29 @@ static void do_refresh_tcon(struct dfs_cache *dc, struct dfs_cache_vol_info *vi,
 		if (IS_ERR(ses)) {
 			rc = PTR_ERR(ses);
 			root_ses = NULL;
-			goto out;
+			goto out_free_path;
 		}
 	} else {
 		ses = tcon->ses;
 	}
 
-	if (unlikely(!ses->server->ops->get_dfs_refer)) {
-		rc = -EOPNOTSUPP;
-	} else {
-		rc = ses->server->ops->get_dfs_refer(xid, ses, path, &refs,
-						     &numrefs, dc->dc_nlsc,
-						     tcon->remap);
-		if (!rc) {
-			mutex_lock(&dfs_cache_list_lock);
-			ce = __update_cache_entry(npath, refs, numrefs);
-			mutex_unlock(&dfs_cache_list_lock);
-			dump_refs(refs, numrefs);
-			free_dfs_info_array(refs, numrefs);
-			if (IS_ERR(ce))
-				rc = PTR_ERR(ce);
-		}
+	rc = get_dfs_referral(xid, ses, cache_nlsc, tcon->remap, npath, &refs,
+			      &numrefs);
+	if (!rc) {
+		dump_refs(refs, numrefs);
+		rc = update_cache_entry(npath, refs, numrefs);
+		free_dfs_info_array(refs, numrefs);
 	}
 
-out:
 	if (root_ses)
 		cifs_put_smb_ses(root_ses);
 
-	free_xid(xid);
+out_free_path:
 	free_normalized_path(path, npath);
+
+out_free_xid:
+	free_xid(xid);
+	return rc;
 }
 
 /*
@@ -1449,30 +1518,61 @@ out:
  */
 static void refresh_cache_worker(struct work_struct *work)
 {
-	struct dfs_cache *dc = container_of(work, struct dfs_cache,
-					    dc_refresh.work);
-	struct dfs_cache_vol_info *vi;
+	struct vol_info *vi, *nvi;
 	struct TCP_Server_Info *server;
-	LIST_HEAD(list);
+	LIST_HEAD(vols);
+	LIST_HEAD(tcons);
 	struct cifs_tcon *tcon, *ntcon;
+	int rc;
 
-	mutex_lock(&dc->dc_lock);
-
-	list_for_each_entry(vi, &dc->dc_vol_list, vi_list) {
-		server = cifs_find_tcp_session(&vi->vi_vol);
-		if (IS_ERR_OR_NULL(server))
+	/*
+	 * Find SMB volumes that are eligible (server->tcpStatus == CifsGood)
+	 * for refreshing.
+	 */
+	spin_lock(&vol_list_lock);
+	list_for_each_entry(vi, &vol_list, list) {
+		server = get_tcp_server(&vi->smb_vol);
+		if (!server)
 			continue;
-		if (server->tcpStatus != CifsGood)
-			goto next;
-		get_tcons(server, &list);
-		list_for_each_entry_safe(tcon, ntcon, &list, ulist) {
-			do_refresh_tcon(dc, vi, tcon);
+
+		kref_get(&vi->refcnt);
+		list_add_tail(&vi->rlist, &vols);
+		put_tcp_server(server);
+	}
+	spin_unlock(&vol_list_lock);
+
+	/* Walk through all TCONs and refresh any expired cache entry */
+	list_for_each_entry_safe(vi, nvi, &vols, rlist) {
+		spin_lock(&vi->smb_vol_lock);
+		server = get_tcp_server(&vi->smb_vol);
+		spin_unlock(&vi->smb_vol_lock);
+
+		if (!server)
+			goto next_vol;
+
+		get_tcons(server, &tcons);
+		rc = 0;
+
+		list_for_each_entry_safe(tcon, ntcon, &tcons, ulist) {
+			/*
+			 * Skip tcp server if any of its tcons failed to refresh
+			 * (possibily due to reconnects).
+			 */
+			if (!rc)
+				rc = refresh_tcon(vi, tcon);
+
 			list_del_init(&tcon->ulist);
 			cifs_put_tcon(tcon);
 		}
-next:
-		cifs_put_tcp_session(server, 0);
+
+		put_tcp_server(server);
+
+next_vol:
+		list_del_init(&vi->rlist);
+		kref_put(&vi->refcnt, vol_release);
 	}
-	queue_delayed_work(cifsiod_wq, &dc->dc_refresh, dc->dc_ttl * HZ);
-	mutex_unlock(&dc->dc_lock);
+
+	spin_lock(&cache_ttl_lock);
+	queue_delayed_work(dfscache_wq, &refresh_task, cache_ttl * HZ);
+	spin_unlock(&cache_ttl_lock);
 }
diff --git a/fs/cifs/dfs_cache.h b/fs/cifs/dfs_cache.h
index 76c732943f5f..bf94d08cfb5a 100644
--- a/fs/cifs/dfs_cache.h
+++ b/fs/cifs/dfs_cache.h
@@ -24,7 +24,7 @@ struct dfs_cache_tgt_iterator {
 
 extern int dfs_cache_init(void);
 extern void dfs_cache_destroy(void);
-extern const struct file_operations dfscache_proc_fops;
+extern const struct proc_ops dfscache_proc_ops;
 
 extern int dfs_cache_find(const unsigned int xid, struct cifs_ses *ses,
 			  const struct nls_table *nls_codepage, int remap,
@@ -49,6 +49,10 @@ extern int dfs_cache_update_vol(const char *fullpath,
 				struct TCP_Server_Info *server);
 extern void dfs_cache_del_vol(const char *fullpath);
 
+extern int dfs_cache_get_tgt_share(const struct dfs_cache_tgt_iterator *it,
+				   const char **share, size_t *share_len,
+				   const char **prefix, size_t *prefix_len);
+
 static inline struct dfs_cache_tgt_iterator *
 dfs_cache_get_next_tgt(struct dfs_cache_tgt_list *tl,
 		       struct dfs_cache_tgt_iterator *it)
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 7ce689d31aa2..36e7b2fd2190 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -244,10 +244,8 @@ cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned int xid,
 		*oplock = REQ_OPLOCK;
 
 	full_path = build_path_from_dentry(direntry);
-	if (full_path == NULL) {
-		rc = -ENOMEM;
-		goto out;
-	}
+	if (!full_path)
+		return -ENOMEM;
 
 	if (tcon->unix_ext && cap_unix(tcon->ses) && !tcon->broken_posix_open &&
 	    (CIFS_UNIX_POSIX_PATH_OPS_CAP &
@@ -357,13 +355,10 @@ cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned int xid,
 	if (!tcon->unix_ext && (mode & S_IWUGO) == 0)
 		create_options |= CREATE_OPTION_READONLY;
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
-
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = desired_access;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, create_options);
 	oparms.disposition = disposition;
 	oparms.path = full_path;
 	oparms.fid = fid;
@@ -560,7 +555,6 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry,
 		if (server->ops->close)
 			server->ops->close(xid, tcon, &fid);
 		cifs_del_pending_open(&open);
-		fput(file);
 		rc = -ENOMEM;
 	}
 
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index fa7b0fa72bb3..75ddce8ef456 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -222,9 +222,6 @@ cifs_nt_open(char *full_path, struct inode *inode, struct cifs_sb_info *cifs_sb,
 	if (!buf)
 		return -ENOMEM;
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
-
 	/* O_SYNC also has bit for O_DSYNC so following check picks up either */
 	if (f_flags & O_SYNC)
 		create_options |= CREATE_WRITE_THROUGH;
@@ -235,7 +232,7 @@ cifs_nt_open(char *full_path, struct inode *inode, struct cifs_sb_info *cifs_sb,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = desired_access;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, create_options);
 	oparms.disposition = disposition;
 	oparms.path = full_path;
 	oparms.fid = fid;
@@ -288,6 +285,8 @@ cifs_down_write(struct rw_semaphore *sem)
 		msleep(10);
 }
 
+static void cifsFileInfo_put_work(struct work_struct *work);
+
 struct cifsFileInfo *
 cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
 		  struct tcon_link *tlink, __u32 oplock)
@@ -313,9 +312,6 @@ cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
 	INIT_LIST_HEAD(&fdlocks->locks);
 	fdlocks->cfile = cfile;
 	cfile->llist = fdlocks;
-	cifs_down_write(&cinode->lock_sem);
-	list_add(&fdlocks->llist, &cinode->llist);
-	up_write(&cinode->lock_sem);
 
 	cfile->count = 1;
 	cfile->pid = current->tgid;
@@ -325,6 +321,7 @@ cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
 	cfile->invalidHandle = false;
 	cfile->tlink = cifs_get_tlink(tlink);
 	INIT_WORK(&cfile->oplock_break, cifs_oplock_break);
+	INIT_WORK(&cfile->put, cifsFileInfo_put_work);
 	mutex_init(&cfile->fh_mutex);
 	spin_lock_init(&cfile->file_info_lock);
 
@@ -339,6 +336,10 @@ cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
 		oplock = 0;
 	}
 
+	cifs_down_write(&cinode->lock_sem);
+	list_add(&fdlocks->llist, &cinode->llist);
+	up_write(&cinode->lock_sem);
+
 	spin_lock(&tcon->open_file_lock);
 	if (fid->pending_open->oplock != CIFS_OPLOCK_NO_CHANGE && oplock)
 		oplock = fid->pending_open->oplock;
@@ -375,6 +376,41 @@ cifsFileInfo_get(struct cifsFileInfo *cifs_file)
 	return cifs_file;
 }
 
+static void cifsFileInfo_put_final(struct cifsFileInfo *cifs_file)
+{
+	struct inode *inode = d_inode(cifs_file->dentry);
+	struct cifsInodeInfo *cifsi = CIFS_I(inode);
+	struct cifsLockInfo *li, *tmp;
+	struct super_block *sb = inode->i_sb;
+
+	/*
+	 * Delete any outstanding lock records. We'll lose them when the file
+	 * is closed anyway.
+	 */
+	cifs_down_write(&cifsi->lock_sem);
+	list_for_each_entry_safe(li, tmp, &cifs_file->llist->locks, llist) {
+		list_del(&li->llist);
+		cifs_del_lock_waiters(li);
+		kfree(li);
+	}
+	list_del(&cifs_file->llist->llist);
+	kfree(cifs_file->llist);
+	up_write(&cifsi->lock_sem);
+
+	cifs_put_tlink(cifs_file->tlink);
+	dput(cifs_file->dentry);
+	cifs_sb_deactive(sb);
+	kfree(cifs_file);
+}
+
+static void cifsFileInfo_put_work(struct work_struct *work)
+{
+	struct cifsFileInfo *cifs_file = container_of(work,
+			struct cifsFileInfo, put);
+
+	cifsFileInfo_put_final(cifs_file);
+}
+
 /**
  * cifsFileInfo_put - release a reference of file priv data
  *
@@ -382,15 +418,15 @@ cifsFileInfo_get(struct cifsFileInfo *cifs_file)
  */
 void cifsFileInfo_put(struct cifsFileInfo *cifs_file)
 {
-	_cifsFileInfo_put(cifs_file, true);
+	_cifsFileInfo_put(cifs_file, true, true);
 }
 
 /**
  * _cifsFileInfo_put - release a reference of file priv data
  *
  * This may involve closing the filehandle @cifs_file out on the
- * server. Must be called without holding tcon->open_file_lock and
- * cifs_file->file_info_lock.
+ * server. Must be called without holding tcon->open_file_lock,
+ * cinode->open_file_lock and cifs_file->file_info_lock.
  *
  * If @wait_for_oplock_handler is true and we are releasing the last
  * reference, wait for any running oplock break handler of the file
@@ -398,7 +434,8 @@ void cifsFileInfo_put(struct cifsFileInfo *cifs_file)
  * oplock break handler, you need to pass false.
  *
  */
-void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_handler)
+void _cifsFileInfo_put(struct cifsFileInfo *cifs_file,
+		       bool wait_oplock_handler, bool offload)
 {
 	struct inode *inode = d_inode(cifs_file->dentry);
 	struct cifs_tcon *tcon = tlink_tcon(cifs_file->tlink);
@@ -406,7 +443,6 @@ void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_handler)
 	struct cifsInodeInfo *cifsi = CIFS_I(inode);
 	struct super_block *sb = inode->i_sb;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
-	struct cifsLockInfo *li, *tmp;
 	struct cifs_fid fid;
 	struct cifs_pending_open open;
 	bool oplock_break_cancelled;
@@ -457,7 +493,9 @@ void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_handler)
 		unsigned int xid;
 
 		xid = get_xid();
-		if (server->ops->close)
+		if (server->ops->close_getattr)
+			server->ops->close_getattr(xid, tcon, cifs_file);
+		else if (server->ops->close)
 			server->ops->close(xid, tcon, &cifs_file->fid);
 		_free_xid(xid);
 	}
@@ -467,24 +505,10 @@ void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_handler)
 
 	cifs_del_pending_open(&open);
 
-	/*
-	 * Delete any outstanding lock records. We'll lose them when the file
-	 * is closed anyway.
-	 */
-	cifs_down_write(&cifsi->lock_sem);
-	list_for_each_entry_safe(li, tmp, &cifs_file->llist->locks, llist) {
-		list_del(&li->llist);
-		cifs_del_lock_waiters(li);
-		kfree(li);
-	}
-	list_del(&cifs_file->llist->llist);
-	kfree(cifs_file->llist);
-	up_write(&cifsi->lock_sem);
-
-	cifs_put_tlink(cifs_file->tlink);
-	dput(cifs_file->dentry);
-	cifs_sb_deactive(sb);
-	kfree(cifs_file);
+	if (offload)
+		queue_work(fileinfo_put_wq, &cifs_file->put);
+	else
+		cifsFileInfo_put_final(cifs_file);
 }
 
 int cifs_open(struct inode *inode, struct file *file)
@@ -725,8 +749,12 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush)
 
 	desired_access = cifs_convert_flags(cfile->f_flags);
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
+	/* O_SYNC also has bit for O_DSYNC so following check picks up either */
+	if (cfile->f_flags & O_SYNC)
+		create_options |= CREATE_WRITE_THROUGH;
+
+	if (cfile->f_flags & O_DIRECT)
+		create_options |= CREATE_NO_BUFFER;
 
 	if (server->ops->get_lease_key)
 		server->ops->get_lease_key(inode, &cfile->fid);
@@ -734,7 +762,7 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush)
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = desired_access;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, create_options);
 	oparms.disposition = disposition;
 	oparms.path = full_path;
 	oparms.fid = &cfile->fid;
@@ -808,7 +836,7 @@ reopen_error_exit:
 int cifs_close(struct inode *inode, struct file *file)
 {
 	if (file->private_data != NULL) {
-		cifsFileInfo_put(file->private_data);
+		_cifsFileInfo_put(file->private_data, true, false);
 		file->private_data = NULL;
 	}
 
@@ -1141,7 +1169,8 @@ try_again:
 	rc = posix_lock_file(file, flock, NULL);
 	up_write(&cinode->lock_sem);
 	if (rc == FILE_LOCK_DEFERRED) {
-		rc = wait_event_interruptible(flock->fl_wait, !flock->fl_blocker);
+		rc = wait_event_interruptible(flock->fl_wait,
+					list_empty(&flock->fl_blocked_member));
 		if (!rc)
 			goto try_again;
 		locks_delete_block(flock);
@@ -1681,7 +1710,7 @@ cifs_setlk(struct file *file, struct file_lock *flock, __u32 type,
 		rc = server->ops->mand_unlock_range(cfile, flock, xid);
 
 out:
-	if (flock->fl_flags & FL_POSIX) {
+	if ((flock->fl_flags & FL_POSIX) || (flock->fl_flags & FL_FLOCK)) {
 		/*
 		 * If this is a request to remove all locks because we
 		 * are closing the file, it doesn't matter if the
@@ -1698,6 +1727,52 @@ out:
 	return rc;
 }
 
+int cifs_flock(struct file *file, int cmd, struct file_lock *fl)
+{
+	int rc, xid;
+	int lock = 0, unlock = 0;
+	bool wait_flag = false;
+	bool posix_lck = false;
+	struct cifs_sb_info *cifs_sb;
+	struct cifs_tcon *tcon;
+	struct cifsFileInfo *cfile;
+	__u32 type;
+
+	rc = -EACCES;
+	xid = get_xid();
+
+	if (!(fl->fl_flags & FL_FLOCK))
+		return -ENOLCK;
+
+	cfile = (struct cifsFileInfo *)file->private_data;
+	tcon = tlink_tcon(cfile->tlink);
+
+	cifs_read_flock(fl, &type, &lock, &unlock, &wait_flag,
+			tcon->ses->server);
+	cifs_sb = CIFS_FILE_SB(file);
+
+	if (cap_unix(tcon->ses) &&
+	    (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) &&
+	    ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0))
+		posix_lck = true;
+
+	if (!lock && !unlock) {
+		/*
+		 * if no lock or unlock then nothing to do since we do not
+		 * know what it is
+		 */
+		free_xid(xid);
+		return -EOPNOTSUPP;
+	}
+
+	rc = cifs_setlk(file, fl, type, wait_flag, posix_lck, lock, unlock,
+			xid);
+	free_xid(xid);
+	return rc;
+
+
+}
+
 int cifs_lock(struct file *file, int cmd, struct file_lock *flock)
 {
 	int rc, xid;
@@ -1884,7 +1959,7 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode,
 
 /* Return -EBADF if no handle is found and general rc otherwise */
 int
-cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only,
+cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, int flags,
 		       struct cifsFileInfo **ret_file)
 {
 	struct cifsFileInfo *open_file, *inv_file = NULL;
@@ -1892,7 +1967,8 @@ cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only,
 	bool any_available = false;
 	int rc = -EBADF;
 	unsigned int refind = 0;
-
+	bool fsuid_only = flags & FIND_WR_FSUID_ONLY;
+	bool with_delete = flags & FIND_WR_WITH_DELETE;
 	*ret_file = NULL;
 
 	/*
@@ -1924,6 +2000,8 @@ refind_writable:
 			continue;
 		if (fsuid_only && !uid_eq(open_file->uid, current_fsuid()))
 			continue;
+		if (with_delete && !(open_file->fid.access & DELETE))
+			continue;
 		if (OPEN_FMODE(open_file->f_flags) & FMODE_WRITE) {
 			if (!open_file->invalidHandle) {
 				/* found a good writable file */
@@ -1971,12 +2049,12 @@ refind_writable:
 }
 
 struct cifsFileInfo *
-find_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only)
+find_writable_file(struct cifsInodeInfo *cifs_inode, int flags)
 {
 	struct cifsFileInfo *cfile;
 	int rc;
 
-	rc = cifs_get_writable_file(cifs_inode, fsuid_only, &cfile);
+	rc = cifs_get_writable_file(cifs_inode, flags, &cfile);
 	if (rc)
 		cifs_dbg(FYI, "couldn't find writable handle rc=%d", rc);
 
@@ -1985,6 +2063,7 @@ find_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only)
 
 int
 cifs_get_writable_path(struct cifs_tcon *tcon, const char *name,
+		       int flags,
 		       struct cifsFileInfo **ret_file)
 {
 	struct list_head *tmp;
@@ -2011,7 +2090,7 @@ cifs_get_writable_path(struct cifs_tcon *tcon, const char *name,
 		kfree(full_path);
 		cinode = CIFS_I(d_inode(cfile->dentry));
 		spin_unlock(&tcon->open_file_lock);
-		return cifs_get_writable_file(cinode, 0, ret_file);
+		return cifs_get_writable_file(cinode, flags, ret_file);
 	}
 
 	spin_unlock(&tcon->open_file_lock);
@@ -2088,7 +2167,8 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to)
 	if (mapping->host->i_size - offset < (loff_t)to)
 		to = (unsigned)(mapping->host->i_size - offset);
 
-	rc = cifs_get_writable_file(CIFS_I(mapping->host), false, &open_file);
+	rc = cifs_get_writable_file(CIFS_I(mapping->host), FIND_WR_ANY,
+				    &open_file);
 	if (!rc) {
 		bytes_written = cifs_write(open_file, open_file->pid,
 					   write_data, to - from, &offset);
@@ -2281,7 +2361,7 @@ retry:
 		if (cfile)
 			cifsFileInfo_put(cfile);
 
-		rc = cifs_get_writable_file(CIFS_I(inode), false, &cfile);
+		rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY, &cfile);
 
 		/* in case of an error store it to return later */
 		if (rc)
@@ -2519,8 +2599,10 @@ int cifs_strict_fsync(struct file *file, loff_t start, loff_t end,
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
 
 	rc = file_write_and_wait_range(file, start, end);
-	if (rc)
+	if (rc) {
+		trace_cifs_fsync_err(inode->i_ino, rc);
 		return rc;
+	}
 
 	xid = get_xid();
 
@@ -2558,8 +2640,10 @@ int cifs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 	struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file);
 
 	rc = file_write_and_wait_range(file, start, end);
-	if (rc)
+	if (rc) {
+		trace_cifs_fsync_err(file_inode(file)->i_ino, rc);
 		return rc;
+	}
 
 	xid = get_xid();
 
@@ -2592,7 +2676,8 @@ int cifs_flush(struct file *file, fl_owner_t id)
 		rc = filemap_write_and_wait(inode->i_mapping);
 
 	cifs_dbg(FYI, "Flush inode %p file %p rc %d\n", inode, file, rc);
-
+	if (rc)
+		trace_cifs_flush_err(inode->i_ino, rc);
 	return rc;
 }
 
@@ -2757,9 +2842,17 @@ cifs_resend_wdata(struct cifs_writedata *wdata, struct list_head *wdata_list,
 		if (!rc) {
 			if (wdata->cfile->invalidHandle)
 				rc = -EAGAIN;
-			else
+			else {
+#ifdef CONFIG_CIFS_SMB_DIRECT
+				if (wdata->mr) {
+					wdata->mr->need_invalidate = true;
+					smbd_deregister_mr(wdata->mr);
+					wdata->mr = NULL;
+				}
+#endif
 				rc = server->ops->async_writev(wdata,
 					cifs_uncached_writedata_release);
+			}
 		}
 
 		/* If the write was successfully sent, we are done */
@@ -2833,7 +2926,7 @@ cifs_write_from_iter(loff_t offset, size_t len, struct iov_iter *from,
 					"direct_writev couldn't get user pages "
 					"(rc=%zd) iter type %d iov_offset %zd "
 					"count %zd\n",
-					result, from->type,
+					result, iov_iter_type(from),
 					from->iov_offset, from->count);
 				dump_stack();
 
@@ -3044,7 +3137,7 @@ static ssize_t __cifs_writev(
 	 * In this case, fall back to non-direct write function.
 	 * this could be improved by getting pages directly in ITER_KVEC
 	 */
-	if (direct && from->type & ITER_KVEC) {
+	if (direct && iov_iter_is_kvec(from)) {
 		cifs_dbg(FYI, "use non-direct cifs_writev for kvec I/O\n");
 		direct = false;
 	}
@@ -3482,8 +3575,16 @@ static int cifs_resend_rdata(struct cifs_readdata *rdata,
 		if (!rc) {
 			if (rdata->cfile->invalidHandle)
 				rc = -EAGAIN;
-			else
+			else {
+#ifdef CONFIG_CIFS_SMB_DIRECT
+				if (rdata->mr) {
+					rdata->mr->need_invalidate = true;
+					smbd_deregister_mr(rdata->mr);
+					rdata->mr = NULL;
+				}
+#endif
 				rc = server->ops->async_readv(rdata);
+			}
 		}
 
 		/* If the read was successfully sent, we are done */
@@ -3556,7 +3657,7 @@ cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file,
 					"couldn't get user pages (rc=%zd)"
 					" iter type %d"
 					" iov_offset %zd count %zd\n",
-					result, direct_iov.type,
+					result, iov_iter_type(&direct_iov),
 					direct_iov.iov_offset,
 					direct_iov.count);
 				dump_stack();
@@ -3740,7 +3841,7 @@ again:
 	if (rc == -ENODATA)
 		rc = 0;
 
-	ctx->rc = (rc == 0) ? ctx->total_len : rc;
+	ctx->rc = (rc == 0) ? (ssize_t)ctx->total_len : rc;
 
 	mutex_unlock(&ctx->aio_mutex);
 
@@ -3767,7 +3868,7 @@ static ssize_t __cifs_readv(
 	 * fall back to data copy read path
 	 * this could be improved by getting pages directly in ITER_KVEC
 	 */
-	if (direct && to->type & ITER_KVEC) {
+	if (direct && iov_iter_is_kvec(to)) {
 		cifs_dbg(FYI, "use non-direct cifs_user_readv for kvec I/O\n");
 		direct = false;
 	}
@@ -3959,7 +4060,7 @@ cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset)
 			 * than it negotiated since it will refuse the read
 			 * then.
 			 */
-			if ((tcon->ses) && !(tcon->ses->capabilities &
+			if (!(tcon->ses->capabilities &
 				tcon->ses->server->vals->cap_large_files)) {
 				current_read_size = min_t(uint,
 					current_read_size, CIFSMaxBufSize);
@@ -4637,12 +4738,13 @@ void cifs_oplock_break(struct work_struct *work)
 	struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
 	struct TCP_Server_Info *server = tcon->ses->server;
 	int rc = 0;
+	bool purge_cache = false;
 
 	wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS,
 			TASK_UNINTERRUPTIBLE);
 
-	server->ops->downgrade_oplock(server, cinode,
-		test_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, &cinode->flags));
+	server->ops->downgrade_oplock(server, cinode, cfile->oplock_level,
+				      cfile->oplock_epoch, &purge_cache);
 
 	if (!CIFS_CACHE_WRITE(cinode) && CIFS_CACHE_READ(cinode) &&
 						cifs_has_mand_locks(cinode)) {
@@ -4657,18 +4759,21 @@ void cifs_oplock_break(struct work_struct *work)
 		else
 			break_lease(inode, O_WRONLY);
 		rc = filemap_fdatawrite(inode->i_mapping);
-		if (!CIFS_CACHE_READ(cinode)) {
+		if (!CIFS_CACHE_READ(cinode) || purge_cache) {
 			rc = filemap_fdatawait(inode->i_mapping);
 			mapping_set_error(inode->i_mapping, rc);
 			cifs_zap_mapping(inode);
 		}
 		cifs_dbg(FYI, "Oplock flush inode %p rc %d\n", inode, rc);
+		if (CIFS_CACHE_WRITE(cinode))
+			goto oplock_break_ack;
 	}
 
 	rc = cifs_push_locks(cfile);
 	if (rc)
 		cifs_dbg(VFS, "Push locks rc = %d\n", rc);
 
+oplock_break_ack:
 	/*
 	 * releasing stale oplock after recent reconnect of smb session using
 	 * a now incorrect file handle is not a data integrity issue but do
@@ -4680,7 +4785,7 @@ void cifs_oplock_break(struct work_struct *work)
 							     cinode);
 		cifs_dbg(FYI, "Oplock release rc = %d\n", rc);
 	}
-	_cifsFileInfo_put(cfile, false /* do not wait for ourself */);
+	_cifsFileInfo_put(cfile, false /* do not wait for ourself */, false);
 	cifs_done_oplock_break(cinode);
 }
 
@@ -4703,6 +4808,60 @@ cifs_direct_io(struct kiocb *iocb, struct iov_iter *iter)
         return -EINVAL;
 }
 
+static int cifs_swap_activate(struct swap_info_struct *sis,
+			      struct file *swap_file, sector_t *span)
+{
+	struct cifsFileInfo *cfile = swap_file->private_data;
+	struct inode *inode = swap_file->f_mapping->host;
+	unsigned long blocks;
+	long long isize;
+
+	cifs_dbg(FYI, "swap activate\n");
+
+	spin_lock(&inode->i_lock);
+	blocks = inode->i_blocks;
+	isize = inode->i_size;
+	spin_unlock(&inode->i_lock);
+	if (blocks*512 < isize) {
+		pr_warn("swap activate: swapfile has holes\n");
+		return -EINVAL;
+	}
+	*span = sis->pages;
+
+	printk_once(KERN_WARNING "Swap support over SMB3 is experimental\n");
+
+	/*
+	 * TODO: consider adding ACL (or documenting how) to prevent other
+	 * users (on this or other systems) from reading it
+	 */
+
+
+	/* TODO: add sk_set_memalloc(inet) or similar */
+
+	if (cfile)
+		cfile->swapfile = true;
+	/*
+	 * TODO: Since file already open, we can't open with DENY_ALL here
+	 * but we could add call to grab a byte range lock to prevent others
+	 * from reading or writing the file
+	 */
+
+	return 0;
+}
+
+static void cifs_swap_deactivate(struct file *file)
+{
+	struct cifsFileInfo *cfile = file->private_data;
+
+	cifs_dbg(FYI, "swap deactivate\n");
+
+	/* TODO: undo sk_set_memalloc(inet) will eventually be needed */
+
+	if (cfile)
+		cfile->swapfile = false;
+
+	/* do we need to unpin (or unlock) the file */
+}
 
 const struct address_space_operations cifs_addr_ops = {
 	.readpage = cifs_readpage,
@@ -4716,6 +4875,13 @@ const struct address_space_operations cifs_addr_ops = {
 	.direct_IO = cifs_direct_io,
 	.invalidatepage = cifs_invalidate_page,
 	.launder_page = cifs_launder_page,
+	/*
+	 * TODO: investigate and if useful we could add an cifs_migratePage
+	 * helper (under an CONFIG_MIGRATION) in the future, and also
+	 * investigate and add an is_dirty_writeback helper if needed
+	 */
+	.swap_activate = cifs_swap_activate,
+	.swap_deactivate = cifs_swap_deactivate,
 };
 
 /*
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index df9377828e2f..5d2965a23730 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -61,7 +61,7 @@ static void cifs_set_ops(struct inode *inode)
 		}
 
 		/* check if server can support readpages */
-		if (cifs_sb_master_tcon(cifs_sb)->ses->server->maxBuf <
+		if (cifs_sb_master_tcon(cifs_sb)->ses->server->max_read <
 				PAGE_SIZE + MAX_CIFS_HDR_SIZE)
 			inode->i_data.a_ops = &cifs_addr_ops_smallbuf;
 		else
@@ -113,6 +113,7 @@ cifs_revalidate_cache(struct inode *inode, struct cifs_fattr *fattr)
 	}
 
 	 /* revalidate if mtime or size have changed */
+	fattr->cf_mtime = timestamp_truncate(fattr->cf_mtime, inode);
 	if (timespec64_equal(&inode->i_mtime, &fattr->cf_mtime) &&
 	    cifs_i->server_eof == fattr->cf_eof) {
 		cifs_dbg(FYI, "%s: inode %llu is unchanged\n",
@@ -162,8 +163,11 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr)
 	cifs_revalidate_cache(inode, fattr);
 
 	spin_lock(&inode->i_lock);
+	fattr->cf_mtime = timestamp_truncate(fattr->cf_mtime, inode);
+	fattr->cf_atime = timestamp_truncate(fattr->cf_atime, inode);
+	fattr->cf_ctime = timestamp_truncate(fattr->cf_ctime, inode);
 	/* we do not want atime to be less than mtime, it broke some apps */
-	if (timespec64_compare(&fattr->cf_atime, &fattr->cf_mtime))
+	if (timespec64_compare(&fattr->cf_atime, &fattr->cf_mtime) < 0)
 		inode->i_atime = fattr->cf_mtime;
 	else
 		inode->i_atime = fattr->cf_atime;
@@ -329,8 +333,7 @@ cifs_create_dfs_fattr(struct cifs_fattr *fattr, struct super_block *sb)
 	fattr->cf_mode = S_IFDIR | S_IXUGO | S_IRWXU;
 	fattr->cf_uid = cifs_sb->mnt_uid;
 	fattr->cf_gid = cifs_sb->mnt_gid;
-	ktime_get_real_ts64(&fattr->cf_mtime);
-	fattr->cf_mtime = timespec64_trunc(fattr->cf_mtime, sb->s_time_gran);
+	ktime_get_coarse_real_ts64(&fattr->cf_mtime);
 	fattr->cf_atime = fattr->cf_ctime = fattr->cf_mtime;
 	fattr->cf_nlink = 2;
 	fattr->cf_flags = CIFS_FATTR_DFS_REFERRAL;
@@ -472,9 +475,7 @@ cifs_sfu_type(struct cifs_fattr *fattr, const char *path,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = GENERIC_READ;
-	oparms.create_options = CREATE_NOT_DIR;
-	if (backup_cred(cifs_sb))
-		oparms.create_options |= CREATE_OPEN_BACKUP_INTENT;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = path;
 	oparms.fid = &fid;
@@ -609,10 +610,8 @@ cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info,
 
 	if (info->LastAccessTime)
 		fattr->cf_atime = cifs_NTtimeToUnix(info->LastAccessTime);
-	else {
-		ktime_get_real_ts64(&fattr->cf_atime);
-		fattr->cf_atime = timespec64_trunc(fattr->cf_atime, sb->s_time_gran);
-	}
+	else
+		ktime_get_coarse_real_ts64(&fattr->cf_atime);
 
 	fattr->cf_ctime = cifs_NTtimeToUnix(info->ChangeTime);
 	fattr->cf_mtime = cifs_NTtimeToUnix(info->LastWriteTime);
@@ -654,8 +653,8 @@ cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info,
 		 */
 		if ((fattr->cf_nlink < 1) && !tcon->unix_ext &&
 		    !info->DeletePending) {
-			cifs_dbg(1, "bogus file nlink value %u\n",
-				fattr->cf_nlink);
+			cifs_dbg(VFS, "bogus file nlink value %u\n",
+				 fattr->cf_nlink);
 			fattr->cf_flags |= CIFS_FATTR_UNKNOWN_NLINK;
 		}
 	}
@@ -727,22 +726,138 @@ static __u64 simple_hashstr(const char *str)
 	return hash;
 }
 
+/**
+ * cifs_backup_query_path_info - SMB1 fallback code to get ino
+ *
+ * Fallback code to get file metadata when we don't have access to
+ * @full_path (EACCES) and have backup creds.
+ *
+ * @data will be set to search info result buffer
+ * @resp_buf will be set to cifs resp buf and needs to be freed with
+ * cifs_buf_release() when done with @data.
+ */
+static int
+cifs_backup_query_path_info(int xid,
+			    struct cifs_tcon *tcon,
+			    struct super_block *sb,
+			    const char *full_path,
+			    void **resp_buf,
+			    FILE_ALL_INFO **data)
+{
+	struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
+	struct cifs_search_info info = {0};
+	u16 flags;
+	int rc;
+
+	*resp_buf = NULL;
+	info.endOfSearch = false;
+	if (tcon->unix_ext)
+		info.info_level = SMB_FIND_FILE_UNIX;
+	else if ((tcon->ses->capabilities &
+		  tcon->ses->server->vals->cap_nt_find) == 0)
+		info.info_level = SMB_FIND_FILE_INFO_STANDARD;
+	else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)
+		info.info_level = SMB_FIND_FILE_ID_FULL_DIR_INFO;
+	else /* no srvino useful for fallback to some netapp */
+		info.info_level = SMB_FIND_FILE_DIRECTORY_INFO;
+
+	flags = CIFS_SEARCH_CLOSE_ALWAYS |
+		CIFS_SEARCH_CLOSE_AT_END |
+		CIFS_SEARCH_BACKUP_SEARCH;
+
+	rc = CIFSFindFirst(xid, tcon, full_path,
+			   cifs_sb, NULL, flags, &info, false);
+	if (rc)
+		return rc;
+
+	*resp_buf = (void *)info.ntwrk_buf_start;
+	*data = (FILE_ALL_INFO *)info.srch_entries_start;
+	return 0;
+}
+
+static void
+cifs_set_fattr_ino(int xid,
+		   struct cifs_tcon *tcon,
+		   struct super_block *sb,
+		   struct inode **inode,
+		   const char *full_path,
+		   FILE_ALL_INFO *data,
+		   struct cifs_fattr *fattr)
+{
+	struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
+	struct TCP_Server_Info *server = tcon->ses->server;
+	int rc;
+
+	if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) {
+		if (*inode)
+			fattr->cf_uniqueid = CIFS_I(*inode)->uniqueid;
+		else
+			fattr->cf_uniqueid = iunique(sb, ROOT_I);
+		return;
+	}
+
+	/*
+	 * If we have an inode pass a NULL tcon to ensure we don't
+	 * make a round trip to the server. This only works for SMB2+.
+	 */
+	rc = server->ops->get_srv_inum(xid,
+				       *inode ? NULL : tcon,
+				       cifs_sb, full_path,
+				       &fattr->cf_uniqueid,
+				       data);
+	if (rc) {
+		/*
+		 * If that fails reuse existing ino or generate one
+		 * and disable server ones
+		 */
+		if (*inode)
+			fattr->cf_uniqueid = CIFS_I(*inode)->uniqueid;
+		else {
+			fattr->cf_uniqueid = iunique(sb, ROOT_I);
+			cifs_autodisable_serverino(cifs_sb);
+		}
+		return;
+	}
+
+	/* If no errors, check for zero root inode (invalid) */
+	if (fattr->cf_uniqueid == 0 && strlen(full_path) == 0) {
+		cifs_dbg(FYI, "Invalid (0) inodenum\n");
+		if (*inode) {
+			/* reuse */
+			fattr->cf_uniqueid = CIFS_I(*inode)->uniqueid;
+		} else {
+			/* make an ino by hashing the UNC */
+			fattr->cf_flags |= CIFS_FATTR_FAKE_ROOT_INO;
+			fattr->cf_uniqueid = simple_hashstr(tcon->treeName);
+		}
+	}
+}
+
+static inline bool is_inode_cache_good(struct inode *ino)
+{
+	return ino && CIFS_CACHE_READ(CIFS_I(ino)) && CIFS_I(ino)->time != 0;
+}
+
 int
-cifs_get_inode_info(struct inode **inode, const char *full_path,
-		    FILE_ALL_INFO *data, struct super_block *sb, int xid,
+cifs_get_inode_info(struct inode **inode,
+		    const char *full_path,
+		    FILE_ALL_INFO *in_data,
+		    struct super_block *sb, int xid,
 		    const struct cifs_fid *fid)
 {
-	__u16 srchflgs;
-	int rc = 0, tmprc = ENOSYS;
+
 	struct cifs_tcon *tcon;
 	struct TCP_Server_Info *server;
 	struct tcon_link *tlink;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
-	char *buf = NULL;
 	bool adjust_tz = false;
-	struct cifs_fattr fattr;
-	struct cifs_search_info *srchinf = NULL;
+	struct cifs_fattr fattr = {0};
 	bool symlink = false;
+	FILE_ALL_INFO *data = in_data;
+	FILE_ALL_INFO *tmp_data = NULL;
+	void *smb1_backup_rsp_buf = NULL;
+	int rc = 0;
+	int tmprc = 0;
 
 	tlink = cifs_sb_tlink(cifs_sb);
 	if (IS_ERR(tlink))
@@ -750,142 +865,88 @@ cifs_get_inode_info(struct inode **inode, const char *full_path,
 	tcon = tlink_tcon(tlink);
 	server = tcon->ses->server;
 
-	cifs_dbg(FYI, "Getting info on %s\n", full_path);
+	/*
+	 * 1. Fetch file metadata if not provided (data)
+	 */
 
-	if ((data == NULL) && (*inode != NULL)) {
-		if (CIFS_CACHE_READ(CIFS_I(*inode)) &&
-		    CIFS_I(*inode)->time != 0) {
+	if (!data) {
+		if (is_inode_cache_good(*inode)) {
 			cifs_dbg(FYI, "No need to revalidate cached inode sizes\n");
-			goto cgii_exit;
-		}
-	}
-
-	/* if inode info is not passed, get it from server */
-	if (data == NULL) {
-		if (!server->ops->query_path_info) {
-			rc = -ENOSYS;
-			goto cgii_exit;
+			goto out;
 		}
-		buf = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL);
-		if (buf == NULL) {
+		tmp_data = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL);
+		if (!tmp_data) {
 			rc = -ENOMEM;
-			goto cgii_exit;
+			goto out;
 		}
-		data = (FILE_ALL_INFO *)buf;
-		rc = server->ops->query_path_info(xid, tcon, cifs_sb, full_path,
-						  data, &adjust_tz, &symlink);
+		rc = server->ops->query_path_info(xid, tcon, cifs_sb,
+						  full_path, tmp_data,
+						  &adjust_tz, &symlink);
+		data = tmp_data;
 	}
 
-	if (!rc) {
-		cifs_all_info_to_fattr(&fattr, data, sb, adjust_tz,
-				       symlink);
-	} else if (rc == -EREMOTE) {
+	/*
+	 * 2. Convert it to internal cifs metadata (fattr)
+	 */
+
+	switch (rc) {
+	case 0:
+		cifs_all_info_to_fattr(&fattr, data, sb, adjust_tz, symlink);
+		break;
+	case -EREMOTE:
+		/* DFS link, no metadata available on this server */
 		cifs_create_dfs_fattr(&fattr, sb);
 		rc = 0;
-	} else if ((rc == -EACCES) && backup_cred(cifs_sb) &&
-		   (strcmp(server->vals->version_string, SMB1_VERSION_STRING)
-		      == 0)) {
+		break;
+	case -EACCES:
 		/*
-		 * For SMB2 and later the backup intent flag is already
-		 * sent if needed on open and there is no path based
-		 * FindFirst operation to use to retry with
+		 * perm errors, try again with backup flags if possible
+		 *
+		 * For SMB2 and later the backup intent flag
+		 * is already sent if needed on open and there
+		 * is no path based FindFirst operation to use
+		 * to retry with
 		 */
+		if (backup_cred(cifs_sb) && is_smb1_server(server)) {
+			/* for easier reading */
+			FILE_DIRECTORY_INFO *fdi;
+			SEARCH_ID_FULL_DIR_INFO *si;
+
+			rc = cifs_backup_query_path_info(xid, tcon, sb,
+							 full_path,
+							 &smb1_backup_rsp_buf,
+							 &data);
+			if (rc)
+				goto out;
 
-		srchinf = kzalloc(sizeof(struct cifs_search_info),
-					GFP_KERNEL);
-		if (srchinf == NULL) {
-			rc = -ENOMEM;
-			goto cgii_exit;
+			fdi = (FILE_DIRECTORY_INFO *)data;
+			si = (SEARCH_ID_FULL_DIR_INFO *)data;
+
+			cifs_dir_info_to_fattr(&fattr, fdi, cifs_sb);
+			fattr.cf_uniqueid = le64_to_cpu(si->UniqueId);
+			/* uniqueid set, skip get inum step */
+			goto handle_mnt_opt;
+		} else {
+			/* nothing we can do, bail out */
+			goto out;
 		}
+		break;
+	default:
+		cifs_dbg(FYI, "%s: unhandled err rc %d\n", __func__, rc);
+		goto out;
+	}
 
-		srchinf->endOfSearch = false;
-		if (tcon->unix_ext)
-			srchinf->info_level = SMB_FIND_FILE_UNIX;
-		else if ((tcon->ses->capabilities &
-			 tcon->ses->server->vals->cap_nt_find) == 0)
-			srchinf->info_level = SMB_FIND_FILE_INFO_STANDARD;
-		else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)
-			srchinf->info_level = SMB_FIND_FILE_ID_FULL_DIR_INFO;
-		else /* no srvino useful for fallback to some netapp */
-			srchinf->info_level = SMB_FIND_FILE_DIRECTORY_INFO;
-
-		srchflgs = CIFS_SEARCH_CLOSE_ALWAYS |
-				CIFS_SEARCH_CLOSE_AT_END |
-				CIFS_SEARCH_BACKUP_SEARCH;
-
-		rc = CIFSFindFirst(xid, tcon, full_path,
-			cifs_sb, NULL, srchflgs, srchinf, false);
-		if (!rc) {
-			data = (FILE_ALL_INFO *)srchinf->srch_entries_start;
+	/*
+	 * 3. Get or update inode number (fattr.cf_uniqueid)
+	 */
 
-			cifs_dir_info_to_fattr(&fattr,
-			(FILE_DIRECTORY_INFO *)data, cifs_sb);
-			fattr.cf_uniqueid = le64_to_cpu(
-			((SEARCH_ID_FULL_DIR_INFO *)data)->UniqueId);
-
-			cifs_buf_release(srchinf->ntwrk_buf_start);
-		}
-		kfree(srchinf);
-		if (rc)
-			goto cgii_exit;
-	} else
-		goto cgii_exit;
+	cifs_set_fattr_ino(xid, tcon, sb, inode, full_path, data, &fattr);
 
 	/*
-	 * If an inode wasn't passed in, then get the inode number
-	 *
-	 * Is an i_ino of zero legal? Can we use that to check if the server
-	 * supports returning inode numbers?  Are there other sanity checks we
-	 * can use to ensure that the server is really filling in that field?
+	 * 4. Tweak fattr based on mount options
 	 */
-	if (*inode == NULL) {
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) {
-			if (server->ops->get_srv_inum)
-				tmprc = server->ops->get_srv_inum(xid,
-								  tcon, cifs_sb, full_path,
-								  &fattr.cf_uniqueid, data);
-			if (tmprc) {
-				cifs_dbg(FYI, "GetSrvInodeNum rc %d\n",
-					 tmprc);
-				fattr.cf_uniqueid = iunique(sb, ROOT_I);
-				cifs_autodisable_serverino(cifs_sb);
-			} else if ((fattr.cf_uniqueid == 0) &&
-				   strlen(full_path) == 0) {
-				/* some servers ret bad root ino ie 0 */
-				cifs_dbg(FYI, "Invalid (0) inodenum\n");
-				fattr.cf_flags |=
-					CIFS_FATTR_FAKE_ROOT_INO;
-				fattr.cf_uniqueid =
-					simple_hashstr(tcon->treeName);
-			}
-		} else
-			fattr.cf_uniqueid = iunique(sb, ROOT_I);
-	} else {
-		if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)
-		    && server->ops->get_srv_inum) {
-			/*
-			 * Pass a NULL tcon to ensure we don't make a round
-			 * trip to the server. This only works for SMB2+.
-			 */
-			tmprc = server->ops->get_srv_inum(xid,
-				NULL, cifs_sb, full_path,
-				&fattr.cf_uniqueid, data);
-			if (tmprc)
-				fattr.cf_uniqueid = CIFS_I(*inode)->uniqueid;
-			else if ((fattr.cf_uniqueid == 0) &&
-					strlen(full_path) == 0) {
-				/*
-				 * Reuse existing root inode num since
-				 * inum zero for root causes ls of . and .. to
-				 * not be returned
-				 */
-				cifs_dbg(FYI, "Srv ret 0 inode num for root\n");
-				fattr.cf_uniqueid = CIFS_I(*inode)->uniqueid;
-			}
-		} else
-			fattr.cf_uniqueid = CIFS_I(*inode)->uniqueid;
-	}
 
+handle_mnt_opt:
 	/* query for SFU type info if supported and needed */
 	if (fattr.cf_cifsattrs & ATTR_SYSTEM &&
 	    cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) {
@@ -900,8 +961,8 @@ cifs_get_inode_info(struct inode **inode, const char *full_path,
 				       full_path, fid);
 		if (rc) {
 			cifs_dbg(FYI, "%s: Get mode from SID failed. rc=%d\n",
-				__func__, rc);
-			goto cgii_exit;
+				 __func__, rc);
+			goto out;
 		}
 	} else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) {
 		rc = cifs_acl_to_fattr(cifs_sb, &fattr, *inode, false,
@@ -909,7 +970,7 @@ cifs_get_inode_info(struct inode **inode, const char *full_path,
 		if (rc) {
 			cifs_dbg(FYI, "%s: Getting ACL failed with error: %d\n",
 				 __func__, rc);
-			goto cgii_exit;
+			goto out;
 		}
 	}
 
@@ -925,6 +986,10 @@ cifs_get_inode_info(struct inode **inode, const char *full_path,
 			cifs_dbg(FYI, "check_mf_symlink: %d\n", tmprc);
 	}
 
+	/*
+	 * 5. Update inode with final fattr data
+	 */
+
 	if (!*inode) {
 		*inode = cifs_iget(sb, &fattr);
 		if (!*inode)
@@ -937,7 +1002,7 @@ cifs_get_inode_info(struct inode **inode, const char *full_path,
 		    CIFS_I(*inode)->uniqueid != fattr.cf_uniqueid)) {
 			CIFS_I(*inode)->time = 0; /* force reval */
 			rc = -ESTALE;
-			goto cgii_exit;
+			goto out;
 		}
 
 		/* if filetype is different, return error */
@@ -945,18 +1010,15 @@ cifs_get_inode_info(struct inode **inode, const char *full_path,
 		    (fattr.cf_mode & S_IFMT))) {
 			CIFS_I(*inode)->time = 0; /* force reval */
 			rc = -ESTALE;
-			goto cgii_exit;
+			goto out;
 		}
 
 		cifs_fattr_to_inode(*inode, &fattr);
 	}
-
-cgii_exit:
-	if ((*inode) && ((*inode)->i_ino == 0))
-		cifs_dbg(FYI, "inode number of zero returned\n");
-
-	kfree(buf);
+out:
+	cifs_buf_release(smb1_backup_rsp_buf);
 	cifs_put_tlink(tlink);
+	kfree(tmp_data);
 	return rc;
 }
 
@@ -1221,7 +1283,7 @@ cifs_rename_pending_delete(const char *full_path, struct dentry *dentry,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = DELETE | FILE_WRITE_ATTRIBUTES;
-	oparms.create_options = CREATE_NOT_DIR;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = full_path;
 	oparms.fid = &fid;
@@ -1586,7 +1648,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, umode_t mode)
 	struct TCP_Server_Info *server;
 	char *full_path;
 
-	cifs_dbg(FYI, "In cifs_mkdir, mode = 0x%hx inode = 0x%p\n",
+	cifs_dbg(FYI, "In cifs_mkdir, mode = %04ho inode = 0x%p\n",
 		 mode, inode);
 
 	cifs_sb = CIFS_SB(inode->i_sb);
@@ -1759,7 +1821,7 @@ cifs_do_rename(const unsigned int xid, struct dentry *from_dentry,
 	oparms.cifs_sb = cifs_sb;
 	/* open the file to be renamed -- we need DELETE perms */
 	oparms.desired_access = DELETE;
-	oparms.create_options = CREATE_NOT_DIR;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = from_path;
 	oparms.fid = &fid;
@@ -1773,6 +1835,8 @@ cifs_do_rename(const unsigned int xid, struct dentry *from_dentry,
 		CIFSSMBClose(xid, tcon, fid.netfid);
 	}
 do_rename_exit:
+	if (rc == 0)
+		d_move(from_dentry, to_dentry);
 	cifs_put_tlink(tlink);
 	return rc;
 }
@@ -1962,6 +2026,10 @@ cifs_revalidate_mapping(struct inode *inode)
 	int rc;
 	unsigned long *flags = &CIFS_I(inode)->flags;
 
+	/* swapfiles are not supposed to be shared */
+	if (IS_SWAPFILE(inode))
+		return 0;
+
 	rc = wait_on_bit_lock_action(flags, CIFS_INO_LOCK, cifs_wait_bit_killable,
 				     TASK_KILLABLE);
 	if (rc)
@@ -2011,6 +2079,7 @@ int cifs_revalidate_dentry_attr(struct dentry *dentry)
 	struct inode *inode = d_inode(dentry);
 	struct super_block *sb = dentry->d_sb;
 	char *full_path = NULL;
+	int count = 0;
 
 	if (inode == NULL)
 		return -ENOENT;
@@ -2032,15 +2101,18 @@ int cifs_revalidate_dentry_attr(struct dentry *dentry)
 		 full_path, inode, inode->i_count.counter,
 		 dentry, cifs_get_time(dentry), jiffies);
 
+again:
 	if (cifs_sb_master_tcon(CIFS_SB(sb))->unix_ext)
 		rc = cifs_get_inode_info_unix(&inode, full_path, sb, xid);
 	else
 		rc = cifs_get_inode_info(&inode, full_path, NULL, sb,
 					 xid, NULL);
-
+	if (rc == -EAGAIN && count++ < 10)
+		goto again;
 out:
 	kfree(full_path);
 	free_xid(xid);
+
 	return rc;
 }
 
@@ -2082,8 +2154,9 @@ int cifs_getattr(const struct path *path, struct kstat *stat,
 	 * We need to be sure that all dirty pages are written and the server
 	 * has actual ctime, mtime and file length.
 	 */
-	if (!CIFS_CACHE_READ(CIFS_I(inode)) && inode->i_mapping &&
-	    inode->i_mapping->nrpages != 0) {
+	if ((request_mask & (STATX_CTIME | STATX_MTIME | STATX_SIZE)) &&
+	    !CIFS_CACHE_READ(CIFS_I(inode)) &&
+	    inode->i_mapping && inode->i_mapping->nrpages != 0) {
 		rc = filemap_fdatawait(inode->i_mapping);
 		if (rc) {
 			mapping_set_error(inode->i_mapping, rc);
@@ -2091,9 +2164,20 @@ int cifs_getattr(const struct path *path, struct kstat *stat,
 		}
 	}
 
-	rc = cifs_revalidate_dentry_attr(dentry);
-	if (rc)
-		return rc;
+	if ((flags & AT_STATX_SYNC_TYPE) == AT_STATX_FORCE_SYNC)
+		CIFS_I(inode)->time = 0; /* force revalidate */
+
+	/*
+	 * If the caller doesn't require syncing, only sync if
+	 * necessary (e.g. due to earlier truncate or setattr
+	 * invalidating the cached metadata)
+	 */
+	if (((flags & AT_STATX_SYNC_TYPE) != AT_STATX_DONT_SYNC) ||
+	    (CIFS_I(inode)->time == 0)) {
+		rc = cifs_revalidate_dentry_attr(dentry);
+		if (rc)
+			return rc;
+	}
 
 	generic_fillattr(inode, stat);
 	stat->blksize = cifs_sb->bsize;
@@ -2125,7 +2209,7 @@ int cifs_getattr(const struct path *path, struct kstat *stat,
 		if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID))
 			stat->gid = current_fsgid();
 	}
-	return rc;
+	return 0;
 }
 
 int cifs_fiemap(struct inode *inode, struct fiemap_extent_info *fei, u64 start,
@@ -2165,7 +2249,7 @@ int cifs_fiemap(struct inode *inode, struct fiemap_extent_info *fei, u64 start,
 	return -ENOTSUPP;
 }
 
-static int cifs_truncate_page(struct address_space *mapping, loff_t from)
+int cifs_truncate_page(struct address_space *mapping, loff_t from)
 {
 	pgoff_t index = from >> PAGE_SHIFT;
 	unsigned offset = from & (PAGE_SIZE - 1);
@@ -2182,7 +2266,7 @@ static int cifs_truncate_page(struct address_space *mapping, loff_t from)
 	return rc;
 }
 
-static void cifs_setsize(struct inode *inode, loff_t offset)
+void cifs_setsize(struct inode *inode, loff_t offset)
 {
 	struct cifsInodeInfo *cifs_i = CIFS_I(inode);
 
@@ -2216,7 +2300,7 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs,
 	 * writebehind data than the SMB timeout for the SetPathInfo
 	 * request would allow
 	 */
-	open_file = find_writable_file(cifsInode, true);
+	open_file = find_writable_file(cifsInode, FIND_WR_FSUID_ONLY);
 	if (open_file) {
 		tcon = tlink_tcon(open_file->tlink);
 		server = tcon->ses->server;
@@ -2366,7 +2450,7 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs)
 		args->ctime = NO_CHANGE_64;
 
 	args->device = 0;
-	open_file = find_writable_file(cifsInode, true);
+	open_file = find_writable_file(cifsInode, FIND_WR_FSUID_ONLY);
 	if (open_file) {
 		u16 nfid = open_file->fid.netfid;
 		u32 npid = open_file->pid;
@@ -2450,26 +2534,27 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs)
 
 	/*
 	 * Attempt to flush data before changing attributes. We need to do
-	 * this for ATTR_SIZE and ATTR_MTIME for sure, and if we change the
-	 * ownership or mode then we may also need to do this. Here, we take
-	 * the safe way out and just do the flush on all setattr requests. If
-	 * the flush returns error, store it to report later and continue.
+	 * this for ATTR_SIZE and ATTR_MTIME.  If the flush of the data
+	 * returns error, store it to report later and continue.
 	 *
 	 * BB: This should be smarter. Why bother flushing pages that
 	 * will be truncated anyway? Also, should we error out here if
-	 * the flush returns error?
+	 * the flush returns error? Do we need to check for ATTR_MTIME_SET flag?
 	 */
-	rc = filemap_write_and_wait(inode->i_mapping);
-	if (is_interrupt_error(rc)) {
-		rc = -ERESTARTSYS;
-		goto cifs_setattr_exit;
+	if (attrs->ia_valid & (ATTR_MTIME | ATTR_SIZE | ATTR_CTIME)) {
+		rc = filemap_write_and_wait(inode->i_mapping);
+		if (is_interrupt_error(rc)) {
+			rc = -ERESTARTSYS;
+			goto cifs_setattr_exit;
+		}
+		mapping_set_error(inode->i_mapping, rc);
 	}
 
-	mapping_set_error(inode->i_mapping, rc);
 	rc = 0;
 
-	if (attrs->ia_valid & ATTR_MTIME) {
-		rc = cifs_get_writable_file(cifsInode, false, &wfile);
+	if ((attrs->ia_valid & ATTR_MTIME) &&
+	    !(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC)) {
+		rc = cifs_get_writable_file(cifsInode, FIND_WR_ANY, &wfile);
 		if (!rc) {
 			tcon = tlink_tcon(wfile->tlink);
 			rc = tcon->ses->server->ops->flush(xid, tcon, &wfile->fid);
diff --git a/fs/cifs/ioctl.c b/fs/cifs/ioctl.c
index 1a01e108d75e..4a73e63c4d43 100644
--- a/fs/cifs/ioctl.c
+++ b/fs/cifs/ioctl.c
@@ -65,7 +65,7 @@ static long cifs_ioctl_query_info(unsigned int xid, struct file *filep,
 
 	if (tcon->ses->server->ops->ioctl_query_info)
 		rc = tcon->ses->server->ops->ioctl_query_info(
-				xid, tcon, utf16_path,
+				xid, tcon, cifs_sb, utf16_path,
 				filep->private_data ? 0 : 1, p);
 	else
 		rc = -EOPNOTSUPP;
@@ -169,6 +169,7 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg)
 	unsigned int xid;
 	struct cifsFileInfo *pSMBFile = filep->private_data;
 	struct cifs_tcon *tcon;
+	struct cifs_sb_info *cifs_sb;
 	__u64	ExtAttrBits = 0;
 	__u64   caps;
 
@@ -299,6 +300,21 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg)
 			else
 				rc = 0;
 			break;
+		case CIFS_IOC_NOTIFY:
+			if (!S_ISDIR(inode->i_mode)) {
+				/* Notify can only be done on directories */
+				rc = -EOPNOTSUPP;
+				break;
+			}
+			cifs_sb = CIFS_SB(inode->i_sb);
+			tcon = tlink_tcon(cifs_sb_tlink(cifs_sb));
+			if (tcon && tcon->ses->server->ops->notify) {
+				rc = tcon->ses->server->ops->notify(xid,
+						filep, (void __user *)arg);
+				cifs_dbg(FYI, "ioctl notify rc %d\n", rc);
+			} else
+				rc = -EOPNOTSUPP;
+			break;
 		default:
 			cifs_dbg(FYI, "unsupported ioctl\n");
 			break;
diff --git a/fs/cifs/link.c b/fs/cifs/link.c
index b736acd3917b..a25ef35b023e 100644
--- a/fs/cifs/link.c
+++ b/fs/cifs/link.c
@@ -315,7 +315,7 @@ cifs_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = GENERIC_READ;
-	oparms.create_options = CREATE_NOT_DIR;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = path;
 	oparms.fid = &fid;
@@ -353,15 +353,11 @@ cifs_create_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 	struct cifs_fid fid;
 	struct cifs_open_parms oparms;
 	struct cifs_io_parms io_parms;
-	int create_options = CREATE_NOT_DIR;
-
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
 
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = GENERIC_WRITE;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR);
 	oparms.disposition = FILE_CREATE;
 	oparms.path = path;
 	oparms.fid = &fid;
@@ -402,9 +398,7 @@ smb3_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = GENERIC_READ;
-	oparms.create_options = CREATE_NOT_DIR;
-	if (backup_cred(cifs_sb))
-		oparms.create_options |= CREATE_OPEN_BACKUP_INTENT;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR);
 	oparms.disposition = FILE_OPEN;
 	oparms.fid = &fid;
 	oparms.reconnect = false;
@@ -422,7 +416,7 @@ smb3_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 	}
 
 	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, pfile_info, NULL,
-		       NULL);
+		       NULL, NULL);
 	if (rc)
 		goto qmf_out_open_fail;
 
@@ -457,14 +451,10 @@ smb3_create_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 	struct cifs_fid fid;
 	struct cifs_open_parms oparms;
 	struct cifs_io_parms io_parms;
-	int create_options = CREATE_NOT_DIR;
 	__le16 *utf16_path;
 	__u8 oplock = SMB2_OPLOCK_LEVEL_NONE;
 	struct kvec iov[2];
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
-
 	cifs_dbg(FYI, "%s: path: %s\n", __func__, path);
 
 	utf16_path = cifs_convert_path_to_utf16(path, cifs_sb);
@@ -474,13 +464,13 @@ smb3_create_mf_symlink(unsigned int xid, struct cifs_tcon *tcon,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = GENERIC_WRITE;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR);
 	oparms.disposition = FILE_CREATE;
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
 	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL,
-		       NULL);
+		       NULL, NULL);
 	if (rc) {
 		kfree(utf16_path);
 		return rc;
diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c
index 5ad83bdb9bea..550ce9020a3e 100644
--- a/fs/cifs/misc.c
+++ b/fs/cifs/misc.c
@@ -31,6 +31,7 @@
 #include "nterr.h"
 #include "cifs_unicode.h"
 #include "smb2pdu.h"
+#include "cifsfs.h"
 
 extern mempool_t *cifs_sm_req_poolp;
 extern mempool_t *cifs_req_poolp;
@@ -488,21 +489,10 @@ is_valid_oplock_break(char *buffer, struct TCP_Server_Info *srv)
 				set_bit(CIFS_INODE_PENDING_OPLOCK_BREAK,
 					&pCifsInode->flags);
 
-				/*
-				 * Set flag if the server downgrades the oplock
-				 * to L2 else clear.
-				 */
-				if (pSMB->OplockLevel)
-					set_bit(
-					   CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2,
-					   &pCifsInode->flags);
-				else
-					clear_bit(
-					   CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2,
-					   &pCifsInode->flags);
-
-				cifs_queue_oplock_break(netfile);
+				netfile->oplock_epoch = 0;
+				netfile->oplock_level = pSMB->OplockLevel;
 				netfile->oplock_break_cancelled = false;
+				cifs_queue_oplock_break(netfile);
 
 				spin_unlock(&tcon->open_file_lock);
 				spin_unlock(&cifs_tcp_ses_lock);
@@ -1033,3 +1023,130 @@ int copy_path_name(char *dst, const char *src)
 	name_len++;
 	return name_len;
 }
+
+struct super_cb_data {
+	void *data;
+	struct super_block *sb;
+};
+
+static void tcp_super_cb(struct super_block *sb, void *arg)
+{
+	struct super_cb_data *sd = arg;
+	struct TCP_Server_Info *server = sd->data;
+	struct cifs_sb_info *cifs_sb;
+	struct cifs_tcon *tcon;
+
+	if (sd->sb)
+		return;
+
+	cifs_sb = CIFS_SB(sb);
+	tcon = cifs_sb_master_tcon(cifs_sb);
+	if (tcon->ses->server == server)
+		sd->sb = sb;
+}
+
+static struct super_block *__cifs_get_super(void (*f)(struct super_block *, void *),
+					    void *data)
+{
+	struct super_cb_data sd = {
+		.data = data,
+		.sb = NULL,
+	};
+
+	iterate_supers_type(&cifs_fs_type, f, &sd);
+
+	if (!sd.sb)
+		return ERR_PTR(-EINVAL);
+	/*
+	 * Grab an active reference in order to prevent automounts (DFS links)
+	 * of expiring and then freeing up our cifs superblock pointer while
+	 * we're doing failover.
+	 */
+	cifs_sb_active(sd.sb);
+	return sd.sb;
+}
+
+static void __cifs_put_super(struct super_block *sb)
+{
+	if (!IS_ERR_OR_NULL(sb))
+		cifs_sb_deactive(sb);
+}
+
+struct super_block *cifs_get_tcp_super(struct TCP_Server_Info *server)
+{
+	return __cifs_get_super(tcp_super_cb, server);
+}
+
+void cifs_put_tcp_super(struct super_block *sb)
+{
+	__cifs_put_super(sb);
+}
+
+#ifdef CONFIG_CIFS_DFS_UPCALL
+static void tcon_super_cb(struct super_block *sb, void *arg)
+{
+	struct super_cb_data *sd = arg;
+	struct cifs_tcon *tcon = sd->data;
+	struct cifs_sb_info *cifs_sb;
+
+	if (sd->sb)
+		return;
+
+	cifs_sb = CIFS_SB(sb);
+	if (tcon->dfs_path && cifs_sb->origin_fullpath &&
+	    !strcasecmp(tcon->dfs_path, cifs_sb->origin_fullpath))
+		sd->sb = sb;
+}
+
+static inline struct super_block *cifs_get_tcon_super(struct cifs_tcon *tcon)
+{
+	return __cifs_get_super(tcon_super_cb, tcon);
+}
+
+static inline void cifs_put_tcon_super(struct super_block *sb)
+{
+	__cifs_put_super(sb);
+}
+#else
+static inline struct super_block *cifs_get_tcon_super(struct cifs_tcon *tcon)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+
+static inline void cifs_put_tcon_super(struct super_block *sb)
+{
+}
+#endif
+
+int update_super_prepath(struct cifs_tcon *tcon, const char *prefix,
+			 size_t prefix_len)
+{
+	struct super_block *sb;
+	struct cifs_sb_info *cifs_sb;
+	int rc = 0;
+
+	sb = cifs_get_tcon_super(tcon);
+	if (IS_ERR(sb))
+		return PTR_ERR(sb);
+
+	cifs_sb = CIFS_SB(sb);
+
+	kfree(cifs_sb->prepath);
+
+	if (*prefix && prefix_len) {
+		cifs_sb->prepath = kstrndup(prefix, prefix_len, GFP_ATOMIC);
+		if (!cifs_sb->prepath) {
+			rc = -ENOMEM;
+			goto out;
+		}
+
+		convert_delimiter(cifs_sb->prepath, CIFS_DIR_SEP(cifs_sb));
+	} else
+		cifs_sb->prepath = NULL;
+
+	cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH;
+
+out:
+	cifs_put_tcon_super(sb);
+	return rc;
+}
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 3925a7bfc74d..50f776a8d4ba 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -32,6 +32,7 @@
 #include "cifs_debug.h"
 #include "cifs_fs_sb.h"
 #include "cifsfs.h"
+#include "smb2proto.h"
 
 /*
  * To be safe - for UCS to UTF-8 with strings loaded with the rare long
@@ -139,6 +140,28 @@ retry:
 	dput(dentry);
 }
 
+static bool reparse_file_needs_reval(const struct cifs_fattr *fattr)
+{
+	if (!(fattr->cf_cifsattrs & ATTR_REPARSE))
+		return false;
+	/*
+	 * The DFS tags should be only intepreted by server side as per
+	 * MS-FSCC 2.1.2.1, but let's include them anyway.
+	 *
+	 * Besides, if cf_cifstag is unset (0), then we still need it to be
+	 * revalidated to know exactly what reparse point it is.
+	 */
+	switch (fattr->cf_cifstag) {
+	case IO_REPARSE_TAG_DFS:
+	case IO_REPARSE_TAG_DFSR:
+	case IO_REPARSE_TAG_SYMLINK:
+	case IO_REPARSE_TAG_NFS:
+	case 0:
+		return true;
+	}
+	return false;
+}
+
 static void
 cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 {
@@ -158,7 +181,7 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 	 * is a symbolic link, DFS referral or a reparse point with a direct
 	 * access like junctions, deduplicated files, NFS symlinks.
 	 */
-	if (fattr->cf_cifsattrs & ATTR_REPARSE)
+	if (reparse_file_needs_reval(fattr))
 		fattr->cf_flags |= CIFS_FATTR_NEED_REVAL;
 
 	/* non-unix readdir doesn't provide nlink */
@@ -174,7 +197,8 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 	 * may look wrong since the inodes may not have timed out by the time
 	 * "ls" does a stat() call on them.
 	 */
-	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL)
+	if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) ||
+	    (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID))
 		fattr->cf_flags |= CIFS_FATTR_NEED_REVAL;
 
 	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL &&
@@ -194,19 +218,91 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 	}
 }
 
-void
-cifs_dir_info_to_fattr(struct cifs_fattr *fattr, FILE_DIRECTORY_INFO *info,
-		       struct cifs_sb_info *cifs_sb)
+/* Fill a cifs_fattr struct with info from SMB_FIND_FILE_POSIX_INFO. */
+static void
+cifs_posix_to_fattr(struct cifs_fattr *fattr, struct smb2_posix_info *info,
+		    struct cifs_sb_info *cifs_sb)
 {
+	struct smb2_posix_info_parsed parsed;
+
+	posix_info_parse(info, NULL, &parsed);
+
 	memset(fattr, 0, sizeof(*fattr));
-	fattr->cf_cifsattrs = le32_to_cpu(info->ExtFileAttributes);
-	fattr->cf_eof = le64_to_cpu(info->EndOfFile);
+	fattr->cf_uniqueid = le64_to_cpu(info->Inode);
 	fattr->cf_bytes = le64_to_cpu(info->AllocationSize);
-	fattr->cf_createtime = le64_to_cpu(info->CreationTime);
+	fattr->cf_eof = le64_to_cpu(info->EndOfFile);
+
 	fattr->cf_atime = cifs_NTtimeToUnix(info->LastAccessTime);
-	fattr->cf_ctime = cifs_NTtimeToUnix(info->ChangeTime);
 	fattr->cf_mtime = cifs_NTtimeToUnix(info->LastWriteTime);
+	fattr->cf_ctime = cifs_NTtimeToUnix(info->CreationTime);
+
+	fattr->cf_nlink = le32_to_cpu(info->HardLinks);
+	fattr->cf_cifsattrs = le32_to_cpu(info->DosAttributes);
+
+	/*
+	 * Since we set the inode type below we need to mask off
+	 * to avoid strange results if bits set above.
+	 * XXX: why not make server&client use the type bits?
+	 */
+	fattr->cf_mode = le32_to_cpu(info->Mode) & ~S_IFMT;
+
+	cifs_dbg(FYI, "posix fattr: dev %d, reparse %d, mode %o",
+		 le32_to_cpu(info->DeviceId),
+		 le32_to_cpu(info->ReparseTag),
+		 le32_to_cpu(info->Mode));
+
+	if (fattr->cf_cifsattrs & ATTR_DIRECTORY) {
+		fattr->cf_mode |= S_IFDIR;
+		fattr->cf_dtype = DT_DIR;
+	} else {
+		/*
+		 * mark anything that is not a dir as regular
+		 * file. special files should have the REPARSE
+		 * attribute and will be marked as needing revaluation
+		 */
+		fattr->cf_mode |= S_IFREG;
+		fattr->cf_dtype = DT_REG;
+	}
+
+	if (reparse_file_needs_reval(fattr))
+		fattr->cf_flags |= CIFS_FATTR_NEED_REVAL;
 
+	/* TODO map SIDs */
+	fattr->cf_uid = cifs_sb->mnt_uid;
+	fattr->cf_gid = cifs_sb->mnt_gid;
+}
+
+static void __dir_info_to_fattr(struct cifs_fattr *fattr, const void *info)
+{
+	const FILE_DIRECTORY_INFO *fi = info;
+
+	memset(fattr, 0, sizeof(*fattr));
+	fattr->cf_cifsattrs = le32_to_cpu(fi->ExtFileAttributes);
+	fattr->cf_eof = le64_to_cpu(fi->EndOfFile);
+	fattr->cf_bytes = le64_to_cpu(fi->AllocationSize);
+	fattr->cf_createtime = le64_to_cpu(fi->CreationTime);
+	fattr->cf_atime = cifs_NTtimeToUnix(fi->LastAccessTime);
+	fattr->cf_ctime = cifs_NTtimeToUnix(fi->ChangeTime);
+	fattr->cf_mtime = cifs_NTtimeToUnix(fi->LastWriteTime);
+}
+
+void
+cifs_dir_info_to_fattr(struct cifs_fattr *fattr, FILE_DIRECTORY_INFO *info,
+		       struct cifs_sb_info *cifs_sb)
+{
+	__dir_info_to_fattr(fattr, info);
+	cifs_fill_common_info(fattr, cifs_sb);
+}
+
+static void cifs_fulldir_info_to_fattr(struct cifs_fattr *fattr,
+				       SEARCH_ID_FULL_DIR_INFO *info,
+				       struct cifs_sb_info *cifs_sb)
+{
+	__dir_info_to_fattr(fattr, info);
+
+	/* See MS-FSCC 2.4.18 FileIdFullDirectoryInformation */
+	if (fattr->cf_cifsattrs & ATTR_REPARSE)
+		fattr->cf_cifstag = le32_to_cpu(info->EaSize);
 	cifs_fill_common_info(fattr, cifs_sb);
 }
 
@@ -318,6 +414,8 @@ ffirst_retry:
 	/* if (cap_unix(tcon->ses) { */
 	if (tcon->unix_ext)
 		cifsFile->srch_inf.info_level = SMB_FIND_FILE_UNIX;
+	else if (tcon->posix_extensions)
+		cifsFile->srch_inf.info_level = SMB_FIND_FILE_POSIX_INFO;
 	else if ((tcon->ses->capabilities &
 		  tcon->ses->server->vals->cap_nt_find) == 0) {
 		cifsFile->srch_inf.info_level = SMB_FIND_FILE_INFO_STANDARD;
@@ -410,6 +508,23 @@ struct cifs_dirent {
 	u64		ino;
 };
 
+static void cifs_fill_dirent_posix(struct cifs_dirent *de,
+				   const struct smb2_posix_info *info)
+{
+	struct smb2_posix_info_parsed parsed;
+
+	/* payload should have already been checked at this point */
+	if (posix_info_parse(info, NULL, &parsed) < 0) {
+		cifs_dbg(VFS, "invalid POSIX info payload");
+		return;
+	}
+
+	de->name = parsed.name;
+	de->namelen = parsed.name_len;
+	de->resume_key = info->Ignored;
+	de->ino = le64_to_cpu(info->Inode);
+}
+
 static void cifs_fill_dirent_unix(struct cifs_dirent *de,
 		const FILE_UNIX_INFO *info, bool is_unicode)
 {
@@ -470,6 +585,9 @@ static int cifs_fill_dirent(struct cifs_dirent *de, const void *info,
 	memset(de, 0, sizeof(*de));
 
 	switch (level) {
+	case SMB_FIND_FILE_POSIX_INFO:
+		cifs_fill_dirent_posix(de, info);
+		break;
 	case SMB_FIND_FILE_UNIX:
 		cifs_fill_dirent_unix(de, info, is_unicode);
 		break;
@@ -745,6 +863,11 @@ static int cifs_filldir(char *find_entry, struct file *file,
 	}
 
 	switch (file_info->srch_inf.info_level) {
+	case SMB_FIND_FILE_POSIX_INFO:
+		cifs_posix_to_fattr(&fattr,
+				    (struct smb2_posix_info *)find_entry,
+				    cifs_sb);
+		break;
 	case SMB_FIND_FILE_UNIX:
 		cifs_unix_basic_to_fattr(&fattr,
 					 &((FILE_UNIX_INFO *)find_entry)->basic,
@@ -755,6 +878,11 @@ static int cifs_filldir(char *find_entry, struct file *file,
 				       (FIND_FILE_STANDARD_INFO *)find_entry,
 				       cifs_sb);
 		break;
+	case SMB_FIND_FILE_ID_FULL_DIR_INFO:
+		cifs_fulldir_info_to_fattr(&fattr,
+					   (SEARCH_ID_FULL_DIR_INFO *)find_entry,
+					   cifs_sb);
+		break;
 	default:
 		cifs_dir_info_to_fattr(&fattr,
 				       (FILE_DIRECTORY_INFO *)find_entry,
diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c
index 85bd644f9773..43a88e26d26b 100644
--- a/fs/cifs/sess.c
+++ b/fs/cifs/sess.c
@@ -31,6 +31,257 @@
 #include <linux/utsname.h>
 #include <linux/slab.h>
 #include "cifs_spnego.h"
+#include "smb2proto.h"
+
+bool
+is_server_using_iface(struct TCP_Server_Info *server,
+		      struct cifs_server_iface *iface)
+{
+	struct sockaddr_in *i4 = (struct sockaddr_in *)&iface->sockaddr;
+	struct sockaddr_in6 *i6 = (struct sockaddr_in6 *)&iface->sockaddr;
+	struct sockaddr_in *s4 = (struct sockaddr_in *)&server->dstaddr;
+	struct sockaddr_in6 *s6 = (struct sockaddr_in6 *)&server->dstaddr;
+
+	if (server->dstaddr.ss_family != iface->sockaddr.ss_family)
+		return false;
+	if (server->dstaddr.ss_family == AF_INET) {
+		if (s4->sin_addr.s_addr != i4->sin_addr.s_addr)
+			return false;
+	} else if (server->dstaddr.ss_family == AF_INET6) {
+		if (memcmp(&s6->sin6_addr, &i6->sin6_addr,
+			   sizeof(i6->sin6_addr)) != 0)
+			return false;
+	} else {
+		/* unknown family.. */
+		return false;
+	}
+	return true;
+}
+
+bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface)
+{
+	int i;
+
+	for (i = 0; i < ses->chan_count; i++) {
+		if (is_server_using_iface(ses->chans[i].server, iface))
+			return true;
+	}
+	return false;
+}
+
+/* returns number of channels added */
+int cifs_try_adding_channels(struct cifs_ses *ses)
+{
+	int old_chan_count = ses->chan_count;
+	int left = ses->chan_max - ses->chan_count;
+	int i = 0;
+	int rc = 0;
+	int tries = 0;
+	struct cifs_server_iface *ifaces = NULL;
+	size_t iface_count;
+
+	if (left <= 0) {
+		cifs_dbg(FYI,
+			 "ses already at max_channels (%zu), nothing to open\n",
+			 ses->chan_max);
+		return 0;
+	}
+
+	if (ses->server->dialect < SMB30_PROT_ID) {
+		cifs_dbg(VFS, "multichannel is not supported on this protocol version, use 3.0 or above\n");
+		return 0;
+	}
+
+	/*
+	 * Make a copy of the iface list at the time and use that
+	 * instead so as to not hold the iface spinlock for opening
+	 * channels
+	 */
+	spin_lock(&ses->iface_lock);
+	iface_count = ses->iface_count;
+	if (iface_count <= 0) {
+		spin_unlock(&ses->iface_lock);
+		cifs_dbg(VFS, "no iface list available to open channels\n");
+		return 0;
+	}
+	ifaces = kmemdup(ses->iface_list, iface_count*sizeof(*ifaces),
+			 GFP_ATOMIC);
+	if (!ifaces) {
+		spin_unlock(&ses->iface_lock);
+		return 0;
+	}
+	spin_unlock(&ses->iface_lock);
+
+	/*
+	 * Keep connecting to same, fastest, iface for all channels as
+	 * long as its RSS. Try next fastest one if not RSS or channel
+	 * creation fails.
+	 */
+	while (left > 0) {
+		struct cifs_server_iface *iface;
+
+		tries++;
+		if (tries > 3*ses->chan_max) {
+			cifs_dbg(FYI, "too many attempt at opening channels (%d channels left to open)\n",
+				 left);
+			break;
+		}
+
+		iface = &ifaces[i];
+		if (is_ses_using_iface(ses, iface) && !iface->rss_capable) {
+			i = (i+1) % iface_count;
+			continue;
+		}
+
+		rc = cifs_ses_add_channel(ses, iface);
+		if (rc) {
+			cifs_dbg(FYI, "failed to open extra channel on iface#%d rc=%d\n",
+				 i, rc);
+			i = (i+1) % iface_count;
+			continue;
+		}
+
+		cifs_dbg(FYI, "successfully opened new channel on iface#%d\n",
+			 i);
+		left--;
+	}
+
+	kfree(ifaces);
+	return ses->chan_count - old_chan_count;
+}
+
+int
+cifs_ses_add_channel(struct cifs_ses *ses, struct cifs_server_iface *iface)
+{
+	struct cifs_chan *chan;
+	struct smb_vol vol = {NULL};
+	static const char unc_fmt[] = "\\%s\\foo";
+	char unc[sizeof(unc_fmt)+SERVER_NAME_LEN_WITH_NULL] = {0};
+	struct sockaddr_in *ipv4 = (struct sockaddr_in *)&iface->sockaddr;
+	struct sockaddr_in6 *ipv6 = (struct sockaddr_in6 *)&iface->sockaddr;
+	int rc;
+	unsigned int xid = get_xid();
+
+	cifs_dbg(FYI, "adding channel to ses %p (speed:%zu bps rdma:%s ",
+		 ses, iface->speed, iface->rdma_capable ? "yes" : "no");
+	if (iface->sockaddr.ss_family == AF_INET)
+		cifs_dbg(FYI, "ip:%pI4)\n", &ipv4->sin_addr);
+	else
+		cifs_dbg(FYI, "ip:%pI6)\n", &ipv6->sin6_addr);
+
+	/*
+	 * Setup a smb_vol with mostly the same info as the existing
+	 * session and overwrite it with the requested iface data.
+	 *
+	 * We need to setup at least the fields used for negprot and
+	 * sesssetup.
+	 *
+	 * We only need the volume here, so we can reuse memory from
+	 * the session and server without caring about memory
+	 * management.
+	 */
+
+	/* Always make new connection for now (TODO?) */
+	vol.nosharesock = true;
+
+	/* Auth */
+	vol.domainauto = ses->domainAuto;
+	vol.domainname = ses->domainName;
+	vol.username = ses->user_name;
+	vol.password = ses->password;
+	vol.sectype = ses->sectype;
+	vol.sign = ses->sign;
+
+	/* UNC and paths */
+	/* XXX: Use ses->server->hostname? */
+	sprintf(unc, unc_fmt, ses->serverName);
+	vol.UNC = unc;
+	vol.prepath = "";
+
+	/* Re-use same version as master connection */
+	vol.vals = ses->server->vals;
+	vol.ops = ses->server->ops;
+
+	vol.noblocksnd = ses->server->noblocksnd;
+	vol.noautotune = ses->server->noautotune;
+	vol.sockopt_tcp_nodelay = ses->server->tcp_nodelay;
+	vol.echo_interval = ses->server->echo_interval / HZ;
+
+	/*
+	 * This will be used for encoding/decoding user/domain/pw
+	 * during sess setup auth.
+	 *
+	 * XXX: We use the default for simplicity but the proper way
+	 * would be to use the one that ses used, which is not
+	 * stored. This might break when dealing with non-ascii
+	 * strings.
+	 */
+	vol.local_nls = load_nls_default();
+
+	/* Use RDMA if possible */
+	vol.rdma = iface->rdma_capable;
+	memcpy(&vol.dstaddr, &iface->sockaddr, sizeof(struct sockaddr_storage));
+
+	/* reuse master con client guid */
+	memcpy(&vol.client_guid, ses->server->client_guid,
+	       SMB2_CLIENT_GUID_SIZE);
+	vol.use_client_guid = true;
+
+	mutex_lock(&ses->session_mutex);
+
+	chan = &ses->chans[ses->chan_count];
+	chan->server = cifs_get_tcp_session(&vol);
+	if (IS_ERR(chan->server)) {
+		rc = PTR_ERR(chan->server);
+		chan->server = NULL;
+		goto out;
+	}
+	spin_lock(&cifs_tcp_ses_lock);
+	chan->server->is_channel = true;
+	spin_unlock(&cifs_tcp_ses_lock);
+
+	/*
+	 * We need to allocate the server crypto now as we will need
+	 * to sign packets before we generate the channel signing key
+	 * (we sign with the session key)
+	 */
+	rc = smb311_crypto_shash_allocate(chan->server);
+	if (rc) {
+		cifs_dbg(VFS, "%s: crypto alloc failed\n", __func__);
+		goto out;
+	}
+
+	ses->binding = true;
+	rc = cifs_negotiate_protocol(xid, ses);
+	if (rc)
+		goto out;
+
+	rc = cifs_setup_session(xid, ses, vol.local_nls);
+	if (rc)
+		goto out;
+
+	/* success, put it on the list
+	 * XXX: sharing ses between 2 tcp server is not possible, the
+	 * way "internal" linked lists works in linux makes element
+	 * only able to belong to one list
+	 *
+	 * the binding session is already established so the rest of
+	 * the code should be able to look it up, no need to add the
+	 * ses to the new server.
+	 */
+
+	ses->chan_count++;
+	atomic_set(&ses->chan_seq, 0);
+out:
+	ses->binding = false;
+	mutex_unlock(&ses->session_mutex);
+
+	if (rc && chan->server)
+		cifs_put_tcp_session(chan->server, 0);
+	unload_nls(vol.local_nls);
+
+	return rc;
+}
 
 static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, SESSION_SETUP_ANDX *pSMB)
 {
@@ -342,6 +593,7 @@ int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len,
 void build_ntlmssp_negotiate_blob(unsigned char *pbuffer,
 					 struct cifs_ses *ses)
 {
+	struct TCP_Server_Info *server = cifs_ses_server(ses);
 	NEGOTIATE_MESSAGE *sec_blob = (NEGOTIATE_MESSAGE *)pbuffer;
 	__u32 flags;
 
@@ -354,9 +606,9 @@ void build_ntlmssp_negotiate_blob(unsigned char *pbuffer,
 		NTLMSSP_NEGOTIATE_128 | NTLMSSP_NEGOTIATE_UNICODE |
 		NTLMSSP_NEGOTIATE_NTLM | NTLMSSP_NEGOTIATE_EXTENDED_SEC |
 		NTLMSSP_NEGOTIATE_SEAL;
-	if (ses->server->sign)
+	if (server->sign)
 		flags |= NTLMSSP_NEGOTIATE_SIGN;
-	if (!ses->server->session_estab || ses->ntlmssp->sesskey_per_smbsess)
+	if (!server->session_estab || ses->ntlmssp->sesskey_per_smbsess)
 		flags |= NTLMSSP_NEGOTIATE_KEY_XCH;
 
 	sec_blob->NegotiateFlags = cpu_to_le32(flags);
diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c
index 514810694c0f..b130efaf8feb 100644
--- a/fs/cifs/smb1ops.c
+++ b/fs/cifs/smb1ops.c
@@ -369,12 +369,10 @@ coalesce_t2(char *second_buf, struct smb_hdr *target_hdr)
 
 static void
 cifs_downgrade_oplock(struct TCP_Server_Info *server,
-			struct cifsInodeInfo *cinode, bool set_level2)
+		      struct cifsInodeInfo *cinode, __u32 oplock,
+		      unsigned int epoch, bool *purge_cache)
 {
-	if (set_level2)
-		cifs_set_oplock_level(cinode, OPLOCK_READ);
-	else
-		cifs_set_oplock_level(cinode, 0);
+	cifs_set_oplock_level(cinode, oplock);
 }
 
 static bool
@@ -506,7 +504,8 @@ cifs_negotiate_rsize(struct cifs_tcon *tcon, struct smb_vol *volume_info)
 }
 
 static void
-cifs_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon)
+cifs_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon,
+	      struct cifs_sb_info *cifs_sb)
 {
 	CIFSSMBQFSDeviceInfo(xid, tcon);
 	CIFSSMBQFSAttributeInfo(xid, tcon);
@@ -567,7 +566,7 @@ cifs_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 		oparms.tcon = tcon;
 		oparms.cifs_sb = cifs_sb;
 		oparms.desired_access = FILE_READ_ATTRIBUTES;
-		oparms.create_options = 0;
+		oparms.create_options = cifs_create_options(cifs_sb, 0);
 		oparms.disposition = FILE_OPEN;
 		oparms.path = full_path;
 		oparms.fid = &fid;
@@ -767,7 +766,7 @@ smb_set_file_info(struct inode *inode, const char *full_path,
 	struct cifs_tcon *tcon;
 
 	/* if the file is already open for write, just use that fileid */
-	open_file = find_writable_file(cinode, true);
+	open_file = find_writable_file(cinode, FIND_WR_FSUID_ONLY);
 	if (open_file) {
 		fid.netfid = open_file->fid.netfid;
 		netpid = open_file->pid;
@@ -795,7 +794,7 @@ smb_set_file_info(struct inode *inode, const char *full_path,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = SYNCHRONIZE | FILE_WRITE_ATTRIBUTES;
-	oparms.create_options = CREATE_NOT_DIR;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = full_path;
 	oparms.fid = &fid;
@@ -874,7 +873,7 @@ cifs_oplock_response(struct cifs_tcon *tcon, struct cifs_fid *fid,
 
 static int
 cifs_queryfs(const unsigned int xid, struct cifs_tcon *tcon,
-	     struct kstatfs *buf)
+	     struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
 {
 	int rc = -EOPNOTSUPP;
 
@@ -972,7 +971,8 @@ cifs_query_symlink(const unsigned int xid, struct cifs_tcon *tcon,
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = FILE_READ_ATTRIBUTES;
-	oparms.create_options = OPEN_REPARSE_POINT;
+	oparms.create_options = cifs_create_options(cifs_sb,
+						    OPEN_REPARSE_POINT);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = full_path;
 	oparms.fid = &fid;
@@ -1031,7 +1031,6 @@ cifs_make_node(unsigned int xid, struct inode *inode,
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
 	struct inode *newinode = NULL;
 	int rc = -EPERM;
-	int create_options = CREATE_NOT_DIR | CREATE_OPTION_SPECIAL;
 	FILE_ALL_INFO *buf = NULL;
 	struct cifs_io_parms io_parms;
 	__u32 oplock = 0;
@@ -1092,13 +1091,11 @@ cifs_make_node(unsigned int xid, struct inode *inode,
 		goto out;
 	}
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
-
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = GENERIC_WRITE;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR |
+						    CREATE_OPTION_SPECIAL);
 	oparms.disposition = FILE_CREATE;
 	oparms.path = full_path;
 	oparms.fid = &fid;
diff --git a/fs/cifs/smb2file.c b/fs/cifs/smb2file.c
index 8b0b512c5792..2fa3ba354cc9 100644
--- a/fs/cifs/smb2file.c
+++ b/fs/cifs/smb2file.c
@@ -62,12 +62,12 @@ smb2_open_file(const unsigned int xid, struct cifs_open_parms *oparms,
 	smb2_oplock = SMB2_OPLOCK_LEVEL_BATCH;
 
 	rc = SMB2_open(xid, oparms, smb2_path, &smb2_oplock, smb2_data, NULL,
-		       NULL);
+		       NULL, NULL);
 	if (rc)
 		goto out;
 
 
-	 if (oparms->tcon->use_resilient) {
+	if (oparms->tcon->use_resilient) {
 		/* default timeout is 0, servers pick default (120 seconds) */
 		nr_ioctl_req.Timeout =
 			cpu_to_le32(oparms->tcon->handle_timeout);
@@ -152,7 +152,12 @@ smb2_unlock_range(struct cifsFileInfo *cfile, struct file_lock *flock,
 		    (li->offset + li->length))
 			continue;
 		if (current->tgid != li->pid)
-			continue;
+			/*
+			 * flock and OFD lock are associated with an open
+			 * file description, not the process.
+			 */
+			if (!(flock->fl_flags & (FL_FLOCK | FL_OFDLCK)))
+				continue;
 		if (cinode->can_cache_brlcks) {
 			/*
 			 * We can cache brlock requests - simply remove a lock
diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c
index 4121ac1163ca..a8c301ae00ed 100644
--- a/fs/cifs/smb2inode.c
+++ b/fs/cifs/smb2inode.c
@@ -95,12 +95,11 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon,
 		goto finished;
 	}
 
+	memset(&oparms, 0, sizeof(struct cifs_open_parms));
 	oparms.tcon = tcon;
 	oparms.desired_access = desired_access;
 	oparms.disposition = create_disposition;
-	oparms.create_options = create_options;
-	if (backup_cred(cifs_sb))
-		oparms.create_options |= CREATE_OPEN_BACKUP_INTENT;
+	oparms.create_options = cifs_create_options(cifs_sb, create_options);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 	oparms.mode = mode;
@@ -313,7 +312,7 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon,
 	rqst[num_rqst].rq_iov = close_iov;
 	rqst[num_rqst].rq_nvec = 1;
 	rc = SMB2_close_init(tcon, &rqst[num_rqst], COMPOUND_FID,
-			     COMPOUND_FID);
+			     COMPOUND_FID, false);
 	smb2_set_related(&rqst[num_rqst]);
 	if (rc)
 		goto finished;
@@ -456,7 +455,7 @@ smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 
 	/* If it is a root and its handle is cached then use it */
 	if (!strlen(full_path) && !no_cached_open) {
-		rc = open_shroot(xid, tcon, &fid);
+		rc = open_shroot(xid, tcon, cifs_sb, &fid);
 		if (rc)
 			goto out;
 
@@ -473,9 +472,6 @@ smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 		goto out;
 	}
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
-
 	cifs_get_readable_path(tcon, full_path, &cfile);
 	rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
 			      FILE_READ_ATTRIBUTES, FILE_OPEN, create_options,
@@ -525,7 +521,7 @@ smb2_mkdir_setinfo(struct inode *inode, const char *name,
 	cifs_i = CIFS_I(inode);
 	dosattrs = cifs_i->cifsAttrs | ATTR_READONLY;
 	data.Attributes = cpu_to_le32(dosattrs);
-	cifs_get_writable_path(tcon, name, &cfile);
+	cifs_get_writable_path(tcon, name, FIND_WR_ANY, &cfile);
 	tmprc = smb2_compound_op(xid, tcon, cifs_sb, name,
 				 FILE_WRITE_ATTRIBUTES, FILE_CREATE,
 				 CREATE_NOT_FILE, ACL_NO_MODE,
@@ -581,7 +577,7 @@ smb2_rename_path(const unsigned int xid, struct cifs_tcon *tcon,
 {
 	struct cifsFileInfo *cfile;
 
-	cifs_get_writable_path(tcon, from_name, &cfile);
+	cifs_get_writable_path(tcon, from_name, FIND_WR_WITH_DELETE, &cfile);
 
 	return smb2_set_path_attr(xid, tcon, from_name, to_name,
 				  cifs_sb, DELETE, SMB2_OP_RENAME, cfile);
diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c
index e311f58dc1c8..497afb0b9960 100644
--- a/fs/cifs/smb2misc.c
+++ b/fs/cifs/smb2misc.c
@@ -29,6 +29,7 @@
 #include "cifs_unicode.h"
 #include "smb2status.h"
 #include "smb2glob.h"
+#include "nterr.h"
 
 static int
 check_smb2_hdr(struct smb2_sync_hdr *shdr, __u64 mid)
@@ -249,16 +250,10 @@ smb2_check_message(char *buf, unsigned int len, struct TCP_Server_Info *srvr)
 		 * of junk. Other servers match RFC1001 len to actual
 		 * SMB2/SMB3 frame length (header + smb2 response specific data)
 		 * Some windows servers also pad up to 8 bytes when compounding.
-		 * If pad is longer than eight bytes, log the server behavior
-		 * (once), since may indicate a problem but allow it and continue
-		 * since the frame is parseable.
 		 */
-		if (clc_len < len) {
-			pr_warn_once(
-			     "srv rsp padded more than expected. Length %d not %d for cmd:%d mid:%llu\n",
-			     len, clc_len, command, mid);
+		if (clc_len < len)
 			return 0;
-		}
+
 		pr_warn_once(
 			"srv rsp too short, len %d not %d. cmd:%d mid:%llu\n",
 			len, clc_len, command, mid);
@@ -534,7 +529,7 @@ smb2_tcon_has_lease(struct cifs_tcon *tcon, struct smb2_lease_break *rsp,
 
 		cifs_dbg(FYI, "found in the open list\n");
 		cifs_dbg(FYI, "lease key match, lease break 0x%x\n",
-			 le32_to_cpu(rsp->NewLeaseState));
+			 lease_state);
 
 		if (ack_req)
 			cfile->oplock_break_cancelled = false;
@@ -543,17 +538,8 @@ smb2_tcon_has_lease(struct cifs_tcon *tcon, struct smb2_lease_break *rsp,
 
 		set_bit(CIFS_INODE_PENDING_OPLOCK_BREAK, &cinode->flags);
 
-		/*
-		 * Set or clear flags depending on the lease state being READ.
-		 * HANDLE caching flag should be added when the client starts
-		 * to defer closing remote file handles with HANDLE leases.
-		 */
-		if (lease_state & SMB2_LEASE_READ_CACHING_HE)
-			set_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2,
-				&cinode->flags);
-		else
-			clear_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2,
-				  &cinode->flags);
+		cfile->oplock_epoch = le16_to_cpu(rsp->Epoch);
+		cfile->oplock_level = lease_state;
 
 		cifs_queue_oplock_break(cfile);
 		kfree(lw);
@@ -576,7 +562,7 @@ smb2_tcon_has_lease(struct cifs_tcon *tcon, struct smb2_lease_break *rsp,
 
 		cifs_dbg(FYI, "found in the pending open list\n");
 		cifs_dbg(FYI, "lease key match, lease break 0x%x\n",
-			 le32_to_cpu(rsp->NewLeaseState));
+			 lease_state);
 
 		open->oplock = lease_state;
 	}
@@ -673,10 +659,10 @@ smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server)
 	spin_lock(&cifs_tcp_ses_lock);
 	list_for_each(tmp, &server->smb_ses_list) {
 		ses = list_entry(tmp, struct cifs_ses, smb_ses_list);
+
 		list_for_each(tmp1, &ses->tcon_list) {
 			tcon = list_entry(tmp1, struct cifs_tcon, tcon_list);
 
-			cifs_stats_inc(&tcon->stats.cifs_stats.num_oplock_brks);
 			spin_lock(&tcon->open_file_lock);
 			list_for_each(tmp2, &tcon->openFileList) {
 				cfile = list_entry(tmp2, struct cifsFileInfo,
@@ -688,6 +674,8 @@ smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server)
 					continue;
 
 				cifs_dbg(FYI, "file id match, oplock break\n");
+				cifs_stats_inc(
+				    &tcon->stats.cifs_stats.num_oplock_brks);
 				cinode = CIFS_I(d_inode(cfile->dentry));
 				spin_lock(&cfile->file_info_lock);
 				if (!CIFS_CACHE_WRITE(cinode) &&
@@ -699,18 +687,9 @@ smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server)
 				set_bit(CIFS_INODE_PENDING_OPLOCK_BREAK,
 					&cinode->flags);
 
-				/*
-				 * Set flag if the server downgrades the oplock
-				 * to L2 else clear.
-				 */
-				if (rsp->OplockLevel)
-					set_bit(
-					   CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2,
-					   &cinode->flags);
-				else
-					clear_bit(
-					   CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2,
-					   &cinode->flags);
+				cfile->oplock_epoch = 0;
+				cfile->oplock_level = rsp->OplockLevel;
+
 				spin_unlock(&cfile->file_info_lock);
 
 				cifs_queue_oplock_break(cfile);
@@ -720,9 +699,6 @@ smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server)
 				return true;
 			}
 			spin_unlock(&tcon->open_file_lock);
-			spin_unlock(&cifs_tcp_ses_lock);
-			cifs_dbg(FYI, "No matching file for oplock break\n");
-			return true;
 		}
 	}
 	spin_unlock(&cifs_tcp_ses_lock);
@@ -735,45 +711,112 @@ smb2_cancelled_close_fid(struct work_struct *work)
 {
 	struct close_cancelled_open *cancelled = container_of(work,
 					struct close_cancelled_open, work);
+	struct cifs_tcon *tcon = cancelled->tcon;
+	int rc;
+
+	if (cancelled->mid)
+		cifs_tcon_dbg(VFS, "Close unmatched open for MID:%llx\n",
+			      cancelled->mid);
+	else
+		cifs_tcon_dbg(VFS, "Close interrupted close\n");
 
-	cifs_dbg(VFS, "Close unmatched open\n");
+	rc = SMB2_close(0, tcon, cancelled->fid.persistent_fid,
+			cancelled->fid.volatile_fid);
+	if (rc)
+		cifs_tcon_dbg(VFS, "Close cancelled mid failed rc:%d\n", rc);
 
-	SMB2_close(0, cancelled->tcon, cancelled->fid.persistent_fid,
-		   cancelled->fid.volatile_fid);
-	cifs_put_tcon(cancelled->tcon);
+	cifs_put_tcon(tcon);
 	kfree(cancelled);
 }
 
+/*
+ * Caller should already has an extra reference to @tcon
+ * This function is used to queue work to close a handle to prevent leaks
+ * on the server.
+ * We handle two cases. If an open was interrupted after we sent the
+ * SMB2_CREATE to the server but before we processed the reply, and second
+ * if a close was interrupted before we sent the SMB2_CLOSE to the server.
+ */
+static int
+__smb2_handle_cancelled_cmd(struct cifs_tcon *tcon, __u16 cmd, __u64 mid,
+			    __u64 persistent_fid, __u64 volatile_fid)
+{
+	struct close_cancelled_open *cancelled;
+
+	cancelled = kzalloc(sizeof(*cancelled), GFP_ATOMIC);
+	if (!cancelled)
+		return -ENOMEM;
+
+	cancelled->fid.persistent_fid = persistent_fid;
+	cancelled->fid.volatile_fid = volatile_fid;
+	cancelled->tcon = tcon;
+	cancelled->cmd = cmd;
+	cancelled->mid = mid;
+	INIT_WORK(&cancelled->work, smb2_cancelled_close_fid);
+	WARN_ON(queue_work(cifsiod_wq, &cancelled->work) == false);
+
+	return 0;
+}
+
+int
+smb2_handle_cancelled_close(struct cifs_tcon *tcon, __u64 persistent_fid,
+			    __u64 volatile_fid)
+{
+	int rc;
+
+	cifs_dbg(FYI, "%s: tc_count=%d\n", __func__, tcon->tc_count);
+	spin_lock(&cifs_tcp_ses_lock);
+	if (tcon->tc_count <= 0) {
+		struct TCP_Server_Info *server = NULL;
+
+		WARN_ONCE(tcon->tc_count < 0, "tcon refcount is negative");
+		spin_unlock(&cifs_tcp_ses_lock);
+
+		if (tcon->ses)
+			server = tcon->ses->server;
+
+		cifs_server_dbg(FYI, "tid=%u: tcon is closing, skipping async close retry of fid %llu %llu\n",
+				tcon->tid, persistent_fid, volatile_fid);
+
+		return 0;
+	}
+	tcon->tc_count++;
+	spin_unlock(&cifs_tcp_ses_lock);
+
+	rc = __smb2_handle_cancelled_cmd(tcon, SMB2_CLOSE_HE, 0,
+					 persistent_fid, volatile_fid);
+	if (rc)
+		cifs_put_tcon(tcon);
+
+	return rc;
+}
+
 int
 smb2_handle_cancelled_mid(char *buffer, struct TCP_Server_Info *server)
 {
 	struct smb2_sync_hdr *sync_hdr = (struct smb2_sync_hdr *)buffer;
 	struct smb2_create_rsp *rsp = (struct smb2_create_rsp *)buffer;
 	struct cifs_tcon *tcon;
-	struct close_cancelled_open *cancelled;
+	int rc;
 
 	if (sync_hdr->Command != SMB2_CREATE ||
 	    sync_hdr->Status != STATUS_SUCCESS)
 		return 0;
 
-	cancelled = kzalloc(sizeof(*cancelled), GFP_KERNEL);
-	if (!cancelled)
-		return -ENOMEM;
-
 	tcon = smb2_find_smb_tcon(server, sync_hdr->SessionId,
 				  sync_hdr->TreeId);
-	if (!tcon) {
-		kfree(cancelled);
+	if (!tcon)
 		return -ENOENT;
-	}
 
-	cancelled->fid.persistent_fid = rsp->PersistentFileId;
-	cancelled->fid.volatile_fid = rsp->VolatileFileId;
-	cancelled->tcon = tcon;
-	INIT_WORK(&cancelled->work, smb2_cancelled_close_fid);
-	queue_work(cifsiod_wq, &cancelled->work);
+	rc = __smb2_handle_cancelled_cmd(tcon,
+					 le16_to_cpu(sync_hdr->Command),
+					 le64_to_cpu(sync_hdr->MessageId),
+					 rsp->PersistentFileId,
+					 rsp->VolatileFileId);
+	if (rc)
+		cifs_put_tcon(tcon);
 
-	return 0;
+	return rc;
 }
 
 /**
@@ -788,23 +831,37 @@ smb311_update_preauth_hash(struct cifs_ses *ses, struct kvec *iov, int nvec)
 	int i, rc;
 	struct sdesc *d;
 	struct smb2_sync_hdr *hdr;
+	struct TCP_Server_Info *server = cifs_ses_server(ses);
 
-	if (ses->server->tcpStatus == CifsGood) {
-		/* skip non smb311 connections */
-		if (ses->server->dialect != SMB311_PROT_ID)
-			return 0;
+	hdr = (struct smb2_sync_hdr *)iov[0].iov_base;
+	/* neg prot are always taken */
+	if (hdr->Command == SMB2_NEGOTIATE)
+		goto ok;
 
-		/* skip last sess setup response */
-		hdr = (struct smb2_sync_hdr *)iov[0].iov_base;
-		if (hdr->Flags & SMB2_FLAGS_SIGNED)
-			return 0;
-	}
+	/*
+	 * If we process a command which wasn't a negprot it means the
+	 * neg prot was already done, so the server dialect was set
+	 * and we can test it. Preauth requires 3.1.1 for now.
+	 */
+	if (server->dialect != SMB311_PROT_ID)
+		return 0;
+
+	if (hdr->Command != SMB2_SESSION_SETUP)
+		return 0;
+
+	/* skip last sess setup response */
+	if ((hdr->Flags & SMB2_FLAGS_SERVER_TO_REDIR)
+	    && (hdr->Status == NT_STATUS_OK
+		|| (hdr->Status !=
+		    cpu_to_le32(NT_STATUS_MORE_PROCESSING_REQUIRED))))
+		return 0;
 
-	rc = smb311_crypto_shash_allocate(ses->server);
+ok:
+	rc = smb311_crypto_shash_allocate(server);
 	if (rc)
 		return rc;
 
-	d = ses->server->secmech.sdescsha512;
+	d = server->secmech.sdescsha512;
 	rc = crypto_shash_init(&d->shash);
 	if (rc) {
 		cifs_dbg(VFS, "%s: could not init sha512 shash\n", __func__);
diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c
index cd55af9b7cc5..f829f4165d38 100644
--- a/fs/cifs/smb2ops.c
+++ b/fs/cifs/smb2ops.c
@@ -10,7 +10,9 @@
 #include <linux/falloc.h>
 #include <linux/scatterlist.h>
 #include <linux/uuid.h>
+#include <linux/sort.h>
 #include <crypto/aead.h>
+#include "cifsfs.h"
 #include "cifsglob.h"
 #include "smb2pdu.h"
 #include "smb2proto.h"
@@ -151,13 +153,7 @@ smb2_get_credits_field(struct TCP_Server_Info *server, const int optype)
 static unsigned int
 smb2_get_credits(struct mid_q_entry *mid)
 {
-	struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)mid->resp_buf;
-
-	if (mid->mid_state == MID_RESPONSE_RECEIVED
-	    || mid->mid_state == MID_RESPONSE_MALFORMED)
-		return le16_to_cpu(shdr->CreditRequest);
-
-	return 0;
+	return mid->credits_received;
 }
 
 static int
@@ -315,7 +311,7 @@ smb2_negotiate(const unsigned int xid, struct cifs_ses *ses)
 {
 	int rc;
 
-	ses->server->CurrentMid = 0;
+	cifs_ses_server(ses)->CurrentMid = 0;
 	rc = SMB2_negotiate(xid, ses);
 	/* BB we probably don't need to retry with modern servers */
 	if (rc == -EAGAIN)
@@ -332,16 +328,6 @@ smb2_negotiate_wsize(struct cifs_tcon *tcon, struct smb_vol *volume_info)
 	/* start with specified wsize, or default */
 	wsize = volume_info->wsize ? volume_info->wsize : CIFS_DEFAULT_IOSIZE;
 	wsize = min_t(unsigned int, wsize, server->max_write);
-#ifdef CONFIG_CIFS_SMB_DIRECT
-	if (server->rdma) {
-		if (server->sign)
-			wsize = min_t(unsigned int,
-				wsize, server->smbd_conn->max_fragmented_send_size);
-		else
-			wsize = min_t(unsigned int,
-				wsize, server->smbd_conn->max_readwrite_size);
-	}
-#endif
 	if (!(server->capabilities & SMB2_GLOBAL_CAP_LARGE_MTU))
 		wsize = min_t(unsigned int, wsize, SMB2_MAX_BUFFER_SIZE);
 
@@ -360,8 +346,15 @@ smb3_negotiate_wsize(struct cifs_tcon *tcon, struct smb_vol *volume_info)
 #ifdef CONFIG_CIFS_SMB_DIRECT
 	if (server->rdma) {
 		if (server->sign)
+			/*
+			 * Account for SMB2 data transfer packet header and
+			 * possible encryption header
+			 */
 			wsize = min_t(unsigned int,
-				wsize, server->smbd_conn->max_fragmented_send_size);
+				wsize,
+				server->smbd_conn->max_fragmented_send_size -
+					SMB2_READWRITE_PDU_HEADER_SIZE -
+					sizeof(struct smb2_transform_hdr));
 		else
 			wsize = min_t(unsigned int,
 				wsize, server->smbd_conn->max_readwrite_size);
@@ -382,16 +375,6 @@ smb2_negotiate_rsize(struct cifs_tcon *tcon, struct smb_vol *volume_info)
 	/* start with specified rsize, or default */
 	rsize = volume_info->rsize ? volume_info->rsize : CIFS_DEFAULT_IOSIZE;
 	rsize = min_t(unsigned int, rsize, server->max_read);
-#ifdef CONFIG_CIFS_SMB_DIRECT
-	if (server->rdma) {
-		if (server->sign)
-			rsize = min_t(unsigned int,
-				rsize, server->smbd_conn->max_fragmented_recv_size);
-		else
-			rsize = min_t(unsigned int,
-				rsize, server->smbd_conn->max_readwrite_size);
-	}
-#endif
 
 	if (!(server->capabilities & SMB2_GLOBAL_CAP_LARGE_MTU))
 		rsize = min_t(unsigned int, rsize, SMB2_MAX_BUFFER_SIZE);
@@ -411,8 +394,15 @@ smb3_negotiate_rsize(struct cifs_tcon *tcon, struct smb_vol *volume_info)
 #ifdef CONFIG_CIFS_SMB_DIRECT
 	if (server->rdma) {
 		if (server->sign)
+			/*
+			 * Account for SMB2 data transfer packet header and
+			 * possible encryption header
+			 */
 			rsize = min_t(unsigned int,
-				rsize, server->smbd_conn->max_fragmented_recv_size);
+				rsize,
+				server->smbd_conn->max_fragmented_recv_size -
+					SMB2_READWRITE_PDU_HEADER_SIZE -
+					sizeof(struct smb2_transform_hdr));
 		else
 			rsize = min_t(unsigned int,
 				rsize, server->smbd_conn->max_readwrite_size);
@@ -558,6 +548,13 @@ out:
 	return rc;
 }
 
+static int compare_iface(const void *ia, const void *ib)
+{
+	const struct cifs_server_iface *a = (struct cifs_server_iface *)ia;
+	const struct cifs_server_iface *b = (struct cifs_server_iface *)ib;
+
+	return a->speed == b->speed ? 0 : (a->speed > b->speed ? -1 : 1);
+}
 
 static int
 SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon)
@@ -587,6 +584,9 @@ SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon)
 	if (rc)
 		goto out;
 
+	/* sort interfaces from fastest to slowest */
+	sort(iface_list, iface_count, sizeof(*iface_list), compare_iface, NULL);
+
 	spin_lock(&ses->iface_lock);
 	kfree(ses->iface_list);
 	ses->iface_list = iface_list;
@@ -611,6 +611,7 @@ smb2_close_cached_fid(struct kref *ref)
 			   cfid->fid->volatile_fid);
 		cfid->is_valid = false;
 		cfid->file_all_info_is_valid = false;
+		cfid->has_lease = false;
 	}
 }
 
@@ -621,19 +622,35 @@ void close_shroot(struct cached_fid *cfid)
 	mutex_unlock(&cfid->fid_mutex);
 }
 
+void close_shroot_lease_locked(struct cached_fid *cfid)
+{
+	if (cfid->has_lease) {
+		cfid->has_lease = false;
+		kref_put(&cfid->refcount, smb2_close_cached_fid);
+	}
+}
+
+void close_shroot_lease(struct cached_fid *cfid)
+{
+	mutex_lock(&cfid->fid_mutex);
+	close_shroot_lease_locked(cfid);
+	mutex_unlock(&cfid->fid_mutex);
+}
+
 void
 smb2_cached_lease_break(struct work_struct *work)
 {
 	struct cached_fid *cfid = container_of(work,
 				struct cached_fid, lease_break);
 
-	close_shroot(cfid);
+	close_shroot_lease(cfid);
 }
 
 /*
  * Open the directory at the root of a share
  */
-int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid)
+int open_shroot(unsigned int xid, struct cifs_tcon *tcon,
+		struct cifs_sb_info *cifs_sb, struct cifs_fid *pfid)
 {
 	struct cifs_ses *ses = tcon->ses;
 	struct TCP_Server_Info *server = ses->server;
@@ -670,6 +687,11 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid)
 	if (smb3_encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
+	if (!server->ops->new_lease_key)
+		return -EIO;
+
+	server->ops->new_lease_key(pfid);
+
 	memset(rqst, 0, sizeof(rqst));
 	resp_buftype[0] = resp_buftype[1] = CIFS_NO_BUFFER;
 	memset(rsp_iov, 0, sizeof(rsp_iov));
@@ -680,7 +702,7 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid)
 	rqst[0].rq_nvec = SMB2_CREATE_IOV_SIZE;
 
 	oparms.tcon = tcon;
-	oparms.create_options = 0;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.desired_access = FILE_READ_ATTRIBUTES;
 	oparms.disposition = FILE_OPEN;
 	oparms.fid = pfid;
@@ -768,9 +790,11 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid)
 	/* BB TBD check to see if oplock level check can be removed below */
 	if (o_rsp->OplockLevel == SMB2_OPLOCK_LEVEL_LEASE) {
 		kref_get(&tcon->crfid.refcount);
+		tcon->crfid.has_lease = true;
 		smb2_parse_contexts(server, o_rsp,
 				&oparms.fid->epoch,
-				oparms.fid->lease_key, &oplock, NULL);
+				    oparms.fid->lease_key, &oplock,
+				    NULL, NULL);
 	} else
 		goto oshr_exit;
 
@@ -782,7 +806,7 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid)
 				sizeof(struct smb2_file_all_info),
 				&rsp_iov[1], sizeof(struct smb2_file_all_info),
 				(char *)&tcon->crfid.file_all_info))
-		tcon->crfid.file_all_info_is_valid = 1;
+		tcon->crfid.file_all_info_is_valid = true;
 
 oshr_exit:
 	mutex_unlock(&tcon->crfid.fid_mutex);
@@ -795,7 +819,8 @@ oshr_free:
 }
 
 static void
-smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon)
+smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon,
+	      struct cifs_sb_info *cifs_sb)
 {
 	int rc;
 	__le16 srch_path = 0; /* Null - open root of share */
@@ -807,15 +832,15 @@ smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon)
 	oparms.tcon = tcon;
 	oparms.desired_access = FILE_READ_ATTRIBUTES;
 	oparms.disposition = FILE_OPEN;
-	oparms.create_options = 0;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
 	if (no_cached_open)
 		rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL,
-			       NULL);
+			       NULL, NULL);
 	else
-		rc = open_shroot(xid, tcon, &fid);
+		rc = open_shroot(xid, tcon, cifs_sb, &fid);
 
 	if (rc)
 		return;
@@ -837,7 +862,8 @@ smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon)
 }
 
 static void
-smb2_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon)
+smb2_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon,
+	      struct cifs_sb_info *cifs_sb)
 {
 	int rc;
 	__le16 srch_path = 0; /* Null - open root of share */
@@ -848,11 +874,12 @@ smb2_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon)
 	oparms.tcon = tcon;
 	oparms.desired_access = FILE_READ_ATTRIBUTES;
 	oparms.disposition = FILE_OPEN;
-	oparms.create_options = 0;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
-	rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL, NULL);
+	rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL,
+		       NULL, NULL);
 	if (rc)
 		return;
 
@@ -883,14 +910,12 @@ smb2_is_path_accessible(const unsigned int xid, struct cifs_tcon *tcon,
 	oparms.tcon = tcon;
 	oparms.desired_access = FILE_READ_ATTRIBUTES;
 	oparms.disposition = FILE_OPEN;
-	if (backup_cred(cifs_sb))
-		oparms.create_options = CREATE_OPEN_BACKUP_INTENT;
-	else
-		oparms.create_options = 0;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
-	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL);
+	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL,
+		       NULL);
 	if (rc) {
 		kfree(utf16_path);
 		return rc;
@@ -1093,7 +1118,8 @@ smb2_set_ea(const unsigned int xid, struct cifs_tcon *tcon,
 	void *data[1];
 	struct smb2_file_full_ea_info *ea = NULL;
 	struct kvec close_iov[1];
-	int rc;
+	struct smb2_query_info_rsp *rsp;
+	int rc, used_len = 0;
 
 	if (smb3_encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
@@ -1116,6 +1142,38 @@ smb2_set_ea(const unsigned int xid, struct cifs_tcon *tcon,
 							     cifs_sb);
 			if (rc == -ENODATA)
 				goto sea_exit;
+		} else {
+			/* If we are adding a attribute we should first check
+			 * if there will be enough space available to store
+			 * the new EA. If not we should not add it since we
+			 * would not be able to even read the EAs back.
+			 */
+			rc = smb2_query_info_compound(xid, tcon, utf16_path,
+				      FILE_READ_EA,
+				      FILE_FULL_EA_INFORMATION,
+				      SMB2_O_INFO_FILE,
+				      CIFSMaxBufSize -
+				      MAX_SMB2_CREATE_RESPONSE_SIZE -
+				      MAX_SMB2_CLOSE_RESPONSE_SIZE,
+				      &rsp_iov[1], &resp_buftype[1], cifs_sb);
+			if (rc == 0) {
+				rsp = (struct smb2_query_info_rsp *)rsp_iov[1].iov_base;
+				used_len = le32_to_cpu(rsp->OutputBufferLength);
+			}
+			free_rsp_buf(resp_buftype[1], rsp_iov[1].iov_base);
+			resp_buftype[1] = CIFS_NO_BUFFER;
+			memset(&rsp_iov[1], 0, sizeof(rsp_iov[1]));
+			rc = 0;
+
+			/* Use a fudge factor of 256 bytes in case we collide
+			 * with a different set_EAs command.
+			 */
+			if(CIFSMaxBufSize - MAX_SMB2_CREATE_RESPONSE_SIZE -
+			   MAX_SMB2_CLOSE_RESPONSE_SIZE - 256 <
+			   used_len + ea_name_len + ea_value_len + 1) {
+				rc = -ENOSPC;
+				goto sea_exit;
+			}
 		}
 	}
 
@@ -1128,10 +1186,7 @@ smb2_set_ea(const unsigned int xid, struct cifs_tcon *tcon,
 	oparms.tcon = tcon;
 	oparms.desired_access = FILE_WRITE_EA;
 	oparms.disposition = FILE_OPEN;
-	if (backup_cred(cifs_sb))
-		oparms.create_options = CREATE_OPEN_BACKUP_INTENT;
-	else
-		oparms.create_options = 0;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
@@ -1173,7 +1228,7 @@ smb2_set_ea(const unsigned int xid, struct cifs_tcon *tcon,
 	memset(&close_iov, 0, sizeof(close_iov));
 	rqst[2].rq_iov = close_iov;
 	rqst[2].rq_nvec = 1;
-	rc = SMB2_close_init(tcon, &rqst[2], COMPOUND_FID, COMPOUND_FID);
+	rc = SMB2_close_init(tcon, &rqst[2], COMPOUND_FID, COMPOUND_FID, false);
 	smb2_set_related(&rqst[2]);
 
 	rc = compound_send_recv(xid, ses, flags, 3, rqst,
@@ -1311,6 +1366,7 @@ smb2_set_fid(struct cifsFileInfo *cfile, struct cifs_fid *fid, __u32 oplock)
 
 	cfile->fid.persistent_fid = fid->persistent_fid;
 	cfile->fid.volatile_fid = fid->volatile_fid;
+	cfile->fid.access = fid->access;
 #ifdef CONFIG_CIFS_DEBUG2
 	cfile->fid.mid = fid->mid;
 #endif /* CIFS_DEBUG2 */
@@ -1327,6 +1383,45 @@ smb2_close_file(const unsigned int xid, struct cifs_tcon *tcon,
 	SMB2_close(xid, tcon, fid->persistent_fid, fid->volatile_fid);
 }
 
+static void
+smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon,
+		   struct cifsFileInfo *cfile)
+{
+	struct smb2_file_network_open_info file_inf;
+	struct inode *inode;
+	int rc;
+
+	rc = __SMB2_close(xid, tcon, cfile->fid.persistent_fid,
+		   cfile->fid.volatile_fid, &file_inf);
+	if (rc)
+		return;
+
+	inode = d_inode(cfile->dentry);
+
+	spin_lock(&inode->i_lock);
+	CIFS_I(inode)->time = jiffies;
+
+	/* Creation time should not need to be updated on close */
+	if (file_inf.LastWriteTime)
+		inode->i_mtime = cifs_NTtimeToUnix(file_inf.LastWriteTime);
+	if (file_inf.ChangeTime)
+		inode->i_ctime = cifs_NTtimeToUnix(file_inf.ChangeTime);
+	if (file_inf.LastAccessTime)
+		inode->i_atime = cifs_NTtimeToUnix(file_inf.LastAccessTime);
+
+	/*
+	 * i_blocks is not related to (i_size / i_blksize),
+	 * but instead 512 byte (2**9) size is required for
+	 * calculating num blocks.
+	 */
+	if (le64_to_cpu(file_inf.AllocationSize) > 4096)
+		inode->i_blocks =
+			(512 - 1 + le64_to_cpu(file_inf.AllocationSize)) >> 9;
+
+	/* End of file and Attributes should not have to be updated on close */
+	spin_unlock(&inode->i_lock);
+}
+
 static int
 SMB2_request_res_key(const unsigned int xid, struct cifs_tcon *tcon,
 		     u64 persistent_fid, u64 volatile_fid,
@@ -1360,6 +1455,7 @@ req_res_key_exit:
 static int
 smb2_ioctl_query_info(const unsigned int xid,
 		      struct cifs_tcon *tcon,
+		      struct cifs_sb_info *cifs_sb,
 		      __le16 *path, int is_dir,
 		      unsigned long p)
 {
@@ -1385,6 +1481,7 @@ smb2_ioctl_query_info(const unsigned int xid,
 	struct kvec close_iov[1];
 	unsigned int size[2];
 	void *data[2];
+	int create_options = is_dir ? CREATE_NOT_FILE : CREATE_NOT_DIR;
 
 	memset(rqst, 0, sizeof(rqst));
 	resp_buftype[0] = resp_buftype[1] = resp_buftype[2] = CIFS_NO_BUFFER;
@@ -1402,15 +1499,10 @@ smb2_ioctl_query_info(const unsigned int xid,
 	if (smb3_encryption_required(tcon))
 		flags |= CIFS_TRANSFORM_REQ;
 
-	buffer = kmalloc(qi.output_buffer_length, GFP_KERNEL);
-	if (buffer == NULL)
-		return -ENOMEM;
-
-	if (copy_from_user(buffer, arg + sizeof(struct smb_query_info),
-			   qi.output_buffer_length)) {
-		rc = -EFAULT;
-		goto iqinf_exit;
-	}
+	buffer = memdup_user(arg + sizeof(struct smb_query_info),
+			     qi.output_buffer_length);
+	if (IS_ERR(buffer))
+		return PTR_ERR(buffer);
 
 	/* Open */
 	memset(&open_iov, 0, sizeof(open_iov));
@@ -1420,10 +1512,7 @@ smb2_ioctl_query_info(const unsigned int xid,
 	memset(&oparms, 0, sizeof(oparms));
 	oparms.tcon = tcon;
 	oparms.disposition = FILE_OPEN;
-	if (is_dir)
-		oparms.create_options = CREATE_NOT_FILE;
-	else
-		oparms.create_options = CREATE_NOT_DIR;
+	oparms.create_options = cifs_create_options(cifs_sb, create_options);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
@@ -1467,7 +1556,9 @@ smb2_ioctl_query_info(const unsigned int xid,
 					     COMPOUND_FID, COMPOUND_FID,
 					     qi.info_type, true, buffer,
 					     qi.output_buffer_length,
-					     CIFSMaxBufSize);
+					     CIFSMaxBufSize -
+					     MAX_SMB2_CREATE_RESPONSE_SIZE -
+					     MAX_SMB2_CLOSE_RESPONSE_SIZE);
 		}
 	} else if (qi.flags == PASSTHRU_SET_INFO) {
 		/* Can eventually relax perm check since server enforces too */
@@ -1512,7 +1603,7 @@ smb2_ioctl_query_info(const unsigned int xid,
 	rqst[2].rq_iov = close_iov;
 	rqst[2].rq_nvec = 1;
 
-	rc = SMB2_close_init(tcon, &rqst[2], COMPOUND_FID, COMPOUND_FID);
+	rc = SMB2_close_init(tcon, &rqst[2], COMPOUND_FID, COMPOUND_FID, false);
 	if (rc)
 		goto iqinf_exit;
 	smb2_set_related(&rqst[2]);
@@ -1529,35 +1620,32 @@ smb2_ioctl_query_info(const unsigned int xid,
 		if (le32_to_cpu(io_rsp->OutputCount) < qi.input_buffer_length)
 			qi.input_buffer_length = le32_to_cpu(io_rsp->OutputCount);
 		if (qi.input_buffer_length > 0 &&
-		    le32_to_cpu(io_rsp->OutputOffset) + qi.input_buffer_length > rsp_iov[1].iov_len) {
-			rc = -EFAULT;
-			goto iqinf_exit;
-		}
-		if (copy_to_user(&pqi->input_buffer_length, &qi.input_buffer_length,
-				 sizeof(qi.input_buffer_length))) {
-			rc = -EFAULT;
-			goto iqinf_exit;
-		}
+		    le32_to_cpu(io_rsp->OutputOffset) + qi.input_buffer_length
+		    > rsp_iov[1].iov_len)
+			goto e_fault;
+
+		if (copy_to_user(&pqi->input_buffer_length,
+				 &qi.input_buffer_length,
+				 sizeof(qi.input_buffer_length)))
+			goto e_fault;
+
 		if (copy_to_user((void __user *)pqi + sizeof(struct smb_query_info),
 				 (const void *)io_rsp + le32_to_cpu(io_rsp->OutputOffset),
-				 qi.input_buffer_length)) {
-			rc = -EFAULT;
-			goto iqinf_exit;
-		}
+				 qi.input_buffer_length))
+			goto e_fault;
 	} else {
 		pqi = (struct smb_query_info __user *)arg;
 		qi_rsp = (struct smb2_query_info_rsp *)rsp_iov[1].iov_base;
 		if (le32_to_cpu(qi_rsp->OutputBufferLength) < qi.input_buffer_length)
 			qi.input_buffer_length = le32_to_cpu(qi_rsp->OutputBufferLength);
-		if (copy_to_user(&pqi->input_buffer_length, &qi.input_buffer_length,
-				 sizeof(qi.input_buffer_length))) {
-			rc = -EFAULT;
-			goto iqinf_exit;
-		}
-		if (copy_to_user(pqi + 1, qi_rsp->Buffer, qi.input_buffer_length)) {
-			rc = -EFAULT;
-			goto iqinf_exit;
-		}
+		if (copy_to_user(&pqi->input_buffer_length,
+				 &qi.input_buffer_length,
+				 sizeof(qi.input_buffer_length)))
+			goto e_fault;
+
+		if (copy_to_user(pqi + 1, qi_rsp->Buffer,
+				 qi.input_buffer_length))
+			goto e_fault;
 	}
 
  iqinf_exit:
@@ -1573,6 +1661,10 @@ smb2_ioctl_query_info(const unsigned int xid,
 	free_rsp_buf(resp_buftype[1], rsp_iov[1].iov_base);
 	free_rsp_buf(resp_buftype[2], rsp_iov[2].iov_base);
 	return rc;
+
+e_fault:
+	rc = -EFAULT;
+	goto iqinf_exit;
 }
 
 static ssize_t
@@ -1989,6 +2081,67 @@ smb3_enum_snapshots(const unsigned int xid, struct cifs_tcon *tcon,
 	return rc;
 }
 
+
+
+static int
+smb3_notify(const unsigned int xid, struct file *pfile,
+	    void __user *ioc_buf)
+{
+	struct smb3_notify notify;
+	struct dentry *dentry = pfile->f_path.dentry;
+	struct inode *inode = file_inode(pfile);
+	struct cifs_sb_info *cifs_sb;
+	struct cifs_open_parms oparms;
+	struct cifs_fid fid;
+	struct cifs_tcon *tcon;
+	unsigned char *path = NULL;
+	__le16 *utf16_path = NULL;
+	u8 oplock = SMB2_OPLOCK_LEVEL_NONE;
+	int rc = 0;
+
+	path = build_path_from_dentry(dentry);
+	if (path == NULL)
+		return -ENOMEM;
+
+	cifs_sb = CIFS_SB(inode->i_sb);
+
+	utf16_path = cifs_convert_path_to_utf16(path + 1, cifs_sb);
+	if (utf16_path == NULL) {
+		rc = -ENOMEM;
+		goto notify_exit;
+	}
+
+	if (copy_from_user(&notify, ioc_buf, sizeof(struct smb3_notify))) {
+		rc = -EFAULT;
+		goto notify_exit;
+	}
+
+	tcon = cifs_sb_master_tcon(cifs_sb);
+	oparms.tcon = tcon;
+	oparms.desired_access = FILE_READ_ATTRIBUTES;
+	oparms.disposition = FILE_OPEN;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
+	oparms.fid = &fid;
+	oparms.reconnect = false;
+
+	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL,
+		       NULL);
+	if (rc)
+		goto notify_exit;
+
+	rc = SMB2_change_notify(xid, tcon, fid.persistent_fid, fid.volatile_fid,
+				notify.watch_tree, notify.completion_filter);
+
+	SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid);
+
+	cifs_dbg(FYI, "change notify for path %s rc %d\n", path, rc);
+
+notify_exit:
+	kfree(path);
+	kfree(utf16_path);
+	return rc;
+}
+
 static int
 smb2_query_dir_first(const unsigned int xid, struct cifs_tcon *tcon,
 		     const char *path, struct cifs_sb_info *cifs_sb,
@@ -1996,40 +2149,111 @@ smb2_query_dir_first(const unsigned int xid, struct cifs_tcon *tcon,
 		     struct cifs_search_info *srch_inf)
 {
 	__le16 *utf16_path;
-	int rc;
-	__u8 oplock = SMB2_OPLOCK_LEVEL_NONE;
+	struct smb_rqst rqst[2];
+	struct kvec rsp_iov[2];
+	int resp_buftype[2];
+	struct kvec open_iov[SMB2_CREATE_IOV_SIZE];
+	struct kvec qd_iov[SMB2_QUERY_DIRECTORY_IOV_SIZE];
+	int rc, flags = 0;
+	u8 oplock = SMB2_OPLOCK_LEVEL_NONE;
 	struct cifs_open_parms oparms;
+	struct smb2_query_directory_rsp *qd_rsp = NULL;
+	struct smb2_create_rsp *op_rsp = NULL;
 
 	utf16_path = cifs_convert_path_to_utf16(path, cifs_sb);
 	if (!utf16_path)
 		return -ENOMEM;
 
+	if (smb3_encryption_required(tcon))
+		flags |= CIFS_TRANSFORM_REQ;
+
+	memset(rqst, 0, sizeof(rqst));
+	resp_buftype[0] = resp_buftype[1] = CIFS_NO_BUFFER;
+	memset(rsp_iov, 0, sizeof(rsp_iov));
+
+	/* Open */
+	memset(&open_iov, 0, sizeof(open_iov));
+	rqst[0].rq_iov = open_iov;
+	rqst[0].rq_nvec = SMB2_CREATE_IOV_SIZE;
+
 	oparms.tcon = tcon;
 	oparms.desired_access = FILE_READ_ATTRIBUTES | FILE_READ_DATA;
 	oparms.disposition = FILE_OPEN;
-	if (backup_cred(cifs_sb))
-		oparms.create_options = CREATE_OPEN_BACKUP_INTENT;
-	else
-		oparms.create_options = 0;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.fid = fid;
 	oparms.reconnect = false;
 
-	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL);
-	kfree(utf16_path);
-	if (rc) {
-		cifs_dbg(FYI, "open dir failed rc=%d\n", rc);
-		return rc;
-	}
+	rc = SMB2_open_init(tcon, &rqst[0], &oplock, &oparms, utf16_path);
+	if (rc)
+		goto qdf_free;
+	smb2_set_next_command(tcon, &rqst[0]);
 
+	/* Query directory */
 	srch_inf->entries_in_buffer = 0;
 	srch_inf->index_of_last_entry = 2;
 
-	rc = SMB2_query_directory(xid, tcon, fid->persistent_fid,
-				  fid->volatile_fid, 0, srch_inf);
-	if (rc) {
-		cifs_dbg(FYI, "query directory failed rc=%d\n", rc);
+	memset(&qd_iov, 0, sizeof(qd_iov));
+	rqst[1].rq_iov = qd_iov;
+	rqst[1].rq_nvec = SMB2_QUERY_DIRECTORY_IOV_SIZE;
+
+	rc = SMB2_query_directory_init(xid, tcon, &rqst[1],
+				       COMPOUND_FID, COMPOUND_FID,
+				       0, srch_inf->info_level);
+	if (rc)
+		goto qdf_free;
+
+	smb2_set_related(&rqst[1]);
+
+	rc = compound_send_recv(xid, tcon->ses, flags, 2, rqst,
+				resp_buftype, rsp_iov);
+
+	/* If the open failed there is nothing to do */
+	op_rsp = (struct smb2_create_rsp *)rsp_iov[0].iov_base;
+	if (op_rsp == NULL || op_rsp->sync_hdr.Status != STATUS_SUCCESS) {
+		cifs_dbg(FYI, "query_dir_first: open failed rc=%d\n", rc);
+		goto qdf_free;
+	}
+	fid->persistent_fid = op_rsp->PersistentFileId;
+	fid->volatile_fid = op_rsp->VolatileFileId;
+
+	/* Anything else than ENODATA means a genuine error */
+	if (rc && rc != -ENODATA) {
 		SMB2_close(xid, tcon, fid->persistent_fid, fid->volatile_fid);
+		cifs_dbg(FYI, "query_dir_first: query directory failed rc=%d\n", rc);
+		trace_smb3_query_dir_err(xid, fid->persistent_fid,
+					 tcon->tid, tcon->ses->Suid, 0, 0, rc);
+		goto qdf_free;
 	}
+
+	atomic_inc(&tcon->num_remote_opens);
+
+	qd_rsp = (struct smb2_query_directory_rsp *)rsp_iov[1].iov_base;
+	if (qd_rsp->sync_hdr.Status == STATUS_NO_MORE_FILES) {
+		trace_smb3_query_dir_done(xid, fid->persistent_fid,
+					  tcon->tid, tcon->ses->Suid, 0, 0);
+		srch_inf->endOfSearch = true;
+		rc = 0;
+		goto qdf_free;
+	}
+
+	rc = smb2_parse_query_directory(tcon, &rsp_iov[1], resp_buftype[1],
+					srch_inf);
+	if (rc) {
+		trace_smb3_query_dir_err(xid, fid->persistent_fid, tcon->tid,
+			tcon->ses->Suid, 0, 0, rc);
+		goto qdf_free;
+	}
+	resp_buftype[1] = CIFS_NO_BUFFER;
+
+	trace_smb3_query_dir_done(xid, fid->persistent_fid, tcon->tid,
+			tcon->ses->Suid, 0, srch_inf->entries_in_buffer);
+
+ qdf_free:
+	kfree(utf16_path);
+	SMB2_open_free(&rqst[0]);
+	SMB2_query_directory_free(&rqst[1]);
+	free_rsp_buf(resp_buftype[0], rsp_iov[0].iov_base);
+	free_rsp_buf(resp_buftype[1], rsp_iov[1].iov_base);
 	return rc;
 }
 
@@ -2211,10 +2435,7 @@ smb2_query_info_compound(const unsigned int xid, struct cifs_tcon *tcon,
 	oparms.tcon = tcon;
 	oparms.desired_access = desired_access;
 	oparms.disposition = FILE_OPEN;
-	if (cifs_sb && backup_cred(cifs_sb))
-		oparms.create_options = CREATE_OPEN_BACKUP_INTENT;
-	else
-		oparms.create_options = 0;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
@@ -2240,7 +2461,7 @@ smb2_query_info_compound(const unsigned int xid, struct cifs_tcon *tcon,
 	rqst[2].rq_iov = close_iov;
 	rqst[2].rq_nvec = 1;
 
-	rc = SMB2_close_init(tcon, &rqst[2], COMPOUND_FID, COMPOUND_FID);
+	rc = SMB2_close_init(tcon, &rqst[2], COMPOUND_FID, COMPOUND_FID, false);
 	if (rc)
 		goto qic_exit;
 	smb2_set_related(&rqst[2]);
@@ -2270,7 +2491,7 @@ smb2_query_info_compound(const unsigned int xid, struct cifs_tcon *tcon,
 
 static int
 smb2_queryfs(const unsigned int xid, struct cifs_tcon *tcon,
-	     struct kstatfs *buf)
+	     struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
 {
 	struct smb2_query_info_rsp *rsp;
 	struct smb2_fs_full_size_info *info = NULL;
@@ -2285,7 +2506,7 @@ smb2_queryfs(const unsigned int xid, struct cifs_tcon *tcon,
 				      FS_FULL_SIZE_INFORMATION,
 				      SMB2_O_INFO_FILESYSTEM,
 				      sizeof(struct smb2_fs_full_size_info),
-				      &rsp_iov, &buftype, NULL);
+				      &rsp_iov, &buftype, cifs_sb);
 	if (rc)
 		goto qfs_exit;
 
@@ -2307,7 +2528,7 @@ qfs_exit:
 
 static int
 smb311_queryfs(const unsigned int xid, struct cifs_tcon *tcon,
-	     struct kstatfs *buf)
+	       struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
 {
 	int rc;
 	__le16 srch_path = 0; /* Null - open root of share */
@@ -2316,16 +2537,17 @@ smb311_queryfs(const unsigned int xid, struct cifs_tcon *tcon,
 	struct cifs_fid fid;
 
 	if (!tcon->posix_extensions)
-		return smb2_queryfs(xid, tcon, buf);
+		return smb2_queryfs(xid, tcon, cifs_sb, buf);
 
 	oparms.tcon = tcon;
 	oparms.desired_access = FILE_READ_ATTRIBUTES;
 	oparms.disposition = FILE_OPEN;
-	oparms.create_options = 0;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
-	rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL, NULL);
+	rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL,
+		       NULL, NULL);
 	if (rc)
 		return rc;
 
@@ -2590,6 +2812,7 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon,
 	struct smb2_create_rsp *create_rsp;
 	struct smb2_ioctl_rsp *ioctl_rsp;
 	struct reparse_data_buffer *reparse_buf;
+	int create_options = is_reparse_point ? OPEN_REPARSE_POINT : 0;
 	u32 plen;
 
 	cifs_dbg(FYI, "%s: path: %s\n", __func__, full_path);
@@ -2616,14 +2839,7 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon,
 	oparms.tcon = tcon;
 	oparms.desired_access = FILE_READ_ATTRIBUTES;
 	oparms.disposition = FILE_OPEN;
-
-	if (backup_cred(cifs_sb))
-		oparms.create_options = CREATE_OPEN_BACKUP_INTENT;
-	else
-		oparms.create_options = 0;
-	if (is_reparse_point)
-		oparms.create_options = OPEN_REPARSE_POINT;
-
+	oparms.create_options = cifs_create_options(cifs_sb, create_options);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
@@ -2640,7 +2856,10 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon,
 
 	rc = SMB2_ioctl_init(tcon, &rqst[1], fid.persistent_fid,
 			     fid.volatile_fid, FSCTL_GET_REPARSE_POINT,
-			     true /* is_fctl */, NULL, 0, CIFSMaxBufSize);
+			     true /* is_fctl */, NULL, 0,
+			     CIFSMaxBufSize -
+			     MAX_SMB2_CREATE_RESPONSE_SIZE -
+			     MAX_SMB2_CLOSE_RESPONSE_SIZE);
 	if (rc)
 		goto querty_exit;
 
@@ -2653,7 +2872,7 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon,
 	rqst[2].rq_iov = close_iov;
 	rqst[2].rq_nvec = 1;
 
-	rc = SMB2_close_init(tcon, &rqst[2], COMPOUND_FID, COMPOUND_FID);
+	rc = SMB2_close_init(tcon, &rqst[2], COMPOUND_FID, COMPOUND_FID, false);
 	if (rc)
 		goto querty_exit;
 
@@ -2799,11 +3018,6 @@ get_smb2_acl_by_path(struct cifs_sb_info *cifs_sb,
 	tcon = tlink_tcon(tlink);
 	xid = get_xid();
 
-	if (backup_cred(cifs_sb))
-		oparms.create_options = CREATE_OPEN_BACKUP_INTENT;
-	else
-		oparms.create_options = 0;
-
 	utf16_path = cifs_convert_path_to_utf16(path, cifs_sb);
 	if (!utf16_path) {
 		rc = -ENOMEM;
@@ -2814,10 +3028,12 @@ get_smb2_acl_by_path(struct cifs_sb_info *cifs_sb,
 	oparms.tcon = tcon;
 	oparms.desired_access = READ_CONTROL;
 	oparms.disposition = FILE_OPEN;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
-	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL);
+	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL,
+		       NULL);
 	kfree(utf16_path);
 	if (!rc) {
 		rc = SMB2_query_acl(xid, tlink_tcon(tlink), fid.persistent_fid,
@@ -2855,11 +3071,6 @@ set_smb2_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 	tcon = tlink_tcon(tlink);
 	xid = get_xid();
 
-	if (backup_cred(cifs_sb))
-		oparms.create_options = CREATE_OPEN_BACKUP_INTENT;
-	else
-		oparms.create_options = 0;
-
 	if (aclflag == CIFS_ACL_OWNER || aclflag == CIFS_ACL_GROUP)
 		access_flags = WRITE_OWNER;
 	else
@@ -2874,12 +3085,14 @@ set_smb2_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 
 	oparms.tcon = tcon;
 	oparms.desired_access = access_flags;
+	oparms.create_options = cifs_create_options(cifs_sb, 0);
 	oparms.disposition = FILE_OPEN;
 	oparms.path = path;
 	oparms.fid = &fid;
 	oparms.reconnect = false;
 
-	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL);
+	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL,
+		       NULL, NULL);
 	kfree(utf16_path);
 	if (!rc) {
 		rc = SMB2_set_acl(xid, tlink_tcon(tlink), fid.persistent_fid,
@@ -3038,28 +3251,36 @@ static long smb3_simple_falloc(struct file *file, struct cifs_tcon *tcon,
 		}
 
 	/*
+	 * Extending the file
+	 */
+	if ((keep_size == false) && i_size_read(inode) < off + len) {
+		rc = inode_newsize_ok(inode, off + len);
+		if (rc)
+			goto out;
+
+		if ((cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE) == 0)
+			smb2_set_sparse(xid, tcon, cfile, inode, false);
+
+		eof = cpu_to_le64(off + len);
+		rc = SMB2_set_eof(xid, tcon, cfile->fid.persistent_fid,
+				  cfile->fid.volatile_fid, cfile->pid, &eof);
+		if (rc == 0) {
+			cifsi->server_eof = off + len;
+			cifs_setsize(inode, off + len);
+			cifs_truncate_page(inode->i_mapping, inode->i_size);
+			truncate_setsize(inode, off + len);
+		}
+		goto out;
+	}
+
+	/*
 	 * Files are non-sparse by default so falloc may be a no-op
-	 * Must check if file sparse. If not sparse, and not extending
-	 * then no need to do anything since file already allocated
+	 * Must check if file sparse. If not sparse, and since we are not
+	 * extending then no need to do anything since file already allocated
 	 */
 	if ((cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE) == 0) {
-		if (keep_size == true)
-			rc = 0;
-		/* check if extending file */
-		else if (i_size_read(inode) >= off + len)
-			/* not extending file and already not sparse */
-			rc = 0;
-		/* BB: in future add else clause to extend file */
-		else
-			rc = -EOPNOTSUPP;
-		if (rc)
-			trace_smb3_falloc_err(xid, cfile->fid.persistent_fid,
-				tcon->tid, tcon->ses->Suid, off, len, rc);
-		else
-			trace_smb3_falloc_done(xid, cfile->fid.persistent_fid,
-				tcon->tid, tcon->ses->Suid, off, len);
-		free_xid(xid);
-		return rc;
+		rc = 0;
+		goto out;
 	}
 
 	if ((keep_size == true) || (i_size_read(inode) >= off + len)) {
@@ -3073,25 +3294,14 @@ static long smb3_simple_falloc(struct file *file, struct cifs_tcon *tcon,
 		 */
 		if ((off > 8192) || (off + len + 8192 < i_size_read(inode))) {
 			rc = -EOPNOTSUPP;
-			trace_smb3_falloc_err(xid, cfile->fid.persistent_fid,
-				tcon->tid, tcon->ses->Suid, off, len, rc);
-			free_xid(xid);
-			return rc;
-		}
-
-		smb2_set_sparse(xid, tcon, cfile, inode, false);
-		rc = 0;
-	} else {
-		smb2_set_sparse(xid, tcon, cfile, inode, false);
-		rc = 0;
-		if (i_size_read(inode) < off + len) {
-			eof = cpu_to_le64(off + len);
-			rc = SMB2_set_eof(xid, tcon, cfile->fid.persistent_fid,
-					  cfile->fid.volatile_fid, cfile->pid,
-					  &eof);
+			goto out;
 		}
 	}
 
+	smb2_set_sparse(xid, tcon, cfile, inode, false);
+	rc = 0;
+
+out:
 	if (rc)
 		trace_smb3_falloc_err(xid, cfile->fid.persistent_fid, tcon->tid,
 				tcon->ses->Suid, off, len, rc);
@@ -3130,7 +3340,7 @@ static loff_t smb3_llseek(struct file *file, struct cifs_tcon *tcon, loff_t offs
 	 * some servers (Windows2016) will not reflect recent writes in
 	 * QUERY_ALLOCATED_RANGES until SMB2_flush is called.
 	 */
-	wrcfile = find_writable_file(cifsi, false);
+	wrcfile = find_writable_file(cifsi, FIND_WR_ANY);
 	if (wrcfile) {
 		filemap_write_and_wait(inode->i_mapping);
 		smb2_flush_file(xid, tcon, &wrcfile->fid);
@@ -3219,7 +3429,7 @@ static int smb3_fiemap(struct cifs_tcon *tcon,
 	if (rc)
 		goto out;
 
-	if (out_data_len < sizeof(struct file_allocated_range_buffer)) {
+	if (out_data_len && out_data_len < sizeof(struct file_allocated_range_buffer)) {
 		rc = -EINVAL;
 		goto out;
 	}
@@ -3281,22 +3491,38 @@ static long smb3_fallocate(struct file *file, struct cifs_tcon *tcon, int mode,
 
 static void
 smb2_downgrade_oplock(struct TCP_Server_Info *server,
-			struct cifsInodeInfo *cinode, bool set_level2)
+		      struct cifsInodeInfo *cinode, __u32 oplock,
+		      unsigned int epoch, bool *purge_cache)
 {
-	if (set_level2)
-		server->ops->set_oplock_level(cinode, SMB2_OPLOCK_LEVEL_II,
-						0, NULL);
-	else
-		server->ops->set_oplock_level(cinode, 0, 0, NULL);
+	server->ops->set_oplock_level(cinode, oplock, 0, NULL);
 }
 
 static void
-smb21_downgrade_oplock(struct TCP_Server_Info *server,
-		       struct cifsInodeInfo *cinode, bool set_level2)
+smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock,
+		       unsigned int epoch, bool *purge_cache);
+
+static void
+smb3_downgrade_oplock(struct TCP_Server_Info *server,
+		       struct cifsInodeInfo *cinode, __u32 oplock,
+		       unsigned int epoch, bool *purge_cache)
 {
-	server->ops->set_oplock_level(cinode,
-				      set_level2 ? SMB2_LEASE_READ_CACHING_HE :
-				      0, 0, NULL);
+	unsigned int old_state = cinode->oplock;
+	unsigned int old_epoch = cinode->epoch;
+	unsigned int new_state;
+
+	if (epoch > old_epoch) {
+		smb21_set_oplock_level(cinode, oplock, 0, NULL);
+		cinode->epoch = epoch;
+	}
+
+	new_state = cinode->oplock;
+	*purge_cache = false;
+
+	if ((old_state & CIFS_CACHE_READ_FLG) != 0 &&
+	    (new_state & CIFS_CACHE_READ_FLG) == 0)
+		*purge_cache = true;
+	else if (old_state == new_state && (epoch - old_epoch > 1))
+		*purge_cache = true;
 }
 
 static void
@@ -3598,14 +3824,16 @@ smb2_get_enc_key(struct TCP_Server_Info *server, __u64 ses_id, int enc, u8 *key)
 	u8 *ses_enc_key;
 
 	spin_lock(&cifs_tcp_ses_lock);
-	list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) {
-		if (ses->Suid != ses_id)
-			continue;
-		ses_enc_key = enc ? ses->smb3encryptionkey :
-							ses->smb3decryptionkey;
-		memcpy(key, ses_enc_key, SMB3_SIGN_KEY_SIZE);
-		spin_unlock(&cifs_tcp_ses_lock);
-		return 0;
+	list_for_each_entry(server, &cifs_tcp_ses_list, tcp_ses_list) {
+		list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) {
+			if (ses->Suid == ses_id) {
+				ses_enc_key = enc ? ses->smb3encryptionkey :
+					ses->smb3decryptionkey;
+				memcpy(key, ses_enc_key, SMB3_SIGN_KEY_SIZE);
+				spin_unlock(&cifs_tcp_ses_lock);
+				return 0;
+			}
+		}
 	}
 	spin_unlock(&cifs_tcp_ses_lock);
 
@@ -3933,7 +4161,6 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid,
 	if (server->ops->is_session_expired &&
 	    server->ops->is_session_expired(buf)) {
 		cifs_reconnect(server);
-		wake_up(&server->response_q);
 		return -1;
 	}
 
@@ -4297,14 +4524,12 @@ smb3_receive_transform(struct TCP_Server_Info *server,
 		cifs_server_dbg(VFS, "Transform message is too small (%u)\n",
 			 pdu_length);
 		cifs_reconnect(server);
-		wake_up(&server->response_q);
 		return -ECONNABORTED;
 	}
 
 	if (pdu_length < orig_len + sizeof(struct smb2_transform_hdr)) {
 		cifs_server_dbg(VFS, "Transform message is broken\n");
 		cifs_reconnect(server);
-		wake_up(&server->response_q);
 		return -ECONNABORTED;
 	}
 
@@ -4345,7 +4570,6 @@ smb2_make_node(unsigned int xid, struct inode *inode,
 {
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
 	int rc = -EPERM;
-	int create_options = CREATE_NOT_DIR | CREATE_OPTION_SPECIAL;
 	FILE_ALL_INFO *buf = NULL;
 	struct cifs_io_parms io_parms;
 	__u32 oplock = 0;
@@ -4381,13 +4605,11 @@ smb2_make_node(unsigned int xid, struct inode *inode,
 		goto out;
 	}
 
-	if (backup_cred(cifs_sb))
-		create_options |= CREATE_OPEN_BACKUP_INTENT;
-
 	oparms.tcon = tcon;
 	oparms.cifs_sb = cifs_sb;
 	oparms.desired_access = GENERIC_WRITE;
-	oparms.create_options = create_options;
+	oparms.create_options = cifs_create_options(cifs_sb, CREATE_NOT_DIR |
+						    CREATE_OPTION_SPECIAL);
 	oparms.disposition = FILE_CREATE;
 	oparms.path = full_path;
 	oparms.fid = &fid;
@@ -4556,7 +4778,7 @@ struct smb_version_operations smb21_operations = {
 	.print_stats = smb2_print_stats,
 	.is_oplock_break = smb2_is_valid_oplock_break,
 	.handle_cancelled_mid = smb2_handle_cancelled_mid,
-	.downgrade_oplock = smb21_downgrade_oplock,
+	.downgrade_oplock = smb2_downgrade_oplock,
 	.need_neg = smb2_need_neg,
 	.negotiate = smb2_negotiate,
 	.negotiate_wsize = smb2_negotiate_wsize,
@@ -4616,6 +4838,7 @@ struct smb_version_operations smb21_operations = {
 	.wp_retry_size = smb2_wp_retry_size,
 	.dir_needs_close = smb2_dir_needs_close,
 	.enum_snapshots = smb3_enum_snapshots,
+	.notify = smb3_notify,
 	.get_dfs_refer = smb2_get_dfs_refer,
 	.select_sectype = smb2_select_sectype,
 #ifdef CONFIG_CIFS_XATTR
@@ -4656,7 +4879,7 @@ struct smb_version_operations smb30_operations = {
 	.dump_share_caps = smb2_dump_share_caps,
 	.is_oplock_break = smb2_is_valid_oplock_break,
 	.handle_cancelled_mid = smb2_handle_cancelled_mid,
-	.downgrade_oplock = smb21_downgrade_oplock,
+	.downgrade_oplock = smb3_downgrade_oplock,
 	.need_neg = smb2_need_neg,
 	.negotiate = smb2_negotiate,
 	.negotiate_wsize = smb3_negotiate_wsize,
@@ -4688,6 +4911,7 @@ struct smb_version_operations smb30_operations = {
 	.open = smb2_open_file,
 	.set_fid = smb2_set_fid,
 	.close = smb2_close_file,
+	.close_getattr = smb2_close_getattr,
 	.flush = smb2_flush_file,
 	.async_readv = smb2_async_readv,
 	.async_writev = smb2_async_writev,
@@ -4721,6 +4945,7 @@ struct smb_version_operations smb30_operations = {
 	.dir_needs_close = smb2_dir_needs_close,
 	.fallocate = smb3_fallocate,
 	.enum_snapshots = smb3_enum_snapshots,
+	.notify = smb3_notify,
 	.init_transform_rq = smb3_init_transform_rq,
 	.is_transform_hdr = smb3_is_transform_hdr,
 	.receive_transform = smb3_receive_transform,
@@ -4764,7 +4989,7 @@ struct smb_version_operations smb311_operations = {
 	.dump_share_caps = smb2_dump_share_caps,
 	.is_oplock_break = smb2_is_valid_oplock_break,
 	.handle_cancelled_mid = smb2_handle_cancelled_mid,
-	.downgrade_oplock = smb21_downgrade_oplock,
+	.downgrade_oplock = smb3_downgrade_oplock,
 	.need_neg = smb2_need_neg,
 	.negotiate = smb2_negotiate,
 	.negotiate_wsize = smb3_negotiate_wsize,
@@ -4797,6 +5022,7 @@ struct smb_version_operations smb311_operations = {
 	.open = smb2_open_file,
 	.set_fid = smb2_set_fid,
 	.close = smb2_close_file,
+	.close_getattr = smb2_close_getattr,
 	.flush = smb2_flush_file,
 	.async_readv = smb2_async_readv,
 	.async_writev = smb2_async_writev,
@@ -4830,6 +5056,7 @@ struct smb_version_operations smb311_operations = {
 	.dir_needs_close = smb2_dir_needs_close,
 	.fallocate = smb3_fallocate,
 	.enum_snapshots = smb3_enum_snapshots,
+	.notify = smb3_notify,
 	.init_transform_rq = smb3_init_transform_rq,
 	.is_transform_hdr = smb3_is_transform_hdr,
 	.receive_transform = smb3_receive_transform,
diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c
index 05149862aea4..b30aa3cdd845 100644
--- a/fs/cifs/smb2pdu.c
+++ b/fs/cifs/smb2pdu.c
@@ -193,9 +193,18 @@ static int __smb2_reconnect(const struct nls_table *nlsc,
 
 	for (it = dfs_cache_get_tgt_iterator(&tl); it;
 	     it = dfs_cache_get_next_tgt(&tl, it)) {
-		const char *tgt = dfs_cache_get_tgt_name(it);
+		const char *share, *prefix;
+		size_t share_len, prefix_len;
 
-		extract_unc_hostname(tgt, &dfs_host, &dfs_host_len);
+		rc = dfs_cache_get_tgt_share(it, &share, &share_len, &prefix,
+					     &prefix_len);
+		if (rc) {
+			cifs_dbg(VFS, "%s: failed to parse target share %d\n",
+				 __func__, rc);
+			continue;
+		}
+
+		extract_unc_hostname(share, &dfs_host, &dfs_host_len);
 
 		if (dfs_host_len != tcp_host_len
 		    || strncasecmp(dfs_host, tcp_host, dfs_host_len) != 0) {
@@ -206,11 +215,13 @@ static int __smb2_reconnect(const struct nls_table *nlsc,
 			continue;
 		}
 
-		scnprintf(tree, MAX_TREE_SIZE, "\\%s", tgt);
+		scnprintf(tree, MAX_TREE_SIZE, "\\%.*s", (int)share_len, share);
 
 		rc = SMB2_tcon(0, tcon->ses, tree, tcon, nlsc);
-		if (!rc)
+		if (!rc) {
+			rc = update_super_prepath(tcon, prefix, prefix_len);
 			break;
+		}
 		if (rc == -EREMOTE)
 			break;
 	}
@@ -252,7 +263,7 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon)
 	if (tcon == NULL)
 		return 0;
 
-	if (smb2_command == SMB2_TREE_CONNECT || smb2_command == SMB2_IOCTL)
+	if (smb2_command == SMB2_TREE_CONNECT)
 		return 0;
 
 	if (tcon->tidStatus == CifsExiting) {
@@ -312,7 +323,7 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon)
 		if (server->tcpStatus != CifsNeedReconnect)
 			break;
 
-		if (--retries)
+		if (retries && --retries)
 			continue;
 
 		/*
@@ -350,9 +361,14 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon)
 	}
 
 	rc = cifs_negotiate_protocol(0, tcon->ses);
-	if (!rc && tcon->ses->need_reconnect)
+	if (!rc && tcon->ses->need_reconnect) {
 		rc = cifs_setup_session(0, tcon->ses, nls_codepage);
-
+		if ((rc == -EACCES) && !tcon->retry) {
+			rc = -EHOSTDOWN;
+			mutex_unlock(&tcon->ses->session_mutex);
+			goto failed;
+		}
+	}
 	if (rc || !tcon->need_reconnect) {
 		mutex_unlock(&tcon->ses->session_mutex);
 		goto out;
@@ -373,7 +389,7 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon)
 	}
 
 	if (smb2_command != SMB2_INTERNAL_CMD)
-		queue_delayed_work(cifsiod_wq, &server->reconnect, 0);
+		mod_delayed_work(cifsiod_wq, &server->reconnect, 0);
 
 	atomic_inc(&tconInfoReconnectCount);
 out:
@@ -397,6 +413,7 @@ out:
 	case SMB2_SET_INFO:
 		rc = -EAGAIN;
 	}
+failed:
 	unload_nls(nls_codepage);
 	return rc;
 }
@@ -426,16 +443,9 @@ fill_small_buf(__le16 smb2_command, struct cifs_tcon *tcon, void *buf,
  * SMB information in the SMB header. If the return code is zero, this
  * function must have filled in request_buf pointer.
  */
-static int
-smb2_plain_req_init(__le16 smb2_command, struct cifs_tcon *tcon,
-		    void **request_buf, unsigned int *total_len)
+static int __smb2_plain_req_init(__le16 smb2_command, struct cifs_tcon *tcon,
+				  void **request_buf, unsigned int *total_len)
 {
-	int rc;
-
-	rc = smb2_reconnect(smb2_command, tcon);
-	if (rc)
-		return rc;
-
 	/* BB eventually switch this to SMB2 specific small buf size */
 	if (smb2_command == SMB2_SET_INFO)
 		*request_buf = cifs_buf_get();
@@ -456,7 +466,31 @@ smb2_plain_req_init(__le16 smb2_command, struct cifs_tcon *tcon,
 		cifs_stats_inc(&tcon->num_smbs_sent);
 	}
 
-	return rc;
+	return 0;
+}
+
+static int smb2_plain_req_init(__le16 smb2_command, struct cifs_tcon *tcon,
+			       void **request_buf, unsigned int *total_len)
+{
+	int rc;
+
+	rc = smb2_reconnect(smb2_command, tcon);
+	if (rc)
+		return rc;
+
+	return __smb2_plain_req_init(smb2_command, tcon, request_buf,
+				     total_len);
+}
+
+static int smb2_ioctl_req_init(u32 opcode, struct cifs_tcon *tcon,
+			       void **request_buf, unsigned int *total_len)
+{
+	/* Skip reconnect only for FSCTL_VALIDATE_NEGOTIATE_INFO IOCTLs */
+	if (opcode == FSCTL_VALIDATE_NEGOTIATE_INFO) {
+		return __smb2_plain_req_init(SMB2_IOCTL, tcon, request_buf,
+					     total_len);
+	}
+	return smb2_plain_req_init(SMB2_IOCTL, tcon, request_buf, total_len);
 }
 
 /* For explanation of negotiate contexts see MS-SMB2 section 2.2.3.1 */
@@ -537,7 +571,7 @@ static void
 assemble_neg_contexts(struct smb2_negotiate_req *req,
 		      struct TCP_Server_Info *server, unsigned int *total_len)
 {
-	char *pneg_ctxt = (char *)req;
+	char *pneg_ctxt;
 	unsigned int ctxt_len;
 
 	if (*total_len > 200) {
@@ -791,7 +825,7 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses)
 	struct kvec rsp_iov;
 	int rc = 0;
 	int resp_buftype;
-	struct TCP_Server_Info *server = ses->server;
+	struct TCP_Server_Info *server = cifs_ses_server(ses);
 	int blob_offset, blob_length;
 	char *security_blob;
 	int flags = CIFS_NEG_OP;
@@ -813,7 +847,7 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses)
 	memset(server->preauth_sha_hash, 0, SMB2_PREAUTH_HASH_SIZE);
 	memset(ses->preauth_sha_hash, 0, SMB2_PREAUTH_HASH_SIZE);
 
-	if (strcmp(ses->server->vals->version_string,
+	if (strcmp(server->vals->version_string,
 		   SMB3ANY_VERSION_STRING) == 0) {
 		req->Dialects[0] = cpu_to_le16(SMB30_PROT_ID);
 		req->Dialects[1] = cpu_to_le16(SMB302_PROT_ID);
@@ -829,7 +863,7 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses)
 		total_len += 8;
 	} else {
 		/* otherwise send specific dialect */
-		req->Dialects[0] = cpu_to_le16(ses->server->vals->protocol_id);
+		req->Dialects[0] = cpu_to_le16(server->vals->protocol_id);
 		req->DialectCount = cpu_to_le16(1);
 		total_len += 2;
 	}
@@ -1171,7 +1205,7 @@ SMB2_sess_alloc_buffer(struct SMB2_sess_data *sess_data)
 	int rc;
 	struct cifs_ses *ses = sess_data->ses;
 	struct smb2_sess_setup_req *req;
-	struct TCP_Server_Info *server = ses->server;
+	struct TCP_Server_Info *server = cifs_ses_server(ses);
 	unsigned int total_len;
 
 	rc = smb2_plain_req_init(SMB2_SESSION_SETUP, NULL, (void **) &req,
@@ -1179,13 +1213,21 @@ SMB2_sess_alloc_buffer(struct SMB2_sess_data *sess_data)
 	if (rc)
 		return rc;
 
-	/* First session, not a reauthenticate */
-	req->sync_hdr.SessionId = 0;
-
-	/* if reconnect, we need to send previous sess id, otherwise it is 0 */
-	req->PreviousSessionId = sess_data->previous_session;
-
-	req->Flags = 0; /* MBZ */
+	if (sess_data->ses->binding) {
+		req->sync_hdr.SessionId = sess_data->ses->Suid;
+		req->sync_hdr.Flags |= SMB2_FLAGS_SIGNED;
+		req->PreviousSessionId = 0;
+		req->Flags = SMB2_SESSION_REQ_FLAG_BINDING;
+	} else {
+		/* First session, not a reauthenticate */
+		req->sync_hdr.SessionId = 0;
+		/*
+		 * if reconnect, we need to send previous sess id
+		 * otherwise it is 0
+		 */
+		req->PreviousSessionId = sess_data->previous_session;
+		req->Flags = 0; /* MBZ */
+	}
 
 	/* enough to enable echos and oplocks and one max size write */
 	req->sync_hdr.CreditRequest = cpu_to_le16(130);
@@ -1258,28 +1300,33 @@ SMB2_sess_establish_session(struct SMB2_sess_data *sess_data)
 {
 	int rc = 0;
 	struct cifs_ses *ses = sess_data->ses;
+	struct TCP_Server_Info *server = cifs_ses_server(ses);
 
-	mutex_lock(&ses->server->srv_mutex);
-	if (ses->server->ops->generate_signingkey) {
-		rc = ses->server->ops->generate_signingkey(ses);
+	mutex_lock(&server->srv_mutex);
+	if (server->ops->generate_signingkey) {
+		rc = server->ops->generate_signingkey(ses);
 		if (rc) {
 			cifs_dbg(FYI,
 				"SMB3 session key generation failed\n");
-			mutex_unlock(&ses->server->srv_mutex);
+			mutex_unlock(&server->srv_mutex);
 			return rc;
 		}
 	}
-	if (!ses->server->session_estab) {
-		ses->server->sequence_number = 0x2;
-		ses->server->session_estab = true;
+	if (!server->session_estab) {
+		server->sequence_number = 0x2;
+		server->session_estab = true;
 	}
-	mutex_unlock(&ses->server->srv_mutex);
+	mutex_unlock(&server->srv_mutex);
 
 	cifs_dbg(FYI, "SMB2/3 session established successfully\n");
-	spin_lock(&GlobalMid_Lock);
-	ses->status = CifsGood;
-	ses->need_reconnect = false;
-	spin_unlock(&GlobalMid_Lock);
+	/* keep existing ses state if binding */
+	if (!ses->binding) {
+		spin_lock(&GlobalMid_Lock);
+		ses->status = CifsGood;
+		ses->need_reconnect = false;
+		spin_unlock(&GlobalMid_Lock);
+	}
+
 	return rc;
 }
 
@@ -1317,16 +1364,19 @@ SMB2_auth_kerberos(struct SMB2_sess_data *sess_data)
 		goto out_put_spnego_key;
 	}
 
-	ses->auth_key.response = kmemdup(msg->data, msg->sesskey_len,
-					 GFP_KERNEL);
-	if (!ses->auth_key.response) {
-		cifs_dbg(VFS,
-			"Kerberos can't allocate (%u bytes) memory",
-			msg->sesskey_len);
-		rc = -ENOMEM;
-		goto out_put_spnego_key;
+	/* keep session key if binding */
+	if (!ses->binding) {
+		ses->auth_key.response = kmemdup(msg->data, msg->sesskey_len,
+						 GFP_KERNEL);
+		if (!ses->auth_key.response) {
+			cifs_dbg(VFS,
+				 "Kerberos can't allocate (%u bytes) memory",
+				 msg->sesskey_len);
+			rc = -ENOMEM;
+			goto out_put_spnego_key;
+		}
+		ses->auth_key.len = msg->sesskey_len;
 	}
-	ses->auth_key.len = msg->sesskey_len;
 
 	sess_data->iov[1].iov_base = msg->data + msg->sesskey_len;
 	sess_data->iov[1].iov_len = msg->secblob_len;
@@ -1336,9 +1386,11 @@ SMB2_auth_kerberos(struct SMB2_sess_data *sess_data)
 		goto out_put_spnego_key;
 
 	rsp = (struct smb2_sess_setup_rsp *)sess_data->iov[0].iov_base;
-	ses->Suid = rsp->sync_hdr.SessionId;
-
-	ses->session_flags = le16_to_cpu(rsp->SessionFlags);
+	/* keep session id and flags if binding */
+	if (!ses->binding) {
+		ses->Suid = rsp->sync_hdr.SessionId;
+		ses->session_flags = le16_to_cpu(rsp->SessionFlags);
+	}
 
 	rc = SMB2_sess_establish_session(sess_data);
 out_put_spnego_key:
@@ -1432,9 +1484,11 @@ SMB2_sess_auth_rawntlmssp_negotiate(struct SMB2_sess_data *sess_data)
 
 	cifs_dbg(FYI, "rawntlmssp session setup challenge phase\n");
 
-
-	ses->Suid = rsp->sync_hdr.SessionId;
-	ses->session_flags = le16_to_cpu(rsp->SessionFlags);
+	/* keep existing ses id and flags if binding */
+	if (!ses->binding) {
+		ses->Suid = rsp->sync_hdr.SessionId;
+		ses->session_flags = le16_to_cpu(rsp->SessionFlags);
+	}
 
 out:
 	kfree(ntlmssp_blob);
@@ -1491,10 +1545,28 @@ SMB2_sess_auth_rawntlmssp_authenticate(struct SMB2_sess_data *sess_data)
 
 	rsp = (struct smb2_sess_setup_rsp *)sess_data->iov[0].iov_base;
 
-	ses->Suid = rsp->sync_hdr.SessionId;
-	ses->session_flags = le16_to_cpu(rsp->SessionFlags);
+	/* keep existing ses id and flags if binding */
+	if (!ses->binding) {
+		ses->Suid = rsp->sync_hdr.SessionId;
+		ses->session_flags = le16_to_cpu(rsp->SessionFlags);
+	}
 
 	rc = SMB2_sess_establish_session(sess_data);
+#ifdef CONFIG_CIFS_DEBUG_DUMP_KEYS
+	if (ses->server->dialect < SMB30_PROT_ID) {
+		cifs_dbg(VFS, "%s: dumping generated SMB2 session keys\n", __func__);
+		/*
+		 * The session id is opaque in terms of endianness, so we can't
+		 * print it as a long long. we dump it as we got it on the wire
+		 */
+		cifs_dbg(VFS, "Session Id    %*ph\n", (int)sizeof(ses->Suid),
+			 &ses->Suid);
+		cifs_dbg(VFS, "Session Key   %*ph\n",
+			 SMB2_NTLMV2_SESSKEY_SIZE, ses->auth_key.response);
+		cifs_dbg(VFS, "Signing Key   %*ph\n",
+			 SMB3_SIGN_KEY_SIZE, ses->auth_key.response);
+	}
+#endif
 out:
 	kfree(ntlmssp_blob);
 	SMB2_sess_free_buffer(sess_data);
@@ -1509,7 +1581,7 @@ SMB2_select_sec(struct cifs_ses *ses, struct SMB2_sess_data *sess_data)
 {
 	int type;
 
-	type = smb2_select_sectype(ses->server, ses->sectype);
+	type = smb2_select_sectype(cifs_ses_server(ses), ses->sectype);
 	cifs_dbg(FYI, "sess setup type %d\n", type);
 	if (type == Unspecified) {
 		cifs_dbg(VFS,
@@ -1537,7 +1609,7 @@ SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses,
 		const struct nls_table *nls_cp)
 {
 	int rc = 0;
-	struct TCP_Server_Info *server = ses->server;
+	struct TCP_Server_Info *server = cifs_ses_server(ses);
 	struct SMB2_sess_data *sess_data;
 
 	cifs_dbg(FYI, "Session Setup\n");
@@ -1563,7 +1635,7 @@ SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses,
 	/*
 	 * Initialize the session hash with the server one.
 	 */
-	memcpy(ses->preauth_sha_hash, ses->server->preauth_sha_hash,
+	memcpy(ses->preauth_sha_hash, server->preauth_sha_hash,
 	       SMB2_PREAUTH_HASH_SIZE);
 
 	while (sess_data->func)
@@ -1807,6 +1879,8 @@ SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon)
 	if ((tcon->need_reconnect) || (tcon->ses->need_reconnect))
 		return 0;
 
+	close_shroot_lease(&tcon->crfid);
+
 	rc = smb2_plain_req_init(SMB2_TREE_DISCONNECT, tcon, (void **) &req,
 			     &total_len);
 	if (rc)
@@ -1891,17 +1965,56 @@ parse_query_id_ctxt(struct create_context *cc, struct smb2_file_all_info *buf)
 	buf->IndexNumber = pdisk_id->DiskFileId;
 }
 
+static void
+parse_posix_ctxt(struct create_context *cc, struct smb2_file_all_info *info,
+		 struct create_posix_rsp *posix)
+{
+	int sid_len;
+	u8 *beg = (u8 *)cc + le16_to_cpu(cc->DataOffset);
+	u8 *end = beg + le32_to_cpu(cc->DataLength);
+	u8 *sid;
+
+	memset(posix, 0, sizeof(*posix));
+
+	posix->nlink = le32_to_cpu(*(__le32 *)(beg + 0));
+	posix->reparse_tag = le32_to_cpu(*(__le32 *)(beg + 4));
+	posix->mode = le32_to_cpu(*(__le32 *)(beg + 8));
+
+	sid = beg + 12;
+	sid_len = posix_info_sid_size(sid, end);
+	if (sid_len < 0) {
+		cifs_dbg(VFS, "bad owner sid in posix create response\n");
+		return;
+	}
+	memcpy(&posix->owner, sid, sid_len);
+
+	sid = sid + sid_len;
+	sid_len = posix_info_sid_size(sid, end);
+	if (sid_len < 0) {
+		cifs_dbg(VFS, "bad group sid in posix create response\n");
+		return;
+	}
+	memcpy(&posix->group, sid, sid_len);
+
+	cifs_dbg(FYI, "nlink=%d mode=%o reparse_tag=%x\n",
+		 posix->nlink, posix->mode, posix->reparse_tag);
+}
+
 void
 smb2_parse_contexts(struct TCP_Server_Info *server,
-		       struct smb2_create_rsp *rsp,
-		       unsigned int *epoch, char *lease_key, __u8 *oplock,
-		       struct smb2_file_all_info *buf)
+		    struct smb2_create_rsp *rsp,
+		    unsigned int *epoch, char *lease_key, __u8 *oplock,
+		    struct smb2_file_all_info *buf,
+		    struct create_posix_rsp *posix)
 {
 	char *data_offset;
 	struct create_context *cc;
 	unsigned int next;
 	unsigned int remaining;
 	char *name;
+	const char smb3_create_tag_posix[] = {0x93, 0xAD, 0x25, 0x50, 0x9C,
+					0xB4, 0x11, 0xE7, 0xB4, 0x23, 0x83,
+					0xDE, 0x96, 0x8B, 0xCD, 0x7C};
 
 	*oplock = 0;
 	data_offset = (char *)rsp + le32_to_cpu(rsp->CreateContextsOffset);
@@ -1921,6 +2034,16 @@ smb2_parse_contexts(struct TCP_Server_Info *server,
 		else if (buf && (le16_to_cpu(cc->NameLength) == 4) &&
 		    strncmp(name, SMB2_CREATE_QUERY_ON_DISK_ID, 4) == 0)
 			parse_query_id_ctxt(cc, buf);
+		else if ((le16_to_cpu(cc->NameLength) == 16)) {
+			if (posix &&
+			    memcmp(name, smb3_create_tag_posix, 16) == 0)
+				parse_posix_ctxt(cc, buf, posix);
+		}
+		/* else {
+			cifs_dbg(FYI, "Context not matched with len %d\n",
+				le16_to_cpu(cc->NameLength));
+			cifs_dump_mem("Cctxt name: ", name, 4);
+		} */
 
 		next = le32_to_cpu(cc->Next);
 		if (!next)
@@ -2149,6 +2272,77 @@ add_twarp_context(struct kvec *iov, unsigned int *num_iovec, __u64 timewarp)
 	return 0;
 }
 
+/* See MS-SMB2 2.2.13.2.2 and MS-DTYP 2.4.6 */
+static struct crt_sd_ctxt *
+create_sd_buf(umode_t mode, unsigned int *len)
+{
+	struct crt_sd_ctxt *buf;
+	struct cifs_ace *pace;
+	unsigned int sdlen, acelen;
+
+	*len = roundup(sizeof(struct crt_sd_ctxt) + sizeof(struct cifs_ace) * 2,
+			8);
+	buf = kzalloc(*len, GFP_KERNEL);
+	if (buf == NULL)
+		return buf;
+
+	sdlen = sizeof(struct smb3_sd) + sizeof(struct smb3_acl) +
+		 2 * sizeof(struct cifs_ace);
+
+	buf->ccontext.DataOffset = cpu_to_le16(offsetof
+					(struct crt_sd_ctxt, sd));
+	buf->ccontext.DataLength = cpu_to_le32(sdlen);
+	buf->ccontext.NameOffset = cpu_to_le16(offsetof
+				(struct crt_sd_ctxt, Name));
+	buf->ccontext.NameLength = cpu_to_le16(4);
+	/* SMB2_CREATE_SD_BUFFER_TOKEN is "SecD" */
+	buf->Name[0] = 'S';
+	buf->Name[1] = 'e';
+	buf->Name[2] = 'c';
+	buf->Name[3] = 'D';
+	buf->sd.Revision = 1;  /* Must be one see MS-DTYP 2.4.6 */
+	/*
+	 * ACL is "self relative" ie ACL is stored in contiguous block of memory
+	 * and "DP" ie the DACL is present
+	 */
+	buf->sd.Control = cpu_to_le16(ACL_CONTROL_SR | ACL_CONTROL_DP);
+
+	/* offset owner, group and Sbz1 and SACL are all zero */
+	buf->sd.OffsetDacl = cpu_to_le32(sizeof(struct smb3_sd));
+	buf->acl.AclRevision = ACL_REVISION; /* See 2.4.4.1 of MS-DTYP */
+
+	/* create one ACE to hold the mode embedded in reserved special SID */
+	pace = (struct cifs_ace *)(sizeof(struct crt_sd_ctxt) + (char *)buf);
+	acelen = setup_special_mode_ACE(pace, (__u64)mode);
+	/* and one more ACE to allow access for authenticated users */
+	pace = (struct cifs_ace *)(acelen + (sizeof(struct crt_sd_ctxt) +
+		(char *)buf));
+	acelen += setup_authusers_ACE(pace);
+	buf->acl.AclSize = cpu_to_le16(sizeof(struct cifs_acl) + acelen);
+	buf->acl.AceCount = cpu_to_le16(2);
+	return buf;
+}
+
+static int
+add_sd_context(struct kvec *iov, unsigned int *num_iovec, umode_t mode)
+{
+	struct smb2_create_req *req = iov[0].iov_base;
+	unsigned int num = *num_iovec;
+	unsigned int len = 0;
+
+	iov[num].iov_base = create_sd_buf(mode, &len);
+	if (iov[num].iov_base == NULL)
+		return -ENOMEM;
+	iov[num].iov_len = len;
+	if (!req->CreateContextsOffset)
+		req->CreateContextsOffset = cpu_to_le32(
+				sizeof(struct smb2_create_req) +
+				iov[num - 1].iov_len);
+	le32_add_cpu(&req->CreateContextsLength, len);
+	*num_iovec = num + 1;
+	return 0;
+}
+
 static struct crt_query_id_ctxt *
 create_query_id_buf(void)
 {
@@ -2521,7 +2715,9 @@ SMB2_open_init(struct cifs_tcon *tcon, struct smb_rqst *rqst, __u8 *oplock,
 			return rc;
 	}
 
-	if ((oparms->disposition == FILE_CREATE) &&
+	if ((oparms->disposition != FILE_OPEN) &&
+	    (oparms->cifs_sb) &&
+	    (oparms->cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) &&
 	    (oparms->mode != ACL_NO_MODE)) {
 		if (n_iov > 2) {
 			struct create_context *ccontext =
@@ -2530,7 +2726,8 @@ SMB2_open_init(struct cifs_tcon *tcon, struct smb_rqst *rqst, __u8 *oplock,
 				cpu_to_le32(iov[n_iov-1].iov_len);
 		}
 
-		/* rc = add_sd_context(iov, &n_iov, oparms->mode); */
+		cifs_dbg(FYI, "add sd with mode 0x%x\n", oparms->mode);
+		rc = add_sd_context(iov, &n_iov, oparms->mode);
 		if (rc)
 			return rc;
 	}
@@ -2565,6 +2762,7 @@ SMB2_open_free(struct smb_rqst *rqst)
 int
 SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 	  __u8 *oplock, struct smb2_file_all_info *buf,
+	  struct create_posix_rsp *posix,
 	  struct kvec *err_iov, int *buftype)
 {
 	struct smb_rqst rqst;
@@ -2627,6 +2825,7 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 	atomic_inc(&tcon->num_remote_opens);
 	oparms->fid->persistent_fid = rsp->PersistentFileId;
 	oparms->fid->volatile_fid = rsp->VolatileFileId;
+	oparms->fid->access = oparms->desired_access;
 #ifdef CONFIG_CIFS_DEBUG2
 	oparms->fid->mid = le64_to_cpu(rsp->sync_hdr.MessageId);
 #endif /* CIFS_DEBUG2 */
@@ -2642,7 +2841,7 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path,
 
 
 	smb2_parse_contexts(server, rsp, &oparms->fid->epoch,
-			    oparms->fid->lease_key, oplock, buf);
+			    oparms->fid->lease_key, oplock, buf, posix);
 creat_exit:
 	SMB2_open_free(&rqst);
 	free_rsp_buf(resp_buftype, rsp);
@@ -2661,7 +2860,7 @@ SMB2_ioctl_init(struct cifs_tcon *tcon, struct smb_rqst *rqst,
 	int rc;
 	char *in_data_buf;
 
-	rc = smb2_plain_req_init(SMB2_IOCTL, tcon, (void **) &req, &total_len);
+	rc = smb2_ioctl_req_init(opcode, tcon, (void **) &req, &total_len);
 	if (rc)
 		return rc;
 
@@ -2890,7 +3089,7 @@ SMB2_set_compression(const unsigned int xid, struct cifs_tcon *tcon,
 
 int
 SMB2_close_init(struct cifs_tcon *tcon, struct smb_rqst *rqst,
-		u64 persistent_fid, u64 volatile_fid)
+		u64 persistent_fid, u64 volatile_fid, bool query_attrs)
 {
 	struct smb2_close_req *req;
 	struct kvec *iov = rqst->rq_iov;
@@ -2903,6 +3102,10 @@ SMB2_close_init(struct cifs_tcon *tcon, struct smb_rqst *rqst,
 
 	req->PersistentFileId = persistent_fid;
 	req->VolatileFileId = volatile_fid;
+	if (query_attrs)
+		req->Flags = SMB2_CLOSE_FLAG_POSTQUERY_ATTRIB;
+	else
+		req->Flags = 0;
 	iov[0].iov_base = (char *)req;
 	iov[0].iov_len = total_len;
 
@@ -2917,8 +3120,9 @@ SMB2_close_free(struct smb_rqst *rqst)
 }
 
 int
-SMB2_close_flags(const unsigned int xid, struct cifs_tcon *tcon,
-		 u64 persistent_fid, u64 volatile_fid, int flags)
+__SMB2_close(const unsigned int xid, struct cifs_tcon *tcon,
+	     u64 persistent_fid, u64 volatile_fid,
+	     struct smb2_file_network_open_info *pbuf)
 {
 	struct smb_rqst rqst;
 	struct smb2_close_rsp *rsp = NULL;
@@ -2927,6 +3131,8 @@ SMB2_close_flags(const unsigned int xid, struct cifs_tcon *tcon,
 	struct kvec rsp_iov;
 	int resp_buftype = CIFS_NO_BUFFER;
 	int rc = 0;
+	int flags = 0;
+	bool query_attrs = false;
 
 	cifs_dbg(FYI, "Close\n");
 
@@ -2941,8 +3147,13 @@ SMB2_close_flags(const unsigned int xid, struct cifs_tcon *tcon,
 	rqst.rq_iov = iov;
 	rqst.rq_nvec = 1;
 
+	/* check if need to ask server to return timestamps in close response */
+	if (pbuf)
+		query_attrs = true;
+
 	trace_smb3_close_enter(xid, persistent_fid, tcon->tid, ses->Suid);
-	rc = SMB2_close_init(tcon, &rqst, persistent_fid, volatile_fid);
+	rc = SMB2_close_init(tcon, &rqst, persistent_fid, volatile_fid,
+			     query_attrs);
 	if (rc)
 		goto close_exit;
 
@@ -2954,25 +3165,40 @@ SMB2_close_flags(const unsigned int xid, struct cifs_tcon *tcon,
 		trace_smb3_close_err(xid, persistent_fid, tcon->tid, ses->Suid,
 				     rc);
 		goto close_exit;
-	} else
+	} else {
 		trace_smb3_close_done(xid, persistent_fid, tcon->tid,
 				      ses->Suid);
+		/*
+		 * Note that have to subtract 4 since struct network_open_info
+		 * has a final 4 byte pad that close response does not have
+		 */
+		if (pbuf)
+			memcpy(pbuf, (char *)&rsp->CreationTime, sizeof(*pbuf) - 4);
+	}
 
 	atomic_dec(&tcon->num_remote_opens);
-
-	/* BB FIXME - decode close response, update inode for caching */
-
 close_exit:
 	SMB2_close_free(&rqst);
 	free_rsp_buf(resp_buftype, rsp);
+
+	/* retry close in a worker thread if this one is interrupted */
+	if (rc == -EINTR) {
+		int tmp_rc;
+
+		tmp_rc = smb2_handle_cancelled_close(tcon, persistent_fid,
+						     volatile_fid);
+		if (tmp_rc)
+			cifs_dbg(VFS, "handle cancelled close fid 0x%llx returned error %d\n",
+				 persistent_fid, tmp_rc);
+	}
 	return rc;
 }
 
 int
 SMB2_close(const unsigned int xid, struct cifs_tcon *tcon,
-	   u64 persistent_fid, u64 volatile_fid)
+		u64 persistent_fid, u64 volatile_fid)
 {
-	return SMB2_close_flags(xid, tcon, persistent_fid, volatile_fid, 0);
+	return __SMB2_close(xid, tcon, persistent_fid, volatile_fid, NULL);
 }
 
 int
@@ -3214,6 +3440,7 @@ SMB2_notify_init(const unsigned int xid, struct smb_rqst *rqst,
 
 	req->PersistentFileId = persistent_fid;
 	req->VolatileFileId = volatile_fid;
+	/* See note 354 of MS-SMB2, 64K max */
 	req->OutputBufferLength =
 		cpu_to_le32(SMB2_MAX_BUFFER_SIZE - MAX_SMB2_HDR_SIZE);
 	req->CompletionFilter = cpu_to_le32(completion_filter);
@@ -3386,7 +3613,7 @@ SMB2_echo(struct TCP_Server_Info *server)
 
 	if (server->tcpStatus == CifsNeedNegotiate) {
 		/* No need to send echo on newly established connections */
-		queue_delayed_work(cifsiod_wq, &server->reconnect, 0);
+		mod_delayed_work(cifsiod_wq, &server->reconnect, 0);
 		return rc;
 	}
 
@@ -3880,6 +4107,9 @@ smb2_writev_callback(struct mid_q_entry *mid)
 				     wdata->cfile->fid.persistent_fid,
 				     tcon->tid, tcon->ses->Suid, wdata->offset,
 				     wdata->bytes, wdata->result);
+		if (wdata->result == -ENOSPC)
+			printk_once(KERN_WARNING "Out of space writing to %s\n",
+				    tcon->treeName);
 	} else
 		trace_smb3_write_done(0 /* no xid */,
 				      wdata->cfile->fid.persistent_fid,
@@ -4110,8 +4340,104 @@ SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms,
 	return rc;
 }
 
+int posix_info_sid_size(const void *beg, const void *end)
+{
+	size_t subauth;
+	int total;
+
+	if (beg + 1 > end)
+		return -1;
+
+	subauth = *(u8 *)(beg+1);
+	if (subauth < 1 || subauth > 15)
+		return -1;
+
+	total = 1 + 1 + 6 + 4*subauth;
+	if (beg + total > end)
+		return -1;
+
+	return total;
+}
+
+int posix_info_parse(const void *beg, const void *end,
+		     struct smb2_posix_info_parsed *out)
+
+{
+	int total_len = 0;
+	int sid_len;
+	int name_len;
+	const void *owner_sid;
+	const void *group_sid;
+	const void *name;
+
+	/* if no end bound given, assume payload to be correct */
+	if (!end) {
+		const struct smb2_posix_info *p = beg;
+
+		end = beg + le32_to_cpu(p->NextEntryOffset);
+		/* last element will have a 0 offset, pick a sensible bound */
+		if (end == beg)
+			end += 0xFFFF;
+	}
+
+	/* check base buf */
+	if (beg + sizeof(struct smb2_posix_info) > end)
+		return -1;
+	total_len = sizeof(struct smb2_posix_info);
+
+	/* check owner sid */
+	owner_sid = beg + total_len;
+	sid_len = posix_info_sid_size(owner_sid, end);
+	if (sid_len < 0)
+		return -1;
+	total_len += sid_len;
+
+	/* check group sid */
+	group_sid = beg + total_len;
+	sid_len = posix_info_sid_size(group_sid, end);
+	if (sid_len < 0)
+		return -1;
+	total_len += sid_len;
+
+	/* check name len */
+	if (beg + total_len + 4 > end)
+		return -1;
+	name_len = le32_to_cpu(*(__le32 *)(beg + total_len));
+	if (name_len < 1 || name_len > 0xFFFF)
+		return -1;
+	total_len += 4;
+
+	/* check name */
+	name = beg + total_len;
+	if (name + name_len > end)
+		return -1;
+	total_len += name_len;
+
+	if (out) {
+		out->base = beg;
+		out->size = total_len;
+		out->name_len = name_len;
+		out->name = name;
+		memcpy(&out->owner, owner_sid,
+		       posix_info_sid_size(owner_sid, end));
+		memcpy(&out->group, group_sid,
+		       posix_info_sid_size(group_sid, end));
+	}
+	return total_len;
+}
+
+static int posix_info_extra_size(const void *beg, const void *end)
+{
+	int len = posix_info_parse(beg, end, NULL);
+
+	if (len < 0)
+		return -1;
+	return len - sizeof(struct smb2_posix_info);
+}
+
 static unsigned int
-num_entries(char *bufstart, char *end_of_buf, char **lastentry, size_t size)
+num_entries(int infotype, char *bufstart, char *end_of_buf, char **lastentry,
+	    size_t size)
 {
 	int len;
 	unsigned int entrycount = 0;
@@ -4135,8 +4461,13 @@ num_entries(char *bufstart, char *end_of_buf, char **lastentry, size_t size)
 		entryptr = entryptr + next_offset;
 		dir_info = (FILE_DIRECTORY_INFO *)entryptr;
 
-		len = le32_to_cpu(dir_info->FileNameLength);
-		if (entryptr + len < entryptr ||
+		if (infotype == SMB_FIND_FILE_POSIX_INFO)
+			len = posix_info_extra_size(entryptr, end_of_buf);
+		else
+			len = le32_to_cpu(dir_info->FileNameLength);
+
+		if (len < 0 ||
+		    entryptr + len < entryptr ||
 		    entryptr + len > end_of_buf ||
 		    entryptr + len + size > end_of_buf) {
 			cifs_dbg(VFS, "directory entry name would overflow frame end of buf %p\n",
@@ -4158,56 +4489,41 @@ num_entries(char *bufstart, char *end_of_buf, char **lastentry, size_t size)
 /*
  * Readdir/FindFirst
  */
-int
-SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
-		     u64 persistent_fid, u64 volatile_fid, int index,
-		     struct cifs_search_info *srch_inf)
+int SMB2_query_directory_init(const unsigned int xid,
+			      struct cifs_tcon *tcon, struct smb_rqst *rqst,
+			      u64 persistent_fid, u64 volatile_fid,
+			      int index, int info_level)
 {
-	struct smb_rqst rqst;
+	struct TCP_Server_Info *server = tcon->ses->server;
 	struct smb2_query_directory_req *req;
-	struct smb2_query_directory_rsp *rsp = NULL;
-	struct kvec iov[2];
-	struct kvec rsp_iov;
-	int rc = 0;
-	int len;
-	int resp_buftype = CIFS_NO_BUFFER;
 	unsigned char *bufptr;
-	struct TCP_Server_Info *server;
-	struct cifs_ses *ses = tcon->ses;
 	__le16 asteriks = cpu_to_le16('*');
-	char *end_of_smb;
-	unsigned int output_size = CIFSMaxBufSize;
-	size_t info_buf_size;
-	int flags = 0;
+	unsigned int output_size = CIFSMaxBufSize -
+		MAX_SMB2_CREATE_RESPONSE_SIZE -
+		MAX_SMB2_CLOSE_RESPONSE_SIZE;
 	unsigned int total_len;
-
-	if (ses && (ses->server))
-		server = ses->server;
-	else
-		return -EIO;
+	struct kvec *iov = rqst->rq_iov;
+	int len, rc;
 
 	rc = smb2_plain_req_init(SMB2_QUERY_DIRECTORY, tcon, (void **) &req,
 			     &total_len);
 	if (rc)
 		return rc;
 
-	if (smb3_encryption_required(tcon))
-		flags |= CIFS_TRANSFORM_REQ;
-
-	switch (srch_inf->info_level) {
+	switch (info_level) {
 	case SMB_FIND_FILE_DIRECTORY_INFO:
 		req->FileInformationClass = FILE_DIRECTORY_INFORMATION;
-		info_buf_size = sizeof(FILE_DIRECTORY_INFO) - 1;
 		break;
 	case SMB_FIND_FILE_ID_FULL_DIR_INFO:
 		req->FileInformationClass = FILEID_FULL_DIRECTORY_INFORMATION;
-		info_buf_size = sizeof(SEARCH_ID_FULL_DIR_INFO) - 1;
+		break;
+	case SMB_FIND_FILE_POSIX_INFO:
+		req->FileInformationClass = SMB_FIND_FILE_POSIX_INFO;
 		break;
 	default:
 		cifs_tcon_dbg(VFS, "info level %u isn't supported\n",
-			 srch_inf->info_level);
-		rc = -EINVAL;
-		goto qdir_exit;
+			info_level);
+		return -EINVAL;
 	}
 
 	req->FileIndex = cpu_to_le32(index);
@@ -4236,39 +4552,55 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 	iov[1].iov_base = (char *)(req->Buffer);
 	iov[1].iov_len = len;
 
-	memset(&rqst, 0, sizeof(struct smb_rqst));
-	rqst.rq_iov = iov;
-	rqst.rq_nvec = 2;
-
 	trace_smb3_query_dir_enter(xid, persistent_fid, tcon->tid,
 			tcon->ses->Suid, index, output_size);
 
-	rc = cifs_send_recv(xid, ses, &rqst, &resp_buftype, flags, &rsp_iov);
-	cifs_small_buf_release(req);
-	rsp = (struct smb2_query_directory_rsp *)rsp_iov.iov_base;
+	return 0;
+}
 
-	if (rc) {
-		if (rc == -ENODATA &&
-		    rsp->sync_hdr.Status == STATUS_NO_MORE_FILES) {
-			trace_smb3_query_dir_done(xid, persistent_fid,
-				tcon->tid, tcon->ses->Suid, index, 0);
-			srch_inf->endOfSearch = true;
-			rc = 0;
-		} else {
-			trace_smb3_query_dir_err(xid, persistent_fid, tcon->tid,
-				tcon->ses->Suid, index, 0, rc);
-			cifs_stats_fail_inc(tcon, SMB2_QUERY_DIRECTORY_HE);
-		}
-		goto qdir_exit;
+void SMB2_query_directory_free(struct smb_rqst *rqst)
+{
+	if (rqst && rqst->rq_iov) {
+		cifs_small_buf_release(rqst->rq_iov[0].iov_base); /* request */
+	}
+}
+
+int
+smb2_parse_query_directory(struct cifs_tcon *tcon,
+			   struct kvec *rsp_iov,
+			   int resp_buftype,
+			   struct cifs_search_info *srch_inf)
+{
+	struct smb2_query_directory_rsp *rsp;
+	size_t info_buf_size;
+	char *end_of_smb;
+	int rc;
+
+	rsp = (struct smb2_query_directory_rsp *)rsp_iov->iov_base;
+
+	switch (srch_inf->info_level) {
+	case SMB_FIND_FILE_DIRECTORY_INFO:
+		info_buf_size = sizeof(FILE_DIRECTORY_INFO) - 1;
+		break;
+	case SMB_FIND_FILE_ID_FULL_DIR_INFO:
+		info_buf_size = sizeof(SEARCH_ID_FULL_DIR_INFO) - 1;
+		break;
+	case SMB_FIND_FILE_POSIX_INFO:
+		/* note that posix payload are variable size */
+		info_buf_size = sizeof(struct smb2_posix_info);
+		break;
+	default:
+		cifs_tcon_dbg(VFS, "info level %u isn't supported\n",
+			 srch_inf->info_level);
+		return -EINVAL;
 	}
 
 	rc = smb2_validate_iov(le16_to_cpu(rsp->OutputBufferOffset),
-			       le32_to_cpu(rsp->OutputBufferLength), &rsp_iov,
+			       le32_to_cpu(rsp->OutputBufferLength), rsp_iov,
 			       info_buf_size);
 	if (rc) {
-		trace_smb3_query_dir_err(xid, persistent_fid, tcon->tid,
-			tcon->ses->Suid, index, 0, rc);
-		goto qdir_exit;
+		cifs_tcon_dbg(VFS, "bad info payload");
+		return rc;
 	}
 
 	srch_inf->unicode = true;
@@ -4282,10 +4614,15 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 	srch_inf->ntwrk_buf_start = (char *)rsp;
 	srch_inf->srch_entries_start = srch_inf->last_entry =
 		(char *)rsp + le16_to_cpu(rsp->OutputBufferOffset);
-	end_of_smb = rsp_iov.iov_len + (char *)rsp;
-	srch_inf->entries_in_buffer =
-			num_entries(srch_inf->srch_entries_start, end_of_smb,
-				    &srch_inf->last_entry, info_buf_size);
+	end_of_smb = rsp_iov->iov_len + (char *)rsp;
+
+	srch_inf->entries_in_buffer = num_entries(
+		srch_inf->info_level,
+		srch_inf->srch_entries_start,
+		end_of_smb,
+		&srch_inf->last_entry,
+		info_buf_size);
+
 	srch_inf->index_of_last_entry += srch_inf->entries_in_buffer;
 	cifs_dbg(FYI, "num entries %d last_index %lld srch start %p srch end %p\n",
 		 srch_inf->entries_in_buffer, srch_inf->index_of_last_entry,
@@ -4297,11 +4634,72 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 	else
 		cifs_tcon_dbg(VFS, "illegal search buffer type\n");
 
+	return 0;
+}
+
+int
+SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
+		     u64 persistent_fid, u64 volatile_fid, int index,
+		     struct cifs_search_info *srch_inf)
+{
+	struct smb_rqst rqst;
+	struct kvec iov[SMB2_QUERY_DIRECTORY_IOV_SIZE];
+	struct smb2_query_directory_rsp *rsp = NULL;
+	int resp_buftype = CIFS_NO_BUFFER;
+	struct kvec rsp_iov;
+	int rc = 0;
+	struct cifs_ses *ses = tcon->ses;
+	int flags = 0;
+
+	if (!ses || !(ses->server))
+		return -EIO;
+
+	if (smb3_encryption_required(tcon))
+		flags |= CIFS_TRANSFORM_REQ;
+
+	memset(&rqst, 0, sizeof(struct smb_rqst));
+	memset(&iov, 0, sizeof(iov));
+	rqst.rq_iov = iov;
+	rqst.rq_nvec = SMB2_QUERY_DIRECTORY_IOV_SIZE;
+
+	rc = SMB2_query_directory_init(xid, tcon, &rqst, persistent_fid,
+				       volatile_fid, index,
+				       srch_inf->info_level);
+	if (rc)
+		goto qdir_exit;
+
+	rc = cifs_send_recv(xid, ses, &rqst, &resp_buftype, flags, &rsp_iov);
+	rsp = (struct smb2_query_directory_rsp *)rsp_iov.iov_base;
+
+	if (rc) {
+		if (rc == -ENODATA &&
+		    rsp->sync_hdr.Status == STATUS_NO_MORE_FILES) {
+			trace_smb3_query_dir_done(xid, persistent_fid,
+				tcon->tid, tcon->ses->Suid, index, 0);
+			srch_inf->endOfSearch = true;
+			rc = 0;
+		} else {
+			trace_smb3_query_dir_err(xid, persistent_fid, tcon->tid,
+				tcon->ses->Suid, index, 0, rc);
+			cifs_stats_fail_inc(tcon, SMB2_QUERY_DIRECTORY_HE);
+		}
+		goto qdir_exit;
+	}
+
+	rc = smb2_parse_query_directory(tcon, &rsp_iov,	resp_buftype,
+					srch_inf);
+	if (rc) {
+		trace_smb3_query_dir_err(xid, persistent_fid, tcon->tid,
+			tcon->ses->Suid, index, 0, rc);
+		goto qdir_exit;
+	}
+	resp_buftype = CIFS_NO_BUFFER;
+
 	trace_smb3_query_dir_done(xid, persistent_fid, tcon->tid,
 			tcon->ses->Suid, index, srch_inf->entries_in_buffer);
-	return rc;
 
 qdir_exit:
+	SMB2_query_directory_free(&rqst);
 	free_rsp_buf(resp_buftype, rsp);
 	return rc;
 }
diff --git a/fs/cifs/smb2pdu.h b/fs/cifs/smb2pdu.h
index 0abfde6d0b05..10acf90f858d 100644
--- a/fs/cifs/smb2pdu.h
+++ b/fs/cifs/smb2pdu.h
@@ -25,6 +25,7 @@
 #define _SMB2PDU_H
 
 #include <net/sock.h>
+#include <cifsacl.h>
 
 /*
  * Note that, due to trying to use names similar to the protocol specifications,
@@ -90,6 +91,7 @@
 
 #define SMB2_PROTO_NUMBER cpu_to_le32(0x424d53fe)
 #define SMB2_TRANSFORM_PROTO_NUM cpu_to_le32(0x424d53fd)
+#define SMB2_COMPRESSION_TRANSFORM_ID cpu_to_le32(0x424d53fc)
 
 /*
  * SMB2 Header Definition
@@ -118,6 +120,9 @@ struct smb2_sync_hdr {
 	__u8   Signature[16];
 } __packed;
 
+/* The total header size for SMB2 read and write */
+#define SMB2_READWRITE_PDU_HEADER_SIZE (48 + sizeof(struct smb2_sync_hdr))
+
 struct smb2_sync_pdu {
 	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize2; /* size of wct area (varies, request specific) */
@@ -126,16 +131,33 @@ struct smb2_sync_pdu {
 #define SMB3_AES128CCM_NONCE 11
 #define SMB3_AES128GCM_NONCE 12
 
+/* Transform flags (for 3.0 dialect this flag indicates CCM */
+#define TRANSFORM_FLAG_ENCRYPTED	0x0001
 struct smb2_transform_hdr {
 	__le32 ProtocolId;	/* 0xFD 'S' 'M' 'B' */
 	__u8   Signature[16];
 	__u8   Nonce[16];
 	__le32 OriginalMessageSize;
 	__u16  Reserved1;
-	__le16 Flags; /* EncryptionAlgorithm */
+	__le16 Flags; /* EncryptionAlgorithm for 3.0, enc enabled for 3.1.1 */
 	__u64  SessionId;
 } __packed;
 
+/* See MS-SMB2 2.2.42.1 */
+struct compression_playload_header {
+	__le16	AlgorithmId;
+	__le16	Reserved;
+	__le32	Length;
+} __packed;
+
+/* See MS-SMB2 2.2.42.2 */
+struct compression_pattern_payload_v1 {
+	__le16	Pattern;
+	__le16	Reserved1;
+	__le16	Reserved2;
+	__le32	Repetitions;
+} __packed;
+
 /*
  *	SMB2 flag definitions
  */
@@ -181,7 +203,7 @@ struct smb2_symlink_err_rsp {
 	__le16 PrintNameOffset;
 	__le16 PrintNameLength;
 	__le32 Flags;
-	__u8  PathBuffer[0];
+	__u8  PathBuffer[];
 } __packed;
 
 /* SMB 3.1.1 and later dialects. See MS-SMB2 section 2.2.2.1 */
@@ -191,6 +213,10 @@ struct smb2_error_context_rsp {
 	__u8  ErrorContextData; /* ErrorDataLength long array */
 } __packed;
 
+/* ErrorId values */
+#define SMB2_ERROR_ID_DEFAULT		0x00000000
+#define SMB2_ERROR_ID_SHARE_REDIRECT	cpu_to_le32(0x72645253)	/* "rdRS" */
+
 /* Defines for Type field below (see MS-SMB2 2.2.2.2.2.1) */
 #define MOVE_DST_IPADDR_V4	cpu_to_le32(0x00000001)
 #define MOVE_DST_IPADDR_V6	cpu_to_le32(0x00000002)
@@ -209,7 +235,7 @@ struct share_redirect_error_context_rsp {
 	__le16 Flags;
 	__le16 TargetType;
 	__le32 IPAddrCount;
-	struct move_dst_ipaddr IpAddrMoveList[0];
+	struct move_dst_ipaddr IpAddrMoveList[];
 	/* __u8 ResourceName[] */ /* Name of share as counted Unicode string */
 } __packed;
 
@@ -306,11 +332,17 @@ struct smb2_encryption_neg_context {
 #define SMB3_COMPRESS_LZNT1	cpu_to_le16(0x0001)
 #define SMB3_COMPRESS_LZ77	cpu_to_le16(0x0002)
 #define SMB3_COMPRESS_LZ77_HUFF	cpu_to_le16(0x0003)
+/* Pattern scanning algorithm See MS-SMB2 3.1.4.4.1 */
+#define SMB3_COMPRESS_PATTERN	cpu_to_le16(0x0004)
+
+/* Compression Flags */
+#define SMB2_COMPRESSION_CAPABILITIES_FLAG_NONE		cpu_to_le32(0x00000000)
+#define SMB2_COMPRESSION_CAPABILITIES_FLAG_CHAINED	cpu_to_le32(0x00000001)
 
 struct smb2_compression_capabilities_context {
 	__le16	ContextType; /* 3 */
 	__le16  DataLength;
-	__u32	Reserved;
+	__u32	Flags;
 	__le16	CompressionAlgorithmCount;
 	__u16	Padding;
 	__u32	Reserved1;
@@ -325,7 +357,7 @@ struct smb2_netname_neg_context {
 	__le16	ContextType; /* 0x100 */
 	__le16	DataLength;
 	__le32	Reserved;
-	__le16	NetName[0]; /* hostname of target converted to UCS-2 */
+	__le16	NetName[]; /* hostname of target converted to UCS-2 */
 } __packed;
 
 #define POSIX_CTXT_DATA_LEN	16
@@ -405,7 +437,7 @@ struct smb2_logoff_rsp {
 struct smb2_tree_connect_req {
 	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize;	/* Must be 9 */
-	__le16 Reserved; /* Flags in SMB3.1.1 */
+	__le16 Flags; /* Reserved MBZ for dialects prior to SMB3.1.1 */
 	__le16 PathOffset;
 	__le16 PathLength;
 	__u8   Buffer[1];	/* variable length */
@@ -420,13 +452,13 @@ struct tree_connect_contexts {
 	__le16 ContextType;
 	__le16 DataLength;
 	__le32 Reserved;
-	__u8   Data[0];
+	__u8   Data[];
 } __packed;
 
 /* Remoted identity tree connect context structures - see MS-SMB2 2.2.9.2.1 */
 struct smb3_blob_data {
 	__le16 BlobSize;
-	__u8   BlobData[0];
+	__u8   BlobData[];
 } __packed;
 
 /* Valid values for Attr */
@@ -476,14 +508,14 @@ struct remoted_identity_tcon_context {
 	__le16 DeviceGroups; /* offset to SID_ARRAY_DATA struct */
 	__le16 UserClaims; /* offset to BLOB_DATA struct */
 	__le16 DeviceClaims; /* offset to BLOB_DATA struct */
-	__u8   TicketInfo[0]; /* variable length buf - remoted identity data */
+	__u8   TicketInfo[]; /* variable length buf - remoted identity data */
 } __packed;
 
 struct smb2_tree_connect_req_extension {
 	__le32 TreeConnectContextOffset;
 	__le16 TreeConnectContextCount;
 	__u8  Reserved[10];
-	__u8  PathName[0]; /* variable sized array */
+	__u8  PathName[]; /* variable sized array */
 	/* followed by array of TreeConnectContexts */
 } __packed;
 
@@ -632,7 +664,7 @@ struct smb2_tree_disconnect_rsp {
 			| FILE_WRITE_EA_LE | FILE_WRITE_ATTRIBUTES_LE)
 #define FILE_EXEC_RIGHTS_LE (FILE_EXECUTE_LE)
 
-/* Impersonation Levels */
+/* Impersonation Levels. See MS-WPO section 9.7 and MSDN-IMPERS */
 #define IL_ANONYMOUS		cpu_to_le32(0x00000000)
 #define IL_IDENTIFICATION	cpu_to_le32(0x00000001)
 #define IL_IMPERSONATION	cpu_to_le32(0x00000002)
@@ -688,7 +720,7 @@ struct smb2_create_req {
 	__le16 NameLength;
 	__le32 CreateContextsOffset;
 	__le32 CreateContextsLength;
-	__u8   Buffer[0];
+	__u8   Buffer[];
 } __packed;
 
 /*
@@ -726,7 +758,7 @@ struct create_context {
 	__le16 Reserved;
 	__le16 DataOffset;
 	__le32 DataLength;
-	__u8 Buffer[0];
+	__u8 Buffer[];
 } __packed;
 
 #define SMB2_LEASE_READ_CACHING_HE	0x01
@@ -738,7 +770,7 @@ struct create_context {
 #define SMB2_LEASE_HANDLE_CACHING	cpu_to_le32(0x02)
 #define SMB2_LEASE_WRITE_CACHING	cpu_to_le32(0x04)
 
-#define SMB2_LEASE_FLAG_BREAK_IN_PROGRESS cpu_to_le32(0x02)
+#define SMB2_LEASE_FLAG_BREAK_IN_PROGRESS cpu_to_le32(0x00000002)
 #define SMB2_LEASE_FLAG_PARENT_LEASE_KEY_SET cpu_to_le32(0x00000004)
 
 #define SMB2_LEASE_KEY_SIZE 16
@@ -855,11 +887,20 @@ struct crt_query_id_ctxt {
 	__u8	Name[8];
 } __packed;
 
+struct crt_sd_ctxt {
+	struct create_context ccontext;
+	__u8	Name[8];
+	struct smb3_sd sd;
+	struct smb3_acl acl;
+	/* Followed by at least 4 ACEs */
+} __packed;
+
+
 #define COPY_CHUNK_RES_KEY_SIZE	24
 struct resume_key_req {
 	char ResumeKey[COPY_CHUNK_RES_KEY_SIZE];
 	__le32	ContextLength;	/* MBZ */
-	char	Context[0];	/* ignored, Windows sets to 4 bytes of zero */
+	char	Context[];	/* ignored, Windows sets to 4 bytes of zero */
 } __packed;
 
 /* this goes in the ioctl buffer when doing a copychunk request */
@@ -921,7 +962,7 @@ struct reparse_data_buffer {
 	__le32	ReparseTag;
 	__le16	ReparseDataLength;
 	__u16	Reserved;
-	__u8	DataBuffer[0]; /* Variable Length */
+	__u8	DataBuffer[]; /* Variable Length */
 } __packed;
 
 struct reparse_guid_data_buffer {
@@ -929,7 +970,7 @@ struct reparse_guid_data_buffer {
 	__le16	ReparseDataLength;
 	__u16	Reserved;
 	__u8	ReparseGuid[16];
-	__u8	DataBuffer[0]; /* Variable Length */
+	__u8	DataBuffer[]; /* Variable Length */
 } __packed;
 
 struct reparse_mount_point_data_buffer {
@@ -940,7 +981,7 @@ struct reparse_mount_point_data_buffer {
 	__le16	SubstituteNameLength;
 	__le16	PrintNameOffset;
 	__le16	PrintNameLength;
-	__u8	PathBuffer[0]; /* Variable Length */
+	__u8	PathBuffer[]; /* Variable Length */
 } __packed;
 
 #define SYMLINK_FLAG_RELATIVE 0x00000001
@@ -954,7 +995,7 @@ struct reparse_symlink_data_buffer {
 	__le16	PrintNameOffset;
 	__le16	PrintNameLength;
 	__le32	Flags;
-	__u8	PathBuffer[0]; /* Variable Length */
+	__u8	PathBuffer[]; /* Variable Length */
 } __packed;
 
 /* See MS-FSCC 2.1.2.6 and cifspdu.h for struct reparse_posix_data */
@@ -1056,7 +1097,7 @@ struct smb2_ioctl_req {
 	__le32 MaxOutputResponse;
 	__le32 Flags;
 	__u32  Reserved2;
-	__u8   Buffer[0];
+	__u8   Buffer[];
 } __packed;
 
 struct smb2_ioctl_rsp {
@@ -1170,7 +1211,7 @@ struct smb2_write_req {
 	__le64 Offset;
 	__u64  PersistentFileId; /* opaque endianness */
 	__u64  VolatileFileId; /* opaque endianness */
-	__le32 Channel; /* Reserved MBZ */
+	__le32 Channel; /* MBZ unless SMB3.02 or later */
 	__le32 RemainingBytes;
 	__le16 WriteChannelInfoOffset;
 	__le16 WriteChannelInfoLength;
@@ -1272,6 +1313,8 @@ struct smb2_echo_rsp {
 #define SMB2_INDEX_SPECIFIED		0x04
 #define SMB2_REOPEN			0x10
 
+#define SMB2_QUERY_DIRECTORY_IOV_SIZE 2
+
 struct smb2_query_directory_req {
 	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 33 */
@@ -1386,7 +1429,7 @@ struct smb2_oplock_break {
 struct smb2_lease_break {
 	struct smb2_sync_hdr sync_hdr;
 	__le16 StructureSize; /* Must be 44 */
-	__le16 Reserved;
+	__le16 Epoch;
 	__le32 Flags;
 	__u8   LeaseKey[16];
 	__le32 CurrentLeaseState;
@@ -1457,7 +1500,7 @@ struct smb3_fs_vol_info {
 	__le32	VolumeLabelLength; /* includes trailing null */
 	__u8	SupportsObjects; /* True if eg like NTFS, supports objects */
 	__u8	Reserved;
-	__u8	VolumeLabel[0]; /* variable len */
+	__u8	VolumeLabel[]; /* variable len */
 } __packed;
 
 /* partial list of QUERY INFO levels */
@@ -1507,6 +1550,7 @@ struct smb3_fs_vol_info {
 #define FILE_NORMALIZED_NAME_INFORMATION 48
 #define FILEID_GLOBAL_TX_DIRECTORY_INFORMATION 50
 #define FILE_STANDARD_LINK_INFORMATION	54
+#define FILE_ID_INFORMATION		59
 
 struct smb2_file_internal_info {
 	__le64 IndexNumber;
@@ -1518,7 +1562,7 @@ struct smb2_file_rename_info { /* encoding of request for level 10 */
 	__u8   Reserved[7];
 	__u64  RootDirectory;  /* MBZ for network operations (why says spec?) */
 	__le32 FileNameLength;
-	char   FileName[0];     /* New name to be assigned */
+	char   FileName[];     /* New name to be assigned */
 } __packed; /* level 10 Set */
 
 struct smb2_file_link_info { /* encoding of request for level 11 */
@@ -1527,7 +1571,7 @@ struct smb2_file_link_info { /* encoding of request for level 11 */
 	__u8   Reserved[7];
 	__u64  RootDirectory;  /* MBZ for network operations (why says spec?) */
 	__le32 FileNameLength;
-	char   FileName[0];     /* Name to be assigned to new link */
+	char   FileName[];     /* Name to be assigned to new link */
 } __packed; /* level 11 Set */
 
 struct smb2_file_full_ea_info { /* encoding of response for level 15 */
@@ -1535,7 +1579,7 @@ struct smb2_file_full_ea_info { /* encoding of response for level 15 */
 	__u8   flags;
 	__u8   ea_name_length;
 	__le16 ea_value_length;
-	char   ea_data[0]; /* \0 terminated name plus value */
+	char   ea_data[]; /* \0 terminated name plus value */
 } __packed; /* level 15 Set */
 
 /*
@@ -1570,6 +1614,77 @@ struct smb2_file_eof_info { /* encoding of request for level 10 */
 	__le64 EndOfFile; /* new end of file value */
 } __packed; /* level 20 Set */
 
+struct smb2_file_network_open_info {
+	__le64 CreationTime;
+	__le64 LastAccessTime;
+	__le64 LastWriteTime;
+	__le64 ChangeTime;
+	__le64 AllocationSize;
+	__le64 EndOfFile;
+	__le32 Attributes;
+	__le32 Reserved;
+} __packed; /* level 34 Query also similar returned in close rsp and open rsp */
+
+/* See MS-FSCC 2.4.43 */
+struct smb2_file_id_information {
+	__le64	VolumeSerialNumber;
+	__u64  PersistentFileId; /* opaque endianness */
+	__u64  VolatileFileId; /* opaque endianness */
+} __packed; /* level 59 */
+
 extern char smb2_padding[7];
 
+/* equivalent of the contents of SMB3.1.1 POSIX open context response */
+struct create_posix_rsp {
+	u32 nlink;
+	u32 reparse_tag;
+	u32 mode;
+	struct cifs_sid owner; /* var-sized on the wire */
+	struct cifs_sid group; /* var-sized on the wire */
+} __packed;
+
+/*
+ * SMB2-only POSIX info level
+ *
+ * See posix_info_sid_size(), posix_info_extra_size() and
+ * posix_info_parse() to help with the handling of this struct.
+ */
+struct smb2_posix_info {
+	__le32 NextEntryOffset;
+	__u32 Ignored;
+	__le64 CreationTime;
+	__le64 LastAccessTime;
+	__le64 LastWriteTime;
+	__le64 ChangeTime;
+	__le64 EndOfFile;
+	__le64 AllocationSize;
+	__le32 DosAttributes;
+	__le64 Inode;
+	__le32 DeviceId;
+	__le32 Zero;
+	/* beginning of POSIX Create Context Response */
+	__le32 HardLinks;
+	__le32 ReparseTag;
+	__le32 Mode;
+	/*
+	 * var sized owner SID
+	 * var sized group SID
+	 * le32 filenamelength
+	 * u8  filename[]
+	 */
+} __packed;
+
+/*
+ * Parsed version of the above struct. Allows direct access to the
+ * variable length fields
+ */
+struct smb2_posix_info_parsed {
+	const struct smb2_posix_info *base;
+	size_t size;
+	struct cifs_sid owner;
+	struct cifs_sid group;
+	int name_len;
+	const u8 *name;
+};
+
 #endif				/* _SMB2PDU_H */
diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h
index 71b2930b8e0b..087d5f14320b 100644
--- a/fs/cifs/smb2proto.h
+++ b/fs/cifs/smb2proto.h
@@ -46,7 +46,8 @@ extern int smb2_verify_signature(struct smb_rqst *, struct TCP_Server_Info *);
 extern int smb2_check_receive(struct mid_q_entry *mid,
 			      struct TCP_Server_Info *server, bool log_error);
 extern struct mid_q_entry *smb2_setup_request(struct cifs_ses *ses,
-			      struct smb_rqst *rqst);
+					      struct TCP_Server_Info *,
+					      struct smb_rqst *rqst);
 extern struct mid_q_entry *smb2_setup_async_request(
 			struct TCP_Server_Info *server, struct smb_rqst *rqst);
 extern struct cifs_ses *smb2_find_smb_ses(struct TCP_Server_Info *server,
@@ -54,9 +55,11 @@ extern struct cifs_ses *smb2_find_smb_ses(struct TCP_Server_Info *server,
 extern struct cifs_tcon *smb2_find_smb_tcon(struct TCP_Server_Info *server,
 						__u64 ses_id, __u32  tid);
 extern int smb2_calc_signature(struct smb_rqst *rqst,
-				struct TCP_Server_Info *server);
+				struct TCP_Server_Info *server,
+				bool allocate_crypto);
 extern int smb3_calc_signature(struct smb_rqst *rqst,
-				struct TCP_Server_Info *server);
+				struct TCP_Server_Info *server,
+				bool allocate_crypto);
 extern void smb2_echo_request(struct work_struct *work);
 extern __le32 smb2_get_lease_state(struct cifsInodeInfo *cinode);
 extern bool smb2_is_valid_oplock_break(char *buffer,
@@ -67,8 +70,10 @@ extern int smb3_handle_read_data(struct TCP_Server_Info *server,
 				 struct mid_q_entry *mid);
 
 extern int open_shroot(unsigned int xid, struct cifs_tcon *tcon,
-			struct cifs_fid *pfid);
+		       struct cifs_sb_info *cifs_sb, struct cifs_fid *pfid);
 extern void close_shroot(struct cached_fid *cfid);
+extern void close_shroot_lease(struct cached_fid *cfid);
+extern void close_shroot_lease_locked(struct cached_fid *cfid);
 extern void move_smb2_info_to_cifs(FILE_ALL_INFO *dst,
 				   struct smb2_file_all_info *src);
 extern int smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
@@ -136,6 +141,7 @@ extern int SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon);
 extern int SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms,
 		     __le16 *path, __u8 *oplock,
 		     struct smb2_file_all_info *buf,
+		     struct create_posix_rsp *posix,
 		     struct kvec *err_iov, int *resp_buftype);
 extern int SMB2_open_init(struct cifs_tcon *tcon, struct smb_rqst *rqst,
 			  __u8 *oplock, struct cifs_open_parms *oparms,
@@ -154,12 +160,13 @@ extern int SMB2_change_notify(const unsigned int xid, struct cifs_tcon *tcon,
 			u64 persistent_fid, u64 volatile_fid, bool watch_tree,
 			u32 completion_filter);
 
+extern int __SMB2_close(const unsigned int xid, struct cifs_tcon *tcon,
+			u64 persistent_fid, u64 volatile_fid,
+			struct smb2_file_network_open_info *pbuf);
 extern int SMB2_close(const unsigned int xid, struct cifs_tcon *tcon,
 		      u64 persistent_file_id, u64 volatile_file_id);
-extern int SMB2_close_flags(const unsigned int xid, struct cifs_tcon *tcon,
-			    u64 persistent_fid, u64 volatile_fid, int flags);
 extern int SMB2_close_init(struct cifs_tcon *tcon, struct smb_rqst *rqst,
-		      u64 persistent_file_id, u64 volatile_file_id);
+		      u64 persistent_fid, u64 volatile_fid, bool query_attrs);
 extern void SMB2_close_free(struct smb_rqst *rqst);
 extern int SMB2_flush(const unsigned int xid, struct cifs_tcon *tcon,
 		      u64 persistent_file_id, u64 volatile_file_id);
@@ -193,6 +200,11 @@ extern int SMB2_echo(struct TCP_Server_Info *server);
 extern int SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon,
 				u64 persistent_fid, u64 volatile_fid, int index,
 				struct cifs_search_info *srch_inf);
+extern int SMB2_query_directory_init(unsigned int xid, struct cifs_tcon *tcon,
+				     struct smb_rqst *rqst,
+				     u64 persistent_fid, u64 volatile_fid,
+				     int index, int info_level);
+extern void SMB2_query_directory_free(struct smb_rqst *rqst);
 extern int SMB2_set_eof(const unsigned int xid, struct cifs_tcon *tcon,
 			u64 persistent_fid, u64 volatile_fid, u32 pid,
 			__le64 *eof);
@@ -212,6 +224,9 @@ extern int SMB2_set_compression(const unsigned int xid, struct cifs_tcon *tcon,
 extern int SMB2_oplock_break(const unsigned int xid, struct cifs_tcon *tcon,
 			     const u64 persistent_fid, const u64 volatile_fid,
 			     const __u8 oplock_level);
+extern int smb2_handle_cancelled_close(struct cifs_tcon *tcon,
+				       __u64 persistent_fid,
+				       __u64 volatile_fid);
 extern int smb2_handle_cancelled_mid(char *buffer,
 					struct TCP_Server_Info *server);
 void smb2_cancelled_close_fid(struct work_struct *work);
@@ -240,7 +255,8 @@ extern enum securityEnum smb2_select_sectype(struct TCP_Server_Info *,
 extern void smb2_parse_contexts(struct TCP_Server_Info *server,
 				struct smb2_create_rsp *rsp,
 				unsigned int *epoch, char *lease_key,
-				__u8 *oplock, struct smb2_file_all_info *buf);
+				__u8 *oplock, struct smb2_file_all_info *buf,
+				struct create_posix_rsp *posix);
 extern int smb3_encryption_required(const struct cifs_tcon *tcon);
 extern int smb2_validate_iov(unsigned int offset, unsigned int buffer_length,
 			     struct kvec *iov, unsigned int min_buf_size);
@@ -260,4 +276,7 @@ extern int smb2_query_info_compound(const unsigned int xid,
 				    u32 class, u32 type, u32 output_len,
 				    struct kvec *rsp, int *buftype,
 				    struct cifs_sb_info *cifs_sb);
+int posix_info_parse(const void *beg, const void *end,
+		     struct smb2_posix_info_parsed *out);
+int posix_info_sid_size(const void *beg, const void *end);
 #endif			/* _SMB2PROTO_H */
diff --git a/fs/cifs/smb2transport.c b/fs/cifs/smb2transport.c
index 148d7942c796..c0348e3b1695 100644
--- a/fs/cifs/smb2transport.c
+++ b/fs/cifs/smb2transport.c
@@ -41,14 +41,6 @@
 #include "smb2glob.h"
 
 static int
-smb2_crypto_shash_allocate(struct TCP_Server_Info *server)
-{
-	return cifs_alloc_hash("hmac(sha256)",
-			       &server->secmech.hmacsha256,
-			       &server->secmech.sdeschmacsha256);
-}
-
-static int
 smb3_crypto_shash_allocate(struct TCP_Server_Info *server)
 {
 	struct cifs_secmech *p = &server->secmech;
@@ -98,6 +90,62 @@ err:
 	return rc;
 }
 
+
+static
+int smb2_get_sign_key(__u64 ses_id, struct TCP_Server_Info *server, u8 *key)
+{
+	struct cifs_chan *chan;
+	struct cifs_ses *ses = NULL;
+	struct TCP_Server_Info *it = NULL;
+	int i;
+	int rc = 0;
+
+	spin_lock(&cifs_tcp_ses_lock);
+
+	list_for_each_entry(it, &cifs_tcp_ses_list, tcp_ses_list) {
+		list_for_each_entry(ses, &it->smb_ses_list, smb_ses_list) {
+			if (ses->Suid == ses_id)
+				goto found;
+		}
+	}
+	cifs_server_dbg(VFS, "%s: Could not find session 0x%llx\n",
+			__func__, ses_id);
+	rc = -ENOENT;
+	goto out;
+
+found:
+	if (ses->binding) {
+		/*
+		 * If we are in the process of binding a new channel
+		 * to an existing session, use the master connection
+		 * session key
+		 */
+		memcpy(key, ses->smb3signingkey, SMB3_SIGN_KEY_SIZE);
+		goto out;
+	}
+
+	/*
+	 * Otherwise, use the channel key.
+	 */
+
+	for (i = 0; i < ses->chan_count; i++) {
+		chan = ses->chans + i;
+		if (chan->server == server) {
+			memcpy(key, chan->signkey, SMB3_SIGN_KEY_SIZE);
+			goto out;
+		}
+	}
+
+	cifs_dbg(VFS,
+		 "%s: Could not find channel signing key for session 0x%llx\n",
+		 __func__, ses_id);
+	rc = -ENOENT;
+
+out:
+	spin_unlock(&cifs_tcp_ses_lock);
+	return rc;
+}
+
 static struct cifs_ses *
 smb2_find_smb_ses_unlocked(struct TCP_Server_Info *server, __u64 ses_id)
 {
@@ -163,7 +211,8 @@ smb2_find_smb_tcon(struct TCP_Server_Info *server, __u64 ses_id, __u32  tid)
 }
 
 int
-smb2_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
+smb2_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server,
+			bool allocate_crypto)
 {
 	int rc;
 	unsigned char smb2_signature[SMB2_HMACSHA256_SIZE];
@@ -172,6 +221,8 @@ smb2_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 	struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)iov[0].iov_base;
 	struct cifs_ses *ses;
 	struct shash_desc *shash;
+	struct crypto_shash *hash;
+	struct sdesc *sdesc = NULL;
 	struct smb_rqst drqst;
 
 	ses = smb2_find_smb_ses(server, shdr->SessionId);
@@ -183,24 +234,32 @@ smb2_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 	memset(smb2_signature, 0x0, SMB2_HMACSHA256_SIZE);
 	memset(shdr->Signature, 0x0, SMB2_SIGNATURE_SIZE);
 
-	rc = smb2_crypto_shash_allocate(server);
-	if (rc) {
-		cifs_server_dbg(VFS, "%s: sha256 alloc failed\n", __func__);
-		return rc;
+	if (allocate_crypto) {
+		rc = cifs_alloc_hash("hmac(sha256)", &hash, &sdesc);
+		if (rc) {
+			cifs_server_dbg(VFS,
+					"%s: sha256 alloc failed\n", __func__);
+			return rc;
+		}
+		shash = &sdesc->shash;
+	} else {
+		hash = server->secmech.hmacsha256;
+		shash = &server->secmech.sdeschmacsha256->shash;
 	}
 
-	rc = crypto_shash_setkey(server->secmech.hmacsha256,
-				 ses->auth_key.response, SMB2_NTLMV2_SESSKEY_SIZE);
+	rc = crypto_shash_setkey(hash, ses->auth_key.response,
+			SMB2_NTLMV2_SESSKEY_SIZE);
 	if (rc) {
-		cifs_server_dbg(VFS, "%s: Could not update with response\n", __func__);
-		return rc;
+		cifs_server_dbg(VFS,
+				"%s: Could not update with response\n",
+				__func__);
+		goto out;
 	}
 
-	shash = &server->secmech.sdeschmacsha256->shash;
 	rc = crypto_shash_init(shash);
 	if (rc) {
 		cifs_server_dbg(VFS, "%s: Could not init sha256", __func__);
-		return rc;
+		goto out;
 	}
 
 	/*
@@ -215,9 +274,10 @@ smb2_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 		rc = crypto_shash_update(shash, iov[0].iov_base,
 					 iov[0].iov_len);
 		if (rc) {
-			cifs_server_dbg(VFS, "%s: Could not update with payload\n",
-				 __func__);
-			return rc;
+			cifs_server_dbg(VFS,
+					"%s: Could not update with payload\n",
+					__func__);
+			goto out;
 		}
 		drqst.rq_iov++;
 		drqst.rq_nvec--;
@@ -227,6 +287,9 @@ smb2_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 	if (!rc)
 		memcpy(shdr->Signature, sigptr, SMB2_SIGNATURE_SIZE);
 
+out:
+	if (allocate_crypto)
+		cifs_free_hash(&hash, &sdesc);
 	return rc;
 }
 
@@ -328,21 +391,45 @@ generate_smb3signingkey(struct cifs_ses *ses,
 {
 	int rc;
 
-	rc = generate_key(ses, ptriplet->signing.label,
-			  ptriplet->signing.context, ses->smb3signingkey,
-			  SMB3_SIGN_KEY_SIZE);
-	if (rc)
-		return rc;
+	/*
+	 * All channels use the same encryption/decryption keys but
+	 * they have their own signing key.
+	 *
+	 * When we generate the keys, check if it is for a new channel
+	 * (binding) in which case we only need to generate a signing
+	 * key and store it in the channel as to not overwrite the
+	 * master connection signing key stored in the session
+	 */
 
-	rc = generate_key(ses, ptriplet->encryption.label,
-			  ptriplet->encryption.context, ses->smb3encryptionkey,
-			  SMB3_SIGN_KEY_SIZE);
-	if (rc)
-		return rc;
+	if (ses->binding) {
+		rc = generate_key(ses, ptriplet->signing.label,
+				  ptriplet->signing.context,
+				  cifs_ses_binding_channel(ses)->signkey,
+				  SMB3_SIGN_KEY_SIZE);
+		if (rc)
+			return rc;
+	} else {
+		rc = generate_key(ses, ptriplet->signing.label,
+				  ptriplet->signing.context,
+				  ses->smb3signingkey,
+				  SMB3_SIGN_KEY_SIZE);
+		if (rc)
+			return rc;
 
-	rc = generate_key(ses, ptriplet->decryption.label,
-			  ptriplet->decryption.context,
-			  ses->smb3decryptionkey, SMB3_SIGN_KEY_SIZE);
+		memcpy(ses->chans[0].signkey, ses->smb3signingkey,
+		       SMB3_SIGN_KEY_SIZE);
+
+		rc = generate_key(ses, ptriplet->encryption.label,
+				  ptriplet->encryption.context,
+				  ses->smb3encryptionkey,
+				  SMB3_SIGN_KEY_SIZE);
+		rc = generate_key(ses, ptriplet->decryption.label,
+				  ptriplet->decryption.context,
+				  ses->smb3decryptionkey,
+				  SMB3_SIGN_KEY_SIZE);
+		if (rc)
+			return rc;
+	}
 
 	if (rc)
 		return rc;
@@ -424,31 +511,42 @@ generate_smb311signingkey(struct cifs_ses *ses)
 }
 
 int
-smb3_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
+smb3_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server,
+			bool allocate_crypto)
 {
 	int rc;
 	unsigned char smb3_signature[SMB2_CMACAES_SIZE];
 	unsigned char *sigptr = smb3_signature;
 	struct kvec *iov = rqst->rq_iov;
 	struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)iov[0].iov_base;
-	struct cifs_ses *ses;
-	struct shash_desc *shash = &server->secmech.sdesccmacaes->shash;
+	struct shash_desc *shash;
+	struct crypto_shash *hash;
+	struct sdesc *sdesc = NULL;
 	struct smb_rqst drqst;
+	u8 key[SMB3_SIGN_KEY_SIZE];
 
-	ses = smb2_find_smb_ses(server, shdr->SessionId);
-	if (!ses) {
-		cifs_server_dbg(VFS, "%s: Could not find session\n", __func__);
+	rc = smb2_get_sign_key(shdr->SessionId, server, key);
+	if (rc)
 		return 0;
+
+	if (allocate_crypto) {
+		rc = cifs_alloc_hash("cmac(aes)", &hash, &sdesc);
+		if (rc)
+			return rc;
+
+		shash = &sdesc->shash;
+	} else {
+		hash = server->secmech.cmacaes;
+		shash = &server->secmech.sdesccmacaes->shash;
 	}
 
 	memset(smb3_signature, 0x0, SMB2_CMACAES_SIZE);
 	memset(shdr->Signature, 0x0, SMB2_SIGNATURE_SIZE);
 
-	rc = crypto_shash_setkey(server->secmech.cmacaes,
-				 ses->smb3signingkey, SMB2_CMACAES_SIZE);
+	rc = crypto_shash_setkey(hash, key, SMB2_CMACAES_SIZE);
 	if (rc) {
 		cifs_server_dbg(VFS, "%s: Could not set key for cmac aes\n", __func__);
-		return rc;
+		goto out;
 	}
 
 	/*
@@ -459,7 +557,7 @@ smb3_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 	rc = crypto_shash_init(shash);
 	if (rc) {
 		cifs_server_dbg(VFS, "%s: Could not init cmac aes\n", __func__);
-		return rc;
+		goto out;
 	}
 
 	/*
@@ -476,7 +574,7 @@ smb3_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 		if (rc) {
 			cifs_server_dbg(VFS, "%s: Could not update with payload\n",
 				 __func__);
-			return rc;
+			goto out;
 		}
 		drqst.rq_iov++;
 		drqst.rq_nvec--;
@@ -486,6 +584,9 @@ smb3_calc_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 	if (!rc)
 		memcpy(shdr->Signature, sigptr, SMB2_SIGNATURE_SIZE);
 
+out:
+	if (allocate_crypto)
+		cifs_free_hash(&hash, &sdesc);
 	return rc;
 }
 
@@ -494,19 +595,28 @@ static int
 smb2_sign_rqst(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 {
 	int rc = 0;
-	struct smb2_sync_hdr *shdr =
-			(struct smb2_sync_hdr *)rqst->rq_iov[0].iov_base;
+	struct smb2_sync_hdr *shdr;
+	struct smb2_sess_setup_req *ssr;
+	bool is_binding;
+	bool is_signed;
 
-	if (!(shdr->Flags & SMB2_FLAGS_SIGNED) ||
-	    server->tcpStatus == CifsNeedNegotiate)
-		return rc;
+	shdr = (struct smb2_sync_hdr *)rqst->rq_iov[0].iov_base;
+	ssr = (struct smb2_sess_setup_req *)shdr;
+
+	is_binding = shdr->Command == SMB2_SESSION_SETUP &&
+		(ssr->Flags & SMB2_SESSION_REQ_FLAG_BINDING);
+	is_signed = shdr->Flags & SMB2_FLAGS_SIGNED;
 
-	if (!server->session_estab) {
+	if (!is_signed)
+		return 0;
+	if (server->tcpStatus == CifsNeedNegotiate)
+		return 0;
+	if (!is_binding && !server->session_estab) {
 		strncpy(shdr->Signature, "BSRSPYL", 8);
-		return rc;
+		return 0;
 	}
 
-	rc = server->ops->calc_signature(rqst, server);
+	rc = server->ops->calc_signature(rqst, server, false);
 
 	return rc;
 }
@@ -515,7 +625,7 @@ int
 smb2_verify_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 {
 	unsigned int rc;
-	char server_response_sig[16];
+	char server_response_sig[SMB2_SIGNATURE_SIZE];
 	struct smb2_sync_hdr *shdr =
 			(struct smb2_sync_hdr *)rqst->rq_iov[0].iov_base;
 
@@ -544,16 +654,16 @@ smb2_verify_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server)
 
 	memset(shdr->Signature, 0, SMB2_SIGNATURE_SIZE);
 
-	mutex_lock(&server->srv_mutex);
-	rc = server->ops->calc_signature(rqst, server);
-	mutex_unlock(&server->srv_mutex);
+	rc = server->ops->calc_signature(rqst, server, true);
 
 	if (rc)
 		return rc;
 
-	if (memcmp(server_response_sig, shdr->Signature, SMB2_SIGNATURE_SIZE))
+	if (memcmp(server_response_sig, shdr->Signature, SMB2_SIGNATURE_SIZE)) {
+		cifs_dbg(VFS, "sign fail cmd 0x%x message id 0x%llx\n",
+			shdr->Command, shdr->MessageId);
 		return -EACCES;
-	else
+	} else
 		return 0;
 }
 
@@ -599,6 +709,8 @@ smb2_mid_entry_alloc(const struct smb2_sync_hdr *shdr,
 	 * The default is for the mid to be synchronous, so the
 	 * default callback just wakes up the current task.
 	 */
+	get_task_struct(current);
+	temp->creator = current;
 	temp->callback = cifs_wake_up_task;
 	temp->callback_data = current;
 
@@ -610,18 +722,18 @@ smb2_mid_entry_alloc(const struct smb2_sync_hdr *shdr,
 }
 
 static int
-smb2_get_mid_entry(struct cifs_ses *ses, struct smb2_sync_hdr *shdr,
-		   struct mid_q_entry **mid)
+smb2_get_mid_entry(struct cifs_ses *ses, struct TCP_Server_Info *server,
+		   struct smb2_sync_hdr *shdr, struct mid_q_entry **mid)
 {
-	if (ses->server->tcpStatus == CifsExiting)
+	if (server->tcpStatus == CifsExiting)
 		return -ENOENT;
 
-	if (ses->server->tcpStatus == CifsNeedReconnect) {
+	if (server->tcpStatus == CifsNeedReconnect) {
 		cifs_dbg(FYI, "tcp session dead - return to caller to retry\n");
 		return -EAGAIN;
 	}
 
-	if (ses->server->tcpStatus == CifsNeedNegotiate &&
+	if (server->tcpStatus == CifsNeedNegotiate &&
 	   shdr->Command != SMB2_NEGOTIATE)
 		return -EAGAIN;
 
@@ -638,11 +750,11 @@ smb2_get_mid_entry(struct cifs_ses *ses, struct smb2_sync_hdr *shdr,
 		/* else ok - we are shutting down the session */
 	}
 
-	*mid = smb2_mid_entry_alloc(shdr, ses->server);
+	*mid = smb2_mid_entry_alloc(shdr, server);
 	if (*mid == NULL)
 		return -ENOMEM;
 	spin_lock(&GlobalMid_Lock);
-	list_add_tail(&(*mid)->qhead, &ses->server->pending_mid_q);
+	list_add_tail(&(*mid)->qhead, &server->pending_mid_q);
 	spin_unlock(&GlobalMid_Lock);
 
 	return 0;
@@ -675,24 +787,25 @@ smb2_check_receive(struct mid_q_entry *mid, struct TCP_Server_Info *server,
 }
 
 struct mid_q_entry *
-smb2_setup_request(struct cifs_ses *ses, struct smb_rqst *rqst)
+smb2_setup_request(struct cifs_ses *ses, struct TCP_Server_Info *server,
+		   struct smb_rqst *rqst)
 {
 	int rc;
 	struct smb2_sync_hdr *shdr =
 			(struct smb2_sync_hdr *)rqst->rq_iov[0].iov_base;
 	struct mid_q_entry *mid;
 
-	smb2_seq_num_into_buf(ses->server, shdr);
+	smb2_seq_num_into_buf(server, shdr);
 
-	rc = smb2_get_mid_entry(ses, shdr, &mid);
+	rc = smb2_get_mid_entry(ses, server, shdr, &mid);
 	if (rc) {
-		revert_current_mid_from_hdr(ses->server, shdr);
+		revert_current_mid_from_hdr(server, shdr);
 		return ERR_PTR(rc);
 	}
 
-	rc = smb2_sign_rqst(rqst, ses->server);
+	rc = smb2_sign_rqst(rqst, server);
 	if (rc) {
-		revert_current_mid_from_hdr(ses->server, shdr);
+		revert_current_mid_from_hdr(server, shdr);
 		cifs_delete_mid(mid);
 		return ERR_PTR(rc);
 	}
diff --git a/fs/cifs/smbdirect.c b/fs/cifs/smbdirect.c
index 3c91fa97c9a8..1a5834a5d597 100644
--- a/fs/cifs/smbdirect.c
+++ b/fs/cifs/smbdirect.c
@@ -284,13 +284,10 @@ static void send_done(struct ib_cq *cq, struct ib_wc *wc)
 			request->sge[i].length,
 			DMA_TO_DEVICE);
 
-	if (request->has_payload) {
-		if (atomic_dec_and_test(&request->info->send_payload_pending))
-			wake_up(&request->info->wait_send_payload_pending);
-	} else {
-		if (atomic_dec_and_test(&request->info->send_pending))
-			wake_up(&request->info->wait_send_pending);
-	}
+	if (atomic_dec_and_test(&request->info->send_pending))
+		wake_up(&request->info->wait_send_pending);
+
+	wake_up(&request->info->wait_post_send);
 
 	mempool_free(request, request->info->request_mempool);
 }
@@ -383,27 +380,6 @@ static bool process_negotiation_response(
 	return true;
 }
 
-/*
- * Check and schedule to send an immediate packet
- * This is used to extend credtis to remote peer to keep the transport busy
- */
-static void check_and_send_immediate(struct smbd_connection *info)
-{
-	if (info->transport_status != SMBD_CONNECTED)
-		return;
-
-	info->send_immediate = true;
-
-	/*
-	 * Promptly send a packet if our peer is running low on receive
-	 * credits
-	 */
-	if (atomic_read(&info->receive_credits) <
-		info->receive_credit_target - 1)
-		queue_delayed_work(
-			info->workqueue, &info->send_immediate_work, 0);
-}
-
 static void smbd_post_send_credits(struct work_struct *work)
 {
 	int ret = 0;
@@ -453,29 +429,16 @@ static void smbd_post_send_credits(struct work_struct *work)
 	info->new_credits_offered += ret;
 	spin_unlock(&info->lock_new_credits_offered);
 
-	atomic_add(ret, &info->receive_credits);
-
-	/* Check if we can post new receive and grant credits to peer */
-	check_and_send_immediate(info);
-}
-
-static void smbd_recv_done_work(struct work_struct *work)
-{
-	struct smbd_connection *info =
-		container_of(work, struct smbd_connection, recv_done_work);
-
-	/*
-	 * We may have new send credits granted from remote peer
-	 * If any sender is blcoked on lack of credets, unblock it
-	 */
-	if (atomic_read(&info->send_credits))
-		wake_up_interruptible(&info->wait_send_queue);
-
-	/*
-	 * Check if we need to send something to remote peer to
-	 * grant more credits or respond to KEEP_ALIVE packet
-	 */
-	check_and_send_immediate(info);
+	/* Promptly send an immediate packet as defined in [MS-SMBD] 3.1.1.1 */
+	info->send_immediate = true;
+	if (atomic_read(&info->receive_credits) <
+		info->receive_credit_target - 1) {
+		if (info->keep_alive_requested == KEEP_ALIVE_PENDING ||
+		    info->send_immediate) {
+			log_keep_alive(INFO, "send an empty message\n");
+			smbd_post_send_empty(info);
+		}
+	}
 }
 
 /* Called from softirq, when recv is done */
@@ -546,8 +509,15 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc)
 		atomic_dec(&info->receive_credits);
 		info->receive_credit_target =
 			le16_to_cpu(data_transfer->credits_requested);
-		atomic_add(le16_to_cpu(data_transfer->credits_granted),
-			&info->send_credits);
+		if (le16_to_cpu(data_transfer->credits_granted)) {
+			atomic_add(le16_to_cpu(data_transfer->credits_granted),
+				&info->send_credits);
+			/*
+			 * We have new send credits granted from remote peer
+			 * If any sender is waiting for credits, unblock it
+			 */
+			wake_up_interruptible(&info->wait_send_queue);
+		}
 
 		log_incoming(INFO, "data flags %d data_offset %d "
 			"data_length %d remaining_data_length %d\n",
@@ -563,7 +533,6 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc)
 			info->keep_alive_requested = KEEP_ALIVE_PENDING;
 		}
 
-		queue_work(info->workqueue, &info->recv_done_work);
 		return;
 
 	default:
@@ -756,7 +725,6 @@ static int smbd_post_send_negotiate_req(struct smbd_connection *info)
 		request->sge[0].addr,
 		request->sge[0].length, request->sge[0].lkey);
 
-	request->has_payload = false;
 	atomic_inc(&info->send_pending);
 	rc = ib_post_send(info->id->qp, &send_wr, NULL);
 	if (!rc)
@@ -813,45 +781,96 @@ static int manage_keep_alive_before_sending(struct smbd_connection *info)
 	return 0;
 }
 
-/*
- * Build and prepare the SMBD packet header
- * This function waits for avaialbe send credits and build a SMBD packet
- * header. The caller then optional append payload to the packet after
- * the header
- * intput values
- * size: the size of the payload
- * remaining_data_length: remaining data to send if this is part of a
- * fragmented packet
- * output values
- * request_out: the request allocated from this function
- * return values: 0 on success, otherwise actual error code returned
- */
-static int smbd_create_header(struct smbd_connection *info,
-		int size, int remaining_data_length,
-		struct smbd_request **request_out)
+/* Post the send request */
+static int smbd_post_send(struct smbd_connection *info,
+		struct smbd_request *request)
+{
+	struct ib_send_wr send_wr;
+	int rc, i;
+
+	for (i = 0; i < request->num_sge; i++) {
+		log_rdma_send(INFO,
+			"rdma_request sge[%d] addr=%llu length=%u\n",
+			i, request->sge[i].addr, request->sge[i].length);
+		ib_dma_sync_single_for_device(
+			info->id->device,
+			request->sge[i].addr,
+			request->sge[i].length,
+			DMA_TO_DEVICE);
+	}
+
+	request->cqe.done = send_done;
+
+	send_wr.next = NULL;
+	send_wr.wr_cqe = &request->cqe;
+	send_wr.sg_list = request->sge;
+	send_wr.num_sge = request->num_sge;
+	send_wr.opcode = IB_WR_SEND;
+	send_wr.send_flags = IB_SEND_SIGNALED;
+
+	rc = ib_post_send(info->id->qp, &send_wr, NULL);
+	if (rc) {
+		log_rdma_send(ERR, "ib_post_send failed rc=%d\n", rc);
+		smbd_disconnect_rdma_connection(info);
+		rc = -EAGAIN;
+	} else
+		/* Reset timer for idle connection after packet is sent */
+		mod_delayed_work(info->workqueue, &info->idle_timer_work,
+			info->keep_alive_interval*HZ);
+
+	return rc;
+}
+
+static int smbd_post_send_sgl(struct smbd_connection *info,
+	struct scatterlist *sgl, int data_length, int remaining_data_length)
 {
+	int num_sgs;
+	int i, rc;
+	int header_length;
 	struct smbd_request *request;
 	struct smbd_data_transfer *packet;
-	int header_length;
-	int rc;
+	int new_credits;
+	struct scatterlist *sg;
 
+wait_credit:
 	/* Wait for send credits. A SMBD packet needs one credit */
 	rc = wait_event_interruptible(info->wait_send_queue,
 		atomic_read(&info->send_credits) > 0 ||
 		info->transport_status != SMBD_CONNECTED);
 	if (rc)
-		return rc;
+		goto err_wait_credit;
+
+	if (info->transport_status != SMBD_CONNECTED) {
+		log_outgoing(ERR, "disconnected not sending on wait_credit\n");
+		rc = -EAGAIN;
+		goto err_wait_credit;
+	}
+	if (unlikely(atomic_dec_return(&info->send_credits) < 0)) {
+		atomic_inc(&info->send_credits);
+		goto wait_credit;
+	}
+
+wait_send_queue:
+	wait_event(info->wait_post_send,
+		atomic_read(&info->send_pending) < info->send_credit_target ||
+		info->transport_status != SMBD_CONNECTED);
 
 	if (info->transport_status != SMBD_CONNECTED) {
-		log_outgoing(ERR, "disconnected not sending\n");
-		return -EAGAIN;
+		log_outgoing(ERR, "disconnected not sending on wait_send_queue\n");
+		rc = -EAGAIN;
+		goto err_wait_send_queue;
+	}
+
+	if (unlikely(atomic_inc_return(&info->send_pending) >
+				info->send_credit_target)) {
+		atomic_dec(&info->send_pending);
+		goto wait_send_queue;
 	}
-	atomic_dec(&info->send_credits);
 
 	request = mempool_alloc(info->request_mempool, GFP_KERNEL);
 	if (!request) {
 		rc = -ENOMEM;
-		goto err;
+		goto err_alloc;
 	}
 
 	request->info = info;
@@ -859,8 +878,11 @@ static int smbd_create_header(struct smbd_connection *info,
 	/* Fill in the packet header */
 	packet = smbd_request_payload(request);
 	packet->credits_requested = cpu_to_le16(info->send_credit_target);
-	packet->credits_granted =
-		cpu_to_le16(manage_credits_prior_sending(info));
+
+	new_credits = manage_credits_prior_sending(info);
+	atomic_add(new_credits, &info->receive_credits);
+	packet->credits_granted = cpu_to_le16(new_credits);
+
 	info->send_immediate = false;
 
 	packet->flags = 0;
@@ -868,11 +890,11 @@ static int smbd_create_header(struct smbd_connection *info,
 		packet->flags |= cpu_to_le16(SMB_DIRECT_RESPONSE_REQUESTED);
 
 	packet->reserved = 0;
-	if (!size)
+	if (!data_length)
 		packet->data_offset = 0;
 	else
 		packet->data_offset = cpu_to_le32(24);
-	packet->data_length = cpu_to_le32(size);
+	packet->data_length = cpu_to_le32(data_length);
 	packet->remaining_data_length = cpu_to_le32(remaining_data_length);
 	packet->padding = 0;
 
@@ -887,7 +909,7 @@ static int smbd_create_header(struct smbd_connection *info,
 	/* Map the packet to DMA */
 	header_length = sizeof(struct smbd_data_transfer);
 	/* If this is a packet without payload, don't send padding */
-	if (!size)
+	if (!data_length)
 		header_length = offsetof(struct smbd_data_transfer, padding);
 
 	request->num_sge = 1;
@@ -896,102 +918,15 @@ static int smbd_create_header(struct smbd_connection *info,
 						 header_length,
 						 DMA_TO_DEVICE);
 	if (ib_dma_mapping_error(info->id->device, request->sge[0].addr)) {
-		mempool_free(request, info->request_mempool);
 		rc = -EIO;
-		goto err;
+		request->sge[0].addr = 0;
+		goto err_dma;
 	}
 
 	request->sge[0].length = header_length;
 	request->sge[0].lkey = info->pd->local_dma_lkey;
 
-	*request_out = request;
-	return 0;
-
-err:
-	atomic_inc(&info->send_credits);
-	return rc;
-}
-
-static void smbd_destroy_header(struct smbd_connection *info,
-		struct smbd_request *request)
-{
-
-	ib_dma_unmap_single(info->id->device,
-			    request->sge[0].addr,
-			    request->sge[0].length,
-			    DMA_TO_DEVICE);
-	mempool_free(request, info->request_mempool);
-	atomic_inc(&info->send_credits);
-}
-
-/* Post the send request */
-static int smbd_post_send(struct smbd_connection *info,
-		struct smbd_request *request, bool has_payload)
-{
-	struct ib_send_wr send_wr;
-	int rc, i;
-
-	for (i = 0; i < request->num_sge; i++) {
-		log_rdma_send(INFO,
-			"rdma_request sge[%d] addr=%llu length=%u\n",
-			i, request->sge[i].addr, request->sge[i].length);
-		ib_dma_sync_single_for_device(
-			info->id->device,
-			request->sge[i].addr,
-			request->sge[i].length,
-			DMA_TO_DEVICE);
-	}
-
-	request->cqe.done = send_done;
-
-	send_wr.next = NULL;
-	send_wr.wr_cqe = &request->cqe;
-	send_wr.sg_list = request->sge;
-	send_wr.num_sge = request->num_sge;
-	send_wr.opcode = IB_WR_SEND;
-	send_wr.send_flags = IB_SEND_SIGNALED;
-
-	if (has_payload) {
-		request->has_payload = true;
-		atomic_inc(&info->send_payload_pending);
-	} else {
-		request->has_payload = false;
-		atomic_inc(&info->send_pending);
-	}
-
-	rc = ib_post_send(info->id->qp, &send_wr, NULL);
-	if (rc) {
-		log_rdma_send(ERR, "ib_post_send failed rc=%d\n", rc);
-		if (has_payload) {
-			if (atomic_dec_and_test(&info->send_payload_pending))
-				wake_up(&info->wait_send_payload_pending);
-		} else {
-			if (atomic_dec_and_test(&info->send_pending))
-				wake_up(&info->wait_send_pending);
-		}
-		smbd_disconnect_rdma_connection(info);
-		rc = -EAGAIN;
-	} else
-		/* Reset timer for idle connection after packet is sent */
-		mod_delayed_work(info->workqueue, &info->idle_timer_work,
-			info->keep_alive_interval*HZ);
-
-	return rc;
-}
-
-static int smbd_post_send_sgl(struct smbd_connection *info,
-	struct scatterlist *sgl, int data_length, int remaining_data_length)
-{
-	int num_sgs;
-	int i, rc;
-	struct smbd_request *request;
-	struct scatterlist *sg;
-
-	rc = smbd_create_header(
-		info, data_length, remaining_data_length, &request);
-	if (rc)
-		return rc;
-
+	/* Fill in the packet data payload */
 	num_sgs = sgl ? sg_nents(sgl) : 0;
 	for_each_sg(sgl, sg, num_sgs, i) {
 		request->sge[i+1].addr =
@@ -1001,25 +936,41 @@ static int smbd_post_send_sgl(struct smbd_connection *info,
 				info->id->device, request->sge[i+1].addr)) {
 			rc = -EIO;
 			request->sge[i+1].addr = 0;
-			goto dma_mapping_failure;
+			goto err_dma;
 		}
 		request->sge[i+1].length = sg->length;
 		request->sge[i+1].lkey = info->pd->local_dma_lkey;
 		request->num_sge++;
 	}
 
-	rc = smbd_post_send(info, request, data_length);
+	rc = smbd_post_send(info, request);
 	if (!rc)
 		return 0;
 
-dma_mapping_failure:
-	for (i = 1; i < request->num_sge; i++)
+err_dma:
+	for (i = 0; i < request->num_sge; i++)
 		if (request->sge[i].addr)
 			ib_dma_unmap_single(info->id->device,
 					    request->sge[i].addr,
 					    request->sge[i].length,
 					    DMA_TO_DEVICE);
-	smbd_destroy_header(info, request);
+	mempool_free(request, info->request_mempool);
+
+	/* roll back receive credits and credits to be offered */
+	spin_lock(&info->lock_new_credits_offered);
+	info->new_credits_offered += new_credits;
+	spin_unlock(&info->lock_new_credits_offered);
+	atomic_sub(new_credits, &info->receive_credits);
+
+err_alloc:
+	if (atomic_dec_and_test(&info->send_pending))
+		wake_up(&info->wait_send_pending);
+
+err_wait_send_queue:
+	/* roll back send credits and pending */
+	atomic_inc(&info->send_credits);
+
+err_wait_credit:
 	return rc;
 }
 
@@ -1069,7 +1020,7 @@ static int smbd_post_send_data(
 
 	if (n_vec > SMBDIRECT_MAX_SGE) {
 		cifs_dbg(VFS, "Can't fit data to SGL, n_vec=%d\n", n_vec);
-		return -ENOMEM;
+		return -EINVAL;
 	}
 
 	sg_init_table(sgl, n_vec);
@@ -1341,25 +1292,6 @@ static void destroy_receive_buffers(struct smbd_connection *info)
 		mempool_free(response, info->response_mempool);
 }
 
-/*
- * Check and send an immediate or keep alive packet
- * The condition to send those packets are defined in [MS-SMBD] 3.1.1.1
- * Connection.KeepaliveRequested and Connection.SendImmediate
- * The idea is to extend credits to server as soon as it becomes available
- */
-static void send_immediate_work(struct work_struct *work)
-{
-	struct smbd_connection *info = container_of(
-					work, struct smbd_connection,
-					send_immediate_work.work);
-
-	if (info->keep_alive_requested == KEEP_ALIVE_PENDING ||
-	    info->send_immediate) {
-		log_keep_alive(INFO, "send an empty message\n");
-		smbd_post_send_empty(info);
-	}
-}
-
 /* Implement idle connection timer [MS-SMBD] 3.1.6.2 */
 static void idle_connection_timer(struct work_struct *work)
 {
@@ -1414,14 +1346,10 @@ void smbd_destroy(struct TCP_Server_Info *server)
 
 	log_rdma_event(INFO, "cancelling idle timer\n");
 	cancel_delayed_work_sync(&info->idle_timer_work);
-	log_rdma_event(INFO, "cancelling send immediate work\n");
-	cancel_delayed_work_sync(&info->send_immediate_work);
 
 	log_rdma_event(INFO, "wait for all send posted to IB to finish\n");
 	wait_event(info->wait_send_pending,
 		atomic_read(&info->send_pending) == 0);
-	wait_event(info->wait_send_payload_pending,
-		atomic_read(&info->send_payload_pending) == 0);
 
 	/* It's not posssible for upper layer to get to reassembly */
 	log_rdma_event(INFO, "drain the reassembly queue\n");
@@ -1476,6 +1404,7 @@ void smbd_destroy(struct TCP_Server_Info *server)
 	info->transport_status = SMBD_DESTROYED;
 
 	destroy_workqueue(info->workqueue);
+	log_rdma_event(INFO,  "rdma session destroyed\n");
 	kfree(info);
 }
 
@@ -1505,8 +1434,9 @@ create_conn:
 	log_rdma_event(INFO, "creating rdma session\n");
 	server->smbd_conn = smbd_get_connection(
 		server, (struct sockaddr *) &server->dstaddr);
-	log_rdma_event(INFO, "created rdma session info=%p\n",
-		server->smbd_conn);
+
+	if (server->smbd_conn)
+		cifs_dbg(VFS, "RDMA transport re-established\n");
 
 	return server->smbd_conn ? 0 : -ENOENT;
 }
@@ -1749,18 +1679,15 @@ static struct smbd_connection *_smbd_get_connection(
 
 	init_waitqueue_head(&info->wait_send_queue);
 	INIT_DELAYED_WORK(&info->idle_timer_work, idle_connection_timer);
-	INIT_DELAYED_WORK(&info->send_immediate_work, send_immediate_work);
 	queue_delayed_work(info->workqueue, &info->idle_timer_work,
 		info->keep_alive_interval*HZ);
 
 	init_waitqueue_head(&info->wait_send_pending);
 	atomic_set(&info->send_pending, 0);
 
-	init_waitqueue_head(&info->wait_send_payload_pending);
-	atomic_set(&info->send_payload_pending, 0);
+	init_waitqueue_head(&info->wait_post_send);
 
 	INIT_WORK(&info->disconnect_work, smbd_disconnect_rdma_work);
-	INIT_WORK(&info->recv_done_work, smbd_recv_done_work);
 	INIT_WORK(&info->post_send_credits_work, smbd_post_send_credits);
 	info->new_credits_offered = 0;
 	spin_lock_init(&info->lock_new_credits_offered);
@@ -1970,7 +1897,7 @@ read_rfc1002_done:
 
 	if (info->transport_status != SMBD_CONNECTED) {
 		log_read(ERR, "disconnected\n");
-		return 0;
+		return -ECONNABORTED;
 	}
 
 	goto again;
@@ -2095,8 +2022,7 @@ int smbd_send(struct TCP_Server_Info *server,
 	for (i = 0; i < num_rqst; i++)
 		remaining_data_length += smb_rqst_len(server, &rqst_array[i]);
 
-	if (remaining_data_length + sizeof(struct smbd_data_transfer) >
-		info->max_fragmented_send_size) {
+	if (remaining_data_length > info->max_fragmented_send_size) {
 		log_write(ERR, "payload size %d > max size %d\n",
 			remaining_data_length, info->max_fragmented_send_size);
 		rc = -EINVAL;
@@ -2233,8 +2159,8 @@ done:
 	 * that means all the I/Os have been out and we are good to return
 	 */
 
-	wait_event(info->wait_send_payload_pending,
-		atomic_read(&info->send_payload_pending) == 0);
+	wait_event(info->wait_send_pending,
+		atomic_read(&info->send_pending) == 0);
 
 	return rc;
 }
@@ -2269,12 +2195,7 @@ static void smbd_mr_recovery_work(struct work_struct *work)
 	int rc;
 
 	list_for_each_entry(smbdirect_mr, &info->mr_list, list) {
-		if (smbdirect_mr->state == MR_INVALIDATED)
-			ib_dma_unmap_sg(
-				info->id->device, smbdirect_mr->sgl,
-				smbdirect_mr->sgl_count,
-				smbdirect_mr->dir);
-		else if (smbdirect_mr->state == MR_ERROR) {
+		if (smbdirect_mr->state == MR_ERROR) {
 
 			/* recover this MR entry */
 			rc = ib_dereg_mr(smbdirect_mr->mr);
@@ -2602,11 +2523,20 @@ int smbd_deregister_mr(struct smbd_mr *smbdirect_mr)
 		 */
 		smbdirect_mr->state = MR_INVALIDATED;
 
-	/*
-	 * Schedule the work to do MR recovery for future I/Os
-	 * MR recovery is slow and we don't want it to block the current I/O
-	 */
-	queue_work(info->workqueue, &info->mr_recovery_work);
+	if (smbdirect_mr->state == MR_INVALIDATED) {
+		ib_dma_unmap_sg(
+			info->id->device, smbdirect_mr->sgl,
+			smbdirect_mr->sgl_count,
+			smbdirect_mr->dir);
+		smbdirect_mr->state = MR_READY;
+		if (atomic_inc_return(&info->mr_ready_count) == 1)
+			wake_up_interruptible(&info->wait_mr);
+	} else
+		/*
+		 * Schedule the work to do MR recovery for future I/Os MR
+		 * recovery is slow and don't want it to block current I/O
+		 */
+		queue_work(info->workqueue, &info->mr_recovery_work);
 
 done:
 	if (atomic_dec_and_test(&info->mr_used_count))
diff --git a/fs/cifs/smbdirect.h b/fs/cifs/smbdirect.h
index 6ff880a1e186..a87fca82a796 100644
--- a/fs/cifs/smbdirect.h
+++ b/fs/cifs/smbdirect.h
@@ -67,7 +67,6 @@ struct smbd_connection {
 	bool negotiate_done;
 
 	struct work_struct disconnect_work;
-	struct work_struct recv_done_work;
 	struct work_struct post_send_credits_work;
 
 	spinlock_t lock_new_credits_offered;
@@ -115,8 +114,7 @@ struct smbd_connection {
 	/* Activity accoutning */
 	atomic_t send_pending;
 	wait_queue_head_t wait_send_pending;
-	atomic_t send_payload_pending;
-	wait_queue_head_t wait_send_payload_pending;
+	wait_queue_head_t wait_post_send;
 
 	/* Receive queue */
 	struct list_head receive_queue;
@@ -155,7 +153,6 @@ struct smbd_connection {
 
 	struct workqueue_struct *workqueue;
 	struct delayed_work idle_timer_work;
-	struct delayed_work send_immediate_work;
 
 	/* Memory pool for preallocating buffers */
 	/* request pool for RDMA send */
@@ -235,9 +232,6 @@ struct smbd_request {
 	struct smbd_connection *info;
 	struct ib_cqe cqe;
 
-	/* true if this request carries upper layer payload */
-	bool has_payload;
-
 	/* the SGE entries for this packet */
 	struct ib_sge sge[SMBDIRECT_MAX_SGE];
 	int num_sge;
diff --git a/fs/cifs/trace.h b/fs/cifs/trace.h
index e7e350b13d6a..4cb0d5f7ce45 100644
--- a/fs/cifs/trace.h
+++ b/fs/cifs/trace.h
@@ -547,6 +547,33 @@ DEFINE_EVENT(smb3_exit_err_class, smb3_##name,    \
 
 DEFINE_SMB3_EXIT_ERR_EVENT(exit_err);
 
+
+DECLARE_EVENT_CLASS(smb3_sync_err_class,
+	TP_PROTO(unsigned long ino,
+		int	rc),
+	TP_ARGS(ino, rc),
+	TP_STRUCT__entry(
+		__field(unsigned long, ino)
+		__field(int, rc)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->rc = rc;
+	),
+	TP_printk("\tino=%lu rc=%d",
+		__entry->ino, __entry->rc)
+)
+
+#define DEFINE_SMB3_SYNC_ERR_EVENT(name)          \
+DEFINE_EVENT(smb3_sync_err_class, cifs_##name,    \
+	TP_PROTO(unsigned long ino,		\
+		int	rc),			\
+	TP_ARGS(ino, rc))
+
+DEFINE_SMB3_SYNC_ERR_EVENT(fsync_err);
+DEFINE_SMB3_SYNC_ERR_EVENT(flush_err);
+
+
 DECLARE_EVENT_CLASS(smb3_enter_exit_class,
 	TP_PROTO(unsigned int xid,
 		const char *func_name),
diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c
index ca3de62688d6..c97570eb2c18 100644
--- a/fs/cifs/transport.c
+++ b/fs/cifs/transport.c
@@ -76,6 +76,8 @@ AllocMidQEntry(const struct smb_hdr *smb_buffer, struct TCP_Server_Info *server)
 	 * The default is for the mid to be synchronous, so the
 	 * default callback just wakes up the current task.
 	 */
+	get_task_struct(current);
+	temp->creator = current;
 	temp->callback = cifs_wake_up_task;
 	temp->callback_data = current;
 
@@ -93,8 +95,14 @@ static void _cifs_mid_q_entry_release(struct kref *refcount)
 	__u16 smb_cmd = le16_to_cpu(midEntry->command);
 	unsigned long now;
 	unsigned long roundtrip_time;
-	struct TCP_Server_Info *server = midEntry->server;
 #endif
+	struct TCP_Server_Info *server = midEntry->server;
+
+	if (midEntry->resp_buf && (midEntry->mid_flags & MID_WAIT_CANCELLED) &&
+	    midEntry->mid_state == MID_RESPONSE_RECEIVED &&
+	    server->ops->handle_cancelled_mid)
+		server->ops->handle_cancelled_mid(midEntry->resp_buf, server);
+
 	midEntry->mid_state = MID_FREE;
 	atomic_dec(&midCount);
 	if (midEntry->large_buf)
@@ -152,6 +160,7 @@ static void _cifs_mid_q_entry_release(struct kref *refcount)
 		}
 	}
 #endif
+	put_task_struct(midEntry->creator);
 
 	mempool_free(midEntry, cifs_mid_poolp);
 }
@@ -319,8 +328,11 @@ __smb_send_rqst(struct TCP_Server_Info *server, int num_rqst,
 	int val = 1;
 	__be32 rfc1002_marker;
 
-	if (cifs_rdma_enabled(server) && server->smbd_conn) {
-		rc = smbd_send(server, num_rqst, rqst);
+	if (cifs_rdma_enabled(server)) {
+		/* return -EAGAIN when connecting or reconnecting */
+		rc = -EAGAIN;
+		if (server->smbd_conn)
+			rc = smbd_send(server, num_rqst, rqst);
 		goto smbd_done;
 	}
 
@@ -454,7 +466,7 @@ smb_send_rqst(struct TCP_Server_Info *server, int num_rqst,
 	      struct smb_rqst *rqst, int flags)
 {
 	struct kvec iov;
-	struct smb2_transform_hdr tr_hdr;
+	struct smb2_transform_hdr *tr_hdr;
 	struct smb_rqst cur_rqst[MAX_COMPOUND];
 	int rc;
 
@@ -464,28 +476,34 @@ smb_send_rqst(struct TCP_Server_Info *server, int num_rqst,
 	if (num_rqst > MAX_COMPOUND - 1)
 		return -ENOMEM;
 
-	memset(&cur_rqst[0], 0, sizeof(cur_rqst));
-	memset(&iov, 0, sizeof(iov));
-	memset(&tr_hdr, 0, sizeof(tr_hdr));
-
-	iov.iov_base = &tr_hdr;
-	iov.iov_len = sizeof(tr_hdr);
-	cur_rqst[0].rq_iov = &iov;
-	cur_rqst[0].rq_nvec = 1;
-
 	if (!server->ops->init_transform_rq) {
 		cifs_server_dbg(VFS, "Encryption requested but transform "
 				"callback is missing\n");
 		return -EIO;
 	}
 
+	tr_hdr = kmalloc(sizeof(*tr_hdr), GFP_NOFS);
+	if (!tr_hdr)
+		return -ENOMEM;
+
+	memset(&cur_rqst[0], 0, sizeof(cur_rqst));
+	memset(&iov, 0, sizeof(iov));
+	memset(tr_hdr, 0, sizeof(*tr_hdr));
+
+	iov.iov_base = tr_hdr;
+	iov.iov_len = sizeof(*tr_hdr);
+	cur_rqst[0].rq_iov = &iov;
+	cur_rqst[0].rq_nvec = 1;
+
 	rc = server->ops->init_transform_rq(server, num_rqst + 1,
 					    &cur_rqst[0], rqst);
 	if (rc)
-		return rc;
+		goto out;
 
 	rc = __smb_send_rqst(server, num_rqst + 1, &cur_rqst[0]);
 	smb3_free_compound_rqst(num_rqst, &cur_rqst[1]);
+out:
+	kfree(tr_hdr);
 	return rc;
 }
 
@@ -927,7 +945,8 @@ cifs_check_receive(struct mid_q_entry *mid, struct TCP_Server_Info *server,
 }
 
 struct mid_q_entry *
-cifs_setup_request(struct cifs_ses *ses, struct smb_rqst *rqst)
+cifs_setup_request(struct cifs_ses *ses, struct TCP_Server_Info *ignored,
+		   struct smb_rqst *rqst)
 {
 	int rc;
 	struct smb_hdr *hdr = (struct smb_hdr *)rqst->rq_iov[0].iov_base;
@@ -999,7 +1018,18 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
 		return -EIO;
 	}
 
-	server = ses->server;
+	if (!ses->binding) {
+		uint index = 0;
+
+		if (ses->chan_count > 1) {
+			index = (uint)atomic_inc_return(&ses->chan_seq);
+			index %= ses->chan_count;
+		}
+		server = ses->chans[index].server;
+	} else {
+		server = cifs_ses_server(ses);
+	}
+
 	if (server->tcpStatus == CifsExiting)
 		return -ENOENT;
 
@@ -1044,7 +1074,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
 	}
 
 	for (i = 0; i < num_rqst; i++) {
-		midQ[i] = server->ops->setup_request(ses, &rqst[i]);
+		midQ[i] = server->ops->setup_request(ses, server, &rqst[i]);
 		if (IS_ERR(midQ[i])) {
 			revert_current_mid(server, i);
 			for (j = 0; j < i; j++)
@@ -1119,8 +1149,8 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
 				 midQ[i]->mid, le16_to_cpu(midQ[i]->command));
 			send_cancel(server, &rqst[i], midQ[i]);
 			spin_lock(&GlobalMid_Lock);
+			midQ[i]->mid_flags |= MID_WAIT_CANCELLED;
 			if (midQ[i]->mid_state == MID_REQUEST_SUBMITTED) {
-				midQ[i]->mid_flags |= MID_WAIT_CANCELLED;
 				midQ[i]->callback = cifs_cancelled_callback;
 				cancelled_mid[i] = true;
 				credits[i].value = 0;
@@ -1287,7 +1317,7 @@ SendReceive(const unsigned int xid, struct cifs_ses *ses,
 
 	rc = allocate_mid(ses, in_buf, &midQ);
 	if (rc) {
-		mutex_unlock(&ses->server->srv_mutex);
+		mutex_unlock(&server->srv_mutex);
 		/* Update # of requests on wire to server */
 		add_credits(server, &credits, 0);
 		return rc;
diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c
index db4ba8f6077e..b8299173ea7e 100644
--- a/fs/cifs/xattr.c
+++ b/fs/cifs/xattr.c
@@ -32,7 +32,8 @@
 #include "cifs_unicode.h"
 
 #define MAX_EA_VALUE_SIZE CIFSMaxBufSize
-#define CIFS_XATTR_CIFS_ACL "system.cifs_acl"
+#define CIFS_XATTR_CIFS_ACL "system.cifs_acl" /* DACL only */
+#define CIFS_XATTR_CIFS_NTSD "system.cifs_ntsd" /* owner plus DACL */
 #define CIFS_XATTR_ATTRIB "cifs.dosattrib"  /* full name: user.cifs.dosattrib */
 #define CIFS_XATTR_CREATETIME "cifs.creationtime"  /* user.cifs.creationtime */
 /*
@@ -40,12 +41,62 @@
  * confusing users and using the 20+ year old term 'cifs' when it is no longer
  * secure, replaced by SMB2 (then even more highly secure SMB3) many years ago
  */
-#define SMB3_XATTR_CIFS_ACL "system.smb3_acl"
+#define SMB3_XATTR_CIFS_ACL "system.smb3_acl" /* DACL only */
+#define SMB3_XATTR_CIFS_NTSD "system.smb3_ntsd" /* owner plus DACL */
 #define SMB3_XATTR_ATTRIB "smb3.dosattrib"  /* full name: user.smb3.dosattrib */
 #define SMB3_XATTR_CREATETIME "smb3.creationtime"  /* user.smb3.creationtime */
 /* BB need to add server (Samba e.g) support for security and trusted prefix */
 
-enum { XATTR_USER, XATTR_CIFS_ACL, XATTR_ACL_ACCESS, XATTR_ACL_DEFAULT };
+enum { XATTR_USER, XATTR_CIFS_ACL, XATTR_ACL_ACCESS, XATTR_ACL_DEFAULT,
+	XATTR_CIFS_NTSD };
+
+static int cifs_attrib_set(unsigned int xid, struct cifs_tcon *pTcon,
+			   struct inode *inode, char *full_path,
+			   const void *value, size_t size)
+{
+	ssize_t rc = -EOPNOTSUPP;
+	__u32 *pattrib = (__u32 *)value;
+	__u32 attrib;
+	FILE_BASIC_INFO info_buf;
+
+	if ((value == NULL) || (size != sizeof(__u32)))
+		return -ERANGE;
+
+	memset(&info_buf, 0, sizeof(info_buf));
+	attrib = *pattrib;
+	info_buf.Attributes = cpu_to_le32(attrib);
+	if (pTcon->ses->server->ops->set_file_info)
+		rc = pTcon->ses->server->ops->set_file_info(inode, full_path,
+				&info_buf, xid);
+	if (rc == 0)
+		CIFS_I(inode)->cifsAttrs = attrib;
+
+	return rc;
+}
+
+static int cifs_creation_time_set(unsigned int xid, struct cifs_tcon *pTcon,
+				  struct inode *inode, char *full_path,
+				  const void *value, size_t size)
+{
+	ssize_t rc = -EOPNOTSUPP;
+	__u64 *pcreation_time = (__u64 *)value;
+	__u64 creation_time;
+	FILE_BASIC_INFO info_buf;
+
+	if ((value == NULL) || (size != sizeof(__u64)))
+		return -ERANGE;
+
+	memset(&info_buf, 0, sizeof(info_buf));
+	creation_time = *pcreation_time;
+	info_buf.CreationTime = cpu_to_le64(creation_time);
+	if (pTcon->ses->server->ops->set_file_info)
+		rc = pTcon->ses->server->ops->set_file_info(inode, full_path,
+				&info_buf, xid);
+	if (rc == 0)
+		CIFS_I(inode)->createtime = creation_time;
+
+	return rc;
+}
 
 static int cifs_xattr_set(const struct xattr_handler *handler,
 			  struct dentry *dentry, struct inode *inode,
@@ -86,6 +137,23 @@ static int cifs_xattr_set(const struct xattr_handler *handler,
 
 	switch (handler->flags) {
 	case XATTR_USER:
+		cifs_dbg(FYI, "%s:setting user xattr %s\n", __func__, name);
+		if ((strcmp(name, CIFS_XATTR_ATTRIB) == 0) ||
+		    (strcmp(name, SMB3_XATTR_ATTRIB) == 0)) {
+			rc = cifs_attrib_set(xid, pTcon, inode, full_path,
+					value, size);
+			if (rc == 0) /* force revalidate of the inode */
+				CIFS_I(inode)->time = 0;
+			break;
+		} else if ((strcmp(name, CIFS_XATTR_CREATETIME) == 0) ||
+			   (strcmp(name, SMB3_XATTR_CREATETIME) == 0)) {
+			rc = cifs_creation_time_set(xid, pTcon, inode,
+					full_path, value, size);
+			if (rc == 0) /* force revalidate of the inode */
+				CIFS_I(inode)->time = 0;
+			break;
+		}
+
 		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR)
 			goto out;
 
@@ -95,7 +163,8 @@ static int cifs_xattr_set(const struct xattr_handler *handler,
 				cifs_sb->local_nls, cifs_sb);
 		break;
 
-	case XATTR_CIFS_ACL: {
+	case XATTR_CIFS_ACL:
+	case XATTR_CIFS_NTSD: {
 		struct cifs_ntsd *pacl;
 
 		if (!value)
@@ -106,12 +175,25 @@ static int cifs_xattr_set(const struct xattr_handler *handler,
 		} else {
 			memcpy(pacl, value, size);
 			if (value &&
-			    pTcon->ses->server->ops->set_acl)
-				rc = pTcon->ses->server->ops->set_acl(pacl,
-						size, inode,
-						full_path, CIFS_ACL_DACL);
-			else
+			    pTcon->ses->server->ops->set_acl) {
+				rc = 0;
+				if (handler->flags == XATTR_CIFS_NTSD) {
+					/* set owner and DACL */
+					rc = pTcon->ses->server->ops->set_acl(
+							pacl, size, inode,
+							full_path,
+							CIFS_ACL_OWNER);
+				}
+				if (rc == 0) {
+					/* set DACL */
+					rc = pTcon->ses->server->ops->set_acl(
+							pacl, size, inode,
+							full_path,
+							CIFS_ACL_DACL);
+				}
+			} else {
 				rc = -EOPNOTSUPP;
+			}
 			if (rc == 0) /* force revalidate of the inode */
 				CIFS_I(inode)->time = 0;
 			kfree(pacl);
@@ -179,7 +261,7 @@ static int cifs_creation_time_get(struct dentry *dentry, struct inode *inode,
 				  void *value, size_t size)
 {
 	ssize_t rc;
-	__u64 * pcreatetime;
+	__u64 *pcreatetime;
 
 	rc = cifs_revalidate_dentry_attr(dentry);
 	if (rc)
@@ -244,7 +326,9 @@ static int cifs_xattr_get(const struct xattr_handler *handler,
 				full_path, name, value, size, cifs_sb);
 		break;
 
-	case XATTR_CIFS_ACL: {
+	case XATTR_CIFS_ACL:
+	case XATTR_CIFS_NTSD: {
+		/* the whole ntsd is fetched regardless */
 		u32 acllen;
 		struct cifs_ntsd *pacl;
 
@@ -382,6 +466,26 @@ static const struct xattr_handler smb3_acl_xattr_handler = {
 	.set = cifs_xattr_set,
 };
 
+static const struct xattr_handler cifs_cifs_ntsd_xattr_handler = {
+	.name = CIFS_XATTR_CIFS_NTSD,
+	.flags = XATTR_CIFS_NTSD,
+	.get = cifs_xattr_get,
+	.set = cifs_xattr_set,
+};
+
+/*
+ * Although this is just an alias for the above, need to move away from
+ * confusing users and using the 20 year old term 'cifs' when it is no
+ * longer secure and was replaced by SMB2/SMB3 a long time ago, and
+ * SMB3 and later are highly secure.
+ */
+static const struct xattr_handler smb3_ntsd_xattr_handler = {
+	.name = SMB3_XATTR_CIFS_NTSD,
+	.flags = XATTR_CIFS_NTSD,
+	.get = cifs_xattr_get,
+	.set = cifs_xattr_set,
+};
+
 static const struct xattr_handler cifs_posix_acl_access_xattr_handler = {
 	.name = XATTR_NAME_POSIX_ACL_ACCESS,
 	.flags = XATTR_ACL_ACCESS,
@@ -401,6 +505,8 @@ const struct xattr_handler *cifs_xattr_handlers[] = {
 	&cifs_os2_xattr_handler,
 	&cifs_cifs_acl_xattr_handler,
 	&smb3_acl_xattr_handler, /* alias for above since avoiding "cifs" */
+	&cifs_cifs_ntsd_xattr_handler,
+	&smb3_ntsd_xattr_handler, /* alias for above since avoiding "cifs" */
 	&cifs_posix_acl_access_xattr_handler,
 	&cifs_posix_acl_default_xattr_handler,
 	NULL
diff --git a/fs/coda/Kconfig b/fs/coda/Kconfig
index ae6759f9594a..c3477eeafb3f 100644
--- a/fs/coda/Kconfig
+++ b/fs/coda/Kconfig
@@ -15,7 +15,7 @@ config CODA_FS
 	  *client*.  You will need user level code as well, both for the
 	  client and server.  Servers are currently user level, i.e. they need
 	  no kernel support.  Please read
-	  <file:Documentation/filesystems/coda.txt> and check out the Coda
+	  <file:Documentation/filesystems/coda.rst> and check out the Coda
 	  home page <http://www.coda.cs.cmu.edu/>.
 
 	  To compile the coda client support as a module, choose M here: the
diff --git a/fs/compat_binfmt_elf.c b/fs/compat_binfmt_elf.c
index b7f9ffa1d5f1..e61f3fe8e32a 100644
--- a/fs/compat_binfmt_elf.c
+++ b/fs/compat_binfmt_elf.c
@@ -17,6 +17,8 @@
 #include <linux/elfcore-compat.h>
 #include <linux/time.h>
 
+#define ELF_COMPAT	1
+
 /*
  * Rename the basic ELF layout types to refer to the 32-bit class of files.
  */
@@ -28,18 +30,20 @@
 #undef	elf_shdr
 #undef	elf_note
 #undef	elf_addr_t
+#undef	ELF_GNU_PROPERTY_ALIGN
 #define elfhdr		elf32_hdr
 #define elf_phdr	elf32_phdr
 #define elf_shdr	elf32_shdr
 #define elf_note	elf32_note
 #define elf_addr_t	Elf32_Addr
+#define ELF_GNU_PROPERTY_ALIGN	ELF32_GNU_PROPERTY_ALIGN
 
 /*
  * Some data types as stored in coredump.
  */
 #define user_long_t		compat_long_t
 #define user_siginfo_t		compat_siginfo_t
-#define copy_siginfo_to_user	copy_siginfo_to_user32
+#define copy_siginfo_to_external	copy_siginfo_to_external32
 
 /*
  * The machine-dependent core note format types are defined in elfcore-compat.h,
@@ -48,8 +52,8 @@
 #define elf_prstatus	compat_elf_prstatus
 #define elf_prpsinfo	compat_elf_prpsinfo
 
-#undef ns_to_timeval
-#define ns_to_timeval ns_to_old_timeval32
+#undef ns_to_kernel_old_timeval
+#define ns_to_kernel_old_timeval ns_to_old_timeval32
 
 /*
  * To use this file, asm/elf.h must define compat_elf_check_arch.
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
deleted file mode 100644
index a7ec2d3dff92..000000000000
--- a/fs/compat_ioctl.c
+++ /dev/null
@@ -1,1096 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * ioctl32.c: Conversion between 32bit and 64bit native ioctls.
- *
- * Copyright (C) 1997-2000  Jakub Jelinek  (jakub@redhat.com)
- * Copyright (C) 1998  Eddie C. Dost  (ecd@skynet.be)
- * Copyright (C) 2001,2002  Andi Kleen, SuSE Labs 
- * Copyright (C) 2003       Pavel Machek (pavel@ucw.cz)
- *
- * These routines maintain argument size conversion between 32bit and 64bit
- * ioctls.
- */
-
-#include <linux/joystick.h>
-
-#include <linux/types.h>
-#include <linux/compat.h>
-#include <linux/kernel.h>
-#include <linux/capability.h>
-#include <linux/compiler.h>
-#include <linux/sched.h>
-#include <linux/smp.h>
-#include <linux/ioctl.h>
-#include <linux/if.h>
-#include <linux/raid/md_u.h>
-#include <linux/falloc.h>
-#include <linux/file.h>
-#include <linux/ppp-ioctl.h>
-#include <linux/if_pppox.h>
-#include <linux/mtio.h>
-#include <linux/tty.h>
-#include <linux/vt_kern.h>
-#include <linux/raw.h>
-#include <linux/blkdev.h>
-#include <linux/rtc.h>
-#include <linux/pci.h>
-#include <linux/serial.h>
-#include <linux/ctype.h>
-#include <linux/syscalls.h>
-#include <linux/gfp.h>
-#include <linux/cec.h>
-
-#include "internal.h"
-
-#include <net/bluetooth/bluetooth.h>
-#include <net/bluetooth/hci_sock.h>
-#include <net/bluetooth/rfcomm.h>
-
-#include <linux/capi.h>
-#include <linux/gigaset_dev.h>
-
-#ifdef CONFIG_BLOCK
-#include <linux/cdrom.h>
-#include <linux/fd.h>
-#include <scsi/scsi.h>
-#include <scsi/scsi_ioctl.h>
-#include <scsi/sg.h>
-#endif
-
-#include <linux/uaccess.h>
-#include <linux/watchdog.h>
-
-#include <linux/soundcard.h>
-
-#include <linux/hiddev.h>
-
-
-#include <linux/sort.h>
-
-#ifdef CONFIG_SPARC
-#include <linux/fb.h>
-#include <asm/fbio.h>
-#endif
-
-#define convert_in_user(srcptr, dstptr)			\
-({							\
-	typeof(*srcptr) val;				\
-							\
-	get_user(val, srcptr) || put_user(val, dstptr);	\
-})
-
-static int do_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
-{
-	int err;
-
-	err = security_file_ioctl(file, cmd, arg);
-	if (err)
-		return err;
-
-	return vfs_ioctl(file, cmd, arg);
-}
-
-#ifdef CONFIG_BLOCK
-typedef struct sg_io_hdr32 {
-	compat_int_t interface_id;	/* [i] 'S' for SCSI generic (required) */
-	compat_int_t dxfer_direction;	/* [i] data transfer direction  */
-	unsigned char cmd_len;		/* [i] SCSI command length ( <= 16 bytes) */
-	unsigned char mx_sb_len;		/* [i] max length to write to sbp */
-	unsigned short iovec_count;	/* [i] 0 implies no scatter gather */
-	compat_uint_t dxfer_len;		/* [i] byte count of data transfer */
-	compat_uint_t dxferp;		/* [i], [*io] points to data transfer memory
-					      or scatter gather list */
-	compat_uptr_t cmdp;		/* [i], [*i] points to command to perform */
-	compat_uptr_t sbp;		/* [i], [*o] points to sense_buffer memory */
-	compat_uint_t timeout;		/* [i] MAX_UINT->no timeout (unit: millisec) */
-	compat_uint_t flags;		/* [i] 0 -> default, see SG_FLAG... */
-	compat_int_t pack_id;		/* [i->o] unused internally (normally) */
-	compat_uptr_t usr_ptr;		/* [i->o] unused internally */
-	unsigned char status;		/* [o] scsi status */
-	unsigned char masked_status;	/* [o] shifted, masked scsi status */
-	unsigned char msg_status;		/* [o] messaging level data (optional) */
-	unsigned char sb_len_wr;		/* [o] byte count actually written to sbp */
-	unsigned short host_status;	/* [o] errors from host adapter */
-	unsigned short driver_status;	/* [o] errors from software driver */
-	compat_int_t resid;		/* [o] dxfer_len - actual_transferred */
-	compat_uint_t duration;		/* [o] time taken by cmd (unit: millisec) */
-	compat_uint_t info;		/* [o] auxiliary information */
-} sg_io_hdr32_t;  /* 64 bytes long (on sparc32) */
-
-typedef struct sg_iovec32 {
-	compat_uint_t iov_base;
-	compat_uint_t iov_len;
-} sg_iovec32_t;
-
-static int sg_build_iovec(sg_io_hdr_t __user *sgio, void __user *dxferp, u16 iovec_count)
-{
-	sg_iovec_t __user *iov = (sg_iovec_t __user *) (sgio + 1);
-	sg_iovec32_t __user *iov32 = dxferp;
-	int i;
-
-	for (i = 0; i < iovec_count; i++) {
-		u32 base, len;
-
-		if (get_user(base, &iov32[i].iov_base) ||
-		    get_user(len, &iov32[i].iov_len) ||
-		    put_user(compat_ptr(base), &iov[i].iov_base) ||
-		    put_user(len, &iov[i].iov_len))
-			return -EFAULT;
-	}
-
-	if (put_user(iov, &sgio->dxferp))
-		return -EFAULT;
-	return 0;
-}
-
-static int sg_ioctl_trans(struct file *file, unsigned int cmd,
-			sg_io_hdr32_t __user *sgio32)
-{
-	sg_io_hdr_t __user *sgio;
-	u16 iovec_count;
-	u32 data;
-	void __user *dxferp;
-	int err;
-	int interface_id;
-
-	if (get_user(interface_id, &sgio32->interface_id))
-		return -EFAULT;
-	if (interface_id != 'S')
-		return do_ioctl(file, cmd, (unsigned long)sgio32);
-
-	if (get_user(iovec_count, &sgio32->iovec_count))
-		return -EFAULT;
-
-	{
-		void __user *top = compat_alloc_user_space(0);
-		void __user *new = compat_alloc_user_space(sizeof(sg_io_hdr_t) +
-				       (iovec_count * sizeof(sg_iovec_t)));
-		if (new > top)
-			return -EINVAL;
-
-		sgio = new;
-	}
-
-	/* Ok, now construct.  */
-	if (copy_in_user(&sgio->interface_id, &sgio32->interface_id,
-			 (2 * sizeof(int)) +
-			 (2 * sizeof(unsigned char)) +
-			 (1 * sizeof(unsigned short)) +
-			 (1 * sizeof(unsigned int))))
-		return -EFAULT;
-
-	if (get_user(data, &sgio32->dxferp))
-		return -EFAULT;
-	dxferp = compat_ptr(data);
-	if (iovec_count) {
-		if (sg_build_iovec(sgio, dxferp, iovec_count))
-			return -EFAULT;
-	} else {
-		if (put_user(dxferp, &sgio->dxferp))
-			return -EFAULT;
-	}
-
-	{
-		unsigned char __user *cmdp;
-		unsigned char __user *sbp;
-
-		if (get_user(data, &sgio32->cmdp))
-			return -EFAULT;
-		cmdp = compat_ptr(data);
-
-		if (get_user(data, &sgio32->sbp))
-			return -EFAULT;
-		sbp = compat_ptr(data);
-
-		if (put_user(cmdp, &sgio->cmdp) ||
-		    put_user(sbp, &sgio->sbp))
-			return -EFAULT;
-	}
-
-	if (copy_in_user(&sgio->timeout, &sgio32->timeout,
-			 3 * sizeof(int)))
-		return -EFAULT;
-
-	if (get_user(data, &sgio32->usr_ptr))
-		return -EFAULT;
-	if (put_user(compat_ptr(data), &sgio->usr_ptr))
-		return -EFAULT;
-
-	err = do_ioctl(file, cmd, (unsigned long) sgio);
-
-	if (err >= 0) {
-		void __user *datap;
-
-		if (copy_in_user(&sgio32->pack_id, &sgio->pack_id,
-				 sizeof(int)) ||
-		    get_user(datap, &sgio->usr_ptr) ||
-		    put_user((u32)(unsigned long)datap,
-			     &sgio32->usr_ptr) ||
-		    copy_in_user(&sgio32->status, &sgio->status,
-				 (4 * sizeof(unsigned char)) +
-				 (2 * sizeof(unsigned short)) +
-				 (3 * sizeof(int))))
-			err = -EFAULT;
-	}
-
-	return err;
-}
-
-struct compat_sg_req_info { /* used by SG_GET_REQUEST_TABLE ioctl() */
-	char req_state;
-	char orphan;
-	char sg_io_owned;
-	char problem;
-	int pack_id;
-	compat_uptr_t usr_ptr;
-	unsigned int duration;
-	int unused;
-};
-
-static int sg_grt_trans(struct file *file,
-		unsigned int cmd, struct compat_sg_req_info __user *o)
-{
-	int err, i;
-	sg_req_info_t __user *r;
-	r = compat_alloc_user_space(sizeof(sg_req_info_t)*SG_MAX_QUEUE);
-	err = do_ioctl(file, cmd, (unsigned long)r);
-	if (err < 0)
-		return err;
-	for (i = 0; i < SG_MAX_QUEUE; i++) {
-		void __user *ptr;
-		int d;
-
-		if (copy_in_user(o + i, r + i, offsetof(sg_req_info_t, usr_ptr)) ||
-		    get_user(ptr, &r[i].usr_ptr) ||
-		    get_user(d, &r[i].duration) ||
-		    put_user((u32)(unsigned long)(ptr), &o[i].usr_ptr) ||
-		    put_user(d, &o[i].duration))
-			return -EFAULT;
-	}
-	return err;
-}
-#endif /* CONFIG_BLOCK */
-
-struct sock_fprog32 {
-	unsigned short	len;
-	compat_caddr_t	filter;
-};
-
-#define PPPIOCSPASS32	_IOW('t', 71, struct sock_fprog32)
-#define PPPIOCSACTIVE32	_IOW('t', 70, struct sock_fprog32)
-
-static int ppp_sock_fprog_ioctl_trans(struct file *file,
-		unsigned int cmd, struct sock_fprog32 __user *u_fprog32)
-{
-	struct sock_fprog __user *u_fprog64 = compat_alloc_user_space(sizeof(struct sock_fprog));
-	void __user *fptr64;
-	u32 fptr32;
-	u16 flen;
-
-	if (get_user(flen, &u_fprog32->len) ||
-	    get_user(fptr32, &u_fprog32->filter))
-		return -EFAULT;
-
-	fptr64 = compat_ptr(fptr32);
-
-	if (put_user(flen, &u_fprog64->len) ||
-	    put_user(fptr64, &u_fprog64->filter))
-		return -EFAULT;
-
-	if (cmd == PPPIOCSPASS32)
-		cmd = PPPIOCSPASS;
-	else
-		cmd = PPPIOCSACTIVE;
-
-	return do_ioctl(file, cmd, (unsigned long) u_fprog64);
-}
-
-struct ppp_option_data32 {
-	compat_caddr_t	ptr;
-	u32			length;
-	compat_int_t		transmit;
-};
-#define PPPIOCSCOMPRESS32	_IOW('t', 77, struct ppp_option_data32)
-
-struct ppp_idle32 {
-	compat_time_t xmit_idle;
-	compat_time_t recv_idle;
-};
-#define PPPIOCGIDLE32		_IOR('t', 63, struct ppp_idle32)
-
-static int ppp_gidle(struct file *file, unsigned int cmd,
-		struct ppp_idle32 __user *idle32)
-{
-	struct ppp_idle __user *idle;
-	__kernel_time_t xmit, recv;
-	int err;
-
-	idle = compat_alloc_user_space(sizeof(*idle));
-
-	err = do_ioctl(file, PPPIOCGIDLE, (unsigned long) idle);
-
-	if (!err) {
-		if (get_user(xmit, &idle->xmit_idle) ||
-		    get_user(recv, &idle->recv_idle) ||
-		    put_user(xmit, &idle32->xmit_idle) ||
-		    put_user(recv, &idle32->recv_idle))
-			err = -EFAULT;
-	}
-	return err;
-}
-
-static int ppp_scompress(struct file *file, unsigned int cmd,
-	struct ppp_option_data32 __user *odata32)
-{
-	struct ppp_option_data __user *odata;
-	__u32 data;
-	void __user *datap;
-
-	odata = compat_alloc_user_space(sizeof(*odata));
-
-	if (get_user(data, &odata32->ptr))
-		return -EFAULT;
-
-	datap = compat_ptr(data);
-	if (put_user(datap, &odata->ptr))
-		return -EFAULT;
-
-	if (copy_in_user(&odata->length, &odata32->length,
-			 sizeof(__u32) + sizeof(int)))
-		return -EFAULT;
-
-	return do_ioctl(file, PPPIOCSCOMPRESS, (unsigned long) odata);
-}
-
-#ifdef CONFIG_BLOCK
-struct mtget32 {
-	compat_long_t	mt_type;
-	compat_long_t	mt_resid;
-	compat_long_t	mt_dsreg;
-	compat_long_t	mt_gstat;
-	compat_long_t	mt_erreg;
-	compat_daddr_t	mt_fileno;
-	compat_daddr_t	mt_blkno;
-};
-#define MTIOCGET32	_IOR('m', 2, struct mtget32)
-
-struct mtpos32 {
-	compat_long_t	mt_blkno;
-};
-#define MTIOCPOS32	_IOR('m', 3, struct mtpos32)
-
-static int mt_ioctl_trans(struct file *file,
-		unsigned int cmd, void __user *argp)
-{
-	/* NULL initialization to make gcc shut up */
-	struct mtget __user *get = NULL;
-	struct mtget32 __user *umget32;
-	struct mtpos __user *pos = NULL;
-	struct mtpos32 __user *upos32;
-	unsigned long kcmd;
-	void *karg;
-	int err = 0;
-
-	switch(cmd) {
-	case MTIOCPOS32:
-		kcmd = MTIOCPOS;
-		pos = compat_alloc_user_space(sizeof(*pos));
-		karg = pos;
-		break;
-	default:	/* MTIOCGET32 */
-		kcmd = MTIOCGET;
-		get = compat_alloc_user_space(sizeof(*get));
-		karg = get;
-		break;
-	}
-	if (karg == NULL)
-		return -EFAULT;
-	err = do_ioctl(file, kcmd, (unsigned long)karg);
-	if (err)
-		return err;
-	switch (cmd) {
-	case MTIOCPOS32:
-		upos32 = argp;
-		err = convert_in_user(&pos->mt_blkno, &upos32->mt_blkno);
-		break;
-	case MTIOCGET32:
-		umget32 = argp;
-		err = convert_in_user(&get->mt_type, &umget32->mt_type);
-		err |= convert_in_user(&get->mt_resid, &umget32->mt_resid);
-		err |= convert_in_user(&get->mt_dsreg, &umget32->mt_dsreg);
-		err |= convert_in_user(&get->mt_gstat, &umget32->mt_gstat);
-		err |= convert_in_user(&get->mt_erreg, &umget32->mt_erreg);
-		err |= convert_in_user(&get->mt_fileno, &umget32->mt_fileno);
-		err |= convert_in_user(&get->mt_blkno, &umget32->mt_blkno);
-		break;
-	}
-	return err ? -EFAULT: 0;
-}
-
-#endif /* CONFIG_BLOCK */
-
-/* Bluetooth ioctls */
-#define HCIUARTSETPROTO		_IOW('U', 200, int)
-#define HCIUARTGETPROTO		_IOR('U', 201, int)
-#define HCIUARTGETDEVICE	_IOR('U', 202, int)
-#define HCIUARTSETFLAGS		_IOW('U', 203, int)
-#define HCIUARTGETFLAGS		_IOR('U', 204, int)
-
-#define RTC_IRQP_READ32		_IOR('p', 0x0b, compat_ulong_t)
-#define RTC_IRQP_SET32		_IOW('p', 0x0c, compat_ulong_t)
-#define RTC_EPOCH_READ32	_IOR('p', 0x0d, compat_ulong_t)
-#define RTC_EPOCH_SET32		_IOW('p', 0x0e, compat_ulong_t)
-
-static int rtc_ioctl(struct file *file,
-		unsigned cmd, void __user *argp)
-{
-	unsigned long __user *valp = compat_alloc_user_space(sizeof(*valp));
-	int ret;
-
-	if (valp == NULL)
-		return -EFAULT;
-	switch (cmd) {
-	case RTC_IRQP_READ32:
-	case RTC_EPOCH_READ32:
-		ret = do_ioctl(file, (cmd == RTC_IRQP_READ32) ?
-					RTC_IRQP_READ : RTC_EPOCH_READ,
-					(unsigned long)valp);
-		if (ret)
-			return ret;
-		return convert_in_user(valp, (unsigned int __user *)argp);
-	case RTC_IRQP_SET32:
-		return do_ioctl(file, RTC_IRQP_SET, (unsigned long)argp);
-	case RTC_EPOCH_SET32:
-		return do_ioctl(file, RTC_EPOCH_SET, (unsigned long)argp);
-	}
-
-	return -ENOIOCTLCMD;
-}
-
-/* on ia32 l_start is on a 32-bit boundary */
-#if defined(CONFIG_IA64) || defined(CONFIG_X86_64)
-struct space_resv_32 {
-	__s16		l_type;
-	__s16		l_whence;
-	__s64		l_start	__attribute__((packed));
-			/* len == 0 means until end of file */
-	__s64		l_len __attribute__((packed));
-	__s32		l_sysid;
-	__u32		l_pid;
-	__s32		l_pad[4];	/* reserve area */
-};
-
-#define FS_IOC_RESVSP_32		_IOW ('X', 40, struct space_resv_32)
-#define FS_IOC_RESVSP64_32	_IOW ('X', 42, struct space_resv_32)
-
-/* just account for different alignment */
-static int compat_ioctl_preallocate(struct file *file,
-			struct space_resv_32    __user *p32)
-{
-	struct space_resv	__user *p = compat_alloc_user_space(sizeof(*p));
-
-	if (copy_in_user(&p->l_type,	&p32->l_type,	sizeof(s16)) ||
-	    copy_in_user(&p->l_whence,	&p32->l_whence, sizeof(s16)) ||
-	    copy_in_user(&p->l_start,	&p32->l_start,	sizeof(s64)) ||
-	    copy_in_user(&p->l_len,	&p32->l_len,	sizeof(s64)) ||
-	    copy_in_user(&p->l_sysid,	&p32->l_sysid,	sizeof(s32)) ||
-	    copy_in_user(&p->l_pid,	&p32->l_pid,	sizeof(u32)) ||
-	    copy_in_user(&p->l_pad,	&p32->l_pad,	4*sizeof(u32)))
-		return -EFAULT;
-
-	return ioctl_preallocate(file, p);
-}
-#endif
-
-/*
- * simple reversible transform to make our table more evenly
- * distributed after sorting.
- */
-#define XFORM(i) (((i) ^ ((i) << 27) ^ ((i) << 17)) & 0xffffffff)
-
-#define COMPATIBLE_IOCTL(cmd) XFORM((u32)cmd),
-/* ioctl should not be warned about even if it's not implemented.
-   Valid reasons to use this:
-   - It is implemented with ->compat_ioctl on some device, but programs
-   call it on others too.
-   - The ioctl is not implemented in the native kernel, but programs
-   call it commonly anyways.
-   Most other reasons are not valid. */
-#define IGNORE_IOCTL(cmd) COMPATIBLE_IOCTL(cmd)
-
-static unsigned int ioctl_pointer[] = {
-/* compatible ioctls first */
-/* Little t */
-COMPATIBLE_IOCTL(TIOCOUTQ)
-/* Little f */
-COMPATIBLE_IOCTL(FIOCLEX)
-COMPATIBLE_IOCTL(FIONCLEX)
-COMPATIBLE_IOCTL(FIOASYNC)
-COMPATIBLE_IOCTL(FIONBIO)
-COMPATIBLE_IOCTL(FIONREAD)  /* This is also TIOCINQ */
-COMPATIBLE_IOCTL(FS_IOC_FIEMAP)
-/* 0x00 */
-COMPATIBLE_IOCTL(FIBMAP)
-COMPATIBLE_IOCTL(FIGETBSZ)
-/* 'X' - originally XFS but some now in the VFS */
-COMPATIBLE_IOCTL(FIFREEZE)
-COMPATIBLE_IOCTL(FITHAW)
-COMPATIBLE_IOCTL(FITRIM)
-#ifdef CONFIG_BLOCK
-/* Big S */
-COMPATIBLE_IOCTL(SCSI_IOCTL_GET_IDLUN)
-COMPATIBLE_IOCTL(SCSI_IOCTL_DOORLOCK)
-COMPATIBLE_IOCTL(SCSI_IOCTL_DOORUNLOCK)
-COMPATIBLE_IOCTL(SCSI_IOCTL_TEST_UNIT_READY)
-COMPATIBLE_IOCTL(SCSI_IOCTL_GET_BUS_NUMBER)
-COMPATIBLE_IOCTL(SCSI_IOCTL_SEND_COMMAND)
-COMPATIBLE_IOCTL(SCSI_IOCTL_PROBE_HOST)
-COMPATIBLE_IOCTL(SCSI_IOCTL_GET_PCI)
-#endif
-/* Big V (don't complain on serial console) */
-IGNORE_IOCTL(VT_OPENQRY)
-IGNORE_IOCTL(VT_GETMODE)
-/* Little p (/dev/rtc, /dev/envctrl, etc.) */
-COMPATIBLE_IOCTL(RTC_AIE_ON)
-COMPATIBLE_IOCTL(RTC_AIE_OFF)
-COMPATIBLE_IOCTL(RTC_UIE_ON)
-COMPATIBLE_IOCTL(RTC_UIE_OFF)
-COMPATIBLE_IOCTL(RTC_PIE_ON)
-COMPATIBLE_IOCTL(RTC_PIE_OFF)
-COMPATIBLE_IOCTL(RTC_WIE_ON)
-COMPATIBLE_IOCTL(RTC_WIE_OFF)
-COMPATIBLE_IOCTL(RTC_ALM_SET)
-COMPATIBLE_IOCTL(RTC_ALM_READ)
-COMPATIBLE_IOCTL(RTC_RD_TIME)
-COMPATIBLE_IOCTL(RTC_SET_TIME)
-COMPATIBLE_IOCTL(RTC_WKALM_SET)
-COMPATIBLE_IOCTL(RTC_WKALM_RD)
-/*
- * These two are only for the sbus rtc driver, but
- * hwclock tries them on every rtc device first when
- * running on sparc.  On other architectures the entries
- * are useless but harmless.
- */
-COMPATIBLE_IOCTL(_IOR('p', 20, int[7])) /* RTCGET */
-COMPATIBLE_IOCTL(_IOW('p', 21, int[7])) /* RTCSET */
-/* Little m */
-COMPATIBLE_IOCTL(MTIOCTOP)
-/* Socket level stuff */
-COMPATIBLE_IOCTL(FIOQSIZE)
-#ifdef CONFIG_BLOCK
-/* md calls this on random blockdevs */
-IGNORE_IOCTL(RAID_VERSION)
-/* qemu/qemu-img might call these two on plain files for probing */
-IGNORE_IOCTL(CDROM_DRIVE_STATUS)
-IGNORE_IOCTL(FDGETPRM32)
-/* SG stuff */
-COMPATIBLE_IOCTL(SG_SET_TIMEOUT)
-COMPATIBLE_IOCTL(SG_GET_TIMEOUT)
-COMPATIBLE_IOCTL(SG_EMULATED_HOST)
-COMPATIBLE_IOCTL(SG_GET_TRANSFORM)
-COMPATIBLE_IOCTL(SG_SET_RESERVED_SIZE)
-COMPATIBLE_IOCTL(SG_GET_RESERVED_SIZE)
-COMPATIBLE_IOCTL(SG_GET_SCSI_ID)
-COMPATIBLE_IOCTL(SG_SET_FORCE_LOW_DMA)
-COMPATIBLE_IOCTL(SG_GET_LOW_DMA)
-COMPATIBLE_IOCTL(SG_SET_FORCE_PACK_ID)
-COMPATIBLE_IOCTL(SG_GET_PACK_ID)
-COMPATIBLE_IOCTL(SG_GET_NUM_WAITING)
-COMPATIBLE_IOCTL(SG_SET_DEBUG)
-COMPATIBLE_IOCTL(SG_GET_SG_TABLESIZE)
-COMPATIBLE_IOCTL(SG_GET_COMMAND_Q)
-COMPATIBLE_IOCTL(SG_SET_COMMAND_Q)
-COMPATIBLE_IOCTL(SG_GET_VERSION_NUM)
-COMPATIBLE_IOCTL(SG_NEXT_CMD_LEN)
-COMPATIBLE_IOCTL(SG_SCSI_RESET)
-COMPATIBLE_IOCTL(SG_GET_REQUEST_TABLE)
-COMPATIBLE_IOCTL(SG_SET_KEEP_ORPHAN)
-COMPATIBLE_IOCTL(SG_GET_KEEP_ORPHAN)
-#endif
-/* PPP stuff */
-COMPATIBLE_IOCTL(PPPIOCGFLAGS)
-COMPATIBLE_IOCTL(PPPIOCSFLAGS)
-COMPATIBLE_IOCTL(PPPIOCGASYNCMAP)
-COMPATIBLE_IOCTL(PPPIOCSASYNCMAP)
-COMPATIBLE_IOCTL(PPPIOCGUNIT)
-COMPATIBLE_IOCTL(PPPIOCGRASYNCMAP)
-COMPATIBLE_IOCTL(PPPIOCSRASYNCMAP)
-COMPATIBLE_IOCTL(PPPIOCGMRU)
-COMPATIBLE_IOCTL(PPPIOCSMRU)
-COMPATIBLE_IOCTL(PPPIOCSMAXCID)
-COMPATIBLE_IOCTL(PPPIOCGXASYNCMAP)
-COMPATIBLE_IOCTL(PPPIOCSXASYNCMAP)
-COMPATIBLE_IOCTL(PPPIOCXFERUNIT)
-/* PPPIOCSCOMPRESS is translated */
-COMPATIBLE_IOCTL(PPPIOCGNPMODE)
-COMPATIBLE_IOCTL(PPPIOCSNPMODE)
-COMPATIBLE_IOCTL(PPPIOCGDEBUG)
-COMPATIBLE_IOCTL(PPPIOCSDEBUG)
-/* PPPIOCSPASS is translated */
-/* PPPIOCSACTIVE is translated */
-/* PPPIOCGIDLE is translated */
-COMPATIBLE_IOCTL(PPPIOCNEWUNIT)
-COMPATIBLE_IOCTL(PPPIOCATTACH)
-COMPATIBLE_IOCTL(PPPIOCDETACH)
-COMPATIBLE_IOCTL(PPPIOCSMRRU)
-COMPATIBLE_IOCTL(PPPIOCCONNECT)
-COMPATIBLE_IOCTL(PPPIOCDISCONN)
-COMPATIBLE_IOCTL(PPPIOCATTCHAN)
-COMPATIBLE_IOCTL(PPPIOCGCHAN)
-COMPATIBLE_IOCTL(PPPIOCGL2TPSTATS)
-/* Big A */
-/* sparc only */
-/* Big Q for sound/OSS */
-COMPATIBLE_IOCTL(SNDCTL_SEQ_RESET)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_SYNC)
-COMPATIBLE_IOCTL(SNDCTL_SYNTH_INFO)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_CTRLRATE)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_GETOUTCOUNT)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_GETINCOUNT)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_PERCMODE)
-COMPATIBLE_IOCTL(SNDCTL_FM_LOAD_INSTR)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_TESTMIDI)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_RESETSAMPLES)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_NRSYNTHS)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_NRMIDIS)
-COMPATIBLE_IOCTL(SNDCTL_MIDI_INFO)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_THRESHOLD)
-COMPATIBLE_IOCTL(SNDCTL_SYNTH_MEMAVL)
-COMPATIBLE_IOCTL(SNDCTL_FM_4OP_ENABLE)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_PANIC)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_OUTOFBAND)
-COMPATIBLE_IOCTL(SNDCTL_SEQ_GETTIME)
-COMPATIBLE_IOCTL(SNDCTL_SYNTH_ID)
-COMPATIBLE_IOCTL(SNDCTL_SYNTH_CONTROL)
-COMPATIBLE_IOCTL(SNDCTL_SYNTH_REMOVESAMPLE)
-/* Big T for sound/OSS */
-COMPATIBLE_IOCTL(SNDCTL_TMR_TIMEBASE)
-COMPATIBLE_IOCTL(SNDCTL_TMR_START)
-COMPATIBLE_IOCTL(SNDCTL_TMR_STOP)
-COMPATIBLE_IOCTL(SNDCTL_TMR_CONTINUE)
-COMPATIBLE_IOCTL(SNDCTL_TMR_TEMPO)
-COMPATIBLE_IOCTL(SNDCTL_TMR_SOURCE)
-COMPATIBLE_IOCTL(SNDCTL_TMR_METRONOME)
-COMPATIBLE_IOCTL(SNDCTL_TMR_SELECT)
-/* Little m for sound/OSS */
-COMPATIBLE_IOCTL(SNDCTL_MIDI_PRETIME)
-COMPATIBLE_IOCTL(SNDCTL_MIDI_MPUMODE)
-COMPATIBLE_IOCTL(SNDCTL_MIDI_MPUCMD)
-/* Big P for sound/OSS */
-COMPATIBLE_IOCTL(SNDCTL_DSP_RESET)
-COMPATIBLE_IOCTL(SNDCTL_DSP_SYNC)
-COMPATIBLE_IOCTL(SNDCTL_DSP_SPEED)
-COMPATIBLE_IOCTL(SNDCTL_DSP_STEREO)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETBLKSIZE)
-COMPATIBLE_IOCTL(SNDCTL_DSP_CHANNELS)
-COMPATIBLE_IOCTL(SOUND_PCM_WRITE_FILTER)
-COMPATIBLE_IOCTL(SNDCTL_DSP_POST)
-COMPATIBLE_IOCTL(SNDCTL_DSP_SUBDIVIDE)
-COMPATIBLE_IOCTL(SNDCTL_DSP_SETFRAGMENT)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETFMTS)
-COMPATIBLE_IOCTL(SNDCTL_DSP_SETFMT)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETOSPACE)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETISPACE)
-COMPATIBLE_IOCTL(SNDCTL_DSP_NONBLOCK)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETCAPS)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETTRIGGER)
-COMPATIBLE_IOCTL(SNDCTL_DSP_SETTRIGGER)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETIPTR)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETOPTR)
-/* SNDCTL_DSP_MAPINBUF,  XXX needs translation */
-/* SNDCTL_DSP_MAPOUTBUF,  XXX needs translation */
-COMPATIBLE_IOCTL(SNDCTL_DSP_SETSYNCRO)
-COMPATIBLE_IOCTL(SNDCTL_DSP_SETDUPLEX)
-COMPATIBLE_IOCTL(SNDCTL_DSP_GETODELAY)
-COMPATIBLE_IOCTL(SNDCTL_DSP_PROFILE)
-COMPATIBLE_IOCTL(SOUND_PCM_READ_RATE)
-COMPATIBLE_IOCTL(SOUND_PCM_READ_CHANNELS)
-COMPATIBLE_IOCTL(SOUND_PCM_READ_BITS)
-COMPATIBLE_IOCTL(SOUND_PCM_READ_FILTER)
-/* Big C for sound/OSS */
-COMPATIBLE_IOCTL(SNDCTL_COPR_RESET)
-COMPATIBLE_IOCTL(SNDCTL_COPR_LOAD)
-COMPATIBLE_IOCTL(SNDCTL_COPR_RDATA)
-COMPATIBLE_IOCTL(SNDCTL_COPR_RCODE)
-COMPATIBLE_IOCTL(SNDCTL_COPR_WDATA)
-COMPATIBLE_IOCTL(SNDCTL_COPR_WCODE)
-COMPATIBLE_IOCTL(SNDCTL_COPR_RUN)
-COMPATIBLE_IOCTL(SNDCTL_COPR_HALT)
-COMPATIBLE_IOCTL(SNDCTL_COPR_SENDMSG)
-COMPATIBLE_IOCTL(SNDCTL_COPR_RCVMSG)
-/* Big M for sound/OSS */
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_VOLUME)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_BASS)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_TREBLE)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_SYNTH)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_PCM)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_SPEAKER)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_LINE)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_MIC)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_CD)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_IMIX)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_ALTPCM)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_RECLEV)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_IGAIN)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_OGAIN)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_LINE1)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_LINE2)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_LINE3)
-COMPATIBLE_IOCTL(MIXER_READ(SOUND_MIXER_DIGITAL1))
-COMPATIBLE_IOCTL(MIXER_READ(SOUND_MIXER_DIGITAL2))
-COMPATIBLE_IOCTL(MIXER_READ(SOUND_MIXER_DIGITAL3))
-COMPATIBLE_IOCTL(MIXER_READ(SOUND_MIXER_PHONEIN))
-COMPATIBLE_IOCTL(MIXER_READ(SOUND_MIXER_PHONEOUT))
-COMPATIBLE_IOCTL(MIXER_READ(SOUND_MIXER_VIDEO))
-COMPATIBLE_IOCTL(MIXER_READ(SOUND_MIXER_RADIO))
-COMPATIBLE_IOCTL(MIXER_READ(SOUND_MIXER_MONITOR))
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_MUTE)
-/* SOUND_MIXER_READ_ENHANCE,  same value as READ_MUTE */
-/* SOUND_MIXER_READ_LOUD,  same value as READ_MUTE */
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_RECSRC)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_DEVMASK)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_RECMASK)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_STEREODEVS)
-COMPATIBLE_IOCTL(SOUND_MIXER_READ_CAPS)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_VOLUME)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_BASS)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_TREBLE)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_SYNTH)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_PCM)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_SPEAKER)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_LINE)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_MIC)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_CD)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_IMIX)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_ALTPCM)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_RECLEV)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_IGAIN)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_OGAIN)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_LINE1)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_LINE2)
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_LINE3)
-COMPATIBLE_IOCTL(MIXER_WRITE(SOUND_MIXER_DIGITAL1))
-COMPATIBLE_IOCTL(MIXER_WRITE(SOUND_MIXER_DIGITAL2))
-COMPATIBLE_IOCTL(MIXER_WRITE(SOUND_MIXER_DIGITAL3))
-COMPATIBLE_IOCTL(MIXER_WRITE(SOUND_MIXER_PHONEIN))
-COMPATIBLE_IOCTL(MIXER_WRITE(SOUND_MIXER_PHONEOUT))
-COMPATIBLE_IOCTL(MIXER_WRITE(SOUND_MIXER_VIDEO))
-COMPATIBLE_IOCTL(MIXER_WRITE(SOUND_MIXER_RADIO))
-COMPATIBLE_IOCTL(MIXER_WRITE(SOUND_MIXER_MONITOR))
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_MUTE)
-/* SOUND_MIXER_WRITE_ENHANCE,  same value as WRITE_MUTE */
-/* SOUND_MIXER_WRITE_LOUD,  same value as WRITE_MUTE */
-COMPATIBLE_IOCTL(SOUND_MIXER_WRITE_RECSRC)
-COMPATIBLE_IOCTL(SOUND_MIXER_INFO)
-COMPATIBLE_IOCTL(SOUND_OLD_MIXER_INFO)
-COMPATIBLE_IOCTL(SOUND_MIXER_ACCESS)
-COMPATIBLE_IOCTL(SOUND_MIXER_AGC)
-COMPATIBLE_IOCTL(SOUND_MIXER_3DSE)
-COMPATIBLE_IOCTL(SOUND_MIXER_PRIVATE1)
-COMPATIBLE_IOCTL(SOUND_MIXER_PRIVATE2)
-COMPATIBLE_IOCTL(SOUND_MIXER_PRIVATE3)
-COMPATIBLE_IOCTL(SOUND_MIXER_PRIVATE4)
-COMPATIBLE_IOCTL(SOUND_MIXER_PRIVATE5)
-COMPATIBLE_IOCTL(SOUND_MIXER_GETLEVELS)
-COMPATIBLE_IOCTL(SOUND_MIXER_SETLEVELS)
-COMPATIBLE_IOCTL(OSS_GETVERSION)
-/* Raw devices */
-COMPATIBLE_IOCTL(RAW_SETBIND)
-COMPATIBLE_IOCTL(RAW_GETBIND)
-/* Watchdog */
-COMPATIBLE_IOCTL(WDIOC_GETSUPPORT)
-COMPATIBLE_IOCTL(WDIOC_GETSTATUS)
-COMPATIBLE_IOCTL(WDIOC_GETBOOTSTATUS)
-COMPATIBLE_IOCTL(WDIOC_GETTEMP)
-COMPATIBLE_IOCTL(WDIOC_SETOPTIONS)
-COMPATIBLE_IOCTL(WDIOC_KEEPALIVE)
-COMPATIBLE_IOCTL(WDIOC_SETTIMEOUT)
-COMPATIBLE_IOCTL(WDIOC_GETTIMEOUT)
-COMPATIBLE_IOCTL(WDIOC_SETPRETIMEOUT)
-COMPATIBLE_IOCTL(WDIOC_GETPRETIMEOUT)
-/* Big R */
-COMPATIBLE_IOCTL(RNDGETENTCNT)
-COMPATIBLE_IOCTL(RNDADDTOENTCNT)
-COMPATIBLE_IOCTL(RNDGETPOOL)
-COMPATIBLE_IOCTL(RNDADDENTROPY)
-COMPATIBLE_IOCTL(RNDZAPENTCNT)
-COMPATIBLE_IOCTL(RNDCLEARPOOL)
-/* Bluetooth */
-COMPATIBLE_IOCTL(HCIDEVUP)
-COMPATIBLE_IOCTL(HCIDEVDOWN)
-COMPATIBLE_IOCTL(HCIDEVRESET)
-COMPATIBLE_IOCTL(HCIDEVRESTAT)
-COMPATIBLE_IOCTL(HCIGETDEVLIST)
-COMPATIBLE_IOCTL(HCIGETDEVINFO)
-COMPATIBLE_IOCTL(HCIGETCONNLIST)
-COMPATIBLE_IOCTL(HCIGETCONNINFO)
-COMPATIBLE_IOCTL(HCIGETAUTHINFO)
-COMPATIBLE_IOCTL(HCISETRAW)
-COMPATIBLE_IOCTL(HCISETSCAN)
-COMPATIBLE_IOCTL(HCISETAUTH)
-COMPATIBLE_IOCTL(HCISETENCRYPT)
-COMPATIBLE_IOCTL(HCISETPTYPE)
-COMPATIBLE_IOCTL(HCISETLINKPOL)
-COMPATIBLE_IOCTL(HCISETLINKMODE)
-COMPATIBLE_IOCTL(HCISETACLMTU)
-COMPATIBLE_IOCTL(HCISETSCOMTU)
-COMPATIBLE_IOCTL(HCIBLOCKADDR)
-COMPATIBLE_IOCTL(HCIUNBLOCKADDR)
-COMPATIBLE_IOCTL(HCIINQUIRY)
-COMPATIBLE_IOCTL(HCIUARTSETPROTO)
-COMPATIBLE_IOCTL(HCIUARTGETPROTO)
-COMPATIBLE_IOCTL(HCIUARTGETDEVICE)
-COMPATIBLE_IOCTL(HCIUARTSETFLAGS)
-COMPATIBLE_IOCTL(HCIUARTGETFLAGS)
-COMPATIBLE_IOCTL(RFCOMMCREATEDEV)
-COMPATIBLE_IOCTL(RFCOMMRELEASEDEV)
-COMPATIBLE_IOCTL(RFCOMMGETDEVLIST)
-COMPATIBLE_IOCTL(RFCOMMGETDEVINFO)
-COMPATIBLE_IOCTL(RFCOMMSTEALDLC)
-/* CAPI */
-COMPATIBLE_IOCTL(CAPI_REGISTER)
-COMPATIBLE_IOCTL(CAPI_GET_MANUFACTURER)
-COMPATIBLE_IOCTL(CAPI_GET_VERSION)
-COMPATIBLE_IOCTL(CAPI_GET_SERIAL)
-COMPATIBLE_IOCTL(CAPI_GET_PROFILE)
-COMPATIBLE_IOCTL(CAPI_MANUFACTURER_CMD)
-COMPATIBLE_IOCTL(CAPI_GET_ERRCODE)
-COMPATIBLE_IOCTL(CAPI_INSTALLED)
-COMPATIBLE_IOCTL(CAPI_GET_FLAGS)
-COMPATIBLE_IOCTL(CAPI_SET_FLAGS)
-COMPATIBLE_IOCTL(CAPI_CLR_FLAGS)
-COMPATIBLE_IOCTL(CAPI_NCCI_OPENCOUNT)
-COMPATIBLE_IOCTL(CAPI_NCCI_GETUNIT)
-/* Misc. */
-COMPATIBLE_IOCTL(0x41545900)		/* ATYIO_CLKR */
-COMPATIBLE_IOCTL(0x41545901)		/* ATYIO_CLKW */
-COMPATIBLE_IOCTL(PCIIOC_CONTROLLER)
-COMPATIBLE_IOCTL(PCIIOC_MMAP_IS_IO)
-COMPATIBLE_IOCTL(PCIIOC_MMAP_IS_MEM)
-COMPATIBLE_IOCTL(PCIIOC_WRITE_COMBINE)
-/* hiddev */
-COMPATIBLE_IOCTL(HIDIOCGVERSION)
-COMPATIBLE_IOCTL(HIDIOCAPPLICATION)
-COMPATIBLE_IOCTL(HIDIOCGDEVINFO)
-COMPATIBLE_IOCTL(HIDIOCGSTRING)
-COMPATIBLE_IOCTL(HIDIOCINITREPORT)
-COMPATIBLE_IOCTL(HIDIOCGREPORT)
-COMPATIBLE_IOCTL(HIDIOCSREPORT)
-COMPATIBLE_IOCTL(HIDIOCGREPORTINFO)
-COMPATIBLE_IOCTL(HIDIOCGFIELDINFO)
-COMPATIBLE_IOCTL(HIDIOCGUSAGE)
-COMPATIBLE_IOCTL(HIDIOCSUSAGE)
-COMPATIBLE_IOCTL(HIDIOCGUCODE)
-COMPATIBLE_IOCTL(HIDIOCGFLAG)
-COMPATIBLE_IOCTL(HIDIOCSFLAG)
-COMPATIBLE_IOCTL(HIDIOCGCOLLECTIONINDEX)
-COMPATIBLE_IOCTL(HIDIOCGCOLLECTIONINFO)
-/* joystick */
-COMPATIBLE_IOCTL(JSIOCGVERSION)
-COMPATIBLE_IOCTL(JSIOCGAXES)
-COMPATIBLE_IOCTL(JSIOCGBUTTONS)
-COMPATIBLE_IOCTL(JSIOCGNAME(0))
-
-/* fat 'r' ioctls. These are handled by fat with ->compat_ioctl,
-   but we don't want warnings on other file systems. So declare
-   them as compatible here. */
-#define VFAT_IOCTL_READDIR_BOTH32       _IOR('r', 1, struct compat_dirent[2])
-#define VFAT_IOCTL_READDIR_SHORT32      _IOR('r', 2, struct compat_dirent[2])
-
-IGNORE_IOCTL(VFAT_IOCTL_READDIR_BOTH32)
-IGNORE_IOCTL(VFAT_IOCTL_READDIR_SHORT32)
-
-#ifdef CONFIG_SPARC
-/* Sparc framebuffers, handled in sbusfb_compat_ioctl() */
-IGNORE_IOCTL(FBIOGTYPE)
-IGNORE_IOCTL(FBIOSATTR)
-IGNORE_IOCTL(FBIOGATTR)
-IGNORE_IOCTL(FBIOSVIDEO)
-IGNORE_IOCTL(FBIOGVIDEO)
-IGNORE_IOCTL(FBIOSCURPOS)
-IGNORE_IOCTL(FBIOGCURPOS)
-IGNORE_IOCTL(FBIOGCURMAX)
-IGNORE_IOCTL(FBIOPUTCMAP32)
-IGNORE_IOCTL(FBIOGETCMAP32)
-IGNORE_IOCTL(FBIOSCURSOR32)
-IGNORE_IOCTL(FBIOGCURSOR32)
-#endif
-};
-
-/*
- * Convert common ioctl arguments based on their command number
- *
- * Please do not add any code in here. Instead, implement
- * a compat_ioctl operation in the place that handleѕ the
- * ioctl for the native case.
- */
-static long do_ioctl_trans(unsigned int cmd,
-		 unsigned long arg, struct file *file)
-{
-	void __user *argp = compat_ptr(arg);
-
-	switch (cmd) {
-	case PPPIOCGIDLE32:
-		return ppp_gidle(file, cmd, argp);
-	case PPPIOCSCOMPRESS32:
-		return ppp_scompress(file, cmd, argp);
-	case PPPIOCSPASS32:
-	case PPPIOCSACTIVE32:
-		return ppp_sock_fprog_ioctl_trans(file, cmd, argp);
-#ifdef CONFIG_BLOCK
-	case SG_IO:
-		return sg_ioctl_trans(file, cmd, argp);
-	case SG_GET_REQUEST_TABLE:
-		return sg_grt_trans(file, cmd, argp);
-	case MTIOCGET32:
-	case MTIOCPOS32:
-		return mt_ioctl_trans(file, cmd, argp);
-#endif
-	/* Not implemented in the native kernel */
-	case RTC_IRQP_READ32:
-	case RTC_IRQP_SET32:
-	case RTC_EPOCH_READ32:
-	case RTC_EPOCH_SET32:
-		return rtc_ioctl(file, cmd, argp);
-	}
-
-	/*
-	 * These take an integer instead of a pointer as 'arg',
-	 * so we must not do a compat_ptr() translation.
-	 */
-	switch (cmd) {
-	/* RAID */
-	case HOT_REMOVE_DISK:
-	case HOT_ADD_DISK:
-	case SET_DISK_FAULTY:
-	case SET_BITMAP_FILE:
-		return vfs_ioctl(file, cmd, arg);
-	}
-
-	return -ENOIOCTLCMD;
-}
-
-static int compat_ioctl_check_table(unsigned int xcmd)
-{
-	int i;
-	const int max = ARRAY_SIZE(ioctl_pointer) - 1;
-
-	BUILD_BUG_ON(max >= (1 << 16));
-
-	/* guess initial offset into table, assuming a
-	   normalized distribution */
-	i = ((xcmd >> 16) * max) >> 16;
-
-	/* do linear search up first, until greater or equal */
-	while (ioctl_pointer[i] < xcmd && i < max)
-		i++;
-
-	/* then do linear search down */
-	while (ioctl_pointer[i] > xcmd && i > 0)
-		i--;
-
-	return ioctl_pointer[i] == xcmd;
-}
-
-COMPAT_SYSCALL_DEFINE3(ioctl, unsigned int, fd, unsigned int, cmd,
-		       compat_ulong_t, arg32)
-{
-	unsigned long arg = arg32;
-	struct fd f = fdget(fd);
-	int error = -EBADF;
-	if (!f.file)
-		goto out;
-
-	/* RED-PEN how should LSM module know it's handling 32bit? */
-	error = security_file_ioctl(f.file, cmd, arg);
-	if (error)
-		goto out_fput;
-
-	/*
-	 * To allow the compat_ioctl handlers to be self contained
-	 * we need to check the common ioctls here first.
-	 * Just handle them with the standard handlers below.
-	 */
-	switch (cmd) {
-	case FIOCLEX:
-	case FIONCLEX:
-	case FIONBIO:
-	case FIOASYNC:
-	case FIOQSIZE:
-		break;
-
-#if defined(CONFIG_IA64) || defined(CONFIG_X86_64)
-	case FS_IOC_RESVSP_32:
-	case FS_IOC_RESVSP64_32:
-		error = compat_ioctl_preallocate(f.file, compat_ptr(arg));
-		goto out_fput;
-#else
-	case FS_IOC_RESVSP:
-	case FS_IOC_RESVSP64:
-		error = ioctl_preallocate(f.file, compat_ptr(arg));
-		goto out_fput;
-#endif
-
-	case FICLONE:
-	case FICLONERANGE:
-	case FIDEDUPERANGE:
-	case FS_IOC_FIEMAP:
-		goto do_ioctl;
-
-	case FIBMAP:
-	case FIGETBSZ:
-	case FIONREAD:
-		if (S_ISREG(file_inode(f.file)->i_mode))
-			break;
-		/*FALL THROUGH*/
-
-	default:
-		if (f.file->f_op->compat_ioctl) {
-			error = f.file->f_op->compat_ioctl(f.file, cmd, arg);
-			if (error != -ENOIOCTLCMD)
-				goto out_fput;
-		}
-
-		if (!f.file->f_op->unlocked_ioctl)
-			goto do_ioctl;
-		break;
-	}
-
-	if (compat_ioctl_check_table(XFORM(cmd)))
-		goto found_handler;
-
-	error = do_ioctl_trans(cmd, arg, f.file);
-	if (error == -ENOIOCTLCMD)
-		error = -ENOTTY;
-
-	goto out_fput;
-
- found_handler:
-	arg = (unsigned long)compat_ptr(arg);
- do_ioctl:
-	error = do_vfs_ioctl(f.file, fd, cmd, arg);
- out_fput:
-	fdput(f);
- out:
-	return error;
-}
-
-static int __init init_sys32_ioctl_cmp(const void *p, const void *q)
-{
-	unsigned int a, b;
-	a = *(unsigned int *)p;
-	b = *(unsigned int *)q;
-	if (a > b)
-		return 1;
-	if (a < b)
-		return -1;
-	return 0;
-}
-
-static int __init init_sys32_ioctl(void)
-{
-	sort(ioctl_pointer, ARRAY_SIZE(ioctl_pointer), sizeof(*ioctl_pointer),
-		init_sys32_ioctl_cmp, NULL);
-	return 0;
-}
-__initcall(init_sys32_ioctl);
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index cf7b7e1d5bd7..cb733652ecca 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -1519,6 +1519,7 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry)
 		spin_lock(&configfs_dirent_lock);
 		configfs_detach_rollback(dentry);
 		spin_unlock(&configfs_dirent_lock);
+		config_item_put(parent_item);
 		return -EINTR;
 	}
 	frag->frag_dead = true;
diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c
index 680aba9c00d5..8bd6a883c94c 100644
--- a/fs/configfs/inode.c
+++ b/fs/configfs/inode.c
@@ -9,7 +9,7 @@
  *
  * configfs Copyright (C) 2005 Oracle.  All rights reserved.
  *
- * Please see Documentation/filesystems/configfs/configfs.txt for more
+ * Please see Documentation/filesystems/configfs.rst for more
  * information.
  */
 
@@ -76,14 +76,11 @@ int configfs_setattr(struct dentry * dentry, struct iattr * iattr)
 	if (ia_valid & ATTR_GID)
 		sd_iattr->ia_gid = iattr->ia_gid;
 	if (ia_valid & ATTR_ATIME)
-		sd_iattr->ia_atime = timestamp_truncate(iattr->ia_atime,
-						      inode);
+		sd_iattr->ia_atime = iattr->ia_atime;
 	if (ia_valid & ATTR_MTIME)
-		sd_iattr->ia_mtime = timestamp_truncate(iattr->ia_mtime,
-						      inode);
+		sd_iattr->ia_mtime = iattr->ia_mtime;
 	if (ia_valid & ATTR_CTIME)
-		sd_iattr->ia_ctime = timestamp_truncate(iattr->ia_ctime,
-						      inode);
+		sd_iattr->ia_ctime = iattr->ia_ctime;
 	if (ia_valid & ATTR_MODE) {
 		umode_t mode = iattr->ia_mode;
 
diff --git a/fs/configfs/item.c b/fs/configfs/item.c
index 6e0f1fcb8a5b..704a4356f137 100644
--- a/fs/configfs/item.c
+++ b/fs/configfs/item.c
@@ -9,7 +9,7 @@
  *
  * configfs Copyright (C) 2005 Oracle.  All rights reserved.
  *
- * Please see the file Documentation/filesystems/configfs/configfs.txt for
+ * Please see the file Documentation/filesystems/configfs.rst for
  * critical information about using the config_item interface.
  */
 
diff --git a/fs/coredump.c b/fs/coredump.c
index b1ea7dfbd149..478a0d810136 100644
--- a/fs/coredump.c
+++ b/fs/coredump.c
@@ -211,6 +211,8 @@ static int format_corename(struct core_name *cn, struct coredump_params *cprm,
 			return -ENOMEM;
 		(*argv)[(*argc)++] = 0;
 		++pat_ptr;
+		if (!(*pat_ptr))
+			return -ENOMEM;
 	}
 
 	/* Repeat as long as we have more pattern to process and more output
@@ -517,7 +519,7 @@ static void wait_for_dump_helpers(struct file *file)
 	pipe_lock(pipe);
 	pipe->readers++;
 	pipe->writers--;
-	wake_up_interruptible_sync(&pipe->wait);
+	wake_up_interruptible_sync(&pipe->rd_wait);
 	kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
 	pipe_unlock(pipe);
 
@@ -525,7 +527,7 @@ static void wait_for_dump_helpers(struct file *file)
 	 * We actually want wait_event_freezable() but then we need
 	 * to clear TIF_SIGPENDING and improve dump_interrupted().
 	 */
-	wait_event_interruptible(pipe->wait, pipe->readers == 1);
+	wait_event_interruptible(pipe->rd_wait, pipe->readers == 1);
 
 	pipe_lock(pipe);
 	pipe->readers--;
@@ -786,6 +788,14 @@ void do_coredump(const kernel_siginfo_t *siginfo)
 	if (displaced)
 		put_files_struct(displaced);
 	if (!dump_interrupted()) {
+		/*
+		 * umh disabled with CONFIG_STATIC_USERMODEHELPER_PATH="" would
+		 * have this set to NULL.
+		 */
+		if (!cprm.file) {
+			pr_info("Core dump to |%s disabled\n", cn.corename);
+			goto close_fail;
+		}
 		file_start_write(cprm.file);
 		core_dumped = binfmt->core_dump(&cprm);
 		file_end_write(cprm.file);
diff --git a/fs/cramfs/Kconfig b/fs/cramfs/Kconfig
index c8bebb70a971..d98cef0dbb6b 100644
--- a/fs/cramfs/Kconfig
+++ b/fs/cramfs/Kconfig
@@ -9,7 +9,7 @@ config CRAMFS
 	  limited to 256MB file systems (with 16MB files), and doesn't support
 	  16/32 bits uid/gid, hard links and timestamps.
 
-	  See <file:Documentation/filesystems/cramfs.txt> and
+	  See <file:Documentation/filesystems/cramfs.rst> and
 	  <file:fs/cramfs/README> for further information.
 
 	  To compile this as a module, choose M here: the module will be called
diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
index 2f04024c3588..912308600d39 100644
--- a/fs/cramfs/inode.c
+++ b/fs/cramfs/inode.c
@@ -534,7 +534,7 @@ static int cramfs_read_super(struct super_block *sb, struct fs_context *fc,
 		/* check for wrong endianness */
 		if (super->magic == CRAMFS_MAGIC_WEND) {
 			if (!silent)
-				errorf(fc, "cramfs: wrong endianness");
+				errorfc(fc, "wrong endianness");
 			return -EINVAL;
 		}
 
@@ -546,22 +546,22 @@ static int cramfs_read_super(struct super_block *sb, struct fs_context *fc,
 		mutex_unlock(&read_mutex);
 		if (super->magic != CRAMFS_MAGIC) {
 			if (super->magic == CRAMFS_MAGIC_WEND && !silent)
-				errorf(fc, "cramfs: wrong endianness");
+				errorfc(fc, "wrong endianness");
 			else if (!silent)
-				errorf(fc, "cramfs: wrong magic");
+				errorfc(fc, "wrong magic");
 			return -EINVAL;
 		}
 	}
 
 	/* get feature flags first */
 	if (super->flags & ~CRAMFS_SUPPORTED_FLAGS) {
-		errorf(fc, "cramfs: unsupported filesystem features");
+		errorfc(fc, "unsupported filesystem features");
 		return -EINVAL;
 	}
 
 	/* Check that the root inode is in a sane state */
 	if (!S_ISDIR(super->root.mode)) {
-		errorf(fc, "cramfs: root is not a directory");
+		errorfc(fc, "root is not a directory");
 		return -EINVAL;
 	}
 	/* correct strange, hard-coded permissions of mkcramfs */
@@ -580,12 +580,12 @@ static int cramfs_read_super(struct super_block *sb, struct fs_context *fc,
 	sbi->magic = super->magic;
 	sbi->flags = super->flags;
 	if (root_offset == 0)
-		infof(fc, "cramfs: empty filesystem");
+		infofc(fc, "empty filesystem");
 	else if (!(super->flags & CRAMFS_FLAG_SHIFTED_ROOT_OFFSET) &&
 		 ((root_offset != sizeof(struct cramfs_super)) &&
 		  (root_offset != 512 + sizeof(struct cramfs_super))))
 	{
-		errorf(fc, "cramfs: bad root offset %lu", root_offset);
+		errorfc(fc, "bad root offset %lu", root_offset);
 		return -EINVAL;
 	}
 
diff --git a/fs/crypto/Kconfig b/fs/crypto/Kconfig
index ff5a1746cbae..8046d7c7a3e9 100644
--- a/fs/crypto/Kconfig
+++ b/fs/crypto/Kconfig
@@ -2,13 +2,8 @@
 config FS_ENCRYPTION
 	bool "FS Encryption (Per-file encryption)"
 	select CRYPTO
-	select CRYPTO_AES
-	select CRYPTO_CBC
-	select CRYPTO_ECB
-	select CRYPTO_XTS
-	select CRYPTO_CTS
-	select CRYPTO_SHA512
-	select CRYPTO_HMAC
+	select CRYPTO_HASH
+	select CRYPTO_SKCIPHER
 	select KEYS
 	help
 	  Enable encryption of files and directories.  This
@@ -16,3 +11,16 @@ config FS_ENCRYPTION
 	  efficient since it avoids caching the encrypted and
 	  decrypted pages in the page cache.  Currently Ext4,
 	  F2FS and UBIFS make use of this feature.
+
+# Filesystems supporting encryption must select this if FS_ENCRYPTION.  This
+# allows the algorithms to be built as modules when all the filesystems are.
+config FS_ENCRYPTION_ALGS
+	tristate
+	select CRYPTO_AES
+	select CRYPTO_CBC
+	select CRYPTO_CTS
+	select CRYPTO_ECB
+	select CRYPTO_HMAC
+	select CRYPTO_SHA256
+	select CRYPTO_SHA512
+	select CRYPTO_XTS
diff --git a/fs/crypto/bio.c b/fs/crypto/bio.c
index 82da2510721f..4fa18fff9c4e 100644
--- a/fs/crypto/bio.c
+++ b/fs/crypto/bio.c
@@ -26,7 +26,7 @@
 #include <linux/namei.h>
 #include "fscrypt_private.h"
 
-static void __fscrypt_decrypt_bio(struct bio *bio, bool done)
+void fscrypt_decrypt_bio(struct bio *bio)
 {
 	struct bio_vec *bv;
 	struct bvec_iter_all iter_all;
@@ -37,84 +37,105 @@ static void __fscrypt_decrypt_bio(struct bio *bio, bool done)
 							   bv->bv_offset);
 		if (ret)
 			SetPageError(page);
-		else if (done)
-			SetPageUptodate(page);
-		if (done)
-			unlock_page(page);
 	}
 }
-
-void fscrypt_decrypt_bio(struct bio *bio)
-{
-	__fscrypt_decrypt_bio(bio, false);
-}
 EXPORT_SYMBOL(fscrypt_decrypt_bio);
 
-static void completion_pages(struct work_struct *work)
-{
-	struct fscrypt_ctx *ctx = container_of(work, struct fscrypt_ctx, work);
-	struct bio *bio = ctx->bio;
-
-	__fscrypt_decrypt_bio(bio, true);
-	fscrypt_release_ctx(ctx);
-	bio_put(bio);
-}
-
-void fscrypt_enqueue_decrypt_bio(struct fscrypt_ctx *ctx, struct bio *bio)
-{
-	INIT_WORK(&ctx->work, completion_pages);
-	ctx->bio = bio;
-	fscrypt_enqueue_decrypt_work(&ctx->work);
-}
-EXPORT_SYMBOL(fscrypt_enqueue_decrypt_bio);
-
+/**
+ * fscrypt_zeroout_range() - zero out a range of blocks in an encrypted file
+ * @inode: the file's inode
+ * @lblk: the first file logical block to zero out
+ * @pblk: the first filesystem physical block to zero out
+ * @len: number of blocks to zero out
+ *
+ * Zero out filesystem blocks in an encrypted regular file on-disk, i.e. write
+ * ciphertext blocks which decrypt to the all-zeroes block.  The blocks must be
+ * both logically and physically contiguous.  It's also assumed that the
+ * filesystem only uses a single block device, ->s_bdev.
+ *
+ * Note that since each block uses a different IV, this involves writing a
+ * different ciphertext to each block; we can't simply reuse the same one.
+ *
+ * Return: 0 on success; -errno on failure.
+ */
 int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
-				sector_t pblk, unsigned int len)
+			  sector_t pblk, unsigned int len)
 {
 	const unsigned int blockbits = inode->i_blkbits;
 	const unsigned int blocksize = 1 << blockbits;
-	struct page *ciphertext_page;
+	const unsigned int blocks_per_page_bits = PAGE_SHIFT - blockbits;
+	const unsigned int blocks_per_page = 1 << blocks_per_page_bits;
+	struct page *pages[16]; /* write up to 16 pages at a time */
+	unsigned int nr_pages;
+	unsigned int i;
+	unsigned int offset;
 	struct bio *bio;
-	int ret, err = 0;
+	int ret, err;
 
-	ciphertext_page = fscrypt_alloc_bounce_page(GFP_NOWAIT);
-	if (!ciphertext_page)
-		return -ENOMEM;
+	if (len == 0)
+		return 0;
 
-	while (len--) {
-		err = fscrypt_crypt_block(inode, FS_ENCRYPT, lblk,
-					  ZERO_PAGE(0), ciphertext_page,
-					  blocksize, 0, GFP_NOFS);
-		if (err)
-			goto errout;
+	BUILD_BUG_ON(ARRAY_SIZE(pages) > BIO_MAX_PAGES);
+	nr_pages = min_t(unsigned int, ARRAY_SIZE(pages),
+			 (len + blocks_per_page - 1) >> blocks_per_page_bits);
+
+	/*
+	 * We need at least one page for ciphertext.  Allocate the first one
+	 * from a mempool, with __GFP_DIRECT_RECLAIM set so that it can't fail.
+	 *
+	 * Any additional page allocations are allowed to fail, as they only
+	 * help performance, and waiting on the mempool for them could deadlock.
+	 */
+	for (i = 0; i < nr_pages; i++) {
+		pages[i] = fscrypt_alloc_bounce_page(i == 0 ? GFP_NOFS :
+						     GFP_NOWAIT | __GFP_NOWARN);
+		if (!pages[i])
+			break;
+	}
+	nr_pages = i;
+	if (WARN_ON(nr_pages <= 0))
+		return -EINVAL;
+
+	/* This always succeeds since __GFP_DIRECT_RECLAIM is set. */
+	bio = bio_alloc(GFP_NOFS, nr_pages);
 
-		bio = bio_alloc(GFP_NOWAIT, 1);
-		if (!bio) {
-			err = -ENOMEM;
-			goto errout;
-		}
+	do {
 		bio_set_dev(bio, inode->i_sb->s_bdev);
 		bio->bi_iter.bi_sector = pblk << (blockbits - 9);
 		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
-		ret = bio_add_page(bio, ciphertext_page, blocksize, 0);
-		if (WARN_ON(ret != blocksize)) {
-			/* should never happen! */
-			bio_put(bio);
-			err = -EIO;
-			goto errout;
-		}
+
+		i = 0;
+		offset = 0;
+		do {
+			err = fscrypt_crypt_block(inode, FS_ENCRYPT, lblk,
+						  ZERO_PAGE(0), pages[i],
+						  blocksize, offset, GFP_NOFS);
+			if (err)
+				goto out;
+			lblk++;
+			pblk++;
+			len--;
+			offset += blocksize;
+			if (offset == PAGE_SIZE || len == 0) {
+				ret = bio_add_page(bio, pages[i++], offset, 0);
+				if (WARN_ON(ret != offset)) {
+					err = -EIO;
+					goto out;
+				}
+				offset = 0;
+			}
+		} while (i != nr_pages && len != 0);
+
 		err = submit_bio_wait(bio);
-		if (err == 0 && bio->bi_status)
-			err = -EIO;
-		bio_put(bio);
 		if (err)
-			goto errout;
-		lblk++;
-		pblk++;
-	}
+			goto out;
+		bio_reset(bio);
+	} while (len != 0);
 	err = 0;
-errout:
-	fscrypt_free_bounce_page(ciphertext_page);
+out:
+	bio_put(bio);
+	for (i = 0; i < nr_pages; i++)
+		fscrypt_free_bounce_page(pages[i]);
 	return err;
 }
 EXPORT_SYMBOL(fscrypt_zeroout_range);
diff --git a/fs/crypto/crypto.c b/fs/crypto/crypto.c
index 32a7ad0098cc..ed015cb66c7c 100644
--- a/fs/crypto/crypto.c
+++ b/fs/crypto/crypto.c
@@ -25,31 +25,20 @@
 #include <linux/module.h>
 #include <linux/scatterlist.h>
 #include <linux/ratelimit.h>
-#include <linux/dcache.h>
-#include <linux/namei.h>
-#include <crypto/aes.h>
 #include <crypto/skcipher.h>
 #include "fscrypt_private.h"
 
 static unsigned int num_prealloc_crypto_pages = 32;
-static unsigned int num_prealloc_crypto_ctxs = 128;
 
 module_param(num_prealloc_crypto_pages, uint, 0444);
 MODULE_PARM_DESC(num_prealloc_crypto_pages,
 		"Number of crypto pages to preallocate");
-module_param(num_prealloc_crypto_ctxs, uint, 0444);
-MODULE_PARM_DESC(num_prealloc_crypto_ctxs,
-		"Number of crypto contexts to preallocate");
 
 static mempool_t *fscrypt_bounce_page_pool = NULL;
 
-static LIST_HEAD(fscrypt_free_ctxs);
-static DEFINE_SPINLOCK(fscrypt_ctx_lock);
-
 static struct workqueue_struct *fscrypt_read_workqueue;
 static DEFINE_MUTEX(fscrypt_init_mutex);
 
-static struct kmem_cache *fscrypt_ctx_cachep;
 struct kmem_cache *fscrypt_info_cachep;
 
 void fscrypt_enqueue_decrypt_work(struct work_struct *work)
@@ -58,62 +47,6 @@ void fscrypt_enqueue_decrypt_work(struct work_struct *work)
 }
 EXPORT_SYMBOL(fscrypt_enqueue_decrypt_work);
 
-/**
- * fscrypt_release_ctx() - Release a decryption context
- * @ctx: The decryption context to release.
- *
- * If the decryption context was allocated from the pre-allocated pool, return
- * it to that pool.  Else, free it.
- */
-void fscrypt_release_ctx(struct fscrypt_ctx *ctx)
-{
-	unsigned long flags;
-
-	if (ctx->flags & FS_CTX_REQUIRES_FREE_ENCRYPT_FL) {
-		kmem_cache_free(fscrypt_ctx_cachep, ctx);
-	} else {
-		spin_lock_irqsave(&fscrypt_ctx_lock, flags);
-		list_add(&ctx->free_list, &fscrypt_free_ctxs);
-		spin_unlock_irqrestore(&fscrypt_ctx_lock, flags);
-	}
-}
-EXPORT_SYMBOL(fscrypt_release_ctx);
-
-/**
- * fscrypt_get_ctx() - Get a decryption context
- * @gfp_flags:   The gfp flag for memory allocation
- *
- * Allocate and initialize a decryption context.
- *
- * Return: A new decryption context on success; an ERR_PTR() otherwise.
- */
-struct fscrypt_ctx *fscrypt_get_ctx(gfp_t gfp_flags)
-{
-	struct fscrypt_ctx *ctx;
-	unsigned long flags;
-
-	/*
-	 * First try getting a ctx from the free list so that we don't have to
-	 * call into the slab allocator.
-	 */
-	spin_lock_irqsave(&fscrypt_ctx_lock, flags);
-	ctx = list_first_entry_or_null(&fscrypt_free_ctxs,
-					struct fscrypt_ctx, free_list);
-	if (ctx)
-		list_del(&ctx->free_list);
-	spin_unlock_irqrestore(&fscrypt_ctx_lock, flags);
-	if (!ctx) {
-		ctx = kmem_cache_zalloc(fscrypt_ctx_cachep, gfp_flags);
-		if (!ctx)
-			return ERR_PTR(-ENOMEM);
-		ctx->flags |= FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
-	} else {
-		ctx->flags &= ~FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
-	}
-	return ctx;
-}
-EXPORT_SYMBOL(fscrypt_get_ctx);
-
 struct page *fscrypt_alloc_bounce_page(gfp_t gfp_flags)
 {
 	return mempool_alloc(fscrypt_bounce_page_pool, gfp_flags);
@@ -121,6 +54,7 @@ struct page *fscrypt_alloc_bounce_page(gfp_t gfp_flags)
 
 /**
  * fscrypt_free_bounce_page() - free a ciphertext bounce page
+ * @bounce_page: the bounce page to free, or NULL
  *
  * Free a bounce page that was allocated by fscrypt_encrypt_pagecache_blocks(),
  * or by fscrypt_alloc_bounce_page() directly.
@@ -138,14 +72,21 @@ EXPORT_SYMBOL(fscrypt_free_bounce_page);
 void fscrypt_generate_iv(union fscrypt_iv *iv, u64 lblk_num,
 			 const struct fscrypt_info *ci)
 {
+	u8 flags = fscrypt_policy_flags(&ci->ci_policy);
+
 	memset(iv, 0, ci->ci_mode->ivsize);
-	iv->lblk_num = cpu_to_le64(lblk_num);
 
-	if (fscrypt_is_direct_key_policy(&ci->ci_policy))
+	if (flags & FSCRYPT_POLICY_FLAG_IV_INO_LBLK_64) {
+		WARN_ON_ONCE(lblk_num > U32_MAX);
+		WARN_ON_ONCE(ci->ci_inode->i_ino > U32_MAX);
+		lblk_num |= (u64)ci->ci_inode->i_ino << 32;
+	} else if (flags & FSCRYPT_POLICY_FLAG_IV_INO_LBLK_32) {
+		WARN_ON_ONCE(lblk_num > U32_MAX);
+		lblk_num = (u32)(ci->ci_hashed_ino + lblk_num);
+	} else if (flags & FSCRYPT_POLICY_FLAG_DIRECT_KEY) {
 		memcpy(iv->nonce, ci->ci_nonce, FS_KEY_DERIVATION_NONCE_SIZE);
-
-	if (ci->ci_essiv_tfm != NULL)
-		crypto_cipher_encrypt_one(ci->ci_essiv_tfm, iv->raw, iv->raw);
+	}
+	iv->lblk_num = cpu_to_le64(lblk_num);
 }
 
 /* Encrypt or decrypt a single filesystem block of file contents */
@@ -196,13 +137,14 @@ int fscrypt_crypt_block(const struct inode *inode, fscrypt_direction_t rw,
 }
 
 /**
- * fscrypt_encrypt_pagecache_blocks() - Encrypt filesystem blocks from a pagecache page
+ * fscrypt_encrypt_pagecache_blocks() - Encrypt filesystem blocks from a
+ *					pagecache page
  * @page:      The locked pagecache page containing the block(s) to encrypt
  * @len:       Total size of the block(s) to encrypt.  Must be a nonzero
  *		multiple of the filesystem's block size.
  * @offs:      Byte offset within @page of the first block to encrypt.  Must be
  *		a multiple of the filesystem's block size.
- * @gfp_flags: Memory allocation flags
+ * @gfp_flags: Memory allocation flags.  See details below.
  *
  * A new bounce page is allocated, and the specified block(s) are encrypted into
  * it.  In the bounce page, the ciphertext block(s) will be located at the same
@@ -212,6 +154,11 @@ int fscrypt_crypt_block(const struct inode *inode, fscrypt_direction_t rw,
  *
  * This is for use by the filesystem's ->writepages() method.
  *
+ * The bounce page allocation is mempool-backed, so it will always succeed when
+ * @gfp_flags includes __GFP_DIRECT_RECLAIM, e.g. when it's GFP_NOFS.  However,
+ * only the first page of each bio can be allocated this way.  To prevent
+ * deadlocks, for any additional pages a mask like GFP_NOWAIT must be used.
+ *
  * Return: the new encrypted bounce page on success; an ERR_PTR() on failure
  */
 struct page *fscrypt_encrypt_pagecache_blocks(struct page *page,
@@ -281,7 +228,8 @@ int fscrypt_encrypt_block_inplace(const struct inode *inode, struct page *page,
 EXPORT_SYMBOL(fscrypt_encrypt_block_inplace);
 
 /**
- * fscrypt_decrypt_pagecache_blocks() - Decrypt filesystem blocks in a pagecache page
+ * fscrypt_decrypt_pagecache_blocks() - Decrypt filesystem blocks in a
+ *					pagecache page
  * @page:      The locked pagecache page containing the block(s) to decrypt
  * @len:       Total size of the block(s) to decrypt.  Must be a nonzero
  *		multiple of the filesystem's block size.
@@ -348,65 +296,6 @@ int fscrypt_decrypt_block_inplace(const struct inode *inode, struct page *page,
 }
 EXPORT_SYMBOL(fscrypt_decrypt_block_inplace);
 
-/*
- * Validate dentries in encrypted directories to make sure we aren't potentially
- * caching stale dentries after a key has been added.
- */
-static int fscrypt_d_revalidate(struct dentry *dentry, unsigned int flags)
-{
-	struct dentry *dir;
-	int err;
-	int valid;
-
-	/*
-	 * Plaintext names are always valid, since fscrypt doesn't support
-	 * reverting to ciphertext names without evicting the directory's inode
-	 * -- which implies eviction of the dentries in the directory.
-	 */
-	if (!(dentry->d_flags & DCACHE_ENCRYPTED_NAME))
-		return 1;
-
-	/*
-	 * Ciphertext name; valid if the directory's key is still unavailable.
-	 *
-	 * Although fscrypt forbids rename() on ciphertext names, we still must
-	 * use dget_parent() here rather than use ->d_parent directly.  That's
-	 * because a corrupted fs image may contain directory hard links, which
-	 * the VFS handles by moving the directory's dentry tree in the dcache
-	 * each time ->lookup() finds the directory and it already has a dentry
-	 * elsewhere.  Thus ->d_parent can be changing, and we must safely grab
-	 * a reference to some ->d_parent to prevent it from being freed.
-	 */
-
-	if (flags & LOOKUP_RCU)
-		return -ECHILD;
-
-	dir = dget_parent(dentry);
-	err = fscrypt_get_encryption_info(d_inode(dir));
-	valid = !fscrypt_has_encryption_key(d_inode(dir));
-	dput(dir);
-
-	if (err < 0)
-		return err;
-
-	return valid;
-}
-
-const struct dentry_operations fscrypt_d_ops = {
-	.d_revalidate = fscrypt_d_revalidate,
-};
-
-static void fscrypt_destroy(void)
-{
-	struct fscrypt_ctx *pos, *n;
-
-	list_for_each_entry_safe(pos, n, &fscrypt_free_ctxs, free_list)
-		kmem_cache_free(fscrypt_ctx_cachep, pos);
-	INIT_LIST_HEAD(&fscrypt_free_ctxs);
-	mempool_destroy(fscrypt_bounce_page_pool);
-	fscrypt_bounce_page_pool = NULL;
-}
-
 /**
  * fscrypt_initialize() - allocate major buffers for fs encryption.
  * @cop_flags:  fscrypt operations flags
@@ -414,11 +303,11 @@ static void fscrypt_destroy(void)
  * We only call this when we start accessing encrypted files, since it
  * results in memory getting allocated that wouldn't otherwise be used.
  *
- * Return: Zero on success, non-zero otherwise.
+ * Return: 0 on success; -errno on failure
  */
 int fscrypt_initialize(unsigned int cop_flags)
 {
-	int i, res = -ENOMEM;
+	int err = 0;
 
 	/* No need to allocate a bounce page pool if this FS won't use it. */
 	if (cop_flags & FS_CFLG_OWN_PAGES)
@@ -426,29 +315,18 @@ int fscrypt_initialize(unsigned int cop_flags)
 
 	mutex_lock(&fscrypt_init_mutex);
 	if (fscrypt_bounce_page_pool)
-		goto already_initialized;
-
-	for (i = 0; i < num_prealloc_crypto_ctxs; i++) {
-		struct fscrypt_ctx *ctx;
-
-		ctx = kmem_cache_zalloc(fscrypt_ctx_cachep, GFP_NOFS);
-		if (!ctx)
-			goto fail;
-		list_add(&ctx->free_list, &fscrypt_free_ctxs);
-	}
+		goto out_unlock;
 
+	err = -ENOMEM;
 	fscrypt_bounce_page_pool =
 		mempool_create_page_pool(num_prealloc_crypto_pages, 0);
 	if (!fscrypt_bounce_page_pool)
-		goto fail;
+		goto out_unlock;
 
-already_initialized:
+	err = 0;
+out_unlock:
 	mutex_unlock(&fscrypt_init_mutex);
-	return 0;
-fail:
-	fscrypt_destroy();
-	mutex_unlock(&fscrypt_init_mutex);
-	return res;
+	return err;
 }
 
 void fscrypt_msg(const struct inode *inode, const char *level,
@@ -475,6 +353,8 @@ void fscrypt_msg(const struct inode *inode, const char *level,
 
 /**
  * fscrypt_init() - Set up for fs encryption.
+ *
+ * Return: 0 on success; -errno on failure
  */
 static int __init fscrypt_init(void)
 {
@@ -494,13 +374,9 @@ static int __init fscrypt_init(void)
 	if (!fscrypt_read_workqueue)
 		goto fail;
 
-	fscrypt_ctx_cachep = KMEM_CACHE(fscrypt_ctx, SLAB_RECLAIM_ACCOUNT);
-	if (!fscrypt_ctx_cachep)
-		goto fail_free_queue;
-
 	fscrypt_info_cachep = KMEM_CACHE(fscrypt_info, SLAB_RECLAIM_ACCOUNT);
 	if (!fscrypt_info_cachep)
-		goto fail_free_ctx;
+		goto fail_free_queue;
 
 	err = fscrypt_init_keyring();
 	if (err)
@@ -510,8 +386,6 @@ static int __init fscrypt_init(void)
 
 fail_free_info:
 	kmem_cache_destroy(fscrypt_info_cachep);
-fail_free_ctx:
-	kmem_cache_destroy(fscrypt_ctx_cachep);
 fail_free_queue:
 	destroy_workqueue(fscrypt_read_workqueue);
 fail:
diff --git a/fs/crypto/fname.c b/fs/crypto/fname.c
index 3da3707c10e3..83ca5f1e7934 100644
--- a/fs/crypto/fname.c
+++ b/fs/crypto/fname.c
@@ -11,10 +11,82 @@
  * This has not yet undergone a rigorous security audit.
  */
 
+#include <linux/namei.h>
 #include <linux/scatterlist.h>
+#include <crypto/hash.h>
+#include <crypto/sha.h>
 #include <crypto/skcipher.h>
 #include "fscrypt_private.h"
 
+/*
+ * struct fscrypt_nokey_name - identifier for directory entry when key is absent
+ *
+ * When userspace lists an encrypted directory without access to the key, the
+ * filesystem must present a unique "no-key name" for each filename that allows
+ * it to find the directory entry again if requested.  Naively, that would just
+ * mean using the ciphertext filenames.  However, since the ciphertext filenames
+ * can contain illegal characters ('\0' and '/'), they must be encoded in some
+ * way.  We use base64.  But that can cause names to exceed NAME_MAX (255
+ * bytes), so we also need to use a strong hash to abbreviate long names.
+ *
+ * The filesystem may also need another kind of hash, the "dirhash", to quickly
+ * find the directory entry.  Since filesystems normally compute the dirhash
+ * over the on-disk filename (i.e. the ciphertext), it's not computable from
+ * no-key names that abbreviate the ciphertext using the strong hash to fit in
+ * NAME_MAX.  It's also not computable if it's a keyed hash taken over the
+ * plaintext (but it may still be available in the on-disk directory entry);
+ * casefolded directories use this type of dirhash.  At least in these cases,
+ * each no-key name must include the name's dirhash too.
+ *
+ * To meet all these requirements, we base64-encode the following
+ * variable-length structure.  It contains the dirhash, or 0's if the filesystem
+ * didn't provide one; up to 149 bytes of the ciphertext name; and for
+ * ciphertexts longer than 149 bytes, also the SHA-256 of the remaining bytes.
+ *
+ * This ensures that each no-key name contains everything needed to find the
+ * directory entry again, contains only legal characters, doesn't exceed
+ * NAME_MAX, is unambiguous unless there's a SHA-256 collision, and that we only
+ * take the performance hit of SHA-256 on very long filenames (which are rare).
+ */
+struct fscrypt_nokey_name {
+	u32 dirhash[2];
+	u8 bytes[149];
+	u8 sha256[SHA256_DIGEST_SIZE];
+}; /* 189 bytes => 252 bytes base64-encoded, which is <= NAME_MAX (255) */
+
+/*
+ * Decoded size of max-size nokey name, i.e. a name that was abbreviated using
+ * the strong hash and thus includes the 'sha256' field.  This isn't simply
+ * sizeof(struct fscrypt_nokey_name), as the padding at the end isn't included.
+ */
+#define FSCRYPT_NOKEY_NAME_MAX	offsetofend(struct fscrypt_nokey_name, sha256)
+
+static struct crypto_shash *sha256_hash_tfm;
+
+static int fscrypt_do_sha256(const u8 *data, unsigned int data_len, u8 *result)
+{
+	struct crypto_shash *tfm = READ_ONCE(sha256_hash_tfm);
+
+	if (unlikely(!tfm)) {
+		struct crypto_shash *prev_tfm;
+
+		tfm = crypto_alloc_shash("sha256", 0, 0);
+		if (IS_ERR(tfm)) {
+			fscrypt_err(NULL,
+				    "Error allocating SHA-256 transform: %ld",
+				    PTR_ERR(tfm));
+			return PTR_ERR(tfm);
+		}
+		prev_tfm = cmpxchg(&sha256_hash_tfm, NULL, tfm);
+		if (prev_tfm) {
+			crypto_free_shash(tfm);
+			tfm = prev_tfm;
+		}
+	}
+
+	return crypto_shash_tfm_digest(tfm, data, data_len, result);
+}
+
 static inline bool fscrypt_is_dot_dotdot(const struct qstr *str)
 {
 	if (str->len == 1 && str->name[0] == '.')
@@ -27,19 +99,22 @@ static inline bool fscrypt_is_dot_dotdot(const struct qstr *str)
 }
 
 /**
- * fname_encrypt() - encrypt a filename
- *
- * The output buffer must be at least as large as the input buffer.
- * Any extra space is filled with NUL padding before encryption.
+ * fscrypt_fname_encrypt() - encrypt a filename
+ * @inode: inode of the parent directory (for regular filenames)
+ *	   or of the symlink (for symlink targets)
+ * @iname: the filename to encrypt
+ * @out: (output) the encrypted filename
+ * @olen: size of the encrypted filename.  It must be at least @iname->len.
+ *	  Any extra space is filled with NUL padding before encryption.
  *
  * Return: 0 on success, -errno on failure
  */
-int fname_encrypt(struct inode *inode, const struct qstr *iname,
-		  u8 *out, unsigned int olen)
+int fscrypt_fname_encrypt(const struct inode *inode, const struct qstr *iname,
+			  u8 *out, unsigned int olen)
 {
 	struct skcipher_request *req = NULL;
 	DECLARE_CRYPTO_WAIT(wait);
-	struct fscrypt_info *ci = inode->i_crypt_info;
+	const struct fscrypt_info *ci = inode->i_crypt_info;
 	struct crypto_skcipher *tfm = ci->ci_ctfm;
 	union fscrypt_iv iv;
 	struct scatterlist sg;
@@ -80,19 +155,22 @@ int fname_encrypt(struct inode *inode, const struct qstr *iname,
 
 /**
  * fname_decrypt() - decrypt a filename
- *
- * The caller must have allocated sufficient memory for the @oname string.
+ * @inode: inode of the parent directory (for regular filenames)
+ *	   or of the symlink (for symlink targets)
+ * @iname: the encrypted filename to decrypt
+ * @oname: (output) the decrypted filename.  The caller must have allocated
+ *	   enough space for this, e.g. using fscrypt_fname_alloc_buffer().
  *
  * Return: 0 on success, -errno on failure
  */
-static int fname_decrypt(struct inode *inode,
-				const struct fscrypt_str *iname,
-				struct fscrypt_str *oname)
+static int fname_decrypt(const struct inode *inode,
+			 const struct fscrypt_str *iname,
+			 struct fscrypt_str *oname)
 {
 	struct skcipher_request *req = NULL;
 	DECLARE_CRYPTO_WAIT(wait);
 	struct scatterlist src_sg, dst_sg;
-	struct fscrypt_info *ci = inode->i_crypt_info;
+	const struct fscrypt_info *ci = inode->i_crypt_info;
 	struct crypto_skcipher *tfm = ci->ci_ctfm;
 	union fscrypt_iv iv;
 	int res;
@@ -129,7 +207,10 @@ static const char lookup_table[65] =
 #define BASE64_CHARS(nbytes)	DIV_ROUND_UP((nbytes) * 4, 3)
 
 /**
- * base64_encode() -
+ * base64_encode() - base64-encode some bytes
+ * @src: the bytes to encode
+ * @len: number of bytes to encode
+ * @dst: (output) the base64-encoded string.  Not NUL-terminated.
  *
  * Encodes the input string using characters from the set [A-Za-z0-9+,].
  * The encoded string is roughly 4/3 times the size of the input string.
@@ -195,7 +276,12 @@ bool fscrypt_fname_encrypted_size(const struct inode *inode, u32 orig_len,
 }
 
 /**
- * fscrypt_fname_alloc_buffer - allocate a buffer for presented filenames
+ * fscrypt_fname_alloc_buffer() - allocate a buffer for presented filenames
+ * @inode: inode of the parent directory (for regular filenames)
+ *	   or of the symlink (for symlink targets)
+ * @max_encrypted_len: maximum length of encrypted filenames the buffer will be
+ *		       used to present
+ * @crypto_str: (output) buffer to allocate
  *
  * Allocate a buffer that is large enough to hold any decrypted or encoded
  * filename (null-terminated), for the given maximum encrypted filename length.
@@ -206,9 +292,7 @@ int fscrypt_fname_alloc_buffer(const struct inode *inode,
 			       u32 max_encrypted_len,
 			       struct fscrypt_str *crypto_str)
 {
-	const u32 max_encoded_len =
-		max_t(u32, BASE64_CHARS(FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE),
-		      1 + BASE64_CHARS(sizeof(struct fscrypt_digested_name)));
+	const u32 max_encoded_len = BASE64_CHARS(FSCRYPT_NOKEY_NAME_MAX);
 	u32 max_presented_len;
 
 	max_presented_len = max(max_encoded_len, max_encrypted_len);
@@ -222,9 +306,10 @@ int fscrypt_fname_alloc_buffer(const struct inode *inode,
 EXPORT_SYMBOL(fscrypt_fname_alloc_buffer);
 
 /**
- * fscrypt_fname_free_buffer - free the buffer for presented filenames
+ * fscrypt_fname_free_buffer() - free a buffer for presented filenames
+ * @crypto_str: the buffer to free
  *
- * Free the buffer allocated by fscrypt_fname_alloc_buffer().
+ * Free a buffer that was allocated by fscrypt_fname_alloc_buffer().
  */
 void fscrypt_fname_free_buffer(struct fscrypt_str *crypto_str)
 {
@@ -236,24 +321,35 @@ void fscrypt_fname_free_buffer(struct fscrypt_str *crypto_str)
 EXPORT_SYMBOL(fscrypt_fname_free_buffer);
 
 /**
- * fscrypt_fname_disk_to_usr() - converts a filename from disk space to user
- * space
+ * fscrypt_fname_disk_to_usr() - convert an encrypted filename to
+ *				 user-presentable form
+ * @inode: inode of the parent directory (for regular filenames)
+ *	   or of the symlink (for symlink targets)
+ * @hash: first part of the name's dirhash, if applicable.  This only needs to
+ *	  be provided if the filename is located in an indexed directory whose
+ *	  encryption key may be unavailable.  Not needed for symlink targets.
+ * @minor_hash: second part of the name's dirhash, if applicable
+ * @iname: encrypted filename to convert.  May also be "." or "..", which
+ *	   aren't actually encrypted.
+ * @oname: output buffer for the user-presentable filename.  The caller must
+ *	   have allocated enough space for this, e.g. using
+ *	   fscrypt_fname_alloc_buffer().
  *
- * The caller must have allocated sufficient memory for the @oname string.
- *
- * If the key is available, we'll decrypt the disk name; otherwise, we'll encode
- * it for presentation.  Short names are directly base64-encoded, while long
- * names are encoded in fscrypt_digested_name format.
+ * If the key is available, we'll decrypt the disk name.  Otherwise, we'll
+ * encode it for presentation in fscrypt_nokey_name format.
+ * See struct fscrypt_nokey_name for details.
  *
  * Return: 0 on success, -errno on failure
  */
-int fscrypt_fname_disk_to_usr(struct inode *inode,
-			u32 hash, u32 minor_hash,
-			const struct fscrypt_str *iname,
-			struct fscrypt_str *oname)
+int fscrypt_fname_disk_to_usr(const struct inode *inode,
+			      u32 hash, u32 minor_hash,
+			      const struct fscrypt_str *iname,
+			      struct fscrypt_str *oname)
 {
 	const struct qstr qname = FSTR_TO_QSTR(iname);
-	struct fscrypt_digested_name digested_name;
+	struct fscrypt_nokey_name nokey_name;
+	u32 size; /* size of the unencoded no-key name */
+	int err;
 
 	if (fscrypt_is_dot_dotdot(&qname)) {
 		oname->name[0] = '.';
@@ -268,24 +364,37 @@ int fscrypt_fname_disk_to_usr(struct inode *inode,
 	if (fscrypt_has_encryption_key(inode))
 		return fname_decrypt(inode, iname, oname);
 
-	if (iname->len <= FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE) {
-		oname->len = base64_encode(iname->name, iname->len,
-					   oname->name);
-		return 0;
-	}
+	/*
+	 * Sanity check that struct fscrypt_nokey_name doesn't have padding
+	 * between fields and that its encoded size never exceeds NAME_MAX.
+	 */
+	BUILD_BUG_ON(offsetofend(struct fscrypt_nokey_name, dirhash) !=
+		     offsetof(struct fscrypt_nokey_name, bytes));
+	BUILD_BUG_ON(offsetofend(struct fscrypt_nokey_name, bytes) !=
+		     offsetof(struct fscrypt_nokey_name, sha256));
+	BUILD_BUG_ON(BASE64_CHARS(FSCRYPT_NOKEY_NAME_MAX) > NAME_MAX);
+
 	if (hash) {
-		digested_name.hash = hash;
-		digested_name.minor_hash = minor_hash;
+		nokey_name.dirhash[0] = hash;
+		nokey_name.dirhash[1] = minor_hash;
+	} else {
+		nokey_name.dirhash[0] = 0;
+		nokey_name.dirhash[1] = 0;
+	}
+	if (iname->len <= sizeof(nokey_name.bytes)) {
+		memcpy(nokey_name.bytes, iname->name, iname->len);
+		size = offsetof(struct fscrypt_nokey_name, bytes[iname->len]);
 	} else {
-		digested_name.hash = 0;
-		digested_name.minor_hash = 0;
+		memcpy(nokey_name.bytes, iname->name, sizeof(nokey_name.bytes));
+		/* Compute strong hash of remaining part of name. */
+		err = fscrypt_do_sha256(&iname->name[sizeof(nokey_name.bytes)],
+					iname->len - sizeof(nokey_name.bytes),
+					nokey_name.sha256);
+		if (err)
+			return err;
+		size = FSCRYPT_NOKEY_NAME_MAX;
 	}
-	memcpy(digested_name.digest,
-	       FSCRYPT_FNAME_DIGEST(iname->name, iname->len),
-	       FSCRYPT_FNAME_DIGEST_SIZE);
-	oname->name[0] = '_';
-	oname->len = 1 + base64_encode((const u8 *)&digested_name,
-				       sizeof(digested_name), oname->name + 1);
+	oname->len = base64_encode((const u8 *)&nokey_name, size, oname->name);
 	return 0;
 }
 EXPORT_SYMBOL(fscrypt_fname_disk_to_usr);
@@ -306,8 +415,7 @@ EXPORT_SYMBOL(fscrypt_fname_disk_to_usr);
  * get the disk_name.
  *
  * Else, for keyless @lookup operations, @iname is the presented ciphertext, so
- * we decode it to get either the ciphertext disk_name (for short names) or the
- * fscrypt_digested_name (for long names).  Non-@lookup operations will be
+ * we decode it to get the fscrypt_nokey_name.  Non-@lookup operations will be
  * impossible in this case, so we fail them with ENOKEY.
  *
  * If successful, fscrypt_free_filename() must be called later to clean up.
@@ -317,8 +425,8 @@ EXPORT_SYMBOL(fscrypt_fname_disk_to_usr);
 int fscrypt_setup_filename(struct inode *dir, const struct qstr *iname,
 			      int lookup, struct fscrypt_name *fname)
 {
+	struct fscrypt_nokey_name *nokey_name;
 	int ret;
-	int digested;
 
 	memset(fname, 0, sizeof(struct fscrypt_name));
 	fname->usr_fname = iname;
@@ -342,8 +450,8 @@ int fscrypt_setup_filename(struct inode *dir, const struct qstr *iname,
 		if (!fname->crypto_buf.name)
 			return -ENOMEM;
 
-		ret = fname_encrypt(dir, iname, fname->crypto_buf.name,
-				    fname->crypto_buf.len);
+		ret = fscrypt_fname_encrypt(dir, iname, fname->crypto_buf.name,
+					    fname->crypto_buf.len);
 		if (ret)
 			goto errout;
 		fname->disk_name.name = fname->crypto_buf.name;
@@ -358,40 +466,31 @@ int fscrypt_setup_filename(struct inode *dir, const struct qstr *iname,
 	 * We don't have the key and we are doing a lookup; decode the
 	 * user-supplied name
 	 */
-	if (iname->name[0] == '_') {
-		if (iname->len !=
-		    1 + BASE64_CHARS(sizeof(struct fscrypt_digested_name)))
-			return -ENOENT;
-		digested = 1;
-	} else {
-		if (iname->len >
-		    BASE64_CHARS(FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE))
-			return -ENOENT;
-		digested = 0;
-	}
 
-	fname->crypto_buf.name =
-		kmalloc(max_t(size_t, FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE,
-			      sizeof(struct fscrypt_digested_name)),
-			GFP_KERNEL);
+	if (iname->len > BASE64_CHARS(FSCRYPT_NOKEY_NAME_MAX))
+		return -ENOENT;
+
+	fname->crypto_buf.name = kmalloc(FSCRYPT_NOKEY_NAME_MAX, GFP_KERNEL);
 	if (fname->crypto_buf.name == NULL)
 		return -ENOMEM;
 
-	ret = base64_decode(iname->name + digested, iname->len - digested,
-			    fname->crypto_buf.name);
-	if (ret < 0) {
+	ret = base64_decode(iname->name, iname->len, fname->crypto_buf.name);
+	if (ret < (int)offsetof(struct fscrypt_nokey_name, bytes[1]) ||
+	    (ret > offsetof(struct fscrypt_nokey_name, sha256) &&
+	     ret != FSCRYPT_NOKEY_NAME_MAX)) {
 		ret = -ENOENT;
 		goto errout;
 	}
 	fname->crypto_buf.len = ret;
-	if (digested) {
-		const struct fscrypt_digested_name *n =
-			(const void *)fname->crypto_buf.name;
-		fname->hash = n->hash;
-		fname->minor_hash = n->minor_hash;
-	} else {
-		fname->disk_name.name = fname->crypto_buf.name;
-		fname->disk_name.len = fname->crypto_buf.len;
+
+	nokey_name = (void *)fname->crypto_buf.name;
+	fname->hash = nokey_name->dirhash[0];
+	fname->minor_hash = nokey_name->dirhash[1];
+	if (ret != FSCRYPT_NOKEY_NAME_MAX) {
+		/* The full ciphertext filename is available. */
+		fname->disk_name.name = nokey_name->bytes;
+		fname->disk_name.len =
+			ret - offsetof(struct fscrypt_nokey_name, bytes);
 	}
 	return 0;
 
@@ -400,3 +499,109 @@ errout:
 	return ret;
 }
 EXPORT_SYMBOL(fscrypt_setup_filename);
+
+/**
+ * fscrypt_match_name() - test whether the given name matches a directory entry
+ * @fname: the name being searched for
+ * @de_name: the name from the directory entry
+ * @de_name_len: the length of @de_name in bytes
+ *
+ * Normally @fname->disk_name will be set, and in that case we simply compare
+ * that to the name stored in the directory entry.  The only exception is that
+ * if we don't have the key for an encrypted directory and the name we're
+ * looking for is very long, then we won't have the full disk_name and instead
+ * we'll need to match against a fscrypt_nokey_name that includes a strong hash.
+ *
+ * Return: %true if the name matches, otherwise %false.
+ */
+bool fscrypt_match_name(const struct fscrypt_name *fname,
+			const u8 *de_name, u32 de_name_len)
+{
+	const struct fscrypt_nokey_name *nokey_name =
+		(const void *)fname->crypto_buf.name;
+	u8 sha256[SHA256_DIGEST_SIZE];
+
+	if (likely(fname->disk_name.name)) {
+		if (de_name_len != fname->disk_name.len)
+			return false;
+		return !memcmp(de_name, fname->disk_name.name, de_name_len);
+	}
+	if (de_name_len <= sizeof(nokey_name->bytes))
+		return false;
+	if (memcmp(de_name, nokey_name->bytes, sizeof(nokey_name->bytes)))
+		return false;
+	if (fscrypt_do_sha256(&de_name[sizeof(nokey_name->bytes)],
+			      de_name_len - sizeof(nokey_name->bytes), sha256))
+		return false;
+	return !memcmp(sha256, nokey_name->sha256, sizeof(sha256));
+}
+EXPORT_SYMBOL_GPL(fscrypt_match_name);
+
+/**
+ * fscrypt_fname_siphash() - calculate the SipHash of a filename
+ * @dir: the parent directory
+ * @name: the filename to calculate the SipHash of
+ *
+ * Given a plaintext filename @name and a directory @dir which uses SipHash as
+ * its dirhash method and has had its fscrypt key set up, this function
+ * calculates the SipHash of that name using the directory's secret dirhash key.
+ *
+ * Return: the SipHash of @name using the hash key of @dir
+ */
+u64 fscrypt_fname_siphash(const struct inode *dir, const struct qstr *name)
+{
+	const struct fscrypt_info *ci = dir->i_crypt_info;
+
+	WARN_ON(!ci->ci_dirhash_key_initialized);
+
+	return siphash(name->name, name->len, &ci->ci_dirhash_key);
+}
+EXPORT_SYMBOL_GPL(fscrypt_fname_siphash);
+
+/*
+ * Validate dentries in encrypted directories to make sure we aren't potentially
+ * caching stale dentries after a key has been added.
+ */
+static int fscrypt_d_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	struct dentry *dir;
+	int err;
+	int valid;
+
+	/*
+	 * Plaintext names are always valid, since fscrypt doesn't support
+	 * reverting to ciphertext names without evicting the directory's inode
+	 * -- which implies eviction of the dentries in the directory.
+	 */
+	if (!(dentry->d_flags & DCACHE_ENCRYPTED_NAME))
+		return 1;
+
+	/*
+	 * Ciphertext name; valid if the directory's key is still unavailable.
+	 *
+	 * Although fscrypt forbids rename() on ciphertext names, we still must
+	 * use dget_parent() here rather than use ->d_parent directly.  That's
+	 * because a corrupted fs image may contain directory hard links, which
+	 * the VFS handles by moving the directory's dentry tree in the dcache
+	 * each time ->lookup() finds the directory and it already has a dentry
+	 * elsewhere.  Thus ->d_parent can be changing, and we must safely grab
+	 * a reference to some ->d_parent to prevent it from being freed.
+	 */
+
+	if (flags & LOOKUP_RCU)
+		return -ECHILD;
+
+	dir = dget_parent(dentry);
+	err = fscrypt_get_encryption_info(d_inode(dir));
+	valid = !fscrypt_has_encryption_key(d_inode(dir));
+	dput(dir);
+
+	if (err < 0)
+		return err;
+
+	return valid;
+}
+
+const struct dentry_operations fscrypt_d_ops = {
+	.d_revalidate = fscrypt_d_revalidate,
+};
diff --git a/fs/crypto/fscrypt_private.h b/fs/crypto/fscrypt_private.h
index e84efc01512e..eb7fcd2b7fb8 100644
--- a/fs/crypto/fscrypt_private.h
+++ b/fs/crypto/fscrypt_private.h
@@ -12,6 +12,7 @@
 #define _FSCRYPT_PRIVATE_H
 
 #include <linux/fscrypt.h>
+#include <linux/siphash.h>
 #include <crypto/hash.h>
 
 #define CONST_STRLEN(str)	(sizeof(str) - 1)
@@ -42,7 +43,7 @@ struct fscrypt_context_v2 {
 	u8 nonce[FS_KEY_DERIVATION_NONCE_SIZE];
 };
 
-/**
+/*
  * fscrypt_context - the encryption context of an inode
  *
  * This is the on-disk equivalent of an fscrypt_policy, stored alongside each
@@ -75,6 +76,26 @@ static inline int fscrypt_context_size(const union fscrypt_context *ctx)
 	return 0;
 }
 
+/* Check whether an fscrypt_context has a recognized version number and size */
+static inline bool fscrypt_context_is_valid(const union fscrypt_context *ctx,
+					    int ctx_size)
+{
+	return ctx_size >= 1 && ctx_size == fscrypt_context_size(ctx);
+}
+
+/* Retrieve the context's nonce, assuming the context was already validated */
+static inline const u8 *fscrypt_context_nonce(const union fscrypt_context *ctx)
+{
+	switch (ctx->version) {
+	case FSCRYPT_CONTEXT_V1:
+		return ctx->v1.nonce;
+	case FSCRYPT_CONTEXT_V2:
+		return ctx->v2.nonce;
+	}
+	WARN_ON(1);
+	return NULL;
+}
+
 #undef fscrypt_policy
 union fscrypt_policy {
 	u8 version;
@@ -136,13 +157,7 @@ fscrypt_policy_flags(const union fscrypt_policy *policy)
 	BUG();
 }
 
-static inline bool
-fscrypt_is_direct_key_policy(const union fscrypt_policy *policy)
-{
-	return fscrypt_policy_flags(policy) & FSCRYPT_POLICY_FLAG_DIRECT_KEY;
-}
-
-/**
+/*
  * For encrypted symlinks, the ciphertext length is stored at the beginning
  * of the string in little-endian format.
  */
@@ -163,11 +178,8 @@ struct fscrypt_info {
 	/* The actual crypto transform used for encryption and decryption */
 	struct crypto_skcipher *ci_ctfm;
 
-	/*
-	 * Cipher for ESSIV IV generation.  Only set for CBC contents
-	 * encryption, otherwise is NULL.
-	 */
-	struct crypto_cipher *ci_essiv_tfm;
+	/* True if the key should be freed when this fscrypt_info is freed */
+	bool ci_owns_key;
 
 	/*
 	 * Encryption mode used for this inode.  It corresponds to either the
@@ -197,11 +209,22 @@ struct fscrypt_info {
 	 */
 	struct fscrypt_direct_key *ci_direct_key;
 
+	/*
+	 * This inode's hash key for filenames.  This is a 128-bit SipHash-2-4
+	 * key.  This is only set for directories that use a keyed dirhash over
+	 * the plaintext filenames -- currently just casefolded directories.
+	 */
+	siphash_key_t ci_dirhash_key;
+	bool ci_dirhash_key_initialized;
+
 	/* The encryption policy used by this inode */
 	union fscrypt_policy ci_policy;
 
 	/* This inode's nonce, copied from the fscrypt_context */
 	u8 ci_nonce[FS_KEY_DERIVATION_NONCE_SIZE];
+
+	/* Hashed inode number.  Only set for IV_INO_LBLK_32 */
+	u32 ci_hashed_ino;
 };
 
 typedef enum {
@@ -209,38 +232,16 @@ typedef enum {
 	FS_ENCRYPT,
 } fscrypt_direction_t;
 
-#define FS_CTX_REQUIRES_FREE_ENCRYPT_FL		0x00000001
-
-static inline bool fscrypt_valid_enc_modes(u32 contents_mode,
-					   u32 filenames_mode)
-{
-	if (contents_mode == FSCRYPT_MODE_AES_128_CBC &&
-	    filenames_mode == FSCRYPT_MODE_AES_128_CTS)
-		return true;
-
-	if (contents_mode == FSCRYPT_MODE_AES_256_XTS &&
-	    filenames_mode == FSCRYPT_MODE_AES_256_CTS)
-		return true;
-
-	if (contents_mode == FSCRYPT_MODE_ADIANTUM &&
-	    filenames_mode == FSCRYPT_MODE_ADIANTUM)
-		return true;
-
-	return false;
-}
-
 /* crypto.c */
 extern struct kmem_cache *fscrypt_info_cachep;
-extern int fscrypt_initialize(unsigned int cop_flags);
-extern int fscrypt_crypt_block(const struct inode *inode,
-			       fscrypt_direction_t rw, u64 lblk_num,
-			       struct page *src_page, struct page *dest_page,
-			       unsigned int len, unsigned int offs,
-			       gfp_t gfp_flags);
-extern struct page *fscrypt_alloc_bounce_page(gfp_t gfp_flags);
-extern const struct dentry_operations fscrypt_d_ops;
-
-extern void __printf(3, 4) __cold
+int fscrypt_initialize(unsigned int cop_flags);
+int fscrypt_crypt_block(const struct inode *inode, fscrypt_direction_t rw,
+			u64 lblk_num, struct page *src_page,
+			struct page *dest_page, unsigned int len,
+			unsigned int offs, gfp_t gfp_flags);
+struct page *fscrypt_alloc_bounce_page(gfp_t gfp_flags);
+
+void __printf(3, 4) __cold
 fscrypt_msg(const struct inode *inode, const char *level, const char *fmt, ...);
 
 #define fscrypt_warn(inode, fmt, ...)		\
@@ -265,11 +266,11 @@ void fscrypt_generate_iv(union fscrypt_iv *iv, u64 lblk_num,
 			 const struct fscrypt_info *ci);
 
 /* fname.c */
-extern int fname_encrypt(struct inode *inode, const struct qstr *iname,
-			 u8 *out, unsigned int olen);
-extern bool fscrypt_fname_encrypted_size(const struct inode *inode,
-					 u32 orig_len, u32 max_len,
-					 u32 *encrypted_len_ret);
+int fscrypt_fname_encrypt(const struct inode *inode, const struct qstr *iname,
+			  u8 *out, unsigned int olen);
+bool fscrypt_fname_encrypted_size(const struct inode *inode, u32 orig_len,
+				  u32 max_len, u32 *encrypted_len_ret);
+extern const struct dentry_operations fscrypt_d_ops;
 
 /* hkdf.c */
 
@@ -277,8 +278,8 @@ struct fscrypt_hkdf {
 	struct crypto_shash *hmac_tfm;
 };
 
-extern int fscrypt_init_hkdf(struct fscrypt_hkdf *hkdf, const u8 *master_key,
-			     unsigned int master_key_size);
+int fscrypt_init_hkdf(struct fscrypt_hkdf *hkdf, const u8 *master_key,
+		      unsigned int master_key_size);
 
 /*
  * The list of contexts in which fscrypt uses HKDF.  These values are used as
@@ -288,14 +289,18 @@ extern int fscrypt_init_hkdf(struct fscrypt_hkdf *hkdf, const u8 *master_key,
  * output doesn't reveal another.
  */
 #define HKDF_CONTEXT_KEY_IDENTIFIER	1
-#define HKDF_CONTEXT_PER_FILE_KEY	2
-#define HKDF_CONTEXT_PER_MODE_KEY	3
+#define HKDF_CONTEXT_PER_FILE_ENC_KEY	2
+#define HKDF_CONTEXT_DIRECT_KEY		3
+#define HKDF_CONTEXT_IV_INO_LBLK_64_KEY	4
+#define HKDF_CONTEXT_DIRHASH_KEY	5
+#define HKDF_CONTEXT_IV_INO_LBLK_32_KEY	6
+#define HKDF_CONTEXT_INODE_HASH_KEY	7
 
-extern int fscrypt_hkdf_expand(struct fscrypt_hkdf *hkdf, u8 context,
-			       const u8 *info, unsigned int infolen,
-			       u8 *okm, unsigned int okmlen);
+int fscrypt_hkdf_expand(const struct fscrypt_hkdf *hkdf, u8 context,
+			const u8 *info, unsigned int infolen,
+			u8 *okm, unsigned int okmlen);
 
-extern void fscrypt_destroy_hkdf(struct fscrypt_hkdf *hkdf);
+void fscrypt_destroy_hkdf(struct fscrypt_hkdf *hkdf);
 
 /* keyring.c */
 
@@ -386,8 +391,17 @@ struct fscrypt_master_key {
 	struct list_head	mk_decrypted_inodes;
 	spinlock_t		mk_decrypted_inodes_lock;
 
-	/* Per-mode tfms for DIRECT_KEY policies, allocated on-demand */
-	struct crypto_skcipher	*mk_mode_keys[__FSCRYPT_MODE_MAX + 1];
+	/*
+	 * Per-mode encryption keys for the various types of encryption policies
+	 * that use them.  Allocated and derived on-demand.
+	 */
+	struct crypto_skcipher *mk_direct_keys[__FSCRYPT_MODE_MAX + 1];
+	struct crypto_skcipher *mk_iv_ino_lblk_64_keys[__FSCRYPT_MODE_MAX + 1];
+	struct crypto_skcipher *mk_iv_ino_lblk_32_keys[__FSCRYPT_MODE_MAX + 1];
+
+	/* Hash key for inode numbers.  Initialized only when needed. */
+	siphash_key_t		mk_ino_hash_key;
+	bool			mk_ino_hash_key_initialized;
 
 } __randomize_layout;
 
@@ -427,14 +441,17 @@ static inline int master_key_spec_len(const struct fscrypt_key_specifier *spec)
 	return 0;
 }
 
-extern struct key *
+struct key *
 fscrypt_find_master_key(struct super_block *sb,
 			const struct fscrypt_key_specifier *mk_spec);
 
-extern int fscrypt_verify_key_added(struct super_block *sb,
-				    const u8 identifier[FSCRYPT_KEY_IDENTIFIER_SIZE]);
+int fscrypt_add_test_dummy_key(struct super_block *sb,
+			       struct fscrypt_key_specifier *key_spec);
+
+int fscrypt_verify_key_added(struct super_block *sb,
+			     const u8 identifier[FSCRYPT_KEY_IDENTIFIER_SIZE]);
 
-extern int __init fscrypt_init_keyring(void);
+int __init fscrypt_init_keyring(void);
 
 /* keysetup.c */
 
@@ -443,40 +460,37 @@ struct fscrypt_mode {
 	const char *cipher_str;
 	int keysize;
 	int ivsize;
-	bool logged_impl_name;
-	bool needs_essiv;
+	int logged_impl_name;
 };
 
-static inline bool
-fscrypt_mode_supports_direct_key(const struct fscrypt_mode *mode)
-{
-	return mode->ivsize >= offsetofend(union fscrypt_iv, nonce);
-}
+extern struct fscrypt_mode fscrypt_modes[];
 
-extern struct crypto_skcipher *
-fscrypt_allocate_skcipher(struct fscrypt_mode *mode, const u8 *raw_key,
-			  const struct inode *inode);
+struct crypto_skcipher *fscrypt_allocate_skcipher(struct fscrypt_mode *mode,
+						  const u8 *raw_key,
+						  const struct inode *inode);
 
-extern int fscrypt_set_derived_key(struct fscrypt_info *ci,
-				   const u8 *derived_key);
+int fscrypt_set_per_file_enc_key(struct fscrypt_info *ci, const u8 *raw_key);
+
+int fscrypt_derive_dirhash_key(struct fscrypt_info *ci,
+			       const struct fscrypt_master_key *mk);
 
 /* keysetup_v1.c */
 
-extern void fscrypt_put_direct_key(struct fscrypt_direct_key *dk);
+void fscrypt_put_direct_key(struct fscrypt_direct_key *dk);
+
+int fscrypt_setup_v1_file_key(struct fscrypt_info *ci,
+			      const u8 *raw_master_key);
 
-extern int fscrypt_setup_v1_file_key(struct fscrypt_info *ci,
-				     const u8 *raw_master_key);
+int fscrypt_setup_v1_file_key_via_subscribed_keyrings(struct fscrypt_info *ci);
 
-extern int fscrypt_setup_v1_file_key_via_subscribed_keyrings(
-					struct fscrypt_info *ci);
 /* policy.c */
 
-extern bool fscrypt_policies_equal(const union fscrypt_policy *policy1,
-				   const union fscrypt_policy *policy2);
-extern bool fscrypt_supported_policy(const union fscrypt_policy *policy_u,
-				     const struct inode *inode);
-extern int fscrypt_policy_from_context(union fscrypt_policy *policy_u,
-				       const union fscrypt_context *ctx_u,
-				       int ctx_size);
+bool fscrypt_policies_equal(const union fscrypt_policy *policy1,
+			    const union fscrypt_policy *policy2);
+bool fscrypt_supported_policy(const union fscrypt_policy *policy_u,
+			      const struct inode *inode);
+int fscrypt_policy_from_context(union fscrypt_policy *policy_u,
+				const union fscrypt_context *ctx_u,
+				int ctx_size);
 
 #endif /* _FSCRYPT_PRIVATE_H */
diff --git a/fs/crypto/hkdf.c b/fs/crypto/hkdf.c
index f21873e1b467..0cba7928446d 100644
--- a/fs/crypto/hkdf.c
+++ b/fs/crypto/hkdf.c
@@ -44,17 +44,13 @@ static int hkdf_extract(struct crypto_shash *hmac_tfm, const u8 *ikm,
 			unsigned int ikmlen, u8 prk[HKDF_HASHLEN])
 {
 	static const u8 default_salt[HKDF_HASHLEN];
-	SHASH_DESC_ON_STACK(desc, hmac_tfm);
 	int err;
 
 	err = crypto_shash_setkey(hmac_tfm, default_salt, HKDF_HASHLEN);
 	if (err)
 		return err;
 
-	desc->tfm = hmac_tfm;
-	err = crypto_shash_digest(desc, ikm, ikmlen, prk);
-	shash_desc_zero(desc);
-	return err;
+	return crypto_shash_tfm_digest(hmac_tfm, ikm, ikmlen, prk);
 }
 
 /*
@@ -112,7 +108,7 @@ out:
  * adds to its application-specific info strings to guarantee that it doesn't
  * accidentally repeat an info string when using HKDF for different purposes.)
  */
-int fscrypt_hkdf_expand(struct fscrypt_hkdf *hkdf, u8 context,
+int fscrypt_hkdf_expand(const struct fscrypt_hkdf *hkdf, u8 context,
 			const u8 *info, unsigned int infolen,
 			u8 *okm, unsigned int okmlen)
 {
diff --git a/fs/crypto/hooks.c b/fs/crypto/hooks.c
index bb3b7fcfdd48..09fb8aa0f2e9 100644
--- a/fs/crypto/hooks.c
+++ b/fs/crypto/hooks.c
@@ -5,10 +5,12 @@
  * Encryption hooks for higher-level filesystem operations.
  */
 
+#include <linux/key.h>
+
 #include "fscrypt_private.h"
 
 /**
- * fscrypt_file_open - prepare to open a possibly-encrypted regular file
+ * fscrypt_file_open() - prepare to open a possibly-encrypted regular file
  * @inode: the inode being opened
  * @filp: the struct file being set up
  *
@@ -122,6 +124,48 @@ int __fscrypt_prepare_lookup(struct inode *dir, struct dentry *dentry,
 }
 EXPORT_SYMBOL_GPL(__fscrypt_prepare_lookup);
 
+/**
+ * fscrypt_prepare_setflags() - prepare to change flags with FS_IOC_SETFLAGS
+ * @inode: the inode on which flags are being changed
+ * @oldflags: the old flags
+ * @flags: the new flags
+ *
+ * The caller should be holding i_rwsem for write.
+ *
+ * Return: 0 on success; -errno if the flags change isn't allowed or if
+ *	   another error occurs.
+ */
+int fscrypt_prepare_setflags(struct inode *inode,
+			     unsigned int oldflags, unsigned int flags)
+{
+	struct fscrypt_info *ci;
+	struct fscrypt_master_key *mk;
+	int err;
+
+	/*
+	 * When the CASEFOLD flag is set on an encrypted directory, we must
+	 * derive the secret key needed for the dirhash.  This is only possible
+	 * if the directory uses a v2 encryption policy.
+	 */
+	if (IS_ENCRYPTED(inode) && (flags & ~oldflags & FS_CASEFOLD_FL)) {
+		err = fscrypt_require_key(inode);
+		if (err)
+			return err;
+		ci = inode->i_crypt_info;
+		if (ci->ci_policy.version != FSCRYPT_POLICY_V2)
+			return -EINVAL;
+		mk = ci->ci_master_key->payload.data[0];
+		down_read(&mk->mk_secret_sem);
+		if (is_master_key_secret_present(&mk->mk_secret))
+			err = fscrypt_derive_dirhash_key(ci, mk);
+		else
+			err = -ENOKEY;
+		up_read(&mk->mk_secret_sem);
+		return err;
+	}
+	return 0;
+}
+
 int __fscrypt_prepare_symlink(struct inode *dir, unsigned int len,
 			      unsigned int max_len,
 			      struct fscrypt_str *disk_link)
@@ -188,7 +232,8 @@ int __fscrypt_encrypt_symlink(struct inode *inode, const char *target,
 	ciphertext_len = disk_link->len - sizeof(*sd);
 	sd->len = cpu_to_le16(ciphertext_len);
 
-	err = fname_encrypt(inode, &iname, sd->encrypted_path, ciphertext_len);
+	err = fscrypt_fname_encrypt(inode, &iname, sd->encrypted_path,
+				    ciphertext_len);
 	if (err)
 		goto err_free_sd;
 
@@ -217,7 +262,7 @@ err_free_sd:
 EXPORT_SYMBOL_GPL(__fscrypt_encrypt_symlink);
 
 /**
- * fscrypt_get_symlink - get the target of an encrypted symlink
+ * fscrypt_get_symlink() - get the target of an encrypted symlink
  * @inode: the symlink inode
  * @caddr: the on-disk contents of the symlink
  * @max_size: size of @caddr buffer
diff --git a/fs/crypto/keyring.c b/fs/crypto/keyring.c
index c34fa7c61b43..e24eb48bfbe1 100644
--- a/fs/crypto/keyring.c
+++ b/fs/crypto/keyring.c
@@ -20,6 +20,7 @@
 
 #include <crypto/skcipher.h>
 #include <linux/key-type.h>
+#include <linux/random.h>
 #include <linux/seq_file.h>
 
 #include "fscrypt_private.h"
@@ -43,8 +44,11 @@ static void free_master_key(struct fscrypt_master_key *mk)
 
 	wipe_master_key_secret(&mk->mk_secret);
 
-	for (i = 0; i < ARRAY_SIZE(mk->mk_mode_keys); i++)
-		crypto_free_skcipher(mk->mk_mode_keys[i]);
+	for (i = 0; i <= __FSCRYPT_MODE_MAX; i++) {
+		crypto_free_skcipher(mk->mk_direct_keys[i]);
+		crypto_free_skcipher(mk->mk_iv_ino_lblk_64_keys[i]);
+		crypto_free_skcipher(mk->mk_iv_ino_lblk_32_keys[i]);
+	}
 
 	key_put(mk->mk_users);
 	kzfree(mk);
@@ -149,7 +153,7 @@ static struct key *search_fscrypt_keyring(struct key *keyring,
 }
 
 #define FSCRYPT_FS_KEYRING_DESCRIPTION_SIZE	\
-	(CONST_STRLEN("fscrypt-") + FIELD_SIZEOF(struct super_block, s_id))
+	(CONST_STRLEN("fscrypt-") + sizeof_field(struct super_block, s_id))
 
 #define FSCRYPT_MK_DESCRIPTION_SIZE	(2 * FSCRYPT_KEY_IDENTIFIER_SIZE + 1)
 
@@ -422,9 +426,9 @@ static int add_existing_master_key(struct fscrypt_master_key *mk,
 	return 0;
 }
 
-static int add_master_key(struct super_block *sb,
-			  struct fscrypt_master_key_secret *secret,
-			  const struct fscrypt_key_specifier *mk_spec)
+static int do_add_master_key(struct super_block *sb,
+			     struct fscrypt_master_key_secret *secret,
+			     const struct fscrypt_key_specifier *mk_spec)
 {
 	static DEFINE_MUTEX(fscrypt_add_key_mutex);
 	struct key *key;
@@ -463,6 +467,138 @@ out_unlock:
 	return err;
 }
 
+static int add_master_key(struct super_block *sb,
+			  struct fscrypt_master_key_secret *secret,
+			  struct fscrypt_key_specifier *key_spec)
+{
+	int err;
+
+	if (key_spec->type == FSCRYPT_KEY_SPEC_TYPE_IDENTIFIER) {
+		err = fscrypt_init_hkdf(&secret->hkdf, secret->raw,
+					secret->size);
+		if (err)
+			return err;
+
+		/*
+		 * Now that the HKDF context is initialized, the raw key is no
+		 * longer needed.
+		 */
+		memzero_explicit(secret->raw, secret->size);
+
+		/* Calculate the key identifier */
+		err = fscrypt_hkdf_expand(&secret->hkdf,
+					  HKDF_CONTEXT_KEY_IDENTIFIER, NULL, 0,
+					  key_spec->u.identifier,
+					  FSCRYPT_KEY_IDENTIFIER_SIZE);
+		if (err)
+			return err;
+	}
+	return do_add_master_key(sb, secret, key_spec);
+}
+
+static int fscrypt_provisioning_key_preparse(struct key_preparsed_payload *prep)
+{
+	const struct fscrypt_provisioning_key_payload *payload = prep->data;
+
+	if (prep->datalen < sizeof(*payload) + FSCRYPT_MIN_KEY_SIZE ||
+	    prep->datalen > sizeof(*payload) + FSCRYPT_MAX_KEY_SIZE)
+		return -EINVAL;
+
+	if (payload->type != FSCRYPT_KEY_SPEC_TYPE_DESCRIPTOR &&
+	    payload->type != FSCRYPT_KEY_SPEC_TYPE_IDENTIFIER)
+		return -EINVAL;
+
+	if (payload->__reserved)
+		return -EINVAL;
+
+	prep->payload.data[0] = kmemdup(payload, prep->datalen, GFP_KERNEL);
+	if (!prep->payload.data[0])
+		return -ENOMEM;
+
+	prep->quotalen = prep->datalen;
+	return 0;
+}
+
+static void fscrypt_provisioning_key_free_preparse(
+					struct key_preparsed_payload *prep)
+{
+	kzfree(prep->payload.data[0]);
+}
+
+static void fscrypt_provisioning_key_describe(const struct key *key,
+					      struct seq_file *m)
+{
+	seq_puts(m, key->description);
+	if (key_is_positive(key)) {
+		const struct fscrypt_provisioning_key_payload *payload =
+			key->payload.data[0];
+
+		seq_printf(m, ": %u [%u]", key->datalen, payload->type);
+	}
+}
+
+static void fscrypt_provisioning_key_destroy(struct key *key)
+{
+	kzfree(key->payload.data[0]);
+}
+
+static struct key_type key_type_fscrypt_provisioning = {
+	.name			= "fscrypt-provisioning",
+	.preparse		= fscrypt_provisioning_key_preparse,
+	.free_preparse		= fscrypt_provisioning_key_free_preparse,
+	.instantiate		= generic_key_instantiate,
+	.describe		= fscrypt_provisioning_key_describe,
+	.destroy		= fscrypt_provisioning_key_destroy,
+};
+
+/*
+ * Retrieve the raw key from the Linux keyring key specified by 'key_id', and
+ * store it into 'secret'.
+ *
+ * The key must be of type "fscrypt-provisioning" and must have the field
+ * fscrypt_provisioning_key_payload::type set to 'type', indicating that it's
+ * only usable with fscrypt with the particular KDF version identified by
+ * 'type'.  We don't use the "logon" key type because there's no way to
+ * completely restrict the use of such keys; they can be used by any kernel API
+ * that accepts "logon" keys and doesn't require a specific service prefix.
+ *
+ * The ability to specify the key via Linux keyring key is intended for cases
+ * where userspace needs to re-add keys after the filesystem is unmounted and
+ * re-mounted.  Most users should just provide the raw key directly instead.
+ */
+static int get_keyring_key(u32 key_id, u32 type,
+			   struct fscrypt_master_key_secret *secret)
+{
+	key_ref_t ref;
+	struct key *key;
+	const struct fscrypt_provisioning_key_payload *payload;
+	int err;
+
+	ref = lookup_user_key(key_id, 0, KEY_NEED_SEARCH);
+	if (IS_ERR(ref))
+		return PTR_ERR(ref);
+	key = key_ref_to_ptr(ref);
+
+	if (key->type != &key_type_fscrypt_provisioning)
+		goto bad_key;
+	payload = key->payload.data[0];
+
+	/* Don't allow fscrypt v1 keys to be used as v2 keys and vice versa. */
+	if (payload->type != type)
+		goto bad_key;
+
+	secret->size = key->datalen - sizeof(*payload);
+	memcpy(secret->raw, payload->raw, secret->size);
+	err = 0;
+	goto out_put;
+
+bad_key:
+	err = -EKEYREJECTED;
+out_put:
+	key_ref_put(ref);
+	return err;
+}
+
 /*
  * Add a master encryption key to the filesystem, causing all files which were
  * encrypted with it to appear "unlocked" (decrypted) when accessed.
@@ -501,61 +637,46 @@ int fscrypt_ioctl_add_key(struct file *filp, void __user *_uarg)
 	if (!valid_key_spec(&arg.key_spec))
 		return -EINVAL;
 
-	if (arg.raw_size < FSCRYPT_MIN_KEY_SIZE ||
-	    arg.raw_size > FSCRYPT_MAX_KEY_SIZE)
-		return -EINVAL;
-
 	if (memchr_inv(arg.__reserved, 0, sizeof(arg.__reserved)))
 		return -EINVAL;
 
-	memset(&secret, 0, sizeof(secret));
-	secret.size = arg.raw_size;
-	err = -EFAULT;
-	if (copy_from_user(secret.raw, uarg->raw, secret.size))
-		goto out_wipe_secret;
-
-	switch (arg.key_spec.type) {
-	case FSCRYPT_KEY_SPEC_TYPE_DESCRIPTOR:
-		/*
-		 * Only root can add keys that are identified by an arbitrary
-		 * descriptor rather than by a cryptographic hash --- since
-		 * otherwise a malicious user could add the wrong key.
-		 */
-		err = -EACCES;
-		if (!capable(CAP_SYS_ADMIN))
-			goto out_wipe_secret;
-		break;
-	case FSCRYPT_KEY_SPEC_TYPE_IDENTIFIER:
-		err = fscrypt_init_hkdf(&secret.hkdf, secret.raw, secret.size);
-		if (err)
-			goto out_wipe_secret;
-
-		/*
-		 * Now that the HKDF context is initialized, the raw key is no
-		 * longer needed.
-		 */
-		memzero_explicit(secret.raw, secret.size);
+	/*
+	 * Only root can add keys that are identified by an arbitrary descriptor
+	 * rather than by a cryptographic hash --- since otherwise a malicious
+	 * user could add the wrong key.
+	 */
+	if (arg.key_spec.type == FSCRYPT_KEY_SPEC_TYPE_DESCRIPTOR &&
+	    !capable(CAP_SYS_ADMIN))
+		return -EACCES;
 
-		/* Calculate the key identifier and return it to userspace. */
-		err = fscrypt_hkdf_expand(&secret.hkdf,
-					  HKDF_CONTEXT_KEY_IDENTIFIER,
-					  NULL, 0, arg.key_spec.u.identifier,
-					  FSCRYPT_KEY_IDENTIFIER_SIZE);
+	memset(&secret, 0, sizeof(secret));
+	if (arg.key_id) {
+		if (arg.raw_size != 0)
+			return -EINVAL;
+		err = get_keyring_key(arg.key_id, arg.key_spec.type, &secret);
 		if (err)
 			goto out_wipe_secret;
+	} else {
+		if (arg.raw_size < FSCRYPT_MIN_KEY_SIZE ||
+		    arg.raw_size > FSCRYPT_MAX_KEY_SIZE)
+			return -EINVAL;
+		secret.size = arg.raw_size;
 		err = -EFAULT;
-		if (copy_to_user(uarg->key_spec.u.identifier,
-				 arg.key_spec.u.identifier,
-				 FSCRYPT_KEY_IDENTIFIER_SIZE))
+		if (copy_from_user(secret.raw, uarg->raw, secret.size))
 			goto out_wipe_secret;
-		break;
-	default:
-		WARN_ON(1);
-		err = -EINVAL;
-		goto out_wipe_secret;
 	}
 
 	err = add_master_key(sb, &secret, &arg.key_spec);
+	if (err)
+		goto out_wipe_secret;
+
+	/* Return the key identifier to userspace, if applicable */
+	err = -EFAULT;
+	if (arg.key_spec.type == FSCRYPT_KEY_SPEC_TYPE_IDENTIFIER &&
+	    copy_to_user(uarg->key_spec.u.identifier, arg.key_spec.u.identifier,
+			 FSCRYPT_KEY_IDENTIFIER_SIZE))
+		goto out_wipe_secret;
+	err = 0;
 out_wipe_secret:
 	wipe_master_key_secret(&secret);
 	return err;
@@ -563,6 +684,29 @@ out_wipe_secret:
 EXPORT_SYMBOL_GPL(fscrypt_ioctl_add_key);
 
 /*
+ * Add the key for '-o test_dummy_encryption' to the filesystem keyring.
+ *
+ * Use a per-boot random key to prevent people from misusing this option.
+ */
+int fscrypt_add_test_dummy_key(struct super_block *sb,
+			       struct fscrypt_key_specifier *key_spec)
+{
+	static u8 test_key[FSCRYPT_MAX_KEY_SIZE];
+	struct fscrypt_master_key_secret secret;
+	int err;
+
+	get_random_once(test_key, FSCRYPT_MAX_KEY_SIZE);
+
+	memset(&secret, 0, sizeof(secret));
+	secret.size = FSCRYPT_MAX_KEY_SIZE;
+	memcpy(secret.raw, test_key, FSCRYPT_MAX_KEY_SIZE);
+
+	err = add_master_key(sb, &secret, key_spec);
+	wipe_master_key_secret(&secret);
+	return err;
+}
+
+/*
  * Verify that the current user has added a master key with the given identifier
  * (returns -ENOKEY if not).  This is needed to prevent a user from encrypting
  * their files using some other user's key which they don't actually know.
@@ -664,9 +808,6 @@ static int check_for_busy_inodes(struct super_block *sb,
 	struct list_head *pos;
 	size_t busy_count = 0;
 	unsigned long ino;
-	struct dentry *dentry;
-	char _path[256];
-	char *path = NULL;
 
 	spin_lock(&mk->mk_decrypted_inodes_lock);
 
@@ -685,22 +826,14 @@ static int check_for_busy_inodes(struct super_block *sb,
 					 struct fscrypt_info,
 					 ci_master_key_link)->ci_inode;
 		ino = inode->i_ino;
-		dentry = d_find_alias(inode);
 	}
 	spin_unlock(&mk->mk_decrypted_inodes_lock);
 
-	if (dentry) {
-		path = dentry_path(dentry, _path, sizeof(_path));
-		dput(dentry);
-	}
-	if (IS_ERR_OR_NULL(path))
-		path = "(unknown)";
-
 	fscrypt_warn(NULL,
-		     "%s: %zu inode(s) still busy after removing key with %s %*phN, including ino %lu (%s)",
+		     "%s: %zu inode(s) still busy after removing key with %s %*phN, including ino %lu",
 		     sb->s_id, busy_count, master_key_spec_type(&mk->mk_spec),
 		     master_key_spec_len(&mk->mk_spec), (u8 *)&mk->mk_spec.u,
-		     ino, path);
+		     ino);
 	return -EBUSY;
 }
 
@@ -976,8 +1109,14 @@ int __init fscrypt_init_keyring(void)
 	if (err)
 		goto err_unregister_fscrypt;
 
+	err = register_key_type(&key_type_fscrypt_provisioning);
+	if (err)
+		goto err_unregister_fscrypt_user;
+
 	return 0;
 
+err_unregister_fscrypt_user:
+	unregister_key_type(&key_type_fscrypt_user);
 err_unregister_fscrypt:
 	unregister_key_type(&key_type_fscrypt);
 	return err;
diff --git a/fs/crypto/keysetup.c b/fs/crypto/keysetup.c
index d71c2d6dd162..1129adfa097d 100644
--- a/fs/crypto/keysetup.c
+++ b/fs/crypto/keysetup.c
@@ -8,16 +8,12 @@
  * Heavily modified since then.
  */
 
-#include <crypto/aes.h>
-#include <crypto/sha.h>
 #include <crypto/skcipher.h>
 #include <linux/key.h>
 
 #include "fscrypt_private.h"
 
-static struct crypto_shash *essiv_hash_tfm;
-
-static struct fscrypt_mode available_modes[] = {
+struct fscrypt_mode fscrypt_modes[] = {
 	[FSCRYPT_MODE_AES_256_XTS] = {
 		.friendly_name = "AES-256-XTS",
 		.cipher_str = "xts(aes)",
@@ -31,11 +27,10 @@ static struct fscrypt_mode available_modes[] = {
 		.ivsize = 16,
 	},
 	[FSCRYPT_MODE_AES_128_CBC] = {
-		.friendly_name = "AES-128-CBC",
-		.cipher_str = "cbc(aes)",
+		.friendly_name = "AES-128-CBC-ESSIV",
+		.cipher_str = "essiv(cbc(aes),sha256)",
 		.keysize = 16,
 		.ivsize = 16,
-		.needs_essiv = true,
 	},
 	[FSCRYPT_MODE_AES_128_CTS] = {
 		.friendly_name = "AES-128-CTS-CBC",
@@ -51,15 +46,17 @@ static struct fscrypt_mode available_modes[] = {
 	},
 };
 
+static DEFINE_MUTEX(fscrypt_mode_key_setup_mutex);
+
 static struct fscrypt_mode *
 select_encryption_mode(const union fscrypt_policy *policy,
 		       const struct inode *inode)
 {
 	if (S_ISREG(inode->i_mode))
-		return &available_modes[fscrypt_policy_contents_mode(policy)];
+		return &fscrypt_modes[fscrypt_policy_contents_mode(policy)];
 
 	if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
-		return &available_modes[fscrypt_policy_fnames_mode(policy)];
+		return &fscrypt_modes[fscrypt_policy_fnames_mode(policy)];
 
 	WARN_ONCE(1, "fscrypt: filesystem tried to load encryption info for inode %lu, which is not encryptable (file type %d)\n",
 		  inode->i_ino, (inode->i_mode & S_IFMT));
@@ -86,18 +83,19 @@ struct crypto_skcipher *fscrypt_allocate_skcipher(struct fscrypt_mode *mode,
 			    mode->cipher_str, PTR_ERR(tfm));
 		return tfm;
 	}
-	if (unlikely(!mode->logged_impl_name)) {
+	if (!xchg(&mode->logged_impl_name, 1)) {
 		/*
 		 * fscrypt performance can vary greatly depending on which
 		 * crypto algorithm implementation is used.  Help people debug
 		 * performance problems by logging the ->cra_driver_name the
-		 * first time a mode is used.  Note that multiple threads can
-		 * race here, but it doesn't really matter.
+		 * first time a mode is used.
 		 */
-		mode->logged_impl_name = true;
 		pr_info("fscrypt: %s using implementation \"%s\"\n",
-			mode->friendly_name,
-			crypto_skcipher_alg(tfm)->base.cra_driver_name);
+			mode->friendly_name, crypto_skcipher_driver_name(tfm));
+	}
+	if (WARN_ON(crypto_skcipher_ivsize(tfm) != mode->ivsize)) {
+		err = -EINVAL;
+		goto err_free_tfm;
 	}
 	crypto_skcipher_set_flags(tfm, CRYPTO_TFM_REQ_FORBID_WEAK_KEYS);
 	err = crypto_skcipher_setkey(tfm, raw_key, mode->keysize);
@@ -111,174 +109,187 @@ err_free_tfm:
 	return ERR_PTR(err);
 }
 
-static int derive_essiv_salt(const u8 *key, int keysize, u8 *salt)
+/* Given a per-file encryption key, set up the file's crypto transform object */
+int fscrypt_set_per_file_enc_key(struct fscrypt_info *ci, const u8 *raw_key)
 {
-	struct crypto_shash *tfm = READ_ONCE(essiv_hash_tfm);
-
-	/* init hash transform on demand */
-	if (unlikely(!tfm)) {
-		struct crypto_shash *prev_tfm;
-
-		tfm = crypto_alloc_shash("sha256", 0, 0);
-		if (IS_ERR(tfm)) {
-			if (PTR_ERR(tfm) == -ENOENT) {
-				fscrypt_warn(NULL,
-					     "Missing crypto API support for SHA-256");
-				return -ENOPKG;
-			}
-			fscrypt_err(NULL,
-				    "Error allocating SHA-256 transform: %ld",
-				    PTR_ERR(tfm));
-			return PTR_ERR(tfm);
-		}
-		prev_tfm = cmpxchg(&essiv_hash_tfm, NULL, tfm);
-		if (prev_tfm) {
-			crypto_free_shash(tfm);
-			tfm = prev_tfm;
-		}
-	}
+	struct crypto_skcipher *tfm;
 
-	{
-		SHASH_DESC_ON_STACK(desc, tfm);
-		desc->tfm = tfm;
+	tfm = fscrypt_allocate_skcipher(ci->ci_mode, raw_key, ci->ci_inode);
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
 
-		return crypto_shash_digest(desc, key, keysize, salt);
-	}
+	ci->ci_ctfm = tfm;
+	ci->ci_owns_key = true;
+	return 0;
 }
 
-static int init_essiv_generator(struct fscrypt_info *ci, const u8 *raw_key,
-				int keysize)
+static int setup_per_mode_enc_key(struct fscrypt_info *ci,
+				  struct fscrypt_master_key *mk,
+				  struct crypto_skcipher **tfms,
+				  u8 hkdf_context, bool include_fs_uuid)
 {
+	const struct inode *inode = ci->ci_inode;
+	const struct super_block *sb = inode->i_sb;
+	struct fscrypt_mode *mode = ci->ci_mode;
+	const u8 mode_num = mode - fscrypt_modes;
+	struct crypto_skcipher *tfm;
+	u8 mode_key[FSCRYPT_MAX_KEY_SIZE];
+	u8 hkdf_info[sizeof(mode_num) + sizeof(sb->s_uuid)];
+	unsigned int hkdf_infolen = 0;
 	int err;
-	struct crypto_cipher *essiv_tfm;
-	u8 salt[SHA256_DIGEST_SIZE];
 
-	if (WARN_ON(ci->ci_mode->ivsize != AES_BLOCK_SIZE))
+	if (WARN_ON(mode_num > __FSCRYPT_MODE_MAX))
 		return -EINVAL;
 
-	essiv_tfm = crypto_alloc_cipher("aes", 0, 0);
-	if (IS_ERR(essiv_tfm))
-		return PTR_ERR(essiv_tfm);
+	/* pairs with smp_store_release() below */
+	tfm = READ_ONCE(tfms[mode_num]);
+	if (likely(tfm != NULL)) {
+		ci->ci_ctfm = tfm;
+		return 0;
+	}
 
-	ci->ci_essiv_tfm = essiv_tfm;
+	mutex_lock(&fscrypt_mode_key_setup_mutex);
 
-	err = derive_essiv_salt(raw_key, keysize, salt);
-	if (err)
-		goto out;
+	if (tfms[mode_num])
+		goto done_unlock;
 
-	/*
-	 * Using SHA256 to derive the salt/key will result in AES-256 being
-	 * used for IV generation. File contents encryption will still use the
-	 * configured keysize (AES-128) nevertheless.
-	 */
-	err = crypto_cipher_setkey(essiv_tfm, salt, sizeof(salt));
+	BUILD_BUG_ON(sizeof(mode_num) != 1);
+	BUILD_BUG_ON(sizeof(sb->s_uuid) != 16);
+	BUILD_BUG_ON(sizeof(hkdf_info) != 17);
+	hkdf_info[hkdf_infolen++] = mode_num;
+	if (include_fs_uuid) {
+		memcpy(&hkdf_info[hkdf_infolen], &sb->s_uuid,
+		       sizeof(sb->s_uuid));
+		hkdf_infolen += sizeof(sb->s_uuid);
+	}
+	err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf,
+				  hkdf_context, hkdf_info, hkdf_infolen,
+				  mode_key, mode->keysize);
 	if (err)
-		goto out;
-
-out:
-	memzero_explicit(salt, sizeof(salt));
+		goto out_unlock;
+	tfm = fscrypt_allocate_skcipher(mode, mode_key, inode);
+	memzero_explicit(mode_key, mode->keysize);
+	if (IS_ERR(tfm)) {
+		err = PTR_ERR(tfm);
+		goto out_unlock;
+	}
+	/* pairs with READ_ONCE() above */
+	smp_store_release(&tfms[mode_num], tfm);
+done_unlock:
+	ci->ci_ctfm = tfm;
+	err = 0;
+out_unlock:
+	mutex_unlock(&fscrypt_mode_key_setup_mutex);
 	return err;
 }
 
-/* Given the per-file key, set up the file's crypto transform object(s) */
-int fscrypt_set_derived_key(struct fscrypt_info *ci, const u8 *derived_key)
+int fscrypt_derive_dirhash_key(struct fscrypt_info *ci,
+			       const struct fscrypt_master_key *mk)
 {
-	struct fscrypt_mode *mode = ci->ci_mode;
-	struct crypto_skcipher *ctfm;
 	int err;
 
-	ctfm = fscrypt_allocate_skcipher(mode, derived_key, ci->ci_inode);
-	if (IS_ERR(ctfm))
-		return PTR_ERR(ctfm);
-
-	ci->ci_ctfm = ctfm;
-
-	if (mode->needs_essiv) {
-		err = init_essiv_generator(ci, derived_key, mode->keysize);
-		if (err) {
-			fscrypt_warn(ci->ci_inode,
-				     "Error initializing ESSIV generator: %d",
-				     err);
-			return err;
-		}
-	}
+	err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf, HKDF_CONTEXT_DIRHASH_KEY,
+				  ci->ci_nonce, FS_KEY_DERIVATION_NONCE_SIZE,
+				  (u8 *)&ci->ci_dirhash_key,
+				  sizeof(ci->ci_dirhash_key));
+	if (err)
+		return err;
+	ci->ci_dirhash_key_initialized = true;
 	return 0;
 }
 
-static int setup_per_mode_key(struct fscrypt_info *ci,
-			      struct fscrypt_master_key *mk)
+static int fscrypt_setup_iv_ino_lblk_32_key(struct fscrypt_info *ci,
+					    struct fscrypt_master_key *mk)
 {
-	struct fscrypt_mode *mode = ci->ci_mode;
-	u8 mode_num = mode - available_modes;
-	struct crypto_skcipher *tfm, *prev_tfm;
-	u8 mode_key[FSCRYPT_MAX_KEY_SIZE];
 	int err;
 
-	if (WARN_ON(mode_num >= ARRAY_SIZE(mk->mk_mode_keys)))
-		return -EINVAL;
-
-	/* pairs with cmpxchg() below */
-	tfm = READ_ONCE(mk->mk_mode_keys[mode_num]);
-	if (likely(tfm != NULL))
-		goto done;
-
-	BUILD_BUG_ON(sizeof(mode_num) != 1);
-	err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf,
-				  HKDF_CONTEXT_PER_MODE_KEY,
-				  &mode_num, sizeof(mode_num),
-				  mode_key, mode->keysize);
+	err = setup_per_mode_enc_key(ci, mk, mk->mk_iv_ino_lblk_32_keys,
+				     HKDF_CONTEXT_IV_INO_LBLK_32_KEY, true);
 	if (err)
 		return err;
-	tfm = fscrypt_allocate_skcipher(mode, mode_key, ci->ci_inode);
-	memzero_explicit(mode_key, mode->keysize);
-	if (IS_ERR(tfm))
-		return PTR_ERR(tfm);
 
-	/* pairs with READ_ONCE() above */
-	prev_tfm = cmpxchg(&mk->mk_mode_keys[mode_num], NULL, tfm);
-	if (prev_tfm != NULL) {
-		crypto_free_skcipher(tfm);
-		tfm = prev_tfm;
+	/* pairs with smp_store_release() below */
+	if (!smp_load_acquire(&mk->mk_ino_hash_key_initialized)) {
+
+		mutex_lock(&fscrypt_mode_key_setup_mutex);
+
+		if (mk->mk_ino_hash_key_initialized)
+			goto unlock;
+
+		err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf,
+					  HKDF_CONTEXT_INODE_HASH_KEY, NULL, 0,
+					  (u8 *)&mk->mk_ino_hash_key,
+					  sizeof(mk->mk_ino_hash_key));
+		if (err)
+			goto unlock;
+		/* pairs with smp_load_acquire() above */
+		smp_store_release(&mk->mk_ino_hash_key_initialized, true);
+unlock:
+		mutex_unlock(&fscrypt_mode_key_setup_mutex);
+		if (err)
+			return err;
 	}
-done:
-	ci->ci_ctfm = tfm;
+
+	ci->ci_hashed_ino = (u32)siphash_1u64(ci->ci_inode->i_ino,
+					      &mk->mk_ino_hash_key);
 	return 0;
 }
 
 static int fscrypt_setup_v2_file_key(struct fscrypt_info *ci,
 				     struct fscrypt_master_key *mk)
 {
-	u8 derived_key[FSCRYPT_MAX_KEY_SIZE];
 	int err;
 
 	if (ci->ci_policy.v2.flags & FSCRYPT_POLICY_FLAG_DIRECT_KEY) {
 		/*
-		 * DIRECT_KEY: instead of deriving per-file keys, the per-file
-		 * nonce will be included in all the IVs.  But unlike v1
-		 * policies, for v2 policies in this case we don't encrypt with
-		 * the master key directly but rather derive a per-mode key.
-		 * This ensures that the master key is consistently used only
-		 * for HKDF, avoiding key reuse issues.
+		 * DIRECT_KEY: instead of deriving per-file encryption keys, the
+		 * per-file nonce will be included in all the IVs.  But unlike
+		 * v1 policies, for v2 policies in this case we don't encrypt
+		 * with the master key directly but rather derive a per-mode
+		 * encryption key.  This ensures that the master key is
+		 * consistently used only for HKDF, avoiding key reuse issues.
 		 */
-		if (!fscrypt_mode_supports_direct_key(ci->ci_mode)) {
-			fscrypt_warn(ci->ci_inode,
-				     "Direct key flag not allowed with %s",
-				     ci->ci_mode->friendly_name);
-			return -EINVAL;
-		}
-		return setup_per_mode_key(ci, mk);
-	}
+		err = setup_per_mode_enc_key(ci, mk, mk->mk_direct_keys,
+					     HKDF_CONTEXT_DIRECT_KEY, false);
+	} else if (ci->ci_policy.v2.flags &
+		   FSCRYPT_POLICY_FLAG_IV_INO_LBLK_64) {
+		/*
+		 * IV_INO_LBLK_64: encryption keys are derived from (master_key,
+		 * mode_num, filesystem_uuid), and inode number is included in
+		 * the IVs.  This format is optimized for use with inline
+		 * encryption hardware compliant with the UFS standard.
+		 */
+		err = setup_per_mode_enc_key(ci, mk, mk->mk_iv_ino_lblk_64_keys,
+					     HKDF_CONTEXT_IV_INO_LBLK_64_KEY,
+					     true);
+	} else if (ci->ci_policy.v2.flags &
+		   FSCRYPT_POLICY_FLAG_IV_INO_LBLK_32) {
+		err = fscrypt_setup_iv_ino_lblk_32_key(ci, mk);
+	} else {
+		u8 derived_key[FSCRYPT_MAX_KEY_SIZE];
+
+		err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf,
+					  HKDF_CONTEXT_PER_FILE_ENC_KEY,
+					  ci->ci_nonce,
+					  FS_KEY_DERIVATION_NONCE_SIZE,
+					  derived_key, ci->ci_mode->keysize);
+		if (err)
+			return err;
 
-	err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf,
-				  HKDF_CONTEXT_PER_FILE_KEY,
-				  ci->ci_nonce, FS_KEY_DERIVATION_NONCE_SIZE,
-				  derived_key, ci->ci_mode->keysize);
+		err = fscrypt_set_per_file_enc_key(ci, derived_key);
+		memzero_explicit(derived_key, ci->ci_mode->keysize);
+	}
 	if (err)
 		return err;
 
-	err = fscrypt_set_derived_key(ci, derived_key);
-	memzero_explicit(derived_key, ci->ci_mode->keysize);
-	return err;
+	/* Derive a secret dirhash key for directories that need it. */
+	if (S_ISDIR(ci->ci_inode->i_mode) && IS_CASEFOLDED(ci->ci_inode)) {
+		err = fscrypt_derive_dirhash_key(ci, mk);
+		if (err)
+			return err;
+	}
+
+	return 0;
 }
 
 /*
@@ -388,13 +399,10 @@ static void put_crypt_info(struct fscrypt_info *ci)
 	if (!ci)
 		return;
 
-	if (ci->ci_direct_key) {
+	if (ci->ci_direct_key)
 		fscrypt_put_direct_key(ci->ci_direct_key);
-	} else if ((ci->ci_ctfm != NULL || ci->ci_essiv_tfm != NULL) &&
-		   !fscrypt_is_direct_key_policy(&ci->ci_policy)) {
+	else if (ci->ci_owns_key)
 		crypto_free_skcipher(ci->ci_ctfm);
-		crypto_free_cipher(ci->ci_essiv_tfm);
-	}
 
 	key = ci->ci_master_key;
 	if (key) {
@@ -415,6 +423,7 @@ static void put_crypt_info(struct fscrypt_info *ci)
 			key_invalidate(key);
 		key_put(key);
 	}
+	memzero_explicit(ci, sizeof(*ci));
 	kmem_cache_free(fscrypt_info_cachep, ci);
 }
 
@@ -435,21 +444,18 @@ int fscrypt_get_encryption_info(struct inode *inode)
 
 	res = inode->i_sb->s_cop->get_context(inode, &ctx, sizeof(ctx));
 	if (res < 0) {
-		if (!fscrypt_dummy_context_enabled(inode) ||
-		    IS_ENCRYPTED(inode)) {
+		const union fscrypt_context *dummy_ctx =
+			fscrypt_get_dummy_context(inode->i_sb);
+
+		if (IS_ENCRYPTED(inode) || !dummy_ctx) {
 			fscrypt_warn(inode,
 				     "Error %d getting encryption context",
 				     res);
 			return res;
 		}
 		/* Fake up a context for an unencrypted directory */
-		memset(&ctx, 0, sizeof(ctx));
-		ctx.version = FSCRYPT_CONTEXT_V1;
-		ctx.v1.contents_encryption_mode = FSCRYPT_MODE_AES_256_XTS;
-		ctx.v1.filenames_encryption_mode = FSCRYPT_MODE_AES_256_CTS;
-		memset(ctx.v1.master_key_descriptor, 0x42,
-		       FSCRYPT_KEY_DESCRIPTOR_SIZE);
-		res = sizeof(ctx.v1);
+		res = fscrypt_context_size(dummy_ctx);
+		memcpy(&ctx, dummy_ctx, res);
 	}
 
 	crypt_info = kmem_cache_zalloc(fscrypt_info_cachep, GFP_NOFS);
@@ -465,20 +471,8 @@ int fscrypt_get_encryption_info(struct inode *inode)
 		goto out;
 	}
 
-	switch (ctx.version) {
-	case FSCRYPT_CONTEXT_V1:
-		memcpy(crypt_info->ci_nonce, ctx.v1.nonce,
-		       FS_KEY_DERIVATION_NONCE_SIZE);
-		break;
-	case FSCRYPT_CONTEXT_V2:
-		memcpy(crypt_info->ci_nonce, ctx.v2.nonce,
-		       FS_KEY_DERIVATION_NONCE_SIZE);
-		break;
-	default:
-		WARN_ON(1);
-		res = -EINVAL;
-		goto out;
-	}
+	memcpy(crypt_info->ci_nonce, fscrypt_context_nonce(&ctx),
+	       FS_KEY_DERIVATION_NONCE_SIZE);
 
 	if (!fscrypt_supported_policy(&crypt_info->ci_policy, inode)) {
 		res = -EINVAL;
@@ -527,7 +521,8 @@ out:
 EXPORT_SYMBOL(fscrypt_get_encryption_info);
 
 /**
- * fscrypt_put_encryption_info - free most of an inode's fscrypt data
+ * fscrypt_put_encryption_info() - free most of an inode's fscrypt data
+ * @inode: an inode being evicted
  *
  * Free the inode's fscrypt_info.  Filesystems must call this when the inode is
  * being evicted.  An RCU grace period need not have elapsed yet.
@@ -540,7 +535,8 @@ void fscrypt_put_encryption_info(struct inode *inode)
 EXPORT_SYMBOL(fscrypt_put_encryption_info);
 
 /**
- * fscrypt_free_inode - free an inode's fscrypt data requiring RCU delay
+ * fscrypt_free_inode() - free an inode's fscrypt data requiring RCU delay
+ * @inode: an inode being freed
  *
  * Free the inode's cached decrypted symlink target, if any.  Filesystems must
  * call this after an RCU grace period, just before they free the inode.
@@ -555,7 +551,8 @@ void fscrypt_free_inode(struct inode *inode)
 EXPORT_SYMBOL(fscrypt_free_inode);
 
 /**
- * fscrypt_drop_inode - check whether the inode's master key has been removed
+ * fscrypt_drop_inode() - check whether the inode's master key has been removed
+ * @inode: an inode being considered for eviction
  *
  * Filesystems supporting fscrypt must call this from their ->drop_inode()
  * method so that encrypted inodes are evicted as soon as they're no longer in
@@ -579,6 +576,15 @@ int fscrypt_drop_inode(struct inode *inode)
 	mk = ci->ci_master_key->payload.data[0];
 
 	/*
+	 * With proper, non-racy use of FS_IOC_REMOVE_ENCRYPTION_KEY, all inodes
+	 * protected by the key were cleaned by sync_filesystem().  But if
+	 * userspace is still using the files, inodes can be dirtied between
+	 * then and now.  We mustn't lose any writes, so skip dirty inodes here.
+	 */
+	if (inode->i_state & I_DIRTY_ALL)
+		return 0;
+
+	/*
 	 * Note: since we aren't holding ->mk_secret_sem, the result here can
 	 * immediately become outdated.  But there's no correctness problem with
 	 * unnecessarily evicting.  Nor is there a correctness problem with not
diff --git a/fs/crypto/keysetup_v1.c b/fs/crypto/keysetup_v1.c
index ad1a36c370c3..801b48c0cd7f 100644
--- a/fs/crypto/keysetup_v1.c
+++ b/fs/crypto/keysetup_v1.c
@@ -9,7 +9,7 @@
  * This file implements compatibility functions for the original encryption
  * policy version ("v1"), including:
  *
- * - Deriving per-file keys using the AES-128-ECB based KDF
+ * - Deriving per-file encryption keys using the AES-128-ECB based KDF
  *   (rather than the new method of using HKDF-SHA512)
  *
  * - Retrieving fscrypt master keys from process-subscribed keyrings
@@ -253,27 +253,8 @@ err_free_dk:
 static int setup_v1_file_key_direct(struct fscrypt_info *ci,
 				    const u8 *raw_master_key)
 {
-	const struct fscrypt_mode *mode = ci->ci_mode;
 	struct fscrypt_direct_key *dk;
 
-	if (!fscrypt_mode_supports_direct_key(mode)) {
-		fscrypt_warn(ci->ci_inode,
-			     "Direct key mode not allowed with %s",
-			     mode->friendly_name);
-		return -EINVAL;
-	}
-
-	if (ci->ci_policy.v1.contents_encryption_mode !=
-	    ci->ci_policy.v1.filenames_encryption_mode) {
-		fscrypt_warn(ci->ci_inode,
-			     "Direct key mode not allowed with different contents and filenames modes");
-		return -EINVAL;
-	}
-
-	/* ESSIV implies 16-byte IVs which implies !DIRECT_KEY */
-	if (WARN_ON(mode->needs_essiv))
-		return -EINVAL;
-
 	dk = fscrypt_get_direct_key(ci, raw_master_key);
 	if (IS_ERR(dk))
 		return PTR_ERR(dk);
@@ -302,7 +283,7 @@ static int setup_v1_file_key_derived(struct fscrypt_info *ci,
 	if (err)
 		goto out;
 
-	err = fscrypt_set_derived_key(ci, derived_key);
+	err = fscrypt_set_per_file_enc_key(ci, derived_key);
 out:
 	kzfree(derived_key);
 	return err;
diff --git a/fs/crypto/policy.c b/fs/crypto/policy.c
index 4072ba644595..d23ff162c78b 100644
--- a/fs/crypto/policy.c
+++ b/fs/crypto/policy.c
@@ -11,12 +11,15 @@
  */
 
 #include <linux/random.h>
+#include <linux/seq_file.h>
 #include <linux/string.h>
 #include <linux/mount.h>
 #include "fscrypt_private.h"
 
 /**
- * fscrypt_policies_equal - check whether two encryption policies are the same
+ * fscrypt_policies_equal() - check whether two encryption policies are the same
+ * @policy1: the first policy
+ * @policy2: the second policy
  *
  * Return: %true if equal, else %false
  */
@@ -29,13 +32,175 @@ bool fscrypt_policies_equal(const union fscrypt_policy *policy1,
 	return !memcmp(policy1, policy2, fscrypt_policy_size(policy1));
 }
 
+static bool fscrypt_valid_enc_modes(u32 contents_mode, u32 filenames_mode)
+{
+	if (contents_mode == FSCRYPT_MODE_AES_256_XTS &&
+	    filenames_mode == FSCRYPT_MODE_AES_256_CTS)
+		return true;
+
+	if (contents_mode == FSCRYPT_MODE_AES_128_CBC &&
+	    filenames_mode == FSCRYPT_MODE_AES_128_CTS)
+		return true;
+
+	if (contents_mode == FSCRYPT_MODE_ADIANTUM &&
+	    filenames_mode == FSCRYPT_MODE_ADIANTUM)
+		return true;
+
+	return false;
+}
+
+static bool supported_direct_key_modes(const struct inode *inode,
+				       u32 contents_mode, u32 filenames_mode)
+{
+	const struct fscrypt_mode *mode;
+
+	if (contents_mode != filenames_mode) {
+		fscrypt_warn(inode,
+			     "Direct key flag not allowed with different contents and filenames modes");
+		return false;
+	}
+	mode = &fscrypt_modes[contents_mode];
+
+	if (mode->ivsize < offsetofend(union fscrypt_iv, nonce)) {
+		fscrypt_warn(inode, "Direct key flag not allowed with %s",
+			     mode->friendly_name);
+		return false;
+	}
+	return true;
+}
+
+static bool supported_iv_ino_lblk_policy(const struct fscrypt_policy_v2 *policy,
+					 const struct inode *inode,
+					 const char *type,
+					 int max_ino_bits, int max_lblk_bits)
+{
+	struct super_block *sb = inode->i_sb;
+	int ino_bits = 64, lblk_bits = 64;
+
+	/*
+	 * It's unsafe to include inode numbers in the IVs if the filesystem can
+	 * potentially renumber inodes, e.g. via filesystem shrinking.
+	 */
+	if (!sb->s_cop->has_stable_inodes ||
+	    !sb->s_cop->has_stable_inodes(sb)) {
+		fscrypt_warn(inode,
+			     "Can't use %s policy on filesystem '%s' because it doesn't have stable inode numbers",
+			     type, sb->s_id);
+		return false;
+	}
+	if (sb->s_cop->get_ino_and_lblk_bits)
+		sb->s_cop->get_ino_and_lblk_bits(sb, &ino_bits, &lblk_bits);
+	if (ino_bits > max_ino_bits) {
+		fscrypt_warn(inode,
+			     "Can't use %s policy on filesystem '%s' because its inode numbers are too long",
+			     type, sb->s_id);
+		return false;
+	}
+	if (lblk_bits > max_lblk_bits) {
+		fscrypt_warn(inode,
+			     "Can't use %s policy on filesystem '%s' because its block numbers are too long",
+			     type, sb->s_id);
+		return false;
+	}
+	return true;
+}
+
+static bool fscrypt_supported_v1_policy(const struct fscrypt_policy_v1 *policy,
+					const struct inode *inode)
+{
+	if (!fscrypt_valid_enc_modes(policy->contents_encryption_mode,
+				     policy->filenames_encryption_mode)) {
+		fscrypt_warn(inode,
+			     "Unsupported encryption modes (contents %d, filenames %d)",
+			     policy->contents_encryption_mode,
+			     policy->filenames_encryption_mode);
+		return false;
+	}
+
+	if (policy->flags & ~(FSCRYPT_POLICY_FLAGS_PAD_MASK |
+			      FSCRYPT_POLICY_FLAG_DIRECT_KEY)) {
+		fscrypt_warn(inode, "Unsupported encryption flags (0x%02x)",
+			     policy->flags);
+		return false;
+	}
+
+	if ((policy->flags & FSCRYPT_POLICY_FLAG_DIRECT_KEY) &&
+	    !supported_direct_key_modes(inode, policy->contents_encryption_mode,
+					policy->filenames_encryption_mode))
+		return false;
+
+	if (IS_CASEFOLDED(inode)) {
+		/* With v1, there's no way to derive dirhash keys. */
+		fscrypt_warn(inode,
+			     "v1 policies can't be used on casefolded directories");
+		return false;
+	}
+
+	return true;
+}
+
+static bool fscrypt_supported_v2_policy(const struct fscrypt_policy_v2 *policy,
+					const struct inode *inode)
+{
+	int count = 0;
+
+	if (!fscrypt_valid_enc_modes(policy->contents_encryption_mode,
+				     policy->filenames_encryption_mode)) {
+		fscrypt_warn(inode,
+			     "Unsupported encryption modes (contents %d, filenames %d)",
+			     policy->contents_encryption_mode,
+			     policy->filenames_encryption_mode);
+		return false;
+	}
+
+	if (policy->flags & ~FSCRYPT_POLICY_FLAGS_VALID) {
+		fscrypt_warn(inode, "Unsupported encryption flags (0x%02x)",
+			     policy->flags);
+		return false;
+	}
+
+	count += !!(policy->flags & FSCRYPT_POLICY_FLAG_DIRECT_KEY);
+	count += !!(policy->flags & FSCRYPT_POLICY_FLAG_IV_INO_LBLK_64);
+	count += !!(policy->flags & FSCRYPT_POLICY_FLAG_IV_INO_LBLK_32);
+	if (count > 1) {
+		fscrypt_warn(inode, "Mutually exclusive encryption flags (0x%02x)",
+			     policy->flags);
+		return false;
+	}
+
+	if ((policy->flags & FSCRYPT_POLICY_FLAG_DIRECT_KEY) &&
+	    !supported_direct_key_modes(inode, policy->contents_encryption_mode,
+					policy->filenames_encryption_mode))
+		return false;
+
+	if ((policy->flags & FSCRYPT_POLICY_FLAG_IV_INO_LBLK_64) &&
+	    !supported_iv_ino_lblk_policy(policy, inode, "IV_INO_LBLK_64",
+					  32, 32))
+		return false;
+
+	if ((policy->flags & FSCRYPT_POLICY_FLAG_IV_INO_LBLK_32) &&
+	    /* This uses hashed inode numbers, so ino_bits doesn't matter. */
+	    !supported_iv_ino_lblk_policy(policy, inode, "IV_INO_LBLK_32",
+					  INT_MAX, 32))
+		return false;
+
+	if (memchr_inv(policy->__reserved, 0, sizeof(policy->__reserved))) {
+		fscrypt_warn(inode, "Reserved bits set in encryption policy");
+		return false;
+	}
+
+	return true;
+}
+
 /**
- * fscrypt_supported_policy - check whether an encryption policy is supported
+ * fscrypt_supported_policy() - check whether an encryption policy is supported
+ * @policy_u: the encryption policy
+ * @inode: the inode on which the policy will be used
  *
  * Given an encryption policy, check whether all its encryption modes and other
- * settings are supported by this kernel.  (But we don't currently don't check
- * for crypto API support here, so attempting to use an algorithm not configured
- * into the crypto API will still fail later.)
+ * settings are supported by this kernel on the given inode.  (But we don't
+ * currently don't check for crypto API support here, so attempting to use an
+ * algorithm not configured into the crypto API will still fail later.)
  *
  * Return: %true if supported, else %false
  */
@@ -43,61 +208,19 @@ bool fscrypt_supported_policy(const union fscrypt_policy *policy_u,
 			      const struct inode *inode)
 {
 	switch (policy_u->version) {
-	case FSCRYPT_POLICY_V1: {
-		const struct fscrypt_policy_v1 *policy = &policy_u->v1;
-
-		if (!fscrypt_valid_enc_modes(policy->contents_encryption_mode,
-					     policy->filenames_encryption_mode)) {
-			fscrypt_warn(inode,
-				     "Unsupported encryption modes (contents %d, filenames %d)",
-				     policy->contents_encryption_mode,
-				     policy->filenames_encryption_mode);
-			return false;
-		}
-
-		if (policy->flags & ~FSCRYPT_POLICY_FLAGS_VALID) {
-			fscrypt_warn(inode,
-				     "Unsupported encryption flags (0x%02x)",
-				     policy->flags);
-			return false;
-		}
-
-		return true;
-	}
-	case FSCRYPT_POLICY_V2: {
-		const struct fscrypt_policy_v2 *policy = &policy_u->v2;
-
-		if (!fscrypt_valid_enc_modes(policy->contents_encryption_mode,
-					     policy->filenames_encryption_mode)) {
-			fscrypt_warn(inode,
-				     "Unsupported encryption modes (contents %d, filenames %d)",
-				     policy->contents_encryption_mode,
-				     policy->filenames_encryption_mode);
-			return false;
-		}
-
-		if (policy->flags & ~FSCRYPT_POLICY_FLAGS_VALID) {
-			fscrypt_warn(inode,
-				     "Unsupported encryption flags (0x%02x)",
-				     policy->flags);
-			return false;
-		}
-
-		if (memchr_inv(policy->__reserved, 0,
-			       sizeof(policy->__reserved))) {
-			fscrypt_warn(inode,
-				     "Reserved bits set in encryption policy");
-			return false;
-		}
-
-		return true;
-	}
+	case FSCRYPT_POLICY_V1:
+		return fscrypt_supported_v1_policy(&policy_u->v1, inode);
+	case FSCRYPT_POLICY_V2:
+		return fscrypt_supported_v2_policy(&policy_u->v2, inode);
 	}
 	return false;
 }
 
 /**
- * fscrypt_new_context_from_policy - create a new fscrypt_context from a policy
+ * fscrypt_new_context_from_policy() - create a new fscrypt_context from
+ *				       an fscrypt_policy
+ * @ctx_u: output context
+ * @policy_u: input policy
  *
  * Create an fscrypt_context for an inode that is being assigned the given
  * encryption policy.  A new nonce is randomly generated.
@@ -147,7 +270,11 @@ static int fscrypt_new_context_from_policy(union fscrypt_context *ctx_u,
 }
 
 /**
- * fscrypt_policy_from_context - convert an fscrypt_context to an fscrypt_policy
+ * fscrypt_policy_from_context() - convert an fscrypt_context to
+ *				   an fscrypt_policy
+ * @policy_u: output policy
+ * @ctx_u: input context
+ * @ctx_size: size of input context in bytes
  *
  * Given an fscrypt_context, build the corresponding fscrypt_policy.
  *
@@ -163,7 +290,7 @@ int fscrypt_policy_from_context(union fscrypt_policy *policy_u,
 {
 	memset(policy_u, 0, sizeof(*policy_u));
 
-	if (ctx_size <= 0 || ctx_size != fscrypt_context_size(ctx_u))
+	if (!fscrypt_context_is_valid(ctx_u, ctx_size))
 		return -EINVAL;
 
 	switch (ctx_u->version) {
@@ -259,6 +386,9 @@ static int set_encryption_policy(struct inode *inode,
 					       policy->v2.master_key_identifier);
 		if (err)
 			return err;
+		if (policy->v2.flags & FSCRYPT_POLICY_FLAG_IV_INO_LBLK_32)
+			pr_warn_once("%s (pid %d) is setting an IV_INO_LBLK_32 encryption policy.  This should only be used if there are certain hardware limitations.\n",
+				     current->comm, current->pid);
 		break;
 	default:
 		WARN_ON(1);
@@ -386,6 +516,25 @@ int fscrypt_ioctl_get_policy_ex(struct file *filp, void __user *uarg)
 }
 EXPORT_SYMBOL_GPL(fscrypt_ioctl_get_policy_ex);
 
+/* FS_IOC_GET_ENCRYPTION_NONCE: retrieve file's encryption nonce for testing */
+int fscrypt_ioctl_get_nonce(struct file *filp, void __user *arg)
+{
+	struct inode *inode = file_inode(filp);
+	union fscrypt_context ctx;
+	int ret;
+
+	ret = inode->i_sb->s_cop->get_context(inode, &ctx, sizeof(ctx));
+	if (ret < 0)
+		return ret;
+	if (!fscrypt_context_is_valid(&ctx, ret))
+		return -EINVAL;
+	if (copy_to_user(arg, fscrypt_context_nonce(&ctx),
+			 FS_KEY_DERIVATION_NONCE_SIZE))
+		return -EFAULT;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(fscrypt_ioctl_get_nonce);
+
 /**
  * fscrypt_has_permitted_context() - is a file's encryption policy permitted
  *				     within its directory?
@@ -491,3 +640,127 @@ int fscrypt_inherit_context(struct inode *parent, struct inode *child,
 	return preload ? fscrypt_get_encryption_info(child): 0;
 }
 EXPORT_SYMBOL(fscrypt_inherit_context);
+
+/**
+ * fscrypt_set_test_dummy_encryption() - handle '-o test_dummy_encryption'
+ * @sb: the filesystem on which test_dummy_encryption is being specified
+ * @arg: the argument to the test_dummy_encryption option.
+ *	 If no argument was specified, then @arg->from == NULL.
+ * @dummy_ctx: the filesystem's current dummy context (input/output, see below)
+ *
+ * Handle the test_dummy_encryption mount option by creating a dummy encryption
+ * context, saving it in @dummy_ctx, and adding the corresponding dummy
+ * encryption key to the filesystem.  If the @dummy_ctx is already set, then
+ * instead validate that it matches @arg.  Don't support changing it via
+ * remount, as that is difficult to do safely.
+ *
+ * The reason we use an fscrypt_context rather than an fscrypt_policy is because
+ * we mustn't generate a new nonce each time we access a dummy-encrypted
+ * directory, as that would change the way filenames are encrypted.
+ *
+ * Return: 0 on success (dummy context set, or the same context is already set);
+ *         -EEXIST if a different dummy context is already set;
+ *         or another -errno value.
+ */
+int fscrypt_set_test_dummy_encryption(struct super_block *sb,
+				      const substring_t *arg,
+				      struct fscrypt_dummy_context *dummy_ctx)
+{
+	const char *argstr = "v2";
+	const char *argstr_to_free = NULL;
+	struct fscrypt_key_specifier key_spec = { 0 };
+	int version;
+	union fscrypt_context *ctx = NULL;
+	int err;
+
+	if (arg->from) {
+		argstr = argstr_to_free = match_strdup(arg);
+		if (!argstr)
+			return -ENOMEM;
+	}
+
+	if (!strcmp(argstr, "v1")) {
+		version = FSCRYPT_CONTEXT_V1;
+		key_spec.type = FSCRYPT_KEY_SPEC_TYPE_DESCRIPTOR;
+		memset(key_spec.u.descriptor, 0x42,
+		       FSCRYPT_KEY_DESCRIPTOR_SIZE);
+	} else if (!strcmp(argstr, "v2")) {
+		version = FSCRYPT_CONTEXT_V2;
+		key_spec.type = FSCRYPT_KEY_SPEC_TYPE_IDENTIFIER;
+		/* key_spec.u.identifier gets filled in when adding the key */
+	} else {
+		err = -EINVAL;
+		goto out;
+	}
+
+	if (dummy_ctx->ctx) {
+		/*
+		 * Note: if we ever make test_dummy_encryption support
+		 * specifying other encryption settings, such as the encryption
+		 * modes, we'll need to compare those settings here.
+		 */
+		if (dummy_ctx->ctx->version == version)
+			err = 0;
+		else
+			err = -EEXIST;
+		goto out;
+	}
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	err = fscrypt_add_test_dummy_key(sb, &key_spec);
+	if (err)
+		goto out;
+
+	ctx->version = version;
+	switch (ctx->version) {
+	case FSCRYPT_CONTEXT_V1:
+		ctx->v1.contents_encryption_mode = FSCRYPT_MODE_AES_256_XTS;
+		ctx->v1.filenames_encryption_mode = FSCRYPT_MODE_AES_256_CTS;
+		memcpy(ctx->v1.master_key_descriptor, key_spec.u.descriptor,
+		       FSCRYPT_KEY_DESCRIPTOR_SIZE);
+		break;
+	case FSCRYPT_CONTEXT_V2:
+		ctx->v2.contents_encryption_mode = FSCRYPT_MODE_AES_256_XTS;
+		ctx->v2.filenames_encryption_mode = FSCRYPT_MODE_AES_256_CTS;
+		memcpy(ctx->v2.master_key_identifier, key_spec.u.identifier,
+		       FSCRYPT_KEY_IDENTIFIER_SIZE);
+		break;
+	default:
+		WARN_ON(1);
+		err = -EINVAL;
+		goto out;
+	}
+	dummy_ctx->ctx = ctx;
+	ctx = NULL;
+	err = 0;
+out:
+	kfree(ctx);
+	kfree(argstr_to_free);
+	return err;
+}
+EXPORT_SYMBOL_GPL(fscrypt_set_test_dummy_encryption);
+
+/**
+ * fscrypt_show_test_dummy_encryption() - show '-o test_dummy_encryption'
+ * @seq: the seq_file to print the option to
+ * @sep: the separator character to use
+ * @sb: the filesystem whose options are being shown
+ *
+ * Show the test_dummy_encryption mount option, if it was specified.
+ * This is mainly used for /proc/mounts.
+ */
+void fscrypt_show_test_dummy_encryption(struct seq_file *seq, char sep,
+					struct super_block *sb)
+{
+	const union fscrypt_context *ctx = fscrypt_get_dummy_context(sb);
+
+	if (!ctx)
+		return;
+	seq_printf(seq, "%ctest_dummy_encryption=v%d", sep, ctx->version);
+}
+EXPORT_SYMBOL_GPL(fscrypt_show_test_dummy_encryption);
diff --git a/fs/dax.c b/fs/dax.c
index 2cc43cd914eb..11b16729b86f 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -937,12 +937,11 @@ static int dax_writeback_one(struct xa_state *xas, struct dax_device *dax_dev,
  * on persistent storage prior to completion of the operation.
  */
 int dax_writeback_mapping_range(struct address_space *mapping,
-		struct block_device *bdev, struct writeback_control *wbc)
+		struct dax_device *dax_dev, struct writeback_control *wbc)
 {
 	XA_STATE(xas, &mapping->i_pages, wbc->range_start >> PAGE_SHIFT);
 	struct inode *inode = mapping->host;
 	pgoff_t end_index = wbc->range_end >> PAGE_SHIFT;
-	struct dax_device *dax_dev;
 	void *entry;
 	int ret = 0;
 	unsigned int scanned = 0;
@@ -953,10 +952,6 @@ int dax_writeback_mapping_range(struct address_space *mapping,
 	if (!mapping->nrexceptional || wbc->sync_mode != WB_SYNC_ALL)
 		return 0;
 
-	dax_dev = dax_get_by_host(bdev->bd_disk->disk_name);
-	if (!dax_dev)
-		return -EIO;
-
 	trace_dax_writeback_range(inode, xas.xa_index, end_index);
 
 	tag_pages_for_writeback(mapping, xas.xa_index, end_index);
@@ -977,7 +972,6 @@ int dax_writeback_mapping_range(struct address_space *mapping,
 		xas_lock_irq(&xas);
 	}
 	xas_unlock_irq(&xas);
-	put_dax(dax_dev);
 	trace_dax_writeback_range_done(inode, xas.xa_index, end_index);
 	return ret;
 }
@@ -1044,54 +1038,47 @@ static vm_fault_t dax_load_hole(struct xa_state *xas,
 	return ret;
 }
 
-static bool dax_range_is_aligned(struct block_device *bdev,
-				 unsigned int offset, unsigned int length)
+int dax_iomap_zero(loff_t pos, unsigned offset, unsigned size,
+		   struct iomap *iomap)
 {
-	unsigned short sector_size = bdev_logical_block_size(bdev);
+	sector_t sector = iomap_sector(iomap, pos & PAGE_MASK);
+	pgoff_t pgoff;
+	long rc, id;
+	void *kaddr;
+	bool page_aligned = false;
 
-	if (!IS_ALIGNED(offset, sector_size))
-		return false;
-	if (!IS_ALIGNED(length, sector_size))
-		return false;
 
-	return true;
-}
+	if (IS_ALIGNED(sector << SECTOR_SHIFT, PAGE_SIZE) &&
+	    IS_ALIGNED(size, PAGE_SIZE))
+		page_aligned = true;
 
-int __dax_zero_page_range(struct block_device *bdev,
-		struct dax_device *dax_dev, sector_t sector,
-		unsigned int offset, unsigned int size)
-{
-	if (dax_range_is_aligned(bdev, offset, size)) {
-		sector_t start_sector = sector + (offset >> 9);
+	rc = bdev_dax_pgoff(iomap->bdev, sector, PAGE_SIZE, &pgoff);
+	if (rc)
+		return rc;
 
-		return blkdev_issue_zeroout(bdev, start_sector,
-				size >> 9, GFP_NOFS, 0);
-	} else {
-		pgoff_t pgoff;
-		long rc, id;
-		void *kaddr;
+	id = dax_read_lock();
 
-		rc = bdev_dax_pgoff(bdev, sector, PAGE_SIZE, &pgoff);
-		if (rc)
-			return rc;
+	if (page_aligned)
+		rc = dax_zero_page_range(iomap->dax_dev, pgoff,
+					 size >> PAGE_SHIFT);
+	else
+		rc = dax_direct_access(iomap->dax_dev, pgoff, 1, &kaddr, NULL);
+	if (rc < 0) {
+		dax_read_unlock(id);
+		return rc;
+	}
 
-		id = dax_read_lock();
-		rc = dax_direct_access(dax_dev, pgoff, 1, &kaddr, NULL);
-		if (rc < 0) {
-			dax_read_unlock(id);
-			return rc;
-		}
+	if (!page_aligned) {
 		memset(kaddr + offset, 0, size);
-		dax_flush(dax_dev, kaddr + offset, size);
-		dax_read_unlock(id);
+		dax_flush(iomap->dax_dev, kaddr + offset, size);
 	}
+	dax_read_unlock(id);
 	return 0;
 }
-EXPORT_SYMBOL_GPL(__dax_zero_page_range);
 
 static loff_t
 dax_iomap_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
-		struct iomap *iomap)
+		struct iomap *iomap, struct iomap *srcmap)
 {
 	struct block_device *bdev = iomap->bdev;
 	struct dax_device *dax_dev = iomap->dax_dev;
@@ -1207,6 +1194,9 @@ dax_iomap_rw(struct kiocb *iocb, struct iov_iter *iter,
 		lockdep_assert_held(&inode->i_rwsem);
 	}
 
+	if (iocb->ki_flags & IOCB_NOWAIT)
+		flags |= IOMAP_NOWAIT;
+
 	while (iov_iter_count(iter)) {
 		ret = iomap_apply(inode, pos, iov_iter_count(iter), flags, ops,
 				iter, dax_iomap_actor);
@@ -1248,7 +1238,8 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp,
 	struct inode *inode = mapping->host;
 	unsigned long vaddr = vmf->address;
 	loff_t pos = (loff_t)vmf->pgoff << PAGE_SHIFT;
-	struct iomap iomap = { 0 };
+	struct iomap iomap = { .type = IOMAP_HOLE };
+	struct iomap srcmap = { .type = IOMAP_HOLE };
 	unsigned flags = IOMAP_FAULT;
 	int error, major = 0;
 	bool write = vmf->flags & FAULT_FLAG_WRITE;
@@ -1293,7 +1284,7 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp,
 	 * the file system block size to be equal the page size, which means
 	 * that we never have to deal with more than a single extent here.
 	 */
-	error = ops->iomap_begin(inode, pos, PAGE_SIZE, flags, &iomap);
+	error = ops->iomap_begin(inode, pos, PAGE_SIZE, flags, &iomap, &srcmap);
 	if (iomap_errp)
 		*iomap_errp = error;
 	if (error) {
@@ -1472,7 +1463,8 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp,
 	unsigned int iomap_flags = (write ? IOMAP_WRITE : 0) | IOMAP_FAULT;
 	struct inode *inode = mapping->host;
 	vm_fault_t result = VM_FAULT_FALLBACK;
-	struct iomap iomap = { 0 };
+	struct iomap iomap = { .type = IOMAP_HOLE };
+	struct iomap srcmap = { .type = IOMAP_HOLE };
 	pgoff_t max_pgoff;
 	void *entry;
 	loff_t pos;
@@ -1547,7 +1539,8 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp,
 	 * to look up our filesystem block.
 	 */
 	pos = (loff_t)xas.xa_index << PAGE_SHIFT;
-	error = ops->iomap_begin(inode, pos, PMD_SIZE, iomap_flags, &iomap);
+	error = ops->iomap_begin(inode, pos, PMD_SIZE, iomap_flags, &iomap,
+			&srcmap);
 	if (error)
 		goto unlock_entry;
 
diff --git a/fs/dcache.c b/fs/dcache.c
index e88cf0554e65..b280e07e162b 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -319,7 +319,7 @@ static inline void __d_set_inode_and_type(struct dentry *dentry,
 	flags = READ_ONCE(dentry->d_flags);
 	flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU);
 	flags |= type_flags;
-	WRITE_ONCE(dentry->d_flags, flags);
+	smp_store_release(&dentry->d_flags, flags);
 }
 
 static inline void __d_clear_type_and_inode(struct dentry *dentry)
@@ -903,17 +903,19 @@ struct dentry *dget_parent(struct dentry *dentry)
 {
 	int gotref;
 	struct dentry *ret;
+	unsigned seq;
 
 	/*
 	 * Do optimistic parent lookup without any
 	 * locking.
 	 */
 	rcu_read_lock();
+	seq = raw_seqcount_begin(&dentry->d_seq);
 	ret = READ_ONCE(dentry->d_parent);
 	gotref = lockref_get_not_zero(&ret->d_lockref);
 	rcu_read_unlock();
 	if (likely(gotref)) {
-		if (likely(ret == READ_ONCE(dentry->d_parent)))
+		if (!read_seqcount_retry(&dentry->d_seq, seq))
 			return ret;
 		dput(ret);
 	}
@@ -1319,7 +1321,7 @@ resume:
 
 		if (!list_empty(&dentry->d_subdirs)) {
 			spin_unlock(&this_parent->d_lock);
-			spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_);
+			spin_release(&dentry->d_lock.dep_map, _RET_IP_);
 			this_parent = dentry;
 			spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_);
 			goto repeat;
@@ -1679,7 +1681,7 @@ EXPORT_SYMBOL(d_invalidate);
  * copied and the copy passed in may be reused after this call.
  */
  
-struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name)
+static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name)
 {
 	struct dentry *dentry;
 	char *dname;
diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
index 87846aad594b..ae49a55bda00 100644
--- a/fs/debugfs/file.c
+++ b/fs/debugfs/file.c
@@ -18,6 +18,7 @@
 #include <linux/slab.h>
 #include <linux/atomic.h>
 #include <linux/device.h>
+#include <linux/pm_runtime.h>
 #include <linux/poll.h>
 #include <linux/security.h>
 
@@ -142,18 +143,21 @@ EXPORT_SYMBOL_GPL(debugfs_file_put);
  * We also need to exclude any file that has ways to write or alter it as root
  * can bypass the permissions check.
  */
-static bool debugfs_is_locked_down(struct inode *inode,
-				   struct file *filp,
-				   const struct file_operations *real_fops)
+static int debugfs_locked_down(struct inode *inode,
+			       struct file *filp,
+			       const struct file_operations *real_fops)
 {
 	if ((inode->i_mode & 07777) == 0444 &&
 	    !(filp->f_mode & FMODE_WRITE) &&
 	    !real_fops->unlocked_ioctl &&
 	    !real_fops->compat_ioctl &&
 	    !real_fops->mmap)
-		return false;
+		return 0;
 
-	return security_locked_down(LOCKDOWN_DEBUGFS);
+	if (security_locked_down(LOCKDOWN_DEBUGFS))
+		return -EPERM;
+
+	return 0;
 }
 
 static int open_proxy_open(struct inode *inode, struct file *filp)
@@ -168,12 +172,17 @@ static int open_proxy_open(struct inode *inode, struct file *filp)
 
 	real_fops = debugfs_real_fops(filp);
 
-	r = debugfs_is_locked_down(inode, filp, real_fops);
+	r = debugfs_locked_down(inode, filp, real_fops);
 	if (r)
 		goto out;
 
-	real_fops = fops_get(real_fops);
-	if (!real_fops) {
+	if (!fops_get(real_fops)) {
+#ifdef MODULE
+		if (real_fops->owner &&
+		    real_fops->owner->state == MODULE_STATE_GOING)
+			goto out;
+#endif
+
 		/* Huh? Module did not clean up after itself at exit? */
 		WARN(1, "debugfs file owner did not clean up at exit: %pd",
 			dentry);
@@ -298,12 +307,17 @@ static int full_proxy_open(struct inode *inode, struct file *filp)
 
 	real_fops = debugfs_real_fops(filp);
 
-	r = debugfs_is_locked_down(inode, filp, real_fops);
+	r = debugfs_locked_down(inode, filp, real_fops);
 	if (r)
 		goto out;
 
-	real_fops = fops_get(real_fops);
-	if (!real_fops) {
+	if (!fops_get(real_fops)) {
+#ifdef MODULE
+		if (real_fops->owner &&
+		    real_fops->owner->state == MODULE_STATE_GOING)
+			goto out;
+#endif
+
 		/* Huh? Module did not cleanup after itself at exit? */
 		WARN(1, "debugfs file owner did not clean up at exit: %pd",
 			dentry);
@@ -420,20 +434,11 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_u8_wo, NULL, debugfs_u8_set, "%llu\n");
  * This function creates a file in debugfs with the given name that
  * contains the value of the variable @value.  If the @mode variable is so
  * set, it can be read from, and written to.
- *
- * This function will return a pointer to a dentry if it succeeds.  This
- * pointer must be passed to the debugfs_remove() function when the file is
- * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
- * returned.
- *
- * If debugfs is not enabled in the kernel, the value %ERR_PTR(-ENODEV) will
- * be returned.
  */
-struct dentry *debugfs_create_u8(const char *name, umode_t mode,
-				 struct dentry *parent, u8 *value)
+void debugfs_create_u8(const char *name, umode_t mode, struct dentry *parent,
+		       u8 *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u8,
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u8,
 				   &fops_u8_ro, &fops_u8_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_u8);
@@ -465,20 +470,11 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_u16_wo, NULL, debugfs_u16_set, "%llu\n");
  * This function creates a file in debugfs with the given name that
  * contains the value of the variable @value.  If the @mode variable is so
  * set, it can be read from, and written to.
- *
- * This function will return a pointer to a dentry if it succeeds.  This
- * pointer must be passed to the debugfs_remove() function when the file is
- * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
- * returned.
- *
- * If debugfs is not enabled in the kernel, the value %ERR_PTR(-ENODEV) will
- * be returned.
  */
-struct dentry *debugfs_create_u16(const char *name, umode_t mode,
-				  struct dentry *parent, u16 *value)
+void debugfs_create_u16(const char *name, umode_t mode, struct dentry *parent,
+			u16 *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u16,
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u16,
 				   &fops_u16_ro, &fops_u16_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_u16);
@@ -510,20 +506,11 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_u32_wo, NULL, debugfs_u32_set, "%llu\n");
  * This function creates a file in debugfs with the given name that
  * contains the value of the variable @value.  If the @mode variable is so
  * set, it can be read from, and written to.
- *
- * This function will return a pointer to a dentry if it succeeds.  This
- * pointer must be passed to the debugfs_remove() function when the file is
- * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
- * returned.
- *
- * If debugfs is not enabled in the kernel, the value %ERR_PTR(-ENODEV) will
- * be returned.
  */
-struct dentry *debugfs_create_u32(const char *name, umode_t mode,
-				 struct dentry *parent, u32 *value)
+void debugfs_create_u32(const char *name, umode_t mode, struct dentry *parent,
+			u32 *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u32,
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u32,
 				   &fops_u32_ro, &fops_u32_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_u32);
@@ -556,20 +543,11 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_u64_wo, NULL, debugfs_u64_set, "%llu\n");
  * This function creates a file in debugfs with the given name that
  * contains the value of the variable @value.  If the @mode variable is so
  * set, it can be read from, and written to.
- *
- * This function will return a pointer to a dentry if it succeeds.  This
- * pointer must be passed to the debugfs_remove() function when the file is
- * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
- * returned.
- *
- * If debugfs is not enabled in the kernel, the value %ERR_PTR(-ENODEV) will
- * be returned.
  */
-struct dentry *debugfs_create_u64(const char *name, umode_t mode,
-				 struct dentry *parent, u64 *value)
+void debugfs_create_u64(const char *name, umode_t mode, struct dentry *parent,
+			u64 *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u64,
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_u64,
 				   &fops_u64_ro, &fops_u64_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_u64);
@@ -608,10 +586,10 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_ulong_wo, NULL, debugfs_ulong_set, "%llu\n");
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
+ * you are responsible here.)  If an error occurs, ERR_PTR(-ERROR) will be
  * returned.
  *
- * If debugfs is not enabled in the kernel, the value %ERR_PTR(-ENODEV) will
+ * If debugfs is not enabled in the kernel, the value ERR_PTR(-ENODEV) will
  * be returned.
  */
 struct dentry *debugfs_create_ulong(const char *name, umode_t mode,
@@ -660,10 +638,10 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_x64_wo, NULL, debugfs_u64_set, "0x%016llx\n");
  * @value: a pointer to the variable that the file should read to and write
  *         from.
  */
-struct dentry *debugfs_create_x8(const char *name, umode_t mode,
-				 struct dentry *parent, u8 *value)
+void debugfs_create_x8(const char *name, umode_t mode, struct dentry *parent,
+		       u8 *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_x8,
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_x8,
 				   &fops_x8_ro, &fops_x8_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_x8);
@@ -678,10 +656,10 @@ EXPORT_SYMBOL_GPL(debugfs_create_x8);
  * @value: a pointer to the variable that the file should read to and write
  *         from.
  */
-struct dentry *debugfs_create_x16(const char *name, umode_t mode,
-				 struct dentry *parent, u16 *value)
+void debugfs_create_x16(const char *name, umode_t mode, struct dentry *parent,
+			u16 *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_x16,
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_x16,
 				   &fops_x16_ro, &fops_x16_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_x16);
@@ -696,10 +674,10 @@ EXPORT_SYMBOL_GPL(debugfs_create_x16);
  * @value: a pointer to the variable that the file should read to and write
  *         from.
  */
-struct dentry *debugfs_create_x32(const char *name, umode_t mode,
-				 struct dentry *parent, u32 *value)
+void debugfs_create_x32(const char *name, umode_t mode, struct dentry *parent,
+			u32 *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_x32,
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_x32,
 				   &fops_x32_ro, &fops_x32_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_x32);
@@ -714,10 +692,10 @@ EXPORT_SYMBOL_GPL(debugfs_create_x32);
  * @value: a pointer to the variable that the file should read to and write
  *         from.
  */
-struct dentry *debugfs_create_x64(const char *name, umode_t mode,
-				 struct dentry *parent, u64 *value)
+void debugfs_create_x64(const char *name, umode_t mode, struct dentry *parent,
+			u64 *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value, &fops_x64,
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_x64,
 				   &fops_x64_ro, &fops_x64_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_x64);
@@ -748,12 +726,11 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_size_t_wo, NULL, debugfs_size_t_set, "%llu\n");
  * @value: a pointer to the variable that the file should read to and write
  *         from.
  */
-struct dentry *debugfs_create_size_t(const char *name, umode_t mode,
-				     struct dentry *parent, size_t *value)
+void debugfs_create_size_t(const char *name, umode_t mode,
+			   struct dentry *parent, size_t *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value,
-					&fops_size_t, &fops_size_t_ro,
-					&fops_size_t_wo);
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_size_t,
+				   &fops_size_t_ro, &fops_size_t_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_size_t);
 
@@ -785,12 +762,11 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_atomic_t_wo, NULL, debugfs_atomic_t_set,
  * @value: a pointer to the variable that the file should read to and write
  *         from.
  */
-struct dentry *debugfs_create_atomic_t(const char *name, umode_t mode,
-				 struct dentry *parent, atomic_t *value)
+void debugfs_create_atomic_t(const char *name, umode_t mode,
+			     struct dentry *parent, atomic_t *value)
 {
-	return debugfs_create_mode_unsafe(name, mode, parent, value,
-					&fops_atomic_t, &fops_atomic_t_ro,
-					&fops_atomic_t_wo);
+	debugfs_create_mode_unsafe(name, mode, parent, value, &fops_atomic_t,
+				   &fops_atomic_t_ro, &fops_atomic_t_wo);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_atomic_t);
 
@@ -875,10 +851,10 @@ static const struct file_operations fops_bool_wo = {
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
+ * you are responsible here.)  If an error occurs, ERR_PTR(-ERROR) will be
  * returned.
  *
- * If debugfs is not enabled in the kernel, the value %ERR_PTR(-ENODEV) will
+ * If debugfs is not enabled in the kernel, the value ERR_PTR(-ENODEV) will
  * be returned.
  */
 struct dentry *debugfs_create_bool(const char *name, umode_t mode,
@@ -928,10 +904,10 @@ static const struct file_operations fops_blob = {
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
+ * you are responsible here.)  If an error occurs, ERR_PTR(-ERROR) will be
  * returned.
  *
- * If debugfs is not enabled in the kernel, the value %ERR_PTR(-ENODEV) will
+ * If debugfs is not enabled in the kernel, the value ERR_PTR(-ENODEV) will
  * be returned.
  */
 struct dentry *debugfs_create_blob(const char *name, umode_t mode,
@@ -1086,7 +1062,14 @@ static int debugfs_show_regset32(struct seq_file *s, void *data)
 {
 	struct debugfs_regset32 *regset = s->private;
 
+	if (regset->dev)
+		pm_runtime_get_sync(regset->dev);
+
 	debugfs_print_regs32(s, regset->regs, regset->nregs, regset->base, "");
+
+	if (regset->dev)
+		pm_runtime_put(regset->dev);
+
 	return 0;
 }
 
@@ -1116,21 +1099,12 @@ static const struct file_operations fops_regset32 = {
  * This function creates a file in debugfs with the given name that reports
  * the names and values of a set of 32-bit registers. If the @mode variable
  * is so set it can be read from. Writing is not supported.
- *
- * This function will return a pointer to a dentry if it succeeds.  This
- * pointer must be passed to the debugfs_remove() function when the file is
- * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
- * returned.
- *
- * If debugfs is not enabled in the kernel, the value %ERR_PTR(-ENODEV) will
- * be returned.
  */
-struct dentry *debugfs_create_regset32(const char *name, umode_t mode,
-				       struct dentry *parent,
-				       struct debugfs_regset32 *regset)
+void debugfs_create_regset32(const char *name, umode_t mode,
+			     struct dentry *parent,
+			     struct debugfs_regset32 *regset)
 {
-	return debugfs_create_file(name, mode, parent, regset, &fops_regset32);
+	debugfs_create_file(name, mode, parent, regset, &fops_regset32);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_regset32);
 
@@ -1187,4 +1161,3 @@ struct dentry *debugfs_create_devm_seqfile(struct device *dev, const char *name,
 				   &debugfs_devm_entry_ops);
 }
 EXPORT_SYMBOL_GPL(debugfs_create_devm_seqfile);
-
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 7b975dbb2bb4..b7f2e971ecbc 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -299,13 +299,9 @@ struct dentry *debugfs_lookup(const char *name, struct dentry *parent)
 	if (!parent)
 		parent = debugfs_mount->mnt_root;
 
-	dentry = lookup_one_len_unlocked(name, parent, strlen(name));
+	dentry = lookup_positive_unlocked(name, parent, strlen(name));
 	if (IS_ERR(dentry))
 		return NULL;
-	if (!d_really_is_positive(dentry)) {
-		dput(dentry);
-		return NULL;
-	}
 	return dentry;
 }
 EXPORT_SYMBOL_GPL(debugfs_lookup);
@@ -336,7 +332,10 @@ static struct dentry *start_creating(const char *name, struct dentry *parent)
 		parent = debugfs_mount->mnt_root;
 
 	inode_lock(d_inode(parent));
-	dentry = lookup_one_len(name, parent, strlen(name));
+	if (unlikely(IS_DEADDIR(d_inode(parent))))
+		dentry = ERR_PTR(-ENOENT);
+	else
+		dentry = lookup_one_len(name, parent, strlen(name));
 	if (!IS_ERR(dentry) && d_really_is_positive(dentry)) {
 		if (d_is_dir(dentry))
 			pr_err("Directory '%s' with parent '%s' already present!\n",
@@ -427,7 +426,7 @@ static struct dentry *__debugfs_create_file(const char *name, umode_t mode,
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
+ * you are responsible here.)  If an error occurs, ERR_PTR(-ERROR) will be
  * returned.
  *
  * If debugfs is not enabled in the kernel, the value -%ENODEV will be
@@ -502,26 +501,16 @@ EXPORT_SYMBOL_GPL(debugfs_create_file_unsafe);
  * wide range of flexibility in creating a file, or a directory (if you want
  * to create a directory, the debugfs_create_dir() function is
  * recommended to be used instead.)
- *
- * This function will return a pointer to a dentry if it succeeds.  This
- * pointer must be passed to the debugfs_remove() function when the file is
- * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
- * returned.
- *
- * If debugfs is not enabled in the kernel, the value -%ENODEV will be
- * returned.
  */
-struct dentry *debugfs_create_file_size(const char *name, umode_t mode,
-					struct dentry *parent, void *data,
-					const struct file_operations *fops,
-					loff_t file_size)
+void debugfs_create_file_size(const char *name, umode_t mode,
+			      struct dentry *parent, void *data,
+			      const struct file_operations *fops,
+			      loff_t file_size)
 {
 	struct dentry *de = debugfs_create_file(name, mode, parent, data, fops);
 
 	if (de)
 		d_inode(de)->i_size = file_size;
-	return de;
 }
 EXPORT_SYMBOL_GPL(debugfs_create_file_size);
 
@@ -538,7 +527,7 @@ EXPORT_SYMBOL_GPL(debugfs_create_file_size);
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR) will be
+ * you are responsible here.)  If an error occurs, ERR_PTR(-ERROR) will be
  * returned.
  *
  * If debugfs is not enabled in the kernel, the value -%ENODEV will be
@@ -631,7 +620,7 @@ EXPORT_SYMBOL(debugfs_create_automount);
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the symbolic
  * link is to be removed (no automatic cleanup happens if your module is
- * unloaded, you are responsible here.)  If an error occurs, %ERR_PTR(-ERROR)
+ * unloaded, you are responsible here.)  If an error occurs, ERR_PTR(-ERROR)
  * will be returned.
  *
  * If debugfs is not enabled in the kernel, the value -%ENODEV will be
@@ -685,62 +674,15 @@ static void __debugfs_file_removed(struct dentry *dentry)
 		wait_for_completion(&fsd->active_users_drained);
 }
 
-static int __debugfs_remove(struct dentry *dentry, struct dentry *parent)
-{
-	int ret = 0;
-
-	if (simple_positive(dentry)) {
-		dget(dentry);
-		if (d_is_dir(dentry)) {
-			ret = simple_rmdir(d_inode(parent), dentry);
-			if (!ret)
-				fsnotify_rmdir(d_inode(parent), dentry);
-		} else {
-			simple_unlink(d_inode(parent), dentry);
-			fsnotify_unlink(d_inode(parent), dentry);
-		}
-		if (!ret)
-			d_delete(dentry);
-		if (d_is_reg(dentry))
-			__debugfs_file_removed(dentry);
-		dput(dentry);
-	}
-	return ret;
-}
-
-/**
- * debugfs_remove - removes a file or directory from the debugfs filesystem
- * @dentry: a pointer to a the dentry of the file or directory to be
- *          removed.  If this parameter is NULL or an error value, nothing
- *          will be done.
- *
- * This function removes a file or directory in debugfs that was previously
- * created with a call to another debugfs function (like
- * debugfs_create_file() or variants thereof.)
- *
- * This function is required to be called in order for the file to be
- * removed, no automatic cleanup of files will happen when a module is
- * removed, you are responsible here.
- */
-void debugfs_remove(struct dentry *dentry)
+static void remove_one(struct dentry *victim)
 {
-	struct dentry *parent;
-	int ret;
-
-	if (IS_ERR_OR_NULL(dentry))
-		return;
-
-	parent = dentry->d_parent;
-	inode_lock(d_inode(parent));
-	ret = __debugfs_remove(dentry, parent);
-	inode_unlock(d_inode(parent));
-	if (!ret)
-		simple_release_fs(&debugfs_mount, &debugfs_mount_count);
+        if (d_is_reg(victim))
+		__debugfs_file_removed(victim);
+	simple_release_fs(&debugfs_mount, &debugfs_mount_count);
 }
-EXPORT_SYMBOL_GPL(debugfs_remove);
 
 /**
- * debugfs_remove_recursive - recursively removes a directory
+ * debugfs_remove - recursively removes a directory
  * @dentry: a pointer to a the dentry of the directory to be removed.  If this
  *          parameter is NULL or an error value, nothing will be done.
  *
@@ -752,65 +694,16 @@ EXPORT_SYMBOL_GPL(debugfs_remove);
  * removed, no automatic cleanup of files will happen when a module is
  * removed, you are responsible here.
  */
-void debugfs_remove_recursive(struct dentry *dentry)
+void debugfs_remove(struct dentry *dentry)
 {
-	struct dentry *child, *parent;
-
 	if (IS_ERR_OR_NULL(dentry))
 		return;
 
-	parent = dentry;
- down:
-	inode_lock(d_inode(parent));
- loop:
-	/*
-	 * The parent->d_subdirs is protected by the d_lock. Outside that
-	 * lock, the child can be unlinked and set to be freed which can
-	 * use the d_u.d_child as the rcu head and corrupt this list.
-	 */
-	spin_lock(&parent->d_lock);
-	list_for_each_entry(child, &parent->d_subdirs, d_child) {
-		if (!simple_positive(child))
-			continue;
-
-		/* perhaps simple_empty(child) makes more sense */
-		if (!list_empty(&child->d_subdirs)) {
-			spin_unlock(&parent->d_lock);
-			inode_unlock(d_inode(parent));
-			parent = child;
-			goto down;
-		}
-
-		spin_unlock(&parent->d_lock);
-
-		if (!__debugfs_remove(child, parent))
-			simple_release_fs(&debugfs_mount, &debugfs_mount_count);
-
-		/*
-		 * The parent->d_lock protects agaist child from unlinking
-		 * from d_subdirs. When releasing the parent->d_lock we can
-		 * no longer trust that the next pointer is valid.
-		 * Restart the loop. We'll skip this one with the
-		 * simple_positive() check.
-		 */
-		goto loop;
-	}
-	spin_unlock(&parent->d_lock);
-
-	inode_unlock(d_inode(parent));
-	child = parent;
-	parent = parent->d_parent;
-	inode_lock(d_inode(parent));
-
-	if (child != dentry)
-		/* go up */
-		goto loop;
-
-	if (!__debugfs_remove(child, parent))
-		simple_release_fs(&debugfs_mount, &debugfs_mount_count);
-	inode_unlock(d_inode(parent));
+	simple_pin_fs(&debug_fs_type, &debugfs_mount, &debugfs_mount_count);
+	simple_recursive_removal(dentry, remove_one);
+	simple_release_fs(&debugfs_mount, &debugfs_mount_count);
 }
-EXPORT_SYMBOL_GPL(debugfs_remove_recursive);
+EXPORT_SYMBOL_GPL(debugfs_remove);
 
 /**
  * debugfs_rename - rename a file/directory in the debugfs filesystem
@@ -910,4 +803,3 @@ static int __init debugfs_init(void)
 	return retval;
 }
 core_initcall(debugfs_init);
-
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 9329ced91f1d..6d5370eac2a8 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -39,6 +39,8 @@
 #include <linux/atomic.h>
 #include <linux/prefetch.h>
 
+#include "internal.h"
+
 /*
  * How many user pages to map in one call to get_user_pages().  This determines
  * the size of a structure in the slab cache
@@ -221,27 +223,6 @@ static inline struct page *dio_get_page(struct dio *dio,
 }
 
 /*
- * Warn about a page cache invalidation failure during a direct io write.
- */
-void dio_warn_stale_pagecache(struct file *filp)
-{
-	static DEFINE_RATELIMIT_STATE(_rs, 86400 * HZ, DEFAULT_RATELIMIT_BURST);
-	char pathname[128];
-	struct inode *inode = file_inode(filp);
-	char *path;
-
-	errseq_set(&inode->i_mapping->wb_err, -EIO);
-	if (__ratelimit(&_rs)) {
-		path = file_path(filp, pathname, sizeof(pathname));
-		if (IS_ERR(path))
-			path = "(unknown)";
-		pr_crit("Page cache invalidation failure on direct I/O.  Possible data corruption due to collision with buffered I/O!\n");
-		pr_crit("File: %s PID: %d Comm: %.20s\n", path, current->pid,
-			current->comm);
-	}
-}
-
-/*
  * dio_complete() - called when all DIO BIO I/O has been completed
  *
  * This drops i_dio_count, lets interested parties know that a DIO operation
@@ -519,7 +500,7 @@ static struct bio *dio_await_one(struct dio *dio)
 		spin_unlock_irqrestore(&dio->bio_lock, flags);
 		if (!(dio->iocb->ki_flags & IOCB_HIPRI) ||
 		    !blk_poll(dio->bio_disk->queue, dio->bio_cookie, true))
-			io_schedule();
+			blk_io_schedule();
 		/* wake up sets us TASK_RUNNING */
 		spin_lock_irqsave(&dio->bio_lock, flags);
 		dio->waiter = NULL;
diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c
index 3951d39b9b75..cdfaf4f0e11a 100644
--- a/fs/dlm/lowcomms.c
+++ b/fs/dlm/lowcomms.c
@@ -1035,7 +1035,7 @@ static void sctp_connect_to_sock(struct connection *con)
 	int result;
 	int addr_len;
 	struct socket *sock;
-	struct timeval tv = { .tv_sec = 5, .tv_usec = 0 };
+	struct __kernel_sock_timeval tv = { .tv_sec = 5, .tv_usec = 0 };
 
 	if (con->nodeid == 0) {
 		log_print("attempt to connect sock 0 foiled");
@@ -1087,12 +1087,12 @@ static void sctp_connect_to_sock(struct connection *con)
 	 * since O_NONBLOCK argument in connect() function does not work here,
 	 * then, we should restore the default value of this attribute.
 	 */
-	kernel_setsockopt(sock, SOL_SOCKET, SO_SNDTIMEO_OLD, (char *)&tv,
+	kernel_setsockopt(sock, SOL_SOCKET, SO_SNDTIMEO_NEW, (char *)&tv,
 			  sizeof(tv));
 	result = sock->ops->connect(sock, (struct sockaddr *)&daddr, addr_len,
 				   0);
 	memset(&tv, 0, sizeof(tv));
-	kernel_setsockopt(sock, SOL_SOCKET, SO_SNDTIMEO_OLD, (char *)&tv,
+	kernel_setsockopt(sock, SOL_SOCKET, SO_SNDTIMEO_NEW, (char *)&tv,
 			  sizeof(tv));
 
 	if (result == -EINPROGRESS)
diff --git a/fs/drop_caches.c b/fs/drop_caches.c
index d31b6c72b476..dc1a1d5d825b 100644
--- a/fs/drop_caches.c
+++ b/fs/drop_caches.c
@@ -35,11 +35,11 @@ static void drop_pagecache_sb(struct super_block *sb, void *unused)
 		spin_unlock(&inode->i_lock);
 		spin_unlock(&sb->s_inode_list_lock);
 
-		cond_resched();
 		invalidate_mapping_pages(inode->i_mapping, 0, -1);
 		iput(toput_inode);
 		toput_inode = inode;
 
+		cond_resched();
 		spin_lock(&sb->s_inode_list_lock);
 	}
 	spin_unlock(&sb->s_inode_list_lock);
diff --git a/fs/ecryptfs/Kconfig b/fs/ecryptfs/Kconfig
index 522c35d5292b..1bdeaa6d5790 100644
--- a/fs/ecryptfs/Kconfig
+++ b/fs/ecryptfs/Kconfig
@@ -7,7 +7,7 @@ config ECRYPT_FS
 	select CRYPTO_MD5
 	help
 	  Encrypted filesystem that operates on the VFS layer.  See
-	  <file:Documentation/filesystems/ecryptfs.txt> to learn more about
+	  <file:Documentation/filesystems/ecryptfs.rst> to learn more about
 	  eCryptfs.  Userspace components are required and can be
 	  obtained from <http://ecryptfs.sf.net>.
 
diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index f91db24bbf3b..0681540c48d9 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -48,18 +48,6 @@ void ecryptfs_from_hex(char *dst, char *src, int dst_size)
 	}
 }
 
-static int ecryptfs_hash_digest(struct crypto_shash *tfm,
-				char *src, int len, char *dst)
-{
-	SHASH_DESC_ON_STACK(desc, tfm);
-	int err;
-
-	desc->tfm = tfm;
-	err = crypto_shash_digest(desc, src, len, dst);
-	shash_desc_zero(desc);
-	return err;
-}
-
 /**
  * ecryptfs_calculate_md5 - calculates the md5 of @src
  * @dst: Pointer to 16 bytes of allocated memory
@@ -74,11 +62,8 @@ static int ecryptfs_calculate_md5(char *dst,
 				  struct ecryptfs_crypt_stat *crypt_stat,
 				  char *src, int len)
 {
-	struct crypto_shash *tfm;
-	int rc = 0;
+	int rc = crypto_shash_tfm_digest(crypt_stat->hash_tfm, src, len, dst);
 
-	tfm = crypt_stat->hash_tfm;
-	rc = ecryptfs_hash_digest(tfm, src, len, dst);
 	if (rc) {
 		printk(KERN_ERR
 		       "%s: Error computing crypto hash; rc = [%d]\n",
@@ -311,8 +296,10 @@ static int crypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat,
 	struct extent_crypt_result ecr;
 	int rc = 0;
 
-	BUG_ON(!crypt_stat || !crypt_stat->tfm
-	       || !(crypt_stat->flags & ECRYPTFS_STRUCT_INITIALIZED));
+	if (!crypt_stat || !crypt_stat->tfm
+	       || !(crypt_stat->flags & ECRYPTFS_STRUCT_INITIALIZED))
+		return -EINVAL;
+
 	if (unlikely(ecryptfs_verbosity > 0)) {
 		ecryptfs_printk(KERN_DEBUG, "Key size [%zd]; key:\n",
 				crypt_stat->key_size);
@@ -1586,7 +1573,7 @@ ecryptfs_process_key_cipher(struct crypto_skcipher **key_tfm,
 	}
 	crypto_skcipher_set_flags(*key_tfm, CRYPTO_TFM_REQ_FORBID_WEAK_KEYS);
 	if (*key_size == 0)
-		*key_size = crypto_skcipher_default_keysize(*key_tfm);
+		*key_size = crypto_skcipher_max_keysize(*key_tfm);
 	get_random_bytes(dummy_key, *key_size);
 	rc = crypto_skcipher_setkey(*key_tfm, dummy_key, *key_size);
 	if (rc) {
diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h
index 1c1a56be7ea2..e6ac78c62ca4 100644
--- a/fs/ecryptfs/ecryptfs_kernel.h
+++ b/fs/ecryptfs/ecryptfs_kernel.h
@@ -8,7 +8,7 @@
  * Copyright (C) 2004-2008 International Business Machines Corp.
  *   Author(s): Michael A. Halcrow <mahalcro@us.ibm.com>
  *              Trevor S. Highland <trevor.highland@gmail.com>
- *              Tyler Hicks <tyhicks@ou.edu>
+ *              Tyler Hicks <code@tyhicks.com>
  */
 
 #ifndef ECRYPTFS_KERNEL_H
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index feecb57defa7..5fb45d865ce5 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -378,6 +378,7 @@ ecryptfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 		return rc;
 
 	switch (cmd) {
+	case FITRIM:
 	case FS_IOC32_GETFLAGS:
 	case FS_IOC32_SETFLAGS:
 	case FS_IOC32_GETVERSION:
diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c
index 216fbe6a4837..af3eb02bbca1 100644
--- a/fs/ecryptfs/keystore.c
+++ b/fs/ecryptfs/keystore.c
@@ -1304,7 +1304,7 @@ parse_tag_1_packet(struct ecryptfs_crypt_stat *crypt_stat,
 		printk(KERN_WARNING "Tag 1 packet contains key larger "
 		       "than ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES\n");
 		rc = -EINVAL;
-		goto out;
+		goto out_free;
 	}
 	memcpy((*new_auth_tok)->session_key.encrypted_key,
 	       &data[(*packet_size)], (body_size - (ECRYPTFS_SIG_SIZE + 2)));
@@ -2204,9 +2204,9 @@ write_tag_3_packet(char *dest, size_t *remaining_bytes,
 	if (mount_crypt_stat->global_default_cipher_key_size == 0) {
 		printk(KERN_WARNING "No key size specified at mount; "
 		       "defaulting to [%d]\n",
-		       crypto_skcipher_default_keysize(tfm));
+		       crypto_skcipher_max_keysize(tfm));
 		mount_crypt_stat->global_default_cipher_key_size =
-			crypto_skcipher_default_keysize(tfm);
+			crypto_skcipher_max_keysize(tfm);
 	}
 	if (crypt_stat->key_size == 0)
 		crypt_stat->key_size =
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index b8a7ce379ffe..e63259fdef28 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -7,7 +7,7 @@
  * Copyright (C) 2004-2007 International Business Machines Corp.
  *   Author(s): Michael A. Halcrow <mahalcro@us.ibm.com>
  *              Michael C. Thompson <mcthomps@us.ibm.com>
- *              Tyler Hicks <tyhicks@ou.edu>
+ *              Tyler Hicks <code@tyhicks.com>
  */
 
 #include <linux/dcache.h>
diff --git a/fs/ecryptfs/messaging.c b/fs/ecryptfs/messaging.c
index d668e60b85b5..8646ba76def3 100644
--- a/fs/ecryptfs/messaging.c
+++ b/fs/ecryptfs/messaging.c
@@ -4,7 +4,7 @@
  *
  * Copyright (C) 2004-2008 International Business Machines Corp.
  *   Author(s): Michael A. Halcrow <mhalcrow@us.ibm.com>
- *		Tyler Hicks <tyhicks@ou.edu>
+ *		Tyler Hicks <code@tyhicks.com>
  */
 #include <linux/sched.h>
 #include <linux/slab.h>
@@ -379,6 +379,7 @@ int __init ecryptfs_init_messaging(void)
 					* ecryptfs_message_buf_len),
 				       GFP_KERNEL);
 	if (!ecryptfs_msg_ctx_arr) {
+		kfree(ecryptfs_daemon_hash);
 		rc = -ENOMEM;
 		goto out;
 	}
diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c
index cffa0c1ec829..019572c6b39a 100644
--- a/fs/ecryptfs/mmap.c
+++ b/fs/ecryptfs/mmap.c
@@ -524,16 +524,12 @@ out:
 
 static sector_t ecryptfs_bmap(struct address_space *mapping, sector_t block)
 {
-	int rc = 0;
-	struct inode *inode;
-	struct inode *lower_inode;
-
-	inode = (struct inode *)mapping->host;
-	lower_inode = ecryptfs_inode_to_lower(inode);
-	if (lower_inode->i_mapping->a_ops->bmap)
-		rc = lower_inode->i_mapping->a_ops->bmap(lower_inode->i_mapping,
-							 block);
-	return rc;
+	struct inode *lower_inode = ecryptfs_inode_to_lower(mapping->host);
+	int ret = bmap(lower_inode, &block);
+
+	if (ret)
+		return 0;
+	return block;
 }
 
 const struct address_space_operations ecryptfs_aops = {
diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c
index fa4f6447ddad..12c66f5d92dd 100644
--- a/fs/efivarfs/super.c
+++ b/fs/efivarfs/super.c
@@ -252,7 +252,7 @@ static struct file_system_type efivarfs_type = {
 
 static __init int efivarfs_init(void)
 {
-	if (!efi_enabled(EFI_RUNTIME_SERVICES))
+	if (!efi_rt_services_supported(EFI_RT_SUPPORTED_VARIABLE_SERVICES))
 		return -ENODEV;
 
 	if (!efivars_kobject())
diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig
index 9d634d3a1845..74b0aaa7114c 100644
--- a/fs/erofs/Kconfig
+++ b/fs/erofs/Kconfig
@@ -3,6 +3,7 @@
 config EROFS_FS
 	tristate "EROFS filesystem support"
 	depends on BLOCK
+	select LIBCRC32C
 	help
 	  EROFS (Enhanced Read-Only File System) is a lightweight
 	  read-only file system with modern designs (eg. page-sized
diff --git a/fs/erofs/data.c b/fs/erofs/data.c
index fc3a8d8064f8..d0542151e8c4 100644
--- a/fs/erofs/data.c
+++ b/fs/erofs/data.c
@@ -280,47 +280,36 @@ static int erofs_raw_access_readpage(struct file *file, struct page *page)
 	return 0;
 }
 
-static int erofs_raw_access_readpages(struct file *filp,
-				      struct address_space *mapping,
-				      struct list_head *pages,
-				      unsigned int nr_pages)
+static void erofs_raw_access_readahead(struct readahead_control *rac)
 {
 	erofs_off_t last_block;
 	struct bio *bio = NULL;
-	gfp_t gfp = readahead_gfp_mask(mapping);
-	struct page *page = list_last_entry(pages, struct page, lru);
-
-	trace_erofs_readpages(mapping->host, page, nr_pages, true);
+	struct page *page;
 
-	for (; nr_pages; --nr_pages) {
-		page = list_entry(pages->prev, struct page, lru);
+	trace_erofs_readpages(rac->mapping->host, readahead_index(rac),
+			readahead_count(rac), true);
 
+	while ((page = readahead_page(rac))) {
 		prefetchw(&page->flags);
-		list_del(&page->lru);
 
-		if (!add_to_page_cache_lru(page, mapping, page->index, gfp)) {
-			bio = erofs_read_raw_page(bio, mapping, page,
-						  &last_block, nr_pages, true);
+		bio = erofs_read_raw_page(bio, rac->mapping, page, &last_block,
+				readahead_count(rac), true);
 
-			/* all the page errors are ignored when readahead */
-			if (IS_ERR(bio)) {
-				pr_err("%s, readahead error at page %lu of nid %llu\n",
-				       __func__, page->index,
-				       EROFS_I(mapping->host)->nid);
+		/* all the page errors are ignored when readahead */
+		if (IS_ERR(bio)) {
+			pr_err("%s, readahead error at page %lu of nid %llu\n",
+			       __func__, page->index,
+			       EROFS_I(rac->mapping->host)->nid);
 
-				bio = NULL;
-			}
+			bio = NULL;
 		}
 
-		/* pages could still be locked */
 		put_page(page);
 	}
-	DBG_BUGON(!list_empty(pages));
 
 	/* the rare case (end in gaps) */
 	if (bio)
 		submit_bio(bio);
-	return 0;
 }
 
 static int erofs_get_block(struct inode *inode, sector_t iblock,
@@ -358,7 +347,7 @@ static sector_t erofs_bmap(struct address_space *mapping, sector_t block)
 /* for uncompressed (aligned) files and raw access for other files */
 const struct address_space_operations erofs_raw_access_aops = {
 	.readpage = erofs_raw_access_readpage,
-	.readpages = erofs_raw_access_readpages,
+	.readahead = erofs_raw_access_readahead,
 	.bmap = erofs_bmap,
 };
 
diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c
index 19f89f9fb10c..7628816f2453 100644
--- a/fs/erofs/decompressor.c
+++ b/fs/erofs/decompressor.c
@@ -73,7 +73,7 @@ static int z_erofs_lz4_prepare_destpages(struct z_erofs_decompress_req *rq,
 			victim = availables[--top];
 			get_page(victim);
 		} else {
-			victim = erofs_allocpage(pagepool, GFP_KERNEL, false);
+			victim = erofs_allocpage(pagepool, GFP_KERNEL);
 			if (!victim)
 				return -ENOMEM;
 			victim->mapping = Z_EROFS_MAPPING_STAGING;
@@ -157,17 +157,27 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out)
 		}
 	}
 
-	ret = LZ4_decompress_safe_partial(src + inputmargin, out,
-					  inlen, rq->outputsize,
-					  rq->outputsize);
-	if (ret < 0) {
-		erofs_err(rq->sb, "failed to decompress, in[%u, %u] out[%u]",
-			  inlen, inputmargin, rq->outputsize);
+	/* legacy format could compress extra data in a pcluster. */
+	if (rq->partial_decoding || !support_0padding)
+		ret = LZ4_decompress_safe_partial(src + inputmargin, out,
+						  inlen, rq->outputsize,
+						  rq->outputsize);
+	else
+		ret = LZ4_decompress_safe(src + inputmargin, out,
+					  inlen, rq->outputsize);
+
+	if (ret != rq->outputsize) {
+		erofs_err(rq->sb, "failed to decompress %d in[%u, %u] out[%u]",
+			  ret, inlen, inputmargin, rq->outputsize);
+
 		WARN_ON(1);
 		print_hex_dump(KERN_DEBUG, "[ in]: ", DUMP_PREFIX_OFFSET,
 			       16, 1, src + inputmargin, inlen, true);
 		print_hex_dump(KERN_DEBUG, "[out]: ", DUMP_PREFIX_OFFSET,
 			       16, 1, out, rq->outputsize, true);
+
+		if (ret >= 0)
+			memset(out + ret, 0, rq->outputsize - ret);
 		ret = -EIO;
 	}
 
@@ -264,7 +274,7 @@ static int z_erofs_decompress_generic(struct z_erofs_decompress_req *rq,
 
 	i = 0;
 	while (1) {
-		dst = vm_map_ram(rq->out, nrpages_out, -1, PAGE_KERNEL);
+		dst = vm_map_ram(rq->out, nrpages_out, -1);
 
 		/* retry two more times (totally 3 times) */
 		if (dst || ++i >= 3)
@@ -306,24 +316,22 @@ static int z_erofs_shifted_transform(const struct z_erofs_decompress_req *rq,
 	}
 
 	src = kmap_atomic(*rq->in);
-	if (!rq->out[0]) {
-		dst = NULL;
-	} else {
+	if (rq->out[0]) {
 		dst = kmap_atomic(rq->out[0]);
 		memcpy(dst + rq->pageofs_out, src, righthalf);
+		kunmap_atomic(dst);
 	}
 
-	if (rq->out[1] == *rq->in) {
-		memmove(src, src + righthalf, rq->pageofs_out);
-	} else if (nrpages_out == 2) {
-		if (dst)
-			kunmap_atomic(dst);
+	if (nrpages_out == 2) {
 		DBG_BUGON(!rq->out[1]);
-		dst = kmap_atomic(rq->out[1]);
-		memcpy(dst, src + righthalf, rq->pageofs_out);
+		if (rq->out[1] == *rq->in) {
+			memmove(src, src + righthalf, rq->pageofs_out);
+		} else {
+			dst = kmap_atomic(rq->out[1]);
+			memcpy(dst, src + righthalf, rq->pageofs_out);
+			kunmap_atomic(dst);
+		}
 	}
-	if (dst)
-		kunmap_atomic(dst);
 	kunmap_atomic(src);
 	return 0;
 }
diff --git a/fs/erofs/erofs_fs.h b/fs/erofs/erofs_fs.h
index b1ee5654750d..385fa49c7749 100644
--- a/fs/erofs/erofs_fs.h
+++ b/fs/erofs/erofs_fs.h
@@ -11,6 +11,8 @@
 
 #define EROFS_SUPER_OFFSET      1024
 
+#define EROFS_FEATURE_COMPAT_SB_CHKSUM          0x00000001
+
 /*
  * Any bits that aren't in EROFS_ALL_FEATURE_INCOMPAT should
  * be incompatible with this kernel version.
@@ -37,7 +39,6 @@ struct erofs_super_block {
 	__u8 uuid[16];          /* 128-bit uuid for volume */
 	__u8 volume_name[16];   /* volume name */
 	__le32 feature_incompat;
-
 	__u8 reserved2[44];
 };
 
diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h
index 544a453f3076..5eead7fdc7a6 100644
--- a/fs/erofs/internal.h
+++ b/fs/erofs/internal.h
@@ -52,8 +52,8 @@ struct erofs_sb_info {
 	struct list_head list;
 	struct mutex umount_mutex;
 
-	/* the dedicated workstation for compression */
-	struct radix_tree_root workstn_tree;
+	/* managed XArray arranged in physical block number */
+	struct xarray managed_pslots;
 
 	/* threshold for decompression synchronously */
 	unsigned int max_sync_decompress_pages;
@@ -85,6 +85,7 @@ struct erofs_sb_info {
 
 	u8 uuid[16];                    /* 128-bit uuid for volume */
 	u8 volume_name[16];             /* volume name */
+	u32 feature_compat;
 	u32 feature_incompat;
 
 	unsigned int mount_opt;
@@ -278,9 +279,7 @@ static inline unsigned int erofs_inode_datalayout(unsigned int value)
 extern const struct super_operations erofs_sops;
 
 extern const struct address_space_operations erofs_raw_access_aops;
-#ifdef CONFIG_EROFS_FS_ZIP
-extern const struct address_space_operations z_erofs_vle_normalaccess_aops;
-#endif
+extern const struct address_space_operations z_erofs_aops;
 
 /*
  * Logical to physical block mapping, used by erofs_map_blocks()
@@ -382,7 +381,7 @@ int erofs_namei(struct inode *dir, struct qstr *name,
 extern const struct file_operations erofs_dir_fops;
 
 /* utils.c / zdata.c */
-struct page *erofs_allocpage(struct list_head *pool, gfp_t gfp, bool nofail);
+struct page *erofs_allocpage(struct list_head *pool, gfp_t gfp);
 
 #if (EROFS_PCPUBUF_NR_PAGES > 0)
 void *erofs_get_pcpubuf(unsigned int pagenr);
@@ -402,9 +401,9 @@ static inline void *erofs_get_pcpubuf(unsigned int pagenr)
 #ifdef CONFIG_EROFS_FS_ZIP
 int erofs_workgroup_put(struct erofs_workgroup *grp);
 struct erofs_workgroup *erofs_find_workgroup(struct super_block *sb,
-					     pgoff_t index, bool *tag);
-int erofs_register_workgroup(struct super_block *sb,
-			     struct erofs_workgroup *grp, bool tag);
+					     pgoff_t index);
+struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb,
+					       struct erofs_workgroup *grp);
 void erofs_workgroup_free_rcu(struct erofs_workgroup *grp);
 void erofs_shrinker_register(struct super_block *sb);
 void erofs_shrinker_unregister(struct super_block *sb);
diff --git a/fs/erofs/super.c b/fs/erofs/super.c
index 0e369494f2f2..b514c67e5fc2 100644
--- a/fs/erofs/super.c
+++ b/fs/erofs/super.c
@@ -9,6 +9,7 @@
 #include <linux/statfs.h>
 #include <linux/parser.h>
 #include <linux/seq_file.h>
+#include <linux/crc32c.h>
 #include "xattr.h"
 
 #define CREATE_TRACE_POINTS
@@ -46,6 +47,30 @@ void _erofs_info(struct super_block *sb, const char *function,
 	va_end(args);
 }
 
+static int erofs_superblock_csum_verify(struct super_block *sb, void *sbdata)
+{
+	struct erofs_super_block *dsb;
+	u32 expected_crc, crc;
+
+	dsb = kmemdup(sbdata + EROFS_SUPER_OFFSET,
+		      EROFS_BLKSIZ - EROFS_SUPER_OFFSET, GFP_KERNEL);
+	if (!dsb)
+		return -ENOMEM;
+
+	expected_crc = le32_to_cpu(dsb->checksum);
+	dsb->checksum = 0;
+	/* to allow for x86 boot sectors and other oddities. */
+	crc = crc32c(~0, dsb, EROFS_BLKSIZ - EROFS_SUPER_OFFSET);
+	kfree(dsb);
+
+	if (crc != expected_crc) {
+		erofs_err(sb, "invalid checksum 0x%08x, 0x%08x expected",
+			  crc, expected_crc);
+		return -EBADMSG;
+	}
+	return 0;
+}
+
 static void erofs_inode_init_once(void *ptr)
 {
 	struct erofs_inode *vi = ptr;
@@ -112,7 +137,7 @@ static int erofs_read_superblock(struct super_block *sb)
 
 	sbi = EROFS_SB(sb);
 
-	data = kmap_atomic(page);
+	data = kmap(page);
 	dsb = (struct erofs_super_block *)(data + EROFS_SUPER_OFFSET);
 
 	ret = -EINVAL;
@@ -121,6 +146,13 @@ static int erofs_read_superblock(struct super_block *sb)
 		goto out;
 	}
 
+	sbi->feature_compat = le32_to_cpu(dsb->feature_compat);
+	if (sbi->feature_compat & EROFS_FEATURE_COMPAT_SB_CHKSUM) {
+		ret = erofs_superblock_csum_verify(sb, data);
+		if (ret)
+			goto out;
+	}
+
 	blkszbits = dsb->blkszbits;
 	/* 9(512 bytes) + LOG_SECTORS_PER_BLOCK == LOG_BLOCK_SIZE */
 	if (blkszbits != LOG_BLOCK_SIZE) {
@@ -155,7 +187,7 @@ static int erofs_read_superblock(struct super_block *sb)
 	}
 	ret = 0;
 out:
-	kunmap_atomic(data);
+	kunmap(page);
 	put_page(page);
 	return ret;
 }
@@ -393,7 +425,7 @@ static int erofs_fill_super(struct super_block *sb, void *data, int silent)
 		sb->s_flags &= ~SB_POSIXACL;
 
 #ifdef CONFIG_EROFS_FS_ZIP
-	INIT_RADIX_TREE(&sbi->workstn_tree, GFP_ATOMIC);
+	xa_init(&sbi->managed_pslots);
 #endif
 
 	/* get the root inode */
@@ -566,9 +598,6 @@ static int erofs_show_options(struct seq_file *seq, struct dentry *root)
 		seq_puts(seq, ",cache_strategy=readahead");
 	} else if (sbi->cache_strategy == EROFS_ZIP_CACHE_READAROUND) {
 		seq_puts(seq, ",cache_strategy=readaround");
-	} else {
-		seq_puts(seq, ",cache_strategy=(unknown)");
-		DBG_BUGON(1);
 	}
 #endif
 	return 0;
diff --git a/fs/erofs/utils.c b/fs/erofs/utils.c
index d92b3e753a6f..52d0be10f1aa 100644
--- a/fs/erofs/utils.c
+++ b/fs/erofs/utils.c
@@ -7,7 +7,7 @@
 #include "internal.h"
 #include <linux/pagevec.h>
 
-struct page *erofs_allocpage(struct list_head *pool, gfp_t gfp, bool nofail)
+struct page *erofs_allocpage(struct list_head *pool, gfp_t gfp)
 {
 	struct page *page;
 
@@ -16,7 +16,7 @@ struct page *erofs_allocpage(struct list_head *pool, gfp_t gfp, bool nofail)
 		DBG_BUGON(page_ref_count(page) != 1);
 		list_del(&page->lru);
 	} else {
-		page = alloc_pages(gfp | (nofail ? __GFP_NOFAIL : 0), 0);
+		page = alloc_page(gfp);
 	}
 	return page;
 }
@@ -37,9 +37,6 @@ void *erofs_get_pcpubuf(unsigned int pagenr)
 /* global shrink count (for all mounted EROFS instances) */
 static atomic_long_t erofs_global_shrink_cnt;
 
-#define __erofs_workgroup_get(grp)	atomic_inc(&(grp)->refcount)
-#define __erofs_workgroup_put(grp)	atomic_dec(&(grp)->refcount)
-
 static int erofs_workgroup_get(struct erofs_workgroup *grp)
 {
 	int o;
@@ -59,18 +56,15 @@ repeat:
 }
 
 struct erofs_workgroup *erofs_find_workgroup(struct super_block *sb,
-					     pgoff_t index, bool *tag)
+					     pgoff_t index)
 {
 	struct erofs_sb_info *sbi = EROFS_SB(sb);
 	struct erofs_workgroup *grp;
 
 repeat:
 	rcu_read_lock();
-	grp = radix_tree_lookup(&sbi->workstn_tree, index);
+	grp = xa_load(&sbi->managed_pslots, index);
 	if (grp) {
-		*tag = xa_pointer_tag(grp);
-		grp = xa_untag_pointer(grp);
-
 		if (erofs_workgroup_get(grp)) {
 			/* prefer to relax rcu read side */
 			rcu_read_unlock();
@@ -83,46 +77,37 @@ repeat:
 	return grp;
 }
 
-int erofs_register_workgroup(struct super_block *sb,
-			     struct erofs_workgroup *grp,
-			     bool tag)
+struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb,
+					       struct erofs_workgroup *grp)
 {
-	struct erofs_sb_info *sbi;
-	int err;
-
-	/* grp shouldn't be broken or used before */
-	if (atomic_read(&grp->refcount) != 1) {
-		DBG_BUGON(1);
-		return -EINVAL;
-	}
-
-	err = radix_tree_preload(GFP_NOFS);
-	if (err)
-		return err;
-
-	sbi = EROFS_SB(sb);
-	xa_lock(&sbi->workstn_tree);
-
-	grp = xa_tag_pointer(grp, tag);
+	struct erofs_sb_info *const sbi = EROFS_SB(sb);
+	struct erofs_workgroup *pre;
 
 	/*
-	 * Bump up reference count before making this workgroup
-	 * visible to other users in order to avoid potential UAF
-	 * without serialized by workstn_lock.
+	 * Bump up a reference count before making this visible
+	 * to others for the XArray in order to avoid potential
+	 * UAF without serialized by xa_lock.
 	 */
-	__erofs_workgroup_get(grp);
-
-	err = radix_tree_insert(&sbi->workstn_tree, grp->index, grp);
-	if (err)
-		/*
-		 * it's safe to decrease since the workgroup isn't visible
-		 * and refcount >= 2 (cannot be freezed).
-		 */
-		__erofs_workgroup_put(grp);
+	atomic_inc(&grp->refcount);
 
-	xa_unlock(&sbi->workstn_tree);
-	radix_tree_preload_end();
-	return err;
+repeat:
+	xa_lock(&sbi->managed_pslots);
+	pre = __xa_cmpxchg(&sbi->managed_pslots, grp->index,
+			   NULL, grp, GFP_NOFS);
+	if (pre) {
+		if (xa_is_err(pre)) {
+			pre = ERR_PTR(xa_err(pre));
+		} else if (erofs_workgroup_get(pre)) {
+			/* try to legitimize the current in-tree one */
+			xa_unlock(&sbi->managed_pslots);
+			cond_resched();
+			goto repeat;
+		}
+		atomic_dec(&grp->refcount);
+		grp = pre;
+	}
+	xa_unlock(&sbi->managed_pslots);
+	return grp;
 }
 
 static void  __erofs_workgroup_free(struct erofs_workgroup *grp)
@@ -149,8 +134,7 @@ static void erofs_workgroup_unfreeze_final(struct erofs_workgroup *grp)
 }
 
 static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi,
-					   struct erofs_workgroup *grp,
-					   bool cleanup)
+					   struct erofs_workgroup *grp)
 {
 	/*
 	 * If managed cache is on, refcount of workgroups
@@ -162,7 +146,7 @@ static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi,
 
 	/*
 	 * Note that all cached pages should be unattached
-	 * before deleted from the radix tree. Otherwise some
+	 * before deleted from the XArray. Otherwise some
 	 * cached pages could be still attached to the orphan
 	 * old workgroup when the new one is available in the tree.
 	 */
@@ -176,8 +160,7 @@ static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi,
 	 * however in order to avoid some race conditions, add a
 	 * DBG_BUGON to observe this in advance.
 	 */
-	DBG_BUGON(xa_untag_pointer(radix_tree_delete(&sbi->workstn_tree,
-						     grp->index)) != grp);
+	DBG_BUGON(xa_erase(&sbi->managed_pslots, grp->index) != grp);
 
 	/*
 	 * If managed cache is on, last refcount should indicate
@@ -188,37 +171,21 @@ static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi,
 }
 
 static unsigned long erofs_shrink_workstation(struct erofs_sb_info *sbi,
-					      unsigned long nr_shrink,
-					      bool cleanup)
+					      unsigned long nr_shrink)
 {
-	pgoff_t first_index = 0;
-	void *batch[PAGEVEC_SIZE];
+	struct erofs_workgroup *grp;
 	unsigned int freed = 0;
+	unsigned long index;
 
-	int i, found;
-repeat:
-	xa_lock(&sbi->workstn_tree);
-
-	found = radix_tree_gang_lookup(&sbi->workstn_tree,
-				       batch, first_index, PAGEVEC_SIZE);
-
-	for (i = 0; i < found; ++i) {
-		struct erofs_workgroup *grp = xa_untag_pointer(batch[i]);
-
-		first_index = grp->index + 1;
-
+	xa_for_each(&sbi->managed_pslots, index, grp) {
 		/* try to shrink each valid workgroup */
-		if (!erofs_try_to_release_workgroup(sbi, grp, cleanup))
+		if (!erofs_try_to_release_workgroup(sbi, grp))
 			continue;
 
 		++freed;
 		if (!--nr_shrink)
 			break;
 	}
-	xa_unlock(&sbi->workstn_tree);
-
-	if (i && nr_shrink)
-		goto repeat;
 	return freed;
 }
 
@@ -245,7 +212,8 @@ void erofs_shrinker_unregister(struct super_block *sb)
 	struct erofs_sb_info *const sbi = EROFS_SB(sb);
 
 	mutex_lock(&sbi->umount_mutex);
-	erofs_shrink_workstation(sbi, ~0UL, true);
+	/* clean up all remaining workgroups in memory */
+	erofs_shrink_workstation(sbi, ~0UL);
 
 	spin_lock(&erofs_sb_list_lock);
 	list_del(&sbi->list);
@@ -294,7 +262,7 @@ static unsigned long erofs_shrink_scan(struct shrinker *shrink,
 		spin_unlock(&erofs_sb_list_lock);
 		sbi->shrinker_run_no = run_no;
 
-		freed += erofs_shrink_workstation(sbi, nr, false);
+		freed += erofs_shrink_workstation(sbi, nr - freed);
 
 		spin_lock(&erofs_sb_list_lock);
 		/* Get the next list element before we move this one */
diff --git a/fs/erofs/xattr.c b/fs/erofs/xattr.c
index a13a78725c57..b766c3ee5fa8 100644
--- a/fs/erofs/xattr.c
+++ b/fs/erofs/xattr.c
@@ -649,6 +649,8 @@ ssize_t erofs_listxattr(struct dentry *dentry,
 	struct listxattr_iter it;
 
 	ret = init_inode_xattrs(d_inode(dentry));
+	if (ret == -ENOATTR)
+		return 0;
 	if (ret)
 		return ret;
 
diff --git a/fs/erofs/xattr.h b/fs/erofs/xattr.h
index 3585b84d2f20..50966f1c676e 100644
--- a/fs/erofs/xattr.h
+++ b/fs/erofs/xattr.h
@@ -46,18 +46,19 @@ extern const struct xattr_handler erofs_xattr_security_handler;
 
 static inline const struct xattr_handler *erofs_xattr_handler(unsigned int idx)
 {
-static const struct xattr_handler *xattr_handler_map[] = {
-	[EROFS_XATTR_INDEX_USER] = &erofs_xattr_user_handler,
+	static const struct xattr_handler *xattr_handler_map[] = {
+		[EROFS_XATTR_INDEX_USER] = &erofs_xattr_user_handler,
 #ifdef CONFIG_EROFS_FS_POSIX_ACL
-	[EROFS_XATTR_INDEX_POSIX_ACL_ACCESS] = &posix_acl_access_xattr_handler,
-	[EROFS_XATTR_INDEX_POSIX_ACL_DEFAULT] =
-		&posix_acl_default_xattr_handler,
+		[EROFS_XATTR_INDEX_POSIX_ACL_ACCESS] =
+			&posix_acl_access_xattr_handler,
+		[EROFS_XATTR_INDEX_POSIX_ACL_DEFAULT] =
+			&posix_acl_default_xattr_handler,
 #endif
-	[EROFS_XATTR_INDEX_TRUSTED] = &erofs_xattr_trusted_handler,
+		[EROFS_XATTR_INDEX_TRUSTED] = &erofs_xattr_trusted_handler,
 #ifdef CONFIG_EROFS_FS_SECURITY
-	[EROFS_XATTR_INDEX_SECURITY] = &erofs_xattr_security_handler,
+		[EROFS_XATTR_INDEX_SECURITY] = &erofs_xattr_security_handler,
 #endif
-};
+	};
 
 	return idx && idx < ARRAY_SIZE(xattr_handler_map) ?
 		xattr_handler_map[idx] : NULL;
diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c
index fad80c97d247..187f93b4900e 100644
--- a/fs/erofs/zdata.c
+++ b/fs/erofs/zdata.c
@@ -67,16 +67,6 @@ static void z_erofs_pcluster_init_once(void *ptr)
 		pcl->compressed_pages[i] = NULL;
 }
 
-static void z_erofs_pcluster_init_always(struct z_erofs_pcluster *pcl)
-{
-	struct z_erofs_collection *cl = z_erofs_primarycollection(pcl);
-
-	atomic_set(&pcl->obj.refcount, 1);
-
-	DBG_BUGON(cl->nr_pages);
-	DBG_BUGON(cl->vcnt);
-}
-
 int __init z_erofs_init_zip_subsystem(void)
 {
 	pcluster_cachep = kmem_cache_create("erofs_compress",
@@ -337,40 +327,31 @@ retry:
 	return COLLECT_PRIMARY;	/* :( better luck next time */
 }
 
-static struct z_erofs_collection *cllookup(struct z_erofs_collector *clt,
-					   struct inode *inode,
-					   struct erofs_map_blocks *map)
+static int z_erofs_lookup_collection(struct z_erofs_collector *clt,
+				     struct inode *inode,
+				     struct erofs_map_blocks *map)
 {
-	struct erofs_workgroup *grp;
-	struct z_erofs_pcluster *pcl;
+	struct z_erofs_pcluster *pcl = clt->pcl;
 	struct z_erofs_collection *cl;
 	unsigned int length;
-	bool tag;
-
-	grp = erofs_find_workgroup(inode->i_sb, map->m_pa >> PAGE_SHIFT, &tag);
-	if (!grp)
-		return NULL;
 
-	pcl = container_of(grp, struct z_erofs_pcluster, obj);
+	/* to avoid unexpected loop formed by corrupted images */
 	if (clt->owned_head == &pcl->next || pcl == clt->tailpcl) {
 		DBG_BUGON(1);
-		erofs_workgroup_put(grp);
-		return ERR_PTR(-EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	cl = z_erofs_primarycollection(pcl);
 	if (cl->pageofs != (map->m_la & ~PAGE_MASK)) {
 		DBG_BUGON(1);
-		erofs_workgroup_put(grp);
-		return ERR_PTR(-EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	length = READ_ONCE(pcl->length);
 	if (length & Z_EROFS_PCLUSTER_FULL_LENGTH) {
 		if ((map->m_llen << Z_EROFS_PCLUSTER_LENGTH_BIT) > length) {
 			DBG_BUGON(1);
-			erofs_workgroup_put(grp);
-			return ERR_PTR(-EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 	} else {
 		unsigned int llen = map->m_llen << Z_EROFS_PCLUSTER_LENGTH_BIT;
@@ -392,25 +373,25 @@ static struct z_erofs_collection *cllookup(struct z_erofs_collector *clt,
 	/* clean tailpcl if the current owned_head is Z_EROFS_PCLUSTER_TAIL */
 	if (clt->owned_head == Z_EROFS_PCLUSTER_TAIL)
 		clt->tailpcl = NULL;
-	clt->pcl = pcl;
 	clt->cl = cl;
-	return cl;
+	return 0;
 }
 
-static struct z_erofs_collection *clregister(struct z_erofs_collector *clt,
-					     struct inode *inode,
-					     struct erofs_map_blocks *map)
+static int z_erofs_register_collection(struct z_erofs_collector *clt,
+				       struct inode *inode,
+				       struct erofs_map_blocks *map)
 {
 	struct z_erofs_pcluster *pcl;
 	struct z_erofs_collection *cl;
+	struct erofs_workgroup *grp;
 	int err;
 
 	/* no available workgroup, let's allocate one */
 	pcl = kmem_cache_alloc(pcluster_cachep, GFP_NOFS);
 	if (!pcl)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 
-	z_erofs_pcluster_init_always(pcl);
+	atomic_set(&pcl->obj.refcount, 1);
 	pcl->obj.index = map->m_pa >> PAGE_SHIFT;
 
 	pcl->length = (map->m_llen << Z_EROFS_PCLUSTER_LENGTH_BIT) |
@@ -430,19 +411,29 @@ static struct z_erofs_collection *clregister(struct z_erofs_collector *clt,
 	clt->mode = COLLECT_PRIMARY_FOLLOWED;
 
 	cl = z_erofs_primarycollection(pcl);
+
+	/* must be cleaned before freeing to slab */
+	DBG_BUGON(cl->nr_pages);
+	DBG_BUGON(cl->vcnt);
+
 	cl->pageofs = map->m_la & ~PAGE_MASK;
 
 	/*
 	 * lock all primary followed works before visible to others
 	 * and mutex_trylock *never* fails for a new pcluster.
 	 */
-	mutex_trylock(&cl->lock);
+	DBG_BUGON(!mutex_trylock(&cl->lock));
 
-	err = erofs_register_workgroup(inode->i_sb, &pcl->obj, 0);
-	if (err) {
-		mutex_unlock(&cl->lock);
-		kmem_cache_free(pcluster_cachep, pcl);
-		return ERR_PTR(-EAGAIN);
+	grp = erofs_insert_workgroup(inode->i_sb, &pcl->obj);
+	if (IS_ERR(grp)) {
+		err = PTR_ERR(grp);
+		goto err_out;
+	}
+
+	if (grp != &pcl->obj) {
+		clt->pcl = container_of(grp, struct z_erofs_pcluster, obj);
+		err = -EEXIST;
+		goto err_out;
 	}
 	/* used to check tail merging loop due to corrupted images */
 	if (clt->owned_head == Z_EROFS_PCLUSTER_TAIL)
@@ -450,14 +441,20 @@ static struct z_erofs_collection *clregister(struct z_erofs_collector *clt,
 	clt->owned_head = &pcl->next;
 	clt->pcl = pcl;
 	clt->cl = cl;
-	return cl;
+	return 0;
+
+err_out:
+	mutex_unlock(&cl->lock);
+	kmem_cache_free(pcluster_cachep, pcl);
+	return err;
 }
 
 static int z_erofs_collector_begin(struct z_erofs_collector *clt,
 				   struct inode *inode,
 				   struct erofs_map_blocks *map)
 {
-	struct z_erofs_collection *cl;
+	struct erofs_workgroup *grp;
+	int ret;
 
 	DBG_BUGON(clt->cl);
 
@@ -470,20 +467,27 @@ static int z_erofs_collector_begin(struct z_erofs_collector *clt,
 		return -EINVAL;
 	}
 
-repeat:
-	cl = cllookup(clt, inode, map);
-	if (!cl) {
-		cl = clregister(clt, inode, map);
+	grp = erofs_find_workgroup(inode->i_sb, map->m_pa >> PAGE_SHIFT);
+	if (grp) {
+		clt->pcl = container_of(grp, struct z_erofs_pcluster, obj);
+	} else {
+		ret = z_erofs_register_collection(clt, inode, map);
 
-		if (cl == ERR_PTR(-EAGAIN))
-			goto repeat;
+		if (!ret)
+			goto out;
+		if (ret != -EEXIST)
+			return ret;
 	}
 
-	if (IS_ERR(cl))
-		return PTR_ERR(cl);
+	ret = z_erofs_lookup_collection(clt, inode, map);
+	if (ret) {
+		erofs_workgroup_put(&clt->pcl->obj);
+		return ret;
+	}
 
+out:
 	z_erofs_pagevec_ctor_init(&clt->vector, Z_EROFS_NR_INLINE_PAGEVECS,
-				  cl->pagevec, cl->vcnt);
+				  clt->cl->pagevec, clt->cl->vcnt);
 
 	clt->compressedpages = clt->pcl->compressed_pages;
 	if (clt->mode <= COLLECT_PRIMARY) /* cannot do in-place I/O */
@@ -543,15 +547,6 @@ static bool z_erofs_collector_end(struct z_erofs_collector *clt)
 	return true;
 }
 
-static inline struct page *__stagingpage_alloc(struct list_head *pagepool,
-					       gfp_t gfp)
-{
-	struct page *page = erofs_allocpage(pagepool, gfp, true);
-
-	page->mapping = Z_EROFS_MAPPING_STAGING;
-	return page;
-}
-
 static bool should_alloc_managed_pages(struct z_erofs_decompress_frontend *fe,
 				       unsigned int cachestrategy,
 				       erofs_off_t la)
@@ -571,7 +566,7 @@ static int z_erofs_do_read_page(struct z_erofs_decompress_frontend *fe,
 				struct list_head *pagepool)
 {
 	struct inode *const inode = fe->inode;
-	struct erofs_sb_info *const sbi __maybe_unused = EROFS_I_SB(inode);
+	struct erofs_sb_info *const sbi = EROFS_I_SB(inode);
 	struct erofs_map_blocks *const map = &fe->map;
 	struct z_erofs_collector *const clt = &fe->clt;
 	const loff_t offset = page_offset(page);
@@ -658,8 +653,9 @@ retry:
 	/* should allocate an additional staging page for pagevec */
 	if (err == -EAGAIN) {
 		struct page *const newpage =
-			__stagingpage_alloc(pagepool, GFP_NOFS);
+			erofs_allocpage(pagepool, GFP_NOFS | __GFP_NOFAIL);
 
+		newpage->mapping = Z_EROFS_MAPPING_STAGING;
 		err = z_erofs_attach_page(clt, newpage,
 					  Z_EROFS_PAGE_TYPE_EXCLUSIVE);
 		if (!err)
@@ -698,13 +694,11 @@ err_out:
 	goto out;
 }
 
-static void z_erofs_vle_unzip_kickoff(void *ptr, int bios)
+static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io,
+				       bool sync, int bios)
 {
-	tagptr1_t t = tagptr_init(tagptr1_t, ptr);
-	struct z_erofs_unzip_io *io = tagptr_unfold_ptr(t);
-	bool background = tagptr_unfold_tags(t);
-
-	if (!background) {
+	/* wake up the caller thread for sync decompression */
+	if (sync) {
 		unsigned long flags;
 
 		spin_lock_irqsave(&io->u.wait.lock, flags);
@@ -718,37 +712,30 @@ static void z_erofs_vle_unzip_kickoff(void *ptr, int bios)
 		queue_work(z_erofs_workqueue, &io->u.work);
 }
 
-static inline void z_erofs_vle_read_endio(struct bio *bio)
+static void z_erofs_decompressqueue_endio(struct bio *bio)
 {
-	struct erofs_sb_info *sbi = NULL;
+	tagptr1_t t = tagptr_init(tagptr1_t, bio->bi_private);
+	struct z_erofs_decompressqueue *q = tagptr_unfold_ptr(t);
 	blk_status_t err = bio->bi_status;
 	struct bio_vec *bvec;
 	struct bvec_iter_all iter_all;
 
 	bio_for_each_segment_all(bvec, bio, iter_all) {
 		struct page *page = bvec->bv_page;
-		bool cachemngd = false;
 
 		DBG_BUGON(PageUptodate(page));
 		DBG_BUGON(!page->mapping);
 
-		if (!sbi && !z_erofs_page_is_staging(page))
-			sbi = EROFS_SB(page->mapping->host->i_sb);
-
-		/* sbi should already be gotten if the page is managed */
-		if (sbi)
-			cachemngd = erofs_page_is_managed(sbi, page);
-
 		if (err)
 			SetPageError(page);
-		else if (cachemngd)
-			SetPageUptodate(page);
 
-		if (cachemngd)
+		if (erofs_page_is_managed(EROFS_SB(q->sb), page)) {
+			if (!err)
+				SetPageUptodate(page);
 			unlock_page(page);
+		}
 	}
-
-	z_erofs_vle_unzip_kickoff(bio->bi_private, -1);
+	z_erofs_decompress_kickoff(q, tagptr_unfold_tags(t), -1);
 	bio_put(bio);
 }
 
@@ -953,9 +940,8 @@ out:
 	return err;
 }
 
-static void z_erofs_vle_unzip_all(struct super_block *sb,
-				  struct z_erofs_unzip_io *io,
-				  struct list_head *pagepool)
+static void z_erofs_decompress_queue(const struct z_erofs_decompressqueue *io,
+				     struct list_head *pagepool)
 {
 	z_erofs_next_pcluster_t owned = io->head;
 
@@ -971,21 +957,21 @@ static void z_erofs_vle_unzip_all(struct super_block *sb,
 		pcl = container_of(owned, struct z_erofs_pcluster, next);
 		owned = READ_ONCE(pcl->next);
 
-		z_erofs_decompress_pcluster(sb, pcl, pagepool);
+		z_erofs_decompress_pcluster(io->sb, pcl, pagepool);
 	}
 }
 
-static void z_erofs_vle_unzip_wq(struct work_struct *work)
+static void z_erofs_decompressqueue_work(struct work_struct *work)
 {
-	struct z_erofs_unzip_io_sb *iosb =
-		container_of(work, struct z_erofs_unzip_io_sb, io.u.work);
+	struct z_erofs_decompressqueue *bgq =
+		container_of(work, struct z_erofs_decompressqueue, u.work);
 	LIST_HEAD(pagepool);
 
-	DBG_BUGON(iosb->io.head == Z_EROFS_PCLUSTER_TAIL_CLOSED);
-	z_erofs_vle_unzip_all(iosb->sb, &iosb->io, &pagepool);
+	DBG_BUGON(bgq->head == Z_EROFS_PCLUSTER_TAIL_CLOSED);
+	z_erofs_decompress_queue(bgq, &pagepool);
 
 	put_pages_list(&pagepool);
-	kvfree(iosb);
+	kvfree(bgq);
 }
 
 static struct page *pickup_page_for_submission(struct z_erofs_pcluster *pcl,
@@ -994,8 +980,6 @@ static struct page *pickup_page_for_submission(struct z_erofs_pcluster *pcl,
 					       struct address_space *mc,
 					       gfp_t gfp)
 {
-	/* determined at compile time to avoid too many #ifdefs */
-	const bool nocache = __builtin_constant_p(mc) ? !mc : false;
 	const pgoff_t index = pcl->obj.index;
 	bool tocache = false;
 
@@ -1016,7 +1000,7 @@ repeat:
 	 * the cached page has not been allocated and
 	 * an placeholder is out there, prepare it now.
 	 */
-	if (!nocache && page == PAGE_UNALLOCATED) {
+	if (page == PAGE_UNALLOCATED) {
 		tocache = true;
 		goto out_allocpage;
 	}
@@ -1029,21 +1013,6 @@ repeat:
 	mapping = READ_ONCE(page->mapping);
 
 	/*
-	 * if managed cache is disabled, it's no way to
-	 * get such a cached-like page.
-	 */
-	if (nocache) {
-		/* if managed cache is disabled, it is impossible `justfound' */
-		DBG_BUGON(justfound);
-
-		/* and it should be locked, not uptodate, and not truncated */
-		DBG_BUGON(!PageLocked(page));
-		DBG_BUGON(PageUptodate(page));
-		DBG_BUGON(!mapping);
-		goto out;
-	}
-
-	/*
 	 * unmanaged (file) pages are all locked solidly,
 	 * therefore it is impossible for `mapping' to be NULL.
 	 */
@@ -1093,50 +1062,52 @@ repeat:
 	unlock_page(page);
 	put_page(page);
 out_allocpage:
-	page = __stagingpage_alloc(pagepool, gfp);
-	if (oldpage != cmpxchg(&pcl->compressed_pages[nr], oldpage, page)) {
-		list_add(&page->lru, pagepool);
-		cpu_relax();
-		goto repeat;
-	}
-	if (nocache || !tocache)
-		goto out;
-	if (add_to_page_cache_lru(page, mc, index + nr, gfp)) {
+	page = erofs_allocpage(pagepool, gfp | __GFP_NOFAIL);
+	if (!tocache || add_to_page_cache_lru(page, mc, index + nr, gfp)) {
+		/* non-LRU / non-movable temporary page is needed */
 		page->mapping = Z_EROFS_MAPPING_STAGING;
-		goto out;
+		tocache = false;
 	}
 
+	if (oldpage != cmpxchg(&pcl->compressed_pages[nr], oldpage, page)) {
+		if (tocache) {
+			/* since it added to managed cache successfully */
+			unlock_page(page);
+			put_page(page);
+		} else {
+			list_add(&page->lru, pagepool);
+		}
+		cond_resched();
+		goto repeat;
+	}
 	set_page_private(page, (unsigned long)pcl);
 	SetPagePrivate(page);
 out:	/* the only exit (for tracing and debugging) */
 	return page;
 }
 
-static struct z_erofs_unzip_io *jobqueue_init(struct super_block *sb,
-					      struct z_erofs_unzip_io *io,
-					      bool foreground)
+static struct z_erofs_decompressqueue *
+jobqueue_init(struct super_block *sb,
+	      struct z_erofs_decompressqueue *fgq, bool *fg)
 {
-	struct z_erofs_unzip_io_sb *iosb;
+	struct z_erofs_decompressqueue *q;
 
-	if (foreground) {
-		/* waitqueue available for foreground io */
-		DBG_BUGON(!io);
-
-		init_waitqueue_head(&io->u.wait);
-		atomic_set(&io->pending_bios, 0);
-		goto out;
+	if (fg && !*fg) {
+		q = kvzalloc(sizeof(*q), GFP_KERNEL | __GFP_NOWARN);
+		if (!q) {
+			*fg = true;
+			goto fg_out;
+		}
+		INIT_WORK(&q->u.work, z_erofs_decompressqueue_work);
+	} else {
+fg_out:
+		q = fgq;
+		init_waitqueue_head(&fgq->u.wait);
+		atomic_set(&fgq->pending_bios, 0);
 	}
-
-	iosb = kvzalloc(sizeof(*iosb), GFP_KERNEL | __GFP_NOFAIL);
-	DBG_BUGON(!iosb);
-
-	/* initialize fields in the allocated descriptor */
-	io = &iosb->io;
-	iosb->sb = sb;
-	INIT_WORK(&io->u.work, z_erofs_vle_unzip_wq);
-out:
-	io->head = Z_EROFS_PCLUSTER_TAIL_CLOSED;
-	return io;
+	q->sb = sb;
+	q->head = Z_EROFS_PCLUSTER_TAIL_CLOSED;
+	return q;
 }
 
 /* define decompression jobqueue types */
@@ -1147,22 +1118,17 @@ enum {
 };
 
 static void *jobqueueset_init(struct super_block *sb,
-			      z_erofs_next_pcluster_t qtail[],
-			      struct z_erofs_unzip_io *q[],
-			      struct z_erofs_unzip_io *fgq,
-			      bool forcefg)
+			      struct z_erofs_decompressqueue *q[],
+			      struct z_erofs_decompressqueue *fgq, bool *fg)
 {
 	/*
 	 * if managed cache is enabled, bypass jobqueue is needed,
 	 * no need to read from device for all pclusters in this queue.
 	 */
-	q[JQ_BYPASS] = jobqueue_init(sb, fgq + JQ_BYPASS, true);
-	qtail[JQ_BYPASS] = &q[JQ_BYPASS]->head;
-
-	q[JQ_SUBMIT] = jobqueue_init(sb, fgq + JQ_SUBMIT, forcefg);
-	qtail[JQ_SUBMIT] = &q[JQ_SUBMIT]->head;
+	q[JQ_BYPASS] = jobqueue_init(sb, fgq + JQ_BYPASS, NULL);
+	q[JQ_SUBMIT] = jobqueue_init(sb, fgq + JQ_SUBMIT, fg);
 
-	return tagptr_cast_ptr(tagptr_fold(tagptr1_t, q[JQ_SUBMIT], !forcefg));
+	return tagptr_cast_ptr(tagptr_fold(tagptr1_t, q[JQ_SUBMIT], *fg));
 }
 
 static void move_to_bypass_jobqueue(struct z_erofs_pcluster *pcl,
@@ -1184,55 +1150,33 @@ static void move_to_bypass_jobqueue(struct z_erofs_pcluster *pcl,
 	qtail[JQ_BYPASS] = &pcl->next;
 }
 
-static bool postsubmit_is_all_bypassed(struct z_erofs_unzip_io *q[],
-				       unsigned int nr_bios,
-				       bool force_fg)
+static void z_erofs_submit_queue(struct super_block *sb,
+				 z_erofs_next_pcluster_t owned_head,
+				 struct list_head *pagepool,
+				 struct z_erofs_decompressqueue *fgq,
+				 bool *force_fg)
 {
-	/*
-	 * although background is preferred, no one is pending for submission.
-	 * don't issue workqueue for decompression but drop it directly instead.
-	 */
-	if (force_fg || nr_bios)
-		return false;
-
-	kvfree(container_of(q[JQ_SUBMIT], struct z_erofs_unzip_io_sb, io));
-	return true;
-}
-
-static bool z_erofs_vle_submit_all(struct super_block *sb,
-				   z_erofs_next_pcluster_t owned_head,
-				   struct list_head *pagepool,
-				   struct z_erofs_unzip_io *fgq,
-				   bool force_fg)
-{
-	struct erofs_sb_info *const sbi __maybe_unused = EROFS_SB(sb);
+	struct erofs_sb_info *const sbi = EROFS_SB(sb);
 	z_erofs_next_pcluster_t qtail[NR_JOBQUEUES];
-	struct z_erofs_unzip_io *q[NR_JOBQUEUES];
-	struct bio *bio;
+	struct z_erofs_decompressqueue *q[NR_JOBQUEUES];
 	void *bi_private;
 	/* since bio will be NULL, no need to initialize last_index */
 	pgoff_t uninitialized_var(last_index);
-	bool force_submit = false;
-	unsigned int nr_bios;
+	unsigned int nr_bios = 0;
+	struct bio *bio = NULL;
 
-	if (owned_head == Z_EROFS_PCLUSTER_TAIL)
-		return false;
-
-	force_submit = false;
-	bio = NULL;
-	nr_bios = 0;
-	bi_private = jobqueueset_init(sb, qtail, q, fgq, force_fg);
+	bi_private = jobqueueset_init(sb, q, fgq, force_fg);
+	qtail[JQ_BYPASS] = &q[JQ_BYPASS]->head;
+	qtail[JQ_SUBMIT] = &q[JQ_SUBMIT]->head;
 
 	/* by default, all need io submission */
 	q[JQ_SUBMIT]->head = owned_head;
 
 	do {
 		struct z_erofs_pcluster *pcl;
-		unsigned int clusterpages;
-		pgoff_t first_index;
-		struct page *page;
-		unsigned int i = 0, bypass = 0;
-		int err;
+		pgoff_t cur, end;
+		unsigned int i = 0;
+		bool bypass = true;
 
 		/* no possible 'owned_head' equals the following */
 		DBG_BUGON(owned_head == Z_EROFS_PCLUSTER_TAIL_CLOSED);
@@ -1240,55 +1184,50 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 
 		pcl = container_of(owned_head, struct z_erofs_pcluster, next);
 
-		clusterpages = BIT(pcl->clusterbits);
+		cur = pcl->obj.index;
+		end = cur + BIT(pcl->clusterbits);
 
 		/* close the main owned chain at first */
 		owned_head = cmpxchg(&pcl->next, Z_EROFS_PCLUSTER_TAIL,
 				     Z_EROFS_PCLUSTER_TAIL_CLOSED);
 
-		first_index = pcl->obj.index;
-		force_submit |= (first_index != last_index + 1);
+		do {
+			struct page *page;
+			int err;
 
-repeat:
-		page = pickup_page_for_submission(pcl, i, pagepool,
-						  MNGD_MAPPING(sbi),
-						  GFP_NOFS);
-		if (!page) {
-			force_submit = true;
-			++bypass;
-			goto skippage;
-		}
+			page = pickup_page_for_submission(pcl, i++, pagepool,
+							  MNGD_MAPPING(sbi),
+							  GFP_NOFS);
+			if (!page)
+				continue;
 
-		if (bio && force_submit) {
+			if (bio && cur != last_index + 1) {
 submit_bio_retry:
-			submit_bio(bio);
-			bio = NULL;
-		}
-
-		if (!bio) {
-			bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES);
+				submit_bio(bio);
+				bio = NULL;
+			}
 
-			bio->bi_end_io = z_erofs_vle_read_endio;
-			bio_set_dev(bio, sb->s_bdev);
-			bio->bi_iter.bi_sector = (sector_t)(first_index + i) <<
-				LOG_SECTORS_PER_BLOCK;
-			bio->bi_private = bi_private;
-			bio->bi_opf = REQ_OP_READ;
+			if (!bio) {
+				bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES);
 
-			++nr_bios;
-		}
+				bio->bi_end_io = z_erofs_decompressqueue_endio;
+				bio_set_dev(bio, sb->s_bdev);
+				bio->bi_iter.bi_sector = (sector_t)cur <<
+					LOG_SECTORS_PER_BLOCK;
+				bio->bi_private = bi_private;
+				bio->bi_opf = REQ_OP_READ;
+				++nr_bios;
+			}
 
-		err = bio_add_page(bio, page, PAGE_SIZE, 0);
-		if (err < PAGE_SIZE)
-			goto submit_bio_retry;
+			err = bio_add_page(bio, page, PAGE_SIZE, 0);
+			if (err < PAGE_SIZE)
+				goto submit_bio_retry;
 
-		force_submit = false;
-		last_index = first_index + i;
-skippage:
-		if (++i < clusterpages)
-			goto repeat;
+			last_index = cur;
+			bypass = false;
+		} while (++cur < end);
 
-		if (bypass < clusterpages)
+		if (!bypass)
 			qtail[JQ_SUBMIT] = &pcl->next;
 		else
 			move_to_bypass_jobqueue(pcl, qtail, owned_head);
@@ -1297,40 +1236,42 @@ skippage:
 	if (bio)
 		submit_bio(bio);
 
-	if (postsubmit_is_all_bypassed(q, nr_bios, force_fg))
-		return true;
-
-	z_erofs_vle_unzip_kickoff(bi_private, nr_bios);
-	return true;
+	/*
+	 * although background is preferred, no one is pending for submission.
+	 * don't issue workqueue for decompression but drop it directly instead.
+	 */
+	if (!*force_fg && !nr_bios) {
+		kvfree(q[JQ_SUBMIT]);
+		return;
+	}
+	z_erofs_decompress_kickoff(q[JQ_SUBMIT], *force_fg, nr_bios);
 }
 
-static void z_erofs_submit_and_unzip(struct super_block *sb,
-				     struct z_erofs_collector *clt,
-				     struct list_head *pagepool,
-				     bool force_fg)
+static void z_erofs_runqueue(struct super_block *sb,
+			     struct z_erofs_collector *clt,
+			     struct list_head *pagepool, bool force_fg)
 {
-	struct z_erofs_unzip_io io[NR_JOBQUEUES];
+	struct z_erofs_decompressqueue io[NR_JOBQUEUES];
 
-	if (!z_erofs_vle_submit_all(sb, clt->owned_head,
-				    pagepool, io, force_fg))
+	if (clt->owned_head == Z_EROFS_PCLUSTER_TAIL)
 		return;
+	z_erofs_submit_queue(sb, clt->owned_head, pagepool, io, &force_fg);
 
-	/* decompress no I/O pclusters immediately */
-	z_erofs_vle_unzip_all(sb, &io[JQ_BYPASS], pagepool);
+	/* handle bypass queue (no i/o pclusters) immediately */
+	z_erofs_decompress_queue(&io[JQ_BYPASS], pagepool);
 
 	if (!force_fg)
 		return;
 
 	/* wait until all bios are completed */
-	wait_event(io[JQ_SUBMIT].u.wait,
-		   !atomic_read(&io[JQ_SUBMIT].pending_bios));
+	io_wait_event(io[JQ_SUBMIT].u.wait,
+		      !atomic_read(&io[JQ_SUBMIT].pending_bios));
 
-	/* let's synchronous decompression */
-	z_erofs_vle_unzip_all(sb, &io[JQ_SUBMIT], pagepool);
+	/* handle synchronous decompress queue in the caller context */
+	z_erofs_decompress_queue(&io[JQ_SUBMIT], pagepool);
 }
 
-static int z_erofs_vle_normalaccess_readpage(struct file *file,
-					     struct page *page)
+static int z_erofs_readpage(struct file *file, struct page *page)
 {
 	struct inode *const inode = page->mapping->host;
 	struct z_erofs_decompress_frontend f = DECOMPRESS_FRONTEND_INIT(inode);
@@ -1345,7 +1286,7 @@ static int z_erofs_vle_normalaccess_readpage(struct file *file,
 	(void)z_erofs_collector_end(&f.clt);
 
 	/* if some compressed cluster ready, need submit them anyway */
-	z_erofs_submit_and_unzip(inode->i_sb, &f.clt, &pagepool, true);
+	z_erofs_runqueue(inode->i_sb, &f.clt, &pagepool, true);
 
 	if (err)
 		erofs_err(inode->i_sb, "failed to read, err [%d]", err);
@@ -1364,30 +1305,23 @@ static bool should_decompress_synchronously(struct erofs_sb_info *sbi,
 	return nr <= sbi->max_sync_decompress_pages;
 }
 
-static int z_erofs_vle_normalaccess_readpages(struct file *filp,
-					      struct address_space *mapping,
-					      struct list_head *pages,
-					      unsigned int nr_pages)
+static void z_erofs_readahead(struct readahead_control *rac)
 {
-	struct inode *const inode = mapping->host;
+	struct inode *const inode = rac->mapping->host;
 	struct erofs_sb_info *const sbi = EROFS_I_SB(inode);
 
-	bool sync = should_decompress_synchronously(sbi, nr_pages);
+	bool sync = should_decompress_synchronously(sbi, readahead_count(rac));
 	struct z_erofs_decompress_frontend f = DECOMPRESS_FRONTEND_INIT(inode);
-	gfp_t gfp = mapping_gfp_constraint(mapping, GFP_KERNEL);
-	struct page *head = NULL;
+	struct page *page, *head = NULL;
 	LIST_HEAD(pagepool);
 
-	trace_erofs_readpages(mapping->host, lru_to_page(pages),
-			      nr_pages, false);
+	trace_erofs_readpages(inode, readahead_index(rac),
+			readahead_count(rac), false);
 
-	f.headoffset = (erofs_off_t)lru_to_page(pages)->index << PAGE_SHIFT;
-
-	for (; nr_pages; --nr_pages) {
-		struct page *page = lru_to_page(pages);
+	f.headoffset = readahead_pos(rac);
 
+	while ((page = readahead_page(rac))) {
 		prefetchw(&page->flags);
-		list_del(&page->lru);
 
 		/*
 		 * A pure asynchronous readahead is indicated if
@@ -1396,11 +1330,6 @@ static int z_erofs_vle_normalaccess_readpages(struct file *filp,
 		 */
 		sync &= !(PageReadahead(page) && !head);
 
-		if (add_to_page_cache_lru(page, mapping, page->index, gfp)) {
-			list_add(&page->lru, &pagepool);
-			continue;
-		}
-
 		set_page_private(page, (unsigned long)head);
 		head = page;
 	}
@@ -1422,18 +1351,17 @@ static int z_erofs_vle_normalaccess_readpages(struct file *filp,
 
 	(void)z_erofs_collector_end(&f.clt);
 
-	z_erofs_submit_and_unzip(inode->i_sb, &f.clt, &pagepool, sync);
+	z_erofs_runqueue(inode->i_sb, &f.clt, &pagepool, sync);
 
 	if (f.map.mpage)
 		put_page(f.map.mpage);
 
 	/* clean up the remaining free pages */
 	put_pages_list(&pagepool);
-	return 0;
 }
 
-const struct address_space_operations z_erofs_vle_normalaccess_aops = {
-	.readpage = z_erofs_vle_normalaccess_readpage,
-	.readpages = z_erofs_vle_normalaccess_readpages,
+const struct address_space_operations z_erofs_aops = {
+	.readpage = z_erofs_readpage,
+	.readahead = z_erofs_readahead,
 };
 
diff --git a/fs/erofs/zdata.h b/fs/erofs/zdata.h
index faf950189bd7..7824f5563a55 100644
--- a/fs/erofs/zdata.h
+++ b/fs/erofs/zdata.h
@@ -84,7 +84,8 @@ struct z_erofs_pcluster {
 
 #define Z_EROFS_WORKGROUP_SIZE  sizeof(struct z_erofs_pcluster)
 
-struct z_erofs_unzip_io {
+struct z_erofs_decompressqueue {
+	struct super_block *sb;
 	atomic_t pending_bios;
 	z_erofs_next_pcluster_t head;
 
@@ -94,11 +95,6 @@ struct z_erofs_unzip_io {
 	} u;
 };
 
-struct z_erofs_unzip_io_sb {
-	struct z_erofs_unzip_io io;
-	struct super_block *sb;
-};
-
 #define MNGD_MAPPING(sbi)	((sbi)->managed_cache->i_mapping)
 static inline bool erofs_page_is_managed(const struct erofs_sb_info *sbi,
 					 struct page *page)
diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c
index 6a26c293ae2d..736db3a4cdef 100644
--- a/fs/erofs/zmap.c
+++ b/fs/erofs/zmap.c
@@ -22,11 +22,11 @@ int z_erofs_fill_inode(struct inode *inode)
 		set_bit(EROFS_I_Z_INITED_BIT, &vi->flags);
 	}
 
-	inode->i_mapping->a_ops = &z_erofs_vle_normalaccess_aops;
+	inode->i_mapping->a_ops = &z_erofs_aops;
 	return 0;
 }
 
-static int fill_inode_lazy(struct inode *inode)
+static int z_erofs_fill_inode_lazy(struct inode *inode)
 {
 	struct erofs_inode *const vi = EROFS_I(inode);
 	struct super_block *const sb = inode->i_sb;
@@ -138,8 +138,8 @@ static int z_erofs_reload_indexes(struct z_erofs_maprecorder *m,
 	return 0;
 }
 
-static int vle_legacy_load_cluster_from_disk(struct z_erofs_maprecorder *m,
-					     unsigned long lcn)
+static int legacy_load_cluster_from_disk(struct z_erofs_maprecorder *m,
+					 unsigned long lcn)
 {
 	struct inode *const inode = m->inode;
 	struct erofs_inode *const vi = EROFS_I(inode);
@@ -311,13 +311,13 @@ out:
 	return unpack_compacted_index(m, amortizedshift, erofs_blkoff(pos));
 }
 
-static int vle_load_cluster_from_disk(struct z_erofs_maprecorder *m,
-				      unsigned int lcn)
+static int z_erofs_load_cluster_from_disk(struct z_erofs_maprecorder *m,
+					  unsigned int lcn)
 {
 	const unsigned int datamode = EROFS_I(m->inode)->datalayout;
 
 	if (datamode == EROFS_INODE_FLAT_COMPRESSION_LEGACY)
-		return vle_legacy_load_cluster_from_disk(m, lcn);
+		return legacy_load_cluster_from_disk(m, lcn);
 
 	if (datamode == EROFS_INODE_FLAT_COMPRESSION)
 		return compacted_load_cluster_from_disk(m, lcn);
@@ -325,8 +325,8 @@ static int vle_load_cluster_from_disk(struct z_erofs_maprecorder *m,
 	return -EINVAL;
 }
 
-static int vle_extent_lookback(struct z_erofs_maprecorder *m,
-			       unsigned int lookback_distance)
+static int z_erofs_extent_lookback(struct z_erofs_maprecorder *m,
+				   unsigned int lookback_distance)
 {
 	struct erofs_inode *const vi = EROFS_I(m->inode);
 	struct erofs_map_blocks *const map = m->map;
@@ -343,7 +343,7 @@ static int vle_extent_lookback(struct z_erofs_maprecorder *m,
 
 	/* load extent head logical cluster if needed */
 	lcn -= lookback_distance;
-	err = vle_load_cluster_from_disk(m, lcn);
+	err = z_erofs_load_cluster_from_disk(m, lcn);
 	if (err)
 		return err;
 
@@ -356,7 +356,7 @@ static int vle_extent_lookback(struct z_erofs_maprecorder *m,
 			DBG_BUGON(1);
 			return -EFSCORRUPTED;
 		}
-		return vle_extent_lookback(m, m->delta[0]);
+		return z_erofs_extent_lookback(m, m->delta[0]);
 	case Z_EROFS_VLE_CLUSTER_TYPE_PLAIN:
 		map->m_flags &= ~EROFS_MAP_ZIPPED;
 		/* fallthrough */
@@ -396,7 +396,7 @@ int z_erofs_map_blocks_iter(struct inode *inode,
 		goto out;
 	}
 
-	err = fill_inode_lazy(inode);
+	err = z_erofs_fill_inode_lazy(inode);
 	if (err)
 		goto out;
 
@@ -405,7 +405,7 @@ int z_erofs_map_blocks_iter(struct inode *inode,
 	m.lcn = ofs >> lclusterbits;
 	endoff = ofs & ((1 << lclusterbits) - 1);
 
-	err = vle_load_cluster_from_disk(&m, m.lcn);
+	err = z_erofs_load_cluster_from_disk(&m, m.lcn);
 	if (err)
 		goto unmap_out;
 
@@ -436,7 +436,7 @@ int z_erofs_map_blocks_iter(struct inode *inode,
 		/* fallthrough */
 	case Z_EROFS_VLE_CLUSTER_TYPE_NONHEAD:
 		/* get the correspoinding first chunk */
-		err = vle_extent_lookback(&m, m.delta[0]);
+		err = z_erofs_extent_lookback(&m, m.delta[0]);
 		if (err)
 			goto unmap_out;
 		break;
diff --git a/fs/eventfd.c b/fs/eventfd.c
index 8aa0ea8c55e8..78e41c7c3d05 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -24,6 +24,8 @@
 #include <linux/seq_file.h>
 #include <linux/idr.h>
 
+DEFINE_PER_CPU(int, eventfd_wake_count);
+
 static DEFINE_IDA(eventfd_ida);
 
 struct eventfd_ctx {
@@ -60,12 +62,25 @@ __u64 eventfd_signal(struct eventfd_ctx *ctx, __u64 n)
 {
 	unsigned long flags;
 
+	/*
+	 * Deadlock or stack overflow issues can happen if we recurse here
+	 * through waitqueue wakeup handlers. If the caller users potentially
+	 * nested waitqueues with custom wakeup handlers, then it should
+	 * check eventfd_signal_count() before calling this function. If
+	 * it returns true, the eventfd_signal() call should be deferred to a
+	 * safe context.
+	 */
+	if (WARN_ON_ONCE(this_cpu_read(eventfd_wake_count)))
+		return 0;
+
 	spin_lock_irqsave(&ctx->wqh.lock, flags);
+	this_cpu_inc(eventfd_wake_count);
 	if (ULLONG_MAX - ctx->count < n)
 		n = ULLONG_MAX - ctx->count;
 	ctx->count += n;
 	if (waitqueue_active(&ctx->wqh))
 		wake_up_locked_poll(&ctx->wqh, EPOLLIN);
+	this_cpu_dec(eventfd_wake_count);
 	spin_unlock_irqrestore(&ctx->wqh.lock, flags);
 
 	return n;
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index c4159bcc05d9..12eebcdea9c8 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -218,13 +218,18 @@ struct eventpoll {
 	struct file *file;
 
 	/* used to optimize loop detection check */
-	int visited;
 	struct list_head visited_list_link;
+	int visited;
 
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	/* used to track busy poll napi_id */
 	unsigned int napi_id;
 #endif
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	/* tracks wakeup nests for lockdep validation */
+	u8 nests;
+#endif
 };
 
 /* Wait structure used by the poll hooks */
@@ -354,12 +359,6 @@ static inline struct epitem *ep_item_from_epqueue(poll_table *p)
 	return container_of(p, struct ep_pqueue, pt)->epi;
 }
 
-/* Tells if the epoll_ctl(2) operation needs an event copy from userspace */
-static inline int ep_op_has_event(int op)
-{
-	return op != EPOLL_CTL_DEL;
-}
-
 /* Initialize the poll safe wake up structure */
 static void ep_nested_calls_init(struct nested_calls *ncalls)
 {
@@ -551,35 +550,47 @@ out_unlock:
  */
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 
-static struct nested_calls poll_safewake_ncalls;
-
-static int ep_poll_wakeup_proc(void *priv, void *cookie, int call_nests)
+static void ep_poll_safewake(struct eventpoll *ep, struct epitem *epi)
 {
+	struct eventpoll *ep_src;
 	unsigned long flags;
-	wait_queue_head_t *wqueue = (wait_queue_head_t *)cookie;
-
-	spin_lock_irqsave_nested(&wqueue->lock, flags, call_nests + 1);
-	wake_up_locked_poll(wqueue, EPOLLIN);
-	spin_unlock_irqrestore(&wqueue->lock, flags);
-
-	return 0;
-}
-
-static void ep_poll_safewake(wait_queue_head_t *wq)
-{
-	int this_cpu = get_cpu();
+	u8 nests = 0;
 
-	ep_call_nested(&poll_safewake_ncalls,
-		       ep_poll_wakeup_proc, NULL, wq, (void *) (long) this_cpu);
-
-	put_cpu();
+	/*
+	 * To set the subclass or nesting level for spin_lock_irqsave_nested()
+	 * it might be natural to create a per-cpu nest count. However, since
+	 * we can recurse on ep->poll_wait.lock, and a non-raw spinlock can
+	 * schedule() in the -rt kernel, the per-cpu variable are no longer
+	 * protected. Thus, we are introducing a per eventpoll nest field.
+	 * If we are not being call from ep_poll_callback(), epi is NULL and
+	 * we are at the first level of nesting, 0. Otherwise, we are being
+	 * called from ep_poll_callback() and if a previous wakeup source is
+	 * not an epoll file itself, we are at depth 1 since the wakeup source
+	 * is depth 0. If the wakeup source is a previous epoll file in the
+	 * wakeup chain then we use its nests value and record ours as
+	 * nests + 1. The previous epoll file nests value is stable since its
+	 * already holding its own poll_wait.lock.
+	 */
+	if (epi) {
+		if ((is_file_epoll(epi->ffd.file))) {
+			ep_src = epi->ffd.file->private_data;
+			nests = ep_src->nests;
+		} else {
+			nests = 1;
+		}
+	}
+	spin_lock_irqsave_nested(&ep->poll_wait.lock, flags, nests);
+	ep->nests = nests + 1;
+	wake_up_locked_poll(&ep->poll_wait, EPOLLIN);
+	ep->nests = 0;
+	spin_unlock_irqrestore(&ep->poll_wait.lock, flags);
 }
 
 #else
 
-static void ep_poll_safewake(wait_queue_head_t *wq)
+static void ep_poll_safewake(struct eventpoll *ep, struct epitem *epi)
 {
-	wake_up_poll(wq, EPOLLIN);
+	wake_up_poll(&ep->poll_wait, EPOLLIN);
 }
 
 #endif
@@ -671,7 +682,6 @@ static __poll_t ep_scan_ready_list(struct eventpoll *ep,
 			      void *priv, int depth, bool ep_locked)
 {
 	__poll_t res;
-	int pwake = 0;
 	struct epitem *epi, *nepi;
 	LIST_HEAD(txlist);
 
@@ -738,26 +748,11 @@ static __poll_t ep_scan_ready_list(struct eventpoll *ep,
 	 */
 	list_splice(&txlist, &ep->rdllist);
 	__pm_relax(ep->ws);
-
-	if (!list_empty(&ep->rdllist)) {
-		/*
-		 * Wake up (if active) both the eventpoll wait list and
-		 * the ->poll() wait list (delayed after we release the lock).
-		 */
-		if (waitqueue_active(&ep->wq))
-			wake_up(&ep->wq);
-		if (waitqueue_active(&ep->poll_wait))
-			pwake++;
-	}
 	write_unlock_irq(&ep->lock);
 
 	if (!ep_locked)
 		mutex_unlock(&ep->mtx);
 
-	/* We have to call this outside the lock */
-	if (pwake)
-		ep_poll_safewake(&ep->poll_wait);
-
 	return res;
 }
 
@@ -816,7 +811,7 @@ static void ep_free(struct eventpoll *ep)
 
 	/* We need to release all tasks waiting for these file */
 	if (waitqueue_active(&ep->poll_wait))
-		ep_poll_safewake(&ep->poll_wait);
+		ep_poll_safewake(ep, NULL);
 
 	/*
 	 * We need to lock this because we could be hit by
@@ -1176,6 +1171,10 @@ static inline bool chain_epi_lockless(struct epitem *epi)
 {
 	struct eventpoll *ep = epi->ep;
 
+	/* Fast preliminary check */
+	if (epi->next != EP_UNACTIVE_PTR)
+		return false;
+
 	/* Check that the same epi has not been just chained from another CPU */
 	if (cmpxchg(&epi->next, EP_UNACTIVE_PTR, NULL) != EP_UNACTIVE_PTR)
 		return false;
@@ -1242,16 +1241,12 @@ static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, v
 	 * chained in ep->ovflist and requeued later on.
 	 */
 	if (READ_ONCE(ep->ovflist) != EP_UNACTIVE_PTR) {
-		if (epi->next == EP_UNACTIVE_PTR &&
-		    chain_epi_lockless(epi))
+		if (chain_epi_lockless(epi))
+			ep_pm_stay_awake_rcu(epi);
+	} else if (!ep_is_linked(epi)) {
+		/* In the usual case, add event to ready list. */
+		if (list_add_tail_lockless(&epi->rdllink, &ep->rdllist))
 			ep_pm_stay_awake_rcu(epi);
-		goto out_unlock;
-	}
-
-	/* If this file is already in the ready list we exit soon */
-	if (!ep_is_linked(epi) &&
-	    list_add_tail_lockless(&epi->rdllink, &ep->rdllist)) {
-		ep_pm_stay_awake_rcu(epi);
 	}
 
 	/*
@@ -1285,7 +1280,7 @@ out_unlock:
 
 	/* We have to call this outside the lock */
 	if (pwake)
-		ep_poll_safewake(&ep->poll_wait);
+		ep_poll_safewake(ep, epi);
 
 	if (!(epi->event.events & EPOLLEXCLUSIVE))
 		ewake = 1;
@@ -1589,7 +1584,7 @@ static int ep_insert(struct eventpoll *ep, const struct epoll_event *event,
 
 	/* We have to call this outside the lock */
 	if (pwake)
-		ep_poll_safewake(&ep->poll_wait);
+		ep_poll_safewake(ep, NULL);
 
 	return 0;
 
@@ -1693,7 +1688,7 @@ static int ep_modify(struct eventpoll *ep, struct epitem *epi,
 
 	/* We have to call this outside the lock */
 	if (pwake)
-		ep_poll_safewake(&ep->poll_wait);
+		ep_poll_safewake(ep, NULL);
 
 	return 0;
 }
@@ -1827,7 +1822,6 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
 {
 	int res = 0, eavail, timed_out = 0;
 	u64 slack = 0;
-	bool waiter = false;
 	wait_queue_entry_t wait;
 	ktime_t expires, *to = NULL;
 
@@ -1872,55 +1866,75 @@ fetch_events:
 	 */
 	ep_reset_busy_poll_napi_id(ep);
 
-	/*
-	 * We don't have any available event to return to the caller.  We need
-	 * to sleep here, and we will be woken by ep_poll_callback() when events
-	 * become available.
-	 */
-	if (!waiter) {
-		waiter = true;
-		init_waitqueue_entry(&wait, current);
-
-		spin_lock_irq(&ep->wq.lock);
-		__add_wait_queue_exclusive(&ep->wq, &wait);
-		spin_unlock_irq(&ep->wq.lock);
-	}
+	do {
+		/*
+		 * Internally init_wait() uses autoremove_wake_function(),
+		 * thus wait entry is removed from the wait queue on each
+		 * wakeup. Why it is important? In case of several waiters
+		 * each new wakeup will hit the next waiter, giving it the
+		 * chance to harvest new event. Otherwise wakeup can be
+		 * lost. This is also good performance-wise, because on
+		 * normal wakeup path no need to call __remove_wait_queue()
+		 * explicitly, thus ep->lock is not taken, which halts the
+		 * event delivery.
+		 */
+		init_wait(&wait);
 
-	for (;;) {
+		write_lock_irq(&ep->lock);
 		/*
-		 * We don't want to sleep if the ep_poll_callback() sends us
-		 * a wakeup in between. That's why we set the task state
-		 * to TASK_INTERRUPTIBLE before doing the checks.
+		 * Barrierless variant, waitqueue_active() is called under
+		 * the same lock on wakeup ep_poll_callback() side, so it
+		 * is safe to avoid an explicit barrier.
 		 */
-		set_current_state(TASK_INTERRUPTIBLE);
+		__set_current_state(TASK_INTERRUPTIBLE);
+
 		/*
-		 * Always short-circuit for fatal signals to allow
-		 * threads to make a timely exit without the chance of
-		 * finding more events available and fetching
-		 * repeatedly.
+		 * Do the final check under the lock. ep_scan_ready_list()
+		 * plays with two lists (->rdllist and ->ovflist) and there
+		 * is always a race when both lists are empty for short
+		 * period of time although events are pending, so lock is
+		 * important.
 		 */
-		if (fatal_signal_pending(current)) {
-			res = -EINTR;
-			break;
+		eavail = ep_events_available(ep);
+		if (!eavail) {
+			if (signal_pending(current))
+				res = -EINTR;
+			else
+				__add_wait_queue_exclusive(&ep->wq, &wait);
 		}
+		write_unlock_irq(&ep->lock);
 
-		eavail = ep_events_available(ep);
-		if (eavail)
-			break;
-		if (signal_pending(current)) {
-			res = -EINTR;
+		if (eavail || res)
 			break;
-		}
 
 		if (!schedule_hrtimeout_range(to, slack, HRTIMER_MODE_ABS)) {
 			timed_out = 1;
 			break;
 		}
-	}
+
+		/* We were woken up, thus go and try to harvest some events */
+		eavail = 1;
+
+	} while (0);
 
 	__set_current_state(TASK_RUNNING);
 
+	if (!list_empty_careful(&wait.entry)) {
+		write_lock_irq(&ep->lock);
+		__remove_wait_queue(&ep->wq, &wait);
+		write_unlock_irq(&ep->lock);
+	}
+
 send_events:
+	if (fatal_signal_pending(current)) {
+		/*
+		 * Always short-circuit for fatal signals to allow
+		 * threads to make a timely exit without the chance of
+		 * finding more events available and fetching
+		 * repeatedly.
+		 */
+		res = -EINTR;
+	}
 	/*
 	 * Try to transfer events to user space. In case we get 0 events and
 	 * there's still timeout left over, we go trying again in search of
@@ -1930,12 +1944,6 @@ send_events:
 	    !(res = ep_send_events(ep, events, maxevents)) && !timed_out)
 		goto fetch_events;
 
-	if (waiter) {
-		spin_lock_irq(&ep->wq.lock);
-		__remove_wait_queue(&ep->wq, &wait);
-		spin_unlock_irq(&ep->wq.lock);
-	}
-
 	return res;
 }
 
@@ -2095,27 +2103,28 @@ SYSCALL_DEFINE1(epoll_create, int, size)
 	return do_epoll_create(0);
 }
 
-/*
- * The following function implements the controller interface for
- * the eventpoll file that enables the insertion/removal/change of
- * file descriptors inside the interest set.
- */
-SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
-		struct epoll_event __user *, event)
+static inline int epoll_mutex_lock(struct mutex *mutex, int depth,
+				   bool nonblock)
+{
+	if (!nonblock) {
+		mutex_lock_nested(mutex, depth);
+		return 0;
+	}
+	if (mutex_trylock(mutex))
+		return 0;
+	return -EAGAIN;
+}
+
+int do_epoll_ctl(int epfd, int op, int fd, struct epoll_event *epds,
+		 bool nonblock)
 {
 	int error;
 	int full_check = 0;
 	struct fd f, tf;
 	struct eventpoll *ep;
 	struct epitem *epi;
-	struct epoll_event epds;
 	struct eventpoll *tep = NULL;
 
-	error = -EFAULT;
-	if (ep_op_has_event(op) &&
-	    copy_from_user(&epds, event, sizeof(struct epoll_event)))
-		goto error_return;
-
 	error = -EBADF;
 	f = fdget(epfd);
 	if (!f.file)
@@ -2133,7 +2142,7 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 
 	/* Check if EPOLLWAKEUP is allowed */
 	if (ep_op_has_event(op))
-		ep_take_care_of_epollwakeup(&epds);
+		ep_take_care_of_epollwakeup(epds);
 
 	/*
 	 * We have to check that the file structure underneath the file descriptor
@@ -2149,11 +2158,11 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 	 * so EPOLLEXCLUSIVE is not allowed for a EPOLL_CTL_MOD operation.
 	 * Also, we do not currently supported nested exclusive wakeups.
 	 */
-	if (ep_op_has_event(op) && (epds.events & EPOLLEXCLUSIVE)) {
+	if (ep_op_has_event(op) && (epds->events & EPOLLEXCLUSIVE)) {
 		if (op == EPOLL_CTL_MOD)
 			goto error_tgt_fput;
 		if (op == EPOLL_CTL_ADD && (is_file_epoll(tf.file) ||
-				(epds.events & ~EPOLLEXCLUSIVE_OK_BITS)))
+				(epds->events & ~EPOLLEXCLUSIVE_OK_BITS)))
 			goto error_tgt_fput;
 	}
 
@@ -2178,13 +2187,17 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 	 * deep wakeup paths from forming in parallel through multiple
 	 * EPOLL_CTL_ADD operations.
 	 */
-	mutex_lock_nested(&ep->mtx, 0);
+	error = epoll_mutex_lock(&ep->mtx, 0, nonblock);
+	if (error)
+		goto error_tgt_fput;
 	if (op == EPOLL_CTL_ADD) {
 		if (!list_empty(&f.file->f_ep_links) ||
 						is_file_epoll(tf.file)) {
-			full_check = 1;
 			mutex_unlock(&ep->mtx);
-			mutex_lock(&epmutex);
+			error = epoll_mutex_lock(&epmutex, 0, nonblock);
+			if (error)
+				goto error_tgt_fput;
+			full_check = 1;
 			if (is_file_epoll(tf.file)) {
 				error = -ELOOP;
 				if (ep_loop_check(ep, tf.file) != 0) {
@@ -2194,10 +2207,19 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 			} else
 				list_add(&tf.file->f_tfile_llink,
 							&tfile_check_list);
-			mutex_lock_nested(&ep->mtx, 0);
+			error = epoll_mutex_lock(&ep->mtx, 0, nonblock);
+			if (error) {
+out_del:
+				list_del(&tf.file->f_tfile_llink);
+				goto error_tgt_fput;
+			}
 			if (is_file_epoll(tf.file)) {
 				tep = tf.file->private_data;
-				mutex_lock_nested(&tep->mtx, 1);
+				error = epoll_mutex_lock(&tep->mtx, 1, nonblock);
+				if (error) {
+					mutex_unlock(&ep->mtx);
+					goto out_del;
+				}
 			}
 		}
 	}
@@ -2213,8 +2235,8 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 	switch (op) {
 	case EPOLL_CTL_ADD:
 		if (!epi) {
-			epds.events |= EPOLLERR | EPOLLHUP;
-			error = ep_insert(ep, &epds, tf.file, fd, full_check);
+			epds->events |= EPOLLERR | EPOLLHUP;
+			error = ep_insert(ep, epds, tf.file, fd, full_check);
 		} else
 			error = -EEXIST;
 		if (full_check)
@@ -2229,8 +2251,8 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 	case EPOLL_CTL_MOD:
 		if (epi) {
 			if (!(epi->event.events & EPOLLEXCLUSIVE)) {
-				epds.events |= EPOLLERR | EPOLLHUP;
-				error = ep_modify(ep, epi, &epds);
+				epds->events |= EPOLLERR | EPOLLHUP;
+				error = ep_modify(ep, epi, epds);
 			}
 		} else
 			error = -ENOENT;
@@ -2253,6 +2275,23 @@ error_return:
 }
 
 /*
+ * The following function implements the controller interface for
+ * the eventpoll file that enables the insertion/removal/change of
+ * file descriptors inside the interest set.
+ */
+SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
+		struct epoll_event __user *, event)
+{
+	struct epoll_event epds;
+
+	if (ep_op_has_event(op) &&
+	    copy_from_user(&epds, event, sizeof(struct epoll_event)))
+		return -EFAULT;
+
+	return do_epoll_ctl(epfd, op, fd, &epds, false);
+}
+
+/*
  * Implement the event wait interface for the eventpoll file. It is the kernel
  * part of the user space epoll_wait(2).
  */
@@ -2370,11 +2409,6 @@ static int __init eventpoll_init(void)
 	 */
 	ep_nested_calls_init(&poll_loop_ncalls);
 
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-	/* Initialize the structure used to perform safe poll wait head wake ups */
-	ep_nested_calls_init(&poll_safewake_ncalls);
-#endif
-
 	/*
 	 * We can have many thousands of epitems, so prevent this from
 	 * using an extra cache line on 64-bit (and smaller) CPUs
diff --git a/fs/exec.c b/fs/exec.c
index 555e93c7dec8..2c465119affc 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -59,7 +59,6 @@
 #include <linux/kmod.h>
 #include <linux/fsnotify.h>
 #include <linux/fs_struct.h>
-#include <linux/pipe_fs_i.h>
 #include <linux/oom.h>
 #include <linux/compat.h>
 #include <linux/vmalloc.h>
@@ -273,7 +272,6 @@ static int __bprm_mm_init(struct linux_binprm *bprm)
 		goto err;
 
 	mm->stack_vm = mm->total_vm = 1;
-	arch_bprm_mm_init(mm, vma);
 	up_write(&mm->mmap_sem);
 	bprm->p = vma->vm_end - sizeof(void *);
 	return 0;
@@ -762,6 +760,11 @@ int setup_arg_pages(struct linux_binprm *bprm,
 		goto out_unlock;
 	BUG_ON(prev != vma);
 
+	if (unlikely(vm_flags & VM_EXEC)) {
+		pr_warn_once("process '%pD4' started with executable stack\n",
+			     bprm->file);
+	}
+
 	/* Move stack pages down in memory. */
 	if (stack_shift) {
 		ret = shift_arg_pages(vma, stack_shift);
@@ -982,6 +985,32 @@ int kernel_read_file_from_path(const char *path, void **buf, loff_t *size,
 }
 EXPORT_SYMBOL_GPL(kernel_read_file_from_path);
 
+int kernel_read_file_from_path_initns(const char *path, void **buf,
+				      loff_t *size, loff_t max_size,
+				      enum kernel_read_file_id id)
+{
+	struct file *file;
+	struct path root;
+	int ret;
+
+	if (!path || !*path)
+		return -EINVAL;
+
+	task_lock(&init_task);
+	get_fs_root(init_task.fs, &root);
+	task_unlock(&init_task);
+
+	file = file_open_root(root.dentry, root.mnt, path, O_RDONLY, 0);
+	path_put(&root);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	ret = kernel_read_file(file, buf, size, max_size, id);
+	fput(file);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(kernel_read_file_from_path_initns);
+
 int kernel_read_file_from_fd(int fd, void **buf, loff_t *size, loff_t max_size,
 			     enum kernel_read_file_id id)
 {
@@ -1007,15 +1036,25 @@ ssize_t read_code(struct file *file, unsigned long addr, loff_t pos, size_t len)
 }
 EXPORT_SYMBOL(read_code);
 
+/*
+ * Maps the mm_struct mm into the current task struct.
+ * On success, this function returns with the mutex
+ * exec_update_mutex locked.
+ */
 static int exec_mmap(struct mm_struct *mm)
 {
 	struct task_struct *tsk;
 	struct mm_struct *old_mm, *active_mm;
+	int ret;
 
 	/* Notify parent that we're no longer interested in the old VM */
 	tsk = current;
 	old_mm = current->mm;
-	mm_release(tsk, old_mm);
+	exec_mm_release(tsk, old_mm);
+
+	ret = mutex_lock_killable(&tsk->signal->exec_update_mutex);
+	if (ret)
+		return ret;
 
 	if (old_mm) {
 		sync_mm_rss(old_mm);
@@ -1028,9 +1067,11 @@ static int exec_mmap(struct mm_struct *mm)
 		down_read(&old_mm->mmap_sem);
 		if (unlikely(old_mm->core_state)) {
 			up_read(&old_mm->mmap_sem);
+			mutex_unlock(&tsk->signal->exec_update_mutex);
 			return -EINTR;
 		}
 	}
+
 	task_lock(tsk);
 	active_mm = tsk->active_mm;
 	membarrier_exec_mmap(mm);
@@ -1132,7 +1173,7 @@ static int de_thread(struct task_struct *tsk)
 		 * also take its birthdate (always earlier than our own).
 		 */
 		tsk->start_time = leader->start_time;
-		tsk->real_start_time = leader->real_start_time;
+		tsk->start_boottime = leader->start_boottime;
 
 		BUG_ON(!same_thread_group(leader, tsk));
 		BUG_ON(has_group_leader_pid(tsk));
@@ -1186,10 +1227,22 @@ no_thread_group:
 	/* we have changed execution domain */
 	tsk->exit_signal = SIGCHLD;
 
-#ifdef CONFIG_POSIX_TIMERS
-	exit_itimers(sig);
-	flush_itimer_signals();
-#endif
+	BUG_ON(!thread_group_leader(tsk));
+	return 0;
+
+killed:
+	/* protects against exit_notify() and __exit_signal() */
+	read_lock(&tasklist_lock);
+	sig->group_exit_task = NULL;
+	sig->notify_count = 0;
+	read_unlock(&tasklist_lock);
+	return -EAGAIN;
+}
+
+
+static int unshare_sighand(struct task_struct *me)
+{
+	struct sighand_struct *oldsighand = me->sighand;
 
 	if (refcount_read(&oldsighand->count) != 1) {
 		struct sighand_struct *newsighand;
@@ -1207,23 +1260,13 @@ no_thread_group:
 
 		write_lock_irq(&tasklist_lock);
 		spin_lock(&oldsighand->siglock);
-		rcu_assign_pointer(tsk->sighand, newsighand);
+		rcu_assign_pointer(me->sighand, newsighand);
 		spin_unlock(&oldsighand->siglock);
 		write_unlock_irq(&tasklist_lock);
 
 		__cleanup_sighand(oldsighand);
 	}
-
-	BUG_ON(!thread_group_leader(tsk));
 	return 0;
-
-killed:
-	/* protects against exit_notify() and __exit_signal() */
-	read_lock(&tasklist_lock);
-	sig->group_exit_task = NULL;
-	sig->notify_count = 0;
-	read_unlock(&tasklist_lock);
-	return -EAGAIN;
 }
 
 char *__get_task_comm(char *buf, size_t buf_size, struct task_struct *tsk)
@@ -1257,13 +1300,13 @@ void __set_task_comm(struct task_struct *tsk, const char *buf, bool exec)
  */
 int flush_old_exec(struct linux_binprm * bprm)
 {
+	struct task_struct *me = current;
 	int retval;
 
 	/*
-	 * Make sure we have a private signal table and that
-	 * we are unassociated from the previous thread group.
+	 * Make this the only thread in the thread group.
 	 */
-	retval = de_thread(current);
+	retval = de_thread(me);
 	if (retval)
 		goto out;
 
@@ -1274,6 +1317,8 @@ int flush_old_exec(struct linux_binprm * bprm)
 	 */
 	set_mm_exe_file(bprm->mm, bprm->file);
 
+	would_dump(bprm, bprm->file);
+
 	/*
 	 * Release all of the old mmap stuff
 	 */
@@ -1283,18 +1328,31 @@ int flush_old_exec(struct linux_binprm * bprm)
 		goto out;
 
 	/*
-	 * After clearing bprm->mm (to mark that current is using the
-	 * prepared mm now), we have nothing left of the original
+	 * After setting bprm->called_exec_mmap (to mark that current is
+	 * using the prepared mm now), we have nothing left of the original
 	 * process. If anything from here on returns an error, the check
 	 * in search_binary_handler() will SEGV current.
 	 */
+	bprm->called_exec_mmap = 1;
 	bprm->mm = NULL;
 
+#ifdef CONFIG_POSIX_TIMERS
+	exit_itimers(me->signal);
+	flush_itimer_signals();
+#endif
+
+	/*
+	 * Make the signal table private.
+	 */
+	retval = unshare_sighand(me);
+	if (retval)
+		goto out;
+
 	set_fs(USER_DS);
-	current->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD |
+	me->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD |
 					PF_NOFREEZE | PF_NO_SETAFFINITY);
 	flush_thread();
-	current->personality &= ~bprm->per_clear;
+	me->personality &= ~bprm->per_clear;
 
 	/*
 	 * We have to apply CLOEXEC before we change whether the process is
@@ -1302,7 +1360,7 @@ int flush_old_exec(struct linux_binprm * bprm)
 	 * trying to access the should-be-closed file descriptors of a process
 	 * undergoing exec(2).
 	 */
-	do_close_on_exec(current->files);
+	do_close_on_exec(me->files);
 	return 0;
 
 out:
@@ -1383,7 +1441,7 @@ void setup_new_exec(struct linux_binprm * bprm)
 
 	/* An exec changes our domain. We are no longer part of the thread
 	   group */
-	current->self_exec_id++;
+	WRITE_ONCE(current->self_exec_id, current->self_exec_id + 1);
 	flush_signal_handlers(current, 0);
 }
 EXPORT_SYMBOL(setup_new_exec);
@@ -1421,6 +1479,8 @@ static void free_bprm(struct linux_binprm *bprm)
 {
 	free_arg_pages(bprm);
 	if (bprm->cred) {
+		if (bprm->called_exec_mmap)
+			mutex_unlock(&current->signal->exec_update_mutex);
 		mutex_unlock(&current->signal->cred_guard_mutex);
 		abort_creds(bprm->cred);
 	}
@@ -1470,6 +1530,7 @@ void install_exec_creds(struct linux_binprm *bprm)
 	 * credentials; any time after this it may be unlocked.
 	 */
 	security_bprm_committed_creds(bprm);
+	mutex_unlock(&current->signal->exec_update_mutex);
 	mutex_unlock(&current->signal->cred_guard_mutex);
 }
 EXPORT_SYMBOL(install_exec_creds);
@@ -1661,7 +1722,7 @@ int search_binary_handler(struct linux_binprm *bprm)
 
 		read_lock(&binfmt_lock);
 		put_binfmt(fmt);
-		if (retval < 0 && !bprm->mm) {
+		if (retval < 0 && bprm->called_exec_mmap) {
 			/* we got to flush_old_exec() and failed after it */
 			read_unlock(&binfmt_lock);
 			force_sigsegv(SIGSEGV);
@@ -1817,8 +1878,6 @@ static int __do_execve_file(int fd, struct filename *filename,
 	if (retval < 0)
 		goto out;
 
-	would_dump(bprm, bprm->file);
-
 	retval = exec_binprm(bprm);
 	if (retval < 0)
 		goto out;
diff --git a/fs/exfat/Kconfig b/fs/exfat/Kconfig
new file mode 100644
index 000000000000..2d3636dc5b8c
--- /dev/null
+++ b/fs/exfat/Kconfig
@@ -0,0 +1,21 @@
+# SPDX-License-Identifier: GPL-2.0-or-later
+
+config EXFAT_FS
+	tristate "exFAT filesystem support"
+	select NLS
+	help
+	  This allows you to mount devices formatted with the exFAT file system.
+	  exFAT is typically used on SD-Cards or USB sticks.
+
+	  To compile this as a module, choose M here: the module will be called
+	  exfat.
+
+config EXFAT_DEFAULT_IOCHARSET
+	string "Default iocharset for exFAT"
+	default "utf8"
+	depends on EXFAT_FS
+	help
+	  Set this to the default input/output character set to use for
+	  converting between the encoding is used for user visible filename and
+	  UTF-16 character that exfat filesystem use, and can be overridden with
+	  the "iocharset" mount option for exFAT filesystems.
diff --git a/fs/exfat/Makefile b/fs/exfat/Makefile
new file mode 100644
index 000000000000..ed51926a4971
--- /dev/null
+++ b/fs/exfat/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-or-later
+#
+# Makefile for the linux exFAT filesystem support.
+#
+obj-$(CONFIG_EXFAT_FS) += exfat.o
+
+exfat-y	:= inode.o namei.o dir.o super.o fatent.o cache.o nls.o misc.o \
+	   file.o balloc.o
diff --git a/fs/exfat/balloc.c b/fs/exfat/balloc.c
new file mode 100644
index 000000000000..6774a5a6ded8
--- /dev/null
+++ b/fs/exfat/balloc.c
@@ -0,0 +1,277 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/blkdev.h>
+#include <linux/slab.h>
+#include <linux/buffer_head.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+static const unsigned char free_bit[] = {
+	0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2,/*  0 ~  19*/
+	0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 5, 0, 1, 0, 2, 0, 1, 0, 3,/* 20 ~  39*/
+	0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2,/* 40 ~  59*/
+	0, 1, 0, 6, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4,/* 60 ~  79*/
+	0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 5, 0, 1, 0, 2,/* 80 ~  99*/
+	0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2, 0, 1, 0, 3,/*100 ~ 119*/
+	0, 1, 0, 2, 0, 1, 0, 7, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2,/*120 ~ 139*/
+	0, 1, 0, 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 5,/*140 ~ 159*/
+	0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2,/*160 ~ 179*/
+	0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 6, 0, 1, 0, 2, 0, 1, 0, 3,/*180 ~ 199*/
+	0, 1, 0, 2, 0, 1, 0, 4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2,/*200 ~ 219*/
+	0, 1, 0, 5, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0, 4,/*220 ~ 239*/
+	0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0                /*240 ~ 254*/
+};
+
+static const unsigned char used_bit[] = {
+	0, 1, 1, 2, 1, 2, 2, 3, 1, 2, 2, 3, 2, 3, 3, 4, 1, 2, 2, 3,/*  0 ~  19*/
+	2, 3, 3, 4, 2, 3, 3, 4, 3, 4, 4, 5, 1, 2, 2, 3, 2, 3, 3, 4,/* 20 ~  39*/
+	2, 3, 3, 4, 3, 4, 4, 5, 2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5,/* 40 ~  59*/
+	4, 5, 5, 6, 1, 2, 2, 3, 2, 3, 3, 4, 2, 3, 3, 4, 3, 4, 4, 5,/* 60 ~  79*/
+	2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6, 2, 3, 3, 4,/* 80 ~  99*/
+	3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6, 3, 4, 4, 5, 4, 5, 5, 6,/*100 ~ 119*/
+	4, 5, 5, 6, 5, 6, 6, 7, 1, 2, 2, 3, 2, 3, 3, 4, 2, 3, 3, 4,/*120 ~ 139*/
+	3, 4, 4, 5, 2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6,/*140 ~ 159*/
+	2, 3, 3, 4, 3, 4, 4, 5, 3, 4, 4, 5, 4, 5, 5, 6, 3, 4, 4, 5,/*160 ~ 179*/
+	4, 5, 5, 6, 4, 5, 5, 6, 5, 6, 6, 7, 2, 3, 3, 4, 3, 4, 4, 5,/*180 ~ 199*/
+	3, 4, 4, 5, 4, 5, 5, 6, 3, 4, 4, 5, 4, 5, 5, 6, 4, 5, 5, 6,/*200 ~ 219*/
+	5, 6, 6, 7, 3, 4, 4, 5, 4, 5, 5, 6, 4, 5, 5, 6, 5, 6, 6, 7,/*220 ~ 239*/
+	4, 5, 5, 6, 5, 6, 6, 7, 5, 6, 6, 7, 6, 7, 7, 8             /*240 ~ 255*/
+};
+
+/*
+ *  Allocation Bitmap Management Functions
+ */
+static int exfat_allocate_bitmap(struct super_block *sb,
+		struct exfat_dentry *ep)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	long long map_size;
+	unsigned int i, need_map_size;
+	sector_t sector;
+
+	sbi->map_clu = le32_to_cpu(ep->dentry.bitmap.start_clu);
+	map_size = le64_to_cpu(ep->dentry.bitmap.size);
+	need_map_size = ((EXFAT_DATA_CLUSTER_COUNT(sbi) - 1) / BITS_PER_BYTE)
+		+ 1;
+	if (need_map_size != map_size) {
+		exfat_msg(sb, KERN_ERR,
+				"bogus allocation bitmap size(need : %u, cur : %lld)",
+				need_map_size, map_size);
+		/*
+		 * Only allowed when bogus allocation
+		 * bitmap size is large
+		 */
+		if (need_map_size > map_size)
+			return -EIO;
+	}
+	sbi->map_sectors = ((need_map_size - 1) >>
+			(sb->s_blocksize_bits)) + 1;
+	sbi->vol_amap = kmalloc_array(sbi->map_sectors,
+				sizeof(struct buffer_head *), GFP_KERNEL);
+	if (!sbi->vol_amap)
+		return -ENOMEM;
+
+	sector = exfat_cluster_to_sector(sbi, sbi->map_clu);
+	for (i = 0; i < sbi->map_sectors; i++) {
+		sbi->vol_amap[i] = sb_bread(sb, sector + i);
+		if (!sbi->vol_amap[i]) {
+			/* release all buffers and free vol_amap */
+			int j = 0;
+
+			while (j < i)
+				brelse(sbi->vol_amap[j++]);
+
+			kfree(sbi->vol_amap);
+			sbi->vol_amap = NULL;
+			return -EIO;
+		}
+	}
+
+	return 0;
+}
+
+int exfat_load_bitmap(struct super_block *sb)
+{
+	unsigned int i, type;
+	struct exfat_chain clu;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	exfat_chain_set(&clu, sbi->root_dir, 0, ALLOC_FAT_CHAIN);
+	while (clu.dir != EXFAT_EOF_CLUSTER) {
+		for (i = 0; i < sbi->dentries_per_clu; i++) {
+			struct exfat_dentry *ep;
+			struct buffer_head *bh;
+
+			ep = exfat_get_dentry(sb, &clu, i, &bh, NULL);
+			if (!ep)
+				return -EIO;
+
+			type = exfat_get_entry_type(ep);
+			if (type == TYPE_UNUSED)
+				break;
+			if (type != TYPE_BITMAP)
+				continue;
+			if (ep->dentry.bitmap.flags == 0x0) {
+				int err;
+
+				err = exfat_allocate_bitmap(sb, ep);
+				brelse(bh);
+				return err;
+			}
+			brelse(bh);
+		}
+
+		if (exfat_get_next_cluster(sb, &clu.dir))
+			return -EIO;
+	}
+
+	return -EINVAL;
+}
+
+void exfat_free_bitmap(struct exfat_sb_info *sbi)
+{
+	int i;
+
+	for (i = 0; i < sbi->map_sectors; i++)
+		__brelse(sbi->vol_amap[i]);
+
+	kfree(sbi->vol_amap);
+}
+
+/*
+ * If the value of "clu" is 0, it means cluster 2 which is the first cluster of
+ * the cluster heap.
+ */
+int exfat_set_bitmap(struct inode *inode, unsigned int clu)
+{
+	int i, b;
+	unsigned int ent_idx;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	WARN_ON(clu < EXFAT_FIRST_CLUSTER);
+	ent_idx = CLUSTER_TO_BITMAP_ENT(clu);
+	i = BITMAP_OFFSET_SECTOR_INDEX(sb, ent_idx);
+	b = BITMAP_OFFSET_BIT_IN_SECTOR(sb, ent_idx);
+
+	set_bit_le(b, sbi->vol_amap[i]->b_data);
+	exfat_update_bh(sb, sbi->vol_amap[i], IS_DIRSYNC(inode));
+	return 0;
+}
+
+/*
+ * If the value of "clu" is 0, it means cluster 2 which is the first cluster of
+ * the cluster heap.
+ */
+void exfat_clear_bitmap(struct inode *inode, unsigned int clu)
+{
+	int i, b;
+	unsigned int ent_idx;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_mount_options *opts = &sbi->options;
+
+	WARN_ON(clu < EXFAT_FIRST_CLUSTER);
+	ent_idx = CLUSTER_TO_BITMAP_ENT(clu);
+	i = BITMAP_OFFSET_SECTOR_INDEX(sb, ent_idx);
+	b = BITMAP_OFFSET_BIT_IN_SECTOR(sb, ent_idx);
+
+	clear_bit_le(b, sbi->vol_amap[i]->b_data);
+	exfat_update_bh(sb, sbi->vol_amap[i], IS_DIRSYNC(inode));
+
+	if (opts->discard) {
+		int ret_discard;
+
+		ret_discard = sb_issue_discard(sb,
+			exfat_cluster_to_sector(sbi, clu +
+						EXFAT_RESERVED_CLUSTERS),
+			(1 << sbi->sect_per_clus_bits), GFP_NOFS, 0);
+
+		if (ret_discard == -EOPNOTSUPP) {
+			exfat_msg(sb, KERN_ERR,
+				"discard not supported by device, disabling");
+			opts->discard = 0;
+		}
+	}
+}
+
+/*
+ * If the value of "clu" is 0, it means cluster 2 which is the first cluster of
+ * the cluster heap.
+ */
+unsigned int exfat_find_free_bitmap(struct super_block *sb, unsigned int clu)
+{
+	unsigned int i, map_i, map_b, ent_idx;
+	unsigned int clu_base, clu_free;
+	unsigned char k, clu_mask;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	WARN_ON(clu < EXFAT_FIRST_CLUSTER);
+	ent_idx = CLUSTER_TO_BITMAP_ENT(clu);
+	clu_base = BITMAP_ENT_TO_CLUSTER(ent_idx & ~(BITS_PER_BYTE_MASK));
+	clu_mask = IGNORED_BITS_REMAINED(clu, clu_base);
+
+	map_i = BITMAP_OFFSET_SECTOR_INDEX(sb, ent_idx);
+	map_b = BITMAP_OFFSET_BYTE_IN_SECTOR(sb, ent_idx);
+
+	for (i = EXFAT_FIRST_CLUSTER; i < sbi->num_clusters;
+	     i += BITS_PER_BYTE) {
+		k = *(sbi->vol_amap[map_i]->b_data + map_b);
+		if (clu_mask > 0) {
+			k |= clu_mask;
+			clu_mask = 0;
+		}
+		if (k < 0xFF) {
+			clu_free = clu_base + free_bit[k];
+			if (clu_free < sbi->num_clusters)
+				return clu_free;
+		}
+		clu_base += BITS_PER_BYTE;
+
+		if (++map_b >= sb->s_blocksize ||
+		    clu_base >= sbi->num_clusters) {
+			if (++map_i >= sbi->map_sectors) {
+				clu_base = EXFAT_FIRST_CLUSTER;
+				map_i = 0;
+			}
+			map_b = 0;
+		}
+	}
+
+	return EXFAT_EOF_CLUSTER;
+}
+
+int exfat_count_used_clusters(struct super_block *sb, unsigned int *ret_count)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	unsigned int count = 0;
+	unsigned int i, map_i = 0, map_b = 0;
+	unsigned int total_clus = EXFAT_DATA_CLUSTER_COUNT(sbi);
+	unsigned int last_mask = total_clus & BITS_PER_BYTE_MASK;
+	unsigned char clu_bits;
+	const unsigned char last_bit_mask[] = {0, 0b00000001, 0b00000011,
+		0b00000111, 0b00001111, 0b00011111, 0b00111111, 0b01111111};
+
+	total_clus &= ~last_mask;
+	for (i = 0; i < total_clus; i += BITS_PER_BYTE) {
+		clu_bits = *(sbi->vol_amap[map_i]->b_data + map_b);
+		count += used_bit[clu_bits];
+		if (++map_b >= (unsigned int)sb->s_blocksize) {
+			map_i++;
+			map_b = 0;
+		}
+	}
+
+	if (last_mask) {
+		clu_bits = *(sbi->vol_amap[map_i]->b_data + map_b);
+		clu_bits &= last_bit_mask[last_mask];
+		count += used_bit[clu_bits];
+	}
+
+	*ret_count = count;
+	return 0;
+}
diff --git a/fs/exfat/cache.c b/fs/exfat/cache.c
new file mode 100644
index 000000000000..03d0824fc368
--- /dev/null
+++ b/fs/exfat/cache.c
@@ -0,0 +1,325 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  linux/fs/fat/cache.c
+ *
+ *  Written 1992,1993 by Werner Almesberger
+ *
+ *  Mar 1999. AV. Changed cache, so that it uses the starting cluster instead
+ *	of inode number.
+ *  May 1999. AV. Fixed the bogosity with FAT32 (read "FAT28"). Fscking lusers.
+ *  Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/slab.h>
+#include <asm/unaligned.h>
+#include <linux/buffer_head.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+#define EXFAT_CACHE_VALID	0
+#define EXFAT_MAX_CACHE		16
+
+struct exfat_cache {
+	struct list_head cache_list;
+	unsigned int nr_contig;	/* number of contiguous clusters */
+	unsigned int fcluster;	/* cluster number in the file. */
+	unsigned int dcluster;	/* cluster number on disk. */
+};
+
+struct exfat_cache_id {
+	unsigned int id;
+	unsigned int nr_contig;
+	unsigned int fcluster;
+	unsigned int dcluster;
+};
+
+static struct kmem_cache *exfat_cachep;
+
+static void exfat_cache_init_once(void *c)
+{
+	struct exfat_cache *cache = (struct exfat_cache *)c;
+
+	INIT_LIST_HEAD(&cache->cache_list);
+}
+
+int exfat_cache_init(void)
+{
+	exfat_cachep = kmem_cache_create("exfat_cache",
+				sizeof(struct exfat_cache),
+				0, SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD,
+				exfat_cache_init_once);
+	if (!exfat_cachep)
+		return -ENOMEM;
+	return 0;
+}
+
+void exfat_cache_shutdown(void)
+{
+	if (!exfat_cachep)
+		return;
+	kmem_cache_destroy(exfat_cachep);
+}
+
+void exfat_cache_init_inode(struct inode *inode)
+{
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+
+	spin_lock_init(&ei->cache_lru_lock);
+	ei->nr_caches = 0;
+	ei->cache_valid_id = EXFAT_CACHE_VALID + 1;
+	INIT_LIST_HEAD(&ei->cache_lru);
+}
+
+static inline struct exfat_cache *exfat_cache_alloc(void)
+{
+	return kmem_cache_alloc(exfat_cachep, GFP_NOFS);
+}
+
+static inline void exfat_cache_free(struct exfat_cache *cache)
+{
+	WARN_ON(!list_empty(&cache->cache_list));
+	kmem_cache_free(exfat_cachep, cache);
+}
+
+static inline void exfat_cache_update_lru(struct inode *inode,
+		struct exfat_cache *cache)
+{
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+
+	if (ei->cache_lru.next != &cache->cache_list)
+		list_move(&cache->cache_list, &ei->cache_lru);
+}
+
+static unsigned int exfat_cache_lookup(struct inode *inode,
+		unsigned int fclus, struct exfat_cache_id *cid,
+		unsigned int *cached_fclus, unsigned int *cached_dclus)
+{
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	static struct exfat_cache nohit = { .fcluster = 0, };
+	struct exfat_cache *hit = &nohit, *p;
+	unsigned int offset = EXFAT_EOF_CLUSTER;
+
+	spin_lock(&ei->cache_lru_lock);
+	list_for_each_entry(p, &ei->cache_lru, cache_list) {
+		/* Find the cache of "fclus" or nearest cache. */
+		if (p->fcluster <= fclus && hit->fcluster < p->fcluster) {
+			hit = p;
+			if (hit->fcluster + hit->nr_contig < fclus) {
+				offset = hit->nr_contig;
+			} else {
+				offset = fclus - hit->fcluster;
+				break;
+			}
+		}
+	}
+	if (hit != &nohit) {
+		exfat_cache_update_lru(inode, hit);
+
+		cid->id = ei->cache_valid_id;
+		cid->nr_contig = hit->nr_contig;
+		cid->fcluster = hit->fcluster;
+		cid->dcluster = hit->dcluster;
+		*cached_fclus = cid->fcluster + offset;
+		*cached_dclus = cid->dcluster + offset;
+	}
+	spin_unlock(&ei->cache_lru_lock);
+
+	return offset;
+}
+
+static struct exfat_cache *exfat_cache_merge(struct inode *inode,
+		struct exfat_cache_id *new)
+{
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct exfat_cache *p;
+
+	list_for_each_entry(p, &ei->cache_lru, cache_list) {
+		/* Find the same part as "new" in cluster-chain. */
+		if (p->fcluster == new->fcluster) {
+			if (new->nr_contig > p->nr_contig)
+				p->nr_contig = new->nr_contig;
+			return p;
+		}
+	}
+	return NULL;
+}
+
+static void exfat_cache_add(struct inode *inode,
+		struct exfat_cache_id *new)
+{
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct exfat_cache *cache, *tmp;
+
+	if (new->fcluster == EXFAT_EOF_CLUSTER) /* dummy cache */
+		return;
+
+	spin_lock(&ei->cache_lru_lock);
+	if (new->id != EXFAT_CACHE_VALID &&
+	    new->id != ei->cache_valid_id)
+		goto unlock;	/* this cache was invalidated */
+
+	cache = exfat_cache_merge(inode, new);
+	if (cache == NULL) {
+		if (ei->nr_caches < EXFAT_MAX_CACHE) {
+			ei->nr_caches++;
+			spin_unlock(&ei->cache_lru_lock);
+
+			tmp = exfat_cache_alloc();
+			if (!tmp) {
+				spin_lock(&ei->cache_lru_lock);
+				ei->nr_caches--;
+				spin_unlock(&ei->cache_lru_lock);
+				return;
+			}
+
+			spin_lock(&ei->cache_lru_lock);
+			cache = exfat_cache_merge(inode, new);
+			if (cache != NULL) {
+				ei->nr_caches--;
+				exfat_cache_free(tmp);
+				goto out_update_lru;
+			}
+			cache = tmp;
+		} else {
+			struct list_head *p = ei->cache_lru.prev;
+
+			cache = list_entry(p,
+					struct exfat_cache, cache_list);
+		}
+		cache->fcluster = new->fcluster;
+		cache->dcluster = new->dcluster;
+		cache->nr_contig = new->nr_contig;
+	}
+out_update_lru:
+	exfat_cache_update_lru(inode, cache);
+unlock:
+	spin_unlock(&ei->cache_lru_lock);
+}
+
+/*
+ * Cache invalidation occurs rarely, thus the LRU chain is not updated. It
+ * fixes itself after a while.
+ */
+static void __exfat_cache_inval_inode(struct inode *inode)
+{
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct exfat_cache *cache;
+
+	while (!list_empty(&ei->cache_lru)) {
+		cache = list_entry(ei->cache_lru.next,
+				   struct exfat_cache, cache_list);
+		list_del_init(&cache->cache_list);
+		ei->nr_caches--;
+		exfat_cache_free(cache);
+	}
+	/* Update. The copy of caches before this id is discarded. */
+	ei->cache_valid_id++;
+	if (ei->cache_valid_id == EXFAT_CACHE_VALID)
+		ei->cache_valid_id++;
+}
+
+void exfat_cache_inval_inode(struct inode *inode)
+{
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+
+	spin_lock(&ei->cache_lru_lock);
+	__exfat_cache_inval_inode(inode);
+	spin_unlock(&ei->cache_lru_lock);
+}
+
+static inline int cache_contiguous(struct exfat_cache_id *cid,
+		unsigned int dclus)
+{
+	cid->nr_contig++;
+	return cid->dcluster + cid->nr_contig == dclus;
+}
+
+static inline void cache_init(struct exfat_cache_id *cid,
+		unsigned int fclus, unsigned int dclus)
+{
+	cid->id = EXFAT_CACHE_VALID;
+	cid->fcluster = fclus;
+	cid->dcluster = dclus;
+	cid->nr_contig = 0;
+}
+
+int exfat_get_cluster(struct inode *inode, unsigned int cluster,
+		unsigned int *fclus, unsigned int *dclus,
+		unsigned int *last_dclus, int allow_eof)
+{
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	unsigned int limit = sbi->num_clusters;
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct exfat_cache_id cid;
+	unsigned int content;
+
+	if (ei->start_clu == EXFAT_FREE_CLUSTER) {
+		exfat_fs_error(sb,
+			"invalid access to exfat cache (entry 0x%08x)",
+			ei->start_clu);
+		return -EIO;
+	}
+
+	*fclus = 0;
+	*dclus = ei->start_clu;
+	*last_dclus = *dclus;
+
+	/*
+	 * Don`t use exfat_cache if zero offset or non-cluster allocation
+	 */
+	if (cluster == 0 || *dclus == EXFAT_EOF_CLUSTER)
+		return 0;
+
+	cache_init(&cid, EXFAT_EOF_CLUSTER, EXFAT_EOF_CLUSTER);
+
+	if (exfat_cache_lookup(inode, cluster, &cid, fclus, dclus) ==
+			EXFAT_EOF_CLUSTER) {
+		/*
+		 * dummy, always not contiguous
+		 * This is reinitialized by cache_init(), later.
+		 */
+		WARN_ON(cid.id != EXFAT_CACHE_VALID ||
+			cid.fcluster != EXFAT_EOF_CLUSTER ||
+			cid.dcluster != EXFAT_EOF_CLUSTER ||
+			cid.nr_contig != 0);
+	}
+
+	if (*fclus == cluster)
+		return 0;
+
+	while (*fclus < cluster) {
+		/* prevent the infinite loop of cluster chain */
+		if (*fclus > limit) {
+			exfat_fs_error(sb,
+				"detected the cluster chain loop (i_pos %u)",
+				(*fclus));
+			return -EIO;
+		}
+
+		if (exfat_ent_get(sb, *dclus, &content))
+			return -EIO;
+
+		*last_dclus = *dclus;
+		*dclus = content;
+		(*fclus)++;
+
+		if (content == EXFAT_EOF_CLUSTER) {
+			if (!allow_eof) {
+				exfat_fs_error(sb,
+				       "invalid cluster chain (i_pos %u, last_clus 0x%08x is EOF)",
+				       *fclus, (*last_dclus));
+				return -EIO;
+			}
+
+			break;
+		}
+
+		if (!cache_contiguous(&cid, *dclus))
+			cache_init(&cid, *fclus, *dclus);
+	}
+
+	exfat_cache_add(inode, &cid);
+	return 0;
+}
diff --git a/fs/exfat/dir.c b/fs/exfat/dir.c
new file mode 100644
index 000000000000..4b91afb0f051
--- /dev/null
+++ b/fs/exfat/dir.c
@@ -0,0 +1,1238 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/slab.h>
+#include <linux/bio.h>
+#include <linux/buffer_head.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+static int exfat_extract_uni_name(struct exfat_dentry *ep,
+		unsigned short *uniname)
+{
+	int i, len = 0;
+
+	for (i = 0; i < EXFAT_FILE_NAME_LEN; i++) {
+		*uniname = le16_to_cpu(ep->dentry.name.unicode_0_14[i]);
+		if (*uniname == 0x0)
+			return len;
+		uniname++;
+		len++;
+	}
+
+	*uniname = 0x0;
+	return len;
+
+}
+
+static void exfat_get_uniname_from_ext_entry(struct super_block *sb,
+		struct exfat_chain *p_dir, int entry, unsigned short *uniname)
+{
+	int i;
+	struct exfat_dentry *ep;
+	struct exfat_entry_set_cache *es;
+
+	es = exfat_get_dentry_set(sb, p_dir, entry, ES_ALL_ENTRIES, &ep);
+	if (!es)
+		return;
+
+	if (es->num_entries < 3)
+		goto free_es;
+
+	ep += 2;
+
+	/*
+	 * First entry  : file entry
+	 * Second entry : stream-extension entry
+	 * Third entry  : first file-name entry
+	 * So, the index of first file-name dentry should start from 2.
+	 */
+	for (i = 2; i < es->num_entries; i++, ep++) {
+		/* end of name entry */
+		if (exfat_get_entry_type(ep) != TYPE_EXTEND)
+			goto free_es;
+
+		exfat_extract_uni_name(ep, uniname);
+		uniname += EXFAT_FILE_NAME_LEN;
+	}
+
+free_es:
+	kfree(es);
+}
+
+/* read a directory entry from the opened directory */
+static int exfat_readdir(struct inode *inode, struct exfat_dir_entry *dir_entry)
+{
+	int i, dentries_per_clu, dentries_per_clu_bits = 0;
+	unsigned int type, clu_offset;
+	sector_t sector;
+	struct exfat_chain dir, clu;
+	struct exfat_uni_name uni_name;
+	struct exfat_dentry *ep;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	unsigned int dentry = ei->rwoffset & 0xFFFFFFFF;
+	struct buffer_head *bh;
+
+	/* check if the given file ID is opened */
+	if (ei->type != TYPE_DIR)
+		return -EPERM;
+
+	if (ei->entry == -1)
+		exfat_chain_set(&dir, sbi->root_dir, 0, ALLOC_FAT_CHAIN);
+	else
+		exfat_chain_set(&dir, ei->start_clu,
+			EXFAT_B_TO_CLU(i_size_read(inode), sbi), ei->flags);
+
+	dentries_per_clu = sbi->dentries_per_clu;
+	dentries_per_clu_bits = ilog2(dentries_per_clu);
+
+	clu_offset = dentry >> dentries_per_clu_bits;
+	exfat_chain_dup(&clu, &dir);
+
+	if (clu.flags == ALLOC_NO_FAT_CHAIN) {
+		clu.dir += clu_offset;
+		clu.size -= clu_offset;
+	} else {
+		/* hint_information */
+		if (clu_offset > 0 && ei->hint_bmap.off != EXFAT_EOF_CLUSTER &&
+		    ei->hint_bmap.off > 0 && clu_offset >= ei->hint_bmap.off) {
+			clu_offset -= ei->hint_bmap.off;
+			clu.dir = ei->hint_bmap.clu;
+		}
+
+		while (clu_offset > 0) {
+			if (exfat_get_next_cluster(sb, &(clu.dir)))
+				return -EIO;
+
+			clu_offset--;
+		}
+	}
+
+	while (clu.dir != EXFAT_EOF_CLUSTER) {
+		i = dentry & (dentries_per_clu - 1);
+
+		for ( ; i < dentries_per_clu; i++, dentry++) {
+			ep = exfat_get_dentry(sb, &clu, i, &bh, &sector);
+			if (!ep)
+				return -EIO;
+
+			type = exfat_get_entry_type(ep);
+			if (type == TYPE_UNUSED) {
+				brelse(bh);
+				break;
+			}
+
+			if (type != TYPE_FILE && type != TYPE_DIR) {
+				brelse(bh);
+				continue;
+			}
+
+			dir_entry->attr = le16_to_cpu(ep->dentry.file.attr);
+			exfat_get_entry_time(sbi, &dir_entry->crtime,
+					ep->dentry.file.create_tz,
+					ep->dentry.file.create_time,
+					ep->dentry.file.create_date,
+					ep->dentry.file.create_time_ms);
+			exfat_get_entry_time(sbi, &dir_entry->mtime,
+					ep->dentry.file.modify_tz,
+					ep->dentry.file.modify_time,
+					ep->dentry.file.modify_date,
+					ep->dentry.file.modify_time_ms);
+			exfat_get_entry_time(sbi, &dir_entry->atime,
+					ep->dentry.file.access_tz,
+					ep->dentry.file.access_time,
+					ep->dentry.file.access_date,
+					0);
+
+			*uni_name.name = 0x0;
+			exfat_get_uniname_from_ext_entry(sb, &dir, dentry,
+				uni_name.name);
+			exfat_utf16_to_nls(sb, &uni_name,
+				dir_entry->namebuf.lfn,
+				dir_entry->namebuf.lfnbuf_len);
+			brelse(bh);
+
+			ep = exfat_get_dentry(sb, &clu, i + 1, &bh, NULL);
+			if (!ep)
+				return -EIO;
+			dir_entry->size =
+				le64_to_cpu(ep->dentry.stream.valid_size);
+			brelse(bh);
+
+			ei->hint_bmap.off = dentry >> dentries_per_clu_bits;
+			ei->hint_bmap.clu = clu.dir;
+
+			ei->rwoffset = ++dentry;
+			return 0;
+		}
+
+		if (clu.flags == ALLOC_NO_FAT_CHAIN) {
+			if (--clu.size > 0)
+				clu.dir++;
+			else
+				clu.dir = EXFAT_EOF_CLUSTER;
+		} else {
+			if (exfat_get_next_cluster(sb, &(clu.dir)))
+				return -EIO;
+		}
+	}
+
+	dir_entry->namebuf.lfn[0] = '\0';
+	ei->rwoffset = dentry;
+	return 0;
+}
+
+static void exfat_init_namebuf(struct exfat_dentry_namebuf *nb)
+{
+	nb->lfn = NULL;
+	nb->lfnbuf_len = 0;
+}
+
+static int exfat_alloc_namebuf(struct exfat_dentry_namebuf *nb)
+{
+	nb->lfn = __getname();
+	if (!nb->lfn)
+		return -ENOMEM;
+	nb->lfnbuf_len = MAX_VFSNAME_BUF_SIZE;
+	return 0;
+}
+
+static void exfat_free_namebuf(struct exfat_dentry_namebuf *nb)
+{
+	if (!nb->lfn)
+		return;
+
+	__putname(nb->lfn);
+	exfat_init_namebuf(nb);
+}
+
+/* skip iterating emit_dots when dir is empty */
+#define ITER_POS_FILLED_DOTS    (2)
+static int exfat_iterate(struct file *filp, struct dir_context *ctx)
+{
+	struct inode *inode = filp->f_path.dentry->d_inode;
+	struct super_block *sb = inode->i_sb;
+	struct inode *tmp;
+	struct exfat_dir_entry de;
+	struct exfat_dentry_namebuf *nb = &(de.namebuf);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	unsigned long inum;
+	loff_t cpos, i_pos;
+	int err = 0, fake_offset = 0;
+
+	exfat_init_namebuf(nb);
+	mutex_lock(&EXFAT_SB(sb)->s_lock);
+
+	cpos = ctx->pos;
+	if (!dir_emit_dots(filp, ctx))
+		goto unlock;
+
+	if (ctx->pos == ITER_POS_FILLED_DOTS) {
+		cpos = 0;
+		fake_offset = 1;
+	}
+
+	if (cpos & (DENTRY_SIZE - 1)) {
+		err = -ENOENT;
+		goto unlock;
+	}
+
+	/* name buffer should be allocated before use */
+	err = exfat_alloc_namebuf(nb);
+	if (err)
+		goto unlock;
+get_new:
+	ei->rwoffset = EXFAT_B_TO_DEN(cpos);
+
+	if (cpos >= i_size_read(inode))
+		goto end_of_dir;
+
+	err = exfat_readdir(inode, &de);
+	if (err) {
+		/*
+		 * At least we tried to read a sector.  Move cpos to next sector
+		 * position (should be aligned).
+		 */
+		if (err == -EIO) {
+			cpos += 1 << (sb->s_blocksize_bits);
+			cpos &= ~(sb->s_blocksize - 1);
+		}
+
+		err = -EIO;
+		goto end_of_dir;
+	}
+
+	cpos = EXFAT_DEN_TO_B(ei->rwoffset);
+
+	if (!nb->lfn[0])
+		goto end_of_dir;
+
+	i_pos = ((loff_t)ei->start_clu << 32) |
+		((ei->rwoffset - 1) & 0xffffffff);
+	tmp = exfat_iget(sb, i_pos);
+	if (tmp) {
+		inum = tmp->i_ino;
+		iput(tmp);
+	} else {
+		inum = iunique(sb, EXFAT_ROOT_INO);
+	}
+
+	/*
+	 * Before calling dir_emit(), sb_lock should be released.
+	 * Because page fault can occur in dir_emit() when the size
+	 * of buffer given from user is larger than one page size.
+	 */
+	mutex_unlock(&EXFAT_SB(sb)->s_lock);
+	if (!dir_emit(ctx, nb->lfn, strlen(nb->lfn), inum,
+			(de.attr & ATTR_SUBDIR) ? DT_DIR : DT_REG))
+		goto out_unlocked;
+	mutex_lock(&EXFAT_SB(sb)->s_lock);
+	ctx->pos = cpos;
+	goto get_new;
+
+end_of_dir:
+	if (!cpos && fake_offset)
+		cpos = ITER_POS_FILLED_DOTS;
+	ctx->pos = cpos;
+unlock:
+	mutex_unlock(&EXFAT_SB(sb)->s_lock);
+out_unlocked:
+	/*
+	 * To improve performance, free namebuf after unlock sb_lock.
+	 * If namebuf is not allocated, this function do nothing
+	 */
+	exfat_free_namebuf(nb);
+	return err;
+}
+
+const struct file_operations exfat_dir_operations = {
+	.llseek		= generic_file_llseek,
+	.read		= generic_read_dir,
+	.iterate	= exfat_iterate,
+	.fsync		= generic_file_fsync,
+};
+
+int exfat_alloc_new_dir(struct inode *inode, struct exfat_chain *clu)
+{
+	int ret;
+
+	exfat_chain_set(clu, EXFAT_EOF_CLUSTER, 0, ALLOC_NO_FAT_CHAIN);
+
+	ret = exfat_alloc_cluster(inode, 1, clu);
+	if (ret)
+		return ret;
+
+	return exfat_zeroed_cluster(inode, clu->dir);
+}
+
+int exfat_calc_num_entries(struct exfat_uni_name *p_uniname)
+{
+	int len;
+
+	len = p_uniname->name_len;
+	if (len == 0)
+		return -EINVAL;
+
+	/* 1 file entry + 1 stream entry + name entries */
+	return ((len - 1) / EXFAT_FILE_NAME_LEN + 3);
+}
+
+unsigned int exfat_get_entry_type(struct exfat_dentry *ep)
+{
+	if (ep->type == EXFAT_UNUSED)
+		return TYPE_UNUSED;
+	if (IS_EXFAT_DELETED(ep->type))
+		return TYPE_DELETED;
+	if (ep->type == EXFAT_INVAL)
+		return TYPE_INVALID;
+	if (IS_EXFAT_CRITICAL_PRI(ep->type)) {
+		if (ep->type == EXFAT_BITMAP)
+			return TYPE_BITMAP;
+		if (ep->type == EXFAT_UPCASE)
+			return TYPE_UPCASE;
+		if (ep->type == EXFAT_VOLUME)
+			return TYPE_VOLUME;
+		if (ep->type == EXFAT_FILE) {
+			if (le16_to_cpu(ep->dentry.file.attr) & ATTR_SUBDIR)
+				return TYPE_DIR;
+			return TYPE_FILE;
+		}
+		return TYPE_CRITICAL_PRI;
+	}
+	if (IS_EXFAT_BENIGN_PRI(ep->type)) {
+		if (ep->type == EXFAT_GUID)
+			return TYPE_GUID;
+		if (ep->type == EXFAT_PADDING)
+			return TYPE_PADDING;
+		if (ep->type == EXFAT_ACLTAB)
+			return TYPE_ACLTAB;
+		return TYPE_BENIGN_PRI;
+	}
+	if (IS_EXFAT_CRITICAL_SEC(ep->type)) {
+		if (ep->type == EXFAT_STREAM)
+			return TYPE_STREAM;
+		if (ep->type == EXFAT_NAME)
+			return TYPE_EXTEND;
+		if (ep->type == EXFAT_ACL)
+			return TYPE_ACL;
+		return TYPE_CRITICAL_SEC;
+	}
+	return TYPE_BENIGN_SEC;
+}
+
+static void exfat_set_entry_type(struct exfat_dentry *ep, unsigned int type)
+{
+	if (type == TYPE_UNUSED) {
+		ep->type = EXFAT_UNUSED;
+	} else if (type == TYPE_DELETED) {
+		ep->type &= EXFAT_DELETE;
+	} else if (type == TYPE_STREAM) {
+		ep->type = EXFAT_STREAM;
+	} else if (type == TYPE_EXTEND) {
+		ep->type = EXFAT_NAME;
+	} else if (type == TYPE_BITMAP) {
+		ep->type = EXFAT_BITMAP;
+	} else if (type == TYPE_UPCASE) {
+		ep->type = EXFAT_UPCASE;
+	} else if (type == TYPE_VOLUME) {
+		ep->type = EXFAT_VOLUME;
+	} else if (type == TYPE_DIR) {
+		ep->type = EXFAT_FILE;
+		ep->dentry.file.attr = cpu_to_le16(ATTR_SUBDIR);
+	} else if (type == TYPE_FILE) {
+		ep->type = EXFAT_FILE;
+		ep->dentry.file.attr = cpu_to_le16(ATTR_ARCHIVE);
+	}
+}
+
+static void exfat_init_stream_entry(struct exfat_dentry *ep,
+		unsigned char flags, unsigned int start_clu,
+		unsigned long long size)
+{
+	exfat_set_entry_type(ep, TYPE_STREAM);
+	ep->dentry.stream.flags = flags;
+	ep->dentry.stream.start_clu = cpu_to_le32(start_clu);
+	ep->dentry.stream.valid_size = cpu_to_le64(size);
+	ep->dentry.stream.size = cpu_to_le64(size);
+}
+
+static void exfat_init_name_entry(struct exfat_dentry *ep,
+		unsigned short *uniname)
+{
+	int i;
+
+	exfat_set_entry_type(ep, TYPE_EXTEND);
+	ep->dentry.name.flags = 0x0;
+
+	for (i = 0; i < EXFAT_FILE_NAME_LEN; i++) {
+		ep->dentry.name.unicode_0_14[i] = cpu_to_le16(*uniname);
+		if (*uniname == 0x0)
+			break;
+		uniname++;
+	}
+}
+
+int exfat_init_dir_entry(struct inode *inode, struct exfat_chain *p_dir,
+		int entry, unsigned int type, unsigned int start_clu,
+		unsigned long long size)
+{
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct timespec64 ts = current_time(inode);
+	sector_t sector;
+	struct exfat_dentry *ep;
+	struct buffer_head *bh;
+
+	/*
+	 * We cannot use exfat_get_dentry_set here because file ep is not
+	 * initialized yet.
+	 */
+	ep = exfat_get_dentry(sb, p_dir, entry, &bh, &sector);
+	if (!ep)
+		return -EIO;
+
+	exfat_set_entry_type(ep, type);
+	exfat_set_entry_time(sbi, &ts,
+			&ep->dentry.file.create_tz,
+			&ep->dentry.file.create_time,
+			&ep->dentry.file.create_date,
+			&ep->dentry.file.create_time_ms);
+	exfat_set_entry_time(sbi, &ts,
+			&ep->dentry.file.modify_tz,
+			&ep->dentry.file.modify_time,
+			&ep->dentry.file.modify_date,
+			&ep->dentry.file.modify_time_ms);
+	exfat_set_entry_time(sbi, &ts,
+			&ep->dentry.file.access_tz,
+			&ep->dentry.file.access_time,
+			&ep->dentry.file.access_date,
+			NULL);
+
+	exfat_update_bh(sb, bh, IS_DIRSYNC(inode));
+	brelse(bh);
+
+	ep = exfat_get_dentry(sb, p_dir, entry + 1, &bh, &sector);
+	if (!ep)
+		return -EIO;
+
+	exfat_init_stream_entry(ep,
+		(type == TYPE_FILE) ? ALLOC_FAT_CHAIN : ALLOC_NO_FAT_CHAIN,
+		start_clu, size);
+	exfat_update_bh(sb, bh, IS_DIRSYNC(inode));
+	brelse(bh);
+
+	return 0;
+}
+
+int exfat_update_dir_chksum(struct inode *inode, struct exfat_chain *p_dir,
+		int entry)
+{
+	struct super_block *sb = inode->i_sb;
+	int ret = 0;
+	int i, num_entries;
+	sector_t sector;
+	unsigned short chksum;
+	struct exfat_dentry *ep, *fep;
+	struct buffer_head *fbh, *bh;
+
+	fep = exfat_get_dentry(sb, p_dir, entry, &fbh, &sector);
+	if (!fep)
+		return -EIO;
+
+	num_entries = fep->dentry.file.num_ext + 1;
+	chksum = exfat_calc_chksum_2byte(fep, DENTRY_SIZE, 0, CS_DIR_ENTRY);
+
+	for (i = 1; i < num_entries; i++) {
+		ep = exfat_get_dentry(sb, p_dir, entry + i, &bh, NULL);
+		if (!ep) {
+			ret = -EIO;
+			goto release_fbh;
+		}
+		chksum = exfat_calc_chksum_2byte(ep, DENTRY_SIZE, chksum,
+				CS_DEFAULT);
+		brelse(bh);
+	}
+
+	fep->dentry.file.checksum = cpu_to_le16(chksum);
+	exfat_update_bh(sb, fbh, IS_DIRSYNC(inode));
+release_fbh:
+	brelse(fbh);
+	return ret;
+}
+
+int exfat_init_ext_entry(struct inode *inode, struct exfat_chain *p_dir,
+		int entry, int num_entries, struct exfat_uni_name *p_uniname)
+{
+	struct super_block *sb = inode->i_sb;
+	int i;
+	sector_t sector;
+	unsigned short *uniname = p_uniname->name;
+	struct exfat_dentry *ep;
+	struct buffer_head *bh;
+	int sync = IS_DIRSYNC(inode);
+
+	ep = exfat_get_dentry(sb, p_dir, entry, &bh, &sector);
+	if (!ep)
+		return -EIO;
+
+	ep->dentry.file.num_ext = (unsigned char)(num_entries - 1);
+	exfat_update_bh(sb, bh, sync);
+	brelse(bh);
+
+	ep = exfat_get_dentry(sb, p_dir, entry + 1, &bh, &sector);
+	if (!ep)
+		return -EIO;
+
+	ep->dentry.stream.name_len = p_uniname->name_len;
+	ep->dentry.stream.name_hash = cpu_to_le16(p_uniname->name_hash);
+	exfat_update_bh(sb, bh, sync);
+	brelse(bh);
+
+	for (i = EXFAT_FIRST_CLUSTER; i < num_entries; i++) {
+		ep = exfat_get_dentry(sb, p_dir, entry + i, &bh, &sector);
+		if (!ep)
+			return -EIO;
+
+		exfat_init_name_entry(ep, uniname);
+		exfat_update_bh(sb, bh, sync);
+		brelse(bh);
+		uniname += EXFAT_FILE_NAME_LEN;
+	}
+
+	exfat_update_dir_chksum(inode, p_dir, entry);
+	return 0;
+}
+
+int exfat_remove_entries(struct inode *inode, struct exfat_chain *p_dir,
+		int entry, int order, int num_entries)
+{
+	struct super_block *sb = inode->i_sb;
+	int i;
+	sector_t sector;
+	struct exfat_dentry *ep;
+	struct buffer_head *bh;
+
+	for (i = order; i < num_entries; i++) {
+		ep = exfat_get_dentry(sb, p_dir, entry + i, &bh, &sector);
+		if (!ep)
+			return -EIO;
+
+		exfat_set_entry_type(ep, TYPE_DELETED);
+		exfat_update_bh(sb, bh, IS_DIRSYNC(inode));
+		brelse(bh);
+	}
+
+	return 0;
+}
+
+int exfat_update_dir_chksum_with_entry_set(struct super_block *sb,
+		struct exfat_entry_set_cache *es, int sync)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct buffer_head *bh;
+	sector_t sec = es->sector;
+	unsigned int off = es->offset;
+	int chksum_type = CS_DIR_ENTRY, i, num_entries = es->num_entries;
+	unsigned int buf_off = (off - es->offset);
+	unsigned int remaining_byte_in_sector, copy_entries, clu;
+	unsigned short chksum = 0;
+
+	for (i = 0; i < num_entries; i++) {
+		chksum = exfat_calc_chksum_2byte(&es->entries[i], DENTRY_SIZE,
+			chksum, chksum_type);
+		chksum_type = CS_DEFAULT;
+	}
+
+	es->entries[0].dentry.file.checksum = cpu_to_le16(chksum);
+
+	while (num_entries) {
+		/* write per sector base */
+		remaining_byte_in_sector = (1 << sb->s_blocksize_bits) - off;
+		copy_entries = min_t(int,
+			EXFAT_B_TO_DEN(remaining_byte_in_sector),
+			num_entries);
+		bh = sb_bread(sb, sec);
+		if (!bh)
+			goto err_out;
+		memcpy(bh->b_data + off,
+			(unsigned char *)&es->entries[0] + buf_off,
+			EXFAT_DEN_TO_B(copy_entries));
+		exfat_update_bh(sb, bh, sync);
+		brelse(bh);
+		num_entries -= copy_entries;
+
+		if (num_entries) {
+			/* get next sector */
+			if (exfat_is_last_sector_in_cluster(sbi, sec)) {
+				clu = exfat_sector_to_cluster(sbi, sec);
+				if (es->alloc_flag == ALLOC_NO_FAT_CHAIN)
+					clu++;
+				else if (exfat_get_next_cluster(sb, &clu))
+					goto err_out;
+				sec = exfat_cluster_to_sector(sbi, clu);
+			} else {
+				sec++;
+			}
+			off = 0;
+			buf_off += EXFAT_DEN_TO_B(copy_entries);
+		}
+	}
+
+	return 0;
+err_out:
+	return -EIO;
+}
+
+static int exfat_walk_fat_chain(struct super_block *sb,
+		struct exfat_chain *p_dir, unsigned int byte_offset,
+		unsigned int *clu)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	unsigned int clu_offset;
+	unsigned int cur_clu;
+
+	clu_offset = EXFAT_B_TO_CLU(byte_offset, sbi);
+	cur_clu = p_dir->dir;
+
+	if (p_dir->flags == ALLOC_NO_FAT_CHAIN) {
+		cur_clu += clu_offset;
+	} else {
+		while (clu_offset > 0) {
+			if (exfat_get_next_cluster(sb, &cur_clu))
+				return -EIO;
+			if (cur_clu == EXFAT_EOF_CLUSTER) {
+				exfat_fs_error(sb,
+					"invalid dentry access beyond EOF (clu : %u, eidx : %d)",
+					p_dir->dir,
+					EXFAT_B_TO_DEN(byte_offset));
+				return -EIO;
+			}
+			clu_offset--;
+		}
+	}
+
+	*clu = cur_clu;
+	return 0;
+}
+
+int exfat_find_location(struct super_block *sb, struct exfat_chain *p_dir,
+		int entry, sector_t *sector, int *offset)
+{
+	int ret;
+	unsigned int off, clu = 0;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	off = EXFAT_DEN_TO_B(entry);
+
+	ret = exfat_walk_fat_chain(sb, p_dir, off, &clu);
+	if (ret)
+		return ret;
+
+	/* byte offset in cluster */
+	off = EXFAT_CLU_OFFSET(off, sbi);
+
+	/* byte offset in sector    */
+	*offset = EXFAT_BLK_OFFSET(off, sb);
+
+	/* sector offset in cluster */
+	*sector = EXFAT_B_TO_BLK(off, sb);
+	*sector += exfat_cluster_to_sector(sbi, clu);
+	return 0;
+}
+
+#define EXFAT_MAX_RA_SIZE     (128*1024)
+static int exfat_dir_readahead(struct super_block *sb, sector_t sec)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct buffer_head *bh;
+	unsigned int max_ra_count = EXFAT_MAX_RA_SIZE >> sb->s_blocksize_bits;
+	unsigned int page_ra_count = PAGE_SIZE >> sb->s_blocksize_bits;
+	unsigned int adj_ra_count = max(sbi->sect_per_clus, page_ra_count);
+	unsigned int ra_count = min(adj_ra_count, max_ra_count);
+
+	/* Read-ahead is not required */
+	if (sbi->sect_per_clus == 1)
+		return 0;
+
+	if (sec < sbi->data_start_sector) {
+		exfat_msg(sb, KERN_ERR,
+			"requested sector is invalid(sect:%llu, root:%llu)",
+			(unsigned long long)sec, sbi->data_start_sector);
+		return -EIO;
+	}
+
+	/* Not sector aligned with ra_count, resize ra_count to page size */
+	if ((sec - sbi->data_start_sector) & (ra_count - 1))
+		ra_count = page_ra_count;
+
+	bh = sb_find_get_block(sb, sec);
+	if (!bh || !buffer_uptodate(bh)) {
+		unsigned int i;
+
+		for (i = 0; i < ra_count; i++)
+			sb_breadahead(sb, (sector_t)(sec + i));
+	}
+	brelse(bh);
+	return 0;
+}
+
+struct exfat_dentry *exfat_get_dentry(struct super_block *sb,
+		struct exfat_chain *p_dir, int entry, struct buffer_head **bh,
+		sector_t *sector)
+{
+	unsigned int dentries_per_page = EXFAT_B_TO_DEN(PAGE_SIZE);
+	int off;
+	sector_t sec;
+
+	if (p_dir->dir == DIR_DELETED) {
+		exfat_msg(sb, KERN_ERR, "abnormal access to deleted dentry\n");
+		return NULL;
+	}
+
+	if (exfat_find_location(sb, p_dir, entry, &sec, &off))
+		return NULL;
+
+	if (p_dir->dir != EXFAT_FREE_CLUSTER &&
+			!(entry & (dentries_per_page - 1)))
+		exfat_dir_readahead(sb, sec);
+
+	*bh = sb_bread(sb, sec);
+	if (!*bh)
+		return NULL;
+
+	if (sector)
+		*sector = sec;
+	return (struct exfat_dentry *)((*bh)->b_data + off);
+}
+
+enum exfat_validate_dentry_mode {
+	ES_MODE_STARTED,
+	ES_MODE_GET_FILE_ENTRY,
+	ES_MODE_GET_STRM_ENTRY,
+	ES_MODE_GET_NAME_ENTRY,
+	ES_MODE_GET_CRITICAL_SEC_ENTRY,
+};
+
+static bool exfat_validate_entry(unsigned int type,
+		enum exfat_validate_dentry_mode *mode)
+{
+	if (type == TYPE_UNUSED || type == TYPE_DELETED)
+		return false;
+
+	switch (*mode) {
+	case ES_MODE_STARTED:
+		if  (type != TYPE_FILE && type != TYPE_DIR)
+			return false;
+		*mode = ES_MODE_GET_FILE_ENTRY;
+		return true;
+	case ES_MODE_GET_FILE_ENTRY:
+		if (type != TYPE_STREAM)
+			return false;
+		*mode = ES_MODE_GET_STRM_ENTRY;
+		return true;
+	case ES_MODE_GET_STRM_ENTRY:
+		if (type != TYPE_EXTEND)
+			return false;
+		*mode = ES_MODE_GET_NAME_ENTRY;
+		return true;
+	case ES_MODE_GET_NAME_ENTRY:
+		if (type == TYPE_STREAM)
+			return false;
+		if (type != TYPE_EXTEND) {
+			if (!(type & TYPE_CRITICAL_SEC))
+				return false;
+			*mode = ES_MODE_GET_CRITICAL_SEC_ENTRY;
+		}
+		return true;
+	case ES_MODE_GET_CRITICAL_SEC_ENTRY:
+		if (type == TYPE_EXTEND || type == TYPE_STREAM)
+			return false;
+		if ((type & TYPE_CRITICAL_SEC) != TYPE_CRITICAL_SEC)
+			return false;
+		return true;
+	default:
+		WARN_ON_ONCE(1);
+		return false;
+	}
+}
+
+/*
+ * Returns a set of dentries for a file or dir.
+ *
+ * Note that this is a copy (dump) of dentries so that user should
+ * call write_entry_set() to apply changes made in this entry set
+ * to the real device.
+ *
+ * in:
+ *   sb+p_dir+entry: indicates a file/dir
+ *   type:  specifies how many dentries should be included.
+ * out:
+ *   file_ep: will point the first dentry(= file dentry) on success
+ * return:
+ *   pointer of entry set on success,
+ *   NULL on failure.
+ */
+struct exfat_entry_set_cache *exfat_get_dentry_set(struct super_block *sb,
+		struct exfat_chain *p_dir, int entry, unsigned int type,
+		struct exfat_dentry **file_ep)
+{
+	int ret;
+	unsigned int off, byte_offset, clu = 0;
+	unsigned int entry_type;
+	sector_t sec;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_entry_set_cache *es;
+	struct exfat_dentry *ep, *pos;
+	unsigned char num_entries;
+	enum exfat_validate_dentry_mode mode = ES_MODE_STARTED;
+	struct buffer_head *bh;
+
+	if (p_dir->dir == DIR_DELETED) {
+		exfat_msg(sb, KERN_ERR, "access to deleted dentry\n");
+		return NULL;
+	}
+
+	byte_offset = EXFAT_DEN_TO_B(entry);
+	ret = exfat_walk_fat_chain(sb, p_dir, byte_offset, &clu);
+	if (ret)
+		return NULL;
+
+	/* byte offset in cluster */
+	byte_offset = EXFAT_CLU_OFFSET(byte_offset, sbi);
+
+	/* byte offset in sector */
+	off = EXFAT_BLK_OFFSET(byte_offset, sb);
+
+	/* sector offset in cluster */
+	sec = EXFAT_B_TO_BLK(byte_offset, sb);
+	sec += exfat_cluster_to_sector(sbi, clu);
+
+	bh = sb_bread(sb, sec);
+	if (!bh)
+		return NULL;
+
+	ep = (struct exfat_dentry *)(bh->b_data + off);
+	entry_type = exfat_get_entry_type(ep);
+
+	if (entry_type != TYPE_FILE && entry_type != TYPE_DIR)
+		goto release_bh;
+
+	num_entries = type == ES_ALL_ENTRIES ?
+		ep->dentry.file.num_ext + 1 : type;
+	es = kmalloc(struct_size(es, entries, num_entries), GFP_KERNEL);
+	if (!es)
+		goto release_bh;
+
+	es->num_entries = num_entries;
+	es->sector = sec;
+	es->offset = off;
+	es->alloc_flag = p_dir->flags;
+
+	pos = &es->entries[0];
+
+	while (num_entries) {
+		if (!exfat_validate_entry(exfat_get_entry_type(ep), &mode))
+			goto free_es;
+
+		/* copy dentry */
+		memcpy(pos, ep, sizeof(struct exfat_dentry));
+
+		if (--num_entries == 0)
+			break;
+
+		if (((off + DENTRY_SIZE) & (sb->s_blocksize - 1)) <
+		    (off & (sb->s_blocksize - 1))) {
+			/* get the next sector */
+			if (exfat_is_last_sector_in_cluster(sbi, sec)) {
+				if (es->alloc_flag == ALLOC_NO_FAT_CHAIN)
+					clu++;
+				else if (exfat_get_next_cluster(sb, &clu))
+					goto free_es;
+				sec = exfat_cluster_to_sector(sbi, clu);
+			} else {
+				sec++;
+			}
+
+			brelse(bh);
+			bh = sb_bread(sb, sec);
+			if (!bh)
+				goto free_es;
+			off = 0;
+			ep = (struct exfat_dentry *)bh->b_data;
+		} else {
+			ep++;
+			off += DENTRY_SIZE;
+		}
+		pos++;
+	}
+
+	if (file_ep)
+		*file_ep = &es->entries[0];
+	brelse(bh);
+	return es;
+
+free_es:
+	kfree(es);
+release_bh:
+	brelse(bh);
+	return NULL;
+}
+
+enum {
+	DIRENT_STEP_FILE,
+	DIRENT_STEP_STRM,
+	DIRENT_STEP_NAME,
+	DIRENT_STEP_SECD,
+};
+
+/*
+ * return values:
+ *   >= 0	: return dir entiry position with the name in dir
+ *   -EEXIST	: (root dir, ".") it is the root dir itself
+ *   -ENOENT	: entry with the name does not exist
+ *   -EIO	: I/O error
+ */
+int exfat_find_dir_entry(struct super_block *sb, struct exfat_inode_info *ei,
+		struct exfat_chain *p_dir, struct exfat_uni_name *p_uniname,
+		int num_entries, unsigned int type)
+{
+	int i, rewind = 0, dentry = 0, end_eidx = 0, num_ext = 0, len;
+	int order, step, name_len = 0;
+	int dentries_per_clu, num_empty = 0;
+	unsigned int entry_type;
+	unsigned short *uniname = NULL;
+	struct exfat_chain clu;
+	struct exfat_hint *hint_stat = &ei->hint_stat;
+	struct exfat_hint_femp candi_empty;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	dentries_per_clu = sbi->dentries_per_clu;
+
+	exfat_chain_dup(&clu, p_dir);
+
+	if (hint_stat->eidx) {
+		clu.dir = hint_stat->clu;
+		dentry = hint_stat->eidx;
+		end_eidx = dentry;
+	}
+
+	candi_empty.eidx = EXFAT_HINT_NONE;
+rewind:
+	order = 0;
+	step = DIRENT_STEP_FILE;
+	while (clu.dir != EXFAT_EOF_CLUSTER) {
+		i = dentry & (dentries_per_clu - 1);
+		for (; i < dentries_per_clu; i++, dentry++) {
+			struct exfat_dentry *ep;
+			struct buffer_head *bh;
+
+			if (rewind && dentry == end_eidx)
+				goto not_found;
+
+			ep = exfat_get_dentry(sb, &clu, i, &bh, NULL);
+			if (!ep)
+				return -EIO;
+
+			entry_type = exfat_get_entry_type(ep);
+
+			if (entry_type == TYPE_UNUSED ||
+			    entry_type == TYPE_DELETED) {
+				step = DIRENT_STEP_FILE;
+
+				num_empty++;
+				if (candi_empty.eidx == EXFAT_HINT_NONE &&
+						num_empty == 1) {
+					exfat_chain_set(&candi_empty.cur,
+						clu.dir, clu.size, clu.flags);
+				}
+
+				if (candi_empty.eidx == EXFAT_HINT_NONE &&
+						num_empty >= num_entries) {
+					candi_empty.eidx =
+						dentry - (num_empty - 1);
+					WARN_ON(candi_empty.eidx < 0);
+					candi_empty.count = num_empty;
+
+					if (ei->hint_femp.eidx ==
+							EXFAT_HINT_NONE ||
+						candi_empty.eidx <=
+							 ei->hint_femp.eidx) {
+						memcpy(&ei->hint_femp,
+							&candi_empty,
+							sizeof(candi_empty));
+					}
+				}
+
+				brelse(bh);
+				if (entry_type == TYPE_UNUSED)
+					goto not_found;
+				continue;
+			}
+
+			num_empty = 0;
+			candi_empty.eidx = EXFAT_HINT_NONE;
+
+			if (entry_type == TYPE_FILE || entry_type == TYPE_DIR) {
+				step = DIRENT_STEP_FILE;
+				if (type == TYPE_ALL || type == entry_type) {
+					num_ext = ep->dentry.file.num_ext;
+					step = DIRENT_STEP_STRM;
+				}
+				brelse(bh);
+				continue;
+			}
+
+			if (entry_type == TYPE_STREAM) {
+				unsigned short name_hash;
+
+				if (step != DIRENT_STEP_STRM) {
+					step = DIRENT_STEP_FILE;
+					brelse(bh);
+					continue;
+				}
+				step = DIRENT_STEP_FILE;
+				name_hash = le16_to_cpu(
+						ep->dentry.stream.name_hash);
+				if (p_uniname->name_hash == name_hash &&
+				    p_uniname->name_len ==
+						ep->dentry.stream.name_len) {
+					step = DIRENT_STEP_NAME;
+					order = 1;
+					name_len = 0;
+				}
+				brelse(bh);
+				continue;
+			}
+
+			brelse(bh);
+			if (entry_type == TYPE_EXTEND) {
+				unsigned short entry_uniname[16], unichar;
+
+				if (step != DIRENT_STEP_NAME) {
+					step = DIRENT_STEP_FILE;
+					continue;
+				}
+
+				if (++order == 2)
+					uniname = p_uniname->name;
+				else
+					uniname += EXFAT_FILE_NAME_LEN;
+
+				len = exfat_extract_uni_name(ep, entry_uniname);
+				name_len += len;
+
+				unichar = *(uniname+len);
+				*(uniname+len) = 0x0;
+
+				if (exfat_uniname_ncmp(sb, uniname,
+					entry_uniname, len)) {
+					step = DIRENT_STEP_FILE;
+				} else if (p_uniname->name_len == name_len) {
+					if (order == num_ext)
+						goto found;
+					step = DIRENT_STEP_SECD;
+				}
+
+				*(uniname+len) = unichar;
+				continue;
+			}
+
+			if (entry_type &
+					(TYPE_CRITICAL_SEC | TYPE_BENIGN_SEC)) {
+				if (step == DIRENT_STEP_SECD) {
+					if (++order == num_ext)
+						goto found;
+					continue;
+				}
+			}
+			step = DIRENT_STEP_FILE;
+		}
+
+		if (clu.flags == ALLOC_NO_FAT_CHAIN) {
+			if (--clu.size > 0)
+				clu.dir++;
+			else
+				clu.dir = EXFAT_EOF_CLUSTER;
+		} else {
+			if (exfat_get_next_cluster(sb, &clu.dir))
+				return -EIO;
+		}
+	}
+
+not_found:
+	/*
+	 * We started at not 0 index,so we should try to find target
+	 * from 0 index to the index we started at.
+	 */
+	if (!rewind && end_eidx) {
+		rewind = 1;
+		dentry = 0;
+		clu.dir = p_dir->dir;
+		/* reset empty hint */
+		num_empty = 0;
+		candi_empty.eidx = EXFAT_HINT_NONE;
+		goto rewind;
+	}
+
+	/* initialized hint_stat */
+	hint_stat->clu = p_dir->dir;
+	hint_stat->eidx = 0;
+	return -ENOENT;
+
+found:
+	/* next dentry we'll find is out of this cluster */
+	if (!((dentry + 1) & (dentries_per_clu - 1))) {
+		int ret = 0;
+
+		if (clu.flags == ALLOC_NO_FAT_CHAIN) {
+			if (--clu.size > 0)
+				clu.dir++;
+			else
+				clu.dir = EXFAT_EOF_CLUSTER;
+		} else {
+			ret = exfat_get_next_cluster(sb, &clu.dir);
+		}
+
+		if (ret || clu.dir != EXFAT_EOF_CLUSTER) {
+			/* just initialized hint_stat */
+			hint_stat->clu = p_dir->dir;
+			hint_stat->eidx = 0;
+			return (dentry - num_ext);
+		}
+	}
+
+	hint_stat->clu = clu.dir;
+	hint_stat->eidx = dentry + 1;
+	return dentry - num_ext;
+}
+
+int exfat_count_ext_entries(struct super_block *sb, struct exfat_chain *p_dir,
+		int entry, struct exfat_dentry *ep)
+{
+	int i, count = 0;
+	unsigned int type;
+	struct exfat_dentry *ext_ep;
+	struct buffer_head *bh;
+
+	for (i = 0, entry++; i < ep->dentry.file.num_ext; i++, entry++) {
+		ext_ep = exfat_get_dentry(sb, p_dir, entry, &bh, NULL);
+		if (!ext_ep)
+			return -EIO;
+
+		type = exfat_get_entry_type(ext_ep);
+		brelse(bh);
+		if (type == TYPE_EXTEND || type == TYPE_STREAM)
+			count++;
+		else
+			break;
+	}
+	return count;
+}
+
+int exfat_count_dir_entries(struct super_block *sb, struct exfat_chain *p_dir)
+{
+	int i, count = 0;
+	int dentries_per_clu;
+	unsigned int entry_type;
+	struct exfat_chain clu;
+	struct exfat_dentry *ep;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct buffer_head *bh;
+
+	dentries_per_clu = sbi->dentries_per_clu;
+
+	exfat_chain_dup(&clu, p_dir);
+
+	while (clu.dir != EXFAT_EOF_CLUSTER) {
+		for (i = 0; i < dentries_per_clu; i++) {
+			ep = exfat_get_dentry(sb, &clu, i, &bh, NULL);
+			if (!ep)
+				return -EIO;
+			entry_type = exfat_get_entry_type(ep);
+			brelse(bh);
+
+			if (entry_type == TYPE_UNUSED)
+				return count;
+			if (entry_type != TYPE_DIR)
+				continue;
+			count++;
+		}
+
+		if (clu.flags == ALLOC_NO_FAT_CHAIN) {
+			if (--clu.size > 0)
+				clu.dir++;
+			else
+				clu.dir = EXFAT_EOF_CLUSTER;
+		} else {
+			if (exfat_get_next_cluster(sb, &(clu.dir)))
+				return -EIO;
+		}
+	}
+
+	return count;
+}
diff --git a/fs/exfat/exfat_fs.h b/fs/exfat/exfat_fs.h
new file mode 100644
index 000000000000..d67fb8a6f770
--- /dev/null
+++ b/fs/exfat/exfat_fs.h
@@ -0,0 +1,520 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#ifndef _EXFAT_FS_H
+#define _EXFAT_FS_H
+
+#include <linux/fs.h>
+#include <linux/ratelimit.h>
+#include <linux/nls.h>
+
+#define EXFAT_SUPER_MAGIC       0x2011BAB0UL
+#define EXFAT_ROOT_INO		1
+
+#define EXFAT_SB_DIRTY		0
+
+#define EXFAT_CLUSTERS_UNTRACKED (~0u)
+
+/*
+ * exfat error flags
+ */
+enum exfat_error_mode {
+	EXFAT_ERRORS_CONT,	/* ignore error and continue */
+	EXFAT_ERRORS_PANIC,	/* panic on error */
+	EXFAT_ERRORS_RO,	/* remount r/o on error */
+};
+
+/*
+ * exfat nls lossy flag
+ */
+enum {
+	NLS_NAME_NO_LOSSY,	/* no lossy */
+	NLS_NAME_LOSSY,		/* just detected incorrect filename(s) */
+	NLS_NAME_OVERLEN,	/* the length is over than its limit */
+};
+
+#define EXFAT_HASH_BITS		8
+#define EXFAT_HASH_SIZE		(1UL << EXFAT_HASH_BITS)
+
+/*
+ * Type Definitions
+ */
+#define ES_2_ENTRIES		2
+#define ES_ALL_ENTRIES		0
+
+#define DIR_DELETED		0xFFFF0321
+
+/* type values */
+#define TYPE_UNUSED		0x0000
+#define TYPE_DELETED		0x0001
+#define TYPE_INVALID		0x0002
+#define TYPE_CRITICAL_PRI	0x0100
+#define TYPE_BITMAP		0x0101
+#define TYPE_UPCASE		0x0102
+#define TYPE_VOLUME		0x0103
+#define TYPE_DIR		0x0104
+#define TYPE_FILE		0x011F
+#define TYPE_CRITICAL_SEC	0x0200
+#define TYPE_STREAM		0x0201
+#define TYPE_EXTEND		0x0202
+#define TYPE_ACL		0x0203
+#define TYPE_BENIGN_PRI		0x0400
+#define TYPE_GUID		0x0401
+#define TYPE_PADDING		0x0402
+#define TYPE_ACLTAB		0x0403
+#define TYPE_BENIGN_SEC		0x0800
+#define TYPE_ALL		0x0FFF
+
+#define MAX_CHARSET_SIZE	6 /* max size of multi-byte character */
+#define MAX_NAME_LENGTH		255 /* max len of file name excluding NULL */
+#define MAX_VFSNAME_BUF_SIZE	((MAX_NAME_LENGTH + 1) * MAX_CHARSET_SIZE)
+
+#define FAT_CACHE_SIZE		128
+#define FAT_CACHE_HASH_SIZE	64
+#define BUF_CACHE_SIZE		256
+#define BUF_CACHE_HASH_SIZE	64
+
+#define EXFAT_HINT_NONE		-1
+#define EXFAT_MIN_SUBDIR	2
+
+/*
+ * helpers for cluster size to byte conversion.
+ */
+#define EXFAT_CLU_TO_B(b, sbi)		((b) << (sbi)->cluster_size_bits)
+#define EXFAT_B_TO_CLU(b, sbi)		((b) >> (sbi)->cluster_size_bits)
+#define EXFAT_B_TO_CLU_ROUND_UP(b, sbi)	\
+	(((b - 1) >> (sbi)->cluster_size_bits) + 1)
+#define EXFAT_CLU_OFFSET(off, sbi)	((off) & ((sbi)->cluster_size - 1))
+
+/*
+ * helpers for block size to byte conversion.
+ */
+#define EXFAT_BLK_TO_B(b, sb)		((b) << (sb)->s_blocksize_bits)
+#define EXFAT_B_TO_BLK(b, sb)		((b) >> (sb)->s_blocksize_bits)
+#define EXFAT_B_TO_BLK_ROUND_UP(b, sb)	\
+	(((b - 1) >> (sb)->s_blocksize_bits) + 1)
+#define EXFAT_BLK_OFFSET(off, sb)	((off) & ((sb)->s_blocksize - 1))
+
+/*
+ * helpers for block size to dentry size conversion.
+ */
+#define EXFAT_B_TO_DEN_IDX(b, sbi)	\
+	((b) << ((sbi)->cluster_size_bits - DENTRY_SIZE_BITS))
+#define EXFAT_B_TO_DEN(b)		((b) >> DENTRY_SIZE_BITS)
+#define EXFAT_DEN_TO_B(b)		((b) << DENTRY_SIZE_BITS)
+
+/*
+ * helpers for fat entry.
+ */
+#define FAT_ENT_SIZE (4)
+#define FAT_ENT_SIZE_BITS (2)
+#define FAT_ENT_OFFSET_SECTOR(sb, loc) (EXFAT_SB(sb)->FAT1_start_sector + \
+	(((u64)loc << FAT_ENT_SIZE_BITS) >> sb->s_blocksize_bits))
+#define FAT_ENT_OFFSET_BYTE_IN_SECTOR(sb, loc)	\
+	((loc << FAT_ENT_SIZE_BITS) & (sb->s_blocksize - 1))
+
+/*
+ * helpers for bitmap.
+ */
+#define CLUSTER_TO_BITMAP_ENT(clu) ((clu) - EXFAT_RESERVED_CLUSTERS)
+#define BITMAP_ENT_TO_CLUSTER(ent) ((ent) + EXFAT_RESERVED_CLUSTERS)
+#define BITS_PER_SECTOR(sb) ((sb)->s_blocksize * BITS_PER_BYTE)
+#define BITS_PER_SECTOR_MASK(sb) (BITS_PER_SECTOR(sb) - 1)
+#define BITMAP_OFFSET_SECTOR_INDEX(sb, ent) \
+	((ent / BITS_PER_BYTE) >> (sb)->s_blocksize_bits)
+#define BITMAP_OFFSET_BIT_IN_SECTOR(sb, ent) (ent & BITS_PER_SECTOR_MASK(sb))
+#define BITMAP_OFFSET_BYTE_IN_SECTOR(sb, ent) \
+	((ent / BITS_PER_BYTE) & ((sb)->s_blocksize - 1))
+#define BITS_PER_BYTE_MASK	0x7
+#define IGNORED_BITS_REMAINED(clu, clu_base) ((1 << ((clu) - (clu_base))) - 1)
+
+struct exfat_dentry_namebuf {
+	char *lfn;
+	int lfnbuf_len; /* usally MAX_UNINAME_BUF_SIZE */
+};
+
+/* unicode name structure */
+struct exfat_uni_name {
+	/* +3 for null and for converting */
+	unsigned short name[MAX_NAME_LENGTH + 3];
+	unsigned short name_hash;
+	unsigned char name_len;
+};
+
+/* directory structure */
+struct exfat_chain {
+	unsigned int dir;
+	unsigned int size;
+	unsigned char flags;
+};
+
+/* first empty entry hint information */
+struct exfat_hint_femp {
+	/* entry index of a directory */
+	int eidx;
+	/* count of continuous empty entry */
+	int count;
+	/* the cluster that first empty slot exists in */
+	struct exfat_chain cur;
+};
+
+/* hint structure */
+struct exfat_hint {
+	unsigned int clu;
+	union {
+		unsigned int off; /* cluster offset */
+		int eidx; /* entry index */
+	};
+};
+
+struct exfat_entry_set_cache {
+	/* sector number that contains file_entry */
+	sector_t sector;
+	/* byte offset in the sector */
+	unsigned int offset;
+	/* flag in stream entry. 01 for cluster chain, 03 for contig. */
+	int alloc_flag;
+	unsigned int num_entries;
+	struct exfat_dentry entries[];
+};
+
+struct exfat_dir_entry {
+	struct exfat_chain dir;
+	int entry;
+	unsigned int type;
+	unsigned int start_clu;
+	unsigned char flags;
+	unsigned short attr;
+	loff_t size;
+	unsigned int num_subdirs;
+	struct timespec64 atime;
+	struct timespec64 mtime;
+	struct timespec64 crtime;
+	struct exfat_dentry_namebuf namebuf;
+};
+
+/*
+ * exfat mount in-memory data
+ */
+struct exfat_mount_options {
+	kuid_t fs_uid;
+	kgid_t fs_gid;
+	unsigned short fs_fmask;
+	unsigned short fs_dmask;
+	/* permission for setting the [am]time */
+	unsigned short allow_utime;
+	/* charset for filename input/display */
+	char *iocharset;
+	/* on error: continue, panic, remount-ro */
+	enum exfat_error_mode errors;
+	unsigned utf8:1, /* Use of UTF-8 character set */
+		 discard:1; /* Issue discard requests on deletions */
+	int time_offset; /* Offset of timestamps from UTC (in minutes) */
+};
+
+/*
+ * EXFAT file system superblock in-memory data
+ */
+struct exfat_sb_info {
+	unsigned long long num_sectors; /* num of sectors in volume */
+	unsigned int num_clusters; /* num of clusters in volume */
+	unsigned int cluster_size; /* cluster size in bytes */
+	unsigned int cluster_size_bits;
+	unsigned int sect_per_clus; /* cluster size in sectors */
+	unsigned int sect_per_clus_bits;
+	unsigned long long FAT1_start_sector; /* FAT1 start sector */
+	unsigned long long FAT2_start_sector; /* FAT2 start sector */
+	unsigned long long data_start_sector; /* data area start sector */
+	unsigned int num_FAT_sectors; /* num of FAT sectors */
+	unsigned int root_dir; /* root dir cluster */
+	unsigned int dentries_per_clu; /* num of dentries per cluster */
+	unsigned int vol_flag; /* volume dirty flag */
+	struct buffer_head *pbr_bh; /* buffer_head of PBR sector */
+
+	unsigned int map_clu; /* allocation bitmap start cluster */
+	unsigned int map_sectors; /* num of allocation bitmap sectors */
+	struct buffer_head **vol_amap; /* allocation bitmap */
+
+	unsigned short *vol_utbl; /* upcase table */
+
+	unsigned int clu_srch_ptr; /* cluster search pointer */
+	unsigned int used_clusters; /* number of used clusters */
+
+	unsigned long s_state;
+	struct mutex s_lock; /* superblock lock */
+	struct exfat_mount_options options;
+	struct nls_table *nls_io; /* Charset used for input and display */
+	struct ratelimit_state ratelimit;
+
+	spinlock_t inode_hash_lock;
+	struct hlist_head inode_hashtable[EXFAT_HASH_SIZE];
+
+	struct rcu_head rcu;
+};
+
+/*
+ * EXFAT file system inode in-memory data
+ */
+struct exfat_inode_info {
+	struct exfat_chain dir;
+	int entry;
+	unsigned int type;
+	unsigned short attr;
+	unsigned int start_clu;
+	unsigned char flags;
+	/*
+	 * the copy of low 32bit of i_version to check
+	 * the validation of hint_stat.
+	 */
+	unsigned int version;
+	/* file offset or dentry index for readdir */
+	loff_t rwoffset;
+
+	/* hint for cluster last accessed */
+	struct exfat_hint hint_bmap;
+	/* hint for entry index we try to lookup next time */
+	struct exfat_hint hint_stat;
+	/* hint for first empty entry */
+	struct exfat_hint_femp hint_femp;
+
+	spinlock_t cache_lru_lock;
+	struct list_head cache_lru;
+	int nr_caches;
+	/* for avoiding the race between alloc and free */
+	unsigned int cache_valid_id;
+
+	/*
+	 * NOTE: i_size_ondisk is 64bits, so must hold ->inode_lock to access.
+	 * physically allocated size.
+	 */
+	loff_t i_size_ondisk;
+	/* block-aligned i_size (used in cont_write_begin) */
+	loff_t i_size_aligned;
+	/* on-disk position of directory entry or 0 */
+	loff_t i_pos;
+	/* hash by i_location */
+	struct hlist_node i_hash_fat;
+	/* protect bmap against truncate */
+	struct rw_semaphore truncate_lock;
+	struct inode vfs_inode;
+	/* File creation time */
+	struct timespec64 i_crtime;
+};
+
+static inline struct exfat_sb_info *EXFAT_SB(struct super_block *sb)
+{
+	return sb->s_fs_info;
+}
+
+static inline struct exfat_inode_info *EXFAT_I(struct inode *inode)
+{
+	return container_of(inode, struct exfat_inode_info, vfs_inode);
+}
+
+/*
+ * If ->i_mode can't hold 0222 (i.e. ATTR_RO), we use ->i_attrs to
+ * save ATTR_RO instead of ->i_mode.
+ *
+ * If it's directory and !sbi->options.rodir, ATTR_RO isn't read-only
+ * bit, it's just used as flag for app.
+ */
+static inline int exfat_mode_can_hold_ro(struct inode *inode)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(inode->i_sb);
+
+	if (S_ISDIR(inode->i_mode))
+		return 0;
+
+	if ((~sbi->options.fs_fmask) & 0222)
+		return 1;
+	return 0;
+}
+
+/* Convert attribute bits and a mask to the UNIX mode. */
+static inline mode_t exfat_make_mode(struct exfat_sb_info *sbi,
+		unsigned short attr, mode_t mode)
+{
+	if ((attr & ATTR_READONLY) && !(attr & ATTR_SUBDIR))
+		mode &= ~0222;
+
+	if (attr & ATTR_SUBDIR)
+		return (mode & ~sbi->options.fs_dmask) | S_IFDIR;
+
+	return (mode & ~sbi->options.fs_fmask) | S_IFREG;
+}
+
+/* Return the FAT attribute byte for this inode */
+static inline unsigned short exfat_make_attr(struct inode *inode)
+{
+	unsigned short attr = EXFAT_I(inode)->attr;
+
+	if (S_ISDIR(inode->i_mode))
+		attr |= ATTR_SUBDIR;
+	if (exfat_mode_can_hold_ro(inode) && !(inode->i_mode & 0222))
+		attr |= ATTR_READONLY;
+	return attr;
+}
+
+static inline void exfat_save_attr(struct inode *inode, unsigned short attr)
+{
+	if (exfat_mode_can_hold_ro(inode))
+		EXFAT_I(inode)->attr = attr & (ATTR_RWMASK | ATTR_READONLY);
+	else
+		EXFAT_I(inode)->attr = attr & ATTR_RWMASK;
+}
+
+static inline bool exfat_is_last_sector_in_cluster(struct exfat_sb_info *sbi,
+		sector_t sec)
+{
+	return ((sec - sbi->data_start_sector + 1) &
+		((1 << sbi->sect_per_clus_bits) - 1)) == 0;
+}
+
+static inline sector_t exfat_cluster_to_sector(struct exfat_sb_info *sbi,
+		unsigned int clus)
+{
+	return ((clus - EXFAT_RESERVED_CLUSTERS) << sbi->sect_per_clus_bits) +
+		sbi->data_start_sector;
+}
+
+static inline int exfat_sector_to_cluster(struct exfat_sb_info *sbi,
+		sector_t sec)
+{
+	return ((sec - sbi->data_start_sector) >> sbi->sect_per_clus_bits) +
+		EXFAT_RESERVED_CLUSTERS;
+}
+
+/* super.c */
+int exfat_set_vol_flags(struct super_block *sb, unsigned short new_flag);
+
+/* fatent.c */
+#define exfat_get_next_cluster(sb, pclu) exfat_ent_get(sb, *(pclu), pclu)
+
+int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc,
+		struct exfat_chain *p_chain);
+int exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain);
+int exfat_ent_get(struct super_block *sb, unsigned int loc,
+		unsigned int *content);
+int exfat_ent_set(struct super_block *sb, unsigned int loc,
+		unsigned int content);
+int exfat_count_ext_entries(struct super_block *sb, struct exfat_chain *p_dir,
+		int entry, struct exfat_dentry *p_entry);
+int exfat_chain_cont_cluster(struct super_block *sb, unsigned int chain,
+		unsigned int len);
+int exfat_zeroed_cluster(struct inode *dir, unsigned int clu);
+int exfat_find_last_cluster(struct super_block *sb, struct exfat_chain *p_chain,
+		unsigned int *ret_clu);
+int exfat_count_num_clusters(struct super_block *sb,
+		struct exfat_chain *p_chain, unsigned int *ret_count);
+
+/* balloc.c */
+int exfat_load_bitmap(struct super_block *sb);
+void exfat_free_bitmap(struct exfat_sb_info *sbi);
+int exfat_set_bitmap(struct inode *inode, unsigned int clu);
+void exfat_clear_bitmap(struct inode *inode, unsigned int clu);
+unsigned int exfat_find_free_bitmap(struct super_block *sb, unsigned int clu);
+int exfat_count_used_clusters(struct super_block *sb, unsigned int *ret_count);
+
+/* file.c */
+extern const struct file_operations exfat_file_operations;
+int __exfat_truncate(struct inode *inode, loff_t new_size);
+void exfat_truncate(struct inode *inode, loff_t size);
+int exfat_setattr(struct dentry *dentry, struct iattr *attr);
+int exfat_getattr(const struct path *path, struct kstat *stat,
+		unsigned int request_mask, unsigned int query_flags);
+
+/* namei.c */
+extern const struct dentry_operations exfat_dentry_ops;
+extern const struct dentry_operations exfat_utf8_dentry_ops;
+
+/* cache.c */
+int exfat_cache_init(void);
+void exfat_cache_shutdown(void);
+void exfat_cache_init_inode(struct inode *inode);
+void exfat_cache_inval_inode(struct inode *inode);
+int exfat_get_cluster(struct inode *inode, unsigned int cluster,
+		unsigned int *fclus, unsigned int *dclus,
+		unsigned int *last_dclus, int allow_eof);
+
+/* dir.c */
+extern const struct inode_operations exfat_dir_inode_operations;
+extern const struct file_operations exfat_dir_operations;
+unsigned int exfat_get_entry_type(struct exfat_dentry *p_entry);
+int exfat_init_dir_entry(struct inode *inode, struct exfat_chain *p_dir,
+		int entry, unsigned int type, unsigned int start_clu,
+		unsigned long long size);
+int exfat_init_ext_entry(struct inode *inode, struct exfat_chain *p_dir,
+		int entry, int num_entries, struct exfat_uni_name *p_uniname);
+int exfat_remove_entries(struct inode *inode, struct exfat_chain *p_dir,
+		int entry, int order, int num_entries);
+int exfat_update_dir_chksum(struct inode *inode, struct exfat_chain *p_dir,
+		int entry);
+int exfat_update_dir_chksum_with_entry_set(struct super_block *sb,
+		struct exfat_entry_set_cache *es, int sync);
+int exfat_calc_num_entries(struct exfat_uni_name *p_uniname);
+int exfat_find_dir_entry(struct super_block *sb, struct exfat_inode_info *ei,
+		struct exfat_chain *p_dir, struct exfat_uni_name *p_uniname,
+		int num_entries, unsigned int type);
+int exfat_alloc_new_dir(struct inode *inode, struct exfat_chain *clu);
+int exfat_find_location(struct super_block *sb, struct exfat_chain *p_dir,
+		int entry, sector_t *sector, int *offset);
+struct exfat_dentry *exfat_get_dentry(struct super_block *sb,
+		struct exfat_chain *p_dir, int entry, struct buffer_head **bh,
+		sector_t *sector);
+struct exfat_entry_set_cache *exfat_get_dentry_set(struct super_block *sb,
+		struct exfat_chain *p_dir, int entry, unsigned int type,
+		struct exfat_dentry **file_ep);
+int exfat_count_dir_entries(struct super_block *sb, struct exfat_chain *p_dir);
+
+/* inode.c */
+extern const struct inode_operations exfat_file_inode_operations;
+void exfat_sync_inode(struct inode *inode);
+struct inode *exfat_build_inode(struct super_block *sb,
+		struct exfat_dir_entry *info, loff_t i_pos);
+void exfat_hash_inode(struct inode *inode, loff_t i_pos);
+void exfat_unhash_inode(struct inode *inode);
+struct inode *exfat_iget(struct super_block *sb, loff_t i_pos);
+int exfat_write_inode(struct inode *inode, struct writeback_control *wbc);
+void exfat_evict_inode(struct inode *inode);
+int exfat_block_truncate_page(struct inode *inode, loff_t from);
+
+/* exfat/nls.c */
+unsigned short exfat_toupper(struct super_block *sb, unsigned short a);
+int exfat_uniname_ncmp(struct super_block *sb, unsigned short *a,
+		unsigned short *b, unsigned int len);
+int exfat_utf16_to_nls(struct super_block *sb,
+		struct exfat_uni_name *uniname, unsigned char *p_cstring,
+		int len);
+int exfat_nls_to_utf16(struct super_block *sb,
+		const unsigned char *p_cstring, const int len,
+		struct exfat_uni_name *uniname, int *p_lossy);
+int exfat_create_upcase_table(struct super_block *sb);
+void exfat_free_upcase_table(struct exfat_sb_info *sbi);
+unsigned short exfat_high_surrogate(unicode_t u);
+unsigned short exfat_low_surrogate(unicode_t u);
+
+/* exfat/misc.c */
+void __exfat_fs_error(struct super_block *sb, int report, const char *fmt, ...)
+		__printf(3, 4) __cold;
+#define exfat_fs_error(sb, fmt, args...)          \
+		__exfat_fs_error(sb, 1, fmt, ## args)
+#define exfat_fs_error_ratelimit(sb, fmt, args...) \
+		__exfat_fs_error(sb, __ratelimit(&EXFAT_SB(sb)->ratelimit), \
+		fmt, ## args)
+void exfat_msg(struct super_block *sb, const char *lv, const char *fmt, ...)
+		__printf(3, 4) __cold;
+void exfat_get_entry_time(struct exfat_sb_info *sbi, struct timespec64 *ts,
+		u8 tz, __le16 time, __le16 date, u8 time_ms);
+void exfat_truncate_atime(struct timespec64 *ts);
+void exfat_set_entry_time(struct exfat_sb_info *sbi, struct timespec64 *ts,
+		u8 *tz, __le16 *time, __le16 *date, u8 *time_ms);
+unsigned short exfat_calc_chksum_2byte(void *data, int len,
+		unsigned short chksum, int type);
+void exfat_update_bh(struct super_block *sb, struct buffer_head *bh, int sync);
+void exfat_chain_set(struct exfat_chain *ec, unsigned int dir,
+		unsigned int size, unsigned char flags);
+void exfat_chain_dup(struct exfat_chain *dup, struct exfat_chain *ec);
+
+#endif /* !_EXFAT_FS_H */
diff --git a/fs/exfat/exfat_raw.h b/fs/exfat/exfat_raw.h
new file mode 100644
index 000000000000..2a841010e649
--- /dev/null
+++ b/fs/exfat/exfat_raw.h
@@ -0,0 +1,184 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#ifndef _EXFAT_RAW_H
+#define _EXFAT_RAW_H
+
+#include <linux/types.h>
+
+#define PBR_SIGNATURE		0xAA55
+
+#define EXFAT_MAX_FILE_LEN	255
+
+#define VOL_CLEAN		0x0000
+#define VOL_DIRTY		0x0002
+
+#define EXFAT_EOF_CLUSTER	0xFFFFFFFFu
+#define EXFAT_BAD_CLUSTER	0xFFFFFFF7u
+#define EXFAT_FREE_CLUSTER	0
+/* Cluster 0, 1 are reserved, the first cluster is 2 in the cluster heap. */
+#define EXFAT_RESERVED_CLUSTERS	2
+#define EXFAT_FIRST_CLUSTER	2
+#define EXFAT_DATA_CLUSTER_COUNT(sbi)	\
+	((sbi)->num_clusters - EXFAT_RESERVED_CLUSTERS)
+
+/* AllocationPossible and NoFatChain field in GeneralSecondaryFlags Field */
+#define ALLOC_FAT_CHAIN		0x01
+#define ALLOC_NO_FAT_CHAIN	0x03
+
+#define DENTRY_SIZE		32 /* directory entry size */
+#define DENTRY_SIZE_BITS	5
+/* exFAT allows 8388608(256MB) directory entries */
+#define MAX_EXFAT_DENTRIES	8388608
+
+/* dentry types */
+#define EXFAT_UNUSED		0x00	/* end of directory */
+#define EXFAT_DELETE		(~0x80)
+#define IS_EXFAT_DELETED(x)	((x) < 0x80) /* deleted file (0x01~0x7F) */
+#define EXFAT_INVAL		0x80	/* invalid value */
+#define EXFAT_BITMAP		0x81	/* allocation bitmap */
+#define EXFAT_UPCASE		0x82	/* upcase table */
+#define EXFAT_VOLUME		0x83	/* volume label */
+#define EXFAT_FILE		0x85	/* file or dir */
+#define EXFAT_GUID		0xA0
+#define EXFAT_PADDING		0xA1
+#define EXFAT_ACLTAB		0xA2
+#define EXFAT_STREAM		0xC0	/* stream entry */
+#define EXFAT_NAME		0xC1	/* file name entry */
+#define EXFAT_ACL		0xC2	/* stream entry */
+
+#define IS_EXFAT_CRITICAL_PRI(x)	(x < 0xA0)
+#define IS_EXFAT_BENIGN_PRI(x)		(x < 0xC0)
+#define IS_EXFAT_CRITICAL_SEC(x)	(x < 0xE0)
+
+/* checksum types */
+#define CS_DIR_ENTRY		0
+#define CS_PBR_SECTOR		1
+#define CS_DEFAULT		2
+
+/* file attributes */
+#define ATTR_READONLY		0x0001
+#define ATTR_HIDDEN		0x0002
+#define ATTR_SYSTEM		0x0004
+#define ATTR_VOLUME		0x0008
+#define ATTR_SUBDIR		0x0010
+#define ATTR_ARCHIVE		0x0020
+
+#define ATTR_RWMASK		(ATTR_HIDDEN | ATTR_SYSTEM | ATTR_VOLUME | \
+				 ATTR_SUBDIR | ATTR_ARCHIVE)
+
+#define PBR64_JUMP_BOOT_LEN		3
+#define PBR64_OEM_NAME_LEN		8
+#define PBR64_RESERVED_LEN		53
+
+#define EXFAT_FILE_NAME_LEN		15
+
+/* EXFAT BIOS parameter block (64 bytes) */
+struct bpb64 {
+	__u8 jmp_boot[PBR64_JUMP_BOOT_LEN];
+	__u8 oem_name[PBR64_OEM_NAME_LEN];
+	__u8 res_zero[PBR64_RESERVED_LEN];
+} __packed;
+
+/* EXFAT EXTEND BIOS parameter block (56 bytes) */
+struct bsx64 {
+	__le64 vol_offset;
+	__le64 vol_length;
+	__le32 fat_offset;
+	__le32 fat_length;
+	__le32 clu_offset;
+	__le32 clu_count;
+	__le32 root_cluster;
+	__le32 vol_serial;
+	__u8 fs_version[2];
+	__le16 vol_flags;
+	__u8 sect_size_bits;
+	__u8 sect_per_clus_bits;
+	__u8 num_fats;
+	__u8 phy_drv_no;
+	__u8 perc_in_use;
+	__u8 reserved2[7];
+} __packed;
+
+/* EXFAT PBR[BPB+BSX] (120 bytes) */
+struct pbr64 {
+	struct bpb64 bpb;
+	struct bsx64 bsx;
+} __packed;
+
+/* Common PBR[Partition Boot Record] (512 bytes) */
+struct pbr {
+	union {
+		__u8 raw[64];
+		struct bpb64 f64;
+	} bpb;
+	union {
+		__u8 raw[56];
+		struct bsx64 f64;
+	} bsx;
+	__u8 boot_code[390];
+	__le16 signature;
+} __packed;
+
+struct exfat_dentry {
+	__u8 type;
+	union {
+		struct {
+			__u8 num_ext;
+			__le16 checksum;
+			__le16 attr;
+			__le16 reserved1;
+			__le16 create_time;
+			__le16 create_date;
+			__le16 modify_time;
+			__le16 modify_date;
+			__le16 access_time;
+			__le16 access_date;
+			__u8 create_time_ms;
+			__u8 modify_time_ms;
+			__u8 create_tz;
+			__u8 modify_tz;
+			__u8 access_tz;
+			__u8 reserved2[7];
+		} __packed file; /* file directory entry */
+		struct {
+			__u8 flags;
+			__u8 reserved1;
+			__u8 name_len;
+			__le16 name_hash;
+			__le16 reserved2;
+			__le64 valid_size;
+			__le32 reserved3;
+			__le32 start_clu;
+			__le64 size;
+		} __packed stream; /* stream extension directory entry */
+		struct {
+			__u8 flags;
+			__le16 unicode_0_14[EXFAT_FILE_NAME_LEN];
+		} __packed name; /* file name directory entry */
+		struct {
+			__u8 flags;
+			__u8 reserved[18];
+			__le32 start_clu;
+			__le64 size;
+		} __packed bitmap; /* allocation bitmap directory entry */
+		struct {
+			__u8 reserved1[3];
+			__le32 checksum;
+			__u8 reserved2[12];
+			__le32 start_clu;
+			__le64 size;
+		} __packed upcase; /* up-case table directory entry */
+	} __packed dentry;
+} __packed;
+
+#define EXFAT_TZ_VALID		(1 << 7)
+
+/* Jan 1 GMT 00:00:00 1980 */
+#define EXFAT_MIN_TIMESTAMP_SECS    315532800LL
+/* Dec 31 GMT 23:59:59 2107 */
+#define EXFAT_MAX_TIMESTAMP_SECS    4354819199LL
+
+#endif /* !_EXFAT_RAW_H */
diff --git a/fs/exfat/fatent.c b/fs/exfat/fatent.c
new file mode 100644
index 000000000000..a855b1769a96
--- /dev/null
+++ b/fs/exfat/fatent.c
@@ -0,0 +1,463 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/slab.h>
+#include <asm/unaligned.h>
+#include <linux/buffer_head.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+static int exfat_mirror_bh(struct super_block *sb, sector_t sec,
+		struct buffer_head *bh)
+{
+	struct buffer_head *c_bh;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	sector_t sec2;
+	int err = 0;
+
+	if (sbi->FAT2_start_sector != sbi->FAT1_start_sector) {
+		sec2 = sec - sbi->FAT1_start_sector + sbi->FAT2_start_sector;
+		c_bh = sb_getblk(sb, sec2);
+		if (!c_bh)
+			return -ENOMEM;
+		memcpy(c_bh->b_data, bh->b_data, sb->s_blocksize);
+		set_buffer_uptodate(c_bh);
+		mark_buffer_dirty(c_bh);
+		if (sb->s_flags & SB_SYNCHRONOUS)
+			err = sync_dirty_buffer(c_bh);
+		brelse(c_bh);
+	}
+
+	return err;
+}
+
+static int __exfat_ent_get(struct super_block *sb, unsigned int loc,
+		unsigned int *content)
+{
+	unsigned int off;
+	sector_t sec;
+	struct buffer_head *bh;
+
+	sec = FAT_ENT_OFFSET_SECTOR(sb, loc);
+	off = FAT_ENT_OFFSET_BYTE_IN_SECTOR(sb, loc);
+
+	bh = sb_bread(sb, sec);
+	if (!bh)
+		return -EIO;
+
+	*content = le32_to_cpu(*(__le32 *)(&bh->b_data[off]));
+
+	/* remap reserved clusters to simplify code */
+	if (*content > EXFAT_BAD_CLUSTER)
+		*content = EXFAT_EOF_CLUSTER;
+
+	brelse(bh);
+	return 0;
+}
+
+int exfat_ent_set(struct super_block *sb, unsigned int loc,
+		unsigned int content)
+{
+	unsigned int off;
+	sector_t sec;
+	__le32 *fat_entry;
+	struct buffer_head *bh;
+
+	sec = FAT_ENT_OFFSET_SECTOR(sb, loc);
+	off = FAT_ENT_OFFSET_BYTE_IN_SECTOR(sb, loc);
+
+	bh = sb_bread(sb, sec);
+	if (!bh)
+		return -EIO;
+
+	fat_entry = (__le32 *)&(bh->b_data[off]);
+	*fat_entry = cpu_to_le32(content);
+	exfat_update_bh(sb, bh, sb->s_flags & SB_SYNCHRONOUS);
+	exfat_mirror_bh(sb, sec, bh);
+	brelse(bh);
+	return 0;
+}
+
+static inline bool is_valid_cluster(struct exfat_sb_info *sbi,
+		unsigned int clus)
+{
+	if (clus < EXFAT_FIRST_CLUSTER || sbi->num_clusters <= clus)
+		return false;
+	return true;
+}
+
+int exfat_ent_get(struct super_block *sb, unsigned int loc,
+		unsigned int *content)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	int err;
+
+	if (!is_valid_cluster(sbi, loc)) {
+		exfat_fs_error(sb, "invalid access to FAT (entry 0x%08x)",
+			loc);
+		return -EIO;
+	}
+
+	err = __exfat_ent_get(sb, loc, content);
+	if (err) {
+		exfat_fs_error(sb,
+			"failed to access to FAT (entry 0x%08x, err:%d)",
+			loc, err);
+		return err;
+	}
+
+	if (*content == EXFAT_FREE_CLUSTER) {
+		exfat_fs_error(sb,
+			"invalid access to FAT free cluster (entry 0x%08x)",
+			loc);
+		return -EIO;
+	}
+
+	if (*content == EXFAT_BAD_CLUSTER) {
+		exfat_fs_error(sb,
+			"invalid access to FAT bad cluster (entry 0x%08x)",
+			loc);
+		return -EIO;
+	}
+
+	if (*content != EXFAT_EOF_CLUSTER && !is_valid_cluster(sbi, *content)) {
+		exfat_fs_error(sb,
+			"invalid access to FAT (entry 0x%08x) bogus content (0x%08x)",
+			loc, *content);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int exfat_chain_cont_cluster(struct super_block *sb, unsigned int chain,
+		unsigned int len)
+{
+	if (!len)
+		return 0;
+
+	while (len > 1) {
+		if (exfat_ent_set(sb, chain, chain + 1))
+			return -EIO;
+		chain++;
+		len--;
+	}
+
+	if (exfat_ent_set(sb, chain, EXFAT_EOF_CLUSTER))
+		return -EIO;
+	return 0;
+}
+
+int exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain)
+{
+	unsigned int num_clusters = 0;
+	unsigned int clu;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	/* invalid cluster number */
+	if (p_chain->dir == EXFAT_FREE_CLUSTER ||
+	    p_chain->dir == EXFAT_EOF_CLUSTER ||
+	    p_chain->dir < EXFAT_FIRST_CLUSTER)
+		return 0;
+
+	/* no cluster to truncate */
+	if (p_chain->size == 0)
+		return 0;
+
+	/* check cluster validation */
+	if (p_chain->dir < 2 && p_chain->dir >= sbi->num_clusters) {
+		exfat_msg(sb, KERN_ERR, "invalid start cluster (%u)",
+				p_chain->dir);
+		return -EIO;
+	}
+
+	set_bit(EXFAT_SB_DIRTY, &sbi->s_state);
+	clu = p_chain->dir;
+
+	if (p_chain->flags == ALLOC_NO_FAT_CHAIN) {
+		do {
+			exfat_clear_bitmap(inode, clu);
+			clu++;
+
+			num_clusters++;
+		} while (num_clusters < p_chain->size);
+	} else {
+		do {
+			exfat_clear_bitmap(inode, clu);
+
+			if (exfat_get_next_cluster(sb, &clu))
+				goto dec_used_clus;
+
+			num_clusters++;
+		} while (clu != EXFAT_EOF_CLUSTER);
+	}
+
+dec_used_clus:
+	sbi->used_clusters -= num_clusters;
+	return 0;
+}
+
+int exfat_find_last_cluster(struct super_block *sb, struct exfat_chain *p_chain,
+		unsigned int *ret_clu)
+{
+	unsigned int clu, next;
+	unsigned int count = 0;
+
+	next = p_chain->dir;
+	if (p_chain->flags == ALLOC_NO_FAT_CHAIN) {
+		*ret_clu = next + p_chain->size - 1;
+		return 0;
+	}
+
+	do {
+		count++;
+		clu = next;
+		if (exfat_ent_get(sb, clu, &next))
+			return -EIO;
+	} while (next != EXFAT_EOF_CLUSTER);
+
+	if (p_chain->size != count) {
+		exfat_fs_error(sb,
+			"bogus directory size (clus : ondisk(%d) != counted(%d))",
+			p_chain->size, count);
+		return -EIO;
+	}
+
+	*ret_clu = clu;
+	return 0;
+}
+
+static inline int exfat_sync_bhs(struct buffer_head **bhs, int nr_bhs)
+{
+	int i, err = 0;
+
+	for (i = 0; i < nr_bhs; i++)
+		write_dirty_buffer(bhs[i], 0);
+
+	for (i = 0; i < nr_bhs; i++) {
+		wait_on_buffer(bhs[i]);
+		if (!err && !buffer_uptodate(bhs[i]))
+			err = -EIO;
+	}
+	return err;
+}
+
+int exfat_zeroed_cluster(struct inode *dir, unsigned int clu)
+{
+	struct super_block *sb = dir->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct buffer_head *bhs[MAX_BUF_PER_PAGE];
+	int nr_bhs = MAX_BUF_PER_PAGE;
+	sector_t blknr, last_blknr;
+	int err, i, n;
+
+	blknr = exfat_cluster_to_sector(sbi, clu);
+	last_blknr = blknr + sbi->sect_per_clus;
+
+	if (last_blknr > sbi->num_sectors && sbi->num_sectors > 0) {
+		exfat_fs_error_ratelimit(sb,
+			"%s: out of range(sect:%llu len:%u)",
+			__func__, (unsigned long long)blknr,
+			sbi->sect_per_clus);
+		return -EIO;
+	}
+
+	/* Zeroing the unused blocks on this cluster */
+	n = 0;
+	while (blknr < last_blknr) {
+		bhs[n] = sb_getblk(sb, blknr);
+		if (!bhs[n]) {
+			err = -ENOMEM;
+			goto release_bhs;
+		}
+		memset(bhs[n]->b_data, 0, sb->s_blocksize);
+		exfat_update_bh(sb, bhs[n], 0);
+
+		n++;
+		blknr++;
+
+		if (n == nr_bhs) {
+			if (IS_DIRSYNC(dir)) {
+				err = exfat_sync_bhs(bhs, n);
+				if (err)
+					goto release_bhs;
+			}
+
+			for (i = 0; i < n; i++)
+				brelse(bhs[i]);
+			n = 0;
+		}
+	}
+
+	if (IS_DIRSYNC(dir)) {
+		err = exfat_sync_bhs(bhs, n);
+		if (err)
+			goto release_bhs;
+	}
+
+	for (i = 0; i < n; i++)
+		brelse(bhs[i]);
+
+	return 0;
+
+release_bhs:
+	exfat_msg(sb, KERN_ERR, "failed zeroed sect %llu\n",
+		(unsigned long long)blknr);
+	for (i = 0; i < n; i++)
+		bforget(bhs[i]);
+	return err;
+}
+
+int exfat_alloc_cluster(struct inode *inode, unsigned int num_alloc,
+		struct exfat_chain *p_chain)
+{
+	int ret = -ENOSPC;
+	unsigned int num_clusters = 0, total_cnt;
+	unsigned int hint_clu, new_clu, last_clu = EXFAT_EOF_CLUSTER;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	total_cnt = EXFAT_DATA_CLUSTER_COUNT(sbi);
+
+	if (unlikely(total_cnt < sbi->used_clusters)) {
+		exfat_fs_error_ratelimit(sb,
+			"%s: invalid used clusters(t:%u,u:%u)\n",
+			__func__, total_cnt, sbi->used_clusters);
+		return -EIO;
+	}
+
+	if (num_alloc > total_cnt - sbi->used_clusters)
+		return -ENOSPC;
+
+	hint_clu = p_chain->dir;
+	/* find new cluster */
+	if (hint_clu == EXFAT_EOF_CLUSTER) {
+		if (sbi->clu_srch_ptr < EXFAT_FIRST_CLUSTER) {
+			exfat_msg(sb, KERN_ERR,
+				"sbi->clu_srch_ptr is invalid (%u)\n",
+				sbi->clu_srch_ptr);
+			sbi->clu_srch_ptr = EXFAT_FIRST_CLUSTER;
+		}
+
+		hint_clu = exfat_find_free_bitmap(sb, sbi->clu_srch_ptr);
+		if (hint_clu == EXFAT_EOF_CLUSTER)
+			return -ENOSPC;
+	}
+
+	/* check cluster validation */
+	if (hint_clu < EXFAT_FIRST_CLUSTER && hint_clu >= sbi->num_clusters) {
+		exfat_msg(sb, KERN_ERR, "hint_cluster is invalid (%u)\n",
+			hint_clu);
+		hint_clu = EXFAT_FIRST_CLUSTER;
+		if (p_chain->flags == ALLOC_NO_FAT_CHAIN) {
+			if (exfat_chain_cont_cluster(sb, p_chain->dir,
+					num_clusters))
+				return -EIO;
+			p_chain->flags = ALLOC_FAT_CHAIN;
+		}
+	}
+
+	set_bit(EXFAT_SB_DIRTY, &sbi->s_state);
+
+	p_chain->dir = EXFAT_EOF_CLUSTER;
+
+	while ((new_clu = exfat_find_free_bitmap(sb, hint_clu)) !=
+	       EXFAT_EOF_CLUSTER) {
+		if (new_clu != hint_clu &&
+		    p_chain->flags == ALLOC_NO_FAT_CHAIN) {
+			if (exfat_chain_cont_cluster(sb, p_chain->dir,
+					num_clusters)) {
+				ret = -EIO;
+				goto free_cluster;
+			}
+			p_chain->flags = ALLOC_FAT_CHAIN;
+		}
+
+		/* update allocation bitmap */
+		if (exfat_set_bitmap(inode, new_clu)) {
+			ret = -EIO;
+			goto free_cluster;
+		}
+
+		num_clusters++;
+
+		/* update FAT table */
+		if (p_chain->flags == ALLOC_FAT_CHAIN) {
+			if (exfat_ent_set(sb, new_clu, EXFAT_EOF_CLUSTER)) {
+				ret = -EIO;
+				goto free_cluster;
+			}
+		}
+
+		if (p_chain->dir == EXFAT_EOF_CLUSTER) {
+			p_chain->dir = new_clu;
+		} else if (p_chain->flags == ALLOC_FAT_CHAIN) {
+			if (exfat_ent_set(sb, last_clu, new_clu)) {
+				ret = -EIO;
+				goto free_cluster;
+			}
+		}
+		last_clu = new_clu;
+
+		if (--num_alloc == 0) {
+			sbi->clu_srch_ptr = hint_clu;
+			sbi->used_clusters += num_clusters;
+
+			p_chain->size += num_clusters;
+			return 0;
+		}
+
+		hint_clu = new_clu + 1;
+		if (hint_clu >= sbi->num_clusters) {
+			hint_clu = EXFAT_FIRST_CLUSTER;
+
+			if (p_chain->flags == ALLOC_NO_FAT_CHAIN) {
+				if (exfat_chain_cont_cluster(sb, p_chain->dir,
+						num_clusters)) {
+					ret = -EIO;
+					goto free_cluster;
+				}
+				p_chain->flags = ALLOC_FAT_CHAIN;
+			}
+		}
+	}
+free_cluster:
+	if (num_clusters)
+		exfat_free_cluster(inode, p_chain);
+	return ret;
+}
+
+int exfat_count_num_clusters(struct super_block *sb,
+		struct exfat_chain *p_chain, unsigned int *ret_count)
+{
+	unsigned int i, count;
+	unsigned int clu;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	if (!p_chain->dir || p_chain->dir == EXFAT_EOF_CLUSTER) {
+		*ret_count = 0;
+		return 0;
+	}
+
+	if (p_chain->flags == ALLOC_NO_FAT_CHAIN) {
+		*ret_count = p_chain->size;
+		return 0;
+	}
+
+	clu = p_chain->dir;
+	count = 0;
+	for (i = EXFAT_FIRST_CLUSTER; i < sbi->num_clusters; i++) {
+		count++;
+		if (exfat_ent_get(sb, clu, &clu))
+			return -EIO;
+		if (clu == EXFAT_EOF_CLUSTER)
+			break;
+	}
+
+	*ret_count = count;
+	return 0;
+}
diff --git a/fs/exfat/file.c b/fs/exfat/file.c
new file mode 100644
index 000000000000..c9db8eb0cfc3
--- /dev/null
+++ b/fs/exfat/file.c
@@ -0,0 +1,363 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/slab.h>
+#include <linux/cred.h>
+#include <linux/buffer_head.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+static int exfat_cont_expand(struct inode *inode, loff_t size)
+{
+	struct address_space *mapping = inode->i_mapping;
+	loff_t start = i_size_read(inode), count = size - i_size_read(inode);
+	int err, err2;
+
+	err = generic_cont_expand_simple(inode, size);
+	if (err)
+		return err;
+
+	inode->i_ctime = inode->i_mtime = current_time(inode);
+	mark_inode_dirty(inode);
+
+	if (!IS_SYNC(inode))
+		return 0;
+
+	err = filemap_fdatawrite_range(mapping, start, start + count - 1);
+	err2 = sync_mapping_buffers(mapping);
+	if (!err)
+		err = err2;
+	err2 = write_inode_now(inode, 1);
+	if (!err)
+		err = err2;
+	if (err)
+		return err;
+
+	return filemap_fdatawait_range(mapping, start, start + count - 1);
+}
+
+static bool exfat_allow_set_time(struct exfat_sb_info *sbi, struct inode *inode)
+{
+	mode_t allow_utime = sbi->options.allow_utime;
+
+	if (!uid_eq(current_fsuid(), inode->i_uid)) {
+		if (in_group_p(inode->i_gid))
+			allow_utime >>= 3;
+		if (allow_utime & MAY_WRITE)
+			return true;
+	}
+
+	/* use a default check */
+	return false;
+}
+
+static int exfat_sanitize_mode(const struct exfat_sb_info *sbi,
+		struct inode *inode, umode_t *mode_ptr)
+{
+	mode_t i_mode, mask, perm;
+
+	i_mode = inode->i_mode;
+
+	mask = (S_ISREG(i_mode) || S_ISLNK(i_mode)) ?
+		sbi->options.fs_fmask : sbi->options.fs_dmask;
+	perm = *mode_ptr & ~(S_IFMT | mask);
+
+	/* Of the r and x bits, all (subject to umask) must be present.*/
+	if ((perm & 0555) != (i_mode & 0555))
+		return -EPERM;
+
+	if (exfat_mode_can_hold_ro(inode)) {
+		/*
+		 * Of the w bits, either all (subject to umask) or none must
+		 * be present.
+		 */
+		if ((perm & 0222) && ((perm & 0222) != (0222 & ~mask)))
+			return -EPERM;
+	} else {
+		/*
+		 * If exfat_mode_can_hold_ro(inode) is false, can't change
+		 * w bits.
+		 */
+		if ((perm & 0222) != (0222 & ~mask))
+			return -EPERM;
+	}
+
+	*mode_ptr &= S_IFMT | perm;
+
+	return 0;
+}
+
+/* resize the file length */
+int __exfat_truncate(struct inode *inode, loff_t new_size)
+{
+	unsigned int num_clusters_new, num_clusters_phys;
+	unsigned int last_clu = EXFAT_FREE_CLUSTER;
+	struct exfat_chain clu;
+	struct exfat_dentry *ep, *ep2;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct exfat_entry_set_cache *es = NULL;
+	int evict = (ei->dir.dir == DIR_DELETED) ? 1 : 0;
+
+	/* check if the given file ID is opened */
+	if (ei->type != TYPE_FILE && ei->type != TYPE_DIR)
+		return -EPERM;
+
+	exfat_set_vol_flags(sb, VOL_DIRTY);
+
+	num_clusters_new = EXFAT_B_TO_CLU_ROUND_UP(i_size_read(inode), sbi);
+	num_clusters_phys =
+		EXFAT_B_TO_CLU_ROUND_UP(EXFAT_I(inode)->i_size_ondisk, sbi);
+
+	exfat_chain_set(&clu, ei->start_clu, num_clusters_phys, ei->flags);
+
+	if (new_size > 0) {
+		/*
+		 * Truncate FAT chain num_clusters after the first cluster
+		 * num_clusters = min(new, phys);
+		 */
+		unsigned int num_clusters =
+			min(num_clusters_new, num_clusters_phys);
+
+		/*
+		 * Follow FAT chain
+		 * (defensive coding - works fine even with corrupted FAT table
+		 */
+		if (clu.flags == ALLOC_NO_FAT_CHAIN) {
+			clu.dir += num_clusters;
+			clu.size -= num_clusters;
+		} else {
+			while (num_clusters > 0) {
+				last_clu = clu.dir;
+				if (exfat_get_next_cluster(sb, &(clu.dir)))
+					return -EIO;
+
+				num_clusters--;
+				clu.size--;
+			}
+		}
+	} else {
+		ei->flags = ALLOC_NO_FAT_CHAIN;
+		ei->start_clu = EXFAT_EOF_CLUSTER;
+	}
+
+	i_size_write(inode, new_size);
+
+	if (ei->type == TYPE_FILE)
+		ei->attr |= ATTR_ARCHIVE;
+
+	/* update the directory entry */
+	if (!evict) {
+		struct timespec64 ts;
+
+		es = exfat_get_dentry_set(sb, &(ei->dir), ei->entry,
+				ES_ALL_ENTRIES, &ep);
+		if (!es)
+			return -EIO;
+		ep2 = ep + 1;
+
+		ts = current_time(inode);
+		exfat_set_entry_time(sbi, &ts,
+				&ep->dentry.file.modify_tz,
+				&ep->dentry.file.modify_time,
+				&ep->dentry.file.modify_date,
+				&ep->dentry.file.modify_time_ms);
+		ep->dentry.file.attr = cpu_to_le16(ei->attr);
+
+		/* File size should be zero if there is no cluster allocated */
+		if (ei->start_clu == EXFAT_EOF_CLUSTER) {
+			ep->dentry.stream.valid_size = 0;
+			ep->dentry.stream.size = 0;
+		} else {
+			ep->dentry.stream.valid_size = cpu_to_le64(new_size);
+			ep->dentry.stream.size = ep->dentry.stream.valid_size;
+		}
+
+		if (new_size == 0) {
+			/* Any directory can not be truncated to zero */
+			WARN_ON(ei->type != TYPE_FILE);
+
+			ep2->dentry.stream.flags = ALLOC_FAT_CHAIN;
+			ep2->dentry.stream.start_clu = EXFAT_FREE_CLUSTER;
+		}
+
+		if (exfat_update_dir_chksum_with_entry_set(sb, es,
+		    inode_needs_sync(inode)))
+			return -EIO;
+		kfree(es);
+	}
+
+	/* cut off from the FAT chain */
+	if (ei->flags == ALLOC_FAT_CHAIN && last_clu != EXFAT_FREE_CLUSTER &&
+			last_clu != EXFAT_EOF_CLUSTER) {
+		if (exfat_ent_set(sb, last_clu, EXFAT_EOF_CLUSTER))
+			return -EIO;
+	}
+
+	/* invalidate cache and free the clusters */
+	/* clear exfat cache */
+	exfat_cache_inval_inode(inode);
+
+	/* hint information */
+	ei->hint_bmap.off = EXFAT_EOF_CLUSTER;
+	ei->hint_bmap.clu = EXFAT_EOF_CLUSTER;
+	if (ei->rwoffset > new_size)
+		ei->rwoffset = new_size;
+
+	/* hint_stat will be used if this is directory. */
+	ei->hint_stat.eidx = 0;
+	ei->hint_stat.clu = ei->start_clu;
+	ei->hint_femp.eidx = EXFAT_HINT_NONE;
+
+	/* free the clusters */
+	if (exfat_free_cluster(inode, &clu))
+		return -EIO;
+
+	exfat_set_vol_flags(sb, VOL_CLEAN);
+
+	return 0;
+}
+
+void exfat_truncate(struct inode *inode, loff_t size)
+{
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	unsigned int blocksize = 1 << inode->i_blkbits;
+	loff_t aligned_size;
+	int err;
+
+	mutex_lock(&sbi->s_lock);
+	if (EXFAT_I(inode)->start_clu == 0) {
+		/*
+		 * Empty start_clu != ~0 (not allocated)
+		 */
+		exfat_fs_error(sb, "tried to truncate zeroed cluster.");
+		goto write_size;
+	}
+
+	err = __exfat_truncate(inode, i_size_read(inode));
+	if (err)
+		goto write_size;
+
+	inode->i_ctime = inode->i_mtime = current_time(inode);
+	if (IS_DIRSYNC(inode))
+		exfat_sync_inode(inode);
+	else
+		mark_inode_dirty(inode);
+
+	inode->i_blocks = ((i_size_read(inode) + (sbi->cluster_size - 1)) &
+			~(sbi->cluster_size - 1)) >> inode->i_blkbits;
+write_size:
+	aligned_size = i_size_read(inode);
+	if (aligned_size & (blocksize - 1)) {
+		aligned_size |= (blocksize - 1);
+		aligned_size++;
+	}
+
+	if (EXFAT_I(inode)->i_size_ondisk > i_size_read(inode))
+		EXFAT_I(inode)->i_size_ondisk = aligned_size;
+
+	if (EXFAT_I(inode)->i_size_aligned > i_size_read(inode))
+		EXFAT_I(inode)->i_size_aligned = aligned_size;
+	mutex_unlock(&sbi->s_lock);
+}
+
+int exfat_getattr(const struct path *path, struct kstat *stat,
+		unsigned int request_mask, unsigned int query_flags)
+{
+	struct inode *inode = d_backing_inode(path->dentry);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+
+	generic_fillattr(inode, stat);
+	exfat_truncate_atime(&stat->atime);
+	stat->result_mask |= STATX_BTIME;
+	stat->btime.tv_sec = ei->i_crtime.tv_sec;
+	stat->btime.tv_nsec = ei->i_crtime.tv_nsec;
+	stat->blksize = EXFAT_SB(inode->i_sb)->cluster_size;
+	return 0;
+}
+
+int exfat_setattr(struct dentry *dentry, struct iattr *attr)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(dentry->d_sb);
+	struct inode *inode = dentry->d_inode;
+	unsigned int ia_valid;
+	int error;
+
+	if ((attr->ia_valid & ATTR_SIZE) &&
+	    attr->ia_size > i_size_read(inode)) {
+		error = exfat_cont_expand(inode, attr->ia_size);
+		if (error || attr->ia_valid == ATTR_SIZE)
+			return error;
+		attr->ia_valid &= ~ATTR_SIZE;
+	}
+
+	/* Check for setting the inode time. */
+	ia_valid = attr->ia_valid;
+	if ((ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET | ATTR_TIMES_SET)) &&
+	    exfat_allow_set_time(sbi, inode)) {
+		attr->ia_valid &= ~(ATTR_MTIME_SET | ATTR_ATIME_SET |
+				ATTR_TIMES_SET);
+	}
+
+	error = setattr_prepare(dentry, attr);
+	attr->ia_valid = ia_valid;
+	if (error)
+		goto out;
+
+	if (((attr->ia_valid & ATTR_UID) &&
+	     !uid_eq(attr->ia_uid, sbi->options.fs_uid)) ||
+	    ((attr->ia_valid & ATTR_GID) &&
+	     !gid_eq(attr->ia_gid, sbi->options.fs_gid)) ||
+	    ((attr->ia_valid & ATTR_MODE) &&
+	     (attr->ia_mode & ~(S_IFREG | S_IFLNK | S_IFDIR | 0777)))) {
+		error = -EPERM;
+		goto out;
+	}
+
+	/*
+	 * We don't return -EPERM here. Yes, strange, but this is too
+	 * old behavior.
+	 */
+	if (attr->ia_valid & ATTR_MODE) {
+		if (exfat_sanitize_mode(sbi, inode, &attr->ia_mode) < 0)
+			attr->ia_valid &= ~ATTR_MODE;
+	}
+
+	if (attr->ia_valid & ATTR_SIZE) {
+		error = exfat_block_truncate_page(inode, attr->ia_size);
+		if (error)
+			goto out;
+
+		down_write(&EXFAT_I(inode)->truncate_lock);
+		truncate_setsize(inode, attr->ia_size);
+		exfat_truncate(inode, attr->ia_size);
+		up_write(&EXFAT_I(inode)->truncate_lock);
+	}
+
+	setattr_copy(inode, attr);
+	exfat_truncate_atime(&inode->i_atime);
+	mark_inode_dirty(inode);
+
+out:
+	return error;
+}
+
+const struct file_operations exfat_file_operations = {
+	.llseek		= generic_file_llseek,
+	.read_iter	= generic_file_read_iter,
+	.write_iter	= generic_file_write_iter,
+	.mmap		= generic_file_mmap,
+	.fsync		= generic_file_fsync,
+	.splice_read	= generic_file_splice_read,
+	.splice_write	= iter_file_splice_write,
+};
+
+const struct inode_operations exfat_file_inode_operations = {
+	.setattr     = exfat_setattr,
+	.getattr     = exfat_getattr,
+};
diff --git a/fs/exfat/inode.c b/fs/exfat/inode.c
new file mode 100644
index 000000000000..785ead346543
--- /dev/null
+++ b/fs/exfat/inode.c
@@ -0,0 +1,670 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/init.h>
+#include <linux/buffer_head.h>
+#include <linux/mpage.h>
+#include <linux/bio.h>
+#include <linux/blkdev.h>
+#include <linux/time.h>
+#include <linux/writeback.h>
+#include <linux/uio.h>
+#include <linux/random.h>
+#include <linux/iversion.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+static int __exfat_write_inode(struct inode *inode, int sync)
+{
+	int ret = -EIO;
+	unsigned long long on_disk_size;
+	struct exfat_dentry *ep, *ep2;
+	struct exfat_entry_set_cache *es = NULL;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	bool is_dir = (ei->type == TYPE_DIR) ? true : false;
+
+	if (inode->i_ino == EXFAT_ROOT_INO)
+		return 0;
+
+	/*
+	 * If the indode is already unlinked, there is no need for updating it.
+	 */
+	if (ei->dir.dir == DIR_DELETED)
+		return 0;
+
+	if (is_dir && ei->dir.dir == sbi->root_dir && ei->entry == -1)
+		return 0;
+
+	exfat_set_vol_flags(sb, VOL_DIRTY);
+
+	/* get the directory entry of given file or directory */
+	es = exfat_get_dentry_set(sb, &(ei->dir), ei->entry, ES_ALL_ENTRIES,
+		&ep);
+	if (!es)
+		return -EIO;
+	ep2 = ep + 1;
+
+	ep->dentry.file.attr = cpu_to_le16(exfat_make_attr(inode));
+
+	/* set FILE_INFO structure using the acquired struct exfat_dentry */
+	exfat_set_entry_time(sbi, &ei->i_crtime,
+			&ep->dentry.file.create_tz,
+			&ep->dentry.file.create_time,
+			&ep->dentry.file.create_date,
+			&ep->dentry.file.create_time_ms);
+	exfat_set_entry_time(sbi, &inode->i_mtime,
+			&ep->dentry.file.modify_tz,
+			&ep->dentry.file.modify_time,
+			&ep->dentry.file.modify_date,
+			&ep->dentry.file.modify_time_ms);
+	exfat_set_entry_time(sbi, &inode->i_atime,
+			&ep->dentry.file.access_tz,
+			&ep->dentry.file.access_time,
+			&ep->dentry.file.access_date,
+			NULL);
+
+	/* File size should be zero if there is no cluster allocated */
+	on_disk_size = i_size_read(inode);
+
+	if (ei->start_clu == EXFAT_EOF_CLUSTER)
+		on_disk_size = 0;
+
+	ep2->dentry.stream.valid_size = cpu_to_le64(on_disk_size);
+	ep2->dentry.stream.size = ep2->dentry.stream.valid_size;
+
+	ret = exfat_update_dir_chksum_with_entry_set(sb, es, sync);
+	kfree(es);
+	return ret;
+}
+
+int exfat_write_inode(struct inode *inode, struct writeback_control *wbc)
+{
+	int ret;
+
+	mutex_lock(&EXFAT_SB(inode->i_sb)->s_lock);
+	ret = __exfat_write_inode(inode, wbc->sync_mode == WB_SYNC_ALL);
+	mutex_unlock(&EXFAT_SB(inode->i_sb)->s_lock);
+
+	return ret;
+}
+
+void exfat_sync_inode(struct inode *inode)
+{
+	lockdep_assert_held(&EXFAT_SB(inode->i_sb)->s_lock);
+	__exfat_write_inode(inode, 1);
+}
+
+/*
+ * Input: inode, (logical) clu_offset, target allocation area
+ * Output: errcode, cluster number
+ * *clu = (~0), if it's unable to allocate a new cluster
+ */
+static int exfat_map_cluster(struct inode *inode, unsigned int clu_offset,
+		unsigned int *clu, int create)
+{
+	int ret, modified = false;
+	unsigned int last_clu;
+	struct exfat_chain new_clu;
+	struct exfat_dentry *ep;
+	struct exfat_entry_set_cache *es = NULL;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	unsigned int local_clu_offset = clu_offset;
+	unsigned int num_to_be_allocated = 0, num_clusters = 0;
+
+	ei->rwoffset = EXFAT_CLU_TO_B(clu_offset, sbi);
+
+	if (EXFAT_I(inode)->i_size_ondisk > 0)
+		num_clusters =
+			EXFAT_B_TO_CLU_ROUND_UP(EXFAT_I(inode)->i_size_ondisk,
+			sbi);
+
+	if (clu_offset >= num_clusters)
+		num_to_be_allocated = clu_offset - num_clusters + 1;
+
+	if (!create && (num_to_be_allocated > 0)) {
+		*clu = EXFAT_EOF_CLUSTER;
+		return 0;
+	}
+
+	*clu = last_clu = ei->start_clu;
+
+	if (ei->flags == ALLOC_NO_FAT_CHAIN) {
+		if (clu_offset > 0 && *clu != EXFAT_EOF_CLUSTER) {
+			last_clu += clu_offset - 1;
+
+			if (clu_offset == num_clusters)
+				*clu = EXFAT_EOF_CLUSTER;
+			else
+				*clu += clu_offset;
+		}
+	} else if (ei->type == TYPE_FILE) {
+		unsigned int fclus = 0;
+		int err = exfat_get_cluster(inode, clu_offset,
+				&fclus, clu, &last_clu, 1);
+		if (err)
+			return -EIO;
+
+		clu_offset -= fclus;
+	} else {
+		/* hint information */
+		if (clu_offset > 0 && ei->hint_bmap.off != EXFAT_EOF_CLUSTER &&
+		    ei->hint_bmap.off > 0 && clu_offset >= ei->hint_bmap.off) {
+			clu_offset -= ei->hint_bmap.off;
+			/* hint_bmap.clu should be valid */
+			WARN_ON(ei->hint_bmap.clu < 2);
+			*clu = ei->hint_bmap.clu;
+		}
+
+		while (clu_offset > 0 && *clu != EXFAT_EOF_CLUSTER) {
+			last_clu = *clu;
+			if (exfat_get_next_cluster(sb, clu))
+				return -EIO;
+			clu_offset--;
+		}
+	}
+
+	if (*clu == EXFAT_EOF_CLUSTER) {
+		exfat_set_vol_flags(sb, VOL_DIRTY);
+
+		new_clu.dir = (last_clu == EXFAT_EOF_CLUSTER) ?
+				EXFAT_EOF_CLUSTER : last_clu + 1;
+		new_clu.size = 0;
+		new_clu.flags = ei->flags;
+
+		/* allocate a cluster */
+		if (num_to_be_allocated < 1) {
+			/* Broken FAT (i_sze > allocated FAT) */
+			exfat_fs_error(sb, "broken FAT chain.");
+			return -EIO;
+		}
+
+		ret = exfat_alloc_cluster(inode, num_to_be_allocated, &new_clu);
+		if (ret)
+			return ret;
+
+		if (new_clu.dir == EXFAT_EOF_CLUSTER ||
+		    new_clu.dir == EXFAT_FREE_CLUSTER) {
+			exfat_fs_error(sb,
+				"bogus cluster new allocated (last_clu : %u, new_clu : %u)",
+				last_clu, new_clu.dir);
+			return -EIO;
+		}
+
+		/* append to the FAT chain */
+		if (last_clu == EXFAT_EOF_CLUSTER) {
+			if (new_clu.flags == ALLOC_FAT_CHAIN)
+				ei->flags = ALLOC_FAT_CHAIN;
+			ei->start_clu = new_clu.dir;
+			modified = true;
+		} else {
+			if (new_clu.flags != ei->flags) {
+				/* no-fat-chain bit is disabled,
+				 * so fat-chain should be synced with
+				 * alloc-bitmap
+				 */
+				exfat_chain_cont_cluster(sb, ei->start_clu,
+					num_clusters);
+				ei->flags = ALLOC_FAT_CHAIN;
+				modified = true;
+			}
+			if (new_clu.flags == ALLOC_FAT_CHAIN)
+				if (exfat_ent_set(sb, last_clu, new_clu.dir))
+					return -EIO;
+		}
+
+		num_clusters += num_to_be_allocated;
+		*clu = new_clu.dir;
+
+		if (ei->dir.dir != DIR_DELETED) {
+			es = exfat_get_dentry_set(sb, &(ei->dir), ei->entry,
+				ES_ALL_ENTRIES, &ep);
+			if (!es)
+				return -EIO;
+			/* get stream entry */
+			ep++;
+
+			/* update directory entry */
+			if (modified) {
+				if (ep->dentry.stream.flags != ei->flags)
+					ep->dentry.stream.flags = ei->flags;
+
+				if (le32_to_cpu(ep->dentry.stream.start_clu) !=
+						ei->start_clu)
+					ep->dentry.stream.start_clu =
+						cpu_to_le32(ei->start_clu);
+
+				ep->dentry.stream.valid_size =
+					cpu_to_le64(i_size_read(inode));
+				ep->dentry.stream.size =
+					ep->dentry.stream.valid_size;
+			}
+
+			if (exfat_update_dir_chksum_with_entry_set(sb, es,
+			    inode_needs_sync(inode)))
+				return -EIO;
+			kfree(es);
+
+		} /* end of if != DIR_DELETED */
+
+		inode->i_blocks +=
+			num_to_be_allocated << sbi->sect_per_clus_bits;
+
+		/*
+		 * Move *clu pointer along FAT chains (hole care) because the
+		 * caller of this function expect *clu to be the last cluster.
+		 * This only works when num_to_be_allocated >= 2,
+		 * *clu = (the first cluster of the allocated chain) =>
+		 * (the last cluster of ...)
+		 */
+		if (ei->flags == ALLOC_NO_FAT_CHAIN) {
+			*clu += num_to_be_allocated - 1;
+		} else {
+			while (num_to_be_allocated > 1) {
+				if (exfat_get_next_cluster(sb, clu))
+					return -EIO;
+				num_to_be_allocated--;
+			}
+		}
+
+	}
+
+	/* hint information */
+	ei->hint_bmap.off = local_clu_offset;
+	ei->hint_bmap.clu = *clu;
+
+	return 0;
+}
+
+static int exfat_map_new_buffer(struct exfat_inode_info *ei,
+		struct buffer_head *bh, loff_t pos)
+{
+	if (buffer_delay(bh) && pos > ei->i_size_aligned)
+		return -EIO;
+	set_buffer_new(bh);
+
+	/*
+	 * Adjust i_size_aligned if i_size_ondisk is bigger than it.
+	 */
+	if (ei->i_size_ondisk > ei->i_size_aligned)
+		ei->i_size_aligned = ei->i_size_ondisk;
+	return 0;
+}
+
+static int exfat_get_block(struct inode *inode, sector_t iblock,
+		struct buffer_head *bh_result, int create)
+{
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	unsigned long max_blocks = bh_result->b_size >> inode->i_blkbits;
+	int err = 0;
+	unsigned long mapped_blocks = 0;
+	unsigned int cluster, sec_offset;
+	sector_t last_block;
+	sector_t phys = 0;
+	loff_t pos;
+
+	mutex_lock(&sbi->s_lock);
+	last_block = EXFAT_B_TO_BLK_ROUND_UP(i_size_read(inode), sb);
+	if (iblock >= last_block && !create)
+		goto done;
+
+	/* Is this block already allocated? */
+	err = exfat_map_cluster(inode, iblock >> sbi->sect_per_clus_bits,
+			&cluster, create);
+	if (err) {
+		if (err != -ENOSPC)
+			exfat_fs_error_ratelimit(sb,
+				"failed to bmap (inode : %p iblock : %llu, err : %d)",
+				inode, (unsigned long long)iblock, err);
+		goto unlock_ret;
+	}
+
+	if (cluster == EXFAT_EOF_CLUSTER)
+		goto done;
+
+	/* sector offset in cluster */
+	sec_offset = iblock & (sbi->sect_per_clus - 1);
+
+	phys = exfat_cluster_to_sector(sbi, cluster) + sec_offset;
+	mapped_blocks = sbi->sect_per_clus - sec_offset;
+	max_blocks = min(mapped_blocks, max_blocks);
+
+	/* Treat newly added block / cluster */
+	if (iblock < last_block)
+		create = 0;
+
+	if (create || buffer_delay(bh_result)) {
+		pos = EXFAT_BLK_TO_B((iblock + 1), sb);
+		if (ei->i_size_ondisk < pos)
+			ei->i_size_ondisk = pos;
+	}
+
+	if (create) {
+		err = exfat_map_new_buffer(ei, bh_result, pos);
+		if (err) {
+			exfat_fs_error(sb,
+					"requested for bmap out of range(pos : (%llu) > i_size_aligned(%llu)\n",
+					pos, ei->i_size_aligned);
+			goto unlock_ret;
+		}
+	}
+
+	if (buffer_delay(bh_result))
+		clear_buffer_delay(bh_result);
+	map_bh(bh_result, sb, phys);
+done:
+	bh_result->b_size = EXFAT_BLK_TO_B(max_blocks, sb);
+unlock_ret:
+	mutex_unlock(&sbi->s_lock);
+	return err;
+}
+
+static int exfat_readpage(struct file *file, struct page *page)
+{
+	return mpage_readpage(page, exfat_get_block);
+}
+
+static void exfat_readahead(struct readahead_control *rac)
+{
+	mpage_readahead(rac, exfat_get_block);
+}
+
+static int exfat_writepage(struct page *page, struct writeback_control *wbc)
+{
+	return block_write_full_page(page, exfat_get_block, wbc);
+}
+
+static int exfat_writepages(struct address_space *mapping,
+		struct writeback_control *wbc)
+{
+	return mpage_writepages(mapping, wbc, exfat_get_block);
+}
+
+static void exfat_write_failed(struct address_space *mapping, loff_t to)
+{
+	struct inode *inode = mapping->host;
+
+	if (to > i_size_read(inode)) {
+		truncate_pagecache(inode, i_size_read(inode));
+		exfat_truncate(inode, EXFAT_I(inode)->i_size_aligned);
+	}
+}
+
+static int exfat_write_begin(struct file *file, struct address_space *mapping,
+		loff_t pos, unsigned int len, unsigned int flags,
+		struct page **pagep, void **fsdata)
+{
+	int ret;
+
+	*pagep = NULL;
+	ret = cont_write_begin(file, mapping, pos, len, flags, pagep, fsdata,
+			       exfat_get_block,
+			       &EXFAT_I(mapping->host)->i_size_ondisk);
+
+	if (ret < 0)
+		exfat_write_failed(mapping, pos+len);
+
+	return ret;
+}
+
+static int exfat_write_end(struct file *file, struct address_space *mapping,
+		loff_t pos, unsigned int len, unsigned int copied,
+		struct page *pagep, void *fsdata)
+{
+	struct inode *inode = mapping->host;
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	int err;
+
+	err = generic_write_end(file, mapping, pos, len, copied, pagep, fsdata);
+
+	if (EXFAT_I(inode)->i_size_aligned < i_size_read(inode)) {
+		exfat_fs_error(inode->i_sb,
+			"invalid size(size(%llu) > aligned(%llu)\n",
+			i_size_read(inode), EXFAT_I(inode)->i_size_aligned);
+		return -EIO;
+	}
+
+	if (err < len)
+		exfat_write_failed(mapping, pos+len);
+
+	if (!(err < 0) && !(ei->attr & ATTR_ARCHIVE)) {
+		inode->i_mtime = inode->i_ctime = current_time(inode);
+		ei->attr |= ATTR_ARCHIVE;
+		mark_inode_dirty(inode);
+	}
+
+	return err;
+}
+
+static ssize_t exfat_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
+{
+	struct address_space *mapping = iocb->ki_filp->f_mapping;
+	struct inode *inode = mapping->host;
+	loff_t size = iocb->ki_pos + iov_iter_count(iter);
+	int rw = iov_iter_rw(iter);
+	ssize_t ret;
+
+	if (rw == WRITE) {
+		/*
+		 * FIXME: blockdev_direct_IO() doesn't use ->write_begin(),
+		 * so we need to update the ->i_size_aligned to block boundary.
+		 *
+		 * But we must fill the remaining area or hole by nul for
+		 * updating ->i_size_aligned
+		 *
+		 * Return 0, and fallback to normal buffered write.
+		 */
+		if (EXFAT_I(inode)->i_size_aligned < size)
+			return 0;
+	}
+
+	/*
+	 * Need to use the DIO_LOCKING for avoiding the race
+	 * condition of exfat_get_block() and ->truncate().
+	 */
+	ret = blockdev_direct_IO(iocb, inode, iter, exfat_get_block);
+	if (ret < 0 && (rw & WRITE))
+		exfat_write_failed(mapping, size);
+	return ret;
+}
+
+static sector_t exfat_aop_bmap(struct address_space *mapping, sector_t block)
+{
+	sector_t blocknr;
+
+	/* exfat_get_cluster() assumes the requested blocknr isn't truncated. */
+	down_read(&EXFAT_I(mapping->host)->truncate_lock);
+	blocknr = generic_block_bmap(mapping, block, exfat_get_block);
+	up_read(&EXFAT_I(mapping->host)->truncate_lock);
+	return blocknr;
+}
+
+/*
+ * exfat_block_truncate_page() zeroes out a mapping from file offset `from'
+ * up to the end of the block which corresponds to `from'.
+ * This is required during truncate to physically zeroout the tail end
+ * of that block so it doesn't yield old data if the file is later grown.
+ * Also, avoid causing failure from fsx for cases of "data past EOF"
+ */
+int exfat_block_truncate_page(struct inode *inode, loff_t from)
+{
+	return block_truncate_page(inode->i_mapping, from, exfat_get_block);
+}
+
+static const struct address_space_operations exfat_aops = {
+	.readpage	= exfat_readpage,
+	.readahead	= exfat_readahead,
+	.writepage	= exfat_writepage,
+	.writepages	= exfat_writepages,
+	.write_begin	= exfat_write_begin,
+	.write_end	= exfat_write_end,
+	.direct_IO	= exfat_direct_IO,
+	.bmap		= exfat_aop_bmap
+};
+
+static inline unsigned long exfat_hash(loff_t i_pos)
+{
+	return hash_32(i_pos, EXFAT_HASH_BITS);
+}
+
+void exfat_hash_inode(struct inode *inode, loff_t i_pos)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(inode->i_sb);
+	struct hlist_head *head = sbi->inode_hashtable + exfat_hash(i_pos);
+
+	spin_lock(&sbi->inode_hash_lock);
+	EXFAT_I(inode)->i_pos = i_pos;
+	hlist_add_head(&EXFAT_I(inode)->i_hash_fat, head);
+	spin_unlock(&sbi->inode_hash_lock);
+}
+
+void exfat_unhash_inode(struct inode *inode)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(inode->i_sb);
+
+	spin_lock(&sbi->inode_hash_lock);
+	hlist_del_init(&EXFAT_I(inode)->i_hash_fat);
+	EXFAT_I(inode)->i_pos = 0;
+	spin_unlock(&sbi->inode_hash_lock);
+}
+
+struct inode *exfat_iget(struct super_block *sb, loff_t i_pos)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *info;
+	struct hlist_head *head = sbi->inode_hashtable + exfat_hash(i_pos);
+	struct inode *inode = NULL;
+
+	spin_lock(&sbi->inode_hash_lock);
+	hlist_for_each_entry(info, head, i_hash_fat) {
+		WARN_ON(info->vfs_inode.i_sb != sb);
+
+		if (i_pos != info->i_pos)
+			continue;
+		inode = igrab(&info->vfs_inode);
+		if (inode)
+			break;
+	}
+	spin_unlock(&sbi->inode_hash_lock);
+	return inode;
+}
+
+/* doesn't deal with root inode */
+static int exfat_fill_inode(struct inode *inode, struct exfat_dir_entry *info)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(inode->i_sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	loff_t size = info->size;
+
+	memcpy(&ei->dir, &info->dir, sizeof(struct exfat_chain));
+	ei->entry = info->entry;
+	ei->attr = info->attr;
+	ei->start_clu = info->start_clu;
+	ei->flags = info->flags;
+	ei->type = info->type;
+
+	ei->version = 0;
+	ei->hint_stat.eidx = 0;
+	ei->hint_stat.clu = info->start_clu;
+	ei->hint_femp.eidx = EXFAT_HINT_NONE;
+	ei->rwoffset = 0;
+	ei->hint_bmap.off = EXFAT_EOF_CLUSTER;
+	ei->i_pos = 0;
+
+	inode->i_uid = sbi->options.fs_uid;
+	inode->i_gid = sbi->options.fs_gid;
+	inode_inc_iversion(inode);
+	inode->i_generation = prandom_u32();
+
+	if (info->attr & ATTR_SUBDIR) { /* directory */
+		inode->i_generation &= ~1;
+		inode->i_mode = exfat_make_mode(sbi, info->attr, 0777);
+		inode->i_op = &exfat_dir_inode_operations;
+		inode->i_fop = &exfat_dir_operations;
+		set_nlink(inode, info->num_subdirs);
+	} else { /* regular file */
+		inode->i_generation |= 1;
+		inode->i_mode = exfat_make_mode(sbi, info->attr, 0777);
+		inode->i_op = &exfat_file_inode_operations;
+		inode->i_fop = &exfat_file_operations;
+		inode->i_mapping->a_ops = &exfat_aops;
+		inode->i_mapping->nrpages = 0;
+	}
+
+	i_size_write(inode, size);
+
+	/* ondisk and aligned size should be aligned with block size */
+	if (size & (inode->i_sb->s_blocksize - 1)) {
+		size |= (inode->i_sb->s_blocksize - 1);
+		size++;
+	}
+
+	ei->i_size_aligned = size;
+	ei->i_size_ondisk = size;
+
+	exfat_save_attr(inode, info->attr);
+
+	inode->i_blocks = ((i_size_read(inode) + (sbi->cluster_size - 1)) &
+		~(sbi->cluster_size - 1)) >> inode->i_blkbits;
+	inode->i_mtime = info->mtime;
+	inode->i_ctime = info->mtime;
+	ei->i_crtime = info->crtime;
+	inode->i_atime = info->atime;
+
+	exfat_cache_init_inode(inode);
+
+	return 0;
+}
+
+struct inode *exfat_build_inode(struct super_block *sb,
+		struct exfat_dir_entry *info, loff_t i_pos)
+{
+	struct inode *inode;
+	int err;
+
+	inode = exfat_iget(sb, i_pos);
+	if (inode)
+		goto out;
+	inode = new_inode(sb);
+	if (!inode) {
+		inode = ERR_PTR(-ENOMEM);
+		goto out;
+	}
+	inode->i_ino = iunique(sb, EXFAT_ROOT_INO);
+	inode_set_iversion(inode, 1);
+	err = exfat_fill_inode(inode, info);
+	if (err) {
+		iput(inode);
+		inode = ERR_PTR(err);
+		goto out;
+	}
+	exfat_hash_inode(inode, i_pos);
+	insert_inode_hash(inode);
+out:
+	return inode;
+}
+
+void exfat_evict_inode(struct inode *inode)
+{
+	truncate_inode_pages(&inode->i_data, 0);
+
+	if (!inode->i_nlink) {
+		i_size_write(inode, 0);
+		mutex_lock(&EXFAT_SB(inode->i_sb)->s_lock);
+		__exfat_truncate(inode, 0);
+		mutex_unlock(&EXFAT_SB(inode->i_sb)->s_lock);
+	}
+
+	invalidate_inode_buffers(inode);
+	clear_inode(inode);
+	exfat_cache_inval_inode(inode);
+	exfat_unhash_inode(inode);
+}
diff --git a/fs/exfat/misc.c b/fs/exfat/misc.c
new file mode 100644
index 000000000000..ebd2cbe3cbc1
--- /dev/null
+++ b/fs/exfat/misc.c
@@ -0,0 +1,175 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  Written 1992,1993 by Werner Almesberger
+ *  22/11/2000 - Fixed fat_date_unix2dos for dates earlier than 01/01/1980
+ *		 and date_dos2unix for date==0 by Igor Zhbanov(bsg@uniyar.ac.ru)
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/buffer_head.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+/*
+ * exfat_fs_error reports a file system problem that might indicate fa data
+ * corruption/inconsistency. Depending on 'errors' mount option the
+ * panic() is called, or error message is printed FAT and nothing is done,
+ * or filesystem is remounted read-only (default behavior).
+ * In case the file system is remounted read-only, it can be made writable
+ * again by remounting it.
+ */
+void __exfat_fs_error(struct super_block *sb, int report, const char *fmt, ...)
+{
+	struct exfat_mount_options *opts = &EXFAT_SB(sb)->options;
+	va_list args;
+	struct va_format vaf;
+
+	if (report) {
+		va_start(args, fmt);
+		vaf.fmt = fmt;
+		vaf.va = &args;
+		exfat_msg(sb, KERN_ERR, "error, %pV\n", &vaf);
+		va_end(args);
+	}
+
+	if (opts->errors == EXFAT_ERRORS_PANIC) {
+		panic("exFAT-fs (%s): fs panic from previous error\n",
+			sb->s_id);
+	} else if (opts->errors == EXFAT_ERRORS_RO && !sb_rdonly(sb)) {
+		sb->s_flags |= SB_RDONLY;
+		exfat_msg(sb, KERN_ERR, "Filesystem has been set read-only");
+	}
+}
+
+/*
+ * exfat_msg() - print preformated EXFAT specific messages.
+ * All logs except what uses exfat_fs_error() should be written by exfat_msg()
+ */
+void exfat_msg(struct super_block *sb, const char *level, const char *fmt, ...)
+{
+	struct va_format vaf;
+	va_list args;
+
+	va_start(args, fmt);
+	vaf.fmt = fmt;
+	vaf.va = &args;
+	/* level means KERN_ pacility level */
+	printk("%sexFAT-fs (%s): %pV\n", level, sb->s_id, &vaf);
+	va_end(args);
+}
+
+#define SECS_PER_MIN    (60)
+#define TIMEZONE_SEC(x)	((x) * 15 * SECS_PER_MIN)
+
+static void exfat_adjust_tz(struct timespec64 *ts, u8 tz_off)
+{
+	if (tz_off <= 0x3F)
+		ts->tv_sec -= TIMEZONE_SEC(tz_off);
+	else /* 0x40 <= (tz_off & 0x7F) <=0x7F */
+		ts->tv_sec += TIMEZONE_SEC(0x80 - tz_off);
+}
+
+/* Convert a EXFAT time/date pair to a UNIX date (seconds since 1 1 70). */
+void exfat_get_entry_time(struct exfat_sb_info *sbi, struct timespec64 *ts,
+		u8 tz, __le16 time, __le16 date, u8 time_ms)
+{
+	u16 t = le16_to_cpu(time);
+	u16 d = le16_to_cpu(date);
+
+	ts->tv_sec = mktime64(1980 + (d >> 9), d >> 5 & 0x000F, d & 0x001F,
+			      t >> 11, (t >> 5) & 0x003F, (t & 0x001F) << 1);
+
+
+	/* time_ms field represent 0 ~ 199(1990 ms) */
+	if (time_ms) {
+		ts->tv_sec += time_ms / 100;
+		ts->tv_nsec = (time_ms % 100) * 10 * NSEC_PER_MSEC;
+	} else
+		ts->tv_nsec = 0;
+
+	if (tz & EXFAT_TZ_VALID)
+		/* Adjust timezone to UTC0. */
+		exfat_adjust_tz(ts, tz & ~EXFAT_TZ_VALID);
+	else
+		/* Convert from local time to UTC using time_offset. */
+		ts->tv_sec -= sbi->options.time_offset * SECS_PER_MIN;
+}
+
+/* Convert linear UNIX date to a EXFAT time/date pair. */
+void exfat_set_entry_time(struct exfat_sb_info *sbi, struct timespec64 *ts,
+		u8 *tz, __le16 *time, __le16 *date, u8 *time_ms)
+{
+	struct tm tm;
+	u16 t, d;
+
+	time64_to_tm(ts->tv_sec, 0, &tm);
+	t = (tm.tm_hour << 11) | (tm.tm_min << 5) | (tm.tm_sec >> 1);
+	d = ((tm.tm_year - 80) <<  9) | ((tm.tm_mon + 1) << 5) | tm.tm_mday;
+
+	*time = cpu_to_le16(t);
+	*date = cpu_to_le16(d);
+
+	/* time_ms field represent 0 ~ 199(1990 ms) */
+	if (time_ms)
+		*time_ms = (tm.tm_sec & 1) * 100 +
+			ts->tv_nsec / (10 * NSEC_PER_MSEC);
+
+	/*
+	 * Record 00h value for OffsetFromUtc field and 1 value for OffsetValid
+	 * to indicate that local time and UTC are the same.
+	 */
+	*tz = EXFAT_TZ_VALID;
+}
+
+/*
+ * The timestamp for access_time has double seconds granularity.
+ * (There is no 10msIncrement field for access_time unlike create/modify_time)
+ * atime also has only a 2-second resolution.
+ */
+void exfat_truncate_atime(struct timespec64 *ts)
+{
+	ts->tv_sec = round_down(ts->tv_sec, 2);
+	ts->tv_nsec = 0;
+}
+
+unsigned short exfat_calc_chksum_2byte(void *data, int len,
+		unsigned short chksum, int type)
+{
+	int i;
+	unsigned char *c = (unsigned char *)data;
+
+	for (i = 0; i < len; i++, c++) {
+		if (((i == 2) || (i == 3)) && (type == CS_DIR_ENTRY))
+			continue;
+		chksum = (((chksum & 1) << 15) | ((chksum & 0xFFFE) >> 1)) +
+			(unsigned short)*c;
+	}
+	return chksum;
+}
+
+void exfat_update_bh(struct super_block *sb, struct buffer_head *bh, int sync)
+{
+	set_bit(EXFAT_SB_DIRTY, &EXFAT_SB(sb)->s_state);
+	set_buffer_uptodate(bh);
+	mark_buffer_dirty(bh);
+
+	if (sync)
+		sync_dirty_buffer(bh);
+}
+
+void exfat_chain_set(struct exfat_chain *ec, unsigned int dir,
+		unsigned int size, unsigned char flags)
+{
+	ec->dir = dir;
+	ec->size = size;
+	ec->flags = flags;
+}
+
+void exfat_chain_dup(struct exfat_chain *dup, struct exfat_chain *ec)
+{
+	return exfat_chain_set(dup, ec->dir, ec->size, ec->flags);
+}
diff --git a/fs/exfat/namei.c b/fs/exfat/namei.c
new file mode 100644
index 000000000000..a2659a8a68a1
--- /dev/null
+++ b/fs/exfat/namei.c
@@ -0,0 +1,1456 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/iversion.h>
+#include <linux/namei.h>
+#include <linux/slab.h>
+#include <linux/buffer_head.h>
+#include <linux/nls.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+static inline unsigned long exfat_d_version(struct dentry *dentry)
+{
+	return (unsigned long) dentry->d_fsdata;
+}
+
+static inline void exfat_d_version_set(struct dentry *dentry,
+		unsigned long version)
+{
+	dentry->d_fsdata = (void *) version;
+}
+
+/*
+ * If new entry was created in the parent, it could create the 8.3 alias (the
+ * shortname of logname).  So, the parent may have the negative-dentry which
+ * matches the created 8.3 alias.
+ *
+ * If it happened, the negative dentry isn't actually negative anymore.  So,
+ * drop it.
+ */
+static int exfat_d_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	int ret;
+
+	if (flags & LOOKUP_RCU)
+		return -ECHILD;
+
+	/*
+	 * This is not negative dentry. Always valid.
+	 *
+	 * Note, rename() to existing directory entry will have ->d_inode, and
+	 * will use existing name which isn't specified name by user.
+	 *
+	 * We may be able to drop this positive dentry here. But dropping
+	 * positive dentry isn't good idea. So it's unsupported like
+	 * rename("filename", "FILENAME") for now.
+	 */
+	if (d_really_is_positive(dentry))
+		return 1;
+
+	/*
+	 * Drop the negative dentry, in order to make sure to use the case
+	 * sensitive name which is specified by user if this is for creation.
+	 */
+	if (flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET))
+		return 0;
+
+	spin_lock(&dentry->d_lock);
+	ret = inode_eq_iversion(d_inode(dentry->d_parent),
+			exfat_d_version(dentry));
+	spin_unlock(&dentry->d_lock);
+	return ret;
+}
+
+/* returns the length of a struct qstr, ignoring trailing dots */
+static unsigned int exfat_striptail_len(unsigned int len, const char *name)
+{
+	while (len && name[len - 1] == '.')
+		len--;
+	return len;
+}
+
+/*
+ * Compute the hash for the exfat name corresponding to the dentry.  If the name
+ * is invalid, we leave the hash code unchanged so that the existing dentry can
+ * be used. The exfat fs routines will return ENOENT or EINVAL as appropriate.
+ */
+static int exfat_d_hash(const struct dentry *dentry, struct qstr *qstr)
+{
+	struct super_block *sb = dentry->d_sb;
+	struct nls_table *t = EXFAT_SB(sb)->nls_io;
+	const unsigned char *name = qstr->name;
+	unsigned int len = exfat_striptail_len(qstr->len, qstr->name);
+	unsigned long hash = init_name_hash(dentry);
+	int i, charlen;
+	wchar_t c;
+
+	for (i = 0; i < len; i += charlen) {
+		charlen = t->char2uni(&name[i], len - i, &c);
+		if (charlen < 0)
+			return charlen;
+		hash = partial_name_hash(exfat_toupper(sb, c), hash);
+	}
+
+	qstr->hash = end_name_hash(hash);
+	return 0;
+}
+
+static int exfat_d_cmp(const struct dentry *dentry, unsigned int len,
+		const char *str, const struct qstr *name)
+{
+	struct super_block *sb = dentry->d_sb;
+	struct nls_table *t = EXFAT_SB(sb)->nls_io;
+	unsigned int alen = exfat_striptail_len(name->len, name->name);
+	unsigned int blen = exfat_striptail_len(len, str);
+	wchar_t c1, c2;
+	int charlen, i;
+
+	if (alen != blen)
+		return 1;
+
+	for (i = 0; i < len; i += charlen) {
+		charlen = t->char2uni(&name->name[i], alen - i, &c1);
+		if (charlen < 0)
+			return 1;
+		if (charlen != t->char2uni(&str[i], blen - i, &c2))
+			return 1;
+
+		if (exfat_toupper(sb, c1) != exfat_toupper(sb, c2))
+			return 1;
+	}
+
+	return 0;
+}
+
+const struct dentry_operations exfat_dentry_ops = {
+	.d_revalidate	= exfat_d_revalidate,
+	.d_hash		= exfat_d_hash,
+	.d_compare	= exfat_d_cmp,
+};
+
+static int exfat_utf8_d_hash(const struct dentry *dentry, struct qstr *qstr)
+{
+	struct super_block *sb = dentry->d_sb;
+	const unsigned char *name = qstr->name;
+	unsigned int len = exfat_striptail_len(qstr->len, qstr->name);
+	unsigned long hash = init_name_hash(dentry);
+	int i, charlen;
+	unicode_t u;
+
+	for (i = 0; i < len; i += charlen) {
+		charlen = utf8_to_utf32(&name[i], len - i, &u);
+		if (charlen < 0)
+			return charlen;
+
+		/*
+		 * Convert to UTF-16: code points above U+FFFF are encoded as
+		 * surrogate pairs.
+		 * exfat_toupper() works only for code points up to the U+FFFF.
+		 */
+		if (u > 0xFFFF) {
+			hash = partial_name_hash(exfat_high_surrogate(u), hash);
+			hash = partial_name_hash(exfat_low_surrogate(u), hash);
+		} else {
+			hash = partial_name_hash(exfat_toupper(sb, u), hash);
+		}
+	}
+
+	qstr->hash = end_name_hash(hash);
+	return 0;
+}
+
+static int exfat_utf8_d_cmp(const struct dentry *dentry, unsigned int len,
+		const char *str, const struct qstr *name)
+{
+	struct super_block *sb = dentry->d_sb;
+	unsigned int alen = exfat_striptail_len(name->len, name->name);
+	unsigned int blen = exfat_striptail_len(len, str);
+	unicode_t u_a, u_b;
+	int charlen, i;
+
+	if (alen != blen)
+		return 1;
+
+	for (i = 0; i < alen; i += charlen) {
+		charlen = utf8_to_utf32(&name->name[i], alen - i, &u_a);
+		if (charlen < 0)
+			return 1;
+		if (charlen != utf8_to_utf32(&str[i], blen - i, &u_b))
+			return 1;
+
+		if (u_a <= 0xFFFF && u_b <= 0xFFFF) {
+			if (exfat_toupper(sb, u_a) != exfat_toupper(sb, u_b))
+				return 1;
+		} else if (u_a > 0xFFFF && u_b > 0xFFFF) {
+			if (exfat_low_surrogate(u_a) !=
+					exfat_low_surrogate(u_b) ||
+			    exfat_high_surrogate(u_a) !=
+					exfat_high_surrogate(u_b))
+				return 1;
+		} else {
+			return 1;
+		}
+	}
+
+	return 0;
+}
+
+const struct dentry_operations exfat_utf8_dentry_ops = {
+	.d_revalidate	= exfat_d_revalidate,
+	.d_hash		= exfat_utf8_d_hash,
+	.d_compare	= exfat_utf8_d_cmp,
+};
+
+/* used only in search empty_slot() */
+#define CNT_UNUSED_NOHIT        (-1)
+#define CNT_UNUSED_HIT          (-2)
+/* search EMPTY CONTINUOUS "num_entries" entries */
+static int exfat_search_empty_slot(struct super_block *sb,
+		struct exfat_hint_femp *hint_femp, struct exfat_chain *p_dir,
+		int num_entries)
+{
+	int i, dentry, num_empty = 0;
+	int dentries_per_clu;
+	unsigned int type;
+	struct exfat_chain clu;
+	struct exfat_dentry *ep;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct buffer_head *bh;
+
+	dentries_per_clu = sbi->dentries_per_clu;
+
+	if (hint_femp->eidx != EXFAT_HINT_NONE) {
+		dentry = hint_femp->eidx;
+		if (num_entries <= hint_femp->count) {
+			hint_femp->eidx = EXFAT_HINT_NONE;
+			return dentry;
+		}
+
+		exfat_chain_dup(&clu, &hint_femp->cur);
+	} else {
+		exfat_chain_dup(&clu, p_dir);
+		dentry = 0;
+	}
+
+	while (clu.dir != EXFAT_EOF_CLUSTER) {
+		i = dentry & (dentries_per_clu - 1);
+
+		for (; i < dentries_per_clu; i++, dentry++) {
+			ep = exfat_get_dentry(sb, &clu, i, &bh, NULL);
+			if (!ep)
+				return -EIO;
+			type = exfat_get_entry_type(ep);
+			brelse(bh);
+
+			if (type == TYPE_UNUSED || type == TYPE_DELETED) {
+				num_empty++;
+				if (hint_femp->eidx == EXFAT_HINT_NONE) {
+					hint_femp->eidx = dentry;
+					hint_femp->count = CNT_UNUSED_NOHIT;
+					exfat_chain_set(&hint_femp->cur,
+						clu.dir, clu.size, clu.flags);
+				}
+
+				if (type == TYPE_UNUSED &&
+				    hint_femp->count != CNT_UNUSED_HIT)
+					hint_femp->count = CNT_UNUSED_HIT;
+			} else {
+				if (hint_femp->eidx != EXFAT_HINT_NONE &&
+				    hint_femp->count == CNT_UNUSED_HIT) {
+					/* unused empty group means
+					 * an empty group which includes
+					 * unused dentry
+					 */
+					exfat_fs_error(sb,
+						"found bogus dentry(%d) beyond unused empty group(%d) (start_clu : %u, cur_clu : %u)",
+						dentry, hint_femp->eidx,
+						p_dir->dir, clu.dir);
+					return -EIO;
+				}
+
+				num_empty = 0;
+				hint_femp->eidx = EXFAT_HINT_NONE;
+			}
+
+			if (num_empty >= num_entries) {
+				/* found and invalidate hint_femp */
+				hint_femp->eidx = EXFAT_HINT_NONE;
+				return (dentry - (num_entries - 1));
+			}
+		}
+
+		if (clu.flags == ALLOC_NO_FAT_CHAIN) {
+			if (--clu.size > 0)
+				clu.dir++;
+			else
+				clu.dir = EXFAT_EOF_CLUSTER;
+		} else {
+			if (exfat_get_next_cluster(sb, &clu.dir))
+				return -EIO;
+		}
+	}
+
+	return -ENOSPC;
+}
+
+static int exfat_check_max_dentries(struct inode *inode)
+{
+	if (EXFAT_B_TO_DEN(i_size_read(inode)) >= MAX_EXFAT_DENTRIES) {
+		/*
+		 * exFAT spec allows a dir to grow upto 8388608(256MB)
+		 * dentries
+		 */
+		return -ENOSPC;
+	}
+	return 0;
+}
+
+/* find empty directory entry.
+ * if there isn't any empty slot, expand cluster chain.
+ */
+static int exfat_find_empty_entry(struct inode *inode,
+		struct exfat_chain *p_dir, int num_entries)
+{
+	int dentry;
+	unsigned int ret, last_clu;
+	sector_t sector;
+	loff_t size = 0;
+	struct exfat_chain clu;
+	struct exfat_dentry *ep = NULL;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct exfat_hint_femp hint_femp;
+
+	hint_femp.eidx = EXFAT_HINT_NONE;
+
+	if (ei->hint_femp.eidx != EXFAT_HINT_NONE) {
+		memcpy(&hint_femp, &ei->hint_femp,
+				sizeof(struct exfat_hint_femp));
+		ei->hint_femp.eidx = EXFAT_HINT_NONE;
+	}
+
+	while ((dentry = exfat_search_empty_slot(sb, &hint_femp, p_dir,
+					num_entries)) < 0) {
+		if (dentry == -EIO)
+			break;
+
+		if (exfat_check_max_dentries(inode))
+			return -ENOSPC;
+
+		/* we trust p_dir->size regardless of FAT type */
+		if (exfat_find_last_cluster(sb, p_dir, &last_clu))
+			return -EIO;
+
+		/*
+		 * Allocate new cluster to this directory
+		 */
+		exfat_chain_set(&clu, last_clu + 1, 0, p_dir->flags);
+
+		/* allocate a cluster */
+		ret = exfat_alloc_cluster(inode, 1, &clu);
+		if (ret)
+			return ret;
+
+		if (exfat_zeroed_cluster(inode, clu.dir))
+			return -EIO;
+
+		/* append to the FAT chain */
+		if (clu.flags != p_dir->flags) {
+			/* no-fat-chain bit is disabled,
+			 * so fat-chain should be synced with alloc-bitmap
+			 */
+			exfat_chain_cont_cluster(sb, p_dir->dir, p_dir->size);
+			p_dir->flags = ALLOC_FAT_CHAIN;
+			hint_femp.cur.flags = ALLOC_FAT_CHAIN;
+		}
+
+		if (clu.flags == ALLOC_FAT_CHAIN)
+			if (exfat_ent_set(sb, last_clu, clu.dir))
+				return -EIO;
+
+		if (hint_femp.eidx == EXFAT_HINT_NONE) {
+			/* the special case that new dentry
+			 * should be allocated from the start of new cluster
+			 */
+			hint_femp.eidx = EXFAT_B_TO_DEN_IDX(p_dir->size, sbi);
+			hint_femp.count = sbi->dentries_per_clu;
+
+			exfat_chain_set(&hint_femp.cur, clu.dir, 0, clu.flags);
+		}
+		hint_femp.cur.size++;
+		p_dir->size++;
+		size = EXFAT_CLU_TO_B(p_dir->size, sbi);
+
+		/* update the directory entry */
+		if (p_dir->dir != sbi->root_dir) {
+			struct buffer_head *bh;
+
+			ep = exfat_get_dentry(sb,
+				&(ei->dir), ei->entry + 1, &bh, &sector);
+			if (!ep)
+				return -EIO;
+
+			ep->dentry.stream.valid_size = cpu_to_le64(size);
+			ep->dentry.stream.size = ep->dentry.stream.valid_size;
+			ep->dentry.stream.flags = p_dir->flags;
+			exfat_update_bh(sb, bh, IS_DIRSYNC(inode));
+			brelse(bh);
+			if (exfat_update_dir_chksum(inode, &(ei->dir),
+			    ei->entry))
+				return -EIO;
+		}
+
+		/* directory inode should be updated in here */
+		i_size_write(inode, size);
+		EXFAT_I(inode)->i_size_ondisk += sbi->cluster_size;
+		EXFAT_I(inode)->i_size_aligned += sbi->cluster_size;
+		EXFAT_I(inode)->flags = p_dir->flags;
+		inode->i_blocks += 1 << sbi->sect_per_clus_bits;
+	}
+
+	return dentry;
+}
+
+/*
+ * Name Resolution Functions :
+ * Zero if it was successful; otherwise nonzero.
+ */
+static int __exfat_resolve_path(struct inode *inode, const unsigned char *path,
+		struct exfat_chain *p_dir, struct exfat_uni_name *p_uniname,
+		int lookup)
+{
+	int namelen;
+	int lossy = NLS_NAME_NO_LOSSY;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+
+	/* strip all trailing periods */
+	namelen = exfat_striptail_len(strlen(path), path);
+	if (!namelen)
+		return -ENOENT;
+
+	if (strlen(path) > (MAX_NAME_LENGTH * MAX_CHARSET_SIZE))
+		return -ENAMETOOLONG;
+
+	/*
+	 * strip all leading spaces :
+	 * "MS windows 7" supports leading spaces.
+	 * So we should skip this preprocessing for compatibility.
+	 */
+
+	/* file name conversion :
+	 * If lookup case, we allow bad-name for compatibility.
+	 */
+	namelen = exfat_nls_to_utf16(sb, path, namelen, p_uniname,
+			&lossy);
+	if (namelen < 0)
+		return namelen; /* return error value */
+
+	if ((lossy && !lookup) || !namelen)
+		return -EINVAL;
+
+	exfat_chain_set(p_dir, ei->start_clu,
+		EXFAT_B_TO_CLU(i_size_read(inode), sbi), ei->flags);
+
+	return 0;
+}
+
+static inline int exfat_resolve_path(struct inode *inode,
+		const unsigned char *path, struct exfat_chain *dir,
+		struct exfat_uni_name *uni)
+{
+	return __exfat_resolve_path(inode, path, dir, uni, 0);
+}
+
+static inline int exfat_resolve_path_for_lookup(struct inode *inode,
+		const unsigned char *path, struct exfat_chain *dir,
+		struct exfat_uni_name *uni)
+{
+	return __exfat_resolve_path(inode, path, dir, uni, 1);
+}
+
+static inline loff_t exfat_make_i_pos(struct exfat_dir_entry *info)
+{
+	return ((loff_t) info->dir.dir << 32) | (info->entry & 0xffffffff);
+}
+
+static int exfat_add_entry(struct inode *inode, const char *path,
+		struct exfat_chain *p_dir, unsigned int type,
+		struct exfat_dir_entry *info)
+{
+	int ret, dentry, num_entries;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_uni_name uniname;
+	struct exfat_chain clu;
+	int clu_size = 0;
+	unsigned int start_clu = EXFAT_FREE_CLUSTER;
+
+	ret = exfat_resolve_path(inode, path, p_dir, &uniname);
+	if (ret)
+		goto out;
+
+	num_entries = exfat_calc_num_entries(&uniname);
+	if (num_entries < 0) {
+		ret = num_entries;
+		goto out;
+	}
+
+	/* exfat_find_empty_entry must be called before alloc_cluster() */
+	dentry = exfat_find_empty_entry(inode, p_dir, num_entries);
+	if (dentry < 0) {
+		ret = dentry; /* -EIO or -ENOSPC */
+		goto out;
+	}
+
+	if (type == TYPE_DIR) {
+		ret = exfat_alloc_new_dir(inode, &clu);
+		if (ret)
+			goto out;
+		start_clu = clu.dir;
+		clu_size = sbi->cluster_size;
+	}
+
+	/* update the directory entry */
+	/* fill the dos name directory entry information of the created file.
+	 * the first cluster is not determined yet. (0)
+	 */
+	ret = exfat_init_dir_entry(inode, p_dir, dentry, type,
+		start_clu, clu_size);
+	if (ret)
+		goto out;
+
+	ret = exfat_init_ext_entry(inode, p_dir, dentry, num_entries, &uniname);
+	if (ret)
+		goto out;
+
+	memcpy(&info->dir, p_dir, sizeof(struct exfat_chain));
+	info->entry = dentry;
+	info->flags = ALLOC_NO_FAT_CHAIN;
+	info->type = type;
+
+	if (type == TYPE_FILE) {
+		info->attr = ATTR_ARCHIVE;
+		info->start_clu = EXFAT_EOF_CLUSTER;
+		info->size = 0;
+		info->num_subdirs = 0;
+	} else {
+		int count;
+		struct exfat_chain cdir;
+
+		info->attr = ATTR_SUBDIR;
+		info->start_clu = start_clu;
+		info->size = clu_size;
+
+		exfat_chain_set(&cdir, info->start_clu,
+			EXFAT_B_TO_CLU(info->size, sbi), info->flags);
+		count = exfat_count_dir_entries(sb, &cdir);
+		if (count < 0)
+			return -EIO;
+		info->num_subdirs = count + EXFAT_MIN_SUBDIR;
+	}
+	memset(&info->crtime, 0, sizeof(info->crtime));
+	memset(&info->mtime, 0, sizeof(info->mtime));
+	memset(&info->atime, 0, sizeof(info->atime));
+out:
+	return ret;
+}
+
+static int exfat_create(struct inode *dir, struct dentry *dentry, umode_t mode,
+		bool excl)
+{
+	struct super_block *sb = dir->i_sb;
+	struct inode *inode;
+	struct exfat_chain cdir;
+	struct exfat_dir_entry info;
+	loff_t i_pos;
+	int err;
+
+	mutex_lock(&EXFAT_SB(sb)->s_lock);
+	exfat_set_vol_flags(sb, VOL_DIRTY);
+	err = exfat_add_entry(dir, dentry->d_name.name, &cdir, TYPE_FILE,
+		&info);
+	exfat_set_vol_flags(sb, VOL_CLEAN);
+	if (err)
+		goto unlock;
+
+	inode_inc_iversion(dir);
+	dir->i_ctime = dir->i_mtime = current_time(dir);
+	if (IS_DIRSYNC(dir))
+		exfat_sync_inode(dir);
+	else
+		mark_inode_dirty(dir);
+
+	i_pos = exfat_make_i_pos(&info);
+	inode = exfat_build_inode(sb, &info, i_pos);
+	if (IS_ERR(inode))
+		goto unlock;
+
+	inode_inc_iversion(inode);
+	inode->i_mtime = inode->i_atime = inode->i_ctime =
+		EXFAT_I(inode)->i_crtime = current_time(inode);
+	exfat_truncate_atime(&inode->i_atime);
+	/* timestamp is already written, so mark_inode_dirty() is unneeded. */
+
+	d_instantiate(dentry, inode);
+unlock:
+	mutex_unlock(&EXFAT_SB(sb)->s_lock);
+	return err;
+}
+
+/* lookup a file */
+static int exfat_find(struct inode *dir, struct qstr *qname,
+		struct exfat_dir_entry *info)
+{
+	int ret, dentry, num_entries, count;
+	struct exfat_chain cdir;
+	struct exfat_uni_name uni_name;
+	struct exfat_dentry *ep, *ep2;
+	struct exfat_entry_set_cache *es = NULL;
+	struct super_block *sb = dir->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(dir);
+
+	if (qname->len == 0)
+		return -ENOENT;
+
+	/* check the validity of directory name in the given pathname */
+	ret = exfat_resolve_path_for_lookup(dir, qname->name, &cdir, &uni_name);
+	if (ret)
+		return ret;
+
+	num_entries = exfat_calc_num_entries(&uni_name);
+	if (num_entries < 0)
+		return num_entries;
+
+	/* check the validation of hint_stat and initialize it if required */
+	if (ei->version != (inode_peek_iversion_raw(dir) & 0xffffffff)) {
+		ei->hint_stat.clu = cdir.dir;
+		ei->hint_stat.eidx = 0;
+		ei->version = (inode_peek_iversion_raw(dir) & 0xffffffff);
+		ei->hint_femp.eidx = EXFAT_HINT_NONE;
+	}
+
+	/* search the file name for directories */
+	dentry = exfat_find_dir_entry(sb, ei, &cdir, &uni_name,
+			num_entries, TYPE_ALL);
+
+	if ((dentry < 0) && (dentry != -EEXIST))
+		return dentry; /* -error value */
+
+	memcpy(&info->dir, &cdir.dir, sizeof(struct exfat_chain));
+	info->entry = dentry;
+	info->num_subdirs = 0;
+
+	/* root directory itself */
+	if (unlikely(dentry == -EEXIST)) {
+		int num_clu = 0;
+
+		info->type = TYPE_DIR;
+		info->attr = ATTR_SUBDIR;
+		info->flags = ALLOC_FAT_CHAIN;
+		info->start_clu = sbi->root_dir;
+		memset(&info->crtime, 0, sizeof(info->crtime));
+		memset(&info->mtime, 0, sizeof(info->mtime));
+		memset(&info->atime, 0, sizeof(info->atime));
+
+		exfat_chain_set(&cdir, sbi->root_dir, 0, ALLOC_FAT_CHAIN);
+		if (exfat_count_num_clusters(sb, &cdir, &num_clu))
+			return -EIO;
+		info->size = num_clu << sbi->cluster_size_bits;
+
+		count = exfat_count_dir_entries(sb, &cdir);
+		if (count < 0)
+			return -EIO;
+
+		info->num_subdirs = count;
+	} else {
+		es = exfat_get_dentry_set(sb, &cdir, dentry, ES_2_ENTRIES, &ep);
+		if (!es)
+			return -EIO;
+		ep2 = ep + 1;
+
+		info->type = exfat_get_entry_type(ep);
+		info->attr = le16_to_cpu(ep->dentry.file.attr);
+		info->size = le64_to_cpu(ep2->dentry.stream.valid_size);
+		if ((info->type == TYPE_FILE) && (info->size == 0)) {
+			info->flags = ALLOC_NO_FAT_CHAIN;
+			info->start_clu = EXFAT_EOF_CLUSTER;
+		} else {
+			info->flags = ep2->dentry.stream.flags;
+			info->start_clu =
+				le32_to_cpu(ep2->dentry.stream.start_clu);
+		}
+
+		if (ei->start_clu == EXFAT_FREE_CLUSTER) {
+			exfat_fs_error(sb,
+				"non-zero size file starts with zero cluster (size : %llu, p_dir : %u, entry : 0x%08x)",
+				i_size_read(dir), ei->dir.dir, ei->entry);
+			kfree(es);
+			return -EIO;
+		}
+
+		exfat_get_entry_time(sbi, &info->crtime,
+				ep->dentry.file.create_tz,
+				ep->dentry.file.create_time,
+				ep->dentry.file.create_date,
+				ep->dentry.file.create_time_ms);
+		exfat_get_entry_time(sbi, &info->mtime,
+				ep->dentry.file.modify_tz,
+				ep->dentry.file.modify_time,
+				ep->dentry.file.modify_date,
+				ep->dentry.file.modify_time_ms);
+		exfat_get_entry_time(sbi, &info->atime,
+				ep->dentry.file.access_tz,
+				ep->dentry.file.access_time,
+				ep->dentry.file.access_date,
+				0);
+		kfree(es);
+
+		if (info->type == TYPE_DIR) {
+			exfat_chain_set(&cdir, info->start_clu,
+				EXFAT_B_TO_CLU(info->size, sbi), info->flags);
+			count = exfat_count_dir_entries(sb, &cdir);
+			if (count < 0)
+				return -EIO;
+
+			info->num_subdirs = count + EXFAT_MIN_SUBDIR;
+		}
+	}
+	return 0;
+}
+
+static int exfat_d_anon_disconn(struct dentry *dentry)
+{
+	return IS_ROOT(dentry) && (dentry->d_flags & DCACHE_DISCONNECTED);
+}
+
+static struct dentry *exfat_lookup(struct inode *dir, struct dentry *dentry,
+		unsigned int flags)
+{
+	struct super_block *sb = dir->i_sb;
+	struct inode *inode;
+	struct dentry *alias;
+	struct exfat_dir_entry info;
+	int err;
+	loff_t i_pos;
+	mode_t i_mode;
+
+	mutex_lock(&EXFAT_SB(sb)->s_lock);
+	err = exfat_find(dir, &dentry->d_name, &info);
+	if (err) {
+		if (err == -ENOENT) {
+			inode = NULL;
+			goto out;
+		}
+		goto unlock;
+	}
+
+	i_pos = exfat_make_i_pos(&info);
+	inode = exfat_build_inode(sb, &info, i_pos);
+	if (IS_ERR(inode)) {
+		err = PTR_ERR(inode);
+		goto unlock;
+	}
+
+	i_mode = inode->i_mode;
+	alias = d_find_alias(inode);
+
+	/*
+	 * Checking "alias->d_parent == dentry->d_parent" to make sure
+	 * FS is not corrupted (especially double linked dir).
+	 */
+	if (alias && alias->d_parent == dentry->d_parent &&
+			!exfat_d_anon_disconn(alias)) {
+
+		/*
+		 * Unhashed alias is able to exist because of revalidate()
+		 * called by lookup_fast. You can easily make this status
+		 * by calling create and lookup concurrently
+		 * In such case, we reuse an alias instead of new dentry
+		 */
+		if (d_unhashed(alias)) {
+			WARN_ON(alias->d_name.hash_len !=
+				dentry->d_name.hash_len);
+			exfat_msg(sb, KERN_INFO,
+				"rehashed a dentry(%p) in read lookup", alias);
+			d_drop(dentry);
+			d_rehash(alias);
+		} else if (!S_ISDIR(i_mode)) {
+			/*
+			 * This inode has non anonymous-DCACHE_DISCONNECTED
+			 * dentry. This means, the user did ->lookup() by an
+			 * another name (longname vs 8.3 alias of it) in past.
+			 *
+			 * Switch to new one for reason of locality if possible.
+			 */
+			d_move(alias, dentry);
+		}
+		iput(inode);
+		mutex_unlock(&EXFAT_SB(sb)->s_lock);
+		return alias;
+	}
+	dput(alias);
+out:
+	mutex_unlock(&EXFAT_SB(sb)->s_lock);
+	if (!inode)
+		exfat_d_version_set(dentry, inode_query_iversion(dir));
+
+	return d_splice_alias(inode, dentry);
+unlock:
+	mutex_unlock(&EXFAT_SB(sb)->s_lock);
+	return ERR_PTR(err);
+}
+
+/* remove an entry, BUT don't truncate */
+static int exfat_unlink(struct inode *dir, struct dentry *dentry)
+{
+	struct exfat_chain cdir;
+	struct exfat_dentry *ep;
+	struct super_block *sb = dir->i_sb;
+	struct inode *inode = dentry->d_inode;
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct buffer_head *bh;
+	sector_t sector;
+	int num_entries, entry, err = 0;
+
+	mutex_lock(&EXFAT_SB(sb)->s_lock);
+	exfat_chain_dup(&cdir, &ei->dir);
+	entry = ei->entry;
+	if (ei->dir.dir == DIR_DELETED) {
+		exfat_msg(sb, KERN_ERR, "abnormal access to deleted dentry");
+		err = -ENOENT;
+		goto unlock;
+	}
+
+	ep = exfat_get_dentry(sb, &cdir, entry, &bh, &sector);
+	if (!ep) {
+		err = -EIO;
+		goto unlock;
+	}
+	num_entries = exfat_count_ext_entries(sb, &cdir, entry, ep);
+	if (num_entries < 0) {
+		err = -EIO;
+		brelse(bh);
+		goto unlock;
+	}
+	num_entries++;
+	brelse(bh);
+
+	exfat_set_vol_flags(sb, VOL_DIRTY);
+	/* update the directory entry */
+	if (exfat_remove_entries(dir, &cdir, entry, 0, num_entries)) {
+		err = -EIO;
+		goto unlock;
+	}
+
+	/* This doesn't modify ei */
+	ei->dir.dir = DIR_DELETED;
+	exfat_set_vol_flags(sb, VOL_CLEAN);
+
+	inode_inc_iversion(dir);
+	dir->i_mtime = dir->i_atime = current_time(dir);
+	exfat_truncate_atime(&dir->i_atime);
+	if (IS_DIRSYNC(dir))
+		exfat_sync_inode(dir);
+	else
+		mark_inode_dirty(dir);
+
+	clear_nlink(inode);
+	inode->i_mtime = inode->i_atime = current_time(inode);
+	exfat_truncate_atime(&inode->i_atime);
+	exfat_unhash_inode(inode);
+	exfat_d_version_set(dentry, inode_query_iversion(dir));
+unlock:
+	mutex_unlock(&EXFAT_SB(sb)->s_lock);
+	return err;
+}
+
+static int exfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
+{
+	struct super_block *sb = dir->i_sb;
+	struct inode *inode;
+	struct exfat_dir_entry info;
+	struct exfat_chain cdir;
+	loff_t i_pos;
+	int err;
+
+	mutex_lock(&EXFAT_SB(sb)->s_lock);
+	exfat_set_vol_flags(sb, VOL_DIRTY);
+	err = exfat_add_entry(dir, dentry->d_name.name, &cdir, TYPE_DIR,
+		&info);
+	exfat_set_vol_flags(sb, VOL_CLEAN);
+	if (err)
+		goto unlock;
+
+	inode_inc_iversion(dir);
+	dir->i_ctime = dir->i_mtime = current_time(dir);
+	if (IS_DIRSYNC(dir))
+		exfat_sync_inode(dir);
+	else
+		mark_inode_dirty(dir);
+	inc_nlink(dir);
+
+	i_pos = exfat_make_i_pos(&info);
+	inode = exfat_build_inode(sb, &info, i_pos);
+	if (IS_ERR(inode)) {
+		err = PTR_ERR(inode);
+		goto unlock;
+	}
+
+	inode_inc_iversion(inode);
+	inode->i_mtime = inode->i_atime = inode->i_ctime =
+		EXFAT_I(inode)->i_crtime = current_time(inode);
+	exfat_truncate_atime(&inode->i_atime);
+	/* timestamp is already written, so mark_inode_dirty() is unneeded. */
+
+	d_instantiate(dentry, inode);
+
+unlock:
+	mutex_unlock(&EXFAT_SB(sb)->s_lock);
+	return err;
+}
+
+static int exfat_check_dir_empty(struct super_block *sb,
+		struct exfat_chain *p_dir)
+{
+	int i, dentries_per_clu;
+	unsigned int type;
+	struct exfat_chain clu;
+	struct exfat_dentry *ep;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct buffer_head *bh;
+
+	dentries_per_clu = sbi->dentries_per_clu;
+
+	exfat_chain_dup(&clu, p_dir);
+
+	while (clu.dir != EXFAT_EOF_CLUSTER) {
+		for (i = 0; i < dentries_per_clu; i++) {
+			ep = exfat_get_dentry(sb, &clu, i, &bh, NULL);
+			if (!ep)
+				return -EIO;
+			type = exfat_get_entry_type(ep);
+			brelse(bh);
+			if (type == TYPE_UNUSED)
+				return 0;
+
+			if (type != TYPE_FILE && type != TYPE_DIR)
+				continue;
+
+			return -ENOTEMPTY;
+		}
+
+		if (clu.flags == ALLOC_NO_FAT_CHAIN) {
+			if (--clu.size > 0)
+				clu.dir++;
+			else
+				clu.dir = EXFAT_EOF_CLUSTER;
+		} else {
+			if (exfat_get_next_cluster(sb, &(clu.dir)))
+				return -EIO;
+		}
+	}
+
+	return 0;
+}
+
+static int exfat_rmdir(struct inode *dir, struct dentry *dentry)
+{
+	struct inode *inode = dentry->d_inode;
+	struct exfat_dentry *ep;
+	struct exfat_chain cdir, clu_to_free;
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct buffer_head *bh;
+	sector_t sector;
+	int num_entries, entry, err;
+
+	mutex_lock(&EXFAT_SB(inode->i_sb)->s_lock);
+
+	exfat_chain_dup(&cdir, &ei->dir);
+	entry = ei->entry;
+
+	if (ei->dir.dir == DIR_DELETED) {
+		exfat_msg(sb, KERN_ERR, "abnormal access to deleted dentry");
+		err = -ENOENT;
+		goto unlock;
+	}
+
+	exfat_set_vol_flags(sb, VOL_DIRTY);
+	exfat_chain_set(&clu_to_free, ei->start_clu,
+		EXFAT_B_TO_CLU_ROUND_UP(i_size_read(inode), sbi), ei->flags);
+
+	err = exfat_check_dir_empty(sb, &clu_to_free);
+	if (err) {
+		if (err == -EIO)
+			exfat_msg(sb, KERN_ERR,
+				"failed to exfat_check_dir_empty : err(%d)",
+				err);
+		goto unlock;
+	}
+
+	ep = exfat_get_dentry(sb, &cdir, entry, &bh, &sector);
+	if (!ep) {
+		err = -EIO;
+		goto unlock;
+	}
+
+	num_entries = exfat_count_ext_entries(sb, &cdir, entry, ep);
+	if (num_entries < 0) {
+		err = -EIO;
+		brelse(bh);
+		goto unlock;
+	}
+	num_entries++;
+	brelse(bh);
+
+	err = exfat_remove_entries(dir, &cdir, entry, 0, num_entries);
+	if (err) {
+		exfat_msg(sb, KERN_ERR,
+				"failed to exfat_remove_entries : err(%d)",
+				err);
+		goto unlock;
+	}
+	ei->dir.dir = DIR_DELETED;
+	exfat_set_vol_flags(sb, VOL_CLEAN);
+
+	inode_inc_iversion(dir);
+	dir->i_mtime = dir->i_atime = current_time(dir);
+	exfat_truncate_atime(&dir->i_atime);
+	if (IS_DIRSYNC(dir))
+		exfat_sync_inode(dir);
+	else
+		mark_inode_dirty(dir);
+	drop_nlink(dir);
+
+	clear_nlink(inode);
+	inode->i_mtime = inode->i_atime = current_time(inode);
+	exfat_truncate_atime(&inode->i_atime);
+	exfat_unhash_inode(inode);
+	exfat_d_version_set(dentry, inode_query_iversion(dir));
+unlock:
+	mutex_unlock(&EXFAT_SB(inode->i_sb)->s_lock);
+	return err;
+}
+
+static int exfat_rename_file(struct inode *inode, struct exfat_chain *p_dir,
+		int oldentry, struct exfat_uni_name *p_uniname,
+		struct exfat_inode_info *ei)
+{
+	int ret, num_old_entries, num_new_entries;
+	sector_t sector_old, sector_new;
+	struct exfat_dentry *epold, *epnew;
+	struct super_block *sb = inode->i_sb;
+	struct buffer_head *new_bh, *old_bh;
+	int sync = IS_DIRSYNC(inode);
+
+	epold = exfat_get_dentry(sb, p_dir, oldentry, &old_bh, &sector_old);
+	if (!epold)
+		return -EIO;
+
+	num_old_entries = exfat_count_ext_entries(sb, p_dir, oldentry, epold);
+	if (num_old_entries < 0)
+		return -EIO;
+	num_old_entries++;
+
+	num_new_entries = exfat_calc_num_entries(p_uniname);
+	if (num_new_entries < 0)
+		return num_new_entries;
+
+	if (num_old_entries < num_new_entries) {
+		int newentry;
+
+		newentry =
+			exfat_find_empty_entry(inode, p_dir, num_new_entries);
+		if (newentry < 0)
+			return newentry; /* -EIO or -ENOSPC */
+
+		epnew = exfat_get_dentry(sb, p_dir, newentry, &new_bh,
+			&sector_new);
+		if (!epnew)
+			return -EIO;
+
+		memcpy(epnew, epold, DENTRY_SIZE);
+		if (exfat_get_entry_type(epnew) == TYPE_FILE) {
+			epnew->dentry.file.attr |= cpu_to_le16(ATTR_ARCHIVE);
+			ei->attr |= ATTR_ARCHIVE;
+		}
+		exfat_update_bh(sb, new_bh, sync);
+		brelse(old_bh);
+		brelse(new_bh);
+
+		epold = exfat_get_dentry(sb, p_dir, oldentry + 1, &old_bh,
+			&sector_old);
+		epnew = exfat_get_dentry(sb, p_dir, newentry + 1, &new_bh,
+			&sector_new);
+		if (!epold || !epnew)
+			return -EIO;
+
+		memcpy(epnew, epold, DENTRY_SIZE);
+		exfat_update_bh(sb, new_bh, sync);
+		brelse(old_bh);
+		brelse(new_bh);
+
+		ret = exfat_init_ext_entry(inode, p_dir, newentry,
+			num_new_entries, p_uniname);
+		if (ret)
+			return ret;
+
+		exfat_remove_entries(inode, p_dir, oldentry, 0,
+			num_old_entries);
+		ei->entry = newentry;
+	} else {
+		if (exfat_get_entry_type(epold) == TYPE_FILE) {
+			epold->dentry.file.attr |= cpu_to_le16(ATTR_ARCHIVE);
+			ei->attr |= ATTR_ARCHIVE;
+		}
+		exfat_update_bh(sb, old_bh, sync);
+		brelse(old_bh);
+		ret = exfat_init_ext_entry(inode, p_dir, oldentry,
+			num_new_entries, p_uniname);
+		if (ret)
+			return ret;
+
+		exfat_remove_entries(inode, p_dir, oldentry, num_new_entries,
+			num_old_entries);
+	}
+	return 0;
+}
+
+static int exfat_move_file(struct inode *inode, struct exfat_chain *p_olddir,
+		int oldentry, struct exfat_chain *p_newdir,
+		struct exfat_uni_name *p_uniname, struct exfat_inode_info *ei)
+{
+	int ret, newentry, num_new_entries, num_old_entries;
+	sector_t sector_mov, sector_new;
+	struct exfat_dentry *epmov, *epnew;
+	struct super_block *sb = inode->i_sb;
+	struct buffer_head *mov_bh, *new_bh;
+
+	epmov = exfat_get_dentry(sb, p_olddir, oldentry, &mov_bh, &sector_mov);
+	if (!epmov)
+		return -EIO;
+
+	/* check if the source and target directory is the same */
+	if (exfat_get_entry_type(epmov) == TYPE_DIR &&
+	    le32_to_cpu(epmov->dentry.stream.start_clu) == p_newdir->dir)
+		return -EINVAL;
+
+	num_old_entries = exfat_count_ext_entries(sb, p_olddir, oldentry,
+		epmov);
+	if (num_old_entries < 0)
+		return -EIO;
+	num_old_entries++;
+
+	num_new_entries = exfat_calc_num_entries(p_uniname);
+	if (num_new_entries < 0)
+		return num_new_entries;
+
+	newentry = exfat_find_empty_entry(inode, p_newdir, num_new_entries);
+	if (newentry < 0)
+		return newentry; /* -EIO or -ENOSPC */
+
+	epnew = exfat_get_dentry(sb, p_newdir, newentry, &new_bh, &sector_new);
+	if (!epnew)
+		return -EIO;
+
+	memcpy(epnew, epmov, DENTRY_SIZE);
+	if (exfat_get_entry_type(epnew) == TYPE_FILE) {
+		epnew->dentry.file.attr |= cpu_to_le16(ATTR_ARCHIVE);
+		ei->attr |= ATTR_ARCHIVE;
+	}
+	exfat_update_bh(sb, new_bh, IS_DIRSYNC(inode));
+	brelse(mov_bh);
+	brelse(new_bh);
+
+	epmov = exfat_get_dentry(sb, p_olddir, oldentry + 1, &mov_bh,
+		&sector_mov);
+	epnew = exfat_get_dentry(sb, p_newdir, newentry + 1, &new_bh,
+		&sector_new);
+	if (!epmov || !epnew)
+		return -EIO;
+
+	memcpy(epnew, epmov, DENTRY_SIZE);
+	exfat_update_bh(sb, new_bh, IS_DIRSYNC(inode));
+	brelse(mov_bh);
+	brelse(new_bh);
+
+	ret = exfat_init_ext_entry(inode, p_newdir, newentry, num_new_entries,
+		p_uniname);
+	if (ret)
+		return ret;
+
+	exfat_remove_entries(inode, p_olddir, oldentry, 0, num_old_entries);
+
+	exfat_chain_set(&ei->dir, p_newdir->dir, p_newdir->size,
+		p_newdir->flags);
+
+	ei->entry = newentry;
+	return 0;
+}
+
+static void exfat_update_parent_info(struct exfat_inode_info *ei,
+		struct inode *parent_inode)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(parent_inode->i_sb);
+	struct exfat_inode_info *parent_ei = EXFAT_I(parent_inode);
+	loff_t parent_isize = i_size_read(parent_inode);
+
+	/*
+	 * the problem that struct exfat_inode_info caches wrong parent info.
+	 *
+	 * because of flag-mismatch of ei->dir,
+	 * there is abnormal traversing cluster chain.
+	 */
+	if (unlikely(parent_ei->flags != ei->dir.flags ||
+		     parent_isize != EXFAT_CLU_TO_B(ei->dir.size, sbi) ||
+		     parent_ei->start_clu != ei->dir.dir)) {
+		exfat_chain_set(&ei->dir, parent_ei->start_clu,
+			EXFAT_B_TO_CLU_ROUND_UP(parent_isize, sbi),
+			parent_ei->flags);
+	}
+}
+
+/* rename or move a old file into a new file */
+static int __exfat_rename(struct inode *old_parent_inode,
+		struct exfat_inode_info *ei, struct inode *new_parent_inode,
+		struct dentry *new_dentry)
+{
+	int ret;
+	int dentry;
+	struct exfat_chain olddir, newdir;
+	struct exfat_chain *p_dir = NULL;
+	struct exfat_uni_name uni_name;
+	struct exfat_dentry *ep;
+	struct super_block *sb = old_parent_inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	const unsigned char *new_path = new_dentry->d_name.name;
+	struct inode *new_inode = new_dentry->d_inode;
+	int num_entries;
+	struct exfat_inode_info *new_ei = NULL;
+	unsigned int new_entry_type = TYPE_UNUSED;
+	int new_entry = 0;
+	struct buffer_head *old_bh, *new_bh = NULL;
+
+	/* check the validity of pointer parameters */
+	if (new_path == NULL || strlen(new_path) == 0)
+		return -EINVAL;
+
+	if (ei->dir.dir == DIR_DELETED) {
+		exfat_msg(sb, KERN_ERR,
+				"abnormal access to deleted source dentry");
+		return -ENOENT;
+	}
+
+	exfat_update_parent_info(ei, old_parent_inode);
+
+	exfat_chain_dup(&olddir, &ei->dir);
+	dentry = ei->entry;
+
+	ep = exfat_get_dentry(sb, &olddir, dentry, &old_bh, NULL);
+	if (!ep) {
+		ret = -EIO;
+		goto out;
+	}
+	brelse(old_bh);
+
+	/* check whether new dir is existing directory and empty */
+	if (new_inode) {
+		ret = -EIO;
+		new_ei = EXFAT_I(new_inode);
+
+		if (new_ei->dir.dir == DIR_DELETED) {
+			exfat_msg(sb, KERN_ERR,
+				"abnormal access to deleted target dentry");
+			goto out;
+		}
+
+		exfat_update_parent_info(new_ei, new_parent_inode);
+
+		p_dir = &(new_ei->dir);
+		new_entry = new_ei->entry;
+		ep = exfat_get_dentry(sb, p_dir, new_entry, &new_bh, NULL);
+		if (!ep)
+			goto out;
+
+		new_entry_type = exfat_get_entry_type(ep);
+		brelse(new_bh);
+
+		/* if new_inode exists, update ei */
+		if (new_entry_type == TYPE_DIR) {
+			struct exfat_chain new_clu;
+
+			new_clu.dir = new_ei->start_clu;
+			new_clu.size =
+				EXFAT_B_TO_CLU_ROUND_UP(i_size_read(new_inode),
+				sbi);
+			new_clu.flags = new_ei->flags;
+
+			ret = exfat_check_dir_empty(sb, &new_clu);
+			if (ret)
+				goto out;
+		}
+	}
+
+	/* check the validity of directory name in the given new pathname */
+	ret = exfat_resolve_path(new_parent_inode, new_path, &newdir,
+			&uni_name);
+	if (ret)
+		goto out;
+
+	exfat_set_vol_flags(sb, VOL_DIRTY);
+
+	if (olddir.dir == newdir.dir)
+		ret = exfat_rename_file(new_parent_inode, &olddir, dentry,
+				&uni_name, ei);
+	else
+		ret = exfat_move_file(new_parent_inode, &olddir, dentry,
+				&newdir, &uni_name, ei);
+
+	if (!ret && new_inode) {
+		/* delete entries of new_dir */
+		ep = exfat_get_dentry(sb, p_dir, new_entry, &new_bh, NULL);
+		if (!ep) {
+			ret = -EIO;
+			goto del_out;
+		}
+
+		num_entries = exfat_count_ext_entries(sb, p_dir, new_entry, ep);
+		if (num_entries < 0) {
+			ret = -EIO;
+			goto del_out;
+		}
+		brelse(new_bh);
+
+		if (exfat_remove_entries(new_inode, p_dir, new_entry, 0,
+				num_entries + 1)) {
+			ret = -EIO;
+			goto del_out;
+		}
+
+		/* Free the clusters if new_inode is a dir(as if exfat_rmdir) */
+		if (new_entry_type == TYPE_DIR) {
+			/* new_ei, new_clu_to_free */
+			struct exfat_chain new_clu_to_free;
+
+			exfat_chain_set(&new_clu_to_free, new_ei->start_clu,
+				EXFAT_B_TO_CLU_ROUND_UP(i_size_read(new_inode),
+				sbi), new_ei->flags);
+
+			if (exfat_free_cluster(new_inode, &new_clu_to_free)) {
+				/* just set I/O error only */
+				ret = -EIO;
+			}
+
+			i_size_write(new_inode, 0);
+			new_ei->start_clu = EXFAT_EOF_CLUSTER;
+			new_ei->flags = ALLOC_NO_FAT_CHAIN;
+		}
+del_out:
+		/* Update new_inode ei
+		 * Prevent syncing removed new_inode
+		 * (new_ei is already initialized above code ("if (new_inode)")
+		 */
+		new_ei->dir.dir = DIR_DELETED;
+	}
+	exfat_set_vol_flags(sb, VOL_CLEAN);
+out:
+	return ret;
+}
+
+static int exfat_rename(struct inode *old_dir, struct dentry *old_dentry,
+		struct inode *new_dir, struct dentry *new_dentry,
+		unsigned int flags)
+{
+	struct inode *old_inode, *new_inode;
+	struct super_block *sb = old_dir->i_sb;
+	loff_t i_pos;
+	int err;
+
+	/*
+	 * The VFS already checks for existence, so for local filesystems
+	 * the RENAME_NOREPLACE implementation is equivalent to plain rename.
+	 * Don't support any other flags
+	 */
+	if (flags & ~RENAME_NOREPLACE)
+		return -EINVAL;
+
+	mutex_lock(&EXFAT_SB(sb)->s_lock);
+	old_inode = old_dentry->d_inode;
+	new_inode = new_dentry->d_inode;
+
+	err = __exfat_rename(old_dir, EXFAT_I(old_inode), new_dir, new_dentry);
+	if (err)
+		goto unlock;
+
+	inode_inc_iversion(new_dir);
+	new_dir->i_ctime = new_dir->i_mtime = new_dir->i_atime =
+		EXFAT_I(new_dir)->i_crtime = current_time(new_dir);
+	exfat_truncate_atime(&new_dir->i_atime);
+	if (IS_DIRSYNC(new_dir))
+		exfat_sync_inode(new_dir);
+	else
+		mark_inode_dirty(new_dir);
+
+	i_pos = ((loff_t)EXFAT_I(old_inode)->dir.dir << 32) |
+		(EXFAT_I(old_inode)->entry & 0xffffffff);
+	exfat_unhash_inode(old_inode);
+	exfat_hash_inode(old_inode, i_pos);
+	if (IS_DIRSYNC(new_dir))
+		exfat_sync_inode(old_inode);
+	else
+		mark_inode_dirty(old_inode);
+
+	if (S_ISDIR(old_inode->i_mode) && old_dir != new_dir) {
+		drop_nlink(old_dir);
+		if (!new_inode)
+			inc_nlink(new_dir);
+	}
+
+	inode_inc_iversion(old_dir);
+	old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
+	if (IS_DIRSYNC(old_dir))
+		exfat_sync_inode(old_dir);
+	else
+		mark_inode_dirty(old_dir);
+
+	if (new_inode) {
+		exfat_unhash_inode(new_inode);
+
+		/* skip drop_nlink if new_inode already has been dropped */
+		if (new_inode->i_nlink) {
+			drop_nlink(new_inode);
+			if (S_ISDIR(new_inode->i_mode))
+				drop_nlink(new_inode);
+		} else {
+			exfat_msg(sb, KERN_WARNING,
+					"abnormal access to an inode dropped");
+			WARN_ON(new_inode->i_nlink == 0);
+		}
+		new_inode->i_ctime = EXFAT_I(new_inode)->i_crtime =
+			current_time(new_inode);
+	}
+
+unlock:
+	mutex_unlock(&EXFAT_SB(sb)->s_lock);
+	return err;
+}
+
+const struct inode_operations exfat_dir_inode_operations = {
+	.create		= exfat_create,
+	.lookup		= exfat_lookup,
+	.unlink		= exfat_unlink,
+	.mkdir		= exfat_mkdir,
+	.rmdir		= exfat_rmdir,
+	.rename		= exfat_rename,
+	.setattr	= exfat_setattr,
+	.getattr	= exfat_getattr,
+};
diff --git a/fs/exfat/nls.c b/fs/exfat/nls.c
new file mode 100644
index 000000000000..6d1c3ae130ff
--- /dev/null
+++ b/fs/exfat/nls.c
@@ -0,0 +1,831 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/buffer_head.h>
+#include <asm/unaligned.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+/* Upcase tabel macro */
+#define EXFAT_NUM_UPCASE	(2918)
+#define UTBL_COUNT		(0x10000)
+
+/*
+ * Upcase table in compressed format (7.2.5.1 Recommended Up-case Table
+ * in exfat specification, See:
+ * https://docs.microsoft.com/en-us/windows/win32/fileio/exfat-specification).
+ */
+static const unsigned short uni_def_upcase[EXFAT_NUM_UPCASE] = {
+	0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007,
+	0x0008, 0x0009, 0x000a, 0x000b, 0x000c, 0x000d, 0x000e, 0x000f,
+	0x0010, 0x0011, 0x0012, 0x0013, 0x0014, 0x0015, 0x0016, 0x0017,
+	0x0018, 0x0019, 0x001a, 0x001b, 0x001c, 0x001d, 0x001e, 0x001f,
+	0x0020, 0x0021, 0x0022, 0x0023, 0x0024, 0x0025, 0x0026, 0x0027,
+	0x0028, 0x0029, 0x002a, 0x002b, 0x002c, 0x002d, 0x002e, 0x002f,
+	0x0030, 0x0031, 0x0032, 0x0033, 0x0034, 0x0035, 0x0036, 0x0037,
+	0x0038, 0x0039, 0x003a, 0x003b, 0x003c, 0x003d, 0x003e, 0x003f,
+	0x0040, 0x0041, 0x0042, 0x0043, 0x0044, 0x0045, 0x0046, 0x0047,
+	0x0048, 0x0049, 0x004a, 0x004b, 0x004c, 0x004d, 0x004e, 0x004f,
+	0x0050, 0x0051, 0x0052, 0x0053, 0x0054, 0x0055, 0x0056, 0x0057,
+	0x0058, 0x0059, 0x005a, 0x005b, 0x005c, 0x005d, 0x005e, 0x005f,
+	0x0060, 0x0041, 0x0042, 0x0043, 0x0044, 0x0045, 0x0046, 0x0047,
+	0x0048, 0x0049, 0x004a, 0x004b, 0x004c, 0x004d, 0x004e, 0x004f,
+	0x0050, 0x0051, 0x0052, 0x0053, 0x0054, 0x0055, 0x0056, 0x0057,
+	0x0058, 0x0059, 0x005a, 0x007b, 0x007c, 0x007d, 0x007e, 0x007f,
+	0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+	0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+	0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+	0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+	0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x00a4, 0x00a5, 0x00a6, 0x00a7,
+	0x00a8, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
+	0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
+	0x00b8, 0x00b9, 0x00ba, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x00bf,
+	0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
+	0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
+	0x00d0, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
+	0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df,
+	0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
+	0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
+	0x00d0, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00f7,
+	0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x0178,
+	0x0100, 0x0100, 0x0102, 0x0102, 0x0104, 0x0104, 0x0106, 0x0106,
+	0x0108, 0x0108, 0x010a, 0x010a, 0x010c, 0x010c, 0x010e, 0x010e,
+	0x0110, 0x0110, 0x0112, 0x0112, 0x0114, 0x0114, 0x0116, 0x0116,
+	0x0118, 0x0118, 0x011a, 0x011a, 0x011c, 0x011c, 0x011e, 0x011e,
+	0x0120, 0x0120, 0x0122, 0x0122, 0x0124, 0x0124, 0x0126, 0x0126,
+	0x0128, 0x0128, 0x012a, 0x012a, 0x012c, 0x012c, 0x012e, 0x012e,
+	0x0130, 0x0131, 0x0132, 0x0132, 0x0134, 0x0134, 0x0136, 0x0136,
+	0x0138, 0x0139, 0x0139, 0x013b, 0x013b, 0x013d, 0x013d, 0x013f,
+	0x013f, 0x0141, 0x0141, 0x0143, 0x0143, 0x0145, 0x0145, 0x0147,
+	0x0147, 0x0149, 0x014a, 0x014a, 0x014c, 0x014c, 0x014e, 0x014e,
+	0x0150, 0x0150, 0x0152, 0x0152, 0x0154, 0x0154, 0x0156, 0x0156,
+	0x0158, 0x0158, 0x015a, 0x015a, 0x015c, 0x015c, 0x015e, 0x015e,
+	0x0160, 0x0160, 0x0162, 0x0162, 0x0164, 0x0164, 0x0166, 0x0166,
+	0x0168, 0x0168, 0x016a, 0x016a, 0x016c, 0x016c, 0x016e, 0x016e,
+	0x0170, 0x0170, 0x0172, 0x0172, 0x0174, 0x0174, 0x0176, 0x0176,
+	0x0178, 0x0179, 0x0179, 0x017b, 0x017b, 0x017d, 0x017d, 0x017f,
+	0x0243, 0x0181, 0x0182, 0x0182, 0x0184, 0x0184, 0x0186, 0x0187,
+	0x0187, 0x0189, 0x018a, 0x018b, 0x018b, 0x018d, 0x018e, 0x018f,
+	0x0190, 0x0191, 0x0191, 0x0193, 0x0194, 0x01f6, 0x0196, 0x0197,
+	0x0198, 0x0198, 0x023d, 0x019b, 0x019c, 0x019d, 0x0220, 0x019f,
+	0x01a0, 0x01a0, 0x01a2, 0x01a2, 0x01a4, 0x01a4, 0x01a6, 0x01a7,
+	0x01a7, 0x01a9, 0x01aa, 0x01ab, 0x01ac, 0x01ac, 0x01ae, 0x01af,
+	0x01af, 0x01b1, 0x01b2, 0x01b3, 0x01b3, 0x01b5, 0x01b5, 0x01b7,
+	0x01b8, 0x01b8, 0x01ba, 0x01bb, 0x01bc, 0x01bc, 0x01be, 0x01f7,
+	0x01c0, 0x01c1, 0x01c2, 0x01c3, 0x01c4, 0x01c5, 0x01c4, 0x01c7,
+	0x01c8, 0x01c7, 0x01ca, 0x01cb, 0x01ca, 0x01cd, 0x01cd, 0x01cf,
+	0x01cf, 0x01d1, 0x01d1, 0x01d3, 0x01d3, 0x01d5, 0x01d5, 0x01d7,
+	0x01d7, 0x01d9, 0x01d9, 0x01db, 0x01db, 0x018e, 0x01de, 0x01de,
+	0x01e0, 0x01e0, 0x01e2, 0x01e2, 0x01e4, 0x01e4, 0x01e6, 0x01e6,
+	0x01e8, 0x01e8, 0x01ea, 0x01ea, 0x01ec, 0x01ec, 0x01ee, 0x01ee,
+	0x01f0, 0x01f1, 0x01f2, 0x01f1, 0x01f4, 0x01f4, 0x01f6, 0x01f7,
+	0x01f8, 0x01f8, 0x01fa, 0x01fa, 0x01fc, 0x01fc, 0x01fe, 0x01fe,
+	0x0200, 0x0200, 0x0202, 0x0202, 0x0204, 0x0204, 0x0206, 0x0206,
+	0x0208, 0x0208, 0x020a, 0x020a, 0x020c, 0x020c, 0x020e, 0x020e,
+	0x0210, 0x0210, 0x0212, 0x0212, 0x0214, 0x0214, 0x0216, 0x0216,
+	0x0218, 0x0218, 0x021a, 0x021a, 0x021c, 0x021c, 0x021e, 0x021e,
+	0x0220, 0x0221, 0x0222, 0x0222, 0x0224, 0x0224, 0x0226, 0x0226,
+	0x0228, 0x0228, 0x022a, 0x022a, 0x022c, 0x022c, 0x022e, 0x022e,
+	0x0230, 0x0230, 0x0232, 0x0232, 0x0234, 0x0235, 0x0236, 0x0237,
+	0x0238, 0x0239, 0x2c65, 0x023b, 0x023b, 0x023d, 0x2c66, 0x023f,
+	0x0240, 0x0241, 0x0241, 0x0243, 0x0244, 0x0245, 0x0246, 0x0246,
+	0x0248, 0x0248, 0x024a, 0x024a, 0x024c, 0x024c, 0x024e, 0x024e,
+	0x0250, 0x0251, 0x0252, 0x0181, 0x0186, 0x0255, 0x0189, 0x018a,
+	0x0258, 0x018f, 0x025a, 0x0190, 0x025c, 0x025d, 0x025e, 0x025f,
+	0x0193, 0x0261, 0x0262, 0x0194, 0x0264, 0x0265, 0x0266, 0x0267,
+	0x0197, 0x0196, 0x026a, 0x2c62, 0x026c, 0x026d, 0x026e, 0x019c,
+	0x0270, 0x0271, 0x019d, 0x0273, 0x0274, 0x019f, 0x0276, 0x0277,
+	0x0278, 0x0279, 0x027a, 0x027b, 0x027c, 0x2c64, 0x027e, 0x027f,
+	0x01a6, 0x0281, 0x0282, 0x01a9, 0x0284, 0x0285, 0x0286, 0x0287,
+	0x01ae, 0x0244, 0x01b1, 0x01b2, 0x0245, 0x028d, 0x028e, 0x028f,
+	0x0290, 0x0291, 0x01b7, 0x0293, 0x0294, 0x0295, 0x0296, 0x0297,
+	0x0298, 0x0299, 0x029a, 0x029b, 0x029c, 0x029d, 0x029e, 0x029f,
+	0x02a0, 0x02a1, 0x02a2, 0x02a3, 0x02a4, 0x02a5, 0x02a6, 0x02a7,
+	0x02a8, 0x02a9, 0x02aa, 0x02ab, 0x02ac, 0x02ad, 0x02ae, 0x02af,
+	0x02b0, 0x02b1, 0x02b2, 0x02b3, 0x02b4, 0x02b5, 0x02b6, 0x02b7,
+	0x02b8, 0x02b9, 0x02ba, 0x02bb, 0x02bc, 0x02bd, 0x02be, 0x02bf,
+	0x02c0, 0x02c1, 0x02c2, 0x02c3, 0x02c4, 0x02c5, 0x02c6, 0x02c7,
+	0x02c8, 0x02c9, 0x02ca, 0x02cb, 0x02cc, 0x02cd, 0x02ce, 0x02cf,
+	0x02d0, 0x02d1, 0x02d2, 0x02d3, 0x02d4, 0x02d5, 0x02d6, 0x02d7,
+	0x02d8, 0x02d9, 0x02da, 0x02db, 0x02dc, 0x02dd, 0x02de, 0x02df,
+	0x02e0, 0x02e1, 0x02e2, 0x02e3, 0x02e4, 0x02e5, 0x02e6, 0x02e7,
+	0x02e8, 0x02e9, 0x02ea, 0x02eb, 0x02ec, 0x02ed, 0x02ee, 0x02ef,
+	0x02f0, 0x02f1, 0x02f2, 0x02f3, 0x02f4, 0x02f5, 0x02f6, 0x02f7,
+	0x02f8, 0x02f9, 0x02fa, 0x02fb, 0x02fc, 0x02fd, 0x02fe, 0x02ff,
+	0x0300, 0x0301, 0x0302, 0x0303, 0x0304, 0x0305, 0x0306, 0x0307,
+	0x0308, 0x0309, 0x030a, 0x030b, 0x030c, 0x030d, 0x030e, 0x030f,
+	0x0310, 0x0311, 0x0312, 0x0313, 0x0314, 0x0315, 0x0316, 0x0317,
+	0x0318, 0x0319, 0x031a, 0x031b, 0x031c, 0x031d, 0x031e, 0x031f,
+	0x0320, 0x0321, 0x0322, 0x0323, 0x0324, 0x0325, 0x0326, 0x0327,
+	0x0328, 0x0329, 0x032a, 0x032b, 0x032c, 0x032d, 0x032e, 0x032f,
+	0x0330, 0x0331, 0x0332, 0x0333, 0x0334, 0x0335, 0x0336, 0x0337,
+	0x0338, 0x0339, 0x033a, 0x033b, 0x033c, 0x033d, 0x033e, 0x033f,
+	0x0340, 0x0341, 0x0342, 0x0343, 0x0344, 0x0345, 0x0346, 0x0347,
+	0x0348, 0x0349, 0x034a, 0x034b, 0x034c, 0x034d, 0x034e, 0x034f,
+	0x0350, 0x0351, 0x0352, 0x0353, 0x0354, 0x0355, 0x0356, 0x0357,
+	0x0358, 0x0359, 0x035a, 0x035b, 0x035c, 0x035d, 0x035e, 0x035f,
+	0x0360, 0x0361, 0x0362, 0x0363, 0x0364, 0x0365, 0x0366, 0x0367,
+	0x0368, 0x0369, 0x036a, 0x036b, 0x036c, 0x036d, 0x036e, 0x036f,
+	0x0370, 0x0371, 0x0372, 0x0373, 0x0374, 0x0375, 0x0376, 0x0377,
+	0x0378, 0x0379, 0x037a, 0x03fd, 0x03fe, 0x03ff, 0x037e, 0x037f,
+	0x0380, 0x0381, 0x0382, 0x0383, 0x0384, 0x0385, 0x0386, 0x0387,
+	0x0388, 0x0389, 0x038a, 0x038b, 0x038c, 0x038d, 0x038e, 0x038f,
+	0x0390, 0x0391, 0x0392, 0x0393, 0x0394, 0x0395, 0x0396, 0x0397,
+	0x0398, 0x0399, 0x039a, 0x039b, 0x039c, 0x039d, 0x039e, 0x039f,
+	0x03a0, 0x03a1, 0x03a2, 0x03a3, 0x03a4, 0x03a5, 0x03a6, 0x03a7,
+	0x03a8, 0x03a9, 0x03aa, 0x03ab, 0x0386, 0x0388, 0x0389, 0x038a,
+	0x03b0, 0x0391, 0x0392, 0x0393, 0x0394, 0x0395, 0x0396, 0x0397,
+	0x0398, 0x0399, 0x039a, 0x039b, 0x039c, 0x039d, 0x039e, 0x039f,
+	0x03a0, 0x03a1, 0x03a3, 0x03a3, 0x03a4, 0x03a5, 0x03a6, 0x03a7,
+	0x03a8, 0x03a9, 0x03aa, 0x03ab, 0x038c, 0x038e, 0x038f, 0x03cf,
+	0x03d0, 0x03d1, 0x03d2, 0x03d3, 0x03d4, 0x03d5, 0x03d6, 0x03d7,
+	0x03d8, 0x03d8, 0x03da, 0x03da, 0x03dc, 0x03dc, 0x03de, 0x03de,
+	0x03e0, 0x03e0, 0x03e2, 0x03e2, 0x03e4, 0x03e4, 0x03e6, 0x03e6,
+	0x03e8, 0x03e8, 0x03ea, 0x03ea, 0x03ec, 0x03ec, 0x03ee, 0x03ee,
+	0x03f0, 0x03f1, 0x03f9, 0x03f3, 0x03f4, 0x03f5, 0x03f6, 0x03f7,
+	0x03f7, 0x03f9, 0x03fa, 0x03fa, 0x03fc, 0x03fd, 0x03fe, 0x03ff,
+	0x0400, 0x0401, 0x0402, 0x0403, 0x0404, 0x0405, 0x0406, 0x0407,
+	0x0408, 0x0409, 0x040a, 0x040b, 0x040c, 0x040d, 0x040e, 0x040f,
+	0x0410, 0x0411, 0x0412, 0x0413, 0x0414, 0x0415, 0x0416, 0x0417,
+	0x0418, 0x0419, 0x041a, 0x041b, 0x041c, 0x041d, 0x041e, 0x041f,
+	0x0420, 0x0421, 0x0422, 0x0423, 0x0424, 0x0425, 0x0426, 0x0427,
+	0x0428, 0x0429, 0x042a, 0x042b, 0x042c, 0x042d, 0x042e, 0x042f,
+	0x0410, 0x0411, 0x0412, 0x0413, 0x0414, 0x0415, 0x0416, 0x0417,
+	0x0418, 0x0419, 0x041a, 0x041b, 0x041c, 0x041d, 0x041e, 0x041f,
+	0x0420, 0x0421, 0x0422, 0x0423, 0x0424, 0x0425, 0x0426, 0x0427,
+	0x0428, 0x0429, 0x042a, 0x042b, 0x042c, 0x042d, 0x042e, 0x042f,
+	0x0400, 0x0401, 0x0402, 0x0403, 0x0404, 0x0405, 0x0406, 0x0407,
+	0x0408, 0x0409, 0x040a, 0x040b, 0x040c, 0x040d, 0x040e, 0x040f,
+	0x0460, 0x0460, 0x0462, 0x0462, 0x0464, 0x0464, 0x0466, 0x0466,
+	0x0468, 0x0468, 0x046a, 0x046a, 0x046c, 0x046c, 0x046e, 0x046e,
+	0x0470, 0x0470, 0x0472, 0x0472, 0x0474, 0x0474, 0x0476, 0x0476,
+	0x0478, 0x0478, 0x047a, 0x047a, 0x047c, 0x047c, 0x047e, 0x047e,
+	0x0480, 0x0480, 0x0482, 0x0483, 0x0484, 0x0485, 0x0486, 0x0487,
+	0x0488, 0x0489, 0x048a, 0x048a, 0x048c, 0x048c, 0x048e, 0x048e,
+	0x0490, 0x0490, 0x0492, 0x0492, 0x0494, 0x0494, 0x0496, 0x0496,
+	0x0498, 0x0498, 0x049a, 0x049a, 0x049c, 0x049c, 0x049e, 0x049e,
+	0x04a0, 0x04a0, 0x04a2, 0x04a2, 0x04a4, 0x04a4, 0x04a6, 0x04a6,
+	0x04a8, 0x04a8, 0x04aa, 0x04aa, 0x04ac, 0x04ac, 0x04ae, 0x04ae,
+	0x04b0, 0x04b0, 0x04b2, 0x04b2, 0x04b4, 0x04b4, 0x04b6, 0x04b6,
+	0x04b8, 0x04b8, 0x04ba, 0x04ba, 0x04bc, 0x04bc, 0x04be, 0x04be,
+	0x04c0, 0x04c1, 0x04c1, 0x04c3, 0x04c3, 0x04c5, 0x04c5, 0x04c7,
+	0x04c7, 0x04c9, 0x04c9, 0x04cb, 0x04cb, 0x04cd, 0x04cd, 0x04c0,
+	0x04d0, 0x04d0, 0x04d2, 0x04d2, 0x04d4, 0x04d4, 0x04d6, 0x04d6,
+	0x04d8, 0x04d8, 0x04da, 0x04da, 0x04dc, 0x04dc, 0x04de, 0x04de,
+	0x04e0, 0x04e0, 0x04e2, 0x04e2, 0x04e4, 0x04e4, 0x04e6, 0x04e6,
+	0x04e8, 0x04e8, 0x04ea, 0x04ea, 0x04ec, 0x04ec, 0x04ee, 0x04ee,
+	0x04f0, 0x04f0, 0x04f2, 0x04f2, 0x04f4, 0x04f4, 0x04f6, 0x04f6,
+	0x04f8, 0x04f8, 0x04fa, 0x04fa, 0x04fc, 0x04fc, 0x04fe, 0x04fe,
+	0x0500, 0x0500, 0x0502, 0x0502, 0x0504, 0x0504, 0x0506, 0x0506,
+	0x0508, 0x0508, 0x050a, 0x050a, 0x050c, 0x050c, 0x050e, 0x050e,
+	0x0510, 0x0510, 0x0512, 0x0512, 0x0514, 0x0515, 0x0516, 0x0517,
+	0x0518, 0x0519, 0x051a, 0x051b, 0x051c, 0x051d, 0x051e, 0x051f,
+	0x0520, 0x0521, 0x0522, 0x0523, 0x0524, 0x0525, 0x0526, 0x0527,
+	0x0528, 0x0529, 0x052a, 0x052b, 0x052c, 0x052d, 0x052e, 0x052f,
+	0x0530, 0x0531, 0x0532, 0x0533, 0x0534, 0x0535, 0x0536, 0x0537,
+	0x0538, 0x0539, 0x053a, 0x053b, 0x053c, 0x053d, 0x053e, 0x053f,
+	0x0540, 0x0541, 0x0542, 0x0543, 0x0544, 0x0545, 0x0546, 0x0547,
+	0x0548, 0x0549, 0x054a, 0x054b, 0x054c, 0x054d, 0x054e, 0x054f,
+	0x0550, 0x0551, 0x0552, 0x0553, 0x0554, 0x0555, 0x0556, 0x0557,
+	0x0558, 0x0559, 0x055a, 0x055b, 0x055c, 0x055d, 0x055e, 0x055f,
+	0x0560, 0x0531, 0x0532, 0x0533, 0x0534, 0x0535, 0x0536, 0x0537,
+	0x0538, 0x0539, 0x053a, 0x053b, 0x053c, 0x053d, 0x053e, 0x053f,
+	0x0540, 0x0541, 0x0542, 0x0543, 0x0544, 0x0545, 0x0546, 0x0547,
+	0x0548, 0x0549, 0x054a, 0x054b, 0x054c, 0x054d, 0x054e, 0x054f,
+	0x0550, 0x0551, 0x0552, 0x0553, 0x0554, 0x0555, 0x0556, 0xffff,
+	0x17f6, 0x2c63, 0x1d7e, 0x1d7f, 0x1d80, 0x1d81, 0x1d82, 0x1d83,
+	0x1d84, 0x1d85, 0x1d86, 0x1d87, 0x1d88, 0x1d89, 0x1d8a, 0x1d8b,
+	0x1d8c, 0x1d8d, 0x1d8e, 0x1d8f, 0x1d90, 0x1d91, 0x1d92, 0x1d93,
+	0x1d94, 0x1d95, 0x1d96, 0x1d97, 0x1d98, 0x1d99, 0x1d9a, 0x1d9b,
+	0x1d9c, 0x1d9d, 0x1d9e, 0x1d9f, 0x1da0, 0x1da1, 0x1da2, 0x1da3,
+	0x1da4, 0x1da5, 0x1da6, 0x1da7, 0x1da8, 0x1da9, 0x1daa, 0x1dab,
+	0x1dac, 0x1dad, 0x1dae, 0x1daf, 0x1db0, 0x1db1, 0x1db2, 0x1db3,
+	0x1db4, 0x1db5, 0x1db6, 0x1db7, 0x1db8, 0x1db9, 0x1dba, 0x1dbb,
+	0x1dbc, 0x1dbd, 0x1dbe, 0x1dbf, 0x1dc0, 0x1dc1, 0x1dc2, 0x1dc3,
+	0x1dc4, 0x1dc5, 0x1dc6, 0x1dc7, 0x1dc8, 0x1dc9, 0x1dca, 0x1dcb,
+	0x1dcc, 0x1dcd, 0x1dce, 0x1dcf, 0x1dd0, 0x1dd1, 0x1dd2, 0x1dd3,
+	0x1dd4, 0x1dd5, 0x1dd6, 0x1dd7, 0x1dd8, 0x1dd9, 0x1dda, 0x1ddb,
+	0x1ddc, 0x1ddd, 0x1dde, 0x1ddf, 0x1de0, 0x1de1, 0x1de2, 0x1de3,
+	0x1de4, 0x1de5, 0x1de6, 0x1de7, 0x1de8, 0x1de9, 0x1dea, 0x1deb,
+	0x1dec, 0x1ded, 0x1dee, 0x1def, 0x1df0, 0x1df1, 0x1df2, 0x1df3,
+	0x1df4, 0x1df5, 0x1df6, 0x1df7, 0x1df8, 0x1df9, 0x1dfa, 0x1dfb,
+	0x1dfc, 0x1dfd, 0x1dfe, 0x1dff, 0x1e00, 0x1e00, 0x1e02, 0x1e02,
+	0x1e04, 0x1e04, 0x1e06, 0x1e06, 0x1e08, 0x1e08, 0x1e0a, 0x1e0a,
+	0x1e0c, 0x1e0c, 0x1e0e, 0x1e0e, 0x1e10, 0x1e10, 0x1e12, 0x1e12,
+	0x1e14, 0x1e14, 0x1e16, 0x1e16, 0x1e18, 0x1e18, 0x1e1a, 0x1e1a,
+	0x1e1c, 0x1e1c, 0x1e1e, 0x1e1e, 0x1e20, 0x1e20, 0x1e22, 0x1e22,
+	0x1e24, 0x1e24, 0x1e26, 0x1e26, 0x1e28, 0x1e28, 0x1e2a, 0x1e2a,
+	0x1e2c, 0x1e2c, 0x1e2e, 0x1e2e, 0x1e30, 0x1e30, 0x1e32, 0x1e32,
+	0x1e34, 0x1e34, 0x1e36, 0x1e36, 0x1e38, 0x1e38, 0x1e3a, 0x1e3a,
+	0x1e3c, 0x1e3c, 0x1e3e, 0x1e3e, 0x1e40, 0x1e40, 0x1e42, 0x1e42,
+	0x1e44, 0x1e44, 0x1e46, 0x1e46, 0x1e48, 0x1e48, 0x1e4a, 0x1e4a,
+	0x1e4c, 0x1e4c, 0x1e4e, 0x1e4e, 0x1e50, 0x1e50, 0x1e52, 0x1e52,
+	0x1e54, 0x1e54, 0x1e56, 0x1e56, 0x1e58, 0x1e58, 0x1e5a, 0x1e5a,
+	0x1e5c, 0x1e5c, 0x1e5e, 0x1e5e, 0x1e60, 0x1e60, 0x1e62, 0x1e62,
+	0x1e64, 0x1e64, 0x1e66, 0x1e66, 0x1e68, 0x1e68, 0x1e6a, 0x1e6a,
+	0x1e6c, 0x1e6c, 0x1e6e, 0x1e6e, 0x1e70, 0x1e70, 0x1e72, 0x1e72,
+	0x1e74, 0x1e74, 0x1e76, 0x1e76, 0x1e78, 0x1e78, 0x1e7a, 0x1e7a,
+	0x1e7c, 0x1e7c, 0x1e7e, 0x1e7e, 0x1e80, 0x1e80, 0x1e82, 0x1e82,
+	0x1e84, 0x1e84, 0x1e86, 0x1e86, 0x1e88, 0x1e88, 0x1e8a, 0x1e8a,
+	0x1e8c, 0x1e8c, 0x1e8e, 0x1e8e, 0x1e90, 0x1e90, 0x1e92, 0x1e92,
+	0x1e94, 0x1e94, 0x1e96, 0x1e97, 0x1e98, 0x1e99, 0x1e9a, 0x1e9b,
+	0x1e9c, 0x1e9d, 0x1e9e, 0x1e9f, 0x1ea0, 0x1ea0, 0x1ea2, 0x1ea2,
+	0x1ea4, 0x1ea4, 0x1ea6, 0x1ea6, 0x1ea8, 0x1ea8, 0x1eaa, 0x1eaa,
+	0x1eac, 0x1eac, 0x1eae, 0x1eae, 0x1eb0, 0x1eb0, 0x1eb2, 0x1eb2,
+	0x1eb4, 0x1eb4, 0x1eb6, 0x1eb6, 0x1eb8, 0x1eb8, 0x1eba, 0x1eba,
+	0x1ebc, 0x1ebc, 0x1ebe, 0x1ebe, 0x1ec0, 0x1ec0, 0x1ec2, 0x1ec2,
+	0x1ec4, 0x1ec4, 0x1ec6, 0x1ec6, 0x1ec8, 0x1ec8, 0x1eca, 0x1eca,
+	0x1ecc, 0x1ecc, 0x1ece, 0x1ece, 0x1ed0, 0x1ed0, 0x1ed2, 0x1ed2,
+	0x1ed4, 0x1ed4, 0x1ed6, 0x1ed6, 0x1ed8, 0x1ed8, 0x1eda, 0x1eda,
+	0x1edc, 0x1edc, 0x1ede, 0x1ede, 0x1ee0, 0x1ee0, 0x1ee2, 0x1ee2,
+	0x1ee4, 0x1ee4, 0x1ee6, 0x1ee6, 0x1ee8, 0x1ee8, 0x1eea, 0x1eea,
+	0x1eec, 0x1eec, 0x1eee, 0x1eee, 0x1ef0, 0x1ef0, 0x1ef2, 0x1ef2,
+	0x1ef4, 0x1ef4, 0x1ef6, 0x1ef6, 0x1ef8, 0x1ef8, 0x1efa, 0x1efb,
+	0x1efc, 0x1efd, 0x1efe, 0x1eff, 0x1f08, 0x1f09, 0x1f0a, 0x1f0b,
+	0x1f0c, 0x1f0d, 0x1f0e, 0x1f0f, 0x1f08, 0x1f09, 0x1f0a, 0x1f0b,
+	0x1f0c, 0x1f0d, 0x1f0e, 0x1f0f, 0x1f18, 0x1f19, 0x1f1a, 0x1f1b,
+	0x1f1c, 0x1f1d, 0x1f16, 0x1f17, 0x1f18, 0x1f19, 0x1f1a, 0x1f1b,
+	0x1f1c, 0x1f1d, 0x1f1e, 0x1f1f, 0x1f28, 0x1f29, 0x1f2a, 0x1f2b,
+	0x1f2c, 0x1f2d, 0x1f2e, 0x1f2f, 0x1f28, 0x1f29, 0x1f2a, 0x1f2b,
+	0x1f2c, 0x1f2d, 0x1f2e, 0x1f2f, 0x1f38, 0x1f39, 0x1f3a, 0x1f3b,
+	0x1f3c, 0x1f3d, 0x1f3e, 0x1f3f, 0x1f38, 0x1f39, 0x1f3a, 0x1f3b,
+	0x1f3c, 0x1f3d, 0x1f3e, 0x1f3f, 0x1f48, 0x1f49, 0x1f4a, 0x1f4b,
+	0x1f4c, 0x1f4d, 0x1f46, 0x1f47, 0x1f48, 0x1f49, 0x1f4a, 0x1f4b,
+	0x1f4c, 0x1f4d, 0x1f4e, 0x1f4f, 0x1f50, 0x1f59, 0x1f52, 0x1f5b,
+	0x1f54, 0x1f5d, 0x1f56, 0x1f5f, 0x1f58, 0x1f59, 0x1f5a, 0x1f5b,
+	0x1f5c, 0x1f5d, 0x1f5e, 0x1f5f, 0x1f68, 0x1f69, 0x1f6a, 0x1f6b,
+	0x1f6c, 0x1f6d, 0x1f6e, 0x1f6f, 0x1f68, 0x1f69, 0x1f6a, 0x1f6b,
+	0x1f6c, 0x1f6d, 0x1f6e, 0x1f6f, 0x1fba, 0x1fbb, 0x1fc8, 0x1fc9,
+	0x1fca, 0x1fcb, 0x1fda, 0x1fdb, 0x1ff8, 0x1ff9, 0x1fea, 0x1feb,
+	0x1ffa, 0x1ffb, 0x1f7e, 0x1f7f, 0x1f88, 0x1f89, 0x1f8a, 0x1f8b,
+	0x1f8c, 0x1f8d, 0x1f8e, 0x1f8f, 0x1f88, 0x1f89, 0x1f8a, 0x1f8b,
+	0x1f8c, 0x1f8d, 0x1f8e, 0x1f8f, 0x1f98, 0x1f99, 0x1f9a, 0x1f9b,
+	0x1f9c, 0x1f9d, 0x1f9e, 0x1f9f, 0x1f98, 0x1f99, 0x1f9a, 0x1f9b,
+	0x1f9c, 0x1f9d, 0x1f9e, 0x1f9f, 0x1fa8, 0x1fa9, 0x1faa, 0x1fab,
+	0x1fac, 0x1fad, 0x1fae, 0x1faf, 0x1fa8, 0x1fa9, 0x1faa, 0x1fab,
+	0x1fac, 0x1fad, 0x1fae, 0x1faf, 0x1fb8, 0x1fb9, 0x1fb2, 0x1fbc,
+	0x1fb4, 0x1fb5, 0x1fb6, 0x1fb7, 0x1fb8, 0x1fb9, 0x1fba, 0x1fbb,
+	0x1fbc, 0x1fbd, 0x1fbe, 0x1fbf, 0x1fc0, 0x1fc1, 0x1fc2, 0x1fc3,
+	0x1fc4, 0x1fc5, 0x1fc6, 0x1fc7, 0x1fc8, 0x1fc9, 0x1fca, 0x1fcb,
+	0x1fc3, 0x1fcd, 0x1fce, 0x1fcf, 0x1fd8, 0x1fd9, 0x1fd2, 0x1fd3,
+	0x1fd4, 0x1fd5, 0x1fd6, 0x1fd7, 0x1fd8, 0x1fd9, 0x1fda, 0x1fdb,
+	0x1fdc, 0x1fdd, 0x1fde, 0x1fdf, 0x1fe8, 0x1fe9, 0x1fe2, 0x1fe3,
+	0x1fe4, 0x1fec, 0x1fe6, 0x1fe7, 0x1fe8, 0x1fe9, 0x1fea, 0x1feb,
+	0x1fec, 0x1fed, 0x1fee, 0x1fef, 0x1ff0, 0x1ff1, 0x1ff2, 0x1ff3,
+	0x1ff4, 0x1ff5, 0x1ff6, 0x1ff7, 0x1ff8, 0x1ff9, 0x1ffa, 0x1ffb,
+	0x1ff3, 0x1ffd, 0x1ffe, 0x1fff, 0x2000, 0x2001, 0x2002, 0x2003,
+	0x2004, 0x2005, 0x2006, 0x2007, 0x2008, 0x2009, 0x200a, 0x200b,
+	0x200c, 0x200d, 0x200e, 0x200f, 0x2010, 0x2011, 0x2012, 0x2013,
+	0x2014, 0x2015, 0x2016, 0x2017, 0x2018, 0x2019, 0x201a, 0x201b,
+	0x201c, 0x201d, 0x201e, 0x201f, 0x2020, 0x2021, 0x2022, 0x2023,
+	0x2024, 0x2025, 0x2026, 0x2027, 0x2028, 0x2029, 0x202a, 0x202b,
+	0x202c, 0x202d, 0x202e, 0x202f, 0x2030, 0x2031, 0x2032, 0x2033,
+	0x2034, 0x2035, 0x2036, 0x2037, 0x2038, 0x2039, 0x203a, 0x203b,
+	0x203c, 0x203d, 0x203e, 0x203f, 0x2040, 0x2041, 0x2042, 0x2043,
+	0x2044, 0x2045, 0x2046, 0x2047, 0x2048, 0x2049, 0x204a, 0x204b,
+	0x204c, 0x204d, 0x204e, 0x204f, 0x2050, 0x2051, 0x2052, 0x2053,
+	0x2054, 0x2055, 0x2056, 0x2057, 0x2058, 0x2059, 0x205a, 0x205b,
+	0x205c, 0x205d, 0x205e, 0x205f, 0x2060, 0x2061, 0x2062, 0x2063,
+	0x2064, 0x2065, 0x2066, 0x2067, 0x2068, 0x2069, 0x206a, 0x206b,
+	0x206c, 0x206d, 0x206e, 0x206f, 0x2070, 0x2071, 0x2072, 0x2073,
+	0x2074, 0x2075, 0x2076, 0x2077, 0x2078, 0x2079, 0x207a, 0x207b,
+	0x207c, 0x207d, 0x207e, 0x207f, 0x2080, 0x2081, 0x2082, 0x2083,
+	0x2084, 0x2085, 0x2086, 0x2087, 0x2088, 0x2089, 0x208a, 0x208b,
+	0x208c, 0x208d, 0x208e, 0x208f, 0x2090, 0x2091, 0x2092, 0x2093,
+	0x2094, 0x2095, 0x2096, 0x2097, 0x2098, 0x2099, 0x209a, 0x209b,
+	0x209c, 0x209d, 0x209e, 0x209f, 0x20a0, 0x20a1, 0x20a2, 0x20a3,
+	0x20a4, 0x20a5, 0x20a6, 0x20a7, 0x20a8, 0x20a9, 0x20aa, 0x20ab,
+	0x20ac, 0x20ad, 0x20ae, 0x20af, 0x20b0, 0x20b1, 0x20b2, 0x20b3,
+	0x20b4, 0x20b5, 0x20b6, 0x20b7, 0x20b8, 0x20b9, 0x20ba, 0x20bb,
+	0x20bc, 0x20bd, 0x20be, 0x20bf, 0x20c0, 0x20c1, 0x20c2, 0x20c3,
+	0x20c4, 0x20c5, 0x20c6, 0x20c7, 0x20c8, 0x20c9, 0x20ca, 0x20cb,
+	0x20cc, 0x20cd, 0x20ce, 0x20cf, 0x20d0, 0x20d1, 0x20d2, 0x20d3,
+	0x20d4, 0x20d5, 0x20d6, 0x20d7, 0x20d8, 0x20d9, 0x20da, 0x20db,
+	0x20dc, 0x20dd, 0x20de, 0x20df, 0x20e0, 0x20e1, 0x20e2, 0x20e3,
+	0x20e4, 0x20e5, 0x20e6, 0x20e7, 0x20e8, 0x20e9, 0x20ea, 0x20eb,
+	0x20ec, 0x20ed, 0x20ee, 0x20ef, 0x20f0, 0x20f1, 0x20f2, 0x20f3,
+	0x20f4, 0x20f5, 0x20f6, 0x20f7, 0x20f8, 0x20f9, 0x20fa, 0x20fb,
+	0x20fc, 0x20fd, 0x20fe, 0x20ff, 0x2100, 0x2101, 0x2102, 0x2103,
+	0x2104, 0x2105, 0x2106, 0x2107, 0x2108, 0x2109, 0x210a, 0x210b,
+	0x210c, 0x210d, 0x210e, 0x210f, 0x2110, 0x2111, 0x2112, 0x2113,
+	0x2114, 0x2115, 0x2116, 0x2117, 0x2118, 0x2119, 0x211a, 0x211b,
+	0x211c, 0x211d, 0x211e, 0x211f, 0x2120, 0x2121, 0x2122, 0x2123,
+	0x2124, 0x2125, 0x2126, 0x2127, 0x2128, 0x2129, 0x212a, 0x212b,
+	0x212c, 0x212d, 0x212e, 0x212f, 0x2130, 0x2131, 0x2132, 0x2133,
+	0x2134, 0x2135, 0x2136, 0x2137, 0x2138, 0x2139, 0x213a, 0x213b,
+	0x213c, 0x213d, 0x213e, 0x213f, 0x2140, 0x2141, 0x2142, 0x2143,
+	0x2144, 0x2145, 0x2146, 0x2147, 0x2148, 0x2149, 0x214a, 0x214b,
+	0x214c, 0x214d, 0x2132, 0x214f, 0x2150, 0x2151, 0x2152, 0x2153,
+	0x2154, 0x2155, 0x2156, 0x2157, 0x2158, 0x2159, 0x215a, 0x215b,
+	0x215c, 0x215d, 0x215e, 0x215f, 0x2160, 0x2161, 0x2162, 0x2163,
+	0x2164, 0x2165, 0x2166, 0x2167, 0x2168, 0x2169, 0x216a, 0x216b,
+	0x216c, 0x216d, 0x216e, 0x216f, 0x2160, 0x2161, 0x2162, 0x2163,
+	0x2164, 0x2165, 0x2166, 0x2167, 0x2168, 0x2169, 0x216a, 0x216b,
+	0x216c, 0x216d, 0x216e, 0x216f, 0x2180, 0x2181, 0x2182, 0x2183,
+	0x2183, 0xffff, 0x034b, 0x24b6, 0x24b7, 0x24b8, 0x24b9, 0x24ba,
+	0x24bb, 0x24bc, 0x24bd, 0x24be, 0x24bf, 0x24c0, 0x24c1, 0x24c2,
+	0x24c3, 0x24c4, 0x24c5, 0x24c6, 0x24c7, 0x24c8, 0x24c9, 0x24ca,
+	0x24cb, 0x24cc, 0x24cd, 0x24ce, 0x24cf, 0xffff, 0x0746, 0x2c00,
+	0x2c01, 0x2c02, 0x2c03, 0x2c04, 0x2c05, 0x2c06, 0x2c07, 0x2c08,
+	0x2c09, 0x2c0a, 0x2c0b, 0x2c0c, 0x2c0d, 0x2c0e, 0x2c0f, 0x2c10,
+	0x2c11, 0x2c12, 0x2c13, 0x2c14, 0x2c15, 0x2c16, 0x2c17, 0x2c18,
+	0x2c19, 0x2c1a, 0x2c1b, 0x2c1c, 0x2c1d, 0x2c1e, 0x2c1f, 0x2c20,
+	0x2c21, 0x2c22, 0x2c23, 0x2c24, 0x2c25, 0x2c26, 0x2c27, 0x2c28,
+	0x2c29, 0x2c2a, 0x2c2b, 0x2c2c, 0x2c2d, 0x2c2e, 0x2c5f, 0x2c60,
+	0x2c60, 0x2c62, 0x2c63, 0x2c64, 0x2c65, 0x2c66, 0x2c67, 0x2c67,
+	0x2c69, 0x2c69, 0x2c6b, 0x2c6b, 0x2c6d, 0x2c6e, 0x2c6f, 0x2c70,
+	0x2c71, 0x2c72, 0x2c73, 0x2c74, 0x2c75, 0x2c75, 0x2c77, 0x2c78,
+	0x2c79, 0x2c7a, 0x2c7b, 0x2c7c, 0x2c7d, 0x2c7e, 0x2c7f, 0x2c80,
+	0x2c80, 0x2c82, 0x2c82, 0x2c84, 0x2c84, 0x2c86, 0x2c86, 0x2c88,
+	0x2c88, 0x2c8a, 0x2c8a, 0x2c8c, 0x2c8c, 0x2c8e, 0x2c8e, 0x2c90,
+	0x2c90, 0x2c92, 0x2c92, 0x2c94, 0x2c94, 0x2c96, 0x2c96, 0x2c98,
+	0x2c98, 0x2c9a, 0x2c9a, 0x2c9c, 0x2c9c, 0x2c9e, 0x2c9e, 0x2ca0,
+	0x2ca0, 0x2ca2, 0x2ca2, 0x2ca4, 0x2ca4, 0x2ca6, 0x2ca6, 0x2ca8,
+	0x2ca8, 0x2caa, 0x2caa, 0x2cac, 0x2cac, 0x2cae, 0x2cae, 0x2cb0,
+	0x2cb0, 0x2cb2, 0x2cb2, 0x2cb4, 0x2cb4, 0x2cb6, 0x2cb6, 0x2cb8,
+	0x2cb8, 0x2cba, 0x2cba, 0x2cbc, 0x2cbc, 0x2cbe, 0x2cbe, 0x2cc0,
+	0x2cc0, 0x2cc2, 0x2cc2, 0x2cc4, 0x2cc4, 0x2cc6, 0x2cc6, 0x2cc8,
+	0x2cc8, 0x2cca, 0x2cca, 0x2ccc, 0x2ccc, 0x2cce, 0x2cce, 0x2cd0,
+	0x2cd0, 0x2cd2, 0x2cd2, 0x2cd4, 0x2cd4, 0x2cd6, 0x2cd6, 0x2cd8,
+	0x2cd8, 0x2cda, 0x2cda, 0x2cdc, 0x2cdc, 0x2cde, 0x2cde, 0x2ce0,
+	0x2ce0, 0x2ce2, 0x2ce2, 0x2ce4, 0x2ce5, 0x2ce6, 0x2ce7, 0x2ce8,
+	0x2ce9, 0x2cea, 0x2ceb, 0x2cec, 0x2ced, 0x2cee, 0x2cef, 0x2cf0,
+	0x2cf1, 0x2cf2, 0x2cf3, 0x2cf4, 0x2cf5, 0x2cf6, 0x2cf7, 0x2cf8,
+	0x2cf9, 0x2cfa, 0x2cfb, 0x2cfc, 0x2cfd, 0x2cfe, 0x2cff, 0x10a0,
+	0x10a1, 0x10a2, 0x10a3, 0x10a4, 0x10a5, 0x10a6, 0x10a7, 0x10a8,
+	0x10a9, 0x10aa, 0x10ab, 0x10ac, 0x10ad, 0x10ae, 0x10af, 0x10b0,
+	0x10b1, 0x10b2, 0x10b3, 0x10b4, 0x10b5, 0x10b6, 0x10b7, 0x10b8,
+	0x10b9, 0x10ba, 0x10bb, 0x10bc, 0x10bd, 0x10be, 0x10bf, 0x10c0,
+	0x10c1, 0x10c2, 0x10c3, 0x10c4, 0x10c5, 0xffff, 0xd21b, 0xff21,
+	0xff22, 0xff23, 0xff24, 0xff25, 0xff26, 0xff27, 0xff28, 0xff29,
+	0xff2a, 0xff2b, 0xff2c, 0xff2d, 0xff2e, 0xff2f, 0xff30, 0xff31,
+	0xff32, 0xff33, 0xff34, 0xff35, 0xff36, 0xff37, 0xff38, 0xff39,
+	0xff3a, 0xff5b, 0xff5c, 0xff5d, 0xff5e, 0xff5f, 0xff60, 0xff61,
+	0xff62, 0xff63, 0xff64, 0xff65, 0xff66, 0xff67, 0xff68, 0xff69,
+	0xff6a, 0xff6b, 0xff6c, 0xff6d, 0xff6e, 0xff6f, 0xff70, 0xff71,
+	0xff72, 0xff73, 0xff74, 0xff75, 0xff76, 0xff77, 0xff78, 0xff79,
+	0xff7a, 0xff7b, 0xff7c, 0xff7d, 0xff7e, 0xff7f, 0xff80, 0xff81,
+	0xff82, 0xff83, 0xff84, 0xff85, 0xff86, 0xff87, 0xff88, 0xff89,
+	0xff8a, 0xff8b, 0xff8c, 0xff8d, 0xff8e, 0xff8f, 0xff90, 0xff91,
+	0xff92, 0xff93, 0xff94, 0xff95, 0xff96, 0xff97, 0xff98, 0xff99,
+	0xff9a, 0xff9b, 0xff9c, 0xff9d, 0xff9e, 0xff9f, 0xffa0, 0xffa1,
+	0xffa2, 0xffa3, 0xffa4, 0xffa5, 0xffa6, 0xffa7, 0xffa8, 0xffa9,
+	0xffaa, 0xffab, 0xffac, 0xffad, 0xffae, 0xffaf, 0xffb0, 0xffb1,
+	0xffb2, 0xffb3, 0xffb4, 0xffb5, 0xffb6, 0xffb7, 0xffb8, 0xffb9,
+	0xffba, 0xffbb, 0xffbc, 0xffbd, 0xffbe, 0xffbf, 0xffc0, 0xffc1,
+	0xffc2, 0xffc3, 0xffc4, 0xffc5, 0xffc6, 0xffc7, 0xffc8, 0xffc9,
+	0xffca, 0xffcb, 0xffcc, 0xffcd, 0xffce, 0xffcf, 0xffd0, 0xffd1,
+	0xffd2, 0xffd3, 0xffd4, 0xffd5, 0xffd6, 0xffd7, 0xffd8, 0xffd9,
+	0xffda, 0xffdb, 0xffdc, 0xffdd, 0xffde, 0xffdf, 0xffe0, 0xffe1,
+	0xffe2, 0xffe3, 0xffe4, 0xffe5, 0xffe6, 0xffe7, 0xffe8, 0xffe9,
+	0xffea, 0xffeb, 0xffec, 0xffed, 0xffee, 0xffef, 0xfff0, 0xfff1,
+	0xfff2, 0xfff3, 0xfff4, 0xfff5, 0xfff6, 0xfff7, 0xfff8, 0xfff9,
+	0xfffa, 0xfffb, 0xfffc, 0xfffd, 0xfffe, 0xffff,
+};
+
+/*
+ * Allow full-width illegal characters :
+ * "MS windows 7" supports full-width-invalid-name-characters.
+ * So we should check half-width-invalid-name-characters(ASCII) only
+ * for compatibility.
+ *
+ * " * / : < > ? \ |
+ */
+static unsigned short bad_uni_chars[] = {
+	0x0022,         0x002A, 0x002F, 0x003A,
+	0x003C, 0x003E, 0x003F, 0x005C, 0x007C,
+	0
+};
+
+static int exfat_convert_char_to_ucs2(struct nls_table *nls,
+		const unsigned char *ch, int ch_len, unsigned short *ucs2,
+		int *lossy)
+{
+	int len;
+
+	*ucs2 = 0x0;
+
+	if (ch[0] < 0x80) {
+		*ucs2 = ch[0];
+		return 1;
+	}
+
+	len = nls->char2uni(ch, ch_len, ucs2);
+	if (len < 0) {
+		/* conversion failed */
+		if (lossy != NULL)
+			*lossy |= NLS_NAME_LOSSY;
+		*ucs2 = '_';
+		return 1;
+	}
+	return len;
+}
+
+static int exfat_convert_ucs2_to_char(struct nls_table *nls,
+		unsigned short ucs2, unsigned char *ch, int *lossy)
+{
+	int len;
+
+	ch[0] = 0x0;
+
+	if (ucs2 < 0x0080) {
+		ch[0] = ucs2;
+		return 1;
+	}
+
+	len = nls->uni2char(ucs2, ch, MAX_CHARSET_SIZE);
+	if (len < 0) {
+		/* conversion failed */
+		if (lossy != NULL)
+			*lossy |= NLS_NAME_LOSSY;
+		ch[0] = '_';
+		return 1;
+	}
+	return len;
+}
+
+unsigned short exfat_toupper(struct super_block *sb, unsigned short a)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	return sbi->vol_utbl[a] ? sbi->vol_utbl[a] : a;
+}
+
+static unsigned short *exfat_wstrchr(unsigned short *str, unsigned short wchar)
+{
+	while (*str) {
+		if (*(str++) == wchar)
+			return str;
+	}
+	return NULL;
+}
+
+int exfat_uniname_ncmp(struct super_block *sb, unsigned short *a,
+		unsigned short *b, unsigned int len)
+{
+	int i;
+
+	for (i = 0; i < len; i++, a++, b++)
+		if (exfat_toupper(sb, *a) != exfat_toupper(sb, *b))
+			return 1;
+	return 0;
+}
+
+static int exfat_utf16_to_utf8(struct super_block *sb,
+		struct exfat_uni_name *p_uniname, unsigned char *p_cstring,
+		int buflen)
+{
+	int len;
+	const unsigned short *uniname = p_uniname->name;
+
+	/* always len >= 0 */
+	len = utf16s_to_utf8s(uniname, MAX_NAME_LENGTH, UTF16_HOST_ENDIAN,
+		p_cstring, buflen);
+	p_cstring[len] = '\0';
+	return len;
+}
+
+static int exfat_utf8_to_utf16(struct super_block *sb,
+		const unsigned char *p_cstring, const int len,
+		struct exfat_uni_name *p_uniname, int *p_lossy)
+{
+	int i, unilen, lossy = NLS_NAME_NO_LOSSY;
+	unsigned short upname[MAX_NAME_LENGTH + 1];
+	unsigned short *uniname = p_uniname->name;
+
+	WARN_ON(!len);
+
+	unilen = utf8s_to_utf16s(p_cstring, len, UTF16_HOST_ENDIAN,
+			(wchar_t *)uniname, MAX_NAME_LENGTH + 2);
+	if (unilen < 0) {
+		exfat_msg(sb, KERN_ERR,
+			"failed to %s (err : %d) nls len : %d",
+			__func__, unilen, len);
+		return unilen;
+	}
+
+	if (unilen > MAX_NAME_LENGTH) {
+		exfat_msg(sb, KERN_ERR,
+			"failed to %s (estr:ENAMETOOLONG) nls len : %d, unilen : %d > %d",
+			__func__, len, unilen, MAX_NAME_LENGTH);
+		return -ENAMETOOLONG;
+	}
+
+	p_uniname->name_len = unilen & 0xFF;
+
+	for (i = 0; i < unilen; i++) {
+		if (*uniname < 0x0020 ||
+		    exfat_wstrchr(bad_uni_chars, *uniname))
+			lossy |= NLS_NAME_LOSSY;
+
+		upname[i] = exfat_toupper(sb, *uniname);
+		uniname++;
+	}
+
+	*uniname = '\0';
+	p_uniname->name_len = unilen;
+	p_uniname->name_hash = exfat_calc_chksum_2byte(upname, unilen << 1, 0,
+			CS_DEFAULT);
+
+	if (p_lossy)
+		*p_lossy = lossy;
+	return unilen;
+}
+
+#define PLANE_SIZE	0x00010000
+#define SURROGATE_MASK	0xfffff800
+#define SURROGATE_PAIR	0x0000d800
+#define SURROGATE_LOW	0x00000400
+#define SURROGATE_BITS	0x000003ff
+
+unsigned short exfat_high_surrogate(unicode_t u)
+{
+	return ((u - PLANE_SIZE) >> 10) + SURROGATE_PAIR;
+}
+
+unsigned short exfat_low_surrogate(unicode_t u)
+{
+	return ((u - PLANE_SIZE) & SURROGATE_BITS) | SURROGATE_PAIR |
+		SURROGATE_LOW;
+}
+
+static int __exfat_utf16_to_nls(struct super_block *sb,
+		struct exfat_uni_name *p_uniname, unsigned char *p_cstring,
+		int buflen)
+{
+	int i, j, len, out_len = 0;
+	unsigned char buf[MAX_CHARSET_SIZE];
+	const unsigned short *uniname = p_uniname->name;
+	struct nls_table *nls = EXFAT_SB(sb)->nls_io;
+
+	i = 0;
+	while (i < MAX_NAME_LENGTH && out_len < (buflen - 1)) {
+		if (*uniname == '\0')
+			break;
+		if ((*uniname & SURROGATE_MASK) != SURROGATE_PAIR) {
+			len = exfat_convert_ucs2_to_char(nls, *uniname, buf,
+				NULL);
+		} else {
+			/* Process UTF-16 surrogate pair as one character */
+			if (!(*uniname & SURROGATE_LOW) &&
+			    i+1 < MAX_NAME_LENGTH &&
+			    (*(uniname+1) & SURROGATE_MASK) == SURROGATE_PAIR &&
+			    (*(uniname+1) & SURROGATE_LOW)) {
+				uniname++;
+				i++;
+			}
+
+			/*
+			 * UTF-16 surrogate pair encodes code points above
+			 * U+FFFF. Code points above U+FFFF are not supported
+			 * by kernel NLS framework therefore use replacement
+			 * character
+			 */
+			len = 1;
+			buf[0] = '_';
+		}
+
+		if (out_len + len >= buflen)
+			len = buflen - 1 - out_len;
+		out_len += len;
+
+		if (len > 1) {
+			for (j = 0; j < len; j++)
+				*p_cstring++ = buf[j];
+		} else { /* len == 1 */
+			*p_cstring++ = *buf;
+		}
+
+		uniname++;
+		i++;
+	}
+
+	*p_cstring = '\0';
+	return out_len;
+}
+
+static int exfat_nls_to_ucs2(struct super_block *sb,
+		const unsigned char *p_cstring, const int len,
+		struct exfat_uni_name *p_uniname, int *p_lossy)
+{
+	int i = 0, unilen = 0, lossy = NLS_NAME_NO_LOSSY;
+	unsigned short upname[MAX_NAME_LENGTH + 1];
+	unsigned short *uniname = p_uniname->name;
+	struct nls_table *nls = EXFAT_SB(sb)->nls_io;
+
+	WARN_ON(!len);
+
+	while (unilen < MAX_NAME_LENGTH && i < len) {
+		i += exfat_convert_char_to_ucs2(nls, p_cstring + i, len - i,
+				uniname, &lossy);
+
+		if (*uniname < 0x0020 ||
+		    exfat_wstrchr(bad_uni_chars, *uniname))
+			lossy |= NLS_NAME_LOSSY;
+
+		upname[unilen] = exfat_toupper(sb, *uniname);
+		uniname++;
+		unilen++;
+	}
+
+	if (p_cstring[i] != '\0')
+		lossy |= NLS_NAME_OVERLEN;
+
+	*uniname = '\0';
+	p_uniname->name_len = unilen;
+	p_uniname->name_hash = exfat_calc_chksum_2byte(upname, unilen << 1, 0,
+			CS_DEFAULT);
+
+	if (p_lossy)
+		*p_lossy = lossy;
+	return unilen;
+}
+
+int exfat_utf16_to_nls(struct super_block *sb, struct exfat_uni_name *uniname,
+		unsigned char *p_cstring, int buflen)
+{
+	if (EXFAT_SB(sb)->options.utf8)
+		return exfat_utf16_to_utf8(sb, uniname, p_cstring,
+				buflen);
+	return __exfat_utf16_to_nls(sb, uniname, p_cstring, buflen);
+}
+
+int exfat_nls_to_utf16(struct super_block *sb, const unsigned char *p_cstring,
+		const int len, struct exfat_uni_name *uniname, int *p_lossy)
+{
+	if (EXFAT_SB(sb)->options.utf8)
+		return exfat_utf8_to_utf16(sb, p_cstring, len,
+				uniname, p_lossy);
+	return exfat_nls_to_ucs2(sb, p_cstring, len, uniname, p_lossy);
+}
+
+static int exfat_load_upcase_table(struct super_block *sb,
+		sector_t sector, unsigned long long num_sectors,
+		unsigned int utbl_checksum)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	unsigned int sect_size = sb->s_blocksize;
+	unsigned int i, index = 0, checksum = 0;
+	int ret;
+	unsigned char skip = false;
+	unsigned short *upcase_table;
+
+	upcase_table = kcalloc(UTBL_COUNT, sizeof(unsigned short), GFP_KERNEL);
+	if (!upcase_table)
+		return -ENOMEM;
+
+	sbi->vol_utbl = upcase_table;
+	num_sectors += sector;
+
+	while (sector < num_sectors) {
+		struct buffer_head *bh;
+
+		bh = sb_bread(sb, sector);
+		if (!bh) {
+			exfat_msg(sb, KERN_ERR,
+				"failed to read sector(0x%llx)\n",
+				(unsigned long long)sector);
+			ret = -EIO;
+			goto free_table;
+		}
+		sector++;
+		for (i = 0; i < sect_size && index <= 0xFFFF; i += 2) {
+			unsigned short uni = get_unaligned_le16(bh->b_data + i);
+
+			checksum = ((checksum & 1) ? 0x80000000 : 0) +
+				(checksum >> 1) +
+				*(((unsigned char *)bh->b_data) + i);
+			checksum = ((checksum & 1) ? 0x80000000 : 0) +
+				(checksum >> 1) +
+				*(((unsigned char *)bh->b_data) + (i + 1));
+
+			if (skip) {
+				index += uni;
+				skip = false;
+			} else if (uni == index) {
+				index++;
+			} else if (uni == 0xFFFF) {
+				skip = true;
+			} else { /* uni != index , uni != 0xFFFF */
+				upcase_table[index] = uni;
+				index++;
+			}
+		}
+		brelse(bh);
+	}
+
+	if (index >= 0xFFFF && utbl_checksum == checksum)
+		return 0;
+
+	exfat_msg(sb, KERN_ERR,
+			"failed to load upcase table (idx : 0x%08x, chksum : 0x%08x, utbl_chksum : 0x%08x)\n",
+			index, checksum, utbl_checksum);
+	ret = -EINVAL;
+free_table:
+	exfat_free_upcase_table(sbi);
+	return ret;
+}
+
+static int exfat_load_default_upcase_table(struct super_block *sb)
+{
+	int i, ret = -EIO;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	unsigned char skip = false;
+	unsigned short uni = 0, *upcase_table;
+	unsigned int index = 0;
+
+	upcase_table = kcalloc(UTBL_COUNT, sizeof(unsigned short), GFP_KERNEL);
+	if (!upcase_table)
+		return -ENOMEM;
+
+	sbi->vol_utbl = upcase_table;
+
+	for (i = 0; index <= 0xFFFF && i < EXFAT_NUM_UPCASE; i++) {
+		uni = uni_def_upcase[i];
+		if (skip) {
+			index += uni;
+			skip = false;
+		} else if (uni == index) {
+			index++;
+		} else if (uni == 0xFFFF) {
+			skip = true;
+		} else {
+			upcase_table[index] = uni;
+			index++;
+		}
+	}
+
+	if (index >= 0xFFFF)
+		return 0;
+
+	/* FATAL error: default upcase table has error */
+	exfat_free_upcase_table(sbi);
+	return ret;
+}
+
+int exfat_create_upcase_table(struct super_block *sb)
+{
+	int i, ret;
+	unsigned int tbl_clu, type;
+	sector_t sector;
+	unsigned long long tbl_size, num_sectors;
+	unsigned char blksize_bits = sb->s_blocksize_bits;
+	struct exfat_chain clu;
+	struct exfat_dentry *ep;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct buffer_head *bh;
+
+	clu.dir = sbi->root_dir;
+	clu.flags = ALLOC_FAT_CHAIN;
+
+	while (clu.dir != EXFAT_EOF_CLUSTER) {
+		for (i = 0; i < sbi->dentries_per_clu; i++) {
+			ep = exfat_get_dentry(sb, &clu, i, &bh, NULL);
+			if (!ep)
+				return -EIO;
+
+			type = exfat_get_entry_type(ep);
+			if (type == TYPE_UNUSED) {
+				brelse(bh);
+				break;
+			}
+
+			if (type != TYPE_UPCASE) {
+				brelse(bh);
+				continue;
+			}
+
+			tbl_clu  = le32_to_cpu(ep->dentry.upcase.start_clu);
+			tbl_size = le64_to_cpu(ep->dentry.upcase.size);
+
+			sector = exfat_cluster_to_sector(sbi, tbl_clu);
+			num_sectors = ((tbl_size - 1) >> blksize_bits) + 1;
+			ret = exfat_load_upcase_table(sb, sector, num_sectors,
+				le32_to_cpu(ep->dentry.upcase.checksum));
+
+			brelse(bh);
+			if (ret && ret != -EIO)
+				goto load_default;
+
+			/* load successfully */
+			return ret;
+		}
+
+		if (exfat_get_next_cluster(sb, &(clu.dir)))
+			return -EIO;
+	}
+
+load_default:
+	/* load default upcase table */
+	return exfat_load_default_upcase_table(sb);
+}
+
+void exfat_free_upcase_table(struct exfat_sb_info *sbi)
+{
+	kfree(sbi->vol_utbl);
+}
diff --git a/fs/exfat/super.c b/fs/exfat/super.c
new file mode 100644
index 000000000000..a846ff555656
--- /dev/null
+++ b/fs/exfat/super.c
@@ -0,0 +1,732 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  Copyright (C) 2012-2013 Samsung Electronics Co., Ltd.
+ */
+
+#include <linux/fs_context.h>
+#include <linux/fs_parser.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/time.h>
+#include <linux/mount.h>
+#include <linux/cred.h>
+#include <linux/statfs.h>
+#include <linux/seq_file.h>
+#include <linux/blkdev.h>
+#include <linux/fs_struct.h>
+#include <linux/iversion.h>
+#include <linux/nls.h>
+#include <linux/buffer_head.h>
+
+#include "exfat_raw.h"
+#include "exfat_fs.h"
+
+static char exfat_default_iocharset[] = CONFIG_EXFAT_DEFAULT_IOCHARSET;
+static struct kmem_cache *exfat_inode_cachep;
+
+static void exfat_free_iocharset(struct exfat_sb_info *sbi)
+{
+	if (sbi->options.iocharset != exfat_default_iocharset)
+		kfree(sbi->options.iocharset);
+}
+
+static void exfat_delayed_free(struct rcu_head *p)
+{
+	struct exfat_sb_info *sbi = container_of(p, struct exfat_sb_info, rcu);
+
+	unload_nls(sbi->nls_io);
+	exfat_free_iocharset(sbi);
+	exfat_free_upcase_table(sbi);
+	kfree(sbi);
+}
+
+static void exfat_put_super(struct super_block *sb)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	mutex_lock(&sbi->s_lock);
+	if (test_and_clear_bit(EXFAT_SB_DIRTY, &sbi->s_state))
+		sync_blockdev(sb->s_bdev);
+	exfat_set_vol_flags(sb, VOL_CLEAN);
+	exfat_free_bitmap(sbi);
+	brelse(sbi->pbr_bh);
+	mutex_unlock(&sbi->s_lock);
+
+	call_rcu(&sbi->rcu, exfat_delayed_free);
+}
+
+static int exfat_sync_fs(struct super_block *sb, int wait)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	int err = 0;
+
+	/* If there are some dirty buffers in the bdev inode */
+	mutex_lock(&sbi->s_lock);
+	if (test_and_clear_bit(EXFAT_SB_DIRTY, &sbi->s_state)) {
+		sync_blockdev(sb->s_bdev);
+		if (exfat_set_vol_flags(sb, VOL_CLEAN))
+			err = -EIO;
+	}
+	mutex_unlock(&sbi->s_lock);
+	return err;
+}
+
+static int exfat_statfs(struct dentry *dentry, struct kstatfs *buf)
+{
+	struct super_block *sb = dentry->d_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	unsigned long long id = huge_encode_dev(sb->s_bdev->bd_dev);
+
+	if (sbi->used_clusters == EXFAT_CLUSTERS_UNTRACKED) {
+		mutex_lock(&sbi->s_lock);
+		if (exfat_count_used_clusters(sb, &sbi->used_clusters)) {
+			mutex_unlock(&sbi->s_lock);
+			return -EIO;
+		}
+		mutex_unlock(&sbi->s_lock);
+	}
+
+	buf->f_type = sb->s_magic;
+	buf->f_bsize = sbi->cluster_size;
+	buf->f_blocks = sbi->num_clusters - 2; /* clu 0 & 1 */
+	buf->f_bfree = buf->f_blocks - sbi->used_clusters;
+	buf->f_bavail = buf->f_bfree;
+	buf->f_fsid.val[0] = (unsigned int)id;
+	buf->f_fsid.val[1] = (unsigned int)(id >> 32);
+	/* Unicode utf16 255 characters */
+	buf->f_namelen = EXFAT_MAX_FILE_LEN * NLS_MAX_CHARSET_SIZE;
+	return 0;
+}
+
+int exfat_set_vol_flags(struct super_block *sb, unsigned short new_flag)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct pbr64 *bpb = (struct pbr64 *)sbi->pbr_bh->b_data;
+	bool sync = 0;
+
+	/* flags are not changed */
+	if (sbi->vol_flag == new_flag)
+		return 0;
+
+	sbi->vol_flag = new_flag;
+
+	/* skip updating volume dirty flag,
+	 * if this volume has been mounted with read-only
+	 */
+	if (sb_rdonly(sb))
+		return 0;
+
+	bpb->bsx.vol_flags = cpu_to_le16(new_flag);
+
+	if (new_flag == VOL_DIRTY && !buffer_dirty(sbi->pbr_bh))
+		sync = true;
+	else
+		sync = false;
+
+	set_buffer_uptodate(sbi->pbr_bh);
+	mark_buffer_dirty(sbi->pbr_bh);
+
+	if (sync)
+		sync_dirty_buffer(sbi->pbr_bh);
+	return 0;
+}
+
+static int exfat_show_options(struct seq_file *m, struct dentry *root)
+{
+	struct super_block *sb = root->d_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_mount_options *opts = &sbi->options;
+
+	/* Show partition info */
+	if (!uid_eq(opts->fs_uid, GLOBAL_ROOT_UID))
+		seq_printf(m, ",uid=%u",
+				from_kuid_munged(&init_user_ns, opts->fs_uid));
+	if (!gid_eq(opts->fs_gid, GLOBAL_ROOT_GID))
+		seq_printf(m, ",gid=%u",
+				from_kgid_munged(&init_user_ns, opts->fs_gid));
+	seq_printf(m, ",fmask=%04o,dmask=%04o", opts->fs_fmask, opts->fs_dmask);
+	if (opts->allow_utime)
+		seq_printf(m, ",allow_utime=%04o", opts->allow_utime);
+	if (opts->utf8)
+		seq_puts(m, ",iocharset=utf8");
+	else if (sbi->nls_io)
+		seq_printf(m, ",iocharset=%s", sbi->nls_io->charset);
+	if (opts->errors == EXFAT_ERRORS_CONT)
+		seq_puts(m, ",errors=continue");
+	else if (opts->errors == EXFAT_ERRORS_PANIC)
+		seq_puts(m, ",errors=panic");
+	else
+		seq_puts(m, ",errors=remount-ro");
+	if (opts->discard)
+		seq_puts(m, ",discard");
+	if (opts->time_offset)
+		seq_printf(m, ",time_offset=%d", opts->time_offset);
+	return 0;
+}
+
+static struct inode *exfat_alloc_inode(struct super_block *sb)
+{
+	struct exfat_inode_info *ei;
+
+	ei = kmem_cache_alloc(exfat_inode_cachep, GFP_NOFS);
+	if (!ei)
+		return NULL;
+
+	init_rwsem(&ei->truncate_lock);
+	return &ei->vfs_inode;
+}
+
+static void exfat_free_inode(struct inode *inode)
+{
+	kmem_cache_free(exfat_inode_cachep, EXFAT_I(inode));
+}
+
+static const struct super_operations exfat_sops = {
+	.alloc_inode	= exfat_alloc_inode,
+	.free_inode	= exfat_free_inode,
+	.write_inode	= exfat_write_inode,
+	.evict_inode	= exfat_evict_inode,
+	.put_super	= exfat_put_super,
+	.sync_fs	= exfat_sync_fs,
+	.statfs		= exfat_statfs,
+	.show_options	= exfat_show_options,
+};
+
+enum {
+	Opt_uid,
+	Opt_gid,
+	Opt_umask,
+	Opt_dmask,
+	Opt_fmask,
+	Opt_allow_utime,
+	Opt_charset,
+	Opt_errors,
+	Opt_discard,
+	Opt_time_offset,
+
+	/* Deprecated options */
+	Opt_utf8,
+	Opt_debug,
+	Opt_namecase,
+	Opt_codepage,
+};
+
+static const struct constant_table exfat_param_enums[] = {
+	{ "continue",		EXFAT_ERRORS_CONT },
+	{ "panic",		EXFAT_ERRORS_PANIC },
+	{ "remount-ro",		EXFAT_ERRORS_RO },
+	{}
+};
+
+static const struct fs_parameter_spec exfat_parameters[] = {
+	fsparam_u32("uid",			Opt_uid),
+	fsparam_u32("gid",			Opt_gid),
+	fsparam_u32oct("umask",			Opt_umask),
+	fsparam_u32oct("dmask",			Opt_dmask),
+	fsparam_u32oct("fmask",			Opt_fmask),
+	fsparam_u32oct("allow_utime",		Opt_allow_utime),
+	fsparam_string("iocharset",		Opt_charset),
+	fsparam_enum("errors",			Opt_errors, exfat_param_enums),
+	fsparam_flag("discard",			Opt_discard),
+	fsparam_s32("time_offset",		Opt_time_offset),
+	__fsparam(NULL, "utf8",			Opt_utf8, fs_param_deprecated,
+		  NULL),
+	__fsparam(NULL, "debug",		Opt_debug, fs_param_deprecated,
+		  NULL),
+	__fsparam(fs_param_is_u32, "namecase",	Opt_namecase,
+		  fs_param_deprecated, NULL),
+	__fsparam(fs_param_is_u32, "codepage",	Opt_codepage,
+		  fs_param_deprecated, NULL),
+	{}
+};
+
+static int exfat_parse_param(struct fs_context *fc, struct fs_parameter *param)
+{
+	struct exfat_sb_info *sbi = fc->s_fs_info;
+	struct exfat_mount_options *opts = &sbi->options;
+	struct fs_parse_result result;
+	int opt;
+
+	opt = fs_parse(fc, exfat_parameters, param, &result);
+	if (opt < 0)
+		return opt;
+
+	switch (opt) {
+	case Opt_uid:
+		opts->fs_uid = make_kuid(current_user_ns(), result.uint_32);
+		break;
+	case Opt_gid:
+		opts->fs_gid = make_kgid(current_user_ns(), result.uint_32);
+		break;
+	case Opt_umask:
+		opts->fs_fmask = result.uint_32;
+		opts->fs_dmask = result.uint_32;
+		break;
+	case Opt_dmask:
+		opts->fs_dmask = result.uint_32;
+		break;
+	case Opt_fmask:
+		opts->fs_fmask = result.uint_32;
+		break;
+	case Opt_allow_utime:
+		opts->allow_utime = result.uint_32 & 0022;
+		break;
+	case Opt_charset:
+		exfat_free_iocharset(sbi);
+		opts->iocharset = kstrdup(param->string, GFP_KERNEL);
+		if (!opts->iocharset)
+			return -ENOMEM;
+		break;
+	case Opt_errors:
+		opts->errors = result.uint_32;
+		break;
+	case Opt_discard:
+		opts->discard = 1;
+		break;
+	case Opt_time_offset:
+		/*
+		 * Make the limit 24 just in case someone invents something
+		 * unusual.
+		 */
+		if (result.int_32 < -24 * 60 || result.int_32 > 24 * 60)
+			return -EINVAL;
+		opts->time_offset = result.int_32;
+		break;
+	case Opt_utf8:
+	case Opt_debug:
+	case Opt_namecase:
+	case Opt_codepage:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void exfat_hash_init(struct super_block *sb)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	int i;
+
+	spin_lock_init(&sbi->inode_hash_lock);
+	for (i = 0; i < EXFAT_HASH_SIZE; i++)
+		INIT_HLIST_HEAD(&sbi->inode_hashtable[i]);
+}
+
+static int exfat_read_root(struct inode *inode)
+{
+	struct super_block *sb = inode->i_sb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct exfat_inode_info *ei = EXFAT_I(inode);
+	struct exfat_chain cdir;
+	int num_subdirs, num_clu = 0;
+
+	exfat_chain_set(&ei->dir, sbi->root_dir, 0, ALLOC_FAT_CHAIN);
+	ei->entry = -1;
+	ei->start_clu = sbi->root_dir;
+	ei->flags = ALLOC_FAT_CHAIN;
+	ei->type = TYPE_DIR;
+	ei->version = 0;
+	ei->rwoffset = 0;
+	ei->hint_bmap.off = EXFAT_EOF_CLUSTER;
+	ei->hint_stat.eidx = 0;
+	ei->hint_stat.clu = sbi->root_dir;
+	ei->hint_femp.eidx = EXFAT_HINT_NONE;
+
+	exfat_chain_set(&cdir, sbi->root_dir, 0, ALLOC_FAT_CHAIN);
+	if (exfat_count_num_clusters(sb, &cdir, &num_clu))
+		return -EIO;
+	i_size_write(inode, num_clu << sbi->cluster_size_bits);
+
+	num_subdirs = exfat_count_dir_entries(sb, &cdir);
+	if (num_subdirs < 0)
+		return -EIO;
+	set_nlink(inode, num_subdirs + EXFAT_MIN_SUBDIR);
+
+	inode->i_uid = sbi->options.fs_uid;
+	inode->i_gid = sbi->options.fs_gid;
+	inode_inc_iversion(inode);
+	inode->i_generation = 0;
+	inode->i_mode = exfat_make_mode(sbi, ATTR_SUBDIR, 0777);
+	inode->i_op = &exfat_dir_inode_operations;
+	inode->i_fop = &exfat_dir_operations;
+
+	inode->i_blocks = ((i_size_read(inode) + (sbi->cluster_size - 1))
+			& ~(sbi->cluster_size - 1)) >> inode->i_blkbits;
+	EXFAT_I(inode)->i_pos = ((loff_t)sbi->root_dir << 32) | 0xffffffff;
+	EXFAT_I(inode)->i_size_aligned = i_size_read(inode);
+	EXFAT_I(inode)->i_size_ondisk = i_size_read(inode);
+
+	exfat_save_attr(inode, ATTR_SUBDIR);
+	inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime =
+		current_time(inode);
+	exfat_truncate_atime(&inode->i_atime);
+	exfat_cache_init_inode(inode);
+	return 0;
+}
+
+static struct pbr *exfat_read_pbr_with_logical_sector(struct super_block *sb)
+{
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+	struct pbr *p_pbr = (struct pbr *) (sbi->pbr_bh)->b_data;
+	unsigned short logical_sect = 0;
+
+	logical_sect = 1 << p_pbr->bsx.f64.sect_size_bits;
+
+	if (!is_power_of_2(logical_sect) ||
+	    logical_sect < 512 || logical_sect > 4096) {
+		exfat_msg(sb, KERN_ERR, "bogus logical sector size %u",
+				logical_sect);
+		return NULL;
+	}
+
+	if (logical_sect < sb->s_blocksize) {
+		exfat_msg(sb, KERN_ERR,
+			"logical sector size too small for device (logical sector size = %u)",
+			logical_sect);
+		return NULL;
+	}
+
+	if (logical_sect > sb->s_blocksize) {
+		brelse(sbi->pbr_bh);
+		sbi->pbr_bh = NULL;
+
+		if (!sb_set_blocksize(sb, logical_sect)) {
+			exfat_msg(sb, KERN_ERR,
+				"unable to set blocksize %u", logical_sect);
+			return NULL;
+		}
+		sbi->pbr_bh = sb_bread(sb, 0);
+		if (!sbi->pbr_bh) {
+			exfat_msg(sb, KERN_ERR,
+				"unable to read boot sector (logical sector size = %lu)",
+				sb->s_blocksize);
+			return NULL;
+		}
+
+		p_pbr = (struct pbr *)sbi->pbr_bh->b_data;
+	}
+	return p_pbr;
+}
+
+/* mount the file system volume */
+static int __exfat_fill_super(struct super_block *sb)
+{
+	int ret;
+	struct pbr *p_pbr;
+	struct pbr64 *p_bpb;
+	struct exfat_sb_info *sbi = EXFAT_SB(sb);
+
+	/* set block size to read super block */
+	sb_min_blocksize(sb, 512);
+
+	/* read boot sector */
+	sbi->pbr_bh = sb_bread(sb, 0);
+	if (!sbi->pbr_bh) {
+		exfat_msg(sb, KERN_ERR, "unable to read boot sector");
+		return -EIO;
+	}
+
+	/* PRB is read */
+	p_pbr = (struct pbr *)sbi->pbr_bh->b_data;
+
+	/* check the validity of PBR */
+	if (le16_to_cpu((p_pbr->signature)) != PBR_SIGNATURE) {
+		exfat_msg(sb, KERN_ERR, "invalid boot record signature");
+		ret = -EINVAL;
+		goto free_bh;
+	}
+
+
+	/* check logical sector size */
+	p_pbr = exfat_read_pbr_with_logical_sector(sb);
+	if (!p_pbr) {
+		ret = -EIO;
+		goto free_bh;
+	}
+
+	/*
+	 * res_zero field must be filled with zero to prevent mounting
+	 * from FAT volume.
+	 */
+	if (memchr_inv(p_pbr->bpb.f64.res_zero, 0,
+			sizeof(p_pbr->bpb.f64.res_zero))) {
+		ret = -EINVAL;
+		goto free_bh;
+	}
+
+	p_bpb = (struct pbr64 *)p_pbr;
+	if (!p_bpb->bsx.num_fats) {
+		exfat_msg(sb, KERN_ERR, "bogus number of FAT structure");
+		ret = -EINVAL;
+		goto free_bh;
+	}
+
+	sbi->sect_per_clus = 1 << p_bpb->bsx.sect_per_clus_bits;
+	sbi->sect_per_clus_bits = p_bpb->bsx.sect_per_clus_bits;
+	sbi->cluster_size_bits = sbi->sect_per_clus_bits + sb->s_blocksize_bits;
+	sbi->cluster_size = 1 << sbi->cluster_size_bits;
+	sbi->num_FAT_sectors = le32_to_cpu(p_bpb->bsx.fat_length);
+	sbi->FAT1_start_sector = le32_to_cpu(p_bpb->bsx.fat_offset);
+	sbi->FAT2_start_sector = p_bpb->bsx.num_fats == 1 ?
+		sbi->FAT1_start_sector :
+			sbi->FAT1_start_sector + sbi->num_FAT_sectors;
+	sbi->data_start_sector = le32_to_cpu(p_bpb->bsx.clu_offset);
+	sbi->num_sectors = le64_to_cpu(p_bpb->bsx.vol_length);
+	/* because the cluster index starts with 2 */
+	sbi->num_clusters = le32_to_cpu(p_bpb->bsx.clu_count) +
+		EXFAT_RESERVED_CLUSTERS;
+
+	sbi->root_dir = le32_to_cpu(p_bpb->bsx.root_cluster);
+	sbi->dentries_per_clu = 1 <<
+		(sbi->cluster_size_bits - DENTRY_SIZE_BITS);
+
+	sbi->vol_flag = le16_to_cpu(p_bpb->bsx.vol_flags);
+	sbi->clu_srch_ptr = EXFAT_FIRST_CLUSTER;
+	sbi->used_clusters = EXFAT_CLUSTERS_UNTRACKED;
+
+	if (le16_to_cpu(p_bpb->bsx.vol_flags) & VOL_DIRTY) {
+		sbi->vol_flag |= VOL_DIRTY;
+		exfat_msg(sb, KERN_WARNING,
+			"Volume was not properly unmounted. Some data may be corrupt. Please run fsck.");
+	}
+
+	/* exFAT file size is limited by a disk volume size */
+	sb->s_maxbytes = (u64)(sbi->num_clusters - EXFAT_RESERVED_CLUSTERS) <<
+		sbi->cluster_size_bits;
+
+	ret = exfat_create_upcase_table(sb);
+	if (ret) {
+		exfat_msg(sb, KERN_ERR, "failed to load upcase table");
+		goto free_bh;
+	}
+
+	ret = exfat_load_bitmap(sb);
+	if (ret) {
+		exfat_msg(sb, KERN_ERR, "failed to load alloc-bitmap");
+		goto free_upcase_table;
+	}
+
+	ret = exfat_count_used_clusters(sb, &sbi->used_clusters);
+	if (ret) {
+		exfat_msg(sb, KERN_ERR, "failed to scan clusters");
+		goto free_alloc_bitmap;
+	}
+
+	return 0;
+
+free_alloc_bitmap:
+	exfat_free_bitmap(sbi);
+free_upcase_table:
+	exfat_free_upcase_table(sbi);
+free_bh:
+	brelse(sbi->pbr_bh);
+	return ret;
+}
+
+static int exfat_fill_super(struct super_block *sb, struct fs_context *fc)
+{
+	struct exfat_sb_info *sbi = sb->s_fs_info;
+	struct exfat_mount_options *opts = &sbi->options;
+	struct inode *root_inode;
+	int err;
+
+	if (opts->allow_utime == (unsigned short)-1)
+		opts->allow_utime = ~opts->fs_dmask & 0022;
+
+	if (opts->discard) {
+		struct request_queue *q = bdev_get_queue(sb->s_bdev);
+
+		if (!blk_queue_discard(q)) {
+			exfat_msg(sb, KERN_WARNING,
+				"mounting with \"discard\" option, but the device does not support discard");
+			opts->discard = 0;
+		}
+	}
+
+	sb->s_flags |= SB_NODIRATIME;
+	sb->s_magic = EXFAT_SUPER_MAGIC;
+	sb->s_op = &exfat_sops;
+
+	sb->s_time_gran = 10 * NSEC_PER_MSEC;
+	sb->s_time_min = EXFAT_MIN_TIMESTAMP_SECS;
+	sb->s_time_max = EXFAT_MAX_TIMESTAMP_SECS;
+
+	err = __exfat_fill_super(sb);
+	if (err) {
+		exfat_msg(sb, KERN_ERR, "failed to recognize exfat type");
+		goto check_nls_io;
+	}
+
+	/* set up enough so that it can read an inode */
+	exfat_hash_init(sb);
+
+	if (!strcmp(sbi->options.iocharset, "utf8"))
+		opts->utf8 = 1;
+	else {
+		sbi->nls_io = load_nls(sbi->options.iocharset);
+		if (!sbi->nls_io) {
+			exfat_msg(sb, KERN_ERR, "IO charset %s not found",
+					sbi->options.iocharset);
+			err = -EINVAL;
+			goto free_table;
+		}
+	}
+
+	if (sbi->options.utf8)
+		sb->s_d_op = &exfat_utf8_dentry_ops;
+	else
+		sb->s_d_op = &exfat_dentry_ops;
+
+	root_inode = new_inode(sb);
+	if (!root_inode) {
+		exfat_msg(sb, KERN_ERR, "failed to allocate root inode.");
+		err = -ENOMEM;
+		goto free_table;
+	}
+
+	root_inode->i_ino = EXFAT_ROOT_INO;
+	inode_set_iversion(root_inode, 1);
+	err = exfat_read_root(root_inode);
+	if (err) {
+		exfat_msg(sb, KERN_ERR, "failed to initialize root inode.");
+		goto put_inode;
+	}
+
+	exfat_hash_inode(root_inode, EXFAT_I(root_inode)->i_pos);
+	insert_inode_hash(root_inode);
+
+	sb->s_root = d_make_root(root_inode);
+	if (!sb->s_root) {
+		exfat_msg(sb, KERN_ERR, "failed to get the root dentry");
+		err = -ENOMEM;
+		goto put_inode;
+	}
+
+	return 0;
+
+put_inode:
+	iput(root_inode);
+	sb->s_root = NULL;
+
+free_table:
+	exfat_free_upcase_table(sbi);
+	exfat_free_bitmap(sbi);
+	brelse(sbi->pbr_bh);
+
+check_nls_io:
+	unload_nls(sbi->nls_io);
+	exfat_free_iocharset(sbi);
+	sb->s_fs_info = NULL;
+	kfree(sbi);
+	return err;
+}
+
+static int exfat_get_tree(struct fs_context *fc)
+{
+	return get_tree_bdev(fc, exfat_fill_super);
+}
+
+static void exfat_free(struct fs_context *fc)
+{
+	kfree(fc->s_fs_info);
+}
+
+static const struct fs_context_operations exfat_context_ops = {
+	.parse_param	= exfat_parse_param,
+	.get_tree	= exfat_get_tree,
+	.free		= exfat_free,
+};
+
+static int exfat_init_fs_context(struct fs_context *fc)
+{
+	struct exfat_sb_info *sbi;
+
+	sbi = kzalloc(sizeof(struct exfat_sb_info), GFP_KERNEL);
+	if (!sbi)
+		return -ENOMEM;
+
+	mutex_init(&sbi->s_lock);
+	ratelimit_state_init(&sbi->ratelimit, DEFAULT_RATELIMIT_INTERVAL,
+			DEFAULT_RATELIMIT_BURST);
+
+	sbi->options.fs_uid = current_uid();
+	sbi->options.fs_gid = current_gid();
+	sbi->options.fs_fmask = current->fs->umask;
+	sbi->options.fs_dmask = current->fs->umask;
+	sbi->options.allow_utime = -1;
+	sbi->options.iocharset = exfat_default_iocharset;
+	sbi->options.errors = EXFAT_ERRORS_RO;
+
+	fc->s_fs_info = sbi;
+	fc->ops = &exfat_context_ops;
+	return 0;
+}
+
+static struct file_system_type exfat_fs_type = {
+	.owner			= THIS_MODULE,
+	.name			= "exfat",
+	.init_fs_context	= exfat_init_fs_context,
+	.parameters		= exfat_parameters,
+	.kill_sb		= kill_block_super,
+	.fs_flags		= FS_REQUIRES_DEV,
+};
+
+static void exfat_inode_init_once(void *foo)
+{
+	struct exfat_inode_info *ei = (struct exfat_inode_info *)foo;
+
+	INIT_HLIST_NODE(&ei->i_hash_fat);
+	inode_init_once(&ei->vfs_inode);
+}
+
+static int __init init_exfat_fs(void)
+{
+	int err;
+
+	err = exfat_cache_init();
+	if (err)
+		return err;
+
+	exfat_inode_cachep = kmem_cache_create("exfat_inode_cache",
+			sizeof(struct exfat_inode_info),
+			0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
+			exfat_inode_init_once);
+	if (!exfat_inode_cachep) {
+		err = -ENOMEM;
+		goto shutdown_cache;
+	}
+
+	err = register_filesystem(&exfat_fs_type);
+	if (err)
+		goto destroy_cache;
+
+	return 0;
+
+destroy_cache:
+	kmem_cache_destroy(exfat_inode_cachep);
+shutdown_cache:
+	exfat_cache_shutdown();
+	return err;
+}
+
+static void __exit exit_exfat_fs(void)
+{
+	/*
+	 * Make sure all delayed rcu free inodes are flushed before we
+	 * destroy cache.
+	 */
+	rcu_barrier();
+	kmem_cache_destroy(exfat_inode_cachep);
+	unregister_filesystem(&exfat_fs_type);
+	exfat_cache_shutdown();
+}
+
+module_init(init_exfat_fs);
+module_exit(exit_exfat_fs);
+
+MODULE_ALIAS_FS("exfat");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("exFAT filesystem support");
+MODULE_AUTHOR("Samsung Electronics Co., Ltd.");
diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c
index e0cc55164505..fa9c951d3471 100644
--- a/fs/ext2/balloc.c
+++ b/fs/ext2/balloc.c
@@ -269,7 +269,7 @@ goal_in_my_reservation(struct ext2_reserve_window *rsv, ext2_grpblk_t grp_goal,
 	ext2_fsblk_t group_first_block, group_last_block;
 
 	group_first_block = ext2_group_first_block_no(sb, group);
-	group_last_block = group_first_block + EXT2_BLOCKS_PER_GROUP(sb) - 1;
+	group_last_block = ext2_group_last_block_no(sb, group);
 
 	if ((rsv->_rsv_start > group_last_block) ||
 	    (rsv->_rsv_end < group_first_block))
@@ -666,37 +666,24 @@ ext2_try_to_allocate(struct super_block *sb, int group,
 			unsigned long *count,
 			struct ext2_reserve_window *my_rsv)
 {
-	ext2_fsblk_t group_first_block;
+	ext2_fsblk_t group_first_block = ext2_group_first_block_no(sb, group);
+	ext2_fsblk_t group_last_block = ext2_group_last_block_no(sb, group);
        	ext2_grpblk_t start, end;
 	unsigned long num = 0;
 
+	start = 0;
+	end = group_last_block - group_first_block + 1;
 	/* we do allocation within the reservation window if we have a window */
 	if (my_rsv) {
-		group_first_block = ext2_group_first_block_no(sb, group);
 		if (my_rsv->_rsv_start >= group_first_block)
 			start = my_rsv->_rsv_start - group_first_block;
-		else
-			/* reservation window cross group boundary */
-			start = 0;
-		end = my_rsv->_rsv_end - group_first_block + 1;
-		if (end > EXT2_BLOCKS_PER_GROUP(sb))
-			/* reservation window crosses group boundary */
-			end = EXT2_BLOCKS_PER_GROUP(sb);
-		if ((start <= grp_goal) && (grp_goal < end))
-			start = grp_goal;
-		else
+		if (my_rsv->_rsv_end < group_last_block)
+			end = my_rsv->_rsv_end - group_first_block + 1;
+		if (grp_goal < start || grp_goal >= end)
 			grp_goal = -1;
-	} else {
-		if (grp_goal > 0)
-			start = grp_goal;
-		else
-			start = 0;
-		end = EXT2_BLOCKS_PER_GROUP(sb);
 	}
-
 	BUG_ON(start > EXT2_BLOCKS_PER_GROUP(sb));
 
-repeat:
 	if (grp_goal < 0) {
 		grp_goal = find_next_usable_block(start, bitmap_bh, end);
 		if (grp_goal < 0)
@@ -711,32 +698,23 @@ repeat:
 				;
 		}
 	}
-	start = grp_goal;
 
-	if (ext2_set_bit_atomic(sb_bgl_lock(EXT2_SB(sb), group), grp_goal,
-			       				bitmap_bh->b_data)) {
-		/*
-		 * The block was allocated by another thread, or it was
-		 * allocated and then freed by another thread
-		 */
-		start++;
-		grp_goal++;
-		if (start >= end)
-			goto fail_access;
-		goto repeat;
-	}
-	num++;
-	grp_goal++;
-	while (num < *count && grp_goal < end
-		&& !ext2_set_bit_atomic(sb_bgl_lock(EXT2_SB(sb), group),
+	for (; num < *count && grp_goal < end; grp_goal++) {
+		if (ext2_set_bit_atomic(sb_bgl_lock(EXT2_SB(sb), group),
 					grp_goal, bitmap_bh->b_data)) {
+			if (num == 0)
+				continue;
+			break;
+		}
 		num++;
-		grp_goal++;
 	}
+
+	if (num == 0)
+		goto fail_access;
+
 	*count = num;
 	return grp_goal - num;
 fail_access:
-	*count = num;
 	return -1;
 }
 
@@ -754,10 +732,9 @@ fail_access:
  *		but we will shift to the place where start_block is,
  *		then start from there, when looking for a reservable space.
  *
- * 	@size: the target new reservation window size
+ *	@sb: the super block.
  *
- * 	@group_first_block: the first block we consider to start
- *			the real search from
+ * 	@start_block: the first block we consider to start the real search from
  *
  * 	@last_block:
  *		the maximum block number that our goal reservable space
@@ -908,7 +885,7 @@ static int alloc_new_reservation(struct ext2_reserve_window_node *my_rsv,
 	spinlock_t *rsv_lock = &EXT2_SB(sb)->s_rsv_window_lock;
 
 	group_first_block = ext2_group_first_block_no(sb, group);
-	group_end_block = group_first_block + (EXT2_BLOCKS_PER_GROUP(sb) - 1);
+	group_end_block = ext2_group_last_block_no(sb, group);
 
 	if (grp_goal < 0)
 		start_block = group_first_block;
@@ -1115,7 +1092,7 @@ ext2_try_to_allocate_with_rsv(struct super_block *sb, unsigned int group,
 	 * first block is the block number of the first block in this group
 	 */
 	group_first_block = ext2_group_first_block_no(sb, group);
-	group_last_block = group_first_block + (EXT2_BLOCKS_PER_GROUP(sb) - 1);
+	group_last_block = ext2_group_last_block_no(sb, group);
 
 	/*
 	 * Basically we will allocate a new block from inode's reservation
@@ -1313,6 +1290,13 @@ retry_alloc:
 	if (free_blocks > 0) {
 		grp_target_blk = ((goal - le32_to_cpu(es->s_first_data_block)) %
 				EXT2_BLOCKS_PER_GROUP(sb));
+		/*
+		 * In case we retry allocation (due to fs reservation not
+		 * working out or fs corruption), the bitmap_bh is non-null
+		 * pointer and we have to release it before calling
+		 * read_block_bitmap().
+		 */
+		brelse(bitmap_bh);
 		bitmap_bh = read_block_bitmap(sb, group_no);
 		if (!bitmap_bh)
 			goto io_error;
@@ -1404,6 +1388,7 @@ allocated:
 		 * use.  So we may want to selectively mark some of the blocks
 		 * as free
 		 */
+		num = *count;
 		goto retry_alloc;
 	}
 
diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index 10ab238de9a6..8178bd38a9d6 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -813,6 +813,18 @@ ext2_group_first_block_no(struct super_block *sb, unsigned long group_no)
 		le32_to_cpu(EXT2_SB(sb)->s_es->s_first_data_block);
 }
 
+static inline ext2_fsblk_t
+ext2_group_last_block_no(struct super_block *sb, unsigned long group_no)
+{
+	struct ext2_sb_info *sbi = EXT2_SB(sb);
+
+	if (group_no == sbi->s_groups_count - 1)
+		return le32_to_cpu(sbi->s_es->s_blocks_count) - 1;
+	else
+		return ext2_group_first_block_no(sb, group_no) +
+			EXT2_BLOCKS_PER_GROUP(sb) - 1;
+}
+
 #define ext2_set_bit	__test_and_set_bit_le
 #define ext2_clear_bit	__test_and_clear_bit_le
 #define ext2_test_bit	test_bit_le
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 7004ce581a32..2875c0a705b5 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -701,10 +701,13 @@ static int ext2_get_blocks(struct inode *inode,
 		if (!partial) {
 			count++;
 			mutex_unlock(&ei->truncate_mutex);
-			if (err)
-				goto cleanup;
 			goto got_it;
 		}
+
+		if (err) {
+			mutex_unlock(&ei->truncate_mutex);
+			goto cleanup;
+		}
 	}
 
 	/*
@@ -801,7 +804,7 @@ int ext2_get_block(struct inode *inode, sector_t iblock,
 
 #ifdef CONFIG_FS_DAX
 static int ext2_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
-		unsigned flags, struct iomap *iomap)
+		unsigned flags, struct iomap *iomap, struct iomap *srcmap)
 {
 	unsigned int blkbits = inode->i_blkbits;
 	unsigned long first_block = offset >> blkbits;
@@ -874,11 +877,9 @@ static int ext2_readpage(struct file *file, struct page *page)
 	return mpage_readpage(page, ext2_get_block);
 }
 
-static int
-ext2_readpages(struct file *file, struct address_space *mapping,
-		struct list_head *pages, unsigned nr_pages)
+static void ext2_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, ext2_get_block);
+	mpage_readahead(rac, ext2_get_block);
 }
 
 static int
@@ -957,13 +958,14 @@ ext2_writepages(struct address_space *mapping, struct writeback_control *wbc)
 static int
 ext2_dax_writepages(struct address_space *mapping, struct writeback_control *wbc)
 {
-	return dax_writeback_mapping_range(mapping,
-			mapping->host->i_sb->s_bdev, wbc);
+	struct ext2_sb_info *sbi = EXT2_SB(mapping->host->i_sb);
+
+	return dax_writeback_mapping_range(mapping, sbi->s_daxdev, wbc);
 }
 
 const struct address_space_operations ext2_aops = {
 	.readpage		= ext2_readpage,
-	.readpages		= ext2_readpages,
+	.readahead		= ext2_readahead,
 	.writepage		= ext2_writepage,
 	.write_begin		= ext2_write_begin,
 	.write_end		= ext2_write_end,
@@ -977,7 +979,7 @@ const struct address_space_operations ext2_aops = {
 
 const struct address_space_operations ext2_nobh_aops = {
 	.readpage		= ext2_readpage,
-	.readpages		= ext2_readpages,
+	.readahead		= ext2_readahead,
 	.writepage		= ext2_nobh_writepage,
 	.write_begin		= ext2_nobh_write_begin,
 	.write_end		= nobh_write_end,
diff --git a/fs/ext2/ioctl.c b/fs/ext2/ioctl.c
index 1b853fb0b163..32a8d10b579d 100644
--- a/fs/ext2/ioctl.c
+++ b/fs/ext2/ioctl.c
@@ -145,10 +145,13 @@ setversion_out:
 		if (ei->i_block_alloc_info){
 			struct ext2_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node;
 			rsv->rsv_goal_size = rsv_window_size;
+		} else {
+			ret = -ENOMEM;
 		}
+
 		mutex_unlock(&ei->truncate_mutex);
 		mnt_drop_write_file(filp);
-		return 0;
+		return ret;
 	}
 	default:
 		return -ENOTTY;
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 30c630d73f0f..4a4ab683250d 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -702,13 +702,7 @@ static int ext2_check_descriptors(struct super_block *sb)
 	for (i = 0; i < sbi->s_groups_count; i++) {
 		struct ext2_group_desc *gdp = ext2_get_group_desc(sb, i, NULL);
 		ext2_fsblk_t first_block = ext2_group_first_block_no(sb, i);
-		ext2_fsblk_t last_block;
-
-		if (i == sbi->s_groups_count - 1)
-			last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1;
-		else
-			last_block = first_block +
-				(EXT2_BLOCKS_PER_GROUP(sb) - 1);
+		ext2_fsblk_t last_block = ext2_group_last_block_no(sb, i);
 
 		if (le32_to_cpu(gdp->bg_block_bitmap) < first_block ||
 		    le32_to_cpu(gdp->bg_block_bitmap) > last_block)
@@ -806,7 +800,6 @@ static unsigned long descriptor_loc(struct super_block *sb,
 {
 	struct ext2_sb_info *sbi = EXT2_SB(sb);
 	unsigned long bg, first_meta_bg;
-	int has_super = 0;
 	
 	first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg);
 
@@ -814,10 +807,8 @@ static unsigned long descriptor_loc(struct super_block *sb,
 	    nr < first_meta_bg)
 		return (logic_sb_block + nr + 1);
 	bg = sbi->s_desc_per_block * nr;
-	if (ext2_bg_has_super(sb, bg))
-		has_super = 1;
 
-	return ext2_group_first_block_no(sb, bg) + has_super;
+	return ext2_group_first_block_no(sb, bg) + ext2_bg_has_super(sb, bg);
 }
 
 static int ext2_fill_super(struct super_block *sb, void *data, int silent)
@@ -1082,9 +1073,9 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 
 	if (EXT2_BLOCKS_PER_GROUP(sb) == 0)
 		goto cantfind_ext2;
- 	sbi->s_groups_count = ((le32_to_cpu(es->s_blocks_count) -
- 				le32_to_cpu(es->s_first_data_block) - 1)
- 					/ EXT2_BLOCKS_PER_GROUP(sb)) + 1;
+	sbi->s_groups_count = ((le32_to_cpu(es->s_blocks_count) -
+				le32_to_cpu(es->s_first_data_block) - 1)
+					/ EXT2_BLOCKS_PER_GROUP(sb)) + 1;
 	db_count = (sbi->s_groups_count + EXT2_DESC_PER_BLOCK(sb) - 1) /
 		   EXT2_DESC_PER_BLOCK(sb);
 	sbi->s_group_desc = kmalloc_array (db_count,
@@ -1147,6 +1138,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 				ext2_count_dirs(sb), GFP_KERNEL);
 	}
 	if (err) {
+		ret = err;
 		ext2_msg(sb, KERN_ERR, "error: insufficient memory");
 		goto failed_mount3;
 	}
diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index 0456bc990b5e..943cc469f42f 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -56,6 +56,7 @@
 
 #include <linux/buffer_head.h>
 #include <linux/init.h>
+#include <linux/printk.h>
 #include <linux/slab.h>
 #include <linux/mbcache.h>
 #include <linux/quotaops.h>
@@ -84,8 +85,8 @@
 		printk("\n"); \
 	} while (0)
 #else
-# define ea_idebug(f...)
-# define ea_bdebug(f...)
+# define ea_idebug(inode, f...)	no_printk(f)
+# define ea_bdebug(bh, f...)	no_printk(f)
 #endif
 
 static int ext2_xattr_set2(struct inode *, struct buffer_head *,
@@ -790,7 +791,15 @@ ext2_xattr_delete_inode(struct inode *inode)
 	struct buffer_head *bh = NULL;
 	struct ext2_sb_info *sbi = EXT2_SB(inode->i_sb);
 
-	down_write(&EXT2_I(inode)->xattr_sem);
+	/*
+	 * We are the only ones holding inode reference. The xattr_sem should
+	 * better be unlocked! We could as well just not acquire xattr_sem at
+	 * all but this makes the code more futureproof. OTOH we need trylock
+	 * here to avoid false-positive warning from lockdep about reclaim
+	 * circular dependency.
+	 */
+	if (WARN_ON_ONCE(!down_write_trylock(&EXT2_I(inode)->xattr_sem)))
+		return;
 	if (!EXT2_I(inode)->i_file_acl)
 		goto cleanup;
 
@@ -864,8 +873,7 @@ ext2_xattr_cache_insert(struct mb_cache *cache, struct buffer_head *bh)
 				      true);
 	if (error) {
 		if (error == -EBUSY) {
-			ea_bdebug(bh, "already in cache (%d cache entries)",
-				atomic_read(&ext2_xattr_cache->c_entry_count));
+			ea_bdebug(bh, "already in cache");
 			error = 0;
 		}
 	} else
diff --git a/fs/ext2/xattr.h b/fs/ext2/xattr.h
index cee888cdc235..16272e6ddcf4 100644
--- a/fs/ext2/xattr.h
+++ b/fs/ext2/xattr.h
@@ -39,7 +39,7 @@ struct ext2_xattr_entry {
 	__le32	e_value_block;	/* disk block attribute is stored on (n/i) */
 	__le32	e_value_size;	/* size of attribute value */
 	__le32	e_hash;		/* hash value of name and value */
-	char	e_name[0];	/* attribute name */
+	char	e_name[];	/* attribute name */
 };
 
 #define EXT2_XATTR_PAD_BITS		2
diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig
index cbb5ca830e57..2a592e38cdfe 100644
--- a/fs/ext4/Kconfig
+++ b/fs/ext4/Kconfig
@@ -4,12 +4,7 @@
 # kernels after the removal of ext3 driver.
 config EXT3_FS
 	tristate "The Extended 3 (ext3) filesystem"
-	# These must match EXT4_FS selects...
 	select EXT4_FS
-	select JBD2
-	select CRC16
-	select CRYPTO
-	select CRYPTO_CRC32C
 	help
 	  This config option is here only for backward compatibility. ext3
 	  filesystem is now handled by the ext4 driver.
@@ -33,12 +28,12 @@ config EXT3_FS_SECURITY
 
 config EXT4_FS
 	tristate "The Extended 4 (ext4) filesystem"
-	# Please update EXT3_FS selects when changing these
 	select JBD2
 	select CRC16
 	select CRYPTO
 	select CRYPTO_CRC32C
 	select FS_IOMAP
+	select FS_ENCRYPTION_ALGS if FS_ENCRYPTION
 	help
 	  This is the next generation of the ext3 filesystem.
 
@@ -106,3 +101,20 @@ config EXT4_DEBUG
 	  If you select Y here, then you will be able to turn on debugging
 	  with a command such as:
 		echo 1 > /sys/module/ext4/parameters/mballoc_debug
+
+config EXT4_KUNIT_TESTS
+	tristate "KUnit tests for ext4"
+	select EXT4_FS
+	depends on KUNIT
+	help
+	  This builds the ext4 KUnit tests.
+
+	  KUnit tests run during boot and output the results to the debug log
+	  in TAP format (http://testanything.org/). Only useful for kernel devs
+	  running KUnit test harness and are not for inclusion into a production
+	  build.
+
+	  For more information on KUnit and unit tests in general please refer
+	  to the KUnit documentation in Documentation/dev-tools/kunit/.
+
+	  If unsure, say N.
diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile
index b17ddc229ac5..4ccb3c9189d8 100644
--- a/fs/ext4/Makefile
+++ b/fs/ext4/Makefile
@@ -13,4 +13,6 @@ ext4-y	:= balloc.o bitmap.o block_validity.o dir.o ext4_jbd2.o extents.o \
 
 ext4-$(CONFIG_EXT4_FS_POSIX_ACL)	+= acl.o
 ext4-$(CONFIG_EXT4_FS_SECURITY)		+= xattr_security.o
+ext4-inode-test-objs			+= inode-test.o
+obj-$(CONFIG_EXT4_KUNIT_TESTS)		+= ext4-inode-test.o
 ext4-$(CONFIG_FS_VERITY)		+= verity.o
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 0b202e00d93f..a32e5f7b5385 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -270,6 +270,7 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb,
 	ext4_group_t ngroups = ext4_get_groups_count(sb);
 	struct ext4_group_desc *desc;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	struct buffer_head *bh_p;
 
 	if (block_group >= ngroups) {
 		ext4_error(sb, "block_group >= groups_count - block_group = %u,"
@@ -280,7 +281,14 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb,
 
 	group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb);
 	offset = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1);
-	if (!sbi->s_group_desc[group_desc]) {
+	bh_p = sbi_array_rcu_deref(sbi, s_group_desc, group_desc);
+	/*
+	 * sbi_array_rcu_deref returns with rcu unlocked, this is ok since
+	 * the pointer being dereferenced won't be dereferenced again. By
+	 * looking at the usage in add_new_gdb() the value isn't modified,
+	 * just the pointer, and so it remains valid.
+	 */
+	if (!bh_p) {
 		ext4_error(sb, "Group descriptor not loaded - "
 			   "block_group = %u, group_desc = %u, desc = %u",
 			   block_group, group_desc, offset);
@@ -288,10 +296,10 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb,
 	}
 
 	desc = (struct ext4_group_desc *)(
-		(__u8 *)sbi->s_group_desc[group_desc]->b_data +
+		(__u8 *)bh_p->b_data +
 		offset * EXT4_DESC_SIZE(sb));
 	if (bh)
-		*bh = sbi->s_group_desc[group_desc];
+		*bh = bh_p;
 	return desc;
 }
 
@@ -371,7 +379,8 @@ static int ext4_validate_block_bitmap(struct super_block *sb,
 	if (buffer_verified(bh))
 		goto verified;
 	if (unlikely(!ext4_block_bitmap_csum_verify(sb, block_group,
-			desc, bh))) {
+						    desc, bh) ||
+		     ext4_simulate_fail(sb, EXT4_SIM_BBITMAP_CRC))) {
 		ext4_unlock_group(sb, block_group);
 		ext4_error(sb, "bg %u: bad block bitmap checksum", block_group);
 		ext4_mark_group_bitmap_corrupted(sb, block_group,
@@ -401,7 +410,7 @@ verified:
  * Read the bitmap for a given block_group,and validate the
  * bits for block/inode/inode tables are set in the bitmaps
  *
- * Return buffer_head on success or NULL in case of failure.
+ * Return buffer_head on success or an ERR_PTR in case of failure.
  */
 struct buffer_head *
 ext4_read_block_bitmap_nowait(struct super_block *sb, ext4_group_t block_group)
@@ -493,7 +502,7 @@ out:
 	return ERR_PTR(err);
 }
 
-/* Returns 0 on success, 1 on error */
+/* Returns 0 on success, -errno on error */
 int ext4_wait_block_bitmap(struct super_block *sb, ext4_group_t block_group,
 			   struct buffer_head *bh)
 {
@@ -505,10 +514,11 @@ int ext4_wait_block_bitmap(struct super_block *sb, ext4_group_t block_group,
 	if (!desc)
 		return -EFSCORRUPTED;
 	wait_on_buffer(bh);
+	ext4_simulate_fail_bh(sb, bh, EXT4_SIM_BBITMAP_EIO);
 	if (!buffer_uptodate(bh)) {
-		ext4_error(sb, "Cannot read block bitmap - "
-			   "block_group = %u, block_bitmap = %llu",
-			   block_group, (unsigned long long) bh->b_blocknr);
+		ext4_error_err(sb, EIO, "Cannot read block bitmap - "
+			       "block_group = %u, block_bitmap = %llu",
+			       block_group, (unsigned long long) bh->b_blocknr);
 		ext4_mark_group_bitmap_corrupted(sb, block_group,
 					EXT4_GROUP_INFO_BBITMAP_CORRUPT);
 		return -EIO;
diff --git a/fs/ext4/block_validity.c b/fs/ext4/block_validity.c
index d4d4fdfac1a6..16e9b2fda03a 100644
--- a/fs/ext4/block_validity.c
+++ b/fs/ext4/block_validity.c
@@ -133,10 +133,13 @@ static void debug_print_tree(struct ext4_sb_info *sbi)
 {
 	struct rb_node *node;
 	struct ext4_system_zone *entry;
+	struct ext4_system_blocks *system_blks;
 	int first = 1;
 
 	printk(KERN_INFO "System zones: ");
-	node = rb_first(&sbi->system_blks->root);
+	rcu_read_lock();
+	system_blks = rcu_dereference(sbi->system_blks);
+	node = rb_first(&system_blks->root);
 	while (node) {
 		entry = rb_entry(node, struct ext4_system_zone, node);
 		printk(KERN_CONT "%s%llu-%llu", first ? "" : ", ",
@@ -144,6 +147,7 @@ static void debug_print_tree(struct ext4_sb_info *sbi)
 		first = 0;
 		node = rb_next(node);
 	}
+	rcu_read_unlock();
 	printk(KERN_CONT "\n");
 }
 
@@ -162,10 +166,8 @@ static int ext4_data_block_valid_rcu(struct ext4_sb_info *sbi,
 
 	if ((start_blk <= le32_to_cpu(sbi->s_es->s_first_data_block)) ||
 	    (start_blk + count < start_blk) ||
-	    (start_blk + count > ext4_blocks_count(sbi->s_es))) {
-		sbi->s_es->s_last_error_block = cpu_to_le64(start_blk);
+	    (start_blk + count > ext4_blocks_count(sbi->s_es)))
 		return 0;
-	}
 
 	if (system_blks == NULL)
 		return 1;
@@ -177,10 +179,8 @@ static int ext4_data_block_valid_rcu(struct ext4_sb_info *sbi,
 			n = n->rb_left;
 		else if (start_blk >= (entry->start_blk + entry->count))
 			n = n->rb_right;
-		else {
-			sbi->s_es->s_last_error_block = cpu_to_le64(start_blk);
+		else
 			return 0;
-		}
 	}
 	return 1;
 }
@@ -203,6 +203,7 @@ static int ext4_protect_reserved_inode(struct super_block *sb,
 		return PTR_ERR(inode);
 	num = (inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits;
 	while (i < num) {
+		cond_resched();
 		map.m_lblk = i;
 		map.m_len = num - i;
 		n = ext4_map_blocks(NULL, inode, &map, 0);
@@ -215,10 +216,12 @@ static int ext4_protect_reserved_inode(struct super_block *sb,
 		} else {
 			if (!ext4_data_block_valid_rcu(sbi, system_blks,
 						map.m_pblk, n)) {
-				ext4_error(sb, "blocks %llu-%llu from inode %u "
-					   "overlap system zone", map.m_pblk,
-					   map.m_pblk + map.m_len - 1, ino);
 				err = -EFSCORRUPTED;
+				__ext4_error(sb, __func__, __LINE__, -err,
+					     map.m_pblk, "blocks %llu-%llu "
+					     "from inode %u overlap system zone",
+					     map.m_pblk,
+					     map.m_pblk + map.m_len - 1, ino);
 				break;
 			}
 			err = add_system_zone(system_blks, map.m_pblk, n);
@@ -360,7 +363,6 @@ int ext4_data_block_valid(struct ext4_sb_info *sbi, ext4_fsblk_t start_blk,
 int ext4_check_blockref(const char *function, unsigned int line,
 			struct inode *inode, __le32 *p, unsigned int max)
 {
-	struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
 	__le32 *bref = p;
 	unsigned int blk;
 
@@ -374,7 +376,6 @@ int ext4_check_blockref(const char *function, unsigned int line,
 		if (blk &&
 		    unlikely(!ext4_data_block_valid(EXT4_SB(inode->i_sb),
 						    blk, 1))) {
-			es->s_last_error_block = cpu_to_le64(blk);
 			ext4_error_inode(inode, function, line, blk,
 					 "invalid block");
 			return -EFSCORRUPTED;
diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index 9fdd2b269d61..c654205f648d 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -72,6 +72,7 @@ int __ext4_check_dir_entry(const char *function, unsigned int line,
 	const char *error_msg = NULL;
 	const int rlen = ext4_rec_len_from_disk(de->rec_len,
 						dir->i_sb->s_blocksize);
+	const int next_offset = ((char *) de - buf) + rlen;
 
 	if (unlikely(rlen < EXT4_DIR_REC_LEN(1)))
 		error_msg = "rec_len is smaller than minimal";
@@ -79,8 +80,11 @@ int __ext4_check_dir_entry(const char *function, unsigned int line,
 		error_msg = "rec_len % 4 != 0";
 	else if (unlikely(rlen < EXT4_DIR_REC_LEN(de->name_len)))
 		error_msg = "rec_len is too small for name_len";
-	else if (unlikely(((char *) de - buf) + rlen > size))
+	else if (unlikely(next_offset > size))
 		error_msg = "directory entry overrun";
+	else if (unlikely(next_offset > size - EXT4_DIR_REC_LEN(1) &&
+			  next_offset != size))
+		error_msg = "directory entry too close to block end";
 	else if (unlikely(le32_to_cpu(de->inode) >
 			le32_to_cpu(EXT4_SB(dir->i_sb)->s_es->s_inodes_count)))
 		error_msg = "inode out of bounds";
@@ -116,7 +120,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 
 	if (IS_ENCRYPTED(inode)) {
 		err = fscrypt_get_encryption_info(inode);
-		if (err && err != -ENOKEY)
+		if (err)
 			return err;
 	}
 
@@ -125,12 +129,14 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 		if (err != ERR_BAD_DX_DIR) {
 			return err;
 		}
-		/*
-		 * We don't set the inode dirty flag since it's not
-		 * critical that it get flushed back to the disk.
-		 */
-		ext4_clear_inode_flag(file_inode(file),
-				      EXT4_INODE_INDEX);
+		/* Can we just clear INDEX flag to ignore htree information? */
+		if (!ext4_has_metadata_csum(sb)) {
+			/*
+			 * We don't set the inode dirty flag since it's not
+			 * critical that it gets flushed back to the disk.
+			 */
+			ext4_clear_inode_flag(inode, EXT4_INODE_INDEX);
+		}
 	}
 
 	if (ext4_has_inline_data(inode)) {
@@ -386,7 +392,7 @@ struct fname {
 	__u32		inode;
 	__u8		name_len;
 	__u8		file_type;
-	char		name[0];
+	char		name[];
 };
 
 /*
@@ -458,7 +464,6 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 	new_fn->name_len = ent_name->len;
 	new_fn->file_type = dirent->file_type;
 	memcpy(new_fn->name, ent_name->name, ent_name->len);
-	new_fn->name[ent_name->len] = 0;
 
 	while (*p) {
 		parent = *p;
@@ -668,9 +673,11 @@ static int ext4_d_compare(const struct dentry *dentry, unsigned int len,
 			  const char *str, const struct qstr *name)
 {
 	struct qstr qstr = {.name = str, .len = len };
-	struct inode *inode = dentry->d_parent->d_inode;
+	const struct dentry *parent = READ_ONCE(dentry->d_parent);
+	const struct inode *inode = READ_ONCE(parent->d_inode);
 
-	if (!IS_CASEFOLDED(inode) || !EXT4_SB(inode->i_sb)->s_encoding) {
+	if (!inode || !IS_CASEFOLDED(inode) ||
+	    !EXT4_SB(inode->i_sb)->s_encoding) {
 		if (len != name->len)
 			return -1;
 		return memcmp(str, name->name, len);
@@ -683,10 +690,11 @@ static int ext4_d_hash(const struct dentry *dentry, struct qstr *str)
 {
 	const struct ext4_sb_info *sbi = EXT4_SB(dentry->d_sb);
 	const struct unicode_map *um = sbi->s_encoding;
+	const struct inode *inode = READ_ONCE(dentry->d_inode);
 	unsigned char *norm;
 	int len, ret = 0;
 
-	if (!IS_CASEFOLDED(dentry->d_inode) || !um)
+	if (!inode || !IS_CASEFOLDED(inode) || !um)
 		return 0;
 
 	norm = kmalloc(PATH_MAX, GFP_ATOMIC);
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 03db3e71676c..15b062efcff1 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -198,6 +198,12 @@ struct ext4_system_blocks {
  */
 #define	EXT4_IO_END_UNWRITTEN	0x0001
 
+struct ext4_io_end_vec {
+	struct list_head list;		/* list of io_end_vec */
+	loff_t offset;			/* offset in the file */
+	ssize_t size;			/* size of the extent */
+};
+
 /*
  * For converting unwritten extents on a work queue. 'handle' is used for
  * buffered writeback.
@@ -211,8 +217,7 @@ typedef struct ext4_io_end {
 						 * bios covering the extent */
 	unsigned int		flag;		/* unwritten or not */
 	atomic_t		count;		/* reference counter */
-	loff_t			offset;		/* offset in the file */
-	ssize_t			size;		/* size of the extent */
+	struct list_head	list_vec;	/* list of ext4_io_end_vec */
 } ext4_io_end_t;
 
 struct ext4_io_submit {
@@ -409,7 +414,7 @@ struct flex_groups {
 #define EXT4_EXTENTS_FL			0x00080000 /* Inode uses extents */
 #define EXT4_VERITY_FL			0x00100000 /* Verity protected inode */
 #define EXT4_EA_INODE_FL	        0x00200000 /* Inode used for large EA */
-#define EXT4_EOFBLOCKS_FL		0x00400000 /* Blocks allocated beyond EOF */
+/* 0x00400000 was formerly EXT4_EOFBLOCKS_FL */
 #define EXT4_INLINE_DATA_FL		0x10000000 /* Inode has inline data. */
 #define EXT4_PROJINHERIT_FL		0x20000000 /* Create with parents projid */
 #define EXT4_CASEFOLD_FL		0x40000000 /* Casefolded file */
@@ -482,7 +487,7 @@ enum {
 	EXT4_INODE_EXTENTS	= 19,	/* Inode uses extents */
 	EXT4_INODE_VERITY	= 20,	/* Verity protected inode */
 	EXT4_INODE_EA_INODE	= 21,	/* Inode used for large EA */
-	EXT4_INODE_EOFBLOCKS	= 22,	/* Blocks allocated beyond EOF */
+/* 22 was formerly EXT4_INODE_EOFBLOCKS */
 	EXT4_INODE_INLINE_DATA	= 28,	/* Data in inode. */
 	EXT4_INODE_PROJINHERIT	= 29,	/* Create with parents projid */
 	EXT4_INODE_RESERVED	= 31,	/* reserved for ext4 lib */
@@ -528,7 +533,6 @@ static inline void ext4_check_flag_values(void)
 	CHECK_FLAG_VALUE(EXTENTS);
 	CHECK_FLAG_VALUE(VERITY);
 	CHECK_FLAG_VALUE(EA_INODE);
-	CHECK_FLAG_VALUE(EOFBLOCKS);
 	CHECK_FLAG_VALUE(INLINE_DATA);
 	CHECK_FLAG_VALUE(PROJINHERIT);
 	CHECK_FLAG_VALUE(RESERVED);
@@ -718,7 +722,7 @@ enum {
 #define EXT4_MAX_BLOCK_FILE_PHYS	0xFFFFFFFF
 
 /* Max logical block we can support */
-#define EXT4_MAX_LOGICAL_BLOCK		0xFFFFFFFF
+#define EXT4_MAX_LOGICAL_BLOCK		0xFFFFFFFE
 
 /*
  * Structure of an inode on the disk
@@ -1047,8 +1051,6 @@ struct ext4_inode_info {
 	/* allocation reservation info for delalloc */
 	/* In case of bigalloc, this refer to clusters rather than blocks */
 	unsigned int i_reserved_data_blocks;
-	ext4_lblk_t i_da_metadata_calc_last_lblock;
-	int i_da_metadata_calc_len;
 
 	/* pending cluster reservations for bigalloc file systems */
 	struct ext4_pending_tree i_pending_tree;
@@ -1338,7 +1340,8 @@ struct ext4_super_block {
 	__u8	s_lastcheck_hi;
 	__u8	s_first_error_time_hi;
 	__u8	s_last_error_time_hi;
-	__u8	s_pad[2];
+	__u8	s_first_error_errcode;
+	__u8    s_last_error_errcode;
 	__le16  s_encoding;		/* Filename charset encoding */
 	__le16  s_encoding_flags;	/* Filename charset encoding flags */
 	__le32	s_reserved[95];		/* Padding to the end of the block */
@@ -1354,11 +1357,9 @@ struct ext4_super_block {
  */
 #define EXT4_MF_MNTDIR_SAMPLED		0x0001
 #define EXT4_MF_FS_ABORTED		0x0002	/* Fatal error detected */
-#define EXT4_MF_TEST_DUMMY_ENCRYPTION	0x0004
 
 #ifdef CONFIG_FS_ENCRYPTION
-#define DUMMY_ENCRYPTION_ENABLED(sbi) (unlikely((sbi)->s_mount_flags & \
-						EXT4_MF_TEST_DUMMY_ENCRYPTION))
+#define DUMMY_ENCRYPTION_ENABLED(sbi) ((sbi)->s_dummy_enc_ctx.ctx != NULL)
 #else
 #define DUMMY_ENCRYPTION_ENABLED(sbi) (0)
 #endif
@@ -1396,7 +1397,7 @@ struct ext4_sb_info {
 	loff_t s_bitmap_maxbytes;	/* max bytes for bitmap files */
 	struct buffer_head * s_sbh;	/* Buffer containing the super block */
 	struct ext4_super_block *s_es;	/* Pointer to the super block in the buffer */
-	struct buffer_head **s_group_desc;
+	struct buffer_head * __rcu *s_group_desc;
 	unsigned int s_mount_opt;
 	unsigned int s_mount_opt2;
 	unsigned int s_mount_flags;
@@ -1458,7 +1459,7 @@ struct ext4_sb_info {
 #endif
 
 	/* for buddy allocator */
-	struct ext4_group_info ***s_group_info;
+	struct ext4_group_info ** __rcu *s_group_info;
 	struct inode *s_buddy_cache;
 	spinlock_t s_md_lock;
 	unsigned short *s_mb_offsets;
@@ -1508,7 +1509,7 @@ struct ext4_sb_info {
 	unsigned int s_extent_max_zeroout_kb;
 
 	unsigned int s_log_groups_per_flex;
-	struct flex_groups *s_flex_groups;
+	struct flex_groups * __rcu *s_flex_groups;
 	ext4_group_t s_flex_groups_allocated;
 
 	/* workqueue for reserved extent conversions (buffered io) */
@@ -1548,9 +1549,18 @@ struct ext4_sb_info {
 	struct ratelimit_state s_warning_ratelimit_state;
 	struct ratelimit_state s_msg_ratelimit_state;
 
-	/* Barrier between changing inodes' journal flags and writepages ops. */
-	struct percpu_rw_semaphore s_journal_flag_rwsem;
+	/* Encryption context for '-o test_dummy_encryption' */
+	struct fscrypt_dummy_context s_dummy_enc_ctx;
+
+	/*
+	 * Barrier between writepages ops and changing any inode's JOURNAL_DATA
+	 * or EXTENTS flag.
+	 */
+	struct percpu_rw_semaphore s_writepages_rwsem;
 	struct dax_device *s_daxdev;
+#ifdef CONFIG_EXT4_DEBUG
+	unsigned long s_simulate_fail;
+#endif
 };
 
 static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb)
@@ -1570,6 +1580,83 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
 }
 
 /*
+ * Returns: sbi->field[index]
+ * Used to access an array element from the following sbi fields which require
+ * rcu protection to avoid dereferencing an invalid pointer due to reassignment
+ * - s_group_desc
+ * - s_group_info
+ * - s_flex_group
+ */
+#define sbi_array_rcu_deref(sbi, field, index)				   \
+({									   \
+	typeof(*((sbi)->field)) _v;					   \
+	rcu_read_lock();						   \
+	_v = ((typeof(_v)*)rcu_dereference((sbi)->field))[index];	   \
+	rcu_read_unlock();						   \
+	_v;								   \
+})
+
+/*
+ * Simulate_fail codes
+ */
+#define EXT4_SIM_BBITMAP_EIO	1
+#define EXT4_SIM_BBITMAP_CRC	2
+#define EXT4_SIM_IBITMAP_EIO	3
+#define EXT4_SIM_IBITMAP_CRC	4
+#define EXT4_SIM_INODE_EIO	5
+#define EXT4_SIM_INODE_CRC	6
+#define EXT4_SIM_DIRBLOCK_EIO	7
+#define EXT4_SIM_DIRBLOCK_CRC	8
+
+static inline bool ext4_simulate_fail(struct super_block *sb,
+				     unsigned long code)
+{
+#ifdef CONFIG_EXT4_DEBUG
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+
+	if (unlikely(sbi->s_simulate_fail == code)) {
+		sbi->s_simulate_fail = 0;
+		return true;
+	}
+#endif
+	return false;
+}
+
+static inline void ext4_simulate_fail_bh(struct super_block *sb,
+					 struct buffer_head *bh,
+					 unsigned long code)
+{
+	if (!IS_ERR(bh) && ext4_simulate_fail(sb, code))
+		clear_buffer_uptodate(bh);
+}
+
+/*
+ * Error number codes for s_{first,last}_error_errno
+ *
+ * Linux errno numbers are architecture specific, so we need to translate
+ * them into something which is architecture independent.   We don't define
+ * codes for all errno's; just the ones which are most likely to be the cause
+ * of an ext4_error() call.
+ */
+#define EXT4_ERR_UNKNOWN	 1
+#define EXT4_ERR_EIO		 2
+#define EXT4_ERR_ENOMEM		 3
+#define EXT4_ERR_EFSBADCRC	 4
+#define EXT4_ERR_EFSCORRUPTED	 5
+#define EXT4_ERR_ENOSPC		 6
+#define EXT4_ERR_ENOKEY		 7
+#define EXT4_ERR_EROFS		 8
+#define EXT4_ERR_EFBIG		 9
+#define EXT4_ERR_EEXIST		10
+#define EXT4_ERR_ERANGE		11
+#define EXT4_ERR_EOVERFLOW	12
+#define EXT4_ERR_EBUSY		13
+#define EXT4_ERR_ENOTDIR	14
+#define EXT4_ERR_ENOTEMPTY	15
+#define EXT4_ERR_ESHUTDOWN	16
+#define EXT4_ERR_EFAULT		17
+
+/*
  * Inode dynamic state flags
  */
 enum {
@@ -1579,7 +1666,6 @@ enum {
 	EXT4_STATE_NO_EXPAND,		/* No space for expansion */
 	EXT4_STATE_DA_ALLOC_CLOSE,	/* Alloc DA blks on close */
 	EXT4_STATE_EXT_MIGRATE,		/* Inode is migrating */
-	EXT4_STATE_DIO_UNWRITTEN,	/* need convert on dio done*/
 	EXT4_STATE_NEWENTRY,		/* File just added to dir */
 	EXT4_STATE_MAY_INLINE_DATA,	/* may have in-inode data */
 	EXT4_STATE_EXT_PRECACHED,	/* extents have been precached */
@@ -1678,6 +1764,7 @@ static inline bool ext4_verity_in_progress(struct inode *inode)
 #define EXT4_FEATURE_COMPAT_RESIZE_INODE	0x0010
 #define EXT4_FEATURE_COMPAT_DIR_INDEX		0x0020
 #define EXT4_FEATURE_COMPAT_SPARSE_SUPER2	0x0200
+#define EXT4_FEATURE_COMPAT_STABLE_INODES	0x0800
 
 #define EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER	0x0001
 #define EXT4_FEATURE_RO_COMPAT_LARGE_FILE	0x0002
@@ -1779,6 +1866,7 @@ EXT4_FEATURE_COMPAT_FUNCS(xattr,		EXT_ATTR)
 EXT4_FEATURE_COMPAT_FUNCS(resize_inode,		RESIZE_INODE)
 EXT4_FEATURE_COMPAT_FUNCS(dir_index,		DIR_INDEX)
 EXT4_FEATURE_COMPAT_FUNCS(sparse_super2,	SPARSE_SUPER2)
+EXT4_FEATURE_COMPAT_FUNCS(stable_inodes,	STABLE_INODES)
 
 EXT4_FEATURE_RO_COMPAT_FUNCS(sparse_super,	SPARSE_SUPER)
 EXT4_FEATURE_RO_COMPAT_FUNCS(large_file,	LARGE_FILE)
@@ -2476,8 +2564,11 @@ void ext4_insert_dentry(struct inode *inode,
 			struct ext4_filename *fname);
 static inline void ext4_update_dx_flag(struct inode *inode)
 {
-	if (!ext4_has_feature_dir_index(inode->i_sb))
+	if (!ext4_has_feature_dir_index(inode->i_sb)) {
+		/* ext4_iget() should have caught this... */
+		WARN_ON_ONCE(ext4_has_feature_metadata_csum(inode->i_sb));
 		ext4_clear_inode_flag(inode, EXT4_INODE_INDEX);
+	}
 }
 static const unsigned char ext4_filetype_table[] = {
 	DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK
@@ -2560,8 +2651,6 @@ int ext4_get_block_unwritten(struct inode *inode, sector_t iblock,
 			     struct buffer_head *bh_result, int create);
 int ext4_get_block(struct inode *inode, sector_t iblock,
 		   struct buffer_head *bh_result, int create);
-int ext4_dio_get_block(struct inode *inode, sector_t iblock,
-		       struct buffer_head *bh_result, int create);
 int ext4_da_get_block_prep(struct inode *inode, sector_t iblock,
 			   struct buffer_head *bh, int create);
 int ext4_walk_page_buffers(handle_t *handle,
@@ -2604,7 +2693,6 @@ extern int ext4_can_truncate(struct inode *inode);
 extern int ext4_truncate(struct inode *);
 extern int ext4_break_layouts(struct inode *);
 extern int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length);
-extern int ext4_truncate_restart_trans(handle_t *, struct inode *, int nblocks);
 extern void ext4_set_inode_flags(struct inode *);
 extern int ext4_alloc_da_blocks(struct inode *inode);
 extern void ext4_set_aops(struct inode *inode);
@@ -2625,7 +2713,6 @@ extern int ext4_issue_zeroout(struct inode *inode, ext4_lblk_t lblk,
 /* indirect.c */
 extern int ext4_ind_map_blocks(handle_t *handle, struct inode *inode,
 				struct ext4_map_blocks *map, int flags);
-extern int ext4_ind_calc_metadata_amount(struct inode *inode, sector_t lblock);
 extern int ext4_ind_trans_blocks(struct inode *inode, int nrblocks);
 extern void ext4_ind_truncate(handle_t *, struct inode *inode);
 extern int ext4_ind_remove_space(handle_t *handle, struct inode *inode,
@@ -2663,6 +2750,7 @@ extern int ext4_generic_delete_entry(handle_t *handle,
 extern bool ext4_empty_dir(struct inode *inode);
 
 /* resize.c */
+extern void ext4_kvfree_array_rcu(void *to_free);
 extern int ext4_group_add(struct super_block *sb,
 				struct ext4_new_group_data *input);
 extern int ext4_group_extend(struct super_block *sb,
@@ -2676,8 +2764,6 @@ extern struct buffer_head *ext4_sb_bread(struct super_block *sb,
 extern int ext4_seq_options_show(struct seq_file *seq, void *offset);
 extern int ext4_calculate_overhead(struct super_block *sb);
 extern void ext4_superblock_csum_set(struct super_block *sb);
-extern void *ext4_kvmalloc(size_t size, gfp_t flags);
-extern void *ext4_kvzalloc(size_t size, gfp_t flags);
 extern int ext4_alloc_flex_bg_array(struct super_block *sb,
 				    ext4_group_t ngroup);
 extern const char *ext4_decode_error(struct super_block *sb, int errno,
@@ -2686,19 +2772,19 @@ extern void ext4_mark_group_bitmap_corrupted(struct super_block *sb,
 					     ext4_group_t block_group,
 					     unsigned int flags);
 
-extern __printf(4, 5)
-void __ext4_error(struct super_block *, const char *, unsigned int,
+extern __printf(6, 7)
+void __ext4_error(struct super_block *, const char *, unsigned int, int, __u64,
 		  const char *, ...);
-extern __printf(5, 6)
-void __ext4_error_inode(struct inode *, const char *, unsigned int, ext4_fsblk_t,
-		      const char *, ...);
+extern __printf(6, 7)
+void __ext4_error_inode(struct inode *, const char *, unsigned int,
+			ext4_fsblk_t, int, const char *, ...);
 extern __printf(5, 6)
 void __ext4_error_file(struct file *, const char *, unsigned int, ext4_fsblk_t,
 		     const char *, ...);
 extern void __ext4_std_error(struct super_block *, const char *,
 			     unsigned int, int);
-extern __printf(4, 5)
-void __ext4_abort(struct super_block *, const char *, unsigned int,
+extern __printf(5, 6)
+void __ext4_abort(struct super_block *, const char *, unsigned int, int,
 		  const char *, ...);
 extern __printf(4, 5)
 void __ext4_warning(struct super_block *, const char *, unsigned int,
@@ -2719,8 +2805,12 @@ void __ext4_grp_locked_error(const char *, unsigned int,
 #define EXT4_ERROR_INODE(inode, fmt, a...) \
 	ext4_error_inode((inode), __func__, __LINE__, 0, (fmt), ## a)
 
-#define EXT4_ERROR_INODE_BLOCK(inode, block, fmt, a...)			\
-	ext4_error_inode((inode), __func__, __LINE__, (block), (fmt), ## a)
+#define EXT4_ERROR_INODE_ERR(inode, err, fmt, a...)			\
+	__ext4_error_inode((inode), __func__, __LINE__, 0, (err), (fmt), ## a)
+
+#define ext4_error_inode_block(inode, block, err, fmt, a...)		\
+	__ext4_error_inode((inode), __func__, __LINE__, (block), (err),	\
+			   (fmt), ## a)
 
 #define EXT4_ERROR_FILE(file, block, fmt, a...)				\
 	ext4_error_file((file), __func__, __LINE__, (block), (fmt), ## a)
@@ -2728,13 +2818,18 @@ void __ext4_grp_locked_error(const char *, unsigned int,
 #ifdef CONFIG_PRINTK
 
 #define ext4_error_inode(inode, func, line, block, fmt, ...)		\
-	__ext4_error_inode(inode, func, line, block, fmt, ##__VA_ARGS__)
+	__ext4_error_inode(inode, func, line, block, 0, fmt, ##__VA_ARGS__)
+#define ext4_error_inode_err(inode, func, line, block, err, fmt, ...)	\
+	__ext4_error_inode((inode), (func), (line), (block), 		\
+			   (err), (fmt), ##__VA_ARGS__)
 #define ext4_error_file(file, func, line, block, fmt, ...)		\
 	__ext4_error_file(file, func, line, block, fmt, ##__VA_ARGS__)
 #define ext4_error(sb, fmt, ...)					\
-	__ext4_error(sb, __func__, __LINE__, fmt, ##__VA_ARGS__)
-#define ext4_abort(sb, fmt, ...)					\
-	__ext4_abort(sb, __func__, __LINE__, fmt, ##__VA_ARGS__)
+	__ext4_error((sb), __func__, __LINE__, 0, 0, (fmt), ##__VA_ARGS__)
+#define ext4_error_err(sb, err, fmt, ...)				\
+	__ext4_error((sb), __func__, __LINE__, (err), 0, (fmt), ##__VA_ARGS__)
+#define ext4_abort(sb, err, fmt, ...)					\
+	__ext4_abort((sb), __func__, __LINE__, (err), (fmt), ##__VA_ARGS__)
 #define ext4_warning(sb, fmt, ...)					\
 	__ext4_warning(sb, __func__, __LINE__, fmt, ##__VA_ARGS__)
 #define ext4_warning_inode(inode, fmt, ...)				\
@@ -2752,7 +2847,12 @@ void __ext4_grp_locked_error(const char *, unsigned int,
 #define ext4_error_inode(inode, func, line, block, fmt, ...)		\
 do {									\
 	no_printk(fmt, ##__VA_ARGS__);					\
-	__ext4_error_inode(inode, "", 0, block, " ");			\
+	__ext4_error_inode(inode, "", 0, block, 0, " ");		\
+} while (0)
+#define ext4_error_inode_err(inode, func, line, block, err, fmt, ...)	\
+do {									\
+	no_printk(fmt, ##__VA_ARGS__);					\
+	__ext4_error_inode(inode, "", 0, block, err, " ");		\
 } while (0)
 #define ext4_error_file(file, func, line, block, fmt, ...)		\
 do {									\
@@ -2762,12 +2862,17 @@ do {									\
 #define ext4_error(sb, fmt, ...)					\
 do {									\
 	no_printk(fmt, ##__VA_ARGS__);					\
-	__ext4_error(sb, "", 0, " ");					\
+	__ext4_error(sb, "", 0, 0, 0, " ");				\
 } while (0)
-#define ext4_abort(sb, fmt, ...)					\
+#define ext4_error_err(sb, err, fmt, ...)				\
 do {									\
 	no_printk(fmt, ##__VA_ARGS__);					\
-	__ext4_abort(sb, "", 0, " ");					\
+	__ext4_error(sb, "", 0, err, 0, " ");				\
+} while (0)
+#define ext4_abort(sb, err, fmt, ...)					\
+do {									\
+	no_printk(fmt, ##__VA_ARGS__);					\
+	__ext4_abort(sb, "", 0, err, " ");				\
 } while (0)
 #define ext4_warning(sb, fmt, ...)					\
 do {									\
@@ -2910,13 +3015,13 @@ static inline
 struct ext4_group_info *ext4_get_group_info(struct super_block *sb,
 					    ext4_group_t group)
 {
-	 struct ext4_group_info ***grp_info;
+	 struct ext4_group_info **grp_info;
 	 long indexv, indexh;
 	 BUG_ON(group >= EXT4_SB(sb)->s_groups_count);
-	 grp_info = EXT4_SB(sb)->s_group_info;
 	 indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb));
 	 indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1);
-	 return grp_info[indexv][indexh];
+	 grp_info = sbi_array_rcu_deref(EXT4_SB(sb), s_group_info, indexv);
+	 return grp_info[indexh];
 }
 
 /*
@@ -2966,7 +3071,7 @@ static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize)
 		     !inode_is_locked(inode));
 	down_write(&EXT4_I(inode)->i_data_sem);
 	if (newsize > EXT4_I(inode)->i_disksize)
-		EXT4_I(inode)->i_disksize = newsize;
+		WRITE_ONCE(EXT4_I(inode)->i_disksize, newsize);
 	up_write(&EXT4_I(inode)->i_data_sem);
 }
 
@@ -3212,9 +3317,8 @@ static inline void ext4_set_de_type(struct super_block *sb,
 }
 
 /* readpages.c */
-extern int ext4_mpage_readpages(struct address_space *mapping,
-				struct list_head *pages, struct page *page,
-				unsigned nr_pages, bool is_readahead);
+extern int ext4_mpage_readpages(struct inode *inode,
+		struct readahead_control *rac, struct page *page);
 extern int __init ext4_init_post_read_processing(void);
 extern void ext4_exit_post_read_processing(void);
 
@@ -3251,7 +3355,6 @@ struct ext4_extent;
 #define EXT_MAX_BLOCKS	0xffffffff
 
 extern int ext4_ext_tree_init(handle_t *handle, struct inode *);
-extern int ext4_ext_writepage_trans_blocks(struct inode *, int);
 extern int ext4_ext_index_trans_blocks(struct inode *inode, int extents);
 extern int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			       struct ext4_map_blocks *map, int flags);
@@ -3264,16 +3367,13 @@ extern long ext4_fallocate(struct file *file, int mode, loff_t offset,
 			  loff_t len);
 extern int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode,
 					  loff_t offset, ssize_t len);
+extern int ext4_convert_unwritten_io_end_vec(handle_t *handle,
+					     ext4_io_end_t *io_end);
 extern int ext4_map_blocks(handle_t *handle, struct inode *inode,
 			   struct ext4_map_blocks *map, int flags);
-extern int ext4_ext_calc_metadata_amount(struct inode *inode,
-					 ext4_lblk_t lblocks);
 extern int ext4_ext_calc_credits_for_single_extent(struct inode *inode,
 						   int num,
 						   struct ext4_ext_path *path);
-extern int ext4_can_extents_be_merged(struct inode *inode,
-				      struct ext4_extent *ex1,
-				      struct ext4_extent *ex2);
 extern int ext4_ext_insert_extent(handle_t *, struct inode *,
 				  struct ext4_ext_path **,
 				  struct ext4_extent *, int);
@@ -3289,13 +3389,15 @@ extern int ext4_get_es_cache(struct inode *inode,
 			     struct fiemap_extent_info *fieinfo,
 			     __u64 start, __u64 len);
 extern int ext4_ext_precache(struct inode *inode);
-extern int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len);
-extern int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len);
 extern int ext4_swap_extents(handle_t *handle, struct inode *inode1,
 				struct inode *inode2, ext4_lblk_t lblk1,
 			     ext4_lblk_t lblk2,  ext4_lblk_t count,
 			     int mark_unwritten,int *err);
 extern int ext4_clu_mapped(struct inode *inode, ext4_lblk_t lclu);
+extern int ext4_datasem_ensure_credits(handle_t *handle, struct inode *inode,
+				       int check_cred, int restart_cred,
+				       int revoke_cred);
+
 
 /* move_extent.c */
 extern void ext4_double_down_write_data_sem(struct inode *first,
@@ -3322,6 +3424,8 @@ extern int ext4_bio_write_page(struct ext4_io_submit *io,
 			       int len,
 			       struct writeback_control *wbc,
 			       bool keep_towrite);
+extern struct ext4_io_end_vec *ext4_alloc_io_end_vec(ext4_io_end_t *io_end);
+extern struct ext4_io_end_vec *ext4_last_io_end_vec(ext4_io_end_t *io_end);
 
 /* mmp.c */
 extern int ext4_multi_mount_protect(struct super_block *, ext4_fsblk_t);
@@ -3379,6 +3483,8 @@ static inline void ext4_clear_io_unwritten_flag(ext4_io_end_t *io_end)
 }
 
 extern const struct iomap_ops ext4_iomap_ops;
+extern const struct iomap_ops ext4_iomap_overwrite_ops;
+extern const struct iomap_ops ext4_iomap_report_ops;
 
 static inline int ext4_buffer_uptodate(struct buffer_head *bh)
 {
diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h
index 98bd0e9ee7df..1c216fcc202a 100644
--- a/fs/ext4/ext4_extents.h
+++ b/fs/ext4/ext4_extents.h
@@ -267,10 +267,5 @@ static inline void ext4_idx_store_pblock(struct ext4_extent_idx *ix,
 				     0xffff);
 }
 
-#define ext4_ext_dirty(handle, inode, path) \
-		__ext4_ext_dirty(__func__, __LINE__, (handle), (inode), (path))
-int __ext4_ext_dirty(const char *where, unsigned int line, handle_t *handle,
-		     struct inode *inode, struct ext4_ext_path *path);
-
 #endif /* _EXT4_EXTENTS */
 
diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c
index 7c70b08d104c..0c76cdd44d90 100644
--- a/fs/ext4/ext4_jbd2.c
+++ b/fs/ext4/ext4_jbd2.c
@@ -7,6 +7,28 @@
 
 #include <trace/events/ext4.h>
 
+int ext4_inode_journal_mode(struct inode *inode)
+{
+	if (EXT4_JOURNAL(inode) == NULL)
+		return EXT4_INODE_WRITEBACK_DATA_MODE;	/* writeback */
+	/* We do not support data journalling with delayed allocation */
+	if (!S_ISREG(inode->i_mode) ||
+	    ext4_test_inode_flag(inode, EXT4_INODE_EA_INODE) ||
+	    test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ||
+	    (ext4_test_inode_flag(inode, EXT4_INODE_JOURNAL_DATA) &&
+	    !test_opt(inode->i_sb, DELALLOC))) {
+		/* We do not support data journalling for encrypted data */
+		if (S_ISREG(inode->i_mode) && IS_ENCRYPTED(inode))
+			return EXT4_INODE_ORDERED_DATA_MODE;  /* ordered */
+		return EXT4_INODE_JOURNAL_DATA_MODE;	/* journal data */
+	}
+	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA)
+		return EXT4_INODE_ORDERED_DATA_MODE;	/* ordered */
+	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)
+		return EXT4_INODE_WRITEBACK_DATA_MODE;	/* writeback */
+	BUG();
+}
+
 /* Just increment the non-pointer handle value */
 static handle_t *ext4_get_nojournal(void)
 {
@@ -58,19 +80,21 @@ static int ext4_journal_check_start(struct super_block *sb)
 	 * take the FS itself readonly cleanly.
 	 */
 	if (journal && is_journal_aborted(journal)) {
-		ext4_abort(sb, "Detected aborted journal");
+		ext4_abort(sb, -journal->j_errno, "Detected aborted journal");
 		return -EROFS;
 	}
 	return 0;
 }
 
 handle_t *__ext4_journal_start_sb(struct super_block *sb, unsigned int line,
-				  int type, int blocks, int rsv_blocks)
+				  int type, int blocks, int rsv_blocks,
+				  int revoke_creds)
 {
 	journal_t *journal;
 	int err;
 
-	trace_ext4_journal_start(sb, blocks, rsv_blocks, _RET_IP_);
+	trace_ext4_journal_start(sb, blocks, rsv_blocks, revoke_creds,
+				 _RET_IP_);
 	err = ext4_journal_check_start(sb);
 	if (err < 0)
 		return ERR_PTR(err);
@@ -78,8 +102,8 @@ handle_t *__ext4_journal_start_sb(struct super_block *sb, unsigned int line,
 	journal = EXT4_SB(sb)->s_journal;
 	if (!journal)
 		return ext4_get_nojournal();
-	return jbd2__journal_start(journal, blocks, rsv_blocks, GFP_NOFS,
-				   type, line);
+	return jbd2__journal_start(journal, blocks, rsv_blocks, revoke_creds,
+				   GFP_NOFS, type, line);
 }
 
 int __ext4_journal_stop(const char *where, unsigned int line, handle_t *handle)
@@ -119,8 +143,8 @@ handle_t *__ext4_journal_start_reserved(handle_t *handle, unsigned int line,
 		return ext4_get_nojournal();
 
 	sb = handle->h_journal->j_private;
-	trace_ext4_journal_start_reserved(sb, handle->h_buffer_credits,
-					  _RET_IP_);
+	trace_ext4_journal_start_reserved(sb,
+				jbd2_handle_buffer_credits(handle), _RET_IP_);
 	err = ext4_journal_check_start(sb);
 	if (err < 0) {
 		jbd2_journal_free_reserved(handle);
@@ -133,6 +157,19 @@ handle_t *__ext4_journal_start_reserved(handle_t *handle, unsigned int line,
 	return handle;
 }
 
+int __ext4_journal_ensure_credits(handle_t *handle, int check_cred,
+				  int extend_cred, int revoke_cred)
+{
+	if (!ext4_handle_valid(handle))
+		return 0;
+	if (jbd2_handle_buffer_credits(handle) >= check_cred &&
+	    handle->h_revoke_credits >= revoke_cred)
+		return 0;
+	extend_cred = max(0, extend_cred - jbd2_handle_buffer_credits(handle));
+	revoke_cred = max(0, revoke_cred - handle->h_revoke_credits);
+	return ext4_journal_extend(handle, extend_cred, revoke_cred);
+}
+
 static void ext4_journal_abort_handle(const char *caller, unsigned int line,
 				      const char *err_fn,
 				      struct buffer_head *bh,
@@ -234,7 +271,7 @@ int __ext4_forget(const char *where, unsigned int line, handle_t *handle,
 	if (err) {
 		ext4_journal_abort_handle(where, line, __func__,
 					  bh, handle, err);
-		__ext4_abort(inode->i_sb, where, line,
+		__ext4_abort(inode->i_sb, where, line, -err,
 			   "error %d when attempting revoke", err);
 	}
 	BUFFER_TRACE(bh, "exit");
@@ -278,7 +315,7 @@ int __ext4_handle_dirty_metadata(const char *where, unsigned int line,
 				       handle->h_type,
 				       handle->h_line_no,
 				       handle->h_requested_credits,
-				       handle->h_buffer_credits, err);
+				       jbd2_handle_buffer_credits(handle), err);
 				return err;
 			}
 			ext4_error_inode(inode, where, line,
@@ -289,9 +326,11 @@ int __ext4_handle_dirty_metadata(const char *where, unsigned int line,
 					 handle->h_type,
 					 handle->h_line_no,
 					 handle->h_requested_credits,
-					 handle->h_buffer_credits, err);
+					 jbd2_handle_buffer_credits(handle),
+					 err);
 		}
 	} else {
+		set_buffer_uptodate(bh);
 		if (inode)
 			mark_buffer_dirty_inode(bh, inode);
 		else
@@ -299,13 +338,8 @@ int __ext4_handle_dirty_metadata(const char *where, unsigned int line,
 		if (inode && inode_needs_sync(inode)) {
 			sync_dirty_buffer(bh);
 			if (buffer_req(bh) && !buffer_uptodate(bh)) {
-				struct ext4_super_block *es;
-
-				es = EXT4_SB(inode->i_sb)->s_es;
-				es->s_last_error_block =
-					cpu_to_le64(bh->b_blocknr);
-				ext4_error_inode(inode, where, line,
-						 bh->b_blocknr,
+				ext4_error_inode_err(inode, where, line,
+						     bh->b_blocknr, EIO,
 					"IO error syncing itable block");
 				err = -EIO;
 			}
diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h
index ef8fcf7d0d3b..4b9002f0e84c 100644
--- a/fs/ext4/ext4_jbd2.h
+++ b/fs/ext4/ext4_jbd2.h
@@ -261,7 +261,8 @@ int __ext4_handle_dirty_super(const char *where, unsigned int line,
 	__ext4_handle_dirty_super(__func__, __LINE__, (handle), (sb))
 
 handle_t *__ext4_journal_start_sb(struct super_block *sb, unsigned int line,
-				  int type, int blocks, int rsv_blocks);
+				  int type, int blocks, int rsv_blocks,
+				  int revoke_creds);
 int __ext4_journal_stop(const char *where, unsigned int line, handle_t *handle);
 
 #define EXT4_NOJOURNAL_MAX_REF_COUNT ((unsigned long) 4096)
@@ -288,28 +289,41 @@ static inline int ext4_handle_is_aborted(handle_t *handle)
 	return 0;
 }
 
-static inline int ext4_handle_has_enough_credits(handle_t *handle, int needed)
+static inline int ext4_free_metadata_revoke_credits(struct super_block *sb,
+						    int blocks)
 {
-	if (ext4_handle_valid(handle) && handle->h_buffer_credits < needed)
-		return 0;
-	return 1;
+	/* Freeing each metadata block can result in freeing one cluster */
+	return blocks * EXT4_SB(sb)->s_cluster_ratio;
+}
+
+static inline int ext4_trans_default_revoke_credits(struct super_block *sb)
+{
+	return ext4_free_metadata_revoke_credits(sb, 8);
 }
 
 #define ext4_journal_start_sb(sb, type, nblocks)			\
-	__ext4_journal_start_sb((sb), __LINE__, (type), (nblocks), 0)
+	__ext4_journal_start_sb((sb), __LINE__, (type), (nblocks), 0,	\
+				ext4_trans_default_revoke_credits(sb))
 
 #define ext4_journal_start(inode, type, nblocks)			\
-	__ext4_journal_start((inode), __LINE__, (type), (nblocks), 0)
+	__ext4_journal_start((inode), __LINE__, (type), (nblocks), 0,	\
+			     ext4_trans_default_revoke_credits((inode)->i_sb))
+
+#define ext4_journal_start_with_reserve(inode, type, blocks, rsv_blocks)\
+	__ext4_journal_start((inode), __LINE__, (type), (blocks), (rsv_blocks),\
+			     ext4_trans_default_revoke_credits((inode)->i_sb))
 
-#define ext4_journal_start_with_reserve(inode, type, blocks, rsv_blocks) \
-	__ext4_journal_start((inode), __LINE__, (type), (blocks), (rsv_blocks))
+#define ext4_journal_start_with_revoke(inode, type, blocks, revoke_creds) \
+	__ext4_journal_start((inode), __LINE__, (type), (blocks), 0,	\
+			     (revoke_creds))
 
 static inline handle_t *__ext4_journal_start(struct inode *inode,
 					     unsigned int line, int type,
-					     int blocks, int rsv_blocks)
+					     int blocks, int rsv_blocks,
+					     int revoke_creds)
 {
 	return __ext4_journal_start_sb(inode->i_sb, line, type, blocks,
-				       rsv_blocks);
+				       rsv_blocks, revoke_creds);
 }
 
 #define ext4_journal_stop(handle) \
@@ -332,20 +346,68 @@ static inline handle_t *ext4_journal_current_handle(void)
 	return journal_current_handle();
 }
 
-static inline int ext4_journal_extend(handle_t *handle, int nblocks)
+static inline int ext4_journal_extend(handle_t *handle, int nblocks, int revoke)
 {
 	if (ext4_handle_valid(handle))
-		return jbd2_journal_extend(handle, nblocks);
+		return jbd2_journal_extend(handle, nblocks, revoke);
 	return 0;
 }
 
-static inline int ext4_journal_restart(handle_t *handle, int nblocks)
+static inline int ext4_journal_restart(handle_t *handle, int nblocks,
+				       int revoke)
 {
 	if (ext4_handle_valid(handle))
-		return jbd2_journal_restart(handle, nblocks);
+		return jbd2__journal_restart(handle, nblocks, revoke, GFP_NOFS);
 	return 0;
 }
 
+int __ext4_journal_ensure_credits(handle_t *handle, int check_cred,
+				  int extend_cred, int revoke_cred);
+
+
+/*
+ * Ensure @handle has at least @check_creds credits available. If not,
+ * transaction will be extended or restarted to contain at least @extend_cred
+ * credits. Before restarting transaction @fn is executed to allow for cleanup
+ * before the transaction is restarted.
+ *
+ * The return value is < 0 in case of error, 0 in case the handle has enough
+ * credits or transaction extension succeeded, 1 in case transaction had to be
+ * restarted.
+ */
+#define ext4_journal_ensure_credits_fn(handle, check_cred, extend_cred,	\
+				       revoke_cred, fn) \
+({									\
+	__label__ __ensure_end;						\
+	int err = __ext4_journal_ensure_credits((handle), (check_cred),	\
+					(extend_cred), (revoke_cred));	\
+									\
+	if (err <= 0)							\
+		goto __ensure_end;					\
+	err = (fn);							\
+	if (err < 0)							\
+		goto __ensure_end;					\
+	err = ext4_journal_restart((handle), (extend_cred), (revoke_cred)); \
+	if (err == 0)							\
+		err = 1;						\
+__ensure_end:								\
+	err;								\
+})
+
+/*
+ * Ensure given handle has at least requested amount of credits available,
+ * possibly restarting transaction if needed. We also make sure the transaction
+ * has space for at least ext4_trans_default_revoke_credits(sb) revoke records
+ * as freeing one or two blocks is very common pattern and requesting this is
+ * very cheap.
+ */
+static inline int ext4_journal_ensure_credits(handle_t *handle, int credits,
+					      int revoke_creds)
+{
+	return ext4_journal_ensure_credits_fn(handle, credits, credits,
+				revoke_creds, 0);
+}
+
 static inline int ext4_journal_blocks_per_page(struct inode *inode)
 {
 	if (EXT4_JOURNAL(inode) != NULL)
@@ -401,26 +463,7 @@ int ext4_force_commit(struct super_block *sb);
 #define EXT4_INODE_ORDERED_DATA_MODE	0x02 /* ordered data mode */
 #define EXT4_INODE_WRITEBACK_DATA_MODE	0x04 /* writeback data mode */
 
-static inline int ext4_inode_journal_mode(struct inode *inode)
-{
-	if (EXT4_JOURNAL(inode) == NULL)
-		return EXT4_INODE_WRITEBACK_DATA_MODE;	/* writeback */
-	/* We do not support data journalling with delayed allocation */
-	if (!S_ISREG(inode->i_mode) ||
-	    test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ||
-	    (ext4_test_inode_flag(inode, EXT4_INODE_JOURNAL_DATA) &&
-	    !test_opt(inode->i_sb, DELALLOC))) {
-		/* We do not support data journalling for encrypted data */
-		if (S_ISREG(inode->i_mode) && IS_ENCRYPTED(inode))
-			return EXT4_INODE_ORDERED_DATA_MODE;  /* ordered */
-		return EXT4_INODE_JOURNAL_DATA_MODE;	/* journal data */
-	}
-	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA)
-		return EXT4_INODE_ORDERED_DATA_MODE;	/* ordered */
-	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)
-		return EXT4_INODE_WRITEBACK_DATA_MODE;	/* writeback */
-	BUG();
-}
+int ext4_inode_journal_mode(struct inode *inode);
 
 static inline int ext4_should_journal_data(struct inode *inode)
 {
@@ -437,6 +480,19 @@ static inline int ext4_should_writeback_data(struct inode *inode)
 	return ext4_inode_journal_mode(inode) & EXT4_INODE_WRITEBACK_DATA_MODE;
 }
 
+static inline int ext4_free_data_revoke_credits(struct inode *inode, int blocks)
+{
+	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA)
+		return 0;
+	if (!ext4_should_journal_data(inode))
+		return 0;
+	/*
+	 * Data blocks in one extent are contiguous, just account for partial
+	 * clusters at extent boundaries
+	 */
+	return blocks + 2*(EXT4_SB(inode->i_sb)->s_cluster_ratio - 1);
+}
+
 /*
  * This function controls whether or not we should try to go down the
  * dioread_nolock code paths, which makes it safe to avoid taking
@@ -456,6 +512,9 @@ static inline int ext4_should_dioread_nolock(struct inode *inode)
 		return 0;
 	if (ext4_should_journal_data(inode))
 		return 0;
+	/* temporary fix to prevent generic/422 test failures */
+	if (!test_opt(inode->i_sb, DELALLOC))
+		return 0;
 	return 1;
 }
 
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index fb0f99dc8c22..2b4b94542e34 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -28,6 +28,7 @@
 #include <linux/uaccess.h>
 #include <linux/fiemap.h>
 #include <linux/backing-dev.h>
+#include <linux/iomap.h>
 #include "ext4_jbd2.h"
 #include "ext4_extents.h"
 #include "xattr.h"
@@ -83,13 +84,6 @@ static void ext4_extent_block_csum_set(struct inode *inode,
 	et->et_checksum = ext4_extent_block_csum(inode, eh);
 }
 
-static int ext4_split_extent(handle_t *handle,
-				struct inode *inode,
-				struct ext4_ext_path **ppath,
-				struct ext4_map_blocks *map,
-				int split_flag,
-				int flags);
-
 static int ext4_split_extent_at(handle_t *handle,
 			     struct inode *inode,
 			     struct ext4_ext_path **ppath,
@@ -97,32 +91,41 @@ static int ext4_split_extent_at(handle_t *handle,
 			     int split_flag,
 			     int flags);
 
-static int ext4_find_delayed_extent(struct inode *inode,
-				    struct extent_status *newes);
-
-static int ext4_ext_truncate_extend_restart(handle_t *handle,
-					    struct inode *inode,
-					    int needed)
+static int ext4_ext_trunc_restart_fn(struct inode *inode, int *dropped)
 {
-	int err;
-
-	if (!ext4_handle_valid(handle))
-		return 0;
-	if (handle->h_buffer_credits >= needed)
-		return 0;
 	/*
-	 * If we need to extend the journal get a few extra blocks
-	 * while we're at it for efficiency's sake.
+	 * Drop i_data_sem to avoid deadlock with ext4_map_blocks.  At this
+	 * moment, get_block can be called only for blocks inside i_size since
+	 * page cache has been already dropped and writes are blocked by
+	 * i_mutex. So we can safely drop the i_data_sem here.
 	 */
-	needed += 3;
-	err = ext4_journal_extend(handle, needed - handle->h_buffer_credits);
-	if (err <= 0)
-		return err;
-	err = ext4_truncate_restart_trans(handle, inode, needed);
-	if (err == 0)
-		err = -EAGAIN;
+	BUG_ON(EXT4_JOURNAL(inode) == NULL);
+	ext4_discard_preallocations(inode);
+	up_write(&EXT4_I(inode)->i_data_sem);
+	*dropped = 1;
+	return 0;
+}
 
-	return err;
+/*
+ * Make sure 'handle' has at least 'check_cred' credits. If not, restart
+ * transaction with 'restart_cred' credits. The function drops i_data_sem
+ * when restarting transaction and gets it after transaction is restarted.
+ *
+ * The function returns 0 on success, 1 if transaction had to be restarted,
+ * and < 0 in case of fatal error.
+ */
+int ext4_datasem_ensure_credits(handle_t *handle, struct inode *inode,
+				int check_cred, int restart_cred,
+				int revoke_cred)
+{
+	int ret;
+	int dropped = 0;
+
+	ret = ext4_journal_ensure_credits_fn(handle, check_cred, restart_cred,
+		revoke_cred, ext4_ext_trunc_restart_fn(inode, &dropped));
+	if (dropped)
+		down_write(&EXT4_I(inode)->i_data_sem);
+	return ret;
 }
 
 /*
@@ -149,8 +152,9 @@ static int ext4_ext_get_access(handle_t *handle, struct inode *inode,
  *  - ENOMEM
  *  - EIO
  */
-int __ext4_ext_dirty(const char *where, unsigned int line, handle_t *handle,
-		     struct inode *inode, struct ext4_ext_path *path)
+static int __ext4_ext_dirty(const char *where, unsigned int line,
+			    handle_t *handle, struct inode *inode,
+			    struct ext4_ext_path *path)
 {
 	int err;
 
@@ -167,6 +171,9 @@ int __ext4_ext_dirty(const char *where, unsigned int line, handle_t *handle,
 	return err;
 }
 
+#define ext4_ext_dirty(handle, inode, path) \
+		__ext4_ext_dirty(__func__, __LINE__, (handle), (inode), (path))
+
 static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode,
 			      struct ext4_ext_path *path,
 			      ext4_lblk_t block)
@@ -297,53 +304,6 @@ ext4_force_split_extent_at(handle_t *handle, struct inode *inode,
 			(nofail ? EXT4_GET_BLOCKS_METADATA_NOFAIL:0));
 }
 
-/*
- * Calculate the number of metadata blocks needed
- * to allocate @blocks
- * Worse case is one block per extent
- */
-int ext4_ext_calc_metadata_amount(struct inode *inode, ext4_lblk_t lblock)
-{
-	struct ext4_inode_info *ei = EXT4_I(inode);
-	int idxs;
-
-	idxs = ((inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header))
-		/ sizeof(struct ext4_extent_idx));
-
-	/*
-	 * If the new delayed allocation block is contiguous with the
-	 * previous da block, it can share index blocks with the
-	 * previous block, so we only need to allocate a new index
-	 * block every idxs leaf blocks.  At ldxs**2 blocks, we need
-	 * an additional index block, and at ldxs**3 blocks, yet
-	 * another index blocks.
-	 */
-	if (ei->i_da_metadata_calc_len &&
-	    ei->i_da_metadata_calc_last_lblock+1 == lblock) {
-		int num = 0;
-
-		if ((ei->i_da_metadata_calc_len % idxs) == 0)
-			num++;
-		if ((ei->i_da_metadata_calc_len % (idxs*idxs)) == 0)
-			num++;
-		if ((ei->i_da_metadata_calc_len % (idxs*idxs*idxs)) == 0) {
-			num++;
-			ei->i_da_metadata_calc_len = 0;
-		} else
-			ei->i_da_metadata_calc_len++;
-		ei->i_da_metadata_calc_last_lblock++;
-		return num;
-	}
-
-	/*
-	 * In the worst case we need a new set of index blocks at
-	 * every level of the inode's extent tree.
-	 */
-	ei->i_da_metadata_calc_len = 1;
-	ei->i_da_metadata_calc_last_lblock = lblock;
-	return ext_depth(inode) + 1;
-}
-
 static int
 ext4_ext_max_entries(struct inode *inode, int depth)
 {
@@ -389,8 +349,8 @@ static int ext4_valid_extent_idx(struct inode *inode,
 }
 
 static int ext4_valid_extent_entries(struct inode *inode,
-				struct ext4_extent_header *eh,
-				int depth)
+				     struct ext4_extent_header *eh,
+				     ext4_fsblk_t *pblk, int depth)
 {
 	unsigned short entries;
 	if (eh->eh_entries == 0)
@@ -401,8 +361,6 @@ static int ext4_valid_extent_entries(struct inode *inode,
 	if (depth == 0) {
 		/* leaf entries */
 		struct ext4_extent *ext = EXT_FIRST_EXTENT(eh);
-		struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
-		ext4_fsblk_t pblock = 0;
 		ext4_lblk_t lblock = 0;
 		ext4_lblk_t prev = 0;
 		int len = 0;
@@ -414,8 +372,7 @@ static int ext4_valid_extent_entries(struct inode *inode,
 			lblock = le32_to_cpu(ext->ee_block);
 			len = ext4_ext_get_actual_len(ext);
 			if ((lblock <= prev) && prev) {
-				pblock = ext4_ext_pblock(ext);
-				es->s_last_error_block = cpu_to_le64(pblock);
+				*pblk = ext4_ext_pblock(ext);
 				return 0;
 			}
 			ext++;
@@ -462,7 +419,7 @@ static int __ext4_ext_check(const char *function, unsigned int line,
 		error_msg = "invalid eh_entries";
 		goto corrupted;
 	}
-	if (!ext4_valid_extent_entries(inode, eh, depth)) {
+	if (!ext4_valid_extent_entries(inode, eh, &pblk, depth)) {
 		error_msg = "invalid extent entries";
 		goto corrupted;
 	}
@@ -480,13 +437,14 @@ static int __ext4_ext_check(const char *function, unsigned int line,
 	return 0;
 
 corrupted:
-	ext4_error_inode(inode, function, line, 0,
-			 "pblk %llu bad header/extent: %s - magic %x, "
-			 "entries %u, max %u(%u), depth %u(%u)",
-			 (unsigned long long) pblk, error_msg,
-			 le16_to_cpu(eh->eh_magic),
-			 le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max),
-			 max, le16_to_cpu(eh->eh_depth), depth);
+	ext4_error_inode_err(inode, function, line, 0, -err,
+			     "pblk %llu bad header/extent: %s - magic %x, "
+			     "entries %u, max %u(%u), depth %u(%u)",
+			     (unsigned long long) pblk, error_msg,
+			     le16_to_cpu(eh->eh_magic),
+			     le16_to_cpu(eh->eh_entries),
+			     le16_to_cpu(eh->eh_max),
+			     max, le16_to_cpu(eh->eh_depth), depth);
 	return err;
 }
 
@@ -498,6 +456,30 @@ int ext4_ext_check_inode(struct inode *inode)
 	return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode), 0);
 }
 
+static void ext4_cache_extents(struct inode *inode,
+			       struct ext4_extent_header *eh)
+{
+	struct ext4_extent *ex = EXT_FIRST_EXTENT(eh);
+	ext4_lblk_t prev = 0;
+	int i;
+
+	for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) {
+		unsigned int status = EXTENT_STATUS_WRITTEN;
+		ext4_lblk_t lblk = le32_to_cpu(ex->ee_block);
+		int len = ext4_ext_get_actual_len(ex);
+
+		if (prev && (prev != lblk))
+			ext4_es_cache_extent(inode, prev, lblk - prev, ~0,
+					     EXTENT_STATUS_HOLE);
+
+		if (ext4_ext_is_unwritten(ex))
+			status = EXTENT_STATUS_UNWRITTEN;
+		ext4_es_cache_extent(inode, lblk, len,
+				     ext4_ext_pblock(ex), status);
+		prev = lblk + len;
+	}
+}
+
 static struct buffer_head *
 __read_extent_tree_block(const char *function, unsigned int line,
 			 struct inode *inode, ext4_fsblk_t pblk, int depth,
@@ -532,26 +514,7 @@ __read_extent_tree_block(const char *function, unsigned int line,
 	 */
 	if (!(flags & EXT4_EX_NOCACHE) && depth == 0) {
 		struct ext4_extent_header *eh = ext_block_hdr(bh);
-		struct ext4_extent *ex = EXT_FIRST_EXTENT(eh);
-		ext4_lblk_t prev = 0;
-		int i;
-
-		for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) {
-			unsigned int status = EXTENT_STATUS_WRITTEN;
-			ext4_lblk_t lblk = le32_to_cpu(ex->ee_block);
-			int len = ext4_ext_get_actual_len(ex);
-
-			if (prev && (prev != lblk))
-				ext4_es_cache_extent(inode, prev,
-						     lblk - prev, ~0,
-						     EXTENT_STATUS_HOLE);
-
-			if (ext4_ext_is_unwritten(ex))
-				status = EXTENT_STATUS_UNWRITTEN;
-			ext4_es_cache_extent(inode, lblk, len,
-					     ext4_ext_pblock(ex), status);
-			prev = lblk + len;
-		}
+		ext4_cache_extents(inode, eh);
 	}
 	return bh;
 errout:
@@ -581,6 +544,12 @@ int ext4_ext_precache(struct inode *inode)
 	down_read(&ei->i_data_sem);
 	depth = ext_depth(inode);
 
+	/* Don't cache anything if there are no external extent blocks */
+	if (!depth) {
+		up_read(&ei->i_data_sem);
+		return ret;
+	}
+
 	path = kcalloc(depth + 1, sizeof(struct ext4_ext_path),
 		       GFP_NOFS);
 	if (path == NULL) {
@@ -588,9 +557,6 @@ int ext4_ext_precache(struct inode *inode)
 		return -ENOMEM;
 	}
 
-	/* Don't cache anything if there are no external extent blocks */
-	if (depth == 0)
-		goto out;
 	path[0].p_hdr = ext_inode_hdr(inode);
 	ret = ext4_ext_check(inode, path[0].p_hdr, depth, 0);
 	if (ret)
@@ -637,8 +603,9 @@ static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path)
 	ext_debug("path:");
 	for (k = 0; k <= l; k++, path++) {
 		if (path->p_idx) {
-		  ext_debug("  %d->%llu", le32_to_cpu(path->p_idx->ei_block),
-			    ext4_idx_pblock(path->p_idx));
+			ext_debug("  %d->%llu",
+				  le32_to_cpu(path->p_idx->ei_block),
+				  ext4_idx_pblock(path->p_idx));
 		} else if (path->p_ext) {
 			ext_debug("  %d:[%d]%d:%llu ",
 				  le32_to_cpu(path->p_ext->ee_block),
@@ -719,11 +686,12 @@ void ext4_ext_drop_refs(struct ext4_ext_path *path)
 	if (!path)
 		return;
 	depth = path->p_depth;
-	for (i = 0; i <= depth; i++, path++)
+	for (i = 0; i <= depth; i++, path++) {
 		if (path->p_bh) {
 			brelse(path->p_bh);
 			path->p_bh = NULL;
 		}
+	}
 }
 
 /*
@@ -765,8 +733,8 @@ ext4_ext_binsearch_idx(struct inode *inode,
 
 		chix = ix = EXT_FIRST_INDEX(eh);
 		for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) {
-		  if (k != 0 &&
-		      le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) {
+			if (k != 0 && le32_to_cpu(ix->ei_block) <=
+			    le32_to_cpu(ix[-1].ei_block)) {
 				printk(KERN_DEBUG "k=%d, ix=0x%p, "
 				       "first=0x%p\n", k,
 				       ix, EXT_FIRST_INDEX(eh));
@@ -899,6 +867,8 @@ ext4_find_extent(struct inode *inode, ext4_lblk_t block,
 	path[0].p_bh = NULL;
 
 	i = depth;
+	if (!(flags & EXT4_EX_NOCACHE) && depth == 0)
+		ext4_cache_extents(inode, eh);
 	/* walk through the tree */
 	while (i) {
 		ext_debug("depth %d: num %d, max %d\n",
@@ -1620,17 +1590,16 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path)
 		return EXT_MAX_BLOCKS;
 
 	while (depth >= 0) {
+		struct ext4_ext_path *p = &path[depth];
+
 		if (depth == path->p_depth) {
 			/* leaf */
-			if (path[depth].p_ext &&
-				path[depth].p_ext !=
-					EXT_LAST_EXTENT(path[depth].p_hdr))
-			  return le32_to_cpu(path[depth].p_ext[1].ee_block);
+			if (p->p_ext && p->p_ext != EXT_LAST_EXTENT(p->p_hdr))
+				return le32_to_cpu(p->p_ext[1].ee_block);
 		} else {
 			/* index */
-			if (path[depth].p_idx !=
-					EXT_LAST_INDEX(path[depth].p_hdr))
-			  return le32_to_cpu(path[depth].p_idx[1].ei_block);
+			if (p->p_idx != EXT_LAST_INDEX(p->p_hdr))
+				return le32_to_cpu(p->p_idx[1].ei_block);
 		}
 		depth--;
 	}
@@ -1730,9 +1699,9 @@ static int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode,
 	return err;
 }
 
-int
-ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
-				struct ext4_extent *ex2)
+static int ext4_can_extents_be_merged(struct inode *inode,
+				      struct ext4_extent *ex1,
+				      struct ext4_extent *ex2)
 {
 	unsigned short ext1_ee_len, ext2_ee_len;
 
@@ -1746,23 +1715,11 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
 			le32_to_cpu(ex2->ee_block))
 		return 0;
 
-	/*
-	 * To allow future support for preallocated extents to be added
-	 * as an RO_COMPAT feature, refuse to merge to extents if
-	 * this can result in the top bit of ee_len being set.
-	 */
 	if (ext1_ee_len + ext2_ee_len > EXT_INIT_MAX_LEN)
 		return 0;
-	/*
-	 * The check for IO to unwritten extent is somewhat racy as we
-	 * increment i_unwritten / set EXT4_STATE_DIO_UNWRITTEN only after
-	 * dropping i_data_sem. But reserved blocks should save us in that
-	 * case.
-	 */
+
 	if (ext4_ext_is_unwritten(ex1) &&
-	    (ext4_test_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN) ||
-	     atomic_read(&EXT4_I(inode)->i_unwritten) ||
-	     (ext1_ee_len + ext2_ee_len > EXT_UNWRITTEN_MAX_LEN)))
+	    ext1_ee_len + ext2_ee_len > EXT_UNWRITTEN_MAX_LEN)
 		return 0;
 #ifdef AGGRESSIVE_TEST
 	if (ext1_ee_len >= 4)
@@ -1840,7 +1797,8 @@ static void ext4_ext_try_to_merge_up(handle_t *handle,
 	 * group descriptor to release the extent tree block.  If we
 	 * can't get the journal credits, give up.
 	 */
-	if (ext4_journal_extend(handle, 2))
+	if (ext4_journal_extend(handle, 2,
+			ext4_free_metadata_revoke_credits(inode->i_sb, 1)))
 		return;
 
 	/*
@@ -1864,13 +1822,14 @@ static void ext4_ext_try_to_merge_up(handle_t *handle,
 }
 
 /*
- * This function tries to merge the @ex extent to neighbours in the tree.
- * return 1 if merge left else 0.
+ * This function tries to merge the @ex extent to neighbours in the tree, then
+ * tries to collapse the extent tree into the inode.
  */
 static void ext4_ext_try_to_merge(handle_t *handle,
 				  struct inode *inode,
 				  struct ext4_ext_path *path,
-				  struct ext4_extent *ex) {
+				  struct ext4_extent *ex)
+{
 	struct ext4_extent_header *eh;
 	unsigned int depth;
 	int merge_done = 0;
@@ -2166,155 +2125,6 @@ cleanup:
 	return err;
 }
 
-static int ext4_fill_fiemap_extents(struct inode *inode,
-				    ext4_lblk_t block, ext4_lblk_t num,
-				    struct fiemap_extent_info *fieinfo)
-{
-	struct ext4_ext_path *path = NULL;
-	struct ext4_extent *ex;
-	struct extent_status es;
-	ext4_lblk_t next, next_del, start = 0, end = 0;
-	ext4_lblk_t last = block + num;
-	int exists, depth = 0, err = 0;
-	unsigned int flags = 0;
-	unsigned char blksize_bits = inode->i_sb->s_blocksize_bits;
-
-	while (block < last && block != EXT_MAX_BLOCKS) {
-		num = last - block;
-		/* find extent for this block */
-		down_read(&EXT4_I(inode)->i_data_sem);
-
-		path = ext4_find_extent(inode, block, &path, 0);
-		if (IS_ERR(path)) {
-			up_read(&EXT4_I(inode)->i_data_sem);
-			err = PTR_ERR(path);
-			path = NULL;
-			break;
-		}
-
-		depth = ext_depth(inode);
-		if (unlikely(path[depth].p_hdr == NULL)) {
-			up_read(&EXT4_I(inode)->i_data_sem);
-			EXT4_ERROR_INODE(inode, "path[%d].p_hdr == NULL", depth);
-			err = -EFSCORRUPTED;
-			break;
-		}
-		ex = path[depth].p_ext;
-		next = ext4_ext_next_allocated_block(path);
-
-		flags = 0;
-		exists = 0;
-		if (!ex) {
-			/* there is no extent yet, so try to allocate
-			 * all requested space */
-			start = block;
-			end = block + num;
-		} else if (le32_to_cpu(ex->ee_block) > block) {
-			/* need to allocate space before found extent */
-			start = block;
-			end = le32_to_cpu(ex->ee_block);
-			if (block + num < end)
-				end = block + num;
-		} else if (block >= le32_to_cpu(ex->ee_block)
-					+ ext4_ext_get_actual_len(ex)) {
-			/* need to allocate space after found extent */
-			start = block;
-			end = block + num;
-			if (end >= next)
-				end = next;
-		} else if (block >= le32_to_cpu(ex->ee_block)) {
-			/*
-			 * some part of requested space is covered
-			 * by found extent
-			 */
-			start = block;
-			end = le32_to_cpu(ex->ee_block)
-				+ ext4_ext_get_actual_len(ex);
-			if (block + num < end)
-				end = block + num;
-			exists = 1;
-		} else {
-			BUG();
-		}
-		BUG_ON(end <= start);
-
-		if (!exists) {
-			es.es_lblk = start;
-			es.es_len = end - start;
-			es.es_pblk = 0;
-		} else {
-			es.es_lblk = le32_to_cpu(ex->ee_block);
-			es.es_len = ext4_ext_get_actual_len(ex);
-			es.es_pblk = ext4_ext_pblock(ex);
-			if (ext4_ext_is_unwritten(ex))
-				flags |= FIEMAP_EXTENT_UNWRITTEN;
-		}
-
-		/*
-		 * Find delayed extent and update es accordingly. We call
-		 * it even in !exists case to find out whether es is the
-		 * last existing extent or not.
-		 */
-		next_del = ext4_find_delayed_extent(inode, &es);
-		if (!exists && next_del) {
-			exists = 1;
-			flags |= (FIEMAP_EXTENT_DELALLOC |
-				  FIEMAP_EXTENT_UNKNOWN);
-		}
-		up_read(&EXT4_I(inode)->i_data_sem);
-
-		if (unlikely(es.es_len == 0)) {
-			EXT4_ERROR_INODE(inode, "es.es_len == 0");
-			err = -EFSCORRUPTED;
-			break;
-		}
-
-		/*
-		 * This is possible iff next == next_del == EXT_MAX_BLOCKS.
-		 * we need to check next == EXT_MAX_BLOCKS because it is
-		 * possible that an extent is with unwritten and delayed
-		 * status due to when an extent is delayed allocated and
-		 * is allocated by fallocate status tree will track both of
-		 * them in a extent.
-		 *
-		 * So we could return a unwritten and delayed extent, and
-		 * its block is equal to 'next'.
-		 */
-		if (next == next_del && next == EXT_MAX_BLOCKS) {
-			flags |= FIEMAP_EXTENT_LAST;
-			if (unlikely(next_del != EXT_MAX_BLOCKS ||
-				     next != EXT_MAX_BLOCKS)) {
-				EXT4_ERROR_INODE(inode,
-						 "next extent == %u, next "
-						 "delalloc extent = %u",
-						 next, next_del);
-				err = -EFSCORRUPTED;
-				break;
-			}
-		}
-
-		if (exists) {
-			err = fiemap_fill_next_extent(fieinfo,
-				(__u64)es.es_lblk << blksize_bits,
-				(__u64)es.es_pblk << blksize_bits,
-				(__u64)es.es_len << blksize_bits,
-				flags);
-			if (err < 0)
-				break;
-			if (err == 1) {
-				err = 0;
-				break;
-			}
-		}
-
-		block = es.es_lblk + es.es_len;
-	}
-
-	ext4_ext_drop_refs(path);
-	kfree(path);
-	return err;
-}
-
 static int ext4_fill_es_cache_info(struct inode *inode,
 				   ext4_lblk_t block, ext4_lblk_t num,
 				   struct fiemap_extent_info *fieinfo)
@@ -2727,7 +2537,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
 {
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 	int err = 0, correct_index = 0;
-	int depth = ext_depth(inode), credits;
+	int depth = ext_depth(inode), credits, revoke_credits;
 	struct ext4_extent_header *eh;
 	ext4_lblk_t a, b;
 	unsigned num;
@@ -2819,10 +2629,23 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
 			credits += (ext_depth(inode)) + 1;
 		}
 		credits += EXT4_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb);
-
-		err = ext4_ext_truncate_extend_restart(handle, inode, credits);
-		if (err)
+		/*
+		 * We may end up freeing some index blocks and data from the
+		 * punched range. Note that partial clusters are accounted for
+		 * by ext4_free_data_revoke_credits().
+		 */
+		revoke_credits =
+			ext4_free_metadata_revoke_credits(inode->i_sb,
+							  ext_depth(inode)) +
+			ext4_free_data_revoke_credits(inode, b - a + 1);
+
+		err = ext4_datasem_ensure_credits(handle, inode, credits,
+						  credits, revoke_credits);
+		if (err) {
+			if (err > 0)
+				err = -EAGAIN;
 			goto out;
+		}
 
 		err = ext4_ext_get_access(handle, inode, path + depth);
 		if (err)
@@ -2948,7 +2771,9 @@ int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start,
 	ext_debug("truncate since %u to %u\n", start, end);
 
 	/* probably first extent we're gonna free will be last in block */
-	handle = ext4_journal_start(inode, EXT4_HT_TRUNCATE, depth + 1);
+	handle = ext4_journal_start_with_revoke(inode, EXT4_HT_TRUNCATE,
+			depth + 1,
+			ext4_free_metadata_revoke_credits(inode->i_sb, depth));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -3549,8 +3374,8 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 		(unsigned long long)map->m_lblk, map_len);
 
 	sbi = EXT4_SB(inode->i_sb);
-	eof_block = (inode->i_size + inode->i_sb->s_blocksize - 1) >>
-		inode->i_sb->s_blocksize_bits;
+	eof_block = (EXT4_I(inode)->i_disksize + inode->i_sb->s_blocksize - 1)
+			>> inode->i_sb->s_blocksize_bits;
 	if (eof_block < map->m_lblk + map_len)
 		eof_block = map->m_lblk + map_len;
 
@@ -3697,9 +3522,6 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 		max_zeroout = sbi->s_extent_max_zeroout_kb >>
 			(inode->i_sb->s_blocksize_bits - 10);
 
-	if (IS_ENCRYPTED(inode))
-		max_zeroout = 0;
-
 	/*
 	 * five cases:
 	 * 1. split the extent into three extents.
@@ -3805,8 +3627,8 @@ static int ext4_split_convert_extents(handle_t *handle,
 		  __func__, inode->i_ino,
 		  (unsigned long long)map->m_lblk, map->m_len);
 
-	eof_block = (inode->i_size + inode->i_sb->s_blocksize - 1) >>
-		inode->i_sb->s_blocksize_bits;
+	eof_block = (EXT4_I(inode)->i_disksize + inode->i_sb->s_blocksize - 1)
+			>> inode->i_sb->s_blocksize_bits;
 	if (eof_block < map->m_lblk + map->m_len)
 		eof_block = map->m_lblk + map->m_len;
 	/*
@@ -3894,64 +3716,11 @@ out:
 	return err;
 }
 
-/*
- * Handle EOFBLOCKS_FL flag, clearing it if necessary
- */
-static int check_eofblocks_fl(handle_t *handle, struct inode *inode,
-			      ext4_lblk_t lblk,
-			      struct ext4_ext_path *path,
-			      unsigned int len)
-{
-	int i, depth;
-	struct ext4_extent_header *eh;
-	struct ext4_extent *last_ex;
-
-	if (!ext4_test_inode_flag(inode, EXT4_INODE_EOFBLOCKS))
-		return 0;
-
-	depth = ext_depth(inode);
-	eh = path[depth].p_hdr;
-
-	/*
-	 * We're going to remove EOFBLOCKS_FL entirely in future so we
-	 * do not care for this case anymore. Simply remove the flag
-	 * if there are no extents.
-	 */
-	if (unlikely(!eh->eh_entries))
-		goto out;
-	last_ex = EXT_LAST_EXTENT(eh);
-	/*
-	 * We should clear the EOFBLOCKS_FL flag if we are writing the
-	 * last block in the last extent in the file.  We test this by
-	 * first checking to see if the caller to
-	 * ext4_ext_get_blocks() was interested in the last block (or
-	 * a block beyond the last block) in the current extent.  If
-	 * this turns out to be false, we can bail out from this
-	 * function immediately.
-	 */
-	if (lblk + len < le32_to_cpu(last_ex->ee_block) +
-	    ext4_ext_get_actual_len(last_ex))
-		return 0;
-	/*
-	 * If the caller does appear to be planning to write at or
-	 * beyond the end of the current extent, we then test to see
-	 * if the current extent is the last extent in the file, by
-	 * checking to make sure it was reached via the rightmost node
-	 * at each level of the tree.
-	 */
-	for (i = depth-1; i >= 0; i--)
-		if (path[i].p_idx != EXT_LAST_INDEX(path[i].p_hdr))
-			return 0;
-out:
-	ext4_clear_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
-	return ext4_mark_inode_dirty(handle, inode);
-}
-
 static int
 convert_initialized_extent(handle_t *handle, struct inode *inode,
 			   struct ext4_map_blocks *map,
 			   struct ext4_ext_path **ppath,
-			   unsigned int allocated)
+			   unsigned int *allocated)
 {
 	struct ext4_ext_path *path = *ppath;
 	struct ext4_extent *ex;
@@ -4011,14 +3780,12 @@ convert_initialized_extent(handle_t *handle, struct inode *inode,
 	ext4_ext_show_leaf(inode, path);
 
 	ext4_update_inode_fsync_trans(handle, inode, 1);
-	err = check_eofblocks_fl(handle, inode, map->m_lblk, path, map->m_len);
-	if (err)
-		return err;
+
 	map->m_flags |= EXT4_MAP_UNWRITTEN;
-	if (allocated > map->m_len)
-		allocated = map->m_len;
-	map->m_len = allocated;
-	return allocated;
+	if (*allocated > map->m_len)
+		*allocated = map->m_len;
+	map->m_len = *allocated;
+	return 0;
 }
 
 static int
@@ -4027,7 +3794,9 @@ ext4_ext_handle_unwritten_extents(handle_t *handle, struct inode *inode,
 			struct ext4_ext_path **ppath, int flags,
 			unsigned int allocated, ext4_fsblk_t newblock)
 {
+#ifdef EXT_DEBUG
 	struct ext4_ext_path *path = *ppath;
+#endif
 	int ret = 0;
 	int err = 0;
 
@@ -4067,11 +3836,9 @@ ext4_ext_handle_unwritten_extents(handle_t *handle, struct inode *inode,
 		}
 		ret = ext4_convert_unwritten_extents_endio(handle, inode, map,
 							   ppath);
-		if (ret >= 0) {
+		if (ret >= 0)
 			ext4_update_inode_fsync_trans(handle, inode, 1);
-			err = check_eofblocks_fl(handle, inode, map->m_lblk,
-						 path, map->m_len);
-		} else
+		else
 			err = ret;
 		map->m_flags |= EXT4_MAP_MAPPED;
 		map->m_pblk = newblock;
@@ -4120,12 +3887,6 @@ out:
 
 map_out:
 	map->m_flags |= EXT4_MAP_MAPPED;
-	if ((flags & EXT4_GET_BLOCKS_KEEP_SIZE) == 0) {
-		err = check_eofblocks_fl(handle, inode, map->m_lblk, path,
-					 map->m_len);
-		if (err < 0)
-			goto out2;
-	}
 out1:
 	if (allocated > map->m_len)
 		allocated = map->m_len;
@@ -4264,12 +4025,11 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 	struct ext4_extent newex, *ex, *ex2;
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 	ext4_fsblk_t newblock = 0;
-	int free_on_err = 0, err = 0, depth, ret;
+	int err = 0, depth, ret;
 	unsigned int allocated = 0, offset = 0;
 	unsigned int allocated_clusters = 0;
 	struct ext4_allocation_request ar;
 	ext4_lblk_t cluster_offset;
-	bool map_from_cluster = false;
 
 	ext_debug("blocks %u/%u requested for inode %lu\n",
 		  map->m_lblk, map->m_len, inode->i_ino);
@@ -4328,12 +4088,12 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			 */
 			if ((!ext4_ext_is_unwritten(ex)) &&
 			    (flags & EXT4_GET_BLOCKS_CONVERT_UNWRITTEN)) {
-				allocated = convert_initialized_extent(
-						handle, inode, map, &path,
-						allocated);
+				err = convert_initialized_extent(handle,
+					inode, map, &path, &allocated);
 				goto out2;
-			} else if (!ext4_ext_is_unwritten(ex))
+			} else if (!ext4_ext_is_unwritten(ex)) {
 				goto out;
+			}
 
 			ret = ext4_ext_handle_unwritten_extents(
 				handle, inode, map, &path, flags,
@@ -4384,7 +4144,6 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 	    get_implied_cluster_alloc(inode->i_sb, map, ex, path)) {
 		ar.len = allocated = map->m_len;
 		newblock = map->m_pblk;
-		map_from_cluster = true;
 		goto got_allocated_blocks;
 	}
 
@@ -4405,7 +4164,6 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 	    get_implied_cluster_alloc(inode->i_sb, map, ex2, path)) {
 		ar.len = allocated = map->m_len;
 		newblock = map->m_pblk;
-		map_from_cluster = true;
 		goto got_allocated_blocks;
 	}
 
@@ -4462,7 +4220,6 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 		goto out2;
 	ext_debug("allocate new block: goal %llu, found %llu/%u\n",
 		  ar.goal, newblock, allocated);
-	free_on_err = 1;
 	allocated_clusters = ar.len;
 	ar.len = EXT4_C2B(sbi, ar.len) - offset;
 	if (ar.len > allocated)
@@ -4473,28 +4230,28 @@ got_allocated_blocks:
 	ext4_ext_store_pblock(&newex, newblock + offset);
 	newex.ee_len = cpu_to_le16(ar.len);
 	/* Mark unwritten */
-	if (flags & EXT4_GET_BLOCKS_UNWRIT_EXT){
+	if (flags & EXT4_GET_BLOCKS_UNWRIT_EXT) {
 		ext4_ext_mark_unwritten(&newex);
 		map->m_flags |= EXT4_MAP_UNWRITTEN;
 	}
 
-	err = 0;
-	if ((flags & EXT4_GET_BLOCKS_KEEP_SIZE) == 0)
-		err = check_eofblocks_fl(handle, inode, map->m_lblk,
-					 path, ar.len);
-	if (!err)
-		err = ext4_ext_insert_extent(handle, inode, &path,
-					     &newex, flags);
-
-	if (err && free_on_err) {
-		int fb_flags = flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE ?
-			EXT4_FREE_BLOCKS_NO_QUOT_UPDATE : 0;
-		/* free data blocks we just allocated */
-		/* not a good idea to call discard here directly,
-		 * but otherwise we'd need to call it every free() */
-		ext4_discard_preallocations(inode);
-		ext4_free_blocks(handle, inode, NULL, newblock,
-				 EXT4_C2B(sbi, allocated_clusters), fb_flags);
+	err = ext4_ext_insert_extent(handle, inode, &path, &newex, flags);
+	if (err) {
+		if (allocated_clusters) {
+			int fb_flags = 0;
+
+			/*
+			 * free data blocks we just allocated.
+			 * not a good idea to call discard here directly,
+			 * but otherwise we'd need to call it every free().
+			 */
+			ext4_discard_preallocations(inode);
+			if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
+				fb_flags = EXT4_FREE_BLOCKS_NO_QUOT_UPDATE;
+			ext4_free_blocks(handle, inode, NULL, newblock,
+					 EXT4_C2B(sbi, allocated_clusters),
+					 fb_flags);
+		}
 		goto out2;
 	}
 
@@ -4511,7 +4268,7 @@ got_allocated_blocks:
 	 * clusters discovered to be delayed allocated.  Once allocated, a
 	 * cluster is not included in the reserved count.
 	 */
-	if (test_opt(inode->i_sb, DELALLOC) && !map_from_cluster) {
+	if (test_opt(inode->i_sb, DELALLOC) && allocated_clusters) {
 		if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) {
 			/*
 			 * When allocating delayed allocated clusters, simply
@@ -4665,10 +4422,6 @@ retry:
 				epos = new_size;
 			if (ext4_update_inode_size(inode, epos) & 0x1)
 				inode->i_mtime = inode->i_ctime;
-		} else {
-			if (epos > inode->i_size)
-				ext4_set_inode_flag(inode,
-						    EXT4_INODE_EOFBLOCKS);
 		}
 		ext4_mark_inode_dirty(handle, inode);
 		ext4_update_inode_fsync_trans(handle, inode, 1);
@@ -4685,6 +4438,10 @@ retry:
 	return ret > 0 ? ret2 : ret;
 }
 
+static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len);
+
+static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len);
+
 static long ext4_zero_range(struct file *file, loff_t offset,
 			    loff_t len, int mode)
 {
@@ -4702,9 +4459,6 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 
 	trace_ext4_zero_range(inode, offset, len, mode);
 
-	if (!S_ISREG(inode->i_mode))
-		return -EINVAL;
-
 	/* Call ext4_force_commit to flush all data in case of data=journal. */
 	if (ext4_should_journal_data(inode)) {
 		ret = ext4_force_commit(inode->i_sb);
@@ -4744,7 +4498,7 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 	}
 
 	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-	    (offset + len > i_size_read(inode) ||
+	    (offset + len > inode->i_size ||
 	     offset + len > EXT4_I(inode)->i_disksize)) {
 		new_size = offset + len;
 		ret = inode_newsize_ok(inode, new_size);
@@ -4821,16 +4575,8 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 	}
 
 	inode->i_mtime = inode->i_ctime = current_time(inode);
-	if (new_size) {
+	if (new_size)
 		ext4_update_inode_size(inode, new_size);
-	} else {
-		/*
-		* Mark that we allocate beyond EOF so the subsequent truncate
-		* can proceed even if the new size is the same as i_size.
-		*/
-		if ((offset + len) > i_size_read(inode))
-			ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
-	}
 	ext4_mark_inode_dirty(handle, inode);
 
 	/* Zero out partial block at the edges of the range */
@@ -4869,14 +4615,9 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 	 * range since we would need to re-encrypt blocks with a
 	 * different IV or XTS tweak (which are based on the logical
 	 * block number).
-	 *
-	 * XXX It's not clear why zero range isn't working, but we'll
-	 * leave it disabled for encrypted inodes for now.  This is a
-	 * bug we should fix....
 	 */
 	if (IS_ENCRYPTED(inode) &&
-	    (mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE |
-		     FALLOC_FL_ZERO_RANGE)))
+	    (mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE)))
 		return -EOPNOTSUPP;
 
 	/* Return error if mode is not supported */
@@ -4920,7 +4661,7 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 	}
 
 	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-	    (offset + len > i_size_read(inode) ||
+	    (offset + len > inode->i_size ||
 	     offset + len > EXT4_I(inode)->i_disksize)) {
 		new_size = offset + len;
 		ret = inode_newsize_ok(inode, new_size);
@@ -4962,23 +4703,13 @@ int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode,
 	int ret = 0;
 	int ret2 = 0;
 	struct ext4_map_blocks map;
-	unsigned int credits, blkbits = inode->i_blkbits;
+	unsigned int blkbits = inode->i_blkbits;
+	unsigned int credits = 0;
 
 	map.m_lblk = offset >> blkbits;
 	max_blocks = EXT4_MAX_BLOCKS(len, offset, blkbits);
 
-	/*
-	 * This is somewhat ugly but the idea is clear: When transaction is
-	 * reserved, everything goes into it. Otherwise we rather start several
-	 * smaller transactions for conversion of each extent separately.
-	 */
-	if (handle) {
-		handle = ext4_journal_start_reserved(handle,
-						     EXT4_HT_EXT_CONVERT);
-		if (IS_ERR(handle))
-			return PTR_ERR(handle);
-		credits = 0;
-	} else {
+	if (!handle) {
 		/*
 		 * credits to insert 1 extent into extent tree
 		 */
@@ -5009,69 +4740,47 @@ int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode,
 		if (ret <= 0 || ret2)
 			break;
 	}
-	if (!credits)
-		ret2 = ext4_journal_stop(handle);
 	return ret > 0 ? ret2 : ret;
 }
 
-/*
- * If newes is not existing extent (newes->ec_pblk equals zero) find
- * delayed extent at start of newes and update newes accordingly and
- * return start of the next delayed extent.
- *
- * If newes is existing extent (newes->ec_pblk is not equal zero)
- * return start of next delayed extent or EXT_MAX_BLOCKS if no delayed
- * extent found. Leave newes unmodified.
- */
-static int ext4_find_delayed_extent(struct inode *inode,
-				    struct extent_status *newes)
+int ext4_convert_unwritten_io_end_vec(handle_t *handle, ext4_io_end_t *io_end)
 {
-	struct extent_status es;
-	ext4_lblk_t block, next_del;
-
-	if (newes->es_pblk == 0) {
-		ext4_es_find_extent_range(inode, &ext4_es_is_delayed,
-					  newes->es_lblk,
-					  newes->es_lblk + newes->es_len - 1,
-					  &es);
+	int ret, err = 0;
+	struct ext4_io_end_vec *io_end_vec;
 
-		/*
-		 * No extent in extent-tree contains block @newes->es_pblk,
-		 * then the block may stay in 1)a hole or 2)delayed-extent.
-		 */
-		if (es.es_len == 0)
-			/* A hole found. */
-			return 0;
-
-		if (es.es_lblk > newes->es_lblk) {
-			/* A hole found. */
-			newes->es_len = min(es.es_lblk - newes->es_lblk,
-					    newes->es_len);
-			return 0;
-		}
+	/*
+	 * This is somewhat ugly but the idea is clear: When transaction is
+	 * reserved, everything goes into it. Otherwise we rather start several
+	 * smaller transactions for conversion of each extent separately.
+	 */
+	if (handle) {
+		handle = ext4_journal_start_reserved(handle,
+						     EXT4_HT_EXT_CONVERT);
+		if (IS_ERR(handle))
+			return PTR_ERR(handle);
+	}
 
-		newes->es_len = es.es_lblk + es.es_len - newes->es_lblk;
+	list_for_each_entry(io_end_vec, &io_end->list_vec, list) {
+		ret = ext4_convert_unwritten_extents(handle, io_end->inode,
+						     io_end_vec->offset,
+						     io_end_vec->size);
+		if (ret)
+			break;
 	}
 
-	block = newes->es_lblk + newes->es_len;
-	ext4_es_find_extent_range(inode, &ext4_es_is_delayed, block,
-				  EXT_MAX_BLOCKS, &es);
-	if (es.es_len == 0)
-		next_del = EXT_MAX_BLOCKS;
-	else
-		next_del = es.es_lblk;
+	if (handle)
+		err = ext4_journal_stop(handle);
 
-	return next_del;
+	return ret < 0 ? ret : err;
 }
 
-static int ext4_xattr_fiemap(struct inode *inode,
-				struct fiemap_extent_info *fieinfo)
+static int ext4_iomap_xattr_fiemap(struct inode *inode, struct iomap *iomap)
 {
 	__u64 physical = 0;
-	__u64 length;
-	__u32 flags = FIEMAP_EXTENT_LAST;
+	__u64 length = 0;
 	int blockbits = inode->i_sb->s_blocksize_bits;
 	int error = 0;
+	u16 iomap_type;
 
 	/* in-inode? */
 	if (ext4_test_inode_state(inode, EXT4_STATE_XATTR)) {
@@ -5086,40 +4795,71 @@ static int ext4_xattr_fiemap(struct inode *inode,
 				EXT4_I(inode)->i_extra_isize;
 		physical += offset;
 		length = EXT4_SB(inode->i_sb)->s_inode_size - offset;
-		flags |= FIEMAP_EXTENT_DATA_INLINE;
 		brelse(iloc.bh);
-	} else { /* external block */
+		iomap_type = IOMAP_INLINE;
+	} else if (EXT4_I(inode)->i_file_acl) { /* external block */
 		physical = (__u64)EXT4_I(inode)->i_file_acl << blockbits;
 		length = inode->i_sb->s_blocksize;
+		iomap_type = IOMAP_MAPPED;
+	} else {
+		/* no in-inode or external block for xattr, so return -ENOENT */
+		error = -ENOENT;
+		goto out;
 	}
 
-	if (physical)
-		error = fiemap_fill_next_extent(fieinfo, 0, physical,
-						length, flags);
-	return (error < 0 ? error : 0);
+	iomap->addr = physical;
+	iomap->offset = 0;
+	iomap->length = length;
+	iomap->type = iomap_type;
+	iomap->flags = 0;
+out:
+	return error;
 }
 
-static int _ext4_fiemap(struct inode *inode,
-			struct fiemap_extent_info *fieinfo,
-			__u64 start, __u64 len,
-			int (*fill)(struct inode *, ext4_lblk_t,
-				    ext4_lblk_t,
-				    struct fiemap_extent_info *))
+static int ext4_iomap_xattr_begin(struct inode *inode, loff_t offset,
+				  loff_t length, unsigned flags,
+				  struct iomap *iomap, struct iomap *srcmap)
 {
-	ext4_lblk_t start_blk;
-	u32 ext4_fiemap_flags = FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR;
+	int error;
 
-	int error = 0;
+	error = ext4_iomap_xattr_fiemap(inode, iomap);
+	if (error == 0 && (offset >= iomap->length))
+		error = -ENOENT;
+	return error;
+}
 
-	if (ext4_has_inline_data(inode)) {
-		int has_inline = 1;
+static const struct iomap_ops ext4_iomap_xattr_ops = {
+	.iomap_begin		= ext4_iomap_xattr_begin,
+};
 
-		error = ext4_inline_data_fiemap(inode, fieinfo, &has_inline,
-						start, len);
+static int ext4_fiemap_check_ranges(struct inode *inode, u64 start, u64 *len)
+{
+	u64 maxbytes;
 
-		if (has_inline)
-			return error;
-	}
+	if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
+		maxbytes = inode->i_sb->s_maxbytes;
+	else
+		maxbytes = EXT4_SB(inode->i_sb)->s_bitmap_maxbytes;
+
+	if (*len == 0)
+		return -EINVAL;
+	if (start > maxbytes)
+		return -EFBIG;
+
+	/*
+	 * Shrink request scope to what the fs can actually handle.
+	 */
+	if (*len > maxbytes || (maxbytes - *len) < start)
+		*len = maxbytes - start;
+	return 0;
+}
+
+static int _ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+			__u64 start, __u64 len, bool from_es_cache)
+{
+	ext4_lblk_t start_blk;
+	u32 ext4_fiemap_flags = FIEMAP_FLAG_SYNC | FIEMAP_FLAG_XATTR;
+	int error = 0;
 
 	if (fieinfo->fi_flags & FIEMAP_FLAG_CACHE) {
 		error = ext4_ext_precache(inode);
@@ -5128,19 +4868,28 @@ static int _ext4_fiemap(struct inode *inode,
 		fieinfo->fi_flags &= ~FIEMAP_FLAG_CACHE;
 	}
 
-	/* fallback to generic here if not in extents fmt */
-	if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) &&
-	    fill == ext4_fill_fiemap_extents)
-		return generic_block_fiemap(inode, fieinfo, start, len,
-			ext4_get_block);
-
-	if (fill == ext4_fill_es_cache_info)
+	if (from_es_cache)
 		ext4_fiemap_flags &= FIEMAP_FLAG_XATTR;
+
 	if (fiemap_check_flags(fieinfo, ext4_fiemap_flags))
 		return -EBADR;
 
+	/*
+	 * For bitmap files the maximum size limit could be smaller than
+	 * s_maxbytes, so check len here manually instead of just relying on the
+	 * generic check.
+	 */
+	error = ext4_fiemap_check_ranges(inode, start, &len);
+	if (error)
+		return error;
+
 	if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) {
-		error = ext4_xattr_fiemap(inode, fieinfo);
+		fieinfo->fi_flags &= ~FIEMAP_FLAG_XATTR;
+		error = iomap_fiemap(inode, fieinfo, start, len,
+				     &ext4_iomap_xattr_ops);
+	} else if (!from_es_cache) {
+		error = iomap_fiemap(inode, fieinfo, start, len,
+				     &ext4_iomap_report_ops);
 	} else {
 		ext4_lblk_t len_blks;
 		__u64 last_blk;
@@ -5155,7 +4904,8 @@ static int _ext4_fiemap(struct inode *inode,
 		 * Walk the extent tree gathering extent information
 		 * and pushing extents back to the user.
 		 */
-		error = fill(inode, start_blk, len_blks, fieinfo);
+		error = ext4_fill_es_cache_info(inode, start_blk, len_blks,
+						fieinfo);
 	}
 	return error;
 }
@@ -5163,8 +4913,7 @@ static int _ext4_fiemap(struct inode *inode,
 int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		__u64 start, __u64 len)
 {
-	return _ext4_fiemap(inode, fieinfo, start, len,
-			    ext4_fill_fiemap_extents);
+	return _ext4_fiemap(inode, fieinfo, start, len, false);
 }
 
 int ext4_get_es_cache(struct inode *inode, struct fiemap_extent_info *fieinfo,
@@ -5180,8 +4929,7 @@ int ext4_get_es_cache(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			return 0;
 	}
 
-	return _ext4_fiemap(inode, fieinfo, start, len,
-			    ext4_fill_es_cache_info);
+	return _ext4_fiemap(inode, fieinfo, start, len, true);
 }
 
 
@@ -5206,13 +4954,10 @@ ext4_access_path(handle_t *handle, struct inode *inode,
 	 * descriptor) for each block group; assume two block
 	 * groups
 	 */
-	if (handle->h_buffer_credits < 7) {
-		credits = ext4_writepage_trans_blocks(inode);
-		err = ext4_ext_truncate_extend_restart(handle, inode, credits);
-		/* EAGAIN is success */
-		if (err && err != -EAGAIN)
-			return err;
-	}
+	credits = ext4_writepage_trans_blocks(inode);
+	err = ext4_datasem_ensure_credits(handle, inode, 7, credits, 0);
+	if (err < 0)
+		return err;
 
 	err = ext4_ext_get_access(handle, inode, path);
 	return err;
@@ -5231,7 +4976,7 @@ ext4_ext_shift_path_extents(struct ext4_ext_path *path, ext4_lblk_t shift,
 {
 	int depth, err = 0;
 	struct ext4_extent *ex_start, *ex_last;
-	bool update = 0;
+	bool update = false;
 	depth = path->p_depth;
 
 	while (depth >= 0) {
@@ -5247,7 +4992,7 @@ ext4_ext_shift_path_extents(struct ext4_ext_path *path, ext4_lblk_t shift,
 				goto out;
 
 			if (ex_start == EXT_FIRST_EXTENT(path[depth].p_hdr))
-				update = 1;
+				update = true;
 
 			while (ex_start <= ex_last) {
 				if (SHIFT == SHIFT_LEFT) {
@@ -5435,7 +5180,7 @@ out:
  * This implements the fallocate's collapse range functionality for ext4
  * Returns: 0 and non-zero on error.
  */
-int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
+static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 {
 	struct super_block *sb = inode->i_sb;
 	ext4_lblk_t punch_start, punch_stop;
@@ -5452,12 +5197,8 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
 		return -EOPNOTSUPP;
 
-	/* Collapse range works only on fs block size aligned offsets. */
-	if (offset & (EXT4_CLUSTER_SIZE(sb) - 1) ||
-	    len & (EXT4_CLUSTER_SIZE(sb) - 1))
-		return -EINVAL;
-
-	if (!S_ISREG(inode->i_mode))
+	/* Collapse range works only on fs cluster size aligned regions. */
+	if (!IS_ALIGNED(offset | len, EXT4_CLUSTER_SIZE(sb)))
 		return -EINVAL;
 
 	trace_ext4_collapse_range(inode, offset, len);
@@ -5477,7 +5218,7 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 	 * There is no need to overlap collapse range with EOF, in which case
 	 * it is effectively a truncate operation
 	 */
-	if (offset + len >= i_size_read(inode)) {
+	if (offset + len >= inode->i_size) {
 		ret = -EINVAL;
 		goto out_mutex;
 	}
@@ -5555,7 +5296,7 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 		goto out_stop;
 	}
 
-	new_size = i_size_read(inode) - len;
+	new_size = inode->i_size - len;
 	i_size_write(inode, new_size);
 	EXT4_I(inode)->i_disksize = new_size;
 
@@ -5583,7 +5324,7 @@ out_mutex:
  * by len bytes.
  * Returns 0 on success, error otherwise.
  */
-int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
+static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
 {
 	struct super_block *sb = inode->i_sb;
 	handle_t *handle;
@@ -5602,14 +5343,10 @@ int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
 	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
 		return -EOPNOTSUPP;
 
-	/* Insert range works only on fs block size aligned offsets. */
-	if (offset & (EXT4_CLUSTER_SIZE(sb) - 1) ||
-			len & (EXT4_CLUSTER_SIZE(sb) - 1))
+	/* Insert range works only on fs cluster size aligned regions. */
+	if (!IS_ALIGNED(offset | len, EXT4_CLUSTER_SIZE(sb)))
 		return -EINVAL;
 
-	if (!S_ISREG(inode->i_mode))
-		return -EOPNOTSUPP;
-
 	trace_ext4_insert_range(inode, offset, len);
 
 	offset_lblk = offset >> EXT4_BLOCK_SIZE_BITS(sb);
@@ -5629,14 +5366,14 @@ int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
 		goto out_mutex;
 	}
 
-	/* Check for wrap through zero */
-	if (inode->i_size + len > inode->i_sb->s_maxbytes) {
+	/* Check whether the maximum file size would be exceeded */
+	if (len > inode->i_sb->s_maxbytes - inode->i_size) {
 		ret = -EFBIG;
 		goto out_mutex;
 	}
 
-	/* Offset should be less than i_size */
-	if (offset >= i_size_read(inode)) {
+	/* Offset must be less than i_size */
+	if (offset >= inode->i_size) {
 		ret = -EINVAL;
 		goto out_mutex;
 	}
diff --git a/fs/ext4/extents_status.h b/fs/ext4/extents_status.h
index 825313c59752..4ec30a798260 100644
--- a/fs/ext4/extents_status.h
+++ b/fs/ext4/extents_status.h
@@ -209,6 +209,12 @@ static inline ext4_fsblk_t ext4_es_pblock(struct extent_status *es)
 	return es->es_pblk & ~ES_MASK;
 }
 
+static inline ext4_fsblk_t ext4_es_show_pblock(struct extent_status *es)
+{
+	ext4_fsblk_t pblock = ext4_es_pblock(es);
+	return pblock == ~ES_MASK ? 0 : pblock;
+}
+
 static inline void ext4_es_store_pblock(struct extent_status *es,
 					ext4_fsblk_t pb)
 {
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 8d2bbcc2d813..0d624250a62b 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -29,10 +29,58 @@
 #include <linux/pagevec.h>
 #include <linux/uio.h>
 #include <linux/mman.h>
+#include <linux/backing-dev.h>
 #include "ext4.h"
 #include "ext4_jbd2.h"
 #include "xattr.h"
 #include "acl.h"
+#include "truncate.h"
+
+static bool ext4_dio_supported(struct inode *inode)
+{
+	if (IS_ENABLED(CONFIG_FS_ENCRYPTION) && IS_ENCRYPTED(inode))
+		return false;
+	if (fsverity_active(inode))
+		return false;
+	if (ext4_should_journal_data(inode))
+		return false;
+	if (ext4_has_inline_data(inode))
+		return false;
+	return true;
+}
+
+static ssize_t ext4_dio_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	ssize_t ret;
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock_shared(inode))
+			return -EAGAIN;
+	} else {
+		inode_lock_shared(inode);
+	}
+
+	if (!ext4_dio_supported(inode)) {
+		inode_unlock_shared(inode);
+		/*
+		 * Fallback to buffered I/O if the operation being performed on
+		 * the inode is not supported by direct I/O. The IOCB_DIRECT
+		 * flag needs to be cleared here in order to ensure that the
+		 * direct I/O path within generic_file_read_iter() is not
+		 * taken.
+		 */
+		iocb->ki_flags &= ~IOCB_DIRECT;
+		return generic_file_read_iter(iocb, to);
+	}
+
+	ret = iomap_dio_rw(iocb, to, &ext4_iomap_ops, NULL,
+			   is_sync_kiocb(iocb));
+	inode_unlock_shared(inode);
+
+	file_accessed(iocb->ki_filp);
+	return ret;
+}
 
 #ifdef CONFIG_FS_DAX
 static ssize_t ext4_dax_read_iter(struct kiocb *iocb, struct iov_iter *to)
@@ -40,9 +88,10 @@ static ssize_t ext4_dax_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	struct inode *inode = file_inode(iocb->ki_filp);
 	ssize_t ret;
 
-	if (!inode_trylock_shared(inode)) {
-		if (iocb->ki_flags & IOCB_NOWAIT)
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock_shared(inode))
 			return -EAGAIN;
+	} else {
 		inode_lock_shared(inode);
 	}
 	/*
@@ -64,16 +113,21 @@ static ssize_t ext4_dax_read_iter(struct kiocb *iocb, struct iov_iter *to)
 
 static ssize_t ext4_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
-	if (unlikely(ext4_forced_shutdown(EXT4_SB(file_inode(iocb->ki_filp)->i_sb))))
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
 		return -EIO;
 
 	if (!iov_iter_count(to))
 		return 0; /* skip atime */
 
 #ifdef CONFIG_FS_DAX
-	if (IS_DAX(file_inode(iocb->ki_filp)))
+	if (IS_DAX(inode))
 		return ext4_dax_read_iter(iocb, to);
 #endif
+	if (iocb->ki_flags & IOCB_DIRECT)
+		return ext4_dio_read_iter(iocb, to);
+
 	return generic_file_read_iter(iocb, to);
 }
 
@@ -103,13 +157,6 @@ static int ext4_release_file(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-static void ext4_unwritten_wait(struct inode *inode)
-{
-	wait_queue_head_t *wq = ext4_ioend_wq(inode);
-
-	wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_unwritten) == 0));
-}
-
 /*
  * This tests whether the IO in question is block-aligned or not.
  * Ext4 utilizes unwritten extents when hole-filling during direct IO, and they
@@ -119,19 +166,25 @@ static void ext4_unwritten_wait(struct inode *inode)
  * threads are at work on the same unwritten block, they must be synchronized
  * or one thread will zero the other's data, causing corruption.
  */
-static int
-ext4_unaligned_aio(struct inode *inode, struct iov_iter *from, loff_t pos)
+static bool
+ext4_unaligned_io(struct inode *inode, struct iov_iter *from, loff_t pos)
 {
 	struct super_block *sb = inode->i_sb;
-	int blockmask = sb->s_blocksize - 1;
-
-	if (pos >= ALIGN(i_size_read(inode), sb->s_blocksize))
-		return 0;
+	unsigned long blockmask = sb->s_blocksize - 1;
 
 	if ((pos | iov_iter_alignment(from)) & blockmask)
-		return 1;
+		return true;
 
-	return 0;
+	return false;
+}
+
+static bool
+ext4_extending_io(struct inode *inode, loff_t offset, size_t len)
+{
+	if (offset + len > i_size_read(inode) ||
+	    offset + len > EXT4_I(inode)->i_disksize)
+		return true;
+	return false;
 }
 
 /* Is IO overwriting allocated and initialized blocks? */
@@ -157,18 +210,19 @@ static bool ext4_overwrite_io(struct inode *inode, loff_t pos, loff_t len)
 	return err == blklen && (map.m_flags & EXT4_MAP_MAPPED);
 }
 
-static ssize_t ext4_write_checks(struct kiocb *iocb, struct iov_iter *from)
+static ssize_t ext4_generic_write_checks(struct kiocb *iocb,
+					 struct iov_iter *from)
 {
 	struct inode *inode = file_inode(iocb->ki_filp);
 	ssize_t ret;
 
+	if (unlikely(IS_IMMUTABLE(inode)))
+		return -EPERM;
+
 	ret = generic_write_checks(iocb, from);
 	if (ret <= 0)
 		return ret;
 
-	if (unlikely(IS_IMMUTABLE(inode)))
-		return -EPERM;
-
 	/*
 	 * If we have encountered a bitmap-format file, the size limit
 	 * is smaller than s_maxbytes, which is for extent-mapped files.
@@ -180,110 +234,413 @@ static ssize_t ext4_write_checks(struct kiocb *iocb, struct iov_iter *from)
 			return -EFBIG;
 		iov_iter_truncate(from, sbi->s_bitmap_maxbytes - iocb->ki_pos);
 	}
+
 	return iov_iter_count(from);
 }
 
-#ifdef CONFIG_FS_DAX
-static ssize_t
-ext4_dax_write_iter(struct kiocb *iocb, struct iov_iter *from)
+static ssize_t ext4_write_checks(struct kiocb *iocb, struct iov_iter *from)
+{
+	ssize_t ret, count;
+
+	count = ext4_generic_write_checks(iocb, from);
+	if (count <= 0)
+		return count;
+
+	ret = file_modified(iocb->ki_filp);
+	if (ret)
+		return ret;
+	return count;
+}
+
+static ssize_t ext4_buffered_write_iter(struct kiocb *iocb,
+					struct iov_iter *from)
 {
-	struct inode *inode = file_inode(iocb->ki_filp);
 	ssize_t ret;
+	struct inode *inode = file_inode(iocb->ki_filp);
 
-	if (!inode_trylock(inode)) {
-		if (iocb->ki_flags & IOCB_NOWAIT)
-			return -EAGAIN;
-		inode_lock(inode);
-	}
+	if (iocb->ki_flags & IOCB_NOWAIT)
+		return -EOPNOTSUPP;
+
+	inode_lock(inode);
 	ret = ext4_write_checks(iocb, from);
 	if (ret <= 0)
 		goto out;
-	ret = file_remove_privs(iocb->ki_filp);
-	if (ret)
-		goto out;
-	ret = file_update_time(iocb->ki_filp);
-	if (ret)
-		goto out;
 
-	ret = dax_iomap_rw(iocb, from, &ext4_iomap_ops);
+	current->backing_dev_info = inode_to_bdi(inode);
+	ret = generic_perform_write(iocb->ki_filp, from, iocb->ki_pos);
+	current->backing_dev_info = NULL;
+
 out:
 	inode_unlock(inode);
-	if (ret > 0)
+	if (likely(ret > 0)) {
+		iocb->ki_pos += ret;
 		ret = generic_write_sync(iocb, ret);
+	}
+
 	return ret;
 }
-#endif
 
-static ssize_t
-ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
+static ssize_t ext4_handle_inode_extension(struct inode *inode, loff_t offset,
+					   ssize_t written, size_t count)
 {
-	struct inode *inode = file_inode(iocb->ki_filp);
-	int o_direct = iocb->ki_flags & IOCB_DIRECT;
-	int unaligned_aio = 0;
-	int overwrite = 0;
-	ssize_t ret;
+	handle_t *handle;
+	bool truncate = false;
+	u8 blkbits = inode->i_blkbits;
+	ext4_lblk_t written_blk, end_blk;
 
-	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
-		return -EIO;
+	/*
+	 * Note that EXT4_I(inode)->i_disksize can get extended up to
+	 * inode->i_size while the I/O was running due to writeback of delalloc
+	 * blocks. But, the code in ext4_iomap_alloc() is careful to use
+	 * zeroed/unwritten extents if this is possible; thus we won't leave
+	 * uninitialized blocks in a file even if we didn't succeed in writing
+	 * as much as we intended.
+	 */
+	WARN_ON_ONCE(i_size_read(inode) < EXT4_I(inode)->i_disksize);
+	if (offset + count <= EXT4_I(inode)->i_disksize) {
+		/*
+		 * We need to ensure that the inode is removed from the orphan
+		 * list if it has been added prematurely, due to writeback of
+		 * delalloc blocks.
+		 */
+		if (!list_empty(&EXT4_I(inode)->i_orphan) && inode->i_nlink) {
+			handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
+
+			if (IS_ERR(handle)) {
+				ext4_orphan_del(NULL, inode);
+				return PTR_ERR(handle);
+			}
+
+			ext4_orphan_del(handle, inode);
+			ext4_journal_stop(handle);
+		}
 
-#ifdef CONFIG_FS_DAX
-	if (IS_DAX(inode))
-		return ext4_dax_write_iter(iocb, from);
-#endif
+		return written;
+	}
 
-	if (!inode_trylock(inode)) {
-		if (iocb->ki_flags & IOCB_NOWAIT)
-			return -EAGAIN;
-		inode_lock(inode);
+	if (written < 0)
+		goto truncate;
+
+	handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
+	if (IS_ERR(handle)) {
+		written = PTR_ERR(handle);
+		goto truncate;
 	}
 
-	ret = ext4_write_checks(iocb, from);
+	if (ext4_update_inode_size(inode, offset + written))
+		ext4_mark_inode_dirty(handle, inode);
+
+	/*
+	 * We may need to truncate allocated but not written blocks beyond EOF.
+	 */
+	written_blk = ALIGN(offset + written, 1 << blkbits);
+	end_blk = ALIGN(offset + count, 1 << blkbits);
+	if (written_blk < end_blk && ext4_can_truncate(inode))
+		truncate = true;
+
+	/*
+	 * Remove the inode from the orphan list if it has been extended and
+	 * everything went OK.
+	 */
+	if (!truncate && inode->i_nlink)
+		ext4_orphan_del(handle, inode);
+	ext4_journal_stop(handle);
+
+	if (truncate) {
+truncate:
+		ext4_truncate_failed_write(inode);
+		/*
+		 * If the truncate operation failed early, then the inode may
+		 * still be on the orphan list. In that case, we need to try
+		 * remove the inode from the in-memory linked list.
+		 */
+		if (inode->i_nlink)
+			ext4_orphan_del(NULL, inode);
+	}
+
+	return written;
+}
+
+static int ext4_dio_write_end_io(struct kiocb *iocb, ssize_t size,
+				 int error, unsigned int flags)
+{
+	loff_t offset = iocb->ki_pos;
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	if (error)
+		return error;
+
+	if (size && flags & IOMAP_DIO_UNWRITTEN)
+		return ext4_convert_unwritten_extents(NULL, inode,
+						      offset, size);
+
+	return 0;
+}
+
+static const struct iomap_dio_ops ext4_dio_write_ops = {
+	.end_io = ext4_dio_write_end_io,
+};
+
+/*
+ * The intention here is to start with shared lock acquired then see if any
+ * condition requires an exclusive inode lock. If yes, then we restart the
+ * whole operation by releasing the shared lock and acquiring exclusive lock.
+ *
+ * - For unaligned_io we never take shared lock as it may cause data corruption
+ *   when two unaligned IO tries to modify the same block e.g. while zeroing.
+ *
+ * - For extending writes case we don't take the shared lock, since it requires
+ *   updating inode i_disksize and/or orphan handling with exclusive lock.
+ *
+ * - shared locking will only be true mostly with overwrites. Otherwise we will
+ *   switch to exclusive i_rwsem lock.
+ */
+static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from,
+				     bool *ilock_shared, bool *extend)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+	loff_t offset;
+	size_t count;
+	ssize_t ret;
+
+restart:
+	ret = ext4_generic_write_checks(iocb, from);
 	if (ret <= 0)
 		goto out;
 
+	offset = iocb->ki_pos;
+	count = ret;
+	if (ext4_extending_io(inode, offset, count))
+		*extend = true;
 	/*
-	 * Unaligned direct AIO must be serialized among each other as zeroing
-	 * of partial blocks of two competing unaligned AIOs can result in data
-	 * corruption.
+	 * Determine whether the IO operation will overwrite allocated
+	 * and initialized blocks.
+	 * We need exclusive i_rwsem for changing security info
+	 * in file_modified().
 	 */
-	if (o_direct && ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS) &&
-	    !is_sync_kiocb(iocb) &&
-	    ext4_unaligned_aio(inode, from, iocb->ki_pos)) {
-		unaligned_aio = 1;
-		ext4_unwritten_wait(inode);
+	if (*ilock_shared && (!IS_NOSEC(inode) || *extend ||
+	     !ext4_overwrite_io(inode, offset, count))) {
+		inode_unlock_shared(inode);
+		*ilock_shared = false;
+		inode_lock(inode);
+		goto restart;
 	}
 
-	iocb->private = &overwrite;
-	/* Check whether we do a DIO overwrite or not */
-	if (o_direct && !unaligned_aio) {
-		if (ext4_overwrite_io(inode, iocb->ki_pos, iov_iter_count(from))) {
-			if (ext4_should_dioread_nolock(inode))
-				overwrite = 1;
-		} else if (iocb->ki_flags & IOCB_NOWAIT) {
-			ret = -EAGAIN;
-			goto out;
+	ret = file_modified(file);
+	if (ret < 0)
+		goto out;
+
+	return count;
+out:
+	if (*ilock_shared)
+		inode_unlock_shared(inode);
+	else
+		inode_unlock(inode);
+	return ret;
+}
+
+static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+	ssize_t ret;
+	handle_t *handle;
+	struct inode *inode = file_inode(iocb->ki_filp);
+	loff_t offset = iocb->ki_pos;
+	size_t count = iov_iter_count(from);
+	const struct iomap_ops *iomap_ops = &ext4_iomap_ops;
+	bool extend = false, unaligned_io = false;
+	bool ilock_shared = true;
+
+	/*
+	 * We initially start with shared inode lock unless it is
+	 * unaligned IO which needs exclusive lock anyways.
+	 */
+	if (ext4_unaligned_io(inode, from, offset)) {
+		unaligned_io = true;
+		ilock_shared = false;
+	}
+	/*
+	 * Quick check here without any i_rwsem lock to see if it is extending
+	 * IO. A more reliable check is done in ext4_dio_write_checks() with
+	 * proper locking in place.
+	 */
+	if (offset + count > i_size_read(inode))
+		ilock_shared = false;
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (ilock_shared) {
+			if (!inode_trylock_shared(inode))
+				return -EAGAIN;
+		} else {
+			if (!inode_trylock(inode))
+				return -EAGAIN;
 		}
+	} else {
+		if (ilock_shared)
+			inode_lock_shared(inode);
+		else
+			inode_lock(inode);
+	}
+
+	/* Fallback to buffered I/O if the inode does not support direct I/O. */
+	if (!ext4_dio_supported(inode)) {
+		if (ilock_shared)
+			inode_unlock_shared(inode);
+		else
+			inode_unlock(inode);
+		return ext4_buffered_write_iter(iocb, from);
 	}
 
-	ret = __generic_file_write_iter(iocb, from);
+	ret = ext4_dio_write_checks(iocb, from, &ilock_shared, &extend);
+	if (ret <= 0)
+		return ret;
+
+	offset = iocb->ki_pos;
+	count = ret;
+
 	/*
-	 * Unaligned direct AIO must be the only IO in flight. Otherwise
-	 * overlapping aligned IO after unaligned might result in data
+	 * Unaligned direct IO must be serialized among each other as zeroing
+	 * of partial blocks of two competing unaligned IOs can result in data
 	 * corruption.
+	 *
+	 * So we make sure we don't allow any unaligned IO in flight.
+	 * For IOs where we need not wait (like unaligned non-AIO DIO),
+	 * below inode_dio_wait() may anyway become a no-op, since we start
+	 * with exclusive lock.
 	 */
-	if (ret == -EIOCBQUEUED && unaligned_aio)
-		ext4_unwritten_wait(inode);
-	inode_unlock(inode);
+	if (unaligned_io)
+		inode_dio_wait(inode);
 
-	if (ret > 0)
-		ret = generic_write_sync(iocb, ret);
+	if (extend) {
+		handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
+		if (IS_ERR(handle)) {
+			ret = PTR_ERR(handle);
+			goto out;
+		}
+
+		ret = ext4_orphan_add(handle, inode);
+		if (ret) {
+			ext4_journal_stop(handle);
+			goto out;
+		}
+
+		ext4_journal_stop(handle);
+	}
+
+	if (ilock_shared)
+		iomap_ops = &ext4_iomap_overwrite_ops;
+	ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops,
+			   is_sync_kiocb(iocb) || unaligned_io || extend);
+
+	if (extend)
+		ret = ext4_handle_inode_extension(inode, offset, ret, count);
+
+out:
+	if (ilock_shared)
+		inode_unlock_shared(inode);
+	else
+		inode_unlock(inode);
+
+	if (ret >= 0 && iov_iter_count(from)) {
+		ssize_t err;
+		loff_t endbyte;
+
+		offset = iocb->ki_pos;
+		err = ext4_buffered_write_iter(iocb, from);
+		if (err < 0)
+			return err;
+
+		/*
+		 * We need to ensure that the pages within the page cache for
+		 * the range covered by this I/O are written to disk and
+		 * invalidated. This is in attempt to preserve the expected
+		 * direct I/O semantics in the case we fallback to buffered I/O
+		 * to complete off the I/O request.
+		 */
+		ret += err;
+		endbyte = offset + err - 1;
+		err = filemap_write_and_wait_range(iocb->ki_filp->f_mapping,
+						   offset, endbyte);
+		if (!err)
+			invalidate_mapping_pages(iocb->ki_filp->f_mapping,
+						 offset >> PAGE_SHIFT,
+						 endbyte >> PAGE_SHIFT);
+	}
 
 	return ret;
+}
+
+#ifdef CONFIG_FS_DAX
+static ssize_t
+ext4_dax_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+	ssize_t ret;
+	size_t count;
+	loff_t offset;
+	handle_t *handle;
+	bool extend = false;
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock(inode))
+			return -EAGAIN;
+	} else {
+		inode_lock(inode);
+	}
+
+	ret = ext4_write_checks(iocb, from);
+	if (ret <= 0)
+		goto out;
+
+	offset = iocb->ki_pos;
+	count = iov_iter_count(from);
+
+	if (offset + count > EXT4_I(inode)->i_disksize) {
+		handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
+		if (IS_ERR(handle)) {
+			ret = PTR_ERR(handle);
+			goto out;
+		}
+
+		ret = ext4_orphan_add(handle, inode);
+		if (ret) {
+			ext4_journal_stop(handle);
+			goto out;
+		}
+
+		extend = true;
+		ext4_journal_stop(handle);
+	}
+
+	ret = dax_iomap_rw(iocb, from, &ext4_iomap_ops);
 
+	if (extend)
+		ret = ext4_handle_inode_extension(inode, offset, ret, count);
 out:
 	inode_unlock(inode);
+	if (ret > 0)
+		ret = generic_write_sync(iocb, ret);
 	return ret;
 }
+#endif
+
+static ssize_t
+ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
+		return -EIO;
+
+#ifdef CONFIG_FS_DAX
+	if (IS_DAX(inode))
+		return ext4_dax_write_iter(iocb, from);
+#endif
+	if (iocb->ki_flags & IOCB_DIRECT)
+		return ext4_dio_write_iter(iocb, from);
+
+	return ext4_buffered_write_iter(iocb, from);
+}
 
 #ifdef CONFIG_FS_DAX
 static vm_fault_t ext4_dax_huge_fault(struct vm_fault *vmf,
@@ -494,12 +851,14 @@ loff_t ext4_llseek(struct file *file, loff_t offset, int whence)
 						maxbytes, i_size_read(inode));
 	case SEEK_HOLE:
 		inode_lock_shared(inode);
-		offset = iomap_seek_hole(inode, offset, &ext4_iomap_ops);
+		offset = iomap_seek_hole(inode, offset,
+					 &ext4_iomap_report_ops);
 		inode_unlock_shared(inode);
 		break;
 	case SEEK_DATA:
 		inode_lock_shared(inode);
-		offset = iomap_seek_data(inode, offset, &ext4_iomap_ops);
+		offset = iomap_seek_data(inode, offset,
+					 &ext4_iomap_report_ops);
 		inode_unlock_shared(inode);
 		break;
 	}
@@ -513,6 +872,7 @@ const struct file_operations ext4_file_operations = {
 	.llseek		= ext4_llseek,
 	.read_iter	= ext4_file_read_iter,
 	.write_iter	= ext4_file_write_iter,
+	.iopoll		= iomap_dio_iopoll,
 	.unlocked_ioctl = ext4_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= ext4_compat_ioctl,
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c
index 5508baa11bb6..35ff9a56db67 100644
--- a/fs/ext4/fsync.c
+++ b/fs/ext4/fsync.c
@@ -80,6 +80,43 @@ static int ext4_sync_parent(struct inode *inode)
 	return ret;
 }
 
+static int ext4_fsync_nojournal(struct inode *inode, bool datasync,
+				bool *needs_barrier)
+{
+	int ret, err;
+
+	ret = sync_mapping_buffers(inode->i_mapping);
+	if (!(inode->i_state & I_DIRTY_ALL))
+		return ret;
+	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
+		return ret;
+
+	err = sync_inode_metadata(inode, 1);
+	if (!ret)
+		ret = err;
+
+	if (!ret)
+		ret = ext4_sync_parent(inode);
+	if (test_opt(inode->i_sb, BARRIER))
+		*needs_barrier = true;
+
+	return ret;
+}
+
+static int ext4_fsync_journal(struct inode *inode, bool datasync,
+			     bool *needs_barrier)
+{
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
+	tid_t commit_tid = datasync ? ei->i_datasync_tid : ei->i_sync_tid;
+
+	if (journal->j_flags & JBD2_BARRIER &&
+	    !jbd2_trans_will_send_data_barrier(journal, commit_tid))
+		*needs_barrier = true;
+
+	return jbd2_complete_transaction(journal, commit_tid);
+}
+
 /*
  * akpm: A new design for ext4_sync_file().
  *
@@ -91,17 +128,14 @@ static int ext4_sync_parent(struct inode *inode)
  * What we do is just kick off a commit and wait on it.  This will snapshot the
  * inode to disk.
  */
-
 int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	struct inode *inode = file->f_mapping->host;
-	struct ext4_inode_info *ei = EXT4_I(inode);
-	journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
 	int ret = 0, err;
-	tid_t commit_tid;
 	bool needs_barrier = false;
+	struct inode *inode = file->f_mapping->host;
+	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 
-	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
+	if (unlikely(ext4_forced_shutdown(sbi)))
 		return -EIO;
 
 	J_ASSERT(ext4_journal_current_handle() == NULL);
@@ -111,23 +145,15 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	if (sb_rdonly(inode->i_sb)) {
 		/* Make sure that we read updated s_mount_flags value */
 		smp_rmb();
-		if (EXT4_SB(inode->i_sb)->s_mount_flags & EXT4_MF_FS_ABORTED)
+		if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED)
 			ret = -EROFS;
 		goto out;
 	}
 
-	if (!journal) {
-		ret = __generic_file_fsync(file, start, end, datasync);
-		if (!ret)
-			ret = ext4_sync_parent(inode);
-		if (test_opt(inode->i_sb, BARRIER))
-			goto issue_flush;
-		goto out;
-	}
-
 	ret = file_write_and_wait_range(file, start, end);
 	if (ret)
 		return ret;
+
 	/*
 	 * data=writeback,ordered:
 	 *  The caller's filemap_fdatawrite()/wait will sync the data.
@@ -142,19 +168,15 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	 *  (they were dirtied by commit).  But that's OK - the blocks are
 	 *  safe in-journal, which is all fsync() needs to ensure.
 	 */
-	if (ext4_should_journal_data(inode)) {
+	if (!sbi->s_journal)
+		ret = ext4_fsync_nojournal(inode, datasync, &needs_barrier);
+	else if (ext4_should_journal_data(inode))
 		ret = ext4_force_commit(inode->i_sb);
-		goto out;
-	}
+	else
+		ret = ext4_fsync_journal(inode, datasync, &needs_barrier);
 
-	commit_tid = datasync ? ei->i_datasync_tid : ei->i_sync_tid;
-	if (journal->j_flags & JBD2_BARRIER &&
-	    !jbd2_trans_will_send_data_barrier(journal, commit_tid))
-		needs_barrier = true;
-	ret = jbd2_complete_transaction(journal, commit_tid);
 	if (needs_barrier) {
-	issue_flush:
-		err = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+		err = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
 		if (!ret)
 			ret = err;
 	}
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 764ff4c56233..499f08d8522e 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -94,7 +94,8 @@ static int ext4_validate_inode_bitmap(struct super_block *sb,
 		goto verified;
 	blk = ext4_inode_bitmap(sb, desc);
 	if (!ext4_inode_bitmap_csum_verify(sb, block_group, desc, bh,
-					   EXT4_INODES_PER_GROUP(sb) / 8)) {
+					   EXT4_INODES_PER_GROUP(sb) / 8) ||
+	    ext4_simulate_fail(sb, EXT4_SIM_IBITMAP_CRC)) {
 		ext4_unlock_group(sb, block_group);
 		ext4_error(sb, "Corrupt inode bitmap - block_group = %u, "
 			   "inode_bitmap = %llu", block_group, blk);
@@ -112,7 +113,7 @@ verified:
  * Read the inode allocation bitmap for a given block_group, reading
  * into the specified slot in the superblock's bitmap cache.
  *
- * Return buffer_head of bitmap on success or NULL.
+ * Return buffer_head of bitmap on success, or an ERR_PTR on error.
  */
 static struct buffer_head *
 ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
@@ -192,11 +193,12 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
 	get_bh(bh);
 	submit_bh(REQ_OP_READ, REQ_META | REQ_PRIO, bh);
 	wait_on_buffer(bh);
+	ext4_simulate_fail_bh(sb, bh, EXT4_SIM_IBITMAP_EIO);
 	if (!buffer_uptodate(bh)) {
 		put_bh(bh);
-		ext4_error(sb, "Cannot read inode bitmap - "
-			   "block_group = %u, inode_bitmap = %llu",
-			   block_group, bitmap_blk);
+		ext4_error_err(sb, EIO, "Cannot read inode bitmap - "
+			       "block_group = %u, inode_bitmap = %llu",
+			       block_group, bitmap_blk);
 		ext4_mark_group_bitmap_corrupted(sb, block_group,
 				EXT4_GROUP_INFO_IBITMAP_CORRUPT);
 		return ERR_PTR(-EIO);
@@ -265,13 +267,8 @@ void ext4_free_inode(handle_t *handle, struct inode *inode)
 	ext4_debug("freeing inode %lu\n", ino);
 	trace_ext4_free_inode(inode);
 
-	/*
-	 * Note: we must free any quota before locking the superblock,
-	 * as writing the quota to disk may need the lock as well.
-	 */
 	dquot_initialize(inode);
 	dquot_free_inode(inode);
-	dquot_drop(inode);
 
 	is_directory = S_ISDIR(inode->i_mode);
 
@@ -330,11 +327,13 @@ void ext4_free_inode(handle_t *handle, struct inode *inode)
 
 	percpu_counter_inc(&sbi->s_freeinodes_counter);
 	if (sbi->s_log_groups_per_flex) {
-		ext4_group_t f = ext4_flex_group(sbi, block_group);
+		struct flex_groups *fg;
 
-		atomic_inc(&sbi->s_flex_groups[f].free_inodes);
+		fg = sbi_array_rcu_deref(sbi, s_flex_groups,
+					 ext4_flex_group(sbi, block_group));
+		atomic_inc(&fg->free_inodes);
 		if (is_directory)
-			atomic_dec(&sbi->s_flex_groups[f].used_dirs);
+			atomic_dec(&fg->used_dirs);
 	}
 	BUFFER_TRACE(bh2, "call ext4_handle_dirty_metadata");
 	fatal = ext4_handle_dirty_metadata(handle, NULL, bh2);
@@ -370,12 +369,13 @@ static void get_orlov_stats(struct super_block *sb, ext4_group_t g,
 			    int flex_size, struct orlov_stats *stats)
 {
 	struct ext4_group_desc *desc;
-	struct flex_groups *flex_group = EXT4_SB(sb)->s_flex_groups;
 
 	if (flex_size > 1) {
-		stats->free_inodes = atomic_read(&flex_group[g].free_inodes);
-		stats->free_clusters = atomic64_read(&flex_group[g].free_clusters);
-		stats->used_dirs = atomic_read(&flex_group[g].used_dirs);
+		struct flex_groups *fg = sbi_array_rcu_deref(EXT4_SB(sb),
+							     s_flex_groups, g);
+		stats->free_inodes = atomic_read(&fg->free_inodes);
+		stats->free_clusters = atomic64_read(&fg->free_clusters);
+		stats->used_dirs = atomic_read(&fg->used_dirs);
 		return;
 	}
 
@@ -662,7 +662,7 @@ static int find_group_other(struct super_block *sb, struct inode *parent,
  * block has been written back to disk.  (Yes, these values are
  * somewhat arbitrary...)
  */
-#define RECENTCY_MIN	5
+#define RECENTCY_MIN	60
 #define RECENTCY_DIRTY	300
 
 static int recently_deleted(struct super_block *sb, ext4_group_t group, int ino)
@@ -711,21 +711,34 @@ out:
 static int find_inode_bit(struct super_block *sb, ext4_group_t group,
 			  struct buffer_head *bitmap, unsigned long *ino)
 {
+	bool check_recently_deleted = EXT4_SB(sb)->s_journal == NULL;
+	unsigned long recently_deleted_ino = EXT4_INODES_PER_GROUP(sb);
+
 next:
 	*ino = ext4_find_next_zero_bit((unsigned long *)
 				       bitmap->b_data,
 				       EXT4_INODES_PER_GROUP(sb), *ino);
 	if (*ino >= EXT4_INODES_PER_GROUP(sb))
-		return 0;
+		goto not_found;
 
-	if ((EXT4_SB(sb)->s_journal == NULL) &&
-	    recently_deleted(sb, group, *ino)) {
+	if (check_recently_deleted && recently_deleted(sb, group, *ino)) {
+		recently_deleted_ino = *ino;
 		*ino = *ino + 1;
 		if (*ino < EXT4_INODES_PER_GROUP(sb))
 			goto next;
-		return 0;
+		goto not_found;
 	}
-
+	return 1;
+not_found:
+	if (recently_deleted_ino >= EXT4_INODES_PER_GROUP(sb))
+		return 0;
+	/*
+	 * Not reusing recently deleted inodes is mostly a preference. We don't
+	 * want to report ENOSPC or skew allocation patterns because of that.
+	 * So return even recently deleted inode if we could find better in the
+	 * given range.
+	 */
+	*ino = recently_deleted_ino;
 	return 1;
 }
 
@@ -926,8 +939,8 @@ repeat_in_this_group:
 		if (!handle) {
 			BUG_ON(nblocks <= 0);
 			handle = __ext4_journal_start_sb(dir->i_sb, line_no,
-							 handle_type, nblocks,
-							 0);
+				 handle_type, nblocks, 0,
+				 ext4_trans_default_revoke_credits(sb));
 			if (IS_ERR(handle)) {
 				err = PTR_ERR(handle);
 				ext4_std_error(sb, err);
@@ -1056,7 +1069,8 @@ got:
 		if (sbi->s_log_groups_per_flex) {
 			ext4_group_t f = ext4_flex_group(sbi, group);
 
-			atomic_inc(&sbi->s_flex_groups[f].used_dirs);
+			atomic_inc(&sbi_array_rcu_deref(sbi, s_flex_groups,
+							f)->used_dirs);
 		}
 	}
 	if (ext4_has_group_desc_csum(sb)) {
@@ -1079,7 +1093,8 @@ got:
 
 	if (sbi->s_log_groups_per_flex) {
 		flex_group = ext4_flex_group(sbi, group);
-		atomic_dec(&sbi->s_flex_groups[flex_group].free_inodes);
+		atomic_dec(&sbi_array_rcu_deref(sbi, s_flex_groups,
+						flex_group)->free_inodes);
 	}
 
 	inode->i_ino = ino + group * EXT4_INODES_PER_GROUP(sb);
@@ -1228,8 +1243,9 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
 	inode = ext4_iget(sb, ino, EXT4_IGET_NORMAL);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
-		ext4_error(sb, "couldn't read orphan inode %lu (err %d)",
-			   ino, err);
+		ext4_error_err(sb, -err,
+			       "couldn't read orphan inode %lu (err %d)",
+			       ino, err);
 		return inode;
 	}
 
@@ -1424,7 +1440,7 @@ int ext4_init_inode_table(struct super_block *sb, ext4_group_t group,
 	if (ret < 0)
 		goto err_out;
 	if (barrier)
-		blkdev_issue_flush(sb->s_bdev, GFP_NOFS, NULL);
+		blkdev_issue_flush(sb->s_bdev, GFP_NOFS);
 
 skip_zeroout:
 	ext4_lock_group(sb, group);
diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c
index 36699a131168..107f0043f67f 100644
--- a/fs/ext4/indirect.c
+++ b/fs/ext4/indirect.c
@@ -331,11 +331,14 @@ static int ext4_alloc_branch(handle_t *handle,
 	for (i = 0; i <= indirect_blks; i++) {
 		if (i == indirect_blks) {
 			new_blocks[i] = ext4_mb_new_blocks(handle, ar, &err);
-		} else
+		} else {
 			ar->goal = new_blocks[i] = ext4_new_meta_blocks(handle,
 					ar->inode, ar->goal,
 					ar->flags & EXT4_MB_DELALLOC_RESERVED,
 					NULL, &err);
+			/* Simplify error cleanup... */
+			branch[i+1].bh = NULL;
+		}
 		if (err) {
 			i--;
 			goto failed;
@@ -377,18 +380,25 @@ static int ext4_alloc_branch(handle_t *handle,
 	}
 	return 0;
 failed:
+	if (i == indirect_blks) {
+		/* Free data blocks */
+		ext4_free_blocks(handle, ar->inode, NULL, new_blocks[i],
+				 ar->len, 0);
+		i--;
+	}
 	for (; i >= 0; i--) {
 		/*
 		 * We want to ext4_forget() only freshly allocated indirect
-		 * blocks.  Buffer for new_blocks[i-1] is at branch[i].bh and
-		 * buffer at branch[0].bh is indirect block / inode already
-		 * existing before ext4_alloc_branch() was called.
+		 * blocks. Buffer for new_blocks[i] is at branch[i+1].bh
+		 * (buffer at branch[0].bh is indirect block / inode already
+		 * existing before ext4_alloc_branch() was called). Also
+		 * because blocks are freshly allocated, we don't need to
+		 * revoke them which is why we don't set
+		 * EXT4_FREE_BLOCKS_METADATA.
 		 */
-		if (i > 0 && i != indirect_blks && branch[i].bh)
-			ext4_forget(handle, 1, ar->inode, branch[i].bh,
-				    branch[i].bh->b_blocknr);
-		ext4_free_blocks(handle, ar->inode, NULL, new_blocks[i],
-				 (i == indirect_blks) ? ar->len : 1, 0);
+		ext4_free_blocks(handle, ar->inode, branch[i+1].bh,
+				 new_blocks[i], 1,
+				 branch[i+1].bh ? EXT4_FREE_BLOCKS_FORGET : 0);
 	}
 	return err;
 }
@@ -650,32 +660,6 @@ out:
 }
 
 /*
- * Calculate the number of metadata blocks need to reserve
- * to allocate a new block at @lblocks for non extent file based file
- */
-int ext4_ind_calc_metadata_amount(struct inode *inode, sector_t lblock)
-{
-	struct ext4_inode_info *ei = EXT4_I(inode);
-	sector_t dind_mask = ~((sector_t)EXT4_ADDR_PER_BLOCK(inode->i_sb) - 1);
-	int blk_bits;
-
-	if (lblock < EXT4_NDIR_BLOCKS)
-		return 0;
-
-	lblock -= EXT4_NDIR_BLOCKS;
-
-	if (ei->i_da_metadata_calc_len &&
-	    (lblock & dind_mask) == ei->i_da_metadata_calc_last_lblock) {
-		ei->i_da_metadata_calc_len++;
-		return 0;
-	}
-	ei->i_da_metadata_calc_last_lblock = lblock & dind_mask;
-	ei->i_da_metadata_calc_len = 1;
-	blk_bits = order_base_2(lblock);
-	return (blk_bits / EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb)) + 1;
-}
-
-/*
  * Calculate number of indirect blocks touched by mapping @nrblocks logically
  * contiguous blocks
  */
@@ -689,27 +673,63 @@ int ext4_ind_trans_blocks(struct inode *inode, int nrblocks)
 	return DIV_ROUND_UP(nrblocks, EXT4_ADDR_PER_BLOCK(inode->i_sb)) + 4;
 }
 
+static int ext4_ind_trunc_restart_fn(handle_t *handle, struct inode *inode,
+				     struct buffer_head *bh, int *dropped)
+{
+	int err;
+
+	if (bh) {
+		BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
+		err = ext4_handle_dirty_metadata(handle, inode, bh);
+		if (unlikely(err))
+			return err;
+	}
+	err = ext4_mark_inode_dirty(handle, inode);
+	if (unlikely(err))
+		return err;
+	/*
+	 * Drop i_data_sem to avoid deadlock with ext4_map_blocks.  At this
+	 * moment, get_block can be called only for blocks inside i_size since
+	 * page cache has been already dropped and writes are blocked by
+	 * i_mutex. So we can safely drop the i_data_sem here.
+	 */
+	BUG_ON(EXT4_JOURNAL(inode) == NULL);
+	ext4_discard_preallocations(inode);
+	up_write(&EXT4_I(inode)->i_data_sem);
+	*dropped = 1;
+	return 0;
+}
+
 /*
  * Truncate transactions can be complex and absolutely huge.  So we need to
  * be able to restart the transaction at a conventient checkpoint to make
  * sure we don't overflow the journal.
  *
  * Try to extend this transaction for the purposes of truncation.  If
- * extend fails, we need to propagate the failure up and restart the
- * transaction in the top-level truncate loop. --sct
- *
- * Returns 0 if we managed to create more room.  If we can't create more
- * room, and the transaction must be restarted we return 1.
+ * extend fails, we restart transaction.
  */
-static int try_to_extend_transaction(handle_t *handle, struct inode *inode)
+static int ext4_ind_truncate_ensure_credits(handle_t *handle,
+					    struct inode *inode,
+					    struct buffer_head *bh,
+					    int revoke_creds)
 {
-	if (!ext4_handle_valid(handle))
-		return 0;
-	if (ext4_handle_has_enough_credits(handle, EXT4_RESERVE_TRANS_BLOCKS+1))
-		return 0;
-	if (!ext4_journal_extend(handle, ext4_blocks_for_truncate(inode)))
-		return 0;
-	return 1;
+	int ret;
+	int dropped = 0;
+
+	ret = ext4_journal_ensure_credits_fn(handle, EXT4_RESERVE_TRANS_BLOCKS,
+			ext4_blocks_for_truncate(inode), revoke_creds,
+			ext4_ind_trunc_restart_fn(handle, inode, bh, &dropped));
+	if (dropped)
+		down_write(&EXT4_I(inode)->i_data_sem);
+	if (ret <= 0)
+		return ret;
+	if (bh) {
+		BUFFER_TRACE(bh, "retaking write access");
+		ret = ext4_journal_get_write_access(handle, bh);
+		if (unlikely(ret))
+			return ret;
+	}
+	return 0;
 }
 
 /*
@@ -844,27 +864,10 @@ static int ext4_clear_blocks(handle_t *handle, struct inode *inode,
 		return 1;
 	}
 
-	if (try_to_extend_transaction(handle, inode)) {
-		if (bh) {
-			BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
-			err = ext4_handle_dirty_metadata(handle, inode, bh);
-			if (unlikely(err))
-				goto out_err;
-		}
-		err = ext4_mark_inode_dirty(handle, inode);
-		if (unlikely(err))
-			goto out_err;
-		err = ext4_truncate_restart_trans(handle, inode,
-					ext4_blocks_for_truncate(inode));
-		if (unlikely(err))
-			goto out_err;
-		if (bh) {
-			BUFFER_TRACE(bh, "retaking write access");
-			err = ext4_journal_get_write_access(handle, bh);
-			if (unlikely(err))
-				goto out_err;
-		}
-	}
+	err = ext4_ind_truncate_ensure_credits(handle, inode, bh,
+				ext4_free_data_revoke_credits(inode, count));
+	if (err < 0)
+		goto out_err;
 
 	for (p = first; p < last; p++)
 		*p = 0;
@@ -1016,7 +1019,7 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 			 * (should be rare).
 			 */
 			if (!bh) {
-				EXT4_ERROR_INODE_BLOCK(inode, nr,
+				ext4_error_inode_block(inode, nr, EIO,
 						       "Read failure");
 				continue;
 			}
@@ -1047,11 +1050,11 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 			 */
 			if (ext4_handle_is_aborted(handle))
 				return;
-			if (try_to_extend_transaction(handle, inode)) {
-				ext4_mark_inode_dirty(handle, inode);
-				ext4_truncate_restart_trans(handle, inode,
-					    ext4_blocks_for_truncate(inode));
-			}
+			if (ext4_ind_truncate_ensure_credits(handle, inode,
+					NULL,
+					ext4_free_metadata_revoke_credits(
+							inode->i_sb, 1)) < 0)
+				return;
 
 			/*
 			 * The forget flag here is critical because if
diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c
index 2fec62d764fa..f35e289e17aa 100644
--- a/fs/ext4/inline.c
+++ b/fs/ext4/inline.c
@@ -98,9 +98,9 @@ int ext4_get_max_inline_size(struct inode *inode)
 
 	error = ext4_get_inode_loc(inode, &iloc);
 	if (error) {
-		ext4_error_inode(inode, __func__, __LINE__, 0,
-				 "can't get inode location %lu",
-				 inode->i_ino);
+		ext4_error_inode_err(inode, __func__, __LINE__, 0, -error,
+				     "can't get inode location %lu",
+				     inode->i_ino);
 		return 0;
 	}
 
@@ -849,7 +849,7 @@ out:
 
 /*
  * Prepare the write for the inline data.
- * If the the data can be written into the inode, we just read
+ * If the data can be written into the inode, we just read
  * the page and make it uptodate, and start the journal.
  * Otherwise read the page, makes it dirty so that it can be
  * handle in writepages(the i_disksize update is left to the
@@ -1761,8 +1761,9 @@ bool empty_inline_dir(struct inode *dir, int *has_inline_data)
 
 	err = ext4_get_inode_loc(dir, &iloc);
 	if (err) {
-		EXT4_ERROR_INODE(dir, "error %d getting inode %lu block",
-				 err, dir->i_ino);
+		EXT4_ERROR_INODE_ERR(dir, -err,
+				     "error %d getting inode %lu block",
+				     err, dir->i_ino);
 		return true;
 	}
 
@@ -1855,47 +1856,6 @@ out:
 	return error;
 }
 
-int ext4_inline_data_fiemap(struct inode *inode,
-			    struct fiemap_extent_info *fieinfo,
-			    int *has_inline, __u64 start, __u64 len)
-{
-	__u64 physical = 0;
-	__u64 inline_len;
-	__u32 flags = FIEMAP_EXTENT_DATA_INLINE | FIEMAP_EXTENT_NOT_ALIGNED |
-		FIEMAP_EXTENT_LAST;
-	int error = 0;
-	struct ext4_iloc iloc;
-
-	down_read(&EXT4_I(inode)->xattr_sem);
-	if (!ext4_has_inline_data(inode)) {
-		*has_inline = 0;
-		goto out;
-	}
-	inline_len = min_t(size_t, ext4_get_inline_size(inode),
-			   i_size_read(inode));
-	if (start >= inline_len)
-		goto out;
-	if (start + len < inline_len)
-		inline_len = start + len;
-	inline_len -= start;
-
-	error = ext4_get_inode_loc(inode, &iloc);
-	if (error)
-		goto out;
-
-	physical = (__u64)iloc.bh->b_blocknr << inode->i_sb->s_blocksize_bits;
-	physical += (char *)ext4_raw_inode(&iloc) - iloc.bh->b_data;
-	physical += offsetof(struct ext4_inode, i_block);
-
-	brelse(iloc.bh);
-out:
-	up_read(&EXT4_I(inode)->xattr_sem);
-	if (physical)
-		error = fiemap_fill_next_extent(fieinfo, start, physical,
-						inline_len, flags);
-	return (error < 0 ? error : 0);
-}
-
 int ext4_inline_data_truncate(struct inode *inode, int *has_inline)
 {
 	handle_t *handle;
diff --git a/fs/ext4/inode-test.c b/fs/ext4/inode-test.c
new file mode 100644
index 000000000000..d62d802c9c12
--- /dev/null
+++ b/fs/ext4/inode-test.c
@@ -0,0 +1,274 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * KUnit test of ext4 inode that verify the seconds part of [a/c/m]
+ * timestamps in ext4 inode structs are decoded correctly.
+ */
+
+#include <kunit/test.h>
+#include <linux/kernel.h>
+#include <linux/time64.h>
+
+#include "ext4.h"
+
+/*
+ * For constructing the nonnegative timestamp lower bound value.
+ * binary: 00000000 00000000 00000000 00000000
+ */
+#define LOWER_MSB_0 0L
+/*
+ * For constructing the nonnegative timestamp upper bound value.
+ * binary: 01111111 11111111 11111111 11111111
+ *
+ */
+#define UPPER_MSB_0 0x7fffffffL
+/*
+ * For constructing the negative timestamp lower bound value.
+ * binary: 10000000 00000000 00000000 00000000
+ */
+#define LOWER_MSB_1 (-(UPPER_MSB_0) - 1L)  /* avoid overflow */
+/*
+ * For constructing the negative timestamp upper bound value.
+ * binary: 11111111 11111111 11111111 11111111
+ */
+#define UPPER_MSB_1 (-1L)
+/*
+ * Upper bound for nanoseconds value supported by the encoding.
+ * binary: 00111111 11111111 11111111 11111111
+ */
+#define MAX_NANOSECONDS ((1L << 30) - 1)
+
+#define CASE_NAME_FORMAT "%s: msb:%x lower_bound:%x extra_bits: %x"
+
+#define LOWER_BOUND_NEG_NO_EXTRA_BITS_CASE\
+	"1901-12-13 Lower bound of 32bit < 0 timestamp, no extra bits"
+#define UPPER_BOUND_NEG_NO_EXTRA_BITS_CASE\
+	"1969-12-31 Upper bound of 32bit < 0 timestamp, no extra bits"
+#define LOWER_BOUND_NONNEG_NO_EXTRA_BITS_CASE\
+	"1970-01-01 Lower bound of 32bit >=0 timestamp, no extra bits"
+#define UPPER_BOUND_NONNEG_NO_EXTRA_BITS_CASE\
+	"2038-01-19 Upper bound of 32bit >=0 timestamp, no extra bits"
+#define LOWER_BOUND_NEG_LO_1_CASE\
+	"2038-01-19 Lower bound of 32bit <0 timestamp, lo extra sec bit on"
+#define UPPER_BOUND_NEG_LO_1_CASE\
+	"2106-02-07 Upper bound of 32bit <0 timestamp, lo extra sec bit on"
+#define LOWER_BOUND_NONNEG_LO_1_CASE\
+	"2106-02-07 Lower bound of 32bit >=0 timestamp, lo extra sec bit on"
+#define UPPER_BOUND_NONNEG_LO_1_CASE\
+	"2174-02-25 Upper bound of 32bit >=0 timestamp, lo extra sec bit on"
+#define LOWER_BOUND_NEG_HI_1_CASE\
+	"2174-02-25 Lower bound of 32bit <0 timestamp, hi extra sec bit on"
+#define UPPER_BOUND_NEG_HI_1_CASE\
+	"2242-03-16 Upper bound of 32bit <0 timestamp, hi extra sec bit on"
+#define LOWER_BOUND_NONNEG_HI_1_CASE\
+	"2242-03-16 Lower bound of 32bit >=0 timestamp, hi extra sec bit on"
+#define UPPER_BOUND_NONNEG_HI_1_CASE\
+	"2310-04-04 Upper bound of 32bit >=0 timestamp, hi extra sec bit on"
+#define UPPER_BOUND_NONNEG_HI_1_NS_1_CASE\
+	"2310-04-04 Upper bound of 32bit>=0 timestamp, hi extra sec bit 1. 1 ns"
+#define LOWER_BOUND_NONNEG_HI_1_NS_MAX_CASE\
+	"2378-04-22 Lower bound of 32bit>= timestamp. Extra sec bits 1. Max ns"
+#define LOWER_BOUND_NONNEG_EXTRA_BITS_1_CASE\
+	"2378-04-22 Lower bound of 32bit >=0 timestamp. All extra sec bits on"
+#define UPPER_BOUND_NONNEG_EXTRA_BITS_1_CASE\
+	"2446-05-10 Upper bound of 32bit >=0 timestamp. All extra sec bits on"
+
+struct timestamp_expectation {
+	const char *test_case_name;
+	struct timespec64 expected;
+	u32 extra_bits;
+	bool msb_set;
+	bool lower_bound;
+};
+
+static time64_t get_32bit_time(const struct timestamp_expectation * const test)
+{
+	if (test->msb_set) {
+		if (test->lower_bound)
+			return LOWER_MSB_1;
+
+		return UPPER_MSB_1;
+	}
+
+	if (test->lower_bound)
+		return LOWER_MSB_0;
+	return UPPER_MSB_0;
+}
+
+
+/*
+ *  Test data is derived from the table in the Inode Timestamps section of
+ *  Documentation/filesystems/ext4/inodes.rst.
+ */
+static void inode_test_xtimestamp_decoding(struct kunit *test)
+{
+	const struct timestamp_expectation test_data[] = {
+		{
+			.test_case_name = LOWER_BOUND_NEG_NO_EXTRA_BITS_CASE,
+			.msb_set = true,
+			.lower_bound = true,
+			.extra_bits = 0,
+			.expected = {.tv_sec = -0x80000000LL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = UPPER_BOUND_NEG_NO_EXTRA_BITS_CASE,
+			.msb_set = true,
+			.lower_bound = false,
+			.extra_bits = 0,
+			.expected = {.tv_sec = -1LL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = LOWER_BOUND_NONNEG_NO_EXTRA_BITS_CASE,
+			.msb_set = false,
+			.lower_bound = true,
+			.extra_bits = 0,
+			.expected = {0LL, 0L},
+		},
+
+		{
+			.test_case_name = UPPER_BOUND_NONNEG_NO_EXTRA_BITS_CASE,
+			.msb_set = false,
+			.lower_bound = false,
+			.extra_bits = 0,
+			.expected = {.tv_sec = 0x7fffffffLL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = LOWER_BOUND_NEG_LO_1_CASE,
+			.msb_set = true,
+			.lower_bound = true,
+			.extra_bits = 1,
+			.expected = {.tv_sec = 0x80000000LL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = UPPER_BOUND_NEG_LO_1_CASE,
+			.msb_set = true,
+			.lower_bound = false,
+			.extra_bits = 1,
+			.expected = {.tv_sec = 0xffffffffLL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = LOWER_BOUND_NONNEG_LO_1_CASE,
+			.msb_set = false,
+			.lower_bound = true,
+			.extra_bits = 1,
+			.expected = {.tv_sec = 0x100000000LL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = UPPER_BOUND_NONNEG_LO_1_CASE,
+			.msb_set = false,
+			.lower_bound = false,
+			.extra_bits = 1,
+			.expected = {.tv_sec = 0x17fffffffLL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = LOWER_BOUND_NEG_HI_1_CASE,
+			.msb_set = true,
+			.lower_bound = true,
+			.extra_bits =  2,
+			.expected = {.tv_sec = 0x180000000LL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = UPPER_BOUND_NEG_HI_1_CASE,
+			.msb_set = true,
+			.lower_bound = false,
+			.extra_bits = 2,
+			.expected = {.tv_sec = 0x1ffffffffLL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = LOWER_BOUND_NONNEG_HI_1_CASE,
+			.msb_set = false,
+			.lower_bound = true,
+			.extra_bits = 2,
+			.expected = {.tv_sec = 0x200000000LL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = UPPER_BOUND_NONNEG_HI_1_CASE,
+			.msb_set = false,
+			.lower_bound = false,
+			.extra_bits = 2,
+			.expected = {.tv_sec = 0x27fffffffLL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = UPPER_BOUND_NONNEG_HI_1_NS_1_CASE,
+			.msb_set = false,
+			.lower_bound = false,
+			.extra_bits = 6,
+			.expected = {.tv_sec = 0x27fffffffLL, .tv_nsec = 1L},
+		},
+
+		{
+			.test_case_name = LOWER_BOUND_NONNEG_HI_1_NS_MAX_CASE,
+			.msb_set = false,
+			.lower_bound = true,
+			.extra_bits = 0xFFFFFFFF,
+			.expected = {.tv_sec = 0x300000000LL,
+				     .tv_nsec = MAX_NANOSECONDS},
+		},
+
+		{
+			.test_case_name = LOWER_BOUND_NONNEG_EXTRA_BITS_1_CASE,
+			.msb_set = false,
+			.lower_bound = true,
+			.extra_bits = 3,
+			.expected = {.tv_sec = 0x300000000LL, .tv_nsec = 0L},
+		},
+
+		{
+			.test_case_name = UPPER_BOUND_NONNEG_EXTRA_BITS_1_CASE,
+			.msb_set = false,
+			.lower_bound = false,
+			.extra_bits = 3,
+			.expected = {.tv_sec = 0x37fffffffLL, .tv_nsec = 0L},
+		}
+	};
+
+	struct timespec64 timestamp;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(test_data); ++i) {
+		timestamp.tv_sec = get_32bit_time(&test_data[i]);
+		ext4_decode_extra_time(&timestamp,
+				       cpu_to_le32(test_data[i].extra_bits));
+
+		KUNIT_EXPECT_EQ_MSG(test,
+				    test_data[i].expected.tv_sec,
+				    timestamp.tv_sec,
+				    CASE_NAME_FORMAT,
+				    test_data[i].test_case_name,
+				    test_data[i].msb_set,
+				    test_data[i].lower_bound,
+				    test_data[i].extra_bits);
+		KUNIT_EXPECT_EQ_MSG(test,
+				    test_data[i].expected.tv_nsec,
+				    timestamp.tv_nsec,
+				    CASE_NAME_FORMAT,
+				    test_data[i].test_case_name,
+				    test_data[i].msb_set,
+				    test_data[i].lower_bound,
+				    test_data[i].extra_bits);
+	}
+}
+
+static struct kunit_case ext4_inode_test_cases[] = {
+	KUNIT_CASE(inode_test_xtimestamp_decoding),
+	{}
+};
+
+static struct kunit_suite ext4_inode_test_suite = {
+	.name = "ext4_inode_test",
+	.test_cases = ext4_inode_test_cases,
+};
+
+kunit_test_suites(&ext4_inode_test_suite);
+
+MODULE_LICENSE("GPL v2");
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 516faa280ced..52be85f96159 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -48,8 +48,6 @@
 
 #include <trace/events/ext4.h>
 
-#define MPAGE_DA_EXTENT_TAIL 0x01
-
 static __u32 ext4_inode_csum(struct inode *inode, struct ext4_inode *raw,
 			      struct ext4_inode_info *ei)
 {
@@ -164,39 +162,18 @@ int ext4_inode_is_fast_symlink(struct inode *inode)
 }
 
 /*
- * Restart the transaction associated with *handle.  This does a commit,
- * so before we call here everything must be consistently dirtied against
- * this transaction.
- */
-int ext4_truncate_restart_trans(handle_t *handle, struct inode *inode,
-				 int nblocks)
-{
-	int ret;
-
-	/*
-	 * Drop i_data_sem to avoid deadlock with ext4_map_blocks.  At this
-	 * moment, get_block can be called only for blocks inside i_size since
-	 * page cache has been already dropped and writes are blocked by
-	 * i_mutex. So we can safely drop the i_data_sem here.
-	 */
-	BUG_ON(EXT4_JOURNAL(inode) == NULL);
-	jbd_debug(2, "restarting handle %p\n", handle);
-	up_write(&EXT4_I(inode)->i_data_sem);
-	ret = ext4_journal_restart(handle, nblocks);
-	down_write(&EXT4_I(inode)->i_data_sem);
-	ext4_discard_preallocations(inode);
-
-	return ret;
-}
-
-/*
  * Called at the last iput() if i_nlink is zero.
  */
 void ext4_evict_inode(struct inode *inode)
 {
 	handle_t *handle;
 	int err;
-	int extra_credits = 3;
+	/*
+	 * Credits for final inode cleanup and freeing:
+	 * sb + inode (ext4_orphan_del()), block bitmap, group descriptor
+	 * (xattr block freeing), bitmap, group descriptor (inode freeing)
+	 */
+	int extra_credits = 6;
 	struct ext4_xattr_inode_array *ea_inode_array = NULL;
 
 	trace_ext4_evict_inode(inode);
@@ -252,8 +229,12 @@ void ext4_evict_inode(struct inode *inode)
 	if (!IS_NOQUOTA(inode))
 		extra_credits += EXT4_MAXQUOTAS_DEL_BLOCKS(inode->i_sb);
 
+	/*
+	 * Block bitmap, group descriptor, and inode are accounted in both
+	 * ext4_blocks_for_truncate() and extra_credits. So subtract 3.
+	 */
 	handle = ext4_journal_start(inode, EXT4_HT_TRUNCATE,
-				 ext4_blocks_for_truncate(inode)+extra_credits);
+			 ext4_blocks_for_truncate(inode) + extra_credits - 3);
 	if (IS_ERR(handle)) {
 		ext4_std_error(inode->i_sb, PTR_ERR(handle));
 		/*
@@ -288,9 +269,9 @@ void ext4_evict_inode(struct inode *inode)
 	if (inode->i_blocks) {
 		err = ext4_truncate(inode);
 		if (err) {
-			ext4_error(inode->i_sb,
-				   "couldn't truncate inode %lu (err %d)",
-				   inode->i_ino, err);
+			ext4_error_err(inode->i_sb, -err,
+				       "couldn't truncate inode %lu (err %d)",
+				       inode->i_ino, err);
 			goto stop_handle;
 		}
 	}
@@ -419,7 +400,7 @@ int ext4_issue_zeroout(struct inode *inode, ext4_lblk_t lblk, ext4_fsblk_t pblk,
 {
 	int ret;
 
-	if (IS_ENCRYPTED(inode))
+	if (IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode))
 		return fscrypt_zeroout_range(inode, lblk, pblk, len);
 
 	ret = sb_issue_zeroout(inode->i_sb, pblk, len, GFP_NOFS);
@@ -827,136 +808,6 @@ int ext4_get_block_unwritten(struct inode *inode, sector_t iblock,
 #define DIO_MAX_BLOCKS 4096
 
 /*
- * Get blocks function for the cases that need to start a transaction -
- * generally difference cases of direct IO and DAX IO. It also handles retries
- * in case of ENOSPC.
- */
-static int ext4_get_block_trans(struct inode *inode, sector_t iblock,
-				struct buffer_head *bh_result, int flags)
-{
-	int dio_credits;
-	handle_t *handle;
-	int retries = 0;
-	int ret;
-
-	/* Trim mapping request to maximum we can map at once for DIO */
-	if (bh_result->b_size >> inode->i_blkbits > DIO_MAX_BLOCKS)
-		bh_result->b_size = DIO_MAX_BLOCKS << inode->i_blkbits;
-	dio_credits = ext4_chunk_trans_blocks(inode,
-				      bh_result->b_size >> inode->i_blkbits);
-retry:
-	handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, dio_credits);
-	if (IS_ERR(handle))
-		return PTR_ERR(handle);
-
-	ret = _ext4_get_block(inode, iblock, bh_result, flags);
-	ext4_journal_stop(handle);
-
-	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
-		goto retry;
-	return ret;
-}
-
-/* Get block function for DIO reads and writes to inodes without extents */
-int ext4_dio_get_block(struct inode *inode, sector_t iblock,
-		       struct buffer_head *bh, int create)
-{
-	/* We don't expect handle for direct IO */
-	WARN_ON_ONCE(ext4_journal_current_handle());
-
-	if (!create)
-		return _ext4_get_block(inode, iblock, bh, 0);
-	return ext4_get_block_trans(inode, iblock, bh, EXT4_GET_BLOCKS_CREATE);
-}
-
-/*
- * Get block function for AIO DIO writes when we create unwritten extent if
- * blocks are not allocated yet. The extent will be converted to written
- * after IO is complete.
- */
-static int ext4_dio_get_block_unwritten_async(struct inode *inode,
-		sector_t iblock, struct buffer_head *bh_result,	int create)
-{
-	int ret;
-
-	/* We don't expect handle for direct IO */
-	WARN_ON_ONCE(ext4_journal_current_handle());
-
-	ret = ext4_get_block_trans(inode, iblock, bh_result,
-				   EXT4_GET_BLOCKS_IO_CREATE_EXT);
-
-	/*
-	 * When doing DIO using unwritten extents, we need io_end to convert
-	 * unwritten extents to written on IO completion. We allocate io_end
-	 * once we spot unwritten extent and store it in b_private. Generic
-	 * DIO code keeps b_private set and furthermore passes the value to
-	 * our completion callback in 'private' argument.
-	 */
-	if (!ret && buffer_unwritten(bh_result)) {
-		if (!bh_result->b_private) {
-			ext4_io_end_t *io_end;
-
-			io_end = ext4_init_io_end(inode, GFP_KERNEL);
-			if (!io_end)
-				return -ENOMEM;
-			bh_result->b_private = io_end;
-			ext4_set_io_unwritten_flag(inode, io_end);
-		}
-		set_buffer_defer_completion(bh_result);
-	}
-
-	return ret;
-}
-
-/*
- * Get block function for non-AIO DIO writes when we create unwritten extent if
- * blocks are not allocated yet. The extent will be converted to written
- * after IO is complete by ext4_direct_IO_write().
- */
-static int ext4_dio_get_block_unwritten_sync(struct inode *inode,
-		sector_t iblock, struct buffer_head *bh_result,	int create)
-{
-	int ret;
-
-	/* We don't expect handle for direct IO */
-	WARN_ON_ONCE(ext4_journal_current_handle());
-
-	ret = ext4_get_block_trans(inode, iblock, bh_result,
-				   EXT4_GET_BLOCKS_IO_CREATE_EXT);
-
-	/*
-	 * Mark inode as having pending DIO writes to unwritten extents.
-	 * ext4_direct_IO_write() checks this flag and converts extents to
-	 * written.
-	 */
-	if (!ret && buffer_unwritten(bh_result))
-		ext4_set_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN);
-
-	return ret;
-}
-
-static int ext4_dio_get_block_overwrite(struct inode *inode, sector_t iblock,
-		   struct buffer_head *bh_result, int create)
-{
-	int ret;
-
-	ext4_debug("ext4_dio_get_block_overwrite: inode %lu, create flag %d\n",
-		   inode->i_ino, create);
-	/* We don't expect handle for direct IO */
-	WARN_ON_ONCE(ext4_journal_current_handle());
-
-	ret = _ext4_get_block(inode, iblock, bh_result, 0);
-	/*
-	 * Blocks should have been preallocated! ext4_file_write_iter() checks
-	 * that.
-	 */
-	WARN_ON_ONCE(!buffer_mapped(bh_result) || buffer_unwritten(bh_result));
-
-	return ret;
-}
-
-
-/*
  * `handle' can be NULL if create is zero
  */
 struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode,
@@ -2122,7 +1973,7 @@ static int ext4_writepage(struct page *page,
 	bool keep_towrite = false;
 
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) {
-		ext4_invalidatepage(page, 0, PAGE_SIZE);
+		inode->i_mapping->a_ops->invalidatepage(page, 0, PAGE_SIZE);
 		unlock_page(page);
 		return -EIO;
 	}
@@ -2341,6 +2192,79 @@ static int mpage_process_page_bufs(struct mpage_da_data *mpd,
 }
 
 /*
+ * mpage_process_page - update page buffers corresponding to changed extent and
+ *		       may submit fully mapped page for IO
+ *
+ * @mpd		- description of extent to map, on return next extent to map
+ * @m_lblk	- logical block mapping.
+ * @m_pblk	- corresponding physical mapping.
+ * @map_bh	- determines on return whether this page requires any further
+ *		  mapping or not.
+ * Scan given page buffers corresponding to changed extent and update buffer
+ * state according to new extent state.
+ * We map delalloc buffers to their physical location, clear unwritten bits.
+ * If the given page is not fully mapped, we update @map to the next extent in
+ * the given page that needs mapping & return @map_bh as true.
+ */
+static int mpage_process_page(struct mpage_da_data *mpd, struct page *page,
+			      ext4_lblk_t *m_lblk, ext4_fsblk_t *m_pblk,
+			      bool *map_bh)
+{
+	struct buffer_head *head, *bh;
+	ext4_io_end_t *io_end = mpd->io_submit.io_end;
+	ext4_lblk_t lblk = *m_lblk;
+	ext4_fsblk_t pblock = *m_pblk;
+	int err = 0;
+	int blkbits = mpd->inode->i_blkbits;
+	ssize_t io_end_size = 0;
+	struct ext4_io_end_vec *io_end_vec = ext4_last_io_end_vec(io_end);
+
+	bh = head = page_buffers(page);
+	do {
+		if (lblk < mpd->map.m_lblk)
+			continue;
+		if (lblk >= mpd->map.m_lblk + mpd->map.m_len) {
+			/*
+			 * Buffer after end of mapped extent.
+			 * Find next buffer in the page to map.
+			 */
+			mpd->map.m_len = 0;
+			mpd->map.m_flags = 0;
+			io_end_vec->size += io_end_size;
+			io_end_size = 0;
+
+			err = mpage_process_page_bufs(mpd, head, bh, lblk);
+			if (err > 0)
+				err = 0;
+			if (!err && mpd->map.m_len && mpd->map.m_lblk > lblk) {
+				io_end_vec = ext4_alloc_io_end_vec(io_end);
+				if (IS_ERR(io_end_vec)) {
+					err = PTR_ERR(io_end_vec);
+					goto out;
+				}
+				io_end_vec->offset = mpd->map.m_lblk << blkbits;
+			}
+			*map_bh = true;
+			goto out;
+		}
+		if (buffer_delay(bh)) {
+			clear_buffer_delay(bh);
+			bh->b_blocknr = pblock++;
+		}
+		clear_buffer_unwritten(bh);
+		io_end_size += (1 << blkbits);
+	} while (lblk++, (bh = bh->b_this_page) != head);
+
+	io_end_vec->size += io_end_size;
+	io_end_size = 0;
+	*map_bh = false;
+out:
+	*m_lblk = lblk;
+	*m_pblk = pblock;
+	return err;
+}
+
+/*
  * mpage_map_buffers - update buffers corresponding to changed extent and
  *		       submit fully mapped pages for IO
  *
@@ -2359,12 +2283,12 @@ static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd)
 	struct pagevec pvec;
 	int nr_pages, i;
 	struct inode *inode = mpd->inode;
-	struct buffer_head *head, *bh;
 	int bpp_bits = PAGE_SHIFT - inode->i_blkbits;
 	pgoff_t start, end;
 	ext4_lblk_t lblk;
-	sector_t pblock;
+	ext4_fsblk_t pblock;
 	int err;
+	bool map_bh = false;
 
 	start = mpd->map.m_lblk >> bpp_bits;
 	end = (mpd->map.m_lblk + mpd->map.m_len - 1) >> bpp_bits;
@@ -2380,50 +2304,19 @@ static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd)
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
 
-			bh = head = page_buffers(page);
-			do {
-				if (lblk < mpd->map.m_lblk)
-					continue;
-				if (lblk >= mpd->map.m_lblk + mpd->map.m_len) {
-					/*
-					 * Buffer after end of mapped extent.
-					 * Find next buffer in the page to map.
-					 */
-					mpd->map.m_len = 0;
-					mpd->map.m_flags = 0;
-					/*
-					 * FIXME: If dioread_nolock supports
-					 * blocksize < pagesize, we need to make
-					 * sure we add size mapped so far to
-					 * io_end->size as the following call
-					 * can submit the page for IO.
-					 */
-					err = mpage_process_page_bufs(mpd, head,
-								      bh, lblk);
-					pagevec_release(&pvec);
-					if (err > 0)
-						err = 0;
-					return err;
-				}
-				if (buffer_delay(bh)) {
-					clear_buffer_delay(bh);
-					bh->b_blocknr = pblock++;
-				}
-				clear_buffer_unwritten(bh);
-			} while (lblk++, (bh = bh->b_this_page) != head);
-
+			err = mpage_process_page(mpd, page, &lblk, &pblock,
+						 &map_bh);
 			/*
-			 * FIXME: This is going to break if dioread_nolock
-			 * supports blocksize < pagesize as we will try to
-			 * convert potentially unmapped parts of inode.
+			 * If map_bh is true, means page may require further bh
+			 * mapping, or maybe the page was submitted for IO.
+			 * So we return to call further extent mapping.
 			 */
-			mpd->io_submit.io_end->size += PAGE_SIZE;
+			if (err < 0 || map_bh == true)
+				goto out;
 			/* Page fully mapped - let IO run! */
 			err = mpage_submit_page(mpd, page);
-			if (err < 0) {
-				pagevec_release(&pvec);
-				return err;
-			}
+			if (err < 0)
+				goto out;
 		}
 		pagevec_release(&pvec);
 	}
@@ -2431,6 +2324,9 @@ static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd)
 	mpd->map.m_len = 0;
 	mpd->map.m_flags = 0;
 	return 0;
+out:
+	pagevec_release(&pvec);
+	return err;
 }
 
 static int mpage_map_one_extent(handle_t *handle, struct mpage_da_data *mpd)
@@ -2510,9 +2406,13 @@ static int mpage_map_and_submit_extent(handle_t *handle,
 	int err;
 	loff_t disksize;
 	int progress = 0;
+	ext4_io_end_t *io_end = mpd->io_submit.io_end;
+	struct ext4_io_end_vec *io_end_vec;
 
-	mpd->io_submit.io_end->offset =
-				((loff_t)map->m_lblk) << inode->i_blkbits;
+	io_end_vec = ext4_alloc_io_end_vec(io_end);
+	if (IS_ERR(io_end_vec))
+		return PTR_ERR(io_end_vec);
+	io_end_vec->offset = ((loff_t)map->m_lblk) << inode->i_blkbits;
 	do {
 		err = mpage_map_one_extent(handle, mpd);
 		if (err < 0) {
@@ -2564,7 +2464,7 @@ update_disksize:
 	 * truncate are avoided by checking i_size under i_data_sem.
 	 */
 	disksize = ((loff_t)mpd->first_page) << PAGE_SHIFT;
-	if (disksize > EXT4_I(inode)->i_disksize) {
+	if (disksize > READ_ONCE(EXT4_I(inode)->i_disksize)) {
 		int err2;
 		loff_t i_size;
 
@@ -2576,10 +2476,11 @@ update_disksize:
 			EXT4_I(inode)->i_disksize = disksize;
 		up_write(&EXT4_I(inode)->i_data_sem);
 		err2 = ext4_mark_inode_dirty(handle, inode);
-		if (err2)
-			ext4_error(inode->i_sb,
-				   "Failed to mark inode %lu dirty",
-				   inode->i_ino);
+		if (err2) {
+			ext4_error_err(inode->i_sb, -err2,
+				       "Failed to mark inode %lu dirty",
+				       inode->i_ino);
+		}
 		if (!err)
 			err = err2;
 	}
@@ -2725,7 +2626,7 @@ static int ext4_writepages(struct address_space *mapping,
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
 		return -EIO;
 
-	percpu_down_read(&sbi->s_journal_flag_rwsem);
+	percpu_down_read(&sbi->s_writepages_rwsem);
 	trace_ext4_writepages(inode, wbc);
 
 	/*
@@ -2946,7 +2847,7 @@ unplug:
 out_writepages:
 	trace_ext4_writepages_result(inode, wbc, ret,
 				     nr_to_write - wbc->nr_to_write);
-	percpu_up_read(&sbi->s_journal_flag_rwsem);
+	percpu_up_read(&sbi->s_writepages_rwsem);
 	return ret;
 }
 
@@ -2961,13 +2862,13 @@ static int ext4_dax_writepages(struct address_space *mapping,
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
 		return -EIO;
 
-	percpu_down_read(&sbi->s_journal_flag_rwsem);
+	percpu_down_read(&sbi->s_writepages_rwsem);
 	trace_ext4_writepages(inode, wbc);
 
-	ret = dax_writeback_mapping_range(mapping, inode->i_sb->s_bdev, wbc);
+	ret = dax_writeback_mapping_range(mapping, sbi->s_daxdev, wbc);
 	trace_ext4_writepages_result(inode, wbc, ret,
 				     nr_to_write - wbc->nr_to_write);
-	percpu_up_read(&sbi->s_journal_flag_rwsem);
+	percpu_up_read(&sbi->s_writepages_rwsem);
 	return ret;
 }
 
@@ -3309,7 +3210,7 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block)
 			return 0;
 	}
 
-	return generic_block_bmap(mapping, block, ext4_get_block);
+	return iomap_bmap(mapping, block, &ext4_iomap_ops);
 }
 
 static int ext4_readpage(struct file *file, struct page *page)
@@ -3323,23 +3224,20 @@ static int ext4_readpage(struct file *file, struct page *page)
 		ret = ext4_readpage_inline(inode, page);
 
 	if (ret == -EAGAIN)
-		return ext4_mpage_readpages(page->mapping, NULL, page, 1,
-						false);
+		return ext4_mpage_readpages(inode, NULL, page);
 
 	return ret;
 }
 
-static int
-ext4_readpages(struct file *file, struct address_space *mapping,
-		struct list_head *pages, unsigned nr_pages)
+static void ext4_readahead(struct readahead_control *rac)
 {
-	struct inode *inode = mapping->host;
+	struct inode *inode = rac->mapping->host;
 
-	/* If the file has inline data, no need to do readpages. */
+	/* If the file has inline data, no need to do readahead. */
 	if (ext4_has_inline_data(inode))
-		return 0;
+		return;
 
-	return ext4_mpage_readpages(mapping, pages, NULL, nr_pages, true);
+	ext4_mpage_readpages(inode, rac, NULL);
 }
 
 static void ext4_invalidatepage(struct page *page, unsigned int offset,
@@ -3406,473 +3304,276 @@ static bool ext4_inode_datasync_dirty(struct inode *inode)
 	return inode->i_state & I_DIRTY_DATASYNC;
 }
 
-static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
-			    unsigned flags, struct iomap *iomap)
+static void ext4_set_iomap(struct inode *inode, struct iomap *iomap,
+			   struct ext4_map_blocks *map, loff_t offset,
+			   loff_t length)
 {
-	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
-	unsigned int blkbits = inode->i_blkbits;
-	unsigned long first_block, last_block;
-	struct ext4_map_blocks map;
-	bool delalloc = false;
-	int ret;
-
-	if ((offset >> blkbits) > EXT4_MAX_LOGICAL_BLOCK)
-		return -EINVAL;
-	first_block = offset >> blkbits;
-	last_block = min_t(loff_t, (offset + length - 1) >> blkbits,
-			   EXT4_MAX_LOGICAL_BLOCK);
-
-	if (flags & IOMAP_REPORT) {
-		if (ext4_has_inline_data(inode)) {
-			ret = ext4_inline_data_iomap(inode, iomap);
-			if (ret != -EAGAIN) {
-				if (ret == 0 && offset >= iomap->length)
-					ret = -ENOENT;
-				return ret;
-			}
-		}
-	} else {
-		if (WARN_ON_ONCE(ext4_has_inline_data(inode)))
-			return -ERANGE;
-	}
-
-	map.m_lblk = first_block;
-	map.m_len = last_block - first_block + 1;
-
-	if (flags & IOMAP_REPORT) {
-		ret = ext4_map_blocks(NULL, inode, &map, 0);
-		if (ret < 0)
-			return ret;
-
-		if (ret == 0) {
-			ext4_lblk_t end = map.m_lblk + map.m_len - 1;
-			struct extent_status es;
-
-			ext4_es_find_extent_range(inode, &ext4_es_is_delayed,
-						  map.m_lblk, end, &es);
+	u8 blkbits = inode->i_blkbits;
 
-			if (!es.es_len || es.es_lblk > end) {
-				/* entire range is a hole */
-			} else if (es.es_lblk > map.m_lblk) {
-				/* range starts with a hole */
-				map.m_len = es.es_lblk - map.m_lblk;
-			} else {
-				ext4_lblk_t offs = 0;
-
-				if (es.es_lblk < map.m_lblk)
-					offs = map.m_lblk - es.es_lblk;
-				map.m_lblk = es.es_lblk + offs;
-				map.m_len = es.es_len - offs;
-				delalloc = true;
-			}
-		}
-	} else if (flags & IOMAP_WRITE) {
-		int dio_credits;
-		handle_t *handle;
-		int retries = 0;
+	/*
+	 * Writes that span EOF might trigger an I/O size update on completion,
+	 * so consider them to be dirty for the purpose of O_DSYNC, even if
+	 * there is no other metadata changes being made or are pending.
+	 */
+	iomap->flags = 0;
+	if (ext4_inode_datasync_dirty(inode) ||
+	    offset + length > i_size_read(inode))
+		iomap->flags |= IOMAP_F_DIRTY;
 
-		/* Trim mapping request to maximum we can map at once for DIO */
-		if (map.m_len > DIO_MAX_BLOCKS)
-			map.m_len = DIO_MAX_BLOCKS;
-		dio_credits = ext4_chunk_trans_blocks(inode, map.m_len);
-retry:
-		/*
-		 * Either we allocate blocks and then we don't get unwritten
-		 * extent so we have reserved enough credits, or the blocks
-		 * are already allocated and unwritten and in that case
-		 * extent conversion fits in the credits as well.
-		 */
-		handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS,
-					    dio_credits);
-		if (IS_ERR(handle))
-			return PTR_ERR(handle);
+	if (map->m_flags & EXT4_MAP_NEW)
+		iomap->flags |= IOMAP_F_NEW;
 
-		ret = ext4_map_blocks(handle, inode, &map,
-				      EXT4_GET_BLOCKS_CREATE_ZERO);
-		if (ret < 0) {
-			ext4_journal_stop(handle);
-			if (ret == -ENOSPC &&
-			    ext4_should_retry_alloc(inode->i_sb, &retries))
-				goto retry;
-			return ret;
-		}
+	iomap->bdev = inode->i_sb->s_bdev;
+	iomap->dax_dev = EXT4_SB(inode->i_sb)->s_daxdev;
+	iomap->offset = (u64) map->m_lblk << blkbits;
+	iomap->length = (u64) map->m_len << blkbits;
 
-		/*
-		 * If we added blocks beyond i_size, we need to make sure they
-		 * will get truncated if we crash before updating i_size in
-		 * ext4_iomap_end(). For faults we don't need to do that (and
-		 * even cannot because for orphan list operations inode_lock is
-		 * required) - if we happen to instantiate block beyond i_size,
-		 * it is because we race with truncate which has already added
-		 * the inode to the orphan list.
-		 */
-		if (!(flags & IOMAP_FAULT) && first_block + map.m_len >
-		    (i_size_read(inode) + (1 << blkbits) - 1) >> blkbits) {
-			int err;
+	if ((map->m_flags & EXT4_MAP_MAPPED) &&
+	    !ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
+		iomap->flags |= IOMAP_F_MERGED;
 
-			err = ext4_orphan_add(handle, inode);
-			if (err < 0) {
-				ext4_journal_stop(handle);
-				return err;
-			}
-		}
-		ext4_journal_stop(handle);
+	/*
+	 * Flags passed to ext4_map_blocks() for direct I/O writes can result
+	 * in m_flags having both EXT4_MAP_MAPPED and EXT4_MAP_UNWRITTEN bits
+	 * set. In order for any allocated unwritten extents to be converted
+	 * into written extents correctly within the ->end_io() handler, we
+	 * need to ensure that the iomap->type is set appropriately. Hence, the
+	 * reason why we need to check whether the EXT4_MAP_UNWRITTEN bit has
+	 * been set first.
+	 */
+	if (map->m_flags & EXT4_MAP_UNWRITTEN) {
+		iomap->type = IOMAP_UNWRITTEN;
+		iomap->addr = (u64) map->m_pblk << blkbits;
+	} else if (map->m_flags & EXT4_MAP_MAPPED) {
+		iomap->type = IOMAP_MAPPED;
+		iomap->addr = (u64) map->m_pblk << blkbits;
 	} else {
-		ret = ext4_map_blocks(NULL, inode, &map, 0);
-		if (ret < 0)
-			return ret;
-	}
-
-	iomap->flags = 0;
-	if (ext4_inode_datasync_dirty(inode))
-		iomap->flags |= IOMAP_F_DIRTY;
-	iomap->bdev = inode->i_sb->s_bdev;
-	iomap->dax_dev = sbi->s_daxdev;
-	iomap->offset = (u64)first_block << blkbits;
-	iomap->length = (u64)map.m_len << blkbits;
-
-	if (ret == 0) {
-		iomap->type = delalloc ? IOMAP_DELALLOC : IOMAP_HOLE;
+		iomap->type = IOMAP_HOLE;
 		iomap->addr = IOMAP_NULL_ADDR;
-	} else {
-		if (map.m_flags & EXT4_MAP_MAPPED) {
-			iomap->type = IOMAP_MAPPED;
-		} else if (map.m_flags & EXT4_MAP_UNWRITTEN) {
-			iomap->type = IOMAP_UNWRITTEN;
-		} else {
-			WARN_ON_ONCE(1);
-			return -EIO;
-		}
-		iomap->addr = (u64)map.m_pblk << blkbits;
 	}
-
-	if (map.m_flags & EXT4_MAP_NEW)
-		iomap->flags |= IOMAP_F_NEW;
-
-	return 0;
 }
 
-static int ext4_iomap_end(struct inode *inode, loff_t offset, loff_t length,
-			  ssize_t written, unsigned flags, struct iomap *iomap)
+static int ext4_iomap_alloc(struct inode *inode, struct ext4_map_blocks *map,
+			    unsigned int flags)
 {
-	int ret = 0;
 	handle_t *handle;
-	int blkbits = inode->i_blkbits;
-	bool truncate = false;
+	u8 blkbits = inode->i_blkbits;
+	int ret, dio_credits, m_flags = 0, retries = 0;
 
-	if (!(flags & IOMAP_WRITE) || (flags & IOMAP_FAULT))
-		return 0;
+	/*
+	 * Trim the mapping request to the maximum value that we can map at
+	 * once for direct I/O.
+	 */
+	if (map->m_len > DIO_MAX_BLOCKS)
+		map->m_len = DIO_MAX_BLOCKS;
+	dio_credits = ext4_chunk_trans_blocks(inode, map->m_len);
 
-	handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
-	if (IS_ERR(handle)) {
-		ret = PTR_ERR(handle);
-		goto orphan_del;
-	}
-	if (ext4_update_inode_size(inode, offset + written))
-		ext4_mark_inode_dirty(handle, inode);
+retry:
 	/*
-	 * We may need to truncate allocated but not written blocks beyond EOF.
+	 * Either we allocate blocks and then don't get an unwritten extent, so
+	 * in that case we have reserved enough credits. Or, the blocks are
+	 * already allocated and unwritten. In that case, the extent conversion
+	 * fits into the credits as well.
 	 */
-	if (iomap->offset + iomap->length > 
-	    ALIGN(inode->i_size, 1 << blkbits)) {
-		ext4_lblk_t written_blk, end_blk;
+	handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, dio_credits);
+	if (IS_ERR(handle))
+		return PTR_ERR(handle);
 
-		written_blk = (offset + written) >> blkbits;
-		end_blk = (offset + length) >> blkbits;
-		if (written_blk < end_blk && ext4_can_truncate(inode))
-			truncate = true;
-	}
 	/*
-	 * Remove inode from orphan list if we were extending a inode and
-	 * everything went fine.
+	 * DAX and direct I/O are the only two operations that are currently
+	 * supported with IOMAP_WRITE.
 	 */
-	if (!truncate && inode->i_nlink &&
-	    !list_empty(&EXT4_I(inode)->i_orphan))
-		ext4_orphan_del(handle, inode);
+	WARN_ON(!IS_DAX(inode) && !(flags & IOMAP_DIRECT));
+	if (IS_DAX(inode))
+		m_flags = EXT4_GET_BLOCKS_CREATE_ZERO;
+	/*
+	 * We use i_size instead of i_disksize here because delalloc writeback
+	 * can complete at any point during the I/O and subsequently push the
+	 * i_disksize out to i_size. This could be beyond where direct I/O is
+	 * happening and thus expose allocated blocks to direct I/O reads.
+	 */
+	else if ((map->m_lblk * (1 << blkbits)) >= i_size_read(inode))
+		m_flags = EXT4_GET_BLOCKS_CREATE;
+	else if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
+		m_flags = EXT4_GET_BLOCKS_IO_CREATE_EXT;
+
+	ret = ext4_map_blocks(handle, inode, map, m_flags);
+
+	/*
+	 * We cannot fill holes in indirect tree based inodes as that could
+	 * expose stale data in the case of a crash. Use the magic error code
+	 * to fallback to buffered I/O.
+	 */
+	if (!m_flags && !ret)
+		ret = -ENOTBLK;
+
 	ext4_journal_stop(handle);
-	if (truncate) {
-		ext4_truncate_failed_write(inode);
-orphan_del:
-		/*
-		 * If truncate failed early the inode might still be on the
-		 * orphan list; we need to make sure the inode is removed from
-		 * the orphan list in that case.
-		 */
-		if (inode->i_nlink)
-			ext4_orphan_del(NULL, inode);
-	}
+	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
+		goto retry;
+
 	return ret;
 }
 
-const struct iomap_ops ext4_iomap_ops = {
-	.iomap_begin		= ext4_iomap_begin,
-	.iomap_end		= ext4_iomap_end,
-};
 
-static int ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
-			    ssize_t size, void *private)
+static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
+		unsigned flags, struct iomap *iomap, struct iomap *srcmap)
 {
-        ext4_io_end_t *io_end = private;
+	int ret;
+	struct ext4_map_blocks map;
+	u8 blkbits = inode->i_blkbits;
 
-	/* if not async direct IO just return */
-	if (!io_end)
-		return 0;
+	if ((offset >> blkbits) > EXT4_MAX_LOGICAL_BLOCK)
+		return -EINVAL;
 
-	ext_debug("ext4_end_io_dio(): io_end 0x%p "
-		  "for inode %lu, iocb 0x%p, offset %llu, size %zd\n",
-		  io_end, io_end->inode->i_ino, iocb, offset, size);
+	if (WARN_ON_ONCE(ext4_has_inline_data(inode)))
+		return -ERANGE;
 
 	/*
-	 * Error during AIO DIO. We cannot convert unwritten extents as the
-	 * data was not written. Just clear the unwritten flag and drop io_end.
+	 * Calculate the first and last logical blocks respectively.
 	 */
-	if (size <= 0) {
-		ext4_clear_io_unwritten_flag(io_end);
-		size = 0;
-	}
-	io_end->offset = offset;
-	io_end->size = size;
-	ext4_put_io_end(io_end);
+	map.m_lblk = offset >> blkbits;
+	map.m_len = min_t(loff_t, (offset + length - 1) >> blkbits,
+			  EXT4_MAX_LOGICAL_BLOCK) - map.m_lblk + 1;
+
+	if (flags & IOMAP_WRITE)
+		ret = ext4_iomap_alloc(inode, &map, flags);
+	else
+		ret = ext4_map_blocks(NULL, inode, &map, 0);
+
+	if (ret < 0)
+		return ret;
+
+	ext4_set_iomap(inode, iomap, &map, offset, length);
 
 	return 0;
 }
 
-/*
- * Handling of direct IO writes.
- *
- * For ext4 extent files, ext4 will do direct-io write even to holes,
- * preallocated extents, and those write extend the file, no need to
- * fall back to buffered IO.
- *
- * For holes, we fallocate those blocks, mark them as unwritten
- * If those blocks were preallocated, we mark sure they are split, but
- * still keep the range to write as unwritten.
- *
- * The unwritten extents will be converted to written when DIO is completed.
- * For async direct IO, since the IO may still pending when return, we
- * set up an end_io call back function, which will do the conversion
- * when async direct IO completed.
- *
- * If the O_DIRECT write will extend the file then add this inode to the
- * orphan list.  So recovery will truncate it back to the original size
- * if the machine crashes during the write.
- *
- */
-static ssize_t ext4_direct_IO_write(struct kiocb *iocb, struct iov_iter *iter)
+static int ext4_iomap_overwrite_begin(struct inode *inode, loff_t offset,
+		loff_t length, unsigned flags, struct iomap *iomap,
+		struct iomap *srcmap)
 {
-	struct file *file = iocb->ki_filp;
-	struct inode *inode = file->f_mapping->host;
-	struct ext4_inode_info *ei = EXT4_I(inode);
-	ssize_t ret;
-	loff_t offset = iocb->ki_pos;
-	size_t count = iov_iter_count(iter);
-	int overwrite = 0;
-	get_block_t *get_block_func = NULL;
-	int dio_flags = 0;
-	loff_t final_size = offset + count;
-	int orphan = 0;
-	handle_t *handle;
-
-	if (final_size > inode->i_size || final_size > ei->i_disksize) {
-		/* Credits for sb + inode write */
-		handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
-		if (IS_ERR(handle)) {
-			ret = PTR_ERR(handle);
-			goto out;
-		}
-		ret = ext4_orphan_add(handle, inode);
-		if (ret) {
-			ext4_journal_stop(handle);
-			goto out;
-		}
-		orphan = 1;
-		ext4_update_i_disksize(inode, inode->i_size);
-		ext4_journal_stop(handle);
-	}
-
-	BUG_ON(iocb->private == NULL);
+	int ret;
 
 	/*
-	 * Make all waiters for direct IO properly wait also for extent
-	 * conversion. This also disallows race between truncate() and
-	 * overwrite DIO as i_dio_count needs to be incremented under i_mutex.
+	 * Even for writes we don't need to allocate blocks, so just pretend
+	 * we are reading to save overhead of starting a transaction.
 	 */
-	inode_dio_begin(inode);
-
-	/* If we do a overwrite dio, i_mutex locking can be released */
-	overwrite = *((int *)iocb->private);
-
-	if (overwrite)
-		inode_unlock(inode);
+	flags &= ~IOMAP_WRITE;
+	ret = ext4_iomap_begin(inode, offset, length, flags, iomap, srcmap);
+	WARN_ON_ONCE(iomap->type != IOMAP_MAPPED);
+	return ret;
+}
 
+static int ext4_iomap_end(struct inode *inode, loff_t offset, loff_t length,
+			  ssize_t written, unsigned flags, struct iomap *iomap)
+{
 	/*
-	 * For extent mapped files we could direct write to holes and fallocate.
-	 *
-	 * Allocated blocks to fill the hole are marked as unwritten to prevent
-	 * parallel buffered read to expose the stale data before DIO complete
-	 * the data IO.
-	 *
-	 * As to previously fallocated extents, ext4 get_block will just simply
-	 * mark the buffer mapped but still keep the extents unwritten.
-	 *
-	 * For non AIO case, we will convert those unwritten extents to written
-	 * after return back from blockdev_direct_IO. That way we save us from
-	 * allocating io_end structure and also the overhead of offloading
-	 * the extent convertion to a workqueue.
-	 *
-	 * For async DIO, the conversion needs to be deferred when the
-	 * IO is completed. The ext4 end_io callback function will be
-	 * called to take care of the conversion work.  Here for async
-	 * case, we allocate an io_end structure to hook to the iocb.
+	 * Check to see whether an error occurred while writing out the data to
+	 * the allocated blocks. If so, return the magic error code so that we
+	 * fallback to buffered I/O and attempt to complete the remainder of
+	 * the I/O. Any blocks that may have been allocated in preparation for
+	 * the direct I/O will be reused during buffered I/O.
 	 */
-	iocb->private = NULL;
-	if (overwrite)
-		get_block_func = ext4_dio_get_block_overwrite;
-	else if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS) ||
-		   round_down(offset, i_blocksize(inode)) >= inode->i_size) {
-		get_block_func = ext4_dio_get_block;
-		dio_flags = DIO_LOCKING | DIO_SKIP_HOLES;
-	} else if (is_sync_kiocb(iocb)) {
-		get_block_func = ext4_dio_get_block_unwritten_sync;
-		dio_flags = DIO_LOCKING;
-	} else {
-		get_block_func = ext4_dio_get_block_unwritten_async;
-		dio_flags = DIO_LOCKING;
-	}
-	ret = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, iter,
-				   get_block_func, ext4_end_io_dio, NULL,
-				   dio_flags);
+	if (flags & (IOMAP_WRITE | IOMAP_DIRECT) && written == 0)
+		return -ENOTBLK;
 
-	if (ret > 0 && !overwrite && ext4_test_inode_state(inode,
-						EXT4_STATE_DIO_UNWRITTEN)) {
-		int err;
-		/*
-		 * for non AIO case, since the IO is already
-		 * completed, we could do the conversion right here
-		 */
-		err = ext4_convert_unwritten_extents(NULL, inode,
-						     offset, ret);
-		if (err < 0)
-			ret = err;
-		ext4_clear_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN);
-	}
+	return 0;
+}
 
-	inode_dio_end(inode);
-	/* take i_mutex locking again if we do a ovewrite dio */
-	if (overwrite)
-		inode_lock(inode);
+const struct iomap_ops ext4_iomap_ops = {
+	.iomap_begin		= ext4_iomap_begin,
+	.iomap_end		= ext4_iomap_end,
+};
 
-	if (ret < 0 && final_size > inode->i_size)
-		ext4_truncate_failed_write(inode);
+const struct iomap_ops ext4_iomap_overwrite_ops = {
+	.iomap_begin		= ext4_iomap_overwrite_begin,
+	.iomap_end		= ext4_iomap_end,
+};
 
-	/* Handle extending of i_size after direct IO write */
-	if (orphan) {
-		int err;
+static bool ext4_iomap_is_delalloc(struct inode *inode,
+				   struct ext4_map_blocks *map)
+{
+	struct extent_status es;
+	ext4_lblk_t offset = 0, end = map->m_lblk + map->m_len - 1;
 
-		/* Credits for sb + inode write */
-		handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
-		if (IS_ERR(handle)) {
-			/*
-			 * We wrote the data but cannot extend
-			 * i_size. Bail out. In async io case, we do
-			 * not return error here because we have
-			 * already submmitted the corresponding
-			 * bio. Returning error here makes the caller
-			 * think that this IO is done and failed
-			 * resulting in race with bio's completion
-			 * handler.
-			 */
-			if (!ret)
-				ret = PTR_ERR(handle);
-			if (inode->i_nlink)
-				ext4_orphan_del(NULL, inode);
+	ext4_es_find_extent_range(inode, &ext4_es_is_delayed,
+				  map->m_lblk, end, &es);
 
-			goto out;
-		}
-		if (inode->i_nlink)
-			ext4_orphan_del(handle, inode);
-		if (ret > 0) {
-			loff_t end = offset + ret;
-			if (end > inode->i_size || end > ei->i_disksize) {
-				ext4_update_i_disksize(inode, end);
-				if (end > inode->i_size)
-					i_size_write(inode, end);
-				/*
-				 * We're going to return a positive `ret'
-				 * here due to non-zero-length I/O, so there's
-				 * no way of reporting error returns from
-				 * ext4_mark_inode_dirty() to userspace.  So
-				 * ignore it.
-				 */
-				ext4_mark_inode_dirty(handle, inode);
-			}
-		}
-		err = ext4_journal_stop(handle);
-		if (ret == 0)
-			ret = err;
+	if (!es.es_len || es.es_lblk > end)
+		return false;
+
+	if (es.es_lblk > map->m_lblk) {
+		map->m_len = es.es_lblk - map->m_lblk;
+		return false;
 	}
-out:
-	return ret;
-}
 
-static ssize_t ext4_direct_IO_read(struct kiocb *iocb, struct iov_iter *iter)
-{
-	struct address_space *mapping = iocb->ki_filp->f_mapping;
-	struct inode *inode = mapping->host;
-	size_t count = iov_iter_count(iter);
-	ssize_t ret;
+	offset = map->m_lblk - es.es_lblk;
+	map->m_len = es.es_len - offset;
 
-	/*
-	 * Shared inode_lock is enough for us - it protects against concurrent
-	 * writes & truncates and since we take care of writing back page cache,
-	 * we are protected against page writeback as well.
-	 */
-	inode_lock_shared(inode);
-	ret = filemap_write_and_wait_range(mapping, iocb->ki_pos,
-					   iocb->ki_pos + count - 1);
-	if (ret)
-		goto out_unlock;
-	ret = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev,
-				   iter, ext4_dio_get_block, NULL, NULL, 0);
-out_unlock:
-	inode_unlock_shared(inode);
-	return ret;
+	return true;
 }
 
-static ssize_t ext4_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
+static int ext4_iomap_begin_report(struct inode *inode, loff_t offset,
+				   loff_t length, unsigned int flags,
+				   struct iomap *iomap, struct iomap *srcmap)
 {
-	struct file *file = iocb->ki_filp;
-	struct inode *inode = file->f_mapping->host;
-	size_t count = iov_iter_count(iter);
-	loff_t offset = iocb->ki_pos;
-	ssize_t ret;
+	int ret;
+	bool delalloc = false;
+	struct ext4_map_blocks map;
+	u8 blkbits = inode->i_blkbits;
 
-#ifdef CONFIG_FS_ENCRYPTION
-	if (IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode))
-		return 0;
-#endif
-	if (fsverity_active(inode))
-		return 0;
+	if ((offset >> blkbits) > EXT4_MAX_LOGICAL_BLOCK)
+		return -EINVAL;
+
+	if (ext4_has_inline_data(inode)) {
+		ret = ext4_inline_data_iomap(inode, iomap);
+		if (ret != -EAGAIN) {
+			if (ret == 0 && offset >= iomap->length)
+				ret = -ENOENT;
+			return ret;
+		}
+	}
 
 	/*
-	 * If we are doing data journalling we don't support O_DIRECT
+	 * Calculate the first and last logical block respectively.
 	 */
-	if (ext4_should_journal_data(inode))
-		return 0;
+	map.m_lblk = offset >> blkbits;
+	map.m_len = min_t(loff_t, (offset + length - 1) >> blkbits,
+			  EXT4_MAX_LOGICAL_BLOCK) - map.m_lblk + 1;
 
-	/* Let buffer I/O handle the inline data case. */
-	if (ext4_has_inline_data(inode))
-		return 0;
+	/*
+	 * Fiemap callers may call for offset beyond s_bitmap_maxbytes.
+	 * So handle it here itself instead of querying ext4_map_blocks().
+	 * Since ext4_map_blocks() will warn about it and will return
+	 * -EIO error.
+	 */
+	if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) {
+		struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 
-	trace_ext4_direct_IO_enter(inode, offset, count, iov_iter_rw(iter));
-	if (iov_iter_rw(iter) == READ)
-		ret = ext4_direct_IO_read(iocb, iter);
-	else
-		ret = ext4_direct_IO_write(iocb, iter);
-	trace_ext4_direct_IO_exit(inode, offset, count, iov_iter_rw(iter), ret);
-	return ret;
+		if (offset >= sbi->s_bitmap_maxbytes) {
+			map.m_flags = 0;
+			goto set_iomap;
+		}
+	}
+
+	ret = ext4_map_blocks(NULL, inode, &map, 0);
+	if (ret < 0)
+		return ret;
+	if (ret == 0)
+		delalloc = ext4_iomap_is_delalloc(inode, &map);
+
+set_iomap:
+	ext4_set_iomap(inode, iomap, &map, offset, length);
+	if (delalloc && iomap->type == IOMAP_HOLE)
+		iomap->type = IOMAP_DELALLOC;
+
+	return 0;
 }
 
+const struct iomap_ops ext4_iomap_report_ops = {
+	.iomap_begin = ext4_iomap_begin_report,
+};
+
 /*
  * Pages can be marked dirty completely asynchronously from ext4's journalling
  * activity.  By filemap_sync_pte(), try_to_unmap_one(), etc.  We cannot do
@@ -3901,7 +3602,7 @@ static int ext4_set_page_dirty(struct page *page)
 
 static const struct address_space_operations ext4_aops = {
 	.readpage		= ext4_readpage,
-	.readpages		= ext4_readpages,
+	.readahead		= ext4_readahead,
 	.writepage		= ext4_writepage,
 	.writepages		= ext4_writepages,
 	.write_begin		= ext4_write_begin,
@@ -3910,7 +3611,7 @@ static const struct address_space_operations ext4_aops = {
 	.bmap			= ext4_bmap,
 	.invalidatepage		= ext4_invalidatepage,
 	.releasepage		= ext4_releasepage,
-	.direct_IO		= ext4_direct_IO,
+	.direct_IO		= noop_direct_IO,
 	.migratepage		= buffer_migrate_page,
 	.is_partially_uptodate  = block_is_partially_uptodate,
 	.error_remove_page	= generic_error_remove_page,
@@ -3918,7 +3619,7 @@ static const struct address_space_operations ext4_aops = {
 
 static const struct address_space_operations ext4_journalled_aops = {
 	.readpage		= ext4_readpage,
-	.readpages		= ext4_readpages,
+	.readahead		= ext4_readahead,
 	.writepage		= ext4_writepage,
 	.writepages		= ext4_writepages,
 	.write_begin		= ext4_write_begin,
@@ -3927,14 +3628,14 @@ static const struct address_space_operations ext4_journalled_aops = {
 	.bmap			= ext4_bmap,
 	.invalidatepage		= ext4_journalled_invalidatepage,
 	.releasepage		= ext4_releasepage,
-	.direct_IO		= ext4_direct_IO,
+	.direct_IO		= noop_direct_IO,
 	.is_partially_uptodate  = block_is_partially_uptodate,
 	.error_remove_page	= generic_error_remove_page,
 };
 
 static const struct address_space_operations ext4_da_aops = {
 	.readpage		= ext4_readpage,
-	.readpages		= ext4_readpages,
+	.readahead		= ext4_readahead,
 	.writepage		= ext4_writepage,
 	.writepages		= ext4_writepages,
 	.write_begin		= ext4_da_write_begin,
@@ -3943,7 +3644,7 @@ static const struct address_space_operations ext4_da_aops = {
 	.bmap			= ext4_bmap,
 	.invalidatepage		= ext4_invalidatepage,
 	.releasepage		= ext4_releasepage,
-	.direct_IO		= ext4_direct_IO,
+	.direct_IO		= noop_direct_IO,
 	.migratepage		= buffer_migrate_page,
 	.is_partially_uptodate  = block_is_partially_uptodate,
 	.error_remove_page	= generic_error_remove_page,
@@ -4037,8 +3738,12 @@ static int __ext4_block_zero_page_range(handle_t *handle,
 		if (S_ISREG(inode->i_mode) && IS_ENCRYPTED(inode)) {
 			/* We expect the key to be set. */
 			BUG_ON(!fscrypt_has_encryption_key(inode));
-			WARN_ON_ONCE(fscrypt_decrypt_pagecache_blocks(
-					page, blocksize, bh_offset(bh)));
+			err = fscrypt_decrypt_pagecache_blocks(page, blocksize,
+							       bh_offset(bh));
+			if (err) {
+				clear_buffer_uptodate(bh);
+				goto unlock;
+			}
 		}
 	}
 	if (ext4_should_journal_data(inode)) {
@@ -4248,9 +3953,6 @@ int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length)
 	unsigned int credits;
 	int ret = 0;
 
-	if (!S_ISREG(inode->i_mode))
-		return -EOPNOTSUPP;
-
 	trace_ext4_punch_hole(inode, offset, length, 0);
 
 	ext4_clear_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
@@ -4457,8 +4159,6 @@ int ext4_truncate(struct inode *inode)
 	if (!ext4_can_truncate(inode))
 		return 0;
 
-	ext4_clear_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
-
 	if (inode->i_size == 0 && !test_opt(inode->i_sb, NO_AUTO_DA_ALLOC))
 		ext4_set_inode_state(inode, EXT4_STATE_DA_ALLOC_CLOSE);
 
@@ -4576,6 +4276,8 @@ static int __ext4_get_inode_loc(struct inode *inode,
 	bh = sb_getblk(sb, block);
 	if (unlikely(!bh))
 		return -ENOMEM;
+	if (ext4_simulate_fail(sb, EXT4_SIM_INODE_EIO))
+		goto simulate_eio;
 	if (!buffer_uptodate(bh)) {
 		lock_buffer(bh);
 
@@ -4659,7 +4361,7 @@ make_io:
 			if (end > table)
 				end = table;
 			while (b <= end)
-				sb_breadahead(sb, b++);
+				sb_breadahead_unmovable(sb, b++);
 		}
 
 		/*
@@ -4674,7 +4376,8 @@ make_io:
 		blk_finish_plug(&plug);
 		wait_on_buffer(bh);
 		if (!buffer_uptodate(bh)) {
-			EXT4_ERROR_INODE_BLOCK(inode, block,
+		simulate_eio:
+			ext4_error_inode_block(inode, block, EIO,
 					       "unable to read itable block");
 			brelse(bh);
 			return -EIO;
@@ -4826,7 +4529,7 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 	    (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))) {
 		if (flags & EXT4_IGET_HANDLE)
 			return ERR_PTR(-ESTALE);
-		__ext4_error(sb, function, line,
+		__ext4_error(sb, function, line, EFSCORRUPTED, 0,
 			     "inode #%lu: comm %s: iget: illegal inode #",
 			     ino, current->comm);
 		return ERR_PTR(-EFSCORRUPTED);
@@ -4887,9 +4590,10 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 					      sizeof(gen));
 	}
 
-	if (!ext4_inode_csum_verify(inode, raw_inode, ei)) {
-		ext4_error_inode(inode, function, line, 0,
-				 "iget: checksum invalid");
+	if (!ext4_inode_csum_verify(inode, raw_inode, ei) ||
+	    ext4_simulate_fail(sb, EXT4_SIM_INODE_CRC)) {
+		ext4_error_inode_err(inode, function, line, 0, EFSBADCRC,
+				     "iget: checksum invalid");
 		ret = -EFSBADCRC;
 		goto bad_inode;
 	}
@@ -4951,6 +4655,18 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 		ret = -EFSCORRUPTED;
 		goto bad_inode;
 	}
+	/*
+	 * If dir_index is not enabled but there's dir with INDEX flag set,
+	 * we'd normally treat htree data as empty space. But with metadata
+	 * checksumming that corrupts checksums so forbid that.
+	 */
+	if (!ext4_has_feature_dir_index(sb) && ext4_has_metadata_csum(sb) &&
+	    ext4_test_inode_flag(inode, EXT4_INODE_INDEX)) {
+		ext4_error_inode(inode, function, line, 0,
+			 "iget: Dir with htree data on filesystem without dir_index feature.");
+		ret = -EFSCORRUPTED;
+		goto bad_inode;
+	}
 	ei->i_disksize = inode->i_size;
 #ifdef CONFIG_QUOTA
 	ei->i_reserved_quota = 0;
@@ -5107,7 +4823,7 @@ static int ext4_inode_blocks_set(handle_t *handle,
 				struct ext4_inode_info *ei)
 {
 	struct inode *inode = &(ei->vfs_inode);
-	u64 i_blocks = inode->i_blocks;
+	u64 i_blocks = READ_ONCE(inode->i_blocks);
 	struct super_block *sb = inode->i_sb;
 
 	if (i_blocks <= ~0U) {
@@ -5277,7 +4993,7 @@ static int ext4_do_update_inode(handle_t *handle,
 		raw_inode->i_file_acl_high =
 			cpu_to_le16(ei->i_file_acl >> 32);
 	raw_inode->i_file_acl_lo = cpu_to_le32(ei->i_file_acl);
-	if (ei->i_disksize != ext4_isize(inode->i_sb, raw_inode)) {
+	if (READ_ONCE(ei->i_disksize) != ext4_isize(inode->i_sb, raw_inode)) {
 		ext4_isize_set(raw_inode, ei->i_disksize);
 		need_datasync = 1;
 	}
@@ -5426,8 +5142,8 @@ int ext4_write_inode(struct inode *inode, struct writeback_control *wbc)
 		if (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync)
 			sync_dirty_buffer(iloc.bh);
 		if (buffer_req(iloc.bh) && !buffer_uptodate(iloc.bh)) {
-			EXT4_ERROR_INODE_BLOCK(inode, iloc.bh->b_blocknr,
-					 "IO error syncing inode");
+			ext4_error_inode_block(inode, iloc.bh->b_blocknr, EIO,
+					       "IO error syncing inode");
 			err = -EIO;
 		}
 		brelse(iloc.bh);
@@ -5450,11 +5166,15 @@ static void ext4_wait_for_tail_page_commit(struct inode *inode)
 
 	offset = inode->i_size & (PAGE_SIZE - 1);
 	/*
-	 * All buffers in the last page remain valid? Then there's nothing to
-	 * do. We do the check mainly to optimize the common PAGE_SIZE ==
-	 * blocksize case
+	 * If the page is fully truncated, we don't need to wait for any commit
+	 * (and we even should not as __ext4_journalled_invalidatepage() may
+	 * strip all buffers from the page but keep the page dirty which can then
+	 * confuse e.g. concurrent ext4_writepage() seeing dirty page without
+	 * buffers). Also we don't need to wait for any commit if all buffers in
+	 * the page remain valid. This is most beneficial for the common case of
+	 * blocksize == PAGESIZE.
 	 */
-	if (offset > PAGE_SIZE - i_blocksize(inode))
+	if (!offset || offset > (PAGE_SIZE - i_blocksize(inode)))
 		return;
 	while (1) {
 		page = find_lock_page(inode->i_mapping,
@@ -5700,7 +5420,8 @@ int ext4_getattr(const struct path *path, struct kstat *stat,
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	unsigned int flags;
 
-	if (EXT4_FITS_IN_INODE(raw_inode, ei, i_crtime)) {
+	if ((request_mask & STATX_BTIME) &&
+	    EXT4_FITS_IN_INODE(raw_inode, ei, i_crtime)) {
 		stat->result_mask |= STATX_BTIME;
 		stat->btime.tv_sec = ei->i_crtime.tv_sec;
 		stat->btime.tv_nsec = ei->i_crtime.tv_nsec;
@@ -5717,12 +5438,15 @@ int ext4_getattr(const struct path *path, struct kstat *stat,
 		stat->attributes |= STATX_ATTR_IMMUTABLE;
 	if (flags & EXT4_NODUMP_FL)
 		stat->attributes |= STATX_ATTR_NODUMP;
+	if (flags & EXT4_VERITY_FL)
+		stat->attributes |= STATX_ATTR_VERITY;
 
 	stat->attributes_mask |= (STATX_ATTR_APPEND |
 				  STATX_ATTR_COMPRESSED |
 				  STATX_ATTR_ENCRYPTED |
 				  STATX_ATTR_IMMUTABLE |
-				  STATX_ATTR_NODUMP);
+				  STATX_ATTR_NODUMP |
+				  STATX_ATTR_VERITY);
 
 	generic_fillattr(inode, stat);
 	return 0;
@@ -5912,8 +5636,23 @@ static int __ext4_expand_extra_isize(struct inode *inode,
 {
 	struct ext4_inode *raw_inode;
 	struct ext4_xattr_ibody_header *header;
+	unsigned int inode_size = EXT4_INODE_SIZE(inode->i_sb);
+	struct ext4_inode_info *ei = EXT4_I(inode);
 	int error;
 
+	/* this was checked at iget time, but double check for good measure */
+	if ((EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > inode_size) ||
+	    (ei->i_extra_isize & 3)) {
+		EXT4_ERROR_INODE(inode, "bad extra_isize %u (inode size %u)",
+				 ei->i_extra_isize,
+				 EXT4_INODE_SIZE(inode->i_sb));
+		return -EFSCORRUPTED;
+	}
+	if ((new_extra_isize < ei->i_extra_isize) ||
+	    (new_extra_isize < 4) ||
+	    (new_extra_isize > inode_size - EXT4_GOOD_OLD_INODE_SIZE))
+		return -EINVAL;	/* Should never happen */
+
 	raw_inode = ext4_raw_inode(iloc);
 
 	header = IHDR(inode, raw_inode);
@@ -5965,9 +5704,8 @@ static int ext4_try_to_expand_extra_isize(struct inode *inode,
 	 * If this is felt to be critical, then e2fsck should be run to
 	 * force a large enough s_min_extra_isize.
 	 */
-	if (ext4_handle_valid(handle) &&
-	    jbd2_journal_extend(handle,
-				EXT4_DATA_TRANS_BLOCKS(inode->i_sb)) != 0)
+	if (ext4_journal_extend(handle,
+				EXT4_DATA_TRANS_BLOCKS(inode->i_sb), 0) != 0)
 		return -ENOSPC;
 
 	if (ext4_write_trylock_xattr(inode, &no_expand) == 0)
@@ -6007,7 +5745,7 @@ int ext4_expand_extra_isize(struct inode *inode,
 	error = ext4_journal_get_write_access(handle, iloc->bh);
 	if (error) {
 		brelse(iloc->bh);
-		goto out_stop;
+		goto out_unlock;
 	}
 
 	error = __ext4_expand_extra_isize(inode, new_extra_isize, iloc,
@@ -6017,8 +5755,8 @@ int ext4_expand_extra_isize(struct inode *inode,
 	if (!error)
 		error = rc;
 
+out_unlock:
 	ext4_write_unlock_xattr(inode, &no_expand);
-out_stop:
 	ext4_journal_stop(handle);
 	return error;
 }
@@ -6133,7 +5871,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val)
 		}
 	}
 
-	percpu_down_write(&sbi->s_journal_flag_rwsem);
+	percpu_down_write(&sbi->s_writepages_rwsem);
 	jbd2_journal_lock_updates(journal);
 
 	/*
@@ -6150,7 +5888,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val)
 		err = jbd2_journal_flush(journal);
 		if (err < 0) {
 			jbd2_journal_unlock_updates(journal);
-			percpu_up_write(&sbi->s_journal_flag_rwsem);
+			percpu_up_write(&sbi->s_writepages_rwsem);
 			return err;
 		}
 		ext4_clear_inode_flag(inode, EXT4_INODE_JOURNAL_DATA);
@@ -6158,7 +5896,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val)
 	ext4_set_aops(inode);
 
 	jbd2_journal_unlock_updates(journal);
-	percpu_up_write(&sbi->s_journal_flag_rwsem);
+	percpu_up_write(&sbi->s_writepages_rwsem);
 
 	if (val)
 		up_write(&EXT4_I(inode)->i_mmap_sem);
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index 0b7f316fd30f..0746532ba463 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -327,18 +327,6 @@ static int ext4_ioctl_setflags(struct inode *inode,
 	if ((flags ^ oldflags) & EXT4_EXTENTS_FL)
 		migrate = 1;
 
-	if (flags & EXT4_EOFBLOCKS_FL) {
-		/* we don't support adding EOFBLOCKS flag */
-		if (!(oldflags & EXT4_EOFBLOCKS_FL)) {
-			err = -EOPNOTSUPP;
-			goto flags_out;
-		}
-	} else if (oldflags & EXT4_EOFBLOCKS_FL) {
-		err = ext4_truncate(inode);
-		if (err)
-			goto flags_out;
-	}
-
 	if ((flags ^ oldflags) & EXT4_CASEFOLD_FL) {
 		if (!ext4_has_feature_casefold(sb)) {
 			err = -EOPNOTSUPP;
@@ -745,29 +733,6 @@ static void ext4_fill_fsxattr(struct inode *inode, struct fsxattr *fa)
 		fa->fsx_projid = from_kprojid(&init_user_ns, ei->i_projid);
 }
 
-/* copied from fs/ioctl.c */
-static int fiemap_check_ranges(struct super_block *sb,
-			       u64 start, u64 len, u64 *new_len)
-{
-	u64 maxbytes = (u64) sb->s_maxbytes;
-
-	*new_len = len;
-
-	if (len == 0)
-		return -EINVAL;
-
-	if (start > maxbytes)
-		return -EFBIG;
-
-	/*
-	 * Shrink request scope to what the fs can actually handle.
-	 */
-	if (len > maxbytes || (maxbytes - len) < start)
-		*new_len = maxbytes - start;
-
-	return 0;
-}
-
 /* So that the fiemap access checks can't overflow on 32 bit machines. */
 #define FIEMAP_MAX_EXTENTS	(UINT_MAX / sizeof(struct fiemap_extent))
 
@@ -777,8 +742,6 @@ static int ext4_ioctl_get_es_cache(struct file *filp, unsigned long arg)
 	struct fiemap __user *ufiemap = (struct fiemap __user *) arg;
 	struct fiemap_extent_info fieinfo = { 0, };
 	struct inode *inode = file_inode(filp);
-	struct super_block *sb = inode->i_sb;
-	u64 len;
 	int error;
 
 	if (copy_from_user(&fiemap, ufiemap, sizeof(fiemap)))
@@ -787,11 +750,6 @@ static int ext4_ioctl_get_es_cache(struct file *filp, unsigned long arg)
 	if (fiemap.fm_extent_count > FIEMAP_MAX_EXTENTS)
 		return -EINVAL;
 
-	error = fiemap_check_ranges(sb, fiemap.fm_start, fiemap.fm_length,
-				    &len);
-	if (error)
-		return error;
-
 	fieinfo.fi_flags = fiemap.fm_flags;
 	fieinfo.fi_extents_max = fiemap.fm_extent_count;
 	fieinfo.fi_extents_start = ufiemap->fm_extents;
@@ -804,7 +762,8 @@ static int ext4_ioctl_get_es_cache(struct file *filp, unsigned long arg)
 	if (fieinfo.fi_flags & FIEMAP_FLAG_SYNC)
 		filemap_write_and_wait(inode->i_mapping);
 
-	error = ext4_get_es_cache(inode, &fieinfo, fiemap.fm_start, len);
+	error = ext4_get_es_cache(inode, &fieinfo, fiemap.fm_start,
+			fiemap.fm_length);
 	fiemap.fm_flags = fieinfo.fi_flags;
 	fiemap.fm_mapped_extents = fieinfo.fi_extents_mapped;
 	if (copy_to_user(ufiemap, &fiemap, sizeof(fiemap)))
@@ -1210,6 +1169,11 @@ resizefs_out:
 			return -EOPNOTSUPP;
 		return fscrypt_ioctl_get_key_status(filp, (void __user *)arg);
 
+	case FS_IOC_GET_ENCRYPTION_NONCE:
+		if (!ext4_has_feature_encrypt(sb))
+			return -EOPNOTSUPP;
+		return fscrypt_ioctl_get_nonce(filp, (void __user *)arg);
+
 	case EXT4_IOC_CLEAR_ES_CACHE:
 	{
 		if (!inode_owner_or_capable(inode))
@@ -1360,6 +1324,7 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	}
 	case EXT4_IOC_MOVE_EXT:
 	case EXT4_IOC_RESIZE_FS:
+	case FITRIM:
 	case EXT4_IOC_PRECACHE_EXTENTS:
 	case EXT4_IOC_SET_ENCRYPTION_POLICY:
 	case EXT4_IOC_GET_ENCRYPTION_PWSALT:
@@ -1369,6 +1334,7 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case FS_IOC_REMOVE_ENCRYPTION_KEY:
 	case FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS:
 	case FS_IOC_GET_ENCRYPTION_KEY_STATUS:
+	case FS_IOC_GET_ENCRYPTION_NONCE:
 	case EXT4_IOC_SHUTDOWN:
 	case FS_IOC_GETFSMAP:
 	case FS_IOC_ENABLE_VERITY:
@@ -1376,6 +1342,8 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case EXT4_IOC_CLEAR_ES_CACHE:
 	case EXT4_IOC_GETSTATE:
 	case EXT4_IOC_GET_ES_CACHE:
+	case EXT4_IOC_FSGETXATTR:
+	case EXT4_IOC_FSSETXATTR:
 		break;
 	default:
 		return -ENOIOCTLCMD;
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index a3e2767bdf2f..30d5d97548c4 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -1901,8 +1901,15 @@ void ext4_mb_simple_scan_group(struct ext4_allocation_context *ac,
 		BUG_ON(buddy == NULL);
 
 		k = mb_find_next_zero_bit(buddy, max, 0);
-		BUG_ON(k >= max);
-
+		if (k >= max) {
+			ext4_grp_locked_error(ac->ac_sb, e4b->bd_group, 0, 0,
+				"%d free clusters of order %d. But found 0",
+				grp->bb_counters[i], i);
+			ext4_mark_group_bitmap_corrupted(ac->ac_sb,
+					 e4b->bd_group,
+					EXT4_GROUP_INFO_BBITMAP_CORRUPT);
+			break;
+		}
 		ac->ac_found++;
 
 		ac->ac_b_ex.fe_len = 1 << i;
@@ -1936,7 +1943,8 @@ void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac,
 	int free;
 
 	free = e4b->bd_info->bb_free;
-	BUG_ON(free <= 0);
+	if (WARN_ON(free <= 0))
+		return;
 
 	i = e4b->bd_info->bb_first_free;
 
@@ -1959,7 +1967,8 @@ void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac,
 		}
 
 		mb_find_extent(e4b, i, ac->ac_g_ex.fe_len, &ex);
-		BUG_ON(ex.fe_len <= 0);
+		if (WARN_ON(ex.fe_len <= 0))
+			break;
 		if (free < ex.fe_len) {
 			ext4_grp_locked_error(sb, e4b->bd_group, 0, 0,
 					"%d free clusters as per "
@@ -2356,7 +2365,7 @@ int ext4_mb_alloc_groupinfo(struct super_block *sb, ext4_group_t ngroups)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	unsigned size;
-	struct ext4_group_info ***new_groupinfo;
+	struct ext4_group_info ***old_groupinfo, ***new_groupinfo;
 
 	size = (ngroups + EXT4_DESC_PER_BLOCK(sb) - 1) >>
 		EXT4_DESC_PER_BLOCK_BITS(sb);
@@ -2369,13 +2378,16 @@ int ext4_mb_alloc_groupinfo(struct super_block *sb, ext4_group_t ngroups)
 		ext4_msg(sb, KERN_ERR, "can't allocate buddy meta group");
 		return -ENOMEM;
 	}
-	if (sbi->s_group_info) {
-		memcpy(new_groupinfo, sbi->s_group_info,
+	rcu_read_lock();
+	old_groupinfo = rcu_dereference(sbi->s_group_info);
+	if (old_groupinfo)
+		memcpy(new_groupinfo, old_groupinfo,
 		       sbi->s_group_info_size * sizeof(*sbi->s_group_info));
-		kvfree(sbi->s_group_info);
-	}
-	sbi->s_group_info = new_groupinfo;
+	rcu_read_unlock();
+	rcu_assign_pointer(sbi->s_group_info, new_groupinfo);
 	sbi->s_group_info_size = size / sizeof(*sbi->s_group_info);
+	if (old_groupinfo)
+		ext4_kvfree_array_rcu(old_groupinfo);
 	ext4_debug("allocated s_groupinfo array for %d meta_bg's\n", 
 		   sbi->s_group_info_size);
 	return 0;
@@ -2387,6 +2399,7 @@ int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group,
 {
 	int i;
 	int metalen = 0;
+	int idx = group >> EXT4_DESC_PER_BLOCK_BITS(sb);
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct ext4_group_info **meta_group_info;
 	struct kmem_cache *cachep = get_groupinfo_cache(sb->s_blocksize_bits);
@@ -2405,12 +2418,12 @@ int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group,
 				 "for a buddy group");
 			goto exit_meta_group_info;
 		}
-		sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)] =
-			meta_group_info;
+		rcu_read_lock();
+		rcu_dereference(sbi->s_group_info)[idx] = meta_group_info;
+		rcu_read_unlock();
 	}
 
-	meta_group_info =
-		sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)];
+	meta_group_info = sbi_array_rcu_deref(sbi, s_group_info, idx);
 	i = group & (EXT4_DESC_PER_BLOCK(sb) - 1);
 
 	meta_group_info[i] = kmem_cache_zalloc(cachep, GFP_NOFS);
@@ -2458,8 +2471,13 @@ int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group,
 exit_group_info:
 	/* If a meta_group_info table has been allocated, release it now */
 	if (group % EXT4_DESC_PER_BLOCK(sb) == 0) {
-		kfree(sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)]);
-		sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)] = NULL;
+		struct ext4_group_info ***group_info;
+
+		rcu_read_lock();
+		group_info = rcu_dereference(sbi->s_group_info);
+		kfree(group_info[idx]);
+		group_info[idx] = NULL;
+		rcu_read_unlock();
 	}
 exit_meta_group_info:
 	return -ENOMEM;
@@ -2472,6 +2490,7 @@ static int ext4_mb_init_backend(struct super_block *sb)
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	int err;
 	struct ext4_group_desc *desc;
+	struct ext4_group_info ***group_info;
 	struct kmem_cache *cachep;
 
 	err = ext4_mb_alloc_groupinfo(sb, ngroups);
@@ -2507,11 +2526,16 @@ err_freebuddy:
 	while (i-- > 0)
 		kmem_cache_free(cachep, ext4_get_group_info(sb, i));
 	i = sbi->s_group_info_size;
+	rcu_read_lock();
+	group_info = rcu_dereference(sbi->s_group_info);
 	while (i-- > 0)
-		kfree(sbi->s_group_info[i]);
+		kfree(group_info[i]);
+	rcu_read_unlock();
 	iput(sbi->s_buddy_cache);
 err_freesgi:
-	kvfree(sbi->s_group_info);
+	rcu_read_lock();
+	kvfree(rcu_dereference(sbi->s_group_info));
+	rcu_read_unlock();
 	return -ENOMEM;
 }
 
@@ -2700,7 +2724,7 @@ int ext4_mb_release(struct super_block *sb)
 	ext4_group_t ngroups = ext4_get_groups_count(sb);
 	ext4_group_t i;
 	int num_meta_group_infos;
-	struct ext4_group_info *grinfo;
+	struct ext4_group_info *grinfo, ***group_info;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct kmem_cache *cachep = get_groupinfo_cache(sb->s_blocksize_bits);
 
@@ -2719,9 +2743,12 @@ int ext4_mb_release(struct super_block *sb)
 		num_meta_group_infos = (ngroups +
 				EXT4_DESC_PER_BLOCK(sb) - 1) >>
 			EXT4_DESC_PER_BLOCK_BITS(sb);
+		rcu_read_lock();
+		group_info = rcu_dereference(sbi->s_group_info);
 		for (i = 0; i < num_meta_group_infos; i++)
-			kfree(sbi->s_group_info[i]);
-		kvfree(sbi->s_group_info);
+			kfree(group_info[i]);
+		kvfree(group_info);
+		rcu_read_unlock();
 	}
 	kfree(sbi->s_mb_offsets);
 	kfree(sbi->s_mb_maxs);
@@ -3020,7 +3047,8 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 		ext4_group_t flex_group = ext4_flex_group(sbi,
 							  ac->ac_b_ex.fe_group);
 		atomic64_sub(ac->ac_b_ex.fe_len,
-			     &sbi->s_flex_groups[flex_group].free_clusters);
+			     &sbi_array_rcu_deref(sbi, s_flex_groups,
+						  flex_group)->free_clusters);
 	}
 
 	err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh);
@@ -3895,8 +3923,9 @@ ext4_mb_discard_group_preallocations(struct super_block *sb,
 	bitmap_bh = ext4_read_block_bitmap(sb, group);
 	if (IS_ERR(bitmap_bh)) {
 		err = PTR_ERR(bitmap_bh);
-		ext4_error(sb, "Error %d reading block bitmap for %u",
-			   err, group);
+		ext4_error_err(sb, -err,
+			       "Error %d reading block bitmap for %u",
+			       err, group);
 		return 0;
 	}
 
@@ -4063,16 +4092,16 @@ repeat:
 		err = ext4_mb_load_buddy_gfp(sb, group, &e4b,
 					     GFP_NOFS|__GFP_NOFAIL);
 		if (err) {
-			ext4_error(sb, "Error %d loading buddy information for %u",
-				   err, group);
+			ext4_error_err(sb, -err, "Error %d loading buddy information for %u",
+				       err, group);
 			continue;
 		}
 
 		bitmap_bh = ext4_read_block_bitmap(sb, group);
 		if (IS_ERR(bitmap_bh)) {
 			err = PTR_ERR(bitmap_bh);
-			ext4_error(sb, "Error %d reading block bitmap for %u",
-					err, group);
+			ext4_error_err(sb, -err, "Error %d reading block bitmap for %u",
+				       err, group);
 			ext4_mb_unload_buddy(&e4b);
 			continue;
 		}
@@ -4280,7 +4309,8 @@ ext4_mb_discard_lg_preallocations(struct super_block *sb,
 
 	spin_lock(&lg->lg_prealloc_lock);
 	list_for_each_entry_rcu(pa, &lg->lg_prealloc_list[order],
-						pa_inode_list) {
+				pa_inode_list,
+				lockdep_is_held(&lg->lg_prealloc_lock)) {
 		spin_lock(&pa->pa_lock);
 		if (atomic_read(&pa->pa_count)) {
 			/*
@@ -4325,8 +4355,8 @@ ext4_mb_discard_lg_preallocations(struct super_block *sb,
 		err = ext4_mb_load_buddy_gfp(sb, group, &e4b,
 					     GFP_NOFS|__GFP_NOFAIL);
 		if (err) {
-			ext4_error(sb, "Error %d loading buddy information for %u",
-				   err, group);
+			ext4_error_err(sb, -err, "Error %d loading buddy information for %u",
+				       err, group);
 			continue;
 		}
 		ext4_lock_group(sb, group);
@@ -4363,7 +4393,8 @@ static void ext4_mb_add_n_trim(struct ext4_allocation_context *ac)
 	/* Add the prealloc space to lg */
 	spin_lock(&lg->lg_prealloc_lock);
 	list_for_each_entry_rcu(tmp_pa, &lg->lg_prealloc_list[order],
-						pa_inode_list) {
+				pa_inode_list,
+				lockdep_is_held(&lg->lg_prealloc_lock)) {
 		spin_lock(&tmp_pa->pa_lock);
 		if (tmp_pa->pa_deleted) {
 			spin_unlock(&tmp_pa->pa_lock);
@@ -4914,7 +4945,8 @@ do_more:
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi, block_group);
 		atomic64_add(count_clusters,
-			     &sbi->s_flex_groups[flex_group].free_clusters);
+			     &sbi_array_rcu_deref(sbi, s_flex_groups,
+						  flex_group)->free_clusters);
 	}
 
 	/*
@@ -5071,7 +5103,8 @@ int ext4_group_add_blocks(handle_t *handle, struct super_block *sb,
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi, block_group);
 		atomic64_add(clusters_freed,
-			     &sbi->s_flex_groups[flex_group].free_clusters);
+			     &sbi_array_rcu_deref(sbi, s_flex_groups,
+						  flex_group)->free_clusters);
 	}
 
 	ext4_mb_unload_buddy(&e4b);
diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c
index b1e4d359f73b..fb6520f37135 100644
--- a/fs/ext4/migrate.c
+++ b/fs/ext4/migrate.c
@@ -50,29 +50,9 @@ static int finish_range(handle_t *handle, struct inode *inode,
 	needed = ext4_ext_calc_credits_for_single_extent(inode,
 		    lb->last_block - lb->first_block + 1, path);
 
-	/*
-	 * Make sure the credit we accumalated is not really high
-	 */
-	if (needed && ext4_handle_has_enough_credits(handle,
-						EXT4_RESERVE_TRANS_BLOCKS)) {
-		up_write((&EXT4_I(inode)->i_data_sem));
-		retval = ext4_journal_restart(handle, needed);
-		down_write((&EXT4_I(inode)->i_data_sem));
-		if (retval)
-			goto err_out;
-	} else if (needed) {
-		retval = ext4_journal_extend(handle, needed);
-		if (retval) {
-			/*
-			 * IF not able to extend the journal restart the journal
-			 */
-			up_write((&EXT4_I(inode)->i_data_sem));
-			retval = ext4_journal_restart(handle, needed);
-			down_write((&EXT4_I(inode)->i_data_sem));
-			if (retval)
-				goto err_out;
-		}
-	}
+	retval = ext4_datasem_ensure_credits(handle, inode, needed, needed, 0);
+	if (retval < 0)
+		goto err_out;
 	retval = ext4_ext_insert_extent(handle, inode, &path, &newext, 0);
 err_out:
 	up_write((&EXT4_I(inode)->i_data_sem));
@@ -196,42 +176,30 @@ static int update_tind_extent_range(handle_t *handle, struct inode *inode,
 
 }
 
-static int extend_credit_for_blkdel(handle_t *handle, struct inode *inode)
-{
-	int retval = 0, needed;
-
-	if (ext4_handle_has_enough_credits(handle, EXT4_RESERVE_TRANS_BLOCKS+1))
-		return 0;
-	/*
-	 * We are freeing a blocks. During this we touch
-	 * superblock, group descriptor and block bitmap.
-	 * So allocate a credit of 3. We may update
-	 * quota (user and group).
-	 */
-	needed = 3 + EXT4_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb);
-
-	if (ext4_journal_extend(handle, needed) != 0)
-		retval = ext4_journal_restart(handle, needed);
-
-	return retval;
-}
-
 static int free_dind_blocks(handle_t *handle,
 				struct inode *inode, __le32 i_data)
 {
 	int i;
 	__le32 *tmp_idata;
 	struct buffer_head *bh;
+	struct super_block *sb = inode->i_sb;
 	unsigned long max_entries = inode->i_sb->s_blocksize >> 2;
+	int err;
 
-	bh = ext4_sb_bread(inode->i_sb, le32_to_cpu(i_data), 0);
+	bh = ext4_sb_bread(sb, le32_to_cpu(i_data), 0);
 	if (IS_ERR(bh))
 		return PTR_ERR(bh);
 
 	tmp_idata = (__le32 *)bh->b_data;
 	for (i = 0; i < max_entries; i++) {
 		if (tmp_idata[i]) {
-			extend_credit_for_blkdel(handle, inode);
+			err = ext4_journal_ensure_credits(handle,
+				EXT4_RESERVE_TRANS_BLOCKS,
+				ext4_free_metadata_revoke_credits(sb, 1));
+			if (err < 0) {
+				put_bh(bh);
+				return err;
+			}
 			ext4_free_blocks(handle, inode, NULL,
 					 le32_to_cpu(tmp_idata[i]), 1,
 					 EXT4_FREE_BLOCKS_METADATA |
@@ -239,7 +207,10 @@ static int free_dind_blocks(handle_t *handle,
 		}
 	}
 	put_bh(bh);
-	extend_credit_for_blkdel(handle, inode);
+	err = ext4_journal_ensure_credits(handle, EXT4_RESERVE_TRANS_BLOCKS,
+				ext4_free_metadata_revoke_credits(sb, 1));
+	if (err < 0)
+		return err;
 	ext4_free_blocks(handle, inode, NULL, le32_to_cpu(i_data), 1,
 			 EXT4_FREE_BLOCKS_METADATA |
 			 EXT4_FREE_BLOCKS_FORGET);
@@ -270,7 +241,10 @@ static int free_tind_blocks(handle_t *handle,
 		}
 	}
 	put_bh(bh);
-	extend_credit_for_blkdel(handle, inode);
+	retval = ext4_journal_ensure_credits(handle, EXT4_RESERVE_TRANS_BLOCKS,
+			ext4_free_metadata_revoke_credits(inode->i_sb, 1));
+	if (retval < 0)
+		return retval;
 	ext4_free_blocks(handle, inode, NULL, le32_to_cpu(i_data), 1,
 			 EXT4_FREE_BLOCKS_METADATA |
 			 EXT4_FREE_BLOCKS_FORGET);
@@ -283,7 +257,11 @@ static int free_ind_block(handle_t *handle, struct inode *inode, __le32 *i_data)
 
 	/* ei->i_data[EXT4_IND_BLOCK] */
 	if (i_data[0]) {
-		extend_credit_for_blkdel(handle, inode);
+		retval = ext4_journal_ensure_credits(handle,
+			EXT4_RESERVE_TRANS_BLOCKS,
+			ext4_free_metadata_revoke_credits(inode->i_sb, 1));
+		if (retval < 0)
+			return retval;
 		ext4_free_blocks(handle, inode, NULL,
 				le32_to_cpu(i_data[0]), 1,
 				 EXT4_FREE_BLOCKS_METADATA |
@@ -318,12 +296,9 @@ static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
 	 * One credit accounted for writing the
 	 * i_data field of the original inode
 	 */
-	retval = ext4_journal_extend(handle, 1);
-	if (retval) {
-		retval = ext4_journal_restart(handle, 1);
-		if (retval)
-			goto err_out;
-	}
+	retval = ext4_journal_ensure_credits(handle, 1, 0);
+	if (retval < 0)
+		goto err_out;
 
 	i_data[0] = ei->i_data[EXT4_IND_BLOCK];
 	i_data[1] = ei->i_data[EXT4_DIND_BLOCK];
@@ -391,15 +366,20 @@ static int free_ext_idx(handle_t *handle, struct inode *inode,
 		ix = EXT_FIRST_INDEX(eh);
 		for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ix++) {
 			retval = free_ext_idx(handle, inode, ix);
-			if (retval)
-				break;
+			if (retval) {
+				put_bh(bh);
+				return retval;
+			}
 		}
 	}
 	put_bh(bh);
-	extend_credit_for_blkdel(handle, inode);
+	retval = ext4_journal_ensure_credits(handle, EXT4_RESERVE_TRANS_BLOCKS,
+			ext4_free_metadata_revoke_credits(inode->i_sb, 1));
+	if (retval < 0)
+		return retval;
 	ext4_free_blocks(handle, inode, NULL, block, 1,
 			 EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET);
-	return retval;
+	return 0;
 }
 
 /*
@@ -427,6 +407,7 @@ static int free_ext_block(handle_t *handle, struct inode *inode)
 
 int ext4_ext_migrate(struct inode *inode)
 {
+	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 	handle_t *handle;
 	int retval = 0, i;
 	__le32 *i_data;
@@ -451,6 +432,8 @@ int ext4_ext_migrate(struct inode *inode)
 		 */
 		return retval;
 
+	percpu_down_write(&sbi->s_writepages_rwsem);
+
 	/*
 	 * Worst case we can touch the allocation bitmaps, a bgd
 	 * block, and a block to link in the orphan list.  We do need
@@ -461,7 +444,7 @@ int ext4_ext_migrate(struct inode *inode)
 
 	if (IS_ERR(handle)) {
 		retval = PTR_ERR(handle);
-		return retval;
+		goto out_unlock;
 	}
 	goal = (((inode->i_ino - 1) / EXT4_INODES_PER_GROUP(inode->i_sb)) *
 		EXT4_INODES_PER_GROUP(inode->i_sb)) + 1;
@@ -472,7 +455,7 @@ int ext4_ext_migrate(struct inode *inode)
 	if (IS_ERR(tmp_inode)) {
 		retval = PTR_ERR(tmp_inode);
 		ext4_journal_stop(handle);
-		return retval;
+		goto out_unlock;
 	}
 	i_size_write(tmp_inode, i_size_read(inode));
 	/*
@@ -514,7 +497,7 @@ int ext4_ext_migrate(struct inode *inode)
 		 */
 		ext4_orphan_del(NULL, tmp_inode);
 		retval = PTR_ERR(handle);
-		goto out;
+		goto out_tmp_inode;
 	}
 
 	ei = EXT4_I(inode);
@@ -574,9 +557,9 @@ err_out:
 	}
 
 	/* We mark the tmp_inode dirty via ext4_ext_tree_init. */
-	if (ext4_journal_extend(handle, 1) != 0)
-		ext4_journal_restart(handle, 1);
-
+	retval = ext4_journal_ensure_credits(handle, 1, 0);
+	if (retval < 0)
+		goto out_stop;
 	/*
 	 * Mark the tmp_inode as of size zero
 	 */
@@ -594,11 +577,13 @@ err_out:
 
 	/* Reset the extent details */
 	ext4_ext_tree_init(handle, tmp_inode);
+out_stop:
 	ext4_journal_stop(handle);
-out:
+out_tmp_inode:
 	unlock_new_inode(tmp_inode);
 	iput(tmp_inode);
-
+out_unlock:
+	percpu_up_write(&sbi->s_writepages_rwsem);
 	return retval;
 }
 
@@ -608,7 +593,8 @@ out:
 int ext4_ind_migrate(struct inode *inode)
 {
 	struct ext4_extent_header	*eh;
-	struct ext4_super_block		*es = EXT4_SB(inode->i_sb)->s_es;
+	struct ext4_sb_info		*sbi = EXT4_SB(inode->i_sb);
+	struct ext4_super_block		*es = sbi->s_es;
 	struct ext4_inode_info		*ei = EXT4_I(inode);
 	struct ext4_extent		*ex;
 	unsigned int			i, len;
@@ -632,9 +618,13 @@ int ext4_ind_migrate(struct inode *inode)
 	if (test_opt(inode->i_sb, DELALLOC))
 		ext4_alloc_da_blocks(inode);
 
+	percpu_down_write(&sbi->s_writepages_rwsem);
+
 	handle = ext4_journal_start(inode, EXT4_HT_MIGRATE, 1);
-	if (IS_ERR(handle))
-		return PTR_ERR(handle);
+	if (IS_ERR(handle)) {
+		ret = PTR_ERR(handle);
+		goto out_unlock;
+	}
 
 	down_write(&EXT4_I(inode)->i_data_sem);
 	ret = ext4_ext_check_inode(inode);
@@ -669,5 +659,7 @@ int ext4_ind_migrate(struct inode *inode)
 errout:
 	ext4_journal_stop(handle);
 	up_write(&EXT4_I(inode)->i_data_sem);
+out_unlock:
+	percpu_up_write(&sbi->s_writepages_rwsem);
 	return ret;
 }
diff --git a/fs/ext4/mmp.c b/fs/ext4/mmp.c
index 2305b4374fd3..d34cb8c46655 100644
--- a/fs/ext4/mmp.c
+++ b/fs/ext4/mmp.c
@@ -120,10 +120,10 @@ void __dump_mmp_msg(struct super_block *sb, struct mmp_struct *mmp,
 {
 	__ext4_warning(sb, function, line, "%s", msg);
 	__ext4_warning(sb, function, line,
-		       "MMP failure info: last update time: %llu, last update "
-		       "node: %s, last update device: %s",
-		       (long long unsigned int) le64_to_cpu(mmp->mmp_time),
-		       mmp->mmp_nodename, mmp->mmp_bdevname);
+		       "MMP failure info: last update time: %llu, last update node: %.*s, last update device: %.*s",
+		       (unsigned long long)le64_to_cpu(mmp->mmp_time),
+		       (int)sizeof(mmp->mmp_nodename), mmp->mmp_nodename,
+		       (int)sizeof(mmp->mmp_bdevname), mmp->mmp_bdevname);
 }
 
 /*
@@ -154,6 +154,7 @@ static int kmmpd(void *data)
 	mmp_check_interval = max(EXT4_MMP_CHECK_MULT * mmp_update_interval,
 				 EXT4_MMP_MIN_CHECK_INTERVAL);
 	mmp->mmp_check_interval = cpu_to_le16(mmp_check_interval);
+	BUILD_BUG_ON(sizeof(mmp->mmp_bdevname) < BDEVNAME_SIZE);
 	bdevname(bh->b_bdev, mmp->mmp_bdevname);
 
 	memcpy(mmp->mmp_nodename, init_utsname()->nodename,
@@ -173,8 +174,10 @@ static int kmmpd(void *data)
 		 * (s_mmp_update_interval * 60) seconds.
 		 */
 		if (retval) {
-			if ((failed_writes % 60) == 0)
-				ext4_error(sb, "Error writing to MMP block");
+			if ((failed_writes % 60) == 0) {
+				ext4_error_err(sb, -retval,
+					       "Error writing to MMP block");
+			}
 			failed_writes++;
 		}
 
@@ -205,8 +208,9 @@ static int kmmpd(void *data)
 
 			retval = read_mmp_block(sb, &bh_check, mmp_block);
 			if (retval) {
-				ext4_error(sb, "error reading MMP data: %d",
-					   retval);
+				ext4_error_err(sb, -retval,
+					       "error reading MMP data: %d",
+					       retval);
 				goto exit_thread;
 			}
 
@@ -218,7 +222,7 @@ static int kmmpd(void *data)
 					     "Error while updating MMP info. "
 					     "The filesystem seems to have been"
 					     " multiply mounted.");
-				ext4_error(sb, "abort");
+				ext4_error_err(sb, EBUSY, "abort");
 				put_bh(bh_check);
 				retval = -EBUSY;
 				goto exit_thread;
@@ -375,7 +379,8 @@ skip:
 	/*
 	 * Start a kernel thread to update the MMP block periodically.
 	 */
-	EXT4_SB(sb)->s_mmp_tsk = kthread_run(kmmpd, mmpd_data, "kmmpd-%s",
+	EXT4_SB(sb)->s_mmp_tsk = kthread_run(kmmpd, mmpd_data, "kmmpd-%.*s",
+					     (int)sizeof(mmp->mmp_bdevname),
 					     bdevname(bh->b_bdev,
 						      mmp->mmp_bdevname));
 	if (IS_ERR(EXT4_SB(sb)->s_mmp_tsk)) {
diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c
index 30ce3dc69378..1ed86fb6c302 100644
--- a/fs/ext4/move_extent.c
+++ b/fs/ext4/move_extent.c
@@ -422,8 +422,8 @@ repair_branches:
 					   block_len_in_page, 0, &err2);
 	ext4_double_up_write_data_sem(orig_inode, donor_inode);
 	if (replaced_count != block_len_in_page) {
-		EXT4_ERROR_INODE_BLOCK(orig_inode, (sector_t)(orig_blk_offset),
-				       "Unable to copy data block,"
+		ext4_error_inode_block(orig_inode, (sector_t)(orig_blk_offset),
+				       EIO, "Unable to copy data block,"
 				       " data will be lost.");
 		*err = -EIO;
 	}
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index a427d2031a8d..a8aca4772aaa 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -109,7 +109,10 @@ static struct buffer_head *__ext4_read_dirblock(struct inode *inode,
 	struct ext4_dir_entry *dirent;
 	int is_dx_block = 0;
 
-	bh = ext4_bread(NULL, inode, block, 0);
+	if (ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_EIO))
+		bh = ERR_PTR(-EIO);
+	else
+		bh = ext4_bread(NULL, inode, block, 0);
 	if (IS_ERR(bh)) {
 		__ext4_warning(inode->i_sb, func, line,
 			       "inode #%lu: lblock %lu: comm %s: "
@@ -153,21 +156,25 @@ static struct buffer_head *__ext4_read_dirblock(struct inode *inode,
 	 * caller is sure it should be an index block.
 	 */
 	if (is_dx_block && type == INDEX) {
-		if (ext4_dx_csum_verify(inode, dirent))
+		if (ext4_dx_csum_verify(inode, dirent) &&
+		    !ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_CRC))
 			set_buffer_verified(bh);
 		else {
-			ext4_error_inode(inode, func, line, block,
-					 "Directory index failed checksum");
+			ext4_error_inode_err(inode, func, line, block,
+					     EFSBADCRC,
+					     "Directory index failed checksum");
 			brelse(bh);
 			return ERR_PTR(-EFSBADCRC);
 		}
 	}
 	if (!is_dx_block) {
-		if (ext4_dirblock_csum_verify(inode, bh))
+		if (ext4_dirblock_csum_verify(inode, bh) &&
+		    !ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_CRC))
 			set_buffer_verified(bh);
 		else {
-			ext4_error_inode(inode, func, line, block,
-					 "Directory block failed checksum");
+			ext4_error_inode_err(inode, func, line, block,
+					     EFSBADCRC,
+					     "Directory block failed checksum");
 			brelse(bh);
 			return ERR_PTR(-EFSBADCRC);
 		}
@@ -226,13 +233,13 @@ struct dx_root
 		u8 unused_flags;
 	}
 	info;
-	struct dx_entry	entries[0];
+	struct dx_entry	entries[];
 };
 
 struct dx_node
 {
 	struct fake_dirent fake;
-	struct dx_entry	entries[0];
+	struct dx_entry	entries[];
 };
 
 
@@ -1002,7 +1009,6 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 	top = (struct ext4_dir_entry_2 *) ((char *) de +
 					   dir->i_sb->s_blocksize -
 					   EXT4_DIR_REC_LEN(0));
-#ifdef CONFIG_FS_ENCRYPTION
 	/* Check if the directory is encrypted */
 	if (IS_ENCRYPTED(dir)) {
 		err = fscrypt_get_encryption_info(dir);
@@ -1017,7 +1023,7 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 			return err;
 		}
 	}
-#endif
+
 	for (; de < top; de = ext4_next_entry(de, dir->i_sb->s_blocksize)) {
 		if (ext4_check_dir_entry(dir, NULL, de, bh,
 				bh->b_data, bh->b_size,
@@ -1065,9 +1071,7 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 	}
 errout:
 	brelse(bh);
-#ifdef CONFIG_FS_ENCRYPTION
 	fscrypt_fname_free_buffer(&fname_crypto_str);
-#endif
 	return count;
 }
 
@@ -1507,6 +1511,7 @@ restart:
 		/*
 		 * We deal with the read-ahead logic here.
 		 */
+		cond_resched();
 		if (ra_ptr >= ra_max) {
 			/* Refill the readahead buffer */
 			ra_ptr = 0;
@@ -1527,8 +1532,9 @@ restart:
 			goto next;
 		wait_on_buffer(bh);
 		if (!buffer_uptodate(bh)) {
-			EXT4_ERROR_INODE(dir, "reading directory lblock %lu",
-					 (unsigned long) block);
+			EXT4_ERROR_INODE_ERR(dir, EIO,
+					     "reading directory lblock %lu",
+					     (unsigned long) block);
 			brelse(bh);
 			ret = ERR_PTR(-EIO);
 			goto cleanup_and_exit;
@@ -1537,8 +1543,9 @@ restart:
 		    !is_dx_internal_node(dir, block,
 					 (struct ext4_dir_entry *)bh->b_data) &&
 		    !ext4_dirblock_csum_verify(dir, bh)) {
-			EXT4_ERROR_INODE(dir, "checksumming directory "
-					 "block %lu", (unsigned long)block);
+			EXT4_ERROR_INODE_ERR(dir, EFSBADCRC,
+					     "checksumming directory "
+					     "block %lu", (unsigned long)block);
 			brelse(bh);
 			ret = ERR_PTR(-EFSBADCRC);
 			goto cleanup_and_exit;
@@ -2164,7 +2171,9 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
 	struct buffer_head *bh = NULL;
 	struct ext4_dir_entry_2 *de;
 	struct super_block *sb;
+#ifdef CONFIG_UNICODE
 	struct ext4_sb_info *sbi;
+#endif
 	struct ext4_filename fname;
 	int	retval;
 	int	dx_fallback=0;
@@ -2176,12 +2185,12 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
 		csum_size = sizeof(struct ext4_dir_entry_tail);
 
 	sb = dir->i_sb;
-	sbi = EXT4_SB(sb);
 	blocksize = sb->s_blocksize;
 	if (!dentry->d_name.len)
 		return -EINVAL;
 
 #ifdef CONFIG_UNICODE
+	sbi = EXT4_SB(sb);
 	if (ext4_has_strict_mode(sbi) && IS_CASEFOLDED(dir) &&
 	    sbi->s_encoding && utf8_validate(sbi->s_encoding, &dentry->d_name))
 		return -EINVAL;
@@ -2205,6 +2214,13 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
 		retval = ext4_dx_add_entry(handle, &fname, dir, inode);
 		if (!retval || (retval != ERR_BAD_DX_DIR))
 			goto out;
+		/* Can we just ignore htree data? */
+		if (ext4_has_metadata_csum(sb)) {
+			EXT4_ERROR_INODE(dir,
+				"Directory has corrupted htree index.");
+			retval = -EFSCORRUPTED;
+			goto out;
+		}
 		ext4_clear_inode_flag(dir, EXT4_INODE_INDEX);
 		dx_fallback++;
 		ext4_mark_inode_dirty(handle, dir);
@@ -2547,18 +2563,29 @@ static void ext4_dec_count(handle_t *handle, struct inode *inode)
 }
 
 
+/*
+ * Add non-directory inode to a directory. On success, the inode reference is
+ * consumed by dentry is instantiation. This is also indicated by clearing of
+ * *inodep pointer. On failure, the caller is responsible for dropping the
+ * inode reference in the safe context.
+ */
 static int ext4_add_nondir(handle_t *handle,
-		struct dentry *dentry, struct inode *inode)
+		struct dentry *dentry, struct inode **inodep)
 {
+	struct inode *dir = d_inode(dentry->d_parent);
+	struct inode *inode = *inodep;
 	int err = ext4_add_entry(handle, dentry, inode);
 	if (!err) {
 		ext4_mark_inode_dirty(handle, inode);
+		if (IS_DIRSYNC(dir))
+			ext4_handle_sync(handle);
 		d_instantiate_new(dentry, inode);
+		*inodep = NULL;
 		return 0;
 	}
 	drop_nlink(inode);
+	ext4_orphan_add(handle, inode);
 	unlock_new_inode(inode);
-	iput(inode);
 	return err;
 }
 
@@ -2592,12 +2619,12 @@ retry:
 		inode->i_op = &ext4_file_inode_operations;
 		inode->i_fop = &ext4_file_operations;
 		ext4_set_aops(inode);
-		err = ext4_add_nondir(handle, dentry, inode);
-		if (!err && IS_DIRSYNC(dir))
-			ext4_handle_sync(handle);
+		err = ext4_add_nondir(handle, dentry, &inode);
 	}
 	if (handle)
 		ext4_journal_stop(handle);
+	if (!IS_ERR_OR_NULL(inode))
+		iput(inode);
 	if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries))
 		goto retry;
 	return err;
@@ -2624,12 +2651,12 @@ retry:
 	if (!IS_ERR(inode)) {
 		init_special_inode(inode, inode->i_mode, rdev);
 		inode->i_op = &ext4_special_inode_operations;
-		err = ext4_add_nondir(handle, dentry, inode);
-		if (!err && IS_DIRSYNC(dir))
-			ext4_handle_sync(handle);
+		err = ext4_add_nondir(handle, dentry, &inode);
 	}
 	if (handle)
 		ext4_journal_stop(handle);
+	if (!IS_ERR_OR_NULL(inode))
+		iput(inode);
 	if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries))
 		goto retry;
 	return err;
@@ -2779,10 +2806,12 @@ retry:
 	if (err) {
 out_clear_inode:
 		clear_nlink(inode);
+		ext4_orphan_add(handle, inode);
 		unlock_new_inode(inode);
 		ext4_mark_inode_dirty(handle, inode);
+		ext4_journal_stop(handle);
 		iput(inode);
-		goto out_stop;
+		goto out_retry;
 	}
 	ext4_inc_count(handle, dir);
 	ext4_update_dx_flag(dir);
@@ -2796,6 +2825,7 @@ out_clear_inode:
 out_stop:
 	if (handle)
 		ext4_journal_stop(handle);
+out_retry:
 	if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries))
 		goto retry;
 	return err;
@@ -2808,7 +2838,7 @@ bool ext4_empty_dir(struct inode *inode)
 {
 	unsigned int offset;
 	struct buffer_head *bh;
-	struct ext4_dir_entry_2 *de, *de1;
+	struct ext4_dir_entry_2 *de;
 	struct super_block *sb;
 
 	if (ext4_has_inline_data(inode)) {
@@ -2833,19 +2863,25 @@ bool ext4_empty_dir(struct inode *inode)
 		return true;
 
 	de = (struct ext4_dir_entry_2 *) bh->b_data;
-	de1 = ext4_next_entry(de, sb->s_blocksize);
-	if (le32_to_cpu(de->inode) != inode->i_ino ||
-			le32_to_cpu(de1->inode) == 0 ||
-			strcmp(".", de->name) || strcmp("..", de1->name)) {
-		ext4_warning_inode(inode, "directory missing '.' and/or '..'");
+	if (ext4_check_dir_entry(inode, NULL, de, bh, bh->b_data, bh->b_size,
+				 0) ||
+	    le32_to_cpu(de->inode) != inode->i_ino || strcmp(".", de->name)) {
+		ext4_warning_inode(inode, "directory missing '.'");
 		brelse(bh);
 		return true;
 	}
-	offset = ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize) +
-		 ext4_rec_len_from_disk(de1->rec_len, sb->s_blocksize);
-	de = ext4_next_entry(de1, sb->s_blocksize);
+	offset = ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize);
+	de = ext4_next_entry(de, sb->s_blocksize);
+	if (ext4_check_dir_entry(inode, NULL, de, bh, bh->b_data, bh->b_size,
+				 offset) ||
+	    le32_to_cpu(de->inode) == 0 || strcmp("..", de->name)) {
+		ext4_warning_inode(inode, "directory missing '..'");
+		brelse(bh);
+		return true;
+	}
+	offset += ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize);
 	while (offset < inode->i_size) {
-		if ((void *) de >= (void *) (bh->b_data+sb->s_blocksize)) {
+		if (!(offset & (sb->s_blocksize - 1))) {
 			unsigned int lblock;
 			brelse(bh);
 			lblock = offset >> EXT4_BLOCK_SIZE_BITS(sb);
@@ -2856,12 +2892,11 @@ bool ext4_empty_dir(struct inode *inode)
 			}
 			if (IS_ERR(bh))
 				return true;
-			de = (struct ext4_dir_entry_2 *) bh->b_data;
 		}
+		de = (struct ext4_dir_entry_2 *) (bh->b_data +
+					(offset & (sb->s_blocksize - 1)));
 		if (ext4_check_dir_entry(inode, NULL, de, bh,
 					 bh->b_data, bh->b_size, offset)) {
-			de = (struct ext4_dir_entry_2 *)(bh->b_data +
-							 sb->s_blocksize);
 			offset = (offset | (sb->s_blocksize - 1)) + 1;
 			continue;
 		}
@@ -2870,7 +2905,6 @@ bool ext4_empty_dir(struct inode *inode)
 			return false;
 		}
 		offset += ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize);
-		de = ext4_next_entry(de, sb->s_blocksize);
 	}
 	brelse(bh);
 	return true;
@@ -3182,18 +3216,17 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry)
 	if (IS_DIRSYNC(dir))
 		ext4_handle_sync(handle);
 
-	if (inode->i_nlink == 0) {
-		ext4_warning_inode(inode, "Deleting file '%.*s' with no links",
-				   dentry->d_name.len, dentry->d_name.name);
-		set_nlink(inode, 1);
-	}
 	retval = ext4_delete_entry(handle, dir, de, bh);
 	if (retval)
 		goto end_unlink;
 	dir->i_ctime = dir->i_mtime = current_time(dir);
 	ext4_update_dx_flag(dir);
 	ext4_mark_inode_dirty(handle, dir);
-	drop_nlink(inode);
+	if (inode->i_nlink == 0)
+		ext4_warning_inode(inode, "Deleting file '%.*s' with no links",
+				   dentry->d_name.len, dentry->d_name.name);
+	else
+		drop_nlink(inode);
 	if (!inode->i_nlink)
 		ext4_orphan_add(handle, inode);
 	inode->i_ctime = current_time(inode);
@@ -3328,12 +3361,11 @@ static int ext4_symlink(struct inode *dir,
 		inode->i_size = disk_link.len - 1;
 	}
 	EXT4_I(inode)->i_disksize = inode->i_size;
-	err = ext4_add_nondir(handle, dentry, inode);
-	if (!err && IS_DIRSYNC(dir))
-		ext4_handle_sync(handle);
-
+	err = ext4_add_nondir(handle, dentry, &inode);
 	if (handle)
 		ext4_journal_stop(handle);
+	if (inode)
+		iput(inode);
 	goto out_free_encrypted_link;
 
 err_drop_inode:
diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c
index 12ceadef32c5..de6fe969f773 100644
--- a/fs/ext4/page-io.c
+++ b/fs/ext4/page-io.c
@@ -31,18 +31,56 @@
 #include "acl.h"
 
 static struct kmem_cache *io_end_cachep;
+static struct kmem_cache *io_end_vec_cachep;
 
 int __init ext4_init_pageio(void)
 {
 	io_end_cachep = KMEM_CACHE(ext4_io_end, SLAB_RECLAIM_ACCOUNT);
 	if (io_end_cachep == NULL)
 		return -ENOMEM;
+
+	io_end_vec_cachep = KMEM_CACHE(ext4_io_end_vec, 0);
+	if (io_end_vec_cachep == NULL) {
+		kmem_cache_destroy(io_end_cachep);
+		return -ENOMEM;
+	}
 	return 0;
 }
 
 void ext4_exit_pageio(void)
 {
 	kmem_cache_destroy(io_end_cachep);
+	kmem_cache_destroy(io_end_vec_cachep);
+}
+
+struct ext4_io_end_vec *ext4_alloc_io_end_vec(ext4_io_end_t *io_end)
+{
+	struct ext4_io_end_vec *io_end_vec;
+
+	io_end_vec = kmem_cache_zalloc(io_end_vec_cachep, GFP_NOFS);
+	if (!io_end_vec)
+		return ERR_PTR(-ENOMEM);
+	INIT_LIST_HEAD(&io_end_vec->list);
+	list_add_tail(&io_end_vec->list, &io_end->list_vec);
+	return io_end_vec;
+}
+
+static void ext4_free_io_end_vec(ext4_io_end_t *io_end)
+{
+	struct ext4_io_end_vec *io_end_vec, *tmp;
+
+	if (list_empty(&io_end->list_vec))
+		return;
+	list_for_each_entry_safe(io_end_vec, tmp, &io_end->list_vec, list) {
+		list_del(&io_end_vec->list);
+		kmem_cache_free(io_end_vec_cachep, io_end_vec);
+	}
+}
+
+struct ext4_io_end_vec *ext4_last_io_end_vec(ext4_io_end_t *io_end)
+{
+	BUG_ON(list_empty(&io_end->list_vec));
+	return list_last_entry(&io_end->list_vec, struct ext4_io_end_vec, list);
 }
 
 /*
@@ -87,11 +125,10 @@ static void ext4_finish_bio(struct bio *bio)
 		}
 		bh = head = page_buffers(page);
 		/*
-		 * We check all buffers in the page under BH_Uptodate_Lock
+		 * We check all buffers in the page under b_uptodate_lock
 		 * to avoid races with other end io clearing async_write flags
 		 */
-		local_irq_save(flags);
-		bit_spin_lock(BH_Uptodate_Lock, &head->b_state);
+		spin_lock_irqsave(&head->b_uptodate_lock, flags);
 		do {
 			if (bh_offset(bh) < bio_start ||
 			    bh_offset(bh) + bh->b_size > bio_end) {
@@ -103,8 +140,7 @@ static void ext4_finish_bio(struct bio *bio)
 			if (bio->bi_status)
 				buffer_io_error(bh);
 		} while ((bh = bh->b_this_page) != head);
-		bit_spin_unlock(BH_Uptodate_Lock, &head->b_state);
-		local_irq_restore(flags);
+		spin_unlock_irqrestore(&head->b_uptodate_lock, flags);
 		if (!under_io) {
 			fscrypt_free_bounce_page(bounce_page);
 			end_page_writeback(page);
@@ -125,6 +161,7 @@ static void ext4_release_io_end(ext4_io_end_t *io_end)
 		ext4_finish_bio(bio);
 		bio_put(bio);
 	}
+	ext4_free_io_end_vec(io_end);
 	kmem_cache_free(io_end_cachep, io_end);
 }
 
@@ -136,29 +173,26 @@ static void ext4_release_io_end(ext4_io_end_t *io_end)
  * cannot get to ext4_ext_truncate() before all IOs overlapping that range are
  * completed (happens from ext4_free_ioend()).
  */
-static int ext4_end_io(ext4_io_end_t *io)
+static int ext4_end_io_end(ext4_io_end_t *io_end)
 {
-	struct inode *inode = io->inode;
-	loff_t offset = io->offset;
-	ssize_t size = io->size;
-	handle_t *handle = io->handle;
+	struct inode *inode = io_end->inode;
+	handle_t *handle = io_end->handle;
 	int ret = 0;
 
-	ext4_debug("ext4_end_io_nolock: io 0x%p from inode %lu,list->next 0x%p,"
+	ext4_debug("ext4_end_io_nolock: io_end 0x%p from inode %lu,list->next 0x%p,"
 		   "list->prev 0x%p\n",
-		   io, inode->i_ino, io->list.next, io->list.prev);
+		   io_end, inode->i_ino, io_end->list.next, io_end->list.prev);
 
-	io->handle = NULL;	/* Following call will use up the handle */
-	ret = ext4_convert_unwritten_extents(handle, inode, offset, size);
+	io_end->handle = NULL;	/* Following call will use up the handle */
+	ret = ext4_convert_unwritten_io_end_vec(handle, io_end);
 	if (ret < 0 && !ext4_forced_shutdown(EXT4_SB(inode->i_sb))) {
 		ext4_msg(inode->i_sb, KERN_EMERG,
 			 "failed to convert unwritten extents to written "
 			 "extents -- potential data loss!  "
-			 "(inode %lu, offset %llu, size %zd, error %d)",
-			 inode->i_ino, offset, size, ret);
+			 "(inode %lu, error %d)", inode->i_ino, ret);
 	}
-	ext4_clear_io_unwritten_flag(io);
-	ext4_release_io_end(io);
+	ext4_clear_io_unwritten_flag(io_end);
+	ext4_release_io_end(io_end);
 	return ret;
 }
 
@@ -166,21 +200,21 @@ static void dump_completed_IO(struct inode *inode, struct list_head *head)
 {
 #ifdef	EXT4FS_DEBUG
 	struct list_head *cur, *before, *after;
-	ext4_io_end_t *io, *io0, *io1;
+	ext4_io_end_t *io_end, *io_end0, *io_end1;
 
 	if (list_empty(head))
 		return;
 
 	ext4_debug("Dump inode %lu completed io list\n", inode->i_ino);
-	list_for_each_entry(io, head, list) {
-		cur = &io->list;
+	list_for_each_entry(io_end, head, list) {
+		cur = &io_end->list;
 		before = cur->prev;
-		io0 = container_of(before, ext4_io_end_t, list);
+		io_end0 = container_of(before, ext4_io_end_t, list);
 		after = cur->next;
-		io1 = container_of(after, ext4_io_end_t, list);
+		io_end1 = container_of(after, ext4_io_end_t, list);
 
 		ext4_debug("io 0x%p from inode %lu,prev 0x%p,next 0x%p\n",
-			    io, inode->i_ino, io0, io1);
+			    io_end, inode->i_ino, io_end0, io_end1);
 	}
 #endif
 }
@@ -207,7 +241,7 @@ static void ext4_add_complete_io(ext4_io_end_t *io_end)
 static int ext4_do_flush_completed_IO(struct inode *inode,
 				      struct list_head *head)
 {
-	ext4_io_end_t *io;
+	ext4_io_end_t *io_end;
 	struct list_head unwritten;
 	unsigned long flags;
 	struct ext4_inode_info *ei = EXT4_I(inode);
@@ -219,11 +253,11 @@ static int ext4_do_flush_completed_IO(struct inode *inode,
 	spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
 
 	while (!list_empty(&unwritten)) {
-		io = list_entry(unwritten.next, ext4_io_end_t, list);
-		BUG_ON(!(io->flag & EXT4_IO_END_UNWRITTEN));
-		list_del_init(&io->list);
+		io_end = list_entry(unwritten.next, ext4_io_end_t, list);
+		BUG_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN));
+		list_del_init(&io_end->list);
 
-		err = ext4_end_io(io);
+		err = ext4_end_io_end(io_end);
 		if (unlikely(!ret && err))
 			ret = err;
 	}
@@ -242,19 +276,22 @@ void ext4_end_io_rsv_work(struct work_struct *work)
 
 ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags)
 {
-	ext4_io_end_t *io = kmem_cache_zalloc(io_end_cachep, flags);
-	if (io) {
-		io->inode = inode;
-		INIT_LIST_HEAD(&io->list);
-		atomic_set(&io->count, 1);
+	ext4_io_end_t *io_end = kmem_cache_zalloc(io_end_cachep, flags);
+
+	if (io_end) {
+		io_end->inode = inode;
+		INIT_LIST_HEAD(&io_end->list);
+		INIT_LIST_HEAD(&io_end->list_vec);
+		atomic_set(&io_end->count, 1);
 	}
-	return io;
+	return io_end;
 }
 
 void ext4_put_io_end_defer(ext4_io_end_t *io_end)
 {
 	if (atomic_dec_and_test(&io_end->count)) {
-		if (!(io_end->flag & EXT4_IO_END_UNWRITTEN) || !io_end->size) {
+		if (!(io_end->flag & EXT4_IO_END_UNWRITTEN) ||
+				list_empty(&io_end->list_vec)) {
 			ext4_release_io_end(io_end);
 			return;
 		}
@@ -268,9 +305,8 @@ int ext4_put_io_end(ext4_io_end_t *io_end)
 
 	if (atomic_dec_and_test(&io_end->count)) {
 		if (io_end->flag & EXT4_IO_END_UNWRITTEN) {
-			err = ext4_convert_unwritten_extents(io_end->handle,
-						io_end->inode, io_end->offset,
-						io_end->size);
+			err = ext4_convert_unwritten_io_end_vec(io_end->handle,
+								io_end);
 			io_end->handle = NULL;
 			ext4_clear_io_unwritten_flag(io_end);
 		}
@@ -307,10 +343,8 @@ static void ext4_end_bio(struct bio *bio)
 		struct inode *inode = io_end->inode;
 
 		ext4_warning(inode->i_sb, "I/O error %d writing to inode %lu "
-			     "(offset %llu size %ld starting block %llu)",
+			     "starting block %llu)",
 			     bio->bi_status, inode->i_ino,
-			     (unsigned long long) io_end->offset,
-			     (long) io_end->size,
 			     (unsigned long long)
 			     bi_sector >> (inode->i_blkbits - 9));
 		mapping_set_error(inode->i_mapping,
@@ -358,14 +392,16 @@ void ext4_io_submit_init(struct ext4_io_submit *io,
 	io->io_end = NULL;
 }
 
-static int io_submit_init_bio(struct ext4_io_submit *io,
-			      struct buffer_head *bh)
+static void io_submit_init_bio(struct ext4_io_submit *io,
+			       struct buffer_head *bh)
 {
 	struct bio *bio;
 
+	/*
+	 * bio_alloc will _always_ be able to allocate a bio if
+	 * __GFP_DIRECT_RECLAIM is set, see comments for bio_alloc_bioset().
+	 */
 	bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES);
-	if (!bio)
-		return -ENOMEM;
 	bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9);
 	bio_set_dev(bio, bh->b_bdev);
 	bio->bi_end_io = ext4_end_bio;
@@ -373,13 +409,12 @@ static int io_submit_init_bio(struct ext4_io_submit *io,
 	io->io_bio = bio;
 	io->io_next_block = bh->b_blocknr;
 	wbc_init_bio(io->io_wbc, bio);
-	return 0;
 }
 
-static int io_submit_add_bh(struct ext4_io_submit *io,
-			    struct inode *inode,
-			    struct page *page,
-			    struct buffer_head *bh)
+static void io_submit_add_bh(struct ext4_io_submit *io,
+			     struct inode *inode,
+			     struct page *page,
+			     struct buffer_head *bh)
 {
 	int ret;
 
@@ -388,9 +423,7 @@ submit_and_retry:
 		ext4_io_submit(io);
 	}
 	if (io->io_bio == NULL) {
-		ret = io_submit_init_bio(io, bh);
-		if (ret)
-			return ret;
+		io_submit_init_bio(io, bh);
 		io->io_bio->bi_write_hint = inode->i_write_hint;
 	}
 	ret = bio_add_page(io->io_bio, page, bh->b_size, bh_offset(bh));
@@ -398,7 +431,6 @@ submit_and_retry:
 		goto submit_and_retry;
 	wbc_account_cgroup_owner(io->io_wbc, page, bh->b_size);
 	io->io_next_block++;
-	return 0;
 }
 
 int ext4_bio_write_page(struct ext4_io_submit *io,
@@ -478,21 +510,36 @@ int ext4_bio_write_page(struct ext4_io_submit *io,
 		gfp_t gfp_flags = GFP_NOFS;
 		unsigned int enc_bytes = round_up(len, i_blocksize(inode));
 
+		/*
+		 * Since bounce page allocation uses a mempool, we can only use
+		 * a waiting mask (i.e. request guaranteed allocation) on the
+		 * first page of the bio.  Otherwise it can deadlock.
+		 */
+		if (io->io_bio)
+			gfp_flags = GFP_NOWAIT | __GFP_NOWARN;
 	retry_encrypt:
 		bounce_page = fscrypt_encrypt_pagecache_blocks(page, enc_bytes,
 							       0, gfp_flags);
 		if (IS_ERR(bounce_page)) {
 			ret = PTR_ERR(bounce_page);
-			if (ret == -ENOMEM && wbc->sync_mode == WB_SYNC_ALL) {
-				if (io->io_bio) {
+			if (ret == -ENOMEM &&
+			    (io->io_bio || wbc->sync_mode == WB_SYNC_ALL)) {
+				gfp_flags = GFP_NOFS;
+				if (io->io_bio)
 					ext4_io_submit(io);
-					congestion_wait(BLK_RW_ASYNC, HZ/50);
-				}
-				gfp_flags |= __GFP_NOFAIL;
+				else
+					gfp_flags |= __GFP_NOFAIL;
+				congestion_wait(BLK_RW_ASYNC, HZ/50);
 				goto retry_encrypt;
 			}
-			bounce_page = NULL;
-			goto out;
+
+			printk_ratelimited(KERN_ERR "%s: ret = %d\n", __func__, ret);
+			redirty_page_for_writepage(wbc, page);
+			do {
+				clear_buffer_async_write(bh);
+				bh = bh->b_this_page;
+			} while (bh != head);
+			goto unlock;
 		}
 	}
 
@@ -500,30 +547,13 @@ int ext4_bio_write_page(struct ext4_io_submit *io,
 	do {
 		if (!buffer_async_write(bh))
 			continue;
-		ret = io_submit_add_bh(io, inode, bounce_page ?: page, bh);
-		if (ret) {
-			/*
-			 * We only get here on ENOMEM.  Not much else
-			 * we can do but mark the page as dirty, and
-			 * better luck next time.
-			 */
-			break;
-		}
+		io_submit_add_bh(io, inode,
+				 bounce_page ? bounce_page : page, bh);
 		nr_submitted++;
 		clear_buffer_dirty(bh);
 	} while ((bh = bh->b_this_page) != head);
 
-	/* Error stopped previous loop? Clean up buffers... */
-	if (ret) {
-	out:
-		fscrypt_free_bounce_page(bounce_page);
-		printk_ratelimited(KERN_ERR "%s: ret = %d\n", __func__, ret);
-		redirty_page_for_writepage(wbc, page);
-		do {
-			clear_buffer_async_write(bh);
-			bh = bh->b_this_page;
-		} while (bh != head);
-	}
+unlock:
 	unlock_page(page);
 	/* Nothing submitted - we have to end page writeback */
 	if (!nr_submitted)
diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c
index a30b203fa461..5761e9961682 100644
--- a/fs/ext4/readpage.c
+++ b/fs/ext4/readpage.c
@@ -7,8 +7,8 @@
  *
  * This was originally taken from fs/mpage.c
  *
- * The intent is the ext4_mpage_readpages() function here is intended
- * to replace mpage_readpages() in the general case, not just for
+ * The ext4_mpage_readpages() function here is intended to
+ * replace mpage_readahead() in the general case, not just for
  * encrypted files.  It has some limitations (see below), where it
  * will fall back to read_block_full_page(), but these limitations
  * should only be hit when page_size != block_size.
@@ -57,6 +57,7 @@ enum bio_post_read_step {
 	STEP_INITIAL = 0,
 	STEP_DECRYPT,
 	STEP_VERITY,
+	STEP_MAX,
 };
 
 struct bio_post_read_ctx {
@@ -106,10 +107,22 @@ static void verity_work(struct work_struct *work)
 {
 	struct bio_post_read_ctx *ctx =
 		container_of(work, struct bio_post_read_ctx, work);
+	struct bio *bio = ctx->bio;
 
-	fsverity_verify_bio(ctx->bio);
+	/*
+	 * fsverity_verify_bio() may call readpages() again, and although verity
+	 * will be disabled for that, decryption may still be needed, causing
+	 * another bio_post_read_ctx to be allocated.  So to guarantee that
+	 * mempool_alloc() never deadlocks we must free the current ctx first.
+	 * This is safe because verity is the last post-read step.
+	 */
+	BUILD_BUG_ON(STEP_VERITY + 1 != STEP_MAX);
+	mempool_free(ctx, bio_post_read_ctx_pool);
+	bio->bi_private = NULL;
 
-	bio_post_read_processing(ctx);
+	fsverity_verify_bio(bio);
+
+	__read_end_io(bio);
 }
 
 static void bio_post_read_processing(struct bio_post_read_ctx *ctx)
@@ -176,12 +189,11 @@ static inline bool ext4_need_verity(const struct inode *inode, pgoff_t idx)
 	       idx < DIV_ROUND_UP(inode->i_size, PAGE_SIZE);
 }
 
-static struct bio_post_read_ctx *get_bio_post_read_ctx(struct inode *inode,
-						       struct bio *bio,
-						       pgoff_t first_idx)
+static void ext4_set_bio_post_read_ctx(struct bio *bio,
+				       const struct inode *inode,
+				       pgoff_t first_idx)
 {
 	unsigned int post_read_steps = 0;
-	struct bio_post_read_ctx *ctx = NULL;
 
 	if (IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode))
 		post_read_steps |= 1 << STEP_DECRYPT;
@@ -190,14 +202,14 @@ static struct bio_post_read_ctx *get_bio_post_read_ctx(struct inode *inode,
 		post_read_steps |= 1 << STEP_VERITY;
 
 	if (post_read_steps) {
-		ctx = mempool_alloc(bio_post_read_ctx_pool, GFP_NOFS);
-		if (!ctx)
-			return ERR_PTR(-ENOMEM);
+		/* Due to the mempool, this never fails. */
+		struct bio_post_read_ctx *ctx =
+			mempool_alloc(bio_post_read_ctx_pool, GFP_NOFS);
+
 		ctx->bio = bio;
 		ctx->enabled_steps = post_read_steps;
 		bio->bi_private = ctx;
 	}
-	return ctx;
 }
 
 static inline loff_t ext4_readpage_limit(struct inode *inode)
@@ -209,14 +221,12 @@ static inline loff_t ext4_readpage_limit(struct inode *inode)
 	return i_size_read(inode);
 }
 
-int ext4_mpage_readpages(struct address_space *mapping,
-			 struct list_head *pages, struct page *page,
-			 unsigned nr_pages, bool is_readahead)
+int ext4_mpage_readpages(struct inode *inode,
+		struct readahead_control *rac, struct page *page)
 {
 	struct bio *bio = NULL;
 	sector_t last_block_in_bio = 0;
 
-	struct inode *inode = mapping->host;
 	const unsigned blkbits = inode->i_blkbits;
 	const unsigned blocks_per_page = PAGE_SIZE >> blkbits;
 	const unsigned blocksize = 1 << blkbits;
@@ -229,6 +239,7 @@ int ext4_mpage_readpages(struct address_space *mapping,
 	int length;
 	unsigned relative_block = 0;
 	struct ext4_map_blocks map;
+	unsigned int nr_pages = rac ? readahead_count(rac) : 1;
 
 	map.m_pblk = 0;
 	map.m_lblk = 0;
@@ -239,14 +250,9 @@ int ext4_mpage_readpages(struct address_space *mapping,
 		int fully_mapped = 1;
 		unsigned first_hole = blocks_per_page;
 
-		if (pages) {
-			page = lru_to_page(pages);
-
+		if (rac) {
+			page = readahead_page(rac);
 			prefetchw(&page->flags);
-			list_del(&page->lru);
-			if (add_to_page_cache_lru(page, mapping, page->index,
-				  readahead_gfp_mask(mapping)))
-				goto next_page;
 		}
 
 		if (page_has_buffers(page))
@@ -358,24 +364,18 @@ int ext4_mpage_readpages(struct address_space *mapping,
 			bio = NULL;
 		}
 		if (bio == NULL) {
-			struct bio_post_read_ctx *ctx;
-
+			/*
+			 * bio_alloc will _always_ be able to allocate a bio if
+			 * __GFP_DIRECT_RECLAIM is set, see bio_alloc_bioset().
+			 */
 			bio = bio_alloc(GFP_KERNEL,
 				min_t(int, nr_pages, BIO_MAX_PAGES));
-			if (!bio)
-				goto set_error_page;
-			ctx = get_bio_post_read_ctx(inode, bio, page->index);
-			if (IS_ERR(ctx)) {
-				bio_put(bio);
-				bio = NULL;
-				goto set_error_page;
-			}
+			ext4_set_bio_post_read_ctx(bio, inode, page->index);
 			bio_set_dev(bio, bdev);
 			bio->bi_iter.bi_sector = blocks[0] << (blkbits - 9);
 			bio->bi_end_io = mpage_end_io;
-			bio->bi_private = ctx;
 			bio_set_op_attrs(bio, REQ_OP_READ,
-						is_readahead ? REQ_RAHEAD : 0);
+						rac ? REQ_RAHEAD : 0);
 		}
 
 		length = first_hole << blkbits;
@@ -400,10 +400,9 @@ int ext4_mpage_readpages(struct address_space *mapping,
 		else
 			unlock_page(page);
 	next_page:
-		if (pages)
+		if (rac)
 			put_page(page);
 	}
-	BUG_ON(pages && !list_empty(pages));
 	if (bio)
 		submit_bio(bio);
 	return 0;
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index c0e9aef376a7..a50b51270ea9 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -17,6 +17,33 @@
 
 #include "ext4_jbd2.h"
 
+struct ext4_rcu_ptr {
+	struct rcu_head rcu;
+	void *ptr;
+};
+
+static void ext4_rcu_ptr_callback(struct rcu_head *head)
+{
+	struct ext4_rcu_ptr *ptr;
+
+	ptr = container_of(head, struct ext4_rcu_ptr, rcu);
+	kvfree(ptr->ptr);
+	kfree(ptr);
+}
+
+void ext4_kvfree_array_rcu(void *to_free)
+{
+	struct ext4_rcu_ptr *ptr = kzalloc(sizeof(*ptr), GFP_KERNEL);
+
+	if (ptr) {
+		ptr->ptr = to_free;
+		call_rcu(&ptr->rcu, ext4_rcu_ptr_callback);
+		return;
+	}
+	synchronize_rcu();
+	kvfree(to_free);
+}
+
 int ext4_resize_begin(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -388,28 +415,10 @@ static struct buffer_head *bclean(handle_t *handle, struct super_block *sb,
 	return bh;
 }
 
-/*
- * If we have fewer than thresh credits, extend by EXT4_MAX_TRANS_DATA.
- * If that fails, restart the transaction & regain write access for the
- * buffer head which is used for block_bitmap modifications.
- */
-static int extend_or_restart_transaction(handle_t *handle, int thresh)
+static int ext4_resize_ensure_credits_batch(handle_t *handle, int credits)
 {
-	int err;
-
-	if (ext4_handle_has_enough_credits(handle, thresh))
-		return 0;
-
-	err = ext4_journal_extend(handle, EXT4_MAX_TRANS_DATA);
-	if (err < 0)
-		return err;
-	if (err) {
-		err = ext4_journal_restart(handle, EXT4_MAX_TRANS_DATA);
-		if (err)
-			return err;
-	}
-
-	return 0;
+	return ext4_journal_ensure_credits_fn(handle, credits,
+		EXT4_MAX_TRANS_DATA, 0, 0);
 }
 
 /*
@@ -451,8 +460,8 @@ static int set_flexbg_block_bitmap(struct super_block *sb, handle_t *handle,
 			continue;
 		}
 
-		err = extend_or_restart_transaction(handle, 1);
-		if (err)
+		err = ext4_resize_ensure_credits_batch(handle, 1);
+		if (err < 0)
 			return err;
 
 		bh = sb_getblk(sb, flex_gd->groups[group].block_bitmap);
@@ -544,8 +553,8 @@ static int setup_new_flex_group_blocks(struct super_block *sb,
 			struct buffer_head *gdb;
 
 			ext4_debug("update backup group %#04llx\n", block);
-			err = extend_or_restart_transaction(handle, 1);
-			if (err)
+			err = ext4_resize_ensure_credits_batch(handle, 1);
+			if (err < 0)
 				goto out;
 
 			gdb = sb_getblk(sb, block);
@@ -560,8 +569,8 @@ static int setup_new_flex_group_blocks(struct super_block *sb,
 				brelse(gdb);
 				goto out;
 			}
-			memcpy(gdb->b_data, sbi->s_group_desc[j]->b_data,
-			       gdb->b_size);
+			memcpy(gdb->b_data, sbi_array_rcu_deref(sbi,
+				s_group_desc, j)->b_data, gdb->b_size);
 			set_buffer_uptodate(gdb);
 
 			err = ext4_handle_dirty_metadata(handle, NULL, gdb);
@@ -602,8 +611,8 @@ handle_bb:
 
 		/* Initialize block bitmap of the @group */
 		block = group_data[i].block_bitmap;
-		err = extend_or_restart_transaction(handle, 1);
-		if (err)
+		err = ext4_resize_ensure_credits_batch(handle, 1);
+		if (err < 0)
 			goto out;
 
 		bh = bclean(handle, sb, block);
@@ -631,8 +640,8 @@ handle_ib:
 
 		/* Initialize inode bitmap of the @group */
 		block = group_data[i].inode_bitmap;
-		err = extend_or_restart_transaction(handle, 1);
-		if (err)
+		err = ext4_resize_ensure_credits_batch(handle, 1);
+		if (err < 0)
 			goto out;
 		/* Mark unused entries in inode bitmap used */
 		bh = bclean(handle, sb, block);
@@ -842,9 +851,8 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	if (unlikely(err))
 		goto errout;
 
-	n_group_desc = ext4_kvmalloc((gdb_num + 1) *
-				     sizeof(struct buffer_head *),
-				     GFP_NOFS);
+	n_group_desc = kvmalloc((gdb_num + 1) * sizeof(struct buffer_head *),
+				GFP_KERNEL);
 	if (!n_group_desc) {
 		err = -ENOMEM;
 		ext4_warning(sb, "not enough memory for %lu groups",
@@ -879,13 +887,15 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	}
 	brelse(dind);
 
-	o_group_desc = EXT4_SB(sb)->s_group_desc;
+	rcu_read_lock();
+	o_group_desc = rcu_dereference(EXT4_SB(sb)->s_group_desc);
 	memcpy(n_group_desc, o_group_desc,
 	       EXT4_SB(sb)->s_gdb_count * sizeof(struct buffer_head *));
+	rcu_read_unlock();
 	n_group_desc[gdb_num] = gdb_bh;
-	EXT4_SB(sb)->s_group_desc = n_group_desc;
+	rcu_assign_pointer(EXT4_SB(sb)->s_group_desc, n_group_desc);
 	EXT4_SB(sb)->s_gdb_count++;
-	kvfree(o_group_desc);
+	ext4_kvfree_array_rcu(o_group_desc);
 
 	le16_add_cpu(&es->s_reserved_gdt_blocks, -1);
 	err = ext4_handle_dirty_super(handle, sb);
@@ -918,9 +928,8 @@ static int add_new_gdb_meta_bg(struct super_block *sb,
 	gdb_bh = ext4_sb_bread(sb, gdblock, 0);
 	if (IS_ERR(gdb_bh))
 		return PTR_ERR(gdb_bh);
-	n_group_desc = ext4_kvmalloc((gdb_num + 1) *
-				     sizeof(struct buffer_head *),
-				     GFP_NOFS);
+	n_group_desc = kvmalloc((gdb_num + 1) * sizeof(struct buffer_head *),
+				GFP_KERNEL);
 	if (!n_group_desc) {
 		brelse(gdb_bh);
 		err = -ENOMEM;
@@ -929,9 +938,11 @@ static int add_new_gdb_meta_bg(struct super_block *sb,
 		return err;
 	}
 
-	o_group_desc = EXT4_SB(sb)->s_group_desc;
+	rcu_read_lock();
+	o_group_desc = rcu_dereference(EXT4_SB(sb)->s_group_desc);
 	memcpy(n_group_desc, o_group_desc,
 	       EXT4_SB(sb)->s_gdb_count * sizeof(struct buffer_head *));
+	rcu_read_unlock();
 	n_group_desc[gdb_num] = gdb_bh;
 
 	BUFFER_TRACE(gdb_bh, "get_write_access");
@@ -942,9 +953,9 @@ static int add_new_gdb_meta_bg(struct super_block *sb,
 		return err;
 	}
 
-	EXT4_SB(sb)->s_group_desc = n_group_desc;
+	rcu_assign_pointer(EXT4_SB(sb)->s_group_desc, n_group_desc);
 	EXT4_SB(sb)->s_gdb_count++;
-	kvfree(o_group_desc);
+	ext4_kvfree_array_rcu(o_group_desc);
 	return err;
 }
 
@@ -1109,10 +1120,8 @@ static void update_backups(struct super_block *sb, sector_t blk_off, char *data,
 		ext4_fsblk_t backup_block;
 
 		/* Out of journal space, and can't get more - abort - so sad */
-		if (ext4_handle_valid(handle) &&
-		    handle->h_buffer_credits == 0 &&
-		    ext4_journal_extend(handle, EXT4_MAX_TRANS_DATA) &&
-		    (err = ext4_journal_restart(handle, EXT4_MAX_TRANS_DATA)))
+		err = ext4_resize_ensure_credits_batch(handle, 1);
+		if (err < 0)
 			break;
 
 		if (meta_bg == 0)
@@ -1210,7 +1219,8 @@ static int ext4_add_new_descs(handle_t *handle, struct super_block *sb,
 		 * use non-sparse filesystems anymore.  This is already checked above.
 		 */
 		if (gdb_off) {
-			gdb_bh = sbi->s_group_desc[gdb_num];
+			gdb_bh = sbi_array_rcu_deref(sbi, s_group_desc,
+						     gdb_num);
 			BUFFER_TRACE(gdb_bh, "get_write_access");
 			err = ext4_journal_get_write_access(handle, gdb_bh);
 
@@ -1292,7 +1302,7 @@ static int ext4_setup_new_descs(handle_t *handle, struct super_block *sb,
 		/*
 		 * get_write_access() has been called on gdb_bh by ext4_add_new_desc().
 		 */
-		gdb_bh = sbi->s_group_desc[gdb_num];
+		gdb_bh = sbi_array_rcu_deref(sbi, s_group_desc, gdb_num);
 		/* Update group descriptor block for new group */
 		gdp = (struct ext4_group_desc *)(gdb_bh->b_data +
 						 gdb_off * EXT4_DESC_SIZE(sb));
@@ -1420,11 +1430,14 @@ static void ext4_update_super(struct super_block *sb,
 		   percpu_counter_read(&sbi->s_freeclusters_counter));
 	if (ext4_has_feature_flex_bg(sb) && sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group;
+		struct flex_groups *fg;
+
 		flex_group = ext4_flex_group(sbi, group_data[0].group);
+		fg = sbi_array_rcu_deref(sbi, s_flex_groups, flex_group);
 		atomic64_add(EXT4_NUM_B2C(sbi, free_blocks),
-			     &sbi->s_flex_groups[flex_group].free_clusters);
+			     &fg->free_clusters);
 		atomic_add(EXT4_INODES_PER_GROUP(sb) * flex_gd->count,
-			   &sbi->s_flex_groups[flex_group].free_inodes);
+			   &fg->free_inodes);
 	}
 
 	/*
@@ -1519,7 +1532,8 @@ exit_journal:
 		for (; gdb_num <= gdb_num_end; gdb_num++) {
 			struct buffer_head *gdb_bh;
 
-			gdb_bh = sbi->s_group_desc[gdb_num];
+			gdb_bh = sbi_array_rcu_deref(sbi, s_group_desc,
+						     gdb_num);
 			if (old_gdb == gdb_bh->b_blocknr)
 				continue;
 			update_backups(sb, gdb_bh->b_blocknr, gdb_bh->b_data,
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index dd654e53ba3d..9824cd8203e8 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -43,7 +43,7 @@
 #include <linux/uaccess.h>
 #include <linux/iversion.h>
 #include <linux/unicode.h>
-
+#include <linux/part_stat.h>
 #include <linux/kthread.h>
 #include <linux/freezer.h>
 
@@ -154,7 +154,7 @@ ext4_sb_bread(struct super_block *sb, sector_t block, int op_flags)
 
 	if (bh == NULL)
 		return ERR_PTR(-ENOMEM);
-	if (buffer_uptodate(bh))
+	if (ext4_buffer_uptodate(bh))
 		return bh;
 	ll_rw_block(REQ_OP_READ, REQ_META | op_flags, 1, &bh);
 	wait_on_buffer(bh);
@@ -204,26 +204,6 @@ void ext4_superblock_csum_set(struct super_block *sb)
 	es->s_checksum = ext4_superblock_csum(sb, es);
 }
 
-void *ext4_kvmalloc(size_t size, gfp_t flags)
-{
-	void *ret;
-
-	ret = kmalloc(size, flags | __GFP_NOWARN);
-	if (!ret)
-		ret = __vmalloc(size, flags, PAGE_KERNEL);
-	return ret;
-}
-
-void *ext4_kvzalloc(size_t size, gfp_t flags)
-{
-	void *ret;
-
-	ret = kzalloc(size, flags | __GFP_NOWARN);
-	if (!ret)
-		ret = __vmalloc(size, flags | __GFP_ZERO, PAGE_KERNEL);
-	return ret;
-}
-
 ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
 			       struct ext4_group_desc *bg)
 {
@@ -355,10 +335,12 @@ static time64_t __ext4_get_tstamp(__le32 *lo, __u8 *hi)
 #define ext4_get_tstamp(es, tstamp) \
 	__ext4_get_tstamp(&(es)->tstamp, &(es)->tstamp ## _hi)
 
-static void __save_error_info(struct super_block *sb, const char *func,
-			    unsigned int line)
+static void __save_error_info(struct super_block *sb, int error,
+			      __u32 ino, __u64 block,
+			      const char *func, unsigned int line)
 {
 	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
+	int err;
 
 	EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
 	if (bdev_read_only(sb->s_bdev))
@@ -367,6 +349,62 @@ static void __save_error_info(struct super_block *sb, const char *func,
 	ext4_update_tstamp(es, s_last_error_time);
 	strncpy(es->s_last_error_func, func, sizeof(es->s_last_error_func));
 	es->s_last_error_line = cpu_to_le32(line);
+	es->s_last_error_ino = cpu_to_le32(ino);
+	es->s_last_error_block = cpu_to_le64(block);
+	switch (error) {
+	case EIO:
+		err = EXT4_ERR_EIO;
+		break;
+	case ENOMEM:
+		err = EXT4_ERR_ENOMEM;
+		break;
+	case EFSBADCRC:
+		err = EXT4_ERR_EFSBADCRC;
+		break;
+	case 0:
+	case EFSCORRUPTED:
+		err = EXT4_ERR_EFSCORRUPTED;
+		break;
+	case ENOSPC:
+		err = EXT4_ERR_ENOSPC;
+		break;
+	case ENOKEY:
+		err = EXT4_ERR_ENOKEY;
+		break;
+	case EROFS:
+		err = EXT4_ERR_EROFS;
+		break;
+	case EFBIG:
+		err = EXT4_ERR_EFBIG;
+		break;
+	case EEXIST:
+		err = EXT4_ERR_EEXIST;
+		break;
+	case ERANGE:
+		err = EXT4_ERR_ERANGE;
+		break;
+	case EOVERFLOW:
+		err = EXT4_ERR_EOVERFLOW;
+		break;
+	case EBUSY:
+		err = EXT4_ERR_EBUSY;
+		break;
+	case ENOTDIR:
+		err = EXT4_ERR_ENOTDIR;
+		break;
+	case ENOTEMPTY:
+		err = EXT4_ERR_ENOTEMPTY;
+		break;
+	case ESHUTDOWN:
+		err = EXT4_ERR_ESHUTDOWN;
+		break;
+	case EFAULT:
+		err = EXT4_ERR_EFAULT;
+		break;
+	default:
+		err = EXT4_ERR_UNKNOWN;
+	}
+	es->s_last_error_errcode = err;
 	if (!es->s_first_error_time) {
 		es->s_first_error_time = es->s_last_error_time;
 		es->s_first_error_time_hi = es->s_last_error_time_hi;
@@ -375,6 +413,7 @@ static void __save_error_info(struct super_block *sb, const char *func,
 		es->s_first_error_line = cpu_to_le32(line);
 		es->s_first_error_ino = es->s_last_error_ino;
 		es->s_first_error_block = es->s_last_error_block;
+		es->s_first_error_errcode = es->s_last_error_errcode;
 	}
 	/*
 	 * Start the daily error reporting function if it hasn't been
@@ -385,11 +424,13 @@ static void __save_error_info(struct super_block *sb, const char *func,
 	le32_add_cpu(&es->s_error_count, 1);
 }
 
-static void save_error_info(struct super_block *sb, const char *func,
-			    unsigned int line)
+static void save_error_info(struct super_block *sb, int error,
+			    __u32 ino, __u64 block,
+			    const char *func, unsigned int line)
 {
-	__save_error_info(sb, func, line);
-	ext4_commit_super(sb, 1);
+	__save_error_info(sb, error, ino, block, func, line);
+	if (!bdev_read_only(sb->s_bdev))
+		ext4_commit_super(sb, 1);
 }
 
 /*
@@ -494,7 +535,8 @@ static void ext4_handle_error(struct super_block *sb)
 			     "EXT4-fs error")
 
 void __ext4_error(struct super_block *sb, const char *function,
-		  unsigned int line, const char *fmt, ...)
+		  unsigned int line, int error, __u64 block,
+		  const char *fmt, ...)
 {
 	struct va_format vaf;
 	va_list args;
@@ -512,24 +554,21 @@ void __ext4_error(struct super_block *sb, const char *function,
 		       sb->s_id, function, line, current->comm, &vaf);
 		va_end(args);
 	}
-	save_error_info(sb, function, line);
+	save_error_info(sb, error, 0, block, function, line);
 	ext4_handle_error(sb);
 }
 
 void __ext4_error_inode(struct inode *inode, const char *function,
-			unsigned int line, ext4_fsblk_t block,
+			unsigned int line, ext4_fsblk_t block, int error,
 			const char *fmt, ...)
 {
 	va_list args;
 	struct va_format vaf;
-	struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
 
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
 		return;
 
 	trace_ext4_error(inode->i_sb, function, line);
-	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
-	es->s_last_error_block = cpu_to_le64(block);
 	if (ext4_error_ratelimit(inode->i_sb)) {
 		va_start(args, fmt);
 		vaf.fmt = fmt;
@@ -546,7 +585,8 @@ void __ext4_error_inode(struct inode *inode, const char *function,
 			       current->comm, &vaf);
 		va_end(args);
 	}
-	save_error_info(inode->i_sb, function, line);
+	save_error_info(inode->i_sb, error, inode->i_ino, block,
+			function, line);
 	ext4_handle_error(inode->i_sb);
 }
 
@@ -556,7 +596,6 @@ void __ext4_error_file(struct file *file, const char *function,
 {
 	va_list args;
 	struct va_format vaf;
-	struct ext4_super_block *es;
 	struct inode *inode = file_inode(file);
 	char pathname[80], *path;
 
@@ -564,8 +603,6 @@ void __ext4_error_file(struct file *file, const char *function,
 		return;
 
 	trace_ext4_error(inode->i_sb, function, line);
-	es = EXT4_SB(inode->i_sb)->s_es;
-	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
 	if (ext4_error_ratelimit(inode->i_sb)) {
 		path = file_path(file, pathname, sizeof(pathname));
 		if (IS_ERR(path))
@@ -587,7 +624,8 @@ void __ext4_error_file(struct file *file, const char *function,
 			       current->comm, path, &vaf);
 		va_end(args);
 	}
-	save_error_info(inode->i_sb, function, line);
+	save_error_info(inode->i_sb, EFSCORRUPTED, inode->i_ino, block,
+			function, line);
 	ext4_handle_error(inode->i_sb);
 }
 
@@ -655,7 +693,7 @@ void __ext4_std_error(struct super_block *sb, const char *function,
 		       sb->s_id, function, line, errstr);
 	}
 
-	save_error_info(sb, function, line);
+	save_error_info(sb, -errno, 0, 0, function, line);
 	ext4_handle_error(sb);
 }
 
@@ -670,7 +708,7 @@ void __ext4_std_error(struct super_block *sb, const char *function,
  */
 
 void __ext4_abort(struct super_block *sb, const char *function,
-		unsigned int line, const char *fmt, ...)
+		  unsigned int line, int error, const char *fmt, ...)
 {
 	struct va_format vaf;
 	va_list args;
@@ -678,7 +716,7 @@ void __ext4_abort(struct super_block *sb, const char *function,
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(sb))))
 		return;
 
-	save_error_info(sb, function, line);
+	save_error_info(sb, error, 0, 0, function, line);
 	va_start(args, fmt);
 	vaf.fmt = fmt;
 	vaf.va = &args;
@@ -697,7 +735,6 @@ void __ext4_abort(struct super_block *sb, const char *function,
 		sb->s_flags |= SB_RDONLY;
 		if (EXT4_SB(sb)->s_journal)
 			jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO);
-		save_error_info(sb, function, line);
 	}
 	if (test_opt(sb, ERRORS_PANIC) && !system_going_down()) {
 		if (EXT4_SB(sb)->s_journal &&
@@ -771,15 +808,12 @@ __acquires(bitlock)
 {
 	struct va_format vaf;
 	va_list args;
-	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
 
 	if (unlikely(ext4_forced_shutdown(EXT4_SB(sb))))
 		return;
 
 	trace_ext4_error(sb, function, line);
-	es->s_last_error_ino = cpu_to_le32(ino);
-	es->s_last_error_block = cpu_to_le64(block);
-	__save_error_info(sb, function, line);
+	__save_error_info(sb, EFSCORRUPTED, ino, block, function, line);
 
 	if (ext4_error_ratelimit(sb)) {
 		va_start(args, fmt);
@@ -883,7 +917,6 @@ void ext4_update_dynamic_rev(struct super_block *sb)
 static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb)
 {
 	struct block_device *bdev;
-	char b[BDEVNAME_SIZE];
 
 	bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, sb);
 	if (IS_ERR(bdev))
@@ -891,8 +924,9 @@ static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb)
 	return bdev;
 
 fail:
-	ext4_msg(sb, KERN_ERR, "failed to open journal device %s: %ld",
-			__bdevname(dev, b), PTR_ERR(bdev));
+	ext4_msg(sb, KERN_ERR,
+		 "failed to open journal device unknown-block(%u,%u) %ld",
+		 MAJOR(dev), MINOR(dev), PTR_ERR(bdev));
 	return NULL;
 }
 
@@ -970,6 +1004,8 @@ static void ext4_put_super(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct ext4_super_block *es = sbi->s_es;
+	struct buffer_head **group_desc;
+	struct flex_groups **flex_groups;
 	int aborted = 0;
 	int i, err;
 
@@ -978,15 +1014,22 @@ static void ext4_put_super(struct super_block *sb)
 
 	destroy_workqueue(sbi->rsv_conversion_wq);
 
+	/*
+	 * Unregister sysfs before destroying jbd2 journal.
+	 * Since we could still access attr_journal_task attribute via sysfs
+	 * path which could have sbi->s_journal->j_task as NULL
+	 */
+	ext4_unregister_sysfs(sb);
+
 	if (sbi->s_journal) {
 		aborted = is_journal_aborted(sbi->s_journal);
 		err = jbd2_journal_destroy(sbi->s_journal);
 		sbi->s_journal = NULL;
-		if ((err < 0) && !aborted)
-			ext4_abort(sb, "Couldn't clean up the journal");
+		if ((err < 0) && !aborted) {
+			ext4_abort(sb, -err, "Couldn't clean up the journal");
+		}
 	}
 
-	ext4_unregister_sysfs(sb);
 	ext4_es_unregister_shrinker(sbi);
 	del_timer_sync(&sbi->s_err_report);
 	ext4_release_system_zone(sb);
@@ -1000,15 +1043,23 @@ static void ext4_put_super(struct super_block *sb)
 	if (!sb_rdonly(sb))
 		ext4_commit_super(sb, 1);
 
+	rcu_read_lock();
+	group_desc = rcu_dereference(sbi->s_group_desc);
 	for (i = 0; i < sbi->s_gdb_count; i++)
-		brelse(sbi->s_group_desc[i]);
-	kvfree(sbi->s_group_desc);
-	kvfree(sbi->s_flex_groups);
+		brelse(group_desc[i]);
+	kvfree(group_desc);
+	flex_groups = rcu_dereference(sbi->s_flex_groups);
+	if (flex_groups) {
+		for (i = 0; i < sbi->s_flex_groups_allocated; i++)
+			kvfree(flex_groups[i]);
+		kvfree(flex_groups);
+	}
+	rcu_read_unlock();
 	percpu_counter_destroy(&sbi->s_freeclusters_counter);
 	percpu_counter_destroy(&sbi->s_freeinodes_counter);
 	percpu_counter_destroy(&sbi->s_dirs_counter);
 	percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
-	percpu_free_rwsem(&sbi->s_journal_flag_rwsem);
+	percpu_free_rwsem(&sbi->s_writepages_rwsem);
 #ifdef CONFIG_QUOTA
 	for (i = 0; i < EXT4_MAXQUOTAS; i++)
 		kfree(get_qf_name(sb, sbi, i));
@@ -1055,6 +1106,7 @@ static void ext4_put_super(struct super_block *sb)
 		crypto_free_shash(sbi->s_chksum_driver);
 	kfree(sbi->s_blockgroup_lock);
 	fs_put_dax(sbi->s_daxdev);
+	fscrypt_free_dummy_context(&sbi->s_dummy_enc_ctx);
 #ifdef CONFIG_UNICODE
 	utf8_unload(sbi->s_encoding);
 #endif
@@ -1085,8 +1137,6 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	ei->i_es_shk_nr = 0;
 	ei->i_es_shrink_lblk = 0;
 	ei->i_reserved_data_blocks = 0;
-	ei->i_da_metadata_calc_len = 0;
-	ei->i_da_metadata_calc_last_lblock = 0;
 	spin_lock_init(&(ei->i_block_reservation_lock));
 	ext4_init_pending_tree(&ei->i_pending_tree);
 #ifdef CONFIG_QUOTA
@@ -1172,9 +1222,9 @@ void ext4_clear_inode(struct inode *inode)
 {
 	invalidate_inode_buffers(inode);
 	clear_inode(inode);
-	dquot_drop(inode);
 	ext4_discard_preallocations(inode);
 	ext4_es_remove_extent(inode, 0, EXT_MAX_BLOCKS);
+	dquot_drop(inode);
 	if (EXT4_I(inode)->jinode) {
 		jbd2_journal_release_jbd_inode(EXT4_JOURNAL(inode),
 					       EXT4_I(inode)->jinode);
@@ -1340,18 +1390,33 @@ retry:
 	return res;
 }
 
-static bool ext4_dummy_context(struct inode *inode)
+static const union fscrypt_context *
+ext4_get_dummy_context(struct super_block *sb)
 {
-	return DUMMY_ENCRYPTION_ENABLED(EXT4_SB(inode->i_sb));
+	return EXT4_SB(sb)->s_dummy_enc_ctx.ctx;
+}
+
+static bool ext4_has_stable_inodes(struct super_block *sb)
+{
+	return ext4_has_feature_stable_inodes(sb);
+}
+
+static void ext4_get_ino_and_lblk_bits(struct super_block *sb,
+				       int *ino_bits_ret, int *lblk_bits_ret)
+{
+	*ino_bits_ret = 8 * sizeof(EXT4_SB(sb)->s_es->s_inodes_count);
+	*lblk_bits_ret = 8 * sizeof(ext4_lblk_t);
 }
 
 static const struct fscrypt_operations ext4_cryptops = {
 	.key_prefix		= "ext4:",
 	.get_context		= ext4_get_context,
 	.set_context		= ext4_set_context,
-	.dummy_context		= ext4_dummy_context,
+	.get_dummy_context	= ext4_get_dummy_context,
 	.empty_dir		= ext4_empty_dir,
 	.max_namelen		= EXT4_NAME_LEN,
+	.has_stable_inodes	= ext4_has_stable_inodes,
+	.get_ino_and_lblk_bits	= ext4_get_ino_and_lblk_bits,
 };
 #endif
 
@@ -1374,7 +1439,6 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type,
 static int ext4_quota_enable(struct super_block *sb, int type, int format_id,
 			     unsigned int flags);
 static int ext4_enable_quotas(struct super_block *sb);
-static int ext4_get_next_id(struct super_block *sb, struct kqid *qid);
 
 static struct dquot **ext4_get_dquots(struct inode *inode)
 {
@@ -1392,7 +1456,7 @@ static const struct dquot_operations ext4_quota_operations = {
 	.destroy_dquot		= dquot_destroy,
 	.get_projid		= ext4_get_projid,
 	.get_inode_usage	= ext4_get_inode_usage,
-	.get_next_id		= ext4_get_next_id,
+	.get_next_id		= dquot_get_next_id,
 };
 
 static const struct quotactl_ops ext4_qctl_operations = {
@@ -1535,6 +1599,7 @@ static const match_table_t tokens = {
 	{Opt_auto_da_alloc, "auto_da_alloc"},
 	{Opt_noauto_da_alloc, "noauto_da_alloc"},
 	{Opt_dioread_nolock, "dioread_nolock"},
+	{Opt_dioread_lock, "nodioread_nolock"},
 	{Opt_dioread_lock, "dioread_lock"},
 	{Opt_discard, "discard"},
 	{Opt_nodiscard, "nodiscard"},
@@ -1542,6 +1607,7 @@ static const match_table_t tokens = {
 	{Opt_init_itable, "init_itable"},
 	{Opt_noinit_itable, "noinit_itable"},
 	{Opt_max_dir_size_kb, "max_dir_size_kb=%u"},
+	{Opt_test_dummy_encryption, "test_dummy_encryption=%s"},
 	{Opt_test_dummy_encryption, "test_dummy_encryption"},
 	{Opt_nombcache, "nombcache"},
 	{Opt_nombcache, "no_mbcache"},	/* for backward compatibility */
@@ -1753,7 +1819,7 @@ static const struct mount_opts {
 	{Opt_jqfmt_vfsv0, QFMT_VFS_V0, MOPT_QFMT},
 	{Opt_jqfmt_vfsv1, QFMT_VFS_V1, MOPT_QFMT},
 	{Opt_max_dir_size_kb, 0, MOPT_GTE0},
-	{Opt_test_dummy_encryption, 0, MOPT_GTE0},
+	{Opt_test_dummy_encryption, 0, MOPT_STRING},
 	{Opt_nombcache, EXT4_MOUNT_NO_MBCACHE, MOPT_SET},
 	{Opt_err, 0, 0}
 };
@@ -1788,6 +1854,48 @@ static int ext4_sb_read_encoding(const struct ext4_super_block *es,
 }
 #endif
 
+static int ext4_set_test_dummy_encryption(struct super_block *sb,
+					  const char *opt,
+					  const substring_t *arg,
+					  bool is_remount)
+{
+#ifdef CONFIG_FS_ENCRYPTION
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	int err;
+
+	/*
+	 * This mount option is just for testing, and it's not worthwhile to
+	 * implement the extra complexity (e.g. RCU protection) that would be
+	 * needed to allow it to be set or changed during remount.  We do allow
+	 * it to be specified during remount, but only if there is no change.
+	 */
+	if (is_remount && !sbi->s_dummy_enc_ctx.ctx) {
+		ext4_msg(sb, KERN_WARNING,
+			 "Can't set test_dummy_encryption on remount");
+		return -1;
+	}
+	err = fscrypt_set_test_dummy_encryption(sb, arg, &sbi->s_dummy_enc_ctx);
+	if (err) {
+		if (err == -EEXIST)
+			ext4_msg(sb, KERN_WARNING,
+				 "Can't change test_dummy_encryption on remount");
+		else if (err == -EINVAL)
+			ext4_msg(sb, KERN_WARNING,
+				 "Value of option \"%s\" is unrecognized", opt);
+		else
+			ext4_msg(sb, KERN_WARNING,
+				 "Error processing option \"%s\" [%d]",
+				 opt, err);
+		return -1;
+	}
+	ext4_msg(sb, KERN_WARNING, "Test dummy encryption mode enabled");
+#else
+	ext4_msg(sb, KERN_WARNING,
+		 "Test dummy encryption mount option ignored");
+#endif
+	return 1;
+}
+
 static int handle_mount_opt(struct super_block *sb, char *opt, int token,
 			    substring_t *args, unsigned long *journal_devnum,
 			    unsigned int *journal_ioprio, int is_remount)
@@ -1887,6 +1995,13 @@ static int handle_mount_opt(struct super_block *sb, char *opt, int token,
 		}
 		sbi->s_commit_interval = HZ * arg;
 	} else if (token == Opt_debug_want_extra_isize) {
+		if ((arg & 1) ||
+		    (arg < 4) ||
+		    (arg > (sbi->s_inode_size - EXT4_GOOD_OLD_INODE_SIZE))) {
+			ext4_msg(sb, KERN_ERR,
+				 "Invalid want_extra_isize %d", arg);
+			return -1;
+		}
 		sbi->s_want_extra_isize = arg;
 	} else if (token == Opt_max_batch_time) {
 		sbi->s_max_batch_time = arg;
@@ -1977,14 +2092,8 @@ static int handle_mount_opt(struct super_block *sb, char *opt, int token,
 		*journal_ioprio =
 			IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, arg);
 	} else if (token == Opt_test_dummy_encryption) {
-#ifdef CONFIG_FS_ENCRYPTION
-		sbi->s_mount_flags |= EXT4_MF_TEST_DUMMY_ENCRYPTION;
-		ext4_msg(sb, KERN_WARNING,
-			 "Test dummy encryption mode enabled");
-#else
-		ext4_msg(sb, KERN_WARNING,
-			 "Test dummy encryption mount option ignored");
-#endif
+		return ext4_set_test_dummy_encryption(sb, opt, &args[0],
+						      is_remount);
 	} else if (m->flags & MOPT_DATAJ) {
 		if (is_remount) {
 			if (!sbi->s_journal)
@@ -2051,7 +2160,7 @@ static int parse_options(char *options, struct super_block *sb,
 			 unsigned int *journal_ioprio,
 			 int is_remount)
 {
-	struct ext4_sb_info *sbi = EXT4_SB(sb);
+	struct ext4_sb_info __maybe_unused *sbi = EXT4_SB(sb);
 	char *p, __maybe_unused *usr_qf_name, __maybe_unused *grp_qf_name;
 	substring_t args[MAX_OPT_ARGS];
 	int token;
@@ -2108,12 +2217,10 @@ static int parse_options(char *options, struct super_block *sb,
 	if (test_opt(sb, DIOREAD_NOLOCK)) {
 		int blocksize =
 			BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size);
-
-		if (blocksize < PAGE_SIZE) {
-			ext4_msg(sb, KERN_ERR, "can't mount with "
-				 "dioread_nolock if block size != PAGE_SIZE");
-			return 0;
-		}
+		if (blocksize < PAGE_SIZE)
+			ext4_msg(sb, KERN_WARNING, "Warning: mounting with an "
+				 "experimental mount option 'dioread_nolock' "
+				 "for blocksize < PAGE_SIZE");
 	}
 	return 1;
 }
@@ -2243,8 +2350,8 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
 		SEQ_OPTS_PRINT("max_dir_size_kb=%u", sbi->s_max_dir_size_kb);
 	if (test_opt(sb, DATA_ERR_ABORT))
 		SEQ_OPTS_PUTS("data_err=abort");
-	if (DUMMY_ENCRYPTION_ENABLED(sbi))
-		SEQ_OPTS_PUTS("test_dummy_encryption");
+
+	fscrypt_show_test_dummy_encryption(seq, sep, sb);
 
 	ext4_show_quota_options(seq, sb);
 	return 0;
@@ -2325,8 +2432,8 @@ done:
 int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	struct flex_groups *new_groups;
-	int size;
+	struct flex_groups **old_groups, **new_groups;
+	int size, i, j;
 
 	if (!sbi->s_log_groups_per_flex)
 		return 0;
@@ -2335,22 +2442,37 @@ int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup)
 	if (size <= sbi->s_flex_groups_allocated)
 		return 0;
 
-	size = roundup_pow_of_two(size * sizeof(struct flex_groups));
-	new_groups = kvzalloc(size, GFP_KERNEL);
+	new_groups = kvzalloc(roundup_pow_of_two(size *
+			      sizeof(*sbi->s_flex_groups)), GFP_KERNEL);
 	if (!new_groups) {
-		ext4_msg(sb, KERN_ERR, "not enough memory for %d flex groups",
-			 size / (int) sizeof(struct flex_groups));
+		ext4_msg(sb, KERN_ERR,
+			 "not enough memory for %d flex group pointers", size);
 		return -ENOMEM;
 	}
-
-	if (sbi->s_flex_groups) {
-		memcpy(new_groups, sbi->s_flex_groups,
-		       (sbi->s_flex_groups_allocated *
-			sizeof(struct flex_groups)));
-		kvfree(sbi->s_flex_groups);
+	for (i = sbi->s_flex_groups_allocated; i < size; i++) {
+		new_groups[i] = kvzalloc(roundup_pow_of_two(
+					 sizeof(struct flex_groups)),
+					 GFP_KERNEL);
+		if (!new_groups[i]) {
+			for (j = sbi->s_flex_groups_allocated; j < i; j++)
+				kvfree(new_groups[j]);
+			kvfree(new_groups);
+			ext4_msg(sb, KERN_ERR,
+				 "not enough memory for %d flex groups", size);
+			return -ENOMEM;
+		}
 	}
-	sbi->s_flex_groups = new_groups;
-	sbi->s_flex_groups_allocated = size / sizeof(struct flex_groups);
+	rcu_read_lock();
+	old_groups = rcu_dereference(sbi->s_flex_groups);
+	if (old_groups)
+		memcpy(new_groups, old_groups,
+		       (sbi->s_flex_groups_allocated *
+			sizeof(struct flex_groups *)));
+	rcu_read_unlock();
+	rcu_assign_pointer(sbi->s_flex_groups, new_groups);
+	sbi->s_flex_groups_allocated = size;
+	if (old_groups)
+		ext4_kvfree_array_rcu(old_groups);
 	return 0;
 }
 
@@ -2358,6 +2480,7 @@ static int ext4_fill_flex_info(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct ext4_group_desc *gdp = NULL;
+	struct flex_groups *fg;
 	ext4_group_t flex_group;
 	int i, err;
 
@@ -2375,12 +2498,11 @@ static int ext4_fill_flex_info(struct super_block *sb)
 		gdp = ext4_get_group_desc(sb, i, NULL);
 
 		flex_group = ext4_flex_group(sbi, i);
-		atomic_add(ext4_free_inodes_count(sb, gdp),
-			   &sbi->s_flex_groups[flex_group].free_inodes);
+		fg = sbi_array_rcu_deref(sbi, s_flex_groups, flex_group);
+		atomic_add(ext4_free_inodes_count(sb, gdp), &fg->free_inodes);
 		atomic64_add(ext4_free_group_clusters(sb, gdp),
-			     &sbi->s_flex_groups[flex_group].free_clusters);
-		atomic_add(ext4_used_dirs_count(sb, gdp),
-			   &sbi->s_flex_groups[flex_group].used_dirs);
+			     &fg->free_clusters);
+		atomic_add(ext4_used_dirs_count(sb, gdp), &fg->used_dirs);
 	}
 
 	return 1;
@@ -2954,17 +3076,11 @@ static int ext4_feature_set_ok(struct super_block *sb, int readonly)
 		return 0;
 	}
 
-#ifndef CONFIG_QUOTA
-	if (ext4_has_feature_quota(sb) && !readonly) {
-		ext4_msg(sb, KERN_ERR,
-			 "Filesystem with quota feature cannot be mounted RDWR "
-			 "without CONFIG_QUOTA");
-		return 0;
-	}
-	if (ext4_has_feature_project(sb) && !readonly) {
+#if !IS_ENABLED(CONFIG_QUOTA) || !IS_ENABLED(CONFIG_QFMT_V2)
+	if (!readonly && (ext4_has_feature_quota(sb) ||
+			  ext4_has_feature_project(sb))) {
 		ext4_msg(sb, KERN_ERR,
-			 "Filesystem with project quota feature cannot be mounted RDWR "
-			 "without CONFIG_QUOTA");
+			 "The kernel was not built with CONFIG_QUOTA and CONFIG_QFMT_V2");
 		return 0;
 	}
 #endif  /* CONFIG_QUOTA */
@@ -3535,7 +3651,8 @@ int ext4_calculate_overhead(struct super_block *sb)
 	 */
 	if (sbi->s_journal && !sbi->journal_bdev)
 		overhead += EXT4_NUM_B2C(sbi, sbi->s_journal->j_maxlen);
-	else if (ext4_has_feature_journal(sb) && !sbi->s_journal) {
+	else if (ext4_has_feature_journal(sb) && !sbi->s_journal && j_inum) {
+		/* j_inum for internal journal is non-zero */
 		j_inode = ext4_get_journal_inode(sb, j_inum);
 		if (j_inode) {
 			j_blocks = j_inode->i_size >> sb->s_blocksize_bits;
@@ -3551,37 +3668,6 @@ int ext4_calculate_overhead(struct super_block *sb)
 	return 0;
 }
 
-static void ext4_clamp_want_extra_isize(struct super_block *sb)
-{
-	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	struct ext4_super_block *es = sbi->s_es;
-
-	/* determine the minimum size of new large inodes, if present */
-	if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE &&
-	    sbi->s_want_extra_isize == 0) {
-		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
-						     EXT4_GOOD_OLD_INODE_SIZE;
-		if (ext4_has_feature_extra_isize(sb)) {
-			if (sbi->s_want_extra_isize <
-			    le16_to_cpu(es->s_want_extra_isize))
-				sbi->s_want_extra_isize =
-					le16_to_cpu(es->s_want_extra_isize);
-			if (sbi->s_want_extra_isize <
-			    le16_to_cpu(es->s_min_extra_isize))
-				sbi->s_want_extra_isize =
-					le16_to_cpu(es->s_min_extra_isize);
-		}
-	}
-	/* Check if enough inode space is available */
-	if (EXT4_GOOD_OLD_INODE_SIZE + sbi->s_want_extra_isize >
-							sbi->s_inode_size) {
-		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
-						       EXT4_GOOD_OLD_INODE_SIZE;
-		ext4_msg(sb, KERN_INFO,
-			 "required extra inode space not available");
-	}
-}
-
 static void ext4_set_resv_clusters(struct super_block *sb)
 {
 	ext4_fsblk_t resv_clusters;
@@ -3616,9 +3702,10 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct dax_device *dax_dev = fs_dax_get_by_bdev(sb->s_bdev);
 	char *orig_data = kstrdup(data, GFP_KERNEL);
-	struct buffer_head *bh;
+	struct buffer_head *bh, **group_desc;
 	struct ext4_super_block *es = NULL;
 	struct ext4_sb_info *sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
+	struct flex_groups **flex_groups;
 	ext4_fsblk_t block;
 	ext4_fsblk_t sb_block = get_sb_block(&data);
 	ext4_fsblk_t logical_sb_block;
@@ -3789,6 +3876,82 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	 */
 	sbi->s_li_wait_mult = EXT4_DEF_LI_WAIT_MULT;
 
+	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
+
+	if (blocksize == PAGE_SIZE)
+		set_opt(sb, DIOREAD_NOLOCK);
+
+	if (blocksize < EXT4_MIN_BLOCK_SIZE ||
+	    blocksize > EXT4_MAX_BLOCK_SIZE) {
+		ext4_msg(sb, KERN_ERR,
+		       "Unsupported filesystem blocksize %d (%d log_block_size)",
+			 blocksize, le32_to_cpu(es->s_log_block_size));
+		goto failed_mount;
+	}
+
+	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) {
+		sbi->s_inode_size = EXT4_GOOD_OLD_INODE_SIZE;
+		sbi->s_first_ino = EXT4_GOOD_OLD_FIRST_INO;
+	} else {
+		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
+		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
+		if (sbi->s_first_ino < EXT4_GOOD_OLD_FIRST_INO) {
+			ext4_msg(sb, KERN_ERR, "invalid first ino: %u",
+				 sbi->s_first_ino);
+			goto failed_mount;
+		}
+		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
+		    (!is_power_of_2(sbi->s_inode_size)) ||
+		    (sbi->s_inode_size > blocksize)) {
+			ext4_msg(sb, KERN_ERR,
+			       "unsupported inode size: %d",
+			       sbi->s_inode_size);
+			ext4_msg(sb, KERN_ERR, "blocksize: %d", blocksize);
+			goto failed_mount;
+		}
+		/*
+		 * i_atime_extra is the last extra field available for
+		 * [acm]times in struct ext4_inode. Checking for that
+		 * field should suffice to ensure we have extra space
+		 * for all three.
+		 */
+		if (sbi->s_inode_size >= offsetof(struct ext4_inode, i_atime_extra) +
+			sizeof(((struct ext4_inode *)0)->i_atime_extra)) {
+			sb->s_time_gran = 1;
+			sb->s_time_max = EXT4_EXTRA_TIMESTAMP_MAX;
+		} else {
+			sb->s_time_gran = NSEC_PER_SEC;
+			sb->s_time_max = EXT4_NON_EXTRA_TIMESTAMP_MAX;
+		}
+		sb->s_time_min = EXT4_TIMESTAMP_MIN;
+	}
+	if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE) {
+		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
+			EXT4_GOOD_OLD_INODE_SIZE;
+		if (ext4_has_feature_extra_isize(sb)) {
+			unsigned v, max = (sbi->s_inode_size -
+					   EXT4_GOOD_OLD_INODE_SIZE);
+
+			v = le16_to_cpu(es->s_want_extra_isize);
+			if (v > max) {
+				ext4_msg(sb, KERN_ERR,
+					 "bad s_want_extra_isize: %d", v);
+				goto failed_mount;
+			}
+			if (sbi->s_want_extra_isize < v)
+				sbi->s_want_extra_isize = v;
+
+			v = le16_to_cpu(es->s_min_extra_isize);
+			if (v > max) {
+				ext4_msg(sb, KERN_ERR,
+					 "bad s_min_extra_isize: %d", v);
+				goto failed_mount;
+			}
+			if (sbi->s_want_extra_isize < v)
+				sbi->s_want_extra_isize = v;
+		}
+	}
+
 	if (sbi->s_es->s_mount_opts[0]) {
 		char *s_mount_opts = kstrndup(sbi->s_es->s_mount_opts,
 					      sizeof(sbi->s_es->s_mount_opts),
@@ -3846,9 +4009,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 #endif
 
 	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) {
-		printk_once(KERN_WARNING "EXT4-fs: Warning: mounting "
-			    "with data=journal disables delayed "
-			    "allocation and O_DIRECT support!\n");
+		printk_once(KERN_WARNING "EXT4-fs: Warning: mounting with data=journal disables delayed allocation, dioread_nolock, and O_DIRECT support!\n");
+		clear_opt(sb, DIOREAD_NOLOCK);
 		if (test_opt2(sb, EXPLICIT_DELALLOC)) {
 			ext4_msg(sb, KERN_ERR, "can't mount with "
 				 "both data=journal and delalloc");
@@ -3947,14 +4109,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	if (!ext4_feature_set_ok(sb, (sb_rdonly(sb))))
 		goto failed_mount;
 
-	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
-	if (blocksize < EXT4_MIN_BLOCK_SIZE ||
-	    blocksize > EXT4_MAX_BLOCK_SIZE) {
-		ext4_msg(sb, KERN_ERR,
-		       "Unsupported filesystem blocksize %d (%d log_block_size)",
-			 blocksize, le32_to_cpu(es->s_log_block_size));
-		goto failed_mount;
-	}
 	if (le32_to_cpu(es->s_log_block_size) >
 	    (EXT4_MAX_BLOCK_LOG_SIZE - EXT4_MIN_BLOCK_LOG_SIZE)) {
 		ext4_msg(sb, KERN_ERR,
@@ -4027,42 +4181,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 						      has_huge_files);
 	sb->s_maxbytes = ext4_max_size(sb->s_blocksize_bits, has_huge_files);
 
-	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) {
-		sbi->s_inode_size = EXT4_GOOD_OLD_INODE_SIZE;
-		sbi->s_first_ino = EXT4_GOOD_OLD_FIRST_INO;
-	} else {
-		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
-		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
-		if (sbi->s_first_ino < EXT4_GOOD_OLD_FIRST_INO) {
-			ext4_msg(sb, KERN_ERR, "invalid first ino: %u",
-				 sbi->s_first_ino);
-			goto failed_mount;
-		}
-		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
-		    (!is_power_of_2(sbi->s_inode_size)) ||
-		    (sbi->s_inode_size > blocksize)) {
-			ext4_msg(sb, KERN_ERR,
-			       "unsupported inode size: %d",
-			       sbi->s_inode_size);
-			goto failed_mount;
-		}
-		/*
-		 * i_atime_extra is the last extra field available for [acm]times in
-		 * struct ext4_inode. Checking for that field should suffice to ensure
-		 * we have extra space for all three.
-		 */
-		if (sbi->s_inode_size >= offsetof(struct ext4_inode, i_atime_extra) +
-			sizeof(((struct ext4_inode *)0)->i_atime_extra)) {
-			sb->s_time_gran = 1;
-			sb->s_time_max = EXT4_EXTRA_TIMESTAMP_MAX;
-		} else {
-			sb->s_time_gran = NSEC_PER_SEC;
-			sb->s_time_max = EXT4_NON_EXTRA_TIMESTAMP_MAX;
-		}
-
-		sb->s_time_min = EXT4_TIMESTAMP_MIN;
-	}
-
 	sbi->s_desc_size = le16_to_cpu(es->s_desc_size);
 	if (ext4_has_feature_64bit(sb)) {
 		if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT ||
@@ -4085,7 +4203,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	if (sbi->s_inodes_per_group < sbi->s_inodes_per_block ||
 	    sbi->s_inodes_per_group > blocksize * 8) {
 		ext4_msg(sb, KERN_ERR, "invalid inodes per group: %lu\n",
-			 sbi->s_blocks_per_group);
+			 sbi->s_inodes_per_group);
 		goto failed_mount;
 	}
 	sbi->s_itb_per_group = sbi->s_inodes_per_group /
@@ -4214,9 +4332,9 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 			EXT4_BLOCKS_PER_GROUP(sb) - 1);
 	do_div(blocks_count, EXT4_BLOCKS_PER_GROUP(sb));
 	if (blocks_count > ((uint64_t)1<<32) - EXT4_DESC_PER_BLOCK(sb)) {
-		ext4_msg(sb, KERN_WARNING, "groups count too large: %u "
+		ext4_msg(sb, KERN_WARNING, "groups count too large: %llu "
 		       "(block count %llu, first data block %u, "
-		       "blocks per group %lu)", sbi->s_groups_count,
+		       "blocks per group %lu)", blocks_count,
 		       ext4_blocks_count(es),
 		       le32_to_cpu(es->s_first_data_block),
 		       EXT4_BLOCKS_PER_GROUP(sb));
@@ -4244,9 +4362,10 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 			goto failed_mount;
 		}
 	}
-	sbi->s_group_desc = kvmalloc_array(db_count,
-					   sizeof(struct buffer_head *),
-					   GFP_KERNEL);
+	rcu_assign_pointer(sbi->s_group_desc,
+			   kvmalloc_array(db_count,
+					  sizeof(struct buffer_head *),
+					  GFP_KERNEL));
 	if (sbi->s_group_desc == NULL) {
 		ext4_msg(sb, KERN_ERR, "not enough memory");
 		ret = -ENOMEM;
@@ -4258,18 +4377,23 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	/* Pre-read the descriptors into the buffer cache */
 	for (i = 0; i < db_count; i++) {
 		block = descriptor_loc(sb, logical_sb_block, i);
-		sb_breadahead(sb, block);
+		sb_breadahead_unmovable(sb, block);
 	}
 
 	for (i = 0; i < db_count; i++) {
+		struct buffer_head *bh;
+
 		block = descriptor_loc(sb, logical_sb_block, i);
-		sbi->s_group_desc[i] = sb_bread_unmovable(sb, block);
-		if (!sbi->s_group_desc[i]) {
+		bh = sb_bread_unmovable(sb, block);
+		if (!bh) {
 			ext4_msg(sb, KERN_ERR,
 			       "can't read group descriptor %d", i);
 			db_count = i;
 			goto failed_mount2;
 		}
+		rcu_read_lock();
+		rcu_dereference(sbi->s_group_desc)[i] = bh;
+		rcu_read_unlock();
 	}
 	sbi->s_gdb_count = db_count;
 	if (!ext4_check_descriptors(sb, logical_sb_block, &first_not_zeroed)) {
@@ -4439,13 +4563,6 @@ no_journal:
 		}
 	}
 
-	if ((DUMMY_ENCRYPTION_ENABLED(sbi) || ext4_has_feature_encrypt(sb)) &&
-	    (blocksize != PAGE_SIZE)) {
-		ext4_msg(sb, KERN_ERR,
-			 "Unsupported blocksize for fs encryption");
-		goto failed_mount_wq;
-	}
-
 	if (ext4_has_feature_verity(sb) && blocksize != PAGE_SIZE) {
 		ext4_msg(sb, KERN_ERR, "Unsupported blocksize for fs-verity");
 		goto failed_mount_wq;
@@ -4518,8 +4635,6 @@ no_journal:
 	} else if (ret)
 		goto failed_mount4a;
 
-	ext4_clamp_want_extra_isize(sb);
-
 	ext4_set_resv_clusters(sb);
 
 	err = ext4_setup_system_zone(sb);
@@ -4557,7 +4672,7 @@ no_journal:
 		err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0,
 					  GFP_KERNEL);
 	if (!err)
-		err = percpu_init_rwsem(&sbi->s_journal_flag_rwsem);
+		err = percpu_init_rwsem(&sbi->s_writepages_rwsem);
 
 	if (err) {
 		ext4_msg(sb, KERN_ERR, "insufficient memory");
@@ -4645,13 +4760,19 @@ failed_mount7:
 	ext4_unregister_li_request(sb);
 failed_mount6:
 	ext4_mb_release(sb);
-	if (sbi->s_flex_groups)
-		kvfree(sbi->s_flex_groups);
+	rcu_read_lock();
+	flex_groups = rcu_dereference(sbi->s_flex_groups);
+	if (flex_groups) {
+		for (i = 0; i < sbi->s_flex_groups_allocated; i++)
+			kvfree(flex_groups[i]);
+		kvfree(flex_groups);
+	}
+	rcu_read_unlock();
 	percpu_counter_destroy(&sbi->s_freeclusters_counter);
 	percpu_counter_destroy(&sbi->s_freeinodes_counter);
 	percpu_counter_destroy(&sbi->s_dirs_counter);
 	percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
-	percpu_free_rwsem(&sbi->s_journal_flag_rwsem);
+	percpu_free_rwsem(&sbi->s_writepages_rwsem);
 failed_mount5:
 	ext4_ext_release(sb);
 	ext4_release_system_zone(sb);
@@ -4680,9 +4801,12 @@ failed_mount3:
 	if (sbi->s_mmp_tsk)
 		kthread_stop(sbi->s_mmp_tsk);
 failed_mount2:
+	rcu_read_lock();
+	group_desc = rcu_dereference(sbi->s_group_desc);
 	for (i = 0; i < db_count; i++)
-		brelse(sbi->s_group_desc[i]);
-	kvfree(sbi->s_group_desc);
+		brelse(group_desc[i]);
+	kvfree(group_desc);
+	rcu_read_unlock();
 failed_mount:
 	if (sbi->s_chksum_driver)
 		crypto_free_shash(sbi->s_chksum_driver);
@@ -4695,6 +4819,7 @@ failed_mount:
 	for (i = 0; i < EXT4_MAXQUOTAS; i++)
 		kfree(get_qf_name(sb, sbi, i));
 #endif
+	fscrypt_free_dummy_context(&sbi->s_dummy_enc_ctx);
 	ext4_blkdev_remove(sbi);
 	brelse(bh);
 out_fail:
@@ -5171,7 +5296,7 @@ static int ext4_sync_fs(struct super_block *sb, int wait)
 		needs_barrier = true;
 	if (needs_barrier) {
 		int err;
-		err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);
+		err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL);
 		if (!ret)
 			ret = err;
 	}
@@ -5307,8 +5432,6 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 		goto restore_opts;
 	}
 
-	ext4_clamp_want_extra_isize(sb);
-
 	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
 	    test_opt(sb, JOURNAL_CHECKSUM)) {
 		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
@@ -5357,7 +5480,7 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 	}
 
 	if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED)
-		ext4_abort(sb, "Abort forced by user");
+		ext4_abort(sb, EXT4_ERR_ESHUTDOWN, "Abort forced by user");
 
 	sb->s_flags = (sb->s_flags & ~SB_POSIXACL) |
 		(test_opt(sb, POSIX_ACL) ? SB_POSIXACL : 0);
@@ -5546,9 +5669,10 @@ static int ext4_statfs_project(struct super_block *sb,
 		return PTR_ERR(dquot);
 	spin_lock(&dquot->dq_dqb_lock);
 
-	limit = (dquot->dq_dqb.dqb_bsoftlimit ?
-		 dquot->dq_dqb.dqb_bsoftlimit :
-		 dquot->dq_dqb.dqb_bhardlimit) >> sb->s_blocksize_bits;
+	limit = min_not_zero(dquot->dq_dqb.dqb_bsoftlimit,
+			     dquot->dq_dqb.dqb_bhardlimit);
+	limit >>= sb->s_blocksize_bits;
+
 	if (limit && buf->f_blocks > limit) {
 		curblock = (dquot->dq_dqb.dqb_curspace +
 			    dquot->dq_dqb.dqb_rsvspace) >> sb->s_blocksize_bits;
@@ -5558,9 +5682,8 @@ static int ext4_statfs_project(struct super_block *sb,
 			 (buf->f_blocks - curblock) : 0;
 	}
 
-	limit = dquot->dq_dqb.dqb_isoftlimit ?
-		dquot->dq_dqb.dqb_isoftlimit :
-		dquot->dq_dqb.dqb_ihardlimit;
+	limit = min_not_zero(dquot->dq_dqb.dqb_isoftlimit,
+			     dquot->dq_dqb.dqb_ihardlimit);
 	if (limit && buf->f_files > limit) {
 		buf->f_files = limit;
 		buf->f_ffree =
@@ -5835,7 +5958,7 @@ static int ext4_quota_enable(struct super_block *sb, int type, int format_id,
 	/* Don't account quota for quota files to avoid recursion */
 	qf_inode->i_flags |= S_NOQUOTA;
 	lockdep_set_quota_inode(qf_inode, I_DATA_SEM_QUOTA);
-	err = dquot_enable(qf_inode, type, format_id, flags);
+	err = dquot_load_quota_inode(qf_inode, type, format_id, flags);
 	if (err)
 		lockdep_set_quota_inode(qf_inode, I_DATA_SEM_NORMAL);
 	iput(qf_inode);
@@ -5993,7 +6116,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type,
 		bh = ext4_bread(handle, inode, blk,
 				EXT4_GET_BLOCKS_CREATE |
 				EXT4_GET_BLOCKS_METADATA_NOFAIL);
-	} while (IS_ERR(bh) && (PTR_ERR(bh) == -ENOSPC) &&
+	} while (PTR_ERR(bh) == -ENOSPC &&
 		 ext4_should_retry_alloc(inode->i_sb, &retries));
 	if (IS_ERR(bh))
 		return PTR_ERR(bh);
@@ -6019,18 +6142,6 @@ out:
 	}
 	return len;
 }
-
-static int ext4_get_next_id(struct super_block *sb, struct kqid *qid)
-{
-	const struct quota_format_ops	*ops;
-
-	if (!sb_has_quota_loaded(sb, qid->type))
-		return -ESRCH;
-	ops = sb_dqopt(sb)->ops[qid->type];
-	if (!ops || !ops->get_next_id)
-		return -ENOSYS;
-	return dquot_get_next_id(sb, qid);
-}
 #endif
 
 static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c
index eb1efad0e20a..6c9fc9e21c13 100644
--- a/fs/ext4/sysfs.c
+++ b/fs/ext4/sysfs.c
@@ -13,6 +13,7 @@
 #include <linux/seq_file.h>
 #include <linux/slab.h>
 #include <linux/proc_fs.h>
+#include <linux/part_stat.h>
 
 #include "ext4.h"
 #include "ext4_jbd2.h"
@@ -29,6 +30,10 @@ typedef enum {
 	attr_last_error_time,
 	attr_feature,
 	attr_pointer_ui,
+	attr_pointer_ul,
+	attr_pointer_u64,
+	attr_pointer_u8,
+	attr_pointer_string,
 	attr_pointer_atomic,
 	attr_journal_task,
 } attr_id_t;
@@ -46,6 +51,7 @@ struct ext4_attr {
 	struct attribute attr;
 	short attr_id;
 	short attr_ptr;
+	unsigned short attr_size;
 	union {
 		int offset;
 		void *explicit_ptr;
@@ -154,12 +160,35 @@ static struct ext4_attr ext4_attr_##_name = {			\
 	},							\
 }
 
+#define EXT4_ATTR_STRING(_name,_mode,_size,_struct,_elname)	\
+static struct ext4_attr ext4_attr_##_name = {			\
+	.attr = {.name = __stringify(_name), .mode = _mode },	\
+	.attr_id = attr_pointer_string,				\
+	.attr_size = _size,					\
+	.attr_ptr = ptr_##_struct##_offset,			\
+	.u = {							\
+		.offset = offsetof(struct _struct, _elname),\
+	},							\
+}
+
 #define EXT4_RO_ATTR_ES_UI(_name,_elname)				\
 	EXT4_ATTR_OFFSET(_name, 0444, pointer_ui, ext4_super_block, _elname)
 
+#define EXT4_RO_ATTR_ES_U8(_name,_elname)				\
+	EXT4_ATTR_OFFSET(_name, 0444, pointer_u8, ext4_super_block, _elname)
+
+#define EXT4_RO_ATTR_ES_U64(_name,_elname)				\
+	EXT4_ATTR_OFFSET(_name, 0444, pointer_u64, ext4_super_block, _elname)
+
+#define EXT4_RO_ATTR_ES_STRING(_name,_elname,_size)			\
+	EXT4_ATTR_STRING(_name, 0444, _size, ext4_super_block, _elname)
+
 #define EXT4_RW_ATTR_SBI_UI(_name,_elname)	\
 	EXT4_ATTR_OFFSET(_name, 0644, pointer_ui, ext4_sb_info, _elname)
 
+#define EXT4_RW_ATTR_SBI_UL(_name,_elname)	\
+	EXT4_ATTR_OFFSET(_name, 0644, pointer_ul, ext4_sb_info, _elname)
+
 #define EXT4_ATTR_PTR(_name,_mode,_id,_ptr) \
 static struct ext4_attr ext4_attr_##_name = {			\
 	.attr = {.name = __stringify(_name), .mode = _mode },	\
@@ -194,7 +223,20 @@ EXT4_RW_ATTR_SBI_UI(warning_ratelimit_interval_ms, s_warning_ratelimit_state.int
 EXT4_RW_ATTR_SBI_UI(warning_ratelimit_burst, s_warning_ratelimit_state.burst);
 EXT4_RW_ATTR_SBI_UI(msg_ratelimit_interval_ms, s_msg_ratelimit_state.interval);
 EXT4_RW_ATTR_SBI_UI(msg_ratelimit_burst, s_msg_ratelimit_state.burst);
+#ifdef CONFIG_EXT4_DEBUG
+EXT4_RW_ATTR_SBI_UL(simulate_fail, s_simulate_fail);
+#endif
 EXT4_RO_ATTR_ES_UI(errors_count, s_error_count);
+EXT4_RO_ATTR_ES_U8(first_error_errcode, s_first_error_errcode);
+EXT4_RO_ATTR_ES_U8(last_error_errcode, s_last_error_errcode);
+EXT4_RO_ATTR_ES_UI(first_error_ino, s_first_error_ino);
+EXT4_RO_ATTR_ES_UI(last_error_ino, s_last_error_ino);
+EXT4_RO_ATTR_ES_U64(first_error_block, s_first_error_block);
+EXT4_RO_ATTR_ES_U64(last_error_block, s_last_error_block);
+EXT4_RO_ATTR_ES_UI(first_error_line, s_first_error_line);
+EXT4_RO_ATTR_ES_UI(last_error_line, s_last_error_line);
+EXT4_RO_ATTR_ES_STRING(first_error_func, s_first_error_func, 32);
+EXT4_RO_ATTR_ES_STRING(last_error_func, s_last_error_func, 32);
 EXT4_ATTR(first_error_time, 0444, first_error_time);
 EXT4_ATTR(last_error_time, 0444, last_error_time);
 EXT4_ATTR(journal_task, 0444, journal_task);
@@ -225,9 +267,22 @@ static struct attribute *ext4_attrs[] = {
 	ATTR_LIST(msg_ratelimit_interval_ms),
 	ATTR_LIST(msg_ratelimit_burst),
 	ATTR_LIST(errors_count),
+	ATTR_LIST(first_error_ino),
+	ATTR_LIST(last_error_ino),
+	ATTR_LIST(first_error_block),
+	ATTR_LIST(last_error_block),
+	ATTR_LIST(first_error_line),
+	ATTR_LIST(last_error_line),
+	ATTR_LIST(first_error_func),
+	ATTR_LIST(last_error_func),
+	ATTR_LIST(first_error_errcode),
+	ATTR_LIST(last_error_errcode),
 	ATTR_LIST(first_error_time),
 	ATTR_LIST(last_error_time),
 	ATTR_LIST(journal_task),
+#ifdef CONFIG_EXT4_DEBUG
+	ATTR_LIST(simulate_fail),
+#endif
 	NULL,
 };
 ATTRIBUTE_GROUPS(ext4);
@@ -238,6 +293,7 @@ EXT4_ATTR_FEATURE(batched_discard);
 EXT4_ATTR_FEATURE(meta_bg_resize);
 #ifdef CONFIG_FS_ENCRYPTION
 EXT4_ATTR_FEATURE(encryption);
+EXT4_ATTR_FEATURE(test_dummy_encryption_v2);
 #endif
 #ifdef CONFIG_UNICODE
 EXT4_ATTR_FEATURE(casefold);
@@ -253,6 +309,7 @@ static struct attribute *ext4_feat_attrs[] = {
 	ATTR_LIST(meta_bg_resize),
 #ifdef CONFIG_FS_ENCRYPTION
 	ATTR_LIST(encryption),
+	ATTR_LIST(test_dummy_encryption_v2),
 #endif
 #ifdef CONFIG_UNICODE
 	ATTR_LIST(casefold),
@@ -280,7 +337,7 @@ static void *calc_ptr(struct ext4_attr *a, struct ext4_sb_info *sbi)
 
 static ssize_t __print_tstamp(char *buf, __le32 lo, __u8 hi)
 {
-	return snprintf(buf, PAGE_SIZE, "%lld",
+	return snprintf(buf, PAGE_SIZE, "%lld\n",
 			((time64_t)hi << 32) + le32_to_cpu(lo));
 }
 
@@ -318,6 +375,30 @@ static ssize_t ext4_attr_show(struct kobject *kobj,
 		else
 			return snprintf(buf, PAGE_SIZE, "%u\n",
 					*((unsigned int *) ptr));
+	case attr_pointer_ul:
+		if (!ptr)
+			return 0;
+		return snprintf(buf, PAGE_SIZE, "%lu\n",
+				*((unsigned long *) ptr));
+	case attr_pointer_u8:
+		if (!ptr)
+			return 0;
+		return snprintf(buf, PAGE_SIZE, "%u\n",
+				*((unsigned char *) ptr));
+	case attr_pointer_u64:
+		if (!ptr)
+			return 0;
+		if (a->attr_ptr == ptr_ext4_super_block_offset)
+			return snprintf(buf, PAGE_SIZE, "%llu\n",
+					le64_to_cpup(ptr));
+		else
+			return snprintf(buf, PAGE_SIZE, "%llu\n",
+					*((unsigned long long *) ptr));
+	case attr_pointer_string:
+		if (!ptr)
+			return 0;
+		return snprintf(buf, PAGE_SIZE, "%.*s\n", a->attr_size,
+				(char *) ptr);
 	case attr_pointer_atomic:
 		if (!ptr)
 			return 0;
@@ -361,6 +442,14 @@ static ssize_t ext4_attr_store(struct kobject *kobj,
 		else
 			*((unsigned int *) ptr) = t;
 		return len;
+	case attr_pointer_ul:
+		if (!ptr)
+			return 0;
+		ret = kstrtoul(skip_spaces(buf), 0, &t);
+		if (ret)
+			return ret;
+		*((unsigned long *) ptr) = t;
+		return len;
 	case attr_inode_readahead:
 		return inode_readahead_blks_store(sbi, buf, len);
 	case attr_trigger_test_error:
diff --git a/fs/ext4/verity.c b/fs/ext4/verity.c
index d0d8a9795dd6..dec1244dd062 100644
--- a/fs/ext4/verity.c
+++ b/fs/ext4/verity.c
@@ -343,11 +343,23 @@ static int ext4_get_verity_descriptor(struct inode *inode, void *buf,
 }
 
 static struct page *ext4_read_merkle_tree_page(struct inode *inode,
-					       pgoff_t index)
+					       pgoff_t index,
+					       unsigned long num_ra_pages)
 {
+	struct page *page;
+
 	index += ext4_verity_metadata_pos(inode) >> PAGE_SHIFT;
 
-	return read_mapping_page(inode->i_mapping, index, NULL);
+	page = find_get_page_flags(inode->i_mapping, index, FGP_ACCESSED);
+	if (!page || !PageUptodate(page)) {
+		if (page)
+			put_page(page);
+		else if (num_ra_pages > 1)
+			page_cache_readahead_unbounded(inode->i_mapping, NULL,
+					index, num_ra_pages, 0);
+		page = read_mapping_page(inode->i_mapping, index, NULL);
+	}
+	return page;
 }
 
 static int ext4_write_merkle_tree_block(struct inode *inode, const void *buf,
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index 491f9ee4040e..21df43a25328 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -245,7 +245,7 @@ __ext4_xattr_check_block(struct inode *inode, struct buffer_head *bh,
 					 bh->b_data);
 errout:
 	if (error)
-		__ext4_error_inode(inode, function, line, 0,
+		__ext4_error_inode(inode, function, line, 0, -error,
 				   "corrupted xattr block %llu",
 				   (unsigned long long) bh->b_blocknr);
 	else
@@ -269,7 +269,7 @@ __xattr_check_inode(struct inode *inode, struct ext4_xattr_ibody_header *header,
 	error = ext4_xattr_check_entries(IFIRST(header), end, IFIRST(header));
 errout:
 	if (error)
-		__ext4_error_inode(inode, function, line, 0,
+		__ext4_error_inode(inode, function, line, 0, -error,
 				   "corrupted in-inode xattr");
 	return error;
 }
@@ -967,55 +967,6 @@ int __ext4_xattr_set_credits(struct super_block *sb, struct inode *inode,
 	return credits;
 }
 
-static int ext4_xattr_ensure_credits(handle_t *handle, struct inode *inode,
-				     int credits, struct buffer_head *bh,
-				     bool dirty, bool block_csum)
-{
-	int error;
-
-	if (!ext4_handle_valid(handle))
-		return 0;
-
-	if (handle->h_buffer_credits >= credits)
-		return 0;
-
-	error = ext4_journal_extend(handle, credits - handle->h_buffer_credits);
-	if (!error)
-		return 0;
-	if (error < 0) {
-		ext4_warning(inode->i_sb, "Extend journal (error %d)", error);
-		return error;
-	}
-
-	if (bh && dirty) {
-		if (block_csum)
-			ext4_xattr_block_csum_set(inode, bh);
-		error = ext4_handle_dirty_metadata(handle, NULL, bh);
-		if (error) {
-			ext4_warning(inode->i_sb, "Handle metadata (error %d)",
-				     error);
-			return error;
-		}
-	}
-
-	error = ext4_journal_restart(handle, credits);
-	if (error) {
-		ext4_warning(inode->i_sb, "Restart journal (error %d)", error);
-		return error;
-	}
-
-	if (bh) {
-		error = ext4_journal_get_write_access(handle, bh);
-		if (error) {
-			ext4_warning(inode->i_sb,
-				     "Get write access failed (error %d)",
-				     error);
-			return error;
-		}
-	}
-	return 0;
-}
-
 static int ext4_xattr_inode_update_ref(handle_t *handle, struct inode *ea_inode,
 				       int ref_change)
 {
@@ -1149,6 +1100,24 @@ cleanup:
 	return saved_err;
 }
 
+static int ext4_xattr_restart_fn(handle_t *handle, struct inode *inode,
+			struct buffer_head *bh, bool block_csum, bool dirty)
+{
+	int error;
+
+	if (bh && dirty) {
+		if (block_csum)
+			ext4_xattr_block_csum_set(inode, bh);
+		error = ext4_handle_dirty_metadata(handle, NULL, bh);
+		if (error) {
+			ext4_warning(inode->i_sb, "Handle metadata (error %d)",
+				     error);
+			return error;
+		}
+	}
+	return 0;
+}
+
 static void
 ext4_xattr_inode_dec_ref_all(handle_t *handle, struct inode *parent,
 			     struct buffer_head *bh,
@@ -1185,13 +1154,24 @@ ext4_xattr_inode_dec_ref_all(handle_t *handle, struct inode *parent,
 			continue;
 		}
 
-		err = ext4_xattr_ensure_credits(handle, parent, credits, bh,
-						dirty, block_csum);
-		if (err) {
+		err = ext4_journal_ensure_credits_fn(handle, credits, credits,
+			ext4_free_metadata_revoke_credits(parent->i_sb, 1),
+			ext4_xattr_restart_fn(handle, parent, bh, block_csum,
+					      dirty));
+		if (err < 0) {
 			ext4_warning_inode(ea_inode, "Ensure credits err=%d",
 					   err);
 			continue;
 		}
+		if (err > 0) {
+			err = ext4_journal_get_write_access(handle, bh);
+			if (err) {
+				ext4_warning_inode(ea_inode,
+						"Re-get write access err=%d",
+						err);
+				continue;
+			}
+		}
 
 		err = ext4_xattr_inode_dec_ref(handle, ea_inode);
 		if (err) {
@@ -1476,7 +1456,7 @@ ext4_xattr_inode_cache_find(struct inode *inode, const void *value,
 	if (!ce)
 		return NULL;
 
-	ea_data = ext4_kvmalloc(value_len, GFP_NOFS);
+	ea_data = kvmalloc(value_len, GFP_KERNEL);
 	if (!ea_data) {
 		mb_cache_entry_put(ea_inode_cache, ce);
 		return NULL;
@@ -2335,7 +2315,7 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
 						   flags & XATTR_CREATE);
 		brelse(bh);
 
-		if (!ext4_handle_has_enough_credits(handle, credits)) {
+		if (jbd2_handle_buffer_credits(handle) < credits) {
 			error = -ENOSPC;
 			goto cleanup;
 		}
@@ -2862,11 +2842,9 @@ int ext4_xattr_delete_inode(handle_t *handle, struct inode *inode,
 	struct inode *ea_inode;
 	int error;
 
-	error = ext4_xattr_ensure_credits(handle, inode, extra_credits,
-					  NULL /* bh */,
-					  false /* dirty */,
-					  false /* block_csum */);
-	if (error) {
+	error = ext4_journal_ensure_credits(handle, extra_credits,
+			ext4_free_metadata_revoke_credits(inode->i_sb, 1));
+	if (error < 0) {
 		EXT4_ERROR_INODE(inode, "ensure credits (error %d)", error);
 		goto cleanup;
 	}
@@ -2901,9 +2879,11 @@ int ext4_xattr_delete_inode(handle_t *handle, struct inode *inode,
 		bh = ext4_sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl, REQ_PRIO);
 		if (IS_ERR(bh)) {
 			error = PTR_ERR(bh);
-			if (error == -EIO)
-				EXT4_ERROR_INODE(inode, "block %llu read error",
-						 EXT4_I(inode)->i_file_acl);
+			if (error == -EIO) {
+				EXT4_ERROR_INODE_ERR(inode, EIO,
+						     "block %llu read error",
+						     EXT4_I(inode)->i_file_acl);
+			}
 			bh = NULL;
 			goto cleanup;
 		}
diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h
index f39cad2abe2a..ffe21ac77f78 100644
--- a/fs/ext4/xattr.h
+++ b/fs/ext4/xattr.h
@@ -48,7 +48,7 @@ struct ext4_xattr_entry {
 	__le32	e_value_inum;	/* inode in which the value is stored */
 	__le32	e_value_size;	/* size of attribute value */
 	__le32	e_hash;		/* hash value of name and value */
-	char	e_name[0];	/* attribute name */
+	char	e_name[];	/* attribute name */
 };
 
 #define EXT4_XATTR_PAD_BITS		2
@@ -118,7 +118,7 @@ struct ext4_xattr_ibody_find {
 
 struct ext4_xattr_inode_array {
 	unsigned int count;		/* # of used items in the array */
-	struct inode *inodes[0];
+	struct inode *inodes[];
 };
 
 extern const struct xattr_handler ext4_xattr_user_handler;
diff --git a/fs/f2fs/Kconfig b/fs/f2fs/Kconfig
index 652fd2e2b23d..bb68d21e1f8c 100644
--- a/fs/f2fs/Kconfig
+++ b/fs/f2fs/Kconfig
@@ -6,6 +6,7 @@ config F2FS_FS
 	select CRYPTO
 	select CRYPTO_CRC32
 	select F2FS_FS_XATTR if FS_ENCRYPTION
+	select FS_ENCRYPTION_ALGS if FS_ENCRYPTION
 	help
 	  F2FS is based on Log-structured File System (LFS), which supports
 	  versatile "flash-friendly" features. The design has been focused on
@@ -21,7 +22,7 @@ config F2FS_FS
 
 config F2FS_STAT_FS
 	bool "F2FS Status Information"
-	depends on F2FS_FS && DEBUG_FS
+	depends on F2FS_FS
 	default y
 	help
 	  /sys/kernel/debug/f2fs/ contains information about all the partitions
@@ -92,3 +93,37 @@ config F2FS_FAULT_INJECTION
 	  Test F2FS to inject faults such as ENOMEM, ENOSPC, and so on.
 
 	  If unsure, say N.
+
+config F2FS_FS_COMPRESSION
+	bool "F2FS compression feature"
+	depends on F2FS_FS
+	help
+	  Enable filesystem-level compression on f2fs regular files,
+	  multiple back-end compression algorithms are supported.
+
+config F2FS_FS_LZO
+	bool "LZO compression support"
+	depends on F2FS_FS_COMPRESSION
+	select LZO_COMPRESS
+	select LZO_DECOMPRESS
+	default y
+	help
+	  Support LZO compress algorithm, if unsure, say Y.
+
+config F2FS_FS_LZ4
+	bool "LZ4 compression support"
+	depends on F2FS_FS_COMPRESSION
+	select LZ4_COMPRESS
+	select LZ4_DECOMPRESS
+	default y
+	help
+	  Support LZ4 compress algorithm, if unsure, say Y.
+
+config F2FS_FS_ZSTD
+	bool "ZSTD compression support"
+	depends on F2FS_FS_COMPRESSION
+	select ZSTD_COMPRESS
+	select ZSTD_DECOMPRESS
+	default y
+	help
+	  Support ZSTD compress algorithm, if unsure, say Y.
diff --git a/fs/f2fs/Makefile b/fs/f2fs/Makefile
index 2aaecc63834f..ee7316b42f69 100644
--- a/fs/f2fs/Makefile
+++ b/fs/f2fs/Makefile
@@ -9,3 +9,4 @@ f2fs-$(CONFIG_F2FS_FS_XATTR) += xattr.o
 f2fs-$(CONFIG_F2FS_FS_POSIX_ACL) += acl.o
 f2fs-$(CONFIG_F2FS_IO_TRACE) += trace.o
 f2fs-$(CONFIG_FS_VERITY) += verity.o
+f2fs-$(CONFIG_F2FS_FS_COMPRESSION) += compress.o
diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index a0eef95b9e0e..852890b72d6a 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -50,9 +50,6 @@ repeat:
 	return page;
 }
 
-/*
- * We guarantee no failure on the returned page.
- */
 static struct page *__get_meta_page(struct f2fs_sb_info *sbi, pgoff_t index,
 							bool is_meta)
 {
@@ -206,7 +203,7 @@ bool f2fs_is_valid_blkaddr(struct f2fs_sb_info *sbi,
 }
 
 /*
- * Readahead CP/NAT/SIT/SSA pages
+ * Readahead CP/NAT/SIT/SSA/POR pages
  */
 int f2fs_ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages,
 							int type, bool sync)
@@ -581,7 +578,7 @@ int f2fs_acquire_orphan_inode(struct f2fs_sb_info *sbi)
 
 	if (time_to_inject(sbi, FAULT_ORPHAN)) {
 		spin_unlock(&im->ino_lock);
-		f2fs_show_injection_info(FAULT_ORPHAN);
+		f2fs_show_injection_info(sbi, FAULT_ORPHAN);
 		return -ENOSPC;
 	}
 
@@ -898,7 +895,7 @@ int f2fs_get_valid_checkpoint(struct f2fs_sb_info *sbi)
 		return -ENOMEM;
 	/*
 	 * Finding out valid cp block involves read both
-	 * sets( cp pack1 and cp pack 2)
+	 * sets( cp pack 1 and cp pack 2)
 	 */
 	cp_start_blk_no = le32_to_cpu(fsb->cp_blkaddr);
 	cp1 = validate_checkpoint(sbi, cp_start_blk_no, &cp1_version);
@@ -1250,20 +1247,20 @@ static void unblock_operations(struct f2fs_sb_info *sbi)
 	f2fs_unlock_all(sbi);
 }
 
-void f2fs_wait_on_all_pages_writeback(struct f2fs_sb_info *sbi)
+void f2fs_wait_on_all_pages(struct f2fs_sb_info *sbi, int type)
 {
 	DEFINE_WAIT(wait);
 
 	for (;;) {
 		prepare_to_wait(&sbi->cp_wait, &wait, TASK_UNINTERRUPTIBLE);
 
-		if (!get_pages(sbi, F2FS_WB_CP_DATA))
+		if (!get_pages(sbi, type))
 			break;
 
 		if (unlikely(f2fs_cp_error(sbi)))
 			break;
 
-		io_schedule_timeout(5*HZ);
+		io_schedule_timeout(DEFAULT_IO_TIMEOUT);
 	}
 	finish_wait(&sbi->cp_wait, &wait);
 }
@@ -1301,10 +1298,14 @@ static void update_ckpt_flags(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	else
 		__clear_ckpt_flags(ckpt, CP_ORPHAN_PRESENT_FLAG);
 
-	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK) ||
-		is_sbi_flag_set(sbi, SBI_IS_RESIZEFS))
+	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK))
 		__set_ckpt_flags(ckpt, CP_FSCK_FLAG);
 
+	if (is_sbi_flag_set(sbi, SBI_IS_RESIZEFS))
+		__set_ckpt_flags(ckpt, CP_RESIZEFS_FLAG);
+	else
+		__clear_ckpt_flags(ckpt, CP_RESIZEFS_FLAG);
+
 	if (is_sbi_flag_set(sbi, SBI_CP_DISABLED))
 		__set_ckpt_flags(ckpt, CP_DISABLED_FLAG);
 	else
@@ -1384,13 +1385,8 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	/* Flush all the NAT/SIT pages */
 	f2fs_sync_meta_pages(sbi, META, LONG_MAX, FS_CP_META_IO);
-	f2fs_bug_on(sbi, get_pages(sbi, F2FS_DIRTY_META) &&
-					!f2fs_cp_error(sbi));
 
-	/*
-	 * modify checkpoint
-	 * version number is already updated
-	 */
+	/* start to update checkpoint, cp ver is already updated previously */
 	ckpt->elapsed_time = cpu_to_le64(get_mtime(sbi, true));
 	ckpt->free_segment_count = cpu_to_le32(free_segments(sbi));
 	for (i = 0; i < NR_CURSEG_NODE_TYPE; i++) {
@@ -1493,11 +1489,11 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	/* Here, we have one bio having CP pack except cp pack 2 page */
 	f2fs_sync_meta_pages(sbi, META, LONG_MAX, FS_CP_META_IO);
-	f2fs_bug_on(sbi, get_pages(sbi, F2FS_DIRTY_META) &&
-					!f2fs_cp_error(sbi));
+	/* Wait for all dirty meta pages to be submitted for IO */
+	f2fs_wait_on_all_pages(sbi, F2FS_DIRTY_META);
 
 	/* wait for previous submitted meta pages writeback */
-	f2fs_wait_on_all_pages_writeback(sbi);
+	f2fs_wait_on_all_pages(sbi, F2FS_WB_CP_DATA);
 
 	/* flush all device cache */
 	err = f2fs_flush_device_cache(sbi);
@@ -1506,13 +1502,13 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	/* barrier and flush checkpoint cp pack 2 page if it can */
 	commit_checkpoint(sbi, ckpt, start_blk);
-	f2fs_wait_on_all_pages_writeback(sbi);
+	f2fs_wait_on_all_pages(sbi, F2FS_WB_CP_DATA);
 
 	/*
-	 * invalidate intermediate page cache borrowed from meta inode
-	 * which are used for migration of encrypted inode's blocks.
+	 * invalidate intermediate page cache borrowed from meta inode which are
+	 * used for migration of encrypted or verity inode's blocks.
 	 */
-	if (f2fs_sb_has_encrypt(sbi))
+	if (f2fs_sb_has_encrypt(sbi) || f2fs_sb_has_verity(sbi))
 		invalidate_mapping_pages(META_MAPPING(sbi),
 				MAIN_BLKADDR(sbi), MAX_BLKADDR(sbi) - 1);
 
@@ -1543,9 +1539,6 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	return unlikely(f2fs_cp_error(sbi)) ? -EIO : 0;
 }
 
-/*
- * We guarantee that this checkpoint procedure will not fail.
- */
 int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 {
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
@@ -1613,7 +1606,6 @@ int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	f2fs_flush_sit_entries(sbi, cpc);
 
-	/* unlock all the fs_lock[] in do_checkpoint() */
 	err = do_checkpoint(sbi, cpc);
 	if (err)
 		f2fs_release_discard_addrs(sbi);
@@ -1626,7 +1618,7 @@ stop:
 	if (cpc->reason & CP_RECOVERY)
 		f2fs_notice(sbi, "checkpoint: version = %llx", ckpt_ver);
 
-	/* do checkpoint periodically */
+	/* update CP_TIME to trigger checkpoint periodically */
 	f2fs_update_time(sbi, CP_TIME);
 	trace_f2fs_write_checkpoint(sbi->sb, cpc->reason, "finish checkpoint");
 out:
diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c
new file mode 100644
index 000000000000..df7b2d15eacd
--- /dev/null
+++ b/fs/f2fs/compress.c
@@ -0,0 +1,1381 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * f2fs compress support
+ *
+ * Copyright (c) 2019 Chao Yu <chao@kernel.org>
+ */
+
+#include <linux/fs.h>
+#include <linux/f2fs_fs.h>
+#include <linux/writeback.h>
+#include <linux/backing-dev.h>
+#include <linux/lzo.h>
+#include <linux/lz4.h>
+#include <linux/zstd.h>
+
+#include "f2fs.h"
+#include "node.h"
+#include <trace/events/f2fs.h>
+
+struct f2fs_compress_ops {
+	int (*init_compress_ctx)(struct compress_ctx *cc);
+	void (*destroy_compress_ctx)(struct compress_ctx *cc);
+	int (*compress_pages)(struct compress_ctx *cc);
+	int (*init_decompress_ctx)(struct decompress_io_ctx *dic);
+	void (*destroy_decompress_ctx)(struct decompress_io_ctx *dic);
+	int (*decompress_pages)(struct decompress_io_ctx *dic);
+};
+
+static unsigned int offset_in_cluster(struct compress_ctx *cc, pgoff_t index)
+{
+	return index & (cc->cluster_size - 1);
+}
+
+static pgoff_t cluster_idx(struct compress_ctx *cc, pgoff_t index)
+{
+	return index >> cc->log_cluster_size;
+}
+
+static pgoff_t start_idx_of_cluster(struct compress_ctx *cc)
+{
+	return cc->cluster_idx << cc->log_cluster_size;
+}
+
+bool f2fs_is_compressed_page(struct page *page)
+{
+	if (!PagePrivate(page))
+		return false;
+	if (!page_private(page))
+		return false;
+	if (IS_ATOMIC_WRITTEN_PAGE(page) || IS_DUMMY_WRITTEN_PAGE(page))
+		return false;
+	f2fs_bug_on(F2FS_M_SB(page->mapping),
+		*((u32 *)page_private(page)) != F2FS_COMPRESSED_PAGE_MAGIC);
+	return true;
+}
+
+static void f2fs_set_compressed_page(struct page *page,
+		struct inode *inode, pgoff_t index, void *data)
+{
+	SetPagePrivate(page);
+	set_page_private(page, (unsigned long)data);
+
+	/* i_crypto_info and iv index */
+	page->index = index;
+	page->mapping = inode->i_mapping;
+}
+
+static void f2fs_put_compressed_page(struct page *page)
+{
+	set_page_private(page, (unsigned long)NULL);
+	ClearPagePrivate(page);
+	page->mapping = NULL;
+	unlock_page(page);
+	put_page(page);
+}
+
+static void f2fs_drop_rpages(struct compress_ctx *cc, int len, bool unlock)
+{
+	int i;
+
+	for (i = 0; i < len; i++) {
+		if (!cc->rpages[i])
+			continue;
+		if (unlock)
+			unlock_page(cc->rpages[i]);
+		else
+			put_page(cc->rpages[i]);
+	}
+}
+
+static void f2fs_put_rpages(struct compress_ctx *cc)
+{
+	f2fs_drop_rpages(cc, cc->cluster_size, false);
+}
+
+static void f2fs_unlock_rpages(struct compress_ctx *cc, int len)
+{
+	f2fs_drop_rpages(cc, len, true);
+}
+
+static void f2fs_put_rpages_mapping(struct compress_ctx *cc,
+				struct address_space *mapping,
+				pgoff_t start, int len)
+{
+	int i;
+
+	for (i = 0; i < len; i++) {
+		struct page *page = find_get_page(mapping, start + i);
+
+		put_page(page);
+		put_page(page);
+	}
+}
+
+static void f2fs_put_rpages_wbc(struct compress_ctx *cc,
+		struct writeback_control *wbc, bool redirty, int unlock)
+{
+	unsigned int i;
+
+	for (i = 0; i < cc->cluster_size; i++) {
+		if (!cc->rpages[i])
+			continue;
+		if (redirty)
+			redirty_page_for_writepage(wbc, cc->rpages[i]);
+		f2fs_put_page(cc->rpages[i], unlock);
+	}
+}
+
+struct page *f2fs_compress_control_page(struct page *page)
+{
+	return ((struct compress_io_ctx *)page_private(page))->rpages[0];
+}
+
+int f2fs_init_compress_ctx(struct compress_ctx *cc)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(cc->inode);
+
+	if (cc->nr_rpages)
+		return 0;
+
+	cc->rpages = f2fs_kzalloc(sbi, sizeof(struct page *) <<
+					cc->log_cluster_size, GFP_NOFS);
+	return cc->rpages ? 0 : -ENOMEM;
+}
+
+void f2fs_destroy_compress_ctx(struct compress_ctx *cc)
+{
+	kfree(cc->rpages);
+	cc->rpages = NULL;
+	cc->nr_rpages = 0;
+	cc->nr_cpages = 0;
+	cc->cluster_idx = NULL_CLUSTER;
+}
+
+void f2fs_compress_ctx_add_page(struct compress_ctx *cc, struct page *page)
+{
+	unsigned int cluster_ofs;
+
+	if (!f2fs_cluster_can_merge_page(cc, page->index))
+		f2fs_bug_on(F2FS_I_SB(cc->inode), 1);
+
+	cluster_ofs = offset_in_cluster(cc, page->index);
+	cc->rpages[cluster_ofs] = page;
+	cc->nr_rpages++;
+	cc->cluster_idx = cluster_idx(cc, page->index);
+}
+
+#ifdef CONFIG_F2FS_FS_LZO
+static int lzo_init_compress_ctx(struct compress_ctx *cc)
+{
+	cc->private = f2fs_kvmalloc(F2FS_I_SB(cc->inode),
+				LZO1X_MEM_COMPRESS, GFP_NOFS);
+	if (!cc->private)
+		return -ENOMEM;
+
+	cc->clen = lzo1x_worst_compress(PAGE_SIZE << cc->log_cluster_size);
+	return 0;
+}
+
+static void lzo_destroy_compress_ctx(struct compress_ctx *cc)
+{
+	kvfree(cc->private);
+	cc->private = NULL;
+}
+
+static int lzo_compress_pages(struct compress_ctx *cc)
+{
+	int ret;
+
+	ret = lzo1x_1_compress(cc->rbuf, cc->rlen, cc->cbuf->cdata,
+					&cc->clen, cc->private);
+	if (ret != LZO_E_OK) {
+		printk_ratelimited("%sF2FS-fs (%s): lzo compress failed, ret:%d\n",
+				KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id, ret);
+		return -EIO;
+	}
+	return 0;
+}
+
+static int lzo_decompress_pages(struct decompress_io_ctx *dic)
+{
+	int ret;
+
+	ret = lzo1x_decompress_safe(dic->cbuf->cdata, dic->clen,
+						dic->rbuf, &dic->rlen);
+	if (ret != LZO_E_OK) {
+		printk_ratelimited("%sF2FS-fs (%s): lzo decompress failed, ret:%d\n",
+				KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id, ret);
+		return -EIO;
+	}
+
+	if (dic->rlen != PAGE_SIZE << dic->log_cluster_size) {
+		printk_ratelimited("%sF2FS-fs (%s): lzo invalid rlen:%zu, "
+					"expected:%lu\n", KERN_ERR,
+					F2FS_I_SB(dic->inode)->sb->s_id,
+					dic->rlen,
+					PAGE_SIZE << dic->log_cluster_size);
+		return -EIO;
+	}
+	return 0;
+}
+
+static const struct f2fs_compress_ops f2fs_lzo_ops = {
+	.init_compress_ctx	= lzo_init_compress_ctx,
+	.destroy_compress_ctx	= lzo_destroy_compress_ctx,
+	.compress_pages		= lzo_compress_pages,
+	.decompress_pages	= lzo_decompress_pages,
+};
+#endif
+
+#ifdef CONFIG_F2FS_FS_LZ4
+static int lz4_init_compress_ctx(struct compress_ctx *cc)
+{
+	cc->private = f2fs_kvmalloc(F2FS_I_SB(cc->inode),
+				LZ4_MEM_COMPRESS, GFP_NOFS);
+	if (!cc->private)
+		return -ENOMEM;
+
+	cc->clen = LZ4_compressBound(PAGE_SIZE << cc->log_cluster_size);
+	return 0;
+}
+
+static void lz4_destroy_compress_ctx(struct compress_ctx *cc)
+{
+	kvfree(cc->private);
+	cc->private = NULL;
+}
+
+static int lz4_compress_pages(struct compress_ctx *cc)
+{
+	int len;
+
+	len = LZ4_compress_default(cc->rbuf, cc->cbuf->cdata, cc->rlen,
+						cc->clen, cc->private);
+	if (!len) {
+		printk_ratelimited("%sF2FS-fs (%s): lz4 compress failed\n",
+				KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id);
+		return -EIO;
+	}
+	cc->clen = len;
+	return 0;
+}
+
+static int lz4_decompress_pages(struct decompress_io_ctx *dic)
+{
+	int ret;
+
+	ret = LZ4_decompress_safe(dic->cbuf->cdata, dic->rbuf,
+						dic->clen, dic->rlen);
+	if (ret < 0) {
+		printk_ratelimited("%sF2FS-fs (%s): lz4 decompress failed, ret:%d\n",
+				KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id, ret);
+		return -EIO;
+	}
+
+	if (ret != PAGE_SIZE << dic->log_cluster_size) {
+		printk_ratelimited("%sF2FS-fs (%s): lz4 invalid rlen:%zu, "
+					"expected:%lu\n", KERN_ERR,
+					F2FS_I_SB(dic->inode)->sb->s_id,
+					dic->rlen,
+					PAGE_SIZE << dic->log_cluster_size);
+		return -EIO;
+	}
+	return 0;
+}
+
+static const struct f2fs_compress_ops f2fs_lz4_ops = {
+	.init_compress_ctx	= lz4_init_compress_ctx,
+	.destroy_compress_ctx	= lz4_destroy_compress_ctx,
+	.compress_pages		= lz4_compress_pages,
+	.decompress_pages	= lz4_decompress_pages,
+};
+#endif
+
+#ifdef CONFIG_F2FS_FS_ZSTD
+#define F2FS_ZSTD_DEFAULT_CLEVEL	1
+
+static int zstd_init_compress_ctx(struct compress_ctx *cc)
+{
+	ZSTD_parameters params;
+	ZSTD_CStream *stream;
+	void *workspace;
+	unsigned int workspace_size;
+
+	params = ZSTD_getParams(F2FS_ZSTD_DEFAULT_CLEVEL, cc->rlen, 0);
+	workspace_size = ZSTD_CStreamWorkspaceBound(params.cParams);
+
+	workspace = f2fs_kvmalloc(F2FS_I_SB(cc->inode),
+					workspace_size, GFP_NOFS);
+	if (!workspace)
+		return -ENOMEM;
+
+	stream = ZSTD_initCStream(params, 0, workspace, workspace_size);
+	if (!stream) {
+		printk_ratelimited("%sF2FS-fs (%s): %s ZSTD_initCStream failed\n",
+				KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id,
+				__func__);
+		kvfree(workspace);
+		return -EIO;
+	}
+
+	cc->private = workspace;
+	cc->private2 = stream;
+
+	cc->clen = cc->rlen - PAGE_SIZE - COMPRESS_HEADER_SIZE;
+	return 0;
+}
+
+static void zstd_destroy_compress_ctx(struct compress_ctx *cc)
+{
+	kvfree(cc->private);
+	cc->private = NULL;
+	cc->private2 = NULL;
+}
+
+static int zstd_compress_pages(struct compress_ctx *cc)
+{
+	ZSTD_CStream *stream = cc->private2;
+	ZSTD_inBuffer inbuf;
+	ZSTD_outBuffer outbuf;
+	int src_size = cc->rlen;
+	int dst_size = src_size - PAGE_SIZE - COMPRESS_HEADER_SIZE;
+	int ret;
+
+	inbuf.pos = 0;
+	inbuf.src = cc->rbuf;
+	inbuf.size = src_size;
+
+	outbuf.pos = 0;
+	outbuf.dst = cc->cbuf->cdata;
+	outbuf.size = dst_size;
+
+	ret = ZSTD_compressStream(stream, &outbuf, &inbuf);
+	if (ZSTD_isError(ret)) {
+		printk_ratelimited("%sF2FS-fs (%s): %s ZSTD_compressStream failed, ret: %d\n",
+				KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id,
+				__func__, ZSTD_getErrorCode(ret));
+		return -EIO;
+	}
+
+	ret = ZSTD_endStream(stream, &outbuf);
+	if (ZSTD_isError(ret)) {
+		printk_ratelimited("%sF2FS-fs (%s): %s ZSTD_endStream returned %d\n",
+				KERN_ERR, F2FS_I_SB(cc->inode)->sb->s_id,
+				__func__, ZSTD_getErrorCode(ret));
+		return -EIO;
+	}
+
+	cc->clen = outbuf.pos;
+	return 0;
+}
+
+static int zstd_init_decompress_ctx(struct decompress_io_ctx *dic)
+{
+	ZSTD_DStream *stream;
+	void *workspace;
+	unsigned int workspace_size;
+
+	workspace_size = ZSTD_DStreamWorkspaceBound(MAX_COMPRESS_WINDOW_SIZE);
+
+	workspace = f2fs_kvmalloc(F2FS_I_SB(dic->inode),
+					workspace_size, GFP_NOFS);
+	if (!workspace)
+		return -ENOMEM;
+
+	stream = ZSTD_initDStream(MAX_COMPRESS_WINDOW_SIZE,
+					workspace, workspace_size);
+	if (!stream) {
+		printk_ratelimited("%sF2FS-fs (%s): %s ZSTD_initDStream failed\n",
+				KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id,
+				__func__);
+		kvfree(workspace);
+		return -EIO;
+	}
+
+	dic->private = workspace;
+	dic->private2 = stream;
+
+	return 0;
+}
+
+static void zstd_destroy_decompress_ctx(struct decompress_io_ctx *dic)
+{
+	kvfree(dic->private);
+	dic->private = NULL;
+	dic->private2 = NULL;
+}
+
+static int zstd_decompress_pages(struct decompress_io_ctx *dic)
+{
+	ZSTD_DStream *stream = dic->private2;
+	ZSTD_inBuffer inbuf;
+	ZSTD_outBuffer outbuf;
+	int ret;
+
+	inbuf.pos = 0;
+	inbuf.src = dic->cbuf->cdata;
+	inbuf.size = dic->clen;
+
+	outbuf.pos = 0;
+	outbuf.dst = dic->rbuf;
+	outbuf.size = dic->rlen;
+
+	ret = ZSTD_decompressStream(stream, &outbuf, &inbuf);
+	if (ZSTD_isError(ret)) {
+		printk_ratelimited("%sF2FS-fs (%s): %s ZSTD_compressStream failed, ret: %d\n",
+				KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id,
+				__func__, ZSTD_getErrorCode(ret));
+		return -EIO;
+	}
+
+	if (dic->rlen != outbuf.pos) {
+		printk_ratelimited("%sF2FS-fs (%s): %s ZSTD invalid rlen:%zu, "
+				"expected:%lu\n", KERN_ERR,
+				F2FS_I_SB(dic->inode)->sb->s_id,
+				__func__, dic->rlen,
+				PAGE_SIZE << dic->log_cluster_size);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static const struct f2fs_compress_ops f2fs_zstd_ops = {
+	.init_compress_ctx	= zstd_init_compress_ctx,
+	.destroy_compress_ctx	= zstd_destroy_compress_ctx,
+	.compress_pages		= zstd_compress_pages,
+	.init_decompress_ctx	= zstd_init_decompress_ctx,
+	.destroy_decompress_ctx	= zstd_destroy_decompress_ctx,
+	.decompress_pages	= zstd_decompress_pages,
+};
+#endif
+
+static const struct f2fs_compress_ops *f2fs_cops[COMPRESS_MAX] = {
+#ifdef CONFIG_F2FS_FS_LZO
+	&f2fs_lzo_ops,
+#else
+	NULL,
+#endif
+#ifdef CONFIG_F2FS_FS_LZ4
+	&f2fs_lz4_ops,
+#else
+	NULL,
+#endif
+#ifdef CONFIG_F2FS_FS_ZSTD
+	&f2fs_zstd_ops,
+#else
+	NULL,
+#endif
+};
+
+bool f2fs_is_compress_backend_ready(struct inode *inode)
+{
+	if (!f2fs_compressed_file(inode))
+		return true;
+	return f2fs_cops[F2FS_I(inode)->i_compress_algorithm];
+}
+
+static struct page *f2fs_grab_page(void)
+{
+	struct page *page;
+
+	page = alloc_page(GFP_NOFS);
+	if (!page)
+		return NULL;
+	lock_page(page);
+	return page;
+}
+
+static int f2fs_compress_pages(struct compress_ctx *cc)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(cc->inode);
+	struct f2fs_inode_info *fi = F2FS_I(cc->inode);
+	const struct f2fs_compress_ops *cops =
+				f2fs_cops[fi->i_compress_algorithm];
+	unsigned int max_len, nr_cpages;
+	int i, ret;
+
+	trace_f2fs_compress_pages_start(cc->inode, cc->cluster_idx,
+				cc->cluster_size, fi->i_compress_algorithm);
+
+	if (cops->init_compress_ctx) {
+		ret = cops->init_compress_ctx(cc);
+		if (ret)
+			goto out;
+	}
+
+	max_len = COMPRESS_HEADER_SIZE + cc->clen;
+	cc->nr_cpages = DIV_ROUND_UP(max_len, PAGE_SIZE);
+
+	cc->cpages = f2fs_kzalloc(sbi, sizeof(struct page *) *
+					cc->nr_cpages, GFP_NOFS);
+	if (!cc->cpages) {
+		ret = -ENOMEM;
+		goto destroy_compress_ctx;
+	}
+
+	for (i = 0; i < cc->nr_cpages; i++) {
+		cc->cpages[i] = f2fs_grab_page();
+		if (!cc->cpages[i]) {
+			ret = -ENOMEM;
+			goto out_free_cpages;
+		}
+	}
+
+	cc->rbuf = vmap(cc->rpages, cc->cluster_size, VM_MAP, PAGE_KERNEL_RO);
+	if (!cc->rbuf) {
+		ret = -ENOMEM;
+		goto out_free_cpages;
+	}
+
+	cc->cbuf = vmap(cc->cpages, cc->nr_cpages, VM_MAP, PAGE_KERNEL);
+	if (!cc->cbuf) {
+		ret = -ENOMEM;
+		goto out_vunmap_rbuf;
+	}
+
+	ret = cops->compress_pages(cc);
+	if (ret)
+		goto out_vunmap_cbuf;
+
+	max_len = PAGE_SIZE * (cc->cluster_size - 1) - COMPRESS_HEADER_SIZE;
+
+	if (cc->clen > max_len) {
+		ret = -EAGAIN;
+		goto out_vunmap_cbuf;
+	}
+
+	cc->cbuf->clen = cpu_to_le32(cc->clen);
+
+	for (i = 0; i < COMPRESS_DATA_RESERVED_SIZE; i++)
+		cc->cbuf->reserved[i] = cpu_to_le32(0);
+
+	nr_cpages = DIV_ROUND_UP(cc->clen + COMPRESS_HEADER_SIZE, PAGE_SIZE);
+
+	/* zero out any unused part of the last page */
+	memset(&cc->cbuf->cdata[cc->clen], 0,
+	       (nr_cpages * PAGE_SIZE) - (cc->clen + COMPRESS_HEADER_SIZE));
+
+	vunmap(cc->cbuf);
+	vunmap(cc->rbuf);
+
+	for (i = nr_cpages; i < cc->nr_cpages; i++) {
+		f2fs_put_compressed_page(cc->cpages[i]);
+		cc->cpages[i] = NULL;
+	}
+
+	if (cops->destroy_compress_ctx)
+		cops->destroy_compress_ctx(cc);
+
+	cc->nr_cpages = nr_cpages;
+
+	trace_f2fs_compress_pages_end(cc->inode, cc->cluster_idx,
+							cc->clen, ret);
+	return 0;
+
+out_vunmap_cbuf:
+	vunmap(cc->cbuf);
+out_vunmap_rbuf:
+	vunmap(cc->rbuf);
+out_free_cpages:
+	for (i = 0; i < cc->nr_cpages; i++) {
+		if (cc->cpages[i])
+			f2fs_put_compressed_page(cc->cpages[i]);
+	}
+	kfree(cc->cpages);
+	cc->cpages = NULL;
+destroy_compress_ctx:
+	if (cops->destroy_compress_ctx)
+		cops->destroy_compress_ctx(cc);
+out:
+	trace_f2fs_compress_pages_end(cc->inode, cc->cluster_idx,
+							cc->clen, ret);
+	return ret;
+}
+
+void f2fs_decompress_pages(struct bio *bio, struct page *page, bool verity)
+{
+	struct decompress_io_ctx *dic =
+			(struct decompress_io_ctx *)page_private(page);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(dic->inode);
+	struct f2fs_inode_info *fi= F2FS_I(dic->inode);
+	const struct f2fs_compress_ops *cops =
+			f2fs_cops[fi->i_compress_algorithm];
+	int ret;
+
+	dec_page_count(sbi, F2FS_RD_DATA);
+
+	if (bio->bi_status || PageError(page))
+		dic->failed = true;
+
+	if (refcount_dec_not_one(&dic->ref))
+		return;
+
+	trace_f2fs_decompress_pages_start(dic->inode, dic->cluster_idx,
+				dic->cluster_size, fi->i_compress_algorithm);
+
+	/* submit partial compressed pages */
+	if (dic->failed) {
+		ret = -EIO;
+		goto out_free_dic;
+	}
+
+	if (cops->init_decompress_ctx) {
+		ret = cops->init_decompress_ctx(dic);
+		if (ret)
+			goto out_free_dic;
+	}
+
+	dic->rbuf = vmap(dic->tpages, dic->cluster_size, VM_MAP, PAGE_KERNEL);
+	if (!dic->rbuf) {
+		ret = -ENOMEM;
+		goto destroy_decompress_ctx;
+	}
+
+	dic->cbuf = vmap(dic->cpages, dic->nr_cpages, VM_MAP, PAGE_KERNEL_RO);
+	if (!dic->cbuf) {
+		ret = -ENOMEM;
+		goto out_vunmap_rbuf;
+	}
+
+	dic->clen = le32_to_cpu(dic->cbuf->clen);
+	dic->rlen = PAGE_SIZE << dic->log_cluster_size;
+
+	if (dic->clen > PAGE_SIZE * dic->nr_cpages - COMPRESS_HEADER_SIZE) {
+		ret = -EFSCORRUPTED;
+		goto out_vunmap_cbuf;
+	}
+
+	ret = cops->decompress_pages(dic);
+
+out_vunmap_cbuf:
+	vunmap(dic->cbuf);
+out_vunmap_rbuf:
+	vunmap(dic->rbuf);
+destroy_decompress_ctx:
+	if (cops->destroy_decompress_ctx)
+		cops->destroy_decompress_ctx(dic);
+out_free_dic:
+	if (verity)
+		refcount_set(&dic->ref, dic->nr_cpages);
+	if (!verity)
+		f2fs_decompress_end_io(dic->rpages, dic->cluster_size,
+								ret, false);
+
+	trace_f2fs_decompress_pages_end(dic->inode, dic->cluster_idx,
+							dic->clen, ret);
+	if (!verity)
+		f2fs_free_dic(dic);
+}
+
+static bool is_page_in_cluster(struct compress_ctx *cc, pgoff_t index)
+{
+	if (cc->cluster_idx == NULL_CLUSTER)
+		return true;
+	return cc->cluster_idx == cluster_idx(cc, index);
+}
+
+bool f2fs_cluster_is_empty(struct compress_ctx *cc)
+{
+	return cc->nr_rpages == 0;
+}
+
+static bool f2fs_cluster_is_full(struct compress_ctx *cc)
+{
+	return cc->cluster_size == cc->nr_rpages;
+}
+
+bool f2fs_cluster_can_merge_page(struct compress_ctx *cc, pgoff_t index)
+{
+	if (f2fs_cluster_is_empty(cc))
+		return true;
+	return is_page_in_cluster(cc, index);
+}
+
+static bool __cluster_may_compress(struct compress_ctx *cc)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(cc->inode);
+	loff_t i_size = i_size_read(cc->inode);
+	unsigned nr_pages = DIV_ROUND_UP(i_size, PAGE_SIZE);
+	int i;
+
+	for (i = 0; i < cc->cluster_size; i++) {
+		struct page *page = cc->rpages[i];
+
+		f2fs_bug_on(sbi, !page);
+
+		if (unlikely(f2fs_cp_error(sbi)))
+			return false;
+		if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
+			return false;
+
+		/* beyond EOF */
+		if (page->index >= nr_pages)
+			return false;
+	}
+	return true;
+}
+
+static int __f2fs_cluster_blocks(struct compress_ctx *cc, bool compr)
+{
+	struct dnode_of_data dn;
+	int ret;
+
+	set_new_dnode(&dn, cc->inode, NULL, NULL, 0);
+	ret = f2fs_get_dnode_of_data(&dn, start_idx_of_cluster(cc),
+							LOOKUP_NODE);
+	if (ret) {
+		if (ret == -ENOENT)
+			ret = 0;
+		goto fail;
+	}
+
+	if (dn.data_blkaddr == COMPRESS_ADDR) {
+		int i;
+
+		ret = 1;
+		for (i = 1; i < cc->cluster_size; i++) {
+			block_t blkaddr;
+
+			blkaddr = data_blkaddr(dn.inode,
+					dn.node_page, dn.ofs_in_node + i);
+			if (compr) {
+				if (__is_valid_data_blkaddr(blkaddr))
+					ret++;
+			} else {
+				if (blkaddr != NULL_ADDR)
+					ret++;
+			}
+		}
+	}
+fail:
+	f2fs_put_dnode(&dn);
+	return ret;
+}
+
+/* return # of compressed blocks in compressed cluster */
+static int f2fs_compressed_blocks(struct compress_ctx *cc)
+{
+	return __f2fs_cluster_blocks(cc, true);
+}
+
+/* return # of valid blocks in compressed cluster */
+static int f2fs_cluster_blocks(struct compress_ctx *cc, bool compr)
+{
+	return __f2fs_cluster_blocks(cc, false);
+}
+
+int f2fs_is_compressed_cluster(struct inode *inode, pgoff_t index)
+{
+	struct compress_ctx cc = {
+		.inode = inode,
+		.log_cluster_size = F2FS_I(inode)->i_log_cluster_size,
+		.cluster_size = F2FS_I(inode)->i_cluster_size,
+		.cluster_idx = index >> F2FS_I(inode)->i_log_cluster_size,
+	};
+
+	return f2fs_cluster_blocks(&cc, false);
+}
+
+static bool cluster_may_compress(struct compress_ctx *cc)
+{
+	if (!f2fs_compressed_file(cc->inode))
+		return false;
+	if (f2fs_is_atomic_file(cc->inode))
+		return false;
+	if (f2fs_is_mmap_file(cc->inode))
+		return false;
+	if (!f2fs_cluster_is_full(cc))
+		return false;
+	return __cluster_may_compress(cc);
+}
+
+static void set_cluster_writeback(struct compress_ctx *cc)
+{
+	int i;
+
+	for (i = 0; i < cc->cluster_size; i++) {
+		if (cc->rpages[i])
+			set_page_writeback(cc->rpages[i]);
+	}
+}
+
+static void set_cluster_dirty(struct compress_ctx *cc)
+{
+	int i;
+
+	for (i = 0; i < cc->cluster_size; i++)
+		if (cc->rpages[i])
+			set_page_dirty(cc->rpages[i]);
+}
+
+static int prepare_compress_overwrite(struct compress_ctx *cc,
+		struct page **pagep, pgoff_t index, void **fsdata)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(cc->inode);
+	struct address_space *mapping = cc->inode->i_mapping;
+	struct page *page;
+	struct dnode_of_data dn;
+	sector_t last_block_in_bio;
+	unsigned fgp_flag = FGP_LOCK | FGP_WRITE | FGP_CREAT;
+	pgoff_t start_idx = start_idx_of_cluster(cc);
+	int i, ret;
+	bool prealloc;
+
+retry:
+	ret = f2fs_cluster_blocks(cc, false);
+	if (ret <= 0)
+		return ret;
+
+	/* compressed case */
+	prealloc = (ret < cc->cluster_size);
+
+	ret = f2fs_init_compress_ctx(cc);
+	if (ret)
+		return ret;
+
+	/* keep page reference to avoid page reclaim */
+	for (i = 0; i < cc->cluster_size; i++) {
+		page = f2fs_pagecache_get_page(mapping, start_idx + i,
+							fgp_flag, GFP_NOFS);
+		if (!page) {
+			ret = -ENOMEM;
+			goto unlock_pages;
+		}
+
+		if (PageUptodate(page))
+			unlock_page(page);
+		else
+			f2fs_compress_ctx_add_page(cc, page);
+	}
+
+	if (!f2fs_cluster_is_empty(cc)) {
+		struct bio *bio = NULL;
+
+		ret = f2fs_read_multi_pages(cc, &bio, cc->cluster_size,
+					&last_block_in_bio, false, true);
+		f2fs_destroy_compress_ctx(cc);
+		if (ret)
+			goto release_pages;
+		if (bio)
+			f2fs_submit_bio(sbi, bio, DATA);
+
+		ret = f2fs_init_compress_ctx(cc);
+		if (ret)
+			goto release_pages;
+	}
+
+	for (i = 0; i < cc->cluster_size; i++) {
+		f2fs_bug_on(sbi, cc->rpages[i]);
+
+		page = find_lock_page(mapping, start_idx + i);
+		f2fs_bug_on(sbi, !page);
+
+		f2fs_wait_on_page_writeback(page, DATA, true, true);
+
+		f2fs_compress_ctx_add_page(cc, page);
+		f2fs_put_page(page, 0);
+
+		if (!PageUptodate(page)) {
+			f2fs_unlock_rpages(cc, i + 1);
+			f2fs_put_rpages_mapping(cc, mapping, start_idx,
+					cc->cluster_size);
+			f2fs_destroy_compress_ctx(cc);
+			goto retry;
+		}
+	}
+
+	if (prealloc) {
+		__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true);
+
+		set_new_dnode(&dn, cc->inode, NULL, NULL, 0);
+
+		for (i = cc->cluster_size - 1; i > 0; i--) {
+			ret = f2fs_get_block(&dn, start_idx + i);
+			if (ret) {
+				i = cc->cluster_size;
+				break;
+			}
+
+			if (dn.data_blkaddr != NEW_ADDR)
+				break;
+		}
+
+		__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false);
+	}
+
+	if (likely(!ret)) {
+		*fsdata = cc->rpages;
+		*pagep = cc->rpages[offset_in_cluster(cc, index)];
+		return cc->cluster_size;
+	}
+
+unlock_pages:
+	f2fs_unlock_rpages(cc, i);
+release_pages:
+	f2fs_put_rpages_mapping(cc, mapping, start_idx, i);
+	f2fs_destroy_compress_ctx(cc);
+	return ret;
+}
+
+int f2fs_prepare_compress_overwrite(struct inode *inode,
+		struct page **pagep, pgoff_t index, void **fsdata)
+{
+	struct compress_ctx cc = {
+		.inode = inode,
+		.log_cluster_size = F2FS_I(inode)->i_log_cluster_size,
+		.cluster_size = F2FS_I(inode)->i_cluster_size,
+		.cluster_idx = index >> F2FS_I(inode)->i_log_cluster_size,
+		.rpages = NULL,
+		.nr_rpages = 0,
+	};
+
+	return prepare_compress_overwrite(&cc, pagep, index, fsdata);
+}
+
+bool f2fs_compress_write_end(struct inode *inode, void *fsdata,
+					pgoff_t index, unsigned copied)
+
+{
+	struct compress_ctx cc = {
+		.log_cluster_size = F2FS_I(inode)->i_log_cluster_size,
+		.cluster_size = F2FS_I(inode)->i_cluster_size,
+		.rpages = fsdata,
+	};
+	bool first_index = (index == cc.rpages[0]->index);
+
+	if (copied)
+		set_cluster_dirty(&cc);
+
+	f2fs_put_rpages_wbc(&cc, NULL, false, 1);
+	f2fs_destroy_compress_ctx(&cc);
+
+	return first_index;
+}
+
+static int f2fs_write_compressed_pages(struct compress_ctx *cc,
+					int *submitted,
+					struct writeback_control *wbc,
+					enum iostat_type io_type)
+{
+	struct inode *inode = cc->inode;
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+	struct f2fs_io_info fio = {
+		.sbi = sbi,
+		.ino = cc->inode->i_ino,
+		.type = DATA,
+		.op = REQ_OP_WRITE,
+		.op_flags = wbc_to_write_flags(wbc),
+		.old_blkaddr = NEW_ADDR,
+		.page = NULL,
+		.encrypted_page = NULL,
+		.compressed_page = NULL,
+		.submitted = false,
+		.io_type = io_type,
+		.io_wbc = wbc,
+		.encrypted = f2fs_encrypted_file(cc->inode),
+	};
+	struct dnode_of_data dn;
+	struct node_info ni;
+	struct compress_io_ctx *cic;
+	pgoff_t start_idx = start_idx_of_cluster(cc);
+	unsigned int last_index = cc->cluster_size - 1;
+	loff_t psize;
+	int i, err;
+
+	if (!f2fs_trylock_op(sbi))
+		return -EAGAIN;
+
+	set_new_dnode(&dn, cc->inode, NULL, NULL, 0);
+
+	err = f2fs_get_dnode_of_data(&dn, start_idx, LOOKUP_NODE);
+	if (err)
+		goto out_unlock_op;
+
+	for (i = 0; i < cc->cluster_size; i++) {
+		if (data_blkaddr(dn.inode, dn.node_page,
+					dn.ofs_in_node + i) == NULL_ADDR)
+			goto out_put_dnode;
+	}
+
+	psize = (loff_t)(cc->rpages[last_index]->index + 1) << PAGE_SHIFT;
+
+	err = f2fs_get_node_info(fio.sbi, dn.nid, &ni);
+	if (err)
+		goto out_put_dnode;
+
+	fio.version = ni.version;
+
+	cic = f2fs_kzalloc(sbi, sizeof(struct compress_io_ctx), GFP_NOFS);
+	if (!cic)
+		goto out_put_dnode;
+
+	cic->magic = F2FS_COMPRESSED_PAGE_MAGIC;
+	cic->inode = inode;
+	refcount_set(&cic->ref, cc->nr_cpages);
+	cic->rpages = f2fs_kzalloc(sbi, sizeof(struct page *) <<
+			cc->log_cluster_size, GFP_NOFS);
+	if (!cic->rpages)
+		goto out_put_cic;
+
+	cic->nr_rpages = cc->cluster_size;
+
+	for (i = 0; i < cc->nr_cpages; i++) {
+		f2fs_set_compressed_page(cc->cpages[i], inode,
+					cc->rpages[i + 1]->index, cic);
+		fio.compressed_page = cc->cpages[i];
+		if (fio.encrypted) {
+			fio.page = cc->rpages[i + 1];
+			err = f2fs_encrypt_one_page(&fio);
+			if (err)
+				goto out_destroy_crypt;
+			cc->cpages[i] = fio.encrypted_page;
+		}
+	}
+
+	set_cluster_writeback(cc);
+
+	for (i = 0; i < cc->cluster_size; i++)
+		cic->rpages[i] = cc->rpages[i];
+
+	for (i = 0; i < cc->cluster_size; i++, dn.ofs_in_node++) {
+		block_t blkaddr;
+
+		blkaddr = f2fs_data_blkaddr(&dn);
+		fio.page = cc->rpages[i];
+		fio.old_blkaddr = blkaddr;
+
+		/* cluster header */
+		if (i == 0) {
+			if (blkaddr == COMPRESS_ADDR)
+				fio.compr_blocks++;
+			if (__is_valid_data_blkaddr(blkaddr))
+				f2fs_invalidate_blocks(sbi, blkaddr);
+			f2fs_update_data_blkaddr(&dn, COMPRESS_ADDR);
+			goto unlock_continue;
+		}
+
+		if (fio.compr_blocks && __is_valid_data_blkaddr(blkaddr))
+			fio.compr_blocks++;
+
+		if (i > cc->nr_cpages) {
+			if (__is_valid_data_blkaddr(blkaddr)) {
+				f2fs_invalidate_blocks(sbi, blkaddr);
+				f2fs_update_data_blkaddr(&dn, NEW_ADDR);
+			}
+			goto unlock_continue;
+		}
+
+		f2fs_bug_on(fio.sbi, blkaddr == NULL_ADDR);
+
+		if (fio.encrypted)
+			fio.encrypted_page = cc->cpages[i - 1];
+		else
+			fio.compressed_page = cc->cpages[i - 1];
+
+		cc->cpages[i - 1] = NULL;
+		f2fs_outplace_write_data(&dn, &fio);
+		(*submitted)++;
+unlock_continue:
+		inode_dec_dirty_pages(cc->inode);
+		unlock_page(fio.page);
+	}
+
+	if (fio.compr_blocks)
+		f2fs_i_compr_blocks_update(inode, fio.compr_blocks - 1, false);
+	f2fs_i_compr_blocks_update(inode, cc->nr_cpages, true);
+
+	set_inode_flag(cc->inode, FI_APPEND_WRITE);
+	if (cc->cluster_idx == 0)
+		set_inode_flag(inode, FI_FIRST_BLOCK_WRITTEN);
+
+	f2fs_put_dnode(&dn);
+	f2fs_unlock_op(sbi);
+
+	spin_lock(&fi->i_size_lock);
+	if (fi->last_disk_size < psize)
+		fi->last_disk_size = psize;
+	spin_unlock(&fi->i_size_lock);
+
+	f2fs_put_rpages(cc);
+	f2fs_destroy_compress_ctx(cc);
+	return 0;
+
+out_destroy_crypt:
+	kfree(cic->rpages);
+
+	for (--i; i >= 0; i--)
+		fscrypt_finalize_bounce_page(&cc->cpages[i]);
+	for (i = 0; i < cc->nr_cpages; i++) {
+		if (!cc->cpages[i])
+			continue;
+		f2fs_put_page(cc->cpages[i], 1);
+	}
+out_put_cic:
+	kfree(cic);
+out_put_dnode:
+	f2fs_put_dnode(&dn);
+out_unlock_op:
+	f2fs_unlock_op(sbi);
+	return -EAGAIN;
+}
+
+void f2fs_compress_write_end_io(struct bio *bio, struct page *page)
+{
+	struct f2fs_sb_info *sbi = bio->bi_private;
+	struct compress_io_ctx *cic =
+			(struct compress_io_ctx *)page_private(page);
+	int i;
+
+	if (unlikely(bio->bi_status))
+		mapping_set_error(cic->inode->i_mapping, -EIO);
+
+	f2fs_put_compressed_page(page);
+
+	dec_page_count(sbi, F2FS_WB_DATA);
+
+	if (refcount_dec_not_one(&cic->ref))
+		return;
+
+	for (i = 0; i < cic->nr_rpages; i++) {
+		WARN_ON(!cic->rpages[i]);
+		clear_cold_data(cic->rpages[i]);
+		end_page_writeback(cic->rpages[i]);
+	}
+
+	kfree(cic->rpages);
+	kfree(cic);
+}
+
+static int f2fs_write_raw_pages(struct compress_ctx *cc,
+					int *submitted,
+					struct writeback_control *wbc,
+					enum iostat_type io_type)
+{
+	struct address_space *mapping = cc->inode->i_mapping;
+	int _submitted, compr_blocks, ret;
+	int i = -1, err = 0;
+
+	compr_blocks = f2fs_compressed_blocks(cc);
+	if (compr_blocks < 0) {
+		err = compr_blocks;
+		goto out_err;
+	}
+
+	for (i = 0; i < cc->cluster_size; i++) {
+		if (!cc->rpages[i])
+			continue;
+retry_write:
+		if (cc->rpages[i]->mapping != mapping) {
+			unlock_page(cc->rpages[i]);
+			continue;
+		}
+
+		BUG_ON(!PageLocked(cc->rpages[i]));
+
+		ret = f2fs_write_single_data_page(cc->rpages[i], &_submitted,
+						NULL, NULL, wbc, io_type,
+						compr_blocks);
+		if (ret) {
+			if (ret == AOP_WRITEPAGE_ACTIVATE) {
+				unlock_page(cc->rpages[i]);
+				ret = 0;
+			} else if (ret == -EAGAIN) {
+				/*
+				 * for quota file, just redirty left pages to
+				 * avoid deadlock caused by cluster update race
+				 * from foreground operation.
+				 */
+				if (IS_NOQUOTA(cc->inode)) {
+					err = 0;
+					goto out_err;
+				}
+				ret = 0;
+				cond_resched();
+				congestion_wait(BLK_RW_ASYNC,
+						DEFAULT_IO_TIMEOUT);
+				lock_page(cc->rpages[i]);
+				clear_page_dirty_for_io(cc->rpages[i]);
+				goto retry_write;
+			}
+			err = ret;
+			goto out_err;
+		}
+
+		*submitted += _submitted;
+	}
+	return 0;
+out_err:
+	for (++i; i < cc->cluster_size; i++) {
+		if (!cc->rpages[i])
+			continue;
+		redirty_page_for_writepage(wbc, cc->rpages[i]);
+		unlock_page(cc->rpages[i]);
+	}
+	return err;
+}
+
+int f2fs_write_multi_pages(struct compress_ctx *cc,
+					int *submitted,
+					struct writeback_control *wbc,
+					enum iostat_type io_type)
+{
+	struct f2fs_inode_info *fi = F2FS_I(cc->inode);
+	const struct f2fs_compress_ops *cops =
+			f2fs_cops[fi->i_compress_algorithm];
+	int err;
+
+	*submitted = 0;
+	if (cluster_may_compress(cc)) {
+		err = f2fs_compress_pages(cc);
+		if (err == -EAGAIN) {
+			goto write;
+		} else if (err) {
+			f2fs_put_rpages_wbc(cc, wbc, true, 1);
+			goto destroy_out;
+		}
+
+		err = f2fs_write_compressed_pages(cc, submitted,
+							wbc, io_type);
+		cops->destroy_compress_ctx(cc);
+		if (!err)
+			return 0;
+		f2fs_bug_on(F2FS_I_SB(cc->inode), err != -EAGAIN);
+	}
+write:
+	f2fs_bug_on(F2FS_I_SB(cc->inode), *submitted);
+
+	err = f2fs_write_raw_pages(cc, submitted, wbc, io_type);
+	f2fs_put_rpages_wbc(cc, wbc, false, 0);
+destroy_out:
+	f2fs_destroy_compress_ctx(cc);
+	return err;
+}
+
+struct decompress_io_ctx *f2fs_alloc_dic(struct compress_ctx *cc)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(cc->inode);
+	struct decompress_io_ctx *dic;
+	pgoff_t start_idx = start_idx_of_cluster(cc);
+	int i;
+
+	dic = f2fs_kzalloc(sbi, sizeof(struct decompress_io_ctx), GFP_NOFS);
+	if (!dic)
+		return ERR_PTR(-ENOMEM);
+
+	dic->rpages = f2fs_kzalloc(sbi, sizeof(struct page *) <<
+			cc->log_cluster_size, GFP_NOFS);
+	if (!dic->rpages) {
+		kfree(dic);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	dic->magic = F2FS_COMPRESSED_PAGE_MAGIC;
+	dic->inode = cc->inode;
+	refcount_set(&dic->ref, cc->nr_cpages);
+	dic->cluster_idx = cc->cluster_idx;
+	dic->cluster_size = cc->cluster_size;
+	dic->log_cluster_size = cc->log_cluster_size;
+	dic->nr_cpages = cc->nr_cpages;
+	dic->failed = false;
+
+	for (i = 0; i < dic->cluster_size; i++)
+		dic->rpages[i] = cc->rpages[i];
+	dic->nr_rpages = cc->cluster_size;
+
+	dic->cpages = f2fs_kzalloc(sbi, sizeof(struct page *) *
+					dic->nr_cpages, GFP_NOFS);
+	if (!dic->cpages)
+		goto out_free;
+
+	for (i = 0; i < dic->nr_cpages; i++) {
+		struct page *page;
+
+		page = f2fs_grab_page();
+		if (!page)
+			goto out_free;
+
+		f2fs_set_compressed_page(page, cc->inode,
+					start_idx + i + 1, dic);
+		dic->cpages[i] = page;
+	}
+
+	dic->tpages = f2fs_kzalloc(sbi, sizeof(struct page *) *
+					dic->cluster_size, GFP_NOFS);
+	if (!dic->tpages)
+		goto out_free;
+
+	for (i = 0; i < dic->cluster_size; i++) {
+		if (cc->rpages[i]) {
+			dic->tpages[i] = cc->rpages[i];
+			continue;
+		}
+
+		dic->tpages[i] = f2fs_grab_page();
+		if (!dic->tpages[i])
+			goto out_free;
+	}
+
+	return dic;
+
+out_free:
+	f2fs_free_dic(dic);
+	return ERR_PTR(-ENOMEM);
+}
+
+void f2fs_free_dic(struct decompress_io_ctx *dic)
+{
+	int i;
+
+	if (dic->tpages) {
+		for (i = 0; i < dic->cluster_size; i++) {
+			if (dic->rpages[i])
+				continue;
+			if (!dic->tpages[i])
+				continue;
+			unlock_page(dic->tpages[i]);
+			put_page(dic->tpages[i]);
+		}
+		kfree(dic->tpages);
+	}
+
+	if (dic->cpages) {
+		for (i = 0; i < dic->nr_cpages; i++) {
+			if (!dic->cpages[i])
+				continue;
+			f2fs_put_compressed_page(dic->cpages[i]);
+		}
+		kfree(dic->cpages);
+	}
+
+	kfree(dic->rpages);
+	kfree(dic);
+}
+
+void f2fs_decompress_end_io(struct page **rpages,
+			unsigned int cluster_size, bool err, bool verity)
+{
+	int i;
+
+	for (i = 0; i < cluster_size; i++) {
+		struct page *rpage = rpages[i];
+
+		if (!rpage)
+			continue;
+
+		if (err || PageError(rpage))
+			goto clear_uptodate;
+
+		if (!verity || fsverity_verify_page(rpage)) {
+			SetPageUptodate(rpage);
+			goto unlock;
+		}
+clear_uptodate:
+		ClearPageUptodate(rpage);
+		ClearPageError(rpage);
+unlock:
+		unlock_page(rpage);
+	}
+}
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 5755e897a5f0..03ec97f28235 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -29,7 +29,45 @@
 #define NUM_PREALLOC_POST_READ_CTXS	128
 
 static struct kmem_cache *bio_post_read_ctx_cache;
+static struct kmem_cache *bio_entry_slab;
 static mempool_t *bio_post_read_ctx_pool;
+static struct bio_set f2fs_bioset;
+
+#define	F2FS_BIO_POOL_SIZE	NR_CURSEG_TYPE
+
+int __init f2fs_init_bioset(void)
+{
+	if (bioset_init(&f2fs_bioset, F2FS_BIO_POOL_SIZE,
+					0, BIOSET_NEED_BVECS))
+		return -ENOMEM;
+	return 0;
+}
+
+void f2fs_destroy_bioset(void)
+{
+	bioset_exit(&f2fs_bioset);
+}
+
+static inline struct bio *__f2fs_bio_alloc(gfp_t gfp_mask,
+						unsigned int nr_iovecs)
+{
+	return bio_alloc_bioset(gfp_mask, nr_iovecs, &f2fs_bioset);
+}
+
+struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi, int npages, bool noio)
+{
+	if (noio) {
+		/* No failure on bio allocation */
+		return __f2fs_bio_alloc(GFP_NOIO, npages);
+	}
+
+	if (time_to_inject(sbi, FAULT_ALLOC_BIO)) {
+		f2fs_show_injection_info(sbi, FAULT_ALLOC_BIO);
+		return NULL;
+	}
+
+	return __f2fs_bio_alloc(GFP_KERNEL, npages);
+}
 
 static bool __is_cp_guaranteed(struct page *page)
 {
@@ -40,6 +78,9 @@ static bool __is_cp_guaranteed(struct page *page)
 	if (!mapping)
 		return false;
 
+	if (f2fs_is_compressed_page(page))
+		return false;
+
 	inode = mapping->host;
 	sbi = F2FS_I_SB(inode);
 
@@ -72,19 +113,19 @@ static enum count_type __read_io_type(struct page *page)
 
 /* postprocessing steps for read bios */
 enum bio_post_read_step {
-	STEP_INITIAL = 0,
 	STEP_DECRYPT,
+	STEP_DECOMPRESS,
 	STEP_VERITY,
 };
 
 struct bio_post_read_ctx {
 	struct bio *bio;
+	struct f2fs_sb_info *sbi;
 	struct work_struct work;
-	unsigned int cur_step;
 	unsigned int enabled_steps;
 };
 
-static void __read_end_io(struct bio *bio)
+static void __read_end_io(struct bio *bio, bool compr, bool verity)
 {
 	struct page *page;
 	struct bio_vec *bv;
@@ -93,6 +134,15 @@ static void __read_end_io(struct bio *bio)
 	bio_for_each_segment_all(bv, bio, iter_all) {
 		page = bv->bv_page;
 
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+		if (compr && f2fs_is_compressed_page(page)) {
+			f2fs_decompress_pages(bio, page, verity);
+			continue;
+		}
+		if (verity)
+			continue;
+#endif
+
 		/* PG_error was set if any post_read step failed */
 		if (bio->bi_status || PageError(page)) {
 			ClearPageUptodate(page);
@@ -104,31 +154,133 @@ static void __read_end_io(struct bio *bio)
 		dec_page_count(F2FS_P_SB(page), __read_io_type(page));
 		unlock_page(page);
 	}
-	if (bio->bi_private)
-		mempool_free(bio->bi_private, bio_post_read_ctx_pool);
-	bio_put(bio);
+}
+
+static void f2fs_release_read_bio(struct bio *bio);
+static void __f2fs_read_end_io(struct bio *bio, bool compr, bool verity)
+{
+	if (!compr)
+		__read_end_io(bio, false, verity);
+	f2fs_release_read_bio(bio);
+}
+
+static void f2fs_decompress_bio(struct bio *bio, bool verity)
+{
+	__read_end_io(bio, true, verity);
 }
 
 static void bio_post_read_processing(struct bio_post_read_ctx *ctx);
 
-static void decrypt_work(struct work_struct *work)
+static void f2fs_decrypt_work(struct bio_post_read_ctx *ctx)
+{
+	fscrypt_decrypt_bio(ctx->bio);
+}
+
+static void f2fs_decompress_work(struct bio_post_read_ctx *ctx)
+{
+	f2fs_decompress_bio(ctx->bio, ctx->enabled_steps & (1 << STEP_VERITY));
+}
+
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+static void f2fs_verify_pages(struct page **rpages, unsigned int cluster_size)
+{
+	f2fs_decompress_end_io(rpages, cluster_size, false, true);
+}
+
+static void f2fs_verify_bio(struct bio *bio)
+{
+	struct bio_vec *bv;
+	struct bvec_iter_all iter_all;
+
+	bio_for_each_segment_all(bv, bio, iter_all) {
+		struct page *page = bv->bv_page;
+		struct decompress_io_ctx *dic;
+
+		dic = (struct decompress_io_ctx *)page_private(page);
+
+		if (dic) {
+			if (refcount_dec_not_one(&dic->ref))
+				continue;
+			f2fs_verify_pages(dic->rpages,
+						dic->cluster_size);
+			f2fs_free_dic(dic);
+			continue;
+		}
+
+		if (bio->bi_status || PageError(page))
+			goto clear_uptodate;
+
+		if (fsverity_verify_page(page)) {
+			SetPageUptodate(page);
+			goto unlock;
+		}
+clear_uptodate:
+		ClearPageUptodate(page);
+		ClearPageError(page);
+unlock:
+		dec_page_count(F2FS_P_SB(page), __read_io_type(page));
+		unlock_page(page);
+	}
+}
+#endif
+
+static void f2fs_verity_work(struct work_struct *work)
 {
 	struct bio_post_read_ctx *ctx =
 		container_of(work, struct bio_post_read_ctx, work);
+	struct bio *bio = ctx->bio;
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	unsigned int enabled_steps = ctx->enabled_steps;
+#endif
 
-	fscrypt_decrypt_bio(ctx->bio);
+	/*
+	 * fsverity_verify_bio() may call readpages() again, and while verity
+	 * will be disabled for this, decryption may still be needed, resulting
+	 * in another bio_post_read_ctx being allocated.  So to prevent
+	 * deadlocks we need to release the current ctx to the mempool first.
+	 * This assumes that verity is the last post-read step.
+	 */
+	mempool_free(ctx, bio_post_read_ctx_pool);
+	bio->bi_private = NULL;
+
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	/* previous step is decompression */
+	if (enabled_steps & (1 << STEP_DECOMPRESS)) {
+		f2fs_verify_bio(bio);
+		f2fs_release_read_bio(bio);
+		return;
+	}
+#endif
 
-	bio_post_read_processing(ctx);
+	fsverity_verify_bio(bio);
+	__f2fs_read_end_io(bio, false, false);
 }
 
-static void verity_work(struct work_struct *work)
+static void f2fs_post_read_work(struct work_struct *work)
 {
 	struct bio_post_read_ctx *ctx =
 		container_of(work, struct bio_post_read_ctx, work);
 
-	fsverity_verify_bio(ctx->bio);
+	if (ctx->enabled_steps & (1 << STEP_DECRYPT))
+		f2fs_decrypt_work(ctx);
 
-	bio_post_read_processing(ctx);
+	if (ctx->enabled_steps & (1 << STEP_DECOMPRESS))
+		f2fs_decompress_work(ctx);
+
+	if (ctx->enabled_steps & (1 << STEP_VERITY)) {
+		INIT_WORK(&ctx->work, f2fs_verity_work);
+		fsverity_enqueue_verify_work(&ctx->work);
+		return;
+	}
+
+	__f2fs_read_end_io(ctx->bio,
+		ctx->enabled_steps & (1 << STEP_DECOMPRESS), false);
+}
+
+static void f2fs_enqueue_post_read_work(struct f2fs_sb_info *sbi,
+						struct work_struct *work)
+{
+	queue_work(sbi->post_read_wq, work);
 }
 
 static void bio_post_read_processing(struct bio_post_read_ctx *ctx)
@@ -138,50 +290,45 @@ static void bio_post_read_processing(struct bio_post_read_ctx *ctx)
 	 * verity may require reading metadata pages that need decryption, and
 	 * we shouldn't recurse to the same workqueue.
 	 */
-	switch (++ctx->cur_step) {
-	case STEP_DECRYPT:
-		if (ctx->enabled_steps & (1 << STEP_DECRYPT)) {
-			INIT_WORK(&ctx->work, decrypt_work);
-			fscrypt_enqueue_decrypt_work(&ctx->work);
-			return;
-		}
-		ctx->cur_step++;
-		/* fall-through */
-	case STEP_VERITY:
-		if (ctx->enabled_steps & (1 << STEP_VERITY)) {
-			INIT_WORK(&ctx->work, verity_work);
-			fsverity_enqueue_verify_work(&ctx->work);
-			return;
-		}
-		ctx->cur_step++;
-		/* fall-through */
-	default:
-		__read_end_io(ctx->bio);
+
+	if (ctx->enabled_steps & (1 << STEP_DECRYPT) ||
+		ctx->enabled_steps & (1 << STEP_DECOMPRESS)) {
+		INIT_WORK(&ctx->work, f2fs_post_read_work);
+		f2fs_enqueue_post_read_work(ctx->sbi, &ctx->work);
+		return;
+	}
+
+	if (ctx->enabled_steps & (1 << STEP_VERITY)) {
+		INIT_WORK(&ctx->work, f2fs_verity_work);
+		fsverity_enqueue_verify_work(&ctx->work);
+		return;
 	}
+
+	__f2fs_read_end_io(ctx->bio, false, false);
 }
 
 static bool f2fs_bio_post_read_required(struct bio *bio)
 {
-	return bio->bi_private && !bio->bi_status;
+	return bio->bi_private;
 }
 
 static void f2fs_read_end_io(struct bio *bio)
 {
-	if (time_to_inject(F2FS_P_SB(bio_first_page_all(bio)),
-						FAULT_READ_IO)) {
-		f2fs_show_injection_info(FAULT_READ_IO);
+	struct f2fs_sb_info *sbi = F2FS_P_SB(bio_first_page_all(bio));
+
+	if (time_to_inject(sbi, FAULT_READ_IO)) {
+		f2fs_show_injection_info(sbi, FAULT_READ_IO);
 		bio->bi_status = BLK_STS_IOERR;
 	}
 
 	if (f2fs_bio_post_read_required(bio)) {
 		struct bio_post_read_ctx *ctx = bio->bi_private;
 
-		ctx->cur_step = STEP_INITIAL;
 		bio_post_read_processing(ctx);
 		return;
 	}
 
-	__read_end_io(bio);
+	__f2fs_read_end_io(bio, false, false);
 }
 
 static void f2fs_write_end_io(struct bio *bio)
@@ -191,7 +338,7 @@ static void f2fs_write_end_io(struct bio *bio)
 	struct bvec_iter_all iter_all;
 
 	if (time_to_inject(sbi, FAULT_WRITE_IO)) {
-		f2fs_show_injection_info(FAULT_WRITE_IO);
+		f2fs_show_injection_info(sbi, FAULT_WRITE_IO);
 		bio->bi_status = BLK_STS_IOERR;
 	}
 
@@ -212,6 +359,13 @@ static void f2fs_write_end_io(struct bio *bio)
 
 		fscrypt_finalize_bounce_page(&page);
 
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+		if (f2fs_is_compressed_page(page)) {
+			f2fs_compress_write_end_io(bio, page);
+			continue;
+		}
+#endif
+
 		if (unlikely(bio->bi_status)) {
 			mapping_set_error(page->mapping, -EIO);
 			if (type == F2FS_WB_CP_DATA)
@@ -234,9 +388,6 @@ static void f2fs_write_end_io(struct bio *bio)
 	bio_put(bio);
 }
 
-/*
- * Return true, if pre_bio's bdev is same as its target device.
- */
 struct block_device *f2fs_target_device(struct f2fs_sb_info *sbi,
 				block_t blk_addr, struct bio *bio)
 {
@@ -273,6 +424,9 @@ int f2fs_target_device_index(struct f2fs_sb_info *sbi, block_t blkaddr)
 	return 0;
 }
 
+/*
+ * Return true, if pre_bio's bdev is same as its target device.
+ */
 static bool __same_bdev(struct f2fs_sb_info *sbi,
 				block_t blk_addr, struct bio *bio)
 {
@@ -280,9 +434,6 @@ static bool __same_bdev(struct f2fs_sb_info *sbi,
 	return bio->bi_disk == b->bd_disk && bio->bi_partno == b->bd_partno;
 }
 
-/*
- * Low-level block read/write IO operations.
- */
 static struct bio *__bio_alloc(struct f2fs_io_info *fio, int npages)
 {
 	struct f2fs_sb_info *sbi = fio->sbi;
@@ -315,7 +466,7 @@ static inline void __submit_bio(struct f2fs_sb_info *sbi,
 		if (type != DATA && type != NODE)
 			goto submit_io;
 
-		if (test_opt(sbi, LFS) && current->plug)
+		if (f2fs_lfs_mode(sbi) && current->plug)
 			blk_finish_plug(current->plug);
 
 		if (F2FS_IO_ALIGNED(sbi))
@@ -356,6 +507,12 @@ submit_io:
 	submit_bio(bio);
 }
 
+void f2fs_submit_bio(struct f2fs_sb_info *sbi,
+				struct bio *bio, enum page_type type)
+{
+	__submit_bio(sbi, bio, type);
+}
+
 static void __submit_merged_bio(struct f2fs_bio_info *io)
 {
 	struct f2fs_io_info *fio = &io->fio;
@@ -378,7 +535,6 @@ static bool __has_merged_page(struct bio *bio, struct inode *inode,
 						struct page *page, nid_t ino)
 {
 	struct bio_vec *bvec;
-	struct page *target;
 	struct bvec_iter_all iter_all;
 
 	if (!bio)
@@ -388,10 +544,18 @@ static bool __has_merged_page(struct bio *bio, struct inode *inode,
 		return true;
 
 	bio_for_each_segment_all(bvec, bio, iter_all) {
+		struct page *target = bvec->bv_page;
 
-		target = bvec->bv_page;
-		if (fscrypt_is_bounce_page(target))
+		if (fscrypt_is_bounce_page(target)) {
 			target = fscrypt_pagecache_page(target);
+			if (IS_ERR(target))
+				continue;
+		}
+		if (f2fs_is_compressed_page(target)) {
+			target = f2fs_compress_control_page(target);
+			if (IS_ERR(target))
+				continue;
+		}
 
 		if (inode && inode == target->mapping->host)
 			return true;
@@ -543,6 +707,127 @@ static bool io_is_mergeable(struct f2fs_sb_info *sbi, struct bio *bio,
 	return io_type_is_mergeable(io, fio);
 }
 
+static void add_bio_entry(struct f2fs_sb_info *sbi, struct bio *bio,
+				struct page *page, enum temp_type temp)
+{
+	struct f2fs_bio_info *io = sbi->write_io[DATA] + temp;
+	struct bio_entry *be;
+
+	be = f2fs_kmem_cache_alloc(bio_entry_slab, GFP_NOFS);
+	be->bio = bio;
+	bio_get(bio);
+
+	if (bio_add_page(bio, page, PAGE_SIZE, 0) != PAGE_SIZE)
+		f2fs_bug_on(sbi, 1);
+
+	down_write(&io->bio_list_lock);
+	list_add_tail(&be->list, &io->bio_list);
+	up_write(&io->bio_list_lock);
+}
+
+static void del_bio_entry(struct bio_entry *be)
+{
+	list_del(&be->list);
+	kmem_cache_free(bio_entry_slab, be);
+}
+
+static int add_ipu_page(struct f2fs_sb_info *sbi, struct bio **bio,
+							struct page *page)
+{
+	enum temp_type temp;
+	bool found = false;
+	int ret = -EAGAIN;
+
+	for (temp = HOT; temp < NR_TEMP_TYPE && !found; temp++) {
+		struct f2fs_bio_info *io = sbi->write_io[DATA] + temp;
+		struct list_head *head = &io->bio_list;
+		struct bio_entry *be;
+
+		down_write(&io->bio_list_lock);
+		list_for_each_entry(be, head, list) {
+			if (be->bio != *bio)
+				continue;
+
+			found = true;
+
+			if (bio_add_page(*bio, page, PAGE_SIZE, 0) ==
+							PAGE_SIZE) {
+				ret = 0;
+				break;
+			}
+
+			/* bio is full */
+			del_bio_entry(be);
+			__submit_bio(sbi, *bio, DATA);
+			break;
+		}
+		up_write(&io->bio_list_lock);
+	}
+
+	if (ret) {
+		bio_put(*bio);
+		*bio = NULL;
+	}
+
+	return ret;
+}
+
+void f2fs_submit_merged_ipu_write(struct f2fs_sb_info *sbi,
+					struct bio **bio, struct page *page)
+{
+	enum temp_type temp;
+	bool found = false;
+	struct bio *target = bio ? *bio : NULL;
+
+	for (temp = HOT; temp < NR_TEMP_TYPE && !found; temp++) {
+		struct f2fs_bio_info *io = sbi->write_io[DATA] + temp;
+		struct list_head *head = &io->bio_list;
+		struct bio_entry *be;
+
+		if (list_empty(head))
+			continue;
+
+		down_read(&io->bio_list_lock);
+		list_for_each_entry(be, head, list) {
+			if (target)
+				found = (target == be->bio);
+			else
+				found = __has_merged_page(be->bio, NULL,
+								page, 0);
+			if (found)
+				break;
+		}
+		up_read(&io->bio_list_lock);
+
+		if (!found)
+			continue;
+
+		found = false;
+
+		down_write(&io->bio_list_lock);
+		list_for_each_entry(be, head, list) {
+			if (target)
+				found = (target == be->bio);
+			else
+				found = __has_merged_page(be->bio, NULL,
+								page, 0);
+			if (found) {
+				target = be->bio;
+				del_bio_entry(be);
+				break;
+			}
+		}
+		up_write(&io->bio_list_lock);
+	}
+
+	if (found)
+		__submit_bio(sbi, target, DATA);
+	if (bio && *bio) {
+		bio_put(*bio);
+		*bio = NULL;
+	}
+}
+
 int f2fs_merge_page_bio(struct f2fs_io_info *fio)
 {
 	struct bio *bio = *fio->bio;
@@ -557,20 +842,17 @@ int f2fs_merge_page_bio(struct f2fs_io_info *fio)
 	f2fs_trace_ios(fio, 0);
 
 	if (bio && !page_is_mergeable(fio->sbi, bio, *fio->last_block,
-						fio->new_blkaddr)) {
-		__submit_bio(fio->sbi, bio, fio->type);
-		bio = NULL;
-	}
+						fio->new_blkaddr))
+		f2fs_submit_merged_ipu_write(fio->sbi, &bio, NULL);
 alloc_new:
 	if (!bio) {
 		bio = __bio_alloc(fio, BIO_MAX_PAGES);
 		bio_set_op_attrs(bio, fio->op, fio->op_flags);
-	}
 
-	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
-		__submit_bio(fio->sbi, bio, fio->type);
-		bio = NULL;
-		goto alloc_new;
+		add_bio_entry(fio->sbi, bio, page, fio->temp);
+	} else {
+		if (add_ipu_page(fio->sbi, &bio, page))
+			goto alloc_new;
 	}
 
 	if (fio->io_wbc)
@@ -584,19 +866,6 @@ alloc_new:
 	return 0;
 }
 
-static void f2fs_submit_ipu_bio(struct f2fs_sb_info *sbi, struct bio **bio,
-							struct page *page)
-{
-	if (!bio)
-		return;
-
-	if (!__has_merged_page(*bio, NULL, page, 0))
-		return;
-
-	__submit_bio(sbi, *bio, DATA);
-	*bio = NULL;
-}
-
 void f2fs_submit_page_write(struct f2fs_io_info *fio)
 {
 	struct f2fs_sb_info *sbi = fio->sbi;
@@ -622,7 +891,12 @@ next:
 
 	verify_fio_blkaddr(fio);
 
-	bio_page = fio->encrypted_page ? fio->encrypted_page : fio->page;
+	if (fio->encrypted_page)
+		bio_page = fio->encrypted_page;
+	else if (fio->compressed_page)
+		bio_page = fio->compressed_page;
+	else
+		bio_page = fio->page;
 
 	/* set submitted = true as a return value */
 	fio->submitted = true;
@@ -675,14 +949,15 @@ static inline bool f2fs_need_verity(const struct inode *inode, pgoff_t idx)
 
 static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 				      unsigned nr_pages, unsigned op_flag,
-				      pgoff_t first_idx)
+				      pgoff_t first_idx, bool for_write)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct bio *bio;
 	struct bio_post_read_ctx *ctx;
 	unsigned int post_read_steps = 0;
 
-	bio = f2fs_bio_alloc(sbi, min_t(int, nr_pages, BIO_MAX_PAGES), false);
+	bio = f2fs_bio_alloc(sbi, min_t(int, nr_pages, BIO_MAX_PAGES),
+								for_write);
 	if (!bio)
 		return ERR_PTR(-ENOMEM);
 	f2fs_target_device(sbi, blkaddr, bio);
@@ -691,17 +966,16 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 
 	if (f2fs_encrypted_file(inode))
 		post_read_steps |= 1 << STEP_DECRYPT;
-
+	if (f2fs_compressed_file(inode))
+		post_read_steps |= 1 << STEP_DECOMPRESS;
 	if (f2fs_need_verity(inode, first_idx))
 		post_read_steps |= 1 << STEP_VERITY;
 
 	if (post_read_steps) {
+		/* Due to the mempool, this never fails. */
 		ctx = mempool_alloc(bio_post_read_ctx_pool, GFP_NOFS);
-		if (!ctx) {
-			bio_put(bio);
-			return ERR_PTR(-ENOMEM);
-		}
 		ctx->bio = bio;
+		ctx->sbi = sbi;
 		ctx->enabled_steps = post_read_steps;
 		bio->bi_private = ctx;
 	}
@@ -709,14 +983,21 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
 	return bio;
 }
 
+static void f2fs_release_read_bio(struct bio *bio)
+{
+	if (bio->bi_private)
+		mempool_free(bio->bi_private, bio_post_read_ctx_pool);
+	bio_put(bio);
+}
+
 /* This can handle encryption stuffs */
 static int f2fs_submit_page_read(struct inode *inode, struct page *page,
-							block_t blkaddr)
+						block_t blkaddr, bool for_write)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct bio *bio;
 
-	bio = f2fs_grab_read_bio(inode, blkaddr, 1, 0, page->index);
+	bio = f2fs_grab_read_bio(inode, blkaddr, 1, 0, page->index, for_write);
 	if (IS_ERR(bio))
 		return PTR_ERR(bio);
 
@@ -788,8 +1069,7 @@ int f2fs_reserve_new_blocks(struct dnode_of_data *dn, blkcnt_t count)
 	f2fs_wait_on_page_writeback(dn->node_page, NODE, true, true);
 
 	for (; count > 0; dn->ofs_in_node++) {
-		block_t blkaddr = datablock_addr(dn->inode,
-					dn->node_page, dn->ofs_in_node);
+		block_t blkaddr = f2fs_data_blkaddr(dn);
 		if (blkaddr == NULL_ADDR) {
 			dn->data_blkaddr = NEW_ADDR;
 			__set_data_blkaddr(dn);
@@ -903,7 +1183,7 @@ got_it:
 		return page;
 	}
 
-	err = f2fs_submit_page_read(inode, page, dn.data_blkaddr);
+	err = f2fs_submit_page_read(inode, page, dn.data_blkaddr, for_write);
 	if (err)
 		goto put_err;
 	return page;
@@ -1041,8 +1321,7 @@ static int __allocate_data_block(struct dnode_of_data *dn, int seg_type)
 	if (err)
 		return err;
 
-	dn->data_blkaddr = datablock_addr(dn->inode,
-				dn->node_page, dn->ofs_in_node);
+	dn->data_blkaddr = f2fs_data_blkaddr(dn);
 	if (dn->data_blkaddr != NULL_ADDR)
 		goto alloc;
 
@@ -1074,19 +1353,6 @@ int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *from)
 	int err = 0;
 	bool direct_io = iocb->ki_flags & IOCB_DIRECT;
 
-	/* convert inline data for Direct I/O*/
-	if (direct_io) {
-		err = f2fs_convert_inline_inode(inode);
-		if (err)
-			return err;
-	}
-
-	if (direct_io && allow_outplace_dio(inode, iocb, from))
-		return 0;
-
-	if (is_inode_flag_set(inode, FI_NO_PREALLOC))
-		return 0;
-
 	map.m_lblk = F2FS_BLK_ALIGN(iocb->ki_pos);
 	map.m_len = F2FS_BYTES_TO_BLK(iocb->ki_pos + iov_iter_count(from));
 	if (map.m_len > map.m_lblk)
@@ -1142,13 +1408,9 @@ void __do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock)
 }
 
 /*
- * f2fs_map_blocks() now supported readahead/bmap/rw direct_IO with
- * f2fs_map_blocks structure.
- * If original data blocks are allocated, then give them to blockdev.
- * Otherwise,
- *     a. preallocate requested block addresses
- *     b. do not use extent cache for better performance
- *     c. give the block addresses to blockdev
+ * f2fs_map_blocks() tries to find or build mapping relationship which
+ * maps continuous logical blocks to physical blocks, and return such
+ * info via f2fs_map_blocks structure.
  */
 int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
 						int create, int flag)
@@ -1176,7 +1438,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
 	end = pgofs + maxblocks;
 
 	if (!create && f2fs_lookup_extent_cache(inode, pgofs, &ei)) {
-		if (test_opt(sbi, LFS) && flag == F2FS_GET_BLOCK_DIO &&
+		if (f2fs_lfs_mode(sbi) && flag == F2FS_GET_BLOCK_DIO &&
 							map->m_may_create)
 			goto next_dnode;
 
@@ -1221,7 +1483,7 @@ next_dnode:
 	end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
 
 next_block:
-	blkaddr = datablock_addr(dn.inode, dn.node_page, dn.ofs_in_node);
+	blkaddr = f2fs_data_blkaddr(&dn);
 
 	if (__is_valid_data_blkaddr(blkaddr) &&
 		!f2fs_is_valid_blkaddr(sbi, blkaddr, DATA_GENERIC_ENHANCE)) {
@@ -1231,7 +1493,7 @@ next_block:
 
 	if (__is_valid_data_blkaddr(blkaddr)) {
 		/* use out-place-update for driect IO under LFS mode */
-		if (test_opt(sbi, LFS) && flag == F2FS_GET_BLOCK_DIO &&
+		if (f2fs_lfs_mode(sbi) && flag == F2FS_GET_BLOCK_DIO &&
 							map->m_may_create) {
 			err = __allocate_data_block(&dn, map->m_seg_type);
 			if (err)
@@ -1734,7 +1996,8 @@ submit_and_realloc:
 	}
 	if (bio == NULL) {
 		bio = f2fs_grab_read_bio(inode, block_nr, nr_pages,
-				is_readahead ? REQ_RAHEAD : 0, page->index);
+				is_readahead ? REQ_RAHEAD : 0, page->index,
+				false);
 		if (IS_ERR(bio)) {
 			ret = PTR_ERR(bio);
 			bio = NULL;
@@ -1766,6 +2029,145 @@ out:
 	return ret;
 }
 
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret,
+				unsigned nr_pages, sector_t *last_block_in_bio,
+				bool is_readahead, bool for_write)
+{
+	struct dnode_of_data dn;
+	struct inode *inode = cc->inode;
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct bio *bio = *bio_ret;
+	unsigned int start_idx = cc->cluster_idx << cc->log_cluster_size;
+	sector_t last_block_in_file;
+	const unsigned blkbits = inode->i_blkbits;
+	const unsigned blocksize = 1 << blkbits;
+	struct decompress_io_ctx *dic = NULL;
+	int i;
+	int ret = 0;
+
+	f2fs_bug_on(sbi, f2fs_cluster_is_empty(cc));
+
+	last_block_in_file = (f2fs_readpage_limit(inode) +
+					blocksize - 1) >> blkbits;
+
+	/* get rid of pages beyond EOF */
+	for (i = 0; i < cc->cluster_size; i++) {
+		struct page *page = cc->rpages[i];
+
+		if (!page)
+			continue;
+		if ((sector_t)page->index >= last_block_in_file) {
+			zero_user_segment(page, 0, PAGE_SIZE);
+			if (!PageUptodate(page))
+				SetPageUptodate(page);
+		} else if (!PageUptodate(page)) {
+			continue;
+		}
+		unlock_page(page);
+		cc->rpages[i] = NULL;
+		cc->nr_rpages--;
+	}
+
+	/* we are done since all pages are beyond EOF */
+	if (f2fs_cluster_is_empty(cc))
+		goto out;
+
+	set_new_dnode(&dn, inode, NULL, NULL, 0);
+	ret = f2fs_get_dnode_of_data(&dn, start_idx, LOOKUP_NODE);
+	if (ret)
+		goto out;
+
+	/* cluster was overwritten as normal cluster */
+	if (dn.data_blkaddr != COMPRESS_ADDR)
+		goto out;
+
+	for (i = 1; i < cc->cluster_size; i++) {
+		block_t blkaddr;
+
+		blkaddr = data_blkaddr(dn.inode, dn.node_page,
+						dn.ofs_in_node + i);
+
+		if (!__is_valid_data_blkaddr(blkaddr))
+			break;
+
+		if (!f2fs_is_valid_blkaddr(sbi, blkaddr, DATA_GENERIC)) {
+			ret = -EFAULT;
+			goto out_put_dnode;
+		}
+		cc->nr_cpages++;
+	}
+
+	/* nothing to decompress */
+	if (cc->nr_cpages == 0) {
+		ret = 0;
+		goto out_put_dnode;
+	}
+
+	dic = f2fs_alloc_dic(cc);
+	if (IS_ERR(dic)) {
+		ret = PTR_ERR(dic);
+		goto out_put_dnode;
+	}
+
+	for (i = 0; i < dic->nr_cpages; i++) {
+		struct page *page = dic->cpages[i];
+		block_t blkaddr;
+
+		blkaddr = data_blkaddr(dn.inode, dn.node_page,
+						dn.ofs_in_node + i + 1);
+
+		if (bio && !page_is_mergeable(sbi, bio,
+					*last_block_in_bio, blkaddr)) {
+submit_and_realloc:
+			__submit_bio(sbi, bio, DATA);
+			bio = NULL;
+		}
+
+		if (!bio) {
+			bio = f2fs_grab_read_bio(inode, blkaddr, nr_pages,
+					is_readahead ? REQ_RAHEAD : 0,
+					page->index, for_write);
+			if (IS_ERR(bio)) {
+				ret = PTR_ERR(bio);
+				bio = NULL;
+				dic->failed = true;
+				if (refcount_sub_and_test(dic->nr_cpages - i,
+							&dic->ref))
+					f2fs_decompress_end_io(dic->rpages,
+							cc->cluster_size, true,
+							false);
+				f2fs_free_dic(dic);
+				f2fs_put_dnode(&dn);
+				*bio_ret = bio;
+				return ret;
+			}
+		}
+
+		f2fs_wait_on_block_writeback(inode, blkaddr);
+
+		if (bio_add_page(bio, page, blocksize, 0) < blocksize)
+			goto submit_and_realloc;
+
+		inc_page_count(sbi, F2FS_RD_DATA);
+		ClearPageError(page);
+		*last_block_in_bio = blkaddr;
+	}
+
+	f2fs_put_dnode(&dn);
+
+	*bio_ret = bio;
+	return 0;
+
+out_put_dnode:
+	f2fs_put_dnode(&dn);
+out:
+	f2fs_decompress_end_io(cc->rpages, cc->cluster_size, true, false);
+	*bio_ret = bio;
+	return ret;
+}
+#endif
+
 /*
  * This function was originally taken from fs/mpage.c, and customized for f2fs.
  * Major change was from block_size == page_size in f2fs by default.
@@ -1775,14 +2177,26 @@ out:
  * use ->readpage() or do the necessary surgery to decouple ->readpages()
  * from read-ahead.
  */
-static int f2fs_mpage_readpages(struct address_space *mapping,
-			struct list_head *pages, struct page *page,
-			unsigned nr_pages, bool is_readahead)
+static int f2fs_mpage_readpages(struct inode *inode,
+		struct readahead_control *rac, struct page *page)
 {
 	struct bio *bio = NULL;
 	sector_t last_block_in_bio = 0;
-	struct inode *inode = mapping->host;
 	struct f2fs_map_blocks map;
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	struct compress_ctx cc = {
+		.inode = inode,
+		.log_cluster_size = F2FS_I(inode)->i_log_cluster_size,
+		.cluster_size = F2FS_I(inode)->i_cluster_size,
+		.cluster_idx = NULL_CLUSTER,
+		.rpages = NULL,
+		.cpages = NULL,
+		.nr_rpages = 0,
+		.nr_cpages = 0,
+	};
+#endif
+	unsigned nr_pages = rac ? readahead_count(rac) : 1;
+	unsigned max_nr_pages = nr_pages;
 	int ret = 0;
 
 	map.m_pblk = 0;
@@ -1795,32 +2209,72 @@ static int f2fs_mpage_readpages(struct address_space *mapping,
 	map.m_may_create = false;
 
 	for (; nr_pages; nr_pages--) {
-		if (pages) {
-			page = list_last_entry(pages, struct page, lru);
-
+		if (rac) {
+			page = readahead_page(rac);
 			prefetchw(&page->flags);
-			list_del(&page->lru);
-			if (add_to_page_cache_lru(page, mapping,
-						  page_index(page),
-						  readahead_gfp_mask(mapping)))
-				goto next_page;
 		}
 
-		ret = f2fs_read_single_page(inode, page, nr_pages, &map, &bio,
-					&last_block_in_bio, is_readahead);
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+		if (f2fs_compressed_file(inode)) {
+			/* there are remained comressed pages, submit them */
+			if (!f2fs_cluster_can_merge_page(&cc, page->index)) {
+				ret = f2fs_read_multi_pages(&cc, &bio,
+							max_nr_pages,
+							&last_block_in_bio,
+							rac != NULL, false);
+				f2fs_destroy_compress_ctx(&cc);
+				if (ret)
+					goto set_error_page;
+			}
+			ret = f2fs_is_compressed_cluster(inode, page->index);
+			if (ret < 0)
+				goto set_error_page;
+			else if (!ret)
+				goto read_single_page;
+
+			ret = f2fs_init_compress_ctx(&cc);
+			if (ret)
+				goto set_error_page;
+
+			f2fs_compress_ctx_add_page(&cc, page);
+
+			goto next_page;
+		}
+read_single_page:
+#endif
+
+		ret = f2fs_read_single_page(inode, page, max_nr_pages, &map,
+					&bio, &last_block_in_bio, rac);
 		if (ret) {
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+set_error_page:
+#endif
 			SetPageError(page);
 			zero_user_segment(page, 0, PAGE_SIZE);
 			unlock_page(page);
 		}
+#ifdef CONFIG_F2FS_FS_COMPRESSION
 next_page:
-		if (pages)
+#endif
+		if (rac)
 			put_page(page);
+
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+		if (f2fs_compressed_file(inode)) {
+			/* last page */
+			if (nr_pages == 1 && !f2fs_cluster_is_empty(&cc)) {
+				ret = f2fs_read_multi_pages(&cc, &bio,
+							max_nr_pages,
+							&last_block_in_bio,
+							rac != NULL, false);
+				f2fs_destroy_compress_ctx(&cc);
+			}
+		}
+#endif
 	}
-	BUG_ON(pages && !list_empty(pages));
 	if (bio)
 		__submit_bio(F2FS_I_SB(inode), bio, DATA);
-	return pages ? 0 : ret;
+	return ret;
 }
 
 static int f2fs_read_data_page(struct file *file, struct page *page)
@@ -1830,52 +2284,57 @@ static int f2fs_read_data_page(struct file *file, struct page *page)
 
 	trace_f2fs_readpage(page, DATA);
 
+	if (!f2fs_is_compress_backend_ready(inode)) {
+		unlock_page(page);
+		return -EOPNOTSUPP;
+	}
+
 	/* If the file has inline data, try to read it directly */
 	if (f2fs_has_inline_data(inode))
 		ret = f2fs_read_inline_data(inode, page);
 	if (ret == -EAGAIN)
-		ret = f2fs_mpage_readpages(page_file_mapping(page),
-						NULL, page, 1, false);
+		ret = f2fs_mpage_readpages(inode, NULL, page);
 	return ret;
 }
 
-static int f2fs_read_data_pages(struct file *file,
-			struct address_space *mapping,
-			struct list_head *pages, unsigned nr_pages)
+static void f2fs_readahead(struct readahead_control *rac)
 {
-	struct inode *inode = mapping->host;
-	struct page *page = list_last_entry(pages, struct page, lru);
+	struct inode *inode = rac->mapping->host;
 
-	trace_f2fs_readpages(inode, page, nr_pages);
+	trace_f2fs_readpages(inode, readahead_index(rac), readahead_count(rac));
+
+	if (!f2fs_is_compress_backend_ready(inode))
+		return;
 
 	/* If the file has inline data, skip readpages */
 	if (f2fs_has_inline_data(inode))
-		return 0;
+		return;
 
-	return f2fs_mpage_readpages(mapping, pages, NULL, nr_pages, true);
+	f2fs_mpage_readpages(inode, rac, NULL);
 }
 
-static int encrypt_one_page(struct f2fs_io_info *fio)
+int f2fs_encrypt_one_page(struct f2fs_io_info *fio)
 {
 	struct inode *inode = fio->page->mapping->host;
-	struct page *mpage;
+	struct page *mpage, *page;
 	gfp_t gfp_flags = GFP_NOFS;
 
 	if (!f2fs_encrypted_file(inode))
 		return 0;
 
+	page = fio->compressed_page ? fio->compressed_page : fio->page;
+
 	/* wait for GCed page writeback via META_MAPPING */
 	f2fs_wait_on_block_writeback(inode, fio->old_blkaddr);
 
 retry_encrypt:
-	fio->encrypted_page = fscrypt_encrypt_pagecache_blocks(fio->page,
-							       PAGE_SIZE, 0,
-							       gfp_flags);
+	fio->encrypted_page = fscrypt_encrypt_pagecache_blocks(page,
+					PAGE_SIZE, 0, gfp_flags);
 	if (IS_ERR(fio->encrypted_page)) {
 		/* flush pending IOs and wait for a while in the ENOMEM case */
 		if (PTR_ERR(fio->encrypted_page) == -ENOMEM) {
 			f2fs_flush_merged_writes(fio->sbi);
-			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT);
 			gfp_flags |= __GFP_NOFAIL;
 			goto retry_encrypt;
 		}
@@ -1946,7 +2405,7 @@ bool f2fs_should_update_outplace(struct inode *inode, struct f2fs_io_info *fio)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 
-	if (test_opt(sbi, LFS))
+	if (f2fs_lfs_mode(sbi))
 		return true;
 	if (S_ISDIR(inode->i_mode))
 		return true;
@@ -2030,7 +2489,7 @@ got_it:
 	if (ipu_force ||
 		(__is_valid_data_blkaddr(fio->old_blkaddr) &&
 					need_inplace_update(fio))) {
-		err = encrypt_one_page(fio);
+		err = f2fs_encrypt_one_page(fio);
 		if (err)
 			goto out_writepage;
 
@@ -2066,13 +2525,16 @@ got_it:
 
 	fio->version = ni.version;
 
-	err = encrypt_one_page(fio);
+	err = f2fs_encrypt_one_page(fio);
 	if (err)
 		goto out_writepage;
 
 	set_page_writeback(page);
 	ClearPageError(page);
 
+	if (fio->compr_blocks && fio->old_blkaddr == COMPRESS_ADDR)
+		f2fs_i_compr_blocks_update(inode, fio->compr_blocks - 1, false);
+
 	/* LFS mode write path */
 	f2fs_outplace_write_data(&dn, fio);
 	trace_f2fs_do_write_data_page(page, OPU);
@@ -2087,18 +2549,19 @@ out:
 	return err;
 }
 
-static int __write_data_page(struct page *page, bool *submitted,
+int f2fs_write_single_data_page(struct page *page, int *submitted,
 				struct bio **bio,
 				sector_t *last_block,
 				struct writeback_control *wbc,
-				enum iostat_type io_type)
+				enum iostat_type io_type,
+				int compr_blocks)
 {
 	struct inode *inode = page->mapping->host;
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	loff_t i_size = i_size_read(inode);
-	const pgoff_t end_index = ((unsigned long long) i_size)
+	const pgoff_t end_index = ((unsigned long long)i_size)
 							>> PAGE_SHIFT;
-	loff_t psize = (page->index + 1) << PAGE_SHIFT;
+	loff_t psize = (loff_t)(page->index + 1) << PAGE_SHIFT;
 	unsigned offset = 0;
 	bool need_balance_fs = false;
 	int err = 0;
@@ -2112,6 +2575,7 @@ static int __write_data_page(struct page *page, bool *submitted,
 		.page = page,
 		.encrypted_page = NULL,
 		.submitted = false,
+		.compr_blocks = compr_blocks,
 		.need_lock = LOCK_RETRY,
 		.io_type = io_type,
 		.io_wbc = wbc,
@@ -2136,7 +2600,9 @@ static int __write_data_page(struct page *page, bool *submitted,
 	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
 		goto redirty_out;
 
-	if (page->index < end_index || f2fs_verity_in_progress(inode))
+	if (page->index < end_index ||
+			f2fs_verity_in_progress(inode) ||
+			compr_blocks)
 		goto write;
 
 	/*
@@ -2189,10 +2655,10 @@ write:
 	if (err) {
 		file_set_keep_isize(inode);
 	} else {
-		down_write(&F2FS_I(inode)->i_sem);
+		spin_lock(&F2FS_I(inode)->i_size_lock);
 		if (F2FS_I(inode)->last_disk_size < psize)
 			F2FS_I(inode)->last_disk_size = psize;
-		up_write(&F2FS_I(inode)->i_sem);
+		spin_unlock(&F2FS_I(inode)->i_size_lock);
 	}
 
 done:
@@ -2212,22 +2678,19 @@ out:
 		f2fs_remove_dirty_inode(inode);
 		submitted = NULL;
 	}
-
 	unlock_page(page);
 	if (!S_ISDIR(inode->i_mode) && !IS_NOQUOTA(inode) &&
-					!F2FS_I(inode)->cp_task) {
-		f2fs_submit_ipu_bio(sbi, bio, page);
+					!F2FS_I(inode)->cp_task)
 		f2fs_balance_fs(sbi, need_balance_fs);
-	}
 
 	if (unlikely(f2fs_cp_error(sbi))) {
-		f2fs_submit_ipu_bio(sbi, bio, page);
 		f2fs_submit_merged_write(sbi, DATA);
+		f2fs_submit_merged_ipu_write(sbi, bio, NULL);
 		submitted = NULL;
 	}
 
 	if (submitted)
-		*submitted = fio.submitted;
+		*submitted = fio.submitted ? 1 : 0;
 
 	return 0;
 
@@ -2248,7 +2711,23 @@ redirty_out:
 static int f2fs_write_data_page(struct page *page,
 					struct writeback_control *wbc)
 {
-	return __write_data_page(page, NULL, NULL, NULL, wbc, FS_DATA_IO);
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	struct inode *inode = page->mapping->host;
+
+	if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
+		goto out;
+
+	if (f2fs_compressed_file(inode)) {
+		if (f2fs_is_compressed_cluster(inode, page->index)) {
+			redirty_page_for_writepage(wbc, page);
+			return AOP_WRITEPAGE_ACTIVATE;
+		}
+	}
+out:
+#endif
+
+	return f2fs_write_single_data_page(page, NULL, NULL, NULL,
+						wbc, FS_DATA_IO, 0);
 }
 
 /*
@@ -2261,11 +2740,27 @@ static int f2fs_write_cache_pages(struct address_space *mapping,
 					enum iostat_type io_type)
 {
 	int ret = 0;
-	int done = 0;
+	int done = 0, retry = 0;
 	struct pagevec pvec;
 	struct f2fs_sb_info *sbi = F2FS_M_SB(mapping);
 	struct bio *bio = NULL;
 	sector_t last_block;
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	struct inode *inode = mapping->host;
+	struct compress_ctx cc = {
+		.inode = inode,
+		.log_cluster_size = F2FS_I(inode)->i_log_cluster_size,
+		.cluster_size = F2FS_I(inode)->i_cluster_size,
+		.cluster_idx = NULL_CLUSTER,
+		.rpages = NULL,
+		.nr_rpages = 0,
+		.cpages = NULL,
+		.rbuf = NULL,
+		.cbuf = NULL,
+		.rlen = PAGE_SIZE * F2FS_I(inode)->i_cluster_size,
+		.private = NULL,
+	};
+#endif
 	int nr_pages;
 	pgoff_t uninitialized_var(writeback_index);
 	pgoff_t index;
@@ -2275,6 +2770,8 @@ static int f2fs_write_cache_pages(struct address_space *mapping,
 	int range_whole = 0;
 	xa_mark_t tag;
 	int nwritten = 0;
+	int submitted = 0;
+	int i;
 
 	pagevec_init(&pvec);
 
@@ -2304,12 +2801,11 @@ static int f2fs_write_cache_pages(struct address_space *mapping,
 	else
 		tag = PAGECACHE_TAG_DIRTY;
 retry:
+	retry = 0;
 	if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
 		tag_pages_for_writeback(mapping, index, end);
 	done_index = index;
-	while (!done && (index <= end)) {
-		int i;
-
+	while (!done && !retry && (index <= end)) {
 		nr_pages = pagevec_lookup_range_tag(&pvec, mapping, &index, end,
 				tag);
 		if (nr_pages == 0)
@@ -2317,15 +2813,62 @@ retry:
 
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
-			bool submitted = false;
+			bool need_readd;
+readd:
+			need_readd = false;
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+			if (f2fs_compressed_file(inode)) {
+				ret = f2fs_init_compress_ctx(&cc);
+				if (ret) {
+					done = 1;
+					break;
+				}
+
+				if (!f2fs_cluster_can_merge_page(&cc,
+								page->index)) {
+					ret = f2fs_write_multi_pages(&cc,
+						&submitted, wbc, io_type);
+					if (!ret)
+						need_readd = true;
+					goto result;
+				}
 
+				if (unlikely(f2fs_cp_error(sbi)))
+					goto lock_page;
+
+				if (f2fs_cluster_is_empty(&cc)) {
+					void *fsdata = NULL;
+					struct page *pagep;
+					int ret2;
+
+					ret2 = f2fs_prepare_compress_overwrite(
+							inode, &pagep,
+							page->index, &fsdata);
+					if (ret2 < 0) {
+						ret = ret2;
+						done = 1;
+						break;
+					} else if (ret2 &&
+						!f2fs_compress_write_end(inode,
+								fsdata, page->index,
+								1)) {
+						retry = 1;
+						break;
+					}
+				} else {
+					goto lock_page;
+				}
+			}
+#endif
 			/* give a priority to WB_SYNC threads */
 			if (atomic_read(&sbi->wb_sync_req[DATA]) &&
 					wbc->sync_mode == WB_SYNC_NONE) {
 				done = 1;
 				break;
 			}
-
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+lock_page:
+#endif
 			done_index = page->index;
 retry_write:
 			lock_page(page);
@@ -2342,57 +2885,81 @@ continue_unlock:
 			}
 
 			if (PageWriteback(page)) {
-				if (wbc->sync_mode != WB_SYNC_NONE) {
+				if (wbc->sync_mode != WB_SYNC_NONE)
 					f2fs_wait_on_page_writeback(page,
 							DATA, true, true);
-					f2fs_submit_ipu_bio(sbi, &bio, page);
-				} else {
+				else
 					goto continue_unlock;
-				}
 			}
 
 			if (!clear_page_dirty_for_io(page))
 				goto continue_unlock;
 
-			ret = __write_data_page(page, &submitted, &bio,
-					&last_block, wbc, io_type);
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+			if (f2fs_compressed_file(inode)) {
+				get_page(page);
+				f2fs_compress_ctx_add_page(&cc, page);
+				continue;
+			}
+#endif
+			ret = f2fs_write_single_data_page(page, &submitted,
+					&bio, &last_block, wbc, io_type, 0);
+			if (ret == AOP_WRITEPAGE_ACTIVATE)
+				unlock_page(page);
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+result:
+#endif
+			nwritten += submitted;
+			wbc->nr_to_write -= submitted;
+
 			if (unlikely(ret)) {
 				/*
 				 * keep nr_to_write, since vfs uses this to
 				 * get # of written pages.
 				 */
 				if (ret == AOP_WRITEPAGE_ACTIVATE) {
-					unlock_page(page);
 					ret = 0;
-					continue;
+					goto next;
 				} else if (ret == -EAGAIN) {
 					ret = 0;
 					if (wbc->sync_mode == WB_SYNC_ALL) {
 						cond_resched();
 						congestion_wait(BLK_RW_ASYNC,
-									HZ/50);
+							DEFAULT_IO_TIMEOUT);
 						goto retry_write;
 					}
-					continue;
+					goto next;
 				}
 				done_index = page->index + 1;
 				done = 1;
 				break;
-			} else if (submitted) {
-				nwritten++;
 			}
 
-			if (--wbc->nr_to_write <= 0 &&
+			if (wbc->nr_to_write <= 0 &&
 					wbc->sync_mode == WB_SYNC_NONE) {
 				done = 1;
 				break;
 			}
+next:
+			if (need_readd)
+				goto readd;
 		}
 		pagevec_release(&pvec);
 		cond_resched();
 	}
-
-	if (!cycled && !done) {
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	/* flush remained pages in compress cluster */
+	if (f2fs_compressed_file(inode) && !f2fs_cluster_is_empty(&cc)) {
+		ret = f2fs_write_multi_pages(&cc, &submitted, wbc, io_type);
+		nwritten += submitted;
+		wbc->nr_to_write -= submitted;
+		if (ret) {
+			done = 1;
+			retry = 0;
+		}
+	}
+#endif
+	if ((!cycled && !done) || retry) {
 		cycled = 1;
 		index = 0;
 		end = writeback_index - 1;
@@ -2406,7 +2973,7 @@ continue_unlock:
 								NULL, 0, DATA);
 	/* submit cached bio of IPU write */
 	if (bio)
-		__submit_bio(sbi, bio, DATA);
+		f2fs_submit_merged_ipu_write(sbi, &bio, NULL);
 
 	return ret;
 }
@@ -2414,13 +2981,17 @@ continue_unlock:
 static inline bool __should_serialize_io(struct inode *inode,
 					struct writeback_control *wbc)
 {
+	/* to avoid deadlock in path of data flush */
+	if (F2FS_I(inode)->cp_task)
+		return false;
+
 	if (!S_ISREG(inode->i_mode))
 		return false;
 	if (IS_NOQUOTA(inode))
 		return false;
-	/* to avoid deadlock in path of data flush */
-	if (F2FS_I(inode)->cp_task)
-		return false;
+
+	if (f2fs_compressed_file(inode))
+		return true;
 	if (wbc->sync_mode != WB_SYNC_ALL)
 		return true;
 	if (get_dirty_pages(inode) >= SM_I(F2FS_I_SB(inode))->min_seq_blocks)
@@ -2511,14 +3082,16 @@ static void f2fs_write_failed(struct address_space *mapping, loff_t to)
 	struct inode *inode = mapping->host;
 	loff_t i_size = i_size_read(inode);
 
+	if (IS_NOQUOTA(inode))
+		return;
+
 	/* In the fs-verity case, f2fs_end_enable_verity() does the truncate */
 	if (to > i_size && !f2fs_verity_in_progress(inode)) {
 		down_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]);
 		down_write(&F2FS_I(inode)->i_mmap_sem);
 
 		truncate_pagecache(inode, i_size);
-		if (!IS_NOQUOTA(inode))
-			f2fs_truncate_blocks(inode, i_size, true);
+		f2fs_truncate_blocks(inode, i_size, true);
 
 		up_write(&F2FS_I(inode)->i_mmap_sem);
 		up_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]);
@@ -2558,6 +3131,7 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi,
 		__do_map_lock(sbi, flag, true);
 		locked = true;
 	}
+
 restart:
 	/* check inline_data */
 	ipage = f2fs_get_node_page(sbi, inode->i_ino);
@@ -2648,6 +3222,24 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 		if (err)
 			goto fail;
 	}
+
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	if (f2fs_compressed_file(inode)) {
+		int ret;
+
+		*fsdata = NULL;
+
+		ret = f2fs_prepare_compress_overwrite(inode, pagep,
+							index, fsdata);
+		if (ret < 0) {
+			err = ret;
+			goto fail;
+		} else if (ret) {
+			return 0;
+		}
+	}
+#endif
+
 repeat:
 	/*
 	 * Do not use grab_cache_page_write_begin() to avoid deadlock due to
@@ -2660,6 +3252,8 @@ repeat:
 		goto fail;
 	}
 
+	/* TODO: cluster can be compressed due to race with .writepage */
+
 	*pagep = page;
 
 	err = prepare_write_begin(sbi, page, pos, len,
@@ -2699,7 +3293,7 @@ repeat:
 			err = -EFSCORRUPTED;
 			goto fail;
 		}
-		err = f2fs_submit_page_read(inode, page, blkaddr);
+		err = f2fs_submit_page_read(inode, page, blkaddr, true);
 		if (err)
 			goto fail;
 
@@ -2743,6 +3337,16 @@ static int f2fs_write_end(struct file *file,
 		else
 			SetPageUptodate(page);
 	}
+
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	/* overwrite compressed file */
+	if (f2fs_compressed_file(inode) && fsdata) {
+		f2fs_compress_write_end(inode, fsdata, page->index, copied);
+		f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
+		return copied;
+	}
+#endif
+
 	if (!copied)
 		goto unlock_out;
 
@@ -2870,7 +3474,8 @@ static ssize_t f2fs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 	err = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev,
 			iter, rw == WRITE ? get_data_block_dio_write :
 			get_data_block_dio, NULL, f2fs_dio_submit_bio,
-			DIO_LOCKING | DIO_SKIP_HOLES);
+			rw == WRITE ? DIO_LOCKING | DIO_SKIP_HOLES :
+			DIO_SKIP_HOLES);
 
 	if (do_opu)
 		up_read(&fi->i_gc_rwsem[READ]);
@@ -3043,7 +3648,8 @@ int f2fs_migrate_page(struct address_space *mapping,
 
 #ifdef CONFIG_SWAP
 /* Copied from generic_swapfile_activate() to check any holes */
-static int check_swap_activate(struct file *swap_file, unsigned int max)
+static int check_swap_activate(struct swap_info_struct *sis,
+				struct file *swap_file, sector_t *span)
 {
 	struct address_space *mapping = swap_file->f_mapping;
 	struct inode *inode = mapping->host;
@@ -3054,6 +3660,8 @@ static int check_swap_activate(struct file *swap_file, unsigned int max)
 	sector_t last_block;
 	sector_t lowest_block = -1;
 	sector_t highest_block = 0;
+	int nr_extents = 0;
+	int ret;
 
 	blkbits = inode->i_blkbits;
 	blocks_per_page = PAGE_SIZE >> blkbits;
@@ -3065,15 +3673,20 @@ static int check_swap_activate(struct file *swap_file, unsigned int max)
 	probe_block = 0;
 	page_no = 0;
 	last_block = i_size_read(inode) >> blkbits;
-	while ((probe_block + blocks_per_page) <= last_block && page_no < max) {
+	while ((probe_block + blocks_per_page) <= last_block &&
+			page_no < sis->max) {
 		unsigned block_in_page;
 		sector_t first_block;
+		sector_t block = 0;
+		int	 err = 0;
 
 		cond_resched();
 
-		first_block = bmap(inode, probe_block);
-		if (first_block == 0)
+		block = probe_block;
+		err = bmap(inode, &block);
+		if (err || !block)
 			goto bad_bmap;
+		first_block = block;
 
 		/*
 		 * It must be PAGE_SIZE aligned on-disk
@@ -3085,11 +3698,13 @@ static int check_swap_activate(struct file *swap_file, unsigned int max)
 
 		for (block_in_page = 1; block_in_page < blocks_per_page;
 					block_in_page++) {
-			sector_t block;
 
-			block = bmap(inode, probe_block + block_in_page);
-			if (block == 0)
+			block = probe_block + block_in_page;
+			err = bmap(inode, &block);
+
+			if (err || !block)
 				goto bad_bmap;
+
 			if (block != first_block + block_in_page) {
 				/* Discontiguity */
 				probe_block++;
@@ -3105,13 +3720,27 @@ static int check_swap_activate(struct file *swap_file, unsigned int max)
 				highest_block = first_block;
 		}
 
+		/*
+		 * We found a PAGE_SIZE-length, PAGE_SIZE-aligned run of blocks
+		 */
+		ret = add_swap_extent(sis, page_no, 1, first_block);
+		if (ret < 0)
+			goto out;
+		nr_extents += ret;
 		page_no++;
 		probe_block += blocks_per_page;
 reprobe:
 		continue;
 	}
-	return 0;
-
+	ret = nr_extents;
+	*span = 1 + highest_block - lowest_block;
+	if (page_no == 0)
+		page_no = 1;	/* force Empty message */
+	sis->max = page_no;
+	sis->pages = page_no - 1;
+	sis->highest_bit = page_no - 1;
+out:
+	return ret;
 bad_bmap:
 	pr_err("swapon: swapfile has holes\n");
 	return -EINVAL;
@@ -3133,14 +3762,17 @@ static int f2fs_swap_activate(struct swap_info_struct *sis, struct file *file,
 	if (ret)
 		return ret;
 
-	ret = check_swap_activate(file, sis->max);
-	if (ret)
+	if (f2fs_disable_compressed_file(inode))
+		return -EINVAL;
+
+	ret = check_swap_activate(sis, file, span);
+	if (ret < 0)
 		return ret;
 
 	set_inode_flag(inode, FI_PIN_FILE);
 	f2fs_precache_extents(inode);
 	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
-	return 0;
+	return ret;
 }
 
 static void f2fs_swap_deactivate(struct file *file)
@@ -3163,7 +3795,7 @@ static void f2fs_swap_deactivate(struct file *file)
 
 const struct address_space_operations f2fs_dblock_aops = {
 	.readpage	= f2fs_read_data_page,
-	.readpages	= f2fs_read_data_pages,
+	.readahead	= f2fs_readahead,
 	.writepage	= f2fs_write_data_page,
 	.writepages	= f2fs_write_data_pages,
 	.write_begin	= f2fs_write_begin,
@@ -3211,8 +3843,43 @@ fail:
 	return -ENOMEM;
 }
 
-void __exit f2fs_destroy_post_read_processing(void)
+void f2fs_destroy_post_read_processing(void)
 {
 	mempool_destroy(bio_post_read_ctx_pool);
 	kmem_cache_destroy(bio_post_read_ctx_cache);
 }
+
+int f2fs_init_post_read_wq(struct f2fs_sb_info *sbi)
+{
+	if (!f2fs_sb_has_encrypt(sbi) &&
+		!f2fs_sb_has_verity(sbi) &&
+		!f2fs_sb_has_compression(sbi))
+		return 0;
+
+	sbi->post_read_wq = alloc_workqueue("f2fs_post_read_wq",
+						 WQ_UNBOUND | WQ_HIGHPRI,
+						 num_online_cpus());
+	if (!sbi->post_read_wq)
+		return -ENOMEM;
+	return 0;
+}
+
+void f2fs_destroy_post_read_wq(struct f2fs_sb_info *sbi)
+{
+	if (sbi->post_read_wq)
+		destroy_workqueue(sbi->post_read_wq);
+}
+
+int __init f2fs_init_bio_entry_cache(void)
+{
+	bio_entry_slab = f2fs_kmem_cache_create("f2fs_bio_entry_slab",
+			sizeof(struct bio_entry));
+	if (!bio_entry_slab)
+		return -ENOMEM;
+	return 0;
+}
+
+void f2fs_destroy_bio_entry_cache(void)
+{
+	kmem_cache_destroy(bio_entry_slab);
+}
diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c
index 9b0bedd82581..0dbcb0f9c019 100644
--- a/fs/f2fs/debug.c
+++ b/fs/f2fs/debug.c
@@ -21,9 +21,45 @@
 #include "gc.h"
 
 static LIST_HEAD(f2fs_stat_list);
-static struct dentry *f2fs_debugfs_root;
 static DEFINE_MUTEX(f2fs_stat_mutex);
+#ifdef CONFIG_DEBUG_FS
+static struct dentry *f2fs_debugfs_root;
+#endif
+
+/*
+ * This function calculates BDF of every segments
+ */
+void f2fs_update_sit_info(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_stat_info *si = F2FS_STAT(sbi);
+	unsigned long long blks_per_sec, hblks_per_sec, total_vblocks;
+	unsigned long long bimodal, dist;
+	unsigned int segno, vblocks;
+	int ndirty = 0;
+
+	bimodal = 0;
+	total_vblocks = 0;
+	blks_per_sec = BLKS_PER_SEC(sbi);
+	hblks_per_sec = blks_per_sec / 2;
+	for (segno = 0; segno < MAIN_SEGS(sbi); segno += sbi->segs_per_sec) {
+		vblocks = get_valid_blocks(sbi, segno, true);
+		dist = abs(vblocks - hblks_per_sec);
+		bimodal += dist * dist;
+
+		if (vblocks > 0 && vblocks < blks_per_sec) {
+			total_vblocks += vblocks;
+			ndirty++;
+		}
+	}
+	dist = div_u64(MAIN_SECS(sbi) * hblks_per_sec * hblks_per_sec, 100);
+	si->bimodal = div64_u64(bimodal, dist);
+	if (si->dirty_count)
+		si->avg_vblocks = div_u64(total_vblocks, ndirty);
+	else
+		si->avg_vblocks = 0;
+}
 
+#ifdef CONFIG_DEBUG_FS
 static void update_general_status(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_stat_info *si = F2FS_STAT(sbi);
@@ -56,7 +92,7 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->nquota_files = sbi->nquota_files;
 	si->ndirty_all = sbi->ndirty_inode[DIRTY_META];
 	si->inmem_pages = get_pages(sbi, F2FS_INMEM_PAGES);
-	si->aw_cnt = atomic_read(&sbi->aw_cnt);
+	si->aw_cnt = sbi->atomic_files;
 	si->vw_cnt = atomic_read(&sbi->vw_cnt);
 	si->max_aw_cnt = atomic_read(&sbi->max_aw_cnt);
 	si->max_vw_cnt = atomic_read(&sbi->max_vw_cnt);
@@ -94,6 +130,8 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->inline_xattr = atomic_read(&sbi->inline_xattr);
 	si->inline_inode = atomic_read(&sbi->inline_inode);
 	si->inline_dir = atomic_read(&sbi->inline_dir);
+	si->compr_inode = atomic_read(&sbi->compr_inode);
+	si->compr_blocks = atomic_read(&sbi->compr_blocks);
 	si->append = sbi->im[APPEND_INO].ino_num;
 	si->update = sbi->im[UPDATE_INO].ino_num;
 	si->orphans = sbi->im[ORPHAN_INO].ino_num;
@@ -114,7 +152,6 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->free_nids = NM_I(sbi)->nid_cnt[FREE_NID];
 	si->avail_nids = NM_I(sbi)->available_nids;
 	si->alloc_nids = NM_I(sbi)->nid_cnt[PREALLOC_NID];
-	si->bg_gc = sbi->bg_gc;
 	si->io_skip_bggc = sbi->io_skip_bggc;
 	si->other_skip_bggc = sbi->other_skip_bggc;
 	si->skipped_atomic_files[BG_GC] = sbi->skipped_atomic_files[BG_GC];
@@ -146,39 +183,6 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 }
 
 /*
- * This function calculates BDF of every segments
- */
-static void update_sit_info(struct f2fs_sb_info *sbi)
-{
-	struct f2fs_stat_info *si = F2FS_STAT(sbi);
-	unsigned long long blks_per_sec, hblks_per_sec, total_vblocks;
-	unsigned long long bimodal, dist;
-	unsigned int segno, vblocks;
-	int ndirty = 0;
-
-	bimodal = 0;
-	total_vblocks = 0;
-	blks_per_sec = BLKS_PER_SEC(sbi);
-	hblks_per_sec = blks_per_sec / 2;
-	for (segno = 0; segno < MAIN_SEGS(sbi); segno += sbi->segs_per_sec) {
-		vblocks = get_valid_blocks(sbi, segno, true);
-		dist = abs(vblocks - hblks_per_sec);
-		bimodal += dist * dist;
-
-		if (vblocks > 0 && vblocks < blks_per_sec) {
-			total_vblocks += vblocks;
-			ndirty++;
-		}
-	}
-	dist = div_u64(MAIN_SECS(sbi) * hblks_per_sec * hblks_per_sec, 100);
-	si->bimodal = div64_u64(bimodal, dist);
-	if (si->dirty_count)
-		si->avg_vblocks = div_u64(total_vblocks, ndirty);
-	else
-		si->avg_vblocks = 0;
-}
-
-/*
  * This function calculates memory footprint.
  */
 static void update_mem_info(struct f2fs_sb_info *sbi)
@@ -297,6 +301,9 @@ static int stat_show(struct seq_file *s, void *v)
 			   si->ssa_area_segs, si->main_area_segs);
 		seq_printf(s, "(OverProv:%d Resv:%d)]\n\n",
 			   si->overp_segs, si->rsvd_segs);
+		seq_printf(s, "Current Time Sec: %llu / Mounted Time Sec: %llu\n\n",
+					ktime_get_boottime_seconds(),
+					SIT_I(si->sbi)->mounted_time);
 		if (test_opt(si->sbi, DISCARD))
 			seq_printf(s, "Utilization: %u%% (%u valid blocks, %u discard blocks)\n",
 				si->utilization, si->valid_count, si->discard_blks);
@@ -315,6 +322,8 @@ static int stat_show(struct seq_file *s, void *v)
 			   si->inline_inode);
 		seq_printf(s, "  - Inline_dentry Inode: %u\n",
 			   si->inline_dir);
+		seq_printf(s, "  - Compressed Inode: %u, Blocks: %u\n",
+			   si->compr_inode, si->compr_blocks);
 		seq_printf(s, "  - Orphan/Append/Update Inode: %u, %u, %u\n",
 			   si->orphans, si->append, si->update);
 		seq_printf(s, "\nMain area: %d segs, %d secs %d zones\n",
@@ -441,7 +450,7 @@ static int stat_show(struct seq_file *s, void *v)
 			   si->block_count[LFS], si->segment_count[LFS]);
 
 		/* segment usage info */
-		update_sit_info(si->sbi);
+		f2fs_update_sit_info(si->sbi);
 		seq_printf(s, "\nBDF: %u, avg. vblocks: %u\n",
 			   si->bimodal, si->avg_vblocks);
 
@@ -461,6 +470,7 @@ static int stat_show(struct seq_file *s, void *v)
 }
 
 DEFINE_SHOW_ATTRIBUTE(stat);
+#endif
 
 int f2fs_build_stats(struct f2fs_sb_info *sbi)
 {
@@ -491,11 +501,12 @@ int f2fs_build_stats(struct f2fs_sb_info *sbi)
 	atomic_set(&sbi->inline_xattr, 0);
 	atomic_set(&sbi->inline_inode, 0);
 	atomic_set(&sbi->inline_dir, 0);
+	atomic_set(&sbi->compr_inode, 0);
+	atomic_set(&sbi->compr_blocks, 0);
 	atomic_set(&sbi->inplace_count, 0);
 	for (i = META_CP; i < META_MAX; i++)
 		atomic_set(&sbi->meta_count[i], 0);
 
-	atomic_set(&sbi->aw_cnt, 0);
 	atomic_set(&sbi->vw_cnt, 0);
 	atomic_set(&sbi->max_aw_cnt, 0);
 	atomic_set(&sbi->max_vw_cnt, 0);
@@ -520,14 +531,18 @@ void f2fs_destroy_stats(struct f2fs_sb_info *sbi)
 
 void __init f2fs_create_root_stats(void)
 {
+#ifdef CONFIG_DEBUG_FS
 	f2fs_debugfs_root = debugfs_create_dir("f2fs", NULL);
 
 	debugfs_create_file("status", S_IRUGO, f2fs_debugfs_root, NULL,
 			    &stat_fops);
+#endif
 }
 
 void f2fs_destroy_root_stats(void)
 {
+#ifdef CONFIG_DEBUG_FS
 	debugfs_remove_recursive(f2fs_debugfs_root);
 	f2fs_debugfs_root = NULL;
+#endif
 }
diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 4033778bcbbf..44bfc464df78 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -471,7 +471,6 @@ struct page *f2fs_init_inode_metadata(struct inode *inode, struct inode *dir,
 			struct page *dpage)
 {
 	struct page *page;
-	int dummy_encrypt = DUMMY_ENCRYPTION_ENABLED(F2FS_I_SB(dir));
 	int err;
 
 	if (is_inode_flag_set(inode, FI_NEW_INODE)) {
@@ -498,8 +497,7 @@ struct page *f2fs_init_inode_metadata(struct inode *inode, struct inode *dir,
 		if (err)
 			goto put_error;
 
-		if ((IS_ENCRYPTED(dir) || dummy_encrypt) &&
-					f2fs_may_encrypt(inode)) {
+		if (IS_ENCRYPTED(inode)) {
 			err = fscrypt_inherit_context(dir, inode, page, false);
 			if (err)
 				goto put_error;
@@ -578,6 +576,20 @@ next:
 	goto next;
 }
 
+bool f2fs_has_enough_room(struct inode *dir, struct page *ipage,
+					struct fscrypt_name *fname)
+{
+	struct f2fs_dentry_ptr d;
+	unsigned int bit_pos;
+	int slots = GET_DENTRY_SLOTS(fname_len(fname));
+
+	make_dentry_ptr_inline(dir, &d, inline_data_addr(dir, ipage));
+
+	bit_pos = f2fs_room_for_filename(d.bitmap, slots, d.max);
+
+	return bit_pos < d.max;
+}
+
 void f2fs_update_dentry(nid_t ino, umode_t mode, struct f2fs_dentry_ptr *d,
 				const struct qstr *name, f2fs_hash_t name_hash,
 				unsigned int bit_pos)
@@ -628,7 +640,7 @@ int f2fs_add_regular_entry(struct inode *dir, const struct qstr *new_name,
 
 start:
 	if (time_to_inject(F2FS_I_SB(dir), FAULT_DIR_DEPTH)) {
-		f2fs_show_injection_info(FAULT_DIR_DEPTH);
+		f2fs_show_injection_info(F2FS_I_SB(dir), FAULT_DIR_DEPTH);
 		return -ENOSPC;
 	}
 
@@ -836,12 +848,6 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
 			0);
 	set_page_dirty(page);
 
-	dir->i_ctime = dir->i_mtime = current_time(dir);
-	f2fs_mark_inode_dirty_sync(dir, false);
-
-	if (inode)
-		f2fs_drop_nlink(dir, inode);
-
 	if (bit_pos == NR_DENTRY_IN_BLOCK &&
 		!f2fs_truncate_hole(dir, page->index, page->index + 1)) {
 		f2fs_clear_page_cache_dirty_tag(page);
@@ -853,6 +859,12 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
 		f2fs_remove_dirty_inode(dir);
 	}
 	f2fs_put_page(page, 1);
+
+	dir->i_ctime = dir->i_mtime = current_time(dir);
+	f2fs_mark_inode_dirty_sync(dir, false);
+
+	if (inode)
+		f2fs_drop_nlink(dir, inode);
 }
 
 bool f2fs_empty_dir(struct inode *dir)
@@ -919,8 +931,9 @@ int f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 			bit_pos++;
 			ctx->pos = start_pos + bit_pos;
 			printk_ratelimited(
-				"%s, invalid namelen(0), ino:%u, run fsck to fix.",
-				KERN_WARNING, le32_to_cpu(de->ino));
+				"%sF2FS-fs (%s): invalid namelen(0), ino:%u, run fsck to fix.",
+				KERN_WARNING, sbi->sb->s_id,
+				le32_to_cpu(de->ino));
 			set_sbi_flag(sbi, SBI_NEED_FSCK);
 			continue;
 		}
@@ -986,7 +999,7 @@ static int f2fs_readdir(struct file *file, struct dir_context *ctx)
 
 	if (IS_ENCRYPTED(inode)) {
 		err = fscrypt_get_encryption_info(inode);
-		if (err && err != -ENOKEY)
+		if (err)
 			goto out;
 
 		err = fscrypt_fname_alloc_buffer(inode, F2FS_NAME_LEN, &fstr);
@@ -1068,24 +1081,27 @@ static int f2fs_d_compare(const struct dentry *dentry, unsigned int len,
 			  const char *str, const struct qstr *name)
 {
 	struct qstr qstr = {.name = str, .len = len };
+	const struct dentry *parent = READ_ONCE(dentry->d_parent);
+	const struct inode *inode = READ_ONCE(parent->d_inode);
 
-	if (!IS_CASEFOLDED(dentry->d_parent->d_inode)) {
+	if (!inode || !IS_CASEFOLDED(inode)) {
 		if (len != name->len)
 			return -1;
-		return memcmp(str, name, len);
+		return memcmp(str, name->name, len);
 	}
 
-	return f2fs_ci_compare(dentry->d_parent->d_inode, name, &qstr, false);
+	return f2fs_ci_compare(inode, name, &qstr, false);
 }
 
 static int f2fs_d_hash(const struct dentry *dentry, struct qstr *str)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(dentry->d_sb);
 	const struct unicode_map *um = sbi->s_encoding;
+	const struct inode *inode = READ_ONCE(dentry->d_inode);
 	unsigned char *norm;
 	int len, ret = 0;
 
-	if (!IS_CASEFOLDED(dentry->d_inode))
+	if (!inode || !IS_CASEFOLDED(inode))
 		return 0;
 
 	norm = f2fs_kmalloc(sbi, PATH_MAX, GFP_ATOMIC);
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 4024790028aa..5c0149d2f46a 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -22,6 +22,7 @@
 #include <linux/bio.h>
 #include <linux/blkdev.h>
 #include <linux/quotaops.h>
+#include <linux/part_stat.h>
 #include <crypto/hash.h>
 
 #include <linux/fscrypt.h>
@@ -74,7 +75,6 @@ extern const char *f2fs_fault_name[FAULT_MAX];
 /*
  * For mount options
  */
-#define F2FS_MOUNT_BG_GC		0x00000001
 #define F2FS_MOUNT_DISABLE_ROLL_FORWARD	0x00000002
 #define F2FS_MOUNT_DISCARD		0x00000004
 #define F2FS_MOUNT_NOHEAP		0x00000008
@@ -88,11 +88,8 @@ extern const char *f2fs_fault_name[FAULT_MAX];
 #define F2FS_MOUNT_NOBARRIER		0x00000800
 #define F2FS_MOUNT_FASTBOOT		0x00001000
 #define F2FS_MOUNT_EXTENT_CACHE		0x00002000
-#define F2FS_MOUNT_FORCE_FG_GC		0x00004000
 #define F2FS_MOUNT_DATA_FLUSH		0x00008000
 #define F2FS_MOUNT_FAULT_INJECTION	0x00010000
-#define F2FS_MOUNT_ADAPTIVE		0x00020000
-#define F2FS_MOUNT_LFS			0x00040000
 #define F2FS_MOUNT_USRQUOTA		0x00080000
 #define F2FS_MOUNT_GRPQUOTA		0x00100000
 #define F2FS_MOUNT_PRJQUOTA		0x00200000
@@ -100,6 +97,7 @@ extern const char *f2fs_fault_name[FAULT_MAX];
 #define F2FS_MOUNT_INLINE_XATTR_SIZE	0x00800000
 #define F2FS_MOUNT_RESERVE_ROOT		0x01000000
 #define F2FS_MOUNT_DISABLE_CHECKPOINT	0x02000000
+#define F2FS_MOUNT_NORECOVERY		0x04000000
 
 #define F2FS_OPTION(sbi)	((sbi)->mount_opt)
 #define clear_opt(sbi, option)	(F2FS_OPTION(sbi).opt &= ~F2FS_MOUNT_##option)
@@ -116,6 +114,8 @@ typedef u32 block_t;	/*
 			 */
 typedef u32 nid_t;
 
+#define COMPRESS_EXT_NUM		16
+
 struct f2fs_mount_info {
 	unsigned int opt;
 	int write_io_size_bits;		/* Write IO size bits */
@@ -136,10 +136,18 @@ struct f2fs_mount_info {
 	int whint_mode;
 	int alloc_mode;			/* segment allocation policy */
 	int fsync_mode;			/* fsync policy */
-	bool test_dummy_encryption;	/* test dummy encryption */
+	int fs_mode;			/* fs mode: LFS or ADAPTIVE */
+	int bggc_mode;			/* bggc mode: off, on or sync */
+	struct fscrypt_dummy_context dummy_enc_ctx; /* test dummy encryption */
 	block_t unusable_cap;		/* Amount of space allowed to be
 					 * unusable when disabling checkpoint
 					 */
+
+	/* For compression */
+	unsigned char compress_algorithm;	/* algorithm type */
+	unsigned compress_log_size;		/* cluster log size */
+	unsigned char compress_ext_cnt;		/* extension count */
+	unsigned char extensions[COMPRESS_EXT_NUM][F2FS_EXTENSION_LEN];	/* extensions */
 };
 
 #define F2FS_FEATURE_ENCRYPT		0x0001
@@ -155,6 +163,7 @@ struct f2fs_mount_info {
 #define F2FS_FEATURE_VERITY		0x0400
 #define F2FS_FEATURE_SB_CHKSUM		0x0800
 #define F2FS_FEATURE_CASEFOLD		0x1000
+#define F2FS_FEATURE_COMPRESSION	0x2000
 
 #define __F2FS_HAS_FEATURE(raw_super, mask)				\
 	((raw_super->feature & cpu_to_le32(mask)) != 0)
@@ -322,8 +331,8 @@ struct discard_policy {
 	bool io_aware;			/* issue discard in idle time */
 	bool sync;			/* submit discard with REQ_SYNC flag */
 	bool ordered;			/* issue discard by lba order */
+	bool timeout;			/* discard timeout for put_super */
 	unsigned int granularity;	/* discard granularity */
-	int timeout;			/* discard timeout for put_super */
 };
 
 struct discard_cmd_control {
@@ -418,6 +427,7 @@ static inline bool __has_cursum_space(struct f2fs_journal *journal,
 #define F2FS_IOC_GET_PIN_FILE		_IOR(F2FS_IOCTL_MAGIC, 14, __u32)
 #define F2FS_IOC_PRECACHE_EXTENTS	_IO(F2FS_IOCTL_MAGIC, 15)
 #define F2FS_IOC_RESIZE_FS		_IOW(F2FS_IOCTL_MAGIC, 16, __u64)
+#define F2FS_IOC_GET_COMPRESS_BLOCKS	_IOR(F2FS_IOCTL_MAGIC, 17, __u64)
 
 #define F2FS_IOC_GET_VOLUME_NAME	FS_IOC_GETFSLABEL
 #define F2FS_IOC_SET_VOLUME_NAME	FS_IOC_SETFSLABEL
@@ -550,6 +560,9 @@ enum {
 
 #define DEFAULT_RETRY_IO_COUNT	8	/* maximum retry read IO count */
 
+/* congestion wait timeout value, default: 20ms */
+#define	DEFAULT_IO_TIMEOUT	(msecs_to_jiffies(20))
+
 /* maximum retry quota flush count */
 #define DEFAULT_RETRY_QUOTA_FLUSH_COUNT		8
 
@@ -666,6 +679,44 @@ enum {
 	MAX_GC_FAILURE
 };
 
+/* used for f2fs_inode_info->flags */
+enum {
+	FI_NEW_INODE,		/* indicate newly allocated inode */
+	FI_DIRTY_INODE,		/* indicate inode is dirty or not */
+	FI_AUTO_RECOVER,	/* indicate inode is recoverable */
+	FI_DIRTY_DIR,		/* indicate directory has dirty pages */
+	FI_INC_LINK,		/* need to increment i_nlink */
+	FI_ACL_MODE,		/* indicate acl mode */
+	FI_NO_ALLOC,		/* should not allocate any blocks */
+	FI_FREE_NID,		/* free allocated nide */
+	FI_NO_EXTENT,		/* not to use the extent cache */
+	FI_INLINE_XATTR,	/* used for inline xattr */
+	FI_INLINE_DATA,		/* used for inline data*/
+	FI_INLINE_DENTRY,	/* used for inline dentry */
+	FI_APPEND_WRITE,	/* inode has appended data */
+	FI_UPDATE_WRITE,	/* inode has in-place-update data */
+	FI_NEED_IPU,		/* used for ipu per file */
+	FI_ATOMIC_FILE,		/* indicate atomic file */
+	FI_ATOMIC_COMMIT,	/* indicate the state of atomical committing */
+	FI_VOLATILE_FILE,	/* indicate volatile file */
+	FI_FIRST_BLOCK_WRITTEN,	/* indicate #0 data block was written */
+	FI_DROP_CACHE,		/* drop dirty page cache */
+	FI_DATA_EXIST,		/* indicate data exists */
+	FI_INLINE_DOTS,		/* indicate inline dot dentries */
+	FI_DO_DEFRAG,		/* indicate defragment is running */
+	FI_DIRTY_FILE,		/* indicate regular/symlink has dirty pages */
+	FI_NO_PREALLOC,		/* indicate skipped preallocated blocks */
+	FI_HOT_DATA,		/* indicate file is hot */
+	FI_EXTRA_ATTR,		/* indicate file has extra attribute */
+	FI_PROJ_INHERIT,	/* indicate file inherits projectid */
+	FI_PIN_FILE,		/* indicate file should not be gced */
+	FI_ATOMIC_REVOKE_REQUEST, /* request to drop atomic data */
+	FI_VERITY_IN_PROGRESS,	/* building fs-verity Merkle tree */
+	FI_COMPRESSED_FILE,	/* indicate file's data can be compressed */
+	FI_MMAP_FILE,		/* indicate file was mmapped */
+	FI_MAX,			/* max flag, never be used */
+};
+
 struct f2fs_inode_info {
 	struct inode vfs_inode;		/* serve a vfs inode */
 	unsigned long i_flags;		/* keep an inode flags for ioctl */
@@ -678,7 +729,7 @@ struct f2fs_inode_info {
 	umode_t i_acl_mode;		/* keep file acl mode temporarily */
 
 	/* Use below internally in f2fs*/
-	unsigned long flags;		/* use to pass per-file flags */
+	unsigned long flags[BITS_TO_LONGS(FI_MAX)];	/* use to pass per-file flags */
 	struct rw_semaphore i_sem;	/* protect fi info */
 	atomic_t dirty_pages;		/* # of dirty pages */
 	f2fs_hash_t chash;		/* hash value of given file name */
@@ -687,6 +738,7 @@ struct f2fs_inode_info {
 	struct task_struct *cp_task;	/* separate cp/wb IO stats*/
 	nid_t i_xattr_nid;		/* node id that contains xattrs */
 	loff_t	last_disk_size;		/* lastly written file size */
+	spinlock_t i_size_lock;		/* protect last_disk_size */
 
 #ifdef CONFIG_QUOTA
 	struct dquot *i_dquot[MAXQUOTAS];
@@ -712,6 +764,12 @@ struct f2fs_inode_info {
 	int i_inline_xattr_size;	/* inline xattr size */
 	struct timespec64 i_crtime;	/* inode creation time */
 	struct timespec64 i_disk_time[4];/* inode disk times */
+
+	/* for file compress */
+	u64 i_compr_blocks;			/* # of compressed blocks */
+	unsigned char i_compress_algorithm;	/* algorithm type */
+	unsigned char i_log_cluster_size;	/* log of cluster size */
+	unsigned int i_cluster_size;		/* cluster size */
 };
 
 static inline void get_extent_info(struct extent_info *ext,
@@ -890,6 +948,7 @@ enum {
 	CURSEG_WARM_NODE,	/* direct node blocks of normal files */
 	CURSEG_COLD_NODE,	/* indirect node blocks */
 	NO_CHECK_TYPE,
+	CURSEG_COLD_DATA_PINNED,/* cold data for pinned file */
 };
 
 struct flush_cmd {
@@ -1017,6 +1076,7 @@ enum need_lock_type {
 enum cp_reason_type {
 	CP_NO_NEEDED,
 	CP_NON_REGULAR,
+	CP_COMPRESSED,
 	CP_HARDLINK,
 	CP_SB_NEED_CP,
 	CP_WRONG_PINO,
@@ -1055,12 +1115,15 @@ struct f2fs_io_info {
 	block_t old_blkaddr;	/* old block address before Cow */
 	struct page *page;	/* page to be written */
 	struct page *encrypted_page;	/* encrypted page */
+	struct page *compressed_page;	/* compressed page */
 	struct list_head list;		/* serialize IOs */
 	bool submitted;		/* indicate IO submission */
 	int need_lock;		/* indicate we need to lock cp_rwsem */
 	bool in_list;		/* indicate fio is in io_list */
 	bool is_por;		/* indicate IO is from recovery or not */
 	bool retry;		/* need to reallocate block address */
+	int compr_blocks;	/* # of compressed block addresses */
+	bool encrypted;		/* indicate file is encrypted */
 	enum iostat_type io_type;	/* io type */
 	struct writeback_control *io_wbc; /* writeback control */
 	struct bio **bio;		/* bio for ipu */
@@ -1068,6 +1131,11 @@ struct f2fs_io_info {
 	unsigned char version;		/* version of the node */
 };
 
+struct bio_entry {
+	struct bio *bio;
+	struct list_head list;
+};
+
 #define is_read_io(rw) ((rw) == READ)
 struct f2fs_bio_info {
 	struct f2fs_sb_info *sbi;	/* f2fs superblock */
@@ -1077,6 +1145,8 @@ struct f2fs_bio_info {
 	struct rw_semaphore io_rwsem;	/* blocking op for bio */
 	spinlock_t io_lock;		/* serialize DATA/NODE IOs */
 	struct list_head io_list;	/* track fios */
+	struct list_head bio_list;	/* bio entry list head */
+	struct rw_semaphore bio_list_lock;	/* lock to protect bio entry list */
 };
 
 #define FDEV(i)				(sbi->devs[i])
@@ -1145,6 +1215,20 @@ enum {
 };
 
 enum {
+	BGGC_MODE_ON,		/* background gc is on */
+	BGGC_MODE_OFF,		/* background gc is off */
+	BGGC_MODE_SYNC,		/*
+				 * background gc is on, migrating blocks
+				 * like foreground gc
+				 */
+};
+
+enum {
+	FS_MODE_ADAPTIVE,	/* use both lfs/ssr allocation */
+	FS_MODE_LFS,		/* use lfs allocation only */
+};
+
+enum {
 	WHINT_MODE_OFF,		/* not pass down write hints */
 	WHINT_MODE_USER,	/* try to pass down hints given by users */
 	WHINT_MODE_FS,		/* pass down hints with F2FS policy */
@@ -1161,13 +1245,98 @@ enum fsync_mode {
 	FSYNC_MODE_NOBARRIER,	/* fsync behaves nobarrier based on posix */
 };
 
+/*
+ * this value is set in page as a private data which indicate that
+ * the page is atomically written, and it is in inmem_pages list.
+ */
+#define ATOMIC_WRITTEN_PAGE		((unsigned long)-1)
+#define DUMMY_WRITTEN_PAGE		((unsigned long)-2)
+
+#define IS_ATOMIC_WRITTEN_PAGE(page)			\
+		(page_private(page) == (unsigned long)ATOMIC_WRITTEN_PAGE)
+#define IS_DUMMY_WRITTEN_PAGE(page)			\
+		(page_private(page) == (unsigned long)DUMMY_WRITTEN_PAGE)
+
 #ifdef CONFIG_FS_ENCRYPTION
 #define DUMMY_ENCRYPTION_ENABLED(sbi) \
-			(unlikely(F2FS_OPTION(sbi).test_dummy_encryption))
+	(unlikely(F2FS_OPTION(sbi).dummy_enc_ctx.ctx != NULL))
 #else
 #define DUMMY_ENCRYPTION_ENABLED(sbi) (0)
 #endif
 
+/* For compression */
+enum compress_algorithm_type {
+	COMPRESS_LZO,
+	COMPRESS_LZ4,
+	COMPRESS_ZSTD,
+	COMPRESS_MAX,
+};
+
+#define COMPRESS_DATA_RESERVED_SIZE		5
+struct compress_data {
+	__le32 clen;			/* compressed data size */
+	__le32 reserved[COMPRESS_DATA_RESERVED_SIZE];	/* reserved */
+	u8 cdata[];			/* compressed data */
+};
+
+#define COMPRESS_HEADER_SIZE	(sizeof(struct compress_data))
+
+#define F2FS_COMPRESSED_PAGE_MAGIC	0xF5F2C000
+
+/* compress context */
+struct compress_ctx {
+	struct inode *inode;		/* inode the context belong to */
+	pgoff_t cluster_idx;		/* cluster index number */
+	unsigned int cluster_size;	/* page count in cluster */
+	unsigned int log_cluster_size;	/* log of cluster size */
+	struct page **rpages;		/* pages store raw data in cluster */
+	unsigned int nr_rpages;		/* total page number in rpages */
+	struct page **cpages;		/* pages store compressed data in cluster */
+	unsigned int nr_cpages;		/* total page number in cpages */
+	void *rbuf;			/* virtual mapped address on rpages */
+	struct compress_data *cbuf;	/* virtual mapped address on cpages */
+	size_t rlen;			/* valid data length in rbuf */
+	size_t clen;			/* valid data length in cbuf */
+	void *private;			/* payload buffer for specified compression algorithm */
+	void *private2;			/* extra payload buffer */
+};
+
+/* compress context for write IO path */
+struct compress_io_ctx {
+	u32 magic;			/* magic number to indicate page is compressed */
+	struct inode *inode;		/* inode the context belong to */
+	struct page **rpages;		/* pages store raw data in cluster */
+	unsigned int nr_rpages;		/* total page number in rpages */
+	refcount_t ref;			/* referrence count of raw page */
+};
+
+/* decompress io context for read IO path */
+struct decompress_io_ctx {
+	u32 magic;			/* magic number to indicate page is compressed */
+	struct inode *inode;		/* inode the context belong to */
+	pgoff_t cluster_idx;		/* cluster index number */
+	unsigned int cluster_size;	/* page count in cluster */
+	unsigned int log_cluster_size;	/* log of cluster size */
+	struct page **rpages;		/* pages store raw data in cluster */
+	unsigned int nr_rpages;		/* total page number in rpages */
+	struct page **cpages;		/* pages store compressed data in cluster */
+	unsigned int nr_cpages;		/* total page number in cpages */
+	struct page **tpages;		/* temp pages to pad holes in cluster */
+	void *rbuf;			/* virtual mapped address on rpages */
+	struct compress_data *cbuf;	/* virtual mapped address on cpages */
+	size_t rlen;			/* valid data length in rbuf */
+	size_t clen;			/* valid data length in cbuf */
+	refcount_t ref;			/* referrence count of compressed page */
+	bool failed;			/* indicate IO error during decompression */
+	void *private;			/* payload buffer for specified decompression algorithm */
+	void *private2;			/* extra payload buffer */
+};
+
+#define NULL_CLUSTER			((unsigned int)(~0))
+#define MIN_COMPRESS_LOG_SIZE		2
+#define MAX_COMPRESS_LOG_SIZE		8
+#define MAX_COMPRESS_WINDOW_SIZE	((PAGE_SIZE) << MAX_COMPRESS_LOG_SIZE)
+
 struct f2fs_sb_info {
 	struct super_block *sb;			/* pointer to VFS super block */
 	struct proc_dir_entry *s_proc;		/* proc entry */
@@ -1283,17 +1452,22 @@ struct f2fs_sb_info {
 	struct f2fs_mount_info mount_opt;	/* mount options */
 
 	/* for cleaning operations */
-	struct mutex gc_mutex;			/* mutex for GC */
+	struct rw_semaphore gc_lock;		/*
+						 * semaphore for GC, avoid
+						 * race between GC and GC or CP
+						 */
 	struct f2fs_gc_kthread	*gc_thread;	/* GC thread */
 	unsigned int cur_victim_sec;		/* current victim section num */
 	unsigned int gc_mode;			/* current GC state */
 	unsigned int next_victim_seg[2];	/* next segment in victim section */
 	/* for skip statistic */
+	unsigned int atomic_files;              /* # of opened atomic file */
 	unsigned long long skipped_atomic_files[2];	/* FG_GC and BG_GC */
 	unsigned long long skipped_gc_rwsem;		/* FG_GC only */
 
 	/* threshold for gc trials on pinned files */
 	u64 gc_pin_file_threshold;
+	struct rw_semaphore pin_sem;
 
 	/* maximum # of trials to find a victim segment for SSR and GC */
 	unsigned int max_victim_search;
@@ -1317,11 +1491,11 @@ struct f2fs_sb_info {
 	atomic_t inline_xattr;			/* # of inline_xattr inodes */
 	atomic_t inline_inode;			/* # of inline_data inodes */
 	atomic_t inline_dir;			/* # of inline_dentry inodes */
-	atomic_t aw_cnt;			/* # of atomic writes */
+	atomic_t compr_inode;			/* # of compressed inodes */
+	atomic_t compr_blocks;			/* # of compressed blocks */
 	atomic_t vw_cnt;			/* # of volatile writes */
 	atomic_t max_aw_cnt;			/* max # of atomic writes */
 	atomic_t max_vw_cnt;			/* max # of volatile writes */
-	int bg_gc;				/* background gc calls */
 	unsigned int io_skip_bggc;		/* skip background gc for in-flight IO */
 	unsigned int other_skip_bggc;		/* skip background gc for other reasons */
 	unsigned int ndirty_inode[NR_INODE_TYPE];	/* # of dirty inodes */
@@ -1355,6 +1529,11 @@ struct f2fs_sb_info {
 
 	/* Precomputed FS UUID checksum for seeding other checksums */
 	__u32 s_chksum_seed;
+
+	struct workqueue_struct *post_read_wq;	/* post read workqueue */
+
+	struct kmem_cache *inline_xattr_slab;	/* inline xattr entry */
+	unsigned int inline_xattr_slab_size;	/* default inline xattr slab size */
 };
 
 struct f2fs_private_dio {
@@ -1365,9 +1544,10 @@ struct f2fs_private_dio {
 };
 
 #ifdef CONFIG_F2FS_FAULT_INJECTION
-#define f2fs_show_injection_info(type)					\
-	printk_ratelimited("%sF2FS-fs : inject %s in %s of %pS\n",	\
-		KERN_INFO, f2fs_fault_name[type],			\
+#define f2fs_show_injection_info(sbi, type)					\
+	printk_ratelimited("%sF2FS-fs (%s) : inject %s in %s of %pS\n",	\
+		KERN_INFO, sbi->sb->s_id,				\
+		f2fs_fault_name[type],					\
 		__func__, __builtin_return_address(0))
 static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type)
 {
@@ -1387,7 +1567,7 @@ static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type)
 	return false;
 }
 #else
-#define f2fs_show_injection_info(type) do { } while (0)
+#define f2fs_show_injection_info(sbi, type) do { } while (0)
 static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type)
 {
 	return false;
@@ -1772,7 +1952,7 @@ static inline int inc_valid_block_count(struct f2fs_sb_info *sbi,
 		return ret;
 
 	if (time_to_inject(sbi, FAULT_BLOCK)) {
-		f2fs_show_injection_info(FAULT_BLOCK);
+		f2fs_show_injection_info(sbi, FAULT_BLOCK);
 		release = *count;
 		goto release_quota;
 	}
@@ -2024,7 +2204,7 @@ static inline int inc_valid_node_count(struct f2fs_sb_info *sbi,
 	}
 
 	if (time_to_inject(sbi, FAULT_BLOCK)) {
-		f2fs_show_injection_info(FAULT_BLOCK);
+		f2fs_show_injection_info(sbi, FAULT_BLOCK);
 		goto enospc;
 	}
 
@@ -2094,7 +2274,7 @@ static inline void dec_valid_node_count(struct f2fs_sb_info *sbi,
 		dquot_free_inode(inode);
 	} else {
 		if (unlikely(inode->i_blocks == 0)) {
-			f2fs_warn(sbi, "Inconsistent i_blocks, ino:%lu, iblocks:%llu",
+			f2fs_warn(sbi, "dec_valid_node_count: inconsistent i_blocks, ino:%lu, iblocks:%llu",
 				  inode->i_ino,
 				  (unsigned long long)inode->i_blocks);
 			set_sbi_flag(sbi, SBI_NEED_FSCK);
@@ -2139,7 +2319,8 @@ static inline struct page *f2fs_grab_cache_page(struct address_space *mapping,
 			return page;
 
 		if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_ALLOC)) {
-			f2fs_show_injection_info(FAULT_PAGE_ALLOC);
+			f2fs_show_injection_info(F2FS_M_SB(mapping),
+							FAULT_PAGE_ALLOC);
 			return NULL;
 		}
 	}
@@ -2154,7 +2335,7 @@ static inline struct page *f2fs_pagecache_get_page(
 				int fgp_flags, gfp_t gfp_mask)
 {
 	if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_GET)) {
-		f2fs_show_injection_info(FAULT_PAGE_GET);
+		f2fs_show_injection_info(F2FS_M_SB(mapping), FAULT_PAGE_GET);
 		return NULL;
 	}
 
@@ -2210,26 +2391,6 @@ static inline void *f2fs_kmem_cache_alloc(struct kmem_cache *cachep,
 	return entry;
 }
 
-static inline struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi,
-						int npages, bool no_fail)
-{
-	struct bio *bio;
-
-	if (no_fail) {
-		/* No failure on bio allocation */
-		bio = bio_alloc(GFP_NOIO, npages);
-		if (!bio)
-			bio = bio_alloc(GFP_NOIO | __GFP_NOFAIL, npages);
-		return bio;
-	}
-	if (time_to_inject(sbi, FAULT_ALLOC_BIO)) {
-		f2fs_show_injection_info(FAULT_ALLOC_BIO);
-		return NULL;
-	}
-
-	return bio_alloc(GFP_KERNEL, npages);
-}
-
 static inline bool is_idle(struct f2fs_sb_info *sbi, int type)
 {
 	if (sbi->gc_mode == GC_URGENT)
@@ -2281,7 +2442,7 @@ static inline __le32 *blkaddr_in_node(struct f2fs_node *node)
 }
 
 static inline int f2fs_has_extra_attr(struct inode *inode);
-static inline block_t datablock_addr(struct inode *inode,
+static inline block_t data_blkaddr(struct inode *inode,
 			struct page *node_page, unsigned int offset)
 {
 	struct f2fs_node *raw_node;
@@ -2291,9 +2452,9 @@ static inline block_t datablock_addr(struct inode *inode,
 
 	raw_node = F2FS_NODE(node_page);
 
-	/* from GC path only */
 	if (is_inode) {
 		if (!inode)
+			/* from GC path only */
 			base = offset_in_addr(&raw_node->i);
 		else if (f2fs_has_extra_attr(inode))
 			base = get_extra_isize(inode);
@@ -2303,6 +2464,11 @@ static inline block_t datablock_addr(struct inode *inode,
 	return le32_to_cpu(addr_array[base + offset]);
 }
 
+static inline block_t f2fs_data_blkaddr(struct dnode_of_data *dn)
+{
+	return data_blkaddr(dn->inode, dn->node_page, dn->ofs_in_node);
+}
+
 static inline int f2fs_test_bit(unsigned int nr, char *addr)
 {
 	int mask;
@@ -2366,11 +2532,13 @@ static inline void f2fs_change_bit(unsigned int nr, char *addr)
 /*
  * On-disk inode flags (f2fs_inode::i_flags)
  */
+#define F2FS_COMPR_FL			0x00000004 /* Compress file */
 #define F2FS_SYNC_FL			0x00000008 /* Synchronous updates */
 #define F2FS_IMMUTABLE_FL		0x00000010 /* Immutable file */
 #define F2FS_APPEND_FL			0x00000020 /* writes to file may only append */
 #define F2FS_NODUMP_FL			0x00000040 /* do not dump file */
 #define F2FS_NOATIME_FL			0x00000080 /* do not update atime */
+#define F2FS_NOCOMP_FL			0x00000400 /* Don't compress */
 #define F2FS_INDEX_FL			0x00001000 /* hash-indexed directory */
 #define F2FS_DIRSYNC_FL			0x00010000 /* dirsync behaviour (directories only) */
 #define F2FS_PROJINHERIT_FL		0x20000000 /* Create with parents projid */
@@ -2379,7 +2547,7 @@ static inline void f2fs_change_bit(unsigned int nr, char *addr)
 /* Flags that should be inherited by new inodes from their parent. */
 #define F2FS_FL_INHERITED (F2FS_SYNC_FL | F2FS_NODUMP_FL | F2FS_NOATIME_FL | \
 			   F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL | \
-			   F2FS_CASEFOLD_FL)
+			   F2FS_CASEFOLD_FL | F2FS_COMPR_FL | F2FS_NOCOMP_FL)
 
 /* Flags that are appropriate for regular files (all but dir-specific ones). */
 #define F2FS_REG_FLMASK		(~(F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL | \
@@ -2398,41 +2566,6 @@ static inline __u32 f2fs_mask_flags(umode_t mode, __u32 flags)
 		return flags & F2FS_OTHER_FLMASK;
 }
 
-/* used for f2fs_inode_info->flags */
-enum {
-	FI_NEW_INODE,		/* indicate newly allocated inode */
-	FI_DIRTY_INODE,		/* indicate inode is dirty or not */
-	FI_AUTO_RECOVER,	/* indicate inode is recoverable */
-	FI_DIRTY_DIR,		/* indicate directory has dirty pages */
-	FI_INC_LINK,		/* need to increment i_nlink */
-	FI_ACL_MODE,		/* indicate acl mode */
-	FI_NO_ALLOC,		/* should not allocate any blocks */
-	FI_FREE_NID,		/* free allocated nide */
-	FI_NO_EXTENT,		/* not to use the extent cache */
-	FI_INLINE_XATTR,	/* used for inline xattr */
-	FI_INLINE_DATA,		/* used for inline data*/
-	FI_INLINE_DENTRY,	/* used for inline dentry */
-	FI_APPEND_WRITE,	/* inode has appended data */
-	FI_UPDATE_WRITE,	/* inode has in-place-update data */
-	FI_NEED_IPU,		/* used for ipu per file */
-	FI_ATOMIC_FILE,		/* indicate atomic file */
-	FI_ATOMIC_COMMIT,	/* indicate the state of atomical committing */
-	FI_VOLATILE_FILE,	/* indicate volatile file */
-	FI_FIRST_BLOCK_WRITTEN,	/* indicate #0 data block was written */
-	FI_DROP_CACHE,		/* drop dirty page cache */
-	FI_DATA_EXIST,		/* indicate data exists */
-	FI_INLINE_DOTS,		/* indicate inline dot dentries */
-	FI_DO_DEFRAG,		/* indicate defragment is running */
-	FI_DIRTY_FILE,		/* indicate regular/symlink has dirty pages */
-	FI_NO_PREALLOC,		/* indicate skipped preallocated blocks */
-	FI_HOT_DATA,		/* indicate file is hot */
-	FI_EXTRA_ATTR,		/* indicate file has extra attribute */
-	FI_PROJ_INHERIT,	/* indicate file inherits projectid */
-	FI_PIN_FILE,		/* indicate file should not be gced */
-	FI_ATOMIC_REVOKE_REQUEST, /* request to drop atomic data */
-	FI_VERITY_IN_PROGRESS,	/* building fs-verity Merkle tree */
-};
-
 static inline void __mark_inode_dirty_flag(struct inode *inode,
 						int flag, bool set)
 {
@@ -2453,20 +2586,18 @@ static inline void __mark_inode_dirty_flag(struct inode *inode,
 
 static inline void set_inode_flag(struct inode *inode, int flag)
 {
-	if (!test_bit(flag, &F2FS_I(inode)->flags))
-		set_bit(flag, &F2FS_I(inode)->flags);
+	test_and_set_bit(flag, F2FS_I(inode)->flags);
 	__mark_inode_dirty_flag(inode, flag, true);
 }
 
 static inline int is_inode_flag_set(struct inode *inode, int flag)
 {
-	return test_bit(flag, &F2FS_I(inode)->flags);
+	return test_bit(flag, F2FS_I(inode)->flags);
 }
 
 static inline void clear_inode_flag(struct inode *inode, int flag)
 {
-	if (test_bit(flag, &F2FS_I(inode)->flags))
-		clear_bit(flag, &F2FS_I(inode)->flags);
+	test_and_clear_bit(flag, F2FS_I(inode)->flags);
 	__mark_inode_dirty_flag(inode, flag, false);
 }
 
@@ -2557,19 +2688,19 @@ static inline void get_inline_info(struct inode *inode, struct f2fs_inode *ri)
 	struct f2fs_inode_info *fi = F2FS_I(inode);
 
 	if (ri->i_inline & F2FS_INLINE_XATTR)
-		set_bit(FI_INLINE_XATTR, &fi->flags);
+		set_bit(FI_INLINE_XATTR, fi->flags);
 	if (ri->i_inline & F2FS_INLINE_DATA)
-		set_bit(FI_INLINE_DATA, &fi->flags);
+		set_bit(FI_INLINE_DATA, fi->flags);
 	if (ri->i_inline & F2FS_INLINE_DENTRY)
-		set_bit(FI_INLINE_DENTRY, &fi->flags);
+		set_bit(FI_INLINE_DENTRY, fi->flags);
 	if (ri->i_inline & F2FS_DATA_EXIST)
-		set_bit(FI_DATA_EXIST, &fi->flags);
+		set_bit(FI_DATA_EXIST, fi->flags);
 	if (ri->i_inline & F2FS_INLINE_DOTS)
-		set_bit(FI_INLINE_DOTS, &fi->flags);
+		set_bit(FI_INLINE_DOTS, fi->flags);
 	if (ri->i_inline & F2FS_EXTRA_ATTR)
-		set_bit(FI_EXTRA_ATTR, &fi->flags);
+		set_bit(FI_EXTRA_ATTR, fi->flags);
 	if (ri->i_inline & F2FS_PIN_FILE)
-		set_bit(FI_PIN_FILE, &fi->flags);
+		set_bit(FI_PIN_FILE, fi->flags);
 }
 
 static inline void set_raw_inline(struct inode *inode, struct f2fs_inode *ri)
@@ -2602,16 +2733,27 @@ static inline int f2fs_has_inline_xattr(struct inode *inode)
 	return is_inode_flag_set(inode, FI_INLINE_XATTR);
 }
 
+static inline int f2fs_compressed_file(struct inode *inode)
+{
+	return S_ISREG(inode->i_mode) &&
+		is_inode_flag_set(inode, FI_COMPRESSED_FILE);
+}
+
 static inline unsigned int addrs_per_inode(struct inode *inode)
 {
 	unsigned int addrs = CUR_ADDRS_PER_INODE(inode) -
 				get_inline_xattr_addrs(inode);
-	return ALIGN_DOWN(addrs, 1);
+
+	if (!f2fs_compressed_file(inode))
+		return addrs;
+	return ALIGN_DOWN(addrs, F2FS_I(inode)->i_cluster_size);
 }
 
 static inline unsigned int addrs_per_block(struct inode *inode)
 {
-	return ALIGN_DOWN(DEF_ADDRS_PER_BLOCK, 1);
+	if (!f2fs_compressed_file(inode))
+		return DEF_ADDRS_PER_BLOCK;
+	return ALIGN_DOWN(DEF_ADDRS_PER_BLOCK, F2FS_I(inode)->i_cluster_size);
 }
 
 static inline void *inline_xattr_addr(struct inode *inode, struct page *page)
@@ -2644,6 +2786,11 @@ static inline int f2fs_has_inline_dots(struct inode *inode)
 	return is_inode_flag_set(inode, FI_INLINE_DOTS);
 }
 
+static inline int f2fs_is_mmap_file(struct inode *inode)
+{
+	return is_inode_flag_set(inode, FI_MMAP_FILE);
+}
+
 static inline bool f2fs_is_pinned_file(struct inode *inode)
 {
 	return is_inode_flag_set(inode, FI_PIN_FILE);
@@ -2704,6 +2851,20 @@ static inline void clear_file(struct inode *inode, int type)
 	f2fs_mark_inode_dirty_sync(inode, true);
 }
 
+static inline bool f2fs_is_time_consistent(struct inode *inode)
+{
+	if (!timespec64_equal(F2FS_I(inode)->i_disk_time, &inode->i_atime))
+		return false;
+	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &inode->i_ctime))
+		return false;
+	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 2, &inode->i_mtime))
+		return false;
+	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 3,
+						&F2FS_I(inode)->i_crtime))
+		return false;
+	return true;
+}
+
 static inline bool f2fs_skip_inode_update(struct inode *inode, int dsync)
 {
 	bool ret;
@@ -2721,19 +2882,12 @@ static inline bool f2fs_skip_inode_update(struct inode *inode, int dsync)
 			i_size_read(inode) & ~PAGE_MASK)
 		return false;
 
-	if (!timespec64_equal(F2FS_I(inode)->i_disk_time, &inode->i_atime))
-		return false;
-	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &inode->i_ctime))
-		return false;
-	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 2, &inode->i_mtime))
-		return false;
-	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 3,
-						&F2FS_I(inode)->i_crtime))
+	if (!f2fs_is_time_consistent(inode))
 		return false;
 
-	down_read(&F2FS_I(inode)->i_sem);
+	spin_lock(&F2FS_I(inode)->i_size_lock);
 	ret = F2FS_I(inode)->last_disk_size == i_size_read(inode);
-	up_read(&F2FS_I(inode)->i_sem);
+	spin_unlock(&F2FS_I(inode)->i_size_lock);
 
 	return ret;
 }
@@ -2764,7 +2918,8 @@ static inline bool f2fs_may_extent_tree(struct inode *inode)
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 
 	if (!test_opt(sbi, EXTENT_CACHE) ||
-			is_inode_flag_set(inode, FI_NO_EXTENT))
+			is_inode_flag_set(inode, FI_NO_EXTENT) ||
+			is_inode_flag_set(inode, FI_COMPRESSED_FILE))
 		return false;
 
 	/*
@@ -2783,7 +2938,7 @@ static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi,
 	void *ret;
 
 	if (time_to_inject(sbi, FAULT_KMALLOC)) {
-		f2fs_show_injection_info(FAULT_KMALLOC);
+		f2fs_show_injection_info(sbi, FAULT_KMALLOC);
 		return NULL;
 	}
 
@@ -2804,7 +2959,7 @@ static inline void *f2fs_kvmalloc(struct f2fs_sb_info *sbi,
 					size_t size, gfp_t flags)
 {
 	if (time_to_inject(sbi, FAULT_KVMALLOC)) {
-		f2fs_show_injection_info(FAULT_KVMALLOC);
+		f2fs_show_injection_info(sbi, FAULT_KVMALLOC);
 		return NULL;
 	}
 
@@ -2884,7 +3039,8 @@ static inline void verify_blkaddr(struct f2fs_sb_info *sbi,
 
 static inline bool __is_valid_data_blkaddr(block_t blkaddr)
 {
-	if (blkaddr == NEW_ADDR || blkaddr == NULL_ADDR)
+	if (blkaddr == NEW_ADDR || blkaddr == NULL_ADDR ||
+			blkaddr == COMPRESS_ADDR)
 		return false;
 	return true;
 }
@@ -2895,19 +3051,12 @@ static inline void f2fs_set_page_private(struct page *page,
 	if (PagePrivate(page))
 		return;
 
-	get_page(page);
-	SetPagePrivate(page);
-	set_page_private(page, data);
+	attach_page_private(page, (void *)data);
 }
 
 static inline void f2fs_clear_page_private(struct page *page)
 {
-	if (!PagePrivate(page))
-		return;
-
-	set_page_private(page, 0);
-	ClearPagePrivate(page);
-	f2fs_put_page(page, 0);
+	detach_page_private(page);
 }
 
 /*
@@ -2982,6 +3131,8 @@ ino_t f2fs_inode_by_name(struct inode *dir, const struct qstr *qstr,
 			struct page **page);
 void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de,
 			struct page *page, struct inode *inode);
+bool f2fs_has_enough_room(struct inode *dir, struct page *ipage,
+			struct fscrypt_name *fname);
 void f2fs_update_dentry(nid_t ino, umode_t mode, struct f2fs_dentry_ptr *d,
 			const struct qstr *name, f2fs_hash_t name_hash,
 			unsigned int bit_pos);
@@ -3083,7 +3234,7 @@ void f2fs_drop_inmem_pages(struct inode *inode);
 void f2fs_drop_inmem_page(struct inode *inode, struct page *page);
 int f2fs_commit_inmem_pages(struct inode *inode);
 void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need);
-void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi);
+void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi, bool from_bg);
 int f2fs_issue_flush(struct f2fs_sb_info *sbi, nid_t ino);
 int f2fs_create_flush_cmd_control(struct f2fs_sb_info *sbi);
 int f2fs_flush_device_cache(struct f2fs_sb_info *sbi);
@@ -3102,7 +3253,7 @@ void f2fs_release_discard_addrs(struct f2fs_sb_info *sbi);
 int f2fs_npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra);
 void allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
 					unsigned int start, unsigned int end);
-void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi);
+void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi, int type);
 int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range);
 bool f2fs_exist_trim_candidates(struct f2fs_sb_info *sbi,
 					struct cp_control *cpc);
@@ -3136,6 +3287,8 @@ void f2fs_write_node_summaries(struct f2fs_sb_info *sbi, block_t start_blk);
 int f2fs_lookup_journal_in_cursum(struct f2fs_journal *journal, int type,
 			unsigned int val, int alloc);
 void f2fs_flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc);
+int f2fs_fix_curseg_write_pointer(struct f2fs_sb_info *sbi);
+int f2fs_check_write_pointer(struct f2fs_sb_info *sbi);
 int f2fs_build_segment_manager(struct f2fs_sb_info *sbi);
 void f2fs_destroy_segment_manager(struct f2fs_sb_info *sbi);
 int __init f2fs_create_segment_manager_caches(void);
@@ -3177,7 +3330,7 @@ int f2fs_get_valid_checkpoint(struct f2fs_sb_info *sbi);
 void f2fs_update_dirty_page(struct inode *inode, struct page *page);
 void f2fs_remove_dirty_inode(struct inode *inode);
 int f2fs_sync_dirty_inodes(struct f2fs_sb_info *sbi, enum inode_type type);
-void f2fs_wait_on_all_pages_writeback(struct f2fs_sb_info *sbi);
+void f2fs_wait_on_all_pages(struct f2fs_sb_info *sbi, int type);
 int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc);
 void f2fs_init_ino_entry_info(struct f2fs_sb_info *sbi);
 int __init f2fs_create_checkpoint_caches(void);
@@ -3186,12 +3339,19 @@ void f2fs_destroy_checkpoint_caches(void);
 /*
  * data.c
  */
-int f2fs_init_post_read_processing(void);
-void f2fs_destroy_post_read_processing(void);
+int __init f2fs_init_bioset(void);
+void f2fs_destroy_bioset(void);
+struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi, int npages, bool noio);
+int f2fs_init_bio_entry_cache(void);
+void f2fs_destroy_bio_entry_cache(void);
+void f2fs_submit_bio(struct f2fs_sb_info *sbi,
+				struct bio *bio, enum page_type type);
 void f2fs_submit_merged_write(struct f2fs_sb_info *sbi, enum page_type type);
 void f2fs_submit_merged_write_cond(struct f2fs_sb_info *sbi,
 				struct inode *inode, struct page *page,
 				nid_t ino, enum page_type type);
+void f2fs_submit_merged_ipu_write(struct f2fs_sb_info *sbi,
+					struct bio **bio, struct page *page);
 void f2fs_flush_merged_writes(struct f2fs_sb_info *sbi);
 int f2fs_submit_page_bio(struct f2fs_io_info *fio);
 int f2fs_merge_page_bio(struct f2fs_io_info *fio);
@@ -3219,8 +3379,14 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
 			int create, int flag);
 int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			u64 start, u64 len);
+int f2fs_encrypt_one_page(struct f2fs_io_info *fio);
 bool f2fs_should_update_inplace(struct inode *inode, struct f2fs_io_info *fio);
 bool f2fs_should_update_outplace(struct inode *inode, struct f2fs_io_info *fio);
+int f2fs_write_single_data_page(struct page *page, int *submitted,
+				struct bio **bio, sector_t *last_block,
+				struct writeback_control *wbc,
+				enum iostat_type io_type,
+				int compr_blocks);
 void f2fs_invalidate_page(struct page *page, unsigned int offset,
 			unsigned int length);
 int f2fs_release_page(struct page *page, gfp_t wait);
@@ -3230,6 +3396,10 @@ int f2fs_migrate_page(struct address_space *mapping, struct page *newpage,
 #endif
 bool f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len);
 void f2fs_clear_page_cache_dirty_tag(struct page *page);
+int f2fs_init_post_read_processing(void);
+void f2fs_destroy_post_read_processing(void);
+int f2fs_init_post_read_wq(struct f2fs_sb_info *sbi);
+void f2fs_destroy_post_read_wq(struct f2fs_sb_info *sbi);
 
 /*
  * gc.c
@@ -3276,6 +3446,7 @@ struct f2fs_stat_info {
 	int nr_discard_cmd;
 	unsigned int undiscard_blks;
 	int inline_xattr, inline_inode, inline_dir, append, update, orphans;
+	int compr_inode, compr_blocks;
 	int aw_cnt, max_aw_cnt, vw_cnt, max_vw_cnt;
 	unsigned int valid_count, valid_node_count, valid_inode_count, discard_blks;
 	unsigned int bimodal, avg_vblocks;
@@ -3307,7 +3478,7 @@ static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
 #define stat_inc_cp_count(si)		((si)->cp_count++)
 #define stat_inc_bg_cp_count(si)	((si)->bg_cp_count++)
 #define stat_inc_call_count(si)		((si)->call_count++)
-#define stat_inc_bggc_count(sbi)	((sbi)->bg_gc++)
+#define stat_inc_bggc_count(si)		((si)->bg_gc++)
 #define stat_io_skip_bggc_count(sbi)	((sbi)->io_skip_bggc++)
 #define stat_other_skip_bggc_count(sbi)	((sbi)->other_skip_bggc++)
 #define stat_inc_dirty_inode(sbi, type)	((sbi)->ndirty_inode[type]++)
@@ -3346,6 +3517,20 @@ static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
 		if (f2fs_has_inline_dentry(inode))			\
 			(atomic_dec(&F2FS_I_SB(inode)->inline_dir));	\
 	} while (0)
+#define stat_inc_compr_inode(inode)					\
+	do {								\
+		if (f2fs_compressed_file(inode))			\
+			(atomic_inc(&F2FS_I_SB(inode)->compr_inode));	\
+	} while (0)
+#define stat_dec_compr_inode(inode)					\
+	do {								\
+		if (f2fs_compressed_file(inode))			\
+			(atomic_dec(&F2FS_I_SB(inode)->compr_inode));	\
+	} while (0)
+#define stat_add_compr_blocks(inode, blocks)				\
+		(atomic_add(blocks, &F2FS_I_SB(inode)->compr_blocks))
+#define stat_sub_compr_blocks(inode, blocks)				\
+		(atomic_sub(blocks, &F2FS_I_SB(inode)->compr_blocks))
 #define stat_inc_meta_count(sbi, blkaddr)				\
 	do {								\
 		if (blkaddr < SIT_I(sbi)->sit_base_addr)		\
@@ -3363,13 +3548,9 @@ static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
 		((sbi)->block_count[(curseg)->alloc_type]++)
 #define stat_inc_inplace_blocks(sbi)					\
 		(atomic_inc(&(sbi)->inplace_count))
-#define stat_inc_atomic_write(inode)					\
-		(atomic_inc(&F2FS_I_SB(inode)->aw_cnt))
-#define stat_dec_atomic_write(inode)					\
-		(atomic_dec(&F2FS_I_SB(inode)->aw_cnt))
 #define stat_update_max_atomic_write(inode)				\
 	do {								\
-		int cur = atomic_read(&F2FS_I_SB(inode)->aw_cnt);	\
+		int cur = F2FS_I_SB(inode)->atomic_files;	\
 		int max = atomic_read(&F2FS_I_SB(inode)->max_aw_cnt);	\
 		if (cur > max)						\
 			atomic_set(&F2FS_I_SB(inode)->max_aw_cnt, cur);	\
@@ -3421,6 +3602,7 @@ int f2fs_build_stats(struct f2fs_sb_info *sbi);
 void f2fs_destroy_stats(struct f2fs_sb_info *sbi);
 void __init f2fs_create_root_stats(void);
 void f2fs_destroy_root_stats(void);
+void f2fs_update_sit_info(struct f2fs_sb_info *sbi);
 #else
 #define stat_inc_cp_count(si)				do { } while (0)
 #define stat_inc_bg_cp_count(si)			do { } while (0)
@@ -3430,8 +3612,8 @@ void f2fs_destroy_root_stats(void);
 #define stat_other_skip_bggc_count(sbi)			do { } while (0)
 #define stat_inc_dirty_inode(sbi, type)			do { } while (0)
 #define stat_dec_dirty_inode(sbi, type)			do { } while (0)
-#define stat_inc_total_hit(sb)				do { } while (0)
-#define stat_inc_rbtree_node_hit(sb)			do { } while (0)
+#define stat_inc_total_hit(sbi)				do { } while (0)
+#define stat_inc_rbtree_node_hit(sbi)			do { } while (0)
 #define stat_inc_largest_node_hit(sbi)			do { } while (0)
 #define stat_inc_cached_node_hit(sbi)			do { } while (0)
 #define stat_inc_inline_xattr(inode)			do { } while (0)
@@ -3440,6 +3622,10 @@ void f2fs_destroy_root_stats(void);
 #define stat_dec_inline_inode(inode)			do { } while (0)
 #define stat_inc_inline_dir(inode)			do { } while (0)
 #define stat_dec_inline_dir(inode)			do { } while (0)
+#define stat_inc_compr_inode(inode)			do { } while (0)
+#define stat_dec_compr_inode(inode)			do { } while (0)
+#define stat_add_compr_blocks(inode, blocks)		do { } while (0)
+#define stat_sub_compr_blocks(inode, blocks)		do { } while (0)
 #define stat_inc_atomic_write(inode)			do { } while (0)
 #define stat_dec_atomic_write(inode)			do { } while (0)
 #define stat_update_max_atomic_write(inode)		do { } while (0)
@@ -3459,6 +3645,7 @@ static inline int f2fs_build_stats(struct f2fs_sb_info *sbi) { return 0; }
 static inline void f2fs_destroy_stats(struct f2fs_sb_info *sbi) { }
 static inline void __init f2fs_create_root_stats(void) { }
 static inline void f2fs_destroy_root_stats(void) { }
+static inline void update_sit_info(struct f2fs_sb_info *sbi) {}
 #endif
 
 extern const struct file_operations f2fs_dir_operations;
@@ -3487,6 +3674,7 @@ void f2fs_truncate_inline_inode(struct inode *inode,
 int f2fs_read_inline_data(struct inode *inode, struct page *page);
 int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page);
 int f2fs_convert_inline_inode(struct inode *inode);
+int f2fs_try_convert_inline_dir(struct inode *dir, struct dentry *dentry);
 int f2fs_write_inline_data(struct inode *inode, struct page *page);
 bool f2fs_recover_inline_data(struct inode *inode, struct page *npage);
 struct f2fs_dir_entry *f2fs_find_in_inline_dir(struct inode *dir,
@@ -3579,7 +3767,91 @@ static inline void f2fs_set_encrypted_inode(struct inode *inode)
  */
 static inline bool f2fs_post_read_required(struct inode *inode)
 {
-	return f2fs_encrypted_file(inode) || fsverity_active(inode);
+	return f2fs_encrypted_file(inode) || fsverity_active(inode) ||
+		f2fs_compressed_file(inode);
+}
+
+/*
+ * compress.c
+ */
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+bool f2fs_is_compressed_page(struct page *page);
+struct page *f2fs_compress_control_page(struct page *page);
+int f2fs_prepare_compress_overwrite(struct inode *inode,
+			struct page **pagep, pgoff_t index, void **fsdata);
+bool f2fs_compress_write_end(struct inode *inode, void *fsdata,
+					pgoff_t index, unsigned copied);
+void f2fs_compress_write_end_io(struct bio *bio, struct page *page);
+bool f2fs_is_compress_backend_ready(struct inode *inode);
+void f2fs_decompress_pages(struct bio *bio, struct page *page, bool verity);
+bool f2fs_cluster_is_empty(struct compress_ctx *cc);
+bool f2fs_cluster_can_merge_page(struct compress_ctx *cc, pgoff_t index);
+void f2fs_compress_ctx_add_page(struct compress_ctx *cc, struct page *page);
+int f2fs_write_multi_pages(struct compress_ctx *cc,
+						int *submitted,
+						struct writeback_control *wbc,
+						enum iostat_type io_type);
+int f2fs_is_compressed_cluster(struct inode *inode, pgoff_t index);
+int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret,
+				unsigned nr_pages, sector_t *last_block_in_bio,
+				bool is_readahead, bool for_write);
+struct decompress_io_ctx *f2fs_alloc_dic(struct compress_ctx *cc);
+void f2fs_free_dic(struct decompress_io_ctx *dic);
+void f2fs_decompress_end_io(struct page **rpages,
+			unsigned int cluster_size, bool err, bool verity);
+int f2fs_init_compress_ctx(struct compress_ctx *cc);
+void f2fs_destroy_compress_ctx(struct compress_ctx *cc);
+void f2fs_init_compress_info(struct f2fs_sb_info *sbi);
+#else
+static inline bool f2fs_is_compressed_page(struct page *page) { return false; }
+static inline bool f2fs_is_compress_backend_ready(struct inode *inode)
+{
+	if (!f2fs_compressed_file(inode))
+		return true;
+	/* not support compression */
+	return false;
+}
+static inline struct page *f2fs_compress_control_page(struct page *page)
+{
+	WARN_ON_ONCE(1);
+	return ERR_PTR(-EINVAL);
+}
+#endif
+
+static inline void set_compress_context(struct inode *inode)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+
+	F2FS_I(inode)->i_compress_algorithm =
+			F2FS_OPTION(sbi).compress_algorithm;
+	F2FS_I(inode)->i_log_cluster_size =
+			F2FS_OPTION(sbi).compress_log_size;
+	F2FS_I(inode)->i_cluster_size =
+			1 << F2FS_I(inode)->i_log_cluster_size;
+	F2FS_I(inode)->i_flags |= F2FS_COMPR_FL;
+	set_inode_flag(inode, FI_COMPRESSED_FILE);
+	stat_inc_compr_inode(inode);
+	f2fs_mark_inode_dirty_sync(inode, true);
+}
+
+static inline u64 f2fs_disable_compressed_file(struct inode *inode)
+{
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+
+	if (!f2fs_compressed_file(inode))
+		return 0;
+	if (S_ISREG(inode->i_mode)) {
+		if (get_dirty_pages(inode))
+			return 1;
+		if (fi->i_compr_blocks)
+			return fi->i_compr_blocks;
+	}
+
+	fi->i_flags &= ~F2FS_COMPR_FL;
+	stat_dec_compr_inode(inode);
+	clear_inode_flag(inode, FI_COMPRESSED_FILE);
+	f2fs_mark_inode_dirty_sync(inode, true);
+	return 0;
 }
 
 #define F2FS_FEATURE_FUNCS(name, flagname) \
@@ -3600,6 +3872,7 @@ F2FS_FEATURE_FUNCS(lost_found, LOST_FOUND);
 F2FS_FEATURE_FUNCS(verity, VERITY);
 F2FS_FEATURE_FUNCS(sb_chksum, SB_CHKSUM);
 F2FS_FEATURE_FUNCS(casefold, CASEFOLD);
+F2FS_FEATURE_FUNCS(compression, COMPRESSION);
 
 #ifdef CONFIG_BLK_DEV_ZONED
 static inline bool f2fs_blkz_is_seq(struct f2fs_sb_info *sbi, int devi,
@@ -3654,31 +3927,49 @@ static inline bool f2fs_hw_is_readonly(struct f2fs_sb_info *sbi)
 	return false;
 }
 
-
-static inline void set_opt_mode(struct f2fs_sb_info *sbi, unsigned int mt)
+static inline bool f2fs_lfs_mode(struct f2fs_sb_info *sbi)
 {
-	clear_opt(sbi, ADAPTIVE);
-	clear_opt(sbi, LFS);
-
-	switch (mt) {
-	case F2FS_MOUNT_ADAPTIVE:
-		set_opt(sbi, ADAPTIVE);
-		break;
-	case F2FS_MOUNT_LFS:
-		set_opt(sbi, LFS);
-		break;
-	}
+	return F2FS_OPTION(sbi).fs_mode == FS_MODE_LFS;
 }
 
-static inline bool f2fs_may_encrypt(struct inode *inode)
+static inline bool f2fs_may_encrypt(struct inode *dir, struct inode *inode)
 {
 #ifdef CONFIG_FS_ENCRYPTION
+	struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
 	umode_t mode = inode->i_mode;
 
-	return (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode));
-#else
-	return false;
+	/*
+	 * If the directory encrypted or dummy encryption enabled,
+	 * then we should encrypt the inode.
+	 */
+	if (IS_ENCRYPTED(dir) || DUMMY_ENCRYPTION_ENABLED(sbi))
+		return (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode));
 #endif
+	return false;
+}
+
+static inline bool f2fs_may_compress(struct inode *inode)
+{
+	if (IS_SWAPFILE(inode) || f2fs_is_pinned_file(inode) ||
+				f2fs_is_atomic_file(inode) ||
+				f2fs_is_volatile_file(inode))
+		return false;
+	return S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode);
+}
+
+static inline void f2fs_i_compr_blocks_update(struct inode *inode,
+						u64 blocks, bool add)
+{
+	int diff = F2FS_I(inode)->i_cluster_size - blocks;
+
+	if (add) {
+		F2FS_I(inode)->i_compr_blocks += diff;
+		stat_add_compr_blocks(inode, diff);
+	} else {
+		F2FS_I(inode)->i_compr_blocks -= diff;
+		stat_sub_compr_blocks(inode, diff);
+	}
+	f2fs_mark_inode_dirty_sync(inode, true);
 }
 
 static inline int block_unaligned_IO(struct inode *inode,
@@ -3698,7 +3989,7 @@ static inline int allow_outplace_dio(struct inode *inode,
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	int rw = iov_iter_rw(iter);
 
-	return (test_opt(sbi, LFS) && (rw == WRITE) &&
+	return (f2fs_lfs_mode(sbi) && (rw == WRITE) &&
 				!block_unaligned_IO(inode, iocb, iter));
 }
 
@@ -3712,13 +4003,15 @@ static inline bool f2fs_force_buffered_io(struct inode *inode,
 		return true;
 	if (f2fs_is_multi_device(sbi))
 		return true;
+	if (f2fs_compressed_file(inode))
+		return true;
 	/*
 	 * for blkzoned device, fallback direct IO to buffered IO, so
 	 * all IOs can be serialized by log-structured write.
 	 */
 	if (f2fs_sb_has_blkzoned(sbi))
 		return true;
-	if (test_opt(sbi, LFS) && (rw == WRITE)) {
+	if (f2fs_lfs_mode(sbi) && (rw == WRITE)) {
 		if (block_unaligned_IO(inode, iocb, iter))
 			return true;
 		if (F2FS_IO_ALIGNED(sbi))
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 29bc0a542759..6ab8f621a3c5 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -50,8 +50,9 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
 	struct page *page = vmf->page;
 	struct inode *inode = file_inode(vmf->vma->vm_file);
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct dnode_of_data dn = { .node_changed = false };
-	int err;
+	struct dnode_of_data dn;
+	bool need_alloc = true;
+	int err = 0;
 
 	if (unlikely(f2fs_cp_error(sbi))) {
 		err = -EIO;
@@ -63,6 +64,26 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
 		goto err;
 	}
 
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	if (f2fs_compressed_file(inode)) {
+		int ret = f2fs_is_compressed_cluster(inode, page->index);
+
+		if (ret < 0) {
+			err = ret;
+			goto err;
+		} else if (ret) {
+			if (ret < F2FS_I(inode)->i_cluster_size) {
+				err = -EAGAIN;
+				goto err;
+			}
+			need_alloc = false;
+		}
+	}
+#endif
+	/* should do out of any locked page */
+	if (need_alloc)
+		f2fs_balance_fs(sbi, true);
+
 	sb_start_pagefault(inode->i_sb);
 
 	f2fs_bug_on(sbi, f2fs_has_inline_data(inode));
@@ -78,18 +99,27 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
 		goto out_sem;
 	}
 
-	/* block allocation */
-	__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true);
-	set_new_dnode(&dn, inode, NULL, NULL, 0);
-	err = f2fs_get_block(&dn, page->index);
-	f2fs_put_dnode(&dn);
-	__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false);
+	if (need_alloc) {
+		/* block allocation */
+		__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true);
+		set_new_dnode(&dn, inode, NULL, NULL, 0);
+		err = f2fs_get_block(&dn, page->index);
+		f2fs_put_dnode(&dn);
+		__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false);
+	}
+
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	if (!need_alloc) {
+		set_new_dnode(&dn, inode, NULL, NULL, 0);
+		err = f2fs_get_dnode_of_data(&dn, page->index, LOOKUP_NODE);
+		f2fs_put_dnode(&dn);
+	}
+#endif
 	if (err) {
 		unlock_page(page);
 		goto out_sem;
 	}
 
-	/* fill the page */
 	f2fs_wait_on_page_writeback(page, DATA, false, true);
 
 	/* wait for GCed page writeback via META_MAPPING */
@@ -120,8 +150,6 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
 out_sem:
 	up_read(&F2FS_I(inode)->i_mmap_sem);
 
-	f2fs_balance_fs(sbi, dn.node_changed);
-
 	sb_end_pagefault(inode->i_sb);
 err:
 	return block_page_mkwrite_return(err);
@@ -155,6 +183,8 @@ static inline enum cp_reason_type need_do_checkpoint(struct inode *inode)
 
 	if (!S_ISREG(inode->i_mode))
 		cp_reason = CP_NON_REGULAR;
+	else if (f2fs_compressed_file(inode))
+		cp_reason = CP_COMPRESSED;
 	else if (inode->i_nlink != 1)
 		cp_reason = CP_HARDLINK;
 	else if (is_sbi_flag_set(sbi, SBI_NEED_CP))
@@ -425,8 +455,7 @@ static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence)
 				data_ofs = (loff_t)pgofs << PAGE_SHIFT) {
 			block_t blkaddr;
 
-			blkaddr = datablock_addr(dn.inode,
-					dn.node_page, dn.ofs_in_node);
+			blkaddr = f2fs_data_blkaddr(&dn);
 
 			if (__is_valid_data_blkaddr(blkaddr) &&
 				!f2fs_is_valid_blkaddr(F2FS_I_SB(inode),
@@ -485,6 +514,9 @@ static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma)
 	if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
 		return -EIO;
 
+	if (!f2fs_is_compress_backend_ready(inode))
+		return -EOPNOTSUPP;
+
 	/* we don't need to use inline_data strictly */
 	err = f2fs_convert_inline_inode(inode);
 	if (err)
@@ -492,6 +524,7 @@ static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma)
 
 	file_accessed(file);
 	vma->vm_ops = &f2fs_file_vm_ops;
+	set_inode_flag(inode, FI_MMAP_FILE);
 	return 0;
 }
 
@@ -502,6 +535,9 @@ static int f2fs_file_open(struct inode *inode, struct file *filp)
 	if (err)
 		return err;
 
+	if (!f2fs_is_compress_backend_ready(inode))
+		return -EOPNOTSUPP;
+
 	err = fsverity_file_open(inode, filp);
 	if (err)
 		return err;
@@ -518,6 +554,9 @@ void f2fs_truncate_data_blocks_range(struct dnode_of_data *dn, int count)
 	int nr_free = 0, ofs = dn->ofs_in_node, len = count;
 	__le32 *addr;
 	int base = 0;
+	bool compressed_cluster = false;
+	int cluster_index = 0, valid_blocks = 0;
+	int cluster_size = F2FS_I(dn->inode)->i_cluster_size;
 
 	if (IS_INODE(dn->node_page) && f2fs_has_extra_attr(dn->inode))
 		base = get_extra_isize(dn->inode);
@@ -525,26 +564,43 @@ void f2fs_truncate_data_blocks_range(struct dnode_of_data *dn, int count)
 	raw_node = F2FS_NODE(dn->node_page);
 	addr = blkaddr_in_node(raw_node) + base + ofs;
 
-	for (; count > 0; count--, addr++, dn->ofs_in_node++) {
+	/* Assumption: truncateion starts with cluster */
+	for (; count > 0; count--, addr++, dn->ofs_in_node++, cluster_index++) {
 		block_t blkaddr = le32_to_cpu(*addr);
 
+		if (f2fs_compressed_file(dn->inode) &&
+					!(cluster_index & (cluster_size - 1))) {
+			if (compressed_cluster)
+				f2fs_i_compr_blocks_update(dn->inode,
+							valid_blocks, false);
+			compressed_cluster = (blkaddr == COMPRESS_ADDR);
+			valid_blocks = 0;
+		}
+
 		if (blkaddr == NULL_ADDR)
 			continue;
 
 		dn->data_blkaddr = NULL_ADDR;
 		f2fs_set_data_blkaddr(dn);
 
-		if (__is_valid_data_blkaddr(blkaddr) &&
-			!f2fs_is_valid_blkaddr(sbi, blkaddr,
+		if (__is_valid_data_blkaddr(blkaddr)) {
+			if (!f2fs_is_valid_blkaddr(sbi, blkaddr,
 					DATA_GENERIC_ENHANCE))
-			continue;
+				continue;
+			if (compressed_cluster)
+				valid_blocks++;
+		}
 
-		f2fs_invalidate_blocks(sbi, blkaddr);
 		if (dn->ofs_in_node == 0 && IS_INODE(dn->node_page))
 			clear_inode_flag(dn->inode, FI_FIRST_BLOCK_WRITTEN);
+
+		f2fs_invalidate_blocks(sbi, blkaddr);
 		nr_free++;
 	}
 
+	if (compressed_cluster)
+		f2fs_i_compr_blocks_update(dn->inode, valid_blocks, false);
+
 	if (nr_free) {
 		pgoff_t fofs;
 		/*
@@ -587,6 +643,9 @@ static int truncate_partial_data_page(struct inode *inode, u64 from,
 		return 0;
 	}
 
+	if (f2fs_compressed_file(inode))
+		return 0;
+
 	page = f2fs_get_lock_data_page(inode, index, true);
 	if (IS_ERR(page))
 		return PTR_ERR(page) == -ENOENT ? 0 : PTR_ERR(page);
@@ -602,7 +661,7 @@ truncate_out:
 	return 0;
 }
 
-int f2fs_truncate_blocks(struct inode *inode, u64 from, bool lock)
+static int do_truncate_blocks(struct inode *inode, u64 from, bool lock)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct dnode_of_data dn;
@@ -667,6 +726,28 @@ free_partial:
 	return err;
 }
 
+int f2fs_truncate_blocks(struct inode *inode, u64 from, bool lock)
+{
+	u64 free_from = from;
+
+	/*
+	 * for compressed file, only support cluster size
+	 * aligned truncation.
+	 */
+	if (f2fs_compressed_file(inode)) {
+		size_t cluster_shift = PAGE_SHIFT +
+					F2FS_I(inode)->i_log_cluster_size;
+		size_t cluster_mask = (1 << cluster_shift) - 1;
+
+		free_from = from >> cluster_shift;
+		if (from & cluster_mask)
+			free_from++;
+		free_from <<= cluster_shift;
+	}
+
+	return do_truncate_blocks(inode, free_from, lock);
+}
+
 int f2fs_truncate(struct inode *inode)
 {
 	int err;
@@ -681,7 +762,7 @@ int f2fs_truncate(struct inode *inode)
 	trace_f2fs_truncate(inode);
 
 	if (time_to_inject(F2FS_I_SB(inode), FAULT_TRUNCATE)) {
-		f2fs_show_injection_info(FAULT_TRUNCATE);
+		f2fs_show_injection_info(F2FS_I_SB(inode), FAULT_TRUNCATE);
 		return -EIO;
 	}
 
@@ -718,6 +799,8 @@ int f2fs_getattr(const struct path *path, struct kstat *stat,
 	}
 
 	flags = fi->i_flags;
+	if (flags & F2FS_COMPR_FL)
+		stat->attributes |= STATX_ATTR_COMPRESSED;
 	if (flags & F2FS_APPEND_FL)
 		stat->attributes |= STATX_ATTR_APPEND;
 	if (IS_ENCRYPTED(inode))
@@ -726,11 +809,15 @@ int f2fs_getattr(const struct path *path, struct kstat *stat,
 		stat->attributes |= STATX_ATTR_IMMUTABLE;
 	if (flags & F2FS_NODUMP_FL)
 		stat->attributes |= STATX_ATTR_NODUMP;
+	if (IS_VERITY(inode))
+		stat->attributes |= STATX_ATTR_VERITY;
 
-	stat->attributes_mask |= (STATX_ATTR_APPEND |
+	stat->attributes_mask |= (STATX_ATTR_COMPRESSED |
+				  STATX_ATTR_APPEND |
 				  STATX_ATTR_ENCRYPTED |
 				  STATX_ATTR_IMMUTABLE |
-				  STATX_ATTR_NODUMP);
+				  STATX_ATTR_NODUMP |
+				  STATX_ATTR_VERITY);
 
 	generic_fillattr(inode, stat);
 
@@ -751,18 +838,12 @@ static void __setattr_copy(struct inode *inode, const struct iattr *attr)
 		inode->i_uid = attr->ia_uid;
 	if (ia_valid & ATTR_GID)
 		inode->i_gid = attr->ia_gid;
-	if (ia_valid & ATTR_ATIME) {
-		inode->i_atime = timestamp_truncate(attr->ia_atime,
-						  inode);
-	}
-	if (ia_valid & ATTR_MTIME) {
-		inode->i_mtime = timestamp_truncate(attr->ia_mtime,
-						  inode);
-	}
-	if (ia_valid & ATTR_CTIME) {
-		inode->i_ctime = timestamp_truncate(attr->ia_ctime,
-						  inode);
-	}
+	if (ia_valid & ATTR_ATIME)
+		inode->i_atime = attr->ia_atime;
+	if (ia_valid & ATTR_MTIME)
+		inode->i_mtime = attr->ia_mtime;
+	if (ia_valid & ATTR_CTIME)
+		inode->i_ctime = attr->ia_ctime;
 	if (ia_valid & ATTR_MODE) {
 		umode_t mode = attr->ia_mode;
 
@@ -783,6 +864,10 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
 	if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
 		return -EIO;
 
+	if ((attr->ia_valid & ATTR_SIZE) &&
+		!f2fs_is_compress_backend_ready(inode))
+		return -EOPNOTSUPP;
+
 	err = setattr_prepare(dentry, attr);
 	if (err)
 		return err;
@@ -853,10 +938,10 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
 		if (err)
 			return err;
 
-		down_write(&F2FS_I(inode)->i_sem);
+		spin_lock(&F2FS_I(inode)->i_size_lock);
 		inode->i_mtime = inode->i_ctime = current_time(inode);
 		F2FS_I(inode)->last_disk_size = i_size_read(inode);
-		up_write(&F2FS_I(inode)->i_sem);
+		spin_unlock(&F2FS_I(inode)->i_size_lock);
 	}
 
 	__setattr_copy(inode, attr);
@@ -1023,8 +1108,8 @@ next_dnode:
 	} else if (ret == -ENOENT) {
 		if (dn.max_level == 0)
 			return -ENOENT;
-		done = min((pgoff_t)ADDRS_PER_BLOCK(inode) - dn.ofs_in_node,
-									len);
+		done = min((pgoff_t)ADDRS_PER_BLOCK(inode) -
+						dn.ofs_in_node, len);
 		blkaddr += done;
 		do_replace += done;
 		goto next;
@@ -1033,8 +1118,7 @@ next_dnode:
 	done = min((pgoff_t)ADDRS_PER_PAGE(dn.node_page, inode) -
 							dn.ofs_in_node, len);
 	for (i = 0; i < done; i++, blkaddr++, do_replace++, dn.ofs_in_node++) {
-		*blkaddr = datablock_addr(dn.inode,
-					dn.node_page, dn.ofs_in_node);
+		*blkaddr = f2fs_data_blkaddr(&dn);
 
 		if (__is_valid_data_blkaddr(*blkaddr) &&
 			!f2fs_is_valid_blkaddr(sbi, *blkaddr,
@@ -1045,7 +1129,7 @@ next_dnode:
 
 		if (!f2fs_is_checkpointed_data(sbi, *blkaddr)) {
 
-			if (test_opt(sbi, LFS)) {
+			if (f2fs_lfs_mode(sbi)) {
 				f2fs_put_dnode(&dn);
 				return -EOPNOTSUPP;
 			}
@@ -1123,8 +1207,7 @@ static int __clone_blkaddrs(struct inode *src_inode, struct inode *dst_inode,
 				ADDRS_PER_PAGE(dn.node_page, dst_inode) -
 						dn.ofs_in_node, len - i);
 			do {
-				dn.data_blkaddr = datablock_addr(dn.inode,
-						dn.node_page, dn.ofs_in_node);
+				dn.data_blkaddr = f2fs_data_blkaddr(&dn);
 				f2fs_truncate_data_blocks_range(&dn, 1);
 
 				if (do_replace[i]) {
@@ -1139,7 +1222,7 @@ static int __clone_blkaddrs(struct inode *src_inode, struct inode *dst_inode,
 				}
 				dn.ofs_in_node++;
 				i++;
-				new_size = (dst + i) << PAGE_SHIFT;
+				new_size = (loff_t)(dst + i) << PAGE_SHIFT;
 				if (dst_inode->i_size < new_size)
 					f2fs_i_size_write(dst_inode, new_size);
 			} while (--ilen && (do_replace[i] || blkaddr[i] == NULL_ADDR));
@@ -1187,13 +1270,13 @@ static int __exchange_data_block(struct inode *src_inode,
 
 		src_blkaddr = f2fs_kvzalloc(F2FS_I_SB(src_inode),
 					array_size(olen, sizeof(block_t)),
-					GFP_KERNEL);
+					GFP_NOFS);
 		if (!src_blkaddr)
 			return -ENOMEM;
 
 		do_replace = f2fs_kvzalloc(F2FS_I_SB(src_inode),
 					array_size(olen, sizeof(int)),
-					GFP_KERNEL);
+					GFP_NOFS);
 		if (!do_replace) {
 			kvfree(src_blkaddr);
 			return -ENOMEM;
@@ -1300,8 +1383,7 @@ static int f2fs_do_zero_range(struct dnode_of_data *dn, pgoff_t start,
 	int ret;
 
 	for (; index < end; index++, dn->ofs_in_node++) {
-		if (datablock_addr(dn->inode, dn->node_page,
-					dn->ofs_in_node) == NULL_ADDR)
+		if (f2fs_data_blkaddr(dn) == NULL_ADDR)
 			count++;
 	}
 
@@ -1312,8 +1394,7 @@ static int f2fs_do_zero_range(struct dnode_of_data *dn, pgoff_t start,
 
 	dn->ofs_in_node = ofs_in_node;
 	for (index = start; index < end; index++, dn->ofs_in_node++) {
-		dn->data_blkaddr = datablock_addr(dn->inode,
-					dn->node_page, dn->ofs_in_node);
+		dn->data_blkaddr = f2fs_data_blkaddr(dn);
 		/*
 		 * f2fs_reserve_new_blocks will not guarantee entire block
 		 * allocation.
@@ -1545,12 +1626,44 @@ static int expand_inode_data(struct inode *inode, loff_t offset,
 	if (off_end)
 		map.m_len++;
 
-	if (f2fs_is_pinned_file(inode))
-		map.m_seg_type = CURSEG_COLD_DATA;
+	if (!map.m_len)
+		return 0;
+
+	if (f2fs_is_pinned_file(inode)) {
+		block_t len = (map.m_len >> sbi->log_blocks_per_seg) <<
+					sbi->log_blocks_per_seg;
+		block_t done = 0;
+
+		if (map.m_len % sbi->blocks_per_seg)
+			len += sbi->blocks_per_seg;
+
+		map.m_len = sbi->blocks_per_seg;
+next_alloc:
+		if (has_not_enough_free_secs(sbi, 0,
+			GET_SEC_FROM_SEG(sbi, overprovision_segments(sbi)))) {
+			down_write(&sbi->gc_lock);
+			err = f2fs_gc(sbi, true, false, NULL_SEGNO);
+			if (err && err != -ENODATA && err != -EAGAIN)
+				goto out_err;
+		}
+
+		down_write(&sbi->pin_sem);
+		map.m_seg_type = CURSEG_COLD_DATA_PINNED;
+		f2fs_allocate_new_segments(sbi, CURSEG_COLD_DATA);
+		err = f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_DIO);
+		up_write(&sbi->pin_sem);
 
-	err = f2fs_map_blocks(inode, &map, 1, (f2fs_is_pinned_file(inode) ?
-						F2FS_GET_BLOCK_PRE_DIO :
-						F2FS_GET_BLOCK_PRE_AIO));
+		done += map.m_len;
+		len -= map.m_len;
+		map.m_lblk += map.m_len;
+		if (!err && len)
+			goto next_alloc;
+
+		map.m_len = done;
+	} else {
+		err = f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_AIO);
+	}
+out_err:
 	if (err) {
 		pgoff_t last_off;
 
@@ -1586,6 +1699,8 @@ static long f2fs_fallocate(struct file *file, int mode,
 		return -EIO;
 	if (!f2fs_is_checkpoint_ready(F2FS_I_SB(inode)))
 		return -ENOSPC;
+	if (!f2fs_is_compress_backend_ready(inode))
+		return -EOPNOTSUPP;
 
 	/* f2fs only support ->fallocate for regular file */
 	if (!S_ISREG(inode->i_mode))
@@ -1595,6 +1710,11 @@ static long f2fs_fallocate(struct file *file, int mode,
 		(mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE)))
 		return -EOPNOTSUPP;
 
+	if (f2fs_compressed_file(inode) &&
+		(mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE |
+			FALLOC_FL_ZERO_RANGE | FALLOC_FL_INSERT_RANGE)))
+		return -EOPNOTSUPP;
+
 	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE |
 			FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE |
 			FALLOC_FL_INSERT_RANGE))
@@ -1672,19 +1792,50 @@ static int f2fs_file_flush(struct file *file, fl_owner_t id)
 static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
 {
 	struct f2fs_inode_info *fi = F2FS_I(inode);
+	u32 masked_flags = fi->i_flags & mask;
+
+	f2fs_bug_on(F2FS_I_SB(inode), (iflags & ~mask));
 
 	/* Is it quota file? Do not allow user to mess with it */
 	if (IS_NOQUOTA(inode))
 		return -EPERM;
 
-	if ((iflags ^ fi->i_flags) & F2FS_CASEFOLD_FL) {
+	if ((iflags ^ masked_flags) & F2FS_CASEFOLD_FL) {
 		if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
 			return -EOPNOTSUPP;
 		if (!f2fs_empty_dir(inode))
 			return -ENOTEMPTY;
 	}
 
+	if (iflags & (F2FS_COMPR_FL | F2FS_NOCOMP_FL)) {
+		if (!f2fs_sb_has_compression(F2FS_I_SB(inode)))
+			return -EOPNOTSUPP;
+		if ((iflags & F2FS_COMPR_FL) && (iflags & F2FS_NOCOMP_FL))
+			return -EINVAL;
+	}
+
+	if ((iflags ^ masked_flags) & F2FS_COMPR_FL) {
+		if (masked_flags & F2FS_COMPR_FL) {
+			if (f2fs_disable_compressed_file(inode))
+				return -EINVAL;
+		}
+		if (iflags & F2FS_NOCOMP_FL)
+			return -EINVAL;
+		if (iflags & F2FS_COMPR_FL) {
+			if (!f2fs_may_compress(inode))
+				return -EINVAL;
+
+			set_compress_context(inode);
+		}
+	}
+	if ((iflags ^ masked_flags) & F2FS_NOCOMP_FL) {
+		if (masked_flags & F2FS_COMPR_FL)
+			return -EINVAL;
+	}
+
 	fi->i_flags = iflags | (fi->i_flags & ~mask);
+	f2fs_bug_on(F2FS_I_SB(inode), (fi->i_flags & F2FS_COMPR_FL) &&
+					(fi->i_flags & F2FS_NOCOMP_FL));
 
 	if (fi->i_flags & F2FS_PROJINHERIT_FL)
 		set_inode_flag(inode, FI_PROJ_INHERIT);
@@ -1710,11 +1861,13 @@ static const struct {
 	u32 iflag;
 	u32 fsflag;
 } f2fs_fsflags_map[] = {
+	{ F2FS_COMPR_FL,	FS_COMPR_FL },
 	{ F2FS_SYNC_FL,		FS_SYNC_FL },
 	{ F2FS_IMMUTABLE_FL,	FS_IMMUTABLE_FL },
 	{ F2FS_APPEND_FL,	FS_APPEND_FL },
 	{ F2FS_NODUMP_FL,	FS_NODUMP_FL },
 	{ F2FS_NOATIME_FL,	FS_NOATIME_FL },
+	{ F2FS_NOCOMP_FL,	FS_NOCOMP_FL },
 	{ F2FS_INDEX_FL,	FS_INDEX_FL },
 	{ F2FS_DIRSYNC_FL,	FS_DIRSYNC_FL },
 	{ F2FS_PROJINHERIT_FL,	FS_PROJINHERIT_FL },
@@ -1722,11 +1875,13 @@ static const struct {
 };
 
 #define F2FS_GETTABLE_FS_FL (		\
+		FS_COMPR_FL |		\
 		FS_SYNC_FL |		\
 		FS_IMMUTABLE_FL |	\
 		FS_APPEND_FL |		\
 		FS_NODUMP_FL |		\
 		FS_NOATIME_FL |		\
+		FS_NOCOMP_FL |		\
 		FS_INDEX_FL |		\
 		FS_DIRSYNC_FL |		\
 		FS_PROJINHERIT_FL |	\
@@ -1737,11 +1892,13 @@ static const struct {
 		FS_CASEFOLD_FL)
 
 #define F2FS_SETTABLE_FS_FL (		\
+		FS_COMPR_FL |		\
 		FS_SYNC_FL |		\
 		FS_IMMUTABLE_FL |	\
 		FS_APPEND_FL |		\
 		FS_NODUMP_FL |		\
 		FS_NOATIME_FL |		\
+		FS_NOCOMP_FL |		\
 		FS_DIRSYNC_FL |		\
 		FS_PROJINHERIT_FL |	\
 		FS_CASEFOLD_FL)
@@ -1862,6 +2019,8 @@ static int f2fs_ioc_start_atomic_write(struct file *filp)
 
 	inode_lock(inode);
 
+	f2fs_disable_compressed_file(inode);
+
 	if (f2fs_is_atomic_file(inode)) {
 		if (is_inode_flag_set(inode, FI_ATOMIC_REVOKE_REQUEST))
 			ret = -EINVAL;
@@ -1890,6 +2049,7 @@ static int f2fs_ioc_start_atomic_write(struct file *filp)
 	spin_lock(&sbi->inode_lock[ATOMIC_FILE]);
 	if (list_empty(&fi->inmem_ilist))
 		list_add_tail(&fi->inmem_ilist, &sbi->inode_list[ATOMIC_FILE]);
+	sbi->atomic_files++;
 	spin_unlock(&sbi->inode_lock[ATOMIC_FILE]);
 
 	/* add inode in inmem_list first and set atomic_file */
@@ -1899,7 +2059,6 @@ static int f2fs_ioc_start_atomic_write(struct file *filp)
 
 	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
 	F2FS_I(inode)->inmem_task = current;
-	stat_inc_atomic_write(inode);
 	stat_update_max_atomic_write(inode);
 out:
 	inode_unlock(inode);
@@ -2267,6 +2426,14 @@ static int f2fs_ioc_get_encryption_key_status(struct file *filp,
 	return fscrypt_ioctl_get_key_status(filp, (void __user *)arg);
 }
 
+static int f2fs_ioc_get_encryption_nonce(struct file *filp, unsigned long arg)
+{
+	if (!f2fs_sb_has_encrypt(F2FS_I_SB(file_inode(filp))))
+		return -EOPNOTSUPP;
+
+	return fscrypt_ioctl_get_nonce(filp, (void __user *)arg);
+}
+
 static int f2fs_ioc_gc(struct file *filp, unsigned long arg)
 {
 	struct inode *inode = file_inode(filp);
@@ -2288,12 +2455,12 @@ static int f2fs_ioc_gc(struct file *filp, unsigned long arg)
 		return ret;
 
 	if (!sync) {
-		if (!mutex_trylock(&sbi->gc_mutex)) {
+		if (!down_write_trylock(&sbi->gc_lock)) {
 			ret = -EBUSY;
 			goto out;
 		}
 	} else {
-		mutex_lock(&sbi->gc_mutex);
+		down_write(&sbi->gc_lock);
 	}
 
 	ret = f2fs_gc(sbi, sync, true, NULL_SEGNO);
@@ -2331,12 +2498,12 @@ static int f2fs_ioc_gc_range(struct file *filp, unsigned long arg)
 
 do_more:
 	if (!range.sync) {
-		if (!mutex_trylock(&sbi->gc_mutex)) {
+		if (!down_write_trylock(&sbi->gc_lock)) {
 			ret = -EBUSY;
 			goto out;
 		}
 	} else {
-		mutex_lock(&sbi->gc_mutex);
+		down_write(&sbi->gc_lock);
 	}
 
 	ret = f2fs_gc(sbi, range.sync, true, GET_SEGNO(sbi, range.start));
@@ -2767,7 +2934,7 @@ static int f2fs_ioc_flush_device(struct file *filp, unsigned long arg)
 	end_segno = min(start_segno + range.segments, dev_end_segno);
 
 	while (start_segno < end_segno) {
-		if (!mutex_trylock(&sbi->gc_mutex)) {
+		if (!down_write_trylock(&sbi->gc_lock)) {
 			ret = -EBUSY;
 			goto out;
 		}
@@ -3062,10 +3229,16 @@ static int f2fs_ioc_set_pin_file(struct file *filp, unsigned long arg)
 		ret = -EAGAIN;
 		goto out;
 	}
+
 	ret = f2fs_convert_inline_inode(inode);
 	if (ret)
 		goto out;
 
+	if (f2fs_disable_compressed_file(inode)) {
+		ret = -EOPNOTSUPP;
+		goto out;
+	}
+
 	set_inode_flag(inode, FI_PIN_FILE);
 	ret = F2FS_I(inode)->i_gc_failures[GC_FAILURE_PIN];
 done:
@@ -3231,6 +3404,21 @@ out:
 	return err;
 }
 
+static int f2fs_get_compress_blocks(struct file *filp, unsigned long arg)
+{
+	struct inode *inode = file_inode(filp);
+	__u64 blocks;
+
+	if (!f2fs_sb_has_compression(F2FS_I_SB(inode)))
+		return -EOPNOTSUPP;
+
+	if (!f2fs_compressed_file(inode))
+		return -EINVAL;
+
+	blocks = F2FS_I(inode)->i_compr_blocks;
+	return put_user(blocks, (u64 __user *)arg);
+}
+
 long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	if (unlikely(f2fs_cp_error(F2FS_I_SB(file_inode(filp)))))
@@ -3275,6 +3463,8 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		return f2fs_ioc_remove_encryption_key_all_users(filp, arg);
 	case FS_IOC_GET_ENCRYPTION_KEY_STATUS:
 		return f2fs_ioc_get_encryption_key_status(filp, arg);
+	case FS_IOC_GET_ENCRYPTION_NONCE:
+		return f2fs_ioc_get_encryption_nonce(filp, arg);
 	case F2FS_IOC_GARBAGE_COLLECT:
 		return f2fs_ioc_gc(filp, arg);
 	case F2FS_IOC_GARBAGE_COLLECT_RANGE:
@@ -3309,11 +3499,24 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		return f2fs_get_volume_name(filp, arg);
 	case F2FS_IOC_SET_VOLUME_NAME:
 		return f2fs_set_volume_name(filp, arg);
+	case F2FS_IOC_GET_COMPRESS_BLOCKS:
+		return f2fs_get_compress_blocks(filp, arg);
 	default:
 		return -ENOTTY;
 	}
 }
 
+static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+
+	if (!f2fs_is_compress_backend_ready(inode))
+		return -EOPNOTSUPP;
+
+	return generic_file_read_iter(iocb, iter);
+}
+
 static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
@@ -3325,6 +3528,11 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 		goto out;
 	}
 
+	if (!f2fs_is_compress_backend_ready(inode)) {
+		ret = -EOPNOTSUPP;
+		goto out;
+	}
+
 	if (iocb->ki_flags & IOCB_NOWAIT) {
 		if (!inode_trylock(inode)) {
 			ret = -EAGAIN;
@@ -3353,18 +3561,41 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 				ret = -EAGAIN;
 				goto out;
 			}
-		} else {
-			preallocated = true;
-			target_size = iocb->ki_pos + iov_iter_count(from);
+			goto write;
+		}
 
-			err = f2fs_preallocate_blocks(iocb, from);
-			if (err) {
-				clear_inode_flag(inode, FI_NO_PREALLOC);
-				inode_unlock(inode);
-				ret = err;
-				goto out;
-			}
+		if (is_inode_flag_set(inode, FI_NO_PREALLOC))
+			goto write;
+
+		if (iocb->ki_flags & IOCB_DIRECT) {
+			/*
+			 * Convert inline data for Direct I/O before entering
+			 * f2fs_direct_IO().
+			 */
+			err = f2fs_convert_inline_inode(inode);
+			if (err)
+				goto out_err;
+			/*
+			 * If force_buffere_io() is true, we have to allocate
+			 * blocks all the time, since f2fs_direct_IO will fall
+			 * back to buffered IO.
+			 */
+			if (!f2fs_force_buffered_io(inode, iocb, from) &&
+					allow_outplace_dio(inode, iocb, from))
+				goto write;
 		}
+		preallocated = true;
+		target_size = iocb->ki_pos + iov_iter_count(from);
+
+		err = f2fs_preallocate_blocks(iocb, from);
+		if (err) {
+out_err:
+			clear_inode_flag(inode, FI_NO_PREALLOC);
+			inode_unlock(inode);
+			ret = err;
+			goto out;
+		}
+write:
 		ret = __generic_file_write_iter(iocb, from);
 		clear_inode_flag(inode, FI_NO_PREALLOC);
 
@@ -3403,6 +3634,7 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case F2FS_IOC_RELEASE_VOLATILE_WRITE:
 	case F2FS_IOC_ABORT_VOLATILE_WRITE:
 	case F2FS_IOC_SHUTDOWN:
+	case FITRIM:
 	case F2FS_IOC_SET_ENCRYPTION_POLICY:
 	case F2FS_IOC_GET_ENCRYPTION_PWSALT:
 	case F2FS_IOC_GET_ENCRYPTION_POLICY:
@@ -3411,6 +3643,7 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case FS_IOC_REMOVE_ENCRYPTION_KEY:
 	case FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS:
 	case FS_IOC_GET_ENCRYPTION_KEY_STATUS:
+	case FS_IOC_GET_ENCRYPTION_NONCE:
 	case F2FS_IOC_GARBAGE_COLLECT:
 	case F2FS_IOC_GARBAGE_COLLECT_RANGE:
 	case F2FS_IOC_WRITE_CHECKPOINT:
@@ -3428,6 +3661,7 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case FS_IOC_MEASURE_VERITY:
 	case F2FS_IOC_GET_VOLUME_NAME:
 	case F2FS_IOC_SET_VOLUME_NAME:
+	case F2FS_IOC_GET_COMPRESS_BLOCKS:
 		break;
 	default:
 		return -ENOIOCTLCMD;
@@ -3438,7 +3672,7 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
 const struct file_operations f2fs_file_operations = {
 	.llseek		= f2fs_llseek,
-	.read_iter	= generic_file_read_iter,
+	.read_iter	= f2fs_file_read_iter,
 	.write_iter	= f2fs_file_write_iter,
 	.open		= f2fs_file_open,
 	.release	= f2fs_release_file,
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 5877bd729689..26248c8936db 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -31,6 +31,8 @@ static int gc_thread_func(void *data)
 
 	set_freezable();
 	do {
+		bool sync_mode;
+
 		wait_event_interruptible_timeout(*wq,
 				kthread_should_stop() || freezing(current) ||
 				gc_th->gc_wake,
@@ -54,7 +56,7 @@ static int gc_thread_func(void *data)
 		}
 
 		if (time_to_inject(sbi, FAULT_CHECKPOINT)) {
-			f2fs_show_injection_info(FAULT_CHECKPOINT);
+			f2fs_show_injection_info(sbi, FAULT_CHECKPOINT);
 			f2fs_stop_checkpoint(sbi, false);
 		}
 
@@ -78,18 +80,18 @@ static int gc_thread_func(void *data)
 		 */
 		if (sbi->gc_mode == GC_URGENT) {
 			wait_ms = gc_th->urgent_sleep_time;
-			mutex_lock(&sbi->gc_mutex);
+			down_write(&sbi->gc_lock);
 			goto do_gc;
 		}
 
-		if (!mutex_trylock(&sbi->gc_mutex)) {
+		if (!down_write_trylock(&sbi->gc_lock)) {
 			stat_other_skip_bggc_count(sbi);
 			goto next;
 		}
 
 		if (!is_idle(sbi, GC_TIME)) {
 			increase_sleep_time(gc_th, &wait_ms);
-			mutex_unlock(&sbi->gc_mutex);
+			up_write(&sbi->gc_lock);
 			stat_io_skip_bggc_count(sbi);
 			goto next;
 		}
@@ -99,17 +101,19 @@ static int gc_thread_func(void *data)
 		else
 			increase_sleep_time(gc_th, &wait_ms);
 do_gc:
-		stat_inc_bggc_count(sbi);
+		stat_inc_bggc_count(sbi->stat_info);
+
+		sync_mode = F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_SYNC;
 
 		/* if return value is not zero, no victim was selected */
-		if (f2fs_gc(sbi, test_opt(sbi, FORCE_FG_GC), true, NULL_SEGNO))
+		if (f2fs_gc(sbi, sync_mode, true, NULL_SEGNO))
 			wait_ms = gc_th->no_gc_sleep_time;
 
 		trace_f2fs_background_gc(sbi->sb, wait_ms,
 				prefree_segments(sbi), free_segments(sbi));
 
 		/* balancing f2fs's metadata periodically */
-		f2fs_balance_fs_bg(sbi);
+		f2fs_balance_fs_bg(sbi, true);
 next:
 		sb_end_write(sbi->sb);
 
@@ -192,7 +196,10 @@ static void select_policy(struct f2fs_sb_info *sbi, int gc_type,
 		p->ofs_unit = sbi->segs_per_sec;
 	}
 
-	/* we need to check every dirty segments in the FG_GC case */
+	/*
+	 * adjust candidates range, should select all dirty segments for
+	 * foreground GC and urgent GC cases.
+	 */
 	if (gc_type != FG_GC &&
 			(sbi->gc_mode != GC_URGENT) &&
 			p->max_search > sbi->max_victim_search)
@@ -634,7 +641,7 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
 	}
 
 	*nofs = ofs_of_node(node_page);
-	source_blkaddr = datablock_addr(NULL, node_page, ofs_in_node);
+	source_blkaddr = data_blkaddr(NULL, node_page, ofs_in_node);
 	f2fs_put_page(node_page, 1);
 
 	if (source_blkaddr != blkaddr) {
@@ -762,7 +769,7 @@ static int move_data_block(struct inode *inode, block_t bidx,
 	struct page *page, *mpage;
 	block_t newaddr;
 	int err = 0;
-	bool lfs_mode = test_opt(fio.sbi, LFS);
+	bool lfs_mode = f2fs_lfs_mode(fio.sbi);
 
 	/* do not read out */
 	page = f2fs_grab_cache_page(inode->i_mapping, bidx, false);
@@ -970,7 +977,8 @@ retry:
 		if (err) {
 			clear_cold_data(page);
 			if (err == -ENOMEM) {
-				congestion_wait(BLK_RW_ASYNC, HZ/50);
+				congestion_wait(BLK_RW_ASYNC,
+						DEFAULT_IO_TIMEOUT);
 				goto retry;
 			}
 			if (is_dirty)
@@ -1012,8 +1020,14 @@ next_step:
 		block_t start_bidx;
 		nid_t nid = le32_to_cpu(entry->nid);
 
-		/* stop BG_GC if there is not enough free sections. */
-		if (gc_type == BG_GC && has_not_enough_free_secs(sbi, 0, 0))
+		/*
+		 * stop BG_GC if there is not enough free sections.
+		 * Or, stop GC if the segment becomes fully valid caused by
+		 * race condition along with SSR block allocation.
+		 */
+		if ((gc_type == BG_GC && has_not_enough_free_secs(sbi, 0, 0)) ||
+				get_valid_blocks(sbi, segno, true) ==
+							BLKS_PER_SEC(sbi))
 			return submitted;
 
 		if (check_valid_map(sbi, segno, off) == 0)
@@ -1043,8 +1057,10 @@ next_step:
 
 		if (phase == 3) {
 			inode = f2fs_iget(sb, dni.ino);
-			if (IS_ERR(inode) || is_bad_inode(inode))
+			if (IS_ERR(inode) || is_bad_inode(inode)) {
+				set_sbi_flag(sbi, SBI_NEED_FSCK);
 				continue;
+			}
 
 			if (!down_write_trylock(
 				&F2FS_I(inode)->i_gc_rwsem[WRITE])) {
@@ -1195,7 +1211,7 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
 
 		if (get_valid_blocks(sbi, segno, false) == 0)
 			goto freed;
-		if (__is_large_section(sbi) &&
+		if (gc_type == BG_GC && __is_large_section(sbi) &&
 				migrated >= sbi->migration_granularity)
 			goto skip;
 		if (!PageUptodate(sum_page) || unlikely(f2fs_cp_error(sbi)))
@@ -1225,12 +1241,12 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
 							segno, gc_type);
 
 		stat_inc_seg_count(sbi, type, gc_type);
+		migrated++;
 
 freed:
 		if (gc_type == FG_GC &&
 				get_valid_blocks(sbi, segno, false) == 0)
 			seg_freed++;
-		migrated++;
 
 		if (__is_large_section(sbi) && segno + 1 < end_segno)
 			sbi->next_victim_seg[gc_type] = segno + 1;
@@ -1362,7 +1378,7 @@ stop:
 				reserved_segments(sbi),
 				prefree_segments(sbi));
 
-	mutex_unlock(&sbi->gc_mutex);
+	up_write(&sbi->gc_lock);
 
 	put_gc_inode(&gc_list);
 
@@ -1401,9 +1417,9 @@ static int free_segment_range(struct f2fs_sb_info *sbi, unsigned int start,
 			.iroot = RADIX_TREE_INIT(gc_list.iroot, GFP_NOFS),
 		};
 
-		mutex_lock(&sbi->gc_mutex);
+		down_write(&sbi->gc_lock);
 		do_garbage_collect(sbi, segno, &gc_list, FG_GC);
-		mutex_unlock(&sbi->gc_mutex);
+		up_write(&sbi->gc_lock);
 		put_gc_inode(&gc_list);
 
 		if (get_valid_blocks(sbi, segno, true))
@@ -1426,22 +1442,40 @@ static int free_segment_range(struct f2fs_sb_info *sbi, unsigned int start,
 static void update_sb_metadata(struct f2fs_sb_info *sbi, int secs)
 {
 	struct f2fs_super_block *raw_sb = F2FS_RAW_SUPER(sbi);
-	int section_count = le32_to_cpu(raw_sb->section_count);
-	int segment_count = le32_to_cpu(raw_sb->segment_count);
-	int segment_count_main = le32_to_cpu(raw_sb->segment_count_main);
-	long long block_count = le64_to_cpu(raw_sb->block_count);
+	int section_count;
+	int segment_count;
+	int segment_count_main;
+	long long block_count;
 	int segs = secs * sbi->segs_per_sec;
 
+	down_write(&sbi->sb_lock);
+
+	section_count = le32_to_cpu(raw_sb->section_count);
+	segment_count = le32_to_cpu(raw_sb->segment_count);
+	segment_count_main = le32_to_cpu(raw_sb->segment_count_main);
+	block_count = le64_to_cpu(raw_sb->block_count);
+
 	raw_sb->section_count = cpu_to_le32(section_count + secs);
 	raw_sb->segment_count = cpu_to_le32(segment_count + segs);
 	raw_sb->segment_count_main = cpu_to_le32(segment_count_main + segs);
 	raw_sb->block_count = cpu_to_le64(block_count +
 					(long long)segs * sbi->blocks_per_seg);
+	if (f2fs_is_multi_device(sbi)) {
+		int last_dev = sbi->s_ndevs - 1;
+		int dev_segs =
+			le32_to_cpu(raw_sb->devs[last_dev].total_segments);
+
+		raw_sb->devs[last_dev].total_segments =
+						cpu_to_le32(dev_segs + segs);
+	}
+
+	up_write(&sbi->sb_lock);
 }
 
 static void update_fs_metadata(struct f2fs_sb_info *sbi, int secs)
 {
 	int segs = secs * sbi->segs_per_sec;
+	long long blks = (long long)segs * sbi->blocks_per_seg;
 	long long user_block_count =
 				le64_to_cpu(F2FS_CKPT(sbi)->user_block_count);
 
@@ -1449,8 +1483,20 @@ static void update_fs_metadata(struct f2fs_sb_info *sbi, int secs)
 	MAIN_SEGS(sbi) = (int)MAIN_SEGS(sbi) + segs;
 	FREE_I(sbi)->free_sections = (int)FREE_I(sbi)->free_sections + secs;
 	FREE_I(sbi)->free_segments = (int)FREE_I(sbi)->free_segments + segs;
-	F2FS_CKPT(sbi)->user_block_count = cpu_to_le64(user_block_count +
-					(long long)segs * sbi->blocks_per_seg);
+	F2FS_CKPT(sbi)->user_block_count = cpu_to_le64(user_block_count + blks);
+
+	if (f2fs_is_multi_device(sbi)) {
+		int last_dev = sbi->s_ndevs - 1;
+
+		FDEV(last_dev).total_segments =
+				(int)FDEV(last_dev).total_segments + segs;
+		FDEV(last_dev).end_blk =
+				(long long)FDEV(last_dev).end_blk + blks;
+#ifdef CONFIG_BLK_DEV_ZONED
+		FDEV(last_dev).nr_blkz = (int)FDEV(last_dev).nr_blkz +
+					(int)(blks >> sbi->log_blocks_per_blkz);
+#endif
+	}
 }
 
 int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count)
@@ -1465,6 +1511,15 @@ int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count)
 	if (block_count > old_block_count)
 		return -EINVAL;
 
+	if (f2fs_is_multi_device(sbi)) {
+		int last_dev = sbi->s_ndevs - 1;
+		__u64 last_segs = FDEV(last_dev).total_segments;
+
+		if (block_count + last_segs * sbi->blocks_per_seg <=
+								old_block_count)
+			return -EINVAL;
+	}
+
 	/* new fs size should align to section size */
 	div_u64_rem(block_count, BLKS_PER_SEC(sbi), &rem);
 	if (rem)
@@ -1532,11 +1587,17 @@ int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count)
 		goto out;
 	}
 
+	mutex_lock(&sbi->cp_mutex);
 	update_fs_metadata(sbi, -secs);
 	clear_sbi_flag(sbi, SBI_IS_RESIZEFS);
+	set_sbi_flag(sbi, SBI_IS_DIRTY);
+	mutex_unlock(&sbi->cp_mutex);
+
 	err = f2fs_sync_fs(sbi->sb, 1);
 	if (err) {
+		mutex_lock(&sbi->cp_mutex);
 		update_fs_metadata(sbi, secs);
+		mutex_unlock(&sbi->cp_mutex);
 		update_sb_metadata(sbi, secs);
 		f2fs_commit_super(sbi, false);
 	}
diff --git a/fs/f2fs/hash.c b/fs/f2fs/hash.c
index 5bc4dcd8fc03..8c4ea5003ef8 100644
--- a/fs/f2fs/hash.c
+++ b/fs/f2fs/hash.c
@@ -12,7 +12,6 @@
 #include <linux/types.h>
 #include <linux/fs.h>
 #include <linux/f2fs_fs.h>
-#include <linux/cryptohash.h>
 #include <linux/pagemap.h>
 #include <linux/unicode.h>
 
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index 896db0416f0e..4167e5408151 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -368,7 +368,7 @@ static int f2fs_move_inline_dirents(struct inode *dir, struct page *ipage,
 	struct f2fs_dentry_ptr src, dst;
 	int err;
 
-	page = f2fs_grab_cache_page(dir->i_mapping, 0, false);
+	page = f2fs_grab_cache_page(dir->i_mapping, 0, true);
 	if (!page) {
 		f2fs_put_page(ipage, 1);
 		return -ENOMEM;
@@ -530,7 +530,7 @@ recover:
 	return err;
 }
 
-static int f2fs_convert_inline_dir(struct inode *dir, struct page *ipage,
+static int do_convert_inline_dir(struct inode *dir, struct page *ipage,
 							void *inline_dentry)
 {
 	if (!F2FS_I(dir)->i_dir_level)
@@ -539,6 +539,44 @@ static int f2fs_convert_inline_dir(struct inode *dir, struct page *ipage,
 		return f2fs_move_rehashed_dirents(dir, ipage, inline_dentry);
 }
 
+int f2fs_try_convert_inline_dir(struct inode *dir, struct dentry *dentry)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
+	struct page *ipage;
+	struct fscrypt_name fname;
+	void *inline_dentry = NULL;
+	int err = 0;
+
+	if (!f2fs_has_inline_dentry(dir))
+		return 0;
+
+	f2fs_lock_op(sbi);
+
+	err = fscrypt_setup_filename(dir, &dentry->d_name, 0, &fname);
+	if (err)
+		goto out;
+
+	ipage = f2fs_get_node_page(sbi, dir->i_ino);
+	if (IS_ERR(ipage)) {
+		err = PTR_ERR(ipage);
+		goto out;
+	}
+
+	if (f2fs_has_enough_room(dir, ipage, &fname)) {
+		f2fs_put_page(ipage, 1);
+		goto out;
+	}
+
+	inline_dentry = inline_data_addr(dir, ipage);
+
+	err = do_convert_inline_dir(dir, ipage, inline_dentry);
+	if (!err)
+		f2fs_put_page(ipage, 1);
+out:
+	f2fs_unlock_op(sbi);
+	return err;
+}
+
 int f2fs_add_inline_entry(struct inode *dir, const struct qstr *new_name,
 				const struct qstr *orig_name,
 				struct inode *inode, nid_t ino, umode_t mode)
@@ -562,7 +600,7 @@ int f2fs_add_inline_entry(struct inode *dir, const struct qstr *new_name,
 
 	bit_pos = f2fs_room_for_filename(d.bitmap, slots, d.max);
 	if (bit_pos >= d.max) {
-		err = f2fs_convert_inline_dir(dir, ipage, inline_dentry);
+		err = do_convert_inline_dir(dir, ipage, inline_dentry);
 		if (err)
 			return err;
 		err = -EAGAIN;
diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index db4fec30c30d..44582a4db513 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -200,6 +200,7 @@ static bool sanity_check_inode(struct inode *inode, struct page *node_page)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct f2fs_inode_info *fi = F2FS_I(inode);
+	struct f2fs_inode *ri = F2FS_INODE(node_page);
 	unsigned long long iblocks;
 
 	iblocks = le64_to_cpu(F2FS_INODE(node_page)->i_blocks);
@@ -286,6 +287,36 @@ static bool sanity_check_inode(struct inode *inode, struct page *node_page)
 		return false;
 	}
 
+	if (f2fs_has_extra_attr(inode) && f2fs_sb_has_compression(sbi) &&
+			fi->i_flags & F2FS_COMPR_FL &&
+			F2FS_FITS_IN_INODE(ri, fi->i_extra_isize,
+						i_log_cluster_size)) {
+		if (ri->i_compress_algorithm >= COMPRESS_MAX) {
+			f2fs_warn(sbi, "%s: inode (ino=%lx) has unsupported "
+				"compress algorithm: %u, run fsck to fix",
+				  __func__, inode->i_ino,
+				  ri->i_compress_algorithm);
+			return false;
+		}
+		if (le64_to_cpu(ri->i_compr_blocks) >
+				SECTOR_TO_BLOCK(inode->i_blocks)) {
+			f2fs_warn(sbi, "%s: inode (ino=%lx) has inconsistent "
+				"i_compr_blocks:%llu, i_blocks:%llu, run fsck to fix",
+				  __func__, inode->i_ino,
+				  le64_to_cpu(ri->i_compr_blocks),
+				  SECTOR_TO_BLOCK(inode->i_blocks));
+			return false;
+		}
+		if (ri->i_log_cluster_size < MIN_COMPRESS_LOG_SIZE ||
+			ri->i_log_cluster_size > MAX_COMPRESS_LOG_SIZE) {
+			f2fs_warn(sbi, "%s: inode (ino=%lx) has unsupported "
+				"log cluster size: %u, run fsck to fix",
+				  __func__, inode->i_ino,
+				  ri->i_log_cluster_size);
+			return false;
+		}
+	}
+
 	return true;
 }
 
@@ -331,7 +362,7 @@ static int do_read_inode(struct inode *inode)
 	fi->i_flags = le32_to_cpu(ri->i_flags);
 	if (S_ISREG(inode->i_mode))
 		fi->i_flags &= ~F2FS_PROJINHERIT_FL;
-	fi->flags = 0;
+	bitmap_zero(fi->flags, FI_MAX);
 	fi->i_advise = ri->i_advise;
 	fi->i_pino = le32_to_cpu(ri->i_pino);
 	fi->i_dir_level = ri->i_dir_level;
@@ -407,6 +438,18 @@ static int do_read_inode(struct inode *inode)
 		fi->i_crtime.tv_nsec = le32_to_cpu(ri->i_crtime_nsec);
 	}
 
+	if (f2fs_has_extra_attr(inode) && f2fs_sb_has_compression(sbi) &&
+					(fi->i_flags & F2FS_COMPR_FL)) {
+		if (F2FS_FITS_IN_INODE(ri, fi->i_extra_isize,
+					i_log_cluster_size)) {
+			fi->i_compr_blocks = le64_to_cpu(ri->i_compr_blocks);
+			fi->i_compress_algorithm = ri->i_compress_algorithm;
+			fi->i_log_cluster_size = ri->i_log_cluster_size;
+			fi->i_cluster_size = 1 << fi->i_log_cluster_size;
+			set_inode_flag(inode, FI_COMPRESSED_FILE);
+		}
+	}
+
 	F2FS_I(inode)->i_disk_time[0] = inode->i_atime;
 	F2FS_I(inode)->i_disk_time[1] = inode->i_ctime;
 	F2FS_I(inode)->i_disk_time[2] = inode->i_mtime;
@@ -416,6 +459,8 @@ static int do_read_inode(struct inode *inode)
 	stat_inc_inline_xattr(inode);
 	stat_inc_inline_inode(inode);
 	stat_inc_inline_dir(inode);
+	stat_inc_compr_inode(inode);
+	stat_add_compr_blocks(inode, F2FS_I(inode)->i_compr_blocks);
 
 	return 0;
 }
@@ -490,7 +535,7 @@ retry:
 	inode = f2fs_iget(sb, ino);
 	if (IS_ERR(inode)) {
 		if (PTR_ERR(inode) == -ENOMEM) {
-			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT);
 			goto retry;
 		}
 	}
@@ -569,6 +614,17 @@ void f2fs_update_inode(struct inode *inode, struct page *node_page)
 			ri->i_crtime_nsec =
 				cpu_to_le32(F2FS_I(inode)->i_crtime.tv_nsec);
 		}
+
+		if (f2fs_sb_has_compression(F2FS_I_SB(inode)) &&
+			F2FS_FITS_IN_INODE(ri, F2FS_I(inode)->i_extra_isize,
+							i_log_cluster_size)) {
+			ri->i_compr_blocks =
+				cpu_to_le64(F2FS_I(inode)->i_compr_blocks);
+			ri->i_compress_algorithm =
+				F2FS_I(inode)->i_compress_algorithm;
+			ri->i_log_cluster_size =
+				F2FS_I(inode)->i_log_cluster_size;
+		}
 	}
 
 	__set_inode_rdev(inode, ri);
@@ -615,7 +671,11 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
 			inode->i_ino == F2FS_META_INO(sbi))
 		return 0;
 
-	if (!is_inode_flag_set(inode, FI_DIRTY_INODE))
+	/*
+	 * atime could be updated without dirtying f2fs inode in lazytime mode
+	 */
+	if (f2fs_is_time_consistent(inode) &&
+		!is_inode_flag_set(inode, FI_DIRTY_INODE))
 		return 0;
 
 	if (!f2fs_is_checkpoint_ready(sbi))
@@ -677,7 +737,7 @@ retry:
 		err = f2fs_truncate(inode);
 
 	if (time_to_inject(sbi, FAULT_EVICT_INODE)) {
-		f2fs_show_injection_info(FAULT_EVICT_INODE);
+		f2fs_show_injection_info(sbi, FAULT_EVICT_INODE);
 		err = -EIO;
 	}
 
@@ -707,6 +767,8 @@ no_delete:
 	stat_dec_inline_xattr(inode);
 	stat_dec_inline_dir(inode);
 	stat_dec_inline_inode(inode);
+	stat_dec_compr_inode(inode);
+	stat_sub_compr_blocks(inode, F2FS_I(inode)->i_compr_blocks);
 
 	if (likely(!f2fs_cp_error(sbi) &&
 				!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
@@ -714,7 +776,7 @@ no_delete:
 	else
 		f2fs_inode_synced(inode);
 
-	/* ino == 0, if f2fs_new_inode() was failed t*/
+	/* for the case f2fs_new_inode() was failed, .i_ino is zero, skip it */
 	if (inode->i_ino)
 		invalidate_mapping_pages(NODE_MAPPING(sbi), inode->i_ino,
 							inode->i_ino);
diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index 4faf06e8bf89..f54119da2217 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -75,9 +75,7 @@ static struct inode *f2fs_new_inode(struct inode *dir, umode_t mode)
 
 	set_inode_flag(inode, FI_NEW_INODE);
 
-	/* If the directory encrypted, then we should encrypt the inode. */
-	if ((IS_ENCRYPTED(dir) || DUMMY_ENCRYPTION_ENABLED(sbi)) &&
-				f2fs_may_encrypt(inode))
+	if (f2fs_may_encrypt(dir, inode))
 		f2fs_set_encrypted_inode(inode);
 
 	if (f2fs_sb_has_extra_attr(sbi)) {
@@ -119,6 +117,13 @@ static struct inode *f2fs_new_inode(struct inode *dir, umode_t mode)
 	if (F2FS_I(inode)->i_flags & F2FS_PROJINHERIT_FL)
 		set_inode_flag(inode, FI_PROJ_INHERIT);
 
+	if (f2fs_sb_has_compression(sbi)) {
+		/* Inherit the compression flag in directory */
+		if ((F2FS_I(dir)->i_flags & F2FS_COMPR_FL) &&
+					f2fs_may_compress(inode))
+			set_compress_context(inode);
+	}
+
 	f2fs_set_inode_flags(inode);
 
 	trace_f2fs_new_inode(inode, 0);
@@ -149,6 +154,9 @@ static inline int is_extension_exist(const unsigned char *s, const char *sub)
 	size_t sublen = strlen(sub);
 	int i;
 
+	if (sublen == 1 && *sub == '*')
+		return 1;
+
 	/*
 	 * filename format of multimedia file should be defined as:
 	 * "filename + '.' + extension + (optional: '.' + temp extension)".
@@ -167,7 +175,7 @@ static inline int is_extension_exist(const unsigned char *s, const char *sub)
 }
 
 /*
- * Set multimedia files as cold files for hot/cold data separation
+ * Set file's temperature for hot/cold data separation
  */
 static inline void set_file_temperature(struct f2fs_sb_info *sbi, struct inode *inode,
 		const unsigned char *name)
@@ -262,6 +270,45 @@ int f2fs_update_extension_list(struct f2fs_sb_info *sbi, const char *name,
 	return 0;
 }
 
+static void set_compress_inode(struct f2fs_sb_info *sbi, struct inode *inode,
+						const unsigned char *name)
+{
+	__u8 (*extlist)[F2FS_EXTENSION_LEN] = sbi->raw_super->extension_list;
+	unsigned char (*ext)[F2FS_EXTENSION_LEN];
+	unsigned int ext_cnt = F2FS_OPTION(sbi).compress_ext_cnt;
+	int i, cold_count, hot_count;
+
+	if (!f2fs_sb_has_compression(sbi) ||
+			is_inode_flag_set(inode, FI_COMPRESSED_FILE) ||
+			F2FS_I(inode)->i_flags & F2FS_NOCOMP_FL ||
+			!f2fs_may_compress(inode))
+		return;
+
+	down_read(&sbi->sb_lock);
+
+	cold_count = le32_to_cpu(sbi->raw_super->extension_count);
+	hot_count = sbi->raw_super->hot_ext_count;
+
+	for (i = cold_count; i < cold_count + hot_count; i++) {
+		if (is_extension_exist(name, extlist[i])) {
+			up_read(&sbi->sb_lock);
+			return;
+		}
+	}
+
+	up_read(&sbi->sb_lock);
+
+	ext = F2FS_OPTION(sbi).extensions;
+
+	for (i = 0; i < ext_cnt; i++) {
+		if (!is_extension_exist(name, ext[i]))
+			continue;
+
+		set_compress_context(inode);
+		return;
+	}
+}
+
 static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 						bool excl)
 {
@@ -286,6 +333,8 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	if (!test_opt(sbi, DISABLE_EXT_IDENTIFY))
 		set_file_temperature(sbi, inode, dentry->d_name.name);
 
+	set_compress_inode(sbi, inode, dentry->d_name.name);
+
 	inode->i_op = &f2fs_file_inode_operations;
 	inode->i_fop = &f2fs_file_operations;
 	inode->i_mapping->a_ops = &f2fs_dblock_aops;
@@ -797,6 +846,7 @@ static int __f2fs_tmpfile(struct inode *dir, struct dentry *dentry,
 
 	if (whiteout) {
 		f2fs_i_links_write(inode, false);
+		inode->i_state |= I_LINKABLE;
 		*whiteout = inode;
 	} else {
 		d_tmpfile(dentry, inode);
@@ -824,12 +874,6 @@ static int f2fs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
 	if (!f2fs_is_checkpoint_ready(sbi))
 		return -ENOSPC;
 
-	if (IS_ENCRYPTED(dir) || DUMMY_ENCRYPTION_ENABLED(sbi)) {
-		int err = fscrypt_get_encryption_info(dir);
-		if (err)
-			return err;
-	}
-
 	return __f2fs_tmpfile(dir, dentry, mode, NULL);
 }
 
@@ -849,12 +893,11 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct inode *old_inode = d_inode(old_dentry);
 	struct inode *new_inode = d_inode(new_dentry);
 	struct inode *whiteout = NULL;
-	struct page *old_dir_page;
+	struct page *old_dir_page = NULL;
 	struct page *old_page, *new_page = NULL;
 	struct f2fs_dir_entry *old_dir_entry = NULL;
 	struct f2fs_dir_entry *old_entry;
 	struct f2fs_dir_entry *new_entry;
-	bool is_old_inline = f2fs_has_inline_dentry(old_dir);
 	int err;
 
 	if (unlikely(f2fs_cp_error(sbi)))
@@ -867,6 +910,26 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			F2FS_I(old_dentry->d_inode)->i_projid)))
 		return -EXDEV;
 
+	/*
+	 * If new_inode is null, the below renaming flow will
+	 * add a link in old_dir which can conver inline_dir.
+	 * After then, if we failed to get the entry due to other
+	 * reasons like ENOMEM, we had to remove the new entry.
+	 * Instead of adding such the error handling routine, let's
+	 * simply convert first here.
+	 */
+	if (old_dir == new_dir && !new_inode) {
+		err = f2fs_try_convert_inline_dir(old_dir, new_dentry);
+		if (err)
+			return err;
+	}
+
+	if (flags & RENAME_WHITEOUT) {
+		err = f2fs_create_whiteout(old_dir, &whiteout);
+		if (err)
+			return err;
+	}
+
 	err = dquot_initialize(old_dir);
 	if (err)
 		goto out;
@@ -898,17 +961,11 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		}
 	}
 
-	if (flags & RENAME_WHITEOUT) {
-		err = f2fs_create_whiteout(old_dir, &whiteout);
-		if (err)
-			goto out_dir;
-	}
-
 	if (new_inode) {
 
 		err = -ENOTEMPTY;
 		if (old_dir_entry && !f2fs_empty_dir(new_inode))
-			goto out_whiteout;
+			goto out_dir;
 
 		err = -ENOENT;
 		new_entry = f2fs_find_entry(new_dir, &new_dentry->d_name,
@@ -916,7 +973,7 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (!new_entry) {
 			if (IS_ERR(new_page))
 				err = PTR_ERR(new_page);
-			goto out_whiteout;
+			goto out_dir;
 		}
 
 		f2fs_balance_fs(sbi, true);
@@ -928,6 +985,7 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			goto put_out_dir;
 
 		f2fs_set_link(new_dir, new_entry, new_page, old_inode);
+		new_page = NULL;
 
 		new_inode->i_ctime = current_time(new_inode);
 		down_write(&F2FS_I(new_inode)->i_sem);
@@ -948,49 +1006,28 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		err = f2fs_add_link(new_dentry, old_inode);
 		if (err) {
 			f2fs_unlock_op(sbi);
-			goto out_whiteout;
+			goto out_dir;
 		}
 
 		if (old_dir_entry)
 			f2fs_i_links_write(new_dir, true);
-
-		/*
-		 * old entry and new entry can locate in the same inline
-		 * dentry in inode, when attaching new entry in inline dentry,
-		 * it could force inline dentry conversion, after that,
-		 * old_entry and old_page will point to wrong address, in
-		 * order to avoid this, let's do the check and update here.
-		 */
-		if (is_old_inline && !f2fs_has_inline_dentry(old_dir)) {
-			f2fs_put_page(old_page, 0);
-			old_page = NULL;
-
-			old_entry = f2fs_find_entry(old_dir,
-						&old_dentry->d_name, &old_page);
-			if (!old_entry) {
-				err = -ENOENT;
-				if (IS_ERR(old_page))
-					err = PTR_ERR(old_page);
-				f2fs_unlock_op(sbi);
-				goto out_whiteout;
-			}
-		}
 	}
 
 	down_write(&F2FS_I(old_inode)->i_sem);
 	if (!old_dir_entry || whiteout)
 		file_lost_pino(old_inode);
 	else
-		F2FS_I(old_inode)->i_pino = new_dir->i_ino;
+		/* adjust dir's i_pino to pass fsck check */
+		f2fs_i_pino_write(old_inode, new_dir->i_ino);
 	up_write(&F2FS_I(old_inode)->i_sem);
 
 	old_inode->i_ctime = current_time(old_inode);
 	f2fs_mark_inode_dirty_sync(old_inode, false);
 
 	f2fs_delete_entry(old_entry, old_page, old_dir, NULL);
+	old_page = NULL;
 
 	if (whiteout) {
-		whiteout->i_state |= I_LINKABLE;
 		set_inode_flag(whiteout, FI_INC_LINK);
 		err = f2fs_add_link(old_dentry, whiteout);
 		if (err)
@@ -1024,17 +1061,15 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 put_out_dir:
 	f2fs_unlock_op(sbi);
-	if (new_page)
-		f2fs_put_page(new_page, 0);
-out_whiteout:
-	if (whiteout)
-		iput(whiteout);
+	f2fs_put_page(new_page, 0);
 out_dir:
 	if (old_dir_entry)
 		f2fs_put_page(old_dir_page, 0);
 out_old:
 	f2fs_put_page(old_page, 0);
 out:
+	if (whiteout)
+		iput(whiteout);
 	return err;
 }
 
@@ -1141,7 +1176,11 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	f2fs_set_link(old_dir, old_entry, old_page, new_inode);
 
 	down_write(&F2FS_I(old_inode)->i_sem);
-	file_lost_pino(old_inode);
+	if (!old_dir_entry)
+		file_lost_pino(old_inode);
+	else
+		/* adjust dir's i_pino to pass fsck check */
+		f2fs_i_pino_write(old_inode, new_dir->i_ino);
 	up_write(&F2FS_I(old_inode)->i_sem);
 
 	old_dir->i_ctime = current_time(old_dir);
@@ -1156,7 +1195,11 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	f2fs_set_link(new_dir, new_entry, new_page, old_inode);
 
 	down_write(&F2FS_I(new_inode)->i_sem);
-	file_lost_pino(new_inode);
+	if (!new_dir_entry)
+		file_lost_pino(new_inode);
+	else
+		/* adjust dir's i_pino to pass fsck check */
+		f2fs_i_pino_write(new_inode, old_dir->i_ino);
 	up_write(&F2FS_I(new_inode)->i_sem);
 
 	new_dir->i_ctime = current_time(new_dir);
diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index 8b66bc4c004b..ecbd6bd14a49 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -510,9 +510,6 @@ int f2fs_try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink)
 	return nr - nr_shrink;
 }
 
-/*
- * This function always returns success
- */
 int f2fs_get_node_info(struct f2fs_sb_info *sbi, nid_t nid,
 						struct node_info *ni)
 {
@@ -716,8 +713,7 @@ got:
 /*
  * Caller should call f2fs_put_dnode(dn).
  * Also, it should grab and release a rwsem by calling f2fs_lock_op() and
- * f2fs_unlock_op() only if ro is not set RDONLY_NODE.
- * In the case of RDONLY_NODE, we don't need to care about mutex.
+ * f2fs_unlock_op() only if mode is set with ALLOC_NODE.
  */
 int f2fs_get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode)
 {
@@ -809,8 +805,7 @@ int f2fs_get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode)
 	dn->nid = nids[level];
 	dn->ofs_in_node = offset[level];
 	dn->node_page = npage[level];
-	dn->data_blkaddr = datablock_addr(dn->inode,
-				dn->node_page, dn->ofs_in_node);
+	dn->data_blkaddr = f2fs_data_blkaddr(dn);
 	return 0;
 
 release_pages:
@@ -875,7 +870,7 @@ static int truncate_dnode(struct dnode_of_data *dn)
 
 	/* get direct node */
 	page = f2fs_get_node_page(F2FS_I_SB(dn->inode), dn->nid);
-	if (IS_ERR(page) && PTR_ERR(page) == -ENOENT)
+	if (PTR_ERR(page) == -ENOENT)
 		return 1;
 	else if (IS_ERR(page))
 		return PTR_ERR(page);
@@ -1188,8 +1183,9 @@ int f2fs_remove_inode_page(struct inode *inode)
 	}
 
 	if (unlikely(inode->i_blocks != 0 && inode->i_blocks != 8)) {
-		f2fs_warn(F2FS_I_SB(inode), "Inconsistent i_blocks, ino:%lu, iblocks:%llu",
-			  inode->i_ino, (unsigned long long)inode->i_blocks);
+		f2fs_warn(F2FS_I_SB(inode),
+			"f2fs_remove_inode_page: inconsistent i_blocks, ino:%lu, iblocks:%llu",
+			inode->i_ino, (unsigned long long)inode->i_blocks);
 		set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_FSCK);
 	}
 
@@ -1562,15 +1558,16 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted,
 	if (atomic && !test_opt(sbi, NOBARRIER))
 		fio.op_flags |= REQ_PREFLUSH | REQ_FUA;
 
-	set_page_writeback(page);
-	ClearPageError(page);
-
+	/* should add to global list before clearing PAGECACHE status */
 	if (f2fs_in_warm_node_list(sbi, page)) {
 		seq = f2fs_add_fsync_node_entry(sbi, page);
 		if (seq_id)
 			*seq_id = seq;
 	}
 
+	set_page_writeback(page);
+	ClearPageError(page);
+
 	fio.old_blkaddr = ni.blk_addr;
 	f2fs_do_write_node_page(nid, &fio);
 	set_node_addr(sbi, &ni, fio.new_blkaddr, is_fsync_dnode(page));
@@ -1979,7 +1976,7 @@ static int f2fs_write_node_pages(struct address_space *mapping,
 		goto skip_write;
 
 	/* balancing f2fs's metadata in background */
-	f2fs_balance_fs_bg(sbi);
+	f2fs_balance_fs_bg(sbi, true);
 
 	/* collect a number of dirty node pages and write together */
 	if (wbc->sync_mode != WB_SYNC_ALL &&
@@ -2349,7 +2346,6 @@ static int __f2fs_build_free_nids(struct f2fs_sb_info *sbi,
 
 			if (ret) {
 				up_read(&nm_i->nat_tree_lock);
-				f2fs_bug_on(sbi, !mount);
 				f2fs_err(sbi, "NAT is corrupt, run fsck to fix it");
 				return ret;
 			}
@@ -2399,7 +2395,7 @@ bool f2fs_alloc_nid(struct f2fs_sb_info *sbi, nid_t *nid)
 	struct free_nid *i = NULL;
 retry:
 	if (time_to_inject(sbi, FAULT_ALLOC_NID)) {
-		f2fs_show_injection_info(FAULT_ALLOC_NID);
+		f2fs_show_injection_info(sbi, FAULT_ALLOC_NID);
 		return false;
 	}
 
@@ -2603,7 +2599,7 @@ int f2fs_recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
 retry:
 	ipage = f2fs_grab_cache_page(NODE_MAPPING(sbi), ino, false);
 	if (!ipage) {
-		congestion_wait(BLK_RW_ASYNC, HZ/50);
+		congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT);
 		goto retry;
 	}
 
@@ -3194,22 +3190,22 @@ void f2fs_destroy_node_manager(struct f2fs_sb_info *sbi)
 
 int __init f2fs_create_node_manager_caches(void)
 {
-	nat_entry_slab = f2fs_kmem_cache_create("nat_entry",
+	nat_entry_slab = f2fs_kmem_cache_create("f2fs_nat_entry",
 			sizeof(struct nat_entry));
 	if (!nat_entry_slab)
 		goto fail;
 
-	free_nid_slab = f2fs_kmem_cache_create("free_nid",
+	free_nid_slab = f2fs_kmem_cache_create("f2fs_free_nid",
 			sizeof(struct free_nid));
 	if (!free_nid_slab)
 		goto destroy_nat_entry;
 
-	nat_entry_set_slab = f2fs_kmem_cache_create("nat_entry_set",
+	nat_entry_set_slab = f2fs_kmem_cache_create("f2fs_nat_entry_set",
 			sizeof(struct nat_entry_set));
 	if (!nat_entry_set_slab)
 		goto destroy_free_nid;
 
-	fsync_node_entry_slab = f2fs_kmem_cache_create("fsync_node_entry",
+	fsync_node_entry_slab = f2fs_kmem_cache_create("f2fs_fsync_node_entry",
 			sizeof(struct fsync_node_entry));
 	if (!fsync_node_entry_slab)
 		goto destroy_nat_entry_set;
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 783773e4560d..dd804c07eeb0 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -496,8 +496,7 @@ out:
 	return 0;
 
 truncate_out:
-	if (datablock_addr(tdn.inode, tdn.node_page,
-					tdn.ofs_in_node) == blkaddr)
+	if (f2fs_data_blkaddr(&tdn) == blkaddr)
 		f2fs_truncate_data_blocks_range(&tdn, 1);
 	if (dn->inode->i_ino == nid && !dn->inode_page_locked)
 		unlock_page(dn->inode_page);
@@ -535,7 +534,7 @@ retry_dn:
 	err = f2fs_get_dnode_of_data(&dn, start, ALLOC_NODE);
 	if (err) {
 		if (err == -ENOMEM) {
-			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT);
 			goto retry_dn;
 		}
 		goto out;
@@ -560,8 +559,8 @@ retry_dn:
 	for (; start < end; start++, dn.ofs_in_node++) {
 		block_t src, dest;
 
-		src = datablock_addr(dn.inode, dn.node_page, dn.ofs_in_node);
-		dest = datablock_addr(dn.inode, page, dn.ofs_in_node);
+		src = f2fs_data_blkaddr(&dn);
+		dest = data_blkaddr(dn.inode, page, dn.ofs_in_node);
 
 		if (__is_valid_data_blkaddr(src) &&
 			!f2fs_is_valid_blkaddr(sbi, src, META_POR)) {
@@ -618,7 +617,8 @@ retry_prev:
 			err = check_index_in_prev_nodes(sbi, dest, &dn);
 			if (err) {
 				if (err == -ENOMEM) {
-					congestion_wait(BLK_RW_ASYNC, HZ/50);
+					congestion_wait(BLK_RW_ASYNC,
+							DEFAULT_IO_TIMEOUT);
 					goto retry_prev;
 				}
 				goto err;
@@ -711,7 +711,7 @@ next:
 		f2fs_put_page(page, 1);
 	}
 	if (!err)
-		f2fs_allocate_new_segments(sbi);
+		f2fs_allocate_new_segments(sbi, NO_CHECK_TYPE);
 	return err;
 }
 
@@ -723,6 +723,7 @@ int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	int ret = 0;
 	unsigned long s_flags = sbi->sb->s_flags;
 	bool need_writecp = false;
+	bool fix_curseg_write_pointer = false;
 #ifdef CONFIG_QUOTA
 	int quota_enabled;
 #endif
@@ -774,6 +775,8 @@ int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 		sbi->sb->s_flags = s_flags;
 	}
 skip:
+	fix_curseg_write_pointer = !check_only || list_empty(&inode_list);
+
 	destroy_fsync_dnodes(&inode_list, err);
 	destroy_fsync_dnodes(&tmp_inode_list, err);
 
@@ -784,9 +787,22 @@ skip:
 	if (err) {
 		truncate_inode_pages_final(NODE_MAPPING(sbi));
 		truncate_inode_pages_final(META_MAPPING(sbi));
-	} else {
-		clear_sbi_flag(sbi, SBI_POR_DOING);
 	}
+
+	/*
+	 * If fsync data succeeds or there is no fsync data to recover,
+	 * and the f2fs is not read only, check and fix zoned block devices'
+	 * write pointer consistency.
+	 */
+	if (!err && fix_curseg_write_pointer && !f2fs_readonly(sbi->sb) &&
+			f2fs_sb_has_blkzoned(sbi)) {
+		err = f2fs_fix_curseg_write_pointer(sbi);
+		ret = err;
+	}
+
+	if (!err)
+		clear_sbi_flag(sbi, SBI_POR_DOING);
+
 	mutex_unlock(&sbi->cp_mutex);
 
 	/* let's drop all the directory inodes for clean checkpoint */
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 808709581481..b7a9421472a7 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -172,7 +172,7 @@ bool f2fs_need_SSR(struct f2fs_sb_info *sbi)
 	int dent_secs = get_blocktype_secs(sbi, F2FS_DIRTY_DENTS);
 	int imeta_secs = get_blocktype_secs(sbi, F2FS_DIRTY_IMETA);
 
-	if (test_opt(sbi, LFS))
+	if (f2fs_lfs_mode(sbi))
 		return false;
 	if (sbi->gc_mode == GC_URGENT)
 		return true;
@@ -245,7 +245,8 @@ retry:
 								LOOKUP_NODE);
 			if (err) {
 				if (err == -ENOMEM) {
-					congestion_wait(BLK_RW_ASYNC, HZ/50);
+					congestion_wait(BLK_RW_ASYNC,
+							DEFAULT_IO_TIMEOUT);
 					cond_resched();
 					goto retry;
 				}
@@ -288,6 +289,8 @@ void f2fs_drop_inmem_pages_all(struct f2fs_sb_info *sbi, bool gc_failure)
 	struct list_head *head = &sbi->inode_list[ATOMIC_FILE];
 	struct inode *inode;
 	struct f2fs_inode_info *fi;
+	unsigned int count = sbi->atomic_files;
+	unsigned int looped = 0;
 next:
 	spin_lock(&sbi->inode_lock[ATOMIC_FILE]);
 	if (list_empty(head)) {
@@ -296,22 +299,26 @@ next:
 	}
 	fi = list_first_entry(head, struct f2fs_inode_info, inmem_ilist);
 	inode = igrab(&fi->vfs_inode);
+	if (inode)
+		list_move_tail(&fi->inmem_ilist, head);
 	spin_unlock(&sbi->inode_lock[ATOMIC_FILE]);
 
 	if (inode) {
 		if (gc_failure) {
-			if (fi->i_gc_failures[GC_FAILURE_ATOMIC])
-				goto drop;
-			goto skip;
+			if (!fi->i_gc_failures[GC_FAILURE_ATOMIC])
+				goto skip;
 		}
-drop:
 		set_inode_flag(inode, FI_ATOMIC_REVOKE_REQUEST);
 		f2fs_drop_inmem_pages(inode);
+skip:
 		iput(inode);
 	}
-skip:
-	congestion_wait(BLK_RW_ASYNC, HZ/50);
+	congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT);
 	cond_resched();
+	if (gc_failure) {
+		if (++looped >= count)
+			return;
+	}
 	goto next;
 }
 
@@ -327,13 +334,15 @@ void f2fs_drop_inmem_pages(struct inode *inode)
 		mutex_unlock(&fi->inmem_lock);
 	}
 
-	clear_inode_flag(inode, FI_ATOMIC_FILE);
 	fi->i_gc_failures[GC_FAILURE_ATOMIC] = 0;
-	stat_dec_atomic_write(inode);
 
 	spin_lock(&sbi->inode_lock[ATOMIC_FILE]);
 	if (!list_empty(&fi->inmem_ilist))
 		list_del_init(&fi->inmem_ilist);
+	if (f2fs_is_atomic_file(inode)) {
+		clear_inode_flag(inode, FI_ATOMIC_FILE);
+		sbi->atomic_files--;
+	}
 	spin_unlock(&sbi->inode_lock[ATOMIC_FILE]);
 }
 
@@ -407,7 +416,8 @@ retry:
 			err = f2fs_do_write_data_page(&fio);
 			if (err) {
 				if (err == -ENOMEM) {
-					congestion_wait(BLK_RW_ASYNC, HZ/50);
+					congestion_wait(BLK_RW_ASYNC,
+							DEFAULT_IO_TIMEOUT);
 					cond_resched();
 					goto retry;
 				}
@@ -480,13 +490,13 @@ int f2fs_commit_inmem_pages(struct inode *inode)
 void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need)
 {
 	if (time_to_inject(sbi, FAULT_CHECKPOINT)) {
-		f2fs_show_injection_info(FAULT_CHECKPOINT);
+		f2fs_show_injection_info(sbi, FAULT_CHECKPOINT);
 		f2fs_stop_checkpoint(sbi, false);
 	}
 
 	/* balance_fs_bg is able to be pending */
 	if (need && excess_cached_nats(sbi))
-		f2fs_balance_fs_bg(sbi);
+		f2fs_balance_fs_bg(sbi, false);
 
 	if (!f2fs_is_checkpoint_ready(sbi))
 		return;
@@ -496,12 +506,12 @@ void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need)
 	 * dir/node pages without enough free segments.
 	 */
 	if (has_not_enough_free_secs(sbi, 0, 0)) {
-		mutex_lock(&sbi->gc_mutex);
+		down_write(&sbi->gc_lock);
 		f2fs_gc(sbi, false, false, NULL_SEGNO);
 	}
 }
 
-void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi)
+void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi, bool from_bg)
 {
 	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
 		return;
@@ -530,7 +540,7 @@ void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi)
 			excess_dirty_nats(sbi) ||
 			excess_dirty_nodes(sbi) ||
 			f2fs_time_over(sbi, CP_TIME)) {
-		if (test_opt(sbi, DATA_FLUSH)) {
+		if (test_opt(sbi, DATA_FLUSH) && from_bg) {
 			struct blk_plug plug;
 
 			mutex_lock(&sbi->flush_lock);
@@ -1008,8 +1018,9 @@ static void __remove_discard_cmd(struct f2fs_sb_info *sbi,
 
 	if (dc->error)
 		printk_ratelimited(
-			"%sF2FS-fs: Issue discard(%u, %u, %u) failed, ret: %d",
-			KERN_INFO, dc->lstart, dc->start, dc->len, dc->error);
+			"%sF2FS-fs (%s): Issue discard(%u, %u, %u) failed, ret: %d",
+			KERN_INFO, sbi->sb->s_id,
+			dc->lstart, dc->start, dc->len, dc->error);
 	__detach_discard_cmd(dcc, dc);
 }
 
@@ -1069,7 +1080,7 @@ static void __init_discard_policy(struct f2fs_sb_info *sbi,
 
 	dpolicy->max_requests = DEF_MAX_DISCARD_REQUEST;
 	dpolicy->io_aware_gran = MAX_PLIST_NUM;
-	dpolicy->timeout = 0;
+	dpolicy->timeout = false;
 
 	if (discard_type == DPOLICY_BG) {
 		dpolicy->min_interval = DEF_MIN_DISCARD_ISSUE_TIME;
@@ -1094,6 +1105,7 @@ static void __init_discard_policy(struct f2fs_sb_info *sbi,
 		dpolicy->io_aware = false;
 		/* we need to issue all to keep CP_TRIMMED_FLAG */
 		dpolicy->granularity = 1;
+		dpolicy->timeout = true;
 	}
 }
 
@@ -1149,7 +1161,7 @@ static int __submit_discard_cmd(struct f2fs_sb_info *sbi,
 		dc->len += len;
 
 		if (time_to_inject(sbi, FAULT_DISCARD)) {
-			f2fs_show_injection_info(FAULT_DISCARD);
+			f2fs_show_injection_info(sbi, FAULT_DISCARD);
 			err = -EIO;
 			goto submit;
 		}
@@ -1462,12 +1474,12 @@ static int __issue_discard_cmd(struct f2fs_sb_info *sbi,
 	int i, issued = 0;
 	bool io_interrupted = false;
 
-	if (dpolicy->timeout != 0)
-		f2fs_update_time(sbi, dpolicy->timeout);
+	if (dpolicy->timeout)
+		f2fs_update_time(sbi, UMOUNT_DISCARD_TIMEOUT);
 
 	for (i = MAX_PLIST_NUM - 1; i >= 0; i--) {
-		if (dpolicy->timeout != 0 &&
-				f2fs_time_over(sbi, dpolicy->timeout))
+		if (dpolicy->timeout &&
+				f2fs_time_over(sbi, UMOUNT_DISCARD_TIMEOUT))
 			break;
 
 		if (i + 1 < dpolicy->granularity)
@@ -1488,8 +1500,8 @@ static int __issue_discard_cmd(struct f2fs_sb_info *sbi,
 		list_for_each_entry_safe(dc, tmp, pend_list, list) {
 			f2fs_bug_on(sbi, dc->state != D_PREP);
 
-			if (dpolicy->timeout != 0 &&
-				f2fs_time_over(sbi, dpolicy->timeout))
+			if (dpolicy->timeout &&
+				f2fs_time_over(sbi, UMOUNT_DISCARD_TIMEOUT))
 				break;
 
 			if (dpolicy->io_aware && i < dpolicy->io_aware_gran &&
@@ -1668,7 +1680,6 @@ bool f2fs_issue_discard_timeout(struct f2fs_sb_info *sbi)
 
 	__init_discard_policy(sbi, &dpolicy, DPOLICY_UMOUNT,
 					dcc->discard_granularity);
-	dpolicy.timeout = UMOUNT_DISCARD_TIMEOUT;
 	__issue_discard_cmd(sbi, &dpolicy);
 	dropped = __drop_discard_cmd(sbi);
 
@@ -1771,7 +1782,8 @@ static int __f2fs_issue_discard_zone(struct f2fs_sb_info *sbi,
 			return -EIO;
 		}
 		trace_f2fs_issue_reset_zone(bdev, blkstart);
-		return blkdev_reset_zones(bdev, sector, nr_sects, GFP_NOFS);
+		return blkdev_zone_mgmt(bdev, REQ_OP_ZONE_RESET,
+					sector, nr_sects, GFP_NOFS);
 	}
 
 	/* For conventional zones, use regular discard if supported */
@@ -1930,7 +1942,7 @@ void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi,
 	unsigned int start = 0, end = -1;
 	unsigned int secno, start_segno;
 	bool force = (cpc->reason & CP_DISCARD);
-	bool need_align = test_opt(sbi, LFS) && __is_large_section(sbi);
+	bool need_align = f2fs_lfs_mode(sbi) && __is_large_section(sbi);
 
 	mutex_lock(&dirty_i->seglist_lock);
 
@@ -1962,7 +1974,7 @@ void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi,
 					(end - 1) <= cpc->trim_end)
 				continue;
 
-		if (!test_opt(sbi, LFS) || !__is_large_section(sbi)) {
+		if (!f2fs_lfs_mode(sbi) || !__is_large_section(sbi)) {
 			f2fs_issue_discard(sbi, START_BLOCK(sbi, start),
 				(end - start) << sbi->log_blocks_per_seg);
 			continue;
@@ -2214,7 +2226,7 @@ void f2fs_invalidate_blocks(struct f2fs_sb_info *sbi, block_t addr)
 	struct sit_info *sit_i = SIT_I(sbi);
 
 	f2fs_bug_on(sbi, addr == NULL_ADDR);
-	if (addr == NEW_ADDR)
+	if (addr == NEW_ADDR || addr == COMPRESS_ADDR)
 		return;
 
 	invalidate_mapping_pages(META_MAPPING(sbi), addr, addr);
@@ -2690,7 +2702,7 @@ unlock:
 	up_read(&SM_I(sbi)->curseg_lock);
 }
 
-void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi)
+void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi, int type)
 {
 	struct curseg_info *curseg;
 	unsigned int old_segno;
@@ -2699,10 +2711,17 @@ void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi)
 	down_write(&SIT_I(sbi)->sentry_lock);
 
 	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++) {
+		if (type != NO_CHECK_TYPE && i != type)
+			continue;
+
 		curseg = CURSEG_I(sbi, i);
-		old_segno = curseg->segno;
-		SIT_I(sbi)->s_ops->allocate_segment(sbi, i, true);
-		locate_dirty_segment(sbi, old_segno);
+		if (type == NO_CHECK_TYPE || curseg->next_blkoff ||
+				get_valid_blocks(sbi, curseg->segno, false) ||
+				get_ckpt_valid_blocks(sbi, curseg->segno)) {
+			old_segno = curseg->segno;
+			SIT_I(sbi)->s_ops->allocate_segment(sbi, i, true);
+			locate_dirty_segment(sbi, old_segno);
+		}
 	}
 
 	up_write(&SIT_I(sbi)->sentry_lock);
@@ -2784,7 +2803,7 @@ next:
 			blk_finish_plug(&plug);
 			mutex_unlock(&dcc->cmd_lock);
 			trimmed += __wait_all_discard_cmd(sbi, NULL);
-			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT);
 			goto next;
 		}
 skip:
@@ -2813,7 +2832,7 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 	struct discard_policy dpolicy;
 	unsigned long long trimmed = 0;
 	int err = 0;
-	bool need_align = test_opt(sbi, LFS) && __is_large_section(sbi);
+	bool need_align = f2fs_lfs_mode(sbi) && __is_large_section(sbi);
 
 	if (start >= MAX_BLKADDR(sbi) || range->len < sbi->blocksize)
 		return -EINVAL;
@@ -2843,9 +2862,9 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 	if (sbi->discard_blks == 0)
 		goto out;
 
-	mutex_lock(&sbi->gc_mutex);
+	down_write(&sbi->gc_lock);
 	err = f2fs_write_checkpoint(sbi, &cpc);
-	mutex_unlock(&sbi->gc_mutex);
+	up_write(&sbi->gc_lock);
 	if (err)
 		goto out;
 
@@ -3018,7 +3037,8 @@ static int __get_segment_type_6(struct f2fs_io_info *fio)
 	if (fio->type == DATA) {
 		struct inode *inode = fio->page->mapping->host;
 
-		if (is_cold_data(fio->page) || file_is_cold(inode))
+		if (is_cold_data(fio->page) || file_is_cold(inode) ||
+				f2fs_compressed_file(inode))
 			return CURSEG_COLD_DATA;
 		if (file_is_hot(inode) ||
 				is_inode_flag_set(inode, FI_HOT_DATA) ||
@@ -3068,6 +3088,19 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 {
 	struct sit_info *sit_i = SIT_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, type);
+	bool put_pin_sem = false;
+
+	if (type == CURSEG_COLD_DATA) {
+		/* GC during CURSEG_COLD_DATA_PINNED allocation */
+		if (down_read_trylock(&sbi->pin_sem)) {
+			put_pin_sem = true;
+		} else {
+			type = CURSEG_WARM_DATA;
+			curseg = CURSEG_I(sbi, type);
+		}
+	} else if (type == CURSEG_COLD_DATA_PINNED) {
+		type = CURSEG_COLD_DATA;
+	}
 
 	down_read(&SM_I(sbi)->curseg_lock);
 
@@ -3133,6 +3166,9 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 	mutex_unlock(&curseg->curseg_mutex);
 
 	up_read(&SM_I(sbi)->curseg_lock);
+
+	if (put_pin_sem)
+		up_read(&sbi->pin_sem);
 }
 
 static void update_device_state(struct f2fs_io_info *fio)
@@ -3159,7 +3195,7 @@ static void update_device_state(struct f2fs_io_info *fio)
 static void do_write_page(struct f2fs_summary *sum, struct f2fs_io_info *fio)
 {
 	int type = __get_segment_type(fio);
-	bool keep_order = (test_opt(fio->sbi, LFS) && type == CURSEG_COLD_DATA);
+	bool keep_order = (f2fs_lfs_mode(fio->sbi) && type == CURSEG_COLD_DATA);
 
 	if (keep_order)
 		down_read(&fio->sbi->io_order_lock);
@@ -3255,7 +3291,7 @@ int f2fs_inplace_write_data(struct f2fs_io_info *fio)
 
 	stat_inc_inplace_blocks(fio->sbi);
 
-	if (fio->bio)
+	if (fio->bio && !(SM_I(sbi)->ipu_policy & (1 << F2FS_IPU_NOCACHE)))
 		err = f2fs_merge_page_bio(fio);
 	else
 		err = f2fs_submit_page_bio(fio);
@@ -3379,7 +3415,10 @@ void f2fs_wait_on_page_writeback(struct page *page,
 	if (PageWriteback(page)) {
 		struct f2fs_sb_info *sbi = F2FS_P_SB(page);
 
+		/* submit cached LFS IO */
 		f2fs_submit_merged_write_cond(sbi, NULL, page, 0, type);
+		/* sbumit cached IPU IO */
+		f2fs_submit_merged_ipu_write(sbi, NULL, page);
 		if (ordered) {
 			wait_on_page_writeback(page);
 			f2fs_bug_on(sbi, locked && PageWriteback(page));
@@ -4034,7 +4073,7 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
 	sit_i->dirty_sentries = 0;
 	sit_i->sents_per_block = SIT_ENTRY_PER_BLOCK;
 	sit_i->elapsed_time = le64_to_cpu(sbi->ckpt->elapsed_time);
-	sit_i->mounted_time = ktime_get_real_seconds();
+	sit_i->mounted_time = ktime_get_boottime_seconds();
 	init_rwsem(&sit_i->sentry_lock);
 	return 0;
 }
@@ -4331,6 +4370,263 @@ out:
 	return 0;
 }
 
+#ifdef CONFIG_BLK_DEV_ZONED
+
+static int check_zone_write_pointer(struct f2fs_sb_info *sbi,
+				    struct f2fs_dev_info *fdev,
+				    struct blk_zone *zone)
+{
+	unsigned int wp_segno, wp_blkoff, zone_secno, zone_segno, segno;
+	block_t zone_block, wp_block, last_valid_block;
+	unsigned int log_sectors_per_block = sbi->log_blocksize - SECTOR_SHIFT;
+	int i, s, b, ret;
+	struct seg_entry *se;
+
+	if (zone->type != BLK_ZONE_TYPE_SEQWRITE_REQ)
+		return 0;
+
+	wp_block = fdev->start_blk + (zone->wp >> log_sectors_per_block);
+	wp_segno = GET_SEGNO(sbi, wp_block);
+	wp_blkoff = wp_block - START_BLOCK(sbi, wp_segno);
+	zone_block = fdev->start_blk + (zone->start >> log_sectors_per_block);
+	zone_segno = GET_SEGNO(sbi, zone_block);
+	zone_secno = GET_SEC_FROM_SEG(sbi, zone_segno);
+
+	if (zone_segno >= MAIN_SEGS(sbi))
+		return 0;
+
+	/*
+	 * Skip check of zones cursegs point to, since
+	 * fix_curseg_write_pointer() checks them.
+	 */
+	for (i = 0; i < NO_CHECK_TYPE; i++)
+		if (zone_secno == GET_SEC_FROM_SEG(sbi,
+						   CURSEG_I(sbi, i)->segno))
+			return 0;
+
+	/*
+	 * Get last valid block of the zone.
+	 */
+	last_valid_block = zone_block - 1;
+	for (s = sbi->segs_per_sec - 1; s >= 0; s--) {
+		segno = zone_segno + s;
+		se = get_seg_entry(sbi, segno);
+		for (b = sbi->blocks_per_seg - 1; b >= 0; b--)
+			if (f2fs_test_bit(b, se->cur_valid_map)) {
+				last_valid_block = START_BLOCK(sbi, segno) + b;
+				break;
+			}
+		if (last_valid_block >= zone_block)
+			break;
+	}
+
+	/*
+	 * If last valid block is beyond the write pointer, report the
+	 * inconsistency. This inconsistency does not cause write error
+	 * because the zone will not be selected for write operation until
+	 * it get discarded. Just report it.
+	 */
+	if (last_valid_block >= wp_block) {
+		f2fs_notice(sbi, "Valid block beyond write pointer: "
+			    "valid block[0x%x,0x%x] wp[0x%x,0x%x]",
+			    GET_SEGNO(sbi, last_valid_block),
+			    GET_BLKOFF_FROM_SEG0(sbi, last_valid_block),
+			    wp_segno, wp_blkoff);
+		return 0;
+	}
+
+	/*
+	 * If there is no valid block in the zone and if write pointer is
+	 * not at zone start, reset the write pointer.
+	 */
+	if (last_valid_block + 1 == zone_block && zone->wp != zone->start) {
+		f2fs_notice(sbi,
+			    "Zone without valid block has non-zero write "
+			    "pointer. Reset the write pointer: wp[0x%x,0x%x]",
+			    wp_segno, wp_blkoff);
+		ret = __f2fs_issue_discard_zone(sbi, fdev->bdev, zone_block,
+					zone->len >> log_sectors_per_block);
+		if (ret) {
+			f2fs_err(sbi, "Discard zone failed: %s (errno=%d)",
+				 fdev->path, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static struct f2fs_dev_info *get_target_zoned_dev(struct f2fs_sb_info *sbi,
+						  block_t zone_blkaddr)
+{
+	int i;
+
+	for (i = 0; i < sbi->s_ndevs; i++) {
+		if (!bdev_is_zoned(FDEV(i).bdev))
+			continue;
+		if (sbi->s_ndevs == 1 || (FDEV(i).start_blk <= zone_blkaddr &&
+				zone_blkaddr <= FDEV(i).end_blk))
+			return &FDEV(i);
+	}
+
+	return NULL;
+}
+
+static int report_one_zone_cb(struct blk_zone *zone, unsigned int idx,
+			      void *data) {
+	memcpy(data, zone, sizeof(struct blk_zone));
+	return 0;
+}
+
+static int fix_curseg_write_pointer(struct f2fs_sb_info *sbi, int type)
+{
+	struct curseg_info *cs = CURSEG_I(sbi, type);
+	struct f2fs_dev_info *zbd;
+	struct blk_zone zone;
+	unsigned int cs_section, wp_segno, wp_blkoff, wp_sector_off;
+	block_t cs_zone_block, wp_block;
+	unsigned int log_sectors_per_block = sbi->log_blocksize - SECTOR_SHIFT;
+	sector_t zone_sector;
+	int err;
+
+	cs_section = GET_SEC_FROM_SEG(sbi, cs->segno);
+	cs_zone_block = START_BLOCK(sbi, GET_SEG_FROM_SEC(sbi, cs_section));
+
+	zbd = get_target_zoned_dev(sbi, cs_zone_block);
+	if (!zbd)
+		return 0;
+
+	/* report zone for the sector the curseg points to */
+	zone_sector = (sector_t)(cs_zone_block - zbd->start_blk)
+		<< log_sectors_per_block;
+	err = blkdev_report_zones(zbd->bdev, zone_sector, 1,
+				  report_one_zone_cb, &zone);
+	if (err != 1) {
+		f2fs_err(sbi, "Report zone failed: %s errno=(%d)",
+			 zbd->path, err);
+		return err;
+	}
+
+	if (zone.type != BLK_ZONE_TYPE_SEQWRITE_REQ)
+		return 0;
+
+	wp_block = zbd->start_blk + (zone.wp >> log_sectors_per_block);
+	wp_segno = GET_SEGNO(sbi, wp_block);
+	wp_blkoff = wp_block - START_BLOCK(sbi, wp_segno);
+	wp_sector_off = zone.wp & GENMASK(log_sectors_per_block - 1, 0);
+
+	if (cs->segno == wp_segno && cs->next_blkoff == wp_blkoff &&
+		wp_sector_off == 0)
+		return 0;
+
+	f2fs_notice(sbi, "Unaligned curseg[%d] with write pointer: "
+		    "curseg[0x%x,0x%x] wp[0x%x,0x%x]",
+		    type, cs->segno, cs->next_blkoff, wp_segno, wp_blkoff);
+
+	f2fs_notice(sbi, "Assign new section to curseg[%d]: "
+		    "curseg[0x%x,0x%x]", type, cs->segno, cs->next_blkoff);
+	allocate_segment_by_default(sbi, type, true);
+
+	/* check consistency of the zone curseg pointed to */
+	if (check_zone_write_pointer(sbi, zbd, &zone))
+		return -EIO;
+
+	/* check newly assigned zone */
+	cs_section = GET_SEC_FROM_SEG(sbi, cs->segno);
+	cs_zone_block = START_BLOCK(sbi, GET_SEG_FROM_SEC(sbi, cs_section));
+
+	zbd = get_target_zoned_dev(sbi, cs_zone_block);
+	if (!zbd)
+		return 0;
+
+	zone_sector = (sector_t)(cs_zone_block - zbd->start_blk)
+		<< log_sectors_per_block;
+	err = blkdev_report_zones(zbd->bdev, zone_sector, 1,
+				  report_one_zone_cb, &zone);
+	if (err != 1) {
+		f2fs_err(sbi, "Report zone failed: %s errno=(%d)",
+			 zbd->path, err);
+		return err;
+	}
+
+	if (zone.type != BLK_ZONE_TYPE_SEQWRITE_REQ)
+		return 0;
+
+	if (zone.wp != zone.start) {
+		f2fs_notice(sbi,
+			    "New zone for curseg[%d] is not yet discarded. "
+			    "Reset the zone: curseg[0x%x,0x%x]",
+			    type, cs->segno, cs->next_blkoff);
+		err = __f2fs_issue_discard_zone(sbi, zbd->bdev,
+				zone_sector >> log_sectors_per_block,
+				zone.len >> log_sectors_per_block);
+		if (err) {
+			f2fs_err(sbi, "Discard zone failed: %s (errno=%d)",
+				 zbd->path, err);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+int f2fs_fix_curseg_write_pointer(struct f2fs_sb_info *sbi)
+{
+	int i, ret;
+
+	for (i = 0; i < NO_CHECK_TYPE; i++) {
+		ret = fix_curseg_write_pointer(sbi, i);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+struct check_zone_write_pointer_args {
+	struct f2fs_sb_info *sbi;
+	struct f2fs_dev_info *fdev;
+};
+
+static int check_zone_write_pointer_cb(struct blk_zone *zone, unsigned int idx,
+				      void *data) {
+	struct check_zone_write_pointer_args *args;
+	args = (struct check_zone_write_pointer_args *)data;
+
+	return check_zone_write_pointer(args->sbi, args->fdev, zone);
+}
+
+int f2fs_check_write_pointer(struct f2fs_sb_info *sbi)
+{
+	int i, ret;
+	struct check_zone_write_pointer_args args;
+
+	for (i = 0; i < sbi->s_ndevs; i++) {
+		if (!bdev_is_zoned(FDEV(i).bdev))
+			continue;
+
+		args.sbi = sbi;
+		args.fdev = &FDEV(i);
+		ret = blkdev_report_zones(FDEV(i).bdev, 0, BLK_ALL_ZONES,
+					  check_zone_write_pointer_cb, &args);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+#else
+int f2fs_fix_curseg_write_pointer(struct f2fs_sb_info *sbi)
+{
+	return 0;
+}
+
+int f2fs_check_write_pointer(struct f2fs_sb_info *sbi)
+{
+	return 0;
+}
+#endif
+
 /*
  * Update min, max modified time for cost-benefit GC algorithm
  */
@@ -4384,7 +4680,7 @@ int f2fs_build_segment_manager(struct f2fs_sb_info *sbi)
 	if (sm_info->rec_prefree_segments > DEF_MAX_RECLAIM_PREFREE_SEGMENTS)
 		sm_info->rec_prefree_segments = DEF_MAX_RECLAIM_PREFREE_SEGMENTS;
 
-	if (!test_opt(sbi, LFS))
+	if (!f2fs_lfs_mode(sbi))
 		sm_info->ipu_policy = 1 << F2FS_IPU_FSYNC;
 	sm_info->min_ipu_util = DEF_MIN_IPU_UTIL;
 	sm_info->min_fsync_blocks = DEF_MIN_FSYNC_BLOCKS;
@@ -4536,22 +4832,22 @@ void f2fs_destroy_segment_manager(struct f2fs_sb_info *sbi)
 
 int __init f2fs_create_segment_manager_caches(void)
 {
-	discard_entry_slab = f2fs_kmem_cache_create("discard_entry",
+	discard_entry_slab = f2fs_kmem_cache_create("f2fs_discard_entry",
 			sizeof(struct discard_entry));
 	if (!discard_entry_slab)
 		goto fail;
 
-	discard_cmd_slab = f2fs_kmem_cache_create("discard_cmd",
+	discard_cmd_slab = f2fs_kmem_cache_create("f2fs_discard_cmd",
 			sizeof(struct discard_cmd));
 	if (!discard_cmd_slab)
 		goto destroy_discard_entry;
 
-	sit_entry_set_slab = f2fs_kmem_cache_create("sit_entry_set",
+	sit_entry_set_slab = f2fs_kmem_cache_create("f2fs_sit_entry_set",
 			sizeof(struct sit_entry_set));
 	if (!sit_entry_set_slab)
 		goto destroy_discard_cmd;
 
-	inmem_entry_slab = f2fs_kmem_cache_create("inmem_page_entry",
+	inmem_entry_slab = f2fs_kmem_cache_create("f2fs_inmem_page_entry",
 			sizeof(struct inmem_pages));
 	if (!inmem_entry_slab)
 		goto destroy_sit_entry_set;
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index 325781a1ae4d..7a83bd530812 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -200,18 +200,6 @@ struct segment_allocation {
 	void (*allocate_segment)(struct f2fs_sb_info *, int, bool);
 };
 
-/*
- * this value is set in page as a private data which indicate that
- * the page is atomically written, and it is in inmem_pages list.
- */
-#define ATOMIC_WRITTEN_PAGE		((unsigned long)-1)
-#define DUMMY_WRITTEN_PAGE		((unsigned long)-2)
-
-#define IS_ATOMIC_WRITTEN_PAGE(page)			\
-		(page_private(page) == (unsigned long)ATOMIC_WRITTEN_PAGE)
-#define IS_DUMMY_WRITTEN_PAGE(page)			\
-		(page_private(page) == (unsigned long)DUMMY_WRITTEN_PAGE)
-
 #define MAX_SKIP_GC_COUNT			16
 
 struct inmem_pages {
@@ -313,6 +301,8 @@ struct sit_entry_set {
  */
 static inline struct curseg_info *CURSEG_I(struct f2fs_sb_info *sbi, int type)
 {
+	if (type == CURSEG_COLD_DATA_PINNED)
+		type = CURSEG_COLD_DATA;
 	return (struct curseg_info *)(SM_I(sbi)->curseg_array + type);
 }
 
@@ -617,8 +607,10 @@ static inline int utilization(struct f2fs_sb_info *sbi)
  *                     threashold,
  * F2FS_IPU_FSYNC - activated in fsync path only for high performance flash
  *                     storages. IPU will be triggered only if the # of dirty
- *                     pages over min_fsync_blocks.
- * F2FS_IPUT_DISABLE - disable IPU. (=default option)
+ *                     pages over min_fsync_blocks. (=default option)
+ * F2FS_IPU_ASYNC - do IPU given by asynchronous write requests.
+ * F2FS_IPU_NOCACHE - disable IPU bio cache.
+ * F2FS_IPUT_DISABLE - disable IPU. (=default option in LFS mode)
  */
 #define DEF_MIN_IPU_UTIL	70
 #define DEF_MIN_FSYNC_BLOCKS	8
@@ -633,6 +625,7 @@ enum {
 	F2FS_IPU_SSR_UTIL,
 	F2FS_IPU_FSYNC,
 	F2FS_IPU_ASYNC,
+	F2FS_IPU_NOCACHE,
 };
 
 static inline unsigned int curseg_segno(struct f2fs_sb_info *sbi,
@@ -763,7 +756,7 @@ static inline unsigned long long get_mtime(struct f2fs_sb_info *sbi,
 						bool base_time)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
-	time64_t diff, now = ktime_get_real_seconds();
+	time64_t diff, now = ktime_get_boottime_seconds();
 
 	if (now >= sit_i->mounted_time)
 		return sit_i->elapsed_time + now - sit_i->mounted_time;
diff --git a/fs/f2fs/shrinker.c b/fs/f2fs/shrinker.c
index a467aca29cfe..d66de5999a26 100644
--- a/fs/f2fs/shrinker.c
+++ b/fs/f2fs/shrinker.c
@@ -58,7 +58,7 @@ unsigned long f2fs_shrink_count(struct shrinker *shrink,
 		/* count extent cache entries */
 		count += __count_extent_cache(sbi);
 
-		/* shrink clean nat cache entries */
+		/* count clean nat cache entries */
 		count += __count_nat_entries(sbi);
 
 		/* count free nids cache entries */
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 1443cee15863..8a9955902d84 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -24,6 +24,7 @@
 #include <linux/sysfs.h>
 #include <linux/quota.h>
 #include <linux/unicode.h>
+#include <linux/part_stat.h>
 
 #include "f2fs.h"
 #include "node.h"
@@ -141,6 +142,9 @@ enum {
 	Opt_checkpoint_disable_cap,
 	Opt_checkpoint_disable_cap_perc,
 	Opt_checkpoint_enable,
+	Opt_compress_algorithm,
+	Opt_compress_log_size,
+	Opt_compress_extension,
 	Opt_err,
 };
 
@@ -198,11 +202,15 @@ static match_table_t f2fs_tokens = {
 	{Opt_whint, "whint_mode=%s"},
 	{Opt_alloc, "alloc_mode=%s"},
 	{Opt_fsync, "fsync_mode=%s"},
+	{Opt_test_dummy_encryption, "test_dummy_encryption=%s"},
 	{Opt_test_dummy_encryption, "test_dummy_encryption"},
 	{Opt_checkpoint_disable, "checkpoint=disable"},
 	{Opt_checkpoint_disable_cap, "checkpoint=disable:%u"},
 	{Opt_checkpoint_disable_cap_perc, "checkpoint=disable:%u%%"},
 	{Opt_checkpoint_enable, "checkpoint=enable"},
+	{Opt_compress_algorithm, "compress_algorithm=%s"},
+	{Opt_compress_log_size, "compress_log_size=%u"},
+	{Opt_compress_extension, "compress_extension=%s"},
 	{Opt_err, NULL},
 };
 
@@ -387,17 +395,61 @@ static int f2fs_check_quota_options(struct f2fs_sb_info *sbi)
 }
 #endif
 
-static int parse_options(struct super_block *sb, char *options)
+static int f2fs_set_test_dummy_encryption(struct super_block *sb,
+					  const char *opt,
+					  const substring_t *arg,
+					  bool is_remount)
+{
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+#ifdef CONFIG_FS_ENCRYPTION
+	int err;
+
+	if (!f2fs_sb_has_encrypt(sbi)) {
+		f2fs_err(sbi, "Encrypt feature is off");
+		return -EINVAL;
+	}
+
+	/*
+	 * This mount option is just for testing, and it's not worthwhile to
+	 * implement the extra complexity (e.g. RCU protection) that would be
+	 * needed to allow it to be set or changed during remount.  We do allow
+	 * it to be specified during remount, but only if there is no change.
+	 */
+	if (is_remount && !F2FS_OPTION(sbi).dummy_enc_ctx.ctx) {
+		f2fs_warn(sbi, "Can't set test_dummy_encryption on remount");
+		return -EINVAL;
+	}
+	err = fscrypt_set_test_dummy_encryption(
+		sb, arg, &F2FS_OPTION(sbi).dummy_enc_ctx);
+	if (err) {
+		if (err == -EEXIST)
+			f2fs_warn(sbi,
+				  "Can't change test_dummy_encryption on remount");
+		else if (err == -EINVAL)
+			f2fs_warn(sbi, "Value of option \"%s\" is unrecognized",
+				  opt);
+		else
+			f2fs_warn(sbi, "Error processing option \"%s\" [%d]",
+				  opt, err);
+		return -EINVAL;
+	}
+	f2fs_warn(sbi, "Test dummy encryption mode enabled");
+#else
+	f2fs_warn(sbi, "Test dummy encryption mount option ignored");
+#endif
+	return 0;
+}
+
+static int parse_options(struct super_block *sb, char *options, bool is_remount)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(sb);
 	substring_t args[MAX_OPT_ARGS];
+	unsigned char (*ext)[F2FS_EXTENSION_LEN];
 	char *p, *name;
-	int arg = 0;
+	int arg = 0, ext_cnt;
 	kuid_t uid;
 	kgid_t gid;
-#ifdef CONFIG_QUOTA
 	int ret;
-#endif
 
 	if (!options)
 		return 0;
@@ -420,14 +472,11 @@ static int parse_options(struct super_block *sb, char *options)
 			if (!name)
 				return -ENOMEM;
 			if (strlen(name) == 2 && !strncmp(name, "on", 2)) {
-				set_opt(sbi, BG_GC);
-				clear_opt(sbi, FORCE_FG_GC);
+				F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_ON;
 			} else if (strlen(name) == 3 && !strncmp(name, "off", 3)) {
-				clear_opt(sbi, BG_GC);
-				clear_opt(sbi, FORCE_FG_GC);
+				F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_OFF;
 			} else if (strlen(name) == 4 && !strncmp(name, "sync", 4)) {
-				set_opt(sbi, BG_GC);
-				set_opt(sbi, FORCE_FG_GC);
+				F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_SYNC;
 			} else {
 				kvfree(name);
 				return -EINVAL;
@@ -439,7 +488,7 @@ static int parse_options(struct super_block *sb, char *options)
 			break;
 		case Opt_norecovery:
 			/* this option mounts f2fs with ro */
-			set_opt(sbi, DISABLE_ROLL_FORWARD);
+			set_opt(sbi, NORECOVERY);
 			if (!f2fs_readonly(sb))
 				return -EINVAL;
 			break;
@@ -593,10 +642,10 @@ static int parse_options(struct super_block *sb, char *options)
 					kvfree(name);
 					return -EINVAL;
 				}
-				set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
+				F2FS_OPTION(sbi).fs_mode = FS_MODE_ADAPTIVE;
 			} else if (strlen(name) == 3 &&
 					!strncmp(name, "lfs", 3)) {
-				set_opt_mode(sbi, F2FS_MOUNT_LFS);
+				F2FS_OPTION(sbi).fs_mode = FS_MODE_LFS;
 			} else {
 				kvfree(name);
 				return -EINVAL;
@@ -773,17 +822,10 @@ static int parse_options(struct super_block *sb, char *options)
 			kvfree(name);
 			break;
 		case Opt_test_dummy_encryption:
-#ifdef CONFIG_FS_ENCRYPTION
-			if (!f2fs_sb_has_encrypt(sbi)) {
-				f2fs_err(sbi, "Encrypt feature is off");
-				return -EINVAL;
-			}
-
-			F2FS_OPTION(sbi).test_dummy_encryption = true;
-			f2fs_info(sbi, "Test dummy encryption mode enabled");
-#else
-			f2fs_info(sbi, "Test dummy encryption mount option ignored");
-#endif
+			ret = f2fs_set_test_dummy_encryption(sb, p, &args[0],
+							     is_remount);
+			if (ret)
+				return ret;
 			break;
 		case Opt_checkpoint_disable_cap_perc:
 			if (args->from && match_int(args, &arg))
@@ -810,6 +852,70 @@ static int parse_options(struct super_block *sb, char *options)
 		case Opt_checkpoint_enable:
 			clear_opt(sbi, DISABLE_CHECKPOINT);
 			break;
+		case Opt_compress_algorithm:
+			if (!f2fs_sb_has_compression(sbi)) {
+				f2fs_err(sbi, "Compression feature if off");
+				return -EINVAL;
+			}
+			name = match_strdup(&args[0]);
+			if (!name)
+				return -ENOMEM;
+			if (strlen(name) == 3 && !strcmp(name, "lzo")) {
+				F2FS_OPTION(sbi).compress_algorithm =
+								COMPRESS_LZO;
+			} else if (strlen(name) == 3 &&
+					!strcmp(name, "lz4")) {
+				F2FS_OPTION(sbi).compress_algorithm =
+								COMPRESS_LZ4;
+			} else if (strlen(name) == 4 &&
+					!strcmp(name, "zstd")) {
+				F2FS_OPTION(sbi).compress_algorithm =
+								COMPRESS_ZSTD;
+			} else {
+				kfree(name);
+				return -EINVAL;
+			}
+			kfree(name);
+			break;
+		case Opt_compress_log_size:
+			if (!f2fs_sb_has_compression(sbi)) {
+				f2fs_err(sbi, "Compression feature is off");
+				return -EINVAL;
+			}
+			if (args->from && match_int(args, &arg))
+				return -EINVAL;
+			if (arg < MIN_COMPRESS_LOG_SIZE ||
+				arg > MAX_COMPRESS_LOG_SIZE) {
+				f2fs_err(sbi,
+					"Compress cluster log size is out of range");
+				return -EINVAL;
+			}
+			F2FS_OPTION(sbi).compress_log_size = arg;
+			break;
+		case Opt_compress_extension:
+			if (!f2fs_sb_has_compression(sbi)) {
+				f2fs_err(sbi, "Compression feature is off");
+				return -EINVAL;
+			}
+			name = match_strdup(&args[0]);
+			if (!name)
+				return -ENOMEM;
+
+			ext = F2FS_OPTION(sbi).extensions;
+			ext_cnt = F2FS_OPTION(sbi).compress_ext_cnt;
+
+			if (strlen(name) >= F2FS_EXTENSION_LEN ||
+				ext_cnt >= COMPRESS_EXT_NUM) {
+				f2fs_err(sbi,
+					"invalid extension length/number");
+				kfree(name);
+				return -EINVAL;
+			}
+
+			strcpy(ext[ext_cnt], name);
+			F2FS_OPTION(sbi).compress_ext_cnt++;
+			kfree(name);
+			break;
 		default:
 			f2fs_err(sbi, "Unrecognized mount option \"%s\" or missing value",
 				 p);
@@ -837,7 +943,7 @@ static int parse_options(struct super_block *sb, char *options)
 	}
 #endif
 
-	if (F2FS_IO_SIZE_BITS(sbi) && !test_opt(sbi, LFS)) {
+	if (F2FS_IO_SIZE_BITS(sbi) && !f2fs_lfs_mode(sbi)) {
 		f2fs_err(sbi, "Should set mode=lfs with %uKB-sized IO",
 			 F2FS_IO_SIZE_KB(sbi));
 		return -EINVAL;
@@ -867,7 +973,7 @@ static int parse_options(struct super_block *sb, char *options)
 		}
 	}
 
-	if (test_opt(sbi, DISABLE_CHECKPOINT) && test_opt(sbi, LFS)) {
+	if (test_opt(sbi, DISABLE_CHECKPOINT) && f2fs_lfs_mode(sbi)) {
 		f2fs_err(sbi, "LFS not compatible with checkpoint=disable\n");
 		return -EINVAL;
 	}
@@ -893,6 +999,7 @@ static struct inode *f2fs_alloc_inode(struct super_block *sb)
 	/* Initialize f2fs-specific inode info */
 	atomic_set(&fi->dirty_pages, 0);
 	init_rwsem(&fi->i_sem);
+	spin_lock_init(&fi->i_size_lock);
 	INIT_LIST_HEAD(&fi->dirty_list);
 	INIT_LIST_HEAD(&fi->gdirty_list);
 	INIT_LIST_HEAD(&fi->inmem_ilist);
@@ -1105,7 +1212,7 @@ static void f2fs_put_super(struct super_block *sb)
 	/* our cp_error case, we can wait for any writeback page */
 	f2fs_flush_merged_writes(sbi);
 
-	f2fs_wait_on_all_pages_writeback(sbi);
+	f2fs_wait_on_all_pages(sbi, F2FS_WB_CP_DATA);
 
 	f2fs_bug_on(sbi, sbi->fsync_node_num);
 
@@ -1125,6 +1232,8 @@ static void f2fs_put_super(struct super_block *sb)
 	f2fs_destroy_node_manager(sbi);
 	f2fs_destroy_segment_manager(sbi);
 
+	f2fs_destroy_post_read_wq(sbi);
+
 	kvfree(sbi->ckpt);
 
 	f2fs_unregister_sysfs(sbi);
@@ -1135,11 +1244,13 @@ static void f2fs_put_super(struct super_block *sb)
 	kvfree(sbi->raw_super);
 
 	destroy_device_list(sbi);
+	f2fs_destroy_xattr_caches(sbi);
 	mempool_destroy(sbi->write_io_dummy);
 #ifdef CONFIG_QUOTA
 	for (i = 0; i < MAXQUOTAS; i++)
 		kvfree(F2FS_OPTION(sbi).s_qf_names[i]);
 #endif
+	fscrypt_free_dummy_context(&F2FS_OPTION(sbi).dummy_enc_ctx);
 	destroy_percpu_info(sbi);
 	for (i = 0; i < NR_PAGE_TYPE; i++)
 		kvfree(sbi->write_io[i]);
@@ -1169,9 +1280,9 @@ int f2fs_sync_fs(struct super_block *sb, int sync)
 
 		cpc.reason = __get_cp_reason(sbi);
 
-		mutex_lock(&sbi->gc_mutex);
+		down_write(&sbi->gc_lock);
 		err = f2fs_write_checkpoint(sbi, &cpc);
-		mutex_unlock(&sbi->gc_mutex);
+		up_write(&sbi->gc_lock);
 	}
 	f2fs_trace_ios(NULL, 1);
 
@@ -1213,9 +1324,11 @@ static int f2fs_statfs_project(struct super_block *sb,
 		return PTR_ERR(dquot);
 	spin_lock(&dquot->dq_dqb_lock);
 
-	limit = (dquot->dq_dqb.dqb_bsoftlimit ?
-		 dquot->dq_dqb.dqb_bsoftlimit :
-		 dquot->dq_dqb.dqb_bhardlimit) >> sb->s_blocksize_bits;
+	limit = min_not_zero(dquot->dq_dqb.dqb_bsoftlimit,
+					dquot->dq_dqb.dqb_bhardlimit);
+	if (limit)
+		limit >>= sb->s_blocksize_bits;
+
 	if (limit && buf->f_blocks > limit) {
 		curblock = dquot->dq_dqb.dqb_curspace >> sb->s_blocksize_bits;
 		buf->f_blocks = limit;
@@ -1224,9 +1337,9 @@ static int f2fs_statfs_project(struct super_block *sb,
 			 (buf->f_blocks - curblock) : 0;
 	}
 
-	limit = dquot->dq_dqb.dqb_isoftlimit ?
-		dquot->dq_dqb.dqb_isoftlimit :
-		dquot->dq_dqb.dqb_ihardlimit;
+	limit = min_not_zero(dquot->dq_dqb.dqb_isoftlimit,
+					dquot->dq_dqb.dqb_ihardlimit);
+
 	if (limit && buf->f_files > limit) {
 		buf->f_files = limit;
 		buf->f_ffree =
@@ -1332,20 +1445,53 @@ static inline void f2fs_show_quota_options(struct seq_file *seq,
 #endif
 }
 
+static inline void f2fs_show_compress_options(struct seq_file *seq,
+							struct super_block *sb)
+{
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+	char *algtype = "";
+	int i;
+
+	if (!f2fs_sb_has_compression(sbi))
+		return;
+
+	switch (F2FS_OPTION(sbi).compress_algorithm) {
+	case COMPRESS_LZO:
+		algtype = "lzo";
+		break;
+	case COMPRESS_LZ4:
+		algtype = "lz4";
+		break;
+	case COMPRESS_ZSTD:
+		algtype = "zstd";
+		break;
+	}
+	seq_printf(seq, ",compress_algorithm=%s", algtype);
+
+	seq_printf(seq, ",compress_log_size=%u",
+			F2FS_OPTION(sbi).compress_log_size);
+
+	for (i = 0; i < F2FS_OPTION(sbi).compress_ext_cnt; i++) {
+		seq_printf(seq, ",compress_extension=%s",
+			F2FS_OPTION(sbi).extensions[i]);
+	}
+}
+
 static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(root->d_sb);
 
-	if (!f2fs_readonly(sbi->sb) && test_opt(sbi, BG_GC)) {
-		if (test_opt(sbi, FORCE_FG_GC))
-			seq_printf(seq, ",background_gc=%s", "sync");
-		else
-			seq_printf(seq, ",background_gc=%s", "on");
-	} else {
+	if (F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_SYNC)
+		seq_printf(seq, ",background_gc=%s", "sync");
+	else if (F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_ON)
+		seq_printf(seq, ",background_gc=%s", "on");
+	else if (F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_OFF)
 		seq_printf(seq, ",background_gc=%s", "off");
-	}
+
 	if (test_opt(sbi, DISABLE_ROLL_FORWARD))
 		seq_puts(seq, ",disable_roll_forward");
+	if (test_opt(sbi, NORECOVERY))
+		seq_puts(seq, ",norecovery");
 	if (test_opt(sbi, DISCARD))
 		seq_puts(seq, ",discard");
 	else
@@ -1397,9 +1543,9 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 		seq_puts(seq, ",data_flush");
 
 	seq_puts(seq, ",mode=");
-	if (test_opt(sbi, ADAPTIVE))
+	if (F2FS_OPTION(sbi).fs_mode == FS_MODE_ADAPTIVE)
 		seq_puts(seq, "adaptive");
-	else if (test_opt(sbi, LFS))
+	else if (F2FS_OPTION(sbi).fs_mode == FS_MODE_LFS)
 		seq_puts(seq, "lfs");
 	seq_printf(seq, ",active_logs=%u", F2FS_OPTION(sbi).active_logs);
 	if (test_opt(sbi, RESERVE_ROOT))
@@ -1435,10 +1581,8 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 		seq_printf(seq, ",whint_mode=%s", "user-based");
 	else if (F2FS_OPTION(sbi).whint_mode == WHINT_MODE_FS)
 		seq_printf(seq, ",whint_mode=%s", "fs-based");
-#ifdef CONFIG_FS_ENCRYPTION
-	if (F2FS_OPTION(sbi).test_dummy_encryption)
-		seq_puts(seq, ",test_dummy_encryption");
-#endif
+
+	fscrypt_show_test_dummy_encryption(seq, ',', sbi->sb);
 
 	if (F2FS_OPTION(sbi).alloc_mode == ALLOC_MODE_DEFAULT)
 		seq_printf(seq, ",alloc_mode=%s", "default");
@@ -1454,6 +1598,8 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 		seq_printf(seq, ",fsync_mode=%s", "strict");
 	else if (F2FS_OPTION(sbi).fsync_mode == FSYNC_MODE_NOBARRIER)
 		seq_printf(seq, ",fsync_mode=%s", "nobarrier");
+
+	f2fs_show_compress_options(seq, sbi->sb);
 	return 0;
 }
 
@@ -1465,11 +1611,13 @@ static void default_options(struct f2fs_sb_info *sbi)
 	F2FS_OPTION(sbi).whint_mode = WHINT_MODE_OFF;
 	F2FS_OPTION(sbi).alloc_mode = ALLOC_MODE_DEFAULT;
 	F2FS_OPTION(sbi).fsync_mode = FSYNC_MODE_POSIX;
-	F2FS_OPTION(sbi).test_dummy_encryption = false;
 	F2FS_OPTION(sbi).s_resuid = make_kuid(&init_user_ns, F2FS_DEF_RESUID);
 	F2FS_OPTION(sbi).s_resgid = make_kgid(&init_user_ns, F2FS_DEF_RESGID);
+	F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZ4;
+	F2FS_OPTION(sbi).compress_log_size = MIN_COMPRESS_LOG_SIZE;
+	F2FS_OPTION(sbi).compress_ext_cnt = 0;
+	F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_ON;
 
-	set_opt(sbi, BG_GC);
 	set_opt(sbi, INLINE_XATTR);
 	set_opt(sbi, INLINE_DATA);
 	set_opt(sbi, INLINE_DENTRY);
@@ -1481,9 +1629,9 @@ static void default_options(struct f2fs_sb_info *sbi)
 	set_opt(sbi, FLUSH_MERGE);
 	set_opt(sbi, DISCARD);
 	if (f2fs_sb_has_blkzoned(sbi))
-		set_opt_mode(sbi, F2FS_MOUNT_LFS);
+		F2FS_OPTION(sbi).fs_mode = FS_MODE_LFS;
 	else
-		set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
+		F2FS_OPTION(sbi).fs_mode = FS_MODE_ADAPTIVE;
 
 #ifdef CONFIG_F2FS_FS_XATTR
 	set_opt(sbi, XATTR_USER);
@@ -1516,7 +1664,7 @@ static int f2fs_disable_checkpoint(struct f2fs_sb_info *sbi)
 	f2fs_update_time(sbi, DISABLE_TIME);
 
 	while (!f2fs_time_over(sbi, DISABLE_TIME)) {
-		mutex_lock(&sbi->gc_mutex);
+		down_write(&sbi->gc_lock);
 		err = f2fs_gc(sbi, true, false, NULL_SEGNO);
 		if (err == -ENODATA) {
 			err = 0;
@@ -1538,7 +1686,7 @@ static int f2fs_disable_checkpoint(struct f2fs_sb_info *sbi)
 		goto restore_flag;
 	}
 
-	mutex_lock(&sbi->gc_mutex);
+	down_write(&sbi->gc_lock);
 	cpc.reason = CP_PAUSE;
 	set_sbi_flag(sbi, SBI_CP_DISABLED);
 	err = f2fs_write_checkpoint(sbi, &cpc);
@@ -1550,20 +1698,20 @@ static int f2fs_disable_checkpoint(struct f2fs_sb_info *sbi)
 	spin_unlock(&sbi->stat_lock);
 
 out_unlock:
-	mutex_unlock(&sbi->gc_mutex);
+	up_write(&sbi->gc_lock);
 restore_flag:
-	sbi->sb->s_flags = s_flags;	/* Restore MS_RDONLY status */
+	sbi->sb->s_flags = s_flags;	/* Restore SB_RDONLY status */
 	return err;
 }
 
 static void f2fs_enable_checkpoint(struct f2fs_sb_info *sbi)
 {
-	mutex_lock(&sbi->gc_mutex);
+	down_write(&sbi->gc_lock);
 	f2fs_dirty_to_prefree(sbi);
 
 	clear_sbi_flag(sbi, SBI_CP_DISABLED);
 	set_sbi_flag(sbi, SBI_IS_DIRTY);
-	mutex_unlock(&sbi->gc_mutex);
+	up_write(&sbi->gc_lock);
 
 	f2fs_sync_fs(sbi->sb, 1);
 }
@@ -1621,7 +1769,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
 	default_options(sbi);
 
 	/* parse mount options */
-	err = parse_options(sb, data);
+	err = parse_options(sb, data, true);
 	if (err)
 		goto restore_opts;
 	checkpoint_changed =
@@ -1675,7 +1823,8 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
 	 * or if background_gc = off is passed in mount
 	 * option. Also sync the filesystem.
 	 */
-	if ((*flags & SB_RDONLY) || !test_opt(sbi, BG_GC)) {
+	if ((*flags & SB_RDONLY) ||
+			F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_OFF) {
 		if (sbi->gc_thread) {
 			f2fs_stop_gc_thread(sbi);
 			need_restart_gc = true;
@@ -1780,7 +1929,8 @@ repeat:
 		page = read_cache_page_gfp(mapping, blkidx, GFP_NOFS);
 		if (IS_ERR(page)) {
 			if (PTR_ERR(page) == -ENOMEM) {
-				congestion_wait(BLK_RW_ASYNC, HZ/50);
+				congestion_wait(BLK_RW_ASYNC,
+						DEFAULT_IO_TIMEOUT);
 				goto repeat;
 			}
 			set_sbi_flag(F2FS_SB(sb), SBI_QUOTA_NEED_REPAIR);
@@ -1822,6 +1972,7 @@ static ssize_t f2fs_quota_write(struct super_block *sb, int type,
 	int offset = off & (sb->s_blocksize - 1);
 	size_t towrite = len;
 	struct page *page;
+	void *fsdata = NULL;
 	char *kaddr;
 	int err = 0;
 	int tocopy;
@@ -1831,10 +1982,11 @@ static ssize_t f2fs_quota_write(struct super_block *sb, int type,
 								towrite);
 retry:
 		err = a_ops->write_begin(NULL, mapping, off, tocopy, 0,
-							&page, NULL);
+							&page, &fsdata);
 		if (unlikely(err)) {
 			if (err == -ENOMEM) {
-				congestion_wait(BLK_RW_ASYNC, HZ/50);
+				congestion_wait(BLK_RW_ASYNC,
+						DEFAULT_IO_TIMEOUT);
 				goto retry;
 			}
 			set_sbi_flag(F2FS_SB(sb), SBI_QUOTA_NEED_REPAIR);
@@ -1847,7 +1999,7 @@ retry:
 		flush_dcache_page(page);
 
 		a_ops->write_end(NULL, mapping, off, tocopy, tocopy,
-						page, NULL);
+						page, fsdata);
 		offset = 0;
 		towrite -= tocopy;
 		off += tocopy;
@@ -1932,7 +2084,7 @@ static int f2fs_quota_enable(struct super_block *sb, int type, int format_id,
 
 	/* Don't account quota for quota files to avoid recursion */
 	qf_inode->i_flags |= S_NOQUOTA;
-	err = dquot_enable(qf_inode, type, format_id, flags);
+	err = dquot_load_quota_inode(qf_inode, type, format_id, flags);
 	iput(qf_inode);
 	return err;
 }
@@ -2150,7 +2302,7 @@ static int f2fs_dquot_commit(struct dquot *dquot)
 	struct f2fs_sb_info *sbi = F2FS_SB(dquot->dq_sb);
 	int ret;
 
-	down_read(&sbi->quota_sem);
+	down_read_nested(&sbi->quota_sem, SINGLE_DEPTH_NESTING);
 	ret = dquot_commit(dquot);
 	if (ret < 0)
 		set_sbi_flag(sbi, SBI_QUOTA_NEED_REPAIR);
@@ -2174,13 +2326,10 @@ static int f2fs_dquot_acquire(struct dquot *dquot)
 static int f2fs_dquot_release(struct dquot *dquot)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(dquot->dq_sb);
-	int ret;
+	int ret = dquot_release(dquot);
 
-	down_read(&sbi->quota_sem);
-	ret = dquot_release(dquot);
 	if (ret < 0)
 		set_sbi_flag(sbi, SBI_QUOTA_NEED_REPAIR);
-	up_read(&sbi->quota_sem);
 	return ret;
 }
 
@@ -2188,29 +2337,22 @@ static int f2fs_dquot_mark_dquot_dirty(struct dquot *dquot)
 {
 	struct super_block *sb = dquot->dq_sb;
 	struct f2fs_sb_info *sbi = F2FS_SB(sb);
-	int ret;
-
-	down_read(&sbi->quota_sem);
-	ret = dquot_mark_dquot_dirty(dquot);
+	int ret = dquot_mark_dquot_dirty(dquot);
 
 	/* if we are using journalled quota */
 	if (is_journalled_quota(sbi))
 		set_sbi_flag(sbi, SBI_QUOTA_NEED_FLUSH);
 
-	up_read(&sbi->quota_sem);
 	return ret;
 }
 
 static int f2fs_dquot_commit_info(struct super_block *sb, int type)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(sb);
-	int ret;
+	int ret = dquot_commit_info(sb, type);
 
-	down_read(&sbi->quota_sem);
-	ret = dquot_commit_info(sb, type);
 	if (ret < 0)
 		set_sbi_flag(sbi, SBI_QUOTA_NEED_REPAIR);
-	up_read(&sbi->quota_sem);
 	return ret;
 }
 
@@ -2303,18 +2445,33 @@ static int f2fs_set_context(struct inode *inode, const void *ctx, size_t len,
 				ctx, len, fs_data, XATTR_CREATE);
 }
 
-static bool f2fs_dummy_context(struct inode *inode)
+static const union fscrypt_context *
+f2fs_get_dummy_context(struct super_block *sb)
 {
-	return DUMMY_ENCRYPTION_ENABLED(F2FS_I_SB(inode));
+	return F2FS_OPTION(F2FS_SB(sb)).dummy_enc_ctx.ctx;
+}
+
+static bool f2fs_has_stable_inodes(struct super_block *sb)
+{
+	return true;
+}
+
+static void f2fs_get_ino_and_lblk_bits(struct super_block *sb,
+				       int *ino_bits_ret, int *lblk_bits_ret)
+{
+	*ino_bits_ret = 8 * sizeof(nid_t);
+	*lblk_bits_ret = 8 * sizeof(block_t);
 }
 
 static const struct fscrypt_operations f2fs_cryptops = {
-	.key_prefix	= "f2fs:",
-	.get_context	= f2fs_get_context,
-	.set_context	= f2fs_set_context,
-	.dummy_context	= f2fs_dummy_context,
-	.empty_dir	= f2fs_empty_dir,
-	.max_namelen	= F2FS_NAME_LEN,
+	.key_prefix		= "f2fs:",
+	.get_context		= f2fs_get_context,
+	.set_context		= f2fs_set_context,
+	.get_dummy_context	= f2fs_get_dummy_context,
+	.empty_dir		= f2fs_empty_dir,
+	.max_namelen		= F2FS_NAME_LEN,
+	.has_stable_inodes	= f2fs_has_stable_inodes,
+	.get_ino_and_lblk_bits	= f2fs_get_ino_and_lblk_bits,
 };
 #endif
 
@@ -2604,6 +2761,21 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi,
 		return -EFSCORRUPTED;
 	}
 
+	if (RDEV(0).path[0]) {
+		block_t dev_seg_count = le32_to_cpu(RDEV(0).total_segments);
+		int i = 1;
+
+		while (i < MAX_DEVICES && RDEV(i).path[0]) {
+			dev_seg_count += le32_to_cpu(RDEV(i).total_segments);
+			i++;
+		}
+		if (segment_count != dev_seg_count) {
+			f2fs_info(sbi, "Segment count (%u) mismatch with total segments from devices (%u)",
+					segment_count, dev_seg_count);
+			return -EFSCORRUPTED;
+		}
+	}
+
 	if (secs_per_zone > total_sections || !secs_per_zone) {
 		f2fs_info(sbi, "Wrong secs_per_zone / total_sections (%u, %u)",
 			  secs_per_zone, total_sections);
@@ -2838,6 +3010,7 @@ static void init_sb_info(struct f2fs_sb_info *sbi)
 	spin_lock_init(&sbi->dev_lock);
 
 	init_rwsem(&sbi->sb_lock);
+	init_rwsem(&sbi->pin_sem);
 }
 
 static int init_percpu_info(struct f2fs_sb_info *sbi)
@@ -2857,15 +3030,21 @@ static int init_percpu_info(struct f2fs_sb_info *sbi)
 }
 
 #ifdef CONFIG_BLK_DEV_ZONED
+static int f2fs_report_zone_cb(struct blk_zone *zone, unsigned int idx,
+			       void *data)
+{
+	struct f2fs_dev_info *dev = data;
+
+	if (zone->type != BLK_ZONE_TYPE_CONVENTIONAL)
+		set_bit(idx, dev->blkz_seq);
+	return 0;
+}
+
 static int init_blkz_info(struct f2fs_sb_info *sbi, int devi)
 {
 	struct block_device *bdev = FDEV(devi).bdev;
 	sector_t nr_sectors = bdev->bd_part->nr_sects;
-	sector_t sector = 0;
-	struct blk_zone *zones;
-	unsigned int i, nr_zones;
-	unsigned int n = 0;
-	int err = -EIO;
+	int ret;
 
 	if (!f2fs_sb_has_blkzoned(sbi))
 		return 0;
@@ -2890,38 +3069,13 @@ static int init_blkz_info(struct f2fs_sb_info *sbi, int devi)
 	if (!FDEV(devi).blkz_seq)
 		return -ENOMEM;
 
-#define F2FS_REPORT_NR_ZONES   4096
-
-	zones = f2fs_kzalloc(sbi,
-			     array_size(F2FS_REPORT_NR_ZONES,
-					sizeof(struct blk_zone)),
-			     GFP_KERNEL);
-	if (!zones)
-		return -ENOMEM;
-
 	/* Get block zones type */
-	while (zones && sector < nr_sectors) {
-
-		nr_zones = F2FS_REPORT_NR_ZONES;
-		err = blkdev_report_zones(bdev, sector, zones, &nr_zones);
-		if (err)
-			break;
-		if (!nr_zones) {
-			err = -EIO;
-			break;
-		}
-
-		for (i = 0; i < nr_zones; i++) {
-			if (zones[i].type != BLK_ZONE_TYPE_CONVENTIONAL)
-				set_bit(n, FDEV(devi).blkz_seq);
-			sector += zones[i].len;
-			n++;
-		}
-	}
-
-	kvfree(zones);
+	ret = blkdev_report_zones(bdev, 0, BLK_ALL_ZONES, f2fs_report_zone_cb,
+				  &FDEV(devi));
+	if (ret < 0)
+		return ret;
 
-	return err;
+	return 0;
 }
 #endif
 
@@ -2951,6 +3105,7 @@ static int read_raw_super_block(struct f2fs_sb_info *sbi,
 			f2fs_err(sbi, "Unable to read %dth superblock",
 				 block + 1);
 			err = -EIO;
+			*recovery = 1;
 			continue;
 		}
 
@@ -2960,6 +3115,7 @@ static int read_raw_super_block(struct f2fs_sb_info *sbi,
 			f2fs_err(sbi, "Can't find valid F2FS filesystem in %dth superblock",
 				 block + 1);
 			brelse(bh);
+			*recovery = 1;
 			continue;
 		}
 
@@ -2972,10 +3128,6 @@ static int read_raw_super_block(struct f2fs_sb_info *sbi,
 		brelse(bh);
 	}
 
-	/* Fail to read any one of the superblocks*/
-	if (err < 0)
-		*recovery = 1;
-
 	/* No valid superblock */
 	if (!*raw_super)
 		kvfree(super);
@@ -3250,7 +3402,7 @@ try_onemore:
 		goto free_sb_buf;
 	}
 
-	err = parse_options(sb, options);
+	err = parse_options(sb, options, false);
 	if (err)
 		goto free_options;
 
@@ -3294,7 +3446,7 @@ try_onemore:
 
 	/* init f2fs-specific super block info */
 	sbi->valid_super_block = valid_super_block;
-	mutex_init(&sbi->gc_mutex);
+	init_rwsem(&sbi->gc_lock);
 	mutex_init(&sbi->writepages);
 	mutex_init(&sbi->cp_mutex);
 	mutex_init(&sbi->resize_mutex);
@@ -3329,6 +3481,8 @@ try_onemore:
 			sbi->write_io[i][j].bio = NULL;
 			spin_lock_init(&sbi->write_io[i][j].io_lock);
 			INIT_LIST_HEAD(&sbi->write_io[i][j].io_list);
+			INIT_LIST_HEAD(&sbi->write_io[i][j].bio_list);
+			init_rwsem(&sbi->write_io[i][j].bio_list_lock);
 		}
 	}
 
@@ -3350,12 +3504,17 @@ try_onemore:
 		}
 	}
 
+	/* init per sbi slab cache */
+	err = f2fs_init_xattr_caches(sbi);
+	if (err)
+		goto free_io_dummy;
+
 	/* get an inode for meta space */
 	sbi->meta_inode = f2fs_iget(sb, F2FS_META_INO(sbi));
 	if (IS_ERR(sbi->meta_inode)) {
 		f2fs_err(sbi, "Failed to read F2FS meta data inode");
 		err = PTR_ERR(sbi->meta_inode);
-		goto free_io_dummy;
+		goto free_xattr_cache;
 	}
 
 	err = f2fs_get_valid_checkpoint(sbi);
@@ -3381,6 +3540,12 @@ try_onemore:
 		goto free_devices;
 	}
 
+	err = f2fs_init_post_read_wq(sbi);
+	if (err) {
+		f2fs_err(sbi, "Failed to initialize post read workqueue");
+		goto free_devices;
+	}
+
 	sbi->total_valid_node_count =
 				le32_to_cpu(sbi->ckpt->valid_node_count);
 	percpu_counter_set(&sbi->total_valid_inode_count,
@@ -3477,7 +3642,7 @@ try_onemore:
 			f2fs_err(sbi, "Cannot turn on quotas: error %d", err);
 	}
 #endif
-	/* if there are nt orphan nodes free them */
+	/* if there are any orphan inodes, free them */
 	err = f2fs_recover_orphan_inodes(sbi);
 	if (err)
 		goto free_meta;
@@ -3486,7 +3651,8 @@ try_onemore:
 		goto reset_checkpoint;
 
 	/* recover fsynced data */
-	if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) {
+	if (!test_opt(sbi, DISABLE_ROLL_FORWARD) &&
+			!test_opt(sbi, NORECOVERY)) {
 		/*
 		 * mount should be failed, when device has readonly mode, and
 		 * previous checkpoint was not done by clean system shutdown.
@@ -3525,6 +3691,17 @@ try_onemore:
 			goto free_meta;
 		}
 	}
+
+	/*
+	 * If the f2fs is not readonly and fsync data recovery succeeds,
+	 * check zoned block devices' write pointer consistency.
+	 */
+	if (!err && !f2fs_readonly(sb) && f2fs_sb_has_blkzoned(sbi)) {
+		err = f2fs_check_write_pointer(sbi);
+		if (err)
+			goto free_meta;
+	}
+
 reset_checkpoint:
 	/* f2fs_recover_fsync_data() cleared this already */
 	clear_sbi_flag(sbi, SBI_POR_DOING);
@@ -3541,7 +3718,7 @@ reset_checkpoint:
 	 * If filesystem is not mounted as read-only then
 	 * do start the gc_thread.
 	 */
-	if (test_opt(sbi, BG_GC) && !f2fs_readonly(sb)) {
+	if (F2FS_OPTION(sbi).bggc_mode != BGGC_MODE_OFF && !f2fs_readonly(sb)) {
 		/* After POR, we can run background GC thread.*/
 		err = f2fs_start_gc_thread(sbi);
 		if (err)
@@ -3602,6 +3779,7 @@ free_nm:
 	f2fs_destroy_node_manager(sbi);
 free_sm:
 	f2fs_destroy_segment_manager(sbi);
+	f2fs_destroy_post_read_wq(sbi);
 free_devices:
 	destroy_device_list(sbi);
 	kvfree(sbi->ckpt);
@@ -3609,6 +3787,8 @@ free_meta_inode:
 	make_bad_inode(sbi->meta_inode);
 	iput(sbi->meta_inode);
 	sbi->meta_inode = NULL;
+free_xattr_cache:
+	f2fs_destroy_xattr_caches(sbi);
 free_io_dummy:
 	mempool_destroy(sbi->write_io_dummy);
 free_percpu:
@@ -3625,6 +3805,7 @@ free_options:
 	for (i = 0; i < MAXQUOTAS; i++)
 		kvfree(F2FS_OPTION(sbi).s_qf_names[i]);
 #endif
+	fscrypt_free_dummy_context(&F2FS_OPTION(sbi).dummy_enc_ctx);
 	kvfree(options);
 free_sb_buf:
 	kvfree(raw_super);
@@ -3740,8 +3921,17 @@ static int __init init_f2fs_fs(void)
 	err = f2fs_init_post_read_processing();
 	if (err)
 		goto free_root_stats;
+	err = f2fs_init_bio_entry_cache();
+	if (err)
+		goto free_post_read;
+	err = f2fs_init_bioset();
+	if (err)
+		goto free_bio_enrty_cache;
 	return 0;
-
+free_bio_enrty_cache:
+	f2fs_destroy_bio_entry_cache();
+free_post_read:
+	f2fs_destroy_post_read_processing();
 free_root_stats:
 	f2fs_destroy_root_stats();
 	unregister_filesystem(&f2fs_fs_type);
@@ -3765,6 +3955,8 @@ fail:
 
 static void __exit exit_f2fs_fs(void)
 {
+	f2fs_destroy_bioset();
+	f2fs_destroy_bio_entry_cache();
 	f2fs_destroy_post_read_processing();
 	f2fs_destroy_root_stats();
 	unregister_filesystem(&f2fs_fs_type);
diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c
index b558b64a4c9c..3162f46b3c9b 100644
--- a/fs/f2fs/sysfs.c
+++ b/fs/f2fs/sysfs.c
@@ -25,6 +25,9 @@ enum {
 	DCC_INFO,	/* struct discard_cmd_control */
 	NM_INFO,	/* struct f2fs_nm_info */
 	F2FS_SBI,	/* struct f2fs_sb_info */
+#ifdef CONFIG_F2FS_STAT_FS
+	STAT_INFO,      /* struct f2fs_stat_info */
+#endif
 #ifdef CONFIG_F2FS_FAULT_INJECTION
 	FAULT_INFO_RATE,	/* struct f2fs_fault_info */
 	FAULT_INFO_TYPE,	/* struct f2fs_fault_info */
@@ -42,6 +45,9 @@ struct f2fs_attr {
 	int id;
 };
 
+static ssize_t f2fs_sbi_show(struct f2fs_attr *a,
+			     struct f2fs_sb_info *sbi, char *buf);
+
 static unsigned char *__struct_ptr(struct f2fs_sb_info *sbi, int struct_type)
 {
 	if (struct_type == GC_THREAD)
@@ -59,41 +65,25 @@ static unsigned char *__struct_ptr(struct f2fs_sb_info *sbi, int struct_type)
 					struct_type == FAULT_INFO_TYPE)
 		return (unsigned char *)&F2FS_OPTION(sbi).fault_info;
 #endif
+#ifdef CONFIG_F2FS_STAT_FS
+	else if (struct_type == STAT_INFO)
+		return (unsigned char *)F2FS_STAT(sbi);
+#endif
 	return NULL;
 }
 
 static ssize_t dirty_segments_show(struct f2fs_attr *a,
 		struct f2fs_sb_info *sbi, char *buf)
 {
-	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		(unsigned long long)(dirty_segments(sbi)));
-}
-
-static ssize_t unusable_show(struct f2fs_attr *a,
-		struct f2fs_sb_info *sbi, char *buf)
-{
-	block_t unusable;
-
-	if (test_opt(sbi, DISABLE_CHECKPOINT))
-		unusable = sbi->unusable_block_count;
-	else
-		unusable = f2fs_get_unusable_blocks(sbi);
-	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		(unsigned long long)unusable);
+	return sprintf(buf, "%llu\n",
+			(unsigned long long)(dirty_segments(sbi)));
 }
 
-static ssize_t encoding_show(struct f2fs_attr *a,
+static ssize_t free_segments_show(struct f2fs_attr *a,
 		struct f2fs_sb_info *sbi, char *buf)
 {
-#ifdef CONFIG_UNICODE
-	if (f2fs_sb_has_casefold(sbi))
-		return snprintf(buf, PAGE_SIZE, "%s (%d.%d.%d)\n",
-			sbi->s_encoding->charset,
-			(sbi->s_encoding->version >> 16) & 0xff,
-			(sbi->s_encoding->version >> 8) & 0xff,
-			sbi->s_encoding->version & 0xff);
-#endif
-	return snprintf(buf, PAGE_SIZE, "(none)");
+	return sprintf(buf, "%llu\n",
+			(unsigned long long)(free_segments(sbi)));
 }
 
 static ssize_t lifetime_write_kbytes_show(struct f2fs_attr *a,
@@ -102,10 +92,10 @@ static ssize_t lifetime_write_kbytes_show(struct f2fs_attr *a,
 	struct super_block *sb = sbi->sb;
 
 	if (!sb->s_bdev->bd_part)
-		return snprintf(buf, PAGE_SIZE, "0\n");
+		return sprintf(buf, "0\n");
 
-	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		(unsigned long long)(sbi->kbytes_written +
+	return sprintf(buf, "%llu\n",
+			(unsigned long long)(sbi->kbytes_written +
 			BD_PART_WRITTEN(sbi)));
 }
 
@@ -116,54 +106,122 @@ static ssize_t features_show(struct f2fs_attr *a,
 	int len = 0;
 
 	if (!sb->s_bdev->bd_part)
-		return snprintf(buf, PAGE_SIZE, "0\n");
+		return sprintf(buf, "0\n");
 
 	if (f2fs_sb_has_encrypt(sbi))
-		len += snprintf(buf, PAGE_SIZE - len, "%s",
+		len += scnprintf(buf, PAGE_SIZE - len, "%s",
 						"encryption");
 	if (f2fs_sb_has_blkzoned(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "blkzoned");
 	if (f2fs_sb_has_extra_attr(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "extra_attr");
 	if (f2fs_sb_has_project_quota(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "projquota");
 	if (f2fs_sb_has_inode_chksum(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "inode_checksum");
 	if (f2fs_sb_has_flexible_inline_xattr(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "flexible_inline_xattr");
 	if (f2fs_sb_has_quota_ino(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "quota_ino");
 	if (f2fs_sb_has_inode_crtime(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "inode_crtime");
 	if (f2fs_sb_has_lost_found(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "lost_found");
 	if (f2fs_sb_has_verity(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "verity");
 	if (f2fs_sb_has_sb_chksum(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "sb_checksum");
 	if (f2fs_sb_has_casefold(sbi))
-		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "casefold");
-	len += snprintf(buf + len, PAGE_SIZE - len, "\n");
+	if (f2fs_sb_has_compression(sbi))
+		len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
+				len ? ", " : "", "compression");
+	len += scnprintf(buf + len, PAGE_SIZE - len, "%s%s",
+				len ? ", " : "", "pin_file");
+	len += scnprintf(buf + len, PAGE_SIZE - len, "\n");
 	return len;
 }
 
 static ssize_t current_reserved_blocks_show(struct f2fs_attr *a,
 					struct f2fs_sb_info *sbi, char *buf)
 {
-	return snprintf(buf, PAGE_SIZE, "%u\n", sbi->current_reserved_blocks);
+	return sprintf(buf, "%u\n", sbi->current_reserved_blocks);
+}
+
+static ssize_t unusable_show(struct f2fs_attr *a,
+		struct f2fs_sb_info *sbi, char *buf)
+{
+	block_t unusable;
+
+	if (test_opt(sbi, DISABLE_CHECKPOINT))
+		unusable = sbi->unusable_block_count;
+	else
+		unusable = f2fs_get_unusable_blocks(sbi);
+	return sprintf(buf, "%llu\n", (unsigned long long)unusable);
 }
 
+static ssize_t encoding_show(struct f2fs_attr *a,
+		struct f2fs_sb_info *sbi, char *buf)
+{
+#ifdef CONFIG_UNICODE
+	if (f2fs_sb_has_casefold(sbi))
+		return snprintf(buf, PAGE_SIZE, "%s (%d.%d.%d)\n",
+			sbi->s_encoding->charset,
+			(sbi->s_encoding->version >> 16) & 0xff,
+			(sbi->s_encoding->version >> 8) & 0xff,
+			sbi->s_encoding->version & 0xff);
+#endif
+	return sprintf(buf, "(none)");
+}
+
+static ssize_t mounted_time_sec_show(struct f2fs_attr *a,
+		struct f2fs_sb_info *sbi, char *buf)
+{
+	return sprintf(buf, "%llu", SIT_I(sbi)->mounted_time);
+}
+
+#ifdef CONFIG_F2FS_STAT_FS
+static ssize_t moved_blocks_foreground_show(struct f2fs_attr *a,
+				struct f2fs_sb_info *sbi, char *buf)
+{
+	struct f2fs_stat_info *si = F2FS_STAT(sbi);
+
+	return sprintf(buf, "%llu\n",
+		(unsigned long long)(si->tot_blks -
+			(si->bg_data_blks + si->bg_node_blks)));
+}
+
+static ssize_t moved_blocks_background_show(struct f2fs_attr *a,
+				struct f2fs_sb_info *sbi, char *buf)
+{
+	struct f2fs_stat_info *si = F2FS_STAT(sbi);
+
+	return sprintf(buf, "%llu\n",
+		(unsigned long long)(si->bg_data_blks + si->bg_node_blks));
+}
+
+static ssize_t avg_vblocks_show(struct f2fs_attr *a,
+		struct f2fs_sb_info *sbi, char *buf)
+{
+	struct f2fs_stat_info *si = F2FS_STAT(sbi);
+
+	si->dirty_count = dirty_segments(sbi);
+	f2fs_update_sit_info(sbi);
+	return sprintf(buf, "%llu\n", (unsigned long long)(si->avg_vblocks));
+}
+#endif
+
 static ssize_t f2fs_sbi_show(struct f2fs_attr *a,
 			struct f2fs_sb_info *sbi, char *buf)
 {
@@ -181,23 +239,23 @@ static ssize_t f2fs_sbi_show(struct f2fs_attr *a,
 		int hot_count = sbi->raw_super->hot_ext_count;
 		int len = 0, i;
 
-		len += snprintf(buf + len, PAGE_SIZE - len,
+		len += scnprintf(buf + len, PAGE_SIZE - len,
 						"cold file extension:\n");
 		for (i = 0; i < cold_count; i++)
-			len += snprintf(buf + len, PAGE_SIZE - len, "%s\n",
+			len += scnprintf(buf + len, PAGE_SIZE - len, "%s\n",
 								extlist[i]);
 
-		len += snprintf(buf + len, PAGE_SIZE - len,
+		len += scnprintf(buf + len, PAGE_SIZE - len,
 						"hot file extension:\n");
 		for (i = cold_count; i < cold_count + hot_count; i++)
-			len += snprintf(buf + len, PAGE_SIZE - len, "%s\n",
+			len += scnprintf(buf + len, PAGE_SIZE - len, "%s\n",
 								extlist[i]);
 		return len;
 	}
 
 	ui = (unsigned int *)(ptr + a->offset);
 
-	return snprintf(buf, PAGE_SIZE, "%u\n", *ui);
+	return sprintf(buf, "%u\n", *ui);
 }
 
 static ssize_t __sbi_store(struct f2fs_attr *a,
@@ -387,6 +445,8 @@ enum feat_id {
 	FEAT_VERITY,
 	FEAT_SB_CHECKSUM,
 	FEAT_CASEFOLD,
+	FEAT_COMPRESSION,
+	FEAT_TEST_DUMMY_ENCRYPTION_V2,
 };
 
 static ssize_t f2fs_feature_show(struct f2fs_attr *a,
@@ -406,7 +466,9 @@ static ssize_t f2fs_feature_show(struct f2fs_attr *a,
 	case FEAT_VERITY:
 	case FEAT_SB_CHECKSUM:
 	case FEAT_CASEFOLD:
-		return snprintf(buf, PAGE_SIZE, "supported\n");
+	case FEAT_COMPRESSION:
+	case FEAT_TEST_DUMMY_ENCRYPTION_V2:
+		return sprintf(buf, "supported\n");
 	}
 	return 0;
 }
@@ -435,6 +497,14 @@ static struct f2fs_attr f2fs_attr_##_name = {			\
 	.id	= _id,						\
 }
 
+#define F2FS_STAT_ATTR(_struct_type, _struct_name, _name, _elname)	\
+static struct f2fs_attr f2fs_attr_##_name = {			\
+	.attr = {.name = __stringify(_name), .mode = 0444 },	\
+	.show = f2fs_sbi_show,					\
+	.struct_type = _struct_type,				\
+	.offset = offsetof(struct _struct_name, _elname),       \
+}
+
 F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_urgent_sleep_time,
 							urgent_sleep_time);
 F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_min_sleep_time, min_sleep_time);
@@ -443,6 +513,7 @@ F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_no_gc_sleep_time, no_gc_sleep_time);
 F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, gc_idle, gc_mode);
 F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, gc_urgent, gc_mode);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, reclaim_segments, rec_prefree_segments);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, main_blkaddr, main_blkaddr);
 F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_small_discards, max_discards);
 F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, discard_granularity, discard_granularity);
 F2FS_RW_ATTR(RESERVED_BLOCKS, f2fs_sb_info, reserved_blocks, reserved_blocks);
@@ -475,14 +546,26 @@ F2FS_RW_ATTR(FAULT_INFO_RATE, f2fs_fault_info, inject_rate, inject_rate);
 F2FS_RW_ATTR(FAULT_INFO_TYPE, f2fs_fault_info, inject_type, inject_type);
 #endif
 F2FS_GENERAL_RO_ATTR(dirty_segments);
+F2FS_GENERAL_RO_ATTR(free_segments);
 F2FS_GENERAL_RO_ATTR(lifetime_write_kbytes);
 F2FS_GENERAL_RO_ATTR(features);
 F2FS_GENERAL_RO_ATTR(current_reserved_blocks);
 F2FS_GENERAL_RO_ATTR(unusable);
 F2FS_GENERAL_RO_ATTR(encoding);
+F2FS_GENERAL_RO_ATTR(mounted_time_sec);
+#ifdef CONFIG_F2FS_STAT_FS
+F2FS_STAT_ATTR(STAT_INFO, f2fs_stat_info, cp_foreground_calls, cp_count);
+F2FS_STAT_ATTR(STAT_INFO, f2fs_stat_info, cp_background_calls, bg_cp_count);
+F2FS_STAT_ATTR(STAT_INFO, f2fs_stat_info, gc_foreground_calls, call_count);
+F2FS_STAT_ATTR(STAT_INFO, f2fs_stat_info, gc_background_calls, bg_gc);
+F2FS_GENERAL_RO_ATTR(moved_blocks_background);
+F2FS_GENERAL_RO_ATTR(moved_blocks_foreground);
+F2FS_GENERAL_RO_ATTR(avg_vblocks);
+#endif
 
 #ifdef CONFIG_FS_ENCRYPTION
 F2FS_FEATURE_RO_ATTR(encryption, FEAT_CRYPTO);
+F2FS_FEATURE_RO_ATTR(test_dummy_encryption_v2, FEAT_TEST_DUMMY_ENCRYPTION_V2);
 #endif
 #ifdef CONFIG_BLK_DEV_ZONED
 F2FS_FEATURE_RO_ATTR(block_zoned, FEAT_BLKZONED);
@@ -500,6 +583,9 @@ F2FS_FEATURE_RO_ATTR(verity, FEAT_VERITY);
 #endif
 F2FS_FEATURE_RO_ATTR(sb_checksum, FEAT_SB_CHECKSUM);
 F2FS_FEATURE_RO_ATTR(casefold, FEAT_CASEFOLD);
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+F2FS_FEATURE_RO_ATTR(compression, FEAT_COMPRESSION);
+#endif
 
 #define ATTR_LIST(name) (&f2fs_attr_##name.attr)
 static struct attribute *f2fs_attrs[] = {
@@ -510,6 +596,7 @@ static struct attribute *f2fs_attrs[] = {
 	ATTR_LIST(gc_idle),
 	ATTR_LIST(gc_urgent),
 	ATTR_LIST(reclaim_segments),
+	ATTR_LIST(main_blkaddr),
 	ATTR_LIST(max_small_discards),
 	ATTR_LIST(discard_granularity),
 	ATTR_LIST(batched_trim_sections),
@@ -539,12 +626,23 @@ static struct attribute *f2fs_attrs[] = {
 	ATTR_LIST(inject_type),
 #endif
 	ATTR_LIST(dirty_segments),
+	ATTR_LIST(free_segments),
 	ATTR_LIST(unusable),
 	ATTR_LIST(lifetime_write_kbytes),
 	ATTR_LIST(features),
 	ATTR_LIST(reserved_blocks),
 	ATTR_LIST(current_reserved_blocks),
 	ATTR_LIST(encoding),
+	ATTR_LIST(mounted_time_sec),
+#ifdef CONFIG_F2FS_STAT_FS
+	ATTR_LIST(cp_foreground_calls),
+	ATTR_LIST(cp_background_calls),
+	ATTR_LIST(gc_foreground_calls),
+	ATTR_LIST(gc_background_calls),
+	ATTR_LIST(moved_blocks_foreground),
+	ATTR_LIST(moved_blocks_background),
+	ATTR_LIST(avg_vblocks),
+#endif
 	NULL,
 };
 ATTRIBUTE_GROUPS(f2fs);
@@ -552,6 +650,7 @@ ATTRIBUTE_GROUPS(f2fs);
 static struct attribute *f2fs_feat_attrs[] = {
 #ifdef CONFIG_FS_ENCRYPTION
 	ATTR_LIST(encryption),
+	ATTR_LIST(test_dummy_encryption_v2),
 #endif
 #ifdef CONFIG_BLK_DEV_ZONED
 	ATTR_LIST(block_zoned),
@@ -569,6 +668,9 @@ static struct attribute *f2fs_feat_attrs[] = {
 #endif
 	ATTR_LIST(sb_checksum),
 	ATTR_LIST(casefold),
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+	ATTR_LIST(compression),
+#endif
 	NULL,
 };
 ATTRIBUTE_GROUPS(f2fs_feat);
@@ -729,10 +831,12 @@ int __init f2fs_init_sysfs(void)
 
 	ret = kobject_init_and_add(&f2fs_feat, &f2fs_feat_ktype,
 				   NULL, "features");
-	if (ret)
+	if (ret) {
+		kobject_put(&f2fs_feat);
 		kset_unregister(&f2fs_kset);
-	else
+	} else {
 		f2fs_proc_root = proc_mkdir("fs/f2fs", NULL);
+	}
 	return ret;
 }
 
@@ -753,8 +857,11 @@ int f2fs_register_sysfs(struct f2fs_sb_info *sbi)
 	init_completion(&sbi->s_kobj_unregister);
 	err = kobject_init_and_add(&sbi->s_kobj, &f2fs_sb_ktype, NULL,
 				"%s", sb->s_id);
-	if (err)
+	if (err) {
+		kobject_put(&sbi->s_kobj);
+		wait_for_completion(&sbi->s_kobj_unregister);
 		return err;
+	}
 
 	if (f2fs_proc_root)
 		sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root);
@@ -782,4 +889,5 @@ void f2fs_unregister_sysfs(struct f2fs_sb_info *sbi)
 		remove_proc_entry(sbi->sb->s_id, f2fs_proc_root);
 	}
 	kobject_del(&sbi->s_kobj);
+	kobject_put(&sbi->s_kobj);
 }
diff --git a/fs/f2fs/verity.c b/fs/f2fs/verity.c
index a401ef72bc82..865c9fb774fb 100644
--- a/fs/f2fs/verity.c
+++ b/fs/f2fs/verity.c
@@ -223,11 +223,23 @@ static int f2fs_get_verity_descriptor(struct inode *inode, void *buf,
 }
 
 static struct page *f2fs_read_merkle_tree_page(struct inode *inode,
-					       pgoff_t index)
+					       pgoff_t index,
+					       unsigned long num_ra_pages)
 {
+	struct page *page;
+
 	index += f2fs_verity_metadata_pos(inode) >> PAGE_SHIFT;
 
-	return read_mapping_page(inode->i_mapping, index, NULL);
+	page = find_get_page_flags(inode->i_mapping, index, FGP_ACCESSED);
+	if (!page || !PageUptodate(page)) {
+		if (page)
+			put_page(page);
+		else if (num_ra_pages > 1)
+			page_cache_readahead_unbounded(inode->i_mapping, NULL,
+					index, num_ra_pages, 0);
+		page = read_mapping_page(inode->i_mapping, index, NULL);
+	}
+	return page;
 }
 
 static int f2fs_write_merkle_tree_block(struct inode *inode, const void *buf,
diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index 181900af2576..4f6582ef7ee3 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -23,6 +23,25 @@
 #include "xattr.h"
 #include "segment.h"
 
+static void *xattr_alloc(struct f2fs_sb_info *sbi, int size, bool *is_inline)
+{
+	if (likely(size == sbi->inline_xattr_slab_size)) {
+		*is_inline = true;
+		return kmem_cache_zalloc(sbi->inline_xattr_slab, GFP_NOFS);
+	}
+	*is_inline = false;
+	return f2fs_kzalloc(sbi, size, GFP_NOFS);
+}
+
+static void xattr_free(struct f2fs_sb_info *sbi, void *xattr_addr,
+							bool is_inline)
+{
+	if (is_inline)
+		kmem_cache_free(sbi->inline_xattr_slab, xattr_addr);
+	else
+		kvfree(xattr_addr);
+}
+
 static int f2fs_xattr_generic_get(const struct xattr_handler *handler,
 		struct dentry *unused, struct inode *inode,
 		const char *name, void *buffer, size_t size)
@@ -301,7 +320,8 @@ static int read_xattr_block(struct inode *inode, void *txattr_addr)
 static int lookup_all_xattrs(struct inode *inode, struct page *ipage,
 				unsigned int index, unsigned int len,
 				const char *name, struct f2fs_xattr_entry **xe,
-				void **base_addr, int *base_size)
+				void **base_addr, int *base_size,
+				bool *is_inline)
 {
 	void *cur_addr, *txattr_addr, *last_txattr_addr;
 	void *last_addr = NULL;
@@ -312,12 +332,12 @@ static int lookup_all_xattrs(struct inode *inode, struct page *ipage,
 	if (!xnid && !inline_size)
 		return -ENODATA;
 
-	*base_size = XATTR_SIZE(xnid, inode) + XATTR_PADDING_SIZE;
-	txattr_addr = f2fs_kzalloc(F2FS_I_SB(inode), *base_size, GFP_NOFS);
+	*base_size = XATTR_SIZE(inode) + XATTR_PADDING_SIZE;
+	txattr_addr = xattr_alloc(F2FS_I_SB(inode), *base_size, is_inline);
 	if (!txattr_addr)
 		return -ENOMEM;
 
-	last_txattr_addr = (void *)txattr_addr + XATTR_SIZE(xnid, inode);
+	last_txattr_addr = (void *)txattr_addr + XATTR_SIZE(inode);
 
 	/* read from inline xattr */
 	if (inline_size) {
@@ -362,7 +382,7 @@ check:
 	*base_addr = txattr_addr;
 	return 0;
 out:
-	kvfree(txattr_addr);
+	xattr_free(F2FS_I_SB(inode), txattr_addr, *is_inline);
 	return err;
 }
 
@@ -499,6 +519,7 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name,
 	unsigned int size, len;
 	void *base_addr = NULL;
 	int base_size;
+	bool is_inline;
 
 	if (name == NULL)
 		return -EINVAL;
@@ -509,7 +530,7 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name,
 
 	down_read(&F2FS_I(inode)->i_xattr_sem);
 	error = lookup_all_xattrs(inode, ipage, index, len, name,
-				&entry, &base_addr, &base_size);
+				&entry, &base_addr, &base_size, &is_inline);
 	up_read(&F2FS_I(inode)->i_xattr_sem);
 	if (error)
 		return error;
@@ -532,7 +553,7 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name,
 	}
 	error = size;
 out:
-	kvfree(base_addr);
+	xattr_free(F2FS_I_SB(inode), base_addr, is_inline);
 	return error;
 }
 
@@ -540,7 +561,7 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
 	struct inode *inode = d_inode(dentry);
 	struct f2fs_xattr_entry *entry;
-	void *base_addr;
+	void *base_addr, *last_base_addr;
 	int error = 0;
 	size_t rest = buffer_size;
 
@@ -550,6 +571,8 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
 	if (error)
 		return error;
 
+	last_base_addr = (void *)base_addr + XATTR_SIZE(inode);
+
 	list_for_each_xattr(entry, base_addr) {
 		const struct xattr_handler *handler =
 			f2fs_xattr_handler(entry->e_name_index);
@@ -557,6 +580,15 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
 		size_t prefix_len;
 		size_t size;
 
+		if ((void *)(entry) + sizeof(__u32) > last_base_addr ||
+			(void *)XATTR_NEXT_ENTRY(entry) > last_base_addr) {
+			f2fs_err(F2FS_I_SB(inode), "inode (%lu) has corrupted xattr",
+						inode->i_ino);
+			set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_FSCK);
+			error = -EFSCORRUPTED;
+			goto cleanup;
+		}
+
 		if (!handler || (handler->list && !handler->list(dentry)))
 			continue;
 
@@ -597,7 +629,6 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 {
 	struct f2fs_xattr_entry *here, *last;
 	void *base_addr, *last_base_addr;
-	nid_t xnid = F2FS_I(inode)->i_xattr_nid;
 	int found, newsize;
 	size_t len;
 	__u32 new_hsize;
@@ -621,7 +652,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 	if (error)
 		return error;
 
-	last_base_addr = (void *)base_addr + XATTR_SIZE(xnid, inode);
+	last_base_addr = (void *)base_addr + XATTR_SIZE(inode);
 
 	/* find entry with wanted name. */
 	here = __find_xattr(base_addr, last_base_addr, index, len, name);
@@ -746,14 +777,34 @@ int f2fs_setxattr(struct inode *inode, int index, const char *name,
 	f2fs_balance_fs(sbi, true);
 
 	f2fs_lock_op(sbi);
-	/* protect xattr_ver */
-	down_write(&F2FS_I(inode)->i_sem);
 	down_write(&F2FS_I(inode)->i_xattr_sem);
 	err = __f2fs_setxattr(inode, index, name, value, size, ipage, flags);
 	up_write(&F2FS_I(inode)->i_xattr_sem);
-	up_write(&F2FS_I(inode)->i_sem);
 	f2fs_unlock_op(sbi);
 
 	f2fs_update_time(sbi, REQ_TIME);
 	return err;
 }
+
+int f2fs_init_xattr_caches(struct f2fs_sb_info *sbi)
+{
+	dev_t dev = sbi->sb->s_bdev->bd_dev;
+	char slab_name[32];
+
+	sprintf(slab_name, "f2fs_xattr_entry-%u:%u", MAJOR(dev), MINOR(dev));
+
+	sbi->inline_xattr_slab_size = F2FS_OPTION(sbi).inline_xattr_size *
+					sizeof(__le32) + XATTR_PADDING_SIZE;
+
+	sbi->inline_xattr_slab = f2fs_kmem_cache_create(slab_name,
+					sbi->inline_xattr_slab_size);
+	if (!sbi->inline_xattr_slab)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void f2fs_destroy_xattr_caches(struct f2fs_sb_info *sbi)
+{
+	kmem_cache_destroy(sbi->inline_xattr_slab);
+}
diff --git a/fs/f2fs/xattr.h b/fs/f2fs/xattr.h
index de0c600b9cab..938fcd20565d 100644
--- a/fs/f2fs/xattr.h
+++ b/fs/f2fs/xattr.h
@@ -49,7 +49,7 @@ struct f2fs_xattr_entry {
 	__u8    e_name_index;
 	__u8    e_name_len;
 	__le16  e_value_size;   /* size of attribute value */
-	char    e_name[0];      /* attribute name */
+	char    e_name[];      /* attribute name */
 };
 
 #define XATTR_HDR(ptr)		((struct f2fs_xattr_header *)(ptr))
@@ -73,7 +73,8 @@ struct f2fs_xattr_entry {
 				entry = XATTR_NEXT_ENTRY(entry))
 #define VALID_XATTR_BLOCK_SIZE	(PAGE_SIZE - sizeof(struct node_footer))
 #define XATTR_PADDING_SIZE	(sizeof(__u32))
-#define XATTR_SIZE(x,i)		(((x) ? VALID_XATTR_BLOCK_SIZE : 0) +	\
+#define XATTR_SIZE(i)		((F2FS_I(i)->i_xattr_nid ?		\
+					VALID_XATTR_BLOCK_SIZE : 0) +	\
 						(inline_xattr_size(i)))
 #define MIN_OFFSET(i)		XATTR_ALIGN(inline_xattr_size(i) +	\
 						VALID_XATTR_BLOCK_SIZE)
@@ -130,6 +131,8 @@ extern int f2fs_setxattr(struct inode *, int, const char *,
 extern int f2fs_getxattr(struct inode *, int, const char *, void *,
 						size_t, struct page *);
 extern ssize_t f2fs_listxattr(struct dentry *, char *, size_t);
+extern int f2fs_init_xattr_caches(struct f2fs_sb_info *);
+extern void f2fs_destroy_xattr_caches(struct f2fs_sb_info *);
 #else
 
 #define f2fs_xattr_handlers	NULL
@@ -150,6 +153,8 @@ static inline ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer,
 {
 	return -EOPNOTSUPP;
 }
+static inline int f2fs_init_xattr_caches(struct f2fs_sb_info *sbi) { return 0; }
+static inline void f2fs_destroy_xattr_caches(struct f2fs_sb_info *sbi) { }
 #endif
 
 #ifdef CONFIG_F2FS_FS_SECURITY
diff --git a/fs/fat/Kconfig b/fs/fat/Kconfig
index 718163d0c621..ca31993dcb47 100644
--- a/fs/fat/Kconfig
+++ b/fs/fat/Kconfig
@@ -69,7 +69,7 @@ config VFAT_FS
 
 	  The VFAT support enlarges your kernel by about 10 KB and it only
 	  works if you said Y to the "DOS FAT fs support" above.  Please read
-	  the file <file:Documentation/filesystems/vfat.txt> for details.  If
+	  the file <file:Documentation/filesystems/vfat.rst> for details.  If
 	  unsure, say Y.
 
 	  To compile this as a module, choose M here: the module will be called
@@ -82,7 +82,7 @@ config FAT_DEFAULT_CODEPAGE
 	help
 	  This option should be set to the codepage of your FAT filesystems.
 	  It can be overridden with the "codepage" mount option.
-	  See <file:Documentation/filesystems/vfat.txt> for more information.
+	  See <file:Documentation/filesystems/vfat.rst> for more information.
 
 config FAT_DEFAULT_IOCHARSET
 	string "Default iocharset for FAT"
@@ -96,7 +96,7 @@ config FAT_DEFAULT_IOCHARSET
 	  Note that "utf8" is not recommended for FAT filesystems.
 	  If unsure, you shouldn't set "utf8" here - select the next option
 	  instead if you would like to use UTF-8 encoded file names by default.
-	  See <file:Documentation/filesystems/vfat.txt> for more information.
+	  See <file:Documentation/filesystems/vfat.rst> for more information.
 
 	  Enable any character sets you need in File Systems/Native Language
 	  Support.
@@ -114,4 +114,4 @@ config FAT_DEFAULT_UTF8
 
 	  Say Y if you use UTF-8 encoding for file names, N otherwise.
 
-	  See <file:Documentation/filesystems/vfat.txt> for more information.
+	  See <file:Documentation/filesystems/vfat.rst> for more information.
diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index 054acd9fd033..b4ddf48fa444 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -804,8 +804,6 @@ static long fat_dir_ioctl(struct file *filp, unsigned int cmd,
 		return fat_generic_ioctl(filp, cmd, arg);
 	}
 
-	if (!access_ok(d1, sizeof(struct __fat_dirent[2])))
-		return -EFAULT;
 	/*
 	 * Yes, we don't need this put_user() absolutely. However old
 	 * code didn't return the right value. So, app use this value,
@@ -844,8 +842,6 @@ static long fat_compat_dir_ioctl(struct file *filp, unsigned cmd,
 		return fat_generic_ioctl(filp, cmd, (unsigned long)arg);
 	}
 
-	if (!access_ok(d1, sizeof(struct compat_dirent[2])))
-		return -EFAULT;
 	/*
 	 * Yes, we don't need this put_user() absolutely. However old
 	 * code didn't return the right value. So, app use this value,
diff --git a/fs/fat/file.c b/fs/fat/file.c
index 4614c0ba5f1c..42134c58c87e 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -172,15 +172,6 @@ long fat_generic_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	}
 }
 
-#ifdef CONFIG_COMPAT
-static long fat_generic_compat_ioctl(struct file *filp, unsigned int cmd,
-				      unsigned long arg)
-
-{
-	return fat_generic_ioctl(filp, cmd, (unsigned long)compat_ptr(arg));
-}
-#endif
-
 static int fat_file_release(struct inode *inode, struct file *filp)
 {
 	if ((filp->f_mode & FMODE_WRITE) &&
@@ -204,7 +195,7 @@ int fat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 	if (err)
 		return err;
 
-	return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+	return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
 }
 
 
@@ -215,9 +206,7 @@ const struct file_operations fat_file_operations = {
 	.mmap		= generic_file_mmap,
 	.release	= fat_file_release,
 	.unlocked_ioctl	= fat_generic_ioctl,
-#ifdef CONFIG_COMPAT
-	.compat_ioctl	= fat_generic_compat_ioctl,
-#endif
+	.compat_ioctl	= compat_ptr_ioctl,
 	.fsync		= fat_file_fsync,
 	.splice_read	= generic_file_splice_read,
 	.splice_write	= iter_file_splice_write,
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 5f04c5c810fb..e6e68b2274a5 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -21,6 +21,7 @@
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
 #include <asm/unaligned.h>
+#include <linux/random.h>
 #include <linux/iversion.h>
 #include "fat.h"
 
@@ -209,10 +210,9 @@ static int fat_readpage(struct file *file, struct page *page)
 	return mpage_readpage(page, fat_get_block);
 }
 
-static int fat_readpages(struct file *file, struct address_space *mapping,
-			 struct list_head *pages, unsigned nr_pages)
+static void fat_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, fat_get_block);
+	mpage_readahead(rac, fat_get_block);
 }
 
 static void fat_write_failed(struct address_space *mapping, loff_t to)
@@ -343,7 +343,7 @@ int fat_block_truncate_page(struct inode *inode, loff_t from)
 
 static const struct address_space_operations fat_aops = {
 	.readpage	= fat_readpage,
-	.readpages	= fat_readpages,
+	.readahead	= fat_readahead,
 	.writepage	= fat_writepage,
 	.writepages	= fat_writepages,
 	.write_begin	= fat_write_begin,
@@ -521,7 +521,7 @@ int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 	inode->i_uid = sbi->options.fs_uid;
 	inode->i_gid = sbi->options.fs_gid;
 	inode_inc_iversion(inode);
-	inode->i_generation = get_seconds();
+	inode->i_generation = prandom_u32();
 
 	if ((de->attr & ATTR_DIR) && !IS_FREE(de->name)) {
 		inode->i_generation &= ~1;
@@ -749,6 +749,13 @@ static struct inode *fat_alloc_inode(struct super_block *sb)
 		return NULL;
 
 	init_rwsem(&ei->truncate_lock);
+	/* Zeroing to allow iput() even if partial initialized inode. */
+	ei->mmu_private = 0;
+	ei->i_start = 0;
+	ei->i_logstart = 0;
+	ei->i_attrs = 0;
+	ei->i_pos = 0;
+
 	return &ei->vfs_inode;
 }
 
@@ -1373,16 +1380,6 @@ out:
 	return 0;
 }
 
-static void fat_dummy_inode_init(struct inode *inode)
-{
-	/* Initialize this dummy inode to work as no-op. */
-	MSDOS_I(inode)->mmu_private = 0;
-	MSDOS_I(inode)->i_start = 0;
-	MSDOS_I(inode)->i_logstart = 0;
-	MSDOS_I(inode)->i_attrs = 0;
-	MSDOS_I(inode)->i_pos = 0;
-}
-
 static int fat_read_root(struct inode *inode)
 {
 	struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb);
@@ -1843,13 +1840,11 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat,
 	fat_inode = new_inode(sb);
 	if (!fat_inode)
 		goto out_fail;
-	fat_dummy_inode_init(fat_inode);
 	sbi->fat_inode = fat_inode;
 
 	fsinfo_inode = new_inode(sb);
 	if (!fsinfo_inode)
 		goto out_fail;
-	fat_dummy_inode_init(fsinfo_inode);
 	fsinfo_inode->i_ino = MSDOS_FSINFO_INO;
 	sbi->fsinfo_inode = fsinfo_inode;
 	insert_inode_hash(fsinfo_inode);
diff --git a/fs/fat/misc.c b/fs/fat/misc.c
index 1e08bd54c5fb..f1b2a1fc2a6a 100644
--- a/fs/fat/misc.c
+++ b/fs/fat/misc.c
@@ -271,6 +271,14 @@ static inline struct timespec64 fat_timespec64_trunc_2secs(struct timespec64 ts)
 {
 	return (struct timespec64){ ts.tv_sec & ~1ULL, 0 };
 }
+
+static inline struct timespec64 fat_timespec64_trunc_10ms(struct timespec64 ts)
+{
+	if (ts.tv_nsec)
+		ts.tv_nsec -= ts.tv_nsec % 10000000UL;
+	return ts;
+}
+
 /*
  * truncate the various times with appropriate granularity:
  *   root inode:
@@ -308,7 +316,7 @@ int fat_truncate_time(struct inode *inode, struct timespec64 *now, int flags)
 	}
 	if (flags & S_CTIME) {
 		if (sbi->options.isvfat)
-			inode->i_ctime = timespec64_trunc(*now, 10000000);
+			inode->i_ctime = fat_timespec64_trunc_10ms(*now);
 		else
 			inode->i_ctime = fat_timespec64_trunc_2secs(*now);
 	}
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 3d40771e8e7c..2e4c0fa2074b 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -261,7 +261,7 @@ static int f_getowner_uids(struct file *filp, unsigned long arg)
 static bool rw_hint_valid(enum rw_hint hint)
 {
 	switch (hint) {
-	case RWF_WRITE_LIFE_NOT_SET:
+	case RWH_WRITE_LIFE_NOT_SET:
 	case RWH_WRITE_LIFE_NONE:
 	case RWH_WRITE_LIFE_SHORT:
 	case RWH_WRITE_LIFE_MEDIUM:
@@ -277,7 +277,7 @@ static long fcntl_rw_hint(struct file *file, unsigned int cmd,
 			  unsigned long arg)
 {
 	struct inode *inode = file_inode(file);
-	u64 *argp = (u64 __user *)arg;
+	u64 __user *argp = (u64 __user *)arg;
 	enum rw_hint hint;
 	u64 h;
 
@@ -735,8 +735,9 @@ static void send_sigio_to_task(struct task_struct *p,
 		return;
 
 	switch (signum) {
-		kernel_siginfo_t si;
-		default:
+		default: {
+			kernel_siginfo_t si;
+
 			/* Queue a rt signal with the appropriate fd as its
 			   value.  We use SI_SIGIO as the source, not 
 			   SI_KERNEL, since kernel signals always get 
@@ -769,6 +770,7 @@ static void send_sigio_to_task(struct task_struct *p,
 			si.si_fd    = fd;
 			if (!do_send_sig_info(signum, &si, p, type))
 				break;
+		}
 		/* fall-through - fall back on the old plain SIGIO signal */
 		case 0:
 			do_send_sig_info(SIGIO, SEND_SIG_PRIV, p, type);
diff --git a/fs/file.c b/fs/file.c
index 3da91a112bab..abb8b7081d7a 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -70,7 +70,7 @@ static void copy_fd_bitmaps(struct fdtable *nfdt, struct fdtable *ofdt,
  */
 static void copy_fdtable(struct fdtable *nfdt, struct fdtable *ofdt)
 {
-	unsigned int cpy, set;
+	size_t cpy, set;
 
 	BUG_ON(nfdt->max_fds < ofdt->max_fds);
 
@@ -540,9 +540,14 @@ static int alloc_fd(unsigned start, unsigned flags)
 	return __alloc_fd(current->files, start, rlimit(RLIMIT_NOFILE), flags);
 }
 
+int __get_unused_fd_flags(unsigned flags, unsigned long nofile)
+{
+	return __alloc_fd(current->files, 0, nofile, flags);
+}
+
 int get_unused_fd_flags(unsigned flags)
 {
-	return __alloc_fd(current->files, 0, rlimit(RLIMIT_NOFILE), flags);
+	return __get_unused_fd_flags(flags, rlimit(RLIMIT_NOFILE));
 }
 EXPORT_SYMBOL(get_unused_fd_flags);
 
@@ -642,7 +647,9 @@ out_unlock:
 EXPORT_SYMBOL(__close_fd); /* for ksys_close() */
 
 /*
- * variant of __close_fd that gets a ref on the file for later fput
+ * variant of __close_fd that gets a ref on the file for later fput.
+ * The caller must ensure that filp_close() called on the file, and then
+ * an fput().
  */
 int __close_fd_get_file(unsigned int fd, struct file **res)
 {
@@ -662,7 +669,7 @@ int __close_fd_get_file(unsigned int fd, struct file **res)
 	spin_unlock(&files->file_lock);
 	get_file(file);
 	*res = file;
-	return filp_close(file, files);
+	return 0;
 
 out_unlock:
 	spin_unlock(&files->file_lock);
@@ -706,9 +713,9 @@ void do_close_on_exec(struct files_struct *files)
 	spin_unlock(&files->file_lock);
 }
 
-static struct file *__fget(unsigned int fd, fmode_t mask, unsigned int refs)
+static struct file *__fget_files(struct files_struct *files, unsigned int fd,
+				 fmode_t mask, unsigned int refs)
 {
-	struct files_struct *files = current->files;
 	struct file *file;
 
 	rcu_read_lock();
@@ -729,6 +736,12 @@ loop:
 	return file;
 }
 
+static inline struct file *__fget(unsigned int fd, fmode_t mask,
+				  unsigned int refs)
+{
+	return __fget_files(current->files, fd, mask, refs);
+}
+
 struct file *fget_many(unsigned int fd, unsigned int refs)
 {
 	return __fget(fd, FMODE_PATH, refs);
@@ -746,6 +759,18 @@ struct file *fget_raw(unsigned int fd)
 }
 EXPORT_SYMBOL(fget_raw);
 
+struct file *fget_task(struct task_struct *task, unsigned int fd)
+{
+	struct file *file = NULL;
+
+	task_lock(task);
+	if (task->files)
+		file = __fget_files(task->files, fd, 0, 1);
+	task_unlock(task);
+
+	return file;
+}
+
 /*
  * Lightweight file lookup - no refcnt increment if fd table isn't shared.
  *
diff --git a/fs/file_table.c b/fs/file_table.c
index 30d55c9a1744..676e620948d2 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -198,6 +198,7 @@ static struct file *alloc_file(const struct path *path, int flags,
 	file->f_inode = path->dentry->d_inode;
 	file->f_mapping = path->dentry->d_inode->i_mapping;
 	file->f_wb_err = filemap_sample_wb_err(file->f_mapping);
+	file->f_sb_err = file_sample_sb_err(file);
 	if ((file->f_mode & FMODE_READ) &&
 	     likely(fop->read || fop->read_iter))
 		file->f_mode |= FMODE_CAN_READ;
diff --git a/fs/filesystems.c b/fs/filesystems.c
index 9135646e41ac..90b8d879fbaf 100644
--- a/fs/filesystems.c
+++ b/fs/filesystems.c
@@ -74,7 +74,8 @@ int register_filesystem(struct file_system_type * fs)
 	int res = 0;
 	struct file_system_type ** p;
 
-	if (fs->parameters && !fs_validate_description(fs->parameters))
+	if (fs->parameters &&
+	    !fs_validate_description(fs->name, fs->parameters))
 		return -EINVAL;
 
 	BUG_ON(strchr(fs->name, '.'));
@@ -271,7 +272,9 @@ struct file_system_type *get_fs_type(const char *name)
 	fs = __get_fs_type(name, len);
 	if (!fs && (request_module("fs-%.*s", len, name) == 0)) {
 		fs = __get_fs_type(name, len);
-		WARN_ONCE(!fs, "request_module fs-%.*s succeeded, but still no fs?\n", len, name);
+		if (!fs)
+			pr_warn_once("request_module fs-%.*s succeeded, but still no fs?\n",
+				     len, name);
 	}
 
 	if (dot && fs && !(fs->fs_flags & FS_HAS_SUBTYPE)) {
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 335607b8c5c0..a750381d554a 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -1070,7 +1070,6 @@ static void bdi_split_work_to_wbs(struct backing_dev_info *bdi,
 static unsigned long get_nr_dirty_pages(void)
 {
 	return global_node_page_state(NR_FILE_DIRTY) +
-		global_node_page_state(NR_UNSTABLE_NFS) +
 		get_nr_dirty_inodes();
 }
 
@@ -2063,7 +2062,7 @@ void wb_workfn(struct work_struct *work)
 						struct bdi_writeback, dwork);
 	long pages_written;
 
-	set_worker_desc("flush-%s", dev_name(wb->bdi->dev));
+	set_worker_desc("flush-%s", bdi_dev_name(wb->bdi));
 	current->flags |= PF_SWAPWRITE;
 
 	if (likely(!current_is_workqueue_rescuer() ||
@@ -2320,7 +2319,7 @@ void __mark_inode_dirty(struct inode *inode, int flags)
 
 			WARN(bdi_cap_writeback_dirty(wb->bdi) &&
 			     !test_bit(WB_registered, &wb->state),
-			     "bdi-%s not registered\n", wb->bdi->name);
+			     "bdi-%s not registered\n", bdi_dev_name(wb->bdi));
 
 			inode->dirtied_when = jiffies;
 			if (dirtytime)
diff --git a/fs/fs_context.c b/fs/fs_context.c
index 138b5b4d621d..7d5c5dd2b1d5 100644
--- a/fs/fs_context.c
+++ b/fs/fs_context.c
@@ -42,9 +42,9 @@ static const struct constant_table common_set_sb_flag[] = {
 	{ "dirsync",	SB_DIRSYNC },
 	{ "lazytime",	SB_LAZYTIME },
 	{ "mand",	SB_MANDLOCK },
-	{ "posixacl",	SB_POSIXACL },
 	{ "ro",		SB_RDONLY },
 	{ "sync",	SB_SYNCHRONOUS },
+	{ },
 };
 
 static const struct constant_table common_clear_sb_flag[] = {
@@ -52,30 +52,7 @@ static const struct constant_table common_clear_sb_flag[] = {
 	{ "nolazytime",	SB_LAZYTIME },
 	{ "nomand",	SB_MANDLOCK },
 	{ "rw",		SB_RDONLY },
-	{ "silent",	SB_SILENT },
-};
-
-static const char *const forbidden_sb_flag[] = {
-	"bind",
-	"dev",
-	"exec",
-	"move",
-	"noatime",
-	"nodev",
-	"nodiratime",
-	"noexec",
-	"norelatime",
-	"nostrictatime",
-	"nosuid",
-	"private",
-	"rec",
-	"relatime",
-	"remount",
-	"shared",
-	"slave",
-	"strictatime",
-	"suid",
-	"unbindable",
+	{ },
 };
 
 /*
@@ -84,11 +61,6 @@ static const char *const forbidden_sb_flag[] = {
 static int vfs_parse_sb_flag(struct fs_context *fc, const char *key)
 {
 	unsigned int token;
-	unsigned int i;
-
-	for (i = 0; i < ARRAY_SIZE(forbidden_sb_flag); i++)
-		if (strcmp(key, forbidden_sb_flag[i]) == 0)
-			return -EINVAL;
 
 	token = lookup_constant(common_set_sb_flag, key, 0);
 	if (token) {
@@ -175,14 +147,15 @@ int vfs_parse_fs_string(struct fs_context *fc, const char *key,
 
 	struct fs_parameter param = {
 		.key	= key,
-		.type	= fs_value_is_string,
+		.type	= fs_value_is_flag,
 		.size	= v_size,
 	};
 
-	if (v_size > 0) {
+	if (value) {
 		param.string = kmemdup_nul(value, v_size, GFP_KERNEL);
 		if (!param.string)
 			return -ENOMEM;
+		param.type = fs_value_is_string;
 	}
 
 	ret = vfs_parse_fs_param(fc, &param);
@@ -268,6 +241,7 @@ static struct fs_context *alloc_fs_context(struct file_system_type *fs_type,
 	fc->fs_type	= get_filesystem(fs_type);
 	fc->cred	= get_current_cred();
 	fc->net_ns	= get_net(current->nsproxy->net_ns);
+	fc->log.prefix	= fs_type->name;
 
 	mutex_init(&fc->uapi_mutex);
 
@@ -361,8 +335,8 @@ struct fs_context *vfs_dup_fs_context(struct fs_context *src_fc)
 	get_net(fc->net_ns);
 	get_user_ns(fc->user_ns);
 	get_cred(fc->cred);
-	if (fc->log)
-		refcount_inc(&fc->log->usage);
+	if (fc->log.log)
+		refcount_inc(&fc->log.log->usage);
 
 	/* Can't call put until we've called ->dup */
 	ret = fc->ops->dup(fc, src_fc);
@@ -385,64 +359,33 @@ EXPORT_SYMBOL(vfs_dup_fs_context);
  * @fc: The filesystem context to log to.
  * @fmt: The format of the buffer.
  */
-void logfc(struct fs_context *fc, const char *fmt, ...)
+void logfc(struct fc_log *log, const char *prefix, char level, const char *fmt, ...)
 {
-	static const char store_failure[] = "OOM: Can't store error string";
-	struct fc_log *log = fc ? fc->log : NULL;
-	const char *p;
 	va_list va;
-	char *q;
-	u8 freeable;
+	struct va_format vaf = {.fmt = fmt, .va = &va};
 
 	va_start(va, fmt);
-	if (!strchr(fmt, '%')) {
-		p = fmt;
-		goto unformatted_string;
-	}
-	if (strcmp(fmt, "%s") == 0) {
-		p = va_arg(va, const char *);
-		goto unformatted_string;
-	}
-
-	q = kvasprintf(GFP_KERNEL, fmt, va);
-copied_string:
-	if (!q)
-		goto store_failure;
-	freeable = 1;
-	goto store_string;
-
-unformatted_string:
-	if ((unsigned long)p >= (unsigned long)__start_rodata &&
-	    (unsigned long)p <  (unsigned long)__end_rodata)
-		goto const_string;
-	if (log && within_module_core((unsigned long)p, log->owner))
-		goto const_string;
-	q = kstrdup(p, GFP_KERNEL);
-	goto copied_string;
-
-store_failure:
-	p = store_failure;
-const_string:
-	q = (char *)p;
-	freeable = 0;
-store_string:
 	if (!log) {
-		switch (fmt[0]) {
+		switch (level) {
 		case 'w':
-			printk(KERN_WARNING "%s\n", q + 2);
+			printk(KERN_WARNING "%s%s%pV\n", prefix ? prefix : "",
+						prefix ? ": " : "", &vaf);
 			break;
 		case 'e':
-			printk(KERN_ERR "%s\n", q + 2);
+			printk(KERN_ERR "%s%s%pV\n", prefix ? prefix : "",
+						prefix ? ": " : "", &vaf);
 			break;
 		default:
-			printk(KERN_NOTICE "%s\n", q + 2);
+			printk(KERN_NOTICE "%s%s%pV\n", prefix ? prefix : "",
+						prefix ? ": " : "", &vaf);
 			break;
 		}
-		if (freeable)
-			kfree(q);
 	} else {
 		unsigned int logsize = ARRAY_SIZE(log->buffer);
 		u8 index;
+		char *q = kasprintf(GFP_KERNEL, "%c %s%s%pV\n", level,
+						prefix ? prefix : "",
+						prefix ? ": " : "", &vaf);
 
 		index = log->head & (logsize - 1);
 		BUILD_BUG_ON(sizeof(log->head) != sizeof(u8) ||
@@ -454,9 +397,11 @@ store_string:
 			log->tail++;
 		}
 
-		log->buffer[index] = q;
-		log->need_free &= ~(1 << index);
-		log->need_free |= freeable << index;
+		log->buffer[index] = q ? q : "OOM: Can't store error string";
+		if (q)
+			log->need_free |= 1 << index;
+		else
+			log->need_free &= ~(1 << index);
 		log->head++;
 	}
 	va_end(va);
@@ -468,12 +413,12 @@ EXPORT_SYMBOL(logfc);
  */
 static void put_fc_log(struct fs_context *fc)
 {
-	struct fc_log *log = fc->log;
+	struct fc_log *log = fc->log.log;
 	int i;
 
 	if (log) {
 		if (refcount_dec_and_test(&log->usage)) {
-			fc->log = NULL;
+			fc->log.log = NULL;
 			for (i = 0; i <= 7; i++)
 				if (log->need_free & (1 << i))
 					kfree(log->buffer[i]);
diff --git a/fs/fs_parser.c b/fs/fs_parser.c
index d1930adce68d..ab53e42a874a 100644
--- a/fs/fs_parser.c
+++ b/fs/fs_parser.c
@@ -20,42 +20,66 @@ static const struct constant_table bool_names[] = {
 	{ "no",		false },
 	{ "true",	true },
 	{ "yes",	true },
+	{ },
 };
 
+static const struct constant_table *
+__lookup_constant(const struct constant_table *tbl, const char *name)
+{
+	for ( ; tbl->name; tbl++)
+		if (strcmp(name, tbl->name) == 0)
+			return tbl;
+	return NULL;
+}
+
 /**
  * lookup_constant - Look up a constant by name in an ordered table
  * @tbl: The table of constants to search.
- * @tbl_size: The size of the table.
  * @name: The name to look up.
  * @not_found: The value to return if the name is not found.
  */
-int __lookup_constant(const struct constant_table *tbl, size_t tbl_size,
-		      const char *name, int not_found)
+int lookup_constant(const struct constant_table *tbl, const char *name, int not_found)
 {
-	unsigned int i;
+	const struct constant_table *p = __lookup_constant(tbl, name);
 
-	for (i = 0; i < tbl_size; i++)
-		if (strcmp(name, tbl[i].name) == 0)
-			return tbl[i].value;
+	return p ? p->value : not_found;
+}
+EXPORT_SYMBOL(lookup_constant);
 
-	return not_found;
+static inline bool is_flag(const struct fs_parameter_spec *p)
+{
+	return p->type == NULL;
 }
-EXPORT_SYMBOL(__lookup_constant);
 
 static const struct fs_parameter_spec *fs_lookup_key(
-	const struct fs_parameter_description *desc,
-	const char *name)
+	const struct fs_parameter_spec *desc,
+	struct fs_parameter *param, bool *negated)
 {
-	const struct fs_parameter_spec *p;
-
-	if (!desc->specs)
-		return NULL;
-
-	for (p = desc->specs; p->name; p++)
-		if (strcmp(p->name, name) == 0)
+	const struct fs_parameter_spec *p, *other = NULL;
+	const char *name = param->key;
+	bool want_flag = param->type == fs_value_is_flag;
+
+	*negated = false;
+	for (p = desc; p->name; p++) {
+		if (strcmp(p->name, name) != 0)
+			continue;
+		if (likely(is_flag(p) == want_flag))
 			return p;
-
-	return NULL;
+		other = p;
+	}
+	if (want_flag) {
+		if (name[0] == 'n' && name[1] == 'o' && name[2]) {
+			for (p = desc; p->name; p++) {
+				if (strcmp(p->name, name + 2) != 0)
+					continue;
+				if (!(p->flags & fs_param_neg_with_no))
+					continue;
+				*negated = true;
+				return p;
+			}
+		}
+	}
+	return other;
 }
 
 /*
@@ -76,172 +100,38 @@ static const struct fs_parameter_spec *fs_lookup_key(
  * unknown parameters are okay and -EINVAL if there was a conversion issue or
  * the parameter wasn't recognised and unknowns aren't okay.
  */
-int fs_parse(struct fs_context *fc,
-	     const struct fs_parameter_description *desc,
+int __fs_parse(struct p_log *log,
+	     const struct fs_parameter_spec *desc,
 	     struct fs_parameter *param,
 	     struct fs_parse_result *result)
 {
 	const struct fs_parameter_spec *p;
-	const struct fs_parameter_enum *e;
-	int ret = -ENOPARAM, b;
 
-	result->has_value = !!param->string;
-	result->negated = false;
 	result->uint_64 = 0;
 
-	p = fs_lookup_key(desc, param->key);
-	if (!p) {
-		/* If we didn't find something that looks like "noxxx", see if
-		 * "xxx" takes the "no"-form negative - but only if there
-		 * wasn't an value.
-		 */
-		if (result->has_value)
-			goto unknown_parameter;
-		if (param->key[0] != 'n' || param->key[1] != 'o' || !param->key[2])
-			goto unknown_parameter;
-
-		p = fs_lookup_key(desc, param->key + 2);
-		if (!p)
-			goto unknown_parameter;
-		if (!(p->flags & fs_param_neg_with_no))
-			goto unknown_parameter;
-		result->boolean = false;
-		result->negated = true;
-	}
+	p = fs_lookup_key(desc, param, &result->negated);
+	if (!p)
+		return -ENOPARAM;
 
 	if (p->flags & fs_param_deprecated)
-		warnf(fc, "%s: Deprecated parameter '%s'",
-		      desc->name, param->key);
-
-	if (result->negated)
-		goto okay;
-
-	/* Certain parameter types only take a string and convert it. */
-	switch (p->type) {
-	case __fs_param_wasnt_defined:
-		return -EINVAL;
-	case fs_param_is_u32:
-	case fs_param_is_u32_octal:
-	case fs_param_is_u32_hex:
-	case fs_param_is_s32:
-	case fs_param_is_u64:
-	case fs_param_is_enum:
-	case fs_param_is_string:
-		if (param->type != fs_value_is_string)
-			goto bad_value;
-		if (!result->has_value) {
-			if (p->flags & fs_param_v_optional)
-				goto okay;
-			goto bad_value;
-		}
-		/* Fall through */
-	default:
-		break;
-	}
+		warn_plog(log, "Deprecated parameter '%s'", param->key);
 
 	/* Try to turn the type we were given into the type desired by the
 	 * parameter and give an error if we can't.
 	 */
-	switch (p->type) {
-	case fs_param_is_flag:
-		if (param->type != fs_value_is_flag &&
-		    (param->type != fs_value_is_string || result->has_value))
-			return invalf(fc, "%s: Unexpected value for '%s'",
-				      desc->name, param->key);
-		result->boolean = true;
-		goto okay;
-
-	case fs_param_is_bool:
-		switch (param->type) {
-		case fs_value_is_flag:
-			result->boolean = true;
-			goto okay;
-		case fs_value_is_string:
-			if (param->size == 0) {
-				result->boolean = true;
-				goto okay;
-			}
-			b = lookup_constant(bool_names, param->string, -1);
-			if (b == -1)
-				goto bad_value;
-			result->boolean = b;
-			goto okay;
-		default:
-			goto bad_value;
-		}
-
-	case fs_param_is_u32:
-		ret = kstrtouint(param->string, 0, &result->uint_32);
-		goto maybe_okay;
-	case fs_param_is_u32_octal:
-		ret = kstrtouint(param->string, 8, &result->uint_32);
-		goto maybe_okay;
-	case fs_param_is_u32_hex:
-		ret = kstrtouint(param->string, 16, &result->uint_32);
-		goto maybe_okay;
-	case fs_param_is_s32:
-		ret = kstrtoint(param->string, 0, &result->int_32);
-		goto maybe_okay;
-	case fs_param_is_u64:
-		ret = kstrtoull(param->string, 0, &result->uint_64);
-		goto maybe_okay;
-
-	case fs_param_is_enum:
-		for (e = desc->enums; e->name[0]; e++) {
-			if (e->opt == p->opt &&
-			    strcmp(e->name, param->string) == 0) {
-				result->uint_32 = e->value;
-				goto okay;
-			}
-		}
-		goto bad_value;
-
-	case fs_param_is_string:
-		goto okay;
-	case fs_param_is_blob:
-		if (param->type != fs_value_is_blob)
-			goto bad_value;
-		goto okay;
-
-	case fs_param_is_fd: {
-		switch (param->type) {
-		case fs_value_is_string:
-			if (!result->has_value)
-				goto bad_value;
-
-			ret = kstrtouint(param->string, 0, &result->uint_32);
-			break;
-		case fs_value_is_file:
-			result->uint_32 = param->dirfd;
-			ret = 0;
-		default:
-			goto bad_value;
-		}
-
-		if (result->uint_32 > INT_MAX)
-			goto bad_value;
-		goto maybe_okay;
-	}
-
-	case fs_param_is_blockdev:
-	case fs_param_is_path:
-		goto okay;
-	default:
-		BUG();
+	if (is_flag(p)) {
+		if (param->type != fs_value_is_flag)
+			return inval_plog(log, "Unexpected value for '%s'",
+				      param->key);
+		result->boolean = !result->negated;
+	} else  {
+		int ret = p->type(log, p, param, result);
+		if (ret)
+			return ret;
 	}
-
-maybe_okay:
-	if (ret < 0)
-		goto bad_value;
-okay:
 	return p->opt;
-
-bad_value:
-	return invalf(fc, "%s: Bad value for '%s'", desc->name, param->key);
-unknown_parameter:
-	return -ENOPARAM;
 }
-EXPORT_SYMBOL(fs_parse);
+EXPORT_SYMBOL(__fs_parse);
 
 /**
  * fs_lookup_param - Look up a path referred to by a parameter
@@ -267,9 +157,6 @@ int fs_lookup_param(struct fs_context *fc,
 			return PTR_ERR(f);
 		put_f = true;
 		break;
-	case fs_value_is_filename_empty:
-		flags = LOOKUP_EMPTY;
-		/* Fall through */
 	case fs_value_is_filename:
 		f = param->name;
 		put_f = false;
@@ -302,6 +189,124 @@ out:
 }
 EXPORT_SYMBOL(fs_lookup_param);
 
+int fs_param_bad_value(struct p_log *log, struct fs_parameter *param)
+{
+	return inval_plog(log, "Bad value for '%s'", param->key);
+}
+
+int fs_param_is_bool(struct p_log *log, const struct fs_parameter_spec *p,
+		     struct fs_parameter *param, struct fs_parse_result *result)
+{
+	int b;
+	if (param->type != fs_value_is_string)
+		return fs_param_bad_value(log, param);
+	b = lookup_constant(bool_names, param->string, -1);
+	if (b == -1)
+		return fs_param_bad_value(log, param);
+	result->boolean = b;
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_bool);
+
+int fs_param_is_u32(struct p_log *log, const struct fs_parameter_spec *p,
+		    struct fs_parameter *param, struct fs_parse_result *result)
+{
+	int base = (unsigned long)p->data;
+	if (param->type != fs_value_is_string ||
+	    kstrtouint(param->string, base, &result->uint_32) < 0)
+		return fs_param_bad_value(log, param);
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_u32);
+
+int fs_param_is_s32(struct p_log *log, const struct fs_parameter_spec *p,
+		    struct fs_parameter *param, struct fs_parse_result *result)
+{
+	if (param->type != fs_value_is_string ||
+	    kstrtoint(param->string, 0, &result->int_32) < 0)
+		return fs_param_bad_value(log, param);
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_s32);
+
+int fs_param_is_u64(struct p_log *log, const struct fs_parameter_spec *p,
+		    struct fs_parameter *param, struct fs_parse_result *result)
+{
+	if (param->type != fs_value_is_string ||
+	    kstrtoull(param->string, 0, &result->uint_64) < 0)
+		return fs_param_bad_value(log, param);
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_u64);
+
+int fs_param_is_enum(struct p_log *log, const struct fs_parameter_spec *p,
+		     struct fs_parameter *param, struct fs_parse_result *result)
+{
+	const struct constant_table *c;
+	if (param->type != fs_value_is_string)
+		return fs_param_bad_value(log, param);
+	c = __lookup_constant(p->data, param->string);
+	if (!c)
+		return fs_param_bad_value(log, param);
+	result->uint_32 = c->value;
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_enum);
+
+int fs_param_is_string(struct p_log *log, const struct fs_parameter_spec *p,
+		       struct fs_parameter *param, struct fs_parse_result *result)
+{
+	if (param->type != fs_value_is_string || !*param->string)
+		return fs_param_bad_value(log, param);
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_string);
+
+int fs_param_is_blob(struct p_log *log, const struct fs_parameter_spec *p,
+		     struct fs_parameter *param, struct fs_parse_result *result)
+{
+	if (param->type != fs_value_is_blob)
+		return fs_param_bad_value(log, param);
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_blob);
+
+int fs_param_is_fd(struct p_log *log, const struct fs_parameter_spec *p,
+		  struct fs_parameter *param, struct fs_parse_result *result)
+{
+	switch (param->type) {
+	case fs_value_is_string:
+		if (kstrtouint(param->string, 0, &result->uint_32) < 0)
+			break;
+		if (result->uint_32 <= INT_MAX)
+			return 0;
+		break;
+	case fs_value_is_file:
+		result->uint_32 = param->dirfd;
+		if (result->uint_32 <= INT_MAX)
+			return 0;
+		break;
+	default:
+		break;
+	}
+	return fs_param_bad_value(log, param);
+}
+EXPORT_SYMBOL(fs_param_is_fd);
+
+int fs_param_is_blockdev(struct p_log *log, const struct fs_parameter_spec *p,
+		  struct fs_parameter *param, struct fs_parse_result *result)
+{
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_blockdev);
+
+int fs_param_is_path(struct p_log *log, const struct fs_parameter_spec *p,
+		     struct fs_parameter *param, struct fs_parse_result *result)
+{
+	return 0;
+}
+EXPORT_SYMBOL(fs_param_is_path);
+
 #ifdef CONFIG_VALIDATE_FS_PARSER
 /**
  * validate_constant_table - Validate a constant table
@@ -357,102 +362,24 @@ bool validate_constant_table(const struct constant_table *tbl, size_t tbl_size,
  * fs_validate_description - Validate a parameter description
  * @desc: The parameter description to validate.
  */
-bool fs_validate_description(const struct fs_parameter_description *desc)
+bool fs_validate_description(const char *name,
+	const struct fs_parameter_spec *desc)
 {
 	const struct fs_parameter_spec *param, *p2;
-	const struct fs_parameter_enum *e;
-	const char *name = desc->name;
-	unsigned int nr_params = 0;
-	bool good = true, enums = false;
-
-	pr_notice("*** VALIDATE %s ***\n", name);
-
-	if (!name[0]) {
-		pr_err("VALIDATE Parser: No name\n");
-		name = "Unknown";
-		good = false;
-	}
-
-	if (desc->specs) {
-		for (param = desc->specs; param->name; param++) {
-			enum fs_parameter_type t = param->type;
-
-			/* Check that the type is in range */
-			if (t == __fs_param_wasnt_defined ||
-			    t >= nr__fs_parameter_type) {
-				pr_err("VALIDATE %s: PARAM[%s] Bad type %u\n",
-				       name, param->name, t);
-				good = false;
-			} else if (t == fs_param_is_enum) {
-				enums = true;
-			}
-
-			/* Check for duplicate parameter names */
-			for (p2 = desc->specs; p2 < param; p2++) {
-				if (strcmp(param->name, p2->name) == 0) {
-					pr_err("VALIDATE %s: PARAM[%s]: Duplicate\n",
-					       name, param->name);
-					good = false;
-				}
-			}
-		}
-
-		nr_params = param - desc->specs;
-	}
-
-	if (desc->enums) {
-		if (!nr_params) {
-			pr_err("VALIDATE %s: Enum table but no parameters\n",
-			       name);
-			good = false;
-			goto no_enums;
-		}
-		if (!enums) {
-			pr_err("VALIDATE %s: Enum table but no enum-type values\n",
-			       name);
-			good = false;
-			goto no_enums;
-		}
-
-		for (e = desc->enums; e->name[0]; e++) {
-			/* Check that all entries in the enum table have at
-			 * least one parameter that uses them.
-			 */
-			for (param = desc->specs; param->name; param++) {
-				if (param->opt == e->opt &&
-				    param->type != fs_param_is_enum) {
-					pr_err("VALIDATE %s: e[%tu] enum val for %s\n",
-					       name, e - desc->enums, param->name);
-					good = false;
-				}
-			}
-		}
+	bool good = true;
 
-		/* Check that all enum-type parameters have at least one enum
-		 * value in the enum table.
-		 */
-		for (param = desc->specs; param->name; param++) {
-			if (param->type != fs_param_is_enum)
-				continue;
-			for (e = desc->enums; e->name[0]; e++)
-				if (e->opt == param->opt)
-					break;
-			if (!e->name[0]) {
-				pr_err("VALIDATE %s: PARAM[%s] enum with no values\n",
+	for (param = desc; param->name; param++) {
+		/* Check for duplicate parameter names */
+		for (p2 = desc; p2 < param; p2++) {
+			if (strcmp(param->name, p2->name) == 0) {
+				if (is_flag(param) != is_flag(p2))
+					continue;
+				pr_err("VALIDATE %s: PARAM[%s]: Duplicate\n",
 				       name, param->name);
 				good = false;
 			}
 		}
-	} else {
-		if (enums) {
-			pr_err("VALIDATE %s: enum-type values, but no enum table\n",
-			       name);
-			good = false;
-			goto no_enums;
-		}
 	}
-
-no_enums:
 	return good;
 }
 #endif /* CONFIG_VALIDATE_FS_PARSER */
diff --git a/fs/fscache/Kconfig b/fs/fscache/Kconfig
index 506c5e643f0d..5e796e6c38e5 100644
--- a/fs/fscache/Kconfig
+++ b/fs/fscache/Kconfig
@@ -8,7 +8,7 @@ config FSCACHE
 	  Different sorts of caches can be plugged in, depending on the
 	  resources available.
 
-	  See Documentation/filesystems/caching/fscache.txt for more information.
+	  See Documentation/filesystems/caching/fscache.rst for more information.
 
 config FSCACHE_STATS
 	bool "Gather statistical information on local caching"
@@ -25,7 +25,7 @@ config FSCACHE_STATS
 	  between CPUs.  On the other hand, the stats are very useful for
 	  debugging purposes.  Saying 'Y' here is recommended.
 
-	  See Documentation/filesystems/caching/fscache.txt for more information.
+	  See Documentation/filesystems/caching/fscache.rst for more information.
 
 config FSCACHE_HISTOGRAM
 	bool "Gather latency information on local caching"
@@ -42,7 +42,7 @@ config FSCACHE_HISTOGRAM
 	  bouncing between CPUs.  On the other hand, the histogram may be
 	  useful for debugging purposes.  Saying 'N' here is recommended.
 
-	  See Documentation/filesystems/caching/fscache.txt for more information.
+	  See Documentation/filesystems/caching/fscache.rst for more information.
 
 config FSCACHE_DEBUG
 	bool "Debug FS-Cache"
@@ -52,7 +52,7 @@ config FSCACHE_DEBUG
 	  management module.  If this is set, the debugging output may be
 	  enabled by setting bits in /sys/modules/fscache/parameter/debug.
 
-	  See Documentation/filesystems/caching/fscache.txt for more information.
+	  See Documentation/filesystems/caching/fscache.rst for more information.
 
 config FSCACHE_OBJECT_LIST
 	bool "Maintain global object list for debugging purposes"
diff --git a/fs/fscache/cache.c b/fs/fscache/cache.c
index f78793f3d21e..fcc136361415 100644
--- a/fs/fscache/cache.c
+++ b/fs/fscache/cache.c
@@ -172,7 +172,7 @@ no_preference:
  *
  * Initialise a record of a cache and fill in the name.
  *
- * See Documentation/filesystems/caching/backend-api.txt for a complete
+ * See Documentation/filesystems/caching/backend-api.rst for a complete
  * description.
  */
 void fscache_init_cache(struct fscache_cache *cache,
@@ -207,7 +207,7 @@ EXPORT_SYMBOL(fscache_init_cache);
  *
  * Add a cache to the system, making it available for netfs's to use.
  *
- * See Documentation/filesystems/caching/backend-api.txt for a complete
+ * See Documentation/filesystems/caching/backend-api.rst for a complete
  * description.
  */
 int fscache_add_cache(struct fscache_cache *cache,
@@ -307,7 +307,7 @@ EXPORT_SYMBOL(fscache_add_cache);
  * Note that an I/O error occurred in a cache and that it should no longer be
  * used for anything.  This also reports the error into the kernel log.
  *
- * See Documentation/filesystems/caching/backend-api.txt for a complete
+ * See Documentation/filesystems/caching/backend-api.rst for a complete
  * description.
  */
 void fscache_io_error(struct fscache_cache *cache)
@@ -355,7 +355,7 @@ static void fscache_withdraw_all_objects(struct fscache_cache *cache,
  * Withdraw a cache from service, unbinding all its cache objects from the
  * netfs cookies they're currently representing.
  *
- * See Documentation/filesystems/caching/backend-api.txt for a complete
+ * See Documentation/filesystems/caching/backend-api.rst for a complete
  * description.
  */
 void fscache_withdraw_cache(struct fscache_cache *cache)
diff --git a/fs/fscache/cookie.c b/fs/fscache/cookie.c
index 0ce39658a620..751bc5b1cddf 100644
--- a/fs/fscache/cookie.c
+++ b/fs/fscache/cookie.c
@@ -4,7 +4,7 @@
  * Copyright (C) 2004-2007 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
- * See Documentation/filesystems/caching/netfs-api.txt for more information on
+ * See Documentation/filesystems/caching/netfs-api.rst for more information on
  * the netfs API.
  */
 
diff --git a/fs/fscache/internal.h b/fs/fscache/internal.h
index 9616af3768e1..08e91efbce53 100644
--- a/fs/fscache/internal.h
+++ b/fs/fscache/internal.h
@@ -111,7 +111,7 @@ extern void fscache_enqueue_object(struct fscache_object *);
  * object-list.c
  */
 #ifdef CONFIG_FSCACHE_OBJECT_LIST
-extern const struct file_operations fscache_objlist_fops;
+extern const struct proc_ops fscache_objlist_proc_ops;
 
 extern void fscache_objlist_add(struct fscache_object *);
 extern void fscache_objlist_remove(struct fscache_object *);
diff --git a/fs/fscache/object-list.c b/fs/fscache/object-list.c
index 72ebfe578f40..e106a1a1600d 100644
--- a/fs/fscache/object-list.c
+++ b/fs/fscache/object-list.c
@@ -7,6 +7,7 @@
 
 #define FSCACHE_DEBUG_LEVEL COOKIE
 #include <linux/module.h>
+#include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/slab.h>
 #include <linux/key.h>
@@ -405,9 +406,9 @@ static int fscache_objlist_release(struct inode *inode, struct file *file)
 	return seq_release(inode, file);
 }
 
-const struct file_operations fscache_objlist_fops = {
-	.open		= fscache_objlist_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= fscache_objlist_release,
+const struct proc_ops fscache_objlist_proc_ops = {
+	.proc_open	= fscache_objlist_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= fscache_objlist_release,
 };
diff --git a/fs/fscache/object.c b/fs/fscache/object.c
index cfeba839a0f2..cb2146e02cd5 100644
--- a/fs/fscache/object.c
+++ b/fs/fscache/object.c
@@ -4,7 +4,7 @@
  * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
- * See Documentation/filesystems/caching/object.txt for a description of the
+ * See Documentation/filesystems/caching/object.rst for a description of the
  * object state machine and the in-kernel representations.
  */
 
@@ -295,7 +295,7 @@ static void fscache_object_work_func(struct work_struct *work)
  *
  * Initialise a cache object description to its basic values.
  *
- * See Documentation/filesystems/caching/backend-api.txt for a complete
+ * See Documentation/filesystems/caching/backend-api.rst for a complete
  * description.
  */
 void fscache_object_init(struct fscache_object *object,
diff --git a/fs/fscache/operation.c b/fs/fscache/operation.c
index 1a22a55f75a0..4a5651d4904e 100644
--- a/fs/fscache/operation.c
+++ b/fs/fscache/operation.c
@@ -4,7 +4,7 @@
  * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
- * See Documentation/filesystems/caching/operations.txt
+ * See Documentation/filesystems/caching/operations.rst
  */
 
 #define FSCACHE_DEBUG_LEVEL OPERATION
diff --git a/fs/fscache/proc.c b/fs/fscache/proc.c
index 5523446e2952..90a7bc22f7e1 100644
--- a/fs/fscache/proc.c
+++ b/fs/fscache/proc.c
@@ -35,7 +35,7 @@ int __init fscache_proc_init(void)
 
 #ifdef CONFIG_FSCACHE_OBJECT_LIST
 	if (!proc_create("fs/fscache/objects", S_IFREG | 0444, NULL,
-			 &fscache_objlist_fops))
+			 &fscache_objlist_proc_ops))
 		goto error_objects;
 #endif
 
diff --git a/fs/fsopen.c b/fs/fsopen.c
index 043ffa8dc263..2fa3f241b762 100644
--- a/fs/fsopen.c
+++ b/fs/fsopen.c
@@ -25,7 +25,7 @@ static ssize_t fscontext_read(struct file *file,
 			      char __user *_buf, size_t len, loff_t *pos)
 {
 	struct fs_context *fc = file->private_data;
-	struct fc_log *log = fc->log;
+	struct fc_log *log = fc->log.log;
 	unsigned int logsize = ARRAY_SIZE(log->buffer);
 	ssize_t ret;
 	char *p;
@@ -97,11 +97,11 @@ static int fscontext_create_fd(struct fs_context *fc, unsigned int o_flags)
 
 static int fscontext_alloc_log(struct fs_context *fc)
 {
-	fc->log = kzalloc(sizeof(*fc->log), GFP_KERNEL);
-	if (!fc->log)
+	fc->log.log = kzalloc(sizeof(*fc->log.log), GFP_KERNEL);
+	if (!fc->log.log)
 		return -ENOMEM;
-	refcount_set(&fc->log->usage, 1);
-	fc->log->owner = fc->fs_type->owner;
+	refcount_set(&fc->log.log->usage, 1);
+	fc->log.log->owner = fc->fs_type->owner;
 	return 0;
 }
 
@@ -321,6 +321,7 @@ SYSCALL_DEFINE5(fsconfig,
 	struct fs_context *fc;
 	struct fd f;
 	int ret;
+	int lookup_flags = 0;
 
 	struct fs_parameter param = {
 		.type	= fs_value_is_undefined,
@@ -409,19 +410,12 @@ SYSCALL_DEFINE5(fsconfig,
 			goto out_key;
 		}
 		break;
+	case FSCONFIG_SET_PATH_EMPTY:
+		lookup_flags = LOOKUP_EMPTY;
+		/* fallthru */
 	case FSCONFIG_SET_PATH:
 		param.type = fs_value_is_filename;
-		param.name = getname_flags(_value, 0, NULL);
-		if (IS_ERR(param.name)) {
-			ret = PTR_ERR(param.name);
-			goto out_key;
-		}
-		param.dirfd = aux;
-		param.size = strlen(param.name->name);
-		break;
-	case FSCONFIG_SET_PATH_EMPTY:
-		param.type = fs_value_is_filename_empty;
-		param.name = getname_flags(_value, LOOKUP_EMPTY, NULL);
+		param.name = getname_flags(_value, lookup_flags, NULL);
 		if (IS_ERR(param.name)) {
 			ret = PTR_ERR(param.name);
 			goto out_key;
diff --git a/fs/fuse/Kconfig b/fs/fuse/Kconfig
index 0635cba19971..774b2618018a 100644
--- a/fs/fuse/Kconfig
+++ b/fs/fuse/Kconfig
@@ -12,7 +12,7 @@ config FUSE_FS
 	  although chances are your distribution already has that library
 	  installed if you've installed the "fuse" package itself.
 
-	  See <file:Documentation/filesystems/fuse.txt> for more information.
+	  See <file:Documentation/filesystems/fuse.rst> for more information.
 	  See <file:Documentation/Changes> for needed library/utility version.
 
 	  If you want to develop a userspace FS, or if you want to use
@@ -34,7 +34,7 @@ config VIRTIO_FS
 	select VIRTIO
 	help
 	  The Virtio Filesystem allows guests to mount file systems from the
-          host.
+	  host.
 
 	  If you want to share files between guests or with the host, answer Y
-          or M.
+	  or M.
diff --git a/fs/fuse/cuse.c b/fs/fuse/cuse.c
index 00015d851382..030f094910c3 100644
--- a/fs/fuse/cuse.c
+++ b/fs/fuse/cuse.c
@@ -451,8 +451,8 @@ static int cuse_send_init(struct cuse_conn *cc)
 	ap->args.out_args[0].size = sizeof(ia->out);
 	ap->args.out_args[0].value = &ia->out;
 	ap->args.out_args[1].size = CUSE_INIT_INFO_MAX;
-	ap->args.out_argvar = 1;
-	ap->args.out_pages = 1;
+	ap->args.out_argvar = true;
+	ap->args.out_pages = true;
 	ap->num_pages = 1;
 	ap->pages = &ia->page;
 	ap->descs = &ia->desc;
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index ed1abc9e33cf..c7a65cf2bcca 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -276,12 +276,10 @@ static void flush_bg_queue(struct fuse_conn *fc)
 void fuse_request_end(struct fuse_conn *fc, struct fuse_req *req)
 {
 	struct fuse_iqueue *fiq = &fc->iq;
-	bool async;
 
 	if (test_and_set_bit(FR_FINISHED, &req->flags))
 		goto put_request;
 
-	async = req->args->end;
 	/*
 	 * test_and_set_bit() implies smp_mb() between bit
 	 * changing and below intr_entry check. Pairs with
@@ -324,7 +322,7 @@ void fuse_request_end(struct fuse_conn *fc, struct fuse_req *req)
 		wake_up(&req->waitq);
 	}
 
-	if (async)
+	if (test_bit(FR_ASYNC, &req->flags))
 		req->args->end(fc, req->args, req->out.h.error);
 put_request:
 	fuse_put_request(fc, req);
@@ -471,6 +469,8 @@ static void fuse_args_to_req(struct fuse_req *req, struct fuse_args *args)
 	req->in.h.opcode = args->opcode;
 	req->in.h.nodeid = args->nodeid;
 	req->args = args;
+	if (args->end)
+		__set_bit(FR_ASYNC, &req->flags);
 }
 
 ssize_t fuse_simple_request(struct fuse_conn *fc, struct fuse_args *args)
@@ -705,7 +705,7 @@ static int fuse_copy_fill(struct fuse_copy_state *cs)
 			cs->pipebufs++;
 			cs->nr_segs--;
 		} else {
-			if (cs->nr_segs == cs->pipe->buffers)
+			if (cs->nr_segs >= cs->pipe->max_usage)
 				return -EIO;
 
 			page = alloc_page(GFP_HIGHUSER);
@@ -881,7 +881,7 @@ static int fuse_ref_page(struct fuse_copy_state *cs, struct page *page,
 	struct pipe_buffer *buf;
 	int err;
 
-	if (cs->nr_segs == cs->pipe->buffers)
+	if (cs->nr_segs >= cs->pipe->max_usage)
 		return -EIO;
 
 	err = unlock_request(cs->req);
@@ -1343,7 +1343,7 @@ static ssize_t fuse_dev_splice_read(struct file *in, loff_t *ppos,
 	if (!fud)
 		return -EPERM;
 
-	bufs = kvmalloc_array(pipe->buffers, sizeof(struct pipe_buffer),
+	bufs = kvmalloc_array(pipe->max_usage, sizeof(struct pipe_buffer),
 			      GFP_KERNEL);
 	if (!bufs)
 		return -ENOMEM;
@@ -1355,7 +1355,7 @@ static ssize_t fuse_dev_splice_read(struct file *in, loff_t *ppos,
 	if (ret < 0)
 		goto out;
 
-	if (pipe->nrbufs + cs.nr_segs > pipe->buffers) {
+	if (pipe_occupancy(pipe->head, pipe->tail) + cs.nr_segs > pipe->max_usage) {
 		ret = -EIO;
 		goto out;
 	}
@@ -1937,6 +1937,7 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe,
 				     struct file *out, loff_t *ppos,
 				     size_t len, unsigned int flags)
 {
+	unsigned int head, tail, mask, count;
 	unsigned nbuf;
 	unsigned idx;
 	struct pipe_buffer *bufs;
@@ -1951,8 +1952,12 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe,
 
 	pipe_lock(pipe);
 
-	bufs = kvmalloc_array(pipe->nrbufs, sizeof(struct pipe_buffer),
-			      GFP_KERNEL);
+	head = pipe->head;
+	tail = pipe->tail;
+	mask = pipe->ring_size - 1;
+	count = head - tail;
+
+	bufs = kvmalloc_array(count, sizeof(struct pipe_buffer), GFP_KERNEL);
 	if (!bufs) {
 		pipe_unlock(pipe);
 		return -ENOMEM;
@@ -1960,8 +1965,8 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe,
 
 	nbuf = 0;
 	rem = 0;
-	for (idx = 0; idx < pipe->nrbufs && rem < len; idx++)
-		rem += pipe->bufs[(pipe->curbuf + idx) & (pipe->buffers - 1)].len;
+	for (idx = tail; idx != head && rem < len; idx++)
+		rem += pipe->bufs[idx & mask].len;
 
 	ret = -EINVAL;
 	if (rem < len)
@@ -1972,16 +1977,16 @@ static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe,
 		struct pipe_buffer *ibuf;
 		struct pipe_buffer *obuf;
 
-		BUG_ON(nbuf >= pipe->buffers);
-		BUG_ON(!pipe->nrbufs);
-		ibuf = &pipe->bufs[pipe->curbuf];
+		BUG_ON(nbuf >= pipe->ring_size);
+		BUG_ON(tail == head);
+		ibuf = &pipe->bufs[tail & mask];
 		obuf = &bufs[nbuf];
 
 		if (rem >= ibuf->len) {
 			*obuf = *ibuf;
 			ibuf->ops = NULL;
-			pipe->curbuf = (pipe->curbuf + 1) & (pipe->buffers - 1);
-			pipe->nrbufs--;
+			tail++;
+			pipe->tail = tail;
 		} else {
 			if (!pipe_buf_get(pipe, ibuf))
 				goto out_free;
@@ -2076,7 +2081,7 @@ static void end_polls(struct fuse_conn *fc)
  * The same effect is usually achievable through killing the filesystem daemon
  * and all users of the filesystem.  The exception is the combination of an
  * asynchronous request and the tricky deadlock (see
- * Documentation/filesystems/fuse.txt).
+ * Documentation/filesystems/fuse.rst).
  *
  * Aborting requests under I/O goes as follows: 1: Separate out unlocked
  * requests, they should be finished off immediately.  Locked requests will be
@@ -2262,7 +2267,7 @@ const struct file_operations fuse_dev_operations = {
 	.release	= fuse_dev_release,
 	.fasync		= fuse_dev_fasync,
 	.unlocked_ioctl = fuse_dev_ioctl,
-	.compat_ioctl   = fuse_dev_ioctl,
+	.compat_ioctl   = compat_ptr_ioctl,
 };
 EXPORT_SYMBOL_GPL(fuse_dev_operations);
 
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 54d638f9ba1c..de1e2fde60bd 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -248,7 +248,8 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags)
 		kfree(forget);
 		if (ret == -ENOMEM)
 			goto out;
-		if (ret || (outarg.attr.mode ^ inode->i_mode) & S_IFMT)
+		if (ret || fuse_invalid_attr(&outarg.attr) ||
+		    (outarg.attr.mode ^ inode->i_mode) & S_IFMT)
 			goto invalid;
 
 		forget_all_cached_acls(inode);
@@ -319,6 +320,12 @@ int fuse_valid_type(int m)
 		S_ISBLK(m) || S_ISFIFO(m) || S_ISSOCK(m);
 }
 
+bool fuse_invalid_attr(struct fuse_attr *attr)
+{
+	return !fuse_valid_type(attr->mode) ||
+		attr->size > LLONG_MAX;
+}
+
 int fuse_lookup_name(struct super_block *sb, u64 nodeid, const struct qstr *name,
 		     struct fuse_entry_out *outarg, struct inode **inode)
 {
@@ -350,7 +357,7 @@ int fuse_lookup_name(struct super_block *sb, u64 nodeid, const struct qstr *name
 	err = -EIO;
 	if (!outarg->nodeid)
 		goto out_put_forget;
-	if (!fuse_valid_type(outarg->attr.mode))
+	if (fuse_invalid_attr(&outarg->attr))
 		goto out_put_forget;
 
 	*inode = fuse_iget(sb, outarg->nodeid, outarg->generation,
@@ -475,7 +482,8 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry,
 		goto out_free_ff;
 
 	err = -EIO;
-	if (!S_ISREG(outentry.attr.mode) || invalid_nodeid(outentry.nodeid))
+	if (!S_ISREG(outentry.attr.mode) || invalid_nodeid(outentry.nodeid) ||
+	    fuse_invalid_attr(&outentry.attr))
 		goto out_free_ff;
 
 	ff->fh = outopen.fh;
@@ -583,7 +591,7 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_args *args,
 		goto out_put_forget_req;
 
 	err = -EIO;
-	if (invalid_nodeid(outarg.nodeid))
+	if (invalid_nodeid(outarg.nodeid) || fuse_invalid_attr(&outarg.attr))
 		goto out_put_forget_req;
 
 	if ((outarg.attr.mode ^ mode) & S_IFMT)
@@ -810,7 +818,7 @@ static int fuse_rename2(struct inode *olddir, struct dentry *oldent,
 	struct fuse_conn *fc = get_fuse_conn(olddir);
 	int err;
 
-	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
+	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT))
 		return -EINVAL;
 
 	if (flags) {
@@ -862,7 +870,8 @@ static int fuse_link(struct dentry *entry, struct inode *newdir,
 
 		spin_lock(&fi->lock);
 		fi->attr_version = atomic64_inc_return(&fc->attr_version);
-		inc_nlink(inode);
+		if (likely(inode->i_nlink < UINT_MAX))
+			inc_nlink(inode);
 		spin_unlock(&fi->lock);
 		fuse_invalidate_attr(inode);
 		fuse_update_ctime(inode);
@@ -942,7 +951,8 @@ static int fuse_do_getattr(struct inode *inode, struct kstat *stat,
 	args.out_args[0].value = &outarg;
 	err = fuse_simple_request(fc, &args);
 	if (!err) {
-		if ((inode->i_mode ^ outarg.attr.mode) & S_IFMT) {
+		if (fuse_invalid_attr(&outarg.attr) ||
+		    (inode->i_mode ^ outarg.attr.mode) & S_IFMT) {
 			make_bad_inode(inode);
 			err = -EIO;
 		} else {
@@ -1563,7 +1573,8 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
 		goto error;
 	}
 
-	if ((inode->i_mode ^ outarg.attr.mode) & S_IFMT) {
+	if (fuse_invalid_attr(&outarg.attr) ||
+	    (inode->i_mode ^ outarg.attr.mode) & S_IFMT) {
 		make_bad_inode(inode);
 		err = -EIO;
 		goto error;
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index db48a5cf8620..bac51c32d660 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -713,8 +713,10 @@ static ssize_t fuse_async_req_send(struct fuse_conn *fc,
 
 	ia->ap.args.end = fuse_aio_complete_req;
 	err = fuse_simple_background(fc, &ia->ap.args, GFP_KERNEL);
+	if (err)
+		fuse_aio_complete_req(fc, &ia->ap.args, err);
 
-	return err ?: num_bytes;
+	return num_bytes;
 }
 
 static ssize_t fuse_send_read(struct fuse_io_args *ia, loff_t pos, size_t count,
@@ -801,6 +803,10 @@ static int fuse_do_readpage(struct file *file, struct page *page)
 
 	attr_ver = fuse_get_attr_version(fc);
 
+	/* Don't overflow end offset */
+	if (pos + (desc.length - 1) == LLONG_MAX)
+		desc.length--;
+
 	fuse_read_args_fill(&ia, file, pos, desc.length, FUSE_READ);
 	res = fuse_simple_request(fc, &ia.ap.args);
 	if (res < 0)
@@ -880,11 +886,20 @@ static void fuse_send_readpages(struct fuse_io_args *ia, struct file *file)
 	struct fuse_args_pages *ap = &ia->ap;
 	loff_t pos = page_offset(ap->pages[0]);
 	size_t count = ap->num_pages << PAGE_SHIFT;
+	ssize_t res;
 	int err;
 
 	ap->args.out_pages = true;
 	ap->args.page_zeroing = true;
 	ap->args.page_replace = true;
+
+	/* Don't overflow end offset */
+	if (pos + (count - 1) == LLONG_MAX) {
+		count--;
+		ap->descs[ap->num_pages - 1].length--;
+	}
+	WARN_ON((loff_t) (pos + count) < 0);
+
 	fuse_read_args_fill(ia, file, pos, count, FUSE_READ);
 	ia->read.attr_ver = fuse_get_attr_version(fc);
 	if (fc->async_read) {
@@ -894,89 +909,46 @@ static void fuse_send_readpages(struct fuse_io_args *ia, struct file *file)
 		if (!err)
 			return;
 	} else {
-		err = fuse_simple_request(fc, &ap->args);
+		res = fuse_simple_request(fc, &ap->args);
+		err = res < 0 ? res : 0;
 	}
 	fuse_readpages_end(fc, &ap->args, err);
 }
 
-struct fuse_fill_data {
-	struct fuse_io_args *ia;
-	struct file *file;
-	struct inode *inode;
-	unsigned int nr_pages;
-	unsigned int max_pages;
-};
-
-static int fuse_readpages_fill(void *_data, struct page *page)
+static void fuse_readahead(struct readahead_control *rac)
 {
-	struct fuse_fill_data *data = _data;
-	struct fuse_io_args *ia = data->ia;
-	struct fuse_args_pages *ap = &ia->ap;
-	struct inode *inode = data->inode;
+	struct inode *inode = rac->mapping->host;
 	struct fuse_conn *fc = get_fuse_conn(inode);
+	unsigned int i, max_pages, nr_pages = 0;
 
-	fuse_wait_on_page_writeback(inode, page->index);
-
-	if (ap->num_pages &&
-	    (ap->num_pages == fc->max_pages ||
-	     (ap->num_pages + 1) * PAGE_SIZE > fc->max_read ||
-	     ap->pages[ap->num_pages - 1]->index + 1 != page->index)) {
-		data->max_pages = min_t(unsigned int, data->nr_pages,
-					fc->max_pages);
-		fuse_send_readpages(ia, data->file);
-		data->ia = ia = fuse_io_alloc(NULL, data->max_pages);
-		if (!ia) {
-			unlock_page(page);
-			return -ENOMEM;
-		}
-		ap = &ia->ap;
-	}
-
-	if (WARN_ON(ap->num_pages >= data->max_pages)) {
-		unlock_page(page);
-		fuse_io_free(ia);
-		return -EIO;
-	}
-
-	get_page(page);
-	ap->pages[ap->num_pages] = page;
-	ap->descs[ap->num_pages].length = PAGE_SIZE;
-	ap->num_pages++;
-	data->nr_pages--;
-	return 0;
-}
-
-static int fuse_readpages(struct file *file, struct address_space *mapping,
-			  struct list_head *pages, unsigned nr_pages)
-{
-	struct inode *inode = mapping->host;
-	struct fuse_conn *fc = get_fuse_conn(inode);
-	struct fuse_fill_data data;
-	int err;
-
-	err = -EIO;
 	if (is_bad_inode(inode))
-		goto out;
+		return;
 
-	data.file = file;
-	data.inode = inode;
-	data.nr_pages = nr_pages;
-	data.max_pages = min_t(unsigned int, nr_pages, fc->max_pages);
-;
-	data.ia = fuse_io_alloc(NULL, data.max_pages);
-	err = -ENOMEM;
-	if (!data.ia)
-		goto out;
+	max_pages = min_t(unsigned int, fc->max_pages,
+			fc->max_read / PAGE_SIZE);
 
-	err = read_cache_pages(mapping, pages, fuse_readpages_fill, &data);
-	if (!err) {
-		if (data.ia->ap.num_pages)
-			fuse_send_readpages(data.ia, file);
-		else
-			fuse_io_free(data.ia);
+	for (;;) {
+		struct fuse_io_args *ia;
+		struct fuse_args_pages *ap;
+
+		nr_pages = readahead_count(rac) - nr_pages;
+		if (nr_pages > max_pages)
+			nr_pages = max_pages;
+		if (nr_pages == 0)
+			break;
+		ia = fuse_io_alloc(NULL, nr_pages);
+		if (!ia)
+			return;
+		ap = &ia->ap;
+		nr_pages = __readahead_batch(rac, ap->pages, nr_pages);
+		for (i = 0; i < nr_pages; i++) {
+			fuse_wait_on_page_writeback(inode,
+						    readahead_index(rac) + i);
+			ap->descs[i].length = PAGE_SIZE;
+		}
+		ap->num_pages = nr_pages;
+		fuse_send_readpages(ia, rac->file);
 	}
-out:
-	return err;
 }
 
 static ssize_t fuse_cache_read_iter(struct kiocb *iocb, struct iov_iter *to)
@@ -1096,6 +1068,8 @@ static ssize_t fuse_send_write_pages(struct fuse_io_args *ia,
 	ia->write.in.flags = fuse_write_flags(iocb);
 
 	err = fuse_simple_request(fc, &ap->args);
+	if (!err && ia->write.out.size > count)
+		err = -EIO;
 
 	offset = ap->descs[0].offset;
 	count = ia->write.out.size;
@@ -1391,9 +1365,9 @@ static int fuse_get_user_pages(struct fuse_args_pages *ap, struct iov_iter *ii,
 	}
 
 	if (write)
-		ap->args.in_pages = 1;
+		ap->args.in_pages = true;
 	else
-		ap->args.out_pages = 1;
+		ap->args.out_pages = true;
 
 	*nbytesp = nbytes;
 
@@ -1459,6 +1433,7 @@ ssize_t fuse_direct_io(struct fuse_io_priv *io, struct iov_iter *iter,
 		}
 		ia = NULL;
 		if (nres < 0) {
+			iov_iter_revert(iter, nbytes);
 			err = nres;
 			break;
 		}
@@ -1467,8 +1442,10 @@ ssize_t fuse_direct_io(struct fuse_io_priv *io, struct iov_iter *iter,
 		count -= nres;
 		res += nres;
 		pos += nres;
-		if (nres != nbytes)
+		if (nres != nbytes) {
+			iov_iter_revert(iter, nbytes - nres);
 			break;
+		}
 		if (count) {
 			max_pages = iov_iter_npages(iter, fc->max_pages);
 			ia = fuse_io_alloc(io, max_pages);
@@ -3352,10 +3329,10 @@ static const struct file_operations fuse_file_operations = {
 
 static const struct address_space_operations fuse_file_aops  = {
 	.readpage	= fuse_readpage,
+	.readahead	= fuse_readahead,
 	.writepage	= fuse_writepage,
 	.writepages	= fuse_writepages,
 	.launder_page	= fuse_launder_page,
-	.readpages	= fuse_readpages,
 	.set_page_dirty	= __set_page_dirty_nobuffers,
 	.bmap		= fuse_bmap,
 	.direct_IO	= fuse_direct_IO,
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index d148188cfca4..ca344bf71404 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -301,6 +301,7 @@ struct fuse_io_priv {
  * FR_SENT:		request is in userspace, waiting for an answer
  * FR_FINISHED:		request is finished
  * FR_PRIVATE:		request is on private list
+ * FR_ASYNC:		request is asynchronous
  */
 enum fuse_req_flag {
 	FR_ISREPLY,
@@ -314,6 +315,7 @@ enum fuse_req_flag {
 	FR_SENT,
 	FR_FINISHED,
 	FR_PRIVATE,
+	FR_ASYNC,
 };
 
 /**
@@ -989,6 +991,8 @@ void fuse_ctl_remove_conn(struct fuse_conn *fc);
  */
 int fuse_valid_type(int m);
 
+bool fuse_invalid_attr(struct fuse_attr *attr);
+
 /**
  * Is current process allowed to perform filesystem operation?
  */
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 16aec32f7f3d..95d712d44ca1 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -448,7 +448,7 @@ enum {
 	OPT_ERR
 };
 
-static const struct fs_parameter_spec fuse_param_specs[] = {
+static const struct fs_parameter_spec fuse_fs_parameters[] = {
 	fsparam_string	("source",		OPT_SOURCE),
 	fsparam_u32	("fd",			OPT_FD),
 	fsparam_u32oct	("rootmode",		OPT_ROOTMODE),
@@ -462,68 +462,63 @@ static const struct fs_parameter_spec fuse_param_specs[] = {
 	{}
 };
 
-static const struct fs_parameter_description fuse_fs_parameters = {
-	.name		= "fuse",
-	.specs		= fuse_param_specs,
-};
-
 static int fuse_parse_param(struct fs_context *fc, struct fs_parameter *param)
 {
 	struct fs_parse_result result;
 	struct fuse_fs_context *ctx = fc->fs_private;
 	int opt;
 
-	opt = fs_parse(fc, &fuse_fs_parameters, param, &result);
+	opt = fs_parse(fc, fuse_fs_parameters, param, &result);
 	if (opt < 0)
 		return opt;
 
 	switch (opt) {
 	case OPT_SOURCE:
 		if (fc->source)
-			return invalf(fc, "fuse: Multiple sources specified");
+			return invalfc(fc, "Multiple sources specified");
 		fc->source = param->string;
 		param->string = NULL;
 		break;
 
 	case OPT_SUBTYPE:
 		if (ctx->subtype)
-			return invalf(fc, "fuse: Multiple subtypes specified");
+			return invalfc(fc, "Multiple subtypes specified");
 		ctx->subtype = param->string;
 		param->string = NULL;
 		return 0;
 
 	case OPT_FD:
 		ctx->fd = result.uint_32;
-		ctx->fd_present = 1;
+		ctx->fd_present = true;
 		break;
 
 	case OPT_ROOTMODE:
 		if (!fuse_valid_type(result.uint_32))
-			return invalf(fc, "fuse: Invalid rootmode");
+			return invalfc(fc, "Invalid rootmode");
 		ctx->rootmode = result.uint_32;
-		ctx->rootmode_present = 1;
+		ctx->rootmode_present = true;
 		break;
 
 	case OPT_USER_ID:
 		ctx->user_id = make_kuid(fc->user_ns, result.uint_32);
 		if (!uid_valid(ctx->user_id))
-			return invalf(fc, "fuse: Invalid user_id");
-		ctx->user_id_present = 1;
+			return invalfc(fc, "Invalid user_id");
+		ctx->user_id_present = true;
 		break;
 
 	case OPT_GROUP_ID:
 		ctx->group_id = make_kgid(fc->user_ns, result.uint_32);
 		if (!gid_valid(ctx->group_id))
-			return invalf(fc, "fuse: Invalid group_id");
-		ctx->group_id_present = 1;
+			return invalfc(fc, "Invalid group_id");
+		ctx->group_id_present = true;
 		break;
 
 	case OPT_DEFAULT_PERMISSIONS:
-		ctx->default_permissions = 1;
+		ctx->default_permissions = true;
 		break;
 
 	case OPT_ALLOW_OTHER:
-		ctx->allow_other = 1;
+		ctx->allow_other = true;
 		break;
 
 	case OPT_MAX_READ:
@@ -532,7 +527,7 @@ static int fuse_parse_param(struct fs_context *fc, struct fs_parameter *param)
 
 	case OPT_BLKSIZE:
 		if (!ctx->is_bdev)
-			return invalf(fc, "fuse: blksize only supported for fuseblk");
+			return invalfc(fc, "blksize only supported for fuseblk");
 		ctx->blksize = result.uint_32;
 		break;
 
@@ -997,7 +992,7 @@ void fuse_send_init(struct fuse_conn *fc)
 	/* Variable length argument used for backward compatibility
 	   with interface version < 7.5.  Rest of init_out is zeroed
 	   by do_get_request(), so a short reply is not a problem */
-	ia->args.out_argvar = 1;
+	ia->args.out_argvar = true;
 	ia->args.out_args[0].size = sizeof(ia->out);
 	ia->args.out_args[0].value = &ia->out;
 	ia->args.force = true;
@@ -1347,7 +1342,7 @@ static struct file_system_type fuse_fs_type = {
 	.name		= "fuse",
 	.fs_flags	= FS_HAS_SUBTYPE | FS_USERNS_MOUNT,
 	.init_fs_context = fuse_init_fs_context,
-	.parameters	= &fuse_fs_parameters,
+	.parameters	= fuse_fs_parameters,
 	.kill_sb	= fuse_kill_sb_anon,
 };
 MODULE_ALIAS_FS("fuse");
@@ -1363,7 +1358,7 @@ static struct file_system_type fuseblk_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "fuseblk",
 	.init_fs_context = fuse_init_fs_context,
-	.parameters	= &fuse_fs_parameters,
+	.parameters	= fuse_fs_parameters,
 	.kill_sb	= fuse_kill_sb_blk,
 	.fs_flags	= FS_REQUIRES_DEV | FS_HAS_SUBTYPE,
 };
diff --git a/fs/fuse/readdir.c b/fs/fuse/readdir.c
index 5c38b9d84c6e..90e3f01bd796 100644
--- a/fs/fuse/readdir.c
+++ b/fs/fuse/readdir.c
@@ -184,7 +184,7 @@ static int fuse_direntplus_link(struct file *file,
 
 	if (invalid_nodeid(o->nodeid))
 		return -EIO;
-	if (!fuse_valid_type(o->attr.mode))
+	if (fuse_invalid_attr(&o->attr))
 		return -EIO;
 
 	fc = get_fuse_conn(dir);
@@ -332,7 +332,7 @@ static int fuse_readdir_uncached(struct file *file, struct dir_context *ctx)
 		return -ENOMEM;
 
 	plus = fuse_use_readdirplus(inode, ctx);
-	ap->args.out_pages = 1;
+	ap->args.out_pages = true;
 	ap->num_pages = 1;
 	ap->pages = &page;
 	ap->descs = &desc;
diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c
index a5c86048b96e..bade74768903 100644
--- a/fs/fuse/virtio_fs.c
+++ b/fs/fuse/virtio_fs.c
@@ -35,6 +35,7 @@ struct virtio_fs_vq {
 	struct fuse_dev *fud;
 	bool connected;
 	long in_flight;
+	struct completion in_flight_zero; /* No inflight requests */
 	char name[24];
 } ____cacheline_aligned_in_smp;
 
@@ -48,11 +49,15 @@ struct virtio_fs {
 	unsigned int num_request_queues; /* number of request queues */
 };
 
-struct virtio_fs_forget {
+struct virtio_fs_forget_req {
 	struct fuse_in_header ih;
 	struct fuse_forget_in arg;
+};
+
+struct virtio_fs_forget {
 	/* This request can be temporarily queued on virt queue */
 	struct list_head list;
+	struct virtio_fs_forget_req req;
 };
 
 static int virtio_fs_enqueue_req(struct virtio_fs_vq *fsvq,
@@ -81,6 +86,8 @@ static inline void dec_in_flight_req(struct virtio_fs_vq *fsvq)
 {
 	WARN_ON(fsvq->in_flight <= 0);
 	fsvq->in_flight--;
+	if (!fsvq->in_flight)
+		complete(&fsvq->in_flight_zero);
 }
 
 static void release_virtio_fs_obj(struct kref *ref)
@@ -111,22 +118,23 @@ static void virtio_fs_drain_queue(struct virtio_fs_vq *fsvq)
 	WARN_ON(fsvq->in_flight < 0);
 
 	/* Wait for in flight requests to finish.*/
-	while (1) {
-		spin_lock(&fsvq->lock);
-		if (!fsvq->in_flight) {
-			spin_unlock(&fsvq->lock);
-			break;
-		}
+	spin_lock(&fsvq->lock);
+	if (fsvq->in_flight) {
+		/* We are holding virtio_fs_mutex. There should not be any
+		 * waiters waiting for completion.
+		 */
+		reinit_completion(&fsvq->in_flight_zero);
+		spin_unlock(&fsvq->lock);
+		wait_for_completion(&fsvq->in_flight_zero);
+	} else {
 		spin_unlock(&fsvq->lock);
-		/* TODO use completion instead of timeout */
-		usleep_range(1000, 2000);
 	}
 
 	flush_work(&fsvq->done_work);
 	flush_delayed_work(&fsvq->dispatch_work);
 }
 
-static void virtio_fs_drain_all_queues(struct virtio_fs *fs)
+static void virtio_fs_drain_all_queues_locked(struct virtio_fs *fs)
 {
 	struct virtio_fs_vq *fsvq;
 	int i;
@@ -137,6 +145,19 @@ static void virtio_fs_drain_all_queues(struct virtio_fs *fs)
 	}
 }
 
+static void virtio_fs_drain_all_queues(struct virtio_fs *fs)
+{
+	/* Provides mutual exclusion between ->remove and ->kill_sb
+	 * paths. We don't want both of these draining queue at the
+	 * same time. Current completion logic reinits completion
+	 * and that means there should not be any other thread
+	 * doing reinit or waiting for completion already.
+	 */
+	mutex_lock(&virtio_fs_mutex);
+	virtio_fs_drain_all_queues_locked(fs);
+	mutex_unlock(&virtio_fs_mutex);
+}
+
 static void virtio_fs_start_all_queues(struct virtio_fs *fs)
 {
 	struct virtio_fs_vq *fsvq;
@@ -313,17 +334,72 @@ static void virtio_fs_request_dispatch_work(struct work_struct *work)
 	}
 }
 
+/*
+ * Returns 1 if queue is full and sender should wait a bit before sending
+ * next request, 0 otherwise.
+ */
+static int send_forget_request(struct virtio_fs_vq *fsvq,
+			       struct virtio_fs_forget *forget,
+			       bool in_flight)
+{
+	struct scatterlist sg;
+	struct virtqueue *vq;
+	int ret = 0;
+	bool notify;
+	struct virtio_fs_forget_req *req = &forget->req;
+
+	spin_lock(&fsvq->lock);
+	if (!fsvq->connected) {
+		if (in_flight)
+			dec_in_flight_req(fsvq);
+		kfree(forget);
+		goto out;
+	}
+
+	sg_init_one(&sg, req, sizeof(*req));
+	vq = fsvq->vq;
+	dev_dbg(&vq->vdev->dev, "%s\n", __func__);
+
+	ret = virtqueue_add_outbuf(vq, &sg, 1, forget, GFP_ATOMIC);
+	if (ret < 0) {
+		if (ret == -ENOMEM || ret == -ENOSPC) {
+			pr_debug("virtio-fs: Could not queue FORGET: err=%d. Will try later\n",
+				 ret);
+			list_add_tail(&forget->list, &fsvq->queued_reqs);
+			schedule_delayed_work(&fsvq->dispatch_work,
+					      msecs_to_jiffies(1));
+			if (!in_flight)
+				inc_in_flight_req(fsvq);
+			/* Queue is full */
+			ret = 1;
+		} else {
+			pr_debug("virtio-fs: Could not queue FORGET: err=%d. Dropping it.\n",
+				 ret);
+			kfree(forget);
+			if (in_flight)
+				dec_in_flight_req(fsvq);
+		}
+		goto out;
+	}
+
+	if (!in_flight)
+		inc_in_flight_req(fsvq);
+	notify = virtqueue_kick_prepare(vq);
+	spin_unlock(&fsvq->lock);
+
+	if (notify)
+		virtqueue_notify(vq);
+	return ret;
+out:
+	spin_unlock(&fsvq->lock);
+	return ret;
+}
+
 static void virtio_fs_hiprio_dispatch_work(struct work_struct *work)
 {
 	struct virtio_fs_forget *forget;
 	struct virtio_fs_vq *fsvq = container_of(work, struct virtio_fs_vq,
 						 dispatch_work.work);
-	struct virtqueue *vq = fsvq->vq;
-	struct scatterlist sg;
-	struct scatterlist *sgs[] = {&sg};
-	bool notify;
-	int ret;
-
 	pr_debug("virtio-fs: worker %s called.\n", __func__);
 	while (1) {
 		spin_lock(&fsvq->lock);
@@ -335,43 +411,9 @@ static void virtio_fs_hiprio_dispatch_work(struct work_struct *work)
 		}
 
 		list_del(&forget->list);
-		if (!fsvq->connected) {
-			dec_in_flight_req(fsvq);
-			spin_unlock(&fsvq->lock);
-			kfree(forget);
-			continue;
-		}
-
-		sg_init_one(&sg, forget, sizeof(*forget));
-
-		/* Enqueue the request */
-		dev_dbg(&vq->vdev->dev, "%s\n", __func__);
-		ret = virtqueue_add_sgs(vq, sgs, 1, 0, forget, GFP_ATOMIC);
-		if (ret < 0) {
-			if (ret == -ENOMEM || ret == -ENOSPC) {
-				pr_debug("virtio-fs: Could not queue FORGET: err=%d. Will try later\n",
-					 ret);
-				list_add_tail(&forget->list,
-						&fsvq->queued_reqs);
-				schedule_delayed_work(&fsvq->dispatch_work,
-						msecs_to_jiffies(1));
-			} else {
-				pr_debug("virtio-fs: Could not queue FORGET: err=%d. Dropping it.\n",
-					 ret);
-				dec_in_flight_req(fsvq);
-				kfree(forget);
-			}
-			spin_unlock(&fsvq->lock);
-			return;
-		}
-
-		notify = virtqueue_kick_prepare(vq);
 		spin_unlock(&fsvq->lock);
-
-		if (notify)
-			virtqueue_notify(vq);
-		pr_debug("virtio-fs: worker %s dispatched one forget request.\n",
-			 __func__);
+		if (send_forget_request(fsvq, forget, true))
+			return;
 	}
 }
 
@@ -556,6 +598,7 @@ static int virtio_fs_setup_vqs(struct virtio_device *vdev,
 	INIT_LIST_HEAD(&fs->vqs[VQ_HIPRIO].end_reqs);
 	INIT_DELAYED_WORK(&fs->vqs[VQ_HIPRIO].dispatch_work,
 			virtio_fs_hiprio_dispatch_work);
+	init_completion(&fs->vqs[VQ_HIPRIO].in_flight_zero);
 	spin_lock_init(&fs->vqs[VQ_HIPRIO].lock);
 
 	/* Initialize the requests virtqueues */
@@ -566,6 +609,7 @@ static int virtio_fs_setup_vqs(struct virtio_device *vdev,
 				  virtio_fs_request_dispatch_work);
 		INIT_LIST_HEAD(&fs->vqs[i].queued_reqs);
 		INIT_LIST_HEAD(&fs->vqs[i].end_reqs);
+		init_completion(&fs->vqs[i].in_flight_zero);
 		snprintf(fs->vqs[i].name, sizeof(fs->vqs[i].name),
 			 "requests.%u", i - VQ_REQUEST);
 		callbacks[i] = virtio_fs_vq_done;
@@ -659,7 +703,7 @@ static void virtio_fs_remove(struct virtio_device *vdev)
 	/* This device is going away. No one should get new reference */
 	list_del_init(&fs->list);
 	virtio_fs_stop_all_queues(fs);
-	virtio_fs_drain_all_queues(fs);
+	virtio_fs_drain_all_queues_locked(fs);
 	vdev->config->reset(vdev);
 	virtio_fs_cleanup_vqs(vdev, fs);
 
@@ -684,12 +728,12 @@ static int virtio_fs_restore(struct virtio_device *vdev)
 }
 #endif /* CONFIG_PM_SLEEP */
 
-const static struct virtio_device_id id_table[] = {
+static const struct virtio_device_id id_table[] = {
 	{ VIRTIO_ID_FS, VIRTIO_DEV_ANY_ID },
 	{},
 };
 
-const static unsigned int feature_table[] = {};
+static const unsigned int feature_table[] = {};
 
 static struct virtio_driver virtio_fs_driver = {
 	.driver.name		= KBUILD_MODNAME,
@@ -710,14 +754,10 @@ __releases(fiq->lock)
 {
 	struct fuse_forget_link *link;
 	struct virtio_fs_forget *forget;
-	struct scatterlist sg;
-	struct scatterlist *sgs[] = {&sg};
+	struct virtio_fs_forget_req *req;
 	struct virtio_fs *fs;
-	struct virtqueue *vq;
 	struct virtio_fs_vq *fsvq;
-	bool notify;
 	u64 unique;
-	int ret;
 
 	link = fuse_dequeue_forget(fiq, 1, NULL);
 	unique = fuse_get_unique(fiq);
@@ -728,57 +768,19 @@ __releases(fiq->lock)
 
 	/* Allocate a buffer for the request */
 	forget = kmalloc(sizeof(*forget), GFP_NOFS | __GFP_NOFAIL);
+	req = &forget->req;
 
-	forget->ih = (struct fuse_in_header){
+	req->ih = (struct fuse_in_header){
 		.opcode = FUSE_FORGET,
 		.nodeid = link->forget_one.nodeid,
 		.unique = unique,
-		.len = sizeof(*forget),
+		.len = sizeof(*req),
 	};
-	forget->arg = (struct fuse_forget_in){
+	req->arg = (struct fuse_forget_in){
 		.nlookup = link->forget_one.nlookup,
 	};
 
-	sg_init_one(&sg, forget, sizeof(*forget));
-
-	/* Enqueue the request */
-	spin_lock(&fsvq->lock);
-
-	if (!fsvq->connected) {
-		kfree(forget);
-		spin_unlock(&fsvq->lock);
-		goto out;
-	}
-
-	vq = fsvq->vq;
-	dev_dbg(&vq->vdev->dev, "%s\n", __func__);
-
-	ret = virtqueue_add_sgs(vq, sgs, 1, 0, forget, GFP_ATOMIC);
-	if (ret < 0) {
-		if (ret == -ENOMEM || ret == -ENOSPC) {
-			pr_debug("virtio-fs: Could not queue FORGET: err=%d. Will try later.\n",
-				 ret);
-			list_add_tail(&forget->list, &fsvq->queued_reqs);
-			schedule_delayed_work(&fsvq->dispatch_work,
-					msecs_to_jiffies(1));
-			inc_in_flight_req(fsvq);
-		} else {
-			pr_debug("virtio-fs: Could not queue FORGET: err=%d. Dropping it.\n",
-				 ret);
-			kfree(forget);
-		}
-		spin_unlock(&fsvq->lock);
-		goto out;
-	}
-
-	inc_in_flight_req(fsvq);
-	notify = virtqueue_kick_prepare(vq);
-
-	spin_unlock(&fsvq->lock);
-
-	if (notify)
-		virtqueue_notify(vq);
-out:
+	send_forget_request(fsvq, forget, false);
 	kfree(link);
 }
 
@@ -1026,7 +1028,7 @@ __releases(fiq->lock)
 	}
 }
 
-const static struct fuse_iqueue_ops virtio_fs_fiq_ops = {
+static const struct fuse_iqueue_ops virtio_fs_fiq_ops = {
 	.wake_forget_and_unlock		= virtio_fs_wake_forget_and_unlock,
 	.wake_interrupt_and_unlock	= virtio_fs_wake_interrupt_and_unlock,
 	.wake_pending_and_unlock	= virtio_fs_wake_pending_and_unlock,
diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c
index 09e6be8aa036..2e939f5fe751 100644
--- a/fs/gfs2/acl.c
+++ b/fs/gfs2/acl.c
@@ -21,6 +21,7 @@
 #include "glock.h"
 #include "inode.h"
 #include "meta_io.h"
+#include "quota.h"
 #include "rgrp.h"
 #include "trans.h"
 #include "util.h"
@@ -116,14 +117,14 @@ int gfs2_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 	if (acl && acl->a_count > GFS2_ACL_MAX_ENTRIES(GFS2_SB(inode)))
 		return -E2BIG;
 
-	ret = gfs2_rsqa_alloc(ip);
+	ret = gfs2_qa_get(ip);
 	if (ret)
 		return ret;
 
 	if (!gfs2_glock_is_locked_by_me(ip->i_gl)) {
 		ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
 		if (ret)
-			return ret;
+			goto out;
 		need_unlock = true;
 	}
 
@@ -143,5 +144,7 @@ int gfs2_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 unlock:
 	if (need_unlock)
 		gfs2_glock_dq_uninit(&gh);
+out:
+	gfs2_qa_put(ip);
 	return ret;
 }
diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index b9fe975d7625..72c9560f4467 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -133,7 +133,7 @@ static int gfs2_write_full_page(struct page *page, get_block_t *get_block,
 	 * the  page size, the remaining memory is zeroed when mapped, and
 	 * writes to that region are not written out to the file."
 	 */
-	offset = i_size & (PAGE_SIZE-1);
+	offset = i_size & (PAGE_SIZE - 1);
 	if (page->index == end_index && offset)
 		zero_user_segment(page, offset, PAGE_SIZE);
 
@@ -183,14 +183,12 @@ static int gfs2_jdata_writepage(struct page *page, struct writeback_control *wbc
 	struct inode *inode = page->mapping->host;
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
-	int ret;
 
 	if (gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(ip->i_gl)))
 		goto out;
 	if (PageChecked(page) || current->journal_info)
 		goto out_ignore;
-	ret = __gfs2_jdata_writepage(page, wbc);
-	return ret;
+	return __gfs2_jdata_writepage(page, wbc);
 
 out_ignore:
 	redirty_page_for_writepage(wbc, page);
@@ -497,7 +495,7 @@ static int __gfs2_readpage(void *file, struct page *page)
 		error = mpage_readpage(page, gfs2_block_map);
 	}
 
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags)))
+	if (unlikely(gfs2_withdrawn(sdp)))
 		return -EIO;
 
 	return error;
@@ -579,7 +577,7 @@ int gfs2_internal_read(struct gfs2_inode *ip, char *buf, loff_t *pos,
 }
 
 /**
- * gfs2_readpages - Read a bunch of pages at once
+ * gfs2_readahead - Read a bunch of pages at once
  * @file: The file to read from
  * @mapping: Address space info
  * @pages: List of pages to read
@@ -592,31 +590,24 @@ int gfs2_internal_read(struct gfs2_inode *ip, char *buf, loff_t *pos,
  *    obviously not something we'd want to do on too regular a basis.
  *    Any I/O we ignore at this time will be done via readpage later.
  * 2. We don't handle stuffed files here we let readpage do the honours.
- * 3. mpage_readpages() does most of the heavy lifting in the common case.
+ * 3. mpage_readahead() does most of the heavy lifting in the common case.
  * 4. gfs2_block_map() is relied upon to set BH_Boundary in the right places.
  */
 
-static int gfs2_readpages(struct file *file, struct address_space *mapping,
-			  struct list_head *pages, unsigned nr_pages)
+static void gfs2_readahead(struct readahead_control *rac)
 {
-	struct inode *inode = mapping->host;
+	struct inode *inode = rac->mapping->host;
 	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct gfs2_holder gh;
-	int ret;
 
 	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
-	ret = gfs2_glock_nq(&gh);
-	if (unlikely(ret))
+	if (gfs2_glock_nq(&gh))
 		goto out_uninit;
 	if (!gfs2_is_stuffed(ip))
-		ret = mpage_readpages(mapping, pages, nr_pages, gfs2_block_map);
+		mpage_readahead(rac, gfs2_block_map);
 	gfs2_glock_dq(&gh);
 out_uninit:
 	gfs2_holder_uninit(&gh);
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags)))
-		ret = -EIO;
-	return ret;
 }
 
 /**
@@ -807,11 +798,16 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask)
 		bd = bh->b_private;
 		if (bd) {
 			gfs2_assert_warn(sdp, bd->bd_bh == bh);
-			if (!list_empty(&bd->bd_list))
-				list_del_init(&bd->bd_list);
 			bd->bd_bh = NULL;
 			bh->b_private = NULL;
-			kmem_cache_free(gfs2_bufdata_cachep, bd);
+			/*
+			 * The bd may still be queued as a revoke, in which
+			 * case we must not dequeue nor free it.
+			 */
+			if (!bd->bd_blkno && !list_empty(&bd->bd_list))
+				list_del_init(&bd->bd_list);
+			if (list_empty(&bd->bd_list))
+				kmem_cache_free(gfs2_bufdata_cachep, bd);
 		}
 
 		bh = bh->b_this_page;
@@ -830,7 +826,7 @@ static const struct address_space_operations gfs2_aops = {
 	.writepage = gfs2_writepage,
 	.writepages = gfs2_writepages,
 	.readpage = gfs2_readpage,
-	.readpages = gfs2_readpages,
+	.readahead = gfs2_readahead,
 	.bmap = gfs2_bmap,
 	.invalidatepage = gfs2_invalidatepage,
 	.releasepage = gfs2_releasepage,
@@ -844,7 +840,7 @@ static const struct address_space_operations gfs2_jdata_aops = {
 	.writepage = gfs2_jdata_writepage,
 	.writepages = gfs2_jdata_writepages,
 	.readpage = gfs2_readpage,
-	.readpages = gfs2_readpages,
+	.readahead = gfs2_readahead,
 	.set_page_dirty = jdata_set_page_dirty,
 	.bmap = gfs2_bmap,
 	.invalidatepage = gfs2_invalidatepage,
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index f63df54a08c6..6306eaae378b 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -528,10 +528,12 @@ lower_metapath:
 
 		/* Advance in metadata tree. */
 		(mp->mp_list[hgt])++;
-		if (mp->mp_list[hgt] >= sdp->sd_inptrs) {
-			if (!hgt)
+		if (hgt) {
+			if (mp->mp_list[hgt] >= sdp->sd_inptrs)
+				goto lower_metapath;
+		} else {
+			if (mp->mp_list[hgt] >= sdp->sd_diptrs)
 				break;
-			goto lower_metapath;
 		}
 
 fill_up_metapath:
@@ -876,10 +878,9 @@ static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length,
 					ret = -ENOENT;
 					goto unlock;
 				} else {
-					/* report a hole */
 					iomap->offset = pos;
 					iomap->length = length;
-					goto do_alloc;
+					goto hole_found;
 				}
 			}
 			iomap->length = size;
@@ -933,8 +934,6 @@ unlock:
 	return ret;
 
 do_alloc:
-	iomap->addr = IOMAP_NULL_ADDR;
-	iomap->type = IOMAP_HOLE;
 	if (flags & IOMAP_REPORT) {
 		if (pos >= size)
 			ret = -ENOENT;
@@ -956,6 +955,9 @@ do_alloc:
 		if (pos < size && height == ip->i_height)
 			ret = gfs2_hole_size(inode, lblock, len, mp, iomap);
 	}
+hole_found:
+	iomap->addr = IOMAP_NULL_ADDR;
+	iomap->type = IOMAP_HOLE;
 	goto out;
 }
 
@@ -1149,7 +1151,8 @@ static inline bool gfs2_iomap_need_write_lock(unsigned flags)
 }
 
 static int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length,
-			    unsigned flags, struct iomap *iomap)
+			    unsigned flags, struct iomap *iomap,
+			    struct iomap *srcmap)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct metapath mp = { .mp_aheight = 1, };
@@ -2182,7 +2185,7 @@ int gfs2_setattr_size(struct inode *inode, u64 newsize)
 
 	inode_dio_wait(inode);
 
-	ret = gfs2_rsqa_alloc(ip);
+	ret = gfs2_qa_get(ip);
 	if (ret)
 		goto out;
 
@@ -2193,7 +2196,8 @@ int gfs2_setattr_size(struct inode *inode, u64 newsize)
 
 	ret = do_shrink(inode, newsize);
 out:
-	gfs2_rsqa_delete(ip, NULL);
+	gfs2_rs_delete(ip, NULL);
+	gfs2_qa_put(ip);
 	return ret;
 }
 
@@ -2222,7 +2226,7 @@ void gfs2_free_journal_extents(struct gfs2_jdesc *jd)
 	struct gfs2_journal_extent *jext;
 
 	while(!list_empty(&jd->extent_list)) {
-		jext = list_entry(jd->extent_list.next, struct gfs2_journal_extent, list);
+		jext = list_first_entry(&jd->extent_list, struct gfs2_journal_extent, list);
 		list_del(&jext->list);
 		kfree(jext);
 	}
@@ -2243,7 +2247,7 @@ static int gfs2_add_jextent(struct gfs2_jdesc *jd, u64 lblock, u64 dblock, u64 b
 	struct gfs2_journal_extent *jext;
 
 	if (!list_empty(&jd->extent_list)) {
-		jext = list_entry(jd->extent_list.prev, struct gfs2_journal_extent, list);
+		jext = list_last_entry(&jd->extent_list, struct gfs2_journal_extent, list);
 		if ((jext->dblock + jext->blocks) == dblock) {
 			jext->blocks += blocks;
 			return 0;
@@ -2440,8 +2444,16 @@ int __gfs2_punch_hole(struct file *file, loff_t offset, loff_t length)
 	struct inode *inode = file_inode(file);
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
+	unsigned int blocksize = i_blocksize(inode);
+	loff_t start, end;
 	int error;
 
+	start = round_down(offset, blocksize);
+	end = round_up(offset + length, blocksize) - 1;
+	error = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (error)
+		return error;
+
 	if (gfs2_is_jdata(ip))
 		error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_JDATA,
 					 GFS2_JTRUNC_REVOKES);
@@ -2455,9 +2467,8 @@ int __gfs2_punch_hole(struct file *file, loff_t offset, loff_t length)
 		if (error)
 			goto out;
 	} else {
-		unsigned int start_off, end_len, blocksize;
+		unsigned int start_off, end_len;
 
-		blocksize = i_blocksize(inode);
 		start_off = offset & (blocksize - 1);
 		end_len = (offset + length) & (blocksize - 1);
 		if (start_off) {
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index eb9c0578978f..c0f2875c946c 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -73,9 +73,6 @@
 #include "bmap.h"
 #include "util.h"
 
-#define IS_LEAF     1 /* Hashed (leaf) directory */
-#define IS_DINODE   2 /* Linear (stuffed dinode block) directory */
-
 #define MAX_RA_BLOCKS 32 /* max read-ahead blocks */
 
 #define gfs2_disk_hash2offset(h) (((u64)(h)) >> 1)
@@ -357,7 +354,7 @@ static __be64 *gfs2_dir_get_hash_table(struct gfs2_inode *ip)
 
 	hc = kmalloc(hsize, GFP_NOFS | __GFP_NOWARN);
 	if (hc == NULL)
-		hc = __vmalloc(hsize, GFP_NOFS, PAGE_KERNEL);
+		hc = __vmalloc(hsize, GFP_NOFS);
 
 	if (hc == NULL)
 		return ERR_PTR(-ENOMEM);
@@ -1169,7 +1166,7 @@ static int dir_double_exhash(struct gfs2_inode *dip)
 
 	hc2 = kmalloc_array(hsize_bytes, 2, GFP_NOFS | __GFP_NOWARN);
 	if (hc2 == NULL)
-		hc2 = __vmalloc(hsize_bytes * 2, GFP_NOFS, PAGE_KERNEL);
+		hc2 = __vmalloc(hsize_bytes * 2, GFP_NOFS);
 
 	if (!hc2)
 		return -ENOMEM;
@@ -1330,7 +1327,7 @@ static void *gfs2_alloc_sort_buffer(unsigned size)
 	if (size < KMALLOC_MAX_SIZE)
 		ptr = kmalloc(size, GFP_NOFS | __GFP_NOWARN);
 	if (!ptr)
-		ptr = __vmalloc(size, GFP_NOFS, PAGE_KERNEL);
+		ptr = __vmalloc(size, GFP_NOFS);
 	return ptr;
 }
 
@@ -1990,8 +1987,7 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
 
 	ht = kzalloc(size, GFP_NOFS | __GFP_NOWARN);
 	if (ht == NULL)
-		ht = __vmalloc(size, GFP_NOFS | __GFP_NOWARN | __GFP_ZERO,
-			       PAGE_KERNEL);
+		ht = __vmalloc(size, GFP_NOFS | __GFP_NOWARN | __GFP_ZERO);
 	if (!ht)
 		return -ENOMEM;
 
@@ -2031,7 +2027,8 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
 
 	error = gfs2_trans_begin(sdp,
 			rg_blocks + (DIV_ROUND_UP(size, sdp->sd_jbsize) + 1) +
-			RES_DINODE + RES_STATFS + RES_QUOTA, l_blocks);
+			RES_DINODE + RES_STATFS + RES_QUOTA, RES_DINODE +
+				 l_blocks);
 	if (error)
 		goto out_rg_gunlock;
 
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 997b326247e2..fe305e4bfd37 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -6,6 +6,7 @@
 
 #include <linux/slab.h>
 #include <linux/spinlock.h>
+#include <linux/compat.h>
 #include <linux/completion.h>
 #include <linux/buffer_head.h>
 #include <linux/pagemap.h>
@@ -354,6 +355,31 @@ static long gfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	return -ENOTTY;
 }
 
+#ifdef CONFIG_COMPAT
+static long gfs2_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
+{
+	switch(cmd) {
+	/* These are just misnamed, they actually get/put from/to user an int */
+	case FS_IOC32_GETFLAGS:
+		cmd = FS_IOC_GETFLAGS;
+		break;
+	case FS_IOC32_SETFLAGS:
+		cmd = FS_IOC_SETFLAGS;
+		break;
+	/* Keep this list in sync with gfs2_ioctl */
+	case FITRIM:
+	case FS_IOC_GETFSLABEL:
+		break;
+	default:
+		return -ENOIOCTLCMD;
+	}
+
+	return gfs2_ioctl(filp, cmd, (unsigned long)compat_ptr(arg));
+}
+#else
+#define gfs2_compat_ioctl NULL
+#endif
+
 /**
  * gfs2_size_hint - Give a hint to the size of a write request
  * @filep: The struct file
@@ -381,27 +407,28 @@ static void gfs2_size_hint(struct file *filep, loff_t offset, size_t size)
 /**
  * gfs2_allocate_page_backing - Allocate blocks for a write fault
  * @page: The (locked) page to allocate backing for
+ * @length: Size of the allocation
  *
  * We try to allocate all the blocks required for the page in one go.  This
  * might fail for various reasons, so we keep trying until all the blocks to
  * back this page are allocated.  If some of the blocks are already allocated,
  * that is ok too.
  */
-static int gfs2_allocate_page_backing(struct page *page)
+static int gfs2_allocate_page_backing(struct page *page, unsigned int length)
 {
 	u64 pos = page_offset(page);
-	u64 size = PAGE_SIZE;
 
 	do {
 		struct iomap iomap = { };
 
-		if (gfs2_iomap_get_alloc(page->mapping->host, pos, 1, &iomap))
+		if (gfs2_iomap_get_alloc(page->mapping->host, pos, length, &iomap))
 			return -EIO;
 
-		iomap.length = min(iomap.length, size);
-		size -= iomap.length;
+		if (length < iomap.length)
+			iomap.length = length;
+		length -= iomap.length;
 		pos += iomap.length;
-	} while (size > 0);
+	} while (length > 0);
 
 	return 0;
 }
@@ -422,33 +449,48 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf)
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct gfs2_alloc_parms ap = { .aflags = 0, };
-	unsigned long last_index;
-	u64 pos = page_offset(page);
+	u64 offset = page_offset(page);
 	unsigned int data_blocks, ind_blocks, rblocks;
 	struct gfs2_holder gh;
+	unsigned int length;
 	loff_t size;
 	int ret;
 
 	sb_start_pagefault(inode->i_sb);
 
-	ret = gfs2_rsqa_alloc(ip);
-	if (ret)
-		goto out;
-
-	gfs2_size_hint(vmf->vma->vm_file, pos, PAGE_SIZE);
-
 	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
 	ret = gfs2_glock_nq(&gh);
 	if (ret)
 		goto out_uninit;
 
+	/* Check page index against inode size */
+	size = i_size_read(inode);
+	if (offset >= size) {
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
 	/* Update file times before taking page lock */
 	file_update_time(vmf->vma->vm_file);
 
+	/* page is wholly or partially inside EOF */
+	if (offset > size - PAGE_SIZE)
+		length = offset_in_page(size);
+	else
+		length = PAGE_SIZE;
+
+	gfs2_size_hint(vmf->vma->vm_file, offset, length);
+
 	set_bit(GLF_DIRTY, &ip->i_gl->gl_flags);
 	set_bit(GIF_SW_PAGED, &ip->i_flags);
 
-	if (!gfs2_write_alloc_required(ip, pos, PAGE_SIZE)) {
+	/*
+	 * iomap_writepage / iomap_writepages currently don't support inline
+	 * files, so always unstuff here.
+	 */
+
+	if (!gfs2_is_stuffed(ip) &&
+	    !gfs2_write_alloc_required(ip, offset, length)) {
 		lock_page(page);
 		if (!PageUptodate(page) || page->mapping != inode->i_mapping) {
 			ret = -EAGAIN;
@@ -461,7 +503,7 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf)
 	if (ret)
 		goto out_unlock;
 
-	gfs2_write_calc_reserv(ip, PAGE_SIZE, &data_blocks, &ind_blocks);
+	gfs2_write_calc_reserv(ip, length, &data_blocks, &ind_blocks);
 	ap.target = data_blocks + ind_blocks;
 	ret = gfs2_quota_lock_check(ip, &ap);
 	if (ret)
@@ -482,13 +524,6 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf)
 		goto out_trans_fail;
 
 	lock_page(page);
-	ret = -EINVAL;
-	size = i_size_read(inode);
-	last_index = (size - 1) >> PAGE_SHIFT;
-	/* Check page index against inode size */
-	if (size == 0 || (page->index > last_index))
-		goto out_trans_end;
-
 	ret = -EAGAIN;
 	/* If truncated, we must retry the operation, we may have raced
 	 * with the glock demotion code.
@@ -501,7 +536,7 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf)
 	if (gfs2_is_stuffed(ip))
 		ret = gfs2_unstuff_dinode(ip, page);
 	if (ret == 0)
-		ret = gfs2_allocate_page_backing(page);
+		ret = gfs2_allocate_page_backing(page, length);
 
 out_trans_end:
 	if (ret)
@@ -519,7 +554,6 @@ out_uninit:
 		set_page_dirty(page);
 		wait_for_stable_page(page);
 	}
-out:
 	sb_end_pagefault(inode->i_sb);
 	return block_page_mkwrite_return(ret);
 }
@@ -596,7 +630,17 @@ int gfs2_open_common(struct inode *inode, struct file *file)
 
 	gfs2_assert_warn(GFS2_SB(inode), !file->private_data);
 	file->private_data = fp;
+	if (file->f_mode & FMODE_WRITE) {
+		ret = gfs2_qa_get(GFS2_I(inode));
+		if (ret)
+			goto fail;
+	}
 	return 0;
+
+fail:
+	kfree(file->private_data);
+	file->private_data = NULL;
+	return ret;
 }
 
 /**
@@ -651,10 +695,10 @@ static int gfs2_release(struct inode *inode, struct file *file)
 	kfree(file->private_data);
 	file->private_data = NULL;
 
-	if (!(file->f_mode & FMODE_WRITE))
-		return 0;
-
-	gfs2_rsqa_delete(ip, &inode->i_writecount);
+	if (file->f_mode & FMODE_WRITE) {
+		gfs2_rs_delete(ip, &inode->i_writecount);
+		gfs2_qa_put(ip);
+	}
 	return 0;
 }
 
@@ -732,7 +776,8 @@ static ssize_t gfs2_file_direct_read(struct kiocb *iocb, struct iov_iter *to)
 	if (ret)
 		goto out_uninit;
 
-	ret = iomap_dio_rw(iocb, to, &gfs2_iomap_ops, NULL);
+	ret = iomap_dio_rw(iocb, to, &gfs2_iomap_ops, NULL,
+			   is_sync_kiocb(iocb));
 
 	gfs2_glock_dq(&gh);
 out_uninit:
@@ -767,7 +812,8 @@ static ssize_t gfs2_file_direct_write(struct kiocb *iocb, struct iov_iter *from)
 	if (offset + len > i_size_read(&ip->i_inode))
 		goto out;
 
-	ret = iomap_dio_rw(iocb, from, &gfs2_iomap_ops, NULL);
+	ret = iomap_dio_rw(iocb, from, &gfs2_iomap_ops, NULL,
+			   is_sync_kiocb(iocb));
 
 out:
 	gfs2_glock_dq(&gh);
@@ -806,11 +852,7 @@ static ssize_t gfs2_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
 	struct gfs2_inode *ip = GFS2_I(inode);
-	ssize_t written = 0, ret;
-
-	ret = gfs2_rsqa_alloc(ip);
-	if (ret)
-		return ret;
+	ssize_t ret;
 
 	gfs2_size_hint(file, iocb->ki_pos, iov_iter_count(from));
 
@@ -826,68 +868,58 @@ static ssize_t gfs2_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	inode_lock(inode);
 	ret = generic_write_checks(iocb, from);
 	if (ret <= 0)
-		goto out;
-
-	/* We can write back this queue in page reclaim */
-	current->backing_dev_info = inode_to_bdi(inode);
+		goto out_unlock;
 
 	ret = file_remove_privs(file);
 	if (ret)
-		goto out2;
+		goto out_unlock;
 
 	ret = file_update_time(file);
 	if (ret)
-		goto out2;
+		goto out_unlock;
 
 	if (iocb->ki_flags & IOCB_DIRECT) {
 		struct address_space *mapping = file->f_mapping;
-		loff_t pos, endbyte;
-		ssize_t buffered;
+		ssize_t buffered, ret2;
 
-		written = gfs2_file_direct_write(iocb, from);
-		if (written < 0 || !iov_iter_count(from))
-			goto out2;
+		ret = gfs2_file_direct_write(iocb, from);
+		if (ret < 0 || !iov_iter_count(from))
+			goto out_unlock;
 
-		ret = iomap_file_buffered_write(iocb, from, &gfs2_iomap_ops);
-		if (unlikely(ret < 0))
-			goto out2;
-		buffered = ret;
+		iocb->ki_flags |= IOCB_DSYNC;
+		current->backing_dev_info = inode_to_bdi(inode);
+		buffered = iomap_file_buffered_write(iocb, from, &gfs2_iomap_ops);
+		current->backing_dev_info = NULL;
+		if (unlikely(buffered <= 0))
+			goto out_unlock;
 
 		/*
 		 * We need to ensure that the page cache pages are written to
 		 * disk and invalidated to preserve the expected O_DIRECT
-		 * semantics.
+		 * semantics.  If the writeback or invalidate fails, only report
+		 * the direct I/O range as we don't know if the buffered pages
+		 * made it to disk.
 		 */
-		pos = iocb->ki_pos;
-		endbyte = pos + buffered - 1;
-		ret = filemap_write_and_wait_range(mapping, pos, endbyte);
-		if (!ret) {
-			iocb->ki_pos += buffered;
-			written += buffered;
-			invalidate_mapping_pages(mapping,
-						 pos >> PAGE_SHIFT,
-						 endbyte >> PAGE_SHIFT);
-		} else {
-			/*
-			 * We don't know how much we wrote, so just return
-			 * the number of bytes which were direct-written
-			 */
-		}
+		iocb->ki_pos += buffered;
+		ret2 = generic_write_sync(iocb, buffered);
+		invalidate_mapping_pages(mapping,
+				(iocb->ki_pos - buffered) >> PAGE_SHIFT,
+				(iocb->ki_pos - 1) >> PAGE_SHIFT);
+		if (!ret || ret2 > 0)
+			ret += ret2;
 	} else {
+		current->backing_dev_info = inode_to_bdi(inode);
 		ret = iomap_file_buffered_write(iocb, from, &gfs2_iomap_ops);
-		if (likely(ret > 0))
+		current->backing_dev_info = NULL;
+		if (likely(ret > 0)) {
 			iocb->ki_pos += ret;
+			ret = generic_write_sync(iocb, ret);
+		}
 	}
 
-out2:
-	current->backing_dev_info = NULL;
-out:
+out_unlock:
 	inode_unlock(inode);
-	if (likely(ret > 0)) {
-		/* Handle various SYNC-type writes */
-		ret = generic_write_sync(iocb, ret);
-	}
-	return written ? written : ret;
+	return ret;
 }
 
 static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len,
@@ -933,6 +965,7 @@ out:
 	brelse(dibh);
 	return error;
 }
+
 /**
  * calc_max_reserv() - Reverse of write_calc_reserv. Given a number of
  *                     blocks, determine how many bytes can be written.
@@ -1117,17 +1150,11 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t le
 	if (mode & FALLOC_FL_PUNCH_HOLE) {
 		ret = __gfs2_punch_hole(file, offset, len);
 	} else {
-		ret = gfs2_rsqa_alloc(ip);
-		if (ret)
-			goto out_putw;
-
 		ret = __gfs2_fallocate(file, mode, offset, len);
-
 		if (ret)
 			gfs2_rs_deltree(&ip->i_res);
 	}
 
-out_putw:
 	put_write_access(inode);
 out_unlock:
 	gfs2_glock_dq(&gh);
@@ -1141,16 +1168,12 @@ static ssize_t gfs2_file_splice_write(struct pipe_inode_info *pipe,
 				      struct file *out, loff_t *ppos,
 				      size_t len, unsigned int flags)
 {
-	int error;
-	struct gfs2_inode *ip = GFS2_I(out->f_mapping->host);
-
-	error = gfs2_rsqa_alloc(ip);
-	if (error)
-		return (ssize_t)error;
+	ssize_t ret;
 
 	gfs2_size_hint(out, *ppos, len);
 
-	return iter_file_splice_write(pipe, out, ppos, len, flags);
+	ret = iter_file_splice_write(pipe, out, ppos, len, flags);
+	return ret;
 }
 
 #ifdef CONFIG_GFS2_FS_LOCKING_DLM
@@ -1180,7 +1203,7 @@ static int gfs2_lock(struct file *file, int cmd, struct file_lock *fl)
 		cmd = F_SETLK;
 		fl->fl_type = F_UNLCK;
 	}
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags))) {
+	if (unlikely(gfs2_withdrawn(sdp))) {
 		if (fl->fl_type == F_UNLCK)
 			locks_lock_file_wait(file, fl);
 		return -EIO;
@@ -1293,6 +1316,7 @@ const struct file_operations gfs2_file_fops = {
 	.write_iter	= gfs2_file_write_iter,
 	.iopoll		= iomap_dio_iopoll,
 	.unlocked_ioctl	= gfs2_ioctl,
+	.compat_ioctl	= gfs2_compat_ioctl,
 	.mmap		= gfs2_mmap,
 	.open		= gfs2_open,
 	.release	= gfs2_release,
@@ -1308,6 +1332,7 @@ const struct file_operations gfs2_file_fops = {
 const struct file_operations gfs2_dir_fops = {
 	.iterate_shared	= gfs2_readdir,
 	.unlocked_ioctl	= gfs2_ioctl,
+	.compat_ioctl	= gfs2_compat_ioctl,
 	.open		= gfs2_open,
 	.release	= gfs2_release,
 	.fsync		= gfs2_fsync,
@@ -1324,6 +1349,7 @@ const struct file_operations gfs2_file_fops_nolock = {
 	.write_iter	= gfs2_file_write_iter,
 	.iopoll		= iomap_dio_iopoll,
 	.unlocked_ioctl	= gfs2_ioctl,
+	.compat_ioctl	= gfs2_compat_ioctl,
 	.mmap		= gfs2_mmap,
 	.open		= gfs2_open,
 	.release	= gfs2_release,
@@ -1337,6 +1363,7 @@ const struct file_operations gfs2_file_fops_nolock = {
 const struct file_operations gfs2_dir_fops_nolock = {
 	.iterate_shared	= gfs2_readdir,
 	.unlocked_ioctl	= gfs2_ioctl,
+	.compat_ioctl	= gfs2_compat_ioctl,
 	.open		= gfs2_open,
 	.release	= gfs2_release,
 	.fsync		= gfs2_fsync,
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 0290a22ebccf..bf70e3b14938 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -133,6 +133,33 @@ static void gfs2_glock_dealloc(struct rcu_head *rcu)
 	}
 }
 
+/**
+ * glock_blocked_by_withdraw - determine if we can still use a glock
+ * @gl: the glock
+ *
+ * We need to allow some glocks to be enqueued, dequeued, promoted, and demoted
+ * when we're withdrawn. For example, to maintain metadata integrity, we should
+ * disallow the use of inode and rgrp glocks when withdrawn. Other glocks, like
+ * iopen or the transaction glocks may be safely used because none of their
+ * metadata goes through the journal. So in general, we should disallow all
+ * glocks that are journaled, and allow all the others. One exception is:
+ * we need to allow our active journal to be promoted and demoted so others
+ * may recover it and we can reacquire it when they're done.
+ */
+static bool glock_blocked_by_withdraw(struct gfs2_glock *gl)
+{
+	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
+
+	if (likely(!gfs2_withdrawn(sdp)))
+		return false;
+	if (gl->gl_ops->go_flags & GLOF_NONDISK)
+		return false;
+	if (!sdp->sd_jdesc ||
+	    gl->gl_name.ln_number == sdp->sd_jdesc->jd_no_addr)
+		return false;
+	return true;
+}
+
 void gfs2_glock_free(struct gfs2_glock *gl)
 {
 	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
@@ -244,7 +271,7 @@ static void __gfs2_glock_put(struct gfs2_glock *gl)
 	gfs2_glock_remove_from_lru(gl);
 	spin_unlock(&gl->gl_lockref.lock);
 	GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders));
-	GLOCK_BUG_ON(gl, mapping && mapping->nrpages);
+	GLOCK_BUG_ON(gl, mapping && mapping->nrpages && !gfs2_withdrawn(sdp));
 	trace_gfs2_glock_put(gl);
 	sdp->sd_lockstruct.ls_ops->lm_put_lock(gl);
 }
@@ -281,7 +308,7 @@ void gfs2_glock_put(struct gfs2_glock *gl)
 
 static inline int may_grant(const struct gfs2_glock *gl, const struct gfs2_holder *gh)
 {
-	const struct gfs2_holder *gh_head = list_entry(gl->gl_holders.next, const struct gfs2_holder, gh_list);
+	const struct gfs2_holder *gh_head = list_first_entry(&gl->gl_holders, const struct gfs2_holder, gh_list);
 	if ((gh->gh_state == LM_ST_EXCLUSIVE ||
 	     gh_head->gh_state == LM_ST_EXCLUSIVE) && gh != gh_head)
 		return 0;
@@ -549,8 +576,8 @@ __acquires(&gl->gl_lockref.lock)
 	unsigned int lck_flags = (unsigned int)(gh ? gh->gh_flags : 0);
 	int ret;
 
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags)) &&
-	    target != LM_ST_UNLOCKED)
+	if (target != LM_ST_UNLOCKED && glock_blocked_by_withdraw(gl) &&
+	    gh && !(gh->gh_flags & LM_FLAG_NOEXP))
 		return;
 	lck_flags &= (LM_FLAG_TRY | LM_FLAG_TRY_1CB | LM_FLAG_NOEXP |
 		      LM_FLAG_PRIORITY);
@@ -558,7 +585,14 @@ __acquires(&gl->gl_lockref.lock)
 	GLOCK_BUG_ON(gl, gl->gl_state == gl->gl_target);
 	if ((target == LM_ST_UNLOCKED || target == LM_ST_DEFERRED) &&
 	    glops->go_inval) {
-		set_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags);
+		/*
+		 * If another process is already doing the invalidate, let that
+		 * finish first.  The glock state machine will get back to this
+		 * holder again later.
+		 */
+		if (test_and_set_bit(GLF_INVALIDATE_IN_PROGRESS,
+				     &gl->gl_flags))
+			return;
 		do_error(gl, 0); /* Fail queued try locks */
 	}
 	gl->gl_req = target;
@@ -568,13 +602,65 @@ __acquires(&gl->gl_lockref.lock)
 	    (lck_flags & (LM_FLAG_TRY|LM_FLAG_TRY_1CB)))
 		clear_bit(GLF_BLOCKING, &gl->gl_flags);
 	spin_unlock(&gl->gl_lockref.lock);
-	if (glops->go_sync)
-		glops->go_sync(gl);
-	if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags))
+	if (glops->go_sync) {
+		ret = glops->go_sync(gl);
+		/* If we had a problem syncing (due to io errors or whatever,
+		 * we should not invalidate the metadata or tell dlm to
+		 * release the glock to other nodes.
+		 */
+		if (ret) {
+			if (cmpxchg(&sdp->sd_log_error, 0, ret)) {
+				fs_err(sdp, "Error %d syncing glock \n", ret);
+				gfs2_dump_glock(NULL, gl, true);
+			}
+			goto skip_inval;
+		}
+	}
+	if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags)) {
+		/*
+		 * The call to go_sync should have cleared out the ail list.
+		 * If there are still items, we have a problem. We ought to
+		 * withdraw, but we can't because the withdraw code also uses
+		 * glocks. Warn about the error, dump the glock, then fall
+		 * through and wait for logd to do the withdraw for us.
+		 */
+		if ((atomic_read(&gl->gl_ail_count) != 0) &&
+		    (!cmpxchg(&sdp->sd_log_error, 0, -EIO))) {
+			gfs2_assert_warn(sdp, !atomic_read(&gl->gl_ail_count));
+			gfs2_dump_glock(NULL, gl, true);
+		}
 		glops->go_inval(gl, target == LM_ST_DEFERRED ? 0 : DIO_METADATA);
-	clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags);
+		clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags);
+	}
 
+skip_inval:
 	gfs2_glock_hold(gl);
+	/*
+	 * Check for an error encountered since we called go_sync and go_inval.
+	 * If so, we can't withdraw from the glock code because the withdraw
+	 * code itself uses glocks (see function signal_our_withdraw) to
+	 * change the mount to read-only. Most importantly, we must not call
+	 * dlm to unlock the glock until the journal is in a known good state
+	 * (after journal replay) otherwise other nodes may use the object
+	 * (rgrp or dinode) and then later, journal replay will corrupt the
+	 * file system. The best we can do here is wait for the logd daemon
+	 * to see sd_log_error and withdraw, and in the meantime, requeue the
+	 * work for later.
+	 *
+	 * However, if we're just unlocking the lock (say, for unmount, when
+	 * gfs2_gl_hash_clear calls clear_glock) and recovery is complete
+	 * then it's okay to tell dlm to unlock it.
+	 */
+	if (unlikely(sdp->sd_log_error && !gfs2_withdrawn(sdp)))
+		gfs2_withdraw_delayed(sdp);
+	if (glock_blocked_by_withdraw(gl)) {
+		if (target != LM_ST_UNLOCKED ||
+		    test_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags)) {
+			gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD);
+			goto out;
+		}
+	}
+
 	if (sdp->sd_lockstruct.ls_ops->lm_lock)	{
 		/* lock_dlm */
 		ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags);
@@ -583,17 +669,15 @@ __acquires(&gl->gl_lockref.lock)
 		    test_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags)) {
 			finish_xmote(gl, target);
 			gfs2_glock_queue_work(gl, 0);
-		}
-		else if (ret) {
+		} else if (ret) {
 			fs_err(sdp, "lm_lock ret %d\n", ret);
-			GLOCK_BUG_ON(gl, !test_bit(SDF_WITHDRAWN,
-						   &sdp->sd_flags));
+			GLOCK_BUG_ON(gl, !gfs2_withdrawn(sdp));
 		}
 	} else { /* lock_nolock */
 		finish_xmote(gl, target);
 		gfs2_glock_queue_work(gl, 0);
 	}
-
+out:
 	spin_lock(&gl->gl_lockref.lock);
 }
 
@@ -607,7 +691,7 @@ static inline struct gfs2_holder *find_first_holder(const struct gfs2_glock *gl)
 	struct gfs2_holder *gh;
 
 	if (!list_empty(&gl->gl_holders)) {
-		gh = list_entry(gl->gl_holders.next, struct gfs2_holder, gh_list);
+		gh = list_first_entry(&gl->gl_holders, struct gfs2_holder, gh_list);
 		if (test_bit(HIF_HOLDER, &gh->gh_iflags))
 			return gh;
 	}
@@ -820,7 +904,7 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
 	memset(&gl->gl_lksb, 0, sizeof(struct dlm_lksb));
 
 	if (glops->go_flags & GLOF_LVB) {
-		gl->gl_lksb.sb_lvbptr = kzalloc(GFS2_MIN_LVB_SIZE, GFP_NOFS);
+		gl->gl_lksb.sb_lvbptr = kzalloc(GDLM_LVB_SIZE, GFP_NOFS);
 		if (!gl->gl_lksb.sb_lvbptr) {
 			kmem_cache_free(cachep, gl);
 			return -ENOMEM;
@@ -1154,7 +1238,7 @@ fail:
 	}
 	list_add_tail(&gh->gh_list, insert_pt);
 do_cancel:
-	gh = list_entry(gl->gl_holders.next, struct gfs2_holder, gh_list);
+	gh = list_first_entry(&gl->gl_holders, struct gfs2_holder, gh_list);
 	if (!(gh->gh_flags & LM_FLAG_PRIORITY)) {
 		spin_unlock(&gl->gl_lockref.lock);
 		if (sdp->sd_lockstruct.ls_ops->lm_cancel)
@@ -1188,10 +1272,9 @@ trap_recursive:
 int gfs2_glock_nq(struct gfs2_holder *gh)
 {
 	struct gfs2_glock *gl = gh->gh_gl;
-	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
 	int error = 0;
 
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags)))
+	if (glock_blocked_by_withdraw(gl) && !(gh->gh_flags & LM_FLAG_NOEXP))
 		return -EIO;
 
 	if (test_bit(GLF_LRU, &gl->gl_flags))
@@ -1235,24 +1318,32 @@ int gfs2_glock_poll(struct gfs2_holder *gh)
 void gfs2_glock_dq(struct gfs2_holder *gh)
 {
 	struct gfs2_glock *gl = gh->gh_gl;
-	const struct gfs2_glock_operations *glops = gl->gl_ops;
+	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
 	unsigned delay = 0;
 	int fast_path = 0;
 
 	spin_lock(&gl->gl_lockref.lock);
+	/*
+	 * If we're in the process of file system withdraw, we cannot just
+	 * dequeue any glocks until our journal is recovered, lest we
+	 * introduce file system corruption. We need two exceptions to this
+	 * rule: We need to allow unlocking of nondisk glocks and the glock
+	 * for our own journal that needs recovery.
+	 */
+	if (test_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags) &&
+	    glock_blocked_by_withdraw(gl) &&
+	    gh->gh_gl != sdp->sd_jinode_gl) {
+		sdp->sd_glock_dqs_held++;
+		might_sleep();
+		wait_on_bit(&sdp->sd_flags, SDF_WITHDRAW_RECOVERY,
+			    TASK_UNINTERRUPTIBLE);
+	}
 	if (gh->gh_flags & GL_NOCACHE)
 		handle_callback(gl, LM_ST_UNLOCKED, 0, false);
 
 	list_del_init(&gh->gh_list);
 	clear_bit(HIF_HOLDER, &gh->gh_iflags);
 	if (find_first_holder(gl) == NULL) {
-		if (glops->go_unlock) {
-			GLOCK_BUG_ON(gl, test_and_set_bit(GLF_LOCK, &gl->gl_flags));
-			spin_unlock(&gl->gl_lockref.lock);
-			glops->go_unlock(gh);
-			spin_lock(&gl->gl_lockref.lock);
-			clear_bit(GLF_LOCK, &gl->gl_flags);
-		}
 		if (list_empty(&gl->gl_holders) &&
 		    !test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) &&
 		    !test_bit(GLF_DEMOTE, &gl->gl_flags))
@@ -1549,7 +1640,7 @@ __acquires(&lru_lock)
 	list_sort(NULL, list, glock_cmp);
 
 	while(!list_empty(list)) {
-		gl = list_entry(list->next, struct gfs2_glock, gl_lru);
+		gl = list_first_entry(list, struct gfs2_glock, gl_lru);
 		list_del_init(&gl->gl_lru);
 		if (!spin_trylock(&gl->gl_lockref.lock)) {
 add_back_to_lru:
@@ -1590,7 +1681,7 @@ static long gfs2_scan_glock_lru(int nr)
 
 	spin_lock(&lru_lock);
 	while ((nr-- >= 0) && !list_empty(&lru_list)) {
-		gl = list_entry(lru_list.next, struct gfs2_glock, gl_lru);
+		gl = list_first_entry(&lru_list, struct gfs2_glock, gl_lru);
 
 		/* Test for being demotable */
 		if (!test_bit(GLF_LOCK, &gl->gl_flags)) {
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index ff213690e364..9e9c7a4b8c66 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -29,6 +29,8 @@
 
 struct workqueue_struct *gfs2_freeze_wq;
 
+extern struct workqueue_struct *gfs2_control_wq;
+
 static void gfs2_ail_error(struct gfs2_glock *gl, const struct buffer_head *bh)
 {
 	fs_err(gl->gl_name.ln_sbd,
@@ -39,7 +41,8 @@ static void gfs2_ail_error(struct gfs2_glock *gl, const struct buffer_head *bh)
 	fs_err(gl->gl_name.ln_sbd, "AIL glock %u:%llu mapping %p\n",
 	       gl->gl_name.ln_type, gl->gl_name.ln_number,
 	       gfs2_glock2aspace(gl));
-	gfs2_lm_withdraw(gl->gl_name.ln_sbd, "AIL error\n");
+	gfs2_lm(gl->gl_name.ln_sbd, "AIL error\n");
+	gfs2_withdraw(gl->gl_name.ln_sbd);
 }
 
 /**
@@ -79,34 +82,62 @@ static void __gfs2_ail_flush(struct gfs2_glock *gl, bool fsync,
 }
 
 
-static void gfs2_ail_empty_gl(struct gfs2_glock *gl)
+static int gfs2_ail_empty_gl(struct gfs2_glock *gl)
 {
 	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
 	struct gfs2_trans tr;
+	int ret;
 
 	memset(&tr, 0, sizeof(tr));
 	INIT_LIST_HEAD(&tr.tr_buf);
 	INIT_LIST_HEAD(&tr.tr_databuf);
 	tr.tr_revokes = atomic_read(&gl->gl_ail_count);
 
-	if (!tr.tr_revokes)
-		return;
+	if (!tr.tr_revokes) {
+		bool have_revokes;
+		bool log_in_flight;
+
+		/*
+		 * We have nothing on the ail, but there could be revokes on
+		 * the sdp revoke queue, in which case, we still want to flush
+		 * the log and wait for it to finish.
+		 *
+		 * If the sdp revoke list is empty too, we might still have an
+		 * io outstanding for writing revokes, so we should wait for
+		 * it before returning.
+		 *
+		 * If none of these conditions are true, our revokes are all
+		 * flushed and we can return.
+		 */
+		gfs2_log_lock(sdp);
+		have_revokes = !list_empty(&sdp->sd_log_revokes);
+		log_in_flight = atomic_read(&sdp->sd_log_in_flight);
+		gfs2_log_unlock(sdp);
+		if (have_revokes)
+			goto flush;
+		if (log_in_flight)
+			log_flush_wait(sdp);
+		return 0;
+	}
 
 	/* A shortened, inline version of gfs2_trans_begin()
          * tr->alloced is not set since the transaction structure is
          * on the stack */
-	tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64));
+	tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes);
 	tr.tr_ip = _RET_IP_;
-	if (gfs2_log_reserve(sdp, tr.tr_reserved) < 0)
-		return;
+	ret = gfs2_log_reserve(sdp, tr.tr_reserved);
+	if (ret < 0)
+		return ret;
 	WARN_ON_ONCE(current->journal_info);
 	current->journal_info = &tr;
 
 	__gfs2_ail_flush(gl, 0, tr.tr_revokes);
 
 	gfs2_trans_end(sdp);
+flush:
 	gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL |
 		       GFS2_LFC_AIL_EMPTY_GL);
+	return 0;
 }
 
 void gfs2_ail_flush(struct gfs2_glock *gl, bool fsync)
@@ -140,35 +171,32 @@ void gfs2_ail_flush(struct gfs2_glock *gl, bool fsync)
  * return to caller to demote/unlock the glock until I/O is complete.
  */
 
-static void rgrp_go_sync(struct gfs2_glock *gl)
+static int rgrp_go_sync(struct gfs2_glock *gl)
 {
 	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
 	struct address_space *mapping = &sdp->sd_aspace;
-	struct gfs2_rgrpd *rgd;
+	struct gfs2_rgrpd *rgd = gfs2_glock2rgrp(gl);
 	int error;
 
-	spin_lock(&gl->gl_lockref.lock);
-	rgd = gl->gl_object;
-	if (rgd)
-		gfs2_rgrp_brelse(rgd);
-	spin_unlock(&gl->gl_lockref.lock);
-
 	if (!test_and_clear_bit(GLF_DIRTY, &gl->gl_flags))
-		return;
+		return 0;
 	GLOCK_BUG_ON(gl, gl->gl_state != LM_ST_EXCLUSIVE);
 
 	gfs2_log_flush(sdp, gl, GFS2_LOG_HEAD_FLUSH_NORMAL |
 		       GFS2_LFC_RGRP_GO_SYNC);
 	filemap_fdatawrite_range(mapping, gl->gl_vm.start, gl->gl_vm.end);
 	error = filemap_fdatawait_range(mapping, gl->gl_vm.start, gl->gl_vm.end);
+	WARN_ON_ONCE(error);
 	mapping_set_error(mapping, error);
-	gfs2_ail_empty_gl(gl);
+	if (!error)
+		error = gfs2_ail_empty_gl(gl);
 
 	spin_lock(&gl->gl_lockref.lock);
 	rgd = gl->gl_object;
 	if (rgd)
 		gfs2_free_clones(rgd);
 	spin_unlock(&gl->gl_lockref.lock);
+	return error;
 }
 
 /**
@@ -191,7 +219,6 @@ static void rgrp_go_inval(struct gfs2_glock *gl, int flags)
 		gfs2_rgrp_brelse(rgd);
 
 	WARN_ON_ONCE(!(flags & DIO_METADATA));
-	gfs2_assert_withdraw(sdp, !atomic_read(&gl->gl_ail_count));
 	truncate_inode_pages_range(mapping, gl->gl_vm.start, gl->gl_vm.end);
 
 	if (rgd)
@@ -236,12 +263,12 @@ static void gfs2_clear_glop_pending(struct gfs2_inode *ip)
  *
  */
 
-static void inode_go_sync(struct gfs2_glock *gl)
+static int inode_go_sync(struct gfs2_glock *gl)
 {
 	struct gfs2_inode *ip = gfs2_glock2inode(gl);
 	int isreg = ip && S_ISREG(ip->i_inode.i_mode);
 	struct address_space *metamapping = gfs2_glock2aspace(gl);
-	int error;
+	int error = 0;
 
 	if (isreg) {
 		if (test_and_clear_bit(GIF_SW_PAGED, &ip->i_flags))
@@ -274,6 +301,7 @@ static void inode_go_sync(struct gfs2_glock *gl)
 
 out:
 	gfs2_clear_glop_pending(ip);
+	return error;
 }
 
 /**
@@ -291,8 +319,6 @@ static void inode_go_inval(struct gfs2_glock *gl, int flags)
 {
 	struct gfs2_inode *ip = gfs2_glock2inode(gl);
 
-	gfs2_assert_withdraw(gl->gl_name.ln_sbd, !atomic_read(&gl->gl_ail_count));
-
 	if (flags & DIO_METADATA) {
 		struct address_space *mapping = gfs2_glock2aspace(gl);
 		truncate_inode_pages(mapping, 0);
@@ -350,7 +376,7 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf)
 		ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major),
 					   be32_to_cpu(str->di_minor));
 		break;
-	};
+	}
 
 	i_uid_write(&ip->i_inode, be32_to_cpu(str->di_uid));
 	i_gid_write(&ip->i_inode, be32_to_cpu(str->di_gid));
@@ -496,24 +522,29 @@ static void inode_go_dump(struct seq_file *seq, struct gfs2_glock *gl,
  *
  */
 
-static void freeze_go_sync(struct gfs2_glock *gl)
+static int freeze_go_sync(struct gfs2_glock *gl)
 {
 	int error = 0;
 	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
 
-	if (gl->gl_state == LM_ST_SHARED &&
+	if (gl->gl_state == LM_ST_SHARED && !gfs2_withdrawn(sdp) &&
 	    test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) {
 		atomic_set(&sdp->sd_freeze_state, SFS_STARTING_FREEZE);
 		error = freeze_super(sdp->sd_vfs);
 		if (error) {
 			fs_info(sdp, "GFS2: couldn't freeze filesystem: %d\n",
 				error);
+			if (gfs2_withdrawn(sdp)) {
+				atomic_set(&sdp->sd_freeze_state, SFS_UNFROZEN);
+				return 0;
+			}
 			gfs2_assert_withdraw(sdp, 0);
 		}
 		queue_work(gfs2_freeze_wq, &sdp->sd_freeze_work);
 		gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_FREEZE |
 			       GFS2_LFC_FREEZE_GO_SYNC);
 	}
+	return 0;
 }
 
 /**
@@ -540,7 +571,7 @@ static int freeze_go_xmote_bh(struct gfs2_glock *gl, struct gfs2_holder *gh)
 			gfs2_consist(sdp);
 
 		/*  Initialize some head of the log stuff  */
-		if (!test_bit(SDF_WITHDRAWN, &sdp->sd_flags)) {
+		if (!gfs2_withdrawn(sdp)) {
 			sdp->sd_log_sequence = head.lh_sequence + 1;
 			gfs2_log_pointers_init(sdp, head.lh_blkno);
 		}
@@ -582,8 +613,76 @@ static void iopen_go_callback(struct gfs2_glock *gl, bool remote)
 	}
 }
 
+/**
+ * inode_go_free - wake up anyone waiting for dlm's unlock ast to free it
+ * @gl: glock being freed
+ *
+ * For now, this is only used for the journal inode glock. In withdraw
+ * situations, we need to wait for the glock to be freed so that we know
+ * other nodes may proceed with recovery / journal replay.
+ */
+static void inode_go_free(struct gfs2_glock *gl)
+{
+	/* Note that we cannot reference gl_object because it's already set
+	 * to NULL by this point in its lifecycle. */
+	if (!test_bit(GLF_FREEING, &gl->gl_flags))
+		return;
+	clear_bit_unlock(GLF_FREEING, &gl->gl_flags);
+	wake_up_bit(&gl->gl_flags, GLF_FREEING);
+}
+
+/**
+ * nondisk_go_callback - used to signal when a node did a withdraw
+ * @gl: the nondisk glock
+ * @remote: true if this came from a different cluster node
+ *
+ */
+static void nondisk_go_callback(struct gfs2_glock *gl, bool remote)
+{
+	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
+
+	/* Ignore the callback unless it's from another node, and it's the
+	   live lock. */
+	if (!remote || gl->gl_name.ln_number != GFS2_LIVE_LOCK)
+		return;
+
+	/* First order of business is to cancel the demote request. We don't
+	 * really want to demote a nondisk glock. At best it's just to inform
+	 * us of another node's withdraw. We'll keep it in SH mode. */
+	clear_bit(GLF_DEMOTE, &gl->gl_flags);
+	clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags);
+
+	/* Ignore the unlock if we're withdrawn, unmounting, or in recovery. */
+	if (test_bit(SDF_NORECOVERY, &sdp->sd_flags) ||
+	    test_bit(SDF_WITHDRAWN, &sdp->sd_flags) ||
+	    test_bit(SDF_REMOTE_WITHDRAW, &sdp->sd_flags))
+		return;
+
+	/* We only care when a node wants us to unlock, because that means
+	 * they want a journal recovered. */
+	if (gl->gl_demote_state != LM_ST_UNLOCKED)
+		return;
+
+	if (sdp->sd_args.ar_spectator) {
+		fs_warn(sdp, "Spectator node cannot recover journals.\n");
+		return;
+	}
+
+	fs_warn(sdp, "Some node has withdrawn; checking for recovery.\n");
+	set_bit(SDF_REMOTE_WITHDRAW, &sdp->sd_flags);
+	/*
+	 * We can't call remote_withdraw directly here or gfs2_recover_journal
+	 * because this is called from the glock unlock function and the
+	 * remote_withdraw needs to enqueue and dequeue the same "live" glock
+	 * we were called from. So we queue it to the control work queue in
+	 * lock_dlm.
+	 */
+	queue_delayed_work(gfs2_control_wq, &sdp->sd_control_work, 0);
+}
+
 const struct gfs2_glock_operations gfs2_meta_glops = {
 	.go_type = LM_TYPE_META,
+	.go_flags = GLOF_NONDISK,
 };
 
 const struct gfs2_glock_operations gfs2_inode_glops = {
@@ -594,13 +693,13 @@ const struct gfs2_glock_operations gfs2_inode_glops = {
 	.go_dump = inode_go_dump,
 	.go_type = LM_TYPE_INODE,
 	.go_flags = GLOF_ASPACE | GLOF_LRU,
+	.go_free = inode_go_free,
 };
 
 const struct gfs2_glock_operations gfs2_rgrp_glops = {
 	.go_sync = rgrp_go_sync,
 	.go_inval = rgrp_go_inval,
 	.go_lock = gfs2_rgrp_go_lock,
-	.go_unlock = gfs2_rgrp_go_unlock,
 	.go_dump = gfs2_rgrp_dump,
 	.go_type = LM_TYPE_RGRP,
 	.go_flags = GLOF_LVB,
@@ -611,30 +710,34 @@ const struct gfs2_glock_operations gfs2_freeze_glops = {
 	.go_xmote_bh = freeze_go_xmote_bh,
 	.go_demote_ok = freeze_go_demote_ok,
 	.go_type = LM_TYPE_NONDISK,
+	.go_flags = GLOF_NONDISK,
 };
 
 const struct gfs2_glock_operations gfs2_iopen_glops = {
 	.go_type = LM_TYPE_IOPEN,
 	.go_callback = iopen_go_callback,
-	.go_flags = GLOF_LRU,
+	.go_flags = GLOF_LRU | GLOF_NONDISK,
 };
 
 const struct gfs2_glock_operations gfs2_flock_glops = {
 	.go_type = LM_TYPE_FLOCK,
-	.go_flags = GLOF_LRU,
+	.go_flags = GLOF_LRU | GLOF_NONDISK,
 };
 
 const struct gfs2_glock_operations gfs2_nondisk_glops = {
 	.go_type = LM_TYPE_NONDISK,
+	.go_flags = GLOF_NONDISK,
+	.go_callback = nondisk_go_callback,
 };
 
 const struct gfs2_glock_operations gfs2_quota_glops = {
 	.go_type = LM_TYPE_QUOTA,
-	.go_flags = GLOF_LVB | GLOF_LRU,
+	.go_flags = GLOF_LVB | GLOF_LRU | GLOF_NONDISK,
 };
 
 const struct gfs2_glock_operations gfs2_journal_glops = {
 	.go_type = LM_TYPE_JOURNAL,
+	.go_flags = GLOF_NONDISK,
 };
 
 const struct gfs2_glock_operations *gfs2_glops_list[] = {
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 5f89c515f5bb..84a824293a78 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -234,20 +234,21 @@ struct lm_lockname {
 
 
 struct gfs2_glock_operations {
-	void (*go_sync) (struct gfs2_glock *gl);
+	int (*go_sync) (struct gfs2_glock *gl);
 	int (*go_xmote_bh) (struct gfs2_glock *gl, struct gfs2_holder *gh);
 	void (*go_inval) (struct gfs2_glock *gl, int flags);
 	int (*go_demote_ok) (const struct gfs2_glock *gl);
 	int (*go_lock) (struct gfs2_holder *gh);
-	void (*go_unlock) (struct gfs2_holder *gh);
 	void (*go_dump)(struct seq_file *seq, struct gfs2_glock *gl,
 			const char *fs_id_buf);
 	void (*go_callback)(struct gfs2_glock *gl, bool remote);
+	void (*go_free)(struct gfs2_glock *gl);
 	const int go_type;
 	const unsigned long go_flags;
-#define GLOF_ASPACE 1
-#define GLOF_LVB    2
-#define GLOF_LRU    4
+#define GLOF_ASPACE 1 /* address space attached */
+#define GLOF_LVB    2 /* Lock Value Block attached */
+#define GLOF_LRU    4 /* LRU managed */
+#define GLOF_NONDISK   8 /* not I/O related */
 };
 
 enum {
@@ -294,6 +295,7 @@ struct gfs2_qadata { /* quota allocation data */
 	struct gfs2_quota_data *qa_qd[2 * GFS2_MAXQUOTAS];
 	struct gfs2_holder qa_qd_ghs[2 * GFS2_MAXQUOTAS];
 	unsigned int qa_qd_num;
+	int qa_ref;
 };
 
 /* Resource group multi-block reservation, in order of appearance:
@@ -343,6 +345,7 @@ enum {
 	GLF_OBJECT			= 14, /* Used only for tracing */
 	GLF_BLOCKING			= 15,
 	GLF_INODE_CREATING		= 16, /* Inode creation occurring */
+	GLF_FREEING			= 18, /* Wait for glock to be freed */
 };
 
 struct gfs2_glock {
@@ -387,8 +390,6 @@ struct gfs2_glock {
 	struct rhash_head gl_node;
 };
 
-#define GFS2_MIN_LVB_SIZE 32	/* Min size of LVB that gfs2 supports */
-
 enum {
 	GIF_INVALID		= 0,
 	GIF_QD_LOCKED		= 1,
@@ -505,6 +506,7 @@ struct gfs2_trans {
 	unsigned int tr_num_buf_rm;
 	unsigned int tr_num_databuf_rm;
 	unsigned int tr_num_revoke;
+	unsigned int tr_num_revoke_rm;
 
 	struct list_head tr_list;
 	struct list_head tr_databuf;
@@ -543,6 +545,7 @@ struct gfs2_jdesc {
 	struct list_head jd_revoke_list;
 	unsigned int jd_replay_tail;
 
+	u64 jd_no_addr;
 };
 
 struct gfs2_statfs_change_host {
@@ -617,8 +620,12 @@ enum {
 	SDF_RORECOVERY		= 7, /* read only recovery */
 	SDF_SKIP_DLM_UNLOCK	= 8,
 	SDF_FORCE_AIL_FLUSH     = 9,
-	SDF_AIL1_IO_ERROR	= 10,
-	SDF_FS_FROZEN           = 11,
+	SDF_FS_FROZEN           = 10,
+	SDF_WITHDRAWING		= 11, /* Will withdraw eventually */
+	SDF_WITHDRAW_IN_PROG	= 12, /* Withdraw is in progress */
+	SDF_REMOTE_WITHDRAW	= 13, /* Performing remote recovery */
+	SDF_WITHDRAW_RECOVERY	= 14, /* Wait for journal recovery when we are
+					 withdrawing */
 };
 
 enum gfs2_freeze_state {
@@ -703,6 +710,7 @@ struct gfs2_sbd {
 	u32 sd_fsb2bb_shift;
 	u32 sd_diptrs;	/* Number of pointers in a dinode */
 	u32 sd_inptrs;	/* Number of pointers in a indirect block */
+	u32 sd_ldptrs;  /* Number of pointers in a log descriptor block */
 	u32 sd_jbsize;	/* Size of a journaled data block */
 	u32 sd_hash_bsize;	/* sizeof(exhash block) */
 	u32 sd_hash_bsize_shift;
@@ -768,6 +776,7 @@ struct gfs2_sbd {
 	struct gfs2_jdesc *sd_jdesc;
 	struct gfs2_holder sd_journal_gh;
 	struct gfs2_holder sd_jinode_gh;
+	struct gfs2_glock *sd_jinode_gl;
 
 	struct gfs2_holder sd_sc_gh;
 	struct gfs2_holder sd_qc_gh;
@@ -803,7 +812,7 @@ struct gfs2_sbd {
 
 	struct gfs2_trans *sd_log_tr;
 	unsigned int sd_log_blks_reserved;
-	int sd_log_commited_revoke;
+	int sd_log_committed_revoke;
 
 	atomic_t sd_log_pinned;
 	unsigned int sd_log_num_revoke;
@@ -828,7 +837,8 @@ struct gfs2_sbd {
 	atomic_t sd_log_in_flight;
 	struct bio *sd_log_bio;
 	wait_queue_head_t sd_log_flush_wait;
-	int sd_log_error;
+	int sd_log_error; /* First log error */
+	wait_queue_head_t sd_withdraw_wait;
 
 	atomic_t sd_reserving_log;
 	wait_queue_head_t sd_reserving_log_wait;
@@ -852,6 +862,7 @@ struct gfs2_sbd {
 
 	unsigned long sd_last_warning;
 	struct dentry *debugfs_dir;    /* debugfs directory */
+	unsigned long sd_glock_dqs_held;
 };
 
 static inline void gfs2_glstats_inc(struct gfs2_glock *gl, int which)
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index e1e18fb587eb..5acd3ce30759 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -136,7 +136,6 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type,
 
 	if (inode->i_state & I_NEW) {
 		struct gfs2_sbd *sdp = GFS2_SB(inode);
-		ip->i_no_formal_ino = no_formal_ino;
 
 		error = gfs2_glock_get(sdp, no_addr, &gfs2_inode_glops, CREATE, &ip->i_gl);
 		if (unlikely(error))
@@ -145,7 +144,7 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type,
 
 		error = gfs2_glock_get(sdp, no_addr, &gfs2_iopen_glops, CREATE, &io_gl);
 		if (unlikely(error))
-			goto fail_put;
+			goto fail;
 
 		if (type == DT_UNKNOWN || blktype != GFS2_BLKST_FREE) {
 			/*
@@ -156,13 +155,13 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type,
 			error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE,
 						   GL_SKIP, &i_gh);
 			if (error)
-				goto fail_put;
+				goto fail;
 
 			if (blktype != GFS2_BLKST_FREE) {
 				error = gfs2_check_blk_type(sdp, no_addr,
 							    blktype);
 				if (error)
-					goto fail_put;
+					goto fail;
 			}
 		}
 
@@ -170,26 +169,27 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type,
 		set_bit(GIF_INVALID, &ip->i_flags);
 		error = gfs2_glock_nq_init(io_gl, LM_ST_SHARED, GL_EXACT, &ip->i_iopen_gh);
 		if (unlikely(error))
-			goto fail_put;
+			goto fail;
 		glock_set_object(ip->i_iopen_gh.gh_gl, ip);
 		gfs2_glock_put(io_gl);
 		io_gl = NULL;
 
+		/* Lowest possible timestamp; will be overwritten in gfs2_dinode_in. */
+		inode->i_atime.tv_sec = 1LL << (8 * sizeof(inode->i_atime.tv_sec) - 1);
+		inode->i_atime.tv_nsec = 0;
+
 		if (type == DT_UNKNOWN) {
 			/* Inode glock must be locked already */
 			error = gfs2_inode_refresh(GFS2_I(inode));
 			if (error)
-				goto fail_refresh;
+				goto fail;
 		} else {
+			ip->i_no_formal_ino = no_formal_ino;
 			inode->i_mode = DT2IF(type);
 		}
 
 		gfs2_set_iop(inode);
 
-		/* Lowest possible timestamp; will be overwritten in gfs2_dinode_in. */
-		inode->i_atime.tv_sec = 1LL << (8 * sizeof(inode->i_atime.tv_sec) - 1);
-		inode->i_atime.tv_nsec = 0;
-
 		unlock_new_inode(inode);
 	}
 
@@ -197,17 +197,11 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type,
 		gfs2_glock_dq_uninit(&i_gh);
 	return inode;
 
-fail_refresh:
-	ip->i_iopen_gh.gh_flags |= GL_NOCACHE;
-	glock_clear_object(ip->i_iopen_gh.gh_gl, ip);
-	gfs2_glock_dq_uninit(&ip->i_iopen_gh);
-fail_put:
+fail:
 	if (io_gl)
 		gfs2_glock_put(io_gl);
-	glock_clear_object(ip->i_gl, ip);
 	if (gfs2_holder_initialized(&i_gh))
 		gfs2_glock_dq_uninit(&i_gh);
-fail:
 	iget_failed(inode);
 	return ERR_PTR(error);
 }
@@ -594,13 +588,13 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 	if (!name->len || name->len > GFS2_FNAMESIZE)
 		return -ENAMETOOLONG;
 
-	error = gfs2_rsqa_alloc(dip);
+	error = gfs2_qa_get(dip);
 	if (error)
 		return error;
 
 	error = gfs2_rindex_update(sdp);
 	if (error)
-		return error;
+		goto fail;
 
 	error = gfs2_glock_nq_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
 	if (error)
@@ -628,7 +622,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 				error = finish_no_open(file, NULL);
 		}
 		gfs2_glock_dq_uninit(ghs);
-		return error;
+		goto fail;
 	} else if (error != -ENOENT) {
 		goto fail_gunlock;
 	}
@@ -647,7 +641,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 		goto fail_gunlock;
 
 	ip = GFS2_I(inode);
-	error = gfs2_rsqa_alloc(ip);
+	error = gfs2_qa_get(ip);
 	if (error)
 		goto fail_free_acls;
 
@@ -656,7 +650,6 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 	inode->i_rdev = dev;
 	inode->i_size = size;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
-	gfs2_set_inode_blocks(inode, 1);
 	munge_mode_uid_gid(dip, inode);
 	check_and_update_goal(dip);
 	ip->i_goal = dip->i_goal;
@@ -712,7 +705,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 
 	error = gfs2_trans_begin(sdp, blocks, 0);
 	if (error)
-		goto fail_gunlock2;
+		goto fail_free_inode;
 
 	if (blocks > 1) {
 		ip->i_eattr = ip->i_no_addr + 1;
@@ -723,7 +716,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 
 	error = gfs2_glock_get(sdp, ip->i_no_addr, &gfs2_iopen_glops, CREATE, &io_gl);
 	if (error)
-		goto fail_gunlock2;
+		goto fail_free_inode;
 
 	BUG_ON(test_and_set_bit(GLF_INODE_CREATING, &io_gl->gl_flags));
 
@@ -732,7 +725,6 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 		goto fail_gunlock2;
 
 	glock_set_object(ip->i_iopen_gh.gh_gl, ip);
-	gfs2_glock_put(io_gl);
 	gfs2_set_iop(inode);
 	insert_inode_hash(inode);
 
@@ -765,28 +757,33 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 
 	mark_inode_dirty(inode);
 	d_instantiate(dentry, inode);
+	/* After instantiate, errors should result in evict which will destroy
+	 * both inode and iopen glocks properly. */
 	if (file) {
 		file->f_mode |= FMODE_CREATED;
 		error = finish_open(file, dentry, gfs2_open_common);
 	}
 	gfs2_glock_dq_uninit(ghs);
+	gfs2_qa_put(ip);
 	gfs2_glock_dq_uninit(ghs + 1);
 	clear_bit(GLF_INODE_CREATING, &io_gl->gl_flags);
+	gfs2_glock_put(io_gl);
+	gfs2_qa_put(dip);
 	return error;
 
 fail_gunlock3:
 	glock_clear_object(io_gl, ip);
 	gfs2_glock_dq_uninit(&ip->i_iopen_gh);
-	gfs2_glock_put(io_gl);
 fail_gunlock2:
-	if (io_gl)
-		clear_bit(GLF_INODE_CREATING, &io_gl->gl_flags);
+	clear_bit(GLF_INODE_CREATING, &io_gl->gl_flags);
+	gfs2_glock_put(io_gl);
 fail_free_inode:
 	if (ip->i_gl) {
 		glock_clear_object(ip->i_gl, ip);
 		gfs2_glock_put(ip->i_gl);
 	}
-	gfs2_rsqa_delete(ip, NULL);
+	gfs2_rs_delete(ip, NULL);
+	gfs2_qa_put(ip);
 fail_free_acls:
 	posix_acl_release(default_acl);
 	posix_acl_release(acl);
@@ -804,6 +801,7 @@ fail_gunlock:
 	if (gfs2_holder_initialized(ghs + 1))
 		gfs2_glock_dq_uninit(ghs + 1);
 fail:
+	gfs2_qa_put(dip);
 	return error;
 }
 
@@ -905,7 +903,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir,
 	if (S_ISDIR(inode->i_mode))
 		return -EPERM;
 
-	error = gfs2_rsqa_alloc(dip);
+	error = gfs2_qa_get(dip);
 	if (error)
 		return error;
 
@@ -1008,6 +1006,7 @@ out_gunlock:
 out_child:
 	gfs2_glock_dq(ghs);
 out_parent:
+	gfs2_qa_put(dip);
 	gfs2_holder_uninit(ghs);
 	gfs2_holder_uninit(ghs + 1);
 	return error;
@@ -1248,7 +1247,7 @@ static int gfs2_atomic_open(struct inode *dir, struct dentry *dentry,
 		if (!(file->f_mode & FMODE_OPENED))
 			return finish_no_open(file, d);
 		dput(d);
-		return 0;
+		return excl && (flags & O_CREAT) ? -EEXIST : 0;
 	}
 
 	BUG_ON(d != NULL);
@@ -1368,7 +1367,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
 	if (error)
 		return error;
 
-	error = gfs2_rsqa_alloc(ndip);
+	error = gfs2_qa_get(ndip);
 	if (error)
 		return error;
 
@@ -1475,7 +1474,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
 			error = -EEXIST;
 		default:
 			goto out_gunlock;
-		};
+		}
 
 		if (odip != ndip) {
 			if (!ndip->i_inode.i_nlink) {
@@ -1568,6 +1567,7 @@ out_gunlock_r:
 	if (gfs2_holder_initialized(&r_gh))
 		gfs2_glock_dq_uninit(&r_gh);
 out:
+	gfs2_qa_put(ndip);
 	return error;
 }
 
@@ -1879,10 +1879,9 @@ static int setattr_chown(struct inode *inode, struct iattr *attr)
 		ouid = nuid = NO_UID_QUOTA_CHANGE;
 	if (!(attr->ia_valid & ATTR_GID) || gid_eq(ogid, ngid))
 		ogid = ngid = NO_GID_QUOTA_CHANGE;
-
-	error = gfs2_rsqa_alloc(ip);
+	error = gfs2_qa_get(ip);
 	if (error)
-		goto out;
+		return error;
 
 	error = gfs2_rindex_update(sdp);
 	if (error)
@@ -1920,6 +1919,7 @@ out_end_trans:
 out_gunlock_q:
 	gfs2_quota_unlock(ip);
 out:
+	gfs2_qa_put(ip);
 	return error;
 }
 
@@ -1941,21 +1941,21 @@ static int gfs2_setattr(struct dentry *dentry, struct iattr *attr)
 	struct gfs2_holder i_gh;
 	int error;
 
-	error = gfs2_rsqa_alloc(ip);
+	error = gfs2_qa_get(ip);
 	if (error)
 		return error;
 
 	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh);
 	if (error)
-		return error;
+		goto out;
 
 	error = -EPERM;
 	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		goto out;
+		goto error;
 
 	error = setattr_prepare(dentry, attr);
 	if (error)
-		goto out;
+		goto error;
 
 	if (attr->ia_valid & ATTR_SIZE)
 		error = gfs2_setattr_size(inode, attr->ia_size);
@@ -1967,10 +1967,12 @@ static int gfs2_setattr(struct dentry *dentry, struct iattr *attr)
 			error = posix_acl_chmod(inode, inode->i_mode);
 	}
 
-out:
+error:
 	if (!error)
 		mark_inode_dirty(inode);
 	gfs2_glock_dq_uninit(&i_gh);
+out:
+	gfs2_qa_put(ip);
 	return error;
 }
 
diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c
index 7c7197343ee2..9f2b5609f225 100644
--- a/fs/gfs2/lock_dlm.c
+++ b/fs/gfs2/lock_dlm.c
@@ -16,6 +16,8 @@
 
 #include "incore.h"
 #include "glock.h"
+#include "glops.h"
+#include "recovery.h"
 #include "util.h"
 #include "sys.h"
 #include "trace_gfs2.h"
@@ -124,6 +126,8 @@ static void gdlm_ast(void *arg)
 
 	switch (gl->gl_lksb.sb_status) {
 	case -DLM_EUNLOCK: /* Unlocked, so glock can be freed */
+		if (gl->gl_ops->go_free)
+			gl->gl_ops->go_free(gl);
 		gfs2_glock_free(gl);
 		return;
 	case -DLM_ECANCEL: /* Cancel while getting lock */
@@ -323,6 +327,7 @@ static void gdlm_cancel(struct gfs2_glock *gl)
 /*
  * dlm/gfs2 recovery coordination using dlm_recover callbacks
  *
+ *  0. gfs2 checks for another cluster node withdraw, needing journal replay
  *  1. dlm_controld sees lockspace members change
  *  2. dlm_controld blocks dlm-kernel locking activity
  *  3. dlm_controld within dlm-kernel notifies gfs2 (recover_prep)
@@ -571,6 +576,28 @@ static int control_lock(struct gfs2_sbd *sdp, int mode, uint32_t flags)
 			 &ls->ls_control_lksb, "control_lock");
 }
 
+/**
+ * remote_withdraw - react to a node withdrawing from the file system
+ * @sdp: The superblock
+ */
+static void remote_withdraw(struct gfs2_sbd *sdp)
+{
+	struct gfs2_jdesc *jd;
+	int ret = 0, count = 0;
+
+	list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
+		if (jd->jd_jid == sdp->sd_lockstruct.ls_jid)
+			continue;
+		ret = gfs2_recover_journal(jd, true);
+		if (ret)
+			break;
+		count++;
+	}
+
+	/* Now drop the additional reference we acquired */
+	fs_err(sdp, "Journals checked: %d, ret = %d.\n", count, ret);
+}
+
 static void gfs2_control_func(struct work_struct *work)
 {
 	struct gfs2_sbd *sdp = container_of(work, struct gfs2_sbd, sd_control_work.work);
@@ -581,6 +608,13 @@ static void gfs2_control_func(struct work_struct *work)
 	int recover_size;
 	int i, error;
 
+	/* First check for other nodes that may have done a withdraw. */
+	if (test_bit(SDF_REMOTE_WITHDRAW, &sdp->sd_flags)) {
+		remote_withdraw(sdp);
+		clear_bit(SDF_REMOTE_WITHDRAW, &sdp->sd_flags);
+		return;
+	}
+
 	spin_lock(&ls->ls_recover_spin);
 	/*
 	 * No MOUNT_DONE means we're still mounting; control_mount()
@@ -1079,6 +1113,10 @@ static void gdlm_recover_prep(void *arg)
 	struct gfs2_sbd *sdp = arg;
 	struct lm_lockstruct *ls = &sdp->sd_lockstruct;
 
+	if (gfs2_withdrawn(sdp)) {
+		fs_err(sdp, "recover_prep ignored due to withdraw.\n");
+		return;
+	}
 	spin_lock(&ls->ls_recover_spin);
 	ls->ls_recover_block = ls->ls_recover_start;
 	set_bit(DFL_DLM_RECOVERY, &ls->ls_recover_flags);
@@ -1101,6 +1139,11 @@ static void gdlm_recover_slot(void *arg, struct dlm_slot *slot)
 	struct lm_lockstruct *ls = &sdp->sd_lockstruct;
 	int jid = slot->slot - 1;
 
+	if (gfs2_withdrawn(sdp)) {
+		fs_err(sdp, "recover_slot jid %d ignored due to withdraw.\n",
+		       jid);
+		return;
+	}
 	spin_lock(&ls->ls_recover_spin);
 	if (ls->ls_recover_size < jid + 1) {
 		fs_err(sdp, "recover_slot jid %d gen %u short size %d\n",
@@ -1125,6 +1168,10 @@ static void gdlm_recover_done(void *arg, struct dlm_slot *slots, int num_slots,
 	struct gfs2_sbd *sdp = arg;
 	struct lm_lockstruct *ls = &sdp->sd_lockstruct;
 
+	if (gfs2_withdrawn(sdp)) {
+		fs_err(sdp, "recover_done ignored due to withdraw.\n");
+		return;
+	}
 	/* ensure the ls jid arrays are large enough */
 	set_recover_size(sdp, slots, num_slots);
 
@@ -1152,6 +1199,11 @@ static void gdlm_recovery_result(struct gfs2_sbd *sdp, unsigned int jid,
 {
 	struct lm_lockstruct *ls = &sdp->sd_lockstruct;
 
+	if (gfs2_withdrawn(sdp)) {
+		fs_err(sdp, "recovery_result jid %d ignored due to withdraw.\n",
+		       jid);
+		return;
+	}
 	if (test_bit(DFL_NO_DLM_OPS, &ls->ls_recover_flags))
 		return;
 
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index 58e237fba565..0644e58c6191 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -31,11 +31,12 @@
 #include "dir.h"
 #include "trace_gfs2.h"
 
+static void gfs2_log_shutdown(struct gfs2_sbd *sdp);
+
 /**
  * gfs2_struct2blk - compute stuff
  * @sdp: the filesystem
  * @nstruct: the number of structures
- * @ssize: the size of the structures
  *
  * Compute the number of log descriptor blocks needed to hold a certain number
  * of structures of a certain size.
@@ -43,18 +44,16 @@
  * Returns: the number of blocks needed (minimum is always 1)
  */
 
-unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct,
-			     unsigned int ssize)
+unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct)
 {
 	unsigned int blks;
 	unsigned int first, second;
 
 	blks = 1;
-	first = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / ssize;
+	first = sdp->sd_ldptrs;
 
 	if (nstruct > first) {
-		second = (sdp->sd_sb.sb_bsize -
-			  sizeof(struct gfs2_meta_header)) / ssize;
+		second = sdp->sd_inptrs;
 		blks += DIV_ROUND_UP(nstruct - first, second);
 	}
 
@@ -89,8 +88,7 @@ static void gfs2_remove_from_ail(struct gfs2_bufdata *bd)
 
 static int gfs2_ail1_start_one(struct gfs2_sbd *sdp,
 			       struct writeback_control *wbc,
-			       struct gfs2_trans *tr,
-			       bool *withdraw)
+			       struct gfs2_trans *tr)
 __releases(&sdp->sd_ail_lock)
 __acquires(&sdp->sd_ail_lock)
 {
@@ -98,6 +96,7 @@ __acquires(&sdp->sd_ail_lock)
 	struct address_space *mapping;
 	struct gfs2_bufdata *bd, *s;
 	struct buffer_head *bh;
+	int ret = 0;
 
 	list_for_each_entry_safe_reverse(bd, s, &tr->tr_ail1_list, bd_ail_st_list) {
 		bh = bd->bd_bh;
@@ -105,16 +104,21 @@ __acquires(&sdp->sd_ail_lock)
 		gfs2_assert(sdp, bd->bd_tr == tr);
 
 		if (!buffer_busy(bh)) {
-			if (!buffer_uptodate(bh) &&
-			    !test_and_set_bit(SDF_AIL1_IO_ERROR,
-					      &sdp->sd_flags)) {
+			if (buffer_uptodate(bh)) {
+				list_move(&bd->bd_ail_st_list,
+					  &tr->tr_ail2_list);
+				continue;
+			}
+			if (!cmpxchg(&sdp->sd_log_error, 0, -EIO)) {
 				gfs2_io_error_bh(sdp, bh);
-				*withdraw = true;
+				gfs2_withdraw_delayed(sdp);
 			}
-			list_move(&bd->bd_ail_st_list, &tr->tr_ail2_list);
-			continue;
 		}
 
+		if (gfs2_withdrawn(sdp)) {
+			gfs2_remove_from_ail(bd);
+			continue;
+		}
 		if (!buffer_dirty(bh))
 			continue;
 		if (gl == bd->bd_gl)
@@ -125,16 +129,50 @@ __acquires(&sdp->sd_ail_lock)
 		if (!mapping)
 			continue;
 		spin_unlock(&sdp->sd_ail_lock);
-		generic_writepages(mapping, wbc);
+		ret = generic_writepages(mapping, wbc);
 		spin_lock(&sdp->sd_ail_lock);
-		if (wbc->nr_to_write <= 0)
+		if (ret || wbc->nr_to_write <= 0)
 			break;
-		return 1;
+		return -EBUSY;
 	}
 
-	return 0;
+	return ret;
 }
 
+static void dump_ail_list(struct gfs2_sbd *sdp)
+{
+	struct gfs2_trans *tr;
+	struct gfs2_bufdata *bd;
+	struct buffer_head *bh;
+
+	fs_err(sdp, "Error: In gfs2_ail1_flush for ten minutes! t=%d\n",
+	       current->journal_info ? 1 : 0);
+
+	list_for_each_entry_reverse(tr, &sdp->sd_ail1_list, tr_list) {
+		list_for_each_entry_reverse(bd, &tr->tr_ail1_list,
+					    bd_ail_st_list) {
+			bh = bd->bd_bh;
+			fs_err(sdp, "bd %p: blk:0x%llx bh=%p ", bd,
+			       (unsigned long long)bd->bd_blkno, bh);
+			if (!bh) {
+				fs_err(sdp, "\n");
+				continue;
+			}
+			fs_err(sdp, "0x%llx up2:%d dirt:%d lkd:%d req:%d "
+			       "map:%d new:%d ar:%d aw:%d delay:%d "
+			       "io err:%d unwritten:%d dfr:%d pin:%d esc:%d\n",
+			       (unsigned long long)bh->b_blocknr,
+			       buffer_uptodate(bh), buffer_dirty(bh),
+			       buffer_locked(bh), buffer_req(bh),
+			       buffer_mapped(bh), buffer_new(bh),
+			       buffer_async_read(bh), buffer_async_write(bh),
+			       buffer_delay(bh), buffer_write_io_error(bh),
+			       buffer_unwritten(bh),
+			       buffer_defer_completion(bh),
+			       buffer_pinned(bh), buffer_escaped(bh));
+		}
+	}
+}
 
 /**
  * gfs2_ail1_flush - start writeback of some ail1 entries 
@@ -150,22 +188,36 @@ void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc)
 	struct list_head *head = &sdp->sd_ail1_list;
 	struct gfs2_trans *tr;
 	struct blk_plug plug;
-	bool withdraw = false;
+	int ret;
+	unsigned long flush_start = jiffies;
 
 	trace_gfs2_ail_flush(sdp, wbc, 1);
 	blk_start_plug(&plug);
 	spin_lock(&sdp->sd_ail_lock);
 restart:
+	ret = 0;
+	if (time_after(jiffies, flush_start + (HZ * 600))) {
+		dump_ail_list(sdp);
+		goto out;
+	}
 	list_for_each_entry_reverse(tr, head, tr_list) {
 		if (wbc->nr_to_write <= 0)
 			break;
-		if (gfs2_ail1_start_one(sdp, wbc, tr, &withdraw))
-			goto restart;
+		ret = gfs2_ail1_start_one(sdp, wbc, tr);
+		if (ret) {
+			if (ret == -EBUSY)
+				goto restart;
+			break;
+		}
 	}
+out:
 	spin_unlock(&sdp->sd_ail_lock);
 	blk_finish_plug(&plug);
-	if (withdraw)
-		gfs2_lm_withdraw(sdp, NULL);
+	if (ret) {
+		gfs2_lm(sdp, "gfs2_ail1_start_one (generic_writepages) "
+			"returned: %d\n", ret);
+		gfs2_withdraw(sdp);
+	}
 	trace_gfs2_ail_flush(sdp, wbc, 0);
 }
 
@@ -189,12 +241,13 @@ static void gfs2_ail1_start(struct gfs2_sbd *sdp)
 /**
  * gfs2_ail1_empty_one - Check whether or not a trans in the AIL has been synced
  * @sdp: the filesystem
- * @ai: the AIL entry
+ * @tr: the transaction
+ * @max_revokes: If nonzero, issue revokes for the bd items for written buffers
  *
  */
 
 static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr,
-				bool *withdraw)
+				int *max_revokes)
 {
 	struct gfs2_bufdata *bd, *s;
 	struct buffer_head *bh;
@@ -203,12 +256,32 @@ static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr,
 					 bd_ail_st_list) {
 		bh = bd->bd_bh;
 		gfs2_assert(sdp, bd->bd_tr == tr);
-		if (buffer_busy(bh))
+		/*
+		 * If another process flagged an io error, e.g. writing to the
+		 * journal, error all other bhs and move them off the ail1 to
+		 * prevent a tight loop when unmount tries to flush ail1,
+		 * regardless of whether they're still busy. If no outside
+		 * errors were found and the buffer is busy, move to the next.
+		 * If the ail buffer is not busy and caught an error, flag it
+		 * for others.
+		 */
+		if (!sdp->sd_log_error && buffer_busy(bh))
 			continue;
 		if (!buffer_uptodate(bh) &&
-		    !test_and_set_bit(SDF_AIL1_IO_ERROR, &sdp->sd_flags)) {
+		    !cmpxchg(&sdp->sd_log_error, 0, -EIO)) {
 			gfs2_io_error_bh(sdp, bh);
-			*withdraw = true;
+			gfs2_withdraw_delayed(sdp);
+		}
+		/*
+		 * If we have space for revokes and the bd is no longer on any
+		 * buf list, we can just add a revoke for it immediately and
+		 * avoid having to put it on the ail2 list, where it would need
+		 * to be revoked later.
+		 */
+		if (*max_revokes && list_empty(&bd->bd_list)) {
+			gfs2_add_revoke(sdp, bd);
+			(*max_revokes)--;
+			continue;
 		}
 		list_move(&bd->bd_ail_st_list, &tr->tr_ail2_list);
 	}
@@ -217,20 +290,20 @@ static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr,
 /**
  * gfs2_ail1_empty - Try to empty the ail1 lists
  * @sdp: The superblock
+ * @max_revokes: If non-zero, add revokes where appropriate
  *
  * Tries to empty the ail1 lists, starting with the oldest first
  */
 
-static int gfs2_ail1_empty(struct gfs2_sbd *sdp)
+static int gfs2_ail1_empty(struct gfs2_sbd *sdp, int max_revokes)
 {
 	struct gfs2_trans *tr, *s;
 	int oldest_tr = 1;
 	int ret;
-	bool withdraw = false;
 
 	spin_lock(&sdp->sd_ail_lock);
 	list_for_each_entry_safe_reverse(tr, s, &sdp->sd_ail1_list, tr_list) {
-		gfs2_ail1_empty_one(sdp, tr, &withdraw);
+		gfs2_ail1_empty_one(sdp, tr, &max_revokes);
 		if (list_empty(&tr->tr_ail1_list) && oldest_tr)
 			list_move(&tr->tr_list, &sdp->sd_ail2_list);
 		else
@@ -239,8 +312,10 @@ static int gfs2_ail1_empty(struct gfs2_sbd *sdp)
 	ret = list_empty(&sdp->sd_ail1_list);
 	spin_unlock(&sdp->sd_ail_lock);
 
-	if (withdraw)
-		gfs2_lm_withdraw(sdp, "fatal: I/O error(s)\n");
+	if (test_bit(SDF_WITHDRAWING, &sdp->sd_flags)) {
+		gfs2_lm(sdp, "fatal: I/O error(s)\n");
+		gfs2_withdraw(sdp);
+	}
 
 	return ret;
 }
@@ -268,20 +343,17 @@ static void gfs2_ail1_wait(struct gfs2_sbd *sdp)
 }
 
 /**
- * gfs2_ail2_empty_one - Check whether or not a trans in the AIL has been synced
- * @sdp: the filesystem
- * @ai: the AIL entry
- *
+ * gfs2_ail_empty_tr - empty one of the ail lists for a transaction
  */
 
-static void gfs2_ail2_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
+static void gfs2_ail_empty_tr(struct gfs2_sbd *sdp, struct gfs2_trans *tr,
+			      struct list_head *head)
 {
-	struct list_head *head = &tr->tr_ail2_list;
 	struct gfs2_bufdata *bd;
 
 	while (!list_empty(head)) {
-		bd = list_entry(head->prev, struct gfs2_bufdata,
-				bd_ail_st_list);
+		bd = list_first_entry(head, struct gfs2_bufdata,
+				      bd_ail_st_list);
 		gfs2_assert(sdp, bd->bd_tr == tr);
 		gfs2_remove_from_ail(bd);
 	}
@@ -303,7 +375,7 @@ static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail)
 		if (!rm)
 			continue;
 
-		gfs2_ail2_empty_one(sdp, tr);
+		gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail2_list);
 		list_del(&tr->tr_list);
 		gfs2_assert_warn(sdp, list_empty(&tr->tr_ail1_list));
 		gfs2_assert_warn(sdp, list_empty(&tr->tr_ail2_list));
@@ -469,9 +541,8 @@ static unsigned int calc_reserved(struct gfs2_sbd *sdp)
 		reserved += DIV_ROUND_UP(dbuf, databuf_limit(sdp));
 	}
 
-	if (sdp->sd_log_commited_revoke > 0)
-		reserved += gfs2_struct2blk(sdp, sdp->sd_log_commited_revoke,
-					  sizeof(u64));
+	if (sdp->sd_log_committed_revoke > 0)
+		reserved += gfs2_struct2blk(sdp, sdp->sd_log_committed_revoke);
 	/* One for the overall header */
 	if (reserved)
 		reserved++;
@@ -488,7 +559,7 @@ static unsigned int current_tail(struct gfs2_sbd *sdp)
 	if (list_empty(&sdp->sd_ail1_list)) {
 		tail = sdp->sd_log_head;
 	} else {
-		tr = list_entry(sdp->sd_ail1_list.prev, struct gfs2_trans,
+		tr = list_last_entry(&sdp->sd_ail1_list, struct gfs2_trans,
 				tr_list);
 		tail = tr->tr_first;
 	}
@@ -513,7 +584,7 @@ static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail)
 }
 
 
-static void log_flush_wait(struct gfs2_sbd *sdp)
+void log_flush_wait(struct gfs2_sbd *sdp)
 {
 	DEFINE_WAIT(wait);
 
@@ -550,7 +621,7 @@ static void gfs2_ordered_write(struct gfs2_sbd *sdp)
 	spin_lock(&sdp->sd_ordered_lock);
 	list_sort(NULL, &sdp->sd_log_ordered, &ip_cmp);
 	while (!list_empty(&sdp->sd_log_ordered)) {
-		ip = list_entry(sdp->sd_log_ordered.next, struct gfs2_inode, i_ordered);
+		ip = list_first_entry(&sdp->sd_log_ordered, struct gfs2_inode, i_ordered);
 		if (ip->i_inode.i_mapping->nrpages == 0) {
 			test_and_clear_bit(GIF_ORDERED, &ip->i_flags);
 			list_del(&ip->i_ordered);
@@ -571,7 +642,7 @@ static void gfs2_ordered_wait(struct gfs2_sbd *sdp)
 
 	spin_lock(&sdp->sd_ordered_lock);
 	while (!list_empty(&sdp->sd_log_ordered)) {
-		ip = list_entry(sdp->sd_log_ordered.next, struct gfs2_inode, i_ordered);
+		ip = list_first_entry(&sdp->sd_log_ordered, struct gfs2_inode, i_ordered);
 		list_del(&ip->i_ordered);
 		WARN_ON(!test_and_clear_bit(GIF_ORDERED, &ip->i_flags));
 		if (ip->i_inode.i_mapping->nrpages == 0)
@@ -598,38 +669,43 @@ void gfs2_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd)
 	struct buffer_head *bh = bd->bd_bh;
 	struct gfs2_glock *gl = bd->bd_gl;
 
+	sdp->sd_log_num_revoke++;
+	if (atomic_inc_return(&gl->gl_revokes) == 1)
+		gfs2_glock_hold(gl);
 	bh->b_private = NULL;
 	bd->bd_blkno = bh->b_blocknr;
 	gfs2_remove_from_ail(bd); /* drops ref on bh */
 	bd->bd_bh = NULL;
-	sdp->sd_log_num_revoke++;
-	if (atomic_inc_return(&gl->gl_revokes) == 1)
-		gfs2_glock_hold(gl);
 	set_bit(GLF_LFLUSH, &gl->gl_flags);
 	list_add(&bd->bd_list, &sdp->sd_log_revokes);
 }
 
+void gfs2_glock_remove_revoke(struct gfs2_glock *gl)
+{
+	if (atomic_dec_return(&gl->gl_revokes) == 0) {
+		clear_bit(GLF_LFLUSH, &gl->gl_flags);
+		gfs2_glock_queue_put(gl);
+	}
+}
+
+/**
+ * gfs2_write_revokes - Add as many revokes to the system transaction as we can
+ * @sdp: The GFS2 superblock
+ *
+ * Our usual strategy is to defer writing revokes as much as we can in the hope
+ * that we'll eventually overwrite the journal, which will make those revokes
+ * go away.  This changes when we flush the log: at that point, there will
+ * likely be some left-over space in the last revoke block of that transaction.
+ * We can fill that space with additional revokes for blocks that have already
+ * been written back.  This will basically come at no cost now, and will save
+ * us from having to keep track of those blocks on the AIL2 list later.
+ */
 void gfs2_write_revokes(struct gfs2_sbd *sdp)
 {
-	struct gfs2_trans *tr;
-	struct gfs2_bufdata *bd, *tmp;
-	int have_revokes = 0;
+	/* number of revokes we still have room for */
 	int max_revokes = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / sizeof(u64);
 
-	gfs2_ail1_empty(sdp);
-	spin_lock(&sdp->sd_ail_lock);
-	list_for_each_entry_reverse(tr, &sdp->sd_ail1_list, tr_list) {
-		list_for_each_entry(bd, &tr->tr_ail2_list, bd_ail_st_list) {
-			if (list_empty(&bd->bd_list)) {
-				have_revokes = 1;
-				goto done;
-			}
-		}
-	}
-done:
-	spin_unlock(&sdp->sd_ail_lock);
-	if (have_revokes == 0)
-		return;
+	gfs2_log_lock(sdp);
 	while (sdp->sd_log_num_revoke > max_revokes)
 		max_revokes += (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header)) / sizeof(u64);
 	max_revokes -= sdp->sd_log_num_revoke;
@@ -640,20 +716,7 @@ done:
 		if (!sdp->sd_log_blks_reserved)
 			atomic_dec(&sdp->sd_log_blks_free);
 	}
-	gfs2_log_lock(sdp);
-	spin_lock(&sdp->sd_ail_lock);
-	list_for_each_entry_reverse(tr, &sdp->sd_ail1_list, tr_list) {
-		list_for_each_entry_safe(bd, tmp, &tr->tr_ail2_list, bd_ail_st_list) {
-			if (max_revokes == 0)
-				goto out_of_blocks;
-			if (!list_empty(&bd->bd_list))
-				continue;
-			gfs2_add_revoke(sdp, bd);
-			max_revokes--;
-		}
-	}
-out_of_blocks:
-	spin_unlock(&sdp->sd_ail_lock);
+	gfs2_ail1_empty(sdp, max_revokes);
 	gfs2_log_unlock(sdp);
 
 	if (!sdp->sd_log_num_revoke) {
@@ -682,12 +745,16 @@ void gfs2_write_log_header(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
 {
 	struct gfs2_log_header *lh;
 	u32 hash, crc;
-	struct page *page = mempool_alloc(gfs2_page_pool, GFP_NOIO);
+	struct page *page;
 	struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;
 	struct timespec64 tv;
 	struct super_block *sb = sdp->sd_vfs;
 	u64 dblock;
 
+	if (gfs2_withdrawn(sdp))
+		goto out;
+
+	page = mempool_alloc(gfs2_page_pool, GFP_NOIO);
 	lh = page_address(page);
 	clear_page(lh);
 
@@ -707,7 +774,7 @@ void gfs2_write_log_header(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
 	lh->lh_nsec = cpu_to_be32(tv.tv_nsec);
 	lh->lh_sec = cpu_to_be64(tv.tv_sec);
 	if (!list_empty(&jd->extent_list))
-		dblock = gfs2_log_bmap(sdp);
+		dblock = gfs2_log_bmap(jd, lblock);
 	else {
 		int ret = gfs2_lblk_to_dblk(jd->jd_inode, lblock, &dblock);
 		if (gfs2_assert_withdraw(sdp, ret == 0))
@@ -740,6 +807,7 @@ void gfs2_write_log_header(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
 
 	gfs2_log_write(sdp, page, sb->s_blocksize, 0, dblock);
 	gfs2_log_submit_bio(&sdp->sd_log_bio, REQ_OP_WRITE | op_flags);
+out:
 	log_flush_wait(sdp);
 }
 
@@ -768,12 +836,47 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags)
 	sdp->sd_log_idle = (tail == sdp->sd_log_flush_head);
 	gfs2_write_log_header(sdp, sdp->sd_jdesc, sdp->sd_log_sequence++, tail,
 			      sdp->sd_log_flush_head, flags, op_flags);
+	gfs2_log_incr_head(sdp);
 
 	if (sdp->sd_log_tail != tail)
 		log_pull_tail(sdp, tail);
 }
 
 /**
+ * ail_drain - drain the ail lists after a withdraw
+ * @sdp: Pointer to GFS2 superblock
+ */
+static void ail_drain(struct gfs2_sbd *sdp)
+{
+	struct gfs2_trans *tr;
+
+	spin_lock(&sdp->sd_ail_lock);
+	/*
+	 * For transactions on the sd_ail1_list we need to drain both the
+	 * ail1 and ail2 lists. That's because function gfs2_ail1_start_one
+	 * (temporarily) moves items from its tr_ail1 list to tr_ail2 list
+	 * before revokes are sent for that block. Items on the sd_ail2_list
+	 * should have already gotten beyond that point, so no need.
+	 */
+	while (!list_empty(&sdp->sd_ail1_list)) {
+		tr = list_first_entry(&sdp->sd_ail1_list, struct gfs2_trans,
+				      tr_list);
+		gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail1_list);
+		gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail2_list);
+		list_del(&tr->tr_list);
+		kfree(tr);
+	}
+	while (!list_empty(&sdp->sd_ail2_list)) {
+		tr = list_first_entry(&sdp->sd_ail2_list, struct gfs2_trans,
+				      tr_list);
+		gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail2_list);
+		list_del(&tr->tr_list);
+		kfree(tr);
+	}
+	spin_unlock(&sdp->sd_ail_lock);
+}
+
+/**
  * gfs2_log_flush - flush incore transaction(s)
  * @sdp: the filesystem
  * @gl: The glock structure to flush.  If NULL, flush the whole incore log
@@ -783,11 +886,18 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags)
 
 void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags)
 {
-	struct gfs2_trans *tr;
+	struct gfs2_trans *tr = NULL;
 	enum gfs2_freeze_state state = atomic_read(&sdp->sd_freeze_state);
 
 	down_write(&sdp->sd_log_flush_lock);
 
+	/*
+	 * Do this check while holding the log_flush_lock to prevent new
+	 * buffers from being added to the ail via gfs2_pin()
+	 */
+	if (gfs2_withdrawn(sdp))
+		goto out;
+
 	/* Log might have been flushed while we waited for the flush lock */
 	if (gl && !test_bit(GLF_LFLUSH, &gl->gl_flags)) {
 		up_write(&sdp->sd_log_flush_lock);
@@ -806,17 +916,27 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags)
 		INIT_LIST_HEAD(&tr->tr_ail2_list);
 		tr->tr_first = sdp->sd_log_flush_head;
 		if (unlikely (state == SFS_FROZEN))
-			gfs2_assert_withdraw(sdp, !tr->tr_num_buf_new && !tr->tr_num_databuf_new);
+			if (gfs2_assert_withdraw_delayed(sdp,
+			       !tr->tr_num_buf_new && !tr->tr_num_databuf_new))
+				goto out;
 	}
 
 	if (unlikely(state == SFS_FROZEN))
-		gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke);
-	gfs2_assert_withdraw(sdp,
-			sdp->sd_log_num_revoke == sdp->sd_log_commited_revoke);
+		if (gfs2_assert_withdraw_delayed(sdp, !sdp->sd_log_num_revoke))
+			goto out;
+	if (gfs2_assert_withdraw_delayed(sdp,
+			sdp->sd_log_num_revoke == sdp->sd_log_committed_revoke))
+		goto out;
 
 	gfs2_ordered_write(sdp);
+	if (gfs2_withdrawn(sdp))
+		goto out;
 	lops_before_commit(sdp, tr);
+	if (gfs2_withdrawn(sdp))
+		goto out;
 	gfs2_log_submit_bio(&sdp->sd_log_bio, REQ_OP_WRITE);
+	if (gfs2_withdrawn(sdp))
+		goto out;
 
 	if (sdp->sd_log_head != sdp->sd_log_flush_head) {
 		log_flush_wait(sdp);
@@ -826,12 +946,14 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags)
 		trace_gfs2_log_blocks(sdp, -1);
 		log_write_header(sdp, flags);
 	}
+	if (gfs2_withdrawn(sdp))
+		goto out;
 	lops_after_commit(sdp, tr);
 
 	gfs2_log_lock(sdp);
 	sdp->sd_log_head = sdp->sd_log_flush_head;
 	sdp->sd_log_blks_reserved = 0;
-	sdp->sd_log_commited_revoke = 0;
+	sdp->sd_log_committed_revoke = 0;
 
 	spin_lock(&sdp->sd_ail_lock);
 	if (tr && !list_empty(&tr->tr_ail1_list)) {
@@ -846,9 +968,11 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags)
 			for (;;) {
 				gfs2_ail1_start(sdp);
 				gfs2_ail1_wait(sdp);
-				if (gfs2_ail1_empty(sdp))
+				if (gfs2_ail1_empty(sdp, 0))
 					break;
 			}
+			if (gfs2_withdrawn(sdp))
+				goto out;
 			atomic_dec(&sdp->sd_log_blks_free); /* Adjust for unreserved buffer */
 			trace_gfs2_log_blocks(sdp, -1);
 			log_write_header(sdp, flags);
@@ -861,6 +985,12 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags)
 			atomic_set(&sdp->sd_freeze_state, SFS_FROZEN);
 	}
 
+out:
+	if (gfs2_withdrawn(sdp)) {
+		ail_drain(sdp); /* frees all transactions */
+		tr = NULL;
+	}
+
 	trace_gfs2_log_flush(sdp, 0, flags);
 	up_write(&sdp->sd_log_flush_lock);
 
@@ -882,6 +1012,7 @@ static void gfs2_merge_trans(struct gfs2_trans *old, struct gfs2_trans *new)
 	old->tr_num_buf_rm	+= new->tr_num_buf_rm;
 	old->tr_num_databuf_rm	+= new->tr_num_databuf_rm;
 	old->tr_num_revoke	+= new->tr_num_revoke;
+	old->tr_num_revoke_rm	+= new->tr_num_revoke_rm;
 
 	list_splice_tail_init(&new->tr_databuf, &old->tr_databuf);
 	list_splice_tail_init(&new->tr_buf, &old->tr_buf);
@@ -903,7 +1034,7 @@ static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 		set_bit(TR_ATTACHED, &tr->tr_flags);
 	}
 
-	sdp->sd_log_commited_revoke += tr->tr_num_revoke;
+	sdp->sd_log_committed_revoke += tr->tr_num_revoke - tr->tr_num_revoke_rm;
 	reserved = calc_reserved(sdp);
 	maxres = sdp->sd_log_blks_reserved + tr->tr_reserved;
 	gfs2_assert_withdraw(sdp, maxres >= reserved);
@@ -948,7 +1079,7 @@ void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
  *
  */
 
-void gfs2_log_shutdown(struct gfs2_sbd *sdp)
+static void gfs2_log_shutdown(struct gfs2_sbd *sdp)
 {
 	gfs2_assert_withdraw(sdp, !sdp->sd_log_blks_reserved);
 	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke);
@@ -1000,18 +1131,24 @@ int gfs2_logd(void *data)
 
 	while (!kthread_should_stop()) {
 
+		if (gfs2_withdrawn(sdp)) {
+			msleep_interruptible(HZ);
+			continue;
+		}
 		/* Check for errors writing to the journal */
 		if (sdp->sd_log_error) {
-			gfs2_lm_withdraw(sdp,
-					 "GFS2: fsid=%s: error %d: "
-					 "withdrawing the file system to "
-					 "prevent further damage.\n",
-					 sdp->sd_fsname, sdp->sd_log_error);
+			gfs2_lm(sdp,
+				"GFS2: fsid=%s: error %d: "
+				"withdrawing the file system to "
+				"prevent further damage.\n",
+				sdp->sd_fsname, sdp->sd_log_error);
+			gfs2_withdraw(sdp);
+			continue;
 		}
 
 		did_flush = false;
 		if (gfs2_jrnl_flush_reqd(sdp) || t == 0) {
-			gfs2_ail1_empty(sdp);
+			gfs2_ail1_empty(sdp, 0);
 			gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL |
 				       GFS2_LFC_LOGD_JFLUSH_REQD);
 			did_flush = true;
@@ -1020,7 +1157,7 @@ int gfs2_logd(void *data)
 		if (gfs2_ail_flush_reqd(sdp)) {
 			gfs2_ail1_start(sdp);
 			gfs2_ail1_wait(sdp);
-			gfs2_ail1_empty(sdp);
+			gfs2_ail1_empty(sdp, 0);
 			gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL |
 				       GFS2_LFC_LOGD_AIL_FLUSH_REQD);
 			did_flush = true;
diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h
index 2315fca47a2b..c1cd6ae17659 100644
--- a/fs/gfs2/log.h
+++ b/fs/gfs2/log.h
@@ -60,9 +60,9 @@ static inline void gfs2_ordered_add_inode(struct gfs2_inode *ip)
 		spin_unlock(&sdp->sd_ordered_lock);
 	}
 }
+
 extern void gfs2_ordered_del_inode(struct gfs2_inode *ip);
-extern unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct,
-			    unsigned int ssize);
+extern unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct);
 
 extern void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks);
 extern int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks);
@@ -73,10 +73,11 @@ extern void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl,
 			   u32 type);
 extern void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans);
 extern void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc);
+extern void log_flush_wait(struct gfs2_sbd *sdp);
 
-extern void gfs2_log_shutdown(struct gfs2_sbd *sdp);
 extern int gfs2_logd(void *data);
 extern void gfs2_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd);
+extern void gfs2_glock_remove_revoke(struct gfs2_glock *gl);
 extern void gfs2_write_revokes(struct gfs2_sbd *sdp);
 
 #endif /* __LOG_DOT_H__ */
diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c
index 5b17979af539..cb2a11b458c6 100644
--- a/fs/gfs2/lops.c
+++ b/fs/gfs2/lops.c
@@ -129,7 +129,7 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh,
 	atomic_dec(&sdp->sd_log_pinned);
 }
 
-static void gfs2_log_incr_head(struct gfs2_sbd *sdp)
+void gfs2_log_incr_head(struct gfs2_sbd *sdp)
 {
 	BUG_ON((sdp->sd_log_flush_head == sdp->sd_log_tail) &&
 	       (sdp->sd_log_flush_head != sdp->sd_log_head));
@@ -138,18 +138,13 @@ static void gfs2_log_incr_head(struct gfs2_sbd *sdp)
 		sdp->sd_log_flush_head = 0;
 }
 
-u64 gfs2_log_bmap(struct gfs2_sbd *sdp)
+u64 gfs2_log_bmap(struct gfs2_jdesc *jd, unsigned int lblock)
 {
-	unsigned int lbn = sdp->sd_log_flush_head;
 	struct gfs2_journal_extent *je;
-	u64 block;
 
-	list_for_each_entry(je, &sdp->sd_jdesc->extent_list, list) {
-		if ((lbn >= je->lblock) && (lbn < (je->lblock + je->blocks))) {
-			block = je->dblock + lbn - je->lblock;
-			gfs2_log_incr_head(sdp);
-			return block;
-		}
+	list_for_each_entry(je, &jd->extent_list, list) {
+		if (lblock >= je->lblock && lblock < je->lblock + je->blocks)
+			return je->dblock + lblock - je->lblock;
 	}
 
 	return -1;
@@ -208,8 +203,12 @@ static void gfs2_end_log_write(struct bio *bio)
 	struct bvec_iter_all iter_all;
 
 	if (bio->bi_status) {
-		fs_err(sdp, "Error %d writing to journal, jid=%u\n",
-		       bio->bi_status, sdp->sd_jdesc->jd_jid);
+		if (!cmpxchg(&sdp->sd_log_error, 0, (int)bio->bi_status))
+			fs_err(sdp, "Error %d writing to journal, jid=%u\n",
+			       bio->bi_status, sdp->sd_jdesc->jd_jid);
+		gfs2_withdraw_delayed(sdp);
+		/* prevent more writes to the journal */
+		clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
 		wake_up(&sdp->sd_logd_waitq);
 	}
 
@@ -264,7 +263,7 @@ static struct bio *gfs2_log_alloc_bio(struct gfs2_sbd *sdp, u64 blkno,
 	struct super_block *sb = sdp->sd_vfs;
 	struct bio *bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES);
 
-	bio->bi_iter.bi_sector = blkno * (sb->s_blocksize >> 9);
+	bio->bi_iter.bi_sector = blkno << sdp->sd_fsb2bb_shift;
 	bio_set_dev(bio, sb->s_bdev);
 	bio->bi_end_io = end_io;
 	bio->bi_private = sdp;
@@ -351,8 +350,11 @@ void gfs2_log_write(struct gfs2_sbd *sdp, struct page *page,
 
 static void gfs2_log_write_bh(struct gfs2_sbd *sdp, struct buffer_head *bh)
 {
-	gfs2_log_write(sdp, bh->b_page, bh->b_size, bh_offset(bh),
-		       gfs2_log_bmap(sdp));
+	u64 dblock;
+
+	dblock = gfs2_log_bmap(sdp->sd_jdesc, sdp->sd_log_flush_head);
+	gfs2_log_incr_head(sdp);
+	gfs2_log_write(sdp, bh->b_page, bh->b_size, bh_offset(bh), dblock);
 }
 
 /**
@@ -369,8 +371,11 @@ static void gfs2_log_write_bh(struct gfs2_sbd *sdp, struct buffer_head *bh)
 void gfs2_log_write_page(struct gfs2_sbd *sdp, struct page *page)
 {
 	struct super_block *sb = sdp->sd_vfs;
-	gfs2_log_write(sdp, page, sb->s_blocksize, 0,
-		       gfs2_log_bmap(sdp));
+	u64 dblock;
+
+	dblock = gfs2_log_bmap(sdp->sd_jdesc, sdp->sd_log_flush_head);
+	gfs2_log_incr_head(sdp);
+	gfs2_log_write(sdp, page, sb->s_blocksize, 0, dblock);
 }
 
 /**
@@ -421,7 +426,7 @@ static bool gfs2_jhead_pg_srch(struct gfs2_jdesc *jd,
 
 	for (offset = 0; offset < PAGE_SIZE; offset += sdp->sd_sb.sb_bsize) {
 		if (!__get_log_header(sdp, kaddr + offset, 0, &lh)) {
-			if (lh.lh_sequence > head->lh_sequence)
+			if (lh.lh_sequence >= head->lh_sequence)
 				*head = lh;
 			else {
 				ret = true;
@@ -471,6 +476,20 @@ static void gfs2_jhead_process_page(struct gfs2_jdesc *jd, unsigned long index,
 	put_page(page); /* Once more for find_or_create_page */
 }
 
+static struct bio *gfs2_chain_bio(struct bio *prev, unsigned int nr_iovecs)
+{
+	struct bio *new;
+
+	new = bio_alloc(GFP_NOIO, nr_iovecs);
+	bio_copy_dev(new, prev);
+	new->bi_iter.bi_sector = bio_end_sector(prev);
+	new->bi_opf = prev->bi_opf;
+	new->bi_write_hint = prev->bi_write_hint;
+	bio_chain(new, prev);
+	submit_bio(prev);
+	return new;
+}
+
 /**
  * gfs2_find_jhead - find the head of a log
  * @jd: The journal descriptor
@@ -487,10 +506,10 @@ int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head,
 	struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
 	struct address_space *mapping = jd->jd_inode->i_mapping;
 	unsigned int block = 0, blocks_submitted = 0, blocks_read = 0;
-	unsigned int bsize = sdp->sd_sb.sb_bsize;
+	unsigned int bsize = sdp->sd_sb.sb_bsize, off;
 	unsigned int bsize_shift = sdp->sd_sb.sb_bsize_shift;
 	unsigned int shift = PAGE_SHIFT - bsize_shift;
-	unsigned int readhead_blocks = BIO_MAX_PAGES << shift;
+	unsigned int max_blocks = 2 * 1024 * 1024 >> bsize_shift;
 	struct gfs2_journal_extent *je;
 	int sz, ret = 0;
 	struct bio *bio = NULL;
@@ -504,9 +523,9 @@ int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head,
 
 	since = filemap_sample_wb_err(mapping);
 	list_for_each_entry(je, &jd->extent_list, list) {
-		for (; block < je->lblock + je->blocks; block++) {
-			u64 dblock;
+		u64 dblock = je->dblock;
 
+		for (; block < je->lblock + je->blocks; block++, dblock++) {
 			if (!page) {
 				page = find_or_create_page(mapping,
 						block >> shift, GFP_NOFS);
@@ -515,35 +534,41 @@ int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head,
 					done = true;
 					goto out;
 				}
+				off = 0;
 			}
 
-			if (bio) {
-				unsigned int off;
-
-				off = (block << bsize_shift) & ~PAGE_MASK;
-				sz = bio_add_page(bio, page, bsize, off);
-				if (sz == bsize) { /* block added */
-					if (off + bsize == PAGE_SIZE) {
-						page = NULL;
-						goto page_added;
-					}
-					continue;
+			if (bio && (off || block < blocks_submitted + max_blocks)) {
+				sector_t sector = dblock << sdp->sd_fsb2bb_shift;
+
+				if (bio_end_sector(bio) == sector) {
+					sz = bio_add_page(bio, page, bsize, off);
+					if (sz == bsize)
+						goto block_added;
+				}
+				if (off) {
+					unsigned int blocks =
+						(PAGE_SIZE - off) >> bsize_shift;
+
+					bio = gfs2_chain_bio(bio, blocks);
+					goto add_block_to_new_bio;
 				}
-				blocks_submitted = block + 1;
+			}
+
+			if (bio) {
+				blocks_submitted = block;
 				submit_bio(bio);
-				bio = NULL;
 			}
 
-			dblock = je->dblock + (block - je->lblock);
 			bio = gfs2_log_alloc_bio(sdp, dblock, gfs2_end_log_read);
 			bio->bi_opf = REQ_OP_READ;
-			sz = bio_add_page(bio, page, bsize, 0);
-			gfs2_assert_warn(sdp, sz == bsize);
-			if (bsize == PAGE_SIZE)
+add_block_to_new_bio:
+			sz = bio_add_page(bio, page, bsize, off);
+			BUG_ON(sz != bsize);
+block_added:
+			off += bsize;
+			if (off == PAGE_SIZE)
 				page = NULL;
-
-page_added:
-			if (blocks_submitted < blocks_read + readhead_blocks) {
+			if (blocks_submitted <= blocks_read + max_blocks) {
 				/* Keep at least one bio in flight */
 				continue;
 			}
@@ -709,7 +734,7 @@ static void buf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 
 	head = &tr->tr_buf;
 	while (!list_empty(head)) {
-		bd = list_entry(head->next, struct gfs2_bufdata, bd_list);
+		bd = list_first_entry(head, struct gfs2_bufdata, bd_list);
 		list_del_init(&bd->bd_list);
 		gfs2_unpin(sdp, bd->bd_bh, tr);
 	}
@@ -845,7 +870,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 	if (!sdp->sd_log_num_revoke)
 		return;
 
-	length = gfs2_struct2blk(sdp, sdp->sd_log_num_revoke, sizeof(u64));
+	length = gfs2_struct2blk(sdp, sdp->sd_log_num_revoke);
 	page = gfs2_get_log_desc(sdp, GFS2_LOG_DESC_REVOKE, length, sdp->sd_log_num_revoke);
 	offset = sizeof(struct gfs2_log_descriptor);
 
@@ -879,13 +904,10 @@ static void revoke_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 	struct gfs2_glock *gl;
 
 	while (!list_empty(head)) {
-		bd = list_entry(head->next, struct gfs2_bufdata, bd_list);
+		bd = list_first_entry(head, struct gfs2_bufdata, bd_list);
 		list_del_init(&bd->bd_list);
 		gl = bd->bd_gl;
-		if (atomic_dec_return(&gl->gl_revokes) == 0) {
-			clear_bit(GLF_LFLUSH, &gl->gl_flags);
-			gfs2_glock_queue_put(gl);
-		}
+		gfs2_glock_remove_revoke(gl);
 		kmem_cache_free(gfs2_bufdata_cachep, bd);
 	}
 }
@@ -1061,7 +1083,7 @@ static void databuf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 
 	head = &tr->tr_databuf;
 	while (!list_empty(head)) {
-		bd = list_entry(head->next, struct gfs2_bufdata, bd_list);
+		bd = list_first_entry(head, struct gfs2_bufdata, bd_list);
 		list_del_init(&bd->bd_list);
 		gfs2_unpin(sdp, bd->bd_bh, tr);
 	}
diff --git a/fs/gfs2/lops.h b/fs/gfs2/lops.h
index 9c059957a733..9c5e4e491e03 100644
--- a/fs/gfs2/lops.h
+++ b/fs/gfs2/lops.h
@@ -18,7 +18,8 @@
 	 ~(2 * sizeof(__be64) - 1))
 
 extern const struct gfs2_log_operations *gfs2_log_ops[];
-extern u64 gfs2_log_bmap(struct gfs2_sbd *sdp);
+extern void gfs2_log_incr_head(struct gfs2_sbd *sdp);
+extern u64 gfs2_log_bmap(struct gfs2_jdesc *jd, unsigned int lbn);
 extern void gfs2_log_write(struct gfs2_sbd *sdp, struct page *page,
 			   unsigned size, unsigned offset, u64 blkno);
 extern void gfs2_log_write_page(struct gfs2_sbd *sdp, struct page *page);
diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c
index 662ef36c1874..9856cc2e0795 100644
--- a/fs/gfs2/meta_io.c
+++ b/fs/gfs2/meta_io.c
@@ -251,7 +251,8 @@ int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags,
 	struct buffer_head *bh, *bhs[2];
 	int num = 0;
 
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags))) {
+	if (unlikely(gfs2_withdrawn(sdp)) &&
+	    (!sdp->sd_jdesc || gl != sdp->sd_jinode_gl)) {
 		*bhp = NULL;
 		return -EIO;
 	}
@@ -309,7 +310,7 @@ int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags,
 
 int gfs2_meta_wait(struct gfs2_sbd *sdp, struct buffer_head *bh)
 {
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags)))
+	if (unlikely(gfs2_withdrawn(sdp)))
 		return -EIO;
 
 	wait_on_buffer(bh);
@@ -320,7 +321,7 @@ int gfs2_meta_wait(struct gfs2_sbd *sdp, struct buffer_head *bh)
 			gfs2_io_error_bh_wd(sdp, bh);
 		return -EIO;
 	}
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags)))
+	if (unlikely(gfs2_withdrawn(sdp)))
 		return -EIO;
 
 	return 0;
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 18daf494abab..e2b69ffcc6a8 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -298,6 +298,8 @@ static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent)
 			  sizeof(struct gfs2_dinode)) / sizeof(u64);
 	sdp->sd_inptrs = (sdp->sd_sb.sb_bsize -
 			  sizeof(struct gfs2_meta_header)) / sizeof(u64);
+	sdp->sd_ldptrs = (sdp->sd_sb.sb_bsize -
+			  sizeof(struct gfs2_log_descriptor)) / sizeof(u64);
 	sdp->sd_jbsize = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header);
 	sdp->sd_hash_bsize = sdp->sd_sb.sb_bsize / 2;
 	sdp->sd_hash_bsize_shift = sdp->sd_sb.sb_bsize_shift - 1;
@@ -550,6 +552,8 @@ static int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh)
 	mutex_lock(&sdp->sd_jindex_mutex);
 
 	for (;;) {
+		struct gfs2_inode *jip;
+
 		error = gfs2_glock_nq_init(dip->i_gl, LM_ST_SHARED, 0, ji_gh);
 		if (error)
 			break;
@@ -589,6 +593,8 @@ static int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh)
 
 		spin_lock(&sdp->sd_jindex_spin);
 		jd->jd_jid = sdp->sd_journals++;
+		jip = GFS2_I(jd->jd_inode);
+		jd->jd_no_addr = jip->i_no_addr;
 		list_add_tail(&jd->jd_list, &sdp->sd_jindex_list);
 		spin_unlock(&sdp->sd_jindex_spin);
 	}
@@ -598,48 +604,6 @@ static int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh)
 	return error;
 }
 
-/**
- * check_journal_clean - Make sure a journal is clean for a spectator mount
- * @sdp: The GFS2 superblock
- * @jd: The journal descriptor
- *
- * Returns: 0 if the journal is clean or locked, else an error
- */
-static int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd)
-{
-	int error;
-	struct gfs2_holder j_gh;
-	struct gfs2_log_header_host head;
-	struct gfs2_inode *ip;
-
-	ip = GFS2_I(jd->jd_inode);
-	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_NOEXP |
-				   GL_EXACT | GL_NOCACHE, &j_gh);
-	if (error) {
-		fs_err(sdp, "Error locking journal for spectator mount.\n");
-		return -EPERM;
-	}
-	error = gfs2_jdesc_check(jd);
-	if (error) {
-		fs_err(sdp, "Error checking journal for spectator mount.\n");
-		goto out_unlock;
-	}
-	error = gfs2_find_jhead(jd, &head, false);
-	if (error) {
-		fs_err(sdp, "Error parsing journal for spectator mount.\n");
-		goto out_unlock;
-	}
-	if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) {
-		error = -EPERM;
-		fs_err(sdp, "jid=%u: Journal is dirty, so the first mounter "
-		       "must not be a spectator.\n", jd->jd_jid);
-	}
-
-out_unlock:
-	gfs2_glock_dq_uninit(&j_gh);
-	return error;
-}
-
 static int init_journal(struct gfs2_sbd *sdp, int undo)
 {
 	struct inode *master = d_inode(sdp->sd_master_dir);
@@ -692,7 +656,8 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 
 		error = gfs2_glock_nq_num(sdp, sdp->sd_lockstruct.ls_jid,
 					  &gfs2_journal_glops,
-					  LM_ST_EXCLUSIVE, LM_FLAG_NOEXP,
+					  LM_ST_EXCLUSIVE,
+					  LM_FLAG_NOEXP | GL_NOCACHE,
 					  &sdp->sd_journal_gh);
 		if (error) {
 			fs_err(sdp, "can't acquire journal glock: %d\n", error);
@@ -700,6 +665,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 		}
 
 		ip = GFS2_I(sdp->sd_jdesc->jd_inode);
+		sdp->sd_jinode_gl = ip->i_gl;
 		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED,
 					   LM_FLAG_NOEXP | GL_EXACT | GL_NOCACHE,
 					   &sdp->sd_jinode_gh);
@@ -730,7 +696,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 			struct gfs2_jdesc *jd = gfs2_jdesc_find(sdp, x);
 
 			if (sdp->sd_args.ar_spectator) {
-				error = check_journal_clean(sdp, jd);
+				error = check_journal_clean(sdp, jd, true);
 				if (error)
 					goto fail_jinode_gh;
 				continue;
@@ -760,10 +726,13 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 	return 0;
 
 fail_jinode_gh:
-	if (!sdp->sd_args.ar_spectator)
+	/* A withdraw may have done dq/uninit so now we need to check it */
+	if (!sdp->sd_args.ar_spectator &&
+	    gfs2_holder_initialized(&sdp->sd_jinode_gh))
 		gfs2_glock_dq_uninit(&sdp->sd_jinode_gh);
 fail_journal_gh:
-	if (!sdp->sd_args.ar_spectator)
+	if (!sdp->sd_args.ar_spectator &&
+	    gfs2_holder_initialized(&sdp->sd_journal_gh))
 		gfs2_glock_dq_uninit(&sdp->sd_journal_gh);
 fail_jindex:
 	gfs2_jindex_free(sdp);
@@ -1006,8 +975,7 @@ hostdata_error:
 void gfs2_lm_unmount(struct gfs2_sbd *sdp)
 {
 	const struct lm_lockops *lm = sdp->sd_lockstruct.ls_ops;
-	if (likely(!test_bit(SDF_WITHDRAWN, &sdp->sd_flags)) &&
-	    lm->lm_unmount)
+	if (likely(!gfs2_withdrawn(sdp)) && lm->lm_unmount)
 		lm->lm_unmount(sdp);
 }
 
@@ -1251,6 +1219,7 @@ enum gfs2_param {
 	Opt_upgrade,
 	Opt_acl,
 	Opt_quota,
+	Opt_quota_flag,
 	Opt_suiddir,
 	Opt_data,
 	Opt_meta,
@@ -1265,17 +1234,11 @@ enum gfs2_param {
 	Opt_loccookie,
 };
 
-enum opt_quota {
-	Opt_quota_unset = 0,
-	Opt_quota_off,
-	Opt_quota_account,
-	Opt_quota_on,
-};
-
-static const unsigned int opt_quota_values[] = {
-	[Opt_quota_off]     = GFS2_QUOTA_OFF,
-	[Opt_quota_account] = GFS2_QUOTA_ACCOUNT,
-	[Opt_quota_on]      = GFS2_QUOTA_ON,
+static const struct constant_table gfs2_param_quota[] = {
+	{"off",        GFS2_QUOTA_OFF},
+	{"account",    GFS2_QUOTA_ACCOUNT},
+	{"on",         GFS2_QUOTA_ON},
+	{}
 };
 
 enum opt_data {
@@ -1283,12 +1246,24 @@ enum opt_data {
 	Opt_data_ordered   = GFS2_DATA_ORDERED,
 };
 
+static const struct constant_table gfs2_param_data[] = {
+	{"writeback",  Opt_data_writeback },
+	{"ordered",    Opt_data_ordered },
+	{}
+};
+
 enum opt_errors {
 	Opt_errors_withdraw = GFS2_ERRORS_WITHDRAW,
 	Opt_errors_panic    = GFS2_ERRORS_PANIC,
 };
 
-static const struct fs_parameter_spec gfs2_param_specs[] = {
+static const struct constant_table gfs2_param_errors[] = {
+	{"withdraw",   Opt_errors_withdraw },
+	{"panic",      Opt_errors_panic },
+	{}
+};
+
+static const struct fs_parameter_spec gfs2_fs_parameters[] = {
 	fsparam_string ("lockproto",          Opt_lockproto),
 	fsparam_string ("locktable",          Opt_locktable),
 	fsparam_string ("hostdata",           Opt_hostdata),
@@ -1301,11 +1276,11 @@ static const struct fs_parameter_spec gfs2_param_specs[] = {
 	fsparam_flag   ("upgrade",            Opt_upgrade),
 	fsparam_flag_no("acl",                Opt_acl),
 	fsparam_flag_no("suiddir",            Opt_suiddir),
-	fsparam_enum   ("data",               Opt_data),
+	fsparam_enum   ("data",               Opt_data, gfs2_param_data),
 	fsparam_flag   ("meta",               Opt_meta),
 	fsparam_flag_no("discard",            Opt_discard),
 	fsparam_s32    ("commit",             Opt_commit),
-	fsparam_enum   ("errors",             Opt_errors),
+	fsparam_enum   ("errors",             Opt_errors, gfs2_param_errors),
 	fsparam_s32    ("statfs_quantum",     Opt_statfs_quantum),
 	fsparam_s32    ("statfs_percent",     Opt_statfs_percent),
 	fsparam_s32    ("quota_quantum",      Opt_quota_quantum),
@@ -1313,27 +1288,11 @@ static const struct fs_parameter_spec gfs2_param_specs[] = {
 	fsparam_flag_no("rgrplvb",            Opt_rgrplvb),
 	fsparam_flag_no("loccookie",          Opt_loccookie),
 	/* quota can be a flag or an enum so it gets special treatment */
-	__fsparam(fs_param_is_enum, "quota", Opt_quota, fs_param_neg_with_no|fs_param_v_optional),
-	{}
-};
-
-static const struct fs_parameter_enum gfs2_param_enums[] = {
-	{ Opt_quota,    "off",        Opt_quota_off },
-	{ Opt_quota,    "account",    Opt_quota_account },
-	{ Opt_quota,    "on",         Opt_quota_on },
-	{ Opt_data,     "writeback",  Opt_data_writeback },
-	{ Opt_data,     "ordered",    Opt_data_ordered },
-	{ Opt_errors,   "withdraw",   Opt_errors_withdraw },
-	{ Opt_errors,   "panic",      Opt_errors_panic },
+	fsparam_flag_no("quota",	      Opt_quota_flag),
+	fsparam_enum("quota",		      Opt_quota, gfs2_param_quota),
 	{}
 };
 
-const struct fs_parameter_description gfs2_fs_parameters = {
-	.name = "gfs2",
-	.specs = gfs2_param_specs,
-	.enums = gfs2_param_enums,
-};
-
 /* Parse a single mount parameter */
 static int gfs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
 {
@@ -1341,7 +1300,7 @@ static int gfs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
 	struct fs_parse_result result;
 	int o;
 
-	o = fs_parse(fc, &gfs2_fs_parameters, param, &result);
+	o = fs_parse(fc, gfs2_fs_parameters, param, &result);
 	if (o < 0)
 		return o;
 
@@ -1369,7 +1328,7 @@ static int gfs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
 		break;
 	case Opt_debug:
 		if (result.boolean && args->ar_errors == GFS2_ERRORS_PANIC)
-			return invalf(fc, "gfs2: -o debug and -o errors=panic are mutually exclusive");
+			return invalfc(fc, "-o debug and -o errors=panic are mutually exclusive");
 		args->ar_debug = result.boolean;
 		break;
 	case Opt_upgrade:
@@ -1378,17 +1337,11 @@ static int gfs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
 	case Opt_acl:
 		args->ar_posix_acl = result.boolean;
 		break;
+	case Opt_quota_flag:
+		args->ar_quota = result.negated ? GFS2_QUOTA_OFF : GFS2_QUOTA_ON;
+		break;
 	case Opt_quota:
-		/* The quota option can be a flag or an enum. A non-zero int_32
-		   result means that we have an enum index. Otherwise we have
-		   to rely on the 'negated' flag to tell us whether 'quota' or
-		   'noquota' was specified. */
-		if (result.negated)
-			args->ar_quota = GFS2_QUOTA_OFF;
-		else if (result.int_32 > 0)
-			args->ar_quota = opt_quota_values[result.int_32];
-		else
-			args->ar_quota = GFS2_QUOTA_ON;
+		args->ar_quota = result.int_32;
 		break;
 	case Opt_suiddir:
 		args->ar_suiddir = result.boolean;
@@ -1405,27 +1358,27 @@ static int gfs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
 		break;
 	case Opt_commit:
 		if (result.int_32 <= 0)
-			return invalf(fc, "gfs2: commit mount option requires a positive numeric argument");
+			return invalfc(fc, "commit mount option requires a positive numeric argument");
 		args->ar_commit = result.int_32;
 		break;
 	case Opt_statfs_quantum:
 		if (result.int_32 < 0)
-			return invalf(fc, "gfs2: statfs_quantum mount option requires a non-negative numeric argument");
+			return invalfc(fc, "statfs_quantum mount option requires a non-negative numeric argument");
 		args->ar_statfs_quantum = result.int_32;
 		break;
 	case Opt_quota_quantum:
 		if (result.int_32 <= 0)
-			return invalf(fc, "gfs2: quota_quantum mount option requires a positive numeric argument");
+			return invalfc(fc, "quota_quantum mount option requires a positive numeric argument");
 		args->ar_quota_quantum = result.int_32;
 		break;
 	case Opt_statfs_percent:
 		if (result.int_32 < 0 || result.int_32 > 100)
-			return invalf(fc, "gfs2: statfs_percent mount option requires a numeric argument between 0 and 100");
+			return invalfc(fc, "statfs_percent mount option requires a numeric argument between 0 and 100");
 		args->ar_statfs_percent = result.int_32;
 		break;
 	case Opt_errors:
 		if (args->ar_debug && result.uint_32 == GFS2_ERRORS_PANIC)
-			return invalf(fc, "gfs2: -o debug and -o errors=panic are mutually exclusive");
+			return invalfc(fc, "-o debug and -o errors=panic are mutually exclusive");
 		args->ar_errors = result.uint_32;
 		break;
 	case Opt_barrier:
@@ -1438,7 +1391,7 @@ static int gfs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
 		args->ar_loccookie = result.boolean;
 		break;
 	default:
-		return invalf(fc, "gfs2: invalid mount option: %s", param->key);
+		return invalfc(fc, "invalid mount option: %s", param->key);
 	}
 	return 0;
 }
@@ -1464,27 +1417,27 @@ static int gfs2_reconfigure(struct fs_context *fc)
 	spin_unlock(&gt->gt_spin);
 
 	if (strcmp(newargs->ar_lockproto, oldargs->ar_lockproto)) {
-		errorf(fc, "gfs2: reconfiguration of locking protocol not allowed");
+		errorfc(fc, "reconfiguration of locking protocol not allowed");
 		return -EINVAL;
 	}
 	if (strcmp(newargs->ar_locktable, oldargs->ar_locktable)) {
-		errorf(fc, "gfs2: reconfiguration of lock table not allowed");
+		errorfc(fc, "reconfiguration of lock table not allowed");
 		return -EINVAL;
 	}
 	if (strcmp(newargs->ar_hostdata, oldargs->ar_hostdata)) {
-		errorf(fc, "gfs2: reconfiguration of host data not allowed");
+		errorfc(fc, "reconfiguration of host data not allowed");
 		return -EINVAL;
 	}
 	if (newargs->ar_spectator != oldargs->ar_spectator) {
-		errorf(fc, "gfs2: reconfiguration of spectator mode not allowed");
+		errorfc(fc, "reconfiguration of spectator mode not allowed");
 		return -EINVAL;
 	}
 	if (newargs->ar_localflocks != oldargs->ar_localflocks) {
-		errorf(fc, "gfs2: reconfiguration of localflocks not allowed");
+		errorfc(fc, "reconfiguration of localflocks not allowed");
 		return -EINVAL;
 	}
 	if (newargs->ar_meta != oldargs->ar_meta) {
-		errorf(fc, "gfs2: switching between gfs2 and gfs2meta not allowed");
+		errorfc(fc, "switching between gfs2 and gfs2meta not allowed");
 		return -EINVAL;
 	}
 	if (oldargs->ar_spectator)
@@ -1494,11 +1447,11 @@ static int gfs2_reconfigure(struct fs_context *fc)
 		if (fc->sb_flags & SB_RDONLY) {
 			error = gfs2_make_fs_ro(sdp);
 			if (error)
-				errorf(fc, "gfs2: unable to remount read-only");
+				errorfc(fc, "unable to remount read-only");
 		} else {
 			error = gfs2_make_fs_rw(sdp);
 			if (error)
-				errorf(fc, "gfs2: unable to remount read-write");
+				errorfc(fc, "unable to remount read-write");
 		}
 	}
 	sdp->sd_args = *newargs;
@@ -1643,7 +1596,7 @@ struct file_system_type gfs2_fs_type = {
 	.name = "gfs2",
 	.fs_flags = FS_REQUIRES_DEV,
 	.init_fs_context = gfs2_init_fs_context,
-	.parameters = &gfs2_fs_parameters,
+	.parameters = gfs2_fs_parameters,
 	.kill_sb = gfs2_kill_sb,
 	.owner = THIS_MODULE,
 };
diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c
index 7c016a082aa6..4b67d47a7e00 100644
--- a/fs/gfs2/quota.c
+++ b/fs/gfs2/quota.c
@@ -115,7 +115,7 @@ static void gfs2_qd_dispose(struct list_head *list)
 	struct gfs2_sbd *sdp;
 
 	while (!list_empty(list)) {
-		qd = list_entry(list->next, struct gfs2_quota_data, qd_lru);
+		qd = list_first_entry(list, struct gfs2_quota_data, qd_lru);
 		sdp = qd->qd_gl->gl_name.ln_sbd;
 
 		list_del(&qd->qd_lru);
@@ -525,11 +525,11 @@ static void qdsb_put(struct gfs2_quota_data *qd)
 }
 
 /**
- * gfs2_qa_alloc - make sure we have a quota allocations data structure,
- *                 if necessary
+ * gfs2_qa_get - make sure we have a quota allocations data structure,
+ *               if necessary
  * @ip: the inode for this reservation
  */
-int gfs2_qa_alloc(struct gfs2_inode *ip)
+int gfs2_qa_get(struct gfs2_inode *ip)
 {
 	int error = 0;
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
@@ -540,17 +540,21 @@ int gfs2_qa_alloc(struct gfs2_inode *ip)
 	down_write(&ip->i_rw_mutex);
 	if (ip->i_qadata == NULL) {
 		ip->i_qadata = kmem_cache_zalloc(gfs2_qadata_cachep, GFP_NOFS);
-		if (!ip->i_qadata)
+		if (!ip->i_qadata) {
 			error = -ENOMEM;
+			goto out;
+		}
 	}
+	ip->i_qadata->qa_ref++;
+out:
 	up_write(&ip->i_rw_mutex);
 	return error;
 }
 
-void gfs2_qa_delete(struct gfs2_inode *ip, atomic_t *wcount)
+void gfs2_qa_put(struct gfs2_inode *ip)
 {
 	down_write(&ip->i_rw_mutex);
-	if (ip->i_qadata && ((wcount == NULL) || (atomic_read(wcount) <= 1))) {
+	if (ip->i_qadata && --ip->i_qadata->qa_ref == 0) {
 		kmem_cache_free(gfs2_qadata_cachep, ip->i_qadata);
 		ip->i_qadata = NULL;
 	}
@@ -566,27 +570,27 @@ int gfs2_quota_hold(struct gfs2_inode *ip, kuid_t uid, kgid_t gid)
 	if (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF)
 		return 0;
 
-	if (ip->i_qadata == NULL) {
-		error = gfs2_rsqa_alloc(ip);
-		if (error)
-			return error;
-	}
+	error = gfs2_qa_get(ip);
+	if (error)
+		return error;
 
 	qd = ip->i_qadata->qa_qd;
 
 	if (gfs2_assert_warn(sdp, !ip->i_qadata->qa_qd_num) ||
-	    gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags)))
-		return -EIO;
+	    gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags))) {
+		error = -EIO;
+		goto out;
+	}
 
 	error = qdsb_get(sdp, make_kqid_uid(ip->i_inode.i_uid), qd);
 	if (error)
-		goto out;
+		goto out_unhold;
 	ip->i_qadata->qa_qd_num++;
 	qd++;
 
 	error = qdsb_get(sdp, make_kqid_gid(ip->i_inode.i_gid), qd);
 	if (error)
-		goto out;
+		goto out_unhold;
 	ip->i_qadata->qa_qd_num++;
 	qd++;
 
@@ -594,7 +598,7 @@ int gfs2_quota_hold(struct gfs2_inode *ip, kuid_t uid, kgid_t gid)
 	    !uid_eq(uid, ip->i_inode.i_uid)) {
 		error = qdsb_get(sdp, make_kqid_uid(uid), qd);
 		if (error)
-			goto out;
+			goto out_unhold;
 		ip->i_qadata->qa_qd_num++;
 		qd++;
 	}
@@ -603,14 +607,15 @@ int gfs2_quota_hold(struct gfs2_inode *ip, kuid_t uid, kgid_t gid)
 	    !gid_eq(gid, ip->i_inode.i_gid)) {
 		error = qdsb_get(sdp, make_kqid_gid(gid), qd);
 		if (error)
-			goto out;
+			goto out_unhold;
 		ip->i_qadata->qa_qd_num++;
 		qd++;
 	}
 
-out:
+out_unhold:
 	if (error)
 		gfs2_quota_unhold(ip);
+out:
 	return error;
 }
 
@@ -621,6 +626,7 @@ void gfs2_quota_unhold(struct gfs2_inode *ip)
 
 	if (ip->i_qadata == NULL)
 		return;
+
 	gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags));
 
 	for (x = 0; x < ip->i_qadata->qa_qd_num; x++) {
@@ -628,6 +634,7 @@ void gfs2_quota_unhold(struct gfs2_inode *ip)
 		ip->i_qadata->qa_qd[x] = NULL;
 	}
 	ip->i_qadata->qa_qd_num = 0;
+	gfs2_qa_put(ip);
 }
 
 static int sort_qd(const void *a, const void *b)
@@ -876,7 +883,7 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda)
 	unsigned int nalloc = 0, blocks;
 	int error;
 
-	error = gfs2_rsqa_alloc(ip);
+	error = gfs2_qa_get(ip);
 	if (error)
 		return error;
 
@@ -884,8 +891,10 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda)
 			      &data_blocks, &ind_blocks);
 
 	ghs = kmalloc_array(num_qd, sizeof(struct gfs2_holder), GFP_NOFS);
-	if (!ghs)
-		return -ENOMEM;
+	if (!ghs) {
+		error = -ENOMEM;
+		goto out;
+	}
 
 	sort(qda, num_qd, sizeof(struct gfs2_quota_data *), sort_qd, NULL);
 	inode_lock(&ip->i_inode);
@@ -893,12 +902,12 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda)
 		error = gfs2_glock_nq_init(qda[qx]->qd_gl, LM_ST_EXCLUSIVE,
 					   GL_NOCACHE, &ghs[qx]);
 		if (error)
-			goto out;
+			goto out_dq;
 	}
 
 	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh);
 	if (error)
-		goto out;
+		goto out_dq;
 
 	for (x = 0; x < num_qd; x++) {
 		offset = qd2offset(qda[x]);
@@ -950,13 +959,15 @@ out_ipres:
 	gfs2_inplace_release(ip);
 out_alloc:
 	gfs2_glock_dq_uninit(&i_gh);
-out:
+out_dq:
 	while (qx--)
 		gfs2_glock_dq_uninit(&ghs[qx]);
 	inode_unlock(&ip->i_inode);
 	kfree(ghs);
 	gfs2_log_flush(ip->i_gl->gl_name.ln_sbd, ip->i_gl,
 		       GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_DO_SYNC);
+out:
+	gfs2_qa_put(ip);
 	return error;
 }
 
@@ -1040,8 +1051,7 @@ int gfs2_quota_lock(struct gfs2_inode *ip, kuid_t uid, kgid_t gid)
 	u32 x;
 	int error = 0;
 
-	if (capable(CAP_SYS_RESOURCE) ||
-	    sdp->sd_args.ar_quota != GFS2_QUOTA_ON)
+	if (sdp->sd_args.ar_quota != GFS2_QUOTA_ON)
 		return 0;
 
 	error = gfs2_quota_hold(ip, uid, gid);
@@ -1114,7 +1124,7 @@ void gfs2_quota_unlock(struct gfs2_inode *ip)
 	int found;
 
 	if (!test_and_clear_bit(GIF_QD_LOCKED, &ip->i_flags))
-		goto out;
+		return;
 
 	for (x = 0; x < ip->i_qadata->qa_qd_num; x++) {
 		struct gfs2_quota_data *qd;
@@ -1151,7 +1161,6 @@ void gfs2_quota_unlock(struct gfs2_inode *ip)
 			qd_unlock(qda[x]);
 	}
 
-out:
 	gfs2_quota_unhold(ip);
 }
 
@@ -1199,9 +1208,6 @@ int gfs2_quota_check(struct gfs2_inode *ip, kuid_t uid, kgid_t gid,
 	if (!test_bit(GIF_QD_LOCKED, &ip->i_flags))
 		return 0;
 
-        if (sdp->sd_args.ar_quota != GFS2_QUOTA_ON)
-                return 0;
-
 	for (x = 0; x < ip->i_qadata->qa_qd_num; x++) {
 		qd = ip->i_qadata->qa_qd[x];
 
@@ -1259,6 +1265,9 @@ void gfs2_quota_change(struct gfs2_inode *ip, s64 change,
 	if (ip->i_diskflags & GFS2_DIF_SYSTEM)
 		return;
 
+	if (gfs2_assert_withdraw(sdp, ip->i_qadata &&
+				 ip->i_qadata->qa_ref > 0))
+		return;
 	for (x = 0; x < ip->i_qadata->qa_qd_num; x++) {
 		qd = ip->i_qadata->qa_qd[x];
 
@@ -1273,7 +1282,7 @@ int gfs2_quota_sync(struct super_block *sb, int type)
 {
 	struct gfs2_sbd *sdp = sb->s_fs_info;
 	struct gfs2_quota_data **qda;
-	unsigned int max_qd = PAGE_SIZE/sizeof(struct gfs2_holder);
+	unsigned int max_qd = PAGE_SIZE / sizeof(struct gfs2_holder);
 	unsigned int num_qd;
 	unsigned int x;
 	int error = 0;
@@ -1356,7 +1365,7 @@ int gfs2_quota_init(struct gfs2_sbd *sdp)
 	sdp->sd_quota_bitmap = kzalloc(bm_size, GFP_NOFS | __GFP_NOWARN);
 	if (sdp->sd_quota_bitmap == NULL)
 		sdp->sd_quota_bitmap = __vmalloc(bm_size, GFP_NOFS |
-						 __GFP_ZERO, PAGE_KERNEL);
+						 __GFP_ZERO);
 	if (!sdp->sd_quota_bitmap)
 		return error;
 
@@ -1441,7 +1450,7 @@ void gfs2_quota_cleanup(struct gfs2_sbd *sdp)
 
 	spin_lock(&qd_lock);
 	while (!list_empty(head)) {
-		qd = list_entry(head->prev, struct gfs2_quota_data, qd_list);
+		qd = list_last_entry(head, struct gfs2_quota_data, qd_list);
 
 		list_del(&qd->qd_list);
 
@@ -1475,9 +1484,9 @@ static void quotad_error(struct gfs2_sbd *sdp, const char *msg, int error)
 {
 	if (error == 0 || error == -EROFS)
 		return;
-	if (!test_bit(SDF_WITHDRAWN, &sdp->sd_flags)) {
-		fs_err(sdp, "gfs2_quotad: %s error %d\n", msg, error);
-		sdp->sd_log_error = error;
+	if (!gfs2_withdrawn(sdp)) {
+		if (!cmpxchg(&sdp->sd_log_error, 0, error))
+			fs_err(sdp, "gfs2_quotad: %s error %d\n", msg, error);
 		wake_up(&sdp->sd_logd_waitq);
 	}
 }
@@ -1504,7 +1513,7 @@ static void quotad_check_trunc_list(struct gfs2_sbd *sdp)
 		ip = NULL;
 		spin_lock(&sdp->sd_trunc_lock);
 		if (!list_empty(&sdp->sd_trunc_list)) {
-			ip = list_entry(sdp->sd_trunc_list.next,
+			ip = list_first_entry(&sdp->sd_trunc_list,
 					struct gfs2_inode, i_trunc_list);
 			list_del_init(&ip->i_trunc_list);
 		}
@@ -1541,6 +1550,8 @@ int gfs2_quotad(void *data)
 
 	while (!kthread_should_stop()) {
 
+		if (gfs2_withdrawn(sdp))
+			goto bypass;
 		/* Update the master statfs file */
 		if (sdp->sd_statfs_force_sync) {
 			int error = gfs2_statfs_sync(sdp->sd_vfs, 0);
@@ -1561,6 +1572,7 @@ int gfs2_quotad(void *data)
 
 		try_to_freeze();
 
+bypass:
 		t = min(quotad_timeo, statfs_timeo);
 
 		prepare_to_wait(&sdp->sd_quota_wait, &wait, TASK_INTERRUPTIBLE);
@@ -1674,7 +1686,7 @@ static int gfs2_set_dqblk(struct super_block *sb, struct kqid qid,
 	if (error)
 		return error;
 
-	error = gfs2_rsqa_alloc(ip);
+	error = gfs2_qa_get(ip);
 	if (error)
 		goto out_put;
 
@@ -1743,6 +1755,7 @@ out_i:
 out_q:
 	gfs2_glock_dq_uninit(&q_gh);
 out_unlockput:
+	gfs2_qa_put(ip);
 	inode_unlock(&ip->i_inode);
 out_put:
 	qd_put(qd);
diff --git a/fs/gfs2/quota.h b/fs/gfs2/quota.h
index 765627d9a91e..21ada332d555 100644
--- a/fs/gfs2/quota.h
+++ b/fs/gfs2/quota.h
@@ -15,8 +15,8 @@ struct gfs2_sbd;
 #define NO_UID_QUOTA_CHANGE INVALID_UID
 #define NO_GID_QUOTA_CHANGE INVALID_GID
 
-extern int gfs2_qa_alloc(struct gfs2_inode *ip);
-extern void gfs2_qa_delete(struct gfs2_inode *ip, atomic_t *wcount);
+extern int gfs2_qa_get(struct gfs2_inode *ip);
+extern void gfs2_qa_put(struct gfs2_inode *ip);
 extern int gfs2_quota_hold(struct gfs2_inode *ip, kuid_t uid, kgid_t gid);
 extern void gfs2_quota_unhold(struct gfs2_inode *ip);
 
@@ -44,7 +44,8 @@ static inline int gfs2_quota_lock_check(struct gfs2_inode *ip,
 	int ret;
 
 	ap->allowed = UINT_MAX; /* Assume we are permitted a whole lot */
-	if (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF)
+	if (capable(CAP_SYS_RESOURCE) ||
+	    sdp->sd_args.ar_quota == GFS2_QUOTA_OFF)
 		return 0;
 	ret = gfs2_quota_lock(ip, NO_UID_QUOTA_CHANGE, NO_GID_QUOTA_CHANGE);
 	if (ret)
diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c
index c529f8749a89..96c345f49273 100644
--- a/fs/gfs2/recovery.c
+++ b/fs/gfs2/recovery.c
@@ -111,7 +111,7 @@ void gfs2_revoke_clean(struct gfs2_jdesc *jd)
 	struct gfs2_revoke_replay *rr;
 
 	while (!list_empty(head)) {
-		rr = list_entry(head->next, struct gfs2_revoke_replay, rr_list);
+		rr = list_first_entry(head, struct gfs2_revoke_replay, rr_list);
 		list_del(&rr->rr_list);
 		kfree(rr);
 	}
@@ -263,11 +263,13 @@ static void clean_journal(struct gfs2_jdesc *jd,
 	u32 lblock = head->lh_blkno;
 
 	gfs2_replay_incr_blk(jd, &lblock);
-	if (jd->jd_jid == sdp->sd_lockstruct.ls_jid)
-		sdp->sd_log_flush_head = lblock;
 	gfs2_write_log_header(sdp, jd, head->lh_sequence + 1, 0, lblock,
 			      GFS2_LOG_HEAD_UNMOUNT | GFS2_LOG_HEAD_RECOVERY,
 			      REQ_PREFLUSH | REQ_FUA | REQ_META | REQ_SYNC);
+	if (jd->jd_jid == sdp->sd_lockstruct.ls_jid) {
+		sdp->sd_log_flush_head = lblock;
+		gfs2_log_incr_head(sdp);
+	}
 }
 
 
@@ -303,6 +305,11 @@ void gfs2_recover_func(struct work_struct *work)
 	int error = 0;
 	int jlocked = 0;
 
+	if (gfs2_withdrawn(sdp)) {
+		fs_err(sdp, "jid=%u: Recovery not attempted due to withdraw.\n",
+		       jd->jd_jid);
+		goto fail;
+	}
 	t_start = ktime_get();
 	if (sdp->sd_args.ar_spectator)
 		goto fail;
@@ -326,7 +333,7 @@ void gfs2_recover_func(struct work_struct *work)
 
 		default:
 			goto fail;
-		};
+		}
 
 		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED,
 					   LM_FLAG_NOEXP | GL_NOCACHE, &ji_gh);
@@ -391,6 +398,10 @@ void gfs2_recover_func(struct work_struct *work)
 		fs_info(sdp, "jid=%u: Replaying journal...0x%x to 0x%x\n",
 			jd->jd_jid, head.lh_tail, head.lh_blkno);
 
+		/* We take the sd_log_flush_lock here primarily to prevent log
+		 * flushes and simultaneous journal replays from stomping on
+		 * each other wrt sd_log_bio. */
+		down_read(&sdp->sd_log_flush_lock);
 		for (pass = 0; pass < 2; pass++) {
 			lops_before_scan(jd, &head, pass);
 			error = foreach_descriptor(jd, head.lh_tail,
@@ -401,6 +412,7 @@ void gfs2_recover_func(struct work_struct *work)
 		}
 
 		clean_journal(jd, &head);
+		up_read(&sdp->sd_log_flush_lock);
 
 		gfs2_glock_dq_uninit(&thaw_gh);
 		t_rep = ktime_get();
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 2466bb44a23c..a321c34e3d6e 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -36,16 +36,6 @@
 #define BFITNOENT ((u32)~0)
 #define NO_BLOCK ((u64)~0)
 
-#if BITS_PER_LONG == 32
-#define LBITMASK   (0x55555555UL)
-#define LBITSKIP55 (0x55555555UL)
-#define LBITSKIP00 (0x00000000UL)
-#else
-#define LBITMASK   (0x5555555555555555UL)
-#define LBITSKIP55 (0x5555555555555555UL)
-#define LBITSKIP00 (0x0000000000000000UL)
-#endif
-
 /*
  * These routines are used by the resource group routines (rgrp.c)
  * to keep track of block allocation.  Each block is represented by two
@@ -467,24 +457,24 @@ void gfs2_rgrp_verify(struct gfs2_rgrpd *rgd)
 	}
 
 	if (count[0] != rgd->rd_free) {
-		if (gfs2_consist_rgrpd(rgd))
-			fs_err(sdp, "free data mismatch:  %u != %u\n",
-			       count[0], rgd->rd_free);
+		gfs2_lm(sdp, "free data mismatch:  %u != %u\n",
+			count[0], rgd->rd_free);
+		gfs2_consist_rgrpd(rgd);
 		return;
 	}
 
 	tmp = rgd->rd_data - rgd->rd_free - rgd->rd_dinodes;
 	if (count[1] != tmp) {
-		if (gfs2_consist_rgrpd(rgd))
-			fs_err(sdp, "used data mismatch:  %u != %u\n",
-			       count[1], tmp);
+		gfs2_lm(sdp, "used data mismatch:  %u != %u\n",
+			count[1], tmp);
+		gfs2_consist_rgrpd(rgd);
 		return;
 	}
 
 	if (count[2] + count[3] != rgd->rd_dinodes) {
-		if (gfs2_consist_rgrpd(rgd))
-			fs_err(sdp, "used metadata mismatch:  %u != %u\n",
-			       count[2] + count[3], rgd->rd_dinodes);
+		gfs2_lm(sdp, "used metadata mismatch:  %u != %u\n",
+			count[2] + count[3], rgd->rd_dinodes);
+		gfs2_consist_rgrpd(rgd);
 		return;
 	}
 }
@@ -600,16 +590,6 @@ void gfs2_free_clones(struct gfs2_rgrpd *rgd)
 	}
 }
 
-/**
- * gfs2_rsqa_alloc - make sure we have a reservation assigned to the inode
- *                 plus a quota allocations data structure, if necessary
- * @ip: the inode for this reservation
- */
-int gfs2_rsqa_alloc(struct gfs2_inode *ip)
-{
-	return gfs2_qa_alloc(ip);
-}
-
 static void dump_rs(struct seq_file *seq, const struct gfs2_blkreserv *rs,
 		    const char *fs_id_buf)
 {
@@ -682,18 +662,17 @@ void gfs2_rs_deltree(struct gfs2_blkreserv *rs)
 }
 
 /**
- * gfs2_rsqa_delete - delete a multi-block reservation and quota allocation
+ * gfs2_rs_delete - delete a multi-block reservation
  * @ip: The inode for this reservation
  * @wcount: The inode's write count, or NULL
  *
  */
-void gfs2_rsqa_delete(struct gfs2_inode *ip, atomic_t *wcount)
+void gfs2_rs_delete(struct gfs2_inode *ip, atomic_t *wcount)
 {
 	down_write(&ip->i_rw_mutex);
 	if ((wcount == NULL) || (atomic_read(wcount) <= 1))
 		gfs2_rs_deltree(&ip->i_res);
 	up_write(&ip->i_rw_mutex);
-	gfs2_qa_delete(ip, wcount);
 }
 
 /**
@@ -730,8 +709,12 @@ void gfs2_clear_rgrpd(struct gfs2_sbd *sdp)
 		rb_erase(n, &sdp->sd_rindex_tree);
 
 		if (gl) {
-			glock_clear_object(gl, rgd);
+			if (gl->gl_state != LM_ST_UNLOCKED) {
+				gfs2_glock_cb(gl, LM_ST_UNLOCKED);
+				flush_delayed_work(&gl->gl_work);
+			}
 			gfs2_rgrp_brelse(rgd);
+			glock_clear_object(gl, rgd);
 			gfs2_glock_put(gl);
 		}
 
@@ -743,17 +726,6 @@ void gfs2_clear_rgrpd(struct gfs2_sbd *sdp)
 	}
 }
 
-static void gfs2_rindex_print(const struct gfs2_rgrpd *rgd)
-{
-	struct gfs2_sbd *sdp = rgd->rd_sbd;
-
-	fs_info(sdp, "ri_addr = %llu\n", (unsigned long long)rgd->rd_addr);
-	fs_info(sdp, "ri_length = %u\n", rgd->rd_length);
-	fs_info(sdp, "ri_data0 = %llu\n", (unsigned long long)rgd->rd_data0);
-	fs_info(sdp, "ri_data = %u\n", rgd->rd_data);
-	fs_info(sdp, "ri_bitbytes = %u\n", rgd->rd_bitbytes);
-}
-
 /**
  * gfs2_compute_bitstructs - Compute the bitmap sizes
  * @rgd: The resource group descriptor
@@ -824,11 +796,20 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd)
 	}
 	bi = rgd->rd_bits + (length - 1);
 	if ((bi->bi_start + bi->bi_bytes) * GFS2_NBBY != rgd->rd_data) {
-		if (gfs2_consist_rgrpd(rgd)) {
-			gfs2_rindex_print(rgd);
-			fs_err(sdp, "start=%u len=%u offset=%u\n",
-			       bi->bi_start, bi->bi_bytes, bi->bi_offset);
-		}
+		gfs2_lm(sdp,
+			"ri_addr = %llu\n"
+			"ri_length = %u\n"
+			"ri_data0 = %llu\n"
+			"ri_data = %u\n"
+			"ri_bitbytes = %u\n"
+			"start=%u len=%u offset=%u\n",
+			(unsigned long long)rgd->rd_addr,
+			rgd->rd_length,
+			(unsigned long long)rgd->rd_data0,
+			rgd->rd_data,
+			rgd->rd_bitbytes,
+			bi->bi_start, bi->bi_bytes, bi->bi_offset);
+		gfs2_consist_rgrpd(rgd);
 		return -EIO;
 	}
 
@@ -1296,23 +1277,6 @@ void gfs2_rgrp_brelse(struct gfs2_rgrpd *rgd)
 			bi->bi_bh = NULL;
 		}
 	}
-
-}
-
-/**
- * gfs2_rgrp_go_unlock - Unlock a rgrp glock
- * @gh: The glock holder for the resource group
- *
- */
-
-void gfs2_rgrp_go_unlock(struct gfs2_holder *gh)
-{
-	struct gfs2_rgrpd *rgd = gh->gh_gl->gl_object;
-	int demote_requested = test_bit(GLF_DEMOTE, &gh->gh_gl->gl_flags) |
-		test_bit(GLF_PENDING_DEMOTE, &gh->gh_gl->gl_flags);
-
-	if (rgd && demote_requested)
-		gfs2_rgrp_brelse(rgd);
 }
 
 int gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset,
@@ -1842,10 +1806,8 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip
 	struct gfs2_rbm rbm = { .rgd = rgd, .bii = 0, .offset = 0 };
 
 	while (1) {
-		down_write(&sdp->sd_log_flush_lock);
 		error = gfs2_rbm_find(&rbm, GFS2_BLKST_UNLINKED, NULL, NULL,
 				      true);
-		up_write(&sdp->sd_log_flush_lock);
 		if (error == -ENOSPC)
 			break;
 		if (WARN_ON_ONCE(error))
diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h
index c14a673ae36f..a1d7e14fc55b 100644
--- a/fs/gfs2/rgrp.h
+++ b/fs/gfs2/rgrp.h
@@ -33,7 +33,6 @@ extern int gfs2_rindex_update(struct gfs2_sbd *sdp);
 extern void gfs2_free_clones(struct gfs2_rgrpd *rgd);
 extern int gfs2_rgrp_go_lock(struct gfs2_holder *gh);
 extern void gfs2_rgrp_brelse(struct gfs2_rgrpd *rgd);
-extern void gfs2_rgrp_go_unlock(struct gfs2_holder *gh);
 
 extern struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip);
 
@@ -45,9 +44,8 @@ extern void gfs2_inplace_release(struct gfs2_inode *ip);
 extern int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *n,
 			     bool dinode, u64 *generation);
 
-extern int gfs2_rsqa_alloc(struct gfs2_inode *ip);
 extern void gfs2_rs_deltree(struct gfs2_blkreserv *rs);
-extern void gfs2_rsqa_delete(struct gfs2_inode *ip, atomic_t *wcount);
+extern void gfs2_rs_delete(struct gfs2_inode *ip, atomic_t *wcount);
 extern void __gfs2_free_blocks(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd,
 			       u64 bstart, u32 blen, int meta);
 extern void gfs2_free_meta(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd,
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 5fa1eec4fb4f..956fced0a8ec 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -61,11 +61,13 @@ void gfs2_jindex_free(struct gfs2_sbd *sdp)
 	sdp->sd_journals = 0;
 	spin_unlock(&sdp->sd_jindex_spin);
 
+	sdp->sd_jdesc = NULL;
 	while (!list_empty(&list)) {
-		jd = list_entry(list.next, struct gfs2_jdesc, jd_list);
+		jd = list_first_entry(&list, struct gfs2_jdesc, jd_list);
 		gfs2_free_journal_extents(jd);
 		list_del(&jd->jd_list);
 		iput(jd->jd_inode);
+		jd->jd_inode = NULL;
 		kfree(jd);
 	}
 }
@@ -171,9 +173,13 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
 		goto fail_threads;
 
 	j_gl->gl_ops->go_inval(j_gl, DIO_METADATA);
+	if (gfs2_withdrawn(sdp)) {
+		error = -EIO;
+		goto fail;
+	}
 
 	error = gfs2_find_jhead(sdp->sd_jdesc, &head, false);
-	if (error)
+	if (error || gfs2_withdrawn(sdp))
 		goto fail;
 
 	if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) {
@@ -187,7 +193,7 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
 	gfs2_log_pointers_init(sdp, head.lh_blkno);
 
 	error = gfs2_quota_init(sdp);
-	if (error)
+	if (error || gfs2_withdrawn(sdp))
 		goto fail;
 
 	set_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
@@ -399,8 +405,7 @@ struct lfcc {
  * Returns: errno
  */
 
-static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp,
-				    struct gfs2_holder *freeze_gh)
+static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp)
 {
 	struct gfs2_inode *ip;
 	struct gfs2_jdesc *jd;
@@ -425,7 +430,9 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp,
 	}
 
 	error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_EXCLUSIVE,
-				   GL_NOCACHE, freeze_gh);
+				   GL_NOCACHE, &sdp->sd_freeze_gh);
+	if (error)
+		goto out;
 
 	list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
 		error = gfs2_jdesc_check(jd);
@@ -441,11 +448,11 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp,
 	}
 
 	if (error)
-		gfs2_glock_dq_uninit(freeze_gh);
+		gfs2_glock_dq_uninit(&sdp->sd_freeze_gh);
 
 out:
 	while (!list_empty(&list)) {
-		lfcc = list_entry(list.next, struct lfcc, list);
+		lfcc = list_first_entry(&list, struct lfcc, list);
 		list_del(&lfcc->list);
 		gfs2_glock_dq_uninit(&lfcc->gh);
 		kfree(lfcc);
@@ -553,7 +560,7 @@ static void gfs2_dirty_inode(struct inode *inode, int flags)
 
 	if (!(flags & I_DIRTY_INODE))
 		return;
-	if (unlikely(test_bit(SDF_WITHDRAWN, &sdp->sd_flags)))
+	if (unlikely(gfs2_withdrawn(sdp)))
 		return;
 	if (!gfs2_glock_is_locked_by_me(ip->i_gl)) {
 		ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
@@ -598,34 +605,63 @@ out:
 int gfs2_make_fs_ro(struct gfs2_sbd *sdp)
 {
 	struct gfs2_holder freeze_gh;
-	int error;
-
-	error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, GL_NOCACHE,
-				   &freeze_gh);
-	if (error && !test_bit(SDF_WITHDRAWN, &sdp->sd_flags))
-		return error;
+	int error = 0;
+	int log_write_allowed = test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
+
+	gfs2_holder_mark_uninitialized(&freeze_gh);
+	if (sdp->sd_freeze_gl &&
+	    !gfs2_glock_is_locked_by_me(sdp->sd_freeze_gl)) {
+		if (!log_write_allowed) {
+			error = gfs2_glock_nq_init(sdp->sd_freeze_gl,
+						   LM_ST_SHARED, GL_NOCACHE |
+						   LM_FLAG_TRY, &freeze_gh);
+			if (error == GLR_TRYFAILED)
+				error = 0;
+		} else {
+			error = gfs2_glock_nq_init(sdp->sd_freeze_gl,
+						   LM_ST_SHARED, GL_NOCACHE,
+						   &freeze_gh);
+			if (error && !gfs2_withdrawn(sdp))
+				return error;
+		}
+	}
 
 	flush_workqueue(gfs2_delete_workqueue);
-	if (sdp->sd_quotad_process)
+	if (!log_write_allowed && current == sdp->sd_quotad_process)
+		fs_warn(sdp, "The quotad daemon is withdrawing.\n");
+	else if (sdp->sd_quotad_process)
 		kthread_stop(sdp->sd_quotad_process);
 	sdp->sd_quotad_process = NULL;
-	if (sdp->sd_logd_process)
+
+	if (!log_write_allowed && current == sdp->sd_logd_process)
+		fs_warn(sdp, "The logd daemon is withdrawing.\n");
+	else if (sdp->sd_logd_process)
 		kthread_stop(sdp->sd_logd_process);
 	sdp->sd_logd_process = NULL;
 
-	gfs2_quota_sync(sdp->sd_vfs, 0);
-	gfs2_statfs_sync(sdp->sd_vfs, 0);
-
-	gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_SHUTDOWN |
-		       GFS2_LFC_MAKE_FS_RO);
-	wait_event(sdp->sd_reserving_log_wait, atomic_read(&sdp->sd_reserving_log) == 0);
-	gfs2_assert_warn(sdp, atomic_read(&sdp->sd_log_blks_free) == sdp->sd_jdesc->jd_blocks);
+	if (log_write_allowed) {
+		gfs2_quota_sync(sdp->sd_vfs, 0);
+		gfs2_statfs_sync(sdp->sd_vfs, 0);
 
+		gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_SHUTDOWN |
+			       GFS2_LFC_MAKE_FS_RO);
+		wait_event(sdp->sd_reserving_log_wait,
+			   atomic_read(&sdp->sd_reserving_log) == 0);
+		gfs2_assert_warn(sdp, atomic_read(&sdp->sd_log_blks_free) ==
+				 sdp->sd_jdesc->jd_blocks);
+	} else {
+		wait_event_timeout(sdp->sd_reserving_log_wait,
+				   atomic_read(&sdp->sd_reserving_log) == 0,
+				   HZ * 5);
+	}
 	if (gfs2_holder_initialized(&freeze_gh))
 		gfs2_glock_dq_uninit(&freeze_gh);
 
 	gfs2_quota_cleanup(sdp);
 
+	if (!log_write_allowed)
+		sdp->sd_vfs->s_flags |= SB_RDONLY;
+
 	return error;
 }
 
@@ -676,8 +712,10 @@ restart:
 	gfs2_glock_put(sdp->sd_freeze_gl);
 
 	if (!sdp->sd_args.ar_spectator) {
-		gfs2_glock_dq_uninit(&sdp->sd_journal_gh);
-		gfs2_glock_dq_uninit(&sdp->sd_jinode_gh);
+		if (gfs2_holder_initialized(&sdp->sd_journal_gh))
+			gfs2_glock_dq_uninit(&sdp->sd_journal_gh);
+		if (gfs2_holder_initialized(&sdp->sd_jinode_gh))
+			gfs2_glock_dq_uninit(&sdp->sd_jinode_gh);
 		gfs2_glock_dq_uninit(&sdp->sd_sc_gh);
 		gfs2_glock_dq_uninit(&sdp->sd_qc_gh);
 		iput(sdp->sd_sc_inode);
@@ -761,21 +799,25 @@ static int gfs2_freeze(struct super_block *sb)
 	if (atomic_read(&sdp->sd_freeze_state) != SFS_UNFROZEN)
 		goto out;
 
-	if (test_bit(SDF_WITHDRAWN, &sdp->sd_flags)) {
-		error = -EINVAL;
-		goto out;
-	}
-
 	for (;;) {
-		error = gfs2_lock_fs_check_clean(sdp, &sdp->sd_freeze_gh);
+		if (gfs2_withdrawn(sdp)) {
+			error = -EINVAL;
+			goto out;
+		}
+
+		error = gfs2_lock_fs_check_clean(sdp);
 		if (!error)
 			break;
 
 		if (error == -EBUSY)
 			fs_err(sdp, "waiting for recovery before freeze\n");
-		else
+		else if (error == -EIO) {
+			fs_err(sdp, "Fatal IO error: cannot freeze gfs2 due "
+			       "to recovery error.\n");
+			goto out;
+		} else {
 			fs_err(sdp, "error freezing FS: %d\n", error);
-
+		}
 		fs_err(sdp, "retrying...\n");
 		msleep(1000);
 	}
@@ -1351,14 +1393,6 @@ out_unlock:
 	if (gfs2_rs_active(&ip->i_res))
 		gfs2_rs_deltree(&ip->i_res);
 
-	if (gfs2_holder_initialized(&ip->i_iopen_gh)) {
-		glock_clear_object(ip->i_iopen_gh.gh_gl, ip);
-		if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags)) {
-			ip->i_iopen_gh.gh_flags |= GL_NOCACHE;
-			gfs2_glock_dq(&ip->i_iopen_gh);
-		}
-		gfs2_holder_uninit(&ip->i_iopen_gh);
-	}
 	if (gfs2_holder_initialized(&gh)) {
 		glock_clear_object(ip->i_gl, ip);
 		gfs2_glock_dq_uninit(&gh);
@@ -1367,22 +1401,29 @@ out_unlock:
 		fs_warn(sdp, "gfs2_evict_inode: %d\n", error);
 out:
 	truncate_inode_pages_final(&inode->i_data);
-	gfs2_rsqa_delete(ip, NULL);
+	if (ip->i_qadata)
+		gfs2_assert_warn(sdp, ip->i_qadata->qa_ref == 0);
+	gfs2_rs_delete(ip, NULL);
 	gfs2_ordered_del_inode(ip);
 	clear_inode(inode);
 	gfs2_dir_hash_inval(ip);
-	glock_clear_object(ip->i_gl, ip);
-	wait_on_bit_io(&ip->i_flags, GIF_GLOP_PENDING, TASK_UNINTERRUPTIBLE);
-	gfs2_glock_add_to_lru(ip->i_gl);
-	gfs2_glock_put_eventually(ip->i_gl);
-	ip->i_gl = NULL;
+	if (ip->i_gl) {
+		glock_clear_object(ip->i_gl, ip);
+		wait_on_bit_io(&ip->i_flags, GIF_GLOP_PENDING, TASK_UNINTERRUPTIBLE);
+		gfs2_glock_add_to_lru(ip->i_gl);
+		gfs2_glock_put_eventually(ip->i_gl);
+		ip->i_gl = NULL;
+	}
 	if (gfs2_holder_initialized(&ip->i_iopen_gh)) {
 		struct gfs2_glock *gl = ip->i_iopen_gh.gh_gl;
 
 		glock_clear_object(gl, ip);
-		ip->i_iopen_gh.gh_flags |= GL_NOCACHE;
+		if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags)) {
+			ip->i_iopen_gh.gh_flags |= GL_NOCACHE;
+			gfs2_glock_dq(&ip->i_iopen_gh);
+		}
 		gfs2_glock_hold(gl);
-		gfs2_glock_dq_uninit(&ip->i_iopen_gh);
+		gfs2_holder_uninit(&ip->i_iopen_gh);
 		gfs2_glock_put_eventually(gl);
 	}
 }
@@ -1396,6 +1437,7 @@ static struct inode *gfs2_alloc_inode(struct super_block *sb)
 		return NULL;
 	ip->i_flags = 0;
 	ip->i_gl = NULL;
+	gfs2_holder_mark_uninitialized(&ip->i_iopen_gh);
 	memset(&ip->i_res, 0, sizeof(ip->i_res));
 	RB_CLEAR_NODE(&ip->i_res.rs_node);
 	ip->i_rahead = 0;
diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h
index b8bf811a1305..51900554ed81 100644
--- a/fs/gfs2/super.h
+++ b/fs/gfs2/super.h
@@ -26,7 +26,6 @@ extern void gfs2_jindex_free(struct gfs2_sbd *sdp);
 
 extern struct gfs2_jdesc *gfs2_jdesc_find(struct gfs2_sbd *sdp, unsigned int jid);
 extern int gfs2_jdesc_check(struct gfs2_jdesc *jd);
-
 extern int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename,
 				     struct gfs2_inode **ipp);
 
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index dd15b8e4af2c..d28c41bd69b0 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -118,7 +118,7 @@ static ssize_t freeze_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
 
 static ssize_t withdraw_show(struct gfs2_sbd *sdp, char *buf)
 {
-	unsigned int b = test_bit(SDF_WITHDRAWN, &sdp->sd_flags);
+	unsigned int b = gfs2_withdrawn(sdp);
 	return snprintf(buf, PAGE_SIZE, "%u\n", b);
 }
 
@@ -136,7 +136,8 @@ static ssize_t withdraw_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
 	if (val != 1)
 		return -EINVAL;
 
-	gfs2_lm_withdraw(sdp, "withdrawing from cluster at user's request\n");
+	gfs2_lm(sdp, "withdrawing from cluster at user's request\n");
+	gfs2_withdraw(sdp);
 
 	return len;
 }
@@ -434,6 +435,8 @@ int gfs2_recover_set(struct gfs2_sbd *sdp, unsigned jid)
 	 * never clear the DFL_BLOCK_LOCKS flag, so all our locks would
 	 * permanently stop working.
 	 */
+	if (!sdp->sd_jdesc)
+		goto out;
 	if (sdp->sd_jdesc->jd_jid == jid && !sdp->sd_args.ar_spectator)
 		goto out;
 	rv = -ENOENT;
diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c
index 35e3059255fe..ffe840505082 100644
--- a/fs/gfs2/trans.c
+++ b/fs/gfs2/trans.c
@@ -49,8 +49,7 @@ int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks,
 	if (blocks)
 		tr->tr_reserved += 6 + blocks;
 	if (revokes)
-		tr->tr_reserved += gfs2_struct2blk(sdp, revokes,
-						   sizeof(u64));
+		tr->tr_reserved += gfs2_struct2blk(sdp, revokes);
 	INIT_LIST_HEAD(&tr->tr_databuf);
 	INIT_LIST_HEAD(&tr->tr_buf);
 
@@ -77,10 +76,10 @@ static void gfs2_print_trans(struct gfs2_sbd *sdp, const struct gfs2_trans *tr)
 	fs_warn(sdp, "blocks=%u revokes=%u reserved=%u touched=%u\n",
 		tr->tr_blocks, tr->tr_revokes, tr->tr_reserved,
 		test_bit(TR_TOUCHED, &tr->tr_flags));
-	fs_warn(sdp, "Buf %u/%u Databuf %u/%u Revoke %u\n",
+	fs_warn(sdp, "Buf %u/%u Databuf %u/%u Revoke %u/%u\n",
 		tr->tr_num_buf_new, tr->tr_num_buf_rm,
 		tr->tr_num_databuf_new, tr->tr_num_databuf_rm,
-		tr->tr_num_revoke);
+		tr->tr_num_revoke, tr->tr_num_revoke_rm);
 }
 
 void gfs2_trans_end(struct gfs2_sbd *sdp)
@@ -229,6 +228,10 @@ void gfs2_trans_add_meta(struct gfs2_glock *gl, struct buffer_head *bh)
 		fs_info(sdp, "GFS2:adding buf while frozen\n");
 		gfs2_assert_withdraw(sdp, 0);
 	}
+	if (unlikely(gfs2_withdrawn(sdp))) {
+		fs_info(sdp, "GFS2:adding buf while withdrawn! 0x%llx\n",
+			(unsigned long long)bd->bd_bh->b_blocknr);
+	}
 	gfs2_pin(sdp, bd->bd_bh);
 	mh->__pad0 = cpu_to_be64(0);
 	mh->mh_jid = cpu_to_be32(sdp->sd_jdesc->jd_jid);
@@ -262,8 +265,10 @@ void gfs2_trans_remove_revoke(struct gfs2_sbd *sdp, u64 blkno, unsigned int len)
 			list_del_init(&bd->bd_list);
 			gfs2_assert_withdraw(sdp, sdp->sd_log_num_revoke);
 			sdp->sd_log_num_revoke--;
+			if (bd->bd_gl)
+				gfs2_glock_remove_revoke(bd->bd_gl);
 			kmem_cache_free(gfs2_bufdata_cachep, bd);
-			tr->tr_num_revoke--;
+			tr->tr_num_revoke_rm++;
 			if (--n == 0)
 				break;
 		}
diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c
index c45159133d8e..aa087a5675af 100644
--- a/fs/gfs2/util.c
+++ b/fs/gfs2/util.c
@@ -11,12 +11,18 @@
 #include <linux/buffer_head.h>
 #include <linux/crc32.h>
 #include <linux/gfs2_ondisk.h>
+#include <linux/delay.h>
 #include <linux/uaccess.h>
 
 #include "gfs2.h"
 #include "incore.h"
 #include "glock.h"
+#include "glops.h"
+#include "log.h"
+#include "lops.h"
+#include "recovery.h"
 #include "rgrp.h"
+#include "super.h"
 #include "util.h"
 
 struct kmem_cache *gfs2_glock_cachep __read_mostly;
@@ -33,32 +39,259 @@ void gfs2_assert_i(struct gfs2_sbd *sdp)
 	fs_emerg(sdp, "fatal assertion failed\n");
 }
 
-int gfs2_lm_withdraw(struct gfs2_sbd *sdp, const char *fmt, ...)
+/**
+ * check_journal_clean - Make sure a journal is clean for a spectator mount
+ * @sdp: The GFS2 superblock
+ * @jd: The journal descriptor
+ *
+ * Returns: 0 if the journal is clean or locked, else an error
+ */
+int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
+			bool verbose)
+{
+	int error;
+	struct gfs2_holder j_gh;
+	struct gfs2_log_header_host head;
+	struct gfs2_inode *ip;
+
+	ip = GFS2_I(jd->jd_inode);
+	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_NOEXP |
+				   GL_EXACT | GL_NOCACHE, &j_gh);
+	if (error) {
+		if (verbose)
+			fs_err(sdp, "Error %d locking journal for spectator "
+			       "mount.\n", error);
+		return -EPERM;
+	}
+	error = gfs2_jdesc_check(jd);
+	if (error) {
+		if (verbose)
+			fs_err(sdp, "Error checking journal for spectator "
+			       "mount.\n");
+		goto out_unlock;
+	}
+	error = gfs2_find_jhead(jd, &head, false);
+	if (error) {
+		if (verbose)
+			fs_err(sdp, "Error parsing journal for spectator "
+			       "mount.\n");
+		goto out_unlock;
+	}
+	if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) {
+		error = -EPERM;
+		if (verbose)
+			fs_err(sdp, "jid=%u: Journal is dirty, so the first "
+			       "mounter must not be a spectator.\n",
+			       jd->jd_jid);
+	}
+
+out_unlock:
+	gfs2_glock_dq_uninit(&j_gh);
+	return error;
+}
+
+static void signal_our_withdraw(struct gfs2_sbd *sdp)
+{
+	struct gfs2_glock *gl = sdp->sd_live_gh.gh_gl;
+	struct inode *inode = sdp->sd_jdesc->jd_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	u64 no_formal_ino = ip->i_no_formal_ino;
+	int ret = 0;
+	int tries;
+
+	if (test_bit(SDF_NORECOVERY, &sdp->sd_flags))
+		return;
+
+	/* Prevent any glock dq until withdraw recovery is complete */
+	set_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags);
+	/*
+	 * Don't tell dlm we're bailing until we have no more buffers in the
+	 * wind. If journal had an IO error, the log code should just purge
+	 * the outstanding buffers rather than submitting new IO. Making the
+	 * file system read-only will flush the journal, etc.
+	 *
+	 * During a normal unmount, gfs2_make_fs_ro calls gfs2_log_shutdown
+	 * which clears SDF_JOURNAL_LIVE. In a withdraw, we must not write
+	 * any UNMOUNT log header, so we can't call gfs2_log_shutdown, and
+	 * therefore we need to clear SDF_JOURNAL_LIVE manually.
+	 */
+	clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
+	if (!sb_rdonly(sdp->sd_vfs))
+		ret = gfs2_make_fs_ro(sdp);
+
+	if (sdp->sd_lockstruct.ls_ops->lm_lock == NULL) { /* lock_nolock */
+		if (!ret)
+			ret = -EIO;
+		clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags);
+		goto skip_recovery;
+	}
+	/*
+	 * Drop the glock for our journal so another node can recover it.
+	 */
+	if (gfs2_holder_initialized(&sdp->sd_journal_gh)) {
+		gfs2_glock_dq_wait(&sdp->sd_journal_gh);
+		gfs2_holder_uninit(&sdp->sd_journal_gh);
+	}
+	sdp->sd_jinode_gh.gh_flags |= GL_NOCACHE;
+	gfs2_glock_dq(&sdp->sd_jinode_gh);
+	if (test_bit(SDF_FS_FROZEN, &sdp->sd_flags)) {
+		/* Make sure gfs2_unfreeze works if partially-frozen */
+		flush_workqueue(gfs2_freeze_wq);
+		atomic_set(&sdp->sd_freeze_state, SFS_FROZEN);
+		thaw_super(sdp->sd_vfs);
+	} else {
+		wait_on_bit(&gl->gl_flags, GLF_DEMOTE, TASK_UNINTERRUPTIBLE);
+	}
+
+	/*
+	 * holder_uninit to force glock_put, to force dlm to let go
+	 */
+	gfs2_holder_uninit(&sdp->sd_jinode_gh);
+
+	/*
+	 * Note: We need to be careful here:
+	 * Our iput of jd_inode will evict it. The evict will dequeue its
+	 * glock, but the glock dq will wait for the withdraw unless we have
+	 * exception code in glock_dq.
+	 */
+	iput(inode);
+	/*
+	 * Wait until the journal inode's glock is freed. This allows try locks
+	 * on other nodes to be successful, otherwise we remain the owner of
+	 * the glock as far as dlm is concerned.
+	 */
+	if (gl->gl_ops->go_free) {
+		set_bit(GLF_FREEING, &gl->gl_flags);
+		wait_on_bit(&gl->gl_flags, GLF_FREEING, TASK_UNINTERRUPTIBLE);
+	}
+
+	/*
+	 * Dequeue the "live" glock, but keep a reference so it's never freed.
+	 */
+	gfs2_glock_hold(gl);
+	gfs2_glock_dq_wait(&sdp->sd_live_gh);
+	/*
+	 * We enqueue the "live" glock in EX so that all other nodes
+	 * get a demote request and act on it. We don't really want the
+	 * lock in EX, so we send a "try" lock with 1CB to produce a callback.
+	 */
+	fs_warn(sdp, "Requesting recovery of jid %d.\n",
+		sdp->sd_lockstruct.ls_jid);
+	gfs2_holder_reinit(LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB | LM_FLAG_NOEXP,
+			   &sdp->sd_live_gh);
+	msleep(GL_GLOCK_MAX_HOLD);
+	/*
+	 * This will likely fail in a cluster, but succeed standalone:
+	 */
+	ret = gfs2_glock_nq(&sdp->sd_live_gh);
+
+	/*
+	 * If we actually got the "live" lock in EX mode, there are no other
+	 * nodes available to replay our journal. So we try to replay it
+	 * ourselves. We hold the "live" glock to prevent other mounters
+	 * during recovery, then just dequeue it and reacquire it in our
+	 * normal SH mode. Just in case the problem that caused us to
+	 * withdraw prevents us from recovering our journal (e.g. io errors
+	 * and such) we still check if the journal is clean before proceeding
+	 * but we may wait forever until another mounter does the recovery.
+	 */
+	if (ret == 0) {
+		fs_warn(sdp, "No other mounters found. Trying to recover our "
+			"own journal jid %d.\n", sdp->sd_lockstruct.ls_jid);
+		if (gfs2_recover_journal(sdp->sd_jdesc, 1))
+			fs_warn(sdp, "Unable to recover our journal jid %d.\n",
+				sdp->sd_lockstruct.ls_jid);
+		gfs2_glock_dq_wait(&sdp->sd_live_gh);
+		gfs2_holder_reinit(LM_ST_SHARED, LM_FLAG_NOEXP | GL_EXACT,
+				   &sdp->sd_live_gh);
+		gfs2_glock_nq(&sdp->sd_live_gh);
+	}
+
+	gfs2_glock_queue_put(gl); /* drop the extra reference we acquired */
+	clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags);
+
+	/*
+	 * At this point our journal is evicted, so we need to get a new inode
+	 * for it. Once done, we need to call gfs2_find_jhead which
+	 * calls gfs2_map_journal_extents to map it for us again.
+	 *
+	 * Note that we don't really want it to look up a FREE block. The
+	 * GFS2_BLKST_FREE simply overrides a block check in gfs2_inode_lookup
+	 * which would otherwise fail because it requires grabbing an rgrp
+	 * glock, which would fail with -EIO because we're withdrawing.
+	 */
+	inode = gfs2_inode_lookup(sdp->sd_vfs, DT_UNKNOWN,
+				  sdp->sd_jdesc->jd_no_addr, no_formal_ino,
+				  GFS2_BLKST_FREE);
+	if (IS_ERR(inode)) {
+		fs_warn(sdp, "Reprocessing of jid %d failed with %ld.\n",
+			sdp->sd_lockstruct.ls_jid, PTR_ERR(inode));
+		goto skip_recovery;
+	}
+	sdp->sd_jdesc->jd_inode = inode;
+
+	/*
+	 * Now wait until recovery is complete.
+	 */
+	for (tries = 0; tries < 10; tries++) {
+		ret = check_journal_clean(sdp, sdp->sd_jdesc, false);
+		if (!ret)
+			break;
+		msleep(HZ);
+		fs_warn(sdp, "Waiting for journal recovery jid %d.\n",
+			sdp->sd_lockstruct.ls_jid);
+	}
+skip_recovery:
+	if (!ret)
+		fs_warn(sdp, "Journal recovery complete for jid %d.\n",
+			sdp->sd_lockstruct.ls_jid);
+	else
+		fs_warn(sdp, "Journal recovery skipped for %d until next "
+			"mount.\n", sdp->sd_lockstruct.ls_jid);
+	fs_warn(sdp, "Glock dequeues delayed: %lu\n", sdp->sd_glock_dqs_held);
+	sdp->sd_glock_dqs_held = 0;
+	wake_up_bit(&sdp->sd_flags, SDF_WITHDRAW_RECOVERY);
+}
+
+void gfs2_lm(struct gfs2_sbd *sdp, const char *fmt, ...)
 {
-	struct lm_lockstruct *ls = &sdp->sd_lockstruct;
-	const struct lm_lockops *lm = ls->ls_ops;
-	va_list args;
 	struct va_format vaf;
+	va_list args;
 
 	if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW &&
-	    test_and_set_bit(SDF_WITHDRAWN, &sdp->sd_flags))
-		return 0;
-
-	if (fmt) {
-		va_start(args, fmt);
+	    test_bit(SDF_WITHDRAWN, &sdp->sd_flags))
+		return;
+
+	va_start(args, fmt);
+	vaf.fmt = fmt;
+	vaf.va = &args;
+	fs_err(sdp, "%pV", &vaf);
+	va_end(args);
+}
 
-		vaf.fmt = fmt;
-		vaf.va = &args;
+int gfs2_withdraw(struct gfs2_sbd *sdp)
+{
+	struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+	const struct lm_lockops *lm = ls->ls_ops;
 
-		fs_err(sdp, "%pV", &vaf);
+	if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW &&
+	    test_and_set_bit(SDF_WITHDRAWN, &sdp->sd_flags)) {
+		if (!test_bit(SDF_WITHDRAW_IN_PROG, &sdp->sd_flags))
+			return -1;
 
-		va_end(args);
+		wait_on_bit(&sdp->sd_flags, SDF_WITHDRAW_IN_PROG,
+			    TASK_UNINTERRUPTIBLE);
+		return -1;
 	}
 
+	set_bit(SDF_WITHDRAW_IN_PROG, &sdp->sd_flags);
+
 	if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW) {
 		fs_err(sdp, "about to withdraw this file system\n");
 		BUG_ON(sdp->sd_args.ar_debug);
 
+		signal_our_withdraw(sdp);
+
 		kobject_uevent(&sdp->sd_kobj, KOBJ_OFFLINE);
 
 		if (!strcmp(sdp->sd_lockstruct.ls_ops->lm_proto_name, "lock_dlm"))
@@ -69,8 +302,11 @@ int gfs2_lm_withdraw(struct gfs2_sbd *sdp, const char *fmt, ...)
 			lm->lm_unmount(sdp);
 		}
 		set_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags);
-		fs_err(sdp, "withdrawn\n");
+		fs_err(sdp, "File system withdrawn\n");
 		dump_stack();
+		clear_bit(SDF_WITHDRAW_IN_PROG, &sdp->sd_flags);
+		smp_mb__after_atomic();
+		wake_up_bit(&sdp->sd_flags, SDF_WITHDRAW_IN_PROG);
 	}
 
 	if (sdp->sd_args.ar_errors == GFS2_ERRORS_PANIC)
@@ -81,35 +317,45 @@ int gfs2_lm_withdraw(struct gfs2_sbd *sdp, const char *fmt, ...)
 
 /**
  * gfs2_assert_withdraw_i - Cause the machine to withdraw if @assertion is false
- * Returns: -1 if this call withdrew the machine,
- *          -2 if it was already withdrawn
  */
 
-int gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion,
-			   const char *function, char *file, unsigned int line)
+void gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion,
+			    const char *function, char *file, unsigned int line,
+			    bool delayed)
 {
-	int me;
-	me = gfs2_lm_withdraw(sdp,
-			      "fatal: assertion \"%s\" failed\n"
-			      "   function = %s, file = %s, line = %u\n",
-			      assertion, function, file, line);
+	if (gfs2_withdrawn(sdp))
+		return;
+
+	fs_err(sdp,
+	       "fatal: assertion \"%s\" failed\n"
+	       "   function = %s, file = %s, line = %u\n",
+	       assertion, function, file, line);
+
+	/*
+	 * If errors=panic was specified on mount, it won't help to delay the
+	 * withdraw.
+	 */
+	if (sdp->sd_args.ar_errors == GFS2_ERRORS_PANIC)
+		delayed = false;
+
+	if (delayed)
+		gfs2_withdraw_delayed(sdp);
+	else
+		gfs2_withdraw(sdp);
 	dump_stack();
-	return (me) ? -1 : -2;
 }
 
 /**
  * gfs2_assert_warn_i - Print a message to the console if @assertion is false
- * Returns: -1 if we printed something
- *          -2 if we didn't
  */
 
-int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion,
-		       const char *function, char *file, unsigned int line)
+void gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion,
+			const char *function, char *file, unsigned int line)
 {
 	if (time_before(jiffies,
 			sdp->sd_last_warning +
 			gfs2_tune_get(sdp, gt_complain_secs) * HZ))
-		return -2;
+		return;
 
 	if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW)
 		fs_warn(sdp, "warning: assertion \"%s\" failed at function = %s, file = %s, line = %u\n",
@@ -127,69 +373,59 @@ int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion,
 		      sdp->sd_fsname, function, file, line);
 
 	sdp->sd_last_warning = jiffies;
-
-	return -1;
 }
 
 /**
  * gfs2_consist_i - Flag a filesystem consistency error and withdraw
- * Returns: -1 if this call withdrew the machine,
- *          0 if it was already withdrawn
  */
 
-int gfs2_consist_i(struct gfs2_sbd *sdp, int cluster_wide, const char *function,
-		   char *file, unsigned int line)
+void gfs2_consist_i(struct gfs2_sbd *sdp, const char *function,
+		    char *file, unsigned int line)
 {
-	int rv;
-	rv = gfs2_lm_withdraw(sdp,
-			      "fatal: filesystem consistency error - function = %s, file = %s, line = %u\n",
-			      function, file, line);
-	return rv;
+	gfs2_lm(sdp,
+		"fatal: filesystem consistency error - function = %s, file = %s, line = %u\n",
+		function, file, line);
+	gfs2_withdraw(sdp);
 }
 
 /**
  * gfs2_consist_inode_i - Flag an inode consistency error and withdraw
- * Returns: -1 if this call withdrew the machine,
- *          0 if it was already withdrawn
  */
 
-int gfs2_consist_inode_i(struct gfs2_inode *ip, int cluster_wide,
-			 const char *function, char *file, unsigned int line)
+void gfs2_consist_inode_i(struct gfs2_inode *ip,
+			  const char *function, char *file, unsigned int line)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
-	int rv;
-	rv = gfs2_lm_withdraw(sdp,
-			      "fatal: filesystem consistency error\n"
-			      "  inode = %llu %llu\n"
-			      "  function = %s, file = %s, line = %u\n",
-			      (unsigned long long)ip->i_no_formal_ino,
-			      (unsigned long long)ip->i_no_addr,
-			      function, file, line);
-	return rv;
+
+	gfs2_lm(sdp,
+		"fatal: filesystem consistency error\n"
+		"  inode = %llu %llu\n"
+		"  function = %s, file = %s, line = %u\n",
+		(unsigned long long)ip->i_no_formal_ino,
+		(unsigned long long)ip->i_no_addr,
+		function, file, line);
+	gfs2_withdraw(sdp);
 }
 
 /**
  * gfs2_consist_rgrpd_i - Flag a RG consistency error and withdraw
- * Returns: -1 if this call withdrew the machine,
- *          0 if it was already withdrawn
  */
 
-int gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, int cluster_wide,
-			 const char *function, char *file, unsigned int line)
+void gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd,
+			  const char *function, char *file, unsigned int line)
 {
 	struct gfs2_sbd *sdp = rgd->rd_sbd;
 	char fs_id_buf[sizeof(sdp->sd_fsname) + 7];
-	int rv;
 
 	sprintf(fs_id_buf, "fsid=%s: ", sdp->sd_fsname);
 	gfs2_rgrp_dump(NULL, rgd->rd_gl, fs_id_buf);
-	rv = gfs2_lm_withdraw(sdp,
-			      "fatal: filesystem consistency error\n"
-			      "  RG = %llu\n"
-			      "  function = %s, file = %s, line = %u\n",
-			      (unsigned long long)rgd->rd_addr,
-			      function, file, line);
-	return rv;
+	gfs2_lm(sdp,
+		"fatal: filesystem consistency error\n"
+		"  RG = %llu\n"
+		"  function = %s, file = %s, line = %u\n",
+		(unsigned long long)rgd->rd_addr,
+		function, file, line);
+	gfs2_withdraw(sdp);
 }
 
 /**
@@ -203,12 +439,14 @@ int gfs2_meta_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh,
 		       unsigned int line)
 {
 	int me;
-	me = gfs2_lm_withdraw(sdp,
-			      "fatal: invalid metadata block\n"
-			      "  bh = %llu (%s)\n"
-			      "  function = %s, file = %s, line = %u\n",
-			      (unsigned long long)bh->b_blocknr, type,
-			      function, file, line);
+
+	gfs2_lm(sdp,
+		"fatal: invalid metadata block\n"
+		"  bh = %llu (%s)\n"
+		"  function = %s, file = %s, line = %u\n",
+		(unsigned long long)bh->b_blocknr, type,
+		function, file, line);
+	me = gfs2_withdraw(sdp);
 	return (me) ? -1 : -2;
 }
 
@@ -223,12 +461,14 @@ int gfs2_metatype_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh,
 			   char *file, unsigned int line)
 {
 	int me;
-	me = gfs2_lm_withdraw(sdp,
-			      "fatal: invalid metadata block\n"
-			      "  bh = %llu (type: exp=%u, found=%u)\n"
-			      "  function = %s, file = %s, line = %u\n",
-			      (unsigned long long)bh->b_blocknr, type, t,
-			      function, file, line);
+
+	gfs2_lm(sdp,
+		"fatal: invalid metadata block\n"
+		"  bh = %llu (type: exp=%u, found=%u)\n"
+		"  function = %s, file = %s, line = %u\n",
+		(unsigned long long)bh->b_blocknr, type, t,
+		function, file, line);
+	me = gfs2_withdraw(sdp);
 	return (me) ? -1 : -2;
 }
 
@@ -241,12 +481,11 @@ int gfs2_metatype_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh,
 int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function, char *file,
 		    unsigned int line)
 {
-	int rv;
-	rv = gfs2_lm_withdraw(sdp,
-			      "fatal: I/O error\n"
-			      "  function = %s, file = %s, line = %u\n",
-			      function, file, line);
-	return rv;
+	gfs2_lm(sdp,
+		"fatal: I/O error\n"
+		"  function = %s, file = %s, line = %u\n",
+		function, file, line);
+	return gfs2_withdraw(sdp);
 }
 
 /**
@@ -258,14 +497,14 @@ void gfs2_io_error_bh_i(struct gfs2_sbd *sdp, struct buffer_head *bh,
 			const char *function, char *file, unsigned int line,
 			bool withdraw)
 {
-	if (!test_bit(SDF_WITHDRAWN, &sdp->sd_flags))
-		fs_err(sdp,
-		       "fatal: I/O error\n"
-		       "  block = %llu\n"
-		       "  function = %s, file = %s, line = %u\n",
-		       (unsigned long long)bh->b_blocknr,
-		       function, file, line);
+	if (gfs2_withdrawn(sdp))
+		return;
+
+	fs_err(sdp, "fatal: I/O error\n"
+	       "  block = %llu\n"
+	       "  function = %s, file = %s, line = %u\n",
+	       (unsigned long long)bh->b_blocknr, function, file, line);
 	if (withdraw)
-		gfs2_lm_withdraw(sdp, NULL);
+		gfs2_withdraw(sdp);
 }
 
diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h
index 4b68b2c1fe56..a3542560da6f 100644
--- a/fs/gfs2/util.h
+++ b/fs/gfs2/util.h
@@ -36,41 +36,59 @@ do { \
 } while (0)
 
 
-int gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion,
-			   const char *function, char *file, unsigned int line);
+void gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion,
+			    const char *function, char *file, unsigned int line,
+			    bool delayed);
 
 #define gfs2_assert_withdraw(sdp, assertion) \
-((likely(assertion)) ? 0 : gfs2_assert_withdraw_i((sdp), #assertion, \
-					__func__, __FILE__, __LINE__))
-
-
-int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion,
-		       const char *function, char *file, unsigned int line);
+	({ \
+		bool _bool = (assertion); \
+		if (unlikely(!_bool)) \
+			gfs2_assert_withdraw_i((sdp), #assertion, \
+					__func__, __FILE__, __LINE__, false); \
+		!_bool; \
+	})
+
+#define gfs2_assert_withdraw_delayed(sdp, assertion) \
+	({ \
+		bool _bool = (assertion); \
+		if (unlikely(!_bool)) \
+			gfs2_assert_withdraw_i((sdp), #assertion, \
+					__func__, __FILE__, __LINE__, true); \
+		!_bool; \
+	})
+
+void gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion,
+			const char *function, char *file, unsigned int line);
 
 #define gfs2_assert_warn(sdp, assertion) \
-((likely(assertion)) ? 0 : gfs2_assert_warn_i((sdp), #assertion, \
-					__func__, __FILE__, __LINE__))
-
+	({ \
+		bool _bool = (assertion); \
+		if (unlikely(!_bool)) \
+			gfs2_assert_warn_i((sdp), #assertion, \
+					__func__, __FILE__, __LINE__); \
+		!_bool; \
+	})
 
-int gfs2_consist_i(struct gfs2_sbd *sdp, int cluster_wide,
-		   const char *function, char *file, unsigned int line);
+void gfs2_consist_i(struct gfs2_sbd *sdp,
+		    const char *function, char *file, unsigned int line);
 
 #define gfs2_consist(sdp) \
-gfs2_consist_i((sdp), 0, __func__, __FILE__, __LINE__)
+gfs2_consist_i((sdp), __func__, __FILE__, __LINE__)
 
 
-int gfs2_consist_inode_i(struct gfs2_inode *ip, int cluster_wide,
-			 const char *function, char *file, unsigned int line);
+void gfs2_consist_inode_i(struct gfs2_inode *ip,
+			  const char *function, char *file, unsigned int line);
 
 #define gfs2_consist_inode(ip) \
-gfs2_consist_inode_i((ip), 0, __func__, __FILE__, __LINE__)
+gfs2_consist_inode_i((ip), __func__, __FILE__, __LINE__)
 
 
-int gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, int cluster_wide,
-			 const char *function, char *file, unsigned int line);
+void gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd,
+			  const char *function, char *file, unsigned int line);
 
 #define gfs2_consist_rgrpd(rgd) \
-gfs2_consist_rgrpd_i((rgd), 0, __func__, __FILE__, __LINE__)
+gfs2_consist_rgrpd_i((rgd), __func__, __FILE__, __LINE__)
 
 
 int gfs2_meta_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh,
@@ -129,6 +147,9 @@ static inline void gfs2_metatype_set(struct buffer_head *bh, u16 type,
 int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function,
 		    char *file, unsigned int line);
 
+extern int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
+			       bool verbose);
+
 #define gfs2_io_error(sdp) \
 gfs2_io_error_i((sdp), __func__, __FILE__, __LINE__);
 
@@ -164,10 +185,30 @@ static inline unsigned int gfs2_tune_get_i(struct gfs2_tune *gt,
 	return x;
 }
 
+/**
+ * gfs2_withdraw_delayed - withdraw as soon as possible without deadlocks
+ * @sdp: the superblock
+ */
+static inline void gfs2_withdraw_delayed(struct gfs2_sbd *sdp)
+{
+	set_bit(SDF_WITHDRAWING, &sdp->sd_flags);
+}
+
+/**
+ * gfs2_withdrawn - test whether the file system is withdrawing or withdrawn
+ * @sdp: the superblock
+ */
+static inline bool gfs2_withdrawn(struct gfs2_sbd *sdp)
+{
+	return test_bit(SDF_WITHDRAWN, &sdp->sd_flags) ||
+		test_bit(SDF_WITHDRAWING, &sdp->sd_flags);
+}
+
 #define gfs2_tune_get(sdp, field) \
 gfs2_tune_get_i(&(sdp)->sd_tune, &(sdp)->sd_tune.field)
 
 __printf(2, 3)
-int gfs2_lm_withdraw(struct gfs2_sbd *sdp, const char *fmt, ...);
+void gfs2_lm(struct gfs2_sbd *sdp, const char *fmt, ...);
+int gfs2_withdraw(struct gfs2_sbd *sdp);
 
 #endif /* __UTIL_DOT_H__ */
diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c
index bbe593d16bea..9d7667bc4292 100644
--- a/fs/gfs2/xattr.c
+++ b/fs/gfs2/xattr.c
@@ -1222,7 +1222,7 @@ static int gfs2_xattr_set(const struct xattr_handler *handler,
 	struct gfs2_holder gh;
 	int ret;
 
-	ret = gfs2_rsqa_alloc(ip);
+	ret = gfs2_qa_get(ip);
 	if (ret)
 		return ret;
 
@@ -1231,15 +1231,19 @@ static int gfs2_xattr_set(const struct xattr_handler *handler,
 	if (!gfs2_glock_is_locked_by_me(ip->i_gl)) {
 		ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
 		if (ret)
-			return ret;
+			goto out;
 	} else {
-		if (WARN_ON_ONCE(ip->i_gl->gl_state != LM_ST_EXCLUSIVE))
-			return -EIO;
+		if (WARN_ON_ONCE(ip->i_gl->gl_state != LM_ST_EXCLUSIVE)) {
+			ret = -EIO;
+			goto out;
+		}
 		gfs2_holder_mark_uninitialized(&gh);
 	}
 	ret = __gfs2_xattr_set(inode, name, value, size, flags, handler->flags);
 	if (gfs2_holder_initialized(&gh))
 		gfs2_glock_dq_uninit(&gh);
+out:
+	gfs2_qa_put(ip);
 	return ret;
 }
 
diff --git a/fs/hfs/Kconfig b/fs/hfs/Kconfig
index 44f6e89bcb75..129926b5142d 100644
--- a/fs/hfs/Kconfig
+++ b/fs/hfs/Kconfig
@@ -6,7 +6,7 @@ config HFS_FS
 	help
 	  If you say Y here, you will be able to mount Macintosh-formatted
 	  floppy disks and hard drive partitions with full read-write access.
-	  Please read <file:Documentation/filesystems/hfs.txt> to learn about
+	  Please read <file:Documentation/filesystems/hfs.rst> to learn about
 	  the available mount options.
 
 	  To compile this file system support as a module, choose M here: the
diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h
index 6d0783e2e276..f71c384064c8 100644
--- a/fs/hfs/hfs_fs.h
+++ b/fs/hfs/hfs_fs.h
@@ -242,19 +242,35 @@ extern void hfs_mark_mdb_dirty(struct super_block *sb);
 /*
  * There are two time systems.  Both are based on seconds since
  * a particular time/date.
- *	Unix:	unsigned lil-endian since 00:00 GMT, Jan. 1, 1970
+ *	Unix:	signed little-endian since 00:00 GMT, Jan. 1, 1970
  *	mac:	unsigned big-endian since 00:00 GMT, Jan. 1, 1904
  *
+ * HFS implementations are highly inconsistent, this one matches the
+ * traditional behavior of 64-bit Linux, giving the most useful
+ * time range between 1970 and 2106, by treating any on-disk timestamp
+ * under HFS_UTC_OFFSET (Jan 1 1970) as a time between 2040 and 2106.
  */
-#define __hfs_u_to_mtime(sec)	cpu_to_be32(sec + 2082844800U - sys_tz.tz_minuteswest * 60)
-#define __hfs_m_to_utime(sec)	(be32_to_cpu(sec) - 2082844800U  + sys_tz.tz_minuteswest * 60)
+#define HFS_UTC_OFFSET 2082844800U
 
+static inline time64_t __hfs_m_to_utime(__be32 mt)
+{
+	time64_t ut = (u32)(be32_to_cpu(mt) - HFS_UTC_OFFSET);
+
+	return ut + sys_tz.tz_minuteswest * 60;
+}
+
+static inline __be32 __hfs_u_to_mtime(time64_t ut)
+{
+	ut -= sys_tz.tz_minuteswest * 60;
+
+	return cpu_to_be32(lower_32_bits(ut) + HFS_UTC_OFFSET);
+}
 #define HFS_I(inode)	(container_of(inode, struct hfs_inode_info, vfs_inode))
 #define HFS_SB(sb)	((struct hfs_sb_info *)(sb)->s_fs_info)
 
-#define hfs_m_to_utime(time)	(struct timespec){ .tv_sec = __hfs_m_to_utime(time) }
-#define hfs_u_to_mtime(time)	__hfs_u_to_mtime((time).tv_sec)
-#define hfs_mtime()		__hfs_u_to_mtime(get_seconds())
+#define hfs_m_to_utime(time)   (struct timespec64){ .tv_sec = __hfs_m_to_utime(time) }
+#define hfs_u_to_mtime(time)   __hfs_u_to_mtime((time).tv_sec)
+#define hfs_mtime()		__hfs_u_to_mtime(ktime_get_real_seconds())
 
 static inline const char *hfs_mdb_name(struct super_block *sb)
 {
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index da243c84e93b..2f224b98ee94 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -351,7 +351,7 @@ static int hfs_read_inode(struct inode *inode, void *data)
 		inode->i_mode &= ~hsb->s_file_umask;
 		inode->i_mode |= S_IFREG;
 		inode->i_ctime = inode->i_atime = inode->i_mtime =
-				timespec_to_timespec64(hfs_m_to_utime(rec->file.MdDat));
+				hfs_m_to_utime(rec->file.MdDat);
 		inode->i_op = &hfs_file_inode_operations;
 		inode->i_fop = &hfs_file_operations;
 		inode->i_mapping->a_ops = &hfs_aops;
@@ -362,7 +362,7 @@ static int hfs_read_inode(struct inode *inode, void *data)
 		HFS_I(inode)->fs_blocks = 0;
 		inode->i_mode = S_IFDIR | (S_IRWXUGO & ~hsb->s_dir_umask);
 		inode->i_ctime = inode->i_atime = inode->i_mtime =
-				timespec_to_timespec64(hfs_m_to_utime(rec->dir.MdDat));
+				hfs_m_to_utime(rec->dir.MdDat);
 		inode->i_op = &hfs_dir_inode_operations;
 		inode->i_fop = &hfs_dir_operations;
 		break;
diff --git a/fs/hfs/mdb.c b/fs/hfs/mdb.c
index 460281b1299e..cdf0edeeb278 100644
--- a/fs/hfs/mdb.c
+++ b/fs/hfs/mdb.c
@@ -32,29 +32,35 @@
 static int hfs_get_last_session(struct super_block *sb,
 				sector_t *start, sector_t *size)
 {
-	struct cdrom_multisession ms_info;
-	struct cdrom_tocentry te;
-	int res;
+	struct cdrom_device_info *cdi = disk_to_cdi(sb->s_bdev->bd_disk);
 
 	/* default values */
 	*start = 0;
 	*size = i_size_read(sb->s_bdev->bd_inode) >> 9;
 
 	if (HFS_SB(sb)->session >= 0) {
+		struct cdrom_tocentry te;
+	
+		if (!cdi)
+			return -EINVAL;
+
 		te.cdte_track = HFS_SB(sb)->session;
 		te.cdte_format = CDROM_LBA;
-		res = ioctl_by_bdev(sb->s_bdev, CDROMREADTOCENTRY, (unsigned long)&te);
-		if (!res && (te.cdte_ctrl & CDROM_DATA_TRACK) == 4) {
-			*start = (sector_t)te.cdte_addr.lba << 2;
-			return 0;
+		if (cdrom_read_tocentry(cdi, &te) ||
+		    (te.cdte_ctrl & CDROM_DATA_TRACK) != 4) {
+			pr_err("invalid session number or type of track\n");
+			return -EINVAL;
 		}
-		pr_err("invalid session number or type of track\n");
-		return -EINVAL;
+
+		*start = (sector_t)te.cdte_addr.lba << 2;
+	} else if (cdi) {
+		struct cdrom_multisession ms_info;
+
+		ms_info.addr_format = CDROM_LBA;
+		if (cdrom_multisession(cdi, &ms_info) == 0 && ms_info.xa_flag)
+			*start = (sector_t)ms_info.addr.lba << 2;
 	}
-	ms_info.addr_format = CDROM_LBA;
-	res = ioctl_by_bdev(sb->s_bdev, CDROMMULTISESSION, (unsigned long)&ms_info);
-	if (!res && ms_info.xa_flag)
-		*start = (sector_t)ms_info.addr.lba << 2;
+
 	return 0;
 }
 
diff --git a/fs/hfsplus/attributes.c b/fs/hfsplus/attributes.c
index e6d554476db4..eeebe80c6be4 100644
--- a/fs/hfsplus/attributes.c
+++ b/fs/hfsplus/attributes.c
@@ -292,6 +292,10 @@ static int __hfsplus_delete_attr(struct inode *inode, u32 cnid,
 		return -ENOENT;
 	}
 
+	/* Avoid btree corruption */
+	hfs_bnode_read(fd->bnode, fd->search_key,
+			fd->keyoffset, fd->keylength);
+
 	err = hfs_brec_remove(fd);
 	if (err)
 		return err;
diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h
index b8471bf05def..3b03fff68543 100644
--- a/fs/hfsplus/hfsplus_fs.h
+++ b/fs/hfsplus/hfsplus_fs.h
@@ -533,13 +533,31 @@ int hfsplus_submit_bio(struct super_block *sb, sector_t sector, void *buf,
 		       void **data, int op, int op_flags);
 int hfsplus_read_wrapper(struct super_block *sb);
 
-/* time macros */
-#define __hfsp_mt2ut(t)		(be32_to_cpu(t) - 2082844800U)
-#define __hfsp_ut2mt(t)		(cpu_to_be32(t + 2082844800U))
+/*
+ * time helpers: convert between 1904-base and 1970-base timestamps
+ *
+ * HFS+ implementations are highly inconsistent, this one matches the
+ * traditional behavior of 64-bit Linux, giving the most useful
+ * time range between 1970 and 2106, by treating any on-disk timestamp
+ * under HFSPLUS_UTC_OFFSET (Jan 1 1970) as a time between 2040 and 2106.
+ */
+#define HFSPLUS_UTC_OFFSET 2082844800U
+
+static inline time64_t __hfsp_mt2ut(__be32 mt)
+{
+	time64_t ut = (u32)(be32_to_cpu(mt) - HFSPLUS_UTC_OFFSET);
+
+	return ut;
+}
+
+static inline __be32 __hfsp_ut2mt(time64_t ut)
+{
+	return cpu_to_be32(lower_32_bits(ut) + HFSPLUS_UTC_OFFSET);
+}
 
 /* compatibility */
-#define hfsp_mt2ut(t)		(struct timespec){ .tv_sec = __hfsp_mt2ut(t) }
+#define hfsp_mt2ut(t)		(struct timespec64){ .tv_sec = __hfsp_mt2ut(t) }
 #define hfsp_ut2mt(t)		__hfsp_ut2mt((t).tv_sec)
-#define hfsp_now2mt()		__hfsp_ut2mt(get_seconds())
+#define hfsp_now2mt()		__hfsp_ut2mt(ktime_get_real_seconds())
 
 #endif
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index d131c8ea7eb6..e3da9e96b835 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -340,7 +340,7 @@ int hfsplus_file_fsync(struct file *file, loff_t start, loff_t end,
 	}
 
 	if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags))
-		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
 
 	inode_unlock(inode);
 
@@ -504,9 +504,9 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd)
 		hfsplus_get_perms(inode, &folder->permissions, 1);
 		set_nlink(inode, 1);
 		inode->i_size = 2 + be32_to_cpu(folder->valence);
-		inode->i_atime = timespec_to_timespec64(hfsp_mt2ut(folder->access_date));
-		inode->i_mtime = timespec_to_timespec64(hfsp_mt2ut(folder->content_mod_date));
-		inode->i_ctime = timespec_to_timespec64(hfsp_mt2ut(folder->attribute_mod_date));
+		inode->i_atime = hfsp_mt2ut(folder->access_date);
+		inode->i_mtime = hfsp_mt2ut(folder->content_mod_date);
+		inode->i_ctime = hfsp_mt2ut(folder->attribute_mod_date);
 		HFSPLUS_I(inode)->create_date = folder->create_date;
 		HFSPLUS_I(inode)->fs_blocks = 0;
 		if (folder->flags & cpu_to_be16(HFSPLUS_HAS_FOLDER_COUNT)) {
@@ -542,9 +542,9 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd)
 			init_special_inode(inode, inode->i_mode,
 					   be32_to_cpu(file->permissions.dev));
 		}
-		inode->i_atime = timespec_to_timespec64(hfsp_mt2ut(file->access_date));
-		inode->i_mtime = timespec_to_timespec64(hfsp_mt2ut(file->content_mod_date));
-		inode->i_ctime = timespec_to_timespec64(hfsp_mt2ut(file->attribute_mod_date));
+		inode->i_atime = hfsp_mt2ut(file->access_date);
+		inode->i_mtime = hfsp_mt2ut(file->content_mod_date);
+		inode->i_ctime = hfsp_mt2ut(file->attribute_mod_date);
 		HFSPLUS_I(inode)->create_date = file->create_date;
 	} else {
 		pr_err("bad catalog entry used to create inode\n");
diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index 2b9e5743105e..129dca3f4b78 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -239,7 +239,7 @@ out:
 	mutex_unlock(&sbi->vh_mutex);
 
 	if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags))
-		blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);
+		blkdev_issue_flush(sb->s_bdev, GFP_KERNEL);
 
 	return error;
 }
diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c
index 08c1580bdf7a..61eec628805d 100644
--- a/fs/hfsplus/wrapper.c
+++ b/fs/hfsplus/wrapper.c
@@ -127,31 +127,34 @@ static int hfsplus_read_mdb(void *bufptr, struct hfsplus_wd *wd)
 static int hfsplus_get_last_session(struct super_block *sb,
 				    sector_t *start, sector_t *size)
 {
-	struct cdrom_multisession ms_info;
-	struct cdrom_tocentry te;
-	int res;
+	struct cdrom_device_info *cdi = disk_to_cdi(sb->s_bdev->bd_disk);
 
 	/* default values */
 	*start = 0;
 	*size = i_size_read(sb->s_bdev->bd_inode) >> 9;
 
 	if (HFSPLUS_SB(sb)->session >= 0) {
+		struct cdrom_tocentry te;
+
+		if (!cdi)
+			return -EINVAL;
+
 		te.cdte_track = HFSPLUS_SB(sb)->session;
 		te.cdte_format = CDROM_LBA;
-		res = ioctl_by_bdev(sb->s_bdev,
-			CDROMREADTOCENTRY, (unsigned long)&te);
-		if (!res && (te.cdte_ctrl & CDROM_DATA_TRACK) == 4) {
-			*start = (sector_t)te.cdte_addr.lba << 2;
-			return 0;
+		if (cdrom_read_tocentry(cdi, &te) ||
+		    (te.cdte_ctrl & CDROM_DATA_TRACK) != 4) {
+			pr_err("invalid session number or type of track\n");
+			return -EINVAL;
 		}
-		pr_err("invalid session number or type of track\n");
-		return -EINVAL;
+		*start = (sector_t)te.cdte_addr.lba << 2;
+	} else if (cdi) {
+		struct cdrom_multisession ms_info;
+
+		ms_info.addr_format = CDROM_LBA;
+		if (cdrom_multisession(cdi, &ms_info) == 0 && ms_info.xa_flag)
+			*start = (sector_t)ms_info.addr.lba << 2;
 	}
-	ms_info.addr_format = CDROM_LBA;
-	res = ioctl_by_bdev(sb->s_bdev, CDROMMULTISESSION,
-		(unsigned long)&ms_info);
-	if (!res && ms_info.xa_flag)
-		*start = (sector_t)ms_info.addr.lba << 2;
+
 	return 0;
 }
 
diff --git a/fs/hostfs/hostfs.h b/fs/hostfs/hostfs.h
index f4295aa19350..69cb796f6270 100644
--- a/fs/hostfs/hostfs.h
+++ b/fs/hostfs/hostfs.h
@@ -37,16 +37,20 @@
  * is on, and remove the appropriate bits from attr->ia_mode (attr is a
  * "struct iattr *"). -BlaisorBlade
  */
+struct hostfs_timespec {
+	long long tv_sec;
+	long long tv_nsec;
+};
 
 struct hostfs_iattr {
-	unsigned int	ia_valid;
-	unsigned short	ia_mode;
-	uid_t		ia_uid;
-	gid_t		ia_gid;
-	loff_t		ia_size;
-	struct timespec	ia_atime;
-	struct timespec	ia_mtime;
-	struct timespec	ia_ctime;
+	unsigned int		ia_valid;
+	unsigned short		ia_mode;
+	uid_t			ia_uid;
+	gid_t			ia_gid;
+	loff_t			ia_size;
+	struct hostfs_timespec	ia_atime;
+	struct hostfs_timespec	ia_mtime;
+	struct hostfs_timespec	ia_ctime;
 };
 
 struct hostfs_stat {
@@ -56,7 +60,7 @@ struct hostfs_stat {
 	unsigned int uid;
 	unsigned int gid;
 	unsigned long long size;
-	struct timespec atime, mtime, ctime;
+	struct hostfs_timespec atime, mtime, ctime;
 	unsigned int blksize;
 	unsigned long long blocks;
 	unsigned int maj;
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 5a7eb0c79839..c070c0d8e3e9 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -139,8 +139,8 @@ static char *inode_name(struct inode *ino)
 
 static char *follow_link(char *link)
 {
-	int len, n;
 	char *name, *resolved, *end;
+	int n;
 
 	name = __getname();
 	if (!name) {
@@ -164,15 +164,13 @@ static char *follow_link(char *link)
 		return name;
 
 	*(end + 1) = '\0';
-	len = strlen(link) + strlen(name) + 1;
 
-	resolved = kmalloc(len, GFP_KERNEL);
+	resolved = kasprintf(GFP_KERNEL, "%s%s", link, name);
 	if (resolved == NULL) {
 		n = -ENOMEM;
 		goto out_free;
 	}
 
-	sprintf(resolved, "%s%s", link, name);
 	__putname(name);
 	kfree(link);
 	return resolved;
@@ -549,9 +547,9 @@ static int read_name(struct inode *ino, char *name)
 	set_nlink(ino, st.nlink);
 	i_uid_write(ino, st.uid);
 	i_gid_write(ino, st.gid);
-	ino->i_atime = timespec_to_timespec64(st.atime);
-	ino->i_mtime = timespec_to_timespec64(st.mtime);
-	ino->i_ctime = timespec_to_timespec64(st.ctime);
+	ino->i_atime = (struct timespec64){ st.atime.tv_sec, st.atime.tv_nsec };
+	ino->i_mtime = (struct timespec64){ st.mtime.tv_sec, st.mtime.tv_nsec };
+	ino->i_ctime = (struct timespec64){ st.ctime.tv_sec, st.ctime.tv_nsec };
 	ino->i_size = st.size;
 	ino->i_blocks = st.blocks;
 	return 0;
@@ -820,15 +818,18 @@ static int hostfs_setattr(struct dentry *dentry, struct iattr *attr)
 	}
 	if (attr->ia_valid & ATTR_ATIME) {
 		attrs.ia_valid |= HOSTFS_ATTR_ATIME;
-		attrs.ia_atime = timespec64_to_timespec(attr->ia_atime);
+		attrs.ia_atime = (struct hostfs_timespec)
+			{ attr->ia_atime.tv_sec, attr->ia_atime.tv_nsec };
 	}
 	if (attr->ia_valid & ATTR_MTIME) {
 		attrs.ia_valid |= HOSTFS_ATTR_MTIME;
-		attrs.ia_mtime = timespec64_to_timespec(attr->ia_mtime);
+		attrs.ia_mtime = (struct hostfs_timespec)
+			{ attr->ia_mtime.tv_sec, attr->ia_mtime.tv_nsec };
 	}
 	if (attr->ia_valid & ATTR_CTIME) {
 		attrs.ia_valid |= HOSTFS_ATTR_CTIME;
-		attrs.ia_ctime = timespec64_to_timespec(attr->ia_ctime);
+		attrs.ia_ctime = (struct hostfs_timespec)
+			{ attr->ia_ctime.tv_sec, attr->ia_ctime.tv_nsec };
 	}
 	if (attr->ia_valid & ATTR_ATIME_SET) {
 		attrs.ia_valid |= HOSTFS_ATTR_ATIME_SET;
@@ -918,18 +919,16 @@ static int hostfs_fill_sb_common(struct super_block *sb, void *d, int silent)
 	sb->s_d_op = &simple_dentry_operations;
 	sb->s_maxbytes = MAX_LFS_FILESIZE;
 
-	/* NULL is printed as <NULL> by sprintf: avoid that. */
+	/* NULL is printed as '(null)' by printf(): avoid that. */
 	if (req_root == NULL)
 		req_root = "";
 
 	err = -ENOMEM;
 	sb->s_fs_info = host_root_path =
-		kmalloc(strlen(root_ino) + strlen(req_root) + 2, GFP_KERNEL);
+		kasprintf(GFP_KERNEL, "%s/%s", root_ino, req_root);
 	if (host_root_path == NULL)
 		goto out;
 
-	sprintf(host_root_path, "%s/%s", root_ino, req_root);
-
 	root_inode = new_inode(sb);
 	if (!root_inode)
 		goto out;
diff --git a/fs/hpfs/Kconfig b/fs/hpfs/Kconfig
index 56aa0336254a..2b36dc6f0a10 100644
--- a/fs/hpfs/Kconfig
+++ b/fs/hpfs/Kconfig
@@ -9,7 +9,7 @@ config HPFS_FS
 	  write files to an OS/2 HPFS partition on your hard drive. OS/2
 	  floppies however are in regular MSDOS format, so you don't need this
 	  option in order to be able to read them. Read
-	  <file:Documentation/filesystems/hpfs.txt>.
+	  <file:Documentation/filesystems/hpfs.rst>.
 
 	  To compile this file system support as a module, choose M here: the
 	  module will be called hpfs.  If unsure, say N.
diff --git a/fs/hpfs/dir.c b/fs/hpfs/dir.c
index d85230c84ef2..f32f15669996 100644
--- a/fs/hpfs/dir.c
+++ b/fs/hpfs/dir.c
@@ -325,4 +325,5 @@ const struct file_operations hpfs_dir_ops =
 	.release	= hpfs_dir_release,
 	.fsync		= hpfs_file_fsync,
 	.unlocked_ioctl	= hpfs_ioctl,
+	.compat_ioctl	= compat_ptr_ioctl,
 };
diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c
index 1ecec124e76f..2de0d3492d15 100644
--- a/fs/hpfs/file.c
+++ b/fs/hpfs/file.c
@@ -125,10 +125,9 @@ static int hpfs_writepage(struct page *page, struct writeback_control *wbc)
 	return block_write_full_page(page, hpfs_get_block, wbc);
 }
 
-static int hpfs_readpages(struct file *file, struct address_space *mapping,
-			  struct list_head *pages, unsigned nr_pages)
+static void hpfs_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, hpfs_get_block);
+	mpage_readahead(rac, hpfs_get_block);
 }
 
 static int hpfs_writepages(struct address_space *mapping,
@@ -198,7 +197,7 @@ static int hpfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 const struct address_space_operations hpfs_aops = {
 	.readpage = hpfs_readpage,
 	.writepage = hpfs_writepage,
-	.readpages = hpfs_readpages,
+	.readahead = hpfs_readahead,
 	.writepages = hpfs_writepages,
 	.write_begin = hpfs_write_begin,
 	.write_end = hpfs_write_end,
@@ -215,6 +214,7 @@ const struct file_operations hpfs_file_ops =
 	.fsync		= hpfs_file_fsync,
 	.splice_read	= generic_file_splice_read,
 	.unlocked_ioctl	= hpfs_ioctl,
+	.compat_ioctl	= compat_ptr_ioctl,
 };
 
 const struct inode_operations hpfs_file_iops =
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index a478df035651..991c60c7ffe0 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -73,7 +73,7 @@ enum hugetlb_param {
 	Opt_uid,
 };
 
-static const struct fs_parameter_spec hugetlb_param_specs[] = {
+static const struct fs_parameter_spec hugetlb_fs_parameters[] = {
 	fsparam_u32   ("gid",		Opt_gid),
 	fsparam_string("min_size",	Opt_min_size),
 	fsparam_u32   ("mode",		Opt_mode),
@@ -84,11 +84,6 @@ static const struct fs_parameter_spec hugetlb_param_specs[] = {
 	{}
 };
 
-static const struct fs_parameter_description hugetlb_fs_parameters = {
-	.name		= "hugetlbfs",
-	.specs		= hugetlb_param_specs,
-};
-
 #ifdef CONFIG_NUMA
 static inline void hugetlb_set_vma_policy(struct vm_area_struct *vma,
 					struct inode *inode, pgoff_t index)
@@ -398,10 +393,9 @@ hugetlb_vmdelete_list(struct rb_root_cached *root, pgoff_t start, pgoff_t end)
  *	In this case, we first scan the range and release found pages.
  *	After releasing pages, hugetlb_unreserve_pages cleans up region/reserv
  *	maps and global counts.  Page faults can not race with truncation
- *	in this routine.  hugetlb_no_page() prevents page faults in the
- *	truncated range.  It checks i_size before allocation, and again after
- *	with the page table lock for the page held.  The same lock must be
- *	acquired to unmap a page.
+ *	in this routine.  hugetlb_no_page() holds i_mmap_rwsem and prevents
+ *	page faults in the truncated range by checking i_size.  i_size is
+ *	modified while holding i_mmap_rwsem.
  * hole punch is indicated if end is not LLONG_MAX
  *	In the hole punch case we scan the range and release found pages.
  *	Only when releasing a page is the associated region/reserv map
@@ -440,8 +434,16 @@ static void remove_inode_hugepages(struct inode *inode, loff_t lstart,
 			u32 hash;
 
 			index = page->index;
-			hash = hugetlb_fault_mutex_hash(h, mapping, index, 0);
-			mutex_lock(&hugetlb_fault_mutex_table[hash]);
+			hash = hugetlb_fault_mutex_hash(mapping, index);
+			if (!truncate_op) {
+				/*
+				 * Only need to hold the fault mutex in the
+				 * hole punch case.  This prevents races with
+				 * page faults.  Races are not possible in the
+				 * case of truncation.
+				 */
+				mutex_lock(&hugetlb_fault_mutex_table[hash]);
+			}
 
 			/*
 			 * If page is mapped, it was faulted in after being
@@ -455,7 +457,9 @@ static void remove_inode_hugepages(struct inode *inode, loff_t lstart,
 			if (unlikely(page_mapped(page))) {
 				BUG_ON(truncate_op);
 
+				mutex_unlock(&hugetlb_fault_mutex_table[hash]);
 				i_mmap_lock_write(mapping);
+				mutex_lock(&hugetlb_fault_mutex_table[hash]);
 				hugetlb_vmdelete_list(&mapping->i_mmap,
 					index * pages_per_huge_page(h),
 					(index + 1) * pages_per_huge_page(h));
@@ -482,7 +486,8 @@ static void remove_inode_hugepages(struct inode *inode, loff_t lstart,
 			}
 
 			unlock_page(page);
-			mutex_unlock(&hugetlb_fault_mutex_table[hash]);
+			if (!truncate_op)
+				mutex_unlock(&hugetlb_fault_mutex_table[hash]);
 		}
 		huge_pagevec_release(&pvec);
 		cond_resched();
@@ -520,8 +525,8 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset)
 	BUG_ON(offset & ~huge_page_mask(h));
 	pgoff = offset >> PAGE_SHIFT;
 
-	i_size_write(inode, offset);
 	i_mmap_lock_write(mapping);
+	i_size_write(inode, offset);
 	if (!RB_EMPTY_ROOT(&mapping->i_mmap.rb_root))
 		hugetlb_vmdelete_list(&mapping->i_mmap, pgoff, 0);
 	i_mmap_unlock_write(mapping);
@@ -643,8 +648,12 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset,
 		/* addr is the offset within the file (zero based) */
 		addr = index * hpage_size;
 
-		/* mutex taken here, fault path and hole punch */
-		hash = hugetlb_fault_mutex_hash(h, mapping, index, addr);
+		/*
+		 * fault mutex taken here, protects against fault path
+		 * and hole punch.  inode_lock previously taken protects
+		 * against truncation.
+		 */
+		hash = hugetlb_fault_mutex_hash(mapping, index);
 		mutex_lock(&hugetlb_fault_mutex_table[hash]);
 
 		/* See if already present in mapping to avoid alloc/free */
@@ -815,8 +824,11 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb,
 /*
  * File creation. Allocate an inode, and we're done..
  */
-static int hugetlbfs_mknod(struct inode *dir,
-			struct dentry *dentry, umode_t mode, dev_t dev)
+static int do_hugetlbfs_mknod(struct inode *dir,
+			struct dentry *dentry,
+			umode_t mode,
+			dev_t dev,
+			bool tmpfile)
 {
 	struct inode *inode;
 	int error = -ENOSPC;
@@ -824,13 +836,23 @@ static int hugetlbfs_mknod(struct inode *dir,
 	inode = hugetlbfs_get_inode(dir->i_sb, dir, mode, dev);
 	if (inode) {
 		dir->i_ctime = dir->i_mtime = current_time(dir);
-		d_instantiate(dentry, inode);
-		dget(dentry);	/* Extra count - pin the dentry in core */
+		if (tmpfile) {
+			d_tmpfile(dentry, inode);
+		} else {
+			d_instantiate(dentry, inode);
+			dget(dentry);/* Extra count - pin the dentry in core */
+		}
 		error = 0;
 	}
 	return error;
 }
 
+static int hugetlbfs_mknod(struct inode *dir,
+			struct dentry *dentry, umode_t mode, dev_t dev)
+{
+	return do_hugetlbfs_mknod(dir, dentry, mode, dev, false);
+}
+
 static int hugetlbfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 {
 	int retval = hugetlbfs_mknod(dir, dentry, mode | S_IFDIR, 0);
@@ -844,6 +866,12 @@ static int hugetlbfs_create(struct inode *dir, struct dentry *dentry, umode_t mo
 	return hugetlbfs_mknod(dir, dentry, mode | S_IFREG, 0);
 }
 
+static int hugetlbfs_tmpfile(struct inode *dir,
+			struct dentry *dentry, umode_t mode)
+{
+	return do_hugetlbfs_mknod(dir, dentry, mode | S_IFREG, 0, true);
+}
+
 static int hugetlbfs_symlink(struct inode *dir,
 			struct dentry *dentry, const char *symname)
 {
@@ -1102,6 +1130,7 @@ static const struct inode_operations hugetlbfs_dir_inode_operations = {
 	.mknod		= hugetlbfs_mknod,
 	.rename		= simple_rename,
 	.setattr	= hugetlbfs_setattr,
+	.tmpfile	= hugetlbfs_tmpfile,
 };
 
 static const struct inode_operations hugetlbfs_inode_operations = {
@@ -1151,7 +1180,7 @@ static int hugetlbfs_parse_param(struct fs_context *fc, struct fs_parameter *par
 	unsigned long ps;
 	int opt;
 
-	opt = fs_parse(fc, &hugetlb_fs_parameters, param, &result);
+	opt = fs_parse(fc, hugetlb_fs_parameters, param, &result);
 	if (opt < 0)
 		return opt;
 
@@ -1213,7 +1242,7 @@ static int hugetlbfs_parse_param(struct fs_context *fc, struct fs_parameter *par
 	}
 
 bad_val:
-	return invalf(fc, "hugetlbfs: Bad value '%s' for mount option '%s'\n",
+	return invalfc(fc, "Bad value '%s' for mount option '%s'\n",
 		      param->string, param->key);
 }
 
@@ -1338,7 +1367,7 @@ static int hugetlbfs_init_fs_context(struct fs_context *fc)
 static struct file_system_type hugetlbfs_fs_type = {
 	.name			= "hugetlbfs",
 	.init_fs_context	= hugetlbfs_init_fs_context,
-	.parameters		= &hugetlb_fs_parameters,
+	.parameters		= hugetlb_fs_parameters,
 	.kill_sb		= kill_litter_super,
 };
 
@@ -1461,28 +1490,43 @@ static int __init init_hugetlbfs_fs(void)
 					sizeof(struct hugetlbfs_inode_info),
 					0, SLAB_ACCOUNT, init_once);
 	if (hugetlbfs_inode_cachep == NULL)
-		goto out2;
+		goto out;
 
 	error = register_filesystem(&hugetlbfs_fs_type);
 	if (error)
-		goto out;
+		goto out_free;
+
+	/* default hstate mount is required */
+	mnt = mount_one_hugetlbfs(&hstates[default_hstate_idx]);
+	if (IS_ERR(mnt)) {
+		error = PTR_ERR(mnt);
+		goto out_unreg;
+	}
+	hugetlbfs_vfsmount[default_hstate_idx] = mnt;
 
+	/* other hstates are optional */
 	i = 0;
 	for_each_hstate(h) {
-		mnt = mount_one_hugetlbfs(h);
-		if (IS_ERR(mnt) && i == 0) {
-			error = PTR_ERR(mnt);
-			goto out;
+		if (i == default_hstate_idx) {
+			i++;
+			continue;
 		}
-		hugetlbfs_vfsmount[i] = mnt;
+
+		mnt = mount_one_hugetlbfs(h);
+		if (IS_ERR(mnt))
+			hugetlbfs_vfsmount[i] = NULL;
+		else
+			hugetlbfs_vfsmount[i] = mnt;
 		i++;
 	}
 
 	return 0;
 
- out:
+ out_unreg:
+	(void)unregister_filesystem(&hugetlbfs_fs_type);
+ out_free:
 	kmem_cache_destroy(hugetlbfs_inode_cachep);
- out2:
+ out:
 	return error;
 }
 fs_initcall(init_hugetlbfs_fs)
diff --git a/fs/inode.c b/fs/inode.c
index fef457a42882..37226a9cfa4f 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -12,6 +12,7 @@
 #include <linux/security.h>
 #include <linux/cdev.h>
 #include <linux/memblock.h>
+#include <linux/fscrypt.h>
 #include <linux/fsnotify.h>
 #include <linux/mount.h>
 #include <linux/posix_acl.h>
@@ -137,6 +138,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode)
 	inode->i_sb = sb;
 	inode->i_blkbits = sb->s_blocksize_bits;
 	inode->i_flags = 0;
+	atomic64_set(&inode->i_sequence, 0);
 	atomic_set(&inode->i_count, 1);
 	inode->i_op = &empty_iops;
 	inode->i_fop = &no_open_fops;
@@ -676,6 +678,7 @@ int invalidate_inodes(struct super_block *sb, bool kill_dirty)
 	struct inode *inode, *next;
 	LIST_HEAD(dispose);
 
+again:
 	spin_lock(&sb->s_inode_list_lock);
 	list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) {
 		spin_lock(&inode->i_lock);
@@ -698,6 +701,12 @@ int invalidate_inodes(struct super_block *sb, bool kill_dirty)
 		inode_lru_list_del(inode);
 		spin_unlock(&inode->i_lock);
 		list_add(&inode->i_lru, &dispose);
+		if (need_resched()) {
+			spin_unlock(&sb->s_inode_list_lock);
+			cond_resched();
+			dispose_list(&dispose);
+			goto again;
+		}
 	}
 	spin_unlock(&sb->s_inode_list_lock);
 
@@ -1591,25 +1600,31 @@ retry:
 }
 EXPORT_SYMBOL(iput);
 
+#ifdef CONFIG_BLOCK
 /**
  *	bmap	- find a block number in a file
- *	@inode: inode of file
- *	@block: block to find
- *
- *	Returns the block number on the device holding the inode that
- *	is the disk block number for the block of the file requested.
- *	That is, asked for block 4 of inode 1 the function will return the
- *	disk block relative to the disk start that holds that block of the
- *	file.
+ *	@inode:  inode owning the block number being requested
+ *	@block: pointer containing the block to find
+ *
+ *	Replaces the value in ``*block`` with the block number on the device holding
+ *	corresponding to the requested block number in the file.
+ *	That is, asked for block 4 of inode 1 the function will replace the
+ *	4 in ``*block``, with disk block relative to the disk start that holds that
+ *	block of the file.
+ *
+ *	Returns -EINVAL in case of error, 0 otherwise. If mapping falls into a
+ *	hole, returns 0 and ``*block`` is also set to 0.
  */
-sector_t bmap(struct inode *inode, sector_t block)
+int bmap(struct inode *inode, sector_t *block)
 {
-	sector_t res = 0;
-	if (inode->i_mapping->a_ops->bmap)
-		res = inode->i_mapping->a_ops->bmap(inode->i_mapping, block);
-	return res;
+	if (!inode->i_mapping->a_ops->bmap)
+		return -EINVAL;
+
+	*block = inode->i_mapping->a_ops->bmap(inode->i_mapping, *block);
+	return 0;
 }
 EXPORT_SYMBOL(bmap);
+#endif
 
 /*
  * With relative atime, only update atime if the previous atime is
@@ -1675,12 +1690,9 @@ EXPORT_SYMBOL(generic_update_time);
  */
 static int update_time(struct inode *inode, struct timespec64 *time, int flags)
 {
-	int (*update_time)(struct inode *, struct timespec64 *, int);
-
-	update_time = inode->i_op->update_time ? inode->i_op->update_time :
-		generic_update_time;
-
-	return update_time(inode, time, flags);
+	if (inode->i_op->update_time)
+		return inode->i_op->update_time(inode, time, flags);
+	return generic_update_time(inode, time, flags);
 }
 
 /**
@@ -2146,30 +2158,6 @@ void inode_nohighmem(struct inode *inode)
 EXPORT_SYMBOL(inode_nohighmem);
 
 /**
- * timespec64_trunc - Truncate timespec64 to a granularity
- * @t: Timespec64
- * @gran: Granularity in ns.
- *
- * Truncate a timespec64 to a granularity. Always rounds down. gran must
- * not be 0 nor greater than a second (NSEC_PER_SEC, or 10^9 ns).
- */
-struct timespec64 timespec64_trunc(struct timespec64 t, unsigned gran)
-{
-	/* Avoid division in the common cases 1 ns and 1 s. */
-	if (gran == 1) {
-		/* nothing */
-	} else if (gran == NSEC_PER_SEC) {
-		t.tv_nsec = 0;
-	} else if (gran > 1 && gran < NSEC_PER_SEC) {
-		t.tv_nsec -= t.tv_nsec % gran;
-	} else {
-		WARN(1, "illegal file time granularity: %u", gran);
-	}
-	return t;
-}
-EXPORT_SYMBOL(timespec64_trunc);
-
-/**
  * timestamp_truncate - Truncate timespec to a granularity
  * @t: Timespec
  * @inode: inode being updated
@@ -2245,7 +2233,7 @@ int vfs_ioc_setflags_prepare(struct inode *inode, unsigned int oldflags,
 	    !capable(CAP_LINUX_IMMUTABLE))
 		return -EPERM;
 
-	return 0;
+	return fscrypt_prepare_setflags(inode, oldflags, flags);
 }
 EXPORT_SYMBOL(vfs_ioc_setflags_prepare);
 
diff --git a/fs/internal.h b/fs/internal.h
index 315fcd8d237c..b89d78f10396 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -38,7 +38,6 @@ static inline int __sync_blockdev(struct block_device *bdev, int wait)
 /*
  * buffer.c
  */
-extern void guard_bio_eod(int rw, struct bio *bio);
 extern int __block_write_begin_int(struct page *page, loff_t pos, unsigned len,
 		get_block_t *get_block, struct iomap *iomap);
 
@@ -61,7 +60,6 @@ extern int finish_clean_context(struct fs_context *fc);
  */
 extern int filename_lookup(int dfd, struct filename *name, unsigned flags,
 			   struct path *path, struct path *root);
-extern int user_path_mountpoint_at(int, const char __user *, unsigned int, struct path *);
 extern int vfs_path_lookup(struct dentry *, struct vfsmount *,
 			   const char *, unsigned int, struct path *);
 long do_mknodat(int dfd, const char __user *filename, umode_t mode,
@@ -124,9 +122,10 @@ extern struct file *do_filp_open(int dfd, struct filename *pathname,
 		const struct open_flags *op);
 extern struct file *do_file_open_root(struct dentry *, struct vfsmount *,
 		const char *, const struct open_flags *);
+extern struct open_how build_open_how(int flags, umode_t mode);
+extern int build_open_flags(const struct open_how *how, struct open_flags *op);
 
 long do_sys_ftruncate(unsigned int fd, loff_t length, int small);
-long do_faccessat(int dfd, const char __user *filename, int mode);
 int do_fchmodat(int dfd, const char __user *filename, umode_t mode);
 int do_fchownat(int dfd, const char __user *filename, uid_t user, gid_t group,
 		int flag);
@@ -151,7 +150,6 @@ extern int invalidate_inodes(struct super_block *, bool);
 /*
  * dcache.c
  */
-extern struct dentry *__d_alloc(struct super_block *, const struct qstr *);
 extern int d_set_mounted(struct dentry *dentry);
 extern long prune_dcache_sb(struct super_block *sb, struct shrink_control *sc);
 extern struct dentry *d_alloc_cursor(struct dentry *);
@@ -181,11 +179,11 @@ extern void mnt_pin_kill(struct mount *m);
  */
 extern const struct dentry_operations ns_dentry_operations;
 
-/*
- * fs/ioctl.c
- */
-extern int do_vfs_ioctl(struct file *file, unsigned int fd, unsigned int cmd,
-		    unsigned long arg);
-
 /* direct-io.c: */
 int sb_init_dio_done_wq(struct super_block *sb);
+
+/*
+ * fs/stat.c:
+ */
+int do_statx(int dfd, const char __user *filename, unsigned flags,
+	     unsigned int mask, struct statx __user *buffer);
diff --git a/fs/io-wq.c b/fs/io-wq.c
new file mode 100644
index 000000000000..4023c9846860
--- /dev/null
+++ b/fs/io-wq.c
@@ -0,0 +1,1138 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Basic worker thread pool for io_uring
+ *
+ * Copyright (C) 2019 Jens Axboe
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/sched/signal.h>
+#include <linux/mm.h>
+#include <linux/mmu_context.h>
+#include <linux/sched/mm.h>
+#include <linux/percpu.h>
+#include <linux/slab.h>
+#include <linux/kthread.h>
+#include <linux/rculist_nulls.h>
+#include <linux/fs_struct.h>
+#include <linux/task_work.h>
+
+#include "io-wq.h"
+
+#define WORKER_IDLE_TIMEOUT	(5 * HZ)
+
+enum {
+	IO_WORKER_F_UP		= 1,	/* up and active */
+	IO_WORKER_F_RUNNING	= 2,	/* account as running */
+	IO_WORKER_F_FREE	= 4,	/* worker on free list */
+	IO_WORKER_F_EXITING	= 8,	/* worker exiting */
+	IO_WORKER_F_FIXED	= 16,	/* static idle worker */
+	IO_WORKER_F_BOUND	= 32,	/* is doing bounded work */
+};
+
+enum {
+	IO_WQ_BIT_EXIT		= 0,	/* wq exiting */
+	IO_WQ_BIT_CANCEL	= 1,	/* cancel work on list */
+	IO_WQ_BIT_ERROR		= 2,	/* error on setup */
+};
+
+enum {
+	IO_WQE_FLAG_STALLED	= 1,	/* stalled on hash */
+};
+
+/*
+ * One for each thread in a wqe pool
+ */
+struct io_worker {
+	refcount_t ref;
+	unsigned flags;
+	struct hlist_nulls_node nulls_node;
+	struct list_head all_list;
+	struct task_struct *task;
+	struct io_wqe *wqe;
+
+	struct io_wq_work *cur_work;
+	spinlock_t lock;
+
+	struct rcu_head rcu;
+	struct mm_struct *mm;
+	const struct cred *cur_creds;
+	const struct cred *saved_creds;
+	struct files_struct *restore_files;
+	struct fs_struct *restore_fs;
+};
+
+#if BITS_PER_LONG == 64
+#define IO_WQ_HASH_ORDER	6
+#else
+#define IO_WQ_HASH_ORDER	5
+#endif
+
+#define IO_WQ_NR_HASH_BUCKETS	(1u << IO_WQ_HASH_ORDER)
+
+struct io_wqe_acct {
+	unsigned nr_workers;
+	unsigned max_workers;
+	atomic_t nr_running;
+};
+
+enum {
+	IO_WQ_ACCT_BOUND,
+	IO_WQ_ACCT_UNBOUND,
+};
+
+/*
+ * Per-node worker thread pool
+ */
+struct io_wqe {
+	struct {
+		spinlock_t lock;
+		struct io_wq_work_list work_list;
+		unsigned long hash_map;
+		unsigned flags;
+	} ____cacheline_aligned_in_smp;
+
+	int node;
+	struct io_wqe_acct acct[2];
+
+	struct hlist_nulls_head free_list;
+	struct list_head all_list;
+
+	struct io_wq *wq;
+	struct io_wq_work *hash_tail[IO_WQ_NR_HASH_BUCKETS];
+};
+
+/*
+ * Per io_wq state
+  */
+struct io_wq {
+	struct io_wqe **wqes;
+	unsigned long state;
+
+	free_work_fn *free_work;
+
+	struct task_struct *manager;
+	struct user_struct *user;
+	refcount_t refs;
+	struct completion done;
+
+	refcount_t use_refs;
+};
+
+static bool io_worker_get(struct io_worker *worker)
+{
+	return refcount_inc_not_zero(&worker->ref);
+}
+
+static void io_worker_release(struct io_worker *worker)
+{
+	if (refcount_dec_and_test(&worker->ref))
+		wake_up_process(worker->task);
+}
+
+/*
+ * Note: drops the wqe->lock if returning true! The caller must re-acquire
+ * the lock in that case. Some callers need to restart handling if this
+ * happens, so we can't just re-acquire the lock on behalf of the caller.
+ */
+static bool __io_worker_unuse(struct io_wqe *wqe, struct io_worker *worker)
+{
+	bool dropped_lock = false;
+
+	if (worker->saved_creds) {
+		revert_creds(worker->saved_creds);
+		worker->cur_creds = worker->saved_creds = NULL;
+	}
+
+	if (current->files != worker->restore_files) {
+		__acquire(&wqe->lock);
+		spin_unlock_irq(&wqe->lock);
+		dropped_lock = true;
+
+		task_lock(current);
+		current->files = worker->restore_files;
+		task_unlock(current);
+	}
+
+	if (current->fs != worker->restore_fs)
+		current->fs = worker->restore_fs;
+
+	/*
+	 * If we have an active mm, we need to drop the wq lock before unusing
+	 * it. If we do, return true and let the caller retry the idle loop.
+	 */
+	if (worker->mm) {
+		if (!dropped_lock) {
+			__acquire(&wqe->lock);
+			spin_unlock_irq(&wqe->lock);
+			dropped_lock = true;
+		}
+		__set_current_state(TASK_RUNNING);
+		set_fs(KERNEL_DS);
+		unuse_mm(worker->mm);
+		mmput(worker->mm);
+		worker->mm = NULL;
+	}
+
+	return dropped_lock;
+}
+
+static inline struct io_wqe_acct *io_work_get_acct(struct io_wqe *wqe,
+						   struct io_wq_work *work)
+{
+	if (work->flags & IO_WQ_WORK_UNBOUND)
+		return &wqe->acct[IO_WQ_ACCT_UNBOUND];
+
+	return &wqe->acct[IO_WQ_ACCT_BOUND];
+}
+
+static inline struct io_wqe_acct *io_wqe_get_acct(struct io_wqe *wqe,
+						  struct io_worker *worker)
+{
+	if (worker->flags & IO_WORKER_F_BOUND)
+		return &wqe->acct[IO_WQ_ACCT_BOUND];
+
+	return &wqe->acct[IO_WQ_ACCT_UNBOUND];
+}
+
+static void io_worker_exit(struct io_worker *worker)
+{
+	struct io_wqe *wqe = worker->wqe;
+	struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker);
+	unsigned nr_workers;
+
+	/*
+	 * If we're not at zero, someone else is holding a brief reference
+	 * to the worker. Wait for that to go away.
+	 */
+	set_current_state(TASK_INTERRUPTIBLE);
+	if (!refcount_dec_and_test(&worker->ref))
+		schedule();
+	__set_current_state(TASK_RUNNING);
+
+	preempt_disable();
+	current->flags &= ~PF_IO_WORKER;
+	if (worker->flags & IO_WORKER_F_RUNNING)
+		atomic_dec(&acct->nr_running);
+	if (!(worker->flags & IO_WORKER_F_BOUND))
+		atomic_dec(&wqe->wq->user->processes);
+	worker->flags = 0;
+	preempt_enable();
+
+	spin_lock_irq(&wqe->lock);
+	hlist_nulls_del_rcu(&worker->nulls_node);
+	list_del_rcu(&worker->all_list);
+	if (__io_worker_unuse(wqe, worker)) {
+		__release(&wqe->lock);
+		spin_lock_irq(&wqe->lock);
+	}
+	acct->nr_workers--;
+	nr_workers = wqe->acct[IO_WQ_ACCT_BOUND].nr_workers +
+			wqe->acct[IO_WQ_ACCT_UNBOUND].nr_workers;
+	spin_unlock_irq(&wqe->lock);
+
+	/* all workers gone, wq exit can proceed */
+	if (!nr_workers && refcount_dec_and_test(&wqe->wq->refs))
+		complete(&wqe->wq->done);
+
+	kfree_rcu(worker, rcu);
+}
+
+static inline bool io_wqe_run_queue(struct io_wqe *wqe)
+	__must_hold(wqe->lock)
+{
+	if (!wq_list_empty(&wqe->work_list) &&
+	    !(wqe->flags & IO_WQE_FLAG_STALLED))
+		return true;
+	return false;
+}
+
+/*
+ * Check head of free list for an available worker. If one isn't available,
+ * caller must wake up the wq manager to create one.
+ */
+static bool io_wqe_activate_free_worker(struct io_wqe *wqe)
+	__must_hold(RCU)
+{
+	struct hlist_nulls_node *n;
+	struct io_worker *worker;
+
+	n = rcu_dereference(hlist_nulls_first_rcu(&wqe->free_list));
+	if (is_a_nulls(n))
+		return false;
+
+	worker = hlist_nulls_entry(n, struct io_worker, nulls_node);
+	if (io_worker_get(worker)) {
+		wake_up_process(worker->task);
+		io_worker_release(worker);
+		return true;
+	}
+
+	return false;
+}
+
+/*
+ * We need a worker. If we find a free one, we're good. If not, and we're
+ * below the max number of workers, wake up the manager to create one.
+ */
+static void io_wqe_wake_worker(struct io_wqe *wqe, struct io_wqe_acct *acct)
+{
+	bool ret;
+
+	/*
+	 * Most likely an attempt to queue unbounded work on an io_wq that
+	 * wasn't setup with any unbounded workers.
+	 */
+	WARN_ON_ONCE(!acct->max_workers);
+
+	rcu_read_lock();
+	ret = io_wqe_activate_free_worker(wqe);
+	rcu_read_unlock();
+
+	if (!ret && acct->nr_workers < acct->max_workers)
+		wake_up_process(wqe->wq->manager);
+}
+
+static void io_wqe_inc_running(struct io_wqe *wqe, struct io_worker *worker)
+{
+	struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker);
+
+	atomic_inc(&acct->nr_running);
+}
+
+static void io_wqe_dec_running(struct io_wqe *wqe, struct io_worker *worker)
+	__must_hold(wqe->lock)
+{
+	struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker);
+
+	if (atomic_dec_and_test(&acct->nr_running) && io_wqe_run_queue(wqe))
+		io_wqe_wake_worker(wqe, acct);
+}
+
+static void io_worker_start(struct io_wqe *wqe, struct io_worker *worker)
+{
+	allow_kernel_signal(SIGINT);
+
+	current->flags |= PF_IO_WORKER;
+
+	worker->flags |= (IO_WORKER_F_UP | IO_WORKER_F_RUNNING);
+	worker->restore_files = current->files;
+	worker->restore_fs = current->fs;
+	io_wqe_inc_running(wqe, worker);
+}
+
+/*
+ * Worker will start processing some work. Move it to the busy list, if
+ * it's currently on the freelist
+ */
+static void __io_worker_busy(struct io_wqe *wqe, struct io_worker *worker,
+			     struct io_wq_work *work)
+	__must_hold(wqe->lock)
+{
+	bool worker_bound, work_bound;
+
+	if (worker->flags & IO_WORKER_F_FREE) {
+		worker->flags &= ~IO_WORKER_F_FREE;
+		hlist_nulls_del_init_rcu(&worker->nulls_node);
+	}
+
+	/*
+	 * If worker is moving from bound to unbound (or vice versa), then
+	 * ensure we update the running accounting.
+	 */
+	worker_bound = (worker->flags & IO_WORKER_F_BOUND) != 0;
+	work_bound = (work->flags & IO_WQ_WORK_UNBOUND) == 0;
+	if (worker_bound != work_bound) {
+		io_wqe_dec_running(wqe, worker);
+		if (work_bound) {
+			worker->flags |= IO_WORKER_F_BOUND;
+			wqe->acct[IO_WQ_ACCT_UNBOUND].nr_workers--;
+			wqe->acct[IO_WQ_ACCT_BOUND].nr_workers++;
+			atomic_dec(&wqe->wq->user->processes);
+		} else {
+			worker->flags &= ~IO_WORKER_F_BOUND;
+			wqe->acct[IO_WQ_ACCT_UNBOUND].nr_workers++;
+			wqe->acct[IO_WQ_ACCT_BOUND].nr_workers--;
+			atomic_inc(&wqe->wq->user->processes);
+		}
+		io_wqe_inc_running(wqe, worker);
+	 }
+}
+
+/*
+ * No work, worker going to sleep. Move to freelist, and unuse mm if we
+ * have one attached. Dropping the mm may potentially sleep, so we drop
+ * the lock in that case and return success. Since the caller has to
+ * retry the loop in that case (we changed task state), we don't regrab
+ * the lock if we return success.
+ */
+static bool __io_worker_idle(struct io_wqe *wqe, struct io_worker *worker)
+	__must_hold(wqe->lock)
+{
+	if (!(worker->flags & IO_WORKER_F_FREE)) {
+		worker->flags |= IO_WORKER_F_FREE;
+		hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list);
+	}
+
+	return __io_worker_unuse(wqe, worker);
+}
+
+static inline unsigned int io_get_work_hash(struct io_wq_work *work)
+{
+	return work->flags >> IO_WQ_HASH_SHIFT;
+}
+
+static struct io_wq_work *io_get_next_work(struct io_wqe *wqe)
+	__must_hold(wqe->lock)
+{
+	struct io_wq_work_node *node, *prev;
+	struct io_wq_work *work, *tail;
+	unsigned int hash;
+
+	wq_list_for_each(node, prev, &wqe->work_list) {
+		work = container_of(node, struct io_wq_work, list);
+
+		/* not hashed, can run anytime */
+		if (!io_wq_is_hashed(work)) {
+			wq_list_del(&wqe->work_list, node, prev);
+			return work;
+		}
+
+		/* hashed, can run if not already running */
+		hash = io_get_work_hash(work);
+		if (!(wqe->hash_map & BIT(hash))) {
+			wqe->hash_map |= BIT(hash);
+			/* all items with this hash lie in [work, tail] */
+			tail = wqe->hash_tail[hash];
+			wqe->hash_tail[hash] = NULL;
+			wq_list_cut(&wqe->work_list, &tail->list, prev);
+			return work;
+		}
+	}
+
+	return NULL;
+}
+
+static void io_wq_switch_mm(struct io_worker *worker, struct io_wq_work *work)
+{
+	if (worker->mm) {
+		unuse_mm(worker->mm);
+		mmput(worker->mm);
+		worker->mm = NULL;
+	}
+	if (!work->mm) {
+		set_fs(KERNEL_DS);
+		return;
+	}
+	if (mmget_not_zero(work->mm)) {
+		use_mm(work->mm);
+		if (!worker->mm)
+			set_fs(USER_DS);
+		worker->mm = work->mm;
+		/* hang on to this mm */
+		work->mm = NULL;
+		return;
+	}
+
+	/* failed grabbing mm, ensure work gets cancelled */
+	work->flags |= IO_WQ_WORK_CANCEL;
+}
+
+static void io_wq_switch_creds(struct io_worker *worker,
+			       struct io_wq_work *work)
+{
+	const struct cred *old_creds = override_creds(work->creds);
+
+	worker->cur_creds = work->creds;
+	if (worker->saved_creds)
+		put_cred(old_creds); /* creds set by previous switch */
+	else
+		worker->saved_creds = old_creds;
+}
+
+static void io_impersonate_work(struct io_worker *worker,
+				struct io_wq_work *work)
+{
+	if (work->files && current->files != work->files) {
+		task_lock(current);
+		current->files = work->files;
+		task_unlock(current);
+	}
+	if (work->fs && current->fs != work->fs)
+		current->fs = work->fs;
+	if (work->mm != worker->mm)
+		io_wq_switch_mm(worker, work);
+	if (worker->cur_creds != work->creds)
+		io_wq_switch_creds(worker, work);
+}
+
+static void io_assign_current_work(struct io_worker *worker,
+				   struct io_wq_work *work)
+{
+	if (work) {
+		/* flush pending signals before assigning new work */
+		if (signal_pending(current))
+			flush_signals(current);
+		cond_resched();
+	}
+
+	spin_lock_irq(&worker->lock);
+	worker->cur_work = work;
+	spin_unlock_irq(&worker->lock);
+}
+
+static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work);
+
+static void io_worker_handle_work(struct io_worker *worker)
+	__releases(wqe->lock)
+{
+	struct io_wqe *wqe = worker->wqe;
+	struct io_wq *wq = wqe->wq;
+
+	do {
+		struct io_wq_work *work;
+		unsigned int hash;
+get_next:
+		/*
+		 * If we got some work, mark us as busy. If we didn't, but
+		 * the list isn't empty, it means we stalled on hashed work.
+		 * Mark us stalled so we don't keep looking for work when we
+		 * can't make progress, any work completion or insertion will
+		 * clear the stalled flag.
+		 */
+		work = io_get_next_work(wqe);
+		if (work)
+			__io_worker_busy(wqe, worker, work);
+		else if (!wq_list_empty(&wqe->work_list))
+			wqe->flags |= IO_WQE_FLAG_STALLED;
+
+		spin_unlock_irq(&wqe->lock);
+		if (!work)
+			break;
+		io_assign_current_work(worker, work);
+
+		/* handle a whole dependent link */
+		do {
+			struct io_wq_work *old_work, *next_hashed, *linked;
+
+			next_hashed = wq_next_work(work);
+			io_impersonate_work(worker, work);
+			/*
+			 * OK to set IO_WQ_WORK_CANCEL even for uncancellable
+			 * work, the worker function will do the right thing.
+			 */
+			if (test_bit(IO_WQ_BIT_CANCEL, &wq->state))
+				work->flags |= IO_WQ_WORK_CANCEL;
+
+			hash = io_get_work_hash(work);
+			linked = old_work = work;
+			linked->func(&linked);
+			linked = (old_work == linked) ? NULL : linked;
+
+			work = next_hashed;
+			if (!work && linked && !io_wq_is_hashed(linked)) {
+				work = linked;
+				linked = NULL;
+			}
+			io_assign_current_work(worker, work);
+			wq->free_work(old_work);
+
+			if (linked)
+				io_wqe_enqueue(wqe, linked);
+
+			if (hash != -1U && !next_hashed) {
+				spin_lock_irq(&wqe->lock);
+				wqe->hash_map &= ~BIT_ULL(hash);
+				wqe->flags &= ~IO_WQE_FLAG_STALLED;
+				/* dependent work is not hashed */
+				hash = -1U;
+				/* skip unnecessary unlock-lock wqe->lock */
+				if (!work)
+					goto get_next;
+				spin_unlock_irq(&wqe->lock);
+			}
+		} while (work);
+
+		spin_lock_irq(&wqe->lock);
+	} while (1);
+}
+
+static int io_wqe_worker(void *data)
+{
+	struct io_worker *worker = data;
+	struct io_wqe *wqe = worker->wqe;
+	struct io_wq *wq = wqe->wq;
+
+	io_worker_start(wqe, worker);
+
+	while (!test_bit(IO_WQ_BIT_EXIT, &wq->state)) {
+		set_current_state(TASK_INTERRUPTIBLE);
+loop:
+		spin_lock_irq(&wqe->lock);
+		if (io_wqe_run_queue(wqe)) {
+			__set_current_state(TASK_RUNNING);
+			io_worker_handle_work(worker);
+			goto loop;
+		}
+		/* drops the lock on success, retry */
+		if (__io_worker_idle(wqe, worker)) {
+			__release(&wqe->lock);
+			goto loop;
+		}
+		spin_unlock_irq(&wqe->lock);
+		if (signal_pending(current))
+			flush_signals(current);
+		if (schedule_timeout(WORKER_IDLE_TIMEOUT))
+			continue;
+		/* timed out, exit unless we're the fixed worker */
+		if (test_bit(IO_WQ_BIT_EXIT, &wq->state) ||
+		    !(worker->flags & IO_WORKER_F_FIXED))
+			break;
+	}
+
+	if (test_bit(IO_WQ_BIT_EXIT, &wq->state)) {
+		spin_lock_irq(&wqe->lock);
+		if (!wq_list_empty(&wqe->work_list))
+			io_worker_handle_work(worker);
+		else
+			spin_unlock_irq(&wqe->lock);
+	}
+
+	io_worker_exit(worker);
+	return 0;
+}
+
+/*
+ * Called when a worker is scheduled in. Mark us as currently running.
+ */
+void io_wq_worker_running(struct task_struct *tsk)
+{
+	struct io_worker *worker = kthread_data(tsk);
+	struct io_wqe *wqe = worker->wqe;
+
+	if (!(worker->flags & IO_WORKER_F_UP))
+		return;
+	if (worker->flags & IO_WORKER_F_RUNNING)
+		return;
+	worker->flags |= IO_WORKER_F_RUNNING;
+	io_wqe_inc_running(wqe, worker);
+}
+
+/*
+ * Called when worker is going to sleep. If there are no workers currently
+ * running and we have work pending, wake up a free one or have the manager
+ * set one up.
+ */
+void io_wq_worker_sleeping(struct task_struct *tsk)
+{
+	struct io_worker *worker = kthread_data(tsk);
+	struct io_wqe *wqe = worker->wqe;
+
+	if (!(worker->flags & IO_WORKER_F_UP))
+		return;
+	if (!(worker->flags & IO_WORKER_F_RUNNING))
+		return;
+
+	worker->flags &= ~IO_WORKER_F_RUNNING;
+
+	spin_lock_irq(&wqe->lock);
+	io_wqe_dec_running(wqe, worker);
+	spin_unlock_irq(&wqe->lock);
+}
+
+static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index)
+{
+	struct io_wqe_acct *acct =&wqe->acct[index];
+	struct io_worker *worker;
+
+	worker = kzalloc_node(sizeof(*worker), GFP_KERNEL, wqe->node);
+	if (!worker)
+		return false;
+
+	refcount_set(&worker->ref, 1);
+	worker->nulls_node.pprev = NULL;
+	worker->wqe = wqe;
+	spin_lock_init(&worker->lock);
+
+	worker->task = kthread_create_on_node(io_wqe_worker, worker, wqe->node,
+				"io_wqe_worker-%d/%d", index, wqe->node);
+	if (IS_ERR(worker->task)) {
+		kfree(worker);
+		return false;
+	}
+
+	spin_lock_irq(&wqe->lock);
+	hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list);
+	list_add_tail_rcu(&worker->all_list, &wqe->all_list);
+	worker->flags |= IO_WORKER_F_FREE;
+	if (index == IO_WQ_ACCT_BOUND)
+		worker->flags |= IO_WORKER_F_BOUND;
+	if (!acct->nr_workers && (worker->flags & IO_WORKER_F_BOUND))
+		worker->flags |= IO_WORKER_F_FIXED;
+	acct->nr_workers++;
+	spin_unlock_irq(&wqe->lock);
+
+	if (index == IO_WQ_ACCT_UNBOUND)
+		atomic_inc(&wq->user->processes);
+
+	wake_up_process(worker->task);
+	return true;
+}
+
+static inline bool io_wqe_need_worker(struct io_wqe *wqe, int index)
+	__must_hold(wqe->lock)
+{
+	struct io_wqe_acct *acct = &wqe->acct[index];
+
+	/* if we have available workers or no work, no need */
+	if (!hlist_nulls_empty(&wqe->free_list) || !io_wqe_run_queue(wqe))
+		return false;
+	return acct->nr_workers < acct->max_workers;
+}
+
+/*
+ * Manager thread. Tasked with creating new workers, if we need them.
+ */
+static int io_wq_manager(void *data)
+{
+	struct io_wq *wq = data;
+	int workers_to_create = num_possible_nodes();
+	int node;
+
+	/* create fixed workers */
+	refcount_set(&wq->refs, workers_to_create);
+	for_each_node(node) {
+		if (!node_online(node))
+			continue;
+		if (!create_io_worker(wq, wq->wqes[node], IO_WQ_ACCT_BOUND))
+			goto err;
+		workers_to_create--;
+	}
+
+	while (workers_to_create--)
+		refcount_dec(&wq->refs);
+
+	complete(&wq->done);
+
+	while (!kthread_should_stop()) {
+		if (current->task_works)
+			task_work_run();
+
+		for_each_node(node) {
+			struct io_wqe *wqe = wq->wqes[node];
+			bool fork_worker[2] = { false, false };
+
+			if (!node_online(node))
+				continue;
+
+			spin_lock_irq(&wqe->lock);
+			if (io_wqe_need_worker(wqe, IO_WQ_ACCT_BOUND))
+				fork_worker[IO_WQ_ACCT_BOUND] = true;
+			if (io_wqe_need_worker(wqe, IO_WQ_ACCT_UNBOUND))
+				fork_worker[IO_WQ_ACCT_UNBOUND] = true;
+			spin_unlock_irq(&wqe->lock);
+			if (fork_worker[IO_WQ_ACCT_BOUND])
+				create_io_worker(wq, wqe, IO_WQ_ACCT_BOUND);
+			if (fork_worker[IO_WQ_ACCT_UNBOUND])
+				create_io_worker(wq, wqe, IO_WQ_ACCT_UNBOUND);
+		}
+		set_current_state(TASK_INTERRUPTIBLE);
+		schedule_timeout(HZ);
+	}
+
+	if (current->task_works)
+		task_work_run();
+
+	return 0;
+err:
+	set_bit(IO_WQ_BIT_ERROR, &wq->state);
+	set_bit(IO_WQ_BIT_EXIT, &wq->state);
+	if (refcount_sub_and_test(workers_to_create, &wq->refs))
+		complete(&wq->done);
+	return 0;
+}
+
+static bool io_wq_can_queue(struct io_wqe *wqe, struct io_wqe_acct *acct,
+			    struct io_wq_work *work)
+{
+	bool free_worker;
+
+	if (!(work->flags & IO_WQ_WORK_UNBOUND))
+		return true;
+	if (atomic_read(&acct->nr_running))
+		return true;
+
+	rcu_read_lock();
+	free_worker = !hlist_nulls_empty(&wqe->free_list);
+	rcu_read_unlock();
+	if (free_worker)
+		return true;
+
+	if (atomic_read(&wqe->wq->user->processes) >= acct->max_workers &&
+	    !(capable(CAP_SYS_RESOURCE) || capable(CAP_SYS_ADMIN)))
+		return false;
+
+	return true;
+}
+
+static void io_run_cancel(struct io_wq_work *work, struct io_wqe *wqe)
+{
+	struct io_wq *wq = wqe->wq;
+
+	do {
+		struct io_wq_work *old_work = work;
+
+		work->flags |= IO_WQ_WORK_CANCEL;
+		work->func(&work);
+		work = (work == old_work) ? NULL : work;
+		wq->free_work(old_work);
+	} while (work);
+}
+
+static void io_wqe_insert_work(struct io_wqe *wqe, struct io_wq_work *work)
+{
+	unsigned int hash;
+	struct io_wq_work *tail;
+
+	if (!io_wq_is_hashed(work)) {
+append:
+		wq_list_add_tail(&work->list, &wqe->work_list);
+		return;
+	}
+
+	hash = io_get_work_hash(work);
+	tail = wqe->hash_tail[hash];
+	wqe->hash_tail[hash] = work;
+	if (!tail)
+		goto append;
+
+	wq_list_add_after(&work->list, &tail->list, &wqe->work_list);
+}
+
+static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work)
+{
+	struct io_wqe_acct *acct = io_work_get_acct(wqe, work);
+	int work_flags;
+	unsigned long flags;
+
+	/*
+	 * Do early check to see if we need a new unbound worker, and if we do,
+	 * if we're allowed to do so. This isn't 100% accurate as there's a
+	 * gap between this check and incrementing the value, but that's OK.
+	 * It's close enough to not be an issue, fork() has the same delay.
+	 */
+	if (unlikely(!io_wq_can_queue(wqe, acct, work))) {
+		io_run_cancel(work, wqe);
+		return;
+	}
+
+	work_flags = work->flags;
+	spin_lock_irqsave(&wqe->lock, flags);
+	io_wqe_insert_work(wqe, work);
+	wqe->flags &= ~IO_WQE_FLAG_STALLED;
+	spin_unlock_irqrestore(&wqe->lock, flags);
+
+	if ((work_flags & IO_WQ_WORK_CONCURRENT) ||
+	    !atomic_read(&acct->nr_running))
+		io_wqe_wake_worker(wqe, acct);
+}
+
+void io_wq_enqueue(struct io_wq *wq, struct io_wq_work *work)
+{
+	struct io_wqe *wqe = wq->wqes[numa_node_id()];
+
+	io_wqe_enqueue(wqe, work);
+}
+
+/*
+ * Work items that hash to the same value will not be done in parallel.
+ * Used to limit concurrent writes, generally hashed by inode.
+ */
+void io_wq_hash_work(struct io_wq_work *work, void *val)
+{
+	unsigned int bit;
+
+	bit = hash_ptr(val, IO_WQ_HASH_ORDER);
+	work->flags |= (IO_WQ_WORK_HASHED | (bit << IO_WQ_HASH_SHIFT));
+}
+
+static bool io_wqe_worker_send_sig(struct io_worker *worker, void *data)
+{
+	send_sig(SIGINT, worker->task, 1);
+	return false;
+}
+
+/*
+ * Iterate the passed in list and call the specific function for each
+ * worker that isn't exiting
+ */
+static bool io_wq_for_each_worker(struct io_wqe *wqe,
+				  bool (*func)(struct io_worker *, void *),
+				  void *data)
+{
+	struct io_worker *worker;
+	bool ret = false;
+
+	list_for_each_entry_rcu(worker, &wqe->all_list, all_list) {
+		if (io_worker_get(worker)) {
+			/* no task if node is/was offline */
+			if (worker->task)
+				ret = func(worker, data);
+			io_worker_release(worker);
+			if (ret)
+				break;
+		}
+	}
+
+	return ret;
+}
+
+void io_wq_cancel_all(struct io_wq *wq)
+{
+	int node;
+
+	set_bit(IO_WQ_BIT_CANCEL, &wq->state);
+
+	rcu_read_lock();
+	for_each_node(node) {
+		struct io_wqe *wqe = wq->wqes[node];
+
+		io_wq_for_each_worker(wqe, io_wqe_worker_send_sig, NULL);
+	}
+	rcu_read_unlock();
+}
+
+struct io_cb_cancel_data {
+	work_cancel_fn *fn;
+	void *data;
+};
+
+static bool io_wq_worker_cancel(struct io_worker *worker, void *data)
+{
+	struct io_cb_cancel_data *match = data;
+	unsigned long flags;
+	bool ret = false;
+
+	/*
+	 * Hold the lock to avoid ->cur_work going out of scope, caller
+	 * may dereference the passed in work.
+	 */
+	spin_lock_irqsave(&worker->lock, flags);
+	if (worker->cur_work &&
+	    !(worker->cur_work->flags & IO_WQ_WORK_NO_CANCEL) &&
+	    match->fn(worker->cur_work, match->data)) {
+		send_sig(SIGINT, worker->task, 1);
+		ret = true;
+	}
+	spin_unlock_irqrestore(&worker->lock, flags);
+
+	return ret;
+}
+
+static enum io_wq_cancel io_wqe_cancel_work(struct io_wqe *wqe,
+					    struct io_cb_cancel_data *match)
+{
+	struct io_wq_work_node *node, *prev;
+	struct io_wq_work *work;
+	unsigned long flags;
+	bool found = false;
+
+	/*
+	 * First check pending list, if we're lucky we can just remove it
+	 * from there. CANCEL_OK means that the work is returned as-new,
+	 * no completion will be posted for it.
+	 */
+	spin_lock_irqsave(&wqe->lock, flags);
+	wq_list_for_each(node, prev, &wqe->work_list) {
+		work = container_of(node, struct io_wq_work, list);
+
+		if (match->fn(work, match->data)) {
+			wq_list_del(&wqe->work_list, node, prev);
+			found = true;
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&wqe->lock, flags);
+
+	if (found) {
+		io_run_cancel(work, wqe);
+		return IO_WQ_CANCEL_OK;
+	}
+
+	/*
+	 * Now check if a free (going busy) or busy worker has the work
+	 * currently running. If we find it there, we'll return CANCEL_RUNNING
+	 * as an indication that we attempt to signal cancellation. The
+	 * completion will run normally in this case.
+	 */
+	rcu_read_lock();
+	found = io_wq_for_each_worker(wqe, io_wq_worker_cancel, match);
+	rcu_read_unlock();
+	return found ? IO_WQ_CANCEL_RUNNING : IO_WQ_CANCEL_NOTFOUND;
+}
+
+enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel,
+				  void *data)
+{
+	struct io_cb_cancel_data match = {
+		.fn	= cancel,
+		.data	= data,
+	};
+	enum io_wq_cancel ret = IO_WQ_CANCEL_NOTFOUND;
+	int node;
+
+	for_each_node(node) {
+		struct io_wqe *wqe = wq->wqes[node];
+
+		ret = io_wqe_cancel_work(wqe, &match);
+		if (ret != IO_WQ_CANCEL_NOTFOUND)
+			break;
+	}
+
+	return ret;
+}
+
+static bool io_wq_io_cb_cancel_data(struct io_wq_work *work, void *data)
+{
+	return work == data;
+}
+
+enum io_wq_cancel io_wq_cancel_work(struct io_wq *wq, struct io_wq_work *cwork)
+{
+	return io_wq_cancel_cb(wq, io_wq_io_cb_cancel_data, (void *)cwork);
+}
+
+static bool io_wq_pid_match(struct io_wq_work *work, void *data)
+{
+	pid_t pid = (pid_t) (unsigned long) data;
+
+	return work->task_pid == pid;
+}
+
+enum io_wq_cancel io_wq_cancel_pid(struct io_wq *wq, pid_t pid)
+{
+	void *data = (void *) (unsigned long) pid;
+
+	return io_wq_cancel_cb(wq, io_wq_pid_match, data);
+}
+
+struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data)
+{
+	int ret = -ENOMEM, node;
+	struct io_wq *wq;
+
+	if (WARN_ON_ONCE(!data->free_work))
+		return ERR_PTR(-EINVAL);
+
+	wq = kzalloc(sizeof(*wq), GFP_KERNEL);
+	if (!wq)
+		return ERR_PTR(-ENOMEM);
+
+	wq->wqes = kcalloc(nr_node_ids, sizeof(struct io_wqe *), GFP_KERNEL);
+	if (!wq->wqes) {
+		kfree(wq);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	wq->free_work = data->free_work;
+
+	/* caller must already hold a reference to this */
+	wq->user = data->user;
+
+	for_each_node(node) {
+		struct io_wqe *wqe;
+		int alloc_node = node;
+
+		if (!node_online(alloc_node))
+			alloc_node = NUMA_NO_NODE;
+		wqe = kzalloc_node(sizeof(struct io_wqe), GFP_KERNEL, alloc_node);
+		if (!wqe)
+			goto err;
+		wq->wqes[node] = wqe;
+		wqe->node = alloc_node;
+		wqe->acct[IO_WQ_ACCT_BOUND].max_workers = bounded;
+		atomic_set(&wqe->acct[IO_WQ_ACCT_BOUND].nr_running, 0);
+		if (wq->user) {
+			wqe->acct[IO_WQ_ACCT_UNBOUND].max_workers =
+					task_rlimit(current, RLIMIT_NPROC);
+		}
+		atomic_set(&wqe->acct[IO_WQ_ACCT_UNBOUND].nr_running, 0);
+		wqe->wq = wq;
+		spin_lock_init(&wqe->lock);
+		INIT_WQ_LIST(&wqe->work_list);
+		INIT_HLIST_NULLS_HEAD(&wqe->free_list, 0);
+		INIT_LIST_HEAD(&wqe->all_list);
+	}
+
+	init_completion(&wq->done);
+
+	wq->manager = kthread_create(io_wq_manager, wq, "io_wq_manager");
+	if (!IS_ERR(wq->manager)) {
+		wake_up_process(wq->manager);
+		wait_for_completion(&wq->done);
+		if (test_bit(IO_WQ_BIT_ERROR, &wq->state)) {
+			ret = -ENOMEM;
+			goto err;
+		}
+		refcount_set(&wq->use_refs, 1);
+		reinit_completion(&wq->done);
+		return wq;
+	}
+
+	ret = PTR_ERR(wq->manager);
+	complete(&wq->done);
+err:
+	for_each_node(node)
+		kfree(wq->wqes[node]);
+	kfree(wq->wqes);
+	kfree(wq);
+	return ERR_PTR(ret);
+}
+
+bool io_wq_get(struct io_wq *wq, struct io_wq_data *data)
+{
+	if (data->free_work != wq->free_work)
+		return false;
+
+	return refcount_inc_not_zero(&wq->use_refs);
+}
+
+static bool io_wq_worker_wake(struct io_worker *worker, void *data)
+{
+	wake_up_process(worker->task);
+	return false;
+}
+
+static void __io_wq_destroy(struct io_wq *wq)
+{
+	int node;
+
+	set_bit(IO_WQ_BIT_EXIT, &wq->state);
+	if (wq->manager)
+		kthread_stop(wq->manager);
+
+	rcu_read_lock();
+	for_each_node(node)
+		io_wq_for_each_worker(wq->wqes[node], io_wq_worker_wake, NULL);
+	rcu_read_unlock();
+
+	wait_for_completion(&wq->done);
+
+	for_each_node(node)
+		kfree(wq->wqes[node]);
+	kfree(wq->wqes);
+	kfree(wq);
+}
+
+void io_wq_destroy(struct io_wq *wq)
+{
+	if (refcount_dec_and_test(&wq->use_refs))
+		__io_wq_destroy(wq);
+}
+
+struct task_struct *io_wq_get_task(struct io_wq *wq)
+{
+	return wq->manager;
+}
diff --git a/fs/io-wq.h b/fs/io-wq.h
new file mode 100644
index 000000000000..5ba12de7572f
--- /dev/null
+++ b/fs/io-wq.h
@@ -0,0 +1,157 @@
+#ifndef INTERNAL_IO_WQ_H
+#define INTERNAL_IO_WQ_H
+
+struct io_wq;
+
+enum {
+	IO_WQ_WORK_CANCEL	= 1,
+	IO_WQ_WORK_HASHED	= 4,
+	IO_WQ_WORK_UNBOUND	= 32,
+	IO_WQ_WORK_NO_CANCEL	= 256,
+	IO_WQ_WORK_CONCURRENT	= 512,
+
+	IO_WQ_HASH_SHIFT	= 24,	/* upper 8 bits are used for hash key */
+};
+
+enum io_wq_cancel {
+	IO_WQ_CANCEL_OK,	/* cancelled before started */
+	IO_WQ_CANCEL_RUNNING,	/* found, running, and attempted cancelled */
+	IO_WQ_CANCEL_NOTFOUND,	/* work not found */
+};
+
+struct io_wq_work_node {
+	struct io_wq_work_node *next;
+};
+
+struct io_wq_work_list {
+	struct io_wq_work_node *first;
+	struct io_wq_work_node *last;
+};
+
+static inline void wq_list_add_after(struct io_wq_work_node *node,
+				     struct io_wq_work_node *pos,
+				     struct io_wq_work_list *list)
+{
+	struct io_wq_work_node *next = pos->next;
+
+	pos->next = node;
+	node->next = next;
+	if (!next)
+		list->last = node;
+}
+
+static inline void wq_list_add_tail(struct io_wq_work_node *node,
+				    struct io_wq_work_list *list)
+{
+	if (!list->first) {
+		list->last = node;
+		WRITE_ONCE(list->first, node);
+	} else {
+		list->last->next = node;
+		list->last = node;
+	}
+}
+
+static inline void wq_list_cut(struct io_wq_work_list *list,
+			       struct io_wq_work_node *last,
+			       struct io_wq_work_node *prev)
+{
+	/* first in the list, if prev==NULL */
+	if (!prev)
+		WRITE_ONCE(list->first, last->next);
+	else
+		prev->next = last->next;
+
+	if (last == list->last)
+		list->last = prev;
+	last->next = NULL;
+}
+
+static inline void wq_list_del(struct io_wq_work_list *list,
+			       struct io_wq_work_node *node,
+			       struct io_wq_work_node *prev)
+{
+	wq_list_cut(list, node, prev);
+}
+
+#define wq_list_for_each(pos, prv, head)			\
+	for (pos = (head)->first, prv = NULL; pos; prv = pos, pos = (pos)->next)
+
+#define wq_list_empty(list)	(READ_ONCE((list)->first) == NULL)
+#define INIT_WQ_LIST(list)	do {				\
+	(list)->first = NULL;					\
+	(list)->last = NULL;					\
+} while (0)
+
+struct io_wq_work {
+	struct io_wq_work_node list;
+	void (*func)(struct io_wq_work **);
+	struct files_struct *files;
+	struct mm_struct *mm;
+	const struct cred *creds;
+	struct fs_struct *fs;
+	unsigned flags;
+	pid_t task_pid;
+};
+
+#define INIT_IO_WORK(work, _func)				\
+	do {							\
+		*(work) = (struct io_wq_work){ .func = _func };	\
+	} while (0)						\
+
+static inline struct io_wq_work *wq_next_work(struct io_wq_work *work)
+{
+	if (!work->list.next)
+		return NULL;
+
+	return container_of(work->list.next, struct io_wq_work, list);
+}
+
+typedef void (free_work_fn)(struct io_wq_work *);
+
+struct io_wq_data {
+	struct user_struct *user;
+
+	free_work_fn *free_work;
+};
+
+struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data);
+bool io_wq_get(struct io_wq *wq, struct io_wq_data *data);
+void io_wq_destroy(struct io_wq *wq);
+
+void io_wq_enqueue(struct io_wq *wq, struct io_wq_work *work);
+void io_wq_hash_work(struct io_wq_work *work, void *val);
+
+static inline bool io_wq_is_hashed(struct io_wq_work *work)
+{
+	return work->flags & IO_WQ_WORK_HASHED;
+}
+
+void io_wq_cancel_all(struct io_wq *wq);
+enum io_wq_cancel io_wq_cancel_work(struct io_wq *wq, struct io_wq_work *cwork);
+enum io_wq_cancel io_wq_cancel_pid(struct io_wq *wq, pid_t pid);
+
+typedef bool (work_cancel_fn)(struct io_wq_work *, void *);
+
+enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel,
+					void *data);
+
+struct task_struct *io_wq_get_task(struct io_wq *wq);
+
+#if defined(CONFIG_IO_WQ)
+extern void io_wq_worker_sleeping(struct task_struct *);
+extern void io_wq_worker_running(struct task_struct *);
+#else
+static inline void io_wq_worker_sleeping(struct task_struct *tsk)
+{
+}
+static inline void io_wq_worker_running(struct task_struct *tsk)
+{
+}
+#endif
+
+static inline bool io_wq_current_is_worker(void)
+{
+	return in_task() && (current->flags & PF_IO_WORKER);
+}
+#endif
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 2c819c3c855d..9d4bd0d3a080 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -44,8 +44,10 @@
 #include <linux/errno.h>
 #include <linux/syscalls.h>
 #include <linux/compat.h>
+#include <net/compat.h>
 #include <linux/refcount.h>
 #include <linux/uio.h>
+#include <linux/bits.h>
 
 #include <linux/sched/signal.h>
 #include <linux/fs.h>
@@ -56,7 +58,6 @@
 #include <linux/mmu_context.h>
 #include <linux/percpu.h>
 #include <linux/slab.h>
-#include <linux/workqueue.h>
 #include <linux/kthread.h>
 #include <linux/blkdev.h>
 #include <linux/bvec.h>
@@ -70,13 +71,33 @@
 #include <linux/nospec.h>
 #include <linux/sizes.h>
 #include <linux/hugetlb.h>
+#include <linux/highmem.h>
+#include <linux/namei.h>
+#include <linux/fsnotify.h>
+#include <linux/fadvise.h>
+#include <linux/eventpoll.h>
+#include <linux/fs_struct.h>
+#include <linux/splice.h>
+#include <linux/task_work.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/io_uring.h>
 
 #include <uapi/linux/io_uring.h>
 
 #include "internal.h"
+#include "io-wq.h"
 
 #define IORING_MAX_ENTRIES	32768
-#define IORING_MAX_FIXED_FILES	1024
+#define IORING_MAX_CQ_ENTRIES	(2 * IORING_MAX_ENTRIES)
+
+/*
+ * Shift of 9 is 512 entries, or exactly one page on 64-bit archs
+ */
+#define IORING_FILE_TABLE_SHIFT	9
+#define IORING_MAX_FILES_TABLE	(1U << IORING_FILE_TABLE_SHIFT)
+#define IORING_FILE_TABLE_MASK	(IORING_MAX_FILES_TABLE - 1)
+#define IORING_MAX_FIXED_FILES	(64 * IORING_MAX_FILES_TABLE)
 
 struct io_uring {
 	u32 head ____cacheline_aligned_in_smp;
@@ -121,7 +142,7 @@ struct io_rings {
 	 */
 	u32			sq_dropped;
 	/*
-	 * Runtime flags
+	 * Runtime SQ flags
 	 *
 	 * Written by the kernel, shouldn't be modified by the
 	 * application.
@@ -131,9 +152,16 @@ struct io_rings {
 	 */
 	u32			sq_flags;
 	/*
+	 * Runtime CQ flags
+	 *
+	 * Written by the application, shouldn't be modified by the
+	 * kernel.
+	 */
+	u32                     cq_flags;
+	/*
 	 * Number of completion events lost because the queue was full;
 	 * this should be avoided by the application by making sure
-	 * there are not more requests pending thatn there is space in
+	 * there are not more requests pending than there is space in
 	 * the completion queue.
 	 *
 	 * Written by the kernel, shouldn't be modified by the
@@ -161,14 +189,34 @@ struct io_mapped_ubuf {
 	unsigned int	nr_bvecs;
 };
 
-struct async_list {
-	spinlock_t		lock;
-	atomic_t		cnt;
-	struct list_head	list;
+struct fixed_file_table {
+	struct file		**files;
+};
 
-	struct file		*file;
-	off_t			io_start;
-	size_t			io_len;
+struct fixed_file_ref_node {
+	struct percpu_ref		refs;
+	struct list_head		node;
+	struct list_head		file_list;
+	struct fixed_file_data		*file_data;
+	struct llist_node		llist;
+};
+
+struct fixed_file_data {
+	struct fixed_file_table		*table;
+	struct io_ring_ctx		*ctx;
+
+	struct percpu_ref		*cur_refs;
+	struct percpu_ref		refs;
+	struct completion		done;
+	struct list_head		ref_list;
+	spinlock_t			lock;
+};
+
+struct io_buffer {
+	struct list_head list;
+	__u64 addr;
+	__s32 len;
+	__u16 bid;
 };
 
 struct io_ring_ctx {
@@ -178,8 +226,11 @@ struct io_ring_ctx {
 
 	struct {
 		unsigned int		flags;
-		bool			compat;
-		bool			account_mem;
+		unsigned int		compat: 1;
+		unsigned int		account_mem: 1;
+		unsigned int		cq_overflow_flushed: 1;
+		unsigned int		drain_next: 1;
+		unsigned int		eventfd_async: 1;
 
 		/*
 		 * Ring buffer of indices into array of io_uring_sqe, which is
@@ -198,39 +249,34 @@ struct io_ring_ctx {
 		unsigned		sq_mask;
 		unsigned		sq_thread_idle;
 		unsigned		cached_sq_dropped;
-		struct io_uring_sqe	*sq_sqes;
+		atomic_t		cached_cq_overflow;
+		unsigned long		sq_check_overflow;
 
 		struct list_head	defer_list;
 		struct list_head	timeout_list;
+		struct list_head	cq_overflow_list;
+
+		wait_queue_head_t	inflight_wait;
+		struct io_uring_sqe	*sq_sqes;
 	} ____cacheline_aligned_in_smp;
 
+	struct io_rings	*rings;
+
 	/* IO offload */
-	struct workqueue_struct	*sqo_wq[2];
+	struct io_wq		*io_wq;
 	struct task_struct	*sqo_thread;	/* if using sq thread polling */
 	struct mm_struct	*sqo_mm;
 	wait_queue_head_t	sqo_wait;
-	struct completion	sqo_thread_started;
-
-	struct {
-		unsigned		cached_cq_tail;
-		atomic_t		cached_cq_overflow;
-		unsigned		cq_entries;
-		unsigned		cq_mask;
-		struct wait_queue_head	cq_wait;
-		struct fasync_struct	*cq_fasync;
-		struct eventfd_ctx	*cq_ev_fd;
-		atomic_t		cq_timeouts;
-	} ____cacheline_aligned_in_smp;
-
-	struct io_rings	*rings;
 
 	/*
 	 * If used, fixed file set. Writers must ensure that ->refs is dead,
 	 * readers must ensure that ->refs is alive as long as the file* is
 	 * used. Only updated through io_uring_register(2).
 	 */
-	struct file		**user_files;
+	struct fixed_file_data	*file_data;
 	unsigned		nr_user_files;
+	int 			ring_fd;
+	struct file 		*ring_file;
 
 	/* if used, fixed mapped user buffers */
 	unsigned		nr_user_bufs;
@@ -238,7 +284,32 @@ struct io_ring_ctx {
 
 	struct user_struct	*user;
 
-	struct completion	ctx_done;
+	const struct cred	*creds;
+
+	struct completion	ref_comp;
+	struct completion	sq_thread_comp;
+
+	/* if all else fails... */
+	struct io_kiocb		*fallback_req;
+
+#if defined(CONFIG_UNIX)
+	struct socket		*ring_sock;
+#endif
+
+	struct idr		io_buffer_idr;
+
+	struct idr		personality_idr;
+
+	struct {
+		unsigned		cached_cq_tail;
+		unsigned		cq_entries;
+		unsigned		cq_mask;
+		atomic_t		cq_timeouts;
+		unsigned long		cq_check_overflow;
+		struct wait_queue_head	cq_wait;
+		struct fasync_struct	*cq_fasync;
+		struct eventfd_ctx	*cq_ev_fd;
+	} ____cacheline_aligned_in_smp;
 
 	struct {
 		struct mutex		uring_lock;
@@ -247,7 +318,7 @@ struct io_ring_ctx {
 
 	struct {
 		spinlock_t		completion_lock;
-		bool			poll_multi_file;
+
 		/*
 		 * ->poll_list is protected by the ctx->uring_lock for
 		 * io_uring instances that don't use IORING_SETUP_SQPOLL.
@@ -255,23 +326,18 @@ struct io_ring_ctx {
 		 * manipulate the list, hence no extra locking is needed there.
 		 */
 		struct list_head	poll_list;
-		struct list_head	cancel_list;
-	} ____cacheline_aligned_in_smp;
+		struct hlist_head	*cancel_hash;
+		unsigned		cancel_hash_bits;
+		bool			poll_multi_file;
 
-	struct async_list	pending_async[2];
+		spinlock_t		inflight_lock;
+		struct list_head	inflight_list;
+	} ____cacheline_aligned_in_smp;
 
-#if defined(CONFIG_UNIX)
-	struct socket		*ring_sock;
-#endif
-};
+	struct delayed_work		file_put_work;
+	struct llist_head		file_put_llist;
 
-struct sqe_submit {
-	const struct io_uring_sqe	*sqe;
-	unsigned short			index;
-	u32				sequence;
-	bool				has_user;
-	bool				needs_lock;
-	bool				needs_fixed_file;
+	struct work_struct		exit_work;
 };
 
 /*
@@ -280,16 +346,261 @@ struct sqe_submit {
  */
 struct io_poll_iocb {
 	struct file			*file;
-	struct wait_queue_head		*head;
+	union {
+		struct wait_queue_head	*head;
+		u64			addr;
+	};
 	__poll_t			events;
 	bool				done;
 	bool				canceled;
 	struct wait_queue_entry		wait;
 };
 
-struct io_timeout {
+struct io_close {
 	struct file			*file;
+	struct file			*put_file;
+	int				fd;
+};
+
+struct io_timeout_data {
+	struct io_kiocb			*req;
 	struct hrtimer			timer;
+	struct timespec64		ts;
+	enum hrtimer_mode		mode;
+};
+
+struct io_accept {
+	struct file			*file;
+	struct sockaddr __user		*addr;
+	int __user			*addr_len;
+	int				flags;
+	unsigned long			nofile;
+};
+
+struct io_sync {
+	struct file			*file;
+	loff_t				len;
+	loff_t				off;
+	int				flags;
+	int				mode;
+};
+
+struct io_cancel {
+	struct file			*file;
+	u64				addr;
+};
+
+struct io_timeout {
+	struct file			*file;
+	u64				addr;
+	int				flags;
+	u32				off;
+	u32				target_seq;
+};
+
+struct io_rw {
+	/* NOTE: kiocb has the file as the first member, so don't do it here */
+	struct kiocb			kiocb;
+	u64				addr;
+	u64				len;
+};
+
+struct io_connect {
+	struct file			*file;
+	struct sockaddr __user		*addr;
+	int				addr_len;
+};
+
+struct io_sr_msg {
+	struct file			*file;
+	union {
+		struct user_msghdr __user *msg;
+		void __user		*buf;
+	};
+	int				msg_flags;
+	int				bgid;
+	size_t				len;
+	struct io_buffer		*kbuf;
+};
+
+struct io_open {
+	struct file			*file;
+	int				dfd;
+	struct filename			*filename;
+	struct open_how			how;
+	unsigned long			nofile;
+};
+
+struct io_files_update {
+	struct file			*file;
+	u64				arg;
+	u32				nr_args;
+	u32				offset;
+};
+
+struct io_fadvise {
+	struct file			*file;
+	u64				offset;
+	u32				len;
+	u32				advice;
+};
+
+struct io_madvise {
+	struct file			*file;
+	u64				addr;
+	u32				len;
+	u32				advice;
+};
+
+struct io_epoll {
+	struct file			*file;
+	int				epfd;
+	int				op;
+	int				fd;
+	struct epoll_event		event;
+};
+
+struct io_splice {
+	struct file			*file_out;
+	struct file			*file_in;
+	loff_t				off_out;
+	loff_t				off_in;
+	u64				len;
+	unsigned int			flags;
+};
+
+struct io_provide_buf {
+	struct file			*file;
+	__u64				addr;
+	__s32				len;
+	__u32				bgid;
+	__u16				nbufs;
+	__u16				bid;
+};
+
+struct io_statx {
+	struct file			*file;
+	int				dfd;
+	unsigned int			mask;
+	unsigned int			flags;
+	const char __user		*filename;
+	struct statx __user		*buffer;
+};
+
+struct io_async_connect {
+	struct sockaddr_storage		address;
+};
+
+struct io_async_msghdr {
+	struct iovec			fast_iov[UIO_FASTIOV];
+	struct iovec			*iov;
+	struct sockaddr __user		*uaddr;
+	struct msghdr			msg;
+	struct sockaddr_storage		addr;
+};
+
+struct io_async_rw {
+	struct iovec			fast_iov[UIO_FASTIOV];
+	struct iovec			*iov;
+	ssize_t				nr_segs;
+	ssize_t				size;
+};
+
+struct io_async_ctx {
+	union {
+		struct io_async_rw	rw;
+		struct io_async_msghdr	msg;
+		struct io_async_connect	connect;
+		struct io_timeout_data	timeout;
+	};
+};
+
+enum {
+	REQ_F_FIXED_FILE_BIT	= IOSQE_FIXED_FILE_BIT,
+	REQ_F_IO_DRAIN_BIT	= IOSQE_IO_DRAIN_BIT,
+	REQ_F_LINK_BIT		= IOSQE_IO_LINK_BIT,
+	REQ_F_HARDLINK_BIT	= IOSQE_IO_HARDLINK_BIT,
+	REQ_F_FORCE_ASYNC_BIT	= IOSQE_ASYNC_BIT,
+	REQ_F_BUFFER_SELECT_BIT	= IOSQE_BUFFER_SELECT_BIT,
+
+	REQ_F_LINK_HEAD_BIT,
+	REQ_F_LINK_NEXT_BIT,
+	REQ_F_FAIL_LINK_BIT,
+	REQ_F_INFLIGHT_BIT,
+	REQ_F_CUR_POS_BIT,
+	REQ_F_NOWAIT_BIT,
+	REQ_F_IOPOLL_COMPLETED_BIT,
+	REQ_F_LINK_TIMEOUT_BIT,
+	REQ_F_TIMEOUT_BIT,
+	REQ_F_ISREG_BIT,
+	REQ_F_MUST_PUNT_BIT,
+	REQ_F_TIMEOUT_NOSEQ_BIT,
+	REQ_F_COMP_LOCKED_BIT,
+	REQ_F_NEED_CLEANUP_BIT,
+	REQ_F_OVERFLOW_BIT,
+	REQ_F_POLLED_BIT,
+	REQ_F_BUFFER_SELECTED_BIT,
+	REQ_F_NO_FILE_TABLE_BIT,
+
+	/* not a real bit, just to check we're not overflowing the space */
+	__REQ_F_LAST_BIT,
+};
+
+enum {
+	/* ctx owns file */
+	REQ_F_FIXED_FILE	= BIT(REQ_F_FIXED_FILE_BIT),
+	/* drain existing IO first */
+	REQ_F_IO_DRAIN		= BIT(REQ_F_IO_DRAIN_BIT),
+	/* linked sqes */
+	REQ_F_LINK		= BIT(REQ_F_LINK_BIT),
+	/* doesn't sever on completion < 0 */
+	REQ_F_HARDLINK		= BIT(REQ_F_HARDLINK_BIT),
+	/* IOSQE_ASYNC */
+	REQ_F_FORCE_ASYNC	= BIT(REQ_F_FORCE_ASYNC_BIT),
+	/* IOSQE_BUFFER_SELECT */
+	REQ_F_BUFFER_SELECT	= BIT(REQ_F_BUFFER_SELECT_BIT),
+
+	/* head of a link */
+	REQ_F_LINK_HEAD		= BIT(REQ_F_LINK_HEAD_BIT),
+	/* already grabbed next link */
+	REQ_F_LINK_NEXT		= BIT(REQ_F_LINK_NEXT_BIT),
+	/* fail rest of links */
+	REQ_F_FAIL_LINK		= BIT(REQ_F_FAIL_LINK_BIT),
+	/* on inflight list */
+	REQ_F_INFLIGHT		= BIT(REQ_F_INFLIGHT_BIT),
+	/* read/write uses file position */
+	REQ_F_CUR_POS		= BIT(REQ_F_CUR_POS_BIT),
+	/* must not punt to workers */
+	REQ_F_NOWAIT		= BIT(REQ_F_NOWAIT_BIT),
+	/* polled IO has completed */
+	REQ_F_IOPOLL_COMPLETED	= BIT(REQ_F_IOPOLL_COMPLETED_BIT),
+	/* has linked timeout */
+	REQ_F_LINK_TIMEOUT	= BIT(REQ_F_LINK_TIMEOUT_BIT),
+	/* timeout request */
+	REQ_F_TIMEOUT		= BIT(REQ_F_TIMEOUT_BIT),
+	/* regular file */
+	REQ_F_ISREG		= BIT(REQ_F_ISREG_BIT),
+	/* must be punted even for NONBLOCK */
+	REQ_F_MUST_PUNT		= BIT(REQ_F_MUST_PUNT_BIT),
+	/* no timeout sequence */
+	REQ_F_TIMEOUT_NOSEQ	= BIT(REQ_F_TIMEOUT_NOSEQ_BIT),
+	/* completion under lock */
+	REQ_F_COMP_LOCKED	= BIT(REQ_F_COMP_LOCKED_BIT),
+	/* needs cleanup */
+	REQ_F_NEED_CLEANUP	= BIT(REQ_F_NEED_CLEANUP_BIT),
+	/* in overflow list */
+	REQ_F_OVERFLOW		= BIT(REQ_F_OVERFLOW_BIT),
+	/* already went through poll handler */
+	REQ_F_POLLED		= BIT(REQ_F_POLLED_BIT),
+	/* buffer already selected */
+	REQ_F_BUFFER_SELECTED	= BIT(REQ_F_BUFFER_SELECTED_BIT),
+	/* doesn't need file table for this request */
+	REQ_F_NO_FILE_TABLE	= BIT(REQ_F_NO_FILE_TABLE_BIT),
+};
+
+struct async_poll {
+	struct io_poll_iocb	poll;
+	struct io_wq_work	work;
 };
 
 /*
@@ -301,37 +612,61 @@ struct io_timeout {
 struct io_kiocb {
 	union {
 		struct file		*file;
-		struct kiocb		rw;
+		struct io_rw		rw;
 		struct io_poll_iocb	poll;
+		struct io_accept	accept;
+		struct io_sync		sync;
+		struct io_cancel	cancel;
 		struct io_timeout	timeout;
+		struct io_connect	connect;
+		struct io_sr_msg	sr_msg;
+		struct io_open		open;
+		struct io_close		close;
+		struct io_files_update	files_update;
+		struct io_fadvise	fadvise;
+		struct io_madvise	madvise;
+		struct io_epoll		epoll;
+		struct io_splice	splice;
+		struct io_provide_buf	pbuf;
+		struct io_statx		statx;
 	};
 
-	struct sqe_submit	submit;
+	struct io_async_ctx		*io;
+	int				cflags;
+	u8				opcode;
+
+	u16				buf_index;
 
 	struct io_ring_ctx	*ctx;
 	struct list_head	list;
-	struct list_head	link_list;
 	unsigned int		flags;
 	refcount_t		refs;
-#define REQ_F_NOWAIT		1	/* must not punt to workers */
-#define REQ_F_IOPOLL_COMPLETED	2	/* polled IO has completed */
-#define REQ_F_FIXED_FILE	4	/* ctx owns file */
-#define REQ_F_SEQ_PREV		8	/* sequential with previous */
-#define REQ_F_IO_DRAIN		16	/* drain existing IO first */
-#define REQ_F_IO_DRAINED	32	/* drain done */
-#define REQ_F_LINK		64	/* linked sqes */
-#define REQ_F_LINK_DONE		128	/* linked sqes done */
-#define REQ_F_FAIL_LINK		256	/* fail rest of links */
-#define REQ_F_SHADOW_DRAIN	512	/* link-drain shadow req */
-#define REQ_F_TIMEOUT		1024	/* timeout request */
-#define REQ_F_ISREG		2048	/* regular file */
-#define REQ_F_MUST_PUNT		4096	/* must be punted even for NONBLOCK */
-#define REQ_F_TIMEOUT_NOSEQ	8192	/* no timeout sequence */
+	struct task_struct	*task;
+	unsigned long		fsize;
 	u64			user_data;
 	u32			result;
 	u32			sequence;
 
-	struct work_struct	work;
+	struct list_head	link_list;
+
+	struct list_head	inflight_entry;
+
+	struct percpu_ref	*fixed_file_refs;
+
+	union {
+		/*
+		 * Only commands that never go async can use the below fields,
+		 * obviously. Right now only IORING_OP_POLL_ADD uses them, and
+		 * async armed poll handlers for regular commands. The latter
+		 * restore the work, if needed.
+		 */
+		struct {
+			struct callback_head	task_work;
+			struct hlist_node	hash_node;
+			struct async_poll	*apoll;
+		};
+		struct io_wq_work	work;
+	};
 };
 
 #define IO_PLUG_THRESHOLD		2
@@ -344,8 +679,7 @@ struct io_submit_state {
 	 * io_kiocb alloc cache
 	 */
 	void			*reqs[IO_IOPOLL_BATCH];
-	unsigned		int free_reqs;
-	unsigned		int cur_req;
+	unsigned int		free_reqs;
 
 	/*
 	 * File reference cache
@@ -357,10 +691,199 @@ struct io_submit_state {
 	unsigned int		ios_left;
 };
 
-static void io_sq_wq_submit_work(struct work_struct *work);
-static void io_cqring_fill_event(struct io_ring_ctx *ctx, u64 ki_user_data,
-				 long res);
-static void __io_free_req(struct io_kiocb *req);
+struct io_op_def {
+	/* needs req->io allocated for deferral/async */
+	unsigned		async_ctx : 1;
+	/* needs current->mm setup, does mm access */
+	unsigned		needs_mm : 1;
+	/* needs req->file assigned */
+	unsigned		needs_file : 1;
+	/* hash wq insertion if file is a regular file */
+	unsigned		hash_reg_file : 1;
+	/* unbound wq insertion if file is a non-regular file */
+	unsigned		unbound_nonreg_file : 1;
+	/* opcode is not supported by this kernel */
+	unsigned		not_supported : 1;
+	/* needs file table */
+	unsigned		file_table : 1;
+	/* needs ->fs */
+	unsigned		needs_fs : 1;
+	/* set if opcode supports polled "wait" */
+	unsigned		pollin : 1;
+	unsigned		pollout : 1;
+	/* op supports buffer selection */
+	unsigned		buffer_select : 1;
+};
+
+static const struct io_op_def io_op_defs[] = {
+	[IORING_OP_NOP] = {},
+	[IORING_OP_READV] = {
+		.async_ctx		= 1,
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollin			= 1,
+		.buffer_select		= 1,
+	},
+	[IORING_OP_WRITEV] = {
+		.async_ctx		= 1,
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.hash_reg_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollout		= 1,
+	},
+	[IORING_OP_FSYNC] = {
+		.needs_file		= 1,
+	},
+	[IORING_OP_READ_FIXED] = {
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollin			= 1,
+	},
+	[IORING_OP_WRITE_FIXED] = {
+		.needs_file		= 1,
+		.hash_reg_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollout		= 1,
+	},
+	[IORING_OP_POLL_ADD] = {
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+	},
+	[IORING_OP_POLL_REMOVE] = {},
+	[IORING_OP_SYNC_FILE_RANGE] = {
+		.needs_file		= 1,
+	},
+	[IORING_OP_SENDMSG] = {
+		.async_ctx		= 1,
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.needs_fs		= 1,
+		.pollout		= 1,
+	},
+	[IORING_OP_RECVMSG] = {
+		.async_ctx		= 1,
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.needs_fs		= 1,
+		.pollin			= 1,
+		.buffer_select		= 1,
+	},
+	[IORING_OP_TIMEOUT] = {
+		.async_ctx		= 1,
+		.needs_mm		= 1,
+	},
+	[IORING_OP_TIMEOUT_REMOVE] = {},
+	[IORING_OP_ACCEPT] = {
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.file_table		= 1,
+		.pollin			= 1,
+	},
+	[IORING_OP_ASYNC_CANCEL] = {},
+	[IORING_OP_LINK_TIMEOUT] = {
+		.async_ctx		= 1,
+		.needs_mm		= 1,
+	},
+	[IORING_OP_CONNECT] = {
+		.async_ctx		= 1,
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollout		= 1,
+	},
+	[IORING_OP_FALLOCATE] = {
+		.needs_file		= 1,
+	},
+	[IORING_OP_OPENAT] = {
+		.file_table		= 1,
+		.needs_fs		= 1,
+	},
+	[IORING_OP_CLOSE] = {
+		.file_table		= 1,
+	},
+	[IORING_OP_FILES_UPDATE] = {
+		.needs_mm		= 1,
+		.file_table		= 1,
+	},
+	[IORING_OP_STATX] = {
+		.needs_mm		= 1,
+		.needs_fs		= 1,
+		.file_table		= 1,
+	},
+	[IORING_OP_READ] = {
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollin			= 1,
+		.buffer_select		= 1,
+	},
+	[IORING_OP_WRITE] = {
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollout		= 1,
+	},
+	[IORING_OP_FADVISE] = {
+		.needs_file		= 1,
+	},
+	[IORING_OP_MADVISE] = {
+		.needs_mm		= 1,
+	},
+	[IORING_OP_SEND] = {
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollout		= 1,
+	},
+	[IORING_OP_RECV] = {
+		.needs_mm		= 1,
+		.needs_file		= 1,
+		.unbound_nonreg_file	= 1,
+		.pollin			= 1,
+		.buffer_select		= 1,
+	},
+	[IORING_OP_OPENAT2] = {
+		.file_table		= 1,
+		.needs_fs		= 1,
+	},
+	[IORING_OP_EPOLL_CTL] = {
+		.unbound_nonreg_file	= 1,
+		.file_table		= 1,
+	},
+	[IORING_OP_SPLICE] = {
+		.needs_file		= 1,
+		.hash_reg_file		= 1,
+		.unbound_nonreg_file	= 1,
+	},
+	[IORING_OP_PROVIDE_BUFFERS] = {},
+	[IORING_OP_REMOVE_BUFFERS] = {},
+	[IORING_OP_TEE] = {
+		.needs_file		= 1,
+		.hash_reg_file		= 1,
+		.unbound_nonreg_file	= 1,
+	},
+};
+
+static void io_wq_submit_work(struct io_wq_work **workptr);
+static void io_cqring_fill_event(struct io_kiocb *req, long res);
+static void io_put_req(struct io_kiocb *req);
+static void __io_double_put_req(struct io_kiocb *req);
+static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req);
+static void io_queue_linked_timeout(struct io_kiocb *req);
+static int __io_sqe_files_update(struct io_ring_ctx *ctx,
+				 struct io_uring_files_update *ip,
+				 unsigned nr_args);
+static int io_grab_files(struct io_kiocb *req);
+static void io_cleanup_req(struct io_kiocb *req);
+static int io_file_get(struct io_submit_state *state, struct io_kiocb *req,
+		       int fd, struct file **out_file, bool fixed);
+static void __io_queue_sqe(struct io_kiocb *req,
+			   const struct io_uring_sqe *sqe);
 
 static struct kmem_cache *req_cachep;
 
@@ -379,135 +902,198 @@ struct sock *io_uring_get_socket(struct file *file)
 }
 EXPORT_SYMBOL(io_uring_get_socket);
 
+static void io_file_put_work(struct work_struct *work);
+
+static inline bool io_async_submit(struct io_ring_ctx *ctx)
+{
+	return ctx->flags & IORING_SETUP_SQPOLL;
+}
+
 static void io_ring_ctx_ref_free(struct percpu_ref *ref)
 {
 	struct io_ring_ctx *ctx = container_of(ref, struct io_ring_ctx, refs);
 
-	complete(&ctx->ctx_done);
+	complete(&ctx->ref_comp);
 }
 
 static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
 {
 	struct io_ring_ctx *ctx;
-	int i;
+	int hash_bits;
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
 		return NULL;
 
+	ctx->fallback_req = kmem_cache_alloc(req_cachep, GFP_KERNEL);
+	if (!ctx->fallback_req)
+		goto err;
+
+	/*
+	 * Use 5 bits less than the max cq entries, that should give us around
+	 * 32 entries per hash list if totally full and uniformly spread.
+	 */
+	hash_bits = ilog2(p->cq_entries);
+	hash_bits -= 5;
+	if (hash_bits <= 0)
+		hash_bits = 1;
+	ctx->cancel_hash_bits = hash_bits;
+	ctx->cancel_hash = kmalloc((1U << hash_bits) * sizeof(struct hlist_head),
+					GFP_KERNEL);
+	if (!ctx->cancel_hash)
+		goto err;
+	__hash_init(ctx->cancel_hash, 1U << hash_bits);
+
 	if (percpu_ref_init(&ctx->refs, io_ring_ctx_ref_free,
-			    PERCPU_REF_ALLOW_REINIT, GFP_KERNEL)) {
-		kfree(ctx);
-		return NULL;
-	}
+			    PERCPU_REF_ALLOW_REINIT, GFP_KERNEL))
+		goto err;
 
 	ctx->flags = p->flags;
+	init_waitqueue_head(&ctx->sqo_wait);
 	init_waitqueue_head(&ctx->cq_wait);
-	init_completion(&ctx->ctx_done);
-	init_completion(&ctx->sqo_thread_started);
+	INIT_LIST_HEAD(&ctx->cq_overflow_list);
+	init_completion(&ctx->ref_comp);
+	init_completion(&ctx->sq_thread_comp);
+	idr_init(&ctx->io_buffer_idr);
+	idr_init(&ctx->personality_idr);
 	mutex_init(&ctx->uring_lock);
 	init_waitqueue_head(&ctx->wait);
-	for (i = 0; i < ARRAY_SIZE(ctx->pending_async); i++) {
-		spin_lock_init(&ctx->pending_async[i].lock);
-		INIT_LIST_HEAD(&ctx->pending_async[i].list);
-		atomic_set(&ctx->pending_async[i].cnt, 0);
-	}
 	spin_lock_init(&ctx->completion_lock);
 	INIT_LIST_HEAD(&ctx->poll_list);
-	INIT_LIST_HEAD(&ctx->cancel_list);
 	INIT_LIST_HEAD(&ctx->defer_list);
 	INIT_LIST_HEAD(&ctx->timeout_list);
+	init_waitqueue_head(&ctx->inflight_wait);
+	spin_lock_init(&ctx->inflight_lock);
+	INIT_LIST_HEAD(&ctx->inflight_list);
+	INIT_DELAYED_WORK(&ctx->file_put_work, io_file_put_work);
+	init_llist_head(&ctx->file_put_llist);
 	return ctx;
+err:
+	if (ctx->fallback_req)
+		kmem_cache_free(req_cachep, ctx->fallback_req);
+	kfree(ctx->cancel_hash);
+	kfree(ctx);
+	return NULL;
 }
 
-static inline bool __io_sequence_defer(struct io_ring_ctx *ctx,
-				       struct io_kiocb *req)
+static inline bool __req_need_defer(struct io_kiocb *req)
 {
-	return req->sequence != ctx->cached_cq_tail + ctx->cached_sq_dropped
-					+ atomic_read(&ctx->cached_cq_overflow);
+	struct io_ring_ctx *ctx = req->ctx;
+
+	return req->sequence != ctx->cached_cq_tail
+				+ atomic_read(&ctx->cached_cq_overflow);
 }
 
-static inline bool io_sequence_defer(struct io_ring_ctx *ctx,
-				     struct io_kiocb *req)
+static inline bool req_need_defer(struct io_kiocb *req)
 {
-	if ((req->flags & (REQ_F_IO_DRAIN|REQ_F_IO_DRAINED)) != REQ_F_IO_DRAIN)
-		return false;
+	if (unlikely(req->flags & REQ_F_IO_DRAIN))
+		return __req_need_defer(req);
 
-	return __io_sequence_defer(ctx, req);
+	return false;
 }
 
-static struct io_kiocb *io_get_deferred_req(struct io_ring_ctx *ctx)
+static void __io_commit_cqring(struct io_ring_ctx *ctx)
 {
-	struct io_kiocb *req;
+	struct io_rings *rings = ctx->rings;
 
-	req = list_first_entry_or_null(&ctx->defer_list, struct io_kiocb, list);
-	if (req && !io_sequence_defer(ctx, req)) {
-		list_del_init(&req->list);
-		return req;
-	}
+	/* order cqe stores with ring update */
+	smp_store_release(&rings->cq.tail, ctx->cached_cq_tail);
 
-	return NULL;
+	if (wq_has_sleeper(&ctx->cq_wait)) {
+		wake_up_interruptible(&ctx->cq_wait);
+		kill_fasync(&ctx->cq_fasync, SIGIO, POLL_IN);
+	}
 }
 
-static struct io_kiocb *io_get_timeout_req(struct io_ring_ctx *ctx)
+static inline void io_req_work_grab_env(struct io_kiocb *req,
+					const struct io_op_def *def)
 {
-	struct io_kiocb *req;
-
-	req = list_first_entry_or_null(&ctx->timeout_list, struct io_kiocb, list);
-	if (req) {
-		if (req->flags & REQ_F_TIMEOUT_NOSEQ)
-			return NULL;
-		if (!__io_sequence_defer(ctx, req)) {
-			list_del_init(&req->list);
-			return req;
+	if (!req->work.mm && def->needs_mm) {
+		mmgrab(current->mm);
+		req->work.mm = current->mm;
+	}
+	if (!req->work.creds)
+		req->work.creds = get_current_cred();
+	if (!req->work.fs && def->needs_fs) {
+		spin_lock(&current->fs->lock);
+		if (!current->fs->in_exec) {
+			req->work.fs = current->fs;
+			req->work.fs->users++;
+		} else {
+			req->work.flags |= IO_WQ_WORK_CANCEL;
 		}
+		spin_unlock(&current->fs->lock);
 	}
-
-	return NULL;
+	if (!req->work.task_pid)
+		req->work.task_pid = task_pid_vnr(current);
 }
 
-static void __io_commit_cqring(struct io_ring_ctx *ctx)
+static inline void io_req_work_drop_env(struct io_kiocb *req)
 {
-	struct io_rings *rings = ctx->rings;
+	if (req->work.mm) {
+		mmdrop(req->work.mm);
+		req->work.mm = NULL;
+	}
+	if (req->work.creds) {
+		put_cred(req->work.creds);
+		req->work.creds = NULL;
+	}
+	if (req->work.fs) {
+		struct fs_struct *fs = req->work.fs;
+
+		spin_lock(&req->work.fs->lock);
+		if (--fs->users)
+			fs = NULL;
+		spin_unlock(&req->work.fs->lock);
+		if (fs)
+			free_fs_struct(fs);
+	}
+}
 
-	if (ctx->cached_cq_tail != READ_ONCE(rings->cq.tail)) {
-		/* order cqe stores with ring update */
-		smp_store_release(&rings->cq.tail, ctx->cached_cq_tail);
+static inline void io_prep_async_work(struct io_kiocb *req,
+				      struct io_kiocb **link)
+{
+	const struct io_op_def *def = &io_op_defs[req->opcode];
 
-		if (wq_has_sleeper(&ctx->cq_wait)) {
-			wake_up_interruptible(&ctx->cq_wait);
-			kill_fasync(&ctx->cq_fasync, SIGIO, POLL_IN);
-		}
+	if (req->flags & REQ_F_ISREG) {
+		if (def->hash_reg_file)
+			io_wq_hash_work(&req->work, file_inode(req->file));
+	} else {
+		if (def->unbound_nonreg_file)
+			req->work.flags |= IO_WQ_WORK_UNBOUND;
 	}
+
+	io_req_work_grab_env(req, def);
+
+	*link = io_prep_linked_timeout(req);
 }
 
-static inline void io_queue_async_work(struct io_ring_ctx *ctx,
-				       struct io_kiocb *req)
+static inline void io_queue_async_work(struct io_kiocb *req)
 {
-	int rw = 0;
+	struct io_ring_ctx *ctx = req->ctx;
+	struct io_kiocb *link;
 
-	if (req->submit.sqe) {
-		switch (req->submit.sqe->opcode) {
-		case IORING_OP_WRITEV:
-		case IORING_OP_WRITE_FIXED:
-			rw = !(req->rw.ki_flags & IOCB_DIRECT);
-			break;
-		}
-	}
+	io_prep_async_work(req, &link);
+
+	trace_io_uring_queue_async_work(ctx, io_wq_is_hashed(&req->work), req,
+					&req->work, req->flags);
+	io_wq_enqueue(ctx->io_wq, &req->work);
 
-	queue_work(ctx->sqo_wq[rw], &req->work);
+	if (link)
+		io_queue_linked_timeout(link);
 }
 
 static void io_kill_timeout(struct io_kiocb *req)
 {
 	int ret;
 
-	ret = hrtimer_try_to_cancel(&req->timeout.timer);
+	ret = hrtimer_try_to_cancel(&req->io->timeout.timer);
 	if (ret != -1) {
 		atomic_inc(&req->ctx->cq_timeouts);
-		list_del(&req->list);
-		io_cqring_fill_event(req->ctx, req->user_data, 0);
-		__io_free_req(req);
+		list_del_init(&req->list);
+		req->flags |= REQ_F_COMP_LOCKED;
+		io_cqring_fill_event(req, 0);
+		io_put_req(req);
 	}
 }
 
@@ -521,24 +1107,43 @@ static void io_kill_timeouts(struct io_ring_ctx *ctx)
 	spin_unlock_irq(&ctx->completion_lock);
 }
 
-static void io_commit_cqring(struct io_ring_ctx *ctx)
+static void __io_queue_deferred(struct io_ring_ctx *ctx)
 {
-	struct io_kiocb *req;
+	do {
+		struct io_kiocb *req = list_first_entry(&ctx->defer_list,
+							struct io_kiocb, list);
+
+		if (req_need_defer(req))
+			break;
+		list_del_init(&req->list);
+		io_queue_async_work(req);
+	} while (!list_empty(&ctx->defer_list));
+}
+
+static void io_flush_timeouts(struct io_ring_ctx *ctx)
+{
+	while (!list_empty(&ctx->timeout_list)) {
+		struct io_kiocb *req = list_first_entry(&ctx->timeout_list,
+							struct io_kiocb, list);
 
-	while ((req = io_get_timeout_req(ctx)) != NULL)
+		if (req->flags & REQ_F_TIMEOUT_NOSEQ)
+			break;
+		if (req->timeout.target_seq != ctx->cached_cq_tail
+					- atomic_read(&ctx->cq_timeouts))
+			break;
+
+		list_del_init(&req->list);
 		io_kill_timeout(req);
+	}
+}
 
+static void io_commit_cqring(struct io_ring_ctx *ctx)
+{
+	io_flush_timeouts(ctx);
 	__io_commit_cqring(ctx);
 
-	while ((req = io_get_deferred_req(ctx)) != NULL) {
-		if (req->flags & REQ_F_SHADOW_DRAIN) {
-			/* Just for drain, free it. */
-			__io_free_req(req);
-			continue;
-		}
-		req->flags |= REQ_F_IO_DRAINED;
-		io_queue_async_work(ctx, req);
-	}
+	if (unlikely(!list_empty(&ctx->defer_list)))
+		__io_queue_deferred(ctx);
 }
 
 static struct io_uring_cqe *io_get_cqring(struct io_ring_ctx *ctx)
@@ -559,63 +1164,170 @@ static struct io_uring_cqe *io_get_cqring(struct io_ring_ctx *ctx)
 	return &rings->cqes[tail & ctx->cq_mask];
 }
 
-static void io_cqring_fill_event(struct io_ring_ctx *ctx, u64 ki_user_data,
-				 long res)
+static inline bool io_should_trigger_evfd(struct io_ring_ctx *ctx)
+{
+	if (!ctx->cq_ev_fd)
+		return false;
+	if (READ_ONCE(ctx->rings->cq_flags) & IORING_CQ_EVENTFD_DISABLED)
+		return false;
+	if (!ctx->eventfd_async)
+		return true;
+	return io_wq_current_is_worker();
+}
+
+static void io_cqring_ev_posted(struct io_ring_ctx *ctx)
+{
+	if (waitqueue_active(&ctx->wait))
+		wake_up(&ctx->wait);
+	if (waitqueue_active(&ctx->sqo_wait))
+		wake_up(&ctx->sqo_wait);
+	if (io_should_trigger_evfd(ctx))
+		eventfd_signal(ctx->cq_ev_fd, 1);
+}
+
+/* Returns true if there are no backlogged entries after the flush */
+static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force)
+{
+	struct io_rings *rings = ctx->rings;
+	struct io_uring_cqe *cqe;
+	struct io_kiocb *req;
+	unsigned long flags;
+	LIST_HEAD(list);
+
+	if (!force) {
+		if (list_empty_careful(&ctx->cq_overflow_list))
+			return true;
+		if ((ctx->cached_cq_tail - READ_ONCE(rings->cq.head) ==
+		    rings->cq_ring_entries))
+			return false;
+	}
+
+	spin_lock_irqsave(&ctx->completion_lock, flags);
+
+	/* if force is set, the ring is going away. always drop after that */
+	if (force)
+		ctx->cq_overflow_flushed = 1;
+
+	cqe = NULL;
+	while (!list_empty(&ctx->cq_overflow_list)) {
+		cqe = io_get_cqring(ctx);
+		if (!cqe && !force)
+			break;
+
+		req = list_first_entry(&ctx->cq_overflow_list, struct io_kiocb,
+						list);
+		list_move(&req->list, &list);
+		req->flags &= ~REQ_F_OVERFLOW;
+		if (cqe) {
+			WRITE_ONCE(cqe->user_data, req->user_data);
+			WRITE_ONCE(cqe->res, req->result);
+			WRITE_ONCE(cqe->flags, req->cflags);
+		} else {
+			WRITE_ONCE(ctx->rings->cq_overflow,
+				atomic_inc_return(&ctx->cached_cq_overflow));
+		}
+	}
+
+	io_commit_cqring(ctx);
+	if (cqe) {
+		clear_bit(0, &ctx->sq_check_overflow);
+		clear_bit(0, &ctx->cq_check_overflow);
+	}
+	spin_unlock_irqrestore(&ctx->completion_lock, flags);
+	io_cqring_ev_posted(ctx);
+
+	while (!list_empty(&list)) {
+		req = list_first_entry(&list, struct io_kiocb, list);
+		list_del(&req->list);
+		io_put_req(req);
+	}
+
+	return cqe != NULL;
+}
+
+static void __io_cqring_fill_event(struct io_kiocb *req, long res, long cflags)
 {
+	struct io_ring_ctx *ctx = req->ctx;
 	struct io_uring_cqe *cqe;
 
+	trace_io_uring_complete(ctx, req->user_data, res);
+
 	/*
 	 * If we can't get a cq entry, userspace overflowed the
 	 * submission (by quite a lot). Increment the overflow count in
 	 * the ring.
 	 */
 	cqe = io_get_cqring(ctx);
-	if (cqe) {
-		WRITE_ONCE(cqe->user_data, ki_user_data);
+	if (likely(cqe)) {
+		WRITE_ONCE(cqe->user_data, req->user_data);
 		WRITE_ONCE(cqe->res, res);
-		WRITE_ONCE(cqe->flags, 0);
-	} else {
+		WRITE_ONCE(cqe->flags, cflags);
+	} else if (ctx->cq_overflow_flushed) {
 		WRITE_ONCE(ctx->rings->cq_overflow,
 				atomic_inc_return(&ctx->cached_cq_overflow));
+	} else {
+		if (list_empty(&ctx->cq_overflow_list)) {
+			set_bit(0, &ctx->sq_check_overflow);
+			set_bit(0, &ctx->cq_check_overflow);
+		}
+		req->flags |= REQ_F_OVERFLOW;
+		refcount_inc(&req->refs);
+		req->result = res;
+		req->cflags = cflags;
+		list_add_tail(&req->list, &ctx->cq_overflow_list);
 	}
 }
 
-static void io_cqring_ev_posted(struct io_ring_ctx *ctx)
+static void io_cqring_fill_event(struct io_kiocb *req, long res)
 {
-	if (waitqueue_active(&ctx->wait))
-		wake_up(&ctx->wait);
-	if (waitqueue_active(&ctx->sqo_wait))
-		wake_up(&ctx->sqo_wait);
-	if (ctx->cq_ev_fd)
-		eventfd_signal(ctx->cq_ev_fd, 1);
+	__io_cqring_fill_event(req, res, 0);
 }
 
-static void io_cqring_add_event(struct io_ring_ctx *ctx, u64 user_data,
-				long res)
+static void __io_cqring_add_event(struct io_kiocb *req, long res, long cflags)
 {
+	struct io_ring_ctx *ctx = req->ctx;
 	unsigned long flags;
 
 	spin_lock_irqsave(&ctx->completion_lock, flags);
-	io_cqring_fill_event(ctx, user_data, res);
+	__io_cqring_fill_event(req, res, cflags);
 	io_commit_cqring(ctx);
 	spin_unlock_irqrestore(&ctx->completion_lock, flags);
 
 	io_cqring_ev_posted(ctx);
 }
 
-static struct io_kiocb *io_get_req(struct io_ring_ctx *ctx,
-				   struct io_submit_state *state)
+static void io_cqring_add_event(struct io_kiocb *req, long res)
+{
+	__io_cqring_add_event(req, res, 0);
+}
+
+static inline bool io_is_fallback_req(struct io_kiocb *req)
+{
+	return req == (struct io_kiocb *)
+			((unsigned long) req->ctx->fallback_req & ~1UL);
+}
+
+static struct io_kiocb *io_get_fallback_req(struct io_ring_ctx *ctx)
 {
-	gfp_t gfp = GFP_KERNEL | __GFP_NOWARN;
 	struct io_kiocb *req;
 
-	if (!percpu_ref_tryget(&ctx->refs))
-		return NULL;
+	req = ctx->fallback_req;
+	if (!test_and_set_bit_lock(0, (unsigned long *) &ctx->fallback_req))
+		return req;
+
+	return NULL;
+}
+
+static struct io_kiocb *io_alloc_req(struct io_ring_ctx *ctx,
+				     struct io_submit_state *state)
+{
+	gfp_t gfp = GFP_KERNEL | __GFP_NOWARN;
+	struct io_kiocb *req;
 
 	if (!state) {
 		req = kmem_cache_alloc(req_cachep, gfp);
 		if (unlikely(!req))
-			goto out;
+			goto fallback;
 	} else if (!state->free_reqs) {
 		size_t sz;
 		int ret;
@@ -630,103 +1342,275 @@ static struct io_kiocb *io_get_req(struct io_ring_ctx *ctx,
 		if (unlikely(ret <= 0)) {
 			state->reqs[0] = kmem_cache_alloc(req_cachep, gfp);
 			if (!state->reqs[0])
-				goto out;
+				goto fallback;
 			ret = 1;
 		}
 		state->free_reqs = ret - 1;
-		state->cur_req = 1;
-		req = state->reqs[0];
+		req = state->reqs[ret - 1];
 	} else {
-		req = state->reqs[state->cur_req];
 		state->free_reqs--;
-		state->cur_req++;
+		req = state->reqs[state->free_reqs];
 	}
 
-	req->file = NULL;
-	req->ctx = ctx;
-	req->flags = 0;
-	/* one is dropped after submission, the other at completion */
-	refcount_set(&req->refs, 2);
-	req->result = 0;
 	return req;
-out:
-	percpu_ref_put(&ctx->refs);
-	return NULL;
+fallback:
+	return io_get_fallback_req(ctx);
 }
 
-static void io_free_req_many(struct io_ring_ctx *ctx, void **reqs, int *nr)
+static inline void io_put_file(struct io_kiocb *req, struct file *file,
+			  bool fixed)
 {
-	if (*nr) {
-		kmem_cache_free_bulk(req_cachep, *nr, reqs);
-		percpu_ref_put_many(&ctx->refs, *nr);
-		*nr = 0;
-	}
+	if (fixed)
+		percpu_ref_put(req->fixed_file_refs);
+	else
+		fput(file);
+}
+
+static void __io_req_aux_free(struct io_kiocb *req)
+{
+	if (req->flags & REQ_F_NEED_CLEANUP)
+		io_cleanup_req(req);
+
+	kfree(req->io);
+	if (req->file)
+		io_put_file(req, req->file, (req->flags & REQ_F_FIXED_FILE));
+	if (req->task)
+		put_task_struct(req->task);
+
+	io_req_work_drop_env(req);
 }
 
 static void __io_free_req(struct io_kiocb *req)
 {
-	if (req->file && !(req->flags & REQ_F_FIXED_FILE))
-		fput(req->file);
+	__io_req_aux_free(req);
+
+	if (req->flags & REQ_F_INFLIGHT) {
+		struct io_ring_ctx *ctx = req->ctx;
+		unsigned long flags;
+
+		spin_lock_irqsave(&ctx->inflight_lock, flags);
+		list_del(&req->inflight_entry);
+		if (waitqueue_active(&ctx->inflight_wait))
+			wake_up(&ctx->inflight_wait);
+		spin_unlock_irqrestore(&ctx->inflight_lock, flags);
+	}
+
 	percpu_ref_put(&req->ctx->refs);
-	kmem_cache_free(req_cachep, req);
+	if (likely(!io_is_fallback_req(req)))
+		kmem_cache_free(req_cachep, req);
+	else
+		clear_bit_unlock(0, (unsigned long *) &req->ctx->fallback_req);
 }
 
-static void io_req_link_next(struct io_kiocb *req)
+struct req_batch {
+	void *reqs[IO_IOPOLL_BATCH];
+	int to_free;
+	int need_iter;
+};
+
+static void io_free_req_many(struct io_ring_ctx *ctx, struct req_batch *rb)
 {
-	struct io_kiocb *nxt;
+	if (!rb->to_free)
+		return;
+	if (rb->need_iter) {
+		int i, inflight = 0;
+		unsigned long flags;
+
+		for (i = 0; i < rb->to_free; i++) {
+			struct io_kiocb *req = rb->reqs[i];
+
+			if (req->flags & REQ_F_INFLIGHT)
+				inflight++;
+			__io_req_aux_free(req);
+		}
+		if (!inflight)
+			goto do_free;
+
+		spin_lock_irqsave(&ctx->inflight_lock, flags);
+		for (i = 0; i < rb->to_free; i++) {
+			struct io_kiocb *req = rb->reqs[i];
+
+			if (req->flags & REQ_F_INFLIGHT) {
+				list_del(&req->inflight_entry);
+				if (!--inflight)
+					break;
+			}
+		}
+		spin_unlock_irqrestore(&ctx->inflight_lock, flags);
+
+		if (waitqueue_active(&ctx->inflight_wait))
+			wake_up(&ctx->inflight_wait);
+	}
+do_free:
+	kmem_cache_free_bulk(req_cachep, rb->to_free, rb->reqs);
+	percpu_ref_put_many(&ctx->refs, rb->to_free);
+	rb->to_free = rb->need_iter = 0;
+}
+
+static bool io_link_cancel_timeout(struct io_kiocb *req)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	int ret;
+
+	ret = hrtimer_try_to_cancel(&req->io->timeout.timer);
+	if (ret != -1) {
+		io_cqring_fill_event(req, -ECANCELED);
+		io_commit_cqring(ctx);
+		req->flags &= ~REQ_F_LINK_HEAD;
+		io_put_req(req);
+		return true;
+	}
+
+	return false;
+}
+
+static void io_req_link_next(struct io_kiocb *req, struct io_kiocb **nxtptr)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	bool wake_ev = false;
+
+	/* Already got next link */
+	if (req->flags & REQ_F_LINK_NEXT)
+		return;
 
 	/*
 	 * The list should never be empty when we are called here. But could
 	 * potentially happen if the chain is messed up, check to be on the
 	 * safe side.
 	 */
-	nxt = list_first_entry_or_null(&req->link_list, struct io_kiocb, list);
-	if (nxt) {
-		list_del(&nxt->list);
-		if (!list_empty(&req->link_list)) {
-			INIT_LIST_HEAD(&nxt->link_list);
-			list_splice(&req->link_list, &nxt->link_list);
-			nxt->flags |= REQ_F_LINK;
+	while (!list_empty(&req->link_list)) {
+		struct io_kiocb *nxt = list_first_entry(&req->link_list,
+						struct io_kiocb, link_list);
+
+		if (unlikely((req->flags & REQ_F_LINK_TIMEOUT) &&
+			     (nxt->flags & REQ_F_TIMEOUT))) {
+			list_del_init(&nxt->link_list);
+			wake_ev |= io_link_cancel_timeout(nxt);
+			req->flags &= ~REQ_F_LINK_TIMEOUT;
+			continue;
 		}
 
-		nxt->flags |= REQ_F_LINK_DONE;
-		INIT_WORK(&nxt->work, io_sq_wq_submit_work);
-		io_queue_async_work(req->ctx, nxt);
+		list_del_init(&req->link_list);
+		if (!list_empty(&nxt->link_list))
+			nxt->flags |= REQ_F_LINK_HEAD;
+		*nxtptr = nxt;
+		break;
 	}
+
+	req->flags |= REQ_F_LINK_NEXT;
+	if (wake_ev)
+		io_cqring_ev_posted(ctx);
 }
 
 /*
- * Called if REQ_F_LINK is set, and we fail the head request
+ * Called if REQ_F_LINK_HEAD is set, and we fail the head request
  */
 static void io_fail_links(struct io_kiocb *req)
 {
-	struct io_kiocb *link;
+	struct io_ring_ctx *ctx = req->ctx;
+	unsigned long flags;
+
+	spin_lock_irqsave(&ctx->completion_lock, flags);
 
 	while (!list_empty(&req->link_list)) {
-		link = list_first_entry(&req->link_list, struct io_kiocb, list);
-		list_del(&link->list);
+		struct io_kiocb *link = list_first_entry(&req->link_list,
+						struct io_kiocb, link_list);
+
+		list_del_init(&link->link_list);
+		trace_io_uring_fail_link(req, link);
 
-		io_cqring_add_event(req->ctx, link->user_data, -ECANCELED);
-		__io_free_req(link);
+		if ((req->flags & REQ_F_LINK_TIMEOUT) &&
+		    link->opcode == IORING_OP_LINK_TIMEOUT) {
+			io_link_cancel_timeout(link);
+		} else {
+			io_cqring_fill_event(link, -ECANCELED);
+			__io_double_put_req(link);
+		}
+		req->flags &= ~REQ_F_LINK_TIMEOUT;
 	}
+
+	io_commit_cqring(ctx);
+	spin_unlock_irqrestore(&ctx->completion_lock, flags);
+	io_cqring_ev_posted(ctx);
 }
 
-static void io_free_req(struct io_kiocb *req)
+static void io_req_find_next(struct io_kiocb *req, struct io_kiocb **nxt)
 {
+	if (likely(!(req->flags & REQ_F_LINK_HEAD)))
+		return;
+
 	/*
 	 * If LINK is set, we have dependent requests in this chain. If we
 	 * didn't fail this request, queue the first one up, moving any other
 	 * dependencies to the next request. In case of failure, fail the rest
 	 * of the chain.
 	 */
-	if (req->flags & REQ_F_LINK) {
-		if (req->flags & REQ_F_FAIL_LINK)
-			io_fail_links(req);
-		else
-			io_req_link_next(req);
+	if (req->flags & REQ_F_FAIL_LINK) {
+		io_fail_links(req);
+	} else if ((req->flags & (REQ_F_LINK_TIMEOUT | REQ_F_COMP_LOCKED)) ==
+			REQ_F_LINK_TIMEOUT) {
+		struct io_ring_ctx *ctx = req->ctx;
+		unsigned long flags;
+
+		/*
+		 * If this is a timeout link, we could be racing with the
+		 * timeout timer. Grab the completion lock for this case to
+		 * protect against that.
+		 */
+		spin_lock_irqsave(&ctx->completion_lock, flags);
+		io_req_link_next(req, nxt);
+		spin_unlock_irqrestore(&ctx->completion_lock, flags);
+	} else {
+		io_req_link_next(req, nxt);
 	}
+}
+
+static void io_free_req(struct io_kiocb *req)
+{
+	struct io_kiocb *nxt = NULL;
 
+	io_req_find_next(req, &nxt);
 	__io_free_req(req);
+
+	if (nxt)
+		io_queue_async_work(nxt);
+}
+
+static void io_link_work_cb(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+	struct io_kiocb *link;
+
+	link = list_first_entry(&req->link_list, struct io_kiocb, link_list);
+	io_queue_linked_timeout(link);
+	io_wq_submit_work(workptr);
+}
+
+static void io_wq_assign_next(struct io_wq_work **workptr, struct io_kiocb *nxt)
+{
+	struct io_kiocb *link;
+	const struct io_op_def *def = &io_op_defs[nxt->opcode];
+
+	if ((nxt->flags & REQ_F_ISREG) && def->hash_reg_file)
+		io_wq_hash_work(&nxt->work, file_inode(nxt->file));
+
+	*workptr = &nxt->work;
+	link = io_prep_linked_timeout(nxt);
+	if (link)
+		nxt->work.func = io_link_work_cb;
+}
+
+/*
+ * Drop reference to request, return next in chain (if there is one) if this
+ * was the last reference to this request.
+ */
+__attribute__((nonnull))
+static void io_put_req_find_next(struct io_kiocb *req, struct io_kiocb **nxtptr)
+{
+	if (refcount_dec_and_test(&req->refs)) {
+		io_req_find_next(req, nxtptr);
+		__io_free_req(req);
+	}
 }
 
 static void io_put_req(struct io_kiocb *req)
@@ -735,11 +1619,63 @@ static void io_put_req(struct io_kiocb *req)
 		io_free_req(req);
 }
 
-static unsigned io_cqring_events(struct io_rings *rings)
+static void io_steal_work(struct io_kiocb *req,
+			  struct io_wq_work **workptr)
+{
+	/*
+	 * It's in an io-wq worker, so there always should be at least
+	 * one reference, which will be dropped in io_put_work() just
+	 * after the current handler returns.
+	 *
+	 * It also means, that if the counter dropped to 1, then there is
+	 * no asynchronous users left, so it's safe to steal the next work.
+	 */
+	if (refcount_read(&req->refs) == 1) {
+		struct io_kiocb *nxt = NULL;
+
+		io_req_find_next(req, &nxt);
+		if (nxt)
+			io_wq_assign_next(workptr, nxt);
+	}
+}
+
+/*
+ * Must only be used if we don't need to care about links, usually from
+ * within the completion handling itself.
+ */
+static void __io_double_put_req(struct io_kiocb *req)
+{
+	/* drop both submit and complete references */
+	if (refcount_sub_and_test(2, &req->refs))
+		__io_free_req(req);
+}
+
+static void io_double_put_req(struct io_kiocb *req)
+{
+	/* drop both submit and complete references */
+	if (refcount_sub_and_test(2, &req->refs))
+		io_free_req(req);
+}
+
+static unsigned io_cqring_events(struct io_ring_ctx *ctx, bool noflush)
 {
+	struct io_rings *rings = ctx->rings;
+
+	if (test_bit(0, &ctx->cq_check_overflow)) {
+		/*
+		 * noflush == true is from the waitqueue handler, just ensure
+		 * we wake up the task, and the next invocation will flush the
+		 * entries. We cannot safely to it from here.
+		 */
+		if (noflush && !list_empty(&ctx->cq_overflow_list))
+			return -1U;
+
+		io_cqring_overflow_flush(ctx, false);
+	}
+
 	/* See comment at the top of this file */
 	smp_rmb();
-	return READ_ONCE(rings->cq.tail) - READ_ONCE(rings->cq.head);
+	return ctx->cached_cq_tail - READ_ONCE(rings->cq.head);
 }
 
 static inline unsigned int io_sqring_entries(struct io_ring_ctx *ctx)
@@ -750,43 +1686,76 @@ static inline unsigned int io_sqring_entries(struct io_ring_ctx *ctx)
 	return smp_load_acquire(&rings->sq.tail) - ctx->cached_sq_head;
 }
 
+static inline bool io_req_multi_free(struct req_batch *rb, struct io_kiocb *req)
+{
+	if ((req->flags & REQ_F_LINK_HEAD) || io_is_fallback_req(req))
+		return false;
+
+	if (req->file || req->io)
+		rb->need_iter++;
+
+	rb->reqs[rb->to_free++] = req;
+	if (unlikely(rb->to_free == ARRAY_SIZE(rb->reqs)))
+		io_free_req_many(req->ctx, rb);
+	return true;
+}
+
+static int io_put_kbuf(struct io_kiocb *req)
+{
+	struct io_buffer *kbuf;
+	int cflags;
+
+	kbuf = (struct io_buffer *) (unsigned long) req->rw.addr;
+	cflags = kbuf->bid << IORING_CQE_BUFFER_SHIFT;
+	cflags |= IORING_CQE_F_BUFFER;
+	req->rw.addr = 0;
+	kfree(kbuf);
+	return cflags;
+}
+
 /*
  * Find and free completed poll iocbs
  */
 static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events,
 			       struct list_head *done)
 {
-	void *reqs[IO_IOPOLL_BATCH];
+	struct req_batch rb;
 	struct io_kiocb *req;
-	int to_free;
 
-	to_free = 0;
+	rb.to_free = rb.need_iter = 0;
 	while (!list_empty(done)) {
+		int cflags = 0;
+
 		req = list_first_entry(done, struct io_kiocb, list);
 		list_del(&req->list);
 
-		io_cqring_fill_event(ctx, req->user_data, req->result);
+		if (req->flags & REQ_F_BUFFER_SELECTED)
+			cflags = io_put_kbuf(req);
+
+		__io_cqring_fill_event(req, req->result, cflags);
 		(*nr_events)++;
 
-		if (refcount_dec_and_test(&req->refs)) {
-			/* If we're not using fixed files, we have to pair the
-			 * completion part with the file put. Use regular
-			 * completions for those, only batch free for fixed
-			 * file and non-linked commands.
-			 */
-			if ((req->flags & (REQ_F_FIXED_FILE|REQ_F_LINK)) ==
-			    REQ_F_FIXED_FILE) {
-				reqs[to_free++] = req;
-				if (to_free == ARRAY_SIZE(reqs))
-					io_free_req_many(ctx, reqs, &to_free);
-			} else {
-				io_free_req(req);
-			}
-		}
+		if (refcount_dec_and_test(&req->refs) &&
+		    !io_req_multi_free(&rb, req))
+			io_free_req(req);
 	}
 
 	io_commit_cqring(ctx);
-	io_free_req_many(ctx, reqs, &to_free);
+	if (ctx->flags & IORING_SETUP_SQPOLL)
+		io_cqring_ev_posted(ctx);
+	io_free_req_many(ctx, &rb);
+}
+
+static void io_iopoll_queue(struct list_head *again)
+{
+	struct io_kiocb *req;
+
+	do {
+		req = list_first_entry(again, struct io_kiocb, list);
+		list_del(&req->list);
+		refcount_inc(&req->refs);
+		io_queue_async_work(req);
+	} while (!list_empty(again));
 }
 
 static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events,
@@ -794,6 +1763,7 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events,
 {
 	struct io_kiocb *req, *tmp;
 	LIST_HEAD(done);
+	LIST_HEAD(again);
 	bool spin;
 	int ret;
 
@@ -805,12 +1775,12 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events,
 
 	ret = 0;
 	list_for_each_entry_safe(req, tmp, &ctx->poll_list, list) {
-		struct kiocb *kiocb = &req->rw;
+		struct kiocb *kiocb = &req->rw.kiocb;
 
 		/*
-		 * Move completed entries to our local list. If we find a
-		 * request that requires polling, break out and complete
-		 * the done list first, if we have entries there.
+		 * Move completed and retryable entries to our local lists.
+		 * If we find a request that requires polling, break out
+		 * and complete those lists first, if we have entries there.
 		 */
 		if (req->flags & REQ_F_IOPOLL_COMPLETED) {
 			list_move_tail(&req->list, &done);
@@ -819,6 +1789,13 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events,
 		if (!list_empty(&done))
 			break;
 
+		if (req->result == -EAGAIN) {
+			list_move_tail(&req->list, &again);
+			continue;
+		}
+		if (!list_empty(&again))
+			break;
+
 		ret = kiocb->ki_filp->f_op->iopoll(kiocb, spin);
 		if (ret < 0)
 			break;
@@ -831,11 +1808,14 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events,
 	if (!list_empty(&done))
 		io_iopoll_complete(ctx, nr_events, &done);
 
+	if (!list_empty(&again))
+		io_iopoll_queue(&again);
+
 	return ret;
 }
 
 /*
- * Poll for a mininum of 'min' events. Note that if min == 0 we consider that a
+ * Poll for a minimum of 'min' events. Note that if min == 0 we consider that a
  * non-spinning poll check - we'll still enter the driver poll loop, but only
  * as a non-spinning completion check.
  */
@@ -879,11 +1859,17 @@ static void io_iopoll_reap_events(struct io_ring_ctx *ctx)
 	mutex_unlock(&ctx->uring_lock);
 }
 
-static int __io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events,
-			    long min)
+static int io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events,
+			   long min)
 {
 	int iters = 0, ret = 0;
 
+	/*
+	 * We disallow the app entering submit/complete with polling, but we
+	 * still need to lock the ring to prevent racing with polled issue
+	 * that got punted to a workqueue.
+	 */
+	mutex_lock(&ctx->uring_lock);
 	do {
 		int tmin = 0;
 
@@ -892,7 +1878,7 @@ static int __io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events,
 		 * If we do, we can potentially be spinning for commands that
 		 * already triggered a CQE (eg in error).
 		 */
-		if (io_cqring_events(ctx->rings))
+		if (io_cqring_events(ctx, false))
 			break;
 
 		/*
@@ -919,21 +1905,6 @@ static int __io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events,
 		ret = 0;
 	} while (min && !*nr_events && !need_resched());
 
-	return ret;
-}
-
-static int io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events,
-			   long min)
-{
-	int ret;
-
-	/*
-	 * We disallow the app entering submit/complete with polling, but we
-	 * still need to lock the ring to prevent racing with polled issue
-	 * that got punted to a workqueue.
-	 */
-	mutex_lock(&ctx->uring_lock);
-	ret = __io_iopoll_check(ctx, nr_events, min);
 	mutex_unlock(&ctx->uring_lock);
 	return ret;
 }
@@ -952,28 +1923,44 @@ static void kiocb_end_write(struct io_kiocb *req)
 	file_end_write(req->file);
 }
 
-static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
+static inline void req_set_fail_links(struct io_kiocb *req)
 {
-	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw);
+	if ((req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) == REQ_F_LINK)
+		req->flags |= REQ_F_FAIL_LINK;
+}
+
+static void io_complete_rw_common(struct kiocb *kiocb, long res)
+{
+	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
+	int cflags = 0;
 
 	if (kiocb->ki_flags & IOCB_WRITE)
 		kiocb_end_write(req);
 
-	if ((req->flags & REQ_F_LINK) && res != req->result)
-		req->flags |= REQ_F_FAIL_LINK;
-	io_cqring_add_event(req->ctx, req->user_data, res);
+	if (res != req->result)
+		req_set_fail_links(req);
+	if (req->flags & REQ_F_BUFFER_SELECTED)
+		cflags = io_put_kbuf(req);
+	__io_cqring_add_event(req, res, cflags);
+}
+
+static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
+{
+	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
+
+	io_complete_rw_common(kiocb, res);
 	io_put_req(req);
 }
 
 static void io_complete_rw_iopoll(struct kiocb *kiocb, long res, long res2)
 {
-	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw);
+	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 
 	if (kiocb->ki_flags & IOCB_WRITE)
 		kiocb_end_write(req);
 
-	if ((req->flags & REQ_F_LINK) && res != req->result)
-		req->flags |= REQ_F_FAIL_LINK;
+	if (res != req->result)
+		req_set_fail_links(req);
 	req->result = res;
 	if (res != -EAGAIN)
 		req->flags |= REQ_F_IOPOLL_COMPLETED;
@@ -1001,7 +1988,7 @@ static void io_iopoll_req_issued(struct io_kiocb *req)
 
 		list_req = list_first_entry(&ctx->poll_list, struct io_kiocb,
 						list);
-		if (list_req->rw.ki_filp != req->rw.ki_filp)
+		if (list_req->file != req->file)
 			ctx->poll_multi_file = true;
 	}
 
@@ -1013,17 +2000,25 @@ static void io_iopoll_req_issued(struct io_kiocb *req)
 		list_add(&req->list, &ctx->poll_list);
 	else
 		list_add_tail(&req->list, &ctx->poll_list);
+
+	if ((ctx->flags & IORING_SETUP_SQPOLL) &&
+	    wq_has_sleeper(&ctx->sqo_wait))
+		wake_up(&ctx->sqo_wait);
 }
 
-static void io_file_put(struct io_submit_state *state)
+static void __io_state_file_put(struct io_submit_state *state)
 {
-	if (state->file) {
-		int diff = state->has_refs - state->used_refs;
+	int diff = state->has_refs - state->used_refs;
 
-		if (diff)
-			fput_many(state->file, diff);
-		state->file = NULL;
-	}
+	if (diff)
+		fput_many(state->file, diff);
+	state->file = NULL;
+}
+
+static inline void io_state_file_put(struct io_submit_state *state)
+{
+	if (state->file)
+		__io_state_file_put(state);
 }
 
 /*
@@ -1031,7 +2026,7 @@ static void io_file_put(struct io_submit_state *state)
  * assuming most submissions are for one file, or at least that each file
  * has more than one submission.
  */
-static struct file *io_file_get(struct io_submit_state *state, int fd)
+static struct file *__io_file_get(struct io_submit_state *state, int fd)
 {
 	if (!state)
 		return fget(fd);
@@ -1042,7 +2037,7 @@ static struct file *io_file_get(struct io_submit_state *state, int fd)
 			state->ios_left--;
 			return state->file;
 		}
-		io_file_put(state);
+		__io_state_file_put(state);
 	}
 	state->file = fget_many(fd, state->ios_left);
 	if (!state->file)
@@ -1060,45 +2055,45 @@ static struct file *io_file_get(struct io_submit_state *state, int fd)
  * any file. For now, just ensure that anything potentially problematic is done
  * inline.
  */
-static bool io_file_supports_async(struct file *file)
+static bool io_file_supports_async(struct file *file, int rw)
 {
 	umode_t mode = file_inode(file)->i_mode;
 
-	if (S_ISBLK(mode) || S_ISCHR(mode))
+	if (S_ISBLK(mode) || S_ISCHR(mode) || S_ISSOCK(mode))
 		return true;
 	if (S_ISREG(mode) && file->f_op != &io_uring_fops)
 		return true;
 
-	return false;
+	if (!(file->f_mode & FMODE_NOWAIT))
+		return false;
+
+	if (rw == READ)
+		return file->f_op->read_iter != NULL;
+
+	return file->f_op->write_iter != NULL;
 }
 
-static int io_prep_rw(struct io_kiocb *req, const struct sqe_submit *s,
+static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 		      bool force_nonblock)
 {
-	const struct io_uring_sqe *sqe = s->sqe;
 	struct io_ring_ctx *ctx = req->ctx;
-	struct kiocb *kiocb = &req->rw;
+	struct kiocb *kiocb = &req->rw.kiocb;
 	unsigned ioprio;
 	int ret;
 
-	if (!req->file)
-		return -EBADF;
-
 	if (S_ISREG(file_inode(req->file)->i_mode))
 		req->flags |= REQ_F_ISREG;
 
-	/*
-	 * If the file doesn't support async, mark it as REQ_F_MUST_PUNT so
-	 * we know to async punt it even if it was opened O_NONBLOCK
-	 */
-	if (force_nonblock && !io_file_supports_async(req->file)) {
-		req->flags |= REQ_F_MUST_PUNT;
-		return -EAGAIN;
-	}
-
 	kiocb->ki_pos = READ_ONCE(sqe->off);
-	kiocb->ki_flags = iocb_flags(kiocb->ki_filp);
+	if (kiocb->ki_pos == -1 && !(req->file->f_mode & FMODE_STREAM)) {
+		req->flags |= REQ_F_CUR_POS;
+		kiocb->ki_pos = req->file->f_pos;
+	}
 	kiocb->ki_hint = ki_hint_validate(file_write_hint(kiocb->ki_filp));
+	kiocb->ki_flags = iocb_flags(kiocb->ki_filp);
+	ret = kiocb_set_rw_flags(kiocb, READ_ONCE(sqe->rw_flags));
+	if (unlikely(ret))
+		return ret;
 
 	ioprio = READ_ONCE(sqe->ioprio);
 	if (ioprio) {
@@ -1110,10 +2105,6 @@ static int io_prep_rw(struct io_kiocb *req, const struct sqe_submit *s,
 	} else
 		kiocb->ki_ioprio = get_current_ioprio();
 
-	ret = kiocb_set_rw_flags(kiocb, READ_ONCE(sqe->rw_flags));
-	if (unlikely(ret))
-		return ret;
-
 	/* don't allow async punt if RWF_NOWAIT was requested */
 	if ((kiocb->ki_flags & IOCB_NOWAIT) ||
 	    (req->file->f_flags & O_NONBLOCK))
@@ -1135,6 +2126,10 @@ static int io_prep_rw(struct io_kiocb *req, const struct sqe_submit *s,
 			return -EINVAL;
 		kiocb->ki_complete = io_complete_rw;
 	}
+
+	req->rw.addr = READ_ONCE(sqe->addr);
+	req->rw.len = READ_ONCE(sqe->len);
+	req->buf_index = READ_ONCE(sqe->buf_index);
 	return 0;
 }
 
@@ -1159,13 +2154,25 @@ static inline void io_rw_done(struct kiocb *kiocb, ssize_t ret)
 	}
 }
 
-static int io_import_fixed(struct io_ring_ctx *ctx, int rw,
-			   const struct io_uring_sqe *sqe,
-			   struct iov_iter *iter)
+static void kiocb_done(struct kiocb *kiocb, ssize_t ret)
 {
-	size_t len = READ_ONCE(sqe->len);
+	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
+
+	if (req->flags & REQ_F_CUR_POS)
+		req->file->f_pos = kiocb->ki_pos;
+	if (ret >= 0 && kiocb->ki_complete == io_complete_rw)
+		io_complete_rw(kiocb, ret, 0);
+	else
+		io_rw_done(kiocb, ret);
+}
+
+static ssize_t io_import_fixed(struct io_kiocb *req, int rw,
+			       struct iov_iter *iter)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	size_t len = req->rw.len;
 	struct io_mapped_ubuf *imu;
-	unsigned index, buf_index;
+	u16 index, buf_index;
 	size_t offset;
 	u64 buf_addr;
 
@@ -1173,13 +2180,13 @@ static int io_import_fixed(struct io_ring_ctx *ctx, int rw,
 	if (unlikely(!ctx->user_bufs))
 		return -EFAULT;
 
-	buf_index = READ_ONCE(sqe->buf_index);
+	buf_index = req->buf_index;
 	if (unlikely(buf_index >= ctx->nr_user_bufs))
 		return -EFAULT;
 
 	index = array_index_nospec(buf_index, ctx->nr_user_bufs);
 	imu = &ctx->user_bufs[index];
-	buf_addr = READ_ONCE(sqe->addr);
+	buf_addr = req->rw.addr;
 
 	/* overflow */
 	if (buf_addr + len < buf_addr)
@@ -1233,100 +2240,201 @@ static int io_import_fixed(struct io_ring_ctx *ctx, int rw,
 	return len;
 }
 
-static ssize_t io_import_iovec(struct io_ring_ctx *ctx, int rw,
-			       const struct sqe_submit *s, struct iovec **iovec,
-			       struct iov_iter *iter)
+static void io_ring_submit_unlock(struct io_ring_ctx *ctx, bool needs_lock)
 {
-	const struct io_uring_sqe *sqe = s->sqe;
-	void __user *buf = u64_to_user_ptr(READ_ONCE(sqe->addr));
-	size_t sqe_len = READ_ONCE(sqe->len);
-	u8 opcode;
+	if (needs_lock)
+		mutex_unlock(&ctx->uring_lock);
+}
 
+static void io_ring_submit_lock(struct io_ring_ctx *ctx, bool needs_lock)
+{
 	/*
-	 * We're reading ->opcode for the second time, but the first read
-	 * doesn't care whether it's _FIXED or not, so it doesn't matter
-	 * whether ->opcode changes concurrently. The first read does care
-	 * about whether it is a READ or a WRITE, so we don't trust this read
-	 * for that purpose and instead let the caller pass in the read/write
-	 * flag.
+	 * "Normal" inline submissions always hold the uring_lock, since we
+	 * grab it from the system call. Same is true for the SQPOLL offload.
+	 * The only exception is when we've detached the request and issue it
+	 * from an async worker thread, grab the lock for that case.
 	 */
-	opcode = READ_ONCE(sqe->opcode);
-	if (opcode == IORING_OP_READ_FIXED ||
-	    opcode == IORING_OP_WRITE_FIXED) {
-		ssize_t ret = io_import_fixed(ctx, rw, sqe, iter);
-		*iovec = NULL;
-		return ret;
+	if (needs_lock)
+		mutex_lock(&ctx->uring_lock);
+}
+
+static struct io_buffer *io_buffer_select(struct io_kiocb *req, size_t *len,
+					  int bgid, struct io_buffer *kbuf,
+					  bool needs_lock)
+{
+	struct io_buffer *head;
+
+	if (req->flags & REQ_F_BUFFER_SELECTED)
+		return kbuf;
+
+	io_ring_submit_lock(req->ctx, needs_lock);
+
+	lockdep_assert_held(&req->ctx->uring_lock);
+
+	head = idr_find(&req->ctx->io_buffer_idr, bgid);
+	if (head) {
+		if (!list_empty(&head->list)) {
+			kbuf = list_last_entry(&head->list, struct io_buffer,
+							list);
+			list_del(&kbuf->list);
+		} else {
+			kbuf = head;
+			idr_remove(&req->ctx->io_buffer_idr, bgid);
+		}
+		if (*len > kbuf->len)
+			*len = kbuf->len;
+	} else {
+		kbuf = ERR_PTR(-ENOBUFS);
 	}
 
-	if (!s->has_user)
-		return -EFAULT;
+	io_ring_submit_unlock(req->ctx, needs_lock);
+
+	return kbuf;
+}
+
+static void __user *io_rw_buffer_select(struct io_kiocb *req, size_t *len,
+					bool needs_lock)
+{
+	struct io_buffer *kbuf;
+	u16 bgid;
+
+	kbuf = (struct io_buffer *) (unsigned long) req->rw.addr;
+	bgid = req->buf_index;
+	kbuf = io_buffer_select(req, len, bgid, kbuf, needs_lock);
+	if (IS_ERR(kbuf))
+		return kbuf;
+	req->rw.addr = (u64) (unsigned long) kbuf;
+	req->flags |= REQ_F_BUFFER_SELECTED;
+	return u64_to_user_ptr(kbuf->addr);
+}
 
 #ifdef CONFIG_COMPAT
-	if (ctx->compat)
-		return compat_import_iovec(rw, buf, sqe_len, UIO_FASTIOV,
-						iovec, iter);
+static ssize_t io_compat_import(struct io_kiocb *req, struct iovec *iov,
+				bool needs_lock)
+{
+	struct compat_iovec __user *uiov;
+	compat_ssize_t clen;
+	void __user *buf;
+	ssize_t len;
+
+	uiov = u64_to_user_ptr(req->rw.addr);
+	if (!access_ok(uiov, sizeof(*uiov)))
+		return -EFAULT;
+	if (__get_user(clen, &uiov->iov_len))
+		return -EFAULT;
+	if (clen < 0)
+		return -EINVAL;
+
+	len = clen;
+	buf = io_rw_buffer_select(req, &len, needs_lock);
+	if (IS_ERR(buf))
+		return PTR_ERR(buf);
+	iov[0].iov_base = buf;
+	iov[0].iov_len = (compat_size_t) len;
+	return 0;
+}
 #endif
 
-	return import_iovec(rw, buf, sqe_len, UIO_FASTIOV, iovec, iter);
+static ssize_t __io_iov_buffer_select(struct io_kiocb *req, struct iovec *iov,
+				      bool needs_lock)
+{
+	struct iovec __user *uiov = u64_to_user_ptr(req->rw.addr);
+	void __user *buf;
+	ssize_t len;
+
+	if (copy_from_user(iov, uiov, sizeof(*uiov)))
+		return -EFAULT;
+
+	len = iov[0].iov_len;
+	if (len < 0)
+		return -EINVAL;
+	buf = io_rw_buffer_select(req, &len, needs_lock);
+	if (IS_ERR(buf))
+		return PTR_ERR(buf);
+	iov[0].iov_base = buf;
+	iov[0].iov_len = len;
+	return 0;
 }
 
-static inline bool io_should_merge(struct async_list *al, struct kiocb *kiocb)
+static ssize_t io_iov_buffer_select(struct io_kiocb *req, struct iovec *iov,
+				    bool needs_lock)
 {
-	if (al->file == kiocb->ki_filp) {
-		off_t start, end;
+	if (req->flags & REQ_F_BUFFER_SELECTED)
+		return 0;
+	if (!req->rw.len)
+		return 0;
+	else if (req->rw.len > 1)
+		return -EINVAL;
 
-		/*
-		 * Allow merging if we're anywhere in the range of the same
-		 * page. Generally this happens for sub-page reads or writes,
-		 * and it's beneficial to allow the first worker to bring the
-		 * page in and the piggy backed work can then work on the
-		 * cached page.
-		 */
-		start = al->io_start & PAGE_MASK;
-		end = (al->io_start + al->io_len + PAGE_SIZE - 1) & PAGE_MASK;
-		if (kiocb->ki_pos >= start && kiocb->ki_pos <= end)
-			return true;
-	}
+#ifdef CONFIG_COMPAT
+	if (req->ctx->compat)
+		return io_compat_import(req, iov, needs_lock);
+#endif
 
-	al->file = NULL;
-	return false;
+	return __io_iov_buffer_select(req, iov, needs_lock);
 }
 
-/*
- * Make a note of the last file/offset/direction we punted to async
- * context. We'll use this information to see if we can piggy back a
- * sequential request onto the previous one, if it's still hasn't been
- * completed by the async worker.
- */
-static void io_async_list_note(int rw, struct io_kiocb *req, size_t len)
+static ssize_t io_import_iovec(int rw, struct io_kiocb *req,
+			       struct iovec **iovec, struct iov_iter *iter,
+			       bool needs_lock)
 {
-	struct async_list *async_list = &req->ctx->pending_async[rw];
-	struct kiocb *kiocb = &req->rw;
-	struct file *filp = kiocb->ki_filp;
+	void __user *buf = u64_to_user_ptr(req->rw.addr);
+	size_t sqe_len = req->rw.len;
+	ssize_t ret;
+	u8 opcode;
 
-	if (io_should_merge(async_list, kiocb)) {
-		unsigned long max_bytes;
+	opcode = req->opcode;
+	if (opcode == IORING_OP_READ_FIXED || opcode == IORING_OP_WRITE_FIXED) {
+		*iovec = NULL;
+		return io_import_fixed(req, rw, iter);
+	}
 
-		/* Use 8x RA size as a decent limiter for both reads/writes */
-		max_bytes = filp->f_ra.ra_pages << (PAGE_SHIFT + 3);
-		if (!max_bytes)
-			max_bytes = VM_READAHEAD_PAGES << (PAGE_SHIFT + 3);
+	/* buffer index only valid with fixed read/write, or buffer select  */
+	if (req->buf_index && !(req->flags & REQ_F_BUFFER_SELECT))
+		return -EINVAL;
 
-		/* If max len are exceeded, reset the state */
-		if (async_list->io_len + len <= max_bytes) {
-			req->flags |= REQ_F_SEQ_PREV;
-			async_list->io_len += len;
-		} else {
-			async_list->file = NULL;
+	if (opcode == IORING_OP_READ || opcode == IORING_OP_WRITE) {
+		if (req->flags & REQ_F_BUFFER_SELECT) {
+			buf = io_rw_buffer_select(req, &sqe_len, needs_lock);
+			if (IS_ERR(buf)) {
+				*iovec = NULL;
+				return PTR_ERR(buf);
+			}
+			req->rw.len = sqe_len;
 		}
+
+		ret = import_single_range(rw, buf, sqe_len, *iovec, iter);
+		*iovec = NULL;
+		return ret < 0 ? ret : sqe_len;
 	}
 
-	/* New file? Reset state. */
-	if (async_list->file != filp) {
-		async_list->io_start = kiocb->ki_pos;
-		async_list->io_len = len;
-		async_list->file = filp;
+	if (req->io) {
+		struct io_async_rw *iorw = &req->io->rw;
+
+		*iovec = iorw->iov;
+		iov_iter_init(iter, rw, *iovec, iorw->nr_segs, iorw->size);
+		if (iorw->iov == iorw->fast_iov)
+			*iovec = NULL;
+		return iorw->size;
+	}
+
+	if (req->flags & REQ_F_BUFFER_SELECT) {
+		ret = io_iov_buffer_select(req, *iovec, needs_lock);
+		if (!ret) {
+			ret = (*iovec)->iov_len;
+			iov_iter_init(iter, rw, *iovec, 1, ret);
+		}
+		*iovec = NULL;
+		return ret;
 	}
+
+#ifdef CONFIG_COMPAT
+	if (req->ctx->compat)
+		return compat_import_iovec(rw, buf, sqe_len, UIO_FASTIOV,
+						iovec, iter);
+#endif
+
+	return import_iovec(rw, buf, sqe_len, UIO_FASTIOV, iovec, iter);
 }
 
 /*
@@ -1349,9 +2457,19 @@ static ssize_t loop_rw_iter(int rw, struct file *file, struct kiocb *kiocb,
 		return -EAGAIN;
 
 	while (iov_iter_count(iter)) {
-		struct iovec iovec = iov_iter_iovec(iter);
+		struct iovec iovec;
 		ssize_t nr;
 
+		if (!iov_iter_is_bvec(iter)) {
+			iovec = iov_iter_iovec(iter);
+		} else {
+			/* fixed buffers import bvec */
+			iovec.iov_base = kmap(iter->bvec->bv_page)
+						+ iter->iov_offset;
+			iovec.iov_len = min(iter->count,
+					iter->bvec->bv_len - iter->iov_offset);
+		}
+
 		if (rw == READ) {
 			nr = file->f_op->read(file, iovec.iov_base,
 					      iovec.iov_len, &kiocb->ki_pos);
@@ -1360,6 +2478,9 @@ static ssize_t loop_rw_iter(int rw, struct file *file, struct kiocb *kiocb,
 					       iovec.iov_len, &kiocb->ki_pos);
 		}
 
+		if (iov_iter_is_bvec(iter))
+			kunmap(iter->bvec->bv_page);
+
 		if (nr < 0) {
 			if (!ret)
 				ret = nr;
@@ -1374,107 +2495,209 @@ static ssize_t loop_rw_iter(int rw, struct file *file, struct kiocb *kiocb,
 	return ret;
 }
 
-static int io_read(struct io_kiocb *req, const struct sqe_submit *s,
-		   bool force_nonblock)
+static void io_req_map_rw(struct io_kiocb *req, ssize_t io_size,
+			  struct iovec *iovec, struct iovec *fast_iov,
+			  struct iov_iter *iter)
 {
-	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
-	struct kiocb *kiocb = &req->rw;
+	req->io->rw.nr_segs = iter->nr_segs;
+	req->io->rw.size = io_size;
+	req->io->rw.iov = iovec;
+	if (!req->io->rw.iov) {
+		req->io->rw.iov = req->io->rw.fast_iov;
+		if (req->io->rw.iov != fast_iov)
+			memcpy(req->io->rw.iov, fast_iov,
+			       sizeof(struct iovec) * iter->nr_segs);
+	} else {
+		req->flags |= REQ_F_NEED_CLEANUP;
+	}
+}
+
+static inline int __io_alloc_async_ctx(struct io_kiocb *req)
+{
+	req->io = kmalloc(sizeof(*req->io), GFP_KERNEL);
+	return req->io == NULL;
+}
+
+static int io_alloc_async_ctx(struct io_kiocb *req)
+{
+	if (!io_op_defs[req->opcode].async_ctx)
+		return 0;
+
+	return  __io_alloc_async_ctx(req);
+}
+
+static int io_setup_async_rw(struct io_kiocb *req, ssize_t io_size,
+			     struct iovec *iovec, struct iovec *fast_iov,
+			     struct iov_iter *iter)
+{
+	if (!io_op_defs[req->opcode].async_ctx)
+		return 0;
+	if (!req->io) {
+		if (__io_alloc_async_ctx(req))
+			return -ENOMEM;
+
+		io_req_map_rw(req, io_size, iovec, fast_iov, iter);
+	}
+	return 0;
+}
+
+static int io_read_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			bool force_nonblock)
+{
+	struct io_async_ctx *io;
 	struct iov_iter iter;
-	struct file *file;
-	size_t iov_count;
-	ssize_t read_size, ret;
+	ssize_t ret;
 
-	ret = io_prep_rw(req, s, force_nonblock);
+	ret = io_prep_rw(req, sqe, force_nonblock);
 	if (ret)
 		return ret;
-	file = kiocb->ki_filp;
 
-	if (unlikely(!(file->f_mode & FMODE_READ)))
+	if (unlikely(!(req->file->f_mode & FMODE_READ)))
 		return -EBADF;
 
-	ret = io_import_iovec(req->ctx, READ, s, &iovec, &iter);
+	/* either don't need iovec imported or already have it */
+	if (!req->io || req->flags & REQ_F_NEED_CLEANUP)
+		return 0;
+
+	io = req->io;
+	io->rw.iov = io->rw.fast_iov;
+	req->io = NULL;
+	ret = io_import_iovec(READ, req, &io->rw.iov, &iter, !force_nonblock);
+	req->io = io;
+	if (ret < 0)
+		return ret;
+
+	io_req_map_rw(req, ret, io->rw.iov, io->rw.fast_iov, &iter);
+	return 0;
+}
+
+static int io_read(struct io_kiocb *req, bool force_nonblock)
+{
+	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
+	struct kiocb *kiocb = &req->rw.kiocb;
+	struct iov_iter iter;
+	size_t iov_count;
+	ssize_t io_size, ret;
+
+	ret = io_import_iovec(READ, req, &iovec, &iter, !force_nonblock);
 	if (ret < 0)
 		return ret;
 
-	read_size = ret;
-	if (req->flags & REQ_F_LINK)
-		req->result = read_size;
+	/* Ensure we clear previously set non-block flag */
+	if (!force_nonblock)
+		kiocb->ki_flags &= ~IOCB_NOWAIT;
+
+	req->result = 0;
+	io_size = ret;
+	if (req->flags & REQ_F_LINK_HEAD)
+		req->result = io_size;
+
+	/*
+	 * If the file doesn't support async, mark it as REQ_F_MUST_PUNT so
+	 * we know to async punt it even if it was opened O_NONBLOCK
+	 */
+	if (force_nonblock && !io_file_supports_async(req->file, READ))
+		goto copy_iov;
 
 	iov_count = iov_iter_count(&iter);
-	ret = rw_verify_area(READ, file, &kiocb->ki_pos, iov_count);
+	ret = rw_verify_area(READ, req->file, &kiocb->ki_pos, iov_count);
 	if (!ret) {
 		ssize_t ret2;
 
-		if (file->f_op->read_iter)
-			ret2 = call_read_iter(file, kiocb, &iter);
+		if (req->file->f_op->read_iter)
+			ret2 = call_read_iter(req->file, kiocb, &iter);
 		else
-			ret2 = loop_rw_iter(READ, file, kiocb, &iter);
+			ret2 = loop_rw_iter(READ, req->file, kiocb, &iter);
 
-		/*
-		 * In case of a short read, punt to async. This can happen
-		 * if we have data partially cached. Alternatively we can
-		 * return the short read, in which case the application will
-		 * need to issue another SQE and wait for it. That SQE will
-		 * need async punt anyway, so it's more efficient to do it
-		 * here.
-		 */
-		if (force_nonblock && !(req->flags & REQ_F_NOWAIT) &&
-		    (req->flags & REQ_F_ISREG) &&
-		    ret2 > 0 && ret2 < read_size)
-			ret2 = -EAGAIN;
 		/* Catch -EAGAIN return for forced non-blocking submission */
 		if (!force_nonblock || ret2 != -EAGAIN) {
-			io_rw_done(kiocb, ret2);
+			kiocb_done(kiocb, ret2);
 		} else {
-			/*
-			 * If ->needs_lock is true, we're already in async
-			 * context.
-			 */
-			if (!s->needs_lock)
-				io_async_list_note(READ, req, iov_count);
-			ret = -EAGAIN;
+copy_iov:
+			ret = io_setup_async_rw(req, io_size, iovec,
+						inline_vecs, &iter);
+			if (ret)
+				goto out_free;
+			/* any defer here is final, must blocking retry */
+			if (!(req->flags & REQ_F_NOWAIT) &&
+			    !file_can_poll(req->file))
+				req->flags |= REQ_F_MUST_PUNT;
+			return -EAGAIN;
 		}
 	}
+out_free:
 	kfree(iovec);
+	req->flags &= ~REQ_F_NEED_CLEANUP;
 	return ret;
 }
 
-static int io_write(struct io_kiocb *req, const struct sqe_submit *s,
-		    bool force_nonblock)
+static int io_write_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			 bool force_nonblock)
 {
-	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
-	struct kiocb *kiocb = &req->rw;
+	struct io_async_ctx *io;
 	struct iov_iter iter;
-	struct file *file;
-	size_t iov_count;
 	ssize_t ret;
 
-	ret = io_prep_rw(req, s, force_nonblock);
+	ret = io_prep_rw(req, sqe, force_nonblock);
 	if (ret)
 		return ret;
 
-	file = kiocb->ki_filp;
-	if (unlikely(!(file->f_mode & FMODE_WRITE)))
+	if (unlikely(!(req->file->f_mode & FMODE_WRITE)))
 		return -EBADF;
 
-	ret = io_import_iovec(req->ctx, WRITE, s, &iovec, &iter);
+	req->fsize = rlimit(RLIMIT_FSIZE);
+
+	/* either don't need iovec imported or already have it */
+	if (!req->io || req->flags & REQ_F_NEED_CLEANUP)
+		return 0;
+
+	io = req->io;
+	io->rw.iov = io->rw.fast_iov;
+	req->io = NULL;
+	ret = io_import_iovec(WRITE, req, &io->rw.iov, &iter, !force_nonblock);
+	req->io = io;
 	if (ret < 0)
 		return ret;
 
-	if (req->flags & REQ_F_LINK)
-		req->result = ret;
+	io_req_map_rw(req, ret, io->rw.iov, io->rw.fast_iov, &iter);
+	return 0;
+}
 
-	iov_count = iov_iter_count(&iter);
+static int io_write(struct io_kiocb *req, bool force_nonblock)
+{
+	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
+	struct kiocb *kiocb = &req->rw.kiocb;
+	struct iov_iter iter;
+	size_t iov_count;
+	ssize_t ret, io_size;
 
-	ret = -EAGAIN;
-	if (force_nonblock && !(kiocb->ki_flags & IOCB_DIRECT)) {
-		/* If ->needs_lock is true, we're already in async context. */
-		if (!s->needs_lock)
-			io_async_list_note(WRITE, req, iov_count);
-		goto out_free;
-	}
+	ret = io_import_iovec(WRITE, req, &iovec, &iter, !force_nonblock);
+	if (ret < 0)
+		return ret;
+
+	/* Ensure we clear previously set non-block flag */
+	if (!force_nonblock)
+		req->rw.kiocb.ki_flags &= ~IOCB_NOWAIT;
+
+	req->result = 0;
+	io_size = ret;
+	if (req->flags & REQ_F_LINK_HEAD)
+		req->result = io_size;
+
+	/*
+	 * If the file doesn't support async, mark it as REQ_F_MUST_PUNT so
+	 * we know to async punt it even if it was opened O_NONBLOCK
+	 */
+	if (force_nonblock && !io_file_supports_async(req->file, WRITE))
+		goto copy_iov;
 
-	ret = rw_verify_area(WRITE, file, &kiocb->ki_pos, iov_count);
+	/* file path doesn't support NOWAIT for non-direct_IO */
+	if (force_nonblock && !(kiocb->ki_flags & IOCB_DIRECT) &&
+	    (req->flags & REQ_F_ISREG))
+		goto copy_iov;
+
+	iov_count = iov_iter_count(&iter);
+	ret = rw_verify_area(WRITE, req->file, &kiocb->ki_pos, iov_count);
 	if (!ret) {
 		ssize_t ret2;
 
@@ -1486,46 +2709,158 @@ static int io_write(struct io_kiocb *req, const struct sqe_submit *s,
 		 * we return to userspace.
 		 */
 		if (req->flags & REQ_F_ISREG) {
-			__sb_start_write(file_inode(file)->i_sb,
+			__sb_start_write(file_inode(req->file)->i_sb,
 						SB_FREEZE_WRITE, true);
-			__sb_writers_release(file_inode(file)->i_sb,
+			__sb_writers_release(file_inode(req->file)->i_sb,
 						SB_FREEZE_WRITE);
 		}
 		kiocb->ki_flags |= IOCB_WRITE;
 
-		if (file->f_op->write_iter)
-			ret2 = call_write_iter(file, kiocb, &iter);
+		if (!force_nonblock)
+			current->signal->rlim[RLIMIT_FSIZE].rlim_cur = req->fsize;
+
+		if (req->file->f_op->write_iter)
+			ret2 = call_write_iter(req->file, kiocb, &iter);
 		else
-			ret2 = loop_rw_iter(WRITE, file, kiocb, &iter);
+			ret2 = loop_rw_iter(WRITE, req->file, kiocb, &iter);
+
+		if (!force_nonblock)
+			current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY;
+
+		/*
+		 * Raw bdev writes will return -EOPNOTSUPP for IOCB_NOWAIT. Just
+		 * retry them without IOCB_NOWAIT.
+		 */
+		if (ret2 == -EOPNOTSUPP && (kiocb->ki_flags & IOCB_NOWAIT))
+			ret2 = -EAGAIN;
 		if (!force_nonblock || ret2 != -EAGAIN) {
-			io_rw_done(kiocb, ret2);
+			kiocb_done(kiocb, ret2);
 		} else {
-			/*
-			 * If ->needs_lock is true, we're already in async
-			 * context.
-			 */
-			if (!s->needs_lock)
-				io_async_list_note(WRITE, req, iov_count);
-			ret = -EAGAIN;
+copy_iov:
+			ret = io_setup_async_rw(req, io_size, iovec,
+						inline_vecs, &iter);
+			if (ret)
+				goto out_free;
+			/* any defer here is final, must blocking retry */
+			if (!file_can_poll(req->file))
+				req->flags |= REQ_F_MUST_PUNT;
+			return -EAGAIN;
 		}
 	}
 out_free:
+	req->flags &= ~REQ_F_NEED_CLEANUP;
 	kfree(iovec);
 	return ret;
 }
 
+static int __io_splice_prep(struct io_kiocb *req,
+			    const struct io_uring_sqe *sqe)
+{
+	struct io_splice* sp = &req->splice;
+	unsigned int valid_flags = SPLICE_F_FD_IN_FIXED | SPLICE_F_ALL;
+	int ret;
+
+	if (req->flags & REQ_F_NEED_CLEANUP)
+		return 0;
+
+	sp->file_in = NULL;
+	sp->len = READ_ONCE(sqe->len);
+	sp->flags = READ_ONCE(sqe->splice_flags);
+
+	if (unlikely(sp->flags & ~valid_flags))
+		return -EINVAL;
+
+	ret = io_file_get(NULL, req, READ_ONCE(sqe->splice_fd_in), &sp->file_in,
+			  (sp->flags & SPLICE_F_FD_IN_FIXED));
+	if (ret)
+		return ret;
+	req->flags |= REQ_F_NEED_CLEANUP;
+
+	if (!S_ISREG(file_inode(sp->file_in)->i_mode))
+		req->work.flags |= IO_WQ_WORK_UNBOUND;
+
+	return 0;
+}
+
+static int io_tee_prep(struct io_kiocb *req,
+		       const struct io_uring_sqe *sqe)
+{
+	if (READ_ONCE(sqe->splice_off_in) || READ_ONCE(sqe->off))
+		return -EINVAL;
+	return __io_splice_prep(req, sqe);
+}
+
+static int io_tee(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_splice *sp = &req->splice;
+	struct file *in = sp->file_in;
+	struct file *out = sp->file_out;
+	unsigned int flags = sp->flags & ~SPLICE_F_FD_IN_FIXED;
+	long ret = 0;
+
+	if (force_nonblock)
+		return -EAGAIN;
+	if (sp->len)
+		ret = do_tee(in, out, sp->len, flags);
+
+	io_put_file(req, in, (sp->flags & SPLICE_F_FD_IN_FIXED));
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+
+	io_cqring_add_event(req, ret);
+	if (ret != sp->len)
+		req_set_fail_links(req);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_splice_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_splice* sp = &req->splice;
+
+	sp->off_in = READ_ONCE(sqe->splice_off_in);
+	sp->off_out = READ_ONCE(sqe->off);
+	return __io_splice_prep(req, sqe);
+}
+
+static int io_splice(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_splice *sp = &req->splice;
+	struct file *in = sp->file_in;
+	struct file *out = sp->file_out;
+	unsigned int flags = sp->flags & ~SPLICE_F_FD_IN_FIXED;
+	loff_t *poff_in, *poff_out;
+	long ret = 0;
+
+	if (force_nonblock)
+		return -EAGAIN;
+
+	poff_in = (sp->off_in == -1) ? NULL : &sp->off_in;
+	poff_out = (sp->off_out == -1) ? NULL : &sp->off_out;
+
+	if (sp->len)
+		ret = do_splice(in, poff_in, out, poff_out, sp->len, flags);
+
+	io_put_file(req, in, (sp->flags & SPLICE_F_FD_IN_FIXED));
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+
+	io_cqring_add_event(req, ret);
+	if (ret != sp->len)
+		req_set_fail_links(req);
+	io_put_req(req);
+	return 0;
+}
+
 /*
  * IORING_OP_NOP just posts a completion event, nothing else.
  */
-static int io_nop(struct io_kiocb *req, u64 user_data)
+static int io_nop(struct io_kiocb *req)
 {
 	struct io_ring_ctx *ctx = req->ctx;
-	long err = 0;
 
 	if (unlikely(ctx->flags & IORING_SETUP_IOPOLL))
 		return -EINVAL;
 
-	io_cqring_add_event(ctx, user_data, err);
+	io_cqring_add_event(req, 0);
 	io_put_req(req);
 	return 0;
 }
@@ -1542,46 +2877,604 @@ static int io_prep_fsync(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index))
 		return -EINVAL;
 
+	req->sync.flags = READ_ONCE(sqe->fsync_flags);
+	if (unlikely(req->sync.flags & ~IORING_FSYNC_DATASYNC))
+		return -EINVAL;
+
+	req->sync.off = READ_ONCE(sqe->off);
+	req->sync.len = READ_ONCE(sqe->len);
+	return 0;
+}
+
+static bool io_req_cancelled(struct io_kiocb *req)
+{
+	if (req->work.flags & IO_WQ_WORK_CANCEL) {
+		req_set_fail_links(req);
+		io_cqring_add_event(req, -ECANCELED);
+		io_put_req(req);
+		return true;
+	}
+
+	return false;
+}
+
+static void __io_fsync(struct io_kiocb *req)
+{
+	loff_t end = req->sync.off + req->sync.len;
+	int ret;
+
+	ret = vfs_fsync_range(req->file, req->sync.off,
+				end > 0 ? end : LLONG_MAX,
+				req->sync.flags & IORING_FSYNC_DATASYNC);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+}
+
+static void io_fsync_finish(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+
+	if (io_req_cancelled(req))
+		return;
+	__io_fsync(req);
+	io_steal_work(req, workptr);
+}
+
+static int io_fsync(struct io_kiocb *req, bool force_nonblock)
+{
+	/* fsync always requires a blocking context */
+	if (force_nonblock) {
+		req->work.func = io_fsync_finish;
+		return -EAGAIN;
+	}
+	__io_fsync(req);
+	return 0;
+}
+
+static void __io_fallocate(struct io_kiocb *req)
+{
+	int ret;
+
+	current->signal->rlim[RLIMIT_FSIZE].rlim_cur = req->fsize;
+	ret = vfs_fallocate(req->file, req->sync.mode, req->sync.off,
+				req->sync.len);
+	current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY;
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+}
+
+static void io_fallocate_finish(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+
+	if (io_req_cancelled(req))
+		return;
+	__io_fallocate(req);
+	io_steal_work(req, workptr);
+}
+
+static int io_fallocate_prep(struct io_kiocb *req,
+			     const struct io_uring_sqe *sqe)
+{
+	if (sqe->ioprio || sqe->buf_index || sqe->rw_flags)
+		return -EINVAL;
+
+	req->sync.off = READ_ONCE(sqe->off);
+	req->sync.len = READ_ONCE(sqe->addr);
+	req->sync.mode = READ_ONCE(sqe->len);
+	req->fsize = rlimit(RLIMIT_FSIZE);
+	return 0;
+}
+
+static int io_fallocate(struct io_kiocb *req, bool force_nonblock)
+{
+	/* fallocate always requiring blocking context */
+	if (force_nonblock) {
+		req->work.func = io_fallocate_finish;
+		return -EAGAIN;
+	}
+
+	__io_fallocate(req);
+	return 0;
+}
+
+static int io_openat_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	const char __user *fname;
+	int ret;
+
+	if (sqe->ioprio || sqe->buf_index)
+		return -EINVAL;
+	if (req->flags & REQ_F_FIXED_FILE)
+		return -EBADF;
+	if (req->flags & REQ_F_NEED_CLEANUP)
+		return 0;
+
+	req->open.dfd = READ_ONCE(sqe->fd);
+	req->open.how.mode = READ_ONCE(sqe->len);
+	fname = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	req->open.how.flags = READ_ONCE(sqe->open_flags);
+	if (force_o_largefile())
+		req->open.how.flags |= O_LARGEFILE;
+
+	req->open.filename = getname(fname);
+	if (IS_ERR(req->open.filename)) {
+		ret = PTR_ERR(req->open.filename);
+		req->open.filename = NULL;
+		return ret;
+	}
+
+	req->open.nofile = rlimit(RLIMIT_NOFILE);
+	req->flags |= REQ_F_NEED_CLEANUP;
 	return 0;
 }
 
-static int io_fsync(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		    bool force_nonblock)
+static int io_openat2_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
-	loff_t sqe_off = READ_ONCE(sqe->off);
-	loff_t sqe_len = READ_ONCE(sqe->len);
-	loff_t end = sqe_off + sqe_len;
-	unsigned fsync_flags;
+	struct open_how __user *how;
+	const char __user *fname;
+	size_t len;
 	int ret;
 
-	fsync_flags = READ_ONCE(sqe->fsync_flags);
-	if (unlikely(fsync_flags & ~IORING_FSYNC_DATASYNC))
+	if (sqe->ioprio || sqe->buf_index)
 		return -EINVAL;
+	if (req->flags & REQ_F_FIXED_FILE)
+		return -EBADF;
+	if (req->flags & REQ_F_NEED_CLEANUP)
+		return 0;
+
+	req->open.dfd = READ_ONCE(sqe->fd);
+	fname = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	how = u64_to_user_ptr(READ_ONCE(sqe->addr2));
+	len = READ_ONCE(sqe->len);
 
-	ret = io_prep_fsync(req, sqe);
+	if (len < OPEN_HOW_SIZE_VER0)
+		return -EINVAL;
+
+	ret = copy_struct_from_user(&req->open.how, sizeof(req->open.how), how,
+					len);
 	if (ret)
 		return ret;
 
-	/* fsync always requires a blocking context */
+	if (!(req->open.how.flags & O_PATH) && force_o_largefile())
+		req->open.how.flags |= O_LARGEFILE;
+
+	req->open.filename = getname(fname);
+	if (IS_ERR(req->open.filename)) {
+		ret = PTR_ERR(req->open.filename);
+		req->open.filename = NULL;
+		return ret;
+	}
+
+	req->open.nofile = rlimit(RLIMIT_NOFILE);
+	req->flags |= REQ_F_NEED_CLEANUP;
+	return 0;
+}
+
+static int io_openat2(struct io_kiocb *req, bool force_nonblock)
+{
+	struct open_flags op;
+	struct file *file;
+	int ret;
+
 	if (force_nonblock)
 		return -EAGAIN;
 
-	ret = vfs_fsync_range(req->rw.ki_filp, sqe_off,
-				end > 0 ? end : LLONG_MAX,
-				fsync_flags & IORING_FSYNC_DATASYNC);
+	ret = build_open_flags(&req->open.how, &op);
+	if (ret)
+		goto err;
 
-	if (ret < 0 && (req->flags & REQ_F_LINK))
-		req->flags |= REQ_F_FAIL_LINK;
-	io_cqring_add_event(req->ctx, sqe->user_data, ret);
+	ret = __get_unused_fd_flags(req->open.how.flags, req->open.nofile);
+	if (ret < 0)
+		goto err;
+
+	file = do_filp_open(req->open.dfd, req->open.filename, &op);
+	if (IS_ERR(file)) {
+		put_unused_fd(ret);
+		ret = PTR_ERR(file);
+	} else {
+		fsnotify_open(file);
+		fd_install(ret, file);
+	}
+err:
+	putname(req->open.filename);
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
 	io_put_req(req);
 	return 0;
 }
 
-static int io_prep_sfr(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+static int io_openat(struct io_kiocb *req, bool force_nonblock)
 {
+	req->open.how = build_open_how(req->open.how.flags, req->open.how.mode);
+	return io_openat2(req, force_nonblock);
+}
+
+static int io_remove_buffers_prep(struct io_kiocb *req,
+				  const struct io_uring_sqe *sqe)
+{
+	struct io_provide_buf *p = &req->pbuf;
+	u64 tmp;
+
+	if (sqe->ioprio || sqe->rw_flags || sqe->addr || sqe->len || sqe->off)
+		return -EINVAL;
+
+	tmp = READ_ONCE(sqe->fd);
+	if (!tmp || tmp > USHRT_MAX)
+		return -EINVAL;
+
+	memset(p, 0, sizeof(*p));
+	p->nbufs = tmp;
+	p->bgid = READ_ONCE(sqe->buf_group);
+	return 0;
+}
+
+static int __io_remove_buffers(struct io_ring_ctx *ctx, struct io_buffer *buf,
+			       int bgid, unsigned nbufs)
+{
+	unsigned i = 0;
+
+	/* shouldn't happen */
+	if (!nbufs)
+		return 0;
+
+	/* the head kbuf is the list itself */
+	while (!list_empty(&buf->list)) {
+		struct io_buffer *nxt;
+
+		nxt = list_first_entry(&buf->list, struct io_buffer, list);
+		list_del(&nxt->list);
+		kfree(nxt);
+		if (++i == nbufs)
+			return i;
+	}
+	i++;
+	kfree(buf);
+	idr_remove(&ctx->io_buffer_idr, bgid);
+
+	return i;
+}
+
+static int io_remove_buffers(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_provide_buf *p = &req->pbuf;
+	struct io_ring_ctx *ctx = req->ctx;
+	struct io_buffer *head;
+	int ret = 0;
+
+	io_ring_submit_lock(ctx, !force_nonblock);
+
+	lockdep_assert_held(&ctx->uring_lock);
+
+	ret = -ENOENT;
+	head = idr_find(&ctx->io_buffer_idr, p->bgid);
+	if (head)
+		ret = __io_remove_buffers(ctx, head, p->bgid, p->nbufs);
+
+	io_ring_submit_lock(ctx, !force_nonblock);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_provide_buffers_prep(struct io_kiocb *req,
+				   const struct io_uring_sqe *sqe)
+{
+	struct io_provide_buf *p = &req->pbuf;
+	u64 tmp;
+
+	if (sqe->ioprio || sqe->rw_flags)
+		return -EINVAL;
+
+	tmp = READ_ONCE(sqe->fd);
+	if (!tmp || tmp > USHRT_MAX)
+		return -E2BIG;
+	p->nbufs = tmp;
+	p->addr = READ_ONCE(sqe->addr);
+	p->len = READ_ONCE(sqe->len);
+
+	if (!access_ok(u64_to_user_ptr(p->addr), p->len))
+		return -EFAULT;
+
+	p->bgid = READ_ONCE(sqe->buf_group);
+	tmp = READ_ONCE(sqe->off);
+	if (tmp > USHRT_MAX)
+		return -E2BIG;
+	p->bid = tmp;
+	return 0;
+}
+
+static int io_add_buffers(struct io_provide_buf *pbuf, struct io_buffer **head)
+{
+	struct io_buffer *buf;
+	u64 addr = pbuf->addr;
+	int i, bid = pbuf->bid;
+
+	for (i = 0; i < pbuf->nbufs; i++) {
+		buf = kmalloc(sizeof(*buf), GFP_KERNEL);
+		if (!buf)
+			break;
+
+		buf->addr = addr;
+		buf->len = pbuf->len;
+		buf->bid = bid;
+		addr += pbuf->len;
+		bid++;
+		if (!*head) {
+			INIT_LIST_HEAD(&buf->list);
+			*head = buf;
+		} else {
+			list_add_tail(&buf->list, &(*head)->list);
+		}
+	}
+
+	return i ? i : -ENOMEM;
+}
+
+static int io_provide_buffers(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_provide_buf *p = &req->pbuf;
 	struct io_ring_ctx *ctx = req->ctx;
+	struct io_buffer *head, *list;
 	int ret = 0;
 
+	io_ring_submit_lock(ctx, !force_nonblock);
+
+	lockdep_assert_held(&ctx->uring_lock);
+
+	list = head = idr_find(&ctx->io_buffer_idr, p->bgid);
+
+	ret = io_add_buffers(p, &head);
+	if (ret < 0)
+		goto out;
+
+	if (!list) {
+		ret = idr_alloc(&ctx->io_buffer_idr, head, p->bgid, p->bgid + 1,
+					GFP_KERNEL);
+		if (ret < 0) {
+			__io_remove_buffers(ctx, head, p->bgid, -1U);
+			goto out;
+		}
+	}
+out:
+	io_ring_submit_unlock(ctx, !force_nonblock);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_epoll_ctl_prep(struct io_kiocb *req,
+			     const struct io_uring_sqe *sqe)
+{
+#if defined(CONFIG_EPOLL)
+	if (sqe->ioprio || sqe->buf_index)
+		return -EINVAL;
+
+	req->epoll.epfd = READ_ONCE(sqe->fd);
+	req->epoll.op = READ_ONCE(sqe->len);
+	req->epoll.fd = READ_ONCE(sqe->off);
+
+	if (ep_op_has_event(req->epoll.op)) {
+		struct epoll_event __user *ev;
+
+		ev = u64_to_user_ptr(READ_ONCE(sqe->addr));
+		if (copy_from_user(&req->epoll.event, ev, sizeof(*ev)))
+			return -EFAULT;
+	}
+
+	return 0;
+#else
+	return -EOPNOTSUPP;
+#endif
+}
+
+static int io_epoll_ctl(struct io_kiocb *req, bool force_nonblock)
+{
+#if defined(CONFIG_EPOLL)
+	struct io_epoll *ie = &req->epoll;
+	int ret;
+
+	ret = do_epoll_ctl(ie->epfd, ie->op, ie->fd, &ie->event, force_nonblock);
+	if (force_nonblock && ret == -EAGAIN)
+		return -EAGAIN;
+
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+	return 0;
+#else
+	return -EOPNOTSUPP;
+#endif
+}
+
+static int io_madvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+#if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU)
+	if (sqe->ioprio || sqe->buf_index || sqe->off)
+		return -EINVAL;
+
+	req->madvise.addr = READ_ONCE(sqe->addr);
+	req->madvise.len = READ_ONCE(sqe->len);
+	req->madvise.advice = READ_ONCE(sqe->fadvise_advice);
+	return 0;
+#else
+	return -EOPNOTSUPP;
+#endif
+}
+
+static int io_madvise(struct io_kiocb *req, bool force_nonblock)
+{
+#if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU)
+	struct io_madvise *ma = &req->madvise;
+	int ret;
+
+	if (force_nonblock)
+		return -EAGAIN;
+
+	ret = do_madvise(ma->addr, ma->len, ma->advice);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+	return 0;
+#else
+	return -EOPNOTSUPP;
+#endif
+}
+
+static int io_fadvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	if (sqe->ioprio || sqe->buf_index || sqe->addr)
+		return -EINVAL;
+
+	req->fadvise.offset = READ_ONCE(sqe->off);
+	req->fadvise.len = READ_ONCE(sqe->len);
+	req->fadvise.advice = READ_ONCE(sqe->fadvise_advice);
+	return 0;
+}
+
+static int io_fadvise(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_fadvise *fa = &req->fadvise;
+	int ret;
+
+	if (force_nonblock) {
+		switch (fa->advice) {
+		case POSIX_FADV_NORMAL:
+		case POSIX_FADV_RANDOM:
+		case POSIX_FADV_SEQUENTIAL:
+			break;
+		default:
+			return -EAGAIN;
+		}
+	}
+
+	ret = vfs_fadvise(req->file, fa->offset, fa->len, fa->advice);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_statx_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	if (sqe->ioprio || sqe->buf_index)
+		return -EINVAL;
+	if (req->flags & REQ_F_FIXED_FILE)
+		return -EBADF;
+
+	req->statx.dfd = READ_ONCE(sqe->fd);
+	req->statx.mask = READ_ONCE(sqe->len);
+	req->statx.filename = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	req->statx.buffer = u64_to_user_ptr(READ_ONCE(sqe->addr2));
+	req->statx.flags = READ_ONCE(sqe->statx_flags);
+
+	return 0;
+}
+
+static int io_statx(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_statx *ctx = &req->statx;
+	int ret;
+
+	if (force_nonblock) {
+		/* only need file table for an actual valid fd */
+		if (ctx->dfd == -1 || ctx->dfd == AT_FDCWD)
+			req->flags |= REQ_F_NO_FILE_TABLE;
+		return -EAGAIN;
+	}
+
+	ret = do_statx(ctx->dfd, ctx->filename, ctx->flags, ctx->mask,
+		       ctx->buffer);
+
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_close_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	/*
+	 * If we queue this for async, it must not be cancellable. That would
+	 * leave the 'file' in an undeterminate state.
+	 */
+	req->work.flags |= IO_WQ_WORK_NO_CANCEL;
+
+	if (sqe->ioprio || sqe->off || sqe->addr || sqe->len ||
+	    sqe->rw_flags || sqe->buf_index)
+		return -EINVAL;
+	if (req->flags & REQ_F_FIXED_FILE)
+		return -EBADF;
+
+	req->close.fd = READ_ONCE(sqe->fd);
+	return 0;
+}
+
+/* only called when __close_fd_get_file() is done */
+static void __io_close_finish(struct io_kiocb *req)
+{
+	int ret;
+
+	ret = filp_close(req->close.put_file, req->work.files);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	fput(req->close.put_file);
+	io_put_req(req);
+}
+
+static void io_close_finish(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+
+	/* not cancellable, don't do io_req_cancelled() */
+	__io_close_finish(req);
+	io_steal_work(req, workptr);
+}
+
+static int io_close(struct io_kiocb *req, bool force_nonblock)
+{
+	int ret;
+
+	req->close.put_file = NULL;
+	ret = __close_fd_get_file(req->close.fd, &req->close.put_file);
+	if (ret < 0)
+		return (ret == -ENOENT) ? -EBADF : ret;
+
+	/* if the file has a flush method, be safe and punt to async */
+	if (req->close.put_file->f_op->flush && force_nonblock) {
+		/* avoid grabbing files - we don't need the files */
+		req->flags |= REQ_F_NO_FILE_TABLE | REQ_F_MUST_PUNT;
+		req->work.func = io_close_finish;
+		return -EAGAIN;
+	}
+
+	/*
+	 * No ->flush(), safely close from here and just punt the
+	 * fput() to async context.
+	 */
+	__io_close_finish(req);
+	return 0;
+}
+
+static int io_prep_sfr(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+
 	if (!req->file)
 		return -EBADF;
 
@@ -1590,44 +3483,150 @@ static int io_prep_sfr(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index))
 		return -EINVAL;
 
-	return ret;
+	req->sync.off = READ_ONCE(sqe->off);
+	req->sync.len = READ_ONCE(sqe->len);
+	req->sync.flags = READ_ONCE(sqe->sync_range_flags);
+	return 0;
 }
 
-static int io_sync_file_range(struct io_kiocb *req,
-			      const struct io_uring_sqe *sqe,
-			      bool force_nonblock)
+static void __io_sync_file_range(struct io_kiocb *req)
 {
-	loff_t sqe_off;
-	loff_t sqe_len;
-	unsigned flags;
 	int ret;
 
-	ret = io_prep_sfr(req, sqe);
-	if (ret)
-		return ret;
+	ret = sync_file_range(req->file, req->sync.off, req->sync.len,
+				req->sync.flags);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+}
+
 
+static void io_sync_file_range_finish(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+
+	if (io_req_cancelled(req))
+		return;
+	__io_sync_file_range(req);
+	io_steal_work(req, workptr);
+}
+
+static int io_sync_file_range(struct io_kiocb *req, bool force_nonblock)
+{
 	/* sync_file_range always requires a blocking context */
-	if (force_nonblock)
+	if (force_nonblock) {
+		req->work.func = io_sync_file_range_finish;
+		return -EAGAIN;
+	}
+
+	__io_sync_file_range(req);
+	return 0;
+}
+
+#if defined(CONFIG_NET)
+static int io_setup_async_msg(struct io_kiocb *req,
+			      struct io_async_msghdr *kmsg)
+{
+	if (req->io)
 		return -EAGAIN;
+	if (io_alloc_async_ctx(req)) {
+		if (kmsg->iov != kmsg->fast_iov)
+			kfree(kmsg->iov);
+		return -ENOMEM;
+	}
+	req->flags |= REQ_F_NEED_CLEANUP;
+	memcpy(&req->io->msg, kmsg, sizeof(*kmsg));
+	return -EAGAIN;
+}
 
-	sqe_off = READ_ONCE(sqe->off);
-	sqe_len = READ_ONCE(sqe->len);
-	flags = READ_ONCE(sqe->sync_range_flags);
+static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_sr_msg *sr = &req->sr_msg;
+	struct io_async_ctx *io = req->io;
+	int ret;
 
-	ret = sync_file_range(req->rw.ki_filp, sqe_off, sqe_len, flags);
+	sr->msg_flags = READ_ONCE(sqe->msg_flags);
+	sr->msg = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	sr->len = READ_ONCE(sqe->len);
 
-	if (ret < 0 && (req->flags & REQ_F_LINK))
-		req->flags |= REQ_F_FAIL_LINK;
-	io_cqring_add_event(req->ctx, sqe->user_data, ret);
+#ifdef CONFIG_COMPAT
+	if (req->ctx->compat)
+		sr->msg_flags |= MSG_CMSG_COMPAT;
+#endif
+
+	if (!io || req->opcode == IORING_OP_SEND)
+		return 0;
+	/* iovec is already imported */
+	if (req->flags & REQ_F_NEED_CLEANUP)
+		return 0;
+
+	io->msg.iov = io->msg.fast_iov;
+	ret = sendmsg_copy_msghdr(&io->msg.msg, sr->msg, sr->msg_flags,
+					&io->msg.iov);
+	if (!ret)
+		req->flags |= REQ_F_NEED_CLEANUP;
+	return ret;
+}
+
+static int io_sendmsg(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_async_msghdr *kmsg = NULL;
+	struct socket *sock;
+	int ret;
+
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+
+	sock = sock_from_file(req->file, &ret);
+	if (sock) {
+		struct io_async_ctx io;
+		unsigned flags;
+
+		if (req->io) {
+			kmsg = &req->io->msg;
+			kmsg->msg.msg_name = &req->io->msg.addr;
+			/* if iov is set, it's allocated already */
+			if (!kmsg->iov)
+				kmsg->iov = kmsg->fast_iov;
+			kmsg->msg.msg_iter.iov = kmsg->iov;
+		} else {
+			struct io_sr_msg *sr = &req->sr_msg;
+
+			kmsg = &io.msg;
+			kmsg->msg.msg_name = &io.msg.addr;
+
+			io.msg.iov = io.msg.fast_iov;
+			ret = sendmsg_copy_msghdr(&io.msg.msg, sr->msg,
+					sr->msg_flags, &io.msg.iov);
+			if (ret)
+				return ret;
+		}
+
+		flags = req->sr_msg.msg_flags;
+		if (flags & MSG_DONTWAIT)
+			req->flags |= REQ_F_NOWAIT;
+		else if (force_nonblock)
+			flags |= MSG_DONTWAIT;
+
+		ret = __sys_sendmsg_sock(sock, &kmsg->msg, flags);
+		if (force_nonblock && ret == -EAGAIN)
+			return io_setup_async_msg(req, kmsg);
+		if (ret == -ERESTARTSYS)
+			ret = -EINTR;
+	}
+
+	if (kmsg && kmsg->iov != kmsg->fast_iov)
+		kfree(kmsg->iov);
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+	io_cqring_add_event(req, ret);
+	if (ret < 0)
+		req_set_fail_links(req);
 	io_put_req(req);
 	return 0;
 }
 
-#if defined(CONFIG_NET)
-static int io_send_recvmsg(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-			   bool force_nonblock,
-		   long (*fn)(struct socket *, struct user_msghdr __user *,
-				unsigned int))
+static int io_send(struct io_kiocb *req, bool force_nonblock)
 {
 	struct socket *sock;
 	int ret;
@@ -1637,256 +3636,1007 @@ static int io_send_recvmsg(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 
 	sock = sock_from_file(req->file, &ret);
 	if (sock) {
-		struct user_msghdr __user *msg;
+		struct io_sr_msg *sr = &req->sr_msg;
+		struct msghdr msg;
+		struct iovec iov;
 		unsigned flags;
 
-		flags = READ_ONCE(sqe->msg_flags);
+		ret = import_single_range(WRITE, sr->buf, sr->len, &iov,
+						&msg.msg_iter);
+		if (ret)
+			return ret;
+
+		msg.msg_name = NULL;
+		msg.msg_control = NULL;
+		msg.msg_controllen = 0;
+		msg.msg_namelen = 0;
+
+		flags = req->sr_msg.msg_flags;
 		if (flags & MSG_DONTWAIT)
 			req->flags |= REQ_F_NOWAIT;
 		else if (force_nonblock)
 			flags |= MSG_DONTWAIT;
 
-		msg = (struct user_msghdr __user *) (unsigned long)
-			READ_ONCE(sqe->addr);
-
-		ret = fn(sock, msg, flags);
+		msg.msg_flags = flags;
+		ret = sock_sendmsg(sock, &msg);
 		if (force_nonblock && ret == -EAGAIN)
-			return ret;
+			return -EAGAIN;
+		if (ret == -ERESTARTSYS)
+			ret = -EINTR;
 	}
 
-	io_cqring_add_event(req->ctx, sqe->user_data, ret);
+	io_cqring_add_event(req, ret);
+	if (ret < 0)
+		req_set_fail_links(req);
 	io_put_req(req);
 	return 0;
 }
+
+static int __io_recvmsg_copy_hdr(struct io_kiocb *req, struct io_async_ctx *io)
+{
+	struct io_sr_msg *sr = &req->sr_msg;
+	struct iovec __user *uiov;
+	size_t iov_len;
+	int ret;
+
+	ret = __copy_msghdr_from_user(&io->msg.msg, sr->msg, &io->msg.uaddr,
+					&uiov, &iov_len);
+	if (ret)
+		return ret;
+
+	if (req->flags & REQ_F_BUFFER_SELECT) {
+		if (iov_len > 1)
+			return -EINVAL;
+		if (copy_from_user(io->msg.iov, uiov, sizeof(*uiov)))
+			return -EFAULT;
+		sr->len = io->msg.iov[0].iov_len;
+		iov_iter_init(&io->msg.msg.msg_iter, READ, io->msg.iov, 1,
+				sr->len);
+		io->msg.iov = NULL;
+	} else {
+		ret = import_iovec(READ, uiov, iov_len, UIO_FASTIOV,
+					&io->msg.iov, &io->msg.msg.msg_iter);
+		if (ret > 0)
+			ret = 0;
+	}
+
+	return ret;
+}
+
+#ifdef CONFIG_COMPAT
+static int __io_compat_recvmsg_copy_hdr(struct io_kiocb *req,
+					struct io_async_ctx *io)
+{
+	struct compat_msghdr __user *msg_compat;
+	struct io_sr_msg *sr = &req->sr_msg;
+	struct compat_iovec __user *uiov;
+	compat_uptr_t ptr;
+	compat_size_t len;
+	int ret;
+
+	msg_compat = (struct compat_msghdr __user *) sr->msg;
+	ret = __get_compat_msghdr(&io->msg.msg, msg_compat, &io->msg.uaddr,
+					&ptr, &len);
+	if (ret)
+		return ret;
+
+	uiov = compat_ptr(ptr);
+	if (req->flags & REQ_F_BUFFER_SELECT) {
+		compat_ssize_t clen;
+
+		if (len > 1)
+			return -EINVAL;
+		if (!access_ok(uiov, sizeof(*uiov)))
+			return -EFAULT;
+		if (__get_user(clen, &uiov->iov_len))
+			return -EFAULT;
+		if (clen < 0)
+			return -EINVAL;
+		sr->len = io->msg.iov[0].iov_len;
+		io->msg.iov = NULL;
+	} else {
+		ret = compat_import_iovec(READ, uiov, len, UIO_FASTIOV,
+						&io->msg.iov,
+						&io->msg.msg.msg_iter);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
 #endif
 
-static int io_sendmsg(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		      bool force_nonblock)
+static int io_recvmsg_copy_hdr(struct io_kiocb *req, struct io_async_ctx *io)
 {
-#if defined(CONFIG_NET)
-	return io_send_recvmsg(req, sqe, force_nonblock, __sys_sendmsg_sock);
-#else
-	return -EOPNOTSUPP;
+	io->msg.iov = io->msg.fast_iov;
+
+#ifdef CONFIG_COMPAT
+	if (req->ctx->compat)
+		return __io_compat_recvmsg_copy_hdr(req, io);
 #endif
+
+	return __io_recvmsg_copy_hdr(req, io);
 }
 
-static int io_recvmsg(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		      bool force_nonblock)
+static struct io_buffer *io_recv_buffer_select(struct io_kiocb *req,
+					       int *cflags, bool needs_lock)
 {
-#if defined(CONFIG_NET)
-	return io_send_recvmsg(req, sqe, force_nonblock, __sys_recvmsg_sock);
-#else
-	return -EOPNOTSUPP;
+	struct io_sr_msg *sr = &req->sr_msg;
+	struct io_buffer *kbuf;
+
+	if (!(req->flags & REQ_F_BUFFER_SELECT))
+		return NULL;
+
+	kbuf = io_buffer_select(req, &sr->len, sr->bgid, sr->kbuf, needs_lock);
+	if (IS_ERR(kbuf))
+		return kbuf;
+
+	sr->kbuf = kbuf;
+	req->flags |= REQ_F_BUFFER_SELECTED;
+
+	*cflags = kbuf->bid << IORING_CQE_BUFFER_SHIFT;
+	*cflags |= IORING_CQE_F_BUFFER;
+	return kbuf;
+}
+
+static int io_recvmsg_prep(struct io_kiocb *req,
+			   const struct io_uring_sqe *sqe)
+{
+	struct io_sr_msg *sr = &req->sr_msg;
+	struct io_async_ctx *io = req->io;
+	int ret;
+
+	sr->msg_flags = READ_ONCE(sqe->msg_flags);
+	sr->msg = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	sr->len = READ_ONCE(sqe->len);
+	sr->bgid = READ_ONCE(sqe->buf_group);
+
+#ifdef CONFIG_COMPAT
+	if (req->ctx->compat)
+		sr->msg_flags |= MSG_CMSG_COMPAT;
 #endif
+
+	if (!io || req->opcode == IORING_OP_RECV)
+		return 0;
+	/* iovec is already imported */
+	if (req->flags & REQ_F_NEED_CLEANUP)
+		return 0;
+
+	ret = io_recvmsg_copy_hdr(req, io);
+	if (!ret)
+		req->flags |= REQ_F_NEED_CLEANUP;
+	return ret;
 }
 
-static void io_poll_remove_one(struct io_kiocb *req)
+static int io_recvmsg(struct io_kiocb *req, bool force_nonblock)
 {
-	struct io_poll_iocb *poll = &req->poll;
+	struct io_async_msghdr *kmsg = NULL;
+	struct socket *sock;
+	int ret, cflags = 0;
 
-	spin_lock(&poll->head->lock);
-	WRITE_ONCE(poll->canceled, true);
-	if (!list_empty(&poll->wait.entry)) {
-		list_del_init(&poll->wait.entry);
-		io_queue_async_work(req->ctx, req);
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+
+	sock = sock_from_file(req->file, &ret);
+	if (sock) {
+		struct io_buffer *kbuf;
+		struct io_async_ctx io;
+		unsigned flags;
+
+		if (req->io) {
+			kmsg = &req->io->msg;
+			kmsg->msg.msg_name = &req->io->msg.addr;
+			/* if iov is set, it's allocated already */
+			if (!kmsg->iov)
+				kmsg->iov = kmsg->fast_iov;
+			kmsg->msg.msg_iter.iov = kmsg->iov;
+		} else {
+			kmsg = &io.msg;
+			kmsg->msg.msg_name = &io.msg.addr;
+
+			ret = io_recvmsg_copy_hdr(req, &io);
+			if (ret)
+				return ret;
+		}
+
+		kbuf = io_recv_buffer_select(req, &cflags, !force_nonblock);
+		if (IS_ERR(kbuf)) {
+			return PTR_ERR(kbuf);
+		} else if (kbuf) {
+			kmsg->fast_iov[0].iov_base = u64_to_user_ptr(kbuf->addr);
+			iov_iter_init(&kmsg->msg.msg_iter, READ, kmsg->iov,
+					1, req->sr_msg.len);
+		}
+
+		flags = req->sr_msg.msg_flags;
+		if (flags & MSG_DONTWAIT)
+			req->flags |= REQ_F_NOWAIT;
+		else if (force_nonblock)
+			flags |= MSG_DONTWAIT;
+
+		ret = __sys_recvmsg_sock(sock, &kmsg->msg, req->sr_msg.msg,
+						kmsg->uaddr, flags);
+		if (force_nonblock && ret == -EAGAIN)
+			return io_setup_async_msg(req, kmsg);
+		if (ret == -ERESTARTSYS)
+			ret = -EINTR;
 	}
-	spin_unlock(&poll->head->lock);
 
-	list_del_init(&req->list);
+	if (kmsg && kmsg->iov != kmsg->fast_iov)
+		kfree(kmsg->iov);
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+	__io_cqring_add_event(req, ret, cflags);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_put_req(req);
+	return 0;
 }
 
-static void io_poll_remove_all(struct io_ring_ctx *ctx)
+static int io_recv(struct io_kiocb *req, bool force_nonblock)
 {
-	struct io_kiocb *req;
+	struct io_buffer *kbuf = NULL;
+	struct socket *sock;
+	int ret, cflags = 0;
 
-	spin_lock_irq(&ctx->completion_lock);
-	while (!list_empty(&ctx->cancel_list)) {
-		req = list_first_entry(&ctx->cancel_list, struct io_kiocb,list);
-		io_poll_remove_one(req);
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+
+	sock = sock_from_file(req->file, &ret);
+	if (sock) {
+		struct io_sr_msg *sr = &req->sr_msg;
+		void __user *buf = sr->buf;
+		struct msghdr msg;
+		struct iovec iov;
+		unsigned flags;
+
+		kbuf = io_recv_buffer_select(req, &cflags, !force_nonblock);
+		if (IS_ERR(kbuf))
+			return PTR_ERR(kbuf);
+		else if (kbuf)
+			buf = u64_to_user_ptr(kbuf->addr);
+
+		ret = import_single_range(READ, buf, sr->len, &iov,
+						&msg.msg_iter);
+		if (ret) {
+			kfree(kbuf);
+			return ret;
+		}
+
+		req->flags |= REQ_F_NEED_CLEANUP;
+		msg.msg_name = NULL;
+		msg.msg_control = NULL;
+		msg.msg_controllen = 0;
+		msg.msg_namelen = 0;
+		msg.msg_iocb = NULL;
+		msg.msg_flags = 0;
+
+		flags = req->sr_msg.msg_flags;
+		if (flags & MSG_DONTWAIT)
+			req->flags |= REQ_F_NOWAIT;
+		else if (force_nonblock)
+			flags |= MSG_DONTWAIT;
+
+		ret = sock_recvmsg(sock, &msg, flags);
+		if (force_nonblock && ret == -EAGAIN)
+			return -EAGAIN;
+		if (ret == -ERESTARTSYS)
+			ret = -EINTR;
 	}
-	spin_unlock_irq(&ctx->completion_lock);
+
+	kfree(kbuf);
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+	__io_cqring_add_event(req, ret, cflags);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_put_req(req);
+	return 0;
 }
 
-/*
- * Find a running poll command that matches one specified in sqe->addr,
- * and remove it if found.
- */
-static int io_poll_remove(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+static int io_accept_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
-	struct io_ring_ctx *ctx = req->ctx;
-	struct io_kiocb *poll_req, *next;
-	int ret = -ENOENT;
+	struct io_accept *accept = &req->accept;
 
-	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+	if (unlikely(req->ctx->flags & (IORING_SETUP_IOPOLL|IORING_SETUP_SQPOLL)))
 		return -EINVAL;
-	if (sqe->ioprio || sqe->off || sqe->len || sqe->buf_index ||
-	    sqe->poll_events)
+	if (sqe->ioprio || sqe->len || sqe->buf_index)
 		return -EINVAL;
 
-	spin_lock_irq(&ctx->completion_lock);
-	list_for_each_entry_safe(poll_req, next, &ctx->cancel_list, list) {
-		if (READ_ONCE(sqe->addr) == poll_req->user_data) {
-			io_poll_remove_one(poll_req);
-			ret = 0;
-			break;
-		}
+	accept->addr = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	accept->addr_len = u64_to_user_ptr(READ_ONCE(sqe->addr2));
+	accept->flags = READ_ONCE(sqe->accept_flags);
+	accept->nofile = rlimit(RLIMIT_NOFILE);
+	return 0;
+}
+
+static int __io_accept(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_accept *accept = &req->accept;
+	unsigned file_flags;
+	int ret;
+
+	file_flags = force_nonblock ? O_NONBLOCK : 0;
+	ret = __sys_accept4_file(req->file, file_flags, accept->addr,
+					accept->addr_len, accept->flags,
+					accept->nofile);
+	if (ret == -EAGAIN && force_nonblock)
+		return -EAGAIN;
+	if (ret == -ERESTARTSYS)
+		ret = -EINTR;
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+	return 0;
+}
+
+static void io_accept_finish(struct io_wq_work **workptr)
+{
+	struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work);
+
+	if (io_req_cancelled(req))
+		return;
+	__io_accept(req, false);
+	io_steal_work(req, workptr);
+}
+
+static int io_accept(struct io_kiocb *req, bool force_nonblock)
+{
+	int ret;
+
+	ret = __io_accept(req, force_nonblock);
+	if (ret == -EAGAIN && force_nonblock) {
+		req->work.func = io_accept_finish;
+		return -EAGAIN;
 	}
-	spin_unlock_irq(&ctx->completion_lock);
+	return 0;
+}
+
+static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_connect *conn = &req->connect;
+	struct io_async_ctx *io = req->io;
+
+	if (unlikely(req->ctx->flags & (IORING_SETUP_IOPOLL|IORING_SETUP_SQPOLL)))
+		return -EINVAL;
+	if (sqe->ioprio || sqe->len || sqe->buf_index || sqe->rw_flags)
+		return -EINVAL;
 
-	io_cqring_add_event(req->ctx, sqe->user_data, ret);
+	conn->addr = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	conn->addr_len =  READ_ONCE(sqe->addr2);
+
+	if (!io)
+		return 0;
+
+	return move_addr_to_kernel(conn->addr, conn->addr_len,
+					&io->connect.address);
+}
+
+static int io_connect(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_async_ctx __io, *io;
+	unsigned file_flags;
+	int ret;
+
+	if (req->io) {
+		io = req->io;
+	} else {
+		ret = move_addr_to_kernel(req->connect.addr,
+						req->connect.addr_len,
+						&__io.connect.address);
+		if (ret)
+			goto out;
+		io = &__io;
+	}
+
+	file_flags = force_nonblock ? O_NONBLOCK : 0;
+
+	ret = __sys_connect_file(req->file, &io->connect.address,
+					req->connect.addr_len, file_flags);
+	if ((ret == -EAGAIN || ret == -EINPROGRESS) && force_nonblock) {
+		if (req->io)
+			return -EAGAIN;
+		if (io_alloc_async_ctx(req)) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		memcpy(&req->io->connect, &__io.connect, sizeof(__io.connect));
+		return -EAGAIN;
+	}
+	if (ret == -ERESTARTSYS)
+		ret = -EINTR;
+out:
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
 	io_put_req(req);
 	return 0;
 }
+#else /* !CONFIG_NET */
+static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	return -EOPNOTSUPP;
+}
 
-static void io_poll_complete(struct io_ring_ctx *ctx, struct io_kiocb *req,
-			     __poll_t mask)
+static int io_sendmsg(struct io_kiocb *req, bool force_nonblock)
 {
-	req->poll.done = true;
-	io_cqring_fill_event(ctx, req->user_data, mangle_poll(mask));
-	io_commit_cqring(ctx);
+	return -EOPNOTSUPP;
 }
 
-static void io_poll_complete_work(struct work_struct *work)
+static int io_send(struct io_kiocb *req, bool force_nonblock)
 {
-	struct io_kiocb *req = container_of(work, struct io_kiocb, work);
-	struct io_poll_iocb *poll = &req->poll;
-	struct poll_table_struct pt = { ._key = poll->events };
-	struct io_ring_ctx *ctx = req->ctx;
-	__poll_t mask = 0;
+	return -EOPNOTSUPP;
+}
+
+static int io_recvmsg_prep(struct io_kiocb *req,
+			   const struct io_uring_sqe *sqe)
+{
+	return -EOPNOTSUPP;
+}
+
+static int io_recvmsg(struct io_kiocb *req, bool force_nonblock)
+{
+	return -EOPNOTSUPP;
+}
 
-	if (!READ_ONCE(poll->canceled))
-		mask = vfs_poll(poll->file, &pt) & poll->events;
+static int io_recv(struct io_kiocb *req, bool force_nonblock)
+{
+	return -EOPNOTSUPP;
+}
+
+static int io_accept_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	return -EOPNOTSUPP;
+}
+
+static int io_accept(struct io_kiocb *req, bool force_nonblock)
+{
+	return -EOPNOTSUPP;
+}
+
+static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	return -EOPNOTSUPP;
+}
+
+static int io_connect(struct io_kiocb *req, bool force_nonblock)
+{
+	return -EOPNOTSUPP;
+}
+#endif /* CONFIG_NET */
+
+struct io_poll_table {
+	struct poll_table_struct pt;
+	struct io_kiocb *req;
+	int error;
+};
+
+static int __io_async_wake(struct io_kiocb *req, struct io_poll_iocb *poll,
+			   __poll_t mask, task_work_func_t func)
+{
+	struct task_struct *tsk;
+	int ret;
+
+	/* for instances that support it check for an event match first: */
+	if (mask && !(mask & poll->events))
+		return 0;
 
+	trace_io_uring_task_add(req->ctx, req->opcode, req->user_data, mask);
+
+	list_del_init(&poll->wait.entry);
+
+	tsk = req->task;
+	req->result = mask;
+	init_task_work(&req->task_work, func);
 	/*
-	 * Note that ->ki_cancel callers also delete iocb from active_reqs after
-	 * calling ->ki_cancel.  We need the ctx_lock roundtrip here to
-	 * synchronize with them.  In the cancellation case the list_del_init
-	 * itself is not actually needed, but harmless so we keep it in to
-	 * avoid further branches in the fast path.
+	 * If this fails, then the task is exiting. When a task exits, the
+	 * work gets canceled, so just cancel this request as well instead
+	 * of executing it. We can't safely execute it anyway, as we may not
+	 * have the needed state needed for it anyway.
 	 */
+	ret = task_work_add(tsk, &req->task_work, true);
+	if (unlikely(ret)) {
+		WRITE_ONCE(poll->canceled, true);
+		tsk = io_wq_get_task(req->ctx->io_wq);
+		task_work_add(tsk, &req->task_work, true);
+	}
+	wake_up_process(tsk);
+	return 1;
+}
+
+static bool io_poll_rewait(struct io_kiocb *req, struct io_poll_iocb *poll)
+	__acquires(&req->ctx->completion_lock)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+
+	if (!req->result && !READ_ONCE(poll->canceled)) {
+		struct poll_table_struct pt = { ._key = poll->events };
+
+		req->result = vfs_poll(req->file, &pt) & poll->events;
+	}
+
 	spin_lock_irq(&ctx->completion_lock);
-	if (!mask && !READ_ONCE(poll->canceled)) {
+	if (!req->result && !READ_ONCE(poll->canceled)) {
 		add_wait_queue(poll->head, &poll->wait);
+		return true;
+	}
+
+	return false;
+}
+
+static void io_poll_remove_double(struct io_kiocb *req)
+{
+	struct io_poll_iocb *poll = (struct io_poll_iocb *) req->io;
+
+	lockdep_assert_held(&req->ctx->completion_lock);
+
+	if (poll && poll->head) {
+		struct wait_queue_head *head = poll->head;
+
+		spin_lock(&head->lock);
+		list_del_init(&poll->wait.entry);
+		if (poll->wait.private)
+			refcount_dec(&req->refs);
+		poll->head = NULL;
+		spin_unlock(&head->lock);
+	}
+}
+
+static void io_poll_complete(struct io_kiocb *req, __poll_t mask, int error)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+
+	io_poll_remove_double(req);
+	req->poll.done = true;
+	io_cqring_fill_event(req, error ? error : mangle_poll(mask));
+	io_commit_cqring(ctx);
+}
+
+static void io_poll_task_handler(struct io_kiocb *req, struct io_kiocb **nxt)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+
+	if (io_poll_rewait(req, &req->poll)) {
 		spin_unlock_irq(&ctx->completion_lock);
 		return;
 	}
-	list_del_init(&req->list);
-	io_poll_complete(ctx, req, mask);
+
+	hash_del(&req->hash_node);
+	io_poll_complete(req, req->result, 0);
+	req->flags |= REQ_F_COMP_LOCKED;
+	io_put_req_find_next(req, nxt);
 	spin_unlock_irq(&ctx->completion_lock);
 
 	io_cqring_ev_posted(ctx);
-	io_put_req(req);
 }
 
-static int io_poll_wake(struct wait_queue_entry *wait, unsigned mode, int sync,
-			void *key)
+static void io_poll_task_func(struct callback_head *cb)
 {
-	struct io_poll_iocb *poll = container_of(wait, struct io_poll_iocb,
-							wait);
-	struct io_kiocb *req = container_of(poll, struct io_kiocb, poll);
-	struct io_ring_ctx *ctx = req->ctx;
+	struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work);
+	struct io_kiocb *nxt = NULL;
+
+	io_poll_task_handler(req, &nxt);
+	if (nxt) {
+		struct io_ring_ctx *ctx = nxt->ctx;
+
+		mutex_lock(&ctx->uring_lock);
+		__io_queue_sqe(nxt, NULL);
+		mutex_unlock(&ctx->uring_lock);
+	}
+}
+
+static int io_poll_double_wake(struct wait_queue_entry *wait, unsigned mode,
+			       int sync, void *key)
+{
+	struct io_kiocb *req = wait->private;
+	struct io_poll_iocb *poll = (struct io_poll_iocb *) req->io;
 	__poll_t mask = key_to_poll(key);
-	unsigned long flags;
 
 	/* for instances that support it check for an event match first: */
 	if (mask && !(mask & poll->events))
 		return 0;
 
-	list_del_init(&poll->wait.entry);
-
-	if (mask && spin_trylock_irqsave(&ctx->completion_lock, flags)) {
-		list_del(&req->list);
-		io_poll_complete(ctx, req, mask);
-		spin_unlock_irqrestore(&ctx->completion_lock, flags);
+	if (req->poll.head) {
+		bool done;
 
-		io_cqring_ev_posted(ctx);
-		io_put_req(req);
-	} else {
-		io_queue_async_work(ctx, req);
+		spin_lock(&req->poll.head->lock);
+		done = list_empty(&req->poll.wait.entry);
+		if (!done)
+			list_del_init(&req->poll.wait.entry);
+		spin_unlock(&req->poll.head->lock);
+		if (!done)
+			__io_async_wake(req, poll, mask, io_poll_task_func);
 	}
-
+	refcount_dec(&req->refs);
 	return 1;
 }
 
-struct io_poll_table {
-	struct poll_table_struct pt;
-	struct io_kiocb *req;
-	int error;
-};
+static void io_init_poll_iocb(struct io_poll_iocb *poll, __poll_t events,
+			      wait_queue_func_t wake_func)
+{
+	poll->head = NULL;
+	poll->done = false;
+	poll->canceled = false;
+	poll->events = events;
+	INIT_LIST_HEAD(&poll->wait.entry);
+	init_waitqueue_func_entry(&poll->wait, wake_func);
+}
 
-static void io_poll_queue_proc(struct file *file, struct wait_queue_head *head,
+static void __io_queue_proc(struct io_poll_iocb *poll, struct io_poll_table *pt,
+			    struct wait_queue_head *head)
+{
+	struct io_kiocb *req = pt->req;
+
+	/*
+	 * If poll->head is already set, it's because the file being polled
+	 * uses multiple waitqueues for poll handling (eg one for read, one
+	 * for write). Setup a separate io_poll_iocb if this happens.
+	 */
+	if (unlikely(poll->head)) {
+		/* already have a 2nd entry, fail a third attempt */
+		if (req->io) {
+			pt->error = -EINVAL;
+			return;
+		}
+		poll = kmalloc(sizeof(*poll), GFP_ATOMIC);
+		if (!poll) {
+			pt->error = -ENOMEM;
+			return;
+		}
+		io_init_poll_iocb(poll, req->poll.events, io_poll_double_wake);
+		refcount_inc(&req->refs);
+		poll->wait.private = req;
+		req->io = (void *) poll;
+	}
+
+	pt->error = 0;
+	poll->head = head;
+	add_wait_queue(head, &poll->wait);
+}
+
+static void io_async_queue_proc(struct file *file, struct wait_queue_head *head,
 			       struct poll_table_struct *p)
 {
 	struct io_poll_table *pt = container_of(p, struct io_poll_table, pt);
 
-	if (unlikely(pt->req->poll.head)) {
-		pt->error = -EINVAL;
+	__io_queue_proc(&pt->req->apoll->poll, pt, head);
+}
+
+static void io_async_task_func(struct callback_head *cb)
+{
+	struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work);
+	struct async_poll *apoll = req->apoll;
+	struct io_ring_ctx *ctx = req->ctx;
+	bool canceled = false;
+
+	trace_io_uring_task_run(req->ctx, req->opcode, req->user_data);
+
+	if (io_poll_rewait(req, &apoll->poll)) {
+		spin_unlock_irq(&ctx->completion_lock);
 		return;
 	}
 
-	pt->error = 0;
-	pt->req->poll.head = head;
-	add_wait_queue(head, &pt->req->poll.wait);
+	/* If req is still hashed, it cannot have been canceled. Don't check. */
+	if (hash_hashed(&req->hash_node)) {
+		hash_del(&req->hash_node);
+	} else {
+		canceled = READ_ONCE(apoll->poll.canceled);
+		if (canceled) {
+			io_cqring_fill_event(req, -ECANCELED);
+			io_commit_cqring(ctx);
+		}
+	}
+
+	spin_unlock_irq(&ctx->completion_lock);
+
+	/* restore ->work in case we need to retry again */
+	memcpy(&req->work, &apoll->work, sizeof(req->work));
+	kfree(apoll);
+
+	if (!canceled) {
+		__set_current_state(TASK_RUNNING);
+		mutex_lock(&ctx->uring_lock);
+		__io_queue_sqe(req, NULL);
+		mutex_unlock(&ctx->uring_lock);
+	} else {
+		io_cqring_ev_posted(ctx);
+		req_set_fail_links(req);
+		io_double_put_req(req);
+	}
 }
 
-static int io_poll_add(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+static int io_async_wake(struct wait_queue_entry *wait, unsigned mode, int sync,
+			void *key)
 {
-	struct io_poll_iocb *poll = &req->poll;
-	struct io_ring_ctx *ctx = req->ctx;
-	struct io_poll_table ipt;
-	bool cancel = false;
-	__poll_t mask;
-	u16 events;
+	struct io_kiocb *req = wait->private;
+	struct io_poll_iocb *poll = &req->apoll->poll;
 
-	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
-		return -EINVAL;
-	if (sqe->addr || sqe->ioprio || sqe->off || sqe->len || sqe->buf_index)
-		return -EINVAL;
-	if (!poll->file)
-		return -EBADF;
+	trace_io_uring_poll_wake(req->ctx, req->opcode, req->user_data,
+					key_to_poll(key));
 
-	req->submit.sqe = NULL;
-	INIT_WORK(&req->work, io_poll_complete_work);
-	events = READ_ONCE(sqe->poll_events);
-	poll->events = demangle_poll(events) | EPOLLERR | EPOLLHUP;
+	return __io_async_wake(req, poll, key_to_poll(key), io_async_task_func);
+}
 
-	poll->head = NULL;
-	poll->done = false;
-	poll->canceled = false;
+static void io_poll_req_insert(struct io_kiocb *req)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	struct hlist_head *list;
 
-	ipt.pt._qproc = io_poll_queue_proc;
-	ipt.pt._key = poll->events;
-	ipt.req = req;
-	ipt.error = -EINVAL; /* same as no support for IOCB_CMD_POLL */
+	list = &ctx->cancel_hash[hash_long(req->user_data, ctx->cancel_hash_bits)];
+	hlist_add_head(&req->hash_node, list);
+}
 
-	/* initialized the list so that we can do list_empty checks */
-	INIT_LIST_HEAD(&poll->wait.entry);
-	init_waitqueue_func_entry(&poll->wait, io_poll_wake);
+static __poll_t __io_arm_poll_handler(struct io_kiocb *req,
+				      struct io_poll_iocb *poll,
+				      struct io_poll_table *ipt, __poll_t mask,
+				      wait_queue_func_t wake_func)
+	__acquires(&ctx->completion_lock)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	bool cancel = false;
 
-	INIT_LIST_HEAD(&req->list);
+	poll->file = req->file;
+	io_init_poll_iocb(poll, mask, wake_func);
+	poll->wait.private = req;
 
-	mask = vfs_poll(poll->file, &ipt.pt) & poll->events;
+	ipt->pt._key = mask;
+	ipt->req = req;
+	ipt->error = -EINVAL;
+
+	mask = vfs_poll(req->file, &ipt->pt) & poll->events;
 
 	spin_lock_irq(&ctx->completion_lock);
 	if (likely(poll->head)) {
 		spin_lock(&poll->head->lock);
 		if (unlikely(list_empty(&poll->wait.entry))) {
-			if (ipt.error)
+			if (ipt->error)
 				cancel = true;
-			ipt.error = 0;
+			ipt->error = 0;
 			mask = 0;
 		}
-		if (mask || ipt.error)
+		if (mask || ipt->error)
 			list_del_init(&poll->wait.entry);
 		else if (cancel)
 			WRITE_ONCE(poll->canceled, true);
 		else if (!poll->done) /* actually waiting for an event */
-			list_add_tail(&req->list, &ctx->cancel_list);
+			io_poll_req_insert(req);
 		spin_unlock(&poll->head->lock);
 	}
+
+	return mask;
+}
+
+static bool io_arm_poll_handler(struct io_kiocb *req)
+{
+	const struct io_op_def *def = &io_op_defs[req->opcode];
+	struct io_ring_ctx *ctx = req->ctx;
+	struct async_poll *apoll;
+	struct io_poll_table ipt;
+	__poll_t mask, ret;
+	bool had_io;
+
+	if (!req->file || !file_can_poll(req->file))
+		return false;
+	if (req->flags & (REQ_F_MUST_PUNT | REQ_F_POLLED))
+		return false;
+	if (!def->pollin && !def->pollout)
+		return false;
+
+	apoll = kmalloc(sizeof(*apoll), GFP_ATOMIC);
+	if (unlikely(!apoll))
+		return false;
+
+	req->flags |= REQ_F_POLLED;
+	memcpy(&apoll->work, &req->work, sizeof(req->work));
+	had_io = req->io != NULL;
+
+	get_task_struct(current);
+	req->task = current;
+	req->apoll = apoll;
+	INIT_HLIST_NODE(&req->hash_node);
+
+	mask = 0;
+	if (def->pollin)
+		mask |= POLLIN | POLLRDNORM;
+	if (def->pollout)
+		mask |= POLLOUT | POLLWRNORM;
+	mask |= POLLERR | POLLPRI;
+
+	ipt.pt._qproc = io_async_queue_proc;
+
+	ret = __io_arm_poll_handler(req, &apoll->poll, &ipt, mask,
+					io_async_wake);
+	if (ret) {
+		ipt.error = 0;
+		/* only remove double add if we did it here */
+		if (!had_io)
+			io_poll_remove_double(req);
+		spin_unlock_irq(&ctx->completion_lock);
+		memcpy(&req->work, &apoll->work, sizeof(req->work));
+		kfree(apoll);
+		return false;
+	}
+	spin_unlock_irq(&ctx->completion_lock);
+	trace_io_uring_poll_arm(ctx, req->opcode, req->user_data, mask,
+					apoll->poll.events);
+	return true;
+}
+
+static bool __io_poll_remove_one(struct io_kiocb *req,
+				 struct io_poll_iocb *poll)
+{
+	bool do_complete = false;
+
+	spin_lock(&poll->head->lock);
+	WRITE_ONCE(poll->canceled, true);
+	if (!list_empty(&poll->wait.entry)) {
+		list_del_init(&poll->wait.entry);
+		do_complete = true;
+	}
+	spin_unlock(&poll->head->lock);
+	hash_del(&req->hash_node);
+	return do_complete;
+}
+
+static bool io_poll_remove_one(struct io_kiocb *req)
+{
+	bool do_complete;
+
+	if (req->opcode == IORING_OP_POLL_ADD) {
+		io_poll_remove_double(req);
+		do_complete = __io_poll_remove_one(req, &req->poll);
+	} else {
+		struct async_poll *apoll = req->apoll;
+
+		/* non-poll requests have submit ref still */
+		do_complete = __io_poll_remove_one(req, &apoll->poll);
+		if (do_complete) {
+			io_put_req(req);
+			/*
+			 * restore ->work because we will call
+			 * io_req_work_drop_env below when dropping the
+			 * final reference.
+			 */
+			memcpy(&req->work, &apoll->work, sizeof(req->work));
+			kfree(apoll);
+		}
+	}
+
+	if (do_complete) {
+		io_cqring_fill_event(req, -ECANCELED);
+		io_commit_cqring(req->ctx);
+		req->flags |= REQ_F_COMP_LOCKED;
+		io_put_req(req);
+	}
+
+	return do_complete;
+}
+
+static void io_poll_remove_all(struct io_ring_ctx *ctx)
+{
+	struct hlist_node *tmp;
+	struct io_kiocb *req;
+	int posted = 0, i;
+
+	spin_lock_irq(&ctx->completion_lock);
+	for (i = 0; i < (1U << ctx->cancel_hash_bits); i++) {
+		struct hlist_head *list;
+
+		list = &ctx->cancel_hash[i];
+		hlist_for_each_entry_safe(req, tmp, list, hash_node)
+			posted += io_poll_remove_one(req);
+	}
+	spin_unlock_irq(&ctx->completion_lock);
+
+	if (posted)
+		io_cqring_ev_posted(ctx);
+}
+
+static int io_poll_cancel(struct io_ring_ctx *ctx, __u64 sqe_addr)
+{
+	struct hlist_head *list;
+	struct io_kiocb *req;
+
+	list = &ctx->cancel_hash[hash_long(sqe_addr, ctx->cancel_hash_bits)];
+	hlist_for_each_entry(req, list, hash_node) {
+		if (sqe_addr != req->user_data)
+			continue;
+		if (io_poll_remove_one(req))
+			return 0;
+		return -EALREADY;
+	}
+
+	return -ENOENT;
+}
+
+static int io_poll_remove_prep(struct io_kiocb *req,
+			       const struct io_uring_sqe *sqe)
+{
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (sqe->ioprio || sqe->off || sqe->len || sqe->buf_index ||
+	    sqe->poll_events)
+		return -EINVAL;
+
+	req->poll.addr = READ_ONCE(sqe->addr);
+	return 0;
+}
+
+/*
+ * Find a running poll command that matches one specified in sqe->addr,
+ * and remove it if found.
+ */
+static int io_poll_remove(struct io_kiocb *req)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	u64 addr;
+	int ret;
+
+	addr = req->poll.addr;
+	spin_lock_irq(&ctx->completion_lock);
+	ret = io_poll_cancel(ctx, addr);
+	spin_unlock_irq(&ctx->completion_lock);
+
+	io_cqring_add_event(req, ret);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_poll_wake(struct wait_queue_entry *wait, unsigned mode, int sync,
+			void *key)
+{
+	struct io_kiocb *req = wait->private;
+	struct io_poll_iocb *poll = &req->poll;
+
+	return __io_async_wake(req, poll, key_to_poll(key), io_poll_task_func);
+}
+
+static void io_poll_queue_proc(struct file *file, struct wait_queue_head *head,
+			       struct poll_table_struct *p)
+{
+	struct io_poll_table *pt = container_of(p, struct io_poll_table, pt);
+
+	__io_queue_proc(&pt->req->poll, pt, head);
+}
+
+static int io_poll_add_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_poll_iocb *poll = &req->poll;
+	u16 events;
+
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (sqe->addr || sqe->ioprio || sqe->off || sqe->len || sqe->buf_index)
+		return -EINVAL;
+	if (!poll->file)
+		return -EBADF;
+
+	events = READ_ONCE(sqe->poll_events);
+	poll->events = demangle_poll(events) | EPOLLERR | EPOLLHUP;
+
+	get_task_struct(current);
+	req->task = current;
+	return 0;
+}
+
+static int io_poll_add(struct io_kiocb *req)
+{
+	struct io_poll_iocb *poll = &req->poll;
+	struct io_ring_ctx *ctx = req->ctx;
+	struct io_poll_table ipt;
+	__poll_t mask;
+
+	INIT_HLIST_NODE(&req->hash_node);
+	INIT_LIST_HEAD(&req->list);
+	ipt.pt._qproc = io_poll_queue_proc;
+
+	mask = __io_arm_poll_handler(req, &req->poll, &ipt, poll->events,
+					io_poll_wake);
+
 	if (mask) { /* no async, we'd stolen it */
 		ipt.error = 0;
-		io_poll_complete(ctx, req, mask);
+		io_poll_complete(req, mask, 0);
 	}
 	spin_unlock_irq(&ctx->completion_lock);
 
@@ -1899,201 +4649,710 @@ static int io_poll_add(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 
 static enum hrtimer_restart io_timeout_fn(struct hrtimer *timer)
 {
-	struct io_ring_ctx *ctx;
-	struct io_kiocb *req, *prev;
+	struct io_timeout_data *data = container_of(timer,
+						struct io_timeout_data, timer);
+	struct io_kiocb *req = data->req;
+	struct io_ring_ctx *ctx = req->ctx;
 	unsigned long flags;
 
-	req = container_of(timer, struct io_kiocb, timeout.timer);
-	ctx = req->ctx;
 	atomic_inc(&ctx->cq_timeouts);
 
 	spin_lock_irqsave(&ctx->completion_lock, flags);
 	/*
-	 * Adjust the reqs sequence before the current one because it
-	 * will consume a slot in the cq_ring and the the cq_tail pointer
-	 * will be increased, otherwise other timeout reqs may return in
-	 * advance without waiting for enough wait_nr.
+	 * We could be racing with timeout deletion. If the list is empty,
+	 * then timeout lookup already found it and will be handling it.
 	 */
-	prev = req;
-	list_for_each_entry_continue_reverse(prev, &ctx->timeout_list, list)
-		prev->sequence++;
-	list_del(&req->list);
+	if (!list_empty(&req->list))
+		list_del_init(&req->list);
 
-	io_cqring_fill_event(ctx, req->user_data, -ETIME);
+	io_cqring_fill_event(req, -ETIME);
 	io_commit_cqring(ctx);
 	spin_unlock_irqrestore(&ctx->completion_lock, flags);
 
 	io_cqring_ev_posted(ctx);
-
+	req_set_fail_links(req);
 	io_put_req(req);
 	return HRTIMER_NORESTART;
 }
 
-static int io_timeout(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+static int io_timeout_cancel(struct io_ring_ctx *ctx, __u64 user_data)
+{
+	struct io_kiocb *req;
+	int ret = -ENOENT;
+
+	list_for_each_entry(req, &ctx->timeout_list, list) {
+		if (user_data == req->user_data) {
+			list_del_init(&req->list);
+			ret = 0;
+			break;
+		}
+	}
+
+	if (ret == -ENOENT)
+		return ret;
+
+	ret = hrtimer_try_to_cancel(&req->io->timeout.timer);
+	if (ret == -1)
+		return -EALREADY;
+
+	req_set_fail_links(req);
+	io_cqring_fill_event(req, -ECANCELED);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_timeout_remove_prep(struct io_kiocb *req,
+				  const struct io_uring_sqe *sqe)
+{
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (sqe->flags || sqe->ioprio || sqe->buf_index || sqe->len)
+		return -EINVAL;
+
+	req->timeout.addr = READ_ONCE(sqe->addr);
+	req->timeout.flags = READ_ONCE(sqe->timeout_flags);
+	if (req->timeout.flags)
+		return -EINVAL;
+
+	return 0;
+}
+
+/*
+ * Remove or update an existing timeout command
+ */
+static int io_timeout_remove(struct io_kiocb *req)
 {
-	unsigned count;
 	struct io_ring_ctx *ctx = req->ctx;
-	struct list_head *entry;
-	struct timespec64 ts;
-	unsigned span = 0;
+	int ret;
 
-	if (unlikely(ctx->flags & IORING_SETUP_IOPOLL))
+	spin_lock_irq(&ctx->completion_lock);
+	ret = io_timeout_cancel(ctx, req->timeout.addr);
+
+	io_cqring_fill_event(req, ret);
+	io_commit_cqring(ctx);
+	spin_unlock_irq(&ctx->completion_lock);
+	io_cqring_ev_posted(ctx);
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_timeout_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			   bool is_timeout_link)
+{
+	struct io_timeout_data *data;
+	unsigned flags;
+	u32 off = READ_ONCE(sqe->off);
+
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (sqe->ioprio || sqe->buf_index || sqe->len != 1)
 		return -EINVAL;
-	if (sqe->flags || sqe->ioprio || sqe->buf_index || sqe->timeout_flags ||
-	    sqe->len != 1)
+	if (off && is_timeout_link)
+		return -EINVAL;
+	flags = READ_ONCE(sqe->timeout_flags);
+	if (flags & ~IORING_TIMEOUT_ABS)
 		return -EINVAL;
 
-	if (get_timespec64(&ts, u64_to_user_ptr(sqe->addr)))
-		return -EFAULT;
+	req->timeout.off = off;
+
+	if (!req->io && io_alloc_async_ctx(req))
+		return -ENOMEM;
 
+	data = &req->io->timeout;
+	data->req = req;
 	req->flags |= REQ_F_TIMEOUT;
 
+	if (get_timespec64(&data->ts, u64_to_user_ptr(sqe->addr)))
+		return -EFAULT;
+
+	if (flags & IORING_TIMEOUT_ABS)
+		data->mode = HRTIMER_MODE_ABS;
+	else
+		data->mode = HRTIMER_MODE_REL;
+
+	hrtimer_init(&data->timer, CLOCK_MONOTONIC, data->mode);
+	return 0;
+}
+
+static int io_timeout(struct io_kiocb *req)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	struct io_timeout_data *data = &req->io->timeout;
+	struct list_head *entry;
+	u32 tail, off = req->timeout.off;
+
+	spin_lock_irq(&ctx->completion_lock);
+
 	/*
 	 * sqe->off holds how many events that need to occur for this
 	 * timeout event to be satisfied. If it isn't set, then this is
 	 * a pure timeout request, sequence isn't used.
 	 */
-	count = READ_ONCE(sqe->off);
-	if (!count) {
+	if (!off) {
 		req->flags |= REQ_F_TIMEOUT_NOSEQ;
-		spin_lock_irq(&ctx->completion_lock);
 		entry = ctx->timeout_list.prev;
 		goto add;
 	}
 
-	req->sequence = ctx->cached_sq_head + count - 1;
-	/* reuse it to store the count */
-	req->submit.sequence = count;
+	tail = ctx->cached_cq_tail - atomic_read(&ctx->cq_timeouts);
+	req->timeout.target_seq = tail + off;
 
 	/*
 	 * Insertion sort, ensuring the first entry in the list is always
 	 * the one we need first.
 	 */
-	spin_lock_irq(&ctx->completion_lock);
 	list_for_each_prev(entry, &ctx->timeout_list) {
 		struct io_kiocb *nxt = list_entry(entry, struct io_kiocb, list);
-		unsigned nxt_sq_head;
-		long long tmp, tmp_nxt;
 
 		if (nxt->flags & REQ_F_TIMEOUT_NOSEQ)
 			continue;
-
-		/*
-		 * Since cached_sq_head + count - 1 can overflow, use type long
-		 * long to store it.
-		 */
-		tmp = (long long)ctx->cached_sq_head + count - 1;
-		nxt_sq_head = nxt->sequence - nxt->submit.sequence + 1;
-		tmp_nxt = (long long)nxt_sq_head + nxt->submit.sequence - 1;
-
-		/*
-		 * cached_sq_head may overflow, and it will never overflow twice
-		 * once there is some timeout req still be valid.
-		 */
-		if (ctx->cached_sq_head < nxt_sq_head)
-			tmp += UINT_MAX;
-
-		if (tmp > tmp_nxt)
+		/* nxt.seq is behind @tail, otherwise would've been completed */
+		if (off >= nxt->timeout.target_seq - tail)
 			break;
-
-		/*
-		 * Sequence of reqs after the insert one and itself should
-		 * be adjusted because each timeout req consumes a slot.
-		 */
-		span++;
-		nxt->sequence++;
 	}
-	req->sequence -= span;
 add:
 	list_add(&req->list, entry);
+	data->timer.function = io_timeout_fn;
+	hrtimer_start(&data->timer, timespec64_to_ktime(data->ts), data->mode);
 	spin_unlock_irq(&ctx->completion_lock);
+	return 0;
+}
+
+static bool io_cancel_cb(struct io_wq_work *work, void *data)
+{
+	struct io_kiocb *req = container_of(work, struct io_kiocb, work);
+
+	return req->user_data == (unsigned long) data;
+}
+
+static int io_async_cancel_one(struct io_ring_ctx *ctx, void *sqe_addr)
+{
+	enum io_wq_cancel cancel_ret;
+	int ret = 0;
+
+	cancel_ret = io_wq_cancel_cb(ctx->io_wq, io_cancel_cb, sqe_addr);
+	switch (cancel_ret) {
+	case IO_WQ_CANCEL_OK:
+		ret = 0;
+		break;
+	case IO_WQ_CANCEL_RUNNING:
+		ret = -EALREADY;
+		break;
+	case IO_WQ_CANCEL_NOTFOUND:
+		ret = -ENOENT;
+		break;
+	}
+
+	return ret;
+}
+
+static void io_async_find_and_cancel(struct io_ring_ctx *ctx,
+				     struct io_kiocb *req, __u64 sqe_addr,
+				     int success_ret)
+{
+	unsigned long flags;
+	int ret;
+
+	ret = io_async_cancel_one(ctx, (void *) (unsigned long) sqe_addr);
+	if (ret != -ENOENT) {
+		spin_lock_irqsave(&ctx->completion_lock, flags);
+		goto done;
+	}
+
+	spin_lock_irqsave(&ctx->completion_lock, flags);
+	ret = io_timeout_cancel(ctx, sqe_addr);
+	if (ret != -ENOENT)
+		goto done;
+	ret = io_poll_cancel(ctx, sqe_addr);
+done:
+	if (!ret)
+		ret = success_ret;
+	io_cqring_fill_event(req, ret);
+	io_commit_cqring(ctx);
+	spin_unlock_irqrestore(&ctx->completion_lock, flags);
+	io_cqring_ev_posted(ctx);
+
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_put_req(req);
+}
+
+static int io_async_cancel_prep(struct io_kiocb *req,
+				const struct io_uring_sqe *sqe)
+{
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (sqe->flags || sqe->ioprio || sqe->off || sqe->len ||
+	    sqe->cancel_flags)
+		return -EINVAL;
 
-	hrtimer_init(&req->timeout.timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-	req->timeout.timer.function = io_timeout_fn;
-	hrtimer_start(&req->timeout.timer, timespec64_to_ktime(ts),
-			HRTIMER_MODE_REL);
+	req->cancel.addr = READ_ONCE(sqe->addr);
 	return 0;
 }
 
-static int io_req_defer(struct io_ring_ctx *ctx, struct io_kiocb *req,
-			const struct io_uring_sqe *sqe)
+static int io_async_cancel(struct io_kiocb *req)
 {
-	struct io_uring_sqe *sqe_copy;
+	struct io_ring_ctx *ctx = req->ctx;
 
-	if (!io_sequence_defer(ctx, req) && list_empty(&ctx->defer_list))
-		return 0;
+	io_async_find_and_cancel(ctx, req, req->cancel.addr, 0);
+	return 0;
+}
+
+static int io_files_update_prep(struct io_kiocb *req,
+				const struct io_uring_sqe *sqe)
+{
+	if (sqe->flags || sqe->ioprio || sqe->rw_flags)
+		return -EINVAL;
 
-	sqe_copy = kmalloc(sizeof(*sqe_copy), GFP_KERNEL);
-	if (!sqe_copy)
+	req->files_update.offset = READ_ONCE(sqe->off);
+	req->files_update.nr_args = READ_ONCE(sqe->len);
+	if (!req->files_update.nr_args)
+		return -EINVAL;
+	req->files_update.arg = READ_ONCE(sqe->addr);
+	return 0;
+}
+
+static int io_files_update(struct io_kiocb *req, bool force_nonblock)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	struct io_uring_files_update up;
+	int ret;
+
+	if (force_nonblock)
 		return -EAGAIN;
 
+	up.offset = req->files_update.offset;
+	up.fds = req->files_update.arg;
+
+	mutex_lock(&ctx->uring_lock);
+	ret = __io_sqe_files_update(ctx, &up, req->files_update.nr_args);
+	mutex_unlock(&ctx->uring_lock);
+
+	if (ret < 0)
+		req_set_fail_links(req);
+	io_cqring_add_event(req, ret);
+	io_put_req(req);
+	return 0;
+}
+
+static int io_req_defer_prep(struct io_kiocb *req,
+			     const struct io_uring_sqe *sqe)
+{
+	ssize_t ret = 0;
+
+	if (!sqe)
+		return 0;
+
+	if (io_op_defs[req->opcode].file_table) {
+		ret = io_grab_files(req);
+		if (unlikely(ret))
+			return ret;
+	}
+
+	io_req_work_grab_env(req, &io_op_defs[req->opcode]);
+
+	switch (req->opcode) {
+	case IORING_OP_NOP:
+		break;
+	case IORING_OP_READV:
+	case IORING_OP_READ_FIXED:
+	case IORING_OP_READ:
+		ret = io_read_prep(req, sqe, true);
+		break;
+	case IORING_OP_WRITEV:
+	case IORING_OP_WRITE_FIXED:
+	case IORING_OP_WRITE:
+		ret = io_write_prep(req, sqe, true);
+		break;
+	case IORING_OP_POLL_ADD:
+		ret = io_poll_add_prep(req, sqe);
+		break;
+	case IORING_OP_POLL_REMOVE:
+		ret = io_poll_remove_prep(req, sqe);
+		break;
+	case IORING_OP_FSYNC:
+		ret = io_prep_fsync(req, sqe);
+		break;
+	case IORING_OP_SYNC_FILE_RANGE:
+		ret = io_prep_sfr(req, sqe);
+		break;
+	case IORING_OP_SENDMSG:
+	case IORING_OP_SEND:
+		ret = io_sendmsg_prep(req, sqe);
+		break;
+	case IORING_OP_RECVMSG:
+	case IORING_OP_RECV:
+		ret = io_recvmsg_prep(req, sqe);
+		break;
+	case IORING_OP_CONNECT:
+		ret = io_connect_prep(req, sqe);
+		break;
+	case IORING_OP_TIMEOUT:
+		ret = io_timeout_prep(req, sqe, false);
+		break;
+	case IORING_OP_TIMEOUT_REMOVE:
+		ret = io_timeout_remove_prep(req, sqe);
+		break;
+	case IORING_OP_ASYNC_CANCEL:
+		ret = io_async_cancel_prep(req, sqe);
+		break;
+	case IORING_OP_LINK_TIMEOUT:
+		ret = io_timeout_prep(req, sqe, true);
+		break;
+	case IORING_OP_ACCEPT:
+		ret = io_accept_prep(req, sqe);
+		break;
+	case IORING_OP_FALLOCATE:
+		ret = io_fallocate_prep(req, sqe);
+		break;
+	case IORING_OP_OPENAT:
+		ret = io_openat_prep(req, sqe);
+		break;
+	case IORING_OP_CLOSE:
+		ret = io_close_prep(req, sqe);
+		break;
+	case IORING_OP_FILES_UPDATE:
+		ret = io_files_update_prep(req, sqe);
+		break;
+	case IORING_OP_STATX:
+		ret = io_statx_prep(req, sqe);
+		break;
+	case IORING_OP_FADVISE:
+		ret = io_fadvise_prep(req, sqe);
+		break;
+	case IORING_OP_MADVISE:
+		ret = io_madvise_prep(req, sqe);
+		break;
+	case IORING_OP_OPENAT2:
+		ret = io_openat2_prep(req, sqe);
+		break;
+	case IORING_OP_EPOLL_CTL:
+		ret = io_epoll_ctl_prep(req, sqe);
+		break;
+	case IORING_OP_SPLICE:
+		ret = io_splice_prep(req, sqe);
+		break;
+	case IORING_OP_PROVIDE_BUFFERS:
+		ret = io_provide_buffers_prep(req, sqe);
+		break;
+	case IORING_OP_REMOVE_BUFFERS:
+		ret = io_remove_buffers_prep(req, sqe);
+		break;
+	case IORING_OP_TEE:
+		ret = io_tee_prep(req, sqe);
+		break;
+	default:
+		printk_once(KERN_WARNING "io_uring: unhandled opcode %d\n",
+				req->opcode);
+		ret = -EINVAL;
+		break;
+	}
+
+	return ret;
+}
+
+static int io_req_defer(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	int ret;
+
+	/* Still need defer if there is pending req in defer list. */
+	if (!req_need_defer(req) && list_empty_careful(&ctx->defer_list))
+		return 0;
+
+	if (!req->io) {
+		if (io_alloc_async_ctx(req))
+			return -EAGAIN;
+		ret = io_req_defer_prep(req, sqe);
+		if (ret < 0)
+			return ret;
+	}
+
 	spin_lock_irq(&ctx->completion_lock);
-	if (!io_sequence_defer(ctx, req) && list_empty(&ctx->defer_list)) {
+	if (!req_need_defer(req) && list_empty(&ctx->defer_list)) {
 		spin_unlock_irq(&ctx->completion_lock);
-		kfree(sqe_copy);
 		return 0;
 	}
 
-	memcpy(sqe_copy, sqe, sizeof(*sqe_copy));
-	req->submit.sqe = sqe_copy;
-
-	INIT_WORK(&req->work, io_sq_wq_submit_work);
+	trace_io_uring_defer(ctx, req, req->user_data);
 	list_add_tail(&req->list, &ctx->defer_list);
 	spin_unlock_irq(&ctx->completion_lock);
 	return -EIOCBQUEUED;
 }
 
-static int __io_submit_sqe(struct io_ring_ctx *ctx, struct io_kiocb *req,
-			   const struct sqe_submit *s, bool force_nonblock)
+static void io_cleanup_req(struct io_kiocb *req)
 {
-	int ret, opcode;
+	struct io_async_ctx *io = req->io;
 
-	req->user_data = READ_ONCE(s->sqe->user_data);
+	switch (req->opcode) {
+	case IORING_OP_READV:
+	case IORING_OP_READ_FIXED:
+	case IORING_OP_READ:
+		if (req->flags & REQ_F_BUFFER_SELECTED)
+			kfree((void *)(unsigned long)req->rw.addr);
+		/* fallthrough */
+	case IORING_OP_WRITEV:
+	case IORING_OP_WRITE_FIXED:
+	case IORING_OP_WRITE:
+		if (io->rw.iov != io->rw.fast_iov)
+			kfree(io->rw.iov);
+		break;
+	case IORING_OP_RECVMSG:
+		if (req->flags & REQ_F_BUFFER_SELECTED)
+			kfree(req->sr_msg.kbuf);
+		/* fallthrough */
+	case IORING_OP_SENDMSG:
+		if (io->msg.iov != io->msg.fast_iov)
+			kfree(io->msg.iov);
+		break;
+	case IORING_OP_RECV:
+		if (req->flags & REQ_F_BUFFER_SELECTED)
+			kfree(req->sr_msg.kbuf);
+		break;
+	case IORING_OP_OPENAT:
+	case IORING_OP_OPENAT2:
+		break;
+	case IORING_OP_SPLICE:
+	case IORING_OP_TEE:
+		io_put_file(req, req->splice.file_in,
+			    (req->splice.flags & SPLICE_F_FD_IN_FIXED));
+		break;
+	}
 
-	if (unlikely(s->index >= ctx->sq_entries))
-		return -EINVAL;
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+}
 
-	opcode = READ_ONCE(s->sqe->opcode);
-	switch (opcode) {
+static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			bool force_nonblock)
+{
+	struct io_ring_ctx *ctx = req->ctx;
+	int ret;
+
+	switch (req->opcode) {
 	case IORING_OP_NOP:
-		ret = io_nop(req, req->user_data);
+		ret = io_nop(req);
 		break;
 	case IORING_OP_READV:
-		if (unlikely(s->sqe->buf_index))
-			return -EINVAL;
-		ret = io_read(req, s, force_nonblock);
-		break;
-	case IORING_OP_WRITEV:
-		if (unlikely(s->sqe->buf_index))
-			return -EINVAL;
-		ret = io_write(req, s, force_nonblock);
-		break;
 	case IORING_OP_READ_FIXED:
-		ret = io_read(req, s, force_nonblock);
+	case IORING_OP_READ:
+		if (sqe) {
+			ret = io_read_prep(req, sqe, force_nonblock);
+			if (ret < 0)
+				break;
+		}
+		ret = io_read(req, force_nonblock);
 		break;
+	case IORING_OP_WRITEV:
 	case IORING_OP_WRITE_FIXED:
-		ret = io_write(req, s, force_nonblock);
+	case IORING_OP_WRITE:
+		if (sqe) {
+			ret = io_write_prep(req, sqe, force_nonblock);
+			if (ret < 0)
+				break;
+		}
+		ret = io_write(req, force_nonblock);
 		break;
 	case IORING_OP_FSYNC:
-		ret = io_fsync(req, s->sqe, force_nonblock);
+		if (sqe) {
+			ret = io_prep_fsync(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_fsync(req, force_nonblock);
 		break;
 	case IORING_OP_POLL_ADD:
-		ret = io_poll_add(req, s->sqe);
+		if (sqe) {
+			ret = io_poll_add_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_poll_add(req);
 		break;
 	case IORING_OP_POLL_REMOVE:
-		ret = io_poll_remove(req, s->sqe);
+		if (sqe) {
+			ret = io_poll_remove_prep(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_poll_remove(req);
 		break;
 	case IORING_OP_SYNC_FILE_RANGE:
-		ret = io_sync_file_range(req, s->sqe, force_nonblock);
+		if (sqe) {
+			ret = io_prep_sfr(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_sync_file_range(req, force_nonblock);
 		break;
 	case IORING_OP_SENDMSG:
-		ret = io_sendmsg(req, s->sqe, force_nonblock);
+	case IORING_OP_SEND:
+		if (sqe) {
+			ret = io_sendmsg_prep(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		if (req->opcode == IORING_OP_SENDMSG)
+			ret = io_sendmsg(req, force_nonblock);
+		else
+			ret = io_send(req, force_nonblock);
 		break;
 	case IORING_OP_RECVMSG:
-		ret = io_recvmsg(req, s->sqe, force_nonblock);
+	case IORING_OP_RECV:
+		if (sqe) {
+			ret = io_recvmsg_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		if (req->opcode == IORING_OP_RECVMSG)
+			ret = io_recvmsg(req, force_nonblock);
+		else
+			ret = io_recv(req, force_nonblock);
 		break;
 	case IORING_OP_TIMEOUT:
-		ret = io_timeout(req, s->sqe);
+		if (sqe) {
+			ret = io_timeout_prep(req, sqe, false);
+			if (ret)
+				break;
+		}
+		ret = io_timeout(req);
+		break;
+	case IORING_OP_TIMEOUT_REMOVE:
+		if (sqe) {
+			ret = io_timeout_remove_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_timeout_remove(req);
+		break;
+	case IORING_OP_ACCEPT:
+		if (sqe) {
+			ret = io_accept_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_accept(req, force_nonblock);
+		break;
+	case IORING_OP_CONNECT:
+		if (sqe) {
+			ret = io_connect_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_connect(req, force_nonblock);
+		break;
+	case IORING_OP_ASYNC_CANCEL:
+		if (sqe) {
+			ret = io_async_cancel_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_async_cancel(req);
+		break;
+	case IORING_OP_FALLOCATE:
+		if (sqe) {
+			ret = io_fallocate_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_fallocate(req, force_nonblock);
+		break;
+	case IORING_OP_OPENAT:
+		if (sqe) {
+			ret = io_openat_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_openat(req, force_nonblock);
+		break;
+	case IORING_OP_CLOSE:
+		if (sqe) {
+			ret = io_close_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_close(req, force_nonblock);
+		break;
+	case IORING_OP_FILES_UPDATE:
+		if (sqe) {
+			ret = io_files_update_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_files_update(req, force_nonblock);
+		break;
+	case IORING_OP_STATX:
+		if (sqe) {
+			ret = io_statx_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_statx(req, force_nonblock);
+		break;
+	case IORING_OP_FADVISE:
+		if (sqe) {
+			ret = io_fadvise_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_fadvise(req, force_nonblock);
+		break;
+	case IORING_OP_MADVISE:
+		if (sqe) {
+			ret = io_madvise_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_madvise(req, force_nonblock);
+		break;
+	case IORING_OP_OPENAT2:
+		if (sqe) {
+			ret = io_openat2_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_openat2(req, force_nonblock);
+		break;
+	case IORING_OP_EPOLL_CTL:
+		if (sqe) {
+			ret = io_epoll_ctl_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_epoll_ctl(req, force_nonblock);
+		break;
+	case IORING_OP_SPLICE:
+		if (sqe) {
+			ret = io_splice_prep(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_splice(req, force_nonblock);
+		break;
+	case IORING_OP_PROVIDE_BUFFERS:
+		if (sqe) {
+			ret = io_provide_buffers_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_provide_buffers(req, force_nonblock);
+		break;
+	case IORING_OP_REMOVE_BUFFERS:
+		if (sqe) {
+			ret = io_remove_buffers_prep(req, sqe);
+			if (ret)
+				break;
+		}
+		ret = io_remove_buffers(req, force_nonblock);
+		break;
+	case IORING_OP_TEE:
+		if (sqe) {
+			ret = io_tee_prep(req, sqe);
+			if (ret < 0)
+				break;
+		}
+		ret = io_tee(req, force_nonblock);
 		break;
 	default:
 		ret = -EINVAL;
@@ -2103,252 +5362,238 @@ static int __io_submit_sqe(struct io_ring_ctx *ctx, struct io_kiocb *req,
 	if (ret)
 		return ret;
 
-	if (ctx->flags & IORING_SETUP_IOPOLL) {
+	/* If the op doesn't have a file, we're not polling for it */
+	if ((ctx->flags & IORING_SETUP_IOPOLL) && req->file) {
+		const bool in_async = io_wq_current_is_worker();
+
 		if (req->result == -EAGAIN)
 			return -EAGAIN;
 
 		/* workqueue context doesn't hold uring_lock, grab it now */
-		if (s->needs_lock)
+		if (in_async)
 			mutex_lock(&ctx->uring_lock);
+
 		io_iopoll_req_issued(req);
-		if (s->needs_lock)
+
+		if (in_async)
 			mutex_unlock(&ctx->uring_lock);
 	}
 
 	return 0;
 }
 
-static struct async_list *io_async_list_from_sqe(struct io_ring_ctx *ctx,
-						 const struct io_uring_sqe *sqe)
+static void io_wq_submit_work(struct io_wq_work **workptr)
 {
-	switch (sqe->opcode) {
-	case IORING_OP_READV:
-	case IORING_OP_READ_FIXED:
-		return &ctx->pending_async[READ];
-	case IORING_OP_WRITEV:
-	case IORING_OP_WRITE_FIXED:
-		return &ctx->pending_async[WRITE];
-	default:
-		return NULL;
+	struct io_wq_work *work = *workptr;
+	struct io_kiocb *req = container_of(work, struct io_kiocb, work);
+	int ret = 0;
+
+	/* if NO_CANCEL is set, we must still run the work */
+	if ((work->flags & (IO_WQ_WORK_CANCEL|IO_WQ_WORK_NO_CANCEL)) ==
+				IO_WQ_WORK_CANCEL) {
+		ret = -ECANCELED;
 	}
+
+	if (!ret) {
+		do {
+			ret = io_issue_sqe(req, NULL, false);
+			/*
+			 * We can get EAGAIN for polled IO even though we're
+			 * forcing a sync submission from here, since we can't
+			 * wait for request slots on the block side.
+			 */
+			if (ret != -EAGAIN)
+				break;
+			cond_resched();
+		} while (1);
+	}
+
+	if (ret) {
+		req_set_fail_links(req);
+		io_cqring_add_event(req, ret);
+		io_put_req(req);
+	}
+
+	io_steal_work(req, workptr);
 }
 
-static inline bool io_sqe_needs_user(const struct io_uring_sqe *sqe)
+static inline struct file *io_file_from_index(struct io_ring_ctx *ctx,
+					      int index)
 {
-	u8 opcode = READ_ONCE(sqe->opcode);
+	struct fixed_file_table *table;
 
-	return !(opcode == IORING_OP_READ_FIXED ||
-		 opcode == IORING_OP_WRITE_FIXED);
+	table = &ctx->file_data->table[index >> IORING_FILE_TABLE_SHIFT];
+	return table->files[index & IORING_FILE_TABLE_MASK];
 }
 
-static void io_sq_wq_submit_work(struct work_struct *work)
+static int io_file_get(struct io_submit_state *state, struct io_kiocb *req,
+			int fd, struct file **out_file, bool fixed)
 {
-	struct io_kiocb *req = container_of(work, struct io_kiocb, work);
 	struct io_ring_ctx *ctx = req->ctx;
-	struct mm_struct *cur_mm = NULL;
-	struct async_list *async_list;
-	LIST_HEAD(req_list);
-	mm_segment_t old_fs;
-	int ret;
-
-	async_list = io_async_list_from_sqe(ctx, req->submit.sqe);
-restart:
-	do {
-		struct sqe_submit *s = &req->submit;
-		const struct io_uring_sqe *sqe = s->sqe;
-		unsigned int flags = req->flags;
-
-		/* Ensure we clear previously set non-block flag */
-		req->rw.ki_flags &= ~IOCB_NOWAIT;
-
-		ret = 0;
-		if (io_sqe_needs_user(sqe) && !cur_mm) {
-			if (!mmget_not_zero(ctx->sqo_mm)) {
-				ret = -EFAULT;
-			} else {
-				cur_mm = ctx->sqo_mm;
-				use_mm(cur_mm);
-				old_fs = get_fs();
-				set_fs(USER_DS);
-			}
-		}
-
-		if (!ret) {
-			s->has_user = cur_mm != NULL;
-			s->needs_lock = true;
-			do {
-				ret = __io_submit_sqe(ctx, req, s, false);
-				/*
-				 * We can get EAGAIN for polled IO even though
-				 * we're forcing a sync submission from here,
-				 * since we can't wait for request slots on the
-				 * block side.
-				 */
-				if (ret != -EAGAIN)
-					break;
-				cond_resched();
-			} while (1);
-		}
+	struct file *file;
 
-		/* drop submission reference */
-		io_put_req(req);
+	if (fixed) {
+		if (unlikely(!ctx->file_data ||
+		    (unsigned) fd >= ctx->nr_user_files))
+			return -EBADF;
+		fd = array_index_nospec(fd, ctx->nr_user_files);
+		file = io_file_from_index(ctx, fd);
+		if (!file)
+			return -EBADF;
+		req->fixed_file_refs = ctx->file_data->cur_refs;
+		percpu_ref_get(req->fixed_file_refs);
+	} else {
+		trace_io_uring_file_get(ctx, fd);
+		file = __io_file_get(state, fd);
+		if (unlikely(!file))
+			return -EBADF;
+	}
 
-		if (ret) {
-			io_cqring_add_event(ctx, sqe->user_data, ret);
-			io_put_req(req);
-		}
+	*out_file = file;
+	return 0;
+}
 
-		/* async context always use a copy of the sqe */
-		kfree(sqe);
+static int io_req_set_file(struct io_submit_state *state, struct io_kiocb *req,
+			   int fd)
+{
+	bool fixed;
 
-		/* req from defer and link list needn't decrease async cnt */
-		if (flags & (REQ_F_IO_DRAINED | REQ_F_LINK_DONE))
-			goto out;
+	fixed = (req->flags & REQ_F_FIXED_FILE) != 0;
+	if (unlikely(!fixed && io_async_submit(req->ctx)))
+		return -EBADF;
 
-		if (!async_list)
-			break;
-		if (!list_empty(&req_list)) {
-			req = list_first_entry(&req_list, struct io_kiocb,
-						list);
-			list_del(&req->list);
-			continue;
-		}
-		if (list_empty(&async_list->list))
-			break;
+	return io_file_get(state, req, fd, &req->file, fixed);
+}
 
-		req = NULL;
-		spin_lock(&async_list->lock);
-		if (list_empty(&async_list->list)) {
-			spin_unlock(&async_list->lock);
-			break;
-		}
-		list_splice_init(&async_list->list, &req_list);
-		spin_unlock(&async_list->lock);
+static int io_grab_files(struct io_kiocb *req)
+{
+	int ret = -EBADF;
+	struct io_ring_ctx *ctx = req->ctx;
 
-		req = list_first_entry(&req_list, struct io_kiocb, list);
-		list_del(&req->list);
-	} while (req);
+	if (req->work.files || (req->flags & REQ_F_NO_FILE_TABLE))
+		return 0;
+	if (!ctx->ring_file)
+		return -EBADF;
 
+	rcu_read_lock();
+	spin_lock_irq(&ctx->inflight_lock);
 	/*
-	 * Rare case of racing with a submitter. If we find the count has
-	 * dropped to zero AND we have pending work items, then restart
-	 * the processing. This is a tiny race window.
+	 * We use the f_ops->flush() handler to ensure that we can flush
+	 * out work accessing these files if the fd is closed. Check if
+	 * the fd has changed since we started down this path, and disallow
+	 * this operation if it has.
 	 */
-	if (async_list) {
-		ret = atomic_dec_return(&async_list->cnt);
-		while (!ret && !list_empty(&async_list->list)) {
-			spin_lock(&async_list->lock);
-			atomic_inc(&async_list->cnt);
-			list_splice_init(&async_list->list, &req_list);
-			spin_unlock(&async_list->lock);
-
-			if (!list_empty(&req_list)) {
-				req = list_first_entry(&req_list,
-							struct io_kiocb, list);
-				list_del(&req->list);
-				goto restart;
-			}
-			ret = atomic_dec_return(&async_list->cnt);
-		}
+	if (fcheck(ctx->ring_fd) == ctx->ring_file) {
+		list_add(&req->inflight_entry, &ctx->inflight_list);
+		req->flags |= REQ_F_INFLIGHT;
+		req->work.files = current->files;
+		ret = 0;
 	}
+	spin_unlock_irq(&ctx->inflight_lock);
+	rcu_read_unlock();
 
-out:
-	if (cur_mm) {
-		set_fs(old_fs);
-		unuse_mm(cur_mm);
-		mmput(cur_mm);
-	}
+	return ret;
 }
 
-/*
- * See if we can piggy back onto previously submitted work, that is still
- * running. We currently only allow this if the new request is sequential
- * to the previous one we punted.
- */
-static bool io_add_to_prev_work(struct async_list *list, struct io_kiocb *req)
+static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer)
 {
-	bool ret;
+	struct io_timeout_data *data = container_of(timer,
+						struct io_timeout_data, timer);
+	struct io_kiocb *req = data->req;
+	struct io_ring_ctx *ctx = req->ctx;
+	struct io_kiocb *prev = NULL;
+	unsigned long flags;
 
-	if (!list)
-		return false;
-	if (!(req->flags & REQ_F_SEQ_PREV))
-		return false;
-	if (!atomic_read(&list->cnt))
-		return false;
+	spin_lock_irqsave(&ctx->completion_lock, flags);
 
-	ret = true;
-	spin_lock(&list->lock);
-	list_add_tail(&req->list, &list->list);
 	/*
-	 * Ensure we see a simultaneous modification from io_sq_wq_submit_work()
+	 * We don't expect the list to be empty, that will only happen if we
+	 * race with the completion of the linked work.
 	 */
-	smp_mb();
-	if (!atomic_read(&list->cnt)) {
-		list_del_init(&req->list);
-		ret = false;
+	if (!list_empty(&req->link_list)) {
+		prev = list_entry(req->link_list.prev, struct io_kiocb,
+				  link_list);
+		if (refcount_inc_not_zero(&prev->refs)) {
+			list_del_init(&req->link_list);
+			prev->flags &= ~REQ_F_LINK_TIMEOUT;
+		} else
+			prev = NULL;
 	}
-	spin_unlock(&list->lock);
-	return ret;
-}
 
-static bool io_op_needs_file(const struct io_uring_sqe *sqe)
-{
-	int op = READ_ONCE(sqe->opcode);
+	spin_unlock_irqrestore(&ctx->completion_lock, flags);
 
-	switch (op) {
-	case IORING_OP_NOP:
-	case IORING_OP_POLL_REMOVE:
-	case IORING_OP_TIMEOUT:
-		return false;
-	default:
-		return true;
+	if (prev) {
+		req_set_fail_links(prev);
+		io_async_find_and_cancel(ctx, req, prev->user_data, -ETIME);
+		io_put_req(prev);
+	} else {
+		io_cqring_add_event(req, -ETIME);
+		io_put_req(req);
 	}
+	return HRTIMER_NORESTART;
 }
 
-static int io_req_set_file(struct io_ring_ctx *ctx, const struct sqe_submit *s,
-			   struct io_submit_state *state, struct io_kiocb *req)
+static void io_queue_linked_timeout(struct io_kiocb *req)
 {
-	unsigned flags;
-	int fd;
-
-	flags = READ_ONCE(s->sqe->flags);
-	fd = READ_ONCE(s->sqe->fd);
+	struct io_ring_ctx *ctx = req->ctx;
 
-	if (flags & IOSQE_IO_DRAIN)
-		req->flags |= REQ_F_IO_DRAIN;
 	/*
-	 * All io need record the previous position, if LINK vs DARIN,
-	 * it can be used to mark the position of the first IO in the
-	 * link list.
+	 * If the list is now empty, then our linked request finished before
+	 * we got a chance to setup the timer
 	 */
-	req->sequence = s->sequence;
-
-	if (!io_op_needs_file(s->sqe))
-		return 0;
+	spin_lock_irq(&ctx->completion_lock);
+	if (!list_empty(&req->link_list)) {
+		struct io_timeout_data *data = &req->io->timeout;
 
-	if (flags & IOSQE_FIXED_FILE) {
-		if (unlikely(!ctx->user_files ||
-		    (unsigned) fd >= ctx->nr_user_files))
-			return -EBADF;
-		req->file = ctx->user_files[fd];
-		req->flags |= REQ_F_FIXED_FILE;
-	} else {
-		if (s->needs_fixed_file)
-			return -EBADF;
-		req->file = io_file_get(state, fd);
-		if (unlikely(!req->file))
-			return -EBADF;
+		data->timer.function = io_link_timeout_fn;
+		hrtimer_start(&data->timer, timespec64_to_ktime(data->ts),
+				data->mode);
 	}
+	spin_unlock_irq(&ctx->completion_lock);
 
-	return 0;
+	/* drop submission reference */
+	io_put_req(req);
+}
+
+static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req)
+{
+	struct io_kiocb *nxt;
+
+	if (!(req->flags & REQ_F_LINK_HEAD))
+		return NULL;
+	/* for polled retry, if flag is set, we already went through here */
+	if (req->flags & REQ_F_POLLED)
+		return NULL;
+
+	nxt = list_first_entry_or_null(&req->link_list, struct io_kiocb,
+					link_list);
+	if (!nxt || nxt->opcode != IORING_OP_LINK_TIMEOUT)
+		return NULL;
+
+	req->flags |= REQ_F_LINK_TIMEOUT;
+	return nxt;
 }
 
-static int __io_queue_sqe(struct io_ring_ctx *ctx, struct io_kiocb *req,
-			struct sqe_submit *s)
+static void __io_queue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
+	struct io_kiocb *linked_timeout;
+	struct io_kiocb *nxt;
+	const struct cred *old_creds = NULL;
 	int ret;
 
-	ret = __io_submit_sqe(ctx, req, s, true);
+again:
+	linked_timeout = io_prep_linked_timeout(req);
+
+	if (req->work.creds && req->work.creds != current_cred()) {
+		if (old_creds)
+			revert_creds(old_creds);
+		if (old_creds == req->work.creds)
+			old_creds = NULL; /* restored original creds */
+		else
+			old_creds = override_creds(req->work.creds);
+	}
+
+	ret = io_issue_sqe(req, sqe, true);
 
 	/*
 	 * We async punt it if the file wasn't marked NOWAIT, or if the file
@@ -2356,135 +5601,104 @@ static int __io_queue_sqe(struct io_ring_ctx *ctx, struct io_kiocb *req,
 	 */
 	if (ret == -EAGAIN && (!(req->flags & REQ_F_NOWAIT) ||
 	    (req->flags & REQ_F_MUST_PUNT))) {
-		struct io_uring_sqe *sqe_copy;
-
-		sqe_copy = kmemdup(s->sqe, sizeof(*sqe_copy), GFP_KERNEL);
-		if (sqe_copy) {
-			struct async_list *list;
-
-			s->sqe = sqe_copy;
-			memcpy(&req->submit, s, sizeof(*s));
-			list = io_async_list_from_sqe(ctx, s->sqe);
-			if (!io_add_to_prev_work(list, req)) {
-				if (list)
-					atomic_inc(&list->cnt);
-				INIT_WORK(&req->work, io_sq_wq_submit_work);
-				io_queue_async_work(ctx, req);
-			}
-
-			/*
-			 * Queued up for async execution, worker will release
-			 * submit reference when the iocb is actually submitted.
-			 */
-			return 0;
+		if (io_arm_poll_handler(req)) {
+			if (linked_timeout)
+				io_queue_linked_timeout(linked_timeout);
+			goto exit;
 		}
+punt:
+		if (io_op_defs[req->opcode].file_table) {
+			ret = io_grab_files(req);
+			if (ret)
+				goto err;
+		}
+
+		/*
+		 * Queued up for async execution, worker will release
+		 * submit reference when the iocb is actually submitted.
+		 */
+		io_queue_async_work(req);
+		goto exit;
 	}
 
+err:
+	nxt = NULL;
 	/* drop submission reference */
-	io_put_req(req);
+	io_put_req_find_next(req, &nxt);
+
+	if (linked_timeout) {
+		if (!ret)
+			io_queue_linked_timeout(linked_timeout);
+		else
+			io_put_req(linked_timeout);
+	}
 
 	/* and drop final reference, if we failed */
 	if (ret) {
-		io_cqring_add_event(ctx, req->user_data, ret);
-		if (req->flags & REQ_F_LINK)
-			req->flags |= REQ_F_FAIL_LINK;
+		io_cqring_add_event(req, ret);
+		req_set_fail_links(req);
 		io_put_req(req);
 	}
+	if (nxt) {
+		req = nxt;
 
-	return ret;
-}
-
-static int io_queue_sqe(struct io_ring_ctx *ctx, struct io_kiocb *req,
-			struct sqe_submit *s)
-{
-	int ret;
-
-	ret = io_req_defer(ctx, req, s->sqe);
-	if (ret) {
-		if (ret != -EIOCBQUEUED) {
-			io_free_req(req);
-			io_cqring_add_event(ctx, s->sqe->user_data, ret);
-		}
-		return 0;
+		if (req->flags & REQ_F_FORCE_ASYNC)
+			goto punt;
+		goto again;
 	}
-
-	return __io_queue_sqe(ctx, req, s);
+exit:
+	if (old_creds)
+		revert_creds(old_creds);
 }
 
-static int io_queue_link_head(struct io_ring_ctx *ctx, struct io_kiocb *req,
-			      struct sqe_submit *s, struct io_kiocb *shadow)
+static void io_queue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	int ret;
-	int need_submit = false;
 
-	if (!shadow)
-		return io_queue_sqe(ctx, req, s);
-
-	/*
-	 * Mark the first IO in link list as DRAIN, let all the following
-	 * IOs enter the defer list. all IO needs to be completed before link
-	 * list.
-	 */
-	req->flags |= REQ_F_IO_DRAIN;
-	ret = io_req_defer(ctx, req, s->sqe);
+	ret = io_req_defer(req, sqe);
 	if (ret) {
 		if (ret != -EIOCBQUEUED) {
-			io_free_req(req);
-			__io_free_req(shadow);
-			io_cqring_add_event(ctx, s->sqe->user_data, ret);
-			return 0;
+fail_req:
+			io_cqring_add_event(req, ret);
+			req_set_fail_links(req);
+			io_double_put_req(req);
 		}
-	} else {
+	} else if (req->flags & REQ_F_FORCE_ASYNC) {
+		if (!req->io) {
+			ret = -EAGAIN;
+			if (io_alloc_async_ctx(req))
+				goto fail_req;
+			ret = io_req_defer_prep(req, sqe);
+			if (unlikely(ret < 0))
+				goto fail_req;
+		}
+
 		/*
-		 * If ret == 0 means that all IOs in front of link io are
-		 * running done. let's queue link head.
+		 * Never try inline submit of IOSQE_ASYNC is set, go straight
+		 * to async execution.
 		 */
-		need_submit = true;
+		req->work.flags |= IO_WQ_WORK_CONCURRENT;
+		io_queue_async_work(req);
+	} else {
+		__io_queue_sqe(req, sqe);
 	}
-
-	/* Insert shadow req to defer_list, blocking next IOs */
-	spin_lock_irq(&ctx->completion_lock);
-	list_add_tail(&shadow->list, &ctx->defer_list);
-	spin_unlock_irq(&ctx->completion_lock);
-
-	if (need_submit)
-		return __io_queue_sqe(ctx, req, s);
-
-	return 0;
 }
 
-#define SQE_VALID_FLAGS	(IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK)
+static inline void io_queue_link_head(struct io_kiocb *req)
+{
+	if (unlikely(req->flags & REQ_F_FAIL_LINK)) {
+		io_cqring_add_event(req, -ECANCELED);
+		io_double_put_req(req);
+	} else
+		io_queue_sqe(req, NULL);
+}
 
-static void io_submit_sqe(struct io_ring_ctx *ctx, struct sqe_submit *s,
-			  struct io_submit_state *state, struct io_kiocb **link)
+static int io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			 struct io_kiocb **link)
 {
-	struct io_uring_sqe *sqe_copy;
-	struct io_kiocb *req;
+	struct io_ring_ctx *ctx = req->ctx;
 	int ret;
 
-	/* enforce forwards compatibility on users */
-	if (unlikely(s->sqe->flags & ~SQE_VALID_FLAGS)) {
-		ret = -EINVAL;
-		goto err;
-	}
-
-	req = io_get_req(ctx, state);
-	if (unlikely(!req)) {
-		ret = -EAGAIN;
-		goto err;
-	}
-
-	ret = io_req_set_file(ctx, s, state, req);
-	if (unlikely(ret)) {
-err_req:
-		io_free_req(req);
-err:
-		io_cqring_add_event(ctx, s->sqe->user_data, ret);
-		return;
-	}
-
-	req->user_data = s->sqe->user_data;
-
 	/*
 	 * If we already have a head request, queue this one for async
 	 * submittal once the head completes. If we don't have a head but
@@ -2493,26 +5707,58 @@ err:
 	 * conditions are true (normal request), then just queue it.
 	 */
 	if (*link) {
-		struct io_kiocb *prev = *link;
+		struct io_kiocb *head = *link;
 
-		sqe_copy = kmemdup(s->sqe, sizeof(*sqe_copy), GFP_KERNEL);
-		if (!sqe_copy) {
-			ret = -EAGAIN;
-			goto err_req;
+		/*
+		 * Taking sequential execution of a link, draining both sides
+		 * of the link also fullfils IOSQE_IO_DRAIN semantics for all
+		 * requests in the link. So, it drains the head and the
+		 * next after the link request. The last one is done via
+		 * drain_next flag to persist the effect across calls.
+		 */
+		if (req->flags & REQ_F_IO_DRAIN) {
+			head->flags |= REQ_F_IO_DRAIN;
+			ctx->drain_next = 1;
 		}
+		if (io_alloc_async_ctx(req))
+			return -EAGAIN;
 
-		s->sqe = sqe_copy;
-		memcpy(&req->submit, s, sizeof(*s));
-		list_add_tail(&req->list, &prev->link_list);
-	} else if (s->sqe->flags & IOSQE_IO_LINK) {
-		req->flags |= REQ_F_LINK;
+		ret = io_req_defer_prep(req, sqe);
+		if (ret) {
+			/* fail even hard links since we don't submit */
+			head->flags |= REQ_F_FAIL_LINK;
+			return ret;
+		}
+		trace_io_uring_link(ctx, req, head);
+		list_add_tail(&req->link_list, &head->link_list);
 
-		memcpy(&req->submit, s, sizeof(*s));
-		INIT_LIST_HEAD(&req->link_list);
-		*link = req;
+		/* last request of a link, enqueue the link */
+		if (!(req->flags & (REQ_F_LINK | REQ_F_HARDLINK))) {
+			io_queue_link_head(head);
+			*link = NULL;
+		}
 	} else {
-		io_queue_sqe(ctx, req, s);
+		if (unlikely(ctx->drain_next)) {
+			req->flags |= REQ_F_IO_DRAIN;
+			ctx->drain_next = 0;
+		}
+		if (req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) {
+			req->flags |= REQ_F_LINK_HEAD;
+			INIT_LIST_HEAD(&req->link_list);
+
+			if (io_alloc_async_ctx(req))
+				return -EAGAIN;
+
+			ret = io_req_defer_prep(req, sqe);
+			if (ret)
+				req->flags |= REQ_F_FAIL_LINK;
+			*link = req;
+		} else {
+			io_queue_sqe(req, sqe);
+		}
 	}
+
+	return 0;
 }
 
 /*
@@ -2521,17 +5767,16 @@ err:
 static void io_submit_state_end(struct io_submit_state *state)
 {
 	blk_finish_plug(&state->plug);
-	io_file_put(state);
+	io_state_file_put(state);
 	if (state->free_reqs)
-		kmem_cache_free_bulk(req_cachep, state->free_reqs,
-					&state->reqs[state->cur_req]);
+		kmem_cache_free_bulk(req_cachep, state->free_reqs, state->reqs);
 }
 
 /*
  * Start submission side cache.
  */
 static void io_submit_state_start(struct io_submit_state *state,
-				  struct io_ring_ctx *ctx, unsigned max_ios)
+				  unsigned int max_ios)
 {
 	blk_start_plug(&state->plug);
 	state->free_reqs = 0;
@@ -2543,27 +5788,24 @@ static void io_commit_sqring(struct io_ring_ctx *ctx)
 {
 	struct io_rings *rings = ctx->rings;
 
-	if (ctx->cached_sq_head != READ_ONCE(rings->sq.head)) {
-		/*
-		 * Ensure any loads from the SQEs are done at this point,
-		 * since once we write the new head, the application could
-		 * write new data to them.
-		 */
-		smp_store_release(&rings->sq.head, ctx->cached_sq_head);
-	}
+	/*
+	 * Ensure any loads from the SQEs are done at this point,
+	 * since once we write the new head, the application could
+	 * write new data to them.
+	 */
+	smp_store_release(&rings->sq.head, ctx->cached_sq_head);
 }
 
 /*
- * Fetch an sqe, if one is available. Note that s->sqe will point to memory
+ * Fetch an sqe, if one is available. Note that sqe_ptr will point to memory
  * that is mapped by userspace. This means that care needs to be taken to
  * ensure that reads are stable, as we cannot rely on userspace always
  * being a good citizen. If members of the sqe are validated and then later
  * used, it's important that those reads are done through READ_ONCE() to
  * prevent a re-load down the line.
  */
-static bool io_get_sqring(struct io_ring_ctx *ctx, struct sqe_submit *s)
+static const struct io_uring_sqe *io_get_sqe(struct io_ring_ctx *ctx)
 {
-	struct io_rings *rings = ctx->rings;
 	u32 *sq_array = ctx->sq_array;
 	unsigned head;
 
@@ -2575,184 +5817,275 @@ static bool io_get_sqring(struct io_ring_ctx *ctx, struct sqe_submit *s)
 	 * 2) allows the kernel side to track the head on its own, even
 	 *    though the application is the one updating it.
 	 */
-	head = ctx->cached_sq_head;
-	/* make sure SQ entry isn't read before tail */
-	if (head == smp_load_acquire(&rings->sq.tail))
-		return false;
-
-	head = READ_ONCE(sq_array[head & ctx->sq_mask]);
-	if (head < ctx->sq_entries) {
-		s->index = head;
-		s->sqe = &ctx->sq_sqes[head];
-		s->sequence = ctx->cached_sq_head;
-		ctx->cached_sq_head++;
-		return true;
-	}
+	head = READ_ONCE(sq_array[ctx->cached_sq_head & ctx->sq_mask]);
+	if (likely(head < ctx->sq_entries))
+		return &ctx->sq_sqes[head];
 
 	/* drop invalid entries */
-	ctx->cached_sq_head++;
 	ctx->cached_sq_dropped++;
-	WRITE_ONCE(rings->sq_dropped, ctx->cached_sq_dropped);
-	return false;
+	WRITE_ONCE(ctx->rings->sq_dropped, ctx->cached_sq_dropped);
+	return NULL;
+}
+
+static inline void io_consume_sqe(struct io_ring_ctx *ctx)
+{
+	ctx->cached_sq_head++;
+}
+
+#define SQE_VALID_FLAGS	(IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK|	\
+				IOSQE_IO_HARDLINK | IOSQE_ASYNC | \
+				IOSQE_BUFFER_SELECT)
+
+static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
+		       const struct io_uring_sqe *sqe,
+		       struct io_submit_state *state)
+{
+	unsigned int sqe_flags;
+	int id;
+
+	/*
+	 * All io need record the previous position, if LINK vs DARIN,
+	 * it can be used to mark the position of the first IO in the
+	 * link list.
+	 */
+	req->sequence = ctx->cached_sq_head - ctx->cached_sq_dropped;
+	req->opcode = READ_ONCE(sqe->opcode);
+	req->user_data = READ_ONCE(sqe->user_data);
+	req->io = NULL;
+	req->file = NULL;
+	req->ctx = ctx;
+	req->flags = 0;
+	/* one is dropped after submission, the other at completion */
+	refcount_set(&req->refs, 2);
+	req->task = NULL;
+	req->result = 0;
+	INIT_IO_WORK(&req->work, io_wq_submit_work);
+
+	if (unlikely(req->opcode >= IORING_OP_LAST))
+		return -EINVAL;
+
+	if (io_op_defs[req->opcode].needs_mm && !current->mm) {
+		if (unlikely(!mmget_not_zero(ctx->sqo_mm)))
+			return -EFAULT;
+		use_mm(ctx->sqo_mm);
+	}
+
+	sqe_flags = READ_ONCE(sqe->flags);
+	/* enforce forwards compatibility on users */
+	if (unlikely(sqe_flags & ~SQE_VALID_FLAGS))
+		return -EINVAL;
+
+	if ((sqe_flags & IOSQE_BUFFER_SELECT) &&
+	    !io_op_defs[req->opcode].buffer_select)
+		return -EOPNOTSUPP;
+
+	id = READ_ONCE(sqe->personality);
+	if (id) {
+		req->work.creds = idr_find(&ctx->personality_idr, id);
+		if (unlikely(!req->work.creds))
+			return -EINVAL;
+		get_cred(req->work.creds);
+	}
+
+	/* same numerical values with corresponding REQ_F_*, safe to copy */
+	req->flags |= sqe_flags;
+
+	if (!io_op_defs[req->opcode].needs_file)
+		return 0;
+
+	return io_req_set_file(state, req, READ_ONCE(sqe->fd));
 }
 
 static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
-			  bool has_user, bool mm_fault)
+			  struct file *ring_file, int ring_fd)
 {
 	struct io_submit_state state, *statep = NULL;
 	struct io_kiocb *link = NULL;
-	struct io_kiocb *shadow_req = NULL;
-	bool prev_was_link = false;
 	int i, submitted = 0;
 
+	/* if we have a backlog and couldn't flush it all, return BUSY */
+	if (test_bit(0, &ctx->sq_check_overflow)) {
+		if (!list_empty(&ctx->cq_overflow_list) &&
+		    !io_cqring_overflow_flush(ctx, false))
+			return -EBUSY;
+	}
+
+	/* make sure SQ entry isn't read before tail */
+	nr = min3(nr, ctx->sq_entries, io_sqring_entries(ctx));
+
+	if (!percpu_ref_tryget_many(&ctx->refs, nr))
+		return -EAGAIN;
+
 	if (nr > IO_PLUG_THRESHOLD) {
-		io_submit_state_start(&state, ctx, nr);
+		io_submit_state_start(&state, nr);
 		statep = &state;
 	}
 
+	ctx->ring_fd = ring_fd;
+	ctx->ring_file = ring_file;
+
 	for (i = 0; i < nr; i++) {
-		struct sqe_submit s;
+		const struct io_uring_sqe *sqe;
+		struct io_kiocb *req;
+		int err;
 
-		if (!io_get_sqring(ctx, &s))
+		sqe = io_get_sqe(ctx);
+		if (unlikely(!sqe)) {
+			io_consume_sqe(ctx);
+			break;
+		}
+		req = io_alloc_req(ctx, statep);
+		if (unlikely(!req)) {
+			if (!submitted)
+				submitted = -EAGAIN;
 			break;
-
-		/*
-		 * If previous wasn't linked and we have a linked command,
-		 * that's the end of the chain. Submit the previous link.
-		 */
-		if (!prev_was_link && link) {
-			io_queue_link_head(ctx, link, &link->submit, shadow_req);
-			link = NULL;
-			shadow_req = NULL;
-		}
-		prev_was_link = (s.sqe->flags & IOSQE_IO_LINK) != 0;
-
-		if (link && (s.sqe->flags & IOSQE_IO_DRAIN)) {
-			if (!shadow_req) {
-				shadow_req = io_get_req(ctx, NULL);
-				if (unlikely(!shadow_req))
-					goto out;
-				shadow_req->flags |= (REQ_F_IO_DRAIN | REQ_F_SHADOW_DRAIN);
-				refcount_dec(&shadow_req->refs);
-			}
-			shadow_req->sequence = s.sequence;
 		}
 
-out:
-		if (unlikely(mm_fault)) {
-			io_cqring_add_event(ctx, s.sqe->user_data,
-						-EFAULT);
-		} else {
-			s.has_user = has_user;
-			s.needs_lock = true;
-			s.needs_fixed_file = true;
-			io_submit_sqe(ctx, &s, statep, &link);
-			submitted++;
+		err = io_init_req(ctx, req, sqe, statep);
+		io_consume_sqe(ctx);
+		/* will complete beyond this point, count as submitted */
+		submitted++;
+
+		if (unlikely(err)) {
+fail_req:
+			io_cqring_add_event(req, err);
+			io_double_put_req(req);
+			break;
 		}
+
+		trace_io_uring_submit_sqe(ctx, req->opcode, req->user_data,
+						true, io_async_submit(ctx));
+		err = io_submit_sqe(req, sqe, &link);
+		if (err)
+			goto fail_req;
 	}
 
+	if (unlikely(submitted != nr)) {
+		int ref_used = (submitted == -EAGAIN) ? 0 : submitted;
+
+		percpu_ref_put_many(&ctx->refs, nr - ref_used);
+	}
 	if (link)
-		io_queue_link_head(ctx, link, &link->submit, shadow_req);
+		io_queue_link_head(link);
 	if (statep)
 		io_submit_state_end(&state);
 
+	 /* Commit SQ ring head once we've consumed and submitted all SQEs */
+	io_commit_sqring(ctx);
+
 	return submitted;
 }
 
+static inline void io_sq_thread_drop_mm(struct io_ring_ctx *ctx)
+{
+	struct mm_struct *mm = current->mm;
+
+	if (mm) {
+		unuse_mm(mm);
+		mmput(mm);
+	}
+}
+
 static int io_sq_thread(void *data)
 {
 	struct io_ring_ctx *ctx = data;
-	struct mm_struct *cur_mm = NULL;
+	const struct cred *old_cred;
 	mm_segment_t old_fs;
 	DEFINE_WAIT(wait);
-	unsigned inflight;
 	unsigned long timeout;
+	int ret = 0;
 
-	complete(&ctx->sqo_thread_started);
+	complete(&ctx->sq_thread_comp);
 
 	old_fs = get_fs();
 	set_fs(USER_DS);
+	old_cred = override_creds(ctx->creds);
 
-	timeout = inflight = 0;
+	timeout = jiffies + ctx->sq_thread_idle;
 	while (!kthread_should_park()) {
-		bool mm_fault = false;
 		unsigned int to_submit;
 
-		if (inflight) {
+		if (!list_empty(&ctx->poll_list)) {
 			unsigned nr_events = 0;
 
-			if (ctx->flags & IORING_SETUP_IOPOLL) {
-				/*
-				 * inflight is the count of the maximum possible
-				 * entries we submitted, but it can be smaller
-				 * if we dropped some of them. If we don't have
-				 * poll entries available, then we know that we
-				 * have nothing left to poll for. Reset the
-				 * inflight count to zero in that case.
-				 */
-				mutex_lock(&ctx->uring_lock);
-				if (!list_empty(&ctx->poll_list))
-					__io_iopoll_check(ctx, &nr_events, 0);
-				else
-					inflight = 0;
-				mutex_unlock(&ctx->uring_lock);
-			} else {
-				/*
-				 * Normal IO, just pretend everything completed.
-				 * We don't have to poll completions for that.
-				 */
-				nr_events = inflight;
-			}
-
-			inflight -= nr_events;
-			if (!inflight)
+			mutex_lock(&ctx->uring_lock);
+			if (!list_empty(&ctx->poll_list))
+				io_iopoll_getevents(ctx, &nr_events, 0);
+			else
 				timeout = jiffies + ctx->sq_thread_idle;
+			mutex_unlock(&ctx->uring_lock);
 		}
 
 		to_submit = io_sqring_entries(ctx);
-		if (!to_submit) {
-			/*
-			 * We're polling. If we're within the defined idle
-			 * period, then let us spin without work before going
-			 * to sleep.
-			 */
-			if (inflight || !time_after(jiffies, timeout)) {
-				cond_resched();
-				continue;
-			}
 
+		/*
+		 * If submit got -EBUSY, flag us as needing the application
+		 * to enter the kernel to reap and flush events.
+		 */
+		if (!to_submit || ret == -EBUSY) {
 			/*
 			 * Drop cur_mm before scheduling, we can't hold it for
 			 * long periods (or over schedule()). Do this before
 			 * adding ourselves to the waitqueue, as the unuse/drop
 			 * may sleep.
 			 */
-			if (cur_mm) {
-				unuse_mm(cur_mm);
-				mmput(cur_mm);
-				cur_mm = NULL;
+			io_sq_thread_drop_mm(ctx);
+
+			/*
+			 * We're polling. If we're within the defined idle
+			 * period, then let us spin without work before going
+			 * to sleep. The exception is if we got EBUSY doing
+			 * more IO, we should wait for the application to
+			 * reap events and wake us up.
+			 */
+			if (!list_empty(&ctx->poll_list) ||
+			    (!time_after(jiffies, timeout) && ret != -EBUSY &&
+			    !percpu_ref_is_dying(&ctx->refs))) {
+				if (current->task_works)
+					task_work_run();
+				cond_resched();
+				continue;
 			}
 
 			prepare_to_wait(&ctx->sqo_wait, &wait,
 						TASK_INTERRUPTIBLE);
 
+			/*
+			 * While doing polled IO, before going to sleep, we need
+			 * to check if there are new reqs added to poll_list, it
+			 * is because reqs may have been punted to io worker and
+			 * will be added to poll_list later, hence check the
+			 * poll_list again.
+			 */
+			if ((ctx->flags & IORING_SETUP_IOPOLL) &&
+			    !list_empty_careful(&ctx->poll_list)) {
+				finish_wait(&ctx->sqo_wait, &wait);
+				continue;
+			}
+
 			/* Tell userspace we may need a wakeup call */
 			ctx->rings->sq_flags |= IORING_SQ_NEED_WAKEUP;
 			/* make sure to read SQ tail after writing flags */
 			smp_mb();
 
 			to_submit = io_sqring_entries(ctx);
-			if (!to_submit) {
+			if (!to_submit || ret == -EBUSY) {
 				if (kthread_should_park()) {
 					finish_wait(&ctx->sqo_wait, &wait);
 					break;
 				}
+				if (current->task_works) {
+					task_work_run();
+					finish_wait(&ctx->sqo_wait, &wait);
+					continue;
+				}
 				if (signal_pending(current))
 					flush_signals(current);
 				schedule();
 				finish_wait(&ctx->sqo_wait, &wait);
 
 				ctx->rings->sq_flags &= ~IORING_SQ_NEED_WAKEUP;
+				ret = 0;
 				continue;
 			}
 			finish_wait(&ctx->sqo_wait, &wait);
@@ -2760,93 +6093,25 @@ static int io_sq_thread(void *data)
 			ctx->rings->sq_flags &= ~IORING_SQ_NEED_WAKEUP;
 		}
 
-		/* Unless all new commands are FIXED regions, grab mm */
-		if (!cur_mm) {
-			mm_fault = !mmget_not_zero(ctx->sqo_mm);
-			if (!mm_fault) {
-				use_mm(ctx->sqo_mm);
-				cur_mm = ctx->sqo_mm;
-			}
-		}
-
-		to_submit = min(to_submit, ctx->sq_entries);
-		inflight += io_submit_sqes(ctx, to_submit, cur_mm != NULL,
-					   mm_fault);
-
-		/* Commit SQ ring head once we've consumed all SQEs */
-		io_commit_sqring(ctx);
+		mutex_lock(&ctx->uring_lock);
+		if (likely(!percpu_ref_is_dying(&ctx->refs)))
+			ret = io_submit_sqes(ctx, to_submit, NULL, -1);
+		mutex_unlock(&ctx->uring_lock);
+		timeout = jiffies + ctx->sq_thread_idle;
 	}
 
+	if (current->task_works)
+		task_work_run();
+
 	set_fs(old_fs);
-	if (cur_mm) {
-		unuse_mm(cur_mm);
-		mmput(cur_mm);
-	}
+	io_sq_thread_drop_mm(ctx);
+	revert_creds(old_cred);
 
 	kthread_parkme();
 
 	return 0;
 }
 
-static int io_ring_submit(struct io_ring_ctx *ctx, unsigned int to_submit)
-{
-	struct io_submit_state state, *statep = NULL;
-	struct io_kiocb *link = NULL;
-	struct io_kiocb *shadow_req = NULL;
-	bool prev_was_link = false;
-	int i, submit = 0;
-
-	if (to_submit > IO_PLUG_THRESHOLD) {
-		io_submit_state_start(&state, ctx, to_submit);
-		statep = &state;
-	}
-
-	for (i = 0; i < to_submit; i++) {
-		struct sqe_submit s;
-
-		if (!io_get_sqring(ctx, &s))
-			break;
-
-		/*
-		 * If previous wasn't linked and we have a linked command,
-		 * that's the end of the chain. Submit the previous link.
-		 */
-		if (!prev_was_link && link) {
-			io_queue_link_head(ctx, link, &link->submit, shadow_req);
-			link = NULL;
-			shadow_req = NULL;
-		}
-		prev_was_link = (s.sqe->flags & IOSQE_IO_LINK) != 0;
-
-		if (link && (s.sqe->flags & IOSQE_IO_DRAIN)) {
-			if (!shadow_req) {
-				shadow_req = io_get_req(ctx, NULL);
-				if (unlikely(!shadow_req))
-					goto out;
-				shadow_req->flags |= (REQ_F_IO_DRAIN | REQ_F_SHADOW_DRAIN);
-				refcount_dec(&shadow_req->refs);
-			}
-			shadow_req->sequence = s.sequence;
-		}
-
-out:
-		s.has_user = true;
-		s.needs_lock = false;
-		s.needs_fixed_file = false;
-		submit++;
-		io_submit_sqe(ctx, &s, statep, &link);
-	}
-
-	if (link)
-		io_queue_link_head(ctx, link, &link->submit, shadow_req);
-	if (statep)
-		io_submit_state_end(statep);
-
-	io_commit_sqring(ctx);
-
-	return submit;
-}
-
 struct io_wait_queue {
 	struct wait_queue_entry wq;
 	struct io_ring_ctx *ctx;
@@ -2854,16 +6119,16 @@ struct io_wait_queue {
 	unsigned nr_timeouts;
 };
 
-static inline bool io_should_wake(struct io_wait_queue *iowq)
+static inline bool io_should_wake(struct io_wait_queue *iowq, bool noflush)
 {
 	struct io_ring_ctx *ctx = iowq->ctx;
 
 	/*
-	 * Wake up if we have enough events, or if a timeout occured since we
+	 * Wake up if we have enough events, or if a timeout occurred since we
 	 * started waiting. For timeouts, we always want to return to userspace,
 	 * regardless of event count.
 	 */
-	return io_cqring_events(ctx->rings) >= iowq->to_wait ||
+	return io_cqring_events(ctx, noflush) >= iowq->to_wait ||
 			atomic_read(&ctx->cq_timeouts) != iowq->nr_timeouts;
 }
 
@@ -2873,7 +6138,8 @@ static int io_wake_function(struct wait_queue_entry *curr, unsigned int mode,
 	struct io_wait_queue *iowq = container_of(curr, struct io_wait_queue,
 							wq);
 
-	if (!io_should_wake(iowq))
+	/* use noflush == true, as we can't safely rely on locking context */
+	if (!io_should_wake(iowq, true))
 		return -1;
 
 	return autoremove_wake_function(curr, mode, wake_flags, key);
@@ -2896,10 +6162,15 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
 		.to_wait	= min_events,
 	};
 	struct io_rings *rings = ctx->rings;
-	int ret;
+	int ret = 0;
 
-	if (io_cqring_events(rings) >= min_events)
-		return 0;
+	do {
+		if (io_cqring_events(ctx, false) >= min_events)
+			return 0;
+		if (!current->task_works)
+			break;
+		task_work_run();
+	} while (1);
 
 	if (sig) {
 #ifdef CONFIG_COMPAT
@@ -2914,24 +6185,24 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
 			return ret;
 	}
 
-	ret = 0;
 	iowq.nr_timeouts = atomic_read(&ctx->cq_timeouts);
+	trace_io_uring_cqring_wait(ctx, min_events);
 	do {
 		prepare_to_wait_exclusive(&ctx->wait, &iowq.wq,
 						TASK_INTERRUPTIBLE);
-		if (io_should_wake(&iowq))
+		if (current->task_works)
+			task_work_run();
+		if (io_should_wake(&iowq, false))
 			break;
 		schedule();
 		if (signal_pending(current)) {
-			ret = -ERESTARTSYS;
+			ret = -EINTR;
 			break;
 		}
 	} while (1);
 	finish_wait(&ctx->wait, &iowq.wq);
 
-	restore_saved_sigmask_unless(ret == -ERESTARTSYS);
-	if (ret == -ERESTARTSYS)
-		ret = -EINTR;
+	restore_saved_sigmask_unless(ret == -EINTR);
 
 	return READ_ONCE(rings->cq.head) == READ_ONCE(rings->cq.tail) ? ret : 0;
 }
@@ -2949,19 +6220,55 @@ static void __io_sqe_files_unregister(struct io_ring_ctx *ctx)
 #else
 	int i;
 
-	for (i = 0; i < ctx->nr_user_files; i++)
-		fput(ctx->user_files[i]);
+	for (i = 0; i < ctx->nr_user_files; i++) {
+		struct file *file;
+
+		file = io_file_from_index(ctx, i);
+		if (file)
+			fput(file);
+	}
 #endif
 }
 
+static void io_file_ref_kill(struct percpu_ref *ref)
+{
+	struct fixed_file_data *data;
+
+	data = container_of(ref, struct fixed_file_data, refs);
+	complete(&data->done);
+}
+
 static int io_sqe_files_unregister(struct io_ring_ctx *ctx)
 {
-	if (!ctx->user_files)
+	struct fixed_file_data *data = ctx->file_data;
+	struct fixed_file_ref_node *ref_node = NULL;
+	unsigned nr_tables, i;
+
+	if (!data)
 		return -ENXIO;
 
+	spin_lock(&data->lock);
+	if (!list_empty(&data->ref_list))
+		ref_node = list_first_entry(&data->ref_list,
+				struct fixed_file_ref_node, node);
+	spin_unlock(&data->lock);
+	if (ref_node)
+		percpu_ref_kill(&ref_node->refs);
+
+	percpu_ref_kill(&data->refs);
+
+	/* wait for all refs nodes to complete */
+	flush_delayed_work(&ctx->file_put_work);
+	wait_for_completion(&data->done);
+
 	__io_sqe_files_unregister(ctx);
-	kfree(ctx->user_files);
-	ctx->user_files = NULL;
+	nr_tables = DIV_ROUND_UP(ctx->nr_user_files, IORING_MAX_FILES_TABLE);
+	for (i = 0; i < nr_tables; i++)
+		kfree(data->table[i].files);
+	kfree(data->table);
+	percpu_ref_exit(&data->refs);
+	kfree(data);
+	ctx->file_data = NULL;
 	ctx->nr_user_files = 0;
 	return 0;
 }
@@ -2969,7 +6276,7 @@ static int io_sqe_files_unregister(struct io_ring_ctx *ctx)
 static void io_sq_thread_stop(struct io_ring_ctx *ctx)
 {
 	if (ctx->sqo_thread) {
-		wait_for_completion(&ctx->sqo_thread_started);
+		wait_for_completion(&ctx->sq_thread_comp);
 		/*
 		 * The park is a bit of a work-around, without it we get
 		 * warning spews on shutdown with SQPOLL set and affinity
@@ -2983,31 +6290,15 @@ static void io_sq_thread_stop(struct io_ring_ctx *ctx)
 
 static void io_finish_async(struct io_ring_ctx *ctx)
 {
-	int i;
-
 	io_sq_thread_stop(ctx);
 
-	for (i = 0; i < ARRAY_SIZE(ctx->sqo_wq); i++) {
-		if (ctx->sqo_wq[i]) {
-			destroy_workqueue(ctx->sqo_wq[i]);
-			ctx->sqo_wq[i] = NULL;
-		}
+	if (ctx->io_wq) {
+		io_wq_destroy(ctx->io_wq);
+		ctx->io_wq = NULL;
 	}
 }
 
 #if defined(CONFIG_UNIX)
-static void io_destruct_skb(struct sk_buff *skb)
-{
-	struct io_ring_ctx *ctx = skb->sk->sk_user_data;
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(ctx->sqo_wq); i++)
-		if (ctx->sqo_wq[i])
-			flush_workqueue(ctx->sqo_wq[i]);
-
-	unix_destruct_scm(skb);
-}
-
 /*
  * Ensure the UNIX gc is aware of our file set, so we are certain that
  * the io_uring can be safely unregistered on process exit, even if we have
@@ -3018,14 +6309,7 @@ static int __io_sqe_files_scm(struct io_ring_ctx *ctx, int nr, int offset)
 	struct sock *sk = ctx->ring_sock->sk;
 	struct scm_fp_list *fpl;
 	struct sk_buff *skb;
-	int i;
-
-	if (!capable(CAP_SYS_RESOURCE) && !capable(CAP_SYS_ADMIN)) {
-		unsigned long inflight = ctx->user->unix_inflight + nr;
-
-		if (inflight > task_rlimit(current, RLIMIT_NOFILE))
-			return -EMFILE;
-	}
+	int i, nr_files;
 
 	fpl = kzalloc(sizeof(*fpl), GFP_KERNEL);
 	if (!fpl)
@@ -3038,21 +6322,33 @@ static int __io_sqe_files_scm(struct io_ring_ctx *ctx, int nr, int offset)
 	}
 
 	skb->sk = sk;
-	skb->destructor = io_destruct_skb;
 
+	nr_files = 0;
 	fpl->user = get_uid(ctx->user);
 	for (i = 0; i < nr; i++) {
-		fpl->fp[i] = get_file(ctx->user_files[i + offset]);
-		unix_inflight(fpl->user, fpl->fp[i]);
+		struct file *file = io_file_from_index(ctx, i + offset);
+
+		if (!file)
+			continue;
+		fpl->fp[nr_files] = get_file(file);
+		unix_inflight(fpl->user, fpl->fp[nr_files]);
+		nr_files++;
 	}
 
-	fpl->max = fpl->count = nr;
-	UNIXCB(skb).fp = fpl;
-	refcount_add(skb->truesize, &sk->sk_wmem_alloc);
-	skb_queue_head(&sk->sk_receive_queue, skb);
+	if (nr_files) {
+		fpl->max = SCM_MAX_FD;
+		fpl->count = nr_files;
+		UNIXCB(skb).fp = fpl;
+		skb->destructor = unix_destruct_scm;
+		refcount_add(skb->truesize, &sk->sk_wmem_alloc);
+		skb_queue_head(&sk->sk_receive_queue, skb);
 
-	for (i = 0; i < nr; i++)
-		fput(fpl->fp[i]);
+		for (i = 0; i < nr_files; i++)
+			fput(fpl->fp[i]);
+	} else {
+		kfree_skb(skb);
+		kfree(fpl);
+	}
 
 	return 0;
 }
@@ -3083,7 +6379,10 @@ static int io_sqe_files_scm(struct io_ring_ctx *ctx)
 		return 0;
 
 	while (total < ctx->nr_user_files) {
-		fput(ctx->user_files[total]);
+		struct file *file = io_file_from_index(ctx, total);
+
+		if (file)
+			fput(file);
 		total++;
 	}
 
@@ -3096,34 +6395,257 @@ static int io_sqe_files_scm(struct io_ring_ctx *ctx)
 }
 #endif
 
+static int io_sqe_alloc_file_tables(struct io_ring_ctx *ctx, unsigned nr_tables,
+				    unsigned nr_files)
+{
+	int i;
+
+	for (i = 0; i < nr_tables; i++) {
+		struct fixed_file_table *table = &ctx->file_data->table[i];
+		unsigned this_files;
+
+		this_files = min(nr_files, IORING_MAX_FILES_TABLE);
+		table->files = kcalloc(this_files, sizeof(struct file *),
+					GFP_KERNEL);
+		if (!table->files)
+			break;
+		nr_files -= this_files;
+	}
+
+	if (i == nr_tables)
+		return 0;
+
+	for (i = 0; i < nr_tables; i++) {
+		struct fixed_file_table *table = &ctx->file_data->table[i];
+		kfree(table->files);
+	}
+	return 1;
+}
+
+static void io_ring_file_put(struct io_ring_ctx *ctx, struct file *file)
+{
+#if defined(CONFIG_UNIX)
+	struct sock *sock = ctx->ring_sock->sk;
+	struct sk_buff_head list, *head = &sock->sk_receive_queue;
+	struct sk_buff *skb;
+	int i;
+
+	__skb_queue_head_init(&list);
+
+	/*
+	 * Find the skb that holds this file in its SCM_RIGHTS. When found,
+	 * remove this entry and rearrange the file array.
+	 */
+	skb = skb_dequeue(head);
+	while (skb) {
+		struct scm_fp_list *fp;
+
+		fp = UNIXCB(skb).fp;
+		for (i = 0; i < fp->count; i++) {
+			int left;
+
+			if (fp->fp[i] != file)
+				continue;
+
+			unix_notinflight(fp->user, fp->fp[i]);
+			left = fp->count - 1 - i;
+			if (left) {
+				memmove(&fp->fp[i], &fp->fp[i + 1],
+						left * sizeof(struct file *));
+			}
+			fp->count--;
+			if (!fp->count) {
+				kfree_skb(skb);
+				skb = NULL;
+			} else {
+				__skb_queue_tail(&list, skb);
+			}
+			fput(file);
+			file = NULL;
+			break;
+		}
+
+		if (!file)
+			break;
+
+		__skb_queue_tail(&list, skb);
+
+		skb = skb_dequeue(head);
+	}
+
+	if (skb_peek(&list)) {
+		spin_lock_irq(&head->lock);
+		while ((skb = __skb_dequeue(&list)) != NULL)
+			__skb_queue_tail(head, skb);
+		spin_unlock_irq(&head->lock);
+	}
+#else
+	fput(file);
+#endif
+}
+
+struct io_file_put {
+	struct list_head list;
+	struct file *file;
+};
+
+static void __io_file_put_work(struct fixed_file_ref_node *ref_node)
+{
+	struct fixed_file_data *file_data = ref_node->file_data;
+	struct io_ring_ctx *ctx = file_data->ctx;
+	struct io_file_put *pfile, *tmp;
+
+	list_for_each_entry_safe(pfile, tmp, &ref_node->file_list, list) {
+		list_del(&pfile->list);
+		io_ring_file_put(ctx, pfile->file);
+		kfree(pfile);
+	}
+
+	spin_lock(&file_data->lock);
+	list_del(&ref_node->node);
+	spin_unlock(&file_data->lock);
+
+	percpu_ref_exit(&ref_node->refs);
+	kfree(ref_node);
+	percpu_ref_put(&file_data->refs);
+}
+
+static void io_file_put_work(struct work_struct *work)
+{
+	struct io_ring_ctx *ctx;
+	struct llist_node *node;
+
+	ctx = container_of(work, struct io_ring_ctx, file_put_work.work);
+	node = llist_del_all(&ctx->file_put_llist);
+
+	while (node) {
+		struct fixed_file_ref_node *ref_node;
+		struct llist_node *next = node->next;
+
+		ref_node = llist_entry(node, struct fixed_file_ref_node, llist);
+		__io_file_put_work(ref_node);
+		node = next;
+	}
+}
+
+static void io_file_data_ref_zero(struct percpu_ref *ref)
+{
+	struct fixed_file_ref_node *ref_node;
+	struct io_ring_ctx *ctx;
+	bool first_add;
+	int delay = HZ;
+
+	ref_node = container_of(ref, struct fixed_file_ref_node, refs);
+	ctx = ref_node->file_data->ctx;
+
+	if (percpu_ref_is_dying(&ctx->file_data->refs))
+		delay = 0;
+
+	first_add = llist_add(&ref_node->llist, &ctx->file_put_llist);
+	if (!delay)
+		mod_delayed_work(system_wq, &ctx->file_put_work, 0);
+	else if (first_add)
+		queue_delayed_work(system_wq, &ctx->file_put_work, delay);
+}
+
+static struct fixed_file_ref_node *alloc_fixed_file_ref_node(
+			struct io_ring_ctx *ctx)
+{
+	struct fixed_file_ref_node *ref_node;
+
+	ref_node = kzalloc(sizeof(*ref_node), GFP_KERNEL);
+	if (!ref_node)
+		return ERR_PTR(-ENOMEM);
+
+	if (percpu_ref_init(&ref_node->refs, io_file_data_ref_zero,
+			    0, GFP_KERNEL)) {
+		kfree(ref_node);
+		return ERR_PTR(-ENOMEM);
+	}
+	INIT_LIST_HEAD(&ref_node->node);
+	INIT_LIST_HEAD(&ref_node->file_list);
+	ref_node->file_data = ctx->file_data;
+	return ref_node;
+}
+
+static void destroy_fixed_file_ref_node(struct fixed_file_ref_node *ref_node)
+{
+	percpu_ref_exit(&ref_node->refs);
+	kfree(ref_node);
+}
+
 static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg,
 				 unsigned nr_args)
 {
 	__s32 __user *fds = (__s32 __user *) arg;
+	unsigned nr_tables;
+	struct file *file;
 	int fd, ret = 0;
 	unsigned i;
+	struct fixed_file_ref_node *ref_node;
 
-	if (ctx->user_files)
+	if (ctx->file_data)
 		return -EBUSY;
 	if (!nr_args)
 		return -EINVAL;
 	if (nr_args > IORING_MAX_FIXED_FILES)
 		return -EMFILE;
 
-	ctx->user_files = kcalloc(nr_args, sizeof(struct file *), GFP_KERNEL);
-	if (!ctx->user_files)
+	ctx->file_data = kzalloc(sizeof(*ctx->file_data), GFP_KERNEL);
+	if (!ctx->file_data)
+		return -ENOMEM;
+	ctx->file_data->ctx = ctx;
+	init_completion(&ctx->file_data->done);
+	INIT_LIST_HEAD(&ctx->file_data->ref_list);
+	spin_lock_init(&ctx->file_data->lock);
+
+	nr_tables = DIV_ROUND_UP(nr_args, IORING_MAX_FILES_TABLE);
+	ctx->file_data->table = kcalloc(nr_tables,
+					sizeof(struct fixed_file_table),
+					GFP_KERNEL);
+	if (!ctx->file_data->table) {
+		kfree(ctx->file_data);
+		ctx->file_data = NULL;
+		return -ENOMEM;
+	}
+
+	if (percpu_ref_init(&ctx->file_data->refs, io_file_ref_kill,
+				PERCPU_REF_ALLOW_REINIT, GFP_KERNEL)) {
+		kfree(ctx->file_data->table);
+		kfree(ctx->file_data);
+		ctx->file_data = NULL;
 		return -ENOMEM;
+	}
+
+	if (io_sqe_alloc_file_tables(ctx, nr_tables, nr_args)) {
+		percpu_ref_exit(&ctx->file_data->refs);
+		kfree(ctx->file_data->table);
+		kfree(ctx->file_data);
+		ctx->file_data = NULL;
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < nr_args; i++, ctx->nr_user_files++) {
+		struct fixed_file_table *table;
+		unsigned index;
 
-	for (i = 0; i < nr_args; i++) {
 		ret = -EFAULT;
 		if (copy_from_user(&fd, &fds[i], sizeof(fd)))
 			break;
+		/* allow sparse sets */
+		if (fd == -1) {
+			ret = 0;
+			continue;
+		}
 
-		ctx->user_files[i] = fget(fd);
+		table = &ctx->file_data->table[i >> IORING_FILE_TABLE_SHIFT];
+		index = i & IORING_FILE_TABLE_MASK;
+		file = fget(fd);
 
 		ret = -EBADF;
-		if (!ctx->user_files[i])
+		if (!file)
 			break;
+
 		/*
 		 * Don't allow io_uring instances to be registered. If UNIX
 		 * isn't enabled, then this causes a reference cycle and this
@@ -3131,28 +6653,264 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg,
 		 * handle it just fine, but there's still no point in allowing
 		 * a ring fd as it doesn't support regular read/write anyway.
 		 */
-		if (ctx->user_files[i]->f_op == &io_uring_fops) {
-			fput(ctx->user_files[i]);
+		if (file->f_op == &io_uring_fops) {
+			fput(file);
 			break;
 		}
-		ctx->nr_user_files++;
 		ret = 0;
+		table->files[index] = file;
 	}
 
 	if (ret) {
-		for (i = 0; i < ctx->nr_user_files; i++)
-			fput(ctx->user_files[i]);
+		for (i = 0; i < ctx->nr_user_files; i++) {
+			file = io_file_from_index(ctx, i);
+			if (file)
+				fput(file);
+		}
+		for (i = 0; i < nr_tables; i++)
+			kfree(ctx->file_data->table[i].files);
 
-		kfree(ctx->user_files);
-		ctx->user_files = NULL;
+		kfree(ctx->file_data->table);
+		kfree(ctx->file_data);
+		ctx->file_data = NULL;
 		ctx->nr_user_files = 0;
 		return ret;
 	}
 
 	ret = io_sqe_files_scm(ctx);
-	if (ret)
+	if (ret) {
 		io_sqe_files_unregister(ctx);
+		return ret;
+	}
+
+	ref_node = alloc_fixed_file_ref_node(ctx);
+	if (IS_ERR(ref_node)) {
+		io_sqe_files_unregister(ctx);
+		return PTR_ERR(ref_node);
+	}
+
+	ctx->file_data->cur_refs = &ref_node->refs;
+	spin_lock(&ctx->file_data->lock);
+	list_add(&ref_node->node, &ctx->file_data->ref_list);
+	spin_unlock(&ctx->file_data->lock);
+	percpu_ref_get(&ctx->file_data->refs);
+	return ret;
+}
+
+static int io_sqe_file_register(struct io_ring_ctx *ctx, struct file *file,
+				int index)
+{
+#if defined(CONFIG_UNIX)
+	struct sock *sock = ctx->ring_sock->sk;
+	struct sk_buff_head *head = &sock->sk_receive_queue;
+	struct sk_buff *skb;
+
+	/*
+	 * See if we can merge this file into an existing skb SCM_RIGHTS
+	 * file set. If there's no room, fall back to allocating a new skb
+	 * and filling it in.
+	 */
+	spin_lock_irq(&head->lock);
+	skb = skb_peek(head);
+	if (skb) {
+		struct scm_fp_list *fpl = UNIXCB(skb).fp;
+
+		if (fpl->count < SCM_MAX_FD) {
+			__skb_unlink(skb, head);
+			spin_unlock_irq(&head->lock);
+			fpl->fp[fpl->count] = get_file(file);
+			unix_inflight(fpl->user, fpl->fp[fpl->count]);
+			fpl->count++;
+			spin_lock_irq(&head->lock);
+			__skb_queue_head(head, skb);
+		} else {
+			skb = NULL;
+		}
+	}
+	spin_unlock_irq(&head->lock);
+
+	if (skb) {
+		fput(file);
+		return 0;
+	}
+
+	return __io_sqe_files_scm(ctx, 1, index);
+#else
+	return 0;
+#endif
+}
+
+static int io_queue_file_removal(struct fixed_file_data *data,
+				 struct file *file)
+{
+	struct io_file_put *pfile;
+	struct percpu_ref *refs = data->cur_refs;
+	struct fixed_file_ref_node *ref_node;
+
+	pfile = kzalloc(sizeof(*pfile), GFP_KERNEL);
+	if (!pfile)
+		return -ENOMEM;
+
+	ref_node = container_of(refs, struct fixed_file_ref_node, refs);
+	pfile->file = file;
+	list_add(&pfile->list, &ref_node->file_list);
+
+	return 0;
+}
+
+static int __io_sqe_files_update(struct io_ring_ctx *ctx,
+				 struct io_uring_files_update *up,
+				 unsigned nr_args)
+{
+	struct fixed_file_data *data = ctx->file_data;
+	struct fixed_file_ref_node *ref_node;
+	struct file *file;
+	__s32 __user *fds;
+	int fd, i, err;
+	__u32 done;
+	bool needs_switch = false;
+
+	if (check_add_overflow(up->offset, nr_args, &done))
+		return -EOVERFLOW;
+	if (done > ctx->nr_user_files)
+		return -EINVAL;
+
+	ref_node = alloc_fixed_file_ref_node(ctx);
+	if (IS_ERR(ref_node))
+		return PTR_ERR(ref_node);
+
+	done = 0;
+	fds = u64_to_user_ptr(up->fds);
+	while (nr_args) {
+		struct fixed_file_table *table;
+		unsigned index;
+
+		err = 0;
+		if (copy_from_user(&fd, &fds[done], sizeof(fd))) {
+			err = -EFAULT;
+			break;
+		}
+		i = array_index_nospec(up->offset, ctx->nr_user_files);
+		table = &ctx->file_data->table[i >> IORING_FILE_TABLE_SHIFT];
+		index = i & IORING_FILE_TABLE_MASK;
+		if (table->files[index]) {
+			file = io_file_from_index(ctx, index);
+			err = io_queue_file_removal(data, file);
+			if (err)
+				break;
+			table->files[index] = NULL;
+			needs_switch = true;
+		}
+		if (fd != -1) {
+			file = fget(fd);
+			if (!file) {
+				err = -EBADF;
+				break;
+			}
+			/*
+			 * Don't allow io_uring instances to be registered. If
+			 * UNIX isn't enabled, then this causes a reference
+			 * cycle and this instance can never get freed. If UNIX
+			 * is enabled we'll handle it just fine, but there's
+			 * still no point in allowing a ring fd as it doesn't
+			 * support regular read/write anyway.
+			 */
+			if (file->f_op == &io_uring_fops) {
+				fput(file);
+				err = -EBADF;
+				break;
+			}
+			table->files[index] = file;
+			err = io_sqe_file_register(ctx, file, i);
+			if (err)
+				break;
+		}
+		nr_args--;
+		done++;
+		up->offset++;
+	}
+
+	if (needs_switch) {
+		percpu_ref_kill(data->cur_refs);
+		spin_lock(&data->lock);
+		list_add(&ref_node->node, &data->ref_list);
+		data->cur_refs = &ref_node->refs;
+		spin_unlock(&data->lock);
+		percpu_ref_get(&ctx->file_data->refs);
+	} else
+		destroy_fixed_file_ref_node(ref_node);
+
+	return done ? done : err;
+}
+
+static int io_sqe_files_update(struct io_ring_ctx *ctx, void __user *arg,
+			       unsigned nr_args)
+{
+	struct io_uring_files_update up;
+
+	if (!ctx->file_data)
+		return -ENXIO;
+	if (!nr_args)
+		return -EINVAL;
+	if (copy_from_user(&up, arg, sizeof(up)))
+		return -EFAULT;
+	if (up.resv)
+		return -EINVAL;
+
+	return __io_sqe_files_update(ctx, &up, nr_args);
+}
+
+static void io_free_work(struct io_wq_work *work)
+{
+	struct io_kiocb *req = container_of(work, struct io_kiocb, work);
+
+	/* Consider that io_steal_work() relies on this ref */
+	io_put_req(req);
+}
 
+static int io_init_wq_offload(struct io_ring_ctx *ctx,
+			      struct io_uring_params *p)
+{
+	struct io_wq_data data;
+	struct fd f;
+	struct io_ring_ctx *ctx_attach;
+	unsigned int concurrency;
+	int ret = 0;
+
+	data.user = ctx->user;
+	data.free_work = io_free_work;
+
+	if (!(p->flags & IORING_SETUP_ATTACH_WQ)) {
+		/* Do QD, or 4 * CPUS, whatever is smallest */
+		concurrency = min(ctx->sq_entries, 4 * num_online_cpus());
+
+		ctx->io_wq = io_wq_create(concurrency, &data);
+		if (IS_ERR(ctx->io_wq)) {
+			ret = PTR_ERR(ctx->io_wq);
+			ctx->io_wq = NULL;
+		}
+		return ret;
+	}
+
+	f = fdget(p->wq_fd);
+	if (!f.file)
+		return -EBADF;
+
+	if (f.file->f_op != &io_uring_fops) {
+		ret = -EINVAL;
+		goto out_fput;
+	}
+
+	ctx_attach = f.file->private_data;
+	/* @io_wq is protected by holding the fd */
+	if (!io_wq_get(ctx_attach->io_wq, &data)) {
+		ret = -EINVAL;
+		goto out_fput;
+	}
+
+	ctx->io_wq = ctx_attach->io_wq;
+out_fput:
+	fdput(f);
 	return ret;
 }
 
@@ -3161,7 +6919,6 @@ static int io_sq_offload_start(struct io_ring_ctx *ctx,
 {
 	int ret;
 
-	init_waitqueue_head(&ctx->sqo_wait);
 	mmgrab(current->mm);
 	ctx->sqo_mm = current->mm;
 
@@ -3202,28 +6959,9 @@ static int io_sq_offload_start(struct io_ring_ctx *ctx,
 		goto err;
 	}
 
-	/* Do QD, or 2 * CPUS, whatever is smallest */
-	ctx->sqo_wq[0] = alloc_workqueue("io_ring-wq",
-			WQ_UNBOUND | WQ_FREEZABLE,
-			min(ctx->sq_entries - 1, 2 * num_online_cpus()));
-	if (!ctx->sqo_wq[0]) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	/*
-	 * This is for buffered writes, where we want to limit the parallelism
-	 * due to file locking in file systems. As "normal" buffered writes
-	 * should parellelize on writeout quite nicely, limit us to having 2
-	 * pending. This avoids massive contention on the inode when doing
-	 * buffered async writes.
-	 */
-	ctx->sqo_wq[1] = alloc_workqueue("io_ring-write-wq",
-						WQ_UNBOUND | WQ_FREEZABLE, 2);
-	if (!ctx->sqo_wq[1]) {
-		ret = -ENOMEM;
+	ret = io_init_wq_offload(ctx, p);
+	if (ret)
 		goto err;
-	}
 
 	return 0;
 err:
@@ -3328,7 +7066,7 @@ static int io_sqe_buffer_unregister(struct io_ring_ctx *ctx)
 		struct io_mapped_ubuf *imu = &ctx->user_bufs[i];
 
 		for (j = 0; j < imu->nr_bvecs; j++)
-			put_user_page(imu->bvec[j].bv_page);
+			unpin_user_page(imu->bvec[j].bv_page);
 
 		if (ctx->account_mem)
 			io_unaccount_mem(ctx->user, imu->nr_bvecs);
@@ -3356,7 +7094,7 @@ static int io_copy_iov(struct io_ring_ctx *ctx, struct iovec *dst,
 		if (copy_from_user(&ciov, &ciovs[index], sizeof(ciov)))
 			return -EFAULT;
 
-		dst->iov_base = (void __user *) (unsigned long) ciov.iov_base;
+		dst->iov_base = u64_to_user_ptr((u64)ciov.iov_base);
 		dst->iov_len = ciov.iov_len;
 		return 0;
 	}
@@ -3449,7 +7187,7 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, void __user *arg,
 
 		ret = 0;
 		down_read(&current->mm->mmap_sem);
-		pret = get_user_pages(ubuf, nr_pages,
+		pret = pin_user_pages(ubuf, nr_pages,
 				      FOLL_WRITE | FOLL_LONGTERM,
 				      pages, vmas);
 		if (pret == nr_pages) {
@@ -3473,7 +7211,7 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, void __user *arg,
 			 * release any pages we did get
 			 */
 			if (pret > 0)
-				put_user_pages(pages, pret);
+				unpin_user_pages(pages, pret);
 			if (ctx->account_mem)
 				io_unaccount_mem(ctx->user, nr_pages);
 			kvfree(imu->bvec);
@@ -3541,6 +7279,21 @@ static int io_eventfd_unregister(struct io_ring_ctx *ctx)
 	return -ENXIO;
 }
 
+static int __io_destroy_buffers(int id, void *p, void *data)
+{
+	struct io_ring_ctx *ctx = data;
+	struct io_buffer *buf = p;
+
+	__io_remove_buffers(ctx, buf, id, -1U);
+	return 0;
+}
+
+static void io_destroy_buffers(struct io_ring_ctx *ctx)
+{
+	idr_for_each(&ctx->io_buffer_idr, __io_destroy_buffers, ctx);
+	idr_destroy(&ctx->io_buffer_idr);
+}
+
 static void io_ring_ctx_free(struct io_ring_ctx *ctx)
 {
 	io_finish_async(ctx);
@@ -3551,6 +7304,8 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx)
 	io_sqe_buffer_unregister(ctx);
 	io_sqe_files_unregister(ctx);
 	io_eventfd_unregister(ctx);
+	io_destroy_buffers(ctx);
+	idr_destroy(&ctx->personality_idr);
 
 #if defined(CONFIG_UNIX)
 	if (ctx->ring_sock) {
@@ -3567,6 +7322,9 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx)
 		io_unaccount_mem(ctx->user,
 				ring_pages(ctx->sq_entries, ctx->cq_entries));
 	free_uid(ctx->user);
+	put_cred(ctx->creds);
+	kfree(ctx->cancel_hash);
+	kmem_cache_free(req_cachep, ctx->fallback_req);
 	kfree(ctx);
 }
 
@@ -3584,7 +7342,7 @@ static __poll_t io_uring_poll(struct file *file, poll_table *wait)
 	if (READ_ONCE(ctx->rings->sq.tail) - ctx->cached_sq_head !=
 	    ctx->rings->sq_ring_entries)
 		mask |= EPOLLOUT | EPOLLWRNORM;
-	if (READ_ONCE(ctx->rings->cq.head) != ctx->cached_cq_tail)
+	if (io_cqring_events(ctx, false))
 		mask |= EPOLLIN | EPOLLRDNORM;
 
 	return mask;
@@ -3597,6 +7355,29 @@ static int io_uring_fasync(int fd, struct file *file, int on)
 	return fasync_helper(fd, file, on, &ctx->cq_fasync);
 }
 
+static int io_remove_personalities(int id, void *p, void *data)
+{
+	struct io_ring_ctx *ctx = data;
+	const struct cred *cred;
+
+	cred = idr_remove(&ctx->personality_idr, id);
+	if (cred)
+		put_cred(cred);
+	return 0;
+}
+
+static void io_ring_exit_work(struct work_struct *work)
+{
+	struct io_ring_ctx *ctx;
+
+	ctx = container_of(work, struct io_ring_ctx, exit_work);
+	if (ctx->rings)
+		io_cqring_overflow_flush(ctx, true);
+
+	wait_for_completion(&ctx->ref_comp);
+	io_ring_ctx_free(ctx);
+}
+
 static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)
 {
 	mutex_lock(&ctx->uring_lock);
@@ -3605,9 +7386,17 @@ static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)
 
 	io_kill_timeouts(ctx);
 	io_poll_remove_all(ctx);
+
+	if (ctx->io_wq)
+		io_wq_cancel_all(ctx->io_wq);
+
 	io_iopoll_reap_events(ctx);
-	wait_for_completion(&ctx->ctx_done);
-	io_ring_ctx_free(ctx);
+	/* if we failed setting up the ctx, we might not have any rings */
+	if (ctx->rings)
+		io_cqring_overflow_flush(ctx, true);
+	idr_for_each(&ctx->personality_idr, io_remove_personalities, ctx);
+	INIT_WORK(&ctx->exit_work, io_ring_exit_work);
+	queue_work(system_wq, &ctx->exit_work);
 }
 
 static int io_uring_release(struct inode *inode, struct file *file)
@@ -3619,12 +7408,84 @@ static int io_uring_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static int io_uring_mmap(struct file *file, struct vm_area_struct *vma)
+static void io_uring_cancel_files(struct io_ring_ctx *ctx,
+				  struct files_struct *files)
+{
+	while (!list_empty_careful(&ctx->inflight_list)) {
+		struct io_kiocb *cancel_req = NULL, *req;
+		DEFINE_WAIT(wait);
+
+		spin_lock_irq(&ctx->inflight_lock);
+		list_for_each_entry(req, &ctx->inflight_list, inflight_entry) {
+			if (req->work.files != files)
+				continue;
+			/* req is being completed, ignore */
+			if (!refcount_inc_not_zero(&req->refs))
+				continue;
+			cancel_req = req;
+			break;
+		}
+		if (cancel_req)
+			prepare_to_wait(&ctx->inflight_wait, &wait,
+						TASK_UNINTERRUPTIBLE);
+		spin_unlock_irq(&ctx->inflight_lock);
+
+		/* We need to keep going until we don't find a matching req */
+		if (!cancel_req)
+			break;
+
+		if (cancel_req->flags & REQ_F_OVERFLOW) {
+			spin_lock_irq(&ctx->completion_lock);
+			list_del(&cancel_req->list);
+			cancel_req->flags &= ~REQ_F_OVERFLOW;
+			if (list_empty(&ctx->cq_overflow_list)) {
+				clear_bit(0, &ctx->sq_check_overflow);
+				clear_bit(0, &ctx->cq_check_overflow);
+			}
+			spin_unlock_irq(&ctx->completion_lock);
+
+			WRITE_ONCE(ctx->rings->cq_overflow,
+				atomic_inc_return(&ctx->cached_cq_overflow));
+
+			/*
+			 * Put inflight ref and overflow ref. If that's
+			 * all we had, then we're done with this request.
+			 */
+			if (refcount_sub_and_test(2, &cancel_req->refs)) {
+				io_free_req(cancel_req);
+				finish_wait(&ctx->inflight_wait, &wait);
+				continue;
+			}
+		} else {
+			io_wq_cancel_work(ctx->io_wq, &cancel_req->work);
+			io_put_req(cancel_req);
+		}
+
+		schedule();
+		finish_wait(&ctx->inflight_wait, &wait);
+	}
+}
+
+static int io_uring_flush(struct file *file, void *data)
 {
-	loff_t offset = (loff_t) vma->vm_pgoff << PAGE_SHIFT;
-	unsigned long sz = vma->vm_end - vma->vm_start;
 	struct io_ring_ctx *ctx = file->private_data;
-	unsigned long pfn;
+
+	io_uring_cancel_files(ctx, data);
+
+	/*
+	 * If the task is going away, cancel work it may have pending
+	 */
+	if (fatal_signal_pending(current) || (current->flags & PF_EXITING))
+		io_wq_cancel_pid(ctx->io_wq, task_pid_vnr(current));
+
+	return 0;
+}
+
+static void *io_uring_validate_mmap_request(struct file *file,
+					    loff_t pgoff, size_t sz)
+{
+	struct io_ring_ctx *ctx = file->private_data;
+	loff_t offset = pgoff << PAGE_SHIFT;
 	struct page *page;
 	void *ptr;
 
@@ -3637,17 +7498,59 @@ static int io_uring_mmap(struct file *file, struct vm_area_struct *vma)
 		ptr = ctx->sq_sqes;
 		break;
 	default:
-		return -EINVAL;
+		return ERR_PTR(-EINVAL);
 	}
 
 	page = virt_to_head_page(ptr);
 	if (sz > page_size(page))
-		return -EINVAL;
+		return ERR_PTR(-EINVAL);
+
+	return ptr;
+}
+
+#ifdef CONFIG_MMU
+
+static int io_uring_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	size_t sz = vma->vm_end - vma->vm_start;
+	unsigned long pfn;
+	void *ptr;
+
+	ptr = io_uring_validate_mmap_request(file, vma->vm_pgoff, sz);
+	if (IS_ERR(ptr))
+		return PTR_ERR(ptr);
 
 	pfn = virt_to_phys(ptr) >> PAGE_SHIFT;
 	return remap_pfn_range(vma, vma->vm_start, pfn, sz, vma->vm_page_prot);
 }
 
+#else /* !CONFIG_MMU */
+
+static int io_uring_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	return vma->vm_flags & (VM_SHARED | VM_MAYSHARE) ? 0 : -EINVAL;
+}
+
+static unsigned int io_uring_nommu_mmap_capabilities(struct file *file)
+{
+	return NOMMU_MAP_DIRECT | NOMMU_MAP_READ | NOMMU_MAP_WRITE;
+}
+
+static unsigned long io_uring_nommu_get_unmapped_area(struct file *file,
+	unsigned long addr, unsigned long len,
+	unsigned long pgoff, unsigned long flags)
+{
+	void *ptr;
+
+	ptr = io_uring_validate_mmap_request(file, pgoff, len);
+	if (IS_ERR(ptr))
+		return PTR_ERR(ptr);
+
+	return (unsigned long) ptr;
+}
+
+#endif /* !CONFIG_MMU */
+
 SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
 		u32, min_complete, u32, flags, const sigset_t __user *, sig,
 		size_t, sigsz)
@@ -3657,6 +7560,9 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
 	int submitted = 0;
 	struct fd f;
 
+	if (current->task_works)
+		task_work_run();
+
 	if (flags & ~(IORING_ENTER_GETEVENTS | IORING_ENTER_SQ_WAKEUP))
 		return -EINVAL;
 
@@ -3680,39 +7586,145 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
 	 */
 	ret = 0;
 	if (ctx->flags & IORING_SETUP_SQPOLL) {
+		if (!list_empty_careful(&ctx->cq_overflow_list))
+			io_cqring_overflow_flush(ctx, false);
 		if (flags & IORING_ENTER_SQ_WAKEUP)
 			wake_up(&ctx->sqo_wait);
 		submitted = to_submit;
 	} else if (to_submit) {
-		to_submit = min(to_submit, ctx->sq_entries);
-
 		mutex_lock(&ctx->uring_lock);
-		submitted = io_ring_submit(ctx, to_submit);
+		submitted = io_submit_sqes(ctx, to_submit, f.file, fd);
 		mutex_unlock(&ctx->uring_lock);
+
+		if (submitted != to_submit)
+			goto out;
 	}
 	if (flags & IORING_ENTER_GETEVENTS) {
 		unsigned nr_events = 0;
 
 		min_complete = min(min_complete, ctx->cq_entries);
 
-		if (ctx->flags & IORING_SETUP_IOPOLL) {
+		/*
+		 * When SETUP_IOPOLL and SETUP_SQPOLL are both enabled, user
+		 * space applications don't need to do io completion events
+		 * polling again, they can rely on io_sq_thread to do polling
+		 * work, which can reduce cpu usage and uring_lock contention.
+		 */
+		if (ctx->flags & IORING_SETUP_IOPOLL &&
+		    !(ctx->flags & IORING_SETUP_SQPOLL)) {
 			ret = io_iopoll_check(ctx, &nr_events, min_complete);
 		} else {
 			ret = io_cqring_wait(ctx, min_complete, sig, sigsz);
 		}
 	}
 
+out:
 	percpu_ref_put(&ctx->refs);
 out_fput:
 	fdput(f);
 	return submitted ? submitted : ret;
 }
 
+#ifdef CONFIG_PROC_FS
+static int io_uring_show_cred(int id, void *p, void *data)
+{
+	const struct cred *cred = p;
+	struct seq_file *m = data;
+	struct user_namespace *uns = seq_user_ns(m);
+	struct group_info *gi;
+	kernel_cap_t cap;
+	unsigned __capi;
+	int g;
+
+	seq_printf(m, "%5d\n", id);
+	seq_put_decimal_ull(m, "\tUid:\t", from_kuid_munged(uns, cred->uid));
+	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->euid));
+	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->suid));
+	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->fsuid));
+	seq_put_decimal_ull(m, "\n\tGid:\t", from_kgid_munged(uns, cred->gid));
+	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->egid));
+	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->sgid));
+	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->fsgid));
+	seq_puts(m, "\n\tGroups:\t");
+	gi = cred->group_info;
+	for (g = 0; g < gi->ngroups; g++) {
+		seq_put_decimal_ull(m, g ? " " : "",
+					from_kgid_munged(uns, gi->gid[g]));
+	}
+	seq_puts(m, "\n\tCapEff:\t");
+	cap = cred->cap_effective;
+	CAP_FOR_EACH_U32(__capi)
+		seq_put_hex_ll(m, NULL, cap.cap[CAP_LAST_U32 - __capi], 8);
+	seq_putc(m, '\n');
+	return 0;
+}
+
+static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m)
+{
+	int i;
+
+	mutex_lock(&ctx->uring_lock);
+	seq_printf(m, "UserFiles:\t%u\n", ctx->nr_user_files);
+	for (i = 0; i < ctx->nr_user_files; i++) {
+		struct fixed_file_table *table;
+		struct file *f;
+
+		table = &ctx->file_data->table[i >> IORING_FILE_TABLE_SHIFT];
+		f = table->files[i & IORING_FILE_TABLE_MASK];
+		if (f)
+			seq_printf(m, "%5u: %s\n", i, file_dentry(f)->d_iname);
+		else
+			seq_printf(m, "%5u: <none>\n", i);
+	}
+	seq_printf(m, "UserBufs:\t%u\n", ctx->nr_user_bufs);
+	for (i = 0; i < ctx->nr_user_bufs; i++) {
+		struct io_mapped_ubuf *buf = &ctx->user_bufs[i];
+
+		seq_printf(m, "%5u: 0x%llx/%u\n", i, buf->ubuf,
+						(unsigned int) buf->len);
+	}
+	if (!idr_is_empty(&ctx->personality_idr)) {
+		seq_printf(m, "Personalities:\n");
+		idr_for_each(&ctx->personality_idr, io_uring_show_cred, m);
+	}
+	seq_printf(m, "PollList:\n");
+	spin_lock_irq(&ctx->completion_lock);
+	for (i = 0; i < (1U << ctx->cancel_hash_bits); i++) {
+		struct hlist_head *list = &ctx->cancel_hash[i];
+		struct io_kiocb *req;
+
+		hlist_for_each_entry(req, list, hash_node)
+			seq_printf(m, "  op=%d, task_works=%d\n", req->opcode,
+					req->task->task_works != NULL);
+	}
+	spin_unlock_irq(&ctx->completion_lock);
+	mutex_unlock(&ctx->uring_lock);
+}
+
+static void io_uring_show_fdinfo(struct seq_file *m, struct file *f)
+{
+	struct io_ring_ctx *ctx = f->private_data;
+
+	if (percpu_ref_tryget(&ctx->refs)) {
+		__io_uring_show_fdinfo(ctx, m);
+		percpu_ref_put(&ctx->refs);
+	}
+}
+#endif
+
 static const struct file_operations io_uring_fops = {
 	.release	= io_uring_release,
+	.flush		= io_uring_flush,
 	.mmap		= io_uring_mmap,
+#ifndef CONFIG_MMU
+	.get_unmapped_area = io_uring_nommu_get_unmapped_area,
+	.mmap_capabilities = io_uring_nommu_mmap_capabilities,
+#endif
 	.poll		= io_uring_poll,
 	.fasync		= io_uring_fasync,
+#ifdef CONFIG_PROC_FS
+	.show_fdinfo	= io_uring_show_fdinfo,
+#endif
 };
 
 static int io_allocate_scq_urings(struct io_ring_ctx *ctx,
@@ -3741,12 +7753,18 @@ static int io_allocate_scq_urings(struct io_ring_ctx *ctx,
 	ctx->cq_entries = rings->cq_ring_entries;
 
 	size = array_size(sizeof(struct io_uring_sqe), p->sq_entries);
-	if (size == SIZE_MAX)
+	if (size == SIZE_MAX) {
+		io_mem_free(ctx->rings);
+		ctx->rings = NULL;
 		return -EOVERFLOW;
+	}
 
 	ctx->sq_sqes = io_mem_alloc(size);
-	if (!ctx->sq_sqes)
+	if (!ctx->sq_sqes) {
+		io_mem_free(ctx->rings);
+		ctx->rings = NULL;
 		return -ENOMEM;
+	}
 
 	return 0;
 }
@@ -3783,7 +7801,6 @@ static int io_uring_get_fd(struct io_ring_ctx *ctx)
 
 #if defined(CONFIG_UNIX)
 	ctx->ring_sock->file = file;
-	ctx->ring_sock->sk->sk_user_data = ctx;
 #endif
 	fd_install(ret, file);
 	return ret;
@@ -3795,24 +7812,48 @@ err:
 	return ret;
 }
 
-static int io_uring_create(unsigned entries, struct io_uring_params *p)
+static int io_uring_create(unsigned entries, struct io_uring_params *p,
+			   struct io_uring_params __user *params)
 {
 	struct user_struct *user = NULL;
 	struct io_ring_ctx *ctx;
 	bool account_mem;
 	int ret;
 
-	if (!entries || entries > IORING_MAX_ENTRIES)
+	if (!entries)
 		return -EINVAL;
+	if (entries > IORING_MAX_ENTRIES) {
+		if (!(p->flags & IORING_SETUP_CLAMP))
+			return -EINVAL;
+		entries = IORING_MAX_ENTRIES;
+	}
 
 	/*
 	 * Use twice as many entries for the CQ ring. It's possible for the
 	 * application to drive a higher depth than the size of the SQ ring,
 	 * since the sqes are only used at submission time. This allows for
-	 * some flexibility in overcommitting a bit.
+	 * some flexibility in overcommitting a bit. If the application has
+	 * set IORING_SETUP_CQSIZE, it will have passed in the desired number
+	 * of CQ ring entries manually.
 	 */
 	p->sq_entries = roundup_pow_of_two(entries);
-	p->cq_entries = 2 * p->sq_entries;
+	if (p->flags & IORING_SETUP_CQSIZE) {
+		/*
+		 * If IORING_SETUP_CQSIZE is set, we do the same roundup
+		 * to a power-of-two, if it isn't already. We do NOT impose
+		 * any cq vs sq ring sizing.
+		 */
+		if (p->cq_entries < p->sq_entries)
+			return -EINVAL;
+		if (p->cq_entries > IORING_MAX_CQ_ENTRIES) {
+			if (!(p->flags & IORING_SETUP_CLAMP))
+				return -EINVAL;
+			p->cq_entries = IORING_MAX_CQ_ENTRIES;
+		}
+		p->cq_entries = roundup_pow_of_two(p->cq_entries);
+	} else {
+		p->cq_entries = 2 * p->sq_entries;
+	}
 
 	user = get_uid(current_user());
 	account_mem = !capable(CAP_IPC_LOCK);
@@ -3837,6 +7878,7 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p)
 	ctx->compat = in_compat_syscall();
 	ctx->account_mem = account_mem;
 	ctx->user = user;
+	ctx->creds = get_current_cred();
 
 	ret = io_allocate_scq_urings(ctx, p);
 	if (ret)
@@ -3862,7 +7904,16 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p)
 	p->cq_off.ring_entries = offsetof(struct io_rings, cq_ring_entries);
 	p->cq_off.overflow = offsetof(struct io_rings, cq_overflow);
 	p->cq_off.cqes = offsetof(struct io_rings, cqes);
+	p->cq_off.flags = offsetof(struct io_rings, cq_flags);
 
+	p->features = IORING_FEAT_SINGLE_MMAP | IORING_FEAT_NODROP |
+			IORING_FEAT_SUBMIT_STABLE | IORING_FEAT_RW_CUR_POS |
+			IORING_FEAT_CUR_PERSONALITY | IORING_FEAT_FAST_POLL;
+
+	if (copy_to_user(params, p, sizeof(*p))) {
+		ret = -EFAULT;
+		goto err;
+	}
 	/*
 	 * Install ring fd as the very last thing, so we don't risk someone
 	 * having closed it before we finish setup
@@ -3871,7 +7922,7 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p)
 	if (ret < 0)
 		goto err;
 
-	p->features = IORING_FEAT_SINGLE_MMAP;
+	trace_io_uring_create(ret, ctx, p->sq_entries, p->cq_entries, p->flags);
 	return ret;
 err:
 	io_ring_ctx_wait_and_kill(ctx);
@@ -3886,7 +7937,6 @@ err:
 static long io_uring_setup(u32 entries, struct io_uring_params __user *params)
 {
 	struct io_uring_params p;
-	long ret;
 	int i;
 
 	if (copy_from_user(&p, params, sizeof(p)))
@@ -3897,17 +7947,11 @@ static long io_uring_setup(u32 entries, struct io_uring_params __user *params)
 	}
 
 	if (p.flags & ~(IORING_SETUP_IOPOLL | IORING_SETUP_SQPOLL |
-			IORING_SETUP_SQ_AFF))
+			IORING_SETUP_SQ_AFF | IORING_SETUP_CQSIZE |
+			IORING_SETUP_CLAMP | IORING_SETUP_ATTACH_WQ))
 		return -EINVAL;
 
-	ret = io_uring_create(entries, &p);
-	if (ret < 0)
-		return ret;
-
-	if (copy_to_user(params, &p, sizeof(p)))
-		return -EFAULT;
-
-	return ret;
+	return  io_uring_create(entries, &p, params);
 }
 
 SYSCALL_DEFINE2(io_uring_setup, u32, entries,
@@ -3916,6 +7960,84 @@ SYSCALL_DEFINE2(io_uring_setup, u32, entries,
 	return io_uring_setup(entries, params);
 }
 
+static int io_probe(struct io_ring_ctx *ctx, void __user *arg, unsigned nr_args)
+{
+	struct io_uring_probe *p;
+	size_t size;
+	int i, ret;
+
+	size = struct_size(p, ops, nr_args);
+	if (size == SIZE_MAX)
+		return -EOVERFLOW;
+	p = kzalloc(size, GFP_KERNEL);
+	if (!p)
+		return -ENOMEM;
+
+	ret = -EFAULT;
+	if (copy_from_user(p, arg, size))
+		goto out;
+	ret = -EINVAL;
+	if (memchr_inv(p, 0, size))
+		goto out;
+
+	p->last_op = IORING_OP_LAST - 1;
+	if (nr_args > IORING_OP_LAST)
+		nr_args = IORING_OP_LAST;
+
+	for (i = 0; i < nr_args; i++) {
+		p->ops[i].op = i;
+		if (!io_op_defs[i].not_supported)
+			p->ops[i].flags = IO_URING_OP_SUPPORTED;
+	}
+	p->ops_len = i;
+
+	ret = 0;
+	if (copy_to_user(arg, p, size))
+		ret = -EFAULT;
+out:
+	kfree(p);
+	return ret;
+}
+
+static int io_register_personality(struct io_ring_ctx *ctx)
+{
+	const struct cred *creds = get_current_cred();
+	int id;
+
+	id = idr_alloc_cyclic(&ctx->personality_idr, (void *) creds, 1,
+				USHRT_MAX, GFP_KERNEL);
+	if (id < 0)
+		put_cred(creds);
+	return id;
+}
+
+static int io_unregister_personality(struct io_ring_ctx *ctx, unsigned id)
+{
+	const struct cred *old_creds;
+
+	old_creds = idr_remove(&ctx->personality_idr, id);
+	if (old_creds) {
+		put_cred(old_creds);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static bool io_register_op_must_quiesce(int op)
+{
+	switch (op) {
+	case IORING_UNREGISTER_FILES:
+	case IORING_REGISTER_FILES_UPDATE:
+	case IORING_REGISTER_PROBE:
+	case IORING_REGISTER_PERSONALITY:
+	case IORING_UNREGISTER_PERSONALITY:
+		return false;
+	default:
+		return true;
+	}
+}
+
 static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
 			       void __user *arg, unsigned nr_args)
 	__releases(ctx->uring_lock)
@@ -3931,18 +8053,26 @@ static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
 	if (percpu_ref_is_dying(&ctx->refs))
 		return -ENXIO;
 
-	percpu_ref_kill(&ctx->refs);
+	if (io_register_op_must_quiesce(opcode)) {
+		percpu_ref_kill(&ctx->refs);
 
-	/*
-	 * Drop uring mutex before waiting for references to exit. If another
-	 * thread is currently inside io_uring_enter() it might need to grab
-	 * the uring_lock to make progress. If we hold it here across the drain
-	 * wait, then we can deadlock. It's safe to drop the mutex here, since
-	 * no new references will come in after we've killed the percpu ref.
-	 */
-	mutex_unlock(&ctx->uring_lock);
-	wait_for_completion(&ctx->ctx_done);
-	mutex_lock(&ctx->uring_lock);
+		/*
+		 * Drop uring mutex before waiting for references to exit. If
+		 * another thread is currently inside io_uring_enter() it might
+		 * need to grab the uring_lock to make progress. If we hold it
+		 * here across the drain wait, then we can deadlock. It's safe
+		 * to drop the mutex here, since no new references will come in
+		 * after we've killed the percpu ref.
+		 */
+		mutex_unlock(&ctx->uring_lock);
+		ret = wait_for_completion_interruptible(&ctx->ref_comp);
+		mutex_lock(&ctx->uring_lock);
+		if (ret) {
+			percpu_ref_resurrect(&ctx->refs);
+			ret = -EINTR;
+			goto out;
+		}
+	}
 
 	switch (opcode) {
 	case IORING_REGISTER_BUFFERS:
@@ -3963,11 +8093,21 @@ static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
 			break;
 		ret = io_sqe_files_unregister(ctx);
 		break;
+	case IORING_REGISTER_FILES_UPDATE:
+		ret = io_sqe_files_update(ctx, arg, nr_args);
+		break;
 	case IORING_REGISTER_EVENTFD:
+	case IORING_REGISTER_EVENTFD_ASYNC:
 		ret = -EINVAL;
 		if (nr_args != 1)
 			break;
 		ret = io_eventfd_register(ctx, arg);
+		if (ret)
+			break;
+		if (opcode == IORING_REGISTER_EVENTFD_ASYNC)
+			ctx->eventfd_async = 1;
+		else
+			ctx->eventfd_async = 0;
 		break;
 	case IORING_UNREGISTER_EVENTFD:
 		ret = -EINVAL;
@@ -3975,14 +8115,35 @@ static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
 			break;
 		ret = io_eventfd_unregister(ctx);
 		break;
+	case IORING_REGISTER_PROBE:
+		ret = -EINVAL;
+		if (!arg || nr_args > 256)
+			break;
+		ret = io_probe(ctx, arg, nr_args);
+		break;
+	case IORING_REGISTER_PERSONALITY:
+		ret = -EINVAL;
+		if (arg || nr_args)
+			break;
+		ret = io_register_personality(ctx);
+		break;
+	case IORING_UNREGISTER_PERSONALITY:
+		ret = -EINVAL;
+		if (arg)
+			break;
+		ret = io_unregister_personality(ctx, nr_args);
+		break;
 	default:
 		ret = -EINVAL;
 		break;
 	}
 
-	/* bring the ctx back to life */
-	reinit_completion(&ctx->ctx_done);
-	percpu_ref_reinit(&ctx->refs);
+	if (io_register_op_must_quiesce(opcode)) {
+		/* bring the ctx back to life */
+		percpu_ref_reinit(&ctx->refs);
+out:
+		reinit_completion(&ctx->ref_comp);
+	}
 	return ret;
 }
 
@@ -4006,6 +8167,8 @@ SYSCALL_DEFINE4(io_uring_register, unsigned int, fd, unsigned int, opcode,
 	mutex_lock(&ctx->uring_lock);
 	ret = __io_uring_register(ctx, opcode, arg, nr_args);
 	mutex_unlock(&ctx->uring_lock);
+	trace_io_uring_register(ctx, opcode, ctx->nr_user_files, ctx->nr_user_bufs,
+							ctx->cq_ev_fd != NULL, ret);
 out_fput:
 	fdput(f);
 	return ret;
@@ -4013,6 +8176,44 @@ out_fput:
 
 static int __init io_uring_init(void)
 {
+#define __BUILD_BUG_VERIFY_ELEMENT(stype, eoffset, etype, ename) do { \
+	BUILD_BUG_ON(offsetof(stype, ename) != eoffset); \
+	BUILD_BUG_ON(sizeof(etype) != sizeof_field(stype, ename)); \
+} while (0)
+
+#define BUILD_BUG_SQE_ELEM(eoffset, etype, ename) \
+	__BUILD_BUG_VERIFY_ELEMENT(struct io_uring_sqe, eoffset, etype, ename)
+	BUILD_BUG_ON(sizeof(struct io_uring_sqe) != 64);
+	BUILD_BUG_SQE_ELEM(0,  __u8,   opcode);
+	BUILD_BUG_SQE_ELEM(1,  __u8,   flags);
+	BUILD_BUG_SQE_ELEM(2,  __u16,  ioprio);
+	BUILD_BUG_SQE_ELEM(4,  __s32,  fd);
+	BUILD_BUG_SQE_ELEM(8,  __u64,  off);
+	BUILD_BUG_SQE_ELEM(8,  __u64,  addr2);
+	BUILD_BUG_SQE_ELEM(16, __u64,  addr);
+	BUILD_BUG_SQE_ELEM(16, __u64,  splice_off_in);
+	BUILD_BUG_SQE_ELEM(24, __u32,  len);
+	BUILD_BUG_SQE_ELEM(28,     __kernel_rwf_t, rw_flags);
+	BUILD_BUG_SQE_ELEM(28, /* compat */   int, rw_flags);
+	BUILD_BUG_SQE_ELEM(28, /* compat */ __u32, rw_flags);
+	BUILD_BUG_SQE_ELEM(28, __u32,  fsync_flags);
+	BUILD_BUG_SQE_ELEM(28, __u16,  poll_events);
+	BUILD_BUG_SQE_ELEM(28, __u32,  sync_range_flags);
+	BUILD_BUG_SQE_ELEM(28, __u32,  msg_flags);
+	BUILD_BUG_SQE_ELEM(28, __u32,  timeout_flags);
+	BUILD_BUG_SQE_ELEM(28, __u32,  accept_flags);
+	BUILD_BUG_SQE_ELEM(28, __u32,  cancel_flags);
+	BUILD_BUG_SQE_ELEM(28, __u32,  open_flags);
+	BUILD_BUG_SQE_ELEM(28, __u32,  statx_flags);
+	BUILD_BUG_SQE_ELEM(28, __u32,  fadvise_advice);
+	BUILD_BUG_SQE_ELEM(28, __u32,  splice_flags);
+	BUILD_BUG_SQE_ELEM(32, __u64,  user_data);
+	BUILD_BUG_SQE_ELEM(40, __u16,  buf_index);
+	BUILD_BUG_SQE_ELEM(42, __u16,  personality);
+	BUILD_BUG_SQE_ELEM(44, __s32,  splice_fd_in);
+
+	BUILD_BUG_ON(ARRAY_SIZE(io_op_defs) != IORING_OP_LAST);
+	BUILD_BUG_ON(__REQ_F_LAST_BIT >= 8 * sizeof(int));
 	req_cachep = KMEM_CACHE(io_kiocb, SLAB_HWCACHE_ALIGN | SLAB_PANIC);
 	return 0;
 };
diff --git a/fs/ioctl.c b/fs/ioctl.c
index fef3a6bf7c78..5e80b40bc1b5 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -8,6 +8,7 @@
 #include <linux/syscalls.h>
 #include <linux/mm.h>
 #include <linux/capability.h>
+#include <linux/compat.h>
 #include <linux/file.h>
 #include <linux/fs.h>
 #include <linux/security.h>
@@ -53,19 +54,40 @@ EXPORT_SYMBOL(vfs_ioctl);
 
 static int ioctl_fibmap(struct file *filp, int __user *p)
 {
-	struct address_space *mapping = filp->f_mapping;
-	int res, block;
+	struct inode *inode = file_inode(filp);
+	struct super_block *sb = inode->i_sb;
+	int error, ur_block;
+	sector_t block;
 
-	/* do we support this mess? */
-	if (!mapping->a_ops->bmap)
-		return -EINVAL;
 	if (!capable(CAP_SYS_RAWIO))
 		return -EPERM;
-	res = get_user(block, p);
-	if (res)
-		return res;
-	res = mapping->a_ops->bmap(mapping, block);
-	return put_user(res, p);
+
+	error = get_user(ur_block, p);
+	if (error)
+		return error;
+
+	if (ur_block < 0)
+		return -EINVAL;
+
+	block = ur_block;
+	error = bmap(inode, &block);
+
+	if (block > INT_MAX) {
+		error = -ERANGE;
+		pr_warn_ratelimited("[%s/%d] FS: %s File: %pD4 would truncate fibmap result\n",
+				    current->comm, task_pid_nr(current),
+				    sb->s_id, filp);
+	}
+
+	if (error)
+		ur_block = 0;
+	else
+		ur_block = block;
+
+	if (put_user(ur_block, p))
+		error = -EFAULT;
+
+	return error;
 }
 
 /**
@@ -174,10 +196,9 @@ static int fiemap_check_ranges(struct super_block *sb,
 	return 0;
 }
 
-static int ioctl_fiemap(struct file *filp, unsigned long arg)
+static int ioctl_fiemap(struct file *filp, struct fiemap __user *ufiemap)
 {
 	struct fiemap fiemap;
-	struct fiemap __user *ufiemap = (struct fiemap __user *) arg;
 	struct fiemap_extent_info fieinfo = { 0, };
 	struct inode *inode = file_inode(filp);
 	struct super_block *sb = inode->i_sb;
@@ -244,7 +265,8 @@ fdput:
 	return ret;
 }
 
-static long ioctl_file_clone_range(struct file *file, void __user *argp)
+static long ioctl_file_clone_range(struct file *file,
+				   struct file_clone_range __user *argp)
 {
 	struct file_clone_range args;
 
@@ -466,7 +488,7 @@ EXPORT_SYMBOL(generic_block_fiemap);
  * Only the l_start, l_len and l_whence fields of the 'struct space_resv'
  * are used here, rest are ignored.
  */
-int ioctl_preallocate(struct file *filp, void __user *argp)
+static int ioctl_preallocate(struct file *filp, int mode, void __user *argp)
 {
 	struct inode *inode = file_inode(filp);
 	struct space_resv sr;
@@ -487,26 +509,55 @@ int ioctl_preallocate(struct file *filp, void __user *argp)
 		return -EINVAL;
 	}
 
-	return vfs_fallocate(filp, FALLOC_FL_KEEP_SIZE, sr.l_start, sr.l_len);
+	return vfs_fallocate(filp, mode | FALLOC_FL_KEEP_SIZE, sr.l_start,
+			sr.l_len);
 }
 
-static int file_ioctl(struct file *filp, unsigned int cmd,
-		unsigned long arg)
+/* on ia32 l_start is on a 32-bit boundary */
+#if defined CONFIG_COMPAT && defined(CONFIG_X86_64)
+/* just account for different alignment */
+static int compat_ioctl_preallocate(struct file *file, int mode,
+				    struct space_resv_32 __user *argp)
 {
-	struct inode *inode = file_inode(filp);
-	int __user *p = (int __user *)arg;
+	struct inode *inode = file_inode(file);
+	struct space_resv_32 sr;
+
+	if (copy_from_user(&sr, argp, sizeof(sr)))
+		return -EFAULT;
 
+	switch (sr.l_whence) {
+	case SEEK_SET:
+		break;
+	case SEEK_CUR:
+		sr.l_start += file->f_pos;
+		break;
+	case SEEK_END:
+		sr.l_start += i_size_read(inode);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return vfs_fallocate(file, mode | FALLOC_FL_KEEP_SIZE, sr.l_start, sr.l_len);
+}
+#endif
+
+static int file_ioctl(struct file *filp, unsigned int cmd, int __user *p)
+{
 	switch (cmd) {
 	case FIBMAP:
 		return ioctl_fibmap(filp, p);
-	case FIONREAD:
-		return put_user(i_size_read(inode) - filp->f_pos, p);
 	case FS_IOC_RESVSP:
 	case FS_IOC_RESVSP64:
-		return ioctl_preallocate(filp, p);
+		return ioctl_preallocate(filp, 0, p);
+	case FS_IOC_UNRESVSP:
+	case FS_IOC_UNRESVSP64:
+		return ioctl_preallocate(filp, FALLOC_FL_PUNCH_HOLE, p);
+	case FS_IOC_ZERO_RANGE:
+		return ioctl_preallocate(filp, FALLOC_FL_ZERO_RANGE, p);
 	}
 
-	return vfs_ioctl(filp, cmd, arg);
+	return -ENOIOCTLCMD;
 }
 
 static int ioctl_fionbio(struct file *filp, int __user *argp)
@@ -584,9 +635,9 @@ static int ioctl_fsthaw(struct file *filp)
 	return thaw_super(sb);
 }
 
-static int ioctl_file_dedupe_range(struct file *file, void __user *arg)
+static int ioctl_file_dedupe_range(struct file *file,
+				   struct file_dedupe_range __user *argp)
 {
-	struct file_dedupe_range __user *argp = arg;
 	struct file_dedupe_range *same = NULL;
 	int ret;
 	unsigned long size;
@@ -625,62 +676,58 @@ out:
 }
 
 /*
- * When you add any new common ioctls to the switches above and below
- * please update compat_sys_ioctl() too.
- *
  * do_vfs_ioctl() is not for drivers and not intended to be EXPORT_SYMBOL()'d.
  * It's just a simple helper for sys_ioctl and compat_sys_ioctl.
+ *
+ * When you add any new common ioctls to the switches above and below,
+ * please ensure they have compatible arguments in compat mode.
  */
-int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
-	     unsigned long arg)
+static int do_vfs_ioctl(struct file *filp, unsigned int fd,
+			unsigned int cmd, unsigned long arg)
 {
-	int error = 0;
-	int __user *argp = (int __user *)arg;
+	void __user *argp = (void __user *)arg;
 	struct inode *inode = file_inode(filp);
 
 	switch (cmd) {
 	case FIOCLEX:
 		set_close_on_exec(fd, 1);
-		break;
+		return 0;
 
 	case FIONCLEX:
 		set_close_on_exec(fd, 0);
-		break;
+		return 0;
 
 	case FIONBIO:
-		error = ioctl_fionbio(filp, argp);
-		break;
+		return ioctl_fionbio(filp, argp);
 
 	case FIOASYNC:
-		error = ioctl_fioasync(fd, filp, argp);
-		break;
+		return ioctl_fioasync(fd, filp, argp);
 
 	case FIOQSIZE:
 		if (S_ISDIR(inode->i_mode) || S_ISREG(inode->i_mode) ||
 		    S_ISLNK(inode->i_mode)) {
 			loff_t res = inode_get_bytes(inode);
-			error = copy_to_user(argp, &res, sizeof(res)) ?
-					-EFAULT : 0;
-		} else
-			error = -ENOTTY;
-		break;
+			return copy_to_user(argp, &res, sizeof(res)) ?
+					    -EFAULT : 0;
+		}
+
+		return -ENOTTY;
 
 	case FIFREEZE:
-		error = ioctl_fsfreeze(filp);
-		break;
+		return ioctl_fsfreeze(filp);
 
 	case FITHAW:
-		error = ioctl_fsthaw(filp);
-		break;
+		return ioctl_fsthaw(filp);
 
 	case FS_IOC_FIEMAP:
-		return ioctl_fiemap(filp, arg);
+		return ioctl_fiemap(filp, argp);
 
 	case FIGETBSZ:
 		/* anon_bdev filesystems may not have a block size */
 		if (!inode->i_sb->s_blocksize)
 			return -EINVAL;
-		return put_user(inode->i_sb->s_blocksize, argp);
+
+		return put_user(inode->i_sb->s_blocksize, (int __user *)argp);
 
 	case FICLONE:
 		return ioctl_file_clone(filp, arg, 0, 0, 0);
@@ -691,26 +738,39 @@ int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
 	case FIDEDUPERANGE:
 		return ioctl_file_dedupe_range(filp, argp);
 
+	case FIONREAD:
+		if (!S_ISREG(inode->i_mode))
+			return vfs_ioctl(filp, cmd, arg);
+
+		return put_user(i_size_read(inode) - filp->f_pos,
+				(int __user *)argp);
+
 	default:
 		if (S_ISREG(inode->i_mode))
-			error = file_ioctl(filp, cmd, arg);
-		else
-			error = vfs_ioctl(filp, cmd, arg);
+			return file_ioctl(filp, cmd, argp);
 		break;
 	}
-	return error;
+
+	return -ENOIOCTLCMD;
 }
 
 int ksys_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
-	int error;
 	struct fd f = fdget(fd);
+	int error;
 
 	if (!f.file)
 		return -EBADF;
+
 	error = security_file_ioctl(f.file, cmd, arg);
-	if (!error)
-		error = do_vfs_ioctl(f.file, fd, cmd, arg);
+	if (error)
+		goto out;
+
+	error = do_vfs_ioctl(f.file, fd, cmd, arg);
+	if (error == -ENOIOCTLCMD)
+		error = vfs_ioctl(f.file, cmd, arg);
+
+out:
 	fdput(f);
 	return error;
 }
@@ -719,3 +779,98 @@ SYSCALL_DEFINE3(ioctl, unsigned int, fd, unsigned int, cmd, unsigned long, arg)
 {
 	return ksys_ioctl(fd, cmd, arg);
 }
+
+#ifdef CONFIG_COMPAT
+/**
+ * compat_ptr_ioctl - generic implementation of .compat_ioctl file operation
+ *
+ * This is not normally called as a function, but instead set in struct
+ * file_operations as
+ *
+ *     .compat_ioctl = compat_ptr_ioctl,
+ *
+ * On most architectures, the compat_ptr_ioctl() just passes all arguments
+ * to the corresponding ->ioctl handler. The exception is arch/s390, where
+ * compat_ptr() clears the top bit of a 32-bit pointer value, so user space
+ * pointers to the second 2GB alias the first 2GB, as is the case for
+ * native 32-bit s390 user space.
+ *
+ * The compat_ptr_ioctl() function must therefore be used only with ioctl
+ * functions that either ignore the argument or pass a pointer to a
+ * compatible data type.
+ *
+ * If any ioctl command handled by fops->unlocked_ioctl passes a plain
+ * integer instead of a pointer, or any of the passed data types
+ * is incompatible between 32-bit and 64-bit architectures, a proper
+ * handler is required instead of compat_ptr_ioctl.
+ */
+long compat_ptr_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	if (!file->f_op->unlocked_ioctl)
+		return -ENOIOCTLCMD;
+
+	return file->f_op->unlocked_ioctl(file, cmd, (unsigned long)compat_ptr(arg));
+}
+EXPORT_SYMBOL(compat_ptr_ioctl);
+
+COMPAT_SYSCALL_DEFINE3(ioctl, unsigned int, fd, unsigned int, cmd,
+		       compat_ulong_t, arg)
+{
+	struct fd f = fdget(fd);
+	int error;
+
+	if (!f.file)
+		return -EBADF;
+
+	/* RED-PEN how should LSM module know it's handling 32bit? */
+	error = security_file_ioctl(f.file, cmd, arg);
+	if (error)
+		goto out;
+
+	switch (cmd) {
+	/* FICLONE takes an int argument, so don't use compat_ptr() */
+	case FICLONE:
+		error = ioctl_file_clone(f.file, arg, 0, 0, 0);
+		break;
+
+#if defined(CONFIG_X86_64)
+	/* these get messy on amd64 due to alignment differences */
+	case FS_IOC_RESVSP_32:
+	case FS_IOC_RESVSP64_32:
+		error = compat_ioctl_preallocate(f.file, 0, compat_ptr(arg));
+		break;
+	case FS_IOC_UNRESVSP_32:
+	case FS_IOC_UNRESVSP64_32:
+		error = compat_ioctl_preallocate(f.file, FALLOC_FL_PUNCH_HOLE,
+				compat_ptr(arg));
+		break;
+	case FS_IOC_ZERO_RANGE_32:
+		error = compat_ioctl_preallocate(f.file, FALLOC_FL_ZERO_RANGE,
+				compat_ptr(arg));
+		break;
+#endif
+
+	/*
+	 * everything else in do_vfs_ioctl() takes either a compatible
+	 * pointer argument or no argument -- call it with a modified
+	 * argument.
+	 */
+	default:
+		error = do_vfs_ioctl(f.file, fd, cmd,
+				     (unsigned long)compat_ptr(arg));
+		if (error != -ENOIOCTLCMD)
+			break;
+
+		if (f.file->f_op->compat_ioctl)
+			error = f.file->f_op->compat_ioctl(f.file, cmd, arg);
+		if (error == -ENOIOCTLCMD)
+			error = -ENOTTY;
+		break;
+	}
+
+ out:
+	fdput(f);
+
+	return error;
+}
+#endif
diff --git a/fs/iomap/Makefile b/fs/iomap/Makefile
index 93cd11938bf5..eef2722d93a1 100644
--- a/fs/iomap/Makefile
+++ b/fs/iomap/Makefile
@@ -3,13 +3,15 @@
 # Copyright (c) 2019 Oracle.
 # All Rights Reserved.
 #
-obj-$(CONFIG_FS_IOMAP)		+= iomap.o
 
-iomap-y				+= \
-					apply.o \
-					buffered-io.o \
-					direct-io.o \
-					fiemap.o \
-					seek.o
+ccflags-y += -I $(srctree)/$(src)		# needed for trace events
+
+obj-$(CONFIG_FS_IOMAP)		+= iomap.o
 
+iomap-y				+= trace.o \
+				   apply.o \
+				   buffered-io.o \
+				   direct-io.o \
+				   fiemap.o \
+				   seek.o
 iomap-$(CONFIG_SWAP)		+= swapfile.o
diff --git a/fs/iomap/apply.c b/fs/iomap/apply.c
index 54c02aecf3cd..76925b40b5fd 100644
--- a/fs/iomap/apply.c
+++ b/fs/iomap/apply.c
@@ -7,6 +7,7 @@
 #include <linux/compiler.h>
 #include <linux/fs.h>
 #include <linux/iomap.h>
+#include "trace.h"
 
 /*
  * Execute a iomap write on a segment of the mapping that spans a
@@ -23,8 +24,12 @@ loff_t
 iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags,
 		const struct iomap_ops *ops, void *data, iomap_actor_t actor)
 {
-	struct iomap iomap = { 0 };
+	struct iomap iomap = { .type = IOMAP_HOLE };
+	struct iomap srcmap = { .type = IOMAP_HOLE };
 	loff_t written = 0, ret;
+	u64 end;
+
+	trace_iomap_apply(inode, pos, length, flags, ops, actor, _RET_IP_);
 
 	/*
 	 * Need to map a range from start position for length bytes. This can
@@ -38,7 +43,7 @@ iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags,
 	 * expose transient stale data. If the reserve fails, we can safely
 	 * back out at this point as there is nothing to undo.
 	 */
-	ret = ops->iomap_begin(inode, pos, length, flags, &iomap);
+	ret = ops->iomap_begin(inode, pos, length, flags, &iomap, &srcmap);
 	if (ret)
 		return ret;
 	if (WARN_ON(iomap.offset > pos))
@@ -46,19 +51,34 @@ iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags,
 	if (WARN_ON(iomap.length == 0))
 		return -EIO;
 
+	trace_iomap_apply_dstmap(inode, &iomap);
+	if (srcmap.type != IOMAP_HOLE)
+		trace_iomap_apply_srcmap(inode, &srcmap);
+
 	/*
 	 * Cut down the length to the one actually provided by the filesystem,
 	 * as it might not be able to give us the whole size that we requested.
 	 */
-	if (iomap.offset + iomap.length < pos + length)
-		length = iomap.offset + iomap.length - pos;
+	end = iomap.offset + iomap.length;
+	if (srcmap.type != IOMAP_HOLE)
+		end = min(end, srcmap.offset + srcmap.length);
+	if (pos + length > end)
+		length = end - pos;
 
 	/*
-	 * Now that we have guaranteed that the space allocation will succeed.
+	 * Now that we have guaranteed that the space allocation will succeed,
 	 * we can do the copy-in page by page without having to worry about
 	 * failures exposing transient data.
+	 *
+	 * To support COW operations, we read in data for partially blocks from
+	 * the srcmap if the file system filled it in.  In that case we the
+	 * length needs to be limited to the earlier of the ends of the iomaps.
+	 * If the file system did not provide a srcmap we pass in the normal
+	 * iomap into the actors so that they don't need to have special
+	 * handling for the two cases.
 	 */
-	written = actor(inode, pos, length, data, &iomap);
+	written = actor(inode, pos, length, data, &iomap,
+			srcmap.type != IOMAP_HOLE ? &srcmap : &iomap);
 
 	/*
 	 * Now the data has been copied, commit the range we've copied.  This
diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c
index e25901ae3ff4..a1ed7620fbac 100644
--- a/fs/iomap/buffered-io.c
+++ b/fs/iomap/buffered-io.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /*
  * Copyright (C) 2010 Red Hat, Inc.
- * Copyright (c) 2016-2018 Christoph Hellwig.
+ * Copyright (C) 2016-2019 Christoph Hellwig.
  */
 #include <linux/module.h>
 #include <linux/compiler.h>
@@ -12,13 +12,35 @@
 #include <linux/buffer_head.h>
 #include <linux/dax.h>
 #include <linux/writeback.h>
+#include <linux/list_sort.h>
 #include <linux/swap.h>
 #include <linux/bio.h>
 #include <linux/sched/signal.h>
 #include <linux/migrate.h>
+#include "trace.h"
 
 #include "../internal.h"
 
+/*
+ * Structure allocated for each page when block size < PAGE_SIZE to track
+ * sub-page uptodate status and I/O completions.
+ */
+struct iomap_page {
+	atomic_t		read_count;
+	atomic_t		write_count;
+	spinlock_t		uptodate_lock;
+	DECLARE_BITMAP(uptodate, PAGE_SIZE / 512);
+};
+
+static inline struct iomap_page *to_iomap_page(struct page *page)
+{
+	if (page_has_private(page))
+		return (struct iomap_page *)page_private(page);
+	return NULL;
+}
+
+static struct bio_set iomap_ioend_bioset;
+
 static struct iomap_page *
 iomap_page_create(struct inode *inode, struct page *page)
 {
@@ -30,30 +52,26 @@ iomap_page_create(struct inode *inode, struct page *page)
 	iop = kmalloc(sizeof(*iop), GFP_NOFS | __GFP_NOFAIL);
 	atomic_set(&iop->read_count, 0);
 	atomic_set(&iop->write_count, 0);
+	spin_lock_init(&iop->uptodate_lock);
 	bitmap_zero(iop->uptodate, PAGE_SIZE / SECTOR_SIZE);
 
 	/*
 	 * migrate_page_move_mapping() assumes that pages with private data have
 	 * their count elevated by 1.
 	 */
-	get_page(page);
-	set_page_private(page, (unsigned long)iop);
-	SetPagePrivate(page);
+	attach_page_private(page, iop);
 	return iop;
 }
 
 static void
 iomap_page_release(struct page *page)
 {
-	struct iomap_page *iop = to_iomap_page(page);
+	struct iomap_page *iop = detach_page_private(page);
 
 	if (!iop)
 		return;
 	WARN_ON_ONCE(atomic_read(&iop->read_count));
 	WARN_ON_ONCE(atomic_read(&iop->write_count));
-	ClearPagePrivate(page);
-	set_page_private(page, 0);
-	put_page(page);
 	kfree(iop);
 }
 
@@ -118,25 +136,38 @@ iomap_adjust_read_range(struct inode *inode, struct iomap_page *iop,
 }
 
 static void
-iomap_set_range_uptodate(struct page *page, unsigned off, unsigned len)
+iomap_iop_set_range_uptodate(struct page *page, unsigned off, unsigned len)
 {
 	struct iomap_page *iop = to_iomap_page(page);
 	struct inode *inode = page->mapping->host;
 	unsigned first = off >> inode->i_blkbits;
 	unsigned last = (off + len - 1) >> inode->i_blkbits;
-	unsigned int i;
 	bool uptodate = true;
+	unsigned long flags;
+	unsigned int i;
 
-	if (iop) {
-		for (i = 0; i < PAGE_SIZE / i_blocksize(inode); i++) {
-			if (i >= first && i <= last)
-				set_bit(i, iop->uptodate);
-			else if (!test_bit(i, iop->uptodate))
-				uptodate = false;
-		}
+	spin_lock_irqsave(&iop->uptodate_lock, flags);
+	for (i = 0; i < PAGE_SIZE / i_blocksize(inode); i++) {
+		if (i >= first && i <= last)
+			set_bit(i, iop->uptodate);
+		else if (!test_bit(i, iop->uptodate))
+			uptodate = false;
 	}
 
-	if (uptodate && !PageError(page))
+	if (uptodate)
+		SetPageUptodate(page);
+	spin_unlock_irqrestore(&iop->uptodate_lock, flags);
+}
+
+static void
+iomap_set_range_uptodate(struct page *page, unsigned off, unsigned len)
+{
+	if (PageError(page))
+		return;
+
+	if (page_has_private(page))
+		iomap_iop_set_range_uptodate(page, off, len);
+	else
 		SetPageUptodate(page);
 }
 
@@ -178,9 +209,8 @@ iomap_read_end_io(struct bio *bio)
 struct iomap_readpage_ctx {
 	struct page		*cur_page;
 	bool			cur_page_in_bio;
-	bool			is_readahead;
 	struct bio		*bio;
-	struct list_head	*pages;
+	struct readahead_control *rac;
 };
 
 static void
@@ -203,9 +233,17 @@ iomap_read_inline_data(struct inode *inode, struct page *page,
 	SetPageUptodate(page);
 }
 
+static inline bool iomap_block_needs_zeroing(struct inode *inode,
+		struct iomap *iomap, loff_t pos)
+{
+	return iomap->type != IOMAP_MAPPED ||
+		(iomap->flags & IOMAP_F_NEW) ||
+		pos >= i_size_read(inode);
+}
+
 static loff_t
 iomap_readpage_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
-		struct iomap *iomap)
+		struct iomap *iomap, struct iomap *srcmap)
 {
 	struct iomap_readpage_ctx *ctx = data;
 	struct page *page = ctx->cur_page;
@@ -226,7 +264,7 @@ iomap_readpage_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
 	if (plen == 0)
 		goto done;
 
-	if (iomap->type != IOMAP_MAPPED || pos >= i_size_read(inode)) {
+	if (iomap_block_needs_zeroing(inode, iomap, pos)) {
 		zero_user(page, poff, plen);
 		iomap_set_range_uptodate(page, poff, plen);
 		goto done;
@@ -258,16 +296,24 @@ iomap_readpage_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
 
 	if (!ctx->bio || !is_contig || bio_full(ctx->bio, plen)) {
 		gfp_t gfp = mapping_gfp_constraint(page->mapping, GFP_KERNEL);
+		gfp_t orig_gfp = gfp;
 		int nr_vecs = (length + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
 		if (ctx->bio)
 			submit_bio(ctx->bio);
 
-		if (ctx->is_readahead) /* same as readahead_gfp_mask */
+		if (ctx->rac) /* same as readahead_gfp_mask */
 			gfp |= __GFP_NORETRY | __GFP_NOWARN;
 		ctx->bio = bio_alloc(gfp, min(BIO_MAX_PAGES, nr_vecs));
+		/*
+		 * If the bio_alloc fails, try it again for a single page to
+		 * avoid having to deal with partial page reads.  This emulates
+		 * what do_mpage_readpage does.
+		 */
+		if (!ctx->bio)
+			ctx->bio = bio_alloc(orig_gfp, 1);
 		ctx->bio->bi_opf = REQ_OP_READ;
-		if (ctx->is_readahead)
+		if (ctx->rac)
 			ctx->bio->bi_opf |= REQ_RAHEAD;
 		ctx->bio->bi_iter.bi_sector = sector;
 		bio_set_dev(ctx->bio, iomap->bdev);
@@ -293,6 +339,8 @@ iomap_readpage(struct page *page, const struct iomap_ops *ops)
 	unsigned poff;
 	loff_t ret;
 
+	trace_iomap_readpage(page->mapping->host, 1);
+
 	for (poff = 0; poff < PAGE_SIZE; poff += ret) {
 		ret = iomap_apply(inode, page_offset(page) + poff,
 				PAGE_SIZE - poff, 0, ops, &ctx,
@@ -313,7 +361,7 @@ iomap_readpage(struct page *page, const struct iomap_ops *ops)
 	}
 
 	/*
-	 * Just like mpage_readpages and block_read_full_page we always
+	 * Just like mpage_readahead and block_read_full_page we always
 	 * return 0 and just mark the page as PageError on errors.  This
 	 * should be cleaned up all through the stack eventually.
 	 */
@@ -321,37 +369,9 @@ iomap_readpage(struct page *page, const struct iomap_ops *ops)
 }
 EXPORT_SYMBOL_GPL(iomap_readpage);
 
-static struct page *
-iomap_next_page(struct inode *inode, struct list_head *pages, loff_t pos,
-		loff_t length, loff_t *done)
-{
-	while (!list_empty(pages)) {
-		struct page *page = lru_to_page(pages);
-
-		if (page_offset(page) >= (u64)pos + length)
-			break;
-
-		list_del(&page->lru);
-		if (!add_to_page_cache_lru(page, inode->i_mapping, page->index,
-				GFP_NOFS))
-			return page;
-
-		/*
-		 * If we already have a page in the page cache at index we are
-		 * done.  Upper layers don't care if it is uptodate after the
-		 * readpages call itself as every page gets checked again once
-		 * actually needed.
-		 */
-		*done += PAGE_SIZE;
-		put_page(page);
-	}
-
-	return NULL;
-}
-
 static loff_t
-iomap_readpages_actor(struct inode *inode, loff_t pos, loff_t length,
-		void *data, struct iomap *iomap)
+iomap_readahead_actor(struct inode *inode, loff_t pos, loff_t length,
+		void *data, struct iomap *iomap, struct iomap *srcmap)
 {
 	struct iomap_readpage_ctx *ctx = data;
 	loff_t done, ret;
@@ -364,43 +384,53 @@ iomap_readpages_actor(struct inode *inode, loff_t pos, loff_t length,
 			ctx->cur_page = NULL;
 		}
 		if (!ctx->cur_page) {
-			ctx->cur_page = iomap_next_page(inode, ctx->pages,
-					pos, length, &done);
-			if (!ctx->cur_page)
-				break;
+			ctx->cur_page = readahead_page(ctx->rac);
 			ctx->cur_page_in_bio = false;
 		}
 		ret = iomap_readpage_actor(inode, pos + done, length - done,
-				ctx, iomap);
+				ctx, iomap, srcmap);
 	}
 
 	return done;
 }
 
-int
-iomap_readpages(struct address_space *mapping, struct list_head *pages,
-		unsigned nr_pages, const struct iomap_ops *ops)
+/**
+ * iomap_readahead - Attempt to read pages from a file.
+ * @rac: Describes the pages to be read.
+ * @ops: The operations vector for the filesystem.
+ *
+ * This function is for filesystems to call to implement their readahead
+ * address_space operation.
+ *
+ * Context: The @ops callbacks may submit I/O (eg to read the addresses of
+ * blocks from disc), and may wait for it.  The caller may be trying to
+ * access a different page, and so sleeping excessively should be avoided.
+ * It may allocate memory, but should avoid costly allocations.  This
+ * function is called with memalloc_nofs set, so allocations will not cause
+ * the filesystem to be reentered.
+ */
+void iomap_readahead(struct readahead_control *rac, const struct iomap_ops *ops)
 {
+	struct inode *inode = rac->mapping->host;
+	loff_t pos = readahead_pos(rac);
+	loff_t length = readahead_length(rac);
 	struct iomap_readpage_ctx ctx = {
-		.pages		= pages,
-		.is_readahead	= true,
+		.rac	= rac,
 	};
-	loff_t pos = page_offset(list_entry(pages->prev, struct page, lru));
-	loff_t last = page_offset(list_entry(pages->next, struct page, lru));
-	loff_t length = last - pos + PAGE_SIZE, ret = 0;
+
+	trace_iomap_readahead(inode, readahead_count(rac));
 
 	while (length > 0) {
-		ret = iomap_apply(mapping->host, pos, length, 0, ops,
-				&ctx, iomap_readpages_actor);
+		loff_t ret = iomap_apply(inode, pos, length, 0, ops,
+				&ctx, iomap_readahead_actor);
 		if (ret <= 0) {
 			WARN_ON_ONCE(ret == 0);
-			goto done;
+			break;
 		}
 		pos += ret;
 		length -= ret;
 	}
-	ret = 0;
-done:
+
 	if (ctx.bio)
 		submit_bio(ctx.bio);
 	if (ctx.cur_page) {
@@ -408,15 +438,8 @@ done:
 			unlock_page(ctx.cur_page);
 		put_page(ctx.cur_page);
 	}
-
-	/*
-	 * Check that we didn't lose a page due to the arcance calling
-	 * conventions..
-	 */
-	WARN_ON_ONCE(!ret && !list_empty(ctx.pages));
-	return ret;
 }
-EXPORT_SYMBOL_GPL(iomap_readpages);
+EXPORT_SYMBOL_GPL(iomap_readahead);
 
 /*
  * iomap_is_partially_uptodate checks whether blocks within a page are
@@ -455,6 +478,9 @@ EXPORT_SYMBOL_GPL(iomap_is_partially_uptodate);
 int
 iomap_releasepage(struct page *page, gfp_t gfp_mask)
 {
+	trace_iomap_releasepage(page->mapping->host, page_offset(page),
+			PAGE_SIZE);
+
 	/*
 	 * mm accommodates an old ext3 case where clean pages might not have had
 	 * the dirty bit cleared. Thus, it can send actual dirty pages to
@@ -470,6 +496,8 @@ EXPORT_SYMBOL_GPL(iomap_releasepage);
 void
 iomap_invalidatepage(struct page *page, unsigned int offset, unsigned int len)
 {
+	trace_iomap_invalidatepage(page->mapping->host, offset, len);
+
 	/*
 	 * If we are invalidating the entire page, clear the dirty state from it
 	 * and release it to avoid unnecessary buildup of the LRU.
@@ -493,14 +521,8 @@ iomap_migrate_page(struct address_space *mapping, struct page *newpage,
 	if (ret != MIGRATEPAGE_SUCCESS)
 		return ret;
 
-	if (page_has_private(page)) {
-		ClearPagePrivate(page);
-		get_page(newpage);
-		set_page_private(newpage, page_private(page));
-		set_page_private(page, 0);
-		put_page(page);
-		SetPagePrivate(newpage);
-	}
+	if (page_has_private(page))
+		attach_page_private(newpage, detach_page_private(page));
 
 	if (mode != MIGRATE_SYNC_NO_COPY)
 		migrate_page_copy(newpage, page);
@@ -511,6 +533,10 @@ iomap_migrate_page(struct address_space *mapping, struct page *newpage,
 EXPORT_SYMBOL_GPL(iomap_migrate_page);
 #endif /* CONFIG_MIGRATION */
 
+enum {
+	IOMAP_WRITE_F_UNSHARE		= (1 << 0),
+};
+
 static void
 iomap_write_failed(struct inode *inode, loff_t pos, unsigned len)
 {
@@ -525,19 +551,12 @@ iomap_write_failed(struct inode *inode, loff_t pos, unsigned len)
 }
 
 static int
-iomap_read_page_sync(struct inode *inode, loff_t block_start, struct page *page,
-		unsigned poff, unsigned plen, unsigned from, unsigned to,
-		struct iomap *iomap)
+iomap_read_page_sync(loff_t block_start, struct page *page, unsigned poff,
+		unsigned plen, struct iomap *iomap)
 {
 	struct bio_vec bvec;
 	struct bio bio;
 
-	if (iomap->type != IOMAP_MAPPED || block_start >= i_size_read(inode)) {
-		zero_user_segments(page, poff, from, to, poff + plen);
-		iomap_set_range_uptodate(page, poff, plen);
-		return 0;
-	}
-
 	bio_init(&bio, &bvec, 1);
 	bio.bi_opf = REQ_OP_READ;
 	bio.bi_iter.bi_sector = iomap_sector(iomap, block_start);
@@ -547,15 +566,15 @@ iomap_read_page_sync(struct inode *inode, loff_t block_start, struct page *page,
 }
 
 static int
-__iomap_write_begin(struct inode *inode, loff_t pos, unsigned len,
-		struct page *page, struct iomap *iomap)
+__iomap_write_begin(struct inode *inode, loff_t pos, unsigned len, int flags,
+		struct page *page, struct iomap *srcmap)
 {
 	struct iomap_page *iop = iomap_page_create(inode, page);
 	loff_t block_size = i_blocksize(inode);
 	loff_t block_start = pos & ~(block_size - 1);
 	loff_t block_end = (pos + len + block_size - 1) & ~(block_size - 1);
 	unsigned from = offset_in_page(pos), to = from + len, poff, plen;
-	int status = 0;
+	int status;
 
 	if (PageUptodate(page))
 		return 0;
@@ -566,29 +585,39 @@ __iomap_write_begin(struct inode *inode, loff_t pos, unsigned len,
 		if (plen == 0)
 			break;
 
-		if ((from > poff && from < poff + plen) ||
-		    (to > poff && to < poff + plen)) {
-			status = iomap_read_page_sync(inode, block_start, page,
-					poff, plen, from, to, iomap);
-			if (status)
-				break;
+		if (!(flags & IOMAP_WRITE_F_UNSHARE) &&
+		    (from <= poff || from >= poff + plen) &&
+		    (to <= poff || to >= poff + plen))
+			continue;
+
+		if (iomap_block_needs_zeroing(inode, srcmap, block_start)) {
+			if (WARN_ON_ONCE(flags & IOMAP_WRITE_F_UNSHARE))
+				return -EIO;
+			zero_user_segments(page, poff, from, to, poff + plen);
+			iomap_set_range_uptodate(page, poff, plen);
+			continue;
 		}
 
+		status = iomap_read_page_sync(block_start, page, poff, plen,
+				srcmap);
+		if (status)
+			return status;
 	} while ((block_start += plen) < block_end);
 
-	return status;
+	return 0;
 }
 
 static int
 iomap_write_begin(struct inode *inode, loff_t pos, unsigned len, unsigned flags,
-		struct page **pagep, struct iomap *iomap)
+		struct page **pagep, struct iomap *iomap, struct iomap *srcmap)
 {
 	const struct iomap_page_ops *page_ops = iomap->page_ops;
-	pgoff_t index = pos >> PAGE_SHIFT;
 	struct page *page;
 	int status = 0;
 
 	BUG_ON(pos + len > iomap->offset + iomap->length);
+	if (srcmap != iomap)
+		BUG_ON(pos + len > srcmap->offset + srcmap->length);
 
 	if (fatal_signal_pending(current))
 		return -EINTR;
@@ -599,18 +628,20 @@ iomap_write_begin(struct inode *inode, loff_t pos, unsigned len, unsigned flags,
 			return status;
 	}
 
-	page = grab_cache_page_write_begin(inode->i_mapping, index, flags);
+	page = grab_cache_page_write_begin(inode->i_mapping, pos >> PAGE_SHIFT,
+			AOP_FLAG_NOFS);
 	if (!page) {
 		status = -ENOMEM;
 		goto out_no_page;
 	}
 
-	if (iomap->type == IOMAP_INLINE)
-		iomap_read_inline_data(inode, page, iomap);
+	if (srcmap->type == IOMAP_INLINE)
+		iomap_read_inline_data(inode, page, srcmap);
 	else if (iomap->flags & IOMAP_F_BUFFER_HEAD)
-		status = __block_write_begin_int(page, pos, len, NULL, iomap);
+		status = __block_write_begin_int(page, pos, len, NULL, srcmap);
 	else
-		status = __iomap_write_begin(inode, pos, len, page, iomap);
+		status = __iomap_write_begin(inode, pos, len, flags, page,
+				srcmap);
 
 	if (unlikely(status))
 		goto out_unlock;
@@ -656,7 +687,7 @@ EXPORT_SYMBOL_GPL(iomap_set_page_dirty);
 
 static int
 __iomap_write_end(struct inode *inode, loff_t pos, unsigned len,
-		unsigned copied, struct page *page, struct iomap *iomap)
+		unsigned copied, struct page *page)
 {
 	flush_dcache_page(page);
 
@@ -696,20 +727,20 @@ iomap_write_end_inline(struct inode *inode, struct page *page,
 }
 
 static int
-iomap_write_end(struct inode *inode, loff_t pos, unsigned len,
-		unsigned copied, struct page *page, struct iomap *iomap)
+iomap_write_end(struct inode *inode, loff_t pos, unsigned len, unsigned copied,
+		struct page *page, struct iomap *iomap, struct iomap *srcmap)
 {
 	const struct iomap_page_ops *page_ops = iomap->page_ops;
 	loff_t old_size = inode->i_size;
 	int ret;
 
-	if (iomap->type == IOMAP_INLINE) {
+	if (srcmap->type == IOMAP_INLINE) {
 		ret = iomap_write_end_inline(inode, page, iomap, pos, copied);
-	} else if (iomap->flags & IOMAP_F_BUFFER_HEAD) {
+	} else if (srcmap->flags & IOMAP_F_BUFFER_HEAD) {
 		ret = block_write_end(NULL, inode->i_mapping, pos, len, copied,
 				page, NULL);
 	} else {
-		ret = __iomap_write_end(inode, pos, len, copied, page, iomap);
+		ret = __iomap_write_end(inode, pos, len, copied, page);
 	}
 
 	/*
@@ -736,12 +767,11 @@ iomap_write_end(struct inode *inode, loff_t pos, unsigned len,
 
 static loff_t
 iomap_write_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
-		struct iomap *iomap)
+		struct iomap *iomap, struct iomap *srcmap)
 {
 	struct iov_iter *i = data;
 	long status = 0;
 	ssize_t written = 0;
-	unsigned int flags = AOP_FLAG_NOFS;
 
 	do {
 		struct page *page;
@@ -771,8 +801,8 @@ again:
 			break;
 		}
 
-		status = iomap_write_begin(inode, pos, bytes, flags, &page,
-				iomap);
+		status = iomap_write_begin(inode, pos, bytes, 0, &page, iomap,
+				srcmap);
 		if (unlikely(status))
 			break;
 
@@ -783,8 +813,8 @@ again:
 
 		flush_dcache_page(page);
 
-		status = iomap_write_end(inode, pos, bytes, copied, page,
-				iomap);
+		status = iomap_write_end(inode, pos, bytes, copied, page, iomap,
+				srcmap);
 		if (unlikely(status < 0))
 			break;
 		copied = status;
@@ -835,50 +865,32 @@ iomap_file_buffered_write(struct kiocb *iocb, struct iov_iter *iter,
 }
 EXPORT_SYMBOL_GPL(iomap_file_buffered_write);
 
-static struct page *
-__iomap_read_page(struct inode *inode, loff_t offset)
-{
-	struct address_space *mapping = inode->i_mapping;
-	struct page *page;
-
-	page = read_mapping_page(mapping, offset >> PAGE_SHIFT, NULL);
-	if (IS_ERR(page))
-		return page;
-	if (!PageUptodate(page)) {
-		put_page(page);
-		return ERR_PTR(-EIO);
-	}
-	return page;
-}
-
 static loff_t
-iomap_dirty_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
-		struct iomap *iomap)
+iomap_unshare_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
+		struct iomap *iomap, struct iomap *srcmap)
 {
 	long status = 0;
 	ssize_t written = 0;
 
-	do {
-		struct page *page, *rpage;
-		unsigned long offset;	/* Offset into pagecache page */
-		unsigned long bytes;	/* Bytes to write to page */
-
-		offset = offset_in_page(pos);
-		bytes = min_t(loff_t, PAGE_SIZE - offset, length);
+	/* don't bother with blocks that are not shared to start with */
+	if (!(iomap->flags & IOMAP_F_SHARED))
+		return length;
+	/* don't bother with holes or unwritten extents */
+	if (srcmap->type == IOMAP_HOLE || srcmap->type == IOMAP_UNWRITTEN)
+		return length;
 
-		rpage = __iomap_read_page(inode, pos);
-		if (IS_ERR(rpage))
-			return PTR_ERR(rpage);
+	do {
+		unsigned long offset = offset_in_page(pos);
+		unsigned long bytes = min_t(loff_t, PAGE_SIZE - offset, length);
+		struct page *page;
 
 		status = iomap_write_begin(inode, pos, bytes,
-					   AOP_FLAG_NOFS, &page, iomap);
-		put_page(rpage);
+				IOMAP_WRITE_F_UNSHARE, &page, iomap, srcmap);
 		if (unlikely(status))
 			return status;
 
-		WARN_ON_ONCE(!PageUptodate(page));
-
-		status = iomap_write_end(inode, pos, bytes, bytes, page, iomap);
+		status = iomap_write_end(inode, pos, bytes, bytes, page, iomap,
+				srcmap);
 		if (unlikely(status <= 0)) {
 			if (WARN_ON_ONCE(status == 0))
 				return -EIO;
@@ -898,14 +910,14 @@ iomap_dirty_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
 }
 
 int
-iomap_file_dirty(struct inode *inode, loff_t pos, loff_t len,
+iomap_file_unshare(struct inode *inode, loff_t pos, loff_t len,
 		const struct iomap_ops *ops)
 {
 	loff_t ret;
 
 	while (len) {
 		ret = iomap_apply(inode, pos, len, IOMAP_WRITE, ops, NULL,
-				iomap_dirty_actor);
+				iomap_unshare_actor);
 		if (ret <= 0)
 			return ret;
 		pos += ret;
@@ -914,42 +926,34 @@ iomap_file_dirty(struct inode *inode, loff_t pos, loff_t len,
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(iomap_file_dirty);
+EXPORT_SYMBOL_GPL(iomap_file_unshare);
 
 static int iomap_zero(struct inode *inode, loff_t pos, unsigned offset,
-		unsigned bytes, struct iomap *iomap)
+		unsigned bytes, struct iomap *iomap, struct iomap *srcmap)
 {
 	struct page *page;
 	int status;
 
-	status = iomap_write_begin(inode, pos, bytes, AOP_FLAG_NOFS, &page,
-				   iomap);
+	status = iomap_write_begin(inode, pos, bytes, 0, &page, iomap, srcmap);
 	if (status)
 		return status;
 
 	zero_user(page, offset, bytes);
 	mark_page_accessed(page);
 
-	return iomap_write_end(inode, pos, bytes, bytes, page, iomap);
-}
-
-static int iomap_dax_zero(loff_t pos, unsigned offset, unsigned bytes,
-		struct iomap *iomap)
-{
-	return __dax_zero_page_range(iomap->bdev, iomap->dax_dev,
-			iomap_sector(iomap, pos & PAGE_MASK), offset, bytes);
+	return iomap_write_end(inode, pos, bytes, bytes, page, iomap, srcmap);
 }
 
 static loff_t
 iomap_zero_range_actor(struct inode *inode, loff_t pos, loff_t count,
-		void *data, struct iomap *iomap)
+		void *data, struct iomap *iomap, struct iomap *srcmap)
 {
 	bool *did_zero = data;
 	loff_t written = 0;
 	int status;
 
 	/* already zeroed?  we're done. */
-	if (iomap->type == IOMAP_HOLE || iomap->type == IOMAP_UNWRITTEN)
+	if (srcmap->type == IOMAP_HOLE || srcmap->type == IOMAP_UNWRITTEN)
 		return count;
 
 	do {
@@ -959,9 +963,10 @@ iomap_zero_range_actor(struct inode *inode, loff_t pos, loff_t count,
 		bytes = min_t(loff_t, PAGE_SIZE - offset, count);
 
 		if (IS_DAX(inode))
-			status = iomap_dax_zero(pos, offset, bytes, iomap);
+			status = dax_iomap_zero(pos, offset, bytes, iomap);
 		else
-			status = iomap_zero(inode, pos, offset, bytes, iomap);
+			status = iomap_zero(inode, pos, offset, bytes, iomap,
+					srcmap);
 		if (status < 0)
 			return status;
 
@@ -1011,7 +1016,7 @@ EXPORT_SYMBOL_GPL(iomap_truncate_page);
 
 static loff_t
 iomap_page_mkwrite_actor(struct inode *inode, loff_t pos, loff_t length,
-		void *data, struct iomap *iomap)
+		void *data, struct iomap *iomap, struct iomap *srcmap)
 {
 	struct page *page = data;
 	int ret;
@@ -1035,23 +1040,14 @@ vm_fault_t iomap_page_mkwrite(struct vm_fault *vmf, const struct iomap_ops *ops)
 	struct page *page = vmf->page;
 	struct inode *inode = file_inode(vmf->vma->vm_file);
 	unsigned long length;
-	loff_t offset, size;
+	loff_t offset;
 	ssize_t ret;
 
 	lock_page(page);
-	size = i_size_read(inode);
-	if ((page->mapping != inode->i_mapping) ||
-	    (page_offset(page) > size)) {
-		/* We overload EFAULT to mean page got truncated */
-		ret = -EFAULT;
+	ret = page_mkwrite_check_truncate(page, inode);
+	if (ret < 0)
 		goto out_unlock;
-	}
-
-	/* page is wholly or partially inside EOF */
-	if (((page->index + 1) << PAGE_SHIFT) > size)
-		length = offset_in_page(size);
-	else
-		length = PAGE_SIZE;
+	length = ret;
 
 	offset = page_offset(page);
 	while (length > 0) {
@@ -1071,3 +1067,552 @@ out_unlock:
 	return block_page_mkwrite_return(ret);
 }
 EXPORT_SYMBOL_GPL(iomap_page_mkwrite);
+
+static void
+iomap_finish_page_writeback(struct inode *inode, struct page *page,
+		int error)
+{
+	struct iomap_page *iop = to_iomap_page(page);
+
+	if (error) {
+		SetPageError(page);
+		mapping_set_error(inode->i_mapping, -EIO);
+	}
+
+	WARN_ON_ONCE(i_blocksize(inode) < PAGE_SIZE && !iop);
+	WARN_ON_ONCE(iop && atomic_read(&iop->write_count) <= 0);
+
+	if (!iop || atomic_dec_and_test(&iop->write_count))
+		end_page_writeback(page);
+}
+
+/*
+ * We're now finished for good with this ioend structure.  Update the page
+ * state, release holds on bios, and finally free up memory.  Do not use the
+ * ioend after this.
+ */
+static void
+iomap_finish_ioend(struct iomap_ioend *ioend, int error)
+{
+	struct inode *inode = ioend->io_inode;
+	struct bio *bio = &ioend->io_inline_bio;
+	struct bio *last = ioend->io_bio, *next;
+	u64 start = bio->bi_iter.bi_sector;
+	loff_t offset = ioend->io_offset;
+	bool quiet = bio_flagged(bio, BIO_QUIET);
+
+	for (bio = &ioend->io_inline_bio; bio; bio = next) {
+		struct bio_vec *bv;
+		struct bvec_iter_all iter_all;
+
+		/*
+		 * For the last bio, bi_private points to the ioend, so we
+		 * need to explicitly end the iteration here.
+		 */
+		if (bio == last)
+			next = NULL;
+		else
+			next = bio->bi_private;
+
+		/* walk each page on bio, ending page IO on them */
+		bio_for_each_segment_all(bv, bio, iter_all)
+			iomap_finish_page_writeback(inode, bv->bv_page, error);
+		bio_put(bio);
+	}
+	/* The ioend has been freed by bio_put() */
+
+	if (unlikely(error && !quiet)) {
+		printk_ratelimited(KERN_ERR
+"%s: writeback error on inode %lu, offset %lld, sector %llu",
+			inode->i_sb->s_id, inode->i_ino, offset, start);
+	}
+}
+
+void
+iomap_finish_ioends(struct iomap_ioend *ioend, int error)
+{
+	struct list_head tmp;
+
+	list_replace_init(&ioend->io_list, &tmp);
+	iomap_finish_ioend(ioend, error);
+
+	while (!list_empty(&tmp)) {
+		ioend = list_first_entry(&tmp, struct iomap_ioend, io_list);
+		list_del_init(&ioend->io_list);
+		iomap_finish_ioend(ioend, error);
+	}
+}
+EXPORT_SYMBOL_GPL(iomap_finish_ioends);
+
+/*
+ * We can merge two adjacent ioends if they have the same set of work to do.
+ */
+static bool
+iomap_ioend_can_merge(struct iomap_ioend *ioend, struct iomap_ioend *next)
+{
+	if (ioend->io_bio->bi_status != next->io_bio->bi_status)
+		return false;
+	if ((ioend->io_flags & IOMAP_F_SHARED) ^
+	    (next->io_flags & IOMAP_F_SHARED))
+		return false;
+	if ((ioend->io_type == IOMAP_UNWRITTEN) ^
+	    (next->io_type == IOMAP_UNWRITTEN))
+		return false;
+	if (ioend->io_offset + ioend->io_size != next->io_offset)
+		return false;
+	return true;
+}
+
+void
+iomap_ioend_try_merge(struct iomap_ioend *ioend, struct list_head *more_ioends,
+		void (*merge_private)(struct iomap_ioend *ioend,
+				struct iomap_ioend *next))
+{
+	struct iomap_ioend *next;
+
+	INIT_LIST_HEAD(&ioend->io_list);
+
+	while ((next = list_first_entry_or_null(more_ioends, struct iomap_ioend,
+			io_list))) {
+		if (!iomap_ioend_can_merge(ioend, next))
+			break;
+		list_move_tail(&next->io_list, &ioend->io_list);
+		ioend->io_size += next->io_size;
+		if (next->io_private && merge_private)
+			merge_private(ioend, next);
+	}
+}
+EXPORT_SYMBOL_GPL(iomap_ioend_try_merge);
+
+static int
+iomap_ioend_compare(void *priv, struct list_head *a, struct list_head *b)
+{
+	struct iomap_ioend *ia = container_of(a, struct iomap_ioend, io_list);
+	struct iomap_ioend *ib = container_of(b, struct iomap_ioend, io_list);
+
+	if (ia->io_offset < ib->io_offset)
+		return -1;
+	if (ia->io_offset > ib->io_offset)
+		return 1;
+	return 0;
+}
+
+void
+iomap_sort_ioends(struct list_head *ioend_list)
+{
+	list_sort(NULL, ioend_list, iomap_ioend_compare);
+}
+EXPORT_SYMBOL_GPL(iomap_sort_ioends);
+
+static void iomap_writepage_end_bio(struct bio *bio)
+{
+	struct iomap_ioend *ioend = bio->bi_private;
+
+	iomap_finish_ioend(ioend, blk_status_to_errno(bio->bi_status));
+}
+
+/*
+ * Submit the final bio for an ioend.
+ *
+ * If @error is non-zero, it means that we have a situation where some part of
+ * the submission process has failed after we have marked paged for writeback
+ * and unlocked them.  In this situation, we need to fail the bio instead of
+ * submitting it.  This typically only happens on a filesystem shutdown.
+ */
+static int
+iomap_submit_ioend(struct iomap_writepage_ctx *wpc, struct iomap_ioend *ioend,
+		int error)
+{
+	ioend->io_bio->bi_private = ioend;
+	ioend->io_bio->bi_end_io = iomap_writepage_end_bio;
+
+	if (wpc->ops->prepare_ioend)
+		error = wpc->ops->prepare_ioend(ioend, error);
+	if (error) {
+		/*
+		 * If we are failing the IO now, just mark the ioend with an
+		 * error and finish it.  This will run IO completion immediately
+		 * as there is only one reference to the ioend at this point in
+		 * time.
+		 */
+		ioend->io_bio->bi_status = errno_to_blk_status(error);
+		bio_endio(ioend->io_bio);
+		return error;
+	}
+
+	submit_bio(ioend->io_bio);
+	return 0;
+}
+
+static struct iomap_ioend *
+iomap_alloc_ioend(struct inode *inode, struct iomap_writepage_ctx *wpc,
+		loff_t offset, sector_t sector, struct writeback_control *wbc)
+{
+	struct iomap_ioend *ioend;
+	struct bio *bio;
+
+	bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, &iomap_ioend_bioset);
+	bio_set_dev(bio, wpc->iomap.bdev);
+	bio->bi_iter.bi_sector = sector;
+	bio->bi_opf = REQ_OP_WRITE | wbc_to_write_flags(wbc);
+	bio->bi_write_hint = inode->i_write_hint;
+	wbc_init_bio(wbc, bio);
+
+	ioend = container_of(bio, struct iomap_ioend, io_inline_bio);
+	INIT_LIST_HEAD(&ioend->io_list);
+	ioend->io_type = wpc->iomap.type;
+	ioend->io_flags = wpc->iomap.flags;
+	ioend->io_inode = inode;
+	ioend->io_size = 0;
+	ioend->io_offset = offset;
+	ioend->io_private = NULL;
+	ioend->io_bio = bio;
+	return ioend;
+}
+
+/*
+ * Allocate a new bio, and chain the old bio to the new one.
+ *
+ * Note that we have to do perform the chaining in this unintuitive order
+ * so that the bi_private linkage is set up in the right direction for the
+ * traversal in iomap_finish_ioend().
+ */
+static struct bio *
+iomap_chain_bio(struct bio *prev)
+{
+	struct bio *new;
+
+	new = bio_alloc(GFP_NOFS, BIO_MAX_PAGES);
+	bio_copy_dev(new, prev);/* also copies over blkcg information */
+	new->bi_iter.bi_sector = bio_end_sector(prev);
+	new->bi_opf = prev->bi_opf;
+	new->bi_write_hint = prev->bi_write_hint;
+
+	bio_chain(prev, new);
+	bio_get(prev);		/* for iomap_finish_ioend */
+	submit_bio(prev);
+	return new;
+}
+
+static bool
+iomap_can_add_to_ioend(struct iomap_writepage_ctx *wpc, loff_t offset,
+		sector_t sector)
+{
+	if ((wpc->iomap.flags & IOMAP_F_SHARED) !=
+	    (wpc->ioend->io_flags & IOMAP_F_SHARED))
+		return false;
+	if (wpc->iomap.type != wpc->ioend->io_type)
+		return false;
+	if (offset != wpc->ioend->io_offset + wpc->ioend->io_size)
+		return false;
+	if (sector != bio_end_sector(wpc->ioend->io_bio))
+		return false;
+	return true;
+}
+
+/*
+ * Test to see if we have an existing ioend structure that we could append to
+ * first, otherwise finish off the current ioend and start another.
+ */
+static void
+iomap_add_to_ioend(struct inode *inode, loff_t offset, struct page *page,
+		struct iomap_page *iop, struct iomap_writepage_ctx *wpc,
+		struct writeback_control *wbc, struct list_head *iolist)
+{
+	sector_t sector = iomap_sector(&wpc->iomap, offset);
+	unsigned len = i_blocksize(inode);
+	unsigned poff = offset & (PAGE_SIZE - 1);
+	bool merged, same_page = false;
+
+	if (!wpc->ioend || !iomap_can_add_to_ioend(wpc, offset, sector)) {
+		if (wpc->ioend)
+			list_add(&wpc->ioend->io_list, iolist);
+		wpc->ioend = iomap_alloc_ioend(inode, wpc, offset, sector, wbc);
+	}
+
+	merged = __bio_try_merge_page(wpc->ioend->io_bio, page, len, poff,
+			&same_page);
+	if (iop && !same_page)
+		atomic_inc(&iop->write_count);
+
+	if (!merged) {
+		if (bio_full(wpc->ioend->io_bio, len)) {
+			wpc->ioend->io_bio =
+				iomap_chain_bio(wpc->ioend->io_bio);
+		}
+		bio_add_page(wpc->ioend->io_bio, page, len, poff);
+	}
+
+	wpc->ioend->io_size += len;
+	wbc_account_cgroup_owner(wbc, page, len);
+}
+
+/*
+ * We implement an immediate ioend submission policy here to avoid needing to
+ * chain multiple ioends and hence nest mempool allocations which can violate
+ * forward progress guarantees we need to provide. The current ioend we are
+ * adding blocks to is cached on the writepage context, and if the new block
+ * does not append to the cached ioend it will create a new ioend and cache that
+ * instead.
+ *
+ * If a new ioend is created and cached, the old ioend is returned and queued
+ * locally for submission once the entire page is processed or an error has been
+ * detected.  While ioends are submitted immediately after they are completed,
+ * batching optimisations are provided by higher level block plugging.
+ *
+ * At the end of a writeback pass, there will be a cached ioend remaining on the
+ * writepage context that the caller will need to submit.
+ */
+static int
+iomap_writepage_map(struct iomap_writepage_ctx *wpc,
+		struct writeback_control *wbc, struct inode *inode,
+		struct page *page, u64 end_offset)
+{
+	struct iomap_page *iop = to_iomap_page(page);
+	struct iomap_ioend *ioend, *next;
+	unsigned len = i_blocksize(inode);
+	u64 file_offset; /* file offset of page */
+	int error = 0, count = 0, i;
+	LIST_HEAD(submit_list);
+
+	WARN_ON_ONCE(i_blocksize(inode) < PAGE_SIZE && !iop);
+	WARN_ON_ONCE(iop && atomic_read(&iop->write_count) != 0);
+
+	/*
+	 * Walk through the page to find areas to write back. If we run off the
+	 * end of the current map or find the current map invalid, grab a new
+	 * one.
+	 */
+	for (i = 0, file_offset = page_offset(page);
+	     i < (PAGE_SIZE >> inode->i_blkbits) && file_offset < end_offset;
+	     i++, file_offset += len) {
+		if (iop && !test_bit(i, iop->uptodate))
+			continue;
+
+		error = wpc->ops->map_blocks(wpc, inode, file_offset);
+		if (error)
+			break;
+		if (WARN_ON_ONCE(wpc->iomap.type == IOMAP_INLINE))
+			continue;
+		if (wpc->iomap.type == IOMAP_HOLE)
+			continue;
+		iomap_add_to_ioend(inode, file_offset, page, iop, wpc, wbc,
+				 &submit_list);
+		count++;
+	}
+
+	WARN_ON_ONCE(!wpc->ioend && !list_empty(&submit_list));
+	WARN_ON_ONCE(!PageLocked(page));
+	WARN_ON_ONCE(PageWriteback(page));
+
+	/*
+	 * We cannot cancel the ioend directly here on error.  We may have
+	 * already set other pages under writeback and hence we have to run I/O
+	 * completion to mark the error state of the pages under writeback
+	 * appropriately.
+	 */
+	if (unlikely(error)) {
+		if (!count) {
+			/*
+			 * If the current page hasn't been added to ioend, it
+			 * won't be affected by I/O completions and we must
+			 * discard and unlock it right here.
+			 */
+			if (wpc->ops->discard_page)
+				wpc->ops->discard_page(page);
+			ClearPageUptodate(page);
+			unlock_page(page);
+			goto done;
+		}
+
+		/*
+		 * If the page was not fully cleaned, we need to ensure that the
+		 * higher layers come back to it correctly.  That means we need
+		 * to keep the page dirty, and for WB_SYNC_ALL writeback we need
+		 * to ensure the PAGECACHE_TAG_TOWRITE index mark is not removed
+		 * so another attempt to write this page in this writeback sweep
+		 * will be made.
+		 */
+		set_page_writeback_keepwrite(page);
+	} else {
+		clear_page_dirty_for_io(page);
+		set_page_writeback(page);
+	}
+
+	unlock_page(page);
+
+	/*
+	 * Preserve the original error if there was one, otherwise catch
+	 * submission errors here and propagate into subsequent ioend
+	 * submissions.
+	 */
+	list_for_each_entry_safe(ioend, next, &submit_list, io_list) {
+		int error2;
+
+		list_del_init(&ioend->io_list);
+		error2 = iomap_submit_ioend(wpc, ioend, error);
+		if (error2 && !error)
+			error = error2;
+	}
+
+	/*
+	 * We can end up here with no error and nothing to write only if we race
+	 * with a partial page truncate on a sub-page block sized filesystem.
+	 */
+	if (!count)
+		end_page_writeback(page);
+done:
+	mapping_set_error(page->mapping, error);
+	return error;
+}
+
+/*
+ * Write out a dirty page.
+ *
+ * For delalloc space on the page we need to allocate space and flush it.
+ * For unwritten space on the page we need to start the conversion to
+ * regular allocated space.
+ */
+static int
+iomap_do_writepage(struct page *page, struct writeback_control *wbc, void *data)
+{
+	struct iomap_writepage_ctx *wpc = data;
+	struct inode *inode = page->mapping->host;
+	pgoff_t end_index;
+	u64 end_offset;
+	loff_t offset;
+
+	trace_iomap_writepage(inode, page_offset(page), PAGE_SIZE);
+
+	/*
+	 * Refuse to write the page out if we are called from reclaim context.
+	 *
+	 * This avoids stack overflows when called from deeply used stacks in
+	 * random callers for direct reclaim or memcg reclaim.  We explicitly
+	 * allow reclaim from kswapd as the stack usage there is relatively low.
+	 *
+	 * This should never happen except in the case of a VM regression so
+	 * warn about it.
+	 */
+	if (WARN_ON_ONCE((current->flags & (PF_MEMALLOC|PF_KSWAPD)) ==
+			PF_MEMALLOC))
+		goto redirty;
+
+	/*
+	 * Given that we do not allow direct reclaim to call us, we should
+	 * never be called in a recursive filesystem reclaim context.
+	 */
+	if (WARN_ON_ONCE(current->flags & PF_MEMALLOC_NOFS))
+		goto redirty;
+
+	/*
+	 * Is this page beyond the end of the file?
+	 *
+	 * The page index is less than the end_index, adjust the end_offset
+	 * to the highest offset that this page should represent.
+	 * -----------------------------------------------------
+	 * |			file mapping	       | <EOF> |
+	 * -----------------------------------------------------
+	 * | Page ... | Page N-2 | Page N-1 |  Page N  |       |
+	 * ^--------------------------------^----------|--------
+	 * |     desired writeback range    |      see else    |
+	 * ---------------------------------^------------------|
+	 */
+	offset = i_size_read(inode);
+	end_index = offset >> PAGE_SHIFT;
+	if (page->index < end_index)
+		end_offset = (loff_t)(page->index + 1) << PAGE_SHIFT;
+	else {
+		/*
+		 * Check whether the page to write out is beyond or straddles
+		 * i_size or not.
+		 * -------------------------------------------------------
+		 * |		file mapping		        | <EOF>  |
+		 * -------------------------------------------------------
+		 * | Page ... | Page N-2 | Page N-1 |  Page N   | Beyond |
+		 * ^--------------------------------^-----------|---------
+		 * |				    |      Straddles     |
+		 * ---------------------------------^-----------|--------|
+		 */
+		unsigned offset_into_page = offset & (PAGE_SIZE - 1);
+
+		/*
+		 * Skip the page if it is fully outside i_size, e.g. due to a
+		 * truncate operation that is in progress. We must redirty the
+		 * page so that reclaim stops reclaiming it. Otherwise
+		 * iomap_vm_releasepage() is called on it and gets confused.
+		 *
+		 * Note that the end_index is unsigned long, it would overflow
+		 * if the given offset is greater than 16TB on 32-bit system
+		 * and if we do check the page is fully outside i_size or not
+		 * via "if (page->index >= end_index + 1)" as "end_index + 1"
+		 * will be evaluated to 0.  Hence this page will be redirtied
+		 * and be written out repeatedly which would result in an
+		 * infinite loop, the user program that perform this operation
+		 * will hang.  Instead, we can verify this situation by checking
+		 * if the page to write is totally beyond the i_size or if it's
+		 * offset is just equal to the EOF.
+		 */
+		if (page->index > end_index ||
+		    (page->index == end_index && offset_into_page == 0))
+			goto redirty;
+
+		/*
+		 * The page straddles i_size.  It must be zeroed out on each
+		 * and every writepage invocation because it may be mmapped.
+		 * "A file is mapped in multiples of the page size.  For a file
+		 * that is not a multiple of the page size, the remaining
+		 * memory is zeroed when mapped, and writes to that region are
+		 * not written out to the file."
+		 */
+		zero_user_segment(page, offset_into_page, PAGE_SIZE);
+
+		/* Adjust the end_offset to the end of file */
+		end_offset = offset;
+	}
+
+	return iomap_writepage_map(wpc, wbc, inode, page, end_offset);
+
+redirty:
+	redirty_page_for_writepage(wbc, page);
+	unlock_page(page);
+	return 0;
+}
+
+int
+iomap_writepage(struct page *page, struct writeback_control *wbc,
+		struct iomap_writepage_ctx *wpc,
+		const struct iomap_writeback_ops *ops)
+{
+	int ret;
+
+	wpc->ops = ops;
+	ret = iomap_do_writepage(page, wbc, wpc);
+	if (!wpc->ioend)
+		return ret;
+	return iomap_submit_ioend(wpc, wpc->ioend, ret);
+}
+EXPORT_SYMBOL_GPL(iomap_writepage);
+
+int
+iomap_writepages(struct address_space *mapping, struct writeback_control *wbc,
+		struct iomap_writepage_ctx *wpc,
+		const struct iomap_writeback_ops *ops)
+{
+	int			ret;
+
+	wpc->ops = ops;
+	ret = write_cache_pages(mapping, wbc, iomap_do_writepage, wpc);
+	if (!wpc->ioend)
+		return ret;
+	return iomap_submit_ioend(wpc, wpc->ioend, ret);
+}
+EXPORT_SYMBOL_GPL(iomap_writepages);
+
+static int __init iomap_init(void)
+{
+	return bioset_init(&iomap_ioend_bioset, 4 * (PAGE_SIZE / SECTOR_SIZE),
+			   offsetof(struct iomap_ioend, io_inline_bio),
+			   BIOSET_NEED_BVECS);
+}
+fs_initcall(iomap_init);
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 1fc28c2da279..fd3bd06fabb6 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -201,12 +201,12 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 	unsigned int blkbits = blksize_bits(bdev_logical_block_size(iomap->bdev));
 	unsigned int fs_block_size = i_blocksize(inode), pad;
 	unsigned int align = iov_iter_alignment(dio->submit.iter);
-	struct iov_iter iter;
 	struct bio *bio;
 	bool need_zeroout = false;
 	bool use_fua = false;
 	int nr_pages, ret = 0;
 	size_t copied = 0;
+	size_t orig_count;
 
 	if ((pos | length | align) & ((1 << blkbits) - 1))
 		return -EINVAL;
@@ -236,15 +236,18 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 	}
 
 	/*
-	 * Operate on a partial iter trimmed to the extent we were called for.
-	 * We'll update the iter in the dio once we're done with this extent.
+	 * Save the original count and trim the iter to just the extent we
+	 * are operating on right now.  The iter will be re-expanded once
+	 * we are done.
 	 */
-	iter = *dio->submit.iter;
-	iov_iter_truncate(&iter, length);
+	orig_count = iov_iter_count(dio->submit.iter);
+	iov_iter_truncate(dio->submit.iter, length);
 
-	nr_pages = iov_iter_npages(&iter, BIO_MAX_PAGES);
-	if (nr_pages <= 0)
-		return nr_pages;
+	nr_pages = iov_iter_npages(dio->submit.iter, BIO_MAX_PAGES);
+	if (nr_pages <= 0) {
+		ret = nr_pages;
+		goto out;
+	}
 
 	if (need_zeroout) {
 		/* zero out from the start of the block to the write offset */
@@ -257,7 +260,8 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 		size_t n;
 		if (dio->error) {
 			iov_iter_revert(dio->submit.iter, copied);
-			return 0;
+			copied = ret = 0;
+			goto out;
 		}
 
 		bio = bio_alloc(GFP_KERNEL, nr_pages);
@@ -268,7 +272,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 		bio->bi_private = dio;
 		bio->bi_end_io = iomap_dio_bio_end_io;
 
-		ret = bio_iov_iter_get_pages(bio, &iter);
+		ret = bio_iov_iter_get_pages(bio, dio->submit.iter);
 		if (unlikely(ret)) {
 			/*
 			 * We have to stop part way through an IO. We must fall
@@ -294,13 +298,11 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
 				bio_set_pages_dirty(bio);
 		}
 
-		iov_iter_advance(dio->submit.iter, n);
-
 		dio->size += n;
 		pos += n;
 		copied += n;
 
-		nr_pages = iov_iter_npages(&iter, BIO_MAX_PAGES);
+		nr_pages = iov_iter_npages(dio->submit.iter, BIO_MAX_PAGES);
 		iomap_dio_submit_bio(dio, iomap, bio);
 	} while (nr_pages);
 
@@ -318,7 +320,12 @@ zero_tail:
 		if (pad)
 			iomap_dio_zero(dio, iomap, pos, fs_block_size - pad);
 	}
-	return copied ? copied : ret;
+out:
+	/* Undo iter limitation to current extent */
+	iov_iter_reexpand(dio->submit.iter, orig_count - copied);
+	if (copied)
+		return copied;
+	return ret;
 }
 
 static loff_t
@@ -358,7 +365,7 @@ iomap_dio_inline_actor(struct inode *inode, loff_t pos, loff_t length,
 
 static loff_t
 iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length,
-		void *data, struct iomap *iomap)
+		void *data, struct iomap *iomap, struct iomap *srcmap)
 {
 	struct iomap_dio *dio = data;
 
@@ -392,15 +399,15 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length,
  */
 ssize_t
 iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
-		const struct iomap_ops *ops, const struct iomap_dio_ops *dops)
+		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
+		bool wait_for_completion)
 {
 	struct address_space *mapping = iocb->ki_filp->f_mapping;
 	struct inode *inode = file_inode(iocb->ki_filp);
 	size_t count = iov_iter_count(iter);
-	loff_t pos = iocb->ki_pos, start = pos;
+	loff_t pos = iocb->ki_pos;
 	loff_t end = iocb->ki_pos + count - 1, ret = 0;
 	unsigned int flags = IOMAP_DIRECT;
-	bool wait_for_completion = is_sync_kiocb(iocb);
 	struct blk_plug plug;
 	struct iomap_dio *dio;
 
@@ -409,6 +416,9 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	if (!count)
 		return 0;
 
+	if (WARN_ON(is_sync_kiocb(iocb) && !wait_for_completion))
+		return -EIO;
+
 	dio = kmalloc(sizeof(*dio), GFP_KERNEL);
 	if (!dio)
 		return -ENOMEM;
@@ -430,7 +440,7 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		if (pos >= dio->i_size)
 			goto out_free_dio;
 
-		if (iter_is_iovec(iter) && iov_iter_rw(iter) == READ)
+		if (iter_is_iovec(iter))
 			dio->flags |= IOMAP_DIO_DIRTY;
 	} else {
 		flags |= IOMAP_WRITE;
@@ -451,14 +461,14 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	}
 
 	if (iocb->ki_flags & IOCB_NOWAIT) {
-		if (filemap_range_has_page(mapping, start, end)) {
+		if (filemap_range_has_page(mapping, pos, end)) {
 			ret = -EAGAIN;
 			goto out_free_dio;
 		}
 		flags |= IOMAP_NOWAIT;
 	}
 
-	ret = filemap_write_and_wait_range(mapping, start, end);
+	ret = filemap_write_and_wait_range(mapping, pos, end);
 	if (ret)
 		goto out_free_dio;
 
@@ -469,7 +479,7 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	 * pretty crazy thing to do, so we don't support it 100%.
 	 */
 	ret = invalidate_inode_pages2_range(mapping,
-			start >> PAGE_SHIFT, end >> PAGE_SHIFT);
+			pos >> PAGE_SHIFT, end >> PAGE_SHIFT);
 	if (ret)
 		dio_warn_stale_pagecache(iocb->ki_filp);
 	ret = 0;
@@ -497,8 +507,15 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		}
 		pos += ret;
 
-		if (iov_iter_rw(iter) == READ && pos >= dio->i_size)
+		if (iov_iter_rw(iter) == READ && pos >= dio->i_size) {
+			/*
+			 * We only report that we've read data up to i_size.
+			 * Revert iter to a state corresponding to that as
+			 * some callers (such as splice code) rely on it.
+			 */
+			iov_iter_revert(iter, pos - dio->i_size);
 			break;
+		}
 	} while ((count = iov_iter_count(iter)) > 0);
 	blk_finish_plug(&plug);
 
@@ -517,8 +534,8 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 
 	/*
 	 * We are about to drop our additional submission reference, which
-	 * might be the last reference to the dio.  There are three three
-	 * different ways we can progress here:
+	 * might be the last reference to the dio.  There are three different
+	 * ways we can progress here:
 	 *
 	 *  (a) If this is the last reference we will always complete and free
 	 *	the dio ourselves.
@@ -544,7 +561,7 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 			    !dio->submit.last_queue ||
 			    !blk_poll(dio->submit.last_queue,
 					 dio->submit.cookie, true))
-				io_schedule();
+				blk_io_schedule();
 		}
 		__set_current_state(TASK_RUNNING);
 	}
diff --git a/fs/iomap/fiemap.c b/fs/iomap/fiemap.c
index f26fdd36e383..d55e8f491a5e 100644
--- a/fs/iomap/fiemap.c
+++ b/fs/iomap/fiemap.c
@@ -44,7 +44,7 @@ static int iomap_to_fiemap(struct fiemap_extent_info *fi,
 
 static loff_t
 iomap_fiemap_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
-		struct iomap *iomap)
+		struct iomap *iomap, struct iomap *srcmap)
 {
 	struct fiemap_ctx *ctx = data;
 	loff_t ret = length;
@@ -111,16 +111,13 @@ EXPORT_SYMBOL_GPL(iomap_fiemap);
 
 static loff_t
 iomap_bmap_actor(struct inode *inode, loff_t pos, loff_t length,
-		void *data, struct iomap *iomap)
+		void *data, struct iomap *iomap, struct iomap *srcmap)
 {
 	sector_t *bno = data, addr;
 
 	if (iomap->type == IOMAP_MAPPED) {
 		addr = (pos - iomap->offset + iomap->addr) >> inode->i_blkbits;
-		if (addr > INT_MAX)
-			WARN(1, "would truncate bmap result\n");
-		else
-			*bno = addr;
+		*bno = addr;
 	}
 	return 0;
 }
@@ -133,12 +130,16 @@ iomap_bmap(struct address_space *mapping, sector_t bno,
 	struct inode *inode = mapping->host;
 	loff_t pos = bno << inode->i_blkbits;
 	unsigned blocksize = i_blocksize(inode);
+	int ret;
 
 	if (filemap_write_and_wait(mapping))
 		return 0;
 
 	bno = 0;
-	iomap_apply(inode, pos, blocksize, 0, ops, &bno, iomap_bmap_actor);
+	ret = iomap_apply(inode, pos, blocksize, 0, ops, &bno,
+			  iomap_bmap_actor);
+	if (ret)
+		return 0;
 	return bno;
 }
 EXPORT_SYMBOL_GPL(iomap_bmap);
diff --git a/fs/iomap/seek.c b/fs/iomap/seek.c
index c04bad4b2b43..89f61d93c0bc 100644
--- a/fs/iomap/seek.c
+++ b/fs/iomap/seek.c
@@ -119,7 +119,7 @@ out:
 
 static loff_t
 iomap_seek_hole_actor(struct inode *inode, loff_t offset, loff_t length,
-		      void *data, struct iomap *iomap)
+		      void *data, struct iomap *iomap, struct iomap *srcmap)
 {
 	switch (iomap->type) {
 	case IOMAP_UNWRITTEN:
@@ -165,7 +165,7 @@ EXPORT_SYMBOL_GPL(iomap_seek_hole);
 
 static loff_t
 iomap_seek_data_actor(struct inode *inode, loff_t offset, loff_t length,
-		      void *data, struct iomap *iomap)
+		      void *data, struct iomap *iomap, struct iomap *srcmap)
 {
 	switch (iomap->type) {
 	case IOMAP_HOLE:
diff --git a/fs/iomap/swapfile.c b/fs/iomap/swapfile.c
index 152a230f668d..a648dbf6991e 100644
--- a/fs/iomap/swapfile.c
+++ b/fs/iomap/swapfile.c
@@ -76,7 +76,8 @@ static int iomap_swapfile_add_extent(struct iomap_swapfile_info *isi)
  * distinction between written and unwritten extents.
  */
 static loff_t iomap_swapfile_activate_actor(struct inode *inode, loff_t pos,
-		loff_t count, void *data, struct iomap *iomap)
+		loff_t count, void *data, struct iomap *iomap,
+		struct iomap *srcmap)
 {
 	struct iomap_swapfile_info *isi = data;
 	int error;
diff --git a/fs/iomap/trace.c b/fs/iomap/trace.c
new file mode 100644
index 000000000000..da217246b1a9
--- /dev/null
+++ b/fs/iomap/trace.c
@@ -0,0 +1,12 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2019 Christoph Hellwig
+ */
+#include <linux/iomap.h>
+
+/*
+ * We include this last to have the helpers above available for the trace
+ * event implementations.
+ */
+#define CREATE_TRACE_POINTS
+#include "trace.h"
diff --git a/fs/iomap/trace.h b/fs/iomap/trace.h
new file mode 100644
index 000000000000..5693a39d52fb
--- /dev/null
+++ b/fs/iomap/trace.h
@@ -0,0 +1,186 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (c) 2009-2019 Christoph Hellwig
+ *
+ * NOTE: none of these tracepoints shall be consider a stable kernel ABI
+ * as they can change at any time.
+ */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM iomap
+
+#if !defined(_IOMAP_TRACE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _IOMAP_TRACE_H
+
+#include <linux/tracepoint.h>
+
+struct inode;
+
+DECLARE_EVENT_CLASS(iomap_readpage_class,
+	TP_PROTO(struct inode *inode, int nr_pages),
+	TP_ARGS(inode, nr_pages),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(u64, ino)
+		__field(int, nr_pages)
+	),
+	TP_fast_assign(
+		__entry->dev = inode->i_sb->s_dev;
+		__entry->ino = inode->i_ino;
+		__entry->nr_pages = nr_pages;
+	),
+	TP_printk("dev %d:%d ino 0x%llx nr_pages %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->nr_pages)
+)
+
+#define DEFINE_READPAGE_EVENT(name)		\
+DEFINE_EVENT(iomap_readpage_class, name,	\
+	TP_PROTO(struct inode *inode, int nr_pages), \
+	TP_ARGS(inode, nr_pages))
+DEFINE_READPAGE_EVENT(iomap_readpage);
+DEFINE_READPAGE_EVENT(iomap_readahead);
+
+DECLARE_EVENT_CLASS(iomap_range_class,
+	TP_PROTO(struct inode *inode, unsigned long off, unsigned int len),
+	TP_ARGS(inode, off, len),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(u64, ino)
+		__field(loff_t, size)
+		__field(unsigned long, offset)
+		__field(unsigned int, length)
+	),
+	TP_fast_assign(
+		__entry->dev = inode->i_sb->s_dev;
+		__entry->ino = inode->i_ino;
+		__entry->size = i_size_read(inode);
+		__entry->offset = off;
+		__entry->length = len;
+	),
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx offset %lx "
+		  "length %x",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->size,
+		  __entry->offset,
+		  __entry->length)
+)
+
+#define DEFINE_RANGE_EVENT(name)		\
+DEFINE_EVENT(iomap_range_class, name,	\
+	TP_PROTO(struct inode *inode, unsigned long off, unsigned int len),\
+	TP_ARGS(inode, off, len))
+DEFINE_RANGE_EVENT(iomap_writepage);
+DEFINE_RANGE_EVENT(iomap_releasepage);
+DEFINE_RANGE_EVENT(iomap_invalidatepage);
+
+#define IOMAP_TYPE_STRINGS \
+	{ IOMAP_HOLE,		"HOLE" }, \
+	{ IOMAP_DELALLOC,	"DELALLOC" }, \
+	{ IOMAP_MAPPED,		"MAPPED" }, \
+	{ IOMAP_UNWRITTEN,	"UNWRITTEN" }, \
+	{ IOMAP_INLINE,		"INLINE" }
+
+#define IOMAP_FLAGS_STRINGS \
+	{ IOMAP_WRITE,		"WRITE" }, \
+	{ IOMAP_ZERO,		"ZERO" }, \
+	{ IOMAP_REPORT,		"REPORT" }, \
+	{ IOMAP_FAULT,		"FAULT" }, \
+	{ IOMAP_DIRECT,		"DIRECT" }, \
+	{ IOMAP_NOWAIT,		"NOWAIT" }
+
+#define IOMAP_F_FLAGS_STRINGS \
+	{ IOMAP_F_NEW,		"NEW" }, \
+	{ IOMAP_F_DIRTY,	"DIRTY" }, \
+	{ IOMAP_F_SHARED,	"SHARED" }, \
+	{ IOMAP_F_MERGED,	"MERGED" }, \
+	{ IOMAP_F_BUFFER_HEAD,	"BH" }, \
+	{ IOMAP_F_SIZE_CHANGED,	"SIZE_CHANGED" }
+
+DECLARE_EVENT_CLASS(iomap_class,
+	TP_PROTO(struct inode *inode, struct iomap *iomap),
+	TP_ARGS(inode, iomap),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(u64, ino)
+		__field(u64, addr)
+		__field(loff_t, offset)
+		__field(u64, length)
+		__field(u16, type)
+		__field(u16, flags)
+		__field(dev_t, bdev)
+	),
+	TP_fast_assign(
+		__entry->dev = inode->i_sb->s_dev;
+		__entry->ino = inode->i_ino;
+		__entry->addr = iomap->addr;
+		__entry->offset = iomap->offset;
+		__entry->length = iomap->length;
+		__entry->type = iomap->type;
+		__entry->flags = iomap->flags;
+		__entry->bdev = iomap->bdev ? iomap->bdev->bd_dev : 0;
+	),
+	TP_printk("dev %d:%d ino 0x%llx bdev %d:%d addr %lld offset %lld "
+		  "length %llu type %s flags %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  MAJOR(__entry->bdev), MINOR(__entry->bdev),
+		  __entry->addr,
+		  __entry->offset,
+		  __entry->length,
+		  __print_symbolic(__entry->type, IOMAP_TYPE_STRINGS),
+		  __print_flags(__entry->flags, "|", IOMAP_F_FLAGS_STRINGS))
+)
+
+#define DEFINE_IOMAP_EVENT(name)		\
+DEFINE_EVENT(iomap_class, name,	\
+	TP_PROTO(struct inode *inode, struct iomap *iomap), \
+	TP_ARGS(inode, iomap))
+DEFINE_IOMAP_EVENT(iomap_apply_dstmap);
+DEFINE_IOMAP_EVENT(iomap_apply_srcmap);
+
+TRACE_EVENT(iomap_apply,
+	TP_PROTO(struct inode *inode, loff_t pos, loff_t length,
+		unsigned int flags, const void *ops, void *actor,
+		unsigned long caller),
+	TP_ARGS(inode, pos, length, flags, ops, actor, caller),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(u64, ino)
+		__field(loff_t, pos)
+		__field(loff_t, length)
+		__field(unsigned int, flags)
+		__field(const void *, ops)
+		__field(void *, actor)
+		__field(unsigned long, caller)
+	),
+	TP_fast_assign(
+		__entry->dev = inode->i_sb->s_dev;
+		__entry->ino = inode->i_ino;
+		__entry->pos = pos;
+		__entry->length = length;
+		__entry->flags = flags;
+		__entry->ops = ops;
+		__entry->actor = actor;
+		__entry->caller = caller;
+	),
+	TP_printk("dev %d:%d ino 0x%llx pos %lld length %lld flags %s (0x%x) "
+		  "ops %ps caller %pS actor %ps",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		   __entry->ino,
+		   __entry->pos,
+		   __entry->length,
+		   __print_flags(__entry->flags, "|", IOMAP_FLAGS_STRINGS),
+		   __entry->flags,
+		   __entry->ops,
+		   (void *)__entry->caller,
+		   __entry->actor)
+);
+
+#endif /* _IOMAP_TRACE_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace
+#include <trace/define_trace.h>
diff --git a/fs/isofs/Kconfig b/fs/isofs/Kconfig
index 5e7419599f50..08ffd37b9bb8 100644
--- a/fs/isofs/Kconfig
+++ b/fs/isofs/Kconfig
@@ -8,7 +8,7 @@ config ISO9660_FS
 	  long Unix filenames and symbolic links are also supported by this
 	  driver.  If you have a CD-ROM drive and want to do more with it than
 	  just listen to audio CDs and watch its LEDs, say Y (and read
-	  <file:Documentation/filesystems/isofs.txt> and the CD-ROM-HOWTO,
+	  <file:Documentation/filesystems/isofs.rst> and the CD-ROM-HOWTO,
 	  available from <http://www.tldp.org/docs.html#howto>), thereby
 	  enlarging your kernel by about 27 KB; otherwise say N.
 
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 62c0462dc89f..d634561f871a 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -544,43 +544,41 @@ static int isofs_show_options(struct seq_file *m, struct dentry *root)
 
 static unsigned int isofs_get_last_session(struct super_block *sb, s32 session)
 {
-	struct cdrom_multisession ms_info;
-	unsigned int vol_desc_start;
-	struct block_device *bdev = sb->s_bdev;
-	int i;
+	struct cdrom_device_info *cdi = disk_to_cdi(sb->s_bdev->bd_disk);
+	unsigned int vol_desc_start = 0;
 
-	vol_desc_start=0;
-	ms_info.addr_format=CDROM_LBA;
 	if (session > 0) {
-		struct cdrom_tocentry Te;
-		Te.cdte_track=session;
-		Te.cdte_format=CDROM_LBA;
-		i = ioctl_by_bdev(bdev, CDROMREADTOCENTRY, (unsigned long) &Te);
-		if (!i) {
+		struct cdrom_tocentry te;
+
+		if (!cdi)
+			return 0;
+
+		te.cdte_track = session;
+		te.cdte_format = CDROM_LBA;
+		if (cdrom_read_tocentry(cdi, &te) == 0) {
 			printk(KERN_DEBUG "ISOFS: Session %d start %d type %d\n",
-				session, Te.cdte_addr.lba,
-				Te.cdte_ctrl&CDROM_DATA_TRACK);
-			if ((Te.cdte_ctrl&CDROM_DATA_TRACK) == 4)
-				return Te.cdte_addr.lba;
+				session, te.cdte_addr.lba,
+				te.cdte_ctrl & CDROM_DATA_TRACK);
+			if ((te.cdte_ctrl & CDROM_DATA_TRACK) == 4)
+				return te.cdte_addr.lba;
 		}
 
 		printk(KERN_ERR "ISOFS: Invalid session number or type of track\n");
 	}
-	i = ioctl_by_bdev(bdev, CDROMMULTISESSION, (unsigned long) &ms_info);
-	if (session > 0)
-		printk(KERN_ERR "ISOFS: Invalid session number\n");
-#if 0
-	printk(KERN_DEBUG "isofs.inode: CDROMMULTISESSION: rc=%d\n",i);
-	if (i==0) {
-		printk(KERN_DEBUG "isofs.inode: XA disk: %s\n",ms_info.xa_flag?"yes":"no");
-		printk(KERN_DEBUG "isofs.inode: vol_desc_start = %d\n", ms_info.addr.lba);
-	}
-#endif
-	if (i==0)
+
+	if (cdi) {
+		struct cdrom_multisession ms_info;
+
+		ms_info.addr_format = CDROM_LBA;
+		if (cdrom_multisession(cdi, &ms_info) == 0) {
 #if WE_OBEY_THE_WRITTEN_STANDARDS
-		if (ms_info.xa_flag) /* necessary for a valid ms_info.addr */
+			/* necessary for a valid ms_info.addr */
+			if (ms_info.xa_flag)
 #endif
-			vol_desc_start=ms_info.addr.lba;
+				vol_desc_start = ms_info.addr.lba;
+		}
+	}
+
 	return vol_desc_start;
 }
 
@@ -1185,10 +1183,9 @@ static int isofs_readpage(struct file *file, struct page *page)
 	return mpage_readpage(page, isofs_get_block);
 }
 
-static int isofs_readpages(struct file *file, struct address_space *mapping,
-			struct list_head *pages, unsigned nr_pages)
+static void isofs_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, isofs_get_block);
+	mpage_readahead(rac, isofs_get_block);
 }
 
 static sector_t _isofs_bmap(struct address_space *mapping, sector_t block)
@@ -1198,7 +1195,7 @@ static sector_t _isofs_bmap(struct address_space *mapping, sector_t block)
 
 static const struct address_space_operations isofs_aops = {
 	.readpage = isofs_readpage,
-	.readpages = isofs_readpages,
+	.readahead = isofs_readahead,
 	.bmap = _isofs_bmap
 };
 
diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index a1909066bde6..263f02ad8ebf 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -110,7 +110,7 @@ void __jbd2_log_wait_for_space(journal_t *journal)
 	int nblocks, space_left;
 	/* assert_spin_locked(&journal->j_state_lock); */
 
-	nblocks = jbd2_space_needed(journal);
+	nblocks = journal->j_max_transaction_buffers;
 	while (jbd2_log_space_left(journal) < nblocks) {
 		write_unlock(&journal->j_state_lock);
 		mutex_lock_io(&journal->j_checkpoint_mutex);
@@ -164,7 +164,7 @@ void __jbd2_log_wait_for_space(journal_t *journal)
 				       "journal space in %s\n", __func__,
 				       journal->j_devname);
 				WARN_ON(1);
-				jbd2_journal_abort(journal, 0);
+				jbd2_journal_abort(journal, -EIO);
 			}
 			write_lock(&journal->j_state_lock);
 		} else {
@@ -414,7 +414,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	 * jbd2_cleanup_journal_tail() doesn't get called all that often.
 	 */
 	if (journal->j_flags & JBD2_BARRIER)
-		blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS, NULL);
+		blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS);
 
 	return __jbd2_update_log_tail(journal, first_tid, blocknr);
 }
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 132fb92098c7..6d2da8ad0e6f 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -482,10 +482,10 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		if (jh->b_committed_data) {
 			struct buffer_head *bh = jh2bh(jh);
 
-			jbd_lock_bh_state(bh);
+			spin_lock(&jh->b_state_lock);
 			jbd2_free(jh->b_committed_data, bh->b_size);
 			jh->b_committed_data = NULL;
-			jbd_unlock_bh_state(bh);
+			spin_unlock(&jh->b_state_lock);
 		}
 		jbd2_journal_refile_buffer(journal, jh);
 	}
@@ -560,8 +560,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	stats.run.rs_logging = jiffies;
 	stats.run.rs_flushing = jbd2_time_diff(stats.run.rs_flushing,
 					       stats.run.rs_logging);
-	stats.run.rs_blocks =
-		atomic_read(&commit_transaction->t_outstanding_credits);
+	stats.run.rs_blocks = commit_transaction->t_nr_buffers;
 	stats.run.rs_blocks_logged = 0;
 
 	J_ASSERT(commit_transaction->t_nr_buffers <=
@@ -642,8 +641,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 
 		/*
 		 * start_this_handle() uses t_outstanding_credits to determine
-		 * the free space in the log, but this counter is changed
-		 * by jbd2_journal_next_log_block() also.
+		 * the free space in the log.
 		 */
 		atomic_dec(&commit_transaction->t_outstanding_credits);
 
@@ -727,7 +725,6 @@ start_journal_io:
 				submit_bh(REQ_OP_WRITE, REQ_SYNC, bh);
 			}
 			cond_resched();
-			stats.run.rs_blocks_logged += bufs;
 
 			/* Force a new descriptor to be generated next
                            time round the loop. */
@@ -778,14 +775,14 @@ start_journal_io:
 	if (commit_transaction->t_need_data_flush &&
 	    (journal->j_fs_dev != journal->j_dev) &&
 	    (journal->j_flags & JBD2_BARRIER))
-		blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS, NULL);
+		blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS);
 
 	/* Done it all: now write the commit record asynchronously. */
 	if (jbd2_has_feature_async_commit(journal)) {
 		err = journal_submit_commit_record(journal, commit_transaction,
 						 &cbh, crc32_sum);
 		if (err)
-			__jbd2_journal_abort_hard(journal);
+			jbd2_journal_abort(journal, err);
 	}
 
 	blk_finish_plug(&plug);
@@ -814,6 +811,7 @@ start_journal_io:
 		if (unlikely(!buffer_uptodate(bh)))
 			err = -EIO;
 		jbd2_unfile_log_bh(bh);
+		stats.run.rs_blocks_logged++;
 
 		/*
 		 * The list contains temporary buffer heads created by
@@ -859,6 +857,7 @@ start_journal_io:
 		BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile");
 		clear_buffer_jwrite(bh);
 		jbd2_unfile_log_bh(bh);
+		stats.run.rs_blocks_logged++;
 		__brelse(bh);		/* One for getblk */
 		/* AKPM: bforget here */
 	}
@@ -876,18 +875,22 @@ start_journal_io:
 		err = journal_submit_commit_record(journal, commit_transaction,
 						&cbh, crc32_sum);
 		if (err)
-			__jbd2_journal_abort_hard(journal);
+			jbd2_journal_abort(journal, err);
 	}
 	if (cbh)
 		err = journal_wait_on_commit_record(journal, cbh);
+	stats.run.rs_blocks_logged++;
 	if (jbd2_has_feature_async_commit(journal) &&
 	    journal->j_flags & JBD2_BARRIER) {
-		blkdev_issue_flush(journal->j_dev, GFP_NOFS, NULL);
+		blkdev_issue_flush(journal->j_dev, GFP_NOFS);
 	}
 
 	if (err)
 		jbd2_journal_abort(journal, err);
 
+	WARN_ON_ONCE(
+		atomic_read(&commit_transaction->t_outstanding_credits) < 0);
+
 	/*
 	 * Now disk caches for filesystem device are flushed so we are safe to
 	 * erase checkpointed transactions from the log by updating journal
@@ -918,6 +921,7 @@ restart_loop:
 		transaction_t *cp_transaction;
 		struct buffer_head *bh;
 		int try_to_free = 0;
+		bool drop_ref;
 
 		jh = commit_transaction->t_forget;
 		spin_unlock(&journal->j_list_lock);
@@ -927,7 +931,7 @@ restart_loop:
 		 * done with it.
 		 */
 		get_bh(bh);
-		jbd_lock_bh_state(bh);
+		spin_lock(&jh->b_state_lock);
 		J_ASSERT_JH(jh,	jh->b_transaction == commit_transaction);
 
 		/*
@@ -972,29 +976,34 @@ restart_loop:
 		 * it. */
 
 		/*
-		* A buffer which has been freed while still being journaled by
-		* a previous transaction.
-		*/
-		if (buffer_freed(bh)) {
+		 * A buffer which has been freed while still being journaled
+		 * by a previous transaction, refile the buffer to BJ_Forget of
+		 * the running transaction. If the just committed transaction
+		 * contains "add to orphan" operation, we can completely
+		 * invalidate the buffer now. We are rather through in that
+		 * since the buffer may be still accessible when blocksize <
+		 * pagesize and it is attached to the last partial page.
+		 */
+		if (buffer_freed(bh) && !jh->b_next_transaction) {
+			struct address_space *mapping;
+
+			clear_buffer_freed(bh);
+			clear_buffer_jbddirty(bh);
+
 			/*
-			 * If the running transaction is the one containing
-			 * "add to orphan" operation (b_next_transaction !=
-			 * NULL), we have to wait for that transaction to
-			 * commit before we can really get rid of the buffer.
-			 * So just clear b_modified to not confuse transaction
-			 * credit accounting and refile the buffer to
-			 * BJ_Forget of the running transaction. If the just
-			 * committed transaction contains "add to orphan"
-			 * operation, we can completely invalidate the buffer
-			 * now. We are rather through in that since the
-			 * buffer may be still accessible when blocksize <
-			 * pagesize and it is attached to the last partial
-			 * page.
+			 * Block device buffers need to stay mapped all the
+			 * time, so it is enough to clear buffer_jbddirty and
+			 * buffer_freed bits. For the file mapping buffers (i.e.
+			 * journalled data) we need to unmap buffer and clear
+			 * more bits. We also need to be careful about the check
+			 * because the data page mapping can get cleared under
+			 * our hands. Note that if mapping == NULL, we don't
+			 * need to make buffer unmapped because the page is
+			 * already detached from the mapping and buffers cannot
+			 * get reused.
 			 */
-			jh->b_modified = 0;
-			if (!jh->b_next_transaction) {
-				clear_buffer_freed(bh);
-				clear_buffer_jbddirty(bh);
+			mapping = READ_ONCE(bh->b_page->mapping);
+			if (mapping && !sb_is_blkdev_sb(mapping->host->i_sb)) {
 				clear_buffer_mapped(bh);
 				clear_buffer_new(bh);
 				clear_buffer_req(bh);
@@ -1022,8 +1031,10 @@ restart_loop:
 				try_to_free = 1;
 		}
 		JBUFFER_TRACE(jh, "refile or unfile buffer");
-		__jbd2_journal_refile_buffer(jh);
-		jbd_unlock_bh_state(bh);
+		drop_ref = __jbd2_journal_refile_buffer(jh);
+		spin_unlock(&jh->b_state_lock);
+		if (drop_ref)
+			jbd2_journal_put_journal_head(jh);
 		if (try_to_free)
 			release_buffer_page(bh);	/* Drops bh reference */
 		else
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 1c58859aa592..a49d0e670ddf 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -96,7 +96,6 @@ EXPORT_SYMBOL(jbd2_journal_release_jbd_inode);
 EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate);
 EXPORT_SYMBOL(jbd2_inode_cache);
 
-static void __journal_abort_soft (journal_t *journal, int errno);
 static int jbd2_journal_create_slab(size_t slab_size);
 
 #ifdef CONFIG_JBD2_DEBUG
@@ -363,7 +362,7 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
 	/* keep subsequent assertions sane */
 	atomic_set(&new_bh->b_count, 1);
 
-	jbd_lock_bh_state(bh_in);
+	spin_lock(&jh_in->b_state_lock);
 repeat:
 	/*
 	 * If a new transaction has already done a buffer copy-out, then
@@ -405,13 +404,13 @@ repeat:
 	if (need_copy_out && !done_copy_out) {
 		char *tmp;
 
-		jbd_unlock_bh_state(bh_in);
+		spin_unlock(&jh_in->b_state_lock);
 		tmp = jbd2_alloc(bh_in->b_size, GFP_NOFS);
 		if (!tmp) {
 			brelse(new_bh);
 			return -ENOMEM;
 		}
-		jbd_lock_bh_state(bh_in);
+		spin_lock(&jh_in->b_state_lock);
 		if (jh_in->b_frozen_data) {
 			jbd2_free(tmp, bh_in->b_size);
 			goto repeat;
@@ -464,7 +463,7 @@ repeat:
 	__jbd2_journal_file_buffer(jh_in, transaction, BJ_Shadow);
 	spin_unlock(&journal->j_list_lock);
 	set_buffer_shadow(bh_in);
-	jbd_unlock_bh_state(bh_in);
+	spin_unlock(&jh_in->b_state_lock);
 
 	return do_escape | (done_copy_out << 1);
 }
@@ -795,18 +794,22 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr,
 {
 	int err = 0;
 	unsigned long long ret;
+	sector_t block = 0;
 
 	if (journal->j_inode) {
-		ret = bmap(journal->j_inode, blocknr);
-		if (ret)
-			*retp = ret;
-		else {
+		block = blocknr;
+		ret = bmap(journal->j_inode, &block);
+
+		if (ret || !block) {
 			printk(KERN_ALERT "%s: journal block not found "
 					"at offset %lu on %s\n",
 			       __func__, blocknr, journal->j_devname);
 			err = -EIO;
-			__journal_abort_soft(journal, err);
+			jbd2_journal_abort(journal, err);
+		} else {
+			*retp = block;
 		}
+
 	} else {
 		*retp = blocknr; /* +journal->j_blk_offset */
 	}
@@ -840,6 +843,7 @@ jbd2_journal_get_descriptor_buffer(transaction_t *transaction, int type)
 	bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
 	if (!bh)
 		return NULL;
+	atomic_dec(&transaction->t_outstanding_credits);
 	lock_buffer(bh);
 	memset(bh->b_data, 0, journal->j_blocksize);
 	header = (journal_header_t *)bh->b_data;
@@ -981,6 +985,7 @@ static void *jbd2_seq_info_start(struct seq_file *seq, loff_t *pos)
 
 static void *jbd2_seq_info_next(struct seq_file *seq, void *v, loff_t *pos)
 {
+	(*pos)++;
 	return NULL;
 }
 
@@ -1073,12 +1078,11 @@ static int jbd2_seq_info_release(struct inode *inode, struct file *file)
 	return seq_release(inode, file);
 }
 
-static const struct file_operations jbd2_seq_info_fops = {
-	.owner		= THIS_MODULE,
-	.open           = jbd2_seq_info_open,
-	.read           = seq_read,
-	.llseek         = seq_lseek,
-	.release        = jbd2_seq_info_release,
+static const struct proc_ops jbd2_info_proc_ops = {
+	.proc_open	= jbd2_seq_info_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= jbd2_seq_info_release,
 };
 
 static struct proc_dir_entry *proc_jbd2_stats;
@@ -1088,7 +1092,7 @@ static void jbd2_stats_proc_init(journal_t *journal)
 	journal->j_proc_entry = proc_mkdir(journal->j_devname, proc_jbd2_stats);
 	if (journal->j_proc_entry) {
 		proc_create_data("info", S_IRUGO, journal->j_proc_entry,
-				 &jbd2_seq_info_fops, journal);
+				 &jbd2_info_proc_ops, journal);
 	}
 }
 
@@ -1098,6 +1102,16 @@ static void jbd2_stats_proc_exit(journal_t *journal)
 	remove_proc_entry(journal->j_devname, proc_jbd2_stats);
 }
 
+/* Minimum size of descriptor tag */
+static int jbd2_min_tag_size(void)
+{
+	/*
+	 * Tag with 32-bit block numbers does not use last four bytes of the
+	 * structure
+	 */
+	return sizeof(journal_block_tag_t) - 4;
+}
+
 /*
  * Management for journal control blocks: functions to create and
  * destroy journal_t structures, and to initialise and read existing
@@ -1156,7 +1170,8 @@ static journal_t *journal_init_common(struct block_device *bdev,
 	journal->j_fs_dev = fs_dev;
 	journal->j_blk_offset = start;
 	journal->j_maxlen = len;
-	n = journal->j_blocksize / sizeof(journal_block_tag_t);
+	/* We need enough buffers to write out full descriptor block. */
+	n = journal->j_blocksize / jbd2_min_tag_size();
 	journal->j_wbufsize = n;
 	journal->j_wbuf = kmalloc_array(n, sizeof(struct buffer_head *),
 					GFP_KERNEL);
@@ -1232,11 +1247,14 @@ journal_t *jbd2_journal_init_dev(struct block_device *bdev,
 journal_t *jbd2_journal_init_inode(struct inode *inode)
 {
 	journal_t *journal;
+	sector_t blocknr;
 	char *p;
-	unsigned long long blocknr;
+	int err = 0;
+
+	blocknr = 0;
+	err = bmap(inode, &blocknr);
 
-	blocknr = bmap(inode, 0);
-	if (!blocknr) {
+	if (err || !blocknr) {
 		pr_err("%s: Cannot locate journal superblock\n",
 			__func__);
 		return NULL;
@@ -1488,6 +1506,21 @@ void jbd2_journal_update_sb_errno(journal_t *journal)
 }
 EXPORT_SYMBOL(jbd2_journal_update_sb_errno);
 
+static int journal_revoke_records_per_block(journal_t *journal)
+{
+	int record_size;
+	int space = journal->j_blocksize - sizeof(jbd2_journal_revoke_header_t);
+
+	if (jbd2_has_feature_64bit(journal))
+		record_size = 8;
+	else
+		record_size = 4;
+
+	if (jbd2_journal_has_csum_v2or3(journal))
+		space -= sizeof(struct jbd2_journal_block_tail);
+	return space / record_size;
+}
+
 /*
  * Read the superblock for a given journal, performing initial
  * validation of the format.
@@ -1596,6 +1629,8 @@ static int journal_get_superblock(journal_t *journal)
 						   sizeof(sb->s_uuid));
 	}
 
+	journal->j_revoke_records_per_block =
+				journal_revoke_records_per_block(journal);
 	set_buffer_verified(bh);
 
 	return 0;
@@ -1681,6 +1716,11 @@ int jbd2_journal_load(journal_t *journal)
 		       journal->j_devname);
 		return -EFSCORRUPTED;
 	}
+	/*
+	 * clear JBD2_ABORT flag initialized in journal_init_common
+	 * here to update log tail information with the newest seq.
+	 */
+	journal->j_flags &= ~JBD2_ABORT;
 
 	/* OK, we've finished with the dynamic journal bits:
 	 * reinitialise the dynamic contents of the superblock in memory
@@ -1688,7 +1728,6 @@ int jbd2_journal_load(journal_t *journal)
 	if (journal_reset(journal))
 		goto recovery_error;
 
-	journal->j_flags &= ~JBD2_ABORT;
 	journal->j_flags |= JBD2_LOADED;
 	return 0;
 
@@ -1916,6 +1955,8 @@ int jbd2_journal_set_features (journal_t *journal, unsigned long compat,
 	sb->s_feature_ro_compat |= cpu_to_be32(ro);
 	sb->s_feature_incompat  |= cpu_to_be32(incompat);
 	unlock_buffer(journal->j_sb_buffer);
+	journal->j_revoke_records_per_block =
+				journal_revoke_records_per_block(journal);
 
 	return 1;
 #undef COMPAT_FEATURE_ON
@@ -1946,6 +1987,8 @@ void jbd2_journal_clear_features(journal_t *journal, unsigned long compat,
 	sb->s_feature_compat    &= ~cpu_to_be32(compat);
 	sb->s_feature_ro_compat &= ~cpu_to_be32(ro);
 	sb->s_feature_incompat  &= ~cpu_to_be32(incompat);
+	journal->j_revoke_records_per_block =
+				journal_revoke_records_per_block(journal);
 }
 EXPORT_SYMBOL(jbd2_journal_clear_features);
 
@@ -2065,67 +2108,6 @@ int jbd2_journal_wipe(journal_t *journal, int write)
 	return err;
 }
 
-/*
- * Journal abort has very specific semantics, which we describe
- * for journal abort.
- *
- * Two internal functions, which provide abort to the jbd layer
- * itself are here.
- */
-
-/*
- * Quick version for internal journal use (doesn't lock the journal).
- * Aborts hard --- we mark the abort as occurred, but do _nothing_ else,
- * and don't attempt to make any other journal updates.
- */
-void __jbd2_journal_abort_hard(journal_t *journal)
-{
-	transaction_t *transaction;
-
-	if (journal->j_flags & JBD2_ABORT)
-		return;
-
-	printk(KERN_ERR "Aborting journal on device %s.\n",
-	       journal->j_devname);
-
-	write_lock(&journal->j_state_lock);
-	journal->j_flags |= JBD2_ABORT;
-	transaction = journal->j_running_transaction;
-	if (transaction)
-		__jbd2_log_start_commit(journal, transaction->t_tid);
-	write_unlock(&journal->j_state_lock);
-}
-
-/* Soft abort: record the abort error status in the journal superblock,
- * but don't do any other IO. */
-static void __journal_abort_soft (journal_t *journal, int errno)
-{
-	int old_errno;
-
-	write_lock(&journal->j_state_lock);
-	old_errno = journal->j_errno;
-	if (!journal->j_errno || errno == -ESHUTDOWN)
-		journal->j_errno = errno;
-
-	if (journal->j_flags & JBD2_ABORT) {
-		write_unlock(&journal->j_state_lock);
-		if (!old_errno && old_errno != -ESHUTDOWN &&
-		    errno == -ESHUTDOWN)
-			jbd2_journal_update_sb_errno(journal);
-		return;
-	}
-	write_unlock(&journal->j_state_lock);
-
-	__jbd2_journal_abort_hard(journal);
-
-	if (errno) {
-		jbd2_journal_update_sb_errno(journal);
-		write_lock(&journal->j_state_lock);
-		journal->j_flags |= JBD2_REC_ERR;
-		write_unlock(&journal->j_state_lock);
-	}
-}
-
 /**
  * void jbd2_journal_abort () - Shutdown the journal immediately.
  * @journal: the journal to shutdown.
@@ -2165,16 +2147,51 @@ static void __journal_abort_soft (journal_t *journal, int errno)
  * failure to disk.  ext3_error, for example, now uses this
  * functionality.
  *
- * Errors which originate from within the journaling layer will NOT
- * supply an errno; a null errno implies that absolutely no further
- * writes are done to the journal (unless there are any already in
- * progress).
- *
  */
 
 void jbd2_journal_abort(journal_t *journal, int errno)
 {
-	__journal_abort_soft(journal, errno);
+	transaction_t *transaction;
+
+	/*
+	 * ESHUTDOWN always takes precedence because a file system check
+	 * caused by any other journal abort error is not required after
+	 * a shutdown triggered.
+	 */
+	write_lock(&journal->j_state_lock);
+	if (journal->j_flags & JBD2_ABORT) {
+		int old_errno = journal->j_errno;
+
+		write_unlock(&journal->j_state_lock);
+		if (old_errno != -ESHUTDOWN && errno == -ESHUTDOWN) {
+			journal->j_errno = errno;
+			jbd2_journal_update_sb_errno(journal);
+		}
+		return;
+	}
+
+	/*
+	 * Mark the abort as occurred and start current running transaction
+	 * to release all journaled buffer.
+	 */
+	pr_err("Aborting journal on device %s.\n", journal->j_devname);
+
+	journal->j_flags |= JBD2_ABORT;
+	journal->j_errno = errno;
+	transaction = journal->j_running_transaction;
+	if (transaction)
+		__jbd2_log_start_commit(journal, transaction->t_tid);
+	write_unlock(&journal->j_state_lock);
+
+	/*
+	 * Record errno to the journal super block, so that fsck and jbd2
+	 * layer could realise that a filesystem check is needed.
+	 */
+	jbd2_journal_update_sb_errno(journal);
+
+	write_lock(&journal->j_state_lock);
+	journal->j_flags |= JBD2_REC_ERR;
+	write_unlock(&journal->j_state_lock);
 }
 
 /**
@@ -2410,6 +2427,8 @@ static struct journal_head *journal_alloc_journal_head(void)
 		ret = kmem_cache_zalloc(jbd2_journal_head_cache,
 				GFP_NOFS | __GFP_NOFAIL);
 	}
+	if (ret)
+		spin_lock_init(&ret->b_state_lock);
 	return ret;
 }
 
@@ -2521,7 +2540,6 @@ static void __journal_remove_journal_head(struct buffer_head *bh)
 {
 	struct journal_head *jh = bh2jh(bh);
 
-	J_ASSERT_JH(jh, jh->b_jcount >= 0);
 	J_ASSERT_JH(jh, jh->b_transaction == NULL);
 	J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
 	J_ASSERT_JH(jh, jh->b_cp_transaction == NULL);
@@ -2529,17 +2547,23 @@ static void __journal_remove_journal_head(struct buffer_head *bh)
 	J_ASSERT_BH(bh, buffer_jbd(bh));
 	J_ASSERT_BH(bh, jh2bh(jh) == bh);
 	BUFFER_TRACE(bh, "remove journal_head");
+
+	/* Unlink before dropping the lock */
+	bh->b_private = NULL;
+	jh->b_bh = NULL;	/* debug, really */
+	clear_buffer_jbd(bh);
+}
+
+static void journal_release_journal_head(struct journal_head *jh, size_t b_size)
+{
 	if (jh->b_frozen_data) {
 		printk(KERN_WARNING "%s: freeing b_frozen_data\n", __func__);
-		jbd2_free(jh->b_frozen_data, bh->b_size);
+		jbd2_free(jh->b_frozen_data, b_size);
 	}
 	if (jh->b_committed_data) {
 		printk(KERN_WARNING "%s: freeing b_committed_data\n", __func__);
-		jbd2_free(jh->b_committed_data, bh->b_size);
+		jbd2_free(jh->b_committed_data, b_size);
 	}
-	bh->b_private = NULL;
-	jh->b_bh = NULL;	/* debug, really */
-	clear_buffer_jbd(bh);
 	journal_free_journal_head(jh);
 }
 
@@ -2557,9 +2581,11 @@ void jbd2_journal_put_journal_head(struct journal_head *jh)
 	if (!jh->b_jcount) {
 		__journal_remove_journal_head(bh);
 		jbd_unlock_bh_journal_head(bh);
+		journal_release_journal_head(jh, bh->b_size);
 		__brelse(bh);
-	} else
+	} else {
 		jbd_unlock_bh_journal_head(bh);
+	}
 }
 
 /*
diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c
index a4967b27ffb6..2ed278f0dced 100644
--- a/fs/jbd2/recovery.c
+++ b/fs/jbd2/recovery.c
@@ -286,7 +286,7 @@ int jbd2_journal_recover(journal_t *journal)
 		err = err2;
 	/* Make sure all replayed data is on permanent storage */
 	if (journal->j_flags & JBD2_BARRIER) {
-		err2 = blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
+		err2 = blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL);
 		if (!err)
 			err = err2;
 	}
diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c
index f08073d7bbf5..fa608788b93d 100644
--- a/fs/jbd2/revoke.c
+++ b/fs/jbd2/revoke.c
@@ -371,6 +371,11 @@ int jbd2_journal_revoke(handle_t *handle, unsigned long long blocknr,
 	}
 #endif
 
+	if (WARN_ON_ONCE(handle->h_revoke_credits <= 0)) {
+		if (!bh_in)
+			brelse(bh);
+		return -EIO;
+	}
 	/* We really ought not ever to revoke twice in a row without
            first having the revoke cancelled: it's illegal to free a
            block twice without allocating it in between! */
@@ -391,6 +396,7 @@ int jbd2_journal_revoke(handle_t *handle, unsigned long long blocknr,
 			__brelse(bh);
 		}
 	}
+	handle->h_revoke_credits--;
 
 	jbd_debug(2, "insert revoke for block %llu, bh_in=%p\n",blocknr, bh_in);
 	err = insert_revoke_hash(journal, blocknr,
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index bee8498d7792..3dccc23cf010 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -63,6 +63,28 @@ void jbd2_journal_free_transaction(transaction_t *transaction)
 }
 
 /*
+ * Base amount of descriptor blocks we reserve for each transaction.
+ */
+static int jbd2_descriptor_blocks_per_trans(journal_t *journal)
+{
+	int tag_space = journal->j_blocksize - sizeof(journal_header_t);
+	int tags_per_block;
+
+	/* Subtract UUID */
+	tag_space -= 16;
+	if (jbd2_journal_has_csum_v2or3(journal))
+		tag_space -= sizeof(struct jbd2_journal_block_tail);
+	/* Commit code leaves a slack space of 16 bytes at the end of block */
+	tags_per_block = (tag_space - 16) / journal_tag_bytes(journal);
+	/*
+	 * Revoke descriptors are accounted separately so we need to reserve
+	 * space for commit block and normal transaction descriptor blocks.
+	 */
+	return 1 + DIV_ROUND_UP(journal->j_max_transaction_buffers,
+				tags_per_block);
+}
+
+/*
  * jbd2_get_transaction: obtain a new transaction_t object.
  *
  * Simply initialise a new transaction. Initialize it in
@@ -88,7 +110,9 @@ static void jbd2_get_transaction(journal_t *journal,
 	spin_lock_init(&transaction->t_handle_lock);
 	atomic_set(&transaction->t_updates, 0);
 	atomic_set(&transaction->t_outstanding_credits,
+		   jbd2_descriptor_blocks_per_trans(journal) +
 		   atomic_read(&journal->j_reserved_credits));
+	atomic_set(&transaction->t_outstanding_revokes, 0);
 	atomic_set(&transaction->t_handle_count, 0);
 	INIT_LIST_HEAD(&transaction->t_inode_list);
 	INIT_LIST_HEAD(&transaction->t_private_list);
@@ -258,12 +282,13 @@ static int add_transaction_credits(journal_t *journal, int blocks,
 	 * *before* starting to dirty potentially checkpointed buffers
 	 * in the new transaction.
 	 */
-	if (jbd2_log_space_left(journal) < jbd2_space_needed(journal)) {
+	if (jbd2_log_space_left(journal) < journal->j_max_transaction_buffers) {
 		atomic_sub(total, &t->t_outstanding_credits);
 		read_unlock(&journal->j_state_lock);
 		jbd2_might_wait_for_commit(journal);
 		write_lock(&journal->j_state_lock);
-		if (jbd2_log_space_left(journal) < jbd2_space_needed(journal))
+		if (jbd2_log_space_left(journal) <
+					journal->j_max_transaction_buffers)
 			__jbd2_log_wait_for_space(journal);
 		write_unlock(&journal->j_state_lock);
 		return 1;
@@ -299,12 +324,12 @@ static int start_this_handle(journal_t *journal, handle_t *handle,
 			     gfp_t gfp_mask)
 {
 	transaction_t	*transaction, *new_transaction = NULL;
-	int		blocks = handle->h_buffer_credits;
+	int		blocks = handle->h_total_credits;
 	int		rsv_blocks = 0;
 	unsigned long ts = jiffies;
 
 	if (handle->h_rsv_handle)
-		rsv_blocks = handle->h_rsv_handle->h_buffer_credits;
+		rsv_blocks = handle->h_rsv_handle->h_total_credits;
 
 	/*
 	 * Limit the number of reserved credits to 1/2 of maximum transaction
@@ -405,6 +430,7 @@ repeat:
 	update_t_max_wait(transaction, ts);
 	handle->h_transaction = transaction;
 	handle->h_requested_credits = blocks;
+	handle->h_revoke_credits_requested = handle->h_revoke_credits;
 	handle->h_start_jiffies = jiffies;
 	atomic_inc(&transaction->t_updates);
 	atomic_inc(&transaction->t_handle_count);
@@ -431,15 +457,15 @@ static handle_t *new_handle(int nblocks)
 	handle_t *handle = jbd2_alloc_handle(GFP_NOFS);
 	if (!handle)
 		return NULL;
-	handle->h_buffer_credits = nblocks;
+	handle->h_total_credits = nblocks;
 	handle->h_ref = 1;
 
 	return handle;
 }
 
 handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int rsv_blocks,
-			      gfp_t gfp_mask, unsigned int type,
-			      unsigned int line_no)
+			      int revoke_records, gfp_t gfp_mask,
+			      unsigned int type, unsigned int line_no)
 {
 	handle_t *handle = journal_current_handle();
 	int err;
@@ -453,6 +479,8 @@ handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int rsv_blocks,
 		return handle;
 	}
 
+	nblocks += DIV_ROUND_UP(revoke_records,
+				journal->j_revoke_records_per_block);
 	handle = new_handle(nblocks);
 	if (!handle)
 		return ERR_PTR(-ENOMEM);
@@ -468,6 +496,7 @@ handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int rsv_blocks,
 		rsv_handle->h_journal = journal;
 		handle->h_rsv_handle = rsv_handle;
 	}
+	handle->h_revoke_credits = revoke_records;
 
 	err = start_this_handle(journal, handle, gfp_mask);
 	if (err < 0) {
@@ -496,7 +525,7 @@ EXPORT_SYMBOL(jbd2__journal_start);
  * modified buffers in the log.  We block until the log can guarantee
  * that much space. Additionally, if rsv_blocks > 0, we also create another
  * handle with rsv_blocks reserved blocks in the journal. This handle is
- * is stored in h_rsv_handle. It is not attached to any particular transaction
+ * stored in h_rsv_handle. It is not attached to any particular transaction
  * and thus doesn't block transaction commit. If the caller uses this reserved
  * handle, it has to set h_rsv_handle to NULL as otherwise jbd2_journal_stop()
  * on the parent handle will dispose the reserved one. Reserved handle has to
@@ -508,16 +537,21 @@ EXPORT_SYMBOL(jbd2__journal_start);
  */
 handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
 {
-	return jbd2__journal_start(journal, nblocks, 0, GFP_NOFS, 0, 0);
+	return jbd2__journal_start(journal, nblocks, 0, 0, GFP_NOFS, 0, 0);
 }
 EXPORT_SYMBOL(jbd2_journal_start);
 
-void jbd2_journal_free_reserved(handle_t *handle)
+static void __jbd2_journal_unreserve_handle(handle_t *handle)
 {
 	journal_t *journal = handle->h_journal;
 
 	WARN_ON(!handle->h_reserved);
-	sub_reserved_credits(journal, handle->h_buffer_credits);
+	sub_reserved_credits(journal, handle->h_total_credits);
+}
+
+void jbd2_journal_free_reserved(handle_t *handle)
+{
+	__jbd2_journal_unreserve_handle(handle);
 	jbd2_free_handle(handle);
 }
 EXPORT_SYMBOL(jbd2_journal_free_reserved);
@@ -571,7 +605,7 @@ int jbd2_journal_start_reserved(handle_t *handle, unsigned int type,
 	handle->h_line_no = line_no;
 	trace_jbd2_handle_start(journal->j_fs_dev->bd_dev,
 				handle->h_transaction->t_tid, type,
-				line_no, handle->h_buffer_credits);
+				line_no, handle->h_total_credits);
 	return 0;
 }
 EXPORT_SYMBOL(jbd2_journal_start_reserved);
@@ -580,6 +614,7 @@ EXPORT_SYMBOL(jbd2_journal_start_reserved);
  * int jbd2_journal_extend() - extend buffer credits.
  * @handle:  handle to 'extend'
  * @nblocks: nr blocks to try to extend by.
+ * @revoke_records: number of revoke records to try to extend by.
  *
  * Some transactions, such as large extends and truncates, can be done
  * atomically all at once or in several stages.  The operation requests
@@ -596,7 +631,7 @@ EXPORT_SYMBOL(jbd2_journal_start_reserved);
  * return code < 0 implies an error
  * return code > 0 implies normal transaction-full status.
  */
-int jbd2_journal_extend(handle_t *handle, int nblocks)
+int jbd2_journal_extend(handle_t *handle, int nblocks, int revoke_records)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal;
@@ -618,6 +653,12 @@ int jbd2_journal_extend(handle_t *handle, int nblocks)
 		goto error_out;
 	}
 
+	nblocks += DIV_ROUND_UP(
+			handle->h_revoke_credits_requested + revoke_records,
+			journal->j_revoke_records_per_block) -
+		DIV_ROUND_UP(
+			handle->h_revoke_credits_requested,
+			journal->j_revoke_records_per_block);
 	spin_lock(&transaction->t_handle_lock);
 	wanted = atomic_add_return(nblocks,
 				   &transaction->t_outstanding_credits);
@@ -629,22 +670,16 @@ int jbd2_journal_extend(handle_t *handle, int nblocks)
 		goto unlock;
 	}
 
-	if (wanted + (wanted >> JBD2_CONTROL_BLOCKS_SHIFT) >
-	    jbd2_log_space_left(journal)) {
-		jbd_debug(3, "denied handle %p %d blocks: "
-			  "insufficient log space\n", handle, nblocks);
-		atomic_sub(nblocks, &transaction->t_outstanding_credits);
-		goto unlock;
-	}
-
 	trace_jbd2_handle_extend(journal->j_fs_dev->bd_dev,
 				 transaction->t_tid,
 				 handle->h_type, handle->h_line_no,
-				 handle->h_buffer_credits,
+				 handle->h_total_credits,
 				 nblocks);
 
-	handle->h_buffer_credits += nblocks;
+	handle->h_total_credits += nblocks;
 	handle->h_requested_credits += nblocks;
+	handle->h_revoke_credits += revoke_records;
+	handle->h_revoke_credits_requested += revoke_records;
 	result = 0;
 
 	jbd_debug(3, "extended handle %p by %d\n", handle, nblocks);
@@ -655,11 +690,55 @@ error_out:
 	return result;
 }
 
+static void stop_this_handle(handle_t *handle)
+{
+	transaction_t *transaction = handle->h_transaction;
+	journal_t *journal = transaction->t_journal;
+	int revokes;
+
+	J_ASSERT(journal_current_handle() == handle);
+	J_ASSERT(atomic_read(&transaction->t_updates) > 0);
+	current->journal_info = NULL;
+	/*
+	 * Subtract necessary revoke descriptor blocks from handle credits. We
+	 * take care to account only for revoke descriptor blocks the
+	 * transaction will really need as large sequences of transactions with
+	 * small numbers of revokes are relatively common.
+	 */
+	revokes = handle->h_revoke_credits_requested - handle->h_revoke_credits;
+	if (revokes) {
+		int t_revokes, revoke_descriptors;
+		int rr_per_blk = journal->j_revoke_records_per_block;
+
+		WARN_ON_ONCE(DIV_ROUND_UP(revokes, rr_per_blk)
+				> handle->h_total_credits);
+		t_revokes = atomic_add_return(revokes,
+				&transaction->t_outstanding_revokes);
+		revoke_descriptors =
+			DIV_ROUND_UP(t_revokes, rr_per_blk) -
+			DIV_ROUND_UP(t_revokes - revokes, rr_per_blk);
+		handle->h_total_credits -= revoke_descriptors;
+	}
+	atomic_sub(handle->h_total_credits,
+		   &transaction->t_outstanding_credits);
+	if (handle->h_rsv_handle)
+		__jbd2_journal_unreserve_handle(handle->h_rsv_handle);
+	if (atomic_dec_and_test(&transaction->t_updates))
+		wake_up(&journal->j_wait_updates);
+
+	rwsem_release(&journal->j_trans_commit_map, _THIS_IP_);
+	/*
+	 * Scope of the GFP_NOFS context is over here and so we can restore the
+	 * original alloc context.
+	 */
+	memalloc_nofs_restore(handle->saved_alloc_context);
+}
 
 /**
  * int jbd2_journal_restart() - restart a handle .
  * @handle:  handle to restart
  * @nblocks: nr credits requested
+ * @revoke_records: number of revoke record credits requested
  * @gfp_mask: memory allocation flags (for start_this_handle)
  *
  * Restart a handle for a multi-transaction filesystem
@@ -672,56 +751,48 @@ error_out:
  * credits. We preserve reserved handle if there's any attached to the
  * passed in handle.
  */
-int jbd2__journal_restart(handle_t *handle, int nblocks, gfp_t gfp_mask)
+int jbd2__journal_restart(handle_t *handle, int nblocks, int revoke_records,
+			  gfp_t gfp_mask)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal;
 	tid_t		tid;
-	int		need_to_start, ret;
+	int		need_to_start;
+	int		ret;
 
 	/* If we've had an abort of any type, don't even think about
 	 * actually doing the restart! */
 	if (is_handle_aborted(handle))
 		return 0;
 	journal = transaction->t_journal;
+	tid = transaction->t_tid;
 
 	/*
 	 * First unlink the handle from its current transaction, and start the
 	 * commit on that.
 	 */
-	J_ASSERT(atomic_read(&transaction->t_updates) > 0);
-	J_ASSERT(journal_current_handle() == handle);
-
-	read_lock(&journal->j_state_lock);
-	spin_lock(&transaction->t_handle_lock);
-	atomic_sub(handle->h_buffer_credits,
-		   &transaction->t_outstanding_credits);
-	if (handle->h_rsv_handle) {
-		sub_reserved_credits(journal,
-				     handle->h_rsv_handle->h_buffer_credits);
-	}
-	if (atomic_dec_and_test(&transaction->t_updates))
-		wake_up(&journal->j_wait_updates);
-	tid = transaction->t_tid;
-	spin_unlock(&transaction->t_handle_lock);
+	jbd_debug(2, "restarting handle %p\n", handle);
+	stop_this_handle(handle);
 	handle->h_transaction = NULL;
-	current->journal_info = NULL;
 
-	jbd_debug(2, "restarting handle %p\n", handle);
+	/*
+	 * TODO: If we use READ_ONCE / WRITE_ONCE for j_commit_request we can
+ 	 * get rid of pointless j_state_lock traffic like this.
+	 */
+	read_lock(&journal->j_state_lock);
 	need_to_start = !tid_geq(journal->j_commit_request, tid);
 	read_unlock(&journal->j_state_lock);
 	if (need_to_start)
 		jbd2_log_start_commit(journal, tid);
-
-	rwsem_release(&journal->j_trans_commit_map, 1, _THIS_IP_);
-	handle->h_buffer_credits = nblocks;
-	/*
-	 * Restore the original nofs context because the journal restart
-	 * is basically the same thing as journal stop and start.
-	 * start_this_handle will start a new nofs context.
-	 */
-	memalloc_nofs_restore(handle->saved_alloc_context);
+	handle->h_total_credits = nblocks +
+		DIV_ROUND_UP(revoke_records,
+			     journal->j_revoke_records_per_block);
+	handle->h_revoke_credits = revoke_records;
 	ret = start_this_handle(journal, handle, gfp_mask);
+	trace_jbd2_handle_restart(journal->j_fs_dev->bd_dev,
+				 ret ? 0 : handle->h_transaction->t_tid,
+				 handle->h_type, handle->h_line_no,
+				 handle->h_total_credits);
 	return ret;
 }
 EXPORT_SYMBOL(jbd2__journal_restart);
@@ -729,7 +800,7 @@ EXPORT_SYMBOL(jbd2__journal_restart);
 
 int jbd2_journal_restart(handle_t *handle, int nblocks)
 {
-	return jbd2__journal_restart(handle, nblocks, GFP_NOFS);
+	return jbd2__journal_restart(handle, nblocks, 0, GFP_NOFS);
 }
 EXPORT_SYMBOL(jbd2_journal_restart);
 
@@ -865,8 +936,6 @@ do_get_write_access(handle_t *handle, struct journal_head *jh,
 	char *frozen_buffer = NULL;
 	unsigned long start_lock, time_lock;
 
-	if (is_handle_aborted(handle))
-		return -EROFS;
 	journal = transaction->t_journal;
 
 	jbd_debug(5, "journal_head %p, force_copy %d\n", jh, force_copy);
@@ -879,7 +948,7 @@ repeat:
 
  	start_lock = jiffies;
 	lock_buffer(bh);
-	jbd_lock_bh_state(bh);
+	spin_lock(&jh->b_state_lock);
 
 	/* If it takes too long to lock the buffer, trace it */
 	time_lock = jbd2_time_diff(start_lock, jiffies);
@@ -929,7 +998,7 @@ repeat:
 
 	error = -EROFS;
 	if (is_handle_aborted(handle)) {
-		jbd_unlock_bh_state(bh);
+		spin_unlock(&jh->b_state_lock);
 		goto out;
 	}
 	error = 0;
@@ -993,7 +1062,7 @@ repeat:
 	 */
 	if (buffer_shadow(bh)) {
 		JBUFFER_TRACE(jh, "on shadow: sleep");
-		jbd_unlock_bh_state(bh);
+		spin_unlock(&jh->b_state_lock);
 		wait_on_bit_io(&bh->b_state, BH_Shadow, TASK_UNINTERRUPTIBLE);
 		goto repeat;
 	}
@@ -1014,7 +1083,7 @@ repeat:
 		JBUFFER_TRACE(jh, "generate frozen data");
 		if (!frozen_buffer) {
 			JBUFFER_TRACE(jh, "allocate memory for buffer");
-			jbd_unlock_bh_state(bh);
+			spin_unlock(&jh->b_state_lock);
 			frozen_buffer = jbd2_alloc(jh2bh(jh)->b_size,
 						   GFP_NOFS | __GFP_NOFAIL);
 			goto repeat;
@@ -1033,7 +1102,7 @@ attach_next:
 	jh->b_next_transaction = transaction;
 
 done:
-	jbd_unlock_bh_state(bh);
+	spin_unlock(&jh->b_state_lock);
 
 	/*
 	 * If we are about to journal a buffer, then any revoke pending on it is
@@ -1081,8 +1150,8 @@ static bool jbd2_write_access_granted(handle_t *handle, struct buffer_head *bh,
 	/* For undo access buffer must have data copied */
 	if (undo && !jh->b_committed_data)
 		goto out;
-	if (jh->b_transaction != handle->h_transaction &&
-	    jh->b_next_transaction != handle->h_transaction)
+	if (READ_ONCE(jh->b_transaction) != handle->h_transaction &&
+	    READ_ONCE(jh->b_next_transaction) != handle->h_transaction)
 		goto out;
 	/*
 	 * There are two reasons for the barrier here:
@@ -1118,6 +1187,9 @@ int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh)
 	struct journal_head *jh;
 	int rc;
 
+	if (is_handle_aborted(handle))
+		return -EROFS;
+
 	if (jbd2_write_access_granted(handle, bh, false))
 		return 0;
 
@@ -1172,7 +1244,7 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
 	 * that case: the transaction must have deleted the buffer for it to be
 	 * reused here.
 	 */
-	jbd_lock_bh_state(bh);
+	spin_lock(&jh->b_state_lock);
 	J_ASSERT_JH(jh, (jh->b_transaction == transaction ||
 		jh->b_transaction == NULL ||
 		(jh->b_transaction == journal->j_committing_transaction &&
@@ -1207,7 +1279,7 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
 		jh->b_next_transaction = transaction;
 		spin_unlock(&journal->j_list_lock);
 	}
-	jbd_unlock_bh_state(bh);
+	spin_unlock(&jh->b_state_lock);
 
 	/*
 	 * akpm: I added this.  ext3_alloc_branch can pick up new indirect
@@ -1255,6 +1327,9 @@ int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
 	struct journal_head *jh;
 	char *committed_data = NULL;
 
+	if (is_handle_aborted(handle))
+		return -EROFS;
+
 	if (jbd2_write_access_granted(handle, bh, true))
 		return 0;
 
@@ -1275,13 +1350,13 @@ repeat:
 		committed_data = jbd2_alloc(jh2bh(jh)->b_size,
 					    GFP_NOFS|__GFP_NOFAIL);
 
-	jbd_lock_bh_state(bh);
+	spin_lock(&jh->b_state_lock);
 	if (!jh->b_committed_data) {
 		/* Copy out the current buffer contents into the
 		 * preserved, committed copy. */
 		JBUFFER_TRACE(jh, "generate b_committed data");
 		if (!committed_data) {
-			jbd_unlock_bh_state(bh);
+			spin_unlock(&jh->b_state_lock);
 			goto repeat;
 		}
 
@@ -1289,7 +1364,7 @@ repeat:
 		committed_data = NULL;
 		memcpy(jh->b_committed_data, bh->b_data, bh->b_size);
 	}
-	jbd_unlock_bh_state(bh);
+	spin_unlock(&jh->b_state_lock);
 out:
 	jbd2_journal_put_journal_head(jh);
 	if (unlikely(committed_data))
@@ -1390,16 +1465,16 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 	 */
 	if (jh->b_transaction != transaction &&
 	    jh->b_next_transaction != transaction) {
-		jbd_lock_bh_state(bh);
+		spin_lock(&jh->b_state_lock);
 		J_ASSERT_JH(jh, jh->b_transaction == transaction ||
 				jh->b_next_transaction == transaction);
-		jbd_unlock_bh_state(bh);
+		spin_unlock(&jh->b_state_lock);
 	}
 	if (jh->b_modified == 1) {
 		/* If it's in our transaction it must be in BJ_Metadata list. */
 		if (jh->b_transaction == transaction &&
 		    jh->b_jlist != BJ_Metadata) {
-			jbd_lock_bh_state(bh);
+			spin_lock(&jh->b_state_lock);
 			if (jh->b_transaction == transaction &&
 			    jh->b_jlist != BJ_Metadata)
 				pr_err("JBD2: assertion failure: h_type=%u "
@@ -1409,13 +1484,13 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 				       jh->b_jlist);
 			J_ASSERT_JH(jh, jh->b_transaction != transaction ||
 					jh->b_jlist == BJ_Metadata);
-			jbd_unlock_bh_state(bh);
+			spin_unlock(&jh->b_state_lock);
 		}
 		goto out;
 	}
 
 	journal = transaction->t_journal;
-	jbd_lock_bh_state(bh);
+	spin_lock(&jh->b_state_lock);
 
 	if (jh->b_modified == 0) {
 		/*
@@ -1423,12 +1498,12 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 		 * of the transaction. This needs to be done
 		 * once a transaction -bzzz
 		 */
-		if (handle->h_buffer_credits <= 0) {
+		if (WARN_ON_ONCE(jbd2_handle_buffer_credits(handle) <= 0)) {
 			ret = -ENOSPC;
 			goto out_unlock_bh;
 		}
 		jh->b_modified = 1;
-		handle->h_buffer_credits--;
+		handle->h_total_credits--;
 	}
 
 	/*
@@ -1501,7 +1576,7 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 	__jbd2_journal_file_buffer(jh, transaction, BJ_Metadata);
 	spin_unlock(&journal->j_list_lock);
 out_unlock_bh:
-	jbd_unlock_bh_state(bh);
+	spin_unlock(&jh->b_state_lock);
 out:
 	JBUFFER_TRACE(jh, "exit");
 	return ret;
@@ -1524,7 +1599,7 @@ out:
  * Allow this call even if the handle has aborted --- it may be part of
  * the caller's cleanup after an abort.
  */
-int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
+int jbd2_journal_forget(handle_t *handle, struct buffer_head *bh)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal;
@@ -1539,18 +1614,20 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 
 	BUFFER_TRACE(bh, "entry");
 
-	jbd_lock_bh_state(bh);
+	jh = jbd2_journal_grab_journal_head(bh);
+	if (!jh) {
+		__bforget(bh);
+		return 0;
+	}
 
-	if (!buffer_jbd(bh))
-		goto not_jbd;
-	jh = bh2jh(bh);
+	spin_lock(&jh->b_state_lock);
 
 	/* Critical error: attempting to delete a bitmap buffer, maybe?
 	 * Don't do any jbd operations, and return an error. */
 	if (!J_EXPECT_JH(jh, !jh->b_committed_data,
 			 "inconsistent data on disk")) {
 		err = -EIO;
-		goto not_jbd;
+		goto drop;
 	}
 
 	/* keep track of whether or not this transaction modified us */
@@ -1598,10 +1675,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 			__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
 		} else {
 			__jbd2_journal_unfile_buffer(jh);
-			if (!buffer_jbd(bh)) {
-				spin_unlock(&journal->j_list_lock);
-				goto not_jbd;
-			}
+			jbd2_journal_put_journal_head(jh);
 		}
 		spin_unlock(&journal->j_list_lock);
 	} else if (jh->b_transaction) {
@@ -1643,7 +1717,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 		if (!jh->b_cp_transaction) {
 			JBUFFER_TRACE(jh, "belongs to none transaction");
 			spin_unlock(&journal->j_list_lock);
-			goto not_jbd;
+			goto drop;
 		}
 
 		/*
@@ -1653,7 +1727,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 		if (!buffer_dirty(bh)) {
 			__jbd2_journal_remove_checkpoint(jh);
 			spin_unlock(&journal->j_list_lock);
-			goto not_jbd;
+			goto drop;
 		}
 
 		/*
@@ -1666,20 +1740,15 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 		__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
 		spin_unlock(&journal->j_list_lock);
 	}
-
-	jbd_unlock_bh_state(bh);
-	__brelse(bh);
 drop:
+	__brelse(bh);
+	spin_unlock(&jh->b_state_lock);
+	jbd2_journal_put_journal_head(jh);
 	if (drop_reserve) {
 		/* no need to reserve log space for this block -bzzz */
-		handle->h_buffer_credits++;
+		handle->h_total_credits++;
 	}
 	return err;
-
-not_jbd:
-	jbd_unlock_bh_state(bh);
-	__bforget(bh);
-	goto drop;
 }
 
 /**
@@ -1706,45 +1775,34 @@ int jbd2_journal_stop(handle_t *handle)
 	tid_t tid;
 	pid_t pid;
 
+	if (--handle->h_ref > 0) {
+		jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1,
+						 handle->h_ref);
+		if (is_handle_aborted(handle))
+			return -EIO;
+		return 0;
+	}
 	if (!transaction) {
 		/*
-		 * Handle is already detached from the transaction so
-		 * there is nothing to do other than decrease a refcount,
-		 * or free the handle if refcount drops to zero
+		 * Handle is already detached from the transaction so there is
+		 * nothing to do other than free the handle.
 		 */
-		if (--handle->h_ref > 0) {
-			jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1,
-							 handle->h_ref);
-			return err;
-		} else {
-			if (handle->h_rsv_handle)
-				jbd2_free_handle(handle->h_rsv_handle);
-			goto free_and_exit;
-		}
+		memalloc_nofs_restore(handle->saved_alloc_context);
+		goto free_and_exit;
 	}
 	journal = transaction->t_journal;
-
-	J_ASSERT(journal_current_handle() == handle);
+	tid = transaction->t_tid;
 
 	if (is_handle_aborted(handle))
 		err = -EIO;
-	else
-		J_ASSERT(atomic_read(&transaction->t_updates) > 0);
-
-	if (--handle->h_ref > 0) {
-		jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1,
-			  handle->h_ref);
-		return err;
-	}
 
 	jbd_debug(4, "Handle %p going down\n", handle);
 	trace_jbd2_handle_stats(journal->j_fs_dev->bd_dev,
-				transaction->t_tid,
-				handle->h_type, handle->h_line_no,
+				tid, handle->h_type, handle->h_line_no,
 				jiffies - handle->h_start_jiffies,
 				handle->h_sync, handle->h_requested_credits,
 				(handle->h_requested_credits -
-				 handle->h_buffer_credits));
+				 handle->h_total_credits));
 
 	/*
 	 * Implement synchronous transaction batching.  If the handle
@@ -1804,19 +1862,13 @@ int jbd2_journal_stop(handle_t *handle)
 
 	if (handle->h_sync)
 		transaction->t_synchronous_commit = 1;
-	current->journal_info = NULL;
-	atomic_sub(handle->h_buffer_credits,
-		   &transaction->t_outstanding_credits);
 
 	/*
 	 * If the handle is marked SYNC, we need to set another commit
-	 * going!  We also want to force a commit if the current
-	 * transaction is occupying too much of the log, or if the
-	 * transaction is too old now.
+	 * going!  We also want to force a commit if the transaction is too
+	 * old now.
 	 */
 	if (handle->h_sync ||
-	    (atomic_read(&transaction->t_outstanding_credits) >
-	     journal->j_max_transaction_buffers) ||
 	    time_after_eq(jiffies, transaction->t_expires)) {
 		/* Do this even for aborted journals: an abort still
 		 * completes the commit thread, it just doesn't write
@@ -1825,7 +1877,7 @@ int jbd2_journal_stop(handle_t *handle)
 		jbd_debug(2, "transaction too old, requesting commit for "
 					"handle %p\n", handle);
 		/* This is non-blocking */
-		jbd2_log_start_commit(journal, transaction->t_tid);
+		jbd2_log_start_commit(journal, tid);
 
 		/*
 		 * Special case: JBD2_SYNC synchronous updates require us
@@ -1836,31 +1888,19 @@ int jbd2_journal_stop(handle_t *handle)
 	}
 
 	/*
-	 * Once we drop t_updates, if it goes to zero the transaction
-	 * could start committing on us and eventually disappear.  So
-	 * once we do this, we must not dereference transaction
-	 * pointer again.
+	 * Once stop_this_handle() drops t_updates, the transaction could start
+	 * committing on us and eventually disappear.  So we must not
+	 * dereference transaction pointer again after calling
+	 * stop_this_handle().
 	 */
-	tid = transaction->t_tid;
-	if (atomic_dec_and_test(&transaction->t_updates)) {
-		wake_up(&journal->j_wait_updates);
-		if (journal->j_barrier_count)
-			wake_up(&journal->j_wait_transaction_locked);
-	}
-
-	rwsem_release(&journal->j_trans_commit_map, 1, _THIS_IP_);
+	stop_this_handle(handle);
 
 	if (wait_for_commit)
 		err = jbd2_log_wait_commit(journal, tid);
 
-	if (handle->h_rsv_handle)
-		jbd2_journal_free_reserved(handle->h_rsv_handle);
 free_and_exit:
-	/*
-	 * Scope of the GFP_NOFS context is over here and so we can restore the
-	 * original alloc context.
-	 */
-	memalloc_nofs_restore(handle->saved_alloc_context);
+	if (handle->h_rsv_handle)
+		jbd2_free_handle(handle->h_rsv_handle);
 	jbd2_free_handle(handle);
 	return err;
 }
@@ -1878,7 +1918,7 @@ free_and_exit:
  *
  * j_list_lock is held.
  *
- * jbd_lock_bh_state(jh2bh(jh)) is held.
+ * jh->b_state_lock is held.
  */
 
 static inline void
@@ -1902,7 +1942,7 @@ __blist_add_buffer(struct journal_head **list, struct journal_head *jh)
  *
  * Called with j_list_lock held, and the journal may not be locked.
  *
- * jbd_lock_bh_state(jh2bh(jh)) is held.
+ * jh->b_state_lock is held.
  */
 
 static inline void
@@ -1934,7 +1974,7 @@ static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
 	transaction_t *transaction;
 	struct buffer_head *bh = jh2bh(jh);
 
-	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
+	lockdep_assert_held(&jh->b_state_lock);
 	transaction = jh->b_transaction;
 	if (transaction)
 		assert_spin_locked(&transaction->t_journal->j_list_lock);
@@ -1971,17 +2011,15 @@ static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
 }
 
 /*
- * Remove buffer from all transactions.
+ * Remove buffer from all transactions. The caller is responsible for dropping
+ * the jh reference that belonged to the transaction.
  *
  * Called with bh_state lock and j_list_lock
- *
- * jh and bh may be already freed when this function returns.
  */
 static void __jbd2_journal_unfile_buffer(struct journal_head *jh)
 {
 	__jbd2_journal_temp_unlink_buffer(jh);
 	jh->b_transaction = NULL;
-	jbd2_journal_put_journal_head(jh);
 }
 
 void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh)
@@ -1990,18 +2028,19 @@ void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh)
 
 	/* Get reference so that buffer cannot be freed before we unlock it */
 	get_bh(bh);
-	jbd_lock_bh_state(bh);
+	spin_lock(&jh->b_state_lock);
 	spin_lock(&journal->j_list_lock);
 	__jbd2_journal_unfile_buffer(jh);
 	spin_unlock(&journal->j_list_lock);
-	jbd_unlock_bh_state(bh);
+	spin_unlock(&jh->b_state_lock);
+	jbd2_journal_put_journal_head(jh);
 	__brelse(bh);
 }
 
 /*
  * Called from jbd2_journal_try_to_free_buffers().
  *
- * Called under jbd_lock_bh_state(bh)
+ * Called under jh->b_state_lock
  */
 static void
 __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh)
@@ -2088,10 +2127,10 @@ int jbd2_journal_try_to_free_buffers(journal_t *journal,
 		if (!jh)
 			continue;
 
-		jbd_lock_bh_state(bh);
+		spin_lock(&jh->b_state_lock);
 		__journal_try_to_free_buffer(journal, bh);
+		spin_unlock(&jh->b_state_lock);
 		jbd2_journal_put_journal_head(jh);
-		jbd_unlock_bh_state(bh);
 		if (buffer_jbd(bh))
 			goto busy;
 	} while ((bh = bh->b_this_page) != head);
@@ -2112,7 +2151,7 @@ busy:
  *
  * Called under j_list_lock.
  *
- * Called under jbd_lock_bh_state(bh).
+ * Called under jh->b_state_lock.
  */
 static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
 {
@@ -2133,6 +2172,7 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
 	} else {
 		JBUFFER_TRACE(jh, "on running transaction");
 		__jbd2_journal_unfile_buffer(jh);
+		jbd2_journal_put_journal_head(jh);
 	}
 	return may_free;
 }
@@ -2199,18 +2239,15 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh,
 	 * holding the page lock. --sct
 	 */
 
-	if (!buffer_jbd(bh))
+	jh = jbd2_journal_grab_journal_head(bh);
+	if (!jh)
 		goto zap_buffer_unlocked;
 
 	/* OK, we have data buffer in journaled mode */
 	write_lock(&journal->j_state_lock);
-	jbd_lock_bh_state(bh);
+	spin_lock(&jh->b_state_lock);
 	spin_lock(&journal->j_list_lock);
 
-	jh = jbd2_journal_grab_journal_head(bh);
-	if (!jh)
-		goto zap_buffer_no_jh;
-
 	/*
 	 * We cannot remove the buffer from checkpoint lists until the
 	 * transaction adding inode to orphan list (let's call it T)
@@ -2289,25 +2326,27 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh,
 		 * for commit and try again.
 		 */
 		if (partial_page) {
-			jbd2_journal_put_journal_head(jh);
 			spin_unlock(&journal->j_list_lock);
-			jbd_unlock_bh_state(bh);
+			spin_unlock(&jh->b_state_lock);
 			write_unlock(&journal->j_state_lock);
+			jbd2_journal_put_journal_head(jh);
 			return -EBUSY;
 		}
 		/*
-		 * OK, buffer won't be reachable after truncate. We just set
-		 * j_next_transaction to the running transaction (if there is
-		 * one) and mark buffer as freed so that commit code knows it
-		 * should clear dirty bits when it is done with the buffer.
+		 * OK, buffer won't be reachable after truncate. We just clear
+		 * b_modified to not confuse transaction credit accounting, and
+		 * set j_next_transaction to the running transaction (if there
+		 * is one) and mark buffer as freed so that commit code knows
+		 * it should clear dirty bits when it is done with the buffer.
 		 */
 		set_buffer_freed(bh);
 		if (journal->j_running_transaction && buffer_jbddirty(bh))
 			jh->b_next_transaction = journal->j_running_transaction;
-		jbd2_journal_put_journal_head(jh);
+		jh->b_modified = 0;
 		spin_unlock(&journal->j_list_lock);
-		jbd_unlock_bh_state(bh);
+		spin_unlock(&jh->b_state_lock);
 		write_unlock(&journal->j_state_lock);
+		jbd2_journal_put_journal_head(jh);
 		return 0;
 	} else {
 		/* Good, the buffer belongs to the running transaction.
@@ -2331,11 +2370,10 @@ zap_buffer:
 	 * here.
 	 */
 	jh->b_modified = 0;
-	jbd2_journal_put_journal_head(jh);
-zap_buffer_no_jh:
 	spin_unlock(&journal->j_list_lock);
-	jbd_unlock_bh_state(bh);
+	spin_unlock(&jh->b_state_lock);
 	write_unlock(&journal->j_state_lock);
+	jbd2_journal_put_journal_head(jh);
 zap_buffer_unlocked:
 	clear_buffer_dirty(bh);
 	J_ASSERT_BH(bh, !buffer_jbddirty(bh));
@@ -2422,7 +2460,7 @@ void __jbd2_journal_file_buffer(struct journal_head *jh,
 	int was_dirty = 0;
 	struct buffer_head *bh = jh2bh(jh);
 
-	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
+	lockdep_assert_held(&jh->b_state_lock);
 	assert_spin_locked(&transaction->t_journal->j_list_lock);
 
 	J_ASSERT_JH(jh, jh->b_jlist < BJ_Types);
@@ -2484,11 +2522,11 @@ void __jbd2_journal_file_buffer(struct journal_head *jh,
 void jbd2_journal_file_buffer(struct journal_head *jh,
 				transaction_t *transaction, int jlist)
 {
-	jbd_lock_bh_state(jh2bh(jh));
+	spin_lock(&jh->b_state_lock);
 	spin_lock(&transaction->t_journal->j_list_lock);
 	__jbd2_journal_file_buffer(jh, transaction, jlist);
 	spin_unlock(&transaction->t_journal->j_list_lock);
-	jbd_unlock_bh_state(jh2bh(jh));
+	spin_unlock(&jh->b_state_lock);
 }
 
 /*
@@ -2498,23 +2536,25 @@ void jbd2_journal_file_buffer(struct journal_head *jh,
  * buffer on that transaction's metadata list.
  *
  * Called under j_list_lock
- * Called under jbd_lock_bh_state(jh2bh(jh))
+ * Called under jh->b_state_lock
  *
- * jh and bh may be already free when this function returns
+ * When this function returns true, there's no next transaction to refile to
+ * and the caller has to drop jh reference through
+ * jbd2_journal_put_journal_head().
  */
-void __jbd2_journal_refile_buffer(struct journal_head *jh)
+bool __jbd2_journal_refile_buffer(struct journal_head *jh)
 {
 	int was_dirty, jlist;
 	struct buffer_head *bh = jh2bh(jh);
 
-	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
+	lockdep_assert_held(&jh->b_state_lock);
 	if (jh->b_transaction)
 		assert_spin_locked(&jh->b_transaction->t_journal->j_list_lock);
 
 	/* If the buffer is now unused, just drop it. */
 	if (jh->b_next_transaction == NULL) {
 		__jbd2_journal_unfile_buffer(jh);
-		return;
+		return true;
 	}
 
 	/*
@@ -2529,8 +2569,8 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh)
 	 * our jh reference and thus __jbd2_journal_file_buffer() must not
 	 * take a new one.
 	 */
-	jh->b_transaction = jh->b_next_transaction;
-	jh->b_next_transaction = NULL;
+	WRITE_ONCE(jh->b_transaction, jh->b_next_transaction);
+	WRITE_ONCE(jh->b_next_transaction, NULL);
 	if (buffer_freed(bh))
 		jlist = BJ_Forget;
 	else if (jh->b_modified)
@@ -2542,6 +2582,7 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh)
 
 	if (was_dirty)
 		set_buffer_jbddirty(bh);
+	return false;
 }
 
 /*
@@ -2552,16 +2593,15 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh)
  */
 void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh)
 {
-	struct buffer_head *bh = jh2bh(jh);
+	bool drop;
 
-	/* Get reference so that buffer cannot be freed before we unlock it */
-	get_bh(bh);
-	jbd_lock_bh_state(bh);
+	spin_lock(&jh->b_state_lock);
 	spin_lock(&journal->j_list_lock);
-	__jbd2_journal_refile_buffer(jh);
-	jbd_unlock_bh_state(bh);
+	drop = __jbd2_journal_refile_buffer(jh);
+	spin_unlock(&jh->b_state_lock);
 	spin_unlock(&journal->j_list_lock);
-	__brelse(bh);
+	if (drop)
+		jbd2_journal_put_journal_head(jh);
 }
 
 /*
diff --git a/fs/jffs2/nodelist.c b/fs/jffs2/nodelist.c
index 021a4a2190ee..b86c78d178c6 100644
--- a/fs/jffs2/nodelist.c
+++ b/fs/jffs2/nodelist.c
@@ -226,7 +226,7 @@ static int jffs2_add_frag_to_fragtree(struct jffs2_sb_info *c, struct rb_root *r
 		lastend = this->ofs + this->size;
 	} else {
 		dbg_fragtree2("lookup gave no frag\n");
-		return -EINVAL;
+		lastend = 0;
 	}
 
 	/* See if we ran off the end of the fragtree */
diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c
index 0e6406c4f362..05d7878dfad1 100644
--- a/fs/jffs2/super.c
+++ b/fs/jffs2/super.c
@@ -167,27 +167,21 @@ enum {
 	Opt_rp_size,
 };
 
-static const struct fs_parameter_spec jffs2_param_specs[] = {
-	fsparam_enum	("compr",	Opt_override_compr),
-	fsparam_u32	("rp_size",	Opt_rp_size),
-	{}
-};
-
-static const struct fs_parameter_enum jffs2_param_enums[] = {
-	{ Opt_override_compr,	"none",	JFFS2_COMPR_MODE_NONE },
+static const struct constant_table jffs2_param_compr[] = {
+	{"none",	JFFS2_COMPR_MODE_NONE },
 #ifdef CONFIG_JFFS2_LZO
-	{ Opt_override_compr,	"lzo",	JFFS2_COMPR_MODE_FORCELZO },
+	{"lzo",		JFFS2_COMPR_MODE_FORCELZO },
 #endif
 #ifdef CONFIG_JFFS2_ZLIB
-	{ Opt_override_compr,	"zlib",	JFFS2_COMPR_MODE_FORCEZLIB },
+	{"zlib",	JFFS2_COMPR_MODE_FORCEZLIB },
 #endif
 	{}
 };
 
-const struct fs_parameter_description jffs2_fs_parameters = {
-	.name		= "jffs2",
-	.specs		= jffs2_param_specs,
-	.enums		= jffs2_param_enums,
+static const struct fs_parameter_spec jffs2_fs_parameters[] = {
+	fsparam_enum	("compr",	Opt_override_compr, jffs2_param_compr),
+	fsparam_u32	("rp_size",	Opt_rp_size),
+	{}
 };
 
 static int jffs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
@@ -196,7 +190,7 @@ static int jffs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
 	struct jffs2_sb_info *c = fc->s_fs_info;
 	int opt;
 
-	opt = fs_parse(fc, &jffs2_fs_parameters, param, &result);
+	opt = fs_parse(fc, jffs2_fs_parameters, param, &result);
 	if (opt < 0)
 		return opt;
 
@@ -339,7 +333,7 @@ static struct file_system_type jffs2_fs_type = {
 	.owner =	THIS_MODULE,
 	.name =		"jffs2",
 	.init_fs_context = jffs2_init_fs_context,
-	.parameters =	&jffs2_fs_parameters,
+	.parameters =	jffs2_fs_parameters,
 	.kill_sb =	jffs2_kill_sb,
 };
 MODULE_ALIAS_FS("jffs2");
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 9486afcdac76..6f65bfa9f18d 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -296,10 +296,9 @@ static int jfs_readpage(struct file *file, struct page *page)
 	return mpage_readpage(page, jfs_get_block);
 }
 
-static int jfs_readpages(struct file *file, struct address_space *mapping,
-		struct list_head *pages, unsigned nr_pages)
+static void jfs_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, jfs_get_block);
+	mpage_readahead(rac, jfs_get_block);
 }
 
 static void jfs_write_failed(struct address_space *mapping, loff_t to)
@@ -358,7 +357,7 @@ static ssize_t jfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 
 const struct address_space_operations jfs_aops = {
 	.readpage	= jfs_readpage,
-	.readpages	= jfs_readpages,
+	.readahead	= jfs_readahead,
 	.writepage	= jfs_writepage,
 	.writepages	= jfs_writepages,
 	.write_begin	= jfs_write_begin,
diff --git a/fs/jfs/jfs_debug.c b/fs/jfs/jfs_debug.c
index 888cdd685a1e..44b62b3c322e 100644
--- a/fs/jfs/jfs_debug.c
+++ b/fs/jfs/jfs_debug.c
@@ -43,12 +43,12 @@ static ssize_t jfs_loglevel_proc_write(struct file *file,
 	return count;
 }
 
-static const struct file_operations jfs_loglevel_proc_fops = {
-	.open		= jfs_loglevel_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-	.write		= jfs_loglevel_proc_write,
+static const struct proc_ops jfs_loglevel_proc_ops = {
+	.proc_open	= jfs_loglevel_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+	.proc_write	= jfs_loglevel_proc_write,
 };
 #endif
 
@@ -68,7 +68,7 @@ void jfs_proc_init(void)
 #endif
 #ifdef CONFIG_JFS_DEBUG
 	proc_create_single("TxAnchor", 0, base, jfs_txanchor_proc_show);
-	proc_create("loglevel", 0, base, &jfs_loglevel_proc_fops);
+	proc_create("loglevel", 0, base, &jfs_loglevel_proc_ops);
 #endif
 }
 
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index caade185e568..7dfcab2a2da6 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -4027,7 +4027,6 @@ static int dbGetL2AGSize(s64 nblocks)
  */
 #define MAXL0PAGES	(1 + LPERCTL)
 #define MAXL1PAGES	(1 + LPERCTL * MAXL0PAGES)
-#define MAXL2PAGES	(1 + LPERCTL * MAXL1PAGES)
 
 /*
  * convert number of map pages to the zero origin top dmapctl level
diff --git a/fs/kernfs/dir.c b/fs/kernfs/dir.c
index 6ebae6bbe6a5..9aec80b9d7c6 100644
--- a/fs/kernfs/dir.c
+++ b/fs/kernfs/dir.c
@@ -438,7 +438,7 @@ void kernfs_put_active(struct kernfs_node *kn)
 		return;
 
 	if (kernfs_lockdep(kn))
-		rwsem_release(&kn->dep_map, 1, _RET_IP_);
+		rwsem_release(&kn->dep_map, _RET_IP_);
 	v = atomic_dec_return(&kn->active);
 	if (likely(v != KN_DEACTIVATED_BIAS))
 		return;
@@ -476,7 +476,7 @@ static void kernfs_drain(struct kernfs_node *kn)
 
 	if (kernfs_lockdep(kn)) {
 		lock_acquired(&kn->dep_map, _RET_IP_);
-		rwsem_release(&kn->dep_map, 1, _RET_IP_);
+		rwsem_release(&kn->dep_map, _RET_IP_);
 	}
 
 	kernfs_drain_open_files(kn);
@@ -508,10 +508,6 @@ void kernfs_put(struct kernfs_node *kn)
 	struct kernfs_node *parent;
 	struct kernfs_root *root;
 
-	/*
-	 * kernfs_node is freed with ->count 0, kernfs_find_and_get_node_by_ino
-	 * depends on this to filter reused stale node
-	 */
 	if (!kn || !atomic_dec_and_test(&kn->count))
 		return;
 	root = kernfs_root(kn);
@@ -536,7 +532,7 @@ void kernfs_put(struct kernfs_node *kn)
 		kmem_cache_free(kernfs_iattrs_cache, kn->iattr);
 	}
 	spin_lock(&kernfs_idr_lock);
-	idr_remove(&root->ino_idr, kn->id.ino);
+	idr_remove(&root->ino_idr, (u32)kernfs_ino(kn));
 	spin_unlock(&kernfs_idr_lock);
 	kmem_cache_free(kernfs_node_cache, kn);
 
@@ -621,8 +617,7 @@ static struct kernfs_node *__kernfs_new_node(struct kernfs_root *root,
 					     unsigned flags)
 {
 	struct kernfs_node *kn;
-	u32 gen;
-	int cursor;
+	u32 id_highbits;
 	int ret;
 
 	name = kstrdup_const(name, GFP_KERNEL);
@@ -635,23 +630,19 @@ static struct kernfs_node *__kernfs_new_node(struct kernfs_root *root,
 
 	idr_preload(GFP_KERNEL);
 	spin_lock(&kernfs_idr_lock);
-	cursor = idr_get_cursor(&root->ino_idr);
 	ret = idr_alloc_cyclic(&root->ino_idr, kn, 1, 0, GFP_ATOMIC);
-	if (ret >= 0 && ret < cursor)
-		root->next_generation++;
-	gen = root->next_generation;
+	if (ret >= 0 && ret < root->last_id_lowbits)
+		root->id_highbits++;
+	id_highbits = root->id_highbits;
+	root->last_id_lowbits = ret;
 	spin_unlock(&kernfs_idr_lock);
 	idr_preload_end();
 	if (ret < 0)
 		goto err_out2;
-	kn->id.ino = ret;
-	kn->id.generation = gen;
 
-	/*
-	 * set ino first. This RELEASE is paired with atomic_inc_not_zero in
-	 * kernfs_find_and_get_node_by_ino
-	 */
-	atomic_set_release(&kn->count, 1);
+	kn->id = (u64)id_highbits << 32 | ret;
+
+	atomic_set(&kn->count, 1);
 	atomic_set(&kn->active, KN_DEACTIVATED_BIAS);
 	RB_CLEAR_NODE(&kn->rb);
 
@@ -680,7 +671,7 @@ static struct kernfs_node *__kernfs_new_node(struct kernfs_root *root,
 	return kn;
 
  err_out3:
-	idr_remove(&root->ino_idr, kn->id.ino);
+	idr_remove(&root->ino_idr, (u32)kernfs_ino(kn));
  err_out2:
 	kmem_cache_free(kernfs_node_cache, kn);
  err_out1:
@@ -705,50 +696,52 @@ struct kernfs_node *kernfs_new_node(struct kernfs_node *parent,
 }
 
 /*
- * kernfs_find_and_get_node_by_ino - get kernfs_node from inode number
+ * kernfs_find_and_get_node_by_id - get kernfs_node from node id
  * @root: the kernfs root
- * @ino: inode number
+ * @id: the target node id
+ *
+ * @id's lower 32bits encode ino and upper gen.  If the gen portion is
+ * zero, all generations are matched.
  *
  * RETURNS:
  * NULL on failure. Return a kernfs node with reference counter incremented
  */
-struct kernfs_node *kernfs_find_and_get_node_by_ino(struct kernfs_root *root,
-						    unsigned int ino)
+struct kernfs_node *kernfs_find_and_get_node_by_id(struct kernfs_root *root,
+						   u64 id)
 {
 	struct kernfs_node *kn;
+	ino_t ino = kernfs_id_ino(id);
+	u32 gen = kernfs_id_gen(id);
 
-	rcu_read_lock();
-	kn = idr_find(&root->ino_idr, ino);
+	spin_lock(&kernfs_idr_lock);
+
+	kn = idr_find(&root->ino_idr, (u32)ino);
 	if (!kn)
-		goto out;
+		goto err_unlock;
 
-	/*
-	 * Since kernfs_node is freed in RCU, it's possible an old node for ino
-	 * is freed, but reused before RCU grace period. But a freed node (see
-	 * kernfs_put) or an incompletedly initialized node (see
-	 * __kernfs_new_node) should have 'count' 0. We can use this fact to
-	 * filter out such node.
-	 */
-	if (!atomic_inc_not_zero(&kn->count)) {
-		kn = NULL;
-		goto out;
+	if (sizeof(ino_t) >= sizeof(u64)) {
+		/* we looked up with the low 32bits, compare the whole */
+		if (kernfs_ino(kn) != ino)
+			goto err_unlock;
+	} else {
+		/* 0 matches all generations */
+		if (unlikely(gen && kernfs_gen(kn) != gen))
+			goto err_unlock;
 	}
 
 	/*
-	 * The node could be a new node or a reused node. If it's a new node,
-	 * we are ok. If it's reused because of RCU (because of
-	 * SLAB_TYPESAFE_BY_RCU), the __kernfs_new_node always sets its 'ino'
-	 * before 'count'. So if 'count' is uptodate, 'ino' should be uptodate,
-	 * hence we can use 'ino' to filter stale node.
+	 * ACTIVATED is protected with kernfs_mutex but it was clear when
+	 * @kn was added to idr and we just wanna see it set.  No need to
+	 * grab kernfs_mutex.
 	 */
-	if (kn->id.ino != ino)
-		goto out;
-	rcu_read_unlock();
+	if (unlikely(!(kn->flags & KERNFS_ACTIVATED) ||
+		     !atomic_inc_not_zero(&kn->count)))
+		goto err_unlock;
 
+	spin_unlock(&kernfs_idr_lock);
 	return kn;
-out:
-	rcu_read_unlock();
-	kernfs_put(kn);
+err_unlock:
+	spin_unlock(&kernfs_idr_lock);
 	return NULL;
 }
 
@@ -962,7 +955,17 @@ struct kernfs_root *kernfs_create_root(struct kernfs_syscall_ops *scops,
 
 	idr_init(&root->ino_idr);
 	INIT_LIST_HEAD(&root->supers);
-	root->next_generation = 1;
+
+	/*
+	 * On 64bit ino setups, id is ino.  On 32bit, low 32bits are ino.
+	 * High bits generation.  The starting value for both ino and
+	 * genenration is 1.  Initialize upper 32bit allocation
+	 * accordingly.
+	 */
+	if (sizeof(ino_t) >= sizeof(u64))
+		root->id_highbits = 0;
+	else
+		root->id_highbits = 1;
 
 	kn = __kernfs_new_node(root, NULL, "", S_IFDIR | S_IRUGO | S_IXUGO,
 			       GLOBAL_ROOT_UID, GLOBAL_ROOT_GID,
@@ -1263,7 +1266,7 @@ void kernfs_activate(struct kernfs_node *kn)
 
 	pos = NULL;
 	while ((pos = kernfs_next_descendant_post(pos, kn))) {
-		if (!pos || (pos->flags & KERNFS_ACTIVATED))
+		if (pos->flags & KERNFS_ACTIVATED)
 			continue;
 
 		WARN_ON_ONCE(pos->parent && RB_EMPTY_NODE(&pos->rb));
@@ -1678,7 +1681,7 @@ static int kernfs_fop_readdir(struct file *file, struct dir_context *ctx)
 		const char *name = pos->name;
 		unsigned int type = dt_type(pos);
 		int len = strlen(name);
-		ino_t ino = pos->id.ino;
+		ino_t ino = kernfs_ino(pos);
 
 		ctx->pos = pos->hash;
 		file->private_data = pos;
diff --git a/fs/kernfs/file.c b/fs/kernfs/file.c
index e8c792b49616..34366db3620d 100644
--- a/fs/kernfs/file.c
+++ b/fs/kernfs/file.c
@@ -892,7 +892,7 @@ repeat:
 		 * have the matching @file available.  Look up the inodes
 		 * and generate the events manually.
 		 */
-		inode = ilookup(info->sb, kn->id.ino);
+		inode = ilookup(info->sb, kernfs_ino(kn));
 		if (!inode)
 			continue;
 
@@ -901,7 +901,7 @@ repeat:
 		if (parent) {
 			struct inode *p_inode;
 
-			p_inode = ilookup(info->sb, parent->id.ino);
+			p_inode = ilookup(info->sb, kernfs_ino(parent));
 			if (p_inode) {
 				fsnotify(p_inode, FS_MODIFY | FS_EVENT_ON_CHILD,
 					 inode, FSNOTIFY_EVENT_INODE, &name, 0);
diff --git a/fs/kernfs/inode.c b/fs/kernfs/inode.c
index f3eaa8869f42..fc2469a20fed 100644
--- a/fs/kernfs/inode.c
+++ b/fs/kernfs/inode.c
@@ -53,6 +53,8 @@ static struct kernfs_iattrs *__kernfs_iattrs(struct kernfs_node *kn, int alloc)
 	kn->iattr->ia_ctime = kn->iattr->ia_atime;
 
 	simple_xattrs_init(&kn->iattr->xattrs);
+	atomic_set(&kn->iattr->nr_user_xattrs, 0);
+	atomic_set(&kn->iattr->user_xattr_size, 0);
 out_unlock:
 	ret = kn->iattr;
 	mutex_unlock(&iattr_mutex);
@@ -160,9 +162,9 @@ static inline void set_inode_attr(struct inode *inode,
 {
 	inode->i_uid = attrs->ia_uid;
 	inode->i_gid = attrs->ia_gid;
-	inode->i_atime = timestamp_truncate(attrs->ia_atime, inode);
-	inode->i_mtime = timestamp_truncate(attrs->ia_mtime, inode);
-	inode->i_ctime = timestamp_truncate(attrs->ia_ctime, inode);
+	inode->i_atime = attrs->ia_atime;
+	inode->i_mtime = attrs->ia_mtime;
+	inode->i_ctime = attrs->ia_ctime;
 }
 
 static void kernfs_refresh_inode(struct kernfs_node *kn, struct inode *inode)
@@ -201,7 +203,7 @@ static void kernfs_init_inode(struct kernfs_node *kn, struct inode *inode)
 	inode->i_private = kn;
 	inode->i_mapping->a_ops = &kernfs_aops;
 	inode->i_op = &kernfs_iops;
-	inode->i_generation = kn->id.generation;
+	inode->i_generation = kernfs_gen(kn);
 
 	set_default_inode_attr(inode, kn->mode);
 	kernfs_refresh_inode(kn, inode);
@@ -247,7 +249,7 @@ struct inode *kernfs_get_inode(struct super_block *sb, struct kernfs_node *kn)
 {
 	struct inode *inode;
 
-	inode = iget_locked(sb, kn->id.ino);
+	inode = iget_locked(sb, kernfs_ino(kn));
 	if (inode && (inode->i_state & I_NEW))
 		kernfs_init_inode(kn, inode);
 
@@ -303,7 +305,7 @@ int kernfs_xattr_set(struct kernfs_node *kn, const char *name,
 	if (!attrs)
 		return -ENOMEM;
 
-	return simple_xattr_set(&attrs->xattrs, name, value, size, flags);
+	return simple_xattr_set(&attrs->xattrs, name, value, size, flags, NULL);
 }
 
 static int kernfs_vfs_xattr_get(const struct xattr_handler *handler,
@@ -327,6 +329,86 @@ static int kernfs_vfs_xattr_set(const struct xattr_handler *handler,
 	return kernfs_xattr_set(kn, name, value, size, flags);
 }
 
+static int kernfs_vfs_user_xattr_add(struct kernfs_node *kn,
+				     const char *full_name,
+				     struct simple_xattrs *xattrs,
+				     const void *value, size_t size, int flags)
+{
+	atomic_t *sz = &kn->iattr->user_xattr_size;
+	atomic_t *nr = &kn->iattr->nr_user_xattrs;
+	ssize_t removed_size;
+	int ret;
+
+	if (atomic_inc_return(nr) > KERNFS_MAX_USER_XATTRS) {
+		ret = -ENOSPC;
+		goto dec_count_out;
+	}
+
+	if (atomic_add_return(size, sz) > KERNFS_USER_XATTR_SIZE_LIMIT) {
+		ret = -ENOSPC;
+		goto dec_size_out;
+	}
+
+	ret = simple_xattr_set(xattrs, full_name, value, size, flags,
+			       &removed_size);
+
+	if (!ret && removed_size >= 0)
+		size = removed_size;
+	else if (!ret)
+		return 0;
+dec_size_out:
+	atomic_sub(size, sz);
+dec_count_out:
+	atomic_dec(nr);
+	return ret;
+}
+
+static int kernfs_vfs_user_xattr_rm(struct kernfs_node *kn,
+				    const char *full_name,
+				    struct simple_xattrs *xattrs,
+				    const void *value, size_t size, int flags)
+{
+	atomic_t *sz = &kn->iattr->user_xattr_size;
+	atomic_t *nr = &kn->iattr->nr_user_xattrs;
+	ssize_t removed_size;
+	int ret;
+
+	ret = simple_xattr_set(xattrs, full_name, value, size, flags,
+			       &removed_size);
+
+	if (removed_size >= 0) {
+		atomic_sub(removed_size, sz);
+		atomic_dec(nr);
+	}
+
+	return ret;
+}
+
+static int kernfs_vfs_user_xattr_set(const struct xattr_handler *handler,
+				     struct dentry *unused, struct inode *inode,
+				     const char *suffix, const void *value,
+				     size_t size, int flags)
+{
+	const char *full_name = xattr_full_name(handler, suffix);
+	struct kernfs_node *kn = inode->i_private;
+	struct kernfs_iattrs *attrs;
+
+	if (!(kernfs_root(kn)->flags & KERNFS_ROOT_SUPPORT_USER_XATTR))
+		return -EOPNOTSUPP;
+
+	attrs = kernfs_iattrs(kn);
+	if (!attrs)
+		return -ENOMEM;
+
+	if (value)
+		return kernfs_vfs_user_xattr_add(kn, full_name, &attrs->xattrs,
+						 value, size, flags);
+	else
+		return kernfs_vfs_user_xattr_rm(kn, full_name, &attrs->xattrs,
+						value, size, flags);
+
+}
+
 static const struct xattr_handler kernfs_trusted_xattr_handler = {
 	.prefix = XATTR_TRUSTED_PREFIX,
 	.get = kernfs_vfs_xattr_get,
@@ -339,8 +421,15 @@ static const struct xattr_handler kernfs_security_xattr_handler = {
 	.set = kernfs_vfs_xattr_set,
 };
 
+static const struct xattr_handler kernfs_user_xattr_handler = {
+	.prefix = XATTR_USER_PREFIX,
+	.get = kernfs_vfs_xattr_get,
+	.set = kernfs_vfs_user_xattr_set,
+};
+
 const struct xattr_handler *kernfs_xattr_handlers[] = {
 	&kernfs_trusted_xattr_handler,
 	&kernfs_security_xattr_handler,
+	&kernfs_user_xattr_handler,
 	NULL
 };
diff --git a/fs/kernfs/kernfs-internal.h b/fs/kernfs/kernfs-internal.h
index 02ce570a9a3c..7ee97ef59184 100644
--- a/fs/kernfs/kernfs-internal.h
+++ b/fs/kernfs/kernfs-internal.h
@@ -26,6 +26,8 @@ struct kernfs_iattrs {
 	struct timespec64	ia_ctime;
 
 	struct simple_xattrs	xattrs;
+	atomic_t		nr_user_xattrs;
+	atomic_t		user_xattr_size;
 };
 
 /* +1 to avoid triggering overflow warning when negating it */
@@ -109,8 +111,6 @@ struct kernfs_node *kernfs_new_node(struct kernfs_node *parent,
 				    const char *name, umode_t mode,
 				    kuid_t uid, kgid_t gid,
 				    unsigned flags);
-struct kernfs_node *kernfs_find_and_get_node_by_ino(struct kernfs_root *root,
-						    unsigned int ino);
 
 /*
  * file.c
diff --git a/fs/kernfs/mount.c b/fs/kernfs/mount.c
index 6c12fac2c287..9dc7e7a64e10 100644
--- a/fs/kernfs/mount.c
+++ b/fs/kernfs/mount.c
@@ -53,63 +53,85 @@ const struct super_operations kernfs_sops = {
 	.show_path	= kernfs_sop_show_path,
 };
 
-/*
- * Similar to kernfs_fh_get_inode, this one gets kernfs node from inode
- * number and generation
- */
-struct kernfs_node *kernfs_get_node_by_id(struct kernfs_root *root,
-	const union kernfs_node_id *id)
+static int kernfs_encode_fh(struct inode *inode, __u32 *fh, int *max_len,
+			    struct inode *parent)
 {
-	struct kernfs_node *kn;
+	struct kernfs_node *kn = inode->i_private;
 
-	kn = kernfs_find_and_get_node_by_ino(root, id->ino);
-	if (!kn)
-		return NULL;
-	if (kn->id.generation != id->generation) {
-		kernfs_put(kn);
-		return NULL;
+	if (*max_len < 2) {
+		*max_len = 2;
+		return FILEID_INVALID;
 	}
-	return kn;
+
+	*max_len = 2;
+	*(u64 *)fh = kn->id;
+	return FILEID_KERNFS;
 }
 
-static struct inode *kernfs_fh_get_inode(struct super_block *sb,
-		u64 ino, u32 generation)
+static struct dentry *__kernfs_fh_to_dentry(struct super_block *sb,
+					    struct fid *fid, int fh_len,
+					    int fh_type, bool get_parent)
 {
 	struct kernfs_super_info *info = kernfs_info(sb);
-	struct inode *inode;
 	struct kernfs_node *kn;
+	struct inode *inode;
+	u64 id;
 
-	if (ino == 0)
-		return ERR_PTR(-ESTALE);
+	if (fh_len < 2)
+		return NULL;
+
+	switch (fh_type) {
+	case FILEID_KERNFS:
+		id = *(u64 *)fid;
+		break;
+	case FILEID_INO32_GEN:
+	case FILEID_INO32_GEN_PARENT:
+		/*
+		 * blk_log_action() exposes "LOW32,HIGH32" pair without
+		 * type and userland can call us with generic fid
+		 * constructed from them.  Combine it back to ID.  See
+		 * blk_log_action().
+		 */
+		id = ((u64)fid->i32.gen << 32) | fid->i32.ino;
+		break;
+	default:
+		return NULL;
+	}
 
-	kn = kernfs_find_and_get_node_by_ino(info->root, ino);
+	kn = kernfs_find_and_get_node_by_id(info->root, id);
 	if (!kn)
 		return ERR_PTR(-ESTALE);
+
+	if (get_parent) {
+		struct kernfs_node *parent;
+
+		parent = kernfs_get_parent(kn);
+		kernfs_put(kn);
+		kn = parent;
+		if (!kn)
+			return ERR_PTR(-ESTALE);
+	}
+
 	inode = kernfs_get_inode(sb, kn);
 	kernfs_put(kn);
 	if (!inode)
 		return ERR_PTR(-ESTALE);
 
-	if (generation && inode->i_generation != generation) {
-		/* we didn't find the right inode.. */
-		iput(inode);
-		return ERR_PTR(-ESTALE);
-	}
-	return inode;
+	return d_obtain_alias(inode);
 }
 
-static struct dentry *kernfs_fh_to_dentry(struct super_block *sb, struct fid *fid,
-		int fh_len, int fh_type)
+static struct dentry *kernfs_fh_to_dentry(struct super_block *sb,
+					  struct fid *fid, int fh_len,
+					  int fh_type)
 {
-	return generic_fh_to_dentry(sb, fid, fh_len, fh_type,
-				    kernfs_fh_get_inode);
+	return __kernfs_fh_to_dentry(sb, fid, fh_len, fh_type, false);
 }
 
-static struct dentry *kernfs_fh_to_parent(struct super_block *sb, struct fid *fid,
-		int fh_len, int fh_type)
+static struct dentry *kernfs_fh_to_parent(struct super_block *sb,
+					  struct fid *fid, int fh_len,
+					  int fh_type)
 {
-	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
-				    kernfs_fh_get_inode);
+	return __kernfs_fh_to_dentry(sb, fid, fh_len, fh_type, true);
 }
 
 static struct dentry *kernfs_get_parent_dentry(struct dentry *child)
@@ -120,6 +142,7 @@ static struct dentry *kernfs_get_parent_dentry(struct dentry *child)
 }
 
 static const struct export_operations kernfs_export_ops = {
+	.encode_fh	= kernfs_encode_fh,
 	.fh_to_dentry	= kernfs_fh_to_dentry,
 	.fh_to_parent	= kernfs_fh_to_parent,
 	.get_parent	= kernfs_get_parent_dentry,
@@ -200,7 +223,7 @@ struct dentry *kernfs_node_dentry(struct kernfs_node *kn,
 			dput(dentry);
 			return ERR_PTR(-EINVAL);
 		}
-		dtmp = lookup_one_len_unlocked(kntmp->name, dentry,
+		dtmp = lookup_positive_unlocked(kntmp->name, dentry,
 					       strlen(kntmp->name));
 		dput(dentry);
 		if (IS_ERR(dtmp))
@@ -363,18 +386,9 @@ void kernfs_kill_sb(struct super_block *sb)
 
 void __init kernfs_init(void)
 {
-
-	/*
-	 * the slab is freed in RCU context, so kernfs_find_and_get_node_by_ino
-	 * can access the slab lock free. This could introduce stale nodes,
-	 * please see how kernfs_find_and_get_node_by_ino filters out stale
-	 * nodes.
-	 */
 	kernfs_node_cache = kmem_cache_create("kernfs_node_cache",
 					      sizeof(struct kernfs_node),
-					      0,
-					      SLAB_PANIC | SLAB_TYPESAFE_BY_RCU,
-					      NULL);
+					      0, SLAB_PANIC, NULL);
 
 	/* Creates slab cache for kernfs inode attributes */
 	kernfs_iattrs_cache  = kmem_cache_create("kernfs_iattrs_cache",
diff --git a/fs/libfs.c b/fs/libfs.c
index 1463b038ffc4..4d08edf19c78 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -19,6 +19,7 @@
 #include <linux/buffer_head.h> /* sync_mapping_buffers */
 #include <linux/fs_context.h>
 #include <linux/pseudo_fs.h>
+#include <linux/fsnotify.h>
 
 #include <linux/uaccess.h>
 
@@ -239,6 +240,75 @@ const struct inode_operations simple_dir_inode_operations = {
 };
 EXPORT_SYMBOL(simple_dir_inode_operations);
 
+static struct dentry *find_next_child(struct dentry *parent, struct dentry *prev)
+{
+	struct dentry *child = NULL;
+	struct list_head *p = prev ? &prev->d_child : &parent->d_subdirs;
+
+	spin_lock(&parent->d_lock);
+	while ((p = p->next) != &parent->d_subdirs) {
+		struct dentry *d = container_of(p, struct dentry, d_child);
+		if (simple_positive(d)) {
+			spin_lock_nested(&d->d_lock, DENTRY_D_LOCK_NESTED);
+			if (simple_positive(d))
+				child = dget_dlock(d);
+			spin_unlock(&d->d_lock);
+			if (likely(child))
+				break;
+		}
+	}
+	spin_unlock(&parent->d_lock);
+	dput(prev);
+	return child;
+}
+
+void simple_recursive_removal(struct dentry *dentry,
+                              void (*callback)(struct dentry *))
+{
+	struct dentry *this = dget(dentry);
+	while (true) {
+		struct dentry *victim = NULL, *child;
+		struct inode *inode = this->d_inode;
+
+		inode_lock(inode);
+		if (d_is_dir(this))
+			inode->i_flags |= S_DEAD;
+		while ((child = find_next_child(this, victim)) == NULL) {
+			// kill and ascend
+			// update metadata while it's still locked
+			inode->i_ctime = current_time(inode);
+			clear_nlink(inode);
+			inode_unlock(inode);
+			victim = this;
+			this = this->d_parent;
+			inode = this->d_inode;
+			inode_lock(inode);
+			if (simple_positive(victim)) {
+				d_invalidate(victim);	// avoid lost mounts
+				if (d_is_dir(victim))
+					fsnotify_rmdir(inode, victim);
+				else
+					fsnotify_unlink(inode, victim);
+				if (callback)
+					callback(victim);
+				dput(victim);		// unpin it
+			}
+			if (victim == dentry) {
+				inode->i_ctime = inode->i_mtime =
+					current_time(inode);
+				if (d_is_dir(dentry))
+					drop_nlink(inode);
+				inode_unlock(inode);
+				dput(dentry);
+				return;
+			}
+		}
+		inode_unlock(inode);
+		this = child;
+	}
+}
+EXPORT_SYMBOL(simple_recursive_removal);
+
 static const struct super_operations simple_super_operations = {
 	.statfs		= simple_statfs,
 };
@@ -821,7 +891,7 @@ int simple_attr_open(struct inode *inode, struct file *file,
 {
 	struct simple_attr *attr;
 
-	attr = kmalloc(sizeof(*attr), GFP_KERNEL);
+	attr = kzalloc(sizeof(*attr), GFP_KERNEL);
 	if (!attr)
 		return -ENOMEM;
 
@@ -861,9 +931,11 @@ ssize_t simple_attr_read(struct file *file, char __user *buf,
 	if (ret)
 		return ret;
 
-	if (*ppos) {		/* continued read */
+	if (*ppos && attr->get_buf[0]) {
+		/* continued read */
 		size = strlen(attr->get_buf);
-	} else {		/* first read */
+	} else {
+		/* first read */
 		u64 val;
 		ret = attr->get(attr->data, &val);
 		if (ret)
@@ -1041,7 +1113,7 @@ int generic_file_fsync(struct file *file, loff_t start, loff_t end,
 	err = __generic_file_fsync(file, start, end, datasync);
 	if (err)
 		return err;
-	return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+	return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
 }
 EXPORT_SYMBOL(generic_file_fsync);
 
diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 7d46fafdbbe5..0afb6d59bad0 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -464,7 +464,8 @@ nlm_bind_host(struct nlm_host *host)
 			.version	= host->h_version,
 			.authflavor	= RPC_AUTH_UNIX,
 			.flags		= (RPC_CLNT_CREATE_NOPING |
-					   RPC_CLNT_CREATE_AUTOBIND),
+					   RPC_CLNT_CREATE_AUTOBIND |
+					   RPC_CLNT_CREATE_REUSEPORT),
 			.cred		= host->h_cred,
 		};
 
diff --git a/fs/lockd/procfs.c b/fs/lockd/procfs.c
index ca9228a56d65..a01f08c8c2f3 100644
--- a/fs/lockd/procfs.c
+++ b/fs/lockd/procfs.c
@@ -60,11 +60,11 @@ nlm_end_grace_read(struct file *file, char __user *buf, size_t size,
 	return simple_read_from_buffer(buf, size, pos, resp, sizeof(resp));
 }
 
-static const struct file_operations lockd_end_grace_operations = {
-	.write		= nlm_end_grace_write,
-	.read		= nlm_end_grace_read,
-	.llseek		= default_llseek,
-	.release	= simple_transaction_release,
+static const struct proc_ops lockd_end_grace_proc_ops = {
+	.proc_write	= nlm_end_grace_write,
+	.proc_read	= nlm_end_grace_read,
+	.proc_lseek	= default_llseek,
+	.proc_release	= simple_transaction_release,
 };
 
 int __init
@@ -76,7 +76,7 @@ lockd_create_procfs(void)
 	if (!entry)
 		return -ENOMEM;
 	entry = proc_create("nlm_end_grace", S_IRUGO|S_IWUSR, entry,
-				 &lockd_end_grace_operations);
+			    &lockd_end_grace_proc_ops);
 	if (!entry) {
 		remove_proc_entry("fs/lockd", NULL);
 		return -ENOMEM;
diff --git a/fs/locks.c b/fs/locks.c
index 6970f55daf54..1d4f4d5da704 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -61,7 +61,7 @@
  *
  *  Initial implementation of mandatory locks. SunOS turned out to be
  *  a rotten model, so I implemented the "obvious" semantics.
- *  See 'Documentation/filesystems/mandatory-locking.txt' for details.
+ *  See 'Documentation/filesystems/mandatory-locking.rst' for details.
  *  Andy Walker (andy@lysaker.kvaerner.no), April 06, 1996.
  *
  *  Don't allow mandatory locks on mmap()'ed files. Added simple functions to
@@ -725,7 +725,6 @@ static void __locks_delete_block(struct file_lock *waiter)
 {
 	locks_delete_global_blocked(waiter);
 	list_del_init(&waiter->fl_blocked_member);
-	waiter->fl_blocker = NULL;
 }
 
 static void __locks_wake_up_blocks(struct file_lock *blocker)
@@ -740,6 +739,13 @@ static void __locks_wake_up_blocks(struct file_lock *blocker)
 			waiter->fl_lmops->lm_notify(waiter);
 		else
 			wake_up(&waiter->fl_wait);
+
+		/*
+		 * The setting of fl_blocker to NULL marks the "done"
+		 * point in deleting a block. Paired with acquire at the top
+		 * of locks_delete_block().
+		 */
+		smp_store_release(&waiter->fl_blocker, NULL);
 	}
 }
 
@@ -754,24 +760,41 @@ int locks_delete_block(struct file_lock *waiter)
 	int status = -ENOENT;
 
 	/*
-	 * If fl_blocker is NULL, it won't be set again as this thread
-	 * "owns" the lock and is the only one that might try to claim
-	 * the lock.  So it is safe to test fl_blocker locklessly.
-	 * Also if fl_blocker is NULL, this waiter is not listed on
-	 * fl_blocked_requests for some lock, so no other request can
-	 * be added to the list of fl_blocked_requests for this
-	 * request.  So if fl_blocker is NULL, it is safe to
-	 * locklessly check if fl_blocked_requests is empty.  If both
-	 * of these checks succeed, there is no need to take the lock.
+	 * If fl_blocker is NULL, it won't be set again as this thread "owns"
+	 * the lock and is the only one that might try to claim the lock.
+	 *
+	 * We use acquire/release to manage fl_blocker so that we can
+	 * optimize away taking the blocked_lock_lock in many cases.
+	 *
+	 * The smp_load_acquire guarantees two things:
+	 *
+	 * 1/ that fl_blocked_requests can be tested locklessly. If something
+	 * was recently added to that list it must have been in a locked region
+	 * *before* the locked region when fl_blocker was set to NULL.
+	 *
+	 * 2/ that no other thread is accessing 'waiter', so it is safe to free
+	 * it.  __locks_wake_up_blocks is careful not to touch waiter after
+	 * fl_blocker is released.
+	 *
+	 * If a lockless check of fl_blocker shows it to be NULL, we know that
+	 * no new locks can be inserted into its fl_blocked_requests list, and
+	 * can avoid doing anything further if the list is empty.
 	 */
-	if (waiter->fl_blocker == NULL &&
+	if (!smp_load_acquire(&waiter->fl_blocker) &&
 	    list_empty(&waiter->fl_blocked_requests))
 		return status;
+
 	spin_lock(&blocked_lock_lock);
 	if (waiter->fl_blocker)
 		status = 0;
 	__locks_wake_up_blocks(waiter);
 	__locks_delete_block(waiter);
+
+	/*
+	 * The setting of fl_blocker to NULL marks the "done" point in deleting
+	 * a block. Paired with acquire at the top of this function.
+	 */
+	smp_store_release(&waiter->fl_blocker, NULL);
 	spin_unlock(&blocked_lock_lock);
 	return status;
 }
@@ -1364,7 +1387,8 @@ static int posix_lock_inode_wait(struct inode *inode, struct file_lock *fl)
 		error = posix_lock_inode(inode, fl, NULL);
 		if (error != FILE_LOCK_DEFERRED)
 			break;
-		error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker);
+		error = wait_event_interruptible(fl->fl_wait,
+					list_empty(&fl->fl_blocked_member));
 		if (error)
 			break;
 	}
@@ -1449,7 +1473,8 @@ int locks_mandatory_area(struct inode *inode, struct file *filp, loff_t start,
 		error = posix_lock_inode(inode, &fl, NULL);
 		if (error != FILE_LOCK_DEFERRED)
 			break;
-		error = wait_event_interruptible(fl.fl_wait, !fl.fl_blocker);
+		error = wait_event_interruptible(fl.fl_wait,
+					list_empty(&fl.fl_blocked_member));
 		if (!error) {
 			/*
 			 * If we've been sleeping someone might have
@@ -1652,7 +1677,8 @@ restart:
 
 	locks_dispose_list(&dispose);
 	error = wait_event_interruptible_timeout(new_fl->fl_wait,
-						!new_fl->fl_blocker, break_time);
+					list_empty(&new_fl->fl_blocked_member),
+					break_time);
 
 	percpu_down_read(&file_rwsem);
 	spin_lock(&ctx->flc_lock);
@@ -2136,7 +2162,8 @@ static int flock_lock_inode_wait(struct inode *inode, struct file_lock *fl)
 		error = flock_lock_inode(inode, fl);
 		if (error != FILE_LOCK_DEFERRED)
 			break;
-		error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker);
+		error = wait_event_interruptible(fl->fl_wait,
+				list_empty(&fl->fl_blocked_member));
 		if (error)
 			break;
 	}
@@ -2413,7 +2440,8 @@ static int do_lock_file_wait(struct file *filp, unsigned int cmd,
 		error = vfs_lock_file(filp, cmd, fl, NULL);
 		if (error != FILE_LOCK_DEFERRED)
 			break;
-		error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker);
+		error = wait_event_interruptible(fl->fl_wait,
+					list_empty(&fl->fl_blocked_member));
 		if (error)
 			break;
 	}
@@ -2853,7 +2881,7 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl,
 	}
 	if (inode) {
 		/* userspace relies on this representation of dev_t */
-		seq_printf(f, "%d %02x:%02x:%ld ", fl_pid,
+		seq_printf(f, "%d %02x:%02x:%lu ", fl_pid,
 				MAJOR(inode->i_sb->s_dev),
 				MINOR(inode->i_sb->s_dev), inode->i_ino);
 	} else {
diff --git a/fs/mount.h b/fs/mount.h
index 711a4093e475..c7abb7b394d8 100644
--- a/fs/mount.h
+++ b/fs/mount.h
@@ -9,7 +9,13 @@ struct mnt_namespace {
 	atomic_t		count;
 	struct ns_common	ns;
 	struct mount *	root;
+	/*
+	 * Traversal and modification of .list is protected by either
+	 * - taking namespace_sem for write, OR
+	 * - taking namespace_sem for read AND taking .ns_lock.
+	 */
 	struct list_head	list;
+	spinlock_t		ns_lock;
 	struct user_namespace	*user_ns;
 	struct ucounts		*ucounts;
 	u64			seq;	/* Sequence number to prevent loops */
@@ -133,9 +139,7 @@ struct proc_mounts {
 	struct mnt_namespace *ns;
 	struct path root;
 	int (*show)(struct seq_file *, struct vfsmount *);
-	void *cached_mount;
-	u64 cached_event;
-	loff_t cached_index;
+	struct mount cursor;
 };
 
 extern const struct seq_operations mounts_op;
@@ -153,3 +157,5 @@ static inline bool is_anon_ns(struct mnt_namespace *ns)
 {
 	return ns->seq == 0;
 }
+
+extern void mnt_cursor_del(struct mnt_namespace *ns, struct mount *cursor);
diff --git a/fs/mpage.c b/fs/mpage.c
index a63620cdb73a..830e6cc2a9e7 100644
--- a/fs/mpage.c
+++ b/fs/mpage.c
@@ -62,7 +62,7 @@ static struct bio *mpage_bio_submit(int op, int op_flags, struct bio *bio)
 {
 	bio->bi_end_io = mpage_end_io;
 	bio_set_op_attrs(bio, op, op_flags);
-	guard_bio_eod(op, bio);
+	guard_bio_eod(bio);
 	submit_bio(bio);
 	return NULL;
 }
@@ -91,7 +91,7 @@ mpage_alloc(struct block_device *bdev,
 }
 
 /*
- * support function for mpage_readpages.  The fs supplied get_block might
+ * support function for mpage_readahead.  The fs supplied get_block might
  * return an up to date buffer.  This is used to map that buffer into
  * the page, which allows readpage to avoid triggering a duplicate call
  * to get_block.
@@ -338,13 +338,8 @@ confused:
 }
 
 /**
- * mpage_readpages - populate an address space with some pages & start reads against them
- * @mapping: the address_space
- * @pages: The address of a list_head which contains the target pages.  These
- *   pages have their ->index populated and are otherwise uninitialised.
- *   The page at @pages->prev has the lowest file offset, and reads should be
- *   issued in @pages->prev to @pages->next order.
- * @nr_pages: The number of pages at *@pages
+ * mpage_readahead - start reads against pages
+ * @rac: Describes which pages to read.
  * @get_block: The filesystem's block mapper function.
  *
  * This function walks the pages and the blocks within each page, building and
@@ -381,36 +376,25 @@ confused:
  *
  * This all causes the disk requests to be issued in the correct order.
  */
-int
-mpage_readpages(struct address_space *mapping, struct list_head *pages,
-				unsigned nr_pages, get_block_t get_block)
+void mpage_readahead(struct readahead_control *rac, get_block_t get_block)
 {
+	struct page *page;
 	struct mpage_readpage_args args = {
 		.get_block = get_block,
 		.is_readahead = true,
 	};
-	unsigned page_idx;
-
-	for (page_idx = 0; page_idx < nr_pages; page_idx++) {
-		struct page *page = lru_to_page(pages);
 
+	while ((page = readahead_page(rac))) {
 		prefetchw(&page->flags);
-		list_del(&page->lru);
-		if (!add_to_page_cache_lru(page, mapping,
-					page->index,
-					readahead_gfp_mask(mapping))) {
-			args.page = page;
-			args.nr_pages = nr_pages - page_idx;
-			args.bio = do_mpage_readpage(&args);
-		}
+		args.page = page;
+		args.nr_pages = readahead_count(rac);
+		args.bio = do_mpage_readpage(&args);
 		put_page(page);
 	}
-	BUG_ON(!list_empty(pages));
 	if (args.bio)
 		mpage_bio_submit(REQ_OP_READ, REQ_RAHEAD, args.bio);
-	return 0;
 }
-EXPORT_SYMBOL(mpage_readpages);
+EXPORT_SYMBOL(mpage_readahead);
 
 /*
  * This isn't called much at all
@@ -563,7 +547,7 @@ static int __mpage_writepage(struct page *page, struct writeback_control *wbc,
 		 * Page has buffers, but they are all unmapped. The page was
 		 * created by pagein or read over a hole which was handled by
 		 * block_read_full_page().  If this address_space is also
-		 * using mpage_readpages then this can rarely happen.
+		 * using mpage_readahead then this can rarely happen.
 		 */
 		goto confused;
 	}
diff --git a/fs/namei.c b/fs/namei.c
index 671c3c1a3425..d81f73ff1a8b 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -491,7 +491,7 @@ struct nameidata {
 	struct path	root;
 	struct inode	*inode; /* path.dentry.d_inode */
 	unsigned int	flags;
-	unsigned	seq, m_seq;
+	unsigned	seq, m_seq, r_seq;
 	int		last_type;
 	unsigned	depth;
 	int		total_link_count;
@@ -503,9 +503,10 @@ struct nameidata {
 	} *stack, internal[EMBEDDED_LEVELS];
 	struct filename	*name;
 	struct nameidata *saved;
-	struct inode	*link_inode;
 	unsigned	root_seq;
 	int		dfd;
+	kuid_t		dir_uid;
+	umode_t		dir_mode;
 } __randomize_layout;
 
 static void set_nameidata(struct nameidata *p, int dfd, struct filename *name)
@@ -530,52 +531,34 @@ static void restore_nameidata(void)
 		kfree(now->stack);
 }
 
-static int __nd_alloc_stack(struct nameidata *nd)
+static bool nd_alloc_stack(struct nameidata *nd)
 {
 	struct saved *p;
 
-	if (nd->flags & LOOKUP_RCU) {
-		p= kmalloc_array(MAXSYMLINKS, sizeof(struct saved),
-				  GFP_ATOMIC);
-		if (unlikely(!p))
-			return -ECHILD;
-	} else {
-		p= kmalloc_array(MAXSYMLINKS, sizeof(struct saved),
-				  GFP_KERNEL);
-		if (unlikely(!p))
-			return -ENOMEM;
-	}
+	p= kmalloc_array(MAXSYMLINKS, sizeof(struct saved),
+			 nd->flags & LOOKUP_RCU ? GFP_ATOMIC : GFP_KERNEL);
+	if (unlikely(!p))
+		return false;
 	memcpy(p, nd->internal, sizeof(nd->internal));
 	nd->stack = p;
-	return 0;
+	return true;
 }
 
 /**
- * path_connected - Verify that a path->dentry is below path->mnt.mnt_root
- * @path: nameidate to verify
+ * path_connected - Verify that a dentry is below mnt.mnt_root
  *
  * Rename can sometimes move a file or directory outside of a bind
  * mount, path_connected allows those cases to be detected.
  */
-static bool path_connected(const struct path *path)
+static bool path_connected(struct vfsmount *mnt, struct dentry *dentry)
 {
-	struct vfsmount *mnt = path->mnt;
 	struct super_block *sb = mnt->mnt_sb;
 
 	/* Bind mounts and multi-root filesystems can have disconnected paths */
 	if (!(sb->s_iflags & SB_I_MULTIROOT) && (mnt->mnt_root == sb->s_root))
 		return true;
 
-	return is_subdir(path->dentry, mnt->mnt_root);
-}
-
-static inline int nd_alloc_stack(struct nameidata *nd)
-{
-	if (likely(nd->depth != EMBEDDED_LEVELS))
-		return 0;
-	if (likely(nd->stack != nd->internal))
-		return 0;
-	return __nd_alloc_stack(nd);
+	return is_subdir(dentry, mnt->mnt_root);
 }
 
 static void drop_links(struct nameidata *nd)
@@ -608,10 +591,9 @@ static void terminate_walk(struct nameidata *nd)
 }
 
 /* path_put is needed afterwards regardless of success or failure */
-static bool legitimize_path(struct nameidata *nd,
-			    struct path *path, unsigned seq)
+static bool __legitimize_path(struct path *path, unsigned seq, unsigned mseq)
 {
-	int res = __legitimize_mnt(path->mnt, nd->m_seq);
+	int res = __legitimize_mnt(path->mnt, mseq);
 	if (unlikely(res)) {
 		if (res > 0)
 			path->mnt = NULL;
@@ -625,6 +607,12 @@ static bool legitimize_path(struct nameidata *nd,
 	return !read_seqcount_retry(&path->dentry->d_seq, seq);
 }
 
+static inline bool legitimize_path(struct nameidata *nd,
+			    struct path *path, unsigned seq)
+{
+	return __legitimize_path(path, seq, nd->m_seq);
+}
+
 static bool legitimize_links(struct nameidata *nd)
 {
 	int i;
@@ -641,6 +629,14 @@ static bool legitimize_links(struct nameidata *nd)
 
 static bool legitimize_root(struct nameidata *nd)
 {
+	/*
+	 * For scoped-lookups (where nd->root has been zeroed), we need to
+	 * restart the whole lookup from scratch -- because set_root() is wrong
+	 * for these lookups (nd->dfd is the root, not the filesystem root).
+	 */
+	if (!nd->root.mnt && (nd->flags & LOOKUP_IS_SCOPED))
+		return false;
+	/* Nothing to do if nd->root is zero or is managed by the VFS user. */
 	if (!nd->root.mnt || (nd->flags & LOOKUP_ROOT))
 		return true;
 	nd->flags |= LOOKUP_ROOT_GRABBED;
@@ -776,12 +772,37 @@ static int complete_walk(struct nameidata *nd)
 	int status;
 
 	if (nd->flags & LOOKUP_RCU) {
-		if (!(nd->flags & LOOKUP_ROOT))
+		/*
+		 * We don't want to zero nd->root for scoped-lookups or
+		 * externally-managed nd->root.
+		 */
+		if (!(nd->flags & (LOOKUP_ROOT | LOOKUP_IS_SCOPED)))
 			nd->root.mnt = NULL;
 		if (unlikely(unlazy_walk(nd)))
 			return -ECHILD;
 	}
 
+	if (unlikely(nd->flags & LOOKUP_IS_SCOPED)) {
+		/*
+		 * While the guarantee of LOOKUP_IS_SCOPED is (roughly) "don't
+		 * ever step outside the root during lookup" and should already
+		 * be guaranteed by the rest of namei, we want to avoid a namei
+		 * BUG resulting in userspace being given a path that was not
+		 * scoped within the root at some point during the lookup.
+		 *
+		 * So, do a final sanity-check to make sure that in the
+		 * worst-case scenario (a complete bypass of LOOKUP_IS_SCOPED)
+		 * we won't silently return an fd completely outside of the
+		 * requested root to userspace.
+		 *
+		 * Userspace could move the path outside the root after this
+		 * check, but as discussed elsewhere this is not a concern (the
+		 * resolved file was inside the root at some point).
+		 */
+		if (!path_is_under(&nd->path, &nd->root))
+			return -EXDEV;
+	}
+
 	if (likely(!(nd->flags & LOOKUP_JUMPED)))
 		return 0;
 
@@ -798,10 +819,18 @@ static int complete_walk(struct nameidata *nd)
 	return status;
 }
 
-static void set_root(struct nameidata *nd)
+static int set_root(struct nameidata *nd)
 {
 	struct fs_struct *fs = current->fs;
 
+	/*
+	 * Jumping to the real root in a scoped-lookup is a BUG in namei, but we
+	 * still have to ensure it doesn't happen because it will cause a breakout
+	 * from the dirfd.
+	 */
+	if (WARN_ON(nd->flags & LOOKUP_IS_SCOPED))
+		return -ENOTRECOVERABLE;
+
 	if (nd->flags & LOOKUP_RCU) {
 		unsigned seq;
 
@@ -814,29 +843,23 @@ static void set_root(struct nameidata *nd)
 		get_fs_root(fs, &nd->root);
 		nd->flags |= LOOKUP_ROOT_GRABBED;
 	}
-}
-
-static void path_put_conditional(struct path *path, struct nameidata *nd)
-{
-	dput(path->dentry);
-	if (path->mnt != nd->path.mnt)
-		mntput(path->mnt);
-}
-
-static inline void path_to_nameidata(const struct path *path,
-					struct nameidata *nd)
-{
-	if (!(nd->flags & LOOKUP_RCU)) {
-		dput(nd->path.dentry);
-		if (nd->path.mnt != path->mnt)
-			mntput(nd->path.mnt);
-	}
-	nd->path.mnt = path->mnt;
-	nd->path.dentry = path->dentry;
+	return 0;
 }
 
 static int nd_jump_root(struct nameidata *nd)
 {
+	if (unlikely(nd->flags & LOOKUP_BENEATH))
+		return -EXDEV;
+	if (unlikely(nd->flags & LOOKUP_NO_XDEV)) {
+		/* Absolute path arguments to path_init() are allowed. */
+		if (nd->path.mnt != NULL && nd->path.mnt != nd->root.mnt)
+			return -EXDEV;
+	}
+	if (!nd->root.mnt) {
+		int error = set_root(nd);
+		if (error)
+			return error;
+	}
 	if (nd->flags & LOOKUP_RCU) {
 		struct dentry *d;
 		nd->path = nd->root;
@@ -859,14 +882,32 @@ static int nd_jump_root(struct nameidata *nd)
  * Helper to directly jump to a known parsed path from ->get_link,
  * caller must have taken a reference to path beforehand.
  */
-void nd_jump_link(struct path *path)
+int nd_jump_link(struct path *path)
 {
+	int error = -ELOOP;
 	struct nameidata *nd = current->nameidata;
-	path_put(&nd->path);
 
+	if (unlikely(nd->flags & LOOKUP_NO_MAGICLINKS))
+		goto err;
+
+	error = -EXDEV;
+	if (unlikely(nd->flags & LOOKUP_NO_XDEV)) {
+		if (nd->path.mnt != path->mnt)
+			goto err;
+	}
+	/* Not currently safe for scoped-lookups. */
+	if (unlikely(nd->flags & LOOKUP_IS_SCOPED))
+		goto err;
+
+	path_put(&nd->path);
 	nd->path = *path;
 	nd->inode = nd->path.dentry->d_inode;
 	nd->flags |= LOOKUP_JUMPED;
+	return 0;
+
+err:
+	path_put(path);
+	return error;
 }
 
 static inline void put_link(struct nameidata *nd)
@@ -897,35 +938,28 @@ int sysctl_protected_regular __read_mostly;
  *
  * Returns 0 if following the symlink is allowed, -ve on error.
  */
-static inline int may_follow_link(struct nameidata *nd)
+static inline int may_follow_link(struct nameidata *nd, const struct inode *inode)
 {
-	const struct inode *inode;
-	const struct inode *parent;
-	kuid_t puid;
-
 	if (!sysctl_protected_symlinks)
 		return 0;
 
 	/* Allowed if owner and follower match. */
-	inode = nd->link_inode;
 	if (uid_eq(current_cred()->fsuid, inode->i_uid))
 		return 0;
 
 	/* Allowed if parent directory not sticky and world-writable. */
-	parent = nd->inode;
-	if ((parent->i_mode & (S_ISVTX|S_IWOTH)) != (S_ISVTX|S_IWOTH))
+	if ((nd->dir_mode & (S_ISVTX|S_IWOTH)) != (S_ISVTX|S_IWOTH))
 		return 0;
 
 	/* Allowed if parent directory and link owner match. */
-	puid = parent->i_uid;
-	if (uid_valid(puid) && uid_eq(puid, inode->i_uid))
+	if (uid_valid(nd->dir_uid) && uid_eq(nd->dir_uid, inode->i_uid))
 		return 0;
 
 	if (nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 
 	audit_inode(nd->name, nd->stack[0].link.dentry, 0);
-	audit_log_link_denied("follow_link");
+	audit_log_path_denied(AUDIT_ANOM_LINK, "follow_link");
 	return -EACCES;
 }
 
@@ -993,7 +1027,7 @@ static int may_linkat(struct path *link)
 	if (safe_hardlink_source(inode) || inode_owner_or_capable(inode))
 		return 0;
 
-	audit_log_link_denied("linkat");
+	audit_log_path_denied(AUDIT_ANOM_LINK, "linkat");
 	return -EPERM;
 }
 
@@ -1001,7 +1035,8 @@ static int may_linkat(struct path *link)
  * may_create_in_sticky - Check whether an O_CREAT open in a sticky directory
  *			  should be allowed, or not, on files that already
  *			  exist.
- * @dir: the sticky parent directory
+ * @dir_mode: mode bits of directory
+ * @dir_uid: owner of directory
  * @inode: the inode of the file to open
  *
  * Block an O_CREAT open of a FIFO (or a regular file) when:
@@ -1017,80 +1052,29 @@ static int may_linkat(struct path *link)
  *
  * Returns 0 if the open is allowed, -ve on error.
  */
-static int may_create_in_sticky(struct dentry * const dir,
+static int may_create_in_sticky(umode_t dir_mode, kuid_t dir_uid,
 				struct inode * const inode)
 {
 	if ((!sysctl_protected_fifos && S_ISFIFO(inode->i_mode)) ||
 	    (!sysctl_protected_regular && S_ISREG(inode->i_mode)) ||
-	    likely(!(dir->d_inode->i_mode & S_ISVTX)) ||
-	    uid_eq(inode->i_uid, dir->d_inode->i_uid) ||
+	    likely(!(dir_mode & S_ISVTX)) ||
+	    uid_eq(inode->i_uid, dir_uid) ||
 	    uid_eq(current_fsuid(), inode->i_uid))
 		return 0;
 
-	if (likely(dir->d_inode->i_mode & 0002) ||
-	    (dir->d_inode->i_mode & 0020 &&
+	if (likely(dir_mode & 0002) ||
+	    (dir_mode & 0020 &&
 	     ((sysctl_protected_fifos >= 2 && S_ISFIFO(inode->i_mode)) ||
 	      (sysctl_protected_regular >= 2 && S_ISREG(inode->i_mode))))) {
+		const char *operation = S_ISFIFO(inode->i_mode) ?
+					"sticky_create_fifo" :
+					"sticky_create_regular";
+		audit_log_path_denied(AUDIT_ANOM_CREAT, operation);
 		return -EACCES;
 	}
 	return 0;
 }
 
-static __always_inline
-const char *get_link(struct nameidata *nd)
-{
-	struct saved *last = nd->stack + nd->depth - 1;
-	struct dentry *dentry = last->link.dentry;
-	struct inode *inode = nd->link_inode;
-	int error;
-	const char *res;
-
-	if (!(nd->flags & LOOKUP_RCU)) {
-		touch_atime(&last->link);
-		cond_resched();
-	} else if (atime_needs_update(&last->link, inode)) {
-		if (unlikely(unlazy_walk(nd)))
-			return ERR_PTR(-ECHILD);
-		touch_atime(&last->link);
-	}
-
-	error = security_inode_follow_link(dentry, inode,
-					   nd->flags & LOOKUP_RCU);
-	if (unlikely(error))
-		return ERR_PTR(error);
-
-	nd->last_type = LAST_BIND;
-	res = READ_ONCE(inode->i_link);
-	if (!res) {
-		const char * (*get)(struct dentry *, struct inode *,
-				struct delayed_call *);
-		get = inode->i_op->get_link;
-		if (nd->flags & LOOKUP_RCU) {
-			res = get(NULL, inode, &last->done);
-			if (res == ERR_PTR(-ECHILD)) {
-				if (unlikely(unlazy_walk(nd)))
-					return ERR_PTR(-ECHILD);
-				res = get(dentry, inode, &last->done);
-			}
-		} else {
-			res = get(dentry, inode, &last->done);
-		}
-		if (IS_ERR_OR_NULL(res))
-			return res;
-	}
-	if (*res == '/') {
-		if (!nd->root.mnt)
-			set_root(nd);
-		if (unlikely(nd_jump_root(nd)))
-			return ERR_PTR(-ECHILD);
-		while (unlikely(*++res == '/'))
-			;
-	}
-	if (!*res)
-		res = NULL;
-	return res;
-}
-
 /*
  * follow_up - Find the mountpoint of path's vfsmount
  *
@@ -1124,19 +1108,59 @@ int follow_up(struct path *path)
 }
 EXPORT_SYMBOL(follow_up);
 
+static bool choose_mountpoint_rcu(struct mount *m, const struct path *root,
+				  struct path *path, unsigned *seqp)
+{
+	while (mnt_has_parent(m)) {
+		struct dentry *mountpoint = m->mnt_mountpoint;
+
+		m = m->mnt_parent;
+		if (unlikely(root->dentry == mountpoint &&
+			     root->mnt == &m->mnt))
+			break;
+		if (mountpoint != m->mnt.mnt_root) {
+			path->mnt = &m->mnt;
+			path->dentry = mountpoint;
+			*seqp = read_seqcount_begin(&mountpoint->d_seq);
+			return true;
+		}
+	}
+	return false;
+}
+
+static bool choose_mountpoint(struct mount *m, const struct path *root,
+			      struct path *path)
+{
+	bool found;
+
+	rcu_read_lock();
+	while (1) {
+		unsigned seq, mseq = read_seqbegin(&mount_lock);
+
+		found = choose_mountpoint_rcu(m, root, path, &seq);
+		if (unlikely(!found)) {
+			if (!read_seqretry(&mount_lock, mseq))
+				break;
+		} else {
+			if (likely(__legitimize_path(path, seq, mseq)))
+				break;
+			rcu_read_unlock();
+			path_put(path);
+			rcu_read_lock();
+		}
+	}
+	rcu_read_unlock();
+	return found;
+}
+
 /*
  * Perform an automount
  * - return -EISDIR to tell follow_managed() to stop and return the path we
  *   were called with.
  */
-static int follow_automount(struct path *path, struct nameidata *nd,
-			    bool *need_mntput)
+static int follow_automount(struct path *path, int *count, unsigned lookup_flags)
 {
-	struct vfsmount *mnt;
-	int err;
-
-	if (!path->dentry->d_op || !path->dentry->d_op->d_automount)
-		return -EREMOTE;
+	struct dentry *dentry = path->dentry;
 
 	/* We don't want to mount if someone's just doing a stat -
 	 * unless they're stat'ing a directory and appended a '/' to
@@ -1149,131 +1173,91 @@ static int follow_automount(struct path *path, struct nameidata *nd,
 	 * as being automount points.  These will need the attentions
 	 * of the daemon to instantiate them before they can be used.
 	 */
-	if (!(nd->flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY |
+	if (!(lookup_flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY |
 			   LOOKUP_OPEN | LOOKUP_CREATE | LOOKUP_AUTOMOUNT)) &&
-	    path->dentry->d_inode)
+	    dentry->d_inode)
 		return -EISDIR;
 
-	nd->total_link_count++;
-	if (nd->total_link_count >= 40)
+	if (count && (*count)++ >= MAXSYMLINKS)
 		return -ELOOP;
 
-	mnt = path->dentry->d_op->d_automount(path);
-	if (IS_ERR(mnt)) {
-		/*
-		 * The filesystem is allowed to return -EISDIR here to indicate
-		 * it doesn't want to automount.  For instance, autofs would do
-		 * this so that its userspace daemon can mount on this dentry.
-		 *
-		 * However, we can only permit this if it's a terminal point in
-		 * the path being looked up; if it wasn't then the remainder of
-		 * the path is inaccessible and we should say so.
-		 */
-		if (PTR_ERR(mnt) == -EISDIR && (nd->flags & LOOKUP_PARENT))
-			return -EREMOTE;
-		return PTR_ERR(mnt);
-	}
-
-	if (!mnt) /* mount collision */
-		return 0;
-
-	if (!*need_mntput) {
-		/* lock_mount() may release path->mnt on error */
-		mntget(path->mnt);
-		*need_mntput = true;
-	}
-	err = finish_automount(mnt, path);
-
-	switch (err) {
-	case -EBUSY:
-		/* Someone else made a mount here whilst we were busy */
-		return 0;
-	case 0:
-		path_put(path);
-		path->mnt = mnt;
-		path->dentry = dget(mnt->mnt_root);
-		return 0;
-	default:
-		return err;
-	}
-
+	return finish_automount(dentry->d_op->d_automount(path), path);
 }
 
 /*
- * Handle a dentry that is managed in some way.
- * - Flagged for transit management (autofs)
- * - Flagged as mountpoint
- * - Flagged as automount point
- *
- * This may only be called in refwalk mode.
- *
- * Serialization is taken care of in namespace.c
+ * mount traversal - out-of-line part.  One note on ->d_flags accesses -
+ * dentries are pinned but not locked here, so negative dentry can go
+ * positive right under us.  Use of smp_load_acquire() provides a barrier
+ * sufficient for ->d_inode and ->d_flags consistency.
  */
-static int follow_managed(struct path *path, struct nameidata *nd)
+static int __traverse_mounts(struct path *path, unsigned flags, bool *jumped,
+			     int *count, unsigned lookup_flags)
 {
-	struct vfsmount *mnt = path->mnt; /* held by caller, must be left alone */
-	unsigned managed;
+	struct vfsmount *mnt = path->mnt;
 	bool need_mntput = false;
 	int ret = 0;
 
-	/* Given that we're not holding a lock here, we retain the value in a
-	 * local variable for each dentry as we look at it so that we don't see
-	 * the components of that value change under us */
-	while (managed = READ_ONCE(path->dentry->d_flags),
-	       managed &= DCACHE_MANAGED_DENTRY,
-	       unlikely(managed != 0)) {
+	while (flags & DCACHE_MANAGED_DENTRY) {
 		/* Allow the filesystem to manage the transit without i_mutex
 		 * being held. */
-		if (managed & DCACHE_MANAGE_TRANSIT) {
-			BUG_ON(!path->dentry->d_op);
-			BUG_ON(!path->dentry->d_op->d_manage);
+		if (flags & DCACHE_MANAGE_TRANSIT) {
 			ret = path->dentry->d_op->d_manage(path, false);
+			flags = smp_load_acquire(&path->dentry->d_flags);
 			if (ret < 0)
 				break;
 		}
 
-		/* Transit to a mounted filesystem. */
-		if (managed & DCACHE_MOUNTED) {
+		if (flags & DCACHE_MOUNTED) {	// something's mounted on it..
 			struct vfsmount *mounted = lookup_mnt(path);
-			if (mounted) {
+			if (mounted) {		// ... in our namespace
 				dput(path->dentry);
 				if (need_mntput)
 					mntput(path->mnt);
 				path->mnt = mounted;
 				path->dentry = dget(mounted->mnt_root);
+				// here we know it's positive
+				flags = path->dentry->d_flags;
 				need_mntput = true;
 				continue;
 			}
-
-			/* Something is mounted on this dentry in another
-			 * namespace and/or whatever was mounted there in this
-			 * namespace got unmounted before lookup_mnt() could
-			 * get it */
 		}
 
-		/* Handle an automount point */
-		if (managed & DCACHE_NEED_AUTOMOUNT) {
-			ret = follow_automount(path, nd, &need_mntput);
-			if (ret < 0)
-				break;
-			continue;
-		}
+		if (!(flags & DCACHE_NEED_AUTOMOUNT))
+			break;
 
-		/* We didn't change the current path point */
-		break;
+		// uncovered automount point
+		ret = follow_automount(path, count, lookup_flags);
+		flags = smp_load_acquire(&path->dentry->d_flags);
+		if (ret < 0)
+			break;
 	}
 
+	if (ret == -EISDIR)
+		ret = 0;
+	// possible if you race with several mount --move
 	if (need_mntput && path->mnt == mnt)
 		mntput(path->mnt);
-	if (ret == -EISDIR || !ret)
-		ret = 1;
-	if (need_mntput)
-		nd->flags |= LOOKUP_JUMPED;
-	if (unlikely(ret < 0))
-		path_put_conditional(path, nd);
+	if (!ret && unlikely(d_flags_negative(flags)))
+		ret = -ENOENT;
+	*jumped = need_mntput;
 	return ret;
 }
 
+static inline int traverse_mounts(struct path *path, bool *jumped,
+				  int *count, unsigned lookup_flags)
+{
+	unsigned flags = smp_load_acquire(&path->dentry->d_flags);
+
+	/* fastpath */
+	if (likely(!(flags & DCACHE_MANAGED_DENTRY))) {
+		*jumped = false;
+		if (unlikely(d_flags_negative(flags)))
+			return -ENOENT;
+		return 0;
+	}
+	return __traverse_mounts(path, flags, jumped, count, lookup_flags);
+}
+
 int follow_down_one(struct path *path)
 {
 	struct vfsmount *mounted;
@@ -1290,11 +1274,22 @@ int follow_down_one(struct path *path)
 }
 EXPORT_SYMBOL(follow_down_one);
 
-static inline int managed_dentry_rcu(const struct path *path)
+/*
+ * Follow down to the covering mount currently visible to userspace.  At each
+ * point, the filesystem owning that dentry may be queried as to whether the
+ * caller is permitted to proceed or not.
+ */
+int follow_down(struct path *path)
 {
-	return (path->dentry->d_flags & DCACHE_MANAGE_TRANSIT) ?
-		path->dentry->d_op->d_manage(path, true) : 0;
+	struct vfsmount *mnt = path->mnt;
+	bool jumped;
+	int ret = traverse_mounts(path, &jumped, NULL, 0);
+
+	if (path->mnt != mnt)
+		mntput(mnt);
+	return ret;
 }
+EXPORT_SYMBOL(follow_down);
 
 /*
  * Try to skip to top of mountpoint pile in rcuwalk mode.  Fail if
@@ -1303,190 +1298,88 @@ static inline int managed_dentry_rcu(const struct path *path)
 static bool __follow_mount_rcu(struct nameidata *nd, struct path *path,
 			       struct inode **inode, unsigned *seqp)
 {
+	struct dentry *dentry = path->dentry;
+	unsigned int flags = dentry->d_flags;
+
+	if (likely(!(flags & DCACHE_MANAGED_DENTRY)))
+		return true;
+
+	if (unlikely(nd->flags & LOOKUP_NO_XDEV))
+		return false;
+
 	for (;;) {
-		struct mount *mounted;
 		/*
 		 * Don't forget we might have a non-mountpoint managed dentry
 		 * that wants to block transit.
 		 */
-		switch (managed_dentry_rcu(path)) {
-		case -ECHILD:
-		default:
-			return false;
-		case -EISDIR:
-			return true;
-		case 0:
-			break;
+		if (unlikely(flags & DCACHE_MANAGE_TRANSIT)) {
+			int res = dentry->d_op->d_manage(path, true);
+			if (res)
+				return res == -EISDIR;
+			flags = dentry->d_flags;
 		}
 
-		if (!d_mountpoint(path->dentry))
-			return !(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT);
-
-		mounted = __lookup_mnt(path->mnt, path->dentry);
-		if (!mounted)
-			break;
-		path->mnt = &mounted->mnt;
-		path->dentry = mounted->mnt.mnt_root;
-		nd->flags |= LOOKUP_JUMPED;
-		*seqp = read_seqcount_begin(&path->dentry->d_seq);
-		/*
-		 * Update the inode too. We don't need to re-check the
-		 * dentry sequence number here after this d_inode read,
-		 * because a mount-point is always pinned.
-		 */
-		*inode = path->dentry->d_inode;
-	}
-	return !read_seqretry(&mount_lock, nd->m_seq) &&
-		!(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT);
-}
-
-static int follow_dotdot_rcu(struct nameidata *nd)
-{
-	struct inode *inode = nd->inode;
-
-	while (1) {
-		if (path_equal(&nd->path, &nd->root))
-			break;
-		if (nd->path.dentry != nd->path.mnt->mnt_root) {
-			struct dentry *old = nd->path.dentry;
-			struct dentry *parent = old->d_parent;
-			unsigned seq;
-
-			inode = parent->d_inode;
-			seq = read_seqcount_begin(&parent->d_seq);
-			if (unlikely(read_seqcount_retry(&old->d_seq, nd->seq)))
-				return -ECHILD;
-			nd->path.dentry = parent;
-			nd->seq = seq;
-			if (unlikely(!path_connected(&nd->path)))
-				return -ENOENT;
-			break;
-		} else {
-			struct mount *mnt = real_mount(nd->path.mnt);
-			struct mount *mparent = mnt->mnt_parent;
-			struct dentry *mountpoint = mnt->mnt_mountpoint;
-			struct inode *inode2 = mountpoint->d_inode;
-			unsigned seq = read_seqcount_begin(&mountpoint->d_seq);
-			if (unlikely(read_seqretry(&mount_lock, nd->m_seq)))
-				return -ECHILD;
-			if (&mparent->mnt == nd->path.mnt)
-				break;
-			/* we know that mountpoint was pinned */
-			nd->path.dentry = mountpoint;
-			nd->path.mnt = &mparent->mnt;
-			inode = inode2;
-			nd->seq = seq;
+		if (flags & DCACHE_MOUNTED) {
+			struct mount *mounted = __lookup_mnt(path->mnt, dentry);
+			if (mounted) {
+				path->mnt = &mounted->mnt;
+				dentry = path->dentry = mounted->mnt.mnt_root;
+				nd->flags |= LOOKUP_JUMPED;
+				*seqp = read_seqcount_begin(&dentry->d_seq);
+				*inode = dentry->d_inode;
+				/*
+				 * We don't need to re-check ->d_seq after this
+				 * ->d_inode read - there will be an RCU delay
+				 * between mount hash removal and ->mnt_root
+				 * becoming unpinned.
+				 */
+				flags = dentry->d_flags;
+				continue;
+			}
+			if (read_seqretry(&mount_lock, nd->m_seq))
+				return false;
 		}
+		return !(flags & DCACHE_NEED_AUTOMOUNT);
 	}
-	while (unlikely(d_mountpoint(nd->path.dentry))) {
-		struct mount *mounted;
-		mounted = __lookup_mnt(nd->path.mnt, nd->path.dentry);
-		if (unlikely(read_seqretry(&mount_lock, nd->m_seq)))
-			return -ECHILD;
-		if (!mounted)
-			break;
-		nd->path.mnt = &mounted->mnt;
-		nd->path.dentry = mounted->mnt.mnt_root;
-		inode = nd->path.dentry->d_inode;
-		nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq);
-	}
-	nd->inode = inode;
-	return 0;
 }
 
-/*
- * Follow down to the covering mount currently visible to userspace.  At each
- * point, the filesystem owning that dentry may be queried as to whether the
- * caller is permitted to proceed or not.
- */
-int follow_down(struct path *path)
+static inline int handle_mounts(struct nameidata *nd, struct dentry *dentry,
+			  struct path *path, struct inode **inode,
+			  unsigned int *seqp)
 {
-	unsigned managed;
+	bool jumped;
 	int ret;
 
-	while (managed = READ_ONCE(path->dentry->d_flags),
-	       unlikely(managed & DCACHE_MANAGED_DENTRY)) {
-		/* Allow the filesystem to manage the transit without i_mutex
-		 * being held.
-		 *
-		 * We indicate to the filesystem if someone is trying to mount
-		 * something here.  This gives autofs the chance to deny anyone
-		 * other than its daemon the right to mount on its
-		 * superstructure.
-		 *
-		 * The filesystem may sleep at this point.
-		 */
-		if (managed & DCACHE_MANAGE_TRANSIT) {
-			BUG_ON(!path->dentry->d_op);
-			BUG_ON(!path->dentry->d_op->d_manage);
-			ret = path->dentry->d_op->d_manage(path, false);
-			if (ret < 0)
-				return ret == -EISDIR ? 0 : ret;
-		}
-
-		/* Transit to a mounted filesystem. */
-		if (managed & DCACHE_MOUNTED) {
-			struct vfsmount *mounted = lookup_mnt(path);
-			if (!mounted)
-				break;
-			dput(path->dentry);
-			mntput(path->mnt);
-			path->mnt = mounted;
-			path->dentry = dget(mounted->mnt_root);
-			continue;
-		}
-
-		/* Don't handle automount points here */
-		break;
+	path->mnt = nd->path.mnt;
+	path->dentry = dentry;
+	if (nd->flags & LOOKUP_RCU) {
+		unsigned int seq = *seqp;
+		if (unlikely(!*inode))
+			return -ENOENT;
+		if (likely(__follow_mount_rcu(nd, path, inode, seqp)))
+			return 0;
+		if (unlazy_child(nd, dentry, seq))
+			return -ECHILD;
+		// *path might've been clobbered by __follow_mount_rcu()
+		path->mnt = nd->path.mnt;
+		path->dentry = dentry;
+	}
+	ret = traverse_mounts(path, &jumped, &nd->total_link_count, nd->flags);
+	if (jumped) {
+		if (unlikely(nd->flags & LOOKUP_NO_XDEV))
+			ret = -EXDEV;
+		else
+			nd->flags |= LOOKUP_JUMPED;
 	}
-	return 0;
-}
-EXPORT_SYMBOL(follow_down);
-
-/*
- * Skip to top of mountpoint pile in refwalk mode for follow_dotdot()
- */
-static void follow_mount(struct path *path)
-{
-	while (d_mountpoint(path->dentry)) {
-		struct vfsmount *mounted = lookup_mnt(path);
-		if (!mounted)
-			break;
+	if (unlikely(ret)) {
 		dput(path->dentry);
-		mntput(path->mnt);
-		path->mnt = mounted;
-		path->dentry = dget(mounted->mnt_root);
-	}
-}
-
-static int path_parent_directory(struct path *path)
-{
-	struct dentry *old = path->dentry;
-	/* rare case of legitimate dget_parent()... */
-	path->dentry = dget_parent(path->dentry);
-	dput(old);
-	if (unlikely(!path_connected(path)))
-		return -ENOENT;
-	return 0;
-}
-
-static int follow_dotdot(struct nameidata *nd)
-{
-	while(1) {
-		if (path_equal(&nd->path, &nd->root))
-			break;
-		if (nd->path.dentry != nd->path.mnt->mnt_root) {
-			int ret = path_parent_directory(&nd->path);
-			if (ret)
-				return ret;
-			break;
-		}
-		if (!follow_up(&nd->path))
-			break;
+		if (path->mnt != nd->path.mnt)
+			mntput(path->mnt);
+	} else {
+		*inode = d_backing_inode(path->dentry);
+		*seqp = 0; /* out of RCU mode, so the value doesn't matter */
 	}
-	follow_mount(&nd->path);
-	nd->inode = nd->path.dentry->d_inode;
-	return 0;
+	return ret;
 }
 
 /*
@@ -1543,14 +1436,12 @@ static struct dentry *__lookup_hash(const struct qstr *name,
 	return dentry;
 }
 
-static int lookup_fast(struct nameidata *nd,
-		       struct path *path, struct inode **inode,
-		       unsigned *seqp)
+static struct dentry *lookup_fast(struct nameidata *nd,
+				  struct inode **inode,
+			          unsigned *seqp)
 {
-	struct vfsmount *mnt = nd->path.mnt;
 	struct dentry *dentry, *parent = nd->path.dentry;
 	int status = 1;
-	int err;
 
 	/*
 	 * Rename seqlock is not required here because in the off chance
@@ -1559,12 +1450,11 @@ static int lookup_fast(struct nameidata *nd,
 	 */
 	if (nd->flags & LOOKUP_RCU) {
 		unsigned seq;
-		bool negative;
 		dentry = __d_lookup_rcu(parent, &nd->last, &seq);
 		if (unlikely(!dentry)) {
 			if (unlazy_walk(nd))
-				return -ECHILD;
-			return 0;
+				return ERR_PTR(-ECHILD);
+			return NULL;
 		}
 
 		/*
@@ -1572,9 +1462,8 @@ static int lookup_fast(struct nameidata *nd,
 		 * the dentry name information from lookup.
 		 */
 		*inode = d_backing_inode(dentry);
-		negative = d_is_negative(dentry);
 		if (unlikely(read_seqcount_retry(&dentry->d_seq, seq)))
-			return -ECHILD;
+			return ERR_PTR(-ECHILD);
 
 		/*
 		 * This sequence count validates that the parent had no
@@ -1584,50 +1473,30 @@ static int lookup_fast(struct nameidata *nd,
 		 *  enough, we can use __read_seqcount_retry here.
 		 */
 		if (unlikely(__read_seqcount_retry(&parent->d_seq, nd->seq)))
-			return -ECHILD;
+			return ERR_PTR(-ECHILD);
 
 		*seqp = seq;
 		status = d_revalidate(dentry, nd->flags);
-		if (likely(status > 0)) {
-			/*
-			 * Note: do negative dentry check after revalidation in
-			 * case that drops it.
-			 */
-			if (unlikely(negative))
-				return -ENOENT;
-			path->mnt = mnt;
-			path->dentry = dentry;
-			if (likely(__follow_mount_rcu(nd, path, inode, seqp)))
-				return 1;
-		}
+		if (likely(status > 0))
+			return dentry;
 		if (unlazy_child(nd, dentry, seq))
-			return -ECHILD;
+			return ERR_PTR(-ECHILD);
 		if (unlikely(status == -ECHILD))
 			/* we'd been told to redo it in non-rcu mode */
 			status = d_revalidate(dentry, nd->flags);
 	} else {
 		dentry = __d_lookup(parent, &nd->last);
 		if (unlikely(!dentry))
-			return 0;
+			return NULL;
 		status = d_revalidate(dentry, nd->flags);
 	}
 	if (unlikely(status <= 0)) {
 		if (!status)
 			d_invalidate(dentry);
 		dput(dentry);
-		return status;
-	}
-	if (unlikely(d_is_negative(dentry))) {
-		dput(dentry);
-		return -ENOENT;
+		return ERR_PTR(status);
 	}
-
-	path->mnt = mnt;
-	path->dentry = dentry;
-	err = follow_managed(path, nd);
-	if (likely(err > 0))
-		*inode = d_backing_inode(path->dentry);
-	return err;
+	return dentry;
 }
 
 /* Fast lookup failed, do it the slow way */
@@ -1647,17 +1516,15 @@ again:
 	if (IS_ERR(dentry))
 		return dentry;
 	if (unlikely(!d_in_lookup(dentry))) {
-		if (!(flags & LOOKUP_NO_REVAL)) {
-			int error = d_revalidate(dentry, flags);
-			if (unlikely(error <= 0)) {
-				if (!error) {
-					d_invalidate(dentry);
-					dput(dentry);
-					goto again;
-				}
+		int error = d_revalidate(dentry, flags);
+		if (unlikely(error <= 0)) {
+			if (!error) {
+				d_invalidate(dentry);
 				dput(dentry);
-				dentry = ERR_PTR(error);
+				goto again;
 			}
+			dput(dentry);
+			dentry = ERR_PTR(error);
 		}
 	} else {
 		old = inode->i_op->lookup(inode, dentry, flags);
@@ -1694,60 +1561,107 @@ static inline int may_lookup(struct nameidata *nd)
 	return inode_permission(nd->inode, MAY_EXEC);
 }
 
-static inline int handle_dots(struct nameidata *nd, int type)
+static int reserve_stack(struct nameidata *nd, struct path *link, unsigned seq)
 {
-	if (type == LAST_DOTDOT) {
-		if (!nd->root.mnt)
-			set_root(nd);
-		if (nd->flags & LOOKUP_RCU) {
-			return follow_dotdot_rcu(nd);
-		} else
-			return follow_dotdot(nd);
+	if (unlikely(nd->total_link_count++ >= MAXSYMLINKS))
+		return -ELOOP;
+
+	if (likely(nd->depth != EMBEDDED_LEVELS))
+		return 0;
+	if (likely(nd->stack != nd->internal))
+		return 0;
+	if (likely(nd_alloc_stack(nd)))
+		return 0;
+
+	if (nd->flags & LOOKUP_RCU) {
+		// we need to grab link before we do unlazy.  And we can't skip
+		// unlazy even if we fail to grab the link - cleanup needs it
+		bool grabbed_link = legitimize_path(nd, link, seq);
+
+		if (unlazy_walk(nd) != 0 || !grabbed_link)
+			return -ECHILD;
+
+		if (nd_alloc_stack(nd))
+			return 0;
 	}
-	return 0;
+	return -ENOMEM;
 }
 
-static int pick_link(struct nameidata *nd, struct path *link,
-		     struct inode *inode, unsigned seq)
+enum {WALK_TRAILING = 1, WALK_MORE = 2, WALK_NOFOLLOW = 4};
+
+static const char *pick_link(struct nameidata *nd, struct path *link,
+		     struct inode *inode, unsigned seq, int flags)
 {
-	int error;
 	struct saved *last;
-	if (unlikely(nd->total_link_count++ >= MAXSYMLINKS)) {
-		path_to_nameidata(link, nd);
-		return -ELOOP;
-	}
-	if (!(nd->flags & LOOKUP_RCU)) {
-		if (link->mnt == nd->path.mnt)
-			mntget(link->mnt);
-	}
-	error = nd_alloc_stack(nd);
+	const char *res;
+	int error = reserve_stack(nd, link, seq);
+
 	if (unlikely(error)) {
-		if (error == -ECHILD) {
-			if (unlikely(!legitimize_path(nd, link, seq))) {
-				drop_links(nd);
-				nd->depth = 0;
-				nd->flags &= ~LOOKUP_RCU;
-				nd->path.mnt = NULL;
-				nd->path.dentry = NULL;
-				rcu_read_unlock();
-			} else if (likely(unlazy_walk(nd)) == 0)
-				error = nd_alloc_stack(nd);
-		}
-		if (error) {
+		if (!(nd->flags & LOOKUP_RCU))
 			path_put(link);
-			return error;
-		}
+		return ERR_PTR(error);
 	}
-
 	last = nd->stack + nd->depth++;
 	last->link = *link;
 	clear_delayed_call(&last->done);
-	nd->link_inode = inode;
 	last->seq = seq;
-	return 1;
-}
 
-enum {WALK_FOLLOW = 1, WALK_MORE = 2};
+	if (flags & WALK_TRAILING) {
+		error = may_follow_link(nd, inode);
+		if (unlikely(error))
+			return ERR_PTR(error);
+	}
+
+	if (unlikely(nd->flags & LOOKUP_NO_SYMLINKS))
+		return ERR_PTR(-ELOOP);
+
+	if (!(nd->flags & LOOKUP_RCU)) {
+		touch_atime(&last->link);
+		cond_resched();
+	} else if (atime_needs_update(&last->link, inode)) {
+		if (unlikely(unlazy_walk(nd)))
+			return ERR_PTR(-ECHILD);
+		touch_atime(&last->link);
+	}
+
+	error = security_inode_follow_link(link->dentry, inode,
+					   nd->flags & LOOKUP_RCU);
+	if (unlikely(error))
+		return ERR_PTR(error);
+
+	res = READ_ONCE(inode->i_link);
+	if (!res) {
+		const char * (*get)(struct dentry *, struct inode *,
+				struct delayed_call *);
+		get = inode->i_op->get_link;
+		if (nd->flags & LOOKUP_RCU) {
+			res = get(NULL, inode, &last->done);
+			if (res == ERR_PTR(-ECHILD)) {
+				if (unlikely(unlazy_walk(nd)))
+					return ERR_PTR(-ECHILD);
+				res = get(link->dentry, inode, &last->done);
+			}
+		} else {
+			res = get(link->dentry, inode, &last->done);
+		}
+		if (!res)
+			goto all_done;
+		if (IS_ERR(res))
+			return res;
+	}
+	if (*res == '/') {
+		error = nd_jump_root(nd);
+		if (unlikely(error))
+			return ERR_PTR(error);
+		while (unlikely(*++res == '/'))
+			;
+	}
+	if (*res)
+		return res;
+all_done: // pure jump
+	put_link(nd);
+	return NULL;
+}
 
 /*
  * Do we need to follow links? We _really_ want to be able
@@ -1755,68 +1669,187 @@ enum {WALK_FOLLOW = 1, WALK_MORE = 2};
  * so we keep a cache of "no, this doesn't need follow_link"
  * for the common case.
  */
-static inline int step_into(struct nameidata *nd, struct path *path,
-			    int flags, struct inode *inode, unsigned seq)
+static const char *step_into(struct nameidata *nd, int flags,
+		     struct dentry *dentry, struct inode *inode, unsigned seq)
 {
-	if (!(flags & WALK_MORE) && nd->depth)
-		put_link(nd);
-	if (likely(!d_is_symlink(path->dentry)) ||
-	   !(flags & WALK_FOLLOW || nd->flags & LOOKUP_FOLLOW)) {
+	struct path path;
+	int err = handle_mounts(nd, dentry, &path, &inode, &seq);
+
+	if (err < 0)
+		return ERR_PTR(err);
+	if (likely(!d_is_symlink(path.dentry)) ||
+	   ((flags & WALK_TRAILING) && !(nd->flags & LOOKUP_FOLLOW)) ||
+	   (flags & WALK_NOFOLLOW)) {
 		/* not a symlink or should not follow */
-		path_to_nameidata(path, nd);
+		if (!(nd->flags & LOOKUP_RCU)) {
+			dput(nd->path.dentry);
+			if (nd->path.mnt != path.mnt)
+				mntput(nd->path.mnt);
+		}
+		nd->path = path;
 		nd->inode = inode;
 		nd->seq = seq;
-		return 0;
+		return NULL;
 	}
-	/* make sure that d_is_symlink above matches inode */
 	if (nd->flags & LOOKUP_RCU) {
-		if (read_seqcount_retry(&path->dentry->d_seq, seq))
-			return -ECHILD;
+		/* make sure that d_is_symlink above matches inode */
+		if (read_seqcount_retry(&path.dentry->d_seq, seq))
+			return ERR_PTR(-ECHILD);
+	} else {
+		if (path.mnt == nd->path.mnt)
+			mntget(path.mnt);
 	}
-	return pick_link(nd, path, inode, seq);
+	return pick_link(nd, &path, inode, seq, flags);
 }
 
-static int walk_component(struct nameidata *nd, int flags)
+static struct dentry *follow_dotdot_rcu(struct nameidata *nd,
+					struct inode **inodep,
+					unsigned *seqp)
 {
-	struct path path;
+	struct dentry *parent, *old;
+
+	if (path_equal(&nd->path, &nd->root))
+		goto in_root;
+	if (unlikely(nd->path.dentry == nd->path.mnt->mnt_root)) {
+		struct path path;
+		unsigned seq;
+		if (!choose_mountpoint_rcu(real_mount(nd->path.mnt),
+					   &nd->root, &path, &seq))
+			goto in_root;
+		if (unlikely(nd->flags & LOOKUP_NO_XDEV))
+			return ERR_PTR(-ECHILD);
+		nd->path = path;
+		nd->inode = path.dentry->d_inode;
+		nd->seq = seq;
+		if (unlikely(read_seqretry(&mount_lock, nd->m_seq)))
+			return ERR_PTR(-ECHILD);
+		/* we know that mountpoint was pinned */
+	}
+	old = nd->path.dentry;
+	parent = old->d_parent;
+	*inodep = parent->d_inode;
+	*seqp = read_seqcount_begin(&parent->d_seq);
+	if (unlikely(read_seqcount_retry(&old->d_seq, nd->seq)))
+		return ERR_PTR(-ECHILD);
+	if (unlikely(!path_connected(nd->path.mnt, parent)))
+		return ERR_PTR(-ECHILD);
+	return parent;
+in_root:
+	if (unlikely(read_seqretry(&mount_lock, nd->m_seq)))
+		return ERR_PTR(-ECHILD);
+	if (unlikely(nd->flags & LOOKUP_BENEATH))
+		return ERR_PTR(-ECHILD);
+	return NULL;
+}
+
+static struct dentry *follow_dotdot(struct nameidata *nd,
+				 struct inode **inodep,
+				 unsigned *seqp)
+{
+	struct dentry *parent;
+
+	if (path_equal(&nd->path, &nd->root))
+		goto in_root;
+	if (unlikely(nd->path.dentry == nd->path.mnt->mnt_root)) {
+		struct path path;
+
+		if (!choose_mountpoint(real_mount(nd->path.mnt),
+				       &nd->root, &path))
+			goto in_root;
+		path_put(&nd->path);
+		nd->path = path;
+		nd->inode = path.dentry->d_inode;
+		if (unlikely(nd->flags & LOOKUP_NO_XDEV))
+			return ERR_PTR(-EXDEV);
+	}
+	/* rare case of legitimate dget_parent()... */
+	parent = dget_parent(nd->path.dentry);
+	if (unlikely(!path_connected(nd->path.mnt, parent))) {
+		dput(parent);
+		return ERR_PTR(-ENOENT);
+	}
+	*seqp = 0;
+	*inodep = parent->d_inode;
+	return parent;
+
+in_root:
+	if (unlikely(nd->flags & LOOKUP_BENEATH))
+		return ERR_PTR(-EXDEV);
+	dget(nd->path.dentry);
+	return NULL;
+}
+
+static const char *handle_dots(struct nameidata *nd, int type)
+{
+	if (type == LAST_DOTDOT) {
+		const char *error = NULL;
+		struct dentry *parent;
+		struct inode *inode;
+		unsigned seq;
+
+		if (!nd->root.mnt) {
+			error = ERR_PTR(set_root(nd));
+			if (error)
+				return error;
+		}
+		if (nd->flags & LOOKUP_RCU)
+			parent = follow_dotdot_rcu(nd, &inode, &seq);
+		else
+			parent = follow_dotdot(nd, &inode, &seq);
+		if (IS_ERR(parent))
+			return ERR_CAST(parent);
+		if (unlikely(!parent))
+			error = step_into(nd, WALK_NOFOLLOW,
+					 nd->path.dentry, nd->inode, nd->seq);
+		else
+			error = step_into(nd, WALK_NOFOLLOW,
+					 parent, inode, seq);
+		if (unlikely(error))
+			return error;
+
+		if (unlikely(nd->flags & LOOKUP_IS_SCOPED)) {
+			/*
+			 * If there was a racing rename or mount along our
+			 * path, then we can't be sure that ".." hasn't jumped
+			 * above nd->root (and so userspace should retry or use
+			 * some fallback).
+			 */
+			smp_rmb();
+			if (unlikely(__read_seqcount_retry(&mount_lock.seqcount, nd->m_seq)))
+				return ERR_PTR(-EAGAIN);
+			if (unlikely(__read_seqcount_retry(&rename_lock.seqcount, nd->r_seq)))
+				return ERR_PTR(-EAGAIN);
+		}
+	}
+	return NULL;
+}
+
+static const char *walk_component(struct nameidata *nd, int flags)
+{
+	struct dentry *dentry;
 	struct inode *inode;
 	unsigned seq;
-	int err;
 	/*
 	 * "." and ".." are special - ".." especially so because it has
 	 * to be able to know about the current root directory and
 	 * parent relationships.
 	 */
 	if (unlikely(nd->last_type != LAST_NORM)) {
-		err = handle_dots(nd, nd->last_type);
 		if (!(flags & WALK_MORE) && nd->depth)
 			put_link(nd);
-		return err;
+		return handle_dots(nd, nd->last_type);
 	}
-	err = lookup_fast(nd, &path, &inode, &seq);
-	if (unlikely(err <= 0)) {
-		if (err < 0)
-			return err;
-		path.dentry = lookup_slow(&nd->last, nd->path.dentry,
-					  nd->flags);
-		if (IS_ERR(path.dentry))
-			return PTR_ERR(path.dentry);
-
-		path.mnt = nd->path.mnt;
-		err = follow_managed(&path, nd);
-		if (unlikely(err < 0))
-			return err;
-
-		if (unlikely(d_is_negative(path.dentry))) {
-			path_to_nameidata(&path, nd);
-			return -ENOENT;
-		}
-
-		seq = 0;	/* we are already out of RCU mode */
-		inode = d_backing_inode(path.dentry);
+	dentry = lookup_fast(nd, &inode, &seq);
+	if (IS_ERR(dentry))
+		return ERR_CAST(dentry);
+	if (unlikely(!dentry)) {
+		dentry = lookup_slow(&nd->last, nd->path.dentry, nd->flags);
+		if (IS_ERR(dentry))
+			return ERR_CAST(dentry);
 	}
-
-	return step_into(nd, &path, flags, inode, seq);
+	if (!(flags & WALK_MORE) && nd->depth)
+		put_link(nd);
+	return step_into(nd, flags, dentry, inode, seq);
 }
 
 /*
@@ -2057,8 +2090,11 @@ static inline u64 hash_name(const void *salt, const char *name)
  */
 static int link_path_walk(const char *name, struct nameidata *nd)
 {
+	int depth = 0; // depth <= nd->depth
 	int err;
 
+	nd->last_type = LAST_ROOT;
+	nd->flags |= LOOKUP_PARENT;
 	if (IS_ERR(name))
 		return PTR_ERR(name);
 	while (*name=='/')
@@ -2068,6 +2104,7 @@ static int link_path_walk(const char *name, struct nameidata *nd)
 
 	/* At this point we know we have a real path component. */
 	for(;;) {
+		const char *link;
 		u64 hash_len;
 		int type;
 
@@ -2117,36 +2154,27 @@ static int link_path_walk(const char *name, struct nameidata *nd)
 		} while (unlikely(*name == '/'));
 		if (unlikely(!*name)) {
 OK:
-			/* pathname body, done */
-			if (!nd->depth)
-				return 0;
-			name = nd->stack[nd->depth - 1].name;
-			/* trailing symlink, done */
-			if (!name)
+			/* pathname or trailing symlink, done */
+			if (!depth) {
+				nd->dir_uid = nd->inode->i_uid;
+				nd->dir_mode = nd->inode->i_mode;
+				nd->flags &= ~LOOKUP_PARENT;
 				return 0;
+			}
 			/* last component of nested symlink */
-			err = walk_component(nd, WALK_FOLLOW);
+			name = nd->stack[--depth].name;
+			link = walk_component(nd, 0);
 		} else {
 			/* not the last component */
-			err = walk_component(nd, WALK_FOLLOW | WALK_MORE);
+			link = walk_component(nd, WALK_MORE);
 		}
-		if (err < 0)
-			return err;
-
-		if (err) {
-			const char *s = get_link(nd);
-
-			if (IS_ERR(s))
-				return PTR_ERR(s);
-			err = 0;
-			if (unlikely(!s)) {
-				/* jumped */
-				put_link(nd);
-			} else {
-				nd->stack[nd->depth - 1].name = name;
-				name = s;
-				continue;
-			}
+		if (unlikely(link)) {
+			if (IS_ERR(link))
+				return PTR_ERR(link);
+			/* a symlink to follow */
+			nd->stack[depth++].name = name;
+			name = link;
+			continue;
 		}
 		if (unlikely(!d_can_lookup(nd->path.dentry))) {
 			if (nd->flags & LOOKUP_RCU) {
@@ -2161,6 +2189,7 @@ OK:
 /* must be paired with terminate_walk() */
 static const char *path_init(struct nameidata *nd, unsigned flags)
 {
+	int error;
 	const char *s = nd->name->name;
 
 	if (!*s)
@@ -2168,9 +2197,13 @@ static const char *path_init(struct nameidata *nd, unsigned flags)
 	if (flags & LOOKUP_RCU)
 		rcu_read_lock();
 
-	nd->last_type = LAST_ROOT; /* if there are only slashes... */
-	nd->flags = flags | LOOKUP_JUMPED | LOOKUP_PARENT;
+	nd->flags = flags | LOOKUP_JUMPED;
 	nd->depth = 0;
+
+	nd->m_seq = __read_seqcount_begin(&mount_lock.seqcount);
+	nd->r_seq = __read_seqcount_begin(&rename_lock.seqcount);
+	smp_rmb();
+
 	if (flags & LOOKUP_ROOT) {
 		struct dentry *root = nd->root.dentry;
 		struct inode *inode = root->d_inode;
@@ -2179,9 +2212,8 @@ static const char *path_init(struct nameidata *nd, unsigned flags)
 		nd->path = nd->root;
 		nd->inode = inode;
 		if (flags & LOOKUP_RCU) {
-			nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
+			nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq);
 			nd->root_seq = nd->seq;
-			nd->m_seq = read_seqbegin(&mount_lock);
 		} else {
 			path_get(&nd->path);
 		}
@@ -2192,13 +2224,16 @@ static const char *path_init(struct nameidata *nd, unsigned flags)
 	nd->path.mnt = NULL;
 	nd->path.dentry = NULL;
 
-	nd->m_seq = read_seqbegin(&mount_lock);
-	if (*s == '/') {
-		set_root(nd);
-		if (likely(!nd_jump_root(nd)))
-			return s;
-		return ERR_PTR(-ECHILD);
-	} else if (nd->dfd == AT_FDCWD) {
+	/* Absolute pathname -- fetch the root (LOOKUP_IN_ROOT uses nd->dfd). */
+	if (*s == '/' && !(flags & LOOKUP_IN_ROOT)) {
+		error = nd_jump_root(nd);
+		if (unlikely(error))
+			return ERR_PTR(error);
+		return s;
+	}
+
+	/* Relative pathname -- get the starting-point it is relative to. */
+	if (nd->dfd == AT_FDCWD) {
 		if (flags & LOOKUP_RCU) {
 			struct fs_struct *fs = current->fs;
 			unsigned seq;
@@ -2213,7 +2248,6 @@ static const char *path_init(struct nameidata *nd, unsigned flags)
 			get_fs_pwd(current->fs, &nd->path);
 			nd->inode = nd->path.dentry->d_inode;
 		}
-		return s;
 	} else {
 		/* Caller must check execute permissions on the starting path component */
 		struct fd f = fdget_raw(nd->dfd);
@@ -2238,58 +2272,35 @@ static const char *path_init(struct nameidata *nd, unsigned flags)
 			nd->inode = nd->path.dentry->d_inode;
 		}
 		fdput(f);
-		return s;
 	}
-}
 
-static const char *trailing_symlink(struct nameidata *nd)
-{
-	const char *s;
-	int error = may_follow_link(nd);
-	if (unlikely(error))
-		return ERR_PTR(error);
-	nd->flags |= LOOKUP_PARENT;
-	nd->stack[0].name = NULL;
-	s = get_link(nd);
-	return s ? s : "";
+	/* For scoped-lookups we need to set the root to the dirfd as well. */
+	if (flags & LOOKUP_IS_SCOPED) {
+		nd->root = nd->path;
+		if (flags & LOOKUP_RCU) {
+			nd->root_seq = nd->seq;
+		} else {
+			path_get(&nd->root);
+			nd->flags |= LOOKUP_ROOT_GRABBED;
+		}
+	}
+	return s;
 }
 
-static inline int lookup_last(struct nameidata *nd)
+static inline const char *lookup_last(struct nameidata *nd)
 {
 	if (nd->last_type == LAST_NORM && nd->last.name[nd->last.len])
 		nd->flags |= LOOKUP_FOLLOW | LOOKUP_DIRECTORY;
 
-	nd->flags &= ~LOOKUP_PARENT;
-	return walk_component(nd, 0);
+	return walk_component(nd, WALK_TRAILING);
 }
 
 static int handle_lookup_down(struct nameidata *nd)
 {
-	struct path path = nd->path;
-	struct inode *inode = nd->inode;
-	unsigned seq = nd->seq;
-	int err;
-
-	if (nd->flags & LOOKUP_RCU) {
-		/*
-		 * don't bother with unlazy_walk on failure - we are
-		 * at the very beginning of walk, so we lose nothing
-		 * if we simply redo everything in non-RCU mode
-		 */
-		if (unlikely(!__follow_mount_rcu(nd, &path, &inode, &seq)))
-			return -ECHILD;
-	} else {
-		dget(path.dentry);
-		err = follow_managed(&path, nd);
-		if (unlikely(err < 0))
-			return err;
-		inode = d_backing_inode(path.dentry);
-		seq = 0;
-	}
-	path_to_nameidata(&path, nd);
-	nd->inode = inode;
-	nd->seq = seq;
-	return 0;
+	if (!(nd->flags & LOOKUP_RCU))
+		dget(nd->path.dentry);
+	return PTR_ERR(step_into(nd, WALK_NOFOLLOW,
+			nd->path.dentry, nd->inode, nd->seq));
 }
 
 /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */
@@ -2304,16 +2315,19 @@ static int path_lookupat(struct nameidata *nd, unsigned flags, struct path *path
 			s = ERR_PTR(err);
 	}
 
-	while (!(err = link_path_walk(s, nd))
-		&& ((err = lookup_last(nd)) > 0)) {
-		s = trailing_symlink(nd);
-	}
+	while (!(err = link_path_walk(s, nd)) &&
+	       (s = lookup_last(nd)) != NULL)
+		;
 	if (!err)
 		err = complete_walk(nd);
 
 	if (!err && nd->flags & LOOKUP_DIRECTORY)
 		if (!d_can_lookup(nd->path.dentry))
 			err = -ENOTDIR;
+	if (!err && unlikely(nd->flags & LOOKUP_MOUNTPOINT)) {
+		err = handle_lookup_down(nd);
+		nd->flags &= ~LOOKUP_JUMPED; // no d_weak_revalidate(), please...
+	}
 	if (!err) {
 		*path = nd->path;
 		nd->path.mnt = NULL;
@@ -2342,7 +2356,8 @@ int filename_lookup(int dfd, struct filename *name, unsigned flags,
 		retval = path_lookupat(&nd, flags | LOOKUP_REVAL, path);
 
 	if (likely(!retval))
-		audit_inode(name, path->dentry, 0);
+		audit_inode(name, path->dentry,
+			    flags & LOOKUP_MOUNTPOINT ? AUDIT_INODE_NOEVAL : 0);
 	restore_nameidata();
 	putname(name);
 	return retval;
@@ -2564,30 +2579,49 @@ struct dentry *lookup_one_len_unlocked(const char *name,
 }
 EXPORT_SYMBOL(lookup_one_len_unlocked);
 
+/*
+ * Like lookup_one_len_unlocked(), except that it yields ERR_PTR(-ENOENT)
+ * on negatives.  Returns known positive or ERR_PTR(); that's what
+ * most of the users want.  Note that pinned negative with unlocked parent
+ * _can_ become positive at any time, so callers of lookup_one_len_unlocked()
+ * need to be very careful; pinned positives have ->d_inode stable, so
+ * this one avoids such problems.
+ */
+struct dentry *lookup_positive_unlocked(const char *name,
+				       struct dentry *base, int len)
+{
+	struct dentry *ret = lookup_one_len_unlocked(name, base, len);
+	if (!IS_ERR(ret) && d_flags_negative(smp_load_acquire(&ret->d_flags))) {
+		dput(ret);
+		ret = ERR_PTR(-ENOENT);
+	}
+	return ret;
+}
+EXPORT_SYMBOL(lookup_positive_unlocked);
+
 #ifdef CONFIG_UNIX98_PTYS
 int path_pts(struct path *path)
 {
 	/* Find something mounted on "pts" in the same directory as
 	 * the input path.
 	 */
-	struct dentry *child, *parent;
-	struct qstr this;
-	int ret;
-
-	ret = path_parent_directory(path);
-	if (ret)
-		return ret;
+	struct dentry *parent = dget_parent(path->dentry);
+	struct dentry *child;
+	struct qstr this = QSTR_INIT("pts", 3);
 
-	parent = path->dentry;
-	this.name = "pts";
-	this.len = 3;
+	if (unlikely(!path_connected(path->mnt, parent))) {
+		dput(parent);
+		return -ENOENT;
+	}
+	dput(path->dentry);
+	path->dentry = parent;
 	child = d_hash_and_lookup(parent, &this);
 	if (!child)
 		return -ENOENT;
 
 	path->dentry = child;
 	dput(parent);
-	follow_mount(path);
+	follow_down(path);
 	return 0;
 }
 #endif
@@ -2600,151 +2634,6 @@ int user_path_at_empty(int dfd, const char __user *name, unsigned flags,
 }
 EXPORT_SYMBOL(user_path_at_empty);
 
-/**
- * mountpoint_last - look up last component for umount
- * @nd:   pathwalk nameidata - currently pointing at parent directory of "last"
- *
- * This is a special lookup_last function just for umount. In this case, we
- * need to resolve the path without doing any revalidation.
- *
- * The nameidata should be the result of doing a LOOKUP_PARENT pathwalk. Since
- * mountpoints are always pinned in the dcache, their ancestors are too. Thus,
- * in almost all cases, this lookup will be served out of the dcache. The only
- * cases where it won't are if nd->last refers to a symlink or the path is
- * bogus and it doesn't exist.
- *
- * Returns:
- * -error: if there was an error during lookup. This includes -ENOENT if the
- *         lookup found a negative dentry.
- *
- * 0:      if we successfully resolved nd->last and found it to not to be a
- *         symlink that needs to be followed.
- *
- * 1:      if we successfully resolved nd->last and found it to be a symlink
- *         that needs to be followed.
- */
-static int
-mountpoint_last(struct nameidata *nd)
-{
-	int error = 0;
-	struct dentry *dir = nd->path.dentry;
-	struct path path;
-
-	/* If we're in rcuwalk, drop out of it to handle last component */
-	if (nd->flags & LOOKUP_RCU) {
-		if (unlazy_walk(nd))
-			return -ECHILD;
-	}
-
-	nd->flags &= ~LOOKUP_PARENT;
-
-	if (unlikely(nd->last_type != LAST_NORM)) {
-		error = handle_dots(nd, nd->last_type);
-		if (error)
-			return error;
-		path.dentry = dget(nd->path.dentry);
-	} else {
-		path.dentry = d_lookup(dir, &nd->last);
-		if (!path.dentry) {
-			/*
-			 * No cached dentry. Mounted dentries are pinned in the
-			 * cache, so that means that this dentry is probably
-			 * a symlink or the path doesn't actually point
-			 * to a mounted dentry.
-			 */
-			path.dentry = lookup_slow(&nd->last, dir,
-					     nd->flags | LOOKUP_NO_REVAL);
-			if (IS_ERR(path.dentry))
-				return PTR_ERR(path.dentry);
-		}
-	}
-	if (d_is_negative(path.dentry)) {
-		dput(path.dentry);
-		return -ENOENT;
-	}
-	path.mnt = nd->path.mnt;
-	return step_into(nd, &path, 0, d_backing_inode(path.dentry), 0);
-}
-
-/**
- * path_mountpoint - look up a path to be umounted
- * @nd:		lookup context
- * @flags:	lookup flags
- * @path:	pointer to container for result
- *
- * Look up the given name, but don't attempt to revalidate the last component.
- * Returns 0 and "path" will be valid on success; Returns error otherwise.
- */
-static int
-path_mountpoint(struct nameidata *nd, unsigned flags, struct path *path)
-{
-	const char *s = path_init(nd, flags);
-	int err;
-
-	while (!(err = link_path_walk(s, nd)) &&
-		(err = mountpoint_last(nd)) > 0) {
-		s = trailing_symlink(nd);
-	}
-	if (!err) {
-		*path = nd->path;
-		nd->path.mnt = NULL;
-		nd->path.dentry = NULL;
-		follow_mount(path);
-	}
-	terminate_walk(nd);
-	return err;
-}
-
-static int
-filename_mountpoint(int dfd, struct filename *name, struct path *path,
-			unsigned int flags)
-{
-	struct nameidata nd;
-	int error;
-	if (IS_ERR(name))
-		return PTR_ERR(name);
-	set_nameidata(&nd, dfd, name);
-	error = path_mountpoint(&nd, flags | LOOKUP_RCU, path);
-	if (unlikely(error == -ECHILD))
-		error = path_mountpoint(&nd, flags, path);
-	if (unlikely(error == -ESTALE))
-		error = path_mountpoint(&nd, flags | LOOKUP_REVAL, path);
-	if (likely(!error))
-		audit_inode(name, path->dentry, AUDIT_INODE_NOEVAL);
-	restore_nameidata();
-	putname(name);
-	return error;
-}
-
-/**
- * user_path_mountpoint_at - lookup a path from userland in order to umount it
- * @dfd:	directory file descriptor
- * @name:	pathname from userland
- * @flags:	lookup flags
- * @path:	pointer to container to hold result
- *
- * A umount is a special case for path walking. We're not actually interested
- * in the inode in this situation, and ESTALE errors can be a problem. We
- * simply want track down the dentry and vfsmount attached at the mountpoint
- * and avoid revalidating the last component.
- *
- * Returns 0 and populates "path" on success.
- */
-int
-user_path_mountpoint_at(int dfd, const char __user *name, unsigned int flags,
-			struct path *path)
-{
-	return filename_mountpoint(dfd, getname(name), path, flags);
-}
-
-int
-kern_path_mountpoint(int dfd, const char *name, struct path *path,
-			unsigned int flags)
-{
-	return filename_mountpoint(dfd, getname_kernel(name), path, flags);
-}
-EXPORT_SYMBOL(kern_path_mountpoint);
-
 int __check_sticky(struct inode *dir, struct inode *inode)
 {
 	kuid_t fsuid = current_fsuid();
@@ -3042,18 +2931,14 @@ static int may_o_create(const struct path *dir, struct dentry *dentry, umode_t m
  *
  * Returns an error code otherwise.
  */
-static int atomic_open(struct nameidata *nd, struct dentry *dentry,
-			struct path *path, struct file *file,
-			const struct open_flags *op,
-			int open_flag, umode_t mode)
+static struct dentry *atomic_open(struct nameidata *nd, struct dentry *dentry,
+				  struct file *file,
+				  int open_flag, umode_t mode)
 {
 	struct dentry *const DENTRY_NOT_SET = (void *) -1UL;
 	struct inode *dir =  nd->path.dentry->d_inode;
 	int error;
 
-	if (!(~open_flag & (O_EXCL | O_CREAT)))	/* both O_EXCL and O_CREAT */
-		open_flag &= ~O_TRUNC;
-
 	if (nd->flags & LOOKUP_DIRECTORY)
 		open_flag |= O_DIRECTORY;
 
@@ -3064,19 +2949,10 @@ static int atomic_open(struct nameidata *nd, struct dentry *dentry,
 	d_lookup_done(dentry);
 	if (!error) {
 		if (file->f_mode & FMODE_OPENED) {
-			/*
-			 * We didn't have the inode before the open, so check open
-			 * permission here.
-			 */
-			int acc_mode = op->acc_mode;
-			if (file->f_mode & FMODE_CREATED) {
-				WARN_ON(!(open_flag & O_CREAT));
-				fsnotify_create(dir, dentry);
-				acc_mode = 0;
+			if (unlikely(dentry != file->f_path.dentry)) {
+				dput(dentry);
+				dentry = dget(file->f_path.dentry);
 			}
-			error = may_open(&file->f_path, acc_mode, open_flag);
-			if (WARN_ON(error > 0))
-				error = -EINVAL;
 		} else if (WARN_ON(file->f_path.dentry == DENTRY_NOT_SET)) {
 			error = -EIO;
 		} else {
@@ -3084,19 +2960,15 @@ static int atomic_open(struct nameidata *nd, struct dentry *dentry,
 				dput(dentry);
 				dentry = file->f_path.dentry;
 			}
-			if (file->f_mode & FMODE_CREATED)
-				fsnotify_create(dir, dentry);
-			if (unlikely(d_is_negative(dentry))) {
+			if (unlikely(d_is_negative(dentry)))
 				error = -ENOENT;
-			} else {
-				path->dentry = dentry;
-				path->mnt = nd->path.mnt;
-				return 0;
-			}
 		}
 	}
-	dput(dentry);
-	return error;
+	if (error) {
+		dput(dentry);
+		dentry = ERR_PTR(error);
+	}
+	return dentry;
 }
 
 /*
@@ -3114,10 +2986,9 @@ static int atomic_open(struct nameidata *nd, struct dentry *dentry,
  *
  * An error code is returned on failure.
  */
-static int lookup_open(struct nameidata *nd, struct path *path,
-			struct file *file,
-			const struct open_flags *op,
-			bool got_write)
+static struct dentry *lookup_open(struct nameidata *nd, struct file *file,
+				  const struct open_flags *op,
+				  bool got_write)
 {
 	struct dentry *dir = nd->path.dentry;
 	struct inode *dir_inode = dir->d_inode;
@@ -3128,7 +2999,7 @@ static int lookup_open(struct nameidata *nd, struct path *path,
 	DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
 
 	if (unlikely(IS_DEADDIR(dir_inode)))
-		return -ENOENT;
+		return ERR_PTR(-ENOENT);
 
 	file->f_mode &= ~FMODE_CREATED;
 	dentry = d_lookup(dir, &nd->last);
@@ -3136,7 +3007,7 @@ static int lookup_open(struct nameidata *nd, struct path *path,
 		if (!dentry) {
 			dentry = d_alloc_parallel(dir, &nd->last, &wq);
 			if (IS_ERR(dentry))
-				return PTR_ERR(dentry);
+				return dentry;
 		}
 		if (d_in_lookup(dentry))
 			break;
@@ -3152,7 +3023,7 @@ static int lookup_open(struct nameidata *nd, struct path *path,
 	}
 	if (dentry->d_inode) {
 		/* Cached positive dentry: will open in f_op->open */
-		goto out_no_open;
+		return dentry;
 	}
 
 	/*
@@ -3164,41 +3035,27 @@ static int lookup_open(struct nameidata *nd, struct path *path,
 	 * Another problem is returing the "right" error value (e.g. for an
 	 * O_EXCL open we want to return EEXIST not EROFS).
 	 */
+	if (unlikely(!got_write))
+		open_flag &= ~O_TRUNC;
 	if (open_flag & O_CREAT) {
+		if (open_flag & O_EXCL)
+			open_flag &= ~O_TRUNC;
 		if (!IS_POSIXACL(dir->d_inode))
 			mode &= ~current_umask();
-		if (unlikely(!got_write)) {
-			create_error = -EROFS;
-			open_flag &= ~O_CREAT;
-			if (open_flag & (O_EXCL | O_TRUNC))
-				goto no_open;
-			/* No side effects, safe to clear O_CREAT */
-		} else {
+		if (likely(got_write))
 			create_error = may_o_create(&nd->path, dentry, mode);
-			if (create_error) {
-				open_flag &= ~O_CREAT;
-				if (open_flag & O_EXCL)
-					goto no_open;
-			}
-		}
-	} else if ((open_flag & (O_TRUNC|O_WRONLY|O_RDWR)) &&
-		   unlikely(!got_write)) {
-		/*
-		 * No O_CREATE -> atomicity not a requirement -> fall
-		 * back to lookup + open
-		 */
-		goto no_open;
+		else
+			create_error = -EROFS;
 	}
-
+	if (create_error)
+		open_flag &= ~O_CREAT;
 	if (dir_inode->i_op->atomic_open) {
-		error = atomic_open(nd, dentry, path, file, op, open_flag,
-				    mode);
-		if (unlikely(error == -ENOENT) && create_error)
-			error = create_error;
-		return error;
+		dentry = atomic_open(nd, dentry, file, open_flag, mode);
+		if (unlikely(create_error) && dentry == ERR_PTR(-ENOENT))
+			dentry = ERR_PTR(create_error);
+		return dentry;
 	}
 
-no_open:
 	if (d_in_lookup(dentry)) {
 		struct dentry *res = dir_inode->i_op->lookup(dir_inode, dentry,
 							     nd->flags);
@@ -3225,76 +3082,60 @@ no_open:
 						open_flag & O_EXCL);
 		if (error)
 			goto out_dput;
-		fsnotify_create(dir_inode, dentry);
 	}
 	if (unlikely(create_error) && !dentry->d_inode) {
 		error = create_error;
 		goto out_dput;
 	}
-out_no_open:
-	path->dentry = dentry;
-	path->mnt = nd->path.mnt;
-	return 0;
+	return dentry;
 
 out_dput:
 	dput(dentry);
-	return error;
+	return ERR_PTR(error);
 }
 
-/*
- * Handle the last step of open()
- */
-static int do_last(struct nameidata *nd,
+static const char *open_last_lookups(struct nameidata *nd,
 		   struct file *file, const struct open_flags *op)
 {
 	struct dentry *dir = nd->path.dentry;
 	int open_flag = op->open_flag;
-	bool will_truncate = (open_flag & O_TRUNC) != 0;
 	bool got_write = false;
-	int acc_mode = op->acc_mode;
 	unsigned seq;
 	struct inode *inode;
-	struct path path;
+	struct dentry *dentry;
+	const char *res;
 	int error;
 
-	nd->flags &= ~LOOKUP_PARENT;
 	nd->flags |= op->intent;
 
 	if (nd->last_type != LAST_NORM) {
-		error = handle_dots(nd, nd->last_type);
-		if (unlikely(error))
-			return error;
-		goto finish_open;
+		if (nd->depth)
+			put_link(nd);
+		return handle_dots(nd, nd->last_type);
 	}
 
 	if (!(open_flag & O_CREAT)) {
 		if (nd->last.name[nd->last.len])
 			nd->flags |= LOOKUP_FOLLOW | LOOKUP_DIRECTORY;
 		/* we _can_ be in RCU mode here */
-		error = lookup_fast(nd, &path, &inode, &seq);
-		if (likely(error > 0))
+		dentry = lookup_fast(nd, &inode, &seq);
+		if (IS_ERR(dentry))
+			return ERR_CAST(dentry);
+		if (likely(dentry))
 			goto finish_lookup;
 
-		if (error < 0)
-			return error;
-
-		BUG_ON(nd->inode != dir->d_inode);
 		BUG_ON(nd->flags & LOOKUP_RCU);
 	} else {
 		/* create side of things */
-		/*
-		 * This will *only* deal with leaving RCU mode - LOOKUP_JUMPED
-		 * has been cleared when we got to the last component we are
-		 * about to look up
-		 */
-		error = complete_walk(nd);
-		if (error)
-			return error;
-
+		if (nd->flags & LOOKUP_RCU) {
+			error = unlazy_walk(nd);
+			if (unlikely(error))
+				return ERR_PTR(error);
+		}
 		audit_inode(nd->name, dir, AUDIT_INODE_PARENT);
 		/* trailing slashes? */
 		if (unlikely(nd->last.name[nd->last.len]))
-			return -EISDIR;
+			return ERR_PTR(-EISDIR);
 	}
 
 	if (open_flag & (O_CREAT | O_TRUNC | O_WRONLY | O_RDWR)) {
@@ -3311,113 +3152,90 @@ static int do_last(struct nameidata *nd,
 		inode_lock(dir->d_inode);
 	else
 		inode_lock_shared(dir->d_inode);
-	error = lookup_open(nd, &path, file, op, got_write);
+	dentry = lookup_open(nd, file, op, got_write);
+	if (!IS_ERR(dentry) && (file->f_mode & FMODE_CREATED))
+		fsnotify_create(dir->d_inode, dentry);
 	if (open_flag & O_CREAT)
 		inode_unlock(dir->d_inode);
 	else
 		inode_unlock_shared(dir->d_inode);
 
-	if (error)
-		goto out;
-
-	if (file->f_mode & FMODE_OPENED) {
-		if ((file->f_mode & FMODE_CREATED) ||
-		    !S_ISREG(file_inode(file)->i_mode))
-			will_truncate = false;
-
-		audit_inode(nd->name, file->f_path.dentry, 0);
-		goto opened;
-	}
-
-	if (file->f_mode & FMODE_CREATED) {
-		/* Don't check for write permission, don't truncate */
-		open_flag &= ~O_TRUNC;
-		will_truncate = false;
-		acc_mode = 0;
-		path_to_nameidata(&path, nd);
-		goto finish_open_created;
-	}
-
-	/*
-	 * If atomic_open() acquired write access it is dropped now due to
-	 * possible mount and symlink following (this might be optimized away if
-	 * necessary...)
-	 */
-	if (got_write) {
+	if (got_write)
 		mnt_drop_write(nd->path.mnt);
-		got_write = false;
-	}
 
-	error = follow_managed(&path, nd);
-	if (unlikely(error < 0))
-		return error;
+	if (IS_ERR(dentry))
+		return ERR_CAST(dentry);
 
-	if (unlikely(d_is_negative(path.dentry))) {
-		path_to_nameidata(&path, nd);
-		return -ENOENT;
+	if (file->f_mode & (FMODE_OPENED | FMODE_CREATED)) {
+		dput(nd->path.dentry);
+		nd->path.dentry = dentry;
+		return NULL;
 	}
 
-	/*
-	 * create/update audit record if it already exists.
-	 */
-	audit_inode(nd->name, path.dentry, 0);
+finish_lookup:
+	if (nd->depth)
+		put_link(nd);
+	res = step_into(nd, WALK_TRAILING, dentry, inode, seq);
+	if (unlikely(res))
+		nd->flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL);
+	return res;
+}
 
-	if (unlikely((open_flag & (O_EXCL | O_CREAT)) == (O_EXCL | O_CREAT))) {
-		path_to_nameidata(&path, nd);
-		return -EEXIST;
-	}
+/*
+ * Handle the last step of open()
+ */
+static int do_open(struct nameidata *nd,
+		   struct file *file, const struct open_flags *op)
+{
+	int open_flag = op->open_flag;
+	bool do_truncate;
+	int acc_mode;
+	int error;
 
-	seq = 0;	/* out of RCU mode, so the value doesn't matter */
-	inode = d_backing_inode(path.dentry);
-finish_lookup:
-	error = step_into(nd, &path, 0, inode, seq);
-	if (unlikely(error))
-		return error;
-finish_open:
-	/* Why this, you ask?  _Now_ we might have grown LOOKUP_JUMPED... */
-	error = complete_walk(nd);
-	if (error)
-		return error;
-	audit_inode(nd->name, nd->path.dentry, 0);
+	if (!(file->f_mode & (FMODE_OPENED | FMODE_CREATED))) {
+		error = complete_walk(nd);
+		if (error)
+			return error;
+	}
+	if (!(file->f_mode & FMODE_CREATED))
+		audit_inode(nd->name, nd->path.dentry, 0);
 	if (open_flag & O_CREAT) {
-		error = -EISDIR;
+		if ((open_flag & O_EXCL) && !(file->f_mode & FMODE_CREATED))
+			return -EEXIST;
 		if (d_is_dir(nd->path.dentry))
-			goto out;
-		error = may_create_in_sticky(dir,
+			return -EISDIR;
+		error = may_create_in_sticky(nd->dir_mode, nd->dir_uid,
 					     d_backing_inode(nd->path.dentry));
 		if (unlikely(error))
-			goto out;
+			return error;
 	}
-	error = -ENOTDIR;
 	if ((nd->flags & LOOKUP_DIRECTORY) && !d_can_lookup(nd->path.dentry))
-		goto out;
-	if (!d_is_reg(nd->path.dentry))
-		will_truncate = false;
+		return -ENOTDIR;
 
-	if (will_truncate) {
+	do_truncate = false;
+	acc_mode = op->acc_mode;
+	if (file->f_mode & FMODE_CREATED) {
+		/* Don't check for write permission, don't truncate */
+		open_flag &= ~O_TRUNC;
+		acc_mode = 0;
+	} else if (d_is_reg(nd->path.dentry) && open_flag & O_TRUNC) {
 		error = mnt_want_write(nd->path.mnt);
 		if (error)
-			goto out;
-		got_write = true;
+			return error;
+		do_truncate = true;
 	}
-finish_open_created:
 	error = may_open(&nd->path, acc_mode, open_flag);
-	if (error)
-		goto out;
-	BUG_ON(file->f_mode & FMODE_OPENED); /* once it's opened, it's opened */
-	error = vfs_open(&nd->path, file);
-	if (error)
-		goto out;
-opened:
-	error = ima_file_check(file, op->acc_mode);
-	if (!error && will_truncate)
+	if (!error && !(file->f_mode & FMODE_OPENED))
+		error = vfs_open(&nd->path, file);
+	if (!error)
+		error = ima_file_check(file, op->acc_mode);
+	if (!error && do_truncate)
 		error = handle_truncate(file);
-out:
 	if (unlikely(error > 0)) {
 		WARN_ON(1);
 		error = -EINVAL;
 	}
-	if (got_write)
+	if (do_truncate)
 		mnt_drop_write(nd->path.mnt);
 	return error;
 }
@@ -3522,10 +3340,10 @@ static struct file *path_openat(struct nameidata *nd,
 	} else {
 		const char *s = path_init(nd, flags);
 		while (!(error = link_path_walk(s, nd)) &&
-			(error = do_last(nd, file, op)) > 0) {
-			nd->flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL);
-			s = trailing_symlink(nd);
-		}
+		       (s = open_last_lookups(nd, file, op)) != NULL)
+			;
+		if (!error)
+			error = do_open(nd, file, op);
 		terminate_walk(nd);
 	}
 	if (likely(!error)) {
@@ -3687,12 +3505,14 @@ EXPORT_SYMBOL(user_path_create);
 
 int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev)
 {
+	bool is_whiteout = S_ISCHR(mode) && dev == WHITEOUT_DEV;
 	int error = may_create(dir, dentry);
 
 	if (error)
 		return error;
 
-	if ((S_ISCHR(mode) || S_ISBLK(mode)) && !capable(CAP_MKNOD))
+	if ((S_ISCHR(mode) || S_ISBLK(mode)) && !is_whiteout &&
+	    !capable(CAP_MKNOD))
 		return -EPERM;
 
 	if (!dir->i_op->mknod)
@@ -4527,9 +4347,6 @@ static int do_renameat2(int olddfd, const char __user *oldname, int newdfd,
 	    (flags & RENAME_EXCHANGE))
 		return -EINVAL;
 
-	if ((flags & RENAME_WHITEOUT) && !capable(CAP_MKNOD))
-		return -EPERM;
-
 	if (flags & RENAME_EXCHANGE)
 		target_flags = 0;
 
@@ -4665,20 +4482,6 @@ SYSCALL_DEFINE2(rename, const char __user *, oldname, const char __user *, newna
 	return do_renameat2(AT_FDCWD, oldname, AT_FDCWD, newname, 0);
 }
 
-int vfs_whiteout(struct inode *dir, struct dentry *dentry)
-{
-	int error = may_create(dir, dentry);
-	if (error)
-		return error;
-
-	if (!dir->i_op->mknod)
-		return -EPERM;
-
-	return dir->i_op->mknod(dir, dentry,
-				S_IFCHR | WHITEOUT_MODE, WHITEOUT_DEV);
-}
-EXPORT_SYMBOL(vfs_whiteout);
-
 int readlink_copy(char __user *buffer, int buflen, const char *link)
 {
 	int len = PTR_ERR(link);
diff --git a/fs/namespace.c b/fs/namespace.c
index 2adfe7b166a3..a6baee3c7904 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -648,6 +648,21 @@ struct vfsmount *lookup_mnt(const struct path *path)
 	return m;
 }
 
+static inline void lock_ns_list(struct mnt_namespace *ns)
+{
+	spin_lock(&ns->ns_lock);
+}
+
+static inline void unlock_ns_list(struct mnt_namespace *ns)
+{
+	spin_unlock(&ns->ns_lock);
+}
+
+static inline bool mnt_is_cursor(struct mount *mnt)
+{
+	return mnt->mnt.mnt_flags & MNT_CURSOR;
+}
+
 /*
  * __is_local_mountpoint - Test to see if dentry is a mountpoint in the
  *                         current mount namespace.
@@ -673,11 +688,15 @@ bool __is_local_mountpoint(struct dentry *dentry)
 		goto out;
 
 	down_read(&namespace_sem);
+	lock_ns_list(ns);
 	list_for_each_entry(mnt, &ns->list, mnt_list) {
+		if (mnt_is_cursor(mnt))
+			continue;
 		is_covered = (mnt->mnt_mountpoint == dentry);
 		if (is_covered)
 			break;
 	}
+	unlock_ns_list(ns);
 	up_read(&namespace_sem);
 out:
 	return is_covered;
@@ -1245,46 +1264,71 @@ struct vfsmount *mnt_clone_internal(const struct path *path)
 }
 
 #ifdef CONFIG_PROC_FS
+static struct mount *mnt_list_next(struct mnt_namespace *ns,
+				   struct list_head *p)
+{
+	struct mount *mnt, *ret = NULL;
+
+	lock_ns_list(ns);
+	list_for_each_continue(p, &ns->list) {
+		mnt = list_entry(p, typeof(*mnt), mnt_list);
+		if (!mnt_is_cursor(mnt)) {
+			ret = mnt;
+			break;
+		}
+	}
+	unlock_ns_list(ns);
+
+	return ret;
+}
+
 /* iterator; we want it to have access to namespace_sem, thus here... */
 static void *m_start(struct seq_file *m, loff_t *pos)
 {
 	struct proc_mounts *p = m->private;
+	struct list_head *prev;
 
 	down_read(&namespace_sem);
-	if (p->cached_event == p->ns->event) {
-		void *v = p->cached_mount;
-		if (*pos == p->cached_index)
-			return v;
-		if (*pos == p->cached_index + 1) {
-			v = seq_list_next(v, &p->ns->list, &p->cached_index);
-			return p->cached_mount = v;
-		}
+	if (!*pos) {
+		prev = &p->ns->list;
+	} else {
+		prev = &p->cursor.mnt_list;
+
+		/* Read after we'd reached the end? */
+		if (list_empty(prev))
+			return NULL;
 	}
 
-	p->cached_event = p->ns->event;
-	p->cached_mount = seq_list_start(&p->ns->list, *pos);
-	p->cached_index = *pos;
-	return p->cached_mount;
+	return mnt_list_next(p->ns, prev);
 }
 
 static void *m_next(struct seq_file *m, void *v, loff_t *pos)
 {
 	struct proc_mounts *p = m->private;
+	struct mount *mnt = v;
 
-	p->cached_mount = seq_list_next(v, &p->ns->list, pos);
-	p->cached_index = *pos;
-	return p->cached_mount;
+	++*pos;
+	return mnt_list_next(p->ns, &mnt->mnt_list);
 }
 
 static void m_stop(struct seq_file *m, void *v)
 {
+	struct proc_mounts *p = m->private;
+	struct mount *mnt = v;
+
+	lock_ns_list(p->ns);
+	if (mnt)
+		list_move_tail(&p->cursor.mnt_list, &mnt->mnt_list);
+	else
+		list_del_init(&p->cursor.mnt_list);
+	unlock_ns_list(p->ns);
 	up_read(&namespace_sem);
 }
 
 static int m_show(struct seq_file *m, void *v)
 {
 	struct proc_mounts *p = m->private;
-	struct mount *r = list_entry(v, struct mount, mnt_list);
+	struct mount *r = v;
 	return p->show(m, &r->mnt);
 }
 
@@ -1294,6 +1338,15 @@ const struct seq_operations mounts_op = {
 	.stop	= m_stop,
 	.show	= m_show,
 };
+
+void mnt_cursor_del(struct mnt_namespace *ns, struct mount *cursor)
+{
+	down_read(&namespace_sem);
+	lock_ns_list(ns);
+	list_del(&cursor->mnt_list);
+	unlock_ns_list(ns);
+	up_read(&namespace_sem);
+}
 #endif  /* CONFIG_PROC_FS */
 
 /**
@@ -1669,7 +1722,7 @@ int ksys_umount(char __user *name, int flags)
 	struct path path;
 	struct mount *mnt;
 	int retval;
-	int lookup_flags = 0;
+	int lookup_flags = LOOKUP_MOUNTPOINT;
 
 	if (flags & ~(MNT_FORCE | MNT_DETACH | MNT_EXPIRE | UMOUNT_NOFOLLOW))
 		return -EINVAL;
@@ -1680,7 +1733,7 @@ int ksys_umount(char __user *name, int flags)
 	if (!(flags & UMOUNT_NOFOLLOW))
 		lookup_flags |= LOOKUP_FOLLOW;
 
-	retval = user_path_mountpoint_at(AT_FDCWD, name, lookup_flags, &path);
+	retval = user_path_at(AT_FDCWD, name, lookup_flags, &path);
 	if (retval)
 		goto out;
 	mnt = real_mount(path.mnt);
@@ -1728,7 +1781,7 @@ static bool is_mnt_ns_file(struct dentry *dentry)
 	       dentry->d_fsdata == &mntns_operations;
 }
 
-struct mnt_namespace *to_mnt_ns(struct ns_common *ns)
+static struct mnt_namespace *to_mnt_ns(struct ns_common *ns)
 {
 	return container_of(ns, struct mnt_namespace, ns);
 }
@@ -2356,7 +2409,7 @@ static struct file *open_detached_copy(struct path *path, bool recursive)
 	return file;
 }
 
-SYSCALL_DEFINE3(open_tree, int, dfd, const char *, filename, unsigned, flags)
+SYSCALL_DEFINE3(open_tree, int, dfd, const char __user *, filename, unsigned, flags)
 {
 	struct file *file;
 	struct path path;
@@ -2697,45 +2750,32 @@ static int do_move_mount_old(struct path *path, const char *old_name)
 /*
  * add a mount into a namespace's mount tree
  */
-static int do_add_mount(struct mount *newmnt, struct path *path, int mnt_flags)
+static int do_add_mount(struct mount *newmnt, struct mountpoint *mp,
+			struct path *path, int mnt_flags)
 {
-	struct mountpoint *mp;
-	struct mount *parent;
-	int err;
+	struct mount *parent = real_mount(path->mnt);
 
 	mnt_flags &= ~MNT_INTERNAL_FLAGS;
 
-	mp = lock_mount(path);
-	if (IS_ERR(mp))
-		return PTR_ERR(mp);
-
-	parent = real_mount(path->mnt);
-	err = -EINVAL;
 	if (unlikely(!check_mnt(parent))) {
 		/* that's acceptable only for automounts done in private ns */
 		if (!(mnt_flags & MNT_SHRINKABLE))
-			goto unlock;
+			return -EINVAL;
 		/* ... and for those we'd better have mountpoint still alive */
 		if (!parent->mnt_ns)
-			goto unlock;
+			return -EINVAL;
 	}
 
 	/* Refuse the same filesystem on the same mount point */
-	err = -EBUSY;
 	if (path->mnt->mnt_sb == newmnt->mnt.mnt_sb &&
 	    path->mnt->mnt_root == path->dentry)
-		goto unlock;
+		return -EBUSY;
 
-	err = -EINVAL;
 	if (d_is_symlink(newmnt->mnt.mnt_root))
-		goto unlock;
+		return -EINVAL;
 
 	newmnt->mnt.mnt_flags = mnt_flags;
-	err = graft_tree(newmnt, parent, mp);
-
-unlock:
-	unlock_mount(mp);
-	return err;
+	return graft_tree(newmnt, parent, mp);
 }
 
 static bool mount_too_revealing(const struct super_block *sb, int *new_mnt_flags);
@@ -2748,6 +2788,7 @@ static int do_new_mount_fc(struct fs_context *fc, struct path *mountpoint,
 			   unsigned int mnt_flags)
 {
 	struct vfsmount *mnt;
+	struct mountpoint *mp;
 	struct super_block *sb = fc->root->d_sb;
 	int error;
 
@@ -2768,7 +2809,13 @@ static int do_new_mount_fc(struct fs_context *fc, struct path *mountpoint,
 
 	mnt_warn_timestamp_expiry(mountpoint, mnt);
 
-	error = do_add_mount(real_mount(mnt), mountpoint, mnt_flags);
+	mp = lock_mount(mountpoint);
+	if (IS_ERR(mp)) {
+		mntput(mnt);
+		return PTR_ERR(mp);
+	}
+	error = do_add_mount(real_mount(mnt), mp, mountpoint, mnt_flags);
+	unlock_mount(mp);
 	if (error < 0)
 		mntput(mnt);
 	return error;
@@ -2829,23 +2876,63 @@ static int do_new_mount(struct path *path, const char *fstype, int sb_flags,
 
 int finish_automount(struct vfsmount *m, struct path *path)
 {
-	struct mount *mnt = real_mount(m);
+	struct dentry *dentry = path->dentry;
+	struct mountpoint *mp;
+	struct mount *mnt;
 	int err;
+
+	if (!m)
+		return 0;
+	if (IS_ERR(m))
+		return PTR_ERR(m);
+
+	mnt = real_mount(m);
 	/* The new mount record should have at least 2 refs to prevent it being
 	 * expired before we get a chance to add it
 	 */
 	BUG_ON(mnt_get_count(mnt) < 2);
 
 	if (m->mnt_sb == path->mnt->mnt_sb &&
-	    m->mnt_root == path->dentry) {
+	    m->mnt_root == dentry) {
 		err = -ELOOP;
-		goto fail;
+		goto discard;
 	}
 
-	err = do_add_mount(mnt, path, path->mnt->mnt_flags | MNT_SHRINKABLE);
-	if (!err)
-		return 0;
-fail:
+	/*
+	 * we don't want to use lock_mount() - in this case finding something
+	 * that overmounts our mountpoint to be means "quitely drop what we've
+	 * got", not "try to mount it on top".
+	 */
+	inode_lock(dentry->d_inode);
+	namespace_lock();
+	if (unlikely(cant_mount(dentry))) {
+		err = -ENOENT;
+		goto discard_locked;
+	}
+	rcu_read_lock();
+	if (unlikely(__lookup_mnt(path->mnt, dentry))) {
+		rcu_read_unlock();
+		err = 0;
+		goto discard_locked;
+	}
+	rcu_read_unlock();
+	mp = get_mountpoint(dentry);
+	if (IS_ERR(mp)) {
+		err = PTR_ERR(mp);
+		goto discard_locked;
+	}
+
+	err = do_add_mount(mnt, mp, path, path->mnt->mnt_flags | MNT_SHRINKABLE);
+	unlock_mount(mp);
+	if (unlikely(err))
+		goto discard;
+	mntput(m);
+	return 0;
+
+discard_locked:
+	namespace_unlock();
+	inode_unlock(dentry->d_inode);
+discard:
 	/* remove m from any expiration list it may be on */
 	if (!list_empty(&mnt->mnt_expire)) {
 		namespace_lock();
@@ -2979,39 +3066,10 @@ static void shrink_submounts(struct mount *mnt)
 	}
 }
 
-/*
- * Some copy_from_user() implementations do not return the exact number of
- * bytes remaining to copy on a fault.  But copy_mount_options() requires that.
- * Note that this function differs from copy_from_user() in that it will oops
- * on bad values of `to', rather than returning a short copy.
- */
-static long exact_copy_from_user(void *to, const void __user * from,
-				 unsigned long n)
-{
-	char *t = to;
-	const char __user *f = from;
-	char c;
-
-	if (!access_ok(from, n))
-		return n;
-
-	while (n) {
-		if (__get_user(c, f)) {
-			memset(t, 0, n);
-			break;
-		}
-		*t++ = c;
-		f++;
-		n--;
-	}
-	return n;
-}
-
 void *copy_mount_options(const void __user * data)
 {
-	int i;
-	unsigned long size;
 	char *copy;
+	unsigned size;
 
 	if (!data)
 		return NULL;
@@ -3020,22 +3078,16 @@ void *copy_mount_options(const void __user * data)
 	if (!copy)
 		return ERR_PTR(-ENOMEM);
 
-	/* We only care that *some* data at the address the user
-	 * gave us is valid.  Just in case, we'll zero
-	 * the remainder of the page.
-	 */
-	/* copy_from_user cannot cross TASK_SIZE ! */
-	size = TASK_SIZE - (unsigned long)untagged_addr(data);
-	if (size > PAGE_SIZE)
-		size = PAGE_SIZE;
+	size = PAGE_SIZE - offset_in_page(data);
 
-	i = size - exact_copy_from_user(copy, data, size);
-	if (!i) {
+	if (copy_from_user(copy, data, size)) {
 		kfree(copy);
 		return ERR_PTR(-EFAULT);
 	}
-	if (i != PAGE_SIZE)
-		memset(copy + i, 0, PAGE_SIZE - i);
+	if (size != PAGE_SIZE) {
+		if (copy_from_user(copy + size, data + size, PAGE_SIZE - size))
+			memset(copy + size, 0, PAGE_SIZE - size);
+	}
 	return copy;
 }
 
@@ -3203,6 +3255,7 @@ static struct mnt_namespace *alloc_mnt_ns(struct user_namespace *user_ns, bool a
 	atomic_set(&new_ns->count, 1);
 	INIT_LIST_HEAD(&new_ns->list);
 	init_waitqueue_head(&new_ns->poll);
+	spin_lock_init(&new_ns->ns_lock);
 	new_ns->user_ns = get_user_ns(user_ns);
 	new_ns->ucounts = ucounts;
 	return new_ns;
@@ -3325,8 +3378,8 @@ struct dentry *mount_subtree(struct vfsmount *m, const char *name)
 }
 EXPORT_SYMBOL(mount_subtree);
 
-int ksys_mount(const char __user *dev_name, const char __user *dir_name,
-	       const char __user *type, unsigned long flags, void __user *data)
+SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name,
+		char __user *, type, unsigned long, flags, void __user *, data)
 {
 	int ret;
 	char *kernel_type;
@@ -3359,12 +3412,6 @@ out_type:
 	return ret;
 }
 
-SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name,
-		char __user *, type, unsigned long, flags, void __user *, data)
-{
-	return ksys_mount(dev_name, dir_name, type, flags, data);
-}
-
 /*
  * Create a kernel mount representation for a new, prepared superblock
  * (specified by fs_fd) and attach to an open_tree-like file descriptor.
@@ -3514,8 +3561,8 @@ err_fsfd:
  * Note the flags value is a combination of MOVE_MOUNT_* flags.
  */
 SYSCALL_DEFINE5(move_mount,
-		int, from_dfd, const char *, from_pathname,
-		int, to_dfd, const char *, to_pathname,
+		int, from_dfd, const char __user *, from_pathname,
+		int, to_dfd, const char __user *, to_pathname,
 		unsigned int, flags)
 {
 	struct path from_path, to_path;
@@ -3602,7 +3649,7 @@ EXPORT_SYMBOL(path_is_under);
  * file system may be mounted on put_old. After all, new_root is a mountpoint.
  *
  * Also, the current root cannot be on the 'rootfs' (initial ramfs) filesystem.
- * See Documentation/filesystems/ramfs-rootfs-initramfs.txt for alternatives
+ * See Documentation/filesystems/ramfs-rootfs-initramfs.rst for alternatives
  * in this situation.
  *
  * Notes:
@@ -3849,10 +3896,14 @@ static bool mnt_already_visible(struct mnt_namespace *ns,
 	bool visible = false;
 
 	down_read(&namespace_sem);
+	lock_ns_list(ns);
 	list_for_each_entry(mnt, &ns->list, mnt_list) {
 		struct mount *child;
 		int mnt_flags;
 
+		if (mnt_is_cursor(mnt))
+			continue;
+
 		if (mnt->mnt.mnt_sb->s_type != sb->s_type)
 			continue;
 
@@ -3900,6 +3951,7 @@ static bool mnt_already_visible(struct mnt_namespace *ns,
 	next:	;
 	}
 found:
+	unlock_ns_list(ns);
 	up_read(&namespace_sem);
 	return visible;
 }
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index 295a7a21b774..88e1763e02f3 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -90,7 +90,7 @@ config NFS_V4
 config NFS_SWAP
 	bool "Provide swap over NFS support"
 	default n
-	depends on NFS_FS
+	depends on NFS_FS && SWAP
 	select SUNRPC_SWAP
 	help
 	  This option enables swapon to work on files located on NFS mounts.
@@ -164,7 +164,7 @@ config ROOT_NFS
 	  If you want your system to mount its root file system via NFS,
 	  choose Y here.  This is common practice for managing systems
 	  without local permanent storage.  For details, read
-	  <file:Documentation/filesystems/nfs/nfsroot.txt>.
+	  <file:Documentation/admin-guide/nfs/nfsroot.rst>.
 
 	  Most people say N here.
 
@@ -196,3 +196,12 @@ config NFS_DEBUG
 	depends on NFS_FS && SUNRPC_DEBUG
 	select CRC32
 	default y
+
+config NFS_DISABLE_UDP_SUPPORT
+       bool "NFS: Disable NFS UDP protocol support"
+       depends on NFS_FS
+       default y
+       help
+	 Choose Y here to disable the use of NFS over UDP. NFS over UDP
+	 on modern networks (1Gb+) can lead to data corruption caused by
+	 fragmentation during high loads.
diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile
index 34cdeaecccf6..2433c3e03cfa 100644
--- a/fs/nfs/Makefile
+++ b/fs/nfs/Makefile
@@ -9,7 +9,7 @@ CFLAGS_nfstrace.o += -I$(src)
 nfs-y 			:= client.o dir.o file.o getroot.o inode.o super.o \
 			   io.o direct.o pagelist.o read.o symlink.o unlink.o \
 			   write.o namespace.o mount_clnt.o nfstrace.o \
-			   export.o sysfs.o
+			   export.o sysfs.o fs_context.o
 nfs-$(CONFIG_ROOT_NFS)	+= nfsroot.o
 nfs-$(CONFIG_SYSCTL)	+= sysctl.o
 nfs-$(CONFIG_NFS_FSCACHE) += fscache.o fscache-index.o
diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 690221747b47..d1a0e2c8b1b4 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -476,7 +476,7 @@ static void bl_free_layout_hdr(struct pnfs_layout_hdr *lo)
 	err = ext_tree_remove(bl, true, 0, LLONG_MAX);
 	WARN_ON(err);
 
-	kfree(bl);
+	kfree_rcu(bl, bl_layout.plh_rcu);
 }
 
 static struct pnfs_layout_hdr *__bl_alloc_layout_hdr(struct inode *inode,
diff --git a/fs/nfs/blocklayout/extent_tree.c b/fs/nfs/blocklayout/extent_tree.c
index 7a57ff2528af..8f7cff7a4293 100644
--- a/fs/nfs/blocklayout/extent_tree.c
+++ b/fs/nfs/blocklayout/extent_tree.c
@@ -582,7 +582,7 @@ retry:
 		if (!arg->layoutupdate_pages)
 			return -ENOMEM;
 
-		start_p = __vmalloc(buffer_size, GFP_NOFS, PAGE_KERNEL);
+		start_p = __vmalloc(buffer_size, GFP_NOFS);
 		if (!start_p) {
 			kfree(arg->layoutupdate_pages);
 			return -ENOMEM;
diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 8f34daf85f70..6a2033131c06 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -72,8 +72,8 @@ struct cb_getattrres {
 	uint32_t bitmap[2];
 	uint64_t size;
 	uint64_t change_attr;
-	struct timespec ctime;
-	struct timespec mtime;
+	struct timespec64 ctime;
+	struct timespec64 mtime;
 };
 
 struct cb_recallargs {
@@ -127,7 +127,9 @@ extern __be32 nfs4_callback_sequence(void *argp, void *resp,
 #define RCA4_TYPE_MASK_OBJ_LAYOUT_MAX  9
 #define RCA4_TYPE_MASK_OTHER_LAYOUT_MIN 12
 #define RCA4_TYPE_MASK_OTHER_LAYOUT_MAX 15
-#define RCA4_TYPE_MASK_ALL 0xf31f
+#define PNFS_FF_RCA4_TYPE_MASK_READ 16
+#define PNFS_FF_RCA4_TYPE_MASK_RW 17
+#define RCA4_TYPE_MASK_ALL 0x3f31f
 
 struct cb_recallanyargs {
 	uint32_t	craa_objs_to_keep;
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index f39924ba050b..e61dbc9b86ae 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -26,7 +26,6 @@ __be32 nfs4_callback_getattr(void *argp, void *resp,
 	struct cb_getattrargs *args = argp;
 	struct cb_getattrres *res = resp;
 	struct nfs_delegation *delegation;
-	struct nfs_inode *nfsi;
 	struct inode *inode;
 
 	res->status = htonl(NFS4ERR_OP_NOT_IN_SESSION);
@@ -47,17 +46,16 @@ __be32 nfs4_callback_getattr(void *argp, void *resp,
 				-ntohl(res->status));
 		goto out;
 	}
-	nfsi = NFS_I(inode);
 	rcu_read_lock();
-	delegation = rcu_dereference(nfsi->delegation);
+	delegation = nfs4_get_valid_delegation(inode);
 	if (delegation == NULL || (delegation->type & FMODE_WRITE) == 0)
 		goto out_iput;
 	res->size = i_size_read(inode);
 	res->change_attr = delegation->change_attr;
 	if (nfs_have_writebacks(inode))
 		res->change_attr++;
-	res->ctime = timespec64_to_timespec(inode->i_ctime);
-	res->mtime = timespec64_to_timespec(inode->i_mtime);
+	res->ctime = inode->i_ctime;
+	res->mtime = inode->i_mtime;
 	res->bitmap[0] = (FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE) &
 		args->bitmap[0];
 	res->bitmap[1] = (FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY) &
@@ -123,31 +121,31 @@ out:
  */
 static struct inode *nfs_layout_find_inode_by_stateid(struct nfs_client *clp,
 		const nfs4_stateid *stateid)
+	__must_hold(RCU)
 {
 	struct nfs_server *server;
 	struct inode *inode;
 	struct pnfs_layout_hdr *lo;
 
+	rcu_read_lock();
 	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		list_for_each_entry(lo, &server->layouts, plh_layouts) {
+		list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) {
+			if (!pnfs_layout_is_valid(lo))
+				continue;
 			if (stateid != NULL &&
 			    !nfs4_stateid_match_other(stateid, &lo->plh_stateid))
 				continue;
+			if (!nfs_sb_active(server->super))
+				continue;
 			inode = igrab(lo->plh_inode);
-			if (!inode)
-				return ERR_PTR(-EAGAIN);
-			if (!nfs_sb_active(inode->i_sb)) {
-				rcu_read_unlock();
-				spin_unlock(&clp->cl_lock);
-				iput(inode);
-				spin_lock(&clp->cl_lock);
-				rcu_read_lock();
-				return ERR_PTR(-EAGAIN);
-			}
-			return inode;
+			rcu_read_unlock();
+			if (inode)
+				return inode;
+			nfs_sb_deactive(server->super);
+			return ERR_PTR(-EAGAIN);
 		}
 	}
-
+	rcu_read_unlock();
 	return ERR_PTR(-ENOENT);
 }
 
@@ -165,28 +163,25 @@ static struct inode *nfs_layout_find_inode_by_fh(struct nfs_client *clp,
 	struct inode *inode;
 	struct pnfs_layout_hdr *lo;
 
+	rcu_read_lock();
 	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		list_for_each_entry(lo, &server->layouts, plh_layouts) {
+		list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) {
 			nfsi = NFS_I(lo->plh_inode);
 			if (nfs_compare_fh(fh, &nfsi->fh))
 				continue;
 			if (nfsi->layout != lo)
 				continue;
+			if (!nfs_sb_active(server->super))
+				continue;
 			inode = igrab(lo->plh_inode);
-			if (!inode)
-				return ERR_PTR(-EAGAIN);
-			if (!nfs_sb_active(inode->i_sb)) {
-				rcu_read_unlock();
-				spin_unlock(&clp->cl_lock);
-				iput(inode);
-				spin_lock(&clp->cl_lock);
-				rcu_read_lock();
-				return ERR_PTR(-EAGAIN);
-			}
-			return inode;
+			rcu_read_unlock();
+			if (inode)
+				return inode;
+			nfs_sb_deactive(server->super);
+			return ERR_PTR(-EAGAIN);
 		}
 	}
-
+	rcu_read_unlock();
 	return ERR_PTR(-ENOENT);
 }
 
@@ -196,14 +191,9 @@ static struct inode *nfs_layout_find_inode(struct nfs_client *clp,
 {
 	struct inode *inode;
 
-	spin_lock(&clp->cl_lock);
-	rcu_read_lock();
 	inode = nfs_layout_find_inode_by_stateid(clp, stateid);
 	if (inode == ERR_PTR(-ENOENT))
 		inode = nfs_layout_find_inode_by_fh(clp, fh);
-	rcu_read_unlock();
-	spin_unlock(&clp->cl_lock);
-
 	return inode;
 }
 
@@ -282,7 +272,7 @@ static u32 initiate_file_draining(struct nfs_client *clp,
 		goto unlock;
 	}
 
-	pnfs_set_layout_stateid(lo, &args->cbl_stateid, true);
+	pnfs_set_layout_stateid(lo, &args->cbl_stateid, NULL, true);
 	switch (pnfs_mark_matching_lsegs_return(lo, &free_me_list,
 				&args->cbl_range,
 				be32_to_cpu(args->cbl_stateid.seqid))) {
@@ -607,6 +597,7 @@ __be32 nfs4_callback_recallany(void *argp, void *resp,
 	struct cb_recallanyargs *args = argp;
 	__be32 status;
 	fmode_t flags = 0;
+	bool schedule_manager = false;
 
 	status = cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION);
 	if (!cps->clp) /* set in cb_sequence */
@@ -629,6 +620,18 @@ __be32 nfs4_callback_recallany(void *argp, void *resp,
 
 	if (args->craa_type_mask & BIT(RCA4_TYPE_MASK_FILE_LAYOUT))
 		pnfs_recall_all_layouts(cps->clp);
+
+	if (args->craa_type_mask & BIT(PNFS_FF_RCA4_TYPE_MASK_READ)) {
+		set_bit(NFS4CLNT_RECALL_ANY_LAYOUT_READ, &cps->clp->cl_state);
+		schedule_manager = true;
+	}
+	if (args->craa_type_mask & BIT(PNFS_FF_RCA4_TYPE_MASK_RW)) {
+		set_bit(NFS4CLNT_RECALL_ANY_LAYOUT_RW, &cps->clp->cl_state);
+		schedule_manager = true;
+	}
+	if (schedule_manager)
+		nfs4_schedule_state_manager(cps->clp);
+
 out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
 	return status;
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 73a5a5ea2976..79ff172eb1c8 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -18,6 +18,7 @@
 #include "callback.h"
 #include "internal.h"
 #include "nfs4session.h"
+#include "nfs4trace.h"
 
 #define CB_OP_TAGLEN_MAXSZ		(512)
 #define CB_OP_HDR_RES_MAXSZ		(2 * 4) // opcode, status
@@ -627,7 +628,7 @@ static __be32 encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, u
 	return 0;
 }
 
-static __be32 encode_attr_time(struct xdr_stream *xdr, const struct timespec *time)
+static __be32 encode_attr_time(struct xdr_stream *xdr, const struct timespec64 *time)
 {
 	__be32 *p;
 
@@ -639,14 +640,14 @@ static __be32 encode_attr_time(struct xdr_stream *xdr, const struct timespec *ti
 	return 0;
 }
 
-static __be32 encode_attr_ctime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time)
+static __be32 encode_attr_ctime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec64 *time)
 {
 	if (!(bitmap[1] & FATTR4_WORD1_TIME_METADATA))
 		return 0;
 	return encode_attr_time(xdr,time);
 }
 
-static __be32 encode_attr_mtime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time)
+static __be32 encode_attr_mtime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec64 *time)
 {
 	if (!(bitmap[1] & FATTR4_WORD1_TIME_MODIFY))
 		return 0;
@@ -946,9 +947,13 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp)
 
 	if (hdr_arg.minorversion == 0) {
 		cps.clp = nfs4_find_client_ident(SVC_NET(rqstp), hdr_arg.cb_ident);
-		if (!cps.clp || !check_gss_callback_principal(cps.clp, rqstp)) {
-			if (cps.clp)
-				nfs_put_client(cps.clp);
+		if (!cps.clp) {
+			trace_nfs_cb_no_clp(rqstp->rq_xid, hdr_arg.cb_ident);
+			goto out_invalidcred;
+		}
+		if (!check_gss_callback_principal(cps.clp, rqstp)) {
+			trace_nfs_cb_badprinc(rqstp->rq_xid, hdr_arg.cb_ident);
+			nfs_put_client(cps.clp);
 			goto out_invalidcred;
 		}
 	}
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 30838304a0bf..f1ff3076e4a4 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -153,6 +153,7 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init)
 	if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL)
 		goto error_0;
 
+	clp->cl_minorversion = cl_init->minorversion;
 	clp->cl_nfs_mod = cl_init->nfs_mod;
 	if (!try_module_get(clp->cl_nfs_mod->owner))
 		goto error_dealloc;
@@ -312,6 +313,12 @@ again:
 		/* Match nfsv4 minorversion */
 		if (clp->cl_minorversion != data->minorversion)
 			continue;
+
+		/* Match request for a dedicated DS */
+		if (test_bit(NFS_CS_DS, &data->init_flags) !=
+		    test_bit(NFS_CS_DS, &clp->cl_flags))
+			continue;
+
 		/* Match the full socket address */
 		if (!rpc_cmp_addr_port(sap, clap))
 			/* Match all xprt_switch full socket addresses */
@@ -468,6 +475,7 @@ void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 			to->to_maxval = to->to_initval;
 		to->to_exponential = 0;
 		break;
+#ifndef CONFIG_NFS_DISABLE_UDP_SUPPORT
 	case XPRT_TRANSPORT_UDP:
 		if (retrans == NFS_UNSPEC_RETRANS)
 			to->to_retries = NFS_DEF_UDP_RETRANS;
@@ -478,6 +486,7 @@ void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 		to->to_maxval = NFS_MAX_UDP_TIMEOUT;
 		to->to_exponential = 1;
 		break;
+#endif
 	default:
 		BUG();
 	}
@@ -515,6 +524,10 @@ int nfs_create_rpc_client(struct nfs_client *clp,
 		args.flags |= RPC_CLNT_CREATE_NONPRIVPORT;
 	if (test_bit(NFS_CS_INFINITE_SLOTS, &clp->cl_flags))
 		args.flags |= RPC_CLNT_CREATE_INFINITE_SLOTS;
+	if (test_bit(NFS_CS_NOPING, &clp->cl_flags))
+		args.flags |= RPC_CLNT_CREATE_NOPING;
+	if (test_bit(NFS_CS_REUSEPORT, &clp->cl_flags))
+		args.flags |= RPC_CLNT_CREATE_REUSEPORT;
 
 	if (!IS_ERR(clp->cl_rpcclient))
 		return 0;
@@ -570,8 +583,10 @@ static int nfs_start_lockd(struct nfs_server *server)
 		default:
 			nlm_init.protocol = IPPROTO_TCP;
 			break;
+#ifndef CONFIG_NFS_DISABLE_UDP_SUPPORT
 		case XPRT_TRANSPORT_UDP:
 			nlm_init.protocol = IPPROTO_UDP;
+#endif
 	}
 
 	host = nlmclnt_init(&nlm_init);
@@ -648,27 +663,28 @@ EXPORT_SYMBOL_GPL(nfs_init_client);
  * Create a version 2 or 3 client
  */
 static int nfs_init_server(struct nfs_server *server,
-			   const struct nfs_parsed_mount_data *data,
-			   struct nfs_subversion *nfs_mod)
+			   const struct fs_context *fc)
 {
+	const struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	struct rpc_timeout timeparms;
 	struct nfs_client_initdata cl_init = {
-		.hostname = data->nfs_server.hostname,
-		.addr = (const struct sockaddr *)&data->nfs_server.address,
-		.addrlen = data->nfs_server.addrlen,
-		.nfs_mod = nfs_mod,
-		.proto = data->nfs_server.protocol,
-		.net = data->net,
+		.hostname = ctx->nfs_server.hostname,
+		.addr = (const struct sockaddr *)&ctx->nfs_server.address,
+		.addrlen = ctx->nfs_server.addrlen,
+		.nfs_mod = ctx->nfs_mod,
+		.proto = ctx->nfs_server.protocol,
+		.net = fc->net_ns,
 		.timeparms = &timeparms,
 		.cred = server->cred,
-		.nconnect = data->nfs_server.nconnect,
+		.nconnect = ctx->nfs_server.nconnect,
+		.init_flags = (1UL << NFS_CS_REUSEPORT),
 	};
 	struct nfs_client *clp;
 	int error;
 
-	nfs_init_timeout_values(&timeparms, data->nfs_server.protocol,
-			data->timeo, data->retrans);
-	if (data->flags & NFS_MOUNT_NORESVPORT)
+	nfs_init_timeout_values(&timeparms, ctx->nfs_server.protocol,
+				ctx->timeo, ctx->retrans);
+	if (ctx->flags & NFS_MOUNT_NORESVPORT)
 		set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
 
 	/* Allocate or find a client reference we can use */
@@ -679,46 +695,46 @@ static int nfs_init_server(struct nfs_server *server,
 	server->nfs_client = clp;
 
 	/* Initialise the client representation from the mount data */
-	server->flags = data->flags;
-	server->options = data->options;
+	server->flags = ctx->flags;
+	server->options = ctx->options;
 	server->caps |= NFS_CAP_HARDLINKS|NFS_CAP_SYMLINKS|NFS_CAP_FILEID|
 		NFS_CAP_MODE|NFS_CAP_NLINK|NFS_CAP_OWNER|NFS_CAP_OWNER_GROUP|
 		NFS_CAP_ATIME|NFS_CAP_CTIME|NFS_CAP_MTIME;
 
-	if (data->rsize)
-		server->rsize = nfs_block_size(data->rsize, NULL);
-	if (data->wsize)
-		server->wsize = nfs_block_size(data->wsize, NULL);
+	if (ctx->rsize)
+		server->rsize = nfs_block_size(ctx->rsize, NULL);
+	if (ctx->wsize)
+		server->wsize = nfs_block_size(ctx->wsize, NULL);
 
-	server->acregmin = data->acregmin * HZ;
-	server->acregmax = data->acregmax * HZ;
-	server->acdirmin = data->acdirmin * HZ;
-	server->acdirmax = data->acdirmax * HZ;
+	server->acregmin = ctx->acregmin * HZ;
+	server->acregmax = ctx->acregmax * HZ;
+	server->acdirmin = ctx->acdirmin * HZ;
+	server->acdirmax = ctx->acdirmax * HZ;
 
 	/* Start lockd here, before we might error out */
 	error = nfs_start_lockd(server);
 	if (error < 0)
 		goto error;
 
-	server->port = data->nfs_server.port;
-	server->auth_info = data->auth_info;
+	server->port = ctx->nfs_server.port;
+	server->auth_info = ctx->auth_info;
 
 	error = nfs_init_server_rpcclient(server, &timeparms,
-					  data->selected_flavor);
+					  ctx->selected_flavor);
 	if (error < 0)
 		goto error;
 
 	/* Preserve the values of mount_server-related mount options */
-	if (data->mount_server.addrlen) {
-		memcpy(&server->mountd_address, &data->mount_server.address,
-			data->mount_server.addrlen);
-		server->mountd_addrlen = data->mount_server.addrlen;
+	if (ctx->mount_server.addrlen) {
+		memcpy(&server->mountd_address, &ctx->mount_server.address,
+			ctx->mount_server.addrlen);
+		server->mountd_addrlen = ctx->mount_server.addrlen;
 	}
-	server->mountd_version = data->mount_server.version;
-	server->mountd_port = data->mount_server.port;
-	server->mountd_protocol = data->mount_server.protocol;
+	server->mountd_version = ctx->mount_server.version;
+	server->mountd_port = ctx->mount_server.port;
+	server->mountd_protocol = ctx->mount_server.protocol;
 
-	server->namelen  = data->namlen;
+	server->namelen  = ctx->namlen;
 	return 0;
 
 error:
@@ -940,9 +956,9 @@ EXPORT_SYMBOL_GPL(nfs_free_server);
  * Create a version 2 or 3 volume record
  * - keyed on server and FSID
  */
-struct nfs_server *nfs_create_server(struct nfs_mount_info *mount_info,
-				     struct nfs_subversion *nfs_mod)
+struct nfs_server *nfs_create_server(struct fs_context *fc)
 {
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	struct nfs_server *server;
 	struct nfs_fattr *fattr;
 	int error;
@@ -959,18 +975,18 @@ struct nfs_server *nfs_create_server(struct nfs_mount_info *mount_info,
 		goto error;
 
 	/* Get a client representation */
-	error = nfs_init_server(server, mount_info->parsed, nfs_mod);
+	error = nfs_init_server(server, fc);
 	if (error < 0)
 		goto error;
 
 	/* Probe the root fh to retrieve its FSID */
-	error = nfs_probe_fsinfo(server, mount_info->mntfh, fattr);
+	error = nfs_probe_fsinfo(server, ctx->mntfh, fattr);
 	if (error < 0)
 		goto error;
 	if (server->nfs_client->rpc_ops->version == 3) {
 		if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN)
 			server->namelen = NFS3_MAXNAMLEN;
-		if (!(mount_info->parsed->flags & NFS_MOUNT_NORDIRPLUS))
+		if (!(ctx->flags & NFS_MOUNT_NORDIRPLUS))
 			server->caps |= NFS_CAP_READDIRPLUS;
 	} else {
 		if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN)
@@ -978,8 +994,8 @@ struct nfs_server *nfs_create_server(struct nfs_mount_info *mount_info,
 	}
 
 	if (!(fattr->valid & NFS_ATTR_FATTR)) {
-		error = nfs_mod->rpc_ops->getattr(server, mount_info->mntfh,
-				fattr, NULL, NULL);
+		error = ctx->nfs_mod->rpc_ops->getattr(server, ctx->mntfh,
+						       fattr, NULL, NULL);
 		if (error < 0) {
 			dprintk("nfs_create_server: getattr error = %d\n", -error);
 			goto error;
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index af549d70ec50..816e1427f17e 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -25,13 +25,46 @@
 #include "internal.h"
 #include "nfs4trace.h"
 
-static void nfs_free_delegation(struct nfs_delegation *delegation)
+#define NFS_DEFAULT_DELEGATION_WATERMARK (5000U)
+
+static atomic_long_t nfs_active_delegations;
+static unsigned nfs_delegation_watermark = NFS_DEFAULT_DELEGATION_WATERMARK;
+
+static void __nfs_free_delegation(struct nfs_delegation *delegation)
 {
 	put_cred(delegation->cred);
 	delegation->cred = NULL;
 	kfree_rcu(delegation, rcu);
 }
 
+static void nfs_mark_delegation_revoked(struct nfs_delegation *delegation)
+{
+	if (!test_and_set_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) {
+		delegation->stateid.type = NFS4_INVALID_STATEID_TYPE;
+		atomic_long_dec(&nfs_active_delegations);
+		if (!test_bit(NFS_DELEGATION_RETURNING, &delegation->flags))
+			nfs_clear_verifier_delegated(delegation->inode);
+	}
+}
+
+static struct nfs_delegation *nfs_get_delegation(struct nfs_delegation *delegation)
+{
+	refcount_inc(&delegation->refcount);
+	return delegation;
+}
+
+static void nfs_put_delegation(struct nfs_delegation *delegation)
+{
+	if (refcount_dec_and_test(&delegation->refcount))
+		__nfs_free_delegation(delegation);
+}
+
+static void nfs_free_delegation(struct nfs_delegation *delegation)
+{
+	nfs_mark_delegation_revoked(delegation);
+	nfs_put_delegation(delegation);
+}
+
 /**
  * nfs_mark_delegation_referenced - set delegation's REFERENCED flag
  * @delegation: delegation to process
@@ -199,7 +232,7 @@ void nfs_inode_reclaim_delegation(struct inode *inode, const struct cred *cred,
 	delegation = rcu_dereference(NFS_I(inode)->delegation);
 	if (delegation != NULL) {
 		spin_lock(&delegation->lock);
-		if (delegation->inode != NULL) {
+		if (nfs4_is_valid_delegation(delegation, 0)) {
 			nfs4_stateid_copy(&delegation->stateid, stateid);
 			delegation->type = type;
 			delegation->pagemod_limit = pagemod_limit;
@@ -222,14 +255,18 @@ void nfs_inode_reclaim_delegation(struct inode *inode, const struct cred *cred,
 
 static int nfs_do_return_delegation(struct inode *inode, struct nfs_delegation *delegation, int issync)
 {
+	const struct cred *cred;
 	int res = 0;
 
-	if (!test_bit(NFS_DELEGATION_REVOKED, &delegation->flags))
-		res = nfs4_proc_delegreturn(inode,
-				delegation->cred,
+	if (!test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) {
+		spin_lock(&delegation->lock);
+		cred = get_cred(delegation->cred);
+		spin_unlock(&delegation->lock);
+		res = nfs4_proc_delegreturn(inode, cred,
 				&delegation->stateid,
 				issync);
-	nfs_free_delegation(delegation);
+		put_cred(cred);
+	}
 	return res;
 }
 
@@ -255,9 +292,13 @@ nfs_start_delegation_return_locked(struct nfs_inode *nfsi)
 	if (delegation == NULL)
 		goto out;
 	spin_lock(&delegation->lock);
-	if (!test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags))
-		ret = delegation;
+	if (!test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) {
+		/* Refcount matched in nfs_end_delegation_return() */
+		ret = nfs_get_delegation(delegation);
+	}
 	spin_unlock(&delegation->lock);
+	if (ret)
+		nfs_clear_verifier_delegated(&nfsi->vfs_inode);
 out:
 	return ret;
 }
@@ -298,7 +339,10 @@ nfs_detach_delegation_locked(struct nfs_inode *nfsi,
 		return NULL;
 
 	spin_lock(&delegation->lock);
-	set_bit(NFS_DELEGATION_RETURNING, &delegation->flags);
+	if (!delegation->inode) {
+		spin_unlock(&delegation->lock);
+		return NULL;
+	}
 	list_del_rcu(&delegation->super_list);
 	delegation->inode = NULL;
 	rcu_assign_pointer(nfsi->delegation, NULL);
@@ -325,10 +369,24 @@ nfs_inode_detach_delegation(struct inode *inode)
 	struct nfs_server *server = NFS_SERVER(inode);
 	struct nfs_delegation *delegation;
 
-	delegation = nfs_start_delegation_return(nfsi);
-	if (delegation == NULL)
-		return NULL;
-	return nfs_detach_delegation(nfsi, delegation, server);
+	rcu_read_lock();
+	delegation = rcu_dereference(nfsi->delegation);
+	if (delegation != NULL)
+		delegation = nfs_detach_delegation(nfsi, delegation, server);
+	rcu_read_unlock();
+	return delegation;
+}
+
+static void
+nfs_update_delegation_cred(struct nfs_delegation *delegation,
+		const struct cred *cred)
+{
+	const struct cred *old;
+
+	if (cred_fscmp(delegation->cred, cred) != 0) {
+		old = xchg(&delegation->cred, get_cred(cred));
+		put_cred(old);
+	}
 }
 
 static void
@@ -339,6 +397,14 @@ nfs_update_inplace_delegation(struct nfs_delegation *delegation,
 		delegation->stateid.seqid = update->stateid.seqid;
 		smp_wmb();
 		delegation->type = update->type;
+		delegation->pagemod_limit = update->pagemod_limit;
+		if (test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) {
+			delegation->change_attr = update->change_attr;
+			nfs_update_delegation_cred(delegation, update->cred);
+			/* smp_mb__before_atomic() is implicit due to xchg() */
+			clear_bit(NFS_DELEGATION_REVOKED, &delegation->flags);
+			atomic_long_inc(&nfs_active_delegations);
+		}
 	}
 }
 
@@ -368,6 +434,7 @@ int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred,
 	if (delegation == NULL)
 		return -ENOMEM;
 	nfs4_stateid_copy(&delegation->stateid, stateid);
+	refcount_set(&delegation->refcount, 1);
 	delegation->type = type;
 	delegation->pagemod_limit = pagemod_limit;
 	delegation->change_attr = inode_peek_iversion_raw(inode);
@@ -379,14 +446,18 @@ int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred,
 	spin_lock(&clp->cl_lock);
 	old_delegation = rcu_dereference_protected(nfsi->delegation,
 					lockdep_is_held(&clp->cl_lock));
-	if (old_delegation != NULL) {
-		/* Is this an update of the existing delegation? */
-		if (nfs4_stateid_match_other(&old_delegation->stateid,
-					&delegation->stateid)) {
-			nfs_update_inplace_delegation(old_delegation,
-					delegation);
-			goto out;
-		}
+	if (old_delegation == NULL)
+		goto add_new;
+	/* Is this an update of the existing delegation? */
+	if (nfs4_stateid_match_other(&old_delegation->stateid,
+				&delegation->stateid)) {
+		spin_lock(&old_delegation->lock);
+		nfs_update_inplace_delegation(old_delegation,
+				delegation);
+		spin_unlock(&old_delegation->lock);
+		goto out;
+	}
+	if (!test_bit(NFS_DELEGATION_REVOKED, &old_delegation->flags)) {
 		/*
 		 * Deal with broken servers that hand out two
 		 * delegations for the same file.
@@ -405,15 +476,17 @@ int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred,
 		if (test_and_set_bit(NFS_DELEGATION_RETURNING,
 					&old_delegation->flags))
 			goto out;
-		freeme = nfs_detach_delegation_locked(nfsi,
-				old_delegation, clp);
-		if (freeme == NULL)
-			goto out;
 	}
+	freeme = nfs_detach_delegation_locked(nfsi, old_delegation, clp);
+	if (freeme == NULL)
+		goto out;
+add_new:
 	list_add_tail_rcu(&delegation->super_list, &server->delegations);
 	rcu_assign_pointer(nfsi->delegation, delegation);
 	delegation = NULL;
 
+	atomic_long_inc(&nfs_active_delegations);
+
 	trace_nfs4_set_delegation(inode, type);
 
 	spin_lock(&inode->i_lock);
@@ -423,9 +496,11 @@ int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred,
 out:
 	spin_unlock(&clp->cl_lock);
 	if (delegation != NULL)
-		nfs_free_delegation(delegation);
-	if (freeme != NULL)
+		__nfs_free_delegation(delegation);
+	if (freeme != NULL) {
 		nfs_do_return_delegation(inode, freeme, 0);
+		nfs_free_delegation(freeme);
+	}
 	return status;
 }
 
@@ -435,7 +510,6 @@ out:
 static int nfs_end_delegation_return(struct inode *inode, struct nfs_delegation *delegation, int issync)
 {
 	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
-	struct nfs_inode *nfsi = NFS_I(inode);
 	int err = 0;
 
 	if (delegation == NULL)
@@ -457,11 +531,11 @@ static int nfs_end_delegation_return(struct inode *inode, struct nfs_delegation
 		nfs_abort_delegation_return(delegation, clp);
 		goto out;
 	}
-	if (!nfs_detach_delegation(nfsi, delegation, NFS_SERVER(inode)))
-		goto out;
 
 	err = nfs_do_return_delegation(inode, delegation, issync);
 out:
+	/* Refcount matched in nfs_start_delegation_return_locked() */
+	nfs_put_delegation(delegation);
 	return err;
 }
 
@@ -469,11 +543,9 @@ static bool nfs_delegation_need_return(struct nfs_delegation *delegation)
 {
 	bool ret = false;
 
-	if (test_bit(NFS_DELEGATION_RETURNING, &delegation->flags))
-		goto out;
 	if (test_and_clear_bit(NFS_DELEGATION_RETURN, &delegation->flags))
 		ret = true;
-	if (test_and_clear_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags) && !ret) {
+	else if (test_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags)) {
 		struct inode *inode;
 
 		spin_lock(&delegation->lock);
@@ -482,25 +554,20 @@ static bool nfs_delegation_need_return(struct nfs_delegation *delegation)
 			ret = true;
 		spin_unlock(&delegation->lock);
 	}
-out:
+	if (ret)
+		clear_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags);
+	if (test_bit(NFS_DELEGATION_RETURNING, &delegation->flags) ||
+	    test_bit(NFS_DELEGATION_REVOKED, &delegation->flags))
+		ret = false;
+
 	return ret;
 }
 
-/**
- * nfs_client_return_marked_delegations - return previously marked delegations
- * @clp: nfs_client to process
- *
- * Note that this function is designed to be called by the state
- * manager thread. For this reason, it cannot flush the dirty data,
- * since that could deadlock in case of a state recovery error.
- *
- * Returns zero on success, or a negative errno value.
- */
-int nfs_client_return_marked_delegations(struct nfs_client *clp)
+static int nfs_server_return_marked_delegations(struct nfs_server *server,
+		void __always_unused *data)
 {
 	struct nfs_delegation *delegation;
 	struct nfs_delegation *prev;
-	struct nfs_server *server;
 	struct inode *inode;
 	struct inode *place_holder = NULL;
 	struct nfs_delegation *place_holder_deleg = NULL;
@@ -510,94 +577,100 @@ restart:
 	/*
 	 * To avoid quadratic looping we hold a reference
 	 * to an inode place_holder.  Each time we restart, we
-	 * list nfs_servers from the server of that inode, and
-	 * delegation in the server from the delegations of that
-	 * inode.
+	 * list delegation in the server from the delegations
+	 * of that inode.
 	 * prev is an RCU-protected pointer to a delegation which
 	 * wasn't marked for return and might be a good choice for
 	 * the next place_holder.
 	 */
-	rcu_read_lock();
 	prev = NULL;
+	delegation = NULL;
+	rcu_read_lock();
 	if (place_holder)
-		server = NFS_SERVER(place_holder);
-	else
-		server = list_entry_rcu(clp->cl_superblocks.next,
-					struct nfs_server, client_link);
-	list_for_each_entry_from_rcu(server, &clp->cl_superblocks, client_link) {
-		delegation = NULL;
-		if (place_holder && server == NFS_SERVER(place_holder))
-			delegation = rcu_dereference(NFS_I(place_holder)->delegation);
-		if (!delegation || delegation != place_holder_deleg)
-			delegation = list_entry_rcu(server->delegations.next,
-						    struct nfs_delegation, super_list);
-		list_for_each_entry_from_rcu(delegation, &server->delegations, super_list) {
-			struct inode *to_put = NULL;
-
-			if (!nfs_delegation_need_return(delegation)) {
+		delegation = rcu_dereference(NFS_I(place_holder)->delegation);
+	if (!delegation || delegation != place_holder_deleg)
+		delegation = list_entry_rcu(server->delegations.next,
+					    struct nfs_delegation, super_list);
+	list_for_each_entry_from_rcu(delegation, &server->delegations, super_list) {
+		struct inode *to_put = NULL;
+
+		if (test_bit(NFS_DELEGATION_INODE_FREEING, &delegation->flags))
+			continue;
+		if (!nfs_delegation_need_return(delegation)) {
+			if (nfs4_is_valid_delegation(delegation, 0))
 				prev = delegation;
-				continue;
-			}
-			if (!nfs_sb_active(server->super))
-				break; /* continue in outer loop */
-
-			if (prev) {
-				struct inode *tmp;
-
-				tmp = nfs_delegation_grab_inode(prev);
-				if (tmp) {
-					to_put = place_holder;
-					place_holder = tmp;
-					place_holder_deleg = prev;
-				}
-			}
+			continue;
+		}
 
-			inode = nfs_delegation_grab_inode(delegation);
-			if (inode == NULL) {
-				rcu_read_unlock();
-				if (to_put)
-					iput(to_put);
-				nfs_sb_deactive(server->super);
-				goto restart;
+		if (prev) {
+			struct inode *tmp = nfs_delegation_grab_inode(prev);
+			if (tmp) {
+				to_put = place_holder;
+				place_holder = tmp;
+				place_holder_deleg = prev;
 			}
-			delegation = nfs_start_delegation_return_locked(NFS_I(inode));
+		}
+
+		inode = nfs_delegation_grab_inode(delegation);
+		if (inode == NULL) {
 			rcu_read_unlock();
+			iput(to_put);
+			goto restart;
+		}
+		delegation = nfs_start_delegation_return_locked(NFS_I(inode));
+		rcu_read_unlock();
 
-			if (to_put)
-				iput(to_put);
+		iput(to_put);
 
-			err = nfs_end_delegation_return(inode, delegation, 0);
-			iput(inode);
-			nfs_sb_deactive(server->super);
-			cond_resched();
-			if (!err)
-				goto restart;
-			set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
-			if (place_holder)
-				iput(place_holder);
-			return err;
-		}
+		err = nfs_end_delegation_return(inode, delegation, 0);
+		iput(inode);
+		cond_resched();
+		if (!err)
+			goto restart;
+		set_bit(NFS4CLNT_DELEGRETURN, &server->nfs_client->cl_state);
+		goto out;
 	}
 	rcu_read_unlock();
-	if (place_holder)
-		iput(place_holder);
-	return 0;
+out:
+	iput(place_holder);
+	return err;
+}
+
+/**
+ * nfs_client_return_marked_delegations - return previously marked delegations
+ * @clp: nfs_client to process
+ *
+ * Note that this function is designed to be called by the state
+ * manager thread. For this reason, it cannot flush the dirty data,
+ * since that could deadlock in case of a state recovery error.
+ *
+ * Returns zero on success, or a negative errno value.
+ */
+int nfs_client_return_marked_delegations(struct nfs_client *clp)
+{
+	return nfs_client_for_each_server(clp,
+			nfs_server_return_marked_delegations, NULL);
 }
 
 /**
- * nfs_inode_return_delegation_noreclaim - return delegation, don't reclaim opens
+ * nfs_inode_evict_delegation - return delegation, don't reclaim opens
  * @inode: inode to process
  *
  * Does not protect against delegation reclaims, therefore really only safe
- * to be called from nfs4_clear_inode().
+ * to be called from nfs4_clear_inode(). Guaranteed to always free
+ * the delegation structure.
  */
-void nfs_inode_return_delegation_noreclaim(struct inode *inode)
+void nfs_inode_evict_delegation(struct inode *inode)
 {
 	struct nfs_delegation *delegation;
 
 	delegation = nfs_inode_detach_delegation(inode);
-	if (delegation != NULL)
+	if (delegation != NULL) {
+		set_bit(NFS_DELEGATION_RETURNING, &delegation->flags);
+		set_bit(NFS_DELEGATION_INODE_FREEING, &delegation->flags);
 		nfs_do_return_delegation(inode, delegation, 1);
+		nfs_free_delegation(delegation);
+	}
 }
 
 /**
@@ -624,6 +697,43 @@ int nfs4_inode_return_delegation(struct inode *inode)
 }
 
 /**
+ * nfs_inode_return_delegation_on_close - asynchronously return a delegation
+ * @inode: inode to process
+ *
+ * This routine is called on file close in order to determine if the
+ * inode delegation needs to be returned immediately.
+ */
+void nfs4_inode_return_delegation_on_close(struct inode *inode)
+{
+	struct nfs_delegation *delegation;
+	struct nfs_delegation *ret = NULL;
+
+	if (!inode)
+		return;
+	rcu_read_lock();
+	delegation = nfs4_get_valid_delegation(inode);
+	if (!delegation)
+		goto out;
+	if (test_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags) ||
+	    atomic_long_read(&nfs_active_delegations) >= nfs_delegation_watermark) {
+		spin_lock(&delegation->lock);
+		if (delegation->inode &&
+		    list_empty(&NFS_I(inode)->open_files) &&
+		    !test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) {
+			clear_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags);
+			/* Refcount matched in nfs_end_delegation_return() */
+			ret = nfs_get_delegation(delegation);
+		}
+		spin_unlock(&delegation->lock);
+		if (ret)
+			nfs_clear_verifier_delegated(inode);
+	}
+out:
+	rcu_read_unlock();
+	nfs_end_delegation_return(inode, ret, 0);
+}
+
+/**
  * nfs4_inode_make_writeable
  * @inode: pointer to inode
  *
@@ -633,10 +743,18 @@ int nfs4_inode_return_delegation(struct inode *inode)
  */
 int nfs4_inode_make_writeable(struct inode *inode)
 {
-	if (!nfs4_has_session(NFS_SERVER(inode)->nfs_client) ||
-	    !nfs4_check_delegation(inode, FMODE_WRITE))
-		return nfs4_inode_return_delegation(inode);
-	return 0;
+	struct nfs_delegation *delegation;
+
+	rcu_read_lock();
+	delegation = nfs4_get_valid_delegation(inode);
+	if (delegation == NULL ||
+	    (nfs4_has_session(NFS_SERVER(inode)->nfs_client) &&
+	     (delegation->type & FMODE_WRITE))) {
+		rcu_read_unlock();
+		return 0;
+	}
+	rcu_read_unlock();
+	return nfs4_inode_return_delegation(inode);
 }
 
 static void nfs_mark_return_if_closed_delegation(struct nfs_server *server,
@@ -739,15 +857,7 @@ static void nfs_client_mark_return_unused_delegation_types(struct nfs_client *cl
 	rcu_read_unlock();
 }
 
-static void nfs_mark_delegation_revoked(struct nfs_server *server,
-		struct nfs_delegation *delegation)
-{
-	set_bit(NFS_DELEGATION_REVOKED, &delegation->flags);
-	delegation->stateid.type = NFS4_INVALID_STATEID_TYPE;
-	nfs_mark_return_delegation(server, delegation);
-}
-
-static bool nfs_revoke_delegation(struct inode *inode,
+static void nfs_revoke_delegation(struct inode *inode,
 		const nfs4_stateid *stateid)
 {
 	struct nfs_delegation *delegation;
@@ -761,29 +871,69 @@ static bool nfs_revoke_delegation(struct inode *inode,
 	if (stateid == NULL) {
 		nfs4_stateid_copy(&tmp, &delegation->stateid);
 		stateid = &tmp;
-	} else if (!nfs4_stateid_match(stateid, &delegation->stateid))
-		goto out;
-	nfs_mark_delegation_revoked(NFS_SERVER(inode), delegation);
+	} else {
+		if (!nfs4_stateid_match_other(stateid, &delegation->stateid))
+			goto out;
+		spin_lock(&delegation->lock);
+		if (stateid->seqid) {
+			if (nfs4_stateid_is_newer(&delegation->stateid, stateid)) {
+				spin_unlock(&delegation->lock);
+				goto out;
+			}
+			delegation->stateid.seqid = stateid->seqid;
+		}
+		spin_unlock(&delegation->lock);
+	}
+	nfs_mark_delegation_revoked(delegation);
 	ret = true;
 out:
 	rcu_read_unlock();
 	if (ret)
 		nfs_inode_find_state_and_recover(inode, stateid);
-	return ret;
 }
 
 void nfs_remove_bad_delegation(struct inode *inode,
 		const nfs4_stateid *stateid)
 {
+	nfs_revoke_delegation(inode, stateid);
+}
+EXPORT_SYMBOL_GPL(nfs_remove_bad_delegation);
+
+void nfs_delegation_mark_returned(struct inode *inode,
+		const nfs4_stateid *stateid)
+{
 	struct nfs_delegation *delegation;
 
-	if (!nfs_revoke_delegation(inode, stateid))
+	if (!inode)
 		return;
-	delegation = nfs_inode_detach_delegation(inode);
-	if (delegation)
-		nfs_free_delegation(delegation);
+
+	rcu_read_lock();
+	delegation = rcu_dereference(NFS_I(inode)->delegation);
+	if (!delegation)
+		goto out_rcu_unlock;
+
+	spin_lock(&delegation->lock);
+	if (!nfs4_stateid_match_other(stateid, &delegation->stateid))
+		goto out_spin_unlock;
+	if (stateid->seqid) {
+		/* If delegation->stateid is newer, dont mark as returned */
+		if (nfs4_stateid_is_newer(&delegation->stateid, stateid))
+			goto out_clear_returning;
+		if (delegation->stateid.seqid != stateid->seqid)
+			delegation->stateid.seqid = stateid->seqid;
+	}
+
+	nfs_mark_delegation_revoked(delegation);
+
+out_clear_returning:
+	clear_bit(NFS_DELEGATION_RETURNING, &delegation->flags);
+out_spin_unlock:
+	spin_unlock(&delegation->lock);
+out_rcu_unlock:
+	rcu_read_unlock();
+
+	nfs_inode_find_state_and_recover(inode, stateid);
 }
-EXPORT_SYMBOL_GPL(nfs_remove_bad_delegation);
 
 /**
  * nfs_expire_unused_delegation_types
@@ -840,7 +990,7 @@ int nfs_async_inode_return_delegation(struct inode *inode,
 	struct nfs_delegation *delegation;
 
 	rcu_read_lock();
-	delegation = rcu_dereference(NFS_I(inode)->delegation);
+	delegation = nfs4_get_valid_delegation(inode);
 	if (delegation == NULL)
 		goto out_enoent;
 	if (stateid != NULL &&
@@ -866,6 +1016,7 @@ nfs_delegation_find_inode_server(struct nfs_server *server,
 	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
 		spin_lock(&delegation->lock);
 		if (delegation->inode != NULL &&
+		    !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags) &&
 		    nfs_compare_fh(fhandle, &NFS_I(delegation->inode)->fh) == 0) {
 			freeme = igrab(delegation->inode);
 			if (freeme && nfs_sb_active(freeme->i_sb))
@@ -941,52 +1092,51 @@ void nfs_delegation_mark_reclaim(struct nfs_client *clp)
 	rcu_read_unlock();
 }
 
-/**
- * nfs_delegation_reap_unclaimed - reap unclaimed delegations after reboot recovery is done
- * @clp: nfs_client to process
- *
- */
-void nfs_delegation_reap_unclaimed(struct nfs_client *clp)
+static int nfs_server_reap_unclaimed_delegations(struct nfs_server *server,
+		void __always_unused *data)
 {
 	struct nfs_delegation *delegation;
-	struct nfs_server *server;
 	struct inode *inode;
-
 restart:
 	rcu_read_lock();
-	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		list_for_each_entry_rcu(delegation, &server->delegations,
-								super_list) {
-			if (test_bit(NFS_DELEGATION_INODE_FREEING,
-						&delegation->flags) ||
-			    test_bit(NFS_DELEGATION_RETURNING,
-						&delegation->flags) ||
-			    test_bit(NFS_DELEGATION_NEED_RECLAIM,
-						&delegation->flags) == 0)
-				continue;
-			if (!nfs_sb_active(server->super))
-				break; /* continue in outer loop */
-			inode = nfs_delegation_grab_inode(delegation);
-			if (inode == NULL) {
-				rcu_read_unlock();
-				nfs_sb_deactive(server->super);
-				goto restart;
-			}
-			delegation = nfs_start_delegation_return_locked(NFS_I(inode));
-			rcu_read_unlock();
-			if (delegation != NULL) {
-				delegation = nfs_detach_delegation(NFS_I(inode),
-					delegation, server);
-				if (delegation != NULL)
-					nfs_free_delegation(delegation);
-			}
-			iput(inode);
-			nfs_sb_deactive(server->super);
-			cond_resched();
-			goto restart;
+restart_locked:
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
+		if (test_bit(NFS_DELEGATION_INODE_FREEING,
+					&delegation->flags) ||
+		    test_bit(NFS_DELEGATION_RETURNING,
+					&delegation->flags) ||
+		    test_bit(NFS_DELEGATION_NEED_RECLAIM,
+					&delegation->flags) == 0)
+			continue;
+		inode = nfs_delegation_grab_inode(delegation);
+		if (inode == NULL)
+			goto restart_locked;
+		delegation = nfs_start_delegation_return_locked(NFS_I(inode));
+		rcu_read_unlock();
+		if (delegation != NULL) {
+			if (nfs_detach_delegation(NFS_I(inode), delegation,
+						server) != NULL)
+				nfs_free_delegation(delegation);
+			/* Match nfs_start_delegation_return_locked */
+			nfs_put_delegation(delegation);
 		}
+		iput(inode);
+		cond_resched();
+		goto restart;
 	}
 	rcu_read_unlock();
+	return 0;
+}
+
+/**
+ * nfs_delegation_reap_unclaimed - reap unclaimed delegations after reboot recovery is done
+ * @clp: nfs_client to process
+ *
+ */
+void nfs_delegation_reap_unclaimed(struct nfs_client *clp)
+{
+	nfs_client_for_each_server(clp, nfs_server_reap_unclaimed_delegations,
+			NULL);
 }
 
 static inline bool nfs4_server_rebooted(const struct nfs_client *clp)
@@ -1072,62 +1222,61 @@ nfs_delegation_test_free_expired(struct inode *inode,
 		nfs_remove_bad_delegation(inode, stateid);
 }
 
-/**
- * nfs_reap_expired_delegations - reap expired delegations
- * @clp: nfs_client to process
- *
- * Iterates through all the delegations associated with this server and
- * checks if they have may have been revoked. This function is usually
- * expected to be called in cases where the server may have lost its
- * lease.
- */
-void nfs_reap_expired_delegations(struct nfs_client *clp)
+static int nfs_server_reap_expired_delegations(struct nfs_server *server,
+		void __always_unused *data)
 {
 	struct nfs_delegation *delegation;
-	struct nfs_server *server;
 	struct inode *inode;
 	const struct cred *cred;
 	nfs4_stateid stateid;
-
 restart:
 	rcu_read_lock();
-	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
-		list_for_each_entry_rcu(delegation, &server->delegations,
-								super_list) {
-			if (test_bit(NFS_DELEGATION_INODE_FREEING,
-						&delegation->flags) ||
-			    test_bit(NFS_DELEGATION_RETURNING,
-						&delegation->flags) ||
-			    test_bit(NFS_DELEGATION_TEST_EXPIRED,
-						&delegation->flags) == 0)
-				continue;
-			if (!nfs_sb_active(server->super))
-				break; /* continue in outer loop */
-			inode = nfs_delegation_grab_inode(delegation);
-			if (inode == NULL) {
-				rcu_read_unlock();
-				nfs_sb_deactive(server->super);
-				goto restart;
-			}
-			cred = get_cred_rcu(delegation->cred);
-			nfs4_stateid_copy(&stateid, &delegation->stateid);
-			clear_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags);
-			rcu_read_unlock();
-			nfs_delegation_test_free_expired(inode, &stateid, cred);
-			put_cred(cred);
-			if (nfs4_server_rebooted(clp)) {
-				nfs_inode_mark_test_expired_delegation(server,inode);
-				iput(inode);
-				nfs_sb_deactive(server->super);
-				return;
-			}
+restart_locked:
+	list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
+		if (test_bit(NFS_DELEGATION_INODE_FREEING,
+					&delegation->flags) ||
+		    test_bit(NFS_DELEGATION_RETURNING,
+					&delegation->flags) ||
+		    test_bit(NFS_DELEGATION_TEST_EXPIRED,
+					&delegation->flags) == 0)
+			continue;
+		inode = nfs_delegation_grab_inode(delegation);
+		if (inode == NULL)
+			goto restart_locked;
+		spin_lock(&delegation->lock);
+		cred = get_cred_rcu(delegation->cred);
+		nfs4_stateid_copy(&stateid, &delegation->stateid);
+		spin_unlock(&delegation->lock);
+		clear_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags);
+		rcu_read_unlock();
+		nfs_delegation_test_free_expired(inode, &stateid, cred);
+		put_cred(cred);
+		if (!nfs4_server_rebooted(server->nfs_client)) {
 			iput(inode);
-			nfs_sb_deactive(server->super);
 			cond_resched();
 			goto restart;
 		}
+		nfs_inode_mark_test_expired_delegation(server,inode);
+		iput(inode);
+		return -EAGAIN;
 	}
 	rcu_read_unlock();
+	return 0;
+}
+
+/**
+ * nfs_reap_expired_delegations - reap expired delegations
+ * @clp: nfs_client to process
+ *
+ * Iterates through all the delegations associated with this server and
+ * checks if they have may have been revoked. This function is usually
+ * expected to be called in cases where the server may have lost its
+ * lease.
+ */
+void nfs_reap_expired_delegations(struct nfs_client *clp)
+{
+	nfs_client_for_each_server(clp, nfs_server_reap_expired_delegations,
+			NULL);
 }
 
 void nfs_inode_find_delegation_state_and_recover(struct inode *inode,
@@ -1140,7 +1289,8 @@ void nfs_inode_find_delegation_state_and_recover(struct inode *inode,
 	rcu_read_lock();
 	delegation = rcu_dereference(NFS_I(inode)->delegation);
 	if (delegation &&
-	    nfs4_stateid_match_other(&delegation->stateid, stateid)) {
+	    nfs4_stateid_match_or_older(&delegation->stateid, stateid) &&
+	    !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) {
 		nfs_mark_test_expired_delegation(NFS_SERVER(inode), delegation);
 		found = true;
 	}
@@ -1189,7 +1339,9 @@ bool nfs4_refresh_delegation_stateid(nfs4_stateid *dst, struct inode *inode)
 	rcu_read_lock();
 	delegation = rcu_dereference(NFS_I(inode)->delegation);
 	if (delegation != NULL &&
-	    nfs4_stateid_match_other(dst, &delegation->stateid)) {
+	    nfs4_stateid_match_other(dst, &delegation->stateid) &&
+	    nfs4_stateid_is_newer(&delegation->stateid, dst) &&
+	    !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) {
 		dst->seqid = delegation->stateid.seqid;
 		ret = true;
 	}
@@ -1213,11 +1365,14 @@ bool nfs4_copy_delegation_stateid(struct inode *inode, fmode_t flags,
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
-	bool ret;
+	bool ret = false;
 
 	flags &= FMODE_READ|FMODE_WRITE;
 	rcu_read_lock();
 	delegation = rcu_dereference(nfsi->delegation);
+	if (!delegation)
+		goto out;
+	spin_lock(&delegation->lock);
 	ret = nfs4_is_valid_delegation(delegation, flags);
 	if (ret) {
 		nfs4_stateid_copy(dst, &delegation->stateid);
@@ -1225,6 +1380,8 @@ bool nfs4_copy_delegation_stateid(struct inode *inode, fmode_t flags,
 		if (cred)
 			*cred = get_cred(delegation->cred);
 	}
+	spin_unlock(&delegation->lock);
+out:
 	rcu_read_unlock();
 	return ret;
 }
@@ -1253,3 +1410,5 @@ out:
 	rcu_read_unlock();
 	return ret;
 }
+
+module_param_named(delegation_watermark, nfs_delegation_watermark, uint, 0644);
diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h
index 8b14d441e699..9b00a0b7f832 100644
--- a/fs/nfs/delegation.h
+++ b/fs/nfs/delegation.h
@@ -22,6 +22,7 @@ struct nfs_delegation {
 	unsigned long pagemod_limit;
 	__u64 change_attr;
 	unsigned long flags;
+	refcount_t refcount;
 	spinlock_t lock;
 	struct rcu_head rcu;
 };
@@ -42,8 +43,9 @@ int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred,
 void nfs_inode_reclaim_delegation(struct inode *inode, const struct cred *cred,
 		fmode_t type, const nfs4_stateid *stateid, unsigned long pagemod_limit);
 int nfs4_inode_return_delegation(struct inode *inode);
+void nfs4_inode_return_delegation_on_close(struct inode *inode);
 int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid);
-void nfs_inode_return_delegation_noreclaim(struct inode *inode);
+void nfs_inode_evict_delegation(struct inode *inode);
 
 struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle);
 void nfs_server_return_all_delegations(struct nfs_server *);
@@ -53,6 +55,7 @@ void nfs_expire_unreferenced_delegations(struct nfs_client *clp);
 int nfs_client_return_marked_delegations(struct nfs_client *clp);
 int nfs_delegations_present(struct nfs_client *clp);
 void nfs_remove_bad_delegation(struct inode *inode, const nfs4_stateid *stateid);
+void nfs_delegation_mark_returned(struct inode *inode, const nfs4_stateid *stateid);
 
 void nfs_delegation_mark_reclaim(struct nfs_client *clp);
 void nfs_delegation_reap_unclaimed(struct nfs_client *clp);
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index e180033e35cf..5a331da5f55a 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -58,7 +58,7 @@ static void nfs_readdir_clear_array(struct page*);
 const struct file_operations nfs_dir_operations = {
 	.llseek		= nfs_llseek_dir,
 	.read		= generic_read_dir,
-	.iterate	= nfs_readdir,
+	.iterate_shared	= nfs_readdir,
 	.open		= nfs_opendir,
 	.release	= nfs_closedir,
 	.fsync		= nfs_fsync_dir,
@@ -141,10 +141,9 @@ struct nfs_cache_array {
 	int size;
 	int eof_index;
 	u64 last_cookie;
-	struct nfs_cache_array_entry array[0];
+	struct nfs_cache_array_entry array[];
 };
 
-typedef int (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, bool);
 typedef struct {
 	struct file	*file;
 	struct page	*page;
@@ -153,8 +152,9 @@ typedef struct {
 	u64		*dir_cookie;
 	u64		last_cookie;
 	loff_t		current_index;
-	decode_dirent_t	decode;
+	loff_t		prev_index;
 
+	unsigned long	dir_verifier;
 	unsigned long	timestamp;
 	unsigned long	gencount;
 	unsigned int	cache_entry_index;
@@ -162,6 +162,17 @@ typedef struct {
 	bool eof;
 } nfs_readdir_descriptor_t;
 
+static
+void nfs_readdir_init_array(struct page *page)
+{
+	struct nfs_cache_array *array;
+
+	array = kmap_atomic(page);
+	memset(array, 0, sizeof(struct nfs_cache_array));
+	array->eof_index = -1;
+	kunmap_atomic(array);
+}
+
 /*
  * we are freeing strings created by nfs_add_to_readdir_array()
  */
@@ -174,6 +185,7 @@ void nfs_readdir_clear_array(struct page *page)
 	array = kmap_atomic(page);
 	for (i = 0; i < array->size; i++)
 		kfree(array->array[i].string.name);
+	array->size = 0;
 	kunmap_atomic(array);
 }
 
@@ -186,7 +198,7 @@ static
 int nfs_readdir_make_qstr(struct qstr *string, const char *name, unsigned int len)
 {
 	string->len = len;
-	string->name = kmemdup(name, len, GFP_KERNEL);
+	string->name = kmemdup_nul(name, len, GFP_KERNEL);
 	if (string->name == NULL)
 		return -ENOMEM;
 	/*
@@ -227,6 +239,25 @@ out:
 	return ret;
 }
 
+static inline
+int is_32bit_api(void)
+{
+#ifdef CONFIG_COMPAT
+	return in_compat_syscall();
+#else
+	return (BITS_PER_LONG == 32);
+#endif
+}
+
+static
+bool nfs_readdir_use_cookie(const struct file *filp)
+{
+	if ((filp->f_mode & FMODE_32BITHASH) ||
+	    (!(filp->f_mode & FMODE_64BITHASH) && is_32bit_api()))
+		return false;
+	return true;
+}
+
 static
 int nfs_readdir_search_for_pos(struct nfs_cache_array *array, nfs_readdir_descriptor_t *desc)
 {
@@ -276,7 +307,7 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des
 			    !nfs_readdir_inode_mapping_valid(nfsi)) {
 				ctx->duped = 0;
 				ctx->attr_gencount = nfsi->attr_gencount;
-			} else if (new_pos < desc->ctx->pos) {
+			} else if (new_pos < desc->prev_index) {
 				if (ctx->duped > 0
 				    && ctx->dup_cookie == *desc->dir_cookie) {
 					if (printk_ratelimit()) {
@@ -292,7 +323,11 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des
 				ctx->dup_cookie = *desc->dir_cookie;
 				ctx->duped = -1;
 			}
-			desc->ctx->pos = new_pos;
+			if (nfs_readdir_use_cookie(desc->file))
+				desc->ctx->pos = *desc->dir_cookie;
+			else
+				desc->ctx->pos = new_pos;
+			desc->prev_index = new_pos;
 			desc->cache_entry_index = i;
 			return 0;
 		}
@@ -341,6 +376,7 @@ int nfs_readdir_xdr_filler(struct page **pages, nfs_readdir_descriptor_t *desc,
  again:
 	timestamp = jiffies;
 	gencount = nfs_inc_attr_generation_counter();
+	desc->dir_verifier = nfs_save_change_attribute(inode);
 	error = NFS_PROTO(inode)->readdir(file_dentry(file), cred, entry->cookie, pages,
 					  NFS_SERVER(inode)->dtsize, desc->plus);
 	if (error < 0) {
@@ -362,9 +398,10 @@ error:
 static int xdr_decode(nfs_readdir_descriptor_t *desc,
 		      struct nfs_entry *entry, struct xdr_stream *xdr)
 {
+	struct inode *inode = file_inode(desc->file);
 	int error;
 
-	error = desc->decode(xdr, entry, desc->plus);
+	error = NFS_PROTO(inode)->decode_dirent(xdr, entry, desc->plus);
 	if (error)
 		return error;
 	entry->fattr->time_start = desc->timestamp;
@@ -437,18 +474,19 @@ void nfs_force_use_readdirplus(struct inode *dir)
 	if (nfs_server_capable(dir, NFS_CAP_READDIRPLUS) &&
 	    !list_empty(&nfsi->open_files)) {
 		set_bit(NFS_INO_ADVISE_RDPLUS, &nfsi->flags);
-		invalidate_mapping_pages(dir->i_mapping, 0, -1);
+		invalidate_mapping_pages(dir->i_mapping,
+			nfsi->page_index + 1, -1);
 	}
 }
 
 static
-void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry)
+void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry,
+		unsigned long dir_verifier)
 {
 	struct qstr filename = QSTR_INIT(entry->name, entry->len);
 	DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
 	struct dentry *dentry;
 	struct dentry *alias;
-	struct inode *dir = d_inode(parent);
 	struct inode *inode;
 	int status;
 
@@ -487,7 +525,7 @@ again:
 		if (nfs_same_file(dentry, entry)) {
 			if (!entry->fh->size)
 				goto out;
-			nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
+			nfs_set_verifier(dentry, dir_verifier);
 			status = nfs_refresh_inode(d_inode(dentry), entry->fattr);
 			if (!status)
 				nfs_setsecurity(d_inode(dentry), entry->fattr, entry->label);
@@ -513,7 +551,7 @@ again:
 		dput(dentry);
 		dentry = alias;
 	}
-	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
+	nfs_set_verifier(dentry, dir_verifier);
 out:
 	dput(dentry);
 }
@@ -551,7 +589,8 @@ int nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *en
 		count++;
 
 		if (desc->plus)
-			nfs_prime_dcache(file_dentry(desc->file), entry);
+			nfs_prime_dcache(file_dentry(desc->file), entry,
+					desc->dir_verifier);
 
 		status = nfs_readdir_add_to_array(entry, page);
 		if (status != 0)
@@ -610,6 +649,8 @@ int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page,
 	int status = -ENOMEM;
 	unsigned int array_size = ARRAY_SIZE(pages);
 
+	nfs_readdir_init_array(page);
+
 	entry.prev_cookie = 0;
 	entry.cookie = desc->last_cookie;
 	entry.eof = 0;
@@ -626,8 +667,6 @@ int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page,
 	}
 
 	array = kmap(page);
-	memset(array, 0, sizeof(struct nfs_cache_array));
-	array->eof_index = -1;
 
 	status = nfs_readdir_alloc_pages(pages, array_size);
 	if (status < 0)
@@ -682,6 +721,7 @@ int nfs_readdir_filler(void *data, struct page* page)
 	unlock_page(page);
 	return 0;
  error:
+	nfs_readdir_clear_array(page);
 	unlock_page(page);
 	return ret;
 }
@@ -689,8 +729,6 @@ int nfs_readdir_filler(void *data, struct page* page)
 static
 void cache_page_release(nfs_readdir_descriptor_t *desc)
 {
-	if (!desc->page->mapping)
-		nfs_readdir_clear_array(desc->page);
 	put_page(desc->page);
 	desc->page = NULL;
 }
@@ -704,19 +742,32 @@ struct page *get_cache_page(nfs_readdir_descriptor_t *desc)
 
 /*
  * Returns 0 if desc->dir_cookie was found on page desc->page_index
+ * and locks the page to prevent removal from the page cache.
  */
 static
-int find_cache_page(nfs_readdir_descriptor_t *desc)
+int find_and_lock_cache_page(nfs_readdir_descriptor_t *desc)
 {
+	struct inode *inode = file_inode(desc->file);
+	struct nfs_inode *nfsi = NFS_I(inode);
 	int res;
 
 	desc->page = get_cache_page(desc);
 	if (IS_ERR(desc->page))
 		return PTR_ERR(desc->page);
-
-	res = nfs_readdir_search_array(desc);
+	res = lock_page_killable(desc->page);
 	if (res != 0)
-		cache_page_release(desc);
+		goto error;
+	res = -EAGAIN;
+	if (desc->page->mapping != NULL) {
+		res = nfs_readdir_search_array(desc);
+		if (res == 0) {
+			nfsi->page_index = desc->page_index;
+			return 0;
+		}
+	}
+	unlock_page(desc->page);
+error:
+	cache_page_release(desc);
 	return res;
 }
 
@@ -728,10 +779,11 @@ int readdir_search_pagecache(nfs_readdir_descriptor_t *desc)
 
 	if (desc->page_index == 0) {
 		desc->current_index = 0;
+		desc->prev_index = 0;
 		desc->last_cookie = 0;
 	}
 	do {
-		res = find_cache_page(desc);
+		res = find_and_lock_cache_page(desc);
 	} while (res == -EAGAIN);
 	return res;
 }
@@ -758,11 +810,14 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc)
 			desc->eof = true;
 			break;
 		}
-		desc->ctx->pos++;
 		if (i < (array->size-1))
 			*desc->dir_cookie = array->array[i+1].cookie;
 		else
 			*desc->dir_cookie = array->last_cookie;
+		if (nfs_readdir_use_cookie(file))
+			desc->ctx->pos = *desc->dir_cookie;
+		else
+			desc->ctx->pos++;
 		if (ctx->duped != 0)
 			ctx->duped = 1;
 	}
@@ -770,7 +825,6 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc)
 		desc->eof = true;
 
 	kunmap(desc->page);
-	cache_page_release(desc);
 	dfprintk(DIRCACHE, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n",
 			(unsigned long long)*desc->dir_cookie, res);
 	return res;
@@ -816,13 +870,13 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc)
 
 	status = nfs_do_filldir(desc);
 
+ out_release:
+	nfs_readdir_clear_array(desc->page);
+	cache_page_release(desc);
  out:
 	dfprintk(DIRCACHE, "NFS: %s: returns %d\n",
 			__func__, status);
 	return status;
- out_release:
-	cache_page_release(desc);
-	goto out;
 }
 
 /* The file offset position represents the dirent entry number.  A
@@ -833,9 +887,14 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
 {
 	struct dentry	*dentry = file_dentry(file);
 	struct inode	*inode = d_inode(dentry);
-	nfs_readdir_descriptor_t my_desc,
-			*desc = &my_desc;
 	struct nfs_open_dir_context *dir_ctx = file->private_data;
+	nfs_readdir_descriptor_t my_desc = {
+		.file = file,
+		.ctx = ctx,
+		.dir_cookie = &dir_ctx->dir_cookie,
+		.plus = nfs_use_readdirplus(inode, ctx),
+	},
+			*desc = &my_desc;
 	int res = 0;
 
 	dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n",
@@ -848,14 +907,6 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
 	 * to either find the entry with the appropriate number or
 	 * revalidate the cookie.
 	 */
-	memset(desc, 0, sizeof(*desc));
-
-	desc->file = file;
-	desc->ctx = ctx;
-	desc->dir_cookie = &dir_ctx->dir_cookie;
-	desc->decode = NFS_PROTO(inode)->decode_dirent;
-	desc->plus = nfs_use_readdirplus(inode, ctx);
-
 	if (ctx->pos == 0 || nfs_attribute_cache_expired(inode))
 		res = nfs_revalidate_mapping(inode, file->f_mapping);
 	if (res < 0)
@@ -887,6 +938,8 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
 			break;
 
 		res = nfs_do_filldir(desc);
+		unlock_page(desc->page);
+		cache_page_release(desc);
 		if (res < 0)
 			break;
 	} while (!desc->eof);
@@ -925,7 +978,10 @@ static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int whence)
 	}
 	if (offset != filp->f_pos) {
 		filp->f_pos = offset;
-		dir_ctx->dir_cookie = 0;
+		if (nfs_readdir_use_cookie(filp))
+			dir_ctx->dir_cookie = offset;
+		else
+			dir_ctx->dir_cookie = 0;
 		dir_ctx->duped = 0;
 	}
 	inode_unlock(inode);
@@ -957,14 +1013,113 @@ static int nfs_fsync_dir(struct file *filp, loff_t start, loff_t end,
  * full lookup on all child dentries of 'dir' whenever a change occurs
  * on the server that might have invalidated our dcache.
  *
+ * Note that we reserve bit '0' as a tag to let us know when a dentry
+ * was revalidated while holding a delegation on its inode.
+ *
  * The caller should be holding dir->i_lock
  */
 void nfs_force_lookup_revalidate(struct inode *dir)
 {
-	NFS_I(dir)->cache_change_attribute++;
+	NFS_I(dir)->cache_change_attribute += 2;
 }
 EXPORT_SYMBOL_GPL(nfs_force_lookup_revalidate);
 
+/**
+ * nfs_verify_change_attribute - Detects NFS remote directory changes
+ * @dir: pointer to parent directory inode
+ * @verf: previously saved change attribute
+ *
+ * Return "false" if the verifiers doesn't match the change attribute.
+ * This would usually indicate that the directory contents have changed on
+ * the server, and that any dentries need revalidating.
+ */
+static bool nfs_verify_change_attribute(struct inode *dir, unsigned long verf)
+{
+	return (verf & ~1UL) == nfs_save_change_attribute(dir);
+}
+
+static void nfs_set_verifier_delegated(unsigned long *verf)
+{
+	*verf |= 1UL;
+}
+
+#if IS_ENABLED(CONFIG_NFS_V4)
+static void nfs_unset_verifier_delegated(unsigned long *verf)
+{
+	*verf &= ~1UL;
+}
+#endif /* IS_ENABLED(CONFIG_NFS_V4) */
+
+static bool nfs_test_verifier_delegated(unsigned long verf)
+{
+	return verf & 1;
+}
+
+static bool nfs_verifier_is_delegated(struct dentry *dentry)
+{
+	return nfs_test_verifier_delegated(dentry->d_time);
+}
+
+static void nfs_set_verifier_locked(struct dentry *dentry, unsigned long verf)
+{
+	struct inode *inode = d_inode(dentry);
+
+	if (!nfs_verifier_is_delegated(dentry) &&
+	    !nfs_verify_change_attribute(d_inode(dentry->d_parent), verf))
+		goto out;
+	if (inode && NFS_PROTO(inode)->have_delegation(inode, FMODE_READ))
+		nfs_set_verifier_delegated(&verf);
+out:
+	dentry->d_time = verf;
+}
+
+/**
+ * nfs_set_verifier - save a parent directory verifier in the dentry
+ * @dentry: pointer to dentry
+ * @verf: verifier to save
+ *
+ * Saves the parent directory verifier in @dentry. If the inode has
+ * a delegation, we also tag the dentry as having been revalidated
+ * while holding a delegation so that we know we don't have to
+ * look it up again after a directory change.
+ */
+void nfs_set_verifier(struct dentry *dentry, unsigned long verf)
+{
+
+	spin_lock(&dentry->d_lock);
+	nfs_set_verifier_locked(dentry, verf);
+	spin_unlock(&dentry->d_lock);
+}
+EXPORT_SYMBOL_GPL(nfs_set_verifier);
+
+#if IS_ENABLED(CONFIG_NFS_V4)
+/**
+ * nfs_clear_verifier_delegated - clear the dir verifier delegation tag
+ * @inode: pointer to inode
+ *
+ * Iterates through the dentries in the inode alias list and clears
+ * the tag used to indicate that the dentry has been revalidated
+ * while holding a delegation.
+ * This function is intended for use when the delegation is being
+ * returned or revoked.
+ */
+void nfs_clear_verifier_delegated(struct inode *inode)
+{
+	struct dentry *alias;
+
+	if (!inode)
+		return;
+	spin_lock(&inode->i_lock);
+	hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) {
+		spin_lock(&alias->d_lock);
+		nfs_unset_verifier_delegated(&alias->d_time);
+		spin_unlock(&alias->d_lock);
+	}
+	spin_unlock(&inode->i_lock);
+}
+EXPORT_SYMBOL_GPL(nfs_clear_verifier_delegated);
+#endif /* IS_ENABLED(CONFIG_NFS_V4) */
+
 /*
  * A check for whether or not the parent directory has changed.
  * In the case it has, we assume that the dentries are untrustworthy
@@ -1133,6 +1288,7 @@ nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry,
 	struct nfs_fh *fhandle;
 	struct nfs_fattr *fattr;
 	struct nfs4_label *label;
+	unsigned long dir_verifier;
 	int ret;
 
 	ret = -ENOMEM;
@@ -1142,10 +1298,18 @@ nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry,
 	if (fhandle == NULL || fattr == NULL || IS_ERR(label))
 		goto out;
 
-	ret = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, label);
+	dir_verifier = nfs_save_change_attribute(dir);
+	ret = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr, label);
 	if (ret < 0) {
-		if (ret == -ESTALE || ret == -ENOENT)
+		switch (ret) {
+		case -ESTALE:
+		case -ENOENT:
 			ret = 0;
+			break;
+		case -ETIMEDOUT:
+			if (NFS_SERVER(inode)->flags & NFS_MOUNT_SOFTREVAL)
+				ret = 1;
+		}
 		goto out;
 	}
 	ret = 0;
@@ -1155,7 +1319,7 @@ nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry,
 		goto out;
 
 	nfs_setsecurity(inode, fattr, label);
-	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
+	nfs_set_verifier(dentry, dir_verifier);
 
 	/* set a readdirplus hint that we had a cache miss */
 	nfs_force_use_readdirplus(dir);
@@ -1197,7 +1361,7 @@ nfs_do_lookup_revalidate(struct inode *dir, struct dentry *dentry,
 		goto out_bad;
 	}
 
-	if (NFS_PROTO(dir)->have_delegation(inode, FMODE_READ))
+	if (nfs_verifier_is_delegated(dentry))
 		return nfs_lookup_revalidate_delegated(dir, dentry, inode);
 
 	/* Force a full look up iff the parent directory has changed */
@@ -1382,6 +1546,7 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in
 	struct nfs_fh *fhandle = NULL;
 	struct nfs_fattr *fattr = NULL;
 	struct nfs4_label *label = NULL;
+	unsigned long dir_verifier;
 	int error;
 
 	dfprintk(VFS, "NFS: lookup(%pd2)\n", dentry);
@@ -1407,8 +1572,9 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in
 	if (IS_ERR(label))
 		goto out;
 
+	dir_verifier = nfs_save_change_attribute(dir);
 	trace_nfs_lookup_enter(dir, dentry, flags);
-	error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, label);
+	error = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr, label);
 	if (error == -ENOENT)
 		goto no_entry;
 	if (error < 0) {
@@ -1430,7 +1596,7 @@ no_entry:
 			goto out_label;
 		dentry = res;
 	}
-	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
+	nfs_set_verifier(dentry, dir_verifier);
 out_label:
 	trace_nfs_lookup_exit(dir, dentry, flags, error);
 	nfs4_label_free(label);
@@ -1635,7 +1801,7 @@ nfs4_do_lookup_revalidate(struct inode *dir, struct dentry *dentry,
 	if (inode == NULL)
 		goto full_reval;
 
-	if (NFS_PROTO(dir)->have_delegation(inode, FMODE_READ))
+	if (nfs_verifier_is_delegated(dentry))
 		return nfs_lookup_revalidate_delegated(dir, dentry, inode);
 
 	/* NFS only supports OPEN on regular files */
@@ -1683,7 +1849,7 @@ nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle,
 	d_drop(dentry);
 
 	if (fhandle->size == 0) {
-		error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, NULL);
+		error = NFS_PROTO(dir)->lookup(dir, dentry, fhandle, fattr, NULL);
 		if (error)
 			goto out_error;
 	}
@@ -2143,7 +2309,7 @@ static DEFINE_SPINLOCK(nfs_access_lru_lock);
 static LIST_HEAD(nfs_access_lru_list);
 static atomic_long_t nfs_access_nr_entries;
 
-static unsigned long nfs_access_max_cachesize = ULONG_MAX;
+static unsigned long nfs_access_max_cachesize = 4*1024*1024;
 module_param(nfs_access_max_cachesize, ulong, 0644);
 MODULE_PARM_DESC(nfs_access_max_cachesize, "NFS access maximum total cache length");
 
@@ -2312,11 +2478,11 @@ static int nfs_access_get_cached(struct inode *inode, const struct cred *cred, s
 		/* Found an entry, is our attribute cache valid? */
 		if (!nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS))
 			break;
+		if (!retry)
+			break;
 		err = -ECHILD;
 		if (!may_block)
 			goto out;
-		if (!retry)
-			goto out_zap;
 		spin_unlock(&inode->i_lock);
 		err = __nfs_revalidate_inode(NFS_SERVER(inode), inode);
 		if (err)
@@ -2350,10 +2516,10 @@ static int nfs_access_get_cached_rcu(struct inode *inode, const struct cred *cre
 	rcu_read_lock();
 	if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS)
 		goto out;
-	lh = rcu_dereference(nfsi->access_cache_entry_lru.prev);
+	lh = rcu_dereference(list_tail_rcu(&nfsi->access_cache_entry_lru));
 	cache = list_entry(lh, struct nfs_access_entry, lru);
 	if (lh == &nfsi->access_cache_entry_lru ||
-	    cred != cache->cred)
+	    cred_fscmp(cred, cache->cred) != 0)
 		cache = NULL;
 	if (cache == NULL)
 		goto out;
@@ -2476,7 +2642,7 @@ static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask)
 {
 	struct nfs_access_entry cache;
 	bool may_block = (mask & MAY_NOT_BLOCK) == 0;
-	int cache_mask;
+	int cache_mask = -1;
 	int status;
 
 	trace_nfs_access_enter(inode);
@@ -2503,9 +2669,10 @@ static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask)
 	status = NFS_PROTO(inode)->access(inode, &cache);
 	if (status != 0) {
 		if (status == -ESTALE) {
-			nfs_zap_caches(inode);
 			if (!S_ISDIR(inode->i_mode))
-				set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
+				nfs_set_inode_stale(inode);
+			else
+				nfs_zap_caches(inode);
 		}
 		goto out;
 	}
@@ -2515,7 +2682,7 @@ out_cached:
 	if ((mask & ~cache_mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) != 0)
 		status = -EACCES;
 out:
-	trace_nfs_access_exit(inode, status);
+	trace_nfs_access_exit(inode, mask, cache_mask, status);
 	return status;
 }
 
@@ -2593,14 +2760,7 @@ force_lookup:
 	if (!NFS_PROTO(inode)->access)
 		goto out_notsup;
 
-	/* Always try fast lookups first */
-	rcu_read_lock();
-	res = nfs_do_access(inode, cred, mask|MAY_NOT_BLOCK);
-	rcu_read_unlock();
-	if (res == -ECHILD && !(mask & MAY_NOT_BLOCK)) {
-		/* Fast lookup failed, try the slow way */
-		res = nfs_do_access(inode, cred, mask);
-	}
+	res = nfs_do_access(inode, cred, mask);
 out:
 	if (!res && (mask & MAY_EXEC))
 		res = nfs_execute_ok(inode, mask);
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 040a50fd9bf3..a57e7c72c7f4 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -94,7 +94,7 @@ struct nfs_direct_req {
 #define NFS_ODIRECT_RESCHED_WRITES	(2)	/* write verification failed */
 	/* for read */
 #define NFS_ODIRECT_SHOULD_DIRTY	(3)	/* dirty user-space page after read */
-	struct nfs_writeverf	verf;		/* unstable write verifier */
+#define NFS_ODIRECT_DONE		INT_MAX	/* write verification failed */
 };
 
 static const struct nfs_pgio_completion_ops nfs_direct_write_completion_ops;
@@ -151,106 +151,6 @@ nfs_direct_count_bytes(struct nfs_direct_req *dreq,
 		dreq->count = dreq_len;
 }
 
-/*
- * nfs_direct_select_verf - select the right verifier
- * @dreq - direct request possibly spanning multiple servers
- * @ds_clp - nfs_client of data server or NULL if MDS / non-pnfs
- * @commit_idx - commit bucket index for the DS
- *
- * returns the correct verifier to use given the role of the server
- */
-static struct nfs_writeverf *
-nfs_direct_select_verf(struct nfs_direct_req *dreq,
-		       struct nfs_client *ds_clp,
-		       int commit_idx)
-{
-	struct nfs_writeverf *verfp = &dreq->verf;
-
-#ifdef CONFIG_NFS_V4_1
-	/*
-	 * pNFS is in use, use the DS verf except commit_through_mds is set
-	 * for layout segment where nbuckets is zero.
-	 */
-	if (ds_clp && dreq->ds_cinfo.nbuckets > 0) {
-		if (commit_idx >= 0 && commit_idx < dreq->ds_cinfo.nbuckets)
-			verfp = &dreq->ds_cinfo.buckets[commit_idx].direct_verf;
-		else
-			WARN_ON_ONCE(1);
-	}
-#endif
-	return verfp;
-}
-
-
-/*
- * nfs_direct_set_hdr_verf - set the write/commit verifier
- * @dreq - direct request possibly spanning multiple servers
- * @hdr - pageio header to validate against previously seen verfs
- *
- * Set the server's (MDS or DS) "seen" verifier
- */
-static void nfs_direct_set_hdr_verf(struct nfs_direct_req *dreq,
-				    struct nfs_pgio_header *hdr)
-{
-	struct nfs_writeverf *verfp;
-
-	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp, hdr->ds_commit_idx);
-	WARN_ON_ONCE(verfp->committed >= 0);
-	memcpy(verfp, &hdr->verf, sizeof(struct nfs_writeverf));
-	WARN_ON_ONCE(verfp->committed < 0);
-}
-
-static int nfs_direct_cmp_verf(const struct nfs_writeverf *v1,
-		const struct nfs_writeverf *v2)
-{
-	return nfs_write_verifier_cmp(&v1->verifier, &v2->verifier);
-}
-
-/*
- * nfs_direct_cmp_hdr_verf - compare verifier for pgio header
- * @dreq - direct request possibly spanning multiple servers
- * @hdr - pageio header to validate against previously seen verf
- *
- * set the server's "seen" verf if not initialized.
- * returns result of comparison between @hdr->verf and the "seen"
- * verf of the server used by @hdr (DS or MDS)
- */
-static int nfs_direct_set_or_cmp_hdr_verf(struct nfs_direct_req *dreq,
-					  struct nfs_pgio_header *hdr)
-{
-	struct nfs_writeverf *verfp;
-
-	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp, hdr->ds_commit_idx);
-	if (verfp->committed < 0) {
-		nfs_direct_set_hdr_verf(dreq, hdr);
-		return 0;
-	}
-	return nfs_direct_cmp_verf(verfp, &hdr->verf);
-}
-
-/*
- * nfs_direct_cmp_commit_data_verf - compare verifier for commit data
- * @dreq - direct request possibly spanning multiple servers
- * @data - commit data to validate against previously seen verf
- *
- * returns result of comparison between @data->verf and the verf of
- * the server used by @data (DS or MDS)
- */
-static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq,
-					   struct nfs_commit_data *data)
-{
-	struct nfs_writeverf *verfp;
-
-	verfp = nfs_direct_select_verf(dreq, data->ds_clp,
-					 data->ds_commit_index);
-
-	/* verifier not set so always fail */
-	if (verfp->committed < 0)
-		return 1;
-
-	return nfs_direct_cmp_verf(verfp, &data->verf);
-}
-
 /**
  * nfs_direct_IO - NFS address space operation for direct I/O
  * @iocb: target I/O control block
@@ -305,7 +205,7 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void)
 	kref_get(&dreq->kref);
 	init_completion(&dreq->completion);
 	INIT_LIST_HEAD(&dreq->mds_cinfo.list);
-	dreq->verf.committed = NFS_INVALID_STABLE_HOW;	/* not set yet */
+	pnfs_init_ds_commit_info(&dreq->ds_cinfo);
 	INIT_WORK(&dreq->work, nfs_direct_write_schedule_work);
 	spin_lock_init(&dreq->lock);
 
@@ -316,7 +216,7 @@ static void nfs_direct_req_free(struct kref *kref)
 {
 	struct nfs_direct_req *dreq = container_of(kref, struct nfs_direct_req, kref);
 
-	nfs_free_pnfs_ds_cinfo(&dreq->ds_cinfo);
+	pnfs_release_ds_info(&dreq->ds_cinfo, dreq->inode);
 	if (dreq->l_ctx != NULL)
 		nfs_put_lock_context(dreq->l_ctx);
 	if (dreq->ctx != NULL)
@@ -571,6 +471,7 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter)
 	l_ctx = nfs_get_lock_context(dreq->ctx);
 	if (IS_ERR(l_ctx)) {
 		result = PTR_ERR(l_ctx);
+		nfs_direct_req_release(dreq);
 		goto out_release;
 	}
 	dreq->l_ctx = l_ctx;
@@ -605,15 +506,30 @@ out:
 }
 
 static void
+nfs_direct_join_group(struct list_head *list, struct inode *inode)
+{
+	struct nfs_page *req, *next;
+
+	list_for_each_entry(req, list, wb_list) {
+		if (req->wb_head != req || req->wb_this_page == req)
+			continue;
+		for (next = req->wb_this_page;
+				next != req->wb_head;
+				next = next->wb_this_page) {
+			nfs_list_remove_request(next);
+			nfs_release_request(next);
+		}
+		nfs_join_page_group(req, inode);
+	}
+}
+
+static void
 nfs_direct_write_scan_commit_list(struct inode *inode,
 				  struct list_head *list,
 				  struct nfs_commit_info *cinfo)
 {
 	mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
-#ifdef CONFIG_NFS_V4_1
-	if (cinfo->ds != NULL && cinfo->ds->nwritten != 0)
-		NFS_SERVER(inode)->pnfs_curr_ld->recover_commit_reqs(list, cinfo);
-#endif
+	pnfs_recover_commit_reqs(list, cinfo);
 	nfs_scan_commit_list(&cinfo->mds->list, list, cinfo, 0);
 	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
 }
@@ -629,11 +545,12 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 	nfs_init_cinfo_from_dreq(&cinfo, dreq);
 	nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo);
 
+	nfs_direct_join_group(&reqs, dreq->inode);
+
 	dreq->count = 0;
 	dreq->max_count = 0;
 	list_for_each_entry(req, &reqs, wb_list)
 		dreq->max_count += req->wb_bytes;
-	dreq->verf.committed = NFS_INVALID_STABLE_HOW;
 	nfs_clear_pnfs_ds_commit_verifiers(&dreq->ds_cinfo);
 	get_dreq(dreq);
 
@@ -670,27 +587,35 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 
 static void nfs_direct_commit_complete(struct nfs_commit_data *data)
 {
+	const struct nfs_writeverf *verf = data->res.verf;
 	struct nfs_direct_req *dreq = data->dreq;
 	struct nfs_commit_info cinfo;
 	struct nfs_page *req;
 	int status = data->task.tk_status;
 
+	if (status < 0) {
+		/* Errors in commit are fatal */
+		dreq->error = status;
+		dreq->max_count = 0;
+		dreq->count = 0;
+		dreq->flags = NFS_ODIRECT_DONE;
+	} else if (dreq->flags == NFS_ODIRECT_DONE)
+		status = dreq->error;
+
 	nfs_init_cinfo_from_dreq(&cinfo, dreq);
-	if (status < 0 || nfs_direct_cmp_commit_data_verf(dreq, data))
-		dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
 
 	while (!list_empty(&data->pages)) {
 		req = nfs_list_entry(data->pages.next);
 		nfs_list_remove_request(req);
-		if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES) {
+		if (status >= 0 && !nfs_write_match_verf(verf, req)) {
+			dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
 			/*
 			 * Despite the reboot, the write was successful,
 			 * so reset wb_nio.
 			 */
 			req->wb_nio = 0;
-			/* Note the rewrite will go through mds */
 			nfs_mark_request_commit(req, NULL, &cinfo, 0);
-		} else
+		} else /* Error or match */
 			nfs_release_request(req);
 		nfs_unlock_and_release_request(req);
 	}
@@ -705,7 +630,8 @@ static void nfs_direct_resched_write(struct nfs_commit_info *cinfo,
 	struct nfs_direct_req *dreq = cinfo->dreq;
 
 	spin_lock(&dreq->lock);
-	dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
+	if (dreq->flags != NFS_ODIRECT_DONE)
+		dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
 	spin_unlock(&dreq->lock);
 	nfs_mark_request_commit(req, NULL, cinfo, 0);
 }
@@ -728,6 +654,23 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 		nfs_direct_write_reschedule(dreq);
 }
 
+static void nfs_direct_write_clear_reqs(struct nfs_direct_req *dreq)
+{
+	struct nfs_commit_info cinfo;
+	struct nfs_page *req;
+	LIST_HEAD(reqs);
+
+	nfs_init_cinfo_from_dreq(&cinfo, dreq);
+	nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo);
+
+	while (!list_empty(&reqs)) {
+		req = nfs_list_entry(reqs.next);
+		nfs_list_remove_request(req);
+		nfs_release_request(req);
+		nfs_unlock_and_release_request(req);
+	}
+}
+
 static void nfs_direct_write_schedule_work(struct work_struct *work)
 {
 	struct nfs_direct_req *dreq = container_of(work, struct nfs_direct_req, work);
@@ -742,6 +685,7 @@ static void nfs_direct_write_schedule_work(struct work_struct *work)
 			nfs_direct_write_reschedule(dreq);
 			break;
 		default:
+			nfs_direct_write_clear_reqs(dreq);
 			nfs_zap_mapping(dreq->inode, dreq->inode->i_mapping);
 			nfs_direct_complete(dreq);
 	}
@@ -768,20 +712,15 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr)
 	}
 
 	nfs_direct_count_bytes(dreq, hdr);
-	if (hdr->good_bytes != 0) {
-		if (nfs_write_need_commit(hdr)) {
-			if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES)
-				request_commit = true;
-			else if (dreq->flags == 0) {
-				nfs_direct_set_hdr_verf(dreq, hdr);
-				request_commit = true;
-				dreq->flags = NFS_ODIRECT_DO_COMMIT;
-			} else if (dreq->flags == NFS_ODIRECT_DO_COMMIT) {
-				request_commit = true;
-				if (nfs_direct_set_or_cmp_hdr_verf(dreq, hdr))
-					dreq->flags =
-						NFS_ODIRECT_RESCHED_WRITES;
-			}
+	if (hdr->good_bytes != 0 && nfs_write_need_commit(hdr)) {
+		switch (dreq->flags) {
+		case 0:
+			dreq->flags = NFS_ODIRECT_DO_COMMIT;
+			request_commit = true;
+			break;
+		case NFS_ODIRECT_RESCHED_WRITES:
+		case NFS_ODIRECT_DO_COMMIT:
+			request_commit = true;
 		}
 	}
 	spin_unlock(&dreq->lock);
@@ -824,7 +763,8 @@ static void nfs_direct_write_reschedule_io(struct nfs_pgio_header *hdr)
 		dreq->flags = NFS_ODIRECT_RESCHED_WRITES;
 		/* fake unstable write to let common nfs resend pages */
 		hdr->verf.committed = NFS_UNSTABLE;
-		hdr->good_bytes = hdr->args.count;
+		hdr->good_bytes = hdr->args.offset + hdr->args.count -
+			hdr->io_start;
 	}
 	spin_unlock(&dreq->lock);
 }
@@ -989,11 +929,13 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
 	l_ctx = nfs_get_lock_context(dreq->ctx);
 	if (IS_ERR(l_ctx)) {
 		result = PTR_ERR(l_ctx);
+		nfs_direct_req_release(dreq);
 		goto out_release;
 	}
 	dreq->l_ctx = l_ctx;
 	if (!is_sync_kiocb(iocb))
 		dreq->iocb = iocb;
+	pnfs_init_ds_commit_info_ops(&dreq->ds_cinfo, inode);
 
 	nfs_start_io_direct(inode);
 
diff --git a/fs/nfs/dns_resolve.c b/fs/nfs/dns_resolve.c
index aec769a500a1..963800037609 100644
--- a/fs/nfs/dns_resolve.c
+++ b/fs/nfs/dns_resolve.c
@@ -93,7 +93,7 @@ static void nfs_dns_ent_init(struct cache_head *cnew,
 	key = container_of(ckey, struct nfs_dns_ent, h);
 
 	kfree(new->hostname);
-	new->hostname = kstrndup(key->hostname, key->namelen, GFP_KERNEL);
+	new->hostname = kmemdup_nul(key->hostname, key->namelen, GFP_KERNEL);
 	if (new->hostname) {
 		new->namelen = key->namelen;
 		nfs_dns_ent_update(cnew, ckey);
@@ -152,12 +152,13 @@ static int nfs_dns_upcall(struct cache_detail *cd,
 		struct cache_head *ch)
 {
 	struct nfs_dns_ent *key = container_of(ch, struct nfs_dns_ent, h);
-	int ret;
 
-	ret = nfs_cache_upcall(cd, key->hostname);
-	if (ret)
-		ret = sunrpc_cache_pipe_upcall(cd, ch);
-	return ret;
+	if (test_and_set_bit(CACHE_PENDING, &ch->flags))
+		return 0;
+	if (!nfs_cache_upcall(cd, key->hostname))
+		return 0;
+	clear_bit(CACHE_PENDING, &ch->flags);
+	return sunrpc_cache_pipe_upcall_timeout(cd, ch);
 }
 
 static int nfs_dns_match(struct cache_head *ca,
diff --git a/fs/nfs/export.c b/fs/nfs/export.c
index deecb67638aa..3430d6891e89 100644
--- a/fs/nfs/export.c
+++ b/fs/nfs/export.c
@@ -105,6 +105,7 @@ nfs_fh_to_dentry(struct super_block *sb, struct fid *fid,
 	ret = rpc_ops->getattr(NFS_SB(sb), server_fh, fattr, label, NULL);
 	if (ret) {
 		dprintk("%s: getattr failed %d\n", __func__, ret);
+		trace_nfs_fh_to_dentry(sb, server_fh, fattr->fileid, ret);
 		dentry = ERR_PTR(ret);
 		goto out_free_label;
 	}
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 95dc90570786..f96367a2463e 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -204,44 +204,39 @@ EXPORT_SYMBOL_GPL(nfs_file_mmap);
 static int
 nfs_file_fsync_commit(struct file *file, int datasync)
 {
-	struct nfs_open_context *ctx = nfs_file_open_context(file);
 	struct inode *inode = file_inode(file);
-	int do_resend, status;
-	int ret = 0;
+	int ret;
 
 	dprintk("NFS: fsync file(%pD2) datasync %d\n", file, datasync);
 
 	nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
-	do_resend = test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags);
-	status = nfs_commit_inode(inode, FLUSH_SYNC);
-	if (status == 0)
-		status = file_check_and_advance_wb_err(file);
-	if (status < 0) {
-		ret = status;
-		goto out;
-	}
-	do_resend |= test_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags);
-	if (do_resend)
-		ret = -EAGAIN;
-out:
-	return ret;
+	ret = nfs_commit_inode(inode, FLUSH_SYNC);
+	if (ret < 0)
+		return ret;
+	return file_check_and_advance_wb_err(file);
 }
 
 int
 nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	int ret;
+	struct nfs_open_context *ctx = nfs_file_open_context(file);
 	struct inode *inode = file_inode(file);
+	int ret;
 
 	trace_nfs_fsync_enter(inode);
 
-	do {
+	for (;;) {
 		ret = file_write_and_wait_range(file, start, end);
 		if (ret != 0)
 			break;
 		ret = nfs_file_fsync_commit(file, datasync);
-		if (!ret)
-			ret = pnfs_sync_inode(inode, !!datasync);
+		if (ret != 0)
+			break;
+		ret = pnfs_sync_inode(inode, !!datasync);
+		if (ret != 0)
+			break;
+		if (!test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags))
+			break;
 		/*
 		 * If nfs_file_fsync_commit detected a server reboot, then
 		 * resend all dirty pages that might have been covered by
@@ -249,7 +244,7 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 		 */
 		start = 0;
 		end = LLONG_MAX;
-	} while (ret == -EAGAIN);
+	}
 
 	trace_nfs_fsync_exit(inode, ret);
 	return ret;
@@ -489,7 +484,19 @@ static int nfs_launder_page(struct page *page)
 static int nfs_swap_activate(struct swap_info_struct *sis, struct file *file,
 						sector_t *span)
 {
+	unsigned long blocks;
+	long long isize;
 	struct rpc_clnt *clnt = NFS_CLIENT(file->f_mapping->host);
+	struct inode *inode = file->f_mapping->host;
+
+	spin_lock(&inode->i_lock);
+	blocks = inode->i_blocks;
+	isize = inode->i_size;
+	spin_unlock(&inode->i_lock);
+	if (blocks*512 < isize) {
+		pr_warn("swap activate: swapfile has holes\n");
+		return -EINVAL;
+	}
 
 	*span = sis->pages;
 
@@ -649,7 +656,7 @@ out:
 
 out_swapfile:
 	printk(KERN_INFO "NFS: attempt to write to active swap file!\n");
-	return -EBUSY;
+	return -ETXTBSY;
 }
 EXPORT_SYMBOL_GPL(nfs_file_write);
 
diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c
index c9b605f6c9cb..a13e69009f19 100644
--- a/fs/nfs/filelayout/filelayout.c
+++ b/fs/nfs/filelayout/filelayout.c
@@ -49,6 +49,7 @@ MODULE_AUTHOR("Dean Hildebrand <dhildebz@umich.edu>");
 MODULE_DESCRIPTION("The NFSv4 file layout driver");
 
 #define FILELAYOUT_POLL_RETRY_MAX     (15*HZ)
+static const struct pnfs_commit_ops filelayout_commit_ops;
 
 static loff_t
 filelayout_get_dense_offset(struct nfs4_filelayout_segment *flseg,
@@ -750,72 +751,17 @@ filelayout_free_lseg(struct pnfs_layout_segment *lseg)
 	/* This assumes a single RW lseg */
 	if (lseg->pls_range.iomode == IOMODE_RW) {
 		struct nfs4_filelayout *flo;
+		struct inode *inode;
 
 		flo = FILELAYOUT_FROM_HDR(lseg->pls_layout);
-		flo->commit_info.nbuckets = 0;
-		kfree(flo->commit_info.buckets);
-		flo->commit_info.buckets = NULL;
+		inode = flo->generic_hdr.plh_inode;
+		spin_lock(&inode->i_lock);
+		pnfs_generic_ds_cinfo_release_lseg(&flo->commit_info, lseg);
+		spin_unlock(&inode->i_lock);
 	}
 	_filelayout_free_lseg(fl);
 }
 
-static int
-filelayout_alloc_commit_info(struct pnfs_layout_segment *lseg,
-			     struct nfs_commit_info *cinfo,
-			     gfp_t gfp_flags)
-{
-	struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
-	struct pnfs_commit_bucket *buckets;
-	int size, i;
-
-	if (fl->commit_through_mds)
-		return 0;
-
-	size = (fl->stripe_type == STRIPE_SPARSE) ?
-		fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
-
-	if (cinfo->ds->nbuckets >= size) {
-		/* This assumes there is only one IOMODE_RW lseg.  What
-		 * we really want to do is have a layout_hdr level
-		 * dictionary of <multipath_list4, fh> keys, each
-		 * associated with a struct list_head, populated by calls
-		 * to filelayout_write_pagelist().
-		 * */
-		return 0;
-	}
-
-	buckets = kcalloc(size, sizeof(struct pnfs_commit_bucket),
-			  gfp_flags);
-	if (!buckets)
-		return -ENOMEM;
-	for (i = 0; i < size; i++) {
-		INIT_LIST_HEAD(&buckets[i].written);
-		INIT_LIST_HEAD(&buckets[i].committing);
-		/* mark direct verifier as unset */
-		buckets[i].direct_verf.committed = NFS_INVALID_STABLE_HOW;
-	}
-
-	spin_lock(&cinfo->inode->i_lock);
-	if (cinfo->ds->nbuckets >= size)
-		goto out;
-	for (i = 0; i < cinfo->ds->nbuckets; i++) {
-		list_splice(&cinfo->ds->buckets[i].written,
-			    &buckets[i].written);
-		list_splice(&cinfo->ds->buckets[i].committing,
-			    &buckets[i].committing);
-		buckets[i].direct_verf.committed =
-			cinfo->ds->buckets[i].direct_verf.committed;
-		buckets[i].wlseg = cinfo->ds->buckets[i].wlseg;
-		buckets[i].clseg = cinfo->ds->buckets[i].clseg;
-	}
-	swap(cinfo->ds->buckets, buckets);
-	cinfo->ds->nbuckets = size;
-out:
-	spin_unlock(&cinfo->inode->i_lock);
-	kfree(buckets);
-	return 0;
-}
-
 static struct pnfs_layout_segment *
 filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
 		      struct nfs4_layoutget_res *lgr,
@@ -938,9 +884,6 @@ static void
 filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 			 struct nfs_page *req)
 {
-	struct nfs_commit_info cinfo;
-	int status;
-
 	pnfs_generic_pg_check_layout(pgio);
 	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode,
@@ -959,17 +902,7 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 
 	/* If no lseg, fall back to write through mds */
 	if (pgio->pg_lseg == NULL)
-		goto out_mds;
-	nfs_init_cinfo(&cinfo, pgio->pg_inode, pgio->pg_dreq);
-	status = filelayout_alloc_commit_info(pgio->pg_lseg, &cinfo, GFP_NOFS);
-	if (status < 0) {
-		pnfs_put_lseg(pgio->pg_lseg);
-		pgio->pg_lseg = NULL;
-		goto out_mds;
-	}
-	return;
-out_mds:
-	nfs_pageio_reset_write_mds(pgio);
+		nfs_pageio_reset_write_mds(pgio);
 }
 
 static const struct nfs_pageio_ops filelayout_pg_read_ops = {
@@ -1078,36 +1011,6 @@ out_err:
 	return -EAGAIN;
 }
 
-/* filelayout_search_commit_reqs - Search lists in @cinfo for the head reqest
- *				   for @page
- * @cinfo - commit info for current inode
- * @page - page to search for matching head request
- *
- * Returns a the head request if one is found, otherwise returns NULL.
- */
-static struct nfs_page *
-filelayout_search_commit_reqs(struct nfs_commit_info *cinfo, struct page *page)
-{
-	struct nfs_page *freq, *t;
-	struct pnfs_commit_bucket *b;
-	int i;
-
-	/* Linearly search the commit lists for each bucket until a matching
-	 * request is found */
-	for (i = 0, b = cinfo->ds->buckets; i < cinfo->ds->nbuckets; i++, b++) {
-		list_for_each_entry_safe(freq, t, &b->written, wb_list) {
-			if (freq->wb_page == page)
-				return freq->wb_head;
-		}
-		list_for_each_entry_safe(freq, t, &b->committing, wb_list) {
-			if (freq->wb_page == page)
-				return freq->wb_head;
-		}
-	}
-
-	return NULL;
-}
-
 static int
 filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
 			   int how, struct nfs_commit_info *cinfo)
@@ -1140,13 +1043,17 @@ filelayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
 	struct nfs4_filelayout *flo;
 
 	flo = kzalloc(sizeof(*flo), gfp_flags);
-	return flo != NULL ? &flo->generic_hdr : NULL;
+	if (flo == NULL)
+		return NULL;
+	pnfs_init_ds_commit_info(&flo->commit_info);
+	flo->commit_info.ops = &filelayout_commit_ops;
+	return &flo->generic_hdr;
 }
 
 static void
 filelayout_free_layout_hdr(struct pnfs_layout_hdr *lo)
 {
-	kfree(FILELAYOUT_FROM_HDR(lo));
+	kfree_rcu(FILELAYOUT_FROM_HDR(lo), generic_hdr.plh_rcu);
 }
 
 static struct pnfs_ds_commit_info *
@@ -1160,6 +1067,46 @@ filelayout_get_ds_info(struct inode *inode)
 		return &FILELAYOUT_FROM_HDR(layout)->commit_info;
 }
 
+static void
+filelayout_setup_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg)
+{
+	struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
+	struct inode *inode = lseg->pls_layout->plh_inode;
+	struct pnfs_commit_array *array, *new;
+	unsigned int size = (fl->stripe_type == STRIPE_SPARSE) ?
+		fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
+
+	new = pnfs_alloc_commit_array(size, GFP_NOIO);
+	if (new) {
+		spin_lock(&inode->i_lock);
+		array = pnfs_add_commit_array(fl_cinfo, new, lseg);
+		spin_unlock(&inode->i_lock);
+		if (array != new)
+			pnfs_free_commit_array(new);
+	}
+}
+
+static void
+filelayout_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
+		struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	pnfs_generic_ds_cinfo_destroy(fl_cinfo);
+	spin_unlock(&inode->i_lock);
+}
+
+static const struct pnfs_commit_ops filelayout_commit_ops = {
+	.setup_ds_info		= filelayout_setup_ds_info,
+	.release_ds_info	= filelayout_release_ds_info,
+	.mark_request_commit	= filelayout_mark_request_commit,
+	.clear_request_commit	= pnfs_generic_clear_request_commit,
+	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
+	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
+	.search_commit_reqs	= pnfs_generic_search_commit_reqs,
+	.commit_pagelist	= filelayout_commit_pagelist,
+};
+
 static struct pnfs_layoutdriver_type filelayout_type = {
 	.id			= LAYOUT_NFSV4_1_FILES,
 	.name			= "LAYOUT_NFSV4_1_FILES",
@@ -1173,12 +1120,6 @@ static struct pnfs_layoutdriver_type filelayout_type = {
 	.pg_read_ops		= &filelayout_pg_read_ops,
 	.pg_write_ops		= &filelayout_pg_write_ops,
 	.get_ds_info		= &filelayout_get_ds_info,
-	.mark_request_commit	= filelayout_mark_request_commit,
-	.clear_request_commit	= pnfs_generic_clear_request_commit,
-	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
-	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
-	.search_commit_reqs	= filelayout_search_commit_reqs,
-	.commit_pagelist	= filelayout_commit_pagelist,
 	.read_pagelist		= filelayout_read_pagelist,
 	.write_pagelist		= filelayout_write_pagelist,
 	.alloc_deviceid_node	= filelayout_alloc_deviceid_node,
diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c
index 5657b7f2611f..7d399f72ebbb 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.c
+++ b/fs/nfs/flexfilelayout/flexfilelayout.c
@@ -32,6 +32,7 @@
 
 static unsigned short io_maxretrans;
 
+static const struct pnfs_commit_ops ff_layout_commit_ops;
 static void ff_layout_read_record_layoutstats_done(struct rpc_task *task,
 		struct nfs_pgio_header *hdr);
 static int ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo,
@@ -48,9 +49,11 @@ ff_layout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
 
 	ffl = kzalloc(sizeof(*ffl), gfp_flags);
 	if (ffl) {
+		pnfs_init_ds_commit_info(&ffl->commit_info);
 		INIT_LIST_HEAD(&ffl->error_list);
 		INIT_LIST_HEAD(&ffl->mirrors);
 		ffl->last_report_time = ktime_get();
+		ffl->commit_info.ops = &ff_layout_commit_ops;
 		return &ffl->generic_hdr;
 	} else
 		return NULL;
@@ -59,14 +62,14 @@ ff_layout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
 static void
 ff_layout_free_layout_hdr(struct pnfs_layout_hdr *lo)
 {
+	struct nfs4_flexfile_layout *ffl = FF_LAYOUT_FROM_HDR(lo);
 	struct nfs4_ff_layout_ds_err *err, *n;
 
-	list_for_each_entry_safe(err, n, &FF_LAYOUT_FROM_HDR(lo)->error_list,
-				 list) {
+	list_for_each_entry_safe(err, n, &ffl->error_list, list) {
 		list_del(&err->list);
 		kfree(err);
 	}
-	kfree(FF_LAYOUT_FROM_HDR(lo));
+	kfree_rcu(ffl, generic_hdr.plh_rcu);
 }
 
 static int decode_pnfs_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid)
@@ -248,36 +251,10 @@ static void ff_layout_put_mirror(struct nfs4_ff_layout_mirror *mirror)
 
 static void ff_layout_free_mirror_array(struct nfs4_ff_layout_segment *fls)
 {
-	int i;
-
-	if (fls->mirror_array) {
-		for (i = 0; i < fls->mirror_array_cnt; i++) {
-			/* normally mirror_ds is freed in
-			 * .free_deviceid_node but we still do it here
-			 * for .alloc_lseg error path */
-			ff_layout_put_mirror(fls->mirror_array[i]);
-		}
-		kfree(fls->mirror_array);
-		fls->mirror_array = NULL;
-	}
-}
-
-static int ff_layout_check_layout(struct nfs4_layoutget_res *lgr)
-{
-	int ret = 0;
+	u32 i;
 
-	dprintk("--> %s\n", __func__);
-
-	/* FIXME: remove this check when layout segment support is added */
-	if (lgr->range.offset != 0 ||
-	    lgr->range.length != NFS4_MAX_UINT64) {
-		dprintk("%s Only whole file layouts supported. Use MDS i/o\n",
-			__func__);
-		ret = -EINVAL;
-	}
-
-	dprintk("--> %s returns %d\n", __func__, ret);
-	return ret;
+	for (i = 0; i < fls->mirror_array_cnt; i++)
+		ff_layout_put_mirror(fls->mirror_array[i]);
 }
 
 static void _ff_layout_free_lseg(struct nfs4_ff_layout_segment *fls)
@@ -289,6 +266,23 @@ static void _ff_layout_free_lseg(struct nfs4_ff_layout_segment *fls)
 }
 
 static bool
+ff_lseg_match_mirrors(struct pnfs_layout_segment *l1,
+		struct pnfs_layout_segment *l2)
+{
+	const struct nfs4_ff_layout_segment *fl1 = FF_LAYOUT_LSEG(l1);
+	const struct nfs4_ff_layout_segment *fl2 = FF_LAYOUT_LSEG(l1);
+	u32 i;
+
+	if (fl1->mirror_array_cnt != fl2->mirror_array_cnt)
+		return false;
+	for (i = 0; i < fl1->mirror_array_cnt; i++) {
+		if (fl1->mirror_array[i] != fl2->mirror_array[i])
+			return false;
+	}
+	return true;
+}
+
+static bool
 ff_lseg_range_is_after(const struct pnfs_layout_range *l1,
 		const struct pnfs_layout_range *l2)
 {
@@ -323,6 +317,8 @@ ff_lseg_merge(struct pnfs_layout_segment *new,
 			new->pls_range.length);
 	if (new_end < old->pls_range.offset)
 		return false;
+	if (!ff_lseg_match_mirrors(new, old))
+		return false;
 
 	/* Mergeable: copy info from 'old' to 'new' */
 	if (new_end < old_end)
@@ -400,16 +396,13 @@ ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 		goto out_err_free;
 
 	rc = -ENOMEM;
-	fls = kzalloc(sizeof(*fls), gfp_flags);
+	fls = kzalloc(struct_size(fls, mirror_array, mirror_array_cnt),
+			gfp_flags);
 	if (!fls)
 		goto out_err_free;
 
 	fls->mirror_array_cnt = mirror_array_cnt;
 	fls->stripe_unit = stripe_unit;
-	fls->mirror_array = kcalloc(fls->mirror_array_cnt,
-				    sizeof(fls->mirror_array[0]), gfp_flags);
-	if (fls->mirror_array == NULL)
-		goto out_err_free;
 
 	for (i = 0; i < fls->mirror_array_cnt; i++) {
 		struct nfs4_ff_layout_mirror *mirror;
@@ -545,9 +538,6 @@ ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh,
 
 out_sort_mirrors:
 	ff_layout_sort_mirrors(fls);
-	rc = ff_layout_check_layout(lgr);
-	if (rc)
-		goto out_err_free;
 	ret = &fls->generic_hdr;
 	dprintk("<-- %s (success)\n", __func__);
 out_free_page:
@@ -560,17 +550,6 @@ out_err_free:
 	goto out_free_page;
 }
 
-static bool ff_layout_has_rw_segments(struct pnfs_layout_hdr *layout)
-{
-	struct pnfs_layout_segment *lseg;
-
-	list_for_each_entry(lseg, &layout->plh_segs, pls_list)
-		if (lseg->pls_range.iomode == IOMODE_RW)
-			return true;
-
-	return false;
-}
-
 static void
 ff_layout_free_lseg(struct pnfs_layout_segment *lseg)
 {
@@ -585,23 +564,12 @@ ff_layout_free_lseg(struct pnfs_layout_segment *lseg)
 		ffl = FF_LAYOUT_FROM_HDR(lseg->pls_layout);
 		inode = ffl->generic_hdr.plh_inode;
 		spin_lock(&inode->i_lock);
-		if (!ff_layout_has_rw_segments(lseg->pls_layout)) {
-			ffl->commit_info.nbuckets = 0;
-			kfree(ffl->commit_info.buckets);
-			ffl->commit_info.buckets = NULL;
-		}
+		pnfs_generic_ds_cinfo_release_lseg(&ffl->commit_info, lseg);
 		spin_unlock(&inode->i_lock);
 	}
 	_ff_layout_free_lseg(fls);
 }
 
-/* Return 1 until we have multiple lsegs support */
-static int
-ff_layout_get_lseg_count(struct nfs4_ff_layout_segment *fls)
-{
-	return 1;
-}
-
 static void
 nfs4_ff_start_busy_timer(struct nfs4_ff_busy_timer *timer, ktime_t now)
 {
@@ -746,52 +714,6 @@ nfs4_ff_layout_stat_io_end_write(struct rpc_task *task,
 	spin_unlock(&mirror->lock);
 }
 
-static int
-ff_layout_alloc_commit_info(struct pnfs_layout_segment *lseg,
-			    struct nfs_commit_info *cinfo,
-			    gfp_t gfp_flags)
-{
-	struct nfs4_ff_layout_segment *fls = FF_LAYOUT_LSEG(lseg);
-	struct pnfs_commit_bucket *buckets;
-	int size;
-
-	if (cinfo->ds->nbuckets != 0) {
-		/* This assumes there is only one RW lseg per file.
-		 * To support multiple lseg per file, we need to
-		 * change struct pnfs_commit_bucket to allow dynamic
-		 * increasing nbuckets.
-		 */
-		return 0;
-	}
-
-	size = ff_layout_get_lseg_count(fls) * FF_LAYOUT_MIRROR_COUNT(lseg);
-
-	buckets = kcalloc(size, sizeof(struct pnfs_commit_bucket),
-			  gfp_flags);
-	if (!buckets)
-		return -ENOMEM;
-	else {
-		int i;
-
-		spin_lock(&cinfo->inode->i_lock);
-		if (cinfo->ds->nbuckets != 0)
-			kfree(buckets);
-		else {
-			cinfo->ds->buckets = buckets;
-			cinfo->ds->nbuckets = size;
-			for (i = 0; i < size; i++) {
-				INIT_LIST_HEAD(&buckets[i].written);
-				INIT_LIST_HEAD(&buckets[i].committing);
-				/* mark direct verifier as unset */
-				buckets[i].direct_verf.committed =
-					NFS_INVALID_STABLE_HOW;
-			}
-		}
-		spin_unlock(&cinfo->inode->i_lock);
-		return 0;
-	}
-}
-
 static void
 ff_layout_mark_ds_unreachable(struct pnfs_layout_segment *lseg, int idx)
 {
@@ -876,8 +798,8 @@ ff_layout_pg_get_read(struct nfs_pageio_descriptor *pgio,
 	pnfs_put_lseg(pgio->pg_lseg);
 	pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 					   nfs_req_openctx(req),
-					   0,
-					   NFS4_MAX_UINT64,
+					   req_offset(req),
+					   req->wb_bytes,
 					   IOMODE_READ,
 					   strict_iomode,
 					   GFP_KERNEL);
@@ -888,6 +810,14 @@ ff_layout_pg_get_read(struct nfs_pageio_descriptor *pgio,
 }
 
 static void
+ff_layout_pg_check_layout(struct nfs_pageio_descriptor *pgio,
+			  struct nfs_page *req)
+{
+	pnfs_generic_pg_check_layout(pgio);
+	pnfs_generic_pg_check_range(pgio, req);
+}
+
+static void
 ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 			struct nfs_page *req)
 {
@@ -897,7 +827,7 @@ ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 	int ds_idx;
 
 retry:
-	pnfs_generic_pg_check_layout(pgio);
+	ff_layout_pg_check_layout(pgio, req);
 	/* Use full layout for now */
 	if (!pgio->pg_lseg) {
 		ff_layout_pg_get_read(pgio, req, false);
@@ -953,18 +883,16 @@ ff_layout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 {
 	struct nfs4_ff_layout_mirror *mirror;
 	struct nfs_pgio_mirror *pgm;
-	struct nfs_commit_info cinfo;
 	struct nfs4_pnfs_ds *ds;
 	int i;
-	int status;
 
 retry:
-	pnfs_generic_pg_check_layout(pgio);
+	ff_layout_pg_check_layout(pgio, req);
 	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   nfs_req_openctx(req),
-						   0,
-						   NFS4_MAX_UINT64,
+						   req_offset(req),
+						   req->wb_bytes,
 						   IOMODE_RW,
 						   false,
 						   GFP_NOFS);
@@ -978,11 +906,6 @@ retry:
 	if (pgio->pg_lseg == NULL)
 		goto out_mds;
 
-	nfs_init_cinfo(&cinfo, pgio->pg_inode, pgio->pg_dreq);
-	status = ff_layout_alloc_commit_info(pgio->pg_lseg, &cinfo, GFP_NOFS);
-	if (status < 0)
-		goto out_mds;
-
 	/* Use a direct mapping of ds_idx to pgio mirror_idx */
 	if (WARN_ON_ONCE(pgio->pg_mirror_count !=
 	    FF_LAYOUT_MIRROR_COUNT(pgio->pg_lseg)))
@@ -1266,9 +1189,10 @@ static int ff_layout_async_handle_error(struct rpc_task *task,
 
 static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg,
 					int idx, u64 offset, u64 length,
-					u32 status, int opnum, int error)
+					u32 *op_status, int opnum, int error)
 {
 	struct nfs4_ff_layout_mirror *mirror;
+	u32 status = *op_status;
 	int err;
 
 	if (status == 0) {
@@ -1286,31 +1210,33 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg,
 		case -ENOBUFS:
 		case -EPIPE:
 		case -EPERM:
-			status = NFS4ERR_NXIO;
+			*op_status = status = NFS4ERR_NXIO;
 			break;
 		case -EACCES:
-			status = NFS4ERR_ACCESS;
+			*op_status = status = NFS4ERR_ACCESS;
 			break;
 		default:
 			return;
 		}
 	}
 
+	mirror = FF_LAYOUT_COMP(lseg, idx);
+	err = ff_layout_track_ds_error(FF_LAYOUT_FROM_HDR(lseg->pls_layout),
+				       mirror, offset, length, status, opnum,
+				       GFP_NOIO);
+
 	switch (status) {
 	case NFS4ERR_DELAY:
 	case NFS4ERR_GRACE:
-		return;
-	default:
 		break;
+	case NFS4ERR_NXIO:
+		ff_layout_mark_ds_unreachable(lseg, idx);
+		/* Fallthrough */
+	default:
+		pnfs_error_mark_layout_for_return(lseg->pls_layout->plh_inode,
+						  lseg);
 	}
 
-	mirror = FF_LAYOUT_COMP(lseg, idx);
-	err = ff_layout_track_ds_error(FF_LAYOUT_FROM_HDR(lseg->pls_layout),
-				       mirror, offset, length, status, opnum,
-				       GFP_NOIO);
-	if (status == NFS4ERR_NXIO)
-		ff_layout_mark_ds_unreachable(lseg, idx);
-	pnfs_error_mark_layout_for_return(lseg->pls_layout->plh_inode, lseg);
 	dprintk("%s: err %d op %d status %u\n", __func__, err, opnum, status);
 }
 
@@ -1321,16 +1247,19 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
 	int new_idx = hdr->pgio_mirror_idx;
 	int err;
 
-	trace_nfs4_pnfs_read(hdr, task->tk_status);
-	if (task->tk_status < 0)
+	if (task->tk_status < 0) {
 		ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx,
 					    hdr->args.offset, hdr->args.count,
-					    hdr->res.op_status, OP_READ,
+					    &hdr->res.op_status, OP_READ,
 					    task->tk_status);
+		trace_ff_layout_read_error(hdr);
+	}
+
 	err = ff_layout_async_handle_error(task, hdr->args.context->state,
 					   hdr->ds_clp, hdr->lseg,
 					   hdr->pgio_mirror_idx);
 
+	trace_nfs4_pnfs_read(hdr, err);
 	clear_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags);
 	clear_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags);
 	switch (err) {
@@ -1494,16 +1423,19 @@ static int ff_layout_write_done_cb(struct rpc_task *task,
 	loff_t end_offs = 0;
 	int err;
 
-	trace_nfs4_pnfs_write(hdr, task->tk_status);
-	if (task->tk_status < 0)
+	if (task->tk_status < 0) {
 		ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx,
 					    hdr->args.offset, hdr->args.count,
-					    hdr->res.op_status, OP_WRITE,
+					    &hdr->res.op_status, OP_WRITE,
 					    task->tk_status);
+		trace_ff_layout_write_error(hdr);
+	}
+
 	err = ff_layout_async_handle_error(task, hdr->args.context->state,
 					   hdr->ds_clp, hdr->lseg,
 					   hdr->pgio_mirror_idx);
 
+	trace_nfs4_pnfs_write(hdr, err);
 	clear_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags);
 	clear_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags);
 	switch (err) {
@@ -1537,15 +1469,18 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
 {
 	int err;
 
-	trace_nfs4_pnfs_commit_ds(data, task->tk_status);
-	if (task->tk_status < 0)
+	if (task->tk_status < 0) {
 		ff_layout_io_track_ds_error(data->lseg, data->ds_commit_index,
 					    data->args.offset, data->args.count,
-					    data->res.op_status, OP_COMMIT,
+					    &data->res.op_status, OP_COMMIT,
 					    task->tk_status);
+		trace_ff_layout_commit_error(data);
+	}
+
 	err = ff_layout_async_handle_error(task, NULL, data->ds_clp,
 					   data->lseg, data->ds_commit_index);
 
+	trace_nfs4_pnfs_commit_ds(data, err);
 	switch (err) {
 	case -NFS4ERR_RESET_TO_PNFS:
 		pnfs_generic_prepare_to_resend_writes(data);
@@ -2002,6 +1937,33 @@ ff_layout_get_ds_info(struct inode *inode)
 }
 
 static void
+ff_layout_setup_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg)
+{
+	struct nfs4_ff_layout_segment *flseg = FF_LAYOUT_LSEG(lseg);
+	struct inode *inode = lseg->pls_layout->plh_inode;
+	struct pnfs_commit_array *array, *new;
+
+	new = pnfs_alloc_commit_array(flseg->mirror_array_cnt, GFP_NOIO);
+	if (new) {
+		spin_lock(&inode->i_lock);
+		array = pnfs_add_commit_array(fl_cinfo, new, lseg);
+		spin_unlock(&inode->i_lock);
+		if (array != new)
+			pnfs_free_commit_array(new);
+	}
+}
+
+static void
+ff_layout_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
+		struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	pnfs_generic_ds_cinfo_destroy(fl_cinfo);
+	spin_unlock(&inode->i_lock);
+}
+
+static void
 ff_layout_free_deviceid_node(struct nfs4_deviceid_node *d)
 {
 	nfs4_ff_layout_free_deviceid(container_of(d, struct nfs4_ff_layout_ds,
@@ -2486,6 +2448,16 @@ ff_layout_set_layoutdriver(struct nfs_server *server,
 	return 0;
 }
 
+static const struct pnfs_commit_ops ff_layout_commit_ops = {
+	.setup_ds_info		= ff_layout_setup_ds_info,
+	.release_ds_info	= ff_layout_release_ds_info,
+	.mark_request_commit	= pnfs_layout_mark_request_commit,
+	.clear_request_commit	= pnfs_generic_clear_request_commit,
+	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
+	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
+	.commit_pagelist	= ff_layout_commit_pagelist,
+};
+
 static struct pnfs_layoutdriver_type flexfilelayout_type = {
 	.id			= LAYOUT_FLEX_FILES,
 	.name			= "LAYOUT_FLEX_FILES",
@@ -2502,11 +2474,6 @@ static struct pnfs_layoutdriver_type flexfilelayout_type = {
 	.pg_write_ops		= &ff_layout_pg_write_ops,
 	.get_ds_info		= ff_layout_get_ds_info,
 	.free_deviceid_node	= ff_layout_free_deviceid_node,
-	.mark_request_commit	= pnfs_layout_mark_request_commit,
-	.clear_request_commit	= pnfs_generic_clear_request_commit,
-	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
-	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
-	.commit_pagelist	= ff_layout_commit_pagelist,
 	.read_pagelist		= ff_layout_read_pagelist,
 	.write_pagelist		= ff_layout_write_pagelist,
 	.alloc_deviceid_node    = ff_layout_alloc_deviceid_node,
diff --git a/fs/nfs/flexfilelayout/flexfilelayout.h b/fs/nfs/flexfilelayout/flexfilelayout.h
index 2f369966abf7..354a031c69b1 100644
--- a/fs/nfs/flexfilelayout/flexfilelayout.h
+++ b/fs/nfs/flexfilelayout/flexfilelayout.h
@@ -99,7 +99,7 @@ struct nfs4_ff_layout_segment {
 	u64				stripe_unit;
 	u32				flags;
 	u32				mirror_array_cnt;
-	struct nfs4_ff_layout_mirror	**mirror_array;
+	struct nfs4_ff_layout_mirror	*mirror_array[];
 };
 
 struct nfs4_flexfile_layout {
diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c
new file mode 100644
index 000000000000..ccc88be88d6a
--- /dev/null
+++ b/fs/nfs/fs_context.c
@@ -0,0 +1,1452 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * linux/fs/nfs/fs_context.c
+ *
+ * Copyright (C) 1992 Rick Sladkey
+ * Conversion to new mount api Copyright (C) David Howells
+ *
+ * NFS mount handling.
+ *
+ * Split from fs/nfs/super.c by David Howells <dhowells@redhat.com>
+ */
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/fs_context.h>
+#include <linux/fs_parser.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
+#include <linux/nfs4_mount.h>
+#include "nfs.h"
+#include "internal.h"
+
+#define NFSDBG_FACILITY		NFSDBG_MOUNT
+
+#if IS_ENABLED(CONFIG_NFS_V3)
+#define NFS_DEFAULT_VERSION 3
+#else
+#define NFS_DEFAULT_VERSION 2
+#endif
+
+#define NFS_MAX_CONNECTIONS 16
+
+enum nfs_param {
+	Opt_ac,
+	Opt_acdirmax,
+	Opt_acdirmin,
+	Opt_acl,
+	Opt_acregmax,
+	Opt_acregmin,
+	Opt_actimeo,
+	Opt_addr,
+	Opt_bg,
+	Opt_bsize,
+	Opt_clientaddr,
+	Opt_cto,
+	Opt_fg,
+	Opt_fscache,
+	Opt_fscache_flag,
+	Opt_hard,
+	Opt_intr,
+	Opt_local_lock,
+	Opt_lock,
+	Opt_lookupcache,
+	Opt_migration,
+	Opt_minorversion,
+	Opt_mountaddr,
+	Opt_mounthost,
+	Opt_mountport,
+	Opt_mountproto,
+	Opt_mountvers,
+	Opt_namelen,
+	Opt_nconnect,
+	Opt_port,
+	Opt_posix,
+	Opt_proto,
+	Opt_rdirplus,
+	Opt_rdma,
+	Opt_resvport,
+	Opt_retrans,
+	Opt_retry,
+	Opt_rsize,
+	Opt_sec,
+	Opt_sharecache,
+	Opt_sloppy,
+	Opt_soft,
+	Opt_softerr,
+	Opt_softreval,
+	Opt_source,
+	Opt_tcp,
+	Opt_timeo,
+	Opt_udp,
+	Opt_v,
+	Opt_vers,
+	Opt_wsize,
+};
+
+enum {
+	Opt_local_lock_all,
+	Opt_local_lock_flock,
+	Opt_local_lock_none,
+	Opt_local_lock_posix,
+};
+
+static const struct constant_table nfs_param_enums_local_lock[] = {
+	{ "all",		Opt_local_lock_all },
+	{ "flock",	Opt_local_lock_flock },
+	{ "none",		Opt_local_lock_none },
+	{}
+};
+
+enum {
+	Opt_lookupcache_all,
+	Opt_lookupcache_none,
+	Opt_lookupcache_positive,
+};
+
+static const struct constant_table nfs_param_enums_lookupcache[] = {
+	{ "all",		Opt_lookupcache_all },
+	{ "none",		Opt_lookupcache_none },
+	{ "pos",		Opt_lookupcache_positive },
+	{ "positive",		Opt_lookupcache_positive },
+	{}
+};
+
+static const struct fs_parameter_spec nfs_fs_parameters[] = {
+	fsparam_flag_no("ac",		Opt_ac),
+	fsparam_u32   ("acdirmax",	Opt_acdirmax),
+	fsparam_u32   ("acdirmin",	Opt_acdirmin),
+	fsparam_flag_no("acl",		Opt_acl),
+	fsparam_u32   ("acregmax",	Opt_acregmax),
+	fsparam_u32   ("acregmin",	Opt_acregmin),
+	fsparam_u32   ("actimeo",	Opt_actimeo),
+	fsparam_string("addr",		Opt_addr),
+	fsparam_flag  ("bg",		Opt_bg),
+	fsparam_u32   ("bsize",		Opt_bsize),
+	fsparam_string("clientaddr",	Opt_clientaddr),
+	fsparam_flag_no("cto",		Opt_cto),
+	fsparam_flag  ("fg",		Opt_fg),
+	fsparam_flag_no("fsc",		Opt_fscache_flag),
+	fsparam_string("fsc",		Opt_fscache),
+	fsparam_flag  ("hard",		Opt_hard),
+	__fsparam(NULL, "intr",		Opt_intr,
+		  fs_param_neg_with_no|fs_param_deprecated, NULL),
+	fsparam_enum  ("local_lock",	Opt_local_lock, nfs_param_enums_local_lock),
+	fsparam_flag_no("lock",		Opt_lock),
+	fsparam_enum  ("lookupcache",	Opt_lookupcache, nfs_param_enums_lookupcache),
+	fsparam_flag_no("migration",	Opt_migration),
+	fsparam_u32   ("minorversion",	Opt_minorversion),
+	fsparam_string("mountaddr",	Opt_mountaddr),
+	fsparam_string("mounthost",	Opt_mounthost),
+	fsparam_u32   ("mountport",	Opt_mountport),
+	fsparam_string("mountproto",	Opt_mountproto),
+	fsparam_u32   ("mountvers",	Opt_mountvers),
+	fsparam_u32   ("namlen",	Opt_namelen),
+	fsparam_u32   ("nconnect",	Opt_nconnect),
+	fsparam_string("nfsvers",	Opt_vers),
+	fsparam_u32   ("port",		Opt_port),
+	fsparam_flag_no("posix",	Opt_posix),
+	fsparam_string("proto",		Opt_proto),
+	fsparam_flag_no("rdirplus",	Opt_rdirplus),
+	fsparam_flag  ("rdma",		Opt_rdma),
+	fsparam_flag_no("resvport",	Opt_resvport),
+	fsparam_u32   ("retrans",	Opt_retrans),
+	fsparam_string("retry",		Opt_retry),
+	fsparam_u32   ("rsize",		Opt_rsize),
+	fsparam_string("sec",		Opt_sec),
+	fsparam_flag_no("sharecache",	Opt_sharecache),
+	fsparam_flag  ("sloppy",	Opt_sloppy),
+	fsparam_flag  ("soft",		Opt_soft),
+	fsparam_flag  ("softerr",	Opt_softerr),
+	fsparam_flag  ("softreval",	Opt_softreval),
+	fsparam_string("source",	Opt_source),
+	fsparam_flag  ("tcp",		Opt_tcp),
+	fsparam_u32   ("timeo",		Opt_timeo),
+	fsparam_flag  ("udp",		Opt_udp),
+	fsparam_flag  ("v2",		Opt_v),
+	fsparam_flag  ("v3",		Opt_v),
+	fsparam_flag  ("v4",		Opt_v),
+	fsparam_flag  ("v4.0",		Opt_v),
+	fsparam_flag  ("v4.1",		Opt_v),
+	fsparam_flag  ("v4.2",		Opt_v),
+	fsparam_string("vers",		Opt_vers),
+	fsparam_u32   ("wsize",		Opt_wsize),
+	{}
+};
+
+enum {
+	Opt_vers_2,
+	Opt_vers_3,
+	Opt_vers_4,
+	Opt_vers_4_0,
+	Opt_vers_4_1,
+	Opt_vers_4_2,
+};
+
+static const struct constant_table nfs_vers_tokens[] = {
+	{ "2",		Opt_vers_2 },
+	{ "3",		Opt_vers_3 },
+	{ "4",		Opt_vers_4 },
+	{ "4.0",	Opt_vers_4_0 },
+	{ "4.1",	Opt_vers_4_1 },
+	{ "4.2",	Opt_vers_4_2 },
+	{}
+};
+
+enum {
+	Opt_xprt_rdma,
+	Opt_xprt_rdma6,
+	Opt_xprt_tcp,
+	Opt_xprt_tcp6,
+	Opt_xprt_udp,
+	Opt_xprt_udp6,
+	nr__Opt_xprt
+};
+
+static const struct constant_table nfs_xprt_protocol_tokens[] = {
+	{ "rdma",	Opt_xprt_rdma },
+	{ "rdma6",	Opt_xprt_rdma6 },
+	{ "tcp",	Opt_xprt_tcp },
+	{ "tcp6",	Opt_xprt_tcp6 },
+	{ "udp",	Opt_xprt_udp },
+	{ "udp6",	Opt_xprt_udp6 },
+	{}
+};
+
+enum {
+	Opt_sec_krb5,
+	Opt_sec_krb5i,
+	Opt_sec_krb5p,
+	Opt_sec_lkey,
+	Opt_sec_lkeyi,
+	Opt_sec_lkeyp,
+	Opt_sec_none,
+	Opt_sec_spkm,
+	Opt_sec_spkmi,
+	Opt_sec_spkmp,
+	Opt_sec_sys,
+	nr__Opt_sec
+};
+
+static const struct constant_table nfs_secflavor_tokens[] = {
+	{ "krb5",	Opt_sec_krb5 },
+	{ "krb5i",	Opt_sec_krb5i },
+	{ "krb5p",	Opt_sec_krb5p },
+	{ "lkey",	Opt_sec_lkey },
+	{ "lkeyi",	Opt_sec_lkeyi },
+	{ "lkeyp",	Opt_sec_lkeyp },
+	{ "none",	Opt_sec_none },
+	{ "null",	Opt_sec_none },
+	{ "spkm3",	Opt_sec_spkm },
+	{ "spkm3i",	Opt_sec_spkmi },
+	{ "spkm3p",	Opt_sec_spkmp },
+	{ "sys",	Opt_sec_sys },
+	{}
+};
+
+/*
+ * Sanity-check a server address provided by the mount command.
+ *
+ * Address family must be initialized, and address must not be
+ * the ANY address for that family.
+ */
+static int nfs_verify_server_address(struct sockaddr *addr)
+{
+	switch (addr->sa_family) {
+	case AF_INET: {
+		struct sockaddr_in *sa = (struct sockaddr_in *)addr;
+		return sa->sin_addr.s_addr != htonl(INADDR_ANY);
+	}
+	case AF_INET6: {
+		struct in6_addr *sa = &((struct sockaddr_in6 *)addr)->sin6_addr;
+		return !ipv6_addr_any(sa);
+	}
+	}
+
+	dfprintk(MOUNT, "NFS: Invalid IP address specified\n");
+	return 0;
+}
+
+/*
+ * Sanity check the NFS transport protocol.
+ *
+ */
+static void nfs_validate_transport_protocol(struct nfs_fs_context *ctx)
+{
+	switch (ctx->nfs_server.protocol) {
+	case XPRT_TRANSPORT_UDP:
+	case XPRT_TRANSPORT_TCP:
+	case XPRT_TRANSPORT_RDMA:
+		break;
+	default:
+		ctx->nfs_server.protocol = XPRT_TRANSPORT_TCP;
+	}
+}
+
+/*
+ * For text based NFSv2/v3 mounts, the mount protocol transport default
+ * settings should depend upon the specified NFS transport.
+ */
+static void nfs_set_mount_transport_protocol(struct nfs_fs_context *ctx)
+{
+	nfs_validate_transport_protocol(ctx);
+
+	if (ctx->mount_server.protocol == XPRT_TRANSPORT_UDP ||
+	    ctx->mount_server.protocol == XPRT_TRANSPORT_TCP)
+			return;
+	switch (ctx->nfs_server.protocol) {
+	case XPRT_TRANSPORT_UDP:
+		ctx->mount_server.protocol = XPRT_TRANSPORT_UDP;
+		break;
+	case XPRT_TRANSPORT_TCP:
+	case XPRT_TRANSPORT_RDMA:
+		ctx->mount_server.protocol = XPRT_TRANSPORT_TCP;
+	}
+}
+
+/*
+ * Add 'flavor' to 'auth_info' if not already present.
+ * Returns true if 'flavor' ends up in the list, false otherwise
+ */
+static int nfs_auth_info_add(struct fs_context *fc,
+			     struct nfs_auth_info *auth_info,
+			     rpc_authflavor_t flavor)
+{
+	unsigned int i;
+	unsigned int max_flavor_len = ARRAY_SIZE(auth_info->flavors);
+
+	/* make sure this flavor isn't already in the list */
+	for (i = 0; i < auth_info->flavor_len; i++) {
+		if (flavor == auth_info->flavors[i])
+			return 0;
+	}
+
+	if (auth_info->flavor_len + 1 >= max_flavor_len)
+		return nfs_invalf(fc, "NFS: too many sec= flavors");
+
+	auth_info->flavors[auth_info->flavor_len++] = flavor;
+	return 0;
+}
+
+/*
+ * Parse the value of the 'sec=' option.
+ */
+static int nfs_parse_security_flavors(struct fs_context *fc,
+				      struct fs_parameter *param)
+{
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	rpc_authflavor_t pseudoflavor;
+	char *string = param->string, *p;
+	int ret;
+
+	dfprintk(MOUNT, "NFS: parsing %s=%s option\n", param->key, param->string);
+
+	while ((p = strsep(&string, ":")) != NULL) {
+		if (!*p)
+			continue;
+		switch (lookup_constant(nfs_secflavor_tokens, p, -1)) {
+		case Opt_sec_none:
+			pseudoflavor = RPC_AUTH_NULL;
+			break;
+		case Opt_sec_sys:
+			pseudoflavor = RPC_AUTH_UNIX;
+			break;
+		case Opt_sec_krb5:
+			pseudoflavor = RPC_AUTH_GSS_KRB5;
+			break;
+		case Opt_sec_krb5i:
+			pseudoflavor = RPC_AUTH_GSS_KRB5I;
+			break;
+		case Opt_sec_krb5p:
+			pseudoflavor = RPC_AUTH_GSS_KRB5P;
+			break;
+		case Opt_sec_lkey:
+			pseudoflavor = RPC_AUTH_GSS_LKEY;
+			break;
+		case Opt_sec_lkeyi:
+			pseudoflavor = RPC_AUTH_GSS_LKEYI;
+			break;
+		case Opt_sec_lkeyp:
+			pseudoflavor = RPC_AUTH_GSS_LKEYP;
+			break;
+		case Opt_sec_spkm:
+			pseudoflavor = RPC_AUTH_GSS_SPKM;
+			break;
+		case Opt_sec_spkmi:
+			pseudoflavor = RPC_AUTH_GSS_SPKMI;
+			break;
+		case Opt_sec_spkmp:
+			pseudoflavor = RPC_AUTH_GSS_SPKMP;
+			break;
+		default:
+			return nfs_invalf(fc, "NFS: sec=%s option not recognized", p);
+		}
+
+		ret = nfs_auth_info_add(fc, &ctx->auth_info, pseudoflavor);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int nfs_parse_version_string(struct fs_context *fc,
+				    const char *string)
+{
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+
+	ctx->flags &= ~NFS_MOUNT_VER3;
+	switch (lookup_constant(nfs_vers_tokens, string, -1)) {
+	case Opt_vers_2:
+		ctx->version = 2;
+		break;
+	case Opt_vers_3:
+		ctx->flags |= NFS_MOUNT_VER3;
+		ctx->version = 3;
+		break;
+	case Opt_vers_4:
+		/* Backward compatibility option. In future,
+		 * the mount program should always supply
+		 * a NFSv4 minor version number.
+		 */
+		ctx->version = 4;
+		break;
+	case Opt_vers_4_0:
+		ctx->version = 4;
+		ctx->minorversion = 0;
+		break;
+	case Opt_vers_4_1:
+		ctx->version = 4;
+		ctx->minorversion = 1;
+		break;
+	case Opt_vers_4_2:
+		ctx->version = 4;
+		ctx->minorversion = 2;
+		break;
+	default:
+		return nfs_invalf(fc, "NFS: Unsupported NFS version");
+	}
+	return 0;
+}
+
+/*
+ * Parse a single mount parameter.
+ */
+static int nfs_fs_context_parse_param(struct fs_context *fc,
+				      struct fs_parameter *param)
+{
+	struct fs_parse_result result;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	unsigned short protofamily, mountfamily;
+	unsigned int len;
+	int ret, opt;
+
+	dfprintk(MOUNT, "NFS:   parsing nfs mount option '%s'\n", param->key);
+
+	opt = fs_parse(fc, nfs_fs_parameters, param, &result);
+	if (opt < 0)
+		return ctx->sloppy ? 1 : opt;
+
+	switch (opt) {
+	case Opt_source:
+		if (fc->source)
+			return nfs_invalf(fc, "NFS: Multiple sources not supported");
+		fc->source = param->string;
+		param->string = NULL;
+		break;
+
+		/*
+		 * boolean options:  foo/nofoo
+		 */
+	case Opt_soft:
+		ctx->flags |= NFS_MOUNT_SOFT;
+		ctx->flags &= ~NFS_MOUNT_SOFTERR;
+		break;
+	case Opt_softerr:
+		ctx->flags |= NFS_MOUNT_SOFTERR | NFS_MOUNT_SOFTREVAL;
+		ctx->flags &= ~NFS_MOUNT_SOFT;
+		break;
+	case Opt_hard:
+		ctx->flags &= ~(NFS_MOUNT_SOFT |
+				NFS_MOUNT_SOFTERR |
+				NFS_MOUNT_SOFTREVAL);
+		break;
+	case Opt_softreval:
+		if (result.negated)
+			ctx->flags &= ~NFS_MOUNT_SOFTREVAL;
+		else
+			ctx->flags &= NFS_MOUNT_SOFTREVAL;
+		break;
+	case Opt_posix:
+		if (result.negated)
+			ctx->flags &= ~NFS_MOUNT_POSIX;
+		else
+			ctx->flags |= NFS_MOUNT_POSIX;
+		break;
+	case Opt_cto:
+		if (result.negated)
+			ctx->flags |= NFS_MOUNT_NOCTO;
+		else
+			ctx->flags &= ~NFS_MOUNT_NOCTO;
+		break;
+	case Opt_ac:
+		if (result.negated)
+			ctx->flags |= NFS_MOUNT_NOAC;
+		else
+			ctx->flags &= ~NFS_MOUNT_NOAC;
+		break;
+	case Opt_lock:
+		if (result.negated) {
+			ctx->flags |= NFS_MOUNT_NONLM;
+			ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL);
+		} else {
+			ctx->flags &= ~NFS_MOUNT_NONLM;
+			ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL);
+		}
+		break;
+	case Opt_udp:
+		ctx->flags &= ~NFS_MOUNT_TCP;
+		ctx->nfs_server.protocol = XPRT_TRANSPORT_UDP;
+		break;
+	case Opt_tcp:
+		ctx->flags |= NFS_MOUNT_TCP;
+		ctx->nfs_server.protocol = XPRT_TRANSPORT_TCP;
+		break;
+	case Opt_rdma:
+		ctx->flags |= NFS_MOUNT_TCP; /* for side protocols */
+		ctx->nfs_server.protocol = XPRT_TRANSPORT_RDMA;
+		xprt_load_transport(param->key);
+		break;
+	case Opt_acl:
+		if (result.negated)
+			ctx->flags |= NFS_MOUNT_NOACL;
+		else
+			ctx->flags &= ~NFS_MOUNT_NOACL;
+		break;
+	case Opt_rdirplus:
+		if (result.negated)
+			ctx->flags |= NFS_MOUNT_NORDIRPLUS;
+		else
+			ctx->flags &= ~NFS_MOUNT_NORDIRPLUS;
+		break;
+	case Opt_sharecache:
+		if (result.negated)
+			ctx->flags |= NFS_MOUNT_UNSHARED;
+		else
+			ctx->flags &= ~NFS_MOUNT_UNSHARED;
+		break;
+	case Opt_resvport:
+		if (result.negated)
+			ctx->flags |= NFS_MOUNT_NORESVPORT;
+		else
+			ctx->flags &= ~NFS_MOUNT_NORESVPORT;
+		break;
+	case Opt_fscache_flag:
+		if (result.negated)
+			ctx->options &= ~NFS_OPTION_FSCACHE;
+		else
+			ctx->options |= NFS_OPTION_FSCACHE;
+		kfree(ctx->fscache_uniq);
+		ctx->fscache_uniq = NULL;
+		break;
+	case Opt_fscache:
+		ctx->options |= NFS_OPTION_FSCACHE;
+		kfree(ctx->fscache_uniq);
+		ctx->fscache_uniq = param->string;
+		param->string = NULL;
+		break;
+	case Opt_migration:
+		if (result.negated)
+			ctx->options &= ~NFS_OPTION_MIGRATION;
+		else
+			ctx->options |= NFS_OPTION_MIGRATION;
+		break;
+
+		/*
+		 * options that take numeric values
+		 */
+	case Opt_port:
+		if (result.uint_32 > USHRT_MAX)
+			goto out_of_bounds;
+		ctx->nfs_server.port = result.uint_32;
+		break;
+	case Opt_rsize:
+		ctx->rsize = result.uint_32;
+		break;
+	case Opt_wsize:
+		ctx->wsize = result.uint_32;
+		break;
+	case Opt_bsize:
+		ctx->bsize = result.uint_32;
+		break;
+	case Opt_timeo:
+		if (result.uint_32 < 1 || result.uint_32 > INT_MAX)
+			goto out_of_bounds;
+		ctx->timeo = result.uint_32;
+		break;
+	case Opt_retrans:
+		if (result.uint_32 > INT_MAX)
+			goto out_of_bounds;
+		ctx->retrans = result.uint_32;
+		break;
+	case Opt_acregmin:
+		ctx->acregmin = result.uint_32;
+		break;
+	case Opt_acregmax:
+		ctx->acregmax = result.uint_32;
+		break;
+	case Opt_acdirmin:
+		ctx->acdirmin = result.uint_32;
+		break;
+	case Opt_acdirmax:
+		ctx->acdirmax = result.uint_32;
+		break;
+	case Opt_actimeo:
+		ctx->acregmin = result.uint_32;
+		ctx->acregmax = result.uint_32;
+		ctx->acdirmin = result.uint_32;
+		ctx->acdirmax = result.uint_32;
+		break;
+	case Opt_namelen:
+		ctx->namlen = result.uint_32;
+		break;
+	case Opt_mountport:
+		if (result.uint_32 > USHRT_MAX)
+			goto out_of_bounds;
+		ctx->mount_server.port = result.uint_32;
+		break;
+	case Opt_mountvers:
+		if (result.uint_32 < NFS_MNT_VERSION ||
+		    result.uint_32 > NFS_MNT3_VERSION)
+			goto out_of_bounds;
+		ctx->mount_server.version = result.uint_32;
+		break;
+	case Opt_minorversion:
+		if (result.uint_32 > NFS4_MAX_MINOR_VERSION)
+			goto out_of_bounds;
+		ctx->minorversion = result.uint_32;
+		break;
+
+		/*
+		 * options that take text values
+		 */
+	case Opt_v:
+		ret = nfs_parse_version_string(fc, param->key + 1);
+		if (ret < 0)
+			return ret;
+		break;
+	case Opt_vers:
+		ret = nfs_parse_version_string(fc, param->string);
+		if (ret < 0)
+			return ret;
+		break;
+	case Opt_sec:
+		ret = nfs_parse_security_flavors(fc, param);
+		if (ret < 0)
+			return ret;
+		break;
+
+	case Opt_proto:
+		protofamily = AF_INET;
+		switch (lookup_constant(nfs_xprt_protocol_tokens, param->string, -1)) {
+		case Opt_xprt_udp6:
+			protofamily = AF_INET6;
+			/* fall through */
+		case Opt_xprt_udp:
+			ctx->flags &= ~NFS_MOUNT_TCP;
+			ctx->nfs_server.protocol = XPRT_TRANSPORT_UDP;
+			break;
+		case Opt_xprt_tcp6:
+			protofamily = AF_INET6;
+			/* fall through */
+		case Opt_xprt_tcp:
+			ctx->flags |= NFS_MOUNT_TCP;
+			ctx->nfs_server.protocol = XPRT_TRANSPORT_TCP;
+			break;
+		case Opt_xprt_rdma6:
+			protofamily = AF_INET6;
+			/* fall through */
+		case Opt_xprt_rdma:
+			/* vector side protocols to TCP */
+			ctx->flags |= NFS_MOUNT_TCP;
+			ctx->nfs_server.protocol = XPRT_TRANSPORT_RDMA;
+			xprt_load_transport(param->string);
+			break;
+		default:
+			return nfs_invalf(fc, "NFS: Unrecognized transport protocol");
+		}
+
+		ctx->protofamily = protofamily;
+		break;
+
+	case Opt_mountproto:
+		mountfamily = AF_INET;
+		switch (lookup_constant(nfs_xprt_protocol_tokens, param->string, -1)) {
+		case Opt_xprt_udp6:
+			mountfamily = AF_INET6;
+			/* fall through */
+		case Opt_xprt_udp:
+			ctx->mount_server.protocol = XPRT_TRANSPORT_UDP;
+			break;
+		case Opt_xprt_tcp6:
+			mountfamily = AF_INET6;
+			/* fall through */
+		case Opt_xprt_tcp:
+			ctx->mount_server.protocol = XPRT_TRANSPORT_TCP;
+			break;
+		case Opt_xprt_rdma: /* not used for side protocols */
+		default:
+			return nfs_invalf(fc, "NFS: Unrecognized transport protocol");
+		}
+		ctx->mountfamily = mountfamily;
+		break;
+
+	case Opt_addr:
+		len = rpc_pton(fc->net_ns, param->string, param->size,
+			       &ctx->nfs_server.address,
+			       sizeof(ctx->nfs_server._address));
+		if (len == 0)
+			goto out_invalid_address;
+		ctx->nfs_server.addrlen = len;
+		break;
+	case Opt_clientaddr:
+		kfree(ctx->client_address);
+		ctx->client_address = param->string;
+		param->string = NULL;
+		break;
+	case Opt_mounthost:
+		kfree(ctx->mount_server.hostname);
+		ctx->mount_server.hostname = param->string;
+		param->string = NULL;
+		break;
+	case Opt_mountaddr:
+		len = rpc_pton(fc->net_ns, param->string, param->size,
+			       &ctx->mount_server.address,
+			       sizeof(ctx->mount_server._address));
+		if (len == 0)
+			goto out_invalid_address;
+		ctx->mount_server.addrlen = len;
+		break;
+	case Opt_nconnect:
+		if (result.uint_32 < 1 || result.uint_32 > NFS_MAX_CONNECTIONS)
+			goto out_of_bounds;
+		ctx->nfs_server.nconnect = result.uint_32;
+		break;
+	case Opt_lookupcache:
+		switch (result.uint_32) {
+		case Opt_lookupcache_all:
+			ctx->flags &= ~(NFS_MOUNT_LOOKUP_CACHE_NONEG|NFS_MOUNT_LOOKUP_CACHE_NONE);
+			break;
+		case Opt_lookupcache_positive:
+			ctx->flags &= ~NFS_MOUNT_LOOKUP_CACHE_NONE;
+			ctx->flags |= NFS_MOUNT_LOOKUP_CACHE_NONEG;
+			break;
+		case Opt_lookupcache_none:
+			ctx->flags |= NFS_MOUNT_LOOKUP_CACHE_NONEG|NFS_MOUNT_LOOKUP_CACHE_NONE;
+			break;
+		default:
+			goto out_invalid_value;
+		}
+		break;
+	case Opt_local_lock:
+		switch (result.uint_32) {
+		case Opt_local_lock_all:
+			ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK |
+				       NFS_MOUNT_LOCAL_FCNTL);
+			break;
+		case Opt_local_lock_flock:
+			ctx->flags |= NFS_MOUNT_LOCAL_FLOCK;
+			break;
+		case Opt_local_lock_posix:
+			ctx->flags |= NFS_MOUNT_LOCAL_FCNTL;
+			break;
+		case Opt_local_lock_none:
+			ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK |
+					NFS_MOUNT_LOCAL_FCNTL);
+			break;
+		default:
+			goto out_invalid_value;
+		}
+		break;
+
+		/*
+		 * Special options
+		 */
+	case Opt_sloppy:
+		ctx->sloppy = true;
+		dfprintk(MOUNT, "NFS:   relaxing parsing rules\n");
+		break;
+	}
+
+	return 0;
+
+out_invalid_value:
+	return nfs_invalf(fc, "NFS: Bad mount option value specified");
+out_invalid_address:
+	return nfs_invalf(fc, "NFS: Bad IP address specified");
+out_of_bounds:
+	return nfs_invalf(fc, "NFS: Value for '%s' out of range", param->key);
+}
+
+/*
+ * Split fc->source into "hostname:export_path".
+ *
+ * The leftmost colon demarks the split between the server's hostname
+ * and the export path.  If the hostname starts with a left square
+ * bracket, then it may contain colons.
+ *
+ * Note: caller frees hostname and export path, even on error.
+ */
+static int nfs_parse_source(struct fs_context *fc,
+			    size_t maxnamlen, size_t maxpathlen)
+{
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	const char *dev_name = fc->source;
+	size_t len;
+	const char *end;
+
+	if (unlikely(!dev_name || !*dev_name)) {
+		dfprintk(MOUNT, "NFS: device name not specified\n");
+		return -EINVAL;
+	}
+
+	/* Is the host name protected with square brakcets? */
+	if (*dev_name == '[') {
+		end = strchr(++dev_name, ']');
+		if (end == NULL || end[1] != ':')
+			goto out_bad_devname;
+
+		len = end - dev_name;
+		end++;
+	} else {
+		const char *comma;
+
+		end = strchr(dev_name, ':');
+		if (end == NULL)
+			goto out_bad_devname;
+		len = end - dev_name;
+
+		/* kill possible hostname list: not supported */
+		comma = memchr(dev_name, ',', len);
+		if (comma)
+			len = comma - dev_name;
+	}
+
+	if (len > maxnamlen)
+		goto out_hostname;
+
+	kfree(ctx->nfs_server.hostname);
+
+	/* N.B. caller will free nfs_server.hostname in all cases */
+	ctx->nfs_server.hostname = kmemdup_nul(dev_name, len, GFP_KERNEL);
+	if (!ctx->nfs_server.hostname)
+		goto out_nomem;
+	len = strlen(++end);
+	if (len > maxpathlen)
+		goto out_path;
+	ctx->nfs_server.export_path = kmemdup_nul(end, len, GFP_KERNEL);
+	if (!ctx->nfs_server.export_path)
+		goto out_nomem;
+
+	dfprintk(MOUNT, "NFS: MNTPATH: '%s'\n", ctx->nfs_server.export_path);
+	return 0;
+
+out_bad_devname:
+	return nfs_invalf(fc, "NFS: device name not in host:path format");
+out_nomem:
+	nfs_errorf(fc, "NFS: not enough memory to parse device name");
+	return -ENOMEM;
+out_hostname:
+	nfs_errorf(fc, "NFS: server hostname too long");
+	return -ENAMETOOLONG;
+out_path:
+	nfs_errorf(fc, "NFS: export pathname too long");
+	return -ENAMETOOLONG;
+}
+
+static inline bool is_remount_fc(struct fs_context *fc)
+{
+	return fc->root != NULL;
+}
+
+/*
+ * Parse monolithic NFS2/NFS3 mount data
+ * - fills in the mount root filehandle
+ *
+ * For option strings, user space handles the following behaviors:
+ *
+ * + DNS: mapping server host name to IP address ("addr=" option)
+ *
+ * + failure mode: how to behave if a mount request can't be handled
+ *   immediately ("fg/bg" option)
+ *
+ * + retry: how often to retry a mount request ("retry=" option)
+ *
+ * + breaking back: trying proto=udp after proto=tcp, v2 after v3,
+ *   mountproto=tcp after mountproto=udp, and so on
+ */
+static int nfs23_parse_monolithic(struct fs_context *fc,
+				  struct nfs_mount_data *data)
+{
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct nfs_fh *mntfh = ctx->mntfh;
+	struct sockaddr *sap = (struct sockaddr *)&ctx->nfs_server.address;
+	int extra_flags = NFS_MOUNT_LEGACY_INTERFACE;
+
+	if (data == NULL)
+		goto out_no_data;
+
+	ctx->version = NFS_DEFAULT_VERSION;
+	switch (data->version) {
+	case 1:
+		data->namlen = 0; /* fall through */
+	case 2:
+		data->bsize = 0; /* fall through */
+	case 3:
+		if (data->flags & NFS_MOUNT_VER3)
+			goto out_no_v3;
+		data->root.size = NFS2_FHSIZE;
+		memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE);
+		/* Turn off security negotiation */
+		extra_flags |= NFS_MOUNT_SECFLAVOUR;
+		/* fall through */
+	case 4:
+		if (data->flags & NFS_MOUNT_SECFLAVOUR)
+			goto out_no_sec;
+		/* fall through */
+	case 5:
+		memset(data->context, 0, sizeof(data->context));
+		/* fall through */
+	case 6:
+		if (data->flags & NFS_MOUNT_VER3) {
+			if (data->root.size > NFS3_FHSIZE || data->root.size == 0)
+				goto out_invalid_fh;
+			mntfh->size = data->root.size;
+			ctx->version = 3;
+		} else {
+			mntfh->size = NFS2_FHSIZE;
+			ctx->version = 2;
+		}
+
+
+		memcpy(mntfh->data, data->root.data, mntfh->size);
+		if (mntfh->size < sizeof(mntfh->data))
+			memset(mntfh->data + mntfh->size, 0,
+			       sizeof(mntfh->data) - mntfh->size);
+
+		/*
+		 * Translate to nfs_fs_context, which nfs_fill_super
+		 * can deal with.
+		 */
+		ctx->flags	= data->flags & NFS_MOUNT_FLAGMASK;
+		ctx->flags	|= extra_flags;
+		ctx->rsize	= data->rsize;
+		ctx->wsize	= data->wsize;
+		ctx->timeo	= data->timeo;
+		ctx->retrans	= data->retrans;
+		ctx->acregmin	= data->acregmin;
+		ctx->acregmax	= data->acregmax;
+		ctx->acdirmin	= data->acdirmin;
+		ctx->acdirmax	= data->acdirmax;
+		ctx->need_mount	= false;
+
+		memcpy(sap, &data->addr, sizeof(data->addr));
+		ctx->nfs_server.addrlen = sizeof(data->addr);
+		ctx->nfs_server.port = ntohs(data->addr.sin_port);
+		if (sap->sa_family != AF_INET ||
+		    !nfs_verify_server_address(sap))
+			goto out_no_address;
+
+		if (!(data->flags & NFS_MOUNT_TCP))
+			ctx->nfs_server.protocol = XPRT_TRANSPORT_UDP;
+		/* N.B. caller will free nfs_server.hostname in all cases */
+		ctx->nfs_server.hostname = kstrdup(data->hostname, GFP_KERNEL);
+		if (!ctx->nfs_server.hostname)
+			goto out_nomem;
+
+		ctx->namlen		= data->namlen;
+		ctx->bsize		= data->bsize;
+
+		if (data->flags & NFS_MOUNT_SECFLAVOUR)
+			ctx->selected_flavor = data->pseudoflavor;
+		else
+			ctx->selected_flavor = RPC_AUTH_UNIX;
+
+		if (!(data->flags & NFS_MOUNT_NONLM))
+			ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK|
+					 NFS_MOUNT_LOCAL_FCNTL);
+		else
+			ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK|
+					NFS_MOUNT_LOCAL_FCNTL);
+
+		/*
+		 * The legacy version 6 binary mount data from userspace has a
+		 * field used only to transport selinux information into the
+		 * the kernel.  To continue to support that functionality we
+		 * have a touch of selinux knowledge here in the NFS code. The
+		 * userspace code converted context=blah to just blah so we are
+		 * converting back to the full string selinux understands.
+		 */
+		if (data->context[0]){
+#ifdef CONFIG_SECURITY_SELINUX
+			int ret;
+
+			data->context[NFS_MAX_CONTEXT_LEN] = '\0';
+			ret = vfs_parse_fs_string(fc, "context",
+						  data->context, strlen(data->context));
+			if (ret < 0)
+				return ret;
+#else
+			return -EINVAL;
+#endif
+		}
+
+		break;
+	default:
+		goto generic;
+	}
+
+	ctx->skip_reconfig_option_check = true;
+	return 0;
+
+generic:
+	return generic_parse_monolithic(fc, data);
+
+out_no_data:
+	if (is_remount_fc(fc)) {
+		ctx->skip_reconfig_option_check = true;
+		return 0;
+	}
+	return nfs_invalf(fc, "NFS: mount program didn't pass any mount data");
+
+out_no_v3:
+	return nfs_invalf(fc, "NFS: nfs_mount_data version does not support v3");
+
+out_no_sec:
+	return nfs_invalf(fc, "NFS: nfs_mount_data version supports only AUTH_SYS");
+
+out_nomem:
+	dfprintk(MOUNT, "NFS: not enough memory to handle mount options");
+	return -ENOMEM;
+
+out_no_address:
+	return nfs_invalf(fc, "NFS: mount program didn't pass remote address");
+
+out_invalid_fh:
+	return nfs_invalf(fc, "NFS: invalid root filehandle");
+}
+
+#if IS_ENABLED(CONFIG_NFS_V4)
+/*
+ * Validate NFSv4 mount options
+ */
+static int nfs4_parse_monolithic(struct fs_context *fc,
+				 struct nfs4_mount_data *data)
+{
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct sockaddr *sap = (struct sockaddr *)&ctx->nfs_server.address;
+	char *c;
+
+	if (data == NULL)
+		goto out_no_data;
+
+	ctx->version = 4;
+
+	switch (data->version) {
+	case 1:
+		if (data->host_addrlen > sizeof(ctx->nfs_server.address))
+			goto out_no_address;
+		if (data->host_addrlen == 0)
+			goto out_no_address;
+		ctx->nfs_server.addrlen = data->host_addrlen;
+		if (copy_from_user(sap, data->host_addr, data->host_addrlen))
+			return -EFAULT;
+		if (!nfs_verify_server_address(sap))
+			goto out_no_address;
+		ctx->nfs_server.port = ntohs(((struct sockaddr_in *)sap)->sin_port);
+
+		if (data->auth_flavourlen) {
+			rpc_authflavor_t pseudoflavor;
+			if (data->auth_flavourlen > 1)
+				goto out_inval_auth;
+			if (copy_from_user(&pseudoflavor,
+					   data->auth_flavours,
+					   sizeof(pseudoflavor)))
+				return -EFAULT;
+			ctx->selected_flavor = pseudoflavor;
+		} else
+			ctx->selected_flavor = RPC_AUTH_UNIX;
+
+		c = strndup_user(data->hostname.data, NFS4_MAXNAMLEN);
+		if (IS_ERR(c))
+			return PTR_ERR(c);
+		ctx->nfs_server.hostname = c;
+
+		c = strndup_user(data->mnt_path.data, NFS4_MAXPATHLEN);
+		if (IS_ERR(c))
+			return PTR_ERR(c);
+		ctx->nfs_server.export_path = c;
+		dfprintk(MOUNT, "NFS: MNTPATH: '%s'\n", c);
+
+		c = strndup_user(data->client_addr.data, 16);
+		if (IS_ERR(c))
+			return PTR_ERR(c);
+		ctx->client_address = c;
+
+		/*
+		 * Translate to nfs_fs_context, which nfs_fill_super
+		 * can deal with.
+		 */
+
+		ctx->flags	= data->flags & NFS4_MOUNT_FLAGMASK;
+		ctx->rsize	= data->rsize;
+		ctx->wsize	= data->wsize;
+		ctx->timeo	= data->timeo;
+		ctx->retrans	= data->retrans;
+		ctx->acregmin	= data->acregmin;
+		ctx->acregmax	= data->acregmax;
+		ctx->acdirmin	= data->acdirmin;
+		ctx->acdirmax	= data->acdirmax;
+		ctx->nfs_server.protocol = data->proto;
+		nfs_validate_transport_protocol(ctx);
+		if (ctx->nfs_server.protocol == XPRT_TRANSPORT_UDP)
+			goto out_invalid_transport_udp;
+
+		break;
+	default:
+		goto generic;
+	}
+
+	ctx->skip_reconfig_option_check = true;
+	return 0;
+
+generic:
+	return generic_parse_monolithic(fc, data);
+
+out_no_data:
+	if (is_remount_fc(fc)) {
+		ctx->skip_reconfig_option_check = true;
+		return 0;
+	}
+	return nfs_invalf(fc, "NFS4: mount program didn't pass any mount data");
+
+out_inval_auth:
+	return nfs_invalf(fc, "NFS4: Invalid number of RPC auth flavours %d",
+		      data->auth_flavourlen);
+
+out_no_address:
+	return nfs_invalf(fc, "NFS4: mount program didn't pass remote address");
+
+out_invalid_transport_udp:
+	return nfs_invalf(fc, "NFS: Unsupported transport protocol udp");
+}
+#endif
+
+/*
+ * Parse a monolithic block of data from sys_mount().
+ */
+static int nfs_fs_context_parse_monolithic(struct fs_context *fc,
+					   void *data)
+{
+	if (fc->fs_type == &nfs_fs_type)
+		return nfs23_parse_monolithic(fc, data);
+
+#if IS_ENABLED(CONFIG_NFS_V4)
+	if (fc->fs_type == &nfs4_fs_type)
+		return nfs4_parse_monolithic(fc, data);
+#endif
+
+	return nfs_invalf(fc, "NFS: Unsupported monolithic data version");
+}
+
+/*
+ * Validate the preparsed information in the config.
+ */
+static int nfs_fs_context_validate(struct fs_context *fc)
+{
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct nfs_subversion *nfs_mod;
+	struct sockaddr *sap = (struct sockaddr *)&ctx->nfs_server.address;
+	int max_namelen = PAGE_SIZE;
+	int max_pathlen = NFS_MAXPATHLEN;
+	int port = 0;
+	int ret;
+
+	if (!fc->source)
+		goto out_no_device_name;
+
+	/* Check for sanity first. */
+	if (ctx->minorversion && ctx->version != 4)
+		goto out_minorversion_mismatch;
+
+	if (ctx->options & NFS_OPTION_MIGRATION &&
+	    (ctx->version != 4 || ctx->minorversion != 0))
+		goto out_migration_misuse;
+
+	/* Verify that any proto=/mountproto= options match the address
+	 * families in the addr=/mountaddr= options.
+	 */
+	if (ctx->protofamily != AF_UNSPEC &&
+	    ctx->protofamily != ctx->nfs_server.address.sa_family)
+		goto out_proto_mismatch;
+
+	if (ctx->mountfamily != AF_UNSPEC) {
+		if (ctx->mount_server.addrlen) {
+			if (ctx->mountfamily != ctx->mount_server.address.sa_family)
+				goto out_mountproto_mismatch;
+		} else {
+			if (ctx->mountfamily != ctx->nfs_server.address.sa_family)
+				goto out_mountproto_mismatch;
+		}
+	}
+
+	if (!nfs_verify_server_address(sap))
+		goto out_no_address;
+
+	if (ctx->version == 4) {
+		if (IS_ENABLED(CONFIG_NFS_V4)) {
+			if (ctx->nfs_server.protocol == XPRT_TRANSPORT_RDMA)
+				port = NFS_RDMA_PORT;
+			else
+				port = NFS_PORT;
+			max_namelen = NFS4_MAXNAMLEN;
+			max_pathlen = NFS4_MAXPATHLEN;
+			nfs_validate_transport_protocol(ctx);
+			if (ctx->nfs_server.protocol == XPRT_TRANSPORT_UDP)
+				goto out_invalid_transport_udp;
+			ctx->flags &= ~(NFS_MOUNT_NONLM | NFS_MOUNT_NOACL |
+					NFS_MOUNT_VER3 | NFS_MOUNT_LOCAL_FLOCK |
+					NFS_MOUNT_LOCAL_FCNTL);
+		} else {
+			goto out_v4_not_compiled;
+		}
+	} else {
+		nfs_set_mount_transport_protocol(ctx);
+#ifdef CONFIG_NFS_DISABLE_UDP_SUPPORT
+	       if (ctx->nfs_server.protocol == XPRT_TRANSPORT_UDP)
+		       goto out_invalid_transport_udp;
+#endif
+		if (ctx->nfs_server.protocol == XPRT_TRANSPORT_RDMA)
+			port = NFS_RDMA_PORT;
+	}
+
+	nfs_set_port(sap, &ctx->nfs_server.port, port);
+
+	ret = nfs_parse_source(fc, max_namelen, max_pathlen);
+	if (ret < 0)
+		return ret;
+
+	/* Load the NFS protocol module if we haven't done so yet */
+	if (!ctx->nfs_mod) {
+		nfs_mod = get_nfs_version(ctx->version);
+		if (IS_ERR(nfs_mod)) {
+			ret = PTR_ERR(nfs_mod);
+			goto out_version_unavailable;
+		}
+		ctx->nfs_mod = nfs_mod;
+	}
+
+	/* Ensure the filesystem context has the correct fs_type */
+	if (fc->fs_type != ctx->nfs_mod->nfs_fs) {
+		module_put(fc->fs_type->owner);
+		__module_get(ctx->nfs_mod->nfs_fs->owner);
+		fc->fs_type = ctx->nfs_mod->nfs_fs;
+	}
+	return 0;
+
+out_no_device_name:
+	return nfs_invalf(fc, "NFS: Device name not specified");
+out_v4_not_compiled:
+	nfs_errorf(fc, "NFS: NFSv4 is not compiled into kernel");
+	return -EPROTONOSUPPORT;
+out_invalid_transport_udp:
+	return nfs_invalf(fc, "NFS: Unsupported transport protocol udp");
+out_no_address:
+	return nfs_invalf(fc, "NFS: mount program didn't pass remote address");
+out_mountproto_mismatch:
+	return nfs_invalf(fc, "NFS: Mount server address does not match mountproto= option");
+out_proto_mismatch:
+	return nfs_invalf(fc, "NFS: Server address does not match proto= option");
+out_minorversion_mismatch:
+	return nfs_invalf(fc, "NFS: Mount option vers=%u does not support minorversion=%u",
+			  ctx->version, ctx->minorversion);
+out_migration_misuse:
+	return nfs_invalf(fc, "NFS: 'Migration' not supported for this NFS version");
+out_version_unavailable:
+	nfs_errorf(fc, "NFS: Version unavailable");
+	return ret;
+}
+
+/*
+ * Create an NFS superblock by the appropriate method.
+ */
+static int nfs_get_tree(struct fs_context *fc)
+{
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	int err = nfs_fs_context_validate(fc);
+
+	if (err)
+		return err;
+	if (!ctx->internal)
+		return ctx->nfs_mod->rpc_ops->try_get_tree(fc);
+	else
+		return nfs_get_tree_common(fc);
+}
+
+/*
+ * Handle duplication of a configuration.  The caller copied *src into *sc, but
+ * it can't deal with resource pointers in the filesystem context, so we have
+ * to do that.  We need to clear pointers, copy data or get extra refs as
+ * appropriate.
+ */
+static int nfs_fs_context_dup(struct fs_context *fc, struct fs_context *src_fc)
+{
+	struct nfs_fs_context *src = nfs_fc2context(src_fc), *ctx;
+
+	ctx = kmemdup(src, sizeof(struct nfs_fs_context), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	ctx->mntfh = nfs_alloc_fhandle();
+	if (!ctx->mntfh) {
+		kfree(ctx);
+		return -ENOMEM;
+	}
+	nfs_copy_fh(ctx->mntfh, src->mntfh);
+
+	__module_get(ctx->nfs_mod->owner);
+	ctx->client_address		= NULL;
+	ctx->mount_server.hostname	= NULL;
+	ctx->nfs_server.export_path	= NULL;
+	ctx->nfs_server.hostname	= NULL;
+	ctx->fscache_uniq		= NULL;
+	ctx->clone_data.fattr		= NULL;
+	fc->fs_private = ctx;
+	return 0;
+}
+
+static void nfs_fs_context_free(struct fs_context *fc)
+{
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+
+	if (ctx) {
+		if (ctx->server)
+			nfs_free_server(ctx->server);
+		if (ctx->nfs_mod)
+			put_nfs_version(ctx->nfs_mod);
+		kfree(ctx->client_address);
+		kfree(ctx->mount_server.hostname);
+		kfree(ctx->nfs_server.export_path);
+		kfree(ctx->nfs_server.hostname);
+		kfree(ctx->fscache_uniq);
+		nfs_free_fhandle(ctx->mntfh);
+		nfs_free_fattr(ctx->clone_data.fattr);
+		kfree(ctx);
+	}
+}
+
+static const struct fs_context_operations nfs_fs_context_ops = {
+	.free			= nfs_fs_context_free,
+	.dup			= nfs_fs_context_dup,
+	.parse_param		= nfs_fs_context_parse_param,
+	.parse_monolithic	= nfs_fs_context_parse_monolithic,
+	.get_tree		= nfs_get_tree,
+	.reconfigure		= nfs_reconfigure,
+};
+
+/*
+ * Prepare superblock configuration.  We use the namespaces attached to the
+ * context.  This may be the current process's namespaces, or it may be a
+ * container's namespaces.
+ */
+static int nfs_init_fs_context(struct fs_context *fc)
+{
+	struct nfs_fs_context *ctx;
+
+	ctx = kzalloc(sizeof(struct nfs_fs_context), GFP_KERNEL);
+	if (unlikely(!ctx))
+		return -ENOMEM;
+
+	ctx->mntfh = nfs_alloc_fhandle();
+	if (unlikely(!ctx->mntfh)) {
+		kfree(ctx);
+		return -ENOMEM;
+	}
+
+	ctx->protofamily	= AF_UNSPEC;
+	ctx->mountfamily	= AF_UNSPEC;
+	ctx->mount_server.port	= NFS_UNSPEC_PORT;
+
+	if (fc->root) {
+		/* reconfigure, start with the current config */
+		struct nfs_server *nfss = fc->root->d_sb->s_fs_info;
+		struct net *net = nfss->nfs_client->cl_net;
+
+		ctx->flags		= nfss->flags;
+		ctx->rsize		= nfss->rsize;
+		ctx->wsize		= nfss->wsize;
+		ctx->retrans		= nfss->client->cl_timeout->to_retries;
+		ctx->selected_flavor	= nfss->client->cl_auth->au_flavor;
+		ctx->acregmin		= nfss->acregmin / HZ;
+		ctx->acregmax		= nfss->acregmax / HZ;
+		ctx->acdirmin		= nfss->acdirmin / HZ;
+		ctx->acdirmax		= nfss->acdirmax / HZ;
+		ctx->timeo		= 10U * nfss->client->cl_timeout->to_initval / HZ;
+		ctx->nfs_server.port	= nfss->port;
+		ctx->nfs_server.addrlen	= nfss->nfs_client->cl_addrlen;
+		ctx->version		= nfss->nfs_client->rpc_ops->version;
+		ctx->minorversion	= nfss->nfs_client->cl_minorversion;
+
+		memcpy(&ctx->nfs_server.address, &nfss->nfs_client->cl_addr,
+			ctx->nfs_server.addrlen);
+
+		if (fc->net_ns != net) {
+			put_net(fc->net_ns);
+			fc->net_ns = get_net(net);
+		}
+
+		ctx->nfs_mod = nfss->nfs_client->cl_nfs_mod;
+		__module_get(ctx->nfs_mod->owner);
+	} else {
+		/* defaults */
+		ctx->timeo		= NFS_UNSPEC_TIMEO;
+		ctx->retrans		= NFS_UNSPEC_RETRANS;
+		ctx->acregmin		= NFS_DEF_ACREGMIN;
+		ctx->acregmax		= NFS_DEF_ACREGMAX;
+		ctx->acdirmin		= NFS_DEF_ACDIRMIN;
+		ctx->acdirmax		= NFS_DEF_ACDIRMAX;
+		ctx->nfs_server.port	= NFS_UNSPEC_PORT;
+		ctx->nfs_server.protocol = XPRT_TRANSPORT_TCP;
+		ctx->selected_flavor	= RPC_AUTH_MAXFLAVOR;
+		ctx->minorversion	= 0;
+		ctx->need_mount		= true;
+	}
+	fc->fs_private = ctx;
+	fc->ops = &nfs_fs_context_ops;
+	return 0;
+}
+
+struct file_system_type nfs_fs_type = {
+	.owner			= THIS_MODULE,
+	.name			= "nfs",
+	.init_fs_context	= nfs_init_fs_context,
+	.parameters		= nfs_fs_parameters,
+	.kill_sb		= nfs_kill_super,
+	.fs_flags		= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
+};
+MODULE_ALIAS_FS("nfs");
+EXPORT_SYMBOL_GPL(nfs_fs_type);
+
+#if IS_ENABLED(CONFIG_NFS_V4)
+struct file_system_type nfs4_fs_type = {
+	.owner			= THIS_MODULE,
+	.name			= "nfs4",
+	.init_fs_context	= nfs_init_fs_context,
+	.parameters		= nfs_fs_parameters,
+	.kill_sb		= nfs_kill_super,
+	.fs_flags		= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
+};
+MODULE_ALIAS_FS("nfs4");
+MODULE_ALIAS("nfs4");
+EXPORT_SYMBOL_GPL(nfs4_fs_type);
+#endif /* CONFIG_NFS_V4 */
diff --git a/fs/nfs/fscache-index.c b/fs/nfs/fscache-index.c
index 15f271401dcc..573b1da9342c 100644
--- a/fs/nfs/fscache-index.c
+++ b/fs/nfs/fscache-index.c
@@ -84,8 +84,10 @@ enum fscache_checkaux nfs_fscache_inode_check_aux(void *cookie_netfs_data,
 		return FSCACHE_CHECKAUX_OBSOLETE;
 
 	memset(&auxdata, 0, sizeof(auxdata));
-	auxdata.mtime = timespec64_to_timespec(nfsi->vfs_inode.i_mtime);
-	auxdata.ctime = timespec64_to_timespec(nfsi->vfs_inode.i_ctime);
+	auxdata.mtime_sec  = nfsi->vfs_inode.i_mtime.tv_sec;
+	auxdata.mtime_nsec = nfsi->vfs_inode.i_mtime.tv_nsec;
+	auxdata.ctime_sec  = nfsi->vfs_inode.i_ctime.tv_sec;
+	auxdata.ctime_nsec = nfsi->vfs_inode.i_ctime.tv_nsec;
 
 	if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4)
 		auxdata.change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode);
diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c
index 3800ab6f08fa..a60df88efc40 100644
--- a/fs/nfs/fscache.c
+++ b/fs/nfs/fscache.c
@@ -31,6 +31,7 @@ static DEFINE_SPINLOCK(nfs_fscache_keys_lock);
 struct nfs_server_key {
 	struct {
 		uint16_t	nfsversion;		/* NFS protocol version */
+		uint32_t	minorversion;		/* NFSv4 minor version */
 		uint16_t	family;			/* address family */
 		__be16		port;			/* IP port */
 	} hdr;
@@ -55,6 +56,7 @@ void nfs_fscache_get_client_cookie(struct nfs_client *clp)
 
 	memset(&key, 0, sizeof(key));
 	key.hdr.nfsversion = clp->rpc_ops->version;
+	key.hdr.minorversion = clp->cl_minorversion;
 	key.hdr.family = clp->cl_addr.ss_family;
 
 	switch (clp->cl_addr.ss_family) {
@@ -116,8 +118,6 @@ void nfs_fscache_get_super_cookie(struct super_block *sb, const char *uniq, int
 
 	nfss->fscache_key = NULL;
 	nfss->fscache = NULL;
-	if (!(nfss->options & NFS_OPTION_FSCACHE))
-		return;
 	if (!uniq) {
 		uniq = "";
 		ulen = 1;
@@ -128,7 +128,7 @@ void nfs_fscache_get_super_cookie(struct super_block *sb, const char *uniq, int
 		return;
 
 	key->nfs_client = nfss->nfs_client;
-	key->key.super.s_flags = sb->s_flags & NFS_MS_MASK;
+	key->key.super.s_flags = sb->s_flags & NFS_SB_MASK;
 	key->key.nfs_server.flags = nfss->flags;
 	key->key.nfs_server.rsize = nfss->rsize;
 	key->key.nfs_server.wsize = nfss->wsize;
@@ -186,7 +186,8 @@ void nfs_fscache_get_super_cookie(struct super_block *sb, const char *uniq, int
 	/* create a cache index for looking up filehandles */
 	nfss->fscache = fscache_acquire_cookie(nfss->nfs_client->fscache,
 					       &nfs_fscache_super_index_def,
-					       key, sizeof(*key) + ulen,
+					       &key->key,
+					       sizeof(key->key) + ulen,
 					       NULL, 0,
 					       nfss, 0, true);
 	dfprintk(FSCACHE, "NFS: get superblock cookie (0x%p/0x%p)\n",
@@ -224,6 +225,19 @@ void nfs_fscache_release_super_cookie(struct super_block *sb)
 	}
 }
 
+static void nfs_fscache_update_auxdata(struct nfs_fscache_inode_auxdata *auxdata,
+				  struct nfs_inode *nfsi)
+{
+	memset(auxdata, 0, sizeof(*auxdata));
+	auxdata->mtime_sec  = nfsi->vfs_inode.i_mtime.tv_sec;
+	auxdata->mtime_nsec = nfsi->vfs_inode.i_mtime.tv_nsec;
+	auxdata->ctime_sec  = nfsi->vfs_inode.i_ctime.tv_sec;
+	auxdata->ctime_nsec = nfsi->vfs_inode.i_ctime.tv_nsec;
+
+	if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4)
+		auxdata->change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode);
+}
+
 /*
  * Initialise the per-inode cache cookie pointer for an NFS inode.
  */
@@ -237,12 +251,7 @@ void nfs_fscache_init_inode(struct inode *inode)
 	if (!(nfss->fscache && S_ISREG(inode->i_mode)))
 		return;
 
-	memset(&auxdata, 0, sizeof(auxdata));
-	auxdata.mtime = timespec64_to_timespec(nfsi->vfs_inode.i_mtime);
-	auxdata.ctime = timespec64_to_timespec(nfsi->vfs_inode.i_ctime);
-
-	if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4)
-		auxdata.change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode);
+	nfs_fscache_update_auxdata(&auxdata, nfsi);
 
 	nfsi->fscache = fscache_acquire_cookie(NFS_SB(inode->i_sb)->fscache,
 					       &nfs_fscache_inode_object_def,
@@ -262,9 +271,7 @@ void nfs_fscache_clear_inode(struct inode *inode)
 
 	dfprintk(FSCACHE, "NFS: clear cookie (0x%p/0x%p)\n", nfsi, cookie);
 
-	memset(&auxdata, 0, sizeof(auxdata));
-	auxdata.mtime = timespec64_to_timespec(nfsi->vfs_inode.i_mtime);
-	auxdata.ctime = timespec64_to_timespec(nfsi->vfs_inode.i_ctime);
+	nfs_fscache_update_auxdata(&auxdata, nfsi);
 	fscache_relinquish_cookie(cookie, &auxdata, false);
 	nfsi->fscache = NULL;
 }
@@ -304,9 +311,7 @@ void nfs_fscache_open_file(struct inode *inode, struct file *filp)
 	if (!fscache_cookie_valid(cookie))
 		return;
 
-	memset(&auxdata, 0, sizeof(auxdata));
-	auxdata.mtime = timespec64_to_timespec(nfsi->vfs_inode.i_mtime);
-	auxdata.ctime = timespec64_to_timespec(nfsi->vfs_inode.i_ctime);
+	nfs_fscache_update_auxdata(&auxdata, nfsi);
 
 	if (inode_is_open_for_write(inode)) {
 		dfprintk(FSCACHE, "NFS: nfsi 0x%p disabling cache\n", nfsi);
diff --git a/fs/nfs/fscache.h b/fs/nfs/fscache.h
index ad041cfbf9ec..6754c8607230 100644
--- a/fs/nfs/fscache.h
+++ b/fs/nfs/fscache.h
@@ -62,9 +62,11 @@ struct nfs_fscache_key {
  * cache object.
  */
 struct nfs_fscache_inode_auxdata {
-	struct timespec	mtime;
-	struct timespec	ctime;
-	u64		change_attr;
+	s64	mtime_sec;
+	s64	mtime_nsec;
+	s64	ctime_sec;
+	s64	ctime_nsec;
+	u64	change_attr;
 };
 
 /*
diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c
index 878c4c5982d9..aaeeb4659bff 100644
--- a/fs/nfs/getroot.c
+++ b/fs/nfs/getroot.c
@@ -64,66 +64,102 @@ static int nfs_superblock_set_dummy_root(struct super_block *sb, struct inode *i
 /*
  * get an NFS2/NFS3 root dentry from the root filehandle
  */
-struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh,
-			    const char *devname)
+int nfs_get_root(struct super_block *s, struct fs_context *fc)
 {
-	struct nfs_server *server = NFS_SB(sb);
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct nfs_server *server = NFS_SB(s);
 	struct nfs_fsinfo fsinfo;
-	struct dentry *ret;
+	struct dentry *root;
 	struct inode *inode;
-	void *name = kstrdup(devname, GFP_KERNEL);
-	int error;
+	char *name;
+	int error = -ENOMEM;
+	unsigned long kflags = 0, kflags_out = 0;
 
+	name = kstrdup(fc->source, GFP_KERNEL);
 	if (!name)
-		return ERR_PTR(-ENOMEM);
+		goto out;
 
 	/* get the actual root for this mount */
 	fsinfo.fattr = nfs_alloc_fattr();
-	if (fsinfo.fattr == NULL) {
-		kfree(name);
-		return ERR_PTR(-ENOMEM);
-	}
+	if (fsinfo.fattr == NULL)
+		goto out_name;
 
-	error = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo);
+	fsinfo.fattr->label = nfs4_label_alloc(server, GFP_KERNEL);
+	if (IS_ERR(fsinfo.fattr->label))
+		goto out_fattr;
+	error = server->nfs_client->rpc_ops->getroot(server, ctx->mntfh, &fsinfo);
 	if (error < 0) {
 		dprintk("nfs_get_root: getattr error = %d\n", -error);
-		ret = ERR_PTR(error);
-		goto out;
+		nfs_errorf(fc, "NFS: Couldn't getattr on root");
+		goto out_label;
 	}
 
-	inode = nfs_fhget(sb, mntfh, fsinfo.fattr, NULL);
+	inode = nfs_fhget(s, ctx->mntfh, fsinfo.fattr, NULL);
 	if (IS_ERR(inode)) {
 		dprintk("nfs_get_root: get root inode failed\n");
-		ret = ERR_CAST(inode);
-		goto out;
+		error = PTR_ERR(inode);
+		nfs_errorf(fc, "NFS: Couldn't get root inode");
+		goto out_label;
 	}
 
-	error = nfs_superblock_set_dummy_root(sb, inode);
-	if (error != 0) {
-		ret = ERR_PTR(error);
-		goto out;
-	}
+	error = nfs_superblock_set_dummy_root(s, inode);
+	if (error != 0)
+		goto out_label;
 
 	/* root dentries normally start off anonymous and get spliced in later
 	 * if the dentry tree reaches them; however if the dentry already
 	 * exists, we'll pick it up at this point and use it as the root
 	 */
-	ret = d_obtain_root(inode);
-	if (IS_ERR(ret)) {
+	root = d_obtain_root(inode);
+	if (IS_ERR(root)) {
 		dprintk("nfs_get_root: get root dentry failed\n");
-		goto out;
+		error = PTR_ERR(root);
+		nfs_errorf(fc, "NFS: Couldn't get root dentry");
+		goto out_label;
 	}
 
-	security_d_instantiate(ret, inode);
-	spin_lock(&ret->d_lock);
-	if (IS_ROOT(ret) && !ret->d_fsdata &&
-	    !(ret->d_flags & DCACHE_NFSFS_RENAMED)) {
-		ret->d_fsdata = name;
+	security_d_instantiate(root, inode);
+	spin_lock(&root->d_lock);
+	if (IS_ROOT(root) && !root->d_fsdata &&
+	    !(root->d_flags & DCACHE_NFSFS_RENAMED)) {
+		root->d_fsdata = name;
 		name = NULL;
 	}
-	spin_unlock(&ret->d_lock);
-out:
-	kfree(name);
+	spin_unlock(&root->d_lock);
+	fc->root = root;
+	if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL)
+		kflags |= SECURITY_LSM_NATIVE_LABELS;
+	if (ctx->clone_data.sb) {
+		if (d_inode(fc->root)->i_fop != &nfs_dir_operations) {
+			error = -ESTALE;
+			goto error_splat_root;
+		}
+		/* clone lsm security options from the parent to the new sb */
+		error = security_sb_clone_mnt_opts(ctx->clone_data.sb,
+						   s, kflags, &kflags_out);
+	} else {
+		error = security_sb_set_mnt_opts(s, fc->security,
+							kflags, &kflags_out);
+	}
+	if (error)
+		goto error_splat_root;
+	if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL &&
+		!(kflags_out & SECURITY_LSM_NATIVE_LABELS))
+		NFS_SB(s)->caps &= ~NFS_CAP_SECURITY_LABEL;
+
+	nfs_setsecurity(inode, fsinfo.fattr, fsinfo.fattr->label);
+	error = 0;
+
+out_label:
+	nfs4_label_free(fsinfo.fattr->label);
+out_fattr:
 	nfs_free_fattr(fsinfo.fattr);
-	return ret;
+out_name:
+	kfree(name);
+out:
+	return error;
+error_splat_root:
+	dput(fc->root);
+	fc->root = NULL;
+	goto out_label;
 }
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 2a03bfeec10a..b9d0921cb4fe 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -62,7 +62,6 @@
 /* Default is to see 64-bit inode numbers */
 static bool enable_ino64 = NFS_64_BIT_INODE_NUMBERS_ENABLED;
 
-static void nfs_invalidate_inode(struct inode *);
 static int nfs_update_inode(struct inode *, struct nfs_fattr *);
 
 static struct kmem_cache * nfs_inode_cachep;
@@ -284,10 +283,18 @@ EXPORT_SYMBOL_GPL(nfs_invalidate_atime);
  * Invalidate, but do not unhash, the inode.
  * NB: must be called with inode->i_lock held!
  */
-static void nfs_invalidate_inode(struct inode *inode)
+static void nfs_set_inode_stale_locked(struct inode *inode)
 {
 	set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
 	nfs_zap_caches_locked(inode);
+	trace_nfs_set_inode_stale(inode);
+}
+
+void nfs_set_inode_stale(struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	nfs_set_inode_stale_locked(inode);
+	spin_unlock(&inode->i_lock);
 }
 
 struct nfs_find_desc {
@@ -504,15 +511,15 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st
 		nfsi->read_cache_jiffies = fattr->time_start;
 		nfsi->attr_gencount = fattr->gencount;
 		if (fattr->valid & NFS_ATTR_FATTR_ATIME)
-			inode->i_atime = timespec_to_timespec64(fattr->atime);
+			inode->i_atime = fattr->atime;
 		else if (nfs_server_capable(inode, NFS_CAP_ATIME))
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME);
 		if (fattr->valid & NFS_ATTR_FATTR_MTIME)
-			inode->i_mtime = timespec_to_timespec64(fattr->mtime);
+			inode->i_mtime = fattr->mtime;
 		else if (nfs_server_capable(inode, NFS_CAP_MTIME))
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME);
 		if (fattr->valid & NFS_ATTR_FATTR_CTIME)
-			inode->i_ctime = timespec_to_timespec64(fattr->ctime);
+			inode->i_ctime = fattr->ctime;
 		else if (nfs_server_capable(inode, NFS_CAP_CTIME))
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_CTIME);
 		if (fattr->valid & NFS_ATTR_FATTR_CHANGE)
@@ -698,7 +705,7 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
 		if ((attr->ia_valid & ATTR_GID) != 0)
 			inode->i_gid = attr->ia_gid;
 		if (fattr->valid & NFS_ATTR_FATTR_CTIME)
-			inode->i_ctime = timespec_to_timespec64(fattr->ctime);
+			inode->i_ctime = fattr->ctime;
 		else
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE
 					| NFS_INO_INVALID_CTIME);
@@ -709,14 +716,14 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
 		NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_ATIME
 				| NFS_INO_INVALID_CTIME);
 		if (fattr->valid & NFS_ATTR_FATTR_ATIME)
-			inode->i_atime = timespec_to_timespec64(fattr->atime);
+			inode->i_atime = fattr->atime;
 		else if (attr->ia_valid & ATTR_ATIME_SET)
 			inode->i_atime = attr->ia_atime;
 		else
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME);
 
 		if (fattr->valid & NFS_ATTR_FATTR_CTIME)
-			inode->i_ctime = timespec_to_timespec64(fattr->ctime);
+			inode->i_ctime = fattr->ctime;
 		else
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE
 					| NFS_INO_INVALID_CTIME);
@@ -725,14 +732,14 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
 		NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_MTIME
 				| NFS_INO_INVALID_CTIME);
 		if (fattr->valid & NFS_ATTR_FATTR_MTIME)
-			inode->i_mtime = timespec_to_timespec64(fattr->mtime);
+			inode->i_mtime = fattr->mtime;
 		else if (attr->ia_valid & ATTR_MTIME_SET)
 			inode->i_mtime = attr->ia_mtime;
 		else
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME);
 
 		if (fattr->valid & NFS_ATTR_FATTR_CTIME)
-			inode->i_ctime = timespec_to_timespec64(fattr->ctime);
+			inode->i_ctime = fattr->ctime;
 		else
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE
 					| NFS_INO_INVALID_CTIME);
@@ -959,16 +966,16 @@ struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry,
 						struct file *filp)
 {
 	struct nfs_open_context *ctx;
-	const struct cred *cred = get_current_cred();
 
 	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
-	if (!ctx) {
-		put_cred(cred);
+	if (!ctx)
 		return ERR_PTR(-ENOMEM);
-	}
 	nfs_sb_active(dentry->d_sb);
 	ctx->dentry = dget(dentry);
-	ctx->cred = cred;
+	if (filp)
+		ctx->cred = get_cred(filp->f_cred);
+	else
+		ctx->cred = get_current_cred();
 	ctx->ll_cred = NULL;
 	ctx->state = NULL;
 	ctx->mode = f_mode;
@@ -1061,7 +1068,7 @@ struct nfs_open_context *nfs_find_open_context(struct inode *inode, const struct
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(pos, &nfsi->open_files, list) {
-		if (cred != NULL && pos->cred != cred)
+		if (cred != NULL && cred_fscmp(pos->cred, cred) != 0)
 			continue;
 		if ((pos->mode & (FMODE_READ|FMODE_WRITE)) != mode)
 			continue;
@@ -1156,10 +1163,17 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
 		dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Lu) getattr failed, error=%d\n",
 			 inode->i_sb->s_id,
 			 (unsigned long long)NFS_FILEID(inode), status);
-		if (status == -ESTALE) {
-			nfs_zap_caches(inode);
+		switch (status) {
+		case -ETIMEDOUT:
+			/* A soft timeout occurred. Use cached information? */
+			if (server->flags & NFS_MOUNT_SOFTREVAL)
+				status = 0;
+			break;
+		case -ESTALE:
 			if (!S_ISDIR(inode->i_mode))
-				set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
+				nfs_set_inode_stale(inode);
+			else
+				nfs_zap_caches(inode);
 		}
 		goto err_out;
 	}
@@ -1351,7 +1365,7 @@ static bool nfs_file_has_buffered_writers(struct nfs_inode *nfsi)
 
 static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 {
-	struct timespec ts;
+	struct timespec64 ts;
 
 	if ((fattr->valid & NFS_ATTR_FATTR_PRECHANGE)
 			&& (fattr->valid & NFS_ATTR_FATTR_CHANGE)
@@ -1361,18 +1375,18 @@ static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA);
 	}
 	/* If we have atomic WCC data, we may update some attributes */
-	ts = timespec64_to_timespec(inode->i_ctime);
+	ts = inode->i_ctime;
 	if ((fattr->valid & NFS_ATTR_FATTR_PRECTIME)
 			&& (fattr->valid & NFS_ATTR_FATTR_CTIME)
-			&& timespec_equal(&ts, &fattr->pre_ctime)) {
-		inode->i_ctime = timespec_to_timespec64(fattr->ctime);
+			&& timespec64_equal(&ts, &fattr->pre_ctime)) {
+		inode->i_ctime = fattr->ctime;
 	}
 
-	ts = timespec64_to_timespec(inode->i_mtime);
+	ts = inode->i_mtime;
 	if ((fattr->valid & NFS_ATTR_FATTR_PREMTIME)
 			&& (fattr->valid & NFS_ATTR_FATTR_MTIME)
-			&& timespec_equal(&ts, &fattr->pre_mtime)) {
-		inode->i_mtime = timespec_to_timespec64(fattr->mtime);
+			&& timespec64_equal(&ts, &fattr->pre_mtime)) {
+		inode->i_mtime = fattr->mtime;
 		if (S_ISDIR(inode->i_mode))
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA);
 	}
@@ -1398,7 +1412,7 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat
 	struct nfs_inode *nfsi = NFS_I(inode);
 	loff_t cur_size, new_isize;
 	unsigned long invalid = 0;
-	struct timespec ts;
+	struct timespec64 ts;
 
 	if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ))
 		return 0;
@@ -1425,12 +1439,12 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat
 			invalid |= NFS_INO_INVALID_CHANGE
 				| NFS_INO_REVAL_PAGECACHE;
 
-		ts = timespec64_to_timespec(inode->i_mtime);
-		if ((fattr->valid & NFS_ATTR_FATTR_MTIME) && !timespec_equal(&ts, &fattr->mtime))
+		ts = inode->i_mtime;
+		if ((fattr->valid & NFS_ATTR_FATTR_MTIME) && !timespec64_equal(&ts, &fattr->mtime))
 			invalid |= NFS_INO_INVALID_MTIME;
 
-		ts = timespec64_to_timespec(inode->i_ctime);
-		if ((fattr->valid & NFS_ATTR_FATTR_CTIME) && !timespec_equal(&ts, &fattr->ctime))
+		ts = inode->i_ctime;
+		if ((fattr->valid & NFS_ATTR_FATTR_CTIME) && !timespec64_equal(&ts, &fattr->ctime))
 			invalid |= NFS_INO_INVALID_CTIME;
 
 		if (fattr->valid & NFS_ATTR_FATTR_SIZE) {
@@ -1460,8 +1474,8 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat
 	if ((fattr->valid & NFS_ATTR_FATTR_NLINK) && inode->i_nlink != fattr->nlink)
 		invalid |= NFS_INO_INVALID_OTHER;
 
-	ts = timespec64_to_timespec(inode->i_atime);
-	if ((fattr->valid & NFS_ATTR_FATTR_ATIME) && !timespec_equal(&ts, &fattr->atime))
+	ts = inode->i_atime;
+	if ((fattr->valid & NFS_ATTR_FATTR_ATIME) && !timespec64_equal(&ts, &fattr->atime))
 		invalid |= NFS_INO_INVALID_ATIME;
 
 	if (invalid != 0)
@@ -1733,12 +1747,12 @@ int nfs_post_op_update_inode_force_wcc_locked(struct inode *inode, struct nfs_fa
 	}
 	if ((fattr->valid & NFS_ATTR_FATTR_CTIME) != 0 &&
 			(fattr->valid & NFS_ATTR_FATTR_PRECTIME) == 0) {
-		fattr->pre_ctime = timespec64_to_timespec(inode->i_ctime);
+		fattr->pre_ctime = inode->i_ctime;
 		fattr->valid |= NFS_ATTR_FATTR_PRECTIME;
 	}
 	if ((fattr->valid & NFS_ATTR_FATTR_MTIME) != 0 &&
 			(fattr->valid & NFS_ATTR_FATTR_PREMTIME) == 0) {
-		fattr->pre_mtime = timespec64_to_timespec(inode->i_mtime);
+		fattr->pre_mtime = inode->i_mtime;
 		fattr->valid |= NFS_ATTR_FATTR_PREMTIME;
 	}
 	if ((fattr->valid & NFS_ATTR_FATTR_SIZE) != 0 &&
@@ -1899,7 +1913,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_MTIME) {
-		inode->i_mtime = timespec_to_timespec64(fattr->mtime);
+		inode->i_mtime = fattr->mtime;
 	} else if (server->caps & NFS_CAP_MTIME) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_MTIME
@@ -1908,7 +1922,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_CTIME) {
-		inode->i_ctime = timespec_to_timespec64(fattr->ctime);
+		inode->i_ctime = fattr->ctime;
 	} else if (server->caps & NFS_CAP_CTIME) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_CTIME
@@ -1946,7 +1960,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 
 
 	if (fattr->valid & NFS_ATTR_FATTR_ATIME)
-		inode->i_atime = timespec_to_timespec64(fattr->atime);
+		inode->i_atime = fattr->atime;
 	else if (server->caps & NFS_CAP_ATIME) {
 		nfsi->cache_validity |= save_cache_validity &
 				(NFS_INO_INVALID_ATIME
@@ -2058,7 +2072,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 	 * lookup validation will know that the inode is bad.
 	 * (But we fall through to invalidate the caches.)
 	 */
-	nfs_invalidate_inode(inode);
+	nfs_set_inode_stale_locked(inode);
 	return -ESTALE;
 }
 
@@ -2108,6 +2122,7 @@ static void init_once(void *foo)
 	init_rwsem(&nfsi->rmdir_sem);
 	mutex_init(&nfsi->commit_mutex);
 	nfs4_init_once(nfsi);
+	nfsi->cache_change_attribute = 0;
 }
 
 static int __init nfs_init_inodecache(void)
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 447a3c17fa8e..6673a77884d9 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -4,17 +4,19 @@
  */
 
 #include "nfs4_fs.h"
-#include <linux/mount.h>
+#include <linux/fs_context.h>
 #include <linux/security.h>
 #include <linux/crc32.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/nfs_page.h>
 #include <linux/wait_bit.h>
 
-#define NFS_MS_MASK (SB_RDONLY|SB_NOSUID|SB_NODEV|SB_NOEXEC|SB_SYNCHRONOUS)
+#define NFS_SB_MASK (SB_RDONLY|SB_NOSUID|SB_NODEV|SB_NOEXEC|SB_SYNCHRONOUS)
 
 extern const struct export_operations nfs_export_ops;
 
 struct nfs_string;
+struct nfs_pageio_descriptor;
 
 static inline void nfs_attr_check_mountpoint(struct super_block *parent, struct nfs_fattr *fattr)
 {
@@ -31,17 +33,14 @@ static inline int nfs_attr_use_mounted_on_fileid(struct nfs_fattr *fattr)
 	return 1;
 }
 
-struct nfs_clone_mount {
-	const struct super_block *sb;
-	const struct dentry *dentry;
-	struct nfs_fh *fh;
-	struct nfs_fattr *fattr;
-	char *hostname;
-	char *mnt_path;
-	struct sockaddr *addr;
-	size_t addrlen;
-	rpc_authflavor_t authflavor;
-};
+static inline bool nfs_lookup_is_soft_revalidate(const struct dentry *dentry)
+{
+	if (!(NFS_SB(dentry->d_sb)->flags & NFS_MOUNT_SOFTREVAL))
+		return false;
+	if (!d_is_positive(dentry) || !NFS_FH(d_inode(dentry))->size)
+		return false;
+	return true;
+}
 
 /*
  * Note: RFC 1813 doesn't limit the number of auth flavors that
@@ -82,12 +81,16 @@ struct nfs_client_initdata {
 /*
  * In-kernel mount arguments
  */
-struct nfs_parsed_mount_data {
-	int			flags;
+struct nfs_fs_context {
+	bool			internal;
+	bool			skip_reconfig_option_check;
+	bool			need_mount;
+	bool			sloppy;
+	unsigned int		flags;		/* NFS{,4}_MOUNT_* flags */
 	unsigned int		rsize, wsize;
 	unsigned int		timeo, retrans;
-	unsigned int		acregmin, acregmax,
-				acdirmin, acdirmax;
+	unsigned int		acregmin, acregmax;
+	unsigned int		acdirmin, acdirmax;
 	unsigned int		namlen;
 	unsigned int		options;
 	unsigned int		bsize;
@@ -97,10 +100,14 @@ struct nfs_parsed_mount_data {
 	unsigned int		version;
 	unsigned int		minorversion;
 	char			*fscache_uniq;
-	bool			need_mount;
+	unsigned short		protofamily;
+	unsigned short		mountfamily;
 
 	struct {
-		struct sockaddr_storage	address;
+		union {
+			struct sockaddr	address;
+			struct sockaddr_storage	_address;
+		};
 		size_t			addrlen;
 		char			*hostname;
 		u32			version;
@@ -109,19 +116,41 @@ struct nfs_parsed_mount_data {
 	} mount_server;
 
 	struct {
-		struct sockaddr_storage	address;
+		union {
+			struct sockaddr	address;
+			struct sockaddr_storage	_address;
+		};
 		size_t			addrlen;
 		char			*hostname;
 		char			*export_path;
 		int			port;
 		unsigned short		protocol;
 		unsigned short		nconnect;
+		unsigned short		export_path_len;
 	} nfs_server;
 
-	void			*lsm_opts;
-	struct net		*net;
+	struct nfs_fh		*mntfh;
+	struct nfs_server	*server;
+	struct nfs_subversion	*nfs_mod;
+
+	/* Information for a cloned mount. */
+	struct nfs_clone_mount {
+		struct super_block	*sb;
+		struct dentry		*dentry;
+		struct nfs_fattr	*fattr;
+		unsigned int		inherited_bsize;
+	} clone_data;
 };
 
+#define nfs_errorf(fc, fmt, ...) errorf(fc, fmt, ## __VA_ARGS__)
+#define nfs_invalf(fc, fmt, ...) invalf(fc, fmt, ## __VA_ARGS__)
+#define nfs_warnf(fc, fmt, ...) warnf(fc, fmt, ## __VA_ARGS__)
+
+static inline struct nfs_fs_context *nfs_fc2context(const struct fs_context *fc)
+{
+	return fc->fs_private;
+}
+
 /* mount_clnt.c */
 struct nfs_mount_request {
 	struct sockaddr		*sap;
@@ -137,14 +166,6 @@ struct nfs_mount_request {
 	struct net		*net;
 };
 
-struct nfs_mount_info {
-	void (*fill_super)(struct super_block *, struct nfs_mount_info *);
-	int (*set_security)(struct super_block *, struct dentry *, struct nfs_mount_info *);
-	struct nfs_parsed_mount_data *parsed;
-	struct nfs_clone_mount *cloned;
-	struct nfs_fh *mntfh;
-};
-
 extern int nfs_mount(struct nfs_mount_request *info);
 extern void nfs_umount(const struct nfs_mount_request *info);
 
@@ -170,13 +191,9 @@ extern struct nfs_client *nfs4_find_client_ident(struct net *, int);
 extern struct nfs_client *
 nfs4_find_client_sessionid(struct net *, const struct sockaddr *,
 				struct nfs4_sessionid *, u32);
-extern struct nfs_server *nfs_create_server(struct nfs_mount_info *,
-					struct nfs_subversion *);
-extern struct nfs_server *nfs4_create_server(
-					struct nfs_mount_info *,
-					struct nfs_subversion *);
-extern struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *,
-						      struct nfs_fh *);
+extern struct nfs_server *nfs_create_server(struct fs_context *);
+extern struct nfs_server *nfs4_create_server(struct fs_context *);
+extern struct nfs_server *nfs4_create_referral_server(struct fs_context *);
 extern int nfs4_update_server(struct nfs_server *server, const char *hostname,
 					struct sockaddr *sap, size_t salen,
 					struct net *net);
@@ -227,7 +244,9 @@ static inline void nfs_fs_proc_exit(void)
 extern const struct svc_version nfs4_callback_version1;
 extern const struct svc_version nfs4_callback_version4;
 
-struct nfs_pageio_descriptor;
+/* fs_context.c */
+extern struct file_system_type nfs_fs_type;
+
 /* pagelist.c */
 extern int __init nfs_init_nfspagecache(void);
 extern void nfs_destroy_nfspagecache(void);
@@ -255,12 +274,6 @@ void nfs_free_request(struct nfs_page *req);
 struct nfs_pgio_mirror *
 nfs_pgio_current_mirror(struct nfs_pageio_descriptor *desc);
 
-static inline bool nfs_pgio_has_mirroring(struct nfs_pageio_descriptor *desc)
-{
-	WARN_ON_ONCE(desc->pg_mirror_count < 1);
-	return desc->pg_mirror_count > 1;
-}
-
 static inline bool nfs_match_open_context(const struct nfs_open_context *ctx1,
 		const struct nfs_open_context *ctx2)
 {
@@ -387,23 +400,10 @@ extern int nfs_wait_atomic_killable(atomic_t *p, unsigned int mode);
 
 /* super.c */
 extern const struct super_operations nfs_sops;
-extern struct file_system_type nfs_fs_type;
-extern struct file_system_type nfs_xdev_fs_type;
-#if IS_ENABLED(CONFIG_NFS_V4)
-extern struct file_system_type nfs4_referral_fs_type;
-#endif
 bool nfs_auth_info_match(const struct nfs_auth_info *, rpc_authflavor_t);
-struct dentry *nfs_try_mount(int, const char *, struct nfs_mount_info *,
-			struct nfs_subversion *);
-int nfs_set_sb_security(struct super_block *, struct dentry *, struct nfs_mount_info *);
-int nfs_clone_sb_security(struct super_block *, struct dentry *, struct nfs_mount_info *);
-struct dentry *nfs_fs_mount_common(struct nfs_server *, int, const char *,
-				   struct nfs_mount_info *, struct nfs_subversion *);
-struct dentry *nfs_fs_mount(struct file_system_type *, int, const char *, void *);
-struct dentry * nfs_xdev_mount_common(struct file_system_type *, int,
-		const char *, struct nfs_mount_info *);
+int nfs_try_get_tree(struct fs_context *);
+int nfs_get_tree_common(struct fs_context *);
 void nfs_kill_super(struct super_block *);
-void nfs_fill_super(struct super_block *, struct nfs_mount_info *);
 
 extern struct rpc_stat nfs_rpcstat;
 
@@ -411,7 +411,9 @@ extern int __init register_nfs_fs(void);
 extern void __exit unregister_nfs_fs(void);
 extern bool nfs_sb_active(struct super_block *sb);
 extern void nfs_sb_deactive(struct super_block *sb);
-
+extern int nfs_client_for_each_server(struct nfs_client *clp,
+				      int (*fn)(struct nfs_server *, void *),
+				      void *data);
 /* io.c */
 extern void nfs_start_io_read(struct inode *inode);
 extern void nfs_end_io_read(struct inode *inode);
@@ -430,18 +432,12 @@ static inline bool nfs_file_io_is_buffered(struct nfs_inode *nfsi)
 extern char *nfs_path(char **p, struct dentry *dentry,
 		      char *buffer, ssize_t buflen, unsigned flags);
 extern struct vfsmount *nfs_d_automount(struct path *path);
-struct vfsmount *nfs_submount(struct nfs_server *, struct dentry *,
-			      struct nfs_fh *, struct nfs_fattr *);
-struct vfsmount *nfs_do_submount(struct dentry *, struct nfs_fh *,
-				 struct nfs_fattr *, rpc_authflavor_t);
+int nfs_submount(struct fs_context *, struct nfs_server *);
+int nfs_do_submount(struct fs_context *);
 
 /* getroot.c */
-extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *,
-				   const char *);
+extern int nfs_get_root(struct super_block *s, struct fs_context *fc);
 #if IS_ENABLED(CONFIG_NFS_V4)
-extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *,
-				    const char *);
-
 extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool);
 #endif
 
@@ -460,7 +456,7 @@ int  nfs_show_options(struct seq_file *, struct dentry *);
 int  nfs_show_devname(struct seq_file *, struct dentry *);
 int  nfs_show_path(struct seq_file *, struct dentry *);
 int  nfs_show_stats(struct seq_file *, struct dentry *);
-int nfs_remount(struct super_block *sb, int *flags, char *raw_data);
+int  nfs_reconfigure(struct fs_context *);
 
 /* write.c */
 extern void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio,
@@ -515,13 +511,25 @@ int nfs_filemap_write_and_wait_range(struct address_space *mapping,
 		loff_t lstart, loff_t lend);
 
 #ifdef CONFIG_NFS_V4_1
+static inline void
+pnfs_bucket_clear_pnfs_ds_commit_verifiers(struct pnfs_commit_bucket *buckets,
+		unsigned int nbuckets)
+{
+	unsigned int i;
+
+	for (i = 0; i < nbuckets; i++)
+		buckets[i].direct_verf.committed = NFS_INVALID_STABLE_HOW;
+}
 static inline
 void nfs_clear_pnfs_ds_commit_verifiers(struct pnfs_ds_commit_info *cinfo)
 {
-	int i;
+	struct pnfs_commit_array *array;
 
-	for (i = 0; i < cinfo->nbuckets; i++)
-		cinfo->buckets[i].direct_verf.committed = NFS_INVALID_STABLE_HOW;
+	rcu_read_lock();
+	list_for_each_entry_rcu(array, &cinfo->commits, cinfo_list)
+		pnfs_bucket_clear_pnfs_ds_commit_verifiers(array->buckets,
+				array->nbuckets);
+	rcu_read_unlock();
 }
 #else
 static inline
@@ -542,6 +550,14 @@ nfs_write_verifier_cmp(const struct nfs_write_verifier *v1,
 	return memcmp(v1->data, v2->data, sizeof(v1->data));
 }
 
+static inline bool
+nfs_write_match_verf(const struct nfs_writeverf *verf,
+		struct nfs_page *req)
+{
+	return verf->committed > NFS_UNSTABLE &&
+		!nfs_write_verifier_cmp(&req->wb_verf, &verf->verifier);
+}
+
 /* unlink.c */
 extern struct rpc_task *
 nfs_async_rename(struct inode *old_dir, struct inode *new_dir,
@@ -652,7 +668,8 @@ void nfs_super_set_maxbytes(struct super_block *sb, __u64 maxfilesize)
 }
 
 /*
- * Record the page as unstable and mark its inode as dirty.
+ * Record the page as unstable (an extra writeback period) and mark its
+ * inode as dirty.
  */
 static inline
 void nfs_mark_page_unstable(struct page *page, struct nfs_commit_info *cinfo)
@@ -660,8 +677,11 @@ void nfs_mark_page_unstable(struct page *page, struct nfs_commit_info *cinfo)
 	if (!cinfo->dreq) {
 		struct inode *inode = page_file_mapping(page)->host;
 
-		inc_node_page_state(page, NR_UNSTABLE_NFS);
-		inc_wb_stat(&inode_to_bdi(inode)->wb, WB_RECLAIMABLE);
+		/* This page is really still in write-back - just that the
+		 * writeback is happening on the server now.
+		 */
+		inc_node_page_state(page, NR_WRITEBACK);
+		inc_wb_stat(&inode_to_bdi(inode)->wb, WB_WRITEBACK);
 		__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
 	}
 }
@@ -706,14 +726,14 @@ unsigned int nfs_page_array_len(unsigned int base, size_t len)
 }
 
 /*
- * Convert a struct timespec into a 64-bit change attribute
+ * Convert a struct timespec64 into a 64-bit change attribute
  *
- * This does approximately the same thing as timespec_to_ns(),
+ * This does approximately the same thing as timespec64_to_ns(),
  * but for calculation efficiency, we multiply the seconds by
  * 1024*1024*1024.
  */
 static inline
-u64 nfs_timespec_to_change_attr(const struct timespec *ts)
+u64 nfs_timespec_to_change_attr(const struct timespec64 *ts)
 {
 	return ((u64)ts->tv_sec << 30) + ts->tv_nsec;
 }
@@ -777,3 +797,16 @@ static inline bool nfs_error_is_fatal_on_server(int err)
 	}
 	return nfs_error_is_fatal(err);
 }
+
+/*
+ * Select between a default port value and a user-specified port value.
+ * If a zero value is set, then autobind will be used.
+ */
+static inline void nfs_set_port(struct sockaddr *sap, int *port,
+				const unsigned short default_port)
+{
+	if (*port == NFS_UNSPEC_PORT)
+		*port = default_port;
+
+	rpc_set_port(sap, *port);
+}
diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index cb7c10e9721e..dda5c3e65d8d 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -29,9 +29,8 @@
  */
 #define encode_dirpath_sz	(1 + XDR_QUADLEN(MNTPATHLEN))
 #define MNT_status_sz		(1)
-#define MNT_fhs_status_sz	(1)
 #define MNT_fhandle_sz		XDR_QUADLEN(NFS2_FHSIZE)
-#define MNT_fhandle3_sz		(1 + XDR_QUADLEN(NFS3_FHSIZE))
+#define MNT_fhandlev3_sz	XDR_QUADLEN(NFS3_FHSIZE)
 #define MNT_authflav3_sz	(1 + NFS_MAX_SECFLAVORS)
 
 /*
@@ -39,7 +38,7 @@
  */
 #define MNT_enc_dirpath_sz	encode_dirpath_sz
 #define MNT_dec_mountres_sz	(MNT_status_sz + MNT_fhandle_sz)
-#define MNT_dec_mountres3_sz	(MNT_status_sz + MNT_fhandle_sz + \
+#define MNT_dec_mountres3_sz	(MNT_status_sz + MNT_fhandlev3_sz + \
 				 MNT_authflav3_sz)
 
 /*
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index 9287eb666322..6b063227e34e 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -19,6 +19,7 @@
 #include <linux/vfs.h>
 #include <linux/sunrpc/gss_api.h>
 #include "internal.h"
+#include "nfs.h"
 
 #define NFSDBG_FACILITY		NFSDBG_VFS
 
@@ -139,31 +140,66 @@ EXPORT_SYMBOL_GPL(nfs_path);
  */
 struct vfsmount *nfs_d_automount(struct path *path)
 {
-	struct vfsmount *mnt;
+	struct nfs_fs_context *ctx;
+	struct fs_context *fc;
+	struct vfsmount *mnt = ERR_PTR(-ENOMEM);
 	struct nfs_server *server = NFS_SERVER(d_inode(path->dentry));
-	struct nfs_fh *fh = NULL;
-	struct nfs_fattr *fattr = NULL;
+	struct nfs_client *client = server->nfs_client;
+	int timeout = READ_ONCE(nfs_mountpoint_expiry_timeout);
+	int ret;
 
 	if (IS_ROOT(path->dentry))
 		return ERR_PTR(-ESTALE);
 
-	mnt = ERR_PTR(-ENOMEM);
-	fh = nfs_alloc_fhandle();
-	fattr = nfs_alloc_fattr();
-	if (fh == NULL || fattr == NULL)
-		goto out;
+	/* Open a new filesystem context, transferring parameters from the
+	 * parent superblock, including the network namespace.
+	 */
+	fc = fs_context_for_submount(path->mnt->mnt_sb->s_type, path->dentry);
+	if (IS_ERR(fc))
+		return ERR_CAST(fc);
 
-	mnt = server->nfs_client->rpc_ops->submount(server, path->dentry, fh, fattr);
+	ctx = nfs_fc2context(fc);
+	ctx->clone_data.dentry	= path->dentry;
+	ctx->clone_data.sb	= path->dentry->d_sb;
+	ctx->clone_data.fattr	= nfs_alloc_fattr();
+	if (!ctx->clone_data.fattr)
+		goto out_fc;
+
+	if (fc->net_ns != client->cl_net) {
+		put_net(fc->net_ns);
+		fc->net_ns = get_net(client->cl_net);
+	}
+
+	/* for submounts we want the same server; referrals will reassign */
+	memcpy(&ctx->nfs_server.address, &client->cl_addr, client->cl_addrlen);
+	ctx->nfs_server.addrlen	= client->cl_addrlen;
+	ctx->nfs_server.port	= server->port;
+
+	ctx->version		= client->rpc_ops->version;
+	ctx->minorversion	= client->cl_minorversion;
+	ctx->nfs_mod		= client->cl_nfs_mod;
+	__module_get(ctx->nfs_mod->owner);
+
+	ret = client->rpc_ops->submount(fc, server);
+	if (ret < 0) {
+		mnt = ERR_PTR(ret);
+		goto out_fc;
+	}
+
+	up_write(&fc->root->d_sb->s_umount);
+	mnt = vfs_create_mount(fc);
 	if (IS_ERR(mnt))
-		goto out;
+		goto out_fc;
 
 	mntget(mnt); /* prevent immediate expiration */
+	if (timeout <= 0)
+		goto out_fc;
+
 	mnt_set_expiry(mnt, &nfs_automount_list);
-	schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
+	schedule_delayed_work(&nfs_automount_task, timeout);
 
-out:
-	nfs_free_fattr(fattr);
-	nfs_free_fhandle(fh);
+out_fc:
+	put_fs_context(fc);
 	return mnt;
 }
 
@@ -198,10 +234,11 @@ const struct inode_operations nfs_referral_inode_operations = {
 static void nfs_expire_automounts(struct work_struct *work)
 {
 	struct list_head *list = &nfs_automount_list;
+	int timeout = READ_ONCE(nfs_mountpoint_expiry_timeout);
 
 	mark_mounts_for_expiry(list);
-	if (!list_empty(list))
-		schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
+	if (!list_empty(list) && timeout > 0)
+		schedule_delayed_work(&nfs_automount_task, timeout);
 }
 
 void nfs_release_automount_timer(void)
@@ -210,64 +247,117 @@ void nfs_release_automount_timer(void)
 		cancel_delayed_work(&nfs_automount_task);
 }
 
-/*
- * Clone a mountpoint of the appropriate type
- */
-static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
-					   const char *devname,
-					   struct nfs_clone_mount *mountdata)
-{
-	return vfs_submount(mountdata->dentry, &nfs_xdev_fs_type, devname, mountdata);
-}
-
 /**
  * nfs_do_submount - set up mountpoint when crossing a filesystem boundary
- * @dentry: parent directory
- * @fh: filehandle for new root dentry
- * @fattr: attributes for new root inode
- * @authflavor: security flavor to use when performing the mount
+ * @fc: pointer to struct nfs_fs_context
  *
  */
-struct vfsmount *nfs_do_submount(struct dentry *dentry, struct nfs_fh *fh,
-				 struct nfs_fattr *fattr, rpc_authflavor_t authflavor)
+int nfs_do_submount(struct fs_context *fc)
 {
-	struct nfs_clone_mount mountdata = {
-		.sb = dentry->d_sb,
-		.dentry = dentry,
-		.fh = fh,
-		.fattr = fattr,
-		.authflavor = authflavor,
-	};
-	struct vfsmount *mnt;
-	char *page = (char *) __get_free_page(GFP_USER);
-	char *devname;
-
-	if (page == NULL)
-		return ERR_PTR(-ENOMEM);
-
-	devname = nfs_devname(dentry, page, PAGE_SIZE);
-	if (IS_ERR(devname))
-		mnt = ERR_CAST(devname);
-	else
-		mnt = nfs_do_clone_mount(NFS_SB(dentry->d_sb), devname, &mountdata);
-
-	free_page((unsigned long)page);
-	return mnt;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct dentry *dentry = ctx->clone_data.dentry;
+	struct nfs_server *server;
+	char *buffer, *p;
+	int ret;
+
+	/* create a new volume representation */
+	server = ctx->nfs_mod->rpc_ops->clone_server(NFS_SB(ctx->clone_data.sb),
+						     ctx->mntfh,
+						     ctx->clone_data.fattr,
+						     ctx->selected_flavor);
+
+	if (IS_ERR(server))
+		return PTR_ERR(server);
+
+	ctx->server = server;
+
+	buffer = kmalloc(4096, GFP_USER);
+	if (!buffer)
+		return -ENOMEM;
+
+	ctx->internal		= true;
+	ctx->clone_data.inherited_bsize = ctx->clone_data.sb->s_blocksize_bits;
+
+	p = nfs_devname(dentry, buffer, 4096);
+	if (IS_ERR(p)) {
+		nfs_errorf(fc, "NFS: Couldn't determine submount pathname");
+		ret = PTR_ERR(p);
+	} else {
+		ret = vfs_parse_fs_string(fc, "source", p, buffer + 4096 - p);
+		if (!ret)
+			ret = vfs_get_tree(fc);
+	}
+	kfree(buffer);
+	return ret;
 }
 EXPORT_SYMBOL_GPL(nfs_do_submount);
 
-struct vfsmount *nfs_submount(struct nfs_server *server, struct dentry *dentry,
-			      struct nfs_fh *fh, struct nfs_fattr *fattr)
+int nfs_submount(struct fs_context *fc, struct nfs_server *server)
 {
-	int err;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct dentry *dentry = ctx->clone_data.dentry;
 	struct dentry *parent = dget_parent(dentry);
+	int err;
 
 	/* Look it up again to get its attributes */
-	err = server->nfs_client->rpc_ops->lookup(d_inode(parent), &dentry->d_name, fh, fattr, NULL);
+	err = server->nfs_client->rpc_ops->lookup(d_inode(parent), dentry,
+						  ctx->mntfh, ctx->clone_data.fattr,
+						  NULL);
 	dput(parent);
 	if (err != 0)
-		return ERR_PTR(err);
+		return err;
 
-	return nfs_do_submount(dentry, fh, fattr, server->client->cl_auth->au_flavor);
+	ctx->selected_flavor = server->client->cl_auth->au_flavor;
+	return nfs_do_submount(fc);
 }
 EXPORT_SYMBOL_GPL(nfs_submount);
+
+static int param_set_nfs_timeout(const char *val, const struct kernel_param *kp)
+{
+	long num;
+	int ret;
+
+	if (!val)
+		return -EINVAL;
+	ret = kstrtol(val, 0, &num);
+	if (ret)
+		return -EINVAL;
+	if (num > 0) {
+		if (num >= INT_MAX / HZ)
+			num = INT_MAX;
+		else
+			num *= HZ;
+		*((int *)kp->arg) = num;
+		if (!list_empty(&nfs_automount_list))
+			mod_delayed_work(system_wq, &nfs_automount_task, num);
+	} else {
+		*((int *)kp->arg) = -1*HZ;
+		cancel_delayed_work(&nfs_automount_task);
+	}
+	return 0;
+}
+
+static int param_get_nfs_timeout(char *buffer, const struct kernel_param *kp)
+{
+	long num = *((int *)kp->arg);
+
+	if (num > 0) {
+		if (num >= INT_MAX - (HZ - 1))
+			num = INT_MAX / HZ;
+		else
+			num = (num + (HZ - 1)) / HZ;
+	} else
+		num = -1;
+	return scnprintf(buffer, PAGE_SIZE, "%li\n", num);
+}
+
+static const struct kernel_param_ops param_ops_nfs_timeout = {
+	.set = param_set_nfs_timeout,
+	.get = param_get_nfs_timeout,
+};
+#define param_check_nfs_timeout(name, p) __param_check(name, p, int);
+
+module_param(nfs_mountpoint_expiry_timeout, nfs_timeout, 0644);
+MODULE_PARM_DESC(nfs_mountpoint_expiry_timeout,
+		"Set the NFS automounted mountpoint timeout value (seconds)."
+		"Values <= 0 turn expiration off.");
diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c
index cbc17a203248..f6676af37d5d 100644
--- a/fs/nfs/nfs2xdr.c
+++ b/fs/nfs/nfs2xdr.c
@@ -209,9 +209,9 @@ static int decode_fhandle(struct xdr_stream *xdr, struct nfs_fh *fh)
  *		unsigned int useconds;
  *	};
  */
-static __be32 *xdr_encode_time(__be32 *p, const struct timespec *timep)
+static __be32 *xdr_encode_time(__be32 *p, const struct timespec64 *timep)
 {
-	*p++ = cpu_to_be32(timep->tv_sec);
+	*p++ = cpu_to_be32((u32)timep->tv_sec);
 	if (timep->tv_nsec != 0)
 		*p++ = cpu_to_be32(timep->tv_nsec / NSEC_PER_USEC);
 	else
@@ -227,14 +227,14 @@ static __be32 *xdr_encode_time(__be32 *p, const struct timespec *timep)
  * Illustrated" by Brent Callaghan, Addison-Wesley, ISBN 0-201-32750-5.
  */
 static __be32 *xdr_encode_current_server_time(__be32 *p,
-					      const struct timespec *timep)
+					      const struct timespec64 *timep)
 {
 	*p++ = cpu_to_be32(timep->tv_sec);
 	*p++ = cpu_to_be32(1000000);
 	return p;
 }
 
-static __be32 *xdr_decode_time(__be32 *p, struct timespec *timep)
+static __be32 *xdr_decode_time(__be32 *p, struct timespec64 *timep)
 {
 	timep->tv_sec = be32_to_cpup(p++);
 	timep->tv_nsec = be32_to_cpup(p++) * NSEC_PER_USEC;
@@ -339,7 +339,6 @@ static __be32 *xdr_time_not_set(__be32 *p)
 static void encode_sattr(struct xdr_stream *xdr, const struct iattr *attr,
 		struct user_namespace *userns)
 {
-	struct timespec ts;
 	__be32 *p;
 
 	p = xdr_reserve_space(xdr, NFS_sattr_sz << 2);
@@ -361,21 +360,17 @@ static void encode_sattr(struct xdr_stream *xdr, const struct iattr *attr,
 	else
 		*p++ = cpu_to_be32(NFS2_SATTR_NOT_SET);
 
-	if (attr->ia_valid & ATTR_ATIME_SET) {
-		ts = timespec64_to_timespec(attr->ia_atime);
-		p = xdr_encode_time(p, &ts);
-	} else if (attr->ia_valid & ATTR_ATIME) {
-		ts = timespec64_to_timespec(attr->ia_atime);
-		p = xdr_encode_current_server_time(p, &ts);
-	} else
+	if (attr->ia_valid & ATTR_ATIME_SET)
+		p = xdr_encode_time(p, &attr->ia_atime);
+	else if (attr->ia_valid & ATTR_ATIME)
+		p = xdr_encode_current_server_time(p, &attr->ia_atime);
+	else
 		p = xdr_time_not_set(p);
-	if (attr->ia_valid & ATTR_MTIME_SET) {
-		ts = timespec64_to_timespec(attr->ia_atime);
-		xdr_encode_time(p, &ts);
-	} else if (attr->ia_valid & ATTR_MTIME) {
-		ts = timespec64_to_timespec(attr->ia_mtime);
-		xdr_encode_current_server_time(p, &ts);
-	} else
+	if (attr->ia_valid & ATTR_MTIME_SET)
+		xdr_encode_time(p, &attr->ia_mtime);
+	else if (attr->ia_valid & ATTR_MTIME)
+		xdr_encode_current_server_time(p, &attr->ia_mtime);
+	else
 		xdr_time_not_set(p);
 }
 
diff --git a/fs/nfs/nfs3_fs.h b/fs/nfs/nfs3_fs.h
index f82e11c4cb56..1b950b66b3bb 100644
--- a/fs/nfs/nfs3_fs.h
+++ b/fs/nfs/nfs3_fs.h
@@ -27,7 +27,7 @@ static inline int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
 #endif /* CONFIG_NFS_V3_ACL */
 
 /* nfs3client.c */
-struct nfs_server *nfs3_create_server(struct nfs_mount_info *, struct nfs_subversion *);
+struct nfs_server *nfs3_create_server(struct fs_context *);
 struct nfs_server *nfs3_clone_server(struct nfs_server *, struct nfs_fh *,
 				     struct nfs_fattr *, rpc_authflavor_t);
 
diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c
index c5c3fc6e6c60..26c94b32d6f4 100644
--- a/fs/nfs/nfs3acl.c
+++ b/fs/nfs/nfs3acl.c
@@ -253,37 +253,45 @@ int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,
 
 int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 {
-	struct posix_acl *alloc = NULL, *dfacl = NULL;
+	struct posix_acl *orig = acl, *dfacl = NULL, *alloc;
 	int status;
 
 	if (S_ISDIR(inode->i_mode)) {
 		switch(type) {
 		case ACL_TYPE_ACCESS:
-			alloc = dfacl = get_acl(inode, ACL_TYPE_DEFAULT);
+			alloc = get_acl(inode, ACL_TYPE_DEFAULT);
 			if (IS_ERR(alloc))
 				goto fail;
+			dfacl = alloc;
 			break;
 
 		case ACL_TYPE_DEFAULT:
-			dfacl = acl;
-			alloc = acl = get_acl(inode, ACL_TYPE_ACCESS);
+			alloc = get_acl(inode, ACL_TYPE_ACCESS);
 			if (IS_ERR(alloc))
 				goto fail;
+			dfacl = acl;
+			acl = alloc;
 			break;
 		}
 	}
 
 	if (acl == NULL) {
-		alloc = acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
+		alloc = posix_acl_from_mode(inode->i_mode, GFP_KERNEL);
 		if (IS_ERR(alloc))
 			goto fail;
+		acl = alloc;
 	}
 	status = __nfs3_proc_setacls(inode, acl, dfacl);
-	posix_acl_release(alloc);
+out:
+	if (acl != orig)
+		posix_acl_release(acl);
+	if (dfacl != orig)
+		posix_acl_release(dfacl);
 	return status;
 
 fail:
-	return PTR_ERR(alloc);
+	status = PTR_ERR(alloc);
+	goto out;
 }
 
 const struct xattr_handler *nfs3_xattr_handlers[] = {
diff --git a/fs/nfs/nfs3client.c b/fs/nfs/nfs3client.c
index 148ceb74d27c..5601e47360c2 100644
--- a/fs/nfs/nfs3client.c
+++ b/fs/nfs/nfs3client.c
@@ -46,10 +46,10 @@ static inline void nfs_init_server_aclclient(struct nfs_server *server)
 }
 #endif
 
-struct nfs_server *nfs3_create_server(struct nfs_mount_info *mount_info,
-				      struct nfs_subversion *nfs_mod)
+struct nfs_server *nfs3_create_server(struct fs_context *fc)
 {
-	struct nfs_server *server = nfs_create_server(mount_info, nfs_mod);
+	struct nfs_server *server = nfs_create_server(fc);
+
 	/* Create a client RPC handle for the NFS v3 ACL management interface */
 	if (!IS_ERR(server))
 		nfs_init_server_aclclient(server);
@@ -106,7 +106,10 @@ struct nfs_client *nfs3_set_ds_client(struct nfs_server *mds_srv,
 		cl_init.nconnect = mds_clp->cl_nconnect;
 
 	if (mds_srv->flags & NFS_MOUNT_NORESVPORT)
-		set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
+		__set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
+
+	__set_bit(NFS_CS_NOPING, &cl_init.init_flags);
+	__set_bit(NFS_CS_DS, &cl_init.init_flags);
 
 	/* Use the MDS nfs_client cl_ipaddr. */
 	nfs_init_timeout_values(&ds_timeout, ds_proto, ds_timeo, ds_retrans);
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 9eb2f1a503ab..a46d1d5d16d8 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -110,10 +110,15 @@ nfs3_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_resp	= fattr,
 	};
 	int	status;
+	unsigned short task_flags = 0;
+
+	/* Is this is an attribute revalidation, subject to softreval? */
+	if (inode && (server->flags & NFS_MOUNT_SOFTREVAL))
+		task_flags |= RPC_TASK_TIMEOUT;
 
 	dprintk("NFS call  getattr\n");
 	nfs_fattr_init(fattr);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = rpc_call_sync(server->client, &msg, task_flags);
 	dprintk("NFS reply getattr: %d\n", status);
 	return status;
 }
@@ -140,23 +145,23 @@ nfs3_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
 	nfs_fattr_init(fattr);
 	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
 	if (status == 0) {
+		nfs_setattr_update_inode(inode, sattr, fattr);
 		if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_ACL)
 			nfs_zap_acl_cache(inode);
-		nfs_setattr_update_inode(inode, sattr, fattr);
 	}
 	dprintk("NFS reply setattr: %d\n", status);
 	return status;
 }
 
 static int
-nfs3_proc_lookup(struct inode *dir, const struct qstr *name,
+nfs3_proc_lookup(struct inode *dir, struct dentry *dentry,
 		 struct nfs_fh *fhandle, struct nfs_fattr *fattr,
 		 struct nfs4_label *label)
 {
 	struct nfs3_diropargs	arg = {
 		.fh		= NFS_FH(dir),
-		.name		= name->name,
-		.len		= name->len
+		.name		= dentry->d_name.name,
+		.len		= dentry->d_name.len
 	};
 	struct nfs3_diropres	res = {
 		.fh		= fhandle,
@@ -168,20 +173,25 @@ nfs3_proc_lookup(struct inode *dir, const struct qstr *name,
 		.rpc_resp	= &res,
 	};
 	int			status;
+	unsigned short task_flags = 0;
 
-	dprintk("NFS call  lookup %s\n", name->name);
+	/* Is this is an attribute revalidation, subject to softreval? */
+	if (nfs_lookup_is_soft_revalidate(dentry))
+		task_flags |= RPC_TASK_TIMEOUT;
+
+	dprintk("NFS call  lookup %pd2\n", dentry);
 	res.dir_attr = nfs_alloc_fattr();
 	if (res.dir_attr == NULL)
 		return -ENOMEM;
 
 	nfs_fattr_init(fattr);
-	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, task_flags);
 	nfs_refresh_inode(dir, res.dir_attr);
 	if (status >= 0 && !(fattr->valid & NFS_ATTR_FATTR)) {
 		msg.rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR];
 		msg.rpc_argp = fhandle;
 		msg.rpc_resp = fattr;
-		status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+		status = rpc_call_sync(NFS_CLIENT(dir), &msg, task_flags);
 	}
 	nfs_free_fattr(res.dir_attr);
 	dprintk("NFS reply lookup: %d\n", status);
@@ -990,7 +1000,7 @@ const struct nfs_rpc_ops nfs_v3_clientops = {
 	.nlmclnt_ops	= &nlmclnt_fl_close_lock_ops,
 	.getroot	= nfs3_proc_get_root,
 	.submount	= nfs_submount,
-	.try_mount	= nfs_try_mount,
+	.try_get_tree	= nfs_try_get_tree,
 	.getattr	= nfs3_proc_getattr,
 	.setattr	= nfs3_proc_setattr,
 	.lookup		= nfs3_proc_lookup,
diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c
index 602767850b36..69971f6c840d 100644
--- a/fs/nfs/nfs3xdr.c
+++ b/fs/nfs/nfs3xdr.c
@@ -456,14 +456,14 @@ static void zero_nfs_fh3(struct nfs_fh *fh)
  *		uint32	nseconds;
  *	};
  */
-static __be32 *xdr_encode_nfstime3(__be32 *p, const struct timespec *timep)
+static __be32 *xdr_encode_nfstime3(__be32 *p, const struct timespec64 *timep)
 {
-	*p++ = cpu_to_be32(timep->tv_sec);
+	*p++ = cpu_to_be32((u32)timep->tv_sec);
 	*p++ = cpu_to_be32(timep->tv_nsec);
 	return p;
 }
 
-static __be32 *xdr_decode_nfstime3(__be32 *p, struct timespec *timep)
+static __be32 *xdr_decode_nfstime3(__be32 *p, struct timespec64 *timep)
 {
 	timep->tv_sec = be32_to_cpup(p++);
 	timep->tv_nsec = be32_to_cpup(p++);
@@ -533,7 +533,6 @@ static __be32 *xdr_decode_nfstime3(__be32 *p, struct timespec *timep)
 static void encode_sattr3(struct xdr_stream *xdr, const struct iattr *attr,
 		struct user_namespace *userns)
 {
-	struct timespec ts;
 	u32 nbytes;
 	__be32 *p;
 
@@ -583,10 +582,8 @@ static void encode_sattr3(struct xdr_stream *xdr, const struct iattr *attr,
 		*p++ = xdr_zero;
 
 	if (attr->ia_valid & ATTR_ATIME_SET) {
-		struct timespec ts;
 		*p++ = xdr_two;
-		ts = timespec64_to_timespec(attr->ia_atime);
-		p = xdr_encode_nfstime3(p, &ts);
+		p = xdr_encode_nfstime3(p, &attr->ia_atime);
 	} else if (attr->ia_valid & ATTR_ATIME) {
 		*p++ = xdr_one;
 	} else
@@ -594,8 +591,7 @@ static void encode_sattr3(struct xdr_stream *xdr, const struct iattr *attr,
 
 	if (attr->ia_valid & ATTR_MTIME_SET) {
 		*p++ = xdr_two;
-		ts = timespec64_to_timespec(attr->ia_mtime);
-		xdr_encode_nfstime3(p, &ts);
+		xdr_encode_nfstime3(p, &attr->ia_mtime);
 	} else if (attr->ia_valid & ATTR_MTIME) {
 		*p = xdr_one;
 	} else
@@ -2338,6 +2334,7 @@ static int nfs3_xdr_dec_commit3res(struct rpc_rqst *req,
 				   void *data)
 {
 	struct nfs_commitres *result = data;
+	struct nfs_writeverf *verf = result->verf;
 	enum nfs_stat status;
 	int error;
 
@@ -2350,7 +2347,9 @@ static int nfs3_xdr_dec_commit3res(struct rpc_rqst *req,
 	result->op_status = status;
 	if (status != NFS3_OK)
 		goto out_status;
-	error = decode_writeverf3(xdr, &result->verf->verifier);
+	error = decode_writeverf3(xdr, &verf->verifier);
+	if (!error)
+		verf->committed = NFS_FILE_SYNC;
 out:
 	return error;
 out_status:
diff --git a/fs/nfs/nfs42.h b/fs/nfs/nfs42.h
index 901cca7542f9..c891af949886 100644
--- a/fs/nfs/nfs42.h
+++ b/fs/nfs/nfs42.h
@@ -13,8 +13,10 @@
 #define PNFS_LAYOUTSTATS_MAXDEV (4)
 
 /* nfs4.2proc.c */
+#ifdef CONFIG_NFS_V4_2
 int nfs42_proc_allocate(struct file *, loff_t, loff_t);
-ssize_t nfs42_proc_copy(struct file *, loff_t, struct file *, loff_t, size_t);
+ssize_t nfs42_proc_copy(struct file *, loff_t, struct file *, loff_t, size_t,
+			struct nl4_server *, nfs4_stateid *, bool);
 int nfs42_proc_deallocate(struct file *, loff_t, loff_t);
 loff_t nfs42_proc_llseek(struct file *, loff_t, int);
 int nfs42_proc_layoutstats_generic(struct nfs_server *,
@@ -23,5 +25,16 @@ int nfs42_proc_clone(struct file *, struct file *, loff_t, loff_t, loff_t);
 int nfs42_proc_layouterror(struct pnfs_layout_segment *lseg,
 			   const struct nfs42_layout_error *errors,
 			   size_t n);
+int nfs42_proc_copy_notify(struct file *, struct file *,
+			   struct nfs42_copy_notify_res *);
+static inline bool nfs42_files_from_same_server(struct file *in,
+						struct file *out)
+{
+	struct nfs_client *c_in = (NFS_SERVER(file_inode(in)))->nfs_client;
+	struct nfs_client *c_out = (NFS_SERVER(file_inode(out)))->nfs_client;
 
+	return nfs4_check_serverowner_major_id(c_in->cl_serverowner,
+					       c_out->cl_serverowner);
+}
+#endif /* CONFIG_NFS_V4_2 */
 #endif /* __LINUX_FS_NFS_NFS4_2_H */
diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
index 5196bfa7894d..e2ae54b35dfe 100644
--- a/fs/nfs/nfs42proc.c
+++ b/fs/nfs/nfs42proc.c
@@ -3,6 +3,7 @@
  * Copyright (c) 2014 Anna Schumaker <Anna.Schumaker@Netapp.com>
  */
 #include <linux/fs.h>
+#include <linux/sunrpc/addr.h>
 #include <linux/sunrpc/sched.h>
 #include <linux/nfs.h>
 #include <linux/nfs3.h>
@@ -15,10 +16,30 @@
 #include "pnfs.h"
 #include "nfs4session.h"
 #include "internal.h"
+#include "delegation.h"
 
 #define NFSDBG_FACILITY NFSDBG_PROC
 static int nfs42_do_offload_cancel_async(struct file *dst, nfs4_stateid *std);
 
+static void nfs42_set_netaddr(struct file *filep, struct nfs42_netaddr *naddr)
+{
+	struct nfs_client *clp = (NFS_SERVER(file_inode(filep)))->nfs_client;
+	unsigned short port = 2049;
+
+	rcu_read_lock();
+	naddr->netid_len = scnprintf(naddr->netid,
+					sizeof(naddr->netid), "%s",
+					rpc_peeraddr2str(clp->cl_rpcclient,
+					RPC_DISPLAY_NETID));
+	naddr->addr_len = scnprintf(naddr->addr,
+					sizeof(naddr->addr),
+					"%s.%u.%u",
+					rpc_peeraddr2str(clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR),
+					port >> 8, port & 255);
+	rcu_read_unlock();
+}
+
 static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 		struct nfs_lock_context *lock, loff_t offset, loff_t len)
 {
@@ -28,7 +49,7 @@ static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 		.falloc_fh	= NFS_FH(inode),
 		.falloc_offset	= offset,
 		.falloc_length	= len,
-		.falloc_bitmask	= server->cache_consistency_bitmask,
+		.falloc_bitmask	= nfs4_fattr_bitmap,
 	};
 	struct nfs42_falloc_res res = {
 		.falloc_server	= server,
@@ -40,8 +61,11 @@ static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 
 	status = nfs4_set_rw_stateid(&args.falloc_stateid, lock->open_context,
 			lock, FMODE_WRITE);
-	if (status)
+	if (status) {
+		if (status == -EAGAIN)
+			status = -NFS4ERR_BAD_STATEID;
 		return status;
+	}
 
 	res.falloc_fattr = nfs_alloc_fattr();
 	if (!res.falloc_fattr)
@@ -132,22 +156,26 @@ out_unlock:
 }
 
 static int handle_async_copy(struct nfs42_copy_res *res,
-			     struct nfs_server *server,
+			     struct nfs_server *dst_server,
+			     struct nfs_server *src_server,
 			     struct file *src,
 			     struct file *dst,
-			     nfs4_stateid *src_stateid)
+			     nfs4_stateid *src_stateid,
+			     bool *restart)
 {
 	struct nfs4_copy_state *copy, *tmp_copy;
 	int status = NFS4_OK;
 	bool found_pending = false;
-	struct nfs_open_context *ctx = nfs_file_open_context(dst);
+	struct nfs_open_context *dst_ctx = nfs_file_open_context(dst);
+	struct nfs_open_context *src_ctx = nfs_file_open_context(src);
 
 	copy = kzalloc(sizeof(struct nfs4_copy_state), GFP_NOFS);
 	if (!copy)
 		return -ENOMEM;
 
-	spin_lock(&server->nfs_client->cl_lock);
-	list_for_each_entry(tmp_copy, &server->nfs_client->pending_cb_stateids,
+	spin_lock(&dst_server->nfs_client->cl_lock);
+	list_for_each_entry(tmp_copy,
+				&dst_server->nfs_client->pending_cb_stateids,
 				copies) {
 		if (memcmp(&res->write_res.stateid, &tmp_copy->stateid,
 				NFS4_STATEID_SIZE))
@@ -157,7 +185,7 @@ static int handle_async_copy(struct nfs42_copy_res *res,
 		break;
 	}
 	if (found_pending) {
-		spin_unlock(&server->nfs_client->cl_lock);
+		spin_unlock(&dst_server->nfs_client->cl_lock);
 		kfree(copy);
 		copy = tmp_copy;
 		goto out;
@@ -165,19 +193,32 @@ static int handle_async_copy(struct nfs42_copy_res *res,
 
 	memcpy(&copy->stateid, &res->write_res.stateid, NFS4_STATEID_SIZE);
 	init_completion(&copy->completion);
-	copy->parent_state = ctx->state;
+	copy->parent_dst_state = dst_ctx->state;
+	copy->parent_src_state = src_ctx->state;
+
+	list_add_tail(&copy->copies, &dst_server->ss_copies);
+	spin_unlock(&dst_server->nfs_client->cl_lock);
 
-	list_add_tail(&copy->copies, &server->ss_copies);
-	spin_unlock(&server->nfs_client->cl_lock);
+	if (dst_server != src_server) {
+		spin_lock(&src_server->nfs_client->cl_lock);
+		list_add_tail(&copy->src_copies, &src_server->ss_copies);
+		spin_unlock(&src_server->nfs_client->cl_lock);
+	}
 
 	status = wait_for_completion_interruptible(&copy->completion);
-	spin_lock(&server->nfs_client->cl_lock);
+	spin_lock(&dst_server->nfs_client->cl_lock);
 	list_del_init(&copy->copies);
-	spin_unlock(&server->nfs_client->cl_lock);
+	spin_unlock(&dst_server->nfs_client->cl_lock);
+	if (dst_server != src_server) {
+		spin_lock(&src_server->nfs_client->cl_lock);
+		list_del_init(&copy->src_copies);
+		spin_unlock(&src_server->nfs_client->cl_lock);
+	}
 	if (status == -ERESTARTSYS) {
 		goto out_cancel;
-	} else if (copy->flags) {
+	} else if (copy->flags || copy->error == NFS4ERR_PARTNER_NO_AUTH) {
 		status = -EAGAIN;
+		*restart = true;
 		goto out_cancel;
 	}
 out:
@@ -185,12 +226,14 @@ out:
 	memcpy(&res->write_res.verifier, &copy->verf, sizeof(copy->verf));
 	status = -copy->error;
 
+out_free:
 	kfree(copy);
 	return status;
 out_cancel:
 	nfs42_do_offload_cancel_async(dst, &copy->stateid);
-	kfree(copy);
-	return status;
+	if (!nfs42_files_from_same_server(src, dst))
+		nfs42_do_offload_cancel_async(src, src_stateid);
+	goto out_free;
 }
 
 static int process_copy_commit(struct file *dst, loff_t pos_dst,
@@ -222,7 +265,10 @@ static ssize_t _nfs42_proc_copy(struct file *src,
 				struct file *dst,
 				struct nfs_lock_context *dst_lock,
 				struct nfs42_copy_args *args,
-				struct nfs42_copy_res *res)
+				struct nfs42_copy_res *res,
+				struct nl4_server *nss,
+				nfs4_stateid *cnr_stateid,
+				bool *restart)
 {
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COPY],
@@ -230,17 +276,26 @@ static ssize_t _nfs42_proc_copy(struct file *src,
 		.rpc_resp = res,
 	};
 	struct inode *dst_inode = file_inode(dst);
-	struct nfs_server *server = NFS_SERVER(dst_inode);
+	struct inode *src_inode = file_inode(src);
+	struct nfs_server *dst_server = NFS_SERVER(dst_inode);
+	struct nfs_server *src_server = NFS_SERVER(src_inode);
 	loff_t pos_src = args->src_pos;
 	loff_t pos_dst = args->dst_pos;
 	size_t count = args->count;
 	ssize_t status;
 
-	status = nfs4_set_rw_stateid(&args->src_stateid, src_lock->open_context,
-				     src_lock, FMODE_READ);
-	if (status)
-		return status;
-
+	if (nss) {
+		args->cp_src = nss;
+		nfs4_stateid_copy(&args->src_stateid, cnr_stateid);
+	} else {
+		status = nfs4_set_rw_stateid(&args->src_stateid,
+				src_lock->open_context, src_lock, FMODE_READ);
+		if (status) {
+			if (status == -EAGAIN)
+				status = -NFS4ERR_BAD_STATEID;
+			return status;
+		}
+	}
 	status = nfs_filemap_write_and_wait_range(file_inode(src)->i_mapping,
 			pos_src, pos_src + (loff_t)count - 1);
 	if (status)
@@ -248,8 +303,11 @@ static ssize_t _nfs42_proc_copy(struct file *src,
 
 	status = nfs4_set_rw_stateid(&args->dst_stateid, dst_lock->open_context,
 				     dst_lock, FMODE_WRITE);
-	if (status)
+	if (status) {
+		if (status == -EAGAIN)
+			status = -NFS4ERR_BAD_STATEID;
 		return status;
+	}
 
 	status = nfs_sync_inode(dst_inode);
 	if (status)
@@ -262,13 +320,15 @@ static ssize_t _nfs42_proc_copy(struct file *src,
 		if (!res->commit_res.verf)
 			return -ENOMEM;
 	}
+	set_bit(NFS_CLNT_SRC_SSC_COPY_STATE,
+		&src_lock->open_context->state->flags);
 	set_bit(NFS_CLNT_DST_SSC_COPY_STATE,
 		&dst_lock->open_context->state->flags);
 
-	status = nfs4_call_sync(server->client, server, &msg,
+	status = nfs4_call_sync(dst_server->client, dst_server, &msg,
 				&args->seq_args, &res->seq_res, 0);
 	if (status == -ENOTSUPP)
-		server->caps &= ~NFS_CAP_COPY;
+		dst_server->caps &= ~NFS_CAP_COPY;
 	if (status)
 		goto out;
 
@@ -280,17 +340,17 @@ static ssize_t _nfs42_proc_copy(struct file *src,
 	}
 
 	if (!res->synchronous) {
-		status = handle_async_copy(res, server, src, dst,
-				&args->src_stateid);
+		status = handle_async_copy(res, dst_server, src_server, src,
+				dst, &args->src_stateid, restart);
 		if (status)
-			return status;
+			goto out;
 	}
 
 	if ((!res->synchronous || !args->sync) &&
 			res->write_res.verifier.committed != NFS_FILE_SYNC) {
 		status = process_copy_commit(dst, pos_dst, res);
 		if (status)
-			return status;
+			goto out;
 	}
 
 	truncate_pagecache_range(dst_inode, pos_dst,
@@ -304,8 +364,9 @@ out:
 }
 
 ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src,
-			struct file *dst, loff_t pos_dst,
-			size_t count)
+			struct file *dst, loff_t pos_dst, size_t count,
+			struct nl4_server *nss,
+			nfs4_stateid *cnr_stateid, bool sync)
 {
 	struct nfs_server *server = NFS_SERVER(file_inode(dst));
 	struct nfs_lock_context *src_lock;
@@ -316,7 +377,7 @@ ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src,
 		.dst_fh		= NFS_FH(file_inode(dst)),
 		.dst_pos	= pos_dst,
 		.count		= count,
-		.sync		= false,
+		.sync		= sync,
 	};
 	struct nfs42_copy_res res;
 	struct nfs4_exception src_exception = {
@@ -328,6 +389,7 @@ ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src,
 		.stateid	= &args.dst_stateid,
 	};
 	ssize_t err, err2;
+	bool restart = false;
 
 	src_lock = nfs_get_lock_context(nfs_file_open_context(src));
 	if (IS_ERR(src_lock))
@@ -347,21 +409,33 @@ ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src,
 		inode_lock(file_inode(dst));
 		err = _nfs42_proc_copy(src, src_lock,
 				dst, dst_lock,
-				&args, &res);
+				&args, &res,
+				nss, cnr_stateid, &restart);
 		inode_unlock(file_inode(dst));
 
 		if (err >= 0)
 			break;
-		if (err == -ENOTSUPP) {
+		if (err == -ENOTSUPP &&
+				nfs42_files_from_same_server(src, dst)) {
 			err = -EOPNOTSUPP;
 			break;
 		} else if (err == -EAGAIN) {
-			dst_exception.retry = 1;
-			continue;
+			if (!restart) {
+				dst_exception.retry = 1;
+				continue;
+			}
+			break;
 		} else if (err == -NFS4ERR_OFFLOAD_NO_REQS && !args.sync) {
 			args.sync = true;
 			dst_exception.retry = 1;
 			continue;
+		} else if ((err == -ESTALE ||
+				err == -NFS4ERR_OFFLOAD_DENIED ||
+				err == -ENOTSUPP) &&
+				!nfs42_files_from_same_server(src, dst)) {
+			nfs42_do_offload_cancel_async(src, &args.src_stateid);
+			err = -EOPNOTSUPP;
+			break;
 		}
 
 		err2 = nfs4_handle_exception(server, err, &src_exception);
@@ -459,6 +533,79 @@ static int nfs42_do_offload_cancel_async(struct file *dst,
 	return status;
 }
 
+static int _nfs42_proc_copy_notify(struct file *src, struct file *dst,
+				   struct nfs42_copy_notify_args *args,
+				   struct nfs42_copy_notify_res *res)
+{
+	struct nfs_server *src_server = NFS_SERVER(file_inode(src));
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COPY_NOTIFY],
+		.rpc_argp = args,
+		.rpc_resp = res,
+	};
+	int status;
+	struct nfs_open_context *ctx;
+	struct nfs_lock_context *l_ctx;
+
+	ctx = get_nfs_open_context(nfs_file_open_context(src));
+	l_ctx = nfs_get_lock_context(ctx);
+	if (IS_ERR(l_ctx))
+		return PTR_ERR(l_ctx);
+
+	status = nfs4_set_rw_stateid(&args->cna_src_stateid, ctx, l_ctx,
+				     FMODE_READ);
+	nfs_put_lock_context(l_ctx);
+	if (status) {
+		if (status == -EAGAIN)
+			status = -NFS4ERR_BAD_STATEID;
+		return status;
+	}
+
+	status = nfs4_call_sync(src_server->client, src_server, &msg,
+				&args->cna_seq_args, &res->cnr_seq_res, 0);
+	if (status == -ENOTSUPP)
+		src_server->caps &= ~NFS_CAP_COPY_NOTIFY;
+
+	put_nfs_open_context(nfs_file_open_context(src));
+	return status;
+}
+
+int nfs42_proc_copy_notify(struct file *src, struct file *dst,
+				struct nfs42_copy_notify_res *res)
+{
+	struct nfs_server *src_server = NFS_SERVER(file_inode(src));
+	struct nfs42_copy_notify_args *args;
+	struct nfs4_exception exception = {
+		.inode = file_inode(src),
+	};
+	int status;
+
+	if (!(src_server->caps & NFS_CAP_COPY_NOTIFY))
+		return -EOPNOTSUPP;
+
+	args = kzalloc(sizeof(struct nfs42_copy_notify_args), GFP_NOFS);
+	if (args == NULL)
+		return -ENOMEM;
+
+	args->cna_src_fh  = NFS_FH(file_inode(src)),
+	args->cna_dst.nl4_type = NL4_NETADDR;
+	nfs42_set_netaddr(dst, &args->cna_dst.u.nl4_addr);
+	exception.stateid = &args->cna_src_stateid;
+
+	do {
+		status = _nfs42_proc_copy_notify(src, dst, args, res);
+		if (status == -ENOTSUPP) {
+			status = -EOPNOTSUPP;
+			goto out;
+		}
+		status = nfs4_handle_exception(src_server, status, &exception);
+	} while (exception.retry);
+
+out:
+	kfree(args);
+	return status;
+}
+
 static loff_t _nfs42_proc_llseek(struct file *filep,
 		struct nfs_lock_context *lock, loff_t offset, int whence)
 {
@@ -483,8 +630,11 @@ static loff_t _nfs42_proc_llseek(struct file *filep,
 
 	status = nfs4_set_rw_stateid(&args.sa_stateid, lock->open_context,
 			lock, FMODE_READ);
-	if (status)
+	if (status) {
+		if (status == -EAGAIN)
+			status = -NFS4ERR_BAD_STATEID;
 		return status;
+	}
 
 	status = nfs_filemap_write_and_wait_range(inode->i_mapping,
 			offset, LLONG_MAX);
@@ -859,13 +1009,18 @@ static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f,
 
 	status = nfs4_set_rw_stateid(&args.src_stateid, src_lock->open_context,
 			src_lock, FMODE_READ);
-	if (status)
+	if (status) {
+		if (status == -EAGAIN)
+			status = -NFS4ERR_BAD_STATEID;
 		return status;
-
+	}
 	status = nfs4_set_rw_stateid(&args.dst_stateid, dst_lock->open_context,
 			dst_lock, FMODE_WRITE);
-	if (status)
+	if (status) {
+		if (status == -EAGAIN)
+			status = -NFS4ERR_BAD_STATEID;
 		return status;
+	}
 
 	res.dst_fattr = nfs_alloc_fattr();
 	if (!res.dst_fattr)
diff --git a/fs/nfs/nfs42xdr.c b/fs/nfs/nfs42xdr.c
index aed865a84629..c03f3246d6c5 100644
--- a/fs/nfs/nfs42xdr.c
+++ b/fs/nfs/nfs42xdr.c
@@ -21,7 +21,10 @@
 #define encode_copy_maxsz		(op_encode_hdr_maxsz +          \
 					 XDR_QUADLEN(NFS4_STATEID_SIZE) + \
 					 XDR_QUADLEN(NFS4_STATEID_SIZE) + \
-					 2 + 2 + 2 + 1 + 1 + 1)
+					 2 + 2 + 2 + 1 + 1 + 1 +\
+					 1 + /* One cnr_source_server */\
+					 1 + /* nl4_type */ \
+					 1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT))
 #define decode_copy_maxsz		(op_decode_hdr_maxsz + \
 					 NFS42_WRITE_RES_SIZE + \
 					 1 /* cr_consecutive */ + \
@@ -29,6 +32,16 @@
 #define encode_offload_cancel_maxsz	(op_encode_hdr_maxsz + \
 					 XDR_QUADLEN(NFS4_STATEID_SIZE))
 #define decode_offload_cancel_maxsz	(op_decode_hdr_maxsz)
+#define encode_copy_notify_maxsz	(op_encode_hdr_maxsz + \
+					 XDR_QUADLEN(NFS4_STATEID_SIZE) + \
+					 1 + /* nl4_type */ \
+					 1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT))
+#define decode_copy_notify_maxsz	(op_decode_hdr_maxsz + \
+					 3 + /* cnr_lease_time */\
+					 XDR_QUADLEN(NFS4_STATEID_SIZE) + \
+					 1 + /* Support 1 cnr_source_server */\
+					 1 + /* nl4_type */ \
+					 1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT))
 #define encode_deallocate_maxsz		(op_encode_hdr_maxsz + \
 					 encode_fallocate_maxsz)
 #define decode_deallocate_maxsz		(op_decode_hdr_maxsz)
@@ -99,6 +112,12 @@
 					 decode_sequence_maxsz + \
 					 decode_putfh_maxsz + \
 					 decode_offload_cancel_maxsz)
+#define NFS4_enc_copy_notify_sz		(compound_encode_hdr_maxsz + \
+					 encode_putfh_maxsz + \
+					 encode_copy_notify_maxsz)
+#define NFS4_dec_copy_notify_sz		(compound_decode_hdr_maxsz + \
+					 decode_putfh_maxsz + \
+					 decode_copy_notify_maxsz)
 #define NFS4_enc_deallocate_sz		(compound_encode_hdr_maxsz + \
 					 encode_sequence_maxsz + \
 					 encode_putfh_maxsz + \
@@ -166,6 +185,26 @@ static void encode_allocate(struct xdr_stream *xdr,
 	encode_fallocate(xdr, args);
 }
 
+static void encode_nl4_server(struct xdr_stream *xdr,
+			      const struct nl4_server *ns)
+{
+	encode_uint32(xdr, ns->nl4_type);
+	switch (ns->nl4_type) {
+	case NL4_NAME:
+	case NL4_URL:
+		encode_string(xdr, ns->u.nl4_str_sz, ns->u.nl4_str);
+		break;
+	case NL4_NETADDR:
+		encode_string(xdr, ns->u.nl4_addr.netid_len,
+			      ns->u.nl4_addr.netid);
+		encode_string(xdr, ns->u.nl4_addr.addr_len,
+			      ns->u.nl4_addr.addr);
+		break;
+	default:
+		WARN_ON_ONCE(1);
+	}
+}
+
 static void encode_copy(struct xdr_stream *xdr,
 			const struct nfs42_copy_args *args,
 			struct compound_hdr *hdr)
@@ -180,7 +219,12 @@ static void encode_copy(struct xdr_stream *xdr,
 
 	encode_uint32(xdr, 1); /* consecutive = true */
 	encode_uint32(xdr, args->sync);
-	encode_uint32(xdr, 0); /* src server list */
+	if (args->cp_src == NULL) { /* intra-ssc */
+		encode_uint32(xdr, 0); /* no src server list */
+		return;
+	}
+	encode_uint32(xdr, 1); /* supporting 1 server */
+	encode_nl4_server(xdr, args->cp_src);
 }
 
 static void encode_offload_cancel(struct xdr_stream *xdr,
@@ -191,6 +235,15 @@ static void encode_offload_cancel(struct xdr_stream *xdr,
 	encode_nfs4_stateid(xdr, &args->osa_stateid);
 }
 
+static void encode_copy_notify(struct xdr_stream *xdr,
+			       const struct nfs42_copy_notify_args *args,
+			       struct compound_hdr *hdr)
+{
+	encode_op_hdr(xdr, OP_COPY_NOTIFY, decode_copy_notify_maxsz, hdr);
+	encode_nfs4_stateid(xdr, &args->cna_src_stateid);
+	encode_nl4_server(xdr, &args->cna_dst);
+}
+
 static void encode_deallocate(struct xdr_stream *xdr,
 			      const struct nfs42_falloc_args *args,
 			      struct compound_hdr *hdr)
@@ -355,6 +408,25 @@ static void nfs4_xdr_enc_offload_cancel(struct rpc_rqst *req,
 }
 
 /*
+ * Encode COPY_NOTIFY request
+ */
+static void nfs4_xdr_enc_copy_notify(struct rpc_rqst *req,
+				     struct xdr_stream *xdr,
+				     const void *data)
+{
+	const struct nfs42_copy_notify_args *args = data;
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->cna_seq_args),
+	};
+
+	encode_compound_hdr(xdr, req, &hdr);
+	encode_sequence(xdr, &args->cna_seq_args, &hdr);
+	encode_putfh(xdr, args->cna_src_fh, &hdr);
+	encode_copy_notify(xdr, args, &hdr);
+	encode_nops(&hdr);
+}
+
+/*
  * Encode DEALLOCATE request
  */
 static void nfs4_xdr_enc_deallocate(struct rpc_rqst *req,
@@ -490,6 +562,58 @@ static int decode_write_response(struct xdr_stream *xdr,
 	return decode_verifier(xdr, &res->verifier.verifier);
 }
 
+static int decode_nl4_server(struct xdr_stream *xdr, struct nl4_server *ns)
+{
+	struct nfs42_netaddr *naddr;
+	uint32_t dummy;
+	char *dummy_str;
+	__be32 *p;
+	int status;
+
+	/* nl_type */
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(!p))
+		return -EIO;
+	ns->nl4_type = be32_to_cpup(p);
+	switch (ns->nl4_type) {
+	case NL4_NAME:
+	case NL4_URL:
+		status = decode_opaque_inline(xdr, &dummy, &dummy_str);
+		if (unlikely(status))
+			return status;
+		if (unlikely(dummy > NFS4_OPAQUE_LIMIT))
+			return -EIO;
+		memcpy(&ns->u.nl4_str, dummy_str, dummy);
+		ns->u.nl4_str_sz = dummy;
+		break;
+	case NL4_NETADDR:
+		naddr = &ns->u.nl4_addr;
+
+		/* netid string */
+		status = decode_opaque_inline(xdr, &dummy, &dummy_str);
+		if (unlikely(status))
+			return status;
+		if (unlikely(dummy > RPCBIND_MAXNETIDLEN))
+			return -EIO;
+		naddr->netid_len = dummy;
+		memcpy(naddr->netid, dummy_str, naddr->netid_len);
+
+		/* uaddr string */
+		status = decode_opaque_inline(xdr, &dummy, &dummy_str);
+		if (unlikely(status))
+			return status;
+		if (unlikely(dummy > RPCBIND_MAXUADDRLEN))
+			return -EIO;
+		naddr->addr_len = dummy;
+		memcpy(naddr->addr, dummy_str, naddr->addr_len);
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		return -EIO;
+	}
+	return 0;
+}
+
 static int decode_copy_requirements(struct xdr_stream *xdr,
 				    struct nfs42_copy_res *res) {
 	__be32 *p;
@@ -529,6 +653,42 @@ static int decode_offload_cancel(struct xdr_stream *xdr,
 	return decode_op_hdr(xdr, OP_OFFLOAD_CANCEL);
 }
 
+static int decode_copy_notify(struct xdr_stream *xdr,
+			      struct nfs42_copy_notify_res *res)
+{
+	__be32 *p;
+	int status, count;
+
+	status = decode_op_hdr(xdr, OP_COPY_NOTIFY);
+	if (status)
+		return status;
+	/* cnr_lease_time */
+	p = xdr_inline_decode(xdr, 12);
+	if (unlikely(!p))
+		return -EIO;
+	p = xdr_decode_hyper(p, &res->cnr_lease_time.seconds);
+	res->cnr_lease_time.nseconds = be32_to_cpup(p);
+
+	status = decode_opaque_fixed(xdr, &res->cnr_stateid, NFS4_STATEID_SIZE);
+	if (unlikely(status))
+		return -EIO;
+
+	/* number of source addresses */
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(!p))
+		return -EIO;
+
+	count = be32_to_cpup(p);
+	if (count > 1)
+		pr_warn("NFS: %s: nsvr %d > Supported. Use first servers\n",
+			 __func__, count);
+
+	status = decode_nl4_server(xdr, &res->cnr_src);
+	if (unlikely(status))
+		return -EIO;
+	return 0;
+}
+
 static int decode_deallocate(struct xdr_stream *xdr, struct nfs42_falloc_res *res)
 {
 	return decode_op_hdr(xdr, OP_DEALLOCATE);
@@ -657,6 +817,32 @@ out:
 }
 
 /*
+ * Decode COPY_NOTIFY response
+ */
+static int nfs4_xdr_dec_copy_notify(struct rpc_rqst *rqstp,
+				    struct xdr_stream *xdr,
+				    void *data)
+{
+	struct nfs42_copy_notify_res *res = data;
+	struct compound_hdr hdr;
+	int status;
+
+	status = decode_compound_hdr(xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(xdr, &res->cnr_seq_res, rqstp);
+	if (status)
+		goto out;
+	status = decode_putfh(xdr);
+	if (status)
+		goto out;
+	status = decode_copy_notify(xdr, res);
+
+out:
+	return status;
+}
+
+/*
  * Decode DEALLOCATE request
  */
 static int nfs4_xdr_dec_deallocate(struct rpc_rqst *rqstp,
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 16b2e5cc3e94..2b7f6dcd2eb8 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -42,7 +42,9 @@ enum nfs4_client_state {
 	NFS4CLNT_LEASE_MOVED,
 	NFS4CLNT_DELEGATION_EXPIRED,
 	NFS4CLNT_RUN_MANAGER,
-	NFS4CLNT_DELEGRETURN_RUNNING,
+	NFS4CLNT_RECALL_RUNNING,
+	NFS4CLNT_RECALL_ANY_LAYOUT_READ,
+	NFS4CLNT_RECALL_ANY_LAYOUT_RW,
 };
 
 #define NFS4_RENEW_TIMEOUT		0x01
@@ -166,9 +168,9 @@ enum {
 	NFS_STATE_RECOVERY_FAILED,	/* OPEN stateid state recovery failed */
 	NFS_STATE_MAY_NOTIFY_LOCK,	/* server may CB_NOTIFY_LOCK */
 	NFS_STATE_CHANGE_WAIT,		/* A state changing operation is outstanding */
-#ifdef CONFIG_NFS_V4_2
 	NFS_CLNT_DST_SSC_COPY_STATE,    /* dst server open state on client*/
-#endif /* CONFIG_NFS_V4_2 */
+	NFS_CLNT_SRC_SSC_COPY_STATE,    /* src server open state on client*/
+	NFS_SRV_SSC_COPY_STATE,		/* ssc state on the dst server */
 };
 
 struct nfs4_state {
@@ -268,14 +270,13 @@ extern const struct dentry_operations nfs4_dentry_operations;
 int nfs_atomic_open(struct inode *, struct dentry *, struct file *,
 		    unsigned, umode_t);
 
-/* super.c */
+/* fs_context.c */
 extern struct file_system_type nfs4_fs_type;
 
 /* nfs4namespace.c */
 struct rpc_clnt *nfs4_negotiate_security(struct rpc_clnt *, struct inode *,
 					 const struct qstr *);
-struct vfsmount *nfs4_submount(struct nfs_server *, struct dentry *,
-			       struct nfs_fh *, struct nfs_fattr *);
+int nfs4_submount(struct fs_context *, struct nfs_server *);
 int nfs4_replace_transport(struct nfs_server *server,
 				const struct nfs4_fs_locations *locations);
 
@@ -303,14 +304,23 @@ extern int nfs4_proc_fs_locations(struct rpc_clnt *, struct inode *, const struc
 extern int nfs4_proc_get_locations(struct inode *, struct nfs4_fs_locations *,
 		struct page *page, const struct cred *);
 extern int nfs4_proc_fsid_present(struct inode *, const struct cred *);
-extern struct rpc_clnt *nfs4_proc_lookup_mountpoint(struct inode *, const struct qstr *,
-			    struct nfs_fh *, struct nfs_fattr *);
+extern struct rpc_clnt *nfs4_proc_lookup_mountpoint(struct inode *,
+						    struct dentry *,
+						    struct nfs_fh *,
+						    struct nfs_fattr *);
 extern int nfs4_proc_secinfo(struct inode *, const struct qstr *, struct nfs4_secinfo_flavors *);
 extern const struct xattr_handler *nfs4_xattr_handlers[];
 extern int nfs4_set_rw_stateid(nfs4_stateid *stateid,
 		const struct nfs_open_context *ctx,
 		const struct nfs_lock_context *l_ctx,
 		fmode_t fmode);
+extern int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
+			     struct nfs_fattr *fattr, struct nfs4_label *label,
+			     struct inode *inode);
+extern int update_open_stateid(struct nfs4_state *state,
+				const nfs4_stateid *open_stateid,
+				const nfs4_stateid *deleg_stateid,
+				fmode_t fmode);
 
 extern int nfs4_proc_get_lease_time(struct nfs_client *clp,
 		struct nfs_fsinfo *fsinfo);
@@ -439,12 +449,12 @@ extern void nfs4_schedule_state_renewal(struct nfs_client *);
 extern void nfs4_renewd_prepare_shutdown(struct nfs_server *);
 extern void nfs4_kill_renewd(struct nfs_client *);
 extern void nfs4_renew_state(struct work_struct *);
-extern void nfs4_set_lease_period(struct nfs_client *clp,
-		unsigned long lease,
-		unsigned long lastrenewed);
+extern void nfs4_set_lease_period(struct nfs_client *clp, unsigned long lease);
 
 
 /* nfs4state.c */
+extern const nfs4_stateid current_stateid;
+
 const struct cred *nfs4_get_clid_cred(struct nfs_client *clp);
 const struct cred *nfs4_get_machine_cred(struct nfs_client *clp);
 const struct cred *nfs4_get_renew_cred(struct nfs_client *clp);
@@ -457,6 +467,8 @@ int nfs41_discover_server_trunking(struct nfs_client *clp,
 			struct nfs_client **, const struct cred *);
 extern void nfs4_schedule_session_recovery(struct nfs4_session *, int);
 extern void nfs41_notify_server(struct nfs_client *);
+bool nfs4_check_serverowner_major_id(struct nfs41_server_owner *o1,
+			struct nfs41_server_owner *o2);
 #else
 static inline void nfs4_schedule_session_recovery(struct nfs4_session *session, int err)
 {
@@ -515,7 +527,6 @@ extern const nfs4_stateid invalid_stateid;
 /* nfs4super.c */
 struct nfs_mount_info;
 extern struct nfs_subversion nfs_v4;
-struct dentry *nfs4_try_mount(int, const char *, struct nfs_mount_info *, struct nfs_subversion *);
 extern bool nfs4_disable_idmapping;
 extern unsigned short max_session_slots;
 extern unsigned short max_session_cb_slots;
@@ -525,6 +536,9 @@ extern bool recover_lost_locks;
 #define NFS4_CLIENT_ID_UNIQ_LEN		(64)
 extern char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN];
 
+extern int nfs4_try_get_tree(struct fs_context *);
+extern int nfs4_get_referral_tree(struct fs_context *);
+
 /* nfs4sysctl.c */
 #ifdef CONFIG_SYSCTL
 int nfs4_register_sysctl(void);
@@ -572,6 +586,12 @@ static inline bool nfs4_stateid_is_newer(const nfs4_stateid *s1, const nfs4_stat
 	return (s32)(be32_to_cpu(s1->seqid) - be32_to_cpu(s2->seqid)) > 0;
 }
 
+static inline bool nfs4_stateid_match_or_older(const nfs4_stateid *dst, const nfs4_stateid *src)
+{
+	return nfs4_stateid_match_other(dst, src) &&
+		!(src->seqid && nfs4_stateid_is_newer(dst, src));
+}
+
 static inline void nfs4_stateid_seqid_inc(nfs4_stateid *s1)
 {
 	u32 seqid = be32_to_cpu(s1->seqid);
diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c
index da6204025a2d..0bd77cc1f639 100644
--- a/fs/nfs/nfs4client.c
+++ b/fs/nfs/nfs4client.c
@@ -216,7 +216,6 @@ struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *cl_init)
 	INIT_LIST_HEAD(&clp->cl_ds_clients);
 	rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client");
 	clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED;
-	clp->cl_minorversion = cl_init->minorversion;
 	clp->cl_mvops = nfs_v4_minor_ops[cl_init->minorversion];
 	clp->cl_mig_gen = 1;
 #if IS_ENABLED(CONFIG_NFS_V4_1)
@@ -629,7 +628,7 @@ out:
 /*
  * Returns true if the server major ids match
  */
-static bool
+bool
 nfs4_check_serverowner_major_id(struct nfs41_server_owner *o1,
 				struct nfs41_server_owner *o2)
 {
@@ -879,14 +878,17 @@ static int nfs4_set_client(struct nfs_server *server,
 	};
 	struct nfs_client *clp;
 
-	if (minorversion > 0 && proto == XPRT_TRANSPORT_TCP)
+	if (minorversion == 0)
+		__set_bit(NFS_CS_REUSEPORT, &cl_init.init_flags);
+	else if (proto == XPRT_TRANSPORT_TCP)
 		cl_init.nconnect = nconnect;
+
 	if (server->flags & NFS_MOUNT_NORESVPORT)
-		set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
+		__set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
 	if (server->options & NFS_OPTION_MIGRATION)
-		set_bit(NFS_CS_MIGRATION, &cl_init.init_flags);
+		__set_bit(NFS_CS_MIGRATION, &cl_init.init_flags);
 	if (test_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status))
-		set_bit(NFS_CS_TSM_POSSIBLE, &cl_init.init_flags);
+		__set_bit(NFS_CS_TSM_POSSIBLE, &cl_init.init_flags);
 	server->port = rpc_get_port(addr);
 
 	/* Allocate or find a client reference we can use */
@@ -1052,66 +1054,64 @@ out:
 /*
  * Create a version 4 volume record
  */
-static int nfs4_init_server(struct nfs_server *server,
-		struct nfs_parsed_mount_data *data)
+static int nfs4_init_server(struct nfs_server *server, struct fs_context *fc)
 {
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	struct rpc_timeout timeparms;
 	int error;
 
-	nfs_init_timeout_values(&timeparms, data->nfs_server.protocol,
-			data->timeo, data->retrans);
+	nfs_init_timeout_values(&timeparms, ctx->nfs_server.protocol,
+				ctx->timeo, ctx->retrans);
 
 	/* Initialise the client representation from the mount data */
-	server->flags = data->flags;
-	server->options = data->options;
-	server->auth_info = data->auth_info;
+	server->flags = ctx->flags;
+	server->options = ctx->options;
+	server->auth_info = ctx->auth_info;
 
 	/* Use the first specified auth flavor. If this flavor isn't
 	 * allowed by the server, use the SECINFO path to try the
 	 * other specified flavors */
-	if (data->auth_info.flavor_len >= 1)
-		data->selected_flavor = data->auth_info.flavors[0];
+	if (ctx->auth_info.flavor_len >= 1)
+		ctx->selected_flavor = ctx->auth_info.flavors[0];
 	else
-		data->selected_flavor = RPC_AUTH_UNIX;
+		ctx->selected_flavor = RPC_AUTH_UNIX;
 
 	/* Get a client record */
 	error = nfs4_set_client(server,
-			data->nfs_server.hostname,
-			(const struct sockaddr *)&data->nfs_server.address,
-			data->nfs_server.addrlen,
-			data->client_address,
-			data->nfs_server.protocol,
-			&timeparms,
-			data->minorversion,
-			data->nfs_server.nconnect,
-			data->net);
+				ctx->nfs_server.hostname,
+				&ctx->nfs_server.address,
+				ctx->nfs_server.addrlen,
+				ctx->client_address,
+				ctx->nfs_server.protocol,
+				&timeparms,
+				ctx->minorversion,
+				ctx->nfs_server.nconnect,
+				fc->net_ns);
 	if (error < 0)
 		return error;
 
-	if (data->rsize)
-		server->rsize = nfs_block_size(data->rsize, NULL);
-	if (data->wsize)
-		server->wsize = nfs_block_size(data->wsize, NULL);
+	if (ctx->rsize)
+		server->rsize = nfs_block_size(ctx->rsize, NULL);
+	if (ctx->wsize)
+		server->wsize = nfs_block_size(ctx->wsize, NULL);
 
-	server->acregmin = data->acregmin * HZ;
-	server->acregmax = data->acregmax * HZ;
-	server->acdirmin = data->acdirmin * HZ;
-	server->acdirmax = data->acdirmax * HZ;
-	server->port     = data->nfs_server.port;
+	server->acregmin = ctx->acregmin * HZ;
+	server->acregmax = ctx->acregmax * HZ;
+	server->acdirmin = ctx->acdirmin * HZ;
+	server->acdirmax = ctx->acdirmax * HZ;
+	server->port     = ctx->nfs_server.port;
 
 	return nfs_init_server_rpcclient(server, &timeparms,
-					 data->selected_flavor);
+					 ctx->selected_flavor);
 }
 
 /*
  * Create a version 4 volume record
  * - keyed on server and FSID
  */
-/*struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
-				      struct nfs_fh *mntfh)*/
-struct nfs_server *nfs4_create_server(struct nfs_mount_info *mount_info,
-				      struct nfs_subversion *nfs_mod)
+struct nfs_server *nfs4_create_server(struct fs_context *fc)
 {
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	struct nfs_server *server;
 	bool auth_probe;
 	int error;
@@ -1122,14 +1122,14 @@ struct nfs_server *nfs4_create_server(struct nfs_mount_info *mount_info,
 
 	server->cred = get_cred(current_cred());
 
-	auth_probe = mount_info->parsed->auth_info.flavor_len < 1;
+	auth_probe = ctx->auth_info.flavor_len < 1;
 
 	/* set up the general RPC client */
-	error = nfs4_init_server(server, mount_info->parsed);
+	error = nfs4_init_server(server, fc);
 	if (error < 0)
 		goto error;
 
-	error = nfs4_server_common_setup(server, mount_info->mntfh, auth_probe);
+	error = nfs4_server_common_setup(server, ctx->mntfh, auth_probe);
 	if (error < 0)
 		goto error;
 
@@ -1143,9 +1143,9 @@ error:
 /*
  * Create an NFS4 referral server record
  */
-struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
-					       struct nfs_fh *mntfh)
+struct nfs_server *nfs4_create_referral_server(struct fs_context *fc)
 {
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	struct nfs_client *parent_client;
 	struct nfs_server *server, *parent_server;
 	bool auth_probe;
@@ -1155,7 +1155,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 	if (!server)
 		return ERR_PTR(-ENOMEM);
 
-	parent_server = NFS_SB(data->sb);
+	parent_server = NFS_SB(ctx->clone_data.sb);
 	parent_client = parent_server->nfs_client;
 
 	server->cred = get_cred(parent_server->cred);
@@ -1165,10 +1165,11 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 
 	/* Get a client representation */
 #if IS_ENABLED(CONFIG_SUNRPC_XPRT_RDMA)
-	rpc_set_port(data->addr, NFS_RDMA_PORT);
-	error = nfs4_set_client(server, data->hostname,
-				data->addr,
-				data->addrlen,
+	rpc_set_port(&ctx->nfs_server.address, NFS_RDMA_PORT);
+	error = nfs4_set_client(server,
+				ctx->nfs_server.hostname,
+				&ctx->nfs_server.address,
+				ctx->nfs_server.addrlen,
 				parent_client->cl_ipaddr,
 				XPRT_TRANSPORT_RDMA,
 				parent_server->client->cl_timeout,
@@ -1179,10 +1180,11 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 		goto init_server;
 #endif	/* IS_ENABLED(CONFIG_SUNRPC_XPRT_RDMA) */
 
-	rpc_set_port(data->addr, NFS_PORT);
-	error = nfs4_set_client(server, data->hostname,
-				data->addr,
-				data->addrlen,
+	rpc_set_port(&ctx->nfs_server.address, NFS_PORT);
+	error = nfs4_set_client(server,
+				ctx->nfs_server.hostname,
+				&ctx->nfs_server.address,
+				ctx->nfs_server.addrlen,
 				parent_client->cl_ipaddr,
 				XPRT_TRANSPORT_TCP,
 				parent_server->client->cl_timeout,
@@ -1195,13 +1197,14 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 #if IS_ENABLED(CONFIG_SUNRPC_XPRT_RDMA)
 init_server:
 #endif
-	error = nfs_init_server_rpcclient(server, parent_server->client->cl_timeout, data->authflavor);
+	error = nfs_init_server_rpcclient(server, parent_server->client->cl_timeout,
+					  ctx->selected_flavor);
 	if (error < 0)
 		goto error;
 
 	auth_probe = parent_server->auth_info.flavor_len < 1;
 
-	error = nfs4_server_common_setup(server, mntfh, auth_probe);
+	error = nfs4_server_common_setup(server, ctx->mntfh, auth_probe);
 	if (error < 0)
 		goto error;
 
diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c
index 339663d04bf8..8e5d6223ddd3 100644
--- a/fs/nfs/nfs4file.c
+++ b/fs/nfs/nfs4file.c
@@ -7,6 +7,7 @@
 #include <linux/fs.h>
 #include <linux/file.h>
 #include <linux/falloc.h>
+#include <linux/mount.h>
 #include <linux/nfs_fs.h>
 #include "delegation.h"
 #include "internal.h"
@@ -86,7 +87,6 @@ nfs4_file_open(struct inode *inode, struct file *filp)
 	if (inode != d_inode(dentry))
 		goto out_drop;
 
-	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
 	nfs_file_set_open_context(filp, ctx);
 	nfs_fscache_open_file(inode, filp);
 	err = 0;
@@ -133,14 +133,55 @@ static ssize_t __nfs4_copy_file_range(struct file *file_in, loff_t pos_in,
 				      struct file *file_out, loff_t pos_out,
 				      size_t count, unsigned int flags)
 {
+	struct nfs42_copy_notify_res *cn_resp = NULL;
+	struct nl4_server *nss = NULL;
+	nfs4_stateid *cnrs = NULL;
+	ssize_t ret;
+	bool sync = false;
+
 	/* Only offload copy if superblock is the same */
-	if (file_inode(file_in)->i_sb != file_inode(file_out)->i_sb)
+	if (file_in->f_op != &nfs4_file_operations)
 		return -EXDEV;
 	if (!nfs_server_capable(file_inode(file_out), NFS_CAP_COPY))
 		return -EOPNOTSUPP;
 	if (file_inode(file_in) == file_inode(file_out))
 		return -EOPNOTSUPP;
-	return nfs42_proc_copy(file_in, pos_in, file_out, pos_out, count);
+	/* if the copy size if smaller than 2 RPC payloads, make it
+	 * synchronous
+	 */
+	if (count <= 2 * NFS_SERVER(file_inode(file_in))->rsize)
+		sync = true;
+retry:
+	if (!nfs42_files_from_same_server(file_in, file_out)) {
+		/* for inter copy, if copy size if smaller than 12 RPC
+		 * payloads, fallback to traditional copy. There are
+		 * 14 RPCs during an NFSv4.x mount between source/dest
+		 * servers.
+		 */
+		if (sync ||
+			count <= 14 * NFS_SERVER(file_inode(file_in))->rsize)
+			return -EOPNOTSUPP;
+		cn_resp = kzalloc(sizeof(struct nfs42_copy_notify_res),
+				GFP_NOFS);
+		if (unlikely(cn_resp == NULL))
+			return -ENOMEM;
+
+		ret = nfs42_proc_copy_notify(file_in, file_out, cn_resp);
+		if (ret) {
+			ret = -EOPNOTSUPP;
+			goto out;
+		}
+		nss = &cn_resp->cnr_src;
+		cnrs = &cn_resp->cnr_stateid;
+	}
+	ret = nfs42_proc_copy(file_in, pos_in, file_out, pos_out, count,
+				nss, cnrs, sync);
+out:
+	if (!nfs42_files_from_same_server(file_in, file_out))
+		kfree(cn_resp);
+	if (ret == -EAGAIN)
+		goto retry;
+	return ret;
 }
 
 static ssize_t nfs4_copy_file_range(struct file *file_in, loff_t pos_in,
@@ -211,6 +252,9 @@ static loff_t nfs42_remap_file_range(struct file *src_file, loff_t src_off,
 	if (remap_flags & ~REMAP_FILE_ADVISORY)
 		return -EINVAL;
 
+	if (IS_SWAPFILE(dst_inode) || IS_SWAPFILE(src_inode))
+		return -ETXTBSY;
+
 	/* check alignment w.r.t. clone_blksize */
 	ret = -EINVAL;
 	if (bs) {
@@ -263,6 +307,102 @@ out_unlock:
 out:
 	return ret < 0 ? ret : count;
 }
+
+static int read_name_gen = 1;
+#define SSC_READ_NAME_BODY "ssc_read_%d"
+
+struct file *
+nfs42_ssc_open(struct vfsmount *ss_mnt, struct nfs_fh *src_fh,
+		nfs4_stateid *stateid)
+{
+	struct nfs_fattr fattr;
+	struct file *filep, *res;
+	struct nfs_server *server;
+	struct inode *r_ino = NULL;
+	struct nfs_open_context *ctx;
+	struct nfs4_state_owner *sp;
+	char *read_name = NULL;
+	int len, status = 0;
+
+	server = NFS_SERVER(ss_mnt->mnt_root->d_inode);
+
+	nfs_fattr_init(&fattr);
+
+	status = nfs4_proc_getattr(server, src_fh, &fattr, NULL, NULL);
+	if (status < 0) {
+		res = ERR_PTR(status);
+		goto out;
+	}
+
+	res = ERR_PTR(-ENOMEM);
+	len = strlen(SSC_READ_NAME_BODY) + 16;
+	read_name = kzalloc(len, GFP_NOFS);
+	if (read_name == NULL)
+		goto out;
+	snprintf(read_name, len, SSC_READ_NAME_BODY, read_name_gen++);
+
+	r_ino = nfs_fhget(ss_mnt->mnt_root->d_inode->i_sb, src_fh, &fattr,
+			NULL);
+	if (IS_ERR(r_ino)) {
+		res = ERR_CAST(r_ino);
+		goto out_free_name;
+	}
+
+	filep = alloc_file_pseudo(r_ino, ss_mnt, read_name, FMODE_READ,
+				     r_ino->i_fop);
+	if (IS_ERR(filep)) {
+		res = ERR_CAST(filep);
+		goto out_free_name;
+	}
+	filep->f_mode |= FMODE_READ;
+
+	ctx = alloc_nfs_open_context(filep->f_path.dentry, filep->f_mode,
+					filep);
+	if (IS_ERR(ctx)) {
+		res = ERR_CAST(ctx);
+		goto out_filep;
+	}
+
+	res = ERR_PTR(-EINVAL);
+	sp = nfs4_get_state_owner(server, ctx->cred, GFP_KERNEL);
+	if (sp == NULL)
+		goto out_ctx;
+
+	ctx->state = nfs4_get_open_state(r_ino, sp);
+	if (ctx->state == NULL)
+		goto out_stateowner;
+
+	set_bit(NFS_SRV_SSC_COPY_STATE, &ctx->state->flags);
+	set_bit(NFS_OPEN_STATE, &ctx->state->flags);
+	memcpy(&ctx->state->open_stateid.other, &stateid->other,
+	       NFS4_STATEID_OTHER_SIZE);
+	update_open_stateid(ctx->state, stateid, NULL, filep->f_mode);
+
+	nfs_file_set_open_context(filep, ctx);
+	put_nfs_open_context(ctx);
+
+	file_ra_state_init(&filep->f_ra, filep->f_mapping->host->i_mapping);
+	res = filep;
+out_free_name:
+	kfree(read_name);
+out:
+	return res;
+out_stateowner:
+	nfs4_put_state_owner(sp);
+out_ctx:
+	put_nfs_open_context(ctx);
+out_filep:
+	fput(filep);
+	goto out_free_name;
+}
+EXPORT_SYMBOL_GPL(nfs42_ssc_open);
+void nfs42_ssc_close(struct file *filep)
+{
+	struct nfs_open_context *ctx = nfs_file_open_context(filep);
+
+	ctx->state->flags = 0;
+}
+EXPORT_SYMBOL_GPL(nfs42_ssc_close);
 #endif /* CONFIG_NFS_V4_2 */
 
 const struct file_operations nfs4_file_operations = {
diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c
index 2e460c33ae48..a3ab6e219061 100644
--- a/fs/nfs/nfs4namespace.c
+++ b/fs/nfs/nfs4namespace.c
@@ -8,6 +8,7 @@
  * NFSv4 namespace
  */
 
+#include <linux/module.h>
 #include <linux/dcache.h>
 #include <linux/mount.h>
 #include <linux/namei.h>
@@ -21,37 +22,64 @@
 #include <linux/inet.h>
 #include "internal.h"
 #include "nfs4_fs.h"
+#include "nfs.h"
 #include "dns_resolve.h"
 
 #define NFSDBG_FACILITY		NFSDBG_VFS
 
 /*
+ * Work out the length that an NFSv4 path would render to as a standard posix
+ * path, with a leading slash but no terminating slash.
+ */
+static ssize_t nfs4_pathname_len(const struct nfs4_pathname *pathname)
+{
+	ssize_t len = 0;
+	int i;
+
+	for (i = 0; i < pathname->ncomponents; i++) {
+		const struct nfs4_string *component = &pathname->components[i];
+
+		if (component->len > NAME_MAX)
+			goto too_long;
+		len += 1 + component->len; /* Adding "/foo" */
+		if (len > PATH_MAX)
+			goto too_long;
+	}
+	return len;
+
+too_long:
+	return -ENAMETOOLONG;
+}
+
+/*
  * Convert the NFSv4 pathname components into a standard posix path.
- *
- * Note that the resulting string will be placed at the end of the buffer
  */
-static inline char *nfs4_pathname_string(const struct nfs4_pathname *pathname,
-					 char *buffer, ssize_t buflen)
+static char *nfs4_pathname_string(const struct nfs4_pathname *pathname,
+				  unsigned short *_len)
 {
-	char *end = buffer + buflen;
-	int n;
+	ssize_t len;
+	char *buf, *p;
+	int i;
+
+	len = nfs4_pathname_len(pathname);
+	if (len < 0)
+		return ERR_PTR(len);
+	*_len = len;
+
+	p = buf = kmalloc(len + 1, GFP_KERNEL);
+	if (!buf)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; i < pathname->ncomponents; i++) {
+		const struct nfs4_string *component = &pathname->components[i];
 
-	*--end = '\0';
-	buflen--;
-
-	n = pathname->ncomponents;
-	while (--n >= 0) {
-		const struct nfs4_string *component = &pathname->components[n];
-		buflen -= component->len + 1;
-		if (buflen < 0)
-			goto Elong;
-		end -= component->len;
-		memcpy(end, component->data, component->len);
-		*--end = '/';
+		*p++ = '/';
+		memcpy(p, component->data, component->len);
+		p += component->len;
 	}
-	return end;
-Elong:
-	return ERR_PTR(-ENAMETOOLONG);
+
+	*p = 0;
+	return buf;
 }
 
 /*
@@ -100,21 +128,36 @@ static char *nfs4_path(struct dentry *dentry, char *buffer, ssize_t buflen)
  */
 static int nfs4_validate_fspath(struct dentry *dentry,
 				const struct nfs4_fs_locations *locations,
-				char *page, char *page2)
+				struct nfs_fs_context *ctx)
 {
-	const char *path, *fs_path;
+	const char *path;
+	char *fs_path;
+	unsigned short len;
+	char *buf;
+	int n;
 
-	path = nfs4_path(dentry, page, PAGE_SIZE);
-	if (IS_ERR(path))
+	buf = kmalloc(4096, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	path = nfs4_path(dentry, buf, 4096);
+	if (IS_ERR(path)) {
+		kfree(buf);
 		return PTR_ERR(path);
+	}
 
-	fs_path = nfs4_pathname_string(&locations->fs_path, page2, PAGE_SIZE);
-	if (IS_ERR(fs_path))
+	fs_path = nfs4_pathname_string(&locations->fs_path, &len);
+	if (IS_ERR(fs_path)) {
+		kfree(buf);
 		return PTR_ERR(fs_path);
+	}
 
-	if (strncmp(path, fs_path, strlen(fs_path)) != 0) {
+	n = strncmp(path, fs_path, len);
+	kfree(buf);
+	kfree(fs_path);
+	if (n != 0) {
 		dprintk("%s: path %s does not begin with fsroot %s\n",
-			__func__, path, fs_path);
+			__func__, path, ctx->nfs_server.export_path);
 		return -ENOENT;
 	}
 
@@ -236,95 +279,102 @@ out:
 	return new;
 }
 
-static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
-				     char *page, char *page2,
-				     const struct nfs4_fs_location *location)
+static int try_location(struct fs_context *fc,
+			const struct nfs4_fs_location *location)
 {
-	const size_t addr_bufsize = sizeof(struct sockaddr_storage);
-	struct net *net = rpc_net_ns(NFS_SB(mountdata->sb)->client);
-	struct vfsmount *mnt = ERR_PTR(-ENOENT);
-	char *mnt_path;
-	unsigned int maxbuflen;
-	unsigned int s;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	unsigned int len, s;
+	char *export_path, *source, *p;
+	int ret = -ENOENT;
+
+	/* Allocate a buffer big enough to hold any of the hostnames plus a
+	 * terminating char and also a buffer big enough to hold the hostname
+	 * plus a colon plus the path.
+	 */
+	len = 0;
+	for (s = 0; s < location->nservers; s++) {
+		const struct nfs4_string *buf = &location->servers[s];
+		if (buf->len > len)
+			len = buf->len;
+	}
 
-	mnt_path = nfs4_pathname_string(&location->rootpath, page2, PAGE_SIZE);
-	if (IS_ERR(mnt_path))
-		return ERR_CAST(mnt_path);
-	mountdata->mnt_path = mnt_path;
-	maxbuflen = mnt_path - 1 - page2;
+	kfree(ctx->nfs_server.hostname);
+	ctx->nfs_server.hostname = kmalloc(len + 1, GFP_KERNEL);
+	if (!ctx->nfs_server.hostname)
+		return -ENOMEM;
 
-	mountdata->addr = kmalloc(addr_bufsize, GFP_KERNEL);
-	if (mountdata->addr == NULL)
-		return ERR_PTR(-ENOMEM);
+	export_path = nfs4_pathname_string(&location->rootpath,
+					   &ctx->nfs_server.export_path_len);
+	if (IS_ERR(export_path))
+		return PTR_ERR(export_path);
+
+	ctx->nfs_server.export_path = export_path;
+
+	source = kmalloc(len + 1 + ctx->nfs_server.export_path_len + 1,
+			 GFP_KERNEL);
+	if (!source)
+		return -ENOMEM;
 
+	kfree(fc->source);
+	fc->source = source;
 	for (s = 0; s < location->nservers; s++) {
 		const struct nfs4_string *buf = &location->servers[s];
 
-		if (buf->len <= 0 || buf->len >= maxbuflen)
-			continue;
-
 		if (memchr(buf->data, IPV6_SCOPE_DELIMITER, buf->len))
 			continue;
 
-		mountdata->addrlen = nfs_parse_server_name(buf->data, buf->len,
-				mountdata->addr, addr_bufsize, net);
-		if (mountdata->addrlen == 0)
+		ctx->nfs_server.addrlen =
+			nfs_parse_server_name(buf->data, buf->len,
+					      &ctx->nfs_server.address,
+					      sizeof(ctx->nfs_server._address),
+					      fc->net_ns);
+		if (ctx->nfs_server.addrlen == 0)
 			continue;
 
-		memcpy(page2, buf->data, buf->len);
-		page2[buf->len] = '\0';
-		mountdata->hostname = page2;
+		rpc_set_port(&ctx->nfs_server.address, NFS_PORT);
 
-		snprintf(page, PAGE_SIZE, "%s:%s",
-				mountdata->hostname,
-				mountdata->mnt_path);
+		memcpy(ctx->nfs_server.hostname, buf->data, buf->len);
+		ctx->nfs_server.hostname[buf->len] = '\0';
 
-		mnt = vfs_submount(mountdata->dentry, &nfs4_referral_fs_type, page, mountdata);
-		if (!IS_ERR(mnt))
-			break;
+		p = source;
+		memcpy(p, buf->data, buf->len);
+		p += buf->len;
+		*p++ = ':';
+		memcpy(p, ctx->nfs_server.export_path, ctx->nfs_server.export_path_len);
+		p += ctx->nfs_server.export_path_len;
+		*p = 0;
+
+		ret = nfs4_get_referral_tree(fc);
+		if (ret == 0)
+			return 0;
 	}
-	kfree(mountdata->addr);
-	return mnt;
+
+	return ret;
 }
 
 /**
  * nfs_follow_referral - set up mountpoint when hitting a referral on moved error
- * @dentry: parent directory
+ * @fc: pointer to struct nfs_fs_context
  * @locations: array of NFSv4 server location information
  *
  */
-static struct vfsmount *nfs_follow_referral(struct dentry *dentry,
-					    const struct nfs4_fs_locations *locations)
+static int nfs_follow_referral(struct fs_context *fc,
+			       const struct nfs4_fs_locations *locations)
 {
-	struct vfsmount *mnt = ERR_PTR(-ENOENT);
-	struct nfs_clone_mount mountdata = {
-		.sb = dentry->d_sb,
-		.dentry = dentry,
-		.authflavor = NFS_SB(dentry->d_sb)->client->cl_auth->au_flavor,
-	};
-	char *page = NULL, *page2 = NULL;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	int loc, error;
 
 	if (locations == NULL || locations->nlocations <= 0)
-		goto out;
-
-	dprintk("%s: referral at %pd2\n", __func__, dentry);
-
-	page = (char *) __get_free_page(GFP_USER);
-	if (!page)
-		goto out;
+		return -ENOENT;
 
-	page2 = (char *) __get_free_page(GFP_USER);
-	if (!page2)
-		goto out;
+	dprintk("%s: referral at %pd2\n", __func__, ctx->clone_data.dentry);
 
 	/* Ensure fs path is a prefix of current dentry path */
-	error = nfs4_validate_fspath(dentry, locations, page, page2);
-	if (error < 0) {
-		mnt = ERR_PTR(error);
-		goto out;
-	}
+	error = nfs4_validate_fspath(ctx->clone_data.dentry, locations, ctx);
+	if (error < 0)
+		return error;
 
+	error = -ENOENT;
 	for (loc = 0; loc < locations->nlocations; loc++) {
 		const struct nfs4_fs_location *location = &locations->locations[loc];
 
@@ -332,15 +382,12 @@ static struct vfsmount *nfs_follow_referral(struct dentry *dentry,
 		    location->rootpath.ncomponents == 0)
 			continue;
 
-		mnt = try_location(&mountdata, page, page2, location);
-		if (!IS_ERR(mnt))
-			break;
+		error = try_location(fc, location);
+		if (error == 0)
+			return 0;
 	}
 
-out:
-	free_page((unsigned long) page);
-	free_page((unsigned long) page2);
-	return mnt;
+	return error;
 }
 
 /*
@@ -348,71 +395,72 @@ out:
  * @dentry - dentry of referral
  *
  */
-static struct vfsmount *nfs_do_refmount(struct rpc_clnt *client, struct dentry *dentry)
+static int nfs_do_refmount(struct fs_context *fc, struct rpc_clnt *client)
 {
-	struct vfsmount *mnt = ERR_PTR(-ENOMEM);
-	struct dentry *parent;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct dentry *dentry, *parent;
 	struct nfs4_fs_locations *fs_locations = NULL;
 	struct page *page;
-	int err;
+	int err = -ENOMEM;
 
 	/* BUG_ON(IS_ROOT(dentry)); */
 	page = alloc_page(GFP_KERNEL);
-	if (page == NULL)
-		return mnt;
+	if (!page)
+		return -ENOMEM;
 
 	fs_locations = kmalloc(sizeof(struct nfs4_fs_locations), GFP_KERNEL);
-	if (fs_locations == NULL)
+	if (!fs_locations)
 		goto out_free;
 
 	/* Get locations */
-	mnt = ERR_PTR(-ENOENT);
-
+	dentry = ctx->clone_data.dentry;
 	parent = dget_parent(dentry);
 	dprintk("%s: getting locations for %pd2\n",
 		__func__, dentry);
 
 	err = nfs4_proc_fs_locations(client, d_inode(parent), &dentry->d_name, fs_locations, page);
 	dput(parent);
-	if (err != 0 ||
-	    fs_locations->nlocations <= 0 ||
+	if (err != 0)
+		goto out_free_2;
+
+	err = -ENOENT;
+	if (fs_locations->nlocations <= 0 ||
 	    fs_locations->fs_path.ncomponents <= 0)
-		goto out_free;
+		goto out_free_2;
 
-	mnt = nfs_follow_referral(dentry, fs_locations);
+	err = nfs_follow_referral(fc, fs_locations);
+out_free_2:
+	kfree(fs_locations);
 out_free:
 	__free_page(page);
-	kfree(fs_locations);
-	return mnt;
+	return err;
 }
 
-struct vfsmount *nfs4_submount(struct nfs_server *server, struct dentry *dentry,
-			       struct nfs_fh *fh, struct nfs_fattr *fattr)
+int nfs4_submount(struct fs_context *fc, struct nfs_server *server)
 {
-	rpc_authflavor_t flavor = server->client->cl_auth->au_flavor;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct dentry *dentry = ctx->clone_data.dentry;
 	struct dentry *parent = dget_parent(dentry);
 	struct inode *dir = d_inode(parent);
-	const struct qstr *name = &dentry->d_name;
 	struct rpc_clnt *client;
-	struct vfsmount *mnt;
+	int ret;
 
 	/* Look it up again to get its attributes and sec flavor */
-	client = nfs4_proc_lookup_mountpoint(dir, name, fh, fattr);
+	client = nfs4_proc_lookup_mountpoint(dir, dentry, ctx->mntfh,
+					     ctx->clone_data.fattr);
 	dput(parent);
 	if (IS_ERR(client))
-		return ERR_CAST(client);
+		return PTR_ERR(client);
 
-	if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) {
-		mnt = nfs_do_refmount(client, dentry);
-		goto out;
+	ctx->selected_flavor = client->cl_auth->au_flavor;
+	if (ctx->clone_data.fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) {
+		ret = nfs_do_refmount(fc, client);
+	} else {
+		ret = nfs_do_submount(fc);
 	}
 
-	if (client->cl_auth->au_flavor != flavor)
-		flavor = client->cl_auth->au_flavor;
-	mnt = nfs_do_submount(dentry, fh, fattr, flavor);
-out:
 	rpc_shutdown_client(client);
-	return mnt;
+	return ret;
 }
 
 /*
@@ -453,7 +501,7 @@ static int nfs4_try_replacing_one_location(struct nfs_server *server,
 		rpc_set_port(sap, NFS_PORT);
 
 		error = -ENOMEM;
-		hostname = kstrndup(buf->data, buf->len, GFP_KERNEL);
+		hostname = kmemdup_nul(buf->data, buf->len, GFP_KERNEL);
 		if (hostname == NULL)
 			break;
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index caacf5e7f5e1..9056f3dd380e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -91,7 +91,6 @@ struct nfs4_opendata;
 static int _nfs4_recover_proc_open(struct nfs4_opendata *data);
 static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
 static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr);
-static int nfs4_proc_getattr(struct nfs_server *, struct nfs_fh *, struct nfs_fattr *, struct nfs4_label *label, struct inode *inode);
 static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct nfs4_label *label, struct inode *inode);
 static int nfs4_do_setattr(struct inode *inode, const struct cred *cred,
 			    struct nfs_fattr *fattr, struct iattr *sattr,
@@ -476,6 +475,7 @@ static int nfs4_do_handle_exception(struct nfs_server *server,
 		case -NFS4ERR_ADMIN_REVOKED:
 		case -NFS4ERR_EXPIRED:
 		case -NFS4ERR_BAD_STATEID:
+		case -NFS4ERR_PARTNER_NO_AUTH:
 			if (inode != NULL && stateid != NULL) {
 				nfs_inode_find_state_and_recover(inode,
 						stateid);
@@ -521,9 +521,7 @@ static int nfs4_do_handle_exception(struct nfs_server *server,
 		case -NFS4ERR_DEADSESSION:
 		case -NFS4ERR_SEQ_FALSE_RETRY:
 		case -NFS4ERR_SEQ_MISORDERED:
-			dprintk("%s ERROR: %d Reset session\n", __func__,
-				errorcode);
-			nfs4_schedule_session_recovery(clp->cl_session, errorcode);
+			/* Handled in nfs41_sequence_process() */
 			goto wait_on_recovery;
 #endif /* defined(CONFIG_NFS_V4_1) */
 		case -NFS4ERR_FILE_OPEN:
@@ -782,6 +780,7 @@ static int nfs41_sequence_process(struct rpc_task *task,
 	struct nfs4_session *session;
 	struct nfs4_slot *slot = res->sr_slot;
 	struct nfs_client *clp;
+	int status;
 	int ret = 1;
 
 	if (slot == NULL)
@@ -793,8 +792,13 @@ static int nfs41_sequence_process(struct rpc_task *task,
 	session = slot->table->session;
 
 	trace_nfs4_sequence_done(session, res);
+
+	status = res->sr_status;
+	if (task->tk_status == -NFS4ERR_DEADSESSION)
+		status = -NFS4ERR_DEADSESSION;
+
 	/* Check the SEQUENCE operation status */
-	switch (res->sr_status) {
+	switch (status) {
 	case 0:
 		/* Mark this sequence number as having been acked */
 		nfs4_slot_sequence_acked(slot, slot->seq_nr);
@@ -866,6 +870,10 @@ static int nfs41_sequence_process(struct rpc_task *task,
 		 */
 		slot->seq_nr = slot->seq_nr_highest_sent;
 		goto out_retry;
+	case -NFS4ERR_BADSESSION:
+	case -NFS4ERR_DEADSESSION:
+	case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+		goto session_recover;
 	default:
 		/* Just update the slot sequence no. */
 		slot->seq_done = 1;
@@ -876,8 +884,10 @@ out:
 out_noaction:
 	return ret;
 session_recover:
-	nfs4_schedule_session_recovery(session, res->sr_status);
-	goto retry_nowait;
+	nfs4_schedule_session_recovery(session, status);
+	dprintk("%s ERROR: %d Reset session\n", __func__, status);
+	nfs41_sequence_free_slot(res);
+	goto out;
 retry_new_seq:
 	++slot->seq_nr;
 retry_nowait:
@@ -1087,11 +1097,12 @@ static int nfs4_call_sync_custom(struct rpc_task_setup *task_setup)
 	return ret;
 }
 
-static int nfs4_call_sync_sequence(struct rpc_clnt *clnt,
-				   struct nfs_server *server,
-				   struct rpc_message *msg,
-				   struct nfs4_sequence_args *args,
-				   struct nfs4_sequence_res *res)
+static int nfs4_do_call_sync(struct rpc_clnt *clnt,
+			     struct nfs_server *server,
+			     struct rpc_message *msg,
+			     struct nfs4_sequence_args *args,
+			     struct nfs4_sequence_res *res,
+			     unsigned short task_flags)
 {
 	struct nfs_client *clp = server->nfs_client;
 	struct nfs4_call_sync_data data = {
@@ -1103,12 +1114,23 @@ static int nfs4_call_sync_sequence(struct rpc_clnt *clnt,
 		.rpc_client = clnt,
 		.rpc_message = msg,
 		.callback_ops = clp->cl_mvops->call_sync_ops,
-		.callback_data = &data
+		.callback_data = &data,
+		.flags = task_flags,
 	};
 
 	return nfs4_call_sync_custom(&task_setup);
 }
 
+static int nfs4_call_sync_sequence(struct rpc_clnt *clnt,
+				   struct nfs_server *server,
+				   struct rpc_message *msg,
+				   struct nfs4_sequence_args *args,
+				   struct nfs4_sequence_res *res)
+{
+	return nfs4_do_call_sync(clnt, server, msg, args, res, 0);
+}
+
+
 int nfs4_call_sync(struct rpc_clnt *clnt,
 		   struct nfs_server *server,
 		   struct rpc_message *msg,
@@ -1716,7 +1738,7 @@ static void nfs_state_clear_delegation(struct nfs4_state *state)
 	write_sequnlock(&state->seqlock);
 }
 
-static int update_open_stateid(struct nfs4_state *state,
+int update_open_stateid(struct nfs4_state *state,
 		const nfs4_stateid *open_stateid,
 		const nfs4_stateid *delegation,
 		fmode_t fmode)
@@ -1737,7 +1759,7 @@ static int update_open_stateid(struct nfs4_state *state,
 		ret = 1;
 	}
 
-	deleg_cur = rcu_dereference(nfsi->delegation);
+	deleg_cur = nfs4_get_valid_delegation(state->inode);
 	if (deleg_cur == NULL)
 		goto no_delegation;
 
@@ -1749,7 +1771,7 @@ static int update_open_stateid(struct nfs4_state *state,
 
 	if (delegation == NULL)
 		delegation = &deleg_cur->stateid;
-	else if (!nfs4_stateid_match(&deleg_cur->stateid, delegation))
+	else if (!nfs4_stateid_match_other(&deleg_cur->stateid, delegation))
 		goto no_delegation_unlock;
 
 	nfs_mark_delegation_referenced(deleg_cur);
@@ -1796,7 +1818,7 @@ static void nfs4_return_incompatible_delegation(struct inode *inode, fmode_t fmo
 
 	fmode &= FMODE_READ|FMODE_WRITE;
 	rcu_read_lock();
-	delegation = rcu_dereference(NFS_I(inode)->delegation);
+	delegation = nfs4_get_valid_delegation(inode);
 	if (delegation == NULL || (delegation->type & fmode) == fmode) {
 		rcu_read_unlock();
 		return;
@@ -2188,7 +2210,6 @@ static int nfs4_handle_delegation_recall_error(struct nfs_server *server, struct
 		case -NFS4ERR_BAD_HIGH_SLOT:
 		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 		case -NFS4ERR_DEADSESSION:
-			nfs4_schedule_session_recovery(server->nfs_client->cl_session, err);
 			return -EAGAIN;
 		case -NFS4ERR_STALE_CLIENTID:
 		case -NFS4ERR_STALE_STATEID:
@@ -2325,7 +2346,7 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data)
 		.callback_ops = &nfs4_open_confirm_ops,
 		.callback_data = data,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	int status;
 
@@ -2490,7 +2511,7 @@ static int nfs4_run_open_task(struct nfs4_opendata *data,
 		.callback_ops = &nfs4_open_ops,
 		.callback_data = data,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	int status;
 
@@ -2769,16 +2790,19 @@ static int nfs41_check_delegation_stateid(struct nfs4_state *state)
 		return NFS_OK;
 	}
 
+	spin_lock(&delegation->lock);
 	nfs4_stateid_copy(&stateid, &delegation->stateid);
 
 	if (!test_and_clear_bit(NFS_DELEGATION_TEST_EXPIRED,
 				&delegation->flags)) {
+		spin_unlock(&delegation->lock);
 		rcu_read_unlock();
 		return NFS_OK;
 	}
 
 	if (delegation->cred)
 		cred = get_cred(delegation->cred);
+	spin_unlock(&delegation->lock);
 	rcu_read_unlock();
 	status = nfs41_test_and_free_expired_stateid(server, &stateid, cred);
 	trace_nfs4_test_delegation_stateid(state, NULL, status);
@@ -2953,10 +2977,13 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
 	struct dentry *dentry;
 	struct nfs4_state *state;
 	fmode_t acc_mode = _nfs4_ctx_to_accessmode(ctx);
+	struct inode *dir = d_inode(opendata->dir);
+	unsigned long dir_verifier;
 	unsigned int seq;
 	int ret;
 
 	seq = raw_seqcount_begin(&sp->so_reclaim_seqcount);
+	dir_verifier = nfs_save_change_attribute(dir);
 
 	ret = _nfs4_proc_open(opendata, ctx);
 	if (ret != 0)
@@ -2984,8 +3011,19 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
 			dput(ctx->dentry);
 			ctx->dentry = dentry = alias;
 		}
-		nfs_set_verifier(dentry,
-				nfs_save_change_attribute(d_inode(opendata->dir)));
+	}
+
+	switch(opendata->o_arg.claim) {
+	default:
+		break;
+	case NFS4_OPEN_CLAIM_NULL:
+	case NFS4_OPEN_CLAIM_DELEGATE_CUR:
+	case NFS4_OPEN_CLAIM_DELEGATE_PREV:
+		if (!opendata->rpc_done)
+			break;
+		if (opendata->o_res.delegation_type != 0)
+			dir_verifier = nfs_save_change_attribute(dir);
+		nfs_set_verifier(dentry, dir_verifier);
 	}
 
 	/* Parse layoutget results before we check for access */
@@ -3178,6 +3216,11 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir,
 			exception.retry = 1;
 			continue;
 		}
+		if (status == -NFS4ERR_EXPIRED) {
+			nfs4_schedule_lease_recovery(server->nfs_client);
+			exception.retry = 1;
+			continue;
+		}
 		if (status == -EAGAIN) {
 			/* We must have found a delegation */
 			exception.retry = 1;
@@ -3230,6 +3273,8 @@ static int _nfs4_do_setattr(struct inode *inode,
 		nfs_put_lock_context(l_ctx);
 		if (status == -EIO)
 			return -EBADF;
+		else if (status == -EAGAIN)
+			goto zero_stateid;
 	} else {
 zero_stateid:
 		nfs4_stateid_copy(&arg->stateid, &zero_stateid);
@@ -3609,7 +3654,7 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait)
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_close_ops,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	int status = -ENOMEM;
 
@@ -3960,7 +4005,7 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *mntfh,
 {
 	int error;
 	struct nfs_fattr *fattr = info->fattr;
-	struct nfs4_label *label = NULL;
+	struct nfs4_label *label = fattr->label;
 
 	error = nfs4_server_capabilities(server, mntfh);
 	if (error < 0) {
@@ -3968,23 +4013,17 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *mntfh,
 		return error;
 	}
 
-	label = nfs4_label_alloc(server, GFP_KERNEL);
-	if (IS_ERR(label))
-		return PTR_ERR(label);
-
 	error = nfs4_proc_getattr(server, mntfh, fattr, label, NULL);
 	if (error < 0) {
 		dprintk("nfs4_get_root: getattr error = %d\n", -error);
-		goto err_free_label;
+		goto out;
 	}
 
 	if (fattr->valid & NFS_ATTR_FATTR_FSID &&
 	    !nfs_fsid_equal(&server->fsid, &fattr->fsid))
 		memcpy(&server->fsid, &fattr->fsid, sizeof(server->fsid));
 
-err_free_label:
-	nfs4_label_free(label);
-
+out:
 	return error;
 }
 
@@ -4055,14 +4094,21 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_argp = &args,
 		.rpc_resp = &res,
 	};
+	unsigned short task_flags = 0;
+
+	/* Is this is an attribute revalidation, subject to softreval? */
+	if (inode && (server->flags & NFS_MOUNT_SOFTREVAL))
+		task_flags |= RPC_TASK_TIMEOUT;
 
 	nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, label), inode);
 
 	nfs_fattr_init(fattr);
-	return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
+	nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 0);
+	return nfs4_do_call_sync(server->client, server, &msg,
+			&args.seq_args, &res.seq_res, task_flags);
 }
 
-static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
+int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
 				struct nfs_fattr *fattr, struct nfs4_label *label,
 				struct inode *inode)
 {
@@ -4147,7 +4193,7 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
 }
 
 static int _nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir,
-		const struct qstr *name, struct nfs_fh *fhandle,
+		struct dentry *dentry, struct nfs_fh *fhandle,
 		struct nfs_fattr *fattr, struct nfs4_label *label)
 {
 	struct nfs_server *server = NFS_SERVER(dir);
@@ -4155,7 +4201,7 @@ static int _nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir,
 	struct nfs4_lookup_arg args = {
 		.bitmask = server->attr_bitmask,
 		.dir_fh = NFS_FH(dir),
-		.name = name,
+		.name = &dentry->d_name,
 	};
 	struct nfs4_lookup_res res = {
 		.server = server,
@@ -4168,13 +4214,20 @@ static int _nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir,
 		.rpc_argp = &args,
 		.rpc_resp = &res,
 	};
+	unsigned short task_flags = 0;
+
+	/* Is this is an attribute revalidation, subject to softreval? */
+	if (nfs_lookup_is_soft_revalidate(dentry))
+		task_flags |= RPC_TASK_TIMEOUT;
 
 	args.bitmask = nfs4_bitmask(server, label);
 
 	nfs_fattr_init(fattr);
 
-	dprintk("NFS call  lookup %s\n", name->name);
-	status = nfs4_call_sync(clnt, server, &msg, &args.seq_args, &res.seq_res, 0);
+	dprintk("NFS call  lookup %pd2\n", dentry);
+	nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 0);
+	status = nfs4_do_call_sync(clnt, server, &msg,
+			&args.seq_args, &res.seq_res, task_flags);
 	dprintk("NFS reply lookup: %d\n", status);
 	return status;
 }
@@ -4188,16 +4241,17 @@ static void nfs_fixup_secinfo_attributes(struct nfs_fattr *fattr)
 }
 
 static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir,
-				   const struct qstr *name, struct nfs_fh *fhandle,
+				   struct dentry *dentry, struct nfs_fh *fhandle,
 				   struct nfs_fattr *fattr, struct nfs4_label *label)
 {
 	struct nfs4_exception exception = {
 		.interruptible = true,
 	};
 	struct rpc_clnt *client = *clnt;
+	const struct qstr *name = &dentry->d_name;
 	int err;
 	do {
-		err = _nfs4_proc_lookup(client, dir, name, fhandle, fattr, label);
+		err = _nfs4_proc_lookup(client, dir, dentry, fhandle, fattr, label);
 		trace_nfs4_lookup(dir, name, err);
 		switch (err) {
 		case -NFS4ERR_BADNAME:
@@ -4232,14 +4286,14 @@ out:
 	return err;
 }
 
-static int nfs4_proc_lookup(struct inode *dir, const struct qstr *name,
+static int nfs4_proc_lookup(struct inode *dir, struct dentry *dentry,
 			    struct nfs_fh *fhandle, struct nfs_fattr *fattr,
 			    struct nfs4_label *label)
 {
 	int status;
 	struct rpc_clnt *client = NFS_CLIENT(dir);
 
-	status = nfs4_proc_lookup_common(&client, dir, name, fhandle, fattr, label);
+	status = nfs4_proc_lookup_common(&client, dir, dentry, fhandle, fattr, label);
 	if (client != NFS_CLIENT(dir)) {
 		rpc_shutdown_client(client);
 		nfs_fixup_secinfo_attributes(fattr);
@@ -4248,13 +4302,13 @@ static int nfs4_proc_lookup(struct inode *dir, const struct qstr *name,
 }
 
 struct rpc_clnt *
-nfs4_proc_lookup_mountpoint(struct inode *dir, const struct qstr *name,
+nfs4_proc_lookup_mountpoint(struct inode *dir, struct dentry *dentry,
 			    struct nfs_fh *fhandle, struct nfs_fattr *fattr)
 {
 	struct rpc_clnt *client = NFS_CLIENT(dir);
 	int status;
 
-	status = nfs4_proc_lookup_common(&client, dir, name, fhandle, fattr, NULL);
+	status = nfs4_proc_lookup_common(&client, dir, dentry, fhandle, fattr, NULL);
 	if (status < 0)
 		return ERR_PTR(status);
 	return (client == NFS_CLIENT(dir)) ? rpc_clone_client(client) : client;
@@ -5010,16 +5064,13 @@ static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, str
 	struct nfs4_exception exception = {
 		.interruptible = true,
 	};
-	unsigned long now = jiffies;
 	int err;
 
 	do {
 		err = _nfs4_do_fsinfo(server, fhandle, fsinfo);
 		trace_nfs4_fsinfo(server, fhandle, fsinfo->fattr, err);
 		if (err == 0) {
-			nfs4_set_lease_period(server->nfs_client,
-					fsinfo->lease_time * HZ,
-					now);
+			nfs4_set_lease_period(server->nfs_client, fsinfo->lease_time * HZ);
 			break;
 		}
 		err = nfs4_handle_exception(server, err, &exception);
@@ -5098,12 +5149,12 @@ static bool nfs4_stateid_is_current(nfs4_stateid *stateid,
 		const struct nfs_lock_context *l_ctx,
 		fmode_t fmode)
 {
-	nfs4_stateid current_stateid;
+	nfs4_stateid _current_stateid;
 
 	/* If the current stateid represents a lost lock, then exit */
-	if (nfs4_set_rw_stateid(&current_stateid, ctx, l_ctx, fmode) == -EIO)
+	if (nfs4_set_rw_stateid(&_current_stateid, ctx, l_ctx, fmode) == -EIO)
 		return true;
-	return nfs4_stateid_match(stateid, &current_stateid);
+	return nfs4_stateid_match(stateid, &_current_stateid);
 }
 
 static bool nfs4_error_stateid_expired(int err)
@@ -5282,7 +5333,7 @@ static void nfs4_proc_write_setup(struct nfs_pgio_header *hdr,
 	hdr->timestamp   = jiffies;
 
 	msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE];
-	nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 1, 0);
+	nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 0, 0);
 	nfs4_state_protect_write(server->nfs_client, clnt, msg, hdr);
 }
 
@@ -5496,7 +5547,7 @@ unwind:
 struct nfs4_cached_acl {
 	int cached;
 	size_t len;
-	char data[0];
+	char data[];
 };
 
 static void nfs4_set_cached_acl(struct inode *inode, struct nfs4_cached_acl *acl)
@@ -5573,10 +5624,9 @@ out:
  */
 static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen)
 {
-	struct page *pages[NFS4ACL_MAXPAGES + 1] = {NULL, };
+	struct page **pages;
 	struct nfs_getaclargs args = {
 		.fh = NFS_FH(inode),
-		.acl_pages = pages,
 		.acl_len = buflen,
 	};
 	struct nfs_getaclres res = {
@@ -5587,11 +5637,19 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 		.rpc_argp = &args,
 		.rpc_resp = &res,
 	};
-	unsigned int npages = DIV_ROUND_UP(buflen, PAGE_SIZE) + 1;
+	unsigned int npages;
 	int ret = -ENOMEM, i;
+	struct nfs_server *server = NFS_SERVER(inode);
 
-	if (npages > ARRAY_SIZE(pages))
-		return -ERANGE;
+	if (buflen == 0)
+		buflen = server->rsize;
+
+	npages = DIV_ROUND_UP(buflen, PAGE_SIZE) + 1;
+	pages = kmalloc_array(npages, sizeof(struct page *), GFP_NOFS);
+	if (!pages)
+		return -ENOMEM;
+
+	args.acl_pages = pages;
 
 	for (i = 0; i < npages; i++) {
 		pages[i] = alloc_page(GFP_KERNEL);
@@ -5637,6 +5695,7 @@ out_free:
 			__free_page(pages[i]);
 	if (res.acl_scratch)
 		__free_page(res.acl_scratch);
+	kfree(pages);
 	return ret;
 }
 
@@ -6075,6 +6134,7 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
 		.callback_data = &setclientid,
 		.flags = RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN,
 	};
+	unsigned long now = jiffies;
 	int status;
 
 	/* nfs_client_id4 */
@@ -6107,6 +6167,9 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
 		clp->cl_acceptor = rpcauth_stringify_acceptor(setclientid.sc_cred);
 		put_rpccred(setclientid.sc_cred);
 	}
+
+	if (status == 0)
+		do_renew_lease(clp, now);
 out:
 	trace_nfs4_setclientid(clp, status);
 	dprintk("NFS reply setclientid: %d\n", status);
@@ -6193,13 +6256,17 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 		/* Fallthrough */
 	case -NFS4ERR_BAD_STATEID:
 	case -NFS4ERR_STALE_STATEID:
+	case -ETIMEDOUT:
 		task->tk_status = 0;
 		break;
 	case -NFS4ERR_OLD_STATEID:
-		if (nfs4_refresh_delegation_stateid(&data->stateid, data->inode))
-			goto out_restart;
-		task->tk_status = 0;
-		break;
+		if (!nfs4_refresh_delegation_stateid(&data->stateid, data->inode))
+			nfs4_stateid_seqid_inc(&data->stateid);
+		if (data->args.bitmask) {
+			data->args.bitmask = NULL;
+			data->res.fattr = NULL;
+		}
+		goto out_restart;
 	case -NFS4ERR_ACCESS:
 		if (data->args.bitmask) {
 			data->args.bitmask = NULL;
@@ -6214,6 +6281,7 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 		if (exception.retry)
 			goto out_restart;
 	}
+	nfs_delegation_mark_returned(data->inode, data->args.stateid);
 	data->rpc_status = task->tk_status;
 	return;
 out_restart:
@@ -6243,8 +6311,10 @@ static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data)
 
 	d_data = (struct nfs4_delegreturndata *)data;
 
-	if (!d_data->lr.roc && nfs4_wait_on_layoutreturn(d_data->inode, task))
+	if (!d_data->lr.roc && nfs4_wait_on_layoutreturn(d_data->inode, task)) {
+		nfs4_sequence_done(task, &d_data->res.seq_res);
 		return;
+	}
 
 	lo = d_data->args.lr_args ? d_data->args.lr_args->layout : NULL;
 	if (lo && !pnfs_layout_is_valid(lo)) {
@@ -6277,7 +6347,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred,
 		.rpc_client = server->client,
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_delegreturn_ops,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_TIMEOUT,
 	};
 	int status = 0;
 
@@ -6844,7 +6914,7 @@ static void nfs4_handle_setlk_error(struct nfs_server *server, struct nfs4_lock_
 	case -NFS4ERR_STALE_STATEID:
 		lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED;
 		nfs4_schedule_lease_recovery(server->nfs_client);
-	};
+	}
 }
 
 static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *fl, int recovery_type)
@@ -6860,7 +6930,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_lock_ops,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	int ret;
 
@@ -7820,6 +7890,22 @@ nfs41_same_server_scope(struct nfs41_server_scope *a,
 static void
 nfs4_bind_one_conn_to_session_done(struct rpc_task *task, void *calldata)
 {
+	struct nfs41_bind_conn_to_session_args *args = task->tk_msg.rpc_argp;
+	struct nfs41_bind_conn_to_session_res *res = task->tk_msg.rpc_resp;
+	struct nfs_client *clp = args->client;
+
+	switch (task->tk_status) {
+	case -NFS4ERR_BADSESSION:
+	case -NFS4ERR_DEADSESSION:
+		nfs4_schedule_session_recovery(clp->cl_session,
+				task->tk_status);
+	}
+	if (args->dir == NFS4_CDFC4_FORE_OR_BOTH &&
+			res->dir != NFS4_CDFS4_BOTH) {
+		rpc_task_close_connection(task);
+		if (args->retries++ < MAX_BIND_CONN_TO_SESSION_RETRIES)
+			rpc_restart_call(task);
+	}
 }
 
 static const struct rpc_call_ops nfs4_bind_one_conn_to_session_ops = {
@@ -7842,6 +7928,7 @@ int nfs4_proc_bind_one_conn_to_session(struct rpc_clnt *clnt,
 	struct nfs41_bind_conn_to_session_args args = {
 		.client = clp,
 		.dir = NFS4_CDFC4_FORE_OR_BOTH,
+		.retries = 0,
 	};
 	struct nfs41_bind_conn_to_session_res res;
 	struct rpc_message msg = {
@@ -8179,6 +8266,7 @@ static int _nfs4_proc_exchange_id(struct nfs_client *clp, const struct cred *cre
 	struct rpc_task *task;
 	struct nfs41_exchange_id_args *argp;
 	struct nfs41_exchange_id_res *resp;
+	unsigned long now = jiffies;
 	int status;
 
 	task = nfs4_run_exchange_id(clp, cred, sp4_how, NULL);
@@ -8199,6 +8287,8 @@ static int _nfs4_proc_exchange_id(struct nfs_client *clp, const struct cred *cre
 	if (status != 0)
 		goto out;
 
+	do_renew_lease(clp, now);
+
 	clp->cl_clientid = resp->clientid;
 	clp->cl_exchange_flags = resp->flags;
 	clp->cl_seqid = resp->seqid;
@@ -8602,7 +8692,7 @@ static int _nfs4_proc_create_session(struct nfs_client *clp,
 	case -EACCES:
 	case -EAGAIN:
 		goto out;
-	};
+	}
 
 	clp->cl_seqid++;
 	if (!status) {
@@ -8867,8 +8957,6 @@ static int nfs41_reclaim_complete_handle_errors(struct rpc_task *task, struct nf
 	case -NFS4ERR_BADSESSION:
 	case -NFS4ERR_DEADSESSION:
 	case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
-		nfs4_schedule_session_recovery(clp->cl_session,
-				task->tk_status);
 		break;
 	default:
 		nfs4_schedule_lease_recovery(clp);
@@ -9094,7 +9182,7 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout)
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_layoutget_call_ops,
 		.callback_data = lgp,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	struct pnfs_layout_segment *lseg = NULL;
 	struct nfs4_exception exception = {
@@ -9111,8 +9199,7 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout)
 	nfs4_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0, 0);
 
 	task = rpc_run_task(&task_setup_data);
-	if (IS_ERR(task))
-		return ERR_CAST(task);
+
 	status = rpc_wait_for_completion_task(task);
 	if (status != 0)
 		goto out;
@@ -9211,6 +9298,7 @@ static void nfs4_layoutreturn_release(void *calldata)
 		lrp->ld_private.ops->free(&lrp->ld_private);
 	pnfs_put_layout_hdr(lrp->args.layout);
 	nfs_iput_and_deactive(lrp->inode);
+	put_cred(lrp->cred);
 	kfree(calldata);
 	dprintk("<-- %s\n", __func__);
 }
@@ -9897,6 +9985,7 @@ static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = {
 		| NFS_CAP_ALLOCATE
 		| NFS_CAP_COPY
 		| NFS_CAP_OFFLOAD_CANCEL
+		| NFS_CAP_COPY_NOTIFY
 		| NFS_CAP_DEALLOCATE
 		| NFS_CAP_SEEK
 		| NFS_CAP_LAYOUTSTATS
@@ -9978,7 +10067,7 @@ const struct nfs_rpc_ops nfs_v4_clientops = {
 	.file_ops	= &nfs4_file_operations,
 	.getroot	= nfs4_proc_get_root,
 	.submount	= nfs4_submount,
-	.try_mount	= nfs4_try_mount,
+	.try_get_tree	= nfs4_try_get_tree,
 	.getattr	= nfs4_proc_getattr,
 	.setattr	= nfs4_proc_setattr,
 	.lookup		= nfs4_proc_lookup,
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index 6ea431b067dd..ff876dda7f06 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -138,15 +138,12 @@ nfs4_kill_renewd(struct nfs_client *clp)
  *
  * @clp: pointer to nfs_client
  * @lease: new value for lease period
- * @lastrenewed: time at which lease was last renewed
  */
 void nfs4_set_lease_period(struct nfs_client *clp,
-		unsigned long lease,
-		unsigned long lastrenewed)
+		unsigned long lease)
 {
 	spin_lock(&clp->cl_lock);
 	clp->cl_lease_time = lease;
-	clp->cl_last_renewal = lastrenewed;
 	spin_unlock(&clp->cl_lock);
 
 	/* Cap maximum reconnect timeout at 1/2 lease period */
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 0c6d53dc3672..a8dc25ce48bb 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -60,6 +60,7 @@
 #include "nfs4session.h"
 #include "pnfs.h"
 #include "netns.h"
+#include "nfs4trace.h"
 
 #define NFSDBG_FACILITY		NFSDBG_STATE
 
@@ -91,17 +92,15 @@ static int nfs4_setup_state_renewal(struct nfs_client *clp)
 {
 	int status;
 	struct nfs_fsinfo fsinfo;
-	unsigned long now;
 
 	if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) {
 		nfs4_schedule_state_renewal(clp);
 		return 0;
 	}
 
-	now = jiffies;
 	status = nfs4_proc_get_lease_time(clp, &fsinfo);
 	if (status == 0) {
-		nfs4_set_lease_period(clp, fsinfo.lease_time * HZ, now);
+		nfs4_set_lease_period(clp, fsinfo.lease_time * HZ);
 		nfs4_schedule_state_renewal(clp);
 	}
 
@@ -735,9 +734,9 @@ nfs4_get_open_state(struct inode *inode, struct nfs4_state_owner *owner)
 		state = new;
 		state->owner = owner;
 		atomic_inc(&owner->so_count);
-		list_add_rcu(&state->inode_states, &nfsi->open_states);
 		ihold(inode);
 		state->inode = inode;
+		list_add_rcu(&state->inode_states, &nfsi->open_states);
 		spin_unlock(&inode->i_lock);
 		/* Note: The reclaim code dictates that we add stateless
 		 * and read-only stateids to the end of the list */
@@ -765,6 +764,7 @@ void nfs4_put_open_state(struct nfs4_state *state)
 	list_del(&state->open_states);
 	spin_unlock(&inode->i_lock);
 	spin_unlock(&owner->so_lock);
+	nfs4_inode_return_delegation_on_close(inode);
 	iput(inode);
 	nfs4_free_open_state(state);
 	nfs4_put_state_owner(owner);
@@ -1134,7 +1134,7 @@ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid)
 		case -NFS4ERR_MOVED:
 			/* Non-seqid mutating errors */
 			return;
-	};
+	}
 	/*
 	 * Note: no locking needed as we are guaranteed to be first
 	 * on the sequence list
@@ -1407,7 +1407,7 @@ nfs_state_find_lock_state_by_stateid(struct nfs4_state *state,
 	list_for_each_entry(pos, &state->lock_states, ls_locks) {
 		if (!test_bit(NFS_LOCK_INITIALIZED, &pos->ls_flags))
 			continue;
-		if (nfs4_stateid_match_other(&pos->ls_stateid, stateid))
+		if (nfs4_stateid_match_or_older(&pos->ls_stateid, stateid))
 			return pos;
 	}
 	return NULL;
@@ -1441,12 +1441,13 @@ void nfs_inode_find_state_and_recover(struct inode *inode,
 		state = ctx->state;
 		if (state == NULL)
 			continue;
-		if (nfs4_stateid_match_other(&state->stateid, stateid) &&
+		if (nfs4_stateid_match_or_older(&state->stateid, stateid) &&
 		    nfs4_state_mark_reclaim_nograce(clp, state)) {
 			found = true;
 			continue;
 		}
-		if (nfs4_stateid_match_other(&state->open_stateid, stateid) &&
+		if (test_bit(NFS_OPEN_STATE, &state->flags) &&
+		    nfs4_stateid_match_or_older(&state->open_stateid, stateid) &&
 		    nfs4_state_mark_reclaim_nograce(clp, state)) {
 			found = true;
 			continue;
@@ -1556,16 +1557,32 @@ static void nfs42_complete_copies(struct nfs4_state_owner *sp, struct nfs4_state
 {
 	struct nfs4_copy_state *copy;
 
-	if (!test_bit(NFS_CLNT_DST_SSC_COPY_STATE, &state->flags))
+	if (!test_bit(NFS_CLNT_DST_SSC_COPY_STATE, &state->flags) &&
+		!test_bit(NFS_CLNT_SRC_SSC_COPY_STATE, &state->flags))
 		return;
 
 	spin_lock(&sp->so_server->nfs_client->cl_lock);
 	list_for_each_entry(copy, &sp->so_server->ss_copies, copies) {
-		if (!nfs4_stateid_match_other(&state->stateid, &copy->parent_state->stateid))
-			continue;
+		if ((test_bit(NFS_CLNT_DST_SSC_COPY_STATE, &state->flags) &&
+				!nfs4_stateid_match_other(&state->stateid,
+				&copy->parent_dst_state->stateid)))
+				continue;
 		copy->flags = 1;
-		complete(&copy->completion);
-		break;
+		if (test_and_clear_bit(NFS_CLNT_DST_SSC_COPY_STATE,
+				&state->flags)) {
+			clear_bit(NFS_CLNT_SRC_SSC_COPY_STATE, &state->flags);
+			complete(&copy->completion);
+		}
+	}
+	list_for_each_entry(copy, &sp->so_server->ss_copies, src_copies) {
+		if ((test_bit(NFS_CLNT_SRC_SSC_COPY_STATE, &state->flags) &&
+				!nfs4_stateid_match_other(&state->stateid,
+				&copy->parent_src_state->stateid)))
+				continue;
+		copy->flags = 1;
+		if (test_and_clear_bit(NFS_CLNT_DST_SSC_COPY_STATE,
+				&state->flags))
+			complete(&copy->completion);
 	}
 	spin_unlock(&sp->so_server->nfs_client->cl_lock);
 }
@@ -1593,6 +1610,7 @@ static int __nfs4_reclaim_open_state(struct nfs4_state_owner *sp, struct nfs4_st
 	if (!test_bit(NFS_DELEGATED_STATE, &state->flags)) {
 		spin_lock(&state->state_lock);
 		list_for_each_entry(lock, &state->lock_states, ls_locks) {
+			trace_nfs4_state_lock_reclaim(state, lock);
 			if (!test_bit(NFS_LOCK_INITIALIZED, &lock->ls_flags))
 				pr_warn_ratelimited("NFS: %s: Lock reclaim failed!\n", __func__);
 		}
@@ -1609,6 +1627,9 @@ static int nfs4_reclaim_open_state(struct nfs4_state_owner *sp, const struct nfs
 	struct nfs4_state *state;
 	unsigned int loop = 0;
 	int status = 0;
+#ifdef CONFIG_NFS_V4_2
+	bool found_ssc_copy_state = false;
+#endif /* CONFIG_NFS_V4_2 */
 
 	/* Note: we rely on the sp->so_states list being ordered 
 	 * so that we always reclaim open(O_RDWR) and/or open(O_WRITE)
@@ -1628,6 +1649,13 @@ restart:
 			continue;
 		if (state->state == 0)
 			continue;
+#ifdef CONFIG_NFS_V4_2
+		if (test_bit(NFS_SRV_SSC_COPY_STATE, &state->flags)) {
+			nfs4_state_mark_recovery_failed(state, -EIO);
+			found_ssc_copy_state = true;
+			continue;
+		}
+#endif /* CONFIG_NFS_V4_2 */
 		refcount_inc(&state->count);
 		spin_unlock(&sp->so_lock);
 		status = __nfs4_reclaim_open_state(sp, state, ops);
@@ -1682,6 +1710,10 @@ restart:
 	}
 	raw_write_seqcount_end(&sp->so_reclaim_seqcount);
 	spin_unlock(&sp->so_lock);
+#ifdef CONFIG_NFS_V4_2
+	if (found_ssc_copy_state)
+		return -EIO;
+#endif /* CONFIG_NFS_V4_2 */
 	return 0;
 out_err:
 	nfs4_put_open_state(state);
@@ -2492,6 +2524,21 @@ static int nfs4_bind_conn_to_session(struct nfs_client *clp)
 	}
 	return 0;
 }
+
+static void nfs4_layoutreturn_any_run(struct nfs_client *clp)
+{
+	int iomode = 0;
+
+	if (test_and_clear_bit(NFS4CLNT_RECALL_ANY_LAYOUT_READ, &clp->cl_state))
+		iomode += IOMODE_READ;
+	if (test_and_clear_bit(NFS4CLNT_RECALL_ANY_LAYOUT_RW, &clp->cl_state))
+		iomode += IOMODE_RW;
+	/* Note: IOMODE_READ + IOMODE_RW == IOMODE_ANY */
+	if (iomode) {
+		pnfs_layout_return_unused_byclid(clp, iomode);
+		set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state);
+	}
+}
 #else /* CONFIG_NFS_V4_1 */
 static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
 
@@ -2499,6 +2546,10 @@ static int nfs4_bind_conn_to_session(struct nfs_client *clp)
 {
 	return 0;
 }
+
+static void nfs4_layoutreturn_any_run(struct nfs_client *clp)
+{
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void nfs4_state_manager(struct nfs_client *clp)
@@ -2508,6 +2559,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
 
 	/* Ensure exclusive access to NFSv4 state */
 	do {
+		trace_nfs4_state_mgr(clp);
 		clear_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state);
 		if (test_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state)) {
 			section = "purge state";
@@ -2602,12 +2654,13 @@ static void nfs4_state_manager(struct nfs_client *clp)
 		nfs4_end_drain_session(clp);
 		nfs4_clear_state_manager_bit(clp);
 
-		if (!test_and_set_bit(NFS4CLNT_DELEGRETURN_RUNNING, &clp->cl_state)) {
+		if (!test_and_set_bit(NFS4CLNT_RECALL_RUNNING, &clp->cl_state)) {
 			if (test_and_clear_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state)) {
 				nfs_client_return_marked_delegations(clp);
 				set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state);
 			}
-			clear_bit(NFS4CLNT_DELEGRETURN_RUNNING, &clp->cl_state);
+			nfs4_layoutreturn_any_run(clp);
+			clear_bit(NFS4CLNT_RECALL_RUNNING, &clp->cl_state);
 		}
 
 		/* Did we race with an attempt to give us more work? */
@@ -2621,6 +2674,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
 out_error:
 	if (strlen(section))
 		section_sep = ": ";
+	trace_nfs4_state_mgr_failed(clp, section, status);
 	pr_warn_ratelimited("NFS: state manager%s%s failed on NFSv4 server %s"
 			" with error %d\n", section_sep, section,
 			clp->cl_hostname, -status);
diff --git a/fs/nfs/nfs4super.c b/fs/nfs/nfs4super.c
index 04c57066a11a..1475f932d7da 100644
--- a/fs/nfs/nfs4super.c
+++ b/fs/nfs/nfs4super.c
@@ -4,6 +4,7 @@
  */
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/mount.h>
 #include <linux/nfs4_mount.h>
 #include <linux/nfs_fs.h>
 #include "delegation.h"
@@ -18,36 +19,6 @@
 
 static int nfs4_write_inode(struct inode *inode, struct writeback_control *wbc);
 static void nfs4_evict_inode(struct inode *inode);
-static struct dentry *nfs4_remote_mount(struct file_system_type *fs_type,
-	int flags, const char *dev_name, void *raw_data);
-static struct dentry *nfs4_referral_mount(struct file_system_type *fs_type,
-	int flags, const char *dev_name, void *raw_data);
-static struct dentry *nfs4_remote_referral_mount(struct file_system_type *fs_type,
-	int flags, const char *dev_name, void *raw_data);
-
-static struct file_system_type nfs4_remote_fs_type = {
-	.owner		= THIS_MODULE,
-	.name		= "nfs4",
-	.mount		= nfs4_remote_mount,
-	.kill_sb	= nfs_kill_super,
-	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
-};
-
-static struct file_system_type nfs4_remote_referral_fs_type = {
-	.owner		= THIS_MODULE,
-	.name		= "nfs4",
-	.mount		= nfs4_remote_referral_mount,
-	.kill_sb	= nfs_kill_super,
-	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
-};
-
-struct file_system_type nfs4_referral_fs_type = {
-	.owner		= THIS_MODULE,
-	.name		= "nfs4",
-	.mount		= nfs4_referral_mount,
-	.kill_sb	= nfs_kill_super,
-	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
-};
 
 static const struct super_operations nfs4_sops = {
 	.alloc_inode	= nfs_alloc_inode,
@@ -61,16 +32,15 @@ static const struct super_operations nfs4_sops = {
 	.show_devname	= nfs_show_devname,
 	.show_path	= nfs_show_path,
 	.show_stats	= nfs_show_stats,
-	.remount_fs	= nfs_remount,
 };
 
 struct nfs_subversion nfs_v4 = {
-	.owner = THIS_MODULE,
-	.nfs_fs   = &nfs4_fs_type,
-	.rpc_vers = &nfs_version4,
-	.rpc_ops  = &nfs_v4_clientops,
-	.sops     = &nfs4_sops,
-	.xattr    = nfs4_xattr_handlers,
+	.owner		= THIS_MODULE,
+	.nfs_fs		= &nfs4_fs_type,
+	.rpc_vers	= &nfs_version4,
+	.rpc_ops	= &nfs_v4_clientops,
+	.sops		= &nfs4_sops,
+	.xattr		= nfs4_xattr_handlers,
 };
 
 static int nfs4_write_inode(struct inode *inode, struct writeback_control *wbc)
@@ -92,8 +62,8 @@ static void nfs4_evict_inode(struct inode *inode)
 {
 	truncate_inode_pages_final(&inode->i_data);
 	clear_inode(inode);
-	/* If we are holding a delegation, return it! */
-	nfs_inode_return_delegation_noreclaim(inode);
+	/* If we are holding a delegation, return and free it */
+	nfs_inode_evict_delegation(inode);
 	/* Note that above delegreturn would trigger pnfs return-on-close */
 	pnfs_return_layout(inode);
 	pnfs_destroy_layout(NFS_I(inode));
@@ -101,53 +71,6 @@ static void nfs4_evict_inode(struct inode *inode)
 	nfs_clear_inode(inode);
 }
 
-/*
- * Get the superblock for the NFS4 root partition
- */
-static struct dentry *
-nfs4_remote_mount(struct file_system_type *fs_type, int flags,
-		  const char *dev_name, void *info)
-{
-	struct nfs_mount_info *mount_info = info;
-	struct nfs_server *server;
-	struct dentry *mntroot = ERR_PTR(-ENOMEM);
-
-	mount_info->set_security = nfs_set_sb_security;
-
-	/* Get a volume representation */
-	server = nfs4_create_server(mount_info, &nfs_v4);
-	if (IS_ERR(server)) {
-		mntroot = ERR_CAST(server);
-		goto out;
-	}
-
-	mntroot = nfs_fs_mount_common(server, flags, dev_name, mount_info, &nfs_v4);
-
-out:
-	return mntroot;
-}
-
-static struct vfsmount *nfs_do_root_mount(struct file_system_type *fs_type,
-		int flags, void *data, const char *hostname)
-{
-	struct vfsmount *root_mnt;
-	char *root_devname;
-	size_t len;
-
-	len = strlen(hostname) + 5;
-	root_devname = kmalloc(len, GFP_KERNEL);
-	if (root_devname == NULL)
-		return ERR_PTR(-ENOMEM);
-	/* Does hostname needs to be enclosed in brackets? */
-	if (strchr(hostname, ':'))
-		snprintf(root_devname, len, "[%s]:/", hostname);
-	else
-		snprintf(root_devname, len, "%s:/", hostname);
-	root_mnt = vfs_kern_mount(fs_type, flags, root_devname, data);
-	kfree(root_devname);
-	return root_mnt;
-}
-
 struct nfs_referral_count {
 	struct list_head list;
 	const struct task_struct *task;
@@ -214,111 +137,125 @@ static void nfs_referral_loop_unprotect(void)
 	kfree(p);
 }
 
-static struct dentry *nfs_follow_remote_path(struct vfsmount *root_mnt,
-		const char *export_path)
+static int do_nfs4_mount(struct nfs_server *server,
+			 struct fs_context *fc,
+			 const char *hostname,
+			 const char *export_path)
 {
+	struct nfs_fs_context *root_ctx;
+	struct fs_context *root_fc;
+	struct vfsmount *root_mnt;
 	struct dentry *dentry;
-	int err;
+	size_t len;
+	int ret;
 
-	if (IS_ERR(root_mnt))
-		return ERR_CAST(root_mnt);
+	struct fs_parameter param = {
+		.key	= "source",
+		.type	= fs_value_is_string,
+		.dirfd	= -1,
+	};
 
-	err = nfs_referral_loop_protect();
-	if (err) {
-		mntput(root_mnt);
-		return ERR_PTR(err);
+	if (IS_ERR(server))
+		return PTR_ERR(server);
+
+	root_fc = vfs_dup_fs_context(fc);
+	if (IS_ERR(root_fc)) {
+		nfs_free_server(server);
+		return PTR_ERR(root_fc);
 	}
+	kfree(root_fc->source);
+	root_fc->source = NULL;
 
-	dentry = mount_subtree(root_mnt, export_path);
-	nfs_referral_loop_unprotect();
+	root_ctx = nfs_fc2context(root_fc);
+	root_ctx->internal = true;
+	root_ctx->server = server;
+	/* We leave export_path unset as it's not used to find the root. */
 
-	return dentry;
-}
+	len = strlen(hostname) + 5;
+	param.string = kmalloc(len, GFP_KERNEL);
+	if (param.string == NULL) {
+		put_fs_context(root_fc);
+		return -ENOMEM;
+	}
 
-struct dentry *nfs4_try_mount(int flags, const char *dev_name,
-			      struct nfs_mount_info *mount_info,
-			      struct nfs_subversion *nfs_mod)
-{
-	char *export_path;
-	struct vfsmount *root_mnt;
-	struct dentry *res;
-	struct nfs_parsed_mount_data *data = mount_info->parsed;
+	/* Does hostname needs to be enclosed in brackets? */
+	if (strchr(hostname, ':'))
+		param.size = snprintf(param.string, len, "[%s]:/", hostname);
+	else
+		param.size = snprintf(param.string, len, "%s:/", hostname);
+	ret = vfs_parse_fs_param(root_fc, &param);
+	kfree(param.string);
+	if (ret < 0) {
+		put_fs_context(root_fc);
+		return ret;
+	}
+	root_mnt = fc_mount(root_fc);
+	put_fs_context(root_fc);
+
+	if (IS_ERR(root_mnt))
+		return PTR_ERR(root_mnt);
 
-	dfprintk(MOUNT, "--> nfs4_try_mount()\n");
+	ret = nfs_referral_loop_protect();
+	if (ret) {
+		mntput(root_mnt);
+		return ret;
+	}
 
-	export_path = data->nfs_server.export_path;
-	data->nfs_server.export_path = "/";
-	root_mnt = nfs_do_root_mount(&nfs4_remote_fs_type, flags, mount_info,
-			data->nfs_server.hostname);
-	data->nfs_server.export_path = export_path;
+	dentry = mount_subtree(root_mnt, export_path);
+	nfs_referral_loop_unprotect();
 
-	res = nfs_follow_remote_path(root_mnt, export_path);
+	if (IS_ERR(dentry))
+		return PTR_ERR(dentry);
 
-	dfprintk(MOUNT, "<-- nfs4_try_mount() = %d%s\n",
-		 PTR_ERR_OR_ZERO(res),
-		 IS_ERR(res) ? " [error]" : "");
-	return res;
+	fc->root = dentry;
+	return 0;
 }
 
-static struct dentry *
-nfs4_remote_referral_mount(struct file_system_type *fs_type, int flags,
-			   const char *dev_name, void *raw_data)
+int nfs4_try_get_tree(struct fs_context *fc)
 {
-	struct nfs_mount_info mount_info = {
-		.fill_super = nfs_fill_super,
-		.set_security = nfs_clone_sb_security,
-		.cloned = raw_data,
-	};
-	struct nfs_server *server;
-	struct dentry *mntroot = ERR_PTR(-ENOMEM);
-
-	dprintk("--> nfs4_referral_get_sb()\n");
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	int err;
 
-	mount_info.mntfh = nfs_alloc_fhandle();
-	if (mount_info.cloned == NULL || mount_info.mntfh == NULL)
-		goto out;
+	dfprintk(MOUNT, "--> nfs4_try_get_tree()\n");
 
-	/* create a new volume representation */
-	server = nfs4_create_referral_server(mount_info.cloned, mount_info.mntfh);
-	if (IS_ERR(server)) {
-		mntroot = ERR_CAST(server);
-		goto out;
+	/* We create a mount for the server's root, walk to the requested
+	 * location and then create another mount for that.
+	 */
+	err= do_nfs4_mount(nfs4_create_server(fc),
+			   fc, ctx->nfs_server.hostname,
+			   ctx->nfs_server.export_path);
+	if (err) {
+		nfs_errorf(fc, "NFS4: Couldn't follow remote path");
+		dfprintk(MOUNT, "<-- nfs4_try_get_tree() = %d [error]\n", err);
+	} else {
+		dfprintk(MOUNT, "<-- nfs4_try_get_tree() = 0\n");
 	}
-
-	mntroot = nfs_fs_mount_common(server, flags, dev_name, &mount_info, &nfs_v4);
-out:
-	nfs_free_fhandle(mount_info.mntfh);
-	return mntroot;
+	return err;
 }
 
 /*
  * Create an NFS4 server record on referral traversal
  */
-static struct dentry *nfs4_referral_mount(struct file_system_type *fs_type,
-		int flags, const char *dev_name, void *raw_data)
+int nfs4_get_referral_tree(struct fs_context *fc)
 {
-	struct nfs_clone_mount *data = raw_data;
-	char *export_path;
-	struct vfsmount *root_mnt;
-	struct dentry *res;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	int err;
 
 	dprintk("--> nfs4_referral_mount()\n");
 
-	export_path = data->mnt_path;
-	data->mnt_path = "/";
-
-	root_mnt = nfs_do_root_mount(&nfs4_remote_referral_fs_type,
-			flags, data, data->hostname);
-	data->mnt_path = export_path;
-
-	res = nfs_follow_remote_path(root_mnt, export_path);
-	dprintk("<-- nfs4_referral_mount() = %d%s\n",
-		PTR_ERR_OR_ZERO(res),
-		IS_ERR(res) ? " [error]" : "");
-	return res;
+	/* create a new volume representation */
+	err = do_nfs4_mount(nfs4_create_referral_server(fc),
+			    fc, ctx->nfs_server.hostname,
+			    ctx->nfs_server.export_path);
+	if (err) {
+		nfs_errorf(fc, "NFS4: Couldn't follow remote path");
+		dfprintk(MOUNT, "<-- nfs4_get_referral_tree() = %d [error]\n", err);
+	} else {
+		dfprintk(MOUNT, "<-- nfs4_get_referral_tree() = 0\n");
+	}
+	return err;
 }
 
-
 static int __init init_nfs_v4(void)
 {
 	int err;
diff --git a/fs/nfs/nfs4trace.c b/fs/nfs/nfs4trace.c
index 1a8f376b3f73..d9ac556bebcf 100644
--- a/fs/nfs/nfs4trace.c
+++ b/fs/nfs/nfs4trace.c
@@ -24,4 +24,8 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_done);
 EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_done);
 EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_pagelist);
 EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_pagelist);
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_read_error);
+EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_write_error);
+EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_commit_error);
 #endif
diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h
index b2f395fa7350..543541173a3d 100644
--- a/fs/nfs/nfs4trace.h
+++ b/fs/nfs/nfs4trace.h
@@ -155,6 +155,9 @@ TRACE_DEFINE_ENUM(NFS4ERR_WRONG_CRED);
 TRACE_DEFINE_ENUM(NFS4ERR_WRONG_TYPE);
 TRACE_DEFINE_ENUM(NFS4ERR_XDEV);
 
+TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_MDS);
+TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_PNFS);
+
 #define show_nfsv4_errors(error) \
 	__print_symbolic(error, \
 		{ NFS4_OK, "OK" }, \
@@ -305,7 +308,10 @@ TRACE_DEFINE_ENUM(NFS4ERR_XDEV);
 		{ NFS4ERR_WRONGSEC, "WRONGSEC" }, \
 		{ NFS4ERR_WRONG_CRED, "WRONG_CRED" }, \
 		{ NFS4ERR_WRONG_TYPE, "WRONG_TYPE" }, \
-		{ NFS4ERR_XDEV, "XDEV" })
+		{ NFS4ERR_XDEV, "XDEV" }, \
+		/* ***** Internal to Linux NFS client ***** */ \
+		{ NFS4ERR_RESET_TO_MDS, "RESET_TO_MDS" }, \
+		{ NFS4ERR_RESET_TO_PNFS, "RESET_TO_PNFS" })
 
 #define show_open_flags(flags) \
 	__print_flags(flags, "|", \
@@ -352,7 +358,7 @@ DECLARE_EVENT_CLASS(nfs4_clientid_event,
 		),
 
 		TP_fast_assign(
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__assign_str(dstaddr, clp->cl_hostname);
 		),
 
@@ -432,7 +438,8 @@ TRACE_EVENT(nfs4_sequence_done,
 			__entry->target_highest_slotid =
 					res->sr_target_highest_slotid;
 			__entry->status_flags = res->sr_status_flags;
-			__entry->error = res->sr_status;
+			__entry->error = res->sr_status < 0 ?
+					-res->sr_status : 0;
 		),
 		TP_printk(
 			"error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u "
@@ -562,11 +569,108 @@ TRACE_EVENT(nfs4_setup_sequence,
 		)
 );
 
+TRACE_DEFINE_ENUM(NFS4CLNT_MANAGER_RUNNING);
+TRACE_DEFINE_ENUM(NFS4CLNT_CHECK_LEASE);
+TRACE_DEFINE_ENUM(NFS4CLNT_LEASE_EXPIRED);
+TRACE_DEFINE_ENUM(NFS4CLNT_RECLAIM_REBOOT);
+TRACE_DEFINE_ENUM(NFS4CLNT_RECLAIM_NOGRACE);
+TRACE_DEFINE_ENUM(NFS4CLNT_DELEGRETURN);
+TRACE_DEFINE_ENUM(NFS4CLNT_SESSION_RESET);
+TRACE_DEFINE_ENUM(NFS4CLNT_LEASE_CONFIRM);
+TRACE_DEFINE_ENUM(NFS4CLNT_SERVER_SCOPE_MISMATCH);
+TRACE_DEFINE_ENUM(NFS4CLNT_PURGE_STATE);
+TRACE_DEFINE_ENUM(NFS4CLNT_BIND_CONN_TO_SESSION);
+TRACE_DEFINE_ENUM(NFS4CLNT_MOVED);
+TRACE_DEFINE_ENUM(NFS4CLNT_LEASE_MOVED);
+TRACE_DEFINE_ENUM(NFS4CLNT_DELEGATION_EXPIRED);
+TRACE_DEFINE_ENUM(NFS4CLNT_RUN_MANAGER);
+TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_RUNNING);
+TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_ANY_LAYOUT_READ);
+TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_ANY_LAYOUT_RW);
+
+#define show_nfs4_clp_state(state) \
+	__print_flags(state, "|", \
+		{ NFS4CLNT_MANAGER_RUNNING,	"MANAGER_RUNNING" }, \
+		{ NFS4CLNT_CHECK_LEASE,		"CHECK_LEASE" }, \
+		{ NFS4CLNT_LEASE_EXPIRED,	"LEASE_EXPIRED" }, \
+		{ NFS4CLNT_RECLAIM_REBOOT,	"RECLAIM_REBOOT" }, \
+		{ NFS4CLNT_RECLAIM_NOGRACE,	"RECLAIM_NOGRACE" }, \
+		{ NFS4CLNT_DELEGRETURN,		"DELEGRETURN" }, \
+		{ NFS4CLNT_SESSION_RESET,	"SESSION_RESET" }, \
+		{ NFS4CLNT_LEASE_CONFIRM,	"LEASE_CONFIRM" }, \
+		{ NFS4CLNT_SERVER_SCOPE_MISMATCH, \
+						"SERVER_SCOPE_MISMATCH" }, \
+		{ NFS4CLNT_PURGE_STATE,		"PURGE_STATE" }, \
+		{ NFS4CLNT_BIND_CONN_TO_SESSION, \
+						"BIND_CONN_TO_SESSION" }, \
+		{ NFS4CLNT_MOVED,		"MOVED" }, \
+		{ NFS4CLNT_LEASE_MOVED,		"LEASE_MOVED" }, \
+		{ NFS4CLNT_DELEGATION_EXPIRED,	"DELEGATION_EXPIRED" }, \
+		{ NFS4CLNT_RUN_MANAGER,		"RUN_MANAGER" }, \
+		{ NFS4CLNT_RECALL_RUNNING,	"RECALL_RUNNING" }, \
+		{ NFS4CLNT_RECALL_ANY_LAYOUT_READ, "RECALL_ANY_LAYOUT_READ" }, \
+		{ NFS4CLNT_RECALL_ANY_LAYOUT_RW, "RECALL_ANY_LAYOUT_RW" })
+
+TRACE_EVENT(nfs4_state_mgr,
+		TP_PROTO(
+			const struct nfs_client *clp
+		),
+
+		TP_ARGS(clp),
+
+		TP_STRUCT__entry(
+			__field(unsigned long, state)
+			__string(hostname, clp->cl_hostname)
+		),
+
+		TP_fast_assign(
+			__entry->state = clp->cl_state;
+			__assign_str(hostname, clp->cl_hostname)
+		),
+
+		TP_printk(
+			"hostname=%s clp state=%s", __get_str(hostname),
+			show_nfs4_clp_state(__entry->state)
+		)
+)
+
+TRACE_EVENT(nfs4_state_mgr_failed,
+		TP_PROTO(
+			const struct nfs_client *clp,
+			const char *section,
+			int status
+		),
+
+		TP_ARGS(clp, section, status),
+
+		TP_STRUCT__entry(
+			__field(unsigned long, error)
+			__field(unsigned long, state)
+			__string(hostname, clp->cl_hostname)
+			__string(section, section)
+		),
+
+		TP_fast_assign(
+			__entry->error = status < 0 ? -status : 0;
+			__entry->state = clp->cl_state;
+			__assign_str(hostname, clp->cl_hostname);
+			__assign_str(section, section);
+		),
+
+		TP_printk(
+			"hostname=%s clp state=%s error=%ld (%s) section=%s",
+			__get_str(hostname),
+			show_nfs4_clp_state(__entry->state), -__entry->error,
+			show_nfsv4_errors(__entry->error), __get_str(section)
+
+		)
+)
+
 TRACE_EVENT(nfs4_xdr_status,
 		TP_PROTO(
 			const struct xdr_stream *xdr,
 			u32 op,
-			int error
+			u32 error
 		),
 
 		TP_ARGS(xdr, op, error),
@@ -598,6 +702,41 @@ TRACE_EVENT(nfs4_xdr_status,
 		)
 );
 
+DECLARE_EVENT_CLASS(nfs4_cb_error_class,
+		TP_PROTO(
+			__be32 xid,
+			u32 cb_ident
+		),
+
+		TP_ARGS(xid, cb_ident),
+
+		TP_STRUCT__entry(
+			__field(u32, xid)
+			__field(u32, cbident)
+		),
+
+		TP_fast_assign(
+			__entry->xid = be32_to_cpu(xid);
+			__entry->cbident = cb_ident;
+		),
+
+		TP_printk(
+			"xid=0x%08x cb_ident=0x%08x",
+			__entry->xid, __entry->cbident
+		)
+);
+
+#define DEFINE_CB_ERROR_EVENT(name) \
+	DEFINE_EVENT(nfs4_cb_error_class, nfs_cb_##name, \
+			TP_PROTO( \
+				__be32 xid, \
+				u32 cb_ident \
+			), \
+			TP_ARGS(xid, cb_ident))
+
+DEFINE_CB_ERROR_EVENT(no_clp);
+DEFINE_CB_ERROR_EVENT(badprinc);
+
 DECLARE_EVENT_CLASS(nfs4_open_event,
 		TP_PROTO(
 			const struct nfs_open_context *ctx,
@@ -756,7 +895,7 @@ TRACE_EVENT(nfs4_close,
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
 			__entry->fmode = (__force unsigned int)state->state;
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(args->stateid.seqid);
 			__entry->stateid_hash =
@@ -821,7 +960,7 @@ DECLARE_EVENT_CLASS(nfs4_lock_event,
 		TP_fast_assign(
 			const struct inode *inode = state->inode;
 
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->cmd = cmd;
 			__entry->type = request->fl_type;
 			__entry->start = request->fl_start;
@@ -893,7 +1032,7 @@ TRACE_EVENT(nfs4_set_lock,
 		TP_fast_assign(
 			const struct inode *inode = state->inode;
 
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->cmd = cmd;
 			__entry->type = request->fl_type;
 			__entry->start = request->fl_start;
@@ -929,6 +1068,88 @@ TRACE_EVENT(nfs4_set_lock,
 		)
 );
 
+TRACE_DEFINE_ENUM(LK_STATE_IN_USE);
+TRACE_DEFINE_ENUM(NFS_DELEGATED_STATE);
+TRACE_DEFINE_ENUM(NFS_OPEN_STATE);
+TRACE_DEFINE_ENUM(NFS_O_RDONLY_STATE);
+TRACE_DEFINE_ENUM(NFS_O_WRONLY_STATE);
+TRACE_DEFINE_ENUM(NFS_O_RDWR_STATE);
+TRACE_DEFINE_ENUM(NFS_STATE_RECLAIM_REBOOT);
+TRACE_DEFINE_ENUM(NFS_STATE_RECLAIM_NOGRACE);
+TRACE_DEFINE_ENUM(NFS_STATE_POSIX_LOCKS);
+TRACE_DEFINE_ENUM(NFS_STATE_RECOVERY_FAILED);
+TRACE_DEFINE_ENUM(NFS_STATE_MAY_NOTIFY_LOCK);
+TRACE_DEFINE_ENUM(NFS_STATE_CHANGE_WAIT);
+TRACE_DEFINE_ENUM(NFS_CLNT_DST_SSC_COPY_STATE);
+TRACE_DEFINE_ENUM(NFS_CLNT_SRC_SSC_COPY_STATE);
+TRACE_DEFINE_ENUM(NFS_SRV_SSC_COPY_STATE);
+
+#define show_nfs4_state_flags(flags) \
+	__print_flags(flags, "|", \
+		{ LK_STATE_IN_USE,		"IN_USE" }, \
+		{ NFS_DELEGATED_STATE,		"DELEGATED" }, \
+		{ NFS_OPEN_STATE,		"OPEN" }, \
+		{ NFS_O_RDONLY_STATE,		"O_RDONLY" }, \
+		{ NFS_O_WRONLY_STATE,		"O_WRONLY" }, \
+		{ NFS_O_RDWR_STATE,		"O_RDWR" }, \
+		{ NFS_STATE_RECLAIM_REBOOT,	"RECLAIM_REBOOT" }, \
+		{ NFS_STATE_RECLAIM_NOGRACE,	"RECLAIM_NOGRACE" }, \
+		{ NFS_STATE_POSIX_LOCKS,	"POSIX_LOCKS" }, \
+		{ NFS_STATE_RECOVERY_FAILED,	"RECOVERY_FAILED" }, \
+		{ NFS_STATE_MAY_NOTIFY_LOCK,	"MAY_NOTIFY_LOCK" }, \
+		{ NFS_STATE_CHANGE_WAIT,	"CHANGE_WAIT" }, \
+		{ NFS_CLNT_DST_SSC_COPY_STATE,	"CLNT_DST_SSC_COPY" }, \
+		{ NFS_CLNT_SRC_SSC_COPY_STATE,	"CLNT_SRC_SSC_COPY" }, \
+		{ NFS_SRV_SSC_COPY_STATE,	"SRV_SSC_COPY" })
+
+#define show_nfs4_lock_flags(flags) \
+	__print_flags(flags, "|", \
+		{ BIT(NFS_LOCK_INITIALIZED),	"INITIALIZED" }, \
+		{ BIT(NFS_LOCK_LOST),		"LOST" })
+
+TRACE_EVENT(nfs4_state_lock_reclaim,
+		TP_PROTO(
+			const struct nfs4_state *state,
+			const struct nfs4_lock_state *lock
+		),
+
+		TP_ARGS(state, lock),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(unsigned long, state_flags)
+			__field(unsigned long, lock_flags)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+		),
+
+		TP_fast_assign(
+			const struct inode *inode = state->inode;
+
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->state_flags = state->flags;
+			__entry->lock_flags = lock->ls_flags;
+			__entry->stateid_seq =
+				be32_to_cpu(state->stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&state->stateid);
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"stateid=%d:0x%08x state_flags=%s lock_flags=%s",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid, __entry->fhandle,
+			__entry->stateid_seq, __entry->stateid_hash,
+			show_nfs4_state_flags(__entry->state_flags),
+			show_nfs4_lock_flags(__entry->lock_flags)
+		)
+)
+
 DECLARE_EVENT_CLASS(nfs4_set_delegation_event,
 		TP_PROTO(
 			const struct inode *inode,
@@ -989,7 +1210,7 @@ TRACE_EVENT(nfs4_delegreturn_exit,
 		TP_fast_assign(
 			__entry->dev = res->server->s_dev;
 			__entry->fhandle = nfs_fhandle_hash(args->fhandle);
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(args->stateid->seqid);
 			__entry->stateid_hash =
@@ -1029,7 +1250,7 @@ DECLARE_EVENT_CLASS(nfs4_test_stateid_event,
 		TP_fast_assign(
 			const struct inode *inode = state->inode;
 
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
@@ -1131,7 +1352,7 @@ TRACE_EVENT(nfs4_lookupp,
 		TP_fast_assign(
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->ino = NFS_FILEID(inode);
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 		),
 
 		TP_printk(
@@ -1167,7 +1388,7 @@ TRACE_EVENT(nfs4_rename,
 			__entry->dev = olddir->i_sb->s_dev;
 			__entry->olddir = NFS_FILEID(olddir);
 			__entry->newdir = NFS_FILEID(newdir);
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__assign_str(oldname, oldname->name);
 			__assign_str(newname, newname->name);
 		),
@@ -1258,7 +1479,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_event,
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = NFS_FILEID(inode);
 			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(stateid->seqid);
 			__entry->stateid_hash =
@@ -1314,7 +1535,7 @@ DECLARE_EVENT_CLASS(nfs4_getattr_event,
 			__entry->valid = fattr->valid;
 			__entry->fhandle = nfs_fhandle_hash(fhandle);
 			__entry->fileid = (fattr->valid & NFS_ATTR_FATTR_FILEID) ? fattr->fileid : 0;
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 		),
 
 		TP_printk(
@@ -1361,7 +1582,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event,
 		),
 
 		TP_fast_assign(
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->fhandle = nfs_fhandle_hash(fhandle);
 			if (!IS_ERR_OR_NULL(inode)) {
 				__entry->fileid = NFS_FILEID(inode);
@@ -1418,7 +1639,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event,
 		),
 
 		TP_fast_assign(
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->fhandle = nfs_fhandle_hash(fhandle);
 			if (!IS_ERR_OR_NULL(inode)) {
 				__entry->fileid = NFS_FILEID(inode);
@@ -1519,7 +1740,8 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 			__field(u32, fhandle)
 			__field(u64, fileid)
 			__field(loff_t, offset)
-			__field(size_t, count)
+			__field(u32, arg_count)
+			__field(u32, res_count)
 			__field(unsigned long, error)
 			__field(int, stateid_seq)
 			__field(u32, stateid_hash)
@@ -1527,13 +1749,18 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 
 		TP_fast_assign(
 			const struct inode *inode = hdr->inode;
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
 			const struct nfs4_state *state =
 				hdr->args.context->state;
+
 			__entry->dev = inode->i_sb->s_dev;
-			__entry->fileid = NFS_FILEID(inode);
-			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
 			__entry->offset = hdr->args.offset;
-			__entry->count = hdr->args.count;
+			__entry->arg_count = hdr->args.count;
+			__entry->res_count = hdr->res.count;
 			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(state->stateid.seqid);
@@ -1543,14 +1770,14 @@ DECLARE_EVENT_CLASS(nfs4_read_event,
 
 		TP_printk(
 			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%zu stateid=%d:0x%08x",
+			"offset=%lld count=%u res=%u stateid=%d:0x%08x",
 			-__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
 			(long long)__entry->offset,
-			__entry->count,
+			__entry->arg_count, __entry->res_count,
 			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
@@ -1579,7 +1806,8 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 			__field(u32, fhandle)
 			__field(u64, fileid)
 			__field(loff_t, offset)
-			__field(size_t, count)
+			__field(u32, arg_count)
+			__field(u32, res_count)
 			__field(unsigned long, error)
 			__field(int, stateid_seq)
 			__field(u32, stateid_hash)
@@ -1587,13 +1815,18 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 
 		TP_fast_assign(
 			const struct inode *inode = hdr->inode;
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
 			const struct nfs4_state *state =
 				hdr->args.context->state;
+
 			__entry->dev = inode->i_sb->s_dev;
-			__entry->fileid = NFS_FILEID(inode);
-			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
 			__entry->offset = hdr->args.offset;
-			__entry->count = hdr->args.count;
+			__entry->arg_count = hdr->args.count;
+			__entry->res_count = hdr->res.count;
 			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(state->stateid.seqid);
@@ -1603,14 +1836,14 @@ DECLARE_EVENT_CLASS(nfs4_write_event,
 
 		TP_printk(
 			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%zu stateid=%d:0x%08x",
+			"offset=%lld count=%u res=%u stateid=%d:0x%08x",
 			-__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
 			(long long)__entry->offset,
-			__entry->count,
+			__entry->arg_count, __entry->res_count,
 			__entry->stateid_seq, __entry->stateid_hash
 		)
 );
@@ -1639,24 +1872,28 @@ DECLARE_EVENT_CLASS(nfs4_commit_event,
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
-			__field(loff_t, offset)
-			__field(size_t, count)
 			__field(unsigned long, error)
+			__field(loff_t, offset)
+			__field(u32, count)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = data->inode;
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = data->args.fh ?
+						  data->args.fh : &nfsi->fh;
+
 			__entry->dev = inode->i_sb->s_dev;
-			__entry->fileid = NFS_FILEID(inode);
-			__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
 			__entry->offset = data->args.offset;
 			__entry->count = data->args.count;
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 		),
 
 		TP_printk(
 			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%zu",
+			"offset=%lld count=%u",
 			-__entry->error,
 			show_nfsv4_errors(__entry->error),
 			MAJOR(__entry->dev), MINOR(__entry->dev),
@@ -1721,7 +1958,7 @@ TRACE_EVENT(nfs4_layoutget,
 			__entry->iomode = args->iomode;
 			__entry->offset = args->offset;
 			__entry->count = args->length;
-			__entry->error = error;
+			__entry->error = error < 0 ? -error : 0;
 			__entry->stateid_seq =
 				be32_to_cpu(state->stateid.seqid);
 			__entry->stateid_hash =
@@ -1919,6 +2156,115 @@ DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_done);
 DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_pagelist);
 DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_pagelist);
 
+DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event,
+		TP_PROTO(
+			const struct nfs_pgio_header *hdr
+		),
+
+		TP_ARGS(hdr),
+
+		TP_STRUCT__entry(
+			__field(unsigned long, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
+			__field(int, stateid_seq)
+			__field(u32, stateid_hash)
+			__string(dstaddr, hdr->ds_clp ?
+				rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown")
+		),
+
+		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
+
+			__entry->error = hdr->res.op_status;
+			__entry->fhandle = nfs_fhandle_hash(hdr->args.fh);
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->offset = hdr->args.offset;
+			__entry->count = hdr->args.count;
+			__entry->stateid_seq =
+				be32_to_cpu(hdr->args.stateid.seqid);
+			__entry->stateid_hash =
+				nfs_stateid_hash(&hdr->args.stateid);
+			__assign_str(dstaddr, hdr->ds_clp ?
+				rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown");
+		),
+
+		TP_printk(
+			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s",
+			-__entry->error,
+			show_nfsv4_errors(__entry->error),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->offset, __entry->count,
+			__entry->stateid_seq, __entry->stateid_hash,
+			__get_str(dstaddr)
+		)
+);
+
+#define DEFINE_NFS4_FLEXFILES_IO_EVENT(name) \
+	DEFINE_EVENT(nfs4_flexfiles_io_event, name, \
+			TP_PROTO( \
+				const struct nfs_pgio_header *hdr \
+			), \
+			TP_ARGS(hdr))
+DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_read_error);
+DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_write_error);
+
+TRACE_EVENT(ff_layout_commit_error,
+		TP_PROTO(
+			const struct nfs_commit_data *data
+		),
+
+		TP_ARGS(data),
+
+		TP_STRUCT__entry(
+			__field(unsigned long, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
+			__string(dstaddr, data->ds_clp ?
+				rpc_peeraddr2str(data->ds_clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown")
+		),
+
+		TP_fast_assign(
+			const struct inode *inode = data->inode;
+
+			__entry->error = data->res.op_status;
+			__entry->fhandle = nfs_fhandle_hash(data->args.fh);
+			__entry->fileid = NFS_FILEID(inode);
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->offset = data->args.offset;
+			__entry->count = data->args.count;
+			__assign_str(dstaddr, data->ds_clp ?
+				rpc_peeraddr2str(data->ds_clp->cl_rpcclient,
+					RPC_DISPLAY_ADDR) : "unknown");
+		),
+
+		TP_printk(
+			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"offset=%llu count=%u dstaddr=%s",
+			-__entry->error,
+			show_nfsv4_errors(__entry->error),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->offset, __entry->count,
+			__get_str(dstaddr)
+		)
+);
+
+
 #endif /* CONFIG_NFS_V4_1 */
 
 #endif /* _TRACE_NFS4_H */
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index ab07db0f07cd..47817ef0aadb 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1059,9 +1059,9 @@ static void encode_nfs4_verifier(struct xdr_stream *xdr, const nfs4_verifier *ve
 }
 
 static __be32 *
-xdr_encode_nfstime4(__be32 *p, const struct timespec *t)
+xdr_encode_nfstime4(__be32 *p, const struct timespec64 *t)
 {
-	p = xdr_encode_hyper(p, (__s64)t->tv_sec);
+	p = xdr_encode_hyper(p, t->tv_sec);
 	*p++ = cpu_to_be32(t->tv_nsec);
 	return p;
 }
@@ -1072,7 +1072,6 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap,
 				const struct nfs_server *server,
 				const uint32_t attrmask[])
 {
-	struct timespec ts;
 	char owner_name[IDMAP_NAMESZ];
 	char owner_group[IDMAP_NAMESZ];
 	int owner_namelen = 0;
@@ -1161,16 +1160,14 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap,
 	if (bmval[1] & FATTR4_WORD1_TIME_ACCESS_SET) {
 		if (iap->ia_valid & ATTR_ATIME_SET) {
 			*p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME);
-			ts = timespec64_to_timespec(iap->ia_atime);
-			p = xdr_encode_nfstime4(p, &ts);
+			p = xdr_encode_nfstime4(p, &iap->ia_atime);
 		} else
 			*p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME);
 	}
 	if (bmval[1] & FATTR4_WORD1_TIME_MODIFY_SET) {
 		if (iap->ia_valid & ATTR_MTIME_SET) {
 			*p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME);
-			ts = timespec64_to_timespec(iap->ia_mtime);
-			p = xdr_encode_nfstime4(p, &ts);
+			p = xdr_encode_nfstime4(p, &iap->ia_mtime);
 		} else
 			*p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME);
 	}
@@ -4065,17 +4062,17 @@ static int decode_attr_space_used(struct xdr_stream *xdr, uint32_t *bitmap, uint
 }
 
 static __be32 *
-xdr_decode_nfstime4(__be32 *p, struct timespec *t)
+xdr_decode_nfstime4(__be32 *p, struct timespec64 *t)
 {
 	__u64 sec;
 
 	p = xdr_decode_hyper(p, &sec);
-	t-> tv_sec = (time_t)sec;
+	t-> tv_sec = sec;
 	t->tv_nsec = be32_to_cpup(p++);
 	return p;
 }
 
-static int decode_attr_time(struct xdr_stream *xdr, struct timespec *time)
+static int decode_attr_time(struct xdr_stream *xdr, struct timespec64 *time)
 {
 	__be32 *p;
 
@@ -4086,7 +4083,7 @@ static int decode_attr_time(struct xdr_stream *xdr, struct timespec *time)
 	return 0;
 }
 
-static int decode_attr_time_access(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec *time)
+static int decode_attr_time_access(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec64 *time)
 {
 	int status = 0;
 
@@ -4100,11 +4097,11 @@ static int decode_attr_time_access(struct xdr_stream *xdr, uint32_t *bitmap, str
 			status = NFS_ATTR_FATTR_ATIME;
 		bitmap[1] &= ~FATTR4_WORD1_TIME_ACCESS;
 	}
-	dprintk("%s: atime=%ld\n", __func__, (long)time->tv_sec);
+	dprintk("%s: atime=%lld\n", __func__, time->tv_sec);
 	return status;
 }
 
-static int decode_attr_time_metadata(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec *time)
+static int decode_attr_time_metadata(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec64 *time)
 {
 	int status = 0;
 
@@ -4118,12 +4115,12 @@ static int decode_attr_time_metadata(struct xdr_stream *xdr, uint32_t *bitmap, s
 			status = NFS_ATTR_FATTR_CTIME;
 		bitmap[1] &= ~FATTR4_WORD1_TIME_METADATA;
 	}
-	dprintk("%s: ctime=%ld\n", __func__, (long)time->tv_sec);
+	dprintk("%s: ctime=%lld\n", __func__, time->tv_sec);
 	return status;
 }
 
 static int decode_attr_time_delta(struct xdr_stream *xdr, uint32_t *bitmap,
-				  struct timespec *time)
+				  struct timespec64 *time)
 {
 	int status = 0;
 
@@ -4135,8 +4132,8 @@ static int decode_attr_time_delta(struct xdr_stream *xdr, uint32_t *bitmap,
 		status = decode_attr_time(xdr, time);
 		bitmap[1] &= ~FATTR4_WORD1_TIME_DELTA;
 	}
-	dprintk("%s: time_delta=%ld %ld\n", __func__, (long)time->tv_sec,
-		(long)time->tv_nsec);
+	dprintk("%s: time_delta=%lld %ld\n", __func__, time->tv_sec,
+		time->tv_nsec);
 	return status;
 }
 
@@ -4186,7 +4183,7 @@ static int decode_attr_security_label(struct xdr_stream *xdr, uint32_t *bitmap,
 	return status;
 }
 
-static int decode_attr_time_modify(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec *time)
+static int decode_attr_time_modify(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec64 *time)
 {
 	int status = 0;
 
@@ -4200,7 +4197,7 @@ static int decode_attr_time_modify(struct xdr_stream *xdr, uint32_t *bitmap, str
 			status = NFS_ATTR_FATTR_MTIME;
 		bitmap[1] &= ~FATTR4_WORD1_TIME_MODIFY;
 	}
-	dprintk("%s: mtime=%ld\n", __func__, (long)time->tv_sec);
+	dprintk("%s: mtime=%lld\n", __func__, time->tv_sec);
 	return status;
 }
 
@@ -4316,11 +4313,14 @@ static int decode_write_verifier(struct xdr_stream *xdr, struct nfs_write_verifi
 
 static int decode_commit(struct xdr_stream *xdr, struct nfs_commitres *res)
 {
+	struct nfs_writeverf *verf = res->verf;
 	int status;
 
 	status = decode_op_hdr(xdr, OP_COMMIT);
 	if (!status)
-		status = decode_write_verifier(xdr, &res->verf->verifier);
+		status = decode_write_verifier(xdr, &verf->verifier);
+	if (!status)
+		verf->committed = NFS_FILE_SYNC;
 	return status;
 }
 
@@ -7581,6 +7581,7 @@ const struct rpc_procinfo nfs4_procedures[] = {
 	PROC42(CLONE,		enc_clone,		dec_clone),
 	PROC42(COPY,		enc_copy,		dec_copy),
 	PROC42(OFFLOAD_CANCEL,	enc_offload_cancel,	dec_offload_cancel),
+	PROC42(COPY_NOTIFY,	enc_copy_notify,	dec_copy_notify),
 	PROC(LOOKUPP,		enc_lookupp,		dec_lookupp),
 	PROC42(LAYOUTERROR,	enc_layouterror,	dec_layouterror),
 };
diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index effaa4247b91..8d3278805602 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -88,7 +88,7 @@
 #define NFS_ROOT		"/tftpboot/%s"
 
 /* Default NFSROOT mount options. */
-#define NFS_DEF_OPTIONS		"vers=2,udp,rsize=4096,wsize=4096"
+#define NFS_DEF_OPTIONS		"vers=2,tcp,rsize=4096,wsize=4096"
 
 /* Parameters passed from the kernel command line */
 static char nfs_root_parms[NFS_MAXPATHLEN + 1] __initdata = "";
diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index 361cc10d6f95..7e7a97ae21ed 100644
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -181,6 +181,7 @@ DECLARE_EVENT_CLASS(nfs_inode_event_done,
 				int error \
 			), \
 			TP_ARGS(inode, error))
+DEFINE_NFS_INODE_EVENT(nfs_set_inode_stale);
 DEFINE_NFS_INODE_EVENT(nfs_refresh_inode_enter);
 DEFINE_NFS_INODE_EVENT_DONE(nfs_refresh_inode_exit);
 DEFINE_NFS_INODE_EVENT(nfs_revalidate_inode_enter);
@@ -198,7 +199,66 @@ DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_inode_exit);
 DEFINE_NFS_INODE_EVENT(nfs_fsync_enter);
 DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit);
 DEFINE_NFS_INODE_EVENT(nfs_access_enter);
-DEFINE_NFS_INODE_EVENT_DONE(nfs_access_exit);
+
+TRACE_EVENT(nfs_access_exit,
+		TP_PROTO(
+			const struct inode *inode,
+			unsigned int mask,
+			unsigned int permitted,
+			int error
+		),
+
+		TP_ARGS(inode, mask, permitted, error),
+
+		TP_STRUCT__entry(
+			__field(unsigned long, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(unsigned char, type)
+			__field(u64, fileid)
+			__field(u64, version)
+			__field(loff_t, size)
+			__field(unsigned long, nfsi_flags)
+			__field(unsigned long, cache_validity)
+			__field(unsigned int, mask)
+			__field(unsigned int, permitted)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			__entry->error = error < 0 ? -error : 0;
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->type = nfs_umode_to_dtype(inode->i_mode);
+			__entry->version = inode_peek_iversion_raw(inode);
+			__entry->size = i_size_read(inode);
+			__entry->nfsi_flags = nfsi->flags;
+			__entry->cache_validity = nfsi->cache_validity;
+			__entry->mask = mask;
+			__entry->permitted = permitted;
+		),
+
+		TP_printk(
+			"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"type=%u (%s) version=%llu size=%lld "
+			"cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s) "
+			"mask=0x%x permitted=0x%x",
+			-__entry->error, nfs_show_status(__entry->error),
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			__entry->type,
+			nfs_show_file_type(__entry->type),
+			(unsigned long long)__entry->version,
+			(long long)__entry->size,
+			__entry->cache_validity,
+			nfs_show_cache_validity(__entry->cache_validity),
+			__entry->nfsi_flags,
+			nfs_show_nfsi_flags(__entry->nfsi_flags),
+			__entry->mask, __entry->permitted
+		)
+);
 
 TRACE_DEFINE_ENUM(LOOKUP_FOLLOW);
 TRACE_DEFINE_ENUM(LOOKUP_DIRECTORY);
@@ -206,7 +266,6 @@ TRACE_DEFINE_ENUM(LOOKUP_AUTOMOUNT);
 TRACE_DEFINE_ENUM(LOOKUP_PARENT);
 TRACE_DEFINE_ENUM(LOOKUP_REVAL);
 TRACE_DEFINE_ENUM(LOOKUP_RCU);
-TRACE_DEFINE_ENUM(LOOKUP_NO_REVAL);
 TRACE_DEFINE_ENUM(LOOKUP_OPEN);
 TRACE_DEFINE_ENUM(LOOKUP_CREATE);
 TRACE_DEFINE_ENUM(LOOKUP_EXCL);
@@ -224,7 +283,6 @@ TRACE_DEFINE_ENUM(LOOKUP_DOWN);
 			{ LOOKUP_PARENT, "PARENT" }, \
 			{ LOOKUP_REVAL, "REVAL" }, \
 			{ LOOKUP_RCU, "RCU" }, \
-			{ LOOKUP_NO_REVAL, "NO_REVAL" }, \
 			{ LOOKUP_OPEN, "OPEN" }, \
 			{ LOOKUP_CREATE, "CREATE" }, \
 			{ LOOKUP_EXCL, "EXCL" }, \
@@ -820,75 +878,85 @@ TRACE_EVENT(nfs_sillyrename_unlink,
 
 TRACE_EVENT(nfs_initiate_read,
 		TP_PROTO(
-			const struct inode *inode,
-			loff_t offset, unsigned long count
+			const struct nfs_pgio_header *hdr
 		),
 
-		TP_ARGS(inode, offset, count),
+		TP_ARGS(hdr),
 
 		TP_STRUCT__entry(
-			__field(loff_t, offset)
-			__field(unsigned long, count)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
 		),
 
 		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
 
-			__entry->offset = offset;
-			__entry->count = count;
+			__entry->offset = hdr->args.offset;
+			__entry->count = hdr->args.count;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%lu",
+			"offset=%lld count=%u",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->count
+			(long long)__entry->offset, __entry->count
 		)
 );
 
 TRACE_EVENT(nfs_readpage_done,
 		TP_PROTO(
-			const struct inode *inode,
-			int status, loff_t offset, bool eof
+			const struct rpc_task *task,
+			const struct nfs_pgio_header *hdr
 		),
 
-		TP_ARGS(inode, status, offset, eof),
+		TP_ARGS(task, hdr),
 
 		TP_STRUCT__entry(
-			__field(int, status)
-			__field(loff_t, offset)
-			__field(bool, eof)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, arg_count)
+			__field(u32, res_count)
+			__field(bool, eof)
+			__field(int, status)
 		),
 
 		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
-
-			__entry->status = status;
-			__entry->offset = offset;
-			__entry->eof = eof;
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
+
+			__entry->status = task->tk_status;
+			__entry->offset = hdr->args.offset;
+			__entry->arg_count = hdr->args.count;
+			__entry->res_count = hdr->res.count;
+			__entry->eof = hdr->res.eof;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld status=%d%s",
+			"offset=%lld count=%u res=%u status=%d%s",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->status,
+			(long long)__entry->offset, __entry->arg_count,
+			__entry->res_count, __entry->status,
 			__entry->eof ? " eof" : ""
 		)
 );
@@ -905,90 +973,144 @@ TRACE_DEFINE_ENUM(NFS_FILE_SYNC);
 
 TRACE_EVENT(nfs_initiate_write,
 		TP_PROTO(
-			const struct inode *inode,
-			loff_t offset, unsigned long count,
-			enum nfs3_stable_how stable
+			const struct nfs_pgio_header *hdr
 		),
 
-		TP_ARGS(inode, offset, count, stable),
+		TP_ARGS(hdr),
 
 		TP_STRUCT__entry(
-			__field(loff_t, offset)
-			__field(unsigned long, count)
-			__field(enum nfs3_stable_how, stable)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
+			__field(enum nfs3_stable_how, stable)
 		),
 
 		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
 
-			__entry->offset = offset;
-			__entry->count = count;
-			__entry->stable = stable;
+			__entry->offset = hdr->args.offset;
+			__entry->count = hdr->args.count;
+			__entry->stable = hdr->args.stable;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%lu stable=%s",
+			"offset=%lld count=%u stable=%s",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->count,
+			(long long)__entry->offset, __entry->count,
 			nfs_show_stable(__entry->stable)
 		)
 );
 
 TRACE_EVENT(nfs_writeback_done,
 		TP_PROTO(
-			const struct inode *inode,
-			int status,
-			loff_t offset,
-			struct nfs_writeverf *writeverf
+			const struct rpc_task *task,
+			const struct nfs_pgio_header *hdr
 		),
 
-		TP_ARGS(inode, status, offset, writeverf),
+		TP_ARGS(task, hdr),
 
 		TP_STRUCT__entry(
-			__field(int, status)
-			__field(loff_t, offset)
-			__field(enum nfs3_stable_how, stable)
-			__field(unsigned long long, verifier)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, arg_count)
+			__field(u32, res_count)
+			__field(int, status)
+			__field(enum nfs3_stable_how, stable)
+			__array(char, verifier, NFS4_VERIFIER_SIZE)
 		),
 
 		TP_fast_assign(
+			const struct inode *inode = hdr->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
-
-			__entry->status = status;
-			__entry->offset = offset;
-			__entry->stable = writeverf->committed;
-			memcpy(&__entry->verifier, &writeverf->verifier,
-			       sizeof(__entry->verifier));
+			const struct nfs_fh *fh = hdr->args.fh ?
+						  hdr->args.fh : &nfsi->fh;
+			const struct nfs_writeverf *verf = hdr->res.verf;
+
+			__entry->status = task->tk_status;
+			__entry->offset = hdr->args.offset;
+			__entry->arg_count = hdr->args.count;
+			__entry->res_count = hdr->res.count;
+			__entry->stable = verf->committed;
+			memcpy(__entry->verifier,
+				&verf->verifier,
+				NFS4_VERIFIER_SIZE);
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld status=%d stable=%s "
-			"verifier 0x%016llx",
+			"offset=%lld count=%u res=%u status=%d stable=%s "
+			"verifier=%s",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->status,
+			(long long)__entry->offset, __entry->arg_count,
+			__entry->res_count, __entry->status,
 			nfs_show_stable(__entry->stable),
-			__entry->verifier
+			__print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE)
+		)
+);
+
+DECLARE_EVENT_CLASS(nfs_page_error_class,
+		TP_PROTO(
+			const struct nfs_page *req,
+			int error
+		),
+
+		TP_ARGS(req, error),
+
+		TP_STRUCT__entry(
+			__field(const void *, req)
+			__field(pgoff_t, index)
+			__field(unsigned int, offset)
+			__field(unsigned int, pgbase)
+			__field(unsigned int, bytes)
+			__field(int, error)
+		),
+
+		TP_fast_assign(
+			__entry->req = req;
+			__entry->index = req->wb_index;
+			__entry->offset = req->wb_offset;
+			__entry->pgbase = req->wb_pgbase;
+			__entry->bytes = req->wb_bytes;
+			__entry->error = error;
+		),
+
+		TP_printk(
+			"req=%p index=%lu offset=%u pgbase=%u bytes=%u error=%d",
+			__entry->req, __entry->index, __entry->offset,
+			__entry->pgbase, __entry->bytes, __entry->error
 		)
 );
 
+#define DEFINE_NFS_PAGEERR_EVENT(name) \
+	DEFINE_EVENT(nfs_page_error_class, name, \
+			TP_PROTO( \
+				const struct nfs_page *req, \
+				int error \
+			), \
+			TP_ARGS(req, error))
+
+DEFINE_NFS_PAGEERR_EVENT(nfs_write_error);
+DEFINE_NFS_PAGEERR_EVENT(nfs_comp_error);
+DEFINE_NFS_PAGEERR_EVENT(nfs_commit_error);
+
 TRACE_EVENT(nfs_initiate_commit,
 		TP_PROTO(
 			const struct nfs_commit_data *data
@@ -997,71 +1119,114 @@ TRACE_EVENT(nfs_initiate_commit,
 		TP_ARGS(data),
 
 		TP_STRUCT__entry(
-			__field(loff_t, offset)
-			__field(unsigned long, count)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(u32, count)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = data->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = data->args.fh ?
+						  data->args.fh : &nfsi->fh;
 
 			__entry->offset = data->args.offset;
 			__entry->count = data->args.count;
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld count=%lu",
+			"offset=%lld count=%u",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->count
+			(long long)__entry->offset, __entry->count
 		)
 );
 
 TRACE_EVENT(nfs_commit_done,
 		TP_PROTO(
+			const struct rpc_task *task,
 			const struct nfs_commit_data *data
 		),
 
-		TP_ARGS(data),
+		TP_ARGS(task, data),
 
 		TP_STRUCT__entry(
-			__field(int, status)
-			__field(loff_t, offset)
-			__field(unsigned long long, verifier)
 			__field(dev_t, dev)
 			__field(u32, fhandle)
 			__field(u64, fileid)
+			__field(loff_t, offset)
+			__field(int, status)
+			__field(enum nfs3_stable_how, stable)
+			__array(char, verifier, NFS4_VERIFIER_SIZE)
 		),
 
 		TP_fast_assign(
 			const struct inode *inode = data->inode;
 			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = data->args.fh ?
+						  data->args.fh : &nfsi->fh;
+			const struct nfs_writeverf *verf = data->res.verf;
 
-			__entry->status = data->res.op_status;
+			__entry->status = task->tk_status;
 			__entry->offset = data->args.offset;
-			memcpy(&__entry->verifier, &data->verf.verifier,
-			       sizeof(__entry->verifier));
+			__entry->stable = verf->committed;
+			memcpy(__entry->verifier,
+				&verf->verifier,
+				NFS4_VERIFIER_SIZE);
 			__entry->dev = inode->i_sb->s_dev;
 			__entry->fileid = nfsi->fileid;
-			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fhandle = nfs_fhandle_hash(fh);
 		),
 
 		TP_printk(
 			"fileid=%02x:%02x:%llu fhandle=0x%08x "
-			"offset=%lld status=%d verifier 0x%016llx",
+			"offset=%lld status=%d stable=%s verifier=%s",
 			MAJOR(__entry->dev), MINOR(__entry->dev),
 			(unsigned long long)__entry->fileid,
 			__entry->fhandle,
-			__entry->offset, __entry->status,
-			__entry->verifier
+			(long long)__entry->offset, __entry->status,
+			nfs_show_stable(__entry->stable),
+			__print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE)
+		)
+);
+
+TRACE_EVENT(nfs_fh_to_dentry,
+		TP_PROTO(
+			const struct super_block *sb,
+			const struct nfs_fh *fh,
+			u64 fileid,
+			int error
+		),
+
+		TP_ARGS(sb, fh, fileid, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+		),
+
+		TP_fast_assign(
+			__entry->error = error;
+			__entry->dev = sb->s_dev;
+			__entry->fileid = fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
+		),
+
+		TP_printk(
+			"error=%d fileid=%02x:%02x:%llu fhandle=0x%08x ",
+			__entry->error,
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle
 		)
 );
 
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 20b3717cd7ca..6ca421cbe19c 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -33,9 +33,7 @@ static const struct rpc_call_ops nfs_pgio_common_ops;
 struct nfs_pgio_mirror *
 nfs_pgio_current_mirror(struct nfs_pageio_descriptor *desc)
 {
-	return nfs_pgio_has_mirroring(desc) ?
-		&desc->pg_mirrors[desc->pg_mirror_idx] :
-		&desc->pg_mirrors[0];
+	return &desc->pg_mirrors[desc->pg_mirror_idx];
 }
 EXPORT_SYMBOL_GPL(nfs_pgio_current_mirror);
 
@@ -133,47 +131,166 @@ nfs_async_iocounter_wait(struct rpc_task *task, struct nfs_lock_context *l_ctx)
 EXPORT_SYMBOL_GPL(nfs_async_iocounter_wait);
 
 /*
- * nfs_page_group_lock - lock the head of the page group
- * @req - request in group that is to be locked
+ * nfs_page_lock_head_request - page lock the head of the page group
+ * @req: any member of the page group
+ */
+struct nfs_page *
+nfs_page_group_lock_head(struct nfs_page *req)
+{
+	struct nfs_page *head = req->wb_head;
+
+	while (!nfs_lock_request(head)) {
+		int ret = nfs_wait_on_request(head);
+		if (ret < 0)
+			return ERR_PTR(ret);
+	}
+	if (head != req)
+		kref_get(&head->wb_kref);
+	return head;
+}
+
+/*
+ * nfs_unroll_locks -  unlock all newly locked reqs and wait on @req
+ * @head: head request of page group, must be holding head lock
+ * @req: request that couldn't lock and needs to wait on the req bit lock
  *
- * this lock must be held when traversing or modifying the page
- * group list
+ * This is a helper function for nfs_lock_and_join_requests
+ * returns 0 on success, < 0 on error.
+ */
+static void
+nfs_unroll_locks(struct nfs_page *head, struct nfs_page *req)
+{
+	struct nfs_page *tmp;
+
+	/* relinquish all the locks successfully grabbed this run */
+	for (tmp = head->wb_this_page ; tmp != req; tmp = tmp->wb_this_page) {
+		if (!kref_read(&tmp->wb_kref))
+			continue;
+		nfs_unlock_and_release_request(tmp);
+	}
+}
+
+/*
+ * nfs_page_group_lock_subreq -  try to lock a subrequest
+ * @head: head request of page group
+ * @subreq: request to lock
  *
- * return 0 on success, < 0 on error
+ * This is a helper function for nfs_lock_and_join_requests which
+ * must be called with the head request and page group both locked.
+ * On error, it returns with the page group unlocked.
  */
-int
-nfs_page_group_lock(struct nfs_page *req)
+static int
+nfs_page_group_lock_subreq(struct nfs_page *head, struct nfs_page *subreq)
 {
-	struct nfs_page *head = req->wb_head;
+	int ret;
+
+	if (!kref_get_unless_zero(&subreq->wb_kref))
+		return 0;
+	while (!nfs_lock_request(subreq)) {
+		nfs_page_group_unlock(head);
+		ret = nfs_wait_on_request(subreq);
+		if (!ret)
+			ret = nfs_page_group_lock(head);
+		if (ret < 0) {
+			nfs_unroll_locks(head, subreq);
+			nfs_release_request(subreq);
+			return ret;
+		}
+	}
+	return 0;
+}
+
+/*
+ * nfs_page_group_lock_subrequests -  try to lock the subrequests
+ * @head: head request of page group
+ *
+ * This is a helper function for nfs_lock_and_join_requests which
+ * must be called with the head request locked.
+ */
+int nfs_page_group_lock_subrequests(struct nfs_page *head)
+{
+	struct nfs_page *subreq;
+	int ret;
 
-	WARN_ON_ONCE(head != head->wb_head);
+	ret = nfs_page_group_lock(head);
+	if (ret < 0)
+		return ret;
+	/* lock each request in the page group */
+	for (subreq = head->wb_this_page; subreq != head;
+			subreq = subreq->wb_this_page) {
+		ret = nfs_page_group_lock_subreq(head, subreq);
+		if (ret < 0)
+			return ret;
+	}
+	nfs_page_group_unlock(head);
+	return 0;
+}
 
-	if (!test_and_set_bit(PG_HEADLOCK, &head->wb_flags))
+/*
+ * nfs_page_set_headlock - set the request PG_HEADLOCK
+ * @req: request that is to be locked
+ *
+ * this lock must be held when modifying req->wb_head
+ *
+ * return 0 on success, < 0 on error
+ */
+int
+nfs_page_set_headlock(struct nfs_page *req)
+{
+	if (!test_and_set_bit(PG_HEADLOCK, &req->wb_flags))
 		return 0;
 
-	set_bit(PG_CONTENDED1, &head->wb_flags);
+	set_bit(PG_CONTENDED1, &req->wb_flags);
 	smp_mb__after_atomic();
-	return wait_on_bit_lock(&head->wb_flags, PG_HEADLOCK,
+	return wait_on_bit_lock(&req->wb_flags, PG_HEADLOCK,
 				TASK_UNINTERRUPTIBLE);
 }
 
 /*
- * nfs_page_group_unlock - unlock the head of the page group
- * @req - request in group that is to be unlocked
+ * nfs_page_clear_headlock - clear the request PG_HEADLOCK
+ * @req: request that is to be locked
  */
 void
-nfs_page_group_unlock(struct nfs_page *req)
+nfs_page_clear_headlock(struct nfs_page *req)
 {
-	struct nfs_page *head = req->wb_head;
-
-	WARN_ON_ONCE(head != head->wb_head);
-
 	smp_mb__before_atomic();
-	clear_bit(PG_HEADLOCK, &head->wb_flags);
+	clear_bit(PG_HEADLOCK, &req->wb_flags);
 	smp_mb__after_atomic();
-	if (!test_bit(PG_CONTENDED1, &head->wb_flags))
+	if (!test_bit(PG_CONTENDED1, &req->wb_flags))
 		return;
-	wake_up_bit(&head->wb_flags, PG_HEADLOCK);
+	wake_up_bit(&req->wb_flags, PG_HEADLOCK);
+}
+
+/*
+ * nfs_page_group_lock - lock the head of the page group
+ * @req: request in group that is to be locked
+ *
+ * this lock must be held when traversing or modifying the page
+ * group list
+ *
+ * return 0 on success, < 0 on error
+ */
+int
+nfs_page_group_lock(struct nfs_page *req)
+{
+	int ret;
+
+	ret = nfs_page_set_headlock(req);
+	if (ret || req->wb_head == req)
+		return ret;
+	return nfs_page_set_headlock(req->wb_head);
+}
+
+/*
+ * nfs_page_group_unlock - unlock the head of the page group
+ * @req: request in group that is to be unlocked
+ */
+void
+nfs_page_group_unlock(struct nfs_page *req)
+{
+	if (req != req->wb_head)
+		nfs_page_clear_headlock(req->wb_head);
+	nfs_page_clear_headlock(req);
 }
 
 /*
@@ -359,15 +476,23 @@ nfs_create_request(struct nfs_open_context *ctx, struct page *page,
 }
 
 static struct nfs_page *
-nfs_create_subreq(struct nfs_page *req, struct nfs_page *last,
-		  unsigned int pgbase, unsigned int offset,
+nfs_create_subreq(struct nfs_page *req,
+		  unsigned int pgbase,
+		  unsigned int offset,
 		  unsigned int count)
 {
+	struct nfs_page *last;
 	struct nfs_page *ret;
 
 	ret = __nfs_create_request(req->wb_lock_context, req->wb_page,
 			pgbase, offset, count);
 	if (!IS_ERR(ret)) {
+		/* find the last request */
+		for (last = req->wb_head;
+		     last->wb_this_page != req->wb_head;
+		     last = last->wb_this_page)
+			;
+
 		nfs_lock_request(ret);
 		ret->wb_index = req->wb_index;
 		nfs_page_group_init(ret, last);
@@ -629,7 +754,6 @@ int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_header *hdr,
 		.workqueue = nfsiod_workqueue,
 		.flags = RPC_TASK_ASYNC | flags,
 	};
-	int ret = 0;
 
 	hdr->rw_ops->rw_initiate(hdr, &msg, rpc_ops, &task_setup_data, how);
 
@@ -641,18 +765,10 @@ int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_header *hdr,
 		(unsigned long long)hdr->args.offset);
 
 	task = rpc_run_task(&task_setup_data);
-	if (IS_ERR(task)) {
-		ret = PTR_ERR(task);
-		goto out;
-	}
-	if (how & FLUSH_SYNC) {
-		ret = rpc_wait_for_completion_task(task);
-		if (ret == 0)
-			ret = task->tk_status;
-	}
+	if (IS_ERR(task))
+		return PTR_ERR(task);
 	rpc_put_task(task);
-out:
-	return ret;
+	return 0;
 }
 EXPORT_SYMBOL_GPL(nfs_initiate_pgio);
 
@@ -834,7 +950,8 @@ static int nfs_generic_pg_pgios(struct nfs_pageio_descriptor *desc)
 					hdr->cred,
 					NFS_PROTO(hdr->inode),
 					desc->pg_rpc_callops,
-					desc->pg_ioflags, 0);
+					desc->pg_ioflags,
+					RPC_TASK_CRED_NOREF);
 	return ret;
 }
 
@@ -886,15 +1003,6 @@ static void nfs_pageio_setup_mirroring(struct nfs_pageio_descriptor *pgio,
 	pgio->pg_mirror_count = mirror_count;
 }
 
-/*
- * nfs_pageio_stop_mirroring - stop using mirroring (set mirror count to 1)
- */
-void nfs_pageio_stop_mirroring(struct nfs_pageio_descriptor *pgio)
-{
-	pgio->pg_mirror_count = 1;
-	pgio->pg_mirror_idx = 0;
-}
-
 static void nfs_pageio_cleanup_mirroring(struct nfs_pageio_descriptor *pgio)
 {
 	pgio->pg_mirror_count = 1;
@@ -911,7 +1019,7 @@ static bool nfs_match_lock_context(const struct nfs_lock_context *l1,
 }
 
 /**
- * nfs_can_coalesce_requests - test two requests for compatibility
+ * nfs_coalesce_size - test two requests for compatibility
  * @prev: pointer to nfs_page
  * @req: pointer to nfs_page
  * @pgio: pointer to nfs_pagio_descriptor
@@ -920,41 +1028,36 @@ static bool nfs_match_lock_context(const struct nfs_lock_context *l1,
  * page data area they describe is contiguous, and that their RPC
  * credentials, NFSv4 open state, and lockowners are the same.
  *
- * Return 'true' if this is the case, else return 'false'.
+ * Returns size of the request that can be coalesced
  */
-static bool nfs_can_coalesce_requests(struct nfs_page *prev,
+static unsigned int nfs_coalesce_size(struct nfs_page *prev,
 				      struct nfs_page *req,
 				      struct nfs_pageio_descriptor *pgio)
 {
-	size_t size;
 	struct file_lock_context *flctx;
 
 	if (prev) {
 		if (!nfs_match_open_context(nfs_req_openctx(req), nfs_req_openctx(prev)))
-			return false;
+			return 0;
 		flctx = d_inode(nfs_req_openctx(req)->dentry)->i_flctx;
 		if (flctx != NULL &&
 		    !(list_empty_careful(&flctx->flc_posix) &&
 		      list_empty_careful(&flctx->flc_flock)) &&
 		    !nfs_match_lock_context(req->wb_lock_context,
 					    prev->wb_lock_context))
-			return false;
+			return 0;
 		if (req_offset(req) != req_offset(prev) + prev->wb_bytes)
-			return false;
+			return 0;
 		if (req->wb_page == prev->wb_page) {
 			if (req->wb_pgbase != prev->wb_pgbase + prev->wb_bytes)
-				return false;
+				return 0;
 		} else {
 			if (req->wb_pgbase != 0 ||
 			    prev->wb_pgbase + prev->wb_bytes != PAGE_SIZE)
-				return false;
+				return 0;
 		}
 	}
-	size = pgio->pg_ops->pg_test(pgio, prev, req);
-	WARN_ON_ONCE(size > req->wb_bytes);
-	if (size && size < req->wb_bytes)
-		req->wb_bytes = size;
-	return size > 0;
+	return pgio->pg_ops->pg_test(pgio, prev, req);
 }
 
 /**
@@ -962,15 +1065,16 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev,
  * @desc: destination io descriptor
  * @req: request
  *
- * Returns true if the request 'req' was successfully coalesced into the
- * existing list of pages 'desc'.
+ * If the request 'req' was successfully coalesced into the existing list
+ * of pages 'desc', it returns the size of req.
  */
-static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc,
-				     struct nfs_page *req)
+static unsigned int
+nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc,
+		struct nfs_page *req)
 {
 	struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc);
-
 	struct nfs_page *prev = NULL;
+	unsigned int size;
 
 	if (mirror->pg_count != 0) {
 		prev = nfs_list_entry(mirror->pg_list.prev);
@@ -990,11 +1094,12 @@ static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc,
 		return 0;
 	}
 
-	if (!nfs_can_coalesce_requests(prev, req, desc))
-		return 0;
+	size = nfs_coalesce_size(prev, req, desc);
+	if (size < req->wb_bytes)
+		return size;
 	nfs_list_move_request(req, &mirror->pg_list);
 	mirror->pg_count += req->wb_bytes;
-	return 1;
+	return req->wb_bytes;
 }
 
 /*
@@ -1034,7 +1139,8 @@ nfs_pageio_cleanup_request(struct nfs_pageio_descriptor *desc,
  * @req: request
  *
  * This may split a request into subrequests which are all part of the
- * same page group.
+ * same page group. If so, it will submit @req as the last one, to ensure
+ * the pointer to @req is still valid in case of failure.
  *
  * Returns true if the request 'req' was successfully coalesced into the
  * existing list of pages 'desc'.
@@ -1043,51 +1149,50 @@ static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 			   struct nfs_page *req)
 {
 	struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc);
-
 	struct nfs_page *subreq;
-	unsigned int bytes_left = 0;
-	unsigned int offset, pgbase;
+	unsigned int size, subreq_size;
 
 	nfs_page_group_lock(req);
 
 	subreq = req;
-	bytes_left = subreq->wb_bytes;
-	offset = subreq->wb_offset;
-	pgbase = subreq->wb_pgbase;
-
-	do {
-		if (!nfs_pageio_do_add_request(desc, subreq)) {
-			/* make sure pg_test call(s) did nothing */
-			WARN_ON_ONCE(subreq->wb_bytes != bytes_left);
-			WARN_ON_ONCE(subreq->wb_offset != offset);
-			WARN_ON_ONCE(subreq->wb_pgbase != pgbase);
-
+	subreq_size = subreq->wb_bytes;
+	for(;;) {
+		size = nfs_pageio_do_add_request(desc, subreq);
+		if (size == subreq_size) {
+			/* We successfully submitted a request */
+			if (subreq == req)
+				break;
+			req->wb_pgbase += size;
+			req->wb_bytes -= size;
+			req->wb_offset += size;
+			subreq_size = req->wb_bytes;
+			subreq = req;
+			continue;
+		}
+		if (WARN_ON_ONCE(subreq != req)) {
+			nfs_page_group_unlock(req);
+			nfs_pageio_cleanup_request(desc, subreq);
+			subreq = req;
+			subreq_size = req->wb_bytes;
+			nfs_page_group_lock(req);
+		}
+		if (!size) {
+			/* Can't coalesce any more, so do I/O */
 			nfs_page_group_unlock(req);
 			desc->pg_moreio = 1;
 			nfs_pageio_doio(desc);
 			if (desc->pg_error < 0 || mirror->pg_recoalesce)
-				goto out_cleanup_subreq;
+				return 0;
 			/* retry add_request for this subreq */
 			nfs_page_group_lock(req);
 			continue;
 		}
-
-		/* check for buggy pg_test call(s) */
-		WARN_ON_ONCE(subreq->wb_bytes + subreq->wb_pgbase > PAGE_SIZE);
-		WARN_ON_ONCE(subreq->wb_bytes > bytes_left);
-		WARN_ON_ONCE(subreq->wb_bytes == 0);
-
-		bytes_left -= subreq->wb_bytes;
-		offset += subreq->wb_bytes;
-		pgbase += subreq->wb_bytes;
-
-		if (bytes_left) {
-			subreq = nfs_create_subreq(req, subreq, pgbase,
-					offset, bytes_left);
-			if (IS_ERR(subreq))
-				goto err_ptr;
-		}
-	} while (bytes_left > 0);
+		subreq = nfs_create_subreq(req, req->wb_pgbase,
+				req->wb_offset, size);
+		if (IS_ERR(subreq))
+			goto err_ptr;
+		subreq_size = size;
+	}
 
 	nfs_page_group_unlock(req);
 	return 1;
@@ -1095,10 +1200,6 @@ err_ptr:
 	desc->pg_error = PTR_ERR(subreq);
 	nfs_page_group_unlock(req);
 	return 0;
-out_cleanup_subreq:
-	if (req != subreq)
-		nfs_pageio_cleanup_request(desc, subreq);
-	return 0;
 }
 
 static int nfs_do_recoalesce(struct nfs_pageio_descriptor *desc)
@@ -1167,7 +1268,7 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 {
 	u32 midx;
 	unsigned int pgbase, offset, bytes;
-	struct nfs_page *dupreq, *lastreq;
+	struct nfs_page *dupreq;
 
 	pgbase = req->wb_pgbase;
 	offset = req->wb_offset;
@@ -1177,38 +1278,32 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
 	if (desc->pg_error < 0)
 		goto out_failed;
 
-	for (midx = 0; midx < desc->pg_mirror_count; midx++) {
-		if (midx) {
-			nfs_page_group_lock(req);
+	/* Create the mirror instances first, and fire them off */
+	for (midx = 1; midx < desc->pg_mirror_count; midx++) {
+		nfs_page_group_lock(req);
 
-			/* find the last request */
-			for (lastreq = req->wb_head;
-			     lastreq->wb_this_page != req->wb_head;
-			     lastreq = lastreq->wb_this_page)
-				;
+		dupreq = nfs_create_subreq(req,
+				pgbase, offset, bytes);
 
-			dupreq = nfs_create_subreq(req, lastreq,
-					pgbase, offset, bytes);
-
-			nfs_page_group_unlock(req);
-			if (IS_ERR(dupreq)) {
-				desc->pg_error = PTR_ERR(dupreq);
-				goto out_failed;
-			}
-		} else
-			dupreq = req;
+		nfs_page_group_unlock(req);
+		if (IS_ERR(dupreq)) {
+			desc->pg_error = PTR_ERR(dupreq);
+			goto out_failed;
+		}
 
-		if (nfs_pgio_has_mirroring(desc))
-			desc->pg_mirror_idx = midx;
+		desc->pg_mirror_idx = midx;
 		if (!nfs_pageio_add_request_mirror(desc, dupreq))
 			goto out_cleanup_subreq;
 	}
 
+	desc->pg_mirror_idx = 0;
+	if (!nfs_pageio_add_request_mirror(desc, req))
+		goto out_failed;
+
 	return 1;
 
 out_cleanup_subreq:
-	if (req != dupreq)
-		nfs_pageio_cleanup_request(desc, dupreq);
+	nfs_pageio_cleanup_request(desc, dupreq);
 out_failed:
 	nfs_pageio_error_cleanup(desc);
 	return 0;
@@ -1226,8 +1321,7 @@ static void nfs_pageio_complete_mirror(struct nfs_pageio_descriptor *desc,
 	struct nfs_pgio_mirror *mirror = &desc->pg_mirrors[mirror_idx];
 	u32 restore_idx = desc->pg_mirror_idx;
 
-	if (nfs_pgio_has_mirroring(desc))
-		desc->pg_mirror_idx = mirror_idx;
+	desc->pg_mirror_idx = mirror_idx;
 	for (;;) {
 		nfs_pageio_doio(desc);
 		if (desc->pg_error < 0 || !mirror->pg_recoalesce)
@@ -1320,6 +1414,14 @@ void nfs_pageio_cond_complete(struct nfs_pageio_descriptor *desc, pgoff_t index)
 	}
 }
 
+/*
+ * nfs_pageio_stop_mirroring - stop using mirroring (set mirror count to 1)
+ */
+void nfs_pageio_stop_mirroring(struct nfs_pageio_descriptor *pgio)
+{
+	nfs_pageio_complete(pgio);
+}
+
 int __init nfs_init_nfspagecache(void)
 {
 	nfs_page_cachep = kmem_cache_create("nfs_page",
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index bb80034a7661..dd2e14f5875d 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -268,11 +268,11 @@ pnfs_free_layout_hdr(struct pnfs_layout_hdr *lo)
 	struct nfs_server *server = NFS_SERVER(lo->plh_inode);
 	struct pnfs_layoutdriver_type *ld = server->pnfs_curr_ld;
 
-	if (!list_empty(&lo->plh_layouts)) {
+	if (test_and_clear_bit(NFS_LAYOUT_HASHED, &lo->plh_flags)) {
 		struct nfs_client *clp = server->nfs_client;
 
 		spin_lock(&clp->cl_lock);
-		list_del_init(&lo->plh_layouts);
+		list_del_rcu(&lo->plh_layouts);
 		spin_unlock(&clp->cl_lock);
 	}
 	put_cred(lo->plh_lc_cred);
@@ -309,6 +309,16 @@ pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo)
 	}
 }
 
+static struct inode *
+pnfs_grab_inode_layout_hdr(struct pnfs_layout_hdr *lo)
+{
+	struct inode *inode = igrab(lo->plh_inode);
+	if (inode)
+		return inode;
+	set_bit(NFS_LAYOUT_INODE_FREEING, &lo->plh_flags);
+	return NULL;
+}
+
 static void
 pnfs_set_plh_return_info(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode,
 			 u32 seq)
@@ -496,6 +506,7 @@ pnfs_init_lseg(struct pnfs_layout_hdr *lo, struct pnfs_layout_segment *lseg,
 {
 	INIT_LIST_HEAD(&lseg->pls_list);
 	INIT_LIST_HEAD(&lseg->pls_lc_list);
+	INIT_LIST_HEAD(&lseg->pls_commits);
 	refcount_set(&lseg->pls_refcount, 1);
 	set_bit(NFS_LSEG_VALID, &lseg->pls_flags);
 	lseg->pls_layout = lo;
@@ -782,9 +793,10 @@ pnfs_layout_bulk_destroy_byserver_locked(struct nfs_client *clp,
 		/* If the sb is being destroyed, just bail */
 		if (!nfs_sb_active(server->super))
 			break;
-		inode = igrab(lo->plh_inode);
+		inode = pnfs_grab_inode_layout_hdr(lo);
 		if (inode != NULL) {
-			list_del_init(&lo->plh_layouts);
+			if (test_and_clear_bit(NFS_LAYOUT_HASHED, &lo->plh_flags))
+				list_del_rcu(&lo->plh_layouts);
 			if (pnfs_layout_add_bulk_destroy_list(inode,
 						layout_list))
 				continue;
@@ -794,7 +806,6 @@ pnfs_layout_bulk_destroy_byserver_locked(struct nfs_client *clp,
 		} else {
 			rcu_read_unlock();
 			spin_unlock(&clp->cl_lock);
-			set_bit(NFS_LAYOUT_INODE_FREEING, &lo->plh_flags);
 		}
 		nfs_sb_deactive(server->super);
 		spin_lock(&clp->cl_lock);
@@ -903,10 +914,21 @@ pnfs_destroy_all_layouts(struct nfs_client *clp)
 	pnfs_destroy_layouts_byclid(clp, false);
 }
 
+static void
+pnfs_set_layout_cred(struct pnfs_layout_hdr *lo, const struct cred *cred)
+{
+	const struct cred *old;
+
+	if (cred && cred_fscmp(lo->plh_lc_cred, cred) != 0) {
+		old = xchg(&lo->plh_lc_cred, get_cred(cred));
+		put_cred(old);
+	}
+}
+
 /* update lo->plh_stateid with new if is more recent */
 void
 pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo, const nfs4_stateid *new,
-			bool update_barrier)
+			const struct cred *cred, bool update_barrier)
 {
 	u32 oldseq, newseq, new_barrier = 0;
 
@@ -914,6 +936,7 @@ pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo, const nfs4_stateid *new,
 	newseq = be32_to_cpu(new->seqid);
 
 	if (!pnfs_layout_is_valid(lo)) {
+		pnfs_set_layout_cred(lo, cred);
 		nfs4_stateid_copy(&lo->plh_stateid, new);
 		lo->plh_barrier = newseq;
 		pnfs_clear_layoutreturn_info(lo);
@@ -1061,7 +1084,7 @@ pnfs_alloc_init_layoutget_args(struct inode *ino,
 	lgp->args.ctx = get_nfs_open_context(ctx);
 	nfs4_stateid_copy(&lgp->args.stateid, stateid);
 	lgp->gfp_flags = gfp_flags;
-	lgp->cred = get_cred(ctx->cred);
+	lgp->cred = ctx->cred;
 	return lgp;
 }
 
@@ -1072,7 +1095,6 @@ void pnfs_layoutget_free(struct nfs4_layoutget *lgp)
 	nfs4_free_pages(lgp->args.layout.pages, max_pages);
 	if (lgp->args.inode)
 		pnfs_put_layout_hdr(NFS_I(lgp->args.inode)->layout);
-	put_cred(lgp->cred);
 	put_nfs_open_context(lgp->args.ctx);
 	kfree(lgp);
 }
@@ -1109,7 +1131,7 @@ void pnfs_layoutreturn_free_lsegs(struct pnfs_layout_hdr *lo,
 
 		pnfs_mark_matching_lsegs_invalid(lo, &freeme, range, seq);
 		pnfs_free_returned_lsegs(lo, &freeme, range, seq);
-		pnfs_set_layout_stateid(lo, stateid, true);
+		pnfs_set_layout_stateid(lo, stateid, NULL, true);
 	} else
 		pnfs_mark_layout_stateid_invalid(lo, &freeme);
 out_unlock:
@@ -1122,6 +1144,7 @@ out_unlock:
 static bool
 pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo,
 		nfs4_stateid *stateid,
+		const struct cred **cred,
 		enum pnfs_iomode *iomode)
 {
 	/* Serialise LAYOUTGET/LAYOUTRETURN */
@@ -1132,18 +1155,17 @@ pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo,
 	set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags);
 	pnfs_get_layout_hdr(lo);
 	if (test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) {
-		if (stateid != NULL) {
-			nfs4_stateid_copy(stateid, &lo->plh_stateid);
-			if (lo->plh_return_seq != 0)
-				stateid->seqid = cpu_to_be32(lo->plh_return_seq);
-		}
+		nfs4_stateid_copy(stateid, &lo->plh_stateid);
+		*cred = get_cred(lo->plh_lc_cred);
+		if (lo->plh_return_seq != 0)
+			stateid->seqid = cpu_to_be32(lo->plh_return_seq);
 		if (iomode != NULL)
 			*iomode = lo->plh_return_iomode;
 		pnfs_clear_layoutreturn_info(lo);
 		return true;
 	}
-	if (stateid != NULL)
-		nfs4_stateid_copy(stateid, &lo->plh_stateid);
+	nfs4_stateid_copy(stateid, &lo->plh_stateid);
+	*cred = get_cred(lo->plh_lc_cred);
 	if (iomode != NULL)
 		*iomode = IOMODE_ANY;
 	return true;
@@ -1167,20 +1189,26 @@ pnfs_init_layoutreturn_args(struct nfs4_layoutreturn_args *args,
 }
 
 static int
-pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, const nfs4_stateid *stateid,
-		       enum pnfs_iomode iomode, bool sync)
+pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo,
+		       const nfs4_stateid *stateid,
+		       const struct cred **pcred,
+		       enum pnfs_iomode iomode,
+		       bool sync)
 {
 	struct inode *ino = lo->plh_inode;
 	struct pnfs_layoutdriver_type *ld = NFS_SERVER(ino)->pnfs_curr_ld;
 	struct nfs4_layoutreturn *lrp;
+	const struct cred *cred = *pcred;
 	int status = 0;
 
+	*pcred = NULL;
 	lrp = kzalloc(sizeof(*lrp), GFP_NOFS);
 	if (unlikely(lrp == NULL)) {
 		status = -ENOMEM;
 		spin_lock(&ino->i_lock);
 		pnfs_clear_layoutreturn_waitbit(lo);
 		spin_unlock(&ino->i_lock);
+		put_cred(cred);
 		pnfs_put_layout_hdr(lo);
 		goto out;
 	}
@@ -1188,7 +1216,7 @@ pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, const nfs4_stateid *stateid,
 	pnfs_init_layoutreturn_args(&lrp->args, lo, stateid, iomode);
 	lrp->args.ld_private = &lrp->ld_private;
 	lrp->clp = NFS_SERVER(ino)->nfs_client;
-	lrp->cred = lo->plh_lc_cred;
+	lrp->cred = cred;
 	if (ld->prepare_layoutreturn)
 		ld->prepare_layoutreturn(&lrp->args);
 
@@ -1233,15 +1261,16 @@ static void pnfs_layoutreturn_before_put_layout_hdr(struct pnfs_layout_hdr *lo)
 		return;
 	spin_lock(&inode->i_lock);
 	if (pnfs_layout_need_return(lo)) {
+		const struct cred *cred;
 		nfs4_stateid stateid;
 		enum pnfs_iomode iomode;
 		bool send;
 
-		send = pnfs_prepare_layoutreturn(lo, &stateid, &iomode);
+		send = pnfs_prepare_layoutreturn(lo, &stateid, &cred, &iomode);
 		spin_unlock(&inode->i_lock);
 		if (send) {
 			/* Send an async layoutreturn so we dont deadlock */
-			pnfs_send_layoutreturn(lo, &stateid, iomode, false);
+			pnfs_send_layoutreturn(lo, &stateid, &cred, iomode, false);
 		}
 	} else
 		spin_unlock(&inode->i_lock);
@@ -1261,6 +1290,7 @@ _pnfs_return_layout(struct inode *ino)
 	struct pnfs_layout_hdr *lo = NULL;
 	struct nfs_inode *nfsi = NFS_I(ino);
 	LIST_HEAD(tmp_list);
+	const struct cred *cred;
 	nfs4_stateid stateid;
 	int status = 0;
 	bool send, valid_layout;
@@ -1302,13 +1332,15 @@ _pnfs_return_layout(struct inode *ino)
 			!valid_layout) {
 		spin_unlock(&ino->i_lock);
 		dprintk("NFS: %s no layout segments to return\n", __func__);
-		goto out_put_layout_hdr;
+		goto out_wait_layoutreturn;
 	}
 
-	send = pnfs_prepare_layoutreturn(lo, &stateid, NULL);
+	send = pnfs_prepare_layoutreturn(lo, &stateid, &cred, NULL);
 	spin_unlock(&ino->i_lock);
 	if (send)
-		status = pnfs_send_layoutreturn(lo, &stateid, IOMODE_ANY, true);
+		status = pnfs_send_layoutreturn(lo, &stateid, &cred, IOMODE_ANY, true);
+out_wait_layoutreturn:
+	wait_on_bit(&lo->plh_flags, NFS_LAYOUT_RETURN, TASK_UNINTERRUPTIBLE);
 out_put_layout_hdr:
 	pnfs_free_lseg_list(&tmp_list);
 	pnfs_put_layout_hdr(lo);
@@ -1354,6 +1386,7 @@ bool pnfs_roc(struct inode *ino,
 	struct nfs4_state *state;
 	struct pnfs_layout_hdr *lo;
 	struct pnfs_layout_segment *lseg, *next;
+	const struct cred *lc_cred;
 	nfs4_stateid stateid;
 	enum pnfs_iomode iomode = 0;
 	bool layoutreturn = false, roc = false;
@@ -1423,15 +1456,16 @@ retry:
 	 * 2. we don't send layoutreturn
 	 */
 	/* lo ref dropped in pnfs_roc_release() */
-	layoutreturn = pnfs_prepare_layoutreturn(lo, &stateid, &iomode);
+	layoutreturn = pnfs_prepare_layoutreturn(lo, &stateid, &lc_cred, &iomode);
 	/* If the creds don't match, we can't compound the layoutreturn */
-	if (!layoutreturn || cred != lo->plh_lc_cred)
+	if (!layoutreturn || cred_fscmp(cred, lc_cred) != 0)
 		goto out_noroc;
 
 	roc = layoutreturn;
 	pnfs_init_layoutreturn_args(args, lo, &stateid, iomode);
 	res->lrs_present = 0;
 	layoutreturn = false;
+	put_cred(lc_cred);
 
 out_noroc:
 	spin_unlock(&ino->i_lock);
@@ -1445,7 +1479,7 @@ out_noroc:
 		return true;
 	}
 	if (layoutreturn)
-		pnfs_send_layoutreturn(lo, &stateid, iomode, true);
+		pnfs_send_layoutreturn(lo, &stateid, &lc_cred, iomode, true);
 	pnfs_put_layout_hdr(lo);
 	return false;
 }
@@ -1859,15 +1893,14 @@ static void pnfs_clear_first_layoutget(struct pnfs_layout_hdr *lo)
 static void _add_to_server_list(struct pnfs_layout_hdr *lo,
 				struct nfs_server *server)
 {
-	if (list_empty(&lo->plh_layouts)) {
+	if (!test_and_set_bit(NFS_LAYOUT_HASHED, &lo->plh_flags)) {
 		struct nfs_client *clp = server->nfs_client;
 
 		/* The lo must be on the clp list if there is any
 		 * chance of a CB_LAYOUTRECALL(FILE) coming in.
 		 */
 		spin_lock(&clp->cl_lock);
-		if (list_empty(&lo->plh_layouts))
-			list_add_tail(&lo->plh_layouts, &server->layouts);
+		list_add_tail_rcu(&lo->plh_layouts, &server->layouts);
 		spin_unlock(&clp->cl_lock);
 	}
 }
@@ -1989,6 +2022,7 @@ lookup_again:
 			goto lookup_again;
 		}
 
+		spin_unlock(&ino->i_lock);
 		first = true;
 		status = nfs4_select_rw_stateid(ctx->state,
 					iomode == IOMODE_RW ? FMODE_WRITE : FMODE_READ,
@@ -1998,14 +2032,12 @@ lookup_again:
 			trace_pnfs_update_layout(ino, pos, count,
 					iomode, lo, lseg,
 					PNFS_UPDATE_LAYOUT_INVALID_OPEN);
-			if (status != -EAGAIN)
-				goto out_unlock;
-			spin_unlock(&ino->i_lock);
 			nfs4_schedule_stateid_recovery(server, ctx->state);
 			pnfs_clear_first_layoutget(lo);
 			pnfs_put_layout_hdr(lo);
 			goto lookup_again;
 		}
+		spin_lock(&ino->i_lock);
 	} else {
 		nfs4_stateid_copy(&stateid, &lo->plh_stateid);
 	}
@@ -2160,8 +2192,6 @@ out_unlock:
 	return NULL;
 }
 
-extern const nfs4_stateid current_stateid;
-
 static void _lgopen_prepare_attached(struct nfs4_opendata *data,
 				     struct nfs_open_context *ctx)
 {
@@ -2327,14 +2357,14 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
 
 	if (!pnfs_layout_is_valid(lo)) {
 		/* We have a completely new layout */
-		pnfs_set_layout_stateid(lo, &res->stateid, true);
+		pnfs_set_layout_stateid(lo, &res->stateid, lgp->cred, true);
 	} else if (nfs4_stateid_match_other(&lo->plh_stateid, &res->stateid)) {
 		/* existing state ID, make sure the sequence number matches. */
 		if (pnfs_layout_stateid_blocked(lo, &res->stateid)) {
 			dprintk("%s forget reply due to sequence\n", __func__);
 			goto out_forget;
 		}
-		pnfs_set_layout_stateid(lo, &res->stateid, false);
+		pnfs_set_layout_stateid(lo, &res->stateid, lgp->cred, false);
 	} else {
 		/*
 		 * We got an entirely new state ID.  Mark all segments for the
@@ -2427,43 +2457,159 @@ pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
 	return -ENOENT;
 }
 
-void pnfs_error_mark_layout_for_return(struct inode *inode,
-				       struct pnfs_layout_segment *lseg)
+static void
+pnfs_mark_layout_for_return(struct inode *inode,
+			    const struct pnfs_layout_range *range)
 {
-	struct pnfs_layout_hdr *lo = NFS_I(inode)->layout;
-	struct pnfs_layout_range range = {
-		.iomode = lseg->pls_range.iomode,
-		.offset = 0,
-		.length = NFS4_MAX_UINT64,
-	};
+	struct pnfs_layout_hdr *lo;
 	bool return_now = false;
 
 	spin_lock(&inode->i_lock);
+	lo = NFS_I(inode)->layout;
 	if (!pnfs_layout_is_valid(lo)) {
 		spin_unlock(&inode->i_lock);
 		return;
 	}
-	pnfs_set_plh_return_info(lo, range.iomode, 0);
+	pnfs_set_plh_return_info(lo, range->iomode, 0);
 	/*
 	 * mark all matching lsegs so that we are sure to have no live
 	 * segments at hand when sending layoutreturn. See pnfs_put_lseg()
 	 * for how it works.
 	 */
-	if (pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs, &range, 0) != -EBUSY) {
+	if (pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs, range, 0) != -EBUSY) {
+		const struct cred *cred;
 		nfs4_stateid stateid;
 		enum pnfs_iomode iomode;
 
-		return_now = pnfs_prepare_layoutreturn(lo, &stateid, &iomode);
+		return_now = pnfs_prepare_layoutreturn(lo, &stateid, &cred, &iomode);
 		spin_unlock(&inode->i_lock);
 		if (return_now)
-			pnfs_send_layoutreturn(lo, &stateid, iomode, false);
+			pnfs_send_layoutreturn(lo, &stateid, &cred, iomode, false);
 	} else {
 		spin_unlock(&inode->i_lock);
 		nfs_commit_inode(inode, 0);
 	}
 }
+
+void pnfs_error_mark_layout_for_return(struct inode *inode,
+				       struct pnfs_layout_segment *lseg)
+{
+	struct pnfs_layout_range range = {
+		.iomode = lseg->pls_range.iomode,
+		.offset = 0,
+		.length = NFS4_MAX_UINT64,
+	};
+
+	pnfs_mark_layout_for_return(inode, &range);
+}
 EXPORT_SYMBOL_GPL(pnfs_error_mark_layout_for_return);
 
+static bool
+pnfs_layout_can_be_returned(struct pnfs_layout_hdr *lo)
+{
+	return pnfs_layout_is_valid(lo) &&
+		!test_bit(NFS_LAYOUT_INODE_FREEING, &lo->plh_flags) &&
+		!test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags);
+}
+
+static struct pnfs_layout_segment *
+pnfs_find_first_lseg(struct pnfs_layout_hdr *lo,
+		     const struct pnfs_layout_range *range,
+		     enum pnfs_iomode iomode)
+{
+	struct pnfs_layout_segment *lseg;
+
+	list_for_each_entry(lseg, &lo->plh_segs, pls_list) {
+		if (!test_bit(NFS_LSEG_VALID, &lseg->pls_flags))
+			continue;
+		if (test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags))
+			continue;
+		if (lseg->pls_range.iomode != iomode && iomode != IOMODE_ANY)
+			continue;
+		if (pnfs_lseg_range_intersecting(&lseg->pls_range, range))
+			return lseg;
+	}
+	return NULL;
+}
+
+/* Find open file states whose mode matches that of the range */
+static bool
+pnfs_should_return_unused_layout(struct pnfs_layout_hdr *lo,
+				 const struct pnfs_layout_range *range)
+{
+	struct list_head *head;
+	struct nfs_open_context *ctx;
+	fmode_t mode = 0;
+
+	if (!pnfs_layout_can_be_returned(lo) ||
+	    !pnfs_find_first_lseg(lo, range, range->iomode))
+		return false;
+
+	head = &NFS_I(lo->plh_inode)->open_files;
+	list_for_each_entry_rcu(ctx, head, list) {
+		if (ctx->state)
+			mode |= ctx->state->state & (FMODE_READ|FMODE_WRITE);
+	}
+
+	switch (range->iomode) {
+	default:
+		break;
+	case IOMODE_READ:
+		mode &= ~FMODE_WRITE;
+		break;
+	case IOMODE_RW:
+		if (pnfs_find_first_lseg(lo, range, IOMODE_READ))
+			mode &= ~FMODE_READ;
+	}
+	return mode == 0;
+}
+
+static int
+pnfs_layout_return_unused_byserver(struct nfs_server *server, void *data)
+{
+	const struct pnfs_layout_range *range = data;
+	struct pnfs_layout_hdr *lo;
+	struct inode *inode;
+restart:
+	rcu_read_lock();
+	list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) {
+		if (!pnfs_layout_can_be_returned(lo) ||
+		    test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags))
+			continue;
+		inode = lo->plh_inode;
+		spin_lock(&inode->i_lock);
+		if (!pnfs_should_return_unused_layout(lo, range)) {
+			spin_unlock(&inode->i_lock);
+			continue;
+		}
+		spin_unlock(&inode->i_lock);
+		inode = pnfs_grab_inode_layout_hdr(lo);
+		if (!inode)
+			continue;
+		rcu_read_unlock();
+		pnfs_mark_layout_for_return(inode, range);
+		iput(inode);
+		cond_resched();
+		goto restart;
+	}
+	rcu_read_unlock();
+	return 0;
+}
+
+void
+pnfs_layout_return_unused_byclid(struct nfs_client *clp,
+				 enum pnfs_iomode iomode)
+{
+	struct pnfs_layout_range range = {
+		.iomode = iomode,
+		.offset = 0,
+		.length = NFS4_MAX_UINT64,
+	};
+
+	nfs_client_for_each_server(clp, pnfs_layout_return_unused_byserver,
+			&range);
+}
+
 void
 pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio)
 {
@@ -2479,7 +2625,7 @@ EXPORT_SYMBOL_GPL(pnfs_generic_pg_check_layout);
  * Check for any intersection between the request and the pgio->pg_lseg,
  * and if none, put this pgio->pg_lseg away.
  */
-static void
+void
 pnfs_generic_pg_check_range(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
 {
 	if (pgio->pg_lseg && !pnfs_lseg_request_intersecting(pgio->pg_lseg, req)) {
@@ -2487,6 +2633,7 @@ pnfs_generic_pg_check_range(struct nfs_pageio_descriptor *pgio, struct nfs_page
 		pgio->pg_lseg = NULL;
 	}
 }
+EXPORT_SYMBOL_GPL(pnfs_generic_pg_check_range);
 
 void
 pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
@@ -3004,10 +3151,10 @@ pnfs_layoutcommit_inode(struct inode *inode, bool sync)
 	end_pos = nfsi->layout->plh_lwb;
 
 	nfs4_stateid_copy(&data->args.stateid, &nfsi->layout->plh_stateid);
+	data->cred = get_cred(nfsi->layout->plh_lc_cred);
 	spin_unlock(&inode->i_lock);
 
 	data->args.inode = inode;
-	data->cred = get_cred(nfsi->layout->plh_lc_cred);
 	nfs_fattr_init(&data->fattr);
 	data->args.bitmask = NFS_SERVER(inode)->cache_consistency_bitmask;
 	data->res.fattr = &data->fattr;
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index f8a38065c7e4..8e0ada581b92 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -66,6 +66,7 @@ struct nfs4_pnfs_ds {
 struct pnfs_layout_segment {
 	struct list_head pls_list;
 	struct list_head pls_lc_list;
+	struct list_head pls_commits;
 	struct pnfs_layout_range pls_range;
 	refcount_t pls_refcount;
 	u32 pls_seq;
@@ -79,6 +80,10 @@ enum pnfs_try_status {
 	PNFS_TRY_AGAIN     = 2,
 };
 
+/* error codes for internal use */
+#define NFS4ERR_RESET_TO_MDS   12001
+#define NFS4ERR_RESET_TO_PNFS  12002
+
 #ifdef CONFIG_NFS_V4_1
 
 #define LAYOUT_NFSV4_1_MODULE_PREFIX "nfs-layouttype4"
@@ -91,10 +96,6 @@ enum pnfs_try_status {
 #define NFS4_DEF_DS_RETRANS 5
 #define PNFS_DEVICE_RETRY_TIMEOUT (120*HZ)
 
-/* error codes for internal use */
-#define NFS4ERR_RESET_TO_MDS   12001
-#define NFS4ERR_RESET_TO_PNFS  12002
-
 enum {
 	NFS_LAYOUT_RO_FAILED = 0,	/* get ro layout failed stop trying */
 	NFS_LAYOUT_RW_FAILED,		/* get rw layout failed stop trying */
@@ -105,6 +106,7 @@ enum {
 	NFS_LAYOUT_INVALID_STID,	/* layout stateid id is invalid */
 	NFS_LAYOUT_FIRST_LAYOUTGET,	/* Serialize first layoutget */
 	NFS_LAYOUT_INODE_FREEING,	/* The inode is being freed */
+	NFS_LAYOUT_HASHED,		/* The layout visible */
 };
 
 enum layoutdriver_policy_flags {
@@ -148,22 +150,6 @@ struct pnfs_layoutdriver_type {
 	const struct nfs_pageio_ops *pg_write_ops;
 
 	struct pnfs_ds_commit_info *(*get_ds_info) (struct inode *inode);
-	void (*mark_request_commit) (struct nfs_page *req,
-				     struct pnfs_layout_segment *lseg,
-				     struct nfs_commit_info *cinfo,
-				     u32 ds_commit_idx);
-	void (*clear_request_commit) (struct nfs_page *req,
-				      struct nfs_commit_info *cinfo);
-	int (*scan_commit_lists) (struct nfs_commit_info *cinfo,
-				  int max);
-	void (*recover_commit_reqs) (struct list_head *list,
-				     struct nfs_commit_info *cinfo);
-	struct nfs_page * (*search_commit_reqs)(struct nfs_commit_info *cinfo,
-						struct page *page);
-	int (*commit_pagelist)(struct inode *inode,
-			       struct list_head *mds_pages,
-			       int how,
-			       struct nfs_commit_info *cinfo);
 
 	int (*sync)(struct inode *inode, bool datasync);
 
@@ -186,6 +172,29 @@ struct pnfs_layoutdriver_type {
 	int (*prepare_layoutstats) (struct nfs42_layoutstat_args *args);
 };
 
+struct pnfs_commit_ops {
+	void (*setup_ds_info)(struct pnfs_ds_commit_info *,
+			      struct pnfs_layout_segment *);
+	void (*release_ds_info)(struct pnfs_ds_commit_info *,
+				struct inode *inode);
+	int (*commit_pagelist)(struct inode *inode,
+			       struct list_head *mds_pages,
+			       int how,
+			       struct nfs_commit_info *cinfo);
+	void (*mark_request_commit) (struct nfs_page *req,
+				     struct pnfs_layout_segment *lseg,
+				     struct nfs_commit_info *cinfo,
+				     u32 ds_commit_idx);
+	void (*clear_request_commit) (struct nfs_page *req,
+				      struct nfs_commit_info *cinfo);
+	int (*scan_commit_lists) (struct nfs_commit_info *cinfo,
+				  int max);
+	void (*recover_commit_reqs) (struct list_head *list,
+				     struct nfs_commit_info *cinfo);
+	struct nfs_page * (*search_commit_reqs)(struct nfs_commit_info *cinfo,
+						struct page *page);
+};
+
 struct pnfs_layout_hdr {
 	refcount_t		plh_refcount;
 	atomic_t		plh_outstanding; /* number of RPCs out */
@@ -203,6 +212,7 @@ struct pnfs_layout_hdr {
 	loff_t			plh_lwb; /* last write byte for layoutcommit */
 	const struct cred	*plh_lc_cred; /* layoutcommit cred */
 	struct inode		*plh_inode;
+	struct rcu_head		plh_rcu;
 };
 
 struct pnfs_device {
@@ -242,6 +252,7 @@ void pnfs_put_lseg(struct pnfs_layout_segment *lseg);
 void set_pnfs_layoutdriver(struct nfs_server *, const struct nfs_fh *, struct nfs_fsinfo *);
 void unset_pnfs_layoutdriver(struct nfs_server *);
 void pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio);
+void pnfs_generic_pg_check_range(struct nfs_pageio_descriptor *pgio, struct nfs_page *req);
 void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *, struct nfs_page *);
 int pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc);
 void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio,
@@ -267,6 +278,7 @@ bool nfs4_layout_refresh_old_stateid(nfs4_stateid *dst,
 void pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo);
 void pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo,
 			     const nfs4_stateid *new,
+			     const struct cred *cred,
 			     bool update_barrier);
 int pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo,
 				struct list_head *tmp_list,
@@ -326,6 +338,9 @@ int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *);
 struct nfs4_threshold *pnfs_mdsthreshold_alloc(void);
 void pnfs_error_mark_layout_for_return(struct inode *inode,
 				       struct pnfs_layout_segment *lseg);
+void pnfs_layout_return_unused_byclid(struct nfs_client *clp,
+				      enum pnfs_iomode iomode);
+
 /* nfs4_deviceid_flags */
 enum {
 	NFS_DEVICEID_INVALID = 0,       /* set when MDS clientid recalled */
@@ -360,6 +375,16 @@ bool nfs4_test_deviceid_unavailable(struct nfs4_deviceid_node *node);
 void nfs4_deviceid_purge_client(const struct nfs_client *);
 
 /* pnfs_nfs.c */
+struct pnfs_commit_array *pnfs_alloc_commit_array(size_t n, gfp_t gfp_flags);
+void pnfs_free_commit_array(struct pnfs_commit_array *p);
+struct pnfs_commit_array *pnfs_add_commit_array(struct pnfs_ds_commit_info *,
+						struct pnfs_commit_array *,
+						struct pnfs_layout_segment *);
+
+void pnfs_generic_ds_cinfo_release_lseg(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg);
+void pnfs_generic_ds_cinfo_destroy(struct pnfs_ds_commit_info *fl_cinfo);
+
 void pnfs_generic_clear_request_commit(struct nfs_page *req,
 				       struct nfs_commit_info *cinfo);
 void pnfs_generic_commit_release(void *calldata);
@@ -367,6 +392,8 @@ void pnfs_generic_prepare_to_resend_writes(struct nfs_commit_data *data);
 void pnfs_generic_rw_release(void *data);
 void pnfs_generic_recover_commit_reqs(struct list_head *dst,
 				      struct nfs_commit_info *cinfo);
+struct nfs_page *pnfs_generic_search_commit_reqs(struct nfs_commit_info *cinfo,
+						 struct page *page);
 int pnfs_generic_commit_pagelist(struct inode *inode,
 				 struct list_head *mds_pages,
 				 int how,
@@ -438,9 +465,11 @@ static inline int
 pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how,
 		 struct nfs_commit_info *cinfo)
 {
-	if (cinfo->ds == NULL || cinfo->ds->ncommitting == 0)
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+
+	if (fl_cinfo == NULL || fl_cinfo->ncommitting == 0)
 		return PNFS_NOT_ATTEMPTED;
-	return NFS_SERVER(inode)->pnfs_curr_ld->commit_pagelist(inode, mds_pages, how, cinfo);
+	return fl_cinfo->ops->commit_pagelist(inode, mds_pages, how, cinfo);
 }
 
 static inline struct pnfs_ds_commit_info *
@@ -454,6 +483,28 @@ pnfs_get_ds_info(struct inode *inode)
 }
 
 static inline void
+pnfs_init_ds_commit_info_ops(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode)
+{
+	struct pnfs_ds_commit_info *inode_cinfo = pnfs_get_ds_info(inode);
+	if (inode_cinfo != NULL)
+		fl_cinfo->ops = inode_cinfo->ops;
+}
+
+static inline void
+pnfs_init_ds_commit_info(struct pnfs_ds_commit_info *fl_cinfo)
+{
+	INIT_LIST_HEAD(&fl_cinfo->commits);
+	fl_cinfo->ops = NULL;
+}
+
+static inline void
+pnfs_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode)
+{
+	if (fl_cinfo->ops != NULL && fl_cinfo->ops->release_ds_info != NULL)
+		fl_cinfo->ops->release_ds_info(fl_cinfo, inode);
+}
+
+static inline void
 pnfs_generic_mark_devid_invalid(struct nfs4_deviceid_node *node)
 {
 	set_bit(NFS_DEVICEID_INVALID, &node->flags);
@@ -463,24 +514,22 @@ static inline bool
 pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
 			 struct nfs_commit_info *cinfo, u32 ds_commit_idx)
 {
-	struct inode *inode = d_inode(nfs_req_openctx(req)->dentry);
-	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
 
-	if (lseg == NULL || ld->mark_request_commit == NULL)
+	if (!lseg || !fl_cinfo->ops->mark_request_commit)
 		return false;
-	ld->mark_request_commit(req, lseg, cinfo, ds_commit_idx);
+	fl_cinfo->ops->mark_request_commit(req, lseg, cinfo, ds_commit_idx);
 	return true;
 }
 
 static inline bool
 pnfs_clear_request_commit(struct nfs_page *req, struct nfs_commit_info *cinfo)
 {
-	struct inode *inode = d_inode(nfs_req_openctx(req)->dentry);
-	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
 
-	if (ld == NULL || ld->clear_request_commit == NULL)
+	if (!fl_cinfo || !fl_cinfo->ops || !fl_cinfo->ops->clear_request_commit)
 		return false;
-	ld->clear_request_commit(req, cinfo);
+	fl_cinfo->ops->clear_request_commit(req, cinfo);
 	return true;
 }
 
@@ -488,21 +537,31 @@ static inline int
 pnfs_scan_commit_lists(struct inode *inode, struct nfs_commit_info *cinfo,
 		       int max)
 {
-	if (cinfo->ds == NULL || cinfo->ds->nwritten == 0)
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+
+	if (!fl_cinfo || fl_cinfo->nwritten == 0)
 		return 0;
-	else
-		return NFS_SERVER(inode)->pnfs_curr_ld->scan_commit_lists(cinfo, max);
+	return fl_cinfo->ops->scan_commit_lists(cinfo, max);
+}
+
+static inline void
+pnfs_recover_commit_reqs(struct list_head *head, struct nfs_commit_info *cinfo)
+{
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+
+	if (fl_cinfo && fl_cinfo->nwritten != 0)
+		fl_cinfo->ops->recover_commit_reqs(head, cinfo);
 }
 
 static inline struct nfs_page *
 pnfs_search_commit_reqs(struct inode *inode, struct nfs_commit_info *cinfo,
 			struct page *page)
 {
-	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
 
-	if (ld == NULL || ld->search_commit_reqs == NULL)
+	if (!fl_cinfo->ops || !fl_cinfo->ops->search_commit_reqs)
 		return NULL;
-	return ld->search_commit_reqs(cinfo, page);
+	return fl_cinfo->ops->search_commit_reqs(cinfo, page);
 }
 
 /* Should the pNFS client commit and return the layout upon a setattr */
@@ -750,6 +809,21 @@ pnfs_get_ds_info(struct inode *inode)
 	return NULL;
 }
 
+static inline void
+pnfs_init_ds_commit_info_ops(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode)
+{
+}
+
+static inline void
+pnfs_init_ds_commit_info(struct pnfs_ds_commit_info *fl_cinfo)
+{
+}
+
+static inline void
+pnfs_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode)
+{
+}
+
 static inline bool
 pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
 			 struct nfs_commit_info *cinfo, u32 ds_commit_idx)
@@ -770,6 +844,11 @@ pnfs_scan_commit_lists(struct inode *inode, struct nfs_commit_info *cinfo,
 	return 0;
 }
 
+static inline void
+pnfs_recover_commit_reqs(struct list_head *head, struct nfs_commit_info *cinfo)
+{
+}
+
 static inline struct nfs_page *
 pnfs_search_commit_reqs(struct inode *inode, struct nfs_commit_info *cinfo,
 			struct page *page)
diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c
index 82af4809b869..679767ac258d 100644
--- a/fs/nfs/pnfs_nfs.c
+++ b/fs/nfs/pnfs_nfs.c
@@ -31,12 +31,11 @@ EXPORT_SYMBOL_GPL(pnfs_generic_rw_release);
 /* Fake up some data that will cause nfs_commit_release to retry the writes. */
 void pnfs_generic_prepare_to_resend_writes(struct nfs_commit_data *data)
 {
-	struct nfs_page *first = nfs_list_entry(data->pages.next);
+	struct nfs_writeverf *verf = data->res.verf;
 
 	data->task.tk_status = 0;
-	memcpy(&data->verf.verifier, &first->wb_verf,
-	       sizeof(data->verf.verifier));
-	data->verf.verifier.data[0]++; /* ensure verifier mismatch */
+	memset(&verf->verifier, 0, sizeof(verf->verifier));
+	verf->committed = NFS_UNSTABLE;
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_prepare_to_resend_writes);
 
@@ -60,6 +59,17 @@ void pnfs_generic_commit_release(void *calldata)
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_commit_release);
 
+static struct pnfs_layout_segment *
+pnfs_free_bucket_lseg(struct pnfs_commit_bucket *bucket)
+{
+	if (list_empty(&bucket->committing) && list_empty(&bucket->written)) {
+		struct pnfs_layout_segment *freeme = bucket->lseg;
+		bucket->lseg = NULL;
+		return freeme;
+	}
+	return NULL;
+}
+
 /* The generic layer is about to remove the req from the commit list.
  * If this will make the bucket empty, it will need to put the lseg reference.
  * Note this must be called holding nfsi->commit_mutex
@@ -79,8 +89,7 @@ pnfs_generic_clear_request_commit(struct nfs_page *req,
 		bucket = list_first_entry(&req->wb_list,
 					  struct pnfs_commit_bucket,
 					  written);
-		freeme = bucket->wlseg;
-		bucket->wlseg = NULL;
+		freeme = pnfs_free_bucket_lseg(bucket);
 	}
 out:
 	nfs_request_remove_commit_list(req, cinfo);
@@ -88,10 +97,154 @@ out:
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_clear_request_commit);
 
+struct pnfs_commit_array *
+pnfs_alloc_commit_array(size_t n, gfp_t gfp_flags)
+{
+	struct pnfs_commit_array *p;
+	struct pnfs_commit_bucket *b;
+
+	p = kmalloc(struct_size(p, buckets, n), gfp_flags);
+	if (!p)
+		return NULL;
+	p->nbuckets = n;
+	INIT_LIST_HEAD(&p->cinfo_list);
+	INIT_LIST_HEAD(&p->lseg_list);
+	p->lseg = NULL;
+	for (b = &p->buckets[0]; n != 0; b++, n--) {
+		INIT_LIST_HEAD(&b->written);
+		INIT_LIST_HEAD(&b->committing);
+		b->lseg = NULL;
+		b->direct_verf.committed = NFS_INVALID_STABLE_HOW;
+	}
+	return p;
+}
+EXPORT_SYMBOL_GPL(pnfs_alloc_commit_array);
+
+void
+pnfs_free_commit_array(struct pnfs_commit_array *p)
+{
+	kfree_rcu(p, rcu);
+}
+EXPORT_SYMBOL_GPL(pnfs_free_commit_array);
+
+static struct pnfs_commit_array *
+pnfs_find_commit_array_by_lseg(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg)
+{
+	struct pnfs_commit_array *array;
+
+	list_for_each_entry_rcu(array, &fl_cinfo->commits, cinfo_list) {
+		if (array->lseg == lseg)
+			return array;
+	}
+	return NULL;
+}
+
+struct pnfs_commit_array *
+pnfs_add_commit_array(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_commit_array *new,
+		struct pnfs_layout_segment *lseg)
+{
+	struct pnfs_commit_array *array;
+
+	array = pnfs_find_commit_array_by_lseg(fl_cinfo, lseg);
+	if (array)
+		return array;
+	new->lseg = lseg;
+	refcount_set(&new->refcount, 1);
+	list_add_rcu(&new->cinfo_list, &fl_cinfo->commits);
+	list_add(&new->lseg_list, &lseg->pls_commits);
+	return new;
+}
+EXPORT_SYMBOL_GPL(pnfs_add_commit_array);
+
+static struct pnfs_commit_array *
+pnfs_lookup_commit_array(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg)
+{
+	struct pnfs_commit_array *array;
+
+	rcu_read_lock();
+	array = pnfs_find_commit_array_by_lseg(fl_cinfo, lseg);
+	if (!array) {
+		rcu_read_unlock();
+		fl_cinfo->ops->setup_ds_info(fl_cinfo, lseg);
+		rcu_read_lock();
+		array = pnfs_find_commit_array_by_lseg(fl_cinfo, lseg);
+	}
+	rcu_read_unlock();
+	return array;
+}
+
+static void
+pnfs_release_commit_array_locked(struct pnfs_commit_array *array)
+{
+	list_del_rcu(&array->cinfo_list);
+	list_del(&array->lseg_list);
+	pnfs_free_commit_array(array);
+}
+
+static void
+pnfs_put_commit_array_locked(struct pnfs_commit_array *array)
+{
+	if (refcount_dec_and_test(&array->refcount))
+		pnfs_release_commit_array_locked(array);
+}
+
+static void
+pnfs_put_commit_array(struct pnfs_commit_array *array, struct inode *inode)
+{
+	if (refcount_dec_and_lock(&array->refcount, &inode->i_lock)) {
+		pnfs_release_commit_array_locked(array);
+		spin_unlock(&inode->i_lock);
+	}
+}
+
+static struct pnfs_commit_array *
+pnfs_get_commit_array(struct pnfs_commit_array *array)
+{
+	if (refcount_inc_not_zero(&array->refcount))
+		return array;
+	return NULL;
+}
+
+static void
+pnfs_remove_and_free_commit_array(struct pnfs_commit_array *array)
+{
+	array->lseg = NULL;
+	list_del_init(&array->lseg_list);
+	pnfs_put_commit_array_locked(array);
+}
+
+void
+pnfs_generic_ds_cinfo_release_lseg(struct pnfs_ds_commit_info *fl_cinfo,
+		struct pnfs_layout_segment *lseg)
+{
+	struct pnfs_commit_array *array, *tmp;
+
+	list_for_each_entry_safe(array, tmp, &lseg->pls_commits, lseg_list)
+		pnfs_remove_and_free_commit_array(array);
+}
+EXPORT_SYMBOL_GPL(pnfs_generic_ds_cinfo_release_lseg);
+
+void
+pnfs_generic_ds_cinfo_destroy(struct pnfs_ds_commit_info *fl_cinfo)
+{
+	struct pnfs_commit_array *array, *tmp;
+
+	list_for_each_entry_safe(array, tmp, &fl_cinfo->commits, cinfo_list)
+		pnfs_remove_and_free_commit_array(array);
+}
+EXPORT_SYMBOL_GPL(pnfs_generic_ds_cinfo_destroy);
+
+/*
+ * Locks the nfs_page requests for commit and moves them to
+ * @bucket->committing.
+ */
 static int
-pnfs_generic_scan_ds_commit_list(struct pnfs_commit_bucket *bucket,
-				 struct nfs_commit_info *cinfo,
-				 int max)
+pnfs_bucket_scan_ds_commit_list(struct pnfs_commit_bucket *bucket,
+				struct nfs_commit_info *cinfo,
+				int max)
 {
 	struct list_head *src = &bucket->written;
 	struct list_head *dst = &bucket->committing;
@@ -102,158 +255,254 @@ pnfs_generic_scan_ds_commit_list(struct pnfs_commit_bucket *bucket,
 	if (ret) {
 		cinfo->ds->nwritten -= ret;
 		cinfo->ds->ncommitting += ret;
-		if (bucket->clseg == NULL)
-			bucket->clseg = pnfs_get_lseg(bucket->wlseg);
-		if (list_empty(src)) {
-			pnfs_put_lseg(bucket->wlseg);
-			bucket->wlseg = NULL;
-		}
 	}
 	return ret;
 }
 
+static int pnfs_bucket_scan_array(struct nfs_commit_info *cinfo,
+				  struct pnfs_commit_bucket *buckets,
+				  unsigned int nbuckets,
+				  int max)
+{
+	unsigned int i;
+	int rv = 0, cnt;
+
+	for (i = 0; i < nbuckets && max != 0; i++) {
+		cnt = pnfs_bucket_scan_ds_commit_list(&buckets[i], cinfo, max);
+		rv += cnt;
+		max -= cnt;
+	}
+	return rv;
+}
+
 /* Move reqs from written to committing lists, returning count
  * of number moved.
  */
-int pnfs_generic_scan_commit_lists(struct nfs_commit_info *cinfo,
-				   int max)
+int pnfs_generic_scan_commit_lists(struct nfs_commit_info *cinfo, int max)
 {
-	int i, rv = 0, cnt;
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+	struct pnfs_commit_array *array;
+	int rv = 0, cnt;
 
-	lockdep_assert_held(&NFS_I(cinfo->inode)->commit_mutex);
-	for (i = 0; i < cinfo->ds->nbuckets && max != 0; i++) {
-		cnt = pnfs_generic_scan_ds_commit_list(&cinfo->ds->buckets[i],
-						       cinfo, max);
-		max -= cnt;
+	rcu_read_lock();
+	list_for_each_entry_rcu(array, &fl_cinfo->commits, cinfo_list) {
+		if (!array->lseg || !pnfs_get_commit_array(array))
+			continue;
+		rcu_read_unlock();
+		cnt = pnfs_bucket_scan_array(cinfo, array->buckets,
+				array->nbuckets, max);
+		rcu_read_lock();
+		pnfs_put_commit_array(array, cinfo->inode);
 		rv += cnt;
+		max -= cnt;
+		if (!max)
+			break;
 	}
+	rcu_read_unlock();
 	return rv;
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_scan_commit_lists);
 
-/* Pull everything off the committing lists and dump into @dst.  */
-void pnfs_generic_recover_commit_reqs(struct list_head *dst,
-				      struct nfs_commit_info *cinfo)
+static unsigned int
+pnfs_bucket_recover_commit_reqs(struct list_head *dst,
+			        struct pnfs_commit_bucket *buckets,
+				unsigned int nbuckets,
+				struct nfs_commit_info *cinfo)
 {
 	struct pnfs_commit_bucket *b;
 	struct pnfs_layout_segment *freeme;
-	int nwritten;
-	int i;
+	unsigned int nwritten, ret = 0;
+	unsigned int i;
 
-	lockdep_assert_held(&NFS_I(cinfo->inode)->commit_mutex);
 restart:
-	for (i = 0, b = cinfo->ds->buckets; i < cinfo->ds->nbuckets; i++, b++) {
+	for (i = 0, b = buckets; i < nbuckets; i++, b++) {
 		nwritten = nfs_scan_commit_list(&b->written, dst, cinfo, 0);
 		if (!nwritten)
 			continue;
-		cinfo->ds->nwritten -= nwritten;
-		if (list_empty(&b->written)) {
-			freeme = b->wlseg;
-			b->wlseg = NULL;
+		ret += nwritten;
+		freeme = pnfs_free_bucket_lseg(b);
+		if (freeme) {
 			pnfs_put_lseg(freeme);
 			goto restart;
 		}
 	}
+	return ret;
+}
+
+/* Pull everything off the committing lists and dump into @dst.  */
+void pnfs_generic_recover_commit_reqs(struct list_head *dst,
+				      struct nfs_commit_info *cinfo)
+{
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+	struct pnfs_commit_array *array;
+	unsigned int nwritten;
+
+	lockdep_assert_held(&NFS_I(cinfo->inode)->commit_mutex);
+	rcu_read_lock();
+	list_for_each_entry_rcu(array, &fl_cinfo->commits, cinfo_list) {
+		if (!array->lseg || !pnfs_get_commit_array(array))
+			continue;
+		rcu_read_unlock();
+		nwritten = pnfs_bucket_recover_commit_reqs(dst,
+							   array->buckets,
+							   array->nbuckets,
+							   cinfo);
+		rcu_read_lock();
+		pnfs_put_commit_array(array, cinfo->inode);
+		fl_cinfo->nwritten -= nwritten;
+	}
+	rcu_read_unlock();
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_recover_commit_reqs);
 
-static void pnfs_generic_retry_commit(struct nfs_commit_info *cinfo, int idx)
+static struct nfs_page *
+pnfs_bucket_search_commit_reqs(struct pnfs_commit_bucket *buckets,
+		unsigned int nbuckets, struct page *page)
+{
+	struct nfs_page *req;
+	struct pnfs_commit_bucket *b;
+	unsigned int i;
+
+	/* Linearly search the commit lists for each bucket until a matching
+	 * request is found */
+	for (i = 0, b = buckets; i < nbuckets; i++, b++) {
+		list_for_each_entry(req, &b->written, wb_list) {
+			if (req->wb_page == page)
+				return req->wb_head;
+		}
+		list_for_each_entry(req, &b->committing, wb_list) {
+			if (req->wb_page == page)
+				return req->wb_head;
+		}
+	}
+	return NULL;
+}
+
+/* pnfs_generic_search_commit_reqs - Search lists in @cinfo for the head reqest
+ *				   for @page
+ * @cinfo - commit info for current inode
+ * @page - page to search for matching head request
+ *
+ * Returns a the head request if one is found, otherwise returns NULL.
+ */
+struct nfs_page *
+pnfs_generic_search_commit_reqs(struct nfs_commit_info *cinfo, struct page *page)
 {
 	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
+	struct pnfs_commit_array *array;
+	struct nfs_page *req;
+
+	list_for_each_entry(array, &fl_cinfo->commits, cinfo_list) {
+		req = pnfs_bucket_search_commit_reqs(array->buckets,
+				array->nbuckets, page);
+		if (req)
+			return req;
+	}
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(pnfs_generic_search_commit_reqs);
+
+static struct pnfs_layout_segment *
+pnfs_bucket_get_committing(struct list_head *head,
+			   struct pnfs_commit_bucket *bucket,
+			   struct nfs_commit_info *cinfo)
+{
+	struct list_head *pos;
+
+	list_for_each(pos, &bucket->committing)
+		cinfo->ds->ncommitting--;
+	list_splice_init(&bucket->committing, head);
+	return pnfs_free_bucket_lseg(bucket);
+}
+
+static struct nfs_commit_data *
+pnfs_bucket_fetch_commitdata(struct pnfs_commit_bucket *bucket,
+			     struct nfs_commit_info *cinfo)
+{
+	struct nfs_commit_data *data = nfs_commitdata_alloc(false);
+
+	if (!data)
+		return NULL;
+	data->lseg = pnfs_bucket_get_committing(&data->pages, bucket, cinfo);
+	if (!data->lseg)
+		data->lseg = pnfs_get_lseg(bucket->lseg);
+	return data;
+}
+
+static void pnfs_generic_retry_commit(struct pnfs_commit_bucket *buckets,
+				      unsigned int nbuckets,
+				      struct nfs_commit_info *cinfo,
+				      unsigned int idx)
+{
 	struct pnfs_commit_bucket *bucket;
 	struct pnfs_layout_segment *freeme;
-	struct list_head *pos;
 	LIST_HEAD(pages);
-	int i;
 
-	mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
-	for (i = idx; i < fl_cinfo->nbuckets; i++) {
-		bucket = &fl_cinfo->buckets[i];
+	for (bucket = buckets; idx < nbuckets; bucket++, idx++) {
 		if (list_empty(&bucket->committing))
 			continue;
-		freeme = bucket->clseg;
-		bucket->clseg = NULL;
-		list_for_each(pos, &bucket->committing)
-			cinfo->ds->ncommitting--;
-		list_splice_init(&bucket->committing, &pages);
+		mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
+		freeme = pnfs_bucket_get_committing(&pages, bucket, cinfo);
 		mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
-		nfs_retry_commit(&pages, freeme, cinfo, i);
+		nfs_retry_commit(&pages, freeme, cinfo, idx);
 		pnfs_put_lseg(freeme);
-		mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
 	}
-	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
 }
 
 static unsigned int
-pnfs_generic_alloc_ds_commits(struct nfs_commit_info *cinfo,
-			      struct list_head *list)
+pnfs_bucket_alloc_ds_commits(struct list_head *list,
+			     struct pnfs_commit_bucket *buckets,
+			     unsigned int nbuckets,
+			     struct nfs_commit_info *cinfo)
 {
-	struct pnfs_ds_commit_info *fl_cinfo;
 	struct pnfs_commit_bucket *bucket;
 	struct nfs_commit_data *data;
-	int i;
+	unsigned int i;
 	unsigned int nreq = 0;
 
-	fl_cinfo = cinfo->ds;
-	bucket = fl_cinfo->buckets;
-	for (i = 0; i < fl_cinfo->nbuckets; i++, bucket++) {
+	for (i = 0, bucket = buckets; i < nbuckets; i++, bucket++) {
 		if (list_empty(&bucket->committing))
 			continue;
-		data = nfs_commitdata_alloc(false);
-		if (!data)
-			break;
-		data->ds_commit_index = i;
-		list_add(&data->pages, list);
-		nreq++;
+		mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
+		if (!list_empty(&bucket->committing)) {
+			data = pnfs_bucket_fetch_commitdata(bucket, cinfo);
+			if (!data)
+				goto out_error;
+			data->ds_commit_index = i;
+			list_add_tail(&data->list, list);
+			atomic_inc(&cinfo->mds->rpcs_out);
+			nreq++;
+		}
+		mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
 	}
-
+	return nreq;
+out_error:
+	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
 	/* Clean up on error */
-	pnfs_generic_retry_commit(cinfo, i);
+	pnfs_generic_retry_commit(buckets, nbuckets, cinfo, i);
 	return nreq;
 }
 
-static inline
-void pnfs_fetch_commit_bucket_list(struct list_head *pages,
-		struct nfs_commit_data *data,
-		struct nfs_commit_info *cinfo)
+static unsigned int
+pnfs_alloc_ds_commits_list(struct list_head *list,
+			   struct pnfs_ds_commit_info *fl_cinfo,
+			   struct nfs_commit_info *cinfo)
 {
-	struct pnfs_commit_bucket *bucket;
-	struct list_head *pos;
-
-	bucket = &cinfo->ds->buckets[data->ds_commit_index];
-	mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
-	list_for_each(pos, &bucket->committing)
-		cinfo->ds->ncommitting--;
-	list_splice_init(&bucket->committing, pages);
-	data->lseg = bucket->clseg;
-	bucket->clseg = NULL;
-	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
-
-}
+	struct pnfs_commit_array *array;
+	unsigned int ret = 0;
 
-/* Helper function for pnfs_generic_commit_pagelist to catch an empty
- * page list. This can happen when two commits race.
- *
- * This must be called instead of nfs_init_commit - call one or the other, but
- * not both!
- */
-static bool
-pnfs_generic_commit_cancel_empty_pagelist(struct list_head *pages,
-					  struct nfs_commit_data *data,
-					  struct nfs_commit_info *cinfo)
-{
-	if (list_empty(pages)) {
-		if (atomic_dec_and_test(&cinfo->mds->rpcs_out))
-			wake_up_var(&cinfo->mds->rpcs_out);
-		/* don't call nfs_commitdata_release - it tries to put
-		 * the open_context which is not acquired until nfs_init_commit
-		 * which has not been called on @data */
-		WARN_ON_ONCE(data->context);
-		nfs_commit_free(data);
-		return true;
+	rcu_read_lock();
+	list_for_each_entry_rcu(array, &fl_cinfo->commits, cinfo_list) {
+		if (!array->lseg || !pnfs_get_commit_array(array))
+			continue;
+		rcu_read_unlock();
+		ret += pnfs_bucket_alloc_ds_commits(list, array->buckets,
+				array->nbuckets, cinfo);
+		rcu_read_lock();
+		pnfs_put_commit_array(array, cinfo->inode);
 	}
-
-	return false;
+	rcu_read_unlock();
+	return ret;
 }
 
 /* This follows nfs_commit_list pretty closely */
@@ -263,6 +512,7 @@ pnfs_generic_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
 			     int (*initiate_commit)(struct nfs_commit_data *data,
 						    int how))
 {
+	struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds;
 	struct nfs_commit_data *data, *tmp;
 	LIST_HEAD(list);
 	unsigned int nreq = 0;
@@ -270,40 +520,26 @@ pnfs_generic_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
 	if (!list_empty(mds_pages)) {
 		data = nfs_commitdata_alloc(true);
 		data->ds_commit_index = -1;
-		list_add(&data->pages, &list);
+		list_splice_init(mds_pages, &data->pages);
+		list_add_tail(&data->list, &list);
+		atomic_inc(&cinfo->mds->rpcs_out);
 		nreq++;
 	}
 
-	nreq += pnfs_generic_alloc_ds_commits(cinfo, &list);
-
+	nreq += pnfs_alloc_ds_commits_list(&list, fl_cinfo, cinfo);
 	if (nreq == 0)
 		goto out;
 
-	atomic_add(nreq, &cinfo->mds->rpcs_out);
-
-	list_for_each_entry_safe(data, tmp, &list, pages) {
-		list_del_init(&data->pages);
+	list_for_each_entry_safe(data, tmp, &list, list) {
+		list_del(&data->list);
 		if (data->ds_commit_index < 0) {
-			/* another commit raced with us */
-			if (pnfs_generic_commit_cancel_empty_pagelist(mds_pages,
-				data, cinfo))
-				continue;
-
-			nfs_init_commit(data, mds_pages, NULL, cinfo);
+			nfs_init_commit(data, NULL, NULL, cinfo);
 			nfs_initiate_commit(NFS_CLIENT(inode), data,
 					    NFS_PROTO(data->inode),
-					    data->mds_ops, how, 0);
+					    data->mds_ops, how,
+					    RPC_TASK_CRED_NOREF);
 		} else {
-			LIST_HEAD(pages);
-
-			pnfs_fetch_commit_bucket_list(&pages, data, cinfo);
-
-			/* another commit raced with us */
-			if (pnfs_generic_commit_cancel_empty_pagelist(&pages,
-				data, cinfo))
-				continue;
-
-			nfs_init_commit(data, &pages, data->lseg, cinfo);
+			nfs_init_commit(data, NULL, data->lseg, cinfo);
 			initiate_commit(data, how);
 		}
 	}
@@ -931,32 +1167,33 @@ pnfs_layout_mark_request_commit(struct nfs_page *req,
 				u32 ds_commit_idx)
 {
 	struct list_head *list;
-	struct pnfs_commit_bucket *buckets;
+	struct pnfs_commit_array *array;
+	struct pnfs_commit_bucket *bucket;
 
 	mutex_lock(&NFS_I(cinfo->inode)->commit_mutex);
-	buckets = cinfo->ds->buckets;
-	list = &buckets[ds_commit_idx].written;
-	if (list_empty(list)) {
-		if (!pnfs_is_valid_lseg(lseg)) {
-			mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
-			cinfo->completion_ops->resched_write(cinfo, req);
-			return;
-		}
-		/* Non-empty buckets hold a reference on the lseg.  That ref
-		 * is normally transferred to the COMMIT call and released
-		 * there.  It could also be released if the last req is pulled
-		 * off due to a rewrite, in which case it will be done in
-		 * pnfs_common_clear_request_commit
-		 */
-		WARN_ON_ONCE(buckets[ds_commit_idx].wlseg != NULL);
-		buckets[ds_commit_idx].wlseg = pnfs_get_lseg(lseg);
-	}
+	array = pnfs_lookup_commit_array(cinfo->ds, lseg);
+	if (!array || !pnfs_is_valid_lseg(lseg))
+		goto out_resched;
+	bucket = &array->buckets[ds_commit_idx];
+	list = &bucket->written;
+	/* Non-empty buckets hold a reference on the lseg.  That ref
+	 * is normally transferred to the COMMIT call and released
+	 * there.  It could also be released if the last req is pulled
+	 * off due to a rewrite, in which case it will be done in
+	 * pnfs_common_clear_request_commit
+	 */
+	if (!bucket->lseg)
+		bucket->lseg = pnfs_get_lseg(lseg);
 	set_bit(PG_COMMIT_TO_DS, &req->wb_flags);
 	cinfo->ds->nwritten++;
 
 	nfs_request_add_commit_list_locked(req, list, cinfo);
 	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
 	nfs_mark_page_unstable(req->wb_page, cinfo);
+	return;
+out_resched:
+	mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex);
+	cinfo->completion_ops->resched_write(cinfo, req);
 }
 EXPORT_SYMBOL_GPL(pnfs_layout_mark_request_commit);
 
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 0f7288b94633..15c865cc837f 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -108,10 +108,15 @@ nfs_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_resp	= fattr,
 	};
 	int	status;
+	unsigned short task_flags = 0;
+
+	/* Is this is an attribute revalidation, subject to softreval? */
+	if (inode && (server->flags & NFS_MOUNT_SOFTREVAL))
+		task_flags |= RPC_TASK_TIMEOUT;
 
 	dprintk("NFS call  getattr\n");
 	nfs_fattr_init(fattr);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = rpc_call_sync(server->client, &msg, task_flags);
 	dprintk("NFS reply getattr: %d\n", status);
 	return status;
 }
@@ -147,14 +152,14 @@ nfs_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
 }
 
 static int
-nfs_proc_lookup(struct inode *dir, const struct qstr *name,
+nfs_proc_lookup(struct inode *dir, struct dentry *dentry,
 		struct nfs_fh *fhandle, struct nfs_fattr *fattr,
 		struct nfs4_label *label)
 {
 	struct nfs_diropargs	arg = {
 		.fh		= NFS_FH(dir),
-		.name		= name->name,
-		.len		= name->len
+		.name		= dentry->d_name.name,
+		.len		= dentry->d_name.len
 	};
 	struct nfs_diropok	res = {
 		.fh		= fhandle,
@@ -166,10 +171,15 @@ nfs_proc_lookup(struct inode *dir, const struct qstr *name,
 		.rpc_resp	= &res,
 	};
 	int			status;
+	unsigned short task_flags = 0;
+
+	/* Is this is an attribute revalidation, subject to softreval? */
+	if (nfs_lookup_is_soft_revalidate(dentry))
+		task_flags |= RPC_TASK_TIMEOUT;
 
-	dprintk("NFS call  lookup %s\n", name->name);
+	dprintk("NFS call  lookup %pd2\n", dentry);
 	nfs_fattr_init(fattr);
-	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	status = rpc_call_sync(NFS_CLIENT(dir), &msg, task_flags);
 	dprintk("NFS reply lookup: %d\n", status);
 	return status;
 }
@@ -710,7 +720,7 @@ const struct nfs_rpc_ops nfs_v2_clientops = {
 	.file_ops	= &nfs_file_operations,
 	.getroot	= nfs_proc_get_root,
 	.submount	= nfs_submount,
-	.try_mount	= nfs_try_mount,
+	.try_get_tree	= nfs_try_get_tree,
 	.getattr	= nfs_proc_getattr,
 	.setattr	= nfs_proc_setattr,
 	.lookup		= nfs_proc_lookup,
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index cfe0b586eadd..13b22e898116 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -214,7 +214,7 @@ static void nfs_initiate_read(struct nfs_pgio_header *hdr,
 
 	task_setup_data->flags |= swap_flags;
 	rpc_ops->read_setup(hdr, msg);
-	trace_nfs_initiate_read(inode, hdr->io_start, hdr->good_bytes);
+	trace_nfs_initiate_read(hdr);
 }
 
 static void
@@ -247,11 +247,10 @@ static int nfs_readpage_done(struct rpc_task *task,
 		return status;
 
 	nfs_add_stats(inode, NFSIOS_SERVERREADBYTES, hdr->res.count);
-	trace_nfs_readpage_done(inode, task->tk_status,
-				hdr->args.offset, hdr->res.eof);
+	trace_nfs_readpage_done(task, hdr);
 
 	if (task->tk_status == -ESTALE) {
-		set_bit(NFS_INO_STALE, &NFS_I(inode)->flags);
+		nfs_set_inode_stale(inode);
 		nfs_mark_for_revalidate(inode);
 	}
 	return 0;
@@ -282,6 +281,8 @@ static void nfs_readpage_retry(struct rpc_task *task,
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
+	resp->count = 0;
+	resp->eof = 0;
 	rpc_restart_call_prepare(task);
 }
 
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index a84df7d63403..7a70287f21a2 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -69,250 +69,6 @@
 #include "nfs.h"
 
 #define NFSDBG_FACILITY		NFSDBG_VFS
-#define NFS_TEXT_DATA		1
-
-#if IS_ENABLED(CONFIG_NFS_V3)
-#define NFS_DEFAULT_VERSION 3
-#else
-#define NFS_DEFAULT_VERSION 2
-#endif
-
-#define NFS_MAX_CONNECTIONS 16
-
-enum {
-	/* Mount options that take no arguments */
-	Opt_soft, Opt_softerr, Opt_hard,
-	Opt_posix, Opt_noposix,
-	Opt_cto, Opt_nocto,
-	Opt_ac, Opt_noac,
-	Opt_lock, Opt_nolock,
-	Opt_udp, Opt_tcp, Opt_rdma,
-	Opt_acl, Opt_noacl,
-	Opt_rdirplus, Opt_nordirplus,
-	Opt_sharecache, Opt_nosharecache,
-	Opt_resvport, Opt_noresvport,
-	Opt_fscache, Opt_nofscache,
-	Opt_migration, Opt_nomigration,
-
-	/* Mount options that take integer arguments */
-	Opt_port,
-	Opt_rsize, Opt_wsize, Opt_bsize,
-	Opt_timeo, Opt_retrans,
-	Opt_acregmin, Opt_acregmax,
-	Opt_acdirmin, Opt_acdirmax,
-	Opt_actimeo,
-	Opt_namelen,
-	Opt_mountport,
-	Opt_mountvers,
-	Opt_minorversion,
-
-	/* Mount options that take string arguments */
-	Opt_nfsvers,
-	Opt_sec, Opt_proto, Opt_mountproto, Opt_mounthost,
-	Opt_addr, Opt_mountaddr, Opt_clientaddr,
-	Opt_nconnect,
-	Opt_lookupcache,
-	Opt_fscache_uniq,
-	Opt_local_lock,
-
-	/* Special mount options */
-	Opt_userspace, Opt_deprecated, Opt_sloppy,
-
-	Opt_err
-};
-
-static const match_table_t nfs_mount_option_tokens = {
-	{ Opt_userspace, "bg" },
-	{ Opt_userspace, "fg" },
-	{ Opt_userspace, "retry=%s" },
-
-	{ Opt_sloppy, "sloppy" },
-
-	{ Opt_soft, "soft" },
-	{ Opt_softerr, "softerr" },
-	{ Opt_hard, "hard" },
-	{ Opt_deprecated, "intr" },
-	{ Opt_deprecated, "nointr" },
-	{ Opt_posix, "posix" },
-	{ Opt_noposix, "noposix" },
-	{ Opt_cto, "cto" },
-	{ Opt_nocto, "nocto" },
-	{ Opt_ac, "ac" },
-	{ Opt_noac, "noac" },
-	{ Opt_lock, "lock" },
-	{ Opt_nolock, "nolock" },
-	{ Opt_udp, "udp" },
-	{ Opt_tcp, "tcp" },
-	{ Opt_rdma, "rdma" },
-	{ Opt_acl, "acl" },
-	{ Opt_noacl, "noacl" },
-	{ Opt_rdirplus, "rdirplus" },
-	{ Opt_nordirplus, "nordirplus" },
-	{ Opt_sharecache, "sharecache" },
-	{ Opt_nosharecache, "nosharecache" },
-	{ Opt_resvport, "resvport" },
-	{ Opt_noresvport, "noresvport" },
-	{ Opt_fscache, "fsc" },
-	{ Opt_nofscache, "nofsc" },
-	{ Opt_migration, "migration" },
-	{ Opt_nomigration, "nomigration" },
-
-	{ Opt_port, "port=%s" },
-	{ Opt_rsize, "rsize=%s" },
-	{ Opt_wsize, "wsize=%s" },
-	{ Opt_bsize, "bsize=%s" },
-	{ Opt_timeo, "timeo=%s" },
-	{ Opt_retrans, "retrans=%s" },
-	{ Opt_acregmin, "acregmin=%s" },
-	{ Opt_acregmax, "acregmax=%s" },
-	{ Opt_acdirmin, "acdirmin=%s" },
-	{ Opt_acdirmax, "acdirmax=%s" },
-	{ Opt_actimeo, "actimeo=%s" },
-	{ Opt_namelen, "namlen=%s" },
-	{ Opt_mountport, "mountport=%s" },
-	{ Opt_mountvers, "mountvers=%s" },
-	{ Opt_minorversion, "minorversion=%s" },
-
-	{ Opt_nfsvers, "nfsvers=%s" },
-	{ Opt_nfsvers, "vers=%s" },
-
-	{ Opt_sec, "sec=%s" },
-	{ Opt_proto, "proto=%s" },
-	{ Opt_mountproto, "mountproto=%s" },
-	{ Opt_addr, "addr=%s" },
-	{ Opt_clientaddr, "clientaddr=%s" },
-	{ Opt_mounthost, "mounthost=%s" },
-	{ Opt_mountaddr, "mountaddr=%s" },
-
-	{ Opt_nconnect, "nconnect=%s" },
-
-	{ Opt_lookupcache, "lookupcache=%s" },
-	{ Opt_fscache_uniq, "fsc=%s" },
-	{ Opt_local_lock, "local_lock=%s" },
-
-	/* The following needs to be listed after all other options */
-	{ Opt_nfsvers, "v%s" },
-
-	{ Opt_err, NULL }
-};
-
-enum {
-	Opt_xprt_udp, Opt_xprt_udp6, Opt_xprt_tcp, Opt_xprt_tcp6, Opt_xprt_rdma,
-	Opt_xprt_rdma6,
-
-	Opt_xprt_err
-};
-
-static const match_table_t nfs_xprt_protocol_tokens = {
-	{ Opt_xprt_udp, "udp" },
-	{ Opt_xprt_udp6, "udp6" },
-	{ Opt_xprt_tcp, "tcp" },
-	{ Opt_xprt_tcp6, "tcp6" },
-	{ Opt_xprt_rdma, "rdma" },
-	{ Opt_xprt_rdma6, "rdma6" },
-
-	{ Opt_xprt_err, NULL }
-};
-
-enum {
-	Opt_sec_none, Opt_sec_sys,
-	Opt_sec_krb5, Opt_sec_krb5i, Opt_sec_krb5p,
-	Opt_sec_lkey, Opt_sec_lkeyi, Opt_sec_lkeyp,
-	Opt_sec_spkm, Opt_sec_spkmi, Opt_sec_spkmp,
-
-	Opt_sec_err
-};
-
-static const match_table_t nfs_secflavor_tokens = {
-	{ Opt_sec_none, "none" },
-	{ Opt_sec_none, "null" },
-	{ Opt_sec_sys, "sys" },
-
-	{ Opt_sec_krb5, "krb5" },
-	{ Opt_sec_krb5i, "krb5i" },
-	{ Opt_sec_krb5p, "krb5p" },
-
-	{ Opt_sec_lkey, "lkey" },
-	{ Opt_sec_lkeyi, "lkeyi" },
-	{ Opt_sec_lkeyp, "lkeyp" },
-
-	{ Opt_sec_spkm, "spkm3" },
-	{ Opt_sec_spkmi, "spkm3i" },
-	{ Opt_sec_spkmp, "spkm3p" },
-
-	{ Opt_sec_err, NULL }
-};
-
-enum {
-	Opt_lookupcache_all, Opt_lookupcache_positive,
-	Opt_lookupcache_none,
-
-	Opt_lookupcache_err
-};
-
-static match_table_t nfs_lookupcache_tokens = {
-	{ Opt_lookupcache_all, "all" },
-	{ Opt_lookupcache_positive, "pos" },
-	{ Opt_lookupcache_positive, "positive" },
-	{ Opt_lookupcache_none, "none" },
-
-	{ Opt_lookupcache_err, NULL }
-};
-
-enum {
-	Opt_local_lock_all, Opt_local_lock_flock, Opt_local_lock_posix,
-	Opt_local_lock_none,
-
-	Opt_local_lock_err
-};
-
-static match_table_t nfs_local_lock_tokens = {
-	{ Opt_local_lock_all, "all" },
-	{ Opt_local_lock_flock, "flock" },
-	{ Opt_local_lock_posix, "posix" },
-	{ Opt_local_lock_none, "none" },
-
-	{ Opt_local_lock_err, NULL }
-};
-
-enum {
-	Opt_vers_2, Opt_vers_3, Opt_vers_4, Opt_vers_4_0,
-	Opt_vers_4_1, Opt_vers_4_2,
-
-	Opt_vers_err
-};
-
-static match_table_t nfs_vers_tokens = {
-	{ Opt_vers_2, "2" },
-	{ Opt_vers_3, "3" },
-	{ Opt_vers_4, "4" },
-	{ Opt_vers_4_0, "4.0" },
-	{ Opt_vers_4_1, "4.1" },
-	{ Opt_vers_4_2, "4.2" },
-
-	{ Opt_vers_err, NULL }
-};
-
-static struct dentry *nfs_xdev_mount(struct file_system_type *fs_type,
-		int flags, const char *dev_name, void *raw_data);
-
-struct file_system_type nfs_fs_type = {
-	.owner		= THIS_MODULE,
-	.name		= "nfs",
-	.mount		= nfs_fs_mount,
-	.kill_sb	= nfs_kill_super,
-	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
-};
-MODULE_ALIAS_FS("nfs");
-EXPORT_SYMBOL_GPL(nfs_fs_type);
-
-struct file_system_type nfs_xdev_fs_type = {
-	.owner		= THIS_MODULE,
-	.name		= "nfs",
-	.mount		= nfs_xdev_mount,
-	.kill_sb	= nfs_kill_super,
-	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
-};
 
 const struct super_operations nfs_sops = {
 	.alloc_inode	= nfs_alloc_inode,
@@ -326,26 +82,10 @@ const struct super_operations nfs_sops = {
 	.show_devname	= nfs_show_devname,
 	.show_path	= nfs_show_path,
 	.show_stats	= nfs_show_stats,
-	.remount_fs	= nfs_remount,
 };
 EXPORT_SYMBOL_GPL(nfs_sops);
 
 #if IS_ENABLED(CONFIG_NFS_V4)
-static void nfs4_validate_mount_flags(struct nfs_parsed_mount_data *);
-static int nfs4_validate_mount_data(void *options,
-	struct nfs_parsed_mount_data *args, const char *dev_name);
-
-struct file_system_type nfs4_fs_type = {
-	.owner		= THIS_MODULE,
-	.name		= "nfs4",
-	.mount		= nfs_fs_mount,
-	.kill_sb	= nfs_kill_super,
-	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_BINARY_MOUNTDATA,
-};
-MODULE_ALIAS_FS("nfs4");
-MODULE_ALIAS("nfs4");
-EXPORT_SYMBOL_GPL(nfs4_fs_type);
-
 static int __init register_nfs4_fs(void)
 {
 	return register_filesystem(&nfs4_fs_type);
@@ -436,6 +176,41 @@ void nfs_sb_deactive(struct super_block *sb)
 }
 EXPORT_SYMBOL_GPL(nfs_sb_deactive);
 
+static int __nfs_list_for_each_server(struct list_head *head,
+		int (*fn)(struct nfs_server *, void *),
+		void *data)
+{
+	struct nfs_server *server, *last = NULL;
+	int ret = 0;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, head, client_link) {
+		if (!(server->super && nfs_sb_active(server->super)))
+			continue;
+		rcu_read_unlock();
+		if (last)
+			nfs_sb_deactive(last->super);
+		last = server;
+		ret = fn(server, data);
+		if (ret)
+			goto out;
+		rcu_read_lock();
+	}
+	rcu_read_unlock();
+out:
+	if (last)
+		nfs_sb_deactive(last->super);
+	return ret;
+}
+
+int nfs_client_for_each_server(struct nfs_client *clp,
+		int (*fn)(struct nfs_server *, void *),
+		void *data)
+{
+	return __nfs_list_for_each_server(&clp->cl_superblocks, fn, data);
+}
+EXPORT_SYMBOL_GPL(nfs_client_for_each_server);
+
 /*
  * Deliver file system statistics to userspace
  */
@@ -635,6 +410,7 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 	} nfs_info[] = {
 		{ NFS_MOUNT_SOFT, ",soft", "" },
 		{ NFS_MOUNT_SOFTERR, ",softerr", "" },
+		{ NFS_MOUNT_SOFTREVAL, ",softreval", "" },
 		{ NFS_MOUNT_POSIX, ",posix", "" },
 		{ NFS_MOUNT_NOCTO, ",nocto", "" },
 		{ NFS_MOUNT_NOAC, ",noac", "" },
@@ -931,141 +707,6 @@ void nfs_umount_begin(struct super_block *sb)
 }
 EXPORT_SYMBOL_GPL(nfs_umount_begin);
 
-static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(void)
-{
-	struct nfs_parsed_mount_data *data;
-
-	data = kzalloc(sizeof(*data), GFP_KERNEL);
-	if (data) {
-		data->timeo		= NFS_UNSPEC_TIMEO;
-		data->retrans		= NFS_UNSPEC_RETRANS;
-		data->acregmin		= NFS_DEF_ACREGMIN;
-		data->acregmax		= NFS_DEF_ACREGMAX;
-		data->acdirmin		= NFS_DEF_ACDIRMIN;
-		data->acdirmax		= NFS_DEF_ACDIRMAX;
-		data->mount_server.port	= NFS_UNSPEC_PORT;
-		data->nfs_server.port	= NFS_UNSPEC_PORT;
-		data->nfs_server.protocol = XPRT_TRANSPORT_TCP;
-		data->selected_flavor	= RPC_AUTH_MAXFLAVOR;
-		data->minorversion	= 0;
-		data->need_mount	= true;
-		data->net		= current->nsproxy->net_ns;
-		data->lsm_opts		= NULL;
-	}
-	return data;
-}
-
-static void nfs_free_parsed_mount_data(struct nfs_parsed_mount_data *data)
-{
-	if (data) {
-		kfree(data->client_address);
-		kfree(data->mount_server.hostname);
-		kfree(data->nfs_server.export_path);
-		kfree(data->nfs_server.hostname);
-		kfree(data->fscache_uniq);
-		security_free_mnt_opts(&data->lsm_opts);
-		kfree(data);
-	}
-}
-
-/*
- * Sanity-check a server address provided by the mount command.
- *
- * Address family must be initialized, and address must not be
- * the ANY address for that family.
- */
-static int nfs_verify_server_address(struct sockaddr *addr)
-{
-	switch (addr->sa_family) {
-	case AF_INET: {
-		struct sockaddr_in *sa = (struct sockaddr_in *)addr;
-		return sa->sin_addr.s_addr != htonl(INADDR_ANY);
-	}
-	case AF_INET6: {
-		struct in6_addr *sa = &((struct sockaddr_in6 *)addr)->sin6_addr;
-		return !ipv6_addr_any(sa);
-	}
-	}
-
-	dfprintk(MOUNT, "NFS: Invalid IP address specified\n");
-	return 0;
-}
-
-/*
- * Select between a default port value and a user-specified port value.
- * If a zero value is set, then autobind will be used.
- */
-static void nfs_set_port(struct sockaddr *sap, int *port,
-				 const unsigned short default_port)
-{
-	if (*port == NFS_UNSPEC_PORT)
-		*port = default_port;
-
-	rpc_set_port(sap, *port);
-}
-
-/*
- * Sanity check the NFS transport protocol.
- *
- */
-static void nfs_validate_transport_protocol(struct nfs_parsed_mount_data *mnt)
-{
-	switch (mnt->nfs_server.protocol) {
-	case XPRT_TRANSPORT_UDP:
-	case XPRT_TRANSPORT_TCP:
-	case XPRT_TRANSPORT_RDMA:
-		break;
-	default:
-		mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
-	}
-}
-
-/*
- * For text based NFSv2/v3 mounts, the mount protocol transport default
- * settings should depend upon the specified NFS transport.
- */
-static void nfs_set_mount_transport_protocol(struct nfs_parsed_mount_data *mnt)
-{
-	nfs_validate_transport_protocol(mnt);
-
-	if (mnt->mount_server.protocol == XPRT_TRANSPORT_UDP ||
-	    mnt->mount_server.protocol == XPRT_TRANSPORT_TCP)
-			return;
-	switch (mnt->nfs_server.protocol) {
-	case XPRT_TRANSPORT_UDP:
-		mnt->mount_server.protocol = XPRT_TRANSPORT_UDP;
-		break;
-	case XPRT_TRANSPORT_TCP:
-	case XPRT_TRANSPORT_RDMA:
-		mnt->mount_server.protocol = XPRT_TRANSPORT_TCP;
-	}
-}
-
-/*
- * Add 'flavor' to 'auth_info' if not already present.
- * Returns true if 'flavor' ends up in the list, false otherwise
- */
-static bool nfs_auth_info_add(struct nfs_auth_info *auth_info,
-			      rpc_authflavor_t flavor)
-{
-	unsigned int i;
-	unsigned int max_flavor_len = ARRAY_SIZE(auth_info->flavors);
-
-	/* make sure this flavor isn't already in the list */
-	for (i = 0; i < auth_info->flavor_len; i++) {
-		if (flavor == auth_info->flavors[i])
-			return true;
-	}
-
-	if (auth_info->flavor_len + 1 >= max_flavor_len) {
-		dfprintk(MOUNT, "NFS: too many sec= flavors\n");
-		return false;
-	}
-
-	auth_info->flavors[auth_info->flavor_len++] = flavor;
-	return true;
-}
-
 /*
  * Return true if 'match' is in auth_info or auth_info is empty.
  * Return false otherwise.
@@ -1087,633 +728,13 @@ bool nfs_auth_info_match(const struct nfs_auth_info *auth_info,
 EXPORT_SYMBOL_GPL(nfs_auth_info_match);
 
 /*
- * Parse the value of the 'sec=' option.
- */
-static int nfs_parse_security_flavors(char *value,
-				      struct nfs_parsed_mount_data *mnt)
-{
-	substring_t args[MAX_OPT_ARGS];
-	rpc_authflavor_t pseudoflavor;
-	char *p;
-
-	dfprintk(MOUNT, "NFS: parsing sec=%s option\n", value);
-
-	while ((p = strsep(&value, ":")) != NULL) {
-		switch (match_token(p, nfs_secflavor_tokens, args)) {
-		case Opt_sec_none:
-			pseudoflavor = RPC_AUTH_NULL;
-			break;
-		case Opt_sec_sys:
-			pseudoflavor = RPC_AUTH_UNIX;
-			break;
-		case Opt_sec_krb5:
-			pseudoflavor = RPC_AUTH_GSS_KRB5;
-			break;
-		case Opt_sec_krb5i:
-			pseudoflavor = RPC_AUTH_GSS_KRB5I;
-			break;
-		case Opt_sec_krb5p:
-			pseudoflavor = RPC_AUTH_GSS_KRB5P;
-			break;
-		case Opt_sec_lkey:
-			pseudoflavor = RPC_AUTH_GSS_LKEY;
-			break;
-		case Opt_sec_lkeyi:
-			pseudoflavor = RPC_AUTH_GSS_LKEYI;
-			break;
-		case Opt_sec_lkeyp:
-			pseudoflavor = RPC_AUTH_GSS_LKEYP;
-			break;
-		case Opt_sec_spkm:
-			pseudoflavor = RPC_AUTH_GSS_SPKM;
-			break;
-		case Opt_sec_spkmi:
-			pseudoflavor = RPC_AUTH_GSS_SPKMI;
-			break;
-		case Opt_sec_spkmp:
-			pseudoflavor = RPC_AUTH_GSS_SPKMP;
-			break;
-		default:
-			dfprintk(MOUNT,
-				 "NFS: sec= option '%s' not recognized\n", p);
-			return 0;
-		}
-
-		if (!nfs_auth_info_add(&mnt->auth_info, pseudoflavor))
-			return 0;
-	}
-
-	return 1;
-}
-
-static int nfs_parse_version_string(char *string,
-		struct nfs_parsed_mount_data *mnt,
-		substring_t *args)
-{
-	mnt->flags &= ~NFS_MOUNT_VER3;
-	switch (match_token(string, nfs_vers_tokens, args)) {
-	case Opt_vers_2:
-		mnt->version = 2;
-		break;
-	case Opt_vers_3:
-		mnt->flags |= NFS_MOUNT_VER3;
-		mnt->version = 3;
-		break;
-	case Opt_vers_4:
-		/* Backward compatibility option. In future,
-		 * the mount program should always supply
-		 * a NFSv4 minor version number.
-		 */
-		mnt->version = 4;
-		break;
-	case Opt_vers_4_0:
-		mnt->version = 4;
-		mnt->minorversion = 0;
-		break;
-	case Opt_vers_4_1:
-		mnt->version = 4;
-		mnt->minorversion = 1;
-		break;
-	case Opt_vers_4_2:
-		mnt->version = 4;
-		mnt->minorversion = 2;
-		break;
-	default:
-		return 0;
-	}
-	return 1;
-}
-
-static int nfs_get_option_str(substring_t args[], char **option)
-{
-	kfree(*option);
-	*option = match_strdup(args);
-	return !*option;
-}
-
-static int nfs_get_option_ul(substring_t args[], unsigned long *option)
-{
-	int rc;
-	char *string;
-
-	string = match_strdup(args);
-	if (string == NULL)
-		return -ENOMEM;
-	rc = kstrtoul(string, 10, option);
-	kfree(string);
-
-	return rc;
-}
-
-static int nfs_get_option_ul_bound(substring_t args[], unsigned long *option,
-		unsigned long l_bound, unsigned long u_bound)
-{
-	int ret;
-
-	ret = nfs_get_option_ul(args, option);
-	if (ret != 0)
-		return ret;
-	if (*option < l_bound || *option > u_bound)
-		return -ERANGE;
-	return 0;
-}
-
-/*
- * Error-check and convert a string of mount options from user space into
- * a data structure.  The whole mount string is processed; bad options are
- * skipped as they are encountered.  If there were no errors, return 1;
- * otherwise return 0 (zero).
- */
-static int nfs_parse_mount_options(char *raw,
-				   struct nfs_parsed_mount_data *mnt)
-{
-	char *p, *string;
-	int rc, sloppy = 0, invalid_option = 0;
-	unsigned short protofamily = AF_UNSPEC;
-	unsigned short mountfamily = AF_UNSPEC;
-
-	if (!raw) {
-		dfprintk(MOUNT, "NFS: mount options string was NULL.\n");
-		return 1;
-	}
-	dfprintk(MOUNT, "NFS: nfs mount opts='%s'\n", raw);
-
-	rc = security_sb_eat_lsm_opts(raw, &mnt->lsm_opts);
-	if (rc)
-		goto out_security_failure;
-
-	while ((p = strsep(&raw, ",")) != NULL) {
-		substring_t args[MAX_OPT_ARGS];
-		unsigned long option;
-		int token;
-
-		if (!*p)
-			continue;
-
-		dfprintk(MOUNT, "NFS:   parsing nfs mount option '%s'\n", p);
-
-		token = match_token(p, nfs_mount_option_tokens, args);
-		switch (token) {
-
-		/*
-		 * boolean options:  foo/nofoo
-		 */
-		case Opt_soft:
-			mnt->flags |= NFS_MOUNT_SOFT;
-			mnt->flags &= ~NFS_MOUNT_SOFTERR;
-			break;
-		case Opt_softerr:
-			mnt->flags |= NFS_MOUNT_SOFTERR;
-			mnt->flags &= ~NFS_MOUNT_SOFT;
-			break;
-		case Opt_hard:
-			mnt->flags &= ~(NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR);
-			break;
-		case Opt_posix:
-			mnt->flags |= NFS_MOUNT_POSIX;
-			break;
-		case Opt_noposix:
-			mnt->flags &= ~NFS_MOUNT_POSIX;
-			break;
-		case Opt_cto:
-			mnt->flags &= ~NFS_MOUNT_NOCTO;
-			break;
-		case Opt_nocto:
-			mnt->flags |= NFS_MOUNT_NOCTO;
-			break;
-		case Opt_ac:
-			mnt->flags &= ~NFS_MOUNT_NOAC;
-			break;
-		case Opt_noac:
-			mnt->flags |= NFS_MOUNT_NOAC;
-			break;
-		case Opt_lock:
-			mnt->flags &= ~NFS_MOUNT_NONLM;
-			mnt->flags &= ~(NFS_MOUNT_LOCAL_FLOCK |
-					NFS_MOUNT_LOCAL_FCNTL);
-			break;
-		case Opt_nolock:
-			mnt->flags |= NFS_MOUNT_NONLM;
-			mnt->flags |= (NFS_MOUNT_LOCAL_FLOCK |
-				       NFS_MOUNT_LOCAL_FCNTL);
-			break;
-		case Opt_udp:
-			mnt->flags &= ~NFS_MOUNT_TCP;
-			mnt->nfs_server.protocol = XPRT_TRANSPORT_UDP;
-			break;
-		case Opt_tcp:
-			mnt->flags |= NFS_MOUNT_TCP;
-			mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
-			break;
-		case Opt_rdma:
-			mnt->flags |= NFS_MOUNT_TCP; /* for side protocols */
-			mnt->nfs_server.protocol = XPRT_TRANSPORT_RDMA;
-			xprt_load_transport(p);
-			break;
-		case Opt_acl:
-			mnt->flags &= ~NFS_MOUNT_NOACL;
-			break;
-		case Opt_noacl:
-			mnt->flags |= NFS_MOUNT_NOACL;
-			break;
-		case Opt_rdirplus:
-			mnt->flags &= ~NFS_MOUNT_NORDIRPLUS;
-			break;
-		case Opt_nordirplus:
-			mnt->flags |= NFS_MOUNT_NORDIRPLUS;
-			break;
-		case Opt_sharecache:
-			mnt->flags &= ~NFS_MOUNT_UNSHARED;
-			break;
-		case Opt_nosharecache:
-			mnt->flags |= NFS_MOUNT_UNSHARED;
-			break;
-		case Opt_resvport:
-			mnt->flags &= ~NFS_MOUNT_NORESVPORT;
-			break;
-		case Opt_noresvport:
-			mnt->flags |= NFS_MOUNT_NORESVPORT;
-			break;
-		case Opt_fscache:
-			mnt->options |= NFS_OPTION_FSCACHE;
-			kfree(mnt->fscache_uniq);
-			mnt->fscache_uniq = NULL;
-			break;
-		case Opt_nofscache:
-			mnt->options &= ~NFS_OPTION_FSCACHE;
-			kfree(mnt->fscache_uniq);
-			mnt->fscache_uniq = NULL;
-			break;
-		case Opt_migration:
-			mnt->options |= NFS_OPTION_MIGRATION;
-			break;
-		case Opt_nomigration:
-			mnt->options &= ~NFS_OPTION_MIGRATION;
-			break;
-
-		/*
-		 * options that take numeric values
-		 */
-		case Opt_port:
-			if (nfs_get_option_ul(args, &option) ||
-			    option > USHRT_MAX)
-				goto out_invalid_value;
-			mnt->nfs_server.port = option;
-			break;
-		case Opt_rsize:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->rsize = option;
-			break;
-		case Opt_wsize:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->wsize = option;
-			break;
-		case Opt_bsize:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->bsize = option;
-			break;
-		case Opt_timeo:
-			if (nfs_get_option_ul_bound(args, &option, 1, INT_MAX))
-				goto out_invalid_value;
-			mnt->timeo = option;
-			break;
-		case Opt_retrans:
-			if (nfs_get_option_ul_bound(args, &option, 0, INT_MAX))
-				goto out_invalid_value;
-			mnt->retrans = option;
-			break;
-		case Opt_acregmin:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->acregmin = option;
-			break;
-		case Opt_acregmax:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->acregmax = option;
-			break;
-		case Opt_acdirmin:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->acdirmin = option;
-			break;
-		case Opt_acdirmax:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->acdirmax = option;
-			break;
-		case Opt_actimeo:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->acregmin = mnt->acregmax =
-			mnt->acdirmin = mnt->acdirmax = option;
-			break;
-		case Opt_namelen:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			mnt->namlen = option;
-			break;
-		case Opt_mountport:
-			if (nfs_get_option_ul(args, &option) ||
-			    option > USHRT_MAX)
-				goto out_invalid_value;
-			mnt->mount_server.port = option;
-			break;
-		case Opt_mountvers:
-			if (nfs_get_option_ul(args, &option) ||
-			    option < NFS_MNT_VERSION ||
-			    option > NFS_MNT3_VERSION)
-				goto out_invalid_value;
-			mnt->mount_server.version = option;
-			break;
-		case Opt_minorversion:
-			if (nfs_get_option_ul(args, &option))
-				goto out_invalid_value;
-			if (option > NFS4_MAX_MINOR_VERSION)
-				goto out_invalid_value;
-			mnt->minorversion = option;
-			break;
-
-		/*
-		 * options that take text values
-		 */
-		case Opt_nfsvers:
-			string = match_strdup(args);
-			if (string == NULL)
-				goto out_nomem;
-			rc = nfs_parse_version_string(string, mnt, args);
-			kfree(string);
-			if (!rc)
-				goto out_invalid_value;
-			break;
-		case Opt_sec:
-			string = match_strdup(args);
-			if (string == NULL)
-				goto out_nomem;
-			rc = nfs_parse_security_flavors(string, mnt);
-			kfree(string);
-			if (!rc) {
-				dfprintk(MOUNT, "NFS:   unrecognized "
-						"security flavor\n");
-				return 0;
-			}
-			break;
-		case Opt_proto:
-			string = match_strdup(args);
-			if (string == NULL)
-				goto out_nomem;
-			token = match_token(string,
-					    nfs_xprt_protocol_tokens, args);
-
-			protofamily = AF_INET;
-			switch (token) {
-			case Opt_xprt_udp6:
-				protofamily = AF_INET6;
-				/* fall through */
-			case Opt_xprt_udp:
-				mnt->flags &= ~NFS_MOUNT_TCP;
-				mnt->nfs_server.protocol = XPRT_TRANSPORT_UDP;
-				break;
-			case Opt_xprt_tcp6:
-				protofamily = AF_INET6;
-				/* fall through */
-			case Opt_xprt_tcp:
-				mnt->flags |= NFS_MOUNT_TCP;
-				mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
-				break;
-			case Opt_xprt_rdma6:
-				protofamily = AF_INET6;
-				/* fall through */
-			case Opt_xprt_rdma:
-				/* vector side protocols to TCP */
-				mnt->flags |= NFS_MOUNT_TCP;
-				mnt->nfs_server.protocol = XPRT_TRANSPORT_RDMA;
-				xprt_load_transport(string);
-				break;
-			default:
-				dfprintk(MOUNT, "NFS:   unrecognized "
-						"transport protocol\n");
-				kfree(string);
-				return 0;
-			}
-			kfree(string);
-			break;
-		case Opt_mountproto:
-			string = match_strdup(args);
-			if (string == NULL)
-				goto out_nomem;
-			token = match_token(string,
-					    nfs_xprt_protocol_tokens, args);
-			kfree(string);
-
-			mountfamily = AF_INET;
-			switch (token) {
-			case Opt_xprt_udp6:
-				mountfamily = AF_INET6;
-				/* fall through */
-			case Opt_xprt_udp:
-				mnt->mount_server.protocol = XPRT_TRANSPORT_UDP;
-				break;
-			case Opt_xprt_tcp6:
-				mountfamily = AF_INET6;
-				/* fall through */
-			case Opt_xprt_tcp:
-				mnt->mount_server.protocol = XPRT_TRANSPORT_TCP;
-				break;
-			case Opt_xprt_rdma: /* not used for side protocols */
-			default:
-				dfprintk(MOUNT, "NFS:   unrecognized "
-						"transport protocol\n");
-				return 0;
-			}
-			break;
-		case Opt_addr:
-			string = match_strdup(args);
-			if (string == NULL)
-				goto out_nomem;
-			mnt->nfs_server.addrlen =
-				rpc_pton(mnt->net, string, strlen(string),
-					(struct sockaddr *)
-					&mnt->nfs_server.address,
-					sizeof(mnt->nfs_server.address));
-			kfree(string);
-			if (mnt->nfs_server.addrlen == 0)
-				goto out_invalid_address;
-			break;
-		case Opt_clientaddr:
-			if (nfs_get_option_str(args, &mnt->client_address))
-				goto out_nomem;
-			break;
-		case Opt_mounthost:
-			if (nfs_get_option_str(args,
-					       &mnt->mount_server.hostname))
-				goto out_nomem;
-			break;
-		case Opt_mountaddr:
-			string = match_strdup(args);
-			if (string == NULL)
-				goto out_nomem;
-			mnt->mount_server.addrlen =
-				rpc_pton(mnt->net, string, strlen(string),
-					(struct sockaddr *)
-					&mnt->mount_server.address,
-					sizeof(mnt->mount_server.address));
-			kfree(string);
-			if (mnt->mount_server.addrlen == 0)
-				goto out_invalid_address;
-			break;
-		case Opt_nconnect:
-			if (nfs_get_option_ul_bound(args, &option, 1, NFS_MAX_CONNECTIONS))
-				goto out_invalid_value;
-			mnt->nfs_server.nconnect = option;
-			break;
-		case Opt_lookupcache:
-			string = match_strdup(args);
-			if (string == NULL)
-				goto out_nomem;
-			token = match_token(string,
-					nfs_lookupcache_tokens, args);
-			kfree(string);
-			switch (token) {
-				case Opt_lookupcache_all:
-					mnt->flags &= ~(NFS_MOUNT_LOOKUP_CACHE_NONEG|NFS_MOUNT_LOOKUP_CACHE_NONE);
-					break;
-				case Opt_lookupcache_positive:
-					mnt->flags &= ~NFS_MOUNT_LOOKUP_CACHE_NONE;
-					mnt->flags |= NFS_MOUNT_LOOKUP_CACHE_NONEG;
-					break;
-				case Opt_lookupcache_none:
-					mnt->flags |= NFS_MOUNT_LOOKUP_CACHE_NONEG|NFS_MOUNT_LOOKUP_CACHE_NONE;
-					break;
-				default:
-					dfprintk(MOUNT, "NFS:   invalid "
-							"lookupcache argument\n");
-					return 0;
-			};
-			break;
-		case Opt_fscache_uniq:
-			if (nfs_get_option_str(args, &mnt->fscache_uniq))
-				goto out_nomem;
-			mnt->options |= NFS_OPTION_FSCACHE;
-			break;
-		case Opt_local_lock:
-			string = match_strdup(args);
-			if (string == NULL)
-				goto out_nomem;
-			token = match_token(string, nfs_local_lock_tokens,
-					args);
-			kfree(string);
-			switch (token) {
-			case Opt_local_lock_all:
-				mnt->flags |= (NFS_MOUNT_LOCAL_FLOCK |
-					       NFS_MOUNT_LOCAL_FCNTL);
-				break;
-			case Opt_local_lock_flock:
-				mnt->flags |= NFS_MOUNT_LOCAL_FLOCK;
-				break;
-			case Opt_local_lock_posix:
-				mnt->flags |= NFS_MOUNT_LOCAL_FCNTL;
-				break;
-			case Opt_local_lock_none:
-				mnt->flags &= ~(NFS_MOUNT_LOCAL_FLOCK |
-						NFS_MOUNT_LOCAL_FCNTL);
-				break;
-			default:
-				dfprintk(MOUNT, "NFS:	invalid	"
-						"local_lock argument\n");
-				return 0;
-			};
-			break;
-
-		/*
-		 * Special options
-		 */
-		case Opt_sloppy:
-			sloppy = 1;
-			dfprintk(MOUNT, "NFS:   relaxing parsing rules\n");
-			break;
-		case Opt_userspace:
-		case Opt_deprecated:
-			dfprintk(MOUNT, "NFS:   ignoring mount option "
-					"'%s'\n", p);
-			break;
-
-		default:
-			invalid_option = 1;
-			dfprintk(MOUNT, "NFS:   unrecognized mount option "
-					"'%s'\n", p);
-		}
-	}
-
-	if (!sloppy && invalid_option)
-		return 0;
-
-	if (mnt->minorversion && mnt->version != 4)
-		goto out_minorversion_mismatch;
-
-	if (mnt->options & NFS_OPTION_MIGRATION &&
-	    (mnt->version != 4 || mnt->minorversion != 0))
-		goto out_migration_misuse;
-
-	/*
-	 * verify that any proto=/mountproto= options match the address
-	 * families in the addr=/mountaddr= options.
-	 */
-	if (protofamily != AF_UNSPEC &&
-	    protofamily != mnt->nfs_server.address.ss_family)
-		goto out_proto_mismatch;
-
-	if (mountfamily != AF_UNSPEC) {
-		if (mnt->mount_server.addrlen) {
-			if (mountfamily != mnt->mount_server.address.ss_family)
-				goto out_mountproto_mismatch;
-		} else {
-			if (mountfamily != mnt->nfs_server.address.ss_family)
-				goto out_mountproto_mismatch;
-		}
-	}
-
-	return 1;
-
-out_mountproto_mismatch:
-	printk(KERN_INFO "NFS: mount server address does not match mountproto= "
-			 "option\n");
-	return 0;
-out_proto_mismatch:
-	printk(KERN_INFO "NFS: server address does not match proto= option\n");
-	return 0;
-out_invalid_address:
-	printk(KERN_INFO "NFS: bad IP address specified: %s\n", p);
-	return 0;
-out_invalid_value:
-	printk(KERN_INFO "NFS: bad mount option value specified: %s\n", p);
-	return 0;
-out_minorversion_mismatch:
-	printk(KERN_INFO "NFS: mount option vers=%u does not support "
-			 "minorversion=%u\n", mnt->version, mnt->minorversion);
-	return 0;
-out_migration_misuse:
-	printk(KERN_INFO
-		"NFS: 'migration' not supported for this NFS version\n");
-	return 0;
-out_nomem:
-	printk(KERN_INFO "NFS: not enough memory to parse option\n");
-	return 0;
-out_security_failure:
-	printk(KERN_INFO "NFS: security options invalid: %d\n", rc);
-	return 0;
-}
-
-/*
- * Ensure that a specified authtype in args->auth_info is supported by
- * the server. Returns 0 and sets args->selected_flavor if it's ok, and
+ * Ensure that a specified authtype in ctx->auth_info is supported by
+ * the server. Returns 0 and sets ctx->selected_flavor if it's ok, and
  * -EACCES if not.
  */
-static int nfs_verify_authflavors(struct nfs_parsed_mount_data *args,
-			rpc_authflavor_t *server_authlist, unsigned int count)
+static int nfs_verify_authflavors(struct nfs_fs_context *ctx,
+				  rpc_authflavor_t *server_authlist,
+				  unsigned int count)
 {
 	rpc_authflavor_t flavor = RPC_AUTH_MAXFLAVOR;
 	bool found_auth_null = false;
@@ -1734,7 +755,7 @@ static int nfs_verify_authflavors(struct nfs_parsed_mount_data *args,
 	for (i = 0; i < count; i++) {
 		flavor = server_authlist[i];
 
-		if (nfs_auth_info_match(&args->auth_info, flavor))
+		if (nfs_auth_info_match(&ctx->auth_info, flavor))
 			goto out;
 
 		if (flavor == RPC_AUTH_NULL)
@@ -1742,7 +763,7 @@ static int nfs_verify_authflavors(struct nfs_parsed_mount_data *args,
 	}
 
 	if (found_auth_null) {
-		flavor = args->auth_info.flavors[0];
+		flavor = ctx->auth_info.flavors[0];
 		goto out;
 	}
 
@@ -1751,8 +772,8 @@ static int nfs_verify_authflavors(struct nfs_parsed_mount_data *args,
 	return -EACCES;
 
 out:
-	args->selected_flavor = flavor;
-	dfprintk(MOUNT, "NFS: using auth flavor %u\n", args->selected_flavor);
+	ctx->selected_flavor = flavor;
+	dfprintk(MOUNT, "NFS: using auth flavor %u\n", ctx->selected_flavor);
 	return 0;
 }
 
@@ -1760,50 +781,51 @@ out:
  * Use the remote server's MOUNT service to request the NFS file handle
  * corresponding to the provided path.
  */
-static int nfs_request_mount(struct nfs_parsed_mount_data *args,
+static int nfs_request_mount(struct fs_context *fc,
 			     struct nfs_fh *root_fh,
 			     rpc_authflavor_t *server_authlist,
 			     unsigned int *server_authlist_len)
 {
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	struct nfs_mount_request request = {
 		.sap		= (struct sockaddr *)
-						&args->mount_server.address,
-		.dirpath	= args->nfs_server.export_path,
-		.protocol	= args->mount_server.protocol,
+						&ctx->mount_server.address,
+		.dirpath	= ctx->nfs_server.export_path,
+		.protocol	= ctx->mount_server.protocol,
 		.fh		= root_fh,
-		.noresvport	= args->flags & NFS_MOUNT_NORESVPORT,
+		.noresvport	= ctx->flags & NFS_MOUNT_NORESVPORT,
 		.auth_flav_len	= server_authlist_len,
 		.auth_flavs	= server_authlist,
-		.net		= args->net,
+		.net		= fc->net_ns,
 	};
 	int status;
 
-	if (args->mount_server.version == 0) {
-		switch (args->version) {
+	if (ctx->mount_server.version == 0) {
+		switch (ctx->version) {
 			default:
-				args->mount_server.version = NFS_MNT3_VERSION;
+				ctx->mount_server.version = NFS_MNT3_VERSION;
 				break;
 			case 2:
-				args->mount_server.version = NFS_MNT_VERSION;
+				ctx->mount_server.version = NFS_MNT_VERSION;
 		}
 	}
-	request.version = args->mount_server.version;
+	request.version = ctx->mount_server.version;
 
-	if (args->mount_server.hostname)
-		request.hostname = args->mount_server.hostname;
+	if (ctx->mount_server.hostname)
+		request.hostname = ctx->mount_server.hostname;
 	else
-		request.hostname = args->nfs_server.hostname;
+		request.hostname = ctx->nfs_server.hostname;
 
 	/*
 	 * Construct the mount server's address.
 	 */
-	if (args->mount_server.address.ss_family == AF_UNSPEC) {
-		memcpy(request.sap, &args->nfs_server.address,
-		       args->nfs_server.addrlen);
-		args->mount_server.addrlen = args->nfs_server.addrlen;
+	if (ctx->mount_server.address.sa_family == AF_UNSPEC) {
+		memcpy(request.sap, &ctx->nfs_server.address,
+		       ctx->nfs_server.addrlen);
+		ctx->mount_server.addrlen = ctx->nfs_server.addrlen;
 	}
-	request.salen = args->mount_server.addrlen;
-	nfs_set_port(request.sap, &args->mount_server.port, 0);
+	request.salen = ctx->mount_server.addrlen;
+	nfs_set_port(request.sap, &ctx->mount_server.port, 0);
 
 	/*
 	 * Now ask the mount server to map our export path
@@ -1819,20 +841,18 @@ static int nfs_request_mount(struct nfs_parsed_mount_data *args,
 	return 0;
 }
 
-static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_info,
-					struct nfs_subversion *nfs_mod)
+static struct nfs_server *nfs_try_mount_request(struct fs_context *fc)
 {
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	int status;
 	unsigned int i;
 	bool tried_auth_unix = false;
 	bool auth_null_in_list = false;
 	struct nfs_server *server = ERR_PTR(-EACCES);
-	struct nfs_parsed_mount_data *args = mount_info->parsed;
 	rpc_authflavor_t authlist[NFS_MAX_SECFLAVORS];
 	unsigned int authlist_len = ARRAY_SIZE(authlist);
 
-	status = nfs_request_mount(args, mount_info->mntfh, authlist,
-					&authlist_len);
+	status = nfs_request_mount(fc, ctx->mntfh, authlist, &authlist_len);
 	if (status)
 		return ERR_PTR(status);
 
@@ -1840,13 +860,13 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf
 	 * Was a sec= authflavor specified in the options? First, verify
 	 * whether the server supports it, and then just try to use it if so.
 	 */
-	if (args->auth_info.flavor_len > 0) {
-		status = nfs_verify_authflavors(args, authlist, authlist_len);
+	if (ctx->auth_info.flavor_len > 0) {
+		status = nfs_verify_authflavors(ctx, authlist, authlist_len);
 		dfprintk(MOUNT, "NFS: using auth flavor %u\n",
-			 args->selected_flavor);
+			 ctx->selected_flavor);
 		if (status)
 			return ERR_PTR(status);
-		return nfs_mod->rpc_ops->create_server(mount_info, nfs_mod);
+		return ctx->nfs_mod->rpc_ops->create_server(fc);
 	}
 
 	/*
@@ -1872,8 +892,8 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf
 			/* Fallthrough */
 		}
 		dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", flavor);
-		args->selected_flavor = flavor;
-		server = nfs_mod->rpc_ops->create_server(mount_info, nfs_mod);
+		ctx->selected_flavor = flavor;
+		server = ctx->nfs_mod->rpc_ops->create_server(fc);
 		if (!IS_ERR(server))
 			return server;
 	}
@@ -1888,348 +908,23 @@ static struct nfs_server *nfs_try_mount_request(struct nfs_mount_info *mount_inf
 
 	/* Last chance! Try AUTH_UNIX */
 	dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", RPC_AUTH_UNIX);
-	args->selected_flavor = RPC_AUTH_UNIX;
-	return nfs_mod->rpc_ops->create_server(mount_info, nfs_mod);
+	ctx->selected_flavor = RPC_AUTH_UNIX;
+	return ctx->nfs_mod->rpc_ops->create_server(fc);
 }
 
-struct dentry *nfs_try_mount(int flags, const char *dev_name,
-			     struct nfs_mount_info *mount_info,
-			     struct nfs_subversion *nfs_mod)
+int nfs_try_get_tree(struct fs_context *fc)
 {
-	struct nfs_server *server;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 
-	if (mount_info->parsed->need_mount)
-		server = nfs_try_mount_request(mount_info, nfs_mod);
+	if (ctx->need_mount)
+		ctx->server = nfs_try_mount_request(fc);
 	else
-		server = nfs_mod->rpc_ops->create_server(mount_info, nfs_mod);
+		ctx->server = ctx->nfs_mod->rpc_ops->create_server(fc);
 
-	if (IS_ERR(server))
-		return ERR_CAST(server);
-
-	return nfs_fs_mount_common(server, flags, dev_name, mount_info, nfs_mod);
+	return nfs_get_tree_common(fc);
 }
-EXPORT_SYMBOL_GPL(nfs_try_mount);
+EXPORT_SYMBOL_GPL(nfs_try_get_tree);
 
-/*
- * Split "dev_name" into "hostname:export_path".
- *
- * The leftmost colon demarks the split between the server's hostname
- * and the export path.  If the hostname starts with a left square
- * bracket, then it may contain colons.
- *
- * Note: caller frees hostname and export path, even on error.
- */
-static int nfs_parse_devname(const char *dev_name,
-			     char **hostname, size_t maxnamlen,
-			     char **export_path, size_t maxpathlen)
-{
-	size_t len;
-	char *end;
-
-	if (unlikely(!dev_name || !*dev_name)) {
-		dfprintk(MOUNT, "NFS: device name not specified\n");
-		return -EINVAL;
-	}
-
-	/* Is the host name protected with square brakcets? */
-	if (*dev_name == '[') {
-		end = strchr(++dev_name, ']');
-		if (end == NULL || end[1] != ':')
-			goto out_bad_devname;
-
-		len = end - dev_name;
-		end++;
-	} else {
-		char *comma;
-
-		end = strchr(dev_name, ':');
-		if (end == NULL)
-			goto out_bad_devname;
-		len = end - dev_name;
-
-		/* kill possible hostname list: not supported */
-		comma = strchr(dev_name, ',');
-		if (comma != NULL && comma < end)
-			len = comma - dev_name;
-	}
-
-	if (len > maxnamlen)
-		goto out_hostname;
-
-	/* N.B. caller will free nfs_server.hostname in all cases */
-	*hostname = kstrndup(dev_name, len, GFP_KERNEL);
-	if (*hostname == NULL)
-		goto out_nomem;
-	len = strlen(++end);
-	if (len > maxpathlen)
-		goto out_path;
-	*export_path = kstrndup(end, len, GFP_KERNEL);
-	if (!*export_path)
-		goto out_nomem;
-
-	dfprintk(MOUNT, "NFS: MNTPATH: '%s'\n", *export_path);
-	return 0;
-
-out_bad_devname:
-	dfprintk(MOUNT, "NFS: device name not in host:path format\n");
-	return -EINVAL;
-
-out_nomem:
-	dfprintk(MOUNT, "NFS: not enough memory to parse device name\n");
-	return -ENOMEM;
-
-out_hostname:
-	dfprintk(MOUNT, "NFS: server hostname too long\n");
-	return -ENAMETOOLONG;
-
-out_path:
-	dfprintk(MOUNT, "NFS: export pathname too long\n");
-	return -ENAMETOOLONG;
-}
-
-/*
- * Validate the NFS2/NFS3 mount data
- * - fills in the mount root filehandle
- *
- * For option strings, user space handles the following behaviors:
- *
- * + DNS: mapping server host name to IP address ("addr=" option)
- *
- * + failure mode: how to behave if a mount request can't be handled
- *   immediately ("fg/bg" option)
- *
- * + retry: how often to retry a mount request ("retry=" option)
- *
- * + breaking back: trying proto=udp after proto=tcp, v2 after v3,
- *   mountproto=tcp after mountproto=udp, and so on
- */
-static int nfs23_validate_mount_data(void *options,
-				     struct nfs_parsed_mount_data *args,
-				     struct nfs_fh *mntfh,
-				     const char *dev_name)
-{
-	struct nfs_mount_data *data = (struct nfs_mount_data *)options;
-	struct sockaddr *sap = (struct sockaddr *)&args->nfs_server.address;
-	int extra_flags = NFS_MOUNT_LEGACY_INTERFACE;
-
-	if (data == NULL)
-		goto out_no_data;
-
-	args->version = NFS_DEFAULT_VERSION;
-	switch (data->version) {
-	case 1:
-		data->namlen = 0; /* fall through */
-	case 2:
-		data->bsize = 0; /* fall through */
-	case 3:
-		if (data->flags & NFS_MOUNT_VER3)
-			goto out_no_v3;
-		data->root.size = NFS2_FHSIZE;
-		memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE);
-		/* Turn off security negotiation */
-		extra_flags |= NFS_MOUNT_SECFLAVOUR;
-		/* fall through */
-	case 4:
-		if (data->flags & NFS_MOUNT_SECFLAVOUR)
-			goto out_no_sec;
-		/* fall through */
-	case 5:
-		memset(data->context, 0, sizeof(data->context));
-		/* fall through */
-	case 6:
-		if (data->flags & NFS_MOUNT_VER3) {
-			if (data->root.size > NFS3_FHSIZE || data->root.size == 0)
-				goto out_invalid_fh;
-			mntfh->size = data->root.size;
-			args->version = 3;
-		} else {
-			mntfh->size = NFS2_FHSIZE;
-			args->version = 2;
-		}
-
-
-		memcpy(mntfh->data, data->root.data, mntfh->size);
-		if (mntfh->size < sizeof(mntfh->data))
-			memset(mntfh->data + mntfh->size, 0,
-			       sizeof(mntfh->data) - mntfh->size);
-
-		/*
-		 * Translate to nfs_parsed_mount_data, which nfs_fill_super
-		 * can deal with.
-		 */
-		args->flags		= data->flags & NFS_MOUNT_FLAGMASK;
-		args->flags		|= extra_flags;
-		args->rsize		= data->rsize;
-		args->wsize		= data->wsize;
-		args->timeo		= data->timeo;
-		args->retrans		= data->retrans;
-		args->acregmin		= data->acregmin;
-		args->acregmax		= data->acregmax;
-		args->acdirmin		= data->acdirmin;
-		args->acdirmax		= data->acdirmax;
-		args->need_mount	= false;
-
-		memcpy(sap, &data->addr, sizeof(data->addr));
-		args->nfs_server.addrlen = sizeof(data->addr);
-		args->nfs_server.port = ntohs(data->addr.sin_port);
-		if (sap->sa_family != AF_INET ||
-		    !nfs_verify_server_address(sap))
-			goto out_no_address;
-
-		if (!(data->flags & NFS_MOUNT_TCP))
-			args->nfs_server.protocol = XPRT_TRANSPORT_UDP;
-		/* N.B. caller will free nfs_server.hostname in all cases */
-		args->nfs_server.hostname = kstrdup(data->hostname, GFP_KERNEL);
-		args->namlen		= data->namlen;
-		args->bsize		= data->bsize;
-
-		if (data->flags & NFS_MOUNT_SECFLAVOUR)
-			args->selected_flavor = data->pseudoflavor;
-		else
-			args->selected_flavor = RPC_AUTH_UNIX;
-		if (!args->nfs_server.hostname)
-			goto out_nomem;
-
-		if (!(data->flags & NFS_MOUNT_NONLM))
-			args->flags &= ~(NFS_MOUNT_LOCAL_FLOCK|
-					 NFS_MOUNT_LOCAL_FCNTL);
-		else
-			args->flags |= (NFS_MOUNT_LOCAL_FLOCK|
-					NFS_MOUNT_LOCAL_FCNTL);
-		/*
-		 * The legacy version 6 binary mount data from userspace has a
-		 * field used only to transport selinux information into the
-		 * the kernel.  To continue to support that functionality we
-		 * have a touch of selinux knowledge here in the NFS code. The
-		 * userspace code converted context=blah to just blah so we are
-		 * converting back to the full string selinux understands.
-		 */
-		if (data->context[0]){
-#ifdef CONFIG_SECURITY_SELINUX
-			int rc;
-			data->context[NFS_MAX_CONTEXT_LEN] = '\0';
-			rc = security_add_mnt_opt("context", data->context,
-					strlen(data->context), &args->lsm_opts);
-			if (rc)
-				return rc;
-#else
-			return -EINVAL;
-#endif
-		}
-
-		break;
-	default:
-		return NFS_TEXT_DATA;
-	}
-
-	return 0;
-
-out_no_data:
-	dfprintk(MOUNT, "NFS: mount program didn't pass any mount data\n");
-	return -EINVAL;
-
-out_no_v3:
-	dfprintk(MOUNT, "NFS: nfs_mount_data version %d does not support v3\n",
-		 data->version);
-	return -EINVAL;
-
-out_no_sec:
-	dfprintk(MOUNT, "NFS: nfs_mount_data version supports only AUTH_SYS\n");
-	return -EINVAL;
-
-out_nomem:
-	dfprintk(MOUNT, "NFS: not enough memory to handle mount options\n");
-	return -ENOMEM;
-
-out_no_address:
-	dfprintk(MOUNT, "NFS: mount program didn't pass remote address\n");
-	return -EINVAL;
-
-out_invalid_fh:
-	dfprintk(MOUNT, "NFS: invalid root filehandle\n");
-	return -EINVAL;
-}
-
-#if IS_ENABLED(CONFIG_NFS_V4)
-static int nfs_validate_mount_data(struct file_system_type *fs_type,
-				   void *options,
-				   struct nfs_parsed_mount_data *args,
-				   struct nfs_fh *mntfh,
-				   const char *dev_name)
-{
-	if (fs_type == &nfs_fs_type)
-		return nfs23_validate_mount_data(options, args, mntfh, dev_name);
-	return nfs4_validate_mount_data(options, args, dev_name);
-}
-#else
-static int nfs_validate_mount_data(struct file_system_type *fs_type,
-				   void *options,
-				   struct nfs_parsed_mount_data *args,
-				   struct nfs_fh *mntfh,
-				   const char *dev_name)
-{
-	return nfs23_validate_mount_data(options, args, mntfh, dev_name);
-}
-#endif
-
-static int nfs_validate_text_mount_data(void *options,
-					struct nfs_parsed_mount_data *args,
-					const char *dev_name)
-{
-	int port = 0;
-	int max_namelen = PAGE_SIZE;
-	int max_pathlen = NFS_MAXPATHLEN;
-	struct sockaddr *sap = (struct sockaddr *)&args->nfs_server.address;
-
-	if (nfs_parse_mount_options((char *)options, args) == 0)
-		return -EINVAL;
-
-	if (!nfs_verify_server_address(sap))
-		goto out_no_address;
-
-	if (args->version == 4) {
-#if IS_ENABLED(CONFIG_NFS_V4)
-		if (args->nfs_server.protocol == XPRT_TRANSPORT_RDMA)
-			port = NFS_RDMA_PORT;
-		else
-			port = NFS_PORT;
-		max_namelen = NFS4_MAXNAMLEN;
-		max_pathlen = NFS4_MAXPATHLEN;
-		nfs_validate_transport_protocol(args);
-		if (args->nfs_server.protocol == XPRT_TRANSPORT_UDP)
-			goto out_invalid_transport_udp;
-		nfs4_validate_mount_flags(args);
-#else
-		goto out_v4_not_compiled;
-#endif /* CONFIG_NFS_V4 */
-	} else {
-		nfs_set_mount_transport_protocol(args);
-		if (args->nfs_server.protocol == XPRT_TRANSPORT_RDMA)
-			port = NFS_RDMA_PORT;
-	}
-
-	nfs_set_port(sap, &args->nfs_server.port, port);
-
-	return nfs_parse_devname(dev_name,
-				   &args->nfs_server.hostname,
-				   max_namelen,
-				   &args->nfs_server.export_path,
-				   max_pathlen);
-
-#if !IS_ENABLED(CONFIG_NFS_V4)
-out_v4_not_compiled:
-	dfprintk(MOUNT, "NFS: NFSv4 is not compiled into kernel\n");
-	return -EPROTONOSUPPORT;
-#else
-out_invalid_transport_udp:
-	dfprintk(MOUNT, "NFSv4: Unsupported transport protocol udp\n");
-	return -EINVAL;
-#endif /* !CONFIG_NFS_V4 */
-
-out_no_address:
-	dfprintk(MOUNT, "NFS: mount program didn't pass remote address\n");
-	return -EINVAL;
-}
 
 #define NFS_REMOUNT_CMP_FLAGMASK ~(NFS_MOUNT_INTR \
 		| NFS_MOUNT_SECURE \
@@ -2246,39 +941,35 @@ out_no_address:
 
 static int
 nfs_compare_remount_data(struct nfs_server *nfss,
-			 struct nfs_parsed_mount_data *data)
-{
-	if ((data->flags ^ nfss->flags) & NFS_REMOUNT_CMP_FLAGMASK ||
-	    data->rsize != nfss->rsize ||
-	    data->wsize != nfss->wsize ||
-	    data->version != nfss->nfs_client->rpc_ops->version ||
-	    data->minorversion != nfss->nfs_client->cl_minorversion ||
-	    data->retrans != nfss->client->cl_timeout->to_retries ||
-	    !nfs_auth_info_match(&data->auth_info, nfss->client->cl_auth->au_flavor) ||
-	    data->acregmin != nfss->acregmin / HZ ||
-	    data->acregmax != nfss->acregmax / HZ ||
-	    data->acdirmin != nfss->acdirmin / HZ ||
-	    data->acdirmax != nfss->acdirmax / HZ ||
-	    data->timeo != (10U * nfss->client->cl_timeout->to_initval / HZ) ||
-	    (data->options & NFS_OPTION_FSCACHE) != (nfss->options & NFS_OPTION_FSCACHE) ||
-	    data->nfs_server.port != nfss->port ||
-	    data->nfs_server.addrlen != nfss->nfs_client->cl_addrlen ||
-	    !rpc_cmp_addr((struct sockaddr *)&data->nfs_server.address,
+			 struct nfs_fs_context *ctx)
+{
+	if ((ctx->flags ^ nfss->flags) & NFS_REMOUNT_CMP_FLAGMASK ||
+	    ctx->rsize != nfss->rsize ||
+	    ctx->wsize != nfss->wsize ||
+	    ctx->version != nfss->nfs_client->rpc_ops->version ||
+	    ctx->minorversion != nfss->nfs_client->cl_minorversion ||
+	    ctx->retrans != nfss->client->cl_timeout->to_retries ||
+	    !nfs_auth_info_match(&ctx->auth_info, nfss->client->cl_auth->au_flavor) ||
+	    ctx->acregmin != nfss->acregmin / HZ ||
+	    ctx->acregmax != nfss->acregmax / HZ ||
+	    ctx->acdirmin != nfss->acdirmin / HZ ||
+	    ctx->acdirmax != nfss->acdirmax / HZ ||
+	    ctx->timeo != (10U * nfss->client->cl_timeout->to_initval / HZ) ||
+	    (ctx->options & NFS_OPTION_FSCACHE) != (nfss->options & NFS_OPTION_FSCACHE) ||
+	    ctx->nfs_server.port != nfss->port ||
+	    ctx->nfs_server.addrlen != nfss->nfs_client->cl_addrlen ||
+	    !rpc_cmp_addr((struct sockaddr *)&ctx->nfs_server.address,
 			  (struct sockaddr *)&nfss->nfs_client->cl_addr))
 		return -EINVAL;
 
 	return 0;
 }
 
-int
-nfs_remount(struct super_block *sb, int *flags, char *raw_data)
+int nfs_reconfigure(struct fs_context *fc)
 {
-	int error;
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
+	struct super_block *sb = fc->root->d_sb;
 	struct nfs_server *nfss = sb->s_fs_info;
-	struct nfs_parsed_mount_data *data;
-	struct nfs_mount_data *options = (struct nfs_mount_data *)raw_data;
-	struct nfs4_mount_data *options4 = (struct nfs4_mount_data *)raw_data;
-	u32 nfsvers = nfss->nfs_client->rpc_ops->version;
 
 	sync_filesystem(sb);
 
@@ -2288,92 +979,38 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
 	 * ones were explicitly specified. Fall back to legacy behavior and
 	 * just return success.
 	 */
-	if ((nfsvers == 4 && (!options4 || options4->version == 1)) ||
-	    (nfsvers <= 3 && (!options || (options->version >= 1 &&
-					   options->version <= 6))))
+	if (ctx->skip_reconfig_option_check)
 		return 0;
 
-	data = nfs_alloc_parsed_mount_data();
-	if (data == NULL)
-		return -ENOMEM;
-
-	/* fill out struct with values from existing mount */
-	data->flags = nfss->flags;
-	data->rsize = nfss->rsize;
-	data->wsize = nfss->wsize;
-	data->retrans = nfss->client->cl_timeout->to_retries;
-	data->selected_flavor = nfss->client->cl_auth->au_flavor;
-	data->acregmin = nfss->acregmin / HZ;
-	data->acregmax = nfss->acregmax / HZ;
-	data->acdirmin = nfss->acdirmin / HZ;
-	data->acdirmax = nfss->acdirmax / HZ;
-	data->timeo = 10U * nfss->client->cl_timeout->to_initval / HZ;
-	data->nfs_server.port = nfss->port;
-	data->nfs_server.addrlen = nfss->nfs_client->cl_addrlen;
-	data->version = nfsvers;
-	data->minorversion = nfss->nfs_client->cl_minorversion;
-	data->net = current->nsproxy->net_ns;
-	memcpy(&data->nfs_server.address, &nfss->nfs_client->cl_addr,
-		data->nfs_server.addrlen);
-
-	/* overwrite those values with any that were specified */
-	error = -EINVAL;
-	if (!nfs_parse_mount_options((char *)options, data))
-		goto out;
-
 	/*
 	 * noac is a special case. It implies -o sync, but that's not
-	 * necessarily reflected in the mtab options. do_remount_sb
+	 * necessarily reflected in the mtab options. reconfigure_super
 	 * will clear SB_SYNCHRONOUS if -o sync wasn't specified in the
 	 * remount options, so we have to explicitly reset it.
 	 */
-	if (data->flags & NFS_MOUNT_NOAC)
-		*flags |= SB_SYNCHRONOUS;
+	if (ctx->flags & NFS_MOUNT_NOAC) {
+		fc->sb_flags |= SB_SYNCHRONOUS;
+		fc->sb_flags_mask |= SB_SYNCHRONOUS;
+	}
 
 	/* compare new mount options with old ones */
-	error = nfs_compare_remount_data(nfss, data);
-	if (!error)
-		error = security_sb_remount(sb, data->lsm_opts);
-out:
-	nfs_free_parsed_mount_data(data);
-	return error;
-}
-EXPORT_SYMBOL_GPL(nfs_remount);
-
-/*
- * Initialise the common bits of the superblock
- */
-static void nfs_initialise_sb(struct super_block *sb)
-{
-	struct nfs_server *server = NFS_SB(sb);
-
-	sb->s_magic = NFS_SUPER_MAGIC;
-
-	/* We probably want something more informative here */
-	snprintf(sb->s_id, sizeof(sb->s_id),
-		 "%u:%u", MAJOR(sb->s_dev), MINOR(sb->s_dev));
-
-	if (sb->s_blocksize == 0)
-		sb->s_blocksize = nfs_block_bits(server->wsize,
-						 &sb->s_blocksize_bits);
-
-	nfs_super_set_maxbytes(sb, server->maxfilesize);
+	return nfs_compare_remount_data(nfss, ctx);
 }
+EXPORT_SYMBOL_GPL(nfs_reconfigure);
 
 /*
- * Finish setting up an NFS2/3 superblock
+ * Finish setting up an NFS superblock
  */
-void nfs_fill_super(struct super_block *sb, struct nfs_mount_info *mount_info)
+static void nfs_fill_super(struct super_block *sb, struct nfs_fs_context *ctx)
 {
-	struct nfs_parsed_mount_data *data = mount_info->parsed;
 	struct nfs_server *server = NFS_SB(sb);
 
 	sb->s_blocksize_bits = 0;
 	sb->s_blocksize = 0;
 	sb->s_xattr = server->nfs_client->cl_nfs_mod->xattr;
 	sb->s_op = server->nfs_client->cl_nfs_mod->sops;
-	if (data && data->bsize)
-		sb->s_blocksize = nfs_block_size(data->bsize, &sb->s_blocksize_bits);
+	if (ctx && ctx->bsize)
+		sb->s_blocksize = nfs_block_size(ctx->bsize, &sb->s_blocksize_bits);
 
 	if (server->nfs_client->rpc_ops->version != 2) {
 		/* The VFS shouldn't apply the umask to mode bits. We will do
@@ -2393,53 +1030,27 @@ void nfs_fill_super(struct super_block *sb, struct nfs_mount_info *mount_info)
 		sb->s_time_max = S64_MAX;
 	}
 
- 	nfs_initialise_sb(sb);
-}
-EXPORT_SYMBOL_GPL(nfs_fill_super);
-
-/*
- * Finish setting up a cloned NFS2/3/4 superblock
- */
-static void nfs_clone_super(struct super_block *sb,
-			    struct nfs_mount_info *mount_info)
-{
-	const struct super_block *old_sb = mount_info->cloned->sb;
-	struct nfs_server *server = NFS_SB(sb);
-
-	sb->s_blocksize_bits = old_sb->s_blocksize_bits;
-	sb->s_blocksize = old_sb->s_blocksize;
-	sb->s_maxbytes = old_sb->s_maxbytes;
-	sb->s_xattr = old_sb->s_xattr;
-	sb->s_op = old_sb->s_op;
-	sb->s_export_op = old_sb->s_export_op;
+	sb->s_magic = NFS_SUPER_MAGIC;
 
-	if (server->nfs_client->rpc_ops->version != 2) {
-		/* The VFS shouldn't apply the umask to mode bits. We will do
-		 * so ourselves when necessary.
-		 */
-		sb->s_flags |= SB_POSIXACL;
-		sb->s_time_gran = 1;
-	} else
-		sb->s_time_gran = 1000;
+	/* We probably want something more informative here */
+	snprintf(sb->s_id, sizeof(sb->s_id),
+		 "%u:%u", MAJOR(sb->s_dev), MINOR(sb->s_dev));
 
-	if (server->nfs_client->rpc_ops->version != 4) {
-		sb->s_time_min = 0;
-		sb->s_time_max = U32_MAX;
-	} else {
-		sb->s_time_min = S64_MIN;
-		sb->s_time_max = S64_MAX;
-	}
+	if (sb->s_blocksize == 0)
+		sb->s_blocksize = nfs_block_bits(server->wsize,
+						 &sb->s_blocksize_bits);
 
- 	nfs_initialise_sb(sb);
+	nfs_super_set_maxbytes(sb, server->maxfilesize);
 }
 
-static int nfs_compare_mount_options(const struct super_block *s, const struct nfs_server *b, int flags)
+static int nfs_compare_mount_options(const struct super_block *s, const struct nfs_server *b,
+				     const struct fs_context *fc)
 {
 	const struct nfs_server *a = s->s_fs_info;
 	const struct rpc_clnt *clnt_a = a->client;
 	const struct rpc_clnt *clnt_b = b->client;
 
-	if ((s->s_flags & NFS_MS_MASK) != (flags & NFS_MS_MASK))
+	if ((s->s_flags & NFS_SB_MASK) != (fc->sb_flags & NFS_SB_MASK))
 		goto Ebusy;
 	if (a->nfs_client != b->nfs_client)
 		goto Ebusy;
@@ -2464,19 +1075,11 @@ Ebusy:
 	return 0;
 }
 
-struct nfs_sb_mountdata {
-	struct nfs_server *server;
-	int mntflags;
-};
-
-static int nfs_set_super(struct super_block *s, void *data)
+static int nfs_set_super(struct super_block *s, struct fs_context *fc)
 {
-	struct nfs_sb_mountdata *sb_mntdata = data;
-	struct nfs_server *server = sb_mntdata->server;
+	struct nfs_server *server = fc->s_fs_info;
 	int ret;
 
-	s->s_flags = sb_mntdata->mntflags;
-	s->s_fs_info = server;
 	s->s_d_op = server->nfs_client->rpc_ops->dentry_ops;
 	ret = set_anon_super(s, server);
 	if (ret == 0)
@@ -2541,11 +1144,9 @@ static int nfs_compare_userns(const struct nfs_server *old,
 	return 1;
 }
 
-static int nfs_compare_super(struct super_block *sb, void *data)
+static int nfs_compare_super(struct super_block *sb, struct fs_context *fc)
 {
-	struct nfs_sb_mountdata *sb_mntdata = data;
-	struct nfs_server *server = sb_mntdata->server, *old = NFS_SB(sb);
-	int mntflags = sb_mntdata->mntflags;
+	struct nfs_server *server = fc->s_fs_info, *old = NFS_SB(sb);
 
 	if (!nfs_compare_super_address(old, server))
 		return 0;
@@ -2556,13 +1157,12 @@ static int nfs_compare_super(struct super_block *sb, void *data)
 		return 0;
 	if (!nfs_compare_userns(old, server))
 		return 0;
-	return nfs_compare_mount_options(sb, server, mntflags);
+	return nfs_compare_mount_options(sb, server, fc);
 }
 
 #ifdef CONFIG_NFS_FSCACHE
 static void nfs_get_cache_cookie(struct super_block *sb,
-				 struct nfs_parsed_mount_data *parsed,
-				 struct nfs_clone_mount *cloned)
+				 struct nfs_fs_context *ctx)
 {
 	struct nfs_server *nfss = NFS_SB(sb);
 	char *uniq = NULL;
@@ -2571,80 +1171,35 @@ static void nfs_get_cache_cookie(struct super_block *sb,
 	nfss->fscache_key = NULL;
 	nfss->fscache = NULL;
 
-	if (parsed) {
-		if (!(parsed->options & NFS_OPTION_FSCACHE))
-			return;
-		if (parsed->fscache_uniq) {
-			uniq = parsed->fscache_uniq;
-			ulen = strlen(parsed->fscache_uniq);
-		}
-	} else if (cloned) {
-		struct nfs_server *mnt_s = NFS_SB(cloned->sb);
+	if (!ctx)
+		return;
+
+	if (ctx->clone_data.sb) {
+		struct nfs_server *mnt_s = NFS_SB(ctx->clone_data.sb);
 		if (!(mnt_s->options & NFS_OPTION_FSCACHE))
 			return;
 		if (mnt_s->fscache_key) {
 			uniq = mnt_s->fscache_key->key.uniquifier;
 			ulen = mnt_s->fscache_key->key.uniq_len;
-		};
-	} else
-		return;
+		}
+	} else {
+		if (!(ctx->options & NFS_OPTION_FSCACHE))
+			return;
+		if (ctx->fscache_uniq) {
+			uniq = ctx->fscache_uniq;
+			ulen = strlen(ctx->fscache_uniq);
+		}
+	}
 
 	nfs_fscache_get_super_cookie(sb, uniq, ulen);
 }
 #else
 static void nfs_get_cache_cookie(struct super_block *sb,
-				 struct nfs_parsed_mount_data *parsed,
-				 struct nfs_clone_mount *cloned)
+				 struct nfs_fs_context *ctx)
 {
 }
 #endif
 
-int nfs_set_sb_security(struct super_block *s, struct dentry *mntroot,
-			struct nfs_mount_info *mount_info)
-{
-	int error;
-	unsigned long kflags = 0, kflags_out = 0;
-	if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL)
-		kflags |= SECURITY_LSM_NATIVE_LABELS;
-
-	error = security_sb_set_mnt_opts(s, mount_info->parsed->lsm_opts,
-						kflags, &kflags_out);
-	if (error)
-		goto err;
-
-	if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL &&
-		!(kflags_out & SECURITY_LSM_NATIVE_LABELS))
-		NFS_SB(s)->caps &= ~NFS_CAP_SECURITY_LABEL;
-err:
-	return error;
-}
-EXPORT_SYMBOL_GPL(nfs_set_sb_security);
-
-int nfs_clone_sb_security(struct super_block *s, struct dentry *mntroot,
-			  struct nfs_mount_info *mount_info)
-{
-	int error;
-	unsigned long kflags = 0, kflags_out = 0;
-
-	/* clone any lsm security options from the parent to the new sb */
-	if (d_inode(mntroot)->i_op != NFS_SB(s)->nfs_client->rpc_ops->dir_inode_ops)
-		return -ESTALE;
-
-	if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL)
-		kflags |= SECURITY_LSM_NATIVE_LABELS;
-
-	error = security_sb_clone_mnt_opts(mount_info->cloned->sb, s, kflags,
-			&kflags_out);
-	if (error)
-		return error;
-
-	if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL &&
-		!(kflags_out & SECURITY_LSM_NATIVE_LABELS))
-		NFS_SB(s)->caps &= ~NFS_CAP_SECURITY_LABEL;
-	return 0;
-}
-EXPORT_SYMBOL_GPL(nfs_clone_sb_security);
-
 static void nfs_set_readahead(struct backing_dev_info *bdi,
 			      unsigned long iomax_pages)
 {
@@ -2652,35 +1207,39 @@ static void nfs_set_readahead(struct backing_dev_info *bdi,
 	bdi->io_pages = iomax_pages;
 }
 
-struct dentry *nfs_fs_mount_common(struct nfs_server *server,
-				   int flags, const char *dev_name,
-				   struct nfs_mount_info *mount_info,
-				   struct nfs_subversion *nfs_mod)
+int nfs_get_tree_common(struct fs_context *fc)
 {
+	struct nfs_fs_context *ctx = nfs_fc2context(fc);
 	struct super_block *s;
-	struct dentry *mntroot = ERR_PTR(-ENOMEM);
-	int (*compare_super)(struct super_block *, void *) = nfs_compare_super;
-	struct nfs_sb_mountdata sb_mntdata = {
-		.mntflags = flags,
-		.server = server,
-	};
+	int (*compare_super)(struct super_block *, struct fs_context *) = nfs_compare_super;
+	struct nfs_server *server = ctx->server;
 	int error;
 
+	ctx->server = NULL;
+	if (IS_ERR(server))
+		return PTR_ERR(server);
+
 	if (server->flags & NFS_MOUNT_UNSHARED)
 		compare_super = NULL;
 
 	/* -o noac implies -o sync */
 	if (server->flags & NFS_MOUNT_NOAC)
-		sb_mntdata.mntflags |= SB_SYNCHRONOUS;
+		fc->sb_flags |= SB_SYNCHRONOUS;
+
+	if (ctx->clone_data.sb)
+		if (ctx->clone_data.sb->s_flags & SB_SYNCHRONOUS)
+			fc->sb_flags |= SB_SYNCHRONOUS;
 
-	if (mount_info->cloned != NULL && mount_info->cloned->sb != NULL)
-		if (mount_info->cloned->sb->s_flags & SB_SYNCHRONOUS)
-			sb_mntdata.mntflags |= SB_SYNCHRONOUS;
+	if (server->caps & NFS_CAP_SECURITY_LABEL)
+		fc->lsm_flags |= SECURITY_LSM_NATIVE_LABELS;
 
 	/* Get a superblock - note that we may end up sharing one that already exists */
-	s = sget(nfs_mod->nfs_fs, compare_super, nfs_set_super, flags, &sb_mntdata);
+	fc->s_fs_info = server;
+	s = sget_fc(fc, compare_super, nfs_set_super);
+	fc->s_fs_info = NULL;
 	if (IS_ERR(s)) {
-		mntroot = ERR_CAST(s);
+		error = PTR_ERR(s);
+		nfs_errorf(fc, "NFS: Couldn't get superblock");
 		goto out_err_nosb;
 	}
 
@@ -2690,88 +1249,42 @@ struct dentry *nfs_fs_mount_common(struct nfs_server *server,
 	} else {
 		error = super_setup_bdi_name(s, "%u:%u", MAJOR(server->s_dev),
 					     MINOR(server->s_dev));
-		if (error) {
-			mntroot = ERR_PTR(error);
+		if (error)
 			goto error_splat_super;
-		}
 		nfs_set_readahead(s->s_bdi, server->rpages);
 		server->super = s;
 	}
 
 	if (!s->s_root) {
+		unsigned bsize = ctx->clone_data.inherited_bsize;
 		/* initial superblock/root creation */
-		mount_info->fill_super(s, mount_info);
-		nfs_get_cache_cookie(s, mount_info->parsed, mount_info->cloned);
-		if (!(server->flags & NFS_MOUNT_UNSHARED))
-			s->s_iflags |= SB_I_MULTIROOT;
+		nfs_fill_super(s, ctx);
+		if (bsize) {
+			s->s_blocksize_bits = bsize;
+			s->s_blocksize = 1U << bsize;
+		}
+		nfs_get_cache_cookie(s, ctx);
 	}
 
-	mntroot = nfs_get_root(s, mount_info->mntfh, dev_name);
-	if (IS_ERR(mntroot))
+	error = nfs_get_root(s, fc);
+	if (error < 0) {
+		nfs_errorf(fc, "NFS: Couldn't get root dentry");
 		goto error_splat_super;
-
-	error = mount_info->set_security(s, mntroot, mount_info);
-	if (error)
-		goto error_splat_root;
+	}
 
 	s->s_flags |= SB_ACTIVE;
+	error = 0;
 
 out:
-	return mntroot;
+	return error;
 
 out_err_nosb:
 	nfs_free_server(server);
 	goto out;
-
-error_splat_root:
-	dput(mntroot);
-	mntroot = ERR_PTR(error);
 error_splat_super:
 	deactivate_locked_super(s);
 	goto out;
 }
-EXPORT_SYMBOL_GPL(nfs_fs_mount_common);
-
-struct dentry *nfs_fs_mount(struct file_system_type *fs_type,
-	int flags, const char *dev_name, void *raw_data)
-{
-	struct nfs_mount_info mount_info = {
-		.fill_super = nfs_fill_super,
-		.set_security = nfs_set_sb_security,
-	};
-	struct dentry *mntroot = ERR_PTR(-ENOMEM);
-	struct nfs_subversion *nfs_mod;
-	int error;
-
-	mount_info.parsed = nfs_alloc_parsed_mount_data();
-	mount_info.mntfh = nfs_alloc_fhandle();
-	if (mount_info.parsed == NULL || mount_info.mntfh == NULL)
-		goto out;
-
-	/* Validate the mount data */
-	error = nfs_validate_mount_data(fs_type, raw_data, mount_info.parsed, mount_info.mntfh, dev_name);
-	if (error == NFS_TEXT_DATA)
-		error = nfs_validate_text_mount_data(raw_data, mount_info.parsed, dev_name);
-	if (error < 0) {
-		mntroot = ERR_PTR(error);
-		goto out;
-	}
-
-	nfs_mod = get_nfs_version(mount_info.parsed->version);
-	if (IS_ERR(nfs_mod)) {
-		mntroot = ERR_CAST(nfs_mod);
-		goto out;
-	}
-
-	mntroot = nfs_mod->rpc_ops->try_mount(flags, dev_name, &mount_info, nfs_mod);
-
-	put_nfs_version(nfs_mod);
-out:
-	nfs_free_parsed_mount_data(mount_info.parsed);
-	nfs_free_fhandle(mount_info.mntfh);
-	return mntroot;
-}
-EXPORT_SYMBOL_GPL(nfs_fs_mount);
 
 /*
  * Destroy an NFS2/3 superblock
@@ -2790,150 +1303,8 @@ void nfs_kill_super(struct super_block *s)
 }
 EXPORT_SYMBOL_GPL(nfs_kill_super);
 
-/*
- * Clone an NFS2/3/4 server record on xdev traversal (FSID-change)
- */
-static struct dentry *
-nfs_xdev_mount(struct file_system_type *fs_type, int flags,
-		const char *dev_name, void *raw_data)
-{
-	struct nfs_clone_mount *data = raw_data;
-	struct nfs_mount_info mount_info = {
-		.fill_super = nfs_clone_super,
-		.set_security = nfs_clone_sb_security,
-		.cloned = data,
-	};
-	struct nfs_server *server;
-	struct dentry *mntroot = ERR_PTR(-ENOMEM);
-	struct nfs_subversion *nfs_mod = NFS_SB(data->sb)->nfs_client->cl_nfs_mod;
-
-	dprintk("--> nfs_xdev_mount()\n");
-
-	mount_info.mntfh = mount_info.cloned->fh;
-
-	/* create a new volume representation */
-	server = nfs_mod->rpc_ops->clone_server(NFS_SB(data->sb), data->fh, data->fattr, data->authflavor);
-
-	if (IS_ERR(server))
-		mntroot = ERR_CAST(server);
-	else
-		mntroot = nfs_fs_mount_common(server, flags,
-				dev_name, &mount_info, nfs_mod);
-
-	dprintk("<-- nfs_xdev_mount() = %ld\n",
-			IS_ERR(mntroot) ? PTR_ERR(mntroot) : 0L);
-	return mntroot;
-}
-
 #if IS_ENABLED(CONFIG_NFS_V4)
 
-static void nfs4_validate_mount_flags(struct nfs_parsed_mount_data *args)
-{
-	args->flags &= ~(NFS_MOUNT_NONLM|NFS_MOUNT_NOACL|NFS_MOUNT_VER3|
-			 NFS_MOUNT_LOCAL_FLOCK|NFS_MOUNT_LOCAL_FCNTL);
-}
-
-/*
- * Validate NFSv4 mount options
- */
-static int nfs4_validate_mount_data(void *options,
-				    struct nfs_parsed_mount_data *args,
-				    const char *dev_name)
-{
-	struct sockaddr *sap = (struct sockaddr *)&args->nfs_server.address;
-	struct nfs4_mount_data *data = (struct nfs4_mount_data *)options;
-	char *c;
-
-	if (data == NULL)
-		goto out_no_data;
-
-	args->version = 4;
-
-	switch (data->version) {
-	case 1:
-		if (data->host_addrlen > sizeof(args->nfs_server.address))
-			goto out_no_address;
-		if (data->host_addrlen == 0)
-			goto out_no_address;
-		args->nfs_server.addrlen = data->host_addrlen;
-		if (copy_from_user(sap, data->host_addr, data->host_addrlen))
-			return -EFAULT;
-		if (!nfs_verify_server_address(sap))
-			goto out_no_address;
-		args->nfs_server.port = ntohs(((struct sockaddr_in *)sap)->sin_port);
-
-		if (data->auth_flavourlen) {
-			rpc_authflavor_t pseudoflavor;
-			if (data->auth_flavourlen > 1)
-				goto out_inval_auth;
-			if (copy_from_user(&pseudoflavor,
-					   data->auth_flavours,
-					   sizeof(pseudoflavor)))
-				return -EFAULT;
-			args->selected_flavor = pseudoflavor;
-		} else
-			args->selected_flavor = RPC_AUTH_UNIX;
-
-		c = strndup_user(data->hostname.data, NFS4_MAXNAMLEN);
-		if (IS_ERR(c))
-			return PTR_ERR(c);
-		args->nfs_server.hostname = c;
-
-		c = strndup_user(data->mnt_path.data, NFS4_MAXPATHLEN);
-		if (IS_ERR(c))
-			return PTR_ERR(c);
-		args->nfs_server.export_path = c;
-		dfprintk(MOUNT, "NFS: MNTPATH: '%s'\n", c);
-
-		c = strndup_user(data->client_addr.data, 16);
-		if (IS_ERR(c))
-			return PTR_ERR(c);
-		args->client_address = c;
-
-		/*
-		 * Translate to nfs_parsed_mount_data, which nfs4_fill_super
-		 * can deal with.
-		 */
-
-		args->flags	= data->flags & NFS4_MOUNT_FLAGMASK;
-		args->rsize	= data->rsize;
-		args->wsize	= data->wsize;
-		args->timeo	= data->timeo;
-		args->retrans	= data->retrans;
-		args->acregmin	= data->acregmin;
-		args->acregmax	= data->acregmax;
-		args->acdirmin	= data->acdirmin;
-		args->acdirmax	= data->acdirmax;
-		args->nfs_server.protocol = data->proto;
-		nfs_validate_transport_protocol(args);
-		if (args->nfs_server.protocol == XPRT_TRANSPORT_UDP)
-			goto out_invalid_transport_udp;
-
-		break;
-	default:
-		return NFS_TEXT_DATA;
-	}
-
-	return 0;
-
-out_no_data:
-	dfprintk(MOUNT, "NFS4: mount program didn't pass any mount data\n");
-	return -EINVAL;
-
-out_inval_auth:
-	dfprintk(MOUNT, "NFS4: Invalid number of RPC auth flavours %d\n",
-		 data->auth_flavourlen);
-	return -EINVAL;
-
-out_no_address:
-	dfprintk(MOUNT, "NFS4: mount program didn't pass remote address\n");
-	return -EINVAL;
-
-out_invalid_transport_udp:
-	dfprintk(MOUNT, "NFSv4: Unsupported transport protocol udp\n");
-	return -EINVAL;
-}
-
 /*
  * NFS v4 module parameters need to stay in the
  * NFS client for backwards compatibility
diff --git a/fs/nfs/sysfs.c b/fs/nfs/sysfs.c
index 4f3390b20239..c489496b5659 100644
--- a/fs/nfs/sysfs.c
+++ b/fs/nfs/sysfs.c
@@ -121,8 +121,7 @@ static void nfs_netns_client_release(struct kobject *kobj)
 			struct nfs_netns_client,
 			kobject);
 
-	if (c->identifier)
-		kfree(c->identifier);
+	kfree(c->identifier);
 	kfree(c);
 }
 
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 0effeee28352..b27ebdccef70 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -98,7 +98,7 @@ static void nfs_do_call_unlink(struct inode *inode, struct nfs_unlinkdata *data)
 		.callback_ops = &nfs_unlink_ops,
 		.callback_data = data,
 		.workqueue = nfsiod_workqueue,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 	struct rpc_task *task;
 	struct inode *dir = d_inode(data->dentry->d_parent);
@@ -341,7 +341,7 @@ nfs_async_rename(struct inode *old_dir, struct inode *new_dir,
 		.callback_ops = &nfs_rename_ops,
 		.workqueue = nfsiod_workqueue,
 		.rpc_client = NFS_CLIENT(old_dir),
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF,
 	};
 
 	data = kzalloc(sizeof(*data), GFP_KERNEL);
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 52cab65f91cf..639c34fec04a 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -149,6 +149,31 @@ static void nfs_io_completion_put(struct nfs_io_completion *ioc)
 		kref_put(&ioc->refcount, nfs_io_completion_release);
 }
 
+static void
+nfs_page_set_inode_ref(struct nfs_page *req, struct inode *inode)
+{
+	if (!test_and_set_bit(PG_INODE_REF, &req->wb_flags)) {
+		kref_get(&req->wb_kref);
+		atomic_long_inc(&NFS_I(inode)->nrequests);
+	}
+}
+
+static int
+nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode)
+{
+	int ret;
+
+	if (!test_bit(PG_REMOVE, &req->wb_flags))
+		return 0;
+	ret = nfs_page_group_lock(req);
+	if (ret)
+		return ret;
+	if (test_and_clear_bit(PG_REMOVE, &req->wb_flags))
+		nfs_page_set_inode_ref(req, inode);
+	nfs_page_group_unlock(req);
+	return 0;
+}
+
 static struct nfs_page *
 nfs_page_private_request(struct page *page)
 {
@@ -218,6 +243,36 @@ static struct nfs_page *nfs_page_find_head_request(struct page *page)
 	return req;
 }
 
+static struct nfs_page *nfs_find_and_lock_page_request(struct page *page)
+{
+	struct inode *inode = page_file_mapping(page)->host;
+	struct nfs_page *req, *head;
+	int ret;
+
+	for (;;) {
+		req = nfs_page_find_head_request(page);
+		if (!req)
+			return req;
+		head = nfs_page_group_lock_head(req);
+		if (head != req)
+			nfs_release_request(req);
+		if (IS_ERR(head))
+			return head;
+		ret = nfs_cancel_remove_inode(head, inode);
+		if (ret < 0) {
+			nfs_unlock_and_release_request(head);
+			return ERR_PTR(ret);
+		}
+		/* Ensure that nobody removed the request before we locked it */
+		if (head == nfs_page_private_request(page))
+			break;
+		if (PageSwapCache(page))
+			break;
+		nfs_unlock_and_release_request(head);
+	}
+	return head;
+}
+
 /* Adjust the file length if we're writing beyond the end */
 static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int count)
 {
@@ -243,13 +298,24 @@ out:
 /* A writeback failed: mark the page as bad, and invalidate the page cache */
 static void nfs_set_pageerror(struct address_space *mapping)
 {
+	struct inode *inode = mapping->host;
+
 	nfs_zap_mapping(mapping->host, mapping);
+	/* Force file size revalidation */
+	spin_lock(&inode->i_lock);
+	NFS_I(inode)->cache_validity |= NFS_INO_REVAL_FORCED |
+					NFS_INO_REVAL_PAGECACHE |
+					NFS_INO_INVALID_SIZE;
+	spin_unlock(&inode->i_lock);
 }
 
 static void nfs_mapping_set_error(struct page *page, int error)
 {
+	struct address_space *mapping = page_file_mapping(page);
+
 	SetPageError(page);
-	mapping_set_error(page_file_mapping(page), error);
+	mapping_set_error(mapping, error);
+	nfs_set_pageerror(mapping);
 }
 
 /*
@@ -369,34 +435,6 @@ static void nfs_end_page_writeback(struct nfs_page *req)
 }
 
 /*
- * nfs_unroll_locks_and_wait -  unlock all newly locked reqs and wait on @req
- *
- * this is a helper function for nfs_lock_and_join_requests
- *
- * @inode - inode associated with request page group, must be holding inode lock
- * @head  - head request of page group, must be holding head lock
- * @req   - request that couldn't lock and needs to wait on the req bit lock
- *
- * NOTE: this must be called holding page_group bit lock
- *       which will be released before returning.
- *
- * returns 0 on success, < 0 on error.
- */
-static void
-nfs_unroll_locks(struct inode *inode, struct nfs_page *head,
-			  struct nfs_page *req)
-{
-	struct nfs_page *tmp;
-
-	/* relinquish all the locks successfully grabbed this run */
-	for (tmp = head->wb_this_page ; tmp != req; tmp = tmp->wb_this_page) {
-		if (!kref_read(&tmp->wb_kref))
-			continue;
-		nfs_unlock_and_release_request(tmp);
-	}
-}
-
-/*
  * nfs_destroy_unlinked_subrequests - destroy recently unlinked subrequests
  *
  * @destroy_list - request list (using wb_this_page) terminated by @old_head
@@ -417,22 +455,29 @@ nfs_destroy_unlinked_subrequests(struct nfs_page *destroy_list,
 		destroy_list = (subreq->wb_this_page == old_head) ?
 				   NULL : subreq->wb_this_page;
 
+		/* Note: lock subreq in order to change subreq->wb_head */
+		nfs_page_set_headlock(subreq);
 		WARN_ON_ONCE(old_head != subreq->wb_head);
 
 		/* make sure old group is not used */
 		subreq->wb_this_page = subreq;
+		subreq->wb_head = subreq;
 
 		clear_bit(PG_REMOVE, &subreq->wb_flags);
 
 		/* Note: races with nfs_page_group_destroy() */
 		if (!kref_read(&subreq->wb_kref)) {
 			/* Check if we raced with nfs_page_group_destroy() */
-			if (test_and_clear_bit(PG_TEARDOWN, &subreq->wb_flags))
+			if (test_and_clear_bit(PG_TEARDOWN, &subreq->wb_flags)) {
+				nfs_page_clear_headlock(subreq);
 				nfs_free_request(subreq);
+			} else
+				nfs_page_clear_headlock(subreq);
 			continue;
 		}
+		nfs_page_clear_headlock(subreq);
 
-		subreq->wb_head = subreq;
+		nfs_release_request(old_head);
 
 		if (test_and_clear_bit(PG_INODE_REF, &subreq->wb_flags)) {
 			nfs_release_request(subreq);
@@ -446,105 +491,43 @@ nfs_destroy_unlinked_subrequests(struct nfs_page *destroy_list,
 }
 
 /*
- * nfs_lock_and_join_requests - join all subreqs to the head req and return
- *                              a locked reference, cancelling any pending
- *                              operations for this page.
- *
- * @page - the page used to lookup the "page group" of nfs_page structures
+ * nfs_join_page_group - destroy subrequests of the head req
+ * @head: the page used to lookup the "page group" of nfs_page structures
+ * @inode: Inode to which the request belongs.
  *
  * This function joins all sub requests to the head request by first
  * locking all requests in the group, cancelling any pending operations
  * and finally updating the head request to cover the whole range covered by
  * the (former) group.  All subrequests are removed from any write or commit
  * lists, unlinked from the group and destroyed.
- *
- * Returns a locked, referenced pointer to the head request - which after
- * this call is guaranteed to be the only request associated with the page.
- * Returns NULL if no requests are found for @page, or a ERR_PTR if an
- * error was encountered.
  */
-static struct nfs_page *
-nfs_lock_and_join_requests(struct page *page)
+void
+nfs_join_page_group(struct nfs_page *head, struct inode *inode)
 {
-	struct inode *inode = page_file_mapping(page)->host;
-	struct nfs_page *head, *subreq;
+	struct nfs_page *subreq;
 	struct nfs_page *destroy_list = NULL;
-	unsigned int total_bytes;
-	int ret;
-
-try_again:
-	/*
-	 * A reference is taken only on the head request which acts as a
-	 * reference to the whole page group - the group will not be destroyed
-	 * until the head reference is released.
-	 */
-	head = nfs_page_find_head_request(page);
-	if (!head)
-		return NULL;
-
-	/* lock the page head first in order to avoid an ABBA inefficiency */
-	if (!nfs_lock_request(head)) {
-		ret = nfs_wait_on_request(head);
-		nfs_release_request(head);
-		if (ret < 0)
-			return ERR_PTR(ret);
-		goto try_again;
-	}
-
-	/* Ensure that nobody removed the request before we locked it */
-	if (head != nfs_page_private_request(page) && !PageSwapCache(page)) {
-		nfs_unlock_and_release_request(head);
-		goto try_again;
-	}
+	unsigned int pgbase, off, bytes;
 
-	ret = nfs_page_group_lock(head);
-	if (ret < 0)
-		goto release_request;
-
-	/* lock each request in the page group */
-	total_bytes = head->wb_bytes;
+	pgbase = head->wb_pgbase;
+	bytes = head->wb_bytes;
+	off = head->wb_offset;
 	for (subreq = head->wb_this_page; subreq != head;
 			subreq = subreq->wb_this_page) {
-
-		if (!kref_get_unless_zero(&subreq->wb_kref)) {
-			if (subreq->wb_offset == head->wb_offset + total_bytes)
-				total_bytes += subreq->wb_bytes;
-			continue;
-		}
-
-		while (!nfs_lock_request(subreq)) {
-			/*
-			 * Unlock page to allow nfs_page_group_sync_on_bit()
-			 * to succeed
-			 */
-			nfs_page_group_unlock(head);
-			ret = nfs_wait_on_request(subreq);
-			if (!ret)
-				ret = nfs_page_group_lock(head);
-			if (ret < 0) {
-				nfs_unroll_locks(inode, head, subreq);
-				nfs_release_request(subreq);
-				goto release_request;
-			}
-		}
-		/*
-		 * Subrequests are always contiguous, non overlapping
-		 * and in order - but may be repeated (mirrored writes).
-		 */
-		if (subreq->wb_offset == (head->wb_offset + total_bytes)) {
-			/* keep track of how many bytes this group covers */
-			total_bytes += subreq->wb_bytes;
-		} else if (WARN_ON_ONCE(subreq->wb_offset < head->wb_offset ||
-			    ((subreq->wb_offset + subreq->wb_bytes) >
-			     (head->wb_offset + total_bytes)))) {
-			nfs_page_group_unlock(head);
-			nfs_unroll_locks(inode, head, subreq);
-			nfs_unlock_and_release_request(subreq);
-			ret = -EIO;
-			goto release_request;
+		/* Subrequests should always form a contiguous range */
+		if (pgbase > subreq->wb_pgbase) {
+			off -= pgbase - subreq->wb_pgbase;
+			bytes += pgbase - subreq->wb_pgbase;
+			pgbase = subreq->wb_pgbase;
 		}
+		bytes = max(subreq->wb_pgbase + subreq->wb_bytes
+				- pgbase, bytes);
 	}
 
+	/* Set the head request's range to cover the former page group */
+	head->wb_pgbase = pgbase;
+	head->wb_bytes = bytes;
+	head->wb_offset = off;
+
 	/* Now that all requests are locked, make sure they aren't on any list.
 	 * Commit list removal accounting is done after locks are dropped */
 	subreq = head;
@@ -558,41 +541,57 @@ try_again:
 		/* destroy list will be terminated by head */
 		destroy_list = head->wb_this_page;
 		head->wb_this_page = head;
-
-		/* change head request to cover whole range that
-		 * the former page group covered */
-		head->wb_bytes = total_bytes;
 	}
 
-	/* Postpone destruction of this request */
-	if (test_and_clear_bit(PG_REMOVE, &head->wb_flags)) {
-		set_bit(PG_INODE_REF, &head->wb_flags);
-		kref_get(&head->wb_kref);
-		atomic_long_inc(&NFS_I(inode)->nrequests);
-	}
+	nfs_destroy_unlinked_subrequests(destroy_list, head, inode);
+}
 
-	nfs_page_group_unlock(head);
+/*
+ * nfs_lock_and_join_requests - join all subreqs to the head req
+ * @page: the page used to lookup the "page group" of nfs_page structures
+ *
+ * This function joins all sub requests to the head request by first
+ * locking all requests in the group, cancelling any pending operations
+ * and finally updating the head request to cover the whole range covered by
+ * the (former) group.  All subrequests are removed from any write or commit
+ * lists, unlinked from the group and destroyed.
+ *
+ * Returns a locked, referenced pointer to the head request - which after
+ * this call is guaranteed to be the only request associated with the page.
+ * Returns NULL if no requests are found for @page, or a ERR_PTR if an
+ * error was encountered.
+ */
+static struct nfs_page *
+nfs_lock_and_join_requests(struct page *page)
+{
+	struct inode *inode = page_file_mapping(page)->host;
+	struct nfs_page *head;
+	int ret;
 
-	nfs_destroy_unlinked_subrequests(destroy_list, head, inode);
+	/*
+	 * A reference is taken only on the head request which acts as a
+	 * reference to the whole page group - the group will not be destroyed
+	 * until the head reference is released.
+	 */
+	head = nfs_find_and_lock_page_request(page);
+	if (IS_ERR_OR_NULL(head))
+		return head;
 
-	/* Did we lose a race with nfs_inode_remove_request()? */
-	if (!(PagePrivate(page) || PageSwapCache(page))) {
+	/* lock each request in the page group */
+	ret = nfs_page_group_lock_subrequests(head);
+	if (ret < 0) {
 		nfs_unlock_and_release_request(head);
-		return NULL;
+		return ERR_PTR(ret);
 	}
 
-	/* still holds ref on head from nfs_page_find_head_request
-	 * and still has lock on head from lock loop */
-	return head;
+	nfs_join_page_group(head, inode);
 
-release_request:
-	nfs_unlock_and_release_request(head);
-	return ERR_PTR(ret);
+	return head;
 }
 
 static void nfs_write_error(struct nfs_page *req, int error)
 {
-	nfs_set_pageerror(page_file_mapping(req->wb_page));
+	trace_nfs_write_error(req, error);
 	nfs_mapping_set_error(req->wb_page, error);
 	nfs_inode_remove_request(req);
 	nfs_end_page_writeback(req);
@@ -947,9 +946,9 @@ nfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg,
 static void
 nfs_clear_page_commit(struct page *page)
 {
-	dec_node_page_state(page, NR_UNSTABLE_NFS);
+	dec_node_page_state(page, NR_WRITEBACK);
 	dec_wb_stat(&inode_to_bdi(page_file_mapping(page)->host)->wb,
-		    WB_RECLAIMABLE);
+		    WB_WRITEBACK);
 }
 
 /* Called holding the request lock on @req */
@@ -998,7 +997,7 @@ static void nfs_write_completion(struct nfs_pgio_header *hdr)
 		nfs_list_remove_request(req);
 		if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) &&
 		    (hdr->good_bytes < bytes)) {
-			nfs_set_pageerror(page_file_mapping(req->wb_page));
+			trace_nfs_comp_error(req, hdr->error);
 			nfs_mapping_set_error(req->wb_page, hdr->error);
 			goto remove_req;
 		}
@@ -1403,8 +1402,7 @@ static void nfs_initiate_write(struct nfs_pgio_header *hdr,
 
 	task_setup_data->priority = priority;
 	rpc_ops->write_setup(hdr, msg, &task_setup_data->rpc_client);
-	trace_nfs_initiate_write(hdr->inode, hdr->io_start, hdr->good_bytes,
-				 hdr->args.stable);
+	trace_nfs_initiate_write(hdr);
 }
 
 /* If a nfs_flush_* function fails, it should remove reqs from @head and
@@ -1568,8 +1566,7 @@ static int nfs_writeback_done(struct rpc_task *task,
 		return status;
 
 	nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, hdr->res.count);
-	trace_nfs_writeback_done(inode, task->tk_status,
-				 hdr->args.offset, hdr->res.verf);
+	trace_nfs_writeback_done(task, hdr);
 
 	if (hdr->res.verf->committed < hdr->args.stable &&
 	    task->tk_status >= 0) {
@@ -1649,6 +1646,8 @@ static void nfs_writeback_result(struct rpc_task *task,
 			 */
 			argp->stable = NFS_FILE_SYNC;
 		}
+		resp->count = 0;
+		resp->verf->committed = 0;
 		rpc_restart_call_prepare(task);
 	}
 }
@@ -1735,14 +1734,19 @@ void nfs_init_commit(struct nfs_commit_data *data,
 		     struct pnfs_layout_segment *lseg,
 		     struct nfs_commit_info *cinfo)
 {
-	struct nfs_page *first = nfs_list_entry(head->next);
-	struct nfs_open_context *ctx = nfs_req_openctx(first);
-	struct inode *inode = d_inode(ctx->dentry);
+	struct nfs_page *first;
+	struct nfs_open_context *ctx;
+	struct inode *inode;
 
 	/* Set up the RPC argument and reply structs
 	 * NB: take care not to mess about with data->commit et al. */
 
-	list_splice_init(head, &data->pages);
+	if (head)
+		list_splice_init(head, &data->pages);
+
+	first = nfs_list_entry(data->pages.next);
+	ctx = nfs_req_openctx(first);
+	inode = d_inode(ctx->dentry);
 
 	data->inode	  = inode;
 	data->cred	  = ctx->cred;
@@ -1809,7 +1813,7 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how,
 	nfs_init_commit(data, head, NULL, cinfo);
 	atomic_inc(&cinfo->mds->rpcs_out);
 	return nfs_initiate_commit(NFS_CLIENT(inode), data, NFS_PROTO(inode),
-				   data->mds_ops, how, 0);
+				   data->mds_ops, how, RPC_TASK_CRED_NOREF);
 }
 
 /*
@@ -1824,11 +1828,12 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata)
 
 	/* Call the NFS version-specific code */
 	NFS_PROTO(data->inode)->commit_done(task, data);
-	trace_nfs_commit_done(data);
+	trace_nfs_commit_done(task, data);
 }
 
 static void nfs_commit_release_pages(struct nfs_commit_data *data)
 {
+	const struct nfs_writeverf *verf = data->res.verf;
 	struct nfs_page	*req;
 	int status = data->task.tk_status;
 	struct nfs_commit_info cinfo;
@@ -1847,6 +1852,7 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data)
 			(long long)req_offset(req));
 		if (status < 0) {
 			if (req->wb_page) {
+				trace_nfs_commit_error(req, status);
 				nfs_mapping_set_error(req->wb_page, status);
 				nfs_inode_remove_request(req);
 			}
@@ -1856,7 +1862,7 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data)
 
 		/* Okay, COMMIT succeeded, apparently. Check the verifier
 		 * returned by the server against all stored verfs. */
-		if (!nfs_write_verifier_cmp(&req->wb_verf, &data->verf.verifier)) {
+		if (nfs_write_match_verf(verf, req)) {
 			/* We have a match */
 			if (req->wb_page)
 				nfs_inode_remove_request(req);
diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
index 10cefb0c07c7..99d2cae91bd6 100644
--- a/fs/nfsd/Kconfig
+++ b/fs/nfsd/Kconfig
@@ -73,7 +73,8 @@ config NFSD_V4
 	select NFSD_V3
 	select FS_POSIX_ACL
 	select SUNRPC_GSS
-	select CRYPTO
+	select CRYPTO_MD5
+	select CRYPTO_SHA256
 	select GRACE_PERIOD
 	help
 	  This option enables support in your system's NFS server for
@@ -133,6 +134,16 @@ config NFSD_FLEXFILELAYOUT
 
 	  If unsure, say N.
 
+config NFSD_V4_2_INTER_SSC
+	bool "NFSv4.2 inter server to server COPY"
+	depends on NFSD_V4 && NFS_V4_1 && NFS_V4_2 && NFS_FS=y
+	help
+	  This option enables support for NFSv4.2 inter server to
+	  server copy where the destination server calls the NFSv4.2
+	  client to read the data to copy from the source server.
+
+	  If unsure, say N.
+
 config NFSD_V4_SECURITY_LABEL
 	bool "Provide Security Label support for NFSv4 server"
 	depends on NFSD_V4 && SECURITY
diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index 15422c951fd1..cb777fe82988 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -23,6 +23,7 @@
 #include "netns.h"
 #include "pnfs.h"
 #include "filecache.h"
+#include "trace.h"
 
 #define NFSDDBG_FACILITY	NFSDDBG_EXPORT
 
@@ -50,6 +51,11 @@ static void expkey_put(struct kref *ref)
 	kfree_rcu(key, ek_rcu);
 }
 
+static int expkey_upcall(struct cache_detail *cd, struct cache_head *h)
+{
+	return sunrpc_cache_pipe_upcall(cd, h);
+}
+
 static void expkey_request(struct cache_detail *cd,
 			   struct cache_head *h,
 			   char **bpp, int *blen)
@@ -140,7 +146,9 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen)
 	if (len == 0) {
 		set_bit(CACHE_NEGATIVE, &key.h.flags);
 		ek = svc_expkey_update(cd, &key, ek);
-		if (!ek)
+		if (ek)
+			trace_nfsd_expkey_update(ek, NULL);
+		else
 			err = -ENOMEM;
 	} else {
 		err = kern_path(buf, 0, &key.ek_path);
@@ -150,7 +158,9 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen)
 		dprintk("Found the path %s\n", buf);
 
 		ek = svc_expkey_update(cd, &key, ek);
-		if (!ek)
+		if (ek)
+			trace_nfsd_expkey_update(ek, buf);
+		else
 			err = -ENOMEM;
 		path_put(&key.ek_path);
 	}
@@ -249,6 +259,7 @@ static const struct cache_detail svc_expkey_cache_template = {
 	.hash_size	= EXPKEY_HASHMAX,
 	.name		= "nfsd.fh",
 	.cache_put	= expkey_put,
+	.cache_upcall	= expkey_upcall,
 	.cache_request	= expkey_request,
 	.cache_parse	= expkey_parse,
 	.cache_show	= expkey_show,
@@ -330,6 +341,11 @@ static void svc_export_put(struct kref *ref)
 	kfree_rcu(exp, ex_rcu);
 }
 
+static int svc_export_upcall(struct cache_detail *cd, struct cache_head *h)
+{
+	return sunrpc_cache_pipe_upcall(cd, h);
+}
+
 static void svc_export_request(struct cache_detail *cd,
 			       struct cache_head *h,
 			       char **bpp, int *blen)
@@ -643,15 +659,17 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen)
 	}
 
 	expp = svc_export_lookup(&exp);
-	if (expp)
-		expp = svc_export_update(&exp, expp);
-	else
-		err = -ENOMEM;
-	cache_flush();
-	if (expp == NULL)
+	if (!expp) {
 		err = -ENOMEM;
-	else
+		goto out4;
+	}
+	expp = svc_export_update(&exp, expp);
+	if (expp) {
+		trace_nfsd_export_update(expp);
+		cache_flush();
 		exp_put(expp);
+	} else
+		err = -ENOMEM;
 out4:
 	nfsd4_fslocs_free(&exp.ex_fslocs);
 	kfree(exp.ex_uuid);
@@ -767,6 +785,7 @@ static const struct cache_detail svc_export_cache_template = {
 	.hash_size	= EXPORT_HASHMAX,
 	.name		= "nfsd.export",
 	.cache_put	= svc_export_put,
+	.cache_upcall	= svc_export_upcall,
 	.cache_request	= svc_export_request,
 	.cache_parse	= svc_export_parse,
 	.cache_show	= svc_export_show,
@@ -832,8 +851,10 @@ exp_find_key(struct cache_detail *cd, struct auth_domain *clp, int fsid_type,
 	if (ek == NULL)
 		return ERR_PTR(-ENOMEM);
 	err = cache_check(cd, &ek->h, reqp);
-	if (err)
+	if (err) {
+		trace_nfsd_exp_find_key(&key, err);
 		return ERR_PTR(err);
+	}
 	return ek;
 }
 
@@ -855,8 +876,10 @@ exp_get_by_name(struct cache_detail *cd, struct auth_domain *clp,
 	if (exp == NULL)
 		return ERR_PTR(-ENOMEM);
 	err = cache_check(cd, &exp->h, reqp);
-	if (err)
+	if (err) {
+		trace_nfsd_exp_get_by_name(&key, err);
 		return ERR_PTR(err);
+	}
 	return exp;
 }
 
diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
index ef55e9b1cd4e..82198d747c4c 100644
--- a/fs/nfsd/filecache.c
+++ b/fs/nfsd/filecache.c
@@ -27,7 +27,6 @@
 #define NFSD_FILE_HASH_SIZE                  (1 << NFSD_FILE_HASH_BITS)
 #define NFSD_LAUNDRETTE_DELAY		     (2 * HZ)
 
-#define NFSD_FILE_LRU_RESCAN		     (0)
 #define NFSD_FILE_SHUTDOWN		     (1)
 #define NFSD_FILE_LRU_THRESHOLD		     (4096UL)
 #define NFSD_FILE_LRU_LIMIT		     (NFSD_FILE_LRU_THRESHOLD << 2)
@@ -44,6 +43,17 @@ struct nfsd_fcache_bucket {
 
 static DEFINE_PER_CPU(unsigned long, nfsd_file_cache_hits);
 
+struct nfsd_fcache_disposal {
+	struct list_head list;
+	struct work_struct work;
+	struct net *net;
+	spinlock_t lock;
+	struct list_head freeme;
+	struct rcu_head rcu;
+};
+
+static struct workqueue_struct *nfsd_filecache_wq __read_mostly;
+
 static struct kmem_cache		*nfsd_file_slab;
 static struct kmem_cache		*nfsd_file_mark_slab;
 static struct nfsd_fcache_bucket	*nfsd_file_hashtbl;
@@ -52,32 +62,21 @@ static long				nfsd_file_lru_flags;
 static struct fsnotify_group		*nfsd_file_fsnotify_group;
 static atomic_long_t			nfsd_filecache_count;
 static struct delayed_work		nfsd_filecache_laundrette;
+static DEFINE_SPINLOCK(laundrette_lock);
+static LIST_HEAD(laundrettes);
 
-enum nfsd_file_laundrette_ctl {
-	NFSD_FILE_LAUNDRETTE_NOFLUSH = 0,
-	NFSD_FILE_LAUNDRETTE_MAY_FLUSH
-};
+static void nfsd_file_gc(void);
 
 static void
-nfsd_file_schedule_laundrette(enum nfsd_file_laundrette_ctl ctl)
+nfsd_file_schedule_laundrette(void)
 {
 	long count = atomic_long_read(&nfsd_filecache_count);
 
 	if (count == 0 || test_bit(NFSD_FILE_SHUTDOWN, &nfsd_file_lru_flags))
 		return;
 
-	/* Be more aggressive about scanning if over the threshold */
-	if (count > NFSD_FILE_LRU_THRESHOLD)
-		mod_delayed_work(system_wq, &nfsd_filecache_laundrette, 0);
-	else
-		schedule_delayed_work(&nfsd_filecache_laundrette, NFSD_LAUNDRETTE_DELAY);
-
-	if (ctl == NFSD_FILE_LAUNDRETTE_NOFLUSH)
-		return;
-
-	/* ...and don't delay flushing if we're out of control */
-	if (count >= NFSD_FILE_LRU_LIMIT)
-		flush_delayed_work(&nfsd_filecache_laundrette);
+	queue_delayed_work(system_wq, &nfsd_filecache_laundrette,
+			NFSD_LAUNDRETTE_DELAY);
 }
 
 static void
@@ -101,7 +100,7 @@ nfsd_file_mark_free(struct fsnotify_mark *mark)
 static struct nfsd_file_mark *
 nfsd_file_mark_get(struct nfsd_file_mark *nfm)
 {
-	if (!atomic_inc_not_zero(&nfm->nfm_ref))
+	if (!refcount_inc_not_zero(&nfm->nfm_ref))
 		return NULL;
 	return nfm;
 }
@@ -109,8 +108,7 @@ nfsd_file_mark_get(struct nfsd_file_mark *nfm)
 static void
 nfsd_file_mark_put(struct nfsd_file_mark *nfm)
 {
-	if (atomic_dec_and_test(&nfm->nfm_ref)) {
-
+	if (refcount_dec_and_test(&nfm->nfm_ref)) {
 		fsnotify_destroy_mark(&nfm->nfm_mark, nfsd_file_fsnotify_group);
 		fsnotify_put_mark(&nfm->nfm_mark);
 	}
@@ -133,9 +131,13 @@ nfsd_file_mark_find_or_create(struct nfsd_file *nf)
 						 struct nfsd_file_mark,
 						 nfm_mark));
 			mutex_unlock(&nfsd_file_fsnotify_group->mark_mutex);
-			fsnotify_put_mark(mark);
-			if (likely(nfm))
+			if (nfm) {
+				fsnotify_put_mark(mark);
 				break;
+			}
+			/* Avoid soft lockup race with nfsd_file_mark_put() */
+			fsnotify_destroy_mark(mark, nfsd_file_fsnotify_group);
+			fsnotify_put_mark(mark);
 		} else
 			mutex_unlock(&nfsd_file_fsnotify_group->mark_mutex);
 
@@ -145,7 +147,7 @@ nfsd_file_mark_find_or_create(struct nfsd_file *nf)
 			return NULL;
 		fsnotify_init_mark(&new->nfm_mark, nfsd_file_fsnotify_group);
 		new->nfm_mark.mask = FS_ATTRIB|FS_DELETE_SELF;
-		atomic_set(&new->nfm_ref, 1);
+		refcount_set(&new->nfm_ref, 1);
 
 		err = fsnotify_add_inode_mark(&new->nfm_mark, inode, 0);
 
@@ -183,7 +185,7 @@ nfsd_file_alloc(struct inode *inode, unsigned int may, unsigned int hashval,
 		nf->nf_flags = 0;
 		nf->nf_inode = inode;
 		nf->nf_hashval = hashval;
-		atomic_set(&nf->nf_ref, 1);
+		refcount_set(&nf->nf_ref, 1);
 		nf->nf_may = may & NFSD_FILE_MAY_MASK;
 		if (may & NFSD_MAY_NOT_BREAK_LEASE) {
 			if (may & NFSD_MAY_WRITE)
@@ -192,6 +194,7 @@ nfsd_file_alloc(struct inode *inode, unsigned int may, unsigned int hashval,
 				__set_bit(NFSD_FILE_BREAK_READ, &nf->nf_flags);
 		}
 		nf->nf_mark = NULL;
+		init_rwsem(&nf->nf_rwsem);
 		trace_nfsd_file_alloc(nf);
 	}
 	return nf;
@@ -238,13 +241,6 @@ nfsd_file_check_write_error(struct nfsd_file *nf)
 	return filemap_check_wb_err(file->f_mapping, READ_ONCE(file->f_wb_err));
 }
 
-static bool
-nfsd_file_in_use(struct nfsd_file *nf)
-{
-	return nfsd_file_check_writeback(nf) ||
-			nfsd_file_check_write_error(nf);
-}
-
 static void
 nfsd_file_do_unhash(struct nfsd_file *nf)
 {
@@ -256,8 +252,6 @@ nfsd_file_do_unhash(struct nfsd_file *nf)
 		nfsd_reset_boot_verifier(net_generic(nf->nf_net, nfsd_net_id));
 	--nfsd_file_hashtbl[nf->nf_hashval].nfb_count;
 	hlist_del_rcu(&nf->nf_node);
-	if (!list_empty(&nf->nf_lru))
-		list_lru_del(&nfsd_file_lru, &nf->nf_lru);
 	atomic_long_dec(&nfsd_filecache_count);
 }
 
@@ -266,6 +260,8 @@ nfsd_file_unhash(struct nfsd_file *nf)
 {
 	if (test_and_clear_bit(NFSD_FILE_HASHED, &nf->nf_flags)) {
 		nfsd_file_do_unhash(nf);
+		if (!list_empty(&nf->nf_lru))
+			list_lru_del(&nfsd_file_lru, &nf->nf_lru);
 		return true;
 	}
 	return false;
@@ -283,42 +279,48 @@ nfsd_file_unhash_and_release_locked(struct nfsd_file *nf, struct list_head *disp
 	if (!nfsd_file_unhash(nf))
 		return false;
 	/* keep final reference for nfsd_file_lru_dispose */
-	if (atomic_add_unless(&nf->nf_ref, -1, 1))
+	if (refcount_dec_not_one(&nf->nf_ref))
 		return true;
 
 	list_add(&nf->nf_lru, dispose);
 	return true;
 }
 
-static int
+static void
 nfsd_file_put_noref(struct nfsd_file *nf)
 {
-	int count;
 	trace_nfsd_file_put(nf);
 
-	count = atomic_dec_return(&nf->nf_ref);
-	if (!count) {
+	if (refcount_dec_and_test(&nf->nf_ref)) {
 		WARN_ON(test_bit(NFSD_FILE_HASHED, &nf->nf_flags));
 		nfsd_file_free(nf);
 	}
-	return count;
 }
 
 void
 nfsd_file_put(struct nfsd_file *nf)
 {
-	bool is_hashed = test_bit(NFSD_FILE_HASHED, &nf->nf_flags) != 0;
-	bool unused = !nfsd_file_in_use(nf);
+	bool is_hashed;
 
 	set_bit(NFSD_FILE_REFERENCED, &nf->nf_flags);
-	if (nfsd_file_put_noref(nf) == 1 && is_hashed && unused)
-		nfsd_file_schedule_laundrette(NFSD_FILE_LAUNDRETTE_MAY_FLUSH);
+	if (refcount_read(&nf->nf_ref) > 2 || !nf->nf_file) {
+		nfsd_file_put_noref(nf);
+		return;
+	}
+
+	filemap_flush(nf->nf_file->f_mapping);
+	is_hashed = test_bit(NFSD_FILE_HASHED, &nf->nf_flags) != 0;
+	nfsd_file_put_noref(nf);
+	if (is_hashed)
+		nfsd_file_schedule_laundrette();
+	if (atomic_long_read(&nfsd_filecache_count) >= NFSD_FILE_LRU_LIMIT)
+		nfsd_file_gc();
 }
 
 struct nfsd_file *
 nfsd_file_get(struct nfsd_file *nf)
 {
-	if (likely(atomic_inc_not_zero(&nf->nf_ref)))
+	if (likely(refcount_inc_not_zero(&nf->nf_ref)))
 		return nf;
 	return NULL;
 }
@@ -344,7 +346,7 @@ nfsd_file_dispose_list_sync(struct list_head *dispose)
 	while(!list_empty(dispose)) {
 		nf = list_first_entry(dispose, struct nfsd_file, nf_lru);
 		list_del(&nf->nf_lru);
-		if (!atomic_dec_and_test(&nf->nf_ref))
+		if (!refcount_dec_and_test(&nf->nf_ref))
 			continue;
 		if (nfsd_file_free(nf))
 			flush = true;
@@ -353,6 +355,58 @@ nfsd_file_dispose_list_sync(struct list_head *dispose)
 		flush_delayed_fput();
 }
 
+static void
+nfsd_file_list_remove_disposal(struct list_head *dst,
+		struct nfsd_fcache_disposal *l)
+{
+	spin_lock(&l->lock);
+	list_splice_init(&l->freeme, dst);
+	spin_unlock(&l->lock);
+}
+
+static void
+nfsd_file_list_add_disposal(struct list_head *files, struct net *net)
+{
+	struct nfsd_fcache_disposal *l;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(l, &laundrettes, list) {
+		if (l->net == net) {
+			spin_lock(&l->lock);
+			list_splice_tail_init(files, &l->freeme);
+			spin_unlock(&l->lock);
+			queue_work(nfsd_filecache_wq, &l->work);
+			break;
+		}
+	}
+	rcu_read_unlock();
+}
+
+static void
+nfsd_file_list_add_pernet(struct list_head *dst, struct list_head *src,
+		struct net *net)
+{
+	struct nfsd_file *nf, *tmp;
+
+	list_for_each_entry_safe(nf, tmp, src, nf_lru) {
+		if (nf->nf_net == net)
+			list_move_tail(&nf->nf_lru, dst);
+	}
+}
+
+static void
+nfsd_file_dispose_list_delayed(struct list_head *dispose)
+{
+	LIST_HEAD(list);
+	struct nfsd_file *nf;
+
+	while(!list_empty(dispose)) {
+		nf = list_first_entry(dispose, struct nfsd_file, nf_lru);
+		nfsd_file_list_add_pernet(&list, dispose, nf->nf_net);
+		nfsd_file_list_add_disposal(&list, nf->nf_net);
+	}
+}
+
 /*
  * Note this can deadlock with nfsd_file_cache_purge.
  */
@@ -375,7 +429,7 @@ nfsd_file_lru_cb(struct list_head *item, struct list_lru_one *lru,
 	 * counter. Here we check the counter and then test and clear the flag.
 	 * That order is deliberate to ensure that we can do this locklessly.
 	 */
-	if (atomic_read(&nf->nf_ref) > 1)
+	if (refcount_read(&nf->nf_ref) > 1)
 		goto out_skip;
 
 	/*
@@ -386,31 +440,51 @@ nfsd_file_lru_cb(struct list_head *item, struct list_lru_one *lru,
 		goto out_skip;
 
 	if (test_and_clear_bit(NFSD_FILE_REFERENCED, &nf->nf_flags))
-		goto out_rescan;
+		goto out_skip;
 
 	if (!test_and_clear_bit(NFSD_FILE_HASHED, &nf->nf_flags))
 		goto out_skip;
 
 	list_lru_isolate_move(lru, &nf->nf_lru, head);
 	return LRU_REMOVED;
-out_rescan:
-	set_bit(NFSD_FILE_LRU_RESCAN, &nfsd_file_lru_flags);
 out_skip:
 	return LRU_SKIP;
 }
 
-static void
-nfsd_file_lru_dispose(struct list_head *head)
+static unsigned long
+nfsd_file_lru_walk_list(struct shrink_control *sc)
 {
-	while(!list_empty(head)) {
-		struct nfsd_file *nf = list_first_entry(head,
-				struct nfsd_file, nf_lru);
-		list_del_init(&nf->nf_lru);
+	LIST_HEAD(head);
+	struct nfsd_file *nf;
+	unsigned long ret;
+
+	if (sc)
+		ret = list_lru_shrink_walk(&nfsd_file_lru, sc,
+				nfsd_file_lru_cb, &head);
+	else
+		ret = list_lru_walk(&nfsd_file_lru,
+				nfsd_file_lru_cb,
+				&head, LONG_MAX);
+	list_for_each_entry(nf, &head, nf_lru) {
 		spin_lock(&nfsd_file_hashtbl[nf->nf_hashval].nfb_lock);
 		nfsd_file_do_unhash(nf);
 		spin_unlock(&nfsd_file_hashtbl[nf->nf_hashval].nfb_lock);
-		nfsd_file_put_noref(nf);
 	}
+	nfsd_file_dispose_list_delayed(&head);
+	return ret;
+}
+
+static void
+nfsd_file_gc(void)
+{
+	nfsd_file_lru_walk_list(NULL);
+}
+
+static void
+nfsd_file_gc_worker(struct work_struct *work)
+{
+	nfsd_file_gc();
+	nfsd_file_schedule_laundrette();
 }
 
 static unsigned long
@@ -422,12 +496,7 @@ nfsd_file_lru_count(struct shrinker *s, struct shrink_control *sc)
 static unsigned long
 nfsd_file_lru_scan(struct shrinker *s, struct shrink_control *sc)
 {
-	LIST_HEAD(head);
-	unsigned long ret;
-
-	ret = list_lru_shrink_walk(&nfsd_file_lru, sc, nfsd_file_lru_cb, &head);
-	nfsd_file_lru_dispose(&head);
-	return ret;
+	return nfsd_file_lru_walk_list(sc);
 }
 
 static struct shrinker	nfsd_file_shrinker = {
@@ -489,7 +558,7 @@ nfsd_file_close_inode(struct inode *inode)
 
 	__nfsd_file_close_inode(inode, hashval, &dispose);
 	trace_nfsd_file_close_inode(inode, hashval, !list_empty(&dispose));
-	nfsd_file_dispose_list(&dispose);
+	nfsd_file_dispose_list_delayed(&dispose);
 }
 
 /**
@@ -505,16 +574,11 @@ static void
 nfsd_file_delayed_close(struct work_struct *work)
 {
 	LIST_HEAD(head);
+	struct nfsd_fcache_disposal *l = container_of(work,
+			struct nfsd_fcache_disposal, work);
 
-	list_lru_walk(&nfsd_file_lru, nfsd_file_lru_cb, &head, LONG_MAX);
-
-	if (test_and_clear_bit(NFSD_FILE_LRU_RESCAN, &nfsd_file_lru_flags))
-		nfsd_file_schedule_laundrette(NFSD_FILE_LAUNDRETTE_NOFLUSH);
-
-	if (!list_empty(&head)) {
-		nfsd_file_lru_dispose(&head);
-		flush_delayed_fput();
-	}
+	nfsd_file_list_remove_disposal(&head, l);
+	nfsd_file_dispose_list(&head);
 }
 
 static int
@@ -575,6 +639,10 @@ nfsd_file_cache_init(void)
 	if (nfsd_file_hashtbl)
 		return 0;
 
+	nfsd_filecache_wq = alloc_workqueue("nfsd_filecache", 0, 0);
+	if (!nfsd_filecache_wq)
+		goto out;
+
 	nfsd_file_hashtbl = kcalloc(NFSD_FILE_HASH_SIZE,
 				sizeof(*nfsd_file_hashtbl), GFP_KERNEL);
 	if (!nfsd_file_hashtbl) {
@@ -628,7 +696,7 @@ nfsd_file_cache_init(void)
 		spin_lock_init(&nfsd_file_hashtbl[i].nfb_lock);
 	}
 
-	INIT_DELAYED_WORK(&nfsd_filecache_laundrette, nfsd_file_delayed_close);
+	INIT_DELAYED_WORK(&nfsd_filecache_laundrette, nfsd_file_gc_worker);
 out:
 	return ret;
 out_notifier:
@@ -644,6 +712,8 @@ out_err:
 	nfsd_file_mark_slab = NULL;
 	kfree(nfsd_file_hashtbl);
 	nfsd_file_hashtbl = NULL;
+	destroy_workqueue(nfsd_filecache_wq);
+	nfsd_filecache_wq = NULL;
 	goto out;
 }
 
@@ -682,11 +752,91 @@ nfsd_file_cache_purge(struct net *net)
 	}
 }
 
+static struct nfsd_fcache_disposal *
+nfsd_alloc_fcache_disposal(struct net *net)
+{
+	struct nfsd_fcache_disposal *l;
+
+	l = kmalloc(sizeof(*l), GFP_KERNEL);
+	if (!l)
+		return NULL;
+	INIT_WORK(&l->work, nfsd_file_delayed_close);
+	l->net = net;
+	spin_lock_init(&l->lock);
+	INIT_LIST_HEAD(&l->freeme);
+	return l;
+}
+
+static void
+nfsd_free_fcache_disposal(struct nfsd_fcache_disposal *l)
+{
+	rcu_assign_pointer(l->net, NULL);
+	cancel_work_sync(&l->work);
+	nfsd_file_dispose_list(&l->freeme);
+	kfree_rcu(l, rcu);
+}
+
+static void
+nfsd_add_fcache_disposal(struct nfsd_fcache_disposal *l)
+{
+	spin_lock(&laundrette_lock);
+	list_add_tail_rcu(&l->list, &laundrettes);
+	spin_unlock(&laundrette_lock);
+}
+
+static void
+nfsd_del_fcache_disposal(struct nfsd_fcache_disposal *l)
+{
+	spin_lock(&laundrette_lock);
+	list_del_rcu(&l->list);
+	spin_unlock(&laundrette_lock);
+}
+
+static int
+nfsd_alloc_fcache_disposal_net(struct net *net)
+{
+	struct nfsd_fcache_disposal *l;
+
+	l = nfsd_alloc_fcache_disposal(net);
+	if (!l)
+		return -ENOMEM;
+	nfsd_add_fcache_disposal(l);
+	return 0;
+}
+
+static void
+nfsd_free_fcache_disposal_net(struct net *net)
+{
+	struct nfsd_fcache_disposal *l;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(l, &laundrettes, list) {
+		if (l->net != net)
+			continue;
+		nfsd_del_fcache_disposal(l);
+		rcu_read_unlock();
+		nfsd_free_fcache_disposal(l);
+		return;
+	}
+	rcu_read_unlock();
+}
+
+int
+nfsd_file_cache_start_net(struct net *net)
+{
+	return nfsd_alloc_fcache_disposal_net(net);
+}
+
 void
-nfsd_file_cache_shutdown(void)
+nfsd_file_cache_shutdown_net(struct net *net)
 {
-	LIST_HEAD(dispose);
+	nfsd_file_cache_purge(net);
+	nfsd_free_fcache_disposal_net(net);
+}
 
+void
+nfsd_file_cache_shutdown(void)
+{
 	set_bit(NFSD_FILE_SHUTDOWN, &nfsd_file_lru_flags);
 
 	lease_unregister_notifier(&nfsd_file_lease_notifier);
@@ -708,6 +858,8 @@ nfsd_file_cache_shutdown(void)
 	nfsd_file_mark_slab = NULL;
 	kfree(nfsd_file_hashtbl);
 	nfsd_file_hashtbl = NULL;
+	destroy_workqueue(nfsd_filecache_wq);
+	nfsd_filecache_wq = NULL;
 }
 
 static bool
@@ -738,7 +890,7 @@ nfsd_file_find_locked(struct inode *inode, unsigned int may_flags,
 	unsigned char need = may_flags & NFSD_FILE_MAY_MASK;
 
 	hlist_for_each_entry_rcu(nf, &nfsd_file_hashtbl[hashval].nfb_head,
-				 nf_node) {
+				 nf_node, lockdep_is_held(&nfsd_file_hashtbl[hashval].nfb_lock)) {
 		if ((need & nf->nf_may) != need)
 			continue;
 		if (nf->nf_inode != inode)
@@ -791,6 +943,7 @@ nfsd_file_acquire(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	struct nfsd_file *nf, *new;
 	struct inode *inode;
 	unsigned int hashval;
+	bool retry = true;
 
 	/* FIXME: skip this if fh_dentry is already set? */
 	status = fh_verify(rqstp, fhp, S_IFREG,
@@ -826,6 +979,11 @@ wait_for_construction:
 
 	/* Did construction of this file fail? */
 	if (!test_bit(NFSD_FILE_HASHED, &nf->nf_flags)) {
+		if (!retry) {
+			status = nfserr_jukebox;
+			goto out;
+		}
+		retry = false;
 		nfsd_file_put_noref(nf);
 		goto retry;
 	}
@@ -860,7 +1018,7 @@ out:
 open_file:
 	nf = new;
 	/* Take reference for the hashtable */
-	atomic_inc(&nf->nf_ref);
+	refcount_inc(&nf->nf_ref);
 	__set_bit(NFSD_FILE_HASHED, &nf->nf_flags);
 	__set_bit(NFSD_FILE_PENDING, &nf->nf_flags);
 	list_lru_add(&nfsd_file_lru, &nf->nf_lru);
@@ -869,7 +1027,8 @@ open_file:
 	nfsd_file_hashtbl[hashval].nfb_maxcount = max(nfsd_file_hashtbl[hashval].nfb_maxcount,
 			nfsd_file_hashtbl[hashval].nfb_count);
 	spin_unlock(&nfsd_file_hashtbl[hashval].nfb_lock);
-	atomic_long_inc(&nfsd_filecache_count);
+	if (atomic_long_inc_return(&nfsd_filecache_count) >= NFSD_FILE_LRU_THRESHOLD)
+		nfsd_file_gc();
 
 	nf->nf_mark = nfsd_file_mark_find_or_create(nf);
 	if (nf->nf_mark)
diff --git a/fs/nfsd/filecache.h b/fs/nfsd/filecache.h
index 851d9abf54c2..7872df5a0fe3 100644
--- a/fs/nfsd/filecache.h
+++ b/fs/nfsd/filecache.h
@@ -19,7 +19,7 @@
  */
 struct nfsd_file_mark {
 	struct fsnotify_mark	nfm_mark;
-	atomic_t		nfm_ref;
+	refcount_t		nfm_ref;
 };
 
 /*
@@ -43,14 +43,17 @@ struct nfsd_file {
 	unsigned long		nf_flags;
 	struct inode		*nf_inode;
 	unsigned int		nf_hashval;
-	atomic_t		nf_ref;
+	refcount_t		nf_ref;
 	unsigned char		nf_may;
 	struct nfsd_file_mark	*nf_mark;
+	struct rw_semaphore	nf_rwsem;
 };
 
 int nfsd_file_cache_init(void);
 void nfsd_file_cache_purge(struct net *);
 void nfsd_file_cache_shutdown(void);
+int nfsd_file_cache_start_net(struct net *net);
+void nfsd_file_cache_shutdown_net(struct net *net);
 void nfsd_file_put(struct nfsd_file *nf);
 struct nfsd_file *nfsd_file_get(struct nfsd_file *nf);
 void nfsd_file_close_inode_sync(struct inode *inode);
diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h
index 9a4ef815fb8c..09aa545825bd 100644
--- a/fs/nfsd/netns.h
+++ b/fs/nfsd/netns.h
@@ -40,7 +40,7 @@ struct nfsd_net {
 
 	struct lock_manager nfsd4_manager;
 	bool grace_ended;
-	time_t boot_time;
+	time64_t boot_time;
 
 	/* internal mount of the "nfsd" pseudofilesystem: */
 	struct vfsmount *nfsd_mnt;
@@ -92,8 +92,8 @@ struct nfsd_net {
 	bool in_grace;
 	const struct nfsd4_client_tracking_ops *client_tracking_ops;
 
-	time_t nfsd4_lease;
-	time_t nfsd4_grace;
+	time64_t nfsd4_lease;
+	time64_t nfsd4_grace;
 	bool somebody_reclaimed;
 
 	bool track_reclaim_completes;
@@ -172,6 +172,8 @@ struct nfsd_net {
 	unsigned int             longest_chain_cachesize;
 
 	struct shrinker		nfsd_reply_cache_shrinker;
+	/* utsname taken from the the process that starts the server */
+	char			nfsd_name[UNX_MAXNODENAME+1];
 };
 
 /* Simple check to find out if a given net was properly initialized */
diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index cea68d8411ac..288bc76b4574 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -203,7 +203,7 @@ nfsd3_proc_write(struct svc_rqst *rqstp)
 		RETURN_STATUS(nfserr_io);
 	nfserr = nfsd_write(rqstp, &resp->fh, argp->offset,
 			    rqstp->rq_vec, nvecs, &cnt,
-			    resp->committed);
+			    resp->committed, resp->verf);
 	resp->count = cnt;
 	RETURN_STATUS(nfserr);
 }
@@ -683,7 +683,8 @@ nfsd3_proc_commit(struct svc_rqst *rqstp)
 		RETURN_STATUS(nfserr_inval);
 
 	fh_copy(&resp->fh, &argp->fh);
-	nfserr = nfsd_commit(rqstp, &resp->fh, argp->offset, argp->count);
+	nfserr = nfsd_commit(rqstp, &resp->fh, argp->offset, argp->count,
+			resp->verf);
 
 	RETURN_STATUS(nfserr);
 }
diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c
index 86e5658651f1..aae514d40b64 100644
--- a/fs/nfsd/nfs3xdr.c
+++ b/fs/nfsd/nfs3xdr.c
@@ -32,14 +32,14 @@ static u32	nfs3_ftypes[] = {
  * XDR functions for basic NFS types
  */
 static __be32 *
-encode_time3(__be32 *p, struct timespec *time)
+encode_time3(__be32 *p, struct timespec64 *time)
 {
 	*p++ = htonl((u32) time->tv_sec); *p++ = htonl(time->tv_nsec);
 	return p;
 }
 
 static __be32 *
-decode_time3(__be32 *p, struct timespec *time)
+decode_time3(__be32 *p, struct timespec64 *time)
 {
 	time->tv_sec = ntohl(*p++);
 	time->tv_nsec = ntohl(*p++);
@@ -167,7 +167,6 @@ encode_fattr3(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp,
 	      struct kstat *stat)
 {
 	struct user_namespace *userns = nfsd_user_namespace(rqstp);
-	struct timespec ts;
 	*p++ = htonl(nfs3_ftypes[(stat->mode & S_IFMT) >> 12]);
 	*p++ = htonl((u32) (stat->mode & S_IALLUGO));
 	*p++ = htonl((u32) stat->nlink);
@@ -183,12 +182,9 @@ encode_fattr3(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp,
 	*p++ = htonl((u32) MINOR(stat->rdev));
 	p = encode_fsid(p, fhp);
 	p = xdr_encode_hyper(p, stat->ino);
-	ts = timespec64_to_timespec(stat->atime);
-	p = encode_time3(p, &ts);
-	ts = timespec64_to_timespec(stat->mtime);
-	p = encode_time3(p, &ts);
-	ts = timespec64_to_timespec(stat->ctime);
-	p = encode_time3(p, &ts);
+	p = encode_time3(p, &stat->atime);
+	p = encode_time3(p, &stat->mtime);
+	p = encode_time3(p, &stat->ctime);
 
 	return p;
 }
@@ -277,8 +273,8 @@ void fill_pre_wcc(struct svc_fh *fhp)
 		stat.size  = inode->i_size;
 	}
 
-	fhp->fh_pre_mtime = timespec64_to_timespec(stat.mtime);
-	fhp->fh_pre_ctime = timespec64_to_timespec(stat.ctime);
+	fhp->fh_pre_mtime = stat.mtime;
+	fhp->fh_pre_ctime = stat.ctime;
 	fhp->fh_pre_size  = stat.size;
 	fhp->fh_pre_change = nfsd4_change_attribute(&stat, inode);
 	fhp->fh_pre_saved = true;
@@ -330,7 +326,7 @@ nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p)
 	p = decode_sattr3(p, &args->attrs, nfsd_user_namespace(rqstp));
 
 	if ((args->check_guard = ntohl(*p++)) != 0) { 
-		struct timespec time; 
+		struct timespec64 time;
 		p = decode_time3(p, &time);
 		args->guardtime = time.tv_sec;
 	}
@@ -751,17 +747,13 @@ int
 nfs3svc_encode_writeres(struct svc_rqst *rqstp, __be32 *p)
 {
 	struct nfsd3_writeres *resp = rqstp->rq_resp;
-	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
-	__be32 verf[2];
 
 	p = encode_wcc_data(rqstp, p, &resp->fh);
 	if (resp->status == 0) {
 		*p++ = htonl(resp->count);
 		*p++ = htonl(resp->committed);
-		/* unique identifier, y2038 overflow can be ignored */
-		nfsd_copy_boot_verifier(verf, nn);
-		*p++ = verf[0];
-		*p++ = verf[1];
+		*p++ = resp->verf[0];
+		*p++ = resp->verf[1];
 	}
 	return xdr_ressize_check(rqstp, p);
 }
@@ -863,13 +855,11 @@ compose_entry_fh(struct nfsd3_readdirres *cd, struct svc_fh *fhp,
 		} else
 			dchild = dget(dparent);
 	} else
-		dchild = lookup_one_len_unlocked(name, dparent, namlen);
+		dchild = lookup_positive_unlocked(name, dparent, namlen);
 	if (IS_ERR(dchild))
 		return rv;
 	if (d_mountpoint(dchild))
 		goto out;
-	if (d_really_is_negative(dchild))
-		goto out;
 	if (dchild->d_inode->i_ino != ino)
 		goto out;
 	rv = fh_compose(fhp, exp, dchild, &cd->fh);
@@ -1127,16 +1117,12 @@ int
 nfs3svc_encode_commitres(struct svc_rqst *rqstp, __be32 *p)
 {
 	struct nfsd3_commitres *resp = rqstp->rq_resp;
-	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
-	__be32 verf[2];
 
 	p = encode_wcc_data(rqstp, p, &resp->fh);
 	/* Write verifier */
 	if (resp->status == 0) {
-		/* unique identifier, y2038 overflow can be ignored */
-		nfsd_copy_boot_verifier(verf, nn);
-		*p++ = verf[0];
-		*p++ = verf[1];
+		*p++ = resp->verf[0];
+		*p++ = resp->verf[1];
 	}
 	return xdr_ressize_check(rqstp, p);
 }
diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 524111420b48..5cf91322de0f 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -823,7 +823,41 @@ static const struct rpc_program cb_program = {
 static int max_cb_time(struct net *net)
 {
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
-	return max(nn->nfsd4_lease/10, (time_t)1) * HZ;
+
+	/*
+	 * nfsd4_lease is set to at most one hour in __nfsd4_write_time,
+	 * so we can use 32-bit math on it. Warn if that assumption
+	 * ever stops being true.
+	 */
+	if (WARN_ON_ONCE(nn->nfsd4_lease > 3600))
+		return 360 * HZ;
+
+	return max(((u32)nn->nfsd4_lease)/10, 1u) * HZ;
+}
+
+static struct workqueue_struct *callback_wq;
+
+static bool nfsd4_queue_cb(struct nfsd4_callback *cb)
+{
+	return queue_work(callback_wq, &cb->cb_work);
+}
+
+static void nfsd41_cb_inflight_begin(struct nfs4_client *clp)
+{
+	atomic_inc(&clp->cl_cb_inflight);
+}
+
+static void nfsd41_cb_inflight_end(struct nfs4_client *clp)
+{
+
+	if (atomic_dec_and_test(&clp->cl_cb_inflight))
+		wake_up_var(&clp->cl_cb_inflight);
+}
+
+static void nfsd41_cb_inflight_wait_complete(struct nfs4_client *clp)
+{
+	wait_var_event(&clp->cl_cb_inflight,
+			!atomic_read(&clp->cl_cb_inflight));
 }
 
 static const struct cred *get_backchannel_cred(struct nfs4_client *clp, struct rpc_clnt *client, struct nfsd4_session *ses)
@@ -937,14 +971,21 @@ static void nfsd4_cb_probe_done(struct rpc_task *task, void *calldata)
 		clp->cl_cb_state = NFSD4_CB_UP;
 }
 
+static void nfsd4_cb_probe_release(void *calldata)
+{
+	struct nfs4_client *clp = container_of(calldata, struct nfs4_client, cl_cb_null);
+
+	nfsd41_cb_inflight_end(clp);
+
+}
+
 static const struct rpc_call_ops nfsd4_cb_probe_ops = {
 	/* XXX: release method to ensure we set the cb channel down if
 	 * necessary on early failure? */
 	.rpc_call_done = nfsd4_cb_probe_done,
+	.rpc_release = nfsd4_cb_probe_release,
 };
 
-static struct workqueue_struct *callback_wq;
-
 /*
  * Poke the callback thread to process any updates to the callback
  * parameters, and send a null probe.
@@ -975,9 +1016,12 @@ void nfsd4_change_callback(struct nfs4_client *clp, struct nfs4_cb_conn *conn)
  * If the slot is available, then mark it busy.  Otherwise, set the
  * thread for sleeping on the callback RPC wait queue.
  */
-static bool nfsd41_cb_get_slot(struct nfs4_client *clp, struct rpc_task *task)
+static bool nfsd41_cb_get_slot(struct nfsd4_callback *cb, struct rpc_task *task)
 {
-	if (test_and_set_bit(0, &clp->cl_cb_slot_busy) != 0) {
+	struct nfs4_client *clp = cb->cb_clp;
+
+	if (!cb->cb_holds_slot &&
+	    test_and_set_bit(0, &clp->cl_cb_slot_busy) != 0) {
 		rpc_sleep_on(&clp->cl_cb_waitq, task, NULL);
 		/* Race breaker */
 		if (test_and_set_bit(0, &clp->cl_cb_slot_busy) != 0) {
@@ -986,9 +1030,31 @@ static bool nfsd41_cb_get_slot(struct nfs4_client *clp, struct rpc_task *task)
 		}
 		rpc_wake_up_queued_task(&clp->cl_cb_waitq, task);
 	}
+	cb->cb_holds_slot = true;
 	return true;
 }
 
+static void nfsd41_cb_release_slot(struct nfsd4_callback *cb)
+{
+	struct nfs4_client *clp = cb->cb_clp;
+
+	if (cb->cb_holds_slot) {
+		cb->cb_holds_slot = false;
+		clear_bit(0, &clp->cl_cb_slot_busy);
+		rpc_wake_up_next(&clp->cl_cb_waitq);
+	}
+}
+
+static void nfsd41_destroy_cb(struct nfsd4_callback *cb)
+{
+	struct nfs4_client *clp = cb->cb_clp;
+
+	nfsd41_cb_release_slot(cb);
+	if (cb->cb_ops && cb->cb_ops->release)
+		cb->cb_ops->release(cb);
+	nfsd41_cb_inflight_end(clp);
+}
+
 /*
  * TODO: cb_sequence should support referring call lists, cachethis, multiple
  * slots, and mark callback channel down on communication errors.
@@ -1005,11 +1071,8 @@ static void nfsd4_cb_prepare(struct rpc_task *task, void *calldata)
 	 */
 	cb->cb_seq_status = 1;
 	cb->cb_status = 0;
-	if (minorversion) {
-		if (!cb->cb_holds_slot && !nfsd41_cb_get_slot(clp, task))
-			return;
-		cb->cb_holds_slot = true;
-	}
+	if (minorversion && !nfsd41_cb_get_slot(cb, task))
+		return;
 	rpc_call_start(task);
 }
 
@@ -1072,13 +1135,12 @@ static bool nfsd4_cb_sequence_done(struct rpc_task *task, struct nfsd4_callback
 		}
 		break;
 	default:
+		nfsd4_mark_cb_fault(cb->cb_clp, cb->cb_seq_status);
 		dprintk("%s: unprocessed error %d\n", __func__,
 			cb->cb_seq_status);
 	}
 
-	cb->cb_holds_slot = false;
-	clear_bit(0, &clp->cl_cb_slot_busy);
-	rpc_wake_up_next(&clp->cl_cb_waitq);
+	nfsd41_cb_release_slot(cb);
 	dprintk("%s: freed slot, new seqid=%d\n", __func__,
 		clp->cl_cb_session->se_cb_seq_nr);
 
@@ -1091,8 +1153,10 @@ retry_nowait:
 		ret = false;
 	goto out;
 need_restart:
-	task->tk_status = 0;
-	cb->cb_need_restart = true;
+	if (!test_bit(NFSD4_CLIENT_CB_KILL, &clp->cl_flags)) {
+		task->tk_status = 0;
+		cb->cb_need_restart = true;
+	}
 	return false;
 }
 
@@ -1134,9 +1198,9 @@ static void nfsd4_cb_release(void *calldata)
 	struct nfsd4_callback *cb = calldata;
 
 	if (cb->cb_need_restart)
-		nfsd4_run_cb(cb);
+		nfsd4_queue_cb(cb);
 	else
-		cb->cb_ops->release(cb);
+		nfsd41_destroy_cb(cb);
 
 }
 
@@ -1170,6 +1234,7 @@ void nfsd4_shutdown_callback(struct nfs4_client *clp)
 	 */
 	nfsd4_run_cb(&clp->cl_cb_null);
 	flush_workqueue(callback_wq);
+	nfsd41_cb_inflight_wait_complete(clp);
 }
 
 /* requires cl_lock: */
@@ -1187,6 +1252,12 @@ static struct nfsd4_conn * __nfsd4_find_backchannel(struct nfs4_client *clp)
 	return NULL;
 }
 
+/*
+ * Note there isn't a lot of locking in this code; instead we depend on
+ * the fact that it is run from the callback_wq, which won't run two
+ * work items at once.  So, for example, callback_wq handles all access
+ * of cl_cb_client and all calls to rpc_create or rpc_shutdown_client.
+ */
 static void nfsd4_process_cb_update(struct nfsd4_callback *cb)
 {
 	struct nfs4_cb_conn conn;
@@ -1241,6 +1312,7 @@ nfsd4_run_cb_work(struct work_struct *work)
 		container_of(work, struct nfsd4_callback, cb_work);
 	struct nfs4_client *clp = cb->cb_clp;
 	struct rpc_clnt *clnt;
+	int flags;
 
 	if (cb->cb_need_restart) {
 		cb->cb_need_restart = false;
@@ -1255,8 +1327,7 @@ nfsd4_run_cb_work(struct work_struct *work)
 	clnt = clp->cl_cb_client;
 	if (!clnt) {
 		/* Callback channel broken, or client killed; give up: */
-		if (cb->cb_ops && cb->cb_ops->release)
-			cb->cb_ops->release(cb);
+		nfsd41_destroy_cb(cb);
 		return;
 	}
 
@@ -1265,11 +1336,13 @@ nfsd4_run_cb_work(struct work_struct *work)
 	 */
 	if (!cb->cb_ops && clp->cl_minorversion) {
 		clp->cl_cb_state = NFSD4_CB_UP;
+		nfsd41_destroy_cb(cb);
 		return;
 	}
 
 	cb->cb_msg.rpc_cred = clp->cl_cb_cred;
-	rpc_call_async(clnt, &cb->cb_msg, RPC_TASK_SOFT | RPC_TASK_SOFTCONN,
+	flags = clp->cl_minorversion ? RPC_TASK_NOCONNECT : RPC_TASK_SOFTCONN;
+	rpc_call_async(clnt, &cb->cb_msg, RPC_TASK_SOFT | flags,
 			cb->cb_ops ? &nfsd4_cb_ops : &nfsd4_cb_probe_ops, cb);
 }
 
@@ -1290,5 +1363,9 @@ void nfsd4_init_cb(struct nfsd4_callback *cb, struct nfs4_client *clp,
 
 void nfsd4_run_cb(struct nfsd4_callback *cb)
 {
-	queue_work(callback_wq, &cb->cb_work);
+	struct nfs4_client *clp = cb->cb_clp;
+
+	nfsd41_cb_inflight_begin(clp);
+	if (!nfsd4_queue_cb(cb))
+		nfsd41_cb_inflight_end(clp);
 }
diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c
index d1f285245af8..9460be8a8321 100644
--- a/fs/nfsd/nfs4idmap.c
+++ b/fs/nfsd/nfs4idmap.c
@@ -122,6 +122,12 @@ idtoname_hash(struct ent *ent)
 	return hash;
 }
 
+static int
+idtoname_upcall(struct cache_detail *cd, struct cache_head *h)
+{
+	return sunrpc_cache_pipe_upcall_timeout(cd, h);
+}
+
 static void
 idtoname_request(struct cache_detail *cd, struct cache_head *ch, char **bpp,
     int *blen)
@@ -184,6 +190,7 @@ static const struct cache_detail idtoname_cache_template = {
 	.hash_size	= ENT_HASHMAX,
 	.name		= "nfs4.idtoname",
 	.cache_put	= ent_put,
+	.cache_upcall	= idtoname_upcall,
 	.cache_request	= idtoname_request,
 	.cache_parse	= idtoname_parse,
 	.cache_show	= idtoname_show,
@@ -295,6 +302,12 @@ nametoid_hash(struct ent *ent)
 	return hash_str(ent->name, ENT_HASHBITS);
 }
 
+static int
+nametoid_upcall(struct cache_detail *cd, struct cache_head *h)
+{
+	return sunrpc_cache_pipe_upcall_timeout(cd, h);
+}
+
 static void
 nametoid_request(struct cache_detail *cd, struct cache_head *ch, char **bpp,
     int *blen)
@@ -347,6 +360,7 @@ static const struct cache_detail nametoid_cache_template = {
 	.hash_size	= ENT_HASHMAX,
 	.name		= "nfs4.nametoid",
 	.cache_put	= ent_put,
+	.cache_upcall	= nametoid_upcall,
 	.cache_request	= nametoid_request,
 	.cache_parse	= nametoid_parse,
 	.cache_show	= nametoid_show,
diff --git a/fs/nfsd/nfs4layouts.c b/fs/nfsd/nfs4layouts.c
index 2681c70283ce..e12409eca7cc 100644
--- a/fs/nfsd/nfs4layouts.c
+++ b/fs/nfsd/nfs4layouts.c
@@ -675,7 +675,7 @@ nfsd4_cb_layout_done(struct nfsd4_callback *cb, struct rpc_task *task)
 
 		/* Client gets 2 lease periods to return it */
 		cutoff = ktime_add_ns(task->tk_start,
-					 nn->nfsd4_lease * NSEC_PER_SEC * 2);
+					 (u64)nn->nfsd4_lease * NSEC_PER_SEC * 2);
 
 		if (ktime_before(now, cutoff)) {
 			rpc_delay(task, HZ/100); /* 10 mili-seconds */
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 4e3e77b76411..0e75f7fb5fec 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -37,6 +37,7 @@
 #include <linux/falloc.h>
 #include <linux/slab.h>
 #include <linux/kthread.h>
+#include <linux/sunrpc/addr.h>
 
 #include "idmap.h"
 #include "cache.h"
@@ -232,7 +233,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, stru
 	if (!*resfh)
 		return nfserr_jukebox;
 	fh_init(*resfh, NFS4_FHSIZE);
-	open->op_truncate = 0;
+	open->op_truncate = false;
 
 	if (open->op_create) {
 		/* FIXME: check session persistence and pnfs flags.
@@ -365,7 +366,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (open->op_create && open->op_claim_type != NFS4_OPEN_CLAIM_NULL)
 		return nfserr_inval;
 
-	open->op_created = 0;
+	open->op_created = false;
 	/*
 	 * RFC5661 18.51.3
 	 * Before RECLAIM_COMPLETE done, server should deny new lock
@@ -503,12 +504,20 @@ nfsd4_putfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	    union nfsd4_op_u *u)
 {
 	struct nfsd4_putfh *putfh = &u->putfh;
+	__be32 ret;
 
 	fh_put(&cstate->current_fh);
 	cstate->current_fh.fh_handle.fh_size = putfh->pf_fhlen;
 	memcpy(&cstate->current_fh.fh_handle.fh_base, putfh->pf_fhval,
 	       putfh->pf_fhlen);
-	return fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_BYPASS_GSS);
+	ret = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_BYPASS_GSS);
+#ifdef CONFIG_NFSD_V4_2_INTER_SSC
+	if (ret == nfserr_stale && putfh->no_verify) {
+		SET_FH_FLAG(&cstate->current_fh, NFSD4_FH_FOREIGN);
+		ret = 0;
+	}
+#endif
+	return ret;
 }
 
 static __be32
@@ -530,9 +539,9 @@ nfsd4_restorefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		return nfserr_restorefh;
 
 	fh_dup2(&cstate->current_fh, &cstate->save_fh);
-	if (HAS_STATE_ID(cstate, SAVED_STATE_ID_FLAG)) {
+	if (HAS_CSTATE_FLAG(cstate, SAVED_STATE_ID_FLAG)) {
 		memcpy(&cstate->current_stateid, &cstate->save_stateid, sizeof(stateid_t));
-		SET_STATE_ID(cstate, CURRENT_STATE_ID_FLAG);
+		SET_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG);
 	}
 	return nfs_ok;
 }
@@ -542,9 +551,9 @@ nfsd4_savefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	     union nfsd4_op_u *u)
 {
 	fh_dup2(&cstate->save_fh, &cstate->current_fh);
-	if (HAS_STATE_ID(cstate, CURRENT_STATE_ID_FLAG)) {
+	if (HAS_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG)) {
 		memcpy(&cstate->save_stateid, &cstate->current_stateid, sizeof(stateid_t));
-		SET_STATE_ID(cstate, SAVED_STATE_ID_FLAG);
+		SET_CSTATE_FLAG(cstate, SAVED_STATE_ID_FLAG);
 	}
 	return nfs_ok;
 }
@@ -581,9 +590,9 @@ nfsd4_commit(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 {
 	struct nfsd4_commit *commit = &u->commit;
 
-	gen_boot_verifier(&commit->co_verf, SVC_NET(rqstp));
 	return nfsd_commit(rqstp, &cstate->current_fh, commit->co_offset,
-			     commit->co_count);
+			     commit->co_count,
+			     (__be32 *)commit->co_verf.data);
 }
 
 static __be32
@@ -776,7 +785,7 @@ nfsd4_read(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	/* check stateid */
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
 					&read->rd_stateid, RD_STATE,
-					&read->rd_nf);
+					&read->rd_nf, NULL);
 	if (status) {
 		dprintk("NFSD: nfsd4_read: couldn't process stateid!\n");
 		goto out;
@@ -948,7 +957,7 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (setattr->sa_iattr.ia_valid & ATTR_SIZE) {
 		status = nfs4_preprocess_stateid_op(rqstp, cstate,
 				&cstate->current_fh, &setattr->sa_stateid,
-				WR_STATE, NULL);
+				WR_STATE, NULL, NULL);
 		if (status) {
 			dprintk("NFSD: nfsd4_setattr: couldn't process stateid!\n");
 			return status;
@@ -975,7 +984,7 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (status)
 		goto out;
 	status = nfsd_setattr(rqstp, &cstate->current_fh, &setattr->sa_iattr,
-				0, (time_t)0);
+				0, (time64_t)0);
 out:
 	fh_drop_write(&cstate->current_fh);
 	return status;
@@ -999,22 +1008,22 @@ nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	trace_nfsd_write_start(rqstp, &cstate->current_fh,
 			       write->wr_offset, cnt);
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
-						stateid, WR_STATE, &nf);
+						stateid, WR_STATE, &nf, NULL);
 	if (status) {
 		dprintk("NFSD: nfsd4_write: couldn't process stateid!\n");
 		return status;
 	}
 
 	write->wr_how_written = write->wr_stable_how;
-	gen_boot_verifier(&write->wr_verifier, SVC_NET(rqstp));
 
 	nvecs = svc_fill_write_vector(rqstp, write->wr_pagelist,
 				      &write->wr_head, write->wr_buflen);
 	WARN_ON_ONCE(nvecs > ARRAY_SIZE(rqstp->rq_vec));
 
-	status = nfsd_vfs_write(rqstp, &cstate->current_fh, nf->nf_file,
+	status = nfsd_vfs_write(rqstp, &cstate->current_fh, nf,
 				write->wr_offset, rqstp->rq_vec, nvecs, &cnt,
-				write->wr_how_written);
+				write->wr_how_written,
+				(__be32 *)write->wr_verifier.data);
 	nfsd_file_put(nf);
 
 	write->wr_bytes_written = cnt;
@@ -1034,14 +1043,14 @@ nfsd4_verify_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		return nfserr_nofilehandle;
 
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->save_fh,
-					    src_stateid, RD_STATE, src);
+					    src_stateid, RD_STATE, src, NULL);
 	if (status) {
 		dprintk("NFSD: %s: couldn't process src stateid!\n", __func__);
 		goto out;
 	}
 
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
-					    dst_stateid, WR_STATE, dst);
+					    dst_stateid, WR_STATE, dst, NULL);
 	if (status) {
 		dprintk("NFSD: %s: couldn't process dst stateid!\n", __func__);
 		goto out_put_src;
@@ -1076,8 +1085,9 @@ nfsd4_clone(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (status)
 		goto out;
 
-	status = nfsd4_clone_file_range(src->nf_file, clone->cl_src_pos,
-			dst->nf_file, clone->cl_dst_pos, clone->cl_count);
+	status = nfsd4_clone_file_range(src, clone->cl_src_pos,
+			dst, clone->cl_dst_pos, clone->cl_count,
+			EX_ISSYNC(cstate->current_fh.fh_export));
 
 	nfsd_file_put(dst);
 	nfsd_file_put(src);
@@ -1134,6 +1144,207 @@ void nfsd4_shutdown_copy(struct nfs4_client *clp)
 	while ((copy = nfsd4_get_copy(clp)) != NULL)
 		nfsd4_stop_copy(copy);
 }
+#ifdef CONFIG_NFSD_V4_2_INTER_SSC
+
+extern struct file *nfs42_ssc_open(struct vfsmount *ss_mnt,
+				   struct nfs_fh *src_fh,
+				   nfs4_stateid *stateid);
+extern void nfs42_ssc_close(struct file *filep);
+
+extern void nfs_sb_deactive(struct super_block *sb);
+
+#define NFSD42_INTERSSC_MOUNTOPS "vers=4.2,addr=%s,sec=sys"
+
+/**
+ * Support one copy source server for now.
+ */
+static __be32
+nfsd4_interssc_connect(struct nl4_server *nss, struct svc_rqst *rqstp,
+		       struct vfsmount **mount)
+{
+	struct file_system_type *type;
+	struct vfsmount *ss_mnt;
+	struct nfs42_netaddr *naddr;
+	struct sockaddr_storage tmp_addr;
+	size_t tmp_addrlen, match_netid_len = 3;
+	char *startsep = "", *endsep = "", *match_netid = "tcp";
+	char *ipaddr, *dev_name, *raw_data;
+	int len, raw_len;
+	__be32 status = nfserr_inval;
+
+	naddr = &nss->u.nl4_addr;
+	tmp_addrlen = rpc_uaddr2sockaddr(SVC_NET(rqstp), naddr->addr,
+					 naddr->addr_len,
+					 (struct sockaddr *)&tmp_addr,
+					 sizeof(tmp_addr));
+	if (tmp_addrlen == 0)
+		goto out_err;
+
+	if (tmp_addr.ss_family == AF_INET6) {
+		startsep = "[";
+		endsep = "]";
+		match_netid = "tcp6";
+		match_netid_len = 4;
+	}
+
+	if (naddr->netid_len != match_netid_len ||
+		strncmp(naddr->netid, match_netid, naddr->netid_len))
+		goto out_err;
+
+	/* Construct the raw data for the vfs_kern_mount call */
+	len = RPC_MAX_ADDRBUFLEN + 1;
+	ipaddr = kzalloc(len, GFP_KERNEL);
+	if (!ipaddr)
+		goto out_err;
+
+	rpc_ntop((struct sockaddr *)&tmp_addr, ipaddr, len);
+
+	/* 2 for ipv6 endsep and startsep. 3 for ":/" and trailing '/0'*/
+
+	raw_len = strlen(NFSD42_INTERSSC_MOUNTOPS) + strlen(ipaddr);
+	raw_data = kzalloc(raw_len, GFP_KERNEL);
+	if (!raw_data)
+		goto out_free_ipaddr;
+
+	snprintf(raw_data, raw_len, NFSD42_INTERSSC_MOUNTOPS, ipaddr);
+
+	status = nfserr_nodev;
+	type = get_fs_type("nfs");
+	if (!type)
+		goto out_free_rawdata;
+
+	/* Set the server:<export> for the vfs_kern_mount call */
+	dev_name = kzalloc(len + 5, GFP_KERNEL);
+	if (!dev_name)
+		goto out_free_rawdata;
+	snprintf(dev_name, len + 5, "%s%s%s:/", startsep, ipaddr, endsep);
+
+	/* Use an 'internal' mount: SB_KERNMOUNT -> MNT_INTERNAL */
+	ss_mnt = vfs_kern_mount(type, SB_KERNMOUNT, dev_name, raw_data);
+	module_put(type->owner);
+	if (IS_ERR(ss_mnt))
+		goto out_free_devname;
+
+	status = 0;
+	*mount = ss_mnt;
+
+out_free_devname:
+	kfree(dev_name);
+out_free_rawdata:
+	kfree(raw_data);
+out_free_ipaddr:
+	kfree(ipaddr);
+out_err:
+	return status;
+}
+
+static void
+nfsd4_interssc_disconnect(struct vfsmount *ss_mnt)
+{
+	nfs_sb_deactive(ss_mnt->mnt_sb);
+	mntput(ss_mnt);
+}
+
+/**
+ * nfsd4_setup_inter_ssc
+ *
+ * Verify COPY destination stateid.
+ * Connect to the source server with NFSv4.1.
+ * Create the source struct file for nfsd_copy_range.
+ * Called with COPY cstate:
+ *    SAVED_FH: source filehandle
+ *    CURRENT_FH: destination filehandle
+ */
+static __be32
+nfsd4_setup_inter_ssc(struct svc_rqst *rqstp,
+		      struct nfsd4_compound_state *cstate,
+		      struct nfsd4_copy *copy, struct vfsmount **mount)
+{
+	struct svc_fh *s_fh = NULL;
+	stateid_t *s_stid = &copy->cp_src_stateid;
+	__be32 status = nfserr_inval;
+
+	/* Verify the destination stateid and set dst struct file*/
+	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
+					    &copy->cp_dst_stateid,
+					    WR_STATE, &copy->nf_dst, NULL);
+	if (status)
+		goto out;
+
+	status = nfsd4_interssc_connect(&copy->cp_src, rqstp, mount);
+	if (status)
+		goto out;
+
+	s_fh = &cstate->save_fh;
+
+	copy->c_fh.size = s_fh->fh_handle.fh_size;
+	memcpy(copy->c_fh.data, &s_fh->fh_handle.fh_base, copy->c_fh.size);
+	copy->stateid.seqid = cpu_to_be32(s_stid->si_generation);
+	memcpy(copy->stateid.other, (void *)&s_stid->si_opaque,
+	       sizeof(stateid_opaque_t));
+
+	status = 0;
+out:
+	return status;
+}
+
+static void
+nfsd4_cleanup_inter_ssc(struct vfsmount *ss_mnt, struct nfsd_file *src,
+			struct nfsd_file *dst)
+{
+	nfs42_ssc_close(src->nf_file);
+	nfsd_file_put(src);
+	nfsd_file_put(dst);
+	mntput(ss_mnt);
+}
+
+#else /* CONFIG_NFSD_V4_2_INTER_SSC */
+
+static __be32
+nfsd4_setup_inter_ssc(struct svc_rqst *rqstp,
+		      struct nfsd4_compound_state *cstate,
+		      struct nfsd4_copy *copy,
+		      struct vfsmount **mount)
+{
+	*mount = NULL;
+	return nfserr_inval;
+}
+
+static void
+nfsd4_cleanup_inter_ssc(struct vfsmount *ss_mnt, struct nfsd_file *src,
+			struct nfsd_file *dst)
+{
+}
+
+static void
+nfsd4_interssc_disconnect(struct vfsmount *ss_mnt)
+{
+}
+
+static struct file *nfs42_ssc_open(struct vfsmount *ss_mnt,
+				   struct nfs_fh *src_fh,
+				   nfs4_stateid *stateid)
+{
+	return NULL;
+}
+#endif /* CONFIG_NFSD_V4_2_INTER_SSC */
+
+static __be32
+nfsd4_setup_intra_ssc(struct svc_rqst *rqstp,
+		      struct nfsd4_compound_state *cstate,
+		      struct nfsd4_copy *copy)
+{
+	return nfsd4_verify_copy(rqstp, cstate, &copy->cp_src_stateid,
+				 &copy->nf_src, &copy->cp_dst_stateid,
+				 &copy->nf_dst);
+}
+
+static void
+nfsd4_cleanup_intra_ssc(struct nfsd_file *src, struct nfsd_file *dst)
+{
+	nfsd_file_put(src);
+	nfsd_file_put(dst);
+}
 
 static void nfsd4_cb_offload_release(struct nfsd4_callback *cb)
 {
@@ -1199,12 +1410,16 @@ static __be32 nfsd4_do_copy(struct nfsd4_copy *copy, bool sync)
 		status = nfs_ok;
 	}
 
-	nfsd_file_put(copy->nf_src);
-	nfsd_file_put(copy->nf_dst);
+	if (!copy->cp_intra) /* Inter server SSC */
+		nfsd4_cleanup_inter_ssc(copy->ss_mnt, copy->nf_src,
+					copy->nf_dst);
+	else
+		nfsd4_cleanup_intra_ssc(copy->nf_src, copy->nf_dst);
+
 	return status;
 }
 
-static void dup_copy_fields(struct nfsd4_copy *src, struct nfsd4_copy *dst)
+static int dup_copy_fields(struct nfsd4_copy *src, struct nfsd4_copy *dst)
 {
 	dst->cp_src_pos = src->cp_src_pos;
 	dst->cp_dst_pos = src->cp_dst_pos;
@@ -1214,15 +1429,25 @@ static void dup_copy_fields(struct nfsd4_copy *src, struct nfsd4_copy *dst)
 	memcpy(&dst->fh, &src->fh, sizeof(src->fh));
 	dst->cp_clp = src->cp_clp;
 	dst->nf_dst = nfsd_file_get(src->nf_dst);
-	dst->nf_src = nfsd_file_get(src->nf_src);
+	dst->cp_intra = src->cp_intra;
+	if (src->cp_intra) /* for inter, file_src doesn't exist yet */
+		dst->nf_src = nfsd_file_get(src->nf_src);
+
 	memcpy(&dst->cp_stateid, &src->cp_stateid, sizeof(src->cp_stateid));
+	memcpy(&dst->cp_src, &src->cp_src, sizeof(struct nl4_server));
+	memcpy(&dst->stateid, &src->stateid, sizeof(src->stateid));
+	memcpy(&dst->c_fh, &src->c_fh, sizeof(src->c_fh));
+	dst->ss_mnt = src->ss_mnt;
+
+	return 0;
 }
 
 static void cleanup_async_copy(struct nfsd4_copy *copy)
 {
-	nfs4_free_cp_state(copy);
+	nfs4_free_copy_state(copy);
 	nfsd_file_put(copy->nf_dst);
-	nfsd_file_put(copy->nf_src);
+	if (copy->cp_intra)
+		nfsd_file_put(copy->nf_src);
 	spin_lock(&copy->cp_clp->async_lock);
 	list_del(&copy->copies);
 	spin_unlock(&copy->cp_clp->async_lock);
@@ -1234,7 +1459,24 @@ static int nfsd4_do_async_copy(void *data)
 	struct nfsd4_copy *copy = (struct nfsd4_copy *)data;
 	struct nfsd4_copy *cb_copy;
 
+	if (!copy->cp_intra) { /* Inter server SSC */
+		copy->nf_src = kzalloc(sizeof(struct nfsd_file), GFP_KERNEL);
+		if (!copy->nf_src) {
+			copy->nfserr = nfserr_serverfault;
+			nfsd4_interssc_disconnect(copy->ss_mnt);
+			goto do_callback;
+		}
+		copy->nf_src->nf_file = nfs42_ssc_open(copy->ss_mnt, &copy->c_fh,
+					      &copy->stateid);
+		if (IS_ERR(copy->nf_src->nf_file)) {
+			copy->nfserr = nfserr_offload_denied;
+			nfsd4_interssc_disconnect(copy->ss_mnt);
+			goto do_callback;
+		}
+	}
+
 	copy->nfserr = nfsd4_do_copy(copy, 0);
+do_callback:
 	cb_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL);
 	if (!cb_copy)
 		goto out;
@@ -1246,6 +1488,8 @@ static int nfsd4_do_async_copy(void *data)
 			&nfsd4_cb_offload_ops, NFSPROC4_CLNT_CB_OFFLOAD);
 	nfsd4_run_cb(&cb_copy->cp_cb);
 out:
+	if (!copy->cp_intra)
+		kfree(copy->nf_src);
 	cleanup_async_copy(copy);
 	return 0;
 }
@@ -1258,11 +1502,20 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	__be32 status;
 	struct nfsd4_copy *async_copy = NULL;
 
-	status = nfsd4_verify_copy(rqstp, cstate, &copy->cp_src_stateid,
-				   &copy->nf_src, &copy->cp_dst_stateid,
-				   &copy->nf_dst);
-	if (status)
-		goto out;
+	if (!copy->cp_intra) { /* Inter server SSC */
+		if (!inter_copy_offload_enable || copy->cp_synchronous) {
+			status = nfserr_notsupp;
+			goto out;
+		}
+		status = nfsd4_setup_inter_ssc(rqstp, cstate, copy,
+				&copy->ss_mnt);
+		if (status)
+			return nfserr_offload_denied;
+	} else {
+		status = nfsd4_setup_intra_ssc(rqstp, cstate, copy);
+		if (status)
+			return status;
+	}
 
 	copy->cp_clp = cstate->clp;
 	memcpy(&copy->fh, &cstate->current_fh.fh_handle,
@@ -1273,15 +1526,15 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		status = nfserrno(-ENOMEM);
 		async_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL);
 		if (!async_copy)
-			goto out;
-		if (!nfs4_init_cp_state(nn, copy)) {
-			kfree(async_copy);
-			goto out;
-		}
+			goto out_err;
+		if (!nfs4_init_copy_state(nn, copy))
+			goto out_err;
 		refcount_set(&async_copy->refcount, 1);
 		memcpy(&copy->cp_res.cb_stateid, &copy->cp_stateid,
 			sizeof(copy->cp_stateid));
-		dup_copy_fields(copy, async_copy);
+		status = dup_copy_fields(copy, async_copy);
+		if (status)
+			goto out_err;
 		async_copy->copy_task = kthread_create(nfsd4_do_async_copy,
 				async_copy, "%s", "copy thread");
 		if (IS_ERR(async_copy->copy_task))
@@ -1292,12 +1545,17 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		spin_unlock(&async_copy->cp_clp->async_lock);
 		wake_up_process(async_copy->copy_task);
 		status = nfs_ok;
-	} else
+	} else {
 		status = nfsd4_do_copy(copy, 1);
+	}
 out:
 	return status;
 out_err:
-	cleanup_async_copy(async_copy);
+	if (async_copy)
+		cleanup_async_copy(async_copy);
+	status = nfserrno(-ENOMEM);
+	if (!copy->cp_intra)
+		nfsd4_interssc_disconnect(copy->ss_mnt);
 	goto out;
 }
 
@@ -1308,7 +1566,7 @@ find_async_copy(struct nfs4_client *clp, stateid_t *stateid)
 
 	spin_lock(&clp->async_lock);
 	list_for_each_entry(copy, &clp->async_copies, copies) {
-		if (memcmp(&copy->cp_stateid, stateid, NFS4_STATEID_SIZE))
+		if (memcmp(&copy->cp_stateid.stid, stateid, NFS4_STATEID_SIZE))
 			continue;
 		refcount_inc(&copy->refcount);
 		spin_unlock(&clp->async_lock);
@@ -1324,16 +1582,61 @@ nfsd4_offload_cancel(struct svc_rqst *rqstp,
 		     union nfsd4_op_u *u)
 {
 	struct nfsd4_offload_status *os = &u->offload_status;
-	__be32 status = 0;
 	struct nfsd4_copy *copy;
 	struct nfs4_client *clp = cstate->clp;
 
 	copy = find_async_copy(clp, &os->stateid);
-	if (copy)
+	if (!copy) {
+		struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
+
+		return manage_cpntf_state(nn, &os->stateid, clp, NULL);
+	} else
 		nfsd4_stop_copy(copy);
-	else
-		status = nfserr_bad_stateid;
 
+	return nfs_ok;
+}
+
+static __be32
+nfsd4_copy_notify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
+		  union nfsd4_op_u *u)
+{
+	struct nfsd4_copy_notify *cn = &u->copy_notify;
+	__be32 status;
+	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
+	struct nfs4_stid *stid;
+	struct nfs4_cpntf_state *cps;
+	struct nfs4_client *clp = cstate->clp;
+
+	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
+					&cn->cpn_src_stateid, RD_STATE, NULL,
+					&stid);
+	if (status)
+		return status;
+
+	cn->cpn_sec = nn->nfsd4_lease;
+	cn->cpn_nsec = 0;
+
+	status = nfserrno(-ENOMEM);
+	cps = nfs4_alloc_init_cpntf_state(nn, stid);
+	if (!cps)
+		goto out;
+	memcpy(&cn->cpn_cnr_stateid, &cps->cp_stateid.stid, sizeof(stateid_t));
+	memcpy(&cps->cp_p_stateid, &stid->sc_stateid, sizeof(stateid_t));
+	memcpy(&cps->cp_p_clid, &clp->cl_clientid, sizeof(clientid_t));
+
+	/* For now, only return one server address in cpn_src, the
+	 * address used by the client to connect to this server.
+	 */
+	cn->cpn_src.nl4_type = NL4_NETADDR;
+	status = nfsd4_set_netaddr((struct sockaddr *)&rqstp->rq_daddr,
+				 &cn->cpn_src.u.nl4_addr);
+	WARN_ON_ONCE(status);
+	if (status) {
+		nfs4_put_cpntf_state(nn, cps);
+		goto out;
+	}
+out:
+	nfs4_put_stid(stid);
 	return status;
 }
 
@@ -1346,7 +1649,7 @@ nfsd4_fallocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
 					    &fallocate->falloc_stateid,
-					    WR_STATE, &nf);
+					    WR_STATE, &nf, NULL);
 	if (status != nfs_ok) {
 		dprintk("NFSD: nfsd4_fallocate: couldn't process stateid!\n");
 		return status;
@@ -1405,7 +1708,7 @@ nfsd4_seek(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
 					    &seek->seek_stateid,
-					    RD_STATE, &nf);
+					    RD_STATE, &nf, NULL);
 	if (status) {
 		dprintk("NFSD: nfsd4_seek: couldn't process stateid!\n");
 		return status;
@@ -1910,6 +2213,45 @@ static void svcxdr_init_encode(struct svc_rqst *rqstp,
 		- rqstp->rq_auth_slack;
 }
 
+#ifdef CONFIG_NFSD_V4_2_INTER_SSC
+static void
+check_if_stalefh_allowed(struct nfsd4_compoundargs *args)
+{
+	struct nfsd4_op	*op, *current_op = NULL, *saved_op = NULL;
+	struct nfsd4_copy *copy;
+	struct nfsd4_putfh *putfh;
+	int i;
+
+	/* traverse all operation and if it's a COPY compound, mark the
+	 * source filehandle to skip verification
+	 */
+	for (i = 0; i < args->opcnt; i++) {
+		op = &args->ops[i];
+		if (op->opnum == OP_PUTFH)
+			current_op = op;
+		else if (op->opnum == OP_SAVEFH)
+			saved_op = current_op;
+		else if (op->opnum == OP_RESTOREFH)
+			current_op = saved_op;
+		else if (op->opnum == OP_COPY) {
+			copy = (struct nfsd4_copy *)&op->u;
+			if (!saved_op) {
+				op->status = nfserr_nofilehandle;
+				return;
+			}
+			putfh = (struct nfsd4_putfh *)&saved_op->u;
+			if (!copy->cp_intra)
+				putfh->no_verify = true;
+		}
+	}
+}
+#else
+static void
+check_if_stalefh_allowed(struct nfsd4_compoundargs *args)
+{
+}
+#endif
+
 /*
  * COMPOUND call.
  */
@@ -1958,6 +2300,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp)
 		resp->opcnt = 1;
 		goto encode_op;
 	}
+	check_if_stalefh_allowed(args);
 
 	trace_nfsd_compound(rqstp, args->opcnt);
 	while (!status && resp->opcnt < args->opcnt) {
@@ -1973,13 +2316,14 @@ nfsd4_proc_compound(struct svc_rqst *rqstp)
 				op->status = nfsd4_open_omfg(rqstp, cstate, op);
 			goto encode_op;
 		}
-
-		if (!current_fh->fh_dentry) {
+		if (!current_fh->fh_dentry &&
+				!HAS_FH_FLAG(current_fh, NFSD4_FH_FOREIGN)) {
 			if (!(op->opdesc->op_flags & ALLOWED_WITHOUT_FH)) {
 				op->status = nfserr_nofilehandle;
 				goto encode_op;
 			}
-		} else if (current_fh->fh_export->ex_fslocs.migrated &&
+		} else if (current_fh->fh_export &&
+			   current_fh->fh_export->ex_fslocs.migrated &&
 			  !(op->opdesc->op_flags & ALLOWED_ON_ABSENT_FS)) {
 			op->status = nfserr_moved;
 			goto encode_op;
@@ -2023,7 +2367,8 @@ nfsd4_proc_compound(struct svc_rqst *rqstp)
 			if (op->opdesc->op_flags & OP_CLEAR_STATEID)
 				clear_current_stateid(cstate);
 
-			if (need_wrongsec_check(rqstp))
+			if (current_fh->fh_export &&
+					need_wrongsec_check(rqstp))
 				op->status = check_nfsd_access(current_fh->fh_export, rqstp);
 		}
 encode_op:
@@ -2290,6 +2635,21 @@ static inline u32 nfsd4_offload_status_rsize(struct svc_rqst *rqstp,
 		1 /* osr_complete<1> optional 0 for now */) * sizeof(__be32);
 }
 
+static inline u32 nfsd4_copy_notify_rsize(struct svc_rqst *rqstp,
+					struct nfsd4_op *op)
+{
+	return (op_encode_hdr_size +
+		3 /* cnr_lease_time */ +
+		1 /* We support one cnr_source_server */ +
+		1 /* cnr_stateid seq */ +
+		op_encode_stateid_maxsz /* cnr_stateid */ +
+		1 /* num cnr_source_server*/ +
+		1 /* nl4_type */ +
+		1 /* nl4 size */ +
+		XDR_QUADLEN(NFS4_OPAQUE_LIMIT) /*nl4_loc + nl4_loc_sz */)
+		* sizeof(__be32);
+}
+
 #ifdef CONFIG_NFSD_PNFS
 static inline u32 nfsd4_getdeviceinfo_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
 {
@@ -2714,6 +3074,12 @@ static const struct nfsd4_operation nfsd4_ops[] = {
 		.op_name = "OP_OFFLOAD_CANCEL",
 		.op_rsize_bop = nfsd4_only_status_rsize,
 	},
+	[OP_COPY_NOTIFY] = {
+		.op_func = nfsd4_copy_notify,
+		.op_flags = OP_MODIFIES_SOMETHING,
+		.op_name = "OP_COPY_NOTIFY",
+		.op_rsize_bop = nfsd4_copy_notify_rsize,
+	},
 };
 
 /**
diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index cdc75ad4438b..9e40dfecf1b1 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -127,16 +127,8 @@ nfs4_make_rec_clidname(char *dname, const struct xdr_netobj *clname)
  		goto out;
 	}
 
-	{
-		SHASH_DESC_ON_STACK(desc, tfm);
-
-		desc->tfm = tfm;
-
-		status = crypto_shash_digest(desc, clname->data, clname->len,
-					     cksum.data);
-		shash_desc_zero(desc);
-	}
-
+	status = crypto_shash_tfm_digest(tfm, clname->data, clname->len,
+					 cksum.data);
 	if (status)
 		goto out;
 
@@ -1148,7 +1140,6 @@ nfsd4_cld_create_v2(struct nfs4_client *clp)
 	struct crypto_shash *tfm = cn->cn_tfm;
 	struct xdr_netobj cksum;
 	char *principal = NULL;
-	SHASH_DESC_ON_STACK(desc, tfm);
 
 	/* Don't upcall if it's already stored */
 	if (test_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags))
@@ -1170,16 +1161,14 @@ nfsd4_cld_create_v2(struct nfs4_client *clp)
 	else if (clp->cl_cred.cr_principal)
 		principal = clp->cl_cred.cr_principal;
 	if (principal) {
-		desc->tfm = tfm;
 		cksum.len = crypto_shash_digestsize(tfm);
 		cksum.data = kmalloc(cksum.len, GFP_KERNEL);
 		if (cksum.data == NULL) {
 			ret = -ENOMEM;
 			goto out;
 		}
-		ret = crypto_shash_digest(desc, principal, strlen(principal),
-					  cksum.data);
-		shash_desc_zero(desc);
+		ret = crypto_shash_tfm_digest(tfm, principal, strlen(principal),
+					      cksum.data);
 		if (ret) {
 			kfree(cksum.data);
 			goto out;
@@ -1343,7 +1332,6 @@ nfsd4_cld_check_v2(struct nfs4_client *clp)
 	struct crypto_shash *tfm = cn->cn_tfm;
 	struct xdr_netobj cksum;
 	char *principal = NULL;
-	SHASH_DESC_ON_STACK(desc, tfm);
 
 	/* did we already find that this client is stable? */
 	if (test_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags))
@@ -1381,14 +1369,12 @@ found:
 			principal = clp->cl_cred.cr_principal;
 		if (principal == NULL)
 			return -ENOENT;
-		desc->tfm = tfm;
 		cksum.len = crypto_shash_digestsize(tfm);
 		cksum.data = kmalloc(cksum.len, GFP_KERNEL);
 		if (cksum.data == NULL)
 			return -ENOENT;
-		status = crypto_shash_digest(desc, principal, strlen(principal),
-					     cksum.data);
-		shash_desc_zero(desc);
+		status = crypto_shash_tfm_digest(tfm, principal,
+						 strlen(principal), cksum.data);
 		if (status) {
 			kfree(cksum.data);
 			return -ENOENT;
@@ -1445,7 +1431,7 @@ nfsd4_cld_grace_done_v0(struct nfsd_net *nn)
 	}
 
 	cup->cu_u.cu_msg.cm_cmd = Cld_GraceDone;
-	cup->cu_u.cu_msg.cm_u.cm_gracetime = (int64_t)nn->boot_time;
+	cup->cu_u.cu_msg.cm_u.cm_gracetime = nn->boot_time;
 	ret = cld_pipe_upcall(cn->cn_pipe, &cup->cu_u.cu_msg);
 	if (!ret)
 		ret = cup->cu_u.cu_msg.cm_status;
@@ -1578,6 +1564,7 @@ nfsd4_cld_tracking_init(struct net *net)
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 	bool running;
 	int retries = 10;
+	struct crypto_shash *tfm;
 
 	status = nfs4_cld_state_init(net);
 	if (status)
@@ -1586,11 +1573,6 @@ nfsd4_cld_tracking_init(struct net *net)
 	status = __nfsd4_init_cld_pipe(net);
 	if (status)
 		goto err_shutdown;
-	nn->cld_net->cn_tfm = crypto_alloc_shash("sha256", 0, 0);
-	if (IS_ERR(nn->cld_net->cn_tfm)) {
-		status = PTR_ERR(nn->cld_net->cn_tfm);
-		goto err_remove;
-	}
 
 	/*
 	 * rpc pipe upcalls take 30 seconds to time out, so we don't want to
@@ -1607,6 +1589,12 @@ nfsd4_cld_tracking_init(struct net *net)
 		status = -ETIMEDOUT;
 		goto err_remove;
 	}
+	tfm = crypto_alloc_shash("sha256", 0, 0);
+	if (IS_ERR(tfm)) {
+		status = PTR_ERR(tfm);
+		goto err_remove;
+	}
+	nn->cld_net->cn_tfm = tfm;
 
 	status = nfsd4_cld_get_version(nn);
 	if (status == -EOPNOTSUPP)
@@ -1780,7 +1768,7 @@ nfsd4_cltrack_client_has_session(struct nfs4_client *clp)
 }
 
 static char *
-nfsd4_cltrack_grace_start(time_t grace_start)
+nfsd4_cltrack_grace_start(time64_t grace_start)
 {
 	int copied;
 	size_t len;
@@ -1793,7 +1781,7 @@ nfsd4_cltrack_grace_start(time_t grace_start)
 	if (!result)
 		return result;
 
-	copied = snprintf(result, len, GRACE_START_ENV_PREFIX "%ld",
+	copied = snprintf(result, len, GRACE_START_ENV_PREFIX "%lld",
 				grace_start);
 	if (copied >= len) {
 		/* just return nothing if output was truncated */
@@ -1850,19 +1838,14 @@ nfsd4_umh_cltrack_upcall(char *cmd, char *arg, char *env0, char *env1)
 static char *
 bin_to_hex_dup(const unsigned char *src, int srclen)
 {
-	int i;
-	char *buf, *hex;
+	char *buf;
 
 	/* +1 for terminating NULL */
-	buf = kmalloc((srclen * 2) + 1, GFP_KERNEL);
+	buf = kzalloc((srclen * 2) + 1, GFP_KERNEL);
 	if (!buf)
 		return buf;
 
-	hex = buf;
-	for (i = 0; i < srclen; i++) {
-		sprintf(hex, "%2.2x", *src++);
-		hex += 2;
-	}
+	bin2hex(buf, src, srclen);
 	return buf;
 }
 
@@ -2007,7 +1990,7 @@ nfsd4_umh_cltrack_grace_done(struct nfsd_net *nn)
 	char *legacy;
 	char timestr[22]; /* FIXME: better way to determine max size? */
 
-	sprintf(timestr, "%ld", nn->boot_time);
+	sprintf(timestr, "%lld", nn->boot_time);
 	legacy = nfsd4_cltrack_legacy_topdir();
 	nfsd4_umh_cltrack_upcall("gracedone", timestr, legacy, NULL);
 	kfree(legacy);
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index c65aeaa812d4..c107caa56525 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -80,6 +80,7 @@ static u64 current_sessionid = 1;
 static bool check_for_locks(struct nfs4_file *fp, struct nfs4_lockowner *lowner);
 static void nfs4_free_ol_stateid(struct nfs4_stid *stid);
 void nfsd4_end_grace(struct nfsd_net *nn);
+static void _free_cpntf_state_locked(struct nfsd_net *nn, struct nfs4_cpntf_state *cps);
 
 /* Locking: */
 
@@ -170,7 +171,7 @@ renew_client_locked(struct nfs4_client *clp)
 			clp->cl_clientid.cl_boot,
 			clp->cl_clientid.cl_id);
 	list_move_tail(&clp->cl_lru, &nn->client_lru);
-	clp->cl_time = get_seconds();
+	clp->cl_time = ktime_get_boottime_seconds();
 }
 
 static void put_client_renew_locked(struct nfs4_client *clp)
@@ -266,6 +267,8 @@ find_or_allocate_block(struct nfs4_lockowner *lo, struct knfsd_fh *fh,
 	if (!nbl) {
 		nbl= kmalloc(sizeof(*nbl), GFP_KERNEL);
 		if (nbl) {
+			INIT_LIST_HEAD(&nbl->nbl_list);
+			INIT_LIST_HEAD(&nbl->nbl_lru);
 			fh_copy_shallow(&nbl->nbl_fh, fh);
 			locks_init_lock(&nbl->nbl_lock);
 			nfsd4_init_cb(&nbl->nbl_cb, lo->lo_owner.so_client,
@@ -493,6 +496,8 @@ find_any_file(struct nfs4_file *f)
 {
 	struct nfsd_file *ret;
 
+	if (!f)
+		return NULL;
 	spin_lock(&f->fi_lock);
 	ret = __nfs4_get_fd(f, O_RDWR);
 	if (!ret) {
@@ -722,6 +727,7 @@ struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct kmem_cache *sla
 	/* Will be incremented before return to client: */
 	refcount_set(&stid->sc_count, 1);
 	spin_lock_init(&stid->sc_lock);
+	INIT_LIST_HEAD(&stid->sc_cp_list);
 
 	/*
 	 * It shouldn't be a problem to reuse an opaque stateid value.
@@ -741,30 +747,76 @@ out_free:
 /*
  * Create a unique stateid_t to represent each COPY.
  */
-int nfs4_init_cp_state(struct nfsd_net *nn, struct nfsd4_copy *copy)
+static int nfs4_init_cp_state(struct nfsd_net *nn, copy_stateid_t *stid,
+			      unsigned char sc_type)
 {
 	int new_id;
 
+	stid->stid.si_opaque.so_clid.cl_boot = (u32)nn->boot_time;
+	stid->stid.si_opaque.so_clid.cl_id = nn->s2s_cp_cl_id;
+	stid->sc_type = sc_type;
+
 	idr_preload(GFP_KERNEL);
 	spin_lock(&nn->s2s_cp_lock);
-	new_id = idr_alloc_cyclic(&nn->s2s_cp_stateids, copy, 0, 0, GFP_NOWAIT);
+	new_id = idr_alloc_cyclic(&nn->s2s_cp_stateids, stid, 0, 0, GFP_NOWAIT);
+	stid->stid.si_opaque.so_id = new_id;
 	spin_unlock(&nn->s2s_cp_lock);
 	idr_preload_end();
 	if (new_id < 0)
 		return 0;
-	copy->cp_stateid.si_opaque.so_id = new_id;
-	copy->cp_stateid.si_opaque.so_clid.cl_boot = nn->boot_time;
-	copy->cp_stateid.si_opaque.so_clid.cl_id = nn->s2s_cp_cl_id;
 	return 1;
 }
 
-void nfs4_free_cp_state(struct nfsd4_copy *copy)
+int nfs4_init_copy_state(struct nfsd_net *nn, struct nfsd4_copy *copy)
+{
+	return nfs4_init_cp_state(nn, &copy->cp_stateid, NFS4_COPY_STID);
+}
+
+struct nfs4_cpntf_state *nfs4_alloc_init_cpntf_state(struct nfsd_net *nn,
+						     struct nfs4_stid *p_stid)
+{
+	struct nfs4_cpntf_state *cps;
+
+	cps = kzalloc(sizeof(struct nfs4_cpntf_state), GFP_KERNEL);
+	if (!cps)
+		return NULL;
+	cps->cpntf_time = ktime_get_boottime_seconds();
+	refcount_set(&cps->cp_stateid.sc_count, 1);
+	if (!nfs4_init_cp_state(nn, &cps->cp_stateid, NFS4_COPYNOTIFY_STID))
+		goto out_free;
+	spin_lock(&nn->s2s_cp_lock);
+	list_add(&cps->cp_list, &p_stid->sc_cp_list);
+	spin_unlock(&nn->s2s_cp_lock);
+	return cps;
+out_free:
+	kfree(cps);
+	return NULL;
+}
+
+void nfs4_free_copy_state(struct nfsd4_copy *copy)
 {
 	struct nfsd_net *nn;
 
+	WARN_ON_ONCE(copy->cp_stateid.sc_type != NFS4_COPY_STID);
 	nn = net_generic(copy->cp_clp->net, nfsd_net_id);
 	spin_lock(&nn->s2s_cp_lock);
-	idr_remove(&nn->s2s_cp_stateids, copy->cp_stateid.si_opaque.so_id);
+	idr_remove(&nn->s2s_cp_stateids,
+		   copy->cp_stateid.stid.si_opaque.so_id);
+	spin_unlock(&nn->s2s_cp_lock);
+}
+
+static void nfs4_free_cpntf_statelist(struct net *net, struct nfs4_stid *stid)
+{
+	struct nfs4_cpntf_state *cps;
+	struct nfsd_net *nn;
+
+	nn = net_generic(net, nfsd_net_id);
+	spin_lock(&nn->s2s_cp_lock);
+	while (!list_empty(&stid->sc_cp_list)) {
+		cps = list_first_entry(&stid->sc_cp_list,
+				       struct nfs4_cpntf_state, cp_list);
+		_free_cpntf_state_locked(nn, cps);
+	}
 	spin_unlock(&nn->s2s_cp_lock);
 }
 
@@ -806,7 +858,7 @@ static void nfs4_free_deleg(struct nfs4_stid *stid)
 static DEFINE_SPINLOCK(blocked_delegations_lock);
 static struct bloom_pair {
 	int	entries, old_entries;
-	time_t	swap_time;
+	time64_t swap_time;
 	int	new; /* index into 'set' */
 	DECLARE_BITMAP(set[2], 256);
 } blocked_delegations;
@@ -818,15 +870,15 @@ static int delegation_blocked(struct knfsd_fh *fh)
 
 	if (bd->entries == 0)
 		return 0;
-	if (seconds_since_boot() - bd->swap_time > 30) {
+	if (ktime_get_seconds() - bd->swap_time > 30) {
 		spin_lock(&blocked_delegations_lock);
-		if (seconds_since_boot() - bd->swap_time > 30) {
+		if (ktime_get_seconds() - bd->swap_time > 30) {
 			bd->entries -= bd->old_entries;
 			bd->old_entries = bd->entries;
 			memset(bd->set[bd->new], 0,
 			       sizeof(bd->set[0]));
 			bd->new = 1-bd->new;
-			bd->swap_time = seconds_since_boot();
+			bd->swap_time = ktime_get_seconds();
 		}
 		spin_unlock(&blocked_delegations_lock);
 	}
@@ -856,7 +908,7 @@ static void block_delegations(struct knfsd_fh *fh)
 	__set_bit((hash>>8)&255, bd->set[bd->new]);
 	__set_bit((hash>>16)&255, bd->set[bd->new]);
 	if (bd->entries == 0)
-		bd->swap_time = seconds_since_boot();
+		bd->swap_time = ktime_get_seconds();
 	bd->entries += 1;
 	spin_unlock(&blocked_delegations_lock);
 }
@@ -915,6 +967,7 @@ nfs4_put_stid(struct nfs4_stid *s)
 		return;
 	}
 	idr_remove(&clp->cl_stateids, s->sc_stateid.si_opaque.so_id);
+	nfs4_free_cpntf_statelist(clp->net, s);
 	spin_unlock(&clp->cl_lock);
 	s->sc_free(s);
 	if (fp)
@@ -1260,6 +1313,12 @@ static void nfs4_put_stateowner(struct nfs4_stateowner *sop)
 	nfs4_free_stateowner(sop);
 }
 
+static bool
+nfs4_ol_stateid_unhashed(const struct nfs4_ol_stateid *stp)
+{
+	return list_empty(&stp->st_perfile);
+}
+
 static bool unhash_ol_stateid(struct nfs4_ol_stateid *stp)
 {
 	struct nfs4_file *fp = stp->st_stid.sc_file;
@@ -1330,9 +1389,11 @@ static bool unhash_lock_stateid(struct nfs4_ol_stateid *stp)
 {
 	lockdep_assert_held(&stp->st_stid.sc_client->cl_lock);
 
+	if (!unhash_ol_stateid(stp))
+		return false;
 	list_del_init(&stp->st_locks);
 	nfs4_unhash_stid(&stp->st_stid);
-	return unhash_ol_stateid(stp);
+	return true;
 }
 
 static void release_lock_stateid(struct nfs4_ol_stateid *stp)
@@ -1397,13 +1458,12 @@ static void release_open_stateid_locks(struct nfs4_ol_stateid *open_stp,
 static bool unhash_open_stateid(struct nfs4_ol_stateid *stp,
 				struct list_head *reaplist)
 {
-	bool unhashed;
-
 	lockdep_assert_held(&stp->st_stid.sc_client->cl_lock);
 
-	unhashed = unhash_ol_stateid(stp);
+	if (!unhash_ol_stateid(stp))
+		return false;
 	release_open_stateid_locks(stp, reaplist);
-	return unhashed;
+	return true;
 }
 
 static void release_open_stateid(struct nfs4_ol_stateid *stp)
@@ -1862,7 +1922,7 @@ STALE_CLIENTID(clientid_t *clid, struct nfsd_net *nn)
 	 */
 	if (clid->cl_boot == (u32)nn->boot_time)
 		return 0;
-	dprintk("NFSD stale clientid (%08x/%08x) boot_time %08lx\n",
+	dprintk("NFSD stale clientid (%08x/%08x) boot_time %08llx\n",
 		clid->cl_boot, clid->cl_id, nn->boot_time);
 	return 1;
 }
@@ -2215,14 +2275,14 @@ static void gen_confirm(struct nfs4_client *clp, struct nfsd_net *nn)
 	 * This is opaque to client, so no need to byte-swap. Use
 	 * __force to keep sparse happy
 	 */
-	verf[0] = (__force __be32)get_seconds();
+	verf[0] = (__force __be32)(u32)ktime_get_real_seconds();
 	verf[1] = (__force __be32)nn->clverifier_counter++;
 	memcpy(clp->cl_confirm.data, verf, sizeof(clp->cl_confirm.data));
 }
 
 static void gen_clid(struct nfs4_client *clp, struct nfsd_net *nn)
 {
-	clp->cl_clientid.cl_boot = nn->boot_time;
+	clp->cl_clientid.cl_boot = (u32)nn->boot_time;
 	clp->cl_clientid.cl_id = nn->clientid_counter++;
 	gen_confirm(clp, nn);
 }
@@ -2292,7 +2352,7 @@ static int client_info_show(struct seq_file *m, void *v)
 					clp->cl_nii_domain.len);
 		seq_printf(m, "\nImplementation name: ");
 		seq_quote_mem(m, clp->cl_nii_name.data, clp->cl_nii_name.len);
-		seq_printf(m, "\nImplementation time: [%ld, %ld]\n",
+		seq_printf(m, "\nImplementation time: [%lld, %ld]\n",
 			clp->cl_nii_time.tv_sec, clp->cl_nii_time.tv_nsec);
 	}
 	drop_client(clp);
@@ -2382,10 +2442,10 @@ static int nfs4_show_open(struct seq_file *s, struct nfs4_stid *st)
 	access = bmap_to_share_mode(ols->st_access_bmap);
 	deny   = bmap_to_share_mode(ols->st_deny_bmap);
 
-	seq_printf(s, "access: \%s\%s, ",
+	seq_printf(s, "access: %s%s, ",
 		access & NFS4_SHARE_ACCESS_READ ? "r" : "-",
 		access & NFS4_SHARE_ACCESS_WRITE ? "w" : "-");
-	seq_printf(s, "deny: \%s\%s, ",
+	seq_printf(s, "deny: %s%s, ",
 		deny & NFS4_SHARE_ACCESS_READ ? "r" : "-",
 		deny & NFS4_SHARE_ACCESS_WRITE ? "w" : "-");
 
@@ -2587,7 +2647,7 @@ static const struct file_operations client_ctl_fops = {
 static const struct tree_descr client_files[] = {
 	[0] = {"info", &client_info_fops, S_IRUSR},
 	[1] = {"states", &client_states_fops, S_IRUSR},
-	[2] = {"ctl", &client_ctl_fops, S_IRUSR|S_IWUSR},
+	[2] = {"ctl", &client_ctl_fops, S_IWUSR},
 	[3] = {""},
 };
 
@@ -2612,7 +2672,7 @@ static struct nfs4_client *create_client(struct xdr_netobj name,
 	gen_clid(clp, nn);
 	kref_init(&clp->cl_nfsdfs.cl_ref);
 	nfsd4_init_cb(&clp->cl_cb_null, clp, NULL, NFSPROC4_CLNT_CB_NULL);
-	clp->cl_time = get_seconds();
+	clp->cl_time = ktime_get_boottime_seconds();
 	clear_bit(0, &clp->cl_cb_slot_busy);
 	copy_verf(clp, verf);
 	memcpy(&clp->cl_addr, sa, sizeof(struct sockaddr_storage));
@@ -2946,8 +3006,7 @@ static __be32 copy_impl_id(struct nfs4_client *clp,
 	xdr_netobj_dup(&clp->cl_nii_name, &exid->nii_name, GFP_KERNEL);
 	if (!clp->cl_nii_name.data)
 		return nfserr_jukebox;
-	clp->cl_nii_time.tv_sec = exid->nii_time.tv_sec;
-	clp->cl_nii_time.tv_nsec = exid->nii_time.tv_nsec;
+	clp->cl_nii_time = exid->nii_time;
 	return 0;
 }
 
@@ -3373,7 +3432,7 @@ static __be32 nfsd4_map_bcts_dir(u32 *dir)
 	case NFS4_CDFC4_BACK_OR_BOTH:
 		*dir = NFS4_CDFC4_BOTH;
 		return nfs_ok;
-	};
+	}
 	return nfserr_inval;
 }
 
@@ -3548,12 +3607,17 @@ static bool replay_matches_cache(struct svc_rqst *rqstp,
 	    (bool)seq->cachethis)
 		return false;
 	/*
-	 * If there's an error than the reply can have fewer ops than
-	 * the call.  But if we cached a reply with *more* ops than the
-	 * call you're sending us now, then this new call is clearly not
-	 * really a replay of the old one:
+	 * If there's an error then the reply can have fewer ops than
+	 * the call.
 	 */
-	if (slot->sl_opcnt < argp->opcnt)
+	if (slot->sl_opcnt < argp->opcnt && !slot->sl_status)
+		return false;
+	/*
+	 * But if we cached a reply with *more* ops than the call you're
+	 * sending us now, then this new call is clearly not really a
+	 * replay of the old one:
+	 */
+	if (slot->sl_opcnt > argp->opcnt)
 		return false;
 	/* This is the only check explicitly called by spec: */
 	if (!same_creds(&rqstp->rq_cred, &slot->sl_cred))
@@ -4278,7 +4342,7 @@ move_to_close_lru(struct nfs4_ol_stateid *s, struct net *net)
 	last = oo->oo_last_closed_stid;
 	oo->oo_last_closed_stid = s;
 	list_move_tail(&oo->oo_close_lru, &nn->close_lru);
-	oo->oo_time = get_seconds();
+	oo->oo_time = ktime_get_boottime_seconds();
 	spin_unlock(&nn->client_lock);
 	if (last)
 		nfs4_put_stid(&last->st_stid);
@@ -4290,7 +4354,8 @@ find_file_locked(struct knfsd_fh *fh, unsigned int hashval)
 {
 	struct nfs4_file *fp;
 
-	hlist_for_each_entry_rcu(fp, &file_hashtbl[hashval], fi_hash) {
+	hlist_for_each_entry_rcu(fp, &file_hashtbl[hashval], fi_hash,
+				lockdep_is_held(&state_lock)) {
 		if (fh_match(&fp->fi_fhandle, fh)) {
 			if (refcount_inc_not_zero(&fp->fi_ref))
 				return fp;
@@ -4373,7 +4438,7 @@ static void nfsd4_cb_recall_prepare(struct nfsd4_callback *cb)
 	 */
 	spin_lock(&state_lock);
 	if (dp->dl_time == 0) {
-		dp->dl_time = get_seconds();
+		dp->dl_time = ktime_get_boottime_seconds();
 		list_add_tail(&dp->dl_recall_lru, &nn->del_recall_lru);
 	}
 	spin_unlock(&state_lock);
@@ -4485,7 +4550,8 @@ static __be32 nfsd4_check_seqid(struct nfsd4_compound_state *cstate, struct nfs4
 
 static __be32 lookup_clientid(clientid_t *clid,
 		struct nfsd4_compound_state *cstate,
-		struct nfsd_net *nn)
+		struct nfsd_net *nn,
+		bool sessions)
 {
 	struct nfs4_client *found;
 
@@ -4506,7 +4572,7 @@ static __be32 lookup_clientid(clientid_t *clid,
 	 */
 	WARN_ON_ONCE(cstate->session);
 	spin_lock(&nn->client_lock);
-	found = find_confirmed_client(clid, false, nn);
+	found = find_confirmed_client(clid, sessions, nn);
 	if (!found) {
 		spin_unlock(&nn->client_lock);
 		return nfserr_expired;
@@ -4539,7 +4605,7 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate,
 	if (open->op_file == NULL)
 		return nfserr_jukebox;
 
-	status = lookup_clientid(clientid, cstate, nn);
+	status = lookup_clientid(clientid, cstate, nn, false);
 	if (status)
 		return status;
 	clp = cstate->clp;
@@ -4667,7 +4733,7 @@ nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh,
 		return 0;
 	if (!(open->op_share_access & NFS4_SHARE_ACCESS_WRITE))
 		return nfserr_inval;
-	return nfsd_setattr(rqstp, fh, &iattr, 0, (time_t)0);
+	return nfsd_setattr(rqstp, fh, &iattr, 0, (time64_t)0);
 }
 
 static __be32 nfs4_get_vfs_file(struct svc_rqst *rqstp, struct nfs4_file *fp,
@@ -5128,7 +5194,7 @@ nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	dprintk("process_renew(%08x/%08x): starting\n", 
 			clid->cl_boot, clid->cl_id);
-	status = lookup_clientid(clid, cstate, nn);
+	status = lookup_clientid(clid, cstate, nn, false);
 	if (status)
 		goto out;
 	clp = cstate->clp;
@@ -5179,9 +5245,8 @@ nfsd4_end_grace(struct nfsd_net *nn)
  */
 static bool clients_still_reclaiming(struct nfsd_net *nn)
 {
-	unsigned long now = get_seconds();
-	unsigned long double_grace_period_end = nn->boot_time +
-						2 * nn->nfsd4_lease;
+	time64_t double_grace_period_end = nn->boot_time +
+					   2 * nn->nfsd4_lease;
 
 	if (nn->track_reclaim_completes &&
 			atomic_read(&nn->nr_reclaim_complete) ==
@@ -5194,12 +5259,12 @@ static bool clients_still_reclaiming(struct nfsd_net *nn)
 	 * If we've given them *two* lease times to reclaim, and they're
 	 * still not done, give up:
 	 */
-	if (time_after(now, double_grace_period_end))
+	if (ktime_get_boottime_seconds() > double_grace_period_end)
 		return false;
 	return true;
 }
 
-static time_t
+static time64_t
 nfs4_laundromat(struct nfsd_net *nn)
 {
 	struct nfs4_client *clp;
@@ -5208,8 +5273,11 @@ nfs4_laundromat(struct nfsd_net *nn)
 	struct nfs4_ol_stateid *stp;
 	struct nfsd4_blocked_lock *nbl;
 	struct list_head *pos, *next, reaplist;
-	time_t cutoff = get_seconds() - nn->nfsd4_lease;
-	time_t t, new_timeo = nn->nfsd4_lease;
+	time64_t cutoff = ktime_get_boottime_seconds() - nn->nfsd4_lease;
+	time64_t t, new_timeo = nn->nfsd4_lease;
+	struct nfs4_cpntf_state *cps;
+	copy_stateid_t *cps_t;
+	int i;
 
 	dprintk("NFSD: laundromat service - starting\n");
 
@@ -5220,10 +5288,20 @@ nfs4_laundromat(struct nfsd_net *nn)
 	dprintk("NFSD: end of grace period\n");
 	nfsd4_end_grace(nn);
 	INIT_LIST_HEAD(&reaplist);
+
+	spin_lock(&nn->s2s_cp_lock);
+	idr_for_each_entry(&nn->s2s_cp_stateids, cps_t, i) {
+		cps = container_of(cps_t, struct nfs4_cpntf_state, cp_stateid);
+		if (cps->cp_stateid.sc_type == NFS4_COPYNOTIFY_STID &&
+				cps->cpntf_time > cutoff)
+			_free_cpntf_state_locked(nn, cps);
+	}
+	spin_unlock(&nn->s2s_cp_lock);
+
 	spin_lock(&nn->client_lock);
 	list_for_each_safe(pos, next, &nn->client_lru) {
 		clp = list_entry(pos, struct nfs4_client, cl_lru);
-		if (time_after((unsigned long)clp->cl_time, (unsigned long)cutoff)) {
+		if (clp->cl_time > cutoff) {
 			t = clp->cl_time - cutoff;
 			new_timeo = min(new_timeo, t);
 			break;
@@ -5246,7 +5324,7 @@ nfs4_laundromat(struct nfsd_net *nn)
 	spin_lock(&state_lock);
 	list_for_each_safe(pos, next, &nn->del_recall_lru) {
 		dp = list_entry (pos, struct nfs4_delegation, dl_recall_lru);
-		if (time_after((unsigned long)dp->dl_time, (unsigned long)cutoff)) {
+		if (dp->dl_time > cutoff) {
 			t = dp->dl_time - cutoff;
 			new_timeo = min(new_timeo, t);
 			break;
@@ -5266,8 +5344,7 @@ nfs4_laundromat(struct nfsd_net *nn)
 	while (!list_empty(&nn->close_lru)) {
 		oo = list_first_entry(&nn->close_lru, struct nfs4_openowner,
 					oo_close_lru);
-		if (time_after((unsigned long)oo->oo_time,
-			       (unsigned long)cutoff)) {
+		if (oo->oo_time > cutoff) {
 			t = oo->oo_time - cutoff;
 			new_timeo = min(new_timeo, t);
 			break;
@@ -5297,8 +5374,7 @@ nfs4_laundromat(struct nfsd_net *nn)
 	while (!list_empty(&nn->blocked_locks_lru)) {
 		nbl = list_first_entry(&nn->blocked_locks_lru,
 					struct nfsd4_blocked_lock, nbl_lru);
-		if (time_after((unsigned long)nbl->nbl_time,
-			       (unsigned long)cutoff)) {
+		if (nbl->nbl_time > cutoff) {
 			t = nbl->nbl_time - cutoff;
 			new_timeo = min(new_timeo, t);
 			break;
@@ -5315,7 +5391,7 @@ nfs4_laundromat(struct nfsd_net *nn)
 		free_blocked_lock(nbl);
 	}
 out:
-	new_timeo = max_t(time_t, new_timeo, NFSD_LAUNDROMAT_MINTIMEOUT);
+	new_timeo = max_t(time64_t, new_timeo, NFSD_LAUNDROMAT_MINTIMEOUT);
 	return new_timeo;
 }
 
@@ -5325,13 +5401,13 @@ static void laundromat_main(struct work_struct *);
 static void
 laundromat_main(struct work_struct *laundry)
 {
-	time_t t;
+	time64_t t;
 	struct delayed_work *dwork = to_delayed_work(laundry);
 	struct nfsd_net *nn = container_of(dwork, struct nfsd_net,
 					   laundromat_work);
 
 	t = nfs4_laundromat(nn);
-	dprintk("NFSD: laundromat_main - sleeping for %ld seconds\n", t);
+	dprintk("NFSD: laundromat_main - sleeping for %lld seconds\n", t);
 	queue_delayed_work(laundry_wq, &nn->laundromat_work, t*HZ);
 }
 
@@ -5457,15 +5533,8 @@ static __be32 nfsd4_validate_stateid(struct nfs4_client *cl, stateid_t *stateid)
 	if (ZERO_STATEID(stateid) || ONE_STATEID(stateid) ||
 		CLOSE_STATEID(stateid))
 		return status;
-	/* Client debugging aid. */
-	if (!same_clid(&stateid->si_opaque.so_clid, &cl->cl_clientid)) {
-		char addr_str[INET6_ADDRSTRLEN];
-		rpc_ntop((struct sockaddr *)&cl->cl_addr, addr_str,
-				 sizeof(addr_str));
-		pr_warn_ratelimited("NFSD: client %s testing state ID "
-					"with incorrect client ID\n", addr_str);
+	if (!same_clid(&stateid->si_opaque.so_clid, &cl->cl_clientid))
 		return status;
-	}
 	spin_lock(&cl->cl_lock);
 	s = find_stateid_locked(cl, stateid);
 	if (!s)
@@ -5516,7 +5585,8 @@ nfsd4_lookup_stateid(struct nfsd4_compound_state *cstate,
 	if (ZERO_STATEID(stateid) || ONE_STATEID(stateid) ||
 		CLOSE_STATEID(stateid))
 		return nfserr_bad_stateid;
-	status = lookup_clientid(&stateid->si_opaque.so_clid, cstate, nn);
+	status = lookup_clientid(&stateid->si_opaque.so_clid, cstate, nn,
+				 false);
 	if (status == nfserr_stale_clientid) {
 		if (cstate->session)
 			return nfserr_bad_stateid;
@@ -5595,6 +5665,85 @@ nfs4_check_file(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfs4_stid *s,
 out:
 	return status;
 }
+static void
+_free_cpntf_state_locked(struct nfsd_net *nn, struct nfs4_cpntf_state *cps)
+{
+	WARN_ON_ONCE(cps->cp_stateid.sc_type != NFS4_COPYNOTIFY_STID);
+	if (!refcount_dec_and_test(&cps->cp_stateid.sc_count))
+		return;
+	list_del(&cps->cp_list);
+	idr_remove(&nn->s2s_cp_stateids,
+		   cps->cp_stateid.stid.si_opaque.so_id);
+	kfree(cps);
+}
+/*
+ * A READ from an inter server to server COPY will have a
+ * copy stateid. Look up the copy notify stateid from the
+ * idr structure and take a reference on it.
+ */
+__be32 manage_cpntf_state(struct nfsd_net *nn, stateid_t *st,
+			  struct nfs4_client *clp,
+			  struct nfs4_cpntf_state **cps)
+{
+	copy_stateid_t *cps_t;
+	struct nfs4_cpntf_state *state = NULL;
+
+	if (st->si_opaque.so_clid.cl_id != nn->s2s_cp_cl_id)
+		return nfserr_bad_stateid;
+	spin_lock(&nn->s2s_cp_lock);
+	cps_t = idr_find(&nn->s2s_cp_stateids, st->si_opaque.so_id);
+	if (cps_t) {
+		state = container_of(cps_t, struct nfs4_cpntf_state,
+				     cp_stateid);
+		if (state->cp_stateid.sc_type != NFS4_COPYNOTIFY_STID) {
+			state = NULL;
+			goto unlock;
+		}
+		if (!clp)
+			refcount_inc(&state->cp_stateid.sc_count);
+		else
+			_free_cpntf_state_locked(nn, state);
+	}
+unlock:
+	spin_unlock(&nn->s2s_cp_lock);
+	if (!state)
+		return nfserr_bad_stateid;
+	if (!clp && state)
+		*cps = state;
+	return 0;
+}
+
+static __be32 find_cpntf_state(struct nfsd_net *nn, stateid_t *st,
+			       struct nfs4_stid **stid)
+{
+	__be32 status;
+	struct nfs4_cpntf_state *cps = NULL;
+	struct nfsd4_compound_state cstate;
+
+	status = manage_cpntf_state(nn, st, NULL, &cps);
+	if (status)
+		return status;
+
+	cps->cpntf_time = ktime_get_boottime_seconds();
+	memset(&cstate, 0, sizeof(cstate));
+	status = lookup_clientid(&cps->cp_p_clid, &cstate, nn, true);
+	if (status)
+		goto out;
+	status = nfsd4_lookup_stateid(&cstate, &cps->cp_p_stateid,
+				NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID,
+				stid, nn);
+	put_client_renew(cstate.clp);
+out:
+	nfs4_put_cpntf_state(nn, cps);
+	return status;
+}
+
+void nfs4_put_cpntf_state(struct nfsd_net *nn, struct nfs4_cpntf_state *cps)
+{
+	spin_lock(&nn->s2s_cp_lock);
+	_free_cpntf_state_locked(nn, cps);
+	spin_unlock(&nn->s2s_cp_lock);
+}
 
 /*
  * Checks for stateid operations
@@ -5602,7 +5751,8 @@ out:
 __be32
 nfs4_preprocess_stateid_op(struct svc_rqst *rqstp,
 		struct nfsd4_compound_state *cstate, struct svc_fh *fhp,
-		stateid_t *stateid, int flags, struct nfsd_file **nfp)
+		stateid_t *stateid, int flags, struct nfsd_file **nfp,
+		struct nfs4_stid **cstid)
 {
 	struct inode *ino = d_inode(fhp->fh_dentry);
 	struct net *net = SVC_NET(rqstp);
@@ -5624,6 +5774,8 @@ nfs4_preprocess_stateid_op(struct svc_rqst *rqstp,
 	status = nfsd4_lookup_stateid(cstate, stateid,
 				NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID,
 				&s, nn);
+	if (status == nfserr_bad_stateid)
+		status = find_cpntf_state(nn, stateid, &s);
 	if (status)
 		return status;
 	status = nfsd4_stid_check_stateid_generation(stateid, s,
@@ -5651,8 +5803,12 @@ done:
 	if (status == nfs_ok && nfp)
 		status = nfs4_check_file(rqstp, fhp, s, nfp, flags);
 out:
-	if (s)
-		nfs4_put_stid(s);
+	if (s) {
+		if (!status && cstid)
+			*cstid = s;
+		else
+			nfs4_put_stid(s);
+	}
 	return status;
 }
 
@@ -6242,21 +6398,21 @@ alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp,
 }
 
 static struct nfs4_ol_stateid *
-find_lock_stateid(struct nfs4_lockowner *lo, struct nfs4_file *fp)
+find_lock_stateid(const struct nfs4_lockowner *lo,
+		  const struct nfs4_ol_stateid *ost)
 {
 	struct nfs4_ol_stateid *lst;
-	struct nfs4_client *clp = lo->lo_owner.so_client;
 
-	lockdep_assert_held(&clp->cl_lock);
+	lockdep_assert_held(&ost->st_stid.sc_client->cl_lock);
 
-	list_for_each_entry(lst, &lo->lo_owner.so_stateids, st_perstateowner) {
-		if (lst->st_stid.sc_type != NFS4_LOCK_STID)
-			continue;
-		if (lst->st_stid.sc_file == fp) {
-			refcount_inc(&lst->st_stid.sc_count);
-			return lst;
+	/* If ost is not hashed, ost->st_locks will not be valid */
+	if (!nfs4_ol_stateid_unhashed(ost))
+		list_for_each_entry(lst, &ost->st_locks, st_locks) {
+			if (lst->st_stateowner == &lo->lo_owner) {
+				refcount_inc(&lst->st_stid.sc_count);
+				return lst;
+			}
 		}
-	}
 	return NULL;
 }
 
@@ -6272,11 +6428,11 @@ init_lock_stateid(struct nfs4_ol_stateid *stp, struct nfs4_lockowner *lo,
 	mutex_lock_nested(&stp->st_mutex, OPEN_STATEID_MUTEX);
 retry:
 	spin_lock(&clp->cl_lock);
-	spin_lock(&fp->fi_lock);
-	retstp = find_lock_stateid(lo, fp);
+	if (nfs4_ol_stateid_unhashed(open_stp))
+		goto out_close;
+	retstp = find_lock_stateid(lo, open_stp);
 	if (retstp)
-		goto out_unlock;
-
+		goto out_found;
 	refcount_inc(&stp->st_stid.sc_count);
 	stp->st_stid.sc_type = NFS4_LOCK_STID;
 	stp->st_stateowner = nfs4_get_stateowner(&lo->lo_owner);
@@ -6285,22 +6441,26 @@ retry:
 	stp->st_access_bmap = 0;
 	stp->st_deny_bmap = open_stp->st_deny_bmap;
 	stp->st_openstp = open_stp;
+	spin_lock(&fp->fi_lock);
 	list_add(&stp->st_locks, &open_stp->st_locks);
 	list_add(&stp->st_perstateowner, &lo->lo_owner.so_stateids);
 	list_add(&stp->st_perfile, &fp->fi_stateids);
-out_unlock:
 	spin_unlock(&fp->fi_lock);
 	spin_unlock(&clp->cl_lock);
-	if (retstp) {
-		if (nfsd4_lock_ol_stateid(retstp) != nfs_ok) {
-			nfs4_put_stid(&retstp->st_stid);
-			goto retry;
-		}
-		/* To keep mutex tracking happy */
-		mutex_unlock(&stp->st_mutex);
-		stp = retstp;
-	}
 	return stp;
+out_found:
+	spin_unlock(&clp->cl_lock);
+	if (nfsd4_lock_ol_stateid(retstp) != nfs_ok) {
+		nfs4_put_stid(&retstp->st_stid);
+		goto retry;
+	}
+	/* To keep mutex tracking happy */
+	mutex_unlock(&stp->st_mutex);
+	return retstp;
+out_close:
+	spin_unlock(&clp->cl_lock);
+	mutex_unlock(&stp->st_mutex);
+	return NULL;
 }
 
 static struct nfs4_ol_stateid *
@@ -6315,7 +6475,7 @@ find_or_create_lock_stateid(struct nfs4_lockowner *lo, struct nfs4_file *fi,
 
 	*new = false;
 	spin_lock(&clp->cl_lock);
-	lst = find_lock_stateid(lo, fi);
+	lst = find_lock_stateid(lo, ost);
 	spin_unlock(&clp->cl_lock);
 	if (lst != NULL) {
 		if (nfsd4_lock_ol_stateid(lst) == nfs_ok)
@@ -6545,7 +6705,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	}
 
 	if (fl_flags & FL_SLEEP) {
-		nbl->nbl_time = jiffies;
+		nbl->nbl_time = ktime_get_boottime_seconds();
 		spin_lock(&nn->blocked_locks_lock);
 		list_add_tail(&nbl->nbl_list, &lock_sop->lo_blocked);
 		list_add_tail(&nbl->nbl_lru, &nn->blocked_locks_lru);
@@ -6652,7 +6812,8 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		 return nfserr_inval;
 
 	if (!nfsd4_has_session(cstate)) {
-		status = lookup_clientid(&lockt->lt_clientid, cstate, nn);
+		status = lookup_clientid(&lockt->lt_clientid, cstate, nn,
+					 false);
 		if (status)
 			goto out;
 	}
@@ -6836,7 +6997,7 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp,
 	dprintk("nfsd4_release_lockowner clientid: (%08x/%08x):\n",
 		clid->cl_boot, clid->cl_id);
 
-	status = lookup_clientid(clid, cstate, nn);
+	status = lookup_clientid(clid, cstate, nn, false);
 	if (status)
 		return status;
 
@@ -6983,7 +7144,7 @@ nfs4_check_open_reclaim(clientid_t *clid,
 	__be32 status;
 
 	/* find clientid in conf_id_hashtbl */
-	status = lookup_clientid(clid, cstate, nn);
+	status = lookup_clientid(clid, cstate, nn, false);
 	if (status)
 		return nfserr_reclaim_bad;
 
@@ -7636,7 +7797,7 @@ static int nfs4_state_create_net(struct net *net)
 		INIT_LIST_HEAD(&nn->sessionid_hashtbl[i]);
 	nn->conf_name_tree = RB_ROOT;
 	nn->unconf_name_tree = RB_ROOT;
-	nn->boot_time = get_seconds();
+	nn->boot_time = ktime_get_real_seconds();
 	nn->grace_ended = false;
 	nn->nfsd4_manager.block_opens = true;
 	INIT_LIST_HEAD(&nn->nfsd4_manager.list);
@@ -7705,7 +7866,7 @@ nfs4_state_start_net(struct net *net)
 	nfsd4_client_tracking_init(net);
 	if (nn->track_reclaim_completes && nn->reclaim_str_hashtbl_size == 0)
 		goto skip_grace;
-	printk(KERN_INFO "NFSD: starting %ld-second grace period (net %x)\n",
+	printk(KERN_INFO "NFSD: starting %lld-second grace period (net %x)\n",
 	       nn->nfsd4_grace, net->ns.inum);
 	queue_delayed_work(laundry_wq, &nn->laundromat_work, nn->nfsd4_grace * HZ);
 	return 0;
@@ -7781,7 +7942,8 @@ nfs4_state_shutdown(void)
 static void
 get_stateid(struct nfsd4_compound_state *cstate, stateid_t *stateid)
 {
-	if (HAS_STATE_ID(cstate, CURRENT_STATE_ID_FLAG) && CURRENT_STATEID(stateid))
+	if (HAS_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG) &&
+	    CURRENT_STATEID(stateid))
 		memcpy(stateid, &cstate->current_stateid, sizeof(stateid_t));
 }
 
@@ -7790,14 +7952,14 @@ put_stateid(struct nfsd4_compound_state *cstate, stateid_t *stateid)
 {
 	if (cstate->minorversion) {
 		memcpy(&cstate->current_stateid, stateid, sizeof(stateid_t));
-		SET_STATE_ID(cstate, CURRENT_STATE_ID_FLAG);
+		SET_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG);
 	}
 }
 
 void
 clear_current_stateid(struct nfsd4_compound_state *cstate)
 {
-	CLEAR_STATE_ID(cstate, CURRENT_STATE_ID_FLAG);
+	CLEAR_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG);
 }
 
 /*
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 533d0fc3c96b..996ac01ee977 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -40,6 +40,7 @@
 #include <linux/utsname.h>
 #include <linux/pagemap.h>
 #include <linux/sunrpc/svcauth_gss.h>
+#include <linux/sunrpc/addr.h>
 
 #include "idmap.h"
 #include "acl.h"
@@ -1744,10 +1745,47 @@ nfsd4_decode_clone(struct nfsd4_compoundargs *argp, struct nfsd4_clone *clone)
 	DECODE_TAIL;
 }
 
+static __be32 nfsd4_decode_nl4_server(struct nfsd4_compoundargs *argp,
+				      struct nl4_server *ns)
+{
+	DECODE_HEAD;
+	struct nfs42_netaddr *naddr;
+
+	READ_BUF(4);
+	ns->nl4_type = be32_to_cpup(p++);
+
+	/* currently support for 1 inter-server source server */
+	switch (ns->nl4_type) {
+	case NL4_NETADDR:
+		naddr = &ns->u.nl4_addr;
+
+		READ_BUF(4);
+		naddr->netid_len = be32_to_cpup(p++);
+		if (naddr->netid_len > RPCBIND_MAXNETIDLEN)
+			goto xdr_error;
+
+		READ_BUF(naddr->netid_len + 4); /* 4 for uaddr len */
+		COPYMEM(naddr->netid, naddr->netid_len);
+
+		naddr->addr_len = be32_to_cpup(p++);
+		if (naddr->addr_len > RPCBIND_MAXUADDRLEN)
+			goto xdr_error;
+
+		READ_BUF(naddr->addr_len);
+		COPYMEM(naddr->addr, naddr->addr_len);
+		break;
+	default:
+		goto xdr_error;
+	}
+	DECODE_TAIL;
+}
+
 static __be32
 nfsd4_decode_copy(struct nfsd4_compoundargs *argp, struct nfsd4_copy *copy)
 {
 	DECODE_HEAD;
+	struct nl4_server *ns_dummy;
+	int i, count;
 
 	status = nfsd4_decode_stateid(argp, &copy->cp_src_stateid);
 	if (status)
@@ -1762,7 +1800,32 @@ nfsd4_decode_copy(struct nfsd4_compoundargs *argp, struct nfsd4_copy *copy)
 	p = xdr_decode_hyper(p, &copy->cp_count);
 	p++; /* ca_consecutive: we always do consecutive copies */
 	copy->cp_synchronous = be32_to_cpup(p++);
-	/* tmp = be32_to_cpup(p); Source server list not supported */
+
+	count = be32_to_cpup(p++);
+
+	copy->cp_intra = false;
+	if (count == 0) { /* intra-server copy */
+		copy->cp_intra = true;
+		goto intra;
+	}
+
+	/* decode all the supplied server addresses but use first */
+	status = nfsd4_decode_nl4_server(argp, &copy->cp_src);
+	if (status)
+		return status;
+
+	ns_dummy = kmalloc(sizeof(struct nl4_server), GFP_KERNEL);
+	if (ns_dummy == NULL)
+		return nfserrno(-ENOMEM);
+	for (i = 0; i < count - 1; i++) {
+		status = nfsd4_decode_nl4_server(argp, ns_dummy);
+		if (status) {
+			kfree(ns_dummy);
+			return status;
+		}
+	}
+	kfree(ns_dummy);
+intra:
 
 	DECODE_TAIL;
 }
@@ -1775,6 +1838,18 @@ nfsd4_decode_offload_status(struct nfsd4_compoundargs *argp,
 }
 
 static __be32
+nfsd4_decode_copy_notify(struct nfsd4_compoundargs *argp,
+			 struct nfsd4_copy_notify *cn)
+{
+	int status;
+
+	status = nfsd4_decode_stateid(argp, &cn->cpn_src_stateid);
+	if (status)
+		return status;
+	return nfsd4_decode_nl4_server(argp, &cn->cpn_dst);
+}
+
+static __be32
 nfsd4_decode_seek(struct nfsd4_compoundargs *argp, struct nfsd4_seek *seek)
 {
 	DECODE_HEAD;
@@ -1875,7 +1950,7 @@ static const nfsd4_dec nfsd4_dec_ops[] = {
 	/* new operations for NFSv4.2 */
 	[OP_ALLOCATE]		= (nfsd4_dec)nfsd4_decode_fallocate,
 	[OP_COPY]		= (nfsd4_dec)nfsd4_decode_copy,
-	[OP_COPY_NOTIFY]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_COPY_NOTIFY]	= (nfsd4_dec)nfsd4_decode_copy_notify,
 	[OP_DEALLOCATE]		= (nfsd4_dec)nfsd4_decode_fallocate,
 	[OP_IO_ADVISE]		= (nfsd4_dec)nfsd4_decode_notsupp,
 	[OP_LAYOUTERROR]	= (nfsd4_dec)nfsd4_decode_notsupp,
@@ -2024,11 +2099,11 @@ static __be32 *encode_change(__be32 *p, struct kstat *stat, struct inode *inode,
  */
 static __be32 *encode_time_delta(__be32 *p, struct inode *inode)
 {
-	struct timespec ts;
+	struct timespec64 ts;
 	u32 ns;
 
 	ns = max_t(u32, NSEC_PER_SEC/HZ, inode->i_sb->s_time_gran);
-	ts = ns_to_timespec(ns);
+	ts = ns_to_timespec64(ns);
 
 	p = xdr_encode_hyper(p, ts.tv_sec);
 	*p++ = cpu_to_be32(ts.tv_nsec);
@@ -2991,18 +3066,9 @@ nfsd4_encode_dirent_fattr(struct xdr_stream *xdr, struct nfsd4_readdir *cd,
 	__be32 nfserr;
 	int ignore_crossmnt = 0;
 
-	dentry = lookup_one_len_unlocked(name, cd->rd_fhp->fh_dentry, namlen);
+	dentry = lookup_positive_unlocked(name, cd->rd_fhp->fh_dentry, namlen);
 	if (IS_ERR(dentry))
 		return nfserrno(PTR_ERR(dentry));
-	if (d_really_is_negative(dentry)) {
-		/*
-		 * we're not holding the i_mutex here, so there's
-		 * a window where this directory entry could have gone
-		 * away.
-		 */
-		dput(dentry);
-		return nfserr_noent;
-	}
 
 	exp_get(exp);
 	/*
@@ -3461,7 +3527,6 @@ static __be32 nfsd4_encode_splice_read(
 	struct xdr_stream *xdr = &resp->xdr;
 	struct xdr_buf *buf = xdr->buf;
 	u32 eof;
-	long len;
 	int space_left;
 	__be32 nfserr;
 	__be32 *p = xdr->p - 2;
@@ -3470,7 +3535,6 @@ static __be32 nfsd4_encode_splice_read(
 	if (xdr->end - xdr->p < 1)
 		return nfserr_resource;
 
-	len = maxcount;
 	nfserr = nfsd_splice_read(read->rd_rqstp, read->rd_fhp,
 				  file, read->rd_offset, &maxcount, &eof);
 	read->rd_length = maxcount;
@@ -3527,23 +3591,22 @@ static __be32 nfsd4_encode_readv(struct nfsd4_compoundres *resp,
 	__be32 nfserr;
 	__be32 tmp;
 	__be32 *p;
-	u32 zzz = 0;
 	int pad;
 
+	/*
+	 * svcrdma requires every READ payload to start somewhere
+	 * in xdr->pages.
+	 */
+	if (xdr->iov == xdr->buf->head) {
+		xdr->iov = NULL;
+		xdr->end = xdr->p;
+	}
+
 	len = maxcount;
 	v = 0;
-
-	thislen = min_t(long, len, ((void *)xdr->end - (void *)xdr->p));
-	p = xdr_reserve_space(xdr, (thislen+3)&~3);
-	WARN_ON_ONCE(!p);
-	resp->rqstp->rq_vec[v].iov_base = p;
-	resp->rqstp->rq_vec[v].iov_len = thislen;
-	v++;
-	len -= thislen;
-
 	while (len) {
 		thislen = min_t(long, len, PAGE_SIZE);
-		p = xdr_reserve_space(xdr, (thislen+3)&~3);
+		p = xdr_reserve_space(xdr, thislen);
 		WARN_ON_ONCE(!p);
 		resp->rqstp->rq_vec[v].iov_base = p;
 		resp->rqstp->rq_vec[v].iov_len = thislen;
@@ -3552,23 +3615,25 @@ static __be32 nfsd4_encode_readv(struct nfsd4_compoundres *resp,
 	}
 	read->rd_vlen = v;
 
-	len = maxcount;
 	nfserr = nfsd_readv(resp->rqstp, read->rd_fhp, file, read->rd_offset,
 			    resp->rqstp->rq_vec, read->rd_vlen, &maxcount,
 			    &eof);
 	read->rd_length = maxcount;
 	if (nfserr)
 		return nfserr;
-	xdr_truncate_encode(xdr, starting_len + 8 + ((maxcount+3)&~3));
+	if (svc_encode_read_payload(resp->rqstp, starting_len + 8, maxcount))
+		return nfserr_io;
+	xdr_truncate_encode(xdr, starting_len + 8 + xdr_align_size(maxcount));
 
 	tmp = htonl(eof);
 	write_bytes_to_xdr_buf(xdr->buf, starting_len    , &tmp, 4);
 	tmp = htonl(maxcount);
 	write_bytes_to_xdr_buf(xdr->buf, starting_len + 4, &tmp, 4);
 
+	tmp = xdr_zero;
 	pad = (maxcount&3) ? 4 - (maxcount&3) : 0;
 	write_bytes_to_xdr_buf(xdr->buf, starting_len + 8 + maxcount,
-								&zzz, pad);
+								&tmp, pad);
 	return 0;
 
 }
@@ -3941,11 +4006,12 @@ nfsd4_encode_exchange_id(struct nfsd4_compoundres *resp, __be32 nfserr,
 	int major_id_sz;
 	int server_scope_sz;
 	uint64_t minor_id = 0;
+	struct nfsd_net *nn = net_generic(SVC_NET(resp->rqstp), nfsd_net_id);
 
-	major_id = utsname()->nodename;
-	major_id_sz = strlen(major_id);
-	server_scope = utsname()->nodename;
-	server_scope_sz = strlen(server_scope);
+	major_id = nn->nfsd_name;
+	major_id_sz = strlen(nn->nfsd_name);
+	server_scope = nn->nfsd_name;
+	server_scope_sz = strlen(nn->nfsd_name);
 
 	p = xdr_reserve_space(xdr,
 		8 /* eir_clientid */ +
@@ -4255,6 +4321,46 @@ nfsd42_encode_write_res(struct nfsd4_compoundres *resp,
 }
 
 static __be32
+nfsd42_encode_nl4_server(struct nfsd4_compoundres *resp, struct nl4_server *ns)
+{
+	struct xdr_stream *xdr = &resp->xdr;
+	struct nfs42_netaddr *addr;
+	__be32 *p;
+
+	p = xdr_reserve_space(xdr, 4);
+	*p++ = cpu_to_be32(ns->nl4_type);
+
+	switch (ns->nl4_type) {
+	case NL4_NETADDR:
+		addr = &ns->u.nl4_addr;
+
+		/* netid_len, netid, uaddr_len, uaddr (port included
+		 * in RPCBIND_MAXUADDRLEN)
+		 */
+		p = xdr_reserve_space(xdr,
+			4 /* netid len */ +
+			(XDR_QUADLEN(addr->netid_len) * 4) +
+			4 /* uaddr len */ +
+			(XDR_QUADLEN(addr->addr_len) * 4));
+		if (!p)
+			return nfserr_resource;
+
+		*p++ = cpu_to_be32(addr->netid_len);
+		p = xdr_encode_opaque_fixed(p, addr->netid,
+					    addr->netid_len);
+		*p++ = cpu_to_be32(addr->addr_len);
+		p = xdr_encode_opaque_fixed(p, addr->addr,
+					addr->addr_len);
+		break;
+	default:
+		WARN_ON_ONCE(ns->nl4_type != NL4_NETADDR);
+		return nfserr_inval;
+	}
+
+	return 0;
+}
+
+static __be32
 nfsd4_encode_copy(struct nfsd4_compoundres *resp, __be32 nfserr,
 		  struct nfsd4_copy *copy)
 {
@@ -4288,6 +4394,40 @@ nfsd4_encode_offload_status(struct nfsd4_compoundres *resp, __be32 nfserr,
 }
 
 static __be32
+nfsd4_encode_copy_notify(struct nfsd4_compoundres *resp, __be32 nfserr,
+			 struct nfsd4_copy_notify *cn)
+{
+	struct xdr_stream *xdr = &resp->xdr;
+	__be32 *p;
+
+	if (nfserr)
+		return nfserr;
+
+	/* 8 sec, 4 nsec */
+	p = xdr_reserve_space(xdr, 12);
+	if (!p)
+		return nfserr_resource;
+
+	/* cnr_lease_time */
+	p = xdr_encode_hyper(p, cn->cpn_sec);
+	*p++ = cpu_to_be32(cn->cpn_nsec);
+
+	/* cnr_stateid */
+	nfserr = nfsd4_encode_stateid(xdr, &cn->cpn_cnr_stateid);
+	if (nfserr)
+		return nfserr;
+
+	/* cnr_src.nl_nsvr */
+	p = xdr_reserve_space(xdr, 4);
+	if (!p)
+		return nfserr_resource;
+
+	*p++ = cpu_to_be32(1);
+
+	return nfsd42_encode_nl4_server(resp, &cn->cpn_src);
+}
+
+static __be32
 nfsd4_encode_seek(struct nfsd4_compoundres *resp, __be32 nfserr,
 		  struct nfsd4_seek *seek)
 {
@@ -4384,7 +4524,7 @@ static const nfsd4_enc nfsd4_enc_ops[] = {
 	/* NFSv4.2 operations */
 	[OP_ALLOCATE]		= (nfsd4_enc)nfsd4_encode_noop,
 	[OP_COPY]		= (nfsd4_enc)nfsd4_encode_copy,
-	[OP_COPY_NOTIFY]	= (nfsd4_enc)nfsd4_encode_noop,
+	[OP_COPY_NOTIFY]	= (nfsd4_enc)nfsd4_encode_copy_notify,
 	[OP_DEALLOCATE]		= (nfsd4_enc)nfsd4_encode_noop,
 	[OP_IO_ADVISE]		= (nfsd4_enc)nfsd4_encode_noop,
 	[OP_LAYOUTERROR]	= (nfsd4_enc)nfsd4_encode_noop,
@@ -4511,8 +4651,6 @@ nfsd4_encode_replay(struct xdr_stream *xdr, struct nfsd4_op *op)
 	__be32 *p;
 	struct nfs4_replay *rp = op->replay;
 
-	BUG_ON(!rp);
-
 	p = xdr_reserve_space(xdr, 8 + rp->rp_buflen);
 	if (!p) {
 		WARN_ON_ONCE(1);
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 11b42c523f04..3bb2db947d29 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -157,11 +157,11 @@ static int exports_proc_open(struct inode *inode, struct file *file)
 	return exports_net_open(current->nsproxy->net_ns, file);
 }
 
-static const struct file_operations exports_proc_operations = {
-	.open		= exports_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= seq_release,
+static const struct proc_ops exports_proc_ops = {
+	.proc_open	= exports_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= seq_release,
 };
 
 static int exports_nfsd_open(struct inode *inode, struct file *file)
@@ -956,7 +956,7 @@ static ssize_t write_maxconn(struct file *file, char *buf, size_t size)
 
 #ifdef CONFIG_NFSD_V4
 static ssize_t __nfsd4_write_time(struct file *file, char *buf, size_t size,
-				  time_t *time, struct nfsd_net *nn)
+				  time64_t *time, struct nfsd_net *nn)
 {
 	char *mesg = buf;
 	int rv, i;
@@ -984,11 +984,11 @@ static ssize_t __nfsd4_write_time(struct file *file, char *buf, size_t size,
 		*time = i;
 	}
 
-	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%ld\n", *time);
+	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%lld\n", *time);
 }
 
 static ssize_t nfsd4_write_time(struct file *file, char *buf, size_t size,
-				time_t *time, struct nfsd_net *nn)
+				time64_t *time, struct nfsd_net *nn)
 {
 	ssize_t rv;
 
@@ -1333,6 +1333,7 @@ void nfsd_client_rmdir(struct dentry *dentry)
 	dget(dentry);
 	ret = simple_rmdir(dir, dentry);
 	WARN_ON_ONCE(ret);
+	fsnotify_rmdir(dir, dentry);
 	d_delete(dentry);
 	inode_unlock(dir);
 }
@@ -1431,8 +1432,7 @@ static int create_proc_exports_entry(void)
 	entry = proc_mkdir("fs/nfs", NULL);
 	if (!entry)
 		return -ENOMEM;
-	entry = proc_create("exports", 0, entry,
-				 &exports_proc_operations);
+	entry = proc_create("exports", 0, entry, &exports_proc_ops);
 	if (!entry) {
 		remove_proc_entry("fs/nfs", NULL);
 		return -ENOMEM;
diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h
index af2947551e9c..2ab5569126b8 100644
--- a/fs/nfsd/nfsd.h
+++ b/fs/nfsd/nfsd.h
@@ -19,6 +19,7 @@
 #include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/svc_xprt.h>
 #include <linux/sunrpc/msg_prot.h>
+#include <linux/sunrpc/addr.h>
 
 #include <uapi/linux/nfsd/debug.h>
 
@@ -142,7 +143,6 @@ int nfs4_state_start(void);
 int nfs4_state_start_net(struct net *net);
 void nfs4_state_shutdown(void);
 void nfs4_state_shutdown_net(struct net *net);
-void nfs4_reset_lease(time_t leasetime);
 int nfs4_reset_recoverydir(char *recdir);
 char * nfs4_recoverydir(void);
 bool nfsd4_spo_must_allow(struct svc_rqst *rqstp);
@@ -153,7 +153,6 @@ static inline int nfs4_state_start(void) { return 0; }
 static inline int nfs4_state_start_net(struct net *net) { return 0; }
 static inline void nfs4_state_shutdown(void) { }
 static inline void nfs4_state_shutdown_net(struct net *net) { }
-static inline void nfs4_reset_lease(time_t leasetime) { }
 static inline int nfs4_reset_recoverydir(char *recdir) { return 0; }
 static inline char * nfs4_recoverydir(void) {return NULL; }
 static inline bool nfsd4_spo_must_allow(struct svc_rqst *rqstp)
@@ -280,7 +279,8 @@ void		nfsd_lockd_shutdown(void);
 #define nfserr_union_notsupp		cpu_to_be32(NFS4ERR_UNION_NOTSUPP)
 #define nfserr_offload_denied		cpu_to_be32(NFS4ERR_OFFLOAD_DENIED)
 #define nfserr_wrong_lfs		cpu_to_be32(NFS4ERR_WRONG_LFS)
-#define nfserr_badlabel		cpu_to_be32(NFS4ERR_BADLABEL)
+#define nfserr_badlabel			cpu_to_be32(NFS4ERR_BADLABEL)
+#define nfserr_file_open		cpu_to_be32(NFS4ERR_FILE_OPEN)
 
 /* error codes for internal use */
 /* if a request fails due to kmalloc failure, it gets dropped.
@@ -386,6 +386,37 @@ void		nfsd_lockd_shutdown(void);
 
 extern const u32 nfsd_suppattrs[3][3];
 
+static inline __be32 nfsd4_set_netaddr(struct sockaddr *addr,
+				    struct nfs42_netaddr *netaddr)
+{
+	struct sockaddr_in *sin = (struct sockaddr_in *)addr;
+	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)addr;
+	unsigned int port;
+	size_t ret_addr, ret_port;
+
+	switch (addr->sa_family) {
+	case AF_INET:
+		port = ntohs(sin->sin_port);
+		sprintf(netaddr->netid, "tcp");
+		netaddr->netid_len = 3;
+		break;
+	case AF_INET6:
+		port = ntohs(sin6->sin6_port);
+		sprintf(netaddr->netid, "tcp6");
+		netaddr->netid_len = 4;
+		break;
+	default:
+		return nfserr_inval;
+	}
+	ret_addr = rpc_ntop(addr, netaddr->addr, sizeof(netaddr->addr));
+	ret_port = snprintf(netaddr->addr + ret_addr,
+			    RPCBIND_MAXUADDRLEN + 1 - ret_addr,
+			    ".%u.%u", port >> 8, port & 0xff);
+	WARN_ON(ret_port >= RPCBIND_MAXUADDRLEN + 1 - ret_addr);
+	netaddr->addr_len = ret_addr + ret_port;
+	return 0;
+}
+
 static inline bool bmval_is_subset(const u32 *bm1, const u32 *bm2)
 {
 	return !((bm1[0] & ~bm2[0]) ||
diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c
index b319080288c3..37bc8f5f4514 100644
--- a/fs/nfsd/nfsfh.c
+++ b/fs/nfsd/nfsfh.c
@@ -14,6 +14,7 @@
 #include "nfsd.h"
 #include "vfs.h"
 #include "auth.h"
+#include "trace.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_FH
 
@@ -209,11 +210,14 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp)
 	}
 
 	error = nfserr_stale;
-	if (PTR_ERR(exp) == -ENOENT)
-		return error;
+	if (IS_ERR(exp)) {
+		trace_nfsd_set_fh_dentry_badexport(rqstp, fhp, PTR_ERR(exp));
+
+		if (PTR_ERR(exp) == -ENOENT)
+			return error;
 
-	if (IS_ERR(exp))
 		return nfserrno(PTR_ERR(exp));
+	}
 
 	if (exp->ex_flags & NFSEXP_NOSUBTREECHECK) {
 		/* Elevate privileges so that the lack of 'r' or 'x'
@@ -267,6 +271,9 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp)
 		dentry = exportfs_decode_fh(exp->ex_path.mnt, fid,
 				data_left, fileid_type,
 				nfsd_acceptable, exp);
+		if (IS_ERR_OR_NULL(dentry))
+			trace_nfsd_set_fh_dentry_badhandle(rqstp, fhp,
+					dentry ?  PTR_ERR(dentry) : -ESTALE);
 	}
 	if (dentry == NULL)
 		goto out;
diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h
index 755e256a9103..56cfbc361561 100644
--- a/fs/nfsd/nfsfh.h
+++ b/fs/nfsd/nfsfh.h
@@ -35,15 +35,15 @@ typedef struct svc_fh {
 
 	bool			fh_locked;	/* inode locked by us */
 	bool			fh_want_write;	/* remount protection taken */
-
+	int			fh_flags;	/* FH flags */
 #ifdef CONFIG_NFSD_V3
 	bool			fh_post_saved;	/* post-op attrs saved */
 	bool			fh_pre_saved;	/* pre-op attrs saved */
 
 	/* Pre-op attributes saved during fh_lock */
 	__u64			fh_pre_size;	/* size before operation */
-	struct timespec		fh_pre_mtime;	/* mtime before oper */
-	struct timespec		fh_pre_ctime;	/* ctime before oper */
+	struct timespec64	fh_pre_mtime;	/* mtime before oper */
+	struct timespec64	fh_pre_ctime;	/* ctime before oper */
 	/*
 	 * pre-op nfsv4 change attr: note must check IS_I_VERSION(inode)
 	 *  to find out if it is valid.
@@ -56,6 +56,9 @@ typedef struct svc_fh {
 #endif /* CONFIG_NFSD_V3 */
 
 } svc_fh;
+#define NFSD4_FH_FOREIGN (1<<0)
+#define SET_FH_FLAG(c, f) ((c)->fh_flags |= (f))
+#define HAS_FH_FLAG(c, f) ((c)->fh_flags & (f))
 
 enum nfsd_fsid {
 	FSID_DEV = 0,
diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c
index c83ddac22f38..543bbe0a556e 100644
--- a/fs/nfsd/nfsproc.c
+++ b/fs/nfsd/nfsproc.c
@@ -94,7 +94,7 @@ nfsd_proc_setattr(struct svc_rqst *rqstp)
 		 * Solaris, at least, doesn't seem to care what the time
 		 * request is.  We require it be within 30 minutes of now.
 		 */
-		time_t delta = iap->ia_atime.tv_sec - get_seconds();
+		time64_t delta = iap->ia_atime.tv_sec - ktime_get_real_seconds();
 
 		nfserr = fh_verify(rqstp, fhp, 0, NFSD_MAY_NOP);
 		if (nfserr)
@@ -113,7 +113,7 @@ nfsd_proc_setattr(struct svc_rqst *rqstp)
 		}
 	}
 
-	nfserr = nfsd_setattr(rqstp, fhp, iap, 0, (time_t)0);
+	nfserr = nfsd_setattr(rqstp, fhp, iap, 0, (time64_t)0);
 done:
 	return nfsd_return_attrs(nfserr, resp);
 }
@@ -226,7 +226,7 @@ nfsd_proc_write(struct svc_rqst *rqstp)
 		return nfserr_io;
 	nfserr = nfsd_write(rqstp, fh_copy(&resp->fh, &argp->fh),
 			    argp->offset, rqstp->rq_vec, nvecs,
-			    &cnt, NFS_DATA_SYNC);
+			    &cnt, NFS_DATA_SYNC, NULL);
 	return nfsd_return_attrs(nfserr, resp);
 }
 
@@ -380,7 +380,7 @@ nfsd_proc_create(struct svc_rqst *rqstp)
 		 */
 		attr->ia_valid &= ATTR_SIZE;
 		if (attr->ia_valid)
-			nfserr = nfsd_setattr(rqstp, newfhp, attr, 0, (time_t)0);
+			nfserr = nfsd_setattr(rqstp, newfhp, attr, 0, (time64_t)0);
 	}
 
 out_unlock:
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index fdf7ed4bd5dd..ca9fd348548b 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -31,6 +31,12 @@
 
 #define NFSDDBG_FACILITY	NFSDDBG_SVC
 
+bool inter_copy_offload_enable;
+EXPORT_SYMBOL_GPL(inter_copy_offload_enable);
+module_param(inter_copy_offload_enable, bool, 0644);
+MODULE_PARM_DESC(inter_copy_offload_enable,
+		 "Enable inter server to server copy offload. Default: false");
+
 extern struct svc_program	nfsd_program;
 static int			nfsd(void *vrqstp);
 #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
@@ -95,12 +101,11 @@ static const struct svc_version *nfsd_acl_version[] = {
 
 #define NFSD_ACL_MINVERS            2
 #define NFSD_ACL_NRVERS		ARRAY_SIZE(nfsd_acl_version)
-static const struct svc_version *nfsd_acl_versions[NFSD_ACL_NRVERS];
 
 static struct svc_program	nfsd_acl_program = {
 	.pg_prog		= NFS_ACL_PROGRAM,
 	.pg_nvers		= NFSD_ACL_NRVERS,
-	.pg_vers		= nfsd_acl_versions,
+	.pg_vers		= nfsd_acl_version,
 	.pg_name		= "nfsacl",
 	.pg_class		= "nfsd",
 	.pg_stats		= &nfsd_acl_svcstats,
@@ -392,20 +397,25 @@ static int nfsd_startup_net(int nrservs, struct net *net, const struct cred *cre
 		ret = lockd_up(net, cred);
 		if (ret)
 			goto out_socks;
-		nn->lockd_up = 1;
+		nn->lockd_up = true;
 	}
 
-	ret = nfs4_state_start_net(net);
+	ret = nfsd_file_cache_start_net(net);
 	if (ret)
 		goto out_lockd;
+	ret = nfs4_state_start_net(net);
+	if (ret)
+		goto out_filecache;
 
 	nn->nfsd_net_up = true;
 	return 0;
 
+out_filecache:
+	nfsd_file_cache_shutdown_net(net);
 out_lockd:
 	if (nn->lockd_up) {
 		lockd_down(net);
-		nn->lockd_up = 0;
+		nn->lockd_up = false;
 	}
 out_socks:
 	nfsd_shutdown_generic();
@@ -416,11 +426,11 @@ static void nfsd_shutdown_net(struct net *net)
 {
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
-	nfsd_file_cache_purge(net);
+	nfsd_file_cache_shutdown_net(net);
 	nfs4_state_shutdown_net(net);
 	if (nn->lockd_up) {
 		lockd_down(net);
-		nn->lockd_up = 0;
+		nn->lockd_up = false;
 	}
 	nn->nfsd_net_up = false;
 	nfsd_shutdown_generic();
@@ -739,6 +749,9 @@ nfsd_svc(int nrservs, struct net *net, const struct cred *cred)
 	if (nrservs == 0 && nn->nfsd_serv == NULL)
 		goto out;
 
+	strlcpy(nn->nfsd_name, utsname()->nodename,
+		sizeof(nn->nfsd_name));
+
 	error = nfsd_create_serv(net);
 	if (error)
 		goto out;
diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h
index 46f56afb6cb8..68d3f30ee760 100644
--- a/fs/nfsd/state.h
+++ b/fs/nfsd/state.h
@@ -56,6 +56,14 @@ typedef struct {
 	stateid_opaque_t        si_opaque;
 } stateid_t;
 
+typedef struct {
+	stateid_t		stid;
+#define NFS4_COPY_STID 1
+#define NFS4_COPYNOTIFY_STID 2
+	unsigned char		sc_type;
+	refcount_t		sc_count;
+} copy_stateid_t;
+
 #define STATEID_FMT	"(%08x/%08x/%08x/%08x)"
 #define STATEID_VAL(s) \
 	(s)->si_opaque.so_clid.cl_boot, \
@@ -96,6 +104,7 @@ struct nfs4_stid {
 #define NFS4_REVOKED_DELEG_STID 16
 #define NFS4_CLOSED_DELEG_STID 32
 #define NFS4_LAYOUT_STID 64
+	struct list_head	sc_cp_list;
 	unsigned char		sc_type;
 	stateid_t		sc_stateid;
 	spinlock_t		sc_lock;
@@ -104,6 +113,17 @@ struct nfs4_stid {
 	void			(*sc_free)(struct nfs4_stid *);
 };
 
+/* Keep a list of stateids issued by the COPY_NOTIFY, associate it with the
+ * parent OPEN/LOCK/DELEG stateid.
+ */
+struct nfs4_cpntf_state {
+	copy_stateid_t		cp_stateid;
+	struct list_head	cp_list;	/* per parent nfs4_stid */
+	stateid_t		cp_p_stateid;	/* copy of parent's stateid */
+	clientid_t		cp_p_clid;	/* copy of parent's clid */
+	time64_t		cpntf_time;	/* last time stateid used */
+};
+
 /*
  * Represents a delegation stateid. The nfs4_client holds references to these
  * and they are put when it is being destroyed or when the delegation is
@@ -132,7 +152,7 @@ struct nfs4_delegation {
 	struct list_head	dl_recall_lru;  /* delegation recalled */
 	struct nfs4_clnt_odstate *dl_clnt_odstate;
 	u32			dl_type;
-	time_t			dl_time;
+	time64_t		dl_time;
 /* For recall: */
 	int			dl_retries;
 	struct nfsd4_callback	dl_recall;
@@ -310,7 +330,7 @@ struct nfs4_client {
 #endif
 	struct xdr_netobj	cl_name; 	/* id generated by client */
 	nfs4_verifier		cl_verifier; 	/* generated by client */
-	time_t                  cl_time;        /* time of last lease renewal */
+	time64_t		cl_time;	/* time of last lease renewal */
 	struct sockaddr_storage	cl_addr; 	/* client ipaddress */
 	bool			cl_mach_cred;	/* SP4_MACH_CRED in force */
 	struct svc_cred		cl_cred; 	/* setclientid principal */
@@ -320,7 +340,7 @@ struct nfs4_client {
 	/* NFSv4.1 client implementation id: */
 	struct xdr_netobj	cl_nii_domain;
 	struct xdr_netobj	cl_nii_name;
-	struct timespec		cl_nii_time;
+	struct timespec64	cl_nii_time;
 
 	/* for v4.0 and v4.1 callbacks: */
 	struct nfs4_cb_conn	cl_cb_conn;
@@ -367,6 +387,7 @@ struct nfs4_client {
 	struct net		*net;
 	struct list_head	async_copies;	/* list of async copies */
 	spinlock_t		async_lock;	/* lock for async copies */
+	atomic_t		cl_cb_inflight;	/* Outstanding callbacks */
 };
 
 /* struct nfs4_client_reset
@@ -448,7 +469,7 @@ struct nfs4_openowner {
 	 */
 	struct list_head	oo_close_lru;
 	struct nfs4_ol_stateid *oo_last_closed_stid;
-	time_t			oo_time; /* time of placement on so_close_lru */
+	time64_t		oo_time; /* time of placement on so_close_lru */
 #define NFS4_OO_CONFIRMED   1
 	unsigned char		oo_flags;
 };
@@ -605,7 +626,7 @@ static inline bool nfsd4_stateid_generation_after(stateid_t *a, stateid_t *b)
 struct nfsd4_blocked_lock {
 	struct list_head	nbl_list;
 	struct list_head	nbl_lru;
-	unsigned long		nbl_time;
+	time64_t		nbl_time;
 	struct file_lock	nbl_lock;
 	struct knfsd_fh		nbl_fh;
 	struct nfsd4_callback	nbl_cb;
@@ -617,14 +638,17 @@ struct nfsd4_copy;
 
 extern __be32 nfs4_preprocess_stateid_op(struct svc_rqst *rqstp,
 		struct nfsd4_compound_state *cstate, struct svc_fh *fhp,
-		stateid_t *stateid, int flags, struct nfsd_file **filp);
+		stateid_t *stateid, int flags, struct nfsd_file **filp,
+		struct nfs4_stid **cstid);
 __be32 nfsd4_lookup_stateid(struct nfsd4_compound_state *cstate,
 		     stateid_t *stateid, unsigned char typemask,
 		     struct nfs4_stid **s, struct nfsd_net *nn);
 struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct kmem_cache *slab,
 				  void (*sc_free)(struct nfs4_stid *));
-int nfs4_init_cp_state(struct nfsd_net *nn, struct nfsd4_copy *copy);
-void nfs4_free_cp_state(struct nfsd4_copy *copy);
+int nfs4_init_copy_state(struct nfsd_net *nn, struct nfsd4_copy *copy);
+void nfs4_free_copy_state(struct nfsd4_copy *copy);
+struct nfs4_cpntf_state *nfs4_alloc_init_cpntf_state(struct nfsd_net *nn,
+			struct nfs4_stid *p_stid);
 void nfs4_unhash_stid(struct nfs4_stid *s);
 void nfs4_put_stid(struct nfs4_stid *s);
 void nfs4_inc_and_copy_stateid(stateid_t *dst, struct nfs4_stid *stid);
@@ -654,6 +678,11 @@ void put_nfs4_file(struct nfs4_file *fi);
 extern void nfs4_put_copy(struct nfsd4_copy *copy);
 extern struct nfsd4_copy *
 find_async_copy(struct nfs4_client *clp, stateid_t *staetid);
+extern void nfs4_put_cpntf_state(struct nfsd_net *nn,
+				 struct nfs4_cpntf_state *cps);
+extern __be32 manage_cpntf_state(struct nfsd_net *nn, stateid_t *st,
+				 struct nfs4_client *clp,
+				 struct nfs4_cpntf_state **cps);
 static inline void get_nfs4_file(struct nfs4_file *fi)
 {
 	refcount_inc(&fi->fi_ref);
diff --git a/fs/nfsd/stats.c b/fs/nfsd/stats.c
index 9bce3b913189..b1bc582b0493 100644
--- a/fs/nfsd/stats.c
+++ b/fs/nfsd/stats.c
@@ -84,17 +84,17 @@ static int nfsd_proc_open(struct inode *inode, struct file *file)
 	return single_open(file, nfsd_proc_show, NULL);
 }
 
-static const struct file_operations nfsd_proc_fops = {
-	.open = nfsd_proc_open,
-	.read  = seq_read,
-	.llseek = seq_lseek,
-	.release = single_release,
+static const struct proc_ops nfsd_proc_ops = {
+	.proc_open	= nfsd_proc_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
 };
 
 void
 nfsd_stat_init(void)
 {
-	svc_proc_register(&init_net, &nfsd_svcstats, &nfsd_proc_fops);
+	svc_proc_register(&init_net, &nfsd_svcstats, &nfsd_proc_ops);
 }
 
 void
diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h
index ffc78a0e28b2..78c574251c60 100644
--- a/fs/nfsd/trace.h
+++ b/fs/nfsd/trace.h
@@ -9,6 +9,7 @@
 #define _NFSD_TRACE_H
 
 #include <linux/tracepoint.h>
+#include "export.h"
 #include "nfsfh.h"
 
 TRACE_EVENT(nfsd_compound,
@@ -50,6 +51,127 @@ TRACE_EVENT(nfsd_compound_status,
 		__get_str(name), __entry->status)
 )
 
+DECLARE_EVENT_CLASS(nfsd_fh_err_class,
+	TP_PROTO(struct svc_rqst *rqstp,
+		 struct svc_fh	*fhp,
+		 int		status),
+	TP_ARGS(rqstp, fhp, status),
+	TP_STRUCT__entry(
+		__field(u32, xid)
+		__field(u32, fh_hash)
+		__field(int, status)
+	),
+	TP_fast_assign(
+		__entry->xid = be32_to_cpu(rqstp->rq_xid);
+		__entry->fh_hash = knfsd_fh_hash(&fhp->fh_handle);
+		__entry->status = status;
+	),
+	TP_printk("xid=0x%08x fh_hash=0x%08x status=%d",
+		  __entry->xid, __entry->fh_hash,
+		  __entry->status)
+)
+
+#define DEFINE_NFSD_FH_ERR_EVENT(name)		\
+DEFINE_EVENT(nfsd_fh_err_class, nfsd_##name,	\
+	TP_PROTO(struct svc_rqst *rqstp,	\
+		 struct svc_fh	*fhp,		\
+		 int		status),	\
+	TP_ARGS(rqstp, fhp, status))
+
+DEFINE_NFSD_FH_ERR_EVENT(set_fh_dentry_badexport);
+DEFINE_NFSD_FH_ERR_EVENT(set_fh_dentry_badhandle);
+
+TRACE_EVENT(nfsd_exp_find_key,
+	TP_PROTO(const struct svc_expkey *key,
+		 int status),
+	TP_ARGS(key, status),
+	TP_STRUCT__entry(
+		__field(int, fsidtype)
+		__array(u32, fsid, 6)
+		__string(auth_domain, key->ek_client->name)
+		__field(int, status)
+	),
+	TP_fast_assign(
+		__entry->fsidtype = key->ek_fsidtype;
+		memcpy(__entry->fsid, key->ek_fsid, 4*6);
+		__assign_str(auth_domain, key->ek_client->name);
+		__entry->status = status;
+	),
+	TP_printk("fsid=%x::%s domain=%s status=%d",
+		__entry->fsidtype,
+		__print_array(__entry->fsid, 6, 4),
+		__get_str(auth_domain),
+		__entry->status
+	)
+);
+
+TRACE_EVENT(nfsd_expkey_update,
+	TP_PROTO(const struct svc_expkey *key, const char *exp_path),
+	TP_ARGS(key, exp_path),
+	TP_STRUCT__entry(
+		__field(int, fsidtype)
+		__array(u32, fsid, 6)
+		__string(auth_domain, key->ek_client->name)
+		__string(path, exp_path)
+		__field(bool, cache)
+	),
+	TP_fast_assign(
+		__entry->fsidtype = key->ek_fsidtype;
+		memcpy(__entry->fsid, key->ek_fsid, 4*6);
+		__assign_str(auth_domain, key->ek_client->name);
+		__assign_str(path, exp_path);
+		__entry->cache = !test_bit(CACHE_NEGATIVE, &key->h.flags);
+	),
+	TP_printk("fsid=%x::%s domain=%s path=%s cache=%s",
+		__entry->fsidtype,
+		__print_array(__entry->fsid, 6, 4),
+		__get_str(auth_domain),
+		__get_str(path),
+		__entry->cache ? "pos" : "neg"
+	)
+);
+
+TRACE_EVENT(nfsd_exp_get_by_name,
+	TP_PROTO(const struct svc_export *key,
+		 int status),
+	TP_ARGS(key, status),
+	TP_STRUCT__entry(
+		__string(path, key->ex_path.dentry->d_name.name)
+		__string(auth_domain, key->ex_client->name)
+		__field(int, status)
+	),
+	TP_fast_assign(
+		__assign_str(path, key->ex_path.dentry->d_name.name);
+		__assign_str(auth_domain, key->ex_client->name);
+		__entry->status = status;
+	),
+	TP_printk("path=%s domain=%s status=%d",
+		__get_str(path),
+		__get_str(auth_domain),
+		__entry->status
+	)
+);
+
+TRACE_EVENT(nfsd_export_update,
+	TP_PROTO(const struct svc_export *key),
+	TP_ARGS(key),
+	TP_STRUCT__entry(
+		__string(path, key->ex_path.dentry->d_name.name)
+		__string(auth_domain, key->ex_client->name)
+		__field(bool, cache)
+	),
+	TP_fast_assign(
+		__assign_str(path, key->ex_path.dentry->d_name.name);
+		__assign_str(auth_domain, key->ex_client->name);
+		__entry->cache = !test_bit(CACHE_NEGATIVE, &key->h.flags);
+	),
+	TP_printk("path=%s domain=%s cache=%s",
+		__get_str(path),
+		__get_str(auth_domain),
+		__entry->cache ? "pos" : "neg"
+	)
+);
+
 DECLARE_EVENT_CLASS(nfsd_io_class,
 	TP_PROTO(struct svc_rqst *rqstp,
 		 struct svc_fh	*fhp,
@@ -166,6 +288,12 @@ DEFINE_STATEID_EVENT(layout_recall_done);
 DEFINE_STATEID_EVENT(layout_recall_fail);
 DEFINE_STATEID_EVENT(layout_recall_release);
 
+TRACE_DEFINE_ENUM(NFSD_FILE_HASHED);
+TRACE_DEFINE_ENUM(NFSD_FILE_PENDING);
+TRACE_DEFINE_ENUM(NFSD_FILE_BREAK_READ);
+TRACE_DEFINE_ENUM(NFSD_FILE_BREAK_WRITE);
+TRACE_DEFINE_ENUM(NFSD_FILE_REFERENCED);
+
 #define show_nf_flags(val)						\
 	__print_flags(val, "|",						\
 		{ 1 << NFSD_FILE_HASHED,	"HASHED" },		\
@@ -195,7 +323,7 @@ DECLARE_EVENT_CLASS(nfsd_file_class,
 	TP_fast_assign(
 		__entry->nf_hashval = nf->nf_hashval;
 		__entry->nf_inode = nf->nf_inode;
-		__entry->nf_ref = atomic_read(&nf->nf_ref);
+		__entry->nf_ref = refcount_read(&nf->nf_ref);
 		__entry->nf_flags = nf->nf_flags;
 		__entry->nf_may = nf->nf_may;
 		__entry->nf_file = nf->nf_file;
@@ -228,7 +356,7 @@ TRACE_EVENT(nfsd_file_acquire,
 	TP_ARGS(rqstp, hash, inode, may_flags, nf, status),
 
 	TP_STRUCT__entry(
-		__field(__be32, xid)
+		__field(u32, xid)
 		__field(unsigned int, hash)
 		__field(void *, inode)
 		__field(unsigned int, may_flags)
@@ -236,27 +364,27 @@ TRACE_EVENT(nfsd_file_acquire,
 		__field(unsigned long, nf_flags)
 		__field(unsigned char, nf_may)
 		__field(struct file *, nf_file)
-		__field(__be32, status)
+		__field(u32, status)
 	),
 
 	TP_fast_assign(
-		__entry->xid = rqstp->rq_xid;
+		__entry->xid = be32_to_cpu(rqstp->rq_xid);
 		__entry->hash = hash;
 		__entry->inode = inode;
 		__entry->may_flags = may_flags;
-		__entry->nf_ref = nf ? atomic_read(&nf->nf_ref) : 0;
+		__entry->nf_ref = nf ? refcount_read(&nf->nf_ref) : 0;
 		__entry->nf_flags = nf ? nf->nf_flags : 0;
 		__entry->nf_may = nf ? nf->nf_may : 0;
 		__entry->nf_file = nf ? nf->nf_file : NULL;
-		__entry->status = status;
+		__entry->status = be32_to_cpu(status);
 	),
 
 	TP_printk("xid=0x%x hash=0x%x inode=0x%p may_flags=%s ref=%d nf_flags=%s nf_may=%s nf_file=0x%p status=%u",
-			be32_to_cpu(__entry->xid), __entry->hash, __entry->inode,
+			__entry->xid, __entry->hash, __entry->inode,
 			show_nf_may(__entry->may_flags), __entry->nf_ref,
 			show_nf_flags(__entry->nf_flags),
 			show_nf_may(__entry->nf_may), __entry->nf_file,
-			be32_to_cpu(__entry->status))
+			__entry->status)
 );
 
 DECLARE_EVENT_CLASS(nfsd_file_search_class,
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index bd0a385df3fc..c3fbab1753ec 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -280,19 +280,25 @@ out:
  * Commit metadata changes to stable storage.
  */
 static int
-commit_metadata(struct svc_fh *fhp)
+commit_inode_metadata(struct inode *inode)
 {
-	struct inode *inode = d_inode(fhp->fh_dentry);
 	const struct export_operations *export_ops = inode->i_sb->s_export_op;
 
-	if (!EX_ISSYNC(fhp->fh_export))
-		return 0;
-
 	if (export_ops->commit_metadata)
 		return export_ops->commit_metadata(inode);
 	return sync_inode_metadata(inode, 1);
 }
 
+static int
+commit_metadata(struct svc_fh *fhp)
+{
+	struct inode *inode = d_inode(fhp->fh_dentry);
+
+	if (!EX_ISSYNC(fhp->fh_export))
+		return 0;
+	return commit_inode_metadata(inode);
+}
+
 /*
  * Go over the attributes and take care of the small differences between
  * NFS semantics and what Linux expects.
@@ -358,7 +364,7 @@ out_nfserrno:
  */
 __be32
 nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
-	     int check_guard, time_t guardtime)
+	     int check_guard, time64_t guardtime)
 {
 	struct dentry	*dentry;
 	struct inode	*inode;
@@ -524,17 +530,39 @@ __be32 nfsd4_set_nfs4_label(struct svc_rqst *rqstp, struct svc_fh *fhp,
 }
 #endif
 
-__be32 nfsd4_clone_file_range(struct file *src, u64 src_pos, struct file *dst,
-		u64 dst_pos, u64 count)
+__be32 nfsd4_clone_file_range(struct nfsd_file *nf_src, u64 src_pos,
+		struct nfsd_file *nf_dst, u64 dst_pos, u64 count, bool sync)
 {
+	struct file *src = nf_src->nf_file;
+	struct file *dst = nf_dst->nf_file;
 	loff_t cloned;
+	__be32 ret = 0;
 
+	down_write(&nf_dst->nf_rwsem);
 	cloned = vfs_clone_file_range(src, src_pos, dst, dst_pos, count, 0);
-	if (cloned < 0)
-		return nfserrno(cloned);
-	if (count && cloned != count)
-		return nfserrno(-EINVAL);
-	return 0;
+	if (cloned < 0) {
+		ret = nfserrno(cloned);
+		goto out_err;
+	}
+	if (count && cloned != count) {
+		ret = nfserrno(-EINVAL);
+		goto out_err;
+	}
+	if (sync) {
+		loff_t dst_end = count ? dst_pos + count - 1 : LLONG_MAX;
+		int status = vfs_fsync_range(dst, dst_pos, dst_end, 0);
+
+		if (!status)
+			status = commit_inode_metadata(file_inode(src));
+		if (status < 0) {
+			nfsd_reset_boot_verifier(net_generic(nf_dst->nf_net,
+						 nfsd_net_id));
+			ret = nfserrno(status);
+		}
+	}
+out_err:
+	up_write(&nf_dst->nf_rwsem);
+	return ret;
 }
 
 ssize_t nfsd_copy_file_range(struct file *src, u64 src_pos, struct file *dst,
@@ -932,10 +960,12 @@ static int wait_for_concurrent_writes(struct file *file)
 }
 
 __be32
-nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
+nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf,
 				loff_t offset, struct kvec *vec, int vlen,
-				unsigned long *cnt, int stable)
+				unsigned long *cnt, int stable,
+				__be32 *verf)
 {
+	struct file		*file = nf->nf_file;
 	struct svc_export	*exp;
 	struct iov_iter		iter;
 	__be32			nfserr;
@@ -949,12 +979,13 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 
 	if (test_bit(RQ_LOCAL, &rqstp->rq_flags))
 		/*
-		 * We want less throttling in balance_dirty_pages()
-		 * and shrink_inactive_list() so that nfs to
+		 * We want throttling in balance_dirty_pages()
+		 * and shrink_inactive_list() to only consider
+		 * the backingdev we are writing to, so that nfs to
 		 * localhost doesn't cause nfsd to lock up due to all
 		 * the client's dirty pages or its congested queue.
 		 */
-		current->flags |= PF_LESS_THROTTLE;
+		current->flags |= PF_LOCAL_THROTTLE;
 
 	exp = fhp->fh_export;
 	use_wgather = (rqstp->rq_vers == 2) && EX_WGATHER(exp);
@@ -966,9 +997,28 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 		flags |= RWF_SYNC;
 
 	iov_iter_kvec(&iter, WRITE, vec, vlen, *cnt);
-	host_err = vfs_iter_write(file, &iter, &pos, flags);
-	if (host_err < 0)
+	if (flags & RWF_SYNC) {
+		down_write(&nf->nf_rwsem);
+		host_err = vfs_iter_write(file, &iter, &pos, flags);
+		if (host_err < 0)
+			nfsd_reset_boot_verifier(net_generic(SVC_NET(rqstp),
+						 nfsd_net_id));
+		up_write(&nf->nf_rwsem);
+	} else {
+		down_read(&nf->nf_rwsem);
+		if (verf)
+			nfsd_copy_boot_verifier(verf,
+					net_generic(SVC_NET(rqstp),
+					nfsd_net_id));
+		host_err = vfs_iter_write(file, &iter, &pos, flags);
+		up_read(&nf->nf_rwsem);
+	}
+	if (host_err < 0) {
+		nfsd_reset_boot_verifier(net_generic(SVC_NET(rqstp),
+					 nfsd_net_id));
 		goto out_nfserr;
+	}
+	*cnt = host_err;
 	nfsdstats.io_write += *cnt;
 	fsnotify_modify(file);
 
@@ -988,7 +1038,7 @@ out_nfserr:
 		nfserr = nfserrno(host_err);
 	}
 	if (test_bit(RQ_LOCAL, &rqstp->rq_flags))
-		current_restore_flags(pflags, PF_LESS_THROTTLE);
+		current_restore_flags(pflags, PF_LOCAL_THROTTLE);
 	return nfserr;
 }
 
@@ -1030,7 +1080,8 @@ __be32 nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
  */
 __be32
 nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
-	   struct kvec *vec, int vlen, unsigned long *cnt, int stable)
+	   struct kvec *vec, int vlen, unsigned long *cnt, int stable,
+	   __be32 *verf)
 {
 	struct nfsd_file *nf;
 	__be32 err;
@@ -1041,8 +1092,8 @@ nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
 	if (err)
 		goto out;
 
-	err = nfsd_vfs_write(rqstp, fhp, nf->nf_file, offset, vec,
-			vlen, cnt, stable);
+	err = nfsd_vfs_write(rqstp, fhp, nf, offset, vec,
+			vlen, cnt, stable, verf);
 	nfsd_file_put(nf);
 out:
 	trace_nfsd_write_done(rqstp, fhp, offset, *cnt);
@@ -1061,7 +1112,7 @@ out:
  */
 __be32
 nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp,
-               loff_t offset, unsigned long count)
+               loff_t offset, unsigned long count, __be32 *verf)
 {
 	struct nfsd_file	*nf;
 	loff_t			end = LLONG_MAX;
@@ -1080,10 +1131,14 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (err)
 		goto out;
 	if (EX_ISSYNC(fhp->fh_export)) {
-		int err2 = vfs_fsync_range(nf->nf_file, offset, end, 0);
+		int err2;
 
+		down_write(&nf->nf_rwsem);
+		err2 = vfs_fsync_range(nf->nf_file, offset, end, 0);
 		switch (err2) {
 		case 0:
+			nfsd_copy_boot_verifier(verf, net_generic(nf->nf_net,
+						nfsd_net_id));
 			break;
 		case -EINVAL:
 			err = nfserr_notsupp;
@@ -1093,7 +1148,10 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp,
 			nfsd_reset_boot_verifier(net_generic(nf->nf_net,
 						 nfsd_net_id));
 		}
-	}
+		up_write(&nf->nf_rwsem);
+	} else
+		nfsd_copy_boot_verifier(verf, net_generic(nf->nf_net,
+					nfsd_net_id));
 
 	nfsd_file_put(nf);
 out:
@@ -1117,7 +1175,7 @@ nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *resfhp,
 	if (!uid_eq(current_fsuid(), GLOBAL_ROOT_UID))
 		iap->ia_valid &= ~(ATTR_UID|ATTR_GID);
 	if (iap->ia_valid)
-		return nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0);
+		return nfsd_setattr(rqstp, resfhp, iap, 0, (time64_t)0);
 	/* Callers expect file metadata to be committed here */
 	return nfserrno(commit_metadata(resfhp));
 }
@@ -1380,7 +1438,7 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 			    && d_inode(dchild)->i_atime.tv_sec == v_atime
 			    && d_inode(dchild)->i_size  == 0 ) {
 				if (created)
-					*created = 1;
+					*created = true;
 				break;
 			}
 			/* fall through */
@@ -1389,7 +1447,7 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 			    && d_inode(dchild)->i_atime.tv_sec == v_atime
 			    && d_inode(dchild)->i_size  == 0 ) {
 				if (created)
-					*created = 1;
+					*created = true;
 				goto set_attr;
 			}
 			/* fall through */
@@ -1406,7 +1464,7 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		goto out_nfserr;
 	}
 	if (created)
-		*created = 1;
+		*created = true;
 
 	nfsd_check_ignore_resizing(iap);
 
@@ -1809,7 +1867,17 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 out_drop_write:
 	fh_drop_write(fhp);
 out_nfserr:
-	err = nfserrno(host_err);
+	if (host_err == -EBUSY) {
+		/* name is mounted-on. There is no perfect
+		 * error status.
+		 */
+		if (nfsd_v4client(rqstp))
+			err = nfserr_file_open;
+		else
+			err = nfserr_acces;
+	} else {
+		err = nfserrno(host_err);
+	}
 out:
 	return err;
 }
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
index a13fd9d7e1f5..3eb660ad80d1 100644
--- a/fs/nfsd/vfs.h
+++ b/fs/nfsd/vfs.h
@@ -34,6 +34,8 @@
 #define NFSD_MAY_CREATE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE)
 #define NFSD_MAY_REMOVE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC)
 
+struct nfsd_file;
+
 /*
  * Callback function for readdir
  */
@@ -48,15 +50,16 @@ __be32		 nfsd_lookup_dentry(struct svc_rqst *, struct svc_fh *,
 				const char *, unsigned int,
 				struct svc_export **, struct dentry **);
 __be32		nfsd_setattr(struct svc_rqst *, struct svc_fh *,
-				struct iattr *, int, time_t);
+				struct iattr *, int, time64_t);
 int nfsd_mountpoint(struct dentry *, struct svc_export *);
 #ifdef CONFIG_NFSD_V4
 __be32          nfsd4_set_nfs4_label(struct svc_rqst *, struct svc_fh *,
 		    struct xdr_netobj *);
 __be32		nfsd4_vfs_fallocate(struct svc_rqst *, struct svc_fh *,
 				    struct file *, loff_t, loff_t, int);
-__be32		nfsd4_clone_file_range(struct file *, u64, struct file *,
-			u64, u64);
+__be32		nfsd4_clone_file_range(struct nfsd_file *nf_src, u64 src_pos,
+				       struct nfsd_file *nf_dst, u64 dst_pos,
+				       u64 count, bool sync);
 #endif /* CONFIG_NFSD_V4 */
 __be32		nfsd_create_locked(struct svc_rqst *, struct svc_fh *,
 				char *name, int len, struct iattr *attrs,
@@ -71,7 +74,7 @@ __be32		do_nfsd_create(struct svc_rqst *, struct svc_fh *,
 				struct svc_fh *res, int createmode,
 				u32 *verifier, bool *truncp, bool *created);
 __be32		nfsd_commit(struct svc_rqst *, struct svc_fh *,
-				loff_t, unsigned long);
+				loff_t, unsigned long, __be32 *verf);
 #endif /* CONFIG_NFSD_V3 */
 int 		nfsd_open_break_lease(struct inode *, int);
 __be32		nfsd_open(struct svc_rqst *, struct svc_fh *, umode_t,
@@ -91,11 +94,12 @@ __be32 		nfsd_read(struct svc_rqst *, struct svc_fh *,
 				loff_t, struct kvec *, int, unsigned long *,
 				u32 *eof);
 __be32 		nfsd_write(struct svc_rqst *, struct svc_fh *, loff_t,
-				struct kvec *, int, unsigned long *, int);
+				struct kvec *, int, unsigned long *,
+				int stable, __be32 *verf);
 __be32		nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
-				struct file *file, loff_t offset,
+				struct nfsd_file *nf, loff_t offset,
 				struct kvec *vec, int vlen, unsigned long *cnt,
-				int stable);
+				int stable, __be32 *verf);
 __be32		nfsd_readlink(struct svc_rqst *, struct svc_fh *,
 				char *, int *);
 __be32		nfsd_symlink(struct svc_rqst *, struct svc_fh *,
diff --git a/fs/nfsd/xdr3.h b/fs/nfsd/xdr3.h
index 99ff9f403ff1..4155fd71714c 100644
--- a/fs/nfsd/xdr3.h
+++ b/fs/nfsd/xdr3.h
@@ -14,7 +14,7 @@ struct nfsd3_sattrargs {
 	struct svc_fh		fh;
 	struct iattr		attrs;
 	int			check_guard;
-	time_t			guardtime;
+	time64_t		guardtime;
 };
 
 struct nfsd3_diropargs {
@@ -159,6 +159,7 @@ struct nfsd3_writeres {
 	struct svc_fh		fh;
 	unsigned long		count;
 	int			committed;
+	__be32			verf[2];
 };
 
 struct nfsd3_renameres {
@@ -223,6 +224,7 @@ struct nfsd3_pathconfres {
 struct nfsd3_commitres {
 	__be32			status;
 	struct svc_fh		fh;
+	__be32			verf[2];
 };
 
 struct nfsd3_getaclres {
diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h
index f4737d66ee98..db63d39b1507 100644
--- a/fs/nfsd/xdr4.h
+++ b/fs/nfsd/xdr4.h
@@ -46,9 +46,9 @@
 #define CURRENT_STATE_ID_FLAG (1<<0)
 #define SAVED_STATE_ID_FLAG (1<<1)
 
-#define SET_STATE_ID(c, f) ((c)->sid_flags |= (f))
-#define HAS_STATE_ID(c, f) ((c)->sid_flags & (f))
-#define CLEAR_STATE_ID(c, f) ((c)->sid_flags &= ~(f))
+#define SET_CSTATE_FLAG(c, f) ((c)->sid_flags |= (f))
+#define HAS_CSTATE_FLAG(c, f) ((c)->sid_flags & (f))
+#define CLEAR_CSTATE_FLAG(c, f) ((c)->sid_flags &= ~(f))
 
 struct nfsd4_compound_state {
 	struct svc_fh		current_fh;
@@ -221,6 +221,7 @@ struct nfsd4_lookup {
 struct nfsd4_putfh {
 	u32		pf_fhlen;           /* request */
 	char		*pf_fhval;          /* request */
+	bool		no_verify;	    /* represents foreigh fh */
 };
 
 struct nfsd4_open {
@@ -518,11 +519,13 @@ struct nfsd42_write_res {
 
 struct nfsd4_copy {
 	/* request */
-	stateid_t	cp_src_stateid;
-	stateid_t	cp_dst_stateid;
-	u64		cp_src_pos;
-	u64		cp_dst_pos;
-	u64		cp_count;
+	stateid_t		cp_src_stateid;
+	stateid_t		cp_dst_stateid;
+	u64			cp_src_pos;
+	u64			cp_dst_pos;
+	u64			cp_count;
+	struct nl4_server	cp_src;
+	bool			cp_intra;
 
 	/* both */
 	bool		cp_synchronous;
@@ -540,13 +543,18 @@ struct nfsd4_copy {
 	struct nfsd_file        *nf_src;
 	struct nfsd_file        *nf_dst;
 
-	stateid_t		cp_stateid;
+	copy_stateid_t		cp_stateid;
 
 	struct list_head	copies;
 	struct task_struct	*copy_task;
 	refcount_t		refcount;
 	bool			stopped;
+
+	struct vfsmount		*ss_mnt;
+	struct nfs_fh		c_fh;
+	nfs4_stateid		stateid;
 };
+extern bool inter_copy_offload_enable;
 
 struct nfsd4_seek {
 	/* request */
@@ -568,6 +576,18 @@ struct nfsd4_offload_status {
 	u32		status;
 };
 
+struct nfsd4_copy_notify {
+	/* request */
+	stateid_t		cpn_src_stateid;
+	struct nl4_server	cpn_dst;
+
+	/* response */
+	stateid_t		cpn_cnr_stateid;
+	u64			cpn_sec;
+	u32			cpn_nsec;
+	struct nl4_server	cpn_src;
+};
+
 struct nfsd4_op {
 	int					opnum;
 	const struct nfsd4_operation *		opdesc;
@@ -627,6 +647,7 @@ struct nfsd4_op {
 		struct nfsd4_clone		clone;
 		struct nfsd4_copy		copy;
 		struct nfsd4_offload_status	offload_status;
+		struct nfsd4_copy_notify	copy_notify;
 		struct nfsd4_seek		seek;
 	} u;
 	struct nfs4_replay *			replay;
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 671085512e0f..ceeb3b441844 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -145,18 +145,9 @@ static int nilfs_readpage(struct file *file, struct page *page)
 	return mpage_readpage(page, nilfs_get_block);
 }
 
-/**
- * nilfs_readpages() - implement readpages() method of nilfs_aops {}
- * address_space_operations.
- * @file - file struct of the file to be read
- * @mapping - address_space struct used for reading multiple pages
- * @pages - the pages to be read
- * @nr_pages - number of pages to be read
- */
-static int nilfs_readpages(struct file *file, struct address_space *mapping,
-			   struct list_head *pages, unsigned int nr_pages)
+static void nilfs_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, nilfs_get_block);
+	mpage_readahead(rac, nilfs_get_block);
 }
 
 static int nilfs_writepages(struct address_space *mapping,
@@ -308,7 +299,7 @@ const struct address_space_operations nilfs_aops = {
 	.readpage		= nilfs_readpage,
 	.writepages		= nilfs_writepages,
 	.set_page_dirty		= nilfs_set_page_dirty,
-	.readpages		= nilfs_readpages,
+	.readahead		= nilfs_readahead,
 	.write_begin		= nilfs_write_begin,
 	.write_end		= nilfs_write_end,
 	/* .releasepage		= nilfs_releasepage, */
diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c
index 91b9dac6b2cc..4ba73dbf3e8d 100644
--- a/fs/nilfs2/ioctl.c
+++ b/fs/nilfs2/ioctl.c
@@ -1354,6 +1354,7 @@ long nilfs_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	case NILFS_IOCTL_SYNC:
 	case NILFS_IOCTL_RESIZE:
 	case NILFS_IOCTL_SET_ALLOC_RANGE:
+	case FITRIM:
 		break;
 	default:
 		return -ENOIOCTLCMD;
diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h
index 380a543c5b19..b55cdeb4d169 100644
--- a/fs/nilfs2/the_nilfs.h
+++ b/fs/nilfs2/the_nilfs.h
@@ -375,7 +375,7 @@ static inline int nilfs_flush_device(struct the_nilfs *nilfs)
 	 */
 	smp_wmb();
 
-	err = blkdev_issue_flush(nilfs->ns_bdev, GFP_KERNEL, NULL);
+	err = blkdev_issue_flush(nilfs->ns_bdev, GFP_KERNEL);
 	if (err != -EIO)
 		err = 0;
 	return err;
diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 5778d1347b35..c18459cea6f4 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -17,6 +17,59 @@
 
 #include "fanotify.h"
 
+static bool fanotify_path_equal(struct path *p1, struct path *p2)
+{
+	return p1->mnt == p2->mnt && p1->dentry == p2->dentry;
+}
+
+static inline bool fanotify_fsid_equal(__kernel_fsid_t *fsid1,
+				       __kernel_fsid_t *fsid2)
+{
+	return fsid1->val[0] == fsid2->val[0] && fsid1->val[1] == fsid2->val[1];
+}
+
+static bool fanotify_fh_equal(struct fanotify_fh *fh1,
+			      struct fanotify_fh *fh2)
+{
+	if (fh1->type != fh2->type || fh1->len != fh2->len)
+		return false;
+
+	/* Do not merge events if we failed to encode fh */
+	if (fh1->type == FILEID_INVALID)
+		return false;
+
+	return !fh1->len ||
+		!memcmp(fanotify_fh_buf(fh1), fanotify_fh_buf(fh2), fh1->len);
+}
+
+static bool fanotify_fid_event_equal(struct fanotify_fid_event *ffe1,
+				     struct fanotify_fid_event *ffe2)
+{
+	/* Do not merge fid events without object fh */
+	if (!ffe1->object_fh.len)
+		return false;
+
+	return fanotify_fsid_equal(&ffe1->fsid, &ffe2->fsid) &&
+		fanotify_fh_equal(&ffe1->object_fh, &ffe2->object_fh);
+}
+
+static bool fanotify_name_event_equal(struct fanotify_name_event *fne1,
+				      struct fanotify_name_event *fne2)
+{
+	/*
+	 * Do not merge name events without dir fh.
+	 * FAN_DIR_MODIFY does not encode object fh, so it may be empty.
+	 */
+	if (!fne1->dir_fh.len)
+		return false;
+
+	if (fne1->name_len != fne2->name_len ||
+	    !fanotify_fh_equal(&fne1->dir_fh, &fne2->dir_fh))
+		return false;
+
+	return !memcmp(fne1->name, fne2->name, fne1->name_len);
+}
+
 static bool should_merge(struct fsnotify_event *old_fsn,
 			 struct fsnotify_event *new_fsn)
 {
@@ -26,14 +79,15 @@ static bool should_merge(struct fsnotify_event *old_fsn,
 	old = FANOTIFY_E(old_fsn);
 	new = FANOTIFY_E(new_fsn);
 
-	if (old_fsn->inode != new_fsn->inode || old->pid != new->pid ||
-	    old->fh_type != new->fh_type || old->fh_len != new->fh_len)
+	if (old_fsn->objectid != new_fsn->objectid ||
+	    old->type != new->type || old->pid != new->pid)
 		return false;
 
-	if (fanotify_event_has_path(old)) {
-		return old->path.mnt == new->path.mnt &&
-			old->path.dentry == new->path.dentry;
-	} else if (fanotify_event_has_fid(old)) {
+	switch (old->type) {
+	case FANOTIFY_EVENT_TYPE_PATH:
+		return fanotify_path_equal(fanotify_event_path(old),
+					   fanotify_event_path(new));
+	case FANOTIFY_EVENT_TYPE_FID:
 		/*
 		 * We want to merge many dirent events in the same dir (i.e.
 		 * creates/unlinks/renames), but we do not want to merge dirent
@@ -42,11 +96,18 @@ static bool should_merge(struct fsnotify_event *old_fsn,
 		 * mask FAN_CREATE|FAN_DELETE|FAN_ONDIR if it describes mkdir+
 		 * unlink pair or rmdir+create pair of events.
 		 */
-		return (old->mask & FS_ISDIR) == (new->mask & FS_ISDIR) &&
-			fanotify_fid_equal(&old->fid, &new->fid, old->fh_len);
+		if ((old->mask & FS_ISDIR) != (new->mask & FS_ISDIR))
+			return false;
+
+		return fanotify_fid_event_equal(FANOTIFY_FE(old),
+						FANOTIFY_FE(new));
+	case FANOTIFY_EVENT_TYPE_FID_NAME:
+		return fanotify_name_event_equal(FANOTIFY_NE(old),
+						 FANOTIFY_NE(new));
+	default:
+		WARN_ON_ONCE(1);
 	}
 
-	/* Do not merge events if we failed to encode fid */
 	return false;
 }
 
@@ -151,7 +212,7 @@ static u32 fanotify_group_event_mask(struct fsnotify_group *group,
 {
 	__u32 marks_mask = 0, marks_ignored_mask = 0;
 	__u32 test_mask, user_mask = FANOTIFY_OUTGOING_EVENTS;
-	const struct path *path = data;
+	const struct path *path = fsnotify_data_path(data, data_type);
 	struct fsnotify_mark *mark;
 	int type;
 
@@ -160,7 +221,7 @@ static u32 fanotify_group_event_mask(struct fsnotify_group *group,
 
 	if (!FAN_GROUP_FLAG(group, FAN_REPORT_FID)) {
 		/* Do we have path to open a file descriptor? */
-		if (data_type != FSNOTIFY_EVENT_PATH)
+		if (!path)
 			return 0;
 		/* Path type events are only relevant for files and dirs */
 		if (!d_is_reg(path->dentry) && !d_can_lookup(path->dentry))
@@ -172,6 +233,13 @@ static u32 fanotify_group_event_mask(struct fsnotify_group *group,
 			continue;
 		mark = iter_info->marks[type];
 		/*
+		 * If the event is on dir and this mark doesn't care about
+		 * events on dir, don't send it!
+		 */
+		if (event_mask & FS_ISDIR && !(mark->mask & FS_ISDIR))
+			continue;
+
+		/*
 		 * If the event is for a child and this mark doesn't care about
 		 * events on a child, don't send it!
 		 */
@@ -187,9 +255,9 @@ static u32 fanotify_group_event_mask(struct fsnotify_group *group,
 	test_mask = event_mask & marks_mask & ~marks_ignored_mask;
 
 	/*
-	 * dirent modification events (create/delete/move) do not carry the
-	 * child entry name/inode information. Instead, we report FAN_ONDIR
-	 * for mkdir/rmdir so user can differentiate them from creat/unlink.
+	 * For dirent modification events (create/delete/move) that do not carry
+	 * the child entry name information, we report FAN_ONDIR for mkdir/rmdir
+	 * so user can differentiate them from creat/unlink.
 	 *
 	 * For backward compatibility and consistency, do not report FAN_ONDIR
 	 * to user in legacy fanotify mode (reporting fd) and report FAN_ONDIR
@@ -203,22 +271,20 @@ static u32 fanotify_group_event_mask(struct fsnotify_group *group,
 		user_mask &= ~FAN_ONDIR;
 	}
 
-	if (event_mask & FS_ISDIR &&
-	    !(marks_mask & FS_ISDIR & ~marks_ignored_mask))
-		return 0;
-
 	return test_mask & user_mask;
 }
 
-static int fanotify_encode_fid(struct fanotify_event *event,
-			       struct inode *inode, gfp_t gfp,
-			       __kernel_fsid_t *fsid)
+static void fanotify_encode_fh(struct fanotify_fh *fh, struct inode *inode,
+			       gfp_t gfp)
 {
-	struct fanotify_fid *fid = &event->fid;
-	int dwords, bytes = 0;
-	int err, type;
+	int dwords, type, bytes = 0;
+	char *ext_buf = NULL;
+	void *buf = fh->buf;
+	int err;
+
+	if (!inode)
+		goto out;
 
-	fid->ext_fh = NULL;
 	dwords = 0;
 	err = -ENOENT;
 	type = exportfs_encode_inode_fh(inode, NULL, &dwords, NULL);
@@ -229,31 +295,33 @@ static int fanotify_encode_fid(struct fanotify_event *event,
 	if (bytes > FANOTIFY_INLINE_FH_LEN) {
 		/* Treat failure to allocate fh as failure to allocate event */
 		err = -ENOMEM;
-		fid->ext_fh = kmalloc(bytes, gfp);
-		if (!fid->ext_fh)
+		ext_buf = kmalloc(bytes, gfp);
+		if (!ext_buf)
 			goto out_err;
+
+		*fanotify_fh_ext_buf_ptr(fh) = ext_buf;
+		buf = ext_buf;
 	}
 
-	type = exportfs_encode_inode_fh(inode, fanotify_fid_fh(fid, bytes),
-					&dwords, NULL);
+	type = exportfs_encode_inode_fh(inode, buf, &dwords, NULL);
 	err = -EINVAL;
 	if (!type || type == FILEID_INVALID || bytes != dwords << 2)
 		goto out_err;
 
-	fid->fsid = *fsid;
-	event->fh_len = bytes;
+	fh->type = type;
+	fh->len = bytes;
 
-	return type;
+	return;
 
 out_err:
-	pr_warn_ratelimited("fanotify: failed to encode fid (fsid=%x.%x, "
-			    "type=%d, bytes=%d, err=%i)\n",
-			    fsid->val[0], fsid->val[1], type, bytes, err);
-	kfree(fid->ext_fh);
-	fid->ext_fh = NULL;
-	event->fh_len = 0;
-
-	return FILEID_INVALID;
+	pr_warn_ratelimited("fanotify: failed to encode fid (type=%d, len=%d, err=%i)\n",
+			    type, bytes, err);
+	kfree(ext_buf);
+	*fanotify_fh_ext_buf_ptr(fh) = NULL;
+out:
+	/* Report the event without a file identifier on encode error */
+	fh->type = FILEID_INVALID;
+	fh->len = 0;
 }
 
 /*
@@ -269,21 +337,22 @@ static struct inode *fanotify_fid_inode(struct inode *to_tell, u32 event_mask,
 {
 	if (event_mask & ALL_FSNOTIFY_DIRENT_EVENTS)
 		return to_tell;
-	else if (data_type == FSNOTIFY_EVENT_INODE)
-		return (struct inode *)data;
-	else if (data_type == FSNOTIFY_EVENT_PATH)
-		return d_inode(((struct path *)data)->dentry);
-	return NULL;
+
+	return (struct inode *)fsnotify_data_inode(data, data_type);
 }
 
 struct fanotify_event *fanotify_alloc_event(struct fsnotify_group *group,
 					    struct inode *inode, u32 mask,
 					    const void *data, int data_type,
+					    const struct qstr *file_name,
 					    __kernel_fsid_t *fsid)
 {
 	struct fanotify_event *event = NULL;
+	struct fanotify_fid_event *ffe = NULL;
+	struct fanotify_name_event *fne = NULL;
 	gfp_t gfp = GFP_KERNEL_ACCOUNT;
 	struct inode *id = fanotify_fid_inode(inode, mask, data, data_type);
+	const struct path *path = fsnotify_data_path(data, data_type);
 
 	/*
 	 * For queues with unlimited length lost events are not expected and
@@ -305,33 +374,81 @@ struct fanotify_event *fanotify_alloc_event(struct fsnotify_group *group,
 		pevent = kmem_cache_alloc(fanotify_perm_event_cachep, gfp);
 		if (!pevent)
 			goto out;
+
 		event = &pevent->fae;
+		event->type = FANOTIFY_EVENT_TYPE_PATH_PERM;
 		pevent->response = 0;
 		pevent->state = FAN_EVENT_INIT;
 		goto init;
 	}
-	event = kmem_cache_alloc(fanotify_event_cachep, gfp);
-	if (!event)
-		goto out;
-init: __maybe_unused
-	fsnotify_init_event(&event->fse, inode);
+
+	/*
+	 * For FAN_DIR_MODIFY event, we report the fid of the directory and
+	 * the name of the modified entry.
+	 * Allocate an fanotify_name_event struct and copy the name.
+	 */
+	if (mask & FAN_DIR_MODIFY && !(WARN_ON_ONCE(!file_name))) {
+		fne = kmalloc(sizeof(*fne) + file_name->len + 1, gfp);
+		if (!fne)
+			goto out;
+
+		event = &fne->fae;
+		event->type = FANOTIFY_EVENT_TYPE_FID_NAME;
+		fne->name_len = file_name->len;
+		strcpy(fne->name, file_name->name);
+		goto init;
+	}
+
+	if (FAN_GROUP_FLAG(group, FAN_REPORT_FID)) {
+		ffe = kmem_cache_alloc(fanotify_fid_event_cachep, gfp);
+		if (!ffe)
+			goto out;
+
+		event = &ffe->fae;
+		event->type = FANOTIFY_EVENT_TYPE_FID;
+	} else {
+		struct fanotify_path_event *pevent;
+
+		pevent = kmem_cache_alloc(fanotify_path_event_cachep, gfp);
+		if (!pevent)
+			goto out;
+
+		event = &pevent->fae;
+		event->type = FANOTIFY_EVENT_TYPE_PATH;
+	}
+
+init:
+	/*
+	 * Use the victim inode instead of the watching inode as the id for
+	 * event queue, so event reported on parent is merged with event
+	 * reported on child when both directory and child watches exist.
+	 */
+	fsnotify_init_event(&event->fse, (unsigned long)id);
 	event->mask = mask;
 	if (FAN_GROUP_FLAG(group, FAN_REPORT_TID))
 		event->pid = get_pid(task_pid(current));
 	else
 		event->pid = get_pid(task_tgid(current));
-	event->fh_len = 0;
-	if (id && FAN_GROUP_FLAG(group, FAN_REPORT_FID)) {
-		/* Report the event without a file identifier on encode error */
-		event->fh_type = fanotify_encode_fid(event, id, gfp, fsid);
-	} else if (data_type == FSNOTIFY_EVENT_PATH) {
-		event->fh_type = FILEID_ROOT;
-		event->path = *((struct path *)data);
-		path_get(&event->path);
-	} else {
-		event->fh_type = FILEID_INVALID;
-		event->path.mnt = NULL;
-		event->path.dentry = NULL;
+
+	if (fsid && fanotify_event_fsid(event))
+		*fanotify_event_fsid(event) = *fsid;
+
+	if (fanotify_event_object_fh(event))
+		fanotify_encode_fh(fanotify_event_object_fh(event), id, gfp);
+
+	if (fanotify_event_dir_fh(event))
+		fanotify_encode_fh(fanotify_event_dir_fh(event), id, gfp);
+
+	if (fanotify_event_has_path(event)) {
+		struct path *p = fanotify_event_path(event);
+
+		if (path) {
+			*p = *path;
+			path_get(path);
+		} else {
+			p->mnt = NULL;
+			p->dentry = NULL;
+		}
 	}
 out:
 	memalloc_unuse_memcg();
@@ -392,6 +509,7 @@ static int fanotify_handle_event(struct fsnotify_group *group,
 	BUILD_BUG_ON(FAN_MOVED_FROM != FS_MOVED_FROM);
 	BUILD_BUG_ON(FAN_CREATE != FS_CREATE);
 	BUILD_BUG_ON(FAN_DELETE != FS_DELETE);
+	BUILD_BUG_ON(FAN_DIR_MODIFY != FS_DIR_MODIFY);
 	BUILD_BUG_ON(FAN_DELETE_SELF != FS_DELETE_SELF);
 	BUILD_BUG_ON(FAN_MOVE_SELF != FS_MOVE_SELF);
 	BUILD_BUG_ON(FAN_EVENT_ON_CHILD != FS_EVENT_ON_CHILD);
@@ -429,7 +547,7 @@ static int fanotify_handle_event(struct fsnotify_group *group,
 	}
 
 	event = fanotify_alloc_event(group, inode, mask, data, data_type,
-				     &fsid);
+				     file_name, &fsid);
 	ret = -ENOMEM;
 	if (unlikely(!event)) {
 		/*
@@ -451,7 +569,7 @@ static int fanotify_handle_event(struct fsnotify_group *group,
 
 		ret = 0;
 	} else if (fanotify_is_perm_event(mask)) {
-		ret = fanotify_get_response(group, FANOTIFY_PE(fsn_event),
+		ret = fanotify_get_response(group, FANOTIFY_PERM(event),
 					    iter_info);
 	}
 finish:
@@ -470,22 +588,58 @@ static void fanotify_free_group_priv(struct fsnotify_group *group)
 	free_uid(user);
 }
 
+static void fanotify_free_path_event(struct fanotify_event *event)
+{
+	path_put(fanotify_event_path(event));
+	kmem_cache_free(fanotify_path_event_cachep, FANOTIFY_PE(event));
+}
+
+static void fanotify_free_perm_event(struct fanotify_event *event)
+{
+	path_put(fanotify_event_path(event));
+	kmem_cache_free(fanotify_perm_event_cachep, FANOTIFY_PERM(event));
+}
+
+static void fanotify_free_fid_event(struct fanotify_event *event)
+{
+	struct fanotify_fid_event *ffe = FANOTIFY_FE(event);
+
+	if (fanotify_fh_has_ext_buf(&ffe->object_fh))
+		kfree(fanotify_fh_ext_buf(&ffe->object_fh));
+	kmem_cache_free(fanotify_fid_event_cachep, ffe);
+}
+
+static void fanotify_free_name_event(struct fanotify_event *event)
+{
+	struct fanotify_name_event *fne = FANOTIFY_NE(event);
+
+	if (fanotify_fh_has_ext_buf(&fne->dir_fh))
+		kfree(fanotify_fh_ext_buf(&fne->dir_fh));
+	kfree(fne);
+}
+
 static void fanotify_free_event(struct fsnotify_event *fsn_event)
 {
 	struct fanotify_event *event;
 
 	event = FANOTIFY_E(fsn_event);
-	if (fanotify_event_has_path(event))
-		path_put(&event->path);
-	else if (fanotify_event_has_ext_fh(event))
-		kfree(event->fid.ext_fh);
 	put_pid(event->pid);
-	if (fanotify_is_perm_event(event->mask)) {
-		kmem_cache_free(fanotify_perm_event_cachep,
-				FANOTIFY_PE(fsn_event));
-		return;
+	switch (event->type) {
+	case FANOTIFY_EVENT_TYPE_PATH:
+		fanotify_free_path_event(event);
+		break;
+	case FANOTIFY_EVENT_TYPE_PATH_PERM:
+		fanotify_free_perm_event(event);
+		break;
+	case FANOTIFY_EVENT_TYPE_FID:
+		fanotify_free_fid_event(event);
+		break;
+	case FANOTIFY_EVENT_TYPE_FID_NAME:
+		fanotify_free_name_event(event);
+		break;
+	default:
+		WARN_ON_ONCE(1);
 	}
-	kmem_cache_free(fanotify_event_cachep, event);
 }
 
 static void fanotify_free_mark(struct fsnotify_mark *fsn_mark)
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index 68b30504284c..35bfbf4a7aac 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -5,7 +5,8 @@
 #include <linux/exportfs.h>
 
 extern struct kmem_cache *fanotify_mark_cache;
-extern struct kmem_cache *fanotify_event_cachep;
+extern struct kmem_cache *fanotify_fid_event_cachep;
+extern struct kmem_cache *fanotify_path_event_cachep;
 extern struct kmem_cache *fanotify_perm_event_cachep;
 
 /* Possible states of the permission event */
@@ -18,94 +19,140 @@ enum {
 
 /*
  * 3 dwords are sufficient for most local fs (64bit ino, 32bit generation).
- * For 32bit arch, fid increases the size of fanotify_event by 12 bytes and
- * fh_* fields increase the size of fanotify_event by another 4 bytes.
- * For 64bit arch, fid increases the size of fanotify_fid by 8 bytes and
- * fh_* fields are packed in a hole after mask.
+ * fh buf should be dword aligned. On 64bit arch, the ext_buf pointer is
+ * stored in either the first or last 2 dwords.
  */
-#if BITS_PER_LONG == 32
 #define FANOTIFY_INLINE_FH_LEN	(3 << 2)
-#else
-#define FANOTIFY_INLINE_FH_LEN	(4 << 2)
-#endif
 
-struct fanotify_fid {
-	__kernel_fsid_t fsid;
-	union {
-		unsigned char fh[FANOTIFY_INLINE_FH_LEN];
-		unsigned char *ext_fh;
-	};
-};
+struct fanotify_fh {
+	unsigned char buf[FANOTIFY_INLINE_FH_LEN];
+	u8 type;
+	u8 len;
+} __aligned(4);
+
+static inline bool fanotify_fh_has_ext_buf(struct fanotify_fh *fh)
+{
+	return fh->len > FANOTIFY_INLINE_FH_LEN;
+}
+
+static inline char **fanotify_fh_ext_buf_ptr(struct fanotify_fh *fh)
+{
+	BUILD_BUG_ON(__alignof__(char *) - 4 + sizeof(char *) >
+		     FANOTIFY_INLINE_FH_LEN);
+	return (char **)ALIGN((unsigned long)(fh->buf), __alignof__(char *));
+}
 
-static inline void *fanotify_fid_fh(struct fanotify_fid *fid,
-				    unsigned int fh_len)
+static inline void *fanotify_fh_ext_buf(struct fanotify_fh *fh)
 {
-	return fh_len <= FANOTIFY_INLINE_FH_LEN ? fid->fh : fid->ext_fh;
+	return *fanotify_fh_ext_buf_ptr(fh);
 }
 
-static inline bool fanotify_fid_equal(struct fanotify_fid *fid1,
-				      struct fanotify_fid *fid2,
-				      unsigned int fh_len)
+static inline void *fanotify_fh_buf(struct fanotify_fh *fh)
 {
-	return fid1->fsid.val[0] == fid2->fsid.val[0] &&
-		fid1->fsid.val[1] == fid2->fsid.val[1] &&
-		!memcmp(fanotify_fid_fh(fid1, fh_len),
-			fanotify_fid_fh(fid2, fh_len), fh_len);
+	return fanotify_fh_has_ext_buf(fh) ? fanotify_fh_ext_buf(fh) : fh->buf;
 }
 
 /*
- * Structure for normal fanotify events. It gets allocated in
+ * Common structure for fanotify events. Concrete structs are allocated in
  * fanotify_handle_event() and freed when the information is retrieved by
- * userspace
+ * userspace. The type of event determines how it was allocated, how it will
+ * be freed and which concrete struct it may be cast to.
  */
+enum fanotify_event_type {
+	FANOTIFY_EVENT_TYPE_FID, /* fixed length */
+	FANOTIFY_EVENT_TYPE_FID_NAME, /* variable length */
+	FANOTIFY_EVENT_TYPE_PATH,
+	FANOTIFY_EVENT_TYPE_PATH_PERM,
+};
+
 struct fanotify_event {
 	struct fsnotify_event fse;
 	u32 mask;
-	/*
-	 * Those fields are outside fanotify_fid to pack fanotify_event nicely
-	 * on 64bit arch and to use fh_type as an indication of whether path
-	 * or fid are used in the union:
-	 * FILEID_ROOT (0) for path, > 0 for fid, FILEID_INVALID for neither.
-	 */
-	u8 fh_type;
-	u8 fh_len;
-	u16 pad;
-	union {
-		/*
-		 * We hold ref to this path so it may be dereferenced at any
-		 * point during this object's lifetime
-		 */
-		struct path path;
-		/*
-		 * With FAN_REPORT_FID, we do not hold any reference on the
-		 * victim object. Instead we store its NFS file handle and its
-		 * filesystem's fsid as a unique identifier.
-		 */
-		struct fanotify_fid fid;
-	};
+	enum fanotify_event_type type;
 	struct pid *pid;
 };
 
-static inline bool fanotify_event_has_path(struct fanotify_event *event)
+struct fanotify_fid_event {
+	struct fanotify_event fae;
+	__kernel_fsid_t fsid;
+	struct fanotify_fh object_fh;
+};
+
+static inline struct fanotify_fid_event *
+FANOTIFY_FE(struct fanotify_event *event)
 {
-	return event->fh_type == FILEID_ROOT;
+	return container_of(event, struct fanotify_fid_event, fae);
 }
 
-static inline bool fanotify_event_has_fid(struct fanotify_event *event)
+struct fanotify_name_event {
+	struct fanotify_event fae;
+	__kernel_fsid_t fsid;
+	struct fanotify_fh dir_fh;
+	u8 name_len;
+	char name[0];
+};
+
+static inline struct fanotify_name_event *
+FANOTIFY_NE(struct fanotify_event *event)
 {
-	return event->fh_type != FILEID_ROOT &&
-		event->fh_type != FILEID_INVALID;
+	return container_of(event, struct fanotify_name_event, fae);
 }
 
-static inline bool fanotify_event_has_ext_fh(struct fanotify_event *event)
+static inline __kernel_fsid_t *fanotify_event_fsid(struct fanotify_event *event)
 {
-	return fanotify_event_has_fid(event) &&
-		event->fh_len > FANOTIFY_INLINE_FH_LEN;
+	if (event->type == FANOTIFY_EVENT_TYPE_FID)
+		return &FANOTIFY_FE(event)->fsid;
+	else if (event->type == FANOTIFY_EVENT_TYPE_FID_NAME)
+		return &FANOTIFY_NE(event)->fsid;
+	else
+		return NULL;
 }
 
-static inline void *fanotify_event_fh(struct fanotify_event *event)
+static inline struct fanotify_fh *fanotify_event_object_fh(
+						struct fanotify_event *event)
 {
-	return fanotify_fid_fh(&event->fid, event->fh_len);
+	if (event->type == FANOTIFY_EVENT_TYPE_FID)
+		return &FANOTIFY_FE(event)->object_fh;
+	else
+		return NULL;
+}
+
+static inline struct fanotify_fh *fanotify_event_dir_fh(
+						struct fanotify_event *event)
+{
+	if (event->type == FANOTIFY_EVENT_TYPE_FID_NAME)
+		return &FANOTIFY_NE(event)->dir_fh;
+	else
+		return NULL;
+}
+
+static inline int fanotify_event_object_fh_len(struct fanotify_event *event)
+{
+	struct fanotify_fh *fh = fanotify_event_object_fh(event);
+
+	return fh ? fh->len : 0;
+}
+
+static inline bool fanotify_event_has_name(struct fanotify_event *event)
+{
+	return event->type == FANOTIFY_EVENT_TYPE_FID_NAME;
+}
+
+static inline int fanotify_event_name_len(struct fanotify_event *event)
+{
+	return fanotify_event_has_name(event) ?
+		FANOTIFY_NE(event)->name_len : 0;
+}
+
+struct fanotify_path_event {
+	struct fanotify_event fae;
+	struct path path;
+};
+
+static inline struct fanotify_path_event *
+FANOTIFY_PE(struct fanotify_event *event)
+{
+	return container_of(event, struct fanotify_path_event, fae);
 }
 
 /*
@@ -117,15 +164,16 @@ static inline void *fanotify_event_fh(struct fanotify_event *event)
  */
 struct fanotify_perm_event {
 	struct fanotify_event fae;
+	struct path path;
 	unsigned short response;	/* userspace answer to the event */
 	unsigned short state;		/* state of the event */
 	int fd;		/* fd we passed to userspace for this event */
 };
 
 static inline struct fanotify_perm_event *
-FANOTIFY_PE(struct fsnotify_event *fse)
+FANOTIFY_PERM(struct fanotify_event *event)
 {
-	return container_of(fse, struct fanotify_perm_event, fae.fse);
+	return container_of(event, struct fanotify_perm_event, fae);
 }
 
 static inline bool fanotify_is_perm_event(u32 mask)
@@ -139,7 +187,24 @@ static inline struct fanotify_event *FANOTIFY_E(struct fsnotify_event *fse)
 	return container_of(fse, struct fanotify_event, fse);
 }
 
+static inline bool fanotify_event_has_path(struct fanotify_event *event)
+{
+	return event->type == FANOTIFY_EVENT_TYPE_PATH ||
+		event->type == FANOTIFY_EVENT_TYPE_PATH_PERM;
+}
+
+static inline struct path *fanotify_event_path(struct fanotify_event *event)
+{
+	if (event->type == FANOTIFY_EVENT_TYPE_PATH)
+		return &FANOTIFY_PE(event)->path;
+	else if (event->type == FANOTIFY_EVENT_TYPE_PATH_PERM)
+		return &FANOTIFY_PERM(event)->path;
+	else
+		return NULL;
+}
+
 struct fanotify_event *fanotify_alloc_event(struct fsnotify_group *group,
 					    struct inode *inode, u32 mask,
 					    const void *data, int data_type,
+					    const struct qstr *file_name,
 					    __kernel_fsid_t *fsid);
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 8508ab575017..42cb794c62ac 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -46,32 +46,53 @@
 extern const struct fsnotify_ops fanotify_fsnotify_ops;
 
 struct kmem_cache *fanotify_mark_cache __read_mostly;
-struct kmem_cache *fanotify_event_cachep __read_mostly;
+struct kmem_cache *fanotify_fid_event_cachep __read_mostly;
+struct kmem_cache *fanotify_path_event_cachep __read_mostly;
 struct kmem_cache *fanotify_perm_event_cachep __read_mostly;
 
 #define FANOTIFY_EVENT_ALIGN 4
+#define FANOTIFY_INFO_HDR_LEN \
+	(sizeof(struct fanotify_event_info_fid) + sizeof(struct file_handle))
+
+static int fanotify_fid_info_len(int fh_len, int name_len)
+{
+	int info_len = fh_len;
+
+	if (name_len)
+		info_len += name_len + 1;
+
+	return roundup(FANOTIFY_INFO_HDR_LEN + info_len, FANOTIFY_EVENT_ALIGN);
+}
 
 static int fanotify_event_info_len(struct fanotify_event *event)
 {
-	if (!fanotify_event_has_fid(event))
-		return 0;
+	int info_len = 0;
+	int fh_len = fanotify_event_object_fh_len(event);
+
+	if (fh_len)
+		info_len += fanotify_fid_info_len(fh_len, 0);
 
-	return roundup(sizeof(struct fanotify_event_info_fid) +
-		       sizeof(struct file_handle) + event->fh_len,
-		       FANOTIFY_EVENT_ALIGN);
+	if (fanotify_event_name_len(event)) {
+		struct fanotify_name_event *fne = FANOTIFY_NE(event);
+
+		info_len += fanotify_fid_info_len(fne->dir_fh.len,
+						  fne->name_len);
+	}
+
+	return info_len;
 }
 
 /*
- * Get an fsnotify notification event if one exists and is small
+ * Get an fanotify notification event if one exists and is small
  * enough to fit in "count". Return an error pointer if the count
  * is not large enough. When permission event is dequeued, its state is
  * updated accordingly.
  */
-static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
+static struct fanotify_event *get_one_event(struct fsnotify_group *group,
 					    size_t count)
 {
 	size_t event_size = FAN_EVENT_METADATA_LEN;
-	struct fsnotify_event *fsn_event = NULL;
+	struct fanotify_event *event = NULL;
 
 	pr_debug("%s: group=%p count=%zd\n", __func__, group, count);
 
@@ -85,26 +106,23 @@ static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
 	}
 
 	if (event_size > count) {
-		fsn_event = ERR_PTR(-EINVAL);
+		event = ERR_PTR(-EINVAL);
 		goto out;
 	}
-	fsn_event = fsnotify_remove_first_event(group);
-	if (fanotify_is_perm_event(FANOTIFY_E(fsn_event)->mask))
-		FANOTIFY_PE(fsn_event)->state = FAN_EVENT_REPORTED;
+	event = FANOTIFY_E(fsnotify_remove_first_event(group));
+	if (fanotify_is_perm_event(event->mask))
+		FANOTIFY_PERM(event)->state = FAN_EVENT_REPORTED;
 out:
 	spin_unlock(&group->notification_lock);
-	return fsn_event;
+	return event;
 }
 
-static int create_fd(struct fsnotify_group *group,
-		     struct fanotify_event *event,
+static int create_fd(struct fsnotify_group *group, struct path *path,
 		     struct file **file)
 {
 	int client_fd;
 	struct file *new_file;
 
-	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
-
 	client_fd = get_unused_fd_flags(group->fanotify_data.f_flags);
 	if (client_fd < 0)
 		return client_fd;
@@ -113,14 +131,9 @@ static int create_fd(struct fsnotify_group *group,
 	 * we need a new file handle for the userspace program so it can read even if it was
 	 * originally opened O_WRONLY.
 	 */
-	/* it's possible this event was an overflow event.  in that case dentry and mnt
-	 * are NULL;  That's fine, just don't call dentry open */
-	if (event->path.dentry && event->path.mnt)
-		new_file = dentry_open(&event->path,
-				       group->fanotify_data.f_flags | FMODE_NONOTIFY,
-				       current_cred());
-	else
-		new_file = ERR_PTR(-EOVERFLOW);
+	new_file = dentry_open(path,
+			       group->fanotify_data.f_flags | FMODE_NONOTIFY,
+			       current_cred());
 	if (IS_ERR(new_file)) {
 		/*
 		 * we still send an event even if we can't open the file.  this
@@ -204,83 +217,111 @@ static int process_access_response(struct fsnotify_group *group,
 	return -ENOENT;
 }
 
-static int copy_fid_to_user(struct fanotify_event *event, char __user *buf)
+static int copy_info_to_user(__kernel_fsid_t *fsid, struct fanotify_fh *fh,
+			     const char *name, size_t name_len,
+			     char __user *buf, size_t count)
 {
 	struct fanotify_event_info_fid info = { };
 	struct file_handle handle = { };
-	unsigned char bounce[FANOTIFY_INLINE_FH_LEN], *fh;
-	size_t fh_len = event->fh_len;
-	size_t len = fanotify_event_info_len(event);
+	unsigned char bounce[FANOTIFY_INLINE_FH_LEN], *fh_buf;
+	size_t fh_len = fh ? fh->len : 0;
+	size_t info_len = fanotify_fid_info_len(fh_len, name_len);
+	size_t len = info_len;
+
+	pr_debug("%s: fh_len=%zu name_len=%zu, info_len=%zu, count=%zu\n",
+		 __func__, fh_len, name_len, info_len, count);
 
-	if (!len)
+	if (!fh_len || (name && !name_len))
 		return 0;
 
-	if (WARN_ON_ONCE(len < sizeof(info) + sizeof(handle) + fh_len))
+	if (WARN_ON_ONCE(len < sizeof(info) || len > count))
 		return -EFAULT;
 
-	/* Copy event info fid header followed by vaiable sized file handle */
-	info.hdr.info_type = FAN_EVENT_INFO_TYPE_FID;
+	/*
+	 * Copy event info fid header followed by variable sized file handle
+	 * and optionally followed by variable sized filename.
+	 */
+	info.hdr.info_type = name_len ? FAN_EVENT_INFO_TYPE_DFID_NAME :
+					FAN_EVENT_INFO_TYPE_FID;
 	info.hdr.len = len;
-	info.fsid = event->fid.fsid;
+	info.fsid = *fsid;
 	if (copy_to_user(buf, &info, sizeof(info)))
 		return -EFAULT;
 
 	buf += sizeof(info);
 	len -= sizeof(info);
-	handle.handle_type = event->fh_type;
+	if (WARN_ON_ONCE(len < sizeof(handle)))
+		return -EFAULT;
+
+	handle.handle_type = fh->type;
 	handle.handle_bytes = fh_len;
 	if (copy_to_user(buf, &handle, sizeof(handle)))
 		return -EFAULT;
 
 	buf += sizeof(handle);
 	len -= sizeof(handle);
+	if (WARN_ON_ONCE(len < fh_len))
+		return -EFAULT;
+
 	/*
-	 * For an inline fh, copy through stack to exclude the copy from
-	 * usercopy hardening protections.
+	 * For an inline fh and inline file name, copy through stack to exclude
+	 * the copy from usercopy hardening protections.
 	 */
-	fh = fanotify_event_fh(event);
+	fh_buf = fanotify_fh_buf(fh);
 	if (fh_len <= FANOTIFY_INLINE_FH_LEN) {
-		memcpy(bounce, fh, fh_len);
-		fh = bounce;
+		memcpy(bounce, fh_buf, fh_len);
+		fh_buf = bounce;
 	}
-	if (copy_to_user(buf, fh, fh_len))
+	if (copy_to_user(buf, fh_buf, fh_len))
 		return -EFAULT;
 
-	/* Pad with 0's */
 	buf += fh_len;
 	len -= fh_len;
+
+	if (name_len) {
+		/* Copy the filename with terminating null */
+		name_len++;
+		if (WARN_ON_ONCE(len < name_len))
+			return -EFAULT;
+
+		if (copy_to_user(buf, name, name_len))
+			return -EFAULT;
+
+		buf += name_len;
+		len -= name_len;
+	}
+
+	/* Pad with 0's */
 	WARN_ON_ONCE(len < 0 || len >= FANOTIFY_EVENT_ALIGN);
 	if (len > 0 && clear_user(buf, len))
 		return -EFAULT;
 
-	return 0;
+	return info_len;
 }
 
 static ssize_t copy_event_to_user(struct fsnotify_group *group,
-				  struct fsnotify_event *fsn_event,
+				  struct fanotify_event *event,
 				  char __user *buf, size_t count)
 {
 	struct fanotify_event_metadata metadata;
-	struct fanotify_event *event;
+	struct path *path = fanotify_event_path(event);
 	struct file *f = NULL;
 	int ret, fd = FAN_NOFD;
 
-	pr_debug("%s: group=%p event=%p\n", __func__, group, fsn_event);
+	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
 
-	event = container_of(fsn_event, struct fanotify_event, fse);
-	metadata.event_len = FAN_EVENT_METADATA_LEN;
+	metadata.event_len = FAN_EVENT_METADATA_LEN +
+					fanotify_event_info_len(event);
 	metadata.metadata_len = FAN_EVENT_METADATA_LEN;
 	metadata.vers = FANOTIFY_METADATA_VERSION;
 	metadata.reserved = 0;
 	metadata.mask = event->mask & FANOTIFY_OUTGOING_EVENTS;
 	metadata.pid = pid_vnr(event->pid);
 
-	if (fanotify_event_has_path(event)) {
-		fd = create_fd(group, event, &f);
+	if (path && path->mnt && path->dentry) {
+		fd = create_fd(group, path, &f);
 		if (fd < 0)
 			return fd;
-	} else if (fanotify_event_has_fid(event)) {
-		metadata.event_len += fanotify_event_info_len(event);
 	}
 	metadata.fd = fd;
 
@@ -295,15 +336,39 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
 	if (copy_to_user(buf, &metadata, FAN_EVENT_METADATA_LEN))
 		goto out_close_fd;
 
+	buf += FAN_EVENT_METADATA_LEN;
+	count -= FAN_EVENT_METADATA_LEN;
+
 	if (fanotify_is_perm_event(event->mask))
-		FANOTIFY_PE(fsn_event)->fd = fd;
+		FANOTIFY_PERM(event)->fd = fd;
 
-	if (fanotify_event_has_path(event)) {
+	if (f)
 		fd_install(fd, f);
-	} else if (fanotify_event_has_fid(event)) {
-		ret = copy_fid_to_user(event, buf + FAN_EVENT_METADATA_LEN);
+
+	/* Event info records order is: dir fid + name, child fid */
+	if (fanotify_event_name_len(event)) {
+		struct fanotify_name_event *fne = FANOTIFY_NE(event);
+
+		ret = copy_info_to_user(fanotify_event_fsid(event),
+					fanotify_event_dir_fh(event),
+					fne->name, fne->name_len,
+					buf, count);
+		if (ret < 0)
+			return ret;
+
+		buf += ret;
+		count -= ret;
+	}
+
+	if (fanotify_event_object_fh_len(event)) {
+		ret = copy_info_to_user(fanotify_event_fsid(event),
+					fanotify_event_object_fh(event),
+					NULL, 0, buf, count);
 		if (ret < 0)
 			return ret;
+
+		buf += ret;
+		count -= ret;
 	}
 
 	return metadata.event_len;
@@ -335,7 +400,7 @@ static ssize_t fanotify_read(struct file *file, char __user *buf,
 			     size_t count, loff_t *pos)
 {
 	struct fsnotify_group *group;
-	struct fsnotify_event *kevent;
+	struct fanotify_event *event;
 	char __user *start;
 	int ret;
 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
@@ -347,13 +412,13 @@ static ssize_t fanotify_read(struct file *file, char __user *buf,
 
 	add_wait_queue(&group->notification_waitq, &wait);
 	while (1) {
-		kevent = get_one_event(group, count);
-		if (IS_ERR(kevent)) {
-			ret = PTR_ERR(kevent);
+		event = get_one_event(group, count);
+		if (IS_ERR(event)) {
+			ret = PTR_ERR(event);
 			break;
 		}
 
-		if (!kevent) {
+		if (!event) {
 			ret = -EAGAIN;
 			if (file->f_flags & O_NONBLOCK)
 				break;
@@ -369,7 +434,7 @@ static ssize_t fanotify_read(struct file *file, char __user *buf,
 			continue;
 		}
 
-		ret = copy_event_to_user(group, kevent, buf, count);
+		ret = copy_event_to_user(group, event, buf, count);
 		if (unlikely(ret == -EOPENSTALE)) {
 			/*
 			 * We cannot report events with stale fd so drop it.
@@ -384,17 +449,17 @@ static ssize_t fanotify_read(struct file *file, char __user *buf,
 		 * Permission events get queued to wait for response.  Other
 		 * events can be destroyed now.
 		 */
-		if (!fanotify_is_perm_event(FANOTIFY_E(kevent)->mask)) {
-			fsnotify_destroy_event(group, kevent);
+		if (!fanotify_is_perm_event(event->mask)) {
+			fsnotify_destroy_event(group, &event->fse);
 		} else {
 			if (ret <= 0) {
 				spin_lock(&group->notification_lock);
 				finish_permission_event(group,
-					FANOTIFY_PE(kevent), FAN_DENY);
+					FANOTIFY_PERM(event), FAN_DENY);
 				wake_up(&group->fanotify_data.access_waitq);
 			} else {
 				spin_lock(&group->notification_lock);
-				list_add_tail(&kevent->list,
+				list_add_tail(&event->fse.list,
 					&group->fanotify_data.access_list);
 				spin_unlock(&group->notification_lock);
 			}
@@ -440,8 +505,6 @@ static ssize_t fanotify_write(struct file *file, const char __user *buf, size_t
 static int fanotify_release(struct inode *ignored, struct file *file)
 {
 	struct fsnotify_group *group = file->private_data;
-	struct fanotify_perm_event *event;
-	struct fsnotify_event *fsn_event;
 
 	/*
 	 * Stop new events from arriving in the notification queue. since
@@ -456,6 +519,8 @@ static int fanotify_release(struct inode *ignored, struct file *file)
 	 */
 	spin_lock(&group->notification_lock);
 	while (!list_empty(&group->fanotify_data.access_list)) {
+		struct fanotify_perm_event *event;
+
 		event = list_first_entry(&group->fanotify_data.access_list,
 				struct fanotify_perm_event, fae.fse.list);
 		list_del_init(&event->fae.fse.list);
@@ -469,12 +534,14 @@ static int fanotify_release(struct inode *ignored, struct file *file)
 	 * response is consumed and fanotify_get_response() returns.
 	 */
 	while (!fsnotify_notify_queue_is_empty(group)) {
-		fsn_event = fsnotify_remove_first_event(group);
-		if (!(FANOTIFY_E(fsn_event)->mask & FANOTIFY_PERM_EVENTS)) {
+		struct fanotify_event *event;
+
+		event = FANOTIFY_E(fsnotify_remove_first_event(group));
+		if (!(event->mask & FANOTIFY_PERM_EVENTS)) {
 			spin_unlock(&group->notification_lock);
-			fsnotify_destroy_event(group, fsn_event);
+			fsnotify_destroy_event(group, &event->fse);
 		} else {
-			finish_permission_event(group, FANOTIFY_PE(fsn_event),
+			finish_permission_event(group, FANOTIFY_PERM(event),
 						FAN_ALLOW);
 		}
 		spin_lock(&group->notification_lock);
@@ -523,7 +590,7 @@ static const struct file_operations fanotify_fops = {
 	.fasync		= NULL,
 	.release	= fanotify_release,
 	.unlocked_ioctl	= fanotify_ioctl,
-	.compat_ioctl	= fanotify_ioctl,
+	.compat_ioctl	= compat_ptr_ioctl,
 	.llseek		= noop_llseek,
 };
 
@@ -824,7 +891,7 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 	group->memcg = get_mem_cgroup_from_mm(current->mm);
 
 	oevent = fanotify_alloc_event(group, NULL, FS_Q_OVERFLOW, NULL,
-				      FSNOTIFY_EVENT_NONE, NULL);
+				      FSNOTIFY_EVENT_NONE, NULL, NULL);
 	if (unlikely(!oevent)) {
 		fd = -ENOMEM;
 		goto out_destroy_group;
@@ -1139,7 +1206,10 @@ static int __init fanotify_user_setup(void)
 
 	fanotify_mark_cache = KMEM_CACHE(fsnotify_mark,
 					 SLAB_PANIC|SLAB_ACCOUNT);
-	fanotify_event_cachep = KMEM_CACHE(fanotify_event, SLAB_PANIC);
+	fanotify_fid_event_cachep = KMEM_CACHE(fanotify_fid_event,
+					       SLAB_PANIC);
+	fanotify_path_event_cachep = KMEM_CACHE(fanotify_path_event,
+						SLAB_PANIC);
 	if (IS_ENABLED(CONFIG_FANOTIFY_ACCESS_PERMISSIONS)) {
 		fanotify_perm_event_cachep =
 			KMEM_CACHE(fanotify_perm_event, SLAB_PANIC);
diff --git a/fs/notify/fdinfo.c b/fs/notify/fdinfo.c
index 1e2bfd26b352..ef83f4020554 100644
--- a/fs/notify/fdinfo.c
+++ b/fs/notify/fdinfo.c
@@ -50,7 +50,7 @@ static void show_mark_fhandle(struct seq_file *m, struct inode *inode)
 	f.handle.handle_bytes = sizeof(f.pad);
 	size = f.handle.handle_bytes >> 2;
 
-	ret = exportfs_encode_inode_fh(inode, (struct fid *)f.handle.f_handle, &size, 0);
+	ret = exportfs_encode_inode_fh(inode, (struct fid *)f.handle.f_handle, &size, NULL);
 	if ((ret == FILEID_INVALID) || (ret < 0)) {
 		WARN_ONCE(1, "Can't encode file handler for inotify: %d\n", ret);
 		return;
diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
index 2ecef6155fc0..72d332ce8e12 100644
--- a/fs/notify/fsnotify.c
+++ b/fs/notify/fsnotify.c
@@ -57,6 +57,9 @@ static void fsnotify_unmount_inodes(struct super_block *sb)
 		 * doing an __iget/iput with SB_ACTIVE clear would actually
 		 * evict all inodes with zero i_count from icache which is
 		 * unnecessarily violent and may in fact be illegal to do.
+		 * However, we should have been called /after/ evict_inodes
+		 * removed all zero refcount inodes, in any case.  Test to
+		 * be sure.
 		 */
 		if (!atomic_read(&inode->i_count)) {
 			spin_unlock(&inode->i_lock);
@@ -77,6 +80,7 @@ static void fsnotify_unmount_inodes(struct super_block *sb)
 
 		iput_inode = inode;
 
+		cond_resched();
 		spin_lock(&sb->s_inode_list_lock);
 	}
 	spin_unlock(&sb->s_inode_list_lock);
@@ -139,15 +143,13 @@ void __fsnotify_update_child_dentry_flags(struct inode *inode)
 }
 
 /* Notify this dentry's parent about a child's events. */
-int __fsnotify_parent(const struct path *path, struct dentry *dentry, __u32 mask)
+int fsnotify_parent(struct dentry *dentry, __u32 mask, const void *data,
+		    int data_type)
 {
 	struct dentry *parent;
 	struct inode *p_inode;
 	int ret = 0;
 
-	if (!dentry)
-		dentry = path->dentry;
-
 	if (!(dentry->d_flags & DCACHE_FSNOTIFY_PARENT_WATCHED))
 		return 0;
 
@@ -164,12 +166,7 @@ int __fsnotify_parent(const struct path *path, struct dentry *dentry, __u32 mask
 		mask |= FS_EVENT_ON_CHILD;
 
 		take_dentry_name_snapshot(&name, dentry);
-		if (path)
-			ret = fsnotify(p_inode, mask, path, FSNOTIFY_EVENT_PATH,
-				       &name.name, 0);
-		else
-			ret = fsnotify(p_inode, mask, dentry->d_inode, FSNOTIFY_EVENT_INODE,
-				       &name.name, 0);
+		ret = fsnotify(p_inode, mask, data, data_type, &name.name, 0);
 		release_dentry_name_snapshot(&name);
 	}
 
@@ -177,7 +174,7 @@ int __fsnotify_parent(const struct path *path, struct dentry *dentry, __u32 mask
 
 	return ret;
 }
-EXPORT_SYMBOL_GPL(__fsnotify_parent);
+EXPORT_SYMBOL_GPL(fsnotify_parent);
 
 static int send_to_group(struct inode *to_tell,
 			 __u32 mask, const void *data,
@@ -314,6 +311,7 @@ static void fsnotify_iter_next(struct fsnotify_iter_info *iter_info)
 int fsnotify(struct inode *to_tell, __u32 mask, const void *data, int data_is,
 	     const struct qstr *file_name, u32 cookie)
 {
+	const struct path *path = fsnotify_data_path(data, data_is);
 	struct fsnotify_iter_info iter_info = {};
 	struct super_block *sb = to_tell->i_sb;
 	struct mount *mnt = NULL;
@@ -321,8 +319,8 @@ int fsnotify(struct inode *to_tell, __u32 mask, const void *data, int data_is,
 	int ret = 0;
 	__u32 test_mask = (mask & ALL_FSNOTIFY_EVENTS);
 
-	if (data_is == FSNOTIFY_EVENT_PATH) {
-		mnt = real_mount(((const struct path *)data)->mnt);
+	if (path) {
+		mnt = real_mount(path->mnt);
 		mnt_or_sb_mask |= mnt->mnt_fsnotify_mask;
 	}
 	/* An event "on child" is not intended for a mount/sb mark */
@@ -381,13 +379,11 @@ out:
 }
 EXPORT_SYMBOL_GPL(fsnotify);
 
-extern struct kmem_cache *fsnotify_mark_connector_cachep;
-
 static __init int fsnotify_init(void)
 {
 	int ret;
 
-	BUILD_BUG_ON(HWEIGHT32(ALL_FSNOTIFY_BITS) != 25);
+	BUILD_BUG_ON(HWEIGHT32(ALL_FSNOTIFY_BITS) != 26);
 
 	ret = init_srcu_struct(&fsnotify_mark_srcu);
 	if (ret)
diff --git a/fs/notify/fsnotify.h b/fs/notify/fsnotify.h
index f3462828a0e2..ff2063ec6b0f 100644
--- a/fs/notify/fsnotify.h
+++ b/fs/notify/fsnotify.h
@@ -65,4 +65,6 @@ extern void __fsnotify_update_child_dentry_flags(struct inode *inode);
 extern struct fsnotify_event_holder *fsnotify_alloc_event_holder(void);
 extern void fsnotify_destroy_event_holder(struct fsnotify_event_holder *holder);
 
+extern struct kmem_cache *fsnotify_mark_connector_cachep;
+
 #endif	/* __FS_NOTIFY_FSNOTIFY_H_ */
diff --git a/fs/notify/inotify/Kconfig b/fs/notify/inotify/Kconfig
index 6736e47d94d8..7715fadd5fff 100644
--- a/fs/notify/inotify/Kconfig
+++ b/fs/notify/inotify/Kconfig
@@ -12,6 +12,6 @@ config INOTIFY_USER
 	  new features including multiple file events, one-shot support, and
 	  unmount notification.
 
-	  For more information, see <file:Documentation/filesystems/inotify.txt>
+	  For more information, see <file:Documentation/filesystems/inotify.rst>
 
 	  If unsure, say Y.
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
index d510223d302c..2ebc89047153 100644
--- a/fs/notify/inotify/inotify_fsnotify.c
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -39,7 +39,7 @@ static bool event_compare(struct fsnotify_event *old_fsn,
 	if (old->mask & FS_IN_IGNORED)
 		return false;
 	if ((old->mask == new->mask) &&
-	    (old_fsn->inode == new_fsn->inode) &&
+	    (old_fsn->objectid == new_fsn->objectid) &&
 	    (old->name_len == new->name_len) &&
 	    (!old->name_len || !strcmp(old->name, new->name)))
 		return true;
@@ -61,6 +61,7 @@ int inotify_handle_event(struct fsnotify_group *group,
 			 const struct qstr *file_name, u32 cookie,
 			 struct fsnotify_iter_info *iter_info)
 {
+	const struct path *path = fsnotify_data_path(data, data_type);
 	struct fsnotify_mark *inode_mark = fsnotify_iter_inode_mark(iter_info);
 	struct inotify_inode_mark *i_mark;
 	struct inotify_event_info *event;
@@ -73,12 +74,9 @@ int inotify_handle_event(struct fsnotify_group *group,
 		return 0;
 
 	if ((inode_mark->mask & FS_EXCL_UNLINK) &&
-	    (data_type == FSNOTIFY_EVENT_PATH)) {
-		const struct path *path = data;
+	    path && d_unlinked(path->dentry))
+		return 0;
 
-		if (d_unlinked(path->dentry))
-			return 0;
-	}
 	if (file_name) {
 		len = file_name->len;
 		alloc_len += len + 1;
@@ -118,7 +116,7 @@ int inotify_handle_event(struct fsnotify_group *group,
 		mask &= ~IN_ISDIR;
 
 	fsn_event = &event->fse;
-	fsnotify_init_event(fsn_event, inode);
+	fsnotify_init_event(fsn_event, (unsigned long)inode);
 	event->mask = mask;
 	event->wd = i_mark->wd;
 	event->sync_cookie = cookie;
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 107537a543fd..81ffc8629fc4 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -635,7 +635,7 @@ static struct fsnotify_group *inotify_new_group(unsigned int max_events)
 		return ERR_PTR(-ENOMEM);
 	}
 	group->overflow_event = &oevent->fse;
-	fsnotify_init_event(group->overflow_event, NULL);
+	fsnotify_init_event(group->overflow_event, 0);
 	oevent->mask = FS_Q_OVERFLOW;
 	oevent->wd = -1;
 	oevent->sync_cookie = 0;
diff --git a/fs/nsfs.c b/fs/nsfs.c
index a0431642c6b5..4f1205725cfe 100644
--- a/fs/nsfs.c
+++ b/fs/nsfs.c
@@ -3,6 +3,7 @@
 #include <linux/pseudo_fs.h>
 #include <linux/file.h>
 #include <linux/fs.h>
+#include <linux/proc_fs.h>
 #include <linux/proc_ns.h>
 #include <linux/magic.h>
 #include <linux/ktime.h>
@@ -11,6 +12,8 @@
 #include <linux/nsfs.h>
 #include <linux/uaccess.h>
 
+#include "internal.h"
+
 static struct vfsmount *nsfs_mnt;
 
 static long ns_ioctl(struct file *filp, unsigned int ioctl,
@@ -52,7 +55,7 @@ static void nsfs_evict(struct inode *inode)
 	ns->ops->put(ns);
 }
 
-static void *__ns_get_path(struct path *path, struct ns_common *ns)
+static int __ns_get_path(struct path *path, struct ns_common *ns)
 {
 	struct vfsmount *mnt = nsfs_mnt;
 	struct dentry *dentry;
@@ -71,13 +74,13 @@ static void *__ns_get_path(struct path *path, struct ns_common *ns)
 got_it:
 	path->mnt = mntget(mnt);
 	path->dentry = dentry;
-	return NULL;
+	return 0;
 slow:
 	rcu_read_unlock();
 	inode = new_inode_pseudo(mnt->mnt_sb);
 	if (!inode) {
 		ns->ops->put(ns);
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 	}
 	inode->i_ino = ns->inum;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
@@ -89,7 +92,7 @@ slow:
 	dentry = d_alloc_anon(mnt->mnt_sb);
 	if (!dentry) {
 		iput(inode);
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 	}
 	d_instantiate(dentry, inode);
 	dentry->d_fsdata = (void *)ns->ops;
@@ -98,23 +101,22 @@ slow:
 		d_delete(dentry);	/* make sure ->d_prune() does nothing */
 		dput(dentry);
 		cpu_relax();
-		return ERR_PTR(-EAGAIN);
+		return -EAGAIN;
 	}
 	goto got_it;
 }
 
-void *ns_get_path_cb(struct path *path, ns_get_path_helper_t *ns_get_cb,
+int ns_get_path_cb(struct path *path, ns_get_path_helper_t *ns_get_cb,
 		     void *private_data)
 {
-	void *ret;
+	int ret;
 
 	do {
 		struct ns_common *ns = ns_get_cb(private_data);
 		if (!ns)
-			return ERR_PTR(-ENOENT);
-
+			return -ENOENT;
 		ret = __ns_get_path(path, ns);
-	} while (ret == ERR_PTR(-EAGAIN));
+	} while (ret == -EAGAIN);
 
 	return ret;
 }
@@ -131,7 +133,7 @@ static struct ns_common *ns_get_path_task(void *private_data)
 	return args->ns_ops->get(args->task);
 }
 
-void *ns_get_path(struct path *path, struct task_struct *task,
+int ns_get_path(struct path *path, struct task_struct *task,
 		  const struct proc_ns_operations *ns_ops)
 {
 	struct ns_get_path_task_args args = {
@@ -147,7 +149,7 @@ int open_related_ns(struct ns_common *ns,
 {
 	struct path path = {};
 	struct file *f;
-	void *err;
+	int err;
 	int fd;
 
 	fd = get_unused_fd_flags(O_CLOEXEC);
@@ -164,11 +166,11 @@ int open_related_ns(struct ns_common *ns,
 		}
 
 		err = __ns_get_path(&path, relative);
-	} while (err == ERR_PTR(-EAGAIN));
+	} while (err == -EAGAIN);
 
-	if (IS_ERR(err)) {
+	if (err) {
 		put_unused_fd(fd);
-		return PTR_ERR(err);
+		return err;
 	}
 
 	f = dentry_open(&path, O_RDONLY, current_cred());
@@ -245,6 +247,20 @@ out_invalid:
 	return ERR_PTR(-EINVAL);
 }
 
+/**
+ * ns_match() - Returns true if current namespace matches dev/ino provided.
+ * @ns_common: current ns
+ * @dev: dev_t from nsfs that will be matched against current nsfs
+ * @ino: ino_t from nsfs that will be matched against current nsfs
+ *
+ * Return: true if dev and ino matches the current nsfs.
+ */
+bool ns_match(const struct ns_common *ns, dev_t dev, ino_t ino)
+{
+	return (ns->inum == ino) && (nsfs_mnt->mnt_sb->s_dev == dev);
+}
+
+
 static int nsfs_show_path(struct seq_file *seq, struct dentry *dentry)
 {
 	struct inode *inode = d_inode(dentry);
diff --git a/fs/ntfs/Kconfig b/fs/ntfs/Kconfig
index de9fb5cff226..1667a7e590d8 100644
--- a/fs/ntfs/Kconfig
+++ b/fs/ntfs/Kconfig
@@ -18,7 +18,7 @@ config NTFS_FS
 	  the Linux 2.4 kernel series is separately available as a patch
 	  from the project web site.
 
-	  For more information see <file:Documentation/filesystems/ntfs.txt>
+	  For more information see <file:Documentation/filesystems/ntfs.rst>
 	  and <http://www.linux-ntfs.org/>.
 
 	  To compile this file system support as a module, choose M here: the
diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
index 7202a1e39d70..bb0a43860ad2 100644
--- a/fs/ntfs/aops.c
+++ b/fs/ntfs/aops.c
@@ -92,8 +92,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
 				"0x%llx.", (unsigned long long)bh->b_blocknr);
 	}
 	first = page_buffers(page);
-	local_irq_save(flags);
-	bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
+	spin_lock_irqsave(&first->b_uptodate_lock, flags);
 	clear_buffer_async_read(bh);
 	unlock_buffer(bh);
 	tmp = bh;
@@ -108,8 +107,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
 		}
 		tmp = tmp->b_this_page;
 	} while (tmp != bh);
-	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
 	/*
 	 * If none of the buffers had errors then we can set the page uptodate,
 	 * but we first have to perform the post read mst fixups, if the
@@ -142,8 +140,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
 	unlock_page(page);
 	return;
 still_busy:
-	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&first->b_uptodate_lock, flags);
 	return;
 }
 
@@ -1735,7 +1732,7 @@ void mark_ntfs_record_dirty(struct page *page, const unsigned int ofs) {
 				bh = bh->b_this_page;
 			} while (bh);
 			tail->b_this_page = head;
-			attach_page_buffers(page, head);
+			attach_page_private(page, head);
 		} else
 			buffers_to_free = bh;
 	}
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 6c7388430ad3..d4359a1df3d5 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -2899,18 +2899,12 @@ int ntfs_setattr(struct dentry *dentry, struct iattr *attr)
 			ia_valid |= ATTR_MTIME | ATTR_CTIME;
 		}
 	}
-	if (ia_valid & ATTR_ATIME) {
-		vi->i_atime = timestamp_truncate(attr->ia_atime,
-					       vi);
-	}
-	if (ia_valid & ATTR_MTIME) {
-		vi->i_mtime = timestamp_truncate(attr->ia_mtime,
-					       vi);
-	}
-	if (ia_valid & ATTR_CTIME) {
-		vi->i_ctime = timestamp_truncate(attr->ia_ctime,
-					       vi);
-	}
+	if (ia_valid & ATTR_ATIME)
+		vi->i_atime = attr->ia_atime;
+	if (ia_valid & ATTR_MTIME)
+		vi->i_mtime = attr->ia_mtime;
+	if (ia_valid & ATTR_CTIME)
+		vi->i_ctime = attr->ia_ctime;
 	mark_inode_dirty(vi);
 out:
 	return err;
diff --git a/fs/ntfs/malloc.h b/fs/ntfs/malloc.h
index 842b0bfc3ac9..7068425735f1 100644
--- a/fs/ntfs/malloc.h
+++ b/fs/ntfs/malloc.h
@@ -34,7 +34,7 @@ static inline void *__ntfs_malloc(unsigned long size, gfp_t gfp_mask)
 		/* return (void *)__get_free_page(gfp_mask); */
 	}
 	if (likely((size >> PAGE_SHIFT) < totalram_pages()))
-		return __vmalloc(size, gfp_mask, PAGE_KERNEL);
+		return __vmalloc(size, gfp_mask);
 	return NULL;
 }
 
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index 3aac5c917afe..fbb9f1bc623d 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -504,7 +504,7 @@ int ntfs_sync_mft_mirror(ntfs_volume *vol, const unsigned long mft_no,
 			bh = bh->b_this_page;
 		} while (bh);
 		tail->b_this_page = head;
-		attach_page_buffers(page, head);
+		attach_page_private(page, head);
 	}
 	bh = head = page_buffers(page);
 	BUG_ON(!bh);
diff --git a/fs/ocfs2/Kconfig b/fs/ocfs2/Kconfig
index 46bba20da6b5..1177c33df895 100644
--- a/fs/ocfs2/Kconfig
+++ b/fs/ocfs2/Kconfig
@@ -21,7 +21,7 @@ config OCFS2_FS
 	  OCFS2 mailing lists: http://oss.oracle.com/projects/ocfs2/mailman/
 
 	  For more information on OCFS2, see the file
-	  <file:Documentation/filesystems/ocfs2.txt>.
+	  <file:Documentation/filesystems/ocfs2.rst>.
 
 config OCFS2_FS_O2CB
 	tristate "O2CB Kernelspace Clustering"
diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c
index 3e7da392aa6f..bb981ec76456 100644
--- a/fs/ocfs2/acl.c
+++ b/fs/ocfs2/acl.c
@@ -327,8 +327,8 @@ int ocfs2_acl_chmod(struct inode *inode, struct buffer_head *bh)
 	down_read(&OCFS2_I(inode)->ip_xattr_sem);
 	acl = ocfs2_get_acl_nolock(inode, ACL_TYPE_ACCESS, bh);
 	up_read(&OCFS2_I(inode)->ip_xattr_sem);
-	if (IS_ERR(acl) || !acl)
-		return PTR_ERR(acl);
+	if (IS_ERR_OR_NULL(acl))
+		return PTR_ERR_OR_ZERO(acl);
 	ret = __posix_acl_chmod(&acl, GFP_KERNEL, inode->i_mode);
 	if (ret)
 		return ret;
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index f9baefc76cf9..2f834add165b 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -1060,7 +1060,6 @@ bail:
 			brelse(bhs[i]);
 			bhs[i] = NULL;
 		}
-		mlog_errno(status);
 	}
 	return status;
 }
@@ -2288,9 +2287,9 @@ static int ocfs2_extend_rotate_transaction(handle_t *handle, int subtree_depth,
 	int ret = 0;
 	int credits = (path->p_tree_depth - subtree_depth) * 2 + 1 + op_credits;
 
-	if (handle->h_buffer_credits < credits)
+	if (jbd2_handle_buffer_credits(handle) < credits)
 		ret = ocfs2_extend_trans(handle,
-					 credits - handle->h_buffer_credits);
+				credits - jbd2_handle_buffer_credits(handle));
 
 	return ret;
 }
@@ -2367,7 +2366,7 @@ static int ocfs2_rotate_tree_right(handle_t *handle,
 				   struct ocfs2_path *right_path,
 				   struct ocfs2_path **ret_left_path)
 {
-	int ret, start, orig_credits = handle->h_buffer_credits;
+	int ret, start, orig_credits = jbd2_handle_buffer_credits(handle);
 	u32 cpos;
 	struct ocfs2_path *left_path = NULL;
 	struct super_block *sb = ocfs2_metadata_cache_get_super(et->et_ci);
@@ -3148,7 +3147,7 @@ static int ocfs2_rotate_tree_left(handle_t *handle,
 				  struct ocfs2_path *path,
 				  struct ocfs2_cached_dealloc_ctxt *dealloc)
 {
-	int ret, orig_credits = handle->h_buffer_credits;
+	int ret, orig_credits = jbd2_handle_buffer_credits(handle);
 	struct ocfs2_path *tmp_path = NULL, *restart_path = NULL;
 	struct ocfs2_extent_block *eb;
 	struct ocfs2_extent_list *el;
@@ -3386,8 +3385,8 @@ static int ocfs2_merge_rec_right(struct ocfs2_path *left_path,
 							right_path);
 
 		ret = ocfs2_extend_rotate_transaction(handle, subtree_index,
-						      handle->h_buffer_credits,
-						      right_path);
+					jbd2_handle_buffer_credits(handle),
+					right_path);
 		if (ret) {
 			mlog_errno(ret);
 			goto out;
@@ -3548,8 +3547,8 @@ static int ocfs2_merge_rec_left(struct ocfs2_path *right_path,
 							right_path);
 
 		ret = ocfs2_extend_rotate_transaction(handle, subtree_index,
-						      handle->h_buffer_credits,
-						      left_path);
+					jbd2_handle_buffer_credits(handle),
+					left_path);
 		if (ret) {
 			mlog_errno(ret);
 			goto out;
@@ -3623,7 +3622,7 @@ static int ocfs2_merge_rec_left(struct ocfs2_path *right_path,
 		    le16_to_cpu(el->l_next_free_rec) == 1) {
 			/* extend credit for ocfs2_remove_rightmost_path */
 			ret = ocfs2_extend_rotate_transaction(handle, 0,
-					handle->h_buffer_credits,
+					jbd2_handle_buffer_credits(handle),
 					right_path);
 			if (ret) {
 				mlog_errno(ret);
@@ -3669,7 +3668,7 @@ static int ocfs2_try_to_merge_extent(handle_t *handle,
 	if (ctxt->c_split_covers_rec && ctxt->c_has_empty_extent) {
 		/* extend credit for ocfs2_remove_rightmost_path */
 		ret = ocfs2_extend_rotate_transaction(handle, 0,
-				handle->h_buffer_credits,
+				jbd2_handle_buffer_credits(handle),
 				path);
 		if (ret) {
 			mlog_errno(ret);
@@ -3725,7 +3724,7 @@ static int ocfs2_try_to_merge_extent(handle_t *handle,
 
 		/* extend credit for ocfs2_remove_rightmost_path */
 		ret = ocfs2_extend_rotate_transaction(handle, 0,
-					handle->h_buffer_credits,
+					jbd2_handle_buffer_credits(handle),
 					path);
 		if (ret) {
 			mlog_errno(ret);
@@ -3755,7 +3754,7 @@ static int ocfs2_try_to_merge_extent(handle_t *handle,
 
 		/* extend credit for ocfs2_remove_rightmost_path */
 		ret = ocfs2_extend_rotate_transaction(handle, 0,
-				handle->h_buffer_credits,
+				jbd2_handle_buffer_credits(handle),
 				path);
 		if (ret) {
 			mlog_errno(ret);
@@ -3799,7 +3798,7 @@ static int ocfs2_try_to_merge_extent(handle_t *handle,
 		if (ctxt->c_split_covers_rec) {
 			/* extend credit for ocfs2_remove_rightmost_path */
 			ret = ocfs2_extend_rotate_transaction(handle, 0,
-					handle->h_buffer_credits,
+					jbd2_handle_buffer_credits(handle),
 					path);
 			if (ret) {
 				mlog_errno(ret);
@@ -3942,7 +3941,7 @@ rotate:
 	 * above.
 	 *
 	 * This leaf needs to have space, either by the empty 1st
-	 * extent record, or by virtue of an l_next_rec < l_count.
+	 * extent record, or by virtue of an l_next_free_rec < l_count.
 	 */
 	ocfs2_rotate_leaf(el, insert_rec);
 }
@@ -5358,7 +5357,7 @@ static int ocfs2_truncate_rec(handle_t *handle,
 	if (ocfs2_is_empty_extent(&el->l_recs[0]) && index > 0) {
 		/* extend credit for ocfs2_remove_rightmost_path */
 		ret = ocfs2_extend_rotate_transaction(handle, 0,
-				handle->h_buffer_credits,
+				jbd2_handle_buffer_credits(handle),
 				path);
 		if (ret) {
 			mlog_errno(ret);
@@ -5427,8 +5426,8 @@ static int ocfs2_truncate_rec(handle_t *handle,
 	}
 
 	ret = ocfs2_extend_rotate_transaction(handle, 0,
-					      handle->h_buffer_credits,
-					      path);
+					jbd2_handle_buffer_credits(handle),
+					path);
 	if (ret) {
 		mlog_errno(ret);
 		goto out;
@@ -7403,6 +7402,10 @@ int ocfs2_truncate_inline(struct inode *inode, struct buffer_head *di_bh,
 	struct ocfs2_dinode *di = (struct ocfs2_dinode *)di_bh->b_data;
 	struct ocfs2_inline_data *idata = &di->id2.i_data;
 
+	/* No need to punch hole beyond i_size. */
+	if (start >= i_size_read(inode))
+		return 0;
+
 	if (end > i_size_read(inode))
 		end = i_size_read(inode);
 
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 9cd0a6815933..3bfb4147895a 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -11,7 +11,6 @@
 #include <linux/pagemap.h>
 #include <asm/byteorder.h>
 #include <linux/swap.h>
-#include <linux/pipe_fs_i.h>
 #include <linux/mpage.h>
 #include <linux/quotaops.h>
 #include <linux/blkdev.h>
@@ -351,14 +350,11 @@ out:
  * grow out to a tree. If need be, detecting boundary extents could
  * trivially be added in a future version of ocfs2_get_block().
  */
-static int ocfs2_readpages(struct file *filp, struct address_space *mapping,
-			   struct list_head *pages, unsigned nr_pages)
+static void ocfs2_readahead(struct readahead_control *rac)
 {
-	int ret, err = -EIO;
-	struct inode *inode = mapping->host;
+	int ret;
+	struct inode *inode = rac->mapping->host;
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
-	loff_t start;
-	struct page *last;
 
 	/*
 	 * Use the nonblocking flag for the dlm code to avoid page
@@ -366,36 +362,31 @@ static int ocfs2_readpages(struct file *filp, struct address_space *mapping,
 	 */
 	ret = ocfs2_inode_lock_full(inode, NULL, 0, OCFS2_LOCK_NONBLOCK);
 	if (ret)
-		return err;
+		return;
 
-	if (down_read_trylock(&oi->ip_alloc_sem) == 0) {
-		ocfs2_inode_unlock(inode, 0);
-		return err;
-	}
+	if (down_read_trylock(&oi->ip_alloc_sem) == 0)
+		goto out_unlock;
 
 	/*
 	 * Don't bother with inline-data. There isn't anything
 	 * to read-ahead in that case anyway...
 	 */
 	if (oi->ip_dyn_features & OCFS2_INLINE_DATA_FL)
-		goto out_unlock;
+		goto out_up;
 
 	/*
 	 * Check whether a remote node truncated this file - we just
 	 * drop out in that case as it's not worth handling here.
 	 */
-	last = lru_to_page(pages);
-	start = (loff_t)last->index << PAGE_SHIFT;
-	if (start >= i_size_read(inode))
-		goto out_unlock;
+	if (readahead_pos(rac) >= i_size_read(inode))
+		goto out_up;
 
-	err = mpage_readpages(mapping, pages, nr_pages, ocfs2_get_block);
+	mpage_readahead(rac, ocfs2_get_block);
 
-out_unlock:
+out_up:
 	up_read(&oi->ip_alloc_sem);
+out_unlock:
 	ocfs2_inode_unlock(inode, 0);
-
-	return err;
 }
 
 /* Note: Because we don't support holes, our allocation has
@@ -2475,7 +2466,7 @@ static ssize_t ocfs2_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 
 const struct address_space_operations ocfs2_aops = {
 	.readpage		= ocfs2_readpage,
-	.readpages		= ocfs2_readpages,
+	.readahead		= ocfs2_readahead,
 	.writepage		= ocfs2_writepage,
 	.write_begin		= ocfs2_write_begin,
 	.write_end		= ocfs2_write_end,
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index a368350d4c27..89d13e0705fe 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -101,8 +101,6 @@ static struct o2hb_callback {
 
 static struct o2hb_callback *hbcall_from_type(enum o2hb_callback_type type);
 
-#define O2HB_DEFAULT_BLOCK_BITS       9
-
 enum o2hb_heartbeat_modes {
 	O2HB_HEARTBEAT_LOCAL		= 0,
 	O2HB_HEARTBEAT_GLOBAL,
@@ -1309,7 +1307,7 @@ static int o2hb_debug_open(struct inode *inode, struct file *file)
 
 	case O2HB_DB_TYPE_REGION_NUMBER:
 		reg = (struct o2hb_region *)db->db_data;
-		out += snprintf(buf + out, PAGE_SIZE - out, "%d\n",
+		out += scnprintf(buf + out, PAGE_SIZE - out, "%d\n",
 				reg->hr_region_num);
 		goto done;
 
@@ -1319,12 +1317,12 @@ static int o2hb_debug_open(struct inode *inode, struct file *file)
 		/* If 0, it has never been set before */
 		if (lts)
 			lts = jiffies_to_msecs(jiffies - lts);
-		out += snprintf(buf + out, PAGE_SIZE - out, "%lu\n", lts);
+		out += scnprintf(buf + out, PAGE_SIZE - out, "%lu\n", lts);
 		goto done;
 
 	case O2HB_DB_TYPE_REGION_PINNED:
 		reg = (struct o2hb_region *)db->db_data;
-		out += snprintf(buf + out, PAGE_SIZE - out, "%u\n",
+		out += scnprintf(buf + out, PAGE_SIZE - out, "%u\n",
 				!!reg->hr_item_pinned);
 		goto done;
 
@@ -1333,8 +1331,8 @@ static int o2hb_debug_open(struct inode *inode, struct file *file)
 	}
 
 	while ((i = find_next_bit(map, db->db_len, i + 1)) < db->db_len)
-		out += snprintf(buf + out, PAGE_SIZE - out, "%d ", i);
-	out += snprintf(buf + out, PAGE_SIZE - out, "\n");
+		out += scnprintf(buf + out, PAGE_SIZE - out, "%d ", i);
+	out += scnprintf(buf + out, PAGE_SIZE - out, "\n");
 
 done:
 	i_size_write(inode, out);
diff --git a/fs/ocfs2/cluster/netdebug.c b/fs/ocfs2/cluster/netdebug.c
index 02bf4a1774cc..667a5c5e1f66 100644
--- a/fs/ocfs2/cluster/netdebug.c
+++ b/fs/ocfs2/cluster/netdebug.c
@@ -443,8 +443,8 @@ static int o2net_fill_bitmap(char *buf, int len)
 	o2net_fill_node_map(map, sizeof(map));
 
 	while ((i = find_next_bit(map, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES)
-		out += snprintf(buf + out, PAGE_SIZE - out, "%d ", i);
-	out += snprintf(buf + out, PAGE_SIZE - out, "\n");
+		out += scnprintf(buf + out, PAGE_SIZE - out, "%d ", i);
+	out += scnprintf(buf + out, PAGE_SIZE - out, "\n");
 
 	return out;
 }
diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c
index 5c424a099280..1ef24574f481 100644
--- a/fs/ocfs2/cluster/quorum.c
+++ b/fs/ocfs2/cluster/quorum.c
@@ -73,7 +73,7 @@ static void o2quo_fence_self(void)
 		       "system by restarting ***\n");
 		emergency_restart();
 		break;
-	};
+	}
 }
 
 /* Indicate that a timeout occurred on a heartbeat region write. The
diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c
index 48a3398f0bf5..2c512b40a940 100644
--- a/fs/ocfs2/cluster/tcp.c
+++ b/fs/ocfs2/cluster/tcp.c
@@ -1570,15 +1570,13 @@ static void o2net_start_connect(struct work_struct *work)
 	struct sockaddr_in myaddr = {0, }, remoteaddr = {0, };
 	int ret = 0, stop;
 	unsigned int timeout;
-	unsigned int noio_flag;
+	unsigned int nofs_flag;
 
 	/*
-	 * sock_create allocates the sock with GFP_KERNEL. We must set
-	 * per-process flag PF_MEMALLOC_NOIO so that all allocations done
-	 * by this process are done as if GFP_NOIO was specified. So we
-	 * are not reentering filesystem while doing memory reclaim.
+	 * sock_create allocates the sock with GFP_KERNEL. We must
+	 * prevent the filesystem from being reentered by memory reclaim.
 	 */
-	noio_flag = memalloc_noio_save();
+	nofs_flag = memalloc_nofs_save();
 	/* if we're greater we initiate tx, otherwise we accept */
 	if (o2nm_this_node() <= o2net_num_from_nn(nn))
 		goto out;
@@ -1683,7 +1681,7 @@ out:
 	if (mynode)
 		o2nm_node_put(mynode);
 
-	memalloc_noio_restore(noio_flag);
+	memalloc_nofs_restore(nofs_flag);
 	return;
 }
 
@@ -1810,15 +1808,13 @@ static int o2net_accept_one(struct socket *sock, int *more)
 	struct o2nm_node *local_node = NULL;
 	struct o2net_sock_container *sc = NULL;
 	struct o2net_node *nn;
-	unsigned int noio_flag;
+	unsigned int nofs_flag;
 
 	/*
-	 * sock_create_lite allocates the sock with GFP_KERNEL. We must set
-	 * per-process flag PF_MEMALLOC_NOIO so that all allocations done
-	 * by this process are done as if GFP_NOIO was specified. So we
-	 * are not reentering filesystem while doing memory reclaim.
+	 * sock_create_lite allocates the sock with GFP_KERNEL. We must
+	 * prevent the filesystem from being reentered by memory reclaim.
 	 */
-	noio_flag = memalloc_noio_save();
+	nofs_flag = memalloc_nofs_save();
 
 	BUG_ON(sock == NULL);
 	*more = 0;
@@ -1934,7 +1930,7 @@ out:
 	if (sc)
 		sc_put(sc);
 
-	memalloc_noio_restore(noio_flag);
+	memalloc_nofs_restore(nofs_flag);
 	return ret;
 }
 
@@ -1948,7 +1944,6 @@ static void o2net_accept_many(struct work_struct *work)
 {
 	struct socket *sock = o2net_listen_sock;
 	int	more;
-	int	err;
 
 	/*
 	 * It is critical to note that due to interrupt moderation
@@ -1963,7 +1958,7 @@ static void o2net_accept_many(struct work_struct *work)
 	 */
 
 	for (;;) {
-		err = o2net_accept_one(sock, &more);
+		o2net_accept_one(sock, &more);
 		if (!more)
 			break;
 		cond_resched();
diff --git a/fs/ocfs2/cluster/tcp.h b/fs/ocfs2/cluster/tcp.h
index de87cbffd175..736338f45c59 100644
--- a/fs/ocfs2/cluster/tcp.h
+++ b/fs/ocfs2/cluster/tcp.h
@@ -32,7 +32,7 @@ struct o2net_msg
 	__be32 status;
 	__be32 key;
 	__be32 msg_num;
-	__u8  buf[0];
+	__u8  buf[];
 };
 
 typedef int (o2net_msg_handler_func)(struct o2net_msg *msg, u32 len, void *data,
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index bdef72c0f099..5761060d2ba8 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -676,7 +676,7 @@ static struct buffer_head *ocfs2_find_entry_el(const char *name, int namelen,
 	int ra_ptr = 0;		/* Current index into readahead
 				   buffer */
 	int num = 0;
-	int nblocks, i, err;
+	int nblocks, i;
 
 	sb = dir->i_sb;
 
@@ -708,7 +708,7 @@ restart:
 				num++;
 
 				bh = NULL;
-				err = ocfs2_read_dir_block(dir, b++, &bh,
+				ocfs2_read_dir_block(dir, b++, &bh,
 							   OCFS2_BH_READAHEAD);
 				bh_use[ra_max] = bh;
 			}
diff --git a/fs/ocfs2/dlm/Makefile b/fs/ocfs2/dlm/Makefile
index 38b224372776..5e700b45d32d 100644
--- a/fs/ocfs2/dlm/Makefile
+++ b/fs/ocfs2/dlm/Makefile
@@ -1,6 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0-only
-ccflags-y := -I $(srctree)/$(src)/..
-
 obj-$(CONFIG_OCFS2_FS_O2CB) += ocfs2_dlm.o
 
 ocfs2_dlm-objs := dlmdomain.o dlmdebug.o dlmthread.o dlmrecovery.o \
diff --git a/fs/ocfs2/dlm/dlmast.c b/fs/ocfs2/dlm/dlmast.c
index 4de89af96abf..6abaded3ff6b 100644
--- a/fs/ocfs2/dlm/dlmast.c
+++ b/fs/ocfs2/dlm/dlmast.c
@@ -23,15 +23,15 @@
 #include <linux/spinlock.h>
 
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
 
 #define MLOG_MASK_PREFIX ML_DLM
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 static void dlm_update_lvb(struct dlm_ctxt *dlm, struct dlm_lock_resource *res,
 			   struct dlm_lock *lock);
diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h
index aaf24548b02a..c8a444622faa 100644
--- a/fs/ocfs2/dlm/dlmcommon.h
+++ b/fs/ocfs2/dlm/dlmcommon.h
@@ -564,7 +564,7 @@ struct dlm_migratable_lockres
 	// 48 bytes
 	u8 lvb[DLM_LVB_LEN];
 	// 112 bytes
-	struct dlm_migratable_lock ml[0];  // 16 bytes each, begins at byte 112
+	struct dlm_migratable_lock ml[];  // 16 bytes each, begins at byte 112
 };
 #define DLM_MIG_LOCKRES_MAX_LEN  \
 	(sizeof(struct dlm_migratable_lockres) + \
@@ -601,7 +601,7 @@ struct dlm_convert_lock
 
 	u8 name[O2NM_MAX_NAME_LEN];
 
-	s8 lvb[0];
+	s8 lvb[];
 };
 #define DLM_CONVERT_LOCK_MAX_LEN  (sizeof(struct dlm_convert_lock)+DLM_LVB_LEN)
 
@@ -616,7 +616,7 @@ struct dlm_unlock_lock
 
 	u8 name[O2NM_MAX_NAME_LEN];
 
-	s8 lvb[0];
+	s8 lvb[];
 };
 #define DLM_UNLOCK_LOCK_MAX_LEN  (sizeof(struct dlm_unlock_lock)+DLM_LVB_LEN)
 
@@ -632,7 +632,7 @@ struct dlm_proxy_ast
 
 	u8 name[O2NM_MAX_NAME_LEN];
 
-	s8 lvb[0];
+	s8 lvb[];
 };
 #define DLM_PROXY_AST_MAX_LEN  (sizeof(struct dlm_proxy_ast)+DLM_LVB_LEN)
 
@@ -688,10 +688,6 @@ struct dlm_begin_reco
 	__be32 pad2;
 };
 
-
-#define BITS_PER_BYTE 8
-#define BITS_TO_BYTES(bits) (((bits)+BITS_PER_BYTE-1)/BITS_PER_BYTE)
-
 struct dlm_query_join_request
 {
 	u8 node_idx;
diff --git a/fs/ocfs2/dlm/dlmconvert.c b/fs/ocfs2/dlm/dlmconvert.c
index 965f45dbe17b..6051edc33aef 100644
--- a/fs/ocfs2/dlm/dlmconvert.c
+++ b/fs/ocfs2/dlm/dlmconvert.c
@@ -23,9 +23,9 @@
 #include <linux/spinlock.h>
 
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
@@ -33,7 +33,7 @@
 #include "dlmconvert.h"
 
 #define MLOG_MASK_PREFIX ML_DLM
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 /* NOTE: __dlmconvert_master is the only function in here that
  * needs a spinlock held on entry (res->spinlock) and it is the
diff --git a/fs/ocfs2/dlm/dlmdebug.c b/fs/ocfs2/dlm/dlmdebug.c
index 4d0b452012b2..4b8b41d23e91 100644
--- a/fs/ocfs2/dlm/dlmdebug.c
+++ b/fs/ocfs2/dlm/dlmdebug.c
@@ -17,9 +17,9 @@
 #include <linux/debugfs.h>
 #include <linux/export.h>
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
@@ -27,7 +27,7 @@
 #include "dlmdebug.h"
 
 #define MLOG_MASK_PREFIX ML_DLM
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 static int stringify_lockname(const char *lockname, int locklen, char *buf,
 			      int len);
@@ -244,11 +244,11 @@ static int stringify_lockname(const char *lockname, int locklen, char *buf,
 		memcpy((__be64 *)&inode_blkno_be,
 		       (char *)&lockname[OCFS2_DENTRY_LOCK_INO_START],
 		       sizeof(__be64));
-		out += snprintf(buf + out, len - out, "%.*s%08x",
+		out += scnprintf(buf + out, len - out, "%.*s%08x",
 				OCFS2_DENTRY_LOCK_INO_START - 1, lockname,
 				(unsigned int)be64_to_cpu(inode_blkno_be));
 	} else
-		out += snprintf(buf + out, len - out, "%.*s",
+		out += scnprintf(buf + out, len - out, "%.*s",
 				locklen, lockname);
 	return out;
 }
@@ -260,7 +260,7 @@ static int stringify_nodemap(unsigned long *nodemap, int maxnodes,
 	int i = -1;
 
 	while ((i = find_next_bit(nodemap, maxnodes, i + 1)) < maxnodes)
-		out += snprintf(buf + out, len - out, "%d ", i);
+		out += scnprintf(buf + out, len - out, "%d ", i);
 
 	return out;
 }
@@ -278,34 +278,34 @@ static int dump_mle(struct dlm_master_list_entry *mle, char *buf, int len)
 		mle_type = "MIG";
 
 	out += stringify_lockname(mle->mname, mle->mnamelen, buf + out, len - out);
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"\t%3s\tmas=%3u\tnew=%3u\tevt=%1d\tuse=%1d\tref=%3d\n",
 			mle_type, mle->master, mle->new_master,
 			!list_empty(&mle->hb_events),
 			!!mle->inuse,
 			kref_read(&mle->mle_refs));
 
-	out += snprintf(buf + out, len - out, "Maybe=");
+	out += scnprintf(buf + out, len - out, "Maybe=");
 	out += stringify_nodemap(mle->maybe_map, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
-	out += snprintf(buf + out, len - out, "Vote=");
+	out += scnprintf(buf + out, len - out, "Vote=");
 	out += stringify_nodemap(mle->vote_map, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
-	out += snprintf(buf + out, len - out, "Response=");
+	out += scnprintf(buf + out, len - out, "Response=");
 	out += stringify_nodemap(mle->response_map, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
-	out += snprintf(buf + out, len - out, "Node=");
+	out += scnprintf(buf + out, len - out, "Node=");
 	out += stringify_nodemap(mle->node_map, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 	return out;
 }
@@ -353,7 +353,7 @@ static int debug_purgelist_print(struct dlm_ctxt *dlm, char *buf, int len)
 	int out = 0;
 	unsigned long total = 0;
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Dumping Purgelist for Domain: %s\n", dlm->name);
 
 	spin_lock(&dlm->spinlock);
@@ -365,13 +365,13 @@ static int debug_purgelist_print(struct dlm_ctxt *dlm, char *buf, int len)
 		out += stringify_lockname(res->lockname.name,
 					  res->lockname.len,
 					  buf + out, len - out);
-		out += snprintf(buf + out, len - out, "\t%ld\n",
+		out += scnprintf(buf + out, len - out, "\t%ld\n",
 				(jiffies - res->last_used)/HZ);
 		spin_unlock(&res->spinlock);
 	}
 	spin_unlock(&dlm->spinlock);
 
-	out += snprintf(buf + out, len - out, "Total on list: %lu\n", total);
+	out += scnprintf(buf + out, len - out, "Total on list: %lu\n", total);
 
 	return out;
 }
@@ -410,7 +410,7 @@ static int debug_mle_print(struct dlm_ctxt *dlm, char *buf, int len)
 	int i, out = 0;
 	unsigned long total = 0, longest = 0, bucket_count = 0;
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Dumping MLEs for Domain: %s\n", dlm->name);
 
 	spin_lock(&dlm->master_lock);
@@ -428,7 +428,7 @@ static int debug_mle_print(struct dlm_ctxt *dlm, char *buf, int len)
 	}
 	spin_unlock(&dlm->master_lock);
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Total: %lu, Longest: %lu\n", total, longest);
 	return out;
 }
@@ -467,7 +467,7 @@ static int dump_lock(struct dlm_lock *lock, int list_type, char *buf, int len)
 
 #define DEBUG_LOCK_VERSION	1
 	spin_lock(&lock->spinlock);
-	out = snprintf(buf, len, "LOCK:%d,%d,%d,%d,%d,%d:%lld,%d,%d,%d,%d,%d,"
+	out = scnprintf(buf, len, "LOCK:%d,%d,%d,%d,%d,%d:%lld,%d,%d,%d,%d,%d,"
 		       "%d,%d,%d,%d\n",
 		       DEBUG_LOCK_VERSION,
 		       list_type, lock->ml.type, lock->ml.convert_type,
@@ -491,13 +491,13 @@ static int dump_lockres(struct dlm_lock_resource *res, char *buf, int len)
 	int i;
 	int out = 0;
 
-	out += snprintf(buf + out, len - out, "NAME:");
+	out += scnprintf(buf + out, len - out, "NAME:");
 	out += stringify_lockname(res->lockname.name, res->lockname.len,
 				  buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 #define DEBUG_LRES_VERSION	1
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"LRES:%d,%d,%d,%ld,%d,%d,%d,%d,%d,%d,%d\n",
 			DEBUG_LRES_VERSION,
 			res->owner, res->state, res->last_used,
@@ -509,17 +509,17 @@ static int dump_lockres(struct dlm_lock_resource *res, char *buf, int len)
 			kref_read(&res->refs));
 
 	/* refmap */
-	out += snprintf(buf + out, len - out, "RMAP:");
+	out += scnprintf(buf + out, len - out, "RMAP:");
 	out += stringify_nodemap(res->refmap, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 	/* lvb */
-	out += snprintf(buf + out, len - out, "LVBX:");
+	out += scnprintf(buf + out, len - out, "LVBX:");
 	for (i = 0; i < DLM_LVB_LEN; i++)
-		out += snprintf(buf + out, len - out,
+		out += scnprintf(buf + out, len - out,
 					"%02x", (unsigned char)res->lvb[i]);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 	/* granted */
 	list_for_each_entry(lock, &res->granted, list)
@@ -533,7 +533,7 @@ static int dump_lockres(struct dlm_lock_resource *res, char *buf, int len)
 	list_for_each_entry(lock, &res->blocked, list)
 		out += dump_lock(lock, 2, buf + out, len - out);
 
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 	return out;
 }
@@ -683,41 +683,41 @@ static int debug_state_print(struct dlm_ctxt *dlm, char *buf, int len)
 	}
 
 	/* Domain: xxxxxxxxxx  Key: 0xdfbac769 */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Domain: %s  Key: 0x%08x  Protocol: %d.%d\n",
 			dlm->name, dlm->key, dlm->dlm_locking_proto.pv_major,
 			dlm->dlm_locking_proto.pv_minor);
 
 	/* Thread Pid: xxx  Node: xxx  State: xxxxx */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Thread Pid: %d  Node: %d  State: %s\n",
 			task_pid_nr(dlm->dlm_thread_task), dlm->node_num, state);
 
 	/* Number of Joins: xxx  Joining Node: xxx */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Number of Joins: %d  Joining Node: %d\n",
 			dlm->num_joins, dlm->joining_node);
 
 	/* Domain Map: xx xx xx */
-	out += snprintf(buf + out, len - out, "Domain Map: ");
+	out += scnprintf(buf + out, len - out, "Domain Map: ");
 	out += stringify_nodemap(dlm->domain_map, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 	/* Exit Domain Map: xx xx xx */
-	out += snprintf(buf + out, len - out, "Exit Domain Map: ");
+	out += scnprintf(buf + out, len - out, "Exit Domain Map: ");
 	out += stringify_nodemap(dlm->exit_domain_map, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 	/* Live Map: xx xx xx */
-	out += snprintf(buf + out, len - out, "Live Map: ");
+	out += scnprintf(buf + out, len - out, "Live Map: ");
 	out += stringify_nodemap(dlm->live_nodes_map, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 	/* Lock Resources: xxx (xxx) */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Lock Resources: %d (%d)\n",
 			atomic_read(&dlm->res_cur_count),
 			atomic_read(&dlm->res_tot_count));
@@ -729,29 +729,29 @@ static int debug_state_print(struct dlm_ctxt *dlm, char *buf, int len)
 		cur_mles += atomic_read(&dlm->mle_cur_count[i]);
 
 	/* MLEs: xxx (xxx) */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"MLEs: %d (%d)\n", cur_mles, tot_mles);
 
 	/*  Blocking: xxx (xxx) */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"  Blocking: %d (%d)\n",
 			atomic_read(&dlm->mle_cur_count[DLM_MLE_BLOCK]),
 			atomic_read(&dlm->mle_tot_count[DLM_MLE_BLOCK]));
 
 	/*  Mastery: xxx (xxx) */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"  Mastery: %d (%d)\n",
 			atomic_read(&dlm->mle_cur_count[DLM_MLE_MASTER]),
 			atomic_read(&dlm->mle_tot_count[DLM_MLE_MASTER]));
 
 	/*  Migration: xxx (xxx) */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"  Migration: %d (%d)\n",
 			atomic_read(&dlm->mle_cur_count[DLM_MLE_MIGRATION]),
 			atomic_read(&dlm->mle_tot_count[DLM_MLE_MIGRATION]));
 
 	/* Lists: Dirty=Empty  Purge=InUse  PendingASTs=Empty  ... */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Lists: Dirty=%s  Purge=%s  PendingASTs=%s  "
 			"PendingBASTs=%s\n",
 			(list_empty(&dlm->dirty_list) ? "Empty" : "InUse"),
@@ -760,12 +760,12 @@ static int debug_state_print(struct dlm_ctxt *dlm, char *buf, int len)
 			(list_empty(&dlm->pending_basts) ? "Empty" : "InUse"));
 
 	/* Purge Count: xxx  Refs: xxx */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Purge Count: %d  Refs: %d\n", dlm->purge_count,
 			kref_read(&dlm->dlm_refs));
 
 	/* Dead Node: xxx */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Dead Node: %d\n", dlm->reco.dead_node);
 
 	/* What about DLM_RECO_STATE_FINALIZE? */
@@ -775,19 +775,19 @@ static int debug_state_print(struct dlm_ctxt *dlm, char *buf, int len)
 		state = "INACTIVE";
 
 	/* Recovery Pid: xxxx  Master: xxx  State: xxxx */
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"Recovery Pid: %d  Master: %d  State: %s\n",
 			task_pid_nr(dlm->dlm_reco_thread_task),
 			dlm->reco.new_master, state);
 
 	/* Recovery Map: xx xx */
-	out += snprintf(buf + out, len - out, "Recovery Map: ");
+	out += scnprintf(buf + out, len - out, "Recovery Map: ");
 	out += stringify_nodemap(dlm->recovery_map, O2NM_MAX_NODES,
 				 buf + out, len - out);
-	out += snprintf(buf + out, len - out, "\n");
+	out += scnprintf(buf + out, len - out, "\n");
 
 	/* Recovery Node State: */
-	out += snprintf(buf + out, len - out, "Recovery Node State:\n");
+	out += scnprintf(buf + out, len - out, "Recovery Node State:\n");
 	list_for_each_entry(node, &dlm->reco.node_data, list) {
 		switch (node->state) {
 		case DLM_RECO_NODE_DATA_INIT:
@@ -815,7 +815,7 @@ static int debug_state_print(struct dlm_ctxt *dlm, char *buf, int len)
 			state = "BAD";
 			break;
 		}
-		out += snprintf(buf + out, len - out, "\t%u - %s\n",
+		out += scnprintf(buf + out, len - out, "\t%u - %s\n",
 				node->node_num, state);
 	}
 
diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index ee6f459f9770..357cfc702ce3 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -20,9 +20,9 @@
 #include <linux/debugfs.h>
 #include <linux/sched/signal.h>
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
@@ -30,7 +30,7 @@
 #include "dlmdebug.h"
 
 #define MLOG_MASK_PREFIX (ML_DLM|ML_DLM_DOMAIN)
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 /*
  * ocfs2 node maps are array of long int, which limits to send them freely
diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c
index baff087f3863..83f0760e4fba 100644
--- a/fs/ocfs2/dlm/dlmlock.c
+++ b/fs/ocfs2/dlm/dlmlock.c
@@ -25,9 +25,9 @@
 #include <linux/delay.h>
 
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
@@ -35,7 +35,7 @@
 #include "dlmconvert.h"
 
 #define MLOG_MASK_PREFIX ML_DLM
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 static struct kmem_cache *dlm_lock_cache;
 
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index 74b768ca1cd8..f105746063ed 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -25,9 +25,9 @@
 #include <linux/delay.h>
 
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
@@ -35,7 +35,7 @@
 #include "dlmdebug.h"
 
 #define MLOG_MASK_PREFIX (ML_DLM|ML_DLM_MASTER)
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 static void dlm_mle_node_down(struct dlm_ctxt *dlm,
 			      struct dlm_master_list_entry *mle,
@@ -2554,8 +2554,6 @@ static int dlm_migrate_lockres(struct dlm_ctxt *dlm,
 	if (!dlm_grab(dlm))
 		return -EINVAL;
 
-	BUG_ON(target == O2NM_MAX_NODES);
-
 	name = res->lockname.name;
 	namelen = res->lockname.len;
 
@@ -2751,8 +2749,6 @@ leave:
 	return ret;
 }
 
-#define DLM_MIGRATION_RETRY_MS  100
-
 /*
  * Should be called only after beginning the domain leave process.
  * There should not be any remaining locks on nonlocal lock resources,
@@ -2764,6 +2760,7 @@ leave:
  * Returns: 1 if dlm->spinlock was dropped/retaken, 0 if never dropped
  */
 int dlm_empty_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res)
+	__must_hold(&dlm->spinlock)
 {
 	int ret;
 	int lock_dropped = 0;
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index 064ce5bbc3f6..4b566e88582f 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -26,16 +26,16 @@
 #include <linux/delay.h>
 
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
 #include "dlmdomain.h"
 
 #define MLOG_MASK_PREFIX (ML_DLM|ML_DLM_RECOVERY)
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 static void dlm_do_local_recovery_cleanup(struct dlm_ctxt *dlm, u8 dead_node);
 
@@ -1668,7 +1668,7 @@ static int dlm_lockres_master_requery(struct dlm_ctxt *dlm,
 int dlm_do_master_requery(struct dlm_ctxt *dlm, struct dlm_lock_resource *res,
 			  u8 nodenum, u8 *real_master)
 {
-	int ret = -EINVAL;
+	int ret;
 	struct dlm_master_requery req;
 	int status = DLM_LOCK_RES_OWNER_UNKNOWN;
 
diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c
index 61c51c268460..5ccc4ff0b82a 100644
--- a/fs/ocfs2/dlm/dlmthread.c
+++ b/fs/ocfs2/dlm/dlmthread.c
@@ -25,22 +25,20 @@
 #include <linux/delay.h>
 
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
 #include "dlmdomain.h"
 
 #define MLOG_MASK_PREFIX (ML_DLM|ML_DLM_THREAD)
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 static int dlm_thread(void *data);
 static void dlm_flush_asts(struct dlm_ctxt *dlm);
 
-#define dlm_lock_is_remote(dlm, lock)     ((lock)->ml.node != (dlm)->node_num)
-
 /* will exit holding res->spinlock, but may drop in function */
 /* waits until flags are cleared on res->state */
 void __dlm_wait_on_lockres_flags(struct dlm_lock_resource *res, int flags)
@@ -680,7 +678,6 @@ static void dlm_flush_asts(struct dlm_ctxt *dlm)
 
 #define DLM_THREAD_TIMEOUT_MS (4 * 1000)
 #define DLM_THREAD_MAX_DIRTY  100
-#define DLM_THREAD_MAX_ASTS   10
 
 static int dlm_thread(void *data)
 {
diff --git a/fs/ocfs2/dlm/dlmunlock.c b/fs/ocfs2/dlm/dlmunlock.c
index 3883633e82eb..dcb17ca8ae74 100644
--- a/fs/ocfs2/dlm/dlmunlock.c
+++ b/fs/ocfs2/dlm/dlmunlock.c
@@ -23,15 +23,15 @@
 #include <linux/spinlock.h>
 #include <linux/delay.h>
 
-#include "cluster/heartbeat.h"
-#include "cluster/nodemanager.h"
-#include "cluster/tcp.h"
+#include "../cluster/heartbeat.h"
+#include "../cluster/nodemanager.h"
+#include "../cluster/tcp.h"
 
 #include "dlmapi.h"
 #include "dlmcommon.h"
 
 #define MLOG_MASK_PREFIX ML_DLM
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 #define DLM_UNLOCK_FREE_LOCK           0x00000001
 #define DLM_UNLOCK_CALL_AST            0x00000002
diff --git a/fs/ocfs2/dlmfs/Makefile b/fs/ocfs2/dlmfs/Makefile
index a9874e441bd4..c7895f65be0e 100644
--- a/fs/ocfs2/dlmfs/Makefile
+++ b/fs/ocfs2/dlmfs/Makefile
@@ -1,6 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0-only
-ccflags-y := -I $(srctree)/$(src)/..
-
 obj-$(CONFIG_OCFS2_FS) += ocfs2_dlmfs.o
 
 ocfs2_dlmfs-objs := userdlm.o dlmfs.o
diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c
index 4f1668c81e1f..ea868c6f9800 100644
--- a/fs/ocfs2/dlmfs/dlmfs.c
+++ b/fs/ocfs2/dlmfs/dlmfs.c
@@ -33,11 +33,11 @@
 
 #include <linux/uaccess.h>
 
-#include "stackglue.h"
+#include "../stackglue.h"
 #include "userdlm.h"
 
 #define MLOG_MASK_PREFIX ML_DLMFS
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 
 static const struct super_operations dlmfs_ops;
@@ -227,7 +227,7 @@ static ssize_t dlmfs_file_read(struct file *filp,
 			       loff_t *ppos)
 {
 	int bytes_left;
-	ssize_t readlen, got;
+	ssize_t got;
 	char *lvb_buf;
 	struct inode *inode = file_inode(filp);
 
@@ -237,36 +237,31 @@ static ssize_t dlmfs_file_read(struct file *filp,
 	if (*ppos >= i_size_read(inode))
 		return 0;
 
+	/* don't read past the lvb */
+	if (count > i_size_read(inode) - *ppos)
+		count = i_size_read(inode) - *ppos;
+
 	if (!count)
 		return 0;
 
-	if (!access_ok(buf, count))
-		return -EFAULT;
-
-	/* don't read past the lvb */
-	if ((count + *ppos) > i_size_read(inode))
-		readlen = i_size_read(inode) - *ppos;
-	else
-		readlen = count;
-
-	lvb_buf = kmalloc(readlen, GFP_NOFS);
+	lvb_buf = kmalloc(count, GFP_NOFS);
 	if (!lvb_buf)
 		return -ENOMEM;
 
-	got = user_dlm_read_lvb(inode, lvb_buf, readlen);
+	got = user_dlm_read_lvb(inode, lvb_buf, count);
 	if (got) {
-		BUG_ON(got != readlen);
-		bytes_left = __copy_to_user(buf, lvb_buf, readlen);
-		readlen -= bytes_left;
+		BUG_ON(got != count);
+		bytes_left = copy_to_user(buf, lvb_buf, count);
+		count -= bytes_left;
 	} else
-		readlen = 0;
+		count = 0;
 
 	kfree(lvb_buf);
 
-	*ppos = *ppos + readlen;
+	*ppos = *ppos + count;
 
-	mlog(0, "read %zd bytes\n", readlen);
-	return readlen;
+	mlog(0, "read %zu bytes\n", count);
+	return count;
 }
 
 static ssize_t dlmfs_file_write(struct file *filp,
@@ -275,7 +270,6 @@ static ssize_t dlmfs_file_write(struct file *filp,
 				loff_t *ppos)
 {
 	int bytes_left;
-	ssize_t writelen;
 	char *lvb_buf;
 	struct inode *inode = file_inode(filp);
 
@@ -285,32 +279,27 @@ static ssize_t dlmfs_file_write(struct file *filp,
 	if (*ppos >= i_size_read(inode))
 		return -ENOSPC;
 
+	/* don't write past the lvb */
+	if (count > i_size_read(inode) - *ppos)
+		count = i_size_read(inode) - *ppos;
+
 	if (!count)
 		return 0;
 
-	if (!access_ok(buf, count))
-		return -EFAULT;
-
-	/* don't write past the lvb */
-	if ((count + *ppos) > i_size_read(inode))
-		writelen = i_size_read(inode) - *ppos;
-	else
-		writelen = count - *ppos;
-
-	lvb_buf = kmalloc(writelen, GFP_NOFS);
+	lvb_buf = kmalloc(count, GFP_NOFS);
 	if (!lvb_buf)
 		return -ENOMEM;
 
-	bytes_left = copy_from_user(lvb_buf, buf, writelen);
-	writelen -= bytes_left;
-	if (writelen)
-		user_dlm_write_lvb(inode, lvb_buf, writelen);
+	bytes_left = copy_from_user(lvb_buf, buf, count);
+	count -= bytes_left;
+	if (count)
+		user_dlm_write_lvb(inode, lvb_buf, count);
 
 	kfree(lvb_buf);
 
-	*ppos = *ppos + writelen;
-	mlog(0, "wrote %zd bytes\n", writelen);
-	return writelen;
+	*ppos = *ppos + count;
+	mlog(0, "wrote %zu bytes\n", count);
+	return count;
 }
 
 static void dlmfs_init_once(void *foo)
diff --git a/fs/ocfs2/dlmfs/userdlm.c b/fs/ocfs2/dlmfs/userdlm.c
index 525b14ddfba5..3df5be25bfb1 100644
--- a/fs/ocfs2/dlmfs/userdlm.c
+++ b/fs/ocfs2/dlmfs/userdlm.c
@@ -21,12 +21,12 @@
 #include <linux/types.h>
 #include <linux/crc32.h>
 
-#include "ocfs2_lockingver.h"
-#include "stackglue.h"
+#include "../ocfs2_lockingver.h"
+#include "../stackglue.h"
 #include "userdlm.h"
 
 #define MLOG_MASK_PREFIX ML_DLMFS
-#include "cluster/masklog.h"
+#include "../cluster/masklog.h"
 
 
 static inline struct user_lock_res *user_lksb_to_lock_res(struct ocfs2_dlm_lksb *lksb)
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 6e774c5ea13b..152a0fc4e905 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -570,7 +570,7 @@ void ocfs2_inode_lock_res_init(struct ocfs2_lock_res *res,
 			mlog_bug_on_msg(1, "type: %d\n", type);
 			ops = NULL; /* thanks, gcc */
 			break;
-	};
+	}
 
 	ocfs2_build_lock_name(type, OCFS2_I(inode)->ip_blkno,
 			      generation, res->l_name);
@@ -1687,7 +1687,7 @@ static void __ocfs2_cluster_unlock(struct ocfs2_super *osb,
 	spin_unlock_irqrestore(&lockres->l_lock, flags);
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 	if (lockres->l_lockdep_map.key != NULL)
-		rwsem_release(&lockres->l_lockdep_map, 1, caller_ip);
+		rwsem_release(&lockres->l_lockdep_map, caller_ip);
 #endif
 }
 
@@ -2133,7 +2133,7 @@ static void ocfs2_downconvert_on_unlock(struct ocfs2_super *osb,
 }
 
 #define OCFS2_SEC_BITS   34
-#define OCFS2_SEC_SHIFT  (64 - 34)
+#define OCFS2_SEC_SHIFT  (64 - OCFS2_SEC_BITS)
 #define OCFS2_NSEC_MASK  ((1ULL << OCFS2_SEC_SHIFT) - 1)
 
 /* LVB only has room for 64 bits of time here so we pack it for
@@ -3282,6 +3282,7 @@ static void ocfs2_dlm_init_debug(struct ocfs2_super *osb)
 
 	debugfs_create_u32("locking_filter", 0600, osb->osb_debug_root,
 			   &dlm_debug->d_filter_secs);
+	ocfs2_get_dlm_debug(dlm_debug);
 }
 
 static void ocfs2_dlm_shutdown_debug(struct ocfs2_super *osb)
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 9876db52913a..85979e2214b3 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -194,7 +194,7 @@ static int ocfs2_sync_file(struct file *file, loff_t start, loff_t end,
 		needs_barrier = true;
 	err = jbd2_complete_transaction(journal, commit_tid);
 	if (needs_barrier) {
-		ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+		ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
 		if (!err)
 			err = ret;
 	}
@@ -2101,17 +2101,15 @@ static int ocfs2_is_io_unaligned(struct inode *inode, size_t count, loff_t pos)
 static int ocfs2_inode_lock_for_extent_tree(struct inode *inode,
 					    struct buffer_head **di_bh,
 					    int meta_level,
-					    int overwrite_io,
 					    int write_sem,
 					    int wait)
 {
 	int ret = 0;
 
 	if (wait)
-		ret = ocfs2_inode_lock(inode, NULL, meta_level);
+		ret = ocfs2_inode_lock(inode, di_bh, meta_level);
 	else
-		ret = ocfs2_try_inode_lock(inode,
-			overwrite_io ? NULL : di_bh, meta_level);
+		ret = ocfs2_try_inode_lock(inode, di_bh, meta_level);
 	if (ret < 0)
 		goto out;
 
@@ -2136,6 +2134,7 @@ static int ocfs2_inode_lock_for_extent_tree(struct inode *inode,
 
 out_unlock:
 	brelse(*di_bh);
+	*di_bh = NULL;
 	ocfs2_inode_unlock(inode, meta_level);
 out:
 	return ret;
@@ -2177,7 +2176,6 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 		ret = ocfs2_inode_lock_for_extent_tree(inode,
 						       &di_bh,
 						       meta_level,
-						       overwrite_io,
 						       write_sem,
 						       wait);
 		if (ret < 0) {
@@ -2233,13 +2231,13 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 							   &di_bh,
 							   meta_level,
 							   write_sem);
+			meta_level = 1;
+			write_sem = 1;
 			ret = ocfs2_inode_lock_for_extent_tree(inode,
 							       &di_bh,
 							       meta_level,
-							       overwrite_io,
-							       1,
+							       write_sem,
 							       wait);
-			write_sem = 1;
 			if (ret < 0) {
 				if (ret != -EAGAIN)
 					mlog_errno(ret);
diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c
index efeea208fdeb..89984172fc4a 100644
--- a/fs/ocfs2/ioctl.c
+++ b/fs/ocfs2/ioctl.c
@@ -985,6 +985,7 @@ long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 			return -EFAULT;
 
 		return ocfs2_info_handle(inode, &info, 1);
+	case FITRIM:
 	case OCFS2_IOC_MOVE_EXT:
 		break;
 	default:
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 699a560efbb0..b425f0b01dce 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -91,7 +91,7 @@ enum ocfs2_replay_state {
 struct ocfs2_replay_map {
 	unsigned int rm_slots;
 	enum ocfs2_replay_state rm_state;
-	unsigned char rm_replay_slots[0];
+	unsigned char rm_replay_slots[];
 };
 
 static void ocfs2_replay_map_set_state(struct ocfs2_super *osb, int state)
@@ -420,14 +420,14 @@ int ocfs2_extend_trans(handle_t *handle, int nblocks)
 	if (!nblocks)
 		return 0;
 
-	old_nblocks = handle->h_buffer_credits;
+	old_nblocks = jbd2_handle_buffer_credits(handle);
 
 	trace_ocfs2_extend_trans(old_nblocks, nblocks);
 
 #ifdef CONFIG_OCFS2_DEBUG_FS
 	status = 1;
 #else
-	status = jbd2_journal_extend(handle, nblocks);
+	status = jbd2_journal_extend(handle, nblocks, 0);
 	if (status < 0) {
 		mlog_errno(status);
 		goto bail;
@@ -461,13 +461,13 @@ int ocfs2_allocate_extend_trans(handle_t *handle, int thresh)
 
 	BUG_ON(!handle);
 
-	old_nblks = handle->h_buffer_credits;
+	old_nblks = jbd2_handle_buffer_credits(handle);
 	trace_ocfs2_allocate_extend_trans(old_nblks, thresh);
 
 	if (old_nblks < thresh)
 		return 0;
 
-	status = jbd2_journal_extend(handle, OCFS2_MAX_TRANS_DATA);
+	status = jbd2_journal_extend(handle, OCFS2_MAX_TRANS_DATA, 0);
 	if (status < 0) {
 		mlog_errno(status);
 		goto bail;
@@ -1066,6 +1066,14 @@ int ocfs2_journal_load(struct ocfs2_journal *journal, int local, int replayed)
 
 	ocfs2_clear_journal_error(osb->sb, journal->j_journal, osb->slot_num);
 
+	if (replayed) {
+		jbd2_journal_lock_updates(journal->j_journal);
+		status = jbd2_journal_flush(journal->j_journal);
+		jbd2_journal_unlock_updates(journal->j_journal);
+		if (status < 0)
+			mlog_errno(status);
+	}
+
 	status = ocfs2_journal_toggle_dirty(osb, 1, replayed);
 	if (status < 0) {
 		mlog_errno(status);
diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h
index 3103ba7f97a2..bfe611ed1b1d 100644
--- a/fs/ocfs2/journal.h
+++ b/fs/ocfs2/journal.h
@@ -597,9 +597,11 @@ static inline void ocfs2_update_inode_fsync_trans(handle_t *handle,
 {
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 
-	oi->i_sync_tid = handle->h_transaction->t_tid;
-	if (datasync)
-		oi->i_datasync_tid = handle->h_transaction->t_tid;
+	if (!is_handle_aborted(handle)) {
+		oi->i_sync_tid = handle->h_transaction->t_tid;
+		if (datasync)
+			oi->i_datasync_tid = handle->h_transaction->t_tid;
+	}
 }
 
 #endif /* OCFS2_JOURNAL_H */
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 8ea51cf27b97..5381020aaa9a 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -406,7 +406,7 @@ static int ocfs2_mknod(struct inode *dir,
 
 	if (status < 0) {
 		mlog_errno(status);
-		goto leave;
+		goto roll_back;
 	}
 
 	if (si.enable) {
@@ -414,7 +414,7 @@ static int ocfs2_mknod(struct inode *dir,
 						 meta_ac, data_ac);
 		if (status < 0) {
 			mlog_errno(status);
-			goto leave;
+			goto roll_back;
 		}
 	}
 
@@ -427,7 +427,7 @@ static int ocfs2_mknod(struct inode *dir,
 					  OCFS2_I(dir)->ip_blkno);
 	if (status) {
 		mlog_errno(status);
-		goto leave;
+		goto roll_back;
 	}
 
 	dl = dentry->d_fsdata;
@@ -437,12 +437,19 @@ static int ocfs2_mknod(struct inode *dir,
 				 &lookup);
 	if (status < 0) {
 		mlog_errno(status);
-		goto leave;
+		goto roll_back;
 	}
 
 	insert_inode_hash(inode);
 	d_instantiate(dentry, inode);
 	status = 0;
+
+roll_back:
+	if (status < 0 && S_ISDIR(mode)) {
+		ocfs2_add_links_count(dirfe, -1);
+		drop_nlink(dir);
+	}
+
 leave:
 	if (status < 0 && did_quota_inode)
 		dquot_free_inode(inode);
@@ -586,8 +593,7 @@ static int __ocfs2_mknod_locked(struct inode *dir,
 			mlog_errno(status);
 	}
 
-	oi->i_sync_tid = handle->h_transaction->t_tid;
-	oi->i_datasync_tid = handle->h_transaction->t_tid;
+	ocfs2_update_inode_fsync_trans(handle, inode, 1);
 
 leave:
 	if (status < 0) {
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index 9150cfa4df7d..ee5d98516212 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -279,6 +279,7 @@ enum ocfs2_mount_options
 	OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT = 1 << 15,  /* Journal Async Commit */
 	OCFS2_MOUNT_ERRORS_CONT = 1 << 16, /* Return EIO to the calling process on error */
 	OCFS2_MOUNT_ERRORS_ROFS = 1 << 17, /* Change filesystem to read-only on error */
+	OCFS2_MOUNT_NOCLUSTER = 1 << 18, /* No cluster aware filesystem mount */
 };
 
 #define OCFS2_OSB_SOFT_RO	0x0001
@@ -673,7 +674,8 @@ static inline int ocfs2_cluster_o2cb_global_heartbeat(struct ocfs2_super *osb)
 
 static inline int ocfs2_mount_local(struct ocfs2_super *osb)
 {
-	return (osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_LOCAL_MOUNT);
+	return ((osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_LOCAL_MOUNT)
+		|| (osb->s_mount_opt & OCFS2_MOUNT_NOCLUSTER));
 }
 
 static inline int ocfs2_uses_extended_slot_map(struct ocfs2_super *osb)
diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h
index 0db4a7ec58a2..0dd8c41bafd4 100644
--- a/fs/ocfs2/ocfs2_fs.h
+++ b/fs/ocfs2/ocfs2_fs.h
@@ -470,7 +470,7 @@ struct ocfs2_extent_list {
 	__le16 l_reserved1;
 	__le64 l_reserved2;		/* Pad to
 					   sizeof(ocfs2_extent_rec) */
-/*10*/	struct ocfs2_extent_rec l_recs[0];	/* Extent records */
+/*10*/	struct ocfs2_extent_rec l_recs[];	/* Extent records */
 };
 
 /*
@@ -484,7 +484,7 @@ struct ocfs2_chain_list {
 	__le16 cl_count;		/* Total chains in this list */
 	__le16 cl_next_free_rec;	/* Next unused chain slot */
 	__le64 cl_reserved1;
-/*10*/	struct ocfs2_chain_rec cl_recs[0];	/* Chain records */
+/*10*/	struct ocfs2_chain_rec cl_recs[];	/* Chain records */
 };
 
 /*
@@ -496,7 +496,7 @@ struct ocfs2_truncate_log {
 /*00*/	__le16 tl_count;		/* Total records in this log */
 	__le16 tl_used;			/* Number of records in use */
 	__le32 tl_reserved1;
-/*08*/	struct ocfs2_truncate_rec tl_recs[0];	/* Truncate records */
+/*08*/	struct ocfs2_truncate_rec tl_recs[];	/* Truncate records */
 };
 
 /*
@@ -640,7 +640,7 @@ struct ocfs2_local_alloc
 	__le16 la_size;		/* Size of included bitmap, in bytes */
 	__le16 la_reserved1;
 	__le64 la_reserved2;
-/*10*/	__u8   la_bitmap[0];
+/*10*/	__u8   la_bitmap[];
 };
 
 /*
@@ -653,7 +653,7 @@ struct ocfs2_inline_data
 				 * for data, starting at id_data */
 	__le16	id_reserved0;
 	__le32	id_reserved1;
-	__u8	id_data[0];	/* Start of user data */
+	__u8	id_data[];	/* Start of user data */
 };
 
 /*
@@ -798,7 +798,7 @@ struct ocfs2_dx_entry_list {
 					 * possible in de_entries */
 	__le16		de_num_used;	/* Current number of
 					 * de_entries entries */
-	struct	ocfs2_dx_entry		de_entries[0];	/* Indexed dir entries
+	struct	ocfs2_dx_entry		de_entries[];	/* Indexed dir entries
 							 * in a packed array of
 							 * length de_num_used */
 };
@@ -935,7 +935,7 @@ struct ocfs2_refcount_list {
 	__le16 rl_used;		/* Current number of used records */
 	__le32 rl_reserved2;
 	__le64 rl_reserved1;	/* Pad to sizeof(ocfs2_refcount_record) */
-/*10*/	struct ocfs2_refcount_rec rl_recs[0];	/* Refcount records */
+/*10*/	struct ocfs2_refcount_rec rl_recs[];	/* Refcount records */
 };
 
 
@@ -1021,7 +1021,7 @@ struct ocfs2_xattr_header {
 						    buckets.  A block uses
 						    xb_check and sets
 						    this field to zero.) */
-	struct ocfs2_xattr_entry xh_entries[0]; /* xattr entry list. */
+	struct ocfs2_xattr_entry xh_entries[]; /* xattr entry list. */
 };
 
 /*
@@ -1207,7 +1207,7 @@ struct ocfs2_local_disk_dqinfo {
 /* Header of one chunk of a quota file */
 struct ocfs2_local_disk_chunk {
 	__le32 dqc_free;	/* Number of free entries in the bitmap */
-	__u8 dqc_bitmap[0];	/* Bitmap of entries in the corresponding
+	__u8 dqc_bitmap[];	/* Bitmap of entries in the corresponding
 				 * chunk of quota file */
 };
 
diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c
index 7a922190a8c7..eda83487c9ec 100644
--- a/fs/ocfs2/quota_global.c
+++ b/fs/ocfs2/quota_global.c
@@ -728,7 +728,7 @@ static int ocfs2_release_dquot(struct dquot *dquot)
 
 	mutex_lock(&dquot->dq_lock);
 	/* Check whether we are not racing with some other dqget() */
-	if (atomic_read(&dquot->dq_count) > 1)
+	if (dquot_is_busy(dquot))
 		goto out;
 	/* Running from downconvert thread? Postpone quota processing to wq */
 	if (current == osb->dc_task) {
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index ee43e51188be..cfb77f70c888 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -154,6 +154,7 @@ ocfs2_refcount_cache_get_super(struct ocfs2_caching_info *ci)
 }
 
 static void ocfs2_refcount_cache_lock(struct ocfs2_caching_info *ci)
+__acquires(&rf->rf_lock)
 {
 	struct ocfs2_refcount_tree *rf = cache_info_to_refcount(ci);
 
@@ -161,6 +162,7 @@ static void ocfs2_refcount_cache_lock(struct ocfs2_caching_info *ci)
 }
 
 static void ocfs2_refcount_cache_unlock(struct ocfs2_caching_info *ci)
+__releases(&rf->rf_lock)
 {
 	struct ocfs2_refcount_tree *rf = cache_info_to_refcount(ci);
 
diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c
index 0249e8ca1028..bf3842e34fb9 100644
--- a/fs/ocfs2/reservations.c
+++ b/fs/ocfs2/reservations.c
@@ -33,9 +33,6 @@
 
 static DEFINE_SPINLOCK(resv_lock);
 
-#define	OCFS2_MIN_RESV_WINDOW_BITS	8
-#define	OCFS2_MAX_RESV_WINDOW_BITS	1024
-
 int ocfs2_dir_resv_allowed(struct ocfs2_super *osb)
 {
 	return (osb->osb_resv_level && osb->osb_dir_resv_level);
diff --git a/fs/ocfs2/slot_map.c b/fs/ocfs2/slot_map.c
index 8caeceeaeda7..4da0e4b1e79b 100644
--- a/fs/ocfs2/slot_map.c
+++ b/fs/ocfs2/slot_map.c
@@ -254,14 +254,16 @@ static int __ocfs2_find_empty_slot(struct ocfs2_slot_info *si,
 	int i, ret = -ENOSPC;
 
 	if ((preferred >= 0) && (preferred < si->si_num_slots)) {
-		if (!si->si_slots[preferred].sl_valid) {
+		if (!si->si_slots[preferred].sl_valid ||
+		    !si->si_slots[preferred].sl_node_num) {
 			ret = preferred;
 			goto out;
 		}
 	}
 
 	for(i = 0; i < si->si_num_slots; i++) {
-		if (!si->si_slots[i].sl_valid) {
+		if (!si->si_slots[i].sl_valid ||
+		    !si->si_slots[i].sl_node_num) {
 			ret = i;
 			break;
 		}
@@ -456,24 +458,30 @@ int ocfs2_find_slot(struct ocfs2_super *osb)
 	spin_lock(&osb->osb_lock);
 	ocfs2_update_slot_info(si);
 
-	/* search for ourselves first and take the slot if it already
-	 * exists. Perhaps we need to mark this in a variable for our
-	 * own journal recovery? Possibly not, though we certainly
-	 * need to warn to the user */
-	slot = __ocfs2_node_num_to_slot(si, osb->node_num);
-	if (slot < 0) {
-		/* if no slot yet, then just take 1st available
-		 * one. */
-		slot = __ocfs2_find_empty_slot(si, osb->preferred_slot);
+	if (ocfs2_mount_local(osb))
+		/* use slot 0 directly in local mode */
+		slot = 0;
+	else {
+		/* search for ourselves first and take the slot if it already
+		 * exists. Perhaps we need to mark this in a variable for our
+		 * own journal recovery? Possibly not, though we certainly
+		 * need to warn to the user */
+		slot = __ocfs2_node_num_to_slot(si, osb->node_num);
 		if (slot < 0) {
-			spin_unlock(&osb->osb_lock);
-			mlog(ML_ERROR, "no free slots available!\n");
-			status = -EINVAL;
-			goto bail;
-		}
-	} else
-		printk(KERN_INFO "ocfs2: Slot %d on device (%s) was already "
-		       "allocated to this node!\n", slot, osb->dev_str);
+			/* if no slot yet, then just take 1st available
+			 * one. */
+			slot = __ocfs2_find_empty_slot(si, osb->preferred_slot);
+			if (slot < 0) {
+				spin_unlock(&osb->osb_lock);
+				mlog(ML_ERROR, "no free slots available!\n");
+				status = -EINVAL;
+				goto bail;
+			}
+		} else
+			printk(KERN_INFO "ocfs2: Slot %d on device (%s) was "
+			       "already allocated to this node!\n",
+			       slot, osb->dev_str);
+	}
 
 	ocfs2_set_slot(si, slot, osb->node_num);
 	osb->slot_num = slot;
diff --git a/fs/ocfs2/stackglue.c b/fs/ocfs2/stackglue.c
index 8aa6a667860c..a191094694c6 100644
--- a/fs/ocfs2/stackglue.c
+++ b/fs/ocfs2/stackglue.c
@@ -656,8 +656,6 @@ error:
  * and easier to preserve the name.
  */
 
-#define FS_OCFS2_NM		1
-
 static struct ctl_table ocfs2_nm_table[] = {
 	{
 		.procname	= "hb_ctl_path",
diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index 69c21a3843af..4836becb7578 100644
--- a/fs/ocfs2/suballoc.c
+++ b/fs/ocfs2/suballoc.c
@@ -696,7 +696,7 @@ static int ocfs2_block_group_alloc(struct ocfs2_super *osb,
 
 	bg_bh = ocfs2_block_group_alloc_contig(osb, handle, alloc_inode,
 					       ac, cl);
-	if (IS_ERR(bg_bh) && (PTR_ERR(bg_bh) == -ENOSPC))
+	if (PTR_ERR(bg_bh) == -ENOSPC)
 		bg_bh = ocfs2_block_group_alloc_discontig(handle,
 							  alloc_inode,
 							  ac, cl);
@@ -1252,6 +1252,7 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh,
 					 int nr)
 {
 	struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) bg_bh->b_data;
+	struct journal_head *jh;
 	int ret;
 
 	if (ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap))
@@ -1260,13 +1261,14 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh,
 	if (!buffer_jbd(bg_bh))
 		return 1;
 
-	jbd_lock_bh_state(bg_bh);
-	bg = (struct ocfs2_group_desc *) bh2jh(bg_bh)->b_committed_data;
+	jh = bh2jh(bg_bh);
+	spin_lock(&jh->b_state_lock);
+	bg = (struct ocfs2_group_desc *) jh->b_committed_data;
 	if (bg)
 		ret = !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
 	else
 		ret = 1;
-	jbd_unlock_bh_state(bg_bh);
+	spin_unlock(&jh->b_state_lock);
 
 	return ret;
 }
@@ -2387,6 +2389,7 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 	int status;
 	unsigned int tmp;
 	struct ocfs2_group_desc *undo_bg = NULL;
+	struct journal_head *jh;
 
 	/* The caller got this descriptor from
 	 * ocfs2_read_group_descriptor().  Any corruption is a code bug. */
@@ -2405,10 +2408,10 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 		goto bail;
 	}
 
+	jh = bh2jh(group_bh);
 	if (undo_fn) {
-		jbd_lock_bh_state(group_bh);
-		undo_bg = (struct ocfs2_group_desc *)
-					bh2jh(group_bh)->b_committed_data;
+		spin_lock(&jh->b_state_lock);
+		undo_bg = (struct ocfs2_group_desc *) jh->b_committed_data;
 		BUG_ON(!undo_bg);
 	}
 
@@ -2423,7 +2426,7 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 	le16_add_cpu(&bg->bg_free_bits_count, num_bits);
 	if (le16_to_cpu(bg->bg_free_bits_count) > le16_to_cpu(bg->bg_bits)) {
 		if (undo_fn)
-			jbd_unlock_bh_state(group_bh);
+			spin_unlock(&jh->b_state_lock);
 		return ocfs2_error(alloc_inode->i_sb, "Group descriptor # %llu has bit count %u but claims %u are freed. num_bits %d\n",
 				   (unsigned long long)le64_to_cpu(bg->bg_blkno),
 				   le16_to_cpu(bg->bg_bits),
@@ -2432,7 +2435,7 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 	}
 
 	if (undo_fn)
-		jbd_unlock_bh_state(group_bh);
+		spin_unlock(&jh->b_state_lock);
 
 	ocfs2_journal_dirty(handle, group_bh);
 bail:
@@ -2506,9 +2509,6 @@ static int _ocfs2_free_suballoc_bits(handle_t *handle,
 
 bail:
 	brelse(group_bh);
-
-	if (status)
-		mlog_errno(status);
 	return status;
 }
 
@@ -2579,8 +2579,6 @@ static int _ocfs2_free_clusters(handle_t *handle,
 					 num_clusters);
 
 out:
-	if (status)
-		mlog_errno(status);
 	return status;
 }
 
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index c81e86c62380..71ea9ce71a6b 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -175,6 +175,7 @@ enum {
 	Opt_dir_resv_level,
 	Opt_journal_async_commit,
 	Opt_err_cont,
+	Opt_nocluster,
 	Opt_err,
 };
 
@@ -208,6 +209,7 @@ static const match_table_t tokens = {
 	{Opt_dir_resv_level, "dir_resv_level=%u"},
 	{Opt_journal_async_commit, "journal_async_commit"},
 	{Opt_err_cont, "errors=continue"},
+	{Opt_nocluster, "nocluster"},
 	{Opt_err, NULL}
 };
 
@@ -220,31 +222,31 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 	int i, out = 0;
 	unsigned long flags;
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => Id: %-s  Uuid: %-s  Gen: 0x%X  Label: %-s\n",
 			"Device", osb->dev_str, osb->uuid_str,
 			osb->fs_generation, osb->vol_label);
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => State: %d  Flags: 0x%lX\n", "Volume",
 			atomic_read(&osb->vol_state), osb->osb_flags);
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => Block: %lu  Cluster: %d\n", "Sizes",
 			osb->sb->s_blocksize, osb->s_clustersize);
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => Compat: 0x%X  Incompat: 0x%X  "
 			"ROcompat: 0x%X\n",
 			"Features", osb->s_feature_compat,
 			osb->s_feature_incompat, osb->s_feature_ro_compat);
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => Opts: 0x%lX  AtimeQuanta: %u\n", "Mount",
 			osb->s_mount_opt, osb->s_atime_quantum);
 
 	if (cconn) {
-		out += snprintf(buf + out, len - out,
+		out += scnprintf(buf + out, len - out,
 				"%10s => Stack: %s  Name: %*s  "
 				"Version: %d.%d\n", "Cluster",
 				(*osb->osb_cluster_stack == '\0' ?
@@ -255,7 +257,7 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 	}
 
 	spin_lock_irqsave(&osb->dc_task_lock, flags);
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => Pid: %d  Count: %lu  WakeSeq: %lu  "
 			"WorkSeq: %lu\n", "DownCnvt",
 			(osb->dc_task ?  task_pid_nr(osb->dc_task) : -1),
@@ -264,32 +266,32 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 	spin_unlock_irqrestore(&osb->dc_task_lock, flags);
 
 	spin_lock(&osb->osb_lock);
-	out += snprintf(buf + out, len - out, "%10s => Pid: %d  Nodes:",
+	out += scnprintf(buf + out, len - out, "%10s => Pid: %d  Nodes:",
 			"Recovery",
 			(osb->recovery_thread_task ?
 			 task_pid_nr(osb->recovery_thread_task) : -1));
 	if (rm->rm_used == 0)
-		out += snprintf(buf + out, len - out, " None\n");
+		out += scnprintf(buf + out, len - out, " None\n");
 	else {
 		for (i = 0; i < rm->rm_used; i++)
-			out += snprintf(buf + out, len - out, " %d",
+			out += scnprintf(buf + out, len - out, " %d",
 					rm->rm_entries[i]);
-		out += snprintf(buf + out, len - out, "\n");
+		out += scnprintf(buf + out, len - out, "\n");
 	}
 	spin_unlock(&osb->osb_lock);
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => Pid: %d  Interval: %lu\n", "Commit",
 			(osb->commit_task ? task_pid_nr(osb->commit_task) : -1),
 			osb->osb_commit_interval);
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => State: %d  TxnId: %lu  NumTxns: %d\n",
 			"Journal", osb->journal->j_state,
 			osb->journal->j_trans_id,
 			atomic_read(&osb->journal->j_num_trans));
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => GlobalAllocs: %d  LocalAllocs: %d  "
 			"SubAllocs: %d  LAWinMoves: %d  SAExtends: %d\n",
 			"Stats",
@@ -299,7 +301,7 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 			atomic_read(&osb->alloc_stats.moves),
 			atomic_read(&osb->alloc_stats.bg_extends));
 
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => State: %u  Descriptor: %llu  Size: %u bits  "
 			"Default: %u bits\n",
 			"LocalAlloc", osb->local_alloc_state,
@@ -307,7 +309,7 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 			osb->local_alloc_bits, osb->local_alloc_default_bits);
 
 	spin_lock(&osb->osb_lock);
-	out += snprintf(buf + out, len - out,
+	out += scnprintf(buf + out, len - out,
 			"%10s => InodeSlot: %d  StolenInodes: %d, "
 			"MetaSlot: %d  StolenMeta: %d\n", "Steal",
 			osb->s_inode_steal_slot,
@@ -316,20 +318,20 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 			atomic_read(&osb->s_num_meta_stolen));
 	spin_unlock(&osb->osb_lock);
 
-	out += snprintf(buf + out, len - out, "OrphanScan => ");
-	out += snprintf(buf + out, len - out, "Local: %u  Global: %u ",
+	out += scnprintf(buf + out, len - out, "OrphanScan => ");
+	out += scnprintf(buf + out, len - out, "Local: %u  Global: %u ",
 			os->os_count, os->os_seqno);
-	out += snprintf(buf + out, len - out, " Last Scan: ");
+	out += scnprintf(buf + out, len - out, " Last Scan: ");
 	if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
-		out += snprintf(buf + out, len - out, "Disabled\n");
+		out += scnprintf(buf + out, len - out, "Disabled\n");
 	else
-		out += snprintf(buf + out, len - out, "%lu seconds ago\n",
+		out += scnprintf(buf + out, len - out, "%lu seconds ago\n",
 				(unsigned long)(ktime_get_seconds() - os->os_scantime));
 
-	out += snprintf(buf + out, len - out, "%10s => %3s  %10s\n",
+	out += scnprintf(buf + out, len - out, "%10s => %3s  %10s\n",
 			"Slots", "Num", "RecoGen");
 	for (i = 0; i < osb->max_slots; ++i) {
-		out += snprintf(buf + out, len - out,
+		out += scnprintf(buf + out, len - out,
 				"%10s  %c %3d  %10d\n",
 				" ",
 				(i == osb->slot_num ? '*' : ' '),
@@ -619,6 +621,13 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
 		goto out;
 	}
 
+	tmp = OCFS2_MOUNT_NOCLUSTER;
+	if ((osb->s_mount_opt & tmp) != (parsed_options.mount_opt & tmp)) {
+		ret = -EINVAL;
+		mlog(ML_ERROR, "Cannot change nocluster option on remount\n");
+		goto out;
+	}
+
 	tmp = OCFS2_MOUNT_HB_LOCAL | OCFS2_MOUNT_HB_GLOBAL |
 		OCFS2_MOUNT_HB_NONE;
 	if ((osb->s_mount_opt & tmp) != (parsed_options.mount_opt & tmp)) {
@@ -859,6 +868,7 @@ static int ocfs2_verify_userspace_stack(struct ocfs2_super *osb,
 	}
 
 	if (ocfs2_userspace_stack(osb) &&
+	    !(osb->s_mount_opt & OCFS2_MOUNT_NOCLUSTER) &&
 	    strncmp(osb->osb_cluster_stack, mopt->cluster_stack,
 		    OCFS2_STACK_LABEL_LEN)) {
 		mlog(ML_ERROR,
@@ -926,8 +936,8 @@ static int ocfs2_enable_quotas(struct ocfs2_super *osb)
 			status = -ENOENT;
 			goto out_quota_off;
 		}
-		status = dquot_enable(inode[type], type, QFMT_OCFS2,
-				      DQUOT_USAGE_ENABLED);
+		status = dquot_load_quota_inode(inode[type], type, QFMT_OCFS2,
+						DQUOT_USAGE_ENABLED);
 		if (status < 0)
 			goto out_quota_off;
 	}
@@ -1139,6 +1149,11 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
 	       osb->s_mount_opt & OCFS2_MOUNT_DATA_WRITEBACK ? "writeback" :
 	       "ordered");
 
+	if ((osb->s_mount_opt & OCFS2_MOUNT_NOCLUSTER) &&
+	   !(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_LOCAL_MOUNT))
+		printk(KERN_NOTICE "ocfs2: The shared device (%s) is mounted "
+		       "without cluster aware mode.\n", osb->dev_str);
+
 	atomic_set(&osb->vol_state, VOLUME_MOUNTED);
 	wake_up(&osb->osb_mount_event);
 
@@ -1445,6 +1460,9 @@ static int ocfs2_parse_options(struct super_block *sb,
 		case Opt_journal_async_commit:
 			mopt->mount_opt |= OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT;
 			break;
+		case Opt_nocluster:
+			mopt->mount_opt |= OCFS2_MOUNT_NOCLUSTER;
+			break;
 		default:
 			mlog(ML_ERROR,
 			     "Unrecognized mount option \"%s\" "
@@ -1556,6 +1574,9 @@ static int ocfs2_show_options(struct seq_file *s, struct dentry *root)
 	if (opts & OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT)
 		seq_printf(s, ",journal_async_commit");
 
+	if (opts & OCFS2_MOUNT_NOCLUSTER)
+		seq_printf(s, ",nocluster");
+
 	return 0;
 }
 
diff --git a/fs/omfs/file.c b/fs/omfs/file.c
index d640b9388238..d7b5f09d298c 100644
--- a/fs/omfs/file.c
+++ b/fs/omfs/file.c
@@ -289,10 +289,9 @@ static int omfs_readpage(struct file *file, struct page *page)
 	return block_read_full_page(page, omfs_get_block);
 }
 
-static int omfs_readpages(struct file *file, struct address_space *mapping,
-		struct list_head *pages, unsigned nr_pages)
+static void omfs_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, omfs_get_block);
+	mpage_readahead(rac, omfs_get_block);
 }
 
 static int omfs_writepage(struct page *page, struct writeback_control *wbc)
@@ -373,7 +372,7 @@ const struct inode_operations omfs_file_inops = {
 
 const struct address_space_operations omfs_aops = {
 	.readpage = omfs_readpage,
-	.readpages = omfs_readpages,
+	.readahead = omfs_readahead,
 	.writepage = omfs_writepage,
 	.writepages = omfs_writepages,
 	.write_begin = omfs_write_begin,
diff --git a/fs/open.c b/fs/open.c
index b62f5c0923a8..6cd48a61cda3 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -345,21 +345,14 @@ SYSCALL_DEFINE4(fallocate, int, fd, int, mode, loff_t, offset, loff_t, len)
  * We do this by temporarily clearing all FS-related capabilities and
  * switching the fsuid/fsgid around to the real ones.
  */
-long do_faccessat(int dfd, const char __user *filename, int mode)
+static const struct cred *access_override_creds(void)
 {
 	const struct cred *old_cred;
 	struct cred *override_cred;
-	struct path path;
-	struct inode *inode;
-	int res;
-	unsigned int lookup_flags = LOOKUP_FOLLOW;
-
-	if (mode & ~S_IRWXO)	/* where's F_OK, X_OK, W_OK, R_OK? */
-		return -EINVAL;
 
 	override_cred = prepare_creds();
 	if (!override_cred)
-		return -ENOMEM;
+		return NULL;
 
 	override_cred->fsuid = override_cred->uid;
 	override_cred->fsgid = override_cred->gid;
@@ -394,6 +387,38 @@ long do_faccessat(int dfd, const char __user *filename, int mode)
 	override_cred->non_rcu = 1;
 
 	old_cred = override_creds(override_cred);
+
+	/* override_cred() gets its own ref */
+	put_cred(override_cred);
+
+	return old_cred;
+}
+
+long do_faccessat(int dfd, const char __user *filename, int mode, int flags)
+{
+	struct path path;
+	struct inode *inode;
+	int res;
+	unsigned int lookup_flags = LOOKUP_FOLLOW;
+	const struct cred *old_cred = NULL;
+
+	if (mode & ~S_IRWXO)	/* where's F_OK, X_OK, W_OK, R_OK? */
+		return -EINVAL;
+
+	if (flags & ~(AT_EACCESS | AT_SYMLINK_NOFOLLOW | AT_EMPTY_PATH))
+		return -EINVAL;
+
+	if (flags & AT_SYMLINK_NOFOLLOW)
+		lookup_flags &= ~LOOKUP_FOLLOW;
+	if (flags & AT_EMPTY_PATH)
+		lookup_flags |= LOOKUP_EMPTY;
+
+	if (!(flags & AT_EACCESS)) {
+		old_cred = access_override_creds();
+		if (!old_cred)
+			return -ENOMEM;
+	}
+
 retry:
 	res = user_path_at(dfd, filename, lookup_flags, &path);
 	if (res)
@@ -435,19 +460,26 @@ out_path_release:
 		goto retry;
 	}
 out:
-	revert_creds(old_cred);
-	put_cred(override_cred);
+	if (old_cred)
+		revert_creds(old_cred);
+
 	return res;
 }
 
 SYSCALL_DEFINE3(faccessat, int, dfd, const char __user *, filename, int, mode)
 {
-	return do_faccessat(dfd, filename, mode);
+	return do_faccessat(dfd, filename, mode, 0);
+}
+
+SYSCALL_DEFINE4(faccessat2, int, dfd, const char __user *, filename, int, mode,
+		int, flags)
+{
+	return do_faccessat(dfd, filename, mode, flags);
 }
 
 SYSCALL_DEFINE2(access, const char __user *, filename, int, mode)
 {
-	return do_faccessat(AT_FDCWD, filename, mode);
+	return do_faccessat(AT_FDCWD, filename, mode, 0);
 }
 
 int ksys_chdir(const char __user *filename)
@@ -743,9 +775,8 @@ static int do_dentry_open(struct file *f,
 	path_get(&f->f_path);
 	f->f_inode = inode;
 	f->f_mapping = inode->i_mapping;
-
-	/* Ensure that we skip any errors that predate opening of the file */
 	f->f_wb_err = filemap_sample_wb_err(f->f_mapping);
+	f->f_sb_err = file_sample_sb_err(f);
 
 	if (unlikely(f->f_flags & O_PATH)) {
 		f->f_mode = FMODE_PATH | FMODE_OPENED;
@@ -860,9 +891,6 @@ cleanup_file:
  * the return value of d_splice_alias(), then the caller needs to perform dput()
  * on it after finish_open().
  *
- * On successful return @file is a fully instantiated open file.  After this, if
- * an error occurs in ->atomic_open(), it needs to clean up with fput().
- *
  * Returns zero on success or -errno if the open failed.
  */
 int finish_open(struct file *file, struct dentry *dentry,
@@ -955,48 +983,83 @@ struct file *open_with_fake_path(const struct path *path, int flags,
 }
 EXPORT_SYMBOL(open_with_fake_path);
 
-static inline int build_open_flags(int flags, umode_t mode, struct open_flags *op)
+#define WILL_CREATE(flags)	(flags & (O_CREAT | __O_TMPFILE))
+#define O_PATH_FLAGS		(O_DIRECTORY | O_NOFOLLOW | O_PATH | O_CLOEXEC)
+
+inline struct open_how build_open_how(int flags, umode_t mode)
 {
+	struct open_how how = {
+		.flags = flags & VALID_OPEN_FLAGS,
+		.mode = mode & S_IALLUGO,
+	};
+
+	/* O_PATH beats everything else. */
+	if (how.flags & O_PATH)
+		how.flags &= O_PATH_FLAGS;
+	/* Modes should only be set for create-like flags. */
+	if (!WILL_CREATE(how.flags))
+		how.mode = 0;
+	return how;
+}
+
+inline int build_open_flags(const struct open_how *how, struct open_flags *op)
+{
+	int flags = how->flags;
 	int lookup_flags = 0;
 	int acc_mode = ACC_MODE(flags);
 
+	/* Must never be set by userspace */
+	flags &= ~(FMODE_NONOTIFY | O_CLOEXEC);
+
 	/*
-	 * Clear out all open flags we don't know about so that we don't report
-	 * them in fcntl(F_GETFD) or similar interfaces.
+	 * Older syscalls implicitly clear all of the invalid flags or argument
+	 * values before calling build_open_flags(), but openat2(2) checks all
+	 * of its arguments.
 	 */
-	flags &= VALID_OPEN_FLAGS;
+	if (flags & ~VALID_OPEN_FLAGS)
+		return -EINVAL;
+	if (how->resolve & ~VALID_RESOLVE_FLAGS)
+		return -EINVAL;
 
-	if (flags & (O_CREAT | __O_TMPFILE))
-		op->mode = (mode & S_IALLUGO) | S_IFREG;
-	else
+	/* Deal with the mode. */
+	if (WILL_CREATE(flags)) {
+		if (how->mode & ~S_IALLUGO)
+			return -EINVAL;
+		op->mode = how->mode | S_IFREG;
+	} else {
+		if (how->mode != 0)
+			return -EINVAL;
 		op->mode = 0;
-
-	/* Must never be set by userspace */
-	flags &= ~FMODE_NONOTIFY & ~O_CLOEXEC;
+	}
 
 	/*
-	 * O_SYNC is implemented as __O_SYNC|O_DSYNC.  As many places only
-	 * check for O_DSYNC if the need any syncing at all we enforce it's
-	 * always set instead of having to deal with possibly weird behaviour
-	 * for malicious applications setting only __O_SYNC.
+	 * In order to ensure programs get explicit errors when trying to use
+	 * O_TMPFILE on old kernels, O_TMPFILE is implemented such that it
+	 * looks like (O_DIRECTORY|O_RDWR & ~O_CREAT) to old kernels. But we
+	 * have to require userspace to explicitly set it.
 	 */
-	if (flags & __O_SYNC)
-		flags |= O_DSYNC;
-
 	if (flags & __O_TMPFILE) {
 		if ((flags & O_TMPFILE_MASK) != O_TMPFILE)
 			return -EINVAL;
 		if (!(acc_mode & MAY_WRITE))
 			return -EINVAL;
-	} else if (flags & O_PATH) {
-		/*
-		 * If we have O_PATH in the open flag. Then we
-		 * cannot have anything other than the below set of flags
-		 */
-		flags &= O_DIRECTORY | O_NOFOLLOW | O_PATH;
+	}
+	if (flags & O_PATH) {
+		/* O_PATH only permits certain other flags to be set. */
+		if (flags & ~O_PATH_FLAGS)
+			return -EINVAL;
 		acc_mode = 0;
 	}
 
+	/*
+	 * O_SYNC is implemented as __O_SYNC|O_DSYNC.  As many places only
+	 * check for O_DSYNC if the need any syncing at all we enforce it's
+	 * always set instead of having to deal with possibly weird behaviour
+	 * for malicious applications setting only __O_SYNC.
+	 */
+	if (flags & __O_SYNC)
+		flags |= O_DSYNC;
+
 	op->open_flag = flags;
 
 	/* O_TRUNC implies we need access checks for write permissions */
@@ -1014,14 +1077,28 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o
 
 	if (flags & O_CREAT) {
 		op->intent |= LOOKUP_CREATE;
-		if (flags & O_EXCL)
+		if (flags & O_EXCL) {
 			op->intent |= LOOKUP_EXCL;
+			flags |= O_NOFOLLOW;
+		}
 	}
 
 	if (flags & O_DIRECTORY)
 		lookup_flags |= LOOKUP_DIRECTORY;
 	if (!(flags & O_NOFOLLOW))
 		lookup_flags |= LOOKUP_FOLLOW;
+
+	if (how->resolve & RESOLVE_NO_XDEV)
+		lookup_flags |= LOOKUP_NO_XDEV;
+	if (how->resolve & RESOLVE_NO_MAGICLINKS)
+		lookup_flags |= LOOKUP_NO_MAGICLINKS;
+	if (how->resolve & RESOLVE_NO_SYMLINKS)
+		lookup_flags |= LOOKUP_NO_SYMLINKS;
+	if (how->resolve & RESOLVE_BENEATH)
+		lookup_flags |= LOOKUP_BENEATH;
+	if (how->resolve & RESOLVE_IN_ROOT)
+		lookup_flags |= LOOKUP_IN_ROOT;
+
 	op->lookup_flags = lookup_flags;
 	return 0;
 }
@@ -1040,8 +1117,11 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o
 struct file *file_open_name(struct filename *name, int flags, umode_t mode)
 {
 	struct open_flags op;
-	int err = build_open_flags(flags, mode, &op);
-	return err ? ERR_PTR(err) : do_filp_open(AT_FDCWD, name, &op);
+	struct open_how how = build_open_how(flags, mode);
+	int err = build_open_flags(&how, &op);
+	if (err)
+		return ERR_PTR(err);
+	return do_filp_open(AT_FDCWD, name, &op);
 }
 
 /**
@@ -1072,17 +1152,19 @@ struct file *file_open_root(struct dentry *dentry, struct vfsmount *mnt,
 			    const char *filename, int flags, umode_t mode)
 {
 	struct open_flags op;
-	int err = build_open_flags(flags, mode, &op);
+	struct open_how how = build_open_how(flags, mode);
+	int err = build_open_flags(&how, &op);
 	if (err)
 		return ERR_PTR(err);
 	return do_file_open_root(dentry, mnt, filename, &op);
 }
 EXPORT_SYMBOL(file_open_root);
 
-long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode)
+static long do_sys_openat2(int dfd, const char __user *filename,
+			   struct open_how *how)
 {
 	struct open_flags op;
-	int fd = build_open_flags(flags, mode, &op);
+	int fd = build_open_flags(how, &op);
 	struct filename *tmp;
 
 	if (fd)
@@ -1092,7 +1174,7 @@ long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode)
 	if (IS_ERR(tmp))
 		return PTR_ERR(tmp);
 
-	fd = get_unused_fd_flags(flags);
+	fd = get_unused_fd_flags(how->flags);
 	if (fd >= 0) {
 		struct file *f = do_filp_open(dfd, tmp, &op);
 		if (IS_ERR(f)) {
@@ -1107,12 +1189,16 @@ long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode)
 	return fd;
 }
 
-SYSCALL_DEFINE3(open, const char __user *, filename, int, flags, umode_t, mode)
+long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode)
 {
-	if (force_o_largefile())
-		flags |= O_LARGEFILE;
+	struct open_how how = build_open_how(flags, mode);
+	return do_sys_openat2(dfd, filename, &how);
+}
 
-	return do_sys_open(AT_FDCWD, filename, flags, mode);
+
+SYSCALL_DEFINE3(open, const char __user *, filename, int, flags, umode_t, mode)
+{
+	return ksys_open(filename, flags, mode);
 }
 
 SYSCALL_DEFINE4(openat, int, dfd, const char __user *, filename, int, flags,
@@ -1120,10 +1206,32 @@ SYSCALL_DEFINE4(openat, int, dfd, const char __user *, filename, int, flags,
 {
 	if (force_o_largefile())
 		flags |= O_LARGEFILE;
-
 	return do_sys_open(dfd, filename, flags, mode);
 }
 
+SYSCALL_DEFINE4(openat2, int, dfd, const char __user *, filename,
+		struct open_how __user *, how, size_t, usize)
+{
+	int err;
+	struct open_how tmp;
+
+	BUILD_BUG_ON(sizeof(struct open_how) < OPEN_HOW_SIZE_VER0);
+	BUILD_BUG_ON(sizeof(struct open_how) != OPEN_HOW_SIZE_LATEST);
+
+	if (unlikely(usize < OPEN_HOW_SIZE_VER0))
+		return -EINVAL;
+
+	err = copy_struct_from_user(&tmp, sizeof(tmp), how, usize);
+	if (err)
+		return err;
+
+	/* O_LARGEFILE is only allowed for non-O_PATH. */
+	if (!(tmp.flags & O_PATH) && force_o_largefile())
+		tmp.flags |= O_LARGEFILE;
+
+	return do_sys_openat2(dfd, filename, &tmp);
+}
+
 #ifdef CONFIG_COMPAT
 /*
  * Exactly like sys_open(), except that it doesn't set the
diff --git a/fs/orangefs/file.c b/fs/orangefs/file.c
index a5612abc0936..af375e049aae 100644
--- a/fs/orangefs/file.c
+++ b/fs/orangefs/file.c
@@ -46,8 +46,9 @@ static int flush_racache(struct inode *inode)
  * Post and wait for the I/O upcall to finish
  */
 ssize_t wait_for_direct_io(enum ORANGEFS_io_type type, struct inode *inode,
-    loff_t *offset, struct iov_iter *iter, size_t total_size,
-    loff_t readahead_size, struct orangefs_write_range *wr, int *index_return)
+	loff_t *offset, struct iov_iter *iter, size_t total_size,
+	loff_t readahead_size, struct orangefs_write_range *wr,
+	int *index_return, struct file *file)
 {
 	struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
 	struct orangefs_khandle *handle = &orangefs_inode->refn.khandle;
@@ -55,6 +56,8 @@ ssize_t wait_for_direct_io(enum ORANGEFS_io_type type, struct inode *inode,
 	int buffer_index;
 	ssize_t ret;
 	size_t copy_amount;
+	int open_for_read;
+	int open_for_write;
 
 	new_op = op_alloc(ORANGEFS_VFS_OP_FILE_IO);
 	if (!new_op)
@@ -90,6 +93,38 @@ populate_shared_memory:
 		new_op->upcall.uid = from_kuid(&init_user_ns, wr->uid);
 		new_op->upcall.gid = from_kgid(&init_user_ns, wr->gid);
 	}
+	/*
+	 * Orangefs has no open, and orangefs checks file permissions
+	 * on each file access. Posix requires that file permissions
+	 * be checked on open and nowhere else. Orangefs-through-the-kernel
+	 * needs to seem posix compliant.
+	 *
+	 * The VFS opens files, even if the filesystem provides no
+	 * method. We can see if a file was successfully opened for
+	 * read and or for write by looking at file->f_mode.
+	 *
+	 * When writes are flowing from the page cache, file is no
+	 * longer available. We can trust the VFS to have checked
+	 * file->f_mode before writing to the page cache.
+	 *
+	 * The mode of a file might change between when it is opened
+	 * and IO commences, or it might be created with an arbitrary mode.
+	 *
+	 * We'll make sure we don't hit EACCES during the IO stage by
+	 * using UID 0. Some of the time we have access without changing
+	 * to UID 0 - how to check?
+	 */
+	if (file) {
+		open_for_write = file->f_mode & FMODE_WRITE;
+		open_for_read = file->f_mode & FMODE_READ;
+	} else {
+		open_for_write = 1;
+		open_for_read = 0; /* not relevant? */
+	}
+	if ((type == ORANGEFS_IO_WRITE) && open_for_write)
+		new_op->upcall.uid = 0;
+	if ((type == ORANGEFS_IO_READ) && open_for_read)
+		new_op->upcall.uid = 0;
 
 	gossip_debug(GOSSIP_FILE_DEBUG,
 		     "%s(%pU): offset: %llu total_size: %zd\n",
@@ -311,23 +346,8 @@ static ssize_t orangefs_file_read_iter(struct kiocb *iocb,
     struct iov_iter *iter)
 {
 	int ret;
-	struct orangefs_read_options *ro;
-
 	orangefs_stats.reads++;
 
-	/*
-	 * Remember how they set "count" in read(2) or pread(2) or whatever -
-	 * users can use count as a knob to control orangefs io size and later
-	 * we can try to help them fill as many pages as possible in readpage.
-	 */
-	if (!iocb->ki_filp->private_data) {
-		iocb->ki_filp->private_data = kmalloc(sizeof *ro, GFP_KERNEL);
-		if (!iocb->ki_filp->private_data)
-			return(ENOMEM);
-		ro = iocb->ki_filp->private_data;
-		ro->blksiz = iter->count;
-	}
-
 	down_read(&file_inode(iocb->ki_filp)->i_rwsem);
 	ret = orangefs_revalidate_mapping(file_inode(iocb->ki_filp));
 	if (ret)
@@ -615,12 +635,6 @@ static int orangefs_lock(struct file *filp, int cmd, struct file_lock *fl)
 	return rc;
 }
 
-static int orangefs_file_open(struct inode * inode, struct file *file)
-{
-	file->private_data = NULL;
-	return generic_file_open(inode, file);
-}
-
 static int orangefs_flush(struct file *file, fl_owner_t id)
 {
 	/*
@@ -631,19 +645,8 @@ static int orangefs_flush(struct file *file, fl_owner_t id)
 	 * on an explicit fsync call.  This duplicates historical OrangeFS
 	 * behavior.
 	 */
-	struct inode *inode = file->f_mapping->host;
 	int r;
 
-	kfree(file->private_data);
-	file->private_data = NULL;
-
-	if (inode->i_state & I_DIRTY_TIME) {
-		spin_lock(&inode->i_lock);
-		inode->i_state &= ~I_DIRTY_TIME;
-		spin_unlock(&inode->i_lock);
-		mark_inode_dirty_sync(inode);
-	}
-
 	r = filemap_write_and_wait_range(file->f_mapping, 0, LLONG_MAX);
 	if (r > 0)
 		return 0;
@@ -659,7 +662,7 @@ const struct file_operations orangefs_file_operations = {
 	.lock		= orangefs_lock,
 	.unlocked_ioctl	= orangefs_ioctl,
 	.mmap		= orangefs_file_mmap,
-	.open		= orangefs_file_open,
+	.open		= generic_file_open,
 	.flush		= orangefs_flush,
 	.release	= orangefs_file_release,
 	.fsync		= orangefs_fsync,
diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c
index efb12197da18..48f0547d4850 100644
--- a/fs/orangefs/inode.c
+++ b/fs/orangefs/inode.c
@@ -55,19 +55,14 @@ static int orangefs_writepage_locked(struct page *page,
 	iov_iter_bvec(&iter, WRITE, &bv, 1, wlen);
 
 	ret = wait_for_direct_io(ORANGEFS_IO_WRITE, inode, &off, &iter, wlen,
-	    len, wr, NULL);
+	    len, wr, NULL, NULL);
 	if (ret < 0) {
 		SetPageError(page);
 		mapping_set_error(page->mapping, ret);
 	} else {
 		ret = 0;
 	}
-	if (wr) {
-		kfree(wr);
-		set_page_private(page, 0);
-		ClearPagePrivate(page);
-		put_page(page);
-	}
+	kfree(detach_page_private(page));
 	return ret;
 }
 
@@ -126,7 +121,7 @@ static int orangefs_writepages_work(struct orangefs_writepages *ow,
 	wr.uid = ow->uid;
 	wr.gid = ow->gid;
 	ret = wait_for_direct_io(ORANGEFS_IO_WRITE, inode, &off, &iter, ow->len,
-	    0, &wr, NULL);
+	    0, &wr, NULL, NULL);
 	if (ret < 0) {
 		for (i = 0; i < ow->npages; i++) {
 			SetPageError(ow->pages[i]);
@@ -259,46 +254,19 @@ static int orangefs_readpage(struct file *file, struct page *page)
 	pgoff_t index; /* which page */
 	struct page *next_page;
 	char *kaddr;
-	struct orangefs_read_options *ro = file->private_data;
 	loff_t read_size;
-	loff_t roundedup;
 	int buffer_index = -1; /* orangefs shared memory slot */
 	int slot_index;   /* index into slot */
 	int remaining;
 
 	/*
-	 * If they set some miniscule size for "count" in read(2)
-	 * (for example) then let's try to read a page, or the whole file
-	 * if it is smaller than a page. Once "count" goes over a page
-	 * then lets round up to the highest page size multiple that is
-	 * less than or equal to "count" and do that much orangefs IO and
-	 * try to fill as many pages as we can from it.
-	 *
-	 * "count" should be represented in ro->blksiz.
-	 *
-	 * inode->i_size = file size.
+	 * Get up to this many bytes from Orangefs at a time and try
+	 * to fill them into the page cache at once. Tests with dd made
+	 * this seem like a reasonable static number, if there was
+	 * interest perhaps this number could be made setable through
+	 * sysfs...
 	 */
-	if (ro) {
-		if (ro->blksiz < PAGE_SIZE) {
-			if (inode->i_size < PAGE_SIZE)
-				read_size = inode->i_size;
-			else
-				read_size = PAGE_SIZE;
-		} else {
-			roundedup = ((PAGE_SIZE - 1) & ro->blksiz) ?
-				((ro->blksiz + PAGE_SIZE) & ~(PAGE_SIZE -1)) :
-				ro->blksiz;
-			if (roundedup > inode->i_size)
-				read_size = inode->i_size;
-			else
-				read_size = roundedup;
-
-		}
-	} else {
-		read_size = PAGE_SIZE;
-	}
-	if (!read_size)
-		read_size = PAGE_SIZE;
+	read_size = 524288;
 
 	if (PageDirty(page))
 		orangefs_launder_page(page);
@@ -311,7 +279,7 @@ static int orangefs_readpage(struct file *file, struct page *page)
 	iov_iter_bvec(&iter, READ, &bv, 1, PAGE_SIZE);
 
 	ret = wait_for_direct_io(ORANGEFS_IO_READ, inode, &off, &iter,
-	    read_size, inode->i_size, NULL, &buffer_index);
+	    read_size, inode->i_size, NULL, &buffer_index, file);
 	remaining = ret;
 	/* this will only zero remaining unread portions of the page data */
 	iov_iter_zero(~0U, &iter);
@@ -436,9 +404,7 @@ static int orangefs_write_begin(struct file *file,
 	wr->len = len;
 	wr->uid = current_fsuid();
 	wr->gid = current_fsgid();
-	SetPagePrivate(page);
-	set_page_private(page, (unsigned long)wr);
-	get_page(page);
+	attach_page_private(page, wr);
 okay:
 	return 0;
 }
@@ -486,18 +452,12 @@ static void orangefs_invalidatepage(struct page *page,
 	wr = (struct orangefs_write_range *)page_private(page);
 
 	if (offset == 0 && length == PAGE_SIZE) {
-		kfree((struct orangefs_write_range *)page_private(page));
-		set_page_private(page, 0);
-		ClearPagePrivate(page);
-		put_page(page);
+		kfree(detach_page_private(page));
 		return;
 	/* write range entirely within invalidate range (or equal) */
 	} else if (page_offset(page) + offset <= wr->pos &&
 	    wr->pos + wr->len <= page_offset(page) + offset + length) {
-		kfree((struct orangefs_write_range *)page_private(page));
-		set_page_private(page, 0);
-		ClearPagePrivate(page);
-		put_page(page);
+		kfree(detach_page_private(page));
 		/* XXX is this right? only caller in fs */
 		cancel_dirty_page(page);
 		return;
@@ -562,12 +522,7 @@ static int orangefs_releasepage(struct page *page, gfp_t foo)
 
 static void orangefs_freepage(struct page *page)
 {
-	if (PagePrivate(page)) {
-		kfree((struct orangefs_write_range *)page_private(page));
-		set_page_private(page, 0);
-		ClearPagePrivate(page);
-		put_page(page);
-	}
+	kfree(detach_page_private(page));
 }
 
 static int orangefs_launder_page(struct page *page)
@@ -651,7 +606,7 @@ static ssize_t orangefs_direct_IO(struct kiocb *iocb,
 			     (int)*offset);
 
 		ret = wait_for_direct_io(type, inode, offset, iter,
-				each_count, 0, NULL, NULL);
+				each_count, 0, NULL, NULL, file);
 		gossip_debug(GOSSIP_FILE_DEBUG,
 			     "%s(%pU): return from wait_for_io:%d\n",
 			     __func__,
@@ -767,9 +722,7 @@ vm_fault_t orangefs_page_mkwrite(struct vm_fault *vmf)
 	wr->len = PAGE_SIZE;
 	wr->uid = current_fsuid();
 	wr->gid = current_fsgid();
-	SetPagePrivate(page);
-	set_page_private(page, (unsigned long)wr);
-	get_page(page);
+	attach_page_private(page, wr);
 okay:
 
 	file_update_time(vmf->vma->vm_file);
diff --git a/fs/orangefs/orangefs-debugfs.c b/fs/orangefs/orangefs-debugfs.c
index 25543a966c48..29eaa4544372 100644
--- a/fs/orangefs/orangefs-debugfs.c
+++ b/fs/orangefs/orangefs-debugfs.c
@@ -273,6 +273,7 @@ static void *help_start(struct seq_file *m, loff_t *pos)
 
 static void *help_next(struct seq_file *m, void *v, loff_t *pos)
 {
+	(*pos)++;
 	gossip_debug(GOSSIP_DEBUGFS_DEBUG, "help_next: start\n");
 
 	return NULL;
diff --git a/fs/orangefs/orangefs-kernel.h b/fs/orangefs/orangefs-kernel.h
index 34a6c99fa29b..e12aeb9623d6 100644
--- a/fs/orangefs/orangefs-kernel.h
+++ b/fs/orangefs/orangefs-kernel.h
@@ -239,10 +239,6 @@ struct orangefs_write_range {
 	kgid_t gid;
 };
 
-struct orangefs_read_options {
-	ssize_t blksiz;
-};
-
 extern struct orangefs_stats orangefs_stats;
 
 /*
@@ -398,7 +394,8 @@ bool __is_daemon_in_service(void);
  */
 int orangefs_revalidate_mapping(struct inode *);
 ssize_t wait_for_direct_io(enum ORANGEFS_io_type, struct inode *, loff_t *,
-    struct iov_iter *, size_t, loff_t, struct orangefs_write_range *, int *);
+    struct iov_iter *, size_t, loff_t, struct orangefs_write_range *, int *,
+    struct file *);
 ssize_t do_readv_writev(enum ORANGEFS_io_type, struct file *, loff_t *,
     struct iov_iter *);
 
diff --git a/fs/overlayfs/Kconfig b/fs/overlayfs/Kconfig
index 444e2da4f60e..dd188c7996b3 100644
--- a/fs/overlayfs/Kconfig
+++ b/fs/overlayfs/Kconfig
@@ -9,7 +9,7 @@ config OVERLAY_FS
 	  'lower' filesystem is either hidden or, in the case of directories,
 	  merged with the 'upper' object.
 
-	  For more information see Documentation/filesystems/overlayfs.txt
+	  For more information see Documentation/filesystems/overlayfs.rst
 
 config OVERLAY_FS_REDIRECT_DIR
 	bool "Overlayfs: turn on redirect directory feature by default"
@@ -38,7 +38,7 @@ config OVERLAY_FS_REDIRECT_ALWAYS_FOLLOW
 	  If backward compatibility is not an issue, then it is safe and
 	  recommended to say N here.
 
-	  For more information, see Documentation/filesystems/overlayfs.txt
+	  For more information, see Documentation/filesystems/overlayfs.rst
 
 	  If unsure, say Y.
 
@@ -93,6 +93,7 @@ config OVERLAY_FS_XINO_AUTO
 	bool "Overlayfs: auto enable inode number mapping"
 	default n
 	depends on OVERLAY_FS
+	depends on 64BIT
 	help
 	  If this config option is enabled then overlay filesystems will use
 	  unused high bits in undelying filesystem inode numbers to map all
@@ -102,7 +103,7 @@ config OVERLAY_FS_XINO_AUTO
 	  If compatibility with applications that expect 32bit inodes is not an
 	  issue, then it is safe and recommended to say Y here.
 
-	  For more information, see Documentation/filesystems/overlayfs.txt
+	  For more information, see Documentation/filesystems/overlayfs.rst
 
 	  If unsure, say N.
 
diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c
index b801c6353100..9709cf22cab3 100644
--- a/fs/overlayfs/copy_up.c
+++ b/fs/overlayfs/copy_up.c
@@ -24,7 +24,7 @@
 
 static int ovl_ccup_set(const char *buf, const struct kernel_param *param)
 {
-	pr_warn("overlayfs: \"check_copy_up\" module option is obsolete\n");
+	pr_warn("\"check_copy_up\" module option is obsolete\n");
 	return 0;
 }
 
@@ -36,6 +36,13 @@ static int ovl_ccup_get(char *buf, const struct kernel_param *param)
 module_param_call(check_copy_up, ovl_ccup_set, ovl_ccup_get, NULL, 0644);
 MODULE_PARM_DESC(check_copy_up, "Obsolete; does nothing");
 
+static bool ovl_must_copy_xattr(const char *name)
+{
+	return !strcmp(name, XATTR_POSIX_ACL_ACCESS) ||
+	       !strcmp(name, XATTR_POSIX_ACL_DEFAULT) ||
+	       !strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN);
+}
+
 int ovl_copy_xattr(struct dentry *old, struct dentry *new)
 {
 	ssize_t list_size, size, value_size = 0;
@@ -107,8 +114,13 @@ retry:
 			continue; /* Discard */
 		}
 		error = vfs_setxattr(new, name, value, size, 0);
-		if (error)
-			break;
+		if (error) {
+			if (error != -EOPNOTSUPP || ovl_must_copy_xattr(name))
+				break;
+
+			/* Ignore failure to copy unknown xattrs */
+			error = 0;
+		}
 	}
 	kfree(value);
 out:
@@ -123,6 +135,9 @@ static int ovl_copy_up_data(struct path *old, struct path *new, loff_t len)
 	loff_t old_pos = 0;
 	loff_t new_pos = 0;
 	loff_t cloned;
+	loff_t data_pos = -1;
+	loff_t hole_len;
+	bool skip_hole = false;
 	int error = 0;
 
 	if (len == 0)
@@ -144,7 +159,11 @@ static int ovl_copy_up_data(struct path *old, struct path *new, loff_t len)
 		goto out;
 	/* Couldn't clone, so now we try to copy the data */
 
-	/* FIXME: copy up sparse files efficiently */
+	/* Check if lower fs supports seek operation */
+	if (old_file->f_mode & FMODE_LSEEK &&
+	    old_file->f_op->llseek)
+		skip_hole = true;
+
 	while (len) {
 		size_t this_len = OVL_COPY_UP_CHUNK_SIZE;
 		long bytes;
@@ -157,6 +176,36 @@ static int ovl_copy_up_data(struct path *old, struct path *new, loff_t len)
 			break;
 		}
 
+		/*
+		 * Fill zero for hole will cost unnecessary disk space
+		 * and meanwhile slow down the copy-up speed, so we do
+		 * an optimization for hole during copy-up, it relies
+		 * on SEEK_DATA implementation in lower fs so if lower
+		 * fs does not support it, copy-up will behave as before.
+		 *
+		 * Detail logic of hole detection as below:
+		 * When we detect next data position is larger than current
+		 * position we will skip that hole, otherwise we copy
+		 * data in the size of OVL_COPY_UP_CHUNK_SIZE. Actually,
+		 * it may not recognize all kind of holes and sometimes
+		 * only skips partial of hole area. However, it will be
+		 * enough for most of the use cases.
+		 */
+
+		if (skip_hole && data_pos < old_pos) {
+			data_pos = vfs_llseek(old_file, old_pos, SEEK_DATA);
+			if (data_pos > old_pos) {
+				hole_len = data_pos - old_pos;
+				len -= hole_len;
+				old_pos = new_pos = data_pos;
+				continue;
+			} else if (data_pos == -ENXIO) {
+				break;
+			} else if (data_pos < 0) {
+				skip_hole = false;
+			}
+		}
+
 		bytes = do_splice_direct(old_file, &old_pos,
 					 new_file, &new_pos,
 					 this_len, SPLICE_F_MOVE);
@@ -227,13 +276,17 @@ int ovl_set_attr(struct dentry *upperdentry, struct kstat *stat)
 struct ovl_fh *ovl_encode_real_fh(struct dentry *real, bool is_upper)
 {
 	struct ovl_fh *fh;
-	int fh_type, fh_len, dwords;
-	void *buf;
+	int fh_type, dwords;
 	int buflen = MAX_HANDLE_SZ;
 	uuid_t *uuid = &real->d_sb->s_uuid;
+	int err;
 
-	buf = kmalloc(buflen, GFP_KERNEL);
-	if (!buf)
+	/* Make sure the real fid stays 32bit aligned */
+	BUILD_BUG_ON(OVL_FH_FID_OFFSET % 4);
+	BUILD_BUG_ON(MAX_HANDLE_SZ + OVL_FH_FID_OFFSET > 255);
+
+	fh = kzalloc(buflen + OVL_FH_FID_OFFSET, GFP_KERNEL);
+	if (!fh)
 		return ERR_PTR(-ENOMEM);
 
 	/*
@@ -242,27 +295,19 @@ struct ovl_fh *ovl_encode_real_fh(struct dentry *real, bool is_upper)
 	 * the price or reconnecting the dentry.
 	 */
 	dwords = buflen >> 2;
-	fh_type = exportfs_encode_fh(real, buf, &dwords, 0);
+	fh_type = exportfs_encode_fh(real, (void *)fh->fb.fid, &dwords, 0);
 	buflen = (dwords << 2);
 
-	fh = ERR_PTR(-EIO);
+	err = -EIO;
 	if (WARN_ON(fh_type < 0) ||
 	    WARN_ON(buflen > MAX_HANDLE_SZ) ||
 	    WARN_ON(fh_type == FILEID_INVALID))
-		goto out;
+		goto out_err;
 
-	BUILD_BUG_ON(MAX_HANDLE_SZ + offsetof(struct ovl_fh, fid) > 255);
-	fh_len = offsetof(struct ovl_fh, fid) + buflen;
-	fh = kmalloc(fh_len, GFP_KERNEL);
-	if (!fh) {
-		fh = ERR_PTR(-ENOMEM);
-		goto out;
-	}
-
-	fh->version = OVL_FH_VERSION;
-	fh->magic = OVL_FH_MAGIC;
-	fh->type = fh_type;
-	fh->flags = OVL_FH_FLAG_CPU_ENDIAN;
+	fh->fb.version = OVL_FH_VERSION;
+	fh->fb.magic = OVL_FH_MAGIC;
+	fh->fb.type = fh_type;
+	fh->fb.flags = OVL_FH_FLAG_CPU_ENDIAN;
 	/*
 	 * When we will want to decode an overlay dentry from this handle
 	 * and all layers are on the same fs, if we get a disconncted real
@@ -270,14 +315,15 @@ struct ovl_fh *ovl_encode_real_fh(struct dentry *real, bool is_upper)
 	 * it to upperdentry or to lowerstack is by checking this flag.
 	 */
 	if (is_upper)
-		fh->flags |= OVL_FH_FLAG_PATH_UPPER;
-	fh->len = fh_len;
-	fh->uuid = *uuid;
-	memcpy(fh->fid, buf, buflen);
+		fh->fb.flags |= OVL_FH_FLAG_PATH_UPPER;
+	fh->fb.len = sizeof(fh->fb) + buflen;
+	fh->fb.uuid = *uuid;
 
-out:
-	kfree(buf);
 	return fh;
+
+out_err:
+	kfree(fh);
+	return ERR_PTR(err);
 }
 
 int ovl_set_origin(struct dentry *dentry, struct dentry *lower,
@@ -300,8 +346,8 @@ int ovl_set_origin(struct dentry *dentry, struct dentry *lower,
 	/*
 	 * Do not fail when upper doesn't support xattrs.
 	 */
-	err = ovl_check_setxattr(dentry, upper, OVL_XATTR_ORIGIN, fh,
-				 fh ? fh->len : 0, 0);
+	err = ovl_check_setxattr(dentry, upper, OVL_XATTR_ORIGIN, fh->buf,
+				 fh ? fh->fb.len : 0, 0);
 	kfree(fh);
 
 	return err;
@@ -317,7 +363,7 @@ static int ovl_set_upper_fh(struct dentry *upper, struct dentry *index)
 	if (IS_ERR(fh))
 		return PTR_ERR(fh);
 
-	err = ovl_do_setxattr(index, OVL_XATTR_UPPER, fh, fh->len, 0);
+	err = ovl_do_setxattr(index, OVL_XATTR_UPPER, fh->buf, fh->fb.len, 0);
 
 	kfree(fh);
 	return err;
@@ -483,7 +529,7 @@ static int ovl_copy_up_inode(struct ovl_copy_up_ctx *c, struct dentry *temp)
 	}
 
 	inode_lock(temp->d_inode);
-	if (c->metacopy)
+	if (S_ISREG(c->stat.mode))
 		err = ovl_set_size(temp, &c->stat);
 	if (!err)
 		err = ovl_set_attr(temp, &c->stat);
diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c
index 702aa63f6774..279009dee366 100644
--- a/fs/overlayfs/dir.c
+++ b/fs/overlayfs/dir.c
@@ -35,14 +35,14 @@ int ovl_cleanup(struct inode *wdir, struct dentry *wdentry)
 	dput(wdentry);
 
 	if (err) {
-		pr_err("overlayfs: cleanup of '%pd2' failed (%i)\n",
+		pr_err("cleanup of '%pd2' failed (%i)\n",
 		       wdentry, err);
 	}
 
 	return err;
 }
 
-static struct dentry *ovl_lookup_temp(struct dentry *workdir)
+struct dentry *ovl_lookup_temp(struct dentry *workdir)
 {
 	struct dentry *temp;
 	char name[20];
@@ -53,7 +53,7 @@ static struct dentry *ovl_lookup_temp(struct dentry *workdir)
 
 	temp = lookup_one_len(name, workdir, strlen(name));
 	if (!IS_ERR(temp) && temp->d_inode) {
-		pr_err("overlayfs: workdir/%s already exists\n", name);
+		pr_err("workdir/%s already exists\n", name);
 		dput(temp);
 		temp = ERR_PTR(-EIO);
 	}
@@ -134,7 +134,7 @@ static int ovl_mkdir_real(struct inode *dir, struct dentry **newdentry,
 	d = lookup_one_len(dentry->d_name.name, dentry->d_parent,
 			   dentry->d_name.len);
 	if (IS_ERR(d)) {
-		pr_warn("overlayfs: failed lookup after mkdir (%pd2, err=%i).\n",
+		pr_warn("failed lookup after mkdir (%pd2, err=%i).\n",
 			dentry, err);
 		return PTR_ERR(d);
 	}
@@ -243,6 +243,9 @@ static int ovl_instantiate(struct dentry *dentry, struct inode *inode,
 
 	ovl_dir_modified(dentry->d_parent, false);
 	ovl_dentry_set_upper_alias(dentry);
+	ovl_dentry_update_reval(dentry, newdentry,
+			DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
+
 	if (!hardlink) {
 		/*
 		 * ovl_obtain_alias() can be called after ovl_create_real()
@@ -267,7 +270,7 @@ static int ovl_instantiate(struct dentry *dentry, struct inode *inode,
 
 	d_instantiate(dentry, inode);
 	if (inode != oip.newinode) {
-		pr_warn_ratelimited("overlayfs: newly created inode found in cache (%pd2)\n",
+		pr_warn_ratelimited("newly created inode found in cache (%pd2)\n",
 				    dentry);
 	}
 
@@ -819,6 +822,28 @@ static bool ovl_pure_upper(struct dentry *dentry)
 	       !ovl_test_flag(OVL_WHITEOUTS, d_inode(dentry));
 }
 
+static void ovl_drop_nlink(struct dentry *dentry)
+{
+	struct inode *inode = d_inode(dentry);
+	struct dentry *alias;
+
+	/* Try to find another, hashed alias */
+	spin_lock(&inode->i_lock);
+	hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) {
+		if (alias != dentry && !d_unhashed(alias))
+			break;
+	}
+	spin_unlock(&inode->i_lock);
+
+	/*
+	 * Changes to underlying layers may cause i_nlink to lose sync with
+	 * reality.  In this case prevent the link count from going to zero
+	 * prematurely.
+	 */
+	if (inode->i_nlink > !!alias)
+		drop_nlink(inode);
+}
+
 static int ovl_do_remove(struct dentry *dentry, bool is_dir)
 {
 	int err;
@@ -856,7 +881,7 @@ static int ovl_do_remove(struct dentry *dentry, bool is_dir)
 		if (is_dir)
 			clear_nlink(dentry->d_inode);
 		else
-			drop_nlink(dentry->d_inode);
+			ovl_drop_nlink(dentry);
 	}
 	ovl_nlink_end(dentry);
 
@@ -1009,7 +1034,7 @@ static int ovl_set_redirect(struct dentry *dentry, bool samedir)
 		spin_unlock(&dentry->d_lock);
 	} else {
 		kfree(redirect);
-		pr_warn_ratelimited("overlayfs: failed to set redirect (%i)\n",
+		pr_warn_ratelimited("failed to set redirect (%i)\n",
 				    err);
 		/* Fall back to userspace copy-up */
 		err = -EXDEV;
@@ -1170,7 +1195,7 @@ static int ovl_rename(struct inode *olddir, struct dentry *old,
 	if (newdentry == trap)
 		goto out_dput;
 
-	if (WARN_ON(olddentry->d_inode == newdentry->d_inode))
+	if (olddentry->d_inode == newdentry->d_inode)
 		goto out_dput;
 
 	err = 0;
@@ -1201,7 +1226,7 @@ static int ovl_rename(struct inode *olddir, struct dentry *old,
 		if (new_is_dir)
 			clear_nlink(d_inode(new));
 		else
-			drop_nlink(d_inode(new));
+			ovl_drop_nlink(new);
 	}
 
 	ovl_dir_modified(old->d_parent, ovl_type_origin(old) ||
diff --git a/fs/overlayfs/export.c b/fs/overlayfs/export.c
index 73c9775215b3..ed5c1078919c 100644
--- a/fs/overlayfs/export.c
+++ b/fs/overlayfs/export.c
@@ -30,7 +30,7 @@ static int ovl_encode_maybe_copy_up(struct dentry *dentry)
 	}
 
 	if (err) {
-		pr_warn_ratelimited("overlayfs: failed to copy up on encode (%pd2, err=%i)\n",
+		pr_warn_ratelimited("failed to copy up on encode (%pd2, err=%i)\n",
 				    dentry, err);
 	}
 
@@ -211,10 +211,11 @@ static int ovl_check_encode_origin(struct dentry *dentry)
 	return 1;
 }
 
-static int ovl_d_to_fh(struct dentry *dentry, char *buf, int buflen)
+static int ovl_dentry_to_fid(struct dentry *dentry, u32 *fid, int buflen)
 {
 	struct ovl_fh *fh = NULL;
 	int err, enc_lower;
+	int len;
 
 	/*
 	 * Check if we should encode a lower or upper file handle and maybe
@@ -231,43 +232,29 @@ static int ovl_d_to_fh(struct dentry *dentry, char *buf, int buflen)
 		return PTR_ERR(fh);
 
 	err = -EOVERFLOW;
-	if (fh->len > buflen)
+	len = OVL_FH_LEN(fh);
+	if (len > buflen)
 		goto fail;
 
-	memcpy(buf, (char *)fh, fh->len);
-	err = fh->len;
+	memcpy(fid, fh, len);
+	err = len;
 
 out:
 	kfree(fh);
 	return err;
 
 fail:
-	pr_warn_ratelimited("overlayfs: failed to encode file handle (%pd2, err=%i, buflen=%d, len=%d, type=%d)\n",
-			    dentry, err, buflen, fh ? (int)fh->len : 0,
-			    fh ? fh->type : 0);
+	pr_warn_ratelimited("failed to encode file handle (%pd2, err=%i, buflen=%d, len=%d, type=%d)\n",
+			    dentry, err, buflen, fh ? (int)fh->fb.len : 0,
+			    fh ? fh->fb.type : 0);
 	goto out;
 }
 
-static int ovl_dentry_to_fh(struct dentry *dentry, u32 *fid, int *max_len)
-{
-	int res, len = *max_len << 2;
-
-	res = ovl_d_to_fh(dentry, (char *)fid, len);
-	if (res <= 0)
-		return FILEID_INVALID;
-
-	len = res;
-
-	/* Round up to dwords */
-	*max_len = (len + 3) >> 2;
-	return OVL_FILEID;
-}
-
 static int ovl_encode_fh(struct inode *inode, u32 *fid, int *max_len,
 			 struct inode *parent)
 {
 	struct dentry *dentry;
-	int type;
+	int bytes = *max_len << 2;
 
 	/* TODO: encode connectable file handles */
 	if (parent)
@@ -277,10 +264,14 @@ static int ovl_encode_fh(struct inode *inode, u32 *fid, int *max_len,
 	if (WARN_ON(!dentry))
 		return FILEID_INVALID;
 
-	type = ovl_dentry_to_fh(dentry, fid, max_len);
-
+	bytes = ovl_dentry_to_fid(dentry, fid, bytes);
 	dput(dentry);
-	return type;
+	if (bytes <= 0)
+		return FILEID_INVALID;
+
+	*max_len = bytes >> 2;
+
+	return OVL_FILEID_V1;
 }
 
 /*
@@ -317,29 +308,35 @@ static struct dentry *ovl_obtain_alias(struct super_block *sb,
 		ovl_set_flag(OVL_UPPERDATA, inode);
 
 	dentry = d_find_any_alias(inode);
-	if (!dentry) {
-		dentry = d_alloc_anon(inode->i_sb);
-		if (!dentry)
-			goto nomem;
-		oe = ovl_alloc_entry(lower ? 1 : 0);
-		if (!oe)
-			goto nomem;
-
-		if (lower) {
-			oe->lowerstack->dentry = dget(lower);
-			oe->lowerstack->layer = lowerpath->layer;
-		}
-		dentry->d_fsdata = oe;
-		if (upper_alias)
-			ovl_dentry_set_upper_alias(dentry);
+	if (dentry)
+		goto out_iput;
+
+	dentry = d_alloc_anon(inode->i_sb);
+	if (unlikely(!dentry))
+		goto nomem;
+	oe = ovl_alloc_entry(lower ? 1 : 0);
+	if (!oe)
+		goto nomem;
+
+	if (lower) {
+		oe->lowerstack->dentry = dget(lower);
+		oe->lowerstack->layer = lowerpath->layer;
 	}
+	dentry->d_fsdata = oe;
+	if (upper_alias)
+		ovl_dentry_set_upper_alias(dentry);
+
+	ovl_dentry_update_reval(dentry, upper,
+			DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
 
 	return d_instantiate_anon(dentry, inode);
 
 nomem:
-	iput(inode);
 	dput(dentry);
-	return ERR_PTR(-ENOMEM);
+	dentry = ERR_PTR(-ENOMEM);
+out_iput:
+	iput(inode);
+	return dentry;
 }
 
 /* Get the upper or lower dentry in stach whose on layer @idx */
@@ -367,7 +364,7 @@ static struct dentry *ovl_dentry_real_at(struct dentry *dentry, int idx)
  */
 static struct dentry *ovl_lookup_real_one(struct dentry *connected,
 					  struct dentry *real,
-					  struct ovl_layer *layer)
+					  const struct ovl_layer *layer)
 {
 	struct inode *dir = d_inode(connected);
 	struct dentry *this, *parent = NULL;
@@ -415,7 +412,7 @@ out:
 	return this;
 
 fail:
-	pr_warn_ratelimited("overlayfs: failed to lookup one by real (%pd2, layer=%d, connected=%pd2, err=%i)\n",
+	pr_warn_ratelimited("failed to lookup one by real (%pd2, layer=%d, connected=%pd2, err=%i)\n",
 			    real, layer->idx, connected, err);
 	this = ERR_PTR(err);
 	goto out;
@@ -423,17 +420,16 @@ fail:
 
 static struct dentry *ovl_lookup_real(struct super_block *sb,
 				      struct dentry *real,
-				      struct ovl_layer *layer);
+				      const struct ovl_layer *layer);
 
 /*
  * Lookup an indexed or hashed overlay dentry by real inode.
  */
 static struct dentry *ovl_lookup_real_inode(struct super_block *sb,
 					    struct dentry *real,
-					    struct ovl_layer *layer)
+					    const struct ovl_layer *layer)
 {
 	struct ovl_fs *ofs = sb->s_fs_info;
-	struct ovl_layer upper_layer = { .mnt = ofs->upper_mnt };
 	struct dentry *index = NULL;
 	struct dentry *this = NULL;
 	struct inode *inode;
@@ -475,7 +471,7 @@ static struct dentry *ovl_lookup_real_inode(struct super_block *sb,
 		 * recursive call walks back from indexed upper to the topmost
 		 * connected/hashed upper parent (or up to root).
 		 */
-		this = ovl_lookup_real(sb, upper, &upper_layer);
+		this = ovl_lookup_real(sb, upper, &ofs->layers[0]);
 		dput(upper);
 	}
 
@@ -496,7 +492,7 @@ static struct dentry *ovl_lookup_real_inode(struct super_block *sb,
  */
 static struct dentry *ovl_lookup_real_ancestor(struct super_block *sb,
 					       struct dentry *real,
-					       struct ovl_layer *layer)
+					       const struct ovl_layer *layer)
 {
 	struct dentry *next, *parent = NULL;
 	struct dentry *ancestor = ERR_PTR(-EIO);
@@ -549,7 +545,7 @@ static struct dentry *ovl_lookup_real_ancestor(struct super_block *sb,
  */
 static struct dentry *ovl_lookup_real(struct super_block *sb,
 				      struct dentry *real,
-				      struct ovl_layer *layer)
+				      const struct ovl_layer *layer)
 {
 	struct dentry *connected;
 	int err = 0;
@@ -640,7 +636,7 @@ static struct dentry *ovl_lookup_real(struct super_block *sb,
 	return connected;
 
 fail:
-	pr_warn_ratelimited("overlayfs: failed to lookup by real (%pd2, layer=%d, connected=%pd2, err=%i)\n",
+	pr_warn_ratelimited("failed to lookup by real (%pd2, layer=%d, connected=%pd2, err=%i)\n",
 			    real, layer->idx, connected, err);
 	dput(connected);
 	return ERR_PTR(err);
@@ -655,8 +651,7 @@ static struct dentry *ovl_get_dentry(struct super_block *sb,
 				     struct dentry *index)
 {
 	struct ovl_fs *ofs = sb->s_fs_info;
-	struct ovl_layer upper_layer = { .mnt = ofs->upper_mnt };
-	struct ovl_layer *layer = upper ? &upper_layer : lowerpath->layer;
+	const struct ovl_layer *layer = upper ? &ofs->layers[0] : lowerpath->layer;
 	struct dentry *real = upper ?: (index ?: lowerpath->dentry);
 
 	/*
@@ -777,24 +772,48 @@ out_err:
 	goto out;
 }
 
+static struct ovl_fh *ovl_fid_to_fh(struct fid *fid, int buflen, int fh_type)
+{
+	struct ovl_fh *fh;
+
+	/* If on-wire inner fid is aligned - nothing to do */
+	if (fh_type == OVL_FILEID_V1)
+		return (struct ovl_fh *)fid;
+
+	if (fh_type != OVL_FILEID_V0)
+		return ERR_PTR(-EINVAL);
+
+	if (buflen <= OVL_FH_WIRE_OFFSET)
+		return ERR_PTR(-EINVAL);
+
+	fh = kzalloc(buflen, GFP_KERNEL);
+	if (!fh)
+		return ERR_PTR(-ENOMEM);
+
+	/* Copy unaligned inner fh into aligned buffer */
+	memcpy(&fh->fb, fid, buflen - OVL_FH_WIRE_OFFSET);
+	return fh;
+}
+
 static struct dentry *ovl_fh_to_dentry(struct super_block *sb, struct fid *fid,
 				       int fh_len, int fh_type)
 {
 	struct dentry *dentry = NULL;
-	struct ovl_fh *fh = (struct ovl_fh *) fid;
+	struct ovl_fh *fh = NULL;
 	int len = fh_len << 2;
 	unsigned int flags = 0;
 	int err;
 
-	err = -EINVAL;
-	if (fh_type != OVL_FILEID)
+	fh = ovl_fid_to_fh(fid, len, fh_type);
+	err = PTR_ERR(fh);
+	if (IS_ERR(fh))
 		goto out_err;
 
 	err = ovl_check_fh_len(fh, len);
 	if (err)
 		goto out_err;
 
-	flags = fh->flags;
+	flags = fh->fb.flags;
 	dentry = (flags & OVL_FH_FLAG_PATH_UPPER) ?
 		 ovl_upper_fh_to_d(sb, fh) :
 		 ovl_lower_fh_to_d(sb, fh);
@@ -802,18 +821,24 @@ static struct dentry *ovl_fh_to_dentry(struct super_block *sb, struct fid *fid,
 	if (IS_ERR(dentry) && err != -ESTALE)
 		goto out_err;
 
+out:
+	/* We may have needed to re-align OVL_FILEID_V0 */
+	if (!IS_ERR_OR_NULL(fh) && fh != (void *)fid)
+		kfree(fh);
+
 	return dentry;
 
 out_err:
-	pr_warn_ratelimited("overlayfs: failed to decode file handle (len=%d, type=%d, flags=%x, err=%i)\n",
-			    len, fh_type, flags, err);
-	return ERR_PTR(err);
+	pr_warn_ratelimited("failed to decode file handle (len=%d, type=%d, flags=%x, err=%i)\n",
+			    fh_len, fh_type, flags, err);
+	dentry = ERR_PTR(err);
+	goto out;
 }
 
 static struct dentry *ovl_fh_to_parent(struct super_block *sb, struct fid *fid,
 				       int fh_len, int fh_type)
 {
-	pr_warn_ratelimited("overlayfs: connectable file handles not supported; use 'no_subtree_check' exportfs option.\n");
+	pr_warn_ratelimited("connectable file handles not supported; use 'no_subtree_check' exportfs option.\n");
 	return ERR_PTR(-EACCES);
 }
 
diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c
index e235a635d9ec..87c362f65448 100644
--- a/fs/overlayfs/file.c
+++ b/fs/overlayfs/file.c
@@ -9,8 +9,19 @@
 #include <linux/xattr.h>
 #include <linux/uio.h>
 #include <linux/uaccess.h>
+#include <linux/splice.h>
+#include <linux/mm.h>
+#include <linux/fs.h>
 #include "overlayfs.h"
 
+struct ovl_aio_req {
+	struct kiocb iocb;
+	struct kiocb *orig_iocb;
+	struct fd fd;
+};
+
+static struct kmem_cache *ovl_aio_request_cachep;
+
 static char ovl_whatisit(struct inode *inode, struct inode *realinode)
 {
 	if (realinode != ovl_inode_upper(inode))
@@ -146,7 +157,7 @@ static loff_t ovl_llseek(struct file *file, loff_t offset, int whence)
 	struct inode *inode = file_inode(file);
 	struct fd real;
 	const struct cred *old_cred;
-	ssize_t ret;
+	loff_t ret;
 
 	/*
 	 * The two special cases below do not need to involve real fs,
@@ -171,7 +182,7 @@ static loff_t ovl_llseek(struct file *file, loff_t offset, int whence)
 	 * limitations that are more strict than ->s_maxbytes for specific
 	 * files, so we use the real file to perform seeks.
 	 */
-	inode_lock(inode);
+	ovl_inode_lock(inode);
 	real.file->f_pos = file->f_pos;
 
 	old_cred = ovl_override_creds(inode->i_sb);
@@ -179,7 +190,7 @@ static loff_t ovl_llseek(struct file *file, loff_t offset, int whence)
 	revert_creds(old_cred);
 
 	file->f_pos = real.file->f_pos;
-	inode_unlock(inode);
+	ovl_inode_unlock(inode);
 
 	fdput(real);
 
@@ -225,6 +236,36 @@ static rwf_t ovl_iocb_to_rwf(struct kiocb *iocb)
 	return flags;
 }
 
+static void ovl_aio_cleanup_handler(struct ovl_aio_req *aio_req)
+{
+	struct kiocb *iocb = &aio_req->iocb;
+	struct kiocb *orig_iocb = aio_req->orig_iocb;
+
+	if (iocb->ki_flags & IOCB_WRITE) {
+		struct inode *inode = file_inode(orig_iocb->ki_filp);
+
+		/* Actually acquired in ovl_write_iter() */
+		__sb_writers_acquired(file_inode(iocb->ki_filp)->i_sb,
+				      SB_FREEZE_WRITE);
+		file_end_write(iocb->ki_filp);
+		ovl_copyattr(ovl_inode_real(inode), inode);
+	}
+
+	orig_iocb->ki_pos = iocb->ki_pos;
+	fdput(aio_req->fd);
+	kmem_cache_free(ovl_aio_request_cachep, aio_req);
+}
+
+static void ovl_aio_rw_complete(struct kiocb *iocb, long res, long res2)
+{
+	struct ovl_aio_req *aio_req = container_of(iocb,
+						   struct ovl_aio_req, iocb);
+	struct kiocb *orig_iocb = aio_req->orig_iocb;
+
+	ovl_aio_cleanup_handler(aio_req);
+	orig_iocb->ki_complete(orig_iocb, res, res2);
+}
+
 static ssize_t ovl_read_iter(struct kiocb *iocb, struct iov_iter *iter)
 {
 	struct file *file = iocb->ki_filp;
@@ -240,10 +281,28 @@ static ssize_t ovl_read_iter(struct kiocb *iocb, struct iov_iter *iter)
 		return ret;
 
 	old_cred = ovl_override_creds(file_inode(file)->i_sb);
-	ret = vfs_iter_read(real.file, iter, &iocb->ki_pos,
-			    ovl_iocb_to_rwf(iocb));
+	if (is_sync_kiocb(iocb)) {
+		ret = vfs_iter_read(real.file, iter, &iocb->ki_pos,
+				    ovl_iocb_to_rwf(iocb));
+	} else {
+		struct ovl_aio_req *aio_req;
+
+		ret = -ENOMEM;
+		aio_req = kmem_cache_zalloc(ovl_aio_request_cachep, GFP_KERNEL);
+		if (!aio_req)
+			goto out;
+
+		aio_req->fd = real;
+		real.flags = 0;
+		aio_req->orig_iocb = iocb;
+		kiocb_clone(&aio_req->iocb, iocb, real.file);
+		aio_req->iocb.ki_complete = ovl_aio_rw_complete;
+		ret = vfs_iocb_iter_read(real.file, &aio_req->iocb, iter);
+		if (ret != -EIOCBQUEUED)
+			ovl_aio_cleanup_handler(aio_req);
+	}
+out:
 	revert_creds(old_cred);
-
 	ovl_file_accessed(file);
 
 	fdput(real);
@@ -274,15 +333,36 @@ static ssize_t ovl_write_iter(struct kiocb *iocb, struct iov_iter *iter)
 		goto out_unlock;
 
 	old_cred = ovl_override_creds(file_inode(file)->i_sb);
-	file_start_write(real.file);
-	ret = vfs_iter_write(real.file, iter, &iocb->ki_pos,
-			     ovl_iocb_to_rwf(iocb));
-	file_end_write(real.file);
+	if (is_sync_kiocb(iocb)) {
+		file_start_write(real.file);
+		ret = vfs_iter_write(real.file, iter, &iocb->ki_pos,
+				     ovl_iocb_to_rwf(iocb));
+		file_end_write(real.file);
+		/* Update size */
+		ovl_copyattr(ovl_inode_real(inode), inode);
+	} else {
+		struct ovl_aio_req *aio_req;
+
+		ret = -ENOMEM;
+		aio_req = kmem_cache_zalloc(ovl_aio_request_cachep, GFP_KERNEL);
+		if (!aio_req)
+			goto out;
+
+		file_start_write(real.file);
+		/* Pacify lockdep, same trick as done in aio_write() */
+		__sb_writers_release(file_inode(real.file)->i_sb,
+				     SB_FREEZE_WRITE);
+		aio_req->fd = real;
+		real.flags = 0;
+		aio_req->orig_iocb = iocb;
+		kiocb_clone(&aio_req->iocb, iocb, real.file);
+		aio_req->iocb.ki_complete = ovl_aio_rw_complete;
+		ret = vfs_iocb_iter_write(real.file, &aio_req->iocb, iter);
+		if (ret != -EIOCBQUEUED)
+			ovl_aio_cleanup_handler(aio_req);
+	}
+out:
 	revert_creds(old_cred);
-
-	/* Update size */
-	ovl_copyattr(ovl_inode_real(inode), inode);
-
 	fdput(real);
 
 out_unlock:
@@ -291,6 +371,48 @@ out_unlock:
 	return ret;
 }
 
+static ssize_t ovl_splice_read(struct file *in, loff_t *ppos,
+			 struct pipe_inode_info *pipe, size_t len,
+			 unsigned int flags)
+{
+	ssize_t ret;
+	struct fd real;
+	const struct cred *old_cred;
+
+	ret = ovl_real_fdget(in, &real);
+	if (ret)
+		return ret;
+
+	old_cred = ovl_override_creds(file_inode(in)->i_sb);
+	ret = generic_file_splice_read(real.file, ppos, pipe, len, flags);
+	revert_creds(old_cred);
+
+	ovl_file_accessed(in);
+	fdput(real);
+	return ret;
+}
+
+static ssize_t
+ovl_splice_write(struct pipe_inode_info *pipe, struct file *out,
+			  loff_t *ppos, size_t len, unsigned int flags)
+{
+	struct fd real;
+	const struct cred *old_cred;
+	ssize_t ret;
+
+	ret = ovl_real_fdget(out, &real);
+	if (ret)
+		return ret;
+
+	old_cred = ovl_override_creds(file_inode(out)->i_sb);
+	ret = iter_file_splice_write(pipe, real.file, ppos, len, flags);
+	revert_creds(old_cred);
+
+	ovl_file_accessed(out);
+	fdput(real);
+	return ret;
+}
+
 static int ovl_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct fd real;
@@ -647,7 +769,25 @@ const struct file_operations ovl_file_operations = {
 	.fadvise	= ovl_fadvise,
 	.unlocked_ioctl	= ovl_ioctl,
 	.compat_ioctl	= ovl_compat_ioctl,
+	.splice_read    = ovl_splice_read,
+	.splice_write   = ovl_splice_write,
 
 	.copy_file_range	= ovl_copy_file_range,
 	.remap_file_range	= ovl_remap_file_range,
 };
+
+int __init ovl_aio_request_cache_init(void)
+{
+	ovl_aio_request_cachep = kmem_cache_create("ovl_aio_req",
+						   sizeof(struct ovl_aio_req),
+						   0, SLAB_HWCACHE_ALIGN, NULL);
+	if (!ovl_aio_request_cachep)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void ovl_aio_request_cache_destroy(void)
+{
+	kmem_cache_destroy(ovl_aio_request_cachep);
+}
diff --git a/fs/overlayfs/inode.c b/fs/overlayfs/inode.c
index bc14781886bf..981f11ec51bc 100644
--- a/fs/overlayfs/inode.c
+++ b/fs/overlayfs/inode.c
@@ -58,6 +58,24 @@ int ovl_setattr(struct dentry *dentry, struct iattr *attr)
 		if (attr->ia_valid & (ATTR_KILL_SUID|ATTR_KILL_SGID))
 			attr->ia_valid &= ~ATTR_MODE;
 
+		/*
+		 * We might have to translate ovl file into real file object
+		 * once use cases emerge.  For now, simply don't let underlying
+		 * filesystem rely on attr->ia_file
+		 */
+		attr->ia_valid &= ~ATTR_FILE;
+
+		/*
+		 * If open(O_TRUNC) is done, VFS calls ->setattr with ATTR_OPEN
+		 * set.  Overlayfs does not pass O_TRUNC flag to underlying
+		 * filesystem during open -> do not pass ATTR_OPEN.  This
+		 * disables optimization in fuse which assumes open(O_TRUNC)
+		 * already set file size to 0.  But we never passed O_TRUNC to
+		 * fuse.  So by clearing ATTR_OPEN, fuse will be forced to send
+		 * setattr request to server.
+		 */
+		attr->ia_valid &= ~ATTR_OPEN;
+
 		inode_lock(upperdentry->d_inode);
 		old_cred = ovl_override_creds(dentry->d_sb);
 		err = notify_change(upperdentry, attr, NULL);
@@ -75,11 +93,11 @@ out:
 	return err;
 }
 
-static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat,
-			   struct ovl_layer *lower_layer)
+static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, int fsid)
 {
-	bool samefs = ovl_same_sb(dentry->d_sb);
+	bool samefs = ovl_same_fs(dentry->d_sb);
 	unsigned int xinobits = ovl_xino_bits(dentry->d_sb);
+	unsigned int xinoshift = 64 - xinobits;
 
 	if (samefs) {
 		/*
@@ -90,24 +108,22 @@ static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat,
 		stat->dev = dentry->d_sb->s_dev;
 		return 0;
 	} else if (xinobits) {
-		unsigned int shift = 64 - xinobits;
 		/*
 		 * All inode numbers of underlying fs should not be using the
 		 * high xinobits, so we use high xinobits to partition the
 		 * overlay st_ino address space. The high bits holds the fsid
-		 * (upper fsid is 0). This way overlay inode numbers are unique
-		 * and all inodes use overlay st_dev. Inode numbers are also
-		 * persistent for a given layer configuration.
+		 * (upper fsid is 0). The lowest xinobit is reserved for mapping
+		 * the non-peresistent inode numbers range in case of overflow.
+		 * This way all overlay inode numbers are unique and use the
+		 * overlay st_dev.
 		 */
-		if (stat->ino >> shift) {
-			pr_warn_ratelimited("overlayfs: inode number too big (%pd2, ino=%llu, xinobits=%d)\n",
-					    dentry, stat->ino, xinobits);
-		} else {
-			if (lower_layer)
-				stat->ino |= ((u64)lower_layer->fsid) << shift;
-
+		if (likely(!(stat->ino >> xinoshift))) {
+			stat->ino |= ((u64)fsid) << (xinoshift + 1);
 			stat->dev = dentry->d_sb->s_dev;
 			return 0;
+		} else if (ovl_xino_warn(dentry->d_sb)) {
+			pr_warn_ratelimited("inode number too big (%pd2, ino=%llu, xinobits=%d)\n",
+					    dentry, stat->ino, xinobits);
 		}
 	}
 
@@ -124,15 +140,14 @@ static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat,
 		 */
 		stat->dev = dentry->d_sb->s_dev;
 		stat->ino = dentry->d_inode->i_ino;
-	} else if (lower_layer && lower_layer->fsid) {
+	} else {
 		/*
 		 * For non-samefs setup, if we cannot map all layers st_ino
 		 * to a unified address space, we need to make sure that st_dev
-		 * is unique per lower fs. Upper layer uses real st_dev and
-		 * lower layers use the unique anonymous bdev assigned to the
-		 * lower fs.
+		 * is unique per underlying fs, so we use the unique anonymous
+		 * bdev assigned to the underlying fs.
 		 */
-		stat->dev = lower_layer->fs->pseudo_dev;
+		stat->dev = OVL_FS(dentry->d_sb)->fs[fsid].pseudo_dev;
 	}
 
 	return 0;
@@ -146,8 +161,7 @@ int ovl_getattr(const struct path *path, struct kstat *stat,
 	struct path realpath;
 	const struct cred *old_cred;
 	bool is_dir = S_ISDIR(dentry->d_inode->i_mode);
-	bool samefs = ovl_same_sb(dentry->d_sb);
-	struct ovl_layer *lower_layer = NULL;
+	int fsid = 0;
 	int err;
 	bool metacopy_blocks = false;
 
@@ -168,9 +182,9 @@ int ovl_getattr(const struct path *path, struct kstat *stat,
 	 * If lower filesystem supports NFS file handles, this also guaranties
 	 * persistent st_ino across mount cycle.
 	 */
-	if (!is_dir || samefs || ovl_xino_bits(dentry->d_sb)) {
+	if (!is_dir || ovl_same_dev(dentry->d_sb)) {
 		if (!OVL_TYPE_UPPER(type)) {
-			lower_layer = ovl_layer_lower(dentry);
+			fsid = ovl_layer_lower(dentry)->fsid;
 		} else if (OVL_TYPE_ORIGIN(type)) {
 			struct kstat lowerstat;
 			u32 lowermask = STATX_INO | STATX_BLOCKS |
@@ -200,8 +214,8 @@ int ovl_getattr(const struct path *path, struct kstat *stat,
 			if (ovl_test_flag(OVL_INDEX, d_inode(dentry)) ||
 			    (!ovl_verify_lower(dentry->d_sb) &&
 			     (is_dir || lowerstat.nlink == 1))) {
+				fsid = ovl_layer_lower(dentry)->fsid;
 				stat->ino = lowerstat.ino;
-				lower_layer = ovl_layer_lower(dentry);
 			}
 
 			/*
@@ -235,7 +249,7 @@ int ovl_getattr(const struct path *path, struct kstat *stat,
 		}
 	}
 
-	err = ovl_map_dev_ino(dentry, stat, lower_layer);
+	err = ovl_map_dev_ino(dentry, stat, fsid);
 	if (err)
 		goto out;
 
@@ -509,7 +523,7 @@ static const struct address_space_operations ovl_aops = {
 
 /*
  * It is possible to stack overlayfs instance on top of another
- * overlayfs instance as lower layer. We need to annonate the
+ * overlayfs instance as lower layer. We need to annotate the
  * stackable i_mutex locks according to stack level of the super
  * block instance. An overlayfs instance can never be in stack
  * depth 0 (there is always a real fs below it).  An overlayfs
@@ -521,6 +535,27 @@ static const struct address_space_operations ovl_aops = {
  * [...] &ovl_i_mutex_dir_key[depth]   (stack_depth=2)
  * [...] &ovl_i_mutex_dir_key[depth]#2 (stack_depth=1)
  * [...] &type->i_mutex_dir_key        (stack_depth=0)
+ *
+ * Locking order w.r.t ovl_want_write() is important for nested overlayfs.
+ *
+ * This chain is valid:
+ * - inode->i_rwsem			(inode_lock[2])
+ * - upper_mnt->mnt_sb->s_writers	(ovl_want_write[0])
+ * - OVL_I(inode)->lock			(ovl_inode_lock[2])
+ * - OVL_I(lowerinode)->lock		(ovl_inode_lock[1])
+ *
+ * And this chain is valid:
+ * - inode->i_rwsem			(inode_lock[2])
+ * - OVL_I(inode)->lock			(ovl_inode_lock[2])
+ * - lowerinode->i_rwsem		(inode_lock[1])
+ * - OVL_I(lowerinode)->lock		(ovl_inode_lock[1])
+ *
+ * But lowerinode->i_rwsem SHOULD NOT be acquired while ovl_want_write() is
+ * held, because it is in reverse order of the non-nested case using the same
+ * upper fs:
+ * - inode->i_rwsem			(inode_lock[1])
+ * - upper_mnt->mnt_sb->s_writers	(ovl_want_write[0])
+ * - OVL_I(inode)->lock			(ovl_inode_lock[1])
  */
 #define OVL_MAX_NESTING FILESYSTEM_MAX_STACK_DEPTH
 
@@ -545,27 +580,73 @@ static inline void ovl_lockdep_annotate_inode_mutex_key(struct inode *inode)
 #endif
 }
 
-static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev,
-			   unsigned long ino, int fsid)
+static void ovl_next_ino(struct inode *inode)
+{
+	struct ovl_fs *ofs = inode->i_sb->s_fs_info;
+
+	inode->i_ino = atomic_long_inc_return(&ofs->last_ino);
+	if (unlikely(!inode->i_ino))
+		inode->i_ino = atomic_long_inc_return(&ofs->last_ino);
+}
+
+static void ovl_map_ino(struct inode *inode, unsigned long ino, int fsid)
 {
 	int xinobits = ovl_xino_bits(inode->i_sb);
+	unsigned int xinoshift = 64 - xinobits;
 
 	/*
 	 * When d_ino is consistent with st_ino (samefs or i_ino has enough
 	 * bits to encode layer), set the same value used for st_ino to i_ino,
 	 * so inode number exposed via /proc/locks and a like will be
 	 * consistent with d_ino and st_ino values. An i_ino value inconsistent
-	 * with d_ino also causes nfsd readdirplus to fail.  When called from
-	 * ovl_new_inode(), ino arg is 0, so i_ino will be updated to real
-	 * upper inode i_ino on ovl_inode_init() or ovl_inode_update().
+	 * with d_ino also causes nfsd readdirplus to fail.
 	 */
-	if (ovl_same_sb(inode->i_sb) || xinobits) {
-		inode->i_ino = ino;
-		if (xinobits && fsid && !(ino >> (64 - xinobits)))
-			inode->i_ino |= (unsigned long)fsid << (64 - xinobits);
-	} else {
-		inode->i_ino = get_next_ino();
+	inode->i_ino = ino;
+	if (ovl_same_fs(inode->i_sb)) {
+		return;
+	} else if (xinobits && likely(!(ino >> xinoshift))) {
+		inode->i_ino |= (unsigned long)fsid << (xinoshift + 1);
+		return;
+	}
+
+	/*
+	 * For directory inodes on non-samefs with xino disabled or xino
+	 * overflow, we allocate a non-persistent inode number, to be used for
+	 * resolving st_ino collisions in ovl_map_dev_ino().
+	 *
+	 * To avoid ino collision with legitimate xino values from upper
+	 * layer (fsid 0), use the lowest xinobit to map the non
+	 * persistent inode numbers to the unified st_ino address space.
+	 */
+	if (S_ISDIR(inode->i_mode)) {
+		ovl_next_ino(inode);
+		if (xinobits) {
+			inode->i_ino &= ~0UL >> xinobits;
+			inode->i_ino |= 1UL << xinoshift;
+		}
 	}
+}
+
+void ovl_inode_init(struct inode *inode, struct ovl_inode_params *oip,
+		    unsigned long ino, int fsid)
+{
+	struct inode *realinode;
+
+	if (oip->upperdentry)
+		OVL_I(inode)->__upperdentry = oip->upperdentry;
+	if (oip->lowerpath && oip->lowerpath->dentry)
+		OVL_I(inode)->lower = igrab(d_inode(oip->lowerpath->dentry));
+	if (oip->lowerdata)
+		OVL_I(inode)->lowerdata = igrab(d_inode(oip->lowerdata));
+
+	realinode = ovl_inode_real(inode);
+	ovl_copyattr(realinode, inode);
+	ovl_copyflags(realinode, inode);
+	ovl_map_ino(inode, ino, fsid);
+}
+
+static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev)
+{
 	inode->i_mode = mode;
 	inode->i_flags |= S_NOCMTIME;
 #ifdef CONFIG_FS_POSIX_ACL
@@ -692,7 +773,7 @@ unsigned int ovl_get_nlink(struct dentry *lowerdentry,
 	return nlink;
 
 fail:
-	pr_warn_ratelimited("overlayfs: failed to get index nlink (%pd2, err=%i)\n",
+	pr_warn_ratelimited("failed to get index nlink (%pd2, err=%i)\n",
 			    upperdentry, err);
 	return fallback;
 }
@@ -703,7 +784,7 @@ struct inode *ovl_new_inode(struct super_block *sb, umode_t mode, dev_t rdev)
 
 	inode = new_inode(sb);
 	if (inode)
-		ovl_fill_inode(inode, mode, rdev, 0, 0);
+		ovl_fill_inode(inode, mode, rdev);
 
 	return inode;
 }
@@ -875,7 +956,7 @@ struct inode *ovl_get_inode(struct super_block *sb,
 	struct dentry *lowerdentry = lowerpath ? lowerpath->dentry : NULL;
 	bool bylower = ovl_hash_bylower(sb, upperdentry, lowerdentry,
 					oip->index);
-	int fsid = bylower ? oip->lowerpath->layer->fsid : 0;
+	int fsid = bylower ? lowerpath->layer->fsid : 0;
 	bool is_dir, metacopy = false;
 	unsigned long ino = 0;
 	int err = oip->newinode ? -EEXIST : -ENOMEM;
@@ -925,9 +1006,11 @@ struct inode *ovl_get_inode(struct super_block *sb,
 			err = -ENOMEM;
 			goto out_err;
 		}
+		ino = realinode->i_ino;
+		fsid = lowerpath->layer->fsid;
 	}
-	ovl_fill_inode(inode, realinode->i_mode, realinode->i_rdev, ino, fsid);
-	ovl_inode_init(inode, upperdentry, lowerdentry, oip->lowerdata);
+	ovl_fill_inode(inode, realinode->i_mode, realinode->i_rdev);
+	ovl_inode_init(inode, oip, ino, fsid);
 
 	if (upperdentry && ovl_is_impuredir(upperdentry))
 		ovl_set_flag(OVL_IMPURE, inode);
@@ -963,7 +1046,7 @@ out:
 	return inode;
 
 out_err:
-	pr_warn_ratelimited("overlayfs: failed to get inode (%i)\n", err);
+	pr_warn_ratelimited("failed to get inode (%i)\n", err);
 	inode = ERR_PTR(err);
 	goto out;
 }
diff --git a/fs/overlayfs/namei.c b/fs/overlayfs/namei.c
index e9717c2f7d45..0db23baf98e7 100644
--- a/fs/overlayfs/namei.c
+++ b/fs/overlayfs/namei.c
@@ -84,21 +84,21 @@ static int ovl_acceptable(void *ctx, struct dentry *dentry)
  * Return -ENODATA for "origin unknown".
  * Return <0 for an invalid file handle.
  */
-int ovl_check_fh_len(struct ovl_fh *fh, int fh_len)
+int ovl_check_fb_len(struct ovl_fb *fb, int fb_len)
 {
-	if (fh_len < sizeof(struct ovl_fh) || fh_len < fh->len)
+	if (fb_len < sizeof(struct ovl_fb) || fb_len < fb->len)
 		return -EINVAL;
 
-	if (fh->magic != OVL_FH_MAGIC)
+	if (fb->magic != OVL_FH_MAGIC)
 		return -EINVAL;
 
 	/* Treat larger version and unknown flags as "origin unknown" */
-	if (fh->version > OVL_FH_VERSION || fh->flags & ~OVL_FH_FLAG_ALL)
+	if (fb->version > OVL_FH_VERSION || fb->flags & ~OVL_FH_FLAG_ALL)
 		return -ENODATA;
 
 	/* Treat endianness mismatch as "origin unknown" */
-	if (!(fh->flags & OVL_FH_FLAG_ANY_ENDIAN) &&
-	    (fh->flags & OVL_FH_FLAG_BIG_ENDIAN) != OVL_FH_FLAG_CPU_ENDIAN)
+	if (!(fb->flags & OVL_FH_FLAG_ANY_ENDIAN) &&
+	    (fb->flags & OVL_FH_FLAG_BIG_ENDIAN) != OVL_FH_FLAG_CPU_ENDIAN)
 		return -ENODATA;
 
 	return 0;
@@ -119,15 +119,15 @@ static struct ovl_fh *ovl_get_fh(struct dentry *dentry, const char *name)
 	if (res == 0)
 		return NULL;
 
-	fh = kzalloc(res, GFP_KERNEL);
+	fh = kzalloc(res + OVL_FH_WIRE_OFFSET, GFP_KERNEL);
 	if (!fh)
 		return ERR_PTR(-ENOMEM);
 
-	res = vfs_getxattr(dentry, name, fh, res);
+	res = vfs_getxattr(dentry, name, fh->buf, res);
 	if (res < 0)
 		goto fail;
 
-	err = ovl_check_fh_len(fh, res);
+	err = ovl_check_fb_len(&fh->fb, res);
 	if (err < 0) {
 		if (err == -ENODATA)
 			goto out;
@@ -141,10 +141,10 @@ out:
 	return NULL;
 
 fail:
-	pr_warn_ratelimited("overlayfs: failed to get origin (%i)\n", res);
+	pr_warn_ratelimited("failed to get origin (%i)\n", res);
 	goto out;
 invalid:
-	pr_warn_ratelimited("overlayfs: invalid origin (%*phN)\n", res, fh);
+	pr_warn_ratelimited("invalid origin (%*phN)\n", res, fh);
 	goto out;
 }
 
@@ -158,12 +158,12 @@ struct dentry *ovl_decode_real_fh(struct ovl_fh *fh, struct vfsmount *mnt,
 	 * Make sure that the stored uuid matches the uuid of the lower
 	 * layer where file handle will be decoded.
 	 */
-	if (!uuid_equal(&fh->uuid, &mnt->mnt_sb->s_uuid))
+	if (!uuid_equal(&fh->fb.uuid, &mnt->mnt_sb->s_uuid))
 		return NULL;
 
-	bytes = (fh->len - offsetof(struct ovl_fh, fid));
-	real = exportfs_decode_fh(mnt, (struct fid *)fh->fid,
-				  bytes >> 2, (int)fh->type,
+	bytes = (fh->fb.len - offsetof(struct ovl_fb, fid));
+	real = exportfs_decode_fh(mnt, (struct fid *)fh->fb.fid,
+				  bytes >> 2, (int)fh->fb.type,
 				  connected ? ovl_acceptable : NULL, mnt);
 	if (IS_ERR(real)) {
 		/*
@@ -173,7 +173,7 @@ struct dentry *ovl_decode_real_fh(struct ovl_fh *fh, struct vfsmount *mnt,
 		 * index entries correctly.
 		 */
 		if (real == ERR_PTR(-ESTALE) &&
-		    !(fh->flags & OVL_FH_FLAG_PATH_UPPER))
+		    !(fh->fb.flags & OVL_FH_FLAG_PATH_UPPER))
 			real = NULL;
 		return real;
 	}
@@ -200,7 +200,7 @@ static int ovl_lookup_single(struct dentry *base, struct ovl_lookup_data *d,
 	int err;
 	bool last_element = !post[0];
 
-	this = lookup_one_len_unlocked(name, base, namelen);
+	this = lookup_positive_unlocked(name, base, namelen);
 	if (IS_ERR(this)) {
 		err = PTR_ERR(this);
 		this = NULL;
@@ -208,8 +208,6 @@ static int ovl_lookup_single(struct dentry *base, struct ovl_lookup_data *d,
 			goto out;
 		goto out_err;
 	}
-	if (!this->d_inode)
-		goto put_and_out;
 
 	if (ovl_dentry_weird(this)) {
 		/* Don't support traversing automounts and other weirdness */
@@ -324,8 +322,16 @@ int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, bool connected,
 	struct dentry *origin = NULL;
 	int i;
 
-	for (i = 0; i < ofs->numlower; i++) {
-		origin = ovl_decode_real_fh(fh, ofs->lower_layers[i].mnt,
+	for (i = 1; i < ofs->numlayer; i++) {
+		/*
+		 * If lower fs uuid is not unique among lower fs we cannot match
+		 * fh->uuid to layer.
+		 */
+		if (ofs->layers[i].fsid &&
+		    ofs->layers[i].fs->bad_uuid)
+			continue;
+
+		origin = ovl_decode_real_fh(fh, ofs->layers[i].mnt,
 					    connected);
 		if (origin)
 			break;
@@ -348,13 +354,13 @@ int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, bool connected,
 	}
 	**stackp = (struct ovl_path){
 		.dentry = origin,
-		.layer = &ofs->lower_layers[i]
+		.layer = &ofs->layers[i]
 	};
 
 	return 0;
 
 invalid:
-	pr_warn_ratelimited("overlayfs: invalid origin (%pd2, ftype=%x, origin ftype=%x).\n",
+	pr_warn_ratelimited("invalid origin (%pd2, ftype=%x, origin ftype=%x).\n",
 			    upperdentry, d_inode(upperdentry)->i_mode & S_IFMT,
 			    d_inode(origin)->i_mode & S_IFMT);
 	dput(origin);
@@ -402,7 +408,7 @@ static int ovl_verify_fh(struct dentry *dentry, const char *name,
 	if (IS_ERR(ofh))
 		return PTR_ERR(ofh);
 
-	if (fh->len != ofh->len || memcmp(fh, ofh, fh->len))
+	if (fh->fb.len != ofh->fb.len || memcmp(&fh->fb, &ofh->fb, fh->fb.len))
 		err = -ESTALE;
 
 	kfree(ofh);
@@ -433,7 +439,7 @@ int ovl_verify_set_fh(struct dentry *dentry, const char *name,
 
 	err = ovl_verify_fh(dentry, name, fh);
 	if (set && err == -ENODATA)
-		err = ovl_do_setxattr(dentry, name, fh, fh->len, 0);
+		err = ovl_do_setxattr(dentry, name, fh->buf, fh->fb.len, 0);
 	if (err)
 		goto fail;
 
@@ -443,7 +449,7 @@ out:
 
 fail:
 	inode = d_inode(real);
-	pr_warn_ratelimited("overlayfs: failed to verify %s (%pd2, ino=%lu, err=%i)\n",
+	pr_warn_ratelimited("failed to verify %s (%pd2, ino=%lu, err=%i)\n",
 			    is_upper ? "upper" : "origin", real,
 			    inode ? inode->i_ino : 0, err);
 	goto out;
@@ -469,7 +475,7 @@ struct dentry *ovl_index_upper(struct ovl_fs *ofs, struct dentry *index)
 		return upper ?: ERR_PTR(-ESTALE);
 
 	if (!d_is_dir(upper)) {
-		pr_warn_ratelimited("overlayfs: invalid index upper (%pd2, upper=%pd2).\n",
+		pr_warn_ratelimited("invalid index upper (%pd2, upper=%pd2).\n",
 				    index, upper);
 		dput(upper);
 		return ERR_PTR(-EIO);
@@ -507,20 +513,20 @@ int ovl_verify_index(struct ovl_fs *ofs, struct dentry *index)
 		goto fail;
 
 	err = -EINVAL;
-	if (index->d_name.len < sizeof(struct ovl_fh)*2)
+	if (index->d_name.len < sizeof(struct ovl_fb)*2)
 		goto fail;
 
 	err = -ENOMEM;
 	len = index->d_name.len / 2;
-	fh = kzalloc(len, GFP_KERNEL);
+	fh = kzalloc(len + OVL_FH_WIRE_OFFSET, GFP_KERNEL);
 	if (!fh)
 		goto fail;
 
 	err = -EINVAL;
-	if (hex2bin((u8 *)fh, index->d_name.name, len))
+	if (hex2bin(fh->buf, index->d_name.name, len))
 		goto fail;
 
-	err = ovl_check_fh_len(fh, len);
+	err = ovl_check_fb_len(&fh->fb, len);
 	if (err)
 		goto fail;
 
@@ -583,12 +589,12 @@ out:
 	return err;
 
 fail:
-	pr_warn_ratelimited("overlayfs: failed to verify index (%pd2, ftype=%x, err=%i)\n",
+	pr_warn_ratelimited("failed to verify index (%pd2, ftype=%x, err=%i)\n",
 			    index, d_inode(index)->i_mode & S_IFMT, err);
 	goto out;
 
 orphan:
-	pr_warn_ratelimited("overlayfs: orphan index entry (%pd2, ftype=%x, nlink=%u)\n",
+	pr_warn_ratelimited("orphan index entry (%pd2, ftype=%x, nlink=%u)\n",
 			    index, d_inode(index)->i_mode & S_IFMT,
 			    d_inode(index)->i_nlink);
 	err = -ENOENT;
@@ -599,11 +605,11 @@ static int ovl_get_index_name_fh(struct ovl_fh *fh, struct qstr *name)
 {
 	char *n, *s;
 
-	n = kcalloc(fh->len, 2, GFP_KERNEL);
+	n = kcalloc(fh->fb.len, 2, GFP_KERNEL);
 	if (!n)
 		return -ENOMEM;
 
-	s  = bin2hex(n, fh, fh->len);
+	s  = bin2hex(n, fh->buf, fh->fb.len);
 	*name = (struct qstr) QSTR_INIT(n, s - n);
 
 	return 0;
@@ -651,7 +657,7 @@ struct dentry *ovl_get_index_fh(struct ovl_fs *ofs, struct ovl_fh *fh)
 	if (err)
 		return ERR_PTR(err);
 
-	index = lookup_one_len_unlocked(name.name, ofs->indexdir, name.len);
+	index = lookup_positive_unlocked(name.name, ofs->indexdir, name.len);
 	kfree(name.name);
 	if (IS_ERR(index)) {
 		if (PTR_ERR(index) == -ENOENT)
@@ -659,9 +665,7 @@ struct dentry *ovl_get_index_fh(struct ovl_fs *ofs, struct ovl_fh *fh)
 		return index;
 	}
 
-	if (d_is_negative(index))
-		err = 0;
-	else if (ovl_is_whiteout(index))
+	if (ovl_is_whiteout(index))
 		err = -ESTALE;
 	else if (ovl_dentry_weird(index))
 		err = -EIO;
@@ -685,14 +689,14 @@ struct dentry *ovl_lookup_index(struct ovl_fs *ofs, struct dentry *upper,
 	if (err)
 		return ERR_PTR(err);
 
-	index = lookup_one_len_unlocked(name.name, ofs->indexdir, name.len);
+	index = lookup_positive_unlocked(name.name, ofs->indexdir, name.len);
 	if (IS_ERR(index)) {
 		err = PTR_ERR(index);
 		if (err == -ENOENT) {
 			index = NULL;
 			goto out;
 		}
-		pr_warn_ratelimited("overlayfs: failed inode index lookup (ino=%lu, key=%.*s, err=%i);\n"
+		pr_warn_ratelimited("failed inode index lookup (ino=%lu, key=%.*s, err=%i);\n"
 				    "overlayfs: mount with '-o index=off' to disable inodes index.\n",
 				    d_inode(origin)->i_ino, name.len, name.name,
 				    err);
@@ -700,9 +704,7 @@ struct dentry *ovl_lookup_index(struct ovl_fs *ofs, struct dentry *upper,
 	}
 
 	inode = d_inode(index);
-	if (d_is_negative(index)) {
-		goto out_dput;
-	} else if (ovl_is_whiteout(index) && !verify) {
+	if (ovl_is_whiteout(index) && !verify) {
 		/*
 		 * When index lookup is called with !verify for decoding an
 		 * overlay file handle, a whiteout index implies that decode
@@ -721,13 +723,13 @@ struct dentry *ovl_lookup_index(struct ovl_fs *ofs, struct dentry *upper,
 		 * unlinked, which means that finding a lower origin on lookup
 		 * whose index is a whiteout should be treated as an error.
 		 */
-		pr_warn_ratelimited("overlayfs: bad index found (index=%pd2, ftype=%x, origin ftype=%x).\n",
+		pr_warn_ratelimited("bad index found (index=%pd2, ftype=%x, origin ftype=%x).\n",
 				    index, d_inode(index)->i_mode & S_IFMT,
 				    d_inode(origin)->i_mode & S_IFMT);
 		goto fail;
 	} else if (is_dir && verify) {
 		if (!upper) {
-			pr_warn_ratelimited("overlayfs: suspected uncovered redirected dir found (origin=%pd2, index=%pd2).\n",
+			pr_warn_ratelimited("suspected uncovered redirected dir found (origin=%pd2, index=%pd2).\n",
 					    origin, index);
 			goto fail;
 		}
@@ -736,7 +738,7 @@ struct dentry *ovl_lookup_index(struct ovl_fs *ofs, struct dentry *upper,
 		err = ovl_verify_upper(index, upper, false);
 		if (err) {
 			if (err == -ESTALE) {
-				pr_warn_ratelimited("overlayfs: suspected multiply redirected dir found (upper=%pd2, origin=%pd2, index=%pd2).\n",
+				pr_warn_ratelimited("suspected multiply redirected dir found (upper=%pd2, origin=%pd2, index=%pd2).\n",
 						    upper, origin, index);
 			}
 			goto fail;
@@ -843,7 +845,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 		if (err)
 			goto out;
 
-		if (upperdentry && unlikely(ovl_dentry_remote(upperdentry))) {
+		if (upperdentry && upperdentry->d_flags & DCACHE_OP_REAL) {
 			dput(upperdentry);
 			err = -EREMOTE;
 			goto out;
@@ -883,7 +885,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 
 	if (!d.stop && poe->numlower) {
 		err = -ENOMEM;
-		stack = kcalloc(ofs->numlower, sizeof(struct ovl_path),
+		stack = kcalloc(ofs->numlayer - 1, sizeof(struct ovl_path),
 				GFP_KERNEL);
 		if (!stack)
 			goto out_put_upper;
@@ -965,7 +967,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 		 */
 		err = -EPERM;
 		if (d.redirect && !ofs->config.redirect_follow) {
-			pr_warn_ratelimited("overlayfs: refusing to follow redirect for (%pd2)\n",
+			pr_warn_ratelimited("refusing to follow redirect for (%pd2)\n",
 					    dentry);
 			goto out_put;
 		}
@@ -992,7 +994,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 
 		err = -EPERM;
 		if (!ofs->config.metacopy) {
-			pr_warn_ratelimited("overlay: refusing to follow metacopy origin for (%pd2)\n",
+			pr_warn_ratelimited("refusing to follow metacopy origin for (%pd2)\n",
 					    dentry);
 			goto out_put;
 		}
@@ -1074,6 +1076,9 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 			goto out_free_oe;
 	}
 
+	ovl_dentry_update_reval(dentry, upperdentry,
+			DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
+
 	revert_creds(old_cred);
 	if (origin_path) {
 		dput(origin_path->dentry);
@@ -1131,7 +1136,7 @@ bool ovl_lower_positive(struct dentry *dentry)
 		struct dentry *this;
 		struct dentry *lowerdir = poe->lowerstack[i].dentry;
 
-		this = lookup_one_len_unlocked(name->name, lowerdir,
+		this = lookup_positive_unlocked(name->name, lowerdir,
 					       name->len);
 		if (IS_ERR(this)) {
 			switch (PTR_ERR(this)) {
@@ -1148,10 +1153,8 @@ bool ovl_lower_positive(struct dentry *dentry)
 				break;
 			}
 		} else {
-			if (this->d_inode) {
-				positive = !ovl_is_whiteout(this);
-				done = true;
-			}
+			positive = !ovl_is_whiteout(this);
+			done = true;
 			dput(this);
 		}
 	}
diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h
index 6934bcf030f0..e6f3670146ed 100644
--- a/fs/overlayfs/overlayfs.h
+++ b/fs/overlayfs/overlayfs.h
@@ -9,6 +9,9 @@
 #include <linux/fs.h>
 #include "ovl_entry.h"
 
+#undef pr_fmt
+#define pr_fmt(fmt) "overlayfs: " fmt
+
 enum ovl_path_type {
 	__OVL_PATH_UPPER	= (1 << 0),
 	__OVL_PATH_MERGE	= (1 << 1),
@@ -45,6 +48,12 @@ enum ovl_entry_flag {
 	OVL_E_CONNECTED,
 };
 
+enum {
+	OVL_XINO_OFF,
+	OVL_XINO_AUTO,
+	OVL_XINO_ON,
+};
+
 /*
  * The tuple (fh,uuid) is a universal unique identifier for a copy up origin,
  * where:
@@ -71,20 +80,36 @@ enum ovl_entry_flag {
 #error Endianness not defined
 #endif
 
-/* The type returned by overlay exportfs ops when encoding an ovl_fh handle */
-#define OVL_FILEID	0xfb
+/* The type used to be returned by overlay exportfs for misaligned fid */
+#define OVL_FILEID_V0	0xfb
+/* The type returned by overlay exportfs for 32bit aligned fid */
+#define OVL_FILEID_V1	0xf8
 
-/* On-disk and in-memeory format for redirect by file handle */
-struct ovl_fh {
+/* On-disk format for "origin" file handle */
+struct ovl_fb {
 	u8 version;	/* 0 */
 	u8 magic;	/* 0xfb */
 	u8 len;		/* size of this header + size of fid */
 	u8 flags;	/* OVL_FH_FLAG_* */
 	u8 type;	/* fid_type of fid */
 	uuid_t uuid;	/* uuid of filesystem */
-	u8 fid[0];	/* file identifier */
+	u32 fid[];	/* file identifier should be 32bit aligned in-memory */
+} __packed;
+
+/* In-memory and on-wire format for overlay file handle */
+struct ovl_fh {
+	u8 padding[3];	/* make sure fb.fid is 32bit aligned */
+	union {
+		struct ovl_fb fb;
+		u8 buf[0];
+	};
 } __packed;
 
+#define OVL_FH_WIRE_OFFSET	offsetof(struct ovl_fh, fb)
+#define OVL_FH_LEN(fh)		(OVL_FH_WIRE_OFFSET + (fh)->fb.len)
+#define OVL_FH_FID_OFFSET	(OVL_FH_WIRE_OFFSET + \
+				 offsetof(struct ovl_fb, fid))
+
 static inline int ovl_do_rmdir(struct inode *dir, struct dentry *dentry)
 {
 	int err = vfs_rmdir(dir, dentry);
@@ -205,13 +230,14 @@ int ovl_want_write(struct dentry *dentry);
 void ovl_drop_write(struct dentry *dentry);
 struct dentry *ovl_workdir(struct dentry *dentry);
 const struct cred *ovl_override_creds(struct super_block *sb);
-struct super_block *ovl_same_sb(struct super_block *sb);
 int ovl_can_decode_fh(struct super_block *sb);
 struct dentry *ovl_indexdir(struct super_block *sb);
 bool ovl_index_all(struct super_block *sb);
 bool ovl_verify_lower(struct super_block *sb);
 struct ovl_entry *ovl_alloc_entry(unsigned int numlower);
 bool ovl_dentry_remote(struct dentry *dentry);
+void ovl_dentry_update_reval(struct dentry *dentry, struct dentry *upperdentry,
+			     unsigned int mask);
 bool ovl_dentry_weird(struct dentry *dentry);
 enum ovl_path_type ovl_path_type(struct dentry *dentry);
 void ovl_path_upper(struct dentry *dentry, struct path *path);
@@ -221,7 +247,7 @@ enum ovl_path_type ovl_path_real(struct dentry *dentry, struct path *path);
 struct dentry *ovl_dentry_upper(struct dentry *dentry);
 struct dentry *ovl_dentry_lower(struct dentry *dentry);
 struct dentry *ovl_dentry_lowerdata(struct dentry *dentry);
-struct ovl_layer *ovl_layer_lower(struct dentry *dentry);
+const struct ovl_layer *ovl_layer_lower(struct dentry *dentry);
 struct dentry *ovl_dentry_real(struct dentry *dentry);
 struct dentry *ovl_i_dentry_upper(struct inode *inode);
 struct inode *ovl_inode_upper(struct inode *inode);
@@ -246,8 +272,6 @@ void ovl_set_upperdata(struct inode *inode);
 bool ovl_redirect_dir(struct super_block *sb);
 const char *ovl_dentry_get_redirect(struct dentry *dentry);
 void ovl_dentry_set_redirect(struct dentry *dentry, const char *redirect);
-void ovl_inode_init(struct inode *inode, struct dentry *upperdentry,
-		    struct dentry *lowerdentry, struct dentry *lowerdata);
 void ovl_inode_update(struct inode *inode, struct dentry *upperdentry);
 void ovl_dir_modified(struct dentry *dentry, bool impurity);
 u64 ovl_dentry_version_get(struct dentry *dentry);
@@ -283,14 +307,39 @@ static inline bool ovl_is_impuredir(struct dentry *dentry)
 	return ovl_check_dir_xattr(dentry, OVL_XATTR_IMPURE);
 }
 
+/*
+ * With xino=auto, we do best effort to keep all inodes on same st_dev and
+ * d_ino consistent with st_ino.
+ * With xino=on, we do the same effort but we warn if we failed.
+ */
+static inline bool ovl_xino_warn(struct super_block *sb)
+{
+	return OVL_FS(sb)->config.xino == OVL_XINO_ON;
+}
+
+/* All layers on same fs? */
+static inline bool ovl_same_fs(struct super_block *sb)
+{
+	return OVL_FS(sb)->xino_mode == 0;
+}
+
+/* All overlay inodes have same st_dev? */
+static inline bool ovl_same_dev(struct super_block *sb)
+{
+	return OVL_FS(sb)->xino_mode >= 0;
+}
+
 static inline unsigned int ovl_xino_bits(struct super_block *sb)
 {
-	struct ovl_fs *ofs = sb->s_fs_info;
+	return ovl_same_dev(sb) ? OVL_FS(sb)->xino_mode : 0;
+}
 
-	return ofs->xino_bits;
+static inline void ovl_inode_lock(struct inode *inode)
+{
+	mutex_lock(&OVL_I(inode)->lock);
 }
 
-static inline int ovl_inode_lock(struct inode *inode)
+static inline int ovl_inode_lock_interruptible(struct inode *inode)
 {
 	return mutex_lock_interruptible(&OVL_I(inode)->lock);
 }
@@ -302,7 +351,13 @@ static inline void ovl_inode_unlock(struct inode *inode)
 
 
 /* namei.c */
-int ovl_check_fh_len(struct ovl_fh *fh, int fh_len);
+int ovl_check_fb_len(struct ovl_fb *fb, int fb_len);
+
+static inline int ovl_check_fh_len(struct ovl_fh *fh, int fh_len)
+{
+	return ovl_check_fb_len(&fh->fb, fh_len - OVL_FH_WIRE_OFFSET);
+}
+
 struct dentry *ovl_decode_real_fh(struct ovl_fh *fh, struct vfsmount *mnt,
 				  bool connected);
 int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, bool connected,
@@ -371,6 +426,8 @@ struct ovl_inode_params {
 	char *redirect;
 	struct dentry *lowerdata;
 };
+void ovl_inode_init(struct inode *inode, struct ovl_inode_params *oip,
+		    unsigned long ino, int fsid);
 struct inode *ovl_new_inode(struct super_block *sb, umode_t mode, dev_t rdev);
 struct inode *ovl_lookup_inode(struct super_block *sb, struct dentry *real,
 			       bool is_upper);
@@ -412,10 +469,13 @@ struct ovl_cattr {
 struct dentry *ovl_create_real(struct inode *dir, struct dentry *newdentry,
 			       struct ovl_cattr *attr);
 int ovl_cleanup(struct inode *dir, struct dentry *dentry);
+struct dentry *ovl_lookup_temp(struct dentry *workdir);
 struct dentry *ovl_create_temp(struct dentry *workdir, struct ovl_cattr *attr);
 
 /* file.c */
 extern const struct file_operations ovl_file_operations;
+int __init ovl_aio_request_cache_init(void);
+void ovl_aio_request_cache_destroy(void);
 
 /* copy_up.c */
 int ovl_copy_up(struct dentry *dentry);
diff --git a/fs/overlayfs/ovl_entry.h b/fs/overlayfs/ovl_entry.h
index a8279280e88d..5762d802fe01 100644
--- a/fs/overlayfs/ovl_entry.h
+++ b/fs/overlayfs/ovl_entry.h
@@ -22,6 +22,10 @@ struct ovl_config {
 struct ovl_sb {
 	struct super_block *sb;
 	dev_t pseudo_dev;
+	/* Unusable (conflicting) uuid */
+	bool bad_uuid;
+	/* Used as a lower layer (but maybe also as upper) */
+	bool is_lower;
 };
 
 struct ovl_layer {
@@ -36,18 +40,18 @@ struct ovl_layer {
 };
 
 struct ovl_path {
-	struct ovl_layer *layer;
+	const struct ovl_layer *layer;
 	struct dentry *dentry;
 };
 
 /* private information held for overlayfs's superblock */
 struct ovl_fs {
 	struct vfsmount *upper_mnt;
-	unsigned int numlower;
-	/* Number of unique lower sb that differ from upper sb */
-	unsigned int numlowerfs;
-	struct ovl_layer *lower_layers;
-	struct ovl_sb *lower_fs;
+	unsigned int numlayer;
+	/* Number of unique fs among layers including upper fs */
+	unsigned int numfs;
+	const struct ovl_layer *layers;
+	struct ovl_sb *fs;
 	/* workbasedir is the path at workdir= mount option */
 	struct dentry *workbasedir;
 	/* workdir is the 'work' directory under workbasedir */
@@ -69,10 +73,17 @@ struct ovl_fs {
 	struct inode *workbasedir_trap;
 	struct inode *workdir_trap;
 	struct inode *indexdir_trap;
-	/* Inode numbers in all layers do not use the high xino_bits */
-	unsigned int xino_bits;
+	/* -1: disabled, 0: same fs, 1..32: number of unused ino bits */
+	int xino_mode;
+	/* For allocation of non-persistent inode numbers */
+	atomic_long_t last_ino;
 };
 
+static inline struct ovl_fs *OVL_FS(struct super_block *sb)
+{
+	return (struct ovl_fs *)sb->s_fs_info;
+}
+
 /* private information held for every overlayfs dentry */
 struct ovl_entry {
 	union {
diff --git a/fs/overlayfs/readdir.c b/fs/overlayfs/readdir.c
index 47a91c9733a5..e452ff7d583d 100644
--- a/fs/overlayfs/readdir.c
+++ b/fs/overlayfs/readdir.c
@@ -438,15 +438,23 @@ static struct ovl_dir_cache *ovl_cache_get(struct dentry *dentry)
 
 /* Map inode number to lower fs unique range */
 static u64 ovl_remap_lower_ino(u64 ino, int xinobits, int fsid,
-			       const char *name, int namelen)
+			       const char *name, int namelen, bool warn)
 {
-	if (ino >> (64 - xinobits)) {
-		pr_warn_ratelimited("overlayfs: d_ino too big (%.*s, ino=%llu, xinobits=%d)\n",
-				    namelen, name, ino, xinobits);
+	unsigned int xinoshift = 64 - xinobits;
+
+	if (unlikely(ino >> xinoshift)) {
+		if (warn) {
+			pr_warn_ratelimited("d_ino too big (%.*s, ino=%llu, xinobits=%d)\n",
+					    namelen, name, ino, xinobits);
+		}
 		return ino;
 	}
 
-	return ino | ((u64)fsid) << (64 - xinobits);
+	/*
+	 * The lowest xinobit is reserved for mapping the non-peresistent inode
+	 * numbers range, but this range is only exposed via st_ino, not here.
+	 */
+	return ino | ((u64)fsid) << (xinoshift + 1);
 }
 
 /*
@@ -469,7 +477,7 @@ static int ovl_cache_update_ino(struct path *path, struct ovl_cache_entry *p)
 	int xinobits = ovl_xino_bits(dir->d_sb);
 	int err = 0;
 
-	if (!ovl_same_sb(dir->d_sb) && !xinobits)
+	if (!ovl_same_dev(dir->d_sb))
 		goto out;
 
 	if (p->name[0] == '.') {
@@ -504,12 +512,19 @@ get:
 		if (err)
 			goto fail;
 
-		WARN_ON_ONCE(dir->d_sb->s_dev != stat.dev);
+		/*
+		 * Directory inode is always on overlay st_dev.
+		 * Non-dir with ovl_same_dev() could be on pseudo st_dev in case
+		 * of xino bits overflow.
+		 */
+		WARN_ON_ONCE(S_ISDIR(stat.mode) &&
+			     dir->d_sb->s_dev != stat.dev);
 		ino = stat.ino;
 	} else if (xinobits && !OVL_TYPE_UPPER(type)) {
 		ino = ovl_remap_lower_ino(ino, xinobits,
 					  ovl_layer_lower(this)->fsid,
-					  p->name, p->len);
+					  p->name, p->len,
+					  ovl_xino_warn(dir->d_sb));
 	}
 
 out:
@@ -518,7 +533,7 @@ out:
 	return err;
 
 fail:
-	pr_warn_ratelimited("overlayfs: failed to look up (%s) for ino (%i)\n",
+	pr_warn_ratelimited("failed to look up (%s) for ino (%i)\n",
 			    p->name, err);
 	goto out;
 }
@@ -639,6 +654,7 @@ struct ovl_readdir_translate {
 	u64 parent_ino;
 	int fsid;
 	int xinobits;
+	bool xinowarn;
 };
 
 static int ovl_fill_real(struct dir_context *ctx, const char *name,
@@ -659,7 +675,7 @@ static int ovl_fill_real(struct dir_context *ctx, const char *name,
 			ino = p->ino;
 	} else if (rdt->xinobits) {
 		ino = ovl_remap_lower_ino(ino, rdt->xinobits, rdt->fsid,
-					  name, namelen);
+					  name, namelen, rdt->xinowarn);
 	}
 
 	return orig_ctx->actor(orig_ctx, name, namelen, offset, ino, d_type);
@@ -685,11 +701,12 @@ static int ovl_iterate_real(struct file *file, struct dir_context *ctx)
 	int err;
 	struct ovl_dir_file *od = file->private_data;
 	struct dentry *dir = file->f_path.dentry;
-	struct ovl_layer *lower_layer = ovl_layer_lower(dir);
+	const struct ovl_layer *lower_layer = ovl_layer_lower(dir);
 	struct ovl_readdir_translate rdt = {
 		.ctx.actor = ovl_fill_real,
 		.orig_ctx = ctx,
 		.xinobits = ovl_xino_bits(dir->d_sb),
+		.xinowarn = ovl_xino_warn(dir->d_sb),
 	};
 
 	if (rdt.xinobits && lower_layer)
@@ -738,7 +755,7 @@ static int ovl_iterate(struct file *file, struct dir_context *ctx)
 		 * entries.
 		 */
 		if (ovl_xino_bits(dentry->d_sb) ||
-		    (ovl_same_sb(dentry->d_sb) &&
+		    (ovl_same_fs(dentry->d_sb) &&
 		     (ovl_is_impure_dir(file) ||
 		      OVL_TYPE_MERGE(ovl_path_type(dentry->d_parent))))) {
 			return ovl_iterate_real(file, ctx);
@@ -965,7 +982,7 @@ void ovl_cleanup_whiteouts(struct dentry *upper, struct list_head *list)
 
 		dentry = lookup_one_len(p->name, upper, p->len);
 		if (IS_ERR(dentry)) {
-			pr_err("overlayfs: lookup '%s/%.*s' failed (%i)\n",
+			pr_err("lookup '%s/%.*s' failed (%i)\n",
 			       upper->d_name.name, p->len, p->name,
 			       (int) PTR_ERR(dentry));
 			continue;
@@ -1147,6 +1164,6 @@ next:
 out:
 	ovl_cache_free(&list);
 	if (err)
-		pr_err("overlayfs: failed index dir cleanup (%i)\n", err);
+		pr_err("failed index dir cleanup (%i)\n", err);
 	return err;
 }
diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index afbcb116a7f1..732ad5495c92 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -113,53 +113,54 @@ bug:
 	return dentry;
 }
 
-static int ovl_dentry_revalidate(struct dentry *dentry, unsigned int flags)
+static int ovl_revalidate_real(struct dentry *d, unsigned int flags, bool weak)
 {
-	struct ovl_entry *oe = dentry->d_fsdata;
-	unsigned int i;
 	int ret = 1;
 
-	for (i = 0; i < oe->numlower; i++) {
-		struct dentry *d = oe->lowerstack[i].dentry;
-
-		if (d->d_flags & DCACHE_OP_REVALIDATE) {
-			ret = d->d_op->d_revalidate(d, flags);
-			if (ret < 0)
-				return ret;
-			if (!ret) {
-				if (!(flags & LOOKUP_RCU))
-					d_invalidate(d);
-				return -ESTALE;
-			}
+	if (weak) {
+		if (d->d_flags & DCACHE_OP_WEAK_REVALIDATE)
+			ret =  d->d_op->d_weak_revalidate(d, flags);
+	} else if (d->d_flags & DCACHE_OP_REVALIDATE) {
+		ret = d->d_op->d_revalidate(d, flags);
+		if (!ret) {
+			if (!(flags & LOOKUP_RCU))
+				d_invalidate(d);
+			ret = -ESTALE;
 		}
 	}
-	return 1;
+	return ret;
 }
 
-static int ovl_dentry_weak_revalidate(struct dentry *dentry, unsigned int flags)
+static int ovl_dentry_revalidate_common(struct dentry *dentry,
+					unsigned int flags, bool weak)
 {
 	struct ovl_entry *oe = dentry->d_fsdata;
+	struct dentry *upper;
 	unsigned int i;
 	int ret = 1;
 
-	for (i = 0; i < oe->numlower; i++) {
-		struct dentry *d = oe->lowerstack[i].dentry;
+	upper = ovl_dentry_upper(dentry);
+	if (upper)
+		ret = ovl_revalidate_real(upper, flags, weak);
 
-		if (d->d_flags & DCACHE_OP_WEAK_REVALIDATE) {
-			ret = d->d_op->d_weak_revalidate(d, flags);
-			if (ret <= 0)
-				break;
-		}
+	for (i = 0; ret > 0 && i < oe->numlower; i++) {
+		ret = ovl_revalidate_real(oe->lowerstack[i].dentry, flags,
+					  weak);
 	}
 	return ret;
 }
 
-static const struct dentry_operations ovl_dentry_operations = {
-	.d_release = ovl_dentry_release,
-	.d_real = ovl_d_real,
-};
+static int ovl_dentry_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	return ovl_dentry_revalidate_common(dentry, flags, false);
+}
 
-static const struct dentry_operations ovl_reval_dentry_operations = {
+static int ovl_dentry_weak_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	return ovl_dentry_revalidate_common(dentry, flags, true);
+}
+
+static const struct dentry_operations ovl_dentry_operations = {
 	.d_release = ovl_dentry_release,
 	.d_real = ovl_d_real,
 	.d_revalidate = ovl_dentry_revalidate,
@@ -224,14 +225,14 @@ static void ovl_free_fs(struct ovl_fs *ofs)
 	if (ofs->upperdir_locked)
 		ovl_inuse_unlock(ofs->upper_mnt->mnt_root);
 	mntput(ofs->upper_mnt);
-	for (i = 0; i < ofs->numlower; i++) {
-		iput(ofs->lower_layers[i].trap);
-		mntput(ofs->lower_layers[i].mnt);
+	for (i = 1; i < ofs->numlayer; i++) {
+		iput(ofs->layers[i].trap);
+		mntput(ofs->layers[i].mnt);
 	}
-	for (i = 0; i < ofs->numlowerfs; i++)
-		free_anon_bdev(ofs->lower_fs[i].pseudo_dev);
-	kfree(ofs->lower_layers);
-	kfree(ofs->lower_fs);
+	kfree(ofs->layers);
+	for (i = 0; i < ofs->numfs; i++)
+		free_anon_bdev(ofs->fs[i].pseudo_dev);
+	kfree(ofs->fs);
 
 	kfree(ofs->config.lowerdir);
 	kfree(ofs->config.upperdir);
@@ -316,12 +317,6 @@ static const char *ovl_redirect_mode_def(void)
 	return ovl_redirect_dir_def ? "on" : "off";
 }
 
-enum {
-	OVL_XINO_OFF,
-	OVL_XINO_AUTO,
-	OVL_XINO_ON,
-};
-
 static const char * const ovl_xino_str[] = {
 	"off",
 	"auto",
@@ -358,7 +353,7 @@ static int ovl_show_options(struct seq_file *m, struct dentry *dentry)
 	if (ofs->config.nfs_export != ovl_nfs_export_def)
 		seq_printf(m, ",nfs_export=%s", ofs->config.nfs_export ?
 						"on" : "off");
-	if (ofs->config.xino != ovl_xino_def())
+	if (ofs->config.xino != ovl_xino_def() && !ovl_same_fs(sb))
 		seq_printf(m, ",xino=%s", ovl_xino_str[ofs->config.xino]);
 	if (ofs->config.metacopy != ovl_metacopy_def)
 		seq_printf(m, ",metacopy=%s",
@@ -462,7 +457,7 @@ static int ovl_parse_redirect_mode(struct ovl_config *config, const char *mode)
 		if (ovl_redirect_always_follow)
 			config->redirect_follow = true;
 	} else if (strcmp(mode, "nofollow") != 0) {
-		pr_err("overlayfs: bad mount option \"redirect_dir=%s\"\n",
+		pr_err("bad mount option \"redirect_dir=%s\"\n",
 		       mode);
 		return -EINVAL;
 	}
@@ -560,14 +555,15 @@ static int ovl_parse_opt(char *opt, struct ovl_config *config)
 			break;
 
 		default:
-			pr_err("overlayfs: unrecognized mount option \"%s\" or missing value\n", p);
+			pr_err("unrecognized mount option \"%s\" or missing value\n",
+					p);
 			return -EINVAL;
 		}
 	}
 
 	/* Workdir is useless in non-upper mount */
 	if (!config->upperdir && config->workdir) {
-		pr_info("overlayfs: option \"workdir=%s\" is useless in a non-upper mount, ignore\n",
+		pr_info("option \"workdir=%s\" is useless in a non-upper mount, ignore\n",
 			config->workdir);
 		kfree(config->workdir);
 		config->workdir = NULL;
@@ -587,7 +583,7 @@ static int ovl_parse_opt(char *opt, struct ovl_config *config)
 	/* Resolve metacopy -> redirect_dir dependency */
 	if (config->metacopy && !config->redirect_dir) {
 		if (metacopy_opt && redirect_opt) {
-			pr_err("overlayfs: conflicting options: metacopy=on,redirect_dir=%s\n",
+			pr_err("conflicting options: metacopy=on,redirect_dir=%s\n",
 			       config->redirect_mode);
 			return -EINVAL;
 		}
@@ -596,7 +592,7 @@ static int ovl_parse_opt(char *opt, struct ovl_config *config)
 			 * There was an explicit redirect_dir=... that resulted
 			 * in this conflict.
 			 */
-			pr_info("overlayfs: disabling metacopy due to redirect_dir=%s\n",
+			pr_info("disabling metacopy due to redirect_dir=%s\n",
 				config->redirect_mode);
 			config->metacopy = false;
 		} else {
@@ -692,7 +688,7 @@ out_unlock:
 out_dput:
 	dput(work);
 out_err:
-	pr_warn("overlayfs: failed to create directory %s/%s (errno: %i); mounting read-only\n",
+	pr_warn("failed to create directory %s/%s (errno: %i); mounting read-only\n",
 		ofs->config.workdir, name, -err);
 	work = NULL;
 	goto out_unlock;
@@ -716,21 +712,21 @@ static int ovl_mount_dir_noesc(const char *name, struct path *path)
 	int err = -EINVAL;
 
 	if (!*name) {
-		pr_err("overlayfs: empty lowerdir\n");
+		pr_err("empty lowerdir\n");
 		goto out;
 	}
 	err = kern_path(name, LOOKUP_FOLLOW, path);
 	if (err) {
-		pr_err("overlayfs: failed to resolve '%s': %i\n", name, err);
+		pr_err("failed to resolve '%s': %i\n", name, err);
 		goto out;
 	}
 	err = -EINVAL;
 	if (ovl_dentry_weird(path->dentry)) {
-		pr_err("overlayfs: filesystem on '%s' not supported\n", name);
+		pr_err("filesystem on '%s' not supported\n", name);
 		goto out_put;
 	}
 	if (!d_is_dir(path->dentry)) {
-		pr_err("overlayfs: '%s' not a directory\n", name);
+		pr_err("'%s' not a directory\n", name);
 		goto out_put;
 	}
 	return 0;
@@ -750,13 +746,12 @@ static int ovl_mount_dir(const char *name, struct path *path)
 		ovl_unescape(tmp);
 		err = ovl_mount_dir_noesc(tmp, path);
 
-		if (!err)
-			if (ovl_dentry_remote(path->dentry)) {
-				pr_err("overlayfs: filesystem on '%s' not supported as upperdir\n",
-				       tmp);
-				path_put_init(path);
-				err = -EINVAL;
-			}
+		if (!err && path->dentry->d_flags & DCACHE_OP_REAL) {
+			pr_err("filesystem on '%s' not supported as upperdir\n",
+			       tmp);
+			path_put_init(path);
+			err = -EINVAL;
+		}
 		kfree(tmp);
 	}
 	return err;
@@ -769,7 +764,7 @@ static int ovl_check_namelen(struct path *path, struct ovl_fs *ofs,
 	int err = vfs_statfs(path, &statfs);
 
 	if (err)
-		pr_err("overlayfs: statfs failed on '%s'\n", name);
+		pr_err("statfs failed on '%s'\n", name);
 	else
 		ofs->namelen = max(ofs->namelen, statfs.f_namelen);
 
@@ -777,7 +772,7 @@ static int ovl_check_namelen(struct path *path, struct ovl_fs *ofs,
 }
 
 static int ovl_lower_dir(const char *name, struct path *path,
-			 struct ovl_fs *ofs, int *stack_depth, bool *remote)
+			 struct ovl_fs *ofs, int *stack_depth)
 {
 	int fh_type;
 	int err;
@@ -792,9 +787,6 @@ static int ovl_lower_dir(const char *name, struct path *path,
 
 	*stack_depth = max(*stack_depth, path->mnt->mnt_sb->s_stack_depth);
 
-	if (ovl_dentry_remote(path->dentry))
-		*remote = true;
-
 	/*
 	 * The inodes index feature and NFS export need to encode and decode
 	 * file handles, so they require that all layers support them.
@@ -804,13 +796,13 @@ static int ovl_lower_dir(const char *name, struct path *path,
 	     (ofs->config.index && ofs->config.upperdir)) && !fh_type) {
 		ofs->config.index = false;
 		ofs->config.nfs_export = false;
-		pr_warn("overlayfs: fs on '%s' does not support file handles, falling back to index=off,nfs_export=off.\n",
+		pr_warn("fs on '%s' does not support file handles, falling back to index=off,nfs_export=off.\n",
 			name);
 	}
 
 	/* Check if lower fs has 32bit inode numbers */
 	if (fh_type != FILEID_INO32_GEN)
-		ofs->xino_bits = 0;
+		ofs->xino_mode = -1;
 
 	return 0;
 
@@ -996,7 +988,7 @@ static int ovl_setup_trap(struct super_block *sb, struct dentry *dir,
 	err = PTR_ERR_OR_ZERO(trap);
 	if (err) {
 		if (err == -ELOOP)
-			pr_err("overlayfs: conflicting %s path\n", name);
+			pr_err("conflicting %s path\n", name);
 		return err;
 	}
 
@@ -1013,11 +1005,11 @@ static int ovl_setup_trap(struct super_block *sb, struct dentry *dir,
 static int ovl_report_in_use(struct ovl_fs *ofs, const char *name)
 {
 	if (ofs->config.index) {
-		pr_err("overlayfs: %s is in-use as upperdir/workdir of another mount, mount with '-o index=off' to override exclusive upperdir protection.\n",
+		pr_err("%s is in-use as upperdir/workdir of another mount, mount with '-o index=off' to override exclusive upperdir protection.\n",
 		       name);
 		return -EBUSY;
 	} else {
-		pr_warn("overlayfs: %s is in-use as upperdir/workdir of another mount, accessing files from both mounts will result in undefined behavior.\n",
+		pr_warn("%s is in-use as upperdir/workdir of another mount, accessing files from both mounts will result in undefined behavior.\n",
 			name);
 		return 0;
 	}
@@ -1035,7 +1027,7 @@ static int ovl_get_upper(struct super_block *sb, struct ovl_fs *ofs,
 
 	/* Upper fs should not be r/o */
 	if (sb_rdonly(upperpath->mnt->mnt_sb)) {
-		pr_err("overlayfs: upper fs is r/o, try multi-lower layers mount\n");
+		pr_err("upper fs is r/o, try multi-lower layers mount\n");
 		err = -EINVAL;
 		goto out;
 	}
@@ -1052,7 +1044,7 @@ static int ovl_get_upper(struct super_block *sb, struct ovl_fs *ofs,
 	upper_mnt = clone_private_mount(upperpath);
 	err = PTR_ERR(upper_mnt);
 	if (IS_ERR(upper_mnt)) {
-		pr_err("overlayfs: failed to clone upperpath\n");
+		pr_err("failed to clone upperpath\n");
 		goto out;
 	}
 
@@ -1073,11 +1065,73 @@ out:
 	return err;
 }
 
+/*
+ * Returns 1 if RENAME_WHITEOUT is supported, 0 if not supported and
+ * negative values if error is encountered.
+ */
+static int ovl_check_rename_whiteout(struct dentry *workdir)
+{
+	struct inode *dir = d_inode(workdir);
+	struct dentry *temp;
+	struct dentry *dest;
+	struct dentry *whiteout;
+	struct name_snapshot name;
+	int err;
+
+	inode_lock_nested(dir, I_MUTEX_PARENT);
+
+	temp = ovl_create_temp(workdir, OVL_CATTR(S_IFREG | 0));
+	err = PTR_ERR(temp);
+	if (IS_ERR(temp))
+		goto out_unlock;
+
+	dest = ovl_lookup_temp(workdir);
+	err = PTR_ERR(dest);
+	if (IS_ERR(dest)) {
+		dput(temp);
+		goto out_unlock;
+	}
+
+	/* Name is inline and stable - using snapshot as a copy helper */
+	take_dentry_name_snapshot(&name, temp);
+	err = ovl_do_rename(dir, temp, dir, dest, RENAME_WHITEOUT);
+	if (err) {
+		if (err == -EINVAL)
+			err = 0;
+		goto cleanup_temp;
+	}
+
+	whiteout = lookup_one_len(name.name.name, workdir, name.name.len);
+	err = PTR_ERR(whiteout);
+	if (IS_ERR(whiteout))
+		goto cleanup_temp;
+
+	err = ovl_is_whiteout(whiteout);
+
+	/* Best effort cleanup of whiteout and temp file */
+	if (err)
+		ovl_cleanup(dir, whiteout);
+	dput(whiteout);
+
+cleanup_temp:
+	ovl_cleanup(dir, temp);
+	release_dentry_name_snapshot(&name);
+	dput(temp);
+	dput(dest);
+
+out_unlock:
+	inode_unlock(dir);
+
+	return err;
+}
+
 static int ovl_make_workdir(struct super_block *sb, struct ovl_fs *ofs,
 			    struct path *workpath)
 {
 	struct vfsmount *mnt = ofs->upper_mnt;
 	struct dentry *temp;
+	bool rename_whiteout;
+	bool d_type;
 	int fh_type;
 	int err;
 
@@ -1103,12 +1157,9 @@ static int ovl_make_workdir(struct super_block *sb, struct ovl_fs *ofs,
 	if (err < 0)
 		goto out;
 
-	/*
-	 * We allowed this configuration and don't want to break users over
-	 * kernel upgrade. So warn instead of erroring out.
-	 */
-	if (!err)
-		pr_warn("overlayfs: upper fs needs to support d_type.\n");
+	d_type = err;
+	if (!d_type)
+		pr_warn("upper fs needs to support d_type.\n");
 
 	/* Check if upper/work fs supports O_TMPFILE */
 	temp = ovl_do_tmpfile(ofs->workdir, S_IFREG | 0);
@@ -1116,7 +1167,17 @@ static int ovl_make_workdir(struct super_block *sb, struct ovl_fs *ofs,
 	if (ofs->tmpfile)
 		dput(temp);
 	else
-		pr_warn("overlayfs: upper fs does not support tmpfile.\n");
+		pr_warn("upper fs does not support tmpfile.\n");
+
+
+	/* Check if upper/work fs supports RENAME_WHITEOUT */
+	err = ovl_check_rename_whiteout(ofs->workdir);
+	if (err < 0)
+		goto out;
+
+	rename_whiteout = err;
+	if (!rename_whiteout)
+		pr_warn("upper fs does not support RENAME_WHITEOUT.\n");
 
 	/*
 	 * Check if upper/work fs supports trusted.overlay.* xattr
@@ -1126,26 +1187,38 @@ static int ovl_make_workdir(struct super_block *sb, struct ovl_fs *ofs,
 		ofs->noxattr = true;
 		ofs->config.index = false;
 		ofs->config.metacopy = false;
-		pr_warn("overlayfs: upper fs does not support xattr, falling back to index=off and metacopy=off.\n");
+		pr_warn("upper fs does not support xattr, falling back to index=off and metacopy=off.\n");
 		err = 0;
 	} else {
 		vfs_removexattr(ofs->workdir, OVL_XATTR_OPAQUE);
 	}
 
+	/*
+	 * We allowed sub-optimal upper fs configuration and don't want to break
+	 * users over kernel upgrade, but we never allowed remote upper fs, so
+	 * we can enforce strict requirements for remote upper fs.
+	 */
+	if (ovl_dentry_remote(ofs->workdir) &&
+	    (!d_type || !rename_whiteout || ofs->noxattr)) {
+		pr_err("upper fs missing required features.\n");
+		err = -EINVAL;
+		goto out;
+	}
+
 	/* Check if upper/work fs supports file handles */
 	fh_type = ovl_can_decode_fh(ofs->workdir->d_sb);
 	if (ofs->config.index && !fh_type) {
 		ofs->config.index = false;
-		pr_warn("overlayfs: upper fs does not support file handles, falling back to index=off.\n");
+		pr_warn("upper fs does not support file handles, falling back to index=off.\n");
 	}
 
 	/* Check if upper fs has 32bit inode numbers */
 	if (fh_type != FILEID_INO32_GEN)
-		ofs->xino_bits = 0;
+		ofs->xino_mode = -1;
 
 	/* NFS export of r/w mount depends on index */
 	if (ofs->config.nfs_export && !ofs->config.index) {
-		pr_warn("overlayfs: NFS export requires \"index=on\", falling back to nfs_export=off.\n");
+		pr_warn("NFS export requires \"index=on\", falling back to nfs_export=off.\n");
 		ofs->config.nfs_export = false;
 	}
 out:
@@ -1165,11 +1238,11 @@ static int ovl_get_workdir(struct super_block *sb, struct ovl_fs *ofs,
 
 	err = -EINVAL;
 	if (upperpath->mnt != workpath.mnt) {
-		pr_err("overlayfs: workdir and upperdir must reside under the same mount\n");
+		pr_err("workdir and upperdir must reside under the same mount\n");
 		goto out;
 	}
 	if (!ovl_workdir_ok(workpath.dentry, upperpath->dentry)) {
-		pr_err("overlayfs: workdir and upperdir must be separate subtrees\n");
+		pr_err("workdir and upperdir must be separate subtrees\n");
 		goto out;
 	}
 
@@ -1210,7 +1283,7 @@ static int ovl_get_indexdir(struct super_block *sb, struct ovl_fs *ofs,
 	err = ovl_verify_origin(upperpath->dentry, oe->lowerstack[0].dentry,
 				true);
 	if (err) {
-		pr_err("overlayfs: failed to verify upper root origin\n");
+		pr_err("failed to verify upper root origin\n");
 		goto out;
 	}
 
@@ -1233,18 +1306,18 @@ static int ovl_get_indexdir(struct super_block *sb, struct ovl_fs *ofs,
 			err = ovl_verify_set_fh(ofs->indexdir, OVL_XATTR_ORIGIN,
 						upperpath->dentry, true, false);
 			if (err)
-				pr_err("overlayfs: failed to verify index dir 'origin' xattr\n");
+				pr_err("failed to verify index dir 'origin' xattr\n");
 		}
 		err = ovl_verify_upper(ofs->indexdir, upperpath->dentry, true);
 		if (err)
-			pr_err("overlayfs: failed to verify index dir 'upper' xattr\n");
+			pr_err("failed to verify index dir 'upper' xattr\n");
 
 		/* Cleanup bad/stale/orphan index entries */
 		if (!err)
 			err = ovl_indexdir_cleanup(ofs);
 	}
 	if (err || !ofs->indexdir)
-		pr_warn("overlayfs: try deleting index dir or mounting with '-o index=off' to disable inodes index.\n");
+		pr_warn("try deleting index dir or mounting with '-o index=off' to disable inodes index.\n");
 
 out:
 	mnt_drop_write(mnt);
@@ -1255,17 +1328,22 @@ static bool ovl_lower_uuid_ok(struct ovl_fs *ofs, const uuid_t *uuid)
 {
 	unsigned int i;
 
-	if (!ofs->config.nfs_export && !(ofs->config.index && ofs->upper_mnt))
+	if (!ofs->config.nfs_export && !ofs->upper_mnt)
 		return true;
 
-	for (i = 0; i < ofs->numlowerfs; i++) {
+	for (i = 0; i < ofs->numfs; i++) {
 		/*
 		 * We use uuid to associate an overlay lower file handle with a
 		 * lower layer, so we can accept lower fs with null uuid as long
 		 * as all lower layers with null uuid are on the same fs.
+		 * if we detect multiple lower fs with the same uuid, we
+		 * disable lower file handle decoding on all of them.
 		 */
-		if (uuid_equal(&ofs->lower_fs[i].sb->s_uuid, uuid))
+		if (ofs->fs[i].is_lower &&
+		    uuid_equal(&ofs->fs[i].sb->s_uuid, uuid)) {
+			ofs->fs[i].bad_uuid = true;
 			return false;
+		}
 	}
 	return true;
 }
@@ -1277,54 +1355,80 @@ static int ovl_get_fsid(struct ovl_fs *ofs, const struct path *path)
 	unsigned int i;
 	dev_t dev;
 	int err;
+	bool bad_uuid = false;
 
-	/* fsid 0 is reserved for upper fs even with non upper overlay */
-	if (ofs->upper_mnt && ofs->upper_mnt->mnt_sb == sb)
-		return 0;
-
-	for (i = 0; i < ofs->numlowerfs; i++) {
-		if (ofs->lower_fs[i].sb == sb)
-			return i + 1;
+	for (i = 0; i < ofs->numfs; i++) {
+		if (ofs->fs[i].sb == sb)
+			return i;
 	}
 
 	if (!ovl_lower_uuid_ok(ofs, &sb->s_uuid)) {
-		ofs->config.index = false;
-		ofs->config.nfs_export = false;
-		pr_warn("overlayfs: %s uuid detected in lower fs '%pd2', falling back to index=off,nfs_export=off.\n",
-			uuid_is_null(&sb->s_uuid) ? "null" : "conflicting",
-			path->dentry);
+		bad_uuid = true;
+		if (ofs->config.index || ofs->config.nfs_export) {
+			ofs->config.index = false;
+			ofs->config.nfs_export = false;
+			pr_warn("%s uuid detected in lower fs '%pd2', falling back to index=off,nfs_export=off.\n",
+				uuid_is_null(&sb->s_uuid) ? "null" :
+							    "conflicting",
+				path->dentry);
+		}
 	}
 
 	err = get_anon_bdev(&dev);
 	if (err) {
-		pr_err("overlayfs: failed to get anonymous bdev for lowerpath\n");
+		pr_err("failed to get anonymous bdev for lowerpath\n");
 		return err;
 	}
 
-	ofs->lower_fs[ofs->numlowerfs].sb = sb;
-	ofs->lower_fs[ofs->numlowerfs].pseudo_dev = dev;
-	ofs->numlowerfs++;
+	ofs->fs[ofs->numfs].sb = sb;
+	ofs->fs[ofs->numfs].pseudo_dev = dev;
+	ofs->fs[ofs->numfs].bad_uuid = bad_uuid;
 
-	return ofs->numlowerfs;
+	return ofs->numfs++;
 }
 
-static int ovl_get_lower_layers(struct super_block *sb, struct ovl_fs *ofs,
-				struct path *stack, unsigned int numlower)
+static int ovl_get_layers(struct super_block *sb, struct ovl_fs *ofs,
+			  struct path *stack, unsigned int numlower)
 {
 	int err;
 	unsigned int i;
+	struct ovl_layer *layers;
 
 	err = -ENOMEM;
-	ofs->lower_layers = kcalloc(numlower, sizeof(struct ovl_layer),
-				    GFP_KERNEL);
-	if (ofs->lower_layers == NULL)
+	layers = kcalloc(numlower + 1, sizeof(struct ovl_layer), GFP_KERNEL);
+	if (!layers)
 		goto out;
+	ofs->layers = layers;
 
-	ofs->lower_fs = kcalloc(numlower, sizeof(struct ovl_sb),
-				GFP_KERNEL);
-	if (ofs->lower_fs == NULL)
+	ofs->fs = kcalloc(numlower + 1, sizeof(struct ovl_sb), GFP_KERNEL);
+	if (ofs->fs == NULL)
 		goto out;
 
+	/* idx/fsid 0 are reserved for upper fs even with lower only overlay */
+	ofs->numfs++;
+
+	layers[0].mnt = ofs->upper_mnt;
+	layers[0].idx = 0;
+	layers[0].fsid = 0;
+	ofs->numlayer = 1;
+
+	/*
+	 * All lower layers that share the same fs as upper layer, use the same
+	 * pseudo_dev as upper layer.  Allocate fs[0].pseudo_dev even for lower
+	 * only overlay to simplify ovl_fs_free().
+	 * is_lower will be set if upper fs is shared with a lower layer.
+	 */
+	err = get_anon_bdev(&ofs->fs[0].pseudo_dev);
+	if (err) {
+		pr_err("failed to get anonymous bdev for upper fs\n");
+		goto out;
+	}
+
+	if (ofs->upper_mnt) {
+		ofs->fs[0].sb = ofs->upper_mnt->mnt_sb;
+		ofs->fs[0].is_lower = false;
+	}
+
 	for (i = 0; i < numlower; i++) {
 		struct vfsmount *mnt;
 		struct inode *trap;
@@ -1347,7 +1451,7 @@ static int ovl_get_lower_layers(struct super_block *sb, struct ovl_fs *ofs,
 		mnt = clone_private_mount(&stack[i]);
 		err = PTR_ERR(mnt);
 		if (IS_ERR(mnt)) {
-			pr_err("overlayfs: failed to clone lowerpath\n");
+			pr_err("failed to clone lowerpath\n");
 			iput(trap);
 			goto out;
 		}
@@ -1358,41 +1462,45 @@ static int ovl_get_lower_layers(struct super_block *sb, struct ovl_fs *ofs,
 		 */
 		mnt->mnt_flags |= MNT_READONLY | MNT_NOATIME;
 
-		ofs->lower_layers[ofs->numlower].trap = trap;
-		ofs->lower_layers[ofs->numlower].mnt = mnt;
-		ofs->lower_layers[ofs->numlower].idx = i + 1;
-		ofs->lower_layers[ofs->numlower].fsid = fsid;
-		if (fsid) {
-			ofs->lower_layers[ofs->numlower].fs =
-				&ofs->lower_fs[fsid - 1];
-		}
-		ofs->numlower++;
+		layers[ofs->numlayer].trap = trap;
+		layers[ofs->numlayer].mnt = mnt;
+		layers[ofs->numlayer].idx = ofs->numlayer;
+		layers[ofs->numlayer].fsid = fsid;
+		layers[ofs->numlayer].fs = &ofs->fs[fsid];
+		ofs->numlayer++;
+		ofs->fs[fsid].is_lower = true;
 	}
 
 	/*
 	 * When all layers on same fs, overlay can use real inode numbers.
-	 * With mount option "xino=on", mounter declares that there are enough
-	 * free high bits in underlying fs to hold the unique fsid.
+	 * With mount option "xino=<on|auto>", mounter declares that there are
+	 * enough free high bits in underlying fs to hold the unique fsid.
 	 * If overlayfs does encounter underlying inodes using the high xino
 	 * bits reserved for fsid, it emits a warning and uses the original
-	 * inode number.
+	 * inode number or a non persistent inode number allocated from a
+	 * dedicated range.
 	 */
-	if (!ofs->numlowerfs || (ofs->numlowerfs == 1 && !ofs->upper_mnt)) {
-		ofs->xino_bits = 0;
-		ofs->config.xino = OVL_XINO_OFF;
-	} else if (ofs->config.xino == OVL_XINO_ON && !ofs->xino_bits) {
+	if (ofs->numfs - !ofs->upper_mnt == 1) {
+		if (ofs->config.xino == OVL_XINO_ON)
+			pr_info("\"xino=on\" is useless with all layers on same fs, ignore.\n");
+		ofs->xino_mode = 0;
+	} else if (ofs->config.xino == OVL_XINO_OFF) {
+		ofs->xino_mode = -1;
+	} else if (ofs->xino_mode < 0) {
 		/*
-		 * This is a roundup of number of bits needed for numlowerfs+1
-		 * (i.e. ilog2(numlowerfs+1 - 1) + 1). fsid 0 is reserved for
-		 * upper fs even with non upper overlay.
+		 * This is a roundup of number of bits needed for encoding
+		 * fsid, where fsid 0 is reserved for upper fs (even with
+		 * lower only overlay) +1 extra bit is reserved for the non
+		 * persistent inode number range that is used for resolving
+		 * xino lower bits overflow.
 		 */
-		BUILD_BUG_ON(ilog2(OVL_MAX_STACK) > 31);
-		ofs->xino_bits = ilog2(ofs->numlowerfs) + 1;
+		BUILD_BUG_ON(ilog2(OVL_MAX_STACK) > 30);
+		ofs->xino_mode = ilog2(ofs->numfs - 1) + 2;
 	}
 
-	if (ofs->xino_bits) {
-		pr_info("overlayfs: \"xino\" feature enabled using %d upper inode bits.\n",
-			ofs->xino_bits);
+	if (ofs->xino_mode > 0) {
+		pr_info("\"xino\" feature enabled using %d upper inode bits.\n",
+			ofs->xino_mode);
 	}
 
 	err = 0;
@@ -1407,7 +1515,6 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 	char *lowertmp, *lower;
 	struct path *stack = NULL;
 	unsigned int stacklen, numlower = 0, i;
-	bool remote = false;
 	struct ovl_entry *oe;
 
 	err = -ENOMEM;
@@ -1418,15 +1525,15 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 	err = -EINVAL;
 	stacklen = ovl_split_lowerdirs(lowertmp);
 	if (stacklen > OVL_MAX_STACK) {
-		pr_err("overlayfs: too many lower directories, limit is %d\n",
+		pr_err("too many lower directories, limit is %d\n",
 		       OVL_MAX_STACK);
 		goto out_err;
 	} else if (!ofs->config.upperdir && stacklen == 1) {
-		pr_err("overlayfs: at least 2 lowerdir are needed while upperdir nonexistent\n");
+		pr_err("at least 2 lowerdir are needed while upperdir nonexistent\n");
 		goto out_err;
 	} else if (!ofs->config.upperdir && ofs->config.nfs_export &&
 		   ofs->config.redirect_follow) {
-		pr_warn("overlayfs: NFS export requires \"redirect_dir=nofollow\" on non-upper mount, falling back to nfs_export=off.\n");
+		pr_warn("NFS export requires \"redirect_dir=nofollow\" on non-upper mount, falling back to nfs_export=off.\n");
 		ofs->config.nfs_export = false;
 	}
 
@@ -1439,7 +1546,7 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 	lower = lowertmp;
 	for (numlower = 0; numlower < stacklen; numlower++) {
 		err = ovl_lower_dir(lower, &stack[numlower], ofs,
-				    &sb->s_stack_depth, &remote);
+				    &sb->s_stack_depth);
 		if (err)
 			goto out_err;
 
@@ -1449,11 +1556,11 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 	err = -EINVAL;
 	sb->s_stack_depth++;
 	if (sb->s_stack_depth > FILESYSTEM_MAX_STACK_DEPTH) {
-		pr_err("overlayfs: maximum fs stacking depth exceeded\n");
+		pr_err("maximum fs stacking depth exceeded\n");
 		goto out_err;
 	}
 
-	err = ovl_get_lower_layers(sb, ofs, stack, numlower);
+	err = ovl_get_layers(sb, ofs, stack, numlower);
 	if (err)
 		goto out_err;
 
@@ -1464,14 +1571,9 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 
 	for (i = 0; i < numlower; i++) {
 		oe->lowerstack[i].dentry = dget(stack[i].dentry);
-		oe->lowerstack[i].layer = &ofs->lower_layers[i];
+		oe->lowerstack[i].layer = &ofs->layers[i+1];
 	}
 
-	if (remote)
-		sb->s_d_op = &ovl_reval_dentry_operations;
-	else
-		sb->s_d_op = &ovl_dentry_operations;
-
 out:
 	for (i = 0; i < numlower; i++)
 		path_put(&stack[i]);
@@ -1505,7 +1607,7 @@ static int ovl_check_layer(struct super_block *sb, struct ovl_fs *ofs,
 	while (!err && parent != next) {
 		if (ovl_lookup_trap_inode(sb, parent)) {
 			err = -ELOOP;
-			pr_err("overlayfs: overlapping %s path\n", name);
+			pr_err("overlapping %s path\n", name);
 		} else if (ovl_is_inuse(parent)) {
 			err = ovl_report_in_use(ofs, name);
 		}
@@ -1545,9 +1647,9 @@ static int ovl_check_overlapping_layers(struct super_block *sb,
 			return err;
 	}
 
-	for (i = 0; i < ofs->numlower; i++) {
+	for (i = 1; i < ofs->numlayer; i++) {
 		err = ovl_check_layer(sb, ofs,
-				      ofs->lower_layers[i].mnt->mnt_root,
+				      ofs->layers[i].mnt->mnt_root,
 				      "lowerdir");
 		if (err)
 			return err;
@@ -1556,6 +1658,44 @@ static int ovl_check_overlapping_layers(struct super_block *sb,
 	return 0;
 }
 
+static struct dentry *ovl_get_root(struct super_block *sb,
+				   struct dentry *upperdentry,
+				   struct ovl_entry *oe)
+{
+	struct dentry *root;
+	struct ovl_path *lowerpath = &oe->lowerstack[0];
+	unsigned long ino = d_inode(lowerpath->dentry)->i_ino;
+	int fsid = lowerpath->layer->fsid;
+	struct ovl_inode_params oip = {
+		.upperdentry = upperdentry,
+		.lowerpath = lowerpath,
+	};
+
+	root = d_make_root(ovl_new_inode(sb, S_IFDIR, 0));
+	if (!root)
+		return NULL;
+
+	root->d_fsdata = oe;
+
+	if (upperdentry) {
+		/* Root inode uses upper st_ino/i_ino */
+		ino = d_inode(upperdentry)->i_ino;
+		fsid = 0;
+		ovl_dentry_set_upper_alias(root);
+		if (ovl_is_impuredir(upperdentry))
+			ovl_set_flag(OVL_IMPURE, d_inode(root));
+	}
+
+	/* Root is always merge -> can have whiteouts */
+	ovl_set_flag(OVL_WHITEOUTS, d_inode(root));
+	ovl_dentry_set_flag(OVL_E_CONNECTED, root);
+	ovl_set_upperdata(d_inode(root));
+	ovl_inode_init(d_inode(root), &oip, ino, fsid);
+	ovl_dentry_update_reval(root, upperdentry, DCACHE_OP_WEAK_REVALIDATE);
+
+	return root;
+}
+
 static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct path upperpath = { };
@@ -1565,6 +1705,8 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 	struct cred *cred;
 	int err;
 
+	sb->s_d_op = &ovl_dentry_operations;
+
 	err = -ENOMEM;
 	ofs = kzalloc(sizeof(struct ovl_fs), GFP_KERNEL);
 	if (!ofs)
@@ -1585,22 +1727,28 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 	err = -EINVAL;
 	if (!ofs->config.lowerdir) {
 		if (!silent)
-			pr_err("overlayfs: missing 'lowerdir'\n");
+			pr_err("missing 'lowerdir'\n");
 		goto out_err;
 	}
 
 	sb->s_stack_depth = 0;
 	sb->s_maxbytes = MAX_LFS_FILESIZE;
+	atomic_long_set(&ofs->last_ino, 1);
 	/* Assume underlaying fs uses 32bit inodes unless proven otherwise */
-	if (ofs->config.xino != OVL_XINO_OFF)
-		ofs->xino_bits = BITS_PER_LONG - 32;
+	if (ofs->config.xino != OVL_XINO_OFF) {
+		ofs->xino_mode = BITS_PER_LONG - 32;
+		if (!ofs->xino_mode) {
+			pr_warn("xino not supported on 32bit kernel, falling back to xino=off.\n");
+			ofs->config.xino = OVL_XINO_OFF;
+		}
+	}
 
 	/* alloc/destroy_inode needed for setting up traps in inode cache */
 	sb->s_op = &ovl_super_operations;
 
 	if (ofs->config.upperdir) {
 		if (!ofs->config.workdir) {
-			pr_err("overlayfs: missing 'workdir'\n");
+			pr_err("missing 'workdir'\n");
 			goto out_err;
 		}
 
@@ -1650,13 +1798,13 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 	if (!ofs->indexdir) {
 		ofs->config.index = false;
 		if (ofs->upper_mnt && ofs->config.nfs_export) {
-			pr_warn("overlayfs: NFS export requires an index dir, falling back to nfs_export=off.\n");
+			pr_warn("NFS export requires an index dir, falling back to nfs_export=off.\n");
 			ofs->config.nfs_export = false;
 		}
 	}
 
 	if (ofs->config.metacopy && ofs->config.nfs_export) {
-		pr_warn("overlayfs: NFS export is not supported with metadata only copy up, falling back to nfs_export=off.\n");
+		pr_warn("NFS export is not supported with metadata only copy up, falling back to nfs_export=off.\n");
 		ofs->config.nfs_export = false;
 	}
 
@@ -1672,25 +1820,11 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_flags |= SB_POSIXACL;
 
 	err = -ENOMEM;
-	root_dentry = d_make_root(ovl_new_inode(sb, S_IFDIR, 0));
+	root_dentry = ovl_get_root(sb, upperpath.dentry, oe);
 	if (!root_dentry)
 		goto out_free_oe;
 
-	root_dentry->d_fsdata = oe;
-
 	mntput(upperpath.mnt);
-	if (upperpath.dentry) {
-		ovl_dentry_set_upper_alias(root_dentry);
-		if (ovl_is_impuredir(upperpath.dentry))
-			ovl_set_flag(OVL_IMPURE, d_inode(root_dentry));
-	}
-
-	/* Root is always merge -> can have whiteouts */
-	ovl_set_flag(OVL_WHITEOUTS, d_inode(root_dentry));
-	ovl_dentry_set_flag(OVL_E_CONNECTED, root_dentry);
-	ovl_set_upperdata(d_inode(root_dentry));
-	ovl_inode_init(d_inode(root_dentry), upperpath.dentry,
-		       ovl_dentry_lower(root_dentry), NULL);
 
 	sb->s_root = root_dentry;
 
@@ -1739,9 +1873,15 @@ static int __init ovl_init(void)
 	if (ovl_inode_cachep == NULL)
 		return -ENOMEM;
 
-	err = register_filesystem(&ovl_fs_type);
-	if (err)
-		kmem_cache_destroy(ovl_inode_cachep);
+	err = ovl_aio_request_cache_init();
+	if (!err) {
+		err = register_filesystem(&ovl_fs_type);
+		if (!err)
+			return 0;
+
+		ovl_aio_request_cache_destroy();
+	}
+	kmem_cache_destroy(ovl_inode_cachep);
 
 	return err;
 }
@@ -1756,7 +1896,7 @@ static void __exit ovl_exit(void)
 	 */
 	rcu_barrier();
 	kmem_cache_destroy(ovl_inode_cachep);
-
+	ovl_aio_request_cache_destroy();
 }
 
 module_init(ovl_init);
diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c
index f5678a3f8350..36b60788ee47 100644
--- a/fs/overlayfs/util.c
+++ b/fs/overlayfs/util.c
@@ -40,18 +40,6 @@ const struct cred *ovl_override_creds(struct super_block *sb)
 	return override_creds(ofs->creator_cred);
 }
 
-struct super_block *ovl_same_sb(struct super_block *sb)
-{
-	struct ovl_fs *ofs = sb->s_fs_info;
-
-	if (!ofs->numlowerfs)
-		return ofs->upper_mnt->mnt_sb;
-	else if (ofs->numlowerfs == 1 && !ofs->upper_mnt)
-		return ofs->lower_fs[0].sb;
-	else
-		return NULL;
-}
-
 /*
  * Check if underlying fs supports file handles and try to determine encoding
  * type, in order to deduce maximum inode number used by fs.
@@ -105,8 +93,24 @@ struct ovl_entry *ovl_alloc_entry(unsigned int numlower)
 bool ovl_dentry_remote(struct dentry *dentry)
 {
 	return dentry->d_flags &
-		(DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE |
-		 DCACHE_OP_REAL);
+		(DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
+}
+
+void ovl_dentry_update_reval(struct dentry *dentry, struct dentry *upperdentry,
+			     unsigned int mask)
+{
+	struct ovl_entry *oe = OVL_E(dentry);
+	unsigned int i, flags = 0;
+
+	if (upperdentry)
+		flags |= upperdentry->d_flags;
+	for (i = 0; i < oe->numlower; i++)
+		flags |= oe->lowerstack[i].dentry->d_flags;
+
+	spin_lock(&dentry->d_lock);
+	dentry->d_flags &= ~mask;
+	dentry->d_flags |= flags & mask;
+	spin_unlock(&dentry->d_lock);
 }
 
 bool ovl_dentry_weird(struct dentry *dentry)
@@ -198,7 +202,7 @@ struct dentry *ovl_dentry_lower(struct dentry *dentry)
 	return oe->numlower ? oe->lowerstack[0].dentry : NULL;
 }
 
-struct ovl_layer *ovl_layer_lower(struct dentry *dentry)
+const struct ovl_layer *ovl_layer_lower(struct dentry *dentry)
 {
 	struct ovl_entry *oe = dentry->d_fsdata;
 
@@ -398,24 +402,6 @@ void ovl_dentry_set_redirect(struct dentry *dentry, const char *redirect)
 	oi->redirect = redirect;
 }
 
-void ovl_inode_init(struct inode *inode, struct dentry *upperdentry,
-		    struct dentry *lowerdentry, struct dentry *lowerdata)
-{
-	struct inode *realinode = d_inode(upperdentry ?: lowerdentry);
-
-	if (upperdentry)
-		OVL_I(inode)->__upperdentry = upperdentry;
-	if (lowerdentry)
-		OVL_I(inode)->lower = igrab(d_inode(lowerdentry));
-	if (lowerdata)
-		OVL_I(inode)->lowerdata = igrab(d_inode(lowerdata));
-
-	ovl_copyattr(realinode, inode);
-	ovl_copyflags(realinode, inode);
-	if (!inode->i_ino)
-		inode->i_ino = realinode->i_ino;
-}
-
 void ovl_inode_update(struct inode *inode, struct dentry *upperdentry)
 {
 	struct inode *upperinode = d_inode(upperdentry);
@@ -428,8 +414,6 @@ void ovl_inode_update(struct inode *inode, struct dentry *upperdentry)
 	smp_wmb();
 	OVL_I(inode)->__upperdentry = upperdentry;
 	if (inode_unhashed(inode)) {
-		if (!inode->i_ino)
-			inode->i_ino = upperinode->i_ino;
 		inode->i_private = upperinode;
 		__insert_inode_hash(inode, (unsigned long) upperinode);
 	}
@@ -521,7 +505,7 @@ int ovl_copy_up_start(struct dentry *dentry, int flags)
 	struct inode *inode = d_inode(dentry);
 	int err;
 
-	err = ovl_inode_lock(inode);
+	err = ovl_inode_lock_interruptible(inode);
 	if (!err && ovl_already_copied_up_locked(dentry, flags)) {
 		err = 1; /* Already copied up */
 		ovl_inode_unlock(inode);
@@ -576,7 +560,7 @@ int ovl_check_setxattr(struct dentry *dentry, struct dentry *upperdentry,
 	err = ovl_do_setxattr(upperdentry, name, value, size, 0);
 
 	if (err == -EOPNOTSUPP) {
-		pr_warn("overlayfs: cannot set %s xattr on upper\n", name);
+		pr_warn("cannot set %s xattr on upper\n", name);
 		ofs->noxattr = true;
 		return xerr;
 	}
@@ -700,7 +684,7 @@ static void ovl_cleanup_index(struct dentry *dentry)
 
 	inode = d_inode(upperdentry);
 	if (!S_ISDIR(inode->i_mode) && inode->i_nlink != 1) {
-		pr_warn_ratelimited("overlayfs: cleanup linked index (%pd2, ino=%lu, nlink=%u)\n",
+		pr_warn_ratelimited("cleanup linked index (%pd2, ino=%lu, nlink=%u)\n",
 				    upperdentry, inode->i_ino, inode->i_nlink);
 		/*
 		 * We either have a bug with persistent union nlink or a lower
@@ -739,7 +723,7 @@ out:
 	return;
 
 fail:
-	pr_err("overlayfs: cleanup index of '%pd2' failed (%i)\n", dentry, err);
+	pr_err("cleanup index of '%pd2' failed (%i)\n", dentry, err);
 	goto out;
 }
 
@@ -776,7 +760,7 @@ int ovl_nlink_start(struct dentry *dentry)
 			return err;
 	}
 
-	err = ovl_inode_lock(inode);
+	err = ovl_inode_lock_interruptible(inode);
 	if (err)
 		return err;
 
@@ -830,7 +814,7 @@ int ovl_lock_rename_workdir(struct dentry *workdir, struct dentry *upperdir)
 err_unlock:
 	unlock_rename(workdir, upperdir);
 err:
-	pr_err("overlayfs: failed to lock workdir+upperdir\n");
+	pr_err("failed to lock workdir+upperdir\n");
 	return -EIO;
 }
 
@@ -852,7 +836,7 @@ int ovl_check_metacopy_xattr(struct dentry *dentry)
 
 	return 1;
 out:
-	pr_warn_ratelimited("overlayfs: failed to get metacopy (%i)\n", res);
+	pr_warn_ratelimited("failed to get metacopy (%i)\n", res);
 	return res;
 }
 
@@ -899,7 +883,7 @@ ssize_t ovl_getxattr(struct dentry *dentry, char *name, char **value,
 	return res;
 
 fail:
-	pr_warn_ratelimited("overlayfs: failed to get xattr %s: err=%zi)\n",
+	pr_warn_ratelimited("failed to get xattr %s: err=%zi)\n",
 			    name, res);
 	kfree(buf);
 	return res;
@@ -931,7 +915,7 @@ char *ovl_get_redirect_xattr(struct dentry *dentry, int padding)
 
 	return buf;
 invalid:
-	pr_warn_ratelimited("overlayfs: invalid redirect (%s)\n", buf);
+	pr_warn_ratelimited("invalid redirect (%s)\n", buf);
 	res = -EINVAL;
 	kfree(buf);
 	return ERR_PTR(res);
diff --git a/fs/pipe.c b/fs/pipe.c
index 8a2ab2f974bd..16fb72e9abf7 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -43,10 +43,12 @@ unsigned long pipe_user_pages_hard;
 unsigned long pipe_user_pages_soft = PIPE_DEF_BUFFERS * INR_OPEN_CUR;
 
 /*
- * We use a start+len construction, which provides full use of the 
- * allocated memory.
- * -- Florian Coosmann (FGC)
- * 
+ * We use head and tail indices that aren't masked off, except at the point of
+ * dereference, but rather they're allowed to wrap naturally.  This means there
+ * isn't a dead spot in the buffer, but the ring has to be a power of two and
+ * <= 2^31.
+ * -- David Howells 2019-09-23.
+ *
  * Reads with count = 0 should always return 0.
  * -- Julian Bradfield 1999-06-07.
  *
@@ -106,16 +108,19 @@ void pipe_double_lock(struct pipe_inode_info *pipe1,
 /* Drop the inode semaphore and wait for a pipe event, atomically */
 void pipe_wait(struct pipe_inode_info *pipe)
 {
-	DEFINE_WAIT(wait);
+	DEFINE_WAIT(rdwait);
+	DEFINE_WAIT(wrwait);
 
 	/*
 	 * Pipes are system-local resources, so sleeping on them
 	 * is considered a noninteractive wait:
 	 */
-	prepare_to_wait(&pipe->wait, &wait, TASK_INTERRUPTIBLE);
+	prepare_to_wait(&pipe->rd_wait, &rdwait, TASK_INTERRUPTIBLE);
+	prepare_to_wait(&pipe->wr_wait, &wrwait, TASK_INTERRUPTIBLE);
 	pipe_unlock(pipe);
 	schedule();
-	finish_wait(&pipe->wait, &wait);
+	finish_wait(&pipe->rd_wait, &rdwait);
+	finish_wait(&pipe->wr_wait, &wrwait);
 	pipe_lock(pipe);
 }
 
@@ -141,7 +146,7 @@ static int anon_pipe_buf_steal(struct pipe_inode_info *pipe,
 	struct page *page = buf->page;
 
 	if (page_count(page) == 1) {
-		memcg_kmem_uncharge(page, 0);
+		memcg_kmem_uncharge_page(page, 0);
 		__SetPageLocked(page);
 		return 0;
 	}
@@ -268,27 +273,48 @@ static bool pipe_buf_can_merge(struct pipe_buffer *buf)
 	return buf->ops == &anon_pipe_buf_ops;
 }
 
+/* Done while waiting without holding the pipe lock - thus the READ_ONCE() */
+static inline bool pipe_readable(const struct pipe_inode_info *pipe)
+{
+	unsigned int head = READ_ONCE(pipe->head);
+	unsigned int tail = READ_ONCE(pipe->tail);
+	unsigned int writers = READ_ONCE(pipe->writers);
+
+	return !pipe_empty(head, tail) || !writers;
+}
+
 static ssize_t
 pipe_read(struct kiocb *iocb, struct iov_iter *to)
 {
 	size_t total_len = iov_iter_count(to);
 	struct file *filp = iocb->ki_filp;
 	struct pipe_inode_info *pipe = filp->private_data;
-	int do_wakeup;
+	bool was_full, wake_next_reader = false;
 	ssize_t ret;
 
 	/* Null read succeeds. */
 	if (unlikely(total_len == 0))
 		return 0;
 
-	do_wakeup = 0;
 	ret = 0;
 	__pipe_lock(pipe);
+
+	/*
+	 * We only wake up writers if the pipe was full when we started
+	 * reading in order to avoid unnecessary wakeups.
+	 *
+	 * But when we do wake up writers, we do so using a sync wakeup
+	 * (WF_SYNC), because we want them to get going and generate more
+	 * data for us.
+	 */
+	was_full = pipe_full(pipe->head, pipe->tail, pipe->max_usage);
 	for (;;) {
-		int bufs = pipe->nrbufs;
-		if (bufs) {
-			int curbuf = pipe->curbuf;
-			struct pipe_buffer *buf = pipe->bufs + curbuf;
+		unsigned int head = pipe->head;
+		unsigned int tail = pipe->tail;
+		unsigned int mask = pipe->ring_size - 1;
+
+		if (!pipe_empty(head, tail)) {
+			struct pipe_buffer *buf = &pipe->bufs[tail & mask];
 			size_t chars = buf->len;
 			size_t written;
 			int error;
@@ -321,50 +347,73 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to)
 
 			if (!buf->len) {
 				pipe_buf_release(pipe, buf);
-				curbuf = (curbuf + 1) & (pipe->buffers - 1);
-				pipe->curbuf = curbuf;
-				pipe->nrbufs = --bufs;
-				do_wakeup = 1;
+				spin_lock_irq(&pipe->rd_wait.lock);
+				tail++;
+				pipe->tail = tail;
+				spin_unlock_irq(&pipe->rd_wait.lock);
 			}
 			total_len -= chars;
 			if (!total_len)
 				break;	/* common path: read succeeded */
+			if (!pipe_empty(head, tail))	/* More to do? */
+				continue;
 		}
-		if (bufs)	/* More to do? */
-			continue;
+
 		if (!pipe->writers)
 			break;
-		if (!pipe->waiting_writers) {
-			/* syscall merging: Usually we must not sleep
-			 * if O_NONBLOCK is set, or if we got some data.
-			 * But if a writer sleeps in kernel space, then
-			 * we can wait for that data without violating POSIX.
-			 */
-			if (ret)
-				break;
-			if (filp->f_flags & O_NONBLOCK) {
-				ret = -EAGAIN;
-				break;
-			}
-		}
-		if (signal_pending(current)) {
-			if (!ret)
-				ret = -ERESTARTSYS;
+		if (ret)
+			break;
+		if (filp->f_flags & O_NONBLOCK) {
+			ret = -EAGAIN;
 			break;
 		}
-		if (do_wakeup) {
-			wake_up_interruptible_sync_poll(&pipe->wait, EPOLLOUT | EPOLLWRNORM);
- 			kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT);
+		__pipe_unlock(pipe);
+
+		/*
+		 * We only get here if we didn't actually read anything.
+		 *
+		 * However, we could have seen (and removed) a zero-sized
+		 * pipe buffer, and might have made space in the buffers
+		 * that way.
+		 *
+		 * You can't make zero-sized pipe buffers by doing an empty
+		 * write (not even in packet mode), but they can happen if
+		 * the writer gets an EFAULT when trying to fill a buffer
+		 * that already got allocated and inserted in the buffer
+		 * array.
+		 *
+		 * So we still need to wake up any pending writers in the
+		 * _very_ unlikely case that the pipe was full, but we got
+		 * no data.
+		 */
+		if (unlikely(was_full)) {
+			wake_up_interruptible_sync_poll(&pipe->wr_wait, EPOLLOUT | EPOLLWRNORM);
+			kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT);
 		}
-		pipe_wait(pipe);
+
+		/*
+		 * But because we didn't read anything, at this point we can
+		 * just return directly with -ERESTARTSYS if we're interrupted,
+		 * since we've done any required wakeups and there's no need
+		 * to mark anything accessed. And we've dropped the lock.
+		 */
+		if (wait_event_interruptible_exclusive(pipe->rd_wait, pipe_readable(pipe)) < 0)
+			return -ERESTARTSYS;
+
+		__pipe_lock(pipe);
+		was_full = pipe_full(pipe->head, pipe->tail, pipe->max_usage);
+		wake_next_reader = true;
 	}
+	if (pipe_empty(pipe->head, pipe->tail))
+		wake_next_reader = false;
 	__pipe_unlock(pipe);
 
-	/* Signal writers asynchronously that there is more room. */
-	if (do_wakeup) {
-		wake_up_interruptible_sync_poll(&pipe->wait, EPOLLOUT | EPOLLWRNORM);
+	if (was_full) {
+		wake_up_interruptible_sync_poll(&pipe->wr_wait, EPOLLOUT | EPOLLWRNORM);
 		kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT);
 	}
+	if (wake_next_reader)
+		wake_up_interruptible_sync_poll(&pipe->rd_wait, EPOLLIN | EPOLLRDNORM);
 	if (ret > 0)
 		file_accessed(filp);
 	return ret;
@@ -375,15 +424,28 @@ static inline int is_packetized(struct file *file)
 	return (file->f_flags & O_DIRECT) != 0;
 }
 
+/* Done while waiting without holding the pipe lock - thus the READ_ONCE() */
+static inline bool pipe_writable(const struct pipe_inode_info *pipe)
+{
+	unsigned int head = READ_ONCE(pipe->head);
+	unsigned int tail = READ_ONCE(pipe->tail);
+	unsigned int max_usage = READ_ONCE(pipe->max_usage);
+
+	return !pipe_full(head, tail, max_usage) ||
+		!READ_ONCE(pipe->readers);
+}
+
 static ssize_t
 pipe_write(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *filp = iocb->ki_filp;
 	struct pipe_inode_info *pipe = filp->private_data;
+	unsigned int head;
 	ssize_t ret = 0;
-	int do_wakeup = 0;
 	size_t total_len = iov_iter_count(from);
 	ssize_t chars;
+	bool was_empty = false;
+	bool wake_next_writer = false;
 
 	/* Null write succeeds. */
 	if (unlikely(total_len == 0))
@@ -397,12 +459,23 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from)
 		goto out;
 	}
 
-	/* We try to merge small writes */
-	chars = total_len & (PAGE_SIZE-1); /* size of the last buffer */
-	if (pipe->nrbufs && chars != 0) {
-		int lastbuf = (pipe->curbuf + pipe->nrbufs - 1) &
-							(pipe->buffers - 1);
-		struct pipe_buffer *buf = pipe->bufs + lastbuf;
+	/*
+	 * Only wake up if the pipe started out empty, since
+	 * otherwise there should be no readers waiting.
+	 *
+	 * If it wasn't empty we try to merge new data into
+	 * the last buffer.
+	 *
+	 * That naturally merges small writes, but it also
+	 * page-aligs the rest of the writes for large writes
+	 * spanning multiple pages.
+	 */
+	head = pipe->head;
+	was_empty = pipe_empty(head, pipe->tail);
+	chars = total_len & (PAGE_SIZE-1);
+	if (chars && !was_empty) {
+		unsigned int mask = pipe->ring_size - 1;
+		struct pipe_buffer *buf = &pipe->bufs[(head - 1) & mask];
 		int offset = buf->offset + buf->len;
 
 		if (pipe_buf_can_merge(buf) && offset + chars <= PAGE_SIZE) {
@@ -415,7 +488,7 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from)
 				ret = -EFAULT;
 				goto out;
 			}
-			do_wakeup = 1;
+
 			buf->len += ret;
 			if (!iov_iter_count(from))
 				goto out;
@@ -423,18 +496,17 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from)
 	}
 
 	for (;;) {
-		int bufs;
-
 		if (!pipe->readers) {
 			send_sig(SIGPIPE, current, 0);
 			if (!ret)
 				ret = -EPIPE;
 			break;
 		}
-		bufs = pipe->nrbufs;
-		if (bufs < pipe->buffers) {
-			int newbuf = (pipe->curbuf + bufs) & (pipe->buffers-1);
-			struct pipe_buffer *buf = pipe->bufs + newbuf;
+
+		head = pipe->head;
+		if (!pipe_full(head, pipe->tail, pipe->max_usage)) {
+			unsigned int mask = pipe->ring_size - 1;
+			struct pipe_buffer *buf = &pipe->bufs[head & mask];
 			struct page *page = pipe->tmp_page;
 			int copied;
 
@@ -446,38 +518,54 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from)
 				}
 				pipe->tmp_page = page;
 			}
-			/* Always wake up, even if the copy fails. Otherwise
-			 * we lock up (O_NONBLOCK-)readers that sleep due to
-			 * syscall merging.
-			 * FIXME! Is this really true?
+
+			/* Allocate a slot in the ring in advance and attach an
+			 * empty buffer.  If we fault or otherwise fail to use
+			 * it, either the reader will consume it or it'll still
+			 * be there for the next write.
 			 */
-			do_wakeup = 1;
-			copied = copy_page_from_iter(page, 0, PAGE_SIZE, from);
-			if (unlikely(copied < PAGE_SIZE && iov_iter_count(from))) {
-				if (!ret)
-					ret = -EFAULT;
-				break;
+			spin_lock_irq(&pipe->rd_wait.lock);
+
+			head = pipe->head;
+			if (pipe_full(head, pipe->tail, pipe->max_usage)) {
+				spin_unlock_irq(&pipe->rd_wait.lock);
+				continue;
 			}
-			ret += copied;
+
+			pipe->head = head + 1;
+			spin_unlock_irq(&pipe->rd_wait.lock);
 
 			/* Insert it into the buffer array */
+			buf = &pipe->bufs[head & mask];
 			buf->page = page;
 			buf->ops = &anon_pipe_buf_ops;
 			buf->offset = 0;
-			buf->len = copied;
+			buf->len = 0;
 			buf->flags = 0;
 			if (is_packetized(filp)) {
 				buf->ops = &packet_pipe_buf_ops;
 				buf->flags = PIPE_BUF_FLAG_PACKET;
 			}
-			pipe->nrbufs = ++bufs;
 			pipe->tmp_page = NULL;
 
+			copied = copy_page_from_iter(page, 0, PAGE_SIZE, from);
+			if (unlikely(copied < PAGE_SIZE && iov_iter_count(from))) {
+				if (!ret)
+					ret = -EFAULT;
+				break;
+			}
+			ret += copied;
+			buf->offset = 0;
+			buf->len = copied;
+
 			if (!iov_iter_count(from))
 				break;
 		}
-		if (bufs < pipe->buffers)
+
+		if (!pipe_full(head, pipe->tail, pipe->max_usage))
 			continue;
+
+		/* Wait for buffer space to become available. */
 		if (filp->f_flags & O_NONBLOCK) {
 			if (!ret)
 				ret = -EAGAIN;
@@ -488,21 +576,43 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from)
 				ret = -ERESTARTSYS;
 			break;
 		}
-		if (do_wakeup) {
-			wake_up_interruptible_sync_poll(&pipe->wait, EPOLLIN | EPOLLRDNORM);
+
+		/*
+		 * We're going to release the pipe lock and wait for more
+		 * space. We wake up any readers if necessary, and then
+		 * after waiting we need to re-check whether the pipe
+		 * become empty while we dropped the lock.
+		 */
+		__pipe_unlock(pipe);
+		if (was_empty) {
+			wake_up_interruptible_sync_poll(&pipe->rd_wait, EPOLLIN | EPOLLRDNORM);
 			kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
-			do_wakeup = 0;
 		}
-		pipe->waiting_writers++;
-		pipe_wait(pipe);
-		pipe->waiting_writers--;
+		wait_event_interruptible_exclusive(pipe->wr_wait, pipe_writable(pipe));
+		__pipe_lock(pipe);
+		was_empty = pipe_empty(pipe->head, pipe->tail);
+		wake_next_writer = true;
 	}
 out:
+	if (pipe_full(pipe->head, pipe->tail, pipe->max_usage))
+		wake_next_writer = false;
 	__pipe_unlock(pipe);
-	if (do_wakeup) {
-		wake_up_interruptible_sync_poll(&pipe->wait, EPOLLIN | EPOLLRDNORM);
+
+	/*
+	 * If we do do a wakeup event, we do a 'sync' wakeup, because we
+	 * want the reader to start processing things asap, rather than
+	 * leave the data pending.
+	 *
+	 * This is particularly important for small writes, because of
+	 * how (for example) the GNU make jobserver uses small writes to
+	 * wake up pending jobs
+	 */
+	if (was_empty) {
+		wake_up_interruptible_sync_poll(&pipe->rd_wait, EPOLLIN | EPOLLRDNORM);
 		kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
 	}
+	if (wake_next_writer)
+		wake_up_interruptible_sync_poll(&pipe->wr_wait, EPOLLOUT | EPOLLWRNORM);
 	if (ret > 0 && sb_start_write_trylock(file_inode(filp)->i_sb)) {
 		int err = file_update_time(filp);
 		if (err)
@@ -515,17 +625,19 @@ out:
 static long pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	struct pipe_inode_info *pipe = filp->private_data;
-	int count, buf, nrbufs;
+	int count, head, tail, mask;
 
 	switch (cmd) {
 		case FIONREAD:
 			__pipe_lock(pipe);
 			count = 0;
-			buf = pipe->curbuf;
-			nrbufs = pipe->nrbufs;
-			while (--nrbufs >= 0) {
-				count += pipe->bufs[buf].len;
-				buf = (buf+1) & (pipe->buffers - 1);
+			head = pipe->head;
+			tail = pipe->tail;
+			mask = pipe->ring_size - 1;
+
+			while (tail != head) {
+				count += pipe->bufs[tail & mask].len;
+				tail++;
 			}
 			__pipe_unlock(pipe);
 
@@ -541,21 +653,38 @@ pipe_poll(struct file *filp, poll_table *wait)
 {
 	__poll_t mask;
 	struct pipe_inode_info *pipe = filp->private_data;
-	int nrbufs;
+	unsigned int head, tail;
+
+	/*
+	 * Reading pipe state only -- no need for acquiring the semaphore.
+	 *
+	 * But because this is racy, the code has to add the
+	 * entry to the poll table _first_ ..
+	 */
+	if (filp->f_mode & FMODE_READ)
+		poll_wait(filp, &pipe->rd_wait, wait);
+	if (filp->f_mode & FMODE_WRITE)
+		poll_wait(filp, &pipe->wr_wait, wait);
 
-	poll_wait(filp, &pipe->wait, wait);
+	/*
+	 * .. and only then can you do the racy tests. That way,
+	 * if something changes and you got it wrong, the poll
+	 * table entry will wake you up and fix it.
+	 */
+	head = READ_ONCE(pipe->head);
+	tail = READ_ONCE(pipe->tail);
 
-	/* Reading only -- no need for acquiring the semaphore.  */
-	nrbufs = pipe->nrbufs;
 	mask = 0;
 	if (filp->f_mode & FMODE_READ) {
-		mask = (nrbufs > 0) ? EPOLLIN | EPOLLRDNORM : 0;
+		if (!pipe_empty(head, tail))
+			mask |= EPOLLIN | EPOLLRDNORM;
 		if (!pipe->writers && filp->f_version != pipe->w_counter)
 			mask |= EPOLLHUP;
 	}
 
 	if (filp->f_mode & FMODE_WRITE) {
-		mask |= (nrbufs < pipe->buffers) ? EPOLLOUT | EPOLLWRNORM : 0;
+		if (!pipe_full(head, tail, pipe->max_usage))
+			mask |= EPOLLOUT | EPOLLWRNORM;
 		/*
 		 * Most Unices do not set EPOLLERR for FIFOs but on Linux they
 		 * behave exactly like pipes for poll().
@@ -593,8 +722,10 @@ pipe_release(struct inode *inode, struct file *file)
 	if (file->f_mode & FMODE_WRITE)
 		pipe->writers--;
 
-	if (pipe->readers || pipe->writers) {
-		wake_up_interruptible_sync_poll(&pipe->wait, EPOLLIN | EPOLLOUT | EPOLLRDNORM | EPOLLWRNORM | EPOLLERR | EPOLLHUP);
+	/* Was that the last reader or writer, but not the other side? */
+	if (!pipe->readers != !pipe->writers) {
+		wake_up_interruptible_all(&pipe->rd_wait);
+		wake_up_interruptible_all(&pipe->wr_wait);
 		kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
 		kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT);
 	}
@@ -677,9 +808,11 @@ struct pipe_inode_info *alloc_pipe_info(void)
 			     GFP_KERNEL_ACCOUNT);
 
 	if (pipe->bufs) {
-		init_waitqueue_head(&pipe->wait);
+		init_waitqueue_head(&pipe->rd_wait);
+		init_waitqueue_head(&pipe->wr_wait);
 		pipe->r_counter = pipe->w_counter = 1;
-		pipe->buffers = pipe_bufs;
+		pipe->max_usage = pipe_bufs;
+		pipe->ring_size = pipe_bufs;
 		pipe->user = user;
 		mutex_init(&pipe->mutex);
 		return pipe;
@@ -697,9 +830,9 @@ void free_pipe_info(struct pipe_inode_info *pipe)
 {
 	int i;
 
-	(void) account_pipe_buffers(pipe->user, pipe->buffers, 0);
+	(void) account_pipe_buffers(pipe->user, pipe->ring_size, 0);
 	free_uid(pipe->user);
-	for (i = 0; i < pipe->buffers; i++) {
+	for (i = 0; i < pipe->ring_size; i++) {
 		struct pipe_buffer *buf = pipe->bufs + i;
 		if (buf->ops)
 			pipe_buf_release(pipe, buf);
@@ -793,6 +926,8 @@ int create_pipe_files(struct file **res, int flags)
 	}
 	res[0]->private_data = inode->i_pipe;
 	res[1] = f;
+	stream_open(inode, res[0]);
+	stream_open(inode, res[1]);
 	return 0;
 }
 
@@ -880,7 +1015,7 @@ SYSCALL_DEFINE1(pipe, int __user *, fildes)
 
 static int wait_for_partner(struct pipe_inode_info *pipe, unsigned int *cnt)
 {
-	int cur = *cnt;	
+	int cur = *cnt;
 
 	while (cur == *cnt) {
 		pipe_wait(pipe);
@@ -892,7 +1027,8 @@ static int wait_for_partner(struct pipe_inode_info *pipe, unsigned int *cnt)
 
 static void wake_up_partner(struct pipe_inode_info *pipe)
 {
-	wake_up_interruptible(&pipe->wait);
+	wake_up_interruptible_all(&pipe->rd_wait);
+	wake_up_interruptible_all(&pipe->wr_wait);
 }
 
 static int fifo_open(struct inode *inode, struct file *filp)
@@ -931,9 +1067,9 @@ static int fifo_open(struct inode *inode, struct file *filp)
 	__pipe_lock(pipe);
 
 	/* We can only do regular read/write on fifos */
-	filp->f_mode &= (FMODE_READ | FMODE_WRITE);
+	stream_open(inode, filp);
 
-	switch (filp->f_mode) {
+	switch (filp->f_mode & (FMODE_READ | FMODE_WRITE)) {
 	case FMODE_READ:
 	/*
 	 *  O_RDONLY
@@ -955,7 +1091,7 @@ static int fifo_open(struct inode *inode, struct file *filp)
 			}
 		}
 		break;
-	
+
 	case FMODE_WRITE:
 	/*
 	 *  O_WRONLY
@@ -975,7 +1111,7 @@ static int fifo_open(struct inode *inode, struct file *filp)
 				goto err_wr;
 		}
 		break;
-	
+
 	case FMODE_READ | FMODE_WRITE:
 	/*
 	 *  O_RDWR
@@ -1003,13 +1139,13 @@ static int fifo_open(struct inode *inode, struct file *filp)
 
 err_rd:
 	if (!--pipe->readers)
-		wake_up_interruptible(&pipe->wait);
+		wake_up_interruptible(&pipe->wr_wait);
 	ret = -ERESTARTSYS;
 	goto err;
 
 err_wr:
 	if (!--pipe->writers)
-		wake_up_interruptible(&pipe->wait);
+		wake_up_interruptible_all(&pipe->rd_wait);
 	ret = -ERESTARTSYS;
 	goto err;
 
@@ -1054,14 +1190,14 @@ unsigned int round_pipe_size(unsigned long size)
 static long pipe_set_size(struct pipe_inode_info *pipe, unsigned long arg)
 {
 	struct pipe_buffer *bufs;
-	unsigned int size, nr_pages;
+	unsigned int size, nr_slots, head, tail, mask, n;
 	unsigned long user_bufs;
 	long ret = 0;
 
 	size = round_pipe_size(arg);
-	nr_pages = size >> PAGE_SHIFT;
+	nr_slots = size >> PAGE_SHIFT;
 
-	if (!nr_pages)
+	if (!nr_slots)
 		return -EINVAL;
 
 	/*
@@ -1071,13 +1207,13 @@ static long pipe_set_size(struct pipe_inode_info *pipe, unsigned long arg)
 	 * Decreasing the pipe capacity is always permitted, even
 	 * if the user is currently over a limit.
 	 */
-	if (nr_pages > pipe->buffers &&
+	if (nr_slots > pipe->ring_size &&
 			size > pipe_max_size && !capable(CAP_SYS_RESOURCE))
 		return -EPERM;
 
-	user_bufs = account_pipe_buffers(pipe->user, pipe->buffers, nr_pages);
+	user_bufs = account_pipe_buffers(pipe->user, pipe->ring_size, nr_slots);
 
-	if (nr_pages > pipe->buffers &&
+	if (nr_slots > pipe->ring_size &&
 			(too_many_pipe_buffers_hard(user_bufs) ||
 			 too_many_pipe_buffers_soft(user_bufs)) &&
 			is_unprivileged_user()) {
@@ -1086,17 +1222,21 @@ static long pipe_set_size(struct pipe_inode_info *pipe, unsigned long arg)
 	}
 
 	/*
-	 * We can shrink the pipe, if arg >= pipe->nrbufs. Since we don't
-	 * expect a lot of shrink+grow operations, just free and allocate
-	 * again like we would do for growing. If the pipe currently
+	 * We can shrink the pipe, if arg is greater than the ring occupancy.
+	 * Since we don't expect a lot of shrink+grow operations, just free and
+	 * allocate again like we would do for growing.  If the pipe currently
 	 * contains more buffers than arg, then return busy.
 	 */
-	if (nr_pages < pipe->nrbufs) {
+	mask = pipe->ring_size - 1;
+	head = pipe->head;
+	tail = pipe->tail;
+	n = pipe_occupancy(pipe->head, pipe->tail);
+	if (nr_slots < n) {
 		ret = -EBUSY;
 		goto out_revert_acct;
 	}
 
-	bufs = kcalloc(nr_pages, sizeof(*bufs),
+	bufs = kcalloc(nr_slots, sizeof(*bufs),
 		       GFP_KERNEL_ACCOUNT | __GFP_NOWARN);
 	if (unlikely(!bufs)) {
 		ret = -ENOMEM;
@@ -1105,33 +1245,40 @@ static long pipe_set_size(struct pipe_inode_info *pipe, unsigned long arg)
 
 	/*
 	 * The pipe array wraps around, so just start the new one at zero
-	 * and adjust the indexes.
+	 * and adjust the indices.
 	 */
-	if (pipe->nrbufs) {
-		unsigned int tail;
-		unsigned int head;
-
-		tail = pipe->curbuf + pipe->nrbufs;
-		if (tail < pipe->buffers)
-			tail = 0;
-		else
-			tail &= (pipe->buffers - 1);
-
-		head = pipe->nrbufs - tail;
-		if (head)
-			memcpy(bufs, pipe->bufs + pipe->curbuf, head * sizeof(struct pipe_buffer));
-		if (tail)
-			memcpy(bufs + head, pipe->bufs, tail * sizeof(struct pipe_buffer));
+	if (n > 0) {
+		unsigned int h = head & mask;
+		unsigned int t = tail & mask;
+		if (h > t) {
+			memcpy(bufs, pipe->bufs + t,
+			       n * sizeof(struct pipe_buffer));
+		} else {
+			unsigned int tsize = pipe->ring_size - t;
+			if (h > 0)
+				memcpy(bufs + tsize, pipe->bufs,
+				       h * sizeof(struct pipe_buffer));
+			memcpy(bufs, pipe->bufs + t,
+			       tsize * sizeof(struct pipe_buffer));
+		}
 	}
 
-	pipe->curbuf = 0;
+	head = n;
+	tail = 0;
+
 	kfree(pipe->bufs);
 	pipe->bufs = bufs;
-	pipe->buffers = nr_pages;
-	return nr_pages * PAGE_SIZE;
+	pipe->ring_size = nr_slots;
+	pipe->max_usage = nr_slots;
+	pipe->tail = tail;
+	pipe->head = head;
+
+	/* This might have made more room for writers */
+	wake_up_interruptible(&pipe->wr_wait);
+	return pipe->max_usage * PAGE_SIZE;
 
 out_revert_acct:
-	(void) account_pipe_buffers(pipe->user, nr_pages, pipe->buffers);
+	(void) account_pipe_buffers(pipe->user, nr_slots, pipe->ring_size);
 	return ret;
 }
 
@@ -1161,7 +1308,7 @@ long pipe_fcntl(struct file *file, unsigned int cmd, unsigned long arg)
 		ret = pipe_set_size(pipe, arg);
 		break;
 	case F_GETPIPE_SZ:
-		ret = pipe->buffers * PAGE_SIZE;
+		ret = pipe->max_usage * PAGE_SIZE;
 		break;
 	default:
 		ret = -EINVAL;
diff --git a/fs/pnode.c b/fs/pnode.c
index 49f6d7ff2139..1106137c747a 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -261,14 +261,13 @@ static int propagate_one(struct mount *m)
 	child = copy_tree(last_source, last_source->mnt.mnt_root, type);
 	if (IS_ERR(child))
 		return PTR_ERR(child);
+	read_seqlock_excl(&mount_lock);
 	mnt_set_mountpoint(m, mp, child);
+	if (m->mnt_master != dest_master)
+		SET_MNT_MARK(m->mnt_master);
+	read_sequnlock_excl(&mount_lock);
 	last_dest = m;
 	last_source = child;
-	if (m->mnt_master != dest_master) {
-		read_seqlock_excl(&mount_lock);
-		SET_MNT_MARK(m->mnt_master);
-		read_sequnlock_excl(&mount_lock);
-	}
 	hlist_add_head(&child->mnt_hash, list);
 	return count_mounts(m->mnt_ns, child);
 }
diff --git a/fs/posix_acl.c b/fs/posix_acl.c
index 84ad1c90d535..249672bf54fe 100644
--- a/fs/posix_acl.c
+++ b/fs/posix_acl.c
@@ -631,12 +631,15 @@ EXPORT_SYMBOL_GPL(posix_acl_create);
 
 /**
  * posix_acl_update_mode  -  update mode in set_acl
+ * @inode: target inode
+ * @mode_p: mode (pointer) for update
+ * @acl: acl pointer
  *
  * Update the file mode when setting an ACL: compute the new file permission
  * bits based on the ACL.  In addition, if the ACL is equivalent to the new
- * file mode, set *acl to NULL to indicate that no ACL should be set.
+ * file mode, set *@acl to NULL to indicate that no ACL should be set.
  *
- * As with chmod, clear the setgit bit if the caller is not in the owning group
+ * As with chmod, clear the setgid bit if the caller is not in the owning group
  * or capable of CAP_FSETID (see inode_change_ok).
  *
  * Called from set_acl inode operations.
diff --git a/fs/proc/Kconfig b/fs/proc/Kconfig
index cb5629bd5fff..971a42f6357d 100644
--- a/fs/proc/Kconfig
+++ b/fs/proc/Kconfig
@@ -23,7 +23,7 @@ config PROC_FS
 	  /proc" or the equivalent line in /etc/fstab does the job.
 
 	  The /proc file system is explained in the file
-	  <file:Documentation/filesystems/proc.txt> and on the proc(5) manpage
+	  <file:Documentation/filesystems/proc.rst> and on the proc(5) manpage
 	  ("man 5 proc").
 
 	  This option will enlarge your kernel by about 67 KB. Several
@@ -42,8 +42,8 @@ config PROC_VMCORE
 	bool "/proc/vmcore support"
 	depends on PROC_FS && CRASH_DUMP
 	default y
-        help
-        Exports the dump image of crashed kernel in ELF format.
+	help
+	  Exports the dump image of crashed kernel in ELF format.
 
 config PROC_VMCORE_DEVICE_DUMP
 	bool "Device Hardware/Firmware Log Collection"
@@ -72,7 +72,7 @@ config PROC_SYSCTL
 	  a recompile of the kernel or reboot of the system.  The primary
 	  interface is through /proc/sys.  If you say Y here a tree of
 	  modifiable sysctl entries will be generated beneath the
-          /proc/sys directory. They are explained in the files
+	  /proc/sys directory. They are explained in the files
 	  in <file:Documentation/admin-guide/sysctl/>.  Note that enabling this
 	  option will enlarge the kernel by at least 8 KB.
 
@@ -88,14 +88,14 @@ config PROC_PAGE_MONITOR
 	  Various /proc files exist to monitor process memory utilization:
 	  /proc/pid/smaps, /proc/pid/clear_refs, /proc/pid/pagemap,
 	  /proc/kpagecount, and /proc/kpageflags. Disabling these
-          interfaces will reduce the size of the kernel by approximately 4kb.
+	  interfaces will reduce the size of the kernel by approximately 4kb.
 
 config PROC_CHILDREN
 	bool "Include /proc/<pid>/task/<tid>/children file"
 	default n
 	help
 	  Provides a fast way to retrieve first level children pids of a task. See
-	  <file:Documentation/filesystems/proc.txt> for more information.
+	  <file:Documentation/filesystems/proc.rst> for more information.
 
 	  Say Y if you are running any user-space software which takes benefit from
 	  this interface. For example, rkt is such a piece of software.
@@ -103,3 +103,7 @@ config PROC_CHILDREN
 config PROC_PID_ARCH_STATUS
 	def_bool n
 	depends on PROC_FS
+
+config PROC_CPU_RESCTRL
+	def_bool n
+	depends on PROC_FS
diff --git a/fs/proc/Makefile b/fs/proc/Makefile
index ead487e80510..bd08616ed8ba 100644
--- a/fs/proc/Makefile
+++ b/fs/proc/Makefile
@@ -33,3 +33,4 @@ proc-$(CONFIG_PROC_KCORE)	+= kcore.o
 proc-$(CONFIG_PROC_VMCORE)	+= vmcore.o
 proc-$(CONFIG_PRINTK)	+= kmsg.o
 proc-$(CONFIG_PROC_PAGE_MONITOR)	+= page.o
+proc-$(CONFIG_BOOT_CONFIG)	+= bootconfig.o
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 46dcb6f0eccf..8e16f14bb05a 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -533,7 +533,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns,
 	nice = task_nice(task);
 
 	/* convert nsec -> ticks */
-	start_time = nsec_to_clock_t(task->real_start_time);
+	start_time = nsec_to_clock_t(task->start_boottime);
 
 	seq_put_decimal_ull(m, "", pid_nr_ns(pid, ns));
 	seq_puts(m, " (");
@@ -635,28 +635,35 @@ int proc_tgid_stat(struct seq_file *m, struct pid_namespace *ns,
 int proc_pid_statm(struct seq_file *m, struct pid_namespace *ns,
 			struct pid *pid, struct task_struct *task)
 {
-	unsigned long size = 0, resident = 0, shared = 0, text = 0, data = 0;
 	struct mm_struct *mm = get_task_mm(task);
 
 	if (mm) {
+		unsigned long size;
+		unsigned long resident = 0;
+		unsigned long shared = 0;
+		unsigned long text = 0;
+		unsigned long data = 0;
+
 		size = task_statm(mm, &shared, &text, &data, &resident);
 		mmput(mm);
-	}
-	/*
-	 * For quick read, open code by putting numbers directly
-	 * expected format is
-	 * seq_printf(m, "%lu %lu %lu %lu 0 %lu 0\n",
-	 *               size, resident, shared, text, data);
-	 */
-	seq_put_decimal_ull(m, "", size);
-	seq_put_decimal_ull(m, " ", resident);
-	seq_put_decimal_ull(m, " ", shared);
-	seq_put_decimal_ull(m, " ", text);
-	seq_put_decimal_ull(m, " ", 0);
-	seq_put_decimal_ull(m, " ", data);
-	seq_put_decimal_ull(m, " ", 0);
-	seq_putc(m, '\n');
 
+		/*
+		 * For quick read, open code by putting numbers directly
+		 * expected format is
+		 * seq_printf(m, "%lu %lu %lu %lu 0 %lu 0\n",
+		 *               size, resident, shared, text, data);
+		 */
+		seq_put_decimal_ull(m, "", size);
+		seq_put_decimal_ull(m, " ", resident);
+		seq_put_decimal_ull(m, " ", shared);
+		seq_put_decimal_ull(m, " ", text);
+		seq_put_decimal_ull(m, " ", 0);
+		seq_put_decimal_ull(m, " ", data);
+		seq_put_decimal_ull(m, " ", 0);
+		seq_putc(m, '\n');
+	} else {
+		seq_write(m, "0 0 0 0 0 0 0\n", 14);
+	}
 	return 0;
 }
 
diff --git a/fs/proc/base.c b/fs/proc/base.c
index ebea9501afb8..eb2255e95f62 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -94,6 +94,8 @@
 #include <linux/sched/debug.h>
 #include <linux/sched/stat.h>
 #include <linux/posix-timers.h>
+#include <linux/time_namespace.h>
+#include <linux/resctrl.h>
 #include <trace/events/oom.h>
 #include "internal.h"
 #include "fd.h"
@@ -403,11 +405,11 @@ print0:
 
 static int lock_trace(struct task_struct *task)
 {
-	int err = mutex_lock_killable(&task->signal->cred_guard_mutex);
+	int err = mutex_lock_killable(&task->signal->exec_update_mutex);
 	if (err)
 		return err;
 	if (!ptrace_may_access(task, PTRACE_MODE_ATTACH_FSCREDS)) {
-		mutex_unlock(&task->signal->cred_guard_mutex);
+		mutex_unlock(&task->signal->exec_update_mutex);
 		return -EPERM;
 	}
 	return 0;
@@ -415,7 +417,7 @@ static int lock_trace(struct task_struct *task)
 
 static void unlock_trace(struct task_struct *task)
 {
-	mutex_unlock(&task->signal->cred_guard_mutex);
+	mutex_unlock(&task->signal->exec_update_mutex);
 }
 
 #ifdef CONFIG_STACKTRACE
@@ -1533,6 +1535,108 @@ static const struct file_operations proc_pid_sched_autogroup_operations = {
 
 #endif /* CONFIG_SCHED_AUTOGROUP */
 
+#ifdef CONFIG_TIME_NS
+static int timens_offsets_show(struct seq_file *m, void *v)
+{
+	struct task_struct *p;
+
+	p = get_proc_task(file_inode(m->file));
+	if (!p)
+		return -ESRCH;
+	proc_timens_show_offsets(p, m);
+
+	put_task_struct(p);
+
+	return 0;
+}
+
+static ssize_t timens_offsets_write(struct file *file, const char __user *buf,
+				    size_t count, loff_t *ppos)
+{
+	struct inode *inode = file_inode(file);
+	struct proc_timens_offset offsets[2];
+	char *kbuf = NULL, *pos, *next_line;
+	struct task_struct *p;
+	int ret, noffsets;
+
+	/* Only allow < page size writes at the beginning of the file */
+	if ((*ppos != 0) || (count >= PAGE_SIZE))
+		return -EINVAL;
+
+	/* Slurp in the user data */
+	kbuf = memdup_user_nul(buf, count);
+	if (IS_ERR(kbuf))
+		return PTR_ERR(kbuf);
+
+	/* Parse the user data */
+	ret = -EINVAL;
+	noffsets = 0;
+	for (pos = kbuf; pos; pos = next_line) {
+		struct proc_timens_offset *off = &offsets[noffsets];
+		char clock[10];
+		int err;
+
+		/* Find the end of line and ensure we don't look past it */
+		next_line = strchr(pos, '\n');
+		if (next_line) {
+			*next_line = '\0';
+			next_line++;
+			if (*next_line == '\0')
+				next_line = NULL;
+		}
+
+		err = sscanf(pos, "%9s %lld %lu", clock,
+				&off->val.tv_sec, &off->val.tv_nsec);
+		if (err != 3 || off->val.tv_nsec >= NSEC_PER_SEC)
+			goto out;
+
+		clock[sizeof(clock) - 1] = 0;
+		if (strcmp(clock, "monotonic") == 0 ||
+		    strcmp(clock, __stringify(CLOCK_MONOTONIC)) == 0)
+			off->clockid = CLOCK_MONOTONIC;
+		else if (strcmp(clock, "boottime") == 0 ||
+			 strcmp(clock, __stringify(CLOCK_BOOTTIME)) == 0)
+			off->clockid = CLOCK_BOOTTIME;
+		else
+			goto out;
+
+		noffsets++;
+		if (noffsets == ARRAY_SIZE(offsets)) {
+			if (next_line)
+				count = next_line - kbuf;
+			break;
+		}
+	}
+
+	ret = -ESRCH;
+	p = get_proc_task(inode);
+	if (!p)
+		goto out;
+	ret = proc_timens_set_offset(file, p, offsets, noffsets);
+	put_task_struct(p);
+	if (ret)
+		goto out;
+
+	ret = count;
+out:
+	kfree(kbuf);
+	return ret;
+}
+
+static int timens_offsets_open(struct inode *inode, struct file *filp)
+{
+	return single_open(filp, timens_offsets_show, inode);
+}
+
+static const struct file_operations proc_timens_offsets_operations = {
+	.open		= timens_offsets_open,
+	.read		= seq_read,
+	.write		= timens_offsets_write,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+#endif /* CONFIG_TIME_NS */
+
 static ssize_t comm_write(struct file *file, const char __user *buf,
 				size_t count, loff_t *offset)
 {
@@ -1626,8 +1730,7 @@ static const char *proc_pid_get_link(struct dentry *dentry,
 	if (error)
 		goto out;
 
-	nd_jump_link(&path);
-	return NULL;
+	error = nd_jump_link(&path);
 out:
 	return ERR_PTR(error);
 }
@@ -1743,11 +1846,25 @@ void task_dump_owner(struct task_struct *task, umode_t mode,
 	*rgid = gid;
 }
 
+void proc_pid_evict_inode(struct proc_inode *ei)
+{
+	struct pid *pid = ei->pid;
+
+	if (S_ISDIR(ei->vfs_inode.i_mode)) {
+		spin_lock(&pid->lock);
+		hlist_del_init_rcu(&ei->sibling_inodes);
+		spin_unlock(&pid->lock);
+	}
+
+	put_pid(pid);
+}
+
 struct inode *proc_pid_make_inode(struct super_block * sb,
 				  struct task_struct *task, umode_t mode)
 {
 	struct inode * inode;
 	struct proc_inode *ei;
+	struct pid *pid;
 
 	/* We need a new inode */
 
@@ -1765,10 +1882,18 @@ struct inode *proc_pid_make_inode(struct super_block * sb,
 	/*
 	 * grab the reference to task.
 	 */
-	ei->pid = get_task_pid(task, PIDTYPE_PID);
-	if (!ei->pid)
+	pid = get_task_pid(task, PIDTYPE_PID);
+	if (!pid)
 		goto out_unlock;
 
+	/* Let the pid remember us for quick removal */
+	ei->pid = pid;
+	if (S_ISDIR(mode)) {
+		spin_lock(&pid->lock);
+		hlist_add_head_rcu(&ei->sibling_inodes, &pid->inodes);
+		spin_unlock(&pid->lock);
+	}
+
 	task_dump_owner(task, 0, &inode->i_uid, &inode->i_gid);
 	security_task_to_inode(task, inode);
 
@@ -2770,7 +2895,7 @@ static int do_io_accounting(struct task_struct *task, struct seq_file *m, int wh
 	unsigned long flags;
 	int result;
 
-	result = mutex_lock_killable(&task->signal->cred_guard_mutex);
+	result = mutex_lock_killable(&task->signal->exec_update_mutex);
 	if (result)
 		return result;
 
@@ -2806,7 +2931,7 @@ static int do_io_accounting(struct task_struct *task, struct seq_file *m, int wh
 	result = 0;
 
 out_unlock:
-	mutex_unlock(&task->signal->cred_guard_mutex);
+	mutex_unlock(&task->signal->exec_update_mutex);
 	return result;
 }
 
@@ -3016,6 +3141,9 @@ static const struct pid_entry tgid_base_stuff[] = {
 #ifdef CONFIG_SCHED_AUTOGROUP
 	REG("autogroup",  S_IRUGO|S_IWUSR, proc_pid_sched_autogroup_operations),
 #endif
+#ifdef CONFIG_TIME_NS
+	REG("timens_offsets",  S_IRUGO|S_IWUSR, proc_timens_offsets_operations),
+#endif
 	REG("comm",      S_IRUGO|S_IWUSR, proc_pid_set_comm_operations),
 #ifdef CONFIG_HAVE_ARCH_TRACEHOOK
 	ONE("syscall",    S_IRUSR, proc_pid_syscall),
@@ -3061,6 +3189,9 @@ static const struct pid_entry tgid_base_stuff[] = {
 #ifdef CONFIG_CGROUPS
 	ONE("cgroup",  S_IRUGO, proc_cgroup_show),
 #endif
+#ifdef CONFIG_PROC_CPU_RESCTRL
+	ONE("cpu_resctrl_groups", S_IRUGO, proc_resctrl_show),
+#endif
 	ONE("oom_score",  S_IRUGO, proc_oom_score),
 	REG("oom_adj",    S_IRUGO|S_IWUSR, proc_oom_adj_operations),
 	REG("oom_score_adj", S_IRUGO|S_IWUSR, proc_oom_score_adj_operations),
@@ -3133,90 +3264,28 @@ static const struct inode_operations proc_tgid_base_inode_operations = {
 	.permission	= proc_pid_permission,
 };
 
-static void proc_flush_task_mnt(struct vfsmount *mnt, pid_t pid, pid_t tgid)
-{
-	struct dentry *dentry, *leader, *dir;
-	char buf[10 + 1];
-	struct qstr name;
-
-	name.name = buf;
-	name.len = snprintf(buf, sizeof(buf), "%u", pid);
-	/* no ->d_hash() rejects on procfs */
-	dentry = d_hash_and_lookup(mnt->mnt_root, &name);
-	if (dentry) {
-		d_invalidate(dentry);
-		dput(dentry);
-	}
-
-	if (pid == tgid)
-		return;
-
-	name.name = buf;
-	name.len = snprintf(buf, sizeof(buf), "%u", tgid);
-	leader = d_hash_and_lookup(mnt->mnt_root, &name);
-	if (!leader)
-		goto out;
-
-	name.name = "task";
-	name.len = strlen(name.name);
-	dir = d_hash_and_lookup(leader, &name);
-	if (!dir)
-		goto out_put_leader;
-
-	name.name = buf;
-	name.len = snprintf(buf, sizeof(buf), "%u", pid);
-	dentry = d_hash_and_lookup(dir, &name);
-	if (dentry) {
-		d_invalidate(dentry);
-		dput(dentry);
-	}
-
-	dput(dir);
-out_put_leader:
-	dput(leader);
-out:
-	return;
-}
-
 /**
- * proc_flush_task -  Remove dcache entries for @task from the /proc dcache.
- * @task: task that should be flushed.
- *
- * When flushing dentries from proc, one needs to flush them from global
- * proc (proc_mnt) and from all the namespaces' procs this task was seen
- * in. This call is supposed to do all of this job.
+ * proc_flush_pid -  Remove dcache entries for @pid from the /proc dcache.
+ * @pid: pid that should be flushed.
  *
- * Looks in the dcache for
- * /proc/@pid
- * /proc/@tgid/task/@pid
- * if either directory is present flushes it and all of it'ts children
- * from the dcache.
+ * This function walks a list of inodes (that belong to any proc
+ * filesystem) that are attached to the pid and flushes them from
+ * the dentry cache.
  *
  * It is safe and reasonable to cache /proc entries for a task until
  * that task exits.  After that they just clog up the dcache with
  * useless entries, possibly causing useful dcache entries to be
- * flushed instead.  This routine is proved to flush those useless
- * dcache entries at process exit time.
+ * flushed instead.  This routine is provided to flush those useless
+ * dcache entries when a process is reaped.
  *
  * NOTE: This routine is just an optimization so it does not guarantee
- *       that no dcache entries will exist at process exit time it
- *       just makes it very unlikely that any will persist.
+ *       that no dcache entries will exist after a process is reaped
+ *       it just makes it very unlikely that any will persist.
  */
 
-void proc_flush_task(struct task_struct *task)
+void proc_flush_pid(struct pid *pid)
 {
-	int i;
-	struct pid *pid, *tgid;
-	struct upid *upid;
-
-	pid = task_pid(task);
-	tgid = task_tgid(task);
-
-	for (i = 0; i <= pid->level; i++) {
-		upid = &pid->numbers[i];
-		proc_flush_task_mnt(upid->ns->proc_mnt, upid->nr,
-					tgid->numbers[i].nr);
-	}
+	proc_invalidate_siblings_dcache(&pid->inodes, &pid->lock);
 }
 
 static struct dentry *proc_pid_instantiate(struct dentry * dentry,
@@ -3461,6 +3530,9 @@ static const struct pid_entry tid_base_stuff[] = {
 #ifdef CONFIG_CGROUPS
 	ONE("cgroup",  S_IRUGO, proc_cgroup_show),
 #endif
+#ifdef CONFIG_PROC_CPU_RESCTRL
+	ONE("cpu_resctrl_groups", S_IRUGO, proc_resctrl_show),
+#endif
 	ONE("oom_score", S_IRUGO, proc_oom_score),
 	REG("oom_adj",   S_IRUGO|S_IWUSR, proc_oom_adj_operations),
 	REG("oom_score_adj", S_IRUGO|S_IWUSR, proc_oom_score_adj_operations),
diff --git a/fs/proc/bootconfig.c b/fs/proc/bootconfig.c
new file mode 100644
index 000000000000..9955d75c0585
--- /dev/null
+++ b/fs/proc/bootconfig.c
@@ -0,0 +1,89 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * /proc/bootconfig - Extra boot configuration
+ */
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/printk.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/bootconfig.h>
+#include <linux/slab.h>
+
+static char *saved_boot_config;
+
+static int boot_config_proc_show(struct seq_file *m, void *v)
+{
+	if (saved_boot_config)
+		seq_puts(m, saved_boot_config);
+	return 0;
+}
+
+/* Rest size of buffer */
+#define rest(dst, end) ((end) > (dst) ? (end) - (dst) : 0)
+
+/* Return the needed total length if @size is 0 */
+static int __init copy_xbc_key_value_list(char *dst, size_t size)
+{
+	struct xbc_node *leaf, *vnode;
+	const char *val;
+	char *key, *end = dst + size;
+	int ret = 0;
+
+	key = kzalloc(XBC_KEYLEN_MAX, GFP_KERNEL);
+
+	xbc_for_each_key_value(leaf, val) {
+		ret = xbc_node_compose_key(leaf, key, XBC_KEYLEN_MAX);
+		if (ret < 0)
+			break;
+		ret = snprintf(dst, rest(dst, end), "%s = ", key);
+		if (ret < 0)
+			break;
+		dst += ret;
+		vnode = xbc_node_get_child(leaf);
+		if (vnode && xbc_node_is_array(vnode)) {
+			xbc_array_for_each_value(vnode, val) {
+				ret = snprintf(dst, rest(dst, end), "\"%s\"%s",
+					val, vnode->next ? ", " : "\n");
+				if (ret < 0)
+					goto out;
+				dst += ret;
+			}
+		} else {
+			ret = snprintf(dst, rest(dst, end), "\"%s\"\n", val);
+			if (ret < 0)
+				break;
+			dst += ret;
+		}
+	}
+out:
+	kfree(key);
+
+	return ret < 0 ? ret : dst - (end - size);
+}
+
+static int __init proc_boot_config_init(void)
+{
+	int len;
+
+	len = copy_xbc_key_value_list(NULL, 0);
+	if (len < 0)
+		return len;
+
+	if (len > 0) {
+		saved_boot_config = kzalloc(len + 1, GFP_KERNEL);
+		if (!saved_boot_config)
+			return -ENOMEM;
+
+		len = copy_xbc_key_value_list(saved_boot_config, len + 1);
+		if (len < 0) {
+			kfree(saved_boot_config);
+			return len;
+		}
+	}
+
+	proc_create_single("bootconfig", 0, NULL, boot_config_proc_show);
+
+	return 0;
+}
+fs_initcall(proc_boot_config_init);
diff --git a/fs/proc/cpuinfo.c b/fs/proc/cpuinfo.c
index 96f1087e372c..d0989a443c77 100644
--- a/fs/proc/cpuinfo.c
+++ b/fs/proc/cpuinfo.c
@@ -16,16 +16,17 @@ static int cpuinfo_open(struct inode *inode, struct file *file)
 	return seq_open(file, &cpuinfo_op);
 }
 
-static const struct file_operations proc_cpuinfo_operations = {
-	.open		= cpuinfo_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= seq_release,
+static const struct proc_ops cpuinfo_proc_ops = {
+	.proc_flags	= PROC_ENTRY_PERMANENT,
+	.proc_open	= cpuinfo_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= seq_release,
 };
 
 static int __init proc_cpuinfo_init(void)
 {
-	proc_create("cpuinfo", 0, NULL, &proc_cpuinfo_operations);
+	proc_create("cpuinfo", 0, NULL, &cpuinfo_proc_ops);
 	return 0;
 }
 fs_initcall(proc_cpuinfo_init);
diff --git a/fs/proc/generic.c b/fs/proc/generic.c
index 64e9ee1b129e..4ed6dabdf6ff 100644
--- a/fs/proc/generic.c
+++ b/fs/proc/generic.c
@@ -138,8 +138,12 @@ static int proc_getattr(const struct path *path, struct kstat *stat,
 {
 	struct inode *inode = d_inode(path->dentry);
 	struct proc_dir_entry *de = PDE(inode);
-	if (de && de->nlink)
-		set_nlink(inode, de->nlink);
+	if (de) {
+		nlink_t nlink = READ_ONCE(de->nlink);
+		if (nlink > 0) {
+			set_nlink(inode, nlink);
+		}
+	}
 
 	generic_fillattr(inode, stat);
 	return 0;
@@ -159,7 +163,6 @@ static int __xlate_proc_name(const char *name, struct proc_dir_entry **ret,
 {
 	const char     		*cp = name, *next;
 	struct proc_dir_entry	*de;
-	unsigned int		len;
 
 	de = *ret;
 	if (!de)
@@ -170,13 +173,12 @@ static int __xlate_proc_name(const char *name, struct proc_dir_entry **ret,
 		if (!next)
 			break;
 
-		len = next - cp;
-		de = pde_subdir_find(de, cp, len);
+		de = pde_subdir_find(de, cp, next - cp);
 		if (!de) {
 			WARN(1, "name '%s'\n", name);
 			return -ENOENT;
 		}
-		cp += len + 1;
+		cp = next + 1;
 	}
 	*residual = cp;
 	*ret = de;
@@ -362,6 +364,7 @@ struct proc_dir_entry *proc_register(struct proc_dir_entry *dir,
 		write_unlock(&proc_subdir_lock);
 		goto out_free_inum;
 	}
+	dir->nlink++;
 	write_unlock(&proc_subdir_lock);
 
 	return dp;
@@ -470,12 +473,9 @@ struct proc_dir_entry *proc_mkdir_data(const char *name, umode_t mode,
 	ent = __proc_create(&parent, name, S_IFDIR | mode, 2);
 	if (ent) {
 		ent->data = data;
-		ent->proc_fops = &proc_dir_operations;
+		ent->proc_dir_ops = &proc_dir_operations;
 		ent->proc_iops = &proc_dir_inode_operations;
-		parent->nlink++;
 		ent = proc_register(parent, ent);
-		if (!ent)
-			parent->nlink--;
 	}
 	return ent;
 }
@@ -503,12 +503,9 @@ struct proc_dir_entry *proc_create_mount_point(const char *name)
 	ent = __proc_create(&parent, name, mode, 2);
 	if (ent) {
 		ent->data = NULL;
-		ent->proc_fops = NULL;
+		ent->proc_dir_ops = NULL;
 		ent->proc_iops = NULL;
-		parent->nlink++;
 		ent = proc_register(parent, ent);
-		if (!ent)
-			parent->nlink--;
 	}
 	return ent;
 }
@@ -534,27 +531,32 @@ struct proc_dir_entry *proc_create_reg(const char *name, umode_t mode,
 	return p;
 }
 
+static inline void pde_set_flags(struct proc_dir_entry *pde)
+{
+	if (pde->proc_ops->proc_flags & PROC_ENTRY_PERMANENT)
+		pde->flags |= PROC_ENTRY_PERMANENT;
+}
+
 struct proc_dir_entry *proc_create_data(const char *name, umode_t mode,
 		struct proc_dir_entry *parent,
-		const struct file_operations *proc_fops, void *data)
+		const struct proc_ops *proc_ops, void *data)
 {
 	struct proc_dir_entry *p;
 
-	BUG_ON(proc_fops == NULL);
-
 	p = proc_create_reg(name, mode, &parent, data);
 	if (!p)
 		return NULL;
-	p->proc_fops = proc_fops;
+	p->proc_ops = proc_ops;
+	pde_set_flags(p);
 	return proc_register(parent, p);
 }
 EXPORT_SYMBOL(proc_create_data);
  
 struct proc_dir_entry *proc_create(const char *name, umode_t mode,
 				   struct proc_dir_entry *parent,
-				   const struct file_operations *proc_fops)
+				   const struct proc_ops *proc_ops)
 {
-	return proc_create_data(name, mode, parent, proc_fops, NULL);
+	return proc_create_data(name, mode, parent, proc_ops, NULL);
 }
 EXPORT_SYMBOL(proc_create);
 
@@ -576,11 +578,12 @@ static int proc_seq_release(struct inode *inode, struct file *file)
 	return seq_release(inode, file);
 }
 
-static const struct file_operations proc_seq_fops = {
-	.open		= proc_seq_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= proc_seq_release,
+static const struct proc_ops proc_seq_ops = {
+	/* not permanent -- can call into arbitrary seq_operations */
+	.proc_open	= proc_seq_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= proc_seq_release,
 };
 
 struct proc_dir_entry *proc_create_seq_private(const char *name, umode_t mode,
@@ -592,7 +595,7 @@ struct proc_dir_entry *proc_create_seq_private(const char *name, umode_t mode,
 	p = proc_create_reg(name, mode, &parent, data);
 	if (!p)
 		return NULL;
-	p->proc_fops = &proc_seq_fops;
+	p->proc_ops = &proc_seq_ops;
 	p->seq_ops = ops;
 	p->state_size = state_size;
 	return proc_register(parent, p);
@@ -606,11 +609,12 @@ static int proc_single_open(struct inode *inode, struct file *file)
 	return single_open(file, de->single_show, de->data);
 }
 
-static const struct file_operations proc_single_fops = {
-	.open		= proc_single_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
+static const struct proc_ops proc_single_ops = {
+	/* not permanent -- can call into arbitrary ->single_show */
+	.proc_open	= proc_single_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
 };
 
 struct proc_dir_entry *proc_create_single_data(const char *name, umode_t mode,
@@ -622,7 +626,7 @@ struct proc_dir_entry *proc_create_single_data(const char *name, umode_t mode,
 	p = proc_create_reg(name, mode, &parent, data);
 	if (!p)
 		return NULL;
-	p->proc_fops = &proc_single_fops;
+	p->proc_ops = &proc_single_ops;
 	p->single_show = show;
 	return proc_register(parent, p);
 }
@@ -666,8 +670,16 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent)
 	len = strlen(fn);
 
 	de = pde_subdir_find(parent, fn, len);
-	if (de)
-		rb_erase(&de->subdir_node, &parent->subdir);
+	if (de) {
+		if (unlikely(pde_is_permanent(de))) {
+			WARN(1, "removing permanent /proc entry '%s'", de->name);
+			de = NULL;
+		} else {
+			rb_erase(&de->subdir_node, &parent->subdir);
+			if (S_ISDIR(de->mode))
+				parent->nlink--;
+		}
+	}
 	write_unlock(&proc_subdir_lock);
 	if (!de) {
 		WARN(1, "name '%s'\n", name);
@@ -676,9 +688,6 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent)
 
 	proc_entry_rundown(de);
 
-	if (S_ISDIR(de->mode))
-		parent->nlink--;
-	de->nlink = 0;
 	WARN(pde_subdir_first(de),
 	     "%s: removing non-empty directory '%s/%s', leaking at least '%s'\n",
 	     __func__, de->parent->name, de->name, pde_subdir_first(de)->name);
@@ -704,23 +713,34 @@ int remove_proc_subtree(const char *name, struct proc_dir_entry *parent)
 		write_unlock(&proc_subdir_lock);
 		return -ENOENT;
 	}
+	if (unlikely(pde_is_permanent(root))) {
+		write_unlock(&proc_subdir_lock);
+		WARN(1, "removing permanent /proc entry '%s/%s'",
+			root->parent->name, root->name);
+		return -EINVAL;
+	}
 	rb_erase(&root->subdir_node, &parent->subdir);
 
 	de = root;
 	while (1) {
 		next = pde_subdir_first(de);
 		if (next) {
+			if (unlikely(pde_is_permanent(root))) {
+				write_unlock(&proc_subdir_lock);
+				WARN(1, "removing permanent /proc entry '%s/%s'",
+					next->parent->name, next->name);
+				return -EINVAL;
+			}
 			rb_erase(&next->subdir_node, &de->subdir);
 			de = next;
 			continue;
 		}
-		write_unlock(&proc_subdir_lock);
-
-		proc_entry_rundown(de);
 		next = de->parent;
 		if (S_ISDIR(de->mode))
 			next->nlink--;
-		de->nlink = 0;
+		write_unlock(&proc_subdir_lock);
+
+		proc_entry_rundown(de);
 		if (de == root)
 			break;
 		pde_put(de);
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index dbe43a50caf2..fb4cace9ea41 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -33,21 +33,27 @@ static void proc_evict_inode(struct inode *inode)
 {
 	struct proc_dir_entry *de;
 	struct ctl_table_header *head;
+	struct proc_inode *ei = PROC_I(inode);
 
 	truncate_inode_pages_final(&inode->i_data);
 	clear_inode(inode);
 
 	/* Stop tracking associated processes */
-	put_pid(PROC_I(inode)->pid);
+	if (ei->pid) {
+		proc_pid_evict_inode(ei);
+		ei->pid = NULL;
+	}
 
 	/* Let go of any associated proc directory entry */
-	de = PDE(inode);
-	if (de)
+	de = ei->pde;
+	if (de) {
 		pde_put(de);
+		ei->pde = NULL;
+	}
 
-	head = PROC_I(inode)->sysctl;
+	head = ei->sysctl;
 	if (head) {
-		RCU_INIT_POINTER(PROC_I(inode)->sysctl, NULL);
+		RCU_INIT_POINTER(ei->sysctl, NULL);
 		proc_sys_evict_inode(inode, head);
 	}
 }
@@ -68,6 +74,7 @@ static struct inode *proc_alloc_inode(struct super_block *sb)
 	ei->pde = NULL;
 	ei->sysctl = NULL;
 	ei->sysctl_entry = NULL;
+	INIT_HLIST_NODE(&ei->sibling_inodes);
 	ei->ns_ops = NULL;
 	return &ei->vfs_inode;
 }
@@ -102,6 +109,62 @@ void __init proc_init_kmemcache(void)
 	BUILD_BUG_ON(sizeof(struct proc_dir_entry) >= SIZEOF_PDE);
 }
 
+void proc_invalidate_siblings_dcache(struct hlist_head *inodes, spinlock_t *lock)
+{
+	struct inode *inode;
+	struct proc_inode *ei;
+	struct hlist_node *node;
+	struct super_block *old_sb = NULL;
+
+	rcu_read_lock();
+	for (;;) {
+		struct super_block *sb;
+		node = hlist_first_rcu(inodes);
+		if (!node)
+			break;
+		ei = hlist_entry(node, struct proc_inode, sibling_inodes);
+		spin_lock(lock);
+		hlist_del_init_rcu(&ei->sibling_inodes);
+		spin_unlock(lock);
+
+		inode = &ei->vfs_inode;
+		sb = inode->i_sb;
+		if ((sb != old_sb) && !atomic_inc_not_zero(&sb->s_active))
+			continue;
+		inode = igrab(inode);
+		rcu_read_unlock();
+		if (sb != old_sb) {
+			if (old_sb)
+				deactivate_super(old_sb);
+			old_sb = sb;
+		}
+		if (unlikely(!inode)) {
+			rcu_read_lock();
+			continue;
+		}
+
+		if (S_ISDIR(inode->i_mode)) {
+			struct dentry *dir = d_find_any_alias(inode);
+			if (dir) {
+				d_invalidate(dir);
+				dput(dir);
+			}
+		} else {
+			struct dentry *dentry;
+			while ((dentry = d_find_alias(inode))) {
+				d_invalidate(dentry);
+				dput(dentry);
+			}
+		}
+		iput(inode);
+
+		rcu_read_lock();
+	}
+	rcu_read_unlock();
+	if (old_sb)
+		deactivate_super(old_sb);
+}
+
 static int proc_show_options(struct seq_file *seq, struct dentry *root)
 {
 	struct super_block *sb = root->d_sb;
@@ -139,6 +202,7 @@ static void unuse_pde(struct proc_dir_entry *pde)
 
 /* pde is locked on entry, unlocked on exit */
 static void close_pdeo(struct proc_dir_entry *pde, struct pde_opener *pdeo)
+	__releases(&pde->pde_unload_lock)
 {
 	/*
 	 * close() (proc_reg_release()) can't delete an entry and proceed:
@@ -163,7 +227,7 @@ static void close_pdeo(struct proc_dir_entry *pde, struct pde_opener *pdeo)
 		pdeo->closing = true;
 		spin_unlock(&pde->pde_unload_lock);
 		file = pdeo->file;
-		pde->proc_fops->release(file_inode(file), file);
+		pde->proc_ops->proc_release(file_inode(file), file);
 		spin_lock(&pde->pde_unload_lock);
 		/* After ->release. */
 		list_del(&pdeo->lh);
@@ -195,135 +259,204 @@ void proc_entry_rundown(struct proc_dir_entry *de)
 	spin_unlock(&de->pde_unload_lock);
 }
 
+static loff_t pde_lseek(struct proc_dir_entry *pde, struct file *file, loff_t offset, int whence)
+{
+	typeof_member(struct proc_ops, proc_lseek) lseek;
+
+	lseek = pde->proc_ops->proc_lseek;
+	if (!lseek)
+		lseek = default_llseek;
+	return lseek(file, offset, whence);
+}
+
 static loff_t proc_reg_llseek(struct file *file, loff_t offset, int whence)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	loff_t rv = -EINVAL;
-	if (use_pde(pde)) {
-		typeof_member(struct file_operations, llseek) llseek;
 
-		llseek = pde->proc_fops->llseek;
-		if (!llseek)
-			llseek = default_llseek;
-		rv = llseek(file, offset, whence);
+	if (pde_is_permanent(pde)) {
+		return pde_lseek(pde, file, offset, whence);
+	} else if (use_pde(pde)) {
+		rv = pde_lseek(pde, file, offset, whence);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
+static ssize_t pde_read(struct proc_dir_entry *pde, struct file *file, char __user *buf, size_t count, loff_t *ppos)
+{
+	typeof_member(struct proc_ops, proc_read) read;
+
+	read = pde->proc_ops->proc_read;
+	if (read)
+		return read(file, buf, count, ppos);
+	return -EIO;
+}
+
 static ssize_t proc_reg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	ssize_t rv = -EIO;
-	if (use_pde(pde)) {
-		typeof_member(struct file_operations, read) read;
 
-		read = pde->proc_fops->read;
-		if (read)
-			rv = read(file, buf, count, ppos);
+	if (pde_is_permanent(pde)) {
+		return pde_read(pde, file, buf, count, ppos);
+	} else if (use_pde(pde)) {
+		rv = pde_read(pde, file, buf, count, ppos);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
+static ssize_t pde_write(struct proc_dir_entry *pde, struct file *file, const char __user *buf, size_t count, loff_t *ppos)
+{
+	typeof_member(struct proc_ops, proc_write) write;
+
+	write = pde->proc_ops->proc_write;
+	if (write)
+		return write(file, buf, count, ppos);
+	return -EIO;
+}
+
 static ssize_t proc_reg_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	ssize_t rv = -EIO;
-	if (use_pde(pde)) {
-		typeof_member(struct file_operations, write) write;
 
-		write = pde->proc_fops->write;
-		if (write)
-			rv = write(file, buf, count, ppos);
+	if (pde_is_permanent(pde)) {
+		return pde_write(pde, file, buf, count, ppos);
+	} else if (use_pde(pde)) {
+		rv = pde_write(pde, file, buf, count, ppos);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
+static __poll_t pde_poll(struct proc_dir_entry *pde, struct file *file, struct poll_table_struct *pts)
+{
+	typeof_member(struct proc_ops, proc_poll) poll;
+
+	poll = pde->proc_ops->proc_poll;
+	if (poll)
+		return poll(file, pts);
+	return DEFAULT_POLLMASK;
+}
+
 static __poll_t proc_reg_poll(struct file *file, struct poll_table_struct *pts)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	__poll_t rv = DEFAULT_POLLMASK;
-	if (use_pde(pde)) {
-		typeof_member(struct file_operations, poll) poll;
 
-		poll = pde->proc_fops->poll;
-		if (poll)
-			rv = poll(file, pts);
+	if (pde_is_permanent(pde)) {
+		return pde_poll(pde, file, pts);
+	} else if (use_pde(pde)) {
+		rv = pde_poll(pde, file, pts);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
+static long pde_ioctl(struct proc_dir_entry *pde, struct file *file, unsigned int cmd, unsigned long arg)
+{
+	typeof_member(struct proc_ops, proc_ioctl) ioctl;
+
+	ioctl = pde->proc_ops->proc_ioctl;
+	if (ioctl)
+		return ioctl(file, cmd, arg);
+	return -ENOTTY;
+}
+
 static long proc_reg_unlocked_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	long rv = -ENOTTY;
-	if (use_pde(pde)) {
-		typeof_member(struct file_operations, unlocked_ioctl) ioctl;
 
-		ioctl = pde->proc_fops->unlocked_ioctl;
-		if (ioctl)
-			rv = ioctl(file, cmd, arg);
+	if (pde_is_permanent(pde)) {
+		return pde_ioctl(pde, file, cmd, arg);
+	} else if (use_pde(pde)) {
+		rv = pde_ioctl(pde, file, cmd, arg);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
 #ifdef CONFIG_COMPAT
+static long pde_compat_ioctl(struct proc_dir_entry *pde, struct file *file, unsigned int cmd, unsigned long arg)
+{
+	typeof_member(struct proc_ops, proc_compat_ioctl) compat_ioctl;
+
+	compat_ioctl = pde->proc_ops->proc_compat_ioctl;
+	if (compat_ioctl)
+		return compat_ioctl(file, cmd, arg);
+	return -ENOTTY;
+}
+
 static long proc_reg_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	long rv = -ENOTTY;
-	if (use_pde(pde)) {
-		typeof_member(struct file_operations, compat_ioctl) compat_ioctl;
-
-		compat_ioctl = pde->proc_fops->compat_ioctl;
-		if (compat_ioctl)
-			rv = compat_ioctl(file, cmd, arg);
+	if (pde_is_permanent(pde)) {
+		return pde_compat_ioctl(pde, file, cmd, arg);
+	} else if (use_pde(pde)) {
+		rv = pde_compat_ioctl(pde, file, cmd, arg);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 #endif
 
+static int pde_mmap(struct proc_dir_entry *pde, struct file *file, struct vm_area_struct *vma)
+{
+	typeof_member(struct proc_ops, proc_mmap) mmap;
+
+	mmap = pde->proc_ops->proc_mmap;
+	if (mmap)
+		return mmap(file, vma);
+	return -EIO;
+}
+
 static int proc_reg_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	int rv = -EIO;
-	if (use_pde(pde)) {
-		typeof_member(struct file_operations, mmap) mmap;
 
-		mmap = pde->proc_fops->mmap;
-		if (mmap)
-			rv = mmap(file, vma);
+	if (pde_is_permanent(pde)) {
+		return pde_mmap(pde, file, vma);
+	} else if (use_pde(pde)) {
+		rv = pde_mmap(pde, file, vma);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
 static unsigned long
-proc_reg_get_unmapped_area(struct file *file, unsigned long orig_addr,
+pde_get_unmapped_area(struct proc_dir_entry *pde, struct file *file, unsigned long orig_addr,
 			   unsigned long len, unsigned long pgoff,
 			   unsigned long flags)
 {
-	struct proc_dir_entry *pde = PDE(file_inode(file));
-	unsigned long rv = -EIO;
-
-	if (use_pde(pde)) {
-		typeof_member(struct file_operations, get_unmapped_area) get_area;
+	typeof_member(struct proc_ops, proc_get_unmapped_area) get_area;
 
-		get_area = pde->proc_fops->get_unmapped_area;
+	get_area = pde->proc_ops->proc_get_unmapped_area;
 #ifdef CONFIG_MMU
-		if (!get_area)
-			get_area = current->mm->get_unmapped_area;
+	if (!get_area)
+		get_area = current->mm->get_unmapped_area;
 #endif
+	if (get_area)
+		return get_area(file, orig_addr, len, pgoff, flags);
+	return orig_addr;
+}
+
+static unsigned long
+proc_reg_get_unmapped_area(struct file *file, unsigned long orig_addr,
+			   unsigned long len, unsigned long pgoff,
+			   unsigned long flags)
+{
+	struct proc_dir_entry *pde = PDE(file_inode(file));
+	unsigned long rv = -EIO;
 
-		if (get_area)
-			rv = get_area(file, orig_addr, len, pgoff, flags);
-		else
-			rv = orig_addr;
+	if (pde_is_permanent(pde)) {
+		return pde_get_unmapped_area(pde, file, orig_addr, len, pgoff, flags);
+	} else if (use_pde(pde)) {
+		rv = pde_get_unmapped_area(pde, file, orig_addr, len, pgoff, flags);
 		unuse_pde(pde);
 	}
 	return rv;
@@ -333,10 +466,17 @@ static int proc_reg_open(struct inode *inode, struct file *file)
 {
 	struct proc_dir_entry *pde = PDE(inode);
 	int rv = 0;
-	typeof_member(struct file_operations, open) open;
-	typeof_member(struct file_operations, release) release;
+	typeof_member(struct proc_ops, proc_open) open;
+	typeof_member(struct proc_ops, proc_release) release;
 	struct pde_opener *pdeo;
 
+	if (pde_is_permanent(pde)) {
+		open = pde->proc_ops->proc_open;
+		if (open)
+			rv = open(inode, file);
+		return rv;
+	}
+
 	/*
 	 * Ensure that
 	 * 1) PDE's ->release hook will be called no matter what
@@ -351,7 +491,7 @@ static int proc_reg_open(struct inode *inode, struct file *file)
 	if (!use_pde(pde))
 		return -ENOENT;
 
-	release = pde->proc_fops->release;
+	release = pde->proc_ops->proc_release;
 	if (release) {
 		pdeo = kmem_cache_alloc(pde_opener_cache, GFP_KERNEL);
 		if (!pdeo) {
@@ -360,7 +500,7 @@ static int proc_reg_open(struct inode *inode, struct file *file)
 		}
 	}
 
-	open = pde->proc_fops->open;
+	open = pde->proc_ops->proc_open;
 	if (open)
 		rv = open(inode, file);
 
@@ -386,6 +526,17 @@ static int proc_reg_release(struct inode *inode, struct file *file)
 {
 	struct proc_dir_entry *pde = PDE(inode);
 	struct pde_opener *pdeo;
+
+	if (pde_is_permanent(pde)) {
+		typeof_member(struct proc_ops, proc_release) release;
+
+		release = pde->proc_ops->proc_release;
+		if (release) {
+			return release(inode, file);
+		}
+		return 0;
+	}
+
 	spin_lock(&pde->pde_unload_lock);
 	list_for_each_entry(pdeo, &pde->pde_openers, lh) {
 		if (pdeo->file == file) {
@@ -468,21 +619,23 @@ struct inode *proc_get_inode(struct super_block *sb, struct proc_dir_entry *de)
 			inode->i_size = de->size;
 		if (de->nlink)
 			set_nlink(inode, de->nlink);
-		WARN_ON(!de->proc_iops);
-		inode->i_op = de->proc_iops;
-		if (de->proc_fops) {
-			if (S_ISREG(inode->i_mode)) {
+
+		if (S_ISREG(inode->i_mode)) {
+			inode->i_op = de->proc_iops;
+			inode->i_fop = &proc_reg_file_ops;
 #ifdef CONFIG_COMPAT
-				if (!de->proc_fops->compat_ioctl)
-					inode->i_fop =
-						&proc_reg_file_ops_no_compat;
-				else
-#endif
-					inode->i_fop = &proc_reg_file_ops;
-			} else {
-				inode->i_fop = de->proc_fops;
+			if (!de->proc_ops->proc_compat_ioctl) {
+				inode->i_fop = &proc_reg_file_ops_no_compat;
 			}
-		}
+#endif
+		} else if (S_ISDIR(inode->i_mode)) {
+			inode->i_op = de->proc_iops;
+			inode->i_fop = de->proc_dir_ops;
+		} else if (S_ISLNK(inode->i_mode)) {
+			inode->i_op = de->proc_iops;
+			inode->i_fop = NULL;
+		} else
+			BUG();
 	} else
 	       pde_put(de);
 	return inode;
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index cd0c8d5ce9a1..917cc85e3466 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -39,7 +39,10 @@ struct proc_dir_entry {
 	spinlock_t pde_unload_lock;
 	struct completion *pde_unload_completion;
 	const struct inode_operations *proc_iops;
-	const struct file_operations *proc_fops;
+	union {
+		const struct proc_ops *proc_ops;
+		const struct file_operations *proc_dir_ops;
+	};
 	const struct dentry_operations *proc_dops;
 	union {
 		const struct seq_operations *seq_ops;
@@ -58,6 +61,7 @@ struct proc_dir_entry {
 	struct rb_node subdir_node;
 	char *name;
 	umode_t mode;
+	u8 flags;
 	u8 namelen;
 	char inline_name[];
 } __randomize_layout;
@@ -70,6 +74,11 @@ struct proc_dir_entry {
 	0)
 #define SIZEOF_PDE_INLINE_NAME (SIZEOF_PDE - sizeof(struct proc_dir_entry))
 
+static inline bool pde_is_permanent(const struct proc_dir_entry *pde)
+{
+	return pde->flags & PROC_ENTRY_PERMANENT;
+}
+
 extern struct kmem_cache *proc_dir_entry_cache;
 void pde_free(struct proc_dir_entry *pde);
 
@@ -88,7 +97,7 @@ struct proc_inode {
 	struct proc_dir_entry *pde;
 	struct ctl_table_header *sysctl;
 	struct ctl_table *sysctl_entry;
-	struct hlist_node sysctl_inodes;
+	struct hlist_node sibling_inodes;
 	const struct proc_ns_operations *ns_ops;
 	struct inode vfs_inode;
 } __randomize_layout;
@@ -155,6 +164,7 @@ extern int proc_pid_statm(struct seq_file *, struct pid_namespace *,
 extern const struct dentry_operations pid_dentry_operations;
 extern int pid_getattr(const struct path *, struct kstat *, u32, unsigned int);
 extern int proc_setattr(struct dentry *, struct iattr *);
+extern void proc_pid_evict_inode(struct proc_inode *);
 extern struct inode *proc_pid_make_inode(struct super_block *, struct task_struct *, umode_t);
 extern void pid_update_inode(struct task_struct *, struct inode *);
 extern int pid_delete_dentry(const struct dentry *);
@@ -197,8 +207,8 @@ extern ssize_t proc_simple_write(struct file *, const char __user *, size_t, lof
  * inode.c
  */
 struct pde_opener {
-	struct file *file;
 	struct list_head lh;
+	struct file *file;
 	bool closing;
 	struct completion *c;
 } __randomize_layout;
@@ -207,6 +217,7 @@ extern const struct inode_operations proc_pid_link_inode_operations;
 extern const struct super_operations proc_sops;
 
 void proc_init_kmemcache(void);
+void proc_invalidate_siblings_dcache(struct hlist_head *inodes, spinlock_t *lock);
 void set_proc_pid_nlink(void);
 extern struct inode *proc_get_inode(struct super_block *, struct proc_dir_entry *);
 extern void proc_entry_rundown(struct proc_dir_entry *);
diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c
index e2ed8e08cc7a..8ba492d44e68 100644
--- a/fs/proc/kcore.c
+++ b/fs/proc/kcore.c
@@ -574,11 +574,11 @@ static int release_kcore(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static const struct file_operations proc_kcore_operations = {
-	.read		= read_kcore,
-	.open		= open_kcore,
-	.release	= release_kcore,
-	.llseek		= default_llseek,
+static const struct proc_ops kcore_proc_ops = {
+	.proc_read	= read_kcore,
+	.proc_open	= open_kcore,
+	.proc_release	= release_kcore,
+	.proc_lseek	= default_llseek,
 };
 
 /* just remember that we have to update kcore */
@@ -637,8 +637,7 @@ static void __init add_modules_range(void)
 
 static int __init proc_kcore_init(void)
 {
-	proc_root_kcore = proc_create("kcore", S_IRUSR, NULL,
-				      &proc_kcore_operations);
+	proc_root_kcore = proc_create("kcore", S_IRUSR, NULL, &kcore_proc_ops);
 	if (!proc_root_kcore) {
 		pr_err("couldn't create /proc/kcore\n");
 		return 0; /* Always returns 0. */
diff --git a/fs/proc/kmsg.c b/fs/proc/kmsg.c
index 4f4a2abb225e..b38ad552887f 100644
--- a/fs/proc/kmsg.c
+++ b/fs/proc/kmsg.c
@@ -49,17 +49,18 @@ static __poll_t kmsg_poll(struct file *file, poll_table *wait)
 }
 
 
-static const struct file_operations proc_kmsg_operations = {
-	.read		= kmsg_read,
-	.poll		= kmsg_poll,
-	.open		= kmsg_open,
-	.release	= kmsg_release,
-	.llseek		= generic_file_llseek,
+static const struct proc_ops kmsg_proc_ops = {
+	.proc_flags	= PROC_ENTRY_PERMANENT,
+	.proc_read	= kmsg_read,
+	.proc_poll	= kmsg_poll,
+	.proc_open	= kmsg_open,
+	.proc_release	= kmsg_release,
+	.proc_lseek	= generic_file_llseek,
 };
 
 static int __init proc_kmsg_init(void)
 {
-	proc_create("kmsg", S_IRUSR, NULL, &proc_kmsg_operations);
+	proc_create("kmsg", S_IRUSR, NULL, &kmsg_proc_ops);
 	return 0;
 }
 fs_initcall(proc_kmsg_init);
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index 8c1f1bb1a5ce..ecc63ce01be7 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -103,11 +103,14 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
 	show_val_kb(m, "SUnreclaim:     ", sunreclaim);
 	seq_printf(m, "KernelStack:    %8lu kB\n",
 		   global_zone_page_state(NR_KERNEL_STACK_KB));
+#ifdef CONFIG_SHADOW_CALL_STACK
+	seq_printf(m, "ShadowCallStack:%8lu kB\n",
+		   global_zone_page_state(NR_KERNEL_SCS_KB));
+#endif
 	show_val_kb(m, "PageTables:     ",
 		    global_zone_page_state(NR_PAGETABLE));
 
-	show_val_kb(m, "NFS_Unstable:   ",
-		    global_node_page_state(NR_UNSTABLE_NFS));
+	show_val_kb(m, "NFS_Unstable:   ", 0);
 	show_val_kb(m, "Bounce:         ",
 		    global_zone_page_state(NR_BOUNCE));
 	show_val_kb(m, "WritebackTmp:   ",
diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c
index dd2b35f78b09..8e159fc78c0a 100644
--- a/fs/proc/namespaces.c
+++ b/fs/proc/namespaces.c
@@ -33,6 +33,10 @@ static const struct proc_ns_operations *ns_entries[] = {
 #ifdef CONFIG_CGROUPS
 	&cgroupns_operations,
 #endif
+#ifdef CONFIG_TIME_NS
+	&timens_operations,
+	&timens_for_children_operations,
+#endif
 };
 
 static const char *proc_ns_get_link(struct dentry *dentry,
@@ -42,22 +46,26 @@ static const char *proc_ns_get_link(struct dentry *dentry,
 	const struct proc_ns_operations *ns_ops = PROC_I(inode)->ns_ops;
 	struct task_struct *task;
 	struct path ns_path;
-	void *error = ERR_PTR(-EACCES);
+	int error = -EACCES;
 
 	if (!dentry)
 		return ERR_PTR(-ECHILD);
 
 	task = get_proc_task(inode);
 	if (!task)
-		return error;
+		return ERR_PTR(-EACCES);
 
-	if (ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)) {
-		error = ns_get_path(&ns_path, task, ns_ops);
-		if (!error)
-			nd_jump_link(&ns_path);
-	}
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS))
+		goto out;
+
+	error = ns_get_path(&ns_path, task, ns_ops);
+	if (error)
+		goto out;
+
+	error = nd_jump_link(&ns_path);
+out:
 	put_task_struct(task);
-	return error;
+	return ERR_PTR(error);
 }
 
 static int proc_ns_readlink(struct dentry *dentry, char __user *buffer, int buflen)
diff --git a/fs/proc/page.c b/fs/proc/page.c
index 7c952ee732e6..f909243d4a66 100644
--- a/fs/proc/page.c
+++ b/fs/proc/page.c
@@ -21,6 +21,21 @@
 #define KPMMASK (KPMSIZE - 1)
 #define KPMBITS (KPMSIZE * BITS_PER_BYTE)
 
+static inline unsigned long get_max_dump_pfn(void)
+{
+#ifdef CONFIG_SPARSEMEM
+	/*
+	 * The memmap of early sections is completely populated and marked
+	 * online even if max_pfn does not fall on a section boundary -
+	 * pfn_to_online_page() will succeed on all pages. Allow inspecting
+	 * these memmaps.
+	 */
+	return round_up(max_pfn, PAGES_PER_SECTION);
+#else
+	return max_pfn;
+#endif
+}
+
 /* /proc/kpagecount - an array exposing page counts
  *
  * Each entry is a u64 representing the corresponding
@@ -29,6 +44,7 @@
 static ssize_t kpagecount_read(struct file *file, char __user *buf,
 			     size_t count, loff_t *ppos)
 {
+	const unsigned long max_dump_pfn = get_max_dump_pfn();
 	u64 __user *out = (u64 __user *)buf;
 	struct page *ppage;
 	unsigned long src = *ppos;
@@ -37,9 +53,11 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
 	u64 pcount;
 
 	pfn = src / KPMSIZE;
-	count = min_t(size_t, count, (max_pfn * KPMSIZE) - src);
 	if (src & KPMMASK || count & KPMMASK)
 		return -EINVAL;
+	if (src >= max_dump_pfn * KPMSIZE)
+		return 0;
+	count = min_t(unsigned long, count, (max_dump_pfn * KPMSIZE) - src);
 
 	while (count > 0) {
 		/*
@@ -71,9 +89,9 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
 	return ret;
 }
 
-static const struct file_operations proc_kpagecount_operations = {
-	.llseek = mem_lseek,
-	.read = kpagecount_read,
+static const struct proc_ops kpagecount_proc_ops = {
+	.proc_lseek	= mem_lseek,
+	.proc_read	= kpagecount_read,
 };
 
 /* /proc/kpageflags - an array exposing page flags
@@ -206,6 +224,7 @@ u64 stable_page_flags(struct page *page)
 static ssize_t kpageflags_read(struct file *file, char __user *buf,
 			     size_t count, loff_t *ppos)
 {
+	const unsigned long max_dump_pfn = get_max_dump_pfn();
 	u64 __user *out = (u64 __user *)buf;
 	struct page *ppage;
 	unsigned long src = *ppos;
@@ -213,9 +232,11 @@ static ssize_t kpageflags_read(struct file *file, char __user *buf,
 	ssize_t ret = 0;
 
 	pfn = src / KPMSIZE;
-	count = min_t(unsigned long, count, (max_pfn * KPMSIZE) - src);
 	if (src & KPMMASK || count & KPMMASK)
 		return -EINVAL;
+	if (src >= max_dump_pfn * KPMSIZE)
+		return 0;
+	count = min_t(unsigned long, count, (max_dump_pfn * KPMSIZE) - src);
 
 	while (count > 0) {
 		/*
@@ -242,15 +263,16 @@ static ssize_t kpageflags_read(struct file *file, char __user *buf,
 	return ret;
 }
 
-static const struct file_operations proc_kpageflags_operations = {
-	.llseek = mem_lseek,
-	.read = kpageflags_read,
+static const struct proc_ops kpageflags_proc_ops = {
+	.proc_lseek	= mem_lseek,
+	.proc_read	= kpageflags_read,
 };
 
 #ifdef CONFIG_MEMCG
 static ssize_t kpagecgroup_read(struct file *file, char __user *buf,
 				size_t count, loff_t *ppos)
 {
+	const unsigned long max_dump_pfn = get_max_dump_pfn();
 	u64 __user *out = (u64 __user *)buf;
 	struct page *ppage;
 	unsigned long src = *ppos;
@@ -259,9 +281,11 @@ static ssize_t kpagecgroup_read(struct file *file, char __user *buf,
 	u64 ino;
 
 	pfn = src / KPMSIZE;
-	count = min_t(unsigned long, count, (max_pfn * KPMSIZE) - src);
 	if (src & KPMMASK || count & KPMMASK)
 		return -EINVAL;
+	if (src >= max_dump_pfn * KPMSIZE)
+		return 0;
+	count = min_t(unsigned long, count, (max_dump_pfn * KPMSIZE) - src);
 
 	while (count > 0) {
 		/*
@@ -293,18 +317,18 @@ static ssize_t kpagecgroup_read(struct file *file, char __user *buf,
 	return ret;
 }
 
-static const struct file_operations proc_kpagecgroup_operations = {
-	.llseek = mem_lseek,
-	.read = kpagecgroup_read,
+static const struct proc_ops kpagecgroup_proc_ops = {
+	.proc_lseek	= mem_lseek,
+	.proc_read	= kpagecgroup_read,
 };
 #endif /* CONFIG_MEMCG */
 
 static int __init proc_page_init(void)
 {
-	proc_create("kpagecount", S_IRUSR, NULL, &proc_kpagecount_operations);
-	proc_create("kpageflags", S_IRUSR, NULL, &proc_kpageflags_operations);
+	proc_create("kpagecount", S_IRUSR, NULL, &kpagecount_proc_ops);
+	proc_create("kpageflags", S_IRUSR, NULL, &kpageflags_proc_ops);
 #ifdef CONFIG_MEMCG
-	proc_create("kpagecgroup", S_IRUSR, NULL, &proc_kpagecgroup_operations);
+	proc_create("kpagecgroup", S_IRUSR, NULL, &kpagecgroup_proc_ops);
 #endif
 	return 0;
 }
diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index 76ae278df1c4..4888c5224442 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -90,12 +90,12 @@ static int seq_release_net(struct inode *ino, struct file *f)
 	return 0;
 }
 
-static const struct file_operations proc_net_seq_fops = {
-	.open		= seq_open_net,
-	.read		= seq_read,
-	.write		= proc_simple_write,
-	.llseek		= seq_lseek,
-	.release	= seq_release_net,
+static const struct proc_ops proc_net_seq_ops = {
+	.proc_open	= seq_open_net,
+	.proc_read	= seq_read,
+	.proc_write	= proc_simple_write,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= seq_release_net,
 };
 
 struct proc_dir_entry *proc_create_net_data(const char *name, umode_t mode,
@@ -108,7 +108,7 @@ struct proc_dir_entry *proc_create_net_data(const char *name, umode_t mode,
 	if (!p)
 		return NULL;
 	pde_force_lookup(p);
-	p->proc_fops = &proc_net_seq_fops;
+	p->proc_ops = &proc_net_seq_ops;
 	p->seq_ops = ops;
 	p->state_size = state_size;
 	return proc_register(parent, p);
@@ -152,7 +152,7 @@ struct proc_dir_entry *proc_create_net_data_write(const char *name, umode_t mode
 	if (!p)
 		return NULL;
 	pde_force_lookup(p);
-	p->proc_fops = &proc_net_seq_fops;
+	p->proc_ops = &proc_net_seq_ops;
 	p->seq_ops = ops;
 	p->state_size = state_size;
 	p->write = write;
@@ -183,12 +183,12 @@ static int single_release_net(struct inode *ino, struct file *f)
 	return single_release(ino, f);
 }
 
-static const struct file_operations proc_net_single_fops = {
-	.open		= single_open_net,
-	.read		= seq_read,
-	.write		= proc_simple_write,
-	.llseek		= seq_lseek,
-	.release	= single_release_net,
+static const struct proc_ops proc_net_single_ops = {
+	.proc_open	= single_open_net,
+	.proc_read	= seq_read,
+	.proc_write	= proc_simple_write,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release_net,
 };
 
 struct proc_dir_entry *proc_create_net_single(const char *name, umode_t mode,
@@ -201,7 +201,7 @@ struct proc_dir_entry *proc_create_net_single(const char *name, umode_t mode,
 	if (!p)
 		return NULL;
 	pde_force_lookup(p);
-	p->proc_fops = &proc_net_single_fops;
+	p->proc_ops = &proc_net_single_ops;
 	p->single_show = show;
 	return proc_register(parent, p);
 }
@@ -244,7 +244,7 @@ struct proc_dir_entry *proc_create_net_single_write(const char *name, umode_t mo
 	if (!p)
 		return NULL;
 	pde_force_lookup(p);
-	p->proc_fops = &proc_net_single_fops;
+	p->proc_ops = &proc_net_single_ops;
 	p->single_show = show;
 	p->write = write;
 	return proc_register(parent, p);
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index d80989b6c344..b6f5d459b087 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -267,42 +267,9 @@ static void unuse_table(struct ctl_table_header *p)
 			complete(p->unregistering);
 }
 
-static void proc_sys_prune_dcache(struct ctl_table_header *head)
+static void proc_sys_invalidate_dcache(struct ctl_table_header *head)
 {
-	struct inode *inode;
-	struct proc_inode *ei;
-	struct hlist_node *node;
-	struct super_block *sb;
-
-	rcu_read_lock();
-	for (;;) {
-		node = hlist_first_rcu(&head->inodes);
-		if (!node)
-			break;
-		ei = hlist_entry(node, struct proc_inode, sysctl_inodes);
-		spin_lock(&sysctl_lock);
-		hlist_del_init_rcu(&ei->sysctl_inodes);
-		spin_unlock(&sysctl_lock);
-
-		inode = &ei->vfs_inode;
-		sb = inode->i_sb;
-		if (!atomic_inc_not_zero(&sb->s_active))
-			continue;
-		inode = igrab(inode);
-		rcu_read_unlock();
-		if (unlikely(!inode)) {
-			deactivate_super(sb);
-			rcu_read_lock();
-			continue;
-		}
-
-		d_prune_aliases(inode);
-		iput(inode);
-		deactivate_super(sb);
-
-		rcu_read_lock();
-	}
-	rcu_read_unlock();
+	proc_invalidate_siblings_dcache(&head->inodes, &sysctl_lock);
 }
 
 /* called under sysctl_lock, will reacquire if has to wait */
@@ -324,10 +291,10 @@ static void start_unregistering(struct ctl_table_header *p)
 		spin_unlock(&sysctl_lock);
 	}
 	/*
-	 * Prune dentries for unregistered sysctls: namespaced sysctls
+	 * Invalidate dentries for unregistered sysctls: namespaced sysctls
 	 * can have duplicate names and contaminate dcache very badly.
 	 */
-	proc_sys_prune_dcache(p);
+	proc_sys_invalidate_dcache(p);
 	/*
 	 * do not remove from the list until nobody holds it; walking the
 	 * list in do_sysctl() relies on that.
@@ -483,7 +450,7 @@ static struct inode *proc_sys_make_inode(struct super_block *sb,
 	}
 	ei->sysctl = head;
 	ei->sysctl_entry = table;
-	hlist_add_head_rcu(&ei->sysctl_inodes, &head->inodes);
+	hlist_add_head_rcu(&ei->sibling_inodes, &head->inodes);
 	head->count++;
 	spin_unlock(&sysctl_lock);
 
@@ -514,7 +481,7 @@ static struct inode *proc_sys_make_inode(struct super_block *sb,
 void proc_sys_evict_inode(struct inode *inode, struct ctl_table_header *head)
 {
 	spin_lock(&sysctl_lock);
-	hlist_del_init_rcu(&PROC_I(inode)->sysctl_inodes);
+	hlist_del_init_rcu(&PROC_I(inode)->sibling_inodes);
 	if (!--head->count)
 		kfree_rcu(head, rcu);
 	spin_unlock(&sysctl_lock);
@@ -1720,7 +1687,7 @@ int __init proc_sys_init(void)
 
 	proc_sys_root = proc_mkdir("sys", NULL);
 	proc_sys_root->proc_iops = &proc_sys_dir_operations;
-	proc_sys_root->proc_fops = &proc_sys_dir_file_operations;
+	proc_sys_root->proc_dir_ops = &proc_sys_dir_file_operations;
 	proc_sys_root->nlink = 0;
 
 	return sysctl_init();
diff --git a/fs/proc/root.c b/fs/proc/root.c
index 0b7c8dffc9ae..cdbe9293ea55 100644
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -41,24 +41,19 @@ enum proc_param {
 	Opt_hidepid,
 };
 
-static const struct fs_parameter_spec proc_param_specs[] = {
+static const struct fs_parameter_spec proc_fs_parameters[] = {
 	fsparam_u32("gid",	Opt_gid),
 	fsparam_u32("hidepid",	Opt_hidepid),
 	{}
 };
 
-static const struct fs_parameter_description proc_fs_parameters = {
-	.name		= "proc",
-	.specs		= proc_param_specs,
-};
-
 static int proc_parse_param(struct fs_context *fc, struct fs_parameter *param)
 {
 	struct proc_fs_context *ctx = fc->fs_private;
 	struct fs_parse_result result;
 	int opt;
 
-	opt = fs_parse(fc, &proc_fs_parameters, param, &result);
+	opt = fs_parse(fc, proc_fs_parameters, param, &result);
 	if (opt < 0)
 		return opt;
 
@@ -71,7 +66,7 @@ static int proc_parse_param(struct fs_context *fc, struct fs_parameter *param)
 		ctx->hidepid = result.uint_32;
 		if (ctx->hidepid < HIDEPID_OFF ||
 		    ctx->hidepid > HIDEPID_INVISIBLE)
-			return invalf(fc, "proc: hidepid value must be between 0 and 2.\n");
+			return invalfc(fc, "hidepid value must be between 0 and 2.\n");
 		break;
 
 	default:
@@ -201,13 +196,20 @@ static void proc_kill_sb(struct super_block *sb)
 	if (ns->proc_thread_self)
 		dput(ns->proc_thread_self);
 	kill_anon_super(sb);
+
+	/* Make the pid namespace safe for the next mount of proc */
+	ns->proc_self = NULL;
+	ns->proc_thread_self = NULL;
+	ns->pid_gid = GLOBAL_ROOT_GID;
+	ns->hide_pid = 0;
+
 	put_pid_ns(ns);
 }
 
 static struct file_system_type proc_fs_type = {
 	.name			= "proc",
 	.init_fs_context	= proc_init_fs_context,
-	.parameters		= &proc_fs_parameters,
+	.parameters		= proc_fs_parameters,
 	.kill_sb		= proc_kill_sb,
 	.fs_flags		= FS_USERNS_MOUNT | FS_DISALLOW_NOTIFY_PERM,
 };
@@ -292,44 +294,8 @@ struct proc_dir_entry proc_root = {
 	.nlink		= 2, 
 	.refcnt		= REFCOUNT_INIT(1),
 	.proc_iops	= &proc_root_inode_operations, 
-	.proc_fops	= &proc_root_operations,
+	.proc_dir_ops	= &proc_root_operations,
 	.parent		= &proc_root,
 	.subdir		= RB_ROOT,
 	.name		= "/proc",
 };
-
-int pid_ns_prepare_proc(struct pid_namespace *ns)
-{
-	struct proc_fs_context *ctx;
-	struct fs_context *fc;
-	struct vfsmount *mnt;
-
-	fc = fs_context_for_mount(&proc_fs_type, SB_KERNMOUNT);
-	if (IS_ERR(fc))
-		return PTR_ERR(fc);
-
-	if (fc->user_ns != ns->user_ns) {
-		put_user_ns(fc->user_ns);
-		fc->user_ns = get_user_ns(ns->user_ns);
-	}
-
-	ctx = fc->fs_private;
-	if (ctx->pid_ns != ns) {
-		put_pid_ns(ctx->pid_ns);
-		get_pid_ns(ns);
-		ctx->pid_ns = ns;
-	}
-
-	mnt = fc_mount(fc);
-	put_fs_context(fc);
-	if (IS_ERR(mnt))
-		return PTR_ERR(mnt);
-
-	ns->proc_mnt = mnt;
-	return 0;
-}
-
-void pid_ns_release_proc(struct pid_namespace *ns)
-{
-	kern_unmount(ns->proc_mnt);
-}
diff --git a/fs/proc/stat.c b/fs/proc/stat.c
index 80c305f206bb..46b3293015fe 100644
--- a/fs/proc/stat.c
+++ b/fs/proc/stat.c
@@ -120,20 +120,23 @@ static int show_stat(struct seq_file *p, void *v)
 	getboottime64(&boottime);
 
 	for_each_possible_cpu(i) {
-		struct kernel_cpustat *kcs = &kcpustat_cpu(i);
-
-		user += kcs->cpustat[CPUTIME_USER];
-		nice += kcs->cpustat[CPUTIME_NICE];
-		system += kcs->cpustat[CPUTIME_SYSTEM];
-		idle += get_idle_time(kcs, i);
-		iowait += get_iowait_time(kcs, i);
-		irq += kcs->cpustat[CPUTIME_IRQ];
-		softirq += kcs->cpustat[CPUTIME_SOFTIRQ];
-		steal += kcs->cpustat[CPUTIME_STEAL];
-		guest += kcs->cpustat[CPUTIME_GUEST];
-		guest_nice += kcs->cpustat[CPUTIME_GUEST_NICE];
-		sum += kstat_cpu_irqs_sum(i);
-		sum += arch_irq_stat_cpu(i);
+		struct kernel_cpustat kcpustat;
+		u64 *cpustat = kcpustat.cpustat;
+
+		kcpustat_cpu_fetch(&kcpustat, i);
+
+		user		+= cpustat[CPUTIME_USER];
+		nice		+= cpustat[CPUTIME_NICE];
+		system		+= cpustat[CPUTIME_SYSTEM];
+		idle		+= get_idle_time(&kcpustat, i);
+		iowait		+= get_iowait_time(&kcpustat, i);
+		irq		+= cpustat[CPUTIME_IRQ];
+		softirq		+= cpustat[CPUTIME_SOFTIRQ];
+		steal		+= cpustat[CPUTIME_STEAL];
+		guest		+= cpustat[CPUTIME_GUEST];
+		guest_nice	+= cpustat[CPUTIME_GUEST_NICE];
+		sum		+= kstat_cpu_irqs_sum(i);
+		sum		+= arch_irq_stat_cpu(i);
 
 		for (j = 0; j < NR_SOFTIRQS; j++) {
 			unsigned int softirq_stat = kstat_softirqs_cpu(j, i);
@@ -157,19 +160,22 @@ static int show_stat(struct seq_file *p, void *v)
 	seq_putc(p, '\n');
 
 	for_each_online_cpu(i) {
-		struct kernel_cpustat *kcs = &kcpustat_cpu(i);
+		struct kernel_cpustat kcpustat;
+		u64 *cpustat = kcpustat.cpustat;
+
+		kcpustat_cpu_fetch(&kcpustat, i);
 
 		/* Copy values here to work around gcc-2.95.3, gcc-2.96 */
-		user = kcs->cpustat[CPUTIME_USER];
-		nice = kcs->cpustat[CPUTIME_NICE];
-		system = kcs->cpustat[CPUTIME_SYSTEM];
-		idle = get_idle_time(kcs, i);
-		iowait = get_iowait_time(kcs, i);
-		irq = kcs->cpustat[CPUTIME_IRQ];
-		softirq = kcs->cpustat[CPUTIME_SOFTIRQ];
-		steal = kcs->cpustat[CPUTIME_STEAL];
-		guest = kcs->cpustat[CPUTIME_GUEST];
-		guest_nice = kcs->cpustat[CPUTIME_GUEST_NICE];
+		user		= cpustat[CPUTIME_USER];
+		nice		= cpustat[CPUTIME_NICE];
+		system		= cpustat[CPUTIME_SYSTEM];
+		idle		= get_idle_time(&kcpustat, i);
+		iowait		= get_iowait_time(&kcpustat, i);
+		irq		= cpustat[CPUTIME_IRQ];
+		softirq		= cpustat[CPUTIME_SOFTIRQ];
+		steal		= cpustat[CPUTIME_STEAL];
+		guest		= cpustat[CPUTIME_GUEST];
+		guest_nice	= cpustat[CPUTIME_GUEST_NICE];
 		seq_printf(p, "cpu%d", i);
 		seq_put_decimal_ull(p, " ", nsec_to_clock_t(user));
 		seq_put_decimal_ull(p, " ", nsec_to_clock_t(nice));
@@ -217,16 +223,17 @@ static int stat_open(struct inode *inode, struct file *file)
 	return single_open_size(file, show_stat, NULL, size);
 }
 
-static const struct file_operations proc_stat_operations = {
-	.open		= stat_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
+static const struct proc_ops stat_proc_ops = {
+	.proc_flags	= PROC_ENTRY_PERMANENT,
+	.proc_open	= stat_open,
+	.proc_read	= seq_read,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
 };
 
 static int __init proc_stat_init(void)
 {
-	proc_create("stat", 0, NULL, &proc_stat_operations);
+	proc_create("stat", 0, NULL, &stat_proc_ops);
 	return 0;
 }
 fs_initcall(proc_stat_init);
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 9442631fd4af..6ad407d5efe2 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -123,38 +123,14 @@ static void release_task_mempolicy(struct proc_maps_private *priv)
 }
 #endif
 
-static void vma_stop(struct proc_maps_private *priv)
-{
-	struct mm_struct *mm = priv->mm;
-
-	release_task_mempolicy(priv);
-	up_read(&mm->mmap_sem);
-	mmput(mm);
-}
-
-static struct vm_area_struct *
-m_next_vma(struct proc_maps_private *priv, struct vm_area_struct *vma)
-{
-	if (vma == priv->tail_vma)
-		return NULL;
-	return vma->vm_next ?: priv->tail_vma;
-}
-
-static void m_cache_vma(struct seq_file *m, struct vm_area_struct *vma)
-{
-	if (m->count < m->size)	/* vma is copied successfully */
-		m->version = m_next_vma(m->private, vma) ? vma->vm_end : -1UL;
-}
-
 static void *m_start(struct seq_file *m, loff_t *ppos)
 {
 	struct proc_maps_private *priv = m->private;
-	unsigned long last_addr = m->version;
+	unsigned long last_addr = *ppos;
 	struct mm_struct *mm;
 	struct vm_area_struct *vma;
-	unsigned int pos = *ppos;
 
-	/* See m_cache_vma(). Zero at the start or after lseek. */
+	/* See m_next(). Zero at the start or after lseek. */
 	if (last_addr == -1UL)
 		return NULL;
 
@@ -163,64 +139,59 @@ static void *m_start(struct seq_file *m, loff_t *ppos)
 		return ERR_PTR(-ESRCH);
 
 	mm = priv->mm;
-	if (!mm || !mmget_not_zero(mm))
+	if (!mm || !mmget_not_zero(mm)) {
+		put_task_struct(priv->task);
+		priv->task = NULL;
 		return NULL;
+	}
 
 	if (down_read_killable(&mm->mmap_sem)) {
 		mmput(mm);
+		put_task_struct(priv->task);
+		priv->task = NULL;
 		return ERR_PTR(-EINTR);
 	}
 
 	hold_task_mempolicy(priv);
 	priv->tail_vma = get_gate_vma(mm);
 
-	if (last_addr) {
-		vma = find_vma(mm, last_addr - 1);
-		if (vma && vma->vm_start <= last_addr)
-			vma = m_next_vma(priv, vma);
-		if (vma)
-			return vma;
-	}
-
-	m->version = 0;
-	if (pos < mm->map_count) {
-		for (vma = mm->mmap; pos; pos--) {
-			m->version = vma->vm_start;
-			vma = vma->vm_next;
-		}
+	vma = find_vma(mm, last_addr);
+	if (vma)
 		return vma;
-	}
 
-	/* we do not bother to update m->version in this case */
-	if (pos == mm->map_count && priv->tail_vma)
-		return priv->tail_vma;
-
-	vma_stop(priv);
-	return NULL;
+	return priv->tail_vma;
 }
 
-static void *m_next(struct seq_file *m, void *v, loff_t *pos)
+static void *m_next(struct seq_file *m, void *v, loff_t *ppos)
 {
 	struct proc_maps_private *priv = m->private;
-	struct vm_area_struct *next;
+	struct vm_area_struct *next, *vma = v;
+
+	if (vma == priv->tail_vma)
+		next = NULL;
+	else if (vma->vm_next)
+		next = vma->vm_next;
+	else
+		next = priv->tail_vma;
+
+	*ppos = next ? next->vm_start : -1UL;
 
-	(*pos)++;
-	next = m_next_vma(priv, v);
-	if (!next)
-		vma_stop(priv);
 	return next;
 }
 
 static void m_stop(struct seq_file *m, void *v)
 {
 	struct proc_maps_private *priv = m->private;
+	struct mm_struct *mm = priv->mm;
 
-	if (!IS_ERR_OR_NULL(v))
-		vma_stop(priv);
-	if (priv->task) {
-		put_task_struct(priv->task);
-		priv->task = NULL;
-	}
+	if (!priv->task)
+		return;
+
+	release_task_mempolicy(priv);
+	up_read(&mm->mmap_sem);
+	mmput(mm);
+	put_task_struct(priv->task);
+	priv->task = NULL;
 }
 
 static int proc_maps_open(struct inode *inode, struct file *file,
@@ -363,7 +334,6 @@ done:
 static int show_map(struct seq_file *m, void *v)
 {
 	show_map_vma(m, v);
-	m_cache_vma(m, v);
 	return 0;
 }
 
@@ -505,7 +475,7 @@ static void smaps_account(struct mem_size_stats *mss, struct page *page,
 
 #ifdef CONFIG_SHMEM
 static int smaps_pte_hole(unsigned long addr, unsigned long end,
-		struct mm_walk *walk)
+			  __always_unused int depth, struct mm_walk *walk)
 {
 	struct mem_size_stats *mss = walk->private;
 
@@ -576,10 +546,17 @@ static void smaps_pmd_entry(pmd_t *pmd, unsigned long addr,
 	struct mem_size_stats *mss = walk->private;
 	struct vm_area_struct *vma = walk->vma;
 	bool locked = !!(vma->vm_flags & VM_LOCKED);
-	struct page *page;
+	struct page *page = NULL;
+
+	if (pmd_present(*pmd)) {
+		/* FOLL_DUMP will return -EFAULT on huge zero page */
+		page = follow_trans_huge_pmd(vma, addr, pmd, FOLL_DUMP);
+	} else if (unlikely(thp_migration_supported() && is_swap_pmd(*pmd))) {
+		swp_entry_t entry = pmd_to_swp_entry(*pmd);
 
-	/* FOLL_DUMP will return -EFAULT on huge zero page */
-	page = follow_trans_huge_pmd(vma, addr, pmd, FOLL_DUMP);
+		if (is_migration_entry(entry))
+			page = migration_entry_to_page(entry);
+	}
 	if (IS_ERR_OR_NULL(page))
 		return;
 	if (PageAnon(page))
@@ -608,8 +585,7 @@ static int smaps_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
 
 	ptl = pmd_trans_huge_lock(pmd, vma);
 	if (ptl) {
-		if (pmd_present(*pmd))
-			smaps_pmd_entry(pmd, addr, walk);
+		smaps_pmd_entry(pmd, addr, walk);
 		spin_unlock(ptl);
 		goto out;
 	}
@@ -652,9 +628,6 @@ static void show_smap_vma_flags(struct seq_file *m, struct vm_area_struct *vma)
 		[ilog2(VM_GROWSDOWN)]	= "gd",
 		[ilog2(VM_PFNMAP)]	= "pf",
 		[ilog2(VM_DENYWRITE)]	= "dw",
-#ifdef CONFIG_X86_INTEL_MPX
-		[ilog2(VM_MPX)]		= "mp",
-#endif
 		[ilog2(VM_LOCKED)]	= "lo",
 		[ilog2(VM_IO)]		= "io",
 		[ilog2(VM_SEQ_READ)]	= "sr",
@@ -668,6 +641,9 @@ static void show_smap_vma_flags(struct seq_file *m, struct vm_area_struct *vma)
 		[ilog2(VM_ARCH_1)]	= "ar",
 		[ilog2(VM_WIPEONFORK)]	= "wf",
 		[ilog2(VM_DONTDUMP)]	= "dd",
+#ifdef CONFIG_ARM64_BTI
+		[ilog2(VM_ARM64_BTI)]	= "bt",
+#endif
 #ifdef CONFIG_MEM_SOFT_DIRTY
 		[ilog2(VM_SOFTDIRTY)]	= "sd",
 #endif
@@ -847,8 +823,6 @@ static int show_smap(struct seq_file *m, void *v)
 		seq_printf(m, "ProtectionKey:  %8u\n", vma_pkey(vma));
 	show_smap_vma_flags(m, vma);
 
-	m_cache_vma(m, vma);
-
 	return 0;
 }
 
@@ -1282,7 +1256,7 @@ static int add_to_pagemap(unsigned long addr, pagemap_entry_t *pme,
 }
 
 static int pagemap_pte_hole(unsigned long start, unsigned long end,
-				struct mm_walk *walk)
+			    __always_unused int depth, struct mm_walk *walk)
 {
 	struct pagemapread *pm = walk->private;
 	unsigned long addr = start;
@@ -1887,7 +1861,6 @@ static int show_numa_map(struct seq_file *m, void *v)
 	seq_printf(m, " kernelpagesize_kB=%lu", vma_kernel_pagesize(vma) >> 10);
 out:
 	seq_putc(m, '\n');
-	m_cache_vma(m, vma);
 	return 0;
 }
 
diff --git a/fs/proc/uptime.c b/fs/proc/uptime.c
index a4c2791ab70b..5a1b228964fb 100644
--- a/fs/proc/uptime.c
+++ b/fs/proc/uptime.c
@@ -5,6 +5,7 @@
 #include <linux/sched.h>
 #include <linux/seq_file.h>
 #include <linux/time.h>
+#include <linux/time_namespace.h>
 #include <linux/kernel_stat.h>
 
 static int uptime_proc_show(struct seq_file *m, void *v)
@@ -20,6 +21,8 @@ static int uptime_proc_show(struct seq_file *m, void *v)
 		nsec += (__force u64) kcpustat_cpu(i).cpustat[CPUTIME_IDLE];
 
 	ktime_get_boottime_ts64(&uptime);
+	timens_add_boottime(&uptime);
+
 	idle.tv_sec = div_u64_rem(nsec, NSEC_PER_SEC, &rem);
 	idle.tv_nsec = rem;
 	seq_printf(m, "%lu.%02lu %lu.%02lu\n",
diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
index 7b13988796e1..c663202da8de 100644
--- a/fs/proc/vmcore.c
+++ b/fs/proc/vmcore.c
@@ -266,7 +266,8 @@ static int vmcoredd_mmap_dumps(struct vm_area_struct *vma, unsigned long dst,
 		if (start < offset + dump->size) {
 			tsz = min(offset + (u64)dump->size - start, (u64)size);
 			buf = dump->buf + start - offset;
-			if (remap_vmalloc_range_partial(vma, dst, buf, tsz)) {
+			if (remap_vmalloc_range_partial(vma, dst, buf, 0,
+							tsz)) {
 				ret = -EFAULT;
 				goto out_unlock;
 			}
@@ -624,7 +625,7 @@ static int mmap_vmcore(struct file *file, struct vm_area_struct *vma)
 		tsz = min(elfcorebuf_sz + elfnotes_sz - (size_t)start, size);
 		kaddr = elfnotes_buf + start - elfcorebuf_sz - vmcoredd_orig_sz;
 		if (remap_vmalloc_range_partial(vma, vma->vm_start + len,
-						kaddr, tsz))
+						kaddr, 0, tsz))
 			goto fail;
 
 		size -= tsz;
@@ -667,10 +668,10 @@ static int mmap_vmcore(struct file *file, struct vm_area_struct *vma)
 }
 #endif
 
-static const struct file_operations proc_vmcore_operations = {
-	.read		= read_vmcore,
-	.llseek		= default_llseek,
-	.mmap		= mmap_vmcore,
+static const struct proc_ops vmcore_proc_ops = {
+	.proc_read	= read_vmcore,
+	.proc_lseek	= default_llseek,
+	.proc_mmap	= mmap_vmcore,
 };
 
 static struct vmcore* __init get_new_element(void)
@@ -1555,7 +1556,7 @@ static int __init vmcore_init(void)
 	elfcorehdr_free(elfcorehdr_addr);
 	elfcorehdr_addr = ELFCORE_ADDR_ERR;
 
-	proc_vmcore = proc_create("vmcore", S_IRUSR, NULL, &proc_vmcore_operations);
+	proc_vmcore = proc_create("vmcore", S_IRUSR, NULL, &vmcore_proc_ops);
 	if (proc_vmcore)
 		proc_vmcore->size = vmcore_size;
 	return 0;
diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c
index 273ee82d8aa9..e4d70c0dffe9 100644
--- a/fs/proc_namespace.c
+++ b/fs/proc_namespace.c
@@ -279,7 +279,8 @@ static int mounts_open_common(struct inode *inode, struct file *file,
 	p->ns = ns;
 	p->root = root;
 	p->show = show;
-	p->cached_event = ~0ULL;
+	INIT_LIST_HEAD(&p->cursor.mnt_list);
+	p->cursor.mnt.mnt_flags = MNT_CURSOR;
 
 	return 0;
 
@@ -296,6 +297,7 @@ static int mounts_release(struct inode *inode, struct file *file)
 	struct seq_file *m = file->private_data;
 	struct proc_mounts *p = m->private;
 	path_put(&p->root);
+	mnt_cursor_del(p->ns, &p->cursor);
 	put_mnt_ns(p->ns);
 	return seq_release_private(inode, file);
 }
diff --git a/fs/pstore/Kconfig b/fs/pstore/Kconfig
index 8f0369aad22a..e16a49ebfe54 100644
--- a/fs/pstore/Kconfig
+++ b/fs/pstore/Kconfig
@@ -153,3 +153,112 @@ config PSTORE_RAM
 	  "ramoops.ko".
 
 	  For more information, see Documentation/admin-guide/ramoops.rst.
+
+config PSTORE_ZONE
+	tristate
+	depends on PSTORE
+	help
+	  The common layer for pstore/blk (and pstore/ram in the future)
+	  to manage storage in zones.
+
+config PSTORE_BLK
+	tristate "Log panic/oops to a block device"
+	depends on PSTORE
+	depends on BLOCK
+	select PSTORE_ZONE
+	default n
+	help
+	  This enables panic and oops message to be logged to a block dev
+	  where it can be read back at some later point.
+
+	  For more information, see Documentation/admin-guide/pstore-blk.rst
+
+	  If unsure, say N.
+
+config PSTORE_BLK_BLKDEV
+	string "block device identifier"
+	depends on PSTORE_BLK
+	default ""
+	help
+	  Which block device should be used for pstore/blk.
+
+	  It accepts the following variants:
+	  1) <hex_major><hex_minor> device number in hexadecimal representation,
+	     with no leading 0x, for example b302.
+	  2) /dev/<disk_name> represents the device name of disk
+	  3) /dev/<disk_name><decimal> represents the device name and number
+	     of partition - device number of disk plus the partition number
+	  4) /dev/<disk_name>p<decimal> - same as the above, this form is
+	     used when disk name of partitioned disk ends with a digit.
+	  5) PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF representing the
+	     unique id of a partition if the partition table provides it.
+	     The UUID may be either an EFI/GPT UUID, or refer to an MSDOS
+	     partition using the format SSSSSSSS-PP, where SSSSSSSS is a zero-
+	     filled hex representation of the 32-bit "NT disk signature", and PP
+	     is a zero-filled hex representation of the 1-based partition number.
+	  6) PARTUUID=<UUID>/PARTNROFF=<int> to select a partition in relation
+	     to a partition with a known unique id.
+	  7) <major>:<minor> major and minor number of the device separated by
+	     a colon.
+
+	  NOTE that, both Kconfig and module parameters can configure
+	  pstore/blk, but module parameters have priority over Kconfig.
+
+config PSTORE_BLK_KMSG_SIZE
+	int "Size in Kbytes of kmsg dump log to store"
+	depends on PSTORE_BLK
+	default 64
+	help
+	  This just sets size of kmsg dump (oops, panic, etc) log for
+	  pstore/blk. The size is in KB and must be a multiple of 4.
+
+	  NOTE that, both Kconfig and module parameters can configure
+	  pstore/blk, but module parameters have priority over Kconfig.
+
+config PSTORE_BLK_MAX_REASON
+	int "Maximum kmsg dump reason to store"
+	depends on PSTORE_BLK
+	default 2
+	help
+	  The maximum reason for kmsg dumps to store. The default is
+	  2 (KMSG_DUMP_OOPS), see include/linux/kmsg_dump.h's
+	  enum kmsg_dump_reason for more details.
+
+	  NOTE that, both Kconfig and module parameters can configure
+	  pstore/blk, but module parameters have priority over Kconfig.
+
+config PSTORE_BLK_PMSG_SIZE
+	int "Size in Kbytes of pmsg to store"
+	depends on PSTORE_BLK
+	depends on PSTORE_PMSG
+	default 64
+	help
+	  This just sets size of pmsg (pmsg_size) for pstore/blk. The size is
+	  in KB and must be a multiple of 4.
+
+	  NOTE that, both Kconfig and module parameters can configure
+	  pstore/blk, but module parameters have priority over Kconfig.
+
+config PSTORE_BLK_CONSOLE_SIZE
+	int "Size in Kbytes of console log to store"
+	depends on PSTORE_BLK
+	depends on PSTORE_CONSOLE
+	default 64
+	help
+	  This just sets size of console log (console_size) to store via
+	  pstore/blk. The size is in KB and must be a multiple of 4.
+
+	  NOTE that, both Kconfig and module parameters can configure
+	  pstore/blk, but module parameters have priority over Kconfig.
+
+config PSTORE_BLK_FTRACE_SIZE
+	int "Size in Kbytes of ftrace log to store"
+	depends on PSTORE_BLK
+	depends on PSTORE_FTRACE
+	default 64
+	help
+	  This just sets size of ftrace log (ftrace_size) for pstore/blk. The
+	  size is in KB and must be a multiple of 4.
+
+	  NOTE that, both Kconfig and module parameters can configure
+	  pstore/blk, but module parameters have priority over Kconfig.
diff --git a/fs/pstore/Makefile b/fs/pstore/Makefile
index 967b5891f325..c270467aeece 100644
--- a/fs/pstore/Makefile
+++ b/fs/pstore/Makefile
@@ -12,3 +12,9 @@ pstore-$(CONFIG_PSTORE_PMSG)	+= pmsg.o
 
 ramoops-objs += ram.o ram_core.o
 obj-$(CONFIG_PSTORE_RAM)	+= ramoops.o
+
+pstore_zone-objs += zone.o
+obj-$(CONFIG_PSTORE_ZONE)	+= pstore_zone.o
+
+pstore_blk-objs += blk.o
+obj-$(CONFIG_PSTORE_BLK)	+= pstore_blk.o
diff --git a/fs/pstore/blk.c b/fs/pstore/blk.c
new file mode 100644
index 000000000000..fcd5563dde06
--- /dev/null
+++ b/fs/pstore/blk.c
@@ -0,0 +1,517 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Implements pstore backend driver that write to block (or non-block) storage
+ * devices, using the pstore/zone API.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include "../../block/blk.h"
+#include <linux/blkdev.h>
+#include <linux/string.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/platform_device.h>
+#include <linux/pstore_blk.h>
+#include <linux/mount.h>
+#include <linux/uio.h>
+
+static long kmsg_size = CONFIG_PSTORE_BLK_KMSG_SIZE;
+module_param(kmsg_size, long, 0400);
+MODULE_PARM_DESC(kmsg_size, "kmsg dump record size in kbytes");
+
+static int max_reason = CONFIG_PSTORE_BLK_MAX_REASON;
+module_param(max_reason, int, 0400);
+MODULE_PARM_DESC(max_reason,
+		 "maximum reason for kmsg dump (default 2: Oops and Panic)");
+
+#if IS_ENABLED(CONFIG_PSTORE_PMSG)
+static long pmsg_size = CONFIG_PSTORE_BLK_PMSG_SIZE;
+#else
+static long pmsg_size = -1;
+#endif
+module_param(pmsg_size, long, 0400);
+MODULE_PARM_DESC(pmsg_size, "pmsg size in kbytes");
+
+#if IS_ENABLED(CONFIG_PSTORE_CONSOLE)
+static long console_size = CONFIG_PSTORE_BLK_CONSOLE_SIZE;
+#else
+static long console_size = -1;
+#endif
+module_param(console_size, long, 0400);
+MODULE_PARM_DESC(console_size, "console size in kbytes");
+
+#if IS_ENABLED(CONFIG_PSTORE_FTRACE)
+static long ftrace_size = CONFIG_PSTORE_BLK_FTRACE_SIZE;
+#else
+static long ftrace_size = -1;
+#endif
+module_param(ftrace_size, long, 0400);
+MODULE_PARM_DESC(ftrace_size, "ftrace size in kbytes");
+
+static bool best_effort;
+module_param(best_effort, bool, 0400);
+MODULE_PARM_DESC(best_effort, "use best effort to write (i.e. do not require storage driver pstore support, default: off)");
+
+/*
+ * blkdev - the block device to use for pstore storage
+ *
+ * Usually, this will be a partition of a block device.
+ *
+ * blkdev accepts the following variants:
+ * 1) <hex_major><hex_minor> device number in hexadecimal representation,
+ *    with no leading 0x, for example b302.
+ * 2) /dev/<disk_name> represents the device number of disk
+ * 3) /dev/<disk_name><decimal> represents the device number
+ *    of partition - device number of disk plus the partition number
+ * 4) /dev/<disk_name>p<decimal> - same as the above, that form is
+ *    used when disk name of partitioned disk ends on a digit.
+ * 5) PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF representing the
+ *    unique id of a partition if the partition table provides it.
+ *    The UUID may be either an EFI/GPT UUID, or refer to an MSDOS
+ *    partition using the format SSSSSSSS-PP, where SSSSSSSS is a zero-
+ *    filled hex representation of the 32-bit "NT disk signature", and PP
+ *    is a zero-filled hex representation of the 1-based partition number.
+ * 6) PARTUUID=<UUID>/PARTNROFF=<int> to select a partition in relation to
+ *    a partition with a known unique id.
+ * 7) <major>:<minor> major and minor number of the device separated by
+ *    a colon.
+ */
+static char blkdev[80] = CONFIG_PSTORE_BLK_BLKDEV;
+module_param_string(blkdev, blkdev, 80, 0400);
+MODULE_PARM_DESC(blkdev, "block device for pstore storage");
+
+/*
+ * All globals must only be accessed under the pstore_blk_lock
+ * during the register/unregister functions.
+ */
+static DEFINE_MUTEX(pstore_blk_lock);
+static struct block_device *psblk_bdev;
+static struct pstore_zone_info *pstore_zone_info;
+static pstore_blk_panic_write_op blkdev_panic_write;
+
+struct bdev_info {
+	dev_t devt;
+	sector_t nr_sects;
+	sector_t start_sect;
+};
+
+#define check_size(name, alignsize) ({				\
+	long _##name_ = (name);					\
+	_##name_ = _##name_ <= 0 ? 0 : (_##name_ * 1024);	\
+	if (_##name_ & ((alignsize) - 1)) {			\
+		pr_info(#name " must align to %d\n",		\
+				(alignsize));			\
+		_##name_ = ALIGN(name, (alignsize));		\
+	}							\
+	_##name_;						\
+})
+
+static int __register_pstore_device(struct pstore_device_info *dev)
+{
+	int ret;
+
+	lockdep_assert_held(&pstore_blk_lock);
+
+	if (!dev || !dev->total_size || !dev->read || !dev->write)
+		return -EINVAL;
+
+	/* someone already registered before */
+	if (pstore_zone_info)
+		return -EBUSY;
+
+	pstore_zone_info = kzalloc(sizeof(struct pstore_zone_info), GFP_KERNEL);
+	if (!pstore_zone_info)
+		return -ENOMEM;
+
+	/* zero means not limit on which backends to attempt to store. */
+	if (!dev->flags)
+		dev->flags = UINT_MAX;
+
+#define verify_size(name, alignsize, enabled) {				\
+		long _##name_;						\
+		if (enabled)						\
+			_##name_ = check_size(name, alignsize);		\
+		else							\
+			_##name_ = 0;					\
+		name = _##name_ / 1024;					\
+		pstore_zone_info->name = _##name_;			\
+	}
+
+	verify_size(kmsg_size, 4096, dev->flags & PSTORE_FLAGS_DMESG);
+	verify_size(pmsg_size, 4096, dev->flags & PSTORE_FLAGS_PMSG);
+	verify_size(console_size, 4096, dev->flags & PSTORE_FLAGS_CONSOLE);
+	verify_size(ftrace_size, 4096, dev->flags & PSTORE_FLAGS_FTRACE);
+#undef verify_size
+
+	pstore_zone_info->total_size = dev->total_size;
+	pstore_zone_info->max_reason = max_reason;
+	pstore_zone_info->read = dev->read;
+	pstore_zone_info->write = dev->write;
+	pstore_zone_info->erase = dev->erase;
+	pstore_zone_info->panic_write = dev->panic_write;
+	pstore_zone_info->name = KBUILD_MODNAME;
+	pstore_zone_info->owner = THIS_MODULE;
+
+	ret = register_pstore_zone(pstore_zone_info);
+	if (ret) {
+		kfree(pstore_zone_info);
+		pstore_zone_info = NULL;
+	}
+	return ret;
+}
+/**
+ * register_pstore_device() - register non-block device to pstore/blk
+ *
+ * @dev: non-block device information
+ *
+ * Return:
+ * * 0		- OK
+ * * Others	- something error.
+ */
+int register_pstore_device(struct pstore_device_info *dev)
+{
+	int ret;
+
+	mutex_lock(&pstore_blk_lock);
+	ret = __register_pstore_device(dev);
+	mutex_unlock(&pstore_blk_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(register_pstore_device);
+
+static void __unregister_pstore_device(struct pstore_device_info *dev)
+{
+	lockdep_assert_held(&pstore_blk_lock);
+	if (pstore_zone_info && pstore_zone_info->read == dev->read) {
+		unregister_pstore_zone(pstore_zone_info);
+		kfree(pstore_zone_info);
+		pstore_zone_info = NULL;
+	}
+}
+
+/**
+ * unregister_pstore_device() - unregister non-block device from pstore/blk
+ *
+ * @dev: non-block device information
+ */
+void unregister_pstore_device(struct pstore_device_info *dev)
+{
+	mutex_lock(&pstore_blk_lock);
+	__unregister_pstore_device(dev);
+	mutex_unlock(&pstore_blk_lock);
+}
+EXPORT_SYMBOL_GPL(unregister_pstore_device);
+
+/**
+ * psblk_get_bdev() - open block device
+ *
+ * @holder:	Exclusive holder identifier
+ * @info:	Information about bdev to fill in
+ *
+ * Return: pointer to block device on success and others on error.
+ *
+ * On success, the returned block_device has reference count of one.
+ */
+static struct block_device *psblk_get_bdev(void *holder,
+					   struct bdev_info *info)
+{
+	struct block_device *bdev = ERR_PTR(-ENODEV);
+	fmode_t mode = FMODE_READ | FMODE_WRITE;
+	sector_t nr_sects;
+
+	lockdep_assert_held(&pstore_blk_lock);
+
+	if (pstore_zone_info)
+		return ERR_PTR(-EBUSY);
+
+	if (!blkdev[0])
+		return ERR_PTR(-ENODEV);
+
+	if (holder)
+		mode |= FMODE_EXCL;
+	bdev = blkdev_get_by_path(blkdev, mode, holder);
+	if (IS_ERR(bdev)) {
+		dev_t devt;
+
+		devt = name_to_dev_t(blkdev);
+		if (devt == 0)
+			return ERR_PTR(-ENODEV);
+		bdev = blkdev_get_by_dev(devt, mode, holder);
+		if (IS_ERR(bdev))
+			return bdev;
+	}
+
+	nr_sects = part_nr_sects_read(bdev->bd_part);
+	if (!nr_sects) {
+		pr_err("not enough space for '%s'\n", blkdev);
+		blkdev_put(bdev, mode);
+		return ERR_PTR(-ENOSPC);
+	}
+
+	if (info) {
+		info->devt = bdev->bd_dev;
+		info->nr_sects = nr_sects;
+		info->start_sect = get_start_sect(bdev);
+	}
+
+	return bdev;
+}
+
+static void psblk_put_bdev(struct block_device *bdev, void *holder)
+{
+	fmode_t mode = FMODE_READ | FMODE_WRITE;
+
+	lockdep_assert_held(&pstore_blk_lock);
+
+	if (!bdev)
+		return;
+
+	if (holder)
+		mode |= FMODE_EXCL;
+	blkdev_put(bdev, mode);
+}
+
+static ssize_t psblk_generic_blk_read(char *buf, size_t bytes, loff_t pos)
+{
+	struct block_device *bdev = psblk_bdev;
+	struct file file;
+	struct kiocb kiocb;
+	struct iov_iter iter;
+	struct kvec iov = {.iov_base = buf, .iov_len = bytes};
+
+	if (!bdev)
+		return -ENODEV;
+
+	memset(&file, 0, sizeof(struct file));
+	file.f_mapping = bdev->bd_inode->i_mapping;
+	file.f_flags = O_DSYNC | __O_SYNC | O_NOATIME;
+	file.f_inode = bdev->bd_inode;
+	file_ra_state_init(&file.f_ra, file.f_mapping);
+
+	init_sync_kiocb(&kiocb, &file);
+	kiocb.ki_pos = pos;
+	iov_iter_kvec(&iter, READ, &iov, 1, bytes);
+
+	return generic_file_read_iter(&kiocb, &iter);
+}
+
+static ssize_t psblk_generic_blk_write(const char *buf, size_t bytes,
+		loff_t pos)
+{
+	struct block_device *bdev = psblk_bdev;
+	struct iov_iter iter;
+	struct kiocb kiocb;
+	struct file file;
+	ssize_t ret;
+	struct kvec iov = {.iov_base = (void *)buf, .iov_len = bytes};
+
+	if (!bdev)
+		return -ENODEV;
+
+	/* Console/Ftrace backend may handle buffer until flush dirty zones */
+	if (in_interrupt() || irqs_disabled())
+		return -EBUSY;
+
+	memset(&file, 0, sizeof(struct file));
+	file.f_mapping = bdev->bd_inode->i_mapping;
+	file.f_flags = O_DSYNC | __O_SYNC | O_NOATIME;
+	file.f_inode = bdev->bd_inode;
+
+	init_sync_kiocb(&kiocb, &file);
+	kiocb.ki_pos = pos;
+	iov_iter_kvec(&iter, WRITE, &iov, 1, bytes);
+
+	inode_lock(bdev->bd_inode);
+	ret = generic_write_checks(&kiocb, &iter);
+	if (ret > 0)
+		ret = generic_perform_write(&file, &iter, pos);
+	inode_unlock(bdev->bd_inode);
+
+	if (likely(ret > 0)) {
+		const struct file_operations f_op = {.fsync = blkdev_fsync};
+
+		file.f_op = &f_op;
+		kiocb.ki_pos += ret;
+		ret = generic_write_sync(&kiocb, ret);
+	}
+	return ret;
+}
+
+static ssize_t psblk_blk_panic_write(const char *buf, size_t size,
+		loff_t off)
+{
+	int ret;
+
+	if (!blkdev_panic_write)
+		return -EOPNOTSUPP;
+
+	/* size and off must align to SECTOR_SIZE for block device */
+	ret = blkdev_panic_write(buf, off >> SECTOR_SHIFT,
+			size >> SECTOR_SHIFT);
+	/* try next zone */
+	if (ret == -ENOMSG)
+		return ret;
+	return ret ? -EIO : size;
+}
+
+static int __register_pstore_blk(struct pstore_blk_info *info)
+{
+	char bdev_name[BDEVNAME_SIZE];
+	struct block_device *bdev;
+	struct pstore_device_info dev;
+	struct bdev_info binfo;
+	void *holder = blkdev;
+	int ret = -ENODEV;
+
+	lockdep_assert_held(&pstore_blk_lock);
+
+	/* hold bdev exclusively */
+	memset(&binfo, 0, sizeof(binfo));
+	bdev = psblk_get_bdev(holder, &binfo);
+	if (IS_ERR(bdev)) {
+		pr_err("failed to open '%s'!\n", blkdev);
+		return PTR_ERR(bdev);
+	}
+
+	/* only allow driver matching the @blkdev */
+	if (!binfo.devt || (!best_effort &&
+			    MAJOR(binfo.devt) != info->major)) {
+		pr_debug("invalid major %u (expect %u)\n",
+				info->major, MAJOR(binfo.devt));
+		ret = -ENODEV;
+		goto err_put_bdev;
+	}
+
+	/* psblk_bdev must be assigned before register to pstore/blk */
+	psblk_bdev = bdev;
+	blkdev_panic_write = info->panic_write;
+
+	/* Copy back block device details. */
+	info->devt = binfo.devt;
+	info->nr_sects = binfo.nr_sects;
+	info->start_sect = binfo.start_sect;
+
+	memset(&dev, 0, sizeof(dev));
+	dev.total_size = info->nr_sects << SECTOR_SHIFT;
+	dev.flags = info->flags;
+	dev.read = psblk_generic_blk_read;
+	dev.write = psblk_generic_blk_write;
+	dev.erase = NULL;
+	dev.panic_write = info->panic_write ? psblk_blk_panic_write : NULL;
+
+	ret = __register_pstore_device(&dev);
+	if (ret)
+		goto err_put_bdev;
+
+	bdevname(bdev, bdev_name);
+	pr_info("attached %s%s\n", bdev_name,
+		info->panic_write ? "" : " (no dedicated panic_write!)");
+	return 0;
+
+err_put_bdev:
+	psblk_bdev = NULL;
+	blkdev_panic_write = NULL;
+	psblk_put_bdev(bdev, holder);
+	return ret;
+}
+
+/**
+ * register_pstore_blk() - register block device to pstore/blk
+ *
+ * @info: details on the desired block device interface
+ *
+ * Return:
+ * * 0		- OK
+ * * Others	- something error.
+ */
+int register_pstore_blk(struct pstore_blk_info *info)
+{
+	int ret;
+
+	mutex_lock(&pstore_blk_lock);
+	ret = __register_pstore_blk(info);
+	mutex_unlock(&pstore_blk_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(register_pstore_blk);
+
+static void __unregister_pstore_blk(unsigned int major)
+{
+	struct pstore_device_info dev = { .read = psblk_generic_blk_read };
+	void *holder = blkdev;
+
+	lockdep_assert_held(&pstore_blk_lock);
+	if (psblk_bdev && MAJOR(psblk_bdev->bd_dev) == major) {
+		__unregister_pstore_device(&dev);
+		psblk_put_bdev(psblk_bdev, holder);
+		blkdev_panic_write = NULL;
+		psblk_bdev = NULL;
+	}
+}
+
+/**
+ * unregister_pstore_blk() - unregister block device from pstore/blk
+ *
+ * @major: the major device number of device
+ */
+void unregister_pstore_blk(unsigned int major)
+{
+	mutex_lock(&pstore_blk_lock);
+	__unregister_pstore_blk(major);
+	mutex_unlock(&pstore_blk_lock);
+}
+EXPORT_SYMBOL_GPL(unregister_pstore_blk);
+
+/* get information of pstore/blk */
+int pstore_blk_get_config(struct pstore_blk_config *info)
+{
+	strncpy(info->device, blkdev, 80);
+	info->max_reason = max_reason;
+	info->kmsg_size = check_size(kmsg_size, 4096);
+	info->pmsg_size = check_size(pmsg_size, 4096);
+	info->ftrace_size = check_size(ftrace_size, 4096);
+	info->console_size = check_size(console_size, 4096);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(pstore_blk_get_config);
+
+static int __init pstore_blk_init(void)
+{
+	struct pstore_blk_info info = { };
+	int ret = 0;
+
+	mutex_lock(&pstore_blk_lock);
+	if (!pstore_zone_info && best_effort && blkdev[0])
+		ret = __register_pstore_blk(&info);
+	mutex_unlock(&pstore_blk_lock);
+
+	return ret;
+}
+late_initcall(pstore_blk_init);
+
+static void __exit pstore_blk_exit(void)
+{
+	mutex_lock(&pstore_blk_lock);
+	if (psblk_bdev)
+		__unregister_pstore_blk(MAJOR(psblk_bdev->bd_dev));
+	else {
+		struct pstore_device_info dev = { };
+
+		if (pstore_zone_info)
+			dev.read = pstore_zone_info->read;
+		__unregister_pstore_device(&dev);
+	}
+	mutex_unlock(&pstore_blk_lock);
+}
+module_exit(pstore_blk_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("WeiXiong Liao <liaoweixiong@allwinnertech.com>");
+MODULE_AUTHOR("Kees Cook <keescook@chromium.org>");
+MODULE_DESCRIPTION("pstore backend for block devices");
diff --git a/fs/pstore/ftrace.c b/fs/pstore/ftrace.c
index bfbfc2698070..5c0450701293 100644
--- a/fs/pstore/ftrace.c
+++ b/fs/pstore/ftrace.c
@@ -16,6 +16,7 @@
 #include <linux/debugfs.h>
 #include <linux/err.h>
 #include <linux/cache.h>
+#include <linux/slab.h>
 #include <asm/barrier.h>
 #include "internal.h"
 
@@ -132,3 +133,56 @@ void pstore_unregister_ftrace(void)
 
 	debugfs_remove_recursive(pstore_ftrace_dir);
 }
+
+ssize_t pstore_ftrace_combine_log(char **dest_log, size_t *dest_log_size,
+				  const char *src_log, size_t src_log_size)
+{
+	size_t dest_size, src_size, total, dest_off, src_off;
+	size_t dest_idx = 0, src_idx = 0, merged_idx = 0;
+	void *merged_buf;
+	struct pstore_ftrace_record *drec, *srec, *mrec;
+	size_t record_size = sizeof(struct pstore_ftrace_record);
+
+	dest_off = *dest_log_size % record_size;
+	dest_size = *dest_log_size - dest_off;
+
+	src_off = src_log_size % record_size;
+	src_size = src_log_size - src_off;
+
+	total = dest_size + src_size;
+	merged_buf = kmalloc(total, GFP_KERNEL);
+	if (!merged_buf)
+		return -ENOMEM;
+
+	drec = (struct pstore_ftrace_record *)(*dest_log + dest_off);
+	srec = (struct pstore_ftrace_record *)(src_log + src_off);
+	mrec = (struct pstore_ftrace_record *)(merged_buf);
+
+	while (dest_size > 0 && src_size > 0) {
+		if (pstore_ftrace_read_timestamp(&drec[dest_idx]) <
+		    pstore_ftrace_read_timestamp(&srec[src_idx])) {
+			mrec[merged_idx++] = drec[dest_idx++];
+			dest_size -= record_size;
+		} else {
+			mrec[merged_idx++] = srec[src_idx++];
+			src_size -= record_size;
+		}
+	}
+
+	while (dest_size > 0) {
+		mrec[merged_idx++] = drec[dest_idx++];
+		dest_size -= record_size;
+	}
+
+	while (src_size > 0) {
+		mrec[merged_idx++] = srec[src_idx++];
+		src_size -= record_size;
+	}
+
+	kfree(*dest_log);
+	*dest_log = merged_buf;
+	*dest_log_size = total;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(pstore_ftrace_combine_log);
diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c
index 7fbe8f058220..c331efe8de95 100644
--- a/fs/pstore/inode.c
+++ b/fs/pstore/inode.c
@@ -22,18 +22,21 @@
 #include <linux/magic.h>
 #include <linux/pstore.h>
 #include <linux/slab.h>
-#include <linux/spinlock.h>
 #include <linux/uaccess.h>
 
 #include "internal.h"
 
 #define	PSTORE_NAMELEN	64
 
-static DEFINE_SPINLOCK(allpstore_lock);
-static LIST_HEAD(allpstore);
+static DEFINE_MUTEX(records_list_lock);
+static LIST_HEAD(records_list);
+
+static DEFINE_MUTEX(pstore_sb_lock);
+static struct super_block *pstore_sb;
 
 struct pstore_private {
 	struct list_head list;
+	struct dentry *dentry;
 	struct pstore_record *record;
 	size_t total_size;
 };
@@ -87,11 +90,11 @@ static void *pstore_ftrace_seq_next(struct seq_file *s, void *v, loff_t *pos)
 	struct pstore_private *ps = s->private;
 	struct pstore_ftrace_seq_data *data = v;
 
+	(*pos)++;
 	data->off += REC_SIZE;
 	if (data->off + REC_SIZE > ps->total_size)
 		return NULL;
 
-	(*pos)++;
 	return data;
 }
 
@@ -101,6 +104,9 @@ static int pstore_ftrace_seq_show(struct seq_file *s, void *v)
 	struct pstore_ftrace_seq_data *data = v;
 	struct pstore_ftrace_record *rec;
 
+	if (!data)
+		return 0;
+
 	rec = (struct pstore_ftrace_record *)(ps->record->buf + data->off);
 
 	seq_printf(s, "CPU:%d ts:%llu %08lx  %08lx  %ps <- %pS\n",
@@ -175,10 +181,22 @@ static int pstore_unlink(struct inode *dir, struct dentry *dentry)
 {
 	struct pstore_private *p = d_inode(dentry)->i_private;
 	struct pstore_record *record = p->record;
+	int rc = 0;
 
 	if (!record->psi->erase)
 		return -EPERM;
 
+	/* Make sure we can't race while removing this file. */
+	mutex_lock(&records_list_lock);
+	if (!list_empty(&p->list))
+		list_del_init(&p->list);
+	else
+		rc = -ENOENT;
+	p->dentry = NULL;
+	mutex_unlock(&records_list_lock);
+	if (rc)
+		return rc;
+
 	mutex_lock(&record->psi->read_mutex);
 	record->psi->erase(record);
 	mutex_unlock(&record->psi->read_mutex);
@@ -189,15 +207,9 @@ static int pstore_unlink(struct inode *dir, struct dentry *dentry)
 static void pstore_evict_inode(struct inode *inode)
 {
 	struct pstore_private	*p = inode->i_private;
-	unsigned long		flags;
 
 	clear_inode(inode);
-	if (p) {
-		spin_lock_irqsave(&allpstore_lock, flags);
-		list_del(&p->list);
-		spin_unlock_irqrestore(&allpstore_lock, flags);
-		free_pstore_private(p);
-	}
+	free_pstore_private(p);
 }
 
 static const struct inode_operations pstore_dir_inode_operations = {
@@ -275,11 +287,54 @@ static const struct super_operations pstore_ops = {
 	.show_options	= pstore_show_options,
 };
 
-static struct super_block *pstore_sb;
+static struct dentry *psinfo_lock_root(void)
+{
+	struct dentry *root;
+
+	mutex_lock(&pstore_sb_lock);
+	/*
+	 * Having no backend is fine -- no records appear.
+	 * Not being mounted is fine -- nothing to do.
+	 */
+	if (!psinfo || !pstore_sb) {
+		mutex_unlock(&pstore_sb_lock);
+		return NULL;
+	}
 
-bool pstore_is_mounted(void)
+	root = pstore_sb->s_root;
+	inode_lock(d_inode(root));
+	mutex_unlock(&pstore_sb_lock);
+
+	return root;
+}
+
+int pstore_put_backend_records(struct pstore_info *psi)
 {
-	return pstore_sb != NULL;
+	struct pstore_private *pos, *tmp;
+	struct dentry *root;
+	int rc = 0;
+
+	root = psinfo_lock_root();
+	if (!root)
+		return 0;
+
+	mutex_lock(&records_list_lock);
+	list_for_each_entry_safe(pos, tmp, &records_list, list) {
+		if (pos->record->psi == psi) {
+			list_del_init(&pos->list);
+			rc = simple_unlink(d_inode(root), pos->dentry);
+			if (WARN_ON(rc))
+				break;
+			d_drop(pos->dentry);
+			dput(pos->dentry);
+			pos->dentry = NULL;
+		}
+	}
+	mutex_unlock(&records_list_lock);
+
+	inode_unlock(d_inode(root));
+
+	return rc;
 }
 
 /*
@@ -294,23 +349,20 @@ int pstore_mkfile(struct dentry *root, struct pstore_record *record)
 	int			rc = 0;
 	char			name[PSTORE_NAMELEN];
 	struct pstore_private	*private, *pos;
-	unsigned long		flags;
 	size_t			size = record->size + record->ecc_notice_size;
 
-	WARN_ON(!inode_is_locked(d_inode(root)));
+	if (WARN_ON(!inode_is_locked(d_inode(root))))
+		return -EINVAL;
 
-	spin_lock_irqsave(&allpstore_lock, flags);
-	list_for_each_entry(pos, &allpstore, list) {
+	rc = -EEXIST;
+	/* Skip records that are already present in the filesystem. */
+	mutex_lock(&records_list_lock);
+	list_for_each_entry(pos, &records_list, list) {
 		if (pos->record->type == record->type &&
 		    pos->record->id == record->id &&
-		    pos->record->psi == record->psi) {
-			rc = -EEXIST;
-			break;
-		}
+		    pos->record->psi == record->psi)
+			goto fail;
 	}
-	spin_unlock_irqrestore(&allpstore_lock, flags);
-	if (rc)
-		return rc;
 
 	rc = -ENOMEM;
 	inode = pstore_get_inode(root->d_sb);
@@ -331,6 +383,7 @@ int pstore_mkfile(struct dentry *root, struct pstore_record *record)
 	if (!dentry)
 		goto fail_private;
 
+	private->dentry = dentry;
 	private->record = record;
 	inode->i_size = private->total_size = size;
 	inode->i_private = private;
@@ -340,9 +393,8 @@ int pstore_mkfile(struct dentry *root, struct pstore_record *record)
 
 	d_add(dentry, inode);
 
-	spin_lock_irqsave(&allpstore_lock, flags);
-	list_add(&private->list, &allpstore);
-	spin_unlock_irqrestore(&allpstore_lock, flags);
+	list_add(&private->list, &records_list);
+	mutex_unlock(&records_list_lock);
 
 	return 0;
 
@@ -350,8 +402,8 @@ fail_private:
 	free_pstore_private(private);
 fail_inode:
 	iput(inode);
-
 fail:
+	mutex_unlock(&records_list_lock);
 	return rc;
 }
 
@@ -363,16 +415,13 @@ fail:
  */
 void pstore_get_records(int quiet)
 {
-	struct pstore_info *psi = psinfo;
 	struct dentry *root;
 
-	if (!psi || !pstore_sb)
+	root = psinfo_lock_root();
+	if (!root)
 		return;
 
-	root = pstore_sb->s_root;
-
-	inode_lock(d_inode(root));
-	pstore_get_backend_records(psi, root, quiet);
+	pstore_get_backend_records(psinfo, root, quiet);
 	inode_unlock(d_inode(root));
 }
 
@@ -380,8 +429,6 @@ static int pstore_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct inode *inode;
 
-	pstore_sb = sb;
-
 	sb->s_maxbytes		= MAX_LFS_FILESIZE;
 	sb->s_blocksize		= PAGE_SIZE;
 	sb->s_blocksize_bits	= PAGE_SHIFT;
@@ -402,6 +449,10 @@ static int pstore_fill_super(struct super_block *sb, void *data, int silent)
 	if (!sb->s_root)
 		return -ENOMEM;
 
+	mutex_lock(&pstore_sb_lock);
+	pstore_sb = sb;
+	mutex_unlock(&pstore_sb_lock);
+
 	pstore_get_records(0);
 
 	return 0;
@@ -415,8 +466,17 @@ static struct dentry *pstore_mount(struct file_system_type *fs_type,
 
 static void pstore_kill_sb(struct super_block *sb)
 {
+	mutex_lock(&pstore_sb_lock);
+	WARN_ON(pstore_sb != sb);
+
 	kill_litter_super(sb);
 	pstore_sb = NULL;
+
+	mutex_lock(&records_list_lock);
+	INIT_LIST_HEAD(&records_list);
+	mutex_unlock(&records_list_lock);
+
+	mutex_unlock(&pstore_sb_lock);
 }
 
 static struct file_system_type pstore_fs_type = {
diff --git a/fs/pstore/internal.h b/fs/pstore/internal.h
index 7062ea4bc57c..7fb219042f13 100644
--- a/fs/pstore/internal.h
+++ b/fs/pstore/internal.h
@@ -12,9 +12,18 @@ extern unsigned long kmsg_bytes;
 #ifdef CONFIG_PSTORE_FTRACE
 extern void pstore_register_ftrace(void);
 extern void pstore_unregister_ftrace(void);
+ssize_t pstore_ftrace_combine_log(char **dest_log, size_t *dest_log_size,
+				  const char *src_log, size_t src_log_size);
 #else
 static inline void pstore_register_ftrace(void) {}
 static inline void pstore_unregister_ftrace(void) {}
+static inline ssize_t
+pstore_ftrace_combine_log(char **dest_log, size_t *dest_log_size,
+			  const char *src_log, size_t src_log_size)
+{
+	*dest_log_size = 0;
+	return 0;
+}
 #endif
 
 #ifdef CONFIG_PSTORE_PMSG
@@ -31,9 +40,9 @@ extern void	pstore_set_kmsg_bytes(int);
 extern void	pstore_get_records(int);
 extern void	pstore_get_backend_records(struct pstore_info *psi,
 					   struct dentry *root, int quiet);
+extern int	pstore_put_backend_records(struct pstore_info *psi);
 extern int	pstore_mkfile(struct dentry *root,
 			      struct pstore_record *record);
-extern bool	pstore_is_mounted(void);
 extern void	pstore_record_init(struct pstore_record *record,
 				   struct pstore_info *psi);
 
diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c
index 3d7024662d29..a9e297eefdff 100644
--- a/fs/pstore/platform.c
+++ b/fs/pstore/platform.c
@@ -44,7 +44,7 @@ static int pstore_update_ms = -1;
 module_param_named(update_ms, pstore_update_ms, int, 0600);
 MODULE_PARM_DESC(update_ms, "milliseconds before pstore updates its content "
 		 "(default is -1, which means runtime updates are disabled; "
-		 "enabling this option is not safe, it may lead to further "
+		 "enabling this option may not be safe; it may lead to further "
 		 "corruption on Oopses)");
 
 /* Names should be in the same order as the enum pstore_type_id */
@@ -69,19 +69,25 @@ static void pstore_dowork(struct work_struct *);
 static DECLARE_WORK(pstore_work, pstore_dowork);
 
 /*
- * pstore_lock just protects "psinfo" during
- * calls to pstore_register()
+ * psinfo_lock protects "psinfo" during calls to
+ * pstore_register(), pstore_unregister(), and
+ * the filesystem mount/unmount routines.
  */
-static DEFINE_SPINLOCK(pstore_lock);
+static DEFINE_MUTEX(psinfo_lock);
 struct pstore_info *psinfo;
 
 static char *backend;
+module_param(backend, charp, 0444);
+MODULE_PARM_DESC(backend, "specific backend to use");
+
 static char *compress =
 #ifdef CONFIG_PSTORE_COMPRESS_DEFAULT
 		CONFIG_PSTORE_COMPRESS_DEFAULT;
 #else
 		NULL;
 #endif
+module_param(compress, charp, 0444);
+MODULE_PARM_DESC(compress, "compression to use");
 
 /* Compression parameters */
 static struct crypto_comp *tfm;
@@ -129,24 +135,12 @@ enum pstore_type_id pstore_name_to_type(const char *name)
 }
 EXPORT_SYMBOL_GPL(pstore_name_to_type);
 
-static const char *get_reason_str(enum kmsg_dump_reason reason)
+static void pstore_timer_kick(void)
 {
-	switch (reason) {
-	case KMSG_DUMP_PANIC:
-		return "Panic";
-	case KMSG_DUMP_OOPS:
-		return "Oops";
-	case KMSG_DUMP_EMERG:
-		return "Emergency";
-	case KMSG_DUMP_RESTART:
-		return "Restart";
-	case KMSG_DUMP_HALT:
-		return "Halt";
-	case KMSG_DUMP_POWEROFF:
-		return "Poweroff";
-	default:
-		return "Unknown";
-	}
+	if (pstore_update_ms < 0)
+		return;
+
+	mod_timer(&pstore_timer, jiffies + msecs_to_jiffies(pstore_update_ms));
 }
 
 /*
@@ -393,7 +387,7 @@ static void pstore_dump(struct kmsg_dumper *dumper,
 	unsigned int	part = 1;
 	int		ret;
 
-	why = get_reason_str(reason);
+	why = kmsg_dump_reason_str(reason);
 
 	if (down_trylock(&psinfo->buf_lock)) {
 		/* Failed to acquire lock: give up if we cannot wait. */
@@ -459,8 +453,10 @@ static void pstore_dump(struct kmsg_dumper *dumper,
 		}
 
 		ret = psinfo->write(&record);
-		if (ret == 0 && reason == KMSG_DUMP_OOPS && pstore_is_mounted())
+		if (ret == 0 && reason == KMSG_DUMP_OOPS) {
 			pstore_new_entry = 1;
+			pstore_timer_kick();
+		}
 
 		total += record.size;
 		part++;
@@ -503,14 +499,20 @@ static void pstore_console_write(struct console *con, const char *s, unsigned c)
 }
 
 static struct console pstore_console = {
-	.name	= "pstore",
 	.write	= pstore_console_write,
-	.flags	= CON_PRINTBUFFER | CON_ENABLED | CON_ANYTIME,
 	.index	= -1,
 };
 
 static void pstore_register_console(void)
 {
+	/* Show which backend is going to get console writes. */
+	strscpy(pstore_console.name, psinfo->name,
+		sizeof(pstore_console.name));
+	/*
+	 * Always initialize flags here since prior unregister_console()
+	 * calls may have changed settings (specifically CON_ENABLED).
+	 */
+	pstore_console.flags = CON_PRINTBUFFER | CON_ENABLED | CON_ANYTIME;
 	register_console(&pstore_console);
 }
 
@@ -555,8 +557,6 @@ out:
  */
 int pstore_register(struct pstore_info *psi)
 {
-	struct module *owner = psi->owner;
-
 	if (backend && strcmp(backend, psi->name)) {
 		pr_warn("ignoring unexpected backend '%s'\n", psi->name);
 		return -EPERM;
@@ -576,11 +576,11 @@ int pstore_register(struct pstore_info *psi)
 		return -EINVAL;
 	}
 
-	spin_lock(&pstore_lock);
+	mutex_lock(&psinfo_lock);
 	if (psinfo) {
 		pr_warn("backend '%s' already loaded: ignoring '%s'\n",
 			psinfo->name, psi->name);
-		spin_unlock(&pstore_lock);
+		mutex_unlock(&psinfo_lock);
 		return -EBUSY;
 	}
 
@@ -589,21 +589,16 @@ int pstore_register(struct pstore_info *psi)
 	psinfo = psi;
 	mutex_init(&psinfo->read_mutex);
 	sema_init(&psinfo->buf_lock, 1);
-	spin_unlock(&pstore_lock);
-
-	if (owner && !try_module_get(owner)) {
-		psinfo = NULL;
-		return -EINVAL;
-	}
 
 	if (psi->flags & PSTORE_FLAGS_DMESG)
 		allocate_buf_for_compression();
 
-	if (pstore_is_mounted())
-		pstore_get_records(0);
+	pstore_get_records(0);
 
-	if (psi->flags & PSTORE_FLAGS_DMESG)
+	if (psi->flags & PSTORE_FLAGS_DMESG) {
+		pstore_dumper.max_reason = psinfo->max_reason;
 		pstore_register_kmsg();
+	}
 	if (psi->flags & PSTORE_FLAGS_CONSOLE)
 		pstore_register_console();
 	if (psi->flags & PSTORE_FLAGS_FTRACE)
@@ -612,33 +607,36 @@ int pstore_register(struct pstore_info *psi)
 		pstore_register_pmsg();
 
 	/* Start watching for new records, if desired. */
-	if (pstore_update_ms >= 0) {
-		pstore_timer.expires = jiffies +
-			msecs_to_jiffies(pstore_update_ms);
-		add_timer(&pstore_timer);
-	}
+	pstore_timer_kick();
 
 	/*
 	 * Update the module parameter backend, so it is visible
 	 * through /sys/module/pstore/parameters/backend
 	 */
-	backend = psi->name;
+	backend = kstrdup(psi->name, GFP_KERNEL);
 
 	pr_info("Registered %s as persistent store backend\n", psi->name);
 
-	module_put(owner);
-
+	mutex_unlock(&psinfo_lock);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(pstore_register);
 
 void pstore_unregister(struct pstore_info *psi)
 {
-	/* Stop timer and make sure all work has finished. */
-	pstore_update_ms = -1;
-	del_timer_sync(&pstore_timer);
-	flush_work(&pstore_work);
+	/* It's okay to unregister nothing. */
+	if (!psi)
+		return;
 
+	mutex_lock(&psinfo_lock);
+
+	/* Only one backend can be registered at a time. */
+	if (WARN_ON(psi != psinfo)) {
+		mutex_unlock(&psinfo_lock);
+		return;
+	}
+
+	/* Unregister all callbacks. */
 	if (psi->flags & PSTORE_FLAGS_PMSG)
 		pstore_unregister_pmsg();
 	if (psi->flags & PSTORE_FLAGS_FTRACE)
@@ -648,10 +646,19 @@ void pstore_unregister(struct pstore_info *psi)
 	if (psi->flags & PSTORE_FLAGS_DMESG)
 		pstore_unregister_kmsg();
 
+	/* Stop timer and make sure all work has finished. */
+	del_timer_sync(&pstore_timer);
+	flush_work(&pstore_work);
+
+	/* Remove all backend records from filesystem tree. */
+	pstore_put_backend_records(psi);
+
 	free_buf_for_compression();
 
 	psinfo = NULL;
+	kfree(backend);
 	backend = NULL;
+	mutex_unlock(&psinfo_lock);
 }
 EXPORT_SYMBOL_GPL(pstore_unregister);
 
@@ -788,12 +795,10 @@ static void pstore_timefunc(struct timer_list *unused)
 		schedule_work(&pstore_work);
 	}
 
-	if (pstore_update_ms >= 0)
-		mod_timer(&pstore_timer,
-			  jiffies + msecs_to_jiffies(pstore_update_ms));
+	pstore_timer_kick();
 }
 
-void __init pstore_choose_compression(void)
+static void __init pstore_choose_compression(void)
 {
 	const struct pstore_zbackend *step;
 
@@ -823,9 +828,9 @@ static int __init pstore_init(void)
 
 	ret = pstore_init_fs();
 	if (ret)
-		return ret;
+		free_buf_for_compression();
 
-	return 0;
+	return ret;
 }
 late_initcall(pstore_init);
 
@@ -835,11 +840,5 @@ static void __exit pstore_exit(void)
 }
 module_exit(pstore_exit)
 
-module_param(compress, charp, 0444);
-MODULE_PARM_DESC(compress, "Pstore compression to use");
-
-module_param(backend, charp, 0444);
-MODULE_PARM_DESC(backend, "Pstore backend to use");
-
 MODULE_AUTHOR("Tony Luck <tony.luck@intel.com>");
 MODULE_LICENSE("GPL");
diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c
index 8caff834f002..ca6d8a867285 100644
--- a/fs/pstore/ram.c
+++ b/fs/pstore/ram.c
@@ -21,6 +21,7 @@
 #include <linux/pstore_ram.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
+#include "internal.h"
 
 #define RAMOOPS_KERNMSG_HDR "===="
 #define MIN_MEM_SIZE 4096UL
@@ -53,22 +54,27 @@ MODULE_PARM_DESC(mem_size,
 		"size of reserved RAM used to store oops/panic logs");
 
 static unsigned int mem_type;
-module_param(mem_type, uint, 0600);
+module_param(mem_type, uint, 0400);
 MODULE_PARM_DESC(mem_type,
 		"set to 1 to try to use unbuffered memory (default 0)");
 
-static int dump_oops = 1;
-module_param(dump_oops, int, 0600);
-MODULE_PARM_DESC(dump_oops,
-		"set to 1 to dump oopses, 0 to only dump panics (default 1)");
+static int ramoops_max_reason = -1;
+module_param_named(max_reason, ramoops_max_reason, int, 0400);
+MODULE_PARM_DESC(max_reason,
+		 "maximum reason for kmsg dump (default 2: Oops and Panic) ");
 
 static int ramoops_ecc;
-module_param_named(ecc, ramoops_ecc, int, 0600);
+module_param_named(ecc, ramoops_ecc, int, 0400);
 MODULE_PARM_DESC(ramoops_ecc,
 		"if non-zero, the option enables ECC support and specifies "
 		"ECC buffer size in bytes (1 is a special value, means 16 "
 		"bytes ECC)");
 
+static int ramoops_dump_oops = -1;
+module_param_named(dump_oops, ramoops_dump_oops, int, 0400);
+MODULE_PARM_DESC(dump_oops,
+		 "(deprecated: use max_reason instead) set to 1 to dump oopses & panics, 0 to only dump panics");
+
 struct ramoops_context {
 	struct persistent_ram_zone **dprzs;	/* Oops dump zones */
 	struct persistent_ram_zone *cprz;	/* Console zone */
@@ -81,7 +87,6 @@ struct ramoops_context {
 	size_t console_size;
 	size_t ftrace_size;
 	size_t pmsg_size;
-	int dump_oops;
 	u32 flags;
 	struct persistent_ram_ecc_info ecc_info;
 	unsigned int max_dump_cnt;
@@ -168,58 +173,6 @@ static bool prz_ok(struct persistent_ram_zone *prz)
 			   persistent_ram_ecc_string(prz, NULL, 0));
 }
 
-static ssize_t ftrace_log_combine(struct persistent_ram_zone *dest,
-				  struct persistent_ram_zone *src)
-{
-	size_t dest_size, src_size, total, dest_off, src_off;
-	size_t dest_idx = 0, src_idx = 0, merged_idx = 0;
-	void *merged_buf;
-	struct pstore_ftrace_record *drec, *srec, *mrec;
-	size_t record_size = sizeof(struct pstore_ftrace_record);
-
-	dest_off = dest->old_log_size % record_size;
-	dest_size = dest->old_log_size - dest_off;
-
-	src_off = src->old_log_size % record_size;
-	src_size = src->old_log_size - src_off;
-
-	total = dest_size + src_size;
-	merged_buf = kmalloc(total, GFP_KERNEL);
-	if (!merged_buf)
-		return -ENOMEM;
-
-	drec = (struct pstore_ftrace_record *)(dest->old_log + dest_off);
-	srec = (struct pstore_ftrace_record *)(src->old_log + src_off);
-	mrec = (struct pstore_ftrace_record *)(merged_buf);
-
-	while (dest_size > 0 && src_size > 0) {
-		if (pstore_ftrace_read_timestamp(&drec[dest_idx]) <
-		    pstore_ftrace_read_timestamp(&srec[src_idx])) {
-			mrec[merged_idx++] = drec[dest_idx++];
-			dest_size -= record_size;
-		} else {
-			mrec[merged_idx++] = srec[src_idx++];
-			src_size -= record_size;
-		}
-	}
-
-	while (dest_size > 0) {
-		mrec[merged_idx++] = drec[dest_idx++];
-		dest_size -= record_size;
-	}
-
-	while (src_size > 0) {
-		mrec[merged_idx++] = srec[src_idx++];
-		src_size -= record_size;
-	}
-
-	kfree(dest->old_log);
-	dest->old_log = merged_buf;
-	dest->old_log_size = total;
-
-	return 0;
-}
-
 static ssize_t ramoops_pstore_read(struct pstore_record *record)
 {
 	ssize_t size = 0;
@@ -291,7 +244,12 @@ static ssize_t ramoops_pstore_read(struct pstore_record *record)
 				tmp_prz->corrected_bytes +=
 						prz_next->corrected_bytes;
 				tmp_prz->bad_blocks += prz_next->bad_blocks;
-				size = ftrace_log_combine(tmp_prz, prz_next);
+
+				size = pstore_ftrace_combine_log(
+						&tmp_prz->old_log,
+						&tmp_prz->old_log_size,
+						prz_next->old_log,
+						prz_next->old_log_size);
 				if (size)
 					goto out;
 			}
@@ -382,16 +340,14 @@ static int notrace ramoops_pstore_write(struct pstore_record *record)
 		return -EINVAL;
 
 	/*
-	 * Out of the various dmesg dump types, ramoops is currently designed
-	 * to only store crash logs, rather than storing general kernel logs.
+	 * We could filter on record->reason here if we wanted to (which
+	 * would duplicate what happened before the "max_reason" setting
+	 * was added), but that would defeat the purpose of a system
+	 * changing printk.always_kmsg_dump, so instead log everything that
+	 * the kmsg dumper sends us, since it should be doing the filtering
+	 * based on the combination of printk.always_kmsg_dump and our
+	 * requested "max_reason".
 	 */
-	if (record->reason != KMSG_DUMP_OOPS &&
-	    record->reason != KMSG_DUMP_PANIC)
-		return -EINVAL;
-
-	/* Skip Oopes when configured to do so. */
-	if (record->reason == KMSG_DUMP_OOPS && !cxt->dump_oops)
-		return -EINVAL;
 
 	/*
 	 * Explicitly only take the first part of any new crash.
@@ -407,6 +363,17 @@ static int notrace ramoops_pstore_write(struct pstore_record *record)
 
 	prz = cxt->dprzs[cxt->dump_write_cnt];
 
+	/*
+	 * Since this is a new crash dump, we need to reset the buffer in
+	 * case it still has an old dump present. Without this, the new dump
+	 * will get appended, which would seriously confuse anything trying
+	 * to check dump file contents. Specifically, ramoops_read_kmsg_hdr()
+	 * expects to find a dump header in the beginning of buffer data, so
+	 * we must to reset the buffer values, in order to ensure that the
+	 * header will be written to the beginning of the buffer.
+	 */
+	persistent_ram_zap(prz);
+
 	/* Build header and append record contents. */
 	hlen = ramoops_write_kmsg_hdr(prz, record);
 	if (!hlen)
@@ -572,6 +539,7 @@ static int ramoops_init_przs(const char *name,
 		prz_ar[i] = persistent_ram_new(*paddr, zone_sz, sig,
 					       &cxt->ecc_info,
 					       cxt->memtype, flags, label);
+		kfree(label);
 		if (IS_ERR(prz_ar[i])) {
 			err = PTR_ERR(prz_ar[i]);
 			dev_err(dev, "failed to request %s mem region (0x%zx@0x%llx): %d\n",
@@ -617,6 +585,7 @@ static int ramoops_init_prz(const char *name,
 	label = kasprintf(GFP_KERNEL, "ramoops:%s", name);
 	*prz = persistent_ram_new(*paddr, sz, sig, &cxt->ecc_info,
 				  cxt->memtype, PRZ_FLAG_ZAP_OLD, label);
+	kfree(label);
 	if (IS_ERR(*prz)) {
 		int err = PTR_ERR(*prz);
 
@@ -631,19 +600,25 @@ static int ramoops_init_prz(const char *name,
 	return 0;
 }
 
-static int ramoops_parse_dt_size(struct platform_device *pdev,
-				 const char *propname, u32 *value)
+/* Read a u32 from a dt property and make sure it's safe for an int. */
+static int ramoops_parse_dt_u32(struct platform_device *pdev,
+				const char *propname,
+				u32 default_value, u32 *value)
 {
 	u32 val32 = 0;
 	int ret;
 
 	ret = of_property_read_u32(pdev->dev.of_node, propname, &val32);
-	if (ret < 0 && ret != -EINVAL) {
+	if (ret == -EINVAL) {
+		/* field is missing, use default value. */
+		val32 = default_value;
+	} else if (ret < 0) {
 		dev_err(&pdev->dev, "failed to parse property %s: %d\n",
 			propname, ret);
 		return ret;
 	}
 
+	/* Sanity check our results. */
 	if (val32 > INT_MAX) {
 		dev_err(&pdev->dev, "%s %u > INT_MAX\n", propname, val32);
 		return -EOVERFLOW;
@@ -674,23 +649,32 @@ static int ramoops_parse_dt(struct platform_device *pdev,
 	pdata->mem_size = resource_size(res);
 	pdata->mem_address = res->start;
 	pdata->mem_type = of_property_read_bool(of_node, "unbuffered");
-	pdata->dump_oops = !of_property_read_bool(of_node, "no-dump-oops");
-
-#define parse_size(name, field) {					\
-		ret = ramoops_parse_dt_size(pdev, name, &value);	\
+	/*
+	 * Setting "no-dump-oops" is deprecated and will be ignored if
+	 * "max_reason" is also specified.
+	 */
+	if (of_property_read_bool(of_node, "no-dump-oops"))
+		pdata->max_reason = KMSG_DUMP_PANIC;
+	else
+		pdata->max_reason = KMSG_DUMP_OOPS;
+
+#define parse_u32(name, field, default_value) {				\
+		ret = ramoops_parse_dt_u32(pdev, name, default_value,	\
+					    &value);			\
 		if (ret < 0)						\
 			return ret;					\
 		field = value;						\
 	}
 
-	parse_size("record-size", pdata->record_size);
-	parse_size("console-size", pdata->console_size);
-	parse_size("ftrace-size", pdata->ftrace_size);
-	parse_size("pmsg-size", pdata->pmsg_size);
-	parse_size("ecc-size", pdata->ecc_info.ecc_size);
-	parse_size("flags", pdata->flags);
+	parse_u32("record-size", pdata->record_size, 0);
+	parse_u32("console-size", pdata->console_size, 0);
+	parse_u32("ftrace-size", pdata->ftrace_size, 0);
+	parse_u32("pmsg-size", pdata->pmsg_size, 0);
+	parse_u32("ecc-size", pdata->ecc_info.ecc_size, 0);
+	parse_u32("flags", pdata->flags, 0);
+	parse_u32("max-reason", pdata->max_reason, pdata->max_reason);
 
-#undef parse_size
+#undef parse_u32
 
 	/*
 	 * Some old Chromebooks relied on the kernel setting the
@@ -772,7 +756,6 @@ static int ramoops_probe(struct platform_device *pdev)
 	cxt->console_size = pdata->console_size;
 	cxt->ftrace_size = pdata->ftrace_size;
 	cxt->pmsg_size = pdata->pmsg_size;
-	cxt->dump_oops = pdata->dump_oops;
 	cxt->flags = pdata->flags;
 	cxt->ecc_info = pdata->ecc_info;
 
@@ -815,8 +798,10 @@ static int ramoops_probe(struct platform_device *pdev)
 	 * the single region size is how to check.
 	 */
 	cxt->pstore.flags = 0;
-	if (cxt->max_dump_cnt)
+	if (cxt->max_dump_cnt) {
 		cxt->pstore.flags |= PSTORE_FLAGS_DMESG;
+		cxt->pstore.max_reason = pdata->max_reason;
+	}
 	if (cxt->console_size)
 		cxt->pstore.flags |= PSTORE_FLAGS_CONSOLE;
 	if (cxt->max_ftrace_cnt)
@@ -852,7 +837,7 @@ static int ramoops_probe(struct platform_device *pdev)
 	mem_size = pdata->mem_size;
 	mem_address = pdata->mem_address;
 	record_size = pdata->record_size;
-	dump_oops = pdata->dump_oops;
+	ramoops_max_reason = pdata->max_reason;
 	ramoops_console_size = pdata->console_size;
 	ramoops_pmsg_size = pdata->pmsg_size;
 	ramoops_ftrace_size = pdata->ftrace_size;
@@ -935,7 +920,16 @@ static void __init ramoops_register_dummy(void)
 	pdata.console_size = ramoops_console_size;
 	pdata.ftrace_size = ramoops_ftrace_size;
 	pdata.pmsg_size = ramoops_pmsg_size;
-	pdata.dump_oops = dump_oops;
+	/* If "max_reason" is set, its value has priority over "dump_oops". */
+	if (ramoops_max_reason >= 0)
+		pdata.max_reason = ramoops_max_reason;
+	/* Otherwise, if "dump_oops" is set, parse it into "max_reason". */
+	else if (ramoops_dump_oops != -1)
+		pdata.max_reason = ramoops_dump_oops ? KMSG_DUMP_OOPS
+						     : KMSG_DUMP_PANIC;
+	/* And if neither are explicitly set, use the default. */
+	else
+		pdata.max_reason = KMSG_DUMP_OOPS;
 	pdata.flags = RAMOOPS_FLAG_FTRACE_PER_CPU;
 
 	/*
@@ -950,7 +944,6 @@ static void __init ramoops_register_dummy(void)
 		pr_info("could not create platform device: %ld\n",
 			PTR_ERR(dummy));
 		dummy = NULL;
-		ramoops_unregister_dummy();
 	}
 }
 
diff --git a/fs/pstore/ram_core.c b/fs/pstore/ram_core.c
index 8823f65888f0..aa8e0b65ff1a 100644
--- a/fs/pstore/ram_core.c
+++ b/fs/pstore/ram_core.c
@@ -34,7 +34,7 @@ struct persistent_ram_buffer {
 	uint32_t    sig;
 	atomic_t    start;
 	atomic_t    size;
-	uint8_t     data[0];
+	uint8_t     data[];
 };
 
 #define PERSISTENT_RAM_SIG (0x43474244) /* DBGC */
@@ -283,7 +283,7 @@ static int notrace persistent_ram_update_user(struct persistent_ram_zone *prz,
 	const void __user *s, unsigned int start, unsigned int count)
 {
 	struct persistent_ram_buffer *buffer = prz->buffer;
-	int ret = unlikely(__copy_from_user(buffer->data + start, s, count)) ?
+	int ret = unlikely(copy_from_user(buffer->data + start, s, count)) ?
 		-EFAULT : 0;
 	persistent_ram_update_ecc(prz, start, count);
 	return ret;
@@ -348,8 +348,6 @@ int notrace persistent_ram_write_user(struct persistent_ram_zone *prz,
 	int rem, ret = 0, c = count;
 	size_t start;
 
-	if (unlikely(!access_ok(s, count)))
-		return -EFAULT;
 	if (unlikely(c > prz->buffer_size)) {
 		s += c - prz->buffer_size;
 		c = prz->buffer_size;
@@ -574,7 +572,7 @@ struct persistent_ram_zone *persistent_ram_new(phys_addr_t start, size_t size,
 	/* Initialize general buffer state. */
 	raw_spin_lock_init(&prz->buffer_lock);
 	prz->flags = flags;
-	prz->label = label;
+	prz->label = kstrdup(label, GFP_KERNEL);
 
 	ret = persistent_ram_buffer_map(start, size, prz, memtype);
 	if (ret)
diff --git a/fs/pstore/zone.c b/fs/pstore/zone.c
new file mode 100644
index 000000000000..819428dfa32f
--- /dev/null
+++ b/fs/pstore/zone.c
@@ -0,0 +1,1465 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Provide a pstore intermediate backend, organized into kernel memory
+ * allocated zones that are then mapped and flushed into a single
+ * contiguous region on a storage backend of some kind (block, mtd, etc).
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/mount.h>
+#include <linux/printk.h>
+#include <linux/fs.h>
+#include <linux/pstore_zone.h>
+#include <linux/kdev_t.h>
+#include <linux/device.h>
+#include <linux/namei.h>
+#include <linux/fcntl.h>
+#include <linux/uio.h>
+#include <linux/writeback.h>
+#include "internal.h"
+
+/**
+ * struct psz_head - header of zone to flush to storage
+ *
+ * @sig: signature to indicate header (PSZ_SIG xor PSZONE-type value)
+ * @datalen: length of data in @data
+ * @start: offset into @data where the beginning of the stored bytes begin
+ * @data: zone data.
+ */
+struct psz_buffer {
+#define PSZ_SIG (0x43474244) /* DBGC */
+	uint32_t sig;
+	atomic_t datalen;
+	atomic_t start;
+	uint8_t data[];
+};
+
+/**
+ * struct psz_kmsg_header - kmsg dump-specific header to flush to storage
+ *
+ * @magic: magic num for kmsg dump header
+ * @time: kmsg dump trigger time
+ * @compressed: whether conpressed
+ * @counter: kmsg dump counter
+ * @reason: the kmsg dump reason (e.g. oops, panic, etc)
+ * @data: pointer to log data
+ *
+ * This is a sub-header for a kmsg dump, trailing after &psz_buffer.
+ */
+struct psz_kmsg_header {
+#define PSTORE_KMSG_HEADER_MAGIC 0x4dfc3ae5 /* Just a random number */
+	uint32_t magic;
+	struct timespec64 time;
+	bool compressed;
+	uint32_t counter;
+	enum kmsg_dump_reason reason;
+	uint8_t data[];
+};
+
+/**
+ * struct pstore_zone - single stored buffer
+ *
+ * @off: zone offset of storage
+ * @type: front-end type for this zone
+ * @name: front-end name for this zone
+ * @buffer: pointer to data buffer managed by this zone
+ * @oldbuf: pointer to old data buffer
+ * @buffer_size: bytes in @buffer->data
+ * @should_recover: whether this zone should recover from storage
+ * @dirty: whether the data in @buffer dirty
+ *
+ * zone structure in memory.
+ */
+struct pstore_zone {
+	loff_t off;
+	const char *name;
+	enum pstore_type_id type;
+
+	struct psz_buffer *buffer;
+	struct psz_buffer *oldbuf;
+	size_t buffer_size;
+	bool should_recover;
+	atomic_t dirty;
+};
+
+/**
+ * struct psz_context - all about running state of pstore/zone
+ *
+ * @kpszs: kmsg dump storage zones
+ * @ppsz: pmsg storage zone
+ * @cpsz: console storage zone
+ * @fpszs: ftrace storage zones
+ * @kmsg_max_cnt: max count of @kpszs
+ * @kmsg_read_cnt: counter of total read kmsg dumps
+ * @kmsg_write_cnt: counter of total kmsg dump writes
+ * @pmsg_read_cnt: counter of total read pmsg zone
+ * @console_read_cnt: counter of total read console zone
+ * @ftrace_max_cnt: max count of @fpszs
+ * @ftrace_read_cnt: counter of max read ftrace zone
+ * @oops_counter: counter of oops dumps
+ * @panic_counter: counter of panic dumps
+ * @recovered: whether finished recovering data from storage
+ * @on_panic: whether panic is happening
+ * @pstore_zone_info_lock: lock to @pstore_zone_info
+ * @pstore_zone_info: information from backend
+ * @pstore: structure for pstore
+ */
+struct psz_context {
+	struct pstore_zone **kpszs;
+	struct pstore_zone *ppsz;
+	struct pstore_zone *cpsz;
+	struct pstore_zone **fpszs;
+	unsigned int kmsg_max_cnt;
+	unsigned int kmsg_read_cnt;
+	unsigned int kmsg_write_cnt;
+	unsigned int pmsg_read_cnt;
+	unsigned int console_read_cnt;
+	unsigned int ftrace_max_cnt;
+	unsigned int ftrace_read_cnt;
+	/*
+	 * These counters should be calculated during recovery.
+	 * It records the oops/panic times after crashes rather than boots.
+	 */
+	unsigned int oops_counter;
+	unsigned int panic_counter;
+	atomic_t recovered;
+	atomic_t on_panic;
+
+	/*
+	 * pstore_zone_info_lock protects this entire structure during calls
+	 * to register_pstore_zone()/unregister_pstore_zone().
+	 */
+	struct mutex pstore_zone_info_lock;
+	struct pstore_zone_info *pstore_zone_info;
+	struct pstore_info pstore;
+};
+static struct psz_context pstore_zone_cxt;
+
+static void psz_flush_all_dirty_zones(struct work_struct *);
+static DECLARE_DELAYED_WORK(psz_cleaner, psz_flush_all_dirty_zones);
+
+/**
+ * enum psz_flush_mode - flush mode for psz_zone_write()
+ *
+ * @FLUSH_NONE: do not flush to storage but update data on memory
+ * @FLUSH_PART: just flush part of data including meta data to storage
+ * @FLUSH_META: just flush meta data of zone to storage
+ * @FLUSH_ALL: flush all of zone
+ */
+enum psz_flush_mode {
+	FLUSH_NONE = 0,
+	FLUSH_PART,
+	FLUSH_META,
+	FLUSH_ALL,
+};
+
+static inline int buffer_datalen(struct pstore_zone *zone)
+{
+	return atomic_read(&zone->buffer->datalen);
+}
+
+static inline int buffer_start(struct pstore_zone *zone)
+{
+	return atomic_read(&zone->buffer->start);
+}
+
+static inline bool is_on_panic(void)
+{
+	return atomic_read(&pstore_zone_cxt.on_panic);
+}
+
+static ssize_t psz_zone_read_buffer(struct pstore_zone *zone, char *buf,
+		size_t len, unsigned long off)
+{
+	if (!buf || !zone || !zone->buffer)
+		return -EINVAL;
+	if (off > zone->buffer_size)
+		return -EINVAL;
+	len = min_t(size_t, len, zone->buffer_size - off);
+	memcpy(buf, zone->buffer->data + off, len);
+	return len;
+}
+
+static int psz_zone_read_oldbuf(struct pstore_zone *zone, char *buf,
+		size_t len, unsigned long off)
+{
+	if (!buf || !zone || !zone->oldbuf)
+		return -EINVAL;
+	if (off > zone->buffer_size)
+		return -EINVAL;
+	len = min_t(size_t, len, zone->buffer_size - off);
+	memcpy(buf, zone->oldbuf->data + off, len);
+	return 0;
+}
+
+static int psz_zone_write(struct pstore_zone *zone,
+		enum psz_flush_mode flush_mode, const char *buf,
+		size_t len, unsigned long off)
+{
+	struct pstore_zone_info *info = pstore_zone_cxt.pstore_zone_info;
+	ssize_t wcnt = 0;
+	ssize_t (*writeop)(const char *buf, size_t bytes, loff_t pos);
+	size_t wlen;
+
+	if (off > zone->buffer_size)
+		return -EINVAL;
+
+	wlen = min_t(size_t, len, zone->buffer_size - off);
+	if (buf && wlen) {
+		memcpy(zone->buffer->data + off, buf, wlen);
+		atomic_set(&zone->buffer->datalen, wlen + off);
+	}
+
+	/* avoid to damage old records */
+	if (!is_on_panic() && !atomic_read(&pstore_zone_cxt.recovered))
+		goto dirty;
+
+	writeop = is_on_panic() ? info->panic_write : info->write;
+	if (!writeop)
+		goto dirty;
+
+	switch (flush_mode) {
+	case FLUSH_NONE:
+		if (unlikely(buf && wlen))
+			goto dirty;
+		return 0;
+	case FLUSH_PART:
+		wcnt = writeop((const char *)zone->buffer->data + off, wlen,
+				zone->off + sizeof(*zone->buffer) + off);
+		if (wcnt != wlen)
+			goto dirty;
+		fallthrough;
+	case FLUSH_META:
+		wlen = sizeof(struct psz_buffer);
+		wcnt = writeop((const char *)zone->buffer, wlen, zone->off);
+		if (wcnt != wlen)
+			goto dirty;
+		break;
+	case FLUSH_ALL:
+		wlen = zone->buffer_size + sizeof(*zone->buffer);
+		wcnt = writeop((const char *)zone->buffer, wlen, zone->off);
+		if (wcnt != wlen)
+			goto dirty;
+		break;
+	}
+
+	return 0;
+dirty:
+	/* no need to mark dirty if going to try next zone */
+	if (wcnt == -ENOMSG)
+		return -ENOMSG;
+	atomic_set(&zone->dirty, true);
+	/* flush dirty zones nicely */
+	if (wcnt == -EBUSY && !is_on_panic())
+		schedule_delayed_work(&psz_cleaner, msecs_to_jiffies(500));
+	return -EBUSY;
+}
+
+static int psz_flush_dirty_zone(struct pstore_zone *zone)
+{
+	int ret;
+
+	if (unlikely(!zone))
+		return -EINVAL;
+
+	if (unlikely(!atomic_read(&pstore_zone_cxt.recovered)))
+		return -EBUSY;
+
+	if (!atomic_xchg(&zone->dirty, false))
+		return 0;
+
+	ret = psz_zone_write(zone, FLUSH_ALL, NULL, 0, 0);
+	if (ret)
+		atomic_set(&zone->dirty, true);
+	return ret;
+}
+
+static int psz_flush_dirty_zones(struct pstore_zone **zones, unsigned int cnt)
+{
+	int i, ret;
+	struct pstore_zone *zone;
+
+	if (!zones)
+		return -EINVAL;
+
+	for (i = 0; i < cnt; i++) {
+		zone = zones[i];
+		if (!zone)
+			return -EINVAL;
+		ret = psz_flush_dirty_zone(zone);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+static int psz_move_zone(struct pstore_zone *old, struct pstore_zone *new)
+{
+	const char *data = (const char *)old->buffer->data;
+	int ret;
+
+	ret = psz_zone_write(new, FLUSH_ALL, data, buffer_datalen(old), 0);
+	if (ret) {
+		atomic_set(&new->buffer->datalen, 0);
+		atomic_set(&new->dirty, false);
+		return ret;
+	}
+	atomic_set(&old->buffer->datalen, 0);
+	return 0;
+}
+
+static void psz_flush_all_dirty_zones(struct work_struct *work)
+{
+	struct psz_context *cxt = &pstore_zone_cxt;
+	int ret = 0;
+
+	if (cxt->ppsz)
+		ret |= psz_flush_dirty_zone(cxt->ppsz);
+	if (cxt->cpsz)
+		ret |= psz_flush_dirty_zone(cxt->cpsz);
+	if (cxt->kpszs)
+		ret |= psz_flush_dirty_zones(cxt->kpszs, cxt->kmsg_max_cnt);
+	if (cxt->fpszs)
+		ret |= psz_flush_dirty_zones(cxt->fpszs, cxt->ftrace_max_cnt);
+	if (ret && cxt->pstore_zone_info)
+		schedule_delayed_work(&psz_cleaner, msecs_to_jiffies(1000));
+}
+
+static int psz_kmsg_recover_data(struct psz_context *cxt)
+{
+	struct pstore_zone_info *info = cxt->pstore_zone_info;
+	struct pstore_zone *zone = NULL;
+	struct psz_buffer *buf;
+	unsigned long i;
+	ssize_t rcnt;
+
+	if (!info->read)
+		return -EINVAL;
+
+	for (i = 0; i < cxt->kmsg_max_cnt; i++) {
+		zone = cxt->kpszs[i];
+		if (unlikely(!zone))
+			return -EINVAL;
+		if (atomic_read(&zone->dirty)) {
+			unsigned int wcnt = cxt->kmsg_write_cnt;
+			struct pstore_zone *new = cxt->kpszs[wcnt];
+			int ret;
+
+			ret = psz_move_zone(zone, new);
+			if (ret) {
+				pr_err("move zone from %lu to %d failed\n",
+						i, wcnt);
+				return ret;
+			}
+			cxt->kmsg_write_cnt = (wcnt + 1) % cxt->kmsg_max_cnt;
+		}
+		if (!zone->should_recover)
+			continue;
+		buf = zone->buffer;
+		rcnt = info->read((char *)buf, zone->buffer_size + sizeof(*buf),
+				zone->off);
+		if (rcnt != zone->buffer_size + sizeof(*buf))
+			return (int)rcnt < 0 ? (int)rcnt : -EIO;
+	}
+	return 0;
+}
+
+static int psz_kmsg_recover_meta(struct psz_context *cxt)
+{
+	struct pstore_zone_info *info = cxt->pstore_zone_info;
+	struct pstore_zone *zone;
+	size_t rcnt, len;
+	struct psz_buffer *buf;
+	struct psz_kmsg_header *hdr;
+	struct timespec64 time = { };
+	unsigned long i;
+	/*
+	 * Recover may on panic, we can't allocate any memory by kmalloc.
+	 * So, we use local array instead.
+	 */
+	char buffer_header[sizeof(*buf) + sizeof(*hdr)] = {0};
+
+	if (!info->read)
+		return -EINVAL;
+
+	len = sizeof(*buf) + sizeof(*hdr);
+	buf = (struct psz_buffer *)buffer_header;
+	for (i = 0; i < cxt->kmsg_max_cnt; i++) {
+		zone = cxt->kpszs[i];
+		if (unlikely(!zone))
+			return -EINVAL;
+
+		rcnt = info->read((char *)buf, len, zone->off);
+		if (rcnt == -ENOMSG) {
+			pr_debug("%s with id %lu may be broken, skip\n",
+					zone->name, i);
+			continue;
+		} else if (rcnt != len) {
+			pr_err("read %s with id %lu failed\n", zone->name, i);
+			return (int)rcnt < 0 ? (int)rcnt : -EIO;
+		}
+
+		if (buf->sig != zone->buffer->sig) {
+			pr_debug("no valid data in kmsg dump zone %lu\n", i);
+			continue;
+		}
+
+		if (zone->buffer_size < atomic_read(&buf->datalen)) {
+			pr_info("found overtop zone: %s: id %lu, off %lld, size %zu\n",
+					zone->name, i, zone->off,
+					zone->buffer_size);
+			continue;
+		}
+
+		hdr = (struct psz_kmsg_header *)buf->data;
+		if (hdr->magic != PSTORE_KMSG_HEADER_MAGIC) {
+			pr_info("found invalid zone: %s: id %lu, off %lld, size %zu\n",
+					zone->name, i, zone->off,
+					zone->buffer_size);
+			continue;
+		}
+
+		/*
+		 * we get the newest zone, and the next one must be the oldest
+		 * or unused zone, because we do write one by one like a circle.
+		 */
+		if (hdr->time.tv_sec >= time.tv_sec) {
+			time.tv_sec = hdr->time.tv_sec;
+			cxt->kmsg_write_cnt = (i + 1) % cxt->kmsg_max_cnt;
+		}
+
+		if (hdr->reason == KMSG_DUMP_OOPS)
+			cxt->oops_counter =
+				max(cxt->oops_counter, hdr->counter);
+		else if (hdr->reason == KMSG_DUMP_PANIC)
+			cxt->panic_counter =
+				max(cxt->panic_counter, hdr->counter);
+
+		if (!atomic_read(&buf->datalen)) {
+			pr_debug("found erased zone: %s: id %lu, off %lld, size %zu, datalen %d\n",
+					zone->name, i, zone->off,
+					zone->buffer_size,
+					atomic_read(&buf->datalen));
+			continue;
+		}
+
+		if (!is_on_panic())
+			zone->should_recover = true;
+		pr_debug("found nice zone: %s: id %lu, off %lld, size %zu, datalen %d\n",
+				zone->name, i, zone->off,
+				zone->buffer_size, atomic_read(&buf->datalen));
+	}
+
+	return 0;
+}
+
+static int psz_kmsg_recover(struct psz_context *cxt)
+{
+	int ret;
+
+	if (!cxt->kpszs)
+		return 0;
+
+	ret = psz_kmsg_recover_meta(cxt);
+	if (ret)
+		goto recover_fail;
+
+	ret = psz_kmsg_recover_data(cxt);
+	if (ret)
+		goto recover_fail;
+
+	return 0;
+recover_fail:
+	pr_debug("psz_recover_kmsg failed\n");
+	return ret;
+}
+
+static int psz_recover_zone(struct psz_context *cxt, struct pstore_zone *zone)
+{
+	struct pstore_zone_info *info = cxt->pstore_zone_info;
+	struct psz_buffer *oldbuf, tmpbuf;
+	int ret = 0;
+	char *buf;
+	ssize_t rcnt, len, start, off;
+
+	if (!zone || zone->oldbuf)
+		return 0;
+
+	if (is_on_panic()) {
+		/* save data as much as possible */
+		psz_flush_dirty_zone(zone);
+		return 0;
+	}
+
+	if (unlikely(!info->read))
+		return -EINVAL;
+
+	len = sizeof(struct psz_buffer);
+	rcnt = info->read((char *)&tmpbuf, len, zone->off);
+	if (rcnt != len) {
+		pr_debug("read zone %s failed\n", zone->name);
+		return (int)rcnt < 0 ? (int)rcnt : -EIO;
+	}
+
+	if (tmpbuf.sig != zone->buffer->sig) {
+		pr_debug("no valid data in zone %s\n", zone->name);
+		return 0;
+	}
+
+	if (zone->buffer_size < atomic_read(&tmpbuf.datalen) ||
+		zone->buffer_size < atomic_read(&tmpbuf.start)) {
+		pr_info("found overtop zone: %s: off %lld, size %zu\n",
+				zone->name, zone->off, zone->buffer_size);
+		/* just keep going */
+		return 0;
+	}
+
+	if (!atomic_read(&tmpbuf.datalen)) {
+		pr_debug("found erased zone: %s: off %lld, size %zu, datalen %d\n",
+				zone->name, zone->off, zone->buffer_size,
+				atomic_read(&tmpbuf.datalen));
+		return 0;
+	}
+
+	pr_debug("found nice zone: %s: off %lld, size %zu, datalen %d\n",
+			zone->name, zone->off, zone->buffer_size,
+			atomic_read(&tmpbuf.datalen));
+
+	len = atomic_read(&tmpbuf.datalen) + sizeof(*oldbuf);
+	oldbuf = kzalloc(len, GFP_KERNEL);
+	if (!oldbuf)
+		return -ENOMEM;
+
+	memcpy(oldbuf, &tmpbuf, sizeof(*oldbuf));
+	buf = (char *)oldbuf + sizeof(*oldbuf);
+	len = atomic_read(&oldbuf->datalen);
+	start = atomic_read(&oldbuf->start);
+	off = zone->off + sizeof(*oldbuf);
+
+	/* get part of data */
+	rcnt = info->read(buf, len - start, off + start);
+	if (rcnt != len - start) {
+		pr_err("read zone %s failed\n", zone->name);
+		ret = (int)rcnt < 0 ? (int)rcnt : -EIO;
+		goto free_oldbuf;
+	}
+
+	/* get the rest of data */
+	rcnt = info->read(buf + len - start, start, off);
+	if (rcnt != start) {
+		pr_err("read zone %s failed\n", zone->name);
+		ret = (int)rcnt < 0 ? (int)rcnt : -EIO;
+		goto free_oldbuf;
+	}
+
+	zone->oldbuf = oldbuf;
+	psz_flush_dirty_zone(zone);
+	return 0;
+
+free_oldbuf:
+	kfree(oldbuf);
+	return ret;
+}
+
+static int psz_recover_zones(struct psz_context *cxt,
+		struct pstore_zone **zones, unsigned int cnt)
+{
+	int ret;
+	unsigned int i;
+	struct pstore_zone *zone;
+
+	if (!zones)
+		return 0;
+
+	for (i = 0; i < cnt; i++) {
+		zone = zones[i];
+		if (unlikely(!zone))
+			continue;
+		ret = psz_recover_zone(cxt, zone);
+		if (ret)
+			goto recover_fail;
+	}
+
+	return 0;
+recover_fail:
+	pr_debug("recover %s[%u] failed\n", zone->name, i);
+	return ret;
+}
+
+/**
+ * psz_recovery() - recover data from storage
+ * @cxt: the context of pstore/zone
+ *
+ * recovery means reading data back from storage after rebooting
+ *
+ * Return: 0 on success, others on failure.
+ */
+static inline int psz_recovery(struct psz_context *cxt)
+{
+	int ret;
+
+	if (atomic_read(&cxt->recovered))
+		return 0;
+
+	ret = psz_kmsg_recover(cxt);
+	if (ret)
+		goto out;
+
+	ret = psz_recover_zone(cxt, cxt->ppsz);
+	if (ret)
+		goto out;
+
+	ret = psz_recover_zone(cxt, cxt->cpsz);
+	if (ret)
+		goto out;
+
+	ret = psz_recover_zones(cxt, cxt->fpszs, cxt->ftrace_max_cnt);
+
+out:
+	if (unlikely(ret))
+		pr_err("recover failed\n");
+	else {
+		pr_debug("recover end!\n");
+		atomic_set(&cxt->recovered, 1);
+	}
+	return ret;
+}
+
+static int psz_pstore_open(struct pstore_info *psi)
+{
+	struct psz_context *cxt = psi->data;
+
+	cxt->kmsg_read_cnt = 0;
+	cxt->pmsg_read_cnt = 0;
+	cxt->console_read_cnt = 0;
+	cxt->ftrace_read_cnt = 0;
+	return 0;
+}
+
+static inline bool psz_old_ok(struct pstore_zone *zone)
+{
+	if (zone && zone->oldbuf && atomic_read(&zone->oldbuf->datalen))
+		return true;
+	return false;
+}
+
+static inline bool psz_ok(struct pstore_zone *zone)
+{
+	if (zone && zone->buffer && buffer_datalen(zone))
+		return true;
+	return false;
+}
+
+static inline int psz_kmsg_erase(struct psz_context *cxt,
+		struct pstore_zone *zone, struct pstore_record *record)
+{
+	struct psz_buffer *buffer = zone->buffer;
+	struct psz_kmsg_header *hdr =
+		(struct psz_kmsg_header *)buffer->data;
+	size_t size;
+
+	if (unlikely(!psz_ok(zone)))
+		return 0;
+
+	/* this zone is already updated, no need to erase */
+	if (record->count != hdr->counter)
+		return 0;
+
+	size = buffer_datalen(zone) + sizeof(*zone->buffer);
+	atomic_set(&zone->buffer->datalen, 0);
+	if (cxt->pstore_zone_info->erase)
+		return cxt->pstore_zone_info->erase(size, zone->off);
+	else
+		return psz_zone_write(zone, FLUSH_META, NULL, 0, 0);
+}
+
+static inline int psz_record_erase(struct psz_context *cxt,
+		struct pstore_zone *zone)
+{
+	if (unlikely(!psz_old_ok(zone)))
+		return 0;
+
+	kfree(zone->oldbuf);
+	zone->oldbuf = NULL;
+	/*
+	 * if there are new data in zone buffer, that means the old data
+	 * are already invalid. It is no need to flush 0 (erase) to
+	 * block device.
+	 */
+	if (!buffer_datalen(zone))
+		return psz_zone_write(zone, FLUSH_META, NULL, 0, 0);
+	psz_flush_dirty_zone(zone);
+	return 0;
+}
+
+static int psz_pstore_erase(struct pstore_record *record)
+{
+	struct psz_context *cxt = record->psi->data;
+
+	switch (record->type) {
+	case PSTORE_TYPE_DMESG:
+		if (record->id >= cxt->kmsg_max_cnt)
+			return -EINVAL;
+		return psz_kmsg_erase(cxt, cxt->kpszs[record->id], record);
+	case PSTORE_TYPE_PMSG:
+		return psz_record_erase(cxt, cxt->ppsz);
+	case PSTORE_TYPE_CONSOLE:
+		return psz_record_erase(cxt, cxt->cpsz);
+	case PSTORE_TYPE_FTRACE:
+		if (record->id >= cxt->ftrace_max_cnt)
+			return -EINVAL;
+		return psz_record_erase(cxt, cxt->fpszs[record->id]);
+	default: return -EINVAL;
+	}
+}
+
+static void psz_write_kmsg_hdr(struct pstore_zone *zone,
+		struct pstore_record *record)
+{
+	struct psz_context *cxt = record->psi->data;
+	struct psz_buffer *buffer = zone->buffer;
+	struct psz_kmsg_header *hdr =
+		(struct psz_kmsg_header *)buffer->data;
+
+	hdr->magic = PSTORE_KMSG_HEADER_MAGIC;
+	hdr->compressed = record->compressed;
+	hdr->time.tv_sec = record->time.tv_sec;
+	hdr->time.tv_nsec = record->time.tv_nsec;
+	hdr->reason = record->reason;
+	if (hdr->reason == KMSG_DUMP_OOPS)
+		hdr->counter = ++cxt->oops_counter;
+	else if (hdr->reason == KMSG_DUMP_PANIC)
+		hdr->counter = ++cxt->panic_counter;
+	else
+		hdr->counter = 0;
+}
+
+/*
+ * In case zone is broken, which may occur to MTD device, we try each zones,
+ * start at cxt->kmsg_write_cnt.
+ */
+static inline int notrace psz_kmsg_write_record(struct psz_context *cxt,
+		struct pstore_record *record)
+{
+	size_t size, hlen;
+	struct pstore_zone *zone;
+	unsigned int i;
+
+	for (i = 0; i < cxt->kmsg_max_cnt; i++) {
+		unsigned int zonenum, len;
+		int ret;
+
+		zonenum = (cxt->kmsg_write_cnt + i) % cxt->kmsg_max_cnt;
+		zone = cxt->kpszs[zonenum];
+		if (unlikely(!zone))
+			return -ENOSPC;
+
+		/* avoid destroying old data, allocate a new one */
+		len = zone->buffer_size + sizeof(*zone->buffer);
+		zone->oldbuf = zone->buffer;
+		zone->buffer = kzalloc(len, GFP_KERNEL);
+		if (!zone->buffer) {
+			zone->buffer = zone->oldbuf;
+			return -ENOMEM;
+		}
+		zone->buffer->sig = zone->oldbuf->sig;
+
+		pr_debug("write %s to zone id %d\n", zone->name, zonenum);
+		psz_write_kmsg_hdr(zone, record);
+		hlen = sizeof(struct psz_kmsg_header);
+		size = min_t(size_t, record->size, zone->buffer_size - hlen);
+		ret = psz_zone_write(zone, FLUSH_ALL, record->buf, size, hlen);
+		if (likely(!ret || ret != -ENOMSG)) {
+			cxt->kmsg_write_cnt = zonenum + 1;
+			cxt->kmsg_write_cnt %= cxt->kmsg_max_cnt;
+			/* no need to try next zone, free last zone buffer */
+			kfree(zone->oldbuf);
+			zone->oldbuf = NULL;
+			return ret;
+		}
+
+		pr_debug("zone %u may be broken, try next dmesg zone\n",
+				zonenum);
+		kfree(zone->buffer);
+		zone->buffer = zone->oldbuf;
+		zone->oldbuf = NULL;
+	}
+
+	return -EBUSY;
+}
+
+static int notrace psz_kmsg_write(struct psz_context *cxt,
+		struct pstore_record *record)
+{
+	int ret;
+
+	/*
+	 * Explicitly only take the first part of any new crash.
+	 * If our buffer is larger than kmsg_bytes, this can never happen,
+	 * and if our buffer is smaller than kmsg_bytes, we don't want the
+	 * report split across multiple records.
+	 */
+	if (record->part != 1)
+		return -ENOSPC;
+
+	if (!cxt->kpszs)
+		return -ENOSPC;
+
+	ret = psz_kmsg_write_record(cxt, record);
+	if (!ret && is_on_panic()) {
+		/* ensure all data are flushed to storage when panic */
+		pr_debug("try to flush other dirty zones\n");
+		psz_flush_all_dirty_zones(NULL);
+	}
+
+	/* always return 0 as we had handled it on buffer */
+	return 0;
+}
+
+static int notrace psz_record_write(struct pstore_zone *zone,
+		struct pstore_record *record)
+{
+	size_t start, rem;
+	bool is_full_data = false;
+	char *buf;
+	int cnt;
+
+	if (!zone || !record)
+		return -ENOSPC;
+
+	if (atomic_read(&zone->buffer->datalen) >= zone->buffer_size)
+		is_full_data = true;
+
+	cnt = record->size;
+	buf = record->buf;
+	if (unlikely(cnt > zone->buffer_size)) {
+		buf += cnt - zone->buffer_size;
+		cnt = zone->buffer_size;
+	}
+
+	start = buffer_start(zone);
+	rem = zone->buffer_size - start;
+	if (unlikely(rem < cnt)) {
+		psz_zone_write(zone, FLUSH_PART, buf, rem, start);
+		buf += rem;
+		cnt -= rem;
+		start = 0;
+		is_full_data = true;
+	}
+
+	atomic_set(&zone->buffer->start, cnt + start);
+	psz_zone_write(zone, FLUSH_PART, buf, cnt, start);
+
+	/**
+	 * psz_zone_write will set datalen as start + cnt.
+	 * It work if actual data length lesser than buffer size.
+	 * If data length greater than buffer size, pmsg will rewrite to
+	 * beginning of zone, which make buffer->datalen wrongly.
+	 * So we should reset datalen as buffer size once actual data length
+	 * greater than buffer size.
+	 */
+	if (is_full_data) {
+		atomic_set(&zone->buffer->datalen, zone->buffer_size);
+		psz_zone_write(zone, FLUSH_META, NULL, 0, 0);
+	}
+	return 0;
+}
+
+static int notrace psz_pstore_write(struct pstore_record *record)
+{
+	struct psz_context *cxt = record->psi->data;
+
+	if (record->type == PSTORE_TYPE_DMESG &&
+			record->reason == KMSG_DUMP_PANIC)
+		atomic_set(&cxt->on_panic, 1);
+
+	/*
+	 * if on panic, do not write except panic records
+	 * Fix case that panic_write prints log which wakes up console backend.
+	 */
+	if (is_on_panic() && record->type != PSTORE_TYPE_DMESG)
+		return -EBUSY;
+
+	switch (record->type) {
+	case PSTORE_TYPE_DMESG:
+		return psz_kmsg_write(cxt, record);
+	case PSTORE_TYPE_CONSOLE:
+		return psz_record_write(cxt->cpsz, record);
+	case PSTORE_TYPE_PMSG:
+		return psz_record_write(cxt->ppsz, record);
+	case PSTORE_TYPE_FTRACE: {
+		int zonenum = smp_processor_id();
+
+		if (!cxt->fpszs)
+			return -ENOSPC;
+		return psz_record_write(cxt->fpszs[zonenum], record);
+	}
+	default:
+		return -EINVAL;
+	}
+}
+
+static struct pstore_zone *psz_read_next_zone(struct psz_context *cxt)
+{
+	struct pstore_zone *zone = NULL;
+
+	while (cxt->kmsg_read_cnt < cxt->kmsg_max_cnt) {
+		zone = cxt->kpszs[cxt->kmsg_read_cnt++];
+		if (psz_ok(zone))
+			return zone;
+	}
+
+	if (cxt->ftrace_read_cnt < cxt->ftrace_max_cnt)
+		/*
+		 * No need psz_old_ok(). Let psz_ftrace_read() do so for
+		 * combination. psz_ftrace_read() should traverse over
+		 * all zones in case of some zone without data.
+		 */
+		return cxt->fpszs[cxt->ftrace_read_cnt++];
+
+	if (cxt->pmsg_read_cnt == 0) {
+		cxt->pmsg_read_cnt++;
+		zone = cxt->ppsz;
+		if (psz_old_ok(zone))
+			return zone;
+	}
+
+	if (cxt->console_read_cnt == 0) {
+		cxt->console_read_cnt++;
+		zone = cxt->cpsz;
+		if (psz_old_ok(zone))
+			return zone;
+	}
+
+	return NULL;
+}
+
+static int psz_kmsg_read_hdr(struct pstore_zone *zone,
+		struct pstore_record *record)
+{
+	struct psz_buffer *buffer = zone->buffer;
+	struct psz_kmsg_header *hdr =
+		(struct psz_kmsg_header *)buffer->data;
+
+	if (hdr->magic != PSTORE_KMSG_HEADER_MAGIC)
+		return -EINVAL;
+	record->compressed = hdr->compressed;
+	record->time.tv_sec = hdr->time.tv_sec;
+	record->time.tv_nsec = hdr->time.tv_nsec;
+	record->reason = hdr->reason;
+	record->count = hdr->counter;
+	return 0;
+}
+
+static ssize_t psz_kmsg_read(struct pstore_zone *zone,
+		struct pstore_record *record)
+{
+	ssize_t size, hlen = 0;
+
+	size = buffer_datalen(zone);
+	/* Clear and skip this kmsg dump record if it has no valid header */
+	if (psz_kmsg_read_hdr(zone, record)) {
+		atomic_set(&zone->buffer->datalen, 0);
+		atomic_set(&zone->dirty, 0);
+		return -ENOMSG;
+	}
+	size -= sizeof(struct psz_kmsg_header);
+
+	if (!record->compressed) {
+		char *buf = kasprintf(GFP_KERNEL, "%s: Total %d times\n",
+				      kmsg_dump_reason_str(record->reason),
+				      record->count);
+		hlen = strlen(buf);
+		record->buf = krealloc(buf, hlen + size, GFP_KERNEL);
+		if (!record->buf) {
+			kfree(buf);
+			return -ENOMEM;
+		}
+	} else {
+		record->buf = kmalloc(size, GFP_KERNEL);
+		if (!record->buf)
+			return -ENOMEM;
+	}
+
+	size = psz_zone_read_buffer(zone, record->buf + hlen, size,
+			sizeof(struct psz_kmsg_header));
+	if (unlikely(size < 0)) {
+		kfree(record->buf);
+		return -ENOMSG;
+	}
+
+	return size + hlen;
+}
+
+/* try to combine all ftrace zones */
+static ssize_t psz_ftrace_read(struct pstore_zone *zone,
+		struct pstore_record *record)
+{
+	struct psz_context *cxt;
+	struct psz_buffer *buf;
+	int ret;
+
+	if (!zone || !record)
+		return -ENOSPC;
+
+	if (!psz_old_ok(zone))
+		goto out;
+
+	buf = (struct psz_buffer *)zone->oldbuf;
+	if (!buf)
+		return -ENOMSG;
+
+	ret = pstore_ftrace_combine_log(&record->buf, &record->size,
+			(char *)buf->data, atomic_read(&buf->datalen));
+	if (unlikely(ret))
+		return ret;
+
+out:
+	cxt = record->psi->data;
+	if (cxt->ftrace_read_cnt < cxt->ftrace_max_cnt)
+		/* then, read next ftrace zone */
+		return -ENOMSG;
+	record->id = 0;
+	return record->size ? record->size : -ENOMSG;
+}
+
+static ssize_t psz_record_read(struct pstore_zone *zone,
+		struct pstore_record *record)
+{
+	size_t len;
+	struct psz_buffer *buf;
+
+	if (!zone || !record)
+		return -ENOSPC;
+
+	buf = (struct psz_buffer *)zone->oldbuf;
+	if (!buf)
+		return -ENOMSG;
+
+	len = atomic_read(&buf->datalen);
+	record->buf = kmalloc(len, GFP_KERNEL);
+	if (!record->buf)
+		return -ENOMEM;
+
+	if (unlikely(psz_zone_read_oldbuf(zone, record->buf, len, 0))) {
+		kfree(record->buf);
+		return -ENOMSG;
+	}
+
+	return len;
+}
+
+static ssize_t psz_pstore_read(struct pstore_record *record)
+{
+	struct psz_context *cxt = record->psi->data;
+	ssize_t (*readop)(struct pstore_zone *zone,
+			struct pstore_record *record);
+	struct pstore_zone *zone;
+	ssize_t ret;
+
+	/* before read, we must recover from storage */
+	ret = psz_recovery(cxt);
+	if (ret)
+		return ret;
+
+next_zone:
+	zone = psz_read_next_zone(cxt);
+	if (!zone)
+		return 0;
+
+	record->type = zone->type;
+	switch (record->type) {
+	case PSTORE_TYPE_DMESG:
+		readop = psz_kmsg_read;
+		record->id = cxt->kmsg_read_cnt - 1;
+		break;
+	case PSTORE_TYPE_FTRACE:
+		readop = psz_ftrace_read;
+		break;
+	case PSTORE_TYPE_CONSOLE:
+		fallthrough;
+	case PSTORE_TYPE_PMSG:
+		readop = psz_record_read;
+		break;
+	default:
+		goto next_zone;
+	}
+
+	ret = readop(zone, record);
+	if (ret == -ENOMSG)
+		goto next_zone;
+	return ret;
+}
+
+static struct psz_context pstore_zone_cxt = {
+	.pstore_zone_info_lock =
+		__MUTEX_INITIALIZER(pstore_zone_cxt.pstore_zone_info_lock),
+	.recovered = ATOMIC_INIT(0),
+	.on_panic = ATOMIC_INIT(0),
+	.pstore = {
+		.owner = THIS_MODULE,
+		.open = psz_pstore_open,
+		.read = psz_pstore_read,
+		.write = psz_pstore_write,
+		.erase = psz_pstore_erase,
+	},
+};
+
+static void psz_free_zone(struct pstore_zone **pszone)
+{
+	struct pstore_zone *zone = *pszone;
+
+	if (!zone)
+		return;
+
+	kfree(zone->buffer);
+	kfree(zone);
+	*pszone = NULL;
+}
+
+static void psz_free_zones(struct pstore_zone ***pszones, unsigned int *cnt)
+{
+	struct pstore_zone **zones = *pszones;
+
+	if (!zones)
+		return;
+
+	while (*cnt > 0) {
+		(*cnt)--;
+		psz_free_zone(&(zones[*cnt]));
+	}
+	kfree(zones);
+	*pszones = NULL;
+}
+
+static void psz_free_all_zones(struct psz_context *cxt)
+{
+	if (cxt->kpszs)
+		psz_free_zones(&cxt->kpszs, &cxt->kmsg_max_cnt);
+	if (cxt->ppsz)
+		psz_free_zone(&cxt->ppsz);
+	if (cxt->cpsz)
+		psz_free_zone(&cxt->cpsz);
+	if (cxt->fpszs)
+		psz_free_zones(&cxt->fpszs, &cxt->ftrace_max_cnt);
+}
+
+static struct pstore_zone *psz_init_zone(enum pstore_type_id type,
+		loff_t *off, size_t size)
+{
+	struct pstore_zone_info *info = pstore_zone_cxt.pstore_zone_info;
+	struct pstore_zone *zone;
+	const char *name = pstore_type_to_name(type);
+
+	if (!size)
+		return NULL;
+
+	if (*off + size > info->total_size) {
+		pr_err("no room for %s (0x%zx@0x%llx over 0x%lx)\n",
+			name, size, *off, info->total_size);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	zone = kzalloc(sizeof(struct pstore_zone), GFP_KERNEL);
+	if (!zone)
+		return ERR_PTR(-ENOMEM);
+
+	zone->buffer = kmalloc(size, GFP_KERNEL);
+	if (!zone->buffer) {
+		kfree(zone);
+		return ERR_PTR(-ENOMEM);
+	}
+	memset(zone->buffer, 0xFF, size);
+	zone->off = *off;
+	zone->name = name;
+	zone->type = type;
+	zone->buffer_size = size - sizeof(struct psz_buffer);
+	zone->buffer->sig = type ^ PSZ_SIG;
+	zone->oldbuf = NULL;
+	atomic_set(&zone->dirty, 0);
+	atomic_set(&zone->buffer->datalen, 0);
+	atomic_set(&zone->buffer->start, 0);
+
+	*off += size;
+
+	pr_debug("pszone %s: off 0x%llx, %zu header, %zu data\n", zone->name,
+			zone->off, sizeof(*zone->buffer), zone->buffer_size);
+	return zone;
+}
+
+static struct pstore_zone **psz_init_zones(enum pstore_type_id type,
+	loff_t *off, size_t total_size, ssize_t record_size,
+	unsigned int *cnt)
+{
+	struct pstore_zone_info *info = pstore_zone_cxt.pstore_zone_info;
+	struct pstore_zone **zones, *zone;
+	const char *name = pstore_type_to_name(type);
+	int c, i;
+
+	*cnt = 0;
+	if (!total_size || !record_size)
+		return NULL;
+
+	if (*off + total_size > info->total_size) {
+		pr_err("no room for zones %s (0x%zx@0x%llx over 0x%lx)\n",
+			name, total_size, *off, info->total_size);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	c = total_size / record_size;
+	zones = kcalloc(c, sizeof(*zones), GFP_KERNEL);
+	if (!zones) {
+		pr_err("allocate for zones %s failed\n", name);
+		return ERR_PTR(-ENOMEM);
+	}
+	memset(zones, 0, c * sizeof(*zones));
+
+	for (i = 0; i < c; i++) {
+		zone = psz_init_zone(type, off, record_size);
+		if (!zone || IS_ERR(zone)) {
+			pr_err("initialize zones %s failed\n", name);
+			psz_free_zones(&zones, &i);
+			return (void *)zone;
+		}
+		zones[i] = zone;
+	}
+
+	*cnt = c;
+	return zones;
+}
+
+static int psz_alloc_zones(struct psz_context *cxt)
+{
+	struct pstore_zone_info *info = cxt->pstore_zone_info;
+	loff_t off = 0;
+	int err;
+	size_t off_size = 0;
+
+	off_size += info->pmsg_size;
+	cxt->ppsz = psz_init_zone(PSTORE_TYPE_PMSG, &off, info->pmsg_size);
+	if (IS_ERR(cxt->ppsz)) {
+		err = PTR_ERR(cxt->ppsz);
+		cxt->ppsz = NULL;
+		goto free_out;
+	}
+
+	off_size += info->console_size;
+	cxt->cpsz = psz_init_zone(PSTORE_TYPE_CONSOLE, &off,
+			info->console_size);
+	if (IS_ERR(cxt->cpsz)) {
+		err = PTR_ERR(cxt->cpsz);
+		cxt->cpsz = NULL;
+		goto free_out;
+	}
+
+	off_size += info->ftrace_size;
+	cxt->fpszs = psz_init_zones(PSTORE_TYPE_FTRACE, &off,
+			info->ftrace_size,
+			info->ftrace_size / nr_cpu_ids,
+			&cxt->ftrace_max_cnt);
+	if (IS_ERR(cxt->fpszs)) {
+		err = PTR_ERR(cxt->fpszs);
+		cxt->fpszs = NULL;
+		goto free_out;
+	}
+
+	cxt->kpszs = psz_init_zones(PSTORE_TYPE_DMESG, &off,
+			info->total_size - off_size,
+			info->kmsg_size, &cxt->kmsg_max_cnt);
+	if (IS_ERR(cxt->kpszs)) {
+		err = PTR_ERR(cxt->kpszs);
+		cxt->kpszs = NULL;
+		goto free_out;
+	}
+
+	return 0;
+free_out:
+	psz_free_all_zones(cxt);
+	return err;
+}
+
+/**
+ * register_pstore_zone() - register to pstore/zone
+ *
+ * @info: back-end driver information. See &struct pstore_zone_info.
+ *
+ * Only one back-end at one time.
+ *
+ * Return: 0 on success, others on failure.
+ */
+int register_pstore_zone(struct pstore_zone_info *info)
+{
+	int err = -EINVAL;
+	struct psz_context *cxt = &pstore_zone_cxt;
+
+	if (info->total_size < 4096) {
+		pr_warn("total_size must be >= 4096\n");
+		return -EINVAL;
+	}
+
+	if (!info->kmsg_size && !info->pmsg_size && !info->console_size &&
+	    !info->ftrace_size) {
+		pr_warn("at least one record size must be non-zero\n");
+		return -EINVAL;
+	}
+
+	if (!info->name || !info->name[0])
+		return -EINVAL;
+
+#define check_size(name, size) {					\
+		if (info->name > 0 && info->name < (size)) {		\
+			pr_err(#name " must be over %d\n", (size));	\
+			return -EINVAL;					\
+		}							\
+		if (info->name & (size - 1)) {				\
+			pr_err(#name " must be a multiple of %d\n",	\
+					(size));			\
+			return -EINVAL;					\
+		}							\
+	}
+
+	check_size(total_size, 4096);
+	check_size(kmsg_size, SECTOR_SIZE);
+	check_size(pmsg_size, SECTOR_SIZE);
+	check_size(console_size, SECTOR_SIZE);
+	check_size(ftrace_size, SECTOR_SIZE);
+
+#undef check_size
+
+	/*
+	 * the @read and @write must be applied.
+	 * if no @read, pstore may mount failed.
+	 * if no @write, pstore do not support to remove record file.
+	 */
+	if (!info->read || !info->write) {
+		pr_err("no valid general read/write interface\n");
+		return -EINVAL;
+	}
+
+	mutex_lock(&cxt->pstore_zone_info_lock);
+	if (cxt->pstore_zone_info) {
+		pr_warn("'%s' already loaded: ignoring '%s'\n",
+				cxt->pstore_zone_info->name, info->name);
+		mutex_unlock(&cxt->pstore_zone_info_lock);
+		return -EBUSY;
+	}
+	cxt->pstore_zone_info = info;
+
+	pr_debug("register %s with properties:\n", info->name);
+	pr_debug("\ttotal size : %ld Bytes\n", info->total_size);
+	pr_debug("\tkmsg size : %ld Bytes\n", info->kmsg_size);
+	pr_debug("\tpmsg size : %ld Bytes\n", info->pmsg_size);
+	pr_debug("\tconsole size : %ld Bytes\n", info->console_size);
+	pr_debug("\tftrace size : %ld Bytes\n", info->ftrace_size);
+
+	err = psz_alloc_zones(cxt);
+	if (err) {
+		pr_err("alloc zones failed\n");
+		goto fail_out;
+	}
+
+	if (info->kmsg_size) {
+		cxt->pstore.bufsize = cxt->kpszs[0]->buffer_size -
+			sizeof(struct psz_kmsg_header);
+		cxt->pstore.buf = kzalloc(cxt->pstore.bufsize, GFP_KERNEL);
+		if (!cxt->pstore.buf) {
+			err = -ENOMEM;
+			goto fail_free;
+		}
+	}
+	cxt->pstore.data = cxt;
+
+	pr_info("registered %s as backend for", info->name);
+	cxt->pstore.max_reason = info->max_reason;
+	cxt->pstore.name = info->name;
+	if (info->kmsg_size) {
+		cxt->pstore.flags |= PSTORE_FLAGS_DMESG;
+		pr_cont(" kmsg(%s",
+			kmsg_dump_reason_str(cxt->pstore.max_reason));
+		if (cxt->pstore_zone_info->panic_write)
+			pr_cont(",panic_write");
+		pr_cont(")");
+	}
+	if (info->pmsg_size) {
+		cxt->pstore.flags |= PSTORE_FLAGS_PMSG;
+		pr_cont(" pmsg");
+	}
+	if (info->console_size) {
+		cxt->pstore.flags |= PSTORE_FLAGS_CONSOLE;
+		pr_cont(" console");
+	}
+	if (info->ftrace_size) {
+		cxt->pstore.flags |= PSTORE_FLAGS_FTRACE;
+		pr_cont(" ftrace");
+	}
+	pr_cont("\n");
+
+	err = pstore_register(&cxt->pstore);
+	if (err) {
+		pr_err("registering with pstore failed\n");
+		goto fail_free;
+	}
+	mutex_unlock(&pstore_zone_cxt.pstore_zone_info_lock);
+
+	return 0;
+
+fail_free:
+	kfree(cxt->pstore.buf);
+	cxt->pstore.buf = NULL;
+	cxt->pstore.bufsize = 0;
+	psz_free_all_zones(cxt);
+fail_out:
+	pstore_zone_cxt.pstore_zone_info = NULL;
+	mutex_unlock(&pstore_zone_cxt.pstore_zone_info_lock);
+	return err;
+}
+EXPORT_SYMBOL_GPL(register_pstore_zone);
+
+/**
+ * unregister_pstore_zone() - unregister to pstore/zone
+ *
+ * @info: back-end driver information. See struct pstore_zone_info.
+ */
+void unregister_pstore_zone(struct pstore_zone_info *info)
+{
+	struct psz_context *cxt = &pstore_zone_cxt;
+
+	mutex_lock(&cxt->pstore_zone_info_lock);
+	if (!cxt->pstore_zone_info) {
+		mutex_unlock(&cxt->pstore_zone_info_lock);
+		return;
+	}
+
+	/* Stop incoming writes from pstore. */
+	pstore_unregister(&cxt->pstore);
+
+	/* Flush any pending writes. */
+	psz_flush_all_dirty_zones(NULL);
+	flush_delayed_work(&psz_cleaner);
+
+	/* Clean up allocations. */
+	kfree(cxt->pstore.buf);
+	cxt->pstore.buf = NULL;
+	cxt->pstore.bufsize = 0;
+	cxt->pstore_zone_info = NULL;
+
+	psz_free_all_zones(cxt);
+
+	/* Clear counters and zone state. */
+	cxt->oops_counter = 0;
+	cxt->panic_counter = 0;
+	atomic_set(&cxt->recovered, 0);
+	atomic_set(&cxt->on_panic, 0);
+
+	mutex_unlock(&cxt->pstore_zone_info_lock);
+}
+EXPORT_SYMBOL_GPL(unregister_pstore_zone);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("WeiXiong Liao <liaoweixiong@allwinnertech.com>");
+MODULE_AUTHOR("Kees Cook <keescook@chromium.org>");
+MODULE_DESCRIPTION("Storage Manager for pstore/blk");
diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c
index 345db56c98fd..755293c8c71a 100644
--- a/fs/qnx6/inode.c
+++ b/fs/qnx6/inode.c
@@ -99,10 +99,9 @@ static int qnx6_readpage(struct file *file, struct page *page)
 	return mpage_readpage(page, qnx6_get_block);
 }
 
-static int qnx6_readpages(struct file *file, struct address_space *mapping,
-		   struct list_head *pages, unsigned nr_pages)
+static void qnx6_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, qnx6_get_block);
+	mpage_readahead(rac, qnx6_get_block);
 }
 
 /*
@@ -499,7 +498,7 @@ static sector_t qnx6_bmap(struct address_space *mapping, sector_t block)
 }
 static const struct address_space_operations qnx6_aops = {
 	.readpage	= qnx6_readpage,
-	.readpages	= qnx6_readpages,
+	.readahead	= qnx6_readahead,
 	.bmap		= qnx6_bmap
 };
 
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c
index 6e826b454082..b6a4f692d345 100644
--- a/fs/quota/dquot.c
+++ b/fs/quota/dquot.c
@@ -497,7 +497,7 @@ int dquot_release(struct dquot *dquot)
 
 	mutex_lock(&dquot->dq_lock);
 	/* Check whether we are not racing with some other dqget() */
-	if (atomic_read(&dquot->dq_count) > 1)
+	if (dquot_is_busy(dquot))
 		goto out_dqlock;
 	if (dqopt->ops[dquot->dq_id.type]->release_dqblk) {
 		ret = dqopt->ops[dquot->dq_id.type]->release_dqblk(dquot);
@@ -595,7 +595,6 @@ int dquot_scan_active(struct super_block *sb,
 		/* Now we have active dquot so we can just increase use count */
 		atomic_inc(&dquot->dq_count);
 		spin_unlock(&dq_list_lock);
-		dqstats_inc(DQST_LOOKUPS);
 		dqput(old_dquot);
 		old_dquot = dquot;
 		/*
@@ -623,7 +622,7 @@ EXPORT_SYMBOL(dquot_scan_active);
 /* Write all dquot structures to quota files */
 int dquot_writeback_dquots(struct super_block *sb, int type)
 {
-	struct list_head *dirty;
+	struct list_head dirty;
 	struct dquot *dquot;
 	struct quota_info *dqopt = sb_dqopt(sb);
 	int cnt;
@@ -637,9 +636,10 @@ int dquot_writeback_dquots(struct super_block *sb, int type)
 		if (!sb_has_quota_active(sb, cnt))
 			continue;
 		spin_lock(&dq_list_lock);
-		dirty = &dqopt->info[cnt].dqi_dirty_list;
-		while (!list_empty(dirty)) {
-			dquot = list_first_entry(dirty, struct dquot,
+		/* Move list away to avoid livelock. */
+		list_replace_init(&dqopt->info[cnt].dqi_dirty_list, &dirty);
+		while (!list_empty(&dirty)) {
+			dquot = list_first_entry(&dirty, struct dquot,
 						 dq_dirty);
 
 			WARN_ON(!test_bit(DQ_ACTIVE_B, &dquot->dq_flags));
@@ -649,7 +649,6 @@ int dquot_writeback_dquots(struct super_block *sb, int type)
 			 * use count */
 			dqgrab(dquot);
 			spin_unlock(&dq_list_lock);
-			dqstats_inc(DQST_LOOKUPS);
 			err = sb->dq_op->write_dquot(dquot);
 			if (err) {
 				/*
@@ -985,6 +984,7 @@ static int add_dquot_ref(struct super_block *sb, int type)
 		 * later.
 		 */
 		old_inode = inode;
+		cond_resched();
 		spin_lock(&sb->s_inode_list_lock);
 	}
 	spin_unlock(&sb->s_inode_list_lock);
@@ -2162,14 +2162,29 @@ int dquot_file_open(struct inode *inode, struct file *file)
 }
 EXPORT_SYMBOL(dquot_file_open);
 
+static void vfs_cleanup_quota_inode(struct super_block *sb, int type)
+{
+	struct quota_info *dqopt = sb_dqopt(sb);
+	struct inode *inode = dqopt->files[type];
+
+	if (!inode)
+		return;
+	if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) {
+		inode_lock(inode);
+		inode->i_flags &= ~S_NOQUOTA;
+		inode_unlock(inode);
+	}
+	dqopt->files[type] = NULL;
+	iput(inode);
+}
+
 /*
  * Turn quota off on a device. type == -1 ==> quotaoff for all types (umount)
  */
 int dquot_disable(struct super_block *sb, int type, unsigned int flags)
 {
-	int cnt, ret = 0;
+	int cnt;
 	struct quota_info *dqopt = sb_dqopt(sb);
-	struct inode *toputinode[MAXQUOTAS];
 
 	/* s_umount should be held in exclusive mode */
 	if (WARN_ON_ONCE(down_read_trylock(&sb->s_umount)))
@@ -2191,7 +2206,6 @@ int dquot_disable(struct super_block *sb, int type, unsigned int flags)
 		return 0;
 
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		toputinode[cnt] = NULL;
 		if (type != -1 && cnt != type)
 			continue;
 		if (!sb_has_quota_loaded(sb, cnt))
@@ -2211,8 +2225,7 @@ int dquot_disable(struct super_block *sb, int type, unsigned int flags)
 				dqopt->flags &=	~dquot_state_flag(
 							DQUOT_SUSPENDED, cnt);
 				spin_unlock(&dq_state_lock);
-				iput(dqopt->files[cnt]);
-				dqopt->files[cnt] = NULL;
+				vfs_cleanup_quota_inode(sb, cnt);
 				continue;
 			}
 			spin_unlock(&dq_state_lock);
@@ -2234,10 +2247,6 @@ int dquot_disable(struct super_block *sb, int type, unsigned int flags)
 		if (dqopt->ops[cnt]->free_file_info)
 			dqopt->ops[cnt]->free_file_info(sb, cnt);
 		put_quota_format(dqopt->info[cnt].dqi_format);
-
-		toputinode[cnt] = dqopt->files[cnt];
-		if (!sb_has_quota_loaded(sb, cnt))
-			dqopt->files[cnt] = NULL;
 		dqopt->info[cnt].dqi_flags = 0;
 		dqopt->info[cnt].dqi_igrace = 0;
 		dqopt->info[cnt].dqi_bgrace = 0;
@@ -2259,32 +2268,22 @@ int dquot_disable(struct super_block *sb, int type, unsigned int flags)
 	 * must also discard the blockdev buffers so that we see the
 	 * changes done by userspace on the next quotaon() */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		/* This can happen when suspending quotas on remount-ro... */
-		if (toputinode[cnt] && !sb_has_quota_loaded(sb, cnt)) {
-			inode_lock(toputinode[cnt]);
-			toputinode[cnt]->i_flags &= ~S_NOQUOTA;
-			truncate_inode_pages(&toputinode[cnt]->i_data, 0);
-			inode_unlock(toputinode[cnt]);
-			mark_inode_dirty_sync(toputinode[cnt]);
+		if (!sb_has_quota_loaded(sb, cnt) && dqopt->files[cnt]) {
+			inode_lock(dqopt->files[cnt]);
+			truncate_inode_pages(&dqopt->files[cnt]->i_data, 0);
+			inode_unlock(dqopt->files[cnt]);
 		}
 	if (sb->s_bdev)
 		invalidate_bdev(sb->s_bdev);
 put_inodes:
+	/* We are done when suspending quotas */
+	if (flags & DQUOT_SUSPENDED)
+		return 0;
+
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		if (toputinode[cnt]) {
-			/* On remount RO, we keep the inode pointer so that we
-			 * can reenable quota on the subsequent remount RW. We
-			 * have to check 'flags' variable and not use sb_has_
-			 * function because another quotaon / quotaoff could
-			 * change global state before we got here. We refuse
-			 * to suspend quotas when there is pending delete on
-			 * the quota file... */
-			if (!(flags & DQUOT_SUSPENDED))
-				iput(toputinode[cnt]);
-			else if (!toputinode[cnt]->i_nlink)
-				ret = -EBUSY;
-		}
-	return ret;
+		if (!sb_has_quota_loaded(sb, cnt))
+			vfs_cleanup_quota_inode(sb, cnt);
+	return 0;
 }
 EXPORT_SYMBOL(dquot_disable);
 
@@ -2299,28 +2298,52 @@ EXPORT_SYMBOL(dquot_quota_off);
  *	Turn quotas on on a device
  */
 
-/*
- * Helper function to turn quotas on when we already have the inode of
- * quota file and no quota information is loaded.
- */
-static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
+static int vfs_setup_quota_inode(struct inode *inode, int type)
+{
+	struct super_block *sb = inode->i_sb;
+	struct quota_info *dqopt = sb_dqopt(sb);
+
+	if (!S_ISREG(inode->i_mode))
+		return -EACCES;
+	if (IS_RDONLY(inode))
+		return -EROFS;
+	if (sb_has_quota_loaded(sb, type))
+		return -EBUSY;
+
+	dqopt->files[type] = igrab(inode);
+	if (!dqopt->files[type])
+		return -EIO;
+	if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) {
+		/* We don't want quota and atime on quota files (deadlocks
+		 * possible) Also nobody should write to the file - we use
+		 * special IO operations which ignore the immutable bit. */
+		inode_lock(inode);
+		inode->i_flags |= S_NOQUOTA;
+		inode_unlock(inode);
+		/*
+		 * When S_NOQUOTA is set, remove dquot references as no more
+		 * references can be added
+		 */
+		__dquot_drop(inode);
+	}
+	return 0;
+}
+
+int dquot_load_quota_sb(struct super_block *sb, int type, int format_id,
 	unsigned int flags)
 {
 	struct quota_format_type *fmt = find_quota_format(format_id);
-	struct super_block *sb = inode->i_sb;
 	struct quota_info *dqopt = sb_dqopt(sb);
 	int error;
 
+	/* Just unsuspend quotas? */
+	BUG_ON(flags & DQUOT_SUSPENDED);
+	/* s_umount should be held in exclusive mode */
+	if (WARN_ON_ONCE(down_read_trylock(&sb->s_umount)))
+		up_read(&sb->s_umount);
+
 	if (!fmt)
 		return -ESRCH;
-	if (!S_ISREG(inode->i_mode)) {
-		error = -EACCES;
-		goto out_fmt;
-	}
-	if (IS_RDONLY(inode)) {
-		error = -EROFS;
-		goto out_fmt;
-	}
 	if (!sb->s_op->quota_write || !sb->s_op->quota_read ||
 	    (type == PRJQUOTA && sb->dq_op->get_projid == NULL)) {
 		error = -EINVAL;
@@ -2352,27 +2375,9 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
 		invalidate_bdev(sb->s_bdev);
 	}
 
-	if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) {
-		/* We don't want quota and atime on quota files (deadlocks
-		 * possible) Also nobody should write to the file - we use
-		 * special IO operations which ignore the immutable bit. */
-		inode_lock(inode);
-		inode->i_flags |= S_NOQUOTA;
-		inode_unlock(inode);
-		/*
-		 * When S_NOQUOTA is set, remove dquot references as no more
-		 * references can be added
-		 */
-		__dquot_drop(inode);
-	}
-
-	error = -EIO;
-	dqopt->files[type] = igrab(inode);
-	if (!dqopt->files[type])
-		goto out_file_flags;
 	error = -EINVAL;
 	if (!fmt->qf_ops->check_quota_file(sb, type))
-		goto out_file_init;
+		goto out_fmt;
 
 	dqopt->ops[type] = fmt->qf_ops;
 	dqopt->info[type].dqi_format = fmt;
@@ -2380,7 +2385,7 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
 	INIT_LIST_HEAD(&dqopt->info[type].dqi_dirty_list);
 	error = dqopt->ops[type]->read_file_info(sb, type);
 	if (error < 0)
-		goto out_file_init;
+		goto out_fmt;
 	if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) {
 		spin_lock(&dq_data_lock);
 		dqopt->info[type].dqi_flags |= DQF_SYS_FILE;
@@ -2395,24 +2400,36 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
 		dquot_disable(sb, type, flags);
 
 	return error;
-out_file_init:
-	dqopt->files[type] = NULL;
-	iput(inode);
-out_file_flags:
-	inode_lock(inode);
-	inode->i_flags &= ~S_NOQUOTA;
-	inode_unlock(inode);
 out_fmt:
 	put_quota_format(fmt);
 
 	return error;
 }
+EXPORT_SYMBOL(dquot_load_quota_sb);
+
+/*
+ * More powerful function for turning on quotas on given quota inode allowing
+ * setting of individual quota flags
+ */
+int dquot_load_quota_inode(struct inode *inode, int type, int format_id,
+	unsigned int flags)
+{
+	int err;
+
+	err = vfs_setup_quota_inode(inode, type);
+	if (err < 0)
+		return err;
+	err = dquot_load_quota_sb(inode->i_sb, type, format_id, flags);
+	if (err < 0)
+		vfs_cleanup_quota_inode(inode->i_sb, type);
+	return err;
+}
+EXPORT_SYMBOL(dquot_load_quota_inode);
 
 /* Reenable quotas on remount RW */
 int dquot_resume(struct super_block *sb, int type)
 {
 	struct quota_info *dqopt = sb_dqopt(sb);
-	struct inode *inode;
 	int ret = 0, cnt;
 	unsigned int flags;
 
@@ -2426,8 +2443,6 @@ int dquot_resume(struct super_block *sb, int type)
 		if (!sb_has_quota_suspended(sb, cnt))
 			continue;
 
-		inode = dqopt->files[cnt];
-		dqopt->files[cnt] = NULL;
 		spin_lock(&dq_state_lock);
 		flags = dqopt->flags & dquot_state_flag(DQUOT_USAGE_ENABLED |
 							DQUOT_LIMITS_ENABLED,
@@ -2436,9 +2451,10 @@ int dquot_resume(struct super_block *sb, int type)
 		spin_unlock(&dq_state_lock);
 
 		flags = dquot_generic_flag(flags, cnt);
-		ret = vfs_load_quota_inode(inode, cnt,
-				dqopt->info[cnt].dqi_fmt_id, flags);
-		iput(inode);
+		ret = dquot_load_quota_sb(sb, cnt, dqopt->info[cnt].dqi_fmt_id,
+					  flags);
+		if (ret < 0)
+			vfs_cleanup_quota_inode(sb, type);
 	}
 
 	return ret;
@@ -2455,7 +2471,7 @@ int dquot_quota_on(struct super_block *sb, int type, int format_id,
 	if (path->dentry->d_sb != sb)
 		error = -EXDEV;
 	else
-		error = vfs_load_quota_inode(d_inode(path->dentry), type,
+		error = dquot_load_quota_inode(d_inode(path->dentry), type,
 					     format_id, DQUOT_USAGE_ENABLED |
 					     DQUOT_LIMITS_ENABLED);
 	return error;
@@ -2463,41 +2479,6 @@ int dquot_quota_on(struct super_block *sb, int type, int format_id,
 EXPORT_SYMBOL(dquot_quota_on);
 
 /*
- * More powerful function for turning on quotas allowing setting
- * of individual quota flags
- */
-int dquot_enable(struct inode *inode, int type, int format_id,
-		 unsigned int flags)
-{
-	struct super_block *sb = inode->i_sb;
-
-	/* Just unsuspend quotas? */
-	BUG_ON(flags & DQUOT_SUSPENDED);
-	/* s_umount should be held in exclusive mode */
-	if (WARN_ON_ONCE(down_read_trylock(&sb->s_umount)))
-		up_read(&sb->s_umount);
-
-	if (!flags)
-		return 0;
-	/* Just updating flags needed? */
-	if (sb_has_quota_loaded(sb, type)) {
-		if (flags & DQUOT_USAGE_ENABLED &&
-		    sb_has_quota_usage_enabled(sb, type))
-			return -EBUSY;
-		if (flags & DQUOT_LIMITS_ENABLED &&
-		    sb_has_quota_limits_enabled(sb, type))
-			return -EBUSY;
-		spin_lock(&dq_state_lock);
-		sb_dqopt(sb)->flags |= dquot_state_flag(flags, type);
-		spin_unlock(&dq_state_lock);
-		return 0;
-	}
-
-	return vfs_load_quota_inode(inode, type, format_id, flags);
-}
-EXPORT_SYMBOL(dquot_enable);
-
-/*
  * This function is used when filesystem needs to initialize quotas
  * during mount time.
  */
@@ -2507,21 +2488,15 @@ int dquot_quota_on_mount(struct super_block *sb, char *qf_name,
 	struct dentry *dentry;
 	int error;
 
-	dentry = lookup_one_len_unlocked(qf_name, sb->s_root, strlen(qf_name));
+	dentry = lookup_positive_unlocked(qf_name, sb->s_root, strlen(qf_name));
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
-	if (d_really_is_negative(dentry)) {
-		error = -ENOENT;
-		goto out;
-	}
-
 	error = security_quota_on(dentry);
 	if (!error)
-		error = vfs_load_quota_inode(d_inode(dentry), type, format_id,
+		error = dquot_load_quota_inode(d_inode(dentry), type, format_id,
 				DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);
 
-out:
 	dput(dentry);
 	return error;
 }
@@ -2543,13 +2518,17 @@ static int dquot_quota_enable(struct super_block *sb, unsigned int flags)
 		if (!(flags & qtype_enforce_flag(type)))
 			continue;
 		/* Can't enforce without accounting */
-		if (!sb_has_quota_usage_enabled(sb, type))
-			return -EINVAL;
-		ret = dquot_enable(dqopt->files[type], type,
-				   dqopt->info[type].dqi_fmt_id,
-				   DQUOT_LIMITS_ENABLED);
-		if (ret < 0)
+		if (!sb_has_quota_usage_enabled(sb, type)) {
+			ret = -EINVAL;
 			goto out_err;
+		}
+		if (sb_has_quota_limits_enabled(sb, type)) {
+			ret = -EBUSY;
+			goto out_err;
+		}
+		spin_lock(&dq_state_lock);
+		dqopt->flags |= dquot_state_flag(DQUOT_LIMITS_ENABLED, type);
+		spin_unlock(&dq_state_lock);
 	}
 	return 0;
 out_err:
@@ -2599,10 +2578,12 @@ static int dquot_quota_disable(struct super_block *sb, unsigned int flags)
 out_err:
 	/* Backout enforcement disabling we already did */
 	for (type--; type >= 0; type--)  {
-		if (flags & qtype_enforce_flag(type))
-			dquot_enable(dqopt->files[type], type,
-				     dqopt->info[type].dqi_fmt_id,
-				     DQUOT_LIMITS_ENABLED);
+		if (flags & qtype_enforce_flag(type)) {
+			spin_lock(&dq_state_lock);
+			dqopt->flags |=
+				dquot_state_flag(DQUOT_LIMITS_ENABLED, type);
+			spin_unlock(&dq_state_lock);
+		}
 	}
 	return ret;
 }
@@ -2800,8 +2781,10 @@ int dquot_get_state(struct super_block *sb, struct qc_state *state)
 			tstate->flags |= QCI_LIMITS_ENFORCED;
 		tstate->spc_timelimit = mi->dqi_bgrace;
 		tstate->ino_timelimit = mi->dqi_igrace;
-		tstate->ino = dqopt->files[type]->i_ino;
-		tstate->blocks = dqopt->files[type]->i_blocks;
+		if (dqopt->files[type]) {
+			tstate->ino = dqopt->files[type]->i_ino;
+			tstate->blocks = dqopt->files[type]->i_blocks;
+		}
 		tstate->nextents = 1;	/* We don't know... */
 		spin_unlock(&dq_data_lock);
 	}
@@ -2860,68 +2843,73 @@ EXPORT_SYMBOL(dquot_quotactl_sysfile_ops);
 static int do_proc_dqstats(struct ctl_table *table, int write,
 		     void __user *buffer, size_t *lenp, loff_t *ppos)
 {
-	unsigned int type = (int *)table->data - dqstats.stat;
+	unsigned int type = (unsigned long *)table->data - dqstats.stat;
+	s64 value = percpu_counter_sum(&dqstats.counter[type]);
+
+	/* Filter negative values for non-monotonic counters */
+	if (value < 0 && (type == DQST_ALLOC_DQUOTS ||
+			  type == DQST_FREE_DQUOTS))
+		value = 0;
 
 	/* Update global table */
-	dqstats.stat[type] =
-			percpu_counter_sum_positive(&dqstats.counter[type]);
-	return proc_dointvec(table, write, buffer, lenp, ppos);
+	dqstats.stat[type] = value;
+	return proc_doulongvec_minmax(table, write, buffer, lenp, ppos);
 }
 
 static struct ctl_table fs_dqstats_table[] = {
 	{
 		.procname	= "lookups",
 		.data		= &dqstats.stat[DQST_LOOKUPS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "drops",
 		.data		= &dqstats.stat[DQST_DROPS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "reads",
 		.data		= &dqstats.stat[DQST_READS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "writes",
 		.data		= &dqstats.stat[DQST_WRITES],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "cache_hits",
 		.data		= &dqstats.stat[DQST_CACHE_HITS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "allocated_dquots",
 		.data		= &dqstats.stat[DQST_ALLOC_DQUOTS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "free_dquots",
 		.data		= &dqstats.stat[DQST_FREE_DQUOTS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "syncs",
 		.data		= &dqstats.stat[DQST_SYNCS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
@@ -2983,11 +2971,7 @@ static int __init dquot_init(void)
 
 	/* Find power-of-two hlist_heads which can fit into allocation */
 	nr_hash = (1UL << order) * PAGE_SIZE / sizeof(struct hlist_head);
-	dq_hash_bits = 0;
-	do {
-		dq_hash_bits++;
-	} while (nr_hash >> dq_hash_bits);
-	dq_hash_bits--;
+	dq_hash_bits = ilog2(nr_hash);
 
 	nr_hash = 1UL << dq_hash_bits;
 	dq_hash_mask = nr_hash - 1;
diff --git a/fs/quota/quota.c b/fs/quota/quota.c
index cb13fb76dbee..5444d3c4d93f 100644
--- a/fs/quota/quota.c
+++ b/fs/quota/quota.c
@@ -60,8 +60,6 @@ static int quota_sync_all(int type)
 {
 	int ret;
 
-	if (type >= MAXQUOTAS)
-		return -EINVAL;
 	ret = security_quotactl(Q_SYNC, type, 0, NULL);
 	if (!ret)
 		iterate_supers(quota_sync_one, &type);
@@ -686,8 +684,6 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id,
 {
 	int ret;
 
-	if (type >= MAXQUOTAS)
-		return -EINVAL;
 	type = array_index_nospec(type, MAXQUOTAS);
 	/*
 	 * Quota not supported on this fs? Check this before s_quota_types
@@ -831,6 +827,9 @@ int kernel_quotactl(unsigned int cmd, const char __user *special,
 	cmds = cmd >> SUBCMDSHIFT;
 	type = cmd & SUBCMDMASK;
 
+	if (type >= MAXQUOTAS)
+		return -EINVAL;
+
 	/*
 	 * As a special case Q_SYNC can be called without a specific device.
 	 * It will iterate all superblocks that have quota enabled and call
diff --git a/fs/quota/quota_v1.c b/fs/quota/quota_v1.c
index c740e5572eb8..cd92e5fa0062 100644
--- a/fs/quota/quota_v1.c
+++ b/fs/quota/quota_v1.c
@@ -217,7 +217,6 @@ static const struct quota_format_ops v1_format_ops = {
 	.check_quota_file	= v1_check_quota_file,
 	.read_file_info		= v1_read_file_info,
 	.write_file_info	= v1_write_file_info,
-	.free_file_info		= NULL,
 	.read_dqblk		= v1_read_dqblk,
 	.commit_dqblk		= v1_commit_dqblk,
 };
diff --git a/fs/quota/quota_v2.c b/fs/quota/quota_v2.c
index 53429c29c784..58fc2a7c7fd1 100644
--- a/fs/quota/quota_v2.c
+++ b/fs/quota/quota_v2.c
@@ -22,8 +22,6 @@ MODULE_AUTHOR("Jan Kara");
 MODULE_DESCRIPTION("Quota format v2 support");
 MODULE_LICENSE("GPL");
 
-#define __QUOTA_V2_PARANOIA
-
 static void v2r0_mem2diskdqb(void *dp, struct dquot *dquot);
 static void v2r0_disk2memdqb(struct dquot *dquot, void *dp);
 static int v2r0_is_id(void *dp, struct dquot *dquot);
diff --git a/fs/quota/quotaio_v1.h b/fs/quota/quotaio_v1.h
index bd11e2c08119..31dca9a89176 100644
--- a/fs/quota/quotaio_v1.h
+++ b/fs/quota/quotaio_v1.h
@@ -25,8 +25,10 @@ struct v1_disk_dqblk {
 	__u32 dqb_ihardlimit;	/* absolute limit on allocated inodes */
 	__u32 dqb_isoftlimit;	/* preferred inode limit */
 	__u32 dqb_curinodes;	/* current # allocated inodes */
-	time_t dqb_btime;	/* time limit for excessive disk use */
-	time_t dqb_itime;	/* time limit for excessive inode use */
+
+	/* below fields differ in length on 32-bit vs 64-bit architectures */
+	unsigned long dqb_btime; /* time limit for excessive disk use */
+	unsigned long dqb_itime; /* time limit for excessive inode use */
 };
 
 #define v1_dqoff(UID)      ((loff_t)((UID) * sizeof (struct v1_disk_dqblk)))
diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c
index d82636e8eb65..ee179a81b3da 100644
--- a/fs/ramfs/inode.c
+++ b/fs/ramfs/inode.c
@@ -181,23 +181,18 @@ enum ramfs_param {
 	Opt_mode,
 };
 
-static const struct fs_parameter_spec ramfs_param_specs[] = {
+const struct fs_parameter_spec ramfs_fs_parameters[] = {
 	fsparam_u32oct("mode",	Opt_mode),
 	{}
 };
 
-const struct fs_parameter_description ramfs_fs_parameters = {
-	.name		= "ramfs",
-	.specs		= ramfs_param_specs,
-};
-
 static int ramfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
 {
 	struct fs_parse_result result;
 	struct ramfs_fs_info *fsi = fc->s_fs_info;
 	int opt;
 
-	opt = fs_parse(fc, &ramfs_fs_parameters, param, &result);
+	opt = fs_parse(fc, ramfs_fs_parameters, param, &result);
 	if (opt < 0) {
 		/*
 		 * We might like to report bad mount options here;
@@ -278,7 +273,7 @@ static void ramfs_kill_sb(struct super_block *sb)
 static struct file_system_type ramfs_fs_type = {
 	.name		= "ramfs",
 	.init_fs_context = ramfs_init_fs_context,
-	.parameters	= &ramfs_fs_parameters,
+	.parameters	= ramfs_fs_parameters,
 	.kill_sb	= ramfs_kill_sb,
 	.fs_flags	= FS_USERNS_MOUNT,
 };
diff --git a/fs/read_write.c b/fs/read_write.c
index 5bbf587f5bc1..bbfa9b12b15e 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -331,7 +331,8 @@ COMPAT_SYSCALL_DEFINE3(lseek, unsigned int, fd, compat_off_t, offset, unsigned i
 }
 #endif
 
-#if !defined(CONFIG_64BIT) || defined(CONFIG_COMPAT)
+#if !defined(CONFIG_64BIT) || defined(CONFIG_COMPAT) || \
+	defined(__ARCH_WANT_SYS_LLSEEK)
 SYSCALL_DEFINE5(llseek, unsigned int, fd, unsigned long, offset_high,
 		unsigned long, offset_low, loff_t __user *, result,
 		unsigned int, whence)
@@ -939,6 +940,34 @@ out:
 	return ret;
 }
 
+ssize_t vfs_iocb_iter_read(struct file *file, struct kiocb *iocb,
+			   struct iov_iter *iter)
+{
+	size_t tot_len;
+	ssize_t ret = 0;
+
+	if (!file->f_op->read_iter)
+		return -EINVAL;
+	if (!(file->f_mode & FMODE_READ))
+		return -EBADF;
+	if (!(file->f_mode & FMODE_CAN_READ))
+		return -EINVAL;
+
+	tot_len = iov_iter_count(iter);
+	if (!tot_len)
+		goto out;
+	ret = rw_verify_area(READ, file, &iocb->ki_pos, tot_len);
+	if (ret < 0)
+		return ret;
+
+	ret = call_read_iter(file, iocb, iter);
+out:
+	if (ret >= 0)
+		fsnotify_access(file);
+	return ret;
+}
+EXPORT_SYMBOL(vfs_iocb_iter_read);
+
 ssize_t vfs_iter_read(struct file *file, struct iov_iter *iter, loff_t *ppos,
 		rwf_t flags)
 {
@@ -975,6 +1004,34 @@ static ssize_t do_iter_write(struct file *file, struct iov_iter *iter,
 	return ret;
 }
 
+ssize_t vfs_iocb_iter_write(struct file *file, struct kiocb *iocb,
+			    struct iov_iter *iter)
+{
+	size_t tot_len;
+	ssize_t ret = 0;
+
+	if (!file->f_op->write_iter)
+		return -EINVAL;
+	if (!(file->f_mode & FMODE_WRITE))
+		return -EBADF;
+	if (!(file->f_mode & FMODE_CAN_WRITE))
+		return -EINVAL;
+
+	tot_len = iov_iter_count(iter);
+	if (!tot_len)
+		return 0;
+	ret = rw_verify_area(WRITE, file, &iocb->ki_pos, tot_len);
+	if (ret < 0)
+		return ret;
+
+	ret = call_write_iter(file, iocb, iter);
+	if (ret > 0)
+		fsnotify_modify(file);
+
+	return ret;
+}
+EXPORT_SYMBOL(vfs_iocb_iter_write);
+
 ssize_t vfs_iter_write(struct file *file, struct iov_iter *iter, loff_t *ppos,
 		rwf_t flags)
 {
@@ -1777,10 +1834,9 @@ static int remap_verify_area(struct file *file, loff_t pos, loff_t len,
  * else.  Assume that the offsets have already been checked for block
  * alignment.
  *
- * For deduplication we always scale down to the previous block because we
- * can't meaningfully compare post-EOF contents.
- *
- * For clone we only link a partial EOF block above the destination file's EOF.
+ * For clone we only link a partial EOF block above or at the destination file's
+ * EOF.  For deduplication we accept a partial EOF block only if it ends at the
+ * destination file's EOF (can not link it into the middle of a file).
  *
  * Shorten the request if possible.
  */
@@ -1796,8 +1852,7 @@ static int generic_remap_check_len(struct inode *inode_in,
 	if ((*len & blkmask) == 0)
 		return 0;
 
-	if ((remap_flags & REMAP_FILE_DEDUP) ||
-	    pos_out + *len < i_size_read(inode_out))
+	if (pos_out + *len < i_size_read(inode_out))
 		new_len &= ~blkmask;
 
 	if (new_len == *len)
diff --git a/fs/readdir.c b/fs/readdir.c
index d26d5ea4de7b..a49f07c11cfb 100644
--- a/fs/readdir.c
+++ b/fs/readdir.c
@@ -102,10 +102,14 @@ EXPORT_SYMBOL(iterate_dir);
  * filename length, and the above "soft error" worry means
  * that it's probably better left alone until we have that
  * issue clarified.
+ *
+ * Note the PATH_MAX check - it's arbitrary but the real
+ * kernel limit on a possible path component, not NAME_MAX,
+ * which is the technical standard limit.
  */
 static int verify_dirent_name(const char *name, int len)
 {
-	if (!len)
+	if (len <= 0 || len >= PATH_MAX)
 		return -EIO;
 	if (memchr(name, '/', len))
 		return -EIO;
@@ -153,17 +157,18 @@ static int fillonedir(struct dir_context *ctx, const char *name, int namlen,
 	}
 	buf->result++;
 	dirent = buf->dirent;
-	if (!access_ok(dirent,
+	if (!user_write_access_begin(dirent,
 			(unsigned long)(dirent->d_name + namlen + 1) -
 				(unsigned long)dirent))
 		goto efault;
-	if (	__put_user(d_ino, &dirent->d_ino) ||
-		__put_user(offset, &dirent->d_offset) ||
-		__put_user(namlen, &dirent->d_namlen) ||
-		__copy_to_user(dirent->d_name, name, namlen) ||
-		__put_user(0, dirent->d_name + namlen))
-		goto efault;
+	unsafe_put_user(d_ino, &dirent->d_ino, efault_end);
+	unsafe_put_user(offset, &dirent->d_offset, efault_end);
+	unsafe_put_user(namlen, &dirent->d_namlen, efault_end);
+	unsafe_copy_dirent_name(dirent->d_name, name, namlen, efault_end);
+	user_write_access_end();
 	return 0;
+efault_end:
+	user_write_access_end();
 efault:
 	buf->result = -EFAULT;
 	return -EFAULT;
@@ -206,7 +211,7 @@ struct linux_dirent {
 struct getdents_callback {
 	struct dir_context ctx;
 	struct linux_dirent __user * current_dir;
-	struct linux_dirent __user * previous;
+	int prev_reclen;
 	int count;
 	int error;
 };
@@ -214,12 +219,13 @@ struct getdents_callback {
 static int filldir(struct dir_context *ctx, const char *name, int namlen,
 		   loff_t offset, u64 ino, unsigned int d_type)
 {
-	struct linux_dirent __user * dirent;
+	struct linux_dirent __user *dirent, *prev;
 	struct getdents_callback *buf =
 		container_of(ctx, struct getdents_callback, ctx);
 	unsigned long d_ino;
 	int reclen = ALIGN(offsetof(struct linux_dirent, d_name) + namlen + 2,
 		sizeof(long));
+	int prev_reclen;
 
 	buf->error = verify_dirent_name(name, namlen);
 	if (unlikely(buf->error))
@@ -232,32 +238,28 @@ static int filldir(struct dir_context *ctx, const char *name, int namlen,
 		buf->error = -EOVERFLOW;
 		return -EOVERFLOW;
 	}
-	dirent = buf->previous;
-	if (dirent && signal_pending(current))
+	prev_reclen = buf->prev_reclen;
+	if (prev_reclen && signal_pending(current))
 		return -EINTR;
-
-	/*
-	 * Note! This range-checks 'previous' (which may be NULL).
-	 * The real range was checked in getdents
-	 */
-	if (!user_access_begin(dirent, sizeof(*dirent)))
-		goto efault;
-	if (dirent)
-		unsafe_put_user(offset, &dirent->d_off, efault_end);
 	dirent = buf->current_dir;
+	prev = (void __user *) dirent - prev_reclen;
+	if (!user_write_access_begin(prev, reclen + prev_reclen))
+		goto efault;
+
+	/* This might be 'dirent->d_off', but if so it will get overwritten */
+	unsafe_put_user(offset, &prev->d_off, efault_end);
 	unsafe_put_user(d_ino, &dirent->d_ino, efault_end);
 	unsafe_put_user(reclen, &dirent->d_reclen, efault_end);
 	unsafe_put_user(d_type, (char __user *) dirent + reclen - 1, efault_end);
 	unsafe_copy_dirent_name(dirent->d_name, name, namlen, efault_end);
-	user_access_end();
+	user_write_access_end();
 
-	buf->previous = dirent;
-	dirent = (void __user *)dirent + reclen;
-	buf->current_dir = dirent;
+	buf->current_dir = (void __user *)dirent + reclen;
+	buf->prev_reclen = reclen;
 	buf->count -= reclen;
 	return 0;
 efault_end:
-	user_access_end();
+	user_write_access_end();
 efault:
 	buf->error = -EFAULT;
 	return -EFAULT;
@@ -267,7 +269,6 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd,
 		struct linux_dirent __user *, dirent, unsigned int, count)
 {
 	struct fd f;
-	struct linux_dirent __user * lastdirent;
 	struct getdents_callback buf = {
 		.ctx.actor = filldir,
 		.count = count,
@@ -275,9 +276,6 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd,
 	};
 	int error;
 
-	if (!access_ok(dirent, count))
-		return -EFAULT;
-
 	f = fdget_pos(fd);
 	if (!f.file)
 		return -EBADF;
@@ -285,8 +283,10 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd,
 	error = iterate_dir(f.file, &buf.ctx);
 	if (error >= 0)
 		error = buf.error;
-	lastdirent = buf.previous;
-	if (lastdirent) {
+	if (buf.prev_reclen) {
+		struct linux_dirent __user * lastdirent;
+		lastdirent = (void __user *)buf.current_dir - buf.prev_reclen;
+
 		if (put_user(buf.ctx.pos, &lastdirent->d_off))
 			error = -EFAULT;
 		else
@@ -299,7 +299,7 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd,
 struct getdents_callback64 {
 	struct dir_context ctx;
 	struct linux_dirent64 __user * current_dir;
-	struct linux_dirent64 __user * previous;
+	int prev_reclen;
 	int count;
 	int error;
 };
@@ -307,11 +307,12 @@ struct getdents_callback64 {
 static int filldir64(struct dir_context *ctx, const char *name, int namlen,
 		     loff_t offset, u64 ino, unsigned int d_type)
 {
-	struct linux_dirent64 __user *dirent;
+	struct linux_dirent64 __user *dirent, *prev;
 	struct getdents_callback64 *buf =
 		container_of(ctx, struct getdents_callback64, ctx);
 	int reclen = ALIGN(offsetof(struct linux_dirent64, d_name) + namlen + 1,
 		sizeof(u64));
+	int prev_reclen;
 
 	buf->error = verify_dirent_name(name, namlen);
 	if (unlikely(buf->error))
@@ -319,32 +320,29 @@ static int filldir64(struct dir_context *ctx, const char *name, int namlen,
 	buf->error = -EINVAL;	/* only used if we fail.. */
 	if (reclen > buf->count)
 		return -EINVAL;
-	dirent = buf->previous;
-	if (dirent && signal_pending(current))
+	prev_reclen = buf->prev_reclen;
+	if (prev_reclen && signal_pending(current))
 		return -EINTR;
-
-	/*
-	 * Note! This range-checks 'previous' (which may be NULL).
-	 * The real range was checked in getdents
-	 */
-	if (!user_access_begin(dirent, sizeof(*dirent)))
-		goto efault;
-	if (dirent)
-		unsafe_put_user(offset, &dirent->d_off, efault_end);
 	dirent = buf->current_dir;
+	prev = (void __user *)dirent - prev_reclen;
+	if (!user_write_access_begin(prev, reclen + prev_reclen))
+		goto efault;
+
+	/* This might be 'dirent->d_off', but if so it will get overwritten */
+	unsafe_put_user(offset, &prev->d_off, efault_end);
 	unsafe_put_user(ino, &dirent->d_ino, efault_end);
 	unsafe_put_user(reclen, &dirent->d_reclen, efault_end);
 	unsafe_put_user(d_type, &dirent->d_type, efault_end);
 	unsafe_copy_dirent_name(dirent->d_name, name, namlen, efault_end);
-	user_access_end();
+	user_write_access_end();
 
-	buf->previous = dirent;
-	dirent = (void __user *)dirent + reclen;
-	buf->current_dir = dirent;
+	buf->prev_reclen = reclen;
+	buf->current_dir = (void __user *)dirent + reclen;
 	buf->count -= reclen;
 	return 0;
+
 efault_end:
-	user_access_end();
+	user_write_access_end();
 efault:
 	buf->error = -EFAULT;
 	return -EFAULT;
@@ -354,7 +352,6 @@ int ksys_getdents64(unsigned int fd, struct linux_dirent64 __user *dirent,
 		    unsigned int count)
 {
 	struct fd f;
-	struct linux_dirent64 __user * lastdirent;
 	struct getdents_callback64 buf = {
 		.ctx.actor = filldir64,
 		.count = count,
@@ -362,9 +359,6 @@ int ksys_getdents64(unsigned int fd, struct linux_dirent64 __user *dirent,
 	};
 	int error;
 
-	if (!access_ok(dirent, count))
-		return -EFAULT;
-
 	f = fdget_pos(fd);
 	if (!f.file)
 		return -EBADF;
@@ -372,10 +366,12 @@ int ksys_getdents64(unsigned int fd, struct linux_dirent64 __user *dirent,
 	error = iterate_dir(f.file, &buf.ctx);
 	if (error >= 0)
 		error = buf.error;
-	lastdirent = buf.previous;
-	if (lastdirent) {
+	if (buf.prev_reclen) {
+		struct linux_dirent64 __user * lastdirent;
 		typeof(lastdirent->d_off) d_off = buf.ctx.pos;
-		if (__put_user(d_off, &lastdirent->d_off))
+
+		lastdirent = (void __user *) buf.current_dir - buf.prev_reclen;
+		if (put_user(d_off, &lastdirent->d_off))
 			error = -EFAULT;
 		else
 			error = count - buf.count;
@@ -423,17 +419,18 @@ static int compat_fillonedir(struct dir_context *ctx, const char *name,
 	}
 	buf->result++;
 	dirent = buf->dirent;
-	if (!access_ok(dirent,
+	if (!user_write_access_begin(dirent,
 			(unsigned long)(dirent->d_name + namlen + 1) -
 				(unsigned long)dirent))
 		goto efault;
-	if (	__put_user(d_ino, &dirent->d_ino) ||
-		__put_user(offset, &dirent->d_offset) ||
-		__put_user(namlen, &dirent->d_namlen) ||
-		__copy_to_user(dirent->d_name, name, namlen) ||
-		__put_user(0, dirent->d_name + namlen))
-		goto efault;
+	unsafe_put_user(d_ino, &dirent->d_ino, efault_end);
+	unsafe_put_user(offset, &dirent->d_offset, efault_end);
+	unsafe_put_user(namlen, &dirent->d_namlen, efault_end);
+	unsafe_copy_dirent_name(dirent->d_name, name, namlen, efault_end);
+	user_write_access_end();
 	return 0;
+efault_end:
+	user_write_access_end();
 efault:
 	buf->result = -EFAULT;
 	return -EFAULT;
@@ -470,7 +467,7 @@ struct compat_linux_dirent {
 struct compat_getdents_callback {
 	struct dir_context ctx;
 	struct compat_linux_dirent __user *current_dir;
-	struct compat_linux_dirent __user *previous;
+	int prev_reclen;
 	int count;
 	int error;
 };
@@ -478,13 +475,17 @@ struct compat_getdents_callback {
 static int compat_filldir(struct dir_context *ctx, const char *name, int namlen,
 		loff_t offset, u64 ino, unsigned int d_type)
 {
-	struct compat_linux_dirent __user * dirent;
+	struct compat_linux_dirent __user *dirent, *prev;
 	struct compat_getdents_callback *buf =
 		container_of(ctx, struct compat_getdents_callback, ctx);
 	compat_ulong_t d_ino;
 	int reclen = ALIGN(offsetof(struct compat_linux_dirent, d_name) +
 		namlen + 2, sizeof(compat_long_t));
+	int prev_reclen;
 
+	buf->error = verify_dirent_name(name, namlen);
+	if (unlikely(buf->error))
+		return buf->error;
 	buf->error = -EINVAL;	/* only used if we fail.. */
 	if (reclen > buf->count)
 		return -EINVAL;
@@ -493,29 +494,27 @@ static int compat_filldir(struct dir_context *ctx, const char *name, int namlen,
 		buf->error = -EOVERFLOW;
 		return -EOVERFLOW;
 	}
-	dirent = buf->previous;
-	if (dirent) {
-		if (signal_pending(current))
-			return -EINTR;
-		if (__put_user(offset, &dirent->d_off))
-			goto efault;
-	}
+	prev_reclen = buf->prev_reclen;
+	if (prev_reclen && signal_pending(current))
+		return -EINTR;
 	dirent = buf->current_dir;
-	if (__put_user(d_ino, &dirent->d_ino))
-		goto efault;
-	if (__put_user(reclen, &dirent->d_reclen))
-		goto efault;
-	if (copy_to_user(dirent->d_name, name, namlen))
-		goto efault;
-	if (__put_user(0, dirent->d_name + namlen))
+	prev = (void __user *) dirent - prev_reclen;
+	if (!user_write_access_begin(prev, reclen + prev_reclen))
 		goto efault;
-	if (__put_user(d_type, (char  __user *) dirent + reclen - 1))
-		goto efault;
-	buf->previous = dirent;
-	dirent = (void __user *)dirent + reclen;
-	buf->current_dir = dirent;
+
+	unsafe_put_user(offset, &prev->d_off, efault_end);
+	unsafe_put_user(d_ino, &dirent->d_ino, efault_end);
+	unsafe_put_user(reclen, &dirent->d_reclen, efault_end);
+	unsafe_put_user(d_type, (char __user *) dirent + reclen - 1, efault_end);
+	unsafe_copy_dirent_name(dirent->d_name, name, namlen, efault_end);
+	user_write_access_end();
+
+	buf->prev_reclen = reclen;
+	buf->current_dir = (void __user *)dirent + reclen;
 	buf->count -= reclen;
 	return 0;
+efault_end:
+	user_write_access_end();
 efault:
 	buf->error = -EFAULT;
 	return -EFAULT;
@@ -525,7 +524,6 @@ COMPAT_SYSCALL_DEFINE3(getdents, unsigned int, fd,
 		struct compat_linux_dirent __user *, dirent, unsigned int, count)
 {
 	struct fd f;
-	struct compat_linux_dirent __user * lastdirent;
 	struct compat_getdents_callback buf = {
 		.ctx.actor = compat_filldir,
 		.current_dir = dirent,
@@ -533,9 +531,6 @@ COMPAT_SYSCALL_DEFINE3(getdents, unsigned int, fd,
 	};
 	int error;
 
-	if (!access_ok(dirent, count))
-		return -EFAULT;
-
 	f = fdget_pos(fd);
 	if (!f.file)
 		return -EBADF;
@@ -543,8 +538,10 @@ COMPAT_SYSCALL_DEFINE3(getdents, unsigned int, fd,
 	error = iterate_dir(f.file, &buf.ctx);
 	if (error >= 0)
 		error = buf.error;
-	lastdirent = buf.previous;
-	if (lastdirent) {
+	if (buf.prev_reclen) {
+		struct compat_linux_dirent __user * lastdirent;
+		lastdirent = (void __user *)buf.current_dir - buf.prev_reclen;
+
 		if (put_user(buf.ctx.pos, &lastdirent->d_off))
 			error = -EFAULT;
 		else
diff --git a/fs/reiserfs/do_balan.c b/fs/reiserfs/do_balan.c
index 4075e41408b4..5129efc6f2e6 100644
--- a/fs/reiserfs/do_balan.c
+++ b/fs/reiserfs/do_balan.c
@@ -842,7 +842,7 @@ static void balance_leaf_paste_right_whole(struct tree_balance *tb,
 	struct item_head *pasted;
 	struct buffer_info bi;
 
-							buffer_info_init_right(tb, &bi);
+	buffer_info_init_right(tb, &bi);
 	leaf_shift_right(tb, tb->rnum[0], tb->rbytes);
 
 	/* append item in R[0] */
diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index 843aadcc123c..0b641ae694f1 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -38,16 +38,10 @@ static int reiserfs_file_release(struct inode *inode, struct file *filp)
 
 	BUG_ON(!S_ISREG(inode->i_mode));
 
-        if (atomic_add_unless(&REISERFS_I(inode)->openers, -1, 1))
+	if (!atomic_dec_and_mutex_lock(&REISERFS_I(inode)->openers,
+				       &REISERFS_I(inode)->tailpack))
 		return 0;
 
-	mutex_lock(&REISERFS_I(inode)->tailpack);
-
-        if (!atomic_dec_and_test(&REISERFS_I(inode)->openers)) {
-		mutex_unlock(&REISERFS_I(inode)->tailpack);
-		return 0;
-	}
-
 	/* fast out for when nothing needs to be done */
 	if ((!(REISERFS_I(inode)->i_flags & i_pack_on_close_mask) ||
 	     !tail_has_to_be_packed(inode)) &&
@@ -165,7 +159,7 @@ static int reiserfs_sync_file(struct file *filp, loff_t start, loff_t end,
 	barrier_done = reiserfs_commit_for_inode(inode);
 	reiserfs_write_unlock(inode->i_sb);
 	if (barrier_done != 1 && reiserfs_barrier_flush(inode->i_sb))
-		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
 	inode_unlock(inode);
 	if (barrier_done < 0)
 		return barrier_done;
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 132ec4406ed0..0031070b3692 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -1160,11 +1160,9 @@ failure:
 	return retval;
 }
 
-static int
-reiserfs_readpages(struct file *file, struct address_space *mapping,
-		   struct list_head *pages, unsigned nr_pages)
+static void reiserfs_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, reiserfs_get_block);
+	mpage_readahead(rac, reiserfs_get_block);
 }
 
 /*
@@ -2097,6 +2095,15 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 		goto out_inserted_sd;
 	}
 
+	/*
+	 * Mark it private if we're creating the privroot
+	 * or something under it.
+	 */
+	if (IS_PRIVATE(dir) || dentry == REISERFS_SB(sb)->priv_root) {
+		inode->i_flags |= S_PRIVATE;
+		inode->i_opflags &= ~IOP_XATTR;
+	}
+
 	if (reiserfs_posixacl(inode->i_sb)) {
 		reiserfs_write_unlock(inode->i_sb);
 		retval = reiserfs_inherit_default_acl(th, dir, dentry, inode);
@@ -2111,8 +2118,7 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 		reiserfs_warning(inode->i_sb, "jdm-13090",
 				 "ACLs aren't enabled in the fs, "
 				 "but vfs thinks they are!");
-	} else if (IS_PRIVATE(dir))
-		inode->i_flags |= S_PRIVATE;
+	}
 
 	if (security->name) {
 		reiserfs_write_unlock(inode->i_sb);
@@ -3426,7 +3432,7 @@ out:
 const struct address_space_operations reiserfs_address_space_operations = {
 	.writepage = reiserfs_writepage,
 	.readpage = reiserfs_readpage,
-	.readpages = reiserfs_readpages,
+	.readahead = reiserfs_readahead,
 	.releasepage = reiserfs_releasepage,
 	.invalidatepage = reiserfs_invalidatepage,
 	.write_begin = reiserfs_write_begin,
diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c
index 45e1a5d11af3..adb21bea3d60 100644
--- a/fs/reiserfs/ioctl.c
+++ b/fs/reiserfs/ioctl.c
@@ -184,11 +184,12 @@ int reiserfs_unpack(struct inode *inode, struct file *filp)
 	}
 
 	/* we need to make sure nobody is changing the file size beneath us */
-{
-	int depth = reiserfs_write_unlock_nested(inode->i_sb);
-	inode_lock(inode);
-	reiserfs_write_lock_nested(inode->i_sb, depth);
-}
+	{
+		int depth = reiserfs_write_unlock_nested(inode->i_sb);
+
+		inode_lock(inode);
+		reiserfs_write_lock_nested(inode->i_sb, depth);
+	}
 
 	reiserfs_write_lock(inode->i_sb);
 
diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index 4b3e3e73b512..5c766330e493 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -56,8 +56,6 @@
 /* gets a struct reiserfs_journal_list * from a list head */
 #define JOURNAL_LIST_ENTRY(h) (list_entry((h), struct reiserfs_journal_list, \
                                j_list))
-#define JOURNAL_WORK_ENTRY(h) (list_entry((h), struct reiserfs_journal_list, \
-                               j_working_list))
 
 /* must be correct to keep the desc and commit structs at 4k */
 #define JOURNAL_TRANS_HALF 1018
@@ -2601,7 +2599,6 @@ static int journal_init_dev(struct super_block *super,
 	int result;
 	dev_t jdev;
 	fmode_t blkdev_mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
-	char b[BDEVNAME_SIZE];
 
 	result = 0;
 
@@ -2623,8 +2620,8 @@ static int journal_init_dev(struct super_block *super,
 			result = PTR_ERR(journal->j_dev_bd);
 			journal->j_dev_bd = NULL;
 			reiserfs_warning(super, "sh-458",
-					 "cannot init journal device '%s': %i",
-					 __bdevname(jdev, b), result);
+					 "cannot init journal device unknown-block(%u,%u): %i",
+					 MAJOR(jdev), MINOR(jdev), result);
 			return result;
 		} else if (jdev != super->s_dev)
 			set_blocksize(journal->j_dev_bd, super->s_blocksize);
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index 97f3fc4fdd79..1594687582f0 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -377,10 +377,13 @@ static struct dentry *reiserfs_lookup(struct inode *dir, struct dentry *dentry,
 
 		/*
 		 * Propagate the private flag so we know we're
-		 * in the priv tree
+		 * in the priv tree.  Also clear IOP_XATTR
+		 * since we don't have xattrs on xattr files.
 		 */
-		if (IS_PRIVATE(dir))
+		if (IS_PRIVATE(dir)) {
 			inode->i_flags |= S_PRIVATE;
+			inode->i_opflags &= ~IOP_XATTR;
+		}
 	}
 	reiserfs_write_unlock(dir->i_sb);
 	if (retval == IO_ERROR) {
@@ -835,10 +838,10 @@ static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode
 	 */
 	INC_DIR_INODE_NLINK(dir)
 
-	    retval = reiserfs_new_inode(&th, dir, mode, NULL /*symlink */ ,
-					old_format_only(dir->i_sb) ?
-					EMPTY_DIR_SIZE_V1 : EMPTY_DIR_SIZE,
-					dentry, inode, &security);
+	retval = reiserfs_new_inode(&th, dir, mode, NULL /*symlink */,
+				    old_format_only(dir->i_sb) ?
+				    EMPTY_DIR_SIZE_V1 : EMPTY_DIR_SIZE,
+				    dentry, inode, &security);
 	if (retval) {
 		DEC_DIR_INODE_NLINK(dir)
 		goto out_failed;
@@ -964,7 +967,7 @@ static int reiserfs_rmdir(struct inode *dir, struct dentry *dentry)
 	reiserfs_update_sd(&th, inode);
 
 	DEC_DIR_INODE_NLINK(dir)
-	    dir->i_size -= (DEH_SIZE + de.de_entrylen);
+	dir->i_size -= (DEH_SIZE + de.de_entrylen);
 	reiserfs_update_sd(&th, dir);
 
 	/* prevent empty directory from getting lost */
diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c
index f2cf3441fdfc..ff336513c254 100644
--- a/fs/reiserfs/procfs.c
+++ b/fs/reiserfs/procfs.c
@@ -63,7 +63,6 @@ static int show_version(struct seq_file *m, void *unused)
 #define MAP( i ) D4C( objectid_map( sb, rs )[ i ] )
 
 #define DJF( x ) le32_to_cpu( rs -> x )
-#define DJV( x ) le32_to_cpu( s_v1 -> x )
 #define DJP( x ) le32_to_cpu( jp -> x )
 #define JF( x ) ( r -> s_journal -> x )
 
diff --git a/fs/reiserfs/reiserfs.h b/fs/reiserfs/reiserfs.h
index e5ca9ed79e54..726580114d55 100644
--- a/fs/reiserfs/reiserfs.h
+++ b/fs/reiserfs/reiserfs.h
@@ -1168,6 +1168,8 @@ static inline int bmap_would_wrap(unsigned bmap_nr)
 	return bmap_nr > ((1LL << 16) - 1);
 }
 
+extern const struct xattr_handler *reiserfs_xattr_handlers[];
+
 /*
  * this says about version of key of all items (but stat data) the
  * object consists of
diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c
index da9ebe33882b..8bf88d690729 100644
--- a/fs/reiserfs/stree.c
+++ b/fs/reiserfs/stree.c
@@ -918,12 +918,6 @@ int comp_items(const struct item_head *stored_ih, const struct treepath *path)
 	return memcmp(stored_ih, ih, IH_SIZE);
 }
 
-/* unformatted nodes are not logged anymore, ever.  This is safe now */
-#define held_by_others(bh) (atomic_read(&(bh)->b_count) > 1)
-
-/* block can not be forgotten as it is in I/O or held by someone */
-#define block_in_use(bh) (buffer_locked(bh) || (held_by_others(bh)))
-
 /* prepare for delete or cut of direct item */
 static inline int prepare_for_direct_item(struct treepath *path,
 					  struct item_head *le_ih,
@@ -2246,7 +2240,8 @@ error_out:
 	/* also releases the path */
 	unfix_nodes(&s_ins_balance);
 #ifdef REISERQUOTA_DEBUG
-	reiserfs_debug(th->t_super, REISERFS_DEBUG_CODE,
+	if (inode)
+		reiserfs_debug(th->t_super, REISERFS_DEBUG_CODE,
 		       "reiserquota insert_item(): freeing %u id=%u type=%c",
 		       quota_bytes, inode->i_uid, head2type(ih));
 #endif
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index d69b4ac0ae2f..a6bce5b1fb1d 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -629,6 +629,7 @@ static void reiserfs_put_super(struct super_block *s)
 	reiserfs_write_unlock(s);
 	mutex_destroy(&REISERFS_SB(s)->lock);
 	destroy_workqueue(REISERFS_SB(s)->commit_wq);
+	kfree(REISERFS_SB(s)->s_jdev);
 	kfree(s->s_fs_info);
 	s->s_fs_info = NULL;
 }
@@ -1947,7 +1948,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 		if (!sbi->s_jdev) {
 			SWARN(silent, s, "", "Cannot allocate memory for "
 				"journal device name");
-			goto error;
+			goto error_unlocked;
 		}
 	}
 #ifdef CONFIG_QUOTA
@@ -2049,6 +2050,8 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	if (replay_only(s))
 		goto error_unlocked;
 
+	s->s_xattr = reiserfs_xattr_handlers;
+
 	if (bdev_read_only(s->s_bdev) && !sb_rdonly(s)) {
 		SWARN(silent, s, "clm-7000",
 		      "Detected readonly device, marking FS readonly");
@@ -2238,6 +2241,7 @@ error_unlocked:
 			kfree(qf_names[j]);
 	}
 #endif
+	kfree(sbi->s_jdev);
 	kfree(sbi);
 
 	s->s_fs_info = NULL;
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index b5b26d8a192c..28b241cd6987 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -122,13 +122,13 @@ static struct dentry *open_xa_root(struct super_block *sb, int flags)
 	struct dentry *xaroot;
 
 	if (d_really_is_negative(privroot))
-		return ERR_PTR(-ENODATA);
+		return ERR_PTR(-EOPNOTSUPP);
 
 	inode_lock_nested(d_inode(privroot), I_MUTEX_XATTR);
 
 	xaroot = dget(REISERFS_SB(sb)->xattr_root);
 	if (!xaroot)
-		xaroot = ERR_PTR(-ENODATA);
+		xaroot = ERR_PTR(-EOPNOTSUPP);
 	else if (d_really_is_negative(xaroot)) {
 		int err = -ENODATA;
 
@@ -319,8 +319,12 @@ static int reiserfs_for_each_xattr(struct inode *inode,
 out_dir:
 	dput(dir);
 out:
-	/* -ENODATA isn't an error */
-	if (err == -ENODATA)
+	/*
+	 * -ENODATA: this object doesn't have any xattrs
+	 * -EOPNOTSUPP: this file system doesn't have xattrs enabled on disk.
+	 * Neither are errors
+	 */
+	if (err == -ENODATA || err == -EOPNOTSUPP)
 		err = 0;
 	return err;
 }
@@ -619,6 +623,10 @@ int reiserfs_xattr_set(struct inode *inode, const char *name,
 	int error, error2;
 	size_t jbegin_count = reiserfs_xattr_nblocks(inode, buffer_size);
 
+	/* Check before we start a transaction and then do nothing. */
+	if (!d_really_is_positive(REISERFS_SB(inode->i_sb)->priv_root))
+		return -EOPNOTSUPP;
+
 	if (!(flags & XATTR_REPLACE))
 		jbegin_count += reiserfs_xattr_jcreate_nblocks(inode);
 
@@ -841,8 +849,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size)
 	if (d_really_is_negative(dentry))
 		return -EINVAL;
 
-	if (!dentry->d_sb->s_xattr ||
-	    get_inode_sd_version(d_inode(dentry)) == STAT_DATA_V1)
+	if (get_inode_sd_version(d_inode(dentry)) == STAT_DATA_V1)
 		return -EOPNOTSUPP;
 
 	dir = open_xa_dir(d_inode(dentry), XATTR_REPLACE);
@@ -882,6 +889,7 @@ static int create_privroot(struct dentry *dentry)
 	}
 
 	d_inode(dentry)->i_flags |= S_PRIVATE;
+	d_inode(dentry)->i_opflags &= ~IOP_XATTR;
 	reiserfs_info(dentry->d_sb, "Created %s - reserved for xattr "
 		      "storage.\n", PRIVROOT_NAME);
 
@@ -895,7 +903,7 @@ static int create_privroot(struct dentry *dentry) { return 0; }
 #endif
 
 /* Actual operations that are exported to VFS-land */
-static const struct xattr_handler *reiserfs_xattr_handlers[] = {
+const struct xattr_handler *reiserfs_xattr_handlers[] = {
 #ifdef CONFIG_REISERFS_FS_XATTR
 	&reiserfs_xattr_user_handler,
 	&reiserfs_xattr_trusted_handler,
@@ -966,8 +974,10 @@ int reiserfs_lookup_privroot(struct super_block *s)
 	if (!IS_ERR(dentry)) {
 		REISERFS_SB(s)->priv_root = dentry;
 		d_set_d_op(dentry, &xattr_lookup_poison_ops);
-		if (d_really_is_positive(dentry))
+		if (d_really_is_positive(dentry)) {
 			d_inode(dentry)->i_flags |= S_PRIVATE;
+			d_inode(dentry)->i_opflags &= ~IOP_XATTR;
+		}
 	} else
 		err = PTR_ERR(dentry);
 	inode_unlock(d_inode(s->s_root));
@@ -996,7 +1006,6 @@ int reiserfs_xattr_init(struct super_block *s, int mount_flags)
 	}
 
 	if (d_really_is_positive(privroot)) {
-		s->s_xattr = reiserfs_xattr_handlers;
 		inode_lock(d_inode(privroot));
 		if (!REISERFS_SB(s)->xattr_root) {
 			struct dentry *dentry;
diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index aa9380bac196..05f666794561 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -320,10 +320,8 @@ reiserfs_inherit_default_acl(struct reiserfs_transaction_handle *th,
 	 * would be useless since permissions are ignored, and a pain because
 	 * it introduces locking cycles
 	 */
-	if (IS_PRIVATE(dir)) {
-		inode->i_flags |= S_PRIVATE;
+	if (IS_PRIVATE(inode))
 		goto apply_umask;
-	}
 
 	err = posix_acl_create(dir, &inode->i_mode, &default_acl, &acl);
 	if (err)
diff --git a/fs/romfs/Kconfig b/fs/romfs/Kconfig
index ad4c45788896..9737b8e68878 100644
--- a/fs/romfs/Kconfig
+++ b/fs/romfs/Kconfig
@@ -6,7 +6,7 @@ config ROMFS_FS
 	  This is a very small read-only file system mainly intended for
 	  initial ram disks of installation disks, but it could be used for
 	  other read-only media as well.  Read
-	  <file:Documentation/filesystems/romfs.txt> for details.
+	  <file:Documentation/filesystems/romfs.rst> for details.
 
 	  To compile this file system support as a module, choose M here: the
 	  module will be called romfs.  Note that the file system of your
diff --git a/fs/select.c b/fs/select.c
index 53a0c149f528..11d0285d46b7 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -321,7 +321,7 @@ static int poll_select_finish(struct timespec64 *end_time,
 	switch (pt_type) {
 	case PT_TIMEVAL:
 		{
-			struct timeval rtv;
+			struct __kernel_old_timeval rtv;
 
 			if (sizeof(rtv) > sizeof(rtv.tv_sec) + sizeof(rtv.tv_usec))
 				memset(&rtv, 0, sizeof(rtv));
@@ -698,10 +698,10 @@ out_nofds:
 }
 
 static int kern_select(int n, fd_set __user *inp, fd_set __user *outp,
-		       fd_set __user *exp, struct timeval __user *tvp)
+		       fd_set __user *exp, struct __kernel_old_timeval __user *tvp)
 {
 	struct timespec64 end_time, *to = NULL;
-	struct timeval tv;
+	struct __kernel_old_timeval tv;
 	int ret;
 
 	if (tvp) {
@@ -720,7 +720,7 @@ static int kern_select(int n, fd_set __user *inp, fd_set __user *outp,
 }
 
 SYSCALL_DEFINE5(select, int, n, fd_set __user *, inp, fd_set __user *, outp,
-		fd_set __user *, exp, struct timeval __user *, tvp)
+		fd_set __user *, exp, struct __kernel_old_timeval __user *, tvp)
 {
 	return kern_select(n, inp, outp, exp, tvp);
 }
@@ -810,7 +810,7 @@ SYSCALL_DEFINE6(pselect6_time32, int, n, fd_set __user *, inp, fd_set __user *,
 struct sel_arg_struct {
 	unsigned long n;
 	fd_set __user *inp, *outp, *exp;
-	struct timeval __user *tvp;
+	struct __kernel_old_timeval __user *tvp;
 };
 
 SYSCALL_DEFINE1(old_select, struct sel_arg_struct __user *, arg)
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 1600034a929b..70f5fdf99bf6 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -68,13 +68,6 @@ int seq_open(struct file *file, const struct seq_operations *op)
 	p->file = file;
 
 	/*
-	 * Wrappers around seq_open(e.g. swaps_open) need to be
-	 * aware of this. If they set f_version themselves, they
-	 * should call seq_open first and then set f_version.
-	 */
-	file->f_version = 0;
-
-	/*
 	 * seq_files support lseek() and pread().  They do not implement
 	 * write() at all, but we clear FMODE_PWRITE here for historical
 	 * reasons.
@@ -94,7 +87,6 @@ static int traverse(struct seq_file *m, loff_t offset)
 	int error = 0;
 	void *p;
 
-	m->version = 0;
 	m->index = 0;
 	m->count = m->from = 0;
 	if (!offset)
@@ -161,25 +153,11 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
 	mutex_lock(&m->lock);
 
 	/*
-	 * seq_file->op->..m_start/m_stop/m_next may do special actions
-	 * or optimisations based on the file->f_version, so we want to
-	 * pass the file->f_version to those methods.
-	 *
-	 * seq_file->version is just copy of f_version, and seq_file
-	 * methods can treat it simply as file version.
-	 * It is copied in first and copied out after all operations.
-	 * It is convenient to have it as  part of structure to avoid the
-	 * need of passing another argument to all the seq_file methods.
-	 */
-	m->version = file->f_version;
-
-	/*
 	 * if request is to read from zero offset, reset iterator to first
 	 * record as it might have been already advanced by previous requests
 	 */
 	if (*ppos == 0) {
 		m->index = 0;
-		m->version = 0;
 		m->count = 0;
 	}
 
@@ -190,7 +168,6 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
 		if (err) {
 			/* With prejudice... */
 			m->read_pos = 0;
-			m->version = 0;
 			m->index = 0;
 			m->count = 0;
 			goto Done;
@@ -243,7 +220,6 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
 		m->buf = seq_buf_alloc(m->size <<= 1);
 		if (!m->buf)
 			goto Enomem;
-		m->version = 0;
 		p = m->op->start(m, &m->index);
 	}
 	m->op->stop(m, p);
@@ -256,9 +232,12 @@ Fill:
 		loff_t pos = m->index;
 
 		p = m->op->next(m, p, &m->index);
-		if (pos == m->index)
-			/* Buggy ->next function */
+		if (pos == m->index) {
+			pr_info_ratelimited("buggy seq_file .next function %ps "
+				"did not updated position index\n",
+				m->op->next);
 			m->index++;
+		}
 		if (!p || IS_ERR(p)) {
 			err = PTR_ERR(p);
 			break;
@@ -287,7 +266,6 @@ Done:
 		*ppos += copied;
 		m->read_pos += copied;
 	}
-	file->f_version = m->version;
 	mutex_unlock(&m->lock);
 	return copied;
 Enomem:
@@ -313,7 +291,6 @@ loff_t seq_lseek(struct file *file, loff_t offset, int whence)
 	loff_t retval = -EINVAL;
 
 	mutex_lock(&m->lock);
-	m->version = file->f_version;
 	switch (whence) {
 	case SEEK_CUR:
 		offset += file->f_pos;
@@ -329,7 +306,6 @@ loff_t seq_lseek(struct file *file, loff_t offset, int whence)
 				/* with extreme prejudice... */
 				file->f_pos = 0;
 				m->read_pos = 0;
-				m->version = 0;
 				m->index = 0;
 				m->count = 0;
 			} else {
@@ -340,7 +316,6 @@ loff_t seq_lseek(struct file *file, loff_t offset, int whence)
 			file->f_pos = offset;
 		}
 	}
-	file->f_version = m->version;
 	mutex_unlock(&m->lock);
 	return retval;
 }
diff --git a/fs/splice.c b/fs/splice.c
index 98412721f056..5013565eb756 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -165,8 +165,8 @@ static const struct pipe_buf_operations user_page_pipe_buf_ops = {
 static void wakeup_pipe_readers(struct pipe_inode_info *pipe)
 {
 	smp_mb();
-	if (waitqueue_active(&pipe->wait))
-		wake_up_interruptible(&pipe->wait);
+	if (waitqueue_active(&pipe->rd_wait))
+		wake_up_interruptible(&pipe->rd_wait);
 	kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
 }
 
@@ -185,6 +185,9 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe,
 		       struct splice_pipe_desc *spd)
 {
 	unsigned int spd_pages = spd->nr_pages;
+	unsigned int tail = pipe->tail;
+	unsigned int head = pipe->head;
+	unsigned int mask = pipe->ring_size - 1;
 	int ret = 0, page_nr = 0;
 
 	if (!spd_pages)
@@ -196,9 +199,8 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe,
 		goto out;
 	}
 
-	while (pipe->nrbufs < pipe->buffers) {
-		int newbuf = (pipe->curbuf + pipe->nrbufs) & (pipe->buffers - 1);
-		struct pipe_buffer *buf = pipe->bufs + newbuf;
+	while (!pipe_full(head, tail, pipe->max_usage)) {
+		struct pipe_buffer *buf = &pipe->bufs[head & mask];
 
 		buf->page = spd->pages[page_nr];
 		buf->offset = spd->partial[page_nr].offset;
@@ -207,7 +209,8 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe,
 		buf->ops = spd->ops;
 		buf->flags = 0;
 
-		pipe->nrbufs++;
+		head++;
+		pipe->head = head;
 		page_nr++;
 		ret += buf->len;
 
@@ -228,17 +231,19 @@ EXPORT_SYMBOL_GPL(splice_to_pipe);
 
 ssize_t add_to_pipe(struct pipe_inode_info *pipe, struct pipe_buffer *buf)
 {
+	unsigned int head = pipe->head;
+	unsigned int tail = pipe->tail;
+	unsigned int mask = pipe->ring_size - 1;
 	int ret;
 
 	if (unlikely(!pipe->readers)) {
 		send_sig(SIGPIPE, current, 0);
 		ret = -EPIPE;
-	} else if (pipe->nrbufs == pipe->buffers) {
+	} else if (pipe_full(head, tail, pipe->max_usage)) {
 		ret = -EAGAIN;
 	} else {
-		int newbuf = (pipe->curbuf + pipe->nrbufs) & (pipe->buffers - 1);
-		pipe->bufs[newbuf] = *buf;
-		pipe->nrbufs++;
+		pipe->bufs[head & mask] = *buf;
+		pipe->head = head + 1;
 		return buf->len;
 	}
 	pipe_buf_release(pipe, buf);
@@ -252,14 +257,14 @@ EXPORT_SYMBOL(add_to_pipe);
  */
 int splice_grow_spd(const struct pipe_inode_info *pipe, struct splice_pipe_desc *spd)
 {
-	unsigned int buffers = READ_ONCE(pipe->buffers);
+	unsigned int max_usage = READ_ONCE(pipe->max_usage);
 
-	spd->nr_pages_max = buffers;
-	if (buffers <= PIPE_DEF_BUFFERS)
+	spd->nr_pages_max = max_usage;
+	if (max_usage <= PIPE_DEF_BUFFERS)
 		return 0;
 
-	spd->pages = kmalloc_array(buffers, sizeof(struct page *), GFP_KERNEL);
-	spd->partial = kmalloc_array(buffers, sizeof(struct partial_page),
+	spd->pages = kmalloc_array(max_usage, sizeof(struct page *), GFP_KERNEL);
+	spd->partial = kmalloc_array(max_usage, sizeof(struct partial_page),
 				     GFP_KERNEL);
 
 	if (spd->pages && spd->partial)
@@ -298,10 +303,11 @@ ssize_t generic_file_splice_read(struct file *in, loff_t *ppos,
 {
 	struct iov_iter to;
 	struct kiocb kiocb;
-	int idx, ret;
+	unsigned int i_head;
+	int ret;
 
 	iov_iter_pipe(&to, READ, pipe, len);
-	idx = to.idx;
+	i_head = to.head;
 	init_sync_kiocb(&kiocb, in);
 	kiocb.ki_pos = *ppos;
 	ret = call_read_iter(in, &kiocb, &to);
@@ -309,7 +315,7 @@ ssize_t generic_file_splice_read(struct file *in, loff_t *ppos,
 		*ppos = kiocb.ki_pos;
 		file_accessed(in);
 	} else if (ret < 0) {
-		to.idx = idx;
+		to.head = i_head;
 		to.iov_offset = 0;
 		iov_iter_advance(&to, 0); /* to free what was emitted */
 		/*
@@ -370,11 +376,12 @@ static ssize_t default_file_splice_read(struct file *in, loff_t *ppos,
 	struct iov_iter to;
 	struct page **pages;
 	unsigned int nr_pages;
+	unsigned int mask;
 	size_t offset, base, copied = 0;
 	ssize_t res;
 	int i;
 
-	if (pipe->nrbufs == pipe->buffers)
+	if (pipe_full(pipe->head, pipe->tail, pipe->max_usage))
 		return -EAGAIN;
 
 	/*
@@ -400,8 +407,9 @@ static ssize_t default_file_splice_read(struct file *in, loff_t *ppos,
 		}
 	}
 
-	pipe->bufs[to.idx].offset = offset;
-	pipe->bufs[to.idx].len -= offset;
+	mask = pipe->ring_size - 1;
+	pipe->bufs[to.head & mask].offset = offset;
+	pipe->bufs[to.head & mask].len -= offset;
 
 	for (i = 0; i < nr_pages; i++) {
 		size_t this_len = min_t(size_t, len, PAGE_SIZE - offset);
@@ -443,7 +451,8 @@ static int pipe_to_sendpage(struct pipe_inode_info *pipe,
 
 	more = (sd->flags & SPLICE_F_MORE) ? MSG_MORE : 0;
 
-	if (sd->len < sd->total_len && pipe->nrbufs > 1)
+	if (sd->len < sd->total_len &&
+	    pipe_occupancy(pipe->head, pipe->tail) > 1)
 		more |= MSG_SENDPAGE_NOTLAST;
 
 	return file->f_op->sendpage(file, buf->page, buf->offset,
@@ -453,8 +462,8 @@ static int pipe_to_sendpage(struct pipe_inode_info *pipe,
 static void wakeup_pipe_writers(struct pipe_inode_info *pipe)
 {
 	smp_mb();
-	if (waitqueue_active(&pipe->wait))
-		wake_up_interruptible(&pipe->wait);
+	if (waitqueue_active(&pipe->wr_wait))
+		wake_up_interruptible(&pipe->wr_wait);
 	kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT);
 }
 
@@ -481,10 +490,13 @@ static void wakeup_pipe_writers(struct pipe_inode_info *pipe)
 static int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_desc *sd,
 			  splice_actor *actor)
 {
+	unsigned int head = pipe->head;
+	unsigned int tail = pipe->tail;
+	unsigned int mask = pipe->ring_size - 1;
 	int ret;
 
-	while (pipe->nrbufs) {
-		struct pipe_buffer *buf = pipe->bufs + pipe->curbuf;
+	while (!pipe_empty(head, tail)) {
+		struct pipe_buffer *buf = &pipe->bufs[tail & mask];
 
 		sd->len = buf->len;
 		if (sd->len > sd->total_len)
@@ -511,8 +523,8 @@ static int splice_from_pipe_feed(struct pipe_inode_info *pipe, struct splice_des
 
 		if (!buf->len) {
 			pipe_buf_release(pipe, buf);
-			pipe->curbuf = (pipe->curbuf + 1) & (pipe->buffers - 1);
-			pipe->nrbufs--;
+			tail++;
+			pipe->tail = tail;
 			if (pipe->files)
 				sd->need_wakeup = true;
 		}
@@ -543,11 +555,11 @@ static int splice_from_pipe_next(struct pipe_inode_info *pipe, struct splice_des
 	if (signal_pending(current))
 		return -ERESTARTSYS;
 
-	while (!pipe->nrbufs) {
+	while (pipe_empty(pipe->head, pipe->tail)) {
 		if (!pipe->writers)
 			return 0;
 
-		if (!pipe->waiting_writers && sd->num_spliced)
+		if (sd->num_spliced)
 			return 0;
 
 		if (sd->flags & SPLICE_F_NONBLOCK)
@@ -686,7 +698,7 @@ iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
 		.pos = *ppos,
 		.u.file = out,
 	};
-	int nbufs = pipe->buffers;
+	int nbufs = pipe->max_usage;
 	struct bio_vec *array = kcalloc(nbufs, sizeof(struct bio_vec),
 					GFP_KERNEL);
 	ssize_t ret;
@@ -699,16 +711,17 @@ iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
 	splice_from_pipe_begin(&sd);
 	while (sd.total_len) {
 		struct iov_iter from;
+		unsigned int head, tail, mask;
 		size_t left;
-		int n, idx;
+		int n;
 
 		ret = splice_from_pipe_next(pipe, &sd);
 		if (ret <= 0)
 			break;
 
-		if (unlikely(nbufs < pipe->buffers)) {
+		if (unlikely(nbufs < pipe->max_usage)) {
 			kfree(array);
-			nbufs = pipe->buffers;
+			nbufs = pipe->max_usage;
 			array = kcalloc(nbufs, sizeof(struct bio_vec),
 					GFP_KERNEL);
 			if (!array) {
@@ -717,18 +730,19 @@ iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
 			}
 		}
 
+		head = pipe->head;
+		tail = pipe->tail;
+		mask = pipe->ring_size - 1;
+
 		/* build the vector */
 		left = sd.total_len;
-		for (n = 0, idx = pipe->curbuf; left && n < pipe->nrbufs; n++, idx++) {
-			struct pipe_buffer *buf = pipe->bufs + idx;
+		for (n = 0; !pipe_empty(head, tail) && left && n < nbufs; tail++, n++) {
+			struct pipe_buffer *buf = &pipe->bufs[tail & mask];
 			size_t this_len = buf->len;
 
 			if (this_len > left)
 				this_len = left;
 
-			if (idx == pipe->buffers - 1)
-				idx = -1;
-
 			ret = pipe_buf_confirm(pipe, buf);
 			if (unlikely(ret)) {
 				if (ret == -ENODATA)
@@ -752,14 +766,15 @@ iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
 		*ppos = sd.pos;
 
 		/* dismiss the fully eaten buffers, adjust the partial one */
+		tail = pipe->tail;
 		while (ret) {
-			struct pipe_buffer *buf = pipe->bufs + pipe->curbuf;
+			struct pipe_buffer *buf = &pipe->bufs[tail & mask];
 			if (ret >= buf->len) {
 				ret -= buf->len;
 				buf->len = 0;
 				pipe_buf_release(pipe, buf);
-				pipe->curbuf = (pipe->curbuf + 1) & (pipe->buffers - 1);
-				pipe->nrbufs--;
+				tail++;
+				pipe->tail = tail;
 				if (pipe->files)
 					sd.need_wakeup = true;
 			} else {
@@ -942,15 +957,17 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd,
 	sd->flags &= ~SPLICE_F_NONBLOCK;
 	more = sd->flags & SPLICE_F_MORE;
 
-	WARN_ON_ONCE(pipe->nrbufs != 0);
+	WARN_ON_ONCE(!pipe_empty(pipe->head, pipe->tail));
 
 	while (len) {
+		unsigned int p_space;
 		size_t read_len;
 		loff_t pos = sd->pos, prev_pos = pos;
 
 		/* Don't try to read more the pipe has space for. */
-		read_len = min_t(size_t, len,
-				 (pipe->buffers - pipe->nrbufs) << PAGE_SHIFT);
+		p_space = pipe->max_usage -
+			pipe_occupancy(pipe->head, pipe->tail);
+		read_len = min_t(size_t, len, p_space << PAGE_SHIFT);
 		ret = do_splice_to(in, &pos, pipe, read_len, flags);
 		if (unlikely(ret <= 0))
 			goto out_release;
@@ -989,7 +1006,7 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd,
 	}
 
 done:
-	pipe->nrbufs = pipe->curbuf = 0;
+	pipe->tail = pipe->head = 0;
 	file_accessed(in);
 	return bytes;
 
@@ -998,8 +1015,8 @@ out_release:
 	 * If we did an incomplete transfer we must release
 	 * the pipe buffers in question:
 	 */
-	for (i = 0; i < pipe->buffers; i++) {
-		struct pipe_buffer *buf = pipe->bufs + i;
+	for (i = 0; i < pipe->ring_size; i++) {
+		struct pipe_buffer *buf = &pipe->bufs[i];
 
 		if (buf->ops)
 			pipe_buf_release(pipe, buf);
@@ -1075,15 +1092,13 @@ static int wait_for_space(struct pipe_inode_info *pipe, unsigned flags)
 			send_sig(SIGPIPE, current, 0);
 			return -EPIPE;
 		}
-		if (pipe->nrbufs != pipe->buffers)
+		if (!pipe_full(pipe->head, pipe->tail, pipe->max_usage))
 			return 0;
 		if (flags & SPLICE_F_NONBLOCK)
 			return -EAGAIN;
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		pipe->waiting_writers++;
 		pipe_wait(pipe);
-		pipe->waiting_writers--;
 	}
 }
 
@@ -1094,15 +1109,19 @@ static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe,
 /*
  * Determine where to splice to/from.
  */
-static long do_splice(struct file *in, loff_t __user *off_in,
-		      struct file *out, loff_t __user *off_out,
-		      size_t len, unsigned int flags)
+long do_splice(struct file *in, loff_t __user *off_in,
+		struct file *out, loff_t __user *off_out,
+		size_t len, unsigned int flags)
 {
 	struct pipe_inode_info *ipipe;
 	struct pipe_inode_info *opipe;
 	loff_t offset;
 	long ret;
 
+	if (unlikely(!(in->f_mode & FMODE_READ) ||
+		     !(out->f_mode & FMODE_WRITE)))
+		return -EBADF;
+
 	ipipe = get_pipe_info(in);
 	opipe = get_pipe_info(out);
 
@@ -1110,12 +1129,6 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 		if (off_in || off_out)
 			return -ESPIPE;
 
-		if (!(in->f_mode & FMODE_READ))
-			return -EBADF;
-
-		if (!(out->f_mode & FMODE_WRITE))
-			return -EBADF;
-
 		/* Splicing to self would be fun, but... */
 		if (ipipe == opipe)
 			return -EINVAL;
@@ -1138,9 +1151,6 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 			offset = out->f_pos;
 		}
 
-		if (unlikely(!(out->f_mode & FMODE_WRITE)))
-			return -EBADF;
-
 		if (unlikely(out->f_flags & O_APPEND))
 			return -EINVAL;
 
@@ -1180,8 +1190,15 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 
 		pipe_lock(opipe);
 		ret = wait_for_space(opipe, flags);
-		if (!ret)
+		if (!ret) {
+			unsigned int p_space;
+
+			/* Don't try to read more the pipe has space for. */
+			p_space = opipe->max_usage - pipe_occupancy(opipe->head, opipe->tail);
+			len = min_t(size_t, len, p_space << PAGE_SHIFT);
+
 			ret = do_splice_to(in, &offset, opipe, len, flags);
+		}
 		pipe_unlock(opipe);
 		if (ret > 0)
 			wakeup_pipe_readers(opipe);
@@ -1418,15 +1435,11 @@ SYSCALL_DEFINE6(splice, int, fd_in, loff_t __user *, off_in,
 	error = -EBADF;
 	in = fdget(fd_in);
 	if (in.file) {
-		if (in.file->f_mode & FMODE_READ) {
-			out = fdget(fd_out);
-			if (out.file) {
-				if (out.file->f_mode & FMODE_WRITE)
-					error = do_splice(in.file, off_in,
-							  out.file, off_out,
-							  len, flags);
-				fdput(out);
-			}
+		out = fdget(fd_out);
+		if (out.file) {
+			error = do_splice(in.file, off_in, out.file, off_out,
+					  len, flags);
+			fdput(out);
 		}
 		fdput(in);
 	}
@@ -1442,27 +1455,25 @@ static int ipipe_prep(struct pipe_inode_info *pipe, unsigned int flags)
 	int ret;
 
 	/*
-	 * Check ->nrbufs without the inode lock first. This function
+	 * Check the pipe occupancy without the inode lock first. This function
 	 * is speculative anyways, so missing one is ok.
 	 */
-	if (pipe->nrbufs)
+	if (!pipe_empty(pipe->head, pipe->tail))
 		return 0;
 
 	ret = 0;
 	pipe_lock(pipe);
 
-	while (!pipe->nrbufs) {
+	while (pipe_empty(pipe->head, pipe->tail)) {
 		if (signal_pending(current)) {
 			ret = -ERESTARTSYS;
 			break;
 		}
 		if (!pipe->writers)
 			break;
-		if (!pipe->waiting_writers) {
-			if (flags & SPLICE_F_NONBLOCK) {
-				ret = -EAGAIN;
-				break;
-			}
+		if (flags & SPLICE_F_NONBLOCK) {
+			ret = -EAGAIN;
+			break;
 		}
 		pipe_wait(pipe);
 	}
@@ -1480,16 +1491,16 @@ static int opipe_prep(struct pipe_inode_info *pipe, unsigned int flags)
 	int ret;
 
 	/*
-	 * Check ->nrbufs without the inode lock first. This function
+	 * Check pipe occupancy without the inode lock first. This function
 	 * is speculative anyways, so missing one is ok.
 	 */
-	if (pipe->nrbufs < pipe->buffers)
+	if (!pipe_full(pipe->head, pipe->tail, pipe->max_usage))
 		return 0;
 
 	ret = 0;
 	pipe_lock(pipe);
 
-	while (pipe->nrbufs >= pipe->buffers) {
+	while (pipe_full(pipe->head, pipe->tail, pipe->max_usage)) {
 		if (!pipe->readers) {
 			send_sig(SIGPIPE, current, 0);
 			ret = -EPIPE;
@@ -1503,9 +1514,7 @@ static int opipe_prep(struct pipe_inode_info *pipe, unsigned int flags)
 			ret = -ERESTARTSYS;
 			break;
 		}
-		pipe->waiting_writers++;
 		pipe_wait(pipe);
-		pipe->waiting_writers--;
 	}
 
 	pipe_unlock(pipe);
@@ -1520,7 +1529,10 @@ static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe,
 			       size_t len, unsigned int flags)
 {
 	struct pipe_buffer *ibuf, *obuf;
-	int ret = 0, nbuf;
+	unsigned int i_head, o_head;
+	unsigned int i_tail, o_tail;
+	unsigned int i_mask, o_mask;
+	int ret = 0;
 	bool input_wakeup = false;
 
 
@@ -1540,7 +1552,14 @@ retry:
 	 */
 	pipe_double_lock(ipipe, opipe);
 
+	i_tail = ipipe->tail;
+	i_mask = ipipe->ring_size - 1;
+	o_head = opipe->head;
+	o_mask = opipe->ring_size - 1;
+
 	do {
+		size_t o_len;
+
 		if (!opipe->readers) {
 			send_sig(SIGPIPE, current, 0);
 			if (!ret)
@@ -1548,14 +1567,18 @@ retry:
 			break;
 		}
 
-		if (!ipipe->nrbufs && !ipipe->writers)
+		i_head = ipipe->head;
+		o_tail = opipe->tail;
+
+		if (pipe_empty(i_head, i_tail) && !ipipe->writers)
 			break;
 
 		/*
 		 * Cannot make any progress, because either the input
 		 * pipe is empty or the output pipe is full.
 		 */
-		if (!ipipe->nrbufs || opipe->nrbufs >= opipe->buffers) {
+		if (pipe_empty(i_head, i_tail) ||
+		    pipe_full(o_head, o_tail, opipe->max_usage)) {
 			/* Already processed some buffers, break */
 			if (ret)
 				break;
@@ -1575,9 +1598,8 @@ retry:
 			goto retry;
 		}
 
-		ibuf = ipipe->bufs + ipipe->curbuf;
-		nbuf = (opipe->curbuf + opipe->nrbufs) & (opipe->buffers - 1);
-		obuf = opipe->bufs + nbuf;
+		ibuf = &ipipe->bufs[i_tail & i_mask];
+		obuf = &opipe->bufs[o_head & o_mask];
 
 		if (len >= ibuf->len) {
 			/*
@@ -1585,10 +1607,12 @@ retry:
 			 */
 			*obuf = *ibuf;
 			ibuf->ops = NULL;
-			opipe->nrbufs++;
-			ipipe->curbuf = (ipipe->curbuf + 1) & (ipipe->buffers - 1);
-			ipipe->nrbufs--;
+			i_tail++;
+			ipipe->tail = i_tail;
 			input_wakeup = true;
+			o_len = obuf->len;
+			o_head++;
+			opipe->head = o_head;
 		} else {
 			/*
 			 * Get a reference to this pipe buffer,
@@ -1610,12 +1634,14 @@ retry:
 			pipe_buf_mark_unmergeable(obuf);
 
 			obuf->len = len;
-			opipe->nrbufs++;
-			ibuf->offset += obuf->len;
-			ibuf->len -= obuf->len;
+			ibuf->offset += len;
+			ibuf->len -= len;
+			o_len = len;
+			o_head++;
+			opipe->head = o_head;
 		}
-		ret += obuf->len;
-		len -= obuf->len;
+		ret += o_len;
+		len -= o_len;
 	} while (len);
 
 	pipe_unlock(ipipe);
@@ -1641,7 +1667,10 @@ static int link_pipe(struct pipe_inode_info *ipipe,
 		     size_t len, unsigned int flags)
 {
 	struct pipe_buffer *ibuf, *obuf;
-	int ret = 0, i = 0, nbuf;
+	unsigned int i_head, o_head;
+	unsigned int i_tail, o_tail;
+	unsigned int i_mask, o_mask;
+	int ret = 0;
 
 	/*
 	 * Potential ABBA deadlock, work around it by ordering lock
@@ -1650,6 +1679,11 @@ static int link_pipe(struct pipe_inode_info *ipipe,
 	 */
 	pipe_double_lock(ipipe, opipe);
 
+	i_tail = ipipe->tail;
+	i_mask = ipipe->ring_size - 1;
+	o_head = opipe->head;
+	o_mask = opipe->ring_size - 1;
+
 	do {
 		if (!opipe->readers) {
 			send_sig(SIGPIPE, current, 0);
@@ -1658,15 +1692,19 @@ static int link_pipe(struct pipe_inode_info *ipipe,
 			break;
 		}
 
+		i_head = ipipe->head;
+		o_tail = opipe->tail;
+
 		/*
-		 * If we have iterated all input buffers or ran out of
+		 * If we have iterated all input buffers or run out of
 		 * output room, break.
 		 */
-		if (i >= ipipe->nrbufs || opipe->nrbufs >= opipe->buffers)
+		if (pipe_empty(i_head, i_tail) ||
+		    pipe_full(o_head, o_tail, opipe->max_usage))
 			break;
 
-		ibuf = ipipe->bufs + ((ipipe->curbuf + i) & (ipipe->buffers-1));
-		nbuf = (opipe->curbuf + opipe->nrbufs) & (opipe->buffers - 1);
+		ibuf = &ipipe->bufs[i_tail & i_mask];
+		obuf = &opipe->bufs[o_head & o_mask];
 
 		/*
 		 * Get a reference to this pipe buffer,
@@ -1678,7 +1716,6 @@ static int link_pipe(struct pipe_inode_info *ipipe,
 			break;
 		}
 
-		obuf = opipe->bufs + nbuf;
 		*obuf = *ibuf;
 
 		/*
@@ -1691,19 +1728,13 @@ static int link_pipe(struct pipe_inode_info *ipipe,
 
 		if (obuf->len > len)
 			obuf->len = len;
-
-		opipe->nrbufs++;
 		ret += obuf->len;
 		len -= obuf->len;
-		i++;
-	} while (len);
 
-	/*
-	 * return EAGAIN if we have the potential of some data in the
-	 * future, otherwise just return 0
-	 */
-	if (!ret && ipipe->waiting_writers && (flags & SPLICE_F_NONBLOCK))
-		ret = -EAGAIN;
+		o_head++;
+		opipe->head = o_head;
+		i_tail++;
+	} while (len);
 
 	pipe_unlock(ipipe);
 	pipe_unlock(opipe);
@@ -1723,13 +1754,16 @@ static int link_pipe(struct pipe_inode_info *ipipe,
  * The 'flags' used are the SPLICE_F_* variants, currently the only
  * applicable one is SPLICE_F_NONBLOCK.
  */
-static long do_tee(struct file *in, struct file *out, size_t len,
-		   unsigned int flags)
+long do_tee(struct file *in, struct file *out, size_t len, unsigned int flags)
 {
 	struct pipe_inode_info *ipipe = get_pipe_info(in);
 	struct pipe_inode_info *opipe = get_pipe_info(out);
 	int ret = -EINVAL;
 
+	if (unlikely(!(in->f_mode & FMODE_READ) ||
+		     !(out->f_mode & FMODE_WRITE)))
+		return -EBADF;
+
 	/*
 	 * Duplicate the contents of ipipe to opipe without actually
 	 * copying the data.
@@ -1755,7 +1789,7 @@ static long do_tee(struct file *in, struct file *out, size_t len,
 
 SYSCALL_DEFINE4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags)
 {
-	struct fd in;
+	struct fd in, out;
 	int error;
 
 	if (unlikely(flags & ~SPLICE_F_ALL))
@@ -1767,14 +1801,10 @@ SYSCALL_DEFINE4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags)
 	error = -EBADF;
 	in = fdget(fdin);
 	if (in.file) {
-		if (in.file->f_mode & FMODE_READ) {
-			struct fd out = fdget(fdout);
-			if (out.file) {
-				if (out.file->f_mode & FMODE_WRITE)
-					error = do_tee(in.file, out.file,
-							len, flags);
-				fdput(out);
-			}
+		out = fdget(fdout);
+		if (out.file) {
+			error = do_tee(in.file, out.file, len, flags);
+			fdput(out);
 		}
  		fdput(in);
  	}
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 4f9b9fb59362..64f61330564a 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -13,6 +13,7 @@
  * datablocks and metadata blocks.
  */
 
+#include <linux/blkdev.h>
 #include <linux/fs.h>
 #include <linux/vfs.h>
 #include <linux/slab.h>
@@ -27,44 +28,103 @@
 #include "page_actor.h"
 
 /*
- * Read the metadata block length, this is stored in the first two
- * bytes of the metadata block.
+ * Returns the amount of bytes copied to the page actor.
  */
-static struct buffer_head *get_block_length(struct super_block *sb,
-			u64 *cur_index, int *offset, int *length)
+static int copy_bio_to_actor(struct bio *bio,
+			     struct squashfs_page_actor *actor,
+			     int offset, int req_length)
+{
+	void *actor_addr = squashfs_first_page(actor);
+	struct bvec_iter_all iter_all = {};
+	struct bio_vec *bvec = bvec_init_iter_all(&iter_all);
+	int copied_bytes = 0;
+	int actor_offset = 0;
+
+	if (WARN_ON_ONCE(!bio_next_segment(bio, &iter_all)))
+		return 0;
+
+	while (copied_bytes < req_length) {
+		int bytes_to_copy = min_t(int, bvec->bv_len - offset,
+					  PAGE_SIZE - actor_offset);
+
+		bytes_to_copy = min_t(int, bytes_to_copy,
+				      req_length - copied_bytes);
+		memcpy(actor_addr + actor_offset,
+		       page_address(bvec->bv_page) + bvec->bv_offset + offset,
+		       bytes_to_copy);
+
+		actor_offset += bytes_to_copy;
+		copied_bytes += bytes_to_copy;
+		offset += bytes_to_copy;
+
+		if (actor_offset >= PAGE_SIZE) {
+			actor_addr = squashfs_next_page(actor);
+			if (!actor_addr)
+				break;
+			actor_offset = 0;
+		}
+		if (offset >= bvec->bv_len) {
+			if (!bio_next_segment(bio, &iter_all))
+				break;
+			offset = 0;
+		}
+	}
+	squashfs_finish_page(actor);
+	return copied_bytes;
+}
+
+static int squashfs_bio_read(struct super_block *sb, u64 index, int length,
+			     struct bio **biop, int *block_offset)
 {
 	struct squashfs_sb_info *msblk = sb->s_fs_info;
-	struct buffer_head *bh;
-
-	bh = sb_bread(sb, *cur_index);
-	if (bh == NULL)
-		return NULL;
-
-	if (msblk->devblksize - *offset == 1) {
-		*length = (unsigned char) bh->b_data[*offset];
-		put_bh(bh);
-		bh = sb_bread(sb, ++(*cur_index));
-		if (bh == NULL)
-			return NULL;
-		*length |= (unsigned char) bh->b_data[0] << 8;
-		*offset = 1;
-	} else {
-		*length = (unsigned char) bh->b_data[*offset] |
-			(unsigned char) bh->b_data[*offset + 1] << 8;
-		*offset += 2;
-
-		if (*offset == msblk->devblksize) {
-			put_bh(bh);
-			bh = sb_bread(sb, ++(*cur_index));
-			if (bh == NULL)
-				return NULL;
-			*offset = 0;
+	const u64 read_start = round_down(index, msblk->devblksize);
+	const sector_t block = read_start >> msblk->devblksize_log2;
+	const u64 read_end = round_up(index + length, msblk->devblksize);
+	const sector_t block_end = read_end >> msblk->devblksize_log2;
+	int offset = read_start - round_down(index, PAGE_SIZE);
+	int total_len = (block_end - block) << msblk->devblksize_log2;
+	const int page_count = DIV_ROUND_UP(total_len + offset, PAGE_SIZE);
+	int error, i;
+	struct bio *bio;
+
+	bio = bio_alloc(GFP_NOIO, page_count);
+	if (!bio)
+		return -ENOMEM;
+
+	bio_set_dev(bio, sb->s_bdev);
+	bio->bi_opf = READ;
+	bio->bi_iter.bi_sector = block * (msblk->devblksize >> SECTOR_SHIFT);
+
+	for (i = 0; i < page_count; ++i) {
+		unsigned int len =
+			min_t(unsigned int, PAGE_SIZE - offset, total_len);
+		struct page *page = alloc_page(GFP_NOIO);
+
+		if (!page) {
+			error = -ENOMEM;
+			goto out_free_bio;
+		}
+		if (!bio_add_page(bio, page, len, offset)) {
+			error = -EIO;
+			goto out_free_bio;
 		}
+		offset = 0;
+		total_len -= len;
 	}
 
-	return bh;
-}
+	error = submit_bio_wait(bio);
+	if (error)
+		goto out_free_bio;
 
+	*biop = bio;
+	*block_offset = index & ((1 << msblk->devblksize_log2) - 1);
+	return 0;
+
+out_free_bio:
+	bio_free_pages(bio);
+	bio_put(bio);
+	return error;
+}
 
 /*
  * Read and decompress a metadata block or datablock.  Length is non-zero
@@ -76,129 +136,88 @@ static struct buffer_head *get_block_length(struct super_block *sb,
  * algorithms).
  */
 int squashfs_read_data(struct super_block *sb, u64 index, int length,
-		u64 *next_index, struct squashfs_page_actor *output)
+		       u64 *next_index, struct squashfs_page_actor *output)
 {
 	struct squashfs_sb_info *msblk = sb->s_fs_info;
-	struct buffer_head **bh;
-	int offset = index & ((1 << msblk->devblksize_log2) - 1);
-	u64 cur_index = index >> msblk->devblksize_log2;
-	int bytes, compressed, b = 0, k = 0, avail, i;
-
-	bh = kcalloc(((output->length + msblk->devblksize - 1)
-		>> msblk->devblksize_log2) + 1, sizeof(*bh), GFP_KERNEL);
-	if (bh == NULL)
-		return -ENOMEM;
+	struct bio *bio = NULL;
+	int compressed;
+	int res;
+	int offset;
 
 	if (length) {
 		/*
 		 * Datablock.
 		 */
-		bytes = -offset;
 		compressed = SQUASHFS_COMPRESSED_BLOCK(length);
 		length = SQUASHFS_COMPRESSED_SIZE_BLOCK(length);
-		if (next_index)
-			*next_index = index + length;
-
 		TRACE("Block @ 0x%llx, %scompressed size %d, src size %d\n",
 			index, compressed ? "" : "un", length, output->length);
-
-		if (length < 0 || length > output->length ||
-				(index + length) > msblk->bytes_used)
-			goto read_failure;
-
-		for (b = 0; bytes < length; b++, cur_index++) {
-			bh[b] = sb_getblk(sb, cur_index);
-			if (bh[b] == NULL)
-				goto block_release;
-			bytes += msblk->devblksize;
-		}
-		ll_rw_block(REQ_OP_READ, 0, b, bh);
 	} else {
 		/*
 		 * Metadata block.
 		 */
-		if ((index + 2) > msblk->bytes_used)
-			goto read_failure;
+		const u8 *data;
+		struct bvec_iter_all iter_all = {};
+		struct bio_vec *bvec = bvec_init_iter_all(&iter_all);
 
-		bh[0] = get_block_length(sb, &cur_index, &offset, &length);
-		if (bh[0] == NULL)
-			goto read_failure;
-		b = 1;
+		if (index + 2 > msblk->bytes_used) {
+			res = -EIO;
+			goto out;
+		}
+		res = squashfs_bio_read(sb, index, 2, &bio, &offset);
+		if (res)
+			goto out;
+
+		if (WARN_ON_ONCE(!bio_next_segment(bio, &iter_all))) {
+			res = -EIO;
+			goto out_free_bio;
+		}
+		/* Extract the length of the metadata block */
+		data = page_address(bvec->bv_page) + bvec->bv_offset;
+		length = data[offset];
+		if (offset <= bvec->bv_len - 1) {
+			length |= data[offset + 1] << 8;
+		} else {
+			if (WARN_ON_ONCE(!bio_next_segment(bio, &iter_all))) {
+				res = -EIO;
+				goto out_free_bio;
+			}
+			data = page_address(bvec->bv_page) + bvec->bv_offset;
+			length |= data[0] << 8;
+		}
+		bio_free_pages(bio);
+		bio_put(bio);
 
-		bytes = msblk->devblksize - offset;
 		compressed = SQUASHFS_COMPRESSED(length);
 		length = SQUASHFS_COMPRESSED_SIZE(length);
-		if (next_index)
-			*next_index = index + length + 2;
+		index += 2;
 
 		TRACE("Block @ 0x%llx, %scompressed size %d\n", index,
-				compressed ? "" : "un", length);
-
-		if (length < 0 || length > output->length ||
-					(index + length) > msblk->bytes_used)
-			goto block_release;
-
-		for (; bytes < length; b++) {
-			bh[b] = sb_getblk(sb, ++cur_index);
-			if (bh[b] == NULL)
-				goto block_release;
-			bytes += msblk->devblksize;
-		}
-		ll_rw_block(REQ_OP_READ, 0, b - 1, bh + 1);
+		      compressed ? "" : "un", length);
 	}
+	if (next_index)
+		*next_index = index + length;
 
-	for (i = 0; i < b; i++) {
-		wait_on_buffer(bh[i]);
-		if (!buffer_uptodate(bh[i]))
-			goto block_release;
-	}
+	res = squashfs_bio_read(sb, index, length, &bio, &offset);
+	if (res)
+		goto out;
 
 	if (compressed) {
-		if (!msblk->stream)
-			goto read_failure;
-		length = squashfs_decompress(msblk, bh, b, offset, length,
-			output);
-		if (length < 0)
-			goto read_failure;
-	} else {
-		/*
-		 * Block is uncompressed.
-		 */
-		int in, pg_offset = 0;
-		void *data = squashfs_first_page(output);
-
-		for (bytes = length; k < b; k++) {
-			in = min(bytes, msblk->devblksize - offset);
-			bytes -= in;
-			while (in) {
-				if (pg_offset == PAGE_SIZE) {
-					data = squashfs_next_page(output);
-					pg_offset = 0;
-				}
-				avail = min_t(int, in, PAGE_SIZE -
-						pg_offset);
-				memcpy(data + pg_offset, bh[k]->b_data + offset,
-						avail);
-				in -= avail;
-				pg_offset += avail;
-				offset += avail;
-			}
-			offset = 0;
-			put_bh(bh[k]);
+		if (!msblk->stream) {
+			res = -EIO;
+			goto out_free_bio;
 		}
-		squashfs_finish_page(output);
+		res = squashfs_decompress(msblk, bio, offset, length, output);
+	} else {
+		res = copy_bio_to_actor(bio, output, offset, length);
 	}
 
-	kfree(bh);
-	return length;
-
-block_release:
-	for (; k < b; k++)
-		put_bh(bh[k]);
+out_free_bio:
+	bio_free_pages(bio);
+	bio_put(bio);
+out:
+	if (res < 0)
+		ERROR("Failed to read block 0x%llx: %d\n", index, res);
 
-read_failure:
-	ERROR("squashfs_read_data failed to read block 0x%llx\n",
-					(unsigned long long) index);
-	kfree(bh);
-	return -EIO;
+	return res;
 }
diff --git a/fs/squashfs/decompressor.h b/fs/squashfs/decompressor.h
index ec8617523e56..1b9ccfd0aa51 100644
--- a/fs/squashfs/decompressor.h
+++ b/fs/squashfs/decompressor.h
@@ -10,13 +10,14 @@
  * decompressor.h
  */
 
+#include <linux/bio.h>
+
 struct squashfs_decompressor {
 	void	*(*init)(struct squashfs_sb_info *, void *);
 	void	*(*comp_opts)(struct squashfs_sb_info *, void *, int);
 	void	(*free)(void *);
 	int	(*decompress)(struct squashfs_sb_info *, void *,
-		struct buffer_head **, int, int, int,
-		struct squashfs_page_actor *);
+		struct bio *, int, int, struct squashfs_page_actor *);
 	int	id;
 	char	*name;
 	int	supported;
diff --git a/fs/squashfs/decompressor_multi.c b/fs/squashfs/decompressor_multi.c
index c181dee235bb..db9f12a3ea05 100644
--- a/fs/squashfs/decompressor_multi.c
+++ b/fs/squashfs/decompressor_multi.c
@@ -6,7 +6,7 @@
 #include <linux/types.h>
 #include <linux/mutex.h>
 #include <linux/slab.h>
-#include <linux/buffer_head.h>
+#include <linux/bio.h>
 #include <linux/sched.h>
 #include <linux/wait.h>
 #include <linux/cpumask.h>
@@ -180,14 +180,15 @@ wait:
 }
 
 
-int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh,
-	int b, int offset, int length, struct squashfs_page_actor *output)
+int squashfs_decompress(struct squashfs_sb_info *msblk, struct bio *bio,
+			int offset, int length,
+			struct squashfs_page_actor *output)
 {
 	int res;
 	struct squashfs_stream *stream = msblk->stream;
 	struct decomp_stream *decomp_stream = get_decomp_stream(msblk, stream);
 	res = msblk->decompressor->decompress(msblk, decomp_stream->stream,
-		bh, b, offset, length, output);
+		bio, offset, length, output);
 	put_decomp_stream(decomp_stream, stream);
 	if (res < 0)
 		ERROR("%s decompression failed, data probably corrupt\n",
diff --git a/fs/squashfs/decompressor_multi_percpu.c b/fs/squashfs/decompressor_multi_percpu.c
index 2a2a2d106440..b881b9283b7f 100644
--- a/fs/squashfs/decompressor_multi_percpu.c
+++ b/fs/squashfs/decompressor_multi_percpu.c
@@ -8,6 +8,7 @@
 #include <linux/slab.h>
 #include <linux/percpu.h>
 #include <linux/buffer_head.h>
+#include <linux/local_lock.h>
 
 #include "squashfs_fs.h"
 #include "squashfs_fs_sb.h"
@@ -20,7 +21,8 @@
  */
 
 struct squashfs_stream {
-	void		*stream;
+	void			*stream;
+	local_lock_t	lock;
 };
 
 void *squashfs_decompressor_create(struct squashfs_sb_info *msblk,
@@ -41,6 +43,7 @@ void *squashfs_decompressor_create(struct squashfs_sb_info *msblk,
 			err = PTR_ERR(stream->stream);
 			goto out;
 		}
+		local_lock_init(&stream->lock);
 	}
 
 	kfree(comp_opts);
@@ -72,15 +75,19 @@ void squashfs_decompressor_destroy(struct squashfs_sb_info *msblk)
 	}
 }
 
-int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh,
-	int b, int offset, int length, struct squashfs_page_actor *output)
+int squashfs_decompress(struct squashfs_sb_info *msblk, struct bio *bio,
+	int offset, int length, struct squashfs_page_actor *output)
 {
-	struct squashfs_stream __percpu *percpu =
-			(struct squashfs_stream __percpu *) msblk->stream;
-	struct squashfs_stream *stream = get_cpu_ptr(percpu);
-	int res = msblk->decompressor->decompress(msblk, stream->stream, bh, b,
-		offset, length, output);
-	put_cpu_ptr(stream);
+	struct squashfs_stream *stream;
+	int res;
+
+	local_lock(&msblk->stream->lock);
+	stream = this_cpu_ptr(msblk->stream);
+
+	res = msblk->decompressor->decompress(msblk, stream->stream, bio,
+					      offset, length, output);
+
+	local_unlock(&msblk->stream->lock);
 
 	if (res < 0)
 		ERROR("%s decompression failed, data probably corrupt\n",
diff --git a/fs/squashfs/decompressor_single.c b/fs/squashfs/decompressor_single.c
index 550c3e592032..4eb3d083d45e 100644
--- a/fs/squashfs/decompressor_single.c
+++ b/fs/squashfs/decompressor_single.c
@@ -7,7 +7,7 @@
 #include <linux/types.h>
 #include <linux/mutex.h>
 #include <linux/slab.h>
-#include <linux/buffer_head.h>
+#include <linux/bio.h>
 
 #include "squashfs_fs.h"
 #include "squashfs_fs_sb.h"
@@ -59,14 +59,15 @@ void squashfs_decompressor_destroy(struct squashfs_sb_info *msblk)
 	}
 }
 
-int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh,
-	int b, int offset, int length, struct squashfs_page_actor *output)
+int squashfs_decompress(struct squashfs_sb_info *msblk, struct bio *bio,
+			int offset, int length,
+			struct squashfs_page_actor *output)
 {
 	int res;
 	struct squashfs_stream *stream = msblk->stream;
 
 	mutex_lock(&stream->mutex);
-	res = msblk->decompressor->decompress(msblk, stream->stream, bh, b,
+	res = msblk->decompressor->decompress(msblk, stream->stream, bio,
 		offset, length, output);
 	mutex_unlock(&stream->mutex);
 
diff --git a/fs/squashfs/lz4_wrapper.c b/fs/squashfs/lz4_wrapper.c
index c4e47e0588c7..233d5582fbee 100644
--- a/fs/squashfs/lz4_wrapper.c
+++ b/fs/squashfs/lz4_wrapper.c
@@ -4,7 +4,7 @@
  * Phillip Lougher <phillip@squashfs.org.uk>
  */
 
-#include <linux/buffer_head.h>
+#include <linux/bio.h>
 #include <linux/mutex.h>
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
@@ -89,20 +89,23 @@ static void lz4_free(void *strm)
 
 
 static int lz4_uncompress(struct squashfs_sb_info *msblk, void *strm,
-	struct buffer_head **bh, int b, int offset, int length,
+	struct bio *bio, int offset, int length,
 	struct squashfs_page_actor *output)
 {
+	struct bvec_iter_all iter_all = {};
+	struct bio_vec *bvec = bvec_init_iter_all(&iter_all);
 	struct squashfs_lz4 *stream = strm;
 	void *buff = stream->input, *data;
-	int avail, i, bytes = length, res;
+	int bytes = length, res;
 
-	for (i = 0; i < b; i++) {
-		avail = min(bytes, msblk->devblksize - offset);
-		memcpy(buff, bh[i]->b_data + offset, avail);
+	while (bio_next_segment(bio, &iter_all)) {
+		int avail = min(bytes, ((int)bvec->bv_len) - offset);
+
+		data = page_address(bvec->bv_page) + bvec->bv_offset;
+		memcpy(buff, data + offset, avail);
 		buff += avail;
 		bytes -= avail;
 		offset = 0;
-		put_bh(bh[i]);
 	}
 
 	res = LZ4_decompress_safe(stream->input, stream->output,
diff --git a/fs/squashfs/lzo_wrapper.c b/fs/squashfs/lzo_wrapper.c
index aa3c3dafc33d..97bb7d92ddcd 100644
--- a/fs/squashfs/lzo_wrapper.c
+++ b/fs/squashfs/lzo_wrapper.c
@@ -9,7 +9,7 @@
  */
 
 #include <linux/mutex.h>
-#include <linux/buffer_head.h>
+#include <linux/bio.h>
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
 #include <linux/lzo.h>
@@ -63,21 +63,24 @@ static void lzo_free(void *strm)
 
 
 static int lzo_uncompress(struct squashfs_sb_info *msblk, void *strm,
-	struct buffer_head **bh, int b, int offset, int length,
+	struct bio *bio, int offset, int length,
 	struct squashfs_page_actor *output)
 {
+	struct bvec_iter_all iter_all = {};
+	struct bio_vec *bvec = bvec_init_iter_all(&iter_all);
 	struct squashfs_lzo *stream = strm;
 	void *buff = stream->input, *data;
-	int avail, i, bytes = length, res;
+	int bytes = length, res;
 	size_t out_len = output->length;
 
-	for (i = 0; i < b; i++) {
-		avail = min(bytes, msblk->devblksize - offset);
-		memcpy(buff, bh[i]->b_data + offset, avail);
+	while (bio_next_segment(bio, &iter_all)) {
+		int avail = min(bytes, ((int)bvec->bv_len) - offset);
+
+		data = page_address(bvec->bv_page) + bvec->bv_offset;
+		memcpy(buff, data + offset, avail);
 		buff += avail;
 		bytes -= avail;
 		offset = 0;
-		put_bh(bh[i]);
 	}
 
 	res = lzo1x_decompress_safe(stream->input, (size_t)length,
diff --git a/fs/squashfs/squashfs.h b/fs/squashfs/squashfs.h
index 2797763ed046..9783e01c8100 100644
--- a/fs/squashfs/squashfs.h
+++ b/fs/squashfs/squashfs.h
@@ -40,8 +40,8 @@ extern void *squashfs_decompressor_setup(struct super_block *, unsigned short);
 /* decompressor_xxx.c */
 extern void *squashfs_decompressor_create(struct squashfs_sb_info *, void *);
 extern void squashfs_decompressor_destroy(struct squashfs_sb_info *);
-extern int squashfs_decompress(struct squashfs_sb_info *, struct buffer_head **,
-	int, int, int, struct squashfs_page_actor *);
+extern int squashfs_decompress(struct squashfs_sb_info *, struct bio *,
+				int, int, struct squashfs_page_actor *);
 extern int squashfs_max_decompressors(void);
 
 /* export.c */
diff --git a/fs/squashfs/xz_wrapper.c b/fs/squashfs/xz_wrapper.c
index 4b2f2051a6dc..e80419aed862 100644
--- a/fs/squashfs/xz_wrapper.c
+++ b/fs/squashfs/xz_wrapper.c
@@ -10,7 +10,7 @@
 
 
 #include <linux/mutex.h>
-#include <linux/buffer_head.h>
+#include <linux/bio.h>
 #include <linux/slab.h>
 #include <linux/xz.h>
 #include <linux/bitops.h>
@@ -117,11 +117,12 @@ static void squashfs_xz_free(void *strm)
 
 
 static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void *strm,
-	struct buffer_head **bh, int b, int offset, int length,
+	struct bio *bio, int offset, int length,
 	struct squashfs_page_actor *output)
 {
-	enum xz_ret xz_err;
-	int avail, total = 0, k = 0;
+	struct bvec_iter_all iter_all = {};
+	struct bio_vec *bvec = bvec_init_iter_all(&iter_all);
+	int total = 0, error = 0;
 	struct squashfs_xz *stream = strm;
 
 	xz_dec_reset(stream->state);
@@ -131,11 +132,23 @@ static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void *strm,
 	stream->buf.out_size = PAGE_SIZE;
 	stream->buf.out = squashfs_first_page(output);
 
-	do {
-		if (stream->buf.in_pos == stream->buf.in_size && k < b) {
-			avail = min(length, msblk->devblksize - offset);
+	for (;;) {
+		enum xz_ret xz_err;
+
+		if (stream->buf.in_pos == stream->buf.in_size) {
+			const void *data;
+			int avail;
+
+			if (!bio_next_segment(bio, &iter_all)) {
+				/* XZ_STREAM_END must be reached. */
+				error = -EIO;
+				break;
+			}
+
+			avail = min(length, ((int)bvec->bv_len) - offset);
+			data = page_address(bvec->bv_page) + bvec->bv_offset;
 			length -= avail;
-			stream->buf.in = bh[k]->b_data + offset;
+			stream->buf.in = data + offset;
 			stream->buf.in_size = avail;
 			stream->buf.in_pos = 0;
 			offset = 0;
@@ -150,23 +163,17 @@ static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void *strm,
 		}
 
 		xz_err = xz_dec_run(stream->state, &stream->buf);
-
-		if (stream->buf.in_pos == stream->buf.in_size && k < b)
-			put_bh(bh[k++]);
-	} while (xz_err == XZ_OK);
+		if (xz_err == XZ_STREAM_END)
+			break;
+		if (xz_err != XZ_OK) {
+			error = -EIO;
+			break;
+		}
+	}
 
 	squashfs_finish_page(output);
 
-	if (xz_err != XZ_STREAM_END || k < b)
-		goto out;
-
-	return total + stream->buf.out_pos;
-
-out:
-	for (; k < b; k++)
-		put_bh(bh[k]);
-
-	return -EIO;
+	return error ? error : total + stream->buf.out_pos;
 }
 
 const struct squashfs_decompressor squashfs_xz_comp_ops = {
diff --git a/fs/squashfs/zlib_wrapper.c b/fs/squashfs/zlib_wrapper.c
index f2226afa1625..bcb881ec47f2 100644
--- a/fs/squashfs/zlib_wrapper.c
+++ b/fs/squashfs/zlib_wrapper.c
@@ -10,7 +10,7 @@
 
 
 #include <linux/mutex.h>
-#include <linux/buffer_head.h>
+#include <linux/bio.h>
 #include <linux/slab.h>
 #include <linux/zlib.h>
 #include <linux/vmalloc.h>
@@ -50,21 +50,35 @@ static void zlib_free(void *strm)
 
 
 static int zlib_uncompress(struct squashfs_sb_info *msblk, void *strm,
-	struct buffer_head **bh, int b, int offset, int length,
+	struct bio *bio, int offset, int length,
 	struct squashfs_page_actor *output)
 {
-	int zlib_err, zlib_init = 0, k = 0;
+	struct bvec_iter_all iter_all = {};
+	struct bio_vec *bvec = bvec_init_iter_all(&iter_all);
+	int zlib_init = 0, error = 0;
 	z_stream *stream = strm;
 
 	stream->avail_out = PAGE_SIZE;
 	stream->next_out = squashfs_first_page(output);
 	stream->avail_in = 0;
 
-	do {
-		if (stream->avail_in == 0 && k < b) {
-			int avail = min(length, msblk->devblksize - offset);
+	for (;;) {
+		int zlib_err;
+
+		if (stream->avail_in == 0) {
+			const void *data;
+			int avail;
+
+			if (!bio_next_segment(bio, &iter_all)) {
+				/* Z_STREAM_END must be reached. */
+				error = -EIO;
+				break;
+			}
+
+			avail = min(length, ((int)bvec->bv_len) - offset);
+			data = page_address(bvec->bv_page) + bvec->bv_offset;
 			length -= avail;
-			stream->next_in = bh[k]->b_data + offset;
+			stream->next_in = data + offset;
 			stream->avail_in = avail;
 			offset = 0;
 		}
@@ -78,37 +92,28 @@ static int zlib_uncompress(struct squashfs_sb_info *msblk, void *strm,
 		if (!zlib_init) {
 			zlib_err = zlib_inflateInit(stream);
 			if (zlib_err != Z_OK) {
-				squashfs_finish_page(output);
-				goto out;
+				error = -EIO;
+				break;
 			}
 			zlib_init = 1;
 		}
 
 		zlib_err = zlib_inflate(stream, Z_SYNC_FLUSH);
-
-		if (stream->avail_in == 0 && k < b)
-			put_bh(bh[k++]);
-	} while (zlib_err == Z_OK);
+		if (zlib_err == Z_STREAM_END)
+			break;
+		if (zlib_err != Z_OK) {
+			error = -EIO;
+			break;
+		}
+	}
 
 	squashfs_finish_page(output);
 
-	if (zlib_err != Z_STREAM_END)
-		goto out;
-
-	zlib_err = zlib_inflateEnd(stream);
-	if (zlib_err != Z_OK)
-		goto out;
-
-	if (k < b)
-		goto out;
-
-	return stream->total_out;
-
-out:
-	for (; k < b; k++)
-		put_bh(bh[k]);
+	if (!error)
+		if (zlib_inflateEnd(stream) != Z_OK)
+			error = -EIO;
 
-	return -EIO;
+	return error ? error : stream->total_out;
 }
 
 const struct squashfs_decompressor squashfs_zlib_comp_ops = {
diff --git a/fs/squashfs/zstd_wrapper.c b/fs/squashfs/zstd_wrapper.c
index b448c2a1d0ed..b7cb1faa652d 100644
--- a/fs/squashfs/zstd_wrapper.c
+++ b/fs/squashfs/zstd_wrapper.c
@@ -9,7 +9,7 @@
  */
 
 #include <linux/mutex.h>
-#include <linux/buffer_head.h>
+#include <linux/bio.h>
 #include <linux/slab.h>
 #include <linux/zstd.h>
 #include <linux/vmalloc.h>
@@ -59,33 +59,44 @@ static void zstd_free(void *strm)
 
 
 static int zstd_uncompress(struct squashfs_sb_info *msblk, void *strm,
-	struct buffer_head **bh, int b, int offset, int length,
+	struct bio *bio, int offset, int length,
 	struct squashfs_page_actor *output)
 {
 	struct workspace *wksp = strm;
 	ZSTD_DStream *stream;
 	size_t total_out = 0;
-	size_t zstd_err;
-	int k = 0;
+	int error = 0;
 	ZSTD_inBuffer in_buf = { NULL, 0, 0 };
 	ZSTD_outBuffer out_buf = { NULL, 0, 0 };
+	struct bvec_iter_all iter_all = {};
+	struct bio_vec *bvec = bvec_init_iter_all(&iter_all);
 
 	stream = ZSTD_initDStream(wksp->window_size, wksp->mem, wksp->mem_size);
 
 	if (!stream) {
 		ERROR("Failed to initialize zstd decompressor\n");
-		goto out;
+		return -EIO;
 	}
 
 	out_buf.size = PAGE_SIZE;
 	out_buf.dst = squashfs_first_page(output);
 
-	do {
-		if (in_buf.pos == in_buf.size && k < b) {
-			int avail = min(length, msblk->devblksize - offset);
+	for (;;) {
+		size_t zstd_err;
 
+		if (in_buf.pos == in_buf.size) {
+			const void *data;
+			int avail;
+
+			if (!bio_next_segment(bio, &iter_all)) {
+				error = -EIO;
+				break;
+			}
+
+			avail = min(length, ((int)bvec->bv_len) - offset);
+			data = page_address(bvec->bv_page) + bvec->bv_offset;
 			length -= avail;
-			in_buf.src = bh[k]->b_data + offset;
+			in_buf.src = data + offset;
 			in_buf.size = avail;
 			in_buf.pos = 0;
 			offset = 0;
@@ -97,8 +108,8 @@ static int zstd_uncompress(struct squashfs_sb_info *msblk, void *strm,
 				/* Shouldn't run out of pages
 				 * before stream is done.
 				 */
-				squashfs_finish_page(output);
-				goto out;
+				error = -EIO;
+				break;
 			}
 			out_buf.pos = 0;
 			out_buf.size = PAGE_SIZE;
@@ -107,29 +118,20 @@ static int zstd_uncompress(struct squashfs_sb_info *msblk, void *strm,
 		total_out -= out_buf.pos;
 		zstd_err = ZSTD_decompressStream(stream, &out_buf, &in_buf);
 		total_out += out_buf.pos; /* add the additional data produced */
-
-		if (in_buf.pos == in_buf.size && k < b)
-			put_bh(bh[k++]);
-	} while (zstd_err != 0 && !ZSTD_isError(zstd_err));
-
-	squashfs_finish_page(output);
-
-	if (ZSTD_isError(zstd_err)) {
-		ERROR("zstd decompression error: %d\n",
-				(int)ZSTD_getErrorCode(zstd_err));
-		goto out;
+		if (zstd_err == 0)
+			break;
+
+		if (ZSTD_isError(zstd_err)) {
+			ERROR("zstd decompression error: %d\n",
+					(int)ZSTD_getErrorCode(zstd_err));
+			error = -EIO;
+			break;
+		}
 	}
 
-	if (k < b)
-		goto out;
-
-	return (int)total_out;
-
-out:
-	for (; k < b; k++)
-		put_bh(bh[k]);
+	squashfs_finish_page(output);
 
-	return -EIO;
+	return error ? error : total_out;
 }
 
 const struct squashfs_decompressor squashfs_zstd_comp_ops = {
diff --git a/fs/stack.c b/fs/stack.c
index 4ef2c056269d..c9830924eb12 100644
--- a/fs/stack.c
+++ b/fs/stack.c
@@ -23,7 +23,7 @@ void fsstack_copy_inode_size(struct inode *dst, struct inode *src)
 
 	/*
 	 * But on 32-bit, we ought to make an effort to keep the two halves of
-	 * i_blocks in sync despite SMP or PREEMPT - though stat's
+	 * i_blocks in sync despite SMP or PREEMPTION - though stat's
 	 * generic_fillattr() doesn't bother, and we won't be applying quotas
 	 * (where i_blocks does become important) at the upper level.
 	 *
@@ -38,14 +38,14 @@ void fsstack_copy_inode_size(struct inode *dst, struct inode *src)
 		spin_unlock(&src->i_lock);
 
 	/*
-	 * If CONFIG_SMP or CONFIG_PREEMPT on 32-bit, it's vital for
+	 * If CONFIG_SMP or CONFIG_PREEMPTION on 32-bit, it's vital for
 	 * fsstack_copy_inode_size() to hold some lock around
 	 * i_size_write(), otherwise i_size_read() may spin forever (see
 	 * include/linux/fs.h).  We don't necessarily hold i_mutex when this
 	 * is called, so take i_lock for that case.
 	 *
 	 * And if on 32-bit, continue our effort to keep the two halves of
-	 * i_blocks in sync despite SMP or PREEMPT: use i_lock  for that case
+	 * i_blocks in sync despite SMP or PREEMPTION: use i_lock for that case
 	 * too, and do both at once by combining the tests.
 	 *
 	 * There is none of this locking overhead in the 64-bit case.
diff --git a/fs/stat.c b/fs/stat.c
index c38e4c2e1221..b86a5c338133 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -21,6 +21,9 @@
 #include <linux/uaccess.h>
 #include <asm/unistd.h>
 
+#include "internal.h"
+#include "mount.h"
+
 /**
  * generic_fillattr - Fill in the basic attributes from the inode struct
  * @inode: Inode to use as the source
@@ -68,11 +71,11 @@ int vfs_getattr_nosec(const struct path *path, struct kstat *stat,
 
 	memset(stat, 0, sizeof(*stat));
 	stat->result_mask |= STATX_BASIC_STATS;
-	request_mask &= STATX_ALL;
 	query_flags &= KSTAT_QUERY_FLAGS;
 
 	/* allow the fs to override these if it really wants to */
-	if (IS_NOATIME(inode))
+	/* SB_NOATIME means filesystem supplies dummy atime value */
+	if (inode->i_sb->s_flags & SB_NOATIME)
 		stat->result_mask &= ~STATX_ATIME;
 	if (IS_AUTOMOUNT(inode))
 		stat->attributes |= STATX_ATTR_AUTOMOUNT;
@@ -150,6 +153,24 @@ int vfs_statx_fd(unsigned int fd, struct kstat *stat,
 }
 EXPORT_SYMBOL(vfs_statx_fd);
 
+static inline unsigned vfs_stat_set_lookup_flags(unsigned *lookup_flags,
+						 int flags)
+{
+	if ((flags & ~(AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT |
+		       AT_EMPTY_PATH | KSTAT_QUERY_FLAGS)) != 0)
+		return -EINVAL;
+
+	*lookup_flags = LOOKUP_FOLLOW | LOOKUP_AUTOMOUNT;
+	if (flags & AT_SYMLINK_NOFOLLOW)
+		*lookup_flags &= ~LOOKUP_FOLLOW;
+	if (flags & AT_NO_AUTOMOUNT)
+		*lookup_flags &= ~LOOKUP_AUTOMOUNT;
+	if (flags & AT_EMPTY_PATH)
+		*lookup_flags |= LOOKUP_EMPTY;
+
+	return 0;
+}
+
 /**
  * vfs_statx - Get basic and extra attributes by filename
  * @dfd: A file descriptor representing the base dir for a relative filename
@@ -170,25 +191,21 @@ int vfs_statx(int dfd, const char __user *filename, int flags,
 {
 	struct path path;
 	int error = -EINVAL;
-	unsigned int lookup_flags = LOOKUP_FOLLOW | LOOKUP_AUTOMOUNT;
+	unsigned lookup_flags;
 
-	if ((flags & ~(AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT |
-		       AT_EMPTY_PATH | KSTAT_QUERY_FLAGS)) != 0)
+	if (vfs_stat_set_lookup_flags(&lookup_flags, flags))
 		return -EINVAL;
-
-	if (flags & AT_SYMLINK_NOFOLLOW)
-		lookup_flags &= ~LOOKUP_FOLLOW;
-	if (flags & AT_NO_AUTOMOUNT)
-		lookup_flags &= ~LOOKUP_AUTOMOUNT;
-	if (flags & AT_EMPTY_PATH)
-		lookup_flags |= LOOKUP_EMPTY;
-
 retry:
 	error = user_path_at(dfd, filename, lookup_flags, &path);
 	if (error)
 		goto out;
 
 	error = vfs_getattr(&path, stat, request_mask, flags);
+	stat->mnt_id = real_mount(path.mnt)->mnt_id;
+	stat->result_mask |= STATX_MNT_ID;
+	if (path.mnt->mnt_root == path.dentry)
+		stat->attributes |= STATX_ATTR_MOUNT_ROOT;
+	stat->attributes_mask |= STATX_ATTR_MOUNT_ROOT;
 	path_put(&path);
 	if (retry_estale(error, lookup_flags)) {
 		lookup_flags |= LOOKUP_REVAL;
@@ -553,10 +570,29 @@ cp_statx(const struct kstat *stat, struct statx __user *buffer)
 	tmp.stx_rdev_minor = MINOR(stat->rdev);
 	tmp.stx_dev_major = MAJOR(stat->dev);
 	tmp.stx_dev_minor = MINOR(stat->dev);
+	tmp.stx_mnt_id = stat->mnt_id;
 
 	return copy_to_user(buffer, &tmp, sizeof(tmp)) ? -EFAULT : 0;
 }
 
+int do_statx(int dfd, const char __user *filename, unsigned flags,
+	     unsigned int mask, struct statx __user *buffer)
+{
+	struct kstat stat;
+	int error;
+
+	if (mask & STATX__RESERVED)
+		return -EINVAL;
+	if ((flags & AT_STATX_SYNC_TYPE) == AT_STATX_SYNC_TYPE)
+		return -EINVAL;
+
+	error = vfs_statx(dfd, filename, flags, &stat, mask);
+	if (error)
+		return error;
+
+	return cp_statx(&stat, buffer);
+}
+
 /**
  * sys_statx - System call to get enhanced stats
  * @dfd: Base directory to pathwalk from *or* fd to stat.
@@ -573,19 +609,7 @@ SYSCALL_DEFINE5(statx,
 		unsigned int, mask,
 		struct statx __user *, buffer)
 {
-	struct kstat stat;
-	int error;
-
-	if (mask & STATX__RESERVED)
-		return -EINVAL;
-	if ((flags & AT_STATX_SYNC_TYPE) == AT_STATX_SYNC_TYPE)
-		return -EINVAL;
-
-	error = vfs_statx(dfd, filename, flags, &stat, mask);
-	if (error)
-		return error;
-
-	return cp_statx(&stat, buffer);
+	return do_statx(dfd, filename, flags, mask, buffer);
 }
 
 #ifdef CONFIG_COMPAT
diff --git a/fs/super.c b/fs/super.c
index cfadab2cbf35..bf3b7685b52a 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -448,10 +448,12 @@ void generic_shutdown_super(struct super_block *sb)
 		sync_filesystem(sb);
 		sb->s_flags &= ~SB_ACTIVE;
 
-		fsnotify_sb_delete(sb);
 		cgroup_writeback_umount();
 
+		/* evict all inodes with zero refcount */
 		evict_inodes(sb);
+		/* only nonzero refcount inodes can have marks */
+		fsnotify_sb_delete(sb);
 
 		if (sb->s_dio_done_wq) {
 			destroy_workqueue(sb->s_dio_done_wq);
@@ -1300,8 +1302,8 @@ int get_tree_bdev(struct fs_context *fc,
 	mutex_lock(&bdev->bd_fsfreeze_mutex);
 	if (bdev->bd_fsfreeze_count > 0) {
 		mutex_unlock(&bdev->bd_fsfreeze_mutex);
-		blkdev_put(bdev, mode);
 		warnf(fc, "%pg: Can't mount, blockdev is frozen", bdev);
+		blkdev_put(bdev, mode);
 		return -EBUSY;
 	}
 
@@ -1596,12 +1598,10 @@ int super_setup_bdi_name(struct super_block *sb, char *fmt, ...)
 	int err;
 	va_list args;
 
-	bdi = bdi_alloc(GFP_KERNEL);
+	bdi = bdi_alloc(NUMA_NO_NODE);
 	if (!bdi)
 		return -ENOMEM;
 
-	bdi->name = sb->s_type->name;
-
 	va_start(args, fmt);
 	err = bdi_register_va(bdi, fmt, args);
 	va_end(args);
diff --git a/fs/sync.c b/fs/sync.c
index 4d1ff010bc5a..c6f6f5be5682 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -161,7 +161,7 @@ SYSCALL_DEFINE1(syncfs, int, fd)
 {
 	struct fd f = fdget(fd);
 	struct super_block *sb;
-	int ret;
+	int ret, ret2;
 
 	if (!f.file)
 		return -EBADF;
@@ -171,8 +171,10 @@ SYSCALL_DEFINE1(syncfs, int, fd)
 	ret = sync_filesystem(sb);
 	up_read(&sb->s_umount);
 
+	ret2 = errseq_check_and_advance(&sb->s_wb_err, &f.file->f_sb_err);
+
 	fdput(f);
-	return ret;
+	return ret ? ret : ret2;
 }
 
 /**
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index aa85f2874a9f..59dffd5ca517 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -6,7 +6,7 @@
  * Copyright (c) 2007 SUSE Linux Products GmbH
  * Copyright (c) 2007 Tejun Heo <teheo@suse.de>
  *
- * Please see Documentation/filesystems/sysfs.txt for more information.
+ * Please see Documentation/filesystems/sysfs.rst for more information.
  */
 
 #define pr_fmt(fmt)	"sysfs: " fmt
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 130fc6fbcc03..f275fcda62fb 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -6,7 +6,7 @@
  * Copyright (c) 2007 SUSE Linux Products GmbH
  * Copyright (c) 2007 Tejun Heo <teheo@suse.de>
  *
- * Please see Documentation/filesystems/sysfs.txt for more information.
+ * Please see Documentation/filesystems/sysfs.rst for more information.
  */
 
 #include <linux/module.h>
@@ -558,3 +558,151 @@ void sysfs_remove_bin_file(struct kobject *kobj,
 	kernfs_remove_by_name(kobj->sd, attr->attr.name);
 }
 EXPORT_SYMBOL_GPL(sysfs_remove_bin_file);
+
+static int internal_change_owner(struct kernfs_node *kn, kuid_t kuid,
+				 kgid_t kgid)
+{
+	struct iattr newattrs = {
+		.ia_valid = ATTR_UID | ATTR_GID,
+		.ia_uid = kuid,
+		.ia_gid = kgid,
+	};
+	return kernfs_setattr(kn, &newattrs);
+}
+
+/**
+ *	sysfs_link_change_owner - change owner of a sysfs file.
+ *	@kobj:	object of the kernfs_node the symlink is located in.
+ *	@targ:	object of the kernfs_node the symlink points to.
+ *	@name:	name of the link.
+ *	@kuid:	new owner's kuid
+ *	@kgid:	new owner's kgid
+ *
+ * This function looks up the sysfs symlink entry @name under @kobj and changes
+ * the ownership to @kuid/@kgid. The symlink is looked up in the namespace of
+ * @targ.
+ *
+ * Returns 0 on success or error code on failure.
+ */
+int sysfs_link_change_owner(struct kobject *kobj, struct kobject *targ,
+			    const char *name, kuid_t kuid, kgid_t kgid)
+{
+	struct kernfs_node *kn = NULL;
+	int error;
+
+	if (!name || !kobj->state_in_sysfs || !targ->state_in_sysfs)
+		return -EINVAL;
+
+	error = -ENOENT;
+	kn = kernfs_find_and_get_ns(kobj->sd, name, targ->sd->ns);
+	if (!kn)
+		goto out;
+
+	error = -EINVAL;
+	if (kernfs_type(kn) != KERNFS_LINK)
+		goto out;
+	if (kn->symlink.target_kn->priv != targ)
+		goto out;
+
+	error = internal_change_owner(kn, kuid, kgid);
+
+out:
+	kernfs_put(kn);
+	return error;
+}
+
+/**
+ *	sysfs_file_change_owner - change owner of a sysfs file.
+ *	@kobj:	object.
+ *	@name:	name of the file to change.
+ *	@kuid:	new owner's kuid
+ *	@kgid:	new owner's kgid
+ *
+ * This function looks up the sysfs entry @name under @kobj and changes the
+ * ownership to @kuid/@kgid.
+ *
+ * Returns 0 on success or error code on failure.
+ */
+int sysfs_file_change_owner(struct kobject *kobj, const char *name, kuid_t kuid,
+			    kgid_t kgid)
+{
+	struct kernfs_node *kn;
+	int error;
+
+	if (!name)
+		return -EINVAL;
+
+	if (!kobj->state_in_sysfs)
+		return -EINVAL;
+
+	kn = kernfs_find_and_get(kobj->sd, name);
+	if (!kn)
+		return -ENOENT;
+
+	error = internal_change_owner(kn, kuid, kgid);
+
+	kernfs_put(kn);
+
+	return error;
+}
+EXPORT_SYMBOL_GPL(sysfs_file_change_owner);
+
+/**
+ *	sysfs_change_owner - change owner of the given object.
+ *	@kobj:	object.
+ *	@kuid:	new owner's kuid
+ *	@kgid:	new owner's kgid
+ *
+ * Change the owner of the default directory, files, groups, and attributes of
+ * @kobj to @kuid/@kgid. Note that sysfs_change_owner mirrors how the sysfs
+ * entries for a kobject are added by driver core. In summary,
+ * sysfs_change_owner() takes care of the default directory entry for @kobj,
+ * the default attributes associated with the ktype of @kobj and the default
+ * attributes associated with the ktype of @kobj.
+ * Additional properties not added by driver core have to be changed by the
+ * driver or subsystem which created them. This is similar to how
+ * driver/subsystem specific entries are removed.
+ *
+ * Returns 0 on success or error code on failure.
+ */
+int sysfs_change_owner(struct kobject *kobj, kuid_t kuid, kgid_t kgid)
+{
+	int error;
+	const struct kobj_type *ktype;
+
+	if (!kobj->state_in_sysfs)
+		return -EINVAL;
+
+	/* Change the owner of the kobject itself. */
+	error = internal_change_owner(kobj->sd, kuid, kgid);
+	if (error)
+		return error;
+
+	ktype = get_ktype(kobj);
+	if (ktype) {
+		struct attribute **kattr;
+
+		/*
+		 * Change owner of the default attributes associated with the
+		 * ktype of @kobj.
+		 */
+		for (kattr = ktype->default_attrs; kattr && *kattr; kattr++) {
+			error = sysfs_file_change_owner(kobj, (*kattr)->name,
+							kuid, kgid);
+			if (error)
+				return error;
+		}
+
+		/*
+		 * Change owner of the default groups associated with the
+		 * ktype of @kobj.
+		 */
+		error = sysfs_groups_change_owner(kobj, ktype->default_groups,
+						  kuid, kgid);
+		if (error)
+			return error;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(sysfs_change_owner);
diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c
index d41c21fef138..64e6a6698935 100644
--- a/fs/sysfs/group.c
+++ b/fs/sysfs/group.c
@@ -13,6 +13,7 @@
 #include <linux/dcache.h>
 #include <linux/namei.h>
 #include <linux/err.h>
+#include <linux/fs.h>
 #include "sysfs.h"
 
 
@@ -415,15 +416,18 @@ void sysfs_remove_link_from_group(struct kobject *kobj, const char *group_name,
 EXPORT_SYMBOL_GPL(sysfs_remove_link_from_group);
 
 /**
- * __compat_only_sysfs_link_entry_to_kobj - add a symlink to a kobject pointing
+ * compat_only_sysfs_link_entry_to_kobj - add a symlink to a kobject pointing
  * to a group or an attribute
  * @kobj:		The kobject containing the group.
  * @target_kobj:	The target kobject.
  * @target_name:	The name of the target group or attribute.
+ * @symlink_name:	The name of the symlink file (target_name will be
+ *			considered if symlink_name is NULL).
  */
-int __compat_only_sysfs_link_entry_to_kobj(struct kobject *kobj,
-				      struct kobject *target_kobj,
-				      const char *target_name)
+int compat_only_sysfs_link_entry_to_kobj(struct kobject *kobj,
+					 struct kobject *target_kobj,
+					 const char *target_name,
+					 const char *symlink_name)
 {
 	struct kernfs_node *target;
 	struct kernfs_node *entry;
@@ -448,12 +452,129 @@ int __compat_only_sysfs_link_entry_to_kobj(struct kobject *kobj,
 		return -ENOENT;
 	}
 
-	link = kernfs_create_link(kobj->sd, target_name, entry);
-	if (IS_ERR(link) && PTR_ERR(link) == -EEXIST)
-		sysfs_warn_dup(kobj->sd, target_name);
+	if (!symlink_name)
+		symlink_name = target_name;
+
+	link = kernfs_create_link(kobj->sd, symlink_name, entry);
+	if (PTR_ERR(link) == -EEXIST)
+		sysfs_warn_dup(kobj->sd, symlink_name);
 
 	kernfs_put(entry);
 	kernfs_put(target);
 	return PTR_ERR_OR_ZERO(link);
 }
-EXPORT_SYMBOL_GPL(__compat_only_sysfs_link_entry_to_kobj);
+EXPORT_SYMBOL_GPL(compat_only_sysfs_link_entry_to_kobj);
+
+static int sysfs_group_attrs_change_owner(struct kernfs_node *grp_kn,
+					  const struct attribute_group *grp,
+					  struct iattr *newattrs)
+{
+	struct kernfs_node *kn;
+	int error;
+
+	if (grp->attrs) {
+		struct attribute *const *attr;
+
+		for (attr = grp->attrs; *attr; attr++) {
+			kn = kernfs_find_and_get(grp_kn, (*attr)->name);
+			if (!kn)
+				return -ENOENT;
+
+			error = kernfs_setattr(kn, newattrs);
+			kernfs_put(kn);
+			if (error)
+				return error;
+		}
+	}
+
+	if (grp->bin_attrs) {
+		struct bin_attribute *const *bin_attr;
+
+		for (bin_attr = grp->bin_attrs; *bin_attr; bin_attr++) {
+			kn = kernfs_find_and_get(grp_kn, (*bin_attr)->attr.name);
+			if (!kn)
+				return -ENOENT;
+
+			error = kernfs_setattr(kn, newattrs);
+			kernfs_put(kn);
+			if (error)
+				return error;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * sysfs_group_change_owner - change owner of an attribute group.
+ * @kobj:	The kobject containing the group.
+ * @grp:	The attribute group.
+ * @kuid:	new owner's kuid
+ * @kgid:	new owner's kgid
+ *
+ * Returns 0 on success or error code on failure.
+ */
+int sysfs_group_change_owner(struct kobject *kobj,
+			     const struct attribute_group *grp, kuid_t kuid,
+			     kgid_t kgid)
+{
+	struct kernfs_node *grp_kn;
+	int error;
+	struct iattr newattrs = {
+		.ia_valid = ATTR_UID | ATTR_GID,
+		.ia_uid = kuid,
+		.ia_gid = kgid,
+	};
+
+	if (!kobj->state_in_sysfs)
+		return -EINVAL;
+
+	if (grp->name) {
+		grp_kn = kernfs_find_and_get(kobj->sd, grp->name);
+	} else {
+		kernfs_get(kobj->sd);
+		grp_kn = kobj->sd;
+	}
+	if (!grp_kn)
+		return -ENOENT;
+
+	error = kernfs_setattr(grp_kn, &newattrs);
+	if (!error)
+		error = sysfs_group_attrs_change_owner(grp_kn, grp, &newattrs);
+
+	kernfs_put(grp_kn);
+
+	return error;
+}
+EXPORT_SYMBOL_GPL(sysfs_group_change_owner);
+
+/**
+ * sysfs_groups_change_owner - change owner of a set of attribute groups.
+ * @kobj:	The kobject containing the groups.
+ * @groups:	The attribute groups.
+ * @kuid:	new owner's kuid
+ * @kgid:	new owner's kgid
+ *
+ * Returns 0 on success or error code on failure.
+ */
+int sysfs_groups_change_owner(struct kobject *kobj,
+			      const struct attribute_group **groups,
+			      kuid_t kuid, kgid_t kgid)
+{
+	int error = 0, i;
+
+	if (!kobj->state_in_sysfs)
+		return -EINVAL;
+
+	if (!groups)
+		return 0;
+
+	for (i = 0; groups[i]; i++) {
+		error = sysfs_group_change_owner(kobj, groups[i], kuid, kgid);
+		if (error)
+			break;
+	}
+
+	return error;
+}
+EXPORT_SYMBOL_GPL(sysfs_groups_change_owner);
diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c
index db81cfbab9d6..e747c135c1d1 100644
--- a/fs/sysfs/mount.c
+++ b/fs/sysfs/mount.c
@@ -6,7 +6,7 @@
  * Copyright (c) 2007 SUSE Linux Products GmbH
  * Copyright (c) 2007 Tejun Heo <teheo@suse.de>
  *
- * Please see Documentation/filesystems/sysfs.txt for more information.
+ * Please see Documentation/filesystems/sysfs.rst for more information.
  */
 
 #include <linux/fs.h>
diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c
index c4deecc80f67..5603530a1a52 100644
--- a/fs/sysfs/symlink.c
+++ b/fs/sysfs/symlink.c
@@ -6,7 +6,7 @@
  * Copyright (c) 2007 SUSE Linux Products GmbH
  * Copyright (c) 2007 Tejun Heo <teheo@suse.de>
  *
- * Please see Documentation/filesystems/sysfs.txt for more information.
+ * Please see Documentation/filesystems/sysfs.rst for more information.
  */
 
 #include <linux/fs.h>
diff --git a/fs/sysv/Kconfig b/fs/sysv/Kconfig
index d4edf7d9ae10..b4e23e03fbeb 100644
--- a/fs/sysv/Kconfig
+++ b/fs/sysv/Kconfig
@@ -28,7 +28,7 @@ config SYSV_FS
 	  tar" or preferably "info tar").  Note also that this option has
 	  nothing whatsoever to do with the option "System V IPC". Read about
 	  the System V file system in
-	  <file:Documentation/filesystems/sysv-fs.txt>.
+	  <file:Documentation/filesystems/sysv-fs.rst>.
 	  Saying Y here will enlarge your kernel by about 27 KB.
 
 	  To compile this as a module, choose M here: the module will be called
diff --git a/fs/timerfd.c b/fs/timerfd.c
index 48305ba41e3c..c5509d2448e3 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -26,6 +26,7 @@
 #include <linux/syscalls.h>
 #include <linux/compat.h>
 #include <linux/rcupdate.h>
+#include <linux/time_namespace.h>
 
 struct timerfd_ctx {
 	union {
@@ -196,6 +197,8 @@ static int timerfd_setup(struct timerfd_ctx *ctx, int flags,
 	}
 
 	if (texp != 0) {
+		if (flags & TFD_TIMER_ABSTIME)
+			texp = timens_ktime_to_host(clockid, texp);
 		if (isalarm(ctx)) {
 			if (flags & TFD_TIMER_ABSTIME)
 				alarm_start(&ctx->t.alarm, texp);
@@ -302,11 +305,11 @@ static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count,
 static void timerfd_show(struct seq_file *m, struct file *file)
 {
 	struct timerfd_ctx *ctx = file->private_data;
-	struct itimerspec t;
+	struct timespec64 value, interval;
 
 	spin_lock_irq(&ctx->wqh.lock);
-	t.it_value = ktime_to_timespec(timerfd_get_remaining(ctx));
-	t.it_interval = ktime_to_timespec(ctx->tintv);
+	value = ktime_to_timespec64(timerfd_get_remaining(ctx));
+	interval = ktime_to_timespec64(ctx->tintv);
 	spin_unlock_irq(&ctx->wqh.lock);
 
 	seq_printf(m,
@@ -318,10 +321,10 @@ static void timerfd_show(struct seq_file *m, struct file *file)
 		   ctx->clockid,
 		   (unsigned long long)ctx->ticks,
 		   ctx->settime_flags,
-		   (unsigned long long)t.it_value.tv_sec,
-		   (unsigned long long)t.it_value.tv_nsec,
-		   (unsigned long long)t.it_interval.tv_sec,
-		   (unsigned long long)t.it_interval.tv_nsec);
+		   (unsigned long long)value.tv_sec,
+		   (unsigned long long)value.tv_nsec,
+		   (unsigned long long)interval.tv_sec,
+		   (unsigned long long)interval.tv_nsec);
 }
 #else
 #define timerfd_show NULL
diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c
index 0caa151cae4e..0ee8c6dfb036 100644
--- a/fs/tracefs/inode.c
+++ b/fs/tracefs/inode.c
@@ -330,7 +330,10 @@ static struct dentry *start_creating(const char *name, struct dentry *parent)
 		parent = tracefs_mount->mnt_root;
 
 	inode_lock(parent->d_inode);
-	dentry = lookup_one_len(name, parent, strlen(name));
+	if (unlikely(IS_DEADDIR(parent->d_inode)))
+		dentry = ERR_PTR(-ENOENT);
+	else
+		dentry = lookup_one_len(name, parent, strlen(name));
 	if (!IS_ERR(dentry) && dentry->d_inode) {
 		dput(dentry);
 		dentry = ERR_PTR(-EEXIST);
@@ -499,122 +502,27 @@ __init struct dentry *tracefs_create_instance_dir(const char *name,
 	return dentry;
 }
 
-static int __tracefs_remove(struct dentry *dentry, struct dentry *parent)
+static void remove_one(struct dentry *victim)
 {
-	int ret = 0;
-
-	if (simple_positive(dentry)) {
-		if (dentry->d_inode) {
-			dget(dentry);
-			switch (dentry->d_inode->i_mode & S_IFMT) {
-			case S_IFDIR:
-				ret = simple_rmdir(parent->d_inode, dentry);
-				if (!ret)
-					fsnotify_rmdir(parent->d_inode, dentry);
-				break;
-			default:
-				simple_unlink(parent->d_inode, dentry);
-				fsnotify_unlink(parent->d_inode, dentry);
-				break;
-			}
-			if (!ret)
-				d_delete(dentry);
-			dput(dentry);
-		}
-	}
-	return ret;
-}
-
-/**
- * tracefs_remove - removes a file or directory from the tracefs filesystem
- * @dentry: a pointer to a the dentry of the file or directory to be
- *          removed.
- *
- * This function removes a file or directory in tracefs that was previously
- * created with a call to another tracefs function (like
- * tracefs_create_file() or variants thereof.)
- */
-void tracefs_remove(struct dentry *dentry)
-{
-	struct dentry *parent;
-	int ret;
-
-	if (IS_ERR_OR_NULL(dentry))
-		return;
-
-	parent = dentry->d_parent;
-	inode_lock(parent->d_inode);
-	ret = __tracefs_remove(dentry, parent);
-	inode_unlock(parent->d_inode);
-	if (!ret)
-		simple_release_fs(&tracefs_mount, &tracefs_mount_count);
+	simple_release_fs(&tracefs_mount, &tracefs_mount_count);
 }
 
 /**
- * tracefs_remove_recursive - recursively removes a directory
+ * tracefs_remove - recursively removes a directory
  * @dentry: a pointer to a the dentry of the directory to be removed.
  *
  * This function recursively removes a directory tree in tracefs that
  * was previously created with a call to another tracefs function
  * (like tracefs_create_file() or variants thereof.)
  */
-void tracefs_remove_recursive(struct dentry *dentry)
+void tracefs_remove(struct dentry *dentry)
 {
-	struct dentry *child, *parent;
-
 	if (IS_ERR_OR_NULL(dentry))
 		return;
 
-	parent = dentry;
- down:
-	inode_lock(parent->d_inode);
- loop:
-	/*
-	 * The parent->d_subdirs is protected by the d_lock. Outside that
-	 * lock, the child can be unlinked and set to be freed which can
-	 * use the d_u.d_child as the rcu head and corrupt this list.
-	 */
-	spin_lock(&parent->d_lock);
-	list_for_each_entry(child, &parent->d_subdirs, d_child) {
-		if (!simple_positive(child))
-			continue;
-
-		/* perhaps simple_empty(child) makes more sense */
-		if (!list_empty(&child->d_subdirs)) {
-			spin_unlock(&parent->d_lock);
-			inode_unlock(parent->d_inode);
-			parent = child;
-			goto down;
-		}
-
-		spin_unlock(&parent->d_lock);
-
-		if (!__tracefs_remove(child, parent))
-			simple_release_fs(&tracefs_mount, &tracefs_mount_count);
-
-		/*
-		 * The parent->d_lock protects agaist child from unlinking
-		 * from d_subdirs. When releasing the parent->d_lock we can
-		 * no longer trust that the next pointer is valid.
-		 * Restart the loop. We'll skip this one with the
-		 * simple_positive() check.
-		 */
-		goto loop;
-	}
-	spin_unlock(&parent->d_lock);
-
-	inode_unlock(parent->d_inode);
-	child = parent;
-	parent = parent->d_parent;
-	inode_lock(parent->d_inode);
-
-	if (child != dentry)
-		/* go up */
-		goto loop;
-
-	if (!__tracefs_remove(child, parent))
-		simple_release_fs(&tracefs_mount, &tracefs_mount_count);
-	inode_unlock(parent->d_inode);
+	simple_pin_fs(&trace_fs_type, &tracefs_mount, &tracefs_mount_count);
+	simple_recursive_removal(dentry, remove_one);
+	simple_release_fs(&tracefs_mount, &tracefs_mount_count);
 }
 
 /**
diff --git a/fs/ubifs/Kconfig b/fs/ubifs/Kconfig
index 69932bcfa920..45d3d207fb99 100644
--- a/fs/ubifs/Kconfig
+++ b/fs/ubifs/Kconfig
@@ -12,6 +12,7 @@ config UBIFS_FS
 	select CRYPTO_ZSTD if UBIFS_FS_ZSTD
 	select CRYPTO_HASH_INFO
 	select UBIFS_FS_XATTR if FS_ENCRYPTION
+	select FS_ENCRYPTION_ALGS if FS_ENCRYPTION
 	depends on MTD_UBI
 	help
 	  UBIFS is a file system for flash devices which works on top of UBI.
diff --git a/fs/ubifs/auth.c b/fs/ubifs/auth.c
index 8cdbd53d780c..cc5c0abfd536 100644
--- a/fs/ubifs/auth.c
+++ b/fs/ubifs/auth.c
@@ -31,15 +31,9 @@ int __ubifs_node_calc_hash(const struct ubifs_info *c, const void *node,
 			    u8 *hash)
 {
 	const struct ubifs_ch *ch = node;
-	SHASH_DESC_ON_STACK(shash, c->hash_tfm);
-	int err;
-
-	shash->tfm = c->hash_tfm;
 
-	err = crypto_shash_digest(shash, node, le32_to_cpu(ch->len), hash);
-	if (err < 0)
-		return err;
-	return 0;
+	return crypto_shash_tfm_digest(c->hash_tfm, node, le32_to_cpu(ch->len),
+				       hash);
 }
 
 /**
@@ -53,15 +47,7 @@ int __ubifs_node_calc_hash(const struct ubifs_info *c, const void *node,
 static int ubifs_hash_calc_hmac(const struct ubifs_info *c, const u8 *hash,
 				 u8 *hmac)
 {
-	SHASH_DESC_ON_STACK(shash, c->hmac_tfm);
-	int err;
-
-	shash->tfm = c->hmac_tfm;
-
-	err = crypto_shash_digest(shash, hash, c->hash_len, hmac);
-	if (err < 0)
-		return err;
-	return 0;
+	return crypto_shash_tfm_digest(c->hmac_tfm, hash, c->hash_len, hmac);
 }
 
 /**
@@ -79,13 +65,9 @@ int ubifs_prepare_auth_node(struct ubifs_info *c, void *node,
 			     struct shash_desc *inhash)
 {
 	struct ubifs_auth_node *auth = node;
-	u8 *hash;
+	u8 hash[UBIFS_HASH_ARR_SZ];
 	int err;
 
-	hash = kmalloc(crypto_shash_descsize(c->hash_tfm), GFP_NOFS);
-	if (!hash)
-		return -ENOMEM;
-
 	{
 		SHASH_DESC_ON_STACK(hash_desc, c->hash_tfm);
 
@@ -94,21 +76,16 @@ int ubifs_prepare_auth_node(struct ubifs_info *c, void *node,
 
 		err = crypto_shash_final(hash_desc, hash);
 		if (err)
-			goto out;
+			return err;
 	}
 
 	err = ubifs_hash_calc_hmac(c, hash, auth->hmac);
 	if (err)
-		goto out;
+		return err;
 
 	auth->ch.node_type = UBIFS_AUTH_NODE;
 	ubifs_prepare_node(c, auth, ubifs_auth_node_sz(c), 0);
-
-	err = 0;
-out:
-	kfree(hash);
-
-	return err;
+	return 0;
 }
 
 static struct shash_desc *ubifs_get_desc(const struct ubifs_info *c,
diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c
index e4b52783819d..31288d8fa2ce 100644
--- a/fs/ubifs/debug.c
+++ b/fs/ubifs/debug.c
@@ -815,7 +815,7 @@ void ubifs_dump_leb(const struct ubifs_info *c, int lnum)
 
 	pr_err("(pid %d) start dumping LEB %d\n", current->pid, lnum);
 
-	buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+	buf = __vmalloc(c->leb_size, GFP_NOFS);
 	if (!buf) {
 		ubifs_err(c, "cannot allocate memory for dumping LEB %d", lnum);
 		return;
@@ -2737,18 +2737,6 @@ static ssize_t dfs_file_write(struct file *file, const char __user *u,
 	struct dentry *dent = file->f_path.dentry;
 	int val;
 
-	/*
-	 * TODO: this is racy - the file-system might have already been
-	 * unmounted and we'd oops in this case. The plan is to fix it with
-	 * help of 'iterate_supers_type()' which we should have in v3.0: when
-	 * a debugfs opened, we rember FS's UUID in file->private_data. Then
-	 * whenever we access the FS via a debugfs file, we iterate all UBIFS
-	 * superblocks and fine the one with the same UUID, and take the
-	 * locking right.
-	 *
-	 * The other way to go suggested by Al Viro is to create a separate
-	 * 'ubifs-debug' file-system instead.
-	 */
 	if (file->f_path.dentry == d->dfs_dump_lprops) {
 		ubifs_dump_lprops(c);
 		return count;
diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c
index 0b98e3c8b461..ef85ec167a84 100644
--- a/fs/ubifs/dir.c
+++ b/fs/ubifs/dir.c
@@ -81,7 +81,7 @@ struct inode *ubifs_new_inode(struct ubifs_info *c, struct inode *dir,
 	struct ubifs_inode *ui;
 	bool encrypted = false;
 
-	if (ubifs_crypt_is_encrypted(dir)) {
+	if (IS_ENCRYPTED(dir)) {
 		err = fscrypt_get_encryption_info(dir);
 		if (err) {
 			ubifs_err(c, "fscrypt_get_encryption_info failed: %i", err);
@@ -225,9 +225,9 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry,
 		goto done;
 	}
 
-	if (nm.hash) {
-		ubifs_assert(c, fname_len(&nm) == 0);
-		ubifs_assert(c, fname_name(&nm) == NULL);
+	if (fname_name(&nm) == NULL) {
+		if (nm.hash & ~UBIFS_S_KEY_HASH_MASK)
+			goto done; /* ENOENT */
 		dent_key_init_hash(c, &key, dir->i_ino, nm.hash);
 		err = ubifs_tnc_lookup_dh(c, &key, dent, nm.minor_hash);
 	} else {
@@ -261,7 +261,7 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry,
 		goto done;
 	}
 
-	if (ubifs_crypt_is_encrypted(dir) &&
+	if (IS_ENCRYPTED(dir) &&
 	    (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) &&
 	    !fscrypt_has_permitted_context(dir, inode)) {
 		ubifs_warn(c, "Inconsistent encryption contexts: %lu/%lu",
@@ -499,7 +499,7 @@ static int ubifs_readdir(struct file *file, struct dir_context *ctx)
 	struct ubifs_dent_node *dent;
 	struct inode *dir = file_inode(file);
 	struct ubifs_info *c = dir->i_sb->s_fs_info;
-	bool encrypted = ubifs_crypt_is_encrypted(dir);
+	bool encrypted = IS_ENCRYPTED(dir);
 
 	dbg_gen("dir ino %lu, f_pos %#llx", dir->i_ino, ctx->pos);
 
@@ -512,7 +512,7 @@ static int ubifs_readdir(struct file *file, struct dir_context *ctx)
 
 	if (encrypted) {
 		err = fscrypt_get_encryption_info(dir);
-		if (err && err != -ENOKEY)
+		if (err)
 			return err;
 
 		err = fscrypt_fname_alloc_buffer(dir, UBIFS_MAX_NLEN, &fstr);
@@ -1618,7 +1618,7 @@ int ubifs_getattr(const struct path *path, struct kstat *stat,
 
 static int ubifs_dir_open(struct inode *dir, struct file *file)
 {
-	if (ubifs_crypt_is_encrypted(dir))
+	if (IS_ENCRYPTED(dir))
 		return fscrypt_get_encryption_info(dir) ? -EACCES : 0;
 
 	return 0;
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index cd52585c8f4f..49fe062ce45e 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -67,7 +67,7 @@ static int read_block(struct inode *inode, void *addr, unsigned int block,
 
 	dlen = le32_to_cpu(dn->ch.len) - UBIFS_DATA_NODE_SZ;
 
-	if (ubifs_crypt_is_encrypted(inode)) {
+	if (IS_ENCRYPTED(inode)) {
 		err = ubifs_decrypt(inode, dn, &dlen, block);
 		if (err)
 			goto dump;
@@ -647,7 +647,7 @@ static int populate_page(struct ubifs_info *c, struct page *page,
 			dlen = le32_to_cpu(dn->ch.len) - UBIFS_DATA_NODE_SZ;
 			out_len = UBIFS_BLOCK_SIZE;
 
-			if (ubifs_crypt_is_encrypted(inode)) {
+			if (IS_ENCRYPTED(inode)) {
 				err = ubifs_decrypt(inode, dn, &dlen, page_block);
 				if (err)
 					goto out_err;
@@ -786,7 +786,9 @@ static int ubifs_do_bulk_read(struct ubifs_info *c, struct bu_info *bu,
 
 		if (page_offset > end_index)
 			break;
-		page = find_or_create_page(mapping, page_offset, ra_gfp_mask);
+		page = pagecache_get_page(mapping, page_offset,
+				 FGP_LOCK|FGP_ACCESSED|FGP_CREAT|FGP_NOWAIT,
+				 ra_gfp_mask);
 		if (!page)
 			break;
 		if (!PageUptodate(page))
@@ -1078,18 +1080,12 @@ static void do_attr_changes(struct inode *inode, const struct iattr *attr)
 		inode->i_uid = attr->ia_uid;
 	if (attr->ia_valid & ATTR_GID)
 		inode->i_gid = attr->ia_gid;
-	if (attr->ia_valid & ATTR_ATIME) {
-		inode->i_atime = timestamp_truncate(attr->ia_atime,
-						  inode);
-	}
-	if (attr->ia_valid & ATTR_MTIME) {
-		inode->i_mtime = timestamp_truncate(attr->ia_mtime,
-						  inode);
-	}
-	if (attr->ia_valid & ATTR_CTIME) {
-		inode->i_ctime = timestamp_truncate(attr->ia_ctime,
-						  inode);
-	}
+	if (attr->ia_valid & ATTR_ATIME)
+		inode->i_atime = attr->ia_atime;
+	if (attr->ia_valid & ATTR_MTIME)
+		inode->i_mtime = attr->ia_mtime;
+	if (attr->ia_valid & ATTR_CTIME)
+		inode->i_ctime = attr->ia_ctime;
 	if (attr->ia_valid & ATTR_MODE) {
 		umode_t mode = attr->ia_mode;
 
@@ -1379,7 +1375,6 @@ int ubifs_update_time(struct inode *inode, struct timespec64 *time,
 	struct ubifs_info *c = inode->i_sb->s_fs_info;
 	struct ubifs_budget_req req = { .dirtied_ino = 1,
 			.dirtied_ino_d = ALIGN(ui->data_len, 8) };
-	int iflags = I_DIRTY_TIME;
 	int err, release;
 
 	if (!IS_ENABLED(CONFIG_UBIFS_ATIME_SUPPORT))
@@ -1397,11 +1392,8 @@ int ubifs_update_time(struct inode *inode, struct timespec64 *time,
 	if (flags & S_MTIME)
 		inode->i_mtime = *time;
 
-	if (!(inode->i_sb->s_flags & SB_LAZYTIME))
-		iflags |= I_DIRTY_SYNC;
-
 	release = ui->dirty;
-	__mark_inode_dirty(inode, iflags);
+	__mark_inode_dirty(inode, I_DIRTY_SYNC);
 	mutex_unlock(&ui->ui_mutex);
 	if (release)
 		ubifs_release_budget(c, &req);
diff --git a/fs/ubifs/io.c b/fs/ubifs/io.c
index 8ceb51478800..7e4bfaf2871f 100644
--- a/fs/ubifs/io.c
+++ b/fs/ubifs/io.c
@@ -225,7 +225,7 @@ int ubifs_is_mapped(const struct ubifs_info *c, int lnum)
 int ubifs_check_node(const struct ubifs_info *c, const void *buf, int lnum,
 		     int offs, int quiet, int must_chk_crc)
 {
-	int err = -EINVAL, type, node_len;
+	int err = -EINVAL, type, node_len, dump_node = 1;
 	uint32_t crc, node_crc, magic;
 	const struct ubifs_ch *ch = buf;
 
@@ -278,10 +278,22 @@ int ubifs_check_node(const struct ubifs_info *c, const void *buf, int lnum,
 out_len:
 	if (!quiet)
 		ubifs_err(c, "bad node length %d", node_len);
+	if (type == UBIFS_DATA_NODE && node_len > UBIFS_DATA_NODE_SZ)
+		dump_node = 0;
 out:
 	if (!quiet) {
 		ubifs_err(c, "bad node at LEB %d:%d", lnum, offs);
-		ubifs_dump_node(c, buf);
+		if (dump_node) {
+			ubifs_dump_node(c, buf);
+		} else {
+			int safe_len = min3(node_len, c->leb_size - offs,
+				(int)UBIFS_MAX_DATA_NODE_SZ);
+			pr_err("\tprevent out-of-bounds memory access\n");
+			pr_err("\ttruncated data node length      %d\n", safe_len);
+			pr_err("\tcorrupted data node:\n");
+			print_hex_dump(KERN_ERR, "\t", DUMP_PREFIX_OFFSET, 32, 1,
+					buf, safe_len, 0);
+		}
 		dump_stack();
 	}
 	return err;
diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c
index 5dc5abca11c7..3df9be2c684c 100644
--- a/fs/ubifs/ioctl.c
+++ b/fs/ubifs/ioctl.c
@@ -17,10 +17,14 @@
 #include "ubifs.h"
 
 /* Need to be kept consistent with checked flags in ioctl2ubifs() */
-#define UBIFS_SUPPORTED_IOCTL_FLAGS \
+#define UBIFS_SETTABLE_IOCTL_FLAGS \
 	(FS_COMPR_FL | FS_SYNC_FL | FS_APPEND_FL | \
 	 FS_IMMUTABLE_FL | FS_DIRSYNC_FL)
 
+/* Need to be kept consistent with checked flags in ubifs2ioctl() */
+#define UBIFS_GETTABLE_IOCTL_FLAGS \
+	(UBIFS_SETTABLE_IOCTL_FLAGS | FS_ENCRYPT_FL)
+
 /**
  * ubifs_set_inode_flags - set VFS inode flags.
  * @inode: VFS inode to set flags for
@@ -91,6 +95,8 @@ static int ubifs2ioctl(int ubifs_flags)
 		ioctl_flags |= FS_IMMUTABLE_FL;
 	if (ubifs_flags & UBIFS_DIRSYNC_FL)
 		ioctl_flags |= FS_DIRSYNC_FL;
+	if (ubifs_flags & UBIFS_CRYPT_FL)
+		ioctl_flags |= FS_ENCRYPT_FL;
 
 	return ioctl_flags;
 }
@@ -113,7 +119,8 @@ static int setflags(struct inode *inode, int flags)
 	if (err)
 		goto out_unlock;
 
-	ui->flags = ioctl2ubifs(flags);
+	ui->flags &= ~ioctl2ubifs(UBIFS_SETTABLE_IOCTL_FLAGS);
+	ui->flags |= ioctl2ubifs(flags);
 	ubifs_set_inode_flags(inode);
 	inode->i_ctime = current_time(inode);
 	release = ui->dirty;
@@ -155,8 +162,9 @@ long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 		if (get_user(flags, (int __user *) arg))
 			return -EFAULT;
 
-		if (flags & ~UBIFS_SUPPORTED_IOCTL_FLAGS)
+		if (flags & ~UBIFS_GETTABLE_IOCTL_FLAGS)
 			return -EOPNOTSUPP;
+		flags &= UBIFS_SETTABLE_IOCTL_FLAGS;
 
 		if (!S_ISDIR(inode->i_mode))
 			flags &= ~FS_DIRSYNC_FL;
@@ -200,6 +208,9 @@ long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case FS_IOC_GET_ENCRYPTION_KEY_STATUS:
 		return fscrypt_ioctl_get_key_status(file, (void __user *)arg);
 
+	case FS_IOC_GET_ENCRYPTION_NONCE:
+		return fscrypt_ioctl_get_nonce(file, (void __user *)arg);
+
 	default:
 		return -ENOTTY;
 	}
@@ -222,6 +233,7 @@ long ubifs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case FS_IOC_REMOVE_ENCRYPTION_KEY:
 	case FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS:
 	case FS_IOC_GET_ENCRYPTION_KEY_STATUS:
+	case FS_IOC_GET_ENCRYPTION_NONCE:
 		break;
 	default:
 		return -ENOIOCTLCMD;
diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c
index 4fd9683b8245..e5ec1afe1c66 100644
--- a/fs/ubifs/journal.c
+++ b/fs/ubifs/journal.c
@@ -503,7 +503,7 @@ static void mark_inode_clean(struct ubifs_info *c, struct ubifs_inode *ui)
 static void set_dent_cookie(struct ubifs_info *c, struct ubifs_dent_node *dent)
 {
 	if (c->double_hash)
-		dent->cookie = prandom_u32();
+		dent->cookie = (__force __le32) prandom_u32();
 	else
 		dent->cookie = 0;
 }
@@ -588,7 +588,7 @@ int ubifs_jnl_update(struct ubifs_info *c, const struct inode *dir,
 
 	if (!xent) {
 		dent->ch.node_type = UBIFS_DENT_NODE;
-		if (nm->hash)
+		if (fname_name(nm) == NULL)
 			dent_key_init_hash(c, &dent_key, dir->i_ino, nm->hash);
 		else
 			dent_key_init(c, &dent_key, dir->i_ino, nm);
@@ -646,7 +646,7 @@ int ubifs_jnl_update(struct ubifs_info *c, const struct inode *dir,
 	ubifs_add_auth_dirt(c, lnum);
 
 	if (deletion) {
-		if (nm->hash)
+		if (fname_name(nm) == NULL)
 			err = ubifs_tnc_remove_dh(c, &dent_key, nm->minor_hash);
 		else
 			err = ubifs_tnc_remove_nm(c, &dent_key, nm);
@@ -727,7 +727,7 @@ int ubifs_jnl_write_data(struct ubifs_info *c, const struct inode *inode,
 	int dlen = COMPRESSED_DATA_NODE_BUF_SZ, allocated = 1;
 	int write_len;
 	struct ubifs_inode *ui = ubifs_inode(inode);
-	bool encrypted = ubifs_crypt_is_encrypted(inode);
+	bool encrypted = IS_ENCRYPTED(inode);
 	u8 hash[UBIFS_HASH_ARR_SZ];
 
 	dbg_jnlk(key, "ino %lu, blk %u, len %d, key ",
@@ -899,12 +899,13 @@ int ubifs_jnl_write_inode(struct ubifs_info *c, const struct inode *inode)
 			fname_name(&nm) = xent->name;
 			fname_len(&nm) = le16_to_cpu(xent->nlen);
 
-			xino = ubifs_iget(c->vfs_sb, xent->inum);
+			xino = ubifs_iget(c->vfs_sb, le64_to_cpu(xent->inum));
 			if (IS_ERR(xino)) {
 				err = PTR_ERR(xino);
 				ubifs_err(c, "dead directory entry '%s', error %d",
 					  xent->name, err);
 				ubifs_ro_mode(c, err);
+				kfree(xent);
 				goto out_release;
 			}
 			ubifs_assert(c, ubifs_inode(xino)->xattr);
@@ -1449,7 +1450,7 @@ static int truncate_data_node(const struct ubifs_info *c, const struct inode *in
 	dlen = old_dlen = le32_to_cpu(dn->ch.len) - UBIFS_DATA_NODE_SZ;
 	compr_type = le16_to_cpu(dn->compr_type);
 
-	if (ubifs_crypt_is_encrypted(inode)) {
+	if (IS_ENCRYPTED(inode)) {
 		err = ubifs_decrypt(inode, dn, &dlen, block);
 		if (err)
 			goto out;
@@ -1465,7 +1466,7 @@ static int truncate_data_node(const struct ubifs_info *c, const struct inode *in
 		ubifs_compress(c, buf, *new_len, &dn->data, &out_len, &compr_type);
 	}
 
-	if (ubifs_crypt_is_encrypted(inode)) {
+	if (IS_ENCRYPTED(inode)) {
 		err = ubifs_encrypt(inode, dn, out_len, &old_dlen, block);
 		if (err)
 			goto out;
diff --git a/fs/ubifs/key.h b/fs/ubifs/key.h
index afa704ff5ca0..8142d9d6fe5d 100644
--- a/fs/ubifs/key.h
+++ b/fs/ubifs/key.h
@@ -150,7 +150,6 @@ static inline void dent_key_init(const struct ubifs_info *c,
 	uint32_t hash = c->key_hash(fname_name(nm), fname_len(nm));
 
 	ubifs_assert(c, !(hash & ~UBIFS_S_KEY_HASH_MASK));
-	ubifs_assert(c, !nm->hash && !nm->minor_hash);
 	key->u32[0] = inum;
 	key->u32[1] = hash | (UBIFS_DENT_KEY << UBIFS_S_KEY_HASH_BITS);
 }
diff --git a/fs/ubifs/lprops.c b/fs/ubifs/lprops.c
index 29826c51883a..22bfda158f7f 100644
--- a/fs/ubifs/lprops.c
+++ b/fs/ubifs/lprops.c
@@ -1095,7 +1095,7 @@ static int scan_check_cb(struct ubifs_info *c,
 		return LPT_SCAN_CONTINUE;
 	}
 
-	buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+	buf = __vmalloc(c->leb_size, GFP_NOFS);
 	if (!buf)
 		return -ENOMEM;
 
diff --git a/fs/ubifs/lpt_commit.c b/fs/ubifs/lpt_commit.c
index ff5e0411cf2d..d76a19e460cd 100644
--- a/fs/ubifs/lpt_commit.c
+++ b/fs/ubifs/lpt_commit.c
@@ -1596,7 +1596,7 @@ static int dbg_check_ltab_lnum(struct ubifs_info *c, int lnum)
 	if (!dbg_is_chk_lprops(c))
 		return 0;
 
-	buf = p = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+	buf = p = __vmalloc(c->leb_size, GFP_NOFS);
 	if (!buf) {
 		ubifs_err(c, "cannot allocate memory for ltab checking");
 		return 0;
@@ -1845,7 +1845,7 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum)
 	void *buf, *p;
 
 	pr_err("(pid %d) start dumping LEB %d\n", current->pid, lnum);
-	buf = p = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+	buf = p = __vmalloc(c->leb_size, GFP_NOFS);
 	if (!buf) {
 		ubifs_err(c, "cannot allocate memory to dump LPT");
 		return;
diff --git a/fs/ubifs/master.c b/fs/ubifs/master.c
index 52a85c01397e..911d0555b9f2 100644
--- a/fs/ubifs/master.c
+++ b/fs/ubifs/master.c
@@ -68,12 +68,9 @@ static int mst_node_check_hash(const struct ubifs_info *c,
 	u8 calc[UBIFS_MAX_HASH_LEN];
 	const void *node = mst;
 
-	SHASH_DESC_ON_STACK(shash, c->hash_tfm);
-
-	shash->tfm = c->hash_tfm;
-
-	crypto_shash_digest(shash, node + sizeof(struct ubifs_ch),
-			    UBIFS_MST_NODE_SZ - sizeof(struct ubifs_ch), calc);
+	crypto_shash_tfm_digest(c->hash_tfm, node + sizeof(struct ubifs_ch),
+				UBIFS_MST_NODE_SZ - sizeof(struct ubifs_ch),
+				calc);
 
 	if (ubifs_check_hash(c, expected, calc))
 		return -EPERM;
diff --git a/fs/ubifs/orphan.c b/fs/ubifs/orphan.c
index 3b4b4114f208..2c294085ffed 100644
--- a/fs/ubifs/orphan.c
+++ b/fs/ubifs/orphan.c
@@ -129,7 +129,7 @@ static void __orphan_drop(struct ubifs_info *c, struct ubifs_orphan *o)
 static void orphan_delete(struct ubifs_info *c, struct ubifs_orphan *orph)
 {
 	if (orph->del) {
-		dbg_gen("deleted twice ino %lu", orph->inum);
+		dbg_gen("deleted twice ino %lu", (unsigned long)orph->inum);
 		return;
 	}
 
@@ -137,7 +137,7 @@ static void orphan_delete(struct ubifs_info *c, struct ubifs_orphan *orph)
 		orph->del = 1;
 		orph->dnext = c->orph_dnext;
 		c->orph_dnext = orph;
-		dbg_gen("delete later ino %lu", orph->inum);
+		dbg_gen("delete later ino %lu", (unsigned long)orph->inum);
 		return;
 	}
 
@@ -157,7 +157,7 @@ int ubifs_add_orphan(struct ubifs_info *c, ino_t inum)
 	int err = 0;
 	ino_t xattr_inum;
 	union ubifs_key key;
-	struct ubifs_dent_node *xent;
+	struct ubifs_dent_node *xent, *pxent = NULL;
 	struct fscrypt_name nm = {0};
 	struct ubifs_orphan *xattr_orphan;
 	struct ubifs_orphan *orphan;
@@ -181,11 +181,16 @@ int ubifs_add_orphan(struct ubifs_info *c, ino_t inum)
 		xattr_inum = le64_to_cpu(xent->inum);
 
 		xattr_orphan = orphan_add(c, xattr_inum, orphan);
-		if (IS_ERR(xattr_orphan))
+		if (IS_ERR(xattr_orphan)) {
+			kfree(xent);
 			return PTR_ERR(xattr_orphan);
+		}
 
+		kfree(pxent);
+		pxent = xent;
 		key_read(c, &xent->key, &key);
 	}
+	kfree(pxent);
 
 	return 0;
 }
@@ -631,12 +636,17 @@ static int do_kill_orphans(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 	ino_t inum;
 	int i, n, err, first = 1;
 
+	ino = kmalloc(UBIFS_MAX_INO_NODE_SZ, GFP_NOFS);
+	if (!ino)
+		return -ENOMEM;
+
 	list_for_each_entry(snod, &sleb->nodes, list) {
 		if (snod->type != UBIFS_ORPH_NODE) {
 			ubifs_err(c, "invalid node type %d in orphan area at %d:%d",
 				  snod->type, sleb->lnum, snod->offs);
 			ubifs_dump_node(c, snod->node);
-			return -EINVAL;
+			err = -EINVAL;
+			goto out_free;
 		}
 
 		orph = snod->node;
@@ -663,20 +673,18 @@ static int do_kill_orphans(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 				ubifs_err(c, "out of order commit number %llu in orphan node at %d:%d",
 					  cmt_no, sleb->lnum, snod->offs);
 				ubifs_dump_node(c, snod->node);
-				return -EINVAL;
+				err = -EINVAL;
+				goto out_free;
 			}
 			dbg_rcvry("out of date LEB %d", sleb->lnum);
 			*outofdate = 1;
-			return 0;
+			err = 0;
+			goto out_free;
 		}
 
 		if (first)
 			first = 0;
 
-		ino = kmalloc(UBIFS_MAX_INO_NODE_SZ, GFP_NOFS);
-		if (!ino)
-			return -ENOMEM;
-
 		n = (le32_to_cpu(orph->ch.len) - UBIFS_ORPH_NODE_SZ) >> 3;
 		for (i = 0; i < n; i++) {
 			union ubifs_key key1, key2;
@@ -685,14 +693,14 @@ static int do_kill_orphans(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 
 			ino_key_init(c, &key1, inum);
 			err = ubifs_tnc_lookup(c, &key1, ino);
-			if (err)
+			if (err && err != -ENOENT)
 				goto out_free;
 
 			/*
 			 * Check whether an inode can really get deleted.
 			 * linkat() with O_TMPFILE allows rebirth of an inode.
 			 */
-			if (ino->nlink == 0) {
+			if (err == 0 && ino->nlink == 0) {
 				dbg_rcvry("deleting orphaned inode %lu",
 					  (unsigned long)inum);
 
@@ -969,7 +977,7 @@ static int dbg_scan_orphans(struct ubifs_info *c, struct check_info *ci)
 	if (c->no_orphs)
 		return 0;
 
-	buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL);
+	buf = __vmalloc(c->leb_size, GFP_NOFS);
 	if (!buf) {
 		ubifs_err(c, "cannot allocate memory to check orphans");
 		return 0;
diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c
index b28ac4dfb407..b69ffac7e415 100644
--- a/fs/ubifs/replay.c
+++ b/fs/ubifs/replay.c
@@ -558,7 +558,7 @@ static int is_last_bud(struct ubifs_info *c, struct ubifs_bud *bud)
 	return data == 0xFFFFFFFF;
 }
 
-/* authenticate_sleb_hash and authenticate_sleb_hmac are split out for stack usage */
+/* authenticate_sleb_hash is split out for stack usage */
 static int authenticate_sleb_hash(struct ubifs_info *c, struct shash_desc *log_hash, u8 *hash)
 {
 	SHASH_DESC_ON_STACK(hash_desc, c->hash_tfm);
@@ -569,15 +569,6 @@ static int authenticate_sleb_hash(struct ubifs_info *c, struct shash_desc *log_h
 	return crypto_shash_final(hash_desc, hash);
 }
 
-static int authenticate_sleb_hmac(struct ubifs_info *c, u8 *hash, u8 *hmac)
-{
-	SHASH_DESC_ON_STACK(hmac_desc, c->hmac_tfm);
-
-	hmac_desc->tfm = c->hmac_tfm;
-
-	return crypto_shash_digest(hmac_desc, hash, c->hash_len, hmac);
-}
-
 /**
  * authenticate_sleb - authenticate one scan LEB
  * @c: UBIFS file-system description object
@@ -601,18 +592,12 @@ static int authenticate_sleb(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 	struct ubifs_scan_node *snod;
 	int n_nodes = 0;
 	int err;
-	u8 *hash, *hmac;
+	u8 hash[UBIFS_HASH_ARR_SZ];
+	u8 hmac[UBIFS_HMAC_ARR_SZ];
 
 	if (!ubifs_authenticated(c))
 		return sleb->nodes_cnt;
 
-	hash = kmalloc(crypto_shash_descsize(c->hash_tfm), GFP_NOFS);
-	hmac = kmalloc(c->hmac_desc_len, GFP_NOFS);
-	if (!hash || !hmac) {
-		err = -ENOMEM;
-		goto out;
-	}
-
 	list_for_each_entry(snod, &sleb->nodes, list) {
 
 		n_nodes++;
@@ -624,7 +609,8 @@ static int authenticate_sleb(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 			if (err)
 				goto out;
 
-			err = authenticate_sleb_hmac(c, hash, hmac);
+			err = crypto_shash_tfm_digest(c->hmac_tfm, hash,
+						      c->hash_len, hmac);
 			if (err)
 				goto out;
 
@@ -662,9 +648,6 @@ static int authenticate_sleb(struct ubifs_info *c, struct ubifs_scan_leb *sleb,
 		err = 0;
 	}
 out:
-	kfree(hash);
-	kfree(hmac);
-
 	return err ? err : n_nodes - n_not_auth;
 }
 
diff --git a/fs/ubifs/sb.c b/fs/ubifs/sb.c
index a551eb3e9b89..4b4b65b48c57 100644
--- a/fs/ubifs/sb.c
+++ b/fs/ubifs/sb.c
@@ -84,7 +84,6 @@ static int create_default_filesystem(struct ubifs_info *c)
 	int idx_node_size;
 	long long tmp64, main_bytes;
 	__le64 tmp_le64;
-	__le32 tmp_le32;
 	struct timespec64 ts;
 	u8 hash[UBIFS_HASH_ARR_SZ];
 	u8 hash_lpt[UBIFS_HASH_ARR_SZ];
@@ -161,7 +160,7 @@ static int create_default_filesystem(struct ubifs_info *c)
 	sup = kzalloc(ALIGN(UBIFS_SB_NODE_SZ, c->min_io_size), GFP_KERNEL);
 	mst = kzalloc(c->mst_node_alsz, GFP_KERNEL);
 	idx_node_size = ubifs_idx_node_sz(c, 1);
-	idx = kzalloc(ALIGN(tmp, c->min_io_size), GFP_KERNEL);
+	idx = kzalloc(ALIGN(idx_node_size, c->min_io_size), GFP_KERNEL);
 	ino = kzalloc(ALIGN(UBIFS_INO_NODE_SZ, c->min_io_size), GFP_KERNEL);
 	cs = kzalloc(ALIGN(UBIFS_CS_NODE_SZ, c->min_io_size), GFP_KERNEL);
 
@@ -184,7 +183,7 @@ static int create_default_filesystem(struct ubifs_info *c)
 		if (err)
 			goto out;
 	} else {
-		sup->hash_algo = 0xffff;
+		sup->hash_algo = cpu_to_le16(0xffff);
 	}
 
 	sup->ch.node_type  = UBIFS_SB_NODE;
@@ -291,16 +290,14 @@ static int create_default_filesystem(struct ubifs_info *c)
 	ino->creat_sqnum = cpu_to_le64(++c->max_sqnum);
 	ino->nlink = cpu_to_le32(2);
 
-	ktime_get_real_ts64(&ts);
-	ts = timespec64_trunc(ts, DEFAULT_TIME_GRAN);
+	ktime_get_coarse_real_ts64(&ts);
 	tmp_le64 = cpu_to_le64(ts.tv_sec);
 	ino->atime_sec   = tmp_le64;
 	ino->ctime_sec   = tmp_le64;
 	ino->mtime_sec   = tmp_le64;
-	tmp_le32 = cpu_to_le32(ts.tv_nsec);
-	ino->atime_nsec  = tmp_le32;
-	ino->ctime_nsec  = tmp_le32;
-	ino->mtime_nsec  = tmp_le32;
+	ino->atime_nsec  = 0;
+	ino->ctime_nsec  = 0;
+	ino->mtime_nsec  = 0;
 	ino->mode = cpu_to_le32(S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO);
 	ino->size = cpu_to_le64(UBIFS_INO_NODE_SZ);
 
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 7d4547e5202d..7fc2f3f07c16 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -1599,6 +1599,7 @@ out_free:
 	vfree(c->ileb_buf);
 	vfree(c->sbuf);
 	kfree(c->bottom_up_buf);
+	kfree(c->sup_node);
 	ubifs_debugging_exit(c);
 	return err;
 }
@@ -1641,6 +1642,7 @@ static void ubifs_umount(struct ubifs_info *c)
 	vfree(c->ileb_buf);
 	vfree(c->sbuf);
 	kfree(c->bottom_up_buf);
+	kfree(c->sup_node);
 	ubifs_debugging_exit(c);
 }
 
@@ -2267,10 +2269,8 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags,
 		}
 	} else {
 		err = ubifs_fill_super(sb, data, flags & SB_SILENT ? 1 : 0);
-		if (err) {
-			kfree(c);
+		if (err)
 			goto out_deact;
-		}
 		/* We do not support atime */
 		sb->s_flags |= SB_ACTIVE;
 		if (IS_ENABLED(CONFIG_UBIFS_ATIME_SUPPORT))
diff --git a/fs/ubifs/tnc_commit.c b/fs/ubifs/tnc_commit.c
index a384a0f9ff32..234be1c4dc87 100644
--- a/fs/ubifs/tnc_commit.c
+++ b/fs/ubifs/tnc_commit.c
@@ -212,7 +212,7 @@ static int is_idx_node_in_use(struct ubifs_info *c, union ubifs_key *key,
 /**
  * layout_leb_in_gaps - layout index nodes using in-the-gaps method.
  * @c: UBIFS file-system description object
- * @p: return LEB number here
+ * @p: return LEB number in @c->gap_lebs[p]
  *
  * This function lays out new index nodes for dirty znodes using in-the-gaps
  * method of TNC commit.
@@ -221,7 +221,7 @@ static int is_idx_node_in_use(struct ubifs_info *c, union ubifs_key *key,
  * This function returns the number of index nodes written into the gaps, or a
  * negative error code on failure.
  */
-static int layout_leb_in_gaps(struct ubifs_info *c, int *p)
+static int layout_leb_in_gaps(struct ubifs_info *c, int p)
 {
 	struct ubifs_scan_leb *sleb;
 	struct ubifs_scan_node *snod;
@@ -236,7 +236,7 @@ static int layout_leb_in_gaps(struct ubifs_info *c, int *p)
 		 * filled, however we do not check there at present.
 		 */
 		return lnum; /* Error code */
-	*p = lnum;
+	c->gap_lebs[p] = lnum;
 	dbg_gc("LEB %d", lnum);
 	/*
 	 * Scan the index LEB.  We use the generic scan for this even though
@@ -355,7 +355,7 @@ static int get_leb_cnt(struct ubifs_info *c, int cnt)
  */
 static int layout_in_gaps(struct ubifs_info *c, int cnt)
 {
-	int err, leb_needed_cnt, written, *p;
+	int err, leb_needed_cnt, written, p = 0, old_idx_lebs, *gap_lebs;
 
 	dbg_gc("%d znodes to write", cnt);
 
@@ -364,9 +364,9 @@ static int layout_in_gaps(struct ubifs_info *c, int cnt)
 	if (!c->gap_lebs)
 		return -ENOMEM;
 
-	p = c->gap_lebs;
+	old_idx_lebs = c->lst.idx_lebs;
 	do {
-		ubifs_assert(c, p < c->gap_lebs + c->lst.idx_lebs);
+		ubifs_assert(c, p < c->lst.idx_lebs);
 		written = layout_leb_in_gaps(c, p);
 		if (written < 0) {
 			err = written;
@@ -392,9 +392,29 @@ static int layout_in_gaps(struct ubifs_info *c, int cnt)
 		leb_needed_cnt = get_leb_cnt(c, cnt);
 		dbg_gc("%d znodes remaining, need %d LEBs, have %d", cnt,
 		       leb_needed_cnt, c->ileb_cnt);
+		/*
+		 * Dynamically change the size of @c->gap_lebs to prevent
+		 * oob, because @c->lst.idx_lebs could be increased by
+		 * function @get_idx_gc_leb (called by layout_leb_in_gaps->
+		 * ubifs_find_dirty_idx_leb) during loop. Only enlarge
+		 * @c->gap_lebs when needed.
+		 *
+		 */
+		if (leb_needed_cnt > c->ileb_cnt && p >= old_idx_lebs &&
+		    old_idx_lebs < c->lst.idx_lebs) {
+			old_idx_lebs = c->lst.idx_lebs;
+			gap_lebs = krealloc(c->gap_lebs, sizeof(int) *
+					       (old_idx_lebs + 1), GFP_NOFS);
+			if (!gap_lebs) {
+				kfree(c->gap_lebs);
+				c->gap_lebs = NULL;
+				return -ENOMEM;
+			}
+			c->gap_lebs = gap_lebs;
+		}
 	} while (leb_needed_cnt > c->ileb_cnt);
 
-	*p = -1;
+	c->gap_lebs[p] = -1;
 	return 0;
 }
 
diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h
index c55f212dcb75..bff682309fbe 100644
--- a/fs/ubifs/ubifs.h
+++ b/fs/ubifs/ubifs.h
@@ -2095,13 +2095,6 @@ int ubifs_decrypt(const struct inode *inode, struct ubifs_data_node *dn,
 
 extern const struct fscrypt_operations ubifs_crypt_operations;
 
-static inline bool ubifs_crypt_is_encrypted(const struct inode *inode)
-{
-	const struct ubifs_inode *ui = ubifs_inode(inode);
-
-	return ui->flags & UBIFS_CRYPT_FL;
-}
-
 /* Normal UBIFS messages */
 __printf(2, 3)
 void ubifs_msg(const struct ubifs_info *c, const char *fmt, ...);
diff --git a/fs/udf/Kconfig b/fs/udf/Kconfig
index 6848de581ce1..26e1a49f3ba7 100644
--- a/fs/udf/Kconfig
+++ b/fs/udf/Kconfig
@@ -9,7 +9,7 @@ config UDF_FS
 	  compatible with standard unix file systems, it is also suitable for
 	  removable USB disks. Say Y if you intend to mount DVD discs or CDRW's
 	  written in packet mode, or if you want to use UDF for removable USB
-	  disks. Please read <file:Documentation/filesystems/udf.txt>.
+	  disks. Please read <file:Documentation/filesystems/udf.rst>.
 
 	  To compile this file system support as a module, choose M here: the
 	  module will be called udf.
diff --git a/fs/udf/ecma_167.h b/fs/udf/ecma_167.h
index fb7f2c7bec9c..736ebc5dc441 100644
--- a/fs/udf/ecma_167.h
+++ b/fs/udf/ecma_167.h
@@ -4,7 +4,8 @@
  * This file is based on ECMA-167 3rd edition (June 1997)
  * http://www.ecma.ch
  *
- * Copyright (c) 2001-2002  Ben Fennema <bfennema@falcon.csc.calpoly.edu>
+ * Copyright (c) 2001-2002  Ben Fennema
+ * Copyright (c) 2017-2019  Pali Rohár <pali@kernel.org>
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -32,11 +33,19 @@
  * SUCH DAMAGE.
  */
 
+/**
+ * @file
+ * ECMA-167r3 defines and structure definitions
+ */
+
 #include <linux/types.h>
 
 #ifndef _ECMA_167_H
 #define _ECMA_167_H 1
 
+/* Character sets and coding - d-characters (ECMA 167r3 1/7.2) */
+typedef uint8_t		dchars;
+
 /* Character set specification (ECMA 167r3 1/7.2.1) */
 struct charspec {
 	uint8_t		charSetType;
@@ -54,6 +63,7 @@ struct charspec {
 #define CHARSPEC_TYPE_CS7		0x07	/* (1/7.2.9) */
 #define CHARSPEC_TYPE_CS8		0x08	/* (1/7.2.10) */
 
+/* Fixed-length character fields - d-string (EMCA 167r3 1/7.2.12) */
 typedef uint8_t		dstring;
 
 /* Timestamp (ECMA 167r3 1/7.3) */
@@ -85,22 +95,8 @@ struct regid {
 } __packed;
 
 /* Flags (ECMA 167r3 1/7.4.1) */
-#define ENTITYID_FLAGS_DIRTY		0x00
-#define ENTITYID_FLAGS_PROTECTED	0x01
-
-/* OSTA UDF 2.1.5.2 */
-#define UDF_ID_COMPLIANT "*OSTA UDF Compliant"
-
-/* OSTA UDF 2.1.5.3 */
-struct domainEntityIDSuffix {
-	uint16_t	revision;
-	uint8_t		flags;
-	uint8_t		reserved[5];
-};
-
-/* OSTA UDF 2.1.5.3 */
-#define ENTITYIDSUFFIX_FLAGS_HARDWRITEPROTECT 0
-#define ENTITYIDSUFFIX_FLAGS_SOFTWRITEPROTECT 1
+#define ENTITYID_FLAGS_DIRTY		0x01
+#define ENTITYID_FLAGS_PROTECTED	0x02
 
 /* Volume Structure Descriptor (ECMA 167r3 2/9.1) */
 #define VSD_STD_ID_LEN			5
@@ -202,6 +198,13 @@ struct NSRDesc {
 	uint8_t		structData[2040];
 } __packed;
 
+/* Generic Descriptor */
+struct genericDesc {
+	struct tag	descTag;
+	__le32		volDescSeqNum;
+	uint8_t		reserved[492];
+} __packed;
+
 /* Primary Volume Descriptor (ECMA 167r3 3/10.1) */
 struct primaryVolDesc {
 	struct tag		descTag;
@@ -316,7 +319,7 @@ struct genericPartitionMap {
 
 /* Partition Map Type (ECMA 167r3 3/10.7.1.1) */
 #define GP_PARTITION_MAP_TYPE_UNDEF	0x00
-#define GP_PARTIITON_MAP_TYPE_1		0x01
+#define GP_PARTITION_MAP_TYPE_1		0x01
 #define GP_PARTITION_MAP_TYPE_2		0x02
 
 /* Type 1 Partition Map (ECMA 167r3 3/10.7.2) */
@@ -723,6 +726,7 @@ struct appUseExtAttr {
 #define EXTATTR_DEV_SPEC		12
 #define EXTATTR_IMP_USE			2048
 #define EXTATTR_APP_USE			65536
+#define EXTATTR_SUBTYPE			1
 
 /* Unallocated Space Entry (ECMA 167r3 4/14.11) */
 struct unallocSpaceEntry {
@@ -754,10 +758,12 @@ struct partitionIntegrityEntry {
 /* Short Allocation Descriptor (ECMA 167r3 4/14.14.1) */
 
 /* Extent Length (ECMA 167r3 4/14.14.1.1) */
+#define EXT_LENGTH_MASK			0x3FFFFFFF
+#define EXT_TYPE_MASK			0xC0000000
 #define EXT_RECORDED_ALLOCATED		0x00000000
 #define EXT_NOT_RECORDED_ALLOCATED	0x40000000
 #define EXT_NOT_RECORDED_NOT_ALLOCATED	0x80000000
-#define EXT_NEXT_EXTENT_ALLOCDECS	0xC0000000
+#define EXT_NEXT_EXTENT_ALLOCDESCS	0xC0000000
 
 /* Long Allocation Descriptor (ECMA 167r3 4/14.14.2) */
 
@@ -774,7 +780,7 @@ struct pathComponent {
 	uint8_t		componentType;
 	uint8_t		lengthComponentIdent;
 	__le16		componentFileVersionNum;
-	dstring		componentIdent[0];
+	dchars		componentIdent[0];
 } __packed;
 
 /* File Entry (ECMA 167r3 4/14.17) */
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index ea80036d7897..adaba8e8b326 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -195,10 +195,9 @@ static int udf_readpage(struct file *file, struct page *page)
 	return mpage_readpage(page, udf_get_block);
 }
 
-static int udf_readpages(struct file *file, struct address_space *mapping,
-			struct list_head *pages, unsigned nr_pages)
+static void udf_readahead(struct readahead_control *rac)
 {
-	return mpage_readpages(mapping, pages, nr_pages, udf_get_block);
+	mpage_readahead(rac, udf_get_block);
 }
 
 static int udf_write_begin(struct file *file, struct address_space *mapping,
@@ -234,7 +233,7 @@ static sector_t udf_bmap(struct address_space *mapping, sector_t block)
 
 const struct address_space_operations udf_aops = {
 	.readpage	= udf_readpage,
-	.readpages	= udf_readpages,
+	.readahead	= udf_readahead,
 	.writepage	= udf_writepage,
 	.writepages	= udf_writepages,
 	.write_begin	= udf_write_begin,
@@ -1981,10 +1980,10 @@ int udf_setup_indirect_aext(struct inode *inode, udf_pblk_t block,
 
 		__udf_add_aext(inode, &nepos, &cp_loc, cp_len, 1);
 		udf_write_aext(inode, epos, &nepos.block,
-			       sb->s_blocksize | EXT_NEXT_EXTENT_ALLOCDECS, 0);
+			       sb->s_blocksize | EXT_NEXT_EXTENT_ALLOCDESCS, 0);
 	} else {
 		__udf_add_aext(inode, epos, &nepos.block,
-			       sb->s_blocksize | EXT_NEXT_EXTENT_ALLOCDECS, 0);
+			       sb->s_blocksize | EXT_NEXT_EXTENT_ALLOCDESCS, 0);
 	}
 
 	brelse(epos->bh);
@@ -2143,7 +2142,7 @@ int8_t udf_next_aext(struct inode *inode, struct extent_position *epos,
 	unsigned int indirections = 0;
 
 	while ((etype = udf_current_aext(inode, epos, eloc, elen, inc)) ==
-	       (EXT_NEXT_EXTENT_ALLOCDECS >> 30)) {
+	       (EXT_NEXT_EXTENT_ALLOCDESCS >> 30)) {
 		udf_pblk_t block;
 
 		if (++indirections > UDF_MAX_INDIR_EXTS) {
diff --git a/fs/udf/lowlevel.c b/fs/udf/lowlevel.c
index 5c7ec121990d..f1094cdcd6cd 100644
--- a/fs/udf/lowlevel.c
+++ b/fs/udf/lowlevel.c
@@ -27,41 +27,38 @@
 
 unsigned int udf_get_last_session(struct super_block *sb)
 {
+	struct cdrom_device_info *cdi = disk_to_cdi(sb->s_bdev->bd_disk);
 	struct cdrom_multisession ms_info;
-	unsigned int vol_desc_start;
-	struct block_device *bdev = sb->s_bdev;
-	int i;
 
-	vol_desc_start = 0;
-	ms_info.addr_format = CDROM_LBA;
-	i = ioctl_by_bdev(bdev, CDROMMULTISESSION, (unsigned long)&ms_info);
+	if (!cdi) {
+		udf_debug("CDROMMULTISESSION not supported.\n");
+		return 0;
+	}
 
-	if (i == 0) {
+	ms_info.addr_format = CDROM_LBA;
+	if (cdrom_multisession(cdi, &ms_info) == 0) {
 		udf_debug("XA disk: %s, vol_desc_start=%d\n",
 			  ms_info.xa_flag ? "yes" : "no", ms_info.addr.lba);
 		if (ms_info.xa_flag) /* necessary for a valid ms_info.addr */
-			vol_desc_start = ms_info.addr.lba;
-	} else {
-		udf_debug("CDROMMULTISESSION not supported: rc=%d\n", i);
+			return ms_info.addr.lba;
 	}
-	return vol_desc_start;
+	return 0;
 }
 
 unsigned long udf_get_last_block(struct super_block *sb)
 {
 	struct block_device *bdev = sb->s_bdev;
+	struct cdrom_device_info *cdi = disk_to_cdi(bdev->bd_disk);
 	unsigned long lblock = 0;
 
 	/*
-	 * ioctl failed or returned obviously bogus value?
+	 * The cdrom layer call failed or returned obviously bogus value?
 	 * Try using the device size...
 	 */
-	if (ioctl_by_bdev(bdev, CDROM_LAST_WRITTEN, (unsigned long) &lblock) ||
-	    lblock == 0)
+	if (!cdi || cdrom_get_last_written(cdi, &lblock) || lblock == 0)
 		lblock = i_size_read(bdev->bd_inode) >> sb->s_blocksize_bits;
 
 	if (lblock)
 		return lblock - 1;
-	else
-		return 0;
+	return 0;
 }
diff --git a/fs/udf/osta_udf.h b/fs/udf/osta_udf.h
index a4da59e38b7f..d5fbfab3ddb6 100644
--- a/fs/udf/osta_udf.h
+++ b/fs/udf/osta_udf.h
@@ -1,10 +1,11 @@
 /*
  * osta_udf.h
  *
- * This file is based on OSTA UDF(tm) 2.50 (April 30, 2003)
+ * This file is based on OSTA UDF(tm) 2.60 (March 1, 2005)
  * http://www.osta.org
  *
- * Copyright (c) 2001-2004  Ben Fennema <bfennema@falcon.csc.calpoly.edu>
+ * Copyright (c) 2001-2004  Ben Fennema
+ * Copyright (c) 2017-2019  Pali Rohár <pali@kernel.org>
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -32,38 +33,57 @@
  * SUCH DAMAGE.
  */
 
+/**
+ * @file
+ * OSTA-UDF defines and structure definitions
+ */
+
 #include "ecma_167.h"
 
 #ifndef _OSTA_UDF_H
 #define _OSTA_UDF_H 1
 
-/* OSTA CS0 Charspec (UDF 2.50 2.1.2) */
+/* OSTA CS0 Charspec (UDF 2.60 2.1.2) */
 #define UDF_CHAR_SET_TYPE		0
 #define UDF_CHAR_SET_INFO		"OSTA Compressed Unicode"
 
-/* Entity Identifier (UDF 2.50 2.1.5) */
-/* Identifiers (UDF 2.50 2.1.5.2) */
+/* Entity Identifier (UDF 2.60 2.1.5) */
+/* Identifiers (UDF 2.60 2.1.5.2) */
+/* Implementation Use Extended Attribute (UDF 2.60 3.3.4.5) */
+/* Virtual Allocation Table (UDF 1.50 2.2.10) */
+/* Logical Volume Extended Information (UDF 1.50 Errata, DCN 5003, 3.3.4.5.1.3) */
+/* OS2EA (UDF 1.50 3.3.4.5.3.1) */
+/* MacUniqueIDTable (UDF 1.50 3.3.4.5.4.3) */
+/* MacResourceFork (UDF 1.50 3.3.4.5.4.4) */
 #define UDF_ID_DEVELOPER		"*Linux UDFFS"
 #define	UDF_ID_COMPLIANT		"*OSTA UDF Compliant"
 #define UDF_ID_LV_INFO			"*UDF LV Info"
 #define UDF_ID_FREE_EA			"*UDF FreeEASpace"
 #define UDF_ID_FREE_APP_EA		"*UDF FreeAppEASpace"
 #define UDF_ID_DVD_CGMS			"*UDF DVD CGMS Info"
+#define UDF_ID_VAT_LVEXTENSION		"*UDF VAT LVExtension"
 #define UDF_ID_OS2_EA			"*UDF OS/2 EA"
 #define UDF_ID_OS2_EA_LENGTH		"*UDF OS/2 EALength"
 #define UDF_ID_MAC_VOLUME		"*UDF Mac VolumeInfo"
 #define UDF_ID_MAC_FINDER		"*UDF Mac FinderInfo"
 #define UDF_ID_MAC_UNIQUE		"*UDF Mac UniqueIDTable"
 #define UDF_ID_MAC_RESOURCE		"*UDF Mac ResourceFork"
+#define UDF_ID_OS400_DIRINFO		"*UDF OS/400 DirInfo"
 #define UDF_ID_VIRTUAL			"*UDF Virtual Partition"
 #define UDF_ID_SPARABLE			"*UDF Sparable Partition"
 #define UDF_ID_ALLOC			"*UDF Virtual Alloc Tbl"
 #define UDF_ID_SPARING			"*UDF Sparing Table"
 #define UDF_ID_METADATA			"*UDF Metadata Partition"
 
-/* Identifier Suffix (UDF 2.50 2.1.5.3) */
-#define IS_DF_HARD_WRITE_PROTECT	0x01
-#define IS_DF_SOFT_WRITE_PROTECT	0x02
+/* Identifier Suffix (UDF 2.60 2.1.5.3) */
+#define DOMAIN_FLAGS_HARD_WRITE_PROTECT	0x01
+#define DOMAIN_FLAGS_SOFT_WRITE_PROTECT	0x02
+
+struct domainIdentSuffix {
+	__le16		UDFRevision;
+	uint8_t		domainFlags;
+	uint8_t		reserved[5];
+} __packed;
 
 struct UDFIdentSuffix {
 	__le16		UDFRevision;
@@ -75,15 +95,15 @@ struct UDFIdentSuffix {
 struct impIdentSuffix {
 	uint8_t		OSClass;
 	uint8_t		OSIdentifier;
-	uint8_t		reserved[6];
+	uint8_t		impUse[6];
 } __packed;
 
 struct appIdentSuffix {
 	uint8_t		impUse[8];
 } __packed;
 
-/* Logical Volume Integrity Descriptor (UDF 2.50 2.2.6) */
-/* Implementation Use (UDF 2.50 2.2.6.4) */
+/* Logical Volume Integrity Descriptor (UDF 2.60 2.2.6) */
+/* Implementation Use (UDF 2.60 2.2.6.4) */
 struct logicalVolIntegrityDescImpUse {
 	struct regid	impIdent;
 	__le32		numFiles;
@@ -94,8 +114,8 @@ struct logicalVolIntegrityDescImpUse {
 	uint8_t		impUse[0];
 } __packed;
 
-/* Implementation Use Volume Descriptor (UDF 2.50 2.2.7) */
-/* Implementation Use (UDF 2.50 2.2.7.2) */
+/* Implementation Use Volume Descriptor (UDF 2.60 2.2.7) */
+/* Implementation Use (UDF 2.60 2.2.7.2) */
 struct impUseVolDescImpUse {
 	struct charspec	LVICharset;
 	dstring		logicalVolIdent[128];
@@ -115,7 +135,7 @@ struct udfPartitionMap2 {
 	__le16		partitionNum;
 } __packed;
 
-/* Virtual Partition Map (UDF 2.50 2.2.8) */
+/* Virtual Partition Map (UDF 2.60 2.2.8) */
 struct virtualPartitionMap {
 	uint8_t		partitionMapType;
 	uint8_t		partitionMapLength;
@@ -126,7 +146,7 @@ struct virtualPartitionMap {
 	uint8_t		reserved2[24];
 } __packed;
 
-/* Sparable Partition Map (UDF 2.50 2.2.9) */
+/* Sparable Partition Map (UDF 2.60 2.2.9) */
 struct sparablePartitionMap {
 	uint8_t partitionMapType;
 	uint8_t partitionMapLength;
@@ -141,7 +161,7 @@ struct sparablePartitionMap {
 	__le32 locSparingTable[4];
 } __packed;
 
-/* Metadata Partition Map (UDF 2.4.0 2.2.10) */
+/* Metadata Partition Map (UDF 2.60 2.2.10) */
 struct metadataPartitionMap {
 	uint8_t		partitionMapType;
 	uint8_t		partitionMapLength;
@@ -160,14 +180,14 @@ struct metadataPartitionMap {
 
 /* Virtual Allocation Table (UDF 1.5 2.2.10) */
 struct virtualAllocationTable15 {
-	__le32		VirtualSector[0];
+	__le32		vatEntry[0];
 	struct regid	vatIdent;
 	__le32		previousVATICBLoc;
 } __packed;
 
 #define ICBTAG_FILE_TYPE_VAT15		0x00U
 
-/* Virtual Allocation Table (UDF 2.50 2.2.11) */
+/* Virtual Allocation Table (UDF 2.60 2.2.11) */
 struct virtualAllocationTable20 {
 	__le16		lengthHeader;
 	__le16		lengthImpUse;
@@ -175,9 +195,9 @@ struct virtualAllocationTable20 {
 	__le32		previousVATICBLoc;
 	__le32		numFiles;
 	__le32		numDirs;
-	__le16		minReadRevision;
-	__le16		minWriteRevision;
-	__le16		maxWriteRevision;
+	__le16		minUDFReadRev;
+	__le16		minUDFWriteRev;
+	__le16		maxUDFWriteRev;
 	__le16		reserved;
 	uint8_t		impUse[0];
 	__le32		vatEntry[0];
@@ -185,7 +205,7 @@ struct virtualAllocationTable20 {
 
 #define ICBTAG_FILE_TYPE_VAT20		0xF8U
 
-/* Sparing Table (UDF 2.50 2.2.12) */
+/* Sparing Table (UDF 2.60 2.2.12) */
 struct sparingEntry {
 	__le32		origLocation;
 	__le32		mappedLocation;
@@ -201,12 +221,12 @@ struct sparingTable {
 			mapEntry[0];
 } __packed;
 
-/* Metadata File (and Metadata Mirror File) (UDF 2.50 2.2.13.1) */
+/* Metadata File (and Metadata Mirror File) (UDF 2.60 2.2.13.1) */
 #define ICBTAG_FILE_TYPE_MAIN		0xFA
 #define ICBTAG_FILE_TYPE_MIRROR		0xFB
 #define ICBTAG_FILE_TYPE_BITMAP		0xFC
 
-/* struct struct long_ad ICB - ADImpUse (UDF 2.50 2.2.4.3) */
+/* struct struct long_ad ICB - ADImpUse (UDF 2.60 2.2.4.3) */
 struct allocDescImpUse {
 	__le16		flags;
 	uint8_t		impUse[4];
@@ -214,17 +234,17 @@ struct allocDescImpUse {
 
 #define AD_IU_EXT_ERASED		0x0001
 
-/* Real-Time Files (UDF 2.50 6.11) */
+/* Real-Time Files (UDF 2.60 6.11) */
 #define ICBTAG_FILE_TYPE_REALTIME	0xF9U
 
-/* Implementation Use Extended Attribute (UDF 2.50 3.3.4.5) */
-/* FreeEASpace (UDF 2.50 3.3.4.5.1.1) */
+/* Implementation Use Extended Attribute (UDF 2.60 3.3.4.5) */
+/* FreeEASpace (UDF 2.60 3.3.4.5.1.1) */
 struct freeEaSpace {
 	__le16		headerChecksum;
 	uint8_t		freeEASpace[0];
 } __packed;
 
-/* DVD Copyright Management Information (UDF 2.50 3.3.4.5.1.2) */
+/* DVD Copyright Management Information (UDF 2.60 3.3.4.5.1.2) */
 struct DVDCopyrightImpUse {
 	__le16		headerChecksum;
 	uint8_t		CGMSInfo;
@@ -232,20 +252,35 @@ struct DVDCopyrightImpUse {
 	uint8_t		protectionSystemInfo[4];
 } __packed;
 
-/* Application Use Extended Attribute (UDF 2.50 3.3.4.6) */
-/* FreeAppEASpace (UDF 2.50 3.3.4.6.1) */
+/* Logical Volume Extended Information (UDF 1.50 Errata, DCN 5003, 3.3.4.5.1.3) */
+struct LVExtensionEA {
+	__le16		headerChecksum;
+	__le64		verificationID;
+	__le32		numFiles;
+	__le32		numDirs;
+	dstring		logicalVolIdent[128];
+} __packed;
+
+/* Application Use Extended Attribute (UDF 2.60 3.3.4.6) */
+/* FreeAppEASpace (UDF 2.60 3.3.4.6.1) */
 struct freeAppEASpace {
 	__le16		headerChecksum;
 	uint8_t		freeEASpace[0];
 } __packed;
 
-/* UDF Defined System Stream (UDF 2.50 3.3.7) */
+/* UDF Defined System Stream (UDF 2.60 3.3.7) */
 #define UDF_ID_UNIQUE_ID		"*UDF Unique ID Mapping Data"
 #define UDF_ID_NON_ALLOC		"*UDF Non-Allocatable Space"
 #define UDF_ID_POWER_CAL		"*UDF Power Cal Table"
 #define UDF_ID_BACKUP			"*UDF Backup"
 
-/* Operating System Identifiers (UDF 2.50 6.3) */
+/* UDF Defined Non-System Streams (UDF 2.60 3.3.8) */
+#define UDF_ID_MAC_RESOURCE_FORK_STREAM	"*UDF Macintosh Resource Fork"
+/* #define UDF_ID_OS2_EA		"*UDF OS/2 EA" */
+#define UDF_ID_NT_ACL			"*UDF NT ACL"
+#define UDF_ID_UNIX_ACL			"*UDF UNIX ACL"
+
+/* Operating System Identifiers (UDF 2.60 6.3) */
 #define UDF_OS_CLASS_UNDEF		0x00U
 #define UDF_OS_CLASS_DOS		0x01U
 #define UDF_OS_CLASS_OS2		0x02U
@@ -270,6 +305,7 @@ struct freeAppEASpace {
 #define UDF_OS_ID_LINUX			0x05U
 #define UDF_OS_ID_MKLINUX		0x06U
 #define UDF_OS_ID_FREEBSD		0x07U
+#define UDF_OS_ID_NETBSD		0x08U
 #define UDF_OS_ID_WIN9X			0x00U
 #define UDF_OS_ID_WINNT			0x00U
 #define UDF_OS_ID_OS400			0x00U
diff --git a/fs/udf/super.c b/fs/udf/super.c
index 8c28e93e9b73..f747bf72edbe 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -767,20 +767,20 @@ static int udf_check_vsd(struct super_block *sb)
 static int udf_verify_domain_identifier(struct super_block *sb,
 					struct regid *ident, char *dname)
 {
-	struct domainEntityIDSuffix *suffix;
+	struct domainIdentSuffix *suffix;
 
 	if (memcmp(ident->ident, UDF_ID_COMPLIANT, strlen(UDF_ID_COMPLIANT))) {
 		udf_warn(sb, "Not OSTA UDF compliant %s descriptor.\n", dname);
 		goto force_ro;
 	}
-	if (ident->flags & (1 << ENTITYID_FLAGS_DIRTY)) {
+	if (ident->flags & ENTITYID_FLAGS_DIRTY) {
 		udf_warn(sb, "Possibly not OSTA UDF compliant %s descriptor.\n",
 			 dname);
 		goto force_ro;
 	}
-	suffix = (struct domainEntityIDSuffix *)ident->identSuffix;
-	if (suffix->flags & (1 << ENTITYIDSUFFIX_FLAGS_HARDWRITEPROTECT) ||
-	    suffix->flags & (1 << ENTITYIDSUFFIX_FLAGS_SOFTWRITEPROTECT)) {
+	suffix = (struct domainIdentSuffix *)ident->identSuffix;
+	if ((suffix->domainFlags & DOMAIN_FLAGS_HARD_WRITE_PROTECT) ||
+	    (suffix->domainFlags & DOMAIN_FLAGS_SOFT_WRITE_PROTECT)) {
 		if (!sb_rdonly(sb)) {
 			udf_warn(sb, "Descriptor for %s marked write protected."
 				 " Forcing read only mount.\n", dname);
@@ -1035,7 +1035,6 @@ static int check_partition_desc(struct super_block *sb,
 	switch (le32_to_cpu(p->accessType)) {
 	case PD_ACCESS_TYPE_READ_ONLY:
 	case PD_ACCESS_TYPE_WRITE_ONCE:
-	case PD_ACCESS_TYPE_REWRITABLE:
 	case PD_ACCESS_TYPE_NONE:
 		goto force_ro;
 	}
@@ -1063,7 +1062,8 @@ static int check_partition_desc(struct super_block *sb,
 		goto force_ro;
 
 	if (map->s_partition_type == UDF_VIRTUAL_MAP15 ||
-	    map->s_partition_type == UDF_VIRTUAL_MAP20)
+	    map->s_partition_type == UDF_VIRTUAL_MAP20 ||
+	    map->s_partition_type == UDF_METADATA_MAP25)
 		goto force_ro;
 
 	return 0;
@@ -2402,6 +2402,10 @@ static int udf_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_blocks = sbi->s_partmaps[sbi->s_partition].s_partition_len;
 	buf->f_bfree = udf_count_free(sb);
 	buf->f_bavail = buf->f_bfree;
+	/*
+	 * Let's pretend each free block is also a free 'inode' since UDF does
+	 * not have separate preallocated table of inodes.
+	 */
 	buf->f_files = (lvidiu != NULL ? (le32_to_cpu(lvidiu->numFiles) +
 					  le32_to_cpu(lvidiu->numDirs)) : 0)
 			+ buf->f_bfree;
@@ -2492,17 +2496,29 @@ static unsigned int udf_count_free_table(struct super_block *sb,
 static unsigned int udf_count_free(struct super_block *sb)
 {
 	unsigned int accum = 0;
-	struct udf_sb_info *sbi;
+	struct udf_sb_info *sbi = UDF_SB(sb);
 	struct udf_part_map *map;
+	unsigned int part = sbi->s_partition;
+	int ptype = sbi->s_partmaps[part].s_partition_type;
+
+	if (ptype == UDF_METADATA_MAP25) {
+		part = sbi->s_partmaps[part].s_type_specific.s_metadata.
+							s_phys_partition_ref;
+	} else if (ptype == UDF_VIRTUAL_MAP15 || ptype == UDF_VIRTUAL_MAP20) {
+		/*
+		 * Filesystems with VAT are append-only and we cannot write to
+ 		 * them. Let's just report 0 here.
+		 */
+		return 0;
+	}
 
-	sbi = UDF_SB(sb);
 	if (sbi->s_lvid_bh) {
 		struct logicalVolIntegrityDesc *lvid =
 			(struct logicalVolIntegrityDesc *)
 			sbi->s_lvid_bh->b_data;
-		if (le32_to_cpu(lvid->numOfPartitions) > sbi->s_partition) {
+		if (le32_to_cpu(lvid->numOfPartitions) > part) {
 			accum = le32_to_cpu(
-					lvid->freeSpaceTable[sbi->s_partition]);
+					lvid->freeSpaceTable[part]);
 			if (accum == 0xFFFFFFFF)
 				accum = 0;
 		}
@@ -2511,7 +2527,7 @@ static unsigned int udf_count_free(struct super_block *sb)
 	if (accum)
 		return accum;
 
-	map = &sbi->s_partmaps[sbi->s_partition];
+	map = &sbi->s_partmaps[part];
 	if (map->s_partition_flags & UDF_PART_FLAG_UNALLOC_BITMAP) {
 		accum += udf_count_free_bitmap(sb,
 					       map->s_uspace.s_bitmap);
diff --git a/fs/udf/truncate.c b/fs/udf/truncate.c
index 63a47f1e1d52..532cda99644e 100644
--- a/fs/udf/truncate.c
+++ b/fs/udf/truncate.c
@@ -241,7 +241,7 @@ int udf_truncate_extents(struct inode *inode)
 
 	while ((etype = udf_current_aext(inode, &epos, &eloc,
 					 &elen, 0)) != -1) {
-		if (etype == (EXT_NEXT_EXTENT_ALLOCDECS >> 30)) {
+		if (etype == (EXT_NEXT_EXTENT_ALLOCDESCS >> 30)) {
 			udf_write_aext(inode, &epos, &neloc, nelen, 0);
 			if (indirect_ext_len) {
 				/* We managed to free all extents in the
diff --git a/fs/udf/udf_sb.h b/fs/udf/udf_sb.h
index 3d83be54c474..758efe557a19 100644
--- a/fs/udf/udf_sb.h
+++ b/fs/udf/udf_sb.h
@@ -83,7 +83,7 @@ struct udf_virtual_data {
 struct udf_bitmap {
 	__u32			s_extPosition;
 	int			s_nr_groups;
-	struct buffer_head 	*s_block_bitmap[0];
+	struct buffer_head	*s_block_bitmap[];
 };
 
 struct udf_part_map {
diff --git a/fs/unicode/.gitignore b/fs/unicode/.gitignore
index 0381e2221480..9b2467e77b2d 100644
--- a/fs/unicode/.gitignore
+++ b/fs/unicode/.gitignore
@@ -1,2 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
 mkutf8data
 utf8data.h
diff --git a/fs/unicode/Makefile b/fs/unicode/Makefile
index d46e9baee285..b88aecc86550 100644
--- a/fs/unicode/Makefile
+++ b/fs/unicode/Makefile
@@ -35,4 +35,4 @@ $(obj)/utf8data.h: $(src)/utf8data.h_shipped FORCE
 endif
 
 targets += utf8data.h
-hostprogs-y += mkutf8data
+hostprogs += mkutf8data
diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index f9fd18670e22..e39fdec8a0b0 100644
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -314,8 +314,11 @@ static inline bool userfaultfd_must_wait(struct userfaultfd_ctx *ctx,
 	if (!pmd_present(_pmd))
 		goto out;
 
-	if (pmd_trans_huge(_pmd))
+	if (pmd_trans_huge(_pmd)) {
+		if (!pmd_write(_pmd) && (reason & VM_UFFD_WP))
+			ret = true;
 		goto out;
+	}
 
 	/*
 	 * the pmd is stable (as in !pmd_trans_unstable) so we can re-read it
@@ -328,12 +331,38 @@ static inline bool userfaultfd_must_wait(struct userfaultfd_ctx *ctx,
 	 */
 	if (pte_none(*pte))
 		ret = true;
+	if (!pte_write(*pte) && (reason & VM_UFFD_WP))
+		ret = true;
 	pte_unmap(pte);
 
 out:
 	return ret;
 }
 
+/* Should pair with userfaultfd_signal_pending() */
+static inline long userfaultfd_get_blocking_state(unsigned int flags)
+{
+	if (flags & FAULT_FLAG_INTERRUPTIBLE)
+		return TASK_INTERRUPTIBLE;
+
+	if (flags & FAULT_FLAG_KILLABLE)
+		return TASK_KILLABLE;
+
+	return TASK_UNINTERRUPTIBLE;
+}
+
+/* Should pair with userfaultfd_get_blocking_state() */
+static inline bool userfaultfd_signal_pending(unsigned int flags)
+{
+	if (flags & FAULT_FLAG_INTERRUPTIBLE)
+		return signal_pending(current);
+
+	if (flags & FAULT_FLAG_KILLABLE)
+		return fatal_signal_pending(current);
+
+	return false;
+}
+
 /*
  * The locking rules involved in returning VM_FAULT_RETRY depending on
  * FAULT_FLAG_ALLOW_RETRY, FAULT_FLAG_RETRY_NOWAIT and
@@ -355,7 +384,7 @@ vm_fault_t handle_userfault(struct vm_fault *vmf, unsigned long reason)
 	struct userfaultfd_ctx *ctx;
 	struct userfaultfd_wait_queue uwq;
 	vm_fault_t ret = VM_FAULT_SIGBUS;
-	bool must_wait, return_to_userland;
+	bool must_wait;
 	long blocking_state;
 
 	/*
@@ -462,11 +491,7 @@ vm_fault_t handle_userfault(struct vm_fault *vmf, unsigned long reason)
 	uwq.ctx = ctx;
 	uwq.waken = false;
 
-	return_to_userland =
-		(vmf->flags & (FAULT_FLAG_USER|FAULT_FLAG_KILLABLE)) ==
-		(FAULT_FLAG_USER|FAULT_FLAG_KILLABLE);
-	blocking_state = return_to_userland ? TASK_INTERRUPTIBLE :
-			 TASK_KILLABLE;
+	blocking_state = userfaultfd_get_blocking_state(vmf->flags);
 
 	spin_lock_irq(&ctx->fault_pending_wqh.lock);
 	/*
@@ -492,8 +517,7 @@ vm_fault_t handle_userfault(struct vm_fault *vmf, unsigned long reason)
 	up_read(&mm->mmap_sem);
 
 	if (likely(must_wait && !READ_ONCE(ctx->released) &&
-		   (return_to_userland ? !signal_pending(current) :
-		    !fatal_signal_pending(current)))) {
+		   !userfaultfd_signal_pending(vmf->flags))) {
 		wake_up_poll(&ctx->fd_wqh, EPOLLIN);
 		schedule();
 		ret |= VM_FAULT_MAJOR;
@@ -515,8 +539,7 @@ vm_fault_t handle_userfault(struct vm_fault *vmf, unsigned long reason)
 			set_current_state(blocking_state);
 			if (READ_ONCE(uwq.waken) ||
 			    READ_ONCE(ctx->released) ||
-			    (return_to_userland ? signal_pending(current) :
-			     fatal_signal_pending(current)))
+			    userfaultfd_signal_pending(vmf->flags))
 				break;
 			schedule();
 		}
@@ -524,30 +547,6 @@ vm_fault_t handle_userfault(struct vm_fault *vmf, unsigned long reason)
 
 	__set_current_state(TASK_RUNNING);
 
-	if (return_to_userland) {
-		if (signal_pending(current) &&
-		    !fatal_signal_pending(current)) {
-			/*
-			 * If we got a SIGSTOP or SIGCONT and this is
-			 * a normal userland page fault, just let
-			 * userland return so the signal will be
-			 * handled and gdb debugging works.  The page
-			 * fault code immediately after we return from
-			 * this function is going to release the
-			 * mmap_sem and it's not depending on it
-			 * (unlike gup would if we were not to return
-			 * VM_FAULT_RETRY).
-			 *
-			 * If a fatal signal is pending we still take
-			 * the streamlined VM_FAULT_RETRY failure path
-			 * and there's no need to retake the mmap_sem
-			 * in such case.
-			 */
-			down_read(&mm->mmap_sem);
-			ret = VM_FAULT_NOPAGE;
-		}
-	}
-
 	/*
 	 * Here we race with the list_del; list_add in
 	 * userfaultfd_ctx_read(), however because we don't ever run
@@ -1293,10 +1292,13 @@ static __always_inline int validate_range(struct mm_struct *mm,
 	return 0;
 }
 
-static inline bool vma_can_userfault(struct vm_area_struct *vma)
+static inline bool vma_can_userfault(struct vm_area_struct *vma,
+				     unsigned long vm_flags)
 {
-	return vma_is_anonymous(vma) || is_vm_hugetlb_page(vma) ||
-		vma_is_shmem(vma);
+	/* FIXME: add WP support to hugetlbfs and shmem */
+	return vma_is_anonymous(vma) ||
+		((is_vm_hugetlb_page(vma) || vma_is_shmem(vma)) &&
+		 !(vm_flags & VM_UFFD_WP));
 }
 
 static int userfaultfd_register(struct userfaultfd_ctx *ctx,
@@ -1328,15 +1330,8 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx,
 	vm_flags = 0;
 	if (uffdio_register.mode & UFFDIO_REGISTER_MODE_MISSING)
 		vm_flags |= VM_UFFD_MISSING;
-	if (uffdio_register.mode & UFFDIO_REGISTER_MODE_WP) {
+	if (uffdio_register.mode & UFFDIO_REGISTER_MODE_WP)
 		vm_flags |= VM_UFFD_WP;
-		/*
-		 * FIXME: remove the below error constraint by
-		 * implementing the wprotect tracking mode.
-		 */
-		ret = -EINVAL;
-		goto out;
-	}
 
 	ret = validate_range(mm, &uffdio_register.range.start,
 			     uffdio_register.range.len);
@@ -1386,7 +1381,7 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx,
 
 		/* check not compatible vmas */
 		ret = -EINVAL;
-		if (!vma_can_userfault(cur))
+		if (!vma_can_userfault(cur, vm_flags))
 			goto out_unlock;
 
 		/*
@@ -1414,6 +1409,8 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx,
 			if (end & (vma_hpagesize - 1))
 				goto out_unlock;
 		}
+		if ((vm_flags & VM_UFFD_WP) && !(cur->vm_flags & VM_MAYWRITE))
+			goto out_unlock;
 
 		/*
 		 * Check that this vma isn't already owned by a
@@ -1443,7 +1440,7 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx,
 	do {
 		cond_resched();
 
-		BUG_ON(!vma_can_userfault(vma));
+		BUG_ON(!vma_can_userfault(vma, vm_flags));
 		BUG_ON(vma->vm_userfaultfd_ctx.ctx &&
 		       vma->vm_userfaultfd_ctx.ctx != ctx);
 		WARN_ON(!(vma->vm_flags & VM_MAYWRITE));
@@ -1460,7 +1457,8 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx,
 			start = vma->vm_start;
 		vma_end = min(end, vma->vm_end);
 
-		new_flags = (vma->vm_flags & ~vm_flags) | vm_flags;
+		new_flags = (vma->vm_flags &
+			     ~(VM_UFFD_MISSING|VM_UFFD_WP)) | vm_flags;
 		prev = vma_merge(mm, prev, start, vma_end, new_flags,
 				 vma->anon_vma, vma->vm_file, vma->vm_pgoff,
 				 vma_policy(vma),
@@ -1497,14 +1495,24 @@ out_unlock:
 	up_write(&mm->mmap_sem);
 	mmput(mm);
 	if (!ret) {
+		__u64 ioctls_out;
+
+		ioctls_out = basic_ioctls ? UFFD_API_RANGE_IOCTLS_BASIC :
+		    UFFD_API_RANGE_IOCTLS;
+
+		/*
+		 * Declare the WP ioctl only if the WP mode is
+		 * specified and all checks passed with the range
+		 */
+		if (!(uffdio_register.mode & UFFDIO_REGISTER_MODE_WP))
+			ioctls_out &= ~((__u64)1 << _UFFDIO_WRITEPROTECT);
+
 		/*
 		 * Now that we scanned all vmas we can already tell
 		 * userland which ioctls methods are guaranteed to
 		 * succeed on this range.
 		 */
-		if (put_user(basic_ioctls ? UFFD_API_RANGE_IOCTLS_BASIC :
-			     UFFD_API_RANGE_IOCTLS,
-			     &user_uffdio_register->ioctls))
+		if (put_user(ioctls_out, &user_uffdio_register->ioctls))
 			ret = -EFAULT;
 	}
 out:
@@ -1580,7 +1588,7 @@ static int userfaultfd_unregister(struct userfaultfd_ctx *ctx,
 		 * provides for more strict behavior to notice
 		 * unregistration errors.
 		 */
-		if (!vma_can_userfault(cur))
+		if (!vma_can_userfault(cur, cur->vm_flags))
 			goto out_unlock;
 
 		found = true;
@@ -1594,7 +1602,7 @@ static int userfaultfd_unregister(struct userfaultfd_ctx *ctx,
 	do {
 		cond_resched();
 
-		BUG_ON(!vma_can_userfault(vma));
+		BUG_ON(!vma_can_userfault(vma, vma->vm_flags));
 
 		/*
 		 * Nothing to do: this vma is already registered into this
@@ -1729,11 +1737,12 @@ static int userfaultfd_copy(struct userfaultfd_ctx *ctx,
 	ret = -EINVAL;
 	if (uffdio_copy.src + uffdio_copy.len <= uffdio_copy.src)
 		goto out;
-	if (uffdio_copy.mode & ~UFFDIO_COPY_MODE_DONTWAKE)
+	if (uffdio_copy.mode & ~(UFFDIO_COPY_MODE_DONTWAKE|UFFDIO_COPY_MODE_WP))
 		goto out;
 	if (mmget_not_zero(ctx->mm)) {
 		ret = mcopy_atomic(ctx->mm, uffdio_copy.dst, uffdio_copy.src,
-				   uffdio_copy.len, &ctx->mmap_changing);
+				   uffdio_copy.len, &ctx->mmap_changing,
+				   uffdio_copy.mode);
 		mmput(ctx->mm);
 	} else {
 		return -ESRCH;
@@ -1806,6 +1815,53 @@ out:
 	return ret;
 }
 
+static int userfaultfd_writeprotect(struct userfaultfd_ctx *ctx,
+				    unsigned long arg)
+{
+	int ret;
+	struct uffdio_writeprotect uffdio_wp;
+	struct uffdio_writeprotect __user *user_uffdio_wp;
+	struct userfaultfd_wake_range range;
+	bool mode_wp, mode_dontwake;
+
+	if (READ_ONCE(ctx->mmap_changing))
+		return -EAGAIN;
+
+	user_uffdio_wp = (struct uffdio_writeprotect __user *) arg;
+
+	if (copy_from_user(&uffdio_wp, user_uffdio_wp,
+			   sizeof(struct uffdio_writeprotect)))
+		return -EFAULT;
+
+	ret = validate_range(ctx->mm, &uffdio_wp.range.start,
+			     uffdio_wp.range.len);
+	if (ret)
+		return ret;
+
+	if (uffdio_wp.mode & ~(UFFDIO_WRITEPROTECT_MODE_DONTWAKE |
+			       UFFDIO_WRITEPROTECT_MODE_WP))
+		return -EINVAL;
+
+	mode_wp = uffdio_wp.mode & UFFDIO_WRITEPROTECT_MODE_WP;
+	mode_dontwake = uffdio_wp.mode & UFFDIO_WRITEPROTECT_MODE_DONTWAKE;
+
+	if (mode_wp && mode_dontwake)
+		return -EINVAL;
+
+	ret = mwriteprotect_range(ctx->mm, uffdio_wp.range.start,
+				  uffdio_wp.range.len, mode_wp,
+				  &ctx->mmap_changing);
+	if (ret)
+		return ret;
+
+	if (!mode_wp && !mode_dontwake) {
+		range.start = uffdio_wp.range.start;
+		range.len = uffdio_wp.range.len;
+		wake_userfault(ctx, &range);
+	}
+	return ret;
+}
+
 static inline unsigned int uffd_ctx_features(__u64 user_features)
 {
 	/*
@@ -1834,13 +1890,12 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
 	if (copy_from_user(&uffdio_api, buf, sizeof(uffdio_api)))
 		goto out;
 	features = uffdio_api.features;
-	if (uffdio_api.api != UFFD_API || (features & ~UFFD_API_FEATURES)) {
-		memset(&uffdio_api, 0, sizeof(uffdio_api));
-		if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api)))
-			goto out;
-		ret = -EINVAL;
-		goto out;
-	}
+	ret = -EINVAL;
+	if (uffdio_api.api != UFFD_API || (features & ~UFFD_API_FEATURES))
+		goto err_out;
+	ret = -EPERM;
+	if ((features & UFFD_FEATURE_EVENT_FORK) && !capable(CAP_SYS_PTRACE))
+		goto err_out;
 	/* report all available features and ioctls to userland */
 	uffdio_api.features = UFFD_API_FEATURES;
 	uffdio_api.ioctls = UFFD_API_IOCTLS;
@@ -1853,6 +1908,11 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
 	ret = 0;
 out:
 	return ret;
+err_out:
+	memset(&uffdio_api, 0, sizeof(uffdio_api));
+	if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api)))
+		ret = -EFAULT;
+	goto out;
 }
 
 static long userfaultfd_ioctl(struct file *file, unsigned cmd,
@@ -1883,6 +1943,9 @@ static long userfaultfd_ioctl(struct file *file, unsigned cmd,
 	case UFFDIO_ZEROPAGE:
 		ret = userfaultfd_zeropage(ctx, arg);
 		break;
+	case UFFDIO_WRITEPROTECT:
+		ret = userfaultfd_writeprotect(ctx, arg);
+		break;
 	}
 	return ret;
 }
@@ -1923,7 +1986,7 @@ static const struct file_operations userfaultfd_fops = {
 	.poll		= userfaultfd_poll,
 	.read		= userfaultfd_read,
 	.unlocked_ioctl = userfaultfd_ioctl,
-	.compat_ioctl	= userfaultfd_ioctl,
+	.compat_ioctl	= compat_ptr_ioctl,
 	.llseek		= noop_llseek,
 };
 
diff --git a/fs/utimes.c b/fs/utimes.c
index 1ba3f7883870..b7b927502d6e 100644
--- a/fs/utimes.c
+++ b/fs/utimes.c
@@ -36,14 +36,14 @@ static int utimes_common(const struct path *path, struct timespec64 *times)
 		if (times[0].tv_nsec == UTIME_OMIT)
 			newattrs.ia_valid &= ~ATTR_ATIME;
 		else if (times[0].tv_nsec != UTIME_NOW) {
-			newattrs.ia_atime = timestamp_truncate(times[0], inode);
+			newattrs.ia_atime = times[0];
 			newattrs.ia_valid |= ATTR_ATIME_SET;
 		}
 
 		if (times[1].tv_nsec == UTIME_OMIT)
 			newattrs.ia_valid &= ~ATTR_MTIME;
 		else if (times[1].tv_nsec != UTIME_NOW) {
-			newattrs.ia_mtime = timestamp_truncate(times[1], inode);
+			newattrs.ia_mtime = times[1];
 			newattrs.ia_valid |= ATTR_MTIME_SET;
 		}
 		/*
@@ -95,13 +95,13 @@ long do_utimes(int dfd, const char __user *filename, struct timespec64 *times,
 		goto out;
 	}
 
-	if (flags & ~AT_SYMLINK_NOFOLLOW)
+	if (flags & ~(AT_SYMLINK_NOFOLLOW | AT_EMPTY_PATH))
 		goto out;
 
 	if (filename == NULL && dfd != AT_FDCWD) {
 		struct fd f;
 
-		if (flags & AT_SYMLINK_NOFOLLOW)
+		if (flags)
 			goto out;
 
 		f = fdget(dfd);
@@ -117,6 +117,8 @@ long do_utimes(int dfd, const char __user *filename, struct timespec64 *times,
 
 		if (!(flags & AT_SYMLINK_NOFOLLOW))
 			lookup_flags |= LOOKUP_FOLLOW;
+		if (flags & AT_EMPTY_PATH)
+			lookup_flags |= LOOKUP_EMPTY;
 retry:
 		error = user_path_at(dfd, filename, lookup_flags, &path);
 		if (error)
@@ -161,9 +163,9 @@ SYSCALL_DEFINE4(utimensat, int, dfd, const char __user *, filename,
  * utimensat() instead.
  */
 static long do_futimesat(int dfd, const char __user *filename,
-			 struct timeval __user *utimes)
+			 struct __kernel_old_timeval __user *utimes)
 {
-	struct timeval times[2];
+	struct __kernel_old_timeval times[2];
 	struct timespec64 tstimes[2];
 
 	if (utimes) {
@@ -190,13 +192,13 @@ static long do_futimesat(int dfd, const char __user *filename,
 
 
 SYSCALL_DEFINE3(futimesat, int, dfd, const char __user *, filename,
-		struct timeval __user *, utimes)
+		struct __kernel_old_timeval __user *, utimes)
 {
 	return do_futimesat(dfd, filename, utimes);
 }
 
 SYSCALL_DEFINE2(utimes, char __user *, filename,
-		struct timeval __user *, utimes)
+		struct __kernel_old_timeval __user *, utimes)
 {
 	return do_futimesat(AT_FDCWD, filename, utimes);
 }
diff --git a/fs/vboxsf/Kconfig b/fs/vboxsf/Kconfig
new file mode 100644
index 000000000000..b84586ae08b3
--- /dev/null
+++ b/fs/vboxsf/Kconfig
@@ -0,0 +1,10 @@
+config VBOXSF_FS
+	tristate "VirtualBox guest shared folder (vboxsf) support"
+	depends on X86 && VBOXGUEST
+	select NLS
+	help
+	  VirtualBox hosts can share folders with guests, this driver
+	  implements the Linux-guest side of this allowing folders exported
+	  by the host to be mounted under Linux.
+
+	  If you want to use shared folders in VirtualBox guests, answer Y or M.
diff --git a/fs/vboxsf/Makefile b/fs/vboxsf/Makefile
new file mode 100644
index 000000000000..9e4328e79623
--- /dev/null
+++ b/fs/vboxsf/Makefile
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: MIT
+
+obj-$(CONFIG_VBOXSF_FS) += vboxsf.o
+
+vboxsf-y := dir.o file.o utils.o vboxsf_wrappers.o super.o
diff --git a/fs/vboxsf/dir.c b/fs/vboxsf/dir.c
new file mode 100644
index 000000000000..dd147b490982
--- /dev/null
+++ b/fs/vboxsf/dir.c
@@ -0,0 +1,427 @@
+// SPDX-License-Identifier: MIT
+/*
+ * VirtualBox Guest Shared Folders support: Directory inode and file operations
+ *
+ * Copyright (C) 2006-2018 Oracle Corporation
+ */
+
+#include <linux/namei.h>
+#include <linux/vbox_utils.h>
+#include "vfsmod.h"
+
+static int vboxsf_dir_open(struct inode *inode, struct file *file)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(inode->i_sb);
+	struct shfl_createparms params = {};
+	struct vboxsf_dir_info *sf_d;
+	int err;
+
+	sf_d = vboxsf_dir_info_alloc();
+	if (!sf_d)
+		return -ENOMEM;
+
+	params.handle = SHFL_HANDLE_NIL;
+	params.create_flags = SHFL_CF_DIRECTORY | SHFL_CF_ACT_OPEN_IF_EXISTS |
+			      SHFL_CF_ACT_FAIL_IF_NEW | SHFL_CF_ACCESS_READ;
+
+	err = vboxsf_create_at_dentry(file_dentry(file), &params);
+	if (err)
+		goto err_free_dir_info;
+
+	if (params.result != SHFL_FILE_EXISTS) {
+		err = -ENOENT;
+		goto err_close;
+	}
+
+	err = vboxsf_dir_read_all(sbi, sf_d, params.handle);
+	if (err)
+		goto err_close;
+
+	vboxsf_close(sbi->root, params.handle);
+	file->private_data = sf_d;
+	return 0;
+
+err_close:
+	vboxsf_close(sbi->root, params.handle);
+err_free_dir_info:
+	vboxsf_dir_info_free(sf_d);
+	return err;
+}
+
+static int vboxsf_dir_release(struct inode *inode, struct file *file)
+{
+	if (file->private_data)
+		vboxsf_dir_info_free(file->private_data);
+
+	return 0;
+}
+
+static unsigned int vboxsf_get_d_type(u32 mode)
+{
+	unsigned int d_type;
+
+	switch (mode & SHFL_TYPE_MASK) {
+	case SHFL_TYPE_FIFO:
+		d_type = DT_FIFO;
+		break;
+	case SHFL_TYPE_DEV_CHAR:
+		d_type = DT_CHR;
+		break;
+	case SHFL_TYPE_DIRECTORY:
+		d_type = DT_DIR;
+		break;
+	case SHFL_TYPE_DEV_BLOCK:
+		d_type = DT_BLK;
+		break;
+	case SHFL_TYPE_FILE:
+		d_type = DT_REG;
+		break;
+	case SHFL_TYPE_SYMLINK:
+		d_type = DT_LNK;
+		break;
+	case SHFL_TYPE_SOCKET:
+		d_type = DT_SOCK;
+		break;
+	case SHFL_TYPE_WHITEOUT:
+		d_type = DT_WHT;
+		break;
+	default:
+		d_type = DT_UNKNOWN;
+		break;
+	}
+	return d_type;
+}
+
+static bool vboxsf_dir_emit(struct file *dir, struct dir_context *ctx)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(file_inode(dir)->i_sb);
+	struct vboxsf_dir_info *sf_d = dir->private_data;
+	struct shfl_dirinfo *info;
+	struct vboxsf_dir_buf *b;
+	unsigned int d_type;
+	loff_t i, cur = 0;
+	ino_t fake_ino;
+	void *end;
+	int err;
+
+	list_for_each_entry(b, &sf_d->info_list, head) {
+try_next_entry:
+		if (ctx->pos >= cur + b->entries) {
+			cur += b->entries;
+			continue;
+		}
+
+		/*
+		 * Note the vboxsf_dir_info objects we are iterating over here
+		 * are variable sized, so the info pointer may end up being
+		 * unaligned. This is how we get the data from the host.
+		 * Since vboxsf is only supported on x86 machines this is not
+		 * a problem.
+		 */
+		for (i = 0, info = b->buf; i < ctx->pos - cur; i++) {
+			end = &info->name.string.utf8[info->name.size];
+			/* Only happens if the host gives us corrupt data */
+			if (WARN_ON(end > (b->buf + b->used)))
+				return false;
+			info = end;
+		}
+
+		end = &info->name.string.utf8[info->name.size];
+		if (WARN_ON(end > (b->buf + b->used)))
+			return false;
+
+		/* Info now points to the right entry, emit it. */
+		d_type = vboxsf_get_d_type(info->info.attr.mode);
+
+		/*
+		 * On 32 bit systems pos is 64 signed, while ino is 32 bit
+		 * unsigned so fake_ino may overflow, check for this.
+		 */
+		if ((ino_t)(ctx->pos + 1) != (u64)(ctx->pos + 1)) {
+			vbg_err("vboxsf: fake ino overflow, truncating dir\n");
+			return false;
+		}
+		fake_ino = ctx->pos + 1;
+
+		if (sbi->nls) {
+			char d_name[NAME_MAX];
+
+			err = vboxsf_nlscpy(sbi, d_name, NAME_MAX,
+					    info->name.string.utf8,
+					    info->name.length);
+			if (err) {
+				/* skip erroneous entry and proceed */
+				ctx->pos += 1;
+				goto try_next_entry;
+			}
+
+			return dir_emit(ctx, d_name, strlen(d_name),
+					fake_ino, d_type);
+		}
+
+		return dir_emit(ctx, info->name.string.utf8, info->name.length,
+				fake_ino, d_type);
+	}
+
+	return false;
+}
+
+static int vboxsf_dir_iterate(struct file *dir, struct dir_context *ctx)
+{
+	bool emitted;
+
+	do {
+		emitted = vboxsf_dir_emit(dir, ctx);
+		if (emitted)
+			ctx->pos += 1;
+	} while (emitted);
+
+	return 0;
+}
+
+const struct file_operations vboxsf_dir_fops = {
+	.open = vboxsf_dir_open,
+	.iterate = vboxsf_dir_iterate,
+	.release = vboxsf_dir_release,
+	.read = generic_read_dir,
+	.llseek = generic_file_llseek,
+};
+
+/*
+ * This is called during name resolution/lookup to check if the @dentry in
+ * the cache is still valid. the job is handled by vboxsf_inode_revalidate.
+ */
+static int vboxsf_dentry_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	if (flags & LOOKUP_RCU)
+		return -ECHILD;
+
+	if (d_really_is_positive(dentry))
+		return vboxsf_inode_revalidate(dentry) == 0;
+	else
+		return vboxsf_stat_dentry(dentry, NULL) == -ENOENT;
+}
+
+const struct dentry_operations vboxsf_dentry_ops = {
+	.d_revalidate = vboxsf_dentry_revalidate
+};
+
+/* iops */
+
+static struct dentry *vboxsf_dir_lookup(struct inode *parent,
+					struct dentry *dentry,
+					unsigned int flags)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(parent->i_sb);
+	struct shfl_fsobjinfo fsinfo;
+	struct inode *inode;
+	int err;
+
+	dentry->d_time = jiffies;
+
+	err = vboxsf_stat_dentry(dentry, &fsinfo);
+	if (err) {
+		inode = (err == -ENOENT) ? NULL : ERR_PTR(err);
+	} else {
+		inode = vboxsf_new_inode(parent->i_sb);
+		if (!IS_ERR(inode))
+			vboxsf_init_inode(sbi, inode, &fsinfo);
+	}
+
+	return d_splice_alias(inode, dentry);
+}
+
+static int vboxsf_dir_instantiate(struct inode *parent, struct dentry *dentry,
+				  struct shfl_fsobjinfo *info)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(parent->i_sb);
+	struct vboxsf_inode *sf_i;
+	struct inode *inode;
+
+	inode = vboxsf_new_inode(parent->i_sb);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
+
+	sf_i = VBOXSF_I(inode);
+	/* The host may have given us different attr then requested */
+	sf_i->force_restat = 1;
+	vboxsf_init_inode(sbi, inode, info);
+
+	d_instantiate(dentry, inode);
+
+	return 0;
+}
+
+static int vboxsf_dir_create(struct inode *parent, struct dentry *dentry,
+			     umode_t mode, int is_dir)
+{
+	struct vboxsf_inode *sf_parent_i = VBOXSF_I(parent);
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(parent->i_sb);
+	struct shfl_createparms params = {};
+	int err;
+
+	params.handle = SHFL_HANDLE_NIL;
+	params.create_flags = SHFL_CF_ACT_CREATE_IF_NEW |
+			      SHFL_CF_ACT_FAIL_IF_EXISTS |
+			      SHFL_CF_ACCESS_READWRITE |
+			      (is_dir ? SHFL_CF_DIRECTORY : 0);
+	params.info.attr.mode = (mode & 0777) |
+				(is_dir ? SHFL_TYPE_DIRECTORY : SHFL_TYPE_FILE);
+	params.info.attr.additional = SHFLFSOBJATTRADD_NOTHING;
+
+	err = vboxsf_create_at_dentry(dentry, &params);
+	if (err)
+		return err;
+
+	if (params.result != SHFL_FILE_CREATED)
+		return -EPERM;
+
+	vboxsf_close(sbi->root, params.handle);
+
+	err = vboxsf_dir_instantiate(parent, dentry, &params.info);
+	if (err)
+		return err;
+
+	/* parent directory access/change time changed */
+	sf_parent_i->force_restat = 1;
+
+	return 0;
+}
+
+static int vboxsf_dir_mkfile(struct inode *parent, struct dentry *dentry,
+			     umode_t mode, bool excl)
+{
+	return vboxsf_dir_create(parent, dentry, mode, 0);
+}
+
+static int vboxsf_dir_mkdir(struct inode *parent, struct dentry *dentry,
+			    umode_t mode)
+{
+	return vboxsf_dir_create(parent, dentry, mode, 1);
+}
+
+static int vboxsf_dir_unlink(struct inode *parent, struct dentry *dentry)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(parent->i_sb);
+	struct vboxsf_inode *sf_parent_i = VBOXSF_I(parent);
+	struct inode *inode = d_inode(dentry);
+	struct shfl_string *path;
+	u32 flags;
+	int err;
+
+	if (S_ISDIR(inode->i_mode))
+		flags = SHFL_REMOVE_DIR;
+	else
+		flags = SHFL_REMOVE_FILE;
+
+	if (S_ISLNK(inode->i_mode))
+		flags |= SHFL_REMOVE_SYMLINK;
+
+	path = vboxsf_path_from_dentry(sbi, dentry);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	err = vboxsf_remove(sbi->root, path, flags);
+	__putname(path);
+	if (err)
+		return err;
+
+	/* parent directory access/change time changed */
+	sf_parent_i->force_restat = 1;
+
+	return 0;
+}
+
+static int vboxsf_dir_rename(struct inode *old_parent,
+			     struct dentry *old_dentry,
+			     struct inode *new_parent,
+			     struct dentry *new_dentry,
+			     unsigned int flags)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(old_parent->i_sb);
+	struct vboxsf_inode *sf_old_parent_i = VBOXSF_I(old_parent);
+	struct vboxsf_inode *sf_new_parent_i = VBOXSF_I(new_parent);
+	u32 shfl_flags = SHFL_RENAME_FILE | SHFL_RENAME_REPLACE_IF_EXISTS;
+	struct shfl_string *old_path, *new_path;
+	int err;
+
+	if (flags)
+		return -EINVAL;
+
+	old_path = vboxsf_path_from_dentry(sbi, old_dentry);
+	if (IS_ERR(old_path))
+		return PTR_ERR(old_path);
+
+	new_path = vboxsf_path_from_dentry(sbi, new_dentry);
+	if (IS_ERR(new_path)) {
+		err = PTR_ERR(new_path);
+		goto err_put_old_path;
+	}
+
+	if (d_inode(old_dentry)->i_mode & S_IFDIR)
+		shfl_flags = 0;
+
+	err = vboxsf_rename(sbi->root, old_path, new_path, shfl_flags);
+	if (err == 0) {
+		/* parent directories access/change time changed */
+		sf_new_parent_i->force_restat = 1;
+		sf_old_parent_i->force_restat = 1;
+	}
+
+	__putname(new_path);
+err_put_old_path:
+	__putname(old_path);
+	return err;
+}
+
+static int vboxsf_dir_symlink(struct inode *parent, struct dentry *dentry,
+			      const char *symname)
+{
+	struct vboxsf_inode *sf_parent_i = VBOXSF_I(parent);
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(parent->i_sb);
+	int symname_size = strlen(symname) + 1;
+	struct shfl_string *path, *ssymname;
+	struct shfl_fsobjinfo info;
+	int err;
+
+	path = vboxsf_path_from_dentry(sbi, dentry);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	ssymname = kmalloc(SHFLSTRING_HEADER_SIZE + symname_size, GFP_KERNEL);
+	if (!ssymname) {
+		__putname(path);
+		return -ENOMEM;
+	}
+	ssymname->length = symname_size - 1;
+	ssymname->size = symname_size;
+	memcpy(ssymname->string.utf8, symname, symname_size);
+
+	err = vboxsf_symlink(sbi->root, path, ssymname, &info);
+	kfree(ssymname);
+	__putname(path);
+	if (err) {
+		/* -EROFS means symlinks are note support -> -EPERM */
+		return (err == -EROFS) ? -EPERM : err;
+	}
+
+	err = vboxsf_dir_instantiate(parent, dentry, &info);
+	if (err)
+		return err;
+
+	/* parent directory access/change time changed */
+	sf_parent_i->force_restat = 1;
+	return 0;
+}
+
+const struct inode_operations vboxsf_dir_iops = {
+	.lookup  = vboxsf_dir_lookup,
+	.create  = vboxsf_dir_mkfile,
+	.mkdir   = vboxsf_dir_mkdir,
+	.rmdir   = vboxsf_dir_unlink,
+	.unlink  = vboxsf_dir_unlink,
+	.rename  = vboxsf_dir_rename,
+	.symlink = vboxsf_dir_symlink,
+	.getattr = vboxsf_getattr,
+	.setattr = vboxsf_setattr,
+};
diff --git a/fs/vboxsf/file.c b/fs/vboxsf/file.c
new file mode 100644
index 000000000000..c4ab5996d97a
--- /dev/null
+++ b/fs/vboxsf/file.c
@@ -0,0 +1,379 @@
+// SPDX-License-Identifier: MIT
+/*
+ * VirtualBox Guest Shared Folders support: Regular file inode and file ops.
+ *
+ * Copyright (C) 2006-2018 Oracle Corporation
+ */
+
+#include <linux/mm.h>
+#include <linux/page-flags.h>
+#include <linux/pagemap.h>
+#include <linux/highmem.h>
+#include <linux/sizes.h>
+#include "vfsmod.h"
+
+struct vboxsf_handle {
+	u64 handle;
+	u32 root;
+	u32 access_flags;
+	struct kref refcount;
+	struct list_head head;
+};
+
+static int vboxsf_file_open(struct inode *inode, struct file *file)
+{
+	struct vboxsf_inode *sf_i = VBOXSF_I(inode);
+	struct shfl_createparms params = {};
+	struct vboxsf_handle *sf_handle;
+	u32 access_flags = 0;
+	int err;
+
+	sf_handle = kmalloc(sizeof(*sf_handle), GFP_KERNEL);
+	if (!sf_handle)
+		return -ENOMEM;
+
+	/*
+	 * We check the value of params.handle afterwards to find out if
+	 * the call succeeded or failed, as the API does not seem to cleanly
+	 * distinguish error and informational messages.
+	 *
+	 * Furthermore, we must set params.handle to SHFL_HANDLE_NIL to
+	 * make the shared folders host service use our mode parameter.
+	 */
+	params.handle = SHFL_HANDLE_NIL;
+	if (file->f_flags & O_CREAT) {
+		params.create_flags |= SHFL_CF_ACT_CREATE_IF_NEW;
+		/*
+		 * We ignore O_EXCL, as the Linux kernel seems to call create
+		 * beforehand itself, so O_EXCL should always fail.
+		 */
+		if (file->f_flags & O_TRUNC)
+			params.create_flags |= SHFL_CF_ACT_OVERWRITE_IF_EXISTS;
+		else
+			params.create_flags |= SHFL_CF_ACT_OPEN_IF_EXISTS;
+	} else {
+		params.create_flags |= SHFL_CF_ACT_FAIL_IF_NEW;
+		if (file->f_flags & O_TRUNC)
+			params.create_flags |= SHFL_CF_ACT_OVERWRITE_IF_EXISTS;
+	}
+
+	switch (file->f_flags & O_ACCMODE) {
+	case O_RDONLY:
+		access_flags |= SHFL_CF_ACCESS_READ;
+		break;
+
+	case O_WRONLY:
+		access_flags |= SHFL_CF_ACCESS_WRITE;
+		break;
+
+	case O_RDWR:
+		access_flags |= SHFL_CF_ACCESS_READWRITE;
+		break;
+
+	default:
+		WARN_ON(1);
+	}
+
+	if (file->f_flags & O_APPEND)
+		access_flags |= SHFL_CF_ACCESS_APPEND;
+
+	params.create_flags |= access_flags;
+	params.info.attr.mode = inode->i_mode;
+
+	err = vboxsf_create_at_dentry(file_dentry(file), &params);
+	if (err == 0 && params.handle == SHFL_HANDLE_NIL)
+		err = (params.result == SHFL_FILE_EXISTS) ? -EEXIST : -ENOENT;
+	if (err) {
+		kfree(sf_handle);
+		return err;
+	}
+
+	/* the host may have given us different attr then requested */
+	sf_i->force_restat = 1;
+
+	/* init our handle struct and add it to the inode's handles list */
+	sf_handle->handle = params.handle;
+	sf_handle->root = VBOXSF_SBI(inode->i_sb)->root;
+	sf_handle->access_flags = access_flags;
+	kref_init(&sf_handle->refcount);
+
+	mutex_lock(&sf_i->handle_list_mutex);
+	list_add(&sf_handle->head, &sf_i->handle_list);
+	mutex_unlock(&sf_i->handle_list_mutex);
+
+	file->private_data = sf_handle;
+	return 0;
+}
+
+static void vboxsf_handle_release(struct kref *refcount)
+{
+	struct vboxsf_handle *sf_handle =
+		container_of(refcount, struct vboxsf_handle, refcount);
+
+	vboxsf_close(sf_handle->root, sf_handle->handle);
+	kfree(sf_handle);
+}
+
+static int vboxsf_file_release(struct inode *inode, struct file *file)
+{
+	struct vboxsf_inode *sf_i = VBOXSF_I(inode);
+	struct vboxsf_handle *sf_handle = file->private_data;
+
+	/*
+	 * When a file is closed on our (the guest) side, we want any subsequent
+	 * accesses done on the host side to see all changes done from our side.
+	 */
+	filemap_write_and_wait(inode->i_mapping);
+
+	mutex_lock(&sf_i->handle_list_mutex);
+	list_del(&sf_handle->head);
+	mutex_unlock(&sf_i->handle_list_mutex);
+
+	kref_put(&sf_handle->refcount, vboxsf_handle_release);
+	return 0;
+}
+
+/*
+ * Write back dirty pages now, because there may not be any suitable
+ * open files later
+ */
+static void vboxsf_vma_close(struct vm_area_struct *vma)
+{
+	filemap_write_and_wait(vma->vm_file->f_mapping);
+}
+
+static const struct vm_operations_struct vboxsf_file_vm_ops = {
+	.close		= vboxsf_vma_close,
+	.fault		= filemap_fault,
+	.map_pages	= filemap_map_pages,
+};
+
+static int vboxsf_file_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	int err;
+
+	err = generic_file_mmap(file, vma);
+	if (!err)
+		vma->vm_ops = &vboxsf_file_vm_ops;
+
+	return err;
+}
+
+/*
+ * Note that since we are accessing files on the host's filesystem, files
+ * may always be changed underneath us by the host!
+ *
+ * The vboxsf API between the guest and the host does not offer any functions
+ * to deal with this. There is no inode-generation to check for changes, no
+ * events / callback on changes and no way to lock files.
+ *
+ * To avoid returning stale data when a file gets *opened* on our (the guest)
+ * side, we do a "stat" on the host side, then compare the mtime with the
+ * last known mtime and invalidate the page-cache if they differ.
+ * This is done from vboxsf_inode_revalidate().
+ *
+ * When reads are done through the read_iter fop, it is possible to do
+ * further cache revalidation then, there are 3 options to deal with this:
+ *
+ * 1)  Rely solely on the revalidation done at open time
+ * 2)  Do another "stat" and compare mtime again. Unfortunately the vboxsf
+ *     host API does not allow stat on handles, so we would need to use
+ *     file->f_path.dentry and the stat will then fail if the file was unlinked
+ *     or renamed (and there is no thing like NFS' silly-rename). So we get:
+ * 2a) "stat" and compare mtime, on stat failure invalidate the cache
+ * 2b) "stat" and compare mtime, on stat failure do nothing
+ * 3)  Simply always call invalidate_inode_pages2_range on the range of the read
+ *
+ * Currently we are keeping things KISS and using option 1. this allows
+ * directly using generic_file_read_iter without wrapping it.
+ *
+ * This means that only data written on the host side before open() on
+ * the guest side is guaranteed to be seen by the guest. If necessary
+ * we may provide other read-cache strategies in the future and make this
+ * configurable through a mount option.
+ */
+const struct file_operations vboxsf_reg_fops = {
+	.llseek = generic_file_llseek,
+	.read_iter = generic_file_read_iter,
+	.write_iter = generic_file_write_iter,
+	.mmap = vboxsf_file_mmap,
+	.open = vboxsf_file_open,
+	.release = vboxsf_file_release,
+	.fsync = noop_fsync,
+	.splice_read = generic_file_splice_read,
+};
+
+const struct inode_operations vboxsf_reg_iops = {
+	.getattr = vboxsf_getattr,
+	.setattr = vboxsf_setattr
+};
+
+static int vboxsf_readpage(struct file *file, struct page *page)
+{
+	struct vboxsf_handle *sf_handle = file->private_data;
+	loff_t off = page_offset(page);
+	u32 nread = PAGE_SIZE;
+	u8 *buf;
+	int err;
+
+	buf = kmap(page);
+
+	err = vboxsf_read(sf_handle->root, sf_handle->handle, off, &nread, buf);
+	if (err == 0) {
+		memset(&buf[nread], 0, PAGE_SIZE - nread);
+		flush_dcache_page(page);
+		SetPageUptodate(page);
+	} else {
+		SetPageError(page);
+	}
+
+	kunmap(page);
+	unlock_page(page);
+	return err;
+}
+
+static struct vboxsf_handle *vboxsf_get_write_handle(struct vboxsf_inode *sf_i)
+{
+	struct vboxsf_handle *h, *sf_handle = NULL;
+
+	mutex_lock(&sf_i->handle_list_mutex);
+	list_for_each_entry(h, &sf_i->handle_list, head) {
+		if (h->access_flags == SHFL_CF_ACCESS_WRITE ||
+		    h->access_flags == SHFL_CF_ACCESS_READWRITE) {
+			kref_get(&h->refcount);
+			sf_handle = h;
+			break;
+		}
+	}
+	mutex_unlock(&sf_i->handle_list_mutex);
+
+	return sf_handle;
+}
+
+static int vboxsf_writepage(struct page *page, struct writeback_control *wbc)
+{
+	struct inode *inode = page->mapping->host;
+	struct vboxsf_inode *sf_i = VBOXSF_I(inode);
+	struct vboxsf_handle *sf_handle;
+	loff_t off = page_offset(page);
+	loff_t size = i_size_read(inode);
+	u32 nwrite = PAGE_SIZE;
+	u8 *buf;
+	int err;
+
+	if (off + PAGE_SIZE > size)
+		nwrite = size & ~PAGE_MASK;
+
+	sf_handle = vboxsf_get_write_handle(sf_i);
+	if (!sf_handle)
+		return -EBADF;
+
+	buf = kmap(page);
+	err = vboxsf_write(sf_handle->root, sf_handle->handle,
+			   off, &nwrite, buf);
+	kunmap(page);
+
+	kref_put(&sf_handle->refcount, vboxsf_handle_release);
+
+	if (err == 0) {
+		ClearPageError(page);
+		/* mtime changed */
+		sf_i->force_restat = 1;
+	} else {
+		ClearPageUptodate(page);
+	}
+
+	unlock_page(page);
+	return err;
+}
+
+static int vboxsf_write_end(struct file *file, struct address_space *mapping,
+			    loff_t pos, unsigned int len, unsigned int copied,
+			    struct page *page, void *fsdata)
+{
+	struct inode *inode = mapping->host;
+	struct vboxsf_handle *sf_handle = file->private_data;
+	unsigned int from = pos & ~PAGE_MASK;
+	u32 nwritten = len;
+	u8 *buf;
+	int err;
+
+	/* zero the stale part of the page if we did a short copy */
+	if (!PageUptodate(page) && copied < len)
+		zero_user(page, from + copied, len - copied);
+
+	buf = kmap(page);
+	err = vboxsf_write(sf_handle->root, sf_handle->handle,
+			   pos, &nwritten, buf + from);
+	kunmap(page);
+
+	if (err) {
+		nwritten = 0;
+		goto out;
+	}
+
+	/* mtime changed */
+	VBOXSF_I(inode)->force_restat = 1;
+
+	if (!PageUptodate(page) && nwritten == PAGE_SIZE)
+		SetPageUptodate(page);
+
+	pos += nwritten;
+	if (pos > inode->i_size)
+		i_size_write(inode, pos);
+
+out:
+	unlock_page(page);
+	put_page(page);
+
+	return nwritten;
+}
+
+/*
+ * Note simple_write_begin does not read the page from disk on partial writes
+ * this is ok since vboxsf_write_end only writes the written parts of the
+ * page and it does not call SetPageUptodate for partial writes.
+ */
+const struct address_space_operations vboxsf_reg_aops = {
+	.readpage = vboxsf_readpage,
+	.writepage = vboxsf_writepage,
+	.set_page_dirty = __set_page_dirty_nobuffers,
+	.write_begin = simple_write_begin,
+	.write_end = vboxsf_write_end,
+};
+
+static const char *vboxsf_get_link(struct dentry *dentry, struct inode *inode,
+				   struct delayed_call *done)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(inode->i_sb);
+	struct shfl_string *path;
+	char *link;
+	int err;
+
+	if (!dentry)
+		return ERR_PTR(-ECHILD);
+
+	path = vboxsf_path_from_dentry(sbi, dentry);
+	if (IS_ERR(path))
+		return ERR_CAST(path);
+
+	link = kzalloc(PATH_MAX, GFP_KERNEL);
+	if (!link) {
+		__putname(path);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	err = vboxsf_readlink(sbi->root, path, PATH_MAX, link);
+	__putname(path);
+	if (err) {
+		kfree(link);
+		return ERR_PTR(err);
+	}
+
+	set_delayed_call(done, kfree_link, link);
+	return link;
+}
+
+const struct inode_operations vboxsf_lnk_iops = {
+	.get_link = vboxsf_get_link
+};
diff --git a/fs/vboxsf/shfl_hostintf.h b/fs/vboxsf/shfl_hostintf.h
new file mode 100644
index 000000000000..aca829062c12
--- /dev/null
+++ b/fs/vboxsf/shfl_hostintf.h
@@ -0,0 +1,901 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * VirtualBox Shared Folders: host interface definition.
+ *
+ * Copyright (C) 2006-2018 Oracle Corporation
+ */
+
+#ifndef SHFL_HOSTINTF_H
+#define SHFL_HOSTINTF_H
+
+#include <linux/vbox_vmmdev_types.h>
+
+/* The max in/out buffer size for a FN_READ or FN_WRITE call */
+#define SHFL_MAX_RW_COUNT           (16 * SZ_1M)
+
+/*
+ * Structures shared between guest and the service
+ * can be relocated and use offsets to point to variable
+ * length parts.
+ *
+ * Shared folders protocol works with handles.
+ * Before doing any action on a file system object,
+ * one have to obtain the object handle via a SHFL_FN_CREATE
+ * request. A handle must be closed with SHFL_FN_CLOSE.
+ */
+
+enum {
+	SHFL_FN_QUERY_MAPPINGS = 1,	/* Query mappings changes. */
+	SHFL_FN_QUERY_MAP_NAME = 2,	/* Query map name. */
+	SHFL_FN_CREATE = 3,		/* Open/create object. */
+	SHFL_FN_CLOSE = 4,		/* Close object handle. */
+	SHFL_FN_READ = 5,		/* Read object content. */
+	SHFL_FN_WRITE = 6,		/* Write new object content. */
+	SHFL_FN_LOCK = 7,		/* Lock/unlock a range in the object. */
+	SHFL_FN_LIST = 8,		/* List object content. */
+	SHFL_FN_INFORMATION = 9,	/* Query/set object information. */
+	/* Note function number 10 is not used! */
+	SHFL_FN_REMOVE = 11,		/* Remove object */
+	SHFL_FN_MAP_FOLDER_OLD = 12,	/* Map folder (legacy) */
+	SHFL_FN_UNMAP_FOLDER = 13,	/* Unmap folder */
+	SHFL_FN_RENAME = 14,		/* Rename object */
+	SHFL_FN_FLUSH = 15,		/* Flush file */
+	SHFL_FN_SET_UTF8 = 16,		/* Select UTF8 filename encoding */
+	SHFL_FN_MAP_FOLDER = 17,	/* Map folder */
+	SHFL_FN_READLINK = 18,		/* Read symlink dest (as of VBox 4.0) */
+	SHFL_FN_SYMLINK = 19,		/* Create symlink (as of VBox 4.0) */
+	SHFL_FN_SET_SYMLINKS = 20,	/* Ask host to show symlinks (4.0+) */
+};
+
+/* Root handles for a mapping are of type u32, Root handles are unique. */
+#define SHFL_ROOT_NIL		UINT_MAX
+
+/* Shared folders handle for an opened object are of type u64. */
+#define SHFL_HANDLE_NIL		ULLONG_MAX
+
+/* Hardcoded maximum length (in chars) of a shared folder name. */
+#define SHFL_MAX_LEN         (256)
+/* Hardcoded maximum number of shared folder mapping available to the guest. */
+#define SHFL_MAX_MAPPINGS    (64)
+
+/** Shared folder string buffer structure. */
+struct shfl_string {
+	/** Allocated size of the string member in bytes. */
+	u16 size;
+
+	/** Length of string without trailing nul in bytes. */
+	u16 length;
+
+	/** UTF-8 or UTF-16 string. Nul terminated. */
+	union {
+		u8 utf8[2];
+		u16 utf16[1];
+		u16 ucs2[1]; /* misnomer, use utf16. */
+	} string;
+};
+VMMDEV_ASSERT_SIZE(shfl_string, 6);
+
+/* The size of shfl_string w/o the string part. */
+#define SHFLSTRING_HEADER_SIZE  4
+
+/* Calculate size of the string. */
+static inline u32 shfl_string_buf_size(const struct shfl_string *string)
+{
+	return string ? SHFLSTRING_HEADER_SIZE + string->size : 0;
+}
+
+/* Set user id on execution (S_ISUID). */
+#define SHFL_UNIX_ISUID             0004000U
+/* Set group id on execution (S_ISGID). */
+#define SHFL_UNIX_ISGID             0002000U
+/* Sticky bit (S_ISVTX / S_ISTXT). */
+#define SHFL_UNIX_ISTXT             0001000U
+
+/* Owner readable (S_IRUSR). */
+#define SHFL_UNIX_IRUSR             0000400U
+/* Owner writable (S_IWUSR). */
+#define SHFL_UNIX_IWUSR             0000200U
+/* Owner executable (S_IXUSR). */
+#define SHFL_UNIX_IXUSR             0000100U
+
+/* Group readable (S_IRGRP). */
+#define SHFL_UNIX_IRGRP             0000040U
+/* Group writable (S_IWGRP). */
+#define SHFL_UNIX_IWGRP             0000020U
+/* Group executable (S_IXGRP). */
+#define SHFL_UNIX_IXGRP             0000010U
+
+/* Other readable (S_IROTH). */
+#define SHFL_UNIX_IROTH             0000004U
+/* Other writable (S_IWOTH). */
+#define SHFL_UNIX_IWOTH             0000002U
+/* Other executable (S_IXOTH). */
+#define SHFL_UNIX_IXOTH             0000001U
+
+/* Named pipe (fifo) (S_IFIFO). */
+#define SHFL_TYPE_FIFO              0010000U
+/* Character device (S_IFCHR). */
+#define SHFL_TYPE_DEV_CHAR          0020000U
+/* Directory (S_IFDIR). */
+#define SHFL_TYPE_DIRECTORY         0040000U
+/* Block device (S_IFBLK). */
+#define SHFL_TYPE_DEV_BLOCK         0060000U
+/* Regular file (S_IFREG). */
+#define SHFL_TYPE_FILE              0100000U
+/* Symbolic link (S_IFLNK). */
+#define SHFL_TYPE_SYMLINK           0120000U
+/* Socket (S_IFSOCK). */
+#define SHFL_TYPE_SOCKET            0140000U
+/* Whiteout (S_IFWHT). */
+#define SHFL_TYPE_WHITEOUT          0160000U
+/* Type mask (S_IFMT). */
+#define SHFL_TYPE_MASK              0170000U
+
+/* Checks the mode flags indicate a directory (S_ISDIR). */
+#define SHFL_IS_DIRECTORY(m)   (((m) & SHFL_TYPE_MASK) == SHFL_TYPE_DIRECTORY)
+/* Checks the mode flags indicate a symbolic link (S_ISLNK). */
+#define SHFL_IS_SYMLINK(m)     (((m) & SHFL_TYPE_MASK) == SHFL_TYPE_SYMLINK)
+
+/** The available additional information in a shfl_fsobjattr object. */
+enum shfl_fsobjattr_add {
+	/** No additional information is available / requested. */
+	SHFLFSOBJATTRADD_NOTHING = 1,
+	/**
+	 * The additional unix attributes (shfl_fsobjattr::u::unix_attr) are
+	 *  available / requested.
+	 */
+	SHFLFSOBJATTRADD_UNIX,
+	/**
+	 * The additional extended attribute size (shfl_fsobjattr::u::size) is
+	 *  available / requested.
+	 */
+	SHFLFSOBJATTRADD_EASIZE,
+	/**
+	 * The last valid item (inclusive).
+	 * The valid range is SHFLFSOBJATTRADD_NOTHING thru
+	 * SHFLFSOBJATTRADD_LAST.
+	 */
+	SHFLFSOBJATTRADD_LAST = SHFLFSOBJATTRADD_EASIZE,
+
+	/** The usual 32-bit hack. */
+	SHFLFSOBJATTRADD_32BIT_SIZE_HACK = 0x7fffffff
+};
+
+/**
+ * Additional unix Attributes, these are available when
+ * shfl_fsobjattr.additional == SHFLFSOBJATTRADD_UNIX.
+ */
+struct shfl_fsobjattr_unix {
+	/**
+	 * The user owning the filesystem object (st_uid).
+	 * This field is ~0U if not supported.
+	 */
+	u32 uid;
+
+	/**
+	 * The group the filesystem object is assigned (st_gid).
+	 * This field is ~0U if not supported.
+	 */
+	u32 gid;
+
+	/**
+	 * Number of hard links to this filesystem object (st_nlink).
+	 * This field is 1 if the filesystem doesn't support hardlinking or
+	 * the information isn't available.
+	 */
+	u32 hardlinks;
+
+	/**
+	 * The device number of the device which this filesystem object resides
+	 * on (st_dev). This field is 0 if this information is not available.
+	 */
+	u32 inode_id_device;
+
+	/**
+	 * The unique identifier (within the filesystem) of this filesystem
+	 * object (st_ino). Together with inode_id_device, this field can be
+	 * used as a OS wide unique id, when both their values are not 0.
+	 * This field is 0 if the information is not available.
+	 */
+	u64 inode_id;
+
+	/**
+	 * User flags (st_flags).
+	 * This field is 0 if this information is not available.
+	 */
+	u32 flags;
+
+	/**
+	 * The current generation number (st_gen).
+	 * This field is 0 if this information is not available.
+	 */
+	u32 generation_id;
+
+	/**
+	 * The device number of a char. or block device type object (st_rdev).
+	 * This field is 0 if the file isn't a char. or block device or when
+	 * the OS doesn't use the major+minor device idenfication scheme.
+	 */
+	u32 device;
+} __packed;
+
+/** Extended attribute size. */
+struct shfl_fsobjattr_easize {
+	/** Size of EAs. */
+	s64 cb;
+} __packed;
+
+/** Shared folder filesystem object attributes. */
+struct shfl_fsobjattr {
+	/** Mode flags (st_mode). SHFL_UNIX_*, SHFL_TYPE_*, and SHFL_DOS_*. */
+	u32 mode;
+
+	/** The additional attributes available. */
+	enum shfl_fsobjattr_add additional;
+
+	/**
+	 * Additional attributes.
+	 *
+	 * Unless explicitly specified to an API, the API can provide additional
+	 * data as it is provided by the underlying OS.
+	 */
+	union {
+		struct shfl_fsobjattr_unix unix_attr;
+		struct shfl_fsobjattr_easize size;
+	} __packed u;
+} __packed;
+VMMDEV_ASSERT_SIZE(shfl_fsobjattr, 44);
+
+struct shfl_timespec {
+	s64 ns_relative_to_unix_epoch;
+};
+
+/** Filesystem object information structure. */
+struct shfl_fsobjinfo {
+	/**
+	 * Logical size (st_size).
+	 * For normal files this is the size of the file.
+	 * For symbolic links, this is the length of the path name contained
+	 * in the symbolic link.
+	 * For other objects this fields needs to be specified.
+	 */
+	s64 size;
+
+	/** Disk allocation size (st_blocks * DEV_BSIZE). */
+	s64 allocated;
+
+	/** Time of last access (st_atime). */
+	struct shfl_timespec access_time;
+
+	/** Time of last data modification (st_mtime). */
+	struct shfl_timespec modification_time;
+
+	/**
+	 * Time of last status change (st_ctime).
+	 * If not available this is set to modification_time.
+	 */
+	struct shfl_timespec change_time;
+
+	/**
+	 * Time of file birth (st_birthtime).
+	 * If not available this is set to change_time.
+	 */
+	struct shfl_timespec birth_time;
+
+	/** Attributes. */
+	struct shfl_fsobjattr attr;
+
+} __packed;
+VMMDEV_ASSERT_SIZE(shfl_fsobjinfo, 92);
+
+/**
+ * result of an open/create request.
+ * Along with handle value the result code
+ * identifies what has happened while
+ * trying to open the object.
+ */
+enum shfl_create_result {
+	SHFL_NO_RESULT,
+	/** Specified path does not exist. */
+	SHFL_PATH_NOT_FOUND,
+	/** Path to file exists, but the last component does not. */
+	SHFL_FILE_NOT_FOUND,
+	/** File already exists and either has been opened or not. */
+	SHFL_FILE_EXISTS,
+	/** New file was created. */
+	SHFL_FILE_CREATED,
+	/** Existing file was replaced or overwritten. */
+	SHFL_FILE_REPLACED
+};
+
+/* No flags. Initialization value. */
+#define SHFL_CF_NONE                  (0x00000000)
+
+/*
+ * Only lookup the object, do not return a handle. When this is set all other
+ * flags are ignored.
+ */
+#define SHFL_CF_LOOKUP                (0x00000001)
+
+/*
+ * Open parent directory of specified object.
+ * Useful for the corresponding Windows FSD flag
+ * and for opening paths like \\dir\\*.* to search the 'dir'.
+ */
+#define SHFL_CF_OPEN_TARGET_DIRECTORY (0x00000002)
+
+/* Create/open a directory. */
+#define SHFL_CF_DIRECTORY             (0x00000004)
+
+/*
+ *  Open/create action to do if object exists
+ *  and if the object does not exists.
+ *  REPLACE file means atomically DELETE and CREATE.
+ *  OVERWRITE file means truncating the file to 0 and
+ *  setting new size.
+ *  When opening an existing directory REPLACE and OVERWRITE
+ *  actions are considered invalid, and cause returning
+ *  FILE_EXISTS with NIL handle.
+ */
+#define SHFL_CF_ACT_MASK_IF_EXISTS      (0x000000f0)
+#define SHFL_CF_ACT_MASK_IF_NEW         (0x00000f00)
+
+/* What to do if object exists. */
+#define SHFL_CF_ACT_OPEN_IF_EXISTS      (0x00000000)
+#define SHFL_CF_ACT_FAIL_IF_EXISTS      (0x00000010)
+#define SHFL_CF_ACT_REPLACE_IF_EXISTS   (0x00000020)
+#define SHFL_CF_ACT_OVERWRITE_IF_EXISTS (0x00000030)
+
+/* What to do if object does not exist. */
+#define SHFL_CF_ACT_CREATE_IF_NEW       (0x00000000)
+#define SHFL_CF_ACT_FAIL_IF_NEW         (0x00000100)
+
+/* Read/write requested access for the object. */
+#define SHFL_CF_ACCESS_MASK_RW          (0x00003000)
+
+/* No access requested. */
+#define SHFL_CF_ACCESS_NONE             (0x00000000)
+/* Read access requested. */
+#define SHFL_CF_ACCESS_READ             (0x00001000)
+/* Write access requested. */
+#define SHFL_CF_ACCESS_WRITE            (0x00002000)
+/* Read/Write access requested. */
+#define SHFL_CF_ACCESS_READWRITE	(0x00003000)
+
+/* Requested share access for the object. */
+#define SHFL_CF_ACCESS_MASK_DENY        (0x0000c000)
+
+/* Allow any access. */
+#define SHFL_CF_ACCESS_DENYNONE         (0x00000000)
+/* Do not allow read. */
+#define SHFL_CF_ACCESS_DENYREAD         (0x00004000)
+/* Do not allow write. */
+#define SHFL_CF_ACCESS_DENYWRITE        (0x00008000)
+/* Do not allow access. */
+#define SHFL_CF_ACCESS_DENYALL          (0x0000c000)
+
+/* Requested access to attributes of the object. */
+#define SHFL_CF_ACCESS_MASK_ATTR        (0x00030000)
+
+/* No access requested. */
+#define SHFL_CF_ACCESS_ATTR_NONE        (0x00000000)
+/* Read access requested. */
+#define SHFL_CF_ACCESS_ATTR_READ        (0x00010000)
+/* Write access requested. */
+#define SHFL_CF_ACCESS_ATTR_WRITE       (0x00020000)
+/* Read/Write access requested. */
+#define SHFL_CF_ACCESS_ATTR_READWRITE   (0x00030000)
+
+/*
+ * The file is opened in append mode.
+ * Ignored if SHFL_CF_ACCESS_WRITE is not set.
+ */
+#define SHFL_CF_ACCESS_APPEND           (0x00040000)
+
+/** Create parameters buffer struct for SHFL_FN_CREATE call */
+struct shfl_createparms {
+	/** Returned handle of opened object. */
+	u64 handle;
+
+	/** Returned result of the operation */
+	enum shfl_create_result result;
+
+	/** SHFL_CF_* */
+	u32 create_flags;
+
+	/**
+	 * Attributes of object to create and
+	 * returned actual attributes of opened/created object.
+	 */
+	struct shfl_fsobjinfo info;
+} __packed;
+
+/** Shared Folder directory information */
+struct shfl_dirinfo {
+	/** Full information about the object. */
+	struct shfl_fsobjinfo info;
+	/**
+	 * The length of the short field (number of UTF16 chars).
+	 * It is 16-bit for reasons of alignment.
+	 */
+	u16 short_name_len;
+	/**
+	 * The short name for 8.3 compatibility.
+	 * Empty string if not available.
+	 */
+	u16 short_name[14];
+	struct shfl_string name;
+};
+
+/** Shared folder filesystem properties. */
+struct shfl_fsproperties {
+	/**
+	 * The maximum size of a filesystem object name.
+	 * This does not include the '\\0'.
+	 */
+	u32 max_component_len;
+
+	/**
+	 * True if the filesystem is remote.
+	 * False if the filesystem is local.
+	 */
+	bool remote;
+
+	/**
+	 * True if the filesystem is case sensitive.
+	 * False if the filesystem is case insensitive.
+	 */
+	bool case_sensitive;
+
+	/**
+	 * True if the filesystem is mounted read only.
+	 * False if the filesystem is mounted read write.
+	 */
+	bool read_only;
+
+	/**
+	 * True if the filesystem can encode unicode object names.
+	 * False if it can't.
+	 */
+	bool supports_unicode;
+
+	/**
+	 * True if the filesystem is compresses.
+	 * False if it isn't or we don't know.
+	 */
+	bool compressed;
+
+	/**
+	 * True if the filesystem compresses of individual files.
+	 * False if it doesn't or we don't know.
+	 */
+	bool file_compression;
+};
+VMMDEV_ASSERT_SIZE(shfl_fsproperties, 12);
+
+struct shfl_volinfo {
+	s64 total_allocation_bytes;
+	s64 available_allocation_bytes;
+	u32 bytes_per_allocation_unit;
+	u32 bytes_per_sector;
+	u32 serial;
+	struct shfl_fsproperties properties;
+};
+
+
+/** SHFL_FN_MAP_FOLDER Parameters structure. */
+struct shfl_map_folder {
+	/**
+	 * pointer, in:
+	 * Points to struct shfl_string buffer.
+	 */
+	struct vmmdev_hgcm_function_parameter path;
+
+	/**
+	 * pointer, out: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * pointer, in: UTF16
+	 * Path delimiter
+	 */
+	struct vmmdev_hgcm_function_parameter delimiter;
+
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Case senstive flag
+	 */
+	struct vmmdev_hgcm_function_parameter case_sensitive;
+
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_MAP_FOLDER (4)
+
+
+/** SHFL_FN_UNMAP_FOLDER Parameters structure. */
+struct shfl_unmap_folder {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_UNMAP_FOLDER (1)
+
+
+/** SHFL_FN_CREATE Parameters structure. */
+struct shfl_create {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * pointer, in:
+	 * Points to struct shfl_string buffer.
+	 */
+	struct vmmdev_hgcm_function_parameter path;
+
+	/**
+	 * pointer, in/out:
+	 * Points to struct shfl_createparms buffer.
+	 */
+	struct vmmdev_hgcm_function_parameter parms;
+
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_CREATE (3)
+
+
+/** SHFL_FN_CLOSE Parameters structure. */
+struct shfl_close {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * value64, in:
+	 * SHFLHANDLE (u64) of object to close.
+	 */
+	struct vmmdev_hgcm_function_parameter handle;
+
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_CLOSE (2)
+
+
+/** SHFL_FN_READ Parameters structure. */
+struct shfl_read {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * value64, in:
+	 * SHFLHANDLE (u64) of object to read from.
+	 */
+	struct vmmdev_hgcm_function_parameter handle;
+
+	/**
+	 * value64, in:
+	 * Offset to read from.
+	 */
+	struct vmmdev_hgcm_function_parameter offset;
+
+	/**
+	 * value64, in/out:
+	 * Bytes to read/How many were read.
+	 */
+	struct vmmdev_hgcm_function_parameter cb;
+
+	/**
+	 * pointer, out:
+	 * Buffer to place data to.
+	 */
+	struct vmmdev_hgcm_function_parameter buffer;
+
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_READ (5)
+
+
+/** SHFL_FN_WRITE Parameters structure. */
+struct shfl_write {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * value64, in:
+	 * SHFLHANDLE (u64) of object to write to.
+	 */
+	struct vmmdev_hgcm_function_parameter handle;
+
+	/**
+	 * value64, in:
+	 * Offset to write to.
+	 */
+	struct vmmdev_hgcm_function_parameter offset;
+
+	/**
+	 * value64, in/out:
+	 * Bytes to write/How many were written.
+	 */
+	struct vmmdev_hgcm_function_parameter cb;
+
+	/**
+	 * pointer, in:
+	 * Data to write.
+	 */
+	struct vmmdev_hgcm_function_parameter buffer;
+
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_WRITE (5)
+
+
+/*
+ * SHFL_FN_LIST
+ * Listing information includes variable length RTDIRENTRY[EX] structures.
+ */
+
+#define SHFL_LIST_NONE			0
+#define SHFL_LIST_RETURN_ONE		1
+
+/** SHFL_FN_LIST Parameters structure. */
+struct shfl_list {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * value64, in:
+	 * SHFLHANDLE (u64) of object to be listed.
+	 */
+	struct vmmdev_hgcm_function_parameter handle;
+
+	/**
+	 * value32, in:
+	 * List flags SHFL_LIST_*.
+	 */
+	struct vmmdev_hgcm_function_parameter flags;
+
+	/**
+	 * value32, in/out:
+	 * Bytes to be used for listing information/How many bytes were used.
+	 */
+	struct vmmdev_hgcm_function_parameter cb;
+
+	/**
+	 * pointer, in/optional
+	 * Points to struct shfl_string buffer that specifies a search path.
+	 */
+	struct vmmdev_hgcm_function_parameter path;
+
+	/**
+	 * pointer, out:
+	 * Buffer to place listing information to. (struct shfl_dirinfo)
+	 */
+	struct vmmdev_hgcm_function_parameter buffer;
+
+	/**
+	 * value32, in/out:
+	 * Indicates a key where the listing must be resumed.
+	 * in: 0 means start from begin of object.
+	 * out: 0 means listing completed.
+	 */
+	struct vmmdev_hgcm_function_parameter resume_point;
+
+	/**
+	 * pointer, out:
+	 * Number of files returned
+	 */
+	struct vmmdev_hgcm_function_parameter file_count;
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_LIST (8)
+
+
+/** SHFL_FN_READLINK Parameters structure. */
+struct shfl_readLink {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * pointer, in:
+	 * Points to struct shfl_string buffer.
+	 */
+	struct vmmdev_hgcm_function_parameter path;
+
+	/**
+	 * pointer, out:
+	 * Buffer to place data to.
+	 */
+	struct vmmdev_hgcm_function_parameter buffer;
+
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_READLINK (3)
+
+
+/* SHFL_FN_INFORMATION */
+
+/* Mask of Set/Get bit. */
+#define SHFL_INFO_MODE_MASK    (0x1)
+/* Get information */
+#define SHFL_INFO_GET          (0x0)
+/* Set information */
+#define SHFL_INFO_SET          (0x1)
+
+/* Get name of the object. */
+#define SHFL_INFO_NAME         (0x2)
+/* Set size of object (extend/trucate); only applies to file objects */
+#define SHFL_INFO_SIZE         (0x4)
+/* Get/Set file object info. */
+#define SHFL_INFO_FILE         (0x8)
+/* Get volume information. */
+#define SHFL_INFO_VOLUME       (0x10)
+
+/** SHFL_FN_INFORMATION Parameters structure. */
+struct shfl_information {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * value64, in:
+	 * SHFLHANDLE (u64) of object to be listed.
+	 */
+	struct vmmdev_hgcm_function_parameter handle;
+
+	/**
+	 * value32, in:
+	 * SHFL_INFO_*
+	 */
+	struct vmmdev_hgcm_function_parameter flags;
+
+	/**
+	 * value32, in/out:
+	 * Bytes to be used for information/How many bytes were used.
+	 */
+	struct vmmdev_hgcm_function_parameter cb;
+
+	/**
+	 * pointer, in/out:
+	 * Information to be set/get (shfl_fsobjinfo or shfl_string). Do not
+	 * forget to set the shfl_fsobjinfo::attr::additional for a get
+	 * operation as well.
+	 */
+	struct vmmdev_hgcm_function_parameter info;
+
+};
+
+/* Number of parameters */
+#define SHFL_CPARMS_INFORMATION (5)
+
+
+/* SHFL_FN_REMOVE */
+
+#define SHFL_REMOVE_FILE        (0x1)
+#define SHFL_REMOVE_DIR         (0x2)
+#define SHFL_REMOVE_SYMLINK     (0x4)
+
+/** SHFL_FN_REMOVE Parameters structure. */
+struct shfl_remove {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * pointer, in:
+	 * Points to struct shfl_string buffer.
+	 */
+	struct vmmdev_hgcm_function_parameter path;
+
+	/**
+	 * value32, in:
+	 * remove flags (file/directory)
+	 */
+	struct vmmdev_hgcm_function_parameter flags;
+
+};
+
+#define SHFL_CPARMS_REMOVE  (3)
+
+
+/* SHFL_FN_RENAME */
+
+#define SHFL_RENAME_FILE                (0x1)
+#define SHFL_RENAME_DIR                 (0x2)
+#define SHFL_RENAME_REPLACE_IF_EXISTS   (0x4)
+
+/** SHFL_FN_RENAME Parameters structure. */
+struct shfl_rename {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * pointer, in:
+	 * Points to struct shfl_string src.
+	 */
+	struct vmmdev_hgcm_function_parameter src;
+
+	/**
+	 * pointer, in:
+	 * Points to struct shfl_string dest.
+	 */
+	struct vmmdev_hgcm_function_parameter dest;
+
+	/**
+	 * value32, in:
+	 * rename flags (file/directory)
+	 */
+	struct vmmdev_hgcm_function_parameter flags;
+
+};
+
+#define SHFL_CPARMS_RENAME  (4)
+
+
+/** SHFL_FN_SYMLINK Parameters structure. */
+struct shfl_symlink {
+	/**
+	 * pointer, in: SHFLROOT (u32)
+	 * Root handle of the mapping which name is queried.
+	 */
+	struct vmmdev_hgcm_function_parameter root;
+
+	/**
+	 * pointer, in:
+	 * Points to struct shfl_string of path for the new symlink.
+	 */
+	struct vmmdev_hgcm_function_parameter new_path;
+
+	/**
+	 * pointer, in:
+	 * Points to struct shfl_string of destination for symlink.
+	 */
+	struct vmmdev_hgcm_function_parameter old_path;
+
+	/**
+	 * pointer, out:
+	 * Information about created symlink.
+	 */
+	struct vmmdev_hgcm_function_parameter info;
+
+};
+
+#define SHFL_CPARMS_SYMLINK  (4)
+
+#endif
diff --git a/fs/vboxsf/super.c b/fs/vboxsf/super.c
new file mode 100644
index 000000000000..8fe03b4a0d2b
--- /dev/null
+++ b/fs/vboxsf/super.c
@@ -0,0 +1,491 @@
+// SPDX-License-Identifier: MIT
+/*
+ * VirtualBox Guest Shared Folders support: Virtual File System.
+ *
+ * Module initialization/finalization
+ * File system registration/deregistration
+ * Superblock reading
+ * Few utility functions
+ *
+ * Copyright (C) 2006-2018 Oracle Corporation
+ */
+
+#include <linux/idr.h>
+#include <linux/fs_parser.h>
+#include <linux/magic.h>
+#include <linux/module.h>
+#include <linux/nls.h>
+#include <linux/statfs.h>
+#include <linux/vbox_utils.h>
+#include "vfsmod.h"
+
+#define VBOXSF_SUPER_MAGIC 0x786f4256 /* 'VBox' little endian */
+
+#define VBSF_MOUNT_SIGNATURE_BYTE_0 ('\000')
+#define VBSF_MOUNT_SIGNATURE_BYTE_1 ('\377')
+#define VBSF_MOUNT_SIGNATURE_BYTE_2 ('\376')
+#define VBSF_MOUNT_SIGNATURE_BYTE_3 ('\375')
+
+static int follow_symlinks;
+module_param(follow_symlinks, int, 0444);
+MODULE_PARM_DESC(follow_symlinks,
+		 "Let host resolve symlinks rather than showing them");
+
+static DEFINE_IDA(vboxsf_bdi_ida);
+static DEFINE_MUTEX(vboxsf_setup_mutex);
+static bool vboxsf_setup_done;
+static struct super_operations vboxsf_super_ops; /* forward declaration */
+static struct kmem_cache *vboxsf_inode_cachep;
+
+static char * const vboxsf_default_nls = CONFIG_NLS_DEFAULT;
+
+enum  { opt_nls, opt_uid, opt_gid, opt_ttl, opt_dmode, opt_fmode,
+	opt_dmask, opt_fmask };
+
+static const struct fs_parameter_spec vboxsf_fs_parameters[] = {
+	fsparam_string	("nls",		opt_nls),
+	fsparam_u32	("uid",		opt_uid),
+	fsparam_u32	("gid",		opt_gid),
+	fsparam_u32	("ttl",		opt_ttl),
+	fsparam_u32oct	("dmode",	opt_dmode),
+	fsparam_u32oct	("fmode",	opt_fmode),
+	fsparam_u32oct	("dmask",	opt_dmask),
+	fsparam_u32oct	("fmask",	opt_fmask),
+	{}
+};
+
+static int vboxsf_parse_param(struct fs_context *fc, struct fs_parameter *param)
+{
+	struct vboxsf_fs_context *ctx = fc->fs_private;
+	struct fs_parse_result result;
+	kuid_t uid;
+	kgid_t gid;
+	int opt;
+
+	opt = fs_parse(fc, vboxsf_fs_parameters, param, &result);
+	if (opt < 0)
+		return opt;
+
+	switch (opt) {
+	case opt_nls:
+		if (ctx->nls_name || fc->purpose != FS_CONTEXT_FOR_MOUNT) {
+			vbg_err("vboxsf: Cannot reconfigure nls option\n");
+			return -EINVAL;
+		}
+		ctx->nls_name = param->string;
+		param->string = NULL;
+		break;
+	case opt_uid:
+		uid = make_kuid(current_user_ns(), result.uint_32);
+		if (!uid_valid(uid))
+			return -EINVAL;
+		ctx->o.uid = uid;
+		break;
+	case opt_gid:
+		gid = make_kgid(current_user_ns(), result.uint_32);
+		if (!gid_valid(gid))
+			return -EINVAL;
+		ctx->o.gid = gid;
+		break;
+	case opt_ttl:
+		ctx->o.ttl = msecs_to_jiffies(result.uint_32);
+		break;
+	case opt_dmode:
+		if (result.uint_32 & ~0777)
+			return -EINVAL;
+		ctx->o.dmode = result.uint_32;
+		ctx->o.dmode_set = true;
+		break;
+	case opt_fmode:
+		if (result.uint_32 & ~0777)
+			return -EINVAL;
+		ctx->o.fmode = result.uint_32;
+		ctx->o.fmode_set = true;
+		break;
+	case opt_dmask:
+		if (result.uint_32 & ~07777)
+			return -EINVAL;
+		ctx->o.dmask = result.uint_32;
+		break;
+	case opt_fmask:
+		if (result.uint_32 & ~07777)
+			return -EINVAL;
+		ctx->o.fmask = result.uint_32;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int vboxsf_fill_super(struct super_block *sb, struct fs_context *fc)
+{
+	struct vboxsf_fs_context *ctx = fc->fs_private;
+	struct shfl_string *folder_name, root_path;
+	struct vboxsf_sbi *sbi;
+	struct dentry *droot;
+	struct inode *iroot;
+	char *nls_name;
+	size_t size;
+	int err;
+
+	if (!fc->source)
+		return -EINVAL;
+
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
+	if (!sbi)
+		return -ENOMEM;
+
+	sbi->o = ctx->o;
+	idr_init(&sbi->ino_idr);
+	spin_lock_init(&sbi->ino_idr_lock);
+	sbi->next_generation = 1;
+	sbi->bdi_id = -1;
+
+	/* Load nls if not utf8 */
+	nls_name = ctx->nls_name ? ctx->nls_name : vboxsf_default_nls;
+	if (strcmp(nls_name, "utf8") != 0) {
+		if (nls_name == vboxsf_default_nls)
+			sbi->nls = load_nls_default();
+		else
+			sbi->nls = load_nls(nls_name);
+
+		if (!sbi->nls) {
+			vbg_err("vboxsf: Count not load '%s' nls\n", nls_name);
+			err = -EINVAL;
+			goto fail_free;
+		}
+	}
+
+	sbi->bdi_id = ida_simple_get(&vboxsf_bdi_ida, 0, 0, GFP_KERNEL);
+	if (sbi->bdi_id < 0) {
+		err = sbi->bdi_id;
+		goto fail_free;
+	}
+
+	err = super_setup_bdi_name(sb, "vboxsf-%d", sbi->bdi_id);
+	if (err)
+		goto fail_free;
+
+	/* Turn source into a shfl_string and map the folder */
+	size = strlen(fc->source) + 1;
+	folder_name = kmalloc(SHFLSTRING_HEADER_SIZE + size, GFP_KERNEL);
+	if (!folder_name) {
+		err = -ENOMEM;
+		goto fail_free;
+	}
+	folder_name->size = size;
+	folder_name->length = size - 1;
+	strlcpy(folder_name->string.utf8, fc->source, size);
+	err = vboxsf_map_folder(folder_name, &sbi->root);
+	kfree(folder_name);
+	if (err) {
+		vbg_err("vboxsf: Host rejected mount of '%s' with error %d\n",
+			fc->source, err);
+		goto fail_free;
+	}
+
+	root_path.length = 1;
+	root_path.size = 2;
+	root_path.string.utf8[0] = '/';
+	root_path.string.utf8[1] = 0;
+	err = vboxsf_stat(sbi, &root_path, &sbi->root_info);
+	if (err)
+		goto fail_unmap;
+
+	sb->s_magic = VBOXSF_SUPER_MAGIC;
+	sb->s_blocksize = 1024;
+	sb->s_maxbytes = MAX_LFS_FILESIZE;
+	sb->s_op = &vboxsf_super_ops;
+	sb->s_d_op = &vboxsf_dentry_ops;
+
+	iroot = iget_locked(sb, 0);
+	if (!iroot) {
+		err = -ENOMEM;
+		goto fail_unmap;
+	}
+	vboxsf_init_inode(sbi, iroot, &sbi->root_info);
+	unlock_new_inode(iroot);
+
+	droot = d_make_root(iroot);
+	if (!droot) {
+		err = -ENOMEM;
+		goto fail_unmap;
+	}
+
+	sb->s_root = droot;
+	sb->s_fs_info = sbi;
+	return 0;
+
+fail_unmap:
+	vboxsf_unmap_folder(sbi->root);
+fail_free:
+	if (sbi->bdi_id >= 0)
+		ida_simple_remove(&vboxsf_bdi_ida, sbi->bdi_id);
+	if (sbi->nls)
+		unload_nls(sbi->nls);
+	idr_destroy(&sbi->ino_idr);
+	kfree(sbi);
+	return err;
+}
+
+static void vboxsf_inode_init_once(void *data)
+{
+	struct vboxsf_inode *sf_i = data;
+
+	mutex_init(&sf_i->handle_list_mutex);
+	inode_init_once(&sf_i->vfs_inode);
+}
+
+static struct inode *vboxsf_alloc_inode(struct super_block *sb)
+{
+	struct vboxsf_inode *sf_i;
+
+	sf_i = kmem_cache_alloc(vboxsf_inode_cachep, GFP_NOFS);
+	if (!sf_i)
+		return NULL;
+
+	sf_i->force_restat = 0;
+	INIT_LIST_HEAD(&sf_i->handle_list);
+
+	return &sf_i->vfs_inode;
+}
+
+static void vboxsf_free_inode(struct inode *inode)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(inode->i_sb);
+	unsigned long flags;
+
+	spin_lock_irqsave(&sbi->ino_idr_lock, flags);
+	idr_remove(&sbi->ino_idr, inode->i_ino);
+	spin_unlock_irqrestore(&sbi->ino_idr_lock, flags);
+	kmem_cache_free(vboxsf_inode_cachep, VBOXSF_I(inode));
+}
+
+static void vboxsf_put_super(struct super_block *sb)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(sb);
+
+	vboxsf_unmap_folder(sbi->root);
+	if (sbi->bdi_id >= 0)
+		ida_simple_remove(&vboxsf_bdi_ida, sbi->bdi_id);
+	if (sbi->nls)
+		unload_nls(sbi->nls);
+
+	/*
+	 * vboxsf_free_inode uses the idr, make sure all delayed rcu free
+	 * inodes are flushed.
+	 */
+	rcu_barrier();
+	idr_destroy(&sbi->ino_idr);
+	kfree(sbi);
+}
+
+static int vboxsf_statfs(struct dentry *dentry, struct kstatfs *stat)
+{
+	struct super_block *sb = dentry->d_sb;
+	struct shfl_volinfo shfl_volinfo;
+	struct vboxsf_sbi *sbi;
+	u32 buf_len;
+	int err;
+
+	sbi = VBOXSF_SBI(sb);
+	buf_len = sizeof(shfl_volinfo);
+	err = vboxsf_fsinfo(sbi->root, 0, SHFL_INFO_GET | SHFL_INFO_VOLUME,
+			    &buf_len, &shfl_volinfo);
+	if (err)
+		return err;
+
+	stat->f_type = VBOXSF_SUPER_MAGIC;
+	stat->f_bsize = shfl_volinfo.bytes_per_allocation_unit;
+
+	do_div(shfl_volinfo.total_allocation_bytes,
+	       shfl_volinfo.bytes_per_allocation_unit);
+	stat->f_blocks = shfl_volinfo.total_allocation_bytes;
+
+	do_div(shfl_volinfo.available_allocation_bytes,
+	       shfl_volinfo.bytes_per_allocation_unit);
+	stat->f_bfree  = shfl_volinfo.available_allocation_bytes;
+	stat->f_bavail = shfl_volinfo.available_allocation_bytes;
+
+	stat->f_files = 1000;
+	/*
+	 * Don't return 0 here since the guest may then think that it is not
+	 * possible to create any more files.
+	 */
+	stat->f_ffree = 1000000;
+	stat->f_fsid.val[0] = 0;
+	stat->f_fsid.val[1] = 0;
+	stat->f_namelen = 255;
+	return 0;
+}
+
+static struct super_operations vboxsf_super_ops = {
+	.alloc_inode	= vboxsf_alloc_inode,
+	.free_inode	= vboxsf_free_inode,
+	.put_super	= vboxsf_put_super,
+	.statfs		= vboxsf_statfs,
+};
+
+static int vboxsf_setup(void)
+{
+	int err;
+
+	mutex_lock(&vboxsf_setup_mutex);
+
+	if (vboxsf_setup_done)
+		goto success;
+
+	vboxsf_inode_cachep =
+		kmem_cache_create("vboxsf_inode_cache",
+				  sizeof(struct vboxsf_inode), 0,
+				  (SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD |
+				   SLAB_ACCOUNT),
+				  vboxsf_inode_init_once);
+	if (!vboxsf_inode_cachep) {
+		err = -ENOMEM;
+		goto fail_nomem;
+	}
+
+	err = vboxsf_connect();
+	if (err) {
+		vbg_err("vboxsf: err %d connecting to guest PCI-device\n", err);
+		vbg_err("vboxsf: make sure you are inside a VirtualBox VM\n");
+		vbg_err("vboxsf: and check dmesg for vboxguest errors\n");
+		goto fail_free_cache;
+	}
+
+	err = vboxsf_set_utf8();
+	if (err) {
+		vbg_err("vboxsf_setutf8 error %d\n", err);
+		goto fail_disconnect;
+	}
+
+	if (!follow_symlinks) {
+		err = vboxsf_set_symlinks();
+		if (err)
+			vbg_warn("vboxsf: Unable to show symlinks: %d\n", err);
+	}
+
+	vboxsf_setup_done = true;
+success:
+	mutex_unlock(&vboxsf_setup_mutex);
+	return 0;
+
+fail_disconnect:
+	vboxsf_disconnect();
+fail_free_cache:
+	kmem_cache_destroy(vboxsf_inode_cachep);
+fail_nomem:
+	mutex_unlock(&vboxsf_setup_mutex);
+	return err;
+}
+
+static int vboxsf_parse_monolithic(struct fs_context *fc, void *data)
+{
+	char *options = data;
+
+	if (options && options[0] == VBSF_MOUNT_SIGNATURE_BYTE_0 &&
+		       options[1] == VBSF_MOUNT_SIGNATURE_BYTE_1 &&
+		       options[2] == VBSF_MOUNT_SIGNATURE_BYTE_2 &&
+		       options[3] == VBSF_MOUNT_SIGNATURE_BYTE_3) {
+		vbg_err("vboxsf: Old binary mount data not supported, remove obsolete mount.vboxsf and/or update your VBoxService.\n");
+		return -EINVAL;
+	}
+
+	return generic_parse_monolithic(fc, data);
+}
+
+static int vboxsf_get_tree(struct fs_context *fc)
+{
+	int err;
+
+	err = vboxsf_setup();
+	if (err)
+		return err;
+
+	return get_tree_nodev(fc, vboxsf_fill_super);
+}
+
+static int vboxsf_reconfigure(struct fs_context *fc)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(fc->root->d_sb);
+	struct vboxsf_fs_context *ctx = fc->fs_private;
+	struct inode *iroot = fc->root->d_sb->s_root->d_inode;
+
+	/* Apply changed options to the root inode */
+	sbi->o = ctx->o;
+	vboxsf_init_inode(sbi, iroot, &sbi->root_info);
+
+	return 0;
+}
+
+static void vboxsf_free_fc(struct fs_context *fc)
+{
+	struct vboxsf_fs_context *ctx = fc->fs_private;
+
+	kfree(ctx->nls_name);
+	kfree(ctx);
+}
+
+static const struct fs_context_operations vboxsf_context_ops = {
+	.free			= vboxsf_free_fc,
+	.parse_param		= vboxsf_parse_param,
+	.parse_monolithic	= vboxsf_parse_monolithic,
+	.get_tree		= vboxsf_get_tree,
+	.reconfigure		= vboxsf_reconfigure,
+};
+
+static int vboxsf_init_fs_context(struct fs_context *fc)
+{
+	struct vboxsf_fs_context *ctx;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	current_uid_gid(&ctx->o.uid, &ctx->o.gid);
+
+	fc->fs_private = ctx;
+	fc->ops = &vboxsf_context_ops;
+	return 0;
+}
+
+static struct file_system_type vboxsf_fs_type = {
+	.owner			= THIS_MODULE,
+	.name			= "vboxsf",
+	.init_fs_context	= vboxsf_init_fs_context,
+	.kill_sb		= kill_anon_super
+};
+
+/* Module initialization/finalization handlers */
+static int __init vboxsf_init(void)
+{
+	return register_filesystem(&vboxsf_fs_type);
+}
+
+static void __exit vboxsf_fini(void)
+{
+	unregister_filesystem(&vboxsf_fs_type);
+
+	mutex_lock(&vboxsf_setup_mutex);
+	if (vboxsf_setup_done) {
+		vboxsf_disconnect();
+		/*
+		 * Make sure all delayed rcu free inodes are flushed
+		 * before we destroy the cache.
+		 */
+		rcu_barrier();
+		kmem_cache_destroy(vboxsf_inode_cachep);
+	}
+	mutex_unlock(&vboxsf_setup_mutex);
+}
+
+module_init(vboxsf_init);
+module_exit(vboxsf_fini);
+
+MODULE_DESCRIPTION("Oracle VM VirtualBox Module for Host File System Access");
+MODULE_AUTHOR("Oracle Corporation");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS_FS("vboxsf");
diff --git a/fs/vboxsf/utils.c b/fs/vboxsf/utils.c
new file mode 100644
index 000000000000..96bd160da48b
--- /dev/null
+++ b/fs/vboxsf/utils.c
@@ -0,0 +1,551 @@
+// SPDX-License-Identifier: MIT
+/*
+ * VirtualBox Guest Shared Folders support: Utility functions.
+ * Mainly conversion from/to VirtualBox/Linux data structures.
+ *
+ * Copyright (C) 2006-2018 Oracle Corporation
+ */
+
+#include <linux/namei.h>
+#include <linux/nls.h>
+#include <linux/sizes.h>
+#include <linux/vfs.h>
+#include "vfsmod.h"
+
+struct inode *vboxsf_new_inode(struct super_block *sb)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(sb);
+	struct inode *inode;
+	unsigned long flags;
+	int cursor, ret;
+	u32 gen;
+
+	inode = new_inode(sb);
+	if (!inode)
+		return ERR_PTR(-ENOMEM);
+
+	idr_preload(GFP_KERNEL);
+	spin_lock_irqsave(&sbi->ino_idr_lock, flags);
+	cursor = idr_get_cursor(&sbi->ino_idr);
+	ret = idr_alloc_cyclic(&sbi->ino_idr, inode, 1, 0, GFP_ATOMIC);
+	if (ret >= 0 && ret < cursor)
+		sbi->next_generation++;
+	gen = sbi->next_generation;
+	spin_unlock_irqrestore(&sbi->ino_idr_lock, flags);
+	idr_preload_end();
+
+	if (ret < 0) {
+		iput(inode);
+		return ERR_PTR(ret);
+	}
+
+	inode->i_ino = ret;
+	inode->i_generation = gen;
+	return inode;
+}
+
+/* set [inode] attributes based on [info], uid/gid based on [sbi] */
+void vboxsf_init_inode(struct vboxsf_sbi *sbi, struct inode *inode,
+		       const struct shfl_fsobjinfo *info)
+{
+	const struct shfl_fsobjattr *attr;
+	s64 allocated;
+	int mode;
+
+	attr = &info->attr;
+
+#define mode_set(r) ((attr->mode & (SHFL_UNIX_##r)) ? (S_##r) : 0)
+
+	mode = mode_set(IRUSR);
+	mode |= mode_set(IWUSR);
+	mode |= mode_set(IXUSR);
+
+	mode |= mode_set(IRGRP);
+	mode |= mode_set(IWGRP);
+	mode |= mode_set(IXGRP);
+
+	mode |= mode_set(IROTH);
+	mode |= mode_set(IWOTH);
+	mode |= mode_set(IXOTH);
+
+#undef mode_set
+
+	/* We use the host-side values for these */
+	inode->i_flags |= S_NOATIME | S_NOCMTIME;
+	inode->i_mapping->a_ops = &vboxsf_reg_aops;
+
+	if (SHFL_IS_DIRECTORY(attr->mode)) {
+		inode->i_mode = sbi->o.dmode_set ? sbi->o.dmode : mode;
+		inode->i_mode &= ~sbi->o.dmask;
+		inode->i_mode |= S_IFDIR;
+		inode->i_op = &vboxsf_dir_iops;
+		inode->i_fop = &vboxsf_dir_fops;
+		/*
+		 * XXX: this probably should be set to the number of entries
+		 * in the directory plus two (. ..)
+		 */
+		set_nlink(inode, 1);
+	} else if (SHFL_IS_SYMLINK(attr->mode)) {
+		inode->i_mode = sbi->o.fmode_set ? sbi->o.fmode : mode;
+		inode->i_mode &= ~sbi->o.fmask;
+		inode->i_mode |= S_IFLNK;
+		inode->i_op = &vboxsf_lnk_iops;
+		set_nlink(inode, 1);
+	} else {
+		inode->i_mode = sbi->o.fmode_set ? sbi->o.fmode : mode;
+		inode->i_mode &= ~sbi->o.fmask;
+		inode->i_mode |= S_IFREG;
+		inode->i_op = &vboxsf_reg_iops;
+		inode->i_fop = &vboxsf_reg_fops;
+		set_nlink(inode, 1);
+	}
+
+	inode->i_uid = sbi->o.uid;
+	inode->i_gid = sbi->o.gid;
+
+	inode->i_size = info->size;
+	inode->i_blkbits = 12;
+	/* i_blocks always in units of 512 bytes! */
+	allocated = info->allocated + 511;
+	do_div(allocated, 512);
+	inode->i_blocks = allocated;
+
+	inode->i_atime = ns_to_timespec64(
+				 info->access_time.ns_relative_to_unix_epoch);
+	inode->i_ctime = ns_to_timespec64(
+				 info->change_time.ns_relative_to_unix_epoch);
+	inode->i_mtime = ns_to_timespec64(
+			   info->modification_time.ns_relative_to_unix_epoch);
+}
+
+int vboxsf_create_at_dentry(struct dentry *dentry,
+			    struct shfl_createparms *params)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(dentry->d_sb);
+	struct shfl_string *path;
+	int err;
+
+	path = vboxsf_path_from_dentry(sbi, dentry);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	err = vboxsf_create(sbi->root, path, params);
+	__putname(path);
+
+	return err;
+}
+
+int vboxsf_stat(struct vboxsf_sbi *sbi, struct shfl_string *path,
+		struct shfl_fsobjinfo *info)
+{
+	struct shfl_createparms params = {};
+	int err;
+
+	params.handle = SHFL_HANDLE_NIL;
+	params.create_flags = SHFL_CF_LOOKUP | SHFL_CF_ACT_FAIL_IF_NEW;
+
+	err = vboxsf_create(sbi->root, path, &params);
+	if (err)
+		return err;
+
+	if (params.result != SHFL_FILE_EXISTS)
+		return -ENOENT;
+
+	if (info)
+		*info = params.info;
+
+	return 0;
+}
+
+int vboxsf_stat_dentry(struct dentry *dentry, struct shfl_fsobjinfo *info)
+{
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(dentry->d_sb);
+	struct shfl_string *path;
+	int err;
+
+	path = vboxsf_path_from_dentry(sbi, dentry);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	err = vboxsf_stat(sbi, path, info);
+	__putname(path);
+	return err;
+}
+
+int vboxsf_inode_revalidate(struct dentry *dentry)
+{
+	struct vboxsf_sbi *sbi;
+	struct vboxsf_inode *sf_i;
+	struct shfl_fsobjinfo info;
+	struct timespec64 prev_mtime;
+	struct inode *inode;
+	int err;
+
+	if (!dentry || !d_really_is_positive(dentry))
+		return -EINVAL;
+
+	inode = d_inode(dentry);
+	prev_mtime = inode->i_mtime;
+	sf_i = VBOXSF_I(inode);
+	sbi = VBOXSF_SBI(dentry->d_sb);
+	if (!sf_i->force_restat) {
+		if (time_before(jiffies, dentry->d_time + sbi->o.ttl))
+			return 0;
+	}
+
+	err = vboxsf_stat_dentry(dentry, &info);
+	if (err)
+		return err;
+
+	dentry->d_time = jiffies;
+	sf_i->force_restat = 0;
+	vboxsf_init_inode(sbi, inode, &info);
+
+	/*
+	 * If the file was changed on the host side we need to invalidate the
+	 * page-cache for it.  Note this also gets triggered by our own writes,
+	 * this is unavoidable.
+	 */
+	if (timespec64_compare(&inode->i_mtime, &prev_mtime) > 0)
+		invalidate_inode_pages2(inode->i_mapping);
+
+	return 0;
+}
+
+int vboxsf_getattr(const struct path *path, struct kstat *kstat,
+		   u32 request_mask, unsigned int flags)
+{
+	int err;
+	struct dentry *dentry = path->dentry;
+	struct inode *inode = d_inode(dentry);
+	struct vboxsf_inode *sf_i = VBOXSF_I(inode);
+
+	switch (flags & AT_STATX_SYNC_TYPE) {
+	case AT_STATX_DONT_SYNC:
+		err = 0;
+		break;
+	case AT_STATX_FORCE_SYNC:
+		sf_i->force_restat = 1;
+		/* fall-through */
+	default:
+		err = vboxsf_inode_revalidate(dentry);
+	}
+	if (err)
+		return err;
+
+	generic_fillattr(d_inode(dentry), kstat);
+	return 0;
+}
+
+int vboxsf_setattr(struct dentry *dentry, struct iattr *iattr)
+{
+	struct vboxsf_inode *sf_i = VBOXSF_I(d_inode(dentry));
+	struct vboxsf_sbi *sbi = VBOXSF_SBI(dentry->d_sb);
+	struct shfl_createparms params = {};
+	struct shfl_fsobjinfo info = {};
+	u32 buf_len;
+	int err;
+
+	params.handle = SHFL_HANDLE_NIL;
+	params.create_flags = SHFL_CF_ACT_OPEN_IF_EXISTS |
+			      SHFL_CF_ACT_FAIL_IF_NEW |
+			      SHFL_CF_ACCESS_ATTR_WRITE;
+
+	/* this is at least required for Posix hosts */
+	if (iattr->ia_valid & ATTR_SIZE)
+		params.create_flags |= SHFL_CF_ACCESS_WRITE;
+
+	err = vboxsf_create_at_dentry(dentry, &params);
+	if (err || params.result != SHFL_FILE_EXISTS)
+		return err ? err : -ENOENT;
+
+#define mode_set(r) ((iattr->ia_mode & (S_##r)) ? SHFL_UNIX_##r : 0)
+
+	/*
+	 * Setting the file size and setting the other attributes has to
+	 * be handled separately.
+	 */
+	if (iattr->ia_valid & (ATTR_MODE | ATTR_ATIME | ATTR_MTIME)) {
+		if (iattr->ia_valid & ATTR_MODE) {
+			info.attr.mode = mode_set(IRUSR);
+			info.attr.mode |= mode_set(IWUSR);
+			info.attr.mode |= mode_set(IXUSR);
+			info.attr.mode |= mode_set(IRGRP);
+			info.attr.mode |= mode_set(IWGRP);
+			info.attr.mode |= mode_set(IXGRP);
+			info.attr.mode |= mode_set(IROTH);
+			info.attr.mode |= mode_set(IWOTH);
+			info.attr.mode |= mode_set(IXOTH);
+
+			if (iattr->ia_mode & S_IFDIR)
+				info.attr.mode |= SHFL_TYPE_DIRECTORY;
+			else
+				info.attr.mode |= SHFL_TYPE_FILE;
+		}
+
+		if (iattr->ia_valid & ATTR_ATIME)
+			info.access_time.ns_relative_to_unix_epoch =
+					    timespec64_to_ns(&iattr->ia_atime);
+
+		if (iattr->ia_valid & ATTR_MTIME)
+			info.modification_time.ns_relative_to_unix_epoch =
+					    timespec64_to_ns(&iattr->ia_mtime);
+
+		/*
+		 * Ignore ctime (inode change time) as it can't be set
+		 * from userland anyway.
+		 */
+
+		buf_len = sizeof(info);
+		err = vboxsf_fsinfo(sbi->root, params.handle,
+				    SHFL_INFO_SET | SHFL_INFO_FILE, &buf_len,
+				    &info);
+		if (err) {
+			vboxsf_close(sbi->root, params.handle);
+			return err;
+		}
+
+		/* the host may have given us different attr then requested */
+		sf_i->force_restat = 1;
+	}
+
+#undef mode_set
+
+	if (iattr->ia_valid & ATTR_SIZE) {
+		memset(&info, 0, sizeof(info));
+		info.size = iattr->ia_size;
+		buf_len = sizeof(info);
+		err = vboxsf_fsinfo(sbi->root, params.handle,
+				    SHFL_INFO_SET | SHFL_INFO_SIZE, &buf_len,
+				    &info);
+		if (err) {
+			vboxsf_close(sbi->root, params.handle);
+			return err;
+		}
+
+		/* the host may have given us different attr then requested */
+		sf_i->force_restat = 1;
+	}
+
+	vboxsf_close(sbi->root, params.handle);
+
+	/* Update the inode with what the host has actually given us. */
+	if (sf_i->force_restat)
+		vboxsf_inode_revalidate(dentry);
+
+	return 0;
+}
+
+/*
+ * [dentry] contains string encoded in coding system that corresponds
+ * to [sbi]->nls, we must convert it to UTF8 here.
+ * Returns a shfl_string allocated through __getname (must be freed using
+ * __putname), or an ERR_PTR on error.
+ */
+struct shfl_string *vboxsf_path_from_dentry(struct vboxsf_sbi *sbi,
+					    struct dentry *dentry)
+{
+	struct shfl_string *shfl_path;
+	int path_len, out_len, nb;
+	char *buf, *path;
+	wchar_t uni;
+	u8 *out;
+
+	buf = __getname();
+	if (!buf)
+		return ERR_PTR(-ENOMEM);
+
+	path = dentry_path_raw(dentry, buf, PATH_MAX);
+	if (IS_ERR(path)) {
+		__putname(buf);
+		return ERR_CAST(path);
+	}
+	path_len = strlen(path);
+
+	if (sbi->nls) {
+		shfl_path = __getname();
+		if (!shfl_path) {
+			__putname(buf);
+			return ERR_PTR(-ENOMEM);
+		}
+
+		out = shfl_path->string.utf8;
+		out_len = PATH_MAX - SHFLSTRING_HEADER_SIZE - 1;
+
+		while (path_len) {
+			nb = sbi->nls->char2uni(path, path_len, &uni);
+			if (nb < 0) {
+				__putname(shfl_path);
+				__putname(buf);
+				return ERR_PTR(-EINVAL);
+			}
+			path += nb;
+			path_len -= nb;
+
+			nb = utf32_to_utf8(uni, out, out_len);
+			if (nb < 0) {
+				__putname(shfl_path);
+				__putname(buf);
+				return ERR_PTR(-ENAMETOOLONG);
+			}
+			out += nb;
+			out_len -= nb;
+		}
+		*out = 0;
+		shfl_path->length = out - shfl_path->string.utf8;
+		shfl_path->size = shfl_path->length + 1;
+		__putname(buf);
+	} else {
+		if ((SHFLSTRING_HEADER_SIZE + path_len + 1) > PATH_MAX) {
+			__putname(buf);
+			return ERR_PTR(-ENAMETOOLONG);
+		}
+		/*
+		 * dentry_path stores the name at the end of buf, but the
+		 * shfl_string string we return must be properly aligned.
+		 */
+		shfl_path = (struct shfl_string *)buf;
+		memmove(shfl_path->string.utf8, path, path_len);
+		shfl_path->string.utf8[path_len] = 0;
+		shfl_path->length = path_len;
+		shfl_path->size = path_len + 1;
+	}
+
+	return shfl_path;
+}
+
+int vboxsf_nlscpy(struct vboxsf_sbi *sbi, char *name, size_t name_bound_len,
+		  const unsigned char *utf8_name, size_t utf8_len)
+{
+	const char *in;
+	char *out;
+	size_t out_len;
+	size_t out_bound_len;
+	size_t in_bound_len;
+
+	in = utf8_name;
+	in_bound_len = utf8_len;
+
+	out = name;
+	out_len = 0;
+	/* Reserve space for terminating 0 */
+	out_bound_len = name_bound_len - 1;
+
+	while (in_bound_len) {
+		int nb;
+		unicode_t uni;
+
+		nb = utf8_to_utf32(in, in_bound_len, &uni);
+		if (nb < 0)
+			return -EINVAL;
+
+		in += nb;
+		in_bound_len -= nb;
+
+		nb = sbi->nls->uni2char(uni, out, out_bound_len);
+		if (nb < 0)
+			return nb;
+
+		out += nb;
+		out_bound_len -= nb;
+		out_len += nb;
+	}
+
+	*out = 0;
+
+	return 0;
+}
+
+static struct vboxsf_dir_buf *vboxsf_dir_buf_alloc(struct list_head *list)
+{
+	struct vboxsf_dir_buf *b;
+
+	b = kmalloc(sizeof(*b), GFP_KERNEL);
+	if (!b)
+		return NULL;
+
+	b->buf = kmalloc(DIR_BUFFER_SIZE, GFP_KERNEL);
+	if (!b->buf) {
+		kfree(b);
+		return NULL;
+	}
+
+	b->entries = 0;
+	b->used = 0;
+	b->free = DIR_BUFFER_SIZE;
+	list_add(&b->head, list);
+
+	return b;
+}
+
+static void vboxsf_dir_buf_free(struct vboxsf_dir_buf *b)
+{
+	list_del(&b->head);
+	kfree(b->buf);
+	kfree(b);
+}
+
+struct vboxsf_dir_info *vboxsf_dir_info_alloc(void)
+{
+	struct vboxsf_dir_info *p;
+
+	p = kmalloc(sizeof(*p), GFP_KERNEL);
+	if (!p)
+		return NULL;
+
+	INIT_LIST_HEAD(&p->info_list);
+	return p;
+}
+
+void vboxsf_dir_info_free(struct vboxsf_dir_info *p)
+{
+	struct list_head *list, *pos, *tmp;
+
+	list = &p->info_list;
+	list_for_each_safe(pos, tmp, list) {
+		struct vboxsf_dir_buf *b;
+
+		b = list_entry(pos, struct vboxsf_dir_buf, head);
+		vboxsf_dir_buf_free(b);
+	}
+	kfree(p);
+}
+
+int vboxsf_dir_read_all(struct vboxsf_sbi *sbi, struct vboxsf_dir_info *sf_d,
+			u64 handle)
+{
+	struct vboxsf_dir_buf *b;
+	u32 entries, size;
+	int err = 0;
+	void *buf;
+
+	/* vboxsf_dirinfo returns 1 on end of dir */
+	while (err == 0) {
+		b = vboxsf_dir_buf_alloc(&sf_d->info_list);
+		if (!b) {
+			err = -ENOMEM;
+			break;
+		}
+
+		buf = b->buf;
+		size = b->free;
+
+		err = vboxsf_dirinfo(sbi->root, handle, NULL, 0, 0,
+				     &size, buf, &entries);
+		if (err < 0)
+			break;
+
+		b->entries += entries;
+		b->free -= size;
+		b->used += size;
+	}
+
+	if (b && b->used == 0)
+		vboxsf_dir_buf_free(b);
+
+	/* -EILSEQ means the host could not translate a filename, ignore */
+	if (err > 0 || err == -EILSEQ)
+		err = 0;
+
+	return err;
+}
diff --git a/fs/vboxsf/vboxsf_wrappers.c b/fs/vboxsf/vboxsf_wrappers.c
new file mode 100644
index 000000000000..bfc78a097dae
--- /dev/null
+++ b/fs/vboxsf/vboxsf_wrappers.c
@@ -0,0 +1,371 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Wrapper functions for the shfl host calls.
+ *
+ * Copyright (C) 2006-2018 Oracle Corporation
+ */
+
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/vbox_err.h>
+#include <linux/vbox_utils.h>
+#include "vfsmod.h"
+
+#define SHFL_REQUEST \
+	(VMMDEV_REQUESTOR_KERNEL | VMMDEV_REQUESTOR_USR_DRV_OTHER | \
+	 VMMDEV_REQUESTOR_CON_DONT_KNOW | VMMDEV_REQUESTOR_TRUST_NOT_GIVEN)
+
+static u32 vboxsf_client_id;
+
+int vboxsf_connect(void)
+{
+	struct vbg_dev *gdev;
+	struct vmmdev_hgcm_service_location loc;
+	int err, vbox_status;
+
+	loc.type = VMMDEV_HGCM_LOC_LOCALHOST_EXISTING;
+	strcpy(loc.u.localhost.service_name, "VBoxSharedFolders");
+
+	gdev = vbg_get_gdev();
+	if (IS_ERR(gdev))
+		return -ENODEV;	/* No guest-device */
+
+	err = vbg_hgcm_connect(gdev, SHFL_REQUEST, &loc,
+			       &vboxsf_client_id, &vbox_status);
+	vbg_put_gdev(gdev);
+
+	return err ? err : vbg_status_code_to_errno(vbox_status);
+}
+
+void vboxsf_disconnect(void)
+{
+	struct vbg_dev *gdev;
+	int vbox_status;
+
+	gdev = vbg_get_gdev();
+	if (IS_ERR(gdev))
+		return;   /* guest-device is gone, already disconnected */
+
+	vbg_hgcm_disconnect(gdev, SHFL_REQUEST, vboxsf_client_id, &vbox_status);
+	vbg_put_gdev(gdev);
+}
+
+static int vboxsf_call(u32 function, void *parms, u32 parm_count, int *status)
+{
+	struct vbg_dev *gdev;
+	int err, vbox_status;
+
+	gdev = vbg_get_gdev();
+	if (IS_ERR(gdev))
+		return -ESHUTDOWN; /* guest-dev removed underneath us */
+
+	err = vbg_hgcm_call(gdev, SHFL_REQUEST, vboxsf_client_id, function,
+			    U32_MAX, parms, parm_count, &vbox_status);
+	vbg_put_gdev(gdev);
+
+	if (err < 0)
+		return err;
+
+	if (status)
+		*status = vbox_status;
+
+	return vbg_status_code_to_errno(vbox_status);
+}
+
+int vboxsf_map_folder(struct shfl_string *folder_name, u32 *root)
+{
+	struct shfl_map_folder parms;
+	int err, status;
+
+	parms.path.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL;
+	parms.path.u.pointer.size = shfl_string_buf_size(folder_name);
+	parms.path.u.pointer.u.linear_addr = (uintptr_t)folder_name;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = 0;
+
+	parms.delimiter.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.delimiter.u.value32 = '/';
+
+	parms.case_sensitive.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.case_sensitive.u.value32 = 1;
+
+	err = vboxsf_call(SHFL_FN_MAP_FOLDER, &parms, SHFL_CPARMS_MAP_FOLDER,
+			  &status);
+	if (err == -ENOSYS && status == VERR_NOT_IMPLEMENTED)
+		vbg_err("%s: Error host is too old\n", __func__);
+
+	*root = parms.root.u.value32;
+	return err;
+}
+
+int vboxsf_unmap_folder(u32 root)
+{
+	struct shfl_unmap_folder parms;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	return vboxsf_call(SHFL_FN_UNMAP_FOLDER, &parms,
+			   SHFL_CPARMS_UNMAP_FOLDER, NULL);
+}
+
+/**
+ * vboxsf_create - Create a new file or folder
+ * @root:         Root of the shared folder in which to create the file
+ * @parsed_path:  The path of the file or folder relative to the shared folder
+ * @param:        create_parms Parameters for file/folder creation.
+ *
+ * Create a new file or folder or open an existing one in a shared folder.
+ * Note this function always returns 0 / success unless an exceptional condition
+ * occurs - out of memory, invalid arguments, etc. If the file or folder could
+ * not be opened or created, create_parms->handle will be set to
+ * SHFL_HANDLE_NIL on return.  In this case the value in create_parms->result
+ * provides information as to why (e.g. SHFL_FILE_EXISTS), create_parms->result
+ * is also set on success as additional information.
+ *
+ * Returns:
+ * 0 or negative errno value.
+ */
+int vboxsf_create(u32 root, struct shfl_string *parsed_path,
+		  struct shfl_createparms *create_parms)
+{
+	struct shfl_create parms;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.path.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL;
+	parms.path.u.pointer.size = shfl_string_buf_size(parsed_path);
+	parms.path.u.pointer.u.linear_addr = (uintptr_t)parsed_path;
+
+	parms.parms.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL;
+	parms.parms.u.pointer.size = sizeof(struct shfl_createparms);
+	parms.parms.u.pointer.u.linear_addr = (uintptr_t)create_parms;
+
+	return vboxsf_call(SHFL_FN_CREATE, &parms, SHFL_CPARMS_CREATE, NULL);
+}
+
+int vboxsf_close(u32 root, u64 handle)
+{
+	struct shfl_close parms;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.handle.type = VMMDEV_HGCM_PARM_TYPE_64BIT;
+	parms.handle.u.value64 = handle;
+
+	return vboxsf_call(SHFL_FN_CLOSE, &parms, SHFL_CPARMS_CLOSE, NULL);
+}
+
+int vboxsf_remove(u32 root, struct shfl_string *parsed_path, u32 flags)
+{
+	struct shfl_remove parms;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.path.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_IN;
+	parms.path.u.pointer.size = shfl_string_buf_size(parsed_path);
+	parms.path.u.pointer.u.linear_addr = (uintptr_t)parsed_path;
+
+	parms.flags.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.flags.u.value32 = flags;
+
+	return vboxsf_call(SHFL_FN_REMOVE, &parms, SHFL_CPARMS_REMOVE, NULL);
+}
+
+int vboxsf_rename(u32 root, struct shfl_string *src_path,
+		  struct shfl_string *dest_path, u32 flags)
+{
+	struct shfl_rename parms;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.src.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_IN;
+	parms.src.u.pointer.size = shfl_string_buf_size(src_path);
+	parms.src.u.pointer.u.linear_addr = (uintptr_t)src_path;
+
+	parms.dest.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_IN;
+	parms.dest.u.pointer.size = shfl_string_buf_size(dest_path);
+	parms.dest.u.pointer.u.linear_addr = (uintptr_t)dest_path;
+
+	parms.flags.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.flags.u.value32 = flags;
+
+	return vboxsf_call(SHFL_FN_RENAME, &parms, SHFL_CPARMS_RENAME, NULL);
+}
+
+int vboxsf_read(u32 root, u64 handle, u64 offset, u32 *buf_len, u8 *buf)
+{
+	struct shfl_read parms;
+	int err;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.handle.type = VMMDEV_HGCM_PARM_TYPE_64BIT;
+	parms.handle.u.value64 = handle;
+	parms.offset.type = VMMDEV_HGCM_PARM_TYPE_64BIT;
+	parms.offset.u.value64 = offset;
+	parms.cb.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.cb.u.value32 = *buf_len;
+	parms.buffer.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_OUT;
+	parms.buffer.u.pointer.size = *buf_len;
+	parms.buffer.u.pointer.u.linear_addr = (uintptr_t)buf;
+
+	err = vboxsf_call(SHFL_FN_READ, &parms, SHFL_CPARMS_READ, NULL);
+
+	*buf_len = parms.cb.u.value32;
+	return err;
+}
+
+int vboxsf_write(u32 root, u64 handle, u64 offset, u32 *buf_len, u8 *buf)
+{
+	struct shfl_write parms;
+	int err;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.handle.type = VMMDEV_HGCM_PARM_TYPE_64BIT;
+	parms.handle.u.value64 = handle;
+	parms.offset.type = VMMDEV_HGCM_PARM_TYPE_64BIT;
+	parms.offset.u.value64 = offset;
+	parms.cb.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.cb.u.value32 = *buf_len;
+	parms.buffer.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_IN;
+	parms.buffer.u.pointer.size = *buf_len;
+	parms.buffer.u.pointer.u.linear_addr = (uintptr_t)buf;
+
+	err = vboxsf_call(SHFL_FN_WRITE, &parms, SHFL_CPARMS_WRITE, NULL);
+
+	*buf_len = parms.cb.u.value32;
+	return err;
+}
+
+/* Returns 0 on success, 1 on end-of-dir, negative errno otherwise */
+int vboxsf_dirinfo(u32 root, u64 handle,
+		   struct shfl_string *parsed_path, u32 flags, u32 index,
+		   u32 *buf_len, struct shfl_dirinfo *buf, u32 *file_count)
+{
+	struct shfl_list parms;
+	int err, status;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.handle.type = VMMDEV_HGCM_PARM_TYPE_64BIT;
+	parms.handle.u.value64 = handle;
+	parms.flags.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.flags.u.value32 = flags;
+	parms.cb.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.cb.u.value32 = *buf_len;
+	if (parsed_path) {
+		parms.path.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_IN;
+		parms.path.u.pointer.size = shfl_string_buf_size(parsed_path);
+		parms.path.u.pointer.u.linear_addr = (uintptr_t)parsed_path;
+	} else {
+		parms.path.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_IN;
+		parms.path.u.pointer.size = 0;
+		parms.path.u.pointer.u.linear_addr = 0;
+	}
+
+	parms.buffer.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_OUT;
+	parms.buffer.u.pointer.size = *buf_len;
+	parms.buffer.u.pointer.u.linear_addr = (uintptr_t)buf;
+
+	parms.resume_point.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.resume_point.u.value32 = index;
+	parms.file_count.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.file_count.u.value32 = 0;	/* out parameter only */
+
+	err = vboxsf_call(SHFL_FN_LIST, &parms, SHFL_CPARMS_LIST, &status);
+	if (err == -ENODATA && status == VERR_NO_MORE_FILES)
+		err = 1;
+
+	*buf_len = parms.cb.u.value32;
+	*file_count = parms.file_count.u.value32;
+	return err;
+}
+
+int vboxsf_fsinfo(u32 root, u64 handle, u32 flags,
+		  u32 *buf_len, void *buf)
+{
+	struct shfl_information parms;
+	int err;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.handle.type = VMMDEV_HGCM_PARM_TYPE_64BIT;
+	parms.handle.u.value64 = handle;
+	parms.flags.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.flags.u.value32 = flags;
+	parms.cb.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.cb.u.value32 = *buf_len;
+	parms.info.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL;
+	parms.info.u.pointer.size = *buf_len;
+	parms.info.u.pointer.u.linear_addr = (uintptr_t)buf;
+
+	err = vboxsf_call(SHFL_FN_INFORMATION, &parms, SHFL_CPARMS_INFORMATION,
+			  NULL);
+
+	*buf_len = parms.cb.u.value32;
+	return err;
+}
+
+int vboxsf_readlink(u32 root, struct shfl_string *parsed_path,
+		    u32 buf_len, u8 *buf)
+{
+	struct shfl_readLink parms;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.path.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_IN;
+	parms.path.u.pointer.size = shfl_string_buf_size(parsed_path);
+	parms.path.u.pointer.u.linear_addr = (uintptr_t)parsed_path;
+
+	parms.buffer.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_OUT;
+	parms.buffer.u.pointer.size = buf_len;
+	parms.buffer.u.pointer.u.linear_addr = (uintptr_t)buf;
+
+	return vboxsf_call(SHFL_FN_READLINK, &parms, SHFL_CPARMS_READLINK,
+			   NULL);
+}
+
+int vboxsf_symlink(u32 root, struct shfl_string *new_path,
+		   struct shfl_string *old_path, struct shfl_fsobjinfo *buf)
+{
+	struct shfl_symlink parms;
+
+	parms.root.type = VMMDEV_HGCM_PARM_TYPE_32BIT;
+	parms.root.u.value32 = root;
+
+	parms.new_path.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_IN;
+	parms.new_path.u.pointer.size = shfl_string_buf_size(new_path);
+	parms.new_path.u.pointer.u.linear_addr = (uintptr_t)new_path;
+
+	parms.old_path.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_IN;
+	parms.old_path.u.pointer.size = shfl_string_buf_size(old_path);
+	parms.old_path.u.pointer.u.linear_addr = (uintptr_t)old_path;
+
+	parms.info.type = VMMDEV_HGCM_PARM_TYPE_LINADDR_KERNEL_OUT;
+	parms.info.u.pointer.size = sizeof(struct shfl_fsobjinfo);
+	parms.info.u.pointer.u.linear_addr = (uintptr_t)buf;
+
+	return vboxsf_call(SHFL_FN_SYMLINK, &parms, SHFL_CPARMS_SYMLINK, NULL);
+}
+
+int vboxsf_set_utf8(void)
+{
+	return vboxsf_call(SHFL_FN_SET_UTF8, NULL, 0, NULL);
+}
+
+int vboxsf_set_symlinks(void)
+{
+	return vboxsf_call(SHFL_FN_SET_SYMLINKS, NULL, 0, NULL);
+}
diff --git a/fs/vboxsf/vfsmod.h b/fs/vboxsf/vfsmod.h
new file mode 100644
index 000000000000..18f95b00fc33
--- /dev/null
+++ b/fs/vboxsf/vfsmod.h
@@ -0,0 +1,137 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * VirtualBox Guest Shared Folders support: module header.
+ *
+ * Copyright (C) 2006-2018 Oracle Corporation
+ */
+
+#ifndef VFSMOD_H
+#define VFSMOD_H
+
+#include <linux/backing-dev.h>
+#include <linux/idr.h>
+#include "shfl_hostintf.h"
+
+#define DIR_BUFFER_SIZE SZ_16K
+
+/* The cast is to prevent assignment of void * to pointers of arbitrary type */
+#define VBOXSF_SBI(sb)	((struct vboxsf_sbi *)(sb)->s_fs_info)
+#define VBOXSF_I(i)	container_of(i, struct vboxsf_inode, vfs_inode)
+
+struct vboxsf_options {
+	unsigned long ttl;
+	kuid_t uid;
+	kgid_t gid;
+	bool dmode_set;
+	bool fmode_set;
+	umode_t dmode;
+	umode_t fmode;
+	umode_t dmask;
+	umode_t fmask;
+};
+
+struct vboxsf_fs_context {
+	struct vboxsf_options o;
+	char *nls_name;
+};
+
+/* per-shared folder information */
+struct vboxsf_sbi {
+	struct vboxsf_options o;
+	struct shfl_fsobjinfo root_info;
+	struct idr ino_idr;
+	spinlock_t ino_idr_lock; /* This protects ino_idr */
+	struct nls_table *nls;
+	u32 next_generation;
+	u32 root;
+	int bdi_id;
+};
+
+/* per-inode information */
+struct vboxsf_inode {
+	/* some information was changed, update data on next revalidate */
+	int force_restat;
+	/* list of open handles for this inode + lock protecting it */
+	struct list_head handle_list;
+	/* This mutex protects handle_list accesses */
+	struct mutex handle_list_mutex;
+	/* The VFS inode struct */
+	struct inode vfs_inode;
+};
+
+struct vboxsf_dir_info {
+	struct list_head info_list;
+};
+
+struct vboxsf_dir_buf {
+	size_t entries;
+	size_t free;
+	size_t used;
+	void *buf;
+	struct list_head head;
+};
+
+/* globals */
+extern const struct inode_operations vboxsf_dir_iops;
+extern const struct inode_operations vboxsf_lnk_iops;
+extern const struct inode_operations vboxsf_reg_iops;
+extern const struct file_operations vboxsf_dir_fops;
+extern const struct file_operations vboxsf_reg_fops;
+extern const struct address_space_operations vboxsf_reg_aops;
+extern const struct dentry_operations vboxsf_dentry_ops;
+
+/* from utils.c */
+struct inode *vboxsf_new_inode(struct super_block *sb);
+void vboxsf_init_inode(struct vboxsf_sbi *sbi, struct inode *inode,
+		       const struct shfl_fsobjinfo *info);
+int vboxsf_create_at_dentry(struct dentry *dentry,
+			    struct shfl_createparms *params);
+int vboxsf_stat(struct vboxsf_sbi *sbi, struct shfl_string *path,
+		struct shfl_fsobjinfo *info);
+int vboxsf_stat_dentry(struct dentry *dentry, struct shfl_fsobjinfo *info);
+int vboxsf_inode_revalidate(struct dentry *dentry);
+int vboxsf_getattr(const struct path *path, struct kstat *kstat,
+		   u32 request_mask, unsigned int query_flags);
+int vboxsf_setattr(struct dentry *dentry, struct iattr *iattr);
+struct shfl_string *vboxsf_path_from_dentry(struct vboxsf_sbi *sbi,
+					    struct dentry *dentry);
+int vboxsf_nlscpy(struct vboxsf_sbi *sbi, char *name, size_t name_bound_len,
+		  const unsigned char *utf8_name, size_t utf8_len);
+struct vboxsf_dir_info *vboxsf_dir_info_alloc(void);
+void vboxsf_dir_info_free(struct vboxsf_dir_info *p);
+int vboxsf_dir_read_all(struct vboxsf_sbi *sbi, struct vboxsf_dir_info *sf_d,
+			u64 handle);
+
+/* from vboxsf_wrappers.c */
+int vboxsf_connect(void);
+void vboxsf_disconnect(void);
+
+int vboxsf_create(u32 root, struct shfl_string *parsed_path,
+		  struct shfl_createparms *create_parms);
+
+int vboxsf_close(u32 root, u64 handle);
+int vboxsf_remove(u32 root, struct shfl_string *parsed_path, u32 flags);
+int vboxsf_rename(u32 root, struct shfl_string *src_path,
+		  struct shfl_string *dest_path, u32 flags);
+
+int vboxsf_read(u32 root, u64 handle, u64 offset, u32 *buf_len, u8 *buf);
+int vboxsf_write(u32 root, u64 handle, u64 offset, u32 *buf_len, u8 *buf);
+
+int vboxsf_dirinfo(u32 root, u64 handle,
+		   struct shfl_string *parsed_path, u32 flags, u32 index,
+		   u32 *buf_len, struct shfl_dirinfo *buf, u32 *file_count);
+int vboxsf_fsinfo(u32 root, u64 handle, u32 flags,
+		  u32 *buf_len, void *buf);
+
+int vboxsf_map_folder(struct shfl_string *folder_name, u32 *root);
+int vboxsf_unmap_folder(u32 root);
+
+int vboxsf_readlink(u32 root, struct shfl_string *parsed_path,
+		    u32 buf_len, u8 *buf);
+int vboxsf_symlink(u32 root, struct shfl_string *new_path,
+		   struct shfl_string *old_path, struct shfl_fsobjinfo *buf);
+
+int vboxsf_set_utf8(void);
+int vboxsf_set_symlinks(void);
+
+#endif
diff --git a/fs/verity/enable.c b/fs/verity/enable.c
index eabc6ac19906..5ab3bbec8108 100644
--- a/fs/verity/enable.c
+++ b/fs/verity/enable.c
@@ -8,18 +8,48 @@
 #include "fsverity_private.h"
 
 #include <crypto/hash.h>
+#include <linux/backing-dev.h>
 #include <linux/mount.h>
 #include <linux/pagemap.h>
 #include <linux/sched/signal.h>
 #include <linux/uaccess.h>
 
-static int build_merkle_tree_level(struct inode *inode, unsigned int level,
+/*
+ * Read a file data page for Merkle tree construction.  Do aggressive readahead,
+ * since we're sequentially reading the entire file.
+ */
+static struct page *read_file_data_page(struct file *filp, pgoff_t index,
+					struct file_ra_state *ra,
+					unsigned long remaining_pages)
+{
+	struct page *page;
+
+	page = find_get_page_flags(filp->f_mapping, index, FGP_ACCESSED);
+	if (!page || !PageUptodate(page)) {
+		if (page)
+			put_page(page);
+		else
+			page_cache_sync_readahead(filp->f_mapping, ra, filp,
+						  index, remaining_pages);
+		page = read_mapping_page(filp->f_mapping, index, NULL);
+		if (IS_ERR(page))
+			return page;
+	}
+	if (PageReadahead(page))
+		page_cache_async_readahead(filp->f_mapping, ra, filp, page,
+					   index, remaining_pages);
+	return page;
+}
+
+static int build_merkle_tree_level(struct file *filp, unsigned int level,
 				   u64 num_blocks_to_hash,
 				   const struct merkle_tree_params *params,
 				   u8 *pending_hashes,
 				   struct ahash_request *req)
 {
+	struct inode *inode = file_inode(filp);
 	const struct fsverity_operations *vops = inode->i_sb->s_vop;
+	struct file_ra_state ra = { 0 };
 	unsigned int pending_size = 0;
 	u64 dst_block_num;
 	u64 i;
@@ -36,6 +66,8 @@ static int build_merkle_tree_level(struct inode *inode, unsigned int level,
 		dst_block_num = 0; /* unused */
 	}
 
+	file_ra_state_init(&ra, filp->f_mapping);
+
 	for (i = 0; i < num_blocks_to_hash; i++) {
 		struct page *src_page;
 
@@ -45,7 +77,8 @@ static int build_merkle_tree_level(struct inode *inode, unsigned int level,
 
 		if (level == 0) {
 			/* Leaf: hashing a data block */
-			src_page = read_mapping_page(inode->i_mapping, i, NULL);
+			src_page = read_file_data_page(filp, i, &ra,
+						       num_blocks_to_hash - i);
 			if (IS_ERR(src_page)) {
 				err = PTR_ERR(src_page);
 				fsverity_err(inode,
@@ -54,9 +87,14 @@ static int build_merkle_tree_level(struct inode *inode, unsigned int level,
 				return err;
 			}
 		} else {
+			unsigned long num_ra_pages =
+				min_t(unsigned long, num_blocks_to_hash - i,
+				      inode->i_sb->s_bdi->io_pages);
+
 			/* Non-leaf: hashing hash block from level below */
 			src_page = vops->read_merkle_tree_page(inode,
-					params->level_start[level - 1] + i);
+					params->level_start[level - 1] + i,
+					num_ra_pages);
 			if (IS_ERR(src_page)) {
 				err = PTR_ERR(src_page);
 				fsverity_err(inode,
@@ -103,17 +141,18 @@ static int build_merkle_tree_level(struct inode *inode, unsigned int level,
 }
 
 /*
- * Build the Merkle tree for the given inode using the given parameters, and
+ * Build the Merkle tree for the given file using the given parameters, and
  * return the root hash in @root_hash.
  *
  * The tree is written to a filesystem-specific location as determined by the
  * ->write_merkle_tree_block() method.  However, the blocks that comprise the
  * tree are the same for all filesystems.
  */
-static int build_merkle_tree(struct inode *inode,
+static int build_merkle_tree(struct file *filp,
 			     const struct merkle_tree_params *params,
 			     u8 *root_hash)
 {
+	struct inode *inode = file_inode(filp);
 	u8 *pending_hashes;
 	struct ahash_request *req;
 	u64 blocks;
@@ -126,9 +165,11 @@ static int build_merkle_tree(struct inode *inode,
 		return 0;
 	}
 
+	/* This allocation never fails, since it's mempool-backed. */
+	req = fsverity_alloc_hash_request(params->hash_alg, GFP_KERNEL);
+
 	pending_hashes = kmalloc(params->block_size, GFP_KERNEL);
-	req = ahash_request_alloc(params->hash_alg->tfm, GFP_KERNEL);
-	if (!pending_hashes || !req)
+	if (!pending_hashes)
 		goto out;
 
 	/*
@@ -139,7 +180,7 @@ static int build_merkle_tree(struct inode *inode,
 	blocks = (inode->i_size + params->block_size - 1) >>
 		 params->log_blocksize;
 	for (level = 0; level <= params->num_levels; level++) {
-		err = build_merkle_tree_level(inode, level, blocks, params,
+		err = build_merkle_tree_level(filp, level, blocks, params,
 					      pending_hashes, req);
 		if (err)
 			goto out;
@@ -150,7 +191,7 @@ static int build_merkle_tree(struct inode *inode,
 	err = 0;
 out:
 	kfree(pending_hashes);
-	ahash_request_free(req);
+	fsverity_free_hash_request(params->hash_alg, req);
 	return err;
 }
 
@@ -175,8 +216,7 @@ static int enable_verity(struct file *filp,
 
 	/* Get the salt if the user provided one */
 	if (arg->salt_size &&
-	    copy_from_user(desc->salt,
-			   (const u8 __user *)(uintptr_t)arg->salt_ptr,
+	    copy_from_user(desc->salt, u64_to_user_ptr(arg->salt_ptr),
 			   arg->salt_size)) {
 		err = -EFAULT;
 		goto out;
@@ -185,8 +225,7 @@ static int enable_verity(struct file *filp,
 
 	/* Get the signature if the user provided one */
 	if (arg->sig_size &&
-	    copy_from_user(desc->signature,
-			   (const u8 __user *)(uintptr_t)arg->sig_ptr,
+	    copy_from_user(desc->signature, u64_to_user_ptr(arg->sig_ptr),
 			   arg->sig_size)) {
 		err = -EFAULT;
 		goto out;
@@ -227,7 +266,7 @@ static int enable_verity(struct file *filp,
 	 */
 	pr_debug("Building Merkle tree...\n");
 	BUILD_BUG_ON(sizeof(desc->root_hash) < FS_VERITY_MAX_DIGEST_SIZE);
-	err = build_merkle_tree(inode, &params, desc->root_hash);
+	err = build_merkle_tree(filp, &params, desc->root_hash);
 	if (err) {
 		fsverity_err(inode, "Error %d building Merkle tree", err);
 		goto rollback;
@@ -290,6 +329,8 @@ rollback:
 
 /**
  * fsverity_ioctl_enable() - enable verity on a file
+ * @filp: file to enable verity on
+ * @uarg: user pointer to fsverity_enable_arg
  *
  * Enable fs-verity on a file.  See the "FS_IOC_ENABLE_VERITY" section of
  * Documentation/filesystems/fsverity.rst for the documentation.
@@ -315,7 +356,7 @@ int fsverity_ioctl_enable(struct file *filp, const void __user *uarg)
 	if (arg.block_size != PAGE_SIZE)
 		return -EINVAL;
 
-	if (arg.salt_size > FIELD_SIZEOF(struct fsverity_descriptor, salt))
+	if (arg.salt_size > sizeof_field(struct fsverity_descriptor, salt))
 		return -EMSGSIZE;
 
 	if (arg.sig_size > FS_VERITY_MAX_SIGNATURE_SIZE)
diff --git a/fs/verity/fsverity_private.h b/fs/verity/fsverity_private.h
index e74c79b64d88..e96d99d5145e 100644
--- a/fs/verity/fsverity_private.h
+++ b/fs/verity/fsverity_private.h
@@ -16,6 +16,7 @@
 
 #include <crypto/sha.h>
 #include <linux/fsverity.h>
+#include <linux/mempool.h>
 
 struct ahash_request;
 
@@ -37,11 +38,12 @@ struct fsverity_hash_alg {
 	const char *name;	  /* crypto API name, e.g. sha256 */
 	unsigned int digest_size; /* digest size in bytes, e.g. 32 for SHA-256 */
 	unsigned int block_size;  /* block size in bytes, e.g. 64 for SHA-256 */
+	mempool_t req_pool;	  /* mempool with a preallocated hash request */
 };
 
 /* Merkle tree parameters: hash algorithm, initial hash state, and topology */
 struct merkle_tree_params {
-	const struct fsverity_hash_alg *hash_alg; /* the hash algorithm */
+	struct fsverity_hash_alg *hash_alg; /* the hash algorithm */
 	const u8 *hashstate;		/* initial hash state or NULL */
 	unsigned int digest_size;	/* same as hash_alg->digest_size */
 	unsigned int block_size;	/* size of data and tree blocks */
@@ -50,6 +52,7 @@ struct merkle_tree_params {
 	unsigned int log_arity;		/* log2(hashes_per_block) */
 	unsigned int num_levels;	/* number of levels in Merkle tree */
 	u64 tree_size;			/* Merkle tree size in bytes */
+	unsigned long level0_blocks;	/* number of blocks in tree level 0 */
 
 	/*
 	 * Starting block index for each tree level, ordered from leaf level (0)
@@ -58,7 +61,7 @@ struct merkle_tree_params {
 	u64 level_start[FS_VERITY_MAX_LEVELS];
 };
 
-/**
+/*
  * fsverity_info - cached verity metadata for an inode
  *
  * When a verity file is first opened, an instance of this struct is allocated
@@ -114,20 +117,24 @@ struct fsverity_signed_digest {
 
 extern struct fsverity_hash_alg fsverity_hash_algs[];
 
-const struct fsverity_hash_alg *fsverity_get_hash_alg(const struct inode *inode,
-						      unsigned int num);
-const u8 *fsverity_prepare_hash_state(const struct fsverity_hash_alg *alg,
+struct fsverity_hash_alg *fsverity_get_hash_alg(const struct inode *inode,
+						unsigned int num);
+struct ahash_request *fsverity_alloc_hash_request(struct fsverity_hash_alg *alg,
+						  gfp_t gfp_flags);
+void fsverity_free_hash_request(struct fsverity_hash_alg *alg,
+				struct ahash_request *req);
+const u8 *fsverity_prepare_hash_state(struct fsverity_hash_alg *alg,
 				      const u8 *salt, size_t salt_size);
 int fsverity_hash_page(const struct merkle_tree_params *params,
 		       const struct inode *inode,
 		       struct ahash_request *req, struct page *page, u8 *out);
-int fsverity_hash_buffer(const struct fsverity_hash_alg *alg,
+int fsverity_hash_buffer(struct fsverity_hash_alg *alg,
 			 const void *data, size_t size, u8 *out);
 void __init fsverity_check_hash_algs(void);
 
 /* init.c */
 
-extern void __printf(3, 4) __cold
+void __printf(3, 4) __cold
 fsverity_msg(const struct inode *inode, const char *level,
 	     const char *fmt, ...);
 
diff --git a/fs/verity/hash_algs.c b/fs/verity/hash_algs.c
index 31e6d7d2389a..c37e186ebeb6 100644
--- a/fs/verity/hash_algs.c
+++ b/fs/verity/hash_algs.c
@@ -24,6 +24,8 @@ struct fsverity_hash_alg fsverity_hash_algs[] = {
 	},
 };
 
+static DEFINE_MUTEX(fsverity_hash_alg_init_mutex);
+
 /**
  * fsverity_get_hash_alg() - validate and prepare a hash algorithm
  * @inode: optional inode for logging purposes
@@ -36,8 +38,8 @@ struct fsverity_hash_alg fsverity_hash_algs[] = {
  *
  * Return: pointer to the hash alg on success, else an ERR_PTR()
  */
-const struct fsverity_hash_alg *fsverity_get_hash_alg(const struct inode *inode,
-						      unsigned int num)
+struct fsverity_hash_alg *fsverity_get_hash_alg(const struct inode *inode,
+						unsigned int num)
 {
 	struct fsverity_hash_alg *alg;
 	struct crypto_ahash *tfm;
@@ -50,10 +52,15 @@ const struct fsverity_hash_alg *fsverity_get_hash_alg(const struct inode *inode,
 	}
 	alg = &fsverity_hash_algs[num];
 
-	/* pairs with cmpxchg() below */
-	tfm = READ_ONCE(alg->tfm);
-	if (likely(tfm != NULL))
+	/* pairs with smp_store_release() below */
+	if (likely(smp_load_acquire(&alg->tfm) != NULL))
 		return alg;
+
+	mutex_lock(&fsverity_hash_alg_init_mutex);
+
+	if (alg->tfm != NULL)
+		goto out_unlock;
+
 	/*
 	 * Using the shash API would make things a bit simpler, but the ahash
 	 * API is preferable as it allows the use of crypto accelerators.
@@ -64,12 +71,14 @@ const struct fsverity_hash_alg *fsverity_get_hash_alg(const struct inode *inode,
 			fsverity_warn(inode,
 				      "Missing crypto API support for hash algorithm \"%s\"",
 				      alg->name);
-			return ERR_PTR(-ENOPKG);
+			alg = ERR_PTR(-ENOPKG);
+			goto out_unlock;
 		}
 		fsverity_err(inode,
 			     "Error allocating hash algorithm \"%s\": %ld",
 			     alg->name, PTR_ERR(tfm));
-		return ERR_CAST(tfm);
+		alg = ERR_CAST(tfm);
+		goto out_unlock;
 	}
 
 	err = -EINVAL;
@@ -78,18 +87,61 @@ const struct fsverity_hash_alg *fsverity_get_hash_alg(const struct inode *inode,
 	if (WARN_ON(alg->block_size != crypto_ahash_blocksize(tfm)))
 		goto err_free_tfm;
 
+	err = mempool_init_kmalloc_pool(&alg->req_pool, 1,
+					sizeof(struct ahash_request) +
+					crypto_ahash_reqsize(tfm));
+	if (err)
+		goto err_free_tfm;
+
 	pr_info("%s using implementation \"%s\"\n",
 		alg->name, crypto_ahash_driver_name(tfm));
 
-	/* pairs with READ_ONCE() above */
-	if (cmpxchg(&alg->tfm, NULL, tfm) != NULL)
-		crypto_free_ahash(tfm);
-
-	return alg;
+	/* pairs with smp_load_acquire() above */
+	smp_store_release(&alg->tfm, tfm);
+	goto out_unlock;
 
 err_free_tfm:
 	crypto_free_ahash(tfm);
-	return ERR_PTR(err);
+	alg = ERR_PTR(err);
+out_unlock:
+	mutex_unlock(&fsverity_hash_alg_init_mutex);
+	return alg;
+}
+
+/**
+ * fsverity_alloc_hash_request() - allocate a hash request object
+ * @alg: the hash algorithm for which to allocate the request
+ * @gfp_flags: memory allocation flags
+ *
+ * This is mempool-backed, so this never fails if __GFP_DIRECT_RECLAIM is set in
+ * @gfp_flags.  However, in that case this might need to wait for all
+ * previously-allocated requests to be freed.  So to avoid deadlocks, callers
+ * must never need multiple requests at a time to make forward progress.
+ *
+ * Return: the request object on success; NULL on failure (but see above)
+ */
+struct ahash_request *fsverity_alloc_hash_request(struct fsverity_hash_alg *alg,
+						  gfp_t gfp_flags)
+{
+	struct ahash_request *req = mempool_alloc(&alg->req_pool, gfp_flags);
+
+	if (req)
+		ahash_request_set_tfm(req, alg->tfm);
+	return req;
+}
+
+/**
+ * fsverity_free_hash_request() - free a hash request object
+ * @alg: the hash algorithm
+ * @req: the hash request object to free
+ */
+void fsverity_free_hash_request(struct fsverity_hash_alg *alg,
+				struct ahash_request *req)
+{
+	if (req) {
+		ahash_request_zero(req);
+		mempool_free(req, &alg->req_pool);
+	}
 }
 
 /**
@@ -101,7 +153,7 @@ err_free_tfm:
  * Return: NULL if the salt is empty, otherwise the kmalloc()'ed precomputed
  *	   initial hash state on success or an ERR_PTR() on failure.
  */
-const u8 *fsverity_prepare_hash_state(const struct fsverity_hash_alg *alg,
+const u8 *fsverity_prepare_hash_state(struct fsverity_hash_alg *alg,
 				      const u8 *salt, size_t salt_size)
 {
 	u8 *hashstate = NULL;
@@ -119,11 +171,8 @@ const u8 *fsverity_prepare_hash_state(const struct fsverity_hash_alg *alg,
 	if (!hashstate)
 		return ERR_PTR(-ENOMEM);
 
-	req = ahash_request_alloc(alg->tfm, GFP_KERNEL);
-	if (!req) {
-		err = -ENOMEM;
-		goto err_free;
-	}
+	/* This allocation never fails, since it's mempool-backed. */
+	req = fsverity_alloc_hash_request(alg, GFP_KERNEL);
 
 	/*
 	 * Zero-pad the salt to the next multiple of the input size of the hash
@@ -158,7 +207,7 @@ const u8 *fsverity_prepare_hash_state(const struct fsverity_hash_alg *alg,
 	if (err)
 		goto err_free;
 out:
-	ahash_request_free(req);
+	fsverity_free_hash_request(alg, req);
 	kfree(padded_salt);
 	return hashstate;
 
@@ -229,7 +278,7 @@ int fsverity_hash_page(const struct merkle_tree_params *params,
  *
  * Return: 0 on success, -errno on failure
  */
-int fsverity_hash_buffer(const struct fsverity_hash_alg *alg,
+int fsverity_hash_buffer(struct fsverity_hash_alg *alg,
 			 const void *data, size_t size, u8 *out)
 {
 	struct ahash_request *req;
@@ -237,9 +286,8 @@ int fsverity_hash_buffer(const struct fsverity_hash_alg *alg,
 	DECLARE_CRYPTO_WAIT(wait);
 	int err;
 
-	req = ahash_request_alloc(alg->tfm, GFP_KERNEL);
-	if (!req)
-		return -ENOMEM;
+	/* This allocation never fails, since it's mempool-backed. */
+	req = fsverity_alloc_hash_request(alg, GFP_KERNEL);
 
 	sg_init_one(&sg, data, size);
 	ahash_request_set_callback(req, CRYPTO_TFM_REQ_MAY_SLEEP |
@@ -249,7 +297,7 @@ int fsverity_hash_buffer(const struct fsverity_hash_alg *alg,
 
 	err = crypto_wait_req(crypto_ahash_digest(req), &wait);
 
-	ahash_request_free(req);
+	fsverity_free_hash_request(alg, req);
 	return err;
 }
 
diff --git a/fs/verity/measure.c b/fs/verity/measure.c
index 05049b68c745..df409a5682ed 100644
--- a/fs/verity/measure.c
+++ b/fs/verity/measure.c
@@ -11,6 +11,8 @@
 
 /**
  * fsverity_ioctl_measure() - get a verity file's measurement
+ * @filp: file to get measurement of
+ * @_uarg: user pointer to fsverity_digest
  *
  * Retrieve the file measurement that the kernel is enforcing for reads from a
  * verity file.  See the "FS_IOC_MEASURE_VERITY" section of
diff --git a/fs/verity/open.c b/fs/verity/open.c
index 63d1004b688c..d007db0c9304 100644
--- a/fs/verity/open.c
+++ b/fs/verity/open.c
@@ -31,7 +31,7 @@ int fsverity_init_merkle_tree_params(struct merkle_tree_params *params,
 				     unsigned int log_blocksize,
 				     const u8 *salt, size_t salt_size)
 {
-	const struct fsverity_hash_alg *hash_alg;
+	struct fsverity_hash_alg *hash_alg;
 	int err;
 	u64 blocks;
 	u64 offset;
@@ -102,6 +102,7 @@ int fsverity_init_merkle_tree_params(struct merkle_tree_params *params,
 		/* temporarily using level_start[] to store blocks in level */
 		params->level_start[params->num_levels++] = blocks;
 	}
+	params->level0_blocks = params->level_start[0];
 
 	/* Compute the starting block of each level */
 	offset = 0;
@@ -126,7 +127,7 @@ out_err:
  * Compute the file measurement by hashing the fsverity_descriptor excluding the
  * signature and with the sig_size field set to 0.
  */
-static int compute_file_measurement(const struct fsverity_hash_alg *hash_alg,
+static int compute_file_measurement(struct fsverity_hash_alg *hash_alg,
 				    struct fsverity_descriptor *desc,
 				    u8 *measurement)
 {
@@ -329,6 +330,7 @@ EXPORT_SYMBOL_GPL(fsverity_prepare_setattr);
 
 /**
  * fsverity_cleanup_inode() - free the inode's verity info, if present
+ * @inode: an inode being evicted
  *
  * Filesystems must call this on inode eviction to free ->i_verity_info.
  */
diff --git a/fs/verity/signature.c b/fs/verity/signature.c
index c8b255232de5..b14ed96387ec 100644
--- a/fs/verity/signature.c
+++ b/fs/verity/signature.c
@@ -28,6 +28,9 @@ static struct key *fsverity_keyring;
 
 /**
  * fsverity_verify_signature() - check a verity file's signature
+ * @vi: the file's fsverity_info
+ * @desc: the file's fsverity_descriptor
+ * @desc_size: size of @desc
  *
  * If the file's fs-verity descriptor includes a signature of the file
  * measurement, verify it against the certificates in the fs-verity keyring.
diff --git a/fs/verity/verify.c b/fs/verity/verify.c
index 3e8f2de44667..a8b68c6f663d 100644
--- a/fs/verity/verify.c
+++ b/fs/verity/verify.c
@@ -84,7 +84,8 @@ static inline int cmp_hashes(const struct fsverity_info *vi,
  * Return: true if the page is valid, else false.
  */
 static bool verify_page(struct inode *inode, const struct fsverity_info *vi,
-			struct ahash_request *req, struct page *data_page)
+			struct ahash_request *req, struct page *data_page,
+			unsigned long level0_ra_pages)
 {
 	const struct merkle_tree_params *params = &vi->tree_params;
 	const unsigned int hsize = params->digest_size;
@@ -117,8 +118,8 @@ static bool verify_page(struct inode *inode, const struct fsverity_info *vi,
 		pr_debug_ratelimited("Level %d: hindex=%lu, hoffset=%u\n",
 				     level, hindex, hoffset);
 
-		hpage = inode->i_sb->s_vop->read_merkle_tree_page(inode,
-								  hindex);
+		hpage = inode->i_sb->s_vop->read_merkle_tree_page(inode, hindex,
+				level == 0 ? level0_ra_pages : 0);
 		if (IS_ERR(hpage)) {
 			err = PTR_ERR(hpage);
 			fsverity_err(inode,
@@ -178,6 +179,7 @@ out:
 
 /**
  * fsverity_verify_page() - verify a data page
+ * @page: the page to verity
  *
  * Verify a page that has just been read from a verity file.  The page must be a
  * pagecache page that is still locked and not yet uptodate.
@@ -191,13 +193,12 @@ bool fsverity_verify_page(struct page *page)
 	struct ahash_request *req;
 	bool valid;
 
-	req = ahash_request_alloc(vi->tree_params.hash_alg->tfm, GFP_NOFS);
-	if (unlikely(!req))
-		return false;
+	/* This allocation never fails, since it's mempool-backed. */
+	req = fsverity_alloc_hash_request(vi->tree_params.hash_alg, GFP_NOFS);
 
-	valid = verify_page(inode, vi, req, page);
+	valid = verify_page(inode, vi, req, page, 0);
 
-	ahash_request_free(req);
+	fsverity_free_hash_request(vi->tree_params.hash_alg, req);
 
 	return valid;
 }
@@ -206,6 +207,7 @@ EXPORT_SYMBOL_GPL(fsverity_verify_page);
 #ifdef CONFIG_BLOCK
 /**
  * fsverity_verify_bio() - verify a 'read' bio that has just completed
+ * @bio: the bio to verify
  *
  * Verify a set of pages that have just been read from a verity file.  The pages
  * must be pagecache pages that are still locked and not yet uptodate.  Pages
@@ -222,31 +224,49 @@ void fsverity_verify_bio(struct bio *bio)
 {
 	struct inode *inode = bio_first_page_all(bio)->mapping->host;
 	const struct fsverity_info *vi = inode->i_verity_info;
+	const struct merkle_tree_params *params = &vi->tree_params;
 	struct ahash_request *req;
 	struct bio_vec *bv;
 	struct bvec_iter_all iter_all;
-
-	req = ahash_request_alloc(vi->tree_params.hash_alg->tfm, GFP_NOFS);
-	if (unlikely(!req)) {
+	unsigned long max_ra_pages = 0;
+
+	/* This allocation never fails, since it's mempool-backed. */
+	req = fsverity_alloc_hash_request(params->hash_alg, GFP_NOFS);
+
+	if (bio->bi_opf & REQ_RAHEAD) {
+		/*
+		 * If this bio is for data readahead, then we also do readahead
+		 * of the first (largest) level of the Merkle tree.  Namely,
+		 * when a Merkle tree page is read, we also try to piggy-back on
+		 * some additional pages -- up to 1/4 the number of data pages.
+		 *
+		 * This improves sequential read performance, as it greatly
+		 * reduces the number of I/O requests made to the Merkle tree.
+		 */
 		bio_for_each_segment_all(bv, bio, iter_all)
-			SetPageError(bv->bv_page);
-		return;
+			max_ra_pages++;
+		max_ra_pages /= 4;
 	}
 
 	bio_for_each_segment_all(bv, bio, iter_all) {
 		struct page *page = bv->bv_page;
+		unsigned long level0_index = page->index >> params->log_arity;
+		unsigned long level0_ra_pages =
+			min(max_ra_pages, params->level0_blocks - level0_index);
 
-		if (!PageError(page) && !verify_page(inode, vi, req, page))
+		if (!PageError(page) &&
+		    !verify_page(inode, vi, req, page, level0_ra_pages))
 			SetPageError(page);
 	}
 
-	ahash_request_free(req);
+	fsverity_free_hash_request(params->hash_alg, req);
 }
 EXPORT_SYMBOL_GPL(fsverity_verify_bio);
 #endif /* CONFIG_BLOCK */
 
 /**
  * fsverity_enqueue_verify_work() - enqueue work on the fs-verity workqueue
+ * @work: the work to enqueue
  *
  * Enqueue verification work for asynchronous processing.
  */
diff --git a/fs/xattr.c b/fs/xattr.c
index 90dd78f0eb27..91608d9bfc6a 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -817,7 +817,7 @@ struct simple_xattr *simple_xattr_alloc(const void *value, size_t size)
 	if (len < sizeof(*new_xattr))
 		return NULL;
 
-	new_xattr = kmalloc(len, GFP_KERNEL);
+	new_xattr = kvmalloc(len, GFP_KERNEL);
 	if (!new_xattr)
 		return NULL;
 
@@ -860,6 +860,7 @@ int simple_xattr_get(struct simple_xattrs *xattrs, const char *name,
  * @value: value of the xattr. If %NULL, will remove the attribute.
  * @size: size of the new xattr
  * @flags: %XATTR_{CREATE|REPLACE}
+ * @removed_size: returns size of the removed xattr, -1 if none removed
  *
  * %XATTR_CREATE is set, the xattr shouldn't exist already; otherwise fails
  * with -EEXIST.  If %XATTR_REPLACE is set, the xattr should exist;
@@ -868,12 +869,16 @@ int simple_xattr_get(struct simple_xattrs *xattrs, const char *name,
  * Returns 0 on success, -errno on failure.
  */
 int simple_xattr_set(struct simple_xattrs *xattrs, const char *name,
-		     const void *value, size_t size, int flags)
+		     const void *value, size_t size, int flags,
+		     ssize_t *removed_size)
 {
 	struct simple_xattr *xattr;
 	struct simple_xattr *new_xattr = NULL;
 	int err = 0;
 
+	if (removed_size)
+		*removed_size = -1;
+
 	/* value == NULL means remove */
 	if (value) {
 		new_xattr = simple_xattr_alloc(value, size);
@@ -882,7 +887,7 @@ int simple_xattr_set(struct simple_xattrs *xattrs, const char *name,
 
 		new_xattr->name = kstrdup(name, GFP_KERNEL);
 		if (!new_xattr->name) {
-			kfree(new_xattr);
+			kvfree(new_xattr);
 			return -ENOMEM;
 		}
 	}
@@ -895,8 +900,12 @@ int simple_xattr_set(struct simple_xattrs *xattrs, const char *name,
 				err = -EEXIST;
 			} else if (new_xattr) {
 				list_replace(&xattr->list, &new_xattr->list);
+				if (removed_size)
+					*removed_size = xattr->size;
 			} else {
 				list_del(&xattr->list);
+				if (removed_size)
+					*removed_size = xattr->size;
 			}
 			goto out;
 		}
@@ -912,7 +921,7 @@ out:
 	spin_unlock(&xattrs->lock);
 	if (xattr) {
 		kfree(xattr->name);
-		kfree(xattr);
+		kvfree(xattr);
 	}
 	return err;
 
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 06b68b6115bc..4f95df476181 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -26,8 +26,8 @@ xfs-y				+= $(addprefix libxfs/, \
 				   xfs_bmap.o \
 				   xfs_bmap_btree.o \
 				   xfs_btree.o \
+				   xfs_btree_staging.o \
 				   xfs_da_btree.o \
-				   xfs_da_format.o \
 				   xfs_defer.o \
 				   xfs_dir2.o \
 				   xfs_dir2_block.o \
diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c
index da031b93e182..f1366475c389 100644
--- a/fs/xfs/kmem.c
+++ b/fs/xfs/kmem.c
@@ -32,7 +32,7 @@ kmem_alloc(size_t size, xfs_km_flags_t flags)
 
 
 /*
- * __vmalloc() will allocate data pages and auxillary structures (e.g.
+ * __vmalloc() will allocate data pages and auxiliary structures (e.g.
  * pagetables) with GFP_KERNEL, yet we may be under GFP_NOFS context here. Hence
  * we need to tell memory reclaim that we are in such a context via
  * PF_MEMALLOC_NOFS to prevent memory reclaim re-entering the filesystem here
@@ -48,7 +48,7 @@ __kmem_vmalloc(size_t size, xfs_km_flags_t flags)
 	if (flags & KM_NOFS)
 		nofs_flag = memalloc_nofs_save();
 
-	ptr = __vmalloc(size, lflags, PAGE_KERNEL);
+	ptr = __vmalloc(size, lflags);
 
 	if (flags & KM_NOFS)
 		memalloc_nofs_restore(nofs_flag);
diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h
index 8170d95cf930..6143117770e9 100644
--- a/fs/xfs/kmem.h
+++ b/fs/xfs/kmem.h
@@ -78,39 +78,9 @@ kmem_zalloc_large(size_t size, xfs_km_flags_t flags)
  * Zone interfaces
  */
 
-#define KM_ZONE_HWALIGN	SLAB_HWCACHE_ALIGN
-#define KM_ZONE_RECLAIM	SLAB_RECLAIM_ACCOUNT
-#define KM_ZONE_SPREAD	SLAB_MEM_SPREAD
-#define KM_ZONE_ACCOUNT	SLAB_ACCOUNT
-
 #define kmem_zone	kmem_cache
 #define kmem_zone_t	struct kmem_cache
 
-static inline kmem_zone_t *
-kmem_zone_init(int size, char *zone_name)
-{
-	return kmem_cache_create(zone_name, size, 0, 0, NULL);
-}
-
-static inline kmem_zone_t *
-kmem_zone_init_flags(int size, char *zone_name, slab_flags_t flags,
-		     void (*construct)(void *))
-{
-	return kmem_cache_create(zone_name, size, 0, flags, construct);
-}
-
-static inline void
-kmem_zone_free(kmem_zone_t *zone, void *ptr)
-{
-	kmem_cache_free(zone, ptr);
-}
-
-static inline void
-kmem_zone_destroy(kmem_zone_t *zone)
-{
-	kmem_cache_destroy(zone);
-}
-
 extern void *kmem_zone_alloc(kmem_zone_t *, xfs_km_flags_t);
 
 static inline void *
diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c
index 14fbdf22b7e7..9d84007a5c65 100644
--- a/fs/xfs/libxfs/xfs_ag.c
+++ b/fs/xfs/libxfs/xfs_ag.c
@@ -23,25 +23,28 @@
 #include "xfs_ag_resv.h"
 #include "xfs_health.h"
 
-static struct xfs_buf *
+static int
 xfs_get_aghdr_buf(
 	struct xfs_mount	*mp,
 	xfs_daddr_t		blkno,
 	size_t			numblks,
+	struct xfs_buf		**bpp,
 	const struct xfs_buf_ops *ops)
 {
 	struct xfs_buf		*bp;
+	int			error;
 
-	bp = xfs_buf_get_uncached(mp->m_ddev_targp, numblks, 0);
-	if (!bp)
-		return NULL;
+	error = xfs_buf_get_uncached(mp->m_ddev_targp, numblks, 0, &bp);
+	if (error)
+		return error;
 
 	xfs_buf_zero(bp, 0, BBTOB(bp->b_length));
 	bp->b_bn = blkno;
 	bp->b_maps[0].bm_bn = blkno;
 	bp->b_ops = ops;
 
-	return bp;
+	*bpp = bp;
+	return 0;
 }
 
 static inline bool is_log_ag(struct xfs_mount *mp, struct aghdr_init_data *id)
@@ -228,7 +231,7 @@ xfs_sbblock_init(
 	struct xfs_buf		*bp,
 	struct aghdr_init_data	*id)
 {
-	struct xfs_dsb		*dsb = XFS_BUF_TO_SBP(bp);
+	struct xfs_dsb		*dsb = bp->b_addr;
 
 	xfs_sb_to_disk(dsb, &mp->m_sb);
 	dsb->sb_inprogress = 1;
@@ -240,7 +243,7 @@ xfs_agfblock_init(
 	struct xfs_buf		*bp,
 	struct aghdr_init_data	*id)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(bp);
+	struct xfs_agf		*agf = bp->b_addr;
 	xfs_extlen_t		tmpsize;
 
 	agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC);
@@ -298,7 +301,7 @@ xfs_agflblock_init(
 		uuid_copy(&agfl->agfl_uuid, &mp->m_sb.sb_meta_uuid);
 	}
 
-	agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, bp);
+	agfl_bno = xfs_buf_to_agfl_bno(bp);
 	for (bucket = 0; bucket < xfs_agfl_size(mp); bucket++)
 		agfl_bno[bucket] = cpu_to_be32(NULLAGBLOCK);
 }
@@ -309,7 +312,7 @@ xfs_agiblock_init(
 	struct xfs_buf		*bp,
 	struct aghdr_init_data	*id)
 {
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(bp);
+	struct xfs_agi		*agi = bp->b_addr;
 	int			bucket;
 
 	agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC);
@@ -340,13 +343,13 @@ xfs_ag_init_hdr(
 	struct aghdr_init_data	*id,
 	aghdr_init_work_f	work,
 	const struct xfs_buf_ops *ops)
-
 {
 	struct xfs_buf		*bp;
+	int			error;
 
-	bp = xfs_get_aghdr_buf(mp, id->daddr, id->numblks, ops);
-	if (!bp)
-		return -ENOMEM;
+	error = xfs_get_aghdr_buf(mp, id->daddr, id->numblks, &bp, ops);
+	if (error)
+		return error;
 
 	(*work)(mp, bp, id);
 
@@ -499,7 +502,7 @@ xfs_ag_extend_space(
 	if (error)
 		return error;
 
-	agi = XFS_BUF_TO_AGI(bp);
+	agi = bp->b_addr;
 	be32_add_cpu(&agi->agi_length, len);
 	ASSERT(id->agno == mp->m_sb.sb_agcount - 1 ||
 	       be32_to_cpu(agi->agi_length) == mp->m_sb.sb_agblocks);
@@ -512,7 +515,7 @@ xfs_ag_extend_space(
 	if (error)
 		return error;
 
-	agf = XFS_BUF_TO_AGF(bp);
+	agf = bp->b_addr;
 	be32_add_cpu(&agf->agf_length, len);
 	ASSERT(agf->agf_length == agi->agi_length);
 	xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH);
@@ -566,11 +569,11 @@ xfs_ag_get_geometry(
 	memset(ageo, 0, sizeof(*ageo));
 	ageo->ag_number = agno;
 
-	agi = XFS_BUF_TO_AGI(agi_bp);
+	agi = agi_bp->b_addr;
 	ageo->ag_icount = be32_to_cpu(agi->agi_count);
 	ageo->ag_ifree = be32_to_cpu(agi->agi_freecount);
 
-	agf = XFS_BUF_TO_AGF(agf_bp);
+	agf = agf_bp->b_addr;
 	ageo->ag_length = be32_to_cpu(agf->agf_length);
 	freeblks = pag->pagf_freeblks +
 		   pag->pagf_flcount +
diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c
index 87a9747f1d36..fdfe6dc0d307 100644
--- a/fs/xfs/libxfs/xfs_ag_resv.c
+++ b/fs/xfs/libxfs/xfs_ag_resv.c
@@ -19,6 +19,8 @@
 #include "xfs_btree.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_ialloc_btree.h"
+#include "xfs_sb.h"
+#include "xfs_ag_resv.h"
 
 /*
  * Per-AG Block Reservations
diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 533b04aaf6f6..203e74fa64aa 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -146,9 +146,13 @@ xfs_alloc_lookup_eq(
 	xfs_extlen_t		len,	/* length of extent */
 	int			*stat)	/* success/failure */
 {
+	int			error;
+
 	cur->bc_rec.a.ar_startblock = bno;
 	cur->bc_rec.a.ar_blockcount = len;
-	return xfs_btree_lookup(cur, XFS_LOOKUP_EQ, stat);
+	error = xfs_btree_lookup(cur, XFS_LOOKUP_EQ, stat);
+	cur->bc_ag.abt.active = (*stat == 1);
+	return error;
 }
 
 /*
@@ -162,9 +166,13 @@ xfs_alloc_lookup_ge(
 	xfs_extlen_t		len,	/* length of extent */
 	int			*stat)	/* success/failure */
 {
+	int			error;
+
 	cur->bc_rec.a.ar_startblock = bno;
 	cur->bc_rec.a.ar_blockcount = len;
-	return xfs_btree_lookup(cur, XFS_LOOKUP_GE, stat);
+	error = xfs_btree_lookup(cur, XFS_LOOKUP_GE, stat);
+	cur->bc_ag.abt.active = (*stat == 1);
+	return error;
 }
 
 /*
@@ -178,9 +186,19 @@ xfs_alloc_lookup_le(
 	xfs_extlen_t		len,	/* length of extent */
 	int			*stat)	/* success/failure */
 {
+	int			error;
 	cur->bc_rec.a.ar_startblock = bno;
 	cur->bc_rec.a.ar_blockcount = len;
-	return xfs_btree_lookup(cur, XFS_LOOKUP_LE, stat);
+	error = xfs_btree_lookup(cur, XFS_LOOKUP_LE, stat);
+	cur->bc_ag.abt.active = (*stat == 1);
+	return error;
+}
+
+static inline bool
+xfs_alloc_cur_active(
+	struct xfs_btree_cur	*cur)
+{
+	return cur && cur->bc_ag.abt.active;
 }
 
 /*
@@ -212,7 +230,7 @@ xfs_alloc_get_rec(
 	int			*stat)	/* output: success/failure */
 {
 	struct xfs_mount	*mp = cur->bc_mp;
-	xfs_agnumber_t		agno = cur->bc_private.a.agno;
+	xfs_agnumber_t		agno = cur->bc_ag.agno;
 	union xfs_btree_rec	*rec;
 	int			error;
 
@@ -313,7 +331,7 @@ xfs_alloc_compute_diff(
 	xfs_extlen_t	newlen1=0;	/* length with newbno1 */
 	xfs_extlen_t	newlen2=0;	/* length with newbno2 */
 	xfs_agblock_t	wantend;	/* end of target extent */
-	bool		userdata = xfs_alloc_is_userdata(datatype);
+	bool		userdata = datatype & XFS_ALLOC_USERDATA;
 
 	ASSERT(freelen >= wantlen);
 	freeend = freebno + freelen;
@@ -433,13 +451,17 @@ xfs_alloc_fixup_trees(
 #ifdef DEBUG
 		if ((error = xfs_alloc_get_rec(cnt_cur, &nfbno1, &nflen1, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp,
-			i == 1 && nfbno1 == fbno && nflen1 == flen);
+		if (XFS_IS_CORRUPT(mp,
+				   i != 1 ||
+				   nfbno1 != fbno ||
+				   nflen1 != flen))
+			return -EFSCORRUPTED;
 #endif
 	} else {
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, fbno, flen, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 	/*
 	 * Look up the record in the by-block tree if necessary.
@@ -448,13 +470,17 @@ xfs_alloc_fixup_trees(
 #ifdef DEBUG
 		if ((error = xfs_alloc_get_rec(bno_cur, &nfbno1, &nflen1, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp,
-			i == 1 && nfbno1 == fbno && nflen1 == flen);
+		if (XFS_IS_CORRUPT(mp,
+				   i != 1 ||
+				   nfbno1 != fbno ||
+				   nflen1 != flen))
+			return -EFSCORRUPTED;
 #endif
 	} else {
 		if ((error = xfs_alloc_lookup_eq(bno_cur, fbno, flen, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 
 #ifdef DEBUG
@@ -465,8 +491,10 @@ xfs_alloc_fixup_trees(
 		bnoblock = XFS_BUF_TO_BLOCK(bno_cur->bc_bufs[0]);
 		cntblock = XFS_BUF_TO_BLOCK(cnt_cur->bc_bufs[0]);
 
-		XFS_WANT_CORRUPTED_RETURN(mp,
-			bnoblock->bb_numrecs == cntblock->bb_numrecs);
+		if (XFS_IS_CORRUPT(mp,
+				   bnoblock->bb_numrecs !=
+				   cntblock->bb_numrecs))
+			return -EFSCORRUPTED;
 	}
 #endif
 
@@ -496,25 +524,30 @@ xfs_alloc_fixup_trees(
 	 */
 	if ((error = xfs_btree_delete(cnt_cur, &i)))
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+	if (XFS_IS_CORRUPT(mp, i != 1))
+		return -EFSCORRUPTED;
 	/*
 	 * Add new by-size btree entry(s).
 	 */
 	if (nfbno1 != NULLAGBLOCK) {
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, nfbno1, nflen1, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 0);
+		if (XFS_IS_CORRUPT(mp, i != 0))
+			return -EFSCORRUPTED;
 		if ((error = xfs_btree_insert(cnt_cur, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 	if (nfbno2 != NULLAGBLOCK) {
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, nfbno2, nflen2, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 0);
+		if (XFS_IS_CORRUPT(mp, i != 0))
+			return -EFSCORRUPTED;
 		if ((error = xfs_btree_insert(cnt_cur, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 	/*
 	 * Fix up the by-block btree entry(s).
@@ -525,7 +558,8 @@ xfs_alloc_fixup_trees(
 		 */
 		if ((error = xfs_btree_delete(bno_cur, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	} else {
 		/*
 		 * Update the by-block entry to start later|be shorter.
@@ -539,10 +573,12 @@ xfs_alloc_fixup_trees(
 		 */
 		if ((error = xfs_alloc_lookup_eq(bno_cur, nfbno2, nflen2, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 0);
+		if (XFS_IS_CORRUPT(mp, i != 0))
+			return -EFSCORRUPTED;
 		if ((error = xfs_btree_insert(bno_cur, &i)))
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -553,6 +589,7 @@ xfs_agfl_verify(
 {
 	struct xfs_mount *mp = bp->b_mount;
 	struct xfs_agfl	*agfl = XFS_BUF_TO_AGFL(bp);
+	__be32		*agfl_bno = xfs_buf_to_agfl_bno(bp);
 	int		i;
 
 	/*
@@ -578,8 +615,8 @@ xfs_agfl_verify(
 		return __this_address;
 
 	for (i = 0; i < xfs_agfl_size(mp); i++) {
-		if (be32_to_cpu(agfl->agfl_bno[i]) != NULLAGBLOCK &&
-		    be32_to_cpu(agfl->agfl_bno[i]) >= mp->m_sb.sb_agblocks)
+		if (be32_to_cpu(agfl_bno[i]) != NULLAGBLOCK &&
+		    be32_to_cpu(agfl_bno[i]) >= mp->m_sb.sb_agblocks)
 			return __this_address;
 	}
 
@@ -677,23 +714,305 @@ xfs_alloc_update_counters(
 	struct xfs_buf		*agbp,
 	long			len)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_agf		*agf = agbp->b_addr;
 
 	pag->pagf_freeblks += len;
 	be32_add_cpu(&agf->agf_freeblks, len);
 
 	xfs_trans_agblocks_delta(tp, len);
 	if (unlikely(be32_to_cpu(agf->agf_freeblks) >
-		     be32_to_cpu(agf->agf_length)))
+		     be32_to_cpu(agf->agf_length))) {
+		xfs_buf_mark_corrupt(agbp);
 		return -EFSCORRUPTED;
+	}
 
 	xfs_alloc_log_agf(tp, agbp, XFS_AGF_FREEBLKS);
 	return 0;
 }
 
 /*
- * Allocation group level functions.
+ * Block allocation algorithm and data structures.
  */
+struct xfs_alloc_cur {
+	struct xfs_btree_cur		*cnt;	/* btree cursors */
+	struct xfs_btree_cur		*bnolt;
+	struct xfs_btree_cur		*bnogt;
+	xfs_extlen_t			cur_len;/* current search length */
+	xfs_agblock_t			rec_bno;/* extent startblock */
+	xfs_extlen_t			rec_len;/* extent length */
+	xfs_agblock_t			bno;	/* alloc bno */
+	xfs_extlen_t			len;	/* alloc len */
+	xfs_extlen_t			diff;	/* diff from search bno */
+	unsigned int			busy_gen;/* busy state */
+	bool				busy;
+};
+
+/*
+ * Set up cursors, etc. in the extent allocation cursor. This function can be
+ * called multiple times to reset an initialized structure without having to
+ * reallocate cursors.
+ */
+static int
+xfs_alloc_cur_setup(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur)
+{
+	int			error;
+	int			i;
+
+	ASSERT(args->alignment == 1 || args->type != XFS_ALLOCTYPE_THIS_BNO);
+
+	acur->cur_len = args->maxlen;
+	acur->rec_bno = 0;
+	acur->rec_len = 0;
+	acur->bno = 0;
+	acur->len = 0;
+	acur->diff = -1;
+	acur->busy = false;
+	acur->busy_gen = 0;
+
+	/*
+	 * Perform an initial cntbt lookup to check for availability of maxlen
+	 * extents. If this fails, we'll return -ENOSPC to signal the caller to
+	 * attempt a small allocation.
+	 */
+	if (!acur->cnt)
+		acur->cnt = xfs_allocbt_init_cursor(args->mp, args->tp,
+					args->agbp, args->agno, XFS_BTNUM_CNT);
+	error = xfs_alloc_lookup_ge(acur->cnt, 0, args->maxlen, &i);
+	if (error)
+		return error;
+
+	/*
+	 * Allocate the bnobt left and right search cursors.
+	 */
+	if (!acur->bnolt)
+		acur->bnolt = xfs_allocbt_init_cursor(args->mp, args->tp,
+					args->agbp, args->agno, XFS_BTNUM_BNO);
+	if (!acur->bnogt)
+		acur->bnogt = xfs_allocbt_init_cursor(args->mp, args->tp,
+					args->agbp, args->agno, XFS_BTNUM_BNO);
+	return i == 1 ? 0 : -ENOSPC;
+}
+
+static void
+xfs_alloc_cur_close(
+	struct xfs_alloc_cur	*acur,
+	bool			error)
+{
+	int			cur_error = XFS_BTREE_NOERROR;
+
+	if (error)
+		cur_error = XFS_BTREE_ERROR;
+
+	if (acur->cnt)
+		xfs_btree_del_cursor(acur->cnt, cur_error);
+	if (acur->bnolt)
+		xfs_btree_del_cursor(acur->bnolt, cur_error);
+	if (acur->bnogt)
+		xfs_btree_del_cursor(acur->bnogt, cur_error);
+	acur->cnt = acur->bnolt = acur->bnogt = NULL;
+}
+
+/*
+ * Check an extent for allocation and track the best available candidate in the
+ * allocation structure. The cursor is deactivated if it has entered an out of
+ * range state based on allocation arguments. Optionally return the extent
+ * extent geometry and allocation status if requested by the caller.
+ */
+static int
+xfs_alloc_cur_check(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur,
+	struct xfs_btree_cur	*cur,
+	int			*new)
+{
+	int			error, i;
+	xfs_agblock_t		bno, bnoa, bnew;
+	xfs_extlen_t		len, lena, diff = -1;
+	bool			busy;
+	unsigned		busy_gen = 0;
+	bool			deactivate = false;
+	bool			isbnobt = cur->bc_btnum == XFS_BTNUM_BNO;
+
+	*new = 0;
+
+	error = xfs_alloc_get_rec(cur, &bno, &len, &i);
+	if (error)
+		return error;
+	if (XFS_IS_CORRUPT(args->mp, i != 1))
+		return -EFSCORRUPTED;
+
+	/*
+	 * Check minlen and deactivate a cntbt cursor if out of acceptable size
+	 * range (i.e., walking backwards looking for a minlen extent).
+	 */
+	if (len < args->minlen) {
+		deactivate = !isbnobt;
+		goto out;
+	}
+
+	busy = xfs_alloc_compute_aligned(args, bno, len, &bnoa, &lena,
+					 &busy_gen);
+	acur->busy |= busy;
+	if (busy)
+		acur->busy_gen = busy_gen;
+	/* deactivate a bnobt cursor outside of locality range */
+	if (bnoa < args->min_agbno || bnoa > args->max_agbno) {
+		deactivate = isbnobt;
+		goto out;
+	}
+	if (lena < args->minlen)
+		goto out;
+
+	args->len = XFS_EXTLEN_MIN(lena, args->maxlen);
+	xfs_alloc_fix_len(args);
+	ASSERT(args->len >= args->minlen);
+	if (args->len < acur->len)
+		goto out;
+
+	/*
+	 * We have an aligned record that satisfies minlen and beats or matches
+	 * the candidate extent size. Compare locality for near allocation mode.
+	 */
+	ASSERT(args->type == XFS_ALLOCTYPE_NEAR_BNO);
+	diff = xfs_alloc_compute_diff(args->agbno, args->len,
+				      args->alignment, args->datatype,
+				      bnoa, lena, &bnew);
+	if (bnew == NULLAGBLOCK)
+		goto out;
+
+	/*
+	 * Deactivate a bnobt cursor with worse locality than the current best.
+	 */
+	if (diff > acur->diff) {
+		deactivate = isbnobt;
+		goto out;
+	}
+
+	ASSERT(args->len > acur->len ||
+	       (args->len == acur->len && diff <= acur->diff));
+	acur->rec_bno = bno;
+	acur->rec_len = len;
+	acur->bno = bnew;
+	acur->len = args->len;
+	acur->diff = diff;
+	*new = 1;
+
+	/*
+	 * We're done if we found a perfect allocation. This only deactivates
+	 * the current cursor, but this is just an optimization to terminate a
+	 * cntbt search that otherwise runs to the edge of the tree.
+	 */
+	if (acur->diff == 0 && acur->len == args->maxlen)
+		deactivate = true;
+out:
+	if (deactivate)
+		cur->bc_ag.abt.active = false;
+	trace_xfs_alloc_cur_check(args->mp, cur->bc_btnum, bno, len, diff,
+				  *new);
+	return 0;
+}
+
+/*
+ * Complete an allocation of a candidate extent. Remove the extent from both
+ * trees and update the args structure.
+ */
+STATIC int
+xfs_alloc_cur_finish(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur)
+{
+	struct xfs_agf __maybe_unused *agf = args->agbp->b_addr;
+	int			error;
+
+	ASSERT(acur->cnt && acur->bnolt);
+	ASSERT(acur->bno >= acur->rec_bno);
+	ASSERT(acur->bno + acur->len <= acur->rec_bno + acur->rec_len);
+	ASSERT(acur->rec_bno + acur->rec_len <= be32_to_cpu(agf->agf_length));
+
+	error = xfs_alloc_fixup_trees(acur->cnt, acur->bnolt, acur->rec_bno,
+				      acur->rec_len, acur->bno, acur->len, 0);
+	if (error)
+		return error;
+
+	args->agbno = acur->bno;
+	args->len = acur->len;
+	args->wasfromfl = 0;
+
+	trace_xfs_alloc_cur(args);
+	return 0;
+}
+
+/*
+ * Locality allocation lookup algorithm. This expects a cntbt cursor and uses
+ * bno optimized lookup to search for extents with ideal size and locality.
+ */
+STATIC int
+xfs_alloc_cntbt_iter(
+	struct xfs_alloc_arg		*args,
+	struct xfs_alloc_cur		*acur)
+{
+	struct xfs_btree_cur	*cur = acur->cnt;
+	xfs_agblock_t		bno;
+	xfs_extlen_t		len, cur_len;
+	int			error;
+	int			i;
+
+	if (!xfs_alloc_cur_active(cur))
+		return 0;
+
+	/* locality optimized lookup */
+	cur_len = acur->cur_len;
+	error = xfs_alloc_lookup_ge(cur, args->agbno, cur_len, &i);
+	if (error)
+		return error;
+	if (i == 0)
+		return 0;
+	error = xfs_alloc_get_rec(cur, &bno, &len, &i);
+	if (error)
+		return error;
+
+	/* check the current record and update search length from it */
+	error = xfs_alloc_cur_check(args, acur, cur, &i);
+	if (error)
+		return error;
+	ASSERT(len >= acur->cur_len);
+	acur->cur_len = len;
+
+	/*
+	 * We looked up the first record >= [agbno, len] above. The agbno is a
+	 * secondary key and so the current record may lie just before or after
+	 * agbno. If it is past agbno, check the previous record too so long as
+	 * the length matches as it may be closer. Don't check a smaller record
+	 * because that could deactivate our cursor.
+	 */
+	if (bno > args->agbno) {
+		error = xfs_btree_decrement(cur, 0, &i);
+		if (!error && i) {
+			error = xfs_alloc_get_rec(cur, &bno, &len, &i);
+			if (!error && i && len == acur->cur_len)
+				error = xfs_alloc_cur_check(args, acur, cur,
+							    &i);
+		}
+		if (error)
+			return error;
+	}
+
+	/*
+	 * Increment the search key until we find at least one allocation
+	 * candidate or if the extent we found was larger. Otherwise, double the
+	 * search key to optimize the search. Efficiency is more important here
+	 * than absolute best locality.
+	 */
+	cur_len <<= 1;
+	if (!acur->len || acur->cur_len >= cur_len)
+		acur->cur_len++;
+	else
+		acur->cur_len = cur_len;
+
+	return error;
+}
 
 /*
  * Deal with the case where only small freespaces remain. Either return the
@@ -708,6 +1027,7 @@ xfs_alloc_ag_vextent_small(
 	xfs_extlen_t		*flenp,	/* result length */
 	int			*stat)	/* status: 0-freelist, 1-normal/none */
 {
+	struct xfs_agf		*agf = args->agbp->b_addr;
 	int			error = 0;
 	xfs_agblock_t		fbno = NULLAGBLOCK;
 	xfs_extlen_t		flen = 0;
@@ -727,14 +1047,16 @@ xfs_alloc_ag_vextent_small(
 		error = xfs_alloc_get_rec(ccur, &fbno, &flen, &i);
 		if (error)
 			goto error;
-		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error);
+		if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 		goto out;
 	}
 
 	if (args->minlen != 1 || args->alignment != 1 ||
 	    args->resv == XFS_AG_RESV_AGFL ||
-	    (be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_flcount) <=
-	     args->minleft))
+	    be32_to_cpu(agf->agf_flcount) <= args->minleft)
 		goto out;
 
 	error = xfs_alloc_get_freelist(args->tp, args->agbp, &fbno, 0);
@@ -744,23 +1066,24 @@ xfs_alloc_ag_vextent_small(
 		goto out;
 
 	xfs_extent_busy_reuse(args->mp, args->agno, fbno, 1,
-			      xfs_alloc_allow_busy_reuse(args->datatype));
+			      (args->datatype & XFS_ALLOC_NOBUSY));
 
-	if (xfs_alloc_is_userdata(args->datatype)) {
+	if (args->datatype & XFS_ALLOC_USERDATA) {
 		struct xfs_buf	*bp;
 
-		bp = xfs_btree_get_bufs(args->mp, args->tp, args->agno, fbno);
-		if (!bp) {
-			error = -EFSCORRUPTED;
+		error = xfs_trans_get_buf(args->tp, args->mp->m_ddev_targp,
+				XFS_AGB_TO_DADDR(args->mp, args->agno, fbno),
+				args->mp->m_bsize, 0, &bp);
+		if (error)
 			goto error;
-		}
 		xfs_trans_binval(args->tp, bp);
 	}
 	*fbnop = args->agbno = fbno;
 	*flenp = args->len = 1;
-	XFS_WANT_CORRUPTED_GOTO(args->mp,
-		fbno < be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length),
-		error);
+	if (XFS_IS_CORRUPT(args->mp, fbno >= be32_to_cpu(agf->agf_length))) {
+		error = -EFSCORRUPTED;
+		goto error;
+	}
 	args->wasfromfl = 1;
 	trace_xfs_alloc_small_freelist(args);
 
@@ -879,6 +1202,7 @@ STATIC int			/* error */
 xfs_alloc_ag_vextent_exact(
 	xfs_alloc_arg_t	*args)	/* allocation argument structure */
 {
+	struct xfs_agf __maybe_unused *agf = args->agbp->b_addr;
 	xfs_btree_cur_t	*bno_cur;/* by block-number btree cursor */
 	xfs_btree_cur_t	*cnt_cur;/* by count btree cursor */
 	int		error;
@@ -915,7 +1239,10 @@ xfs_alloc_ag_vextent_exact(
 	error = xfs_alloc_get_rec(bno_cur, &fbno, &flen, &i);
 	if (error)
 		goto error0;
-	XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	ASSERT(fbno <= args->agbno);
 
 	/*
@@ -954,8 +1281,7 @@ xfs_alloc_ag_vextent_exact(
 	 */
 	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
 		args->agno, XFS_BTNUM_CNT);
-	ASSERT(args->agbno + args->len <=
-		be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
+	ASSERT(args->agbno + args->len <= be32_to_cpu(agf->agf_length));
 	error = xfs_alloc_fixup_trees(cnt_cur, bno_cur, fbno, flen, args->agbno,
 				      args->len, XFSA_FIXUP_BNO_OK);
 	if (error) {
@@ -984,98 +1310,243 @@ error0:
 }
 
 /*
- * Search the btree in a given direction via the search cursor and compare
- * the records found against the good extent we've already found.
+ * Search a given number of btree records in a given direction. Check each
+ * record against the good extent we've already found.
  */
 STATIC int
-xfs_alloc_find_best_extent(
-	struct xfs_alloc_arg	*args,	/* allocation argument structure */
-	struct xfs_btree_cur	**gcur,	/* good cursor */
-	struct xfs_btree_cur	**scur,	/* searching cursor */
-	xfs_agblock_t		gdiff,	/* difference for search comparison */
-	xfs_agblock_t		*sbno,	/* extent found by search */
-	xfs_extlen_t		*slen,	/* extent length */
-	xfs_agblock_t		*sbnoa,	/* aligned extent found by search */
-	xfs_extlen_t		*slena,	/* aligned extent length */
-	int			dir)	/* 0 = search right, 1 = search left */
+xfs_alloc_walk_iter(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur,
+	struct xfs_btree_cur	*cur,
+	bool			increment,
+	bool			find_one, /* quit on first candidate */
+	int			count,    /* rec count (-1 for infinite) */
+	int			*stat)
 {
-	xfs_agblock_t		new;
-	xfs_agblock_t		sdiff;
 	int			error;
 	int			i;
-	unsigned		busy_gen;
 
-	/* The good extent is perfect, no need to  search. */
-	if (!gdiff)
-		goto out_use_good;
+	*stat = 0;
 
 	/*
-	 * Look until we find a better one, run out of space or run off the end.
+	 * Search so long as the cursor is active or we find a better extent.
+	 * The cursor is deactivated if it extends beyond the range of the
+	 * current allocation candidate.
 	 */
-	do {
-		error = xfs_alloc_get_rec(*scur, sbno, slen, &i);
+	while (xfs_alloc_cur_active(cur) && count) {
+		error = xfs_alloc_cur_check(args, acur, cur, &i);
 		if (error)
-			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-		xfs_alloc_compute_aligned(args, *sbno, *slen,
-				sbnoa, slena, &busy_gen);
+			return error;
+		if (i == 1) {
+			*stat = 1;
+			if (find_one)
+				break;
+		}
+		if (!xfs_alloc_cur_active(cur))
+			break;
+
+		if (increment)
+			error = xfs_btree_increment(cur, 0, &i);
+		else
+			error = xfs_btree_decrement(cur, 0, &i);
+		if (error)
+			return error;
+		if (i == 0)
+			cur->bc_ag.abt.active = false;
+
+		if (count > 0)
+			count--;
+	}
+
+	return 0;
+}
+
+/*
+ * Search the by-bno and by-size btrees in parallel in search of an extent with
+ * ideal locality based on the NEAR mode ->agbno locality hint.
+ */
+STATIC int
+xfs_alloc_ag_vextent_locality(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur,
+	int			*stat)
+{
+	struct xfs_btree_cur	*fbcur = NULL;
+	int			error;
+	int			i;
+	bool			fbinc;
+
+	ASSERT(acur->len == 0);
+	ASSERT(args->type == XFS_ALLOCTYPE_NEAR_BNO);
+
+	*stat = 0;
+
+	error = xfs_alloc_lookup_ge(acur->cnt, args->agbno, acur->cur_len, &i);
+	if (error)
+		return error;
+	error = xfs_alloc_lookup_le(acur->bnolt, args->agbno, 0, &i);
+	if (error)
+		return error;
+	error = xfs_alloc_lookup_ge(acur->bnogt, args->agbno, 0, &i);
+	if (error)
+		return error;
+
+	/*
+	 * Search the bnobt and cntbt in parallel. Search the bnobt left and
+	 * right and lookup the closest extent to the locality hint for each
+	 * extent size key in the cntbt. The entire search terminates
+	 * immediately on a bnobt hit because that means we've found best case
+	 * locality. Otherwise the search continues until the cntbt cursor runs
+	 * off the end of the tree. If no allocation candidate is found at this
+	 * point, give up on locality, walk backwards from the end of the cntbt
+	 * and take the first available extent.
+	 *
+	 * The parallel tree searches balance each other out to provide fairly
+	 * consistent performance for various situations. The bnobt search can
+	 * have pathological behavior in the worst case scenario of larger
+	 * allocation requests and fragmented free space. On the other hand, the
+	 * bnobt is able to satisfy most smaller allocation requests much more
+	 * quickly than the cntbt. The cntbt search can sift through fragmented
+	 * free space and sets of free extents for larger allocation requests
+	 * more quickly than the bnobt. Since the locality hint is just a hint
+	 * and we don't want to scan the entire bnobt for perfect locality, the
+	 * cntbt search essentially bounds the bnobt search such that we can
+	 * find good enough locality at reasonable performance in most cases.
+	 */
+	while (xfs_alloc_cur_active(acur->bnolt) ||
+	       xfs_alloc_cur_active(acur->bnogt) ||
+	       xfs_alloc_cur_active(acur->cnt)) {
+
+		trace_xfs_alloc_cur_lookup(args);
 
 		/*
-		 * The good extent is closer than this one.
+		 * Search the bnobt left and right. In the case of a hit, finish
+		 * the search in the opposite direction and we're done.
 		 */
-		if (!dir) {
-			if (*sbnoa > args->max_agbno)
-				goto out_use_good;
-			if (*sbnoa >= args->agbno + gdiff)
-				goto out_use_good;
-		} else {
-			if (*sbnoa < args->min_agbno)
-				goto out_use_good;
-			if (*sbnoa <= args->agbno - gdiff)
-				goto out_use_good;
+		error = xfs_alloc_walk_iter(args, acur, acur->bnolt, false,
+					    true, 1, &i);
+		if (error)
+			return error;
+		if (i == 1) {
+			trace_xfs_alloc_cur_left(args);
+			fbcur = acur->bnogt;
+			fbinc = true;
+			break;
+		}
+		error = xfs_alloc_walk_iter(args, acur, acur->bnogt, true, true,
+					    1, &i);
+		if (error)
+			return error;
+		if (i == 1) {
+			trace_xfs_alloc_cur_right(args);
+			fbcur = acur->bnolt;
+			fbinc = false;
+			break;
 		}
 
 		/*
-		 * Same distance, compare length and pick the best.
+		 * Check the extent with best locality based on the current
+		 * extent size search key and keep track of the best candidate.
 		 */
-		if (*slena >= args->minlen) {
-			args->len = XFS_EXTLEN_MIN(*slena, args->maxlen);
-			xfs_alloc_fix_len(args);
-
-			sdiff = xfs_alloc_compute_diff(args->agbno, args->len,
-						       args->alignment,
-						       args->datatype, *sbnoa,
-						       *slena, &new);
+		error = xfs_alloc_cntbt_iter(args, acur);
+		if (error)
+			return error;
+		if (!xfs_alloc_cur_active(acur->cnt)) {
+			trace_xfs_alloc_cur_lookup_done(args);
+			break;
+		}
+	}
 
-			/*
-			 * Choose closer size and invalidate other cursor.
-			 */
-			if (sdiff < gdiff)
-				goto out_use_search;
-			goto out_use_good;
+	/*
+	 * If we failed to find anything due to busy extents, return empty
+	 * handed so the caller can flush and retry. If no busy extents were
+	 * found, walk backwards from the end of the cntbt as a last resort.
+	 */
+	if (!xfs_alloc_cur_active(acur->cnt) && !acur->len && !acur->busy) {
+		error = xfs_btree_decrement(acur->cnt, 0, &i);
+		if (error)
+			return error;
+		if (i) {
+			acur->cnt->bc_ag.abt.active = true;
+			fbcur = acur->cnt;
+			fbinc = false;
 		}
+	}
 
-		if (!dir)
-			error = xfs_btree_increment(*scur, 0, &i);
-		else
-			error = xfs_btree_decrement(*scur, 0, &i);
+	/*
+	 * Search in the opposite direction for a better entry in the case of
+	 * a bnobt hit or walk backwards from the end of the cntbt.
+	 */
+	if (fbcur) {
+		error = xfs_alloc_walk_iter(args, acur, fbcur, fbinc, true, -1,
+					    &i);
 		if (error)
-			goto error0;
-	} while (i);
+			return error;
+	}
 
-out_use_good:
-	xfs_btree_del_cursor(*scur, XFS_BTREE_NOERROR);
-	*scur = NULL;
-	return 0;
+	if (acur->len)
+		*stat = 1;
 
-out_use_search:
-	xfs_btree_del_cursor(*gcur, XFS_BTREE_NOERROR);
-	*gcur = NULL;
 	return 0;
+}
 
-error0:
-	/* caller invalidates cursors */
-	return error;
+/* Check the last block of the cnt btree for allocations. */
+static int
+xfs_alloc_ag_vextent_lastblock(
+	struct xfs_alloc_arg	*args,
+	struct xfs_alloc_cur	*acur,
+	xfs_agblock_t		*bno,
+	xfs_extlen_t		*len,
+	bool			*allocated)
+{
+	int			error;
+	int			i;
+
+#ifdef DEBUG
+	/* Randomly don't execute the first algorithm. */
+	if (prandom_u32() & 1)
+		return 0;
+#endif
+
+	/*
+	 * Start from the entry that lookup found, sequence through all larger
+	 * free blocks.  If we're actually pointing at a record smaller than
+	 * maxlen, go to the start of this block, and skip all those smaller
+	 * than minlen.
+	 */
+	if (*len || args->alignment > 1) {
+		acur->cnt->bc_ptrs[0] = 1;
+		do {
+			error = xfs_alloc_get_rec(acur->cnt, bno, len, &i);
+			if (error)
+				return error;
+			if (XFS_IS_CORRUPT(args->mp, i != 1))
+				return -EFSCORRUPTED;
+			if (*len >= args->minlen)
+				break;
+			error = xfs_btree_increment(acur->cnt, 0, &i);
+			if (error)
+				return error;
+		} while (i);
+		ASSERT(*len >= args->minlen);
+		if (!i)
+			return 0;
+	}
+
+	error = xfs_alloc_walk_iter(args, acur, acur->cnt, true, false, -1, &i);
+	if (error)
+		return error;
+
+	/*
+	 * It didn't work.  We COULD be in a case where there's a good record
+	 * somewhere, so try again.
+	 */
+	if (acur->len == 0)
+		return 0;
+
+	trace_xfs_alloc_near_first(args);
+	*allocated = true;
+	return 0;
 }
 
 /*
@@ -1084,41 +1555,17 @@ error0:
  * and of the form k * prod + mod unless there's nothing that large.
  * Return the starting a.g. block, or NULLAGBLOCK if we can't do it.
  */
-STATIC int				/* error */
+STATIC int
 xfs_alloc_ag_vextent_near(
-	xfs_alloc_arg_t	*args)		/* allocation argument structure */
+	struct xfs_alloc_arg	*args)
 {
-	xfs_btree_cur_t	*bno_cur_gt;	/* cursor for bno btree, right side */
-	xfs_btree_cur_t	*bno_cur_lt;	/* cursor for bno btree, left side */
-	xfs_btree_cur_t	*cnt_cur;	/* cursor for count btree */
-	xfs_agblock_t	gtbno;		/* start bno of right side entry */
-	xfs_agblock_t	gtbnoa;		/* aligned ... */
-	xfs_extlen_t	gtdiff;		/* difference to right side entry */
-	xfs_extlen_t	gtlen;		/* length of right side entry */
-	xfs_extlen_t	gtlena;		/* aligned ... */
-	xfs_agblock_t	gtnew;		/* useful start bno of right side */
-	int		error;		/* error code */
-	int		i;		/* result code, temporary */
-	int		j;		/* result code, temporary */
-	xfs_agblock_t	ltbno;		/* start bno of left side entry */
-	xfs_agblock_t	ltbnoa;		/* aligned ... */
-	xfs_extlen_t	ltdiff;		/* difference to left side entry */
-	xfs_extlen_t	ltlen;		/* length of left side entry */
-	xfs_extlen_t	ltlena;		/* aligned ... */
-	xfs_agblock_t	ltnew;		/* useful start bno of left side */
-	xfs_extlen_t	rlen;		/* length of returned extent */
-	bool		busy;
-	unsigned	busy_gen;
-#ifdef DEBUG
-	/*
-	 * Randomly don't execute the first algorithm.
-	 */
-	int		dofirst;	/* set to do first algorithm */
-
-	dofirst = prandom_u32() & 1;
-#endif
+	struct xfs_alloc_cur	acur = {};
+	int			error;		/* error code */
+	int			i;		/* result code, temporary */
+	xfs_agblock_t		bno;
+	xfs_extlen_t		len;
 
-	/* handle unitialized agbno range so caller doesn't have to */
+	/* handle uninitialized agbno range so caller doesn't have to */
 	if (!args->min_agbno && !args->max_agbno)
 		args->max_agbno = args->mp->m_sb.sb_agblocks - 1;
 	ASSERT(args->min_agbno <= args->max_agbno);
@@ -1130,40 +1577,27 @@ xfs_alloc_ag_vextent_near(
 		args->agbno = args->max_agbno;
 
 restart:
-	bno_cur_lt = NULL;
-	bno_cur_gt = NULL;
-	ltlen = 0;
-	gtlena = 0;
-	ltlena = 0;
-	busy = false;
+	len = 0;
 
 	/*
-	 * Get a cursor for the by-size btree.
-	 */
-	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_CNT);
-
-	/*
-	 * See if there are any free extents as big as maxlen.
-	 */
-	if ((error = xfs_alloc_lookup_ge(cnt_cur, 0, args->maxlen, &i)))
-		goto error0;
-	/*
-	 * If none, then pick up the last entry in the tree unless the
-	 * tree is empty.
+	 * Set up cursors and see if there are any free extents as big as
+	 * maxlen. If not, pick the last entry in the tree unless the tree is
+	 * empty.
 	 */
-	if (!i) {
-		if ((error = xfs_alloc_ag_vextent_small(args, cnt_cur, &ltbno,
-				&ltlen, &i)))
-			goto error0;
-		if (i == 0 || ltlen == 0) {
-			xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
+	error = xfs_alloc_cur_setup(args, &acur);
+	if (error == -ENOSPC) {
+		error = xfs_alloc_ag_vextent_small(args, acur.cnt, &bno,
+				&len, &i);
+		if (error)
+			goto out;
+		if (i == 0 || len == 0) {
 			trace_xfs_alloc_near_noentry(args);
-			return 0;
+			goto out;
 		}
 		ASSERT(i == 1);
+	} else if (error) {
+		goto out;
 	}
-	args->wasfromfl = 0;
 
 	/*
 	 * First algorithm.
@@ -1172,311 +1606,47 @@ restart:
 	 * near the right edge of the tree.  If it's in the last btree leaf
 	 * block, then we just examine all the entries in that block
 	 * that are big enough, and pick the best one.
-	 * This is written as a while loop so we can break out of it,
-	 * but we never loop back to the top.
 	 */
-	while (xfs_btree_islastblock(cnt_cur, 0)) {
-		xfs_extlen_t	bdiff;
-		int		besti=0;
-		xfs_extlen_t	blen=0;
-		xfs_agblock_t	bnew=0;
+	if (xfs_btree_islastblock(acur.cnt, 0)) {
+		bool		allocated = false;
 
-#ifdef DEBUG
-		if (dofirst)
-			break;
-#endif
-		/*
-		 * Start from the entry that lookup found, sequence through
-		 * all larger free blocks.  If we're actually pointing at a
-		 * record smaller than maxlen, go to the start of this block,
-		 * and skip all those smaller than minlen.
-		 */
-		if (ltlen || args->alignment > 1) {
-			cnt_cur->bc_ptrs[0] = 1;
-			do {
-				if ((error = xfs_alloc_get_rec(cnt_cur, &ltbno,
-						&ltlen, &i)))
-					goto error0;
-				XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-				if (ltlen >= args->minlen)
-					break;
-				if ((error = xfs_btree_increment(cnt_cur, 0, &i)))
-					goto error0;
-			} while (i);
-			ASSERT(ltlen >= args->minlen);
-			if (!i)
-				break;
-		}
-		i = cnt_cur->bc_ptrs[0];
-		for (j = 1, blen = 0, bdiff = 0;
-		     !error && j && (blen < args->maxlen || bdiff > 0);
-		     error = xfs_btree_increment(cnt_cur, 0, &j)) {
-			/*
-			 * For each entry, decide if it's better than
-			 * the previous best entry.
-			 */
-			if ((error = xfs_alloc_get_rec(cnt_cur, &ltbno, &ltlen, &i)))
-				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			busy = xfs_alloc_compute_aligned(args, ltbno, ltlen,
-					&ltbnoa, &ltlena, &busy_gen);
-			if (ltlena < args->minlen)
-				continue;
-			if (ltbnoa < args->min_agbno || ltbnoa > args->max_agbno)
-				continue;
-			args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
-			xfs_alloc_fix_len(args);
-			ASSERT(args->len >= args->minlen);
-			if (args->len < blen)
-				continue;
-			ltdiff = xfs_alloc_compute_diff(args->agbno, args->len,
-				args->alignment, args->datatype, ltbnoa,
-				ltlena, &ltnew);
-			if (ltnew != NULLAGBLOCK &&
-			    (args->len > blen || ltdiff < bdiff)) {
-				bdiff = ltdiff;
-				bnew = ltnew;
-				blen = args->len;
-				besti = cnt_cur->bc_ptrs[0];
-			}
-		}
-		/*
-		 * It didn't work.  We COULD be in a case where
-		 * there's a good record somewhere, so try again.
-		 */
-		if (blen == 0)
-			break;
-		/*
-		 * Point at the best entry, and retrieve it again.
-		 */
-		cnt_cur->bc_ptrs[0] = besti;
-		if ((error = xfs_alloc_get_rec(cnt_cur, &ltbno, &ltlen, &i)))
-			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-		ASSERT(ltbno + ltlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
-		args->len = blen;
-
-		/*
-		 * We are allocating starting at bnew for blen blocks.
-		 */
-		args->agbno = bnew;
-		ASSERT(bnew >= ltbno);
-		ASSERT(bnew + blen <= ltbno + ltlen);
-		/*
-		 * Set up a cursor for the by-bno tree.
-		 */
-		bno_cur_lt = xfs_allocbt_init_cursor(args->mp, args->tp,
-			args->agbp, args->agno, XFS_BTNUM_BNO);
-		/*
-		 * Fix up the btree entries.
-		 */
-		if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno,
-				ltlen, bnew, blen, XFSA_FIXUP_CNT_OK)))
-			goto error0;
-		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-		xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_NOERROR);
-
-		trace_xfs_alloc_near_first(args);
-		return 0;
-	}
-	/*
-	 * Second algorithm.
-	 * Search in the by-bno tree to the left and to the right
-	 * simultaneously, until in each case we find a space big enough,
-	 * or run into the edge of the tree.  When we run into the edge,
-	 * we deallocate that cursor.
-	 * If both searches succeed, we compare the two spaces and pick
-	 * the better one.
-	 * With alignment, it's possible for both to fail; the upper
-	 * level algorithm that picks allocation groups for allocations
-	 * is not supposed to do this.
-	 */
-	/*
-	 * Allocate and initialize the cursor for the leftward search.
-	 */
-	bno_cur_lt = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
-		args->agno, XFS_BTNUM_BNO);
-	/*
-	 * Lookup <= bno to find the leftward search's starting point.
-	 */
-	if ((error = xfs_alloc_lookup_le(bno_cur_lt, args->agbno, args->maxlen, &i)))
-		goto error0;
-	if (!i) {
-		/*
-		 * Didn't find anything; use this cursor for the rightward
-		 * search.
-		 */
-		bno_cur_gt = bno_cur_lt;
-		bno_cur_lt = NULL;
-	}
-	/*
-	 * Found something.  Duplicate the cursor for the rightward search.
-	 */
-	else if ((error = xfs_btree_dup_cursor(bno_cur_lt, &bno_cur_gt)))
-		goto error0;
-	/*
-	 * Increment the cursor, so we will point at the entry just right
-	 * of the leftward entry if any, or to the leftmost entry.
-	 */
-	if ((error = xfs_btree_increment(bno_cur_gt, 0, &i)))
-		goto error0;
-	if (!i) {
-		/*
-		 * It failed, there are no rightward entries.
-		 */
-		xfs_btree_del_cursor(bno_cur_gt, XFS_BTREE_NOERROR);
-		bno_cur_gt = NULL;
+		error = xfs_alloc_ag_vextent_lastblock(args, &acur, &bno, &len,
+				&allocated);
+		if (error)
+			goto out;
+		if (allocated)
+			goto alloc_finish;
 	}
-	/*
-	 * Loop going left with the leftward cursor, right with the
-	 * rightward cursor, until either both directions give up or
-	 * we find an entry at least as big as minlen.
-	 */
-	do {
-		if (bno_cur_lt) {
-			if ((error = xfs_alloc_get_rec(bno_cur_lt, &ltbno, &ltlen, &i)))
-				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			busy |= xfs_alloc_compute_aligned(args, ltbno, ltlen,
-					&ltbnoa, &ltlena, &busy_gen);
-			if (ltlena >= args->minlen && ltbnoa >= args->min_agbno)
-				break;
-			if ((error = xfs_btree_decrement(bno_cur_lt, 0, &i)))
-				goto error0;
-			if (!i || ltbnoa < args->min_agbno) {
-				xfs_btree_del_cursor(bno_cur_lt,
-						     XFS_BTREE_NOERROR);
-				bno_cur_lt = NULL;
-			}
-		}
-		if (bno_cur_gt) {
-			if ((error = xfs_alloc_get_rec(bno_cur_gt, &gtbno, &gtlen, &i)))
-				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			busy |= xfs_alloc_compute_aligned(args, gtbno, gtlen,
-					&gtbnoa, &gtlena, &busy_gen);
-			if (gtlena >= args->minlen && gtbnoa <= args->max_agbno)
-				break;
-			if ((error = xfs_btree_increment(bno_cur_gt, 0, &i)))
-				goto error0;
-			if (!i || gtbnoa > args->max_agbno) {
-				xfs_btree_del_cursor(bno_cur_gt,
-						     XFS_BTREE_NOERROR);
-				bno_cur_gt = NULL;
-			}
-		}
-	} while (bno_cur_lt || bno_cur_gt);
 
 	/*
-	 * Got both cursors still active, need to find better entry.
+	 * Second algorithm. Combined cntbt and bnobt search to find ideal
+	 * locality.
 	 */
-	if (bno_cur_lt && bno_cur_gt) {
-		if (ltlena >= args->minlen) {
-			/*
-			 * Left side is good, look for a right side entry.
-			 */
-			args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
-			xfs_alloc_fix_len(args);
-			ltdiff = xfs_alloc_compute_diff(args->agbno, args->len,
-				args->alignment, args->datatype, ltbnoa,
-				ltlena, &ltnew);
-
-			error = xfs_alloc_find_best_extent(args,
-						&bno_cur_lt, &bno_cur_gt,
-						ltdiff, &gtbno, &gtlen,
-						&gtbnoa, &gtlena,
-						0 /* search right */);
-		} else {
-			ASSERT(gtlena >= args->minlen);
-
-			/*
-			 * Right side is good, look for a left side entry.
-			 */
-			args->len = XFS_EXTLEN_MIN(gtlena, args->maxlen);
-			xfs_alloc_fix_len(args);
-			gtdiff = xfs_alloc_compute_diff(args->agbno, args->len,
-				args->alignment, args->datatype, gtbnoa,
-				gtlena, &gtnew);
-
-			error = xfs_alloc_find_best_extent(args,
-						&bno_cur_gt, &bno_cur_lt,
-						gtdiff, &ltbno, &ltlen,
-						&ltbnoa, &ltlena,
-						1 /* search left */);
-		}
-
-		if (error)
-			goto error0;
-	}
+	error = xfs_alloc_ag_vextent_locality(args, &acur, &i);
+	if (error)
+		goto out;
 
 	/*
 	 * If we couldn't get anything, give up.
 	 */
-	if (bno_cur_lt == NULL && bno_cur_gt == NULL) {
-		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-
-		if (busy) {
+	if (!acur.len) {
+		if (acur.busy) {
 			trace_xfs_alloc_near_busy(args);
-			xfs_extent_busy_flush(args->mp, args->pag, busy_gen);
+			xfs_extent_busy_flush(args->mp, args->pag,
+					      acur.busy_gen);
 			goto restart;
 		}
 		trace_xfs_alloc_size_neither(args);
 		args->agbno = NULLAGBLOCK;
-		return 0;
+		goto out;
 	}
 
-	/*
-	 * At this point we have selected a freespace entry, either to the
-	 * left or to the right.  If it's on the right, copy all the
-	 * useful variables to the "left" set so we only have one
-	 * copy of this code.
-	 */
-	if (bno_cur_gt) {
-		bno_cur_lt = bno_cur_gt;
-		bno_cur_gt = NULL;
-		ltbno = gtbno;
-		ltbnoa = gtbnoa;
-		ltlen = gtlen;
-		ltlena = gtlena;
-		j = 1;
-	} else
-		j = 0;
+alloc_finish:
+	/* fix up btrees on a successful allocation */
+	error = xfs_alloc_cur_finish(args, &acur);
 
-	/*
-	 * Fix up the length and compute the useful address.
-	 */
-	args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
-	xfs_alloc_fix_len(args);
-	rlen = args->len;
-	(void)xfs_alloc_compute_diff(args->agbno, rlen, args->alignment,
-				     args->datatype, ltbnoa, ltlena, &ltnew);
-	ASSERT(ltnew >= ltbno);
-	ASSERT(ltnew + rlen <= ltbnoa + ltlena);
-	ASSERT(ltnew + rlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
-	ASSERT(ltnew >= args->min_agbno && ltnew <= args->max_agbno);
-	args->agbno = ltnew;
-
-	if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno, ltlen,
-			ltnew, rlen, XFSA_FIXUP_BNO_OK)))
-		goto error0;
-
-	if (j)
-		trace_xfs_alloc_near_greater(args);
-	else
-		trace_xfs_alloc_near_lesser(args);
-
-	xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-	xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_NOERROR);
-	return 0;
-
- error0:
-	trace_xfs_alloc_near_error(args);
-	if (cnt_cur != NULL)
-		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_ERROR);
-	if (bno_cur_lt != NULL)
-		xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_ERROR);
-	if (bno_cur_gt != NULL)
-		xfs_btree_del_cursor(bno_cur_gt, XFS_BTREE_ERROR);
+out:
+	xfs_alloc_cur_close(&acur, error);
 	return error;
 }
 
@@ -1490,6 +1660,7 @@ STATIC int				/* error */
 xfs_alloc_ag_vextent_size(
 	xfs_alloc_arg_t	*args)		/* allocation argument structure */
 {
+	struct xfs_agf	*agf = args->agbp->b_addr;
 	xfs_btree_cur_t	*bno_cur;	/* cursor for bno btree */
 	xfs_btree_cur_t	*cnt_cur;	/* cursor for cnt btree */
 	int		error;		/* error result */
@@ -1545,7 +1716,10 @@ restart:
 			error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, &i);
 			if (error)
 				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
+			if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 
 			busy = xfs_alloc_compute_aligned(args, fbno, flen,
 					&rbno, &rlen, &busy_gen);
@@ -1579,8 +1753,13 @@ restart:
 	 * This can't happen in the second case above.
 	 */
 	rlen = XFS_EXTLEN_MIN(args->maxlen, rlen);
-	XFS_WANT_CORRUPTED_GOTO(args->mp, rlen == 0 ||
-			(rlen <= flen && rbno + rlen <= fbno + flen), error0);
+	if (XFS_IS_CORRUPT(args->mp,
+			   rlen != 0 &&
+			   (rlen > flen ||
+			    rbno + rlen > fbno + flen))) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	if (rlen < args->maxlen) {
 		xfs_agblock_t	bestfbno;
 		xfs_extlen_t	bestflen;
@@ -1599,15 +1778,22 @@ restart:
 			if ((error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen,
 					&i)))
 				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
+			if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 			if (flen < bestrlen)
 				break;
 			busy = xfs_alloc_compute_aligned(args, fbno, flen,
 					&rbno, &rlen, &busy_gen);
 			rlen = XFS_EXTLEN_MIN(args->maxlen, rlen);
-			XFS_WANT_CORRUPTED_GOTO(args->mp, rlen == 0 ||
-				(rlen <= flen && rbno + rlen <= fbno + flen),
-				error0);
+			if (XFS_IS_CORRUPT(args->mp,
+					   rlen != 0 &&
+					   (rlen > flen ||
+					    rbno + rlen > fbno + flen))) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 			if (rlen > bestrlen) {
 				bestrlen = rlen;
 				bestrbno = rbno;
@@ -1620,7 +1806,10 @@ restart:
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, bestfbno, bestflen,
 				&i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(args->mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		rlen = bestrlen;
 		rbno = bestrbno;
 		flen = bestflen;
@@ -1643,7 +1832,10 @@ restart:
 	xfs_alloc_fix_len(args);
 
 	rlen = args->len;
-	XFS_WANT_CORRUPTED_GOTO(args->mp, rlen <= flen, error0);
+	if (XFS_IS_CORRUPT(args->mp, rlen > flen)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	/*
 	 * Allocate and initialize a cursor for the by-block tree.
 	 */
@@ -1657,10 +1849,12 @@ restart:
 	cnt_cur = bno_cur = NULL;
 	args->len = rlen;
 	args->agbno = rbno;
-	XFS_WANT_CORRUPTED_GOTO(args->mp,
-		args->agbno + args->len <=
-			be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length),
-		error0);
+	if (XFS_IS_CORRUPT(args->mp,
+			   args->agbno + args->len >
+			   be32_to_cpu(agf->agf_length))) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	trace_xfs_alloc_size_done(args);
 	return 0;
 
@@ -1732,7 +1926,10 @@ xfs_free_ag_extent(
 		 */
 		if ((error = xfs_alloc_get_rec(bno_cur, &ltbno, &ltlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * It's not contiguous, though.
 		 */
@@ -1744,8 +1941,10 @@ xfs_free_ag_extent(
 			 * space was invalid, it's (partly) already free.
 			 * Very bad.
 			 */
-			XFS_WANT_CORRUPTED_GOTO(mp,
-						ltbno + ltlen <= bno, error0);
+			if (XFS_IS_CORRUPT(mp, ltbno + ltlen > bno)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 		}
 	}
 	/*
@@ -1760,7 +1959,10 @@ xfs_free_ag_extent(
 		 */
 		if ((error = xfs_alloc_get_rec(bno_cur, &gtbno, &gtlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * It's not contiguous, though.
 		 */
@@ -1772,7 +1974,10 @@ xfs_free_ag_extent(
 			 * space was invalid, it's (partly) already free.
 			 * Very bad.
 			 */
-			XFS_WANT_CORRUPTED_GOTO(mp, gtbno >= bno + len, error0);
+			if (XFS_IS_CORRUPT(mp, bno + len > gtbno)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 		}
 	}
 	/*
@@ -1789,31 +1994,49 @@ xfs_free_ag_extent(
 		 */
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, ltbno, ltlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Delete the old by-size entry on the right.
 		 */
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, gtbno, gtlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Delete the old by-block entry for the right block.
 		 */
 		if ((error = xfs_btree_delete(bno_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Move the by-block cursor back to the left neighbor.
 		 */
 		if ((error = xfs_btree_decrement(bno_cur, 0, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 #ifdef DEBUG
 		/*
 		 * Check that this is the right record: delete didn't
@@ -1826,9 +2049,13 @@ xfs_free_ag_extent(
 			if ((error = xfs_alloc_get_rec(bno_cur, &xxbno, &xxlen,
 					&i)))
 				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(mp,
-				i == 1 && xxbno == ltbno && xxlen == ltlen,
-				error0);
+			if (XFS_IS_CORRUPT(mp,
+					   i != 1 ||
+					   xxbno != ltbno ||
+					   xxlen != ltlen)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 		}
 #endif
 		/*
@@ -1849,17 +2076,26 @@ xfs_free_ag_extent(
 		 */
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, ltbno, ltlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Back up the by-block cursor to the left neighbor, and
 		 * update its length.
 		 */
 		if ((error = xfs_btree_decrement(bno_cur, 0, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		nbno = ltbno;
 		nlen = len + ltlen;
 		if ((error = xfs_alloc_update(bno_cur, nbno, nlen)))
@@ -1875,10 +2111,16 @@ xfs_free_ag_extent(
 		 */
 		if ((error = xfs_alloc_lookup_eq(cnt_cur, gtbno, gtlen, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_delete(cnt_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		/*
 		 * Update the starting block and length of the right
 		 * neighbor in the by-block tree.
@@ -1897,7 +2139,10 @@ xfs_free_ag_extent(
 		nlen = len;
 		if ((error = xfs_btree_insert(bno_cur, &i)))
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 	}
 	xfs_btree_del_cursor(bno_cur, XFS_BTREE_NOERROR);
 	bno_cur = NULL;
@@ -1906,10 +2151,16 @@ xfs_free_ag_extent(
 	 */
 	if ((error = xfs_alloc_lookup_eq(cnt_cur, nbno, nlen, &i)))
 		goto error0;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 0, error0);
+	if (XFS_IS_CORRUPT(mp, i != 0)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	if ((error = xfs_btree_insert(cnt_cur, &i)))
 		goto error0;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
 	cnt_cur = NULL;
 
@@ -1989,30 +2240,39 @@ xfs_alloc_longest_free_extent(
 	 * reservations and AGFL rules in place, we can return this extent.
 	 */
 	if (pag->pagf_longest > delta)
-		return pag->pagf_longest - delta;
+		return min_t(xfs_extlen_t, pag->pag_mount->m_ag_max_usable,
+				pag->pagf_longest - delta);
 
 	/* Otherwise, let the caller try for 1 block if there's space. */
 	return pag->pagf_flcount > 0 || pag->pagf_longest > 0;
 }
 
+/*
+ * Compute the minimum length of the AGFL in the given AG.  If @pag is NULL,
+ * return the largest possible minimum length.
+ */
 unsigned int
 xfs_alloc_min_freelist(
 	struct xfs_mount	*mp,
 	struct xfs_perag	*pag)
 {
+	/* AG btrees have at least 1 level. */
+	static const uint8_t	fake_levels[XFS_BTNUM_AGF] = {1, 1, 1};
+	const uint8_t		*levels = pag ? pag->pagf_levels : fake_levels;
 	unsigned int		min_free;
 
+	ASSERT(mp->m_ag_maxlevels > 0);
+
 	/* space needed by-bno freespace btree */
-	min_free = min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_BNOi] + 1,
+	min_free = min_t(unsigned int, levels[XFS_BTNUM_BNOi] + 1,
 				       mp->m_ag_maxlevels);
 	/* space needed by-size freespace btree */
-	min_free += min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_CNTi] + 1,
+	min_free += min_t(unsigned int, levels[XFS_BTNUM_CNTi] + 1,
 				       mp->m_ag_maxlevels);
 	/* space needed reverse mapping used space btree */
 	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
-		min_free += min_t(unsigned int,
-				  pag->pagf_levels[XFS_BTNUM_RMAPi] + 1,
-				  mp->m_rmap_maxlevels);
+		min_free += min_t(unsigned int, levels[XFS_BTNUM_RMAPi] + 1,
+						mp->m_rmap_maxlevels);
 
 	return min_free;
 }
@@ -2086,9 +2346,11 @@ xfs_free_agfl_block(
 	if (error)
 		return error;
 
-	bp = xfs_btree_get_bufs(tp->t_mountp, tp, agno, agbno);
-	if (!bp)
-		return -EFSCORRUPTED;
+	error = xfs_trans_get_buf(tp, tp->t_mountp->m_ddev_targp,
+			XFS_AGB_TO_DADDR(tp->t_mountp, agno, agbno),
+			tp->t_mountp->m_bsize, 0, &bp);
+	if (error)
+		return error;
 	xfs_trans_binval(tp, bp);
 
 	return 0;
@@ -2161,7 +2423,7 @@ xfs_agfl_reset(
 	struct xfs_perag	*pag)
 {
 	struct xfs_mount	*mp = tp->t_mountp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_agf		*agf = agbp->b_addr;
 
 	ASSERT(pag->pagf_agflreset);
 	trace_xfs_agfl_reset(mp, agf, 0, _RET_IP_);
@@ -2239,12 +2501,11 @@ xfs_alloc_fix_freelist(
 
 	if (!pag->pagf_init) {
 		error = xfs_alloc_read_agf(mp, tp, args->agno, flags, &agbp);
-		if (error)
+		if (error) {
+			/* Couldn't lock the AGF so skip this AG. */
+			if (error == -EAGAIN)
+				error = 0;
 			goto out_no_agbp;
-		if (!pag->pagf_init) {
-			ASSERT(flags & XFS_ALLOC_FLAG_TRYLOCK);
-			ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING));
-			goto out_agbp_relse;
 		}
 	}
 
@@ -2253,7 +2514,7 @@ xfs_alloc_fix_freelist(
 	 * somewhere else if we are not being asked to try harder at this
 	 * point
 	 */
-	if (pag->pagf_metadata && xfs_alloc_is_userdata(args->datatype) &&
+	if (pag->pagf_metadata && (args->datatype & XFS_ALLOC_USERDATA) &&
 	    (flags & XFS_ALLOC_FLAG_TRYLOCK)) {
 		ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING));
 		goto out_agbp_relse;
@@ -2270,11 +2531,10 @@ xfs_alloc_fix_freelist(
 	 */
 	if (!agbp) {
 		error = xfs_alloc_read_agf(mp, tp, args->agno, flags, &agbp);
-		if (error)
-			goto out_no_agbp;
-		if (!agbp) {
-			ASSERT(flags & XFS_ALLOC_FLAG_TRYLOCK);
-			ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING));
+		if (error) {
+			/* Couldn't lock the AGF so skip this AG. */
+			if (error == -EAGAIN)
+				error = 0;
 			goto out_no_agbp;
 		}
 	}
@@ -2394,7 +2654,7 @@ xfs_alloc_get_freelist(
 	xfs_agblock_t	*bnop,	/* block address retrieved from freelist */
 	int		btreeblk) /* destination is a AGF btree */
 {
-	xfs_agf_t	*agf;	/* a.g. freespace structure */
+	struct xfs_agf	*agf = agbp->b_addr;
 	xfs_buf_t	*agflbp;/* buffer for a.g. freelist structure */
 	xfs_agblock_t	bno;	/* block number returned */
 	__be32		*agfl_bno;
@@ -2406,7 +2666,6 @@ xfs_alloc_get_freelist(
 	/*
 	 * Freelist is empty, give up.
 	 */
-	agf = XFS_BUF_TO_AGF(agbp);
 	if (!agf->agf_flcount) {
 		*bnop = NULLAGBLOCK;
 		return 0;
@@ -2423,7 +2682,7 @@ xfs_alloc_get_freelist(
 	/*
 	 * Get the block number and update the data structures.
 	 */
-	agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, agflbp);
+	agfl_bno = xfs_buf_to_agfl_bno(agflbp);
 	bno = be32_to_cpu(agfl_bno[be32_to_cpu(agf->agf_flfirst)]);
 	be32_add_cpu(&agf->agf_flfirst, 1);
 	xfs_trans_brelse(tp, agflbp);
@@ -2484,7 +2743,7 @@ xfs_alloc_log_agf(
 		sizeof(xfs_agf_t)
 	};
 
-	trace_xfs_agf(tp->t_mountp, XFS_BUF_TO_AGF(bp), fields, _RET_IP_);
+	trace_xfs_agf(tp->t_mountp, bp->b_addr, fields, _RET_IP_);
 
 	xfs_trans_buf_set_type(tp, bp, XFS_BLFT_AGF_BUF);
 
@@ -2505,11 +2764,10 @@ xfs_alloc_pagf_init(
 	xfs_buf_t		*bp;
 	int			error;
 
-	if ((error = xfs_alloc_read_agf(mp, tp, agno, flags, &bp)))
-		return error;
-	if (bp)
+	error = xfs_alloc_read_agf(mp, tp, agno, flags, &bp);
+	if (!error)
 		xfs_trans_brelse(tp, bp);
-	return 0;
+	return error;
 }
 
 /*
@@ -2523,18 +2781,15 @@ xfs_alloc_put_freelist(
 	xfs_agblock_t		bno,	/* block being freed */
 	int			btreeblk) /* block came from a AGF btree */
 {
-	xfs_agf_t		*agf;	/* a.g. freespace structure */
+	struct xfs_mount	*mp = tp->t_mountp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	__be32			*blockp;/* pointer to array entry */
 	int			error;
 	int			logflags;
-	xfs_mount_t		*mp;	/* mount structure */
 	xfs_perag_t		*pag;	/* per allocation group data */
 	__be32			*agfl_bno;
 	int			startoff;
 
-	agf = XFS_BUF_TO_AGF(agbp);
-	mp = tp->t_mountp;
-
 	if (!agflbp && (error = xfs_alloc_read_agfl(mp, tp,
 			be32_to_cpu(agf->agf_seqno), &agflbp)))
 		return error;
@@ -2560,7 +2815,7 @@ xfs_alloc_put_freelist(
 
 	ASSERT(be32_to_cpu(agf->agf_flcount) <= xfs_agfl_size(mp));
 
-	agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, agflbp);
+	agfl_bno = xfs_buf_to_agfl_bno(agflbp);
 	blockp = &agfl_bno[be32_to_cpu(agf->agf_fllast)];
 	*blockp = cpu_to_be32(bno);
 	startoff = (char *)blockp - (char *)agflbp->b_addr;
@@ -2578,13 +2833,12 @@ xfs_agf_verify(
 	struct xfs_buf		*bp)
 {
 	struct xfs_mount	*mp = bp->b_mount;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(bp);
+	struct xfs_agf		*agf = bp->b_addr;
 
 	if (xfs_sb_version_hascrc(&mp->m_sb)) {
 		if (!uuid_equal(&agf->agf_uuid, &mp->m_sb.sb_meta_uuid))
 			return __this_address;
-		if (!xfs_log_check_lsn(mp,
-				be64_to_cpu(XFS_BUF_TO_AGF(bp)->agf_lsn)))
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(agf->agf_lsn)))
 			return __this_address;
 	}
 
@@ -2598,6 +2852,13 @@ xfs_agf_verify(
 	      be32_to_cpu(agf->agf_flcount) <= xfs_agfl_size(mp)))
 		return __this_address;
 
+	if (be32_to_cpu(agf->agf_length) > mp->m_sb.sb_dblocks)
+		return __this_address;
+
+	if (be32_to_cpu(agf->agf_freeblks) < be32_to_cpu(agf->agf_longest) ||
+	    be32_to_cpu(agf->agf_freeblks) > be32_to_cpu(agf->agf_length))
+		return __this_address;
+
 	if (be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) < 1 ||
 	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) < 1 ||
 	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) > XFS_BTREE_MAXLEVELS ||
@@ -2609,6 +2870,10 @@ xfs_agf_verify(
 	     be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) > XFS_BTREE_MAXLEVELS))
 		return __this_address;
 
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb) &&
+	    be32_to_cpu(agf->agf_rmap_blocks) > be32_to_cpu(agf->agf_length))
+		return __this_address;
+
 	/*
 	 * during growfs operations, the perag is not fully initialised,
 	 * so we can't use it for any useful checking. growfs ensures we can't
@@ -2623,6 +2888,11 @@ xfs_agf_verify(
 		return __this_address;
 
 	if (xfs_sb_version_hasreflink(&mp->m_sb) &&
+	    be32_to_cpu(agf->agf_refcount_blocks) >
+	    be32_to_cpu(agf->agf_length))
+		return __this_address;
+
+	if (xfs_sb_version_hasreflink(&mp->m_sb) &&
 	    (be32_to_cpu(agf->agf_refcount_level) < 1 ||
 	     be32_to_cpu(agf->agf_refcount_level) > XFS_BTREE_MAXLEVELS))
 		return __this_address;
@@ -2654,6 +2924,7 @@ xfs_agf_write_verify(
 {
 	struct xfs_mount	*mp = bp->b_mount;
 	struct xfs_buf_log_item	*bip = bp->b_log_item;
+	struct xfs_agf		*agf = bp->b_addr;
 	xfs_failaddr_t		fa;
 
 	fa = xfs_agf_verify(bp);
@@ -2666,7 +2937,7 @@ xfs_agf_write_verify(
 		return;
 
 	if (bip)
-		XFS_BUF_TO_AGF(bp)->agf_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+		agf->agf_lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
 	xfs_buf_update_cksum(bp, XFS_AGF_CRC_OFF);
 }
@@ -2695,14 +2966,11 @@ xfs_read_agf(
 	trace_xfs_read_agf(mp, agno);
 
 	ASSERT(agno != NULLAGNUMBER);
-	error = xfs_trans_read_buf(
-			mp, tp, mp->m_ddev_targp,
+	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
 			XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
 			XFS_FSS_TO_BB(mp, 1), flags, bpp, &xfs_agf_buf_ops);
 	if (error)
 		return error;
-	if (!*bpp)
-		return 0;
 
 	ASSERT(!(*bpp)->b_error);
 	xfs_buf_set_ref(*bpp, XFS_AGF_REF);
@@ -2726,17 +2994,18 @@ xfs_alloc_read_agf(
 
 	trace_xfs_alloc_read_agf(mp, agno);
 
+	/* We don't support trylock when freeing. */
+	ASSERT((flags & (XFS_ALLOC_FLAG_FREEING | XFS_ALLOC_FLAG_TRYLOCK)) !=
+			(XFS_ALLOC_FLAG_FREEING | XFS_ALLOC_FLAG_TRYLOCK));
 	ASSERT(agno != NULLAGNUMBER);
 	error = xfs_read_agf(mp, tp, agno,
 			(flags & XFS_ALLOC_FLAG_TRYLOCK) ? XBF_TRYLOCK : 0,
 			bpp);
 	if (error)
 		return error;
-	if (!*bpp)
-		return 0;
 	ASSERT(!(*bpp)->b_error);
 
-	agf = XFS_BUF_TO_AGF(*bpp);
+	agf = (*bpp)->b_addr;
 	pag = xfs_perag_get(mp, agno);
 	if (!pag->pagf_init) {
 		pag->pagf_freeblks = be32_to_cpu(agf->agf_freeblks);
@@ -2956,13 +3225,6 @@ xfs_alloc_vextent(
 			args->len);
 #endif
 
-		/* Zero the extent if we were asked to do so */
-		if (args->datatype & XFS_ALLOC_USERDATA_ZERO) {
-			error = xfs_zero_extent(args->ip, args->fsbno, args->len);
-			if (error)
-				goto error0;
-		}
-
 	}
 	xfs_perag_put(args->pag);
 	return 0;
@@ -3024,6 +3286,7 @@ __xfs_free_extent(
 	struct xfs_buf			*agbp;
 	xfs_agnumber_t			agno = XFS_FSB_TO_AGNO(mp, bno);
 	xfs_agblock_t			agbno = XFS_FSB_TO_AGBNO(mp, bno);
+	struct xfs_agf			*agf;
 	int				error;
 	unsigned int			busy_flags = 0;
 
@@ -3037,13 +3300,18 @@ __xfs_free_extent(
 	error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
 	if (error)
 		return error;
+	agf = agbp->b_addr;
 
-	XFS_WANT_CORRUPTED_GOTO(mp, agbno < mp->m_sb.sb_agblocks, err);
+	if (XFS_IS_CORRUPT(mp, agbno >= mp->m_sb.sb_agblocks)) {
+		error = -EFSCORRUPTED;
+		goto err;
+	}
 
 	/* validate the extent size is legal now we have the agf locked */
-	XFS_WANT_CORRUPTED_GOTO(mp,
-		agbno + len <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_length),
-				err);
+	if (XFS_IS_CORRUPT(mp, agbno + len > be32_to_cpu(agf->agf_length))) {
+		error = -EFSCORRUPTED;
+		goto err;
+	}
 
 	error = xfs_free_ag_extent(tp, agbp, agno, agbno, len, oinfo, type);
 	if (error)
@@ -3151,7 +3419,7 @@ xfs_agfl_walk(
 	unsigned int		i;
 	int			error;
 
-	agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, agflbp);
+	agfl_bno = xfs_buf_to_agfl_bno(agflbp);
 	i = be32_to_cpu(agf->agf_flfirst);
 
 	/* Nothing to walk in an empty AGFL. */
diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index d6ed5d2c07c2..a851bf77f17b 100644
--- a/fs/xfs/libxfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
@@ -54,7 +54,6 @@ typedef struct xfs_alloc_arg {
 	struct xfs_mount *mp;		/* file system mount point */
 	struct xfs_buf	*agbp;		/* buffer for a.g. freelist header */
 	struct xfs_perag *pag;		/* per-ag struct for this agno */
-	struct xfs_inode *ip;		/* for userdata zeroing method */
 	xfs_fsblock_t	fsbno;		/* file system block number */
 	xfs_agnumber_t	agno;		/* allocation group number */
 	xfs_agblock_t	agbno;		/* allocation group-relative block # */
@@ -83,20 +82,7 @@ typedef struct xfs_alloc_arg {
  */
 #define XFS_ALLOC_USERDATA		(1 << 0)/* allocation is for user data*/
 #define XFS_ALLOC_INITIAL_USER_DATA	(1 << 1)/* special case start of file */
-#define XFS_ALLOC_USERDATA_ZERO		(1 << 2)/* zero extent on allocation */
-#define XFS_ALLOC_NOBUSY		(1 << 3)/* Busy extents not allowed */
-
-static inline bool
-xfs_alloc_is_userdata(int datatype)
-{
-	return (datatype & ~XFS_ALLOC_NOBUSY) != 0;
-}
-
-static inline bool
-xfs_alloc_allow_busy_reuse(int datatype)
-{
-	return (datatype & XFS_ALLOC_NOBUSY) == 0;
-}
+#define XFS_ALLOC_NOBUSY		(1 << 2)/* Busy extents not allowed */
 
 /* freespace limit calculations */
 #define XFS_ALLOC_AGFL_RESERVE	4
@@ -250,4 +236,13 @@ typedef int (*xfs_agfl_walk_fn)(struct xfs_mount *mp, xfs_agblock_t bno,
 int xfs_agfl_walk(struct xfs_mount *mp, struct xfs_agf *agf,
 		struct xfs_buf *agflbp, xfs_agfl_walk_fn walk_fn, void *priv);
 
+static inline __be32 *
+xfs_buf_to_agfl_bno(
+	struct xfs_buf		*bp)
+{
+	if (xfs_sb_version_hascrc(&bp->b_mount->m_sb))
+		return bp->b_addr + sizeof(struct xfs_agfl);
+	return bp->b_addr;
+}
+
 #endif	/* __XFS_ALLOC_H__ */
diff --git a/fs/xfs/libxfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c
index 2a94543857a1..60c453cb3ee3 100644
--- a/fs/xfs/libxfs/xfs_alloc_btree.c
+++ b/fs/xfs/libxfs/xfs_alloc_btree.c
@@ -12,6 +12,7 @@
 #include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
+#include "xfs_btree_staging.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
@@ -25,7 +26,7 @@ xfs_allocbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_allocbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_private.a.agbp, cur->bc_private.a.agno,
+			cur->bc_ag.agbp, cur->bc_ag.agno,
 			cur->bc_btnum);
 }
 
@@ -35,8 +36,8 @@ xfs_allocbt_set_root(
 	union xfs_btree_ptr	*ptr,
 	int			inc)
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_agnumber_t		seqno = be32_to_cpu(agf->agf_seqno);
 	int			btnum = cur->bc_btnum;
 	struct xfs_perag	*pag = xfs_perag_get(cur->bc_mp, seqno);
@@ -62,7 +63,7 @@ xfs_allocbt_alloc_block(
 	xfs_agblock_t		bno;
 
 	/* Allocate the new block from the freelist. If we can't, give up.  */
-	error = xfs_alloc_get_freelist(cur->bc_tp, cur->bc_private.a.agbp,
+	error = xfs_alloc_get_freelist(cur->bc_tp, cur->bc_ag.agbp,
 				       &bno, 1);
 	if (error)
 		return error;
@@ -72,7 +73,7 @@ xfs_allocbt_alloc_block(
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_private.a.agno, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agno, bno, 1, false);
 
 	xfs_trans_agbtree_delta(cur->bc_tp, 1);
 	new->s = cpu_to_be32(bno);
@@ -86,8 +87,8 @@ xfs_allocbt_free_block(
 	struct xfs_btree_cur	*cur,
 	struct xfs_buf		*bp)
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_agblock_t		bno;
 	int			error;
 
@@ -113,7 +114,7 @@ xfs_allocbt_update_lastrec(
 	int			ptr,
 	int			reason)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp);
+	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
 	xfs_agnumber_t		seqno = be32_to_cpu(agf->agf_seqno);
 	struct xfs_perag	*pag;
 	__be32			len;
@@ -162,7 +163,7 @@ xfs_allocbt_update_lastrec(
 	pag = xfs_perag_get(cur->bc_mp, seqno);
 	pag->pagf_longest = be32_to_cpu(len);
 	xfs_perag_put(pag);
-	xfs_alloc_log_agf(cur->bc_tp, cur->bc_private.a.agbp, XFS_AGF_LONGEST);
+	xfs_alloc_log_agf(cur->bc_tp, cur->bc_ag.agbp, XFS_AGF_LONGEST);
 }
 
 STATIC int
@@ -226,9 +227,9 @@ xfs_allocbt_init_ptr_from_cur(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*ptr)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp);
+	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agf->agf_seqno));
+	ASSERT(cur->bc_ag.agno == be32_to_cpu(agf->agf_seqno));
 
 	ptr->s = agf->agf_roots[cur->bc_btnum];
 }
@@ -471,18 +472,14 @@ static const struct xfs_btree_ops xfs_cntbt_ops = {
 	.recs_inorder		= xfs_cntbt_recs_inorder,
 };
 
-/*
- * Allocate a new allocation btree cursor.
- */
-struct xfs_btree_cur *			/* new alloc btree cursor */
-xfs_allocbt_init_cursor(
-	struct xfs_mount	*mp,		/* file system mount point */
-	struct xfs_trans	*tp,		/* transaction pointer */
-	struct xfs_buf		*agbp,		/* buffer for agf structure */
-	xfs_agnumber_t		agno,		/* allocation group number */
-	xfs_btnum_t		btnum)		/* btree identifier */
+/* Allocate most of a new allocation btree cursor. */
+STATIC struct xfs_btree_cur *
+xfs_allocbt_init_common(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	xfs_agnumber_t		agno,
+	xfs_btnum_t		btnum)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
 	struct xfs_btree_cur	*cur;
 
 	ASSERT(btnum == XFS_BTNUM_BNO || btnum == XFS_BTNUM_CNT);
@@ -495,18 +492,16 @@ xfs_allocbt_init_cursor(
 	cur->bc_blocklog = mp->m_sb.sb_blocklog;
 
 	if (btnum == XFS_BTNUM_CNT) {
-		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_abtc_2);
 		cur->bc_ops = &xfs_cntbt_ops;
-		cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]);
+		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_abtc_2);
 		cur->bc_flags = XFS_BTREE_LASTREC_UPDATE;
 	} else {
-		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_abtb_2);
 		cur->bc_ops = &xfs_bnobt_ops;
-		cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]);
+		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_abtb_2);
 	}
 
-	cur->bc_private.a.agbp = agbp;
-	cur->bc_private.a.agno = agno;
+	cur->bc_ag.agno = agno;
+	cur->bc_ag.abt.active = false;
 
 	if (xfs_sb_version_hascrc(&mp->m_sb))
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
@@ -515,6 +510,73 @@ xfs_allocbt_init_cursor(
 }
 
 /*
+ * Allocate a new allocation btree cursor.
+ */
+struct xfs_btree_cur *			/* new alloc btree cursor */
+xfs_allocbt_init_cursor(
+	struct xfs_mount	*mp,		/* file system mount point */
+	struct xfs_trans	*tp,		/* transaction pointer */
+	struct xfs_buf		*agbp,		/* buffer for agf structure */
+	xfs_agnumber_t		agno,		/* allocation group number */
+	xfs_btnum_t		btnum)		/* btree identifier */
+{
+	struct xfs_agf		*agf = agbp->b_addr;
+	struct xfs_btree_cur	*cur;
+
+	cur = xfs_allocbt_init_common(mp, tp, agno, btnum);
+	if (btnum == XFS_BTNUM_CNT)
+		cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]);
+	else
+		cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]);
+
+	cur->bc_ag.agbp = agbp;
+
+	return cur;
+}
+
+/* Create a free space btree cursor with a fake root for staging. */
+struct xfs_btree_cur *
+xfs_allocbt_stage_cursor(
+	struct xfs_mount	*mp,
+	struct xbtree_afakeroot	*afake,
+	xfs_agnumber_t		agno,
+	xfs_btnum_t		btnum)
+{
+	struct xfs_btree_cur	*cur;
+
+	cur = xfs_allocbt_init_common(mp, NULL, agno, btnum);
+	xfs_btree_stage_afakeroot(cur, afake);
+	return cur;
+}
+
+/*
+ * Install a new free space btree root.  Caller is responsible for invalidating
+ * and freeing the old btree blocks.
+ */
+void
+xfs_allocbt_commit_staged_btree(
+	struct xfs_btree_cur	*cur,
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp)
+{
+	struct xfs_agf		*agf = agbp->b_addr;
+	struct xbtree_afakeroot	*afake = cur->bc_ag.afake;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	agf->agf_roots[cur->bc_btnum] = cpu_to_be32(afake->af_root);
+	agf->agf_levels[cur->bc_btnum] = cpu_to_be32(afake->af_levels);
+	xfs_alloc_log_agf(tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS);
+
+	if (cur->bc_btnum == XFS_BTNUM_BNO) {
+		xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_bnobt_ops);
+	} else {
+		cur->bc_flags |= XFS_BTREE_LASTREC_UPDATE;
+		xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_cntbt_ops);
+	}
+}
+
+/*
  * Calculate number of records in an alloc btree block.
  */
 int
diff --git a/fs/xfs/libxfs/xfs_alloc_btree.h b/fs/xfs/libxfs/xfs_alloc_btree.h
index c9305ebb69f6..047f09f0be3c 100644
--- a/fs/xfs/libxfs/xfs_alloc_btree.h
+++ b/fs/xfs/libxfs/xfs_alloc_btree.h
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xbtree_afakeroot;
 
 /*
  * Btree block header size depends on a superblock flag.
@@ -48,8 +49,14 @@ struct xfs_mount;
 extern struct xfs_btree_cur *xfs_allocbt_init_cursor(struct xfs_mount *,
 		struct xfs_trans *, struct xfs_buf *,
 		xfs_agnumber_t, xfs_btnum_t);
+struct xfs_btree_cur *xfs_allocbt_stage_cursor(struct xfs_mount *mp,
+		struct xbtree_afakeroot *afake, xfs_agnumber_t agno,
+		xfs_btnum_t btnum);
 extern int xfs_allocbt_maxrecs(struct xfs_mount *, int, int);
 extern xfs_extlen_t xfs_allocbt_calc_size(struct xfs_mount *mp,
 		unsigned long long len);
 
+void xfs_allocbt_commit_staged_btree(struct xfs_btree_cur *cur,
+		struct xfs_trans *tp, struct xfs_buf *agbp);
+
 #endif	/* __XFS_ALLOC_BTREE_H__ */
diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c
index 510ca6974604..e4fe3dca9883 100644
--- a/fs/xfs/libxfs/xfs_attr.c
+++ b/fs/xfs/libxfs/xfs_attr.c
@@ -56,32 +56,6 @@ STATIC int xfs_attr_node_removename(xfs_da_args_t *args);
 STATIC int xfs_attr_fillstate(xfs_da_state_t *state);
 STATIC int xfs_attr_refillstate(xfs_da_state_t *state);
 
-
-STATIC int
-xfs_attr_args_init(
-	struct xfs_da_args	*args,
-	struct xfs_inode	*dp,
-	const unsigned char	*name,
-	int			flags)
-{
-
-	if (!name)
-		return -EINVAL;
-
-	memset(args, 0, sizeof(*args));
-	args->geo = dp->i_mount->m_attr_geo;
-	args->whichfork = XFS_ATTR_FORK;
-	args->dp = dp;
-	args->flags = flags;
-	args->name = name;
-	args->namelen = strlen((const char *)name);
-	if (args->namelen >= MAXNAMELEN)
-		return -EFAULT;		/* match IRIX behaviour */
-
-	args->hashval = xfs_da_hashname(args->name, args->namelen);
-	return 0;
-}
-
 int
 xfs_inode_hasattr(
 	struct xfs_inode	*ip)
@@ -103,84 +77,60 @@ xfs_inode_hasattr(
  */
 int
 xfs_attr_get_ilocked(
-	struct xfs_inode	*ip,
 	struct xfs_da_args	*args)
 {
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL));
+	ASSERT(xfs_isilocked(args->dp, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL));
 
-	if (!xfs_inode_hasattr(ip))
+	if (!xfs_inode_hasattr(args->dp))
 		return -ENOATTR;
-	else if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL)
+
+	if (args->dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL)
 		return xfs_attr_shortform_getvalue(args);
-	else if (xfs_bmap_one_block(ip, XFS_ATTR_FORK))
+	if (xfs_bmap_one_block(args->dp, XFS_ATTR_FORK))
 		return xfs_attr_leaf_get(args);
-	else
-		return xfs_attr_node_get(args);
+	return xfs_attr_node_get(args);
 }
 
 /*
  * Retrieve an extended attribute by name, and its value if requested.
  *
- * If ATTR_KERNOVAL is set in @flags, then the caller does not want the value,
- * just an indication whether the attribute exists and the size of the value if
- * it exists. The size is returned in @valuelenp,
+ * If args->valuelen is zero, then the caller does not want the value, just an
+ * indication whether the attribute exists and the size of the value if it
+ * exists. The size is returned in args.valuelen.
  *
- * If the attribute is found, but exceeds the size limit set by the caller in
- * @valuelenp, return -ERANGE with the size of the attribute that was found in
- * @valuelenp.
+ * If args->value is NULL but args->valuelen is non-zero, allocate the buffer
+ * for the value after existence of the attribute has been determined. The
+ * caller always has to free args->value if it is set, no matter if this
+ * function was successful or not.
  *
- * If ATTR_ALLOC is set in @flags, allocate the buffer for the value after
- * existence of the attribute has been determined. On success, return that
- * buffer to the caller and leave them to free it. On failure, free any
- * allocated buffer and ensure the buffer pointer returned to the caller is
- * null.
+ * If the attribute is found, but exceeds the size limit set by the caller in
+ * args->valuelen, return -ERANGE with the size of the attribute that was found
+ * in args->valuelen.
  */
 int
 xfs_attr_get(
-	struct xfs_inode	*ip,
-	const unsigned char	*name,
-	unsigned char		**value,
-	int			*valuelenp,
-	int			flags)
+	struct xfs_da_args	*args)
 {
-	struct xfs_da_args	args;
 	uint			lock_mode;
 	int			error;
 
-	ASSERT((flags & (ATTR_ALLOC | ATTR_KERNOVAL)) || *value);
-
-	XFS_STATS_INC(ip->i_mount, xs_attr_get);
+	XFS_STATS_INC(args->dp->i_mount, xs_attr_get);
 
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+	if (XFS_FORCED_SHUTDOWN(args->dp->i_mount))
 		return -EIO;
 
-	error = xfs_attr_args_init(&args, ip, name, flags);
-	if (error)
-		return error;
+	args->geo = args->dp->i_mount->m_attr_geo;
+	args->whichfork = XFS_ATTR_FORK;
+	args->hashval = xfs_da_hashname(args->name, args->namelen);
 
 	/* Entirely possible to look up a name which doesn't exist */
-	args.op_flags = XFS_DA_OP_OKNOENT;
-	if (flags & ATTR_ALLOC)
-		args.op_flags |= XFS_DA_OP_ALLOCVAL;
-	else
-		args.value = *value;
-	args.valuelen = *valuelenp;
+	args->op_flags = XFS_DA_OP_OKNOENT;
 
-	lock_mode = xfs_ilock_attr_map_shared(ip);
-	error = xfs_attr_get_ilocked(ip, &args);
-	xfs_iunlock(ip, lock_mode);
-	*valuelenp = args.valuelen;
+	lock_mode = xfs_ilock_attr_map_shared(args->dp);
+	error = xfs_attr_get_ilocked(args);
+	xfs_iunlock(args->dp, lock_mode);
 
-	/* on error, we have to clean up allocated value buffers */
-	if (error) {
-		if (flags & ATTR_ALLOC) {
-			kmem_free(args.value);
-			*value = NULL;
-		}
-		return error;
-	}
-	*value = args.value;
-	return 0;
+	return error;
 }
 
 /*
@@ -236,7 +186,7 @@ xfs_attr_try_sf_addname(
 	 * Commit the shortform mods, and we're done.
 	 * NOTE: this is also the error path (EEXIST, etc).
 	 */
-	if (!error && (args->flags & ATTR_KERNOTIME) == 0)
+	if (!error && !(args->op_flags & XFS_DA_OP_NOTIME))
 		xfs_trans_ichgtime(args->trans, dp, XFS_ICHGTIME_CHG);
 
 	if (mp->m_flags & XFS_MOUNT_WSYNC)
@@ -334,186 +284,127 @@ xfs_attr_remove_args(
 	return error;
 }
 
+/*
+ * Note: If args->value is NULL the attribute will be removed, just like the
+ * Linux ->setattr API.
+ */
 int
 xfs_attr_set(
-	struct xfs_inode	*dp,
-	const unsigned char	*name,
-	unsigned char		*value,
-	int			valuelen,
-	int			flags)
+	struct xfs_da_args	*args)
 {
+	struct xfs_inode	*dp = args->dp;
 	struct xfs_mount	*mp = dp->i_mount;
-	struct xfs_da_args	args;
 	struct xfs_trans_res	tres;
-	int			rsvd = (flags & ATTR_ROOT) != 0;
+	bool			rsvd = (args->attr_filter & XFS_ATTR_ROOT);
 	int			error, local;
-
-	XFS_STATS_INC(mp, xs_attr_set);
+	unsigned int		total;
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
 		return -EIO;
 
-	error = xfs_attr_args_init(&args, dp, name, flags);
-	if (error)
-		return error;
-
-	args.value = value;
-	args.valuelen = valuelen;
-	args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
-	args.total = xfs_attr_calc_size(&args, &local);
-
 	error = xfs_qm_dqattach(dp);
 	if (error)
 		return error;
 
-	/*
-	 * If the inode doesn't have an attribute fork, add one.
-	 * (inode must not be locked when we call this routine)
-	 */
-	if (XFS_IFORK_Q(dp) == 0) {
-		int sf_size = sizeof(xfs_attr_sf_hdr_t) +
-			XFS_ATTR_SF_ENTSIZE_BYNAME(args.namelen, valuelen);
-
-		error = xfs_bmap_add_attrfork(dp, sf_size, rsvd);
-		if (error)
-			return error;
-	}
-
-	tres.tr_logres = M_RES(mp)->tr_attrsetm.tr_logres +
-			 M_RES(mp)->tr_attrsetrt.tr_logres * args.total;
-	tres.tr_logcount = XFS_ATTRSET_LOG_COUNT;
-	tres.tr_logflags = XFS_TRANS_PERM_LOG_RES;
-
-	/*
-	 * Root fork attributes can use reserved data blocks for this
-	 * operation if necessary
-	 */
-	error = xfs_trans_alloc(mp, &tres, args.total, 0,
-			rsvd ? XFS_TRANS_RESERVE : 0, &args.trans);
-	if (error)
-		return error;
-
-	xfs_ilock(dp, XFS_ILOCK_EXCL);
-	error = xfs_trans_reserve_quota_nblks(args.trans, dp, args.total, 0,
-				rsvd ? XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
-				       XFS_QMOPT_RES_REGBLKS);
-	if (error)
-		goto out_trans_cancel;
-
-	xfs_trans_ijoin(args.trans, dp, 0);
-	error = xfs_attr_set_args(&args);
-	if (error)
-		goto out_trans_cancel;
-	if (!args.trans) {
-		/* shortform attribute has already been committed */
-		goto out_unlock;
-	}
-
-	/*
-	 * If this is a synchronous mount, make sure that the
-	 * transaction goes to disk before returning to the user.
-	 */
-	if (mp->m_flags & XFS_MOUNT_WSYNC)
-		xfs_trans_set_sync(args.trans);
-
-	if ((flags & ATTR_KERNOTIME) == 0)
-		xfs_trans_ichgtime(args.trans, dp, XFS_ICHGTIME_CHG);
+	args->geo = mp->m_attr_geo;
+	args->whichfork = XFS_ATTR_FORK;
+	args->hashval = xfs_da_hashname(args->name, args->namelen);
 
 	/*
-	 * Commit the last in the sequence of transactions.
+	 * We have no control over the attribute names that userspace passes us
+	 * to remove, so we have to allow the name lookup prior to attribute
+	 * removal to fail as well.
 	 */
-	xfs_trans_log_inode(args.trans, dp, XFS_ILOG_CORE);
-	error = xfs_trans_commit(args.trans);
-out_unlock:
-	xfs_iunlock(dp, XFS_ILOCK_EXCL);
-	return error;
+	args->op_flags = XFS_DA_OP_OKNOENT;
 
-out_trans_cancel:
-	if (args.trans)
-		xfs_trans_cancel(args.trans);
-	goto out_unlock;
-}
+	if (args->value) {
+		XFS_STATS_INC(mp, xs_attr_set);
 
-/*
- * Generic handler routine to remove a name from an attribute list.
- * Transitions attribute list from Btree to shortform as necessary.
- */
-int
-xfs_attr_remove(
-	struct xfs_inode	*dp,
-	const unsigned char	*name,
-	int			flags)
-{
-	struct xfs_mount	*mp = dp->i_mount;
-	struct xfs_da_args	args;
-	int			error;
-
-	XFS_STATS_INC(mp, xs_attr_remove);
+		args->op_flags |= XFS_DA_OP_ADDNAME;
+		args->total = xfs_attr_calc_size(args, &local);
 
-	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return -EIO;
+		/*
+		 * If the inode doesn't have an attribute fork, add one.
+		 * (inode must not be locked when we call this routine)
+		 */
+		if (XFS_IFORK_Q(dp) == 0) {
+			int sf_size = sizeof(struct xfs_attr_sf_hdr) +
+				XFS_ATTR_SF_ENTSIZE_BYNAME(args->namelen,
+						args->valuelen);
 
-	error = xfs_attr_args_init(&args, dp, name, flags);
-	if (error)
-		return error;
+			error = xfs_bmap_add_attrfork(dp, sf_size, rsvd);
+			if (error)
+				return error;
+		}
 
-	/*
-	 * we have no control over the attribute names that userspace passes us
-	 * to remove, so we have to allow the name lookup prior to attribute
-	 * removal to fail.
-	 */
-	args.op_flags = XFS_DA_OP_OKNOENT;
+		tres.tr_logres = M_RES(mp)->tr_attrsetm.tr_logres +
+				 M_RES(mp)->tr_attrsetrt.tr_logres *
+					args->total;
+		tres.tr_logcount = XFS_ATTRSET_LOG_COUNT;
+		tres.tr_logflags = XFS_TRANS_PERM_LOG_RES;
+		total = args->total;
+	} else {
+		XFS_STATS_INC(mp, xs_attr_remove);
 
-	error = xfs_qm_dqattach(dp);
-	if (error)
-		return error;
+		tres = M_RES(mp)->tr_attrrm;
+		total = XFS_ATTRRM_SPACE_RES(mp);
+	}
 
 	/*
 	 * Root fork attributes can use reserved data blocks for this
 	 * operation if necessary
 	 */
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_attrrm,
-			XFS_ATTRRM_SPACE_RES(mp), 0,
-			(flags & ATTR_ROOT) ? XFS_TRANS_RESERVE : 0,
-			&args.trans);
+	error = xfs_trans_alloc(mp, &tres, total, 0,
+			rsvd ? XFS_TRANS_RESERVE : 0, &args->trans);
 	if (error)
 		return error;
 
 	xfs_ilock(dp, XFS_ILOCK_EXCL);
-	/*
-	 * No need to make quota reservations here. We expect to release some
-	 * blocks not allocate in the common case.
-	 */
-	xfs_trans_ijoin(args.trans, dp, 0);
-
-	error = xfs_attr_remove_args(&args);
-	if (error)
-		goto out;
+	xfs_trans_ijoin(args->trans, dp, 0);
+	if (args->value) {
+		unsigned int	quota_flags = XFS_QMOPT_RES_REGBLKS;
+
+		if (rsvd)
+			quota_flags |= XFS_QMOPT_FORCE_RES;
+		error = xfs_trans_reserve_quota_nblks(args->trans, dp,
+				args->total, 0, quota_flags);
+		if (error)
+			goto out_trans_cancel;
+		error = xfs_attr_set_args(args);
+		if (error)
+			goto out_trans_cancel;
+		/* shortform attribute has already been committed */
+		if (!args->trans)
+			goto out_unlock;
+	} else {
+		error = xfs_attr_remove_args(args);
+		if (error)
+			goto out_trans_cancel;
+	}
 
 	/*
 	 * If this is a synchronous mount, make sure that the
 	 * transaction goes to disk before returning to the user.
 	 */
 	if (mp->m_flags & XFS_MOUNT_WSYNC)
-		xfs_trans_set_sync(args.trans);
+		xfs_trans_set_sync(args->trans);
 
-	if ((flags & ATTR_KERNOTIME) == 0)
-		xfs_trans_ichgtime(args.trans, dp, XFS_ICHGTIME_CHG);
+	if (!(args->op_flags & XFS_DA_OP_NOTIME))
+		xfs_trans_ichgtime(args->trans, dp, XFS_ICHGTIME_CHG);
 
 	/*
 	 * Commit the last in the sequence of transactions.
 	 */
-	xfs_trans_log_inode(args.trans, dp, XFS_ILOG_CORE);
-	error = xfs_trans_commit(args.trans);
+	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE);
+	error = xfs_trans_commit(args->trans);
+out_unlock:
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
-
 	return error;
 
-out:
-	if (args.trans)
-		xfs_trans_cancel(args.trans);
-	xfs_iunlock(dp, XFS_ILOCK_EXCL);
-	return error;
+out_trans_cancel:
+	if (args->trans)
+		xfs_trans_cancel(args->trans);
+	goto out_unlock;
 }
 
 /*========================================================================
@@ -532,10 +423,10 @@ xfs_attr_shortform_addname(xfs_da_args_t *args)
 	trace_xfs_attr_sf_addname(args);
 
 	retval = xfs_attr_shortform_lookup(args);
-	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
+	if (retval == -ENOATTR && (args->attr_flags & XATTR_REPLACE))
 		return retval;
-	} else if (retval == -EEXIST) {
-		if (args->flags & ATTR_CREATE)
+	if (retval == -EEXIST) {
+		if (args->attr_flags & XATTR_CREATE)
 			return retval;
 		retval = xfs_attr_shortform_remove(args);
 		if (retval)
@@ -545,7 +436,7 @@ xfs_attr_shortform_addname(xfs_da_args_t *args)
 		 * that the leaf format add routine won't trip over the attr
 		 * not being around.
 		 */
-		args->flags &= ~ATTR_REPLACE;
+		args->attr_flags &= ~XATTR_REPLACE;
 	}
 
 	if (args->namelen >= XFS_ATTR_SF_ENTSIZE_MAX ||
@@ -589,7 +480,7 @@ xfs_attr_leaf_addname(
 	 */
 	dp = args->dp;
 	args->blkno = 0;
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
@@ -598,14 +489,11 @@ xfs_attr_leaf_addname(
 	 * the given flags produce an error or call for an atomic rename.
 	 */
 	retval = xfs_attr3_leaf_lookup_int(bp, args);
-	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
-		xfs_trans_brelse(args->trans, bp);
-		return retval;
-	} else if (retval == -EEXIST) {
-		if (args->flags & ATTR_CREATE) {	/* pure create op */
-			xfs_trans_brelse(args->trans, bp);
-			return retval;
-		}
+	if (retval == -ENOATTR && (args->attr_flags & XATTR_REPLACE))
+		goto out_brelse;
+	if (retval == -EEXIST) {
+		if (args->attr_flags & XATTR_CREATE)
+			goto out_brelse;
 
 		trace_xfs_attr_leaf_replace(args);
 
@@ -715,7 +603,7 @@ xfs_attr_leaf_addname(
 		 * remove the "old" attr from that block (neat, huh!)
 		 */
 		error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno,
-					   -1, &bp);
+					   &bp);
 		if (error)
 			return error;
 
@@ -746,6 +634,9 @@ xfs_attr_leaf_addname(
 		error = xfs_attr3_leaf_clearflag(args);
 	}
 	return error;
+out_brelse:
+	xfs_trans_brelse(args->trans, bp);
+	return retval;
 }
 
 /*
@@ -769,7 +660,7 @@ xfs_attr_leaf_removename(
 	 */
 	dp = args->dp;
 	args->blkno = 0;
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
@@ -813,7 +704,7 @@ xfs_attr_leaf_get(xfs_da_args_t *args)
 	trace_xfs_attr_leaf_get(args);
 
 	args->blkno = 0;
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
@@ -872,10 +763,10 @@ restart:
 		goto out;
 	blk = &state->path.blk[ state->path.active-1 ];
 	ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
-	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
+	if (retval == -ENOATTR && (args->attr_flags & XATTR_REPLACE))
 		goto out;
-	} else if (retval == -EEXIST) {
-		if (args->flags & ATTR_CREATE)
+	if (retval == -EEXIST) {
+		if (args->attr_flags & XATTR_CREATE)
 			goto out;
 
 		trace_xfs_attr_node_replace(args);
@@ -1007,7 +898,7 @@ restart:
 		 * The INCOMPLETE flag means that we will find the "old"
 		 * attr, not the "new" one.
 		 */
-		args->flags |= XFS_ATTR_INCOMPLETE;
+		args->attr_filter |= XFS_ATTR_INCOMPLETE;
 		state = xfs_da_state_alloc();
 		state->args = args;
 		state->mp = mp;
@@ -1173,7 +1064,7 @@ xfs_attr_node_removename(
 		ASSERT(state->path.blk[0].bp);
 		state->path.blk[0].bp = NULL;
 
-		error = xfs_attr3_leaf_read(args->trans, args->dp, 0, -1, &bp);
+		error = xfs_attr3_leaf_read(args->trans, args->dp, 0, &bp);
 		if (error)
 			goto out;
 
@@ -1266,10 +1157,9 @@ xfs_attr_refillstate(xfs_da_state_t *state)
 	ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
 	for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
 		if (blk->disk_blkno) {
-			error = xfs_da3_node_read(state->args->trans,
-						state->args->dp,
-						blk->blkno, blk->disk_blkno,
-						&blk->bp, XFS_ATTR_FORK);
+			error = xfs_da3_node_read_mapped(state->args->trans,
+					state->args->dp, blk->disk_blkno,
+					&blk->bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
 		} else {
@@ -1285,10 +1175,9 @@ xfs_attr_refillstate(xfs_da_state_t *state)
 	ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
 	for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
 		if (blk->disk_blkno) {
-			error = xfs_da3_node_read(state->args->trans,
-						state->args->dp,
-						blk->blkno, blk->disk_blkno,
-						&blk->bp, XFS_ATTR_FORK);
+			error = xfs_da3_node_read_mapped(state->args->trans,
+					state->args->dp, blk->disk_blkno,
+					&blk->bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
 		} else {
diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h
index 94badfa1743e..0d2d05908537 100644
--- a/fs/xfs/libxfs/xfs_attr.h
+++ b/fs/xfs/libxfs/xfs_attr.h
@@ -21,36 +21,6 @@ struct xfs_attr_list_context;
  * as possible so as to fit into the literal area of the inode.
  */
 
-/*========================================================================
- * External interfaces
- *========================================================================*/
-
-
-#define ATTR_DONTFOLLOW	0x0001	/* -- unused, from IRIX -- */
-#define ATTR_ROOT	0x0002	/* use attrs in root (trusted) namespace */
-#define ATTR_TRUST	0x0004	/* -- unused, from IRIX -- */
-#define ATTR_SECURE	0x0008	/* use attrs in security namespace */
-#define ATTR_CREATE	0x0010	/* pure create: fail if attr already exists */
-#define ATTR_REPLACE	0x0020	/* pure set: fail if attr does not exist */
-
-#define ATTR_KERNOTIME	0x1000	/* [kernel] don't update inode timestamps */
-#define ATTR_KERNOVAL	0x2000	/* [kernel] get attr size only, not value */
-
-#define ATTR_INCOMPLETE	0x4000	/* [kernel] return INCOMPLETE attr keys */
-#define ATTR_ALLOC	0x8000	/* allocate xattr buffer on demand */
-
-#define XFS_ATTR_FLAGS \
-	{ ATTR_DONTFOLLOW, 	"DONTFOLLOW" }, \
-	{ ATTR_ROOT,		"ROOT" }, \
-	{ ATTR_TRUST,		"TRUST" }, \
-	{ ATTR_SECURE,		"SECURE" }, \
-	{ ATTR_CREATE,		"CREATE" }, \
-	{ ATTR_REPLACE,		"REPLACE" }, \
-	{ ATTR_KERNOTIME,	"KERNOTIME" }, \
-	{ ATTR_KERNOVAL,	"KERNOVAL" }, \
-	{ ATTR_INCOMPLETE,	"INCOMPLETE" }, \
-	{ ATTR_ALLOC,		"ALLOC" }
-
 /*
  * The maximum size (into the kernel or returned from the kernel) of an
  * attribute value or the buffer used for an attr_list() call.  Larger
@@ -59,45 +29,16 @@ struct xfs_attr_list_context;
 #define	ATTR_MAX_VALUELEN	(64*1024)	/* max length of a value */
 
 /*
- * Define how lists of attribute names are returned to the user from
- * the attr_list() call.  A large, 32bit aligned, buffer is passed in
- * along with its size.  We put an array of offsets at the top that each
- * reference an attrlist_ent_t and pack the attrlist_ent_t's at the bottom.
- */
-typedef struct attrlist {
-	__s32	al_count;	/* number of entries in attrlist */
-	__s32	al_more;	/* T/F: more attrs (do call again) */
-	__s32	al_offset[1];	/* byte offsets of attrs [var-sized] */
-} attrlist_t;
-
-/*
- * Show the interesting info about one attribute.  This is what the
- * al_offset[i] entry points to.
- */
-typedef struct attrlist_ent {	/* data from attr_list() */
-	__u32	a_valuelen;	/* number bytes in value of attr */
-	char	a_name[1];	/* attr name (NULL terminated) */
-} attrlist_ent_t;
-
-/*
- * Given a pointer to the (char*) buffer containing the attr_list() result,
- * and an index, return a pointer to the indicated attribute in the buffer.
- */
-#define	ATTR_ENTRY(buffer, index)		\
-	((attrlist_ent_t *)			\
-	 &((char *)buffer)[ ((attrlist_t *)(buffer))->al_offset[index] ])
-
-/*
  * Kernel-internal version of the attrlist cursor.
  */
-typedef struct attrlist_cursor_kern {
+struct xfs_attrlist_cursor_kern {
 	__u32	hashval;	/* hash value of next entry to add */
 	__u32	blkno;		/* block containing entry (suggestion) */
 	__u32	offset;		/* offset in list of equal-hashvals */
 	__u16	pad1;		/* padding to match user-level */
 	__u8	pad2;		/* padding to match user-level */
 	__u8	initted;	/* T/F: cursor has been initialized */
-} attrlist_cursor_kern_t;
+};
 
 
 /*========================================================================
@@ -109,27 +50,28 @@ typedef struct attrlist_cursor_kern {
 typedef void (*put_listent_func_t)(struct xfs_attr_list_context *, int,
 			      unsigned char *, int, int);
 
-typedef struct xfs_attr_list_context {
-	struct xfs_trans		*tp;
-	struct xfs_inode		*dp;		/* inode */
-	struct attrlist_cursor_kern	*cursor;	/* position in list */
-	char				*alist;		/* output buffer */
+struct xfs_attr_list_context {
+	struct xfs_trans	*tp;
+	struct xfs_inode	*dp;		/* inode */
+	struct xfs_attrlist_cursor_kern cursor;	/* position in list */
+	void			*buffer;	/* output buffer */
 
 	/*
 	 * Abort attribute list iteration if non-zero.  Can be used to pass
 	 * error values to the xfs_attr_list caller.
 	 */
-	int				seen_enough;
+	int			seen_enough;
+	bool			allow_incomplete;
 
-	ssize_t				count;		/* num used entries */
-	int				dupcnt;		/* count dup hashvals seen */
-	int				bufsize;	/* total buffer size */
-	int				firstu;		/* first used byte in buffer */
-	int				flags;		/* from VOP call */
-	int				resynch;	/* T/F: resynch with cursor */
-	put_listent_func_t		put_listent;	/* list output fmt function */
-	int				index;		/* index into output buffer */
-} xfs_attr_list_context_t;
+	ssize_t			count;		/* num used entries */
+	int			dupcnt;		/* count dup hashvals seen */
+	int			bufsize;	/* total buffer size */
+	int			firstu;		/* first used byte in buffer */
+	unsigned int		attr_filter;	/* XFS_ATTR_{ROOT,SECURE} */
+	int			resynch;	/* T/F: resynch with cursor */
+	put_listent_func_t	put_listent;	/* list output fmt function */
+	int			index;		/* index into output buffer */
+};
 
 
 /*========================================================================
@@ -140,19 +82,14 @@ typedef struct xfs_attr_list_context {
  * Overall external interface routines.
  */
 int xfs_attr_inactive(struct xfs_inode *dp);
-int xfs_attr_list_int_ilocked(struct xfs_attr_list_context *);
-int xfs_attr_list_int(struct xfs_attr_list_context *);
+int xfs_attr_list_ilocked(struct xfs_attr_list_context *);
+int xfs_attr_list(struct xfs_attr_list_context *);
 int xfs_inode_hasattr(struct xfs_inode *ip);
-int xfs_attr_get_ilocked(struct xfs_inode *ip, struct xfs_da_args *args);
-int xfs_attr_get(struct xfs_inode *ip, const unsigned char *name,
-		 unsigned char **value, int *valuelenp, int flags);
-int xfs_attr_set(struct xfs_inode *dp, const unsigned char *name,
-		 unsigned char *value, int valuelen, int flags);
+int xfs_attr_get_ilocked(struct xfs_da_args *args);
+int xfs_attr_get(struct xfs_da_args *args);
+int xfs_attr_set(struct xfs_da_args *args);
 int xfs_attr_set_args(struct xfs_da_args *args);
-int xfs_attr_remove(struct xfs_inode *dp, const unsigned char *name, int flags);
 int xfs_attr_remove_args(struct xfs_da_args *args);
-int xfs_attr_list(struct xfs_inode *dp, char *buffer, int bufsize,
-		  int flags, struct attrlist_cursor_kern *cursor);
 bool xfs_attr_namecheck(const void *name, size_t length);
 
 #endif	/* __XFS_ATTR_H__ */
diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c
index f0089e862216..863444e2dda7 100644
--- a/fs/xfs/libxfs/xfs_attr_leaf.c
+++ b/fs/xfs/libxfs/xfs_attr_leaf.c
@@ -233,6 +233,61 @@ xfs_attr3_leaf_hdr_to_disk(
 }
 
 static xfs_failaddr_t
+xfs_attr3_leaf_verify_entry(
+	struct xfs_mount			*mp,
+	char					*buf_end,
+	struct xfs_attr_leafblock		*leaf,
+	struct xfs_attr3_icleaf_hdr		*leafhdr,
+	struct xfs_attr_leaf_entry		*ent,
+	int					idx,
+	__u32					*last_hashval)
+{
+	struct xfs_attr_leaf_name_local		*lentry;
+	struct xfs_attr_leaf_name_remote	*rentry;
+	char					*name_end;
+	unsigned int				nameidx;
+	unsigned int				namesize;
+	__u32					hashval;
+
+	/* hash order check */
+	hashval = be32_to_cpu(ent->hashval);
+	if (hashval < *last_hashval)
+		return __this_address;
+	*last_hashval = hashval;
+
+	nameidx = be16_to_cpu(ent->nameidx);
+	if (nameidx < leafhdr->firstused || nameidx >= mp->m_attr_geo->blksize)
+		return __this_address;
+
+	/*
+	 * Check the name information.  The namelen fields are u8 so we can't
+	 * possibly exceed the maximum name length of 255 bytes.
+	 */
+	if (ent->flags & XFS_ATTR_LOCAL) {
+		lentry = xfs_attr3_leaf_name_local(leaf, idx);
+		namesize = xfs_attr_leaf_entsize_local(lentry->namelen,
+				be16_to_cpu(lentry->valuelen));
+		name_end = (char *)lentry + namesize;
+		if (lentry->namelen == 0)
+			return __this_address;
+	} else {
+		rentry = xfs_attr3_leaf_name_remote(leaf, idx);
+		namesize = xfs_attr_leaf_entsize_remote(rentry->namelen);
+		name_end = (char *)rentry + namesize;
+		if (rentry->namelen == 0)
+			return __this_address;
+		if (!(ent->flags & XFS_ATTR_INCOMPLETE) &&
+		    rentry->valueblk == 0)
+			return __this_address;
+	}
+
+	if (name_end > buf_end)
+		return __this_address;
+
+	return NULL;
+}
+
+static xfs_failaddr_t
 xfs_attr3_leaf_verify(
 	struct xfs_buf			*bp)
 {
@@ -240,7 +295,10 @@ xfs_attr3_leaf_verify(
 	struct xfs_mount		*mp = bp->b_mount;
 	struct xfs_attr_leafblock	*leaf = bp->b_addr;
 	struct xfs_attr_leaf_entry	*entries;
+	struct xfs_attr_leaf_entry	*ent;
+	char				*buf_end;
 	uint32_t			end;	/* must be 32bit - see below */
+	__u32				last_hashval = 0;
 	int				i;
 	xfs_failaddr_t			fa;
 
@@ -273,8 +331,13 @@ xfs_attr3_leaf_verify(
 	    (char *)bp->b_addr + ichdr.firstused)
 		return __this_address;
 
-	/* XXX: need to range check rest of attr header values */
-	/* XXX: hash order check? */
+	buf_end = (char *)bp->b_addr + mp->m_attr_geo->blksize;
+	for (i = 0, ent = entries; i < ichdr.count; ent++, i++) {
+		fa = xfs_attr3_leaf_verify_entry(mp, buf_end, leaf, &ichdr,
+				ent, i, &last_hashval);
+		if (fa)
+			return fa;
+	}
 
 	/*
 	 * Quickly check the freemap information.  Attribute data has to be
@@ -367,13 +430,12 @@ xfs_attr3_leaf_read(
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp)
 {
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
-				XFS_ATTR_FORK, &xfs_attr3_leaf_buf_ops);
+	err = xfs_da_read_buf(tp, dp, bno, 0, bpp, XFS_ATTR_FORK,
+			&xfs_attr3_leaf_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_ATTR_LEAF_BUF);
 	return err;
@@ -383,14 +445,25 @@ xfs_attr3_leaf_read(
  * Namespace helper routines
  *========================================================================*/
 
-/*
- * If namespace bits don't match return 0.
- * If all match then return 1.
- */
-STATIC int
-xfs_attr_namesp_match(int arg_flags, int ondisk_flags)
+static bool
+xfs_attr_match(
+	struct xfs_da_args	*args,
+	uint8_t			namelen,
+	unsigned char		*name,
+	int			flags)
 {
-	return XFS_ATTR_NSP_ONDISK(ondisk_flags) == XFS_ATTR_NSP_ARGS_TO_ONDISK(arg_flags);
+	if (args->namelen != namelen)
+		return false;
+	if (memcmp(args->name, name, namelen) != 0)
+		return false;
+	/*
+	 * If we are looking for incomplete entries, show only those, else only
+	 * show complete entries.
+	 */
+	if (args->attr_filter !=
+	    (flags & (XFS_ATTR_NSP_ONDISK_MASK | XFS_ATTR_INCOMPLETE)))
+		return false;
+	return true;
 }
 
 static int
@@ -402,7 +475,7 @@ xfs_attr_copy_value(
 	/*
 	 * No copy if all we have to do is get the length
 	 */
-	if (args->flags & ATTR_KERNOVAL) {
+	if (!args->valuelen) {
 		args->valuelen = valuelen;
 		return 0;
 	}
@@ -415,7 +488,7 @@ xfs_attr_copy_value(
 		return -ERANGE;
 	}
 
-	if (args->op_flags & XFS_DA_OP_ALLOCVAL) {
+	if (!args->value) {
 		args->value = kmem_alloc_large(valuelen, 0);
 		if (!args->value)
 			return -ENOMEM;
@@ -453,16 +526,18 @@ xfs_attr_copy_value(
  * special case for dev/uuid inodes, they have fixed size data forks.
  */
 int
-xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
+xfs_attr_shortform_bytesfit(
+	struct xfs_inode	*dp,
+	int			bytes)
 {
-	int offset;
-	int minforkoff;	/* lower limit on valid forkoff locations */
-	int maxforkoff;	/* upper limit on valid forkoff locations */
-	int dsize;
-	xfs_mount_t *mp = dp->i_mount;
+	struct xfs_mount	*mp = dp->i_mount;
+	int64_t			dsize;
+	int			minforkoff;
+	int			maxforkoff;
+	int			offset;
 
 	/* rounded down */
-	offset = (XFS_LITINO(mp, dp->i_d.di_version) - bytes) >> 3;
+	offset = (XFS_LITINO(mp) - bytes) >> 3;
 
 	if (dp->i_d.di_format == XFS_DINODE_FMT_DEV) {
 		minforkoff = roundup(sizeof(xfs_dev_t), 8) >> 3;
@@ -525,12 +600,11 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
 	 * A data fork btree root must have space for at least
 	 * MINDBTPTRS key/ptr pairs if the data fork is small or empty.
 	 */
-	minforkoff = max(dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS));
+	minforkoff = max_t(int64_t, dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS));
 	minforkoff = roundup(minforkoff, 8) >> 3;
 
 	/* attr fork btree root can have at least this many key/ptr pairs */
-	maxforkoff = XFS_LITINO(mp, dp->i_d.di_version) -
-			XFS_BMDR_SPACE_CALC(MINABTPTRS);
+	maxforkoff = XFS_LITINO(mp) - XFS_BMDR_SPACE_CALC(MINABTPTRS);
 	maxforkoff = maxforkoff >> 3;	/* rounded down */
 
 	if (offset >= maxforkoff)
@@ -614,15 +688,8 @@ xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff)
 	sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data;
 	sfe = &sf->list[0];
 	for (i = 0; i < sf->hdr.count; sfe = XFS_ATTR_SF_NEXTENTRY(sfe), i++) {
-#ifdef DEBUG
-		if (sfe->namelen != args->namelen)
-			continue;
-		if (memcmp(args->name, sfe->nameval, args->namelen) != 0)
-			continue;
-		if (!xfs_attr_namesp_match(args->flags, sfe->flags))
-			continue;
-		ASSERT(0);
-#endif
+		ASSERT(!xfs_attr_match(args, sfe->namelen, sfe->nameval,
+			sfe->flags));
 	}
 
 	offset = (char *)sfe - (char *)sf;
@@ -633,7 +700,7 @@ xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff)
 
 	sfe->namelen = args->namelen;
 	sfe->valuelen = args->valuelen;
-	sfe->flags = XFS_ATTR_NSP_ARGS_TO_ONDISK(args->flags);
+	sfe->flags = args->attr_filter;
 	memcpy(sfe->nameval, args->name, args->namelen);
 	memcpy(&sfe->nameval[args->namelen], args->value, args->valuelen);
 	sf->hdr.count++;
@@ -685,13 +752,9 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 	for (i = 0; i < end; sfe = XFS_ATTR_SF_NEXTENTRY(sfe),
 					base += size, i++) {
 		size = XFS_ATTR_SF_ENTSIZE(sfe);
-		if (sfe->namelen != args->namelen)
-			continue;
-		if (memcmp(sfe->nameval, args->name, args->namelen) != 0)
-			continue;
-		if (!xfs_attr_namesp_match(args->flags, sfe->flags))
-			continue;
-		break;
+		if (xfs_attr_match(args, sfe->namelen, sfe->nameval,
+				sfe->flags))
+			break;
 	}
 	if (i == end)
 		return -ENOATTR;
@@ -752,23 +815,19 @@ xfs_attr_shortform_lookup(xfs_da_args_t *args)
 	sfe = &sf->list[0];
 	for (i = 0; i < sf->hdr.count;
 				sfe = XFS_ATTR_SF_NEXTENTRY(sfe), i++) {
-		if (sfe->namelen != args->namelen)
-			continue;
-		if (memcmp(args->name, sfe->nameval, args->namelen) != 0)
-			continue;
-		if (!xfs_attr_namesp_match(args->flags, sfe->flags))
-			continue;
-		return -EEXIST;
+		if (xfs_attr_match(args, sfe->namelen, sfe->nameval,
+				sfe->flags))
+			return -EEXIST;
 	}
 	return -ENOATTR;
 }
 
 /*
- * Retreive the attribute value and length.
+ * Retrieve the attribute value and length.
  *
- * If ATTR_KERNOVAL is specified, only the length needs to be returned.
- * Unlike a lookup, we only return an error if the attribute does not
- * exist or we can't retrieve the value.
+ * If args->valuelen is zero, only the length needs to be returned.  Unlike a
+ * lookup, we only return an error if the attribute does not exist or we can't
+ * retrieve the value.
  */
 int
 xfs_attr_shortform_getvalue(
@@ -783,14 +842,10 @@ xfs_attr_shortform_getvalue(
 	sfe = &sf->list[0];
 	for (i = 0; i < sf->hdr.count;
 				sfe = XFS_ATTR_SF_NEXTENTRY(sfe), i++) {
-		if (sfe->namelen != args->namelen)
-			continue;
-		if (memcmp(args->name, sfe->nameval, args->namelen) != 0)
-			continue;
-		if (!xfs_attr_namesp_match(args->flags, sfe->flags))
-			continue;
-		return xfs_attr_copy_value(args, &sfe->nameval[args->namelen],
-						sfe->valuelen);
+		if (xfs_attr_match(args, sfe->namelen, sfe->nameval,
+				sfe->flags))
+			return xfs_attr_copy_value(args,
+				&sfe->nameval[args->namelen], sfe->valuelen);
 	}
 	return -ENOATTR;
 }
@@ -854,7 +909,7 @@ xfs_attr_shortform_to_leaf(
 		nargs.valuelen = sfe->valuelen;
 		nargs.hashval = xfs_da_hashname(sfe->nameval,
 						sfe->namelen);
-		nargs.flags = XFS_ATTR_NSP_ONDISK_TO_ARGS(sfe->flags);
+		nargs.attr_filter = sfe->flags & XFS_ATTR_NSP_ONDISK_MASK;
 		error = xfs_attr3_leaf_lookup_int(bp, &nargs); /* set a->index */
 		ASSERT(error == -ENOATTR);
 		error = xfs_attr3_leaf_add(bp, &nargs);
@@ -924,7 +979,7 @@ xfs_attr_shortform_verify(
 	char				*endp;
 	struct xfs_ifork		*ifp;
 	int				i;
-	int				size;
+	int64_t				size;
 
 	ASSERT(ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL);
 	ifp = XFS_IFORK_PTR(ip, XFS_ATTR_FORK);
@@ -1060,7 +1115,7 @@ xfs_attr3_leaf_to_shortform(
 		nargs.value = &name_loc->nameval[nargs.namelen];
 		nargs.valuelen = be16_to_cpu(name_loc->valuelen);
 		nargs.hashval = be32_to_cpu(entry->hashval);
-		nargs.flags = XFS_ATTR_NSP_ONDISK_TO_ARGS(entry->flags);
+		nargs.attr_filter = entry->flags & XFS_ATTR_NSP_ONDISK_MASK;
 		xfs_attr_shortform_add(&nargs, forkoff);
 	}
 	error = 0;
@@ -1080,7 +1135,6 @@ xfs_attr3_leaf_to_node(
 	struct xfs_attr_leafblock *leaf;
 	struct xfs_attr3_icleaf_hdr icleafhdr;
 	struct xfs_attr_leaf_entry *entries;
-	struct xfs_da_node_entry *btree;
 	struct xfs_da3_icnode_hdr icnodehdr;
 	struct xfs_da_intnode	*node;
 	struct xfs_inode	*dp = args->dp;
@@ -1095,11 +1149,11 @@ xfs_attr3_leaf_to_node(
 	error = xfs_da_grow_inode(args, &blkno);
 	if (error)
 		goto out;
-	error = xfs_attr3_leaf_read(args->trans, dp, 0, -1, &bp1);
+	error = xfs_attr3_leaf_read(args->trans, dp, 0, &bp1);
 	if (error)
 		goto out;
 
-	error = xfs_da_get_buf(args->trans, dp, blkno, -1, &bp2, XFS_ATTR_FORK);
+	error = xfs_da_get_buf(args->trans, dp, blkno, &bp2, XFS_ATTR_FORK);
 	if (error)
 		goto out;
 
@@ -1120,18 +1174,17 @@ xfs_attr3_leaf_to_node(
 	if (error)
 		goto out;
 	node = bp1->b_addr;
-	dp->d_ops->node_hdr_from_disk(&icnodehdr, node);
-	btree = dp->d_ops->node_tree_p(node);
+	xfs_da3_node_hdr_from_disk(mp, &icnodehdr, node);
 
 	leaf = bp2->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(args->geo, &icleafhdr, leaf);
 	entries = xfs_attr3_leaf_entryp(leaf);
 
 	/* both on-disk, don't endian-flip twice */
-	btree[0].hashval = entries[icleafhdr.count - 1].hashval;
-	btree[0].before = cpu_to_be32(blkno);
+	icnodehdr.btree[0].hashval = entries[icleafhdr.count - 1].hashval;
+	icnodehdr.btree[0].before = cpu_to_be32(blkno);
 	icnodehdr.count = 1;
-	dp->d_ops->node_hdr_to_disk(node, &icnodehdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &icnodehdr);
 	xfs_trans_log_buf(args->trans, bp1, 0, args->geo->blksize - 1);
 	error = 0;
 out:
@@ -1161,7 +1214,7 @@ xfs_attr3_leaf_create(
 
 	trace_xfs_attr_leaf_create(args);
 
-	error = xfs_da_get_buf(args->trans, args->dp, blkno, -1, &bp,
+	error = xfs_da_get_buf(args->trans, args->dp, blkno, &bp,
 					    XFS_ATTR_FORK);
 	if (error)
 		return error;
@@ -1387,8 +1440,9 @@ xfs_attr3_leaf_add_work(
 	entry->nameidx = cpu_to_be16(ichdr->freemap[mapindex].base +
 				     ichdr->freemap[mapindex].size);
 	entry->hashval = cpu_to_be32(args->hashval);
-	entry->flags = tmp ? XFS_ATTR_LOCAL : 0;
-	entry->flags |= XFS_ATTR_NSP_ARGS_TO_ONDISK(args->flags);
+	entry->flags = args->attr_filter;
+	if (tmp)
+		entry->flags |= XFS_ATTR_LOCAL;
 	if (args->op_flags & XFS_DA_OP_RENAME) {
 		entry->flags |= XFS_ATTR_INCOMPLETE;
 		if ((args->blkno2 == args->blkno) &&
@@ -1447,7 +1501,9 @@ xfs_attr3_leaf_add_work(
 	for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
 		if (ichdr->freemap[i].base == tmp) {
 			ichdr->freemap[i].base += sizeof(xfs_attr_leaf_entry_t);
-			ichdr->freemap[i].size -= sizeof(xfs_attr_leaf_entry_t);
+			ichdr->freemap[i].size -=
+				min_t(uint16_t, ichdr->freemap[i].size,
+						sizeof(xfs_attr_leaf_entry_t));
 		}
 	}
 	ichdr->usedbytes += xfs_attr_leaf_entsize(leaf, args->index);
@@ -1931,7 +1987,7 @@ xfs_attr3_leaf_toosmall(
 		if (blkno == 0)
 			continue;
 		error = xfs_attr3_leaf_read(state->args->trans, state->args->dp,
-					blkno, -1, &bp);
+					blkno, &bp);
 		if (error)
 			return error;
 
@@ -2281,8 +2337,10 @@ xfs_attr3_leaf_lookup_int(
 	leaf = bp->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(args->geo, &ichdr, leaf);
 	entries = xfs_attr3_leaf_entryp(leaf);
-	if (ichdr.count >= args->geo->blksize / 8)
+	if (ichdr.count >= args->geo->blksize / 8) {
+		xfs_buf_mark_corrupt(bp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Binary search.  (note: small blocks will skip this loop)
@@ -2298,10 +2356,14 @@ xfs_attr3_leaf_lookup_int(
 		else
 			break;
 	}
-	if (!(probe >= 0 && (!ichdr.count || probe < ichdr.count)))
+	if (!(probe >= 0 && (!ichdr.count || probe < ichdr.count))) {
+		xfs_buf_mark_corrupt(bp);
 		return -EFSCORRUPTED;
-	if (!(span <= 4 || be32_to_cpu(entry->hashval) == hashval))
+	}
+	if (!(span <= 4 || be32_to_cpu(entry->hashval) == hashval)) {
+		xfs_buf_mark_corrupt(bp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Since we may have duplicate hashval's, find the first matching
@@ -2329,33 +2391,17 @@ xfs_attr3_leaf_lookup_int(
 /*
  * GROT: Add code to remove incomplete entries.
  */
-		/*
-		 * If we are looking for INCOMPLETE entries, show only those.
-		 * If we are looking for complete entries, show only those.
-		 */
-		if ((args->flags & XFS_ATTR_INCOMPLETE) !=
-		    (entry->flags & XFS_ATTR_INCOMPLETE)) {
-			continue;
-		}
 		if (entry->flags & XFS_ATTR_LOCAL) {
 			name_loc = xfs_attr3_leaf_name_local(leaf, probe);
-			if (name_loc->namelen != args->namelen)
-				continue;
-			if (memcmp(args->name, name_loc->nameval,
-							args->namelen) != 0)
-				continue;
-			if (!xfs_attr_namesp_match(args->flags, entry->flags))
+			if (!xfs_attr_match(args, name_loc->namelen,
+					name_loc->nameval, entry->flags))
 				continue;
 			args->index = probe;
 			return -EEXIST;
 		} else {
 			name_rmt = xfs_attr3_leaf_name_remote(leaf, probe);
-			if (name_rmt->namelen != args->namelen)
-				continue;
-			if (memcmp(args->name, name_rmt->name,
-							args->namelen) != 0)
-				continue;
-			if (!xfs_attr_namesp_match(args->flags, entry->flags))
+			if (!xfs_attr_match(args, name_rmt->namelen,
+					name_rmt->name, entry->flags))
 				continue;
 			args->index = probe;
 			args->rmtvaluelen = be32_to_cpu(name_rmt->valuelen);
@@ -2374,9 +2420,9 @@ xfs_attr3_leaf_lookup_int(
  * Get the value associated with an attribute name from a leaf attribute
  * list structure.
  *
- * If ATTR_KERNOVAL is specified, only the length needs to be returned.
- * Unlike a lookup, we only return an error if the attribute does not
- * exist or we can't retrieve the value.
+ * If args->valuelen is zero, only the length needs to be returned.  Unlike a
+ * lookup, we only return an error if the attribute does not exist or we can't
+ * retrieve the value.
  */
 int
 xfs_attr3_leaf_getvalue(
@@ -2661,7 +2707,7 @@ xfs_attr3_leaf_clearflag(
 	/*
 	 * Set up the operation.
 	 */
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
@@ -2728,7 +2774,7 @@ xfs_attr3_leaf_setflag(
 	/*
 	 * Set up the operation.
 	 */
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp);
 	if (error)
 		return error;
 
@@ -2790,7 +2836,7 @@ xfs_attr3_leaf_flipflags(
 	/*
 	 * Read the block containing the "old" attr
 	 */
-	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp1);
+	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, &bp1);
 	if (error)
 		return error;
 
@@ -2799,7 +2845,7 @@ xfs_attr3_leaf_flipflags(
 	 */
 	if (args->blkno2 != args->blkno) {
 		error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno2,
-					   -1, &bp2);
+					   &bp2);
 		if (error)
 			return error;
 	} else {
diff --git a/fs/xfs/libxfs/xfs_attr_leaf.h b/fs/xfs/libxfs/xfs_attr_leaf.h
index 7b74e18becff..6dd2d937a42a 100644
--- a/fs/xfs/libxfs/xfs_attr_leaf.h
+++ b/fs/xfs/libxfs/xfs_attr_leaf.h
@@ -8,7 +8,6 @@
 #define	__XFS_ATTR_LEAF_H__
 
 struct attrlist;
-struct attrlist_cursor_kern;
 struct xfs_attr_list_context;
 struct xfs_da_args;
 struct xfs_da_state;
@@ -17,13 +16,27 @@ struct xfs_inode;
 struct xfs_trans;
 
 /*
- * Used to keep a list of "remote value" extents when unlinking an inode.
+ * Incore version of the attribute leaf header.
  */
-typedef struct xfs_attr_inactive_list {
-	xfs_dablk_t	valueblk;	/* block number of value bytes */
-	int		valuelen;	/* number of bytes in value */
-} xfs_attr_inactive_list_t;
-
+struct xfs_attr3_icleaf_hdr {
+	uint32_t	forw;
+	uint32_t	back;
+	uint16_t	magic;
+	uint16_t	count;
+	uint16_t	usedbytes;
+	/*
+	 * Firstused is 32-bit here instead of 16-bit like the on-disk variant
+	 * to support maximum fsb size of 64k without overflow issues throughout
+	 * the attr code. Instead, the overflow condition is handled on
+	 * conversion to/from disk.
+	 */
+	uint32_t	firstused;
+	__u8		holes;
+	struct {
+		uint16_t	base;
+		uint16_t	size;
+	} freemap[XFS_ATTR_LEAF_MAPSIZE];
+};
 
 /*========================================================================
  * Function prototypes for the kernel.
@@ -67,8 +80,8 @@ int	xfs_attr3_leaf_add(struct xfs_buf *leaf_buffer,
 				 struct xfs_da_args *args);
 int	xfs_attr3_leaf_remove(struct xfs_buf *leaf_buffer,
 				    struct xfs_da_args *args);
-void	xfs_attr3_leaf_list_int(struct xfs_buf *bp,
-				      struct xfs_attr_list_context *context);
+int	xfs_attr3_leaf_list_int(struct xfs_buf *bp,
+				struct xfs_attr_list_context *context);
 
 /*
  * Routines used for shrinking the Btree.
@@ -85,8 +98,7 @@ int	xfs_attr_leaf_order(struct xfs_buf *leaf1_bp,
 				   struct xfs_buf *leaf2_bp);
 int	xfs_attr_leaf_newentsize(struct xfs_da_args *args, int *local);
 int	xfs_attr3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
-			xfs_dablk_t bno, xfs_daddr_t mappedbno,
-			struct xfs_buf **bpp);
+			xfs_dablk_t bno, struct xfs_buf **bpp);
 void	xfs_attr3_leaf_hdr_from_disk(struct xfs_da_geometry *geo,
 				     struct xfs_attr3_icleaf_hdr *to,
 				     struct xfs_attr_leafblock *from);
diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c
index 3e39b7d40f25..01ad7f353e08 100644
--- a/fs/xfs/libxfs/xfs_attr_remote.c
+++ b/fs/xfs/libxfs/xfs_attr_remote.c
@@ -19,12 +19,30 @@
 #include "xfs_trans.h"
 #include "xfs_bmap.h"
 #include "xfs_attr.h"
+#include "xfs_attr_remote.h"
 #include "xfs_trace.h"
 #include "xfs_error.h"
 
 #define ATTR_RMTVALUE_MAPSIZE	1	/* # of map entries at once */
 
 /*
+ * Remote Attribute Values
+ * =======================
+ *
+ * Remote extended attribute values are conceptually simple -- they're written
+ * to data blocks mapped by an inode's attribute fork, and they have an upper
+ * size limit of 64k.  Setting a value does not involve the XFS log.
+ *
+ * However, on a v5 filesystem, maximally sized remote attr values require one
+ * block more than 64k worth of space to hold both the remote attribute value
+ * header (64 bytes).  On a 4k block filesystem this results in a 68k buffer;
+ * on a 64k block filesystem, this would be a 128k buffer.  Note that the log
+ * format can only handle a dirty buffer of XFS_MAX_BLOCKSIZE length (64k).
+ * Therefore, we /must/ ensure that remote attribute value buffers never touch
+ * the logging system and therefore never have a log item.
+ */
+
+/*
  * Each contiguous block has a header, so it is not just a simple attribute
  * length to FSB conversion.
  */
@@ -379,7 +397,7 @@ xfs_attr_rmtval_get(
 
 	trace_xfs_attr_rmtval_get(args);
 
-	ASSERT(!(args->flags & ATTR_KERNOVAL));
+	ASSERT(args->valuelen != 0);
 	ASSERT(args->rmtvaluelen == args->valuelen);
 
 	valuelen = args->rmtvaluelen;
@@ -400,17 +418,15 @@ xfs_attr_rmtval_get(
 			       (map[i].br_startblock != HOLESTARTBLOCK));
 			dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
 			dblkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
-			error = xfs_trans_read_buf(mp, args->trans,
-						   mp->m_ddev_targp,
-						   dblkno, dblkcnt, 0, &bp,
-						   &xfs_attr3_rmt_buf_ops);
+			error = xfs_buf_read(mp->m_ddev_targp, dblkno, dblkcnt,
+					0, &bp, &xfs_attr3_rmt_buf_ops);
 			if (error)
 				return error;
 
 			error = xfs_attr_rmtval_copyout(mp, bp, args->dp->i_ino,
 							&offset, &valuelen,
 							&dst);
-			xfs_trans_brelse(args->trans, bp);
+			xfs_buf_relse(bp);
 			if (error)
 				return error;
 
@@ -529,9 +545,9 @@ xfs_attr_rmtval_set(
 		dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
 		dblkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
 
-		bp = xfs_buf_get(mp->m_ddev_targp, dblkno, dblkcnt);
-		if (!bp)
-			return -ENOMEM;
+		error = xfs_buf_get(mp->m_ddev_targp, dblkno, dblkcnt, &bp);
+		if (error)
+			return error;
 		bp->b_ops = &xfs_attr3_rmt_buf_ops;
 
 		xfs_attr_rmtval_copyin(mp, bp, args->dp->i_ino, &offset,
@@ -551,6 +567,33 @@ xfs_attr_rmtval_set(
 	return 0;
 }
 
+/* Mark stale any incore buffers for the remote value. */
+int
+xfs_attr_rmtval_stale(
+	struct xfs_inode	*ip,
+	struct xfs_bmbt_irec	*map,
+	xfs_buf_flags_t		incore_flags)
+{
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_buf		*bp;
+
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
+
+	if (XFS_IS_CORRUPT(mp, map->br_startblock == DELAYSTARTBLOCK) ||
+	    XFS_IS_CORRUPT(mp, map->br_startblock == HOLESTARTBLOCK))
+		return -EFSCORRUPTED;
+
+	bp = xfs_buf_incore(mp->m_ddev_targp,
+			XFS_FSB_TO_DADDR(mp, map->br_startblock),
+			XFS_FSB_TO_BB(mp, map->br_blockcount), incore_flags);
+	if (bp) {
+		xfs_buf_stale(bp);
+		xfs_buf_relse(bp);
+	}
+
+	return 0;
+}
+
 /*
  * Remove the value associated with an attribute by deleting the
  * out-of-line buffer that it is stored on.
@@ -559,7 +602,6 @@ int
 xfs_attr_rmtval_remove(
 	struct xfs_da_args	*args)
 {
-	struct xfs_mount	*mp = args->dp->i_mount;
 	xfs_dablk_t		lblkno;
 	int			blkcnt;
 	int			error;
@@ -574,9 +616,6 @@ xfs_attr_rmtval_remove(
 	blkcnt = args->rmtblkcnt;
 	while (blkcnt > 0) {
 		struct xfs_bmbt_irec	map;
-		struct xfs_buf		*bp;
-		xfs_daddr_t		dblkno;
-		int			dblkcnt;
 		int			nmap;
 
 		/*
@@ -587,22 +626,11 @@ xfs_attr_rmtval_remove(
 				       blkcnt, &map, &nmap, XFS_BMAPI_ATTRFORK);
 		if (error)
 			return error;
-		ASSERT(nmap == 1);
-		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
-		       (map.br_startblock != HOLESTARTBLOCK));
-
-		dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
-		dblkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
-
-		/*
-		 * If the "remote" value is in the cache, remove it.
-		 */
-		bp = xfs_buf_incore(mp->m_ddev_targp, dblkno, dblkcnt, XBF_TRYLOCK);
-		if (bp) {
-			xfs_buf_stale(bp);
-			xfs_buf_relse(bp);
-			bp = NULL;
-		}
+		if (XFS_IS_CORRUPT(args->dp->i_mount, nmap != 1))
+			return -EFSCORRUPTED;
+		error = xfs_attr_rmtval_stale(args->dp, &map, XBF_TRYLOCK);
+		if (error)
+			return error;
 
 		lblkno += map.br_blockcount;
 		blkcnt -= map.br_blockcount;
diff --git a/fs/xfs/libxfs/xfs_attr_remote.h b/fs/xfs/libxfs/xfs_attr_remote.h
index 9d20b66ad379..6fb4572845ce 100644
--- a/fs/xfs/libxfs/xfs_attr_remote.h
+++ b/fs/xfs/libxfs/xfs_attr_remote.h
@@ -11,5 +11,7 @@ int xfs_attr3_rmt_blocks(struct xfs_mount *mp, int attrlen);
 int xfs_attr_rmtval_get(struct xfs_da_args *args);
 int xfs_attr_rmtval_set(struct xfs_da_args *args);
 int xfs_attr_rmtval_remove(struct xfs_da_args *args);
+int xfs_attr_rmtval_stale(struct xfs_inode *ip, struct xfs_bmbt_irec *map,
+		xfs_buf_flags_t incore_flags);
 
 #endif /* __XFS_ATTR_REMOTE_H__ */
diff --git a/fs/xfs/libxfs/xfs_bit.c b/fs/xfs/libxfs/xfs_bit.c
index 7071ff98fdbc..40ce5f3094d1 100644
--- a/fs/xfs/libxfs/xfs_bit.c
+++ b/fs/xfs/libxfs/xfs_bit.c
@@ -5,6 +5,7 @@
  */
 #include "xfs.h"
 #include "xfs_log_format.h"
+#include "xfs_bit.h"
 
 /*
  * XFS bit manipulation routines, used in non-realtime code.
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 02469d59c787..fda13cd7add0 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -34,6 +34,7 @@
 #include "xfs_ag_resv.h"
 #include "xfs_refcount.h"
 #include "xfs_icache.h"
+#include "xfs_iomap.h"
 
 
 kmem_zone_t		*xfs_bmap_free_item_zone;
@@ -192,14 +193,12 @@ xfs_default_attroffset(
 	struct xfs_mount	*mp = ip->i_mount;
 	uint			offset;
 
-	if (mp->m_sb.sb_inodesize == 256) {
-		offset = XFS_LITINO(mp, ip->i_d.di_version) -
-				XFS_BMDR_SPACE_CALC(MINABTPTRS);
-	} else {
+	if (mp->m_sb.sb_inodesize == 256)
+		offset = XFS_LITINO(mp) - XFS_BMDR_SPACE_CALC(MINABTPTRS);
+	else
 		offset = XFS_BMDR_SPACE_CALC(6 * MINABTPTRS);
-	}
 
-	ASSERT(offset < XFS_LITINO(mp, ip->i_d.di_version));
+	ASSERT(offset < XFS_LITINO(mp));
 	return offset;
 }
 
@@ -383,8 +382,10 @@ xfs_bmap_check_leaf_extents(
 		xfs_check_block(block, mp, 0, 0);
 		pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
 		bno = be64_to_cpu(*pp);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-					xfs_verify_fsbno(mp, bno), error0);
+		if (XFS_IS_CORRUPT(mp, !xfs_verify_fsbno(mp, bno))) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if (bp_release) {
 			bp_release = 0;
 			xfs_trans_brelse(NULL, bp);
@@ -611,8 +612,8 @@ xfs_bmap_btree_to_extents(
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, rblock, 1, ifp->if_broot_bytes);
 	cbno = be64_to_cpu(*pp);
 #ifdef DEBUG
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp,
-			xfs_btree_check_lptr(cur, cbno, 1));
+	if (XFS_IS_CORRUPT(cur->bc_mp, !xfs_btree_check_lptr(cur, cbno, 1)))
+		return -EFSCORRUPTED;
 #endif
 	error = xfs_btree_read_bufl(mp, tp, cbno, &cbp, XFS_BMAP_BTREE_REF,
 				&xfs_bmbt_buf_ops);
@@ -687,7 +688,7 @@ xfs_bmap_extents_to_btree(
 	 * Need a cursor.  Can't allocate until bb_level is filled in.
 	 */
 	cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
-	cur->bc_private.b.flags = wasdel ? XFS_BTCUR_BPRV_WASDEL : 0;
+	cur->bc_ino.flags = wasdel ? XFS_BTCUR_BMBT_WASDEL : 0;
 	/*
 	 * Convert to a btree with two levels, one record in root.
 	 */
@@ -724,14 +725,14 @@ xfs_bmap_extents_to_btree(
 	ASSERT(tp->t_firstblock == NULLFSBLOCK ||
 	       args.agno >= XFS_FSB_TO_AGNO(mp, tp->t_firstblock));
 	tp->t_firstblock = args.fsbno;
-	cur->bc_private.b.allocated++;
+	cur->bc_ino.allocated++;
 	ip->i_d.di_nblocks++;
 	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, 1L);
-	abp = xfs_btree_get_bufl(mp, tp, args.fsbno);
-	if (!abp) {
-		error = -EFSCORRUPTED;
+	error = xfs_trans_get_buf(tp, mp->m_ddev_targp,
+			XFS_FSB_TO_DADDR(mp, args.fsbno),
+			mp->m_bsize, 0, &abp);
+	if (error)
 		goto out_unreserve_dquot;
-	}
 
 	/*
 	 * Fill in the child block.
@@ -875,7 +876,11 @@ xfs_bmap_local_to_extents(
 	ASSERT(args.fsbno != NULLFSBLOCK);
 	ASSERT(args.len == 1);
 	tp->t_firstblock = args.fsbno;
-	bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno);
+	error = xfs_trans_get_buf(tp, args.mp->m_ddev_targp,
+			XFS_FSB_TO_DADDR(args.mp, args.fsbno),
+			args.mp->m_bsize, 0, &bp);
+	if (error)
+		goto done;
 
 	/*
 	 * Initialize the block, copy the data and log the remote buffer.
@@ -936,14 +941,17 @@ xfs_bmap_add_attrfork_btree(
 		if (error)
 			goto error0;
 		/* must be at least one entry */
-		XFS_WANT_CORRUPTED_GOTO(mp, stat == 1, error0);
+		if (XFS_IS_CORRUPT(mp, stat != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if ((error = xfs_btree_new_iroot(cur, flags, &stat)))
 			goto error0;
 		if (stat == 0) {
 			xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
 			return -ENOSPC;
 		}
-		cur->bc_private.b.allocated = 0;
+		cur->bc_ino.allocated = 0;
 		xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
 	}
 	return 0;
@@ -970,7 +978,7 @@ xfs_bmap_add_attrfork_extents(
 	error = xfs_bmap_extents_to_btree(tp, ip, &cur, 0, flags,
 					  XFS_DATA_FORK);
 	if (cur) {
-		cur->bc_private.b.allocated = 0;
+		cur->bc_ino.allocated = 0;
 		xfs_btree_del_cursor(cur, error);
 	}
 	return error;
@@ -1083,7 +1091,7 @@ xfs_bmap_add_attrfork(
 		goto trans_cancel;
 	if (XFS_IFORK_Q(ip))
 		goto trans_cancel;
-	if (ip->i_d.di_anextents != 0) {
+	if (XFS_IS_CORRUPT(mp, ip->i_d.di_anextents != 0)) {
 		error = -EFSCORRUPTED;
 		goto trans_cancel;
 	}
@@ -1154,6 +1162,65 @@ trans_cancel:
  * Internal and external extent tree search functions.
  */
 
+struct xfs_iread_state {
+	struct xfs_iext_cursor	icur;
+	xfs_extnum_t		loaded;
+};
+
+/* Stuff every bmbt record from this block into the incore extent map. */
+static int
+xfs_iread_bmbt_block(
+	struct xfs_btree_cur	*cur,
+	int			level,
+	void			*priv)
+{
+	struct xfs_iread_state	*ir = priv;
+	struct xfs_mount	*mp = cur->bc_mp;
+	struct xfs_inode	*ip = cur->bc_ino.ip;
+	struct xfs_btree_block	*block;
+	struct xfs_buf		*bp;
+	struct xfs_bmbt_rec	*frp;
+	xfs_extnum_t		num_recs;
+	xfs_extnum_t		j;
+	int			whichfork = cur->bc_ino.whichfork;
+
+	block = xfs_btree_get_block(cur, level, &bp);
+
+	/* Abort if we find more records than nextents. */
+	num_recs = xfs_btree_get_numrecs(block);
+	if (unlikely(ir->loaded + num_recs >
+		     XFS_IFORK_NEXTENTS(ip, whichfork))) {
+		xfs_warn(ip->i_mount, "corrupt dinode %llu, (btree extents).",
+				(unsigned long long)ip->i_ino);
+		xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__, block,
+				sizeof(*block), __this_address);
+		return -EFSCORRUPTED;
+	}
+
+	/* Copy records into the incore cache. */
+	frp = XFS_BMBT_REC_ADDR(mp, block, 1);
+	for (j = 0; j < num_recs; j++, frp++, ir->loaded++) {
+		struct xfs_bmbt_irec	new;
+		xfs_failaddr_t		fa;
+
+		xfs_bmbt_disk_get_all(frp, &new);
+		fa = xfs_bmap_validate_extent(ip, whichfork, &new);
+		if (fa) {
+			xfs_inode_verifier_error(ip, -EFSCORRUPTED,
+					"xfs_iread_extents(2)", frp,
+					sizeof(*frp), fa);
+			return -EFSCORRUPTED;
+		}
+		xfs_iext_insert(ip, &ir->icur, &new,
+				xfs_bmap_fork_to_state(whichfork));
+		trace_xfs_read_extent(ip, &ir->icur,
+				xfs_bmap_fork_to_state(whichfork), _THIS_IP_);
+		xfs_iext_next(XFS_IFORK_PTR(ip, whichfork), &ir->icur);
+	}
+
+	return 0;
+}
+
 /*
  * Read in extents from a btree-format inode.
  */
@@ -1163,134 +1230,39 @@ xfs_iread_extents(
 	struct xfs_inode	*ip,
 	int			whichfork)
 {
-	struct xfs_mount	*mp = ip->i_mount;
-	int			state = xfs_bmap_fork_to_state(whichfork);
+	struct xfs_iread_state	ir;
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
-	xfs_extnum_t		nextents = XFS_IFORK_NEXTENTS(ip, whichfork);
-	struct xfs_btree_block	*block = ifp->if_broot;
-	struct xfs_iext_cursor	icur;
-	struct xfs_bmbt_irec	new;
-	xfs_fsblock_t		bno;
-	struct xfs_buf		*bp;
-	xfs_extnum_t		i, j;
-	int			level;
-	__be64			*pp;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_btree_cur	*cur;
 	int			error;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
-	if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		return -EFSCORRUPTED;
-	}
-
-	/*
-	 * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
-	 */
-	level = be16_to_cpu(block->bb_level);
-	if (unlikely(level == 0)) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		return -EFSCORRUPTED;
-	}
-	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
-	bno = be64_to_cpu(*pp);
-
-	/*
-	 * Go down the tree until leaf level is reached, following the first
-	 * pointer (leftmost) at each level.
-	 */
-	while (level-- > 0) {
-		error = xfs_btree_read_bufl(mp, tp, bno, &bp,
-				XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
-		if (error)
-			goto out;
-		block = XFS_BUF_TO_BLOCK(bp);
-		if (level == 0)
-			break;
-		pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
-		bno = be64_to_cpu(*pp);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-			xfs_verify_fsbno(mp, bno), out_brelse);
-		xfs_trans_brelse(tp, bp);
+	if (XFS_IS_CORRUPT(mp,
+			   XFS_IFORK_FORMAT(ip, whichfork) !=
+			   XFS_DINODE_FMT_BTREE)) {
+		error = -EFSCORRUPTED;
+		goto out;
 	}
 
-	/*
-	 * Here with bp and block set to the leftmost leaf node in the tree.
-	 */
-	i = 0;
-	xfs_iext_first(ifp, &icur);
-
-	/*
-	 * Loop over all leaf nodes.  Copy information to the extent records.
-	 */
-	for (;;) {
-		xfs_bmbt_rec_t	*frp;
-		xfs_fsblock_t	nextbno;
-		xfs_extnum_t	num_recs;
-
-		num_recs = xfs_btree_get_numrecs(block);
-		if (unlikely(i + num_recs > nextents)) {
-			xfs_warn(ip->i_mount,
-				"corrupt dinode %Lu, (btree extents).",
-				(unsigned long long) ip->i_ino);
-			xfs_inode_verifier_error(ip, -EFSCORRUPTED,
-					__func__, block, sizeof(*block),
-					__this_address);
-			error = -EFSCORRUPTED;
-			goto out_brelse;
-		}
-		/*
-		 * Read-ahead the next leaf block, if any.
-		 */
-		nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
-		if (nextbno != NULLFSBLOCK)
-			xfs_btree_reada_bufl(mp, nextbno, 1,
-					     &xfs_bmbt_buf_ops);
-		/*
-		 * Copy records into the extent records.
-		 */
-		frp = XFS_BMBT_REC_ADDR(mp, block, 1);
-		for (j = 0; j < num_recs; j++, frp++, i++) {
-			xfs_failaddr_t	fa;
-
-			xfs_bmbt_disk_get_all(frp, &new);
-			fa = xfs_bmap_validate_extent(ip, whichfork, &new);
-			if (fa) {
-				error = -EFSCORRUPTED;
-				xfs_inode_verifier_error(ip, error,
-						"xfs_iread_extents(2)",
-						frp, sizeof(*frp), fa);
-				goto out_brelse;
-			}
-			xfs_iext_insert(ip, &icur, &new, state);
-			trace_xfs_read_extent(ip, &icur, state, _THIS_IP_);
-			xfs_iext_next(ifp, &icur);
-		}
-		xfs_trans_brelse(tp, bp);
-		bno = nextbno;
-		/*
-		 * If we've reached the end, stop.
-		 */
-		if (bno == NULLFSBLOCK)
-			break;
-		error = xfs_btree_read_bufl(mp, tp, bno, &bp,
-				XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
-		if (error)
-			goto out;
-		block = XFS_BUF_TO_BLOCK(bp);
-	}
+	ir.loaded = 0;
+	xfs_iext_first(ifp, &ir.icur);
+	cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
+	error = xfs_btree_visit_blocks(cur, xfs_iread_bmbt_block,
+			XFS_BTREE_VISIT_RECORDS, &ir);
+	xfs_btree_del_cursor(cur, error);
+	if (error)
+		goto out;
 
-	if (i != XFS_IFORK_NEXTENTS(ip, whichfork)) {
+	if (XFS_IS_CORRUPT(mp,
+			   ir.loaded != XFS_IFORK_NEXTENTS(ip, whichfork))) {
 		error = -EFSCORRUPTED;
 		goto out;
 	}
-	ASSERT(i == xfs_iext_count(ifp));
+	ASSERT(ir.loaded == xfs_iext_count(ifp));
 
 	ifp->if_flags |= XFS_IFEXTENTS;
 	return 0;
-
-out_brelse:
-	xfs_trans_brelse(tp, bp);
 out:
 	xfs_iext_destroy(ifp);
 	return error;
@@ -1317,8 +1289,7 @@ xfs_bmap_first_unused(
 	xfs_fileoff_t		lowest, max;
 	int			error;
 
-	ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE ||
-	       XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ||
+	ASSERT(xfs_ifork_has_extents(ip, whichfork) ||
 	       XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL);
 
 	if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
@@ -1374,7 +1345,8 @@ xfs_bmap_last_before(
 	case XFS_DINODE_FMT_EXTENTS:
 		break;
 	default:
-		return -EIO;
+		ASSERT(0);
+		return -EFSCORRUPTED;
 	}
 
 	if (!(ifp->if_flags & XFS_IFEXTENTS)) {
@@ -1473,9 +1445,8 @@ xfs_bmap_last_offset(
 	if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL)
 		return 0;
 
-	if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
-	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
-	       return -EIO;
+	if (XFS_IS_CORRUPT(ip->i_mount, !xfs_ifork_has_extents(ip, whichfork)))
+		return -EFSCORRUPTED;
 
 	error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, &is_empty);
 	if (error || is_empty)
@@ -1555,7 +1526,7 @@ xfs_bmap_add_extent_delay_real(
 
 	ASSERT(!isnullstartblock(new->br_startblock));
 	ASSERT(!bma->cur ||
-	       (bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
+	       (bma->cur->bc_ino.flags & XFS_BTCUR_BMBT_WASDEL));
 
 	XFS_STATS_INC(mp, xs_add_exlist);
 
@@ -1652,15 +1623,24 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, &RIGHT, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_delete(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_decrement(bma->cur, 0, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &LEFT);
 			if (error)
 				goto done;
@@ -1686,7 +1666,10 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &LEFT);
 			if (error)
 				goto done;
@@ -1716,7 +1699,10 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, &RIGHT, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &PREV);
 			if (error)
 				goto done;
@@ -1741,11 +1727,17 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 
@@ -1776,7 +1768,10 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &LEFT);
 			if (error)
 				goto done;
@@ -1797,11 +1792,17 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 
 		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
@@ -1815,7 +1816,7 @@ xfs_bmap_add_extent_delay_real(
 		temp = PREV.br_blockcount - new->br_blockcount;
 		da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
 			startblockval(PREV.br_startblock) -
-			(bma->cur ? bma->cur->bc_private.b.allocated : 0));
+			(bma->cur ? bma->cur->bc_ino.allocated : 0));
 
 		PREV.br_startoff = new_endoff;
 		PREV.br_blockcount = temp;
@@ -1842,7 +1843,10 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(bma->cur, &RIGHT);
 			if (error)
 				goto done;
@@ -1874,11 +1878,17 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 
 		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
@@ -1892,7 +1902,7 @@ xfs_bmap_add_extent_delay_real(
 		temp = PREV.br_blockcount - new->br_blockcount;
 		da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
 			startblockval(PREV.br_startblock) -
-			(bma->cur ? bma->cur->bc_private.b.allocated : 0));
+			(bma->cur ? bma->cur->bc_ino.allocated : 0));
 
 		PREV.br_startblock = nullstartblock(da_new);
 		PREV.br_blockcount = temp;
@@ -1954,11 +1964,17 @@ xfs_bmap_add_extent_delay_real(
 			error = xfs_bmbt_lookup_eq(bma->cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(bma->cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 
 		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
@@ -2007,8 +2023,8 @@ xfs_bmap_add_extent_delay_real(
 		xfs_mod_delalloc(mp, (int64_t)da_new - da_old);
 
 	if (bma->cur) {
-		da_new += bma->cur->bc_private.b.allocated;
-		bma->cur->bc_private.b.allocated = 0;
+		da_new += bma->cur->bc_ino.allocated;
+		bma->cur->bc_ino.allocated = 0;
 	}
 
 	/* adjust for changes in reserved delayed indirect blocks */
@@ -2152,19 +2168,34 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, &RIGHT, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_delete(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_decrement(cur, 0, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_delete(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_decrement(cur, 0, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &LEFT);
 			if (error)
 				goto done;
@@ -2190,13 +2221,22 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, &PREV, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_delete(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_decrement(cur, 0, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &LEFT);
 			if (error)
 				goto done;
@@ -2225,13 +2265,22 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, &RIGHT, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_delete(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_decrement(cur, 0, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
@@ -2254,7 +2303,10 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
@@ -2284,7 +2336,10 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
@@ -2318,14 +2373,20 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
 			cur->bc_rec.b = *new;
 			if ((error = xfs_btree_insert(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 
@@ -2352,7 +2413,10 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
@@ -2386,17 +2450,26 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &PREV);
 			if (error)
 				goto done;
 			error = xfs_bmbt_lookup_eq(cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			if ((error = xfs_btree_insert(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 
@@ -2430,7 +2503,10 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			/* new right extent - oldext */
 			error = xfs_bmbt_update(cur, &r[1]);
 			if (error)
@@ -2439,7 +2515,10 @@ xfs_bmap_add_extent_unwritten_real(
 			cur->bc_rec.b = PREV;
 			if ((error = xfs_btree_insert(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			/*
 			 * Reset the cursor to the position of the new extent
 			 * we are about to insert as we can't trust it after
@@ -2448,11 +2527,17 @@ xfs_bmap_add_extent_unwritten_real(
 			error = xfs_bmbt_lookup_eq(cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			/* new middle extent - newext */
 			if ((error = xfs_btree_insert(cur, &i)))
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 
@@ -2486,7 +2571,7 @@ xfs_bmap_add_extent_unwritten_real(
 
 	/* clear out the allocated field, done with it now in any case. */
 	if (cur) {
-		cur->bc_private.b.allocated = 0;
+		cur->bc_ino.allocated = 0;
 		*curp = cur;
 	}
 
@@ -2665,7 +2750,7 @@ xfs_bmap_add_extent_hole_real(
 	struct xfs_bmbt_irec	old;
 
 	ASSERT(!isnullstartblock(new->br_startblock));
-	ASSERT(!cur || !(cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
+	ASSERT(!cur || !(cur->bc_ino.flags & XFS_BTCUR_BMBT_WASDEL));
 
 	XFS_STATS_INC(mp, xs_add_exlist);
 
@@ -2735,15 +2820,24 @@ xfs_bmap_add_extent_hole_real(
 			error = xfs_bmbt_lookup_eq(cur, &right, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_delete(cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_decrement(cur, 0, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &left);
 			if (error)
 				goto done;
@@ -2769,7 +2863,10 @@ xfs_bmap_add_extent_hole_real(
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &left);
 			if (error)
 				goto done;
@@ -2796,7 +2893,10 @@ xfs_bmap_add_extent_hole_real(
 			error = xfs_bmbt_lookup_eq(cur, &old, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_bmbt_update(cur, &right);
 			if (error)
 				goto done;
@@ -2819,11 +2919,17 @@ xfs_bmap_add_extent_hole_real(
 			error = xfs_bmbt_lookup_eq(cur, new, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+			if (XFS_IS_CORRUPT(mp, i != 0)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 			error = xfs_btree_insert(cur, &i);
 			if (error)
 				goto done;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		}
 		break;
 	}
@@ -2847,7 +2953,7 @@ xfs_bmap_add_extent_hole_real(
 
 	/* clear out the allocated field, done with it now in any case. */
 	if (cur)
-		cur->bc_private.b.allocated = 0;
+		cur->bc_ino.allocated = 0;
 
 	xfs_bmap_check_leaf_extents(cur, ip, whichfork);
 done:
@@ -3058,7 +3164,7 @@ xfs_bmap_adjacent(
 	mp = ap->ip->i_mount;
 	nullfb = ap->tp->t_firstblock == NULLFSBLOCK;
 	rt = XFS_IS_REALTIME_INODE(ap->ip) &&
-		xfs_alloc_is_userdata(ap->datatype);
+		(ap->datatype & XFS_ALLOC_USERDATA);
 	fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp,
 							ap->tp->t_firstblock);
 	/*
@@ -3203,11 +3309,12 @@ xfs_bmap_longest_free_extent(
 	pag = xfs_perag_get(mp, ag);
 	if (!pag->pagf_init) {
 		error = xfs_alloc_pagf_init(mp, tp, ag, XFS_ALLOC_FLAG_TRYLOCK);
-		if (error)
-			goto out;
-
-		if (!pag->pagf_init) {
-			*notinit = 1;
+		if (error) {
+			/* Couldn't lock the AGF, so skip this AG. */
+			if (error == -EAGAIN) {
+				*notinit = 1;
+				error = 0;
+			}
 			goto out;
 		}
 	}
@@ -3411,7 +3518,7 @@ xfs_bmap_btalloc(
 
 	if (ap->flags & XFS_BMAPI_COWFORK)
 		align = xfs_get_cowextsz_hint(ap->ip);
-	else if (xfs_alloc_is_userdata(ap->datatype))
+	else if (ap->datatype & XFS_ALLOC_USERDATA)
 		align = xfs_get_extsz_hint(ap->ip);
 	if (align) {
 		error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev,
@@ -3426,7 +3533,7 @@ xfs_bmap_btalloc(
 	fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp,
 							ap->tp->t_firstblock);
 	if (nullfb) {
-		if (xfs_alloc_is_userdata(ap->datatype) &&
+		if ((ap->datatype & XFS_ALLOC_USERDATA) &&
 		    xfs_inode_is_filestream(ap->ip)) {
 			ag = xfs_filestream_lookup_ag(ap->ip);
 			ag = (ag != NULLAGNUMBER) ? ag : 0;
@@ -3466,7 +3573,7 @@ xfs_bmap_btalloc(
 		 * enough for the request.  If one isn't found, then adjust
 		 * the minimum allocation size to the largest space found.
 		 */
-		if (xfs_alloc_is_userdata(ap->datatype) &&
+		if ((ap->datatype & XFS_ALLOC_USERDATA) &&
 		    xfs_inode_is_filestream(ap->ip))
 			error = xfs_bmap_btalloc_filestreams(ap, &args, &blen);
 		else
@@ -3500,13 +3607,11 @@ xfs_bmap_btalloc(
 			args.mod = args.prod - args.mod;
 	}
 	/*
-	 * If we are not low on available data blocks, and the
-	 * underlying logical volume manager is a stripe, and
-	 * the file offset is zero then try to allocate data
-	 * blocks on stripe unit boundary.
-	 * NOTE: ap->aeof is only set if the allocation length
-	 * is >= the stripe unit and the allocation offset is
-	 * at the end of file.
+	 * If we are not low on available data blocks, and the underlying
+	 * logical volume manager is a stripe, and the file offset is zero then
+	 * try to allocate data blocks on stripe unit boundary. NOTE: ap->aeof
+	 * is only set if the allocation length is >= the stripe unit and the
+	 * allocation offset is at the end of file.
 	 */
 	if (!(ap->tp->t_flags & XFS_TRANS_LOWMODE) && ap->aeof) {
 		if (!ap->offset) {
@@ -3514,9 +3619,11 @@ xfs_bmap_btalloc(
 			atype = args.type;
 			isaligned = 1;
 			/*
-			 * Adjust for alignment
+			 * Adjust minlen to try and preserve alignment if we
+			 * can't guarantee an aligned maxlen extent.
 			 */
-			if (blen > args.alignment && blen <= args.maxlen)
+			if (blen > args.alignment &&
+			    blen <= args.maxlen + args.alignment)
 				args.minlen = blen - args.alignment;
 			args.minalignslop = 0;
 		} else {
@@ -3554,8 +3661,6 @@ xfs_bmap_btalloc(
 	args.wasdel = ap->wasdel;
 	args.resv = XFS_AG_RESV_NONE;
 	args.datatype = ap->datatype;
-	if (ap->datatype & XFS_ALLOC_USERDATA_ZERO)
-		args.ip = ap->ip;
 
 	error = xfs_alloc_vextent(&args);
 	if (error)
@@ -3640,20 +3745,6 @@ xfs_bmap_btalloc(
 	return 0;
 }
 
-/*
- * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
- * It figures out where to ask the underlying allocator to put the new extent.
- */
-STATIC int
-xfs_bmap_alloc(
-	struct xfs_bmalloca	*ap)	/* bmap alloc argument struct */
-{
-	if (XFS_IS_REALTIME_INODE(ap->ip) &&
-	    xfs_alloc_is_userdata(ap->datatype))
-		return xfs_bmap_rtalloc(ap);
-	return xfs_bmap_btalloc(ap);
-}
-
 /* Trim extent to fit a logical block range. */
 void
 xfs_trim_extent(
@@ -3815,11 +3906,8 @@ xfs_bmapi_read(
 			   XFS_BMAPI_COWFORK)));
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED|XFS_ILOCK_EXCL));
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT("xfs_bmapi_read", XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
@@ -4010,6 +4098,39 @@ out_unreserve_quota:
 }
 
 static int
+xfs_bmap_alloc_userdata(
+	struct xfs_bmalloca	*bma)
+{
+	struct xfs_mount	*mp = bma->ip->i_mount;
+	int			whichfork = xfs_bmapi_whichfork(bma->flags);
+	int			error;
+
+	/*
+	 * Set the data type being allocated. For the data fork, the first data
+	 * in the file is treated differently to all other allocations. For the
+	 * attribute fork, we only need to ensure the allocated range is not on
+	 * the busy list.
+	 */
+	bma->datatype = XFS_ALLOC_NOBUSY;
+	if (whichfork == XFS_DATA_FORK) {
+		bma->datatype |= XFS_ALLOC_USERDATA;
+		if (bma->offset == 0)
+			bma->datatype |= XFS_ALLOC_INITIAL_USER_DATA;
+
+		if (mp->m_dalign && bma->length >= mp->m_dalign) {
+			error = xfs_bmap_isaeof(bma, whichfork);
+			if (error)
+				return error;
+		}
+
+		if (XFS_IS_REALTIME_INODE(bma->ip))
+			return xfs_bmap_rtalloc(bma);
+	}
+
+	return xfs_bmap_btalloc(bma);
+}
+
+static int
 xfs_bmapi_allocate(
 	struct xfs_bmalloca	*bma)
 {
@@ -4028,7 +4149,8 @@ xfs_bmapi_allocate(
 	if (bma->wasdel) {
 		bma->length = (xfs_extlen_t)bma->got.br_blockcount;
 		bma->offset = bma->got.br_startoff;
-		xfs_iext_peek_prev_extent(ifp, &bma->icur, &bma->prev);
+		if (!xfs_iext_peek_prev_extent(ifp, &bma->icur, &bma->prev))
+			bma->prev.br_startoff = NULLFILEOFF;
 	} else {
 		bma->length = XFS_FILBLKS_MIN(bma->length, MAXEXTLEN);
 		if (!bma->eof)
@@ -4036,43 +4158,24 @@ xfs_bmapi_allocate(
 					bma->got.br_startoff - bma->offset);
 	}
 
-	/*
-	 * Set the data type being allocated. For the data fork, the first data
-	 * in the file is treated differently to all other allocations. For the
-	 * attribute fork, we only need to ensure the allocated range is not on
-	 * the busy list.
-	 */
-	if (!(bma->flags & XFS_BMAPI_METADATA)) {
-		bma->datatype = XFS_ALLOC_NOBUSY;
-		if (whichfork == XFS_DATA_FORK) {
-			if (bma->offset == 0)
-				bma->datatype |= XFS_ALLOC_INITIAL_USER_DATA;
-			else
-				bma->datatype |= XFS_ALLOC_USERDATA;
-		}
-		if (bma->flags & XFS_BMAPI_ZERO)
-			bma->datatype |= XFS_ALLOC_USERDATA_ZERO;
-	}
+	if (bma->flags & XFS_BMAPI_CONTIG)
+		bma->minlen = bma->length;
+	else
+		bma->minlen = 1;
 
-	bma->minlen = (bma->flags & XFS_BMAPI_CONTIG) ? bma->length : 1;
+	if (bma->flags & XFS_BMAPI_METADATA)
+		error = xfs_bmap_btalloc(bma);
+	else
+		error = xfs_bmap_alloc_userdata(bma);
+	if (error || bma->blkno == NULLFSBLOCK)
+		return error;
 
-	/*
-	 * Only want to do the alignment at the eof if it is userdata and
-	 * allocation length is larger than a stripe unit.
-	 */
-	if (mp->m_dalign && bma->length >= mp->m_dalign &&
-	    !(bma->flags & XFS_BMAPI_METADATA) && whichfork == XFS_DATA_FORK) {
-		error = xfs_bmap_isaeof(bma, whichfork);
+	if (bma->flags & XFS_BMAPI_ZERO) {
+		error = xfs_zero_extent(bma->ip, bma->blkno, bma->length);
 		if (error)
 			return error;
 	}
 
-	error = xfs_bmap_alloc(bma);
-	if (error)
-		return error;
-
-	if (bma->blkno == NULLFSBLOCK)
-		return 0;
 	if ((ifp->if_flags & XFS_IFBROOT) && !bma->cur)
 		bma->cur = xfs_bmbt_init_cursor(mp, bma->tp, bma->ip, whichfork);
 	/*
@@ -4082,8 +4185,8 @@ xfs_bmapi_allocate(
 	bma->nallocs++;
 
 	if (bma->cur)
-		bma->cur->bc_private.b.flags =
-			bma->wasdel ? XFS_BTCUR_BPRV_WASDEL : 0;
+		bma->cur->bc_ino.flags =
+			bma->wasdel ? XFS_BTCUR_BMBT_WASDEL : 0;
 
 	bma->got.br_startoff = bma->offset;
 	bma->got.br_startblock = bma->blkno;
@@ -4312,11 +4415,8 @@ xfs_bmapi_write(
 	ASSERT((flags & (XFS_BMAPI_PREALLOC | XFS_BMAPI_ZERO)) !=
 			(XFS_BMAPI_PREALLOC | XFS_BMAPI_ZERO));
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT("xfs_bmapi_write", XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
@@ -4456,16 +4556,21 @@ int
 xfs_bmapi_convert_delalloc(
 	struct xfs_inode	*ip,
 	int			whichfork,
-	xfs_fileoff_t		offset_fsb,
-	struct xfs_bmbt_irec	*imap,
+	xfs_off_t		offset,
+	struct iomap		*iomap,
 	unsigned int		*seq)
 {
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
 	struct xfs_mount	*mp = ip->i_mount;
+	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	struct xfs_bmalloca	bma = { NULL };
+	uint16_t		flags = 0;
 	struct xfs_trans	*tp;
 	int			error;
 
+	if (whichfork == XFS_COW_FORK)
+		flags |= IOMAP_F_SHARED;
+
 	/*
 	 * Space for the extent and indirect blocks was reserved when the
 	 * delalloc extent was created so there's no need to do so here.
@@ -4495,7 +4600,7 @@ xfs_bmapi_convert_delalloc(
 	 * the extent.  Just return the real extent at this offset.
 	 */
 	if (!isnullstartblock(bma.got.br_startblock)) {
-		*imap = bma.got;
+		xfs_bmbt_to_iomap(ip, iomap, &bma.got, flags);
 		*seq = READ_ONCE(ifp->if_seq);
 		goto out_trans_cancel;
 	}
@@ -4505,7 +4610,6 @@ xfs_bmapi_convert_delalloc(
 	bma.wasdel = true;
 	bma.offset = bma.got.br_startoff;
 	bma.length = max_t(xfs_filblks_t, bma.got.br_blockcount, MAXEXTLEN);
-	bma.total = XFS_EXTENTADD_SPACE_RES(ip->i_mount, XFS_DATA_FORK);
 	bma.minleft = xfs_bmapi_minleft(tp, ip, whichfork);
 	if (whichfork == XFS_COW_FORK)
 		bma.flags = XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC;
@@ -4528,7 +4632,7 @@ xfs_bmapi_convert_delalloc(
 	XFS_STATS_INC(mp, xs_xstrat_quick);
 
 	ASSERT(!isnullstartblock(bma.got.br_startblock));
-	*imap = bma.got;
+	xfs_bmbt_to_iomap(ip, iomap, &bma.got, flags);
 	*seq = READ_ONCE(ifp->if_seq);
 
 	if (whichfork == XFS_COW_FORK)
@@ -4578,11 +4682,8 @@ xfs_bmapi_remap(
 	ASSERT((flags & (XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC)) !=
 			(XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC));
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT("xfs_bmapi_remap", XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
@@ -4606,7 +4707,7 @@ xfs_bmapi_remap(
 
 	if (ifp->if_flags & XFS_IFBROOT) {
 		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
-		cur->bc_private.b.flags = 0;
+		cur->bc_ino.flags = 0;
 	}
 
 	got.br_startoff = bno;
@@ -5013,7 +5114,10 @@ xfs_bmap_del_extent_real(
 		error = xfs_bmbt_lookup_eq(cur, &got, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 	}
 
 	if (got.br_startoff == del->br_startoff)
@@ -5037,7 +5141,10 @@ xfs_bmap_del_extent_real(
 		}
 		if ((error = xfs_btree_delete(cur, &i)))
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		break;
 	case BMAP_LEFT_FILLING:
 		/*
@@ -5108,7 +5215,10 @@ xfs_bmap_del_extent_real(
 				error = xfs_bmbt_lookup_eq(cur, &got, &i);
 				if (error)
 					goto done;
-				XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+				if (XFS_IS_CORRUPT(mp, i != 1)) {
+					error = -EFSCORRUPTED;
+					goto done;
+				}
 				/*
 				 * Update the btree record back
 				 * to the original value.
@@ -5125,7 +5235,10 @@ xfs_bmap_del_extent_real(
 				error = -ENOSPC;
 				goto done;
 			}
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto done;
+			}
 		} else
 			flags |= xfs_ilog_fext(whichfork);
 		XFS_IFORK_NEXT_SET(ip, whichfork,
@@ -5192,7 +5305,7 @@ __xfs_bunmapi(
 	int			isrt;		/* freeing in rt area */
 	int			logflags;	/* transaction logging flags */
 	xfs_extlen_t		mod;		/* rt extent offset */
-	struct xfs_mount	*mp;		/* mount structure */
+	struct xfs_mount	*mp = ip->i_mount;
 	int			tmp_logflags;	/* partial logging flags */
 	int			wasdel;		/* was a delayed alloc extent */
 	int			whichfork;	/* data or attribute fork */
@@ -5209,14 +5322,8 @@ __xfs_bunmapi(
 	whichfork = xfs_bmapi_whichfork(flags);
 	ASSERT(whichfork != XFS_COW_FORK);
 	ifp = XFS_IFORK_PTR(ip, whichfork);
-	if (unlikely(
-	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
-		XFS_ERROR_REPORT("xfs_bunmapi", XFS_ERRLEVEL_LOW,
-				 ip->i_mount);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)))
 		return -EFSCORRUPTED;
-	}
-	mp = ip->i_mount;
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
 
@@ -5255,7 +5362,7 @@ __xfs_bunmapi(
 	if (ifp->if_flags & XFS_IFBROOT) {
 		ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE);
 		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
-		cur->bc_private.b.flags = 0;
+		cur->bc_ino.flags = 0;
 	} else
 		cur = NULL;
 
@@ -5300,7 +5407,7 @@ __xfs_bunmapi(
 		 * Make sure we don't touch multiple AGF headers out of order
 		 * in a single transaction, as that could cause AB-BA deadlocks.
 		 */
-		if (!wasdel) {
+		if (!wasdel && !isrt) {
 			agno = XFS_FSB_TO_AGNO(mp, del.br_startblock);
 			if (prev_agno != NULLAGNUMBER && prev_agno > agno)
 				break;
@@ -5376,16 +5483,17 @@ __xfs_bunmapi(
 		}
 		div_u64_rem(del.br_startblock, mp->m_sb.sb_rextsize, &mod);
 		if (mod) {
+			xfs_extlen_t off = mp->m_sb.sb_rextsize - mod;
+
 			/*
 			 * Realtime extent is lined up at the end but not
 			 * at the front.  We'll get rid of full extents if
 			 * we can.
 			 */
-			mod = mp->m_sb.sb_rextsize - mod;
-			if (del.br_blockcount > mod) {
-				del.br_blockcount -= mod;
-				del.br_startoff += mod;
-				del.br_startblock += mod;
+			if (del.br_blockcount > off) {
+				del.br_blockcount -= off;
+				del.br_startoff += off;
+				del.br_startblock += off;
 			} else if (del.br_startoff == start &&
 				   (del.br_state == XFS_EXT_UNWRITTEN ||
 				    tp->t_blk_res == 0)) {
@@ -5403,6 +5511,7 @@ __xfs_bunmapi(
 				continue;
 			} else if (del.br_state == XFS_EXT_UNWRITTEN) {
 				struct xfs_bmbt_irec	prev;
+				xfs_fileoff_t		unwrite_start;
 
 				/*
 				 * This one is already unwritten.
@@ -5416,12 +5525,13 @@ __xfs_bunmapi(
 				ASSERT(!isnullstartblock(prev.br_startblock));
 				ASSERT(del.br_startblock ==
 				       prev.br_startblock + prev.br_blockcount);
-				if (prev.br_startoff < start) {
-					mod = start - prev.br_startoff;
-					prev.br_blockcount -= mod;
-					prev.br_startblock += mod;
-					prev.br_startoff = start;
-				}
+				unwrite_start = max3(start,
+						     del.br_startoff - mod,
+						     prev.br_startoff);
+				mod = unwrite_start - prev.br_startoff;
+				prev.br_startoff = unwrite_start;
+				prev.br_startblock += mod;
+				prev.br_blockcount -= mod;
 				prev.br_state = XFS_EXT_UNWRITTEN;
 				error = xfs_bmap_add_extent_unwritten_real(tp,
 						ip, whichfork, &icur, &cur,
@@ -5508,7 +5618,7 @@ error0:
 		xfs_trans_log_inode(tp, ip, logflags);
 	if (cur) {
 		if (!error)
-			cur->bc_private.b.allocated = 0;
+			cur->bc_ino.allocated = 0;
 		xfs_btree_del_cursor(cur, error);
 	}
 	return error;
@@ -5610,18 +5720,21 @@ xfs_bmse_merge(
 	error = xfs_bmbt_lookup_eq(cur, got, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+	if (XFS_IS_CORRUPT(mp, i != 1))
+		return -EFSCORRUPTED;
 
 	error = xfs_btree_delete(cur, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+	if (XFS_IS_CORRUPT(mp, i != 1))
+		return -EFSCORRUPTED;
 
 	/* lookup and update size of the previous extent */
 	error = xfs_bmbt_lookup_eq(cur, left, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+	if (XFS_IS_CORRUPT(mp, i != 1))
+		return -EFSCORRUPTED;
 
 	error = xfs_bmbt_update(cur, &new);
 	if (error)
@@ -5669,7 +5782,8 @@ xfs_bmap_shift_update_extent(
 		error = xfs_bmbt_lookup_eq(cur, &prev, &i);
 		if (error)
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(mp, i == 1);
+		if (XFS_IS_CORRUPT(mp, i != 1))
+			return -EFSCORRUPTED;
 
 		error = xfs_bmbt_update(cur, got);
 		if (error)
@@ -5705,11 +5819,8 @@ xfs_bmap_collapse_extents(
 	int			error = 0;
 	int			logflags = 0;
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
@@ -5726,15 +5837,17 @@ xfs_bmap_collapse_extents(
 
 	if (ifp->if_flags & XFS_IFBROOT) {
 		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
-		cur->bc_private.b.flags = 0;
+		cur->bc_ino.flags = 0;
 	}
 
 	if (!xfs_iext_lookup_extent(ip, ifp, *next_fsb, &icur, &got)) {
 		*done = true;
 		goto del_cursor;
 	}
-	XFS_WANT_CORRUPTED_GOTO(mp, !isnullstartblock(got.br_startblock),
-				del_cursor);
+	if (XFS_IS_CORRUPT(mp, isnullstartblock(got.br_startblock))) {
+		error = -EFSCORRUPTED;
+		goto del_cursor;
+	}
 
 	new_startoff = got.br_startoff - offset_shift_fsb;
 	if (xfs_iext_peek_prev_extent(ifp, &icur, &prev)) {
@@ -5823,11 +5936,8 @@ xfs_bmap_insert_extents(
 	int			error = 0;
 	int			logflags = 0;
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
@@ -5844,7 +5954,7 @@ xfs_bmap_insert_extents(
 
 	if (ifp->if_flags & XFS_IFBROOT) {
 		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
-		cur->bc_private.b.flags = 0;
+		cur->bc_ino.flags = 0;
 	}
 
 	if (*next_fsb == NULLFSBLOCK) {
@@ -5860,11 +5970,13 @@ xfs_bmap_insert_extents(
 			goto del_cursor;
 		}
 	}
-	XFS_WANT_CORRUPTED_GOTO(mp, !isnullstartblock(got.br_startblock),
-				del_cursor);
+	if (XFS_IS_CORRUPT(mp, isnullstartblock(got.br_startblock))) {
+		error = -EFSCORRUPTED;
+		goto del_cursor;
+	}
 
-	if (stop_fsb >= got.br_startoff + got.br_blockcount) {
-		error = -EIO;
+	if (XFS_IS_CORRUPT(mp, stop_fsb > got.br_startoff)) {
+		error = -EFSCORRUPTED;
 		goto del_cursor;
 	}
 
@@ -5911,8 +6023,8 @@ del_cursor:
  * @split_fsb is a block where the extents is split.  If split_fsb lies in a
  * hole or the first block of extents, just return 0.
  */
-STATIC int
-xfs_bmap_split_extent_at(
+int
+xfs_bmap_split_extent(
 	struct xfs_trans	*tp,
 	struct xfs_inode	*ip,
 	xfs_fileoff_t		split_fsb)
@@ -5929,12 +6041,8 @@ xfs_bmap_split_extent_at(
 	int				logflags = 0;
 	int				i = 0;
 
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT("xfs_bmap_split_extent_at",
-				 XFS_ERRLEVEL_LOW, mp);
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, whichfork)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
 		return -EFSCORRUPTED;
 	}
 
@@ -5964,11 +6072,14 @@ xfs_bmap_split_extent_at(
 
 	if (ifp->if_flags & XFS_IFBROOT) {
 		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
-		cur->bc_private.b.flags = 0;
+		cur->bc_ino.flags = 0;
 		error = xfs_bmbt_lookup_eq(cur, &got, &i);
 		if (error)
 			goto del_cursor;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, del_cursor);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto del_cursor;
+		}
 	}
 
 	got.br_blockcount = gotblkcnt;
@@ -5993,11 +6104,17 @@ xfs_bmap_split_extent_at(
 		error = xfs_bmbt_lookup_eq(cur, &new, &i);
 		if (error)
 			goto del_cursor;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 0, del_cursor);
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			error = -EFSCORRUPTED;
+			goto del_cursor;
+		}
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto del_cursor;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, del_cursor);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto del_cursor;
+		}
 	}
 
 	/*
@@ -6014,7 +6131,7 @@ xfs_bmap_split_extent_at(
 
 del_cursor:
 	if (cur) {
-		cur->bc_private.b.allocated = 0;
+		cur->bc_ino.allocated = 0;
 		xfs_btree_del_cursor(cur, error);
 	}
 
@@ -6023,34 +6140,6 @@ del_cursor:
 	return error;
 }
 
-int
-xfs_bmap_split_extent(
-	struct xfs_inode        *ip,
-	xfs_fileoff_t           split_fsb)
-{
-	struct xfs_mount        *mp = ip->i_mount;
-	struct xfs_trans        *tp;
-	int                     error;
-
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write,
-			XFS_DIOSTRAT_SPACE_RES(mp, 0), 0, 0, &tp);
-	if (error)
-		return error;
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-
-	error = xfs_bmap_split_extent_at(tp, ip, split_fsb);
-	if (error)
-		goto out;
-
-	return xfs_trans_commit(tp);
-
-out:
-	xfs_trans_cancel(tp);
-	return error;
-}
-
 /* Deferred mapping is only for real extents in the data fork. */
 static bool
 xfs_bmap_is_update_needed(
diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index e2798c6f3a5f..f3259ad5c22c 100644
--- a/fs/xfs/libxfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -222,14 +222,14 @@ int	xfs_bmap_can_insert_extents(struct xfs_inode *ip, xfs_fileoff_t off,
 int	xfs_bmap_insert_extents(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_fileoff_t *next_fsb, xfs_fileoff_t offset_shift_fsb,
 		bool *done, xfs_fileoff_t stop_fsb);
-int	xfs_bmap_split_extent(struct xfs_inode *ip, xfs_fileoff_t split_offset);
+int	xfs_bmap_split_extent(struct xfs_trans *tp, struct xfs_inode *ip,
+		xfs_fileoff_t split_offset);
 int	xfs_bmapi_reserve_delalloc(struct xfs_inode *ip, int whichfork,
 		xfs_fileoff_t off, xfs_filblks_t len, xfs_filblks_t prealloc,
 		struct xfs_bmbt_irec *got, struct xfs_iext_cursor *cur,
 		int eof);
 int	xfs_bmapi_convert_delalloc(struct xfs_inode *ip, int whichfork,
-		xfs_fileoff_t offset_fsb, struct xfs_bmbt_irec *imap,
-		unsigned int *seq);
+		xfs_off_t offset, struct iomap *iomap, unsigned int *seq);
 int	xfs_bmap_add_extent_unwritten_real(struct xfs_trans *tp,
 		struct xfs_inode *ip, int whichfork,
 		struct xfs_iext_cursor *icur, struct xfs_btree_cur **curp,
diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c
index ffe608d2a2d9..295a59cf8840 100644
--- a/fs/xfs/libxfs/xfs_bmap_btree.c
+++ b/fs/xfs/libxfs/xfs_bmap_btree.c
@@ -166,13 +166,13 @@ xfs_bmbt_dup_cursor(
 	struct xfs_btree_cur	*new;
 
 	new = xfs_bmbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_private.b.ip, cur->bc_private.b.whichfork);
+			cur->bc_ino.ip, cur->bc_ino.whichfork);
 
 	/*
 	 * Copy the firstblock, dfops, and flags values,
 	 * since init cursor doesn't get them.
 	 */
-	new->bc_private.b.flags = cur->bc_private.b.flags;
+	new->bc_ino.flags = cur->bc_ino.flags;
 
 	return new;
 }
@@ -183,12 +183,12 @@ xfs_bmbt_update_cursor(
 	struct xfs_btree_cur	*dst)
 {
 	ASSERT((dst->bc_tp->t_firstblock != NULLFSBLOCK) ||
-	       (dst->bc_private.b.ip->i_d.di_flags & XFS_DIFLAG_REALTIME));
+	       (dst->bc_ino.ip->i_d.di_flags & XFS_DIFLAG_REALTIME));
 
-	dst->bc_private.b.allocated += src->bc_private.b.allocated;
+	dst->bc_ino.allocated += src->bc_ino.allocated;
 	dst->bc_tp->t_firstblock = src->bc_tp->t_firstblock;
 
-	src->bc_private.b.allocated = 0;
+	src->bc_ino.allocated = 0;
 }
 
 STATIC int
@@ -205,8 +205,8 @@ xfs_bmbt_alloc_block(
 	args.tp = cur->bc_tp;
 	args.mp = cur->bc_mp;
 	args.fsbno = cur->bc_tp->t_firstblock;
-	xfs_rmap_ino_bmbt_owner(&args.oinfo, cur->bc_private.b.ip->i_ino,
-			cur->bc_private.b.whichfork);
+	xfs_rmap_ino_bmbt_owner(&args.oinfo, cur->bc_ino.ip->i_ino,
+			cur->bc_ino.whichfork);
 
 	if (args.fsbno == NULLFSBLOCK) {
 		args.fsbno = be64_to_cpu(start->l);
@@ -230,7 +230,7 @@ xfs_bmbt_alloc_block(
 	}
 
 	args.minlen = args.maxlen = args.prod = 1;
-	args.wasdel = cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL;
+	args.wasdel = cur->bc_ino.flags & XFS_BTCUR_BMBT_WASDEL;
 	if (!args.wasdel && args.tp->t_blk_res == 0) {
 		error = -ENOSPC;
 		goto error0;
@@ -259,10 +259,10 @@ xfs_bmbt_alloc_block(
 
 	ASSERT(args.len == 1);
 	cur->bc_tp->t_firstblock = args.fsbno;
-	cur->bc_private.b.allocated++;
-	cur->bc_private.b.ip->i_d.di_nblocks++;
-	xfs_trans_log_inode(args.tp, cur->bc_private.b.ip, XFS_ILOG_CORE);
-	xfs_trans_mod_dquot_byino(args.tp, cur->bc_private.b.ip,
+	cur->bc_ino.allocated++;
+	cur->bc_ino.ip->i_d.di_nblocks++;
+	xfs_trans_log_inode(args.tp, cur->bc_ino.ip, XFS_ILOG_CORE);
+	xfs_trans_mod_dquot_byino(args.tp, cur->bc_ino.ip,
 			XFS_TRANS_DQ_BCOUNT, 1L);
 
 	new->l = cpu_to_be64(args.fsbno);
@@ -280,12 +280,12 @@ xfs_bmbt_free_block(
 	struct xfs_buf		*bp)
 {
 	struct xfs_mount	*mp = cur->bc_mp;
-	struct xfs_inode	*ip = cur->bc_private.b.ip;
+	struct xfs_inode	*ip = cur->bc_ino.ip;
 	struct xfs_trans	*tp = cur->bc_tp;
 	xfs_fsblock_t		fsbno = XFS_DADDR_TO_FSB(mp, XFS_BUF_ADDR(bp));
 	struct xfs_owner_info	oinfo;
 
-	xfs_rmap_ino_bmbt_owner(&oinfo, ip->i_ino, cur->bc_private.b.whichfork);
+	xfs_rmap_ino_bmbt_owner(&oinfo, ip->i_ino, cur->bc_ino.whichfork);
 	xfs_bmap_add_free(cur->bc_tp, fsbno, 1, &oinfo);
 	ip->i_d.di_nblocks--;
 
@@ -302,8 +302,8 @@ xfs_bmbt_get_minrecs(
 	if (level == cur->bc_nlevels - 1) {
 		struct xfs_ifork	*ifp;
 
-		ifp = XFS_IFORK_PTR(cur->bc_private.b.ip,
-				    cur->bc_private.b.whichfork);
+		ifp = XFS_IFORK_PTR(cur->bc_ino.ip,
+				    cur->bc_ino.whichfork);
 
 		return xfs_bmbt_maxrecs(cur->bc_mp,
 					ifp->if_broot_bytes, level == 0) / 2;
@@ -320,8 +320,8 @@ xfs_bmbt_get_maxrecs(
 	if (level == cur->bc_nlevels - 1) {
 		struct xfs_ifork	*ifp;
 
-		ifp = XFS_IFORK_PTR(cur->bc_private.b.ip,
-				    cur->bc_private.b.whichfork);
+		ifp = XFS_IFORK_PTR(cur->bc_ino.ip,
+				    cur->bc_ino.whichfork);
 
 		return xfs_bmbt_maxrecs(cur->bc_mp,
 					ifp->if_broot_bytes, level == 0);
@@ -347,7 +347,7 @@ xfs_bmbt_get_dmaxrecs(
 {
 	if (level != cur->bc_nlevels - 1)
 		return cur->bc_mp->m_bmap_dmxr[level != 0];
-	return xfs_bmdr_maxrecs(cur->bc_private.b.forksize, level == 0);
+	return xfs_bmdr_maxrecs(cur->bc_ino.forksize, level == 0);
 }
 
 STATIC void
@@ -566,11 +566,11 @@ xfs_bmbt_init_cursor(
 	if (xfs_sb_version_hascrc(&mp->m_sb))
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
-	cur->bc_private.b.forksize = XFS_IFORK_SIZE(ip, whichfork);
-	cur->bc_private.b.ip = ip;
-	cur->bc_private.b.allocated = 0;
-	cur->bc_private.b.flags = 0;
-	cur->bc_private.b.whichfork = whichfork;
+	cur->bc_ino.forksize = XFS_IFORK_SIZE(ip, whichfork);
+	cur->bc_ino.ip = ip;
+	cur->bc_ino.allocated = 0;
+	cur->bc_ino.flags = 0;
+	cur->bc_ino.whichfork = whichfork;
 
 	return cur;
 }
@@ -644,7 +644,7 @@ xfs_bmbt_change_owner(
 	cur = xfs_bmbt_init_cursor(ip->i_mount, tp, ip, whichfork);
 	if (!cur)
 		return -ENOMEM;
-	cur->bc_private.b.flags |= XFS_BTCUR_BPRV_INVALID_OWNER;
+	cur->bc_ino.flags |= XFS_BTCUR_BMBT_INVALID_OWNER;
 
 	error = xfs_btree_change_owner(cur, new_owner, buffer_list);
 	xfs_btree_del_cursor(cur, error);
diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index 71de937f9e64..2d25bab68764 100644
--- a/fs/xfs/libxfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -20,6 +20,7 @@
 #include "xfs_trace.h"
 #include "xfs_alloc.h"
 #include "xfs_log.h"
+#include "xfs_btree_staging.h"
 
 /*
  * Cursor allocation zone.
@@ -105,11 +106,10 @@ xfs_btree_check_lblock(
 	xfs_failaddr_t		fa;
 
 	fa = __xfs_btree_check_lblock(cur, block, level, bp);
-	if (unlikely(XFS_TEST_ERROR(fa != NULL, mp,
-			XFS_ERRTAG_BTREE_CHECK_LBLOCK))) {
+	if (XFS_IS_CORRUPT(mp, fa != NULL) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BTREE_CHECK_LBLOCK)) {
 		if (bp)
 			trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
@@ -169,11 +169,10 @@ xfs_btree_check_sblock(
 	xfs_failaddr_t		fa;
 
 	fa = __xfs_btree_check_sblock(cur, block, level, bp);
-	if (unlikely(XFS_TEST_ERROR(fa != NULL, mp,
-			XFS_ERRTAG_BTREE_CHECK_SBLOCK))) {
+	if (XFS_IS_CORRUPT(mp, fa != NULL) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BTREE_CHECK_SBLOCK)) {
 		if (bp)
 			trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
@@ -216,7 +215,7 @@ xfs_btree_check_sptr(
 {
 	if (level <= 0)
 		return false;
-	return xfs_verify_agbno(cur->bc_mp, cur->bc_private.a.agno, agbno);
+	return xfs_verify_agbno(cur->bc_mp, cur->bc_ag.agno, agbno);
 }
 
 /*
@@ -236,8 +235,8 @@ xfs_btree_check_ptr(
 			return 0;
 		xfs_err(cur->bc_mp,
 "Inode %llu fork %d: Corrupt btree %d pointer at level %d index %d.",
-				cur->bc_private.b.ip->i_ino,
-				cur->bc_private.b.whichfork, cur->bc_btnum,
+				cur->bc_ino.ip->i_ino,
+				cur->bc_ino.whichfork, cur->bc_btnum,
 				level, index);
 	} else {
 		if (xfs_btree_check_sptr(cur, be32_to_cpu((&ptr->s)[index]),
@@ -245,7 +244,7 @@ xfs_btree_check_ptr(
 			return 0;
 		xfs_err(cur->bc_mp,
 "AG %u: Corrupt btree %d pointer at level %d index %d.",
-				cur->bc_private.a.agno, cur->bc_btnum,
+				cur->bc_ag.agno, cur->bc_btnum,
 				level, index);
 	}
 
@@ -380,11 +379,13 @@ xfs_btree_del_cursor(
 	 * allocated indirect blocks' accounting.
 	 */
 	ASSERT(cur->bc_btnum != XFS_BTNUM_BMAP ||
-	       cur->bc_private.b.allocated == 0);
+	       cur->bc_ino.allocated == 0);
 	/*
 	 * Free the cursor.
 	 */
-	kmem_zone_free(xfs_btree_cur_zone, cur);
+	if (unlikely(cur->bc_flags & XFS_BTREE_STAGING))
+		kmem_free((void *)cur->bc_ops);
+	kmem_cache_free(xfs_btree_cur_zone, cur);
 }
 
 /*
@@ -644,6 +645,17 @@ xfs_btree_ptr_addr(
 		((char *)block + xfs_btree_ptr_offset(cur, n, level));
 }
 
+struct xfs_ifork *
+xfs_btree_ifork_ptr(
+	struct xfs_btree_cur	*cur)
+{
+	ASSERT(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE);
+
+	if (cur->bc_flags & XFS_BTREE_STAGING)
+		return cur->bc_ino.ifake->if_fork;
+	return XFS_IFORK_PTR(cur->bc_ino.ip, cur->bc_ino.whichfork);
+}
+
 /*
  * Get the root block which is stored in the inode.
  *
@@ -654,9 +666,8 @@ STATIC struct xfs_btree_block *
 xfs_btree_get_iroot(
 	struct xfs_btree_cur	*cur)
 {
-	struct xfs_ifork	*ifp;
+	struct xfs_ifork	*ifp = xfs_btree_ifork_ptr(cur);
 
-	ifp = XFS_IFORK_PTR(cur->bc_private.b.ip, cur->bc_private.b.whichfork);
 	return (struct xfs_btree_block *)ifp->if_broot;
 }
 
@@ -681,61 +692,6 @@ xfs_btree_get_block(
 }
 
 /*
- * Get a buffer for the block, return it with no data read.
- * Long-form addressing.
- */
-xfs_buf_t *				/* buffer for fsbno */
-xfs_btree_get_bufl(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_fsblock_t	fsbno)		/* file system block number */
-{
-	xfs_daddr_t		d;		/* real disk block address */
-
-	ASSERT(fsbno != NULLFSBLOCK);
-	d = XFS_FSB_TO_DADDR(mp, fsbno);
-	return xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, 0);
-}
-
-/*
- * Get a buffer for the block, return it with no data read.
- * Short-form addressing.
- */
-xfs_buf_t *				/* buffer for agno/agbno */
-xfs_btree_get_bufs(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_agnumber_t	agno,		/* allocation group number */
-	xfs_agblock_t	agbno)		/* allocation group block number */
-{
-	xfs_daddr_t		d;		/* real disk block address */
-
-	ASSERT(agno != NULLAGNUMBER);
-	ASSERT(agbno != NULLAGBLOCK);
-	d = XFS_AGB_TO_DADDR(mp, agno, agbno);
-	return xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, 0);
-}
-
-/*
- * Check for the cursor referring to the last block at the given level.
- */
-int					/* 1=is last block, 0=not last block */
-xfs_btree_islastblock(
-	xfs_btree_cur_t		*cur,	/* btree cursor */
-	int			level)	/* level to check */
-{
-	struct xfs_btree_block	*block;	/* generic btree block pointer */
-	xfs_buf_t		*bp;	/* buffer containing block */
-
-	block = xfs_btree_get_block(cur, level, &bp);
-	xfs_btree_check_block(cur, block, level, bp);
-	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK);
-	else
-		return block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK);
-}
-
-/*
  * Change the cursor to point to the first record at the given level.
  * Other levels are unaffected.
  */
@@ -938,13 +894,13 @@ xfs_btree_readahead_sblock(
 
 
 	if ((lr & XFS_BTCUR_LEFTRA) && left != NULLAGBLOCK) {
-		xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno,
+		xfs_btree_reada_bufs(cur->bc_mp, cur->bc_ag.agno,
 				     left, 1, cur->bc_ops->buf_ops);
 		rval++;
 	}
 
 	if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLAGBLOCK) {
-		xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno,
+		xfs_btree_reada_bufs(cur->bc_mp, cur->bc_ag.agno,
 				     right, 1, cur->bc_ops->buf_ops);
 		rval++;
 	}
@@ -1002,7 +958,7 @@ xfs_btree_ptr_to_daddr(
 		*daddr = XFS_FSB_TO_DADDR(cur->bc_mp, fsbno);
 	} else {
 		agbno = be32_to_cpu(ptr->s);
-		*daddr = XFS_AGB_TO_DADDR(cur->bc_mp, cur->bc_private.a.agno,
+		*daddr = XFS_AGB_TO_DADDR(cur->bc_mp, cur->bc_ag.agno,
 				agbno);
 	}
 
@@ -1071,7 +1027,7 @@ xfs_btree_ptr_is_null(
 		return ptr->s == cpu_to_be32(NULLAGBLOCK);
 }
 
-STATIC void
+void
 xfs_btree_set_ptr_null(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*ptr)
@@ -1107,7 +1063,7 @@ xfs_btree_get_sibling(
 	}
 }
 
-STATIC void
+void
 xfs_btree_set_sibling(
 	struct xfs_btree_cur	*cur,
 	struct xfs_btree_block	*block,
@@ -1185,7 +1141,7 @@ xfs_btree_init_block(
 				 btnum, level, numrecs, owner, 0);
 }
 
-STATIC void
+void
 xfs_btree_init_block_cur(
 	struct xfs_btree_cur	*cur,
 	struct xfs_buf		*bp,
@@ -1201,9 +1157,9 @@ xfs_btree_init_block_cur(
 	 * code.
 	 */
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		owner = cur->bc_private.b.ip->i_ino;
+		owner = cur->bc_ino.ip->i_ino;
 	else
-		owner = cur->bc_private.a.agno;
+		owner = cur->bc_ag.agno;
 
 	xfs_btree_init_block_int(cur->bc_mp, XFS_BUF_TO_BLOCK(bp), bp->b_bn,
 				 cur->bc_btnum, level, numrecs,
@@ -1277,7 +1233,7 @@ xfs_btree_set_refs(
 	}
 }
 
-STATIC int
+int
 xfs_btree_get_buf_block(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*ptr,
@@ -1291,11 +1247,10 @@ xfs_btree_get_buf_block(
 	error = xfs_btree_ptr_to_daddr(cur, ptr, &d);
 	if (error)
 		return error;
-	*bpp = xfs_trans_get_buf(cur->bc_tp, mp->m_ddev_targp, d,
-				 mp->m_bsize, 0);
-
-	if (!*bpp)
-		return -ENOMEM;
+	error = xfs_trans_get_buf(cur->bc_tp, mp->m_ddev_targp, d, mp->m_bsize,
+			0, bpp);
+	if (error)
+		return error;
 
 	(*bpp)->b_ops = cur->bc_ops->buf_ops;
 	*block = XFS_BUF_TO_BLOCK(*bpp);
@@ -1338,7 +1293,7 @@ xfs_btree_read_buf_block(
 /*
  * Copy keys from one btree block to another.
  */
-STATIC void
+void
 xfs_btree_copy_keys(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_key	*dst_key,
@@ -1366,11 +1321,11 @@ xfs_btree_copy_recs(
 /*
  * Copy block pointers from one btree block to another.
  */
-STATIC void
+void
 xfs_btree_copy_ptrs(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*dst_ptr,
-	union xfs_btree_ptr	*src_ptr,
+	const union xfs_btree_ptr *src_ptr,
 	int			numptrs)
 {
 	ASSERT(numptrs >= 0);
@@ -1451,8 +1406,8 @@ xfs_btree_log_keys(
 				  xfs_btree_key_offset(cur, first),
 				  xfs_btree_key_offset(cur, last + 1) - 1);
 	} else {
-		xfs_trans_log_inode(cur->bc_tp, cur->bc_private.b.ip,
-				xfs_ilog_fbroot(cur->bc_private.b.whichfork));
+		xfs_trans_log_inode(cur->bc_tp, cur->bc_ino.ip,
+				xfs_ilog_fbroot(cur->bc_ino.whichfork));
 	}
 }
 
@@ -1494,8 +1449,8 @@ xfs_btree_log_ptrs(
 				xfs_btree_ptr_offset(cur, first, level),
 				xfs_btree_ptr_offset(cur, last + 1, level) - 1);
 	} else {
-		xfs_trans_log_inode(cur->bc_tp, cur->bc_private.b.ip,
-			xfs_ilog_fbroot(cur->bc_private.b.whichfork));
+		xfs_trans_log_inode(cur->bc_tp, cur->bc_ino.ip,
+			xfs_ilog_fbroot(cur->bc_ino.whichfork));
 	}
 
 }
@@ -1563,8 +1518,8 @@ xfs_btree_log_block(
 		xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF);
 		xfs_trans_log_buf(cur->bc_tp, bp, first, last);
 	} else {
-		xfs_trans_log_inode(cur->bc_tp, cur->bc_private.b.ip,
-			xfs_ilog_fbroot(cur->bc_private.b.whichfork));
+		xfs_trans_log_inode(cur->bc_tp, cur->bc_ino.ip,
+			xfs_ilog_fbroot(cur->bc_ino.whichfork));
 	}
 }
 
@@ -1801,10 +1756,10 @@ xfs_btree_lookup_get_block(
 
 	/* Check the inode owner since the verifiers don't. */
 	if (xfs_sb_version_hascrc(&cur->bc_mp->m_sb) &&
-	    !(cur->bc_private.b.flags & XFS_BTCUR_BPRV_INVALID_OWNER) &&
+	    !(cur->bc_ino.flags & XFS_BTCUR_BMBT_INVALID_OWNER) &&
 	    (cur->bc_flags & XFS_BTREE_LONG_PTRS) &&
 	    be64_to_cpu((*blkp)->bb_u.l.bb_owner) !=
-			cur->bc_private.b.ip->i_ino)
+			cur->bc_ino.ip->i_ino)
 		goto out_bad;
 
 	/* Did we get the level we were looking for? */
@@ -1820,6 +1775,7 @@ xfs_btree_lookup_get_block(
 
 out_bad:
 	*blkp = NULL;
+	xfs_buf_mark_corrupt(bp);
 	xfs_trans_brelse(cur->bc_tp, bp);
 	return -EFSCORRUPTED;
 }
@@ -1867,7 +1823,7 @@ xfs_btree_lookup(
 	XFS_BTREE_STATS_INC(cur, lookup);
 
 	/* No such thing as a zero-level tree. */
-	if (cur->bc_nlevels == 0)
+	if (XFS_IS_CORRUPT(cur->bc_mp, cur->bc_nlevels == 0))
 		return -EFSCORRUPTED;
 
 	block = NULL;
@@ -1987,7 +1943,8 @@ xfs_btree_lookup(
 			error = xfs_btree_increment(cur, 0, &i);
 			if (error)
 				goto error0;
-			XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+			if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+				return -EFSCORRUPTED;
 			*stat = 1;
 			return 0;
 		}
@@ -2408,8 +2365,6 @@ xfs_btree_lshift(
 	XFS_BTREE_STATS_ADD(cur, moves, rrecs - 1);
 	if (level > 0) {
 		/* It's a nonleaf. operate on keys and ptrs */
-		int			i;		/* loop index */
-
 		for (i = 0; i < rrecs; i++) {
 			error = xfs_btree_debug_check_ptr(cur, rpp, i + 1, level);
 			if (error)
@@ -2442,7 +2397,10 @@ xfs_btree_lshift(
 		if (error)
 			goto error0;
 		i = xfs_btree_firstrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(tcur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(tcur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		error = xfs_btree_decrement(tcur, level, &i);
 		if (error)
@@ -2609,7 +2567,10 @@ xfs_btree_rshift(
 	if (error)
 		goto error0;
 	i = xfs_btree_lastrec(tcur, level);
-	XFS_WANT_CORRUPTED_GOTO(tcur->bc_mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(tcur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 
 	error = xfs_btree_increment(tcur, level, &i);
 	if (error)
@@ -2990,9 +2951,9 @@ xfs_btree_new_iroot(
 
 	xfs_btree_copy_ptrs(cur, pp, &nptr, 1);
 
-	xfs_iroot_realloc(cur->bc_private.b.ip,
+	xfs_iroot_realloc(cur->bc_ino.ip,
 			  1 - xfs_btree_get_numrecs(cblock),
-			  cur->bc_private.b.whichfork);
+			  cur->bc_ino.whichfork);
 
 	xfs_btree_setbuf(cur, level, cbp);
 
@@ -3005,7 +2966,7 @@ xfs_btree_new_iroot(
 	xfs_btree_log_ptrs(cur, cbp, 1, be16_to_cpu(cblock->bb_numrecs));
 
 	*logflags |=
-		XFS_ILOG_CORE | xfs_ilog_fbroot(cur->bc_private.b.whichfork);
+		XFS_ILOG_CORE | xfs_ilog_fbroot(cur->bc_ino.whichfork);
 	*stat = 1;
 	return 0;
 error0:
@@ -3157,11 +3118,11 @@ xfs_btree_make_block_unfull(
 
 	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
 	    level == cur->bc_nlevels - 1) {
-		struct xfs_inode *ip = cur->bc_private.b.ip;
+		struct xfs_inode *ip = cur->bc_ino.ip;
 
 		if (numrecs < cur->bc_ops->get_dmaxrecs(cur, level)) {
 			/* A root block that can be made bigger. */
-			xfs_iroot_realloc(ip, 1, cur->bc_private.b.whichfork);
+			xfs_iroot_realloc(ip, 1, cur->bc_ino.whichfork);
 			*stat = 1;
 		} else {
 			/* A root block that needs replacing */
@@ -3463,7 +3424,10 @@ xfs_btree_insert(
 			goto error0;
 		}
 
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		level++;
 
 		/*
@@ -3504,8 +3468,8 @@ STATIC int
 xfs_btree_kill_iroot(
 	struct xfs_btree_cur	*cur)
 {
-	int			whichfork = cur->bc_private.b.whichfork;
-	struct xfs_inode	*ip = cur->bc_private.b.ip;
+	int			whichfork = cur->bc_ino.whichfork;
+	struct xfs_inode	*ip = cur->bc_ino.ip;
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
 	struct xfs_btree_block	*block;
 	struct xfs_btree_block	*cblock;
@@ -3563,8 +3527,8 @@ xfs_btree_kill_iroot(
 
 	index = numrecs - cur->bc_ops->get_maxrecs(cur, level);
 	if (index) {
-		xfs_iroot_realloc(cur->bc_private.b.ip, index,
-				  cur->bc_private.b.whichfork);
+		xfs_iroot_realloc(cur->bc_ino.ip, index,
+				  cur->bc_ino.whichfork);
 		block = ifp->if_broot;
 	}
 
@@ -3593,7 +3557,7 @@ xfs_btree_kill_iroot(
 	cur->bc_bufs[level - 1] = NULL;
 	be16_add_cpu(&block->bb_level, -1);
 	xfs_trans_log_inode(cur->bc_tp, ip,
-		XFS_ILOG_CORE | xfs_ilog_fbroot(cur->bc_private.b.whichfork));
+		XFS_ILOG_CORE | xfs_ilog_fbroot(cur->bc_ino.whichfork));
 	cur->bc_nlevels--;
 out0:
 	return 0;
@@ -3761,8 +3725,8 @@ xfs_btree_delrec(
 	 */
 	if (level == cur->bc_nlevels - 1) {
 		if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
-			xfs_iroot_realloc(cur->bc_private.b.ip, -1,
-					  cur->bc_private.b.whichfork);
+			xfs_iroot_realloc(cur->bc_ino.ip, -1,
+					  cur->bc_ino.whichfork);
 
 			error = xfs_btree_kill_iroot(cur);
 			if (error)
@@ -3867,15 +3831,24 @@ xfs_btree_delrec(
 		 * Actually any entry but the first would suffice.
 		 */
 		i = xfs_btree_lastrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		error = xfs_btree_increment(tcur, level, &i);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		i = xfs_btree_lastrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		/* Grab a pointer to the block. */
 		right = xfs_btree_get_block(tcur, level, &rbp);
@@ -3919,12 +3892,18 @@ xfs_btree_delrec(
 		rrecs = xfs_btree_get_numrecs(right);
 		if (!xfs_btree_ptr_is_null(cur, &lptr)) {
 			i = xfs_btree_firstrec(tcur, level);
-			XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+			if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 
 			error = xfs_btree_decrement(tcur, level, &i);
 			if (error)
 				goto error0;
-			XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+			if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto error0;
+			}
 		}
 	}
 
@@ -3938,13 +3917,19 @@ xfs_btree_delrec(
 		 * previous block.
 		 */
 		i = xfs_btree_firstrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		error = xfs_btree_decrement(tcur, level, &i);
 		if (error)
 			goto error0;
 		i = xfs_btree_firstrec(tcur, level);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		/* Grab a pointer to the block. */
 		left = xfs_btree_get_block(tcur, level, &lbp);
@@ -4286,6 +4271,7 @@ int
 xfs_btree_visit_blocks(
 	struct xfs_btree_cur		*cur,
 	xfs_btree_visit_blocks_fn	fn,
+	unsigned int			flags,
 	void				*data)
 {
 	union xfs_btree_ptr		lptr;
@@ -4311,6 +4297,11 @@ xfs_btree_visit_blocks(
 
 			/* save for the next iteration of the loop */
 			xfs_btree_copy_ptrs(cur, &lptr, ptr, 1);
+
+			if (!(flags & XFS_BTREE_VISIT_LEAVES))
+				continue;
+		} else if (!(flags & XFS_BTREE_VISIT_RECORDS)) {
+			continue;
 		}
 
 		/* for each buffer in the level */
@@ -4413,7 +4404,7 @@ xfs_btree_change_owner(
 	bbcoi.buffer_list = buffer_list;
 
 	return xfs_btree_visit_blocks(cur, xfs_btree_block_change_owner,
-			&bbcoi);
+			XFS_BTREE_VISIT_ALL, &bbcoi);
 }
 
 /* Verify the v5 fields of a long-format btree block. */
@@ -4865,7 +4856,7 @@ xfs_btree_count_blocks(
 {
 	*blocks = 0;
 	return xfs_btree_visit_blocks(cur, xfs_btree_count_blocks_helper,
-			blocks);
+			XFS_BTREE_VISIT_ALL, blocks);
 }
 
 /* Compare two btree pointers. */
diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
index ced1e65d1483..8626c5a81aad 100644
--- a/fs/xfs/libxfs/xfs_btree.h
+++ b/fs/xfs/libxfs/xfs_btree.h
@@ -10,6 +10,7 @@ struct xfs_buf;
 struct xfs_inode;
 struct xfs_mount;
 struct xfs_trans;
+struct xfs_ifork;
 
 extern kmem_zone_t	*xfs_btree_cur_zone;
 
@@ -177,12 +178,37 @@ union xfs_btree_irec {
 	struct xfs_refcount_irec	rc;
 };
 
-/* Per-AG btree private information. */
-union xfs_btree_cur_private {
-	struct {
-		unsigned long	nr_ops;		/* # record updates */
-		int		shape_changes;	/* # of extent splits */
-	} refc;
+/* Per-AG btree information. */
+struct xfs_btree_cur_ag {
+	union {
+		struct xfs_buf		*agbp;
+		struct xbtree_afakeroot	*afake;	/* for staging cursor */
+	};
+	xfs_agnumber_t		agno;
+	union {
+		struct {
+			unsigned long nr_ops;	/* # record updates */
+			int	shape_changes;	/* # of extent splits */
+		} refc;
+		struct {
+			bool	active;		/* allocation cursor state */
+		} abt;
+	};
+};
+
+/* Btree-in-inode cursor information */
+struct xfs_btree_cur_ino {
+	struct xfs_inode		*ip;
+	struct xbtree_ifakeroot		*ifake;	/* for staging cursor */
+	int				allocated;
+	short				forksize;
+	char				whichfork;
+	char				flags;
+/* We are converting a delalloc reservation */
+#define	XFS_BTCUR_BMBT_WASDEL		(1 << 0)
+
+/* For extent swap, ignore owner check in verifier */
+#define	XFS_BTCUR_BMBT_INVALID_OWNER	(1 << 1)
 };
 
 /*
@@ -206,21 +232,9 @@ typedef struct xfs_btree_cur
 	xfs_btnum_t	bc_btnum;	/* identifies which btree type */
 	int		bc_statoff;	/* offset of btre stats array */
 	union {
-		struct {			/* needed for BNO, CNT, INO */
-			struct xfs_buf	*agbp;	/* agf/agi buffer pointer */
-			xfs_agnumber_t	agno;	/* ag number */
-			union xfs_btree_cur_private	priv;
-		} a;
-		struct {			/* needed for BMAP */
-			struct xfs_inode *ip;	/* pointer to our inode */
-			int		allocated;	/* count of alloced */
-			short		forksize;	/* fork's inode space */
-			char		whichfork;	/* data or attr fork */
-			char		flags;		/* flags */
-#define	XFS_BTCUR_BPRV_WASDEL		(1<<0)		/* was delayed */
-#define	XFS_BTCUR_BPRV_INVALID_OWNER	(1<<1)		/* for ext swap */
-		} b;
-	}		bc_private;	/* per-btree type data */
+		struct xfs_btree_cur_ag	bc_ag;
+		struct xfs_btree_cur_ino bc_ino;
+	};
 } xfs_btree_cur_t;
 
 /* cursor flags */
@@ -229,6 +243,12 @@ typedef struct xfs_btree_cur
 #define XFS_BTREE_LASTREC_UPDATE	(1<<2)	/* track last rec externally */
 #define XFS_BTREE_CRC_BLOCKS		(1<<3)	/* uses extended btree blocks */
 #define XFS_BTREE_OVERLAPPING		(1<<4)	/* overlapping intervals */
+/*
+ * The root of this btree is a fakeroot structure so that we can stage a btree
+ * rebuild without leaving it accessible via primary metadata.  The ops struct
+ * is dynamically allocated and must be freed when the cursor is deleted.
+ */
+#define XFS_BTREE_STAGING		(1<<5)
 
 
 #define	XFS_BTREE_NOERROR	0
@@ -294,35 +314,6 @@ xfs_btree_dup_cursor(
 	xfs_btree_cur_t		**ncur);/* output cursor */
 
 /*
- * Get a buffer for the block, return it with no data read.
- * Long-form addressing.
- */
-struct xfs_buf *				/* buffer for fsbno */
-xfs_btree_get_bufl(
-	struct xfs_mount	*mp,	/* file system mount point */
-	struct xfs_trans	*tp,	/* transaction pointer */
-	xfs_fsblock_t		fsbno);	/* file system block number */
-
-/*
- * Get a buffer for the block, return it with no data read.
- * Short-form addressing.
- */
-struct xfs_buf *				/* buffer for agno/agbno */
-xfs_btree_get_bufs(
-	struct xfs_mount	*mp,	/* file system mount point */
-	struct xfs_trans	*tp,	/* transaction pointer */
-	xfs_agnumber_t		agno,	/* allocation group number */
-	xfs_agblock_t		agbno);	/* allocation group block number */
-
-/*
- * Check for the cursor referring to the last block at the given level.
- */
-int					/* 1=is last block, 0=not last block */
-xfs_btree_islastblock(
-	xfs_btree_cur_t		*cur,	/* btree cursor */
-	int			level);	/* level to check */
-
-/*
  * Compute first and last byte offsets for the fields given.
  * Interprets the offsets table, which contains struct field offsets.
  */
@@ -482,8 +473,15 @@ int xfs_btree_query_all(struct xfs_btree_cur *cur, xfs_btree_query_range_fn fn,
 
 typedef int (*xfs_btree_visit_blocks_fn)(struct xfs_btree_cur *cur, int level,
 		void *data);
+/* Visit record blocks. */
+#define XFS_BTREE_VISIT_RECORDS		(1 << 0)
+/* Visit leaf blocks. */
+#define XFS_BTREE_VISIT_LEAVES		(1 << 1)
+/* Visit all blocks. */
+#define XFS_BTREE_VISIT_ALL		(XFS_BTREE_VISIT_RECORDS | \
+					 XFS_BTREE_VISIT_LEAVES)
 int xfs_btree_visit_blocks(struct xfs_btree_cur *cur,
-		xfs_btree_visit_blocks_fn fn, void *data);
+		xfs_btree_visit_blocks_fn fn, unsigned int flags, void *data);
 
 int xfs_btree_count_blocks(struct xfs_btree_cur *cur, xfs_extlen_t *blocks);
 
@@ -513,5 +511,39 @@ union xfs_btree_key *xfs_btree_high_key_from_key(struct xfs_btree_cur *cur,
 int xfs_btree_has_record(struct xfs_btree_cur *cur, union xfs_btree_irec *low,
 		union xfs_btree_irec *high, bool *exists);
 bool xfs_btree_has_more_records(struct xfs_btree_cur *cur);
+struct xfs_ifork *xfs_btree_ifork_ptr(struct xfs_btree_cur *cur);
+
+/* Does this cursor point to the last block in the given level? */
+static inline bool
+xfs_btree_islastblock(
+	xfs_btree_cur_t		*cur,
+	int			level)
+{
+	struct xfs_btree_block	*block;
+	struct xfs_buf		*bp;
+
+	block = xfs_btree_get_block(cur, level, &bp);
+	ASSERT(block && xfs_btree_check_block(cur, block, level, bp) == 0);
+
+	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
+		return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK);
+	return block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK);
+}
+
+void xfs_btree_set_ptr_null(struct xfs_btree_cur *cur,
+		union xfs_btree_ptr *ptr);
+int xfs_btree_get_buf_block(struct xfs_btree_cur *cur, union xfs_btree_ptr *ptr,
+		struct xfs_btree_block **block, struct xfs_buf **bpp);
+void xfs_btree_set_sibling(struct xfs_btree_cur *cur,
+		struct xfs_btree_block *block, union xfs_btree_ptr *ptr,
+		int lr);
+void xfs_btree_init_block_cur(struct xfs_btree_cur *cur,
+		struct xfs_buf *bp, int level, int numrecs);
+void xfs_btree_copy_ptrs(struct xfs_btree_cur *cur,
+		union xfs_btree_ptr *dst_ptr,
+		const union xfs_btree_ptr *src_ptr, int numptrs);
+void xfs_btree_copy_keys(struct xfs_btree_cur *cur,
+		union xfs_btree_key *dst_key, union xfs_btree_key *src_key,
+		int numkeys);
 
 #endif	/* __XFS_BTREE_H__ */
diff --git a/fs/xfs/libxfs/xfs_btree_staging.c b/fs/xfs/libxfs/xfs_btree_staging.c
new file mode 100644
index 000000000000..f464a7c7cf22
--- /dev/null
+++ b/fs/xfs/libxfs/xfs_btree_staging.c
@@ -0,0 +1,879 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2020 Oracle.  All Rights Reserved.
+ * Author: Darrick J. Wong <darrick.wong@oracle.com>
+ */
+#include "xfs.h"
+#include "xfs_fs.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
+#include "xfs_bit.h"
+#include "xfs_mount.h"
+#include "xfs_inode.h"
+#include "xfs_trans.h"
+#include "xfs_btree.h"
+#include "xfs_trace.h"
+#include "xfs_btree_staging.h"
+
+/*
+ * Staging Cursors and Fake Roots for Btrees
+ * =========================================
+ *
+ * A staging btree cursor is a special type of btree cursor that callers must
+ * use to construct a new btree index using the btree bulk loader code.  The
+ * bulk loading code uses the staging btree cursor to abstract the details of
+ * initializing new btree blocks and filling them with records or key/ptr
+ * pairs.  Regular btree operations (e.g. queries and modifications) are not
+ * supported with staging cursors, and callers must not invoke them.
+ *
+ * Fake root structures contain all the information about a btree that is under
+ * construction by the bulk loading code.  Staging btree cursors point to fake
+ * root structures instead of the usual AG header or inode structure.
+ *
+ * Callers are expected to initialize a fake root structure and pass it into
+ * the _stage_cursor function for a specific btree type.  When bulk loading is
+ * complete, callers should call the _commit_staged_btree function for that
+ * specific btree type to commit the new btree into the filesystem.
+ */
+
+/*
+ * Don't allow staging cursors to be duplicated because they're supposed to be
+ * kept private to a single thread.
+ */
+STATIC struct xfs_btree_cur *
+xfs_btree_fakeroot_dup_cursor(
+	struct xfs_btree_cur	*cur)
+{
+	ASSERT(0);
+	return NULL;
+}
+
+/*
+ * Don't allow block allocation for a staging cursor, because staging cursors
+ * do not support regular btree modifications.
+ *
+ * Bulk loading uses a separate callback to obtain new blocks from a
+ * preallocated list, which prevents ENOSPC failures during loading.
+ */
+STATIC int
+xfs_btree_fakeroot_alloc_block(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_ptr	*start_bno,
+	union xfs_btree_ptr	*new_bno,
+	int			*stat)
+{
+	ASSERT(0);
+	return -EFSCORRUPTED;
+}
+
+/*
+ * Don't allow block freeing for a staging cursor, because staging cursors
+ * do not support regular btree modifications.
+ */
+STATIC int
+xfs_btree_fakeroot_free_block(
+	struct xfs_btree_cur	*cur,
+	struct xfs_buf		*bp)
+{
+	ASSERT(0);
+	return -EFSCORRUPTED;
+}
+
+/* Initialize a pointer to the root block from the fakeroot. */
+STATIC void
+xfs_btree_fakeroot_init_ptr_from_cur(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_ptr	*ptr)
+{
+	struct xbtree_afakeroot	*afake;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	afake = cur->bc_ag.afake;
+	ptr->s = cpu_to_be32(afake->af_root);
+}
+
+/*
+ * Bulk Loading for AG Btrees
+ * ==========================
+ *
+ * For a btree rooted in an AG header, pass a xbtree_afakeroot structure to the
+ * staging cursor.  Callers should initialize this to zero.
+ *
+ * The _stage_cursor() function for a specific btree type should call
+ * xfs_btree_stage_afakeroot to set up the in-memory cursor as a staging
+ * cursor.  The corresponding _commit_staged_btree() function should log the
+ * new root and call xfs_btree_commit_afakeroot() to transform the staging
+ * cursor into a regular btree cursor.
+ */
+
+/* Update the btree root information for a per-AG fake root. */
+STATIC void
+xfs_btree_afakeroot_set_root(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_ptr	*ptr,
+	int			inc)
+{
+	struct xbtree_afakeroot	*afake = cur->bc_ag.afake;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+	afake->af_root = be32_to_cpu(ptr->s);
+	afake->af_levels += inc;
+}
+
+/*
+ * Initialize a AG-rooted btree cursor with the given AG btree fake root.
+ * The btree cursor's bc_ops will be overridden as needed to make the staging
+ * functionality work.
+ */
+void
+xfs_btree_stage_afakeroot(
+	struct xfs_btree_cur		*cur,
+	struct xbtree_afakeroot		*afake)
+{
+	struct xfs_btree_ops		*nops;
+
+	ASSERT(!(cur->bc_flags & XFS_BTREE_STAGING));
+	ASSERT(!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE));
+	ASSERT(cur->bc_tp == NULL);
+
+	nops = kmem_alloc(sizeof(struct xfs_btree_ops), KM_NOFS);
+	memcpy(nops, cur->bc_ops, sizeof(struct xfs_btree_ops));
+	nops->alloc_block = xfs_btree_fakeroot_alloc_block;
+	nops->free_block = xfs_btree_fakeroot_free_block;
+	nops->init_ptr_from_cur = xfs_btree_fakeroot_init_ptr_from_cur;
+	nops->set_root = xfs_btree_afakeroot_set_root;
+	nops->dup_cursor = xfs_btree_fakeroot_dup_cursor;
+
+	cur->bc_ag.afake = afake;
+	cur->bc_nlevels = afake->af_levels;
+	cur->bc_ops = nops;
+	cur->bc_flags |= XFS_BTREE_STAGING;
+}
+
+/*
+ * Transform an AG-rooted staging btree cursor back into a regular cursor by
+ * substituting a real btree root for the fake one and restoring normal btree
+ * cursor ops.  The caller must log the btree root change prior to calling
+ * this.
+ */
+void
+xfs_btree_commit_afakeroot(
+	struct xfs_btree_cur		*cur,
+	struct xfs_trans		*tp,
+	struct xfs_buf			*agbp,
+	const struct xfs_btree_ops	*ops)
+{
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+	ASSERT(cur->bc_tp == NULL);
+
+	trace_xfs_btree_commit_afakeroot(cur);
+
+	kmem_free((void *)cur->bc_ops);
+	cur->bc_ag.agbp = agbp;
+	cur->bc_ops = ops;
+	cur->bc_flags &= ~XFS_BTREE_STAGING;
+	cur->bc_tp = tp;
+}
+
+/*
+ * Bulk Loading for Inode-Rooted Btrees
+ * ====================================
+ *
+ * For a btree rooted in an inode fork, pass a xbtree_ifakeroot structure to
+ * the staging cursor.  This structure should be initialized as follows:
+ *
+ * - if_fork_size field should be set to the number of bytes available to the
+ *   fork in the inode.
+ *
+ * - if_fork should point to a freshly allocated struct xfs_ifork.
+ *
+ * - if_format should be set to the appropriate fork type (e.g.
+ *   XFS_DINODE_FMT_BTREE).
+ *
+ * All other fields must be zero.
+ *
+ * The _stage_cursor() function for a specific btree type should call
+ * xfs_btree_stage_ifakeroot to set up the in-memory cursor as a staging
+ * cursor.  The corresponding _commit_staged_btree() function should log the
+ * new root and call xfs_btree_commit_ifakeroot() to transform the staging
+ * cursor into a regular btree cursor.
+ */
+
+/*
+ * Initialize an inode-rooted btree cursor with the given inode btree fake
+ * root.  The btree cursor's bc_ops will be overridden as needed to make the
+ * staging functionality work.  If new_ops is not NULL, these new ops will be
+ * passed out to the caller for further overriding.
+ */
+void
+xfs_btree_stage_ifakeroot(
+	struct xfs_btree_cur		*cur,
+	struct xbtree_ifakeroot		*ifake,
+	struct xfs_btree_ops		**new_ops)
+{
+	struct xfs_btree_ops		*nops;
+
+	ASSERT(!(cur->bc_flags & XFS_BTREE_STAGING));
+	ASSERT(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE);
+	ASSERT(cur->bc_tp == NULL);
+
+	nops = kmem_alloc(sizeof(struct xfs_btree_ops), KM_NOFS);
+	memcpy(nops, cur->bc_ops, sizeof(struct xfs_btree_ops));
+	nops->alloc_block = xfs_btree_fakeroot_alloc_block;
+	nops->free_block = xfs_btree_fakeroot_free_block;
+	nops->init_ptr_from_cur = xfs_btree_fakeroot_init_ptr_from_cur;
+	nops->dup_cursor = xfs_btree_fakeroot_dup_cursor;
+
+	cur->bc_ino.ifake = ifake;
+	cur->bc_nlevels = ifake->if_levels;
+	cur->bc_ops = nops;
+	cur->bc_flags |= XFS_BTREE_STAGING;
+
+	if (new_ops)
+		*new_ops = nops;
+}
+
+/*
+ * Transform an inode-rooted staging btree cursor back into a regular cursor by
+ * substituting a real btree root for the fake one and restoring normal btree
+ * cursor ops.  The caller must log the btree root change prior to calling
+ * this.
+ */
+void
+xfs_btree_commit_ifakeroot(
+	struct xfs_btree_cur		*cur,
+	struct xfs_trans		*tp,
+	int				whichfork,
+	const struct xfs_btree_ops	*ops)
+{
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+	ASSERT(cur->bc_tp == NULL);
+
+	trace_xfs_btree_commit_ifakeroot(cur);
+
+	kmem_free((void *)cur->bc_ops);
+	cur->bc_ino.ifake = NULL;
+	cur->bc_ino.whichfork = whichfork;
+	cur->bc_ops = ops;
+	cur->bc_flags &= ~XFS_BTREE_STAGING;
+	cur->bc_tp = tp;
+}
+
+/*
+ * Bulk Loading of Staged Btrees
+ * =============================
+ *
+ * This interface is used with a staged btree cursor to create a totally new
+ * btree with a large number of records (i.e. more than what would fit in a
+ * single root block).  When the creation is complete, the new root can be
+ * linked atomically into the filesystem by committing the staged cursor.
+ *
+ * Creation of a new btree proceeds roughly as follows:
+ *
+ * The first step is to initialize an appropriate fake btree root structure and
+ * then construct a staged btree cursor.  Refer to the block comments about
+ * "Bulk Loading for AG Btrees" and "Bulk Loading for Inode-Rooted Btrees" for
+ * more information about how to do this.
+ *
+ * The second step is to initialize a struct xfs_btree_bload context as
+ * documented in the structure definition.
+ *
+ * The third step is to call xfs_btree_bload_compute_geometry to compute the
+ * height of and the number of blocks needed to construct the btree.  See the
+ * section "Computing the Geometry of the New Btree" for details about this
+ * computation.
+ *
+ * In step four, the caller must allocate xfs_btree_bload.nr_blocks blocks and
+ * save them for later use by ->claim_block().  Bulk loading requires all
+ * blocks to be allocated beforehand to avoid ENOSPC failures midway through a
+ * rebuild, and to minimize seek distances of the new btree.
+ *
+ * Step five is to call xfs_btree_bload() to start constructing the btree.
+ *
+ * The final step is to commit the staging btree cursor, which logs the new
+ * btree root and turns the staging cursor into a regular cursor.  The caller
+ * is responsible for cleaning up the previous btree blocks, if any.
+ *
+ * Computing the Geometry of the New Btree
+ * =======================================
+ *
+ * The number of items placed in each btree block is computed via the following
+ * algorithm: For leaf levels, the number of items for the level is nr_records
+ * in the bload structure.  For node levels, the number of items for the level
+ * is the number of blocks in the next lower level of the tree.  For each
+ * level, the desired number of items per block is defined as:
+ *
+ * desired = max(minrecs, maxrecs - slack factor)
+ *
+ * The number of blocks for the level is defined to be:
+ *
+ * blocks = floor(nr_items / desired)
+ *
+ * Note this is rounded down so that the npb calculation below will never fall
+ * below minrecs.  The number of items that will actually be loaded into each
+ * btree block is defined as:
+ *
+ * npb =  nr_items / blocks
+ *
+ * Some of the leftmost blocks in the level will contain one extra record as
+ * needed to handle uneven division.  If the number of records in any block
+ * would exceed maxrecs for that level, blocks is incremented and npb is
+ * recalculated.
+ *
+ * In other words, we compute the number of blocks needed to satisfy a given
+ * loading level, then spread the items as evenly as possible.
+ *
+ * The height and number of fs blocks required to create the btree are computed
+ * and returned via btree_height and nr_blocks.
+ */
+
+/*
+ * Put a btree block that we're loading onto the ordered list and release it.
+ * The btree blocks will be written to disk when bulk loading is finished.
+ */
+static void
+xfs_btree_bload_drop_buf(
+	struct list_head	*buffers_list,
+	struct xfs_buf		**bpp)
+{
+	if (*bpp == NULL)
+		return;
+
+	if (!xfs_buf_delwri_queue(*bpp, buffers_list))
+		ASSERT(0);
+
+	xfs_buf_relse(*bpp);
+	*bpp = NULL;
+}
+
+/*
+ * Allocate and initialize one btree block for bulk loading.
+ *
+ * The new btree block will have its level and numrecs fields set to the values
+ * of the level and nr_this_block parameters, respectively.
+ *
+ * The caller should ensure that ptrp, bpp, and blockp refer to the left
+ * sibling of the new block, if there is any.  On exit, ptrp, bpp, and blockp
+ * will all point to the new block.
+ */
+STATIC int
+xfs_btree_bload_prep_block(
+	struct xfs_btree_cur		*cur,
+	struct xfs_btree_bload		*bbl,
+	struct list_head		*buffers_list,
+	unsigned int			level,
+	unsigned int			nr_this_block,
+	union xfs_btree_ptr		*ptrp, /* in/out */
+	struct xfs_buf			**bpp, /* in/out */
+	struct xfs_btree_block		**blockp, /* in/out */
+	void				*priv)
+{
+	union xfs_btree_ptr		new_ptr;
+	struct xfs_buf			*new_bp;
+	struct xfs_btree_block		*new_block;
+	int				ret;
+
+	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
+	    level == cur->bc_nlevels - 1) {
+		struct xfs_ifork	*ifp = xfs_btree_ifork_ptr(cur);
+		size_t			new_size;
+
+		ASSERT(*bpp == NULL);
+
+		/* Allocate a new incore btree root block. */
+		new_size = bbl->iroot_size(cur, nr_this_block, priv);
+		ifp->if_broot = kmem_zalloc(new_size, 0);
+		ifp->if_broot_bytes = (int)new_size;
+		ifp->if_flags |= XFS_IFBROOT;
+
+		/* Initialize it and send it out. */
+		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
+				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
+				nr_this_block, cur->bc_ino.ip->i_ino,
+				cur->bc_flags);
+
+		*bpp = NULL;
+		*blockp = ifp->if_broot;
+		xfs_btree_set_ptr_null(cur, ptrp);
+		return 0;
+	}
+
+	/* Claim one of the caller's preallocated blocks. */
+	xfs_btree_set_ptr_null(cur, &new_ptr);
+	ret = bbl->claim_block(cur, &new_ptr, priv);
+	if (ret)
+		return ret;
+
+	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
+
+	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
+	if (ret)
+		return ret;
+
+	/*
+	 * The previous block (if any) is the left sibling of the new block,
+	 * so set its right sibling pointer to the new block and drop it.
+	 */
+	if (*blockp)
+		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
+	xfs_btree_bload_drop_buf(buffers_list, bpp);
+
+	/* Initialize the new btree block. */
+	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
+	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
+
+	/* Set the out parameters. */
+	*bpp = new_bp;
+	*blockp = new_block;
+	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
+	return 0;
+}
+
+/* Load one leaf block. */
+STATIC int
+xfs_btree_bload_leaf(
+	struct xfs_btree_cur		*cur,
+	unsigned int			recs_this_block,
+	xfs_btree_bload_get_record_fn	get_record,
+	struct xfs_btree_block		*block,
+	void				*priv)
+{
+	unsigned int			j;
+	int				ret;
+
+	/* Fill the leaf block with records. */
+	for (j = 1; j <= recs_this_block; j++) {
+		union xfs_btree_rec	*block_rec;
+
+		ret = get_record(cur, priv);
+		if (ret)
+			return ret;
+		block_rec = xfs_btree_rec_addr(cur, j, block);
+		cur->bc_ops->init_rec_from_cur(cur, block_rec);
+	}
+
+	return 0;
+}
+
+/*
+ * Load one node block with key/ptr pairs.
+ *
+ * child_ptr must point to a block within the next level down in the tree.  A
+ * key/ptr entry will be created in the new node block to the block pointed to
+ * by child_ptr.  On exit, child_ptr points to the next block on the child
+ * level that needs processing.
+ */
+STATIC int
+xfs_btree_bload_node(
+	struct xfs_btree_cur	*cur,
+	unsigned int		recs_this_block,
+	union xfs_btree_ptr	*child_ptr,
+	struct xfs_btree_block	*block)
+{
+	unsigned int		j;
+	int			ret;
+
+	/* Fill the node block with keys and pointers. */
+	for (j = 1; j <= recs_this_block; j++) {
+		union xfs_btree_key	child_key;
+		union xfs_btree_ptr	*block_ptr;
+		union xfs_btree_key	*block_key;
+		struct xfs_btree_block	*child_block;
+		struct xfs_buf		*child_bp;
+
+		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
+
+		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
+				&child_bp);
+		if (ret)
+			return ret;
+
+		block_ptr = xfs_btree_ptr_addr(cur, j, block);
+		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
+
+		block_key = xfs_btree_key_addr(cur, j, block);
+		xfs_btree_get_keys(cur, child_block, &child_key);
+		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
+
+		xfs_btree_get_sibling(cur, child_block, child_ptr,
+				XFS_BB_RIGHTSIB);
+		xfs_buf_relse(child_bp);
+	}
+
+	return 0;
+}
+
+/*
+ * Compute the maximum number of records (or keyptrs) per block that we want to
+ * install at this level in the btree.  Caller is responsible for having set
+ * @cur->bc_ino.forksize to the desired fork size, if appropriate.
+ */
+STATIC unsigned int
+xfs_btree_bload_max_npb(
+	struct xfs_btree_cur	*cur,
+	struct xfs_btree_bload	*bbl,
+	unsigned int		level)
+{
+	unsigned int		ret;
+
+	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
+		return cur->bc_ops->get_dmaxrecs(cur, level);
+
+	ret = cur->bc_ops->get_maxrecs(cur, level);
+	if (level == 0)
+		ret -= bbl->leaf_slack;
+	else
+		ret -= bbl->node_slack;
+	return ret;
+}
+
+/*
+ * Compute the desired number of records (or keyptrs) per block that we want to
+ * install at this level in the btree, which must be somewhere between minrecs
+ * and max_npb.  The caller is free to install fewer records per block.
+ */
+STATIC unsigned int
+xfs_btree_bload_desired_npb(
+	struct xfs_btree_cur	*cur,
+	struct xfs_btree_bload	*bbl,
+	unsigned int		level)
+{
+	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
+
+	/* Root blocks are not subject to minrecs rules. */
+	if (level == cur->bc_nlevels - 1)
+		return max(1U, npb);
+
+	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
+}
+
+/*
+ * Compute the number of records to be stored in each block at this level and
+ * the number of blocks for this level.  For leaf levels, we must populate an
+ * empty root block even if there are no records, so we have to have at least
+ * one block.
+ */
+STATIC void
+xfs_btree_bload_level_geometry(
+	struct xfs_btree_cur	*cur,
+	struct xfs_btree_bload	*bbl,
+	unsigned int		level,
+	uint64_t		nr_this_level,
+	unsigned int		*avg_per_block,
+	uint64_t		*blocks,
+	uint64_t		*blocks_with_extra)
+{
+	uint64_t		npb;
+	uint64_t		dontcare;
+	unsigned int		desired_npb;
+	unsigned int		maxnr;
+
+	maxnr = cur->bc_ops->get_maxrecs(cur, level);
+
+	/*
+	 * Compute the number of blocks we need to fill each block with the
+	 * desired number of records/keyptrs per block.  Because desired_npb
+	 * could be minrecs, we use regular integer division (which rounds
+	 * the block count down) so that in the next step the effective # of
+	 * items per block will never be less than desired_npb.
+	 */
+	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
+	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
+	*blocks = max(1ULL, *blocks);
+
+	/*
+	 * Compute the number of records that we will actually put in each
+	 * block, assuming that we want to spread the records evenly between
+	 * the blocks.  Take care that the effective # of items per block (npb)
+	 * won't exceed maxrecs even for the blocks that get an extra record,
+	 * since desired_npb could be maxrecs, and in the previous step we
+	 * rounded the block count down.
+	 */
+	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
+	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
+		(*blocks)++;
+		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
+	}
+
+	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
+
+	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
+			*avg_per_block, desired_npb, *blocks,
+			*blocks_with_extra);
+}
+
+/*
+ * Ensure a slack value is appropriate for the btree.
+ *
+ * If the slack value is negative, set slack so that we fill the block to
+ * halfway between minrecs and maxrecs.  Make sure the slack is never so large
+ * that we can underflow minrecs.
+ */
+static void
+xfs_btree_bload_ensure_slack(
+	struct xfs_btree_cur	*cur,
+	int			*slack,
+	int			level)
+{
+	int			maxr;
+	int			minr;
+
+	maxr = cur->bc_ops->get_maxrecs(cur, level);
+	minr = cur->bc_ops->get_minrecs(cur, level);
+
+	/*
+	 * If slack is negative, automatically set slack so that we load the
+	 * btree block approximately halfway between minrecs and maxrecs.
+	 * Generally, this will net us 75% loading.
+	 */
+	if (*slack < 0)
+		*slack = maxr - ((maxr + minr) >> 1);
+
+	*slack = min(*slack, maxr - minr);
+}
+
+/*
+ * Prepare a btree cursor for a bulk load operation by computing the geometry
+ * fields in bbl.  Caller must ensure that the btree cursor is a staging
+ * cursor.  This function can be called multiple times.
+ */
+int
+xfs_btree_bload_compute_geometry(
+	struct xfs_btree_cur	*cur,
+	struct xfs_btree_bload	*bbl,
+	uint64_t		nr_records)
+{
+	uint64_t		nr_blocks = 0;
+	uint64_t		nr_this_level;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	/*
+	 * Make sure that the slack values make sense for traditional leaf and
+	 * node blocks.  Inode-rooted btrees will return different minrecs and
+	 * maxrecs values for the root block (bc_nlevels == level - 1).  We're
+	 * checking levels 0 and 1 here, so set bc_nlevels such that the btree
+	 * code doesn't interpret either as the root level.
+	 */
+	cur->bc_nlevels = XFS_BTREE_MAXLEVELS - 1;
+	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
+	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
+
+	bbl->nr_records = nr_this_level = nr_records;
+	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
+		uint64_t	level_blocks;
+		uint64_t	dontcare64;
+		unsigned int	level = cur->bc_nlevels - 1;
+		unsigned int	avg_per_block;
+
+		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
+				&avg_per_block, &level_blocks, &dontcare64);
+
+		if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
+			/*
+			 * If all the items we want to store at this level
+			 * would fit in the inode root block, then we have our
+			 * btree root and are done.
+			 *
+			 * Note that bmap btrees forbid records in the root.
+			 */
+			if (level != 0 && nr_this_level <= avg_per_block) {
+				nr_blocks++;
+				break;
+			}
+
+			/*
+			 * Otherwise, we have to store all the items for this
+			 * level in traditional btree blocks and therefore need
+			 * another level of btree to point to those blocks.
+			 *
+			 * We have to re-compute the geometry for each level of
+			 * an inode-rooted btree because the geometry differs
+			 * between a btree root in an inode fork and a
+			 * traditional btree block.
+			 *
+			 * This distinction is made in the btree code based on
+			 * whether level == bc_nlevels - 1.  Based on the
+			 * previous root block size check against the root
+			 * block geometry, we know that we aren't yet ready to
+			 * populate the root.  Increment bc_nevels and
+			 * recalculate the geometry for a traditional
+			 * block-based btree level.
+			 */
+			cur->bc_nlevels++;
+			xfs_btree_bload_level_geometry(cur, bbl, level,
+					nr_this_level, &avg_per_block,
+					&level_blocks, &dontcare64);
+		} else {
+			/*
+			 * If all the items we want to store at this level
+			 * would fit in a single root block, we're done.
+			 */
+			if (nr_this_level <= avg_per_block) {
+				nr_blocks++;
+				break;
+			}
+
+			/* Otherwise, we need another level of btree. */
+			cur->bc_nlevels++;
+		}
+
+		nr_blocks += level_blocks;
+		nr_this_level = level_blocks;
+	}
+
+	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
+		return -EOVERFLOW;
+
+	bbl->btree_height = cur->bc_nlevels;
+	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
+		bbl->nr_blocks = nr_blocks - 1;
+	else
+		bbl->nr_blocks = nr_blocks;
+	return 0;
+}
+
+/* Bulk load a btree given the parameters and geometry established in bbl. */
+int
+xfs_btree_bload(
+	struct xfs_btree_cur		*cur,
+	struct xfs_btree_bload		*bbl,
+	void				*priv)
+{
+	struct list_head		buffers_list;
+	union xfs_btree_ptr		child_ptr;
+	union xfs_btree_ptr		ptr;
+	struct xfs_buf			*bp = NULL;
+	struct xfs_btree_block		*block = NULL;
+	uint64_t			nr_this_level = bbl->nr_records;
+	uint64_t			blocks;
+	uint64_t			i;
+	uint64_t			blocks_with_extra;
+	uint64_t			total_blocks = 0;
+	unsigned int			avg_per_block;
+	unsigned int			level = 0;
+	int				ret;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	INIT_LIST_HEAD(&buffers_list);
+	cur->bc_nlevels = bbl->btree_height;
+	xfs_btree_set_ptr_null(cur, &child_ptr);
+	xfs_btree_set_ptr_null(cur, &ptr);
+
+	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
+			&avg_per_block, &blocks, &blocks_with_extra);
+
+	/* Load each leaf block. */
+	for (i = 0; i < blocks; i++) {
+		unsigned int		nr_this_block = avg_per_block;
+
+		/*
+		 * Due to rounding, btree blocks will not be evenly populated
+		 * in most cases.  blocks_with_extra tells us how many blocks
+		 * will receive an extra record to distribute the excess across
+		 * the current level as evenly as possible.
+		 */
+		if (i < blocks_with_extra)
+			nr_this_block++;
+
+		ret = xfs_btree_bload_prep_block(cur, bbl, &buffers_list, level,
+				nr_this_block, &ptr, &bp, &block, priv);
+		if (ret)
+			goto out;
+
+		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
+				nr_this_block);
+
+		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_record,
+				block, priv);
+		if (ret)
+			goto out;
+
+		/*
+		 * Record the leftmost leaf pointer so we know where to start
+		 * with the first node level.
+		 */
+		if (i == 0)
+			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
+	}
+	total_blocks += blocks;
+	xfs_btree_bload_drop_buf(&buffers_list, &bp);
+
+	/* Populate the internal btree nodes. */
+	for (level = 1; level < cur->bc_nlevels; level++) {
+		union xfs_btree_ptr	first_ptr;
+
+		nr_this_level = blocks;
+		block = NULL;
+		xfs_btree_set_ptr_null(cur, &ptr);
+
+		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
+				&avg_per_block, &blocks, &blocks_with_extra);
+
+		/* Load each node block. */
+		for (i = 0; i < blocks; i++) {
+			unsigned int	nr_this_block = avg_per_block;
+
+			if (i < blocks_with_extra)
+				nr_this_block++;
+
+			ret = xfs_btree_bload_prep_block(cur, bbl,
+					&buffers_list, level, nr_this_block,
+					&ptr, &bp, &block, priv);
+			if (ret)
+				goto out;
+
+			trace_xfs_btree_bload_block(cur, level, i, blocks,
+					&ptr, nr_this_block);
+
+			ret = xfs_btree_bload_node(cur, nr_this_block,
+					&child_ptr, block);
+			if (ret)
+				goto out;
+
+			/*
+			 * Record the leftmost node pointer so that we know
+			 * where to start the next node level above this one.
+			 */
+			if (i == 0)
+				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
+		}
+		total_blocks += blocks;
+		xfs_btree_bload_drop_buf(&buffers_list, &bp);
+		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
+	}
+
+	/* Initialize the new root. */
+	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
+		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
+		cur->bc_ino.ifake->if_levels = cur->bc_nlevels;
+		cur->bc_ino.ifake->if_blocks = total_blocks - 1;
+	} else {
+		cur->bc_ag.afake->af_root = be32_to_cpu(ptr.s);
+		cur->bc_ag.afake->af_levels = cur->bc_nlevels;
+		cur->bc_ag.afake->af_blocks = total_blocks;
+	}
+
+	/*
+	 * Write the new blocks to disk.  If the ordered list isn't empty after
+	 * that, then something went wrong and we have to fail.  This should
+	 * never happen, but we'll check anyway.
+	 */
+	ret = xfs_buf_delwri_submit(&buffers_list);
+	if (ret)
+		goto out;
+	if (!list_empty(&buffers_list)) {
+		ASSERT(list_empty(&buffers_list));
+		ret = -EIO;
+	}
+
+out:
+	xfs_buf_delwri_cancel(&buffers_list);
+	if (bp)
+		xfs_buf_relse(bp);
+	return ret;
+}
diff --git a/fs/xfs/libxfs/xfs_btree_staging.h b/fs/xfs/libxfs/xfs_btree_staging.h
new file mode 100644
index 000000000000..643f0f9b2994
--- /dev/null
+++ b/fs/xfs/libxfs/xfs_btree_staging.h
@@ -0,0 +1,123 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2020 Oracle.  All Rights Reserved.
+ * Author: Darrick J. Wong <darrick.wong@oracle.com>
+ */
+#ifndef __XFS_BTREE_STAGING_H__
+#define __XFS_BTREE_STAGING_H__
+
+/* Fake root for an AG-rooted btree. */
+struct xbtree_afakeroot {
+	/* AG block number of the new btree root. */
+	xfs_agblock_t		af_root;
+
+	/* Height of the new btree. */
+	unsigned int		af_levels;
+
+	/* Number of blocks used by the btree. */
+	unsigned int		af_blocks;
+};
+
+/* Cursor interactions with with fake roots for AG-rooted btrees. */
+void xfs_btree_stage_afakeroot(struct xfs_btree_cur *cur,
+		struct xbtree_afakeroot *afake);
+void xfs_btree_commit_afakeroot(struct xfs_btree_cur *cur, struct xfs_trans *tp,
+		struct xfs_buf *agbp, const struct xfs_btree_ops *ops);
+
+/* Fake root for an inode-rooted btree. */
+struct xbtree_ifakeroot {
+	/* Fake inode fork. */
+	struct xfs_ifork	*if_fork;
+
+	/* Number of blocks used by the btree. */
+	int64_t			if_blocks;
+
+	/* Height of the new btree. */
+	unsigned int		if_levels;
+
+	/* Number of bytes available for this fork in the inode. */
+	unsigned int		if_fork_size;
+
+	/* Fork format. */
+	unsigned int		if_format;
+
+	/* Number of records. */
+	unsigned int		if_extents;
+};
+
+/* Cursor interactions with with fake roots for inode-rooted btrees. */
+void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
+		struct xbtree_ifakeroot *ifake,
+		struct xfs_btree_ops **new_ops);
+void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, struct xfs_trans *tp,
+		int whichfork, const struct xfs_btree_ops *ops);
+
+/* Bulk loading of staged btrees. */
+typedef int (*xfs_btree_bload_get_record_fn)(struct xfs_btree_cur *cur, void *priv);
+typedef int (*xfs_btree_bload_claim_block_fn)(struct xfs_btree_cur *cur,
+		union xfs_btree_ptr *ptr, void *priv);
+typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
+		unsigned int nr_this_level, void *priv);
+
+struct xfs_btree_bload {
+	/*
+	 * This function will be called nr_records times to load records into
+	 * the btree.  The function does this by setting the cursor's bc_rec
+	 * field in in-core format.  Records must be returned in sort order.
+	 */
+	xfs_btree_bload_get_record_fn	get_record;
+
+	/*
+	 * This function will be called nr_blocks times to obtain a pointer
+	 * to a new btree block on disk.  Callers must preallocate all space
+	 * for the new btree before calling xfs_btree_bload, and this function
+	 * is what claims that reservation.
+	 */
+	xfs_btree_bload_claim_block_fn	claim_block;
+
+	/*
+	 * This function should return the size of the in-core btree root
+	 * block.  It is only necessary for XFS_BTREE_ROOT_IN_INODE btree
+	 * types.
+	 */
+	xfs_btree_bload_iroot_size_fn	iroot_size;
+
+	/*
+	 * The caller should set this to the number of records that will be
+	 * stored in the new btree.
+	 */
+	uint64_t			nr_records;
+
+	/*
+	 * Number of free records to leave in each leaf block.  If the caller
+	 * sets this to -1, the slack value will be calculated to be be halfway
+	 * between maxrecs and minrecs.  This typically leaves the block 75%
+	 * full.  Note that slack values are not enforced on inode root blocks.
+	 */
+	int				leaf_slack;
+
+	/*
+	 * Number of free key/ptrs pairs to leave in each node block.  This
+	 * field has the same semantics as leaf_slack.
+	 */
+	int				node_slack;
+
+	/*
+	 * The xfs_btree_bload_compute_geometry function will set this to the
+	 * number of btree blocks needed to store nr_records records.
+	 */
+	uint64_t			nr_blocks;
+
+	/*
+	 * The xfs_btree_bload_compute_geometry function will set this to the
+	 * height of the new btree.
+	 */
+	unsigned int			btree_height;
+};
+
+int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
+		struct xfs_btree_bload *bbl, uint64_t nr_records);
+int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
+		void *priv);
+
+#endif	/* __XFS_BTREE_STAGING_H__ */
diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c
index 4fd1223c1bd5..897749c41f36 100644
--- a/fs/xfs/libxfs/xfs_da_btree.c
+++ b/fs/xfs/libxfs/xfs_da_btree.c
@@ -12,9 +12,9 @@
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
 #include "xfs_mount.h"
+#include "xfs_inode.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
-#include "xfs_inode.h"
 #include "xfs_trans.h"
 #include "xfs_bmap.h"
 #include "xfs_attr_leaf.h"
@@ -107,7 +107,66 @@ xfs_da_state_free(xfs_da_state_t *state)
 #ifdef DEBUG
 	memset((char *)state, 0, sizeof(*state));
 #endif /* DEBUG */
-	kmem_zone_free(xfs_da_state_zone, state);
+	kmem_cache_free(xfs_da_state_zone, state);
+}
+
+static inline int xfs_dabuf_nfsb(struct xfs_mount *mp, int whichfork)
+{
+	if (whichfork == XFS_DATA_FORK)
+		return mp->m_dir_geo->fsbcount;
+	return mp->m_attr_geo->fsbcount;
+}
+
+void
+xfs_da3_node_hdr_from_disk(
+	struct xfs_mount		*mp,
+	struct xfs_da3_icnode_hdr	*to,
+	struct xfs_da_intnode		*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_da3_intnode	*from3 = (struct xfs_da3_intnode *)from;
+
+		to->forw = be32_to_cpu(from3->hdr.info.hdr.forw);
+		to->back = be32_to_cpu(from3->hdr.info.hdr.back);
+		to->magic = be16_to_cpu(from3->hdr.info.hdr.magic);
+		to->count = be16_to_cpu(from3->hdr.__count);
+		to->level = be16_to_cpu(from3->hdr.__level);
+		to->btree = from3->__btree;
+		ASSERT(to->magic == XFS_DA3_NODE_MAGIC);
+	} else {
+		to->forw = be32_to_cpu(from->hdr.info.forw);
+		to->back = be32_to_cpu(from->hdr.info.back);
+		to->magic = be16_to_cpu(from->hdr.info.magic);
+		to->count = be16_to_cpu(from->hdr.__count);
+		to->level = be16_to_cpu(from->hdr.__level);
+		to->btree = from->__btree;
+		ASSERT(to->magic == XFS_DA_NODE_MAGIC);
+	}
+}
+
+void
+xfs_da3_node_hdr_to_disk(
+	struct xfs_mount		*mp,
+	struct xfs_da_intnode		*to,
+	struct xfs_da3_icnode_hdr	*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_da3_intnode	*to3 = (struct xfs_da3_intnode *)to;
+
+		ASSERT(from->magic == XFS_DA3_NODE_MAGIC);
+		to3->hdr.info.hdr.forw = cpu_to_be32(from->forw);
+		to3->hdr.info.hdr.back = cpu_to_be32(from->back);
+		to3->hdr.info.hdr.magic = cpu_to_be16(from->magic);
+		to3->hdr.__count = cpu_to_be16(from->count);
+		to3->hdr.__level = cpu_to_be16(from->level);
+	} else {
+		ASSERT(from->magic == XFS_DA_NODE_MAGIC);
+		to->hdr.info.forw = cpu_to_be32(from->forw);
+		to->hdr.info.back = cpu_to_be32(from->back);
+		to->hdr.info.magic = cpu_to_be16(from->magic);
+		to->hdr.__count = cpu_to_be16(from->count);
+		to->hdr.__level = cpu_to_be16(from->level);
+	}
 }
 
 /*
@@ -145,12 +204,9 @@ xfs_da3_node_verify(
 	struct xfs_mount	*mp = bp->b_mount;
 	struct xfs_da_intnode	*hdr = bp->b_addr;
 	struct xfs_da3_icnode_hdr ichdr;
-	const struct xfs_dir_ops *ops;
 	xfs_failaddr_t		fa;
 
-	ops = xfs_dir_get_ops(mp, NULL);
-
-	ops->node_hdr_from_disk(&ichdr, hdr);
+	xfs_da3_node_hdr_from_disk(mp, &ichdr, hdr);
 
 	fa = xfs_da3_blkinfo_verify(bp, bp->b_addr);
 	if (fa)
@@ -275,46 +331,76 @@ const struct xfs_buf_ops xfs_da3_node_buf_ops = {
 	.verify_struct = xfs_da3_node_verify_struct,
 };
 
+static int
+xfs_da3_node_set_type(
+	struct xfs_trans	*tp,
+	struct xfs_buf		*bp)
+{
+	struct xfs_da_blkinfo	*info = bp->b_addr;
+
+	switch (be16_to_cpu(info->magic)) {
+	case XFS_DA_NODE_MAGIC:
+	case XFS_DA3_NODE_MAGIC:
+		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DA_NODE_BUF);
+		return 0;
+	case XFS_ATTR_LEAF_MAGIC:
+	case XFS_ATTR3_LEAF_MAGIC:
+		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_ATTR_LEAF_BUF);
+		return 0;
+	case XFS_DIR2_LEAFN_MAGIC:
+	case XFS_DIR3_LEAFN_MAGIC:
+		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_LEAFN_BUF);
+		return 0;
+	default:
+		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, tp->t_mountp,
+				info, sizeof(*info));
+		xfs_trans_brelse(tp, bp);
+		return -EFSCORRUPTED;
+	}
+}
+
 int
 xfs_da3_node_read(
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
+	struct xfs_buf		**bpp,
+	int			whichfork)
+{
+	int			error;
+
+	error = xfs_da_read_buf(tp, dp, bno, 0, bpp, whichfork,
+			&xfs_da3_node_buf_ops);
+	if (error || !*bpp || !tp)
+		return error;
+	return xfs_da3_node_set_type(tp, *bpp);
+}
+
+int
+xfs_da3_node_read_mapped(
+	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp,
-	int			which_fork)
+	int			whichfork)
 {
-	int			err;
+	struct xfs_mount	*mp = dp->i_mount;
+	int			error;
 
-	err = xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
-					which_fork, &xfs_da3_node_buf_ops);
-	if (!err && tp && *bpp) {
-		struct xfs_da_blkinfo	*info = (*bpp)->b_addr;
-		int			type;
+	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, mappedbno,
+			XFS_FSB_TO_BB(mp, xfs_dabuf_nfsb(mp, whichfork)), 0,
+			bpp, &xfs_da3_node_buf_ops);
+	if (error || !*bpp)
+		return error;
 
-		switch (be16_to_cpu(info->magic)) {
-		case XFS_DA_NODE_MAGIC:
-		case XFS_DA3_NODE_MAGIC:
-			type = XFS_BLFT_DA_NODE_BUF;
-			break;
-		case XFS_ATTR_LEAF_MAGIC:
-		case XFS_ATTR3_LEAF_MAGIC:
-			type = XFS_BLFT_ATTR_LEAF_BUF;
-			break;
-		case XFS_DIR2_LEAFN_MAGIC:
-		case XFS_DIR3_LEAFN_MAGIC:
-			type = XFS_BLFT_DIR_LEAFN_BUF;
-			break;
-		default:
-			XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-					tp->t_mountp, info, sizeof(*info));
-			xfs_trans_brelse(tp, *bpp);
-			*bpp = NULL;
-			return -EFSCORRUPTED;
-		}
-		xfs_trans_buf_set_type(tp, *bpp, type);
-	}
-	return err;
+	if (whichfork == XFS_ATTR_FORK)
+		xfs_buf_set_ref(*bpp, XFS_ATTR_BTREE_REF);
+	else
+		xfs_buf_set_ref(*bpp, XFS_DIR_BTREE_REF);
+
+	if (!tp)
+		return 0;
+	return xfs_da3_node_set_type(tp, *bpp);
 }
 
 /*========================================================================
@@ -343,7 +429,7 @@ xfs_da3_node_create(
 	trace_xfs_da_node_create(args);
 	ASSERT(level <= XFS_DA_NODE_MAXDEPTH);
 
-	error = xfs_da_get_buf(tp, dp, blkno, -1, &bp, whichfork);
+	error = xfs_da_get_buf(tp, dp, blkno, &bp, whichfork);
 	if (error)
 		return error;
 	bp->b_ops = &xfs_da3_node_buf_ops;
@@ -363,9 +449,9 @@ xfs_da3_node_create(
 	}
 	ichdr.level = level;
 
-	dp->d_ops->node_hdr_to_disk(node, &ichdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &ichdr);
 	xfs_trans_log_buf(tp, bp,
-		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
+		XFS_DA_LOGRANGE(node, &node->hdr, args->geo->node_hdr_size));
 
 	*bpp = bp;
 	return 0;
@@ -504,6 +590,7 @@ xfs_da3_split(
 	node = oldblk->bp->b_addr;
 	if (node->hdr.info.forw) {
 		if (be32_to_cpu(node->hdr.info.forw) != addblk->blkno) {
+			xfs_buf_mark_corrupt(oldblk->bp);
 			error = -EFSCORRUPTED;
 			goto out;
 		}
@@ -516,6 +603,7 @@ xfs_da3_split(
 	node = oldblk->bp->b_addr;
 	if (node->hdr.info.back) {
 		if (be32_to_cpu(node->hdr.info.back) != addblk->blkno) {
+			xfs_buf_mark_corrupt(oldblk->bp);
 			error = -EFSCORRUPTED;
 			goto out;
 		}
@@ -568,7 +656,7 @@ xfs_da3_root_split(
 
 	dp = args->dp;
 	tp = args->trans;
-	error = xfs_da_get_buf(tp, dp, blkno, -1, &bp, args->whichfork);
+	error = xfs_da_get_buf(tp, dp, blkno, &bp, args->whichfork);
 	if (error)
 		return error;
 	node = bp->b_addr;
@@ -577,8 +665,8 @@ xfs_da3_root_split(
 	    oldroot->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) {
 		struct xfs_da3_icnode_hdr icnodehdr;
 
-		dp->d_ops->node_hdr_from_disk(&icnodehdr, oldroot);
-		btree = dp->d_ops->node_tree_p(oldroot);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &icnodehdr, oldroot);
+		btree = icnodehdr.btree;
 		size = (int)((char *)&btree[icnodehdr.count] - (char *)oldroot);
 		level = icnodehdr.level;
 
@@ -589,15 +677,14 @@ xfs_da3_root_split(
 		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DA_NODE_BUF);
 	} else {
 		struct xfs_dir3_icleaf_hdr leafhdr;
-		struct xfs_dir2_leaf_entry *ents;
 
 		leaf = (xfs_dir2_leaf_t *)oldroot;
-		dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-		ents = dp->d_ops->leaf_ents_p(leaf);
+		xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 
 		ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 		       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
-		size = (int)((char *)&ents[leafhdr.count] - (char *)leaf);
+		size = (int)((char *)&leafhdr.ents[leafhdr.count] -
+			(char *)leaf);
 		level = 0;
 
 		/*
@@ -637,14 +724,14 @@ xfs_da3_root_split(
 		return error;
 
 	node = bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-	btree = dp->d_ops->node_tree_p(node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
+	btree = nodehdr.btree;
 	btree[0].hashval = cpu_to_be32(blk1->hashval);
 	btree[0].before = cpu_to_be32(blk1->blkno);
 	btree[1].hashval = cpu_to_be32(blk2->hashval);
 	btree[1].before = cpu_to_be32(blk2->blkno);
 	nodehdr.count = 2;
-	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &nodehdr);
 
 #ifdef DEBUG
 	if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
@@ -686,7 +773,7 @@ xfs_da3_node_split(
 	trace_xfs_da_node_split(state->args);
 
 	node = oldblk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
 
 	/*
 	 * With V2 dirs the extra block is data or freespace.
@@ -733,7 +820,7 @@ xfs_da3_node_split(
 	 * If we had double-split op below us, then add the extra block too.
 	 */
 	node = oldblk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
 	if (oldblk->index <= nodehdr.count) {
 		oldblk->index++;
 		xfs_da3_node_add(state, oldblk, addblk);
@@ -788,10 +875,10 @@ xfs_da3_node_rebalance(
 
 	node1 = blk1->bp->b_addr;
 	node2 = blk2->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
-	dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
-	btree1 = dp->d_ops->node_tree_p(node1);
-	btree2 = dp->d_ops->node_tree_p(node2);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr1, node1);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr2, node2);
+	btree1 = nodehdr1.btree;
+	btree2 = nodehdr2.btree;
 
 	/*
 	 * Figure out how many entries need to move, and in which direction.
@@ -804,10 +891,10 @@ xfs_da3_node_rebalance(
 		tmpnode = node1;
 		node1 = node2;
 		node2 = tmpnode;
-		dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
-		dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
-		btree1 = dp->d_ops->node_tree_p(node1);
-		btree2 = dp->d_ops->node_tree_p(node2);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr1, node1);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr2, node2);
+		btree1 = nodehdr1.btree;
+		btree2 = nodehdr2.btree;
 		swap = 1;
 	}
 
@@ -869,14 +956,15 @@ xfs_da3_node_rebalance(
 	/*
 	 * Log header of node 1 and all current bits of node 2.
 	 */
-	dp->d_ops->node_hdr_to_disk(node1, &nodehdr1);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node1, &nodehdr1);
 	xfs_trans_log_buf(tp, blk1->bp,
-		XFS_DA_LOGRANGE(node1, &node1->hdr, dp->d_ops->node_hdr_size));
+		XFS_DA_LOGRANGE(node1, &node1->hdr,
+				state->args->geo->node_hdr_size));
 
-	dp->d_ops->node_hdr_to_disk(node2, &nodehdr2);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node2, &nodehdr2);
 	xfs_trans_log_buf(tp, blk2->bp,
 		XFS_DA_LOGRANGE(node2, &node2->hdr,
-				dp->d_ops->node_hdr_size +
+				state->args->geo->node_hdr_size +
 				(sizeof(btree2[0]) * nodehdr2.count)));
 
 	/*
@@ -886,10 +974,10 @@ xfs_da3_node_rebalance(
 	if (swap) {
 		node1 = blk1->bp->b_addr;
 		node2 = blk2->bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
-		dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
-		btree1 = dp->d_ops->node_tree_p(node1);
-		btree2 = dp->d_ops->node_tree_p(node2);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr1, node1);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr2, node2);
+		btree1 = nodehdr1.btree;
+		btree2 = nodehdr2.btree;
 	}
 	blk1->hashval = be32_to_cpu(btree1[nodehdr1.count - 1].hashval);
 	blk2->hashval = be32_to_cpu(btree2[nodehdr2.count - 1].hashval);
@@ -921,8 +1009,8 @@ xfs_da3_node_add(
 	trace_xfs_da_node_add(state->args);
 
 	node = oldblk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-	btree = dp->d_ops->node_tree_p(node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
+	btree = nodehdr.btree;
 
 	ASSERT(oldblk->index >= 0 && oldblk->index <= nodehdr.count);
 	ASSERT(newblk->blkno != 0);
@@ -945,9 +1033,10 @@ xfs_da3_node_add(
 				tmp + sizeof(*btree)));
 
 	nodehdr.count += 1;
-	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &nodehdr);
 	xfs_trans_log_buf(state->args->trans, oldblk->bp,
-		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
+		XFS_DA_LOGRANGE(node, &node->hdr,
+				state->args->geo->node_hdr_size));
 
 	/*
 	 * Copy the last hash value from the oldblk to propagate upwards.
@@ -1082,7 +1171,6 @@ xfs_da3_root_join(
 	xfs_dablk_t		child;
 	struct xfs_buf		*bp;
 	struct xfs_da3_icnode_hdr oldroothdr;
-	struct xfs_da_node_entry *btree;
 	int			error;
 	struct xfs_inode	*dp = state->args->dp;
 
@@ -1092,7 +1180,7 @@ xfs_da3_root_join(
 
 	args = state->args;
 	oldroot = root_blk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&oldroothdr, oldroot);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &oldroothdr, oldroot);
 	ASSERT(oldroothdr.forw == 0);
 	ASSERT(oldroothdr.back == 0);
 
@@ -1106,11 +1194,9 @@ xfs_da3_root_join(
 	 * Read in the (only) child block, then copy those bytes into
 	 * the root block's buffer and free the original child block.
 	 */
-	btree = dp->d_ops->node_tree_p(oldroot);
-	child = be32_to_cpu(btree[0].before);
+	child = be32_to_cpu(oldroothdr.btree[0].before);
 	ASSERT(child != 0);
-	error = xfs_da3_node_read(args->trans, dp, child, -1, &bp,
-					     args->whichfork);
+	error = xfs_da3_node_read(args->trans, dp, child, &bp, args->whichfork);
 	if (error)
 		return error;
 	xfs_da_blkinfo_onlychild_validate(bp->b_addr, oldroothdr.level);
@@ -1172,7 +1258,7 @@ xfs_da3_node_toosmall(
 	blk = &state->path.blk[ state->path.active-1 ];
 	info = blk->bp->b_addr;
 	node = (xfs_da_intnode_t *)info;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
 	if (nodehdr.count > (state->args->geo->node_ents >> 1)) {
 		*action = 0;	/* blk over 50%, don't try to join */
 		return 0;	/* blk over 50%, don't try to join */
@@ -1224,13 +1310,13 @@ xfs_da3_node_toosmall(
 			blkno = nodehdr.back;
 		if (blkno == 0)
 			continue;
-		error = xfs_da3_node_read(state->args->trans, dp,
-					blkno, -1, &bp, state->args->whichfork);
+		error = xfs_da3_node_read(state->args->trans, dp, blkno, &bp,
+				state->args->whichfork);
 		if (error)
 			return error;
 
 		node = bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&thdr, node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &thdr, node);
 		xfs_trans_brelse(state->args->trans, bp);
 
 		if (count - thdr.count >= 0)
@@ -1272,18 +1358,14 @@ xfs_da3_node_lasthash(
 	struct xfs_buf		*bp,
 	int			*count)
 {
-	struct xfs_da_intnode	 *node;
-	struct xfs_da_node_entry *btree;
 	struct xfs_da3_icnode_hdr nodehdr;
 
-	node = bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, bp->b_addr);
 	if (count)
 		*count = nodehdr.count;
 	if (!nodehdr.count)
 		return 0;
-	btree = dp->d_ops->node_tree_p(node);
-	return be32_to_cpu(btree[nodehdr.count - 1].hashval);
+	return be32_to_cpu(nodehdr.btree[nodehdr.count - 1].hashval);
 }
 
 /*
@@ -1328,8 +1410,8 @@ xfs_da3_fixhashpath(
 		struct xfs_da3_icnode_hdr nodehdr;
 
 		node = blk->bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-		btree = dp->d_ops->node_tree_p(node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
+		btree = nodehdr.btree;
 		if (be32_to_cpu(btree[blk->index].hashval) == lasthash)
 			break;
 		blk->hashval = lasthash;
@@ -1360,7 +1442,7 @@ xfs_da3_node_remove(
 	trace_xfs_da_node_remove(state->args);
 
 	node = drop_blk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
 	ASSERT(drop_blk->index < nodehdr.count);
 	ASSERT(drop_blk->index >= 0);
 
@@ -1368,7 +1450,7 @@ xfs_da3_node_remove(
 	 * Copy over the offending entry, or just zero it out.
 	 */
 	index = drop_blk->index;
-	btree = dp->d_ops->node_tree_p(node);
+	btree = nodehdr.btree;
 	if (index < nodehdr.count - 1) {
 		tmp  = nodehdr.count - index - 1;
 		tmp *= (uint)sizeof(xfs_da_node_entry_t);
@@ -1381,9 +1463,9 @@ xfs_da3_node_remove(
 	xfs_trans_log_buf(state->args->trans, drop_blk->bp,
 	    XFS_DA_LOGRANGE(node, &btree[index], sizeof(btree[index])));
 	nodehdr.count -= 1;
-	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, node, &nodehdr);
 	xfs_trans_log_buf(state->args->trans, drop_blk->bp,
-	    XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
+	    XFS_DA_LOGRANGE(node, &node->hdr, state->args->geo->node_hdr_size));
 
 	/*
 	 * Copy the last hash value from the block to propagate upwards.
@@ -1416,10 +1498,10 @@ xfs_da3_node_unbalance(
 
 	drop_node = drop_blk->bp->b_addr;
 	save_node = save_blk->bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&drop_hdr, drop_node);
-	dp->d_ops->node_hdr_from_disk(&save_hdr, save_node);
-	drop_btree = dp->d_ops->node_tree_p(drop_node);
-	save_btree = dp->d_ops->node_tree_p(save_node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &drop_hdr, drop_node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &save_hdr, save_node);
+	drop_btree = drop_hdr.btree;
+	save_btree = save_hdr.btree;
 	tp = state->args->trans;
 
 	/*
@@ -1453,10 +1535,10 @@ xfs_da3_node_unbalance(
 	memcpy(&save_btree[sindex], &drop_btree[0], tmp);
 	save_hdr.count += drop_hdr.count;
 
-	dp->d_ops->node_hdr_to_disk(save_node, &save_hdr);
+	xfs_da3_node_hdr_to_disk(dp->i_mount, save_node, &save_hdr);
 	xfs_trans_log_buf(tp, save_blk->bp,
 		XFS_DA_LOGRANGE(save_node, &save_node->hdr,
-				dp->d_ops->node_hdr_size));
+				state->args->geo->node_hdr_size));
 
 	/*
 	 * Save the last hashval in the remaining block for upward propagation.
@@ -1517,7 +1599,7 @@ xfs_da3_node_lookup_int(
 		 */
 		blk->blkno = blkno;
 		error = xfs_da3_node_read(args->trans, args->dp, blkno,
-					-1, &blk->bp, args->whichfork);
+					&blk->bp, args->whichfork);
 		if (error) {
 			blk->blkno = 0;
 			state->path.active--;
@@ -1541,8 +1623,10 @@ xfs_da3_node_lookup_int(
 			break;
 		}
 
-		if (magic != XFS_DA_NODE_MAGIC && magic != XFS_DA3_NODE_MAGIC)
+		if (magic != XFS_DA_NODE_MAGIC && magic != XFS_DA3_NODE_MAGIC) {
+			xfs_buf_mark_corrupt(blk->bp);
 			return -EFSCORRUPTED;
+		}
 
 		blk->magic = XFS_DA_NODE_MAGIC;
 
@@ -1550,19 +1634,22 @@ xfs_da3_node_lookup_int(
 		 * Search an intermediate node for a match.
 		 */
 		node = blk->bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-		btree = dp->d_ops->node_tree_p(node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr, node);
+		btree = nodehdr.btree;
 
 		/* Tree taller than we can handle; bail out! */
-		if (nodehdr.level >= XFS_DA_NODE_MAXDEPTH)
+		if (nodehdr.level >= XFS_DA_NODE_MAXDEPTH) {
+			xfs_buf_mark_corrupt(blk->bp);
 			return -EFSCORRUPTED;
+		}
 
 		/* Check the level from the root. */
 		if (blkno == args->geo->leafblk)
 			expected_level = nodehdr.level - 1;
-		else if (expected_level != nodehdr.level)
+		else if (expected_level != nodehdr.level) {
+			xfs_buf_mark_corrupt(blk->bp);
 			return -EFSCORRUPTED;
-		else
+		} else
 			expected_level--;
 
 		max = nodehdr.count;
@@ -1612,11 +1699,11 @@ xfs_da3_node_lookup_int(
 		}
 
 		/* We can't point back to the root. */
-		if (blkno == args->geo->leafblk)
+		if (XFS_IS_CORRUPT(dp->i_mount, blkno == args->geo->leafblk))
 			return -EFSCORRUPTED;
 	}
 
-	if (expected_level != 0)
+	if (XFS_IS_CORRUPT(dp->i_mount, expected_level != 0))
 		return -EFSCORRUPTED;
 
 	/*
@@ -1678,10 +1765,10 @@ xfs_da3_node_order(
 
 	node1 = node1_bp->b_addr;
 	node2 = node2_bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&node1hdr, node1);
-	dp->d_ops->node_hdr_from_disk(&node2hdr, node2);
-	btree1 = dp->d_ops->node_tree_p(node1);
-	btree2 = dp->d_ops->node_tree_p(node2);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &node1hdr, node1);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &node2hdr, node2);
+	btree1 = node1hdr.btree;
+	btree2 = node2hdr.btree;
 
 	if (node1hdr.count > 0 && node2hdr.count > 0 &&
 	    ((be32_to_cpu(btree2[0].hashval) < be32_to_cpu(btree1[0].hashval)) ||
@@ -1746,7 +1833,7 @@ xfs_da3_blk_link(
 		if (old_info->back) {
 			error = xfs_da3_node_read(args->trans, dp,
 						be32_to_cpu(old_info->back),
-						-1, &bp, args->whichfork);
+						&bp, args->whichfork);
 			if (error)
 				return error;
 			ASSERT(bp != NULL);
@@ -1767,7 +1854,7 @@ xfs_da3_blk_link(
 		if (old_info->forw) {
 			error = xfs_da3_node_read(args->trans, dp,
 						be32_to_cpu(old_info->forw),
-						-1, &bp, args->whichfork);
+						&bp, args->whichfork);
 			if (error)
 				return error;
 			ASSERT(bp != NULL);
@@ -1826,7 +1913,7 @@ xfs_da3_blk_unlink(
 		if (drop_info->back) {
 			error = xfs_da3_node_read(args->trans, args->dp,
 						be32_to_cpu(drop_info->back),
-						-1, &bp, args->whichfork);
+						&bp, args->whichfork);
 			if (error)
 				return error;
 			ASSERT(bp != NULL);
@@ -1843,7 +1930,7 @@ xfs_da3_blk_unlink(
 		if (drop_info->forw) {
 			error = xfs_da3_node_read(args->trans, args->dp,
 						be32_to_cpu(drop_info->forw),
-						-1, &bp, args->whichfork);
+						&bp, args->whichfork);
 			if (error)
 				return error;
 			ASSERT(bp != NULL);
@@ -1878,7 +1965,6 @@ xfs_da3_path_shift(
 {
 	struct xfs_da_state_blk	*blk;
 	struct xfs_da_blkinfo	*info;
-	struct xfs_da_intnode	*node;
 	struct xfs_da_args	*args;
 	struct xfs_da_node_entry *btree;
 	struct xfs_da3_icnode_hdr nodehdr;
@@ -1900,18 +1986,18 @@ xfs_da3_path_shift(
 	ASSERT(path != NULL);
 	ASSERT((path->active > 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
 	level = (path->active-1) - 1;	/* skip bottom layer in path */
-	for (blk = &path->blk[level]; level >= 0; blk--, level--) {
-		node = blk->bp->b_addr;
-		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-		btree = dp->d_ops->node_tree_p(node);
+	for (; level >= 0; level--) {
+		blk = &path->blk[level];
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr,
+					   blk->bp->b_addr);
 
 		if (forward && (blk->index < nodehdr.count - 1)) {
 			blk->index++;
-			blkno = be32_to_cpu(btree[blk->index].before);
+			blkno = be32_to_cpu(nodehdr.btree[blk->index].before);
 			break;
 		} else if (!forward && (blk->index > 0)) {
 			blk->index--;
-			blkno = be32_to_cpu(btree[blk->index].before);
+			blkno = be32_to_cpu(nodehdr.btree[blk->index].before);
 			break;
 		}
 	}
@@ -1929,7 +2015,7 @@ xfs_da3_path_shift(
 		/*
 		 * Read the next child block into a local buffer.
 		 */
-		error = xfs_da3_node_read(args->trans, dp, blkno, -1, &bp,
+		error = xfs_da3_node_read(args->trans, dp, blkno, &bp,
 					  args->whichfork);
 		if (error)
 			return error;
@@ -1962,9 +2048,9 @@ xfs_da3_path_shift(
 		case XFS_DA_NODE_MAGIC:
 		case XFS_DA3_NODE_MAGIC:
 			blk->magic = XFS_DA_NODE_MAGIC;
-			node = (xfs_da_intnode_t *)info;
-			dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-			btree = dp->d_ops->node_tree_p(node);
+			xfs_da3_node_hdr_from_disk(dp->i_mount, &nodehdr,
+						   bp->b_addr);
+			btree = nodehdr.btree;
 			blk->hashval = be32_to_cpu(btree[nodehdr.count - 1].hashval);
 			if (forward)
 				blk->index = 0;
@@ -2044,18 +2130,6 @@ xfs_da_compname(
 					XFS_CMP_EXACT : XFS_CMP_DIFFERENT;
 }
 
-static xfs_dahash_t
-xfs_default_hashname(
-	struct xfs_name	*name)
-{
-	return xfs_da_hashname(name->name, name->len);
-}
-
-const struct xfs_nameops xfs_default_nameops = {
-	.hashname	= xfs_default_hashname,
-	.compname	= xfs_da_compname
-};
-
 int
 xfs_da_grow_inode_int(
 	struct xfs_da_args	*args,
@@ -2213,16 +2287,13 @@ xfs_da3_swap_lastblock(
 	error = xfs_bmap_last_before(tp, dp, &lastoff, w);
 	if (error)
 		return error;
-	if (unlikely(lastoff == 0)) {
-		XFS_ERROR_REPORT("xfs_da_swap_lastblock(1)", XFS_ERRLEVEL_LOW,
-				 mp);
+	if (XFS_IS_CORRUPT(mp, lastoff == 0))
 		return -EFSCORRUPTED;
-	}
 	/*
 	 * Read the last block in the btree space.
 	 */
 	last_blkno = (xfs_dablk_t)lastoff - args->geo->fsbcount;
-	error = xfs_da3_node_read(tp, dp, last_blkno, -1, &last_buf, w);
+	error = xfs_da3_node_read(tp, dp, last_blkno, &last_buf, w);
 	if (error)
 		return error;
 	/*
@@ -2240,16 +2311,17 @@ xfs_da3_swap_lastblock(
 		struct xfs_dir2_leaf_entry *ents;
 
 		dead_leaf2 = (xfs_dir2_leaf_t *)dead_info;
-		dp->d_ops->leaf_hdr_from_disk(&leafhdr, dead_leaf2);
-		ents = dp->d_ops->leaf_ents_p(dead_leaf2);
+		xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr,
+					    dead_leaf2);
+		ents = leafhdr.ents;
 		dead_level = 0;
 		dead_hash = be32_to_cpu(ents[leafhdr.count - 1].hashval);
 	} else {
 		struct xfs_da3_icnode_hdr deadhdr;
 
 		dead_node = (xfs_da_intnode_t *)dead_info;
-		dp->d_ops->node_hdr_from_disk(&deadhdr, dead_node);
-		btree = dp->d_ops->node_tree_p(dead_node);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &deadhdr, dead_node);
+		btree = deadhdr.btree;
 		dead_level = deadhdr.level;
 		dead_hash = be32_to_cpu(btree[deadhdr.count - 1].hashval);
 	}
@@ -2258,15 +2330,13 @@ xfs_da3_swap_lastblock(
 	 * If the moved block has a left sibling, fix up the pointers.
 	 */
 	if ((sib_blkno = be32_to_cpu(dead_info->back))) {
-		error = xfs_da3_node_read(tp, dp, sib_blkno, -1, &sib_buf, w);
+		error = xfs_da3_node_read(tp, dp, sib_blkno, &sib_buf, w);
 		if (error)
 			goto done;
 		sib_info = sib_buf->b_addr;
-		if (unlikely(
-		    be32_to_cpu(sib_info->forw) != last_blkno ||
-		    sib_info->magic != dead_info->magic)) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(2)",
-					 XFS_ERRLEVEL_LOW, mp);
+		if (XFS_IS_CORRUPT(mp,
+				   be32_to_cpu(sib_info->forw) != last_blkno ||
+				   sib_info->magic != dead_info->magic)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
@@ -2280,15 +2350,13 @@ xfs_da3_swap_lastblock(
 	 * If the moved block has a right sibling, fix up the pointers.
 	 */
 	if ((sib_blkno = be32_to_cpu(dead_info->forw))) {
-		error = xfs_da3_node_read(tp, dp, sib_blkno, -1, &sib_buf, w);
+		error = xfs_da3_node_read(tp, dp, sib_blkno, &sib_buf, w);
 		if (error)
 			goto done;
 		sib_info = sib_buf->b_addr;
-		if (unlikely(
-		       be32_to_cpu(sib_info->back) != last_blkno ||
-		       sib_info->magic != dead_info->magic)) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(3)",
-					 XFS_ERRLEVEL_LOW, mp);
+		if (XFS_IS_CORRUPT(mp,
+				   be32_to_cpu(sib_info->back) != last_blkno ||
+				   sib_info->magic != dead_info->magic)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
@@ -2304,27 +2372,24 @@ xfs_da3_swap_lastblock(
 	 * Walk down the tree looking for the parent of the moved block.
 	 */
 	for (;;) {
-		error = xfs_da3_node_read(tp, dp, par_blkno, -1, &par_buf, w);
+		error = xfs_da3_node_read(tp, dp, par_blkno, &par_buf, w);
 		if (error)
 			goto done;
 		par_node = par_buf->b_addr;
-		dp->d_ops->node_hdr_from_disk(&par_hdr, par_node);
-		if (level >= 0 && level != par_hdr.level + 1) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(4)",
-					 XFS_ERRLEVEL_LOW, mp);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &par_hdr, par_node);
+		if (XFS_IS_CORRUPT(mp,
+				   level >= 0 && level != par_hdr.level + 1)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
 		level = par_hdr.level;
-		btree = dp->d_ops->node_tree_p(par_node);
+		btree = par_hdr.btree;
 		for (entno = 0;
 		     entno < par_hdr.count &&
 		     be32_to_cpu(btree[entno].hashval) < dead_hash;
 		     entno++)
 			continue;
-		if (entno == par_hdr.count) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(5)",
-					 XFS_ERRLEVEL_LOW, mp);
+		if (XFS_IS_CORRUPT(mp, entno == par_hdr.count)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
@@ -2349,24 +2414,20 @@ xfs_da3_swap_lastblock(
 		par_blkno = par_hdr.forw;
 		xfs_trans_brelse(tp, par_buf);
 		par_buf = NULL;
-		if (unlikely(par_blkno == 0)) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(6)",
-					 XFS_ERRLEVEL_LOW, mp);
+		if (XFS_IS_CORRUPT(mp, par_blkno == 0)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
-		error = xfs_da3_node_read(tp, dp, par_blkno, -1, &par_buf, w);
+		error = xfs_da3_node_read(tp, dp, par_blkno, &par_buf, w);
 		if (error)
 			goto done;
 		par_node = par_buf->b_addr;
-		dp->d_ops->node_hdr_from_disk(&par_hdr, par_node);
-		if (par_hdr.level != level) {
-			XFS_ERROR_REPORT("xfs_da_swap_lastblock(7)",
-					 XFS_ERRLEVEL_LOW, mp);
+		xfs_da3_node_hdr_from_disk(dp->i_mount, &par_hdr, par_node);
+		if (XFS_IS_CORRUPT(mp, par_hdr.level != level)) {
 			error = -EFSCORRUPTED;
 			goto done;
 		}
-		btree = dp->d_ops->node_tree_p(par_node);
+		btree = par_hdr.btree;
 		entno = 0;
 	}
 	/*
@@ -2429,159 +2490,86 @@ xfs_da_shrink_inode(
 	return error;
 }
 
-/*
- * See if the mapping(s) for this btree block are valid, i.e.
- * don't contain holes, are logically contiguous, and cover the whole range.
- */
-STATIC int
-xfs_da_map_covers_blocks(
-	int		nmap,
-	xfs_bmbt_irec_t	*mapp,
-	xfs_dablk_t	bno,
-	int		count)
-{
-	int		i;
-	xfs_fileoff_t	off;
-
-	for (i = 0, off = bno; i < nmap; i++) {
-		if (mapp[i].br_startblock == HOLESTARTBLOCK ||
-		    mapp[i].br_startblock == DELAYSTARTBLOCK) {
-			return 0;
-		}
-		if (off != mapp[i].br_startoff) {
-			return 0;
-		}
-		off += mapp[i].br_blockcount;
-	}
-	return off == bno + count;
-}
-
-/*
- * Convert a struct xfs_bmbt_irec to a struct xfs_buf_map.
- *
- * For the single map case, it is assumed that the caller has provided a pointer
- * to a valid xfs_buf_map.  For the multiple map case, this function will
- * allocate the xfs_buf_map to hold all the maps and replace the caller's single
- * map pointer with the allocated map.
- */
 static int
-xfs_buf_map_from_irec(
-	struct xfs_mount	*mp,
+xfs_dabuf_map(
+	struct xfs_inode	*dp,
+	xfs_dablk_t		bno,
+	unsigned int		flags,
+	int			whichfork,
 	struct xfs_buf_map	**mapp,
-	int			*nmaps,
-	struct xfs_bmbt_irec	*irecs,
-	int			nirecs)
+	int			*nmaps)
 {
-	struct xfs_buf_map	*map;
-	int			i;
-
-	ASSERT(*nmaps == 1);
-	ASSERT(nirecs >= 1);
+	struct xfs_mount	*mp = dp->i_mount;
+	int			nfsb = xfs_dabuf_nfsb(mp, whichfork);
+	struct xfs_bmbt_irec	irec, *irecs = &irec;
+	struct xfs_buf_map	*map = *mapp;
+	xfs_fileoff_t		off = bno;
+	int			error = 0, nirecs, i;
+
+	if (nfsb > 1)
+		irecs = kmem_zalloc(sizeof(irec) * nfsb, KM_NOFS);
+
+	nirecs = nfsb;
+	error = xfs_bmapi_read(dp, bno, nfsb, irecs, &nirecs,
+			xfs_bmapi_aflag(whichfork));
+	if (error)
+		goto out_free_irecs;
 
+	/*
+	 * Use the caller provided map for the single map case, else allocate a
+	 * larger one that needs to be free by the caller.
+	 */
 	if (nirecs > 1) {
-		map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map),
-				  KM_NOFS);
-		if (!map)
-			return -ENOMEM;
+		map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map), KM_NOFS);
+		if (!map) {
+			error = -ENOMEM;
+			goto out_free_irecs;
+		}
 		*mapp = map;
 	}
 
-	*nmaps = nirecs;
-	map = *mapp;
-	for (i = 0; i < *nmaps; i++) {
-		ASSERT(irecs[i].br_startblock != DELAYSTARTBLOCK &&
-		       irecs[i].br_startblock != HOLESTARTBLOCK);
+	for (i = 0; i < nirecs; i++) {
+		if (irecs[i].br_startblock == HOLESTARTBLOCK ||
+		    irecs[i].br_startblock == DELAYSTARTBLOCK)
+			goto invalid_mapping;
+		if (off != irecs[i].br_startoff)
+			goto invalid_mapping;
+
 		map[i].bm_bn = XFS_FSB_TO_DADDR(mp, irecs[i].br_startblock);
 		map[i].bm_len = XFS_FSB_TO_BB(mp, irecs[i].br_blockcount);
+		off += irecs[i].br_blockcount;
 	}
-	return 0;
-}
 
-/*
- * Map the block we are given ready for reading. There are three possible return
- * values:
- *	-1 - will be returned if we land in a hole and mappedbno == -2 so the
- *	     caller knows not to execute a subsequent read.
- *	 0 - if we mapped the block successfully
- *	>0 - positive error number if there was an error.
- */
-static int
-xfs_dabuf_map(
-	struct xfs_inode	*dp,
-	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
-	int			whichfork,
-	struct xfs_buf_map	**map,
-	int			*nmaps)
-{
-	struct xfs_mount	*mp = dp->i_mount;
-	int			nfsb;
-	int			error = 0;
-	struct xfs_bmbt_irec	irec;
-	struct xfs_bmbt_irec	*irecs = &irec;
-	int			nirecs;
+	if (off != bno + nfsb)
+		goto invalid_mapping;
 
-	ASSERT(map && *map);
-	ASSERT(*nmaps == 1);
-
-	if (whichfork == XFS_DATA_FORK)
-		nfsb = mp->m_dir_geo->fsbcount;
-	else
-		nfsb = mp->m_attr_geo->fsbcount;
-
-	/*
-	 * Caller doesn't have a mapping.  -2 means don't complain
-	 * if we land in a hole.
-	 */
-	if (mappedbno == -1 || mappedbno == -2) {
-		/*
-		 * Optimize the one-block case.
-		 */
-		if (nfsb != 1)
-			irecs = kmem_zalloc(sizeof(irec) * nfsb,
-					    KM_NOFS);
+	*nmaps = nirecs;
+out_free_irecs:
+	if (irecs != &irec)
+		kmem_free(irecs);
+	return error;
 
-		nirecs = nfsb;
-		error = xfs_bmapi_read(dp, (xfs_fileoff_t)bno, nfsb, irecs,
-				       &nirecs, xfs_bmapi_aflag(whichfork));
-		if (error)
-			goto out;
-	} else {
-		irecs->br_startblock = XFS_DADDR_TO_FSB(mp, mappedbno);
-		irecs->br_startoff = (xfs_fileoff_t)bno;
-		irecs->br_blockcount = nfsb;
-		irecs->br_state = 0;
-		nirecs = 1;
-	}
+invalid_mapping:
+	/* Caller ok with no mapping. */
+	if (XFS_IS_CORRUPT(mp, !(flags & XFS_DABUF_MAP_HOLE_OK))) {
+		error = -EFSCORRUPTED;
+		if (xfs_error_level >= XFS_ERRLEVEL_LOW) {
+			xfs_alert(mp, "%s: bno %u inode %llu",
+					__func__, bno, dp->i_ino);
 
-	if (!xfs_da_map_covers_blocks(nirecs, irecs, bno, nfsb)) {
-		error = mappedbno == -2 ? -1 : -EFSCORRUPTED;
-		if (unlikely(error == -EFSCORRUPTED)) {
-			if (xfs_error_level >= XFS_ERRLEVEL_LOW) {
-				int i;
-				xfs_alert(mp, "%s: bno %lld dir: inode %lld",
-					__func__, (long long)bno,
-					(long long)dp->i_ino);
-				for (i = 0; i < *nmaps; i++) {
-					xfs_alert(mp,
+			for (i = 0; i < nirecs; i++) {
+				xfs_alert(mp,
 "[%02d] br_startoff %lld br_startblock %lld br_blockcount %lld br_state %d",
-						i,
-						(long long)irecs[i].br_startoff,
-						(long long)irecs[i].br_startblock,
-						(long long)irecs[i].br_blockcount,
-						irecs[i].br_state);
-				}
+					i, irecs[i].br_startoff,
+					irecs[i].br_startblock,
+					irecs[i].br_blockcount,
+					irecs[i].br_state);
 			}
-			XFS_ERROR_REPORT("xfs_da_do_buf(1)",
-					 XFS_ERRLEVEL_LOW, mp);
 		}
-		goto out;
+	} else {
+		*nmaps = 0;
 	}
-	error = xfs_buf_map_from_irec(mp, map, nmaps, irecs, nirecs);
-out:
-	if (irecs != &irec)
-		kmem_free(irecs);
-	return error;
+	goto out_free_irecs;
 }
 
 /*
@@ -2589,39 +2577,26 @@ out:
  */
 int
 xfs_da_get_buf(
-	struct xfs_trans	*trans,
+	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp,
 	int			whichfork)
 {
+	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_buf		*bp;
-	struct xfs_buf_map	map;
-	struct xfs_buf_map	*mapp;
-	int			nmap;
+	struct xfs_buf_map	map, *mapp = &map;
+	int			nmap = 1;
 	int			error;
 
 	*bpp = NULL;
-	mapp = &map;
-	nmap = 1;
-	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
-				&mapp, &nmap);
-	if (error) {
-		/* mapping a hole is not an error, but we don't continue */
-		if (error == -1)
-			error = 0;
+	error = xfs_dabuf_map(dp, bno, 0, whichfork, &mapp, &nmap);
+	if (error || nmap == 0)
 		goto out_free;
-	}
 
-	bp = xfs_trans_get_buf_map(trans, dp->i_mount->m_ddev_targp,
-				    mapp, nmap, 0);
-	error = bp ? bp->b_error : -EIO;
-	if (error) {
-		if (bp)
-			xfs_trans_brelse(trans, bp);
+	error = xfs_trans_get_buf_map(tp, mp->m_ddev_targp, mapp, nmap, 0, &bp);
+	if (error)
 		goto out_free;
-	}
 
 	*bpp = bp;
 
@@ -2637,35 +2612,27 @@ out_free:
  */
 int
 xfs_da_read_buf(
-	struct xfs_trans	*trans,
+	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
+	unsigned int		flags,
 	struct xfs_buf		**bpp,
 	int			whichfork,
 	const struct xfs_buf_ops *ops)
 {
+	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_buf		*bp;
-	struct xfs_buf_map	map;
-	struct xfs_buf_map	*mapp;
-	int			nmap;
+	struct xfs_buf_map	map, *mapp = &map;
+	int			nmap = 1;
 	int			error;
 
 	*bpp = NULL;
-	mapp = &map;
-	nmap = 1;
-	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
-				&mapp, &nmap);
-	if (error) {
-		/* mapping a hole is not an error, but we don't continue */
-		if (error == -1)
-			error = 0;
+	error = xfs_dabuf_map(dp, bno, flags, whichfork, &mapp, &nmap);
+	if (error || !nmap)
 		goto out_free;
-	}
 
-	error = xfs_trans_read_buf_map(dp->i_mount, trans,
-					dp->i_mount->m_ddev_targp,
-					mapp, nmap, 0, &bp, ops);
+	error = xfs_trans_read_buf_map(mp, tp, mp->m_ddev_targp, mapp, nmap, 0,
+			&bp, ops);
 	if (error)
 		goto out_free;
 
@@ -2688,7 +2655,7 @@ int
 xfs_da_reada_buf(
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mappedbno,
+	unsigned int		flags,
 	int			whichfork,
 	const struct xfs_buf_ops *ops)
 {
@@ -2699,16 +2666,10 @@ xfs_da_reada_buf(
 
 	mapp = &map;
 	nmap = 1;
-	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
-				&mapp, &nmap);
-	if (error) {
-		/* mapping a hole is not an error, but we don't continue */
-		if (error == -1)
-			error = 0;
+	error = xfs_dabuf_map(dp, bno, flags, whichfork, &mapp, &nmap);
+	if (error || !nmap)
 		goto out_free;
-	}
 
-	mappedbno = mapp[0].bm_bn;
 	xfs_buf_readahead_map(dp->i_mount->m_ddev_targp, mapp, nmap, ops);
 
 out_free:
diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h
index ae0bbd20d9ca..53e503b6f186 100644
--- a/fs/xfs/libxfs/xfs_da_btree.h
+++ b/fs/xfs/libxfs/xfs_da_btree.h
@@ -10,7 +10,6 @@
 struct xfs_inode;
 struct xfs_trans;
 struct zone;
-struct xfs_dir_ops;
 
 /*
  * Directory/attribute geometry information. There will be one of these for each
@@ -18,15 +17,23 @@ struct xfs_dir_ops;
  * structures will be attached to the xfs_mount.
  */
 struct xfs_da_geometry {
-	int		blksize;	/* da block size in bytes */
-	int		fsbcount;	/* da block size in filesystem blocks */
+	unsigned int	blksize;	/* da block size in bytes */
+	unsigned int	fsbcount;	/* da block size in filesystem blocks */
 	uint8_t		fsblog;		/* log2 of _filesystem_ block size */
 	uint8_t		blklog;		/* log2 of da block size */
-	uint		node_ents;	/* # of entries in a danode */
-	int		magicpct;	/* 37% of block size in bytes */
+	unsigned int	node_hdr_size;	/* danode header size in bytes */
+	unsigned int	node_ents;	/* # of entries in a danode */
+	unsigned int	magicpct;	/* 37% of block size in bytes */
 	xfs_dablk_t	datablk;	/* blockno of dir data v2 */
+	unsigned int	leaf_hdr_size;	/* dir2 leaf header size */
+	unsigned int	leaf_max_ents;	/* # of entries in dir2 leaf */
 	xfs_dablk_t	leafblk;	/* blockno of leaf data v2 */
+	unsigned int	free_hdr_size;	/* dir2 free header size */
+	unsigned int	free_max_bests;	/* # of bests entries in dir2 free */
 	xfs_dablk_t	freeblk;	/* blockno of free data v2 */
+
+	xfs_dir2_data_aoff_t data_first_offset;
+	size_t		data_entry_offset;
 };
 
 /*========================================================================
@@ -50,9 +57,10 @@ typedef struct xfs_da_args {
 	const uint8_t		*name;		/* string (maybe not NULL terminated) */
 	int		namelen;	/* length of string (maybe no NULL) */
 	uint8_t		filetype;	/* filetype of inode for directories */
-	uint8_t		*value;		/* set of bytes (maybe contain NULLs) */
+	void		*value;		/* set of bytes (maybe contain NULLs) */
 	int		valuelen;	/* length of value */
-	int		flags;		/* argument flags (eg: ATTR_NOCREATE) */
+	unsigned int	attr_filter;	/* XFS_ATTR_{ROOT,SECURE,INCOMPLETE} */
+	unsigned int	attr_flags;	/* XATTR_{CREATE,REPLACE} */
 	xfs_dahash_t	hashval;	/* hash value of name */
 	xfs_ino_t	inumber;	/* input/output inode number */
 	struct xfs_inode *dp;		/* directory inode to manipulate */
@@ -81,7 +89,7 @@ typedef struct xfs_da_args {
 #define XFS_DA_OP_ADDNAME	0x0004	/* this is an add operation */
 #define XFS_DA_OP_OKNOENT	0x0008	/* lookup/add op, ENOENT ok, else die */
 #define XFS_DA_OP_CILOOKUP	0x0010	/* lookup to return CI name if found */
-#define XFS_DA_OP_ALLOCVAL	0x0020	/* lookup to alloc buffer if found  */
+#define XFS_DA_OP_NOTIME	0x0020	/* don't update inode timestamps */
 
 #define XFS_DA_OP_FLAGS \
 	{ XFS_DA_OP_JUSTCHECK,	"JUSTCHECK" }, \
@@ -89,7 +97,7 @@ typedef struct xfs_da_args {
 	{ XFS_DA_OP_ADDNAME,	"ADDNAME" }, \
 	{ XFS_DA_OP_OKNOENT,	"OKNOENT" }, \
 	{ XFS_DA_OP_CILOOKUP,	"CILOOKUP" }, \
-	{ XFS_DA_OP_ALLOCVAL,	"ALLOCVAL" }
+	{ XFS_DA_OP_NOTIME,	"NOTIME" }
 
 /*
  * Storage for holding state during Btree searches and split/join ops.
@@ -125,6 +133,25 @@ typedef struct xfs_da_state {
 } xfs_da_state_t;
 
 /*
+ * In-core version of the node header to abstract the differences in the v2 and
+ * v3 disk format of the headers. Callers need to convert to/from disk format as
+ * appropriate.
+ */
+struct xfs_da3_icnode_hdr {
+	uint32_t		forw;
+	uint32_t		back;
+	uint16_t		magic;
+	uint16_t		count;
+	uint16_t		level;
+
+	/*
+	 * Pointer to the on-disk format entries, which are behind the
+	 * variable size (v4 vs v5) header in the on-disk block.
+	 */
+	struct xfs_da_node_entry *btree;
+};
+
+/*
  * Utility macros to aid in logging changed structure fields.
  */
 #define XFS_DA_LOGOFF(BASE, ADDR)	((char *)(ADDR) - (char *)(BASE))
@@ -132,16 +159,6 @@ typedef struct xfs_da_state {
 		(uint)(XFS_DA_LOGOFF(BASE, ADDR)), \
 		(uint)(XFS_DA_LOGOFF(BASE, ADDR)+(SIZE)-1)
 
-/*
- * Name ops for directory and/or attr name operations
- */
-struct xfs_nameops {
-	xfs_dahash_t	(*hashname)(struct xfs_name *);
-	enum xfs_dacmp	(*compname)(struct xfs_da_args *,
-					const unsigned char *, int);
-};
-
-
 /*========================================================================
  * Function prototypes.
  *========================================================================*/
@@ -172,25 +189,28 @@ int	xfs_da3_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
 int	xfs_da3_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
 				       xfs_da_state_blk_t *new_blk);
 int	xfs_da3_node_read(struct xfs_trans *tp, struct xfs_inode *dp,
-			 xfs_dablk_t bno, xfs_daddr_t mappedbno,
-			 struct xfs_buf **bpp, int which_fork);
+			xfs_dablk_t bno, struct xfs_buf **bpp, int whichfork);
+int	xfs_da3_node_read_mapped(struct xfs_trans *tp, struct xfs_inode *dp,
+			xfs_daddr_t mappedbno, struct xfs_buf **bpp,
+			int whichfork);
 
 /*
  * Utility routines.
  */
+
+#define XFS_DABUF_MAP_HOLE_OK	(1 << 0)
+
 int	xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno);
 int	xfs_da_grow_inode_int(struct xfs_da_args *args, xfs_fileoff_t *bno,
 			      int count);
 int	xfs_da_get_buf(struct xfs_trans *trans, struct xfs_inode *dp,
-			      xfs_dablk_t bno, xfs_daddr_t mappedbno,
-			      struct xfs_buf **bp, int whichfork);
+		xfs_dablk_t bno, struct xfs_buf **bp, int whichfork);
 int	xfs_da_read_buf(struct xfs_trans *trans, struct xfs_inode *dp,
-			       xfs_dablk_t bno, xfs_daddr_t mappedbno,
-			       struct xfs_buf **bpp, int whichfork,
-			       const struct xfs_buf_ops *ops);
+		xfs_dablk_t bno, unsigned int flags, struct xfs_buf **bpp,
+		int whichfork, const struct xfs_buf_ops *ops);
 int	xfs_da_reada_buf(struct xfs_inode *dp, xfs_dablk_t bno,
-				xfs_daddr_t mapped_bno, int whichfork,
-				const struct xfs_buf_ops *ops);
+		unsigned int flags, int whichfork,
+		const struct xfs_buf_ops *ops);
 int	xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
 					  struct xfs_buf *dead_buf);
 
@@ -202,7 +222,11 @@ enum xfs_dacmp xfs_da_compname(struct xfs_da_args *args,
 xfs_da_state_t *xfs_da_state_alloc(void);
 void xfs_da_state_free(xfs_da_state_t *state);
 
+void	xfs_da3_node_hdr_from_disk(struct xfs_mount *mp,
+		struct xfs_da3_icnode_hdr *to, struct xfs_da_intnode *from);
+void	xfs_da3_node_hdr_to_disk(struct xfs_mount *mp,
+		struct xfs_da_intnode *to, struct xfs_da3_icnode_hdr *from);
+
 extern struct kmem_zone *xfs_da_state_zone;
-extern const struct xfs_nameops xfs_default_nameops;
 
 #endif	/* __XFS_DA_BTREE_H__ */
diff --git a/fs/xfs/libxfs/xfs_da_format.c b/fs/xfs/libxfs/xfs_da_format.c
deleted file mode 100644
index b1ae572496b6..000000000000
--- a/fs/xfs/libxfs/xfs_da_format.c
+++ /dev/null
@@ -1,888 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc.
- * Copyright (c) 2013 Red Hat, Inc.
- * All Rights Reserved.
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_shared.h"
-#include "xfs_format.h"
-#include "xfs_log_format.h"
-#include "xfs_trans_resv.h"
-#include "xfs_mount.h"
-#include "xfs_inode.h"
-#include "xfs_dir2.h"
-
-/*
- * Shortform directory ops
- */
-static int
-xfs_dir2_sf_entsize(
-	struct xfs_dir2_sf_hdr	*hdr,
-	int			len)
-{
-	int count = sizeof(struct xfs_dir2_sf_entry);	/* namelen + offset */
-
-	count += len;					/* name */
-	count += hdr->i8count ? XFS_INO64_SIZE : XFS_INO32_SIZE; /* ino # */
-	return count;
-}
-
-static int
-xfs_dir3_sf_entsize(
-	struct xfs_dir2_sf_hdr	*hdr,
-	int			len)
-{
-	return xfs_dir2_sf_entsize(hdr, len) + sizeof(uint8_t);
-}
-
-static struct xfs_dir2_sf_entry *
-xfs_dir2_sf_nextentry(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return (struct xfs_dir2_sf_entry *)
-		((char *)sfep + xfs_dir2_sf_entsize(hdr, sfep->namelen));
-}
-
-static struct xfs_dir2_sf_entry *
-xfs_dir3_sf_nextentry(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return (struct xfs_dir2_sf_entry *)
-		((char *)sfep + xfs_dir3_sf_entsize(hdr, sfep->namelen));
-}
-
-
-/*
- * For filetype enabled shortform directories, the file type field is stored at
- * the end of the name.  Because it's only a single byte, endian conversion is
- * not necessary. For non-filetype enable directories, the type is always
- * unknown and we never store the value.
- */
-static uint8_t
-xfs_dir2_sfe_get_ftype(
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return XFS_DIR3_FT_UNKNOWN;
-}
-
-static void
-xfs_dir2_sfe_put_ftype(
-	struct xfs_dir2_sf_entry *sfep,
-	uint8_t			ftype)
-{
-	ASSERT(ftype < XFS_DIR3_FT_MAX);
-}
-
-static uint8_t
-xfs_dir3_sfe_get_ftype(
-	struct xfs_dir2_sf_entry *sfep)
-{
-	uint8_t		ftype;
-
-	ftype = sfep->name[sfep->namelen];
-	if (ftype >= XFS_DIR3_FT_MAX)
-		return XFS_DIR3_FT_UNKNOWN;
-	return ftype;
-}
-
-static void
-xfs_dir3_sfe_put_ftype(
-	struct xfs_dir2_sf_entry *sfep,
-	uint8_t			ftype)
-{
-	ASSERT(ftype < XFS_DIR3_FT_MAX);
-
-	sfep->name[sfep->namelen] = ftype;
-}
-
-/*
- * Inode numbers in short-form directories can come in two versions,
- * either 4 bytes or 8 bytes wide.  These helpers deal with the
- * two forms transparently by looking at the headers i8count field.
- *
- * For 64-bit inode number the most significant byte must be zero.
- */
-static xfs_ino_t
-xfs_dir2_sf_get_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	uint8_t			*from)
-{
-	if (hdr->i8count)
-		return get_unaligned_be64(from) & 0x00ffffffffffffffULL;
-	else
-		return get_unaligned_be32(from);
-}
-
-static void
-xfs_dir2_sf_put_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	uint8_t			*to,
-	xfs_ino_t		ino)
-{
-	ASSERT((ino & 0xff00000000000000ULL) == 0);
-
-	if (hdr->i8count)
-		put_unaligned_be64(ino, to);
-	else
-		put_unaligned_be32(ino, to);
-}
-
-static xfs_ino_t
-xfs_dir2_sf_get_parent_ino(
-	struct xfs_dir2_sf_hdr	*hdr)
-{
-	return xfs_dir2_sf_get_ino(hdr, hdr->parent);
-}
-
-static void
-xfs_dir2_sf_put_parent_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	xfs_ino_t		ino)
-{
-	xfs_dir2_sf_put_ino(hdr, hdr->parent, ino);
-}
-
-/*
- * In short-form directory entries the inode numbers are stored at variable
- * offset behind the entry name. If the entry stores a filetype value, then it
- * sits between the name and the inode number. Hence the inode numbers may only
- * be accessed through the helpers below.
- */
-static xfs_ino_t
-xfs_dir2_sfe_get_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return xfs_dir2_sf_get_ino(hdr, &sfep->name[sfep->namelen]);
-}
-
-static void
-xfs_dir2_sfe_put_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep,
-	xfs_ino_t		ino)
-{
-	xfs_dir2_sf_put_ino(hdr, &sfep->name[sfep->namelen], ino);
-}
-
-static xfs_ino_t
-xfs_dir3_sfe_get_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return xfs_dir2_sf_get_ino(hdr, &sfep->name[sfep->namelen + 1]);
-}
-
-static void
-xfs_dir3_sfe_put_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep,
-	xfs_ino_t		ino)
-{
-	xfs_dir2_sf_put_ino(hdr, &sfep->name[sfep->namelen + 1], ino);
-}
-
-
-/*
- * Directory data block operations
- */
-
-/*
- * For special situations, the dirent size ends up fixed because we always know
- * what the size of the entry is. That's true for the "." and "..", and
- * therefore we know that they are a fixed size and hence their offsets are
- * constant, as is the first entry.
- *
- * Hence, this calculation is written as a macro to be able to be calculated at
- * compile time and so certain offsets can be calculated directly in the
- * structure initaliser via the macro. There are two macros - one for dirents
- * with ftype and without so there are no unresolvable conditionals in the
- * calculations. We also use round_up() as XFS_DIR2_DATA_ALIGN is always a power
- * of 2 and the compiler doesn't reject it (unlike roundup()).
- */
-#define XFS_DIR2_DATA_ENTSIZE(n)					\
-	round_up((offsetof(struct xfs_dir2_data_entry, name[0]) + (n) +	\
-		 sizeof(xfs_dir2_data_off_t)), XFS_DIR2_DATA_ALIGN)
-
-#define XFS_DIR3_DATA_ENTSIZE(n)					\
-	round_up((offsetof(struct xfs_dir2_data_entry, name[0]) + (n) +	\
-		 sizeof(xfs_dir2_data_off_t) + sizeof(uint8_t)),	\
-		XFS_DIR2_DATA_ALIGN)
-
-static int
-xfs_dir2_data_entsize(
-	int			n)
-{
-	return XFS_DIR2_DATA_ENTSIZE(n);
-}
-
-static int
-xfs_dir3_data_entsize(
-	int			n)
-{
-	return XFS_DIR3_DATA_ENTSIZE(n);
-}
-
-static uint8_t
-xfs_dir2_data_get_ftype(
-	struct xfs_dir2_data_entry *dep)
-{
-	return XFS_DIR3_FT_UNKNOWN;
-}
-
-static void
-xfs_dir2_data_put_ftype(
-	struct xfs_dir2_data_entry *dep,
-	uint8_t			ftype)
-{
-	ASSERT(ftype < XFS_DIR3_FT_MAX);
-}
-
-static uint8_t
-xfs_dir3_data_get_ftype(
-	struct xfs_dir2_data_entry *dep)
-{
-	uint8_t		ftype = dep->name[dep->namelen];
-
-	if (ftype >= XFS_DIR3_FT_MAX)
-		return XFS_DIR3_FT_UNKNOWN;
-	return ftype;
-}
-
-static void
-xfs_dir3_data_put_ftype(
-	struct xfs_dir2_data_entry *dep,
-	uint8_t			type)
-{
-	ASSERT(type < XFS_DIR3_FT_MAX);
-	ASSERT(dep->namelen != 0);
-
-	dep->name[dep->namelen] = type;
-}
-
-/*
- * Pointer to an entry's tag word.
- */
-static __be16 *
-xfs_dir2_data_entry_tag_p(
-	struct xfs_dir2_data_entry *dep)
-{
-	return (__be16 *)((char *)dep +
-		xfs_dir2_data_entsize(dep->namelen) - sizeof(__be16));
-}
-
-static __be16 *
-xfs_dir3_data_entry_tag_p(
-	struct xfs_dir2_data_entry *dep)
-{
-	return (__be16 *)((char *)dep +
-		xfs_dir3_data_entsize(dep->namelen) - sizeof(__be16));
-}
-
-/*
- * location of . and .. in data space (always block 0)
- */
-static struct xfs_dir2_data_entry *
-xfs_dir2_data_dot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_data_dotdot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR2_DATA_ENTSIZE(1));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_data_first_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR2_DATA_ENTSIZE(1) +
-				XFS_DIR2_DATA_ENTSIZE(2));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_ftype_data_dotdot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_ftype_data_first_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1) +
-				XFS_DIR3_DATA_ENTSIZE(2));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir3_data_dot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir3_data_dotdot_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir3_data_first_entry_p(
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1) +
-				XFS_DIR3_DATA_ENTSIZE(2));
-}
-
-static struct xfs_dir2_data_free *
-xfs_dir2_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return hdr->bestfree;
-}
-
-static struct xfs_dir2_data_free *
-xfs_dir3_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return ((struct xfs_dir3_data_hdr *)hdr)->best_free;
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir2_data_entry_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
-}
-
-static struct xfs_dir2_data_unused *
-xfs_dir2_data_unused_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_unused *)
-		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
-}
-
-static struct xfs_dir2_data_entry *
-xfs_dir3_data_entry_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
-}
-
-static struct xfs_dir2_data_unused *
-xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_unused *)
-		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
-}
-
-
-/*
- * Directory Leaf block operations
- */
-static int
-xfs_dir2_max_leaf_ents(struct xfs_da_geometry *geo)
-{
-	return (geo->blksize - sizeof(struct xfs_dir2_leaf_hdr)) /
-		(uint)sizeof(struct xfs_dir2_leaf_entry);
-}
-
-static struct xfs_dir2_leaf_entry *
-xfs_dir2_leaf_ents_p(struct xfs_dir2_leaf *lp)
-{
-	return lp->__ents;
-}
-
-static int
-xfs_dir3_max_leaf_ents(struct xfs_da_geometry *geo)
-{
-	return (geo->blksize - sizeof(struct xfs_dir3_leaf_hdr)) /
-		(uint)sizeof(struct xfs_dir2_leaf_entry);
-}
-
-static struct xfs_dir2_leaf_entry *
-xfs_dir3_leaf_ents_p(struct xfs_dir2_leaf *lp)
-{
-	return ((struct xfs_dir3_leaf *)lp)->__ents;
-}
-
-static void
-xfs_dir2_leaf_hdr_from_disk(
-	struct xfs_dir3_icleaf_hdr	*to,
-	struct xfs_dir2_leaf		*from)
-{
-	to->forw = be32_to_cpu(from->hdr.info.forw);
-	to->back = be32_to_cpu(from->hdr.info.back);
-	to->magic = be16_to_cpu(from->hdr.info.magic);
-	to->count = be16_to_cpu(from->hdr.count);
-	to->stale = be16_to_cpu(from->hdr.stale);
-
-	ASSERT(to->magic == XFS_DIR2_LEAF1_MAGIC ||
-	       to->magic == XFS_DIR2_LEAFN_MAGIC);
-}
-
-static void
-xfs_dir2_leaf_hdr_to_disk(
-	struct xfs_dir2_leaf		*to,
-	struct xfs_dir3_icleaf_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DIR2_LEAF1_MAGIC ||
-	       from->magic == XFS_DIR2_LEAFN_MAGIC);
-
-	to->hdr.info.forw = cpu_to_be32(from->forw);
-	to->hdr.info.back = cpu_to_be32(from->back);
-	to->hdr.info.magic = cpu_to_be16(from->magic);
-	to->hdr.count = cpu_to_be16(from->count);
-	to->hdr.stale = cpu_to_be16(from->stale);
-}
-
-static void
-xfs_dir3_leaf_hdr_from_disk(
-	struct xfs_dir3_icleaf_hdr	*to,
-	struct xfs_dir2_leaf		*from)
-{
-	struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)from;
-
-	to->forw = be32_to_cpu(hdr3->info.hdr.forw);
-	to->back = be32_to_cpu(hdr3->info.hdr.back);
-	to->magic = be16_to_cpu(hdr3->info.hdr.magic);
-	to->count = be16_to_cpu(hdr3->count);
-	to->stale = be16_to_cpu(hdr3->stale);
-
-	ASSERT(to->magic == XFS_DIR3_LEAF1_MAGIC ||
-	       to->magic == XFS_DIR3_LEAFN_MAGIC);
-}
-
-static void
-xfs_dir3_leaf_hdr_to_disk(
-	struct xfs_dir2_leaf		*to,
-	struct xfs_dir3_icleaf_hdr	*from)
-{
-	struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)to;
-
-	ASSERT(from->magic == XFS_DIR3_LEAF1_MAGIC ||
-	       from->magic == XFS_DIR3_LEAFN_MAGIC);
-
-	hdr3->info.hdr.forw = cpu_to_be32(from->forw);
-	hdr3->info.hdr.back = cpu_to_be32(from->back);
-	hdr3->info.hdr.magic = cpu_to_be16(from->magic);
-	hdr3->count = cpu_to_be16(from->count);
-	hdr3->stale = cpu_to_be16(from->stale);
-}
-
-
-/*
- * Directory/Attribute Node block operations
- */
-static struct xfs_da_node_entry *
-xfs_da2_node_tree_p(struct xfs_da_intnode *dap)
-{
-	return dap->__btree;
-}
-
-static struct xfs_da_node_entry *
-xfs_da3_node_tree_p(struct xfs_da_intnode *dap)
-{
-	return ((struct xfs_da3_intnode *)dap)->__btree;
-}
-
-static void
-xfs_da2_node_hdr_from_disk(
-	struct xfs_da3_icnode_hdr	*to,
-	struct xfs_da_intnode		*from)
-{
-	ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
-	to->forw = be32_to_cpu(from->hdr.info.forw);
-	to->back = be32_to_cpu(from->hdr.info.back);
-	to->magic = be16_to_cpu(from->hdr.info.magic);
-	to->count = be16_to_cpu(from->hdr.__count);
-	to->level = be16_to_cpu(from->hdr.__level);
-}
-
-static void
-xfs_da2_node_hdr_to_disk(
-	struct xfs_da_intnode		*to,
-	struct xfs_da3_icnode_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DA_NODE_MAGIC);
-	to->hdr.info.forw = cpu_to_be32(from->forw);
-	to->hdr.info.back = cpu_to_be32(from->back);
-	to->hdr.info.magic = cpu_to_be16(from->magic);
-	to->hdr.__count = cpu_to_be16(from->count);
-	to->hdr.__level = cpu_to_be16(from->level);
-}
-
-static void
-xfs_da3_node_hdr_from_disk(
-	struct xfs_da3_icnode_hdr	*to,
-	struct xfs_da_intnode		*from)
-{
-	struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)from;
-
-	ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC));
-	to->forw = be32_to_cpu(hdr3->info.hdr.forw);
-	to->back = be32_to_cpu(hdr3->info.hdr.back);
-	to->magic = be16_to_cpu(hdr3->info.hdr.magic);
-	to->count = be16_to_cpu(hdr3->__count);
-	to->level = be16_to_cpu(hdr3->__level);
-}
-
-static void
-xfs_da3_node_hdr_to_disk(
-	struct xfs_da_intnode		*to,
-	struct xfs_da3_icnode_hdr	*from)
-{
-	struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)to;
-
-	ASSERT(from->magic == XFS_DA3_NODE_MAGIC);
-	hdr3->info.hdr.forw = cpu_to_be32(from->forw);
-	hdr3->info.hdr.back = cpu_to_be32(from->back);
-	hdr3->info.hdr.magic = cpu_to_be16(from->magic);
-	hdr3->__count = cpu_to_be16(from->count);
-	hdr3->__level = cpu_to_be16(from->level);
-}
-
-
-/*
- * Directory free space block operations
- */
-static int
-xfs_dir2_free_max_bests(struct xfs_da_geometry *geo)
-{
-	return (geo->blksize - sizeof(struct xfs_dir2_free_hdr)) /
-		sizeof(xfs_dir2_data_off_t);
-}
-
-static __be16 *
-xfs_dir2_free_bests_p(struct xfs_dir2_free *free)
-{
-	return (__be16 *)((char *)free + sizeof(struct xfs_dir2_free_hdr));
-}
-
-/*
- * Convert data space db to the corresponding free db.
- */
-static xfs_dir2_db_t
-xfs_dir2_db_to_fdb(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
-{
-	return xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET) +
-			(db / xfs_dir2_free_max_bests(geo));
-}
-
-/*
- * Convert data space db to the corresponding index in a free db.
- */
-static int
-xfs_dir2_db_to_fdindex(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
-{
-	return db % xfs_dir2_free_max_bests(geo);
-}
-
-static int
-xfs_dir3_free_max_bests(struct xfs_da_geometry *geo)
-{
-	return (geo->blksize - sizeof(struct xfs_dir3_free_hdr)) /
-		sizeof(xfs_dir2_data_off_t);
-}
-
-static __be16 *
-xfs_dir3_free_bests_p(struct xfs_dir2_free *free)
-{
-	return (__be16 *)((char *)free + sizeof(struct xfs_dir3_free_hdr));
-}
-
-/*
- * Convert data space db to the corresponding free db.
- */
-static xfs_dir2_db_t
-xfs_dir3_db_to_fdb(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
-{
-	return xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET) +
-			(db / xfs_dir3_free_max_bests(geo));
-}
-
-/*
- * Convert data space db to the corresponding index in a free db.
- */
-static int
-xfs_dir3_db_to_fdindex(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
-{
-	return db % xfs_dir3_free_max_bests(geo);
-}
-
-static void
-xfs_dir2_free_hdr_from_disk(
-	struct xfs_dir3_icfree_hdr	*to,
-	struct xfs_dir2_free		*from)
-{
-	to->magic = be32_to_cpu(from->hdr.magic);
-	to->firstdb = be32_to_cpu(from->hdr.firstdb);
-	to->nvalid = be32_to_cpu(from->hdr.nvalid);
-	to->nused = be32_to_cpu(from->hdr.nused);
-	ASSERT(to->magic == XFS_DIR2_FREE_MAGIC);
-}
-
-static void
-xfs_dir2_free_hdr_to_disk(
-	struct xfs_dir2_free		*to,
-	struct xfs_dir3_icfree_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DIR2_FREE_MAGIC);
-
-	to->hdr.magic = cpu_to_be32(from->magic);
-	to->hdr.firstdb = cpu_to_be32(from->firstdb);
-	to->hdr.nvalid = cpu_to_be32(from->nvalid);
-	to->hdr.nused = cpu_to_be32(from->nused);
-}
-
-static void
-xfs_dir3_free_hdr_from_disk(
-	struct xfs_dir3_icfree_hdr	*to,
-	struct xfs_dir2_free		*from)
-{
-	struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)from;
-
-	to->magic = be32_to_cpu(hdr3->hdr.magic);
-	to->firstdb = be32_to_cpu(hdr3->firstdb);
-	to->nvalid = be32_to_cpu(hdr3->nvalid);
-	to->nused = be32_to_cpu(hdr3->nused);
-
-	ASSERT(to->magic == XFS_DIR3_FREE_MAGIC);
-}
-
-static void
-xfs_dir3_free_hdr_to_disk(
-	struct xfs_dir2_free		*to,
-	struct xfs_dir3_icfree_hdr	*from)
-{
-	struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)to;
-
-	ASSERT(from->magic == XFS_DIR3_FREE_MAGIC);
-
-	hdr3->hdr.magic = cpu_to_be32(from->magic);
-	hdr3->firstdb = cpu_to_be32(from->firstdb);
-	hdr3->nvalid = cpu_to_be32(from->nvalid);
-	hdr3->nused = cpu_to_be32(from->nused);
-}
-
-static const struct xfs_dir_ops xfs_dir2_ops = {
-	.sf_entsize = xfs_dir2_sf_entsize,
-	.sf_nextentry = xfs_dir2_sf_nextentry,
-	.sf_get_ftype = xfs_dir2_sfe_get_ftype,
-	.sf_put_ftype = xfs_dir2_sfe_put_ftype,
-	.sf_get_ino = xfs_dir2_sfe_get_ino,
-	.sf_put_ino = xfs_dir2_sfe_put_ino,
-	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
-	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
-
-	.data_entsize = xfs_dir2_data_entsize,
-	.data_get_ftype = xfs_dir2_data_get_ftype,
-	.data_put_ftype = xfs_dir2_data_put_ftype,
-	.data_entry_tag_p = xfs_dir2_data_entry_tag_p,
-	.data_bestfree_p = xfs_dir2_data_bestfree_p,
-
-	.data_dot_offset = sizeof(struct xfs_dir2_data_hdr),
-	.data_dotdot_offset = sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR2_DATA_ENTSIZE(1),
-	.data_first_offset =  sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR2_DATA_ENTSIZE(1) +
-				XFS_DIR2_DATA_ENTSIZE(2),
-	.data_entry_offset = sizeof(struct xfs_dir2_data_hdr),
-
-	.data_dot_entry_p = xfs_dir2_data_dot_entry_p,
-	.data_dotdot_entry_p = xfs_dir2_data_dotdot_entry_p,
-	.data_first_entry_p = xfs_dir2_data_first_entry_p,
-	.data_entry_p = xfs_dir2_data_entry_p,
-	.data_unused_p = xfs_dir2_data_unused_p,
-
-	.leaf_hdr_size = sizeof(struct xfs_dir2_leaf_hdr),
-	.leaf_hdr_to_disk = xfs_dir2_leaf_hdr_to_disk,
-	.leaf_hdr_from_disk = xfs_dir2_leaf_hdr_from_disk,
-	.leaf_max_ents = xfs_dir2_max_leaf_ents,
-	.leaf_ents_p = xfs_dir2_leaf_ents_p,
-
-	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
-	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
-	.node_tree_p = xfs_da2_node_tree_p,
-
-	.free_hdr_size = sizeof(struct xfs_dir2_free_hdr),
-	.free_hdr_to_disk = xfs_dir2_free_hdr_to_disk,
-	.free_hdr_from_disk = xfs_dir2_free_hdr_from_disk,
-	.free_max_bests = xfs_dir2_free_max_bests,
-	.free_bests_p = xfs_dir2_free_bests_p,
-	.db_to_fdb = xfs_dir2_db_to_fdb,
-	.db_to_fdindex = xfs_dir2_db_to_fdindex,
-};
-
-static const struct xfs_dir_ops xfs_dir2_ftype_ops = {
-	.sf_entsize = xfs_dir3_sf_entsize,
-	.sf_nextentry = xfs_dir3_sf_nextentry,
-	.sf_get_ftype = xfs_dir3_sfe_get_ftype,
-	.sf_put_ftype = xfs_dir3_sfe_put_ftype,
-	.sf_get_ino = xfs_dir3_sfe_get_ino,
-	.sf_put_ino = xfs_dir3_sfe_put_ino,
-	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
-	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
-
-	.data_entsize = xfs_dir3_data_entsize,
-	.data_get_ftype = xfs_dir3_data_get_ftype,
-	.data_put_ftype = xfs_dir3_data_put_ftype,
-	.data_entry_tag_p = xfs_dir3_data_entry_tag_p,
-	.data_bestfree_p = xfs_dir2_data_bestfree_p,
-
-	.data_dot_offset = sizeof(struct xfs_dir2_data_hdr),
-	.data_dotdot_offset = sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1),
-	.data_first_offset =  sizeof(struct xfs_dir2_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1) +
-				XFS_DIR3_DATA_ENTSIZE(2),
-	.data_entry_offset = sizeof(struct xfs_dir2_data_hdr),
-
-	.data_dot_entry_p = xfs_dir2_data_dot_entry_p,
-	.data_dotdot_entry_p = xfs_dir2_ftype_data_dotdot_entry_p,
-	.data_first_entry_p = xfs_dir2_ftype_data_first_entry_p,
-	.data_entry_p = xfs_dir2_data_entry_p,
-	.data_unused_p = xfs_dir2_data_unused_p,
-
-	.leaf_hdr_size = sizeof(struct xfs_dir2_leaf_hdr),
-	.leaf_hdr_to_disk = xfs_dir2_leaf_hdr_to_disk,
-	.leaf_hdr_from_disk = xfs_dir2_leaf_hdr_from_disk,
-	.leaf_max_ents = xfs_dir2_max_leaf_ents,
-	.leaf_ents_p = xfs_dir2_leaf_ents_p,
-
-	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
-	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
-	.node_tree_p = xfs_da2_node_tree_p,
-
-	.free_hdr_size = sizeof(struct xfs_dir2_free_hdr),
-	.free_hdr_to_disk = xfs_dir2_free_hdr_to_disk,
-	.free_hdr_from_disk = xfs_dir2_free_hdr_from_disk,
-	.free_max_bests = xfs_dir2_free_max_bests,
-	.free_bests_p = xfs_dir2_free_bests_p,
-	.db_to_fdb = xfs_dir2_db_to_fdb,
-	.db_to_fdindex = xfs_dir2_db_to_fdindex,
-};
-
-static const struct xfs_dir_ops xfs_dir3_ops = {
-	.sf_entsize = xfs_dir3_sf_entsize,
-	.sf_nextentry = xfs_dir3_sf_nextentry,
-	.sf_get_ftype = xfs_dir3_sfe_get_ftype,
-	.sf_put_ftype = xfs_dir3_sfe_put_ftype,
-	.sf_get_ino = xfs_dir3_sfe_get_ino,
-	.sf_put_ino = xfs_dir3_sfe_put_ino,
-	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
-	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
-
-	.data_entsize = xfs_dir3_data_entsize,
-	.data_get_ftype = xfs_dir3_data_get_ftype,
-	.data_put_ftype = xfs_dir3_data_put_ftype,
-	.data_entry_tag_p = xfs_dir3_data_entry_tag_p,
-	.data_bestfree_p = xfs_dir3_data_bestfree_p,
-
-	.data_dot_offset = sizeof(struct xfs_dir3_data_hdr),
-	.data_dotdot_offset = sizeof(struct xfs_dir3_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1),
-	.data_first_offset =  sizeof(struct xfs_dir3_data_hdr) +
-				XFS_DIR3_DATA_ENTSIZE(1) +
-				XFS_DIR3_DATA_ENTSIZE(2),
-	.data_entry_offset = sizeof(struct xfs_dir3_data_hdr),
-
-	.data_dot_entry_p = xfs_dir3_data_dot_entry_p,
-	.data_dotdot_entry_p = xfs_dir3_data_dotdot_entry_p,
-	.data_first_entry_p = xfs_dir3_data_first_entry_p,
-	.data_entry_p = xfs_dir3_data_entry_p,
-	.data_unused_p = xfs_dir3_data_unused_p,
-
-	.leaf_hdr_size = sizeof(struct xfs_dir3_leaf_hdr),
-	.leaf_hdr_to_disk = xfs_dir3_leaf_hdr_to_disk,
-	.leaf_hdr_from_disk = xfs_dir3_leaf_hdr_from_disk,
-	.leaf_max_ents = xfs_dir3_max_leaf_ents,
-	.leaf_ents_p = xfs_dir3_leaf_ents_p,
-
-	.node_hdr_size = sizeof(struct xfs_da3_node_hdr),
-	.node_hdr_to_disk = xfs_da3_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da3_node_hdr_from_disk,
-	.node_tree_p = xfs_da3_node_tree_p,
-
-	.free_hdr_size = sizeof(struct xfs_dir3_free_hdr),
-	.free_hdr_to_disk = xfs_dir3_free_hdr_to_disk,
-	.free_hdr_from_disk = xfs_dir3_free_hdr_from_disk,
-	.free_max_bests = xfs_dir3_free_max_bests,
-	.free_bests_p = xfs_dir3_free_bests_p,
-	.db_to_fdb = xfs_dir3_db_to_fdb,
-	.db_to_fdindex = xfs_dir3_db_to_fdindex,
-};
-
-static const struct xfs_dir_ops xfs_dir2_nondir_ops = {
-	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
-	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
-	.node_tree_p = xfs_da2_node_tree_p,
-};
-
-static const struct xfs_dir_ops xfs_dir3_nondir_ops = {
-	.node_hdr_size = sizeof(struct xfs_da3_node_hdr),
-	.node_hdr_to_disk = xfs_da3_node_hdr_to_disk,
-	.node_hdr_from_disk = xfs_da3_node_hdr_from_disk,
-	.node_tree_p = xfs_da3_node_tree_p,
-};
-
-/*
- * Return the ops structure according to the current config.  If we are passed
- * an inode, then that overrides the default config we use which is based on
- * feature bits.
- */
-const struct xfs_dir_ops *
-xfs_dir_get_ops(
-	struct xfs_mount	*mp,
-	struct xfs_inode	*dp)
-{
-	if (dp)
-		return dp->d_ops;
-	if (mp->m_dir_inode_ops)
-		return mp->m_dir_inode_ops;
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		return &xfs_dir3_ops;
-	if (xfs_sb_version_hasftype(&mp->m_sb))
-		return &xfs_dir2_ftype_ops;
-	return &xfs_dir2_ops;
-}
-
-const struct xfs_dir_ops *
-xfs_nondir_get_ops(
-	struct xfs_mount	*mp,
-	struct xfs_inode	*dp)
-{
-	if (dp)
-		return dp->d_ops;
-	if (mp->m_nondir_inode_ops)
-		return mp->m_nondir_inode_ops;
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		return &xfs_dir3_nondir_ops;
-	return &xfs_dir2_nondir_ops;
-}
diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h
index ae654e06b2fb..08c0a4d98b89 100644
--- a/fs/xfs/libxfs/xfs_da_format.h
+++ b/fs/xfs/libxfs/xfs_da_format.h
@@ -94,19 +94,6 @@ struct xfs_da3_intnode {
 };
 
 /*
- * In-core version of the node header to abstract the differences in the v2 and
- * v3 disk format of the headers. Callers need to convert to/from disk format as
- * appropriate.
- */
-struct xfs_da3_icnode_hdr {
-	uint32_t	forw;
-	uint32_t	back;
-	uint16_t	magic;
-	uint16_t	count;
-	uint16_t	level;
-};
-
-/*
  * Directory version 2.
  *
  * There are 4 possible formats:
@@ -230,7 +217,7 @@ typedef struct xfs_dir2_sf_entry {
 	 * A 64-bit or 32-bit inode number follows here, at a variable offset
 	 * after the name.
 	 */
-} xfs_dir2_sf_entry_t;
+} __packed xfs_dir2_sf_entry_t;
 
 static inline int xfs_dir2_sf_hdr_size(int i8count)
 {
@@ -434,14 +421,6 @@ struct xfs_dir3_leaf_hdr {
 	__be32			pad;		/* 64 bit alignment */
 };
 
-struct xfs_dir3_icleaf_hdr {
-	uint32_t		forw;
-	uint32_t		back;
-	uint16_t		magic;
-	uint16_t		count;
-	uint16_t		stale;
-};
-
 /*
  * Leaf block entry.
  */
@@ -482,7 +461,7 @@ xfs_dir2_leaf_bests_p(struct xfs_dir2_leaf_tail *ltp)
 }
 
 /*
- * Free space block defintions for the node format.
+ * Free space block definitions for the node format.
  */
 
 /*
@@ -521,19 +500,6 @@ struct xfs_dir3_free {
 #define XFS_DIR3_FREE_CRC_OFF  offsetof(struct xfs_dir3_free, hdr.hdr.crc)
 
 /*
- * In core version of the free block header, abstracted away from on-disk format
- * differences. Use this in the code, and convert to/from the disk version using
- * xfs_dir3_free_hdr_from_disk/xfs_dir3_free_hdr_to_disk.
- */
-struct xfs_dir3_icfree_hdr {
-	uint32_t	magic;
-	uint32_t	firstdb;
-	uint32_t	nvalid;
-	uint32_t	nused;
-
-};
-
-/*
  * Single block format.
  *
  * The single block format looks like the following drawing on disk:
@@ -710,29 +676,6 @@ struct xfs_attr3_leafblock {
 };
 
 /*
- * incore, neutral version of the attribute leaf header
- */
-struct xfs_attr3_icleaf_hdr {
-	uint32_t	forw;
-	uint32_t	back;
-	uint16_t	magic;
-	uint16_t	count;
-	uint16_t	usedbytes;
-	/*
-	 * firstused is 32-bit here instead of 16-bit like the on-disk variant
-	 * to support maximum fsb size of 64k without overflow issues throughout
-	 * the attr code. Instead, the overflow condition is handled on
-	 * conversion to/from disk.
-	 */
-	uint32_t	firstused;
-	__u8		holes;
-	struct {
-		uint16_t	base;
-		uint16_t	size;
-	} freemap[XFS_ATTR_LEAF_MAPSIZE];
-};
-
-/*
  * Special value to represent fs block size in the leaf header firstused field.
  * Only used when block size overflows the 2-bytes available on disk.
  */
@@ -740,8 +683,6 @@ struct xfs_attr3_icleaf_hdr {
 
 /*
  * Flags used in the leaf_entry[i].flags field.
- * NOTE: the INCOMPLETE bit must not collide with the flags bits specified
- * on the system call, they are "or"ed together for various operations.
  */
 #define	XFS_ATTR_LOCAL_BIT	0	/* attr is stored locally */
 #define	XFS_ATTR_ROOT_BIT	1	/* limit access to trusted attrs */
@@ -751,19 +692,7 @@ struct xfs_attr3_icleaf_hdr {
 #define XFS_ATTR_ROOT		(1 << XFS_ATTR_ROOT_BIT)
 #define XFS_ATTR_SECURE		(1 << XFS_ATTR_SECURE_BIT)
 #define XFS_ATTR_INCOMPLETE	(1 << XFS_ATTR_INCOMPLETE_BIT)
-
-/*
- * Conversion macros for converting namespace bits from argument flags
- * to ondisk flags.
- */
-#define XFS_ATTR_NSP_ARGS_MASK		(ATTR_ROOT | ATTR_SECURE)
 #define XFS_ATTR_NSP_ONDISK_MASK	(XFS_ATTR_ROOT | XFS_ATTR_SECURE)
-#define XFS_ATTR_NSP_ONDISK(flags)	((flags) & XFS_ATTR_NSP_ONDISK_MASK)
-#define XFS_ATTR_NSP_ARGS(flags)	((flags) & XFS_ATTR_NSP_ARGS_MASK)
-#define XFS_ATTR_NSP_ARGS_TO_ONDISK(x)	(((x) & ATTR_ROOT ? XFS_ATTR_ROOT : 0) |\
-					 ((x) & ATTR_SECURE ? XFS_ATTR_SECURE : 0))
-#define XFS_ATTR_NSP_ONDISK_TO_ARGS(x)	(((x) & XFS_ATTR_ROOT ? ATTR_ROOT : 0) |\
-					 ((x) & XFS_ATTR_SECURE ? ATTR_SECURE : 0))
 
 /*
  * Alignment for namelist and valuelist entries (since they are mixed
diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c
index 867c5dee0751..dd6fcaaea318 100644
--- a/fs/xfs/libxfs/xfs_dir2.c
+++ b/fs/xfs/libxfs/xfs_dir2.c
@@ -52,7 +52,7 @@ xfs_mode_to_ftype(
  * ASCII case-insensitive (ie. A-Z) support for directories that was
  * used in IRIX.
  */
-STATIC xfs_dahash_t
+xfs_dahash_t
 xfs_ascii_ci_hashname(
 	struct xfs_name	*name)
 {
@@ -65,14 +65,14 @@ xfs_ascii_ci_hashname(
 	return hash;
 }
 
-STATIC enum xfs_dacmp
+enum xfs_dacmp
 xfs_ascii_ci_compname(
-	struct xfs_da_args *args,
-	const unsigned char *name,
-	int		len)
+	struct xfs_da_args	*args,
+	const unsigned char	*name,
+	int			len)
 {
-	enum xfs_dacmp	result;
-	int		i;
+	enum xfs_dacmp		result;
+	int			i;
 
 	if (args->namelen != len)
 		return XFS_CMP_DIFFERENT;
@@ -89,26 +89,16 @@ xfs_ascii_ci_compname(
 	return result;
 }
 
-static const struct xfs_nameops xfs_ascii_ci_nameops = {
-	.hashname	= xfs_ascii_ci_hashname,
-	.compname	= xfs_ascii_ci_compname,
-};
-
 int
 xfs_da_mount(
 	struct xfs_mount	*mp)
 {
 	struct xfs_da_geometry	*dageo;
-	int			nodehdr_size;
 
 
 	ASSERT(mp->m_sb.sb_versionnum & XFS_SB_VERSION_DIRV2BIT);
 	ASSERT(xfs_dir2_dirblock_bytes(&mp->m_sb) <= XFS_MAX_BLOCKSIZE);
 
-	mp->m_dir_inode_ops = xfs_dir_get_ops(mp, NULL);
-	mp->m_nondir_inode_ops = xfs_nondir_get_ops(mp, NULL);
-
-	nodehdr_size = mp->m_dir_inode_ops->node_hdr_size;
 	mp->m_dir_geo = kmem_zalloc(sizeof(struct xfs_da_geometry),
 				    KM_MAYFAIL);
 	mp->m_attr_geo = kmem_zalloc(sizeof(struct xfs_da_geometry),
@@ -125,6 +115,27 @@ xfs_da_mount(
 	dageo->fsblog = mp->m_sb.sb_blocklog;
 	dageo->blksize = xfs_dir2_dirblock_bytes(&mp->m_sb);
 	dageo->fsbcount = 1 << mp->m_sb.sb_dirblklog;
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		dageo->node_hdr_size = sizeof(struct xfs_da3_node_hdr);
+		dageo->leaf_hdr_size = sizeof(struct xfs_dir3_leaf_hdr);
+		dageo->free_hdr_size = sizeof(struct xfs_dir3_free_hdr);
+		dageo->data_entry_offset =
+				sizeof(struct xfs_dir3_data_hdr);
+	} else {
+		dageo->node_hdr_size = sizeof(struct xfs_da_node_hdr);
+		dageo->leaf_hdr_size = sizeof(struct xfs_dir2_leaf_hdr);
+		dageo->free_hdr_size = sizeof(struct xfs_dir2_free_hdr);
+		dageo->data_entry_offset =
+				sizeof(struct xfs_dir2_data_hdr);
+	}
+	dageo->leaf_max_ents = (dageo->blksize - dageo->leaf_hdr_size) /
+			sizeof(struct xfs_dir2_leaf_entry);
+	dageo->free_max_bests = (dageo->blksize - dageo->free_hdr_size) /
+			sizeof(xfs_dir2_data_off_t);
+
+	dageo->data_first_offset = dageo->data_entry_offset +
+			xfs_dir2_data_entsize(mp, 1) +
+			xfs_dir2_data_entsize(mp, 2);
 
 	/*
 	 * Now we've set up the block conversion variables, we can calculate the
@@ -133,7 +144,7 @@ xfs_da_mount(
 	dageo->datablk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_DATA_OFFSET);
 	dageo->leafblk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_LEAF_OFFSET);
 	dageo->freeblk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_FREE_OFFSET);
-	dageo->node_ents = (dageo->blksize - nodehdr_size) /
+	dageo->node_ents = (dageo->blksize - dageo->node_hdr_size) /
 				(uint)sizeof(xfs_da_node_entry_t);
 	dageo->magicpct = (dageo->blksize * 37) / 100;
 
@@ -143,15 +154,10 @@ xfs_da_mount(
 	dageo->fsblog = mp->m_sb.sb_blocklog;
 	dageo->blksize = 1 << dageo->blklog;
 	dageo->fsbcount = 1;
-	dageo->node_ents = (dageo->blksize - nodehdr_size) /
+	dageo->node_hdr_size = mp->m_dir_geo->node_hdr_size;
+	dageo->node_ents = (dageo->blksize - dageo->node_hdr_size) /
 				(uint)sizeof(xfs_da_node_entry_t);
 	dageo->magicpct = (dageo->blksize * 37) / 100;
-
-	if (xfs_sb_version_hasasciici(&mp->m_sb))
-		mp->m_dirnameops = &xfs_ascii_ci_nameops;
-	else
-		mp->m_dirnameops = &xfs_default_nameops;
-
 	return 0;
 }
 
@@ -191,10 +197,10 @@ xfs_dir_ino_validate(
 {
 	bool		ino_ok = xfs_verify_dir_ino(mp, ino);
 
-	if (unlikely(XFS_TEST_ERROR(!ino_ok, mp, XFS_ERRTAG_DIR_INO_VALIDATE))) {
+	if (XFS_IS_CORRUPT(mp, !ino_ok) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_DIR_INO_VALIDATE)) {
 		xfs_warn(mp, "Invalid inode number 0x%Lx",
 				(unsigned long long) ino);
-		XFS_ERROR_REPORT("xfs_dir_ino_validate", XFS_ERRLEVEL_LOW, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
@@ -262,7 +268,7 @@ xfs_dir_createname(
 	args->name = name->name;
 	args->namelen = name->len;
 	args->filetype = name->type;
-	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->hashval = xfs_dir2_hashname(dp->i_mount, name);
 	args->inumber = inum;
 	args->dp = dp;
 	args->total = total;
@@ -358,7 +364,7 @@ xfs_dir_lookup(
 	args->name = name->name;
 	args->namelen = name->len;
 	args->filetype = name->type;
-	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->hashval = xfs_dir2_hashname(dp->i_mount, name);
 	args->dp = dp;
 	args->whichfork = XFS_DATA_FORK;
 	args->trans = tp;
@@ -430,7 +436,7 @@ xfs_dir_removename(
 	args->name = name->name;
 	args->namelen = name->len;
 	args->filetype = name->type;
-	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->hashval = xfs_dir2_hashname(dp->i_mount, name);
 	args->inumber = ino;
 	args->dp = dp;
 	args->total = total;
@@ -491,7 +497,7 @@ xfs_dir_replace(
 	args->name = name->name;
 	args->namelen = name->len;
 	args->filetype = name->type;
-	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->hashval = xfs_dir2_hashname(dp->i_mount, name);
 	args->inumber = inum;
 	args->dp = dp;
 	args->total = total;
@@ -600,7 +606,9 @@ xfs_dir2_isblock(
 	if ((rval = xfs_bmap_last_offset(args->dp, &last, XFS_DATA_FORK)))
 		return rval;
 	rval = XFS_FSB_TO_B(args->dp->i_mount, last) == args->geo->blksize;
-	if (rval != 0 && args->dp->i_d.di_size != args->geo->blksize)
+	if (XFS_IS_CORRUPT(args->dp->i_mount,
+			   rval != 0 &&
+			   args->dp->i_d.di_size != args->geo->blksize))
 		return -EFSCORRUPTED;
 	*vp = rval;
 	return 0;
@@ -716,3 +724,24 @@ xfs_dir2_namecheck(
 	/* There shouldn't be any slashes or nulls here */
 	return !memchr(name, '/', length) && !memchr(name, 0, length);
 }
+
+xfs_dahash_t
+xfs_dir2_hashname(
+	struct xfs_mount	*mp,
+	struct xfs_name		*name)
+{
+	if (unlikely(xfs_sb_version_hasasciici(&mp->m_sb)))
+		return xfs_ascii_ci_hashname(name);
+	return xfs_da_hashname(name->name, name->len);
+}
+
+enum xfs_dacmp
+xfs_dir2_compname(
+	struct xfs_da_args	*args,
+	const unsigned char	*name,
+	int			len)
+{
+	if (unlikely(xfs_sb_version_hasasciici(&args->dp->i_mount->m_sb)))
+		return xfs_ascii_ci_compname(args, name, len);
+	return xfs_da_compname(args, name, len);
+}
diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h
index f54244779492..033777e282f2 100644
--- a/fs/xfs/libxfs/xfs_dir2.h
+++ b/fs/xfs/libxfs/xfs_dir2.h
@@ -18,6 +18,8 @@ struct xfs_dir2_sf_entry;
 struct xfs_dir2_data_hdr;
 struct xfs_dir2_data_entry;
 struct xfs_dir2_data_unused;
+struct xfs_dir3_icfree_hdr;
+struct xfs_dir3_icleaf_hdr;
 
 extern struct xfs_name	xfs_name_dotdot;
 
@@ -27,85 +29,6 @@ extern struct xfs_name	xfs_name_dotdot;
 extern unsigned char xfs_mode_to_ftype(int mode);
 
 /*
- * directory operations vector for encode/decode routines
- */
-struct xfs_dir_ops {
-	int	(*sf_entsize)(struct xfs_dir2_sf_hdr *hdr, int len);
-	struct xfs_dir2_sf_entry *
-		(*sf_nextentry)(struct xfs_dir2_sf_hdr *hdr,
-				struct xfs_dir2_sf_entry *sfep);
-	uint8_t (*sf_get_ftype)(struct xfs_dir2_sf_entry *sfep);
-	void	(*sf_put_ftype)(struct xfs_dir2_sf_entry *sfep,
-				uint8_t ftype);
-	xfs_ino_t (*sf_get_ino)(struct xfs_dir2_sf_hdr *hdr,
-				struct xfs_dir2_sf_entry *sfep);
-	void	(*sf_put_ino)(struct xfs_dir2_sf_hdr *hdr,
-			      struct xfs_dir2_sf_entry *sfep,
-			      xfs_ino_t ino);
-	xfs_ino_t (*sf_get_parent_ino)(struct xfs_dir2_sf_hdr *hdr);
-	void	(*sf_put_parent_ino)(struct xfs_dir2_sf_hdr *hdr,
-				     xfs_ino_t ino);
-
-	int	(*data_entsize)(int len);
-	uint8_t (*data_get_ftype)(struct xfs_dir2_data_entry *dep);
-	void	(*data_put_ftype)(struct xfs_dir2_data_entry *dep,
-				uint8_t ftype);
-	__be16 * (*data_entry_tag_p)(struct xfs_dir2_data_entry *dep);
-	struct xfs_dir2_data_free *
-		(*data_bestfree_p)(struct xfs_dir2_data_hdr *hdr);
-
-	xfs_dir2_data_aoff_t data_dot_offset;
-	xfs_dir2_data_aoff_t data_dotdot_offset;
-	xfs_dir2_data_aoff_t data_first_offset;
-	size_t	data_entry_offset;
-
-	struct xfs_dir2_data_entry *
-		(*data_dot_entry_p)(struct xfs_dir2_data_hdr *hdr);
-	struct xfs_dir2_data_entry *
-		(*data_dotdot_entry_p)(struct xfs_dir2_data_hdr *hdr);
-	struct xfs_dir2_data_entry *
-		(*data_first_entry_p)(struct xfs_dir2_data_hdr *hdr);
-	struct xfs_dir2_data_entry *
-		(*data_entry_p)(struct xfs_dir2_data_hdr *hdr);
-	struct xfs_dir2_data_unused *
-		(*data_unused_p)(struct xfs_dir2_data_hdr *hdr);
-
-	int	leaf_hdr_size;
-	void	(*leaf_hdr_to_disk)(struct xfs_dir2_leaf *to,
-				    struct xfs_dir3_icleaf_hdr *from);
-	void	(*leaf_hdr_from_disk)(struct xfs_dir3_icleaf_hdr *to,
-				      struct xfs_dir2_leaf *from);
-	int	(*leaf_max_ents)(struct xfs_da_geometry *geo);
-	struct xfs_dir2_leaf_entry *
-		(*leaf_ents_p)(struct xfs_dir2_leaf *lp);
-
-	int	node_hdr_size;
-	void	(*node_hdr_to_disk)(struct xfs_da_intnode *to,
-				    struct xfs_da3_icnode_hdr *from);
-	void	(*node_hdr_from_disk)(struct xfs_da3_icnode_hdr *to,
-				      struct xfs_da_intnode *from);
-	struct xfs_da_node_entry *
-		(*node_tree_p)(struct xfs_da_intnode *dap);
-
-	int	free_hdr_size;
-	void	(*free_hdr_to_disk)(struct xfs_dir2_free *to,
-				    struct xfs_dir3_icfree_hdr *from);
-	void	(*free_hdr_from_disk)(struct xfs_dir3_icfree_hdr *to,
-				      struct xfs_dir2_free *from);
-	int	(*free_max_bests)(struct xfs_da_geometry *geo);
-	__be16 * (*free_bests_p)(struct xfs_dir2_free *free);
-	xfs_dir2_db_t (*db_to_fdb)(struct xfs_da_geometry *geo,
-				   xfs_dir2_db_t db);
-	int	(*db_to_fdindex)(struct xfs_da_geometry *geo,
-				 xfs_dir2_db_t db);
-};
-
-extern const struct xfs_dir_ops *
-	xfs_dir_get_ops(struct xfs_mount *mp, struct xfs_inode *dp);
-extern const struct xfs_dir_ops *
-	xfs_nondir_get_ops(struct xfs_mount *mp, struct xfs_inode *dp);
-
-/*
  * Generic directory interface routines
  */
 extern void xfs_dir_startup(void);
@@ -124,6 +47,8 @@ extern int xfs_dir_lookup(struct xfs_trans *tp, struct xfs_inode *dp,
 extern int xfs_dir_removename(struct xfs_trans *tp, struct xfs_inode *dp,
 				struct xfs_name *name, xfs_ino_t ino,
 				xfs_extlen_t tot);
+extern bool xfs_dir2_sf_replace_needblock(struct xfs_inode *dp,
+				xfs_ino_t inum);
 extern int xfs_dir_replace(struct xfs_trans *tp, struct xfs_inode *dp,
 				struct xfs_name *name, xfs_ino_t inum,
 				xfs_extlen_t tot);
@@ -143,10 +68,7 @@ extern int xfs_dir2_isleaf(struct xfs_da_args *args, int *r);
 extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
 				struct xfs_buf *bp);
 
-extern void xfs_dir2_data_freescan_int(struct xfs_da_geometry *geo,
-		const struct xfs_dir_ops *ops,
-		struct xfs_dir2_data_hdr *hdr, int *loghead);
-extern void xfs_dir2_data_freescan(struct xfs_inode *dp,
+extern void xfs_dir2_data_freescan(struct xfs_mount *mp,
 		struct xfs_dir2_data_hdr *hdr, int *loghead);
 extern void xfs_dir2_data_log_entry(struct xfs_da_args *args,
 		struct xfs_buf *bp, struct xfs_dir2_data_entry *dep);
@@ -324,7 +246,7 @@ xfs_dir2_leaf_tail_p(struct xfs_da_geometry *geo, struct xfs_dir2_leaf *lp)
 #define XFS_READDIR_BUFSIZE	(32768)
 
 unsigned char xfs_dir3_get_dtype(struct xfs_mount *mp, uint8_t filetype);
-void *xfs_dir3_data_endp(struct xfs_da_geometry *geo,
+unsigned int xfs_dir3_data_end_offset(struct xfs_da_geometry *geo,
 		struct xfs_dir2_data_hdr *hdr);
 bool xfs_dir2_namecheck(const void *name, size_t length);
 
diff --git a/fs/xfs/libxfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c
index 49e4bc39e7bb..1dbf2f980a26 100644
--- a/fs/xfs/libxfs/xfs_dir2_block.c
+++ b/fs/xfs/libxfs/xfs_dir2_block.c
@@ -114,6 +114,23 @@ const struct xfs_buf_ops xfs_dir3_block_buf_ops = {
 	.verify_struct = xfs_dir3_block_verify,
 };
 
+static xfs_failaddr_t
+xfs_dir3_block_header_check(
+	struct xfs_inode	*dp,
+	struct xfs_buf		*bp)
+{
+	struct xfs_mount	*mp = dp->i_mount;
+
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
+
+		if (be64_to_cpu(hdr3->owner) != dp->i_ino)
+			return __this_address;
+	}
+
+	return NULL;
+}
+
 int
 xfs_dir3_block_read(
 	struct xfs_trans	*tp,
@@ -121,12 +138,24 @@ xfs_dir3_block_read(
 	struct xfs_buf		**bpp)
 {
 	struct xfs_mount	*mp = dp->i_mount;
+	xfs_failaddr_t		fa;
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, mp->m_dir_geo->datablk, -1, bpp,
+	err = xfs_da_read_buf(tp, dp, mp->m_dir_geo->datablk, 0, bpp,
 				XFS_DATA_FORK, &xfs_dir3_block_buf_ops);
-	if (!err && tp && *bpp)
-		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_BLOCK_BUF);
+	if (err || !*bpp)
+		return err;
+
+	/* Check things that we can't do in the verifier. */
+	fa = xfs_dir3_block_header_check(dp, *bpp);
+	if (fa) {
+		__xfs_buf_mark_corrupt(*bpp, fa);
+		xfs_trans_brelse(tp, *bpp);
+		*bpp = NULL;
+		return -EFSCORRUPTED;
+	}
+
+	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_BLOCK_BUF);
 	return err;
 }
 
@@ -172,7 +201,7 @@ xfs_dir2_block_need_space(
 	struct xfs_dir2_data_unused	*enddup = NULL;
 
 	*compact = 0;
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 
 	/*
 	 * If there are stale entries we'll use one for the leaf.
@@ -311,7 +340,7 @@ xfs_dir2_block_compact(
 	 * This needs to happen before the next call to use_free.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(args->dp, hdr, needlog);
+		xfs_dir2_data_freescan(args->dp->i_mount, hdr, needlog);
 }
 
 /*
@@ -355,7 +384,7 @@ xfs_dir2_block_addname(
 	if (error)
 		return error;
 
-	len = dp->d_ops->data_entsize(args->namelen);
+	len = xfs_dir2_data_entsize(dp->i_mount, args->namelen);
 
 	/*
 	 * Set up pointers to parts of the block.
@@ -458,7 +487,7 @@ xfs_dir2_block_addname(
 		 * This needs to happen before the next call to use_free.
 		 */
 		if (needscan) {
-			xfs_dir2_data_freescan(dp, hdr, &needlog);
+			xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 			needscan = 0;
 		}
 		/*
@@ -541,14 +570,14 @@ xfs_dir2_block_addname(
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, args->namelen);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
+	tagp = xfs_dir2_data_entry_tag_p(dp->i_mount, dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
 	/*
 	 * Clean up the bestfree array and log the header, tail, and entry.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, bp);
 	xfs_dir2_block_log_tail(tp, bp);
@@ -633,7 +662,7 @@ xfs_dir2_block_lookup(
 	 * Fill in inode number, CI name if appropriate, release the block.
 	 */
 	args->inumber = be64_to_cpu(dep->inumber);
-	args->filetype = dp->d_ops->data_get_ftype(dep);
+	args->filetype = xfs_dir2_data_get_ftype(dp->i_mount, dep);
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(args->trans, bp);
 	return error;
@@ -660,13 +689,11 @@ xfs_dir2_block_lookup_int(
 	int			high;		/* binary search high index */
 	int			low;		/* binary search low index */
 	int			mid;		/* binary search current idx */
-	xfs_mount_t		*mp;		/* filesystem mount point */
 	xfs_trans_t		*tp;		/* transaction pointer */
 	enum xfs_dacmp		cmp;		/* comparison result */
 
 	dp = args->dp;
 	tp = args->trans;
-	mp = dp->i_mount;
 
 	error = xfs_dir3_block_read(tp, dp, &bp);
 	if (error)
@@ -718,7 +745,7 @@ xfs_dir2_block_lookup_int(
 		 * and buffer. If it's the first case-insensitive match, store
 		 * the index and buffer and continue looking for an exact match.
 		 */
-		cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+		cmp = xfs_dir2_compname(args, dep->name, dep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			args->cmpresult = cmp;
 			*bpp = bp;
@@ -791,7 +818,8 @@ xfs_dir2_block_removename(
 	needlog = needscan = 0;
 	xfs_dir2_data_make_free(args, bp,
 		(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
-		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
+		xfs_dir2_data_entsize(dp->i_mount, dep->namelen), &needlog,
+		&needscan);
 	/*
 	 * Fix up the block tail.
 	 */
@@ -806,7 +834,7 @@ xfs_dir2_block_removename(
 	 * Fix up bestfree, log the header if necessary.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, bp);
 	xfs_dir3_data_check(dp, bp);
@@ -864,7 +892,7 @@ xfs_dir2_block_replace(
 	 * Change the inode number to the new value.
 	 */
 	dep->inumber = cpu_to_be64(args->inumber);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
 	xfs_dir2_data_log_entry(args, bp, dep);
 	xfs_dir3_data_check(dp, bp);
 	return 0;
@@ -914,7 +942,6 @@ xfs_dir2_leaf_to_block(
 	__be16			*tagp;		/* end of entry (tag) */
 	int			to;		/* block/leaf to index */
 	xfs_trans_t		*tp;		/* transaction pointer */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_leaf_to_block(args);
@@ -923,8 +950,7 @@ xfs_dir2_leaf_to_block(
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = lbp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
 	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAF1_MAGIC ||
@@ -938,7 +964,7 @@ xfs_dir2_leaf_to_block(
 	while (dp->i_d.di_size > args->geo->blksize) {
 		int hdrsz;
 
-		hdrsz = dp->d_ops->data_entry_offset;
+		hdrsz = args->geo->data_entry_offset;
 		bestsp = xfs_dir2_leaf_bests_p(ltp);
 		if (be16_to_cpu(bestsp[be32_to_cpu(ltp->bestcount) - 1]) ==
 					    args->geo->blksize - hdrsz) {
@@ -953,7 +979,7 @@ xfs_dir2_leaf_to_block(
 	 * Read the data block if we don't already have it, give up if it fails.
 	 */
 	if (!dbp) {
-		error = xfs_dir3_data_read(tp, dp, args->geo->datablk, -1, &dbp);
+		error = xfs_dir3_data_read(tp, dp, args->geo->datablk, 0, &dbp);
 		if (error)
 			return error;
 	}
@@ -1004,9 +1030,10 @@ xfs_dir2_leaf_to_block(
 	 */
 	lep = xfs_dir2_block_leaf_p(btp);
 	for (from = to = 0; from < leafhdr.count; from++) {
-		if (ents[from].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+		if (leafhdr.ents[from].address ==
+		    cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			continue;
-		lep[to++] = ents[from];
+		lep[to++] = leafhdr.ents[from];
 	}
 	ASSERT(to == be32_to_cpu(btp->count));
 	xfs_dir2_block_log_leaf(tp, dbp, 0, be32_to_cpu(btp->count) - 1);
@@ -1014,7 +1041,7 @@ xfs_dir2_leaf_to_block(
 	 * Scan the bestfree if we need it and log the data block header.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, dbp);
 	/*
@@ -1039,47 +1066,38 @@ xfs_dir2_leaf_to_block(
  */
 int						/* error */
 xfs_dir2_sf_to_block(
-	xfs_da_args_t		*args)		/* operation arguments */
+	struct xfs_da_args	*args)
 {
+	struct xfs_trans	*tp = args->trans;
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(dp, XFS_DATA_FORK);
+	struct xfs_da_geometry	*geo = args->geo;
 	xfs_dir2_db_t		blkno;		/* dir-relative block # (0) */
 	xfs_dir2_data_hdr_t	*hdr;		/* block header */
 	xfs_dir2_leaf_entry_t	*blp;		/* block leaf entries */
 	struct xfs_buf		*bp;		/* block buffer */
 	xfs_dir2_block_tail_t	*btp;		/* block tail pointer */
 	xfs_dir2_data_entry_t	*dep;		/* data entry pointer */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			dummy;		/* trash */
 	xfs_dir2_data_unused_t	*dup;		/* unused entry pointer */
 	int			endoffset;	/* end of data objects */
 	int			error;		/* error return value */
 	int			i;		/* index */
-	xfs_mount_t		*mp;		/* filesystem mount point */
 	int			needlog;	/* need to log block header */
 	int			needscan;	/* need to scan block freespc */
 	int			newoffset;	/* offset from current entry */
-	int			offset;		/* target block offset */
+	unsigned int		offset = geo->data_entry_offset;
 	xfs_dir2_sf_entry_t	*sfep;		/* sf entry pointer */
 	xfs_dir2_sf_hdr_t	*oldsfp;	/* old shortform header  */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform header  */
 	__be16			*tagp;		/* end of data entry */
-	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_name		name;
-	struct xfs_ifork	*ifp;
 
 	trace_xfs_dir2_sf_to_block(args);
 
-	dp = args->dp;
-	tp = args->trans;
-	mp = dp->i_mount;
-	ifp = XFS_IFORK_PTR(dp, XFS_DATA_FORK);
 	ASSERT(ifp->if_flags & XFS_IFINLINE);
-	/*
-	 * Bomb out if the shortform directory is way too short.
-	 */
-	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		return -EIO;
-	}
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
 
 	oldsfp = (xfs_dir2_sf_hdr_t *)ifp->if_u1.if_data;
 
@@ -1123,7 +1141,7 @@ xfs_dir2_sf_to_block(
 	 * The whole thing is initialized to free by the init routine.
 	 * Say we're using the leaf and tail area.
 	 */
-	dup = dp->d_ops->data_unused_p(hdr);
+	dup = bp->b_addr + offset;
 	needlog = needscan = 0;
 	error = xfs_dir2_data_use_free(args, bp, dup, args->geo->blksize - i,
 			i, &needlog, &needscan);
@@ -1146,35 +1164,37 @@ xfs_dir2_sf_to_block(
 			be16_to_cpu(dup->length), &needlog, &needscan);
 	if (error)
 		goto out_free;
+
 	/*
 	 * Create entry for .
 	 */
-	dep = dp->d_ops->data_dot_entry_p(hdr);
+	dep = bp->b_addr + offset;
 	dep->inumber = cpu_to_be64(dp->i_ino);
 	dep->namelen = 1;
 	dep->name[0] = '.';
-	dp->d_ops->data_put_ftype(dep, XFS_DIR3_FT_DIR);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
-	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+	xfs_dir2_data_put_ftype(mp, dep, XFS_DIR3_FT_DIR);
+	tagp = xfs_dir2_data_entry_tag_p(mp, dep);
+	*tagp = cpu_to_be16(offset);
 	xfs_dir2_data_log_entry(args, bp, dep);
 	blp[0].hashval = cpu_to_be32(xfs_dir_hash_dot);
-	blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
-				(char *)dep - (char *)hdr));
+	blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(offset));
+	offset += xfs_dir2_data_entsize(mp, dep->namelen);
+
 	/*
 	 * Create entry for ..
 	 */
-	dep = dp->d_ops->data_dotdot_entry_p(hdr);
-	dep->inumber = cpu_to_be64(dp->d_ops->sf_get_parent_ino(sfp));
+	dep = bp->b_addr + offset;
+	dep->inumber = cpu_to_be64(xfs_dir2_sf_get_parent_ino(sfp));
 	dep->namelen = 2;
 	dep->name[0] = dep->name[1] = '.';
-	dp->d_ops->data_put_ftype(dep, XFS_DIR3_FT_DIR);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
-	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+	xfs_dir2_data_put_ftype(mp, dep, XFS_DIR3_FT_DIR);
+	tagp = xfs_dir2_data_entry_tag_p(mp, dep);
+	*tagp = cpu_to_be16(offset);
 	xfs_dir2_data_log_entry(args, bp, dep);
 	blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot);
-	blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
-				(char *)dep - (char *)hdr));
-	offset = dp->d_ops->data_first_offset;
+	blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(offset));
+	offset += xfs_dir2_data_entsize(mp, dep->namelen);
+
 	/*
 	 * Loop over existing entries, stuff them in.
 	 */
@@ -1183,6 +1203,7 @@ xfs_dir2_sf_to_block(
 		sfep = NULL;
 	else
 		sfep = xfs_dir2_sf_firstentry(sfp);
+
 	/*
 	 * Need to preserve the existing offset values in the sf directory.
 	 * Insert holes (unused entries) where necessary.
@@ -1199,40 +1220,39 @@ xfs_dir2_sf_to_block(
 		 * There should be a hole here, make one.
 		 */
 		if (offset < newoffset) {
-			dup = (xfs_dir2_data_unused_t *)((char *)hdr + offset);
+			dup = bp->b_addr + offset;
 			dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 			dup->length = cpu_to_be16(newoffset - offset);
-			*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16(
-				((char *)dup - (char *)hdr));
+			*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16(offset);
 			xfs_dir2_data_log_unused(args, bp, dup);
 			xfs_dir2_data_freeinsert(hdr,
-						 dp->d_ops->data_bestfree_p(hdr),
-						 dup, &dummy);
+					xfs_dir2_data_bestfree_p(mp, hdr),
+					dup, &dummy);
 			offset += be16_to_cpu(dup->length);
 			continue;
 		}
 		/*
 		 * Copy a real entry.
 		 */
-		dep = (xfs_dir2_data_entry_t *)((char *)hdr + newoffset);
-		dep->inumber = cpu_to_be64(dp->d_ops->sf_get_ino(sfp, sfep));
+		dep = bp->b_addr + newoffset;
+		dep->inumber = cpu_to_be64(xfs_dir2_sf_get_ino(mp, sfp, sfep));
 		dep->namelen = sfep->namelen;
-		dp->d_ops->data_put_ftype(dep, dp->d_ops->sf_get_ftype(sfep));
+		xfs_dir2_data_put_ftype(mp, dep,
+				xfs_dir2_sf_get_ftype(mp, sfep));
 		memcpy(dep->name, sfep->name, dep->namelen);
-		tagp = dp->d_ops->data_entry_tag_p(dep);
-		*tagp = cpu_to_be16((char *)dep - (char *)hdr);
+		tagp = xfs_dir2_data_entry_tag_p(mp, dep);
+		*tagp = cpu_to_be16(newoffset);
 		xfs_dir2_data_log_entry(args, bp, dep);
 		name.name = sfep->name;
 		name.len = sfep->namelen;
-		blp[2 + i].hashval = cpu_to_be32(mp->m_dirnameops->
-							hashname(&name));
-		blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
-						 (char *)dep - (char *)hdr));
+		blp[2 + i].hashval = cpu_to_be32(xfs_dir2_hashname(mp, &name));
+		blp[2 + i].address =
+			cpu_to_be32(xfs_dir2_byte_to_dataptr(newoffset));
 		offset = (int)((char *)(tagp + 1) - (char *)hdr);
 		if (++i == sfp->count)
 			sfep = NULL;
 		else
-			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+			sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 	}
 	/* Done with the temporary buffer */
 	kmem_free(sfp);
diff --git a/fs/xfs/libxfs/xfs_dir2_data.c b/fs/xfs/libxfs/xfs_dir2_data.c
index 2c79be4c3153..375b3edb2ad2 100644
--- a/fs/xfs/libxfs/xfs_dir2_data.c
+++ b/fs/xfs/libxfs/xfs_dir2_data.c
@@ -13,6 +13,7 @@
 #include "xfs_mount.h"
 #include "xfs_inode.h"
 #include "xfs_dir2.h"
+#include "xfs_dir2_priv.h"
 #include "xfs_error.h"
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
@@ -23,6 +24,71 @@ static xfs_failaddr_t xfs_dir2_data_freefind_verify(
 		struct xfs_dir2_data_unused *dup,
 		struct xfs_dir2_data_free **bf_ent);
 
+struct xfs_dir2_data_free *
+xfs_dir2_data_bestfree_p(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_hdr	*hdr)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb))
+		return ((struct xfs_dir3_data_hdr *)hdr)->best_free;
+	return hdr->bestfree;
+}
+
+/*
+ * Pointer to an entry's tag word.
+ */
+__be16 *
+xfs_dir2_data_entry_tag_p(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_entry	*dep)
+{
+	return (__be16 *)((char *)dep +
+		xfs_dir2_data_entsize(mp, dep->namelen) - sizeof(__be16));
+}
+
+uint8_t
+xfs_dir2_data_get_ftype(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_entry	*dep)
+{
+	if (xfs_sb_version_hasftype(&mp->m_sb)) {
+		uint8_t			ftype = dep->name[dep->namelen];
+
+		if (likely(ftype < XFS_DIR3_FT_MAX))
+			return ftype;
+	}
+
+	return XFS_DIR3_FT_UNKNOWN;
+}
+
+void
+xfs_dir2_data_put_ftype(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_entry	*dep,
+	uint8_t				ftype)
+{
+	ASSERT(ftype < XFS_DIR3_FT_MAX);
+	ASSERT(dep->namelen != 0);
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		dep->name[dep->namelen] = ftype;
+}
+
+/*
+ * The number of leaf entries is limited by the size of the block and the amount
+ * of space used by the data entries.  We don't know how much space is used by
+ * the data entries yet, so just ensure that the count falls somewhere inside
+ * the block right now.
+ */
+static inline unsigned int
+xfs_dir2_data_max_leaf_entries(
+	struct xfs_da_geometry		*geo)
+{
+	return (geo->blksize - sizeof(struct xfs_dir2_block_tail) -
+		geo->data_entry_offset) /
+			sizeof(struct xfs_dir2_leaf_entry);
+}
+
 /*
  * Check the consistency of the data block.
  * The input can also be a block-format directory.
@@ -38,40 +104,27 @@ __xfs_dir3_data_check(
 	xfs_dir2_block_tail_t	*btp=NULL;	/* block tail */
 	int			count;		/* count of entries found */
 	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
-	xfs_dir2_data_entry_t	*dep;		/* data entry */
 	xfs_dir2_data_free_t	*dfp;		/* bestfree entry */
-	xfs_dir2_data_unused_t	*dup;		/* unused entry */
-	char			*endp;		/* end of useful data */
 	int			freeseen;	/* mask of bestfrees seen */
 	xfs_dahash_t		hash;		/* hash of current name */
 	int			i;		/* leaf index */
 	int			lastfree;	/* last entry was unused */
 	xfs_dir2_leaf_entry_t	*lep=NULL;	/* block leaf entries */
 	struct xfs_mount	*mp = bp->b_mount;
-	char			*p;		/* current data position */
 	int			stale;		/* count of stale leaves */
 	struct xfs_name		name;
-	const struct xfs_dir_ops *ops;
-	struct xfs_da_geometry	*geo;
-
-	geo = mp->m_dir_geo;
-
-	/*
-	 * We can be passed a null dp here from a verifier, so we need to go the
-	 * hard way to get them.
-	 */
-	ops = xfs_dir_get_ops(mp, dp);
+	unsigned int		offset;
+	unsigned int		end;
+	struct xfs_da_geometry	*geo = mp->m_dir_geo;
 
 	/*
-	 * If this isn't a directory, or we don't get handed the dir ops,
-	 * something is seriously wrong.  Bail out.
+	 * If this isn't a directory, something is seriously wrong.  Bail out.
 	 */
-	if ((dp && !S_ISDIR(VFS_I(dp)->i_mode)) ||
-	    ops != xfs_dir_get_ops(mp, NULL))
+	if (dp && !S_ISDIR(VFS_I(dp)->i_mode))
 		return __this_address;
 
 	hdr = bp->b_addr;
-	p = (char *)ops->data_entry_p(hdr);
+	offset = geo->data_entry_offset;
 
 	switch (hdr->magic) {
 	case cpu_to_be32(XFS_DIR3_BLOCK_MAGIC):
@@ -79,15 +132,8 @@ __xfs_dir3_data_check(
 		btp = xfs_dir2_block_tail_p(geo, hdr);
 		lep = xfs_dir2_block_leaf_p(btp);
 
-		/*
-		 * The number of leaf entries is limited by the size of the
-		 * block and the amount of space used by the data entries.
-		 * We don't know how much space is used by the data entries yet,
-		 * so just ensure that the count falls somewhere inside the
-		 * block right now.
-		 */
 		if (be32_to_cpu(btp->count) >=
-		    ((char *)btp - p) / sizeof(struct xfs_dir2_leaf_entry))
+		    xfs_dir2_data_max_leaf_entries(geo))
 			return __this_address;
 		break;
 	case cpu_to_be32(XFS_DIR3_DATA_MAGIC):
@@ -96,14 +142,14 @@ __xfs_dir3_data_check(
 	default:
 		return __this_address;
 	}
-	endp = xfs_dir3_data_endp(geo, hdr);
-	if (!endp)
+	end = xfs_dir3_data_end_offset(geo, hdr);
+	if (!end)
 		return __this_address;
 
 	/*
 	 * Account for zero bestfree entries.
 	 */
-	bf = ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(mp, hdr);
 	count = lastfree = freeseen = 0;
 	if (!bf[0].length) {
 		if (bf[0].offset)
@@ -128,8 +174,10 @@ __xfs_dir3_data_check(
 	/*
 	 * Loop over the data/unused entries.
 	 */
-	while (p < endp) {
-		dup = (xfs_dir2_data_unused_t *)p;
+	while (offset < end) {
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
+
 		/*
 		 * If it's unused, look for the space in the bestfree table.
 		 * If we find it, account for that, else make sure it
@@ -140,10 +188,10 @@ __xfs_dir3_data_check(
 
 			if (lastfree != 0)
 				return __this_address;
-			if (endp < p + be16_to_cpu(dup->length))
+			if (offset + be16_to_cpu(dup->length) > end)
 				return __this_address;
 			if (be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)) !=
-			    (char *)dup - (char *)hdr)
+			    offset)
 				return __this_address;
 			fa = xfs_dir2_data_freefind_verify(hdr, bf, dup, &dfp);
 			if (fa)
@@ -158,7 +206,7 @@ __xfs_dir3_data_check(
 				    be16_to_cpu(bf[2].length))
 					return __this_address;
 			}
-			p += be16_to_cpu(dup->length);
+			offset += be16_to_cpu(dup->length);
 			lastfree = 1;
 			continue;
 		}
@@ -168,17 +216,15 @@ __xfs_dir3_data_check(
 		 * in the leaf section of the block.
 		 * The linear search is crude but this is DEBUG code.
 		 */
-		dep = (xfs_dir2_data_entry_t *)p;
 		if (dep->namelen == 0)
 			return __this_address;
 		if (xfs_dir_ino_validate(mp, be64_to_cpu(dep->inumber)))
 			return __this_address;
-		if (endp < p + ops->data_entsize(dep->namelen))
+		if (offset + xfs_dir2_data_entsize(mp, dep->namelen) > end)
 			return __this_address;
-		if (be16_to_cpu(*ops->data_entry_tag_p(dep)) !=
-		    (char *)dep - (char *)hdr)
+		if (be16_to_cpu(*xfs_dir2_data_entry_tag_p(mp, dep)) != offset)
 			return __this_address;
-		if (ops->data_get_ftype(dep) >= XFS_DIR3_FT_MAX)
+		if (xfs_dir2_data_get_ftype(mp, dep) >= XFS_DIR3_FT_MAX)
 			return __this_address;
 		count++;
 		lastfree = 0;
@@ -189,7 +235,7 @@ __xfs_dir3_data_check(
 						((char *)dep - (char *)hdr));
 			name.name = dep->name;
 			name.len = dep->namelen;
-			hash = mp->m_dirnameops->hashname(&name);
+			hash = xfs_dir2_hashname(mp, &name);
 			for (i = 0; i < be32_to_cpu(btp->count); i++) {
 				if (be32_to_cpu(lep[i].address) == addr &&
 				    be32_to_cpu(lep[i].hashval) == hash)
@@ -198,7 +244,7 @@ __xfs_dir3_data_check(
 			if (i >= be32_to_cpu(btp->count))
 				return __this_address;
 		}
-		p += ops->data_entsize(dep->namelen);
+		offset += xfs_dir2_data_entsize(mp, dep->namelen);
 	}
 	/*
 	 * Need to have seen all the entries and all the bestfree slots.
@@ -348,21 +394,49 @@ static const struct xfs_buf_ops xfs_dir3_data_reada_buf_ops = {
 	.verify_write = xfs_dir3_data_write_verify,
 };
 
+static xfs_failaddr_t
+xfs_dir3_data_header_check(
+	struct xfs_inode	*dp,
+	struct xfs_buf		*bp)
+{
+	struct xfs_mount	*mp = dp->i_mount;
+
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_data_hdr *hdr3 = bp->b_addr;
+
+		if (be64_to_cpu(hdr3->hdr.owner) != dp->i_ino)
+			return __this_address;
+	}
+
+	return NULL;
+}
 
 int
 xfs_dir3_data_read(
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mapped_bno,
+	unsigned int		flags,
 	struct xfs_buf		**bpp)
 {
+	xfs_failaddr_t		fa;
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, bno, mapped_bno, bpp,
-				XFS_DATA_FORK, &xfs_dir3_data_buf_ops);
-	if (!err && tp && *bpp)
-		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_DATA_BUF);
+	err = xfs_da_read_buf(tp, dp, bno, flags, bpp, XFS_DATA_FORK,
+			&xfs_dir3_data_buf_ops);
+	if (err || !*bpp)
+		return err;
+
+	/* Check things that we can't do in the verifier. */
+	fa = xfs_dir3_data_header_check(dp, *bpp);
+	if (fa) {
+		__xfs_buf_mark_corrupt(*bpp, fa);
+		xfs_trans_brelse(tp, *bpp);
+		*bpp = NULL;
+		return -EFSCORRUPTED;
+	}
+
+	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_DATA_BUF);
 	return err;
 }
 
@@ -370,10 +444,10 @@ int
 xfs_dir3_data_readahead(
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
-	xfs_daddr_t		mapped_bno)
+	unsigned int		flags)
 {
-	return xfs_da_reada_buf(dp, bno, mapped_bno,
-				XFS_DATA_FORK, &xfs_dir3_data_reada_buf_ops);
+	return xfs_da_reada_buf(dp, bno, flags, XFS_DATA_FORK,
+				&xfs_dir3_data_reada_buf_ops);
 }
 
 /*
@@ -561,17 +635,16 @@ xfs_dir2_data_freeremove(
  * Given a data block, reconstruct its bestfree map.
  */
 void
-xfs_dir2_data_freescan_int(
-	struct xfs_da_geometry	*geo,
-	const struct xfs_dir_ops *ops,
-	struct xfs_dir2_data_hdr *hdr,
-	int			*loghead)
+xfs_dir2_data_freescan(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_data_hdr	*hdr,
+	int				*loghead)
 {
-	xfs_dir2_data_entry_t	*dep;		/* active data entry */
-	xfs_dir2_data_unused_t	*dup;		/* unused data entry */
-	struct xfs_dir2_data_free *bf;
-	char			*endp;		/* end of block's data */
-	char			*p;		/* current entry pointer */
+	struct xfs_da_geometry		*geo = mp->m_dir_geo;
+	struct xfs_dir2_data_free	*bf = xfs_dir2_data_bestfree_p(mp, hdr);
+	void				*addr = hdr;
+	unsigned int			offset = geo->data_entry_offset;
+	unsigned int			end;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
@@ -581,79 +654,60 @@ xfs_dir2_data_freescan_int(
 	/*
 	 * Start by clearing the table.
 	 */
-	bf = ops->data_bestfree_p(hdr);
 	memset(bf, 0, sizeof(*bf) * XFS_DIR2_DATA_FD_COUNT);
 	*loghead = 1;
-	/*
-	 * Set up pointers.
-	 */
-	p = (char *)ops->data_entry_p(hdr);
-	endp = xfs_dir3_data_endp(geo, hdr);
-	/*
-	 * Loop over the block's entries.
-	 */
-	while (p < endp) {
-		dup = (xfs_dir2_data_unused_t *)p;
+
+	end = xfs_dir3_data_end_offset(geo, addr);
+	while (offset < end) {
+		struct xfs_dir2_data_unused	*dup = addr + offset;
+		struct xfs_dir2_data_entry	*dep = addr + offset;
+
 		/*
 		 * If it's a free entry, insert it.
 		 */
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
-			ASSERT((char *)dup - (char *)hdr ==
+			ASSERT(offset ==
 			       be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)));
 			xfs_dir2_data_freeinsert(hdr, bf, dup, loghead);
-			p += be16_to_cpu(dup->length);
+			offset += be16_to_cpu(dup->length);
+			continue;
 		}
+
 		/*
 		 * For active entries, check their tags and skip them.
 		 */
-		else {
-			dep = (xfs_dir2_data_entry_t *)p;
-			ASSERT((char *)dep - (char *)hdr ==
-			       be16_to_cpu(*ops->data_entry_tag_p(dep)));
-			p += ops->data_entsize(dep->namelen);
-		}
+		ASSERT(offset ==
+		       be16_to_cpu(*xfs_dir2_data_entry_tag_p(mp, dep)));
+		offset += xfs_dir2_data_entsize(mp, dep->namelen);
 	}
 }
 
-void
-xfs_dir2_data_freescan(
-	struct xfs_inode	*dp,
-	struct xfs_dir2_data_hdr *hdr,
-	int			*loghead)
-{
-	return xfs_dir2_data_freescan_int(dp->i_mount->m_dir_geo, dp->d_ops,
-			hdr, loghead);
-}
-
 /*
  * Initialize a data block at the given block number in the directory.
  * Give back the buffer for the created block.
  */
 int						/* error */
 xfs_dir3_data_init(
-	xfs_da_args_t		*args,		/* directory operation args */
-	xfs_dir2_db_t		blkno,		/* logical dir block number */
-	struct xfs_buf		**bpp)		/* output block buffer */
+	struct xfs_da_args		*args,	/* directory operation args */
+	xfs_dir2_db_t			blkno,	/* logical dir block number */
+	struct xfs_buf			**bpp)	/* output block buffer */
 {
-	struct xfs_buf		*bp;		/* block buffer */
-	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
-	xfs_inode_t		*dp;		/* incore directory inode */
-	xfs_dir2_data_unused_t	*dup;		/* unused entry pointer */
-	struct xfs_dir2_data_free *bf;
-	int			error;		/* error return value */
-	int			i;		/* bestfree index */
-	xfs_mount_t		*mp;		/* filesystem mount point */
-	xfs_trans_t		*tp;		/* transaction pointer */
-	int                     t;              /* temp */
-
-	dp = args->dp;
-	mp = dp->i_mount;
-	tp = args->trans;
+	struct xfs_trans		*tp = args->trans;
+	struct xfs_inode		*dp = args->dp;
+	struct xfs_mount		*mp = dp->i_mount;
+	struct xfs_da_geometry		*geo = args->geo;
+	struct xfs_buf			*bp;
+	struct xfs_dir2_data_hdr	*hdr;
+	struct xfs_dir2_data_unused	*dup;
+	struct xfs_dir2_data_free 	*bf;
+	int				error;
+	int				i;
+
 	/*
 	 * Get the buffer set up for the block.
 	 */
 	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, blkno),
-			       -1, &bp, XFS_DATA_FORK);
+			       &bp, XFS_DATA_FORK);
 	if (error)
 		return error;
 	bp->b_ops = &xfs_dir3_data_buf_ops;
@@ -675,8 +729,9 @@ xfs_dir3_data_init(
 	} else
 		hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
 
-	bf = dp->d_ops->data_bestfree_p(hdr);
-	bf[0].offset = cpu_to_be16(dp->d_ops->data_entry_offset);
+	bf = xfs_dir2_data_bestfree_p(mp, hdr);
+	bf[0].offset = cpu_to_be16(geo->data_entry_offset);
+	bf[0].length = cpu_to_be16(geo->blksize - geo->data_entry_offset);
 	for (i = 1; i < XFS_DIR2_DATA_FD_COUNT; i++) {
 		bf[i].length = 0;
 		bf[i].offset = 0;
@@ -685,13 +740,11 @@ xfs_dir3_data_init(
 	/*
 	 * Set up an unused entry for the block's body.
 	 */
-	dup = dp->d_ops->data_unused_p(hdr);
+	dup = bp->b_addr + geo->data_entry_offset;
 	dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
-
-	t = args->geo->blksize - (uint)dp->d_ops->data_entry_offset;
-	bf[0].length = cpu_to_be16(t);
-	dup->length = cpu_to_be16(t);
+	dup->length = bf[0].length;
 	*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)hdr);
+
 	/*
 	 * Log it and return it.
 	 */
@@ -710,6 +763,7 @@ xfs_dir2_data_log_entry(
 	struct xfs_buf		*bp,
 	xfs_dir2_data_entry_t	*dep)		/* data entry pointer */
 {
+	struct xfs_mount	*mp = bp->b_mount;
 	struct xfs_dir2_data_hdr *hdr = bp->b_addr;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
@@ -718,7 +772,7 @@ xfs_dir2_data_log_entry(
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 
 	xfs_trans_log_buf(args->trans, bp, (uint)((char *)dep - (char *)hdr),
-		(uint)((char *)(args->dp->d_ops->data_entry_tag_p(dep) + 1) -
+		(uint)((char *)(xfs_dir2_data_entry_tag_p(mp, dep) + 1) -
 		       (char *)hdr - 1));
 }
 
@@ -739,8 +793,7 @@ xfs_dir2_data_log_header(
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 #endif
 
-	xfs_trans_log_buf(args->trans, bp, 0,
-			  args->dp->d_ops->data_entry_offset - 1);
+	xfs_trans_log_buf(args->trans, bp, 0, args->geo->data_entry_offset - 1);
 }
 
 /*
@@ -789,11 +842,11 @@ xfs_dir2_data_make_free(
 {
 	xfs_dir2_data_hdr_t	*hdr;		/* data block pointer */
 	xfs_dir2_data_free_t	*dfp;		/* bestfree pointer */
-	char			*endptr;	/* end of data area */
 	int			needscan;	/* need to regen bestfree */
 	xfs_dir2_data_unused_t	*newdup;	/* new unused entry */
 	xfs_dir2_data_unused_t	*postdup;	/* unused entry after us */
 	xfs_dir2_data_unused_t	*prevdup;	/* unused entry before us */
+	unsigned int		end;
 	struct xfs_dir2_data_free *bf;
 
 	hdr = bp->b_addr;
@@ -801,14 +854,14 @@ xfs_dir2_data_make_free(
 	/*
 	 * Figure out where the end of the data area is.
 	 */
-	endptr = xfs_dir3_data_endp(args->geo, hdr);
-	ASSERT(endptr != NULL);
+	end = xfs_dir3_data_end_offset(args->geo, hdr);
+	ASSERT(end != 0);
 
 	/*
 	 * If this isn't the start of the block, then back up to
 	 * the previous entry and see if it's free.
 	 */
-	if (offset > args->dp->d_ops->data_entry_offset) {
+	if (offset > args->geo->data_entry_offset) {
 		__be16			*tagp;	/* tag just before us */
 
 		tagp = (__be16 *)((char *)hdr + offset) - 1;
@@ -821,7 +874,7 @@ xfs_dir2_data_make_free(
 	 * If this isn't the end of the block, see if the entry after
 	 * us is free.
 	 */
-	if ((char *)hdr + offset + len < endptr) {
+	if (offset + len < end) {
 		postdup =
 			(xfs_dir2_data_unused_t *)((char *)hdr + offset + len);
 		if (be16_to_cpu(postdup->freetag) != XFS_DIR2_DATA_FREE_TAG)
@@ -834,7 +887,7 @@ xfs_dir2_data_make_free(
 	 * Previous and following entries are both free,
 	 * merge everything into a single free entry.
 	 */
-	bf = args->dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(args->dp->i_mount, hdr);
 	if (prevdup && postdup) {
 		xfs_dir2_data_free_t	*dfp2;	/* another bestfree pointer */
 
@@ -1025,7 +1078,7 @@ xfs_dir2_data_use_free(
 	 * Look up the entry in the bestfree table.
 	 */
 	oldlen = be16_to_cpu(dup->length);
-	bf = args->dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(args->dp->i_mount, hdr);
 	dfp = xfs_dir2_data_freefind(hdr, bf, dup);
 	ASSERT(dfp || oldlen <= be16_to_cpu(bf[2].length));
 	/*
@@ -1149,19 +1202,22 @@ corrupt:
 }
 
 /* Find the end of the entry data in a data/block format dir block. */
-void *
-xfs_dir3_data_endp(
+unsigned int
+xfs_dir3_data_end_offset(
 	struct xfs_da_geometry		*geo,
 	struct xfs_dir2_data_hdr	*hdr)
 {
+	void				*p;
+
 	switch (hdr->magic) {
 	case cpu_to_be32(XFS_DIR3_BLOCK_MAGIC):
 	case cpu_to_be32(XFS_DIR2_BLOCK_MAGIC):
-		return xfs_dir2_block_leaf_p(xfs_dir2_block_tail_p(geo, hdr));
+		p = xfs_dir2_block_leaf_p(xfs_dir2_block_tail_p(geo, hdr));
+		return p - (void *)hdr;
 	case cpu_to_be32(XFS_DIR3_DATA_MAGIC):
 	case cpu_to_be32(XFS_DIR2_DATA_MAGIC):
-		return (char *)hdr + geo->blksize;
+		return geo->blksize;
 	default:
-		return NULL;
+		return 0;
 	}
 }
diff --git a/fs/xfs/libxfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c
index a53e4585a2f3..95d2a3f92d75 100644
--- a/fs/xfs/libxfs/xfs_dir2_leaf.c
+++ b/fs/xfs/libxfs/xfs_dir2_leaf.c
@@ -24,12 +24,73 @@
  * Local function declarations.
  */
 static int xfs_dir2_leaf_lookup_int(xfs_da_args_t *args, struct xfs_buf **lbpp,
-				    int *indexp, struct xfs_buf **dbpp);
+				    int *indexp, struct xfs_buf **dbpp,
+				    struct xfs_dir3_icleaf_hdr *leafhdr);
 static void xfs_dir3_leaf_log_bests(struct xfs_da_args *args,
 				    struct xfs_buf *bp, int first, int last);
 static void xfs_dir3_leaf_log_tail(struct xfs_da_args *args,
 				   struct xfs_buf *bp);
 
+void
+xfs_dir2_leaf_hdr_from_disk(
+	struct xfs_mount		*mp,
+	struct xfs_dir3_icleaf_hdr	*to,
+	struct xfs_dir2_leaf		*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_leaf *from3 = (struct xfs_dir3_leaf *)from;
+
+		to->forw = be32_to_cpu(from3->hdr.info.hdr.forw);
+		to->back = be32_to_cpu(from3->hdr.info.hdr.back);
+		to->magic = be16_to_cpu(from3->hdr.info.hdr.magic);
+		to->count = be16_to_cpu(from3->hdr.count);
+		to->stale = be16_to_cpu(from3->hdr.stale);
+		to->ents = from3->__ents;
+
+		ASSERT(to->magic == XFS_DIR3_LEAF1_MAGIC ||
+		       to->magic == XFS_DIR3_LEAFN_MAGIC);
+	} else {
+		to->forw = be32_to_cpu(from->hdr.info.forw);
+		to->back = be32_to_cpu(from->hdr.info.back);
+		to->magic = be16_to_cpu(from->hdr.info.magic);
+		to->count = be16_to_cpu(from->hdr.count);
+		to->stale = be16_to_cpu(from->hdr.stale);
+		to->ents = from->__ents;
+
+		ASSERT(to->magic == XFS_DIR2_LEAF1_MAGIC ||
+		       to->magic == XFS_DIR2_LEAFN_MAGIC);
+	}
+}
+
+void
+xfs_dir2_leaf_hdr_to_disk(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_leaf		*to,
+	struct xfs_dir3_icleaf_hdr	*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_leaf *to3 = (struct xfs_dir3_leaf *)to;
+
+		ASSERT(from->magic == XFS_DIR3_LEAF1_MAGIC ||
+		       from->magic == XFS_DIR3_LEAFN_MAGIC);
+
+		to3->hdr.info.hdr.forw = cpu_to_be32(from->forw);
+		to3->hdr.info.hdr.back = cpu_to_be32(from->back);
+		to3->hdr.info.hdr.magic = cpu_to_be16(from->magic);
+		to3->hdr.count = cpu_to_be16(from->count);
+		to3->hdr.stale = cpu_to_be16(from->stale);
+	} else {
+		ASSERT(from->magic == XFS_DIR2_LEAF1_MAGIC ||
+		       from->magic == XFS_DIR2_LEAFN_MAGIC);
+
+		to->hdr.info.forw = cpu_to_be32(from->forw);
+		to->hdr.info.back = cpu_to_be32(from->back);
+		to->hdr.info.magic = cpu_to_be16(from->magic);
+		to->hdr.count = cpu_to_be16(from->count);
+		to->hdr.stale = cpu_to_be16(from->stale);
+	}
+}
+
 /*
  * Check the internal consistency of a leaf1 block.
  * Pop an assert if something is wrong.
@@ -43,7 +104,7 @@ xfs_dir3_leaf1_check(
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 
 	if (leafhdr.magic == XFS_DIR3_LEAF1_MAGIC) {
 		struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
@@ -52,7 +113,7 @@ xfs_dir3_leaf1_check(
 	} else if (leafhdr.magic != XFS_DIR2_LEAF1_MAGIC)
 		return __this_address;
 
-	return xfs_dir3_leaf_check_int(dp->i_mount, dp, &leafhdr, leaf);
+	return xfs_dir3_leaf_check_int(dp->i_mount, &leafhdr, leaf);
 }
 
 static inline void
@@ -76,31 +137,15 @@ xfs_dir3_leaf_check(
 
 xfs_failaddr_t
 xfs_dir3_leaf_check_int(
-	struct xfs_mount	*mp,
-	struct xfs_inode	*dp,
-	struct xfs_dir3_icleaf_hdr *hdr,
-	struct xfs_dir2_leaf	*leaf)
+	struct xfs_mount		*mp,
+	struct xfs_dir3_icleaf_hdr	*hdr,
+	struct xfs_dir2_leaf		*leaf)
 {
-	struct xfs_dir2_leaf_entry *ents;
-	xfs_dir2_leaf_tail_t	*ltp;
-	int			stale;
-	int			i;
-	const struct xfs_dir_ops *ops;
-	struct xfs_dir3_icleaf_hdr leafhdr;
-	struct xfs_da_geometry	*geo = mp->m_dir_geo;
-
-	/*
-	 * we can be passed a null dp here from a verifier, so we need to go the
-	 * hard way to get them.
-	 */
-	ops = xfs_dir_get_ops(mp, dp);
+	struct xfs_da_geometry		*geo = mp->m_dir_geo;
+	xfs_dir2_leaf_tail_t		*ltp;
+	int				stale;
+	int				i;
 
-	if (!hdr) {
-		ops->leaf_hdr_from_disk(&leafhdr, leaf);
-		hdr = &leafhdr;
-	}
-
-	ents = ops->leaf_ents_p(leaf);
 	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 
 	/*
@@ -108,23 +153,23 @@ xfs_dir3_leaf_check_int(
 	 * Should factor in the size of the bests table as well.
 	 * We can deduce a value for that from di_size.
 	 */
-	if (hdr->count > ops->leaf_max_ents(geo))
+	if (hdr->count > geo->leaf_max_ents)
 		return __this_address;
 
 	/* Leaves and bests don't overlap in leaf format. */
 	if ((hdr->magic == XFS_DIR2_LEAF1_MAGIC ||
 	     hdr->magic == XFS_DIR3_LEAF1_MAGIC) &&
-	    (char *)&ents[hdr->count] > (char *)xfs_dir2_leaf_bests_p(ltp))
+	    (char *)&hdr->ents[hdr->count] > (char *)xfs_dir2_leaf_bests_p(ltp))
 		return __this_address;
 
 	/* Check hash value order, count stale entries.  */
 	for (i = stale = 0; i < hdr->count; i++) {
 		if (i + 1 < hdr->count) {
-			if (be32_to_cpu(ents[i].hashval) >
-					be32_to_cpu(ents[i + 1].hashval))
+			if (be32_to_cpu(hdr->ents[i].hashval) >
+					be32_to_cpu(hdr->ents[i + 1].hashval))
 				return __this_address;
 		}
-		if (ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+		if (hdr->ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			stale++;
 	}
 	if (hdr->stale != stale)
@@ -139,17 +184,18 @@ xfs_dir3_leaf_check_int(
  */
 static xfs_failaddr_t
 xfs_dir3_leaf_verify(
-	struct xfs_buf		*bp)
+	struct xfs_buf			*bp)
 {
-	struct xfs_mount	*mp = bp->b_mount;
-	struct xfs_dir2_leaf	*leaf = bp->b_addr;
-	xfs_failaddr_t		fa;
+	struct xfs_mount		*mp = bp->b_mount;
+	struct xfs_dir3_icleaf_hdr	leafhdr;
+	xfs_failaddr_t			fa;
 
 	fa = xfs_da3_blkinfo_verify(bp, bp->b_addr);
 	if (fa)
 		return fa;
 
-	return xfs_dir3_leaf_check_int(mp, NULL, NULL, leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, bp->b_addr);
+	return xfs_dir3_leaf_check_int(mp, &leafhdr, bp->b_addr);
 }
 
 static void
@@ -216,13 +262,12 @@ xfs_dir3_leaf_read(
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		fbno,
-	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp)
 {
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
-				XFS_DATA_FORK, &xfs_dir3_leaf1_buf_ops);
+	err = xfs_da_read_buf(tp, dp, fbno, 0, bpp, XFS_DATA_FORK,
+			&xfs_dir3_leaf1_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAF1_BUF);
 	return err;
@@ -233,13 +278,12 @@ xfs_dir3_leafn_read(
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		fbno,
-	xfs_daddr_t		mappedbno,
 	struct xfs_buf		**bpp)
 {
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
-				XFS_DATA_FORK, &xfs_dir3_leafn_buf_ops);
+	err = xfs_da_read_buf(tp, dp, fbno, 0, bpp, XFS_DATA_FORK,
+			&xfs_dir3_leafn_buf_ops);
 	if (!err && tp && *bpp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAFN_BUF);
 	return err;
@@ -311,7 +355,7 @@ xfs_dir3_leaf_get_buf(
 	       bno < xfs_dir2_byte_to_db(args->geo, XFS_DIR2_FREE_OFFSET));
 
 	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, bno),
-			       -1, &bp, XFS_DATA_FORK);
+			       &bp, XFS_DATA_FORK);
 	if (error)
 		return error;
 
@@ -346,7 +390,6 @@ xfs_dir2_block_to_leaf(
 	int			needscan;	/* need to rescan bestfree */
 	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_dir2_data_free *bf;
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_block_to_leaf(args);
@@ -375,24 +418,24 @@ xfs_dir2_block_to_leaf(
 	xfs_dir3_data_check(dp, dbp);
 	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
-	bf = dp->d_ops->data_bestfree_p(hdr);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 
 	/*
 	 * Set the counts in the leaf header.
 	 */
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 	leafhdr.count = be32_to_cpu(btp->count);
 	leafhdr.stale = be32_to_cpu(btp->stale);
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, lbp);
 
 	/*
 	 * Could compact these but I think we always do the conversion
 	 * after squeezing out stale entries.
 	 */
-	memcpy(ents, blp, be32_to_cpu(btp->count) * sizeof(xfs_dir2_leaf_entry_t));
-	xfs_dir3_leaf_log_ents(args, lbp, 0, leafhdr.count - 1);
+	memcpy(leafhdr.ents, blp,
+		be32_to_cpu(btp->count) * sizeof(struct xfs_dir2_leaf_entry));
+	xfs_dir3_leaf_log_ents(args, &leafhdr, lbp, 0, leafhdr.count - 1);
 	needscan = 0;
 	needlog = 1;
 	/*
@@ -415,7 +458,7 @@ xfs_dir2_block_to_leaf(
 		hdr->magic = cpu_to_be32(XFS_DIR3_DATA_MAGIC);
 
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	/*
 	 * Set up leaf tail and bests table.
 	 */
@@ -594,7 +637,7 @@ xfs_dir2_leaf_addname(
 
 	trace_xfs_dir2_leaf_addname(args);
 
-	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, -1, &lbp);
+	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, &lbp);
 	if (error)
 		return error;
 
@@ -607,10 +650,10 @@ xfs_dir2_leaf_addname(
 	index = xfs_dir2_leaf_search_hash(args, lbp);
 	leaf = lbp->b_addr;
 	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
+	ents = leafhdr.ents;
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
-	length = dp->d_ops->data_entsize(args->namelen);
+	length = xfs_dir2_data_entsize(dp->i_mount, args->namelen);
 
 	/*
 	 * See if there are any entries with the same hash value
@@ -773,7 +816,7 @@ xfs_dir2_leaf_addname(
 		else
 			xfs_dir3_leaf_log_bests(args, lbp, use_block, use_block);
 		hdr = dbp->b_addr;
-		bf = dp->d_ops->data_bestfree_p(hdr);
+		bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 		bestsp[use_block] = bf[0].length;
 		grown = 1;
 	} else {
@@ -783,13 +826,13 @@ xfs_dir2_leaf_addname(
 		 */
 		error = xfs_dir3_data_read(tp, dp,
 				   xfs_dir2_db_to_da(args->geo, use_block),
-				   -1, &dbp);
+				   0, &dbp);
 		if (error) {
 			xfs_trans_brelse(tp, lbp);
 			return error;
 		}
 		hdr = dbp->b_addr;
-		bf = dp->d_ops->data_bestfree_p(hdr);
+		bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 		grown = 0;
 	}
 	/*
@@ -815,14 +858,14 @@ xfs_dir2_leaf_addname(
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, dep->namelen);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
+	tagp = xfs_dir2_data_entry_tag_p(dp->i_mount, dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
 	/*
 	 * Need to scan fix up the bestfree table.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	/*
 	 * Need to log the data block's header.
 	 */
@@ -852,9 +895,9 @@ xfs_dir2_leaf_addname(
 	/*
 	 * Log the leaf fields and give up the buffers.
 	 */
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, lbp);
-	xfs_dir3_leaf_log_ents(args, lbp, lfloglow, lfloghigh);
+	xfs_dir3_leaf_log_ents(args, &leafhdr, lbp, lfloglow, lfloghigh);
 	xfs_dir3_leaf_check(dp, lbp);
 	xfs_dir3_data_check(dp, dbp);
 	return 0;
@@ -874,7 +917,6 @@ xfs_dir3_leaf_compact(
 	xfs_dir2_leaf_t	*leaf;		/* leaf structure */
 	int		loglow;		/* first leaf entry to log */
 	int		to;		/* target leaf index */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_inode *dp = args->dp;
 
 	leaf = bp->b_addr;
@@ -884,9 +926,9 @@ xfs_dir3_leaf_compact(
 	/*
 	 * Compress out the stale entries in place.
 	 */
-	ents = dp->d_ops->leaf_ents_p(leaf);
 	for (from = to = 0, loglow = -1; from < leafhdr->count; from++) {
-		if (ents[from].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+		if (leafhdr->ents[from].address ==
+		    cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			continue;
 		/*
 		 * Only actually copy the entries that are different.
@@ -894,7 +936,7 @@ xfs_dir3_leaf_compact(
 		if (from > to) {
 			if (loglow == -1)
 				loglow = to;
-			ents[to] = ents[from];
+			leafhdr->ents[to] = leafhdr->ents[from];
 		}
 		to++;
 	}
@@ -905,10 +947,10 @@ xfs_dir3_leaf_compact(
 	leafhdr->count -= leafhdr->stale;
 	leafhdr->stale = 0;
 
-	dp->d_ops->leaf_hdr_to_disk(leaf, leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, leafhdr);
 	xfs_dir3_leaf_log_header(args, bp);
 	if (loglow != -1)
-		xfs_dir3_leaf_log_ents(args, bp, loglow, to - 1);
+		xfs_dir3_leaf_log_ents(args, leafhdr, bp, loglow, to - 1);
 }
 
 /*
@@ -1037,6 +1079,7 @@ xfs_dir3_leaf_log_bests(
 void
 xfs_dir3_leaf_log_ents(
 	struct xfs_da_args	*args,
+	struct xfs_dir3_icleaf_hdr *hdr,
 	struct xfs_buf		*bp,
 	int			first,
 	int			last)
@@ -1044,16 +1087,14 @@ xfs_dir3_leaf_log_ents(
 	xfs_dir2_leaf_entry_t	*firstlep;	/* pointer to first entry */
 	xfs_dir2_leaf_entry_t	*lastlep;	/* pointer to last entry */
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
-	struct xfs_dir2_leaf_entry *ents;
 
 	ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
 
-	ents = args->dp->d_ops->leaf_ents_p(leaf);
-	firstlep = &ents[first];
-	lastlep = &ents[last];
+	firstlep = &hdr->ents[first];
+	lastlep = &hdr->ents[last];
 	xfs_trans_log_buf(args->trans, bp,
 		(uint)((char *)firstlep - (char *)leaf),
 		(uint)((char *)lastlep - (char *)leaf + sizeof(*lastlep) - 1));
@@ -1076,7 +1117,7 @@ xfs_dir3_leaf_log_header(
 
 	xfs_trans_log_buf(args->trans, bp,
 			  (uint)((char *)&leaf->hdr - (char *)leaf),
-			  args->dp->d_ops->leaf_hdr_size - 1);
+			  args->geo->leaf_hdr_size - 1);
 }
 
 /*
@@ -1115,28 +1156,27 @@ xfs_dir2_leaf_lookup(
 	int			error;		/* error return code */
 	int			index;		/* found entry index */
 	struct xfs_buf		*lbp;		/* leaf buffer */
-	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
 	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
-	struct xfs_dir2_leaf_entry *ents;
+	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_leaf_lookup(args);
 
 	/*
 	 * Look up name in the leaf block, returning both buffers and index.
 	 */
-	if ((error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp))) {
+	error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp, &leafhdr);
+	if (error)
 		return error;
-	}
+
 	tp = args->trans;
 	dp = args->dp;
 	xfs_dir3_leaf_check(dp, lbp);
-	leaf = lbp->b_addr;
-	ents = dp->d_ops->leaf_ents_p(leaf);
+
 	/*
 	 * Get to the leaf entry and contained data entry address.
 	 */
-	lep = &ents[index];
+	lep = &leafhdr.ents[index];
 
 	/*
 	 * Point to the data entry.
@@ -1148,7 +1188,7 @@ xfs_dir2_leaf_lookup(
 	 * Return the found inode number & CI name if appropriate
 	 */
 	args->inumber = be64_to_cpu(dep->inumber);
-	args->filetype = dp->d_ops->data_get_ftype(dep);
+	args->filetype = xfs_dir2_data_get_ftype(dp->i_mount, dep);
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(tp, dbp);
 	xfs_trans_brelse(tp, lbp);
@@ -1166,7 +1206,8 @@ xfs_dir2_leaf_lookup_int(
 	xfs_da_args_t		*args,		/* operation arguments */
 	struct xfs_buf		**lbpp,		/* out: leaf buffer */
 	int			*indexp,	/* out: index in leaf block */
-	struct xfs_buf		**dbpp)		/* out: data buffer */
+	struct xfs_buf		**dbpp,		/* out: data buffer */
+	struct xfs_dir3_icleaf_hdr *leafhdr)
 {
 	xfs_dir2_db_t		curdb = -1;	/* current data block number */
 	struct xfs_buf		*dbp = NULL;	/* data buffer */
@@ -1182,22 +1223,19 @@ xfs_dir2_leaf_lookup_int(
 	xfs_trans_t		*tp;		/* transaction pointer */
 	xfs_dir2_db_t		cidb = -1;	/* case match data block no. */
 	enum xfs_dacmp		cmp;		/* name compare result */
-	struct xfs_dir2_leaf_entry *ents;
-	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
 
-	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, -1, &lbp);
+	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, &lbp);
 	if (error)
 		return error;
 
 	*lbpp = lbp;
 	leaf = lbp->b_addr;
 	xfs_dir3_leaf_check(dp, lbp);
-	ents = dp->d_ops->leaf_ents_p(leaf);
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, leafhdr, leaf);
 
 	/*
 	 * Look for the first leaf entry with our hash value.
@@ -1207,8 +1245,9 @@ xfs_dir2_leaf_lookup_int(
 	 * Loop over all the entries with the right hash value
 	 * looking to match the name.
 	 */
-	for (lep = &ents[index];
-	     index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
+	for (lep = &leafhdr->ents[index];
+	     index < leafhdr->count &&
+			be32_to_cpu(lep->hashval) == args->hashval;
 	     lep++, index++) {
 		/*
 		 * Skip over stale leaf entries.
@@ -1229,7 +1268,7 @@ xfs_dir2_leaf_lookup_int(
 				xfs_trans_brelse(tp, dbp);
 			error = xfs_dir3_data_read(tp, dp,
 					   xfs_dir2_db_to_da(args->geo, newdb),
-					   -1, &dbp);
+					   0, &dbp);
 			if (error) {
 				xfs_trans_brelse(tp, lbp);
 				return error;
@@ -1247,7 +1286,7 @@ xfs_dir2_leaf_lookup_int(
 		 * and buffer. If it's the first case-insensitive match, store
 		 * the index and buffer and continue looking for an exact match.
 		 */
-		cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+		cmp = xfs_dir2_compname(args, dep->name, dep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			args->cmpresult = cmp;
 			*indexp = index;
@@ -1271,7 +1310,7 @@ xfs_dir2_leaf_lookup_int(
 			xfs_trans_brelse(tp, dbp);
 			error = xfs_dir3_data_read(tp, dp,
 					   xfs_dir2_db_to_da(args->geo, cidb),
-					   -1, &dbp);
+					   0, &dbp);
 			if (error) {
 				xfs_trans_brelse(tp, lbp);
 				return error;
@@ -1297,6 +1336,7 @@ int						/* error */
 xfs_dir2_leaf_removename(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
+	struct xfs_da_geometry	*geo = args->geo;
 	__be16			*bestsp;	/* leaf block best freespace */
 	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
 	xfs_dir2_db_t		db;		/* data block number */
@@ -1314,7 +1354,6 @@ xfs_dir2_leaf_removename(
 	int			needscan;	/* need to rescan data frees */
 	xfs_dir2_data_off_t	oldbest;	/* old value of best free */
 	struct xfs_dir2_data_free *bf;		/* bestfree table */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_leaf_removename(args);
@@ -1322,51 +1361,54 @@ xfs_dir2_leaf_removename(
 	/*
 	 * Lookup the leaf entry, get the leaf and data blocks read in.
 	 */
-	if ((error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp))) {
+	error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp, &leafhdr);
+	if (error)
 		return error;
-	}
+
 	dp = args->dp;
 	leaf = lbp->b_addr;
 	hdr = dbp->b_addr;
 	xfs_dir3_data_check(dp, dbp);
-	bf = dp->d_ops->data_bestfree_p(hdr);
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
+
 	/*
 	 * Point to the leaf entry, use that to point to the data entry.
 	 */
-	lep = &ents[index];
-	db = xfs_dir2_dataptr_to_db(args->geo, be32_to_cpu(lep->address));
+	lep = &leafhdr.ents[index];
+	db = xfs_dir2_dataptr_to_db(geo, be32_to_cpu(lep->address));
 	dep = (xfs_dir2_data_entry_t *)((char *)hdr +
-		xfs_dir2_dataptr_to_off(args->geo, be32_to_cpu(lep->address)));
+		xfs_dir2_dataptr_to_off(geo, be32_to_cpu(lep->address)));
 	needscan = needlog = 0;
 	oldbest = be16_to_cpu(bf[0].length);
-	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
+	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
-	if (be16_to_cpu(bestsp[db]) != oldbest)
+	if (be16_to_cpu(bestsp[db]) != oldbest) {
+		xfs_buf_mark_corrupt(lbp);
 		return -EFSCORRUPTED;
+	}
 	/*
 	 * Mark the former data entry unused.
 	 */
 	xfs_dir2_data_make_free(args, dbp,
 		(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
-		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
+		xfs_dir2_data_entsize(dp->i_mount, dep->namelen), &needlog,
+		&needscan);
 	/*
 	 * We just mark the leaf entry stale by putting a null in it.
 	 */
 	leafhdr.stale++;
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, lbp);
 
 	lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
-	xfs_dir3_leaf_log_ents(args, lbp, index, index);
+	xfs_dir3_leaf_log_ents(args, &leafhdr, lbp, index, index);
 
 	/*
 	 * Scan the freespace in the data block again if necessary,
 	 * log the data block header if necessary.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, dbp);
 	/*
@@ -1382,8 +1424,8 @@ xfs_dir2_leaf_removename(
 	 * If the data block is now empty then get rid of the data block.
 	 */
 	if (be16_to_cpu(bf[0].length) ==
-			args->geo->blksize - dp->d_ops->data_entry_offset) {
-		ASSERT(db != args->geo->datablk);
+	    geo->blksize - geo->data_entry_offset) {
+		ASSERT(db != geo->datablk);
 		if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
 			/*
 			 * Nope, can't get rid of it because it caused
@@ -1425,7 +1467,7 @@ xfs_dir2_leaf_removename(
 	/*
 	 * If the data block was not the first one, drop it.
 	 */
-	else if (db != args->geo->datablk)
+	else if (db != geo->datablk)
 		dbp = NULL;
 
 	xfs_dir3_leaf_check(dp, lbp);
@@ -1448,26 +1490,24 @@ xfs_dir2_leaf_replace(
 	int			error;		/* error return code */
 	int			index;		/* index of leaf entry */
 	struct xfs_buf		*lbp;		/* leaf buffer */
-	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
 	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
-	struct xfs_dir2_leaf_entry *ents;
+	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	trace_xfs_dir2_leaf_replace(args);
 
 	/*
 	 * Look up the entry.
 	 */
-	if ((error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp))) {
+	error = xfs_dir2_leaf_lookup_int(args, &lbp, &index, &dbp, &leafhdr);
+	if (error)
 		return error;
-	}
+
 	dp = args->dp;
-	leaf = lbp->b_addr;
-	ents = dp->d_ops->leaf_ents_p(leaf);
 	/*
 	 * Point to the leaf entry, get data address from it.
 	 */
-	lep = &ents[index];
+	lep = &leafhdr.ents[index];
 	/*
 	 * Point to the data entry.
 	 */
@@ -1479,7 +1519,7 @@ xfs_dir2_leaf_replace(
 	 * Put the new inode number in, log it.
 	 */
 	dep->inumber = cpu_to_be64(args->inumber);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
 	tp = args->trans;
 	xfs_dir2_data_log_entry(args, dbp, dep);
 	xfs_dir3_leaf_check(dp, lbp);
@@ -1501,21 +1541,17 @@ xfs_dir2_leaf_search_hash(
 	xfs_dahash_t		hashwant;	/* hash value looking for */
 	int			high;		/* high leaf index */
 	int			low;		/* low leaf index */
-	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
 	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry */
 	int			mid=0;		/* current leaf index */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	leaf = lbp->b_addr;
-	ents = args->dp->d_ops->leaf_ents_p(leaf);
-	args->dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(args->dp->i_mount, &leafhdr, lbp->b_addr);
 
 	/*
 	 * Note, the table cannot be empty, so we have to go through the loop.
 	 * Binary search the leaf entries looking for our hash value.
 	 */
-	for (lep = ents, low = 0, high = leafhdr.count - 1,
+	for (lep = leafhdr.ents, low = 0, high = leafhdr.count - 1,
 		hashwant = args->hashval;
 	     low <= high; ) {
 		mid = (low + high) >> 1;
@@ -1552,6 +1588,7 @@ xfs_dir2_leaf_trim_data(
 	struct xfs_buf		*lbp,		/* leaf buffer */
 	xfs_dir2_db_t		db)		/* data block number */
 {
+	struct xfs_da_geometry	*geo = args->geo;
 	__be16			*bestsp;	/* leaf bests table */
 	struct xfs_buf		*dbp;		/* data block buffer */
 	xfs_inode_t		*dp;		/* incore directory inode */
@@ -1565,23 +1602,23 @@ xfs_dir2_leaf_trim_data(
 	/*
 	 * Read the offending data block.  We need its buffer.
 	 */
-	error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(args->geo, db),
-				   -1, &dbp);
+	error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(geo, db), 0, &dbp);
 	if (error)
 		return error;
 
 	leaf = lbp->b_addr;
-	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
+	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 
 #ifdef DEBUG
 {
 	struct xfs_dir2_data_hdr *hdr = dbp->b_addr;
-	struct xfs_dir2_data_free *bf = dp->d_ops->data_bestfree_p(hdr);
+	struct xfs_dir2_data_free *bf =
+		xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC));
 	ASSERT(be16_to_cpu(bf[0].length) ==
-	       args->geo->blksize - dp->d_ops->data_entry_offset);
+	       geo->blksize - geo->data_entry_offset);
 	ASSERT(db == be32_to_cpu(ltp->bestcount) - 1);
 }
 #endif
@@ -1639,7 +1676,6 @@ xfs_dir2_node_to_leaf(
 	int			error;		/* error return code */
 	struct xfs_buf		*fbp;		/* buffer for freespace block */
 	xfs_fileoff_t		fo;		/* freespace file offset */
-	xfs_dir2_free_t		*free;		/* freespace structure */
 	struct xfs_buf		*lbp;		/* buffer for leaf block */
 	xfs_dir2_leaf_tail_t	*ltp;		/* tail of leaf structure */
 	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
@@ -1697,7 +1733,7 @@ xfs_dir2_node_to_leaf(
 		return 0;
 	lbp = state->path.blk[0].bp;
 	leaf = lbp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 	       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
@@ -1708,8 +1744,7 @@ xfs_dir2_node_to_leaf(
 	error = xfs_dir2_free_read(tp, dp,  args->geo->freeblk, &fbp);
 	if (error)
 		return error;
-	free = fbp->b_addr;
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
+	xfs_dir2_free_hdr_from_disk(mp, &freehdr, fbp->b_addr);
 
 	ASSERT(!freehdr.firstdb);
 
@@ -1743,10 +1778,10 @@ xfs_dir2_node_to_leaf(
 	/*
 	 * Set up the leaf bests table.
 	 */
-	memcpy(xfs_dir2_leaf_bests_p(ltp), dp->d_ops->free_bests_p(free),
+	memcpy(xfs_dir2_leaf_bests_p(ltp), freehdr.bests,
 		freehdr.nvalid * sizeof(xfs_dir2_data_off_t));
 
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(mp, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, lbp);
 	xfs_dir3_leaf_log_bests(args, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
 	xfs_dir3_leaf_log_tail(args, lbp);
diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c
index 705c4f562758..6ac4aad98cd7 100644
--- a/fs/xfs/libxfs/xfs_dir2_node.c
+++ b/fs/xfs/libxfs/xfs_dir2_node.c
@@ -34,6 +34,25 @@ static int xfs_dir2_leafn_remove(xfs_da_args_t *args, struct xfs_buf *bp,
 				 int *rval);
 
 /*
+ * Convert data space db to the corresponding free db.
+ */
+static xfs_dir2_db_t
+xfs_dir2_db_to_fdb(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
+{
+	return xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET) +
+			(db / geo->free_max_bests);
+}
+
+/*
+ * Convert data space db to the corresponding index in a free db.
+ */
+static int
+xfs_dir2_db_to_fdindex(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
+{
+	return db % geo->free_max_bests;
+}
+
+/*
  * Check internal consistency of a leafn block.
  */
 #ifdef DEBUG
@@ -45,7 +64,7 @@ xfs_dir3_leafn_check(
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 
 	if (leafhdr.magic == XFS_DIR3_LEAFN_MAGIC) {
 		struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
@@ -54,7 +73,7 @@ xfs_dir3_leafn_check(
 	} else if (leafhdr.magic != XFS_DIR2_LEAFN_MAGIC)
 		return __this_address;
 
-	return xfs_dir3_leaf_check_int(dp->i_mount, dp, &leafhdr, leaf);
+	return xfs_dir3_leaf_check_int(dp->i_mount, &leafhdr, leaf);
 }
 
 static inline void
@@ -160,10 +179,9 @@ xfs_dir3_free_header_check(
 	struct xfs_buf		*bp)
 {
 	struct xfs_mount	*mp = dp->i_mount;
+	int			maxbests = mp->m_dir_geo->free_max_bests;
 	unsigned int		firstdb;
-	int			maxbests;
 
-	maxbests = dp->d_ops->free_max_bests(mp->m_dir_geo);
 	firstdb = (xfs_dir2_da_to_db(mp->m_dir_geo, fbno) -
 		   xfs_dir2_byte_to_db(mp->m_dir_geo, XFS_DIR2_FREE_OFFSET)) *
 			maxbests;
@@ -176,6 +194,8 @@ xfs_dir3_free_header_check(
 			return __this_address;
 		if (be32_to_cpu(hdr3->nvalid) < be32_to_cpu(hdr3->nused))
 			return __this_address;
+		if (be64_to_cpu(hdr3->hdr.owner) != dp->i_ino)
+			return __this_address;
 	} else {
 		struct xfs_dir2_free_hdr *hdr = bp->b_addr;
 
@@ -194,22 +214,23 @@ __xfs_dir3_free_read(
 	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		fbno,
-	xfs_daddr_t		mappedbno,
+	unsigned int		flags,
 	struct xfs_buf		**bpp)
 {
 	xfs_failaddr_t		fa;
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
-				XFS_DATA_FORK, &xfs_dir3_free_buf_ops);
+	err = xfs_da_read_buf(tp, dp, fbno, flags, bpp, XFS_DATA_FORK,
+			&xfs_dir3_free_buf_ops);
 	if (err || !*bpp)
 		return err;
 
 	/* Check things that we can't do in the verifier. */
 	fa = xfs_dir3_free_header_check(dp, fbno, *bpp);
 	if (fa) {
-		xfs_verifier_error(*bpp, -EFSCORRUPTED, fa);
+		__xfs_buf_mark_corrupt(*bpp, fa);
 		xfs_trans_brelse(tp, *bpp);
+		*bpp = NULL;
 		return -EFSCORRUPTED;
 	}
 
@@ -220,6 +241,58 @@ __xfs_dir3_free_read(
 	return 0;
 }
 
+void
+xfs_dir2_free_hdr_from_disk(
+	struct xfs_mount		*mp,
+	struct xfs_dir3_icfree_hdr	*to,
+	struct xfs_dir2_free		*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_free	*from3 = (struct xfs_dir3_free *)from;
+
+		to->magic = be32_to_cpu(from3->hdr.hdr.magic);
+		to->firstdb = be32_to_cpu(from3->hdr.firstdb);
+		to->nvalid = be32_to_cpu(from3->hdr.nvalid);
+		to->nused = be32_to_cpu(from3->hdr.nused);
+		to->bests = from3->bests;
+
+		ASSERT(to->magic == XFS_DIR3_FREE_MAGIC);
+	} else {
+		to->magic = be32_to_cpu(from->hdr.magic);
+		to->firstdb = be32_to_cpu(from->hdr.firstdb);
+		to->nvalid = be32_to_cpu(from->hdr.nvalid);
+		to->nused = be32_to_cpu(from->hdr.nused);
+		to->bests = from->bests;
+
+		ASSERT(to->magic == XFS_DIR2_FREE_MAGIC);
+	}
+}
+
+static void
+xfs_dir2_free_hdr_to_disk(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_free		*to,
+	struct xfs_dir3_icfree_hdr	*from)
+{
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		struct xfs_dir3_free	*to3 = (struct xfs_dir3_free *)to;
+
+		ASSERT(from->magic == XFS_DIR3_FREE_MAGIC);
+
+		to3->hdr.hdr.magic = cpu_to_be32(from->magic);
+		to3->hdr.firstdb = cpu_to_be32(from->firstdb);
+		to3->hdr.nvalid = cpu_to_be32(from->nvalid);
+		to3->hdr.nused = cpu_to_be32(from->nused);
+	} else {
+		ASSERT(from->magic == XFS_DIR2_FREE_MAGIC);
+
+		to->hdr.magic = cpu_to_be32(from->magic);
+		to->hdr.firstdb = cpu_to_be32(from->firstdb);
+		to->hdr.nvalid = cpu_to_be32(from->nvalid);
+		to->hdr.nused = cpu_to_be32(from->nused);
+	}
+}
+
 int
 xfs_dir2_free_read(
 	struct xfs_trans	*tp,
@@ -227,7 +300,7 @@ xfs_dir2_free_read(
 	xfs_dablk_t		fbno,
 	struct xfs_buf		**bpp)
 {
-	return __xfs_dir3_free_read(tp, dp, fbno, -1, bpp);
+	return __xfs_dir3_free_read(tp, dp, fbno, 0, bpp);
 }
 
 static int
@@ -237,7 +310,7 @@ xfs_dir2_free_try_read(
 	xfs_dablk_t		fbno,
 	struct xfs_buf		**bpp)
 {
-	return __xfs_dir3_free_read(tp, dp, fbno, -2, bpp);
+	return __xfs_dir3_free_read(tp, dp, fbno, XFS_DABUF_MAP_HOLE_OK, bpp);
 }
 
 static int
@@ -254,7 +327,7 @@ xfs_dir3_free_get_buf(
 	struct xfs_dir3_icfree_hdr hdr;
 
 	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, fbno),
-				   -1, &bp, XFS_DATA_FORK);
+			&bp, XFS_DATA_FORK);
 	if (error)
 		return error;
 
@@ -278,7 +351,7 @@ xfs_dir3_free_get_buf(
 		uuid_copy(&hdr3->hdr.uuid, &mp->m_sb.sb_meta_uuid);
 	} else
 		hdr.magic = XFS_DIR2_FREE_MAGIC;
-	dp->d_ops->free_hdr_to_disk(bp->b_addr, &hdr);
+	xfs_dir2_free_hdr_to_disk(mp, bp->b_addr, &hdr);
 	*bpp = bp;
 	return 0;
 }
@@ -289,21 +362,19 @@ xfs_dir3_free_get_buf(
 STATIC void
 xfs_dir2_free_log_bests(
 	struct xfs_da_args	*args,
+	struct xfs_dir3_icfree_hdr *hdr,
 	struct xfs_buf		*bp,
 	int			first,		/* first entry to log */
 	int			last)		/* last entry to log */
 {
-	xfs_dir2_free_t		*free;		/* freespace structure */
-	__be16			*bests;
+	struct xfs_dir2_free	*free = bp->b_addr;
 
-	free = bp->b_addr;
-	bests = args->dp->d_ops->free_bests_p(free);
 	ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 	       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 	xfs_trans_log_buf(args->trans, bp,
-		(uint)((char *)&bests[first] - (char *)free),
-		(uint)((char *)&bests[last] - (char *)free +
-		       sizeof(bests[0]) - 1));
+			  (char *)&hdr->bests[first] - (char *)free,
+			  (char *)&hdr->bests[last] - (char *)free +
+			   sizeof(hdr->bests[0]) - 1);
 }
 
 /*
@@ -322,7 +393,7 @@ xfs_dir2_free_log_header(
 	       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 #endif
 	xfs_trans_log_buf(args->trans, bp, 0,
-			  args->dp->d_ops->free_hdr_size - 1);
+			  args->geo->free_hdr_size - 1);
 }
 
 /*
@@ -339,14 +410,12 @@ xfs_dir2_leaf_to_node(
 	int			error;		/* error return value */
 	struct xfs_buf		*fbp;		/* freespace buffer */
 	xfs_dir2_db_t		fdb;		/* freespace block number */
-	xfs_dir2_free_t		*free;		/* freespace structure */
 	__be16			*from;		/* pointer to freespace entry */
 	int			i;		/* leaf freespace index */
 	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
 	xfs_dir2_leaf_tail_t	*ltp;		/* leaf tail structure */
 	int			n;		/* count of live freespc ents */
 	xfs_dir2_data_off_t	off;		/* freespace entry value */
-	__be16			*to;		/* pointer to freespace entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_dir3_icfree_hdr freehdr;
 
@@ -368,24 +437,25 @@ xfs_dir2_leaf_to_node(
 	if (error)
 		return error;
 
-	free = fbp->b_addr;
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
+	xfs_dir2_free_hdr_from_disk(dp->i_mount, &freehdr, fbp->b_addr);
 	leaf = lbp->b_addr;
 	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 	if (be32_to_cpu(ltp->bestcount) >
-				(uint)dp->i_d.di_size / args->geo->blksize)
+				(uint)dp->i_d.di_size / args->geo->blksize) {
+		xfs_buf_mark_corrupt(lbp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Copy freespace entries from the leaf block to the new block.
 	 * Count active entries.
 	 */
 	from = xfs_dir2_leaf_bests_p(ltp);
-	to = dp->d_ops->free_bests_p(free);
-	for (i = n = 0; i < be32_to_cpu(ltp->bestcount); i++, from++, to++) {
-		if ((off = be16_to_cpu(*from)) != NULLDATAOFF)
+	for (i = n = 0; i < be32_to_cpu(ltp->bestcount); i++, from++) {
+		off = be16_to_cpu(*from);
+		if (off != NULLDATAOFF)
 			n++;
-		*to = cpu_to_be16(off);
+		freehdr.bests[i] = cpu_to_be16(off);
 	}
 
 	/*
@@ -394,8 +464,8 @@ xfs_dir2_leaf_to_node(
 	freehdr.nused = n;
 	freehdr.nvalid = be32_to_cpu(ltp->bestcount);
 
-	dp->d_ops->free_hdr_to_disk(fbp->b_addr, &freehdr);
-	xfs_dir2_free_log_bests(args, fbp, 0, freehdr.nvalid - 1);
+	xfs_dir2_free_hdr_to_disk(dp->i_mount, fbp->b_addr, &freehdr);
+	xfs_dir2_free_log_bests(args, &freehdr, fbp, 0, freehdr.nvalid - 1);
 	xfs_dir2_free_log_header(args, fbp);
 
 	/*
@@ -438,15 +508,17 @@ xfs_dir2_leafn_add(
 
 	trace_xfs_dir2_leafn_add(args, index);
 
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
+	ents = leafhdr.ents;
 
 	/*
 	 * Quick check just to make sure we are not going to index
 	 * into other peoples memory
 	 */
-	if (index < 0)
+	if (index < 0) {
+		xfs_buf_mark_corrupt(bp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * If there are already the maximum number of leaf entries in
@@ -455,7 +527,7 @@ xfs_dir2_leafn_add(
 	 * a compact.
 	 */
 
-	if (leafhdr.count == dp->d_ops->leaf_max_ents(args->geo)) {
+	if (leafhdr.count == args->geo->leaf_max_ents) {
 		if (!leafhdr.stale)
 			return -ENOSPC;
 		compact = leafhdr.stale > 1;
@@ -493,9 +565,9 @@ xfs_dir2_leafn_add(
 	lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(args->geo,
 				args->blkno, args->index));
 
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, bp);
-	xfs_dir3_leaf_log_ents(args, bp, lfloglow, lfloghigh);
+	xfs_dir3_leaf_log_ents(args, &leafhdr, bp, lfloglow, lfloghigh);
 	xfs_dir3_leaf_check(dp, bp);
 	return 0;
 }
@@ -509,10 +581,9 @@ xfs_dir2_free_hdr_check(
 {
 	struct xfs_dir3_icfree_hdr hdr;
 
-	dp->d_ops->free_hdr_from_disk(&hdr, bp->b_addr);
+	xfs_dir2_free_hdr_from_disk(dp->i_mount, &hdr, bp->b_addr);
 
-	ASSERT((hdr.firstdb %
-		dp->d_ops->free_max_bests(dp->i_mount->m_dir_geo)) == 0);
+	ASSERT((hdr.firstdb % dp->i_mount->m_dir_geo->free_max_bests) == 0);
 	ASSERT(hdr.firstdb <= db);
 	ASSERT(db < hdr.firstdb + hdr.nvalid);
 }
@@ -530,11 +601,9 @@ xfs_dir2_leaf_lasthash(
 	struct xfs_buf	*bp,			/* leaf buffer */
 	int		*count)			/* count of entries in leaf */
 {
-	struct xfs_dir2_leaf	*leaf = bp->b_addr;
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, bp->b_addr);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 	       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC ||
@@ -545,9 +614,7 @@ xfs_dir2_leaf_lasthash(
 		*count = leafhdr.count;
 	if (!leafhdr.count)
 		return 0;
-
-	ents = dp->d_ops->leaf_ents_p(leaf);
-	return be32_to_cpu(ents[leafhdr.count - 1].hashval);
+	return be32_to_cpu(leafhdr.ents[leafhdr.count - 1].hashval);
 }
 
 /*
@@ -576,15 +643,13 @@ xfs_dir2_leafn_lookup_for_addname(
 	xfs_dir2_db_t		newdb;		/* new data block number */
 	xfs_dir2_db_t		newfdb;		/* new free block number */
 	xfs_trans_t		*tp;		/* transaction pointer */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
 
 	xfs_dir3_leaf_check(dp, bp);
 	ASSERT(leafhdr.count > 0);
@@ -604,11 +669,11 @@ xfs_dir2_leafn_lookup_for_addname(
 		ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 		       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 	}
-	length = dp->d_ops->data_entsize(args->namelen);
+	length = xfs_dir2_data_entsize(mp, args->namelen);
 	/*
 	 * Loop over leaf entries with the right hash value.
 	 */
-	for (lep = &ents[index];
+	for (lep = &leafhdr.ents[index];
 	     index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
 	     lep++, index++) {
 		/*
@@ -630,14 +695,14 @@ xfs_dir2_leafn_lookup_for_addname(
 		 * in hand, take a look at it.
 		 */
 		if (newdb != curdb) {
-			__be16 *bests;
+			struct xfs_dir3_icfree_hdr freehdr;
 
 			curdb = newdb;
 			/*
 			 * Convert the data block to the free block
 			 * holding its freespace information.
 			 */
-			newfdb = dp->d_ops->db_to_fdb(args->geo, newdb);
+			newfdb = xfs_dir2_db_to_fdb(args->geo, newdb);
 			/*
 			 * If it's not the one we have in hand, read it in.
 			 */
@@ -661,20 +726,20 @@ xfs_dir2_leafn_lookup_for_addname(
 			/*
 			 * Get the index for our entry.
 			 */
-			fi = dp->d_ops->db_to_fdindex(args->geo, curdb);
+			fi = xfs_dir2_db_to_fdindex(args->geo, curdb);
 			/*
 			 * If it has room, return it.
 			 */
-			bests = dp->d_ops->free_bests_p(free);
-			if (unlikely(bests[fi] == cpu_to_be16(NULLDATAOFF))) {
-				XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int",
-							XFS_ERRLEVEL_LOW, mp);
+			xfs_dir2_free_hdr_from_disk(mp, &freehdr, free);
+			if (XFS_IS_CORRUPT(mp,
+					   freehdr.bests[fi] ==
+					   cpu_to_be16(NULLDATAOFF))) {
 				if (curfdb != newfdb)
 					xfs_trans_brelse(tp, curbp);
 				return -EFSCORRUPTED;
 			}
 			curfdb = newfdb;
-			if (be16_to_cpu(bests[fi]) >= length)
+			if (be16_to_cpu(freehdr.bests[fi]) >= length)
 				goto out;
 		}
 	}
@@ -728,19 +793,19 @@ xfs_dir2_leafn_lookup_for_entry(
 	xfs_dir2_db_t		newdb;		/* new data block number */
 	xfs_trans_t		*tp;		/* transaction pointer */
 	enum xfs_dacmp		cmp;		/* comparison result */
-	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(mp, &leafhdr, leaf);
 
 	xfs_dir3_leaf_check(dp, bp);
-	if (leafhdr.count <= 0)
+	if (leafhdr.count <= 0) {
+		xfs_buf_mark_corrupt(bp);
 		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * Look up the hash value in the leaf entries.
@@ -756,7 +821,7 @@ xfs_dir2_leafn_lookup_for_entry(
 	/*
 	 * Loop over leaf entries with the right hash value.
 	 */
-	for (lep = &ents[index];
+	for (lep = &leafhdr.ents[index];
 	     index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
 	     lep++, index++) {
 		/*
@@ -795,7 +860,7 @@ xfs_dir2_leafn_lookup_for_entry(
 				error = xfs_dir3_data_read(tp, dp,
 						xfs_dir2_db_to_da(args->geo,
 								  newdb),
-						-1, &curbp);
+						0, &curbp);
 				if (error)
 					return error;
 			}
@@ -813,7 +878,7 @@ xfs_dir2_leafn_lookup_for_entry(
 		 * EEXIST immediately. If it's the first case-insensitive
 		 * match, store the block & inode number and continue looking.
 		 */
-		cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+		cmp = xfs_dir2_compname(args, dep->name, dep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			/* If there is a CI match block, drop it */
 			if (args->cmpresult != XFS_CMP_DIFFERENT &&
@@ -821,7 +886,7 @@ xfs_dir2_leafn_lookup_for_entry(
 				xfs_trans_brelse(tp, state->extrablk.bp);
 			args->cmpresult = cmp;
 			args->inumber = be64_to_cpu(dep->inumber);
-			args->filetype = dp->d_ops->data_get_ftype(dep);
+			args->filetype = xfs_dir2_data_get_ftype(mp, dep);
 			*indexp = index;
 			state->extravalid = 1;
 			state->extrablk.bp = curbp;
@@ -911,7 +976,7 @@ xfs_dir3_leafn_moveents(
 	if (start_d < dhdr->count) {
 		memmove(&dents[start_d + count], &dents[start_d],
 			(dhdr->count - start_d) * sizeof(xfs_dir2_leaf_entry_t));
-		xfs_dir3_leaf_log_ents(args, bp_d, start_d + count,
+		xfs_dir3_leaf_log_ents(args, dhdr, bp_d, start_d + count,
 				       count + dhdr->count - 1);
 	}
 	/*
@@ -933,7 +998,7 @@ xfs_dir3_leafn_moveents(
 	 */
 	memcpy(&dents[start_d], &sents[start_s],
 		count * sizeof(xfs_dir2_leaf_entry_t));
-	xfs_dir3_leaf_log_ents(args, bp_d, start_d, start_d + count - 1);
+	xfs_dir3_leaf_log_ents(args, dhdr, bp_d, start_d, start_d + count - 1);
 
 	/*
 	 * If there are source entries after the ones we copied,
@@ -942,7 +1007,8 @@ xfs_dir3_leafn_moveents(
 	if (start_s + count < shdr->count) {
 		memmove(&sents[start_s], &sents[start_s + count],
 			count * sizeof(xfs_dir2_leaf_entry_t));
-		xfs_dir3_leaf_log_ents(args, bp_s, start_s, start_s + count - 1);
+		xfs_dir3_leaf_log_ents(args, shdr, bp_s, start_s,
+				       start_s + count - 1);
 	}
 
 	/*
@@ -971,10 +1037,10 @@ xfs_dir2_leafn_order(
 	struct xfs_dir3_icleaf_hdr hdr1;
 	struct xfs_dir3_icleaf_hdr hdr2;
 
-	dp->d_ops->leaf_hdr_from_disk(&hdr1, leaf1);
-	dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf2);
-	ents1 = dp->d_ops->leaf_ents_p(leaf1);
-	ents2 = dp->d_ops->leaf_ents_p(leaf2);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr1, leaf1);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr2, leaf2);
+	ents1 = hdr1.ents;
+	ents2 = hdr2.ents;
 
 	if (hdr1.count > 0 && hdr2.count > 0 &&
 	    (be32_to_cpu(ents2[0].hashval) < be32_to_cpu(ents1[0].hashval) ||
@@ -1024,10 +1090,10 @@ xfs_dir2_leafn_rebalance(
 
 	leaf1 = blk1->bp->b_addr;
 	leaf2 = blk2->bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&hdr1, leaf1);
-	dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf2);
-	ents1 = dp->d_ops->leaf_ents_p(leaf1);
-	ents2 = dp->d_ops->leaf_ents_p(leaf2);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr1, leaf1);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr2, leaf2);
+	ents1 = hdr1.ents;
+	ents2 = hdr2.ents;
 
 	oldsum = hdr1.count + hdr2.count;
 #if defined(DEBUG) || defined(XFS_WARN)
@@ -1073,8 +1139,8 @@ xfs_dir2_leafn_rebalance(
 	ASSERT(hdr1.stale + hdr2.stale == oldstale);
 
 	/* log the changes made when moving the entries */
-	dp->d_ops->leaf_hdr_to_disk(leaf1, &hdr1);
-	dp->d_ops->leaf_hdr_to_disk(leaf2, &hdr2);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf1, &hdr1);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf2, &hdr2);
 	xfs_dir3_leaf_log_header(args, blk1->bp);
 	xfs_dir3_leaf_log_header(args, blk2->bp);
 
@@ -1120,19 +1186,17 @@ xfs_dir3_data_block_free(
 	int			longest)
 {
 	int			logfree = 0;
-	__be16			*bests;
 	struct xfs_dir3_icfree_hdr freehdr;
 	struct xfs_inode	*dp = args->dp;
 
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
-	bests = dp->d_ops->free_bests_p(free);
+	xfs_dir2_free_hdr_from_disk(dp->i_mount, &freehdr, free);
 	if (hdr) {
 		/*
 		 * Data block is not empty, just set the free entry to the new
 		 * value.
 		 */
-		bests[findex] = cpu_to_be16(longest);
-		xfs_dir2_free_log_bests(args, fbp, findex, findex);
+		freehdr.bests[findex] = cpu_to_be16(longest);
+		xfs_dir2_free_log_bests(args, &freehdr, fbp, findex, findex);
 		return 0;
 	}
 
@@ -1148,18 +1212,18 @@ xfs_dir3_data_block_free(
 		int	i;		/* free entry index */
 
 		for (i = findex - 1; i >= 0; i--) {
-			if (bests[i] != cpu_to_be16(NULLDATAOFF))
+			if (freehdr.bests[i] != cpu_to_be16(NULLDATAOFF))
 				break;
 		}
 		freehdr.nvalid = i + 1;
 		logfree = 0;
 	} else {
 		/* Not the last entry, just punch it out.  */
-		bests[findex] = cpu_to_be16(NULLDATAOFF);
+		freehdr.bests[findex] = cpu_to_be16(NULLDATAOFF);
 		logfree = 1;
 	}
 
-	dp->d_ops->free_hdr_to_disk(free, &freehdr);
+	xfs_dir2_free_hdr_to_disk(dp->i_mount, free, &freehdr);
 	xfs_dir2_free_log_header(args, fbp);
 
 	/*
@@ -1184,7 +1248,7 @@ xfs_dir3_data_block_free(
 
 	/* Log the free entry that changed, unless we got rid of it.  */
 	if (logfree)
-		xfs_dir2_free_log_bests(args, fbp, findex, findex);
+		xfs_dir2_free_log_bests(args, &freehdr, fbp, findex, findex);
 	return 0;
 }
 
@@ -1201,6 +1265,7 @@ xfs_dir2_leafn_remove(
 	xfs_da_state_blk_t	*dblk,		/* data block */
 	int			*rval)		/* resulting block needs join */
 {
+	struct xfs_da_geometry	*geo = args->geo;
 	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
 	xfs_dir2_db_t		db;		/* data block number */
 	struct xfs_buf		*dbp;		/* data block buffer */
@@ -1215,27 +1280,25 @@ xfs_dir2_leafn_remove(
 	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_dir2_data_free *bf;		/* bestfree table */
 	struct xfs_dir3_icleaf_hdr leafhdr;
-	struct xfs_dir2_leaf_entry *ents;
 
 	trace_xfs_dir2_leafn_remove(args, index);
 
 	dp = args->dp;
 	tp = args->trans;
 	leaf = bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
 
 	/*
 	 * Point to the entry we're removing.
 	 */
-	lep = &ents[index];
+	lep = &leafhdr.ents[index];
 
 	/*
 	 * Extract the data block and offset from the entry.
 	 */
-	db = xfs_dir2_dataptr_to_db(args->geo, be32_to_cpu(lep->address));
+	db = xfs_dir2_dataptr_to_db(geo, be32_to_cpu(lep->address));
 	ASSERT(dblk->blkno == db);
-	off = xfs_dir2_dataptr_to_off(args->geo, be32_to_cpu(lep->address));
+	off = xfs_dir2_dataptr_to_off(geo, be32_to_cpu(lep->address));
 	ASSERT(dblk->index == off);
 
 	/*
@@ -1243,11 +1306,11 @@ xfs_dir2_leafn_remove(
 	 * Log the leaf block changes.
 	 */
 	leafhdr.stale++;
-	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, leaf, &leafhdr);
 	xfs_dir3_leaf_log_header(args, bp);
 
 	lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
-	xfs_dir3_leaf_log_ents(args, bp, index, index);
+	xfs_dir3_leaf_log_ents(args, &leafhdr, bp, index, index);
 
 	/*
 	 * Make the data entry free.  Keep track of the longest freespace
@@ -1256,17 +1319,18 @@ xfs_dir2_leafn_remove(
 	dbp = dblk->bp;
 	hdr = dbp->b_addr;
 	dep = (xfs_dir2_data_entry_t *)((char *)hdr + off);
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 	longest = be16_to_cpu(bf[0].length);
 	needlog = needscan = 0;
 	xfs_dir2_data_make_free(args, dbp, off,
-		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
+		xfs_dir2_data_entsize(dp->i_mount, dep->namelen), &needlog,
+		&needscan);
 	/*
 	 * Rescan the data block freespaces for bestfree.
 	 * Log the data block header if needed.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, dbp);
 	xfs_dir3_data_check(dp, dbp);
@@ -1285,9 +1349,8 @@ xfs_dir2_leafn_remove(
 		 * Convert the data block number to a free block,
 		 * read in the free block.
 		 */
-		fdb = dp->d_ops->db_to_fdb(args->geo, db);
-		error = xfs_dir2_free_read(tp, dp,
-					   xfs_dir2_db_to_da(args->geo, fdb),
+		fdb = xfs_dir2_db_to_fdb(geo, db);
+		error = xfs_dir2_free_read(tp, dp, xfs_dir2_db_to_da(geo, fdb),
 					   &fbp);
 		if (error)
 			return error;
@@ -1295,23 +1358,22 @@ xfs_dir2_leafn_remove(
 #ifdef DEBUG
 	{
 		struct xfs_dir3_icfree_hdr freehdr;
-		dp->d_ops->free_hdr_from_disk(&freehdr, free);
-		ASSERT(freehdr.firstdb == dp->d_ops->free_max_bests(args->geo) *
-			(fdb - xfs_dir2_byte_to_db(args->geo,
-						   XFS_DIR2_FREE_OFFSET)));
+
+		xfs_dir2_free_hdr_from_disk(dp->i_mount, &freehdr, free);
+		ASSERT(freehdr.firstdb == geo->free_max_bests *
+			(fdb - xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET)));
 	}
 #endif
 		/*
 		 * Calculate which entry we need to fix.
 		 */
-		findex = dp->d_ops->db_to_fdindex(args->geo, db);
+		findex = xfs_dir2_db_to_fdindex(geo, db);
 		longest = be16_to_cpu(bf[0].length);
 		/*
 		 * If the data block is now empty we can get rid of it
 		 * (usually).
 		 */
-		if (longest == args->geo->blksize -
-			       dp->d_ops->data_entry_offset) {
+		if (longest == geo->blksize - geo->data_entry_offset) {
 			/*
 			 * Try to punch out the data block.
 			 */
@@ -1343,9 +1405,9 @@ xfs_dir2_leafn_remove(
 	 * Return indication of whether this leaf block is empty enough
 	 * to justify trying to join it with a neighbor.
 	 */
-	*rval = (dp->d_ops->leaf_hdr_size +
-		 (uint)sizeof(ents[0]) * (leafhdr.count - leafhdr.stale)) <
-		args->geo->magicpct;
+	*rval = (geo->leaf_hdr_size +
+		 (uint)sizeof(leafhdr.ents) * (leafhdr.count - leafhdr.stale)) <
+		geo->magicpct;
 	return 0;
 }
 
@@ -1444,12 +1506,12 @@ xfs_dir2_leafn_toosmall(
 	 */
 	blk = &state->path.blk[state->path.active - 1];
 	leaf = blk->bp->b_addr;
-	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &leafhdr, leaf);
+	ents = leafhdr.ents;
 	xfs_dir3_leaf_check(dp, blk->bp);
 
 	count = leafhdr.count - leafhdr.stale;
-	bytes = dp->d_ops->leaf_hdr_size + count * sizeof(ents[0]);
+	bytes = state->args->geo->leaf_hdr_size + count * sizeof(ents[0]);
 	if (bytes > (state->args->geo->blksize >> 1)) {
 		/*
 		 * Blk over 50%, don't try to join.
@@ -1494,8 +1556,7 @@ xfs_dir2_leafn_toosmall(
 		/*
 		 * Read the sibling leaf block.
 		 */
-		error = xfs_dir3_leafn_read(state->args->trans, dp,
-					    blkno, -1, &bp);
+		error = xfs_dir3_leafn_read(state->args->trans, dp, blkno, &bp);
 		if (error)
 			return error;
 
@@ -1507,8 +1568,8 @@ xfs_dir2_leafn_toosmall(
 			(state->args->geo->blksize >> 2);
 
 		leaf = bp->b_addr;
-		dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf);
-		ents = dp->d_ops->leaf_ents_p(leaf);
+		xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &hdr2, leaf);
+		ents = hdr2.ents;
 		count += hdr2.count - hdr2.stale;
 		bytes -= count * sizeof(ents[0]);
 
@@ -1570,10 +1631,10 @@ xfs_dir2_leafn_unbalance(
 	drop_leaf = drop_blk->bp->b_addr;
 	save_leaf = save_blk->bp->b_addr;
 
-	dp->d_ops->leaf_hdr_from_disk(&savehdr, save_leaf);
-	dp->d_ops->leaf_hdr_from_disk(&drophdr, drop_leaf);
-	sents = dp->d_ops->leaf_ents_p(save_leaf);
-	dents = dp->d_ops->leaf_ents_p(drop_leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &savehdr, save_leaf);
+	xfs_dir2_leaf_hdr_from_disk(dp->i_mount, &drophdr, drop_leaf);
+	sents = savehdr.ents;
+	dents = drophdr.ents;
 
 	/*
 	 * If there are any stale leaf entries, take this opportunity
@@ -1599,8 +1660,8 @@ xfs_dir2_leafn_unbalance(
 	save_blk->hashval = be32_to_cpu(sents[savehdr.count - 1].hashval);
 
 	/* log the changes made when moving the entries */
-	dp->d_ops->leaf_hdr_to_disk(save_leaf, &savehdr);
-	dp->d_ops->leaf_hdr_to_disk(drop_leaf, &drophdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, save_leaf, &savehdr);
+	xfs_dir2_leaf_hdr_to_disk(dp->i_mount, drop_leaf, &drophdr);
 	xfs_dir3_leaf_log_header(args, save_blk->bp);
 	xfs_dir3_leaf_log_header(args, drop_blk->bp);
 
@@ -1619,19 +1680,16 @@ xfs_dir2_node_add_datablk(
 	xfs_dir2_db_t		*dbno,
 	struct xfs_buf		**dbpp,
 	struct xfs_buf		**fbpp,
+	struct xfs_dir3_icfree_hdr *hdr,
 	int			*findex)
 {
 	struct xfs_inode	*dp = args->dp;
 	struct xfs_trans	*tp = args->trans;
 	struct xfs_mount	*mp = dp->i_mount;
-	struct xfs_dir3_icfree_hdr freehdr;
 	struct xfs_dir2_data_free *bf;
-	struct xfs_dir2_data_hdr *hdr;
-	struct xfs_dir2_free	*free = NULL;
 	xfs_dir2_db_t		fbno;
 	struct xfs_buf		*fbp;
 	struct xfs_buf		*dbp;
-	__be16			*bests = NULL;
 	int			error;
 
 	/* Not allowed to allocate, return failure. */
@@ -1650,7 +1708,7 @@ xfs_dir2_node_add_datablk(
 	 * Get the freespace block corresponding to the data block
 	 * that was just allocated.
 	 */
-	fbno = dp->d_ops->db_to_fdb(args->geo, *dbno);
+	fbno = xfs_dir2_db_to_fdb(args->geo, *dbno);
 	error = xfs_dir2_free_try_read(tp, dp,
 			       xfs_dir2_db_to_da(args->geo, fbno), &fbp);
 	if (error)
@@ -1665,11 +1723,13 @@ xfs_dir2_node_add_datablk(
 		if (error)
 			return error;
 
-		if (dp->d_ops->db_to_fdb(args->geo, *dbno) != fbno) {
+		if (XFS_IS_CORRUPT(mp,
+				   xfs_dir2_db_to_fdb(args->geo, *dbno) !=
+				   fbno)) {
 			xfs_alert(mp,
 "%s: dir ino %llu needed freesp block %lld for data block %lld, got %lld",
 				__func__, (unsigned long long)dp->i_ino,
-				(long long)dp->d_ops->db_to_fdb(args->geo, *dbno),
+				(long long)xfs_dir2_db_to_fdb(args->geo, *dbno),
 				(long long)*dbno, (long long)fbno);
 			if (fblk) {
 				xfs_alert(mp,
@@ -1679,7 +1739,6 @@ xfs_dir2_node_add_datablk(
 			} else {
 				xfs_alert(mp, " ... fblk is NULL");
 			}
-			XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 			return -EFSCORRUPTED;
 		}
 
@@ -1687,44 +1746,39 @@ xfs_dir2_node_add_datablk(
 		error = xfs_dir3_free_get_buf(args, fbno, &fbp);
 		if (error)
 			return error;
-		free = fbp->b_addr;
-		bests = dp->d_ops->free_bests_p(free);
-		dp->d_ops->free_hdr_from_disk(&freehdr, free);
+		xfs_dir2_free_hdr_from_disk(mp, hdr, fbp->b_addr);
 
 		/* Remember the first slot as our empty slot. */
-		freehdr.firstdb = (fbno - xfs_dir2_byte_to_db(args->geo,
+		hdr->firstdb = (fbno - xfs_dir2_byte_to_db(args->geo,
 							XFS_DIR2_FREE_OFFSET)) *
-				dp->d_ops->free_max_bests(args->geo);
+				args->geo->free_max_bests;
 	} else {
-		free = fbp->b_addr;
-		bests = dp->d_ops->free_bests_p(free);
-		dp->d_ops->free_hdr_from_disk(&freehdr, free);
+		xfs_dir2_free_hdr_from_disk(mp, hdr, fbp->b_addr);
 	}
 
 	/* Set the freespace block index from the data block number. */
-	*findex = dp->d_ops->db_to_fdindex(args->geo, *dbno);
+	*findex = xfs_dir2_db_to_fdindex(args->geo, *dbno);
 
 	/* Extend the freespace table if the new data block is off the end. */
-	if (*findex >= freehdr.nvalid) {
-		ASSERT(*findex < dp->d_ops->free_max_bests(args->geo));
-		freehdr.nvalid = *findex + 1;
-		bests[*findex] = cpu_to_be16(NULLDATAOFF);
+	if (*findex >= hdr->nvalid) {
+		ASSERT(*findex < args->geo->free_max_bests);
+		hdr->nvalid = *findex + 1;
+		hdr->bests[*findex] = cpu_to_be16(NULLDATAOFF);
 	}
 
 	/*
 	 * If this entry was for an empty data block (this should always be
 	 * true) then update the header.
 	 */
-	if (bests[*findex] == cpu_to_be16(NULLDATAOFF)) {
-		freehdr.nused++;
-		dp->d_ops->free_hdr_to_disk(fbp->b_addr, &freehdr);
+	if (hdr->bests[*findex] == cpu_to_be16(NULLDATAOFF)) {
+		hdr->nused++;
+		xfs_dir2_free_hdr_to_disk(mp, fbp->b_addr, hdr);
 		xfs_dir2_free_log_header(args, fbp);
 	}
 
 	/* Update the freespace value for the new block in the table. */
-	hdr = dbp->b_addr;
-	bf = dp->d_ops->data_bestfree_p(hdr);
-	bests[*findex] = bf[0].length;
+	bf = xfs_dir2_data_bestfree_p(mp, dbp->b_addr);
+	hdr->bests[*findex] = bf[0].length;
 
 	*dbpp = dbp;
 	*fbpp = fbp;
@@ -1737,11 +1791,10 @@ xfs_dir2_node_find_freeblk(
 	struct xfs_da_state_blk	*fblk,
 	xfs_dir2_db_t		*dbnop,
 	struct xfs_buf		**fbpp,
+	struct xfs_dir3_icfree_hdr *hdr,
 	int			*findexp,
 	int			length)
 {
-	struct xfs_dir3_icfree_hdr freehdr;
-	struct xfs_dir2_free	*free = NULL;
 	struct xfs_inode	*dp = args->dp;
 	struct xfs_trans	*tp = args->trans;
 	struct xfs_buf		*fbp = NULL;
@@ -1751,7 +1804,6 @@ xfs_dir2_node_find_freeblk(
 	xfs_dir2_db_t		dbno = -1;
 	xfs_dir2_db_t		fbno;
 	xfs_fileoff_t		fo;
-	__be16			*bests = NULL;
 	int			findex = 0;
 	int			error;
 
@@ -1762,17 +1814,14 @@ xfs_dir2_node_find_freeblk(
 	 */
 	if (fblk) {
 		fbp = fblk->bp;
-		free = fbp->b_addr;
 		findex = fblk->index;
+		xfs_dir2_free_hdr_from_disk(dp->i_mount, hdr, fbp->b_addr);
 		if (findex >= 0) {
 			/* caller already found the freespace for us. */
-			bests = dp->d_ops->free_bests_p(free);
-			dp->d_ops->free_hdr_from_disk(&freehdr, free);
-
-			ASSERT(findex < freehdr.nvalid);
-			ASSERT(be16_to_cpu(bests[findex]) != NULLDATAOFF);
-			ASSERT(be16_to_cpu(bests[findex]) >= length);
-			dbno = freehdr.firstdb + findex;
+			ASSERT(findex < hdr->nvalid);
+			ASSERT(be16_to_cpu(hdr->bests[findex]) != NULLDATAOFF);
+			ASSERT(be16_to_cpu(hdr->bests[findex]) >= length);
+			dbno = hdr->firstdb + findex;
 			goto found_block;
 		}
 
@@ -1814,15 +1863,13 @@ xfs_dir2_node_find_freeblk(
 		if (!fbp)
 			continue;
 
-		free = fbp->b_addr;
-		bests = dp->d_ops->free_bests_p(free);
-		dp->d_ops->free_hdr_from_disk(&freehdr, free);
+		xfs_dir2_free_hdr_from_disk(dp->i_mount, hdr, fbp->b_addr);
 
 		/* Scan the free entry array for a large enough free space. */
-		for (findex = freehdr.nvalid - 1; findex >= 0; findex--) {
-			if (be16_to_cpu(bests[findex]) != NULLDATAOFF &&
-			    be16_to_cpu(bests[findex]) >= length) {
-				dbno = freehdr.firstdb + findex;
+		for (findex = hdr->nvalid - 1; findex >= 0; findex--) {
+			if (be16_to_cpu(hdr->bests[findex]) != NULLDATAOFF &&
+			    be16_to_cpu(hdr->bests[findex]) >= length) {
+				dbno = hdr->firstdb + findex;
 				goto found_block;
 			}
 		}
@@ -1838,7 +1885,6 @@ found_block:
 	return 0;
 }
 
-
 /*
  * Add the data entry for a node-format directory name addition.
  * The leaf entry is added in xfs_dir2_leafn_add.
@@ -1853,9 +1899,9 @@ xfs_dir2_node_addname_int(
 	struct xfs_dir2_data_entry *dep;	/* data entry pointer */
 	struct xfs_dir2_data_hdr *hdr;		/* data block header */
 	struct xfs_dir2_data_free *bf;
-	struct xfs_dir2_free	*free = NULL;	/* freespace block structure */
 	struct xfs_trans	*tp = args->trans;
 	struct xfs_inode	*dp = args->dp;
+	struct xfs_dir3_icfree_hdr freehdr;
 	struct xfs_buf		*dbp;		/* data block buffer */
 	struct xfs_buf		*fbp;		/* freespace buffer */
 	xfs_dir2_data_aoff_t	aoff;
@@ -1867,11 +1913,10 @@ xfs_dir2_node_addname_int(
 	int			needlog = 0;	/* need to log data header */
 	int			needscan = 0;	/* need to rescan data frees */
 	__be16			*tagp;		/* data entry tag pointer */
-	__be16			*bests;
 
-	length = dp->d_ops->data_entsize(args->namelen);
-	error = xfs_dir2_node_find_freeblk(args, fblk, &dbno, &fbp, &findex,
-					   length);
+	length = xfs_dir2_data_entsize(dp->i_mount, args->namelen);
+	error = xfs_dir2_node_find_freeblk(args, fblk, &dbno, &fbp, &freehdr,
+					   &findex, length);
 	if (error)
 		return error;
 
@@ -1893,19 +1938,19 @@ xfs_dir2_node_addname_int(
 		/* we're going to have to log the free block index later */
 		logfree = 1;
 		error = xfs_dir2_node_add_datablk(args, fblk, &dbno, &dbp, &fbp,
-						  &findex);
+						  &freehdr, &findex);
 	} else {
 		/* Read the data block in. */
 		error = xfs_dir3_data_read(tp, dp,
 					   xfs_dir2_db_to_da(args->geo, dbno),
-					   -1, &dbp);
+					   0, &dbp);
 	}
 	if (error)
 		return error;
 
 	/* setup for data block up now */
 	hdr = dbp->b_addr;
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = xfs_dir2_data_bestfree_p(dp->i_mount, hdr);
 	ASSERT(be16_to_cpu(bf[0].length) >= length);
 
 	/* Point to the existing unused space. */
@@ -1926,28 +1971,26 @@ xfs_dir2_node_addname_int(
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, dep->namelen);
-	dp->d_ops->data_put_ftype(dep, args->filetype);
-	tagp = dp->d_ops->data_entry_tag_p(dep);
+	xfs_dir2_data_put_ftype(dp->i_mount, dep, args->filetype);
+	tagp = xfs_dir2_data_entry_tag_p(dp->i_mount, dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
 	xfs_dir2_data_log_entry(args, dbp, dep);
 
 	/* Rescan the freespace and log the data block if needed. */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp->i_mount, hdr, &needlog);
 	if (needlog)
 		xfs_dir2_data_log_header(args, dbp);
 
 	/* If the freespace block entry is now wrong, update it. */
-	free = fbp->b_addr;
-	bests = dp->d_ops->free_bests_p(free);
-	if (bests[findex] != bf[0].length) {
-		bests[findex] = bf[0].length;
+	if (freehdr.bests[findex] != bf[0].length) {
+		freehdr.bests[findex] = bf[0].length;
 		logfree = 1;
 	}
 
 	/* Log the freespace entry if needed. */
 	if (logfree)
-		xfs_dir2_free_log_bests(args, fbp, findex, findex);
+		xfs_dir2_free_log_bests(args, &freehdr, fbp, findex, findex);
 
 	/* Return the data block and offset in args. */
 	args->blkno = (xfs_dablk_t)dbno;
@@ -2155,8 +2198,6 @@ xfs_dir2_node_replace(
 	int			i;		/* btree level */
 	xfs_ino_t		inum;		/* new inode number */
 	int			ftype;		/* new file type */
-	xfs_dir2_leaf_t		*leaf;		/* leaf structure */
-	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry being changed */
 	int			rval;		/* internal return value */
 	xfs_da_state_t		*state;		/* btree cursor */
 
@@ -2188,16 +2229,17 @@ xfs_dir2_node_replace(
 	 * and locked it.  But paranoia is good.
 	 */
 	if (rval == -EEXIST) {
-		struct xfs_dir2_leaf_entry *ents;
+		struct xfs_dir3_icleaf_hdr	leafhdr;
+
 		/*
 		 * Find the leaf entry.
 		 */
 		blk = &state->path.blk[state->path.active - 1];
 		ASSERT(blk->magic == XFS_DIR2_LEAFN_MAGIC);
-		leaf = blk->bp->b_addr;
-		ents = args->dp->d_ops->leaf_ents_p(leaf);
-		lep = &ents[blk->index];
 		ASSERT(state->extravalid);
+
+		xfs_dir2_leaf_hdr_from_disk(state->mp, &leafhdr,
+					    blk->bp->b_addr);
 		/*
 		 * Point to the data entry.
 		 */
@@ -2207,13 +2249,13 @@ xfs_dir2_node_replace(
 		dep = (xfs_dir2_data_entry_t *)
 		      ((char *)hdr +
 		       xfs_dir2_dataptr_to_off(args->geo,
-					       be32_to_cpu(lep->address)));
+				be32_to_cpu(leafhdr.ents[blk->index].address)));
 		ASSERT(inum != be64_to_cpu(dep->inumber));
 		/*
 		 * Fill in the new inode number and log the entry.
 		 */
 		dep->inumber = cpu_to_be64(inum);
-		args->dp->d_ops->data_put_ftype(dep, ftype);
+		xfs_dir2_data_put_ftype(state->mp, dep, ftype);
 		xfs_dir2_data_log_entry(args, state->extrablk.bp, dep);
 		rval = 0;
 	}
@@ -2270,7 +2312,7 @@ xfs_dir2_node_trim_free(
 	if (!bp)
 		return 0;
 	free = bp->b_addr;
-	dp->d_ops->free_hdr_from_disk(&freehdr, free);
+	xfs_dir2_free_hdr_from_disk(dp->i_mount, &freehdr, free);
 
 	/*
 	 * If there are used entries, there's nothing to do.
diff --git a/fs/xfs/libxfs/xfs_dir2_priv.h b/fs/xfs/libxfs/xfs_dir2_priv.h
index 59f9fb2241a5..01ee0b926572 100644
--- a/fs/xfs/libxfs/xfs_dir2_priv.h
+++ b/fs/xfs/libxfs/xfs_dir2_priv.h
@@ -8,7 +8,41 @@
 
 struct dir_context;
 
+/*
+ * In-core version of the leaf and free block headers to abstract the
+ * differences in the v2 and v3 disk format of the headers.
+ */
+struct xfs_dir3_icleaf_hdr {
+	uint32_t		forw;
+	uint32_t		back;
+	uint16_t		magic;
+	uint16_t		count;
+	uint16_t		stale;
+
+	/*
+	 * Pointer to the on-disk format entries, which are behind the
+	 * variable size (v4 vs v5) header in the on-disk block.
+	 */
+	struct xfs_dir2_leaf_entry *ents;
+};
+
+struct xfs_dir3_icfree_hdr {
+	uint32_t		magic;
+	uint32_t		firstdb;
+	uint32_t		nvalid;
+	uint32_t		nused;
+
+	/*
+	 * Pointer to the on-disk format entries, which are behind the
+	 * variable size (v4 vs v5) header in the on-disk block.
+	 */
+	__be16			*bests;
+};
+
 /* xfs_dir2.c */
+xfs_dahash_t xfs_ascii_ci_hashname(struct xfs_name *name);
+enum xfs_dacmp xfs_ascii_ci_compname(struct xfs_da_args *args,
+		const unsigned char *name, int len);
 extern int xfs_dir2_grow_inode(struct xfs_da_args *args, int space,
 				xfs_dir2_db_t *dbp);
 extern int xfs_dir_cilookup_result(struct xfs_da_args *args,
@@ -26,6 +60,15 @@ extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args,
 		struct xfs_buf *lbp, struct xfs_buf *dbp);
 
 /* xfs_dir2_data.c */
+struct xfs_dir2_data_free *xfs_dir2_data_bestfree_p(struct xfs_mount *mp,
+		struct xfs_dir2_data_hdr *hdr);
+__be16 *xfs_dir2_data_entry_tag_p(struct xfs_mount *mp,
+		struct xfs_dir2_data_entry *dep);
+uint8_t xfs_dir2_data_get_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_data_entry *dep);
+void xfs_dir2_data_put_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_data_entry *dep, uint8_t ftype);
+
 #ifdef DEBUG
 extern void xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
 #else
@@ -34,10 +77,10 @@ extern void xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
 
 extern xfs_failaddr_t __xfs_dir3_data_check(struct xfs_inode *dp,
 		struct xfs_buf *bp);
-extern int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
-		xfs_dablk_t bno, xfs_daddr_t mapped_bno, struct xfs_buf **bpp);
-extern int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno,
-		xfs_daddr_t mapped_bno);
+int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
+		xfs_dablk_t bno, unsigned int flags, struct xfs_buf **bpp);
+int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno,
+		unsigned int flags);
 
 extern struct xfs_dir2_data_free *
 xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr,
@@ -47,10 +90,14 @@ extern int xfs_dir3_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno,
 		struct xfs_buf **bpp);
 
 /* xfs_dir2_leaf.c */
-extern int xfs_dir3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
-		xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp);
-extern int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
-		xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp);
+void xfs_dir2_leaf_hdr_from_disk(struct xfs_mount *mp,
+		struct xfs_dir3_icleaf_hdr *to, struct xfs_dir2_leaf *from);
+void xfs_dir2_leaf_hdr_to_disk(struct xfs_mount *mp, struct xfs_dir2_leaf *to,
+		struct xfs_dir3_icleaf_hdr *from);
+int xfs_dir3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
+		xfs_dablk_t fbno, struct xfs_buf **bpp);
+int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
+		xfs_dablk_t fbno, struct xfs_buf **bpp);
 extern int xfs_dir2_block_to_leaf(struct xfs_da_args *args,
 		struct xfs_buf *dbp);
 extern int xfs_dir2_leaf_addname(struct xfs_da_args *args);
@@ -62,7 +109,8 @@ extern void xfs_dir3_leaf_compact_x1(struct xfs_dir3_icleaf_hdr *leafhdr,
 extern int xfs_dir3_leaf_get_buf(struct xfs_da_args *args, xfs_dir2_db_t bno,
 		struct xfs_buf **bpp, uint16_t magic);
 extern void xfs_dir3_leaf_log_ents(struct xfs_da_args *args,
-		struct xfs_buf *bp, int first, int last);
+		struct xfs_dir3_icleaf_hdr *hdr, struct xfs_buf *bp, int first,
+		int last);
 extern void xfs_dir3_leaf_log_header(struct xfs_da_args *args,
 		struct xfs_buf *bp);
 extern int xfs_dir2_leaf_lookup(struct xfs_da_args *args);
@@ -79,10 +127,11 @@ xfs_dir3_leaf_find_entry(struct xfs_dir3_icleaf_hdr *leafhdr,
 extern int xfs_dir2_node_to_leaf(struct xfs_da_state *state);
 
 extern xfs_failaddr_t xfs_dir3_leaf_check_int(struct xfs_mount *mp,
-		struct xfs_inode *dp, struct xfs_dir3_icleaf_hdr *hdr,
-		struct xfs_dir2_leaf *leaf);
+		struct xfs_dir3_icleaf_hdr *hdr, struct xfs_dir2_leaf *leaf);
 
 /* xfs_dir2_node.c */
+void xfs_dir2_free_hdr_from_disk(struct xfs_mount *mp,
+		struct xfs_dir3_icfree_hdr *to, struct xfs_dir2_free *from);
 extern int xfs_dir2_leaf_to_node(struct xfs_da_args *args,
 		struct xfs_buf *lbp);
 extern xfs_dahash_t xfs_dir2_leaf_lasthash(struct xfs_inode *dp,
@@ -108,6 +157,14 @@ extern int xfs_dir2_free_read(struct xfs_trans *tp, struct xfs_inode *dp,
 		xfs_dablk_t fbno, struct xfs_buf **bpp);
 
 /* xfs_dir2_sf.c */
+xfs_ino_t xfs_dir2_sf_get_ino(struct xfs_mount *mp, struct xfs_dir2_sf_hdr *hdr,
+		struct xfs_dir2_sf_entry *sfep);
+xfs_ino_t xfs_dir2_sf_get_parent_ino(struct xfs_dir2_sf_hdr *hdr);
+void xfs_dir2_sf_put_parent_ino(struct xfs_dir2_sf_hdr *hdr, xfs_ino_t ino);
+uint8_t xfs_dir2_sf_get_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_sf_entry *sfep);
+struct xfs_dir2_sf_entry *xfs_dir2_sf_nextentry(struct xfs_mount *mp,
+		struct xfs_dir2_sf_hdr *hdr, struct xfs_dir2_sf_entry *sfep);
 extern int xfs_dir2_block_sfsize(struct xfs_inode *dp,
 		struct xfs_dir2_data_hdr *block, struct xfs_dir2_sf_hdr *sfhp);
 extern int xfs_dir2_block_to_sf(struct xfs_da_args *args, struct xfs_buf *bp,
@@ -118,9 +175,33 @@ extern int xfs_dir2_sf_lookup(struct xfs_da_args *args);
 extern int xfs_dir2_sf_removename(struct xfs_da_args *args);
 extern int xfs_dir2_sf_replace(struct xfs_da_args *args);
 extern xfs_failaddr_t xfs_dir2_sf_verify(struct xfs_inode *ip);
+int xfs_dir2_sf_entsize(struct xfs_mount *mp,
+		struct xfs_dir2_sf_hdr *hdr, int len);
+void xfs_dir2_sf_put_ino(struct xfs_mount *mp, struct xfs_dir2_sf_hdr *hdr,
+		struct xfs_dir2_sf_entry *sfep, xfs_ino_t ino);
+void xfs_dir2_sf_put_ftype(struct xfs_mount *mp,
+		struct xfs_dir2_sf_entry *sfep, uint8_t ftype);
 
 /* xfs_dir2_readdir.c */
 extern int xfs_readdir(struct xfs_trans *tp, struct xfs_inode *dp,
 		       struct dir_context *ctx, size_t bufsize);
 
+static inline unsigned int
+xfs_dir2_data_entsize(
+	struct xfs_mount	*mp,
+	unsigned int		namelen)
+{
+	unsigned int		len;
+
+	len = offsetof(struct xfs_dir2_data_entry, name[0]) + namelen +
+			sizeof(xfs_dir2_data_off_t) /* tag */;
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		len += sizeof(uint8_t);
+	return round_up(len, XFS_DIR2_DATA_ALIGN);
+}
+
+xfs_dahash_t xfs_dir2_hashname(struct xfs_mount *mp, struct xfs_name *name);
+enum xfs_dacmp xfs_dir2_compname(struct xfs_da_args *args,
+		const unsigned char *name, int len);
+
 #endif /* __XFS_DIR2_PRIV_H__ */
diff --git a/fs/xfs/libxfs/xfs_dir2_sf.c b/fs/xfs/libxfs/xfs_dir2_sf.c
index 85f14fc2a8da..7b7f6fb2ea3b 100644
--- a/fs/xfs/libxfs/xfs_dir2_sf.c
+++ b/fs/xfs/libxfs/xfs_dir2_sf.c
@@ -37,6 +37,126 @@ static void xfs_dir2_sf_check(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino4(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
 
+int
+xfs_dir2_sf_entsize(
+	struct xfs_mount	*mp,
+	struct xfs_dir2_sf_hdr	*hdr,
+	int			len)
+{
+	int			count = len;
+
+	count += sizeof(struct xfs_dir2_sf_entry);	/* namelen + offset */
+	count += hdr->i8count ? XFS_INO64_SIZE : XFS_INO32_SIZE; /* ino # */
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		count += sizeof(uint8_t);
+	return count;
+}
+
+struct xfs_dir2_sf_entry *
+xfs_dir2_sf_nextentry(
+	struct xfs_mount	*mp,
+	struct xfs_dir2_sf_hdr	*hdr,
+	struct xfs_dir2_sf_entry *sfep)
+{
+	return (void *)sfep + xfs_dir2_sf_entsize(mp, hdr, sfep->namelen);
+}
+
+/*
+ * In short-form directory entries the inode numbers are stored at variable
+ * offset behind the entry name. If the entry stores a filetype value, then it
+ * sits between the name and the inode number.  The actual inode numbers can
+ * come in two formats as well, either 4 bytes or 8 bytes wide.
+ */
+xfs_ino_t
+xfs_dir2_sf_get_ino(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_sf_hdr		*hdr,
+	struct xfs_dir2_sf_entry	*sfep)
+{
+	uint8_t				*from = sfep->name + sfep->namelen;
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		from++;
+
+	if (!hdr->i8count)
+		return get_unaligned_be32(from);
+	return get_unaligned_be64(from) & XFS_MAXINUMBER;
+}
+
+void
+xfs_dir2_sf_put_ino(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_sf_hdr		*hdr,
+	struct xfs_dir2_sf_entry	*sfep,
+	xfs_ino_t			ino)
+{
+	uint8_t				*to = sfep->name + sfep->namelen;
+
+	ASSERT(ino <= XFS_MAXINUMBER);
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		to++;
+
+	if (hdr->i8count)
+		put_unaligned_be64(ino, to);
+	else
+		put_unaligned_be32(ino, to);
+}
+
+xfs_ino_t
+xfs_dir2_sf_get_parent_ino(
+	struct xfs_dir2_sf_hdr	*hdr)
+{
+	if (!hdr->i8count)
+		return get_unaligned_be32(hdr->parent);
+	return get_unaligned_be64(hdr->parent) & XFS_MAXINUMBER;
+}
+
+void
+xfs_dir2_sf_put_parent_ino(
+	struct xfs_dir2_sf_hdr		*hdr,
+	xfs_ino_t			ino)
+{
+	ASSERT(ino <= XFS_MAXINUMBER);
+
+	if (hdr->i8count)
+		put_unaligned_be64(ino, hdr->parent);
+	else
+		put_unaligned_be32(ino, hdr->parent);
+}
+
+/*
+ * The file type field is stored at the end of the name for filetype enabled
+ * shortform directories, or not at all otherwise.
+ */
+uint8_t
+xfs_dir2_sf_get_ftype(
+	struct xfs_mount		*mp,
+	struct xfs_dir2_sf_entry	*sfep)
+{
+	if (xfs_sb_version_hasftype(&mp->m_sb)) {
+		uint8_t			ftype = sfep->name[sfep->namelen];
+
+		if (ftype < XFS_DIR3_FT_MAX)
+			return ftype;
+	}
+
+	return XFS_DIR3_FT_UNKNOWN;
+}
+
+void
+xfs_dir2_sf_put_ftype(
+	struct xfs_mount	*mp,
+	struct xfs_dir2_sf_entry *sfep,
+	uint8_t			ftype)
+{
+	ASSERT(ftype < XFS_DIR3_FT_MAX);
+
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		sfep->name[sfep->namelen] = ftype;
+}
+
 /*
  * Given a block directory (dp/block), calculate its size as a shortform (sf)
  * directory and a header for the sf directory, if it will fit it the
@@ -125,7 +245,7 @@ xfs_dir2_block_sfsize(
 	 */
 	sfhp->count = count;
 	sfhp->i8count = i8count;
-	dp->d_ops->sf_put_parent_ino(sfhp, parent);
+	xfs_dir2_sf_put_parent_ino(sfhp, parent);
 	return size;
 }
 
@@ -135,64 +255,48 @@ xfs_dir2_block_sfsize(
  */
 int						/* error */
 xfs_dir2_block_to_sf(
-	xfs_da_args_t		*args,		/* operation arguments */
+	struct xfs_da_args	*args,		/* operation arguments */
 	struct xfs_buf		*bp,
 	int			size,		/* shortform directory size */
-	xfs_dir2_sf_hdr_t	*sfhp)		/* shortform directory hdr */
+	struct xfs_dir2_sf_hdr	*sfhp)		/* shortform directory hdr */
 {
-	xfs_dir2_data_hdr_t	*hdr;		/* block header */
-	xfs_dir2_data_entry_t	*dep;		/* data entry pointer */
-	xfs_inode_t		*dp;		/* incore directory inode */
-	xfs_dir2_data_unused_t	*dup;		/* unused data pointer */
-	char			*endptr;	/* end of data entries */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			error;		/* error return value */
 	int			logflags;	/* inode logging flags */
-	xfs_mount_t		*mp;		/* filesystem mount point */
-	char			*ptr;		/* current data pointer */
-	xfs_dir2_sf_entry_t	*sfep;		/* shortform entry */
-	xfs_dir2_sf_hdr_t	*sfp;		/* shortform directory header */
-	xfs_dir2_sf_hdr_t	*dst;		/* temporary data buffer */
+	struct xfs_dir2_sf_entry *sfep;		/* shortform entry */
+	struct xfs_dir2_sf_hdr	*sfp;		/* shortform directory header */
+	unsigned int		offset = args->geo->data_entry_offset;
+	unsigned int		end;
 
 	trace_xfs_dir2_block_to_sf(args);
 
-	dp = args->dp;
-	mp = dp->i_mount;
-
-	/*
-	 * allocate a temporary destination buffer the size of the inode
-	 * to format the data into. Once we have formatted the data, we
-	 * can free the block and copy the formatted data into the inode literal
-	 * area.
-	 */
-	dst = kmem_alloc(mp->m_sb.sb_inodesize, 0);
-	hdr = bp->b_addr;
-
 	/*
-	 * Copy the header into the newly allocate local space.
+	 * Allocate a temporary destination buffer the size of the inode to
+	 * format the data into.  Once we have formatted the data, we can free
+	 * the block and copy the formatted data into the inode literal area.
 	 */
-	sfp = (xfs_dir2_sf_hdr_t *)dst;
+	sfp = kmem_alloc(mp->m_sb.sb_inodesize, 0);
 	memcpy(sfp, sfhp, xfs_dir2_sf_hdr_size(sfhp->i8count));
 
 	/*
-	 * Set up to loop over the block's entries.
+	 * Loop over the active and unused entries.  Stop when we reach the
+	 * leaf/tail portion of the block.
 	 */
-	ptr = (char *)dp->d_ops->data_entry_p(hdr);
-	endptr = xfs_dir3_data_endp(args->geo, hdr);
+	end = xfs_dir3_data_end_offset(args->geo, bp->b_addr);
 	sfep = xfs_dir2_sf_firstentry(sfp);
-	/*
-	 * Loop over the active and unused entries.
-	 * Stop when we reach the leaf/tail portion of the block.
-	 */
-	while (ptr < endptr) {
+	while (offset < end) {
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
+
 		/*
 		 * If it's unused, just skip over it.
 		 */
-		dup = (xfs_dir2_data_unused_t *)ptr;
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
-			ptr += be16_to_cpu(dup->length);
+			offset += be16_to_cpu(dup->length);
 			continue;
 		}
-		dep = (xfs_dir2_data_entry_t *)ptr;
+
 		/*
 		 * Skip .
 		 */
@@ -204,24 +308,22 @@ xfs_dir2_block_to_sf(
 		else if (dep->namelen == 2 &&
 			 dep->name[0] == '.' && dep->name[1] == '.')
 			ASSERT(be64_to_cpu(dep->inumber) ==
-			       dp->d_ops->sf_get_parent_ino(sfp));
+			       xfs_dir2_sf_get_parent_ino(sfp));
 		/*
 		 * Normal entry, copy it into shortform.
 		 */
 		else {
 			sfep->namelen = dep->namelen;
-			xfs_dir2_sf_put_offset(sfep,
-				(xfs_dir2_data_aoff_t)
-				((char *)dep - (char *)hdr));
+			xfs_dir2_sf_put_offset(sfep, offset);
 			memcpy(sfep->name, dep->name, dep->namelen);
-			dp->d_ops->sf_put_ino(sfp, sfep,
+			xfs_dir2_sf_put_ino(mp, sfp, sfep,
 					      be64_to_cpu(dep->inumber));
-			dp->d_ops->sf_put_ftype(sfep,
-					dp->d_ops->data_get_ftype(dep));
+			xfs_dir2_sf_put_ftype(mp, sfep,
+					xfs_dir2_data_get_ftype(mp, dep));
 
-			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+			sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 		}
-		ptr += dp->d_ops->data_entsize(dep->namelen);
+		offset += xfs_dir2_data_entsize(mp, dep->namelen);
 	}
 	ASSERT((char *)sfep - (char *)sfp == size);
 
@@ -240,7 +342,7 @@ xfs_dir2_block_to_sf(
 	 * Convert the inode to local format and copy the data in.
 	 */
 	ASSERT(dp->i_df.if_bytes == 0);
-	xfs_init_local_fork(dp, XFS_DATA_FORK, dst, size);
+	xfs_init_local_fork(dp, XFS_DATA_FORK, sfp, size);
 	dp->i_d.di_format = XFS_DINODE_FMT_LOCAL;
 	dp->i_d.di_size = size;
 
@@ -248,7 +350,7 @@ xfs_dir2_block_to_sf(
 	xfs_dir2_sf_check(args);
 out:
 	xfs_trans_log_inode(args->trans, dp, logflags);
-	kmem_free(dst);
+	kmem_free(sfp);
 	return error;
 }
 
@@ -277,13 +379,7 @@ xfs_dir2_sf_addname(
 	ASSERT(xfs_dir2_sf_lookup(args) == -ENOENT);
 	dp = args->dp;
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
-	 * Make sure the shortform value has some of its header.
-	 */
-	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return -EIO;
-	}
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
@@ -291,7 +387,7 @@ xfs_dir2_sf_addname(
 	/*
 	 * Compute entry (and change in) size.
 	 */
-	incr_isize = dp->d_ops->sf_entsize(sfp, args->namelen);
+	incr_isize = xfs_dir2_sf_entsize(dp->i_mount, sfp, args->namelen);
 	objchange = 0;
 
 	/*
@@ -364,18 +460,17 @@ xfs_dir2_sf_addname_easy(
 	xfs_dir2_data_aoff_t	offset,		/* offset to use for new ent */
 	int			new_isize)	/* new directory size */
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			byteoff;	/* byte offset in sf dir */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
 
-	dp = args->dp;
-
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	byteoff = (int)((char *)sfep - (char *)sfp);
 	/*
 	 * Grow the in-inode space.
 	 */
-	xfs_idata_realloc(dp, dp->d_ops->sf_entsize(sfp, args->namelen),
+	xfs_idata_realloc(dp, xfs_dir2_sf_entsize(mp, sfp, args->namelen),
 			  XFS_DATA_FORK);
 	/*
 	 * Need to set up again due to realloc of the inode data.
@@ -388,8 +483,8 @@ xfs_dir2_sf_addname_easy(
 	sfep->namelen = args->namelen;
 	xfs_dir2_sf_put_offset(sfep, offset);
 	memcpy(sfep->name, args->name, sfep->namelen);
-	dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
-	dp->d_ops->sf_put_ftype(sfep, args->filetype);
+	xfs_dir2_sf_put_ino(mp, sfp, sfep, args->inumber);
+	xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 
 	/*
 	 * Update the header and inode.
@@ -416,9 +511,10 @@ xfs_dir2_sf_addname_hard(
 	int			objchange,	/* changing inode number size */
 	int			new_isize)	/* new directory size */
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			add_datasize;	/* data size need for new ent */
 	char			*buf;		/* buffer for old */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			eof;		/* reached end of old dir */
 	int			nbytes;		/* temp for byte copies */
 	xfs_dir2_data_aoff_t	new_offset;	/* next offset value */
@@ -432,8 +528,6 @@ xfs_dir2_sf_addname_hard(
 	/*
 	 * Copy the old directory to the stack buffer.
 	 */
-	dp = args->dp;
-
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	old_isize = (int)dp->i_d.di_size;
 	buf = kmem_alloc(old_isize, 0);
@@ -444,13 +538,13 @@ xfs_dir2_sf_addname_hard(
 	 * to insert the new entry.
 	 * If it's going to end up at the end then oldsfep will point there.
 	 */
-	for (offset = dp->d_ops->data_first_offset,
+	for (offset = args->geo->data_first_offset,
 	      oldsfep = xfs_dir2_sf_firstentry(oldsfp),
-	      add_datasize = dp->d_ops->data_entsize(args->namelen),
+	      add_datasize = xfs_dir2_data_entsize(mp, args->namelen),
 	      eof = (char *)oldsfep == &buf[old_isize];
 	     !eof;
-	     offset = new_offset + dp->d_ops->data_entsize(oldsfep->namelen),
-	      oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep),
+	     offset = new_offset + xfs_dir2_data_entsize(mp, oldsfep->namelen),
+	      oldsfep = xfs_dir2_sf_nextentry(mp, oldsfp, oldsfep),
 	      eof = (char *)oldsfep == &buf[old_isize]) {
 		new_offset = xfs_dir2_sf_get_offset(oldsfep);
 		if (offset + add_datasize <= new_offset)
@@ -479,8 +573,8 @@ xfs_dir2_sf_addname_hard(
 	sfep->namelen = args->namelen;
 	xfs_dir2_sf_put_offset(sfep, offset);
 	memcpy(sfep->name, args->name, sfep->namelen);
-	dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
-	dp->d_ops->sf_put_ftype(sfep, args->filetype);
+	xfs_dir2_sf_put_ino(mp, sfp, sfep, args->inumber);
+	xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 	sfp->count++;
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && !objchange)
 		sfp->i8count++;
@@ -488,7 +582,7 @@ xfs_dir2_sf_addname_hard(
 	 * If there's more left to copy, do that.
 	 */
 	if (!eof) {
-		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+		sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 		memcpy(sfep, oldsfep, old_isize - nbytes);
 	}
 	kmem_free(buf);
@@ -510,7 +604,8 @@ xfs_dir2_sf_addname_pick(
 	xfs_dir2_sf_entry_t	**sfepp,	/* out(1): new entry ptr */
 	xfs_dir2_data_aoff_t	*offsetp)	/* out(1): new offset */
 {
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			holefit;	/* found hole it will fit in */
 	int			i;		/* entry number */
 	xfs_dir2_data_aoff_t	offset;		/* data block offset */
@@ -519,11 +614,9 @@ xfs_dir2_sf_addname_pick(
 	int			size;		/* entry's data size */
 	int			used;		/* data bytes used */
 
-	dp = args->dp;
-
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-	size = dp->d_ops->data_entsize(args->namelen);
-	offset = dp->d_ops->data_first_offset;
+	size = xfs_dir2_data_entsize(mp, args->namelen);
+	offset = args->geo->data_first_offset;
 	sfep = xfs_dir2_sf_firstentry(sfp);
 	holefit = 0;
 	/*
@@ -535,8 +628,8 @@ xfs_dir2_sf_addname_pick(
 		if (!holefit)
 			holefit = offset + size <= xfs_dir2_sf_get_offset(sfep);
 		offset = xfs_dir2_sf_get_offset(sfep) +
-			 dp->d_ops->data_entsize(sfep->namelen);
-		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+			 xfs_dir2_data_entsize(mp, sfep->namelen);
+		sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 	}
 	/*
 	 * Calculate data bytes used excluding the new entry, if this
@@ -578,7 +671,8 @@ static void
 xfs_dir2_sf_check(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			i;		/* entry number */
 	int			i8count;	/* number of big inode#s */
 	xfs_ino_t		ino;		/* entry inode number */
@@ -586,23 +680,21 @@ xfs_dir2_sf_check(
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform dir entry */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
 
-	dp = args->dp;
-
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-	offset = dp->d_ops->data_first_offset;
-	ino = dp->d_ops->sf_get_parent_ino(sfp);
+	offset = args->geo->data_first_offset;
+	ino = xfs_dir2_sf_get_parent_ino(sfp);
 	i8count = ino > XFS_DIR2_MAX_SHORT_INUM;
 
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp);
 	     i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep)) {
 		ASSERT(xfs_dir2_sf_get_offset(sfep) >= offset);
-		ino = dp->d_ops->sf_get_ino(sfp, sfep);
+		ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
 		i8count += ino > XFS_DIR2_MAX_SHORT_INUM;
 		offset =
 			xfs_dir2_sf_get_offset(sfep) +
-			dp->d_ops->data_entsize(sfep->namelen);
-		ASSERT(dp->d_ops->sf_get_ftype(sfep) < XFS_DIR3_FT_MAX);
+			xfs_dir2_data_entsize(mp, sfep->namelen);
+		ASSERT(xfs_dir2_sf_get_ftype(mp, sfep) < XFS_DIR3_FT_MAX);
 	}
 	ASSERT(i8count == sfp->i8count);
 	ASSERT((char *)sfep - (char *)sfp == dp->i_d.di_size);
@@ -622,22 +714,16 @@ xfs_dir2_sf_verify(
 	struct xfs_dir2_sf_entry	*sfep;
 	struct xfs_dir2_sf_entry	*next_sfep;
 	char				*endp;
-	const struct xfs_dir_ops	*dops;
 	struct xfs_ifork		*ifp;
 	xfs_ino_t			ino;
 	int				i;
 	int				i8count;
 	int				offset;
-	int				size;
+	int64_t				size;
 	int				error;
 	uint8_t				filetype;
 
 	ASSERT(ip->i_d.di_format == XFS_DINODE_FMT_LOCAL);
-	/*
-	 * xfs_iread calls us before xfs_setup_inode sets up ip->d_ops,
-	 * so we can only trust the mountpoint to have the right pointer.
-	 */
-	dops = xfs_dir_get_ops(mp, NULL);
 
 	ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
 	sfp = (struct xfs_dir2_sf_hdr *)ifp->if_u1.if_data;
@@ -653,12 +739,12 @@ xfs_dir2_sf_verify(
 	endp = (char *)sfp + size;
 
 	/* Check .. entry */
-	ino = dops->sf_get_parent_ino(sfp);
+	ino = xfs_dir2_sf_get_parent_ino(sfp);
 	i8count = ino > XFS_DIR2_MAX_SHORT_INUM;
 	error = xfs_dir_ino_validate(mp, ino);
 	if (error)
 		return __this_address;
-	offset = dops->data_first_offset;
+	offset = mp->m_dir_geo->data_first_offset;
 
 	/* Check all reported entries */
 	sfep = xfs_dir2_sf_firstentry(sfp);
@@ -680,7 +766,7 @@ xfs_dir2_sf_verify(
 		 * within the data buffer.  The next entry starts after the
 		 * name component, so nextentry is an acceptable test.
 		 */
-		next_sfep = dops->sf_nextentry(sfp, sfep);
+		next_sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 		if (endp < (char *)next_sfep)
 			return __this_address;
 
@@ -689,19 +775,19 @@ xfs_dir2_sf_verify(
 			return __this_address;
 
 		/* Check the inode number. */
-		ino = dops->sf_get_ino(sfp, sfep);
+		ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
 		i8count += ino > XFS_DIR2_MAX_SHORT_INUM;
 		error = xfs_dir_ino_validate(mp, ino);
 		if (error)
 			return __this_address;
 
 		/* Check the file type. */
-		filetype = dops->sf_get_ftype(sfep);
+		filetype = xfs_dir2_sf_get_ftype(mp, sfep);
 		if (filetype >= XFS_DIR3_FT_MAX)
 			return __this_address;
 
 		offset = xfs_dir2_sf_get_offset(sfep) +
-				dops->data_entsize(sfep->namelen);
+				xfs_dir2_data_entsize(mp, sfep->namelen);
 
 		sfep = next_sfep;
 	}
@@ -763,7 +849,7 @@ xfs_dir2_sf_create(
 	/*
 	 * Now can put in the inode number, since i8count is set.
 	 */
-	dp->d_ops->sf_put_parent_ino(sfp, pino);
+	xfs_dir2_sf_put_parent_ino(sfp, pino);
 	sfp->count = 0;
 	dp->i_d.di_size = size;
 	xfs_dir2_sf_check(args);
@@ -779,7 +865,8 @@ int						/* error */
 xfs_dir2_sf_lookup(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			i;		/* entry index */
 	int			error;
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
@@ -790,16 +877,9 @@ xfs_dir2_sf_lookup(
 	trace_xfs_dir2_sf_lookup(args);
 
 	xfs_dir2_sf_check(args);
-	dp = args->dp;
 
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
-	 * Bail out if the directory is way too short.
-	 */
-	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return -EIO;
-	}
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
@@ -818,7 +898,7 @@ xfs_dir2_sf_lookup(
 	 */
 	if (args->namelen == 2 &&
 	    args->name[0] == '.' && args->name[1] == '.') {
-		args->inumber = dp->d_ops->sf_get_parent_ino(sfp);
+		args->inumber = xfs_dir2_sf_get_parent_ino(sfp);
 		args->cmpresult = XFS_CMP_EXACT;
 		args->filetype = XFS_DIR3_FT_DIR;
 		return -EEXIST;
@@ -828,18 +908,17 @@ xfs_dir2_sf_lookup(
 	 */
 	ci_sfep = NULL;
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep)) {
 		/*
 		 * Compare name and if it's an exact match, return the inode
 		 * number. If it's the first case-insensitive match, store the
 		 * inode number and continue looking for an exact match.
 		 */
-		cmp = dp->i_mount->m_dirnameops->compname(args, sfep->name,
-								sfep->namelen);
+		cmp = xfs_dir2_compname(args, sfep->name, sfep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			args->cmpresult = cmp;
-			args->inumber = dp->d_ops->sf_get_ino(sfp, sfep);
-			args->filetype = dp->d_ops->sf_get_ftype(sfep);
+			args->inumber = xfs_dir2_sf_get_ino(mp, sfp, sfep);
+			args->filetype = xfs_dir2_sf_get_ftype(mp, sfep);
 			if (cmp == XFS_CMP_EXACT)
 				return -EEXIST;
 			ci_sfep = sfep;
@@ -864,8 +943,9 @@ int						/* error */
 xfs_dir2_sf_removename(
 	xfs_da_args_t		*args)
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			byteoff;	/* offset of removed entry */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			entsize;	/* this entry's size */
 	int			i;		/* shortform entry index */
 	int			newsize;	/* new inode size */
@@ -875,17 +955,9 @@ xfs_dir2_sf_removename(
 
 	trace_xfs_dir2_sf_removename(args);
 
-	dp = args->dp;
-
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
 	oldsize = (int)dp->i_d.di_size;
-	/*
-	 * Bail out if the directory is way too short.
-	 */
-	if (oldsize < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return -EIO;
-	}
+	ASSERT(oldsize >= offsetof(struct xfs_dir2_sf_hdr, parent));
 	ASSERT(dp->i_df.if_bytes == oldsize);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
@@ -895,10 +967,10 @@ xfs_dir2_sf_removename(
 	 * Find the one we're deleting.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep)) {
 		if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
 								XFS_CMP_EXACT) {
-			ASSERT(dp->d_ops->sf_get_ino(sfp, sfep) ==
+			ASSERT(xfs_dir2_sf_get_ino(mp, sfp, sfep) ==
 			       args->inumber);
 			break;
 		}
@@ -912,7 +984,7 @@ xfs_dir2_sf_removename(
 	 * Calculate sizes.
 	 */
 	byteoff = (int)((char *)sfep - (char *)sfp);
-	entsize = dp->d_ops->sf_entsize(sfp, args->namelen);
+	entsize = xfs_dir2_sf_entsize(mp, sfp, args->namelen);
 	newsize = oldsize - entsize;
 	/*
 	 * Copy the part if any after the removed entry, sliding it down.
@@ -945,13 +1017,35 @@ xfs_dir2_sf_removename(
 }
 
 /*
+ * Check whether the sf dir replace operation need more blocks.
+ */
+bool
+xfs_dir2_sf_replace_needblock(
+	struct xfs_inode	*dp,
+	xfs_ino_t		inum)
+{
+	int			newsize;
+	struct xfs_dir2_sf_hdr	*sfp;
+
+	if (dp->i_d.di_format != XFS_DINODE_FMT_LOCAL)
+		return false;
+
+	sfp = (struct xfs_dir2_sf_hdr *)dp->i_df.if_u1.if_data;
+	newsize = dp->i_df.if_bytes + (sfp->count + 1) * XFS_INO64_DIFF;
+
+	return inum > XFS_DIR2_MAX_SHORT_INUM &&
+	       sfp->i8count == 0 && newsize > XFS_IFORK_DSIZE(dp);
+}
+
+/*
  * Replace the inode number of an entry in a shortform directory.
  */
 int						/* error */
 xfs_dir2_sf_replace(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
-	xfs_inode_t		*dp;		/* incore directory inode */
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	int			i;		/* entry index */
 	xfs_ino_t		ino=0;		/* entry old inode number */
 	int			i8elevated;	/* sf_toino8 set i8count=1 */
@@ -960,16 +1054,8 @@ xfs_dir2_sf_replace(
 
 	trace_xfs_dir2_sf_replace(args);
 
-	dp = args->dp;
-
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
-	/*
-	 * Bail out if the shortform directory is way too small.
-	 */
-	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return -EIO;
-	}
+	ASSERT(dp->i_d.di_size >= offsetof(struct xfs_dir2_sf_hdr, parent));
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
@@ -980,17 +1066,14 @@ xfs_dir2_sf_replace(
 	 */
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && sfp->i8count == 0) {
 		int	error;			/* error return value */
-		int	newsize;		/* new inode size */
 
-		newsize = dp->i_df.if_bytes + (sfp->count + 1) * XFS_INO64_DIFF;
 		/*
 		 * Won't fit as shortform, convert to block then do replace.
 		 */
-		if (newsize > XFS_IFORK_DSIZE(dp)) {
+		if (xfs_dir2_sf_replace_needblock(dp, args->inumber)) {
 			error = xfs_dir2_sf_to_block(args);
-			if (error) {
+			if (error)
 				return error;
-			}
 			return xfs_dir2_block_replace(args);
 		}
 		/*
@@ -1008,22 +1091,23 @@ xfs_dir2_sf_replace(
 	 */
 	if (args->namelen == 2 &&
 	    args->name[0] == '.' && args->name[1] == '.') {
-		ino = dp->d_ops->sf_get_parent_ino(sfp);
+		ino = xfs_dir2_sf_get_parent_ino(sfp);
 		ASSERT(args->inumber != ino);
-		dp->d_ops->sf_put_parent_ino(sfp, args->inumber);
+		xfs_dir2_sf_put_parent_ino(sfp, args->inumber);
 	}
 	/*
 	 * Normal entry, look for the name.
 	 */
 	else {
 		for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-		     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
+		     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep)) {
 			if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
 								XFS_CMP_EXACT) {
-				ino = dp->d_ops->sf_get_ino(sfp, sfep);
+				ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
 				ASSERT(args->inumber != ino);
-				dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
-				dp->d_ops->sf_put_ftype(sfep, args->filetype);
+				xfs_dir2_sf_put_ino(mp, sfp, sfep,
+						args->inumber);
+				xfs_dir2_sf_put_ftype(mp, sfep, args->filetype);
 				break;
 			}
 		}
@@ -1076,8 +1160,9 @@ static void
 xfs_dir2_sf_toino4(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	char			*buf;		/* old dir's buffer */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			i;		/* entry index */
 	int			newsize;	/* new inode size */
 	xfs_dir2_sf_entry_t	*oldsfep;	/* old sf entry */
@@ -1088,8 +1173,6 @@ xfs_dir2_sf_toino4(
 
 	trace_xfs_dir2_sf_toino4(args);
 
-	dp = args->dp;
-
 	/*
 	 * Copy the old directory to the buffer.
 	 * Then nuke it from the inode, and add the new buffer to the inode.
@@ -1116,21 +1199,22 @@ xfs_dir2_sf_toino4(
 	 */
 	sfp->count = oldsfp->count;
 	sfp->i8count = 0;
-	dp->d_ops->sf_put_parent_ino(sfp, dp->d_ops->sf_get_parent_ino(oldsfp));
+	xfs_dir2_sf_put_parent_ino(sfp, xfs_dir2_sf_get_parent_ino(oldsfp));
 	/*
 	 * Copy the entries field by field.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp),
 		    oldsfep = xfs_dir2_sf_firstentry(oldsfp);
 	     i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep),
-		  oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep),
+		  oldsfep = xfs_dir2_sf_nextentry(mp, oldsfp, oldsfep)) {
 		sfep->namelen = oldsfep->namelen;
 		memcpy(sfep->offset, oldsfep->offset, sizeof(sfep->offset));
 		memcpy(sfep->name, oldsfep->name, sfep->namelen);
-		dp->d_ops->sf_put_ino(sfp, sfep,
-				      dp->d_ops->sf_get_ino(oldsfp, oldsfep));
-		dp->d_ops->sf_put_ftype(sfep, dp->d_ops->sf_get_ftype(oldsfep));
+		xfs_dir2_sf_put_ino(mp, sfp, sfep,
+				xfs_dir2_sf_get_ino(mp, oldsfp, oldsfep));
+		xfs_dir2_sf_put_ftype(mp, sfep,
+				xfs_dir2_sf_get_ftype(mp, oldsfep));
 	}
 	/*
 	 * Clean up the inode.
@@ -1149,8 +1233,9 @@ static void
 xfs_dir2_sf_toino8(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
+	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	char			*buf;		/* old dir's buffer */
-	xfs_inode_t		*dp;		/* incore directory inode */
 	int			i;		/* entry index */
 	int			newsize;	/* new inode size */
 	xfs_dir2_sf_entry_t	*oldsfep;	/* old sf entry */
@@ -1161,8 +1246,6 @@ xfs_dir2_sf_toino8(
 
 	trace_xfs_dir2_sf_toino8(args);
 
-	dp = args->dp;
-
 	/*
 	 * Copy the old directory to the buffer.
 	 * Then nuke it from the inode, and add the new buffer to the inode.
@@ -1189,21 +1272,22 @@ xfs_dir2_sf_toino8(
 	 */
 	sfp->count = oldsfp->count;
 	sfp->i8count = 1;
-	dp->d_ops->sf_put_parent_ino(sfp, dp->d_ops->sf_get_parent_ino(oldsfp));
+	xfs_dir2_sf_put_parent_ino(sfp, xfs_dir2_sf_get_parent_ino(oldsfp));
 	/*
 	 * Copy the entries field by field.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp),
 		    oldsfep = xfs_dir2_sf_firstentry(oldsfp);
 	     i < sfp->count;
-	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep),
-		  oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep)) {
+	     i++, sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep),
+		  oldsfep = xfs_dir2_sf_nextentry(mp, oldsfp, oldsfep)) {
 		sfep->namelen = oldsfep->namelen;
 		memcpy(sfep->offset, oldsfep->offset, sizeof(sfep->offset));
 		memcpy(sfep->name, oldsfep->name, sfep->namelen);
-		dp->d_ops->sf_put_ino(sfp, sfep,
-				      dp->d_ops->sf_get_ino(oldsfp, oldsfep));
-		dp->d_ops->sf_put_ftype(sfep, dp->d_ops->sf_get_ftype(oldsfep));
+		xfs_dir2_sf_put_ino(mp, sfp, sfep,
+				xfs_dir2_sf_get_ino(mp, oldsfp, oldsfep));
+		xfs_dir2_sf_put_ftype(mp, sfep,
+				xfs_dir2_sf_get_ftype(mp, oldsfep));
 	}
 	/*
 	 * Clean up the inode.
diff --git a/fs/xfs/libxfs/xfs_dquot_buf.c b/fs/xfs/libxfs/xfs_dquot_buf.c
index e8bd688a4073..bedc1e752b60 100644
--- a/fs/xfs/libxfs/xfs_dquot_buf.c
+++ b/fs/xfs/libxfs/xfs_dquot_buf.c
@@ -35,10 +35,10 @@ xfs_calc_dquots_per_chunk(
 
 xfs_failaddr_t
 xfs_dquot_verify(
-	struct xfs_mount *mp,
-	xfs_disk_dquot_t *ddq,
-	xfs_dqid_t	 id,
-	uint		 type)	  /* used only during quotacheck */
+	struct xfs_mount	*mp,
+	struct xfs_disk_dquot	*ddq,
+	xfs_dqid_t		id,
+	uint			type)	/* used only during quotacheck */
 {
 	/*
 	 * We can encounter an uninitialized dquot buffer for 2 reasons:
diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index c968b60cee15..045556e78ee2 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -497,6 +497,23 @@ static inline bool xfs_sb_version_hascrc(struct xfs_sb *sbp)
 	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
 }
 
+/*
+ * v5 file systems support V3 inodes only, earlier file systems support
+ * v2 and v1 inodes.
+ */
+static inline bool xfs_sb_version_has_v3inode(struct xfs_sb *sbp)
+{
+	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
+}
+
+static inline bool xfs_dinode_good_version(struct xfs_sb *sbp,
+		uint8_t version)
+{
+	if (xfs_sb_version_has_v3inode(sbp))
+		return version == 3;
+	return version == 1 || version == 2;
+}
+
 static inline bool xfs_sb_version_has_pquotino(struct xfs_sb *sbp)
 {
 	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
@@ -560,7 +577,6 @@ xfs_is_quota_inode(struct xfs_sb *sbp, xfs_ino_t ino)
 
 #define XFS_SB_DADDR		((xfs_daddr_t)0) /* daddr in filesystem/ag */
 #define	XFS_SB_BLOCK(mp)	XFS_HDR_BLOCK(mp, XFS_SB_DADDR)
-#define XFS_BUF_TO_SBP(bp)	((xfs_dsb_t *)((bp)->b_addr))
 
 #define	XFS_HDR_BLOCK(mp,d)	((xfs_agblock_t)XFS_BB_TO_FSBT(mp,d))
 #define	XFS_DADDR_TO_FSB(mp,d)	XFS_AGB_TO_FSB(mp, \
@@ -707,7 +723,6 @@ typedef struct xfs_agf {
 /* disk block (xfs_daddr_t) in the AG */
 #define XFS_AGF_DADDR(mp)	((xfs_daddr_t)(1 << (mp)->m_sectbb_log))
 #define	XFS_AGF_BLOCK(mp)	XFS_HDR_BLOCK(mp, XFS_AGF_DADDR(mp))
-#define	XFS_BUF_TO_AGF(bp)	((xfs_agf_t *)((bp)->b_addr))
 
 /*
  * Size of the unlinked inode hash table in the agi.
@@ -775,7 +790,6 @@ typedef struct xfs_agi {
 /* disk block (xfs_daddr_t) in the AG */
 #define XFS_AGI_DADDR(mp)	((xfs_daddr_t)(2 << (mp)->m_sectbb_log))
 #define	XFS_AGI_BLOCK(mp)	XFS_HDR_BLOCK(mp, XFS_AGI_DADDR(mp))
-#define	XFS_BUF_TO_AGI(bp)	((xfs_agi_t *)((bp)->b_addr))
 
 /*
  * The third a.g. block contains the a.g. freelist, an array
@@ -783,21 +797,15 @@ typedef struct xfs_agi {
  */
 #define XFS_AGFL_DADDR(mp)	((xfs_daddr_t)(3 << (mp)->m_sectbb_log))
 #define	XFS_AGFL_BLOCK(mp)	XFS_HDR_BLOCK(mp, XFS_AGFL_DADDR(mp))
-#define	XFS_BUF_TO_AGFL(bp)	((xfs_agfl_t *)((bp)->b_addr))
-
-#define XFS_BUF_TO_AGFL_BNO(mp, bp) \
-	(xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
-		&(XFS_BUF_TO_AGFL(bp)->agfl_bno[0]) : \
-		(__be32 *)(bp)->b_addr)
+#define	XFS_BUF_TO_AGFL(bp)	((struct xfs_agfl *)((bp)->b_addr))
 
-typedef struct xfs_agfl {
+struct xfs_agfl {
 	__be32		agfl_magicnum;
 	__be32		agfl_seqno;
 	uuid_t		agfl_uuid;
 	__be64		agfl_lsn;
 	__be32		agfl_crc;
-	__be32		agfl_bno[];	/* actually xfs_agfl_size(mp) */
-} __attribute__((packed)) xfs_agfl_t;
+} __attribute__((packed));
 
 #define XFS_AGFL_CRC_OFF	offsetof(struct xfs_agfl, agfl_crc)
 
@@ -920,13 +928,13 @@ static inline uint xfs_dinode_size(int version)
  * This enum is used in string mapping in xfs_trace.h; please keep the
  * TRACE_DEFINE_ENUMs for it up to date.
  */
-typedef enum xfs_dinode_fmt {
+enum xfs_dinode_fmt {
 	XFS_DINODE_FMT_DEV,		/* xfs_dev_t */
 	XFS_DINODE_FMT_LOCAL,		/* bulk data */
 	XFS_DINODE_FMT_EXTENTS,		/* struct xfs_bmbt_rec */
 	XFS_DINODE_FMT_BTREE,		/* struct xfs_bmdr_block */
 	XFS_DINODE_FMT_UUID		/* added long ago, but never used */
-} xfs_dinode_fmt_t;
+};
 
 #define XFS_INODE_FORMAT_STR \
 	{ XFS_DINODE_FMT_DEV,		"dev" }, \
@@ -946,8 +954,12 @@ typedef enum xfs_dinode_fmt {
 /*
  * Inode size for given fs.
  */
-#define XFS_LITINO(mp, version) \
-	((int)(((mp)->m_sb.sb_inodesize) - xfs_dinode_size(version)))
+#define XFS_DINODE_SIZE(sbp) \
+	(xfs_sb_version_has_v3inode(sbp) ? \
+		sizeof(struct xfs_dinode) : \
+		offsetof(struct xfs_dinode, di_crc))
+#define XFS_LITINO(mp) \
+	((mp)->m_sb.sb_inodesize - XFS_DINODE_SIZE(&(mp)->m_sb))
 
 /*
  * Inode data & attribute fork sizes, per inode.
@@ -956,13 +968,9 @@ typedef enum xfs_dinode_fmt {
 #define XFS_DFORK_BOFF(dip)		((int)((dip)->di_forkoff << 3))
 
 #define XFS_DFORK_DSIZE(dip,mp) \
-	(XFS_DFORK_Q(dip) ? \
-		XFS_DFORK_BOFF(dip) : \
-		XFS_LITINO(mp, (dip)->di_version))
+	(XFS_DFORK_Q(dip) ? XFS_DFORK_BOFF(dip) : XFS_LITINO(mp))
 #define XFS_DFORK_ASIZE(dip,mp) \
-	(XFS_DFORK_Q(dip) ? \
-		XFS_LITINO(mp, (dip)->di_version) - XFS_DFORK_BOFF(dip) : \
-		0)
+	(XFS_DFORK_Q(dip) ? XFS_LITINO(mp) - XFS_DFORK_BOFF(dip) : 0)
 #define XFS_DFORK_SIZE(dip,mp,w) \
 	((w) == XFS_DATA_FORK ? \
 		XFS_DFORK_DSIZE(dip, mp) : \
@@ -1144,11 +1152,11 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev)
 
 /*
  * This is the main portion of the on-disk representation of quota
- * information for a user. This is the q_core of the xfs_dquot_t that
+ * information for a user. This is the q_core of the struct xfs_dquot that
  * is kept in kernel memory. We pad this with some more expansion room
  * to construct the on disk structure.
  */
-typedef struct	xfs_disk_dquot {
+struct xfs_disk_dquot {
 	__be16		d_magic;	/* dquot magic = XFS_DQUOT_MAGIC */
 	__u8		d_version;	/* dquot version */
 	__u8		d_flags;	/* XFS_DQ_USER/PROJ/GROUP */
@@ -1171,15 +1179,15 @@ typedef struct	xfs_disk_dquot {
 	__be32		d_rtbtimer;	/* similar to above; for RT disk blocks */
 	__be16		d_rtbwarns;	/* warnings issued wrt RT disk blocks */
 	__be16		d_pad;
-} xfs_disk_dquot_t;
+};
 
 /*
  * This is what goes on disk. This is separated from the xfs_disk_dquot because
  * carrying the unnecessary padding would be a waste of memory.
  */
 typedef struct xfs_dqblk {
-	xfs_disk_dquot_t  dd_diskdq;	/* portion that lives incore as well */
-	char		  dd_fill[4];	/* filling for posterity */
+	struct xfs_disk_dquot	dd_diskdq; /* portion living incore as well */
+	char			dd_fill[4];/* filling for posterity */
 
 	/*
 	 * These two are only present on filesystems with the CRC bits set.
@@ -1540,6 +1548,13 @@ typedef struct xfs_bmdr_block {
 #define BMBT_BLOCKCOUNT_BITLEN	21
 
 #define BMBT_STARTOFF_MASK	((1ULL << BMBT_STARTOFF_BITLEN) - 1)
+#define BMBT_BLOCKCOUNT_MASK	((1ULL << BMBT_BLOCKCOUNT_BITLEN) - 1)
+
+/*
+ * bmbt records have a file offset (block) field that is 54 bits wide, so this
+ * is the largest xfs_fileoff_t that we ever expect to see.
+ */
+#define XFS_MAX_FILEOFF		(BMBT_STARTOFF_MASK + BMBT_BLOCKCOUNT_MASK)
 
 typedef struct xfs_bmbt_rec {
 	__be64			l0, l1;
diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h
index e9371a8e0e26..245188e4f6d3 100644
--- a/fs/xfs/libxfs/xfs_fs.h
+++ b/fs/xfs/libxfs/xfs_fs.h
@@ -324,7 +324,7 @@ typedef struct xfs_growfs_rt {
  * Structures returned from ioctl XFS_IOC_FSBULKSTAT & XFS_IOC_FSBULKSTAT_SINGLE
  */
 typedef struct xfs_bstime {
-	time_t		tv_sec;		/* seconds		*/
+	__kernel_long_t tv_sec;		/* seconds		*/
 	__s32		tv_nsec;	/* and nanoseconds	*/
 } xfs_bstime_t;
 
@@ -416,7 +416,7 @@ struct xfs_bulkstat {
 
 /*
  * Project quota id helpers (previously projid was 16bit only
- * and using two 16bit values to hold new 32bit projid was choosen
+ * and using two 16bit values to hold new 32bit projid was chosen
  * to retain compatibility with "old" filesystems).
  */
 static inline uint32_t
@@ -568,10 +568,40 @@ typedef struct xfs_fsop_setdm_handlereq {
 	struct fsdmidata		__user *data;	/* DMAPI data	*/
 } xfs_fsop_setdm_handlereq_t;
 
+/*
+ * Flags passed in xfs_attr_multiop.am_flags for the attr ioctl interface.
+ *
+ * NOTE: Must match the values declared in libattr without the XFS_IOC_ prefix.
+ */
+#define XFS_IOC_ATTR_ROOT	0x0002	/* use attrs in root namespace */
+#define XFS_IOC_ATTR_SECURE	0x0008	/* use attrs in security namespace */
+#define XFS_IOC_ATTR_CREATE	0x0010	/* fail if attr already exists */
+#define XFS_IOC_ATTR_REPLACE	0x0020	/* fail if attr does not exist */
+
 typedef struct xfs_attrlist_cursor {
 	__u32		opaque[4];
 } xfs_attrlist_cursor_t;
 
+/*
+ * Define how lists of attribute names are returned to userspace from the
+ * XFS_IOC_ATTRLIST_BY_HANDLE ioctl.  struct xfs_attrlist is the header at the
+ * beginning of the returned buffer, and a each entry in al_offset contains the
+ * relative offset of an xfs_attrlist_ent containing the actual entry.
+ *
+ * NOTE: struct xfs_attrlist must match struct attrlist defined in libattr, and
+ * struct xfs_attrlist_ent must match struct attrlist_ent defined in libattr.
+ */
+struct xfs_attrlist {
+	__s32	al_count;	/* number of entries in attrlist */
+	__s32	al_more;	/* T/F: more attrs (do call again) */
+	__s32	al_offset[1];	/* byte offsets of attrs [var-sized] */
+};
+
+struct xfs_attrlist_ent {	/* data from attr_list() */
+	__u32	a_valuelen;	/* number bytes in value of attr */
+	char	a_name[1];	/* attr name (NULL terminated) */
+};
+
 typedef struct xfs_fsop_attrlist_handlereq {
 	struct xfs_fsop_handlereq	hreq; /* handle interface structure */
 	struct xfs_attrlist_cursor	pos; /* opaque cookie, list offset */
@@ -589,7 +619,7 @@ typedef struct xfs_attr_multiop {
 	void		__user *am_attrname;
 	void		__user *am_attrvalue;
 	__u32		am_length;
-	__u32		am_flags;
+	__u32		am_flags; /* XFS_IOC_ATTR_* */
 } xfs_attr_multiop_t;
 
 typedef struct xfs_fsop_attrmulti_handlereq {
diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index 588d44613094..7fcf62b324b0 100644
--- a/fs/xfs/libxfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -105,7 +105,7 @@ xfs_inobt_get_rec(
 	int				*stat)
 {
 	struct xfs_mount		*mp = cur->bc_mp;
-	xfs_agnumber_t			agno = cur->bc_private.a.agno;
+	xfs_agnumber_t			agno = cur->bc_ag.agno;
 	union xfs_btree_rec		*rec;
 	int				error;
 	uint64_t			realfree;
@@ -177,7 +177,7 @@ xfs_inobt_insert(
 	xfs_btnum_t		btnum)
 {
 	struct xfs_btree_cur	*cur;
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
+	struct xfs_agi		*agi = agbp->b_addr;
 	xfs_agnumber_t		agno = be32_to_cpu(agi->agi_seqno);
 	xfs_agino_t		thisino;
 	int			i;
@@ -276,6 +276,7 @@ xfs_ialloc_inode_init(
 	int			i, j;
 	xfs_daddr_t		d;
 	xfs_ino_t		ino = 0;
+	int			error;
 
 	/*
 	 * Loop over the new block(s), filling in the inodes.  For small block
@@ -303,7 +304,7 @@ xfs_ialloc_inode_init(
 	 * That means for v3 inode we log the entire buffer rather than just the
 	 * inode cores.
 	 */
-	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+	if (xfs_sb_version_has_v3inode(&mp->m_sb)) {
 		version = 3;
 		ino = XFS_AGINO_TO_INO(mp, agno, XFS_AGB_TO_AGINO(mp, agbno));
 
@@ -327,19 +328,18 @@ xfs_ialloc_inode_init(
 		 */
 		d = XFS_AGB_TO_DADDR(mp, agno, agbno +
 				(j * M_IGEO(mp)->blocks_per_cluster));
-		fbuf = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
-					 mp->m_bsize *
-					 M_IGEO(mp)->blocks_per_cluster,
-					 XBF_UNMAPPED);
-		if (!fbuf)
-			return -ENOMEM;
+		error = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
+				mp->m_bsize * M_IGEO(mp)->blocks_per_cluster,
+				XBF_UNMAPPED, &fbuf);
+		if (error)
+			return error;
 
 		/* Initialize the inode buffers and log them appropriately. */
 		fbuf->b_ops = &xfs_inode_buf_ops;
 		xfs_buf_zero(fbuf, 0, BBTOB(fbuf->b_length));
 		for (i = 0; i < M_IGEO(mp)->inodes_per_cluster; i++) {
 			int	ioffset = i << mp->m_sb.sb_inodelog;
-			uint	isize = xfs_dinode_size(version);
+			uint	isize = XFS_DINODE_SIZE(&mp->m_sb);
 
 			free = xfs_make_iptr(mp, fbuf, i);
 			free->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
@@ -525,7 +525,7 @@ xfs_inobt_insert_sprec(
 	bool				merge)	/* merge or replace */
 {
 	struct xfs_btree_cur		*cur;
-	struct xfs_agi			*agi = XFS_BUF_TO_AGI(agbp);
+	struct xfs_agi			*agi = agbp->b_addr;
 	xfs_agnumber_t			agno = be32_to_cpu(agi->agi_seqno);
 	int				error;
 	int				i;
@@ -544,7 +544,10 @@ xfs_inobt_insert_sprec(
 					     nrec->ir_free, &i);
 		if (error)
 			goto error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		goto out;
 	}
@@ -557,17 +560,23 @@ xfs_inobt_insert_sprec(
 		error = xfs_inobt_get_rec(cur, &rec, &i);
 		if (error)
 			goto error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-					rec.ir_startino == nrec->ir_startino,
-					error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
+		if (XFS_IS_CORRUPT(mp, rec.ir_startino != nrec->ir_startino)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		/*
 		 * This should never fail. If we have coexisting records that
 		 * cannot merge, something is seriously wrong.
 		 */
-		XFS_WANT_CORRUPTED_GOTO(mp, __xfs_inobt_can_merge(nrec, &rec),
-					error);
+		if (XFS_IS_CORRUPT(mp, !__xfs_inobt_can_merge(nrec, &rec))) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		trace_xfs_irec_merge_pre(mp, agno, rec.ir_startino,
 					 rec.ir_holemask, nrec->ir_startino,
@@ -649,7 +658,7 @@ xfs_ialloc_ag_alloc(
 	 * chunk of inodes.  If the filesystem is striped, this will fill
 	 * an entire stripe unit with inodes.
 	 */
-	agi = XFS_BUF_TO_AGI(agbp);
+	agi = agbp->b_addr;
 	newino = be32_to_cpu(agi->agi_newino);
 	agno = be32_to_cpu(agi->agi_seqno);
 	args.agbno = XFS_AGINO_TO_AGBNO(args.mp, newino) +
@@ -1057,7 +1066,8 @@ xfs_ialloc_next_rec(
 		error = xfs_inobt_get_rec(cur, rec, &i);
 		if (error)
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 
 	return 0;
@@ -1081,7 +1091,8 @@ xfs_ialloc_get_rec(
 		error = xfs_inobt_get_rec(cur, rec, &i);
 		if (error)
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+			return -EFSCORRUPTED;
 	}
 
 	return 0;
@@ -1119,7 +1130,7 @@ xfs_dialloc_ag_inobt(
 	xfs_ino_t		*inop)
 {
 	struct xfs_mount	*mp = tp->t_mountp;
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
+	struct xfs_agi		*agi = agbp->b_addr;
 	xfs_agnumber_t		agno = be32_to_cpu(agi->agi_seqno);
 	xfs_agnumber_t		pagno = XFS_INO_TO_AGNO(mp, parent);
 	xfs_agino_t		pagino = XFS_INO_TO_AGINO(mp, parent);
@@ -1161,12 +1172,18 @@ xfs_dialloc_ag_inobt(
 		error = xfs_inobt_lookup(cur, pagino, XFS_LOOKUP_LE, &i);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		error = xfs_inobt_get_rec(cur, &rec, &j);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, j == 1, error0);
+		if (XFS_IS_CORRUPT(mp, j != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 
 		if (rec.ir_freecount > 0) {
 			/*
@@ -1321,19 +1338,28 @@ xfs_dialloc_ag_inobt(
 	error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
 	if (error)
 		goto error0;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 
 	for (;;) {
 		error = xfs_inobt_get_rec(cur, &rec, &i);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 		if (rec.ir_freecount > 0)
 			break;
 		error = xfs_btree_increment(cur, 0, &i);
 		if (error)
 			goto error0;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto error0;
+		}
 	}
 
 alloc_inode:
@@ -1393,7 +1419,8 @@ xfs_dialloc_ag_finobt_near(
 		error = xfs_inobt_get_rec(lcur, rec, &i);
 		if (error)
 			return error;
-		XFS_WANT_CORRUPTED_RETURN(lcur->bc_mp, i == 1);
+		if (XFS_IS_CORRUPT(lcur->bc_mp, i != 1))
+			return -EFSCORRUPTED;
 
 		/*
 		 * See if we've landed in the parent inode record. The finobt
@@ -1416,10 +1443,16 @@ xfs_dialloc_ag_finobt_near(
 		error = xfs_inobt_get_rec(rcur, &rrec, &j);
 		if (error)
 			goto error_rcur;
-		XFS_WANT_CORRUPTED_GOTO(lcur->bc_mp, j == 1, error_rcur);
+		if (XFS_IS_CORRUPT(lcur->bc_mp, j != 1)) {
+			error = -EFSCORRUPTED;
+			goto error_rcur;
+		}
 	}
 
-	XFS_WANT_CORRUPTED_GOTO(lcur->bc_mp, i == 1 || j == 1, error_rcur);
+	if (XFS_IS_CORRUPT(lcur->bc_mp, i != 1 && j != 1)) {
+		error = -EFSCORRUPTED;
+		goto error_rcur;
+	}
 	if (i == 1 && j == 1) {
 		/*
 		 * Both the left and right records are valid. Choose the closer
@@ -1472,7 +1505,8 @@ xfs_dialloc_ag_finobt_newino(
 			error = xfs_inobt_get_rec(cur, rec, &i);
 			if (error)
 				return error;
-			XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+			if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+				return -EFSCORRUPTED;
 			return 0;
 		}
 	}
@@ -1483,12 +1517,14 @@ xfs_dialloc_ag_finobt_newino(
 	error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+		return -EFSCORRUPTED;
 
 	error = xfs_inobt_get_rec(cur, rec, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+		return -EFSCORRUPTED;
 
 	return 0;
 }
@@ -1510,20 +1546,24 @@ xfs_dialloc_ag_update_inobt(
 	error = xfs_inobt_lookup(cur, frec->ir_startino, XFS_LOOKUP_EQ, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+		return -EFSCORRUPTED;
 
 	error = xfs_inobt_get_rec(cur, &rec, &i);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1))
+		return -EFSCORRUPTED;
 	ASSERT((XFS_AGINO_TO_OFFSET(cur->bc_mp, rec.ir_startino) %
 				   XFS_INODES_PER_CHUNK) == 0);
 
 	rec.ir_free &= ~XFS_INOBT_MASK(offset);
 	rec.ir_freecount--;
 
-	XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, (rec.ir_free == frec->ir_free) &&
-				  (rec.ir_freecount == frec->ir_freecount));
+	if (XFS_IS_CORRUPT(cur->bc_mp,
+			   rec.ir_free != frec->ir_free ||
+			   rec.ir_freecount != frec->ir_freecount))
+		return -EFSCORRUPTED;
 
 	return xfs_inobt_update(cur, &rec);
 }
@@ -1543,7 +1583,7 @@ xfs_dialloc_ag(
 	xfs_ino_t		*inop)
 {
 	struct xfs_mount		*mp = tp->t_mountp;
-	struct xfs_agi			*agi = XFS_BUF_TO_AGI(agbp);
+	struct xfs_agi			*agi = agbp->b_addr;
 	xfs_agnumber_t			agno = be32_to_cpu(agi->agi_seqno);
 	xfs_agnumber_t			pagno = XFS_INO_TO_AGNO(mp, parent);
 	xfs_agino_t			pagino = XFS_INO_TO_AGINO(mp, parent);
@@ -1903,7 +1943,7 @@ xfs_difree_inobt(
 	struct xfs_icluster		*xic,
 	struct xfs_inobt_rec_incore	*orec)
 {
-	struct xfs_agi			*agi = XFS_BUF_TO_AGI(agbp);
+	struct xfs_agi			*agi = agbp->b_addr;
 	xfs_agnumber_t			agno = be32_to_cpu(agi->agi_seqno);
 	struct xfs_perag		*pag;
 	struct xfs_btree_cur		*cur;
@@ -1933,14 +1973,20 @@ xfs_difree_inobt(
 			__func__, error);
 		goto error0;
 	}
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	error = xfs_inobt_get_rec(cur, &rec, &i);
 	if (error) {
 		xfs_warn(mp, "%s: xfs_inobt_get_rec() returned error %d.",
 			__func__, error);
 		goto error0;
 	}
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error0);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error0;
+	}
 	/*
 	 * Get the offset in the inode chunk.
 	 */
@@ -2033,7 +2079,7 @@ xfs_difree_finobt(
 	xfs_agino_t			agino,
 	struct xfs_inobt_rec_incore	*ibtrec) /* inobt record */
 {
-	struct xfs_agi			*agi = XFS_BUF_TO_AGI(agbp);
+	struct xfs_agi			*agi = agbp->b_addr;
 	xfs_agnumber_t			agno = be32_to_cpu(agi->agi_seqno);
 	struct xfs_btree_cur		*cur;
 	struct xfs_inobt_rec_incore	rec;
@@ -2052,7 +2098,10 @@ xfs_difree_finobt(
 		 * freed an inode in a previously fully allocated chunk. If not,
 		 * something is out of sync.
 		 */
-		XFS_WANT_CORRUPTED_GOTO(mp, ibtrec->ir_freecount == 1, error);
+		if (XFS_IS_CORRUPT(mp, ibtrec->ir_freecount != 1)) {
+			error = -EFSCORRUPTED;
+			goto error;
+		}
 
 		error = xfs_inobt_insert_rec(cur, ibtrec->ir_holemask,
 					     ibtrec->ir_count,
@@ -2075,14 +2124,20 @@ xfs_difree_finobt(
 	error = xfs_inobt_get_rec(cur, &rec, &i);
 	if (error)
 		goto error;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, error);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto error;
+	}
 
 	rec.ir_free |= XFS_INOBT_MASK(offset);
 	rec.ir_freecount++;
 
-	XFS_WANT_CORRUPTED_GOTO(mp, (rec.ir_free == ibtrec->ir_free) &&
-				(rec.ir_freecount == ibtrec->ir_freecount),
-				error);
+	if (XFS_IS_CORRUPT(mp,
+			   rec.ir_free != ibtrec->ir_free ||
+			   rec.ir_freecount != ibtrec->ir_freecount)) {
+		error = -EFSCORRUPTED;
+		goto error;
+	}
 
 	/*
 	 * The content of inobt records should always match between the inobt
@@ -2434,9 +2489,8 @@ xfs_ialloc_log_agi(
 		sizeof(xfs_agi_t)
 	};
 #ifdef DEBUG
-	xfs_agi_t		*agi;	/* allocation group header */
+	struct xfs_agi		*agi = bp->b_addr;
 
-	agi = XFS_BUF_TO_AGI(bp);
 	ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
 #endif
 
@@ -2468,14 +2522,13 @@ xfs_agi_verify(
 	struct xfs_buf	*bp)
 {
 	struct xfs_mount *mp = bp->b_mount;
-	struct xfs_agi	*agi = XFS_BUF_TO_AGI(bp);
+	struct xfs_agi	*agi = bp->b_addr;
 	int		i;
 
 	if (xfs_sb_version_hascrc(&mp->m_sb)) {
 		if (!uuid_equal(&agi->agi_uuid, &mp->m_sb.sb_meta_uuid))
 			return __this_address;
-		if (!xfs_log_check_lsn(mp,
-				be64_to_cpu(XFS_BUF_TO_AGI(bp)->agi_lsn)))
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(agi->agi_lsn)))
 			return __this_address;
 	}
 
@@ -2538,6 +2591,7 @@ xfs_agi_write_verify(
 {
 	struct xfs_mount	*mp = bp->b_mount;
 	struct xfs_buf_log_item	*bip = bp->b_log_item;
+	struct xfs_agi		*agi = bp->b_addr;
 	xfs_failaddr_t		fa;
 
 	fa = xfs_agi_verify(bp);
@@ -2550,7 +2604,7 @@ xfs_agi_write_verify(
 		return;
 
 	if (bip)
-		XFS_BUF_TO_AGI(bp)->agi_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+		agi->agi_lsn = cpu_to_be64(bip->bli_item.li_lsn);
 	xfs_buf_update_cksum(bp, XFS_AGI_CRC_OFF);
 }
 
@@ -2606,7 +2660,7 @@ xfs_ialloc_read_agi(
 	if (error)
 		return error;
 
-	agi = XFS_BUF_TO_AGI(*bpp);
+	agi = (*bpp)->b_addr;
 	pag = xfs_perag_get(mp, agno);
 	if (!pag->pagi_init) {
 		pag->pagi_freecount = be32_to_cpu(agi->agi_freecount);
@@ -2818,7 +2872,7 @@ xfs_ialloc_setup_geometry(
 	 * cannot change the behavior.
 	 */
 	igeo->inode_cluster_size_raw = XFS_INODE_BIG_CLUSTER_SIZE;
-	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+	if (xfs_sb_version_has_v3inode(&mp->m_sb)) {
 		int	new_size = igeo->inode_cluster_size_raw;
 
 		new_size *= mp->m_sb.sb_inodesize / XFS_DINODE_MIN_SIZE;
@@ -2854,3 +2908,67 @@ xfs_ialloc_setup_geometry(
 	else
 		igeo->ialloc_align = 0;
 }
+
+/* Compute the location of the root directory inode that is laid out by mkfs. */
+xfs_ino_t
+xfs_ialloc_calc_rootino(
+	struct xfs_mount	*mp,
+	int			sunit)
+{
+	struct xfs_ino_geometry	*igeo = M_IGEO(mp);
+	xfs_agblock_t		first_bno;
+
+	/*
+	 * Pre-calculate the geometry of AG 0.  We know what it looks like
+	 * because libxfs knows how to create allocation groups now.
+	 *
+	 * first_bno is the first block in which mkfs could possibly have
+	 * allocated the root directory inode, once we factor in the metadata
+	 * that mkfs formats before it.  Namely, the four AG headers...
+	 */
+	first_bno = howmany(4 * mp->m_sb.sb_sectsize, mp->m_sb.sb_blocksize);
+
+	/* ...the two free space btree roots... */
+	first_bno += 2;
+
+	/* ...the inode btree root... */
+	first_bno += 1;
+
+	/* ...the initial AGFL... */
+	first_bno += xfs_alloc_min_freelist(mp, NULL);
+
+	/* ...the free inode btree root... */
+	if (xfs_sb_version_hasfinobt(&mp->m_sb))
+		first_bno++;
+
+	/* ...the reverse mapping btree root... */
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
+		first_bno++;
+
+	/* ...the reference count btree... */
+	if (xfs_sb_version_hasreflink(&mp->m_sb))
+		first_bno++;
+
+	/*
+	 * ...and the log, if it is allocated in the first allocation group.
+	 *
+	 * This can happen with filesystems that only have a single
+	 * allocation group, or very odd geometries created by old mkfs
+	 * versions on very small filesystems.
+	 */
+	if (mp->m_sb.sb_logstart &&
+	    XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart) == 0)
+		 first_bno += mp->m_sb.sb_logblocks;
+
+	/*
+	 * Now round first_bno up to whatever allocation alignment is given
+	 * by the filesystem or was passed in.
+	 */
+	if (xfs_sb_version_hasdalign(&mp->m_sb) && igeo->ialloc_align > 0)
+		first_bno = roundup(first_bno, sunit);
+	else if (xfs_sb_version_hasalign(&mp->m_sb) &&
+			mp->m_sb.sb_inoalignmt > 1)
+		first_bno = roundup(first_bno, mp->m_sb.sb_inoalignmt);
+
+	return XFS_AGINO_TO_INO(mp, 0, XFS_AGB_TO_AGINO(mp, first_bno));
+}
diff --git a/fs/xfs/libxfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h
index 323592d563d5..72b3468b97b1 100644
--- a/fs/xfs/libxfs/xfs_ialloc.h
+++ b/fs/xfs/libxfs/xfs_ialloc.h
@@ -152,5 +152,6 @@ int xfs_inobt_insert_rec(struct xfs_btree_cur *cur, uint16_t holemask,
 
 int xfs_ialloc_cluster_alignment(struct xfs_mount *mp);
 void xfs_ialloc_setup_geometry(struct xfs_mount *mp);
+xfs_ino_t xfs_ialloc_calc_rootino(struct xfs_mount *mp, int sunit);
 
 #endif	/* __XFS_IALLOC_H__ */
diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c
index b82992f795aa..b2c122ad8f0e 100644
--- a/fs/xfs/libxfs/xfs_ialloc_btree.c
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.c
@@ -12,6 +12,7 @@
 #include "xfs_bit.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
+#include "xfs_btree_staging.h"
 #include "xfs_ialloc.h"
 #include "xfs_ialloc_btree.h"
 #include "xfs_alloc.h"
@@ -20,7 +21,6 @@
 #include "xfs_trans.h"
 #include "xfs_rmap.h"
 
-
 STATIC int
 xfs_inobt_get_minrecs(
 	struct xfs_btree_cur	*cur,
@@ -34,7 +34,7 @@ xfs_inobt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_private.a.agbp, cur->bc_private.a.agno,
+			cur->bc_ag.agbp, cur->bc_ag.agno,
 			cur->bc_btnum);
 }
 
@@ -44,8 +44,8 @@ xfs_inobt_set_root(
 	union xfs_btree_ptr	*nptr,
 	int			inc)	/* level change */
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agi		*agi = agbp->b_addr;
 
 	agi->agi_root = nptr->s;
 	be32_add_cpu(&agi->agi_level, inc);
@@ -58,8 +58,8 @@ xfs_finobt_set_root(
 	union xfs_btree_ptr	*nptr,
 	int			inc)	/* level change */
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agi		*agi = agbp->b_addr;
 
 	agi->agi_free_root = nptr->s;
 	be32_add_cpu(&agi->agi_free_level, inc);
@@ -83,7 +83,7 @@ __xfs_inobt_alloc_block(
 	args.tp = cur->bc_tp;
 	args.mp = cur->bc_mp;
 	args.oinfo = XFS_RMAP_OINFO_INOBT;
-	args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.a.agno, sbno);
+	args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_ag.agno, sbno);
 	args.minlen = 1;
 	args.maxlen = 1;
 	args.prod = 1;
@@ -212,9 +212,9 @@ xfs_inobt_init_ptr_from_cur(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*ptr)
 {
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);
+	struct xfs_agi		*agi = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agi->agi_seqno));
+	ASSERT(cur->bc_ag.agno == be32_to_cpu(agi->agi_seqno));
 
 	ptr->s = agi->agi_root;
 }
@@ -224,9 +224,9 @@ xfs_finobt_init_ptr_from_cur(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*ptr)
 {
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);
+	struct xfs_agi		*agi = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agi->agi_seqno));
+	ASSERT(cur->bc_ag.agno == be32_to_cpu(agi->agi_seqno));
 	ptr->s = agi->agi_free_root;
 }
 
@@ -400,32 +400,27 @@ static const struct xfs_btree_ops xfs_finobt_ops = {
 };
 
 /*
- * Allocate a new inode btree cursor.
+ * Initialize a new inode btree cursor.
  */
-struct xfs_btree_cur *				/* new inode btree cursor */
-xfs_inobt_init_cursor(
+static struct xfs_btree_cur *
+xfs_inobt_init_common(
 	struct xfs_mount	*mp,		/* file system mount point */
 	struct xfs_trans	*tp,		/* transaction pointer */
-	struct xfs_buf		*agbp,		/* buffer for agi structure */
 	xfs_agnumber_t		agno,		/* allocation group number */
 	xfs_btnum_t		btnum)		/* ialloc or free ino btree */
 {
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
 	struct xfs_btree_cur	*cur;
 
 	cur = kmem_zone_zalloc(xfs_btree_cur_zone, KM_NOFS);
-
 	cur->bc_tp = tp;
 	cur->bc_mp = mp;
 	cur->bc_btnum = btnum;
 	if (btnum == XFS_BTNUM_INO) {
-		cur->bc_nlevels = be32_to_cpu(agi->agi_level);
-		cur->bc_ops = &xfs_inobt_ops;
 		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_ibt_2);
+		cur->bc_ops = &xfs_inobt_ops;
 	} else {
-		cur->bc_nlevels = be32_to_cpu(agi->agi_free_level);
-		cur->bc_ops = &xfs_finobt_ops;
 		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_fibt_2);
+		cur->bc_ops = &xfs_finobt_ops;
 	}
 
 	cur->bc_blocklog = mp->m_sb.sb_blocklog;
@@ -433,12 +428,75 @@ xfs_inobt_init_cursor(
 	if (xfs_sb_version_hascrc(&mp->m_sb))
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
-	cur->bc_private.a.agbp = agbp;
-	cur->bc_private.a.agno = agno;
+	cur->bc_ag.agno = agno;
+	return cur;
+}
+
+/* Create an inode btree cursor. */
+struct xfs_btree_cur *
+xfs_inobt_init_cursor(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp,
+	xfs_agnumber_t		agno,
+	xfs_btnum_t		btnum)
+{
+	struct xfs_btree_cur	*cur;
+	struct xfs_agi		*agi = agbp->b_addr;
 
+	cur = xfs_inobt_init_common(mp, tp, agno, btnum);
+	if (btnum == XFS_BTNUM_INO)
+		cur->bc_nlevels = be32_to_cpu(agi->agi_level);
+	else
+		cur->bc_nlevels = be32_to_cpu(agi->agi_free_level);
+	cur->bc_ag.agbp = agbp;
 	return cur;
 }
 
+/* Create an inode btree cursor with a fake root for staging. */
+struct xfs_btree_cur *
+xfs_inobt_stage_cursor(
+	struct xfs_mount	*mp,
+	struct xbtree_afakeroot	*afake,
+	xfs_agnumber_t		agno,
+	xfs_btnum_t		btnum)
+{
+	struct xfs_btree_cur	*cur;
+
+	cur = xfs_inobt_init_common(mp, NULL, agno, btnum);
+	xfs_btree_stage_afakeroot(cur, afake);
+	return cur;
+}
+
+/*
+ * Install a new inobt btree root.  Caller is responsible for invalidating
+ * and freeing the old btree blocks.
+ */
+void
+xfs_inobt_commit_staged_btree(
+	struct xfs_btree_cur	*cur,
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp)
+{
+	struct xfs_agi		*agi = agbp->b_addr;
+	struct xbtree_afakeroot	*afake = cur->bc_ag.afake;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	if (cur->bc_btnum == XFS_BTNUM_INO) {
+		agi->agi_root = cpu_to_be32(afake->af_root);
+		agi->agi_level = cpu_to_be32(afake->af_levels);
+		xfs_ialloc_log_agi(tp, agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL);
+		xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_inobt_ops);
+	} else {
+		agi->agi_free_root = cpu_to_be32(afake->af_root);
+		agi->agi_free_level = cpu_to_be32(afake->af_levels);
+		xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREE_ROOT |
+					     XFS_AGI_FREE_LEVEL);
+		xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_finobt_ops);
+	}
+}
+
 /*
  * Calculate number of records in an inobt btree block.
  */
diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.h b/fs/xfs/libxfs/xfs_ialloc_btree.h
index 951305ecaae1..35bbd978c272 100644
--- a/fs/xfs/libxfs/xfs_ialloc_btree.h
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.h
@@ -48,6 +48,9 @@ struct xfs_mount;
 extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *,
 		struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t,
 		xfs_btnum_t);
+struct xfs_btree_cur *xfs_inobt_stage_cursor(struct xfs_mount *mp,
+		struct xbtree_afakeroot *afake, xfs_agnumber_t agno,
+		xfs_btnum_t btnum);
 extern int xfs_inobt_maxrecs(struct xfs_mount *, int, int);
 
 /* ir_holemask to inode allocation bitmap conversion */
@@ -68,4 +71,7 @@ int xfs_inobt_cur(struct xfs_mount *mp, struct xfs_trans *tp,
 		xfs_agnumber_t agno, xfs_btnum_t btnum,
 		struct xfs_btree_cur **curpp, struct xfs_buf **agi_bpp);
 
+void xfs_inobt_commit_staged_btree(struct xfs_btree_cur *cur,
+		struct xfs_trans *tp, struct xfs_buf *agbp);
+
 #endif	/* __XFS_IALLOC_BTREE_H__ */
diff --git a/fs/xfs/libxfs/xfs_iext_tree.c b/fs/xfs/libxfs/xfs_iext_tree.c
index 7bc87408f1a0..52451809c478 100644
--- a/fs/xfs/libxfs/xfs_iext_tree.c
+++ b/fs/xfs/libxfs/xfs_iext_tree.c
@@ -596,7 +596,7 @@ xfs_iext_realloc_root(
 	struct xfs_ifork	*ifp,
 	struct xfs_iext_cursor	*cur)
 {
-	size_t new_size = ifp->if_bytes + sizeof(struct xfs_iext_rec);
+	int64_t new_size = ifp->if_bytes + sizeof(struct xfs_iext_rec);
 	void *new;
 
 	/* account for the prev/next pointers */
diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
index 28ab3c5255e1..39c5a6e24915 100644
--- a/fs/xfs/libxfs/xfs_inode_buf.c
+++ b/fs/xfs/libxfs/xfs_inode_buf.c
@@ -44,17 +44,6 @@ xfs_inobp_check(
 }
 #endif
 
-bool
-xfs_dinode_good_version(
-	struct xfs_mount *mp,
-	__u8		version)
-{
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		return version == 3;
-
-	return version == 1 || version == 2;
-}
-
 /*
  * If we are doing readahead on an inode buffer, we might be in log recovery
  * reading an inode allocation buffer that hasn't yet been replayed, and hence
@@ -93,7 +82,7 @@ xfs_inode_buf_verify(
 		dip = xfs_buf_offset(bp, (i << mp->m_sb.sb_inodelog));
 		unlinked_ino = be32_to_cpu(dip->di_next_unlinked);
 		di_ok = xfs_verify_magic16(bp, dip->di_magic) &&
-			xfs_dinode_good_version(mp, dip->di_version) &&
+			xfs_dinode_good_version(&mp->m_sb, dip->di_version) &&
 			xfs_verify_agino_or_null(mp, agno, unlinked_ino);
 		if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
 						XFS_ERRTAG_ITOBP_INOTOBP))) {
@@ -205,26 +194,23 @@ xfs_inode_from_disk(
 	struct xfs_icdinode	*to = &ip->i_d;
 	struct inode		*inode = VFS_I(ip);
 
-
 	/*
 	 * Convert v1 inodes immediately to v2 inode format as this is the
 	 * minimum inode version format we support in the rest of the code.
+	 * They will also be unconditionally written back to disk as v2 inodes.
 	 */
-	to->di_version = from->di_version;
-	if (to->di_version == 1) {
+	if (unlikely(from->di_version == 1)) {
 		set_nlink(inode, be16_to_cpu(from->di_onlink));
-		to->di_projid_lo = 0;
-		to->di_projid_hi = 0;
-		to->di_version = 2;
+		to->di_projid = 0;
 	} else {
 		set_nlink(inode, be32_to_cpu(from->di_nlink));
-		to->di_projid_lo = be16_to_cpu(from->di_projid_lo);
-		to->di_projid_hi = be16_to_cpu(from->di_projid_hi);
+		to->di_projid = (prid_t)be16_to_cpu(from->di_projid_hi) << 16 |
+					be16_to_cpu(from->di_projid_lo);
 	}
 
 	to->di_format = from->di_format;
-	to->di_uid = be32_to_cpu(from->di_uid);
-	to->di_gid = be32_to_cpu(from->di_gid);
+	i_uid_write(inode, be32_to_cpu(from->di_uid));
+	i_gid_write(inode, be32_to_cpu(from->di_gid));
 	to->di_flushiter = be16_to_cpu(from->di_flushiter);
 
 	/*
@@ -253,11 +239,11 @@ xfs_inode_from_disk(
 	to->di_dmstate	= be16_to_cpu(from->di_dmstate);
 	to->di_flags	= be16_to_cpu(from->di_flags);
 
-	if (to->di_version == 3) {
+	if (xfs_sb_version_has_v3inode(&ip->i_mount->m_sb)) {
 		inode_set_iversion_queried(inode,
 					   be64_to_cpu(from->di_changecount));
-		to->di_crtime.t_sec = be32_to_cpu(from->di_crtime.t_sec);
-		to->di_crtime.t_nsec = be32_to_cpu(from->di_crtime.t_nsec);
+		to->di_crtime.tv_sec = be32_to_cpu(from->di_crtime.t_sec);
+		to->di_crtime.tv_nsec = be32_to_cpu(from->di_crtime.t_nsec);
 		to->di_flags2 = be64_to_cpu(from->di_flags2);
 		to->di_cowextsize = be32_to_cpu(from->di_cowextsize);
 	}
@@ -275,12 +261,11 @@ xfs_inode_to_disk(
 	to->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
 	to->di_onlink = 0;
 
-	to->di_version = from->di_version;
 	to->di_format = from->di_format;
-	to->di_uid = cpu_to_be32(from->di_uid);
-	to->di_gid = cpu_to_be32(from->di_gid);
-	to->di_projid_lo = cpu_to_be16(from->di_projid_lo);
-	to->di_projid_hi = cpu_to_be16(from->di_projid_hi);
+	to->di_uid = cpu_to_be32(i_uid_read(inode));
+	to->di_gid = cpu_to_be32(i_gid_read(inode));
+	to->di_projid_lo = cpu_to_be16(from->di_projid & 0xffff);
+	to->di_projid_hi = cpu_to_be16(from->di_projid >> 16);
 
 	memset(to->di_pad, 0, sizeof(to->di_pad));
 	to->di_atime.t_sec = cpu_to_be32(inode->i_atime.tv_sec);
@@ -304,10 +289,11 @@ xfs_inode_to_disk(
 	to->di_dmstate = cpu_to_be16(from->di_dmstate);
 	to->di_flags = cpu_to_be16(from->di_flags);
 
-	if (from->di_version == 3) {
+	if (xfs_sb_version_has_v3inode(&ip->i_mount->m_sb)) {
+		to->di_version = 3;
 		to->di_changecount = cpu_to_be64(inode_peek_iversion(inode));
-		to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.t_sec);
-		to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.t_nsec);
+		to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.tv_sec);
+		to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.tv_nsec);
 		to->di_flags2 = cpu_to_be64(from->di_flags2);
 		to->di_cowextsize = cpu_to_be32(from->di_cowextsize);
 		to->di_ino = cpu_to_be64(ip->i_ino);
@@ -316,6 +302,7 @@ xfs_inode_to_disk(
 		uuid_copy(&to->di_uuid, &ip->i_mount->m_sb.sb_meta_uuid);
 		to->di_flushiter = 0;
 	} else {
+		to->di_version = 2;
 		to->di_flushiter = cpu_to_be16(from->di_flushiter);
 	}
 }
@@ -429,7 +416,7 @@ xfs_dinode_verify_forkoff(
 	case XFS_DINODE_FMT_LOCAL:	/* fall through ... */
 	case XFS_DINODE_FMT_EXTENTS:    /* fall through ... */
 	case XFS_DINODE_FMT_BTREE:
-		if (dip->di_forkoff >= (XFS_LITINO(mp, dip->di_version) >> 3))
+		if (dip->di_forkoff >= (XFS_LITINO(mp) >> 3))
 			return __this_address;
 		break;
 	default:
@@ -455,7 +442,7 @@ xfs_dinode_verify(
 
 	/* Verify v3 integrity information first */
 	if (dip->di_version >= 3) {
-		if (!xfs_sb_version_hascrc(&mp->m_sb))
+		if (!xfs_sb_version_has_v3inode(&mp->m_sb))
 			return __this_address;
 		if (!xfs_verify_cksum((char *)dip, mp->m_sb.sb_inodesize,
 				      XFS_DINODE_CRC_OFF))
@@ -630,12 +617,9 @@ xfs_iread(
 
 	/* shortcut IO on inode allocation if possible */
 	if ((iget_flags & XFS_IGET_CREATE) &&
-	    xfs_sb_version_hascrc(&mp->m_sb) &&
+	    xfs_sb_version_has_v3inode(&mp->m_sb) &&
 	    !(mp->m_flags & XFS_MOUNT_IKEEP)) {
-		/* initialise the on-disk inode core */
-		memset(&ip->i_d, 0, sizeof(ip->i_d));
 		VFS_I(ip)->i_generation = prandom_u32();
-		ip->i_d.di_version = 3;
 		return 0;
 	}
 
@@ -677,7 +661,6 @@ xfs_iread(
 		 * Partial initialisation of the in-core inode. Just the bits
 		 * that xfs_ialloc won't overwrite or relies on being correct.
 		 */
-		ip->i_d.di_version = dip->di_version;
 		VFS_I(ip)->i_generation = be32_to_cpu(dip->di_gen);
 		ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter);
 
@@ -691,7 +674,6 @@ xfs_iread(
 		VFS_I(ip)->i_mode = 0;
 	}
 
-	ASSERT(ip->i_d.di_version >= 2);
 	ip->i_delayed_blks = 0;
 
 	/*
diff --git a/fs/xfs/libxfs/xfs_inode_buf.h b/fs/xfs/libxfs/xfs_inode_buf.h
index ab0f84165317..9b373dcf9e34 100644
--- a/fs/xfs/libxfs/xfs_inode_buf.h
+++ b/fs/xfs/libxfs/xfs_inode_buf.h
@@ -16,13 +16,9 @@ struct xfs_dinode;
  * format specific structures at the appropriate time.
  */
 struct xfs_icdinode {
-	int8_t		di_version;	/* inode version */
 	int8_t		di_format;	/* format of di_c data */
 	uint16_t	di_flushiter;	/* incremented on flush */
-	uint32_t	di_uid;		/* owner's user id */
-	uint32_t	di_gid;		/* owner's group id */
-	uint16_t	di_projid_lo;	/* lower part of owner's project id */
-	uint16_t	di_projid_hi;	/* higher part of owner's project id */
+	uint32_t	di_projid;	/* owner's project id */
 	xfs_fsize_t	di_size;	/* number of bytes in file */
 	xfs_rfsblock_t	di_nblocks;	/* # of direct & btree blocks used */
 	xfs_extlen_t	di_extsize;	/* basic/minimum extent size for file */
@@ -37,7 +33,7 @@ struct xfs_icdinode {
 	uint64_t	di_flags2;	/* more random flags */
 	uint32_t	di_cowextsize;	/* basic cow extent size for file */
 
-	xfs_ictimestamp_t di_crtime;	/* time created */
+	struct timespec64 di_crtime;	/* time created */
 };
 
 /*
@@ -62,8 +58,6 @@ void	xfs_inode_from_disk(struct xfs_inode *ip, struct xfs_dinode *from);
 void	xfs_log_dinode_to_disk(struct xfs_log_dinode *from,
 			       struct xfs_dinode *to);
 
-bool	xfs_dinode_good_version(struct xfs_mount *mp, __u8 version);
-
 #if defined(DEBUG)
 void	xfs_inobp_check(struct xfs_mount *, struct xfs_buf *);
 #else
diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c
index c643beeb5a24..518c6f0ec3a6 100644
--- a/fs/xfs/libxfs/xfs_inode_fork.c
+++ b/fs/xfs/libxfs/xfs_inode_fork.c
@@ -75,11 +75,15 @@ xfs_iformat_fork(
 			error = xfs_iformat_btree(ip, dip, XFS_DATA_FORK);
 			break;
 		default:
+			xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__,
+					dip, sizeof(*dip), __this_address);
 			return -EFSCORRUPTED;
 		}
 		break;
 
 	default:
+		xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__, dip,
+				sizeof(*dip), __this_address);
 		return -EFSCORRUPTED;
 	}
 	if (error)
@@ -110,14 +114,16 @@ xfs_iformat_fork(
 		error = xfs_iformat_btree(ip, dip, XFS_ATTR_FORK);
 		break;
 	default:
+		xfs_inode_verifier_error(ip, error, __func__, dip,
+				sizeof(*dip), __this_address);
 		error = -EFSCORRUPTED;
 		break;
 	}
 	if (error) {
-		kmem_zone_free(xfs_ifork_zone, ip->i_afp);
+		kmem_cache_free(xfs_ifork_zone, ip->i_afp);
 		ip->i_afp = NULL;
 		if (ip->i_cowfp)
-			kmem_zone_free(xfs_ifork_zone, ip->i_cowfp);
+			kmem_cache_free(xfs_ifork_zone, ip->i_cowfp);
 		ip->i_cowfp = NULL;
 		xfs_idestroy_fork(ip, XFS_DATA_FORK);
 	}
@@ -129,7 +135,7 @@ xfs_init_local_fork(
 	struct xfs_inode	*ip,
 	int			whichfork,
 	const void		*data,
-	int			size)
+	int64_t			size)
 {
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
 	int			mem_size = size, real_size = 0;
@@ -177,7 +183,7 @@ xfs_iformat_local(
 	 */
 	if (unlikely(size > XFS_DFORK_SIZE(dip, ip->i_mount, whichfork))) {
 		xfs_warn(ip->i_mount,
-	"corrupt inode %Lu (bad size %d for local fork, size = %d).",
+	"corrupt inode %Lu (bad size %d for local fork, size = %zd).",
 			(unsigned long long) ip->i_ino, size,
 			XFS_DFORK_SIZE(dip, ip->i_mount, whichfork));
 		xfs_inode_verifier_error(ip, -EFSCORRUPTED,
@@ -467,11 +473,11 @@ xfs_iroot_realloc(
 void
 xfs_idata_realloc(
 	struct xfs_inode	*ip,
-	int			byte_diff,
+	int64_t			byte_diff,
 	int			whichfork)
 {
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
-	int			new_size = (int)ifp->if_bytes + byte_diff;
+	int64_t			new_size = ifp->if_bytes + byte_diff;
 
 	ASSERT(new_size >= 0);
 	ASSERT(new_size <= XFS_IFORK_SIZE(ip, whichfork));
@@ -525,10 +531,10 @@ xfs_idestroy_fork(
 	}
 
 	if (whichfork == XFS_ATTR_FORK) {
-		kmem_zone_free(xfs_ifork_zone, ip->i_afp);
+		kmem_cache_free(xfs_ifork_zone, ip->i_afp);
 		ip->i_afp = NULL;
 	} else if (whichfork == XFS_COW_FORK) {
-		kmem_zone_free(xfs_ifork_zone, ip->i_cowfp);
+		kmem_cache_free(xfs_ifork_zone, ip->i_cowfp);
 		ip->i_cowfp = NULL;
 	}
 }
@@ -552,7 +558,7 @@ xfs_iextents_copy(
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
 	struct xfs_iext_cursor	icur;
 	struct xfs_bmbt_irec	rec;
-	int			copied = 0;
+	int64_t			copied = 0;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL | XFS_ILOCK_SHARED));
 	ASSERT(ifp->if_bytes > 0);
diff --git a/fs/xfs/libxfs/xfs_inode_fork.h b/fs/xfs/libxfs/xfs_inode_fork.h
index 00c62ce170d0..668ee942be22 100644
--- a/fs/xfs/libxfs/xfs_inode_fork.h
+++ b/fs/xfs/libxfs/xfs_inode_fork.h
@@ -13,16 +13,16 @@ struct xfs_dinode;
  * File incore extent information, present for each of data & attr forks.
  */
 struct xfs_ifork {
-	int			if_bytes;	/* bytes in if_u1 */
-	unsigned int		if_seq;		/* fork mod counter */
+	int64_t			if_bytes;	/* bytes in if_u1 */
 	struct xfs_btree_block	*if_broot;	/* file's incore btree root */
-	short			if_broot_bytes;	/* bytes allocated for root */
-	unsigned char		if_flags;	/* per-fork flags */
+	unsigned int		if_seq;		/* fork mod counter */
 	int			if_height;	/* height of the extent tree */
 	union {
 		void		*if_root;	/* extent tree root */
 		char		*if_data;	/* inline file data */
 	} if_u1;
+	short			if_broot_bytes;	/* bytes allocated for root */
+	unsigned char		if_flags;	/* per-fork flags */
 };
 
 /*
@@ -46,14 +46,9 @@ struct xfs_ifork {
 			(ip)->i_afp : \
 			(ip)->i_cowfp))
 #define XFS_IFORK_DSIZE(ip) \
-	(XFS_IFORK_Q(ip) ? \
-		XFS_IFORK_BOFF(ip) : \
-		XFS_LITINO((ip)->i_mount, (ip)->i_d.di_version))
+	(XFS_IFORK_Q(ip) ? XFS_IFORK_BOFF(ip) : XFS_LITINO((ip)->i_mount))
 #define XFS_IFORK_ASIZE(ip) \
-	(XFS_IFORK_Q(ip) ? \
-		XFS_LITINO((ip)->i_mount, (ip)->i_d.di_version) - \
-			XFS_IFORK_BOFF(ip) : \
-		0)
+	(XFS_IFORK_Q(ip) ? XFS_LITINO((ip)->i_mount) - XFS_IFORK_BOFF(ip) : 0)
 #define XFS_IFORK_SIZE(ip,w) \
 	((w) == XFS_DATA_FORK ? \
 		XFS_IFORK_DSIZE(ip) : \
@@ -87,18 +82,24 @@ struct xfs_ifork {
 #define XFS_IFORK_MAXEXT(ip, w) \
 	(XFS_IFORK_SIZE(ip, w) / sizeof(xfs_bmbt_rec_t))
 
+#define xfs_ifork_has_extents(ip, w) \
+	(XFS_IFORK_FORMAT((ip), (w)) == XFS_DINODE_FMT_EXTENTS || \
+	 XFS_IFORK_FORMAT((ip), (w)) == XFS_DINODE_FMT_BTREE)
+
 struct xfs_ifork *xfs_iext_state_to_fork(struct xfs_inode *ip, int state);
 
 int		xfs_iformat_fork(struct xfs_inode *, struct xfs_dinode *);
 void		xfs_iflush_fork(struct xfs_inode *, struct xfs_dinode *,
 				struct xfs_inode_log_item *, int);
 void		xfs_idestroy_fork(struct xfs_inode *, int);
-void		xfs_idata_realloc(struct xfs_inode *, int, int);
+void		xfs_idata_realloc(struct xfs_inode *ip, int64_t byte_diff,
+				int whichfork);
 void		xfs_iroot_realloc(struct xfs_inode *, int, int);
 int		xfs_iread_extents(struct xfs_trans *, struct xfs_inode *, int);
 int		xfs_iextents_copy(struct xfs_inode *, struct xfs_bmbt_rec *,
 				  int);
-void		xfs_init_local_fork(struct xfs_inode *, int, const void *, int);
+void		xfs_init_local_fork(struct xfs_inode *ip, int whichfork,
+				const void *data, int64_t size);
 
 xfs_extnum_t	xfs_iext_count(struct xfs_ifork *ifp);
 void		xfs_iext_insert(struct xfs_inode *, struct xfs_iext_cursor *cur,
diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h
index e5f97c69b320..e3400c9c71cd 100644
--- a/fs/xfs/libxfs/xfs_log_format.h
+++ b/fs/xfs/libxfs/xfs_log_format.h
@@ -424,17 +424,15 @@ struct xfs_log_dinode {
 	/* structure must be padded to 64 bit alignment */
 };
 
-static inline uint xfs_log_dinode_size(int version)
-{
-	if (version == 3)
-		return sizeof(struct xfs_log_dinode);
-	return offsetof(struct xfs_log_dinode, di_next_unlinked);
-}
+#define xfs_log_dinode_size(mp)						\
+	(xfs_sb_version_has_v3inode(&(mp)->m_sb) ?			\
+		sizeof(struct xfs_log_dinode) :				\
+		offsetof(struct xfs_log_dinode, di_next_unlinked))
 
 /*
- * Buffer Log Format defintions
+ * Buffer Log Format definitions
  *
- * These are the physical dirty bitmap defintions for the log format structure.
+ * These are the physical dirty bitmap definitions for the log format structure.
  */
 #define	XFS_BLF_CHUNK		128
 #define	XFS_BLF_SHIFT		7
@@ -462,11 +460,20 @@ static inline uint xfs_log_dinode_size(int version)
 #define	XFS_BLF_GDQUOT_BUF	(1<<4)
 
 /*
- * This is the structure used to lay out a buf log item in the
- * log.  The data map describes which 128 byte chunks of the buffer
- * have been logged.
- */
-#define XFS_BLF_DATAMAP_SIZE	((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) / NBWORD)
+ * This is the structure used to lay out a buf log item in the log.  The data
+ * map describes which 128 byte chunks of the buffer have been logged.
+ *
+ * The placement of blf_map_size causes blf_data_map to start at an odd
+ * multiple of sizeof(unsigned int) offset within the struct.  Because the data
+ * bitmap size will always be an even number, the end of the data_map (and
+ * therefore the structure) will also be at an odd multiple of sizeof(unsigned
+ * int).  Some 64-bit compilers will insert padding at the end of the struct to
+ * ensure 64-bit alignment of blf_blkno, but 32-bit ones will not.  Therefore,
+ * XFS_BLF_DATAMAP_SIZE must be an odd number to make the padding explicit and
+ * keep the structure size consistent between 32-bit and 64-bit platforms.
+ */
+#define __XFS_BLF_DATAMAP_SIZE	((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) / NBWORD)
+#define XFS_BLF_DATAMAP_SIZE	(__XFS_BLF_DATAMAP_SIZE + 1)
 
 typedef struct xfs_buf_log_format {
 	unsigned short	blf_type;	/* buf log item type indicator */
diff --git a/fs/xfs/libxfs/xfs_log_recover.h b/fs/xfs/libxfs/xfs_log_recover.h
index f3d18eaecebb..3bf671637a91 100644
--- a/fs/xfs/libxfs/xfs_log_recover.h
+++ b/fs/xfs/libxfs/xfs_log_recover.h
@@ -30,14 +30,14 @@ typedef struct xlog_recover_item {
 	xfs_log_iovec_t		*ri_buf;	/* ptr to regions buffer */
 } xlog_recover_item_t;
 
-typedef struct xlog_recover {
+struct xlog_recover {
 	struct hlist_node	r_list;
 	xlog_tid_t		r_log_tid;	/* log's transaction id */
 	xfs_trans_header_t	r_theader;	/* trans header for partial */
 	int			r_state;	/* not needed */
 	xfs_lsn_t		r_lsn;		/* xact lsn */
 	struct list_head	r_itemq;	/* q for items */
-} xlog_recover_t;
+};
 
 #define ITEM_TYPE(i)	(*(unsigned short *)(i)->ri_buf[0].i_addr)
 
diff --git a/fs/xfs/libxfs/xfs_refcount.c b/fs/xfs/libxfs/xfs_refcount.c
index 9a7fadb1361c..2076627243b0 100644
--- a/fs/xfs/libxfs/xfs_refcount.c
+++ b/fs/xfs/libxfs/xfs_refcount.c
@@ -46,7 +46,7 @@ xfs_refcount_lookup_le(
 	xfs_agblock_t		bno,
 	int			*stat)
 {
-	trace_xfs_refcount_lookup(cur->bc_mp, cur->bc_private.a.agno, bno,
+	trace_xfs_refcount_lookup(cur->bc_mp, cur->bc_ag.agno, bno,
 			XFS_LOOKUP_LE);
 	cur->bc_rec.rc.rc_startblock = bno;
 	cur->bc_rec.rc.rc_blockcount = 0;
@@ -63,7 +63,7 @@ xfs_refcount_lookup_ge(
 	xfs_agblock_t		bno,
 	int			*stat)
 {
-	trace_xfs_refcount_lookup(cur->bc_mp, cur->bc_private.a.agno, bno,
+	trace_xfs_refcount_lookup(cur->bc_mp, cur->bc_ag.agno, bno,
 			XFS_LOOKUP_GE);
 	cur->bc_rec.rc.rc_startblock = bno;
 	cur->bc_rec.rc.rc_blockcount = 0;
@@ -80,7 +80,7 @@ xfs_refcount_lookup_eq(
 	xfs_agblock_t		bno,
 	int			*stat)
 {
-	trace_xfs_refcount_lookup(cur->bc_mp, cur->bc_private.a.agno, bno,
+	trace_xfs_refcount_lookup(cur->bc_mp, cur->bc_ag.agno, bno,
 			XFS_LOOKUP_LE);
 	cur->bc_rec.rc.rc_startblock = bno;
 	cur->bc_rec.rc.rc_blockcount = 0;
@@ -108,7 +108,7 @@ xfs_refcount_get_rec(
 	int				*stat)
 {
 	struct xfs_mount		*mp = cur->bc_mp;
-	xfs_agnumber_t			agno = cur->bc_private.a.agno;
+	xfs_agnumber_t			agno = cur->bc_ag.agno;
 	union xfs_btree_rec		*rec;
 	int				error;
 	xfs_agblock_t			realstart;
@@ -119,7 +119,7 @@ xfs_refcount_get_rec(
 
 	xfs_refcount_btrec_to_irec(rec, irec);
 
-	agno = cur->bc_private.a.agno;
+	agno = cur->bc_ag.agno;
 	if (irec->rc_blockcount == 0 || irec->rc_blockcount > MAXREFCEXTLEN)
 		goto out_bad_rec;
 
@@ -144,7 +144,7 @@ xfs_refcount_get_rec(
 	if (irec->rc_refcount == 0 || irec->rc_refcount > MAXREFCOUNT)
 		goto out_bad_rec;
 
-	trace_xfs_refcount_get(cur->bc_mp, cur->bc_private.a.agno, irec);
+	trace_xfs_refcount_get(cur->bc_mp, cur->bc_ag.agno, irec);
 	return 0;
 
 out_bad_rec:
@@ -169,14 +169,14 @@ xfs_refcount_update(
 	union xfs_btree_rec	rec;
 	int			error;
 
-	trace_xfs_refcount_update(cur->bc_mp, cur->bc_private.a.agno, irec);
+	trace_xfs_refcount_update(cur->bc_mp, cur->bc_ag.agno, irec);
 	rec.refc.rc_startblock = cpu_to_be32(irec->rc_startblock);
 	rec.refc.rc_blockcount = cpu_to_be32(irec->rc_blockcount);
 	rec.refc.rc_refcount = cpu_to_be32(irec->rc_refcount);
 	error = xfs_btree_update(cur, &rec);
 	if (error)
 		trace_xfs_refcount_update_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -193,19 +193,22 @@ xfs_refcount_insert(
 {
 	int				error;
 
-	trace_xfs_refcount_insert(cur->bc_mp, cur->bc_private.a.agno, irec);
+	trace_xfs_refcount_insert(cur->bc_mp, cur->bc_ag.agno, irec);
 	cur->bc_rec.rc.rc_startblock = irec->rc_startblock;
 	cur->bc_rec.rc.rc_blockcount = irec->rc_blockcount;
 	cur->bc_rec.rc.rc_refcount = irec->rc_refcount;
 	error = xfs_btree_insert(cur, i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, *i == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, *i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 out_error:
 	if (error)
 		trace_xfs_refcount_insert_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -227,17 +230,23 @@ xfs_refcount_delete(
 	error = xfs_refcount_get_rec(cur, &irec, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
-	trace_xfs_refcount_delete(cur->bc_mp, cur->bc_private.a.agno, &irec);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
+	trace_xfs_refcount_delete(cur->bc_mp, cur->bc_ag.agno, &irec);
 	error = xfs_btree_delete(cur, i);
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, *i == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, *i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	if (error)
 		goto out_error;
 	error = xfs_refcount_lookup_ge(cur, irec.rc_startblock, &found_rec);
 out_error:
 	if (error)
 		trace_xfs_refcount_delete_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -349,12 +358,15 @@ xfs_refcount_split_extent(
 	error = xfs_refcount_get_rec(cur, &rcext, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	if (rcext.rc_startblock == agbno || xfs_refc_next(&rcext) <= agbno)
 		return 0;
 
 	*shape_changed = true;
-	trace_xfs_refcount_split_extent(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_refcount_split_extent(cur->bc_mp, cur->bc_ag.agno,
 			&rcext, agbno);
 
 	/* Establish the right extent. */
@@ -371,12 +383,15 @@ xfs_refcount_split_extent(
 	error = xfs_refcount_insert(cur, &tmp, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	return error;
 
 out_error:
 	trace_xfs_refcount_split_extent_error(cur->bc_mp,
-			cur->bc_private.a.agno, error, _RET_IP_);
+			cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -396,7 +411,7 @@ xfs_refcount_merge_center_extents(
 	int				found_rec;
 
 	trace_xfs_refcount_merge_center_extents(cur->bc_mp,
-			cur->bc_private.a.agno, left, center, right);
+			cur->bc_ag.agno, left, center, right);
 
 	/*
 	 * Make sure the center and right extents are not in the btree.
@@ -410,19 +425,27 @@ xfs_refcount_merge_center_extents(
 			&found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	error = xfs_refcount_delete(cur, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	if (center->rc_refcount > 1) {
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the left extent. */
@@ -430,7 +453,10 @@ xfs_refcount_merge_center_extents(
 			&found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	left->rc_blockcount = extlen;
 	error = xfs_refcount_update(cur, left);
@@ -442,7 +468,7 @@ xfs_refcount_merge_center_extents(
 
 out_error:
 	trace_xfs_refcount_merge_center_extents_error(cur->bc_mp,
-			cur->bc_private.a.agno, error, _RET_IP_);
+			cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -461,7 +487,7 @@ xfs_refcount_merge_left_extent(
 	int				found_rec;
 
 	trace_xfs_refcount_merge_left_extent(cur->bc_mp,
-			cur->bc_private.a.agno, left, cleft);
+			cur->bc_ag.agno, left, cleft);
 
 	/* If the extent at agbno (cleft) wasn't synthesized, remove it. */
 	if (cleft->rc_refcount > 1) {
@@ -469,14 +495,18 @@ xfs_refcount_merge_left_extent(
 				&found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the left extent. */
@@ -484,7 +514,10 @@ xfs_refcount_merge_left_extent(
 			&found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	left->rc_blockcount += cleft->rc_blockcount;
 	error = xfs_refcount_update(cur, left);
@@ -497,7 +530,7 @@ xfs_refcount_merge_left_extent(
 
 out_error:
 	trace_xfs_refcount_merge_left_extent_error(cur->bc_mp,
-			cur->bc_private.a.agno, error, _RET_IP_);
+			cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -515,7 +548,7 @@ xfs_refcount_merge_right_extent(
 	int				found_rec;
 
 	trace_xfs_refcount_merge_right_extent(cur->bc_mp,
-			cur->bc_private.a.agno, cright, right);
+			cur->bc_ag.agno, cright, right);
 
 	/*
 	 * If the extent ending at agbno+aglen (cright) wasn't synthesized,
@@ -526,14 +559,18 @@ xfs_refcount_merge_right_extent(
 			&found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* Enlarge the right extent. */
@@ -541,7 +578,10 @@ xfs_refcount_merge_right_extent(
 			&found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	right->rc_startblock -= cright->rc_blockcount;
 	right->rc_blockcount += cright->rc_blockcount;
@@ -554,7 +594,7 @@ xfs_refcount_merge_right_extent(
 
 out_error:
 	trace_xfs_refcount_merge_right_extent_error(cur->bc_mp,
-			cur->bc_private.a.agno, error, _RET_IP_);
+			cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -587,7 +627,10 @@ xfs_refcount_find_left_extents(
 	error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	if (xfs_refc_next(&tmp) != agbno)
 		return 0;
@@ -605,8 +648,10 @@ xfs_refcount_find_left_extents(
 		error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		/* if tmp starts at the end of our range, just use that */
 		if (tmp.rc_startblock == agbno)
@@ -634,13 +679,13 @@ xfs_refcount_find_left_extents(
 		cleft->rc_blockcount = aglen;
 		cleft->rc_refcount = 1;
 	}
-	trace_xfs_refcount_find_left_extent(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_refcount_find_left_extent(cur->bc_mp, cur->bc_ag.agno,
 			left, cleft, agbno);
 	return error;
 
 out_error:
 	trace_xfs_refcount_find_left_extent_error(cur->bc_mp,
-			cur->bc_private.a.agno, error, _RET_IP_);
+			cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -671,7 +716,10 @@ xfs_refcount_find_right_extents(
 	error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	if (tmp.rc_startblock != agbno + aglen)
 		return 0;
@@ -689,8 +737,10 @@ xfs_refcount_find_right_extents(
 		error = xfs_refcount_get_rec(cur, &tmp, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1,
-				out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		/* if tmp ends at the end of our range, just use that */
 		if (xfs_refc_next(&tmp) == agbno + aglen)
@@ -718,13 +768,13 @@ xfs_refcount_find_right_extents(
 		cright->rc_blockcount = aglen;
 		cright->rc_refcount = 1;
 	}
-	trace_xfs_refcount_find_right_extent(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_refcount_find_right_extent(cur->bc_mp, cur->bc_ag.agno,
 			cright, right, agbno + aglen);
 	return error;
 
 out_error:
 	trace_xfs_refcount_find_right_extent_error(cur->bc_mp,
-			cur->bc_private.a.agno, error, _RET_IP_);
+			cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -833,7 +883,7 @@ xfs_refcount_still_have_space(
 {
 	unsigned long			overhead;
 
-	overhead = cur->bc_private.a.priv.refc.shape_changes *
+	overhead = cur->bc_ag.refc.shape_changes *
 			xfs_allocfree_log_count(cur->bc_mp, 1);
 	overhead *= cur->bc_mp->m_sb.sb_blocksize;
 
@@ -841,17 +891,17 @@ xfs_refcount_still_have_space(
 	 * Only allow 2 refcount extent updates per transaction if the
 	 * refcount continue update "error" has been injected.
 	 */
-	if (cur->bc_private.a.priv.refc.nr_ops > 2 &&
+	if (cur->bc_ag.refc.nr_ops > 2 &&
 	    XFS_TEST_ERROR(false, cur->bc_mp,
 			XFS_ERRTAG_REFCOUNT_CONTINUE_UPDATE))
 		return false;
 
-	if (cur->bc_private.a.priv.refc.nr_ops == 0)
+	if (cur->bc_ag.refc.nr_ops == 0)
 		return true;
 	else if (overhead > cur->bc_tp->t_log_res)
 		return false;
 	return  cur->bc_tp->t_log_res - overhead >
-		cur->bc_private.a.priv.refc.nr_ops * XFS_REFCOUNT_ITEM_OVERHEAD;
+		cur->bc_ag.refc.nr_ops * XFS_REFCOUNT_ITEM_OVERHEAD;
 }
 
 /*
@@ -902,7 +952,7 @@ xfs_refcount_adjust_extents(
 					ext.rc_startblock - *agbno);
 			tmp.rc_refcount = 1 + adj;
 			trace_xfs_refcount_modify_extent(cur->bc_mp,
-					cur->bc_private.a.agno, &tmp);
+					cur->bc_ag.agno, &tmp);
 
 			/*
 			 * Either cover the hole (increment) or
@@ -913,12 +963,15 @@ xfs_refcount_adjust_extents(
 						&found_tmp);
 				if (error)
 					goto out_error;
-				XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-						found_tmp == 1, out_error);
-				cur->bc_private.a.priv.refc.nr_ops++;
+				if (XFS_IS_CORRUPT(cur->bc_mp,
+						   found_tmp != 1)) {
+					error = -EFSCORRUPTED;
+					goto out_error;
+				}
+				cur->bc_ag.refc.nr_ops++;
 			} else {
 				fsbno = XFS_AGB_TO_FSB(cur->bc_mp,
-						cur->bc_private.a.agno,
+						cur->bc_ag.agno,
 						tmp.rc_startblock);
 				xfs_bmap_add_free(cur->bc_tp, fsbno,
 						  tmp.rc_blockcount, oinfo);
@@ -945,23 +998,25 @@ xfs_refcount_adjust_extents(
 			goto skip;
 		ext.rc_refcount += adj;
 		trace_xfs_refcount_modify_extent(cur->bc_mp,
-				cur->bc_private.a.agno, &ext);
+				cur->bc_ag.agno, &ext);
 		if (ext.rc_refcount > 1) {
 			error = xfs_refcount_update(cur, &ext);
 			if (error)
 				goto out_error;
-			cur->bc_private.a.priv.refc.nr_ops++;
+			cur->bc_ag.refc.nr_ops++;
 		} else if (ext.rc_refcount == 1) {
 			error = xfs_refcount_delete(cur, &found_rec);
 			if (error)
 				goto out_error;
-			XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-					found_rec == 1, out_error);
-			cur->bc_private.a.priv.refc.nr_ops++;
+			if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+				error = -EFSCORRUPTED;
+				goto out_error;
+			}
+			cur->bc_ag.refc.nr_ops++;
 			goto advloop;
 		} else {
 			fsbno = XFS_AGB_TO_FSB(cur->bc_mp,
-					cur->bc_private.a.agno,
+					cur->bc_ag.agno,
 					ext.rc_startblock);
 			xfs_bmap_add_free(cur->bc_tp, fsbno, ext.rc_blockcount,
 					  oinfo);
@@ -980,7 +1035,7 @@ advloop:
 	return error;
 out_error:
 	trace_xfs_refcount_modify_extent_error(cur->bc_mp,
-			cur->bc_private.a.agno, error, _RET_IP_);
+			cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -1002,10 +1057,10 @@ xfs_refcount_adjust(
 	*new_agbno = agbno;
 	*new_aglen = aglen;
 	if (adj == XFS_REFCOUNT_ADJUST_INCREASE)
-		trace_xfs_refcount_increase(cur->bc_mp, cur->bc_private.a.agno,
+		trace_xfs_refcount_increase(cur->bc_mp, cur->bc_ag.agno,
 				agbno, aglen);
 	else
-		trace_xfs_refcount_decrease(cur->bc_mp, cur->bc_private.a.agno,
+		trace_xfs_refcount_decrease(cur->bc_mp, cur->bc_ag.agno,
 				agbno, aglen);
 
 	/*
@@ -1033,7 +1088,7 @@ xfs_refcount_adjust(
 	if (shape_changed)
 		shape_changes++;
 	if (shape_changes)
-		cur->bc_private.a.priv.refc.shape_changes++;
+		cur->bc_ag.refc.shape_changes++;
 
 	/* Now that we've taken care of the ends, adjust the middle extents */
 	error = xfs_refcount_adjust_extents(cur, new_agbno, new_aglen,
@@ -1044,7 +1099,7 @@ xfs_refcount_adjust(
 	return 0;
 
 out_error:
-	trace_xfs_refcount_adjust_error(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_refcount_adjust_error(cur->bc_mp, cur->bc_ag.agno,
 			error, _RET_IP_);
 	return error;
 }
@@ -1060,7 +1115,7 @@ xfs_refcount_finish_one_cleanup(
 
 	if (rcur == NULL)
 		return;
-	agbp = rcur->bc_private.a.agbp;
+	agbp = rcur->bc_ag.agbp;
 	xfs_btree_del_cursor(rcur, error);
 	if (error)
 		xfs_trans_brelse(tp, agbp);
@@ -1110,9 +1165,9 @@ xfs_refcount_finish_one(
 	 * the startblock, get one now.
 	 */
 	rcur = *pcur;
-	if (rcur != NULL && rcur->bc_private.a.agno != agno) {
-		nr_ops = rcur->bc_private.a.priv.refc.nr_ops;
-		shape_changes = rcur->bc_private.a.priv.refc.shape_changes;
+	if (rcur != NULL && rcur->bc_ag.agno != agno) {
+		nr_ops = rcur->bc_ag.refc.nr_ops;
+		shape_changes = rcur->bc_ag.refc.shape_changes;
 		xfs_refcount_finish_one_cleanup(tp, rcur, 0);
 		rcur = NULL;
 		*pcur = NULL;
@@ -1122,16 +1177,14 @@ xfs_refcount_finish_one(
 				XFS_ALLOC_FLAG_FREEING, &agbp);
 		if (error)
 			return error;
-		if (!agbp)
-			return -EFSCORRUPTED;
 
 		rcur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
 		if (!rcur) {
 			error = -ENOMEM;
 			goto out_cur;
 		}
-		rcur->bc_private.a.priv.refc.nr_ops = nr_ops;
-		rcur->bc_private.a.priv.refc.shape_changes = shape_changes;
+		rcur->bc_ag.refc.nr_ops = nr_ops;
+		rcur->bc_ag.refc.shape_changes = shape_changes;
 	}
 	*pcur = rcur;
 
@@ -1250,7 +1303,7 @@ xfs_refcount_find_shared(
 	int				have;
 	int				error;
 
-	trace_xfs_refcount_find_shared(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_refcount_find_shared(cur->bc_mp, cur->bc_ag.agno,
 			agbno, aglen);
 
 	/* By default, skip the whole range */
@@ -1272,7 +1325,10 @@ xfs_refcount_find_shared(
 	error = xfs_refcount_get_rec(cur, &tmp, &i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, out_error);
+	if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* If the extent ends before the start, look at the next one */
 	if (tmp.rc_startblock + tmp.rc_blockcount <= agbno) {
@@ -1284,7 +1340,10 @@ xfs_refcount_find_shared(
 		error = xfs_refcount_get_rec(cur, &tmp, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
 	/* If the extent starts after the range we want, bail out */
@@ -1312,7 +1371,10 @@ xfs_refcount_find_shared(
 		error = xfs_refcount_get_rec(cur, &tmp, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		if (tmp.rc_startblock >= agbno + aglen ||
 		    tmp.rc_startblock != *fbno + *flen)
 			break;
@@ -1321,12 +1383,12 @@ xfs_refcount_find_shared(
 
 done:
 	trace_xfs_refcount_find_shared_result(cur->bc_mp,
-			cur->bc_private.a.agno, *fbno, *flen);
+			cur->bc_ag.agno, *fbno, *flen);
 
 out_error:
 	if (error)
 		trace_xfs_refcount_find_shared_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -1413,39 +1475,52 @@ xfs_refcount_adjust_cow_extents(
 	switch (adj) {
 	case XFS_REFCOUNT_ADJUST_COW_ALLOC:
 		/* Adding a CoW reservation, there should be nothing here. */
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-				ext.rc_startblock >= agbno + aglen, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp,
+				   agbno + aglen > ext.rc_startblock)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		tmp.rc_startblock = agbno;
 		tmp.rc_blockcount = aglen;
 		tmp.rc_refcount = 1;
 		trace_xfs_refcount_modify_extent(cur->bc_mp,
-				cur->bc_private.a.agno, &tmp);
+				cur->bc_ag.agno, &tmp);
 
 		error = xfs_refcount_insert(cur, &tmp,
 				&found_tmp);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-				found_tmp == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_tmp != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		break;
 	case XFS_REFCOUNT_ADJUST_COW_FREE:
 		/* Removing a CoW reservation, there should be one extent. */
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-			ext.rc_startblock == agbno, out_error);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-			ext.rc_blockcount == aglen, out_error);
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-			ext.rc_refcount == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, ext.rc_startblock != agbno)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
+		if (XFS_IS_CORRUPT(cur->bc_mp, ext.rc_blockcount != aglen)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
+		if (XFS_IS_CORRUPT(cur->bc_mp, ext.rc_refcount != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		ext.rc_refcount = 0;
 		trace_xfs_refcount_modify_extent(cur->bc_mp,
-				cur->bc_private.a.agno, &ext);
+				cur->bc_ag.agno, &ext);
 		error = xfs_refcount_delete(cur, &found_rec);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(cur->bc_mp,
-				found_rec == 1, out_error);
+		if (XFS_IS_CORRUPT(cur->bc_mp, found_rec != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		break;
 	default:
 		ASSERT(0);
@@ -1454,7 +1529,7 @@ xfs_refcount_adjust_cow_extents(
 	return error;
 out_error:
 	trace_xfs_refcount_modify_extent_error(cur->bc_mp,
-			cur->bc_private.a.agno, error, _RET_IP_);
+			cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -1500,7 +1575,7 @@ xfs_refcount_adjust_cow(
 	return 0;
 
 out_error:
-	trace_xfs_refcount_adjust_cow_error(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_refcount_adjust_cow_error(cur->bc_mp, cur->bc_ag.agno,
 			error, _RET_IP_);
 	return error;
 }
@@ -1514,7 +1589,7 @@ __xfs_refcount_cow_alloc(
 	xfs_agblock_t		agbno,
 	xfs_extlen_t		aglen)
 {
-	trace_xfs_refcount_cow_increase(rcur->bc_mp, rcur->bc_private.a.agno,
+	trace_xfs_refcount_cow_increase(rcur->bc_mp, rcur->bc_ag.agno,
 			agbno, aglen);
 
 	/* Add refcount btree reservation */
@@ -1531,7 +1606,7 @@ __xfs_refcount_cow_free(
 	xfs_agblock_t		agbno,
 	xfs_extlen_t		aglen)
 {
-	trace_xfs_refcount_cow_decrease(rcur->bc_mp, rcur->bc_private.a.agno,
+	trace_xfs_refcount_cow_decrease(rcur->bc_mp, rcur->bc_ag.agno,
 			agbno, aglen);
 
 	/* Remove refcount btree reservation */
@@ -1584,14 +1659,15 @@ struct xfs_refcount_recovery {
 /* Stuff an extent on the recovery list. */
 STATIC int
 xfs_refcount_recover_extent(
-	struct xfs_btree_cur 		*cur,
+	struct xfs_btree_cur		*cur,
 	union xfs_btree_rec		*rec,
 	void				*priv)
 {
 	struct list_head		*debris = priv;
 	struct xfs_refcount_recovery	*rr;
 
-	if (be32_to_cpu(rec->refc.rc_refcount) != 1)
+	if (XFS_IS_CORRUPT(cur->bc_mp,
+			   be32_to_cpu(rec->refc.rc_refcount) != 1))
 		return -EFSCORRUPTED;
 
 	rr = kmem_alloc(sizeof(struct xfs_refcount_recovery), 0);
@@ -1640,10 +1716,6 @@ xfs_refcount_recover_cow_leftovers(
 	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
 	if (error)
 		goto out_trans;
-	if (!agbp) {
-		error = -ENOMEM;
-		goto out_trans;
-	}
 	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
 
 	/* Find all the leftover CoW staging extents. */
diff --git a/fs/xfs/libxfs/xfs_refcount_btree.c b/fs/xfs/libxfs/xfs_refcount_btree.c
index 38529dbacd55..7fd6044a4f78 100644
--- a/fs/xfs/libxfs/xfs_refcount_btree.c
+++ b/fs/xfs/libxfs/xfs_refcount_btree.c
@@ -12,6 +12,7 @@
 #include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_btree.h"
+#include "xfs_btree_staging.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
@@ -25,7 +26,7 @@ xfs_refcountbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_refcountbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_private.a.agbp, cur->bc_private.a.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno);
 }
 
 STATIC void
@@ -34,8 +35,8 @@ xfs_refcountbt_set_root(
 	union xfs_btree_ptr	*ptr,
 	int			inc)
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_agnumber_t		seqno = be32_to_cpu(agf->agf_seqno);
 	struct xfs_perag	*pag = xfs_perag_get(cur->bc_mp, seqno);
 
@@ -57,8 +58,8 @@ xfs_refcountbt_alloc_block(
 	union xfs_btree_ptr	*new,
 	int			*stat)
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	struct xfs_alloc_arg	args;		/* block allocation args */
 	int			error;		/* error return value */
 
@@ -66,7 +67,7 @@ xfs_refcountbt_alloc_block(
 	args.tp = cur->bc_tp;
 	args.mp = cur->bc_mp;
 	args.type = XFS_ALLOCTYPE_NEAR_BNO;
-	args.fsbno = XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_private.a.agno,
+	args.fsbno = XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_ag.agno,
 			xfs_refc_block(args.mp));
 	args.oinfo = XFS_RMAP_OINFO_REFC;
 	args.minlen = args.maxlen = args.prod = 1;
@@ -75,13 +76,13 @@ xfs_refcountbt_alloc_block(
 	error = xfs_alloc_vextent(&args);
 	if (error)
 		goto out_error;
-	trace_xfs_refcountbt_alloc_block(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_refcountbt_alloc_block(cur->bc_mp, cur->bc_ag.agno,
 			args.agbno, 1);
 	if (args.fsbno == NULLFSBLOCK) {
 		*stat = 0;
 		return 0;
 	}
-	ASSERT(args.agno == cur->bc_private.a.agno);
+	ASSERT(args.agno == cur->bc_ag.agno);
 	ASSERT(args.len == 1);
 
 	new->s = cpu_to_be32(args.agbno);
@@ -101,12 +102,12 @@ xfs_refcountbt_free_block(
 	struct xfs_buf		*bp)
 {
 	struct xfs_mount	*mp = cur->bc_mp;
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_fsblock_t		fsbno = XFS_DADDR_TO_FSB(mp, XFS_BUF_ADDR(bp));
 	int			error;
 
-	trace_xfs_refcountbt_free_block(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_refcountbt_free_block(cur->bc_mp, cur->bc_ag.agno,
 			XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno), 1);
 	be32_add_cpu(&agf->agf_refcount_blocks, -1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_REFCOUNT_BLOCKS);
@@ -169,9 +170,9 @@ xfs_refcountbt_init_ptr_from_cur(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*ptr)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp);
+	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agf->agf_seqno));
+	ASSERT(cur->bc_ag.agno == be32_to_cpu(agf->agf_seqno));
 
 	ptr->s = agf->agf_refcount_root;
 }
@@ -311,41 +312,90 @@ static const struct xfs_btree_ops xfs_refcountbt_ops = {
 };
 
 /*
- * Allocate a new refcount btree cursor.
+ * Initialize a new refcount btree cursor.
  */
-struct xfs_btree_cur *
-xfs_refcountbt_init_cursor(
+static struct xfs_btree_cur *
+xfs_refcountbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	struct xfs_buf		*agbp,
 	xfs_agnumber_t		agno)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
 	struct xfs_btree_cur	*cur;
 
 	ASSERT(agno != NULLAGNUMBER);
 	ASSERT(agno < mp->m_sb.sb_agcount);
-	cur = kmem_zone_zalloc(xfs_btree_cur_zone, KM_NOFS);
 
+	cur = kmem_zone_zalloc(xfs_btree_cur_zone, KM_NOFS);
 	cur->bc_tp = tp;
 	cur->bc_mp = mp;
 	cur->bc_btnum = XFS_BTNUM_REFC;
 	cur->bc_blocklog = mp->m_sb.sb_blocklog;
-	cur->bc_ops = &xfs_refcountbt_ops;
 	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_refcbt_2);
 
-	cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
-
-	cur->bc_private.a.agbp = agbp;
-	cur->bc_private.a.agno = agno;
+	cur->bc_ag.agno = agno;
 	cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
-	cur->bc_private.a.priv.refc.nr_ops = 0;
-	cur->bc_private.a.priv.refc.shape_changes = 0;
+	cur->bc_ag.refc.nr_ops = 0;
+	cur->bc_ag.refc.shape_changes = 0;
+	cur->bc_ops = &xfs_refcountbt_ops;
+	return cur;
+}
 
+/* Create a btree cursor. */
+struct xfs_btree_cur *
+xfs_refcountbt_init_cursor(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp,
+	xfs_agnumber_t		agno)
+{
+	struct xfs_agf		*agf = agbp->b_addr;
+	struct xfs_btree_cur	*cur;
+
+	cur = xfs_refcountbt_init_common(mp, tp, agno);
+	cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
+	cur->bc_ag.agbp = agbp;
 	return cur;
 }
 
+/* Create a btree cursor with a fake root for staging. */
+struct xfs_btree_cur *
+xfs_refcountbt_stage_cursor(
+	struct xfs_mount	*mp,
+	struct xbtree_afakeroot	*afake,
+	xfs_agnumber_t		agno)
+{
+	struct xfs_btree_cur	*cur;
+
+	cur = xfs_refcountbt_init_common(mp, NULL, agno);
+	xfs_btree_stage_afakeroot(cur, afake);
+	return cur;
+}
+
+/*
+ * Swap in the new btree root.  Once we pass this point the newly rebuilt btree
+ * is in place and we have to kill off all the old btree blocks.
+ */
+void
+xfs_refcountbt_commit_staged_btree(
+	struct xfs_btree_cur	*cur,
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp)
+{
+	struct xfs_agf		*agf = agbp->b_addr;
+	struct xbtree_afakeroot	*afake = cur->bc_ag.afake;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	agf->agf_refcount_root = cpu_to_be32(afake->af_root);
+	agf->agf_refcount_level = cpu_to_be32(afake->af_levels);
+	agf->agf_refcount_blocks = cpu_to_be32(afake->af_blocks);
+	xfs_alloc_log_agf(tp, agbp, XFS_AGF_REFCOUNT_BLOCKS |
+				    XFS_AGF_REFCOUNT_ROOT |
+				    XFS_AGF_REFCOUNT_LEVEL);
+	xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_refcountbt_ops);
+}
+
 /*
  * Calculate the number of records in a refcount btree block.
  */
@@ -420,7 +470,7 @@ xfs_refcountbt_calc_reserves(
 	if (error)
 		return error;
 
-	agf = XFS_BUF_TO_AGF(agbp);
+	agf = agbp->b_addr;
 	agblocks = be32_to_cpu(agf->agf_length);
 	tree_len = be32_to_cpu(agf->agf_refcount_blocks);
 	xfs_trans_brelse(tp, agbp);
diff --git a/fs/xfs/libxfs/xfs_refcount_btree.h b/fs/xfs/libxfs/xfs_refcount_btree.h
index ba416f71c824..69dc515db671 100644
--- a/fs/xfs/libxfs/xfs_refcount_btree.h
+++ b/fs/xfs/libxfs/xfs_refcount_btree.h
@@ -13,6 +13,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xbtree_afakeroot;
 
 /*
  * Btree block header size
@@ -46,6 +47,8 @@ struct xfs_mount;
 extern struct xfs_btree_cur *xfs_refcountbt_init_cursor(struct xfs_mount *mp,
 		struct xfs_trans *tp, struct xfs_buf *agbp,
 		xfs_agnumber_t agno);
+struct xfs_btree_cur *xfs_refcountbt_stage_cursor(struct xfs_mount *mp,
+		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
 extern int xfs_refcountbt_maxrecs(int blocklen, bool leaf);
 extern void xfs_refcountbt_compute_maxlevels(struct xfs_mount *mp);
 
@@ -58,4 +61,7 @@ extern int xfs_refcountbt_calc_reserves(struct xfs_mount *mp,
 		struct xfs_trans *tp, xfs_agnumber_t agno, xfs_extlen_t *ask,
 		xfs_extlen_t *used);
 
+void xfs_refcountbt_commit_staged_btree(struct xfs_btree_cur *cur,
+		struct xfs_trans *tp, struct xfs_buf *agbp);
+
 #endif	/* __XFS_REFCOUNT_BTREE_H__ */
diff --git a/fs/xfs/libxfs/xfs_rmap.c b/fs/xfs/libxfs/xfs_rmap.c
index 38e9414878b3..27c39268c31f 100644
--- a/fs/xfs/libxfs/xfs_rmap.c
+++ b/fs/xfs/libxfs/xfs_rmap.c
@@ -79,7 +79,7 @@ xfs_rmap_update(
 	union xfs_btree_rec	rec;
 	int			error;
 
-	trace_xfs_rmap_update(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_rmap_update(cur->bc_mp, cur->bc_ag.agno,
 			irec->rm_startblock, irec->rm_blockcount,
 			irec->rm_owner, irec->rm_offset, irec->rm_flags);
 
@@ -91,7 +91,7 @@ xfs_rmap_update(
 	error = xfs_btree_update(cur, &rec);
 	if (error)
 		trace_xfs_rmap_update_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -107,13 +107,16 @@ xfs_rmap_insert(
 	int			i;
 	int			error;
 
-	trace_xfs_rmap_insert(rcur->bc_mp, rcur->bc_private.a.agno, agbno,
+	trace_xfs_rmap_insert(rcur->bc_mp, rcur->bc_ag.agno, agbno,
 			len, owner, offset, flags);
 
 	error = xfs_rmap_lookup_eq(rcur, agbno, len, owner, offset, flags, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(rcur->bc_mp, i == 0, done);
+	if (XFS_IS_CORRUPT(rcur->bc_mp, i != 0)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	rcur->bc_rec.r.rm_startblock = agbno;
 	rcur->bc_rec.r.rm_blockcount = len;
@@ -123,11 +126,14 @@ xfs_rmap_insert(
 	error = xfs_btree_insert(rcur, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(rcur->bc_mp, i == 1, done);
+	if (XFS_IS_CORRUPT(rcur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 done:
 	if (error)
 		trace_xfs_rmap_insert_error(rcur->bc_mp,
-				rcur->bc_private.a.agno, error, _RET_IP_);
+				rcur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -143,22 +149,28 @@ xfs_rmap_delete(
 	int			i;
 	int			error;
 
-	trace_xfs_rmap_delete(rcur->bc_mp, rcur->bc_private.a.agno, agbno,
+	trace_xfs_rmap_delete(rcur->bc_mp, rcur->bc_ag.agno, agbno,
 			len, owner, offset, flags);
 
 	error = xfs_rmap_lookup_eq(rcur, agbno, len, owner, offset, flags, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(rcur->bc_mp, i == 1, done);
+	if (XFS_IS_CORRUPT(rcur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	error = xfs_btree_delete(rcur, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(rcur->bc_mp, i == 1, done);
+	if (XFS_IS_CORRUPT(rcur->bc_mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 done:
 	if (error)
 		trace_xfs_rmap_delete_error(rcur->bc_mp,
-				rcur->bc_private.a.agno, error, _RET_IP_);
+				rcur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -185,7 +197,7 @@ xfs_rmap_get_rec(
 	int			*stat)
 {
 	struct xfs_mount	*mp = cur->bc_mp;
-	xfs_agnumber_t		agno = cur->bc_private.a.agno;
+	xfs_agnumber_t		agno = cur->bc_ag.agno;
 	union xfs_btree_rec	*rec;
 	int			error;
 
@@ -248,7 +260,7 @@ xfs_rmap_find_left_neighbor_helper(
 	struct xfs_find_left_neighbor_info	*info = priv;
 
 	trace_xfs_rmap_find_left_neighbor_candidate(cur->bc_mp,
-			cur->bc_private.a.agno, rec->rm_startblock,
+			cur->bc_ag.agno, rec->rm_startblock,
 			rec->rm_blockcount, rec->rm_owner, rec->rm_offset,
 			rec->rm_flags);
 
@@ -300,7 +312,7 @@ xfs_rmap_find_left_neighbor(
 	info.stat = stat;
 
 	trace_xfs_rmap_find_left_neighbor_query(cur->bc_mp,
-			cur->bc_private.a.agno, bno, 0, owner, offset, flags);
+			cur->bc_ag.agno, bno, 0, owner, offset, flags);
 
 	error = xfs_rmap_query_range(cur, &info.high, &info.high,
 			xfs_rmap_find_left_neighbor_helper, &info);
@@ -308,7 +320,7 @@ xfs_rmap_find_left_neighbor(
 		error = 0;
 	if (*stat)
 		trace_xfs_rmap_find_left_neighbor_result(cur->bc_mp,
-				cur->bc_private.a.agno, irec->rm_startblock,
+				cur->bc_ag.agno, irec->rm_startblock,
 				irec->rm_blockcount, irec->rm_owner,
 				irec->rm_offset, irec->rm_flags);
 	return error;
@@ -324,7 +336,7 @@ xfs_rmap_lookup_le_range_helper(
 	struct xfs_find_left_neighbor_info	*info = priv;
 
 	trace_xfs_rmap_lookup_le_range_candidate(cur->bc_mp,
-			cur->bc_private.a.agno, rec->rm_startblock,
+			cur->bc_ag.agno, rec->rm_startblock,
 			rec->rm_blockcount, rec->rm_owner, rec->rm_offset,
 			rec->rm_flags);
 
@@ -373,14 +385,14 @@ xfs_rmap_lookup_le_range(
 	info.stat = stat;
 
 	trace_xfs_rmap_lookup_le_range(cur->bc_mp,
-			cur->bc_private.a.agno, bno, 0, owner, offset, flags);
+			cur->bc_ag.agno, bno, 0, owner, offset, flags);
 	error = xfs_rmap_query_range(cur, &info.high, &info.high,
 			xfs_rmap_lookup_le_range_helper, &info);
 	if (error == -ECANCELED)
 		error = 0;
 	if (*stat)
 		trace_xfs_rmap_lookup_le_range_result(cur->bc_mp,
-				cur->bc_private.a.agno, irec->rm_startblock,
+				cur->bc_ag.agno, irec->rm_startblock,
 				irec->rm_blockcount, irec->rm_owner,
 				irec->rm_offset, irec->rm_flags);
 	return error;
@@ -406,24 +418,39 @@ xfs_rmap_free_check_owner(
 		return 0;
 
 	/* Make sure the unwritten flag matches. */
-	XFS_WANT_CORRUPTED_GOTO(mp, (flags & XFS_RMAP_UNWRITTEN) ==
-			(rec->rm_flags & XFS_RMAP_UNWRITTEN), out);
+	if (XFS_IS_CORRUPT(mp,
+			   (flags & XFS_RMAP_UNWRITTEN) !=
+			   (rec->rm_flags & XFS_RMAP_UNWRITTEN))) {
+		error = -EFSCORRUPTED;
+		goto out;
+	}
 
 	/* Make sure the owner matches what we expect to find in the tree. */
-	XFS_WANT_CORRUPTED_GOTO(mp, owner == rec->rm_owner, out);
+	if (XFS_IS_CORRUPT(mp, owner != rec->rm_owner)) {
+		error = -EFSCORRUPTED;
+		goto out;
+	}
 
 	/* Check the offset, if necessary. */
 	if (XFS_RMAP_NON_INODE_OWNER(owner))
 		goto out;
 
 	if (flags & XFS_RMAP_BMBT_BLOCK) {
-		XFS_WANT_CORRUPTED_GOTO(mp, rec->rm_flags & XFS_RMAP_BMBT_BLOCK,
-				out);
+		if (XFS_IS_CORRUPT(mp,
+				   !(rec->rm_flags & XFS_RMAP_BMBT_BLOCK))) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
 	} else {
-		XFS_WANT_CORRUPTED_GOTO(mp, rec->rm_offset <= offset, out);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-				ltoff + rec->rm_blockcount >= offset + len,
-				out);
+		if (XFS_IS_CORRUPT(mp, rec->rm_offset > offset)) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
+		if (XFS_IS_CORRUPT(mp,
+				   offset + len > ltoff + rec->rm_blockcount)) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
 	}
 
 out:
@@ -471,7 +498,7 @@ xfs_rmap_unmap(
 			(flags & XFS_RMAP_BMBT_BLOCK);
 	if (unwritten)
 		flags |= XFS_RMAP_UNWRITTEN;
-	trace_xfs_rmap_unmap(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_unmap(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 
 	/*
@@ -482,14 +509,20 @@ xfs_rmap_unmap(
 	error = xfs_rmap_lookup_le(cur, bno, len, owner, offset, flags, &i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	error = xfs_rmap_get_rec(cur, &ltrec, &i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	trace_xfs_rmap_lookup_le_range_result(cur->bc_mp,
-			cur->bc_private.a.agno, ltrec.rm_startblock,
+			cur->bc_ag.agno, ltrec.rm_startblock,
 			ltrec.rm_blockcount, ltrec.rm_owner,
 			ltrec.rm_offset, ltrec.rm_flags);
 	ltoff = ltrec.rm_offset;
@@ -502,8 +535,12 @@ xfs_rmap_unmap(
 	 * be the case that the "left" extent goes all the way to EOFS.
 	 */
 	if (owner == XFS_RMAP_OWN_NULL) {
-		XFS_WANT_CORRUPTED_GOTO(mp, bno >= ltrec.rm_startblock +
-						ltrec.rm_blockcount, out_error);
+		if (XFS_IS_CORRUPT(mp,
+				   bno <
+				   ltrec.rm_startblock + ltrec.rm_blockcount)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		goto out_done;
 	}
 
@@ -526,15 +563,22 @@ xfs_rmap_unmap(
 		error = xfs_rmap_get_rec(cur, &rtrec, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		if (rtrec.rm_startblock >= bno + len)
 			goto out_done;
 	}
 
 	/* Make sure the extent we found covers the entire freeing range. */
-	XFS_WANT_CORRUPTED_GOTO(mp, ltrec.rm_startblock <= bno &&
-			ltrec.rm_startblock + ltrec.rm_blockcount >=
-			bno + len, out_error);
+	if (XFS_IS_CORRUPT(mp,
+			   ltrec.rm_startblock > bno ||
+			   ltrec.rm_startblock + ltrec.rm_blockcount <
+			   bno + len)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* Check owner information. */
 	error = xfs_rmap_free_check_owner(mp, ltoff, &ltrec, len, owner,
@@ -544,14 +588,17 @@ xfs_rmap_unmap(
 
 	if (ltrec.rm_startblock == bno && ltrec.rm_blockcount == len) {
 		/* exact match, simply remove the record from rmap tree */
-		trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
+		trace_xfs_rmap_delete(mp, cur->bc_ag.agno,
 				ltrec.rm_startblock, ltrec.rm_blockcount,
 				ltrec.rm_owner, ltrec.rm_offset,
 				ltrec.rm_flags);
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	} else if (ltrec.rm_startblock == bno) {
 		/*
 		 * overlap left hand side of extent: move the start, trim the
@@ -619,7 +666,7 @@ xfs_rmap_unmap(
 		else
 			cur->bc_rec.r.rm_offset = offset + len;
 		cur->bc_rec.r.rm_flags = flags;
-		trace_xfs_rmap_insert(mp, cur->bc_private.a.agno,
+		trace_xfs_rmap_insert(mp, cur->bc_ag.agno,
 				cur->bc_rec.r.rm_startblock,
 				cur->bc_rec.r.rm_blockcount,
 				cur->bc_rec.r.rm_owner,
@@ -631,11 +678,11 @@ xfs_rmap_unmap(
 	}
 
 out_done:
-	trace_xfs_rmap_unmap_done(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_unmap_done(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 out_error:
 	if (error)
-		trace_xfs_rmap_unmap_error(mp, cur->bc_private.a.agno,
+		trace_xfs_rmap_unmap_error(mp, cur->bc_ag.agno,
 				error, _RET_IP_);
 	return error;
 }
@@ -726,7 +773,7 @@ xfs_rmap_map(
 			(flags & XFS_RMAP_BMBT_BLOCK);
 	if (unwritten)
 		flags |= XFS_RMAP_UNWRITTEN;
-	trace_xfs_rmap_map(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_map(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 	ASSERT(!xfs_rmap_should_skip_owner_update(oinfo));
 
@@ -743,9 +790,12 @@ xfs_rmap_map(
 		error = xfs_rmap_get_rec(cur, &ltrec, &have_lt);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, have_lt == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, have_lt != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		trace_xfs_rmap_lookup_le_range_result(cur->bc_mp,
-				cur->bc_private.a.agno, ltrec.rm_startblock,
+				cur->bc_ag.agno, ltrec.rm_startblock,
 				ltrec.rm_blockcount, ltrec.rm_owner,
 				ltrec.rm_offset, ltrec.rm_flags);
 
@@ -753,9 +803,12 @@ xfs_rmap_map(
 			have_lt = 0;
 	}
 
-	XFS_WANT_CORRUPTED_GOTO(mp,
-		have_lt == 0 ||
-		ltrec.rm_startblock + ltrec.rm_blockcount <= bno, out_error);
+	if (XFS_IS_CORRUPT(mp,
+			   have_lt != 0 &&
+			   ltrec.rm_startblock + ltrec.rm_blockcount > bno)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/*
 	 * Increment the cursor to see if we have a right-adjacent record to our
@@ -769,11 +822,16 @@ xfs_rmap_map(
 		error = xfs_rmap_get_rec(cur, &gtrec, &have_gt);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, have_gt == 1, out_error);
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= gtrec.rm_startblock,
-					out_error);
+		if (XFS_IS_CORRUPT(mp, have_gt != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
+		if (XFS_IS_CORRUPT(mp, bno + len > gtrec.rm_startblock)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		trace_xfs_rmap_find_right_neighbor_result(cur->bc_mp,
-			cur->bc_private.a.agno, gtrec.rm_startblock,
+			cur->bc_ag.agno, gtrec.rm_startblock,
 			gtrec.rm_blockcount, gtrec.rm_owner,
 			gtrec.rm_offset, gtrec.rm_flags);
 		if (!xfs_rmap_is_mergeable(&gtrec, owner, flags))
@@ -812,7 +870,7 @@ xfs_rmap_map(
 			 * result: |rrrrrrrrrrrrrrrrrrrrrrrrrrrrr|
 			 */
 			ltrec.rm_blockcount += gtrec.rm_blockcount;
-			trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
+			trace_xfs_rmap_delete(mp, cur->bc_ag.agno,
 					gtrec.rm_startblock,
 					gtrec.rm_blockcount,
 					gtrec.rm_owner,
@@ -821,7 +879,10 @@ xfs_rmap_map(
 			error = xfs_btree_delete(cur, &i);
 			if (error)
 				goto out_error;
-			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+			if (XFS_IS_CORRUPT(mp, i != 1)) {
+				error = -EFSCORRUPTED;
+				goto out_error;
+			}
 		}
 
 		/* point the cursor back to the left record and update */
@@ -860,19 +921,22 @@ xfs_rmap_map(
 		cur->bc_rec.r.rm_owner = owner;
 		cur->bc_rec.r.rm_offset = offset;
 		cur->bc_rec.r.rm_flags = flags;
-		trace_xfs_rmap_insert(mp, cur->bc_private.a.agno, bno, len,
+		trace_xfs_rmap_insert(mp, cur->bc_ag.agno, bno, len,
 			owner, offset, flags);
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 	}
 
-	trace_xfs_rmap_map_done(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_map_done(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 out_error:
 	if (error)
-		trace_xfs_rmap_map_error(mp, cur->bc_private.a.agno,
+		trace_xfs_rmap_map_error(mp, cur->bc_ag.agno,
 				error, _RET_IP_);
 	return error;
 }
@@ -946,7 +1010,7 @@ xfs_rmap_convert(
 			(flags & (XFS_RMAP_ATTR_FORK | XFS_RMAP_BMBT_BLOCK))));
 	oldext = unwritten ? XFS_RMAP_UNWRITTEN : 0;
 	new_endoff = offset + len;
-	trace_xfs_rmap_convert(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_convert(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 
 	/*
@@ -957,14 +1021,20 @@ xfs_rmap_convert(
 	error = xfs_rmap_lookup_le(cur, bno, len, owner, offset, oldext, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	error = xfs_rmap_get_rec(cur, &PREV, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 	trace_xfs_rmap_lookup_le_range_result(cur->bc_mp,
-			cur->bc_private.a.agno, PREV.rm_startblock,
+			cur->bc_ag.agno, PREV.rm_startblock,
 			PREV.rm_blockcount, PREV.rm_owner,
 			PREV.rm_offset, PREV.rm_flags);
 
@@ -995,12 +1065,18 @@ xfs_rmap_convert(
 		error = xfs_rmap_get_rec(cur, &LEFT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		XFS_WANT_CORRUPTED_GOTO(mp,
-				LEFT.rm_startblock + LEFT.rm_blockcount <= bno,
-				done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		if (XFS_IS_CORRUPT(mp,
+				   LEFT.rm_startblock + LEFT.rm_blockcount >
+				   bno)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_find_left_neighbor_result(cur->bc_mp,
-				cur->bc_private.a.agno, LEFT.rm_startblock,
+				cur->bc_ag.agno, LEFT.rm_startblock,
 				LEFT.rm_blockcount, LEFT.rm_owner,
 				LEFT.rm_offset, LEFT.rm_flags);
 		if (LEFT.rm_startblock + LEFT.rm_blockcount == bno &&
@@ -1017,7 +1093,10 @@ xfs_rmap_convert(
 	error = xfs_btree_increment(cur, 0, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 	error = xfs_btree_increment(cur, 0, &i);
 	if (error)
 		goto done;
@@ -1026,11 +1105,16 @@ xfs_rmap_convert(
 		error = xfs_rmap_get_rec(cur, &RIGHT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= RIGHT.rm_startblock,
-					done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		if (XFS_IS_CORRUPT(mp, bno + len > RIGHT.rm_startblock)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_find_right_neighbor_result(cur->bc_mp,
-				cur->bc_private.a.agno, RIGHT.rm_startblock,
+				cur->bc_ag.agno, RIGHT.rm_startblock,
 				RIGHT.rm_blockcount, RIGHT.rm_owner,
 				RIGHT.rm_offset, RIGHT.rm_flags);
 		if (bno + len == RIGHT.rm_startblock &&
@@ -1048,14 +1132,17 @@ xfs_rmap_convert(
 	     RIGHT.rm_blockcount > XFS_RMAP_LEN_MAX)
 		state &= ~RMAP_RIGHT_CONTIG;
 
-	trace_xfs_rmap_convert_state(mp, cur->bc_private.a.agno, state,
+	trace_xfs_rmap_convert_state(mp, cur->bc_ag.agno, state,
 			_RET_IP_);
 
 	/* reset the cursor back to PREV */
 	error = xfs_rmap_lookup_le(cur, bno, len, owner, offset, oldext, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	/*
 	 * Switch out based on the FILLING and CONTIG state bits.
@@ -1071,31 +1158,46 @@ xfs_rmap_convert(
 		error = xfs_btree_increment(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		trace_xfs_rmap_delete(mp, cur->bc_ag.agno,
 				RIGHT.rm_startblock, RIGHT.rm_blockcount,
 				RIGHT.rm_owner, RIGHT.rm_offset,
 				RIGHT.rm_flags);
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		trace_xfs_rmap_delete(mp, cur->bc_ag.agno,
 				PREV.rm_startblock, PREV.rm_blockcount,
 				PREV.rm_owner, PREV.rm_offset,
 				PREV.rm_flags);
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW = LEFT;
 		NEW.rm_blockcount += PREV.rm_blockcount + RIGHT.rm_blockcount;
 		error = xfs_rmap_update(cur, &NEW);
@@ -1108,18 +1210,24 @@ xfs_rmap_convert(
 		 * Setting all of a previous oldext extent to newext.
 		 * The left neighbor is contiguous, the right is not.
 		 */
-		trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
+		trace_xfs_rmap_delete(mp, cur->bc_ag.agno,
 				PREV.rm_startblock, PREV.rm_blockcount,
 				PREV.rm_owner, PREV.rm_offset,
 				PREV.rm_flags);
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW = LEFT;
 		NEW.rm_blockcount += PREV.rm_blockcount;
 		error = xfs_rmap_update(cur, &NEW);
@@ -1135,19 +1243,28 @@ xfs_rmap_convert(
 		error = xfs_btree_increment(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		trace_xfs_rmap_delete(mp, cur->bc_private.a.agno,
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		trace_xfs_rmap_delete(mp, cur->bc_ag.agno,
 				RIGHT.rm_startblock, RIGHT.rm_blockcount,
 				RIGHT.rm_owner, RIGHT.rm_offset,
 				RIGHT.rm_flags);
 		error = xfs_btree_delete(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW = PREV;
 		NEW.rm_blockcount = len + RIGHT.rm_blockcount;
 		NEW.rm_flags = newext;
@@ -1209,12 +1326,15 @@ xfs_rmap_convert(
 		NEW.rm_blockcount = len;
 		NEW.rm_flags = newext;
 		cur->bc_rec.r = NEW;
-		trace_xfs_rmap_insert(mp, cur->bc_private.a.agno, bno,
+		trace_xfs_rmap_insert(mp, cur->bc_ag.agno, bno,
 				len, owner, offset, newext);
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		break;
 
 	case RMAP_RIGHT_FILLING | RMAP_RIGHT_CONTIG:
@@ -1253,19 +1373,25 @@ xfs_rmap_convert(
 				oldext, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_startblock = bno;
 		NEW.rm_owner = owner;
 		NEW.rm_offset = offset;
 		NEW.rm_blockcount = len;
 		NEW.rm_flags = newext;
 		cur->bc_rec.r = NEW;
-		trace_xfs_rmap_insert(mp, cur->bc_private.a.agno, bno,
+		trace_xfs_rmap_insert(mp, cur->bc_ag.agno, bno,
 				len, owner, offset, newext);
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		break;
 
 	case 0:
@@ -1288,14 +1414,17 @@ xfs_rmap_convert(
 		NEW = PREV;
 		NEW.rm_blockcount = offset - PREV.rm_offset;
 		cur->bc_rec.r = NEW;
-		trace_xfs_rmap_insert(mp, cur->bc_private.a.agno,
+		trace_xfs_rmap_insert(mp, cur->bc_ag.agno,
 				NEW.rm_startblock, NEW.rm_blockcount,
 				NEW.rm_owner, NEW.rm_offset,
 				NEW.rm_flags);
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		/*
 		 * Reset the cursor to the position of the new extent
 		 * we are about to insert as we can't trust it after
@@ -1305,16 +1434,22 @@ xfs_rmap_convert(
 				oldext, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
+		if (XFS_IS_CORRUPT(mp, i != 0)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		/* new middle extent - newext */
 		cur->bc_rec.r.rm_flags &= ~XFS_RMAP_UNWRITTEN;
 		cur->bc_rec.r.rm_flags |= newext;
-		trace_xfs_rmap_insert(mp, cur->bc_private.a.agno, bno, len,
+		trace_xfs_rmap_insert(mp, cur->bc_ag.agno, bno, len,
 				owner, offset, newext);
 		error = xfs_btree_insert(cur, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		break;
 
 	case RMAP_LEFT_FILLING | RMAP_LEFT_CONTIG | RMAP_RIGHT_CONTIG:
@@ -1330,12 +1465,12 @@ xfs_rmap_convert(
 		ASSERT(0);
 	}
 
-	trace_xfs_rmap_convert_done(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_convert_done(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 done:
 	if (error)
 		trace_xfs_rmap_convert_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -1371,7 +1506,7 @@ xfs_rmap_convert_shared(
 			(flags & (XFS_RMAP_ATTR_FORK | XFS_RMAP_BMBT_BLOCK))));
 	oldext = unwritten ? XFS_RMAP_UNWRITTEN : 0;
 	new_endoff = offset + len;
-	trace_xfs_rmap_convert(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_convert(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 
 	/*
@@ -1383,7 +1518,10 @@ xfs_rmap_convert_shared(
 			&PREV, &i);
 	if (error)
 		goto done;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto done;
+	}
 
 	ASSERT(PREV.rm_offset <= offset);
 	ASSERT(PREV.rm_offset + PREV.rm_blockcount >= new_endoff);
@@ -1406,9 +1544,12 @@ xfs_rmap_convert_shared(
 		goto done;
 	if (i) {
 		state |= RMAP_LEFT_VALID;
-		XFS_WANT_CORRUPTED_GOTO(mp,
-				LEFT.rm_startblock + LEFT.rm_blockcount <= bno,
-				done);
+		if (XFS_IS_CORRUPT(mp,
+				   LEFT.rm_startblock + LEFT.rm_blockcount >
+				   bno)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		if (xfs_rmap_is_mergeable(&LEFT, owner, newext))
 			state |= RMAP_LEFT_CONTIG;
 	}
@@ -1423,11 +1564,16 @@ xfs_rmap_convert_shared(
 		error = xfs_rmap_get_rec(cur, &RIGHT, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-		XFS_WANT_CORRUPTED_GOTO(mp, bno + len <= RIGHT.rm_startblock,
-				done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
+		if (XFS_IS_CORRUPT(mp, bno + len > RIGHT.rm_startblock)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		trace_xfs_rmap_find_right_neighbor_result(cur->bc_mp,
-				cur->bc_private.a.agno, RIGHT.rm_startblock,
+				cur->bc_ag.agno, RIGHT.rm_startblock,
 				RIGHT.rm_blockcount, RIGHT.rm_owner,
 				RIGHT.rm_offset, RIGHT.rm_flags);
 		if (xfs_rmap_is_mergeable(&RIGHT, owner, newext))
@@ -1443,7 +1589,7 @@ xfs_rmap_convert_shared(
 	     RIGHT.rm_blockcount > XFS_RMAP_LEN_MAX)
 		state &= ~RMAP_RIGHT_CONTIG;
 
-	trace_xfs_rmap_convert_state(mp, cur->bc_private.a.agno, state,
+	trace_xfs_rmap_convert_state(mp, cur->bc_ag.agno, state,
 			_RET_IP_);
 	/*
 	 * Switch out based on the FILLING and CONTIG state bits.
@@ -1472,7 +1618,10 @@ xfs_rmap_convert_shared(
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount += PREV.rm_blockcount + RIGHT.rm_blockcount;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
@@ -1495,7 +1644,10 @@ xfs_rmap_convert_shared(
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount += PREV.rm_blockcount;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
@@ -1518,7 +1670,10 @@ xfs_rmap_convert_shared(
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount += RIGHT.rm_blockcount;
 		NEW.rm_flags = RIGHT.rm_flags;
 		error = xfs_rmap_update(cur, &NEW);
@@ -1538,7 +1693,10 @@ xfs_rmap_convert_shared(
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_flags = newext;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
@@ -1570,7 +1728,10 @@ xfs_rmap_convert_shared(
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount += len;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
@@ -1612,7 +1773,10 @@ xfs_rmap_convert_shared(
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount = offset - NEW.rm_offset;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
@@ -1644,7 +1808,10 @@ xfs_rmap_convert_shared(
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount -= len;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
@@ -1679,7 +1846,10 @@ xfs_rmap_convert_shared(
 				NEW.rm_offset, NEW.rm_flags, &i);
 		if (error)
 			goto done;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto done;
+		}
 		NEW.rm_blockcount = offset - NEW.rm_offset;
 		error = xfs_rmap_update(cur, &NEW);
 		if (error)
@@ -1710,12 +1880,12 @@ xfs_rmap_convert_shared(
 		ASSERT(0);
 	}
 
-	trace_xfs_rmap_convert_done(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_convert_done(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 done:
 	if (error)
 		trace_xfs_rmap_convert_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -1753,7 +1923,7 @@ xfs_rmap_unmap_shared(
 	xfs_owner_info_unpack(oinfo, &owner, &offset, &flags);
 	if (unwritten)
 		flags |= XFS_RMAP_UNWRITTEN;
-	trace_xfs_rmap_unmap(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_unmap(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 
 	/*
@@ -1765,25 +1935,44 @@ xfs_rmap_unmap_shared(
 			&ltrec, &i);
 	if (error)
 		goto out_error;
-	XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+	if (XFS_IS_CORRUPT(mp, i != 1)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 	ltoff = ltrec.rm_offset;
 
 	/* Make sure the extent we found covers the entire freeing range. */
-	XFS_WANT_CORRUPTED_GOTO(mp, ltrec.rm_startblock <= bno &&
-		ltrec.rm_startblock + ltrec.rm_blockcount >=
-		bno + len, out_error);
+	if (XFS_IS_CORRUPT(mp,
+			   ltrec.rm_startblock > bno ||
+			   ltrec.rm_startblock + ltrec.rm_blockcount <
+			   bno + len)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* Make sure the owner matches what we expect to find in the tree. */
-	XFS_WANT_CORRUPTED_GOTO(mp, owner == ltrec.rm_owner, out_error);
+	if (XFS_IS_CORRUPT(mp, owner != ltrec.rm_owner)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* Make sure the unwritten flag matches. */
-	XFS_WANT_CORRUPTED_GOTO(mp, (flags & XFS_RMAP_UNWRITTEN) ==
-			(ltrec.rm_flags & XFS_RMAP_UNWRITTEN), out_error);
+	if (XFS_IS_CORRUPT(mp,
+			   (flags & XFS_RMAP_UNWRITTEN) !=
+			   (ltrec.rm_flags & XFS_RMAP_UNWRITTEN))) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	/* Check the offset. */
-	XFS_WANT_CORRUPTED_GOTO(mp, ltrec.rm_offset <= offset, out_error);
-	XFS_WANT_CORRUPTED_GOTO(mp, offset <= ltoff + ltrec.rm_blockcount,
-			out_error);
+	if (XFS_IS_CORRUPT(mp, ltrec.rm_offset > offset)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
+	if (XFS_IS_CORRUPT(mp, offset > ltoff + ltrec.rm_blockcount)) {
+		error = -EFSCORRUPTED;
+		goto out_error;
+	}
 
 	if (ltrec.rm_startblock == bno && ltrec.rm_blockcount == len) {
 		/* Exact match, simply remove the record from rmap tree. */
@@ -1836,7 +2025,10 @@ xfs_rmap_unmap_shared(
 				ltrec.rm_offset, ltrec.rm_flags, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		ltrec.rm_blockcount -= len;
 		error = xfs_rmap_update(cur, &ltrec);
 		if (error)
@@ -1862,7 +2054,10 @@ xfs_rmap_unmap_shared(
 				ltrec.rm_offset, ltrec.rm_flags, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		ltrec.rm_blockcount = bno - ltrec.rm_startblock;
 		error = xfs_rmap_update(cur, &ltrec);
 		if (error)
@@ -1877,12 +2072,12 @@ xfs_rmap_unmap_shared(
 			goto out_error;
 	}
 
-	trace_xfs_rmap_unmap_done(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_unmap_done(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 out_error:
 	if (error)
 		trace_xfs_rmap_unmap_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -1917,7 +2112,7 @@ xfs_rmap_map_shared(
 	xfs_owner_info_unpack(oinfo, &owner, &offset, &flags);
 	if (unwritten)
 		flags |= XFS_RMAP_UNWRITTEN;
-	trace_xfs_rmap_map(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_map(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 
 	/* Is there a left record that abuts our range? */
@@ -1938,9 +2133,12 @@ xfs_rmap_map_shared(
 		error = xfs_rmap_get_rec(cur, &gtrec, &have_gt);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, have_gt == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, have_gt != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 		trace_xfs_rmap_find_right_neighbor_result(cur->bc_mp,
-			cur->bc_private.a.agno, gtrec.rm_startblock,
+			cur->bc_ag.agno, gtrec.rm_startblock,
 			gtrec.rm_blockcount, gtrec.rm_owner,
 			gtrec.rm_offset, gtrec.rm_flags);
 
@@ -1987,7 +2185,10 @@ xfs_rmap_map_shared(
 				ltrec.rm_offset, ltrec.rm_flags, &i);
 		if (error)
 			goto out_error;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_error);
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_error;
+		}
 
 		error = xfs_rmap_update(cur, &ltrec);
 		if (error)
@@ -2030,12 +2231,12 @@ xfs_rmap_map_shared(
 			goto out_error;
 	}
 
-	trace_xfs_rmap_map_done(mp, cur->bc_private.a.agno, bno, len,
+	trace_xfs_rmap_map_done(mp, cur->bc_ag.agno, bno, len,
 			unwritten, oinfo);
 out_error:
 	if (error)
 		trace_xfs_rmap_map_error(cur->bc_mp,
-				cur->bc_private.a.agno, error, _RET_IP_);
+				cur->bc_ag.agno, error, _RET_IP_);
 	return error;
 }
 
@@ -2135,7 +2336,7 @@ xfs_rmap_finish_one_cleanup(
 
 	if (rcur == NULL)
 		return;
-	agbp = rcur->bc_private.a.agbp;
+	agbp = rcur->bc_ag.agbp;
 	xfs_btree_del_cursor(rcur, error);
 	if (error)
 		xfs_trans_brelse(tp, agbp);
@@ -2185,7 +2386,7 @@ xfs_rmap_finish_one(
 	 * the startblock, get one now.
 	 */
 	rcur = *pcur;
-	if (rcur != NULL && rcur->bc_private.a.agno != agno) {
+	if (rcur != NULL && rcur->bc_ag.agno != agno) {
 		xfs_rmap_finish_one_cleanup(tp, rcur, 0);
 		rcur = NULL;
 		*pcur = NULL;
@@ -2199,7 +2400,7 @@ xfs_rmap_finish_one(
 		error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
 		if (error)
 			return error;
-		if (!agbp)
+		if (XFS_IS_CORRUPT(tp->t_mountp, !agbp))
 			return -EFSCORRUPTED;
 
 		rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
@@ -2493,7 +2694,6 @@ struct xfs_rmap_key_state {
 	uint64_t			owner;
 	uint64_t			offset;
 	unsigned int			flags;
-	bool				has_rmap;
 };
 
 /* For each rmap given, figure out if it doesn't match the key we want. */
@@ -2508,7 +2708,6 @@ xfs_rmap_has_other_keys_helper(
 	if (rks->owner == rec->rm_owner && rks->offset == rec->rm_offset &&
 	    ((rks->flags & rec->rm_flags) & XFS_RMAP_KEY_FLAGS) == rks->flags)
 		return 0;
-	rks->has_rmap = true;
 	return -ECANCELED;
 }
 
@@ -2530,7 +2729,7 @@ xfs_rmap_has_other_keys(
 	int				error;
 
 	xfs_owner_info_unpack(oinfo, &rks.owner, &rks.offset, &rks.flags);
-	rks.has_rmap = false;
+	*has_rmap = false;
 
 	low.rm_startblock = bno;
 	memset(&high, 0xFF, sizeof(high));
@@ -2538,11 +2737,12 @@ xfs_rmap_has_other_keys(
 
 	error = xfs_rmap_query_range(cur, &low, &high,
 			xfs_rmap_has_other_keys_helper, &rks);
-	if (error < 0)
-		return error;
+	if (error == -ECANCELED) {
+		*has_rmap = true;
+		return 0;
+	}
 
-	*has_rmap = rks.has_rmap;
-	return 0;
+	return error;
 }
 
 const struct xfs_owner_info XFS_RMAP_OINFO_SKIP_UPDATE = {
diff --git a/fs/xfs/libxfs/xfs_rmap_btree.c b/fs/xfs/libxfs/xfs_rmap_btree.c
index fc78efa52c94..b7c05314d07c 100644
--- a/fs/xfs/libxfs/xfs_rmap_btree.c
+++ b/fs/xfs/libxfs/xfs_rmap_btree.c
@@ -14,6 +14,7 @@
 #include "xfs_trans.h"
 #include "xfs_alloc.h"
 #include "xfs_btree.h"
+#include "xfs_btree_staging.h"
 #include "xfs_rmap.h"
 #include "xfs_rmap_btree.h"
 #include "xfs_trace.h"
@@ -51,7 +52,7 @@ xfs_rmapbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_private.a.agbp, cur->bc_private.a.agno);
+			cur->bc_ag.agbp, cur->bc_ag.agno);
 }
 
 STATIC void
@@ -60,8 +61,8 @@ xfs_rmapbt_set_root(
 	union xfs_btree_ptr	*ptr,
 	int			inc)
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_agnumber_t		seqno = be32_to_cpu(agf->agf_seqno);
 	int			btnum = cur->bc_btnum;
 	struct xfs_perag	*pag = xfs_perag_get(cur->bc_mp, seqno);
@@ -83,25 +84,25 @@ xfs_rmapbt_alloc_block(
 	union xfs_btree_ptr	*new,
 	int			*stat)
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	int			error;
 	xfs_agblock_t		bno;
 
 	/* Allocate the new block from the freelist. If we can't, give up.  */
-	error = xfs_alloc_get_freelist(cur->bc_tp, cur->bc_private.a.agbp,
+	error = xfs_alloc_get_freelist(cur->bc_tp, cur->bc_ag.agbp,
 				       &bno, 1);
 	if (error)
 		return error;
 
-	trace_xfs_rmapbt_alloc_block(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_rmapbt_alloc_block(cur->bc_mp, cur->bc_ag.agno,
 			bno, 1);
 	if (bno == NULLAGBLOCK) {
 		*stat = 0;
 		return 0;
 	}
 
-	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_private.a.agno, bno, 1,
+	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_ag.agno, bno, 1,
 			false);
 
 	xfs_trans_agbtree_delta(cur->bc_tp, 1);
@@ -109,7 +110,7 @@ xfs_rmapbt_alloc_block(
 	be32_add_cpu(&agf->agf_rmap_blocks, 1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
 
-	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, cur->bc_private.a.agno);
+	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, cur->bc_ag.agno);
 
 	*stat = 1;
 	return 0;
@@ -120,13 +121,13 @@ xfs_rmapbt_free_block(
 	struct xfs_btree_cur	*cur,
 	struct xfs_buf		*bp)
 {
-	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
+	struct xfs_buf		*agbp = cur->bc_ag.agbp;
+	struct xfs_agf		*agf = agbp->b_addr;
 	xfs_agblock_t		bno;
 	int			error;
 
 	bno = xfs_daddr_to_agbno(cur->bc_mp, XFS_BUF_ADDR(bp));
-	trace_xfs_rmapbt_free_block(cur->bc_mp, cur->bc_private.a.agno,
+	trace_xfs_rmapbt_free_block(cur->bc_mp, cur->bc_ag.agno,
 			bno, 1);
 	be32_add_cpu(&agf->agf_rmap_blocks, -1);
 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
@@ -138,7 +139,7 @@ xfs_rmapbt_free_block(
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 	xfs_trans_agbtree_delta(cur->bc_tp, -1);
 
-	xfs_ag_resv_rmapbt_free(cur->bc_mp, cur->bc_private.a.agno);
+	xfs_ag_resv_rmapbt_free(cur->bc_mp, cur->bc_ag.agno);
 
 	return 0;
 }
@@ -215,9 +216,9 @@ xfs_rmapbt_init_ptr_from_cur(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*ptr)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp);
+	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
 
-	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agf->agf_seqno));
+	ASSERT(cur->bc_ag.agno == be32_to_cpu(agf->agf_seqno));
 
 	ptr->s = agf->agf_roots[cur->bc_btnum];
 }
@@ -448,17 +449,12 @@ static const struct xfs_btree_ops xfs_rmapbt_ops = {
 	.recs_inorder		= xfs_rmapbt_recs_inorder,
 };
 
-/*
- * Allocate a new allocation btree cursor.
- */
-struct xfs_btree_cur *
-xfs_rmapbt_init_cursor(
+static struct xfs_btree_cur *
+xfs_rmapbt_init_common(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	struct xfs_buf		*agbp,
 	xfs_agnumber_t		agno)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agbp);
 	struct xfs_btree_cur	*cur;
 
 	cur = kmem_zone_zalloc(xfs_btree_cur_zone, KM_NOFS);
@@ -468,17 +464,68 @@ xfs_rmapbt_init_cursor(
 	cur->bc_btnum = XFS_BTNUM_RMAP;
 	cur->bc_flags = XFS_BTREE_CRC_BLOCKS | XFS_BTREE_OVERLAPPING;
 	cur->bc_blocklog = mp->m_sb.sb_blocklog;
+	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
+	cur->bc_ag.agno = agno;
 	cur->bc_ops = &xfs_rmapbt_ops;
+
+	return cur;
+}
+
+/* Create a new reverse mapping btree cursor. */
+struct xfs_btree_cur *
+xfs_rmapbt_init_cursor(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp,
+	xfs_agnumber_t		agno)
+{
+	struct xfs_agf		*agf = agbp->b_addr;
+	struct xfs_btree_cur	*cur;
+
+	cur = xfs_rmapbt_init_common(mp, tp, agno);
 	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
-	cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_rmap_2);
+	cur->bc_ag.agbp = agbp;
+	return cur;
+}
 
-	cur->bc_private.a.agbp = agbp;
-	cur->bc_private.a.agno = agno;
+/* Create a new reverse mapping btree cursor with a fake root for staging. */
+struct xfs_btree_cur *
+xfs_rmapbt_stage_cursor(
+	struct xfs_mount	*mp,
+	struct xbtree_afakeroot	*afake,
+	xfs_agnumber_t		agno)
+{
+	struct xfs_btree_cur	*cur;
 
+	cur = xfs_rmapbt_init_common(mp, NULL, agno);
+	xfs_btree_stage_afakeroot(cur, afake);
 	return cur;
 }
 
 /*
+ * Install a new reverse mapping btree root.  Caller is responsible for
+ * invalidating and freeing the old btree blocks.
+ */
+void
+xfs_rmapbt_commit_staged_btree(
+	struct xfs_btree_cur	*cur,
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp)
+{
+	struct xfs_agf		*agf = agbp->b_addr;
+	struct xbtree_afakeroot	*afake = cur->bc_ag.afake;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	agf->agf_roots[cur->bc_btnum] = cpu_to_be32(afake->af_root);
+	agf->agf_levels[cur->bc_btnum] = cpu_to_be32(afake->af_levels);
+	agf->agf_rmap_blocks = cpu_to_be32(afake->af_blocks);
+	xfs_alloc_log_agf(tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS |
+				    XFS_AGF_RMAP_BLOCKS);
+	xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_rmapbt_ops);
+}
+
+/*
  * Calculate number of records in an rmap btree block.
  */
 int
@@ -569,7 +616,7 @@ xfs_rmapbt_calc_reserves(
 	if (error)
 		return error;
 
-	agf = XFS_BUF_TO_AGF(agbp);
+	agf = agbp->b_addr;
 	agblocks = be32_to_cpu(agf->agf_length);
 	tree_len = be32_to_cpu(agf->agf_rmap_blocks);
 	xfs_trans_brelse(tp, agbp);
diff --git a/fs/xfs/libxfs/xfs_rmap_btree.h b/fs/xfs/libxfs/xfs_rmap_btree.h
index 820d668b063d..115c3455a734 100644
--- a/fs/xfs/libxfs/xfs_rmap_btree.h
+++ b/fs/xfs/libxfs/xfs_rmap_btree.h
@@ -9,6 +9,7 @@
 struct xfs_buf;
 struct xfs_btree_cur;
 struct xfs_mount;
+struct xbtree_afakeroot;
 
 /* rmaps only exist on crc enabled filesystems */
 #define XFS_RMAP_BLOCK_LEN	XFS_BTREE_SBLOCK_CRC_LEN
@@ -43,6 +44,10 @@ struct xfs_mount;
 struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
 				struct xfs_trans *tp, struct xfs_buf *bp,
 				xfs_agnumber_t agno);
+struct xfs_btree_cur *xfs_rmapbt_stage_cursor(struct xfs_mount *mp,
+		struct xbtree_afakeroot *afake, xfs_agnumber_t agno);
+void xfs_rmapbt_commit_staged_btree(struct xfs_btree_cur *cur,
+		struct xfs_trans *tp, struct xfs_buf *agbp);
 int xfs_rmapbt_maxrecs(int blocklen, int leaf);
 extern void xfs_rmapbt_compute_maxlevels(struct xfs_mount *mp);
 
diff --git a/fs/xfs/libxfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c
index 8ea1efc97b41..f42c74cb8be5 100644
--- a/fs/xfs/libxfs/xfs_rtbitmap.c
+++ b/fs/xfs/libxfs/xfs_rtbitmap.c
@@ -15,7 +15,7 @@
 #include "xfs_bmap.h"
 #include "xfs_trans.h"
 #include "xfs_rtalloc.h"
-
+#include "xfs_error.h"
 
 /*
  * Realtime allocator bitmap functions shared with userspace.
@@ -70,7 +70,7 @@ xfs_rtbuf_get(
 	if (error)
 		return error;
 
-	if (nmap == 0 || !xfs_bmap_is_real_extent(&map))
+	if (XFS_IS_CORRUPT(mp, nmap == 0 || !xfs_bmap_is_real_extent(&map)))
 		return -EFSCORRUPTED;
 
 	ASSERT(map.br_startblock != NULLFSBLOCK);
diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c
index ac6cdca63e15..c526c5e5ab76 100644
--- a/fs/xfs/libxfs/xfs_sb.c
+++ b/fs/xfs/libxfs/xfs_sb.c
@@ -10,6 +10,7 @@
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
 #include "xfs_bit.h"
+#include "xfs_sb.h"
 #include "xfs_mount.h"
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
@@ -219,7 +220,7 @@ xfs_validate_sb_common(
 	struct xfs_buf		*bp,
 	struct xfs_sb		*sbp)
 {
-	struct xfs_dsb		*dsb = XFS_BUF_TO_SBP(bp);
+	struct xfs_dsb		*dsb = bp->b_addr;
 	uint32_t		agcount = 0;
 	uint32_t		rem;
 
@@ -327,6 +328,38 @@ xfs_validate_sb_common(
 		return -EFSCORRUPTED;
 	}
 
+	/* Validate the realtime geometry; stolen from xfs_repair */
+	if (sbp->sb_rextsize * sbp->sb_blocksize > XFS_MAX_RTEXTSIZE ||
+	    sbp->sb_rextsize * sbp->sb_blocksize < XFS_MIN_RTEXTSIZE) {
+		xfs_notice(mp,
+			"realtime extent sanity check failed");
+		return -EFSCORRUPTED;
+	}
+
+	if (sbp->sb_rblocks == 0) {
+		if (sbp->sb_rextents != 0 || sbp->sb_rbmblocks != 0 ||
+		    sbp->sb_rextslog != 0 || sbp->sb_frextents != 0) {
+			xfs_notice(mp,
+				"realtime zeroed geometry check failed");
+			return -EFSCORRUPTED;
+		}
+	} else {
+		uint64_t	rexts;
+		uint64_t	rbmblocks;
+
+		rexts = div_u64(sbp->sb_rblocks, sbp->sb_rextsize);
+		rbmblocks = howmany_64(sbp->sb_rextents,
+				       NBBY * sbp->sb_blocksize);
+
+		if (sbp->sb_rextents != rexts ||
+		    sbp->sb_rextslog != xfs_highbit32(sbp->sb_rextents) ||
+		    sbp->sb_rbmblocks != rbmblocks) {
+			xfs_notice(mp,
+				"realtime geometry sanity check failed");
+			return -EFSCORRUPTED;
+		}
+	}
+
 	if (sbp->sb_unit) {
 		if (!xfs_sb_version_hasdalign(sbp) ||
 		    sbp->sb_unit > sbp->sb_width ||
@@ -680,7 +713,7 @@ xfs_sb_read_verify(
 {
 	struct xfs_sb		sb;
 	struct xfs_mount	*mp = bp->b_mount;
-	struct xfs_dsb		*dsb = XFS_BUF_TO_SBP(bp);
+	struct xfs_dsb		*dsb = bp->b_addr;
 	int			error;
 
 	/*
@@ -706,7 +739,7 @@ xfs_sb_read_verify(
 	 * Check all the superblock fields.  Don't byteswap the xquota flags
 	 * because _verify_common checks the on-disk values.
 	 */
-	__xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp), false);
+	__xfs_sb_from_disk(&sb, dsb, false);
 	error = xfs_validate_sb_common(mp, bp, &sb);
 	if (error)
 		goto out_error;
@@ -729,7 +762,7 @@ static void
 xfs_sb_quiet_read_verify(
 	struct xfs_buf	*bp)
 {
-	struct xfs_dsb	*dsb = XFS_BUF_TO_SBP(bp);
+	struct xfs_dsb	*dsb = bp->b_addr;
 
 	if (dsb->sb_magicnum == cpu_to_be32(XFS_SB_MAGIC)) {
 		/* XFS filesystem, verify noisily! */
@@ -747,13 +780,14 @@ xfs_sb_write_verify(
 	struct xfs_sb		sb;
 	struct xfs_mount	*mp = bp->b_mount;
 	struct xfs_buf_log_item	*bip = bp->b_log_item;
+	struct xfs_dsb		*dsb = bp->b_addr;
 	int			error;
 
 	/*
 	 * Check all the superblock fields.  Don't byteswap the xquota flags
 	 * because _verify_common checks the on-disk values.
 	 */
-	__xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp), false);
+	__xfs_sb_from_disk(&sb, dsb, false);
 	error = xfs_validate_sb_common(mp, bp, &sb);
 	if (error)
 		goto out_error;
@@ -765,7 +799,7 @@ xfs_sb_write_verify(
 		return;
 
 	if (bip)
-		XFS_BUF_TO_SBP(bp)->sb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+		dsb->sb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
 	xfs_buf_update_cksum(bp, XFS_SB_CRC_OFF);
 	return;
@@ -926,7 +960,7 @@ xfs_log_sb(
 	mp->m_sb.sb_ifree = percpu_counter_sum(&mp->m_ifree);
 	mp->m_sb.sb_fdblocks = percpu_counter_sum(&mp->m_fdblocks);
 
-	xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb);
+	xfs_sb_to_disk(bp->b_addr, &mp->m_sb);
 	xfs_trans_buf_set_type(tp, bp, XFS_BLFT_SB_BUF);
 	xfs_trans_log_buf(tp, bp, 0, sizeof(struct xfs_dsb) - 1);
 }
@@ -984,9 +1018,9 @@ xfs_update_secondary_sbs(
 	for (agno = 1; agno < mp->m_sb.sb_agcount; agno++) {
 		struct xfs_buf		*bp;
 
-		bp = xfs_buf_get(mp->m_ddev_targp,
+		error = xfs_buf_get(mp->m_ddev_targp,
 				 XFS_AG_DADDR(mp, agno, XFS_SB_DADDR),
-				 XFS_FSS_TO_BB(mp, 1));
+				 XFS_FSS_TO_BB(mp, 1), &bp);
 		/*
 		 * If we get an error reading or writing alternate superblocks,
 		 * continue.  xfs_repair chooses the "best" superblock based
@@ -994,19 +1028,19 @@ xfs_update_secondary_sbs(
 		 * superblocks un-updated than updated, and xfs_repair may
 		 * pick them over the properly-updated primary.
 		 */
-		if (!bp) {
+		if (error) {
 			xfs_warn(mp,
 		"error allocating secondary superblock for ag %d",
 				agno);
 			if (!saved_error)
-				saved_error = -ENOMEM;
+				saved_error = error;
 			continue;
 		}
 
 		bp->b_ops = &xfs_sb_buf_ops;
 		xfs_buf_oneshot(bp);
 		xfs_buf_zero(bp, 0, BBTOB(bp->b_length));
-		xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb);
+		xfs_sb_to_disk(bp->b_addr, &mp->m_sb);
 		xfs_buf_delwri_queue(bp, &buffer_list);
 		xfs_buf_relse(bp);
 
@@ -1184,13 +1218,14 @@ xfs_sb_get_secondary(
 	struct xfs_buf		**bpp)
 {
 	struct xfs_buf		*bp;
+	int			error;
 
 	ASSERT(agno != 0 && agno != NULLAGNUMBER);
-	bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
+	error = xfs_trans_get_buf(tp, mp->m_ddev_targp,
 			XFS_AG_DADDR(mp, agno, XFS_SB_BLOCK(mp)),
-			XFS_FSS_TO_BB(mp, 1), 0);
-	if (!bp)
-		return -ENOMEM;
+			XFS_FSS_TO_BB(mp, 1), 0, &bp);
+	if (error)
+		return error;
 	bp->b_ops = &xfs_sb_buf_ops;
 	xfs_buf_oneshot(bp);
 	*bpp = bp;
diff --git a/fs/xfs/libxfs/xfs_trans_inode.c b/fs/xfs/libxfs/xfs_trans_inode.c
index a9ad90926b87..2b8ccb5b975d 100644
--- a/fs/xfs/libxfs/xfs_trans_inode.c
+++ b/fs/xfs/libxfs/xfs_trans_inode.c
@@ -55,7 +55,7 @@ xfs_trans_ichgtime(
 	int			flags)
 {
 	struct inode		*inode = VFS_I(ip);
-	struct timespec64 tv;
+	struct timespec64	tv;
 
 	ASSERT(tp);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
@@ -66,10 +66,8 @@ xfs_trans_ichgtime(
 		inode->i_mtime = tv;
 	if (flags & XFS_ICHGTIME_CHG)
 		inode->i_ctime = tv;
-	if (flags & XFS_ICHGTIME_CREATE) {
-		ip->i_d.di_crtime.t_sec = (int32_t)tv.tv_sec;
-		ip->i_d.di_crtime.t_nsec = (int32_t)tv.tv_nsec;
-	}
+	if (flags & XFS_ICHGTIME_CREATE)
+		ip->i_d.di_crtime = tv;
 }
 
 /*
diff --git a/fs/xfs/libxfs/xfs_trans_resv.c b/fs/xfs/libxfs/xfs_trans_resv.c
index d12bbd526e7c..d1a0848cb52e 100644
--- a/fs/xfs/libxfs/xfs_trans_resv.c
+++ b/fs/xfs/libxfs/xfs_trans_resv.c
@@ -187,7 +187,7 @@ xfs_calc_inode_chunk_res(
 			       XFS_FSB_TO_B(mp, 1));
 	if (alloc) {
 		/* icreate tx uses ordered buffers */
-		if (xfs_sb_version_hascrc(&mp->m_sb))
+		if (xfs_sb_version_has_v3inode(&mp->m_sb))
 			return res;
 		size = XFS_FSB_TO_B(mp, 1);
 	}
@@ -197,6 +197,24 @@ xfs_calc_inode_chunk_res(
 }
 
 /*
+ * Per-extent log reservation for the btree changes involved in freeing or
+ * allocating a realtime extent.  We have to be able to log as many rtbitmap
+ * blocks as needed to mark inuse MAXEXTLEN blocks' worth of realtime extents,
+ * as well as the realtime summary block.
+ */
+static unsigned int
+xfs_rtalloc_log_count(
+	struct xfs_mount	*mp,
+	unsigned int		num_ops)
+{
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+	unsigned int		rtbmp_bytes;
+
+	rtbmp_bytes = (MAXEXTLEN / mp->m_sb.sb_rextsize) / NBBY;
+	return (howmany(rtbmp_bytes, blksz) + 1) * num_ops;
+}
+
+/*
  * Various log reservation values.
  *
  * These are based on the size of the file system block because that is what
@@ -218,13 +236,21 @@ xfs_calc_inode_chunk_res(
 
 /*
  * In a write transaction we can allocate a maximum of 2
- * extents.  This gives:
+ * extents.  This gives (t1):
  *    the inode getting the new extents: inode size
  *    the inode's bmap btree: max depth * block size
  *    the agfs of the ags from which the extents are allocated: 2 * sector
  *    the superblock free block counter: sector size
  *    the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
- * And the bmap_finish transaction can free bmap blocks in a join:
+ * Or, if we're writing to a realtime file (t2):
+ *    the inode getting the new extents: inode size
+ *    the inode's bmap btree: max depth * block size
+ *    the agfs of the ags from which the extents are allocated: 2 * sector
+ *    the superblock free block counter: sector size
+ *    the realtime bitmap: ((MAXEXTLEN / rtextsize) / NBBY) bytes
+ *    the realtime summary: 1 block
+ *    the allocation btrees: 2 trees * (2 * max depth - 1) * block size
+ * And the bmap_finish transaction can free bmap blocks in a join (t3):
  *    the agfs of the ags containing the blocks: 2 * sector size
  *    the agfls of the ags containing the blocks: 2 * sector size
  *    the super block free block counter: sector size
@@ -234,40 +260,72 @@ STATIC uint
 xfs_calc_write_reservation(
 	struct xfs_mount	*mp)
 {
-	return XFS_DQUOT_LOGRES(mp) +
-		max((xfs_calc_inode_res(mp, 1) +
+	unsigned int		t1, t2, t3;
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+
+	t1 = xfs_calc_inode_res(mp, 1) +
+	     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK), blksz) +
+	     xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+
+	if (xfs_sb_version_hasrealtime(&mp->m_sb)) {
+		t2 = xfs_calc_inode_res(mp, 1) +
 		     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK),
-				      XFS_FSB_TO_B(mp, 1)) +
+				     blksz) +
 		     xfs_calc_buf_res(3, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2),
-				      XFS_FSB_TO_B(mp, 1))),
-		    (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2),
-				      XFS_FSB_TO_B(mp, 1))));
+		     xfs_calc_buf_res(xfs_rtalloc_log_count(mp, 1), blksz) +
+		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 1), blksz);
+	} else {
+		t2 = 0;
+	}
+
+	t3 = xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+
+	return XFS_DQUOT_LOGRES(mp) + max3(t1, t2, t3);
 }
 
 /*
- * In truncating a file we free up to two extents at once.  We can modify:
+ * In truncating a file we free up to two extents at once.  We can modify (t1):
  *    the inode being truncated: inode size
  *    the inode's bmap btree: (max depth + 1) * block size
- * And the bmap_finish transaction can free the blocks and bmap blocks:
+ * And the bmap_finish transaction can free the blocks and bmap blocks (t2):
  *    the agf for each of the ags: 4 * sector size
  *    the agfl for each of the ags: 4 * sector size
  *    the super block to reflect the freed blocks: sector size
  *    worst case split in allocation btrees per extent assuming 4 extents:
  *		4 exts * 2 trees * (2 * max depth - 1) * block size
+ * Or, if it's a realtime file (t3):
+ *    the agf for each of the ags: 2 * sector size
+ *    the agfl for each of the ags: 2 * sector size
+ *    the super block to reflect the freed blocks: sector size
+ *    the realtime bitmap: 2 exts * ((MAXEXTLEN / rtextsize) / NBBY) bytes
+ *    the realtime summary: 2 exts * 1 block
+ *    worst case split in allocation btrees per extent assuming 2 extents:
+ *		2 exts * 2 trees * (2 * max depth - 1) * block size
  */
 STATIC uint
 xfs_calc_itruncate_reservation(
 	struct xfs_mount	*mp)
 {
-	return XFS_DQUOT_LOGRES(mp) +
-		max((xfs_calc_inode_res(mp, 1) +
-		     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1,
-				      XFS_FSB_TO_B(mp, 1))),
-		    (xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) +
-		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 4),
-				      XFS_FSB_TO_B(mp, 1))));
+	unsigned int		t1, t2, t3;
+	unsigned int		blksz = XFS_FSB_TO_B(mp, 1);
+
+	t1 = xfs_calc_inode_res(mp, 1) +
+	     xfs_calc_buf_res(XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1, blksz);
+
+	t2 = xfs_calc_buf_res(9, mp->m_sb.sb_sectsize) +
+	     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 4), blksz);
+
+	if (xfs_sb_version_hasrealtime(&mp->m_sb)) {
+		t3 = xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
+		     xfs_calc_buf_res(xfs_rtalloc_log_count(mp, 2), blksz) +
+		     xfs_calc_buf_res(xfs_allocfree_log_count(mp, 2), blksz);
+	} else {
+		t3 = 0;
+	}
+
+	return XFS_DQUOT_LOGRES(mp) + max3(t1, t2, t3);
 }
 
 /*
@@ -718,7 +776,7 @@ xfs_calc_clear_agi_bucket_reservation(
 
 /*
  * Adjusting quota limits.
- *    the xfs_disk_dquot_t: sizeof(struct xfs_disk_dquot)
+ *    the disk quota buffer: sizeof(struct xfs_disk_dquot)
  */
 STATIC uint
 xfs_calc_qm_setqlim_reservation(void)
@@ -742,7 +800,7 @@ xfs_calc_qm_dqalloc_reservation(
 
 /*
  * Turning off quotas.
- *    the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2
+ *    the quota off logitems: sizeof(struct xfs_qoff_logitem) * 2
  *    the superblock for the quota flags: sector size
  */
 STATIC uint
@@ -755,7 +813,7 @@ xfs_calc_qm_quotaoff_reservation(
 
 /*
  * End of turning off quotas.
- *    the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2
+ *    the quota off logitems: sizeof(struct xfs_qoff_logitem) * 2
  */
 STATIC uint
 xfs_calc_qm_quotaoff_end_reservation(void)
diff --git a/fs/xfs/libxfs/xfs_types.h b/fs/xfs/libxfs/xfs_types.h
index 300b3e91ca3a..397d94775440 100644
--- a/fs/xfs/libxfs/xfs_types.h
+++ b/fs/xfs/libxfs/xfs_types.h
@@ -21,7 +21,6 @@ typedef int32_t		xfs_suminfo_t;	/* type of bitmap summary info */
 typedef uint32_t	xfs_rtword_t;	/* word type for bitmap manipulations */
 
 typedef int64_t		xfs_lsn_t;	/* log sequence number */
-typedef int32_t		xfs_tid_t;	/* transaction identifier */
 
 typedef uint32_t	xfs_dablk_t;	/* dir/attr block number (in file) */
 typedef uint32_t	xfs_dahash_t;	/* dir/attr hash value */
@@ -33,7 +32,6 @@ typedef uint64_t	xfs_fileoff_t;	/* block number in a file */
 typedef uint64_t	xfs_filblks_t;	/* number of blocks in a file */
 
 typedef int64_t		xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
-typedef int64_t		xfs_sfiloff_t;	/* signed block number in a file */
 
 /*
  * New verifiers will return the instruction address of the failing check.
diff --git a/fs/xfs/scrub/agheader.c b/fs/xfs/scrub/agheader.c
index ba0f747c82e8..e9bcf1faa183 100644
--- a/fs/xfs/scrub/agheader.c
+++ b/fs/xfs/scrub/agheader.c
@@ -92,7 +92,7 @@ xchk_superblock(
 	if (!xchk_process_error(sc, agno, XFS_SB_BLOCK(mp), &error))
 		return error;
 
-	sb = XFS_BUF_TO_SBP(bp);
+	sb = bp->b_addr;
 
 	/*
 	 * Verify the geometries match.  Fields that are permanently
@@ -358,7 +358,7 @@ static inline void
 xchk_agf_xref_freeblks(
 	struct xfs_scrub	*sc)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(sc->sa.agf_bp);
+	struct xfs_agf		*agf = sc->sa.agf_bp->b_addr;
 	xfs_extlen_t		blocks = 0;
 	int			error;
 
@@ -378,7 +378,7 @@ static inline void
 xchk_agf_xref_cntbt(
 	struct xfs_scrub	*sc)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(sc->sa.agf_bp);
+	struct xfs_agf		*agf = sc->sa.agf_bp->b_addr;
 	xfs_agblock_t		agbno;
 	xfs_extlen_t		blocks;
 	int			have;
@@ -410,7 +410,7 @@ STATIC void
 xchk_agf_xref_btreeblks(
 	struct xfs_scrub	*sc)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(sc->sa.agf_bp);
+	struct xfs_agf		*agf = sc->sa.agf_bp->b_addr;
 	struct xfs_mount	*mp = sc->mp;
 	xfs_agblock_t		blocks;
 	xfs_agblock_t		btreeblks;
@@ -456,7 +456,7 @@ static inline void
 xchk_agf_xref_refcblks(
 	struct xfs_scrub	*sc)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(sc->sa.agf_bp);
+	struct xfs_agf		*agf = sc->sa.agf_bp->b_addr;
 	xfs_agblock_t		blocks;
 	int			error;
 
@@ -525,7 +525,7 @@ xchk_agf(
 		goto out;
 	xchk_buffer_recheck(sc, sc->sa.agf_bp);
 
-	agf = XFS_BUF_TO_AGF(sc->sa.agf_bp);
+	agf = sc->sa.agf_bp->b_addr;
 
 	/* Check the AG length */
 	eoag = be32_to_cpu(agf->agf_length);
@@ -711,7 +711,7 @@ xchk_agfl(
 		goto out;
 
 	/* Allocate buffer to ensure uniqueness of AGFL entries. */
-	agf = XFS_BUF_TO_AGF(sc->sa.agf_bp);
+	agf = sc->sa.agf_bp->b_addr;
 	agflcount = be32_to_cpu(agf->agf_flcount);
 	if (agflcount > xfs_agfl_size(sc->mp)) {
 		xchk_block_set_corrupt(sc, sc->sa.agf_bp);
@@ -728,7 +728,7 @@ xchk_agfl(
 	}
 
 	/* Check the blocks in the AGFL. */
-	error = xfs_agfl_walk(sc->mp, XFS_BUF_TO_AGF(sc->sa.agf_bp),
+	error = xfs_agfl_walk(sc->mp, sc->sa.agf_bp->b_addr,
 			sc->sa.agfl_bp, xchk_agfl_block, &sai);
 	if (error == -ECANCELED) {
 		error = 0;
@@ -765,7 +765,7 @@ static inline void
 xchk_agi_xref_icounts(
 	struct xfs_scrub	*sc)
 {
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(sc->sa.agi_bp);
+	struct xfs_agi		*agi = sc->sa.agi_bp->b_addr;
 	xfs_agino_t		icount;
 	xfs_agino_t		freecount;
 	int			error;
@@ -834,7 +834,7 @@ xchk_agi(
 		goto out;
 	xchk_buffer_recheck(sc, sc->sa.agi_bp);
 
-	agi = XFS_BUF_TO_AGI(sc->sa.agi_bp);
+	agi = sc->sa.agi_bp->b_addr;
 
 	/* Check the AG length */
 	eoag = be32_to_cpu(agi->agi_length);
diff --git a/fs/xfs/scrub/agheader_repair.c b/fs/xfs/scrub/agheader_repair.c
index 7a1a38b636a9..bca2ab1d4be9 100644
--- a/fs/xfs/scrub/agheader_repair.c
+++ b/fs/xfs/scrub/agheader_repair.c
@@ -49,7 +49,7 @@ xrep_superblock(
 
 	/* Copy AG 0's superblock to this one. */
 	xfs_buf_zero(bp, 0, BBTOB(bp->b_length));
-	xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb);
+	xfs_sb_to_disk(bp->b_addr, &mp->m_sb);
 
 	/* Write this to disk. */
 	xfs_trans_buf_set_type(sc->tp, bp, XFS_BLFT_SB_BUF);
@@ -140,7 +140,7 @@ xrep_agf_find_btrees(
 	struct xrep_find_ag_btree	*fab,
 	struct xfs_buf			*agfl_bp)
 {
-	struct xfs_agf			*old_agf = XFS_BUF_TO_AGF(agf_bp);
+	struct xfs_agf			*old_agf = agf_bp->b_addr;
 	int				error;
 
 	/* Go find the root data. */
@@ -181,7 +181,7 @@ xrep_agf_init_header(
 	struct xfs_agf		*old_agf)
 {
 	struct xfs_mount	*mp = sc->mp;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agf_bp);
+	struct xfs_agf		*agf = agf_bp->b_addr;
 
 	memcpy(old_agf, agf, sizeof(*old_agf));
 	memset(agf, 0, BBTOB(agf_bp->b_length));
@@ -238,7 +238,7 @@ xrep_agf_calc_from_btrees(
 {
 	struct xrep_agf_allocbt	raa = { .sc = sc };
 	struct xfs_btree_cur	*cur = NULL;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agf_bp);
+	struct xfs_agf		*agf = agf_bp->b_addr;
 	struct xfs_mount	*mp = sc->mp;
 	xfs_agblock_t		btreeblks;
 	xfs_agblock_t		blocks;
@@ -302,7 +302,7 @@ xrep_agf_commit_new(
 	struct xfs_buf		*agf_bp)
 {
 	struct xfs_perag	*pag;
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agf_bp);
+	struct xfs_agf		*agf = agf_bp->b_addr;
 
 	/* Trigger fdblocks recalculation */
 	xfs_force_summary_recalc(sc->mp);
@@ -376,7 +376,7 @@ xrep_agf(
 	if (error)
 		return error;
 	agf_bp->b_ops = &xfs_agf_buf_ops;
-	agf = XFS_BUF_TO_AGF(agf_bp);
+	agf = agf_bp->b_addr;
 
 	/*
 	 * Load the AGFL so that we can screen out OWN_AG blocks that are on
@@ -395,7 +395,7 @@ xrep_agf(
 	 * Spot-check the AGFL blocks; if they're obviously corrupt then
 	 * there's nothing we can do but bail out.
 	 */
-	error = xfs_agfl_walk(sc->mp, XFS_BUF_TO_AGF(agf_bp), agfl_bp,
+	error = xfs_agfl_walk(sc->mp, agf_bp->b_addr, agfl_bp,
 			xrep_agf_check_agfl_block, sc);
 	if (error)
 		return error;
@@ -429,10 +429,10 @@ out_revert:
 
 struct xrep_agfl {
 	/* Bitmap of other OWN_AG metadata blocks. */
-	struct xfs_bitmap	agmetablocks;
+	struct xbitmap		agmetablocks;
 
 	/* Bitmap of free space. */
-	struct xfs_bitmap	*freesp;
+	struct xbitmap		*freesp;
 
 	struct xfs_scrub	*sc;
 };
@@ -453,14 +453,14 @@ xrep_agfl_walk_rmap(
 
 	/* Record all the OWN_AG blocks. */
 	if (rec->rm_owner == XFS_RMAP_OWN_AG) {
-		fsb = XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_private.a.agno,
+		fsb = XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_ag.agno,
 				rec->rm_startblock);
-		error = xfs_bitmap_set(ra->freesp, fsb, rec->rm_blockcount);
+		error = xbitmap_set(ra->freesp, fsb, rec->rm_blockcount);
 		if (error)
 			return error;
 	}
 
-	return xfs_bitmap_set_btcur_path(&ra->agmetablocks, cur);
+	return xbitmap_set_btcur_path(&ra->agmetablocks, cur);
 }
 
 /*
@@ -476,19 +476,17 @@ STATIC int
 xrep_agfl_collect_blocks(
 	struct xfs_scrub	*sc,
 	struct xfs_buf		*agf_bp,
-	struct xfs_bitmap	*agfl_extents,
+	struct xbitmap		*agfl_extents,
 	xfs_agblock_t		*flcount)
 {
 	struct xrep_agfl	ra;
 	struct xfs_mount	*mp = sc->mp;
 	struct xfs_btree_cur	*cur;
-	struct xfs_bitmap_range	*br;
-	struct xfs_bitmap_range	*n;
 	int			error;
 
 	ra.sc = sc;
 	ra.freesp = agfl_extents;
-	xfs_bitmap_init(&ra.agmetablocks);
+	xbitmap_init(&ra.agmetablocks);
 
 	/* Find all space used by the free space btrees & rmapbt. */
 	cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno);
@@ -500,7 +498,7 @@ xrep_agfl_collect_blocks(
 	/* Find all blocks currently being used by the bnobt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
 			XFS_BTNUM_BNO);
-	error = xfs_bitmap_set_btblocks(&ra.agmetablocks, cur);
+	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
 	xfs_btree_del_cursor(cur, error);
@@ -508,7 +506,7 @@ xrep_agfl_collect_blocks(
 	/* Find all blocks currently being used by the cntbt. */
 	cur = xfs_allocbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno,
 			XFS_BTNUM_CNT);
-	error = xfs_bitmap_set_btblocks(&ra.agmetablocks, cur);
+	error = xbitmap_set_btblocks(&ra.agmetablocks, cur);
 	if (error)
 		goto err;
 
@@ -518,8 +516,8 @@ xrep_agfl_collect_blocks(
 	 * Drop the freesp meta blocks that are in use by btrees.
 	 * The remaining blocks /should/ be AGFL blocks.
 	 */
-	error = xfs_bitmap_disunion(agfl_extents, &ra.agmetablocks);
-	xfs_bitmap_destroy(&ra.agmetablocks);
+	error = xbitmap_disunion(agfl_extents, &ra.agmetablocks);
+	xbitmap_destroy(&ra.agmetablocks);
 	if (error)
 		return error;
 
@@ -527,18 +525,12 @@ xrep_agfl_collect_blocks(
 	 * Calculate the new AGFL size.  If we found more blocks than fit in
 	 * the AGFL we'll free them later.
 	 */
-	*flcount = 0;
-	for_each_xfs_bitmap_extent(br, n, agfl_extents) {
-		*flcount += br->len;
-		if (*flcount > xfs_agfl_size(mp))
-			break;
-	}
-	if (*flcount > xfs_agfl_size(mp))
-		*flcount = xfs_agfl_size(mp);
+	*flcount = min_t(uint64_t, xbitmap_hweight(agfl_extents),
+			 xfs_agfl_size(mp));
 	return 0;
 
 err:
-	xfs_bitmap_destroy(&ra.agmetablocks);
+	xbitmap_destroy(&ra.agmetablocks);
 	xfs_btree_del_cursor(cur, error);
 	return error;
 }
@@ -550,7 +542,7 @@ xrep_agfl_update_agf(
 	struct xfs_buf		*agf_bp,
 	xfs_agblock_t		flcount)
 {
-	struct xfs_agf		*agf = XFS_BUF_TO_AGF(agf_bp);
+	struct xfs_agf		*agf = agf_bp->b_addr;
 
 	ASSERT(flcount <= xfs_agfl_size(sc->mp));
 
@@ -573,13 +565,13 @@ STATIC void
 xrep_agfl_init_header(
 	struct xfs_scrub	*sc,
 	struct xfs_buf		*agfl_bp,
-	struct xfs_bitmap	*agfl_extents,
+	struct xbitmap		*agfl_extents,
 	xfs_agblock_t		flcount)
 {
 	struct xfs_mount	*mp = sc->mp;
 	__be32			*agfl_bno;
-	struct xfs_bitmap_range	*br;
-	struct xfs_bitmap_range	*n;
+	struct xbitmap_range	*br;
+	struct xbitmap_range	*n;
 	struct xfs_agfl		*agfl;
 	xfs_agblock_t		agbno;
 	unsigned int		fl_off;
@@ -602,8 +594,8 @@ xrep_agfl_init_header(
 	 * step.
 	 */
 	fl_off = 0;
-	agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, agfl_bp);
-	for_each_xfs_bitmap_extent(br, n, agfl_extents) {
+	agfl_bno = xfs_buf_to_agfl_bno(agfl_bp);
+	for_each_xbitmap_extent(br, n, agfl_extents) {
 		agbno = XFS_FSB_TO_AGBNO(mp, br->start);
 
 		trace_xrep_agfl_insert(mp, sc->sa.agno, agbno, br->len);
@@ -637,7 +629,7 @@ int
 xrep_agfl(
 	struct xfs_scrub	*sc)
 {
-	struct xfs_bitmap	agfl_extents;
+	struct xbitmap		agfl_extents;
 	struct xfs_mount	*mp = sc->mp;
 	struct xfs_buf		*agf_bp;
 	struct xfs_buf		*agfl_bp;
@@ -649,7 +641,7 @@ xrep_agfl(
 		return -EOPNOTSUPP;
 
 	xchk_perag_get(sc->mp, &sc->sa);
-	xfs_bitmap_init(&agfl_extents);
+	xbitmap_init(&agfl_extents);
 
 	/*
 	 * Read the AGF so that we can query the rmapbt.  We hope that there's
@@ -659,8 +651,6 @@ xrep_agfl(
 	error = xfs_alloc_read_agf(mp, sc->tp, sc->sa.agno, 0, &agf_bp);
 	if (error)
 		return error;
-	if (!agf_bp)
-		return -ENOMEM;
 
 	/*
 	 * Make sure we have the AGFL buffer, as scrub might have decided it
@@ -698,10 +688,10 @@ xrep_agfl(
 		goto err;
 
 	/* Dump any AGFL overflow. */
-	return xrep_reap_extents(sc, &agfl_extents, &XFS_RMAP_OINFO_AG,
+	error = xrep_reap_extents(sc, &agfl_extents, &XFS_RMAP_OINFO_AG,
 			XFS_AG_RESV_AGFL);
 err:
-	xfs_bitmap_destroy(&agfl_extents);
+	xbitmap_destroy(&agfl_extents);
 	return error;
 }
 
@@ -735,8 +725,6 @@ xrep_agi_find_btrees(
 	error = xfs_alloc_read_agf(mp, sc->tp, sc->sa.agno, 0, &agf_bp);
 	if (error)
 		return error;
-	if (!agf_bp)
-		return -ENOMEM;
 
 	/* Find the btree roots. */
 	error = xrep_find_ag_btree_roots(sc, agf_bp, fab, NULL);
@@ -765,7 +753,7 @@ xrep_agi_init_header(
 	struct xfs_buf		*agi_bp,
 	struct xfs_agi		*old_agi)
 {
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agi_bp);
+	struct xfs_agi		*agi = agi_bp->b_addr;
 	struct xfs_mount	*mp = sc->mp;
 
 	memcpy(old_agi, agi, sizeof(*old_agi));
@@ -811,7 +799,7 @@ xrep_agi_calc_from_btrees(
 	struct xfs_buf		*agi_bp)
 {
 	struct xfs_btree_cur	*cur;
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agi_bp);
+	struct xfs_agi		*agi = agi_bp->b_addr;
 	struct xfs_mount	*mp = sc->mp;
 	xfs_agino_t		count;
 	xfs_agino_t		freecount;
@@ -839,7 +827,7 @@ xrep_agi_commit_new(
 	struct xfs_buf		*agi_bp)
 {
 	struct xfs_perag	*pag;
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agi_bp);
+	struct xfs_agi		*agi = agi_bp->b_addr;
 
 	/* Trigger inode count recalculation */
 	xfs_force_summary_recalc(sc->mp);
@@ -896,7 +884,7 @@ xrep_agi(
 	if (error)
 		return error;
 	agi_bp->b_ops = &xfs_agi_buf_ops;
-	agi = XFS_BUF_TO_AGI(agi_bp);
+	agi = agi_bp->b_addr;
 
 	/* Find the AGI btree roots. */
 	error = xrep_agi_find_btrees(sc, fab);
diff --git a/fs/xfs/scrub/alloc.c b/fs/xfs/scrub/alloc.c
index 5533e48e605d..73d924e47565 100644
--- a/fs/xfs/scrub/alloc.c
+++ b/fs/xfs/scrub/alloc.c
@@ -94,7 +94,7 @@ xchk_allocbt_rec(
 	union xfs_btree_rec	*rec)
 {
 	struct xfs_mount	*mp = bs->cur->bc_mp;
-	xfs_agnumber_t		agno = bs->cur->bc_private.a.agno;
+	xfs_agnumber_t		agno = bs->cur->bc_ag.agno;
 	xfs_agblock_t		bno;
 	xfs_extlen_t		len;
 
diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c
index 0edc7f8eb96e..9faddb334a2c 100644
--- a/fs/xfs/scrub/attr.c
+++ b/fs/xfs/scrub/attr.c
@@ -98,7 +98,7 @@ struct xchk_xattr {
 /*
  * Check that an extended attribute key can be looked up by hash.
  *
- * We use the XFS attribute list iterator (i.e. xfs_attr_list_int_ilocked)
+ * We use the XFS attribute list iterator (i.e. xfs_attr_list_ilocked)
  * to call this function for every attribute key in an inode.  Once
  * we're here, we load the attribute value to see if any errors happen,
  * or if we get more or less data than we expected.
@@ -147,11 +147,8 @@ xchk_xattr_listent(
 		return;
 	}
 
-	args.flags = ATTR_KERNOTIME;
-	if (flags & XFS_ATTR_ROOT)
-		args.flags |= ATTR_ROOT;
-	else if (flags & XFS_ATTR_SECURE)
-		args.flags |= ATTR_SECURE;
+	args.op_flags = XFS_DA_OP_NOTIME;
+	args.attr_filter = flags & XFS_ATTR_NSP_ONDISK_MASK;
 	args.geo = context->dp->i_mount->m_attr_geo;
 	args.whichfork = XFS_ATTR_FORK;
 	args.dp = context->dp;
@@ -162,7 +159,10 @@ xchk_xattr_listent(
 	args.value = xchk_xattr_valuebuf(sx->sc);
 	args.valuelen = valuelen;
 
-	error = xfs_attr_get_ilocked(context->dp, &args);
+	error = xfs_attr_get_ilocked(&args);
+	/* ENODATA means the hash lookup failed and the attr is bad */
+	if (error == -ENODATA)
+		error = -EFSCORRUPTED;
 	if (!xchk_fblock_process_error(sx->sc, XFS_ATTR_FORK, args.blkno,
 			&error))
 		goto fail_xref;
@@ -398,15 +398,14 @@ out:
 STATIC int
 xchk_xattr_rec(
 	struct xchk_da_btree		*ds,
-	int				level,
-	void				*rec)
+	int				level)
 {
 	struct xfs_mount		*mp = ds->state->mp;
-	struct xfs_attr_leaf_entry	*ent = rec;
-	struct xfs_da_state_blk		*blk;
+	struct xfs_da_state_blk		*blk = &ds->state->path.blk[level];
 	struct xfs_attr_leaf_name_local	*lentry;
 	struct xfs_attr_leaf_name_remote	*rentry;
 	struct xfs_buf			*bp;
+	struct xfs_attr_leaf_entry	*ent;
 	xfs_dahash_t			calc_hash;
 	xfs_dahash_t			hash;
 	int				nameidx;
@@ -414,7 +413,9 @@ xchk_xattr_rec(
 	unsigned int			badflags;
 	int				error;
 
-	blk = &ds->state->path.blk[level];
+	ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
+
+	ent = xfs_attr3_leaf_entryp(blk->bp->b_addr) + blk->index;
 
 	/* Check the whole block, if necessary. */
 	error = xchk_xattr_block(ds, level);
@@ -473,7 +474,6 @@ xchk_xattr(
 	struct xfs_scrub		*sc)
 {
 	struct xchk_xattr		sx;
-	struct attrlist_cursor_kern	cursor = { 0 };
 	xfs_dablk_t			last_checked = -1U;
 	int				error = 0;
 
@@ -492,11 +492,10 @@ xchk_xattr(
 
 	/* Check that every attr key can also be looked up by hash. */
 	sx.context.dp = sc->ip;
-	sx.context.cursor = &cursor;
 	sx.context.resynch = 1;
 	sx.context.put_listent = xchk_xattr_listent;
 	sx.context.tp = sc->tp;
-	sx.context.flags = ATTR_INCOMPLETE;
+	sx.context.allow_incomplete = true;
 	sx.sc = sc;
 
 	/*
@@ -515,7 +514,7 @@ xchk_xattr(
 	 * iteration, which doesn't really follow the usual buffer
 	 * locking order.
 	 */
-	error = xfs_attr_list_int_ilocked(&sx.context);
+	error = xfs_attr_list_ilocked(&sx.context);
 	if (!xchk_fblock_process_error(sc, XFS_ATTR_FORK, 0, &error))
 		goto out;
 
diff --git a/fs/xfs/scrub/bitmap.c b/fs/xfs/scrub/bitmap.c
index 3d47d111be5a..f88694f22d05 100644
--- a/fs/xfs/scrub/bitmap.c
+++ b/fs/xfs/scrub/bitmap.c
@@ -18,14 +18,14 @@
  * This is the logical equivalent of bitmap |= mask(start, len).
  */
 int
-xfs_bitmap_set(
-	struct xfs_bitmap	*bitmap,
+xbitmap_set(
+	struct xbitmap		*bitmap,
 	uint64_t		start,
 	uint64_t		len)
 {
-	struct xfs_bitmap_range	*bmr;
+	struct xbitmap_range	*bmr;
 
-	bmr = kmem_alloc(sizeof(struct xfs_bitmap_range), KM_MAYFAIL);
+	bmr = kmem_alloc(sizeof(struct xbitmap_range), KM_MAYFAIL);
 	if (!bmr)
 		return -ENOMEM;
 
@@ -39,13 +39,13 @@ xfs_bitmap_set(
 
 /* Free everything related to this bitmap. */
 void
-xfs_bitmap_destroy(
-	struct xfs_bitmap	*bitmap)
+xbitmap_destroy(
+	struct xbitmap		*bitmap)
 {
-	struct xfs_bitmap_range	*bmr;
-	struct xfs_bitmap_range	*n;
+	struct xbitmap_range	*bmr;
+	struct xbitmap_range	*n;
 
-	for_each_xfs_bitmap_extent(bmr, n, bitmap) {
+	for_each_xbitmap_extent(bmr, n, bitmap) {
 		list_del(&bmr->list);
 		kmem_free(bmr);
 	}
@@ -53,24 +53,24 @@ xfs_bitmap_destroy(
 
 /* Set up a per-AG block bitmap. */
 void
-xfs_bitmap_init(
-	struct xfs_bitmap	*bitmap)
+xbitmap_init(
+	struct xbitmap		*bitmap)
 {
 	INIT_LIST_HEAD(&bitmap->list);
 }
 
 /* Compare two btree extents. */
 static int
-xfs_bitmap_range_cmp(
+xbitmap_range_cmp(
 	void			*priv,
 	struct list_head	*a,
 	struct list_head	*b)
 {
-	struct xfs_bitmap_range	*ap;
-	struct xfs_bitmap_range	*bp;
+	struct xbitmap_range	*ap;
+	struct xbitmap_range	*bp;
 
-	ap = container_of(a, struct xfs_bitmap_range, list);
-	bp = container_of(b, struct xfs_bitmap_range, list);
+	ap = container_of(a, struct xbitmap_range, list);
+	bp = container_of(b, struct xbitmap_range, list);
 
 	if (ap->start > bp->start)
 		return 1;
@@ -96,14 +96,14 @@ xfs_bitmap_range_cmp(
 #define LEFT_ALIGNED	(1 << 0)
 #define RIGHT_ALIGNED	(1 << 1)
 int
-xfs_bitmap_disunion(
-	struct xfs_bitmap	*bitmap,
-	struct xfs_bitmap	*sub)
+xbitmap_disunion(
+	struct xbitmap		*bitmap,
+	struct xbitmap		*sub)
 {
 	struct list_head	*lp;
-	struct xfs_bitmap_range	*br;
-	struct xfs_bitmap_range	*new_br;
-	struct xfs_bitmap_range	*sub_br;
+	struct xbitmap_range	*br;
+	struct xbitmap_range	*new_br;
+	struct xbitmap_range	*sub_br;
 	uint64_t		sub_start;
 	uint64_t		sub_len;
 	int			state;
@@ -113,8 +113,8 @@ xfs_bitmap_disunion(
 		return 0;
 	ASSERT(!list_empty(&sub->list));
 
-	list_sort(NULL, &bitmap->list, xfs_bitmap_range_cmp);
-	list_sort(NULL, &sub->list, xfs_bitmap_range_cmp);
+	list_sort(NULL, &bitmap->list, xbitmap_range_cmp);
+	list_sort(NULL, &sub->list, xbitmap_range_cmp);
 
 	/*
 	 * Now that we've sorted both lists, we iterate bitmap once, rolling
@@ -124,11 +124,11 @@ xfs_bitmap_disunion(
 	 * list traversal is similar to merge sort, but we're deleting
 	 * instead.  In this manner we avoid O(n^2) operations.
 	 */
-	sub_br = list_first_entry(&sub->list, struct xfs_bitmap_range,
+	sub_br = list_first_entry(&sub->list, struct xbitmap_range,
 			list);
 	lp = bitmap->list.next;
 	while (lp != &bitmap->list) {
-		br = list_entry(lp, struct xfs_bitmap_range, list);
+		br = list_entry(lp, struct xbitmap_range, list);
 
 		/*
 		 * Advance sub_br and/or br until we find a pair that
@@ -181,7 +181,7 @@ xfs_bitmap_disunion(
 			 * Deleting from the middle: add the new right extent
 			 * and then shrink the left extent.
 			 */
-			new_br = kmem_alloc(sizeof(struct xfs_bitmap_range),
+			new_br = kmem_alloc(sizeof(struct xbitmap_range),
 					KM_MAYFAIL);
 			if (!new_br) {
 				error = -ENOMEM;
@@ -247,8 +247,8 @@ out:
  * blocks going from the leaf towards the root.
  */
 int
-xfs_bitmap_set_btcur_path(
-	struct xfs_bitmap	*bitmap,
+xbitmap_set_btcur_path(
+	struct xbitmap		*bitmap,
 	struct xfs_btree_cur	*cur)
 {
 	struct xfs_buf		*bp;
@@ -261,7 +261,7 @@ xfs_bitmap_set_btcur_path(
 		if (!bp)
 			continue;
 		fsb = XFS_DADDR_TO_FSB(cur->bc_mp, bp->b_bn);
-		error = xfs_bitmap_set(bitmap, fsb, 1);
+		error = xbitmap_set(bitmap, fsb, 1);
 		if (error)
 			return error;
 	}
@@ -271,12 +271,12 @@ xfs_bitmap_set_btcur_path(
 
 /* Collect a btree's block in the bitmap. */
 STATIC int
-xfs_bitmap_collect_btblock(
+xbitmap_collect_btblock(
 	struct xfs_btree_cur	*cur,
 	int			level,
 	void			*priv)
 {
-	struct xfs_bitmap	*bitmap = priv;
+	struct xbitmap		*bitmap = priv;
 	struct xfs_buf		*bp;
 	xfs_fsblock_t		fsbno;
 
@@ -285,14 +285,30 @@ xfs_bitmap_collect_btblock(
 		return 0;
 
 	fsbno = XFS_DADDR_TO_FSB(cur->bc_mp, bp->b_bn);
-	return xfs_bitmap_set(bitmap, fsbno, 1);
+	return xbitmap_set(bitmap, fsbno, 1);
 }
 
 /* Walk the btree and mark the bitmap wherever a btree block is found. */
 int
-xfs_bitmap_set_btblocks(
-	struct xfs_bitmap	*bitmap,
+xbitmap_set_btblocks(
+	struct xbitmap		*bitmap,
 	struct xfs_btree_cur	*cur)
 {
-	return xfs_btree_visit_blocks(cur, xfs_bitmap_collect_btblock, bitmap);
+	return xfs_btree_visit_blocks(cur, xbitmap_collect_btblock,
+			XFS_BTREE_VISIT_ALL, bitmap);
+}
+
+/* How many bits are set in this bitmap? */
+uint64_t
+xbitmap_hweight(
+	struct xbitmap		*bitmap)
+{
+	struct xbitmap_range	*bmr;
+	struct xbitmap_range	*n;
+	uint64_t		ret = 0;
+
+	for_each_xbitmap_extent(bmr, n, bitmap)
+		ret += bmr->len;
+
+	return ret;
 }
diff --git a/fs/xfs/scrub/bitmap.h b/fs/xfs/scrub/bitmap.h
index ae8ecbce6fa6..900646b72de1 100644
--- a/fs/xfs/scrub/bitmap.h
+++ b/fs/xfs/scrub/bitmap.h
@@ -6,31 +6,32 @@
 #ifndef __XFS_SCRUB_BITMAP_H__
 #define __XFS_SCRUB_BITMAP_H__
 
-struct xfs_bitmap_range {
+struct xbitmap_range {
 	struct list_head	list;
 	uint64_t		start;
 	uint64_t		len;
 };
 
-struct xfs_bitmap {
+struct xbitmap {
 	struct list_head	list;
 };
 
-void xfs_bitmap_init(struct xfs_bitmap *bitmap);
-void xfs_bitmap_destroy(struct xfs_bitmap *bitmap);
+void xbitmap_init(struct xbitmap *bitmap);
+void xbitmap_destroy(struct xbitmap *bitmap);
 
-#define for_each_xfs_bitmap_extent(bex, n, bitmap) \
+#define for_each_xbitmap_extent(bex, n, bitmap) \
 	list_for_each_entry_safe((bex), (n), &(bitmap)->list, list)
 
-#define for_each_xfs_bitmap_block(b, bex, n, bitmap) \
+#define for_each_xbitmap_block(b, bex, n, bitmap) \
 	list_for_each_entry_safe((bex), (n), &(bitmap)->list, list) \
-		for ((b) = bex->start; (b) < bex->start + bex->len; (b)++)
+		for ((b) = (bex)->start; (b) < (bex)->start + (bex)->len; (b)++)
 
-int xfs_bitmap_set(struct xfs_bitmap *bitmap, uint64_t start, uint64_t len);
-int xfs_bitmap_disunion(struct xfs_bitmap *bitmap, struct xfs_bitmap *sub);
-int xfs_bitmap_set_btcur_path(struct xfs_bitmap *bitmap,
+int xbitmap_set(struct xbitmap *bitmap, uint64_t start, uint64_t len);
+int xbitmap_disunion(struct xbitmap *bitmap, struct xbitmap *sub);
+int xbitmap_set_btcur_path(struct xbitmap *bitmap,
 		struct xfs_btree_cur *cur);
-int xfs_bitmap_set_btblocks(struct xfs_bitmap *bitmap,
+int xbitmap_set_btblocks(struct xbitmap *bitmap,
 		struct xfs_btree_cur *cur);
+uint64_t xbitmap_hweight(struct xbitmap *bitmap);
 
 #endif	/* __XFS_SCRUB_BITMAP_H__ */
diff --git a/fs/xfs/scrub/bmap.c b/fs/xfs/scrub/bmap.c
index fa6ea6407992..add8598eacd5 100644
--- a/fs/xfs/scrub/bmap.c
+++ b/fs/xfs/scrub/bmap.c
@@ -374,7 +374,7 @@ xchk_bmapbt_rec(
 	struct xfs_bmbt_irec	iext_irec;
 	struct xfs_iext_cursor	icur;
 	struct xchk_bmap_info	*info = bs->private;
-	struct xfs_inode	*ip = bs->cur->bc_private.b.ip;
+	struct xfs_inode	*ip = bs->cur->bc_ino.ip;
 	struct xfs_buf		*bp = NULL;
 	struct xfs_btree_block	*block;
 	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, info->whichfork);
@@ -501,7 +501,7 @@ xchk_bmap_check_rmap(
 			xchk_fblock_set_corrupt(sc, sbcri->whichfork,
 					rec->rm_offset);
 		if (irec.br_startblock != XFS_AGB_TO_FSB(sc->mp,
-				cur->bc_private.a.agno, rec->rm_startblock))
+				cur->bc_ag.agno, rec->rm_startblock))
 			xchk_fblock_set_corrupt(sc, sbcri->whichfork,
 					rec->rm_offset);
 		if (irec.br_blockcount > rec->rm_blockcount)
diff --git a/fs/xfs/scrub/common.h b/fs/xfs/scrub/common.h
index 003a772cd26c..2e50d146105d 100644
--- a/fs/xfs/scrub/common.h
+++ b/fs/xfs/scrub/common.h
@@ -14,8 +14,15 @@
 static inline bool
 xchk_should_terminate(
 	struct xfs_scrub	*sc,
-	int				*error)
+	int			*error)
 {
+	/*
+	 * If preemption is disabled, we need to yield to the scheduler every
+	 * few seconds so that we don't run afoul of the soft lockup watchdog
+	 * or RCU stall detector.
+	 */
+	cond_resched();
+
 	if (fatal_signal_pending(current)) {
 		if (*error == 0)
 			*error = -EAGAIN;
diff --git a/fs/xfs/scrub/dabtree.c b/fs/xfs/scrub/dabtree.c
index 77ff9f97bcda..9a2e27ac1300 100644
--- a/fs/xfs/scrub/dabtree.c
+++ b/fs/xfs/scrub/dabtree.c
@@ -77,40 +77,18 @@ xchk_da_set_corrupt(
 			__return_address);
 }
 
-/* Find an entry at a certain level in a da btree. */
-STATIC void *
-xchk_da_btree_entry(
-	struct xchk_da_btree	*ds,
-	int			level,
-	int			rec)
+static struct xfs_da_node_entry *
+xchk_da_btree_node_entry(
+	struct xchk_da_btree		*ds,
+	int				level)
 {
-	char			*ents;
-	struct xfs_da_state_blk	*blk;
-	void			*baddr;
+	struct xfs_da_state_blk		*blk = &ds->state->path.blk[level];
+	struct xfs_da3_icnode_hdr	hdr;
 
-	/* Dispatch the entry finding function. */
-	blk = &ds->state->path.blk[level];
-	baddr = blk->bp->b_addr;
-	switch (blk->magic) {
-	case XFS_ATTR_LEAF_MAGIC:
-	case XFS_ATTR3_LEAF_MAGIC:
-		ents = (char *)xfs_attr3_leaf_entryp(baddr);
-		return ents + (rec * sizeof(struct xfs_attr_leaf_entry));
-	case XFS_DIR2_LEAFN_MAGIC:
-	case XFS_DIR3_LEAFN_MAGIC:
-		ents = (char *)ds->dargs.dp->d_ops->leaf_ents_p(baddr);
-		return ents + (rec * sizeof(struct xfs_dir2_leaf_entry));
-	case XFS_DIR2_LEAF1_MAGIC:
-	case XFS_DIR3_LEAF1_MAGIC:
-		ents = (char *)ds->dargs.dp->d_ops->leaf_ents_p(baddr);
-		return ents + (rec * sizeof(struct xfs_dir2_leaf_entry));
-	case XFS_DA_NODE_MAGIC:
-	case XFS_DA3_NODE_MAGIC:
-		ents = (char *)ds->dargs.dp->d_ops->node_tree_p(baddr);
-		return ents + (rec * sizeof(struct xfs_da_node_entry));
-	}
+	ASSERT(blk->magic == XFS_DA_NODE_MAGIC);
 
-	return NULL;
+	xfs_da3_node_hdr_from_disk(ds->sc->mp, &hdr, blk->bp->b_addr);
+	return hdr.btree + blk->index;
 }
 
 /* Scrub a da btree hash (key). */
@@ -120,7 +98,6 @@ xchk_da_btree_hash(
 	int				level,
 	__be32				*hashp)
 {
-	struct xfs_da_state_blk		*blks;
 	struct xfs_da_node_entry	*entry;
 	xfs_dahash_t			hash;
 	xfs_dahash_t			parent_hash;
@@ -135,8 +112,7 @@ xchk_da_btree_hash(
 		return 0;
 
 	/* Is this hash no larger than the parent hash? */
-	blks = ds->state->path.blk;
-	entry = xchk_da_btree_entry(ds, level - 1, blks[level - 1].index);
+	entry = xchk_da_btree_node_entry(ds, level - 1);
 	parent_hash = be32_to_cpu(entry->hashval);
 	if (parent_hash < hash)
 		xchk_da_set_corrupt(ds, level);
@@ -243,19 +219,21 @@ xchk_da_btree_block_check_sibling(
 	int			direction,
 	xfs_dablk_t		sibling)
 {
+	struct xfs_da_state_path *path = &ds->state->path;
+	struct xfs_da_state_path *altpath = &ds->state->altpath;
 	int			retval;
+	int			plevel;
 	int			error;
 
-	memcpy(&ds->state->altpath, &ds->state->path,
-			sizeof(ds->state->altpath));
+	memcpy(altpath, path, sizeof(ds->state->altpath));
 
 	/*
 	 * If the pointer is null, we shouldn't be able to move the upper
 	 * level pointer anywhere.
 	 */
 	if (sibling == 0) {
-		error = xfs_da3_path_shift(ds->state, &ds->state->altpath,
-				direction, false, &retval);
+		error = xfs_da3_path_shift(ds->state, altpath, direction,
+				false, &retval);
 		if (error == 0 && retval == 0)
 			xchk_da_set_corrupt(ds, level);
 		error = 0;
@@ -263,27 +241,33 @@ xchk_da_btree_block_check_sibling(
 	}
 
 	/* Move the alternate cursor one block in the direction given. */
-	error = xfs_da3_path_shift(ds->state, &ds->state->altpath,
-			direction, false, &retval);
+	error = xfs_da3_path_shift(ds->state, altpath, direction, false,
+			&retval);
 	if (!xchk_da_process_error(ds, level, &error))
-		return error;
+		goto out;
 	if (retval) {
 		xchk_da_set_corrupt(ds, level);
-		return error;
+		goto out;
 	}
-	if (ds->state->altpath.blk[level].bp)
-		xchk_buffer_recheck(ds->sc,
-				ds->state->altpath.blk[level].bp);
+	if (altpath->blk[level].bp)
+		xchk_buffer_recheck(ds->sc, altpath->blk[level].bp);
 
 	/* Compare upper level pointer to sibling pointer. */
-	if (ds->state->altpath.blk[level].blkno != sibling)
+	if (altpath->blk[level].blkno != sibling)
 		xchk_da_set_corrupt(ds, level);
-	if (ds->state->altpath.blk[level].bp) {
-		xfs_trans_brelse(ds->dargs.trans,
-				ds->state->altpath.blk[level].bp);
-		ds->state->altpath.blk[level].bp = NULL;
-	}
+
 out:
+	/* Free all buffers in the altpath that aren't referenced from path. */
+	for (plevel = 0; plevel < altpath->active; plevel++) {
+		if (altpath->blk[plevel].bp == NULL ||
+		    (plevel < path->active &&
+		     altpath->blk[plevel].bp == path->blk[plevel].bp))
+			continue;
+
+		xfs_trans_brelse(ds->dargs.trans, altpath->blk[plevel].bp);
+		altpath->blk[plevel].bp = NULL;
+	}
+
 	return error;
 }
 
@@ -355,8 +339,8 @@ xchk_da_btree_block(
 		goto out_nobuf;
 
 	/* Read the buffer. */
-	error = xfs_da_read_buf(dargs->trans, dargs->dp, blk->blkno, -2,
-			&blk->bp, dargs->whichfork,
+	error = xfs_da_read_buf(dargs->trans, dargs->dp, blk->blkno,
+			XFS_DABUF_MAP_HOLE_OK, &blk->bp, dargs->whichfork,
 			&xchk_da_btree_buf_ops);
 	if (!xchk_da_process_error(ds, level, &error))
 		goto out_nobuf;
@@ -433,8 +417,8 @@ xchk_da_btree_block(
 				XFS_BLFT_DA_NODE_BUF);
 		blk->magic = XFS_DA_NODE_MAGIC;
 		node = blk->bp->b_addr;
-		ip->d_ops->node_hdr_from_disk(&nodehdr, node);
-		btree = ip->d_ops->node_tree_p(node);
+		xfs_da3_node_hdr_from_disk(ip->i_mount, &nodehdr, node);
+		btree = nodehdr.btree;
 		*pmaxrecs = nodehdr.count;
 		blk->hashval = be32_to_cpu(btree[*pmaxrecs - 1].hashval);
 		if (level == 0) {
@@ -479,14 +463,12 @@ xchk_da_btree(
 	struct xfs_mount		*mp = sc->mp;
 	struct xfs_da_state_blk		*blks;
 	struct xfs_da_node_entry	*key;
-	void				*rec;
 	xfs_dablk_t			blkno;
 	int				level;
 	int				error;
 
 	/* Skip short format data structures; no btree to scan. */
-	if (XFS_IFORK_FORMAT(sc->ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
-	    XFS_IFORK_FORMAT(sc->ip, whichfork) != XFS_DINODE_FMT_BTREE)
+	if (!xfs_ifork_has_extents(sc->ip, whichfork))
 		return 0;
 
 	/* Set up initial da state. */
@@ -538,9 +520,7 @@ xchk_da_btree(
 			}
 
 			/* Dispatch record scrubbing. */
-			rec = xchk_da_btree_entry(&ds, level,
-					blks[level].index);
-			error = scrub_fn(&ds, level, rec);
+			error = scrub_fn(&ds, level);
 			if (error)
 				break;
 			if (xchk_should_terminate(sc, &error) ||
@@ -562,7 +542,7 @@ xchk_da_btree(
 		}
 
 		/* Hashes in order for scrub? */
-		key = xchk_da_btree_entry(&ds, level, blks[level].index);
+		key = xchk_da_btree_node_entry(&ds, level);
 		error = xchk_da_btree_hash(&ds, level, &key->hashval);
 		if (error)
 			goto out;
diff --git a/fs/xfs/scrub/dabtree.h b/fs/xfs/scrub/dabtree.h
index cb3f0003245b..1f3515c6d5a8 100644
--- a/fs/xfs/scrub/dabtree.h
+++ b/fs/xfs/scrub/dabtree.h
@@ -28,8 +28,7 @@ struct xchk_da_btree {
 	int			tree_level;
 };
 
-typedef int (*xchk_da_btree_rec_fn)(struct xchk_da_btree *ds,
-		int level, void *rec);
+typedef int (*xchk_da_btree_rec_fn)(struct xchk_da_btree *ds, int level);
 
 /* Check for da btree operation errors. */
 bool xchk_da_process_error(struct xchk_da_btree *ds, int level, int *error);
diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c
index 1e2e11721eb9..fe2a6e030c8a 100644
--- a/fs/xfs/scrub/dir.c
+++ b/fs/xfs/scrub/dir.c
@@ -113,6 +113,9 @@ xchk_dir_actor(
 	offset = xfs_dir2_db_to_da(mp->m_dir_geo,
 			xfs_dir2_dataptr_to_db(mp->m_dir_geo, pos));
 
+	if (xchk_should_terminate(sdc->sc, &error))
+		return error;
+
 	/* Does this inode number make sense? */
 	if (!xfs_verify_dir_ino(mp, ino)) {
 		xchk_fblock_set_corrupt(sdc->sc, XFS_DATA_FORK, offset);
@@ -152,6 +155,9 @@ xchk_dir_actor(
 	xname.type = XFS_DIR3_FT_UNKNOWN;
 
 	error = xfs_dir_lookup(sdc->sc->tp, ip, &xname, &lookup_ino, NULL);
+	/* ENOENT means the hash lookup failed and the dir is corrupt */
+	if (error == -ENOENT)
+		error = -EFSCORRUPTED;
 	if (!xchk_fblock_process_error(sdc->sc, XFS_DATA_FORK, offset,
 			&error))
 		goto out;
@@ -179,15 +185,17 @@ out:
 STATIC int
 xchk_dir_rec(
 	struct xchk_da_btree		*ds,
-	int				level,
-	void				*rec)
+	int				level)
 {
+	struct xfs_da_state_blk		*blk = &ds->state->path.blk[level];
 	struct xfs_mount		*mp = ds->state->mp;
-	struct xfs_dir2_leaf_entry	*ent = rec;
 	struct xfs_inode		*dp = ds->dargs.dp;
+	struct xfs_da_geometry		*geo = mp->m_dir_geo;
 	struct xfs_dir2_data_entry	*dent;
 	struct xfs_buf			*bp;
-	char				*p, *endp;
+	struct xfs_dir2_leaf_entry	*ent;
+	unsigned int			end;
+	unsigned int			iter_off;
 	xfs_ino_t			ino;
 	xfs_dablk_t			rec_bno;
 	xfs_dir2_db_t			db;
@@ -195,9 +203,16 @@ xchk_dir_rec(
 	xfs_dir2_dataptr_t		ptr;
 	xfs_dahash_t			calc_hash;
 	xfs_dahash_t			hash;
+	struct xfs_dir3_icleaf_hdr	hdr;
 	unsigned int			tag;
 	int				error;
 
+	ASSERT(blk->magic == XFS_DIR2_LEAF1_MAGIC ||
+	       blk->magic == XFS_DIR2_LEAFN_MAGIC);
+
+	xfs_dir2_leaf_hdr_from_disk(mp, &hdr, blk->bp->b_addr);
+	ent = hdr.ents + blk->index;
+
 	/* Check the hash of the entry. */
 	error = xchk_da_btree_hash(ds, level, &ent->hashval);
 	if (error)
@@ -209,15 +224,16 @@ xchk_dir_rec(
 		return 0;
 
 	/* Find the directory entry's location. */
-	db = xfs_dir2_dataptr_to_db(mp->m_dir_geo, ptr);
-	off = xfs_dir2_dataptr_to_off(mp->m_dir_geo, ptr);
-	rec_bno = xfs_dir2_db_to_da(mp->m_dir_geo, db);
+	db = xfs_dir2_dataptr_to_db(geo, ptr);
+	off = xfs_dir2_dataptr_to_off(geo, ptr);
+	rec_bno = xfs_dir2_db_to_da(geo, db);
 
-	if (rec_bno >= mp->m_dir_geo->leafblk) {
+	if (rec_bno >= geo->leafblk) {
 		xchk_da_set_corrupt(ds, level);
 		goto out;
 	}
-	error = xfs_dir3_data_read(ds->dargs.trans, dp, rec_bno, -2, &bp);
+	error = xfs_dir3_data_read(ds->dargs.trans, dp, rec_bno,
+			XFS_DABUF_MAP_HOLE_OK, &bp);
 	if (!xchk_fblock_process_error(ds->sc, XFS_DATA_FORK, rec_bno,
 			&error))
 		goto out;
@@ -230,38 +246,37 @@ xchk_dir_rec(
 	if (ds->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
 		goto out_relse;
 
-	dent = (struct xfs_dir2_data_entry *)(((char *)bp->b_addr) + off);
+	dent = bp->b_addr + off;
 
 	/* Make sure we got a real directory entry. */
-	p = (char *)mp->m_dir_inode_ops->data_entry_p(bp->b_addr);
-	endp = xfs_dir3_data_endp(mp->m_dir_geo, bp->b_addr);
-	if (!endp) {
+	iter_off = geo->data_entry_offset;
+	end = xfs_dir3_data_end_offset(geo, bp->b_addr);
+	if (!end) {
 		xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
 		goto out_relse;
 	}
-	while (p < endp) {
-		struct xfs_dir2_data_entry	*dep;
-		struct xfs_dir2_data_unused	*dup;
+	for (;;) {
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + iter_off;
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + iter_off;
+
+		if (iter_off >= end) {
+			xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
+			goto out_relse;
+		}
 
-		dup = (struct xfs_dir2_data_unused *)p;
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
-			p += be16_to_cpu(dup->length);
+			iter_off += be16_to_cpu(dup->length);
 			continue;
 		}
-		dep = (struct xfs_dir2_data_entry *)p;
 		if (dep == dent)
 			break;
-		p += mp->m_dir_inode_ops->data_entsize(dep->namelen);
-	}
-	if (p >= endp) {
-		xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
-		goto out_relse;
+		iter_off += xfs_dir2_data_entsize(mp, dep->namelen);
 	}
 
 	/* Retrieve the entry, sanity check it, and compare hashes. */
 	ino = be64_to_cpu(dent->inumber);
 	hash = be32_to_cpu(ent->hashval);
-	tag = be16_to_cpup(dp->d_ops->data_entry_tag_p(dent));
+	tag = be16_to_cpup(xfs_dir2_data_entry_tag_p(mp, dent));
 	if (!xfs_verify_dir_ino(mp, ino) || tag != off)
 		xchk_fblock_set_corrupt(ds->sc, XFS_DATA_FORK, rec_bno);
 	if (dent->namelen == 0) {
@@ -319,19 +334,15 @@ xchk_directory_data_bestfree(
 	struct xfs_buf			*bp;
 	struct xfs_dir2_data_free	*bf;
 	struct xfs_mount		*mp = sc->mp;
-	const struct xfs_dir_ops	*d_ops;
-	char				*ptr;
-	char				*endptr;
 	u16				tag;
 	unsigned int			nr_bestfrees = 0;
 	unsigned int			nr_frees = 0;
 	unsigned int			smallest_bestfree;
 	int				newlen;
-	int				offset;
+	unsigned int			offset;
+	unsigned int			end;
 	int				error;
 
-	d_ops = sc->ip->d_ops;
-
 	if (is_block) {
 		/* dir block format */
 		if (lblk != XFS_B_TO_FSBT(mp, XFS_DIR2_DATA_OFFSET))
@@ -339,7 +350,7 @@ xchk_directory_data_bestfree(
 		error = xfs_dir3_block_read(sc->tp, sc->ip, &bp);
 	} else {
 		/* dir data format */
-		error = xfs_dir3_data_read(sc->tp, sc->ip, lblk, -1, &bp);
+		error = xfs_dir3_data_read(sc->tp, sc->ip, lblk, 0, &bp);
 	}
 	if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error))
 		goto out;
@@ -351,7 +362,7 @@ xchk_directory_data_bestfree(
 		goto out_buf;
 
 	/* Do the bestfrees correspond to actual free space? */
-	bf = d_ops->data_bestfree_p(bp->b_addr);
+	bf = xfs_dir2_data_bestfree_p(mp, bp->b_addr);
 	smallest_bestfree = UINT_MAX;
 	for (dfp = &bf[0]; dfp < &bf[XFS_DIR2_DATA_FD_COUNT]; dfp++) {
 		offset = be16_to_cpu(dfp->offset);
@@ -361,13 +372,13 @@ xchk_directory_data_bestfree(
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
-		dup = (struct xfs_dir2_data_unused *)(bp->b_addr + offset);
+		dup = bp->b_addr + offset;
 		tag = be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup));
 
 		/* bestfree doesn't match the entry it points at? */
 		if (dup->freetag != cpu_to_be16(XFS_DIR2_DATA_FREE_TAG) ||
 		    be16_to_cpu(dup->length) != be16_to_cpu(dfp->length) ||
-		    tag != ((char *)dup - (char *)bp->b_addr)) {
+		    tag != offset) {
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
@@ -383,30 +394,30 @@ xchk_directory_data_bestfree(
 	}
 
 	/* Make sure the bestfrees are actually the best free spaces. */
-	ptr = (char *)d_ops->data_entry_p(bp->b_addr);
-	endptr = xfs_dir3_data_endp(mp->m_dir_geo, bp->b_addr);
+	offset = mp->m_dir_geo->data_entry_offset;
+	end = xfs_dir3_data_end_offset(mp->m_dir_geo, bp->b_addr);
 
 	/* Iterate the entries, stopping when we hit or go past the end. */
-	while (ptr < endptr) {
-		dup = (struct xfs_dir2_data_unused *)ptr;
+	while (offset < end) {
+		dup = bp->b_addr + offset;
+
 		/* Skip real entries */
 		if (dup->freetag != cpu_to_be16(XFS_DIR2_DATA_FREE_TAG)) {
-			struct xfs_dir2_data_entry	*dep;
+			struct xfs_dir2_data_entry *dep = bp->b_addr + offset;
 
-			dep = (struct xfs_dir2_data_entry *)ptr;
-			newlen = d_ops->data_entsize(dep->namelen);
+			newlen = xfs_dir2_data_entsize(mp, dep->namelen);
 			if (newlen <= 0) {
 				xchk_fblock_set_corrupt(sc, XFS_DATA_FORK,
 						lblk);
 				goto out_buf;
 			}
-			ptr += newlen;
+			offset += newlen;
 			continue;
 		}
 
 		/* Spot check this free entry */
 		tag = be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup));
-		if (tag != ((char *)dup - (char *)bp->b_addr)) {
+		if (tag != offset) {
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
@@ -425,13 +436,13 @@ xchk_directory_data_bestfree(
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 			goto out_buf;
 		}
-		ptr += newlen;
-		if (ptr <= endptr)
+		offset += newlen;
+		if (offset <= end)
 			nr_frees++;
 	}
 
 	/* We're required to fill all the space. */
-	if (ptr != endptr)
+	if (offset != end)
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 
 	/* Did we see at least as many free slots as there are bestfrees? */
@@ -458,7 +469,7 @@ xchk_directory_check_freesp(
 {
 	struct xfs_dir2_data_free	*dfp;
 
-	dfp = sc->ip->d_ops->data_bestfree_p(dbp->b_addr);
+	dfp = xfs_dir2_data_bestfree_p(sc->mp, dbp->b_addr);
 
 	if (len != be16_to_cpu(dfp->length))
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
@@ -475,12 +486,10 @@ xchk_directory_leaf1_bestfree(
 	xfs_dablk_t			lblk)
 {
 	struct xfs_dir3_icleaf_hdr	leafhdr;
-	struct xfs_dir2_leaf_entry	*ents;
 	struct xfs_dir2_leaf_tail	*ltp;
 	struct xfs_dir2_leaf		*leaf;
 	struct xfs_buf			*dbp;
 	struct xfs_buf			*bp;
-	const struct xfs_dir_ops	*d_ops = sc->ip->d_ops;
 	struct xfs_da_geometry		*geo = sc->mp->m_dir_geo;
 	__be16				*bestp;
 	__u16				best;
@@ -492,14 +501,13 @@ xchk_directory_leaf1_bestfree(
 	int				error;
 
 	/* Read the free space block. */
-	error = xfs_dir3_leaf_read(sc->tp, sc->ip, lblk, -1, &bp);
+	error = xfs_dir3_leaf_read(sc->tp, sc->ip, lblk, &bp);
 	if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error))
-		goto out;
+		return error;
 	xchk_buffer_recheck(sc, bp);
 
 	leaf = bp->b_addr;
-	d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = d_ops->leaf_ents_p(leaf);
+	xfs_dir2_leaf_hdr_from_disk(sc->ip->i_mount, &leafhdr, leaf);
 	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 	bestcount = be32_to_cpu(ltp->bestcount);
 	bestp = xfs_dir2_leaf_bests_p(ltp);
@@ -521,24 +529,25 @@ xchk_directory_leaf1_bestfree(
 	}
 
 	/* Is the leaf count even remotely sane? */
-	if (leafhdr.count > d_ops->leaf_max_ents(geo)) {
+	if (leafhdr.count > geo->leaf_max_ents) {
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 		goto out;
 	}
 
 	/* Leaves and bests don't overlap in leaf format. */
-	if ((char *)&ents[leafhdr.count] > (char *)bestp) {
+	if ((char *)&leafhdr.ents[leafhdr.count] > (char *)bestp) {
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 		goto out;
 	}
 
 	/* Check hash value order, count stale entries.  */
 	for (i = 0; i < leafhdr.count; i++) {
-		hash = be32_to_cpu(ents[i].hashval);
+		hash = be32_to_cpu(leafhdr.ents[i].hashval);
 		if (i > 0 && lasthash > hash)
 			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 		lasthash = hash;
-		if (ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+		if (leafhdr.ents[i].address ==
+		    cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			stale++;
 	}
 	if (leafhdr.stale != stale)
@@ -552,16 +561,17 @@ xchk_directory_leaf1_bestfree(
 		if (best == NULLDATAOFF)
 			continue;
 		error = xfs_dir3_data_read(sc->tp, sc->ip,
-				i * args->geo->fsbcount, -1, &dbp);
+				i * args->geo->fsbcount, 0, &dbp);
 		if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk,
 				&error))
 			break;
 		xchk_directory_check_freesp(sc, lblk, dbp, best);
 		xfs_trans_brelse(sc->tp, dbp);
 		if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
-			goto out;
+			break;
 	}
 out:
+	xfs_trans_brelse(sc->tp, bp);
 	return error;
 }
 
@@ -575,7 +585,6 @@ xchk_directory_free_bestfree(
 	struct xfs_dir3_icfree_hdr	freehdr;
 	struct xfs_buf			*dbp;
 	struct xfs_buf			*bp;
-	__be16				*bestp;
 	__u16				best;
 	unsigned int			stale = 0;
 	int				i;
@@ -584,7 +593,7 @@ xchk_directory_free_bestfree(
 	/* Read the free space block */
 	error = xfs_dir2_free_read(sc->tp, sc->ip, lblk, &bp);
 	if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error))
-		goto out;
+		return error;
 	xchk_buffer_recheck(sc, bp);
 
 	if (xfs_sb_version_hascrc(&sc->mp->m_sb)) {
@@ -595,20 +604,19 @@ xchk_directory_free_bestfree(
 	}
 
 	/* Check all the entries. */
-	sc->ip->d_ops->free_hdr_from_disk(&freehdr, bp->b_addr);
-	bestp = sc->ip->d_ops->free_bests_p(bp->b_addr);
-	for (i = 0; i < freehdr.nvalid; i++, bestp++) {
-		best = be16_to_cpu(*bestp);
+	xfs_dir2_free_hdr_from_disk(sc->ip->i_mount, &freehdr, bp->b_addr);
+	for (i = 0; i < freehdr.nvalid; i++) {
+		best = be16_to_cpu(freehdr.bests[i]);
 		if (best == NULLDATAOFF) {
 			stale++;
 			continue;
 		}
 		error = xfs_dir3_data_read(sc->tp, sc->ip,
 				(freehdr.firstdb + i) * args->geo->fsbcount,
-				-1, &dbp);
+				0, &dbp);
 		if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, lblk,
 				&error))
-			break;
+			goto out;
 		xchk_directory_check_freesp(sc, lblk, dbp, best);
 		xfs_trans_brelse(sc->tp, dbp);
 	}
@@ -616,6 +624,7 @@ xchk_directory_free_bestfree(
 	if (freehdr.nused + stale != freehdr.nvalid)
 		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
 out:
+	xfs_trans_brelse(sc->tp, bp);
 	return error;
 }
 
diff --git a/fs/xfs/scrub/fscounters.c b/fs/xfs/scrub/fscounters.c
index 98f82d7c8b40..ec2064ed3c30 100644
--- a/fs/xfs/scrub/fscounters.c
+++ b/fs/xfs/scrub/fscounters.c
@@ -83,9 +83,6 @@ xchk_fscount_warmup(
 		error = xfs_alloc_read_agf(mp, sc->tp, agno, 0, &agf_bp);
 		if (error)
 			break;
-		error = -ENOMEM;
-		if (!agf_bp || !agi_bp)
-			break;
 
 		/*
 		 * These are supposed to be initialized by the header read
@@ -104,7 +101,7 @@ next_loop_perag:
 		pag = NULL;
 		error = 0;
 
-		if (fatal_signal_pending(current))
+		if (xchk_should_terminate(sc, &error))
 			break;
 	}
 
@@ -163,6 +160,7 @@ xchk_fscount_aggregate_agcounts(
 	uint64_t		delayed;
 	xfs_agnumber_t		agno;
 	int			tries = 8;
+	int			error = 0;
 
 retry:
 	fsc->icount = 0;
@@ -196,10 +194,13 @@ retry:
 
 		xfs_perag_put(pag);
 
-		if (fatal_signal_pending(current))
+		if (xchk_should_terminate(sc, &error))
 			break;
 	}
 
+	if (error)
+		return error;
+
 	/*
 	 * The global incore space reservation is taken from the incore
 	 * counters, so leave that out of the computation.
diff --git a/fs/xfs/scrub/health.c b/fs/xfs/scrub/health.c
index b2f602811e9d..83d27cdf579b 100644
--- a/fs/xfs/scrub/health.c
+++ b/fs/xfs/scrub/health.c
@@ -11,6 +11,7 @@
 #include "xfs_sb.h"
 #include "xfs_health.h"
 #include "scrub/scrub.h"
+#include "scrub/health.h"
 
 /*
  * Scrub and In-Core Filesystem Health Assessments
diff --git a/fs/xfs/scrub/ialloc.c b/fs/xfs/scrub/ialloc.c
index 681758704fda..64c217eb06a7 100644
--- a/fs/xfs/scrub/ialloc.c
+++ b/fs/xfs/scrub/ialloc.c
@@ -104,7 +104,7 @@ xchk_iallocbt_chunk(
 	xfs_extlen_t			len)
 {
 	struct xfs_mount		*mp = bs->cur->bc_mp;
-	xfs_agnumber_t			agno = bs->cur->bc_private.a.agno;
+	xfs_agnumber_t			agno = bs->cur->bc_ag.agno;
 	xfs_agblock_t			bno;
 
 	bno = XFS_AGINO_TO_AGBNO(mp, agino);
@@ -164,7 +164,7 @@ xchk_iallocbt_check_cluster_ifree(
 	 * the record, compute which fs inode we're talking about.
 	 */
 	agino = irec->ir_startino + irec_ino;
-	fsino = XFS_AGINO_TO_INO(mp, bs->cur->bc_private.a.agno, agino);
+	fsino = XFS_AGINO_TO_INO(mp, bs->cur->bc_ag.agno, agino);
 	irec_free = (irec->ir_free & XFS_INOBT_MASK(irec_ino));
 
 	if (be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC ||
@@ -215,7 +215,7 @@ xchk_iallocbt_check_cluster(
 	struct xfs_dinode		*dip;
 	struct xfs_buf			*cluster_bp;
 	unsigned int			nr_inodes;
-	xfs_agnumber_t			agno = bs->cur->bc_private.a.agno;
+	xfs_agnumber_t			agno = bs->cur->bc_ag.agno;
 	xfs_agblock_t			agbno;
 	unsigned int			cluster_index;
 	uint16_t			cluster_mask = 0;
@@ -426,7 +426,7 @@ xchk_iallocbt_rec(
 	struct xchk_iallocbt		*iabt = bs->private;
 	struct xfs_inobt_rec_incore	irec;
 	uint64_t			holes;
-	xfs_agnumber_t			agno = bs->cur->bc_private.a.agno;
+	xfs_agnumber_t			agno = bs->cur->bc_ag.agno;
 	xfs_agino_t			agino;
 	xfs_extlen_t			len;
 	int				holecount;
diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c
index c962bd534690..5705adc43a75 100644
--- a/fs/xfs/scrub/parent.c
+++ b/fs/xfs/scrub/parent.c
@@ -32,8 +32,10 @@ xchk_setup_parent(
 
 struct xchk_parent_ctx {
 	struct dir_context	dc;
+	struct xfs_scrub	*sc;
 	xfs_ino_t		ino;
 	xfs_nlink_t		nlink;
+	bool			cancelled;
 };
 
 /* Look for a single entry in a directory pointing to an inode. */
@@ -47,11 +49,21 @@ xchk_parent_actor(
 	unsigned		type)
 {
 	struct xchk_parent_ctx	*spc;
+	int			error = 0;
 
 	spc = container_of(dc, struct xchk_parent_ctx, dc);
 	if (spc->ino == ino)
 		spc->nlink++;
-	return 0;
+
+	/*
+	 * If we're facing a fatal signal, bail out.  Store the cancellation
+	 * status separately because the VFS readdir code squashes error codes
+	 * into short directory reads.
+	 */
+	if (xchk_should_terminate(spc->sc, &error))
+		spc->cancelled = true;
+
+	return error;
 }
 
 /* Count the number of dentries in the parent dir that point to this inode. */
@@ -62,10 +74,9 @@ xchk_parent_count_parent_dentries(
 	xfs_nlink_t		*nlink)
 {
 	struct xchk_parent_ctx	spc = {
-		.dc.actor = xchk_parent_actor,
-		.dc.pos = 0,
-		.ino = sc->ip->i_ino,
-		.nlink = 0,
+		.dc.actor	= xchk_parent_actor,
+		.ino		= sc->ip->i_ino,
+		.sc		= sc,
 	};
 	size_t			bufsize;
 	loff_t			oldpos;
@@ -80,7 +91,7 @@ xchk_parent_count_parent_dentries(
 	 */
 	lock_mode = xfs_ilock_data_map_shared(parent);
 	if (parent->i_d.di_nextents > 0)
-		error = xfs_dir3_data_readahead(parent, 0, -1);
+		error = xfs_dir3_data_readahead(parent, 0, 0);
 	xfs_iunlock(parent, lock_mode);
 	if (error)
 		return error;
@@ -97,6 +108,10 @@ xchk_parent_count_parent_dentries(
 		error = xfs_readdir(sc->tp, parent, &spc.dc, bufsize);
 		if (error)
 			goto out;
+		if (spc.cancelled) {
+			error = -EAGAIN;
+			goto out;
+		}
 		if (oldpos == spc.dc.pos)
 			break;
 		oldpos = spc.dc.pos;
diff --git a/fs/xfs/scrub/quota.c b/fs/xfs/scrub/quota.c
index 0a33b4421c32..905a34558361 100644
--- a/fs/xfs/scrub/quota.c
+++ b/fs/xfs/scrub/quota.c
@@ -93,6 +93,10 @@ xchk_quota_item(
 	unsigned long long	rcount;
 	xfs_ino_t		fs_icount;
 	xfs_dqid_t		id = be32_to_cpu(d->d_id);
+	int			error = 0;
+
+	if (xchk_should_terminate(sc, &error))
+		return error;
 
 	/*
 	 * Except for the root dquot, the actual dquot we got must either have
@@ -178,6 +182,9 @@ xchk_quota_item(
 	if (id != 0 && rhard != 0 && rcount > rhard)
 		xchk_fblock_set_warning(sc, XFS_DATA_FORK, offset);
 
+	if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
+		return -EFSCORRUPTED;
+
 	return 0;
 }
 
diff --git a/fs/xfs/scrub/refcount.c b/fs/xfs/scrub/refcount.c
index 0cab11a5d390..beaeb6fa3119 100644
--- a/fs/xfs/scrub/refcount.c
+++ b/fs/xfs/scrub/refcount.c
@@ -336,7 +336,7 @@ xchk_refcountbt_rec(
 {
 	struct xfs_mount	*mp = bs->cur->bc_mp;
 	xfs_agblock_t		*cow_blocks = bs->private;
-	xfs_agnumber_t		agno = bs->cur->bc_private.a.agno;
+	xfs_agnumber_t		agno = bs->cur->bc_ag.agno;
 	xfs_agblock_t		bno;
 	xfs_extlen_t		len;
 	xfs_nlink_t		refcount;
diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c
index b70a88bc975e..db3cfd12803d 100644
--- a/fs/xfs/scrub/repair.c
+++ b/fs/xfs/scrub/repair.c
@@ -208,8 +208,10 @@ xrep_calc_ag_resblks(
 	/* Now grab the block counters from the AGF. */
 	error = xfs_alloc_read_agf(mp, NULL, sm->sm_agno, 0, &bp);
 	if (!error) {
-		aglen = be32_to_cpu(XFS_BUF_TO_AGF(bp)->agf_length);
-		freelen = be32_to_cpu(XFS_BUF_TO_AGF(bp)->agf_freeblks);
+		struct xfs_agf	*agf = bp->b_addr;
+
+		aglen = be32_to_cpu(agf->agf_length);
+		freelen = be32_to_cpu(agf->agf_freeblks);
 		usedlen = aglen - freelen;
 		xfs_buf_relse(bp);
 	}
@@ -341,13 +343,17 @@ xrep_init_btblock(
 	struct xfs_trans		*tp = sc->tp;
 	struct xfs_mount		*mp = sc->mp;
 	struct xfs_buf			*bp;
+	int				error;
 
 	trace_xrep_init_btblock(mp, XFS_FSB_TO_AGNO(mp, fsb),
 			XFS_FSB_TO_AGBNO(mp, fsb), btnum);
 
 	ASSERT(XFS_FSB_TO_AGNO(mp, fsb) == sc->sa.agno);
-	bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, XFS_FSB_TO_DADDR(mp, fsb),
-			XFS_FSB_TO_BB(mp, 1), 0);
+	error = xfs_trans_get_buf(tp, mp->m_ddev_targp,
+			XFS_FSB_TO_DADDR(mp, fsb), XFS_FSB_TO_BB(mp, 1), 0,
+			&bp);
+	if (error)
+		return error;
 	xfs_buf_zero(bp, 0, BBTOB(bp->b_length));
 	xfs_btree_init_block(mp, bp, btnum, 0, 0, sc->sa.agno);
 	xfs_trans_buf_set_type(tp, bp, XFS_BLFT_BTREE_BUF);
@@ -430,10 +436,10 @@ xrep_init_btblock(
 int
 xrep_invalidate_blocks(
 	struct xfs_scrub	*sc,
-	struct xfs_bitmap	*bitmap)
+	struct xbitmap		*bitmap)
 {
-	struct xfs_bitmap_range	*bmr;
-	struct xfs_bitmap_range	*n;
+	struct xbitmap_range	*bmr;
+	struct xbitmap_range	*n;
 	struct xfs_buf		*bp;
 	xfs_fsblock_t		fsbno;
 
@@ -445,7 +451,7 @@ xrep_invalidate_blocks(
 	 * because we never own those; and if we can't TRYLOCK the buffer we
 	 * assume it's owned by someone else.
 	 */
-	for_each_xfs_bitmap_block(fsbno, bmr, n, bitmap) {
+	for_each_xbitmap_block(fsbno, bmr, n, bitmap) {
 		/* Skip AG headers and post-EOFS blocks */
 		if (!xfs_verify_fsbno(sc->mp, fsbno))
 			continue;
@@ -542,8 +548,6 @@ xrep_reap_block(
 		error = xfs_alloc_read_agf(sc->mp, sc->tp, agno, 0, &agf_bp);
 		if (error)
 			return error;
-		if (!agf_bp)
-			return -ENOMEM;
 	} else {
 		agf_bp = sc->sa.agf_bp;
 	}
@@ -593,18 +597,18 @@ out_free:
 int
 xrep_reap_extents(
 	struct xfs_scrub		*sc,
-	struct xfs_bitmap		*bitmap,
+	struct xbitmap			*bitmap,
 	const struct xfs_owner_info	*oinfo,
 	enum xfs_ag_resv_type		type)
 {
-	struct xfs_bitmap_range		*bmr;
-	struct xfs_bitmap_range		*n;
+	struct xbitmap_range		*bmr;
+	struct xbitmap_range		*n;
 	xfs_fsblock_t			fsbno;
 	int				error = 0;
 
 	ASSERT(xfs_sb_version_hasrmapbt(&sc->mp->m_sb));
 
-	for_each_xfs_bitmap_block(fsbno, bmr, n, bitmap) {
+	for_each_xbitmap_block(fsbno, bmr, n, bitmap) {
 		ASSERT(sc->ip != NULL ||
 		       XFS_FSB_TO_AGNO(sc->mp, fsbno) == sc->sa.agno);
 		trace_xrep_dispose_btree_extent(sc->mp,
@@ -613,11 +617,9 @@ xrep_reap_extents(
 
 		error = xrep_reap_block(sc, fsbno, oinfo, type);
 		if (error)
-			goto out;
+			break;
 	}
 
-out:
-	xfs_bitmap_destroy(bitmap);
 	return error;
 }
 
@@ -877,7 +879,7 @@ xrep_find_ag_btree_roots(
 
 	ri.sc = sc;
 	ri.btree_info = btree_info;
-	ri.agf = XFS_BUF_TO_AGF(agf_bp);
+	ri.agf = agf_bp->b_addr;
 	ri.agfl_bp = agfl_bp;
 	for (fab = btree_info; fab->buf_ops; fab++) {
 		ASSERT(agfl_bp || fab->rmap_owner != XFS_RMAP_OWN_AG);
diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h
index 60c61d7052a8..04a47d45605b 100644
--- a/fs/xfs/scrub/repair.h
+++ b/fs/xfs/scrub/repair.h
@@ -28,11 +28,11 @@ int xrep_init_btblock(struct xfs_scrub *sc, xfs_fsblock_t fsb,
 		struct xfs_buf **bpp, xfs_btnum_t btnum,
 		const struct xfs_buf_ops *ops);
 
-struct xfs_bitmap;
+struct xbitmap;
 
 int xrep_fix_freelist(struct xfs_scrub *sc, bool can_shrink);
-int xrep_invalidate_blocks(struct xfs_scrub *sc, struct xfs_bitmap *btlist);
-int xrep_reap_extents(struct xfs_scrub *sc, struct xfs_bitmap *exlist,
+int xrep_invalidate_blocks(struct xfs_scrub *sc, struct xbitmap *btlist);
+int xrep_reap_extents(struct xfs_scrub *sc, struct xbitmap *exlist,
 		const struct xfs_owner_info *oinfo, enum xfs_ag_resv_type type);
 
 struct xrep_find_ag_btree {
@@ -75,7 +75,6 @@ static inline xfs_extlen_t
 xrep_calc_ag_resblks(
 	struct xfs_scrub	*sc)
 {
-	ASSERT(!(sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR));
 	return 0;
 }
 
diff --git a/fs/xfs/scrub/rmap.c b/fs/xfs/scrub/rmap.c
index 8d4cefd761c1..f4fcb4719f41 100644
--- a/fs/xfs/scrub/rmap.c
+++ b/fs/xfs/scrub/rmap.c
@@ -92,7 +92,7 @@ xchk_rmapbt_rec(
 {
 	struct xfs_mount	*mp = bs->cur->bc_mp;
 	struct xfs_rmap_irec	irec;
-	xfs_agnumber_t		agno = bs->cur->bc_private.a.agno;
+	xfs_agnumber_t		agno = bs->cur->bc_ag.agno;
 	bool			non_inode;
 	bool			is_unwritten;
 	bool			is_bmbt;
diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c
index 15c8c5f3f688..8ebf35b115ce 100644
--- a/fs/xfs/scrub/scrub.c
+++ b/fs/xfs/scrub/scrub.c
@@ -16,6 +16,7 @@
 #include "xfs_qm.h"
 #include "xfs_errortag.h"
 #include "xfs_error.h"
+#include "xfs_scrub.h"
 #include "scrub/scrub.h"
 #include "scrub/common.h"
 #include "scrub/trace.h"
@@ -167,6 +168,7 @@ xchk_teardown(
 			xfs_irele(sc->ip);
 		sc->ip = NULL;
 	}
+	sb_end_write(sc->mp->m_super);
 	if (sc->flags & XCHK_REAPING_DISABLED)
 		xchk_start_reaping(sc);
 	if (sc->flags & XCHK_HAS_QUOTAOFFLOCK) {
@@ -489,6 +491,14 @@ xfs_scrub_metadata(
 	sc.ops = &meta_scrub_ops[sm->sm_type];
 	sc.sick_mask = xchk_health_mask_for_scrub_type(sm->sm_type);
 retry_op:
+	/*
+	 * If freeze runs concurrently with a scrub, the freeze can be delayed
+	 * indefinitely as we walk the filesystem and iterate over metadata
+	 * buffers.  Freeze quiesces the log (which waits for the buffer LRU to
+	 * be emptied) and that won't happen while checking is running.
+	 */
+	sb_start_write(mp->m_super);
+
 	/* Set up for the operation. */
 	error = sc.ops->setup(&sc, ip);
 	if (error)
diff --git a/fs/xfs/scrub/trace.c b/fs/xfs/scrub/trace.c
index 9eaab2eb5ed3..2c6c248be823 100644
--- a/fs/xfs/scrub/trace.c
+++ b/fs/xfs/scrub/trace.c
@@ -24,9 +24,9 @@ xchk_btree_cur_fsbno(
 		return XFS_DADDR_TO_FSB(cur->bc_mp, cur->bc_bufs[level]->b_bn);
 	else if (level == cur->bc_nlevels - 1 &&
 		 cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		return XFS_INO_TO_FSB(cur->bc_mp, cur->bc_private.b.ip->i_ino);
+		return XFS_INO_TO_FSB(cur->bc_mp, cur->bc_ino.ip->i_ino);
 	else if (!(cur->bc_flags & XFS_BTREE_LONG_PTRS))
-		return XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_private.a.agno, 0);
+		return XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_ag.agno, 0);
 	return NULLFSBLOCK;
 }
 
diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h
index 3362bae28b46..e46f5cef90da 100644
--- a/fs/xfs/scrub/trace.h
+++ b/fs/xfs/scrub/trace.h
@@ -329,7 +329,7 @@ TRACE_EVENT(xchk_btree_op_error,
 		__field(int, level)
 		__field(xfs_agnumber_t, agno)
 		__field(xfs_agblock_t, bno)
-		__field(int, ptr);
+		__field(int, ptr)
 		__field(int, error)
 		__field(void *, ret_ip)
 	),
@@ -379,7 +379,7 @@ TRACE_EVENT(xchk_ifork_btree_op_error,
 		xfs_fsblock_t fsbno = xchk_btree_cur_fsbno(cur, level);
 		__entry->dev = sc->mp->m_super->s_dev;
 		__entry->ino = sc->ip->i_ino;
-		__entry->whichfork = cur->bc_private.b.whichfork;
+		__entry->whichfork = cur->bc_ino.whichfork;
 		__entry->type = sc->sm->sm_type;
 		__entry->btnum = cur->bc_btnum;
 		__entry->level = level;
@@ -414,7 +414,7 @@ TRACE_EVENT(xchk_btree_error,
 		__field(int, level)
 		__field(xfs_agnumber_t, agno)
 		__field(xfs_agblock_t, bno)
-		__field(int, ptr);
+		__field(int, ptr)
 		__field(void *, ret_ip)
 	),
 	TP_fast_assign(
@@ -452,14 +452,14 @@ TRACE_EVENT(xchk_ifork_btree_error,
 		__field(int, level)
 		__field(xfs_agnumber_t, agno)
 		__field(xfs_agblock_t, bno)
-		__field(int, ptr);
+		__field(int, ptr)
 		__field(void *, ret_ip)
 	),
 	TP_fast_assign(
 		xfs_fsblock_t fsbno = xchk_btree_cur_fsbno(cur, level);
 		__entry->dev = sc->mp->m_super->s_dev;
 		__entry->ino = sc->ip->i_ino;
-		__entry->whichfork = cur->bc_private.b.whichfork;
+		__entry->whichfork = cur->bc_ino.whichfork;
 		__entry->type = sc->sm->sm_type;
 		__entry->btnum = cur->bc_btnum;
 		__entry->level = level;
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 96d7071cfa46..d4c687b5cd06 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -12,8 +12,12 @@
 #include "xfs_inode.h"
 #include "xfs_attr.h"
 #include "xfs_trace.h"
-#include <linux/posix_acl_xattr.h>
+#include "xfs_error.h"
+#include "xfs_acl.h"
+#include "xfs_da_format.h"
+#include "xfs_da_btree.h"
 
+#include <linux/posix_acl_xattr.h>
 
 /*
  * Locking scheme:
@@ -23,6 +27,7 @@
 
 STATIC struct posix_acl *
 xfs_acl_from_disk(
+	struct xfs_mount	*mp,
 	const struct xfs_acl	*aclp,
 	int			len,
 	int			max_entries)
@@ -32,11 +37,18 @@ xfs_acl_from_disk(
 	const struct xfs_acl_entry *ace;
 	unsigned int count, i;
 
-	if (len < sizeof(*aclp))
+	if (len < sizeof(*aclp)) {
+		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, aclp,
+				len);
 		return ERR_PTR(-EFSCORRUPTED);
+	}
+
 	count = be32_to_cpu(aclp->acl_cnt);
-	if (count > max_entries || XFS_ACL_SIZE(count) != len)
+	if (count > max_entries || XFS_ACL_SIZE(count) != len) {
+		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, aclp,
+				len);
 		return ERR_PTR(-EFSCORRUPTED);
+	}
 
 	acl = posix_acl_alloc(count, GFP_KERNEL);
 	if (!acl)
@@ -57,10 +69,12 @@ xfs_acl_from_disk(
 
 		switch (acl_e->e_tag) {
 		case ACL_USER:
-			acl_e->e_uid = xfs_uid_to_kuid(be32_to_cpu(ace->ae_id));
+			acl_e->e_uid = make_kuid(&init_user_ns,
+						 be32_to_cpu(ace->ae_id));
 			break;
 		case ACL_GROUP:
-			acl_e->e_gid = xfs_gid_to_kgid(be32_to_cpu(ace->ae_id));
+			acl_e->e_gid = make_kgid(&init_user_ns,
+						 be32_to_cpu(ace->ae_id));
 			break;
 		case ACL_USER_OBJ:
 		case ACL_GROUP_OBJ:
@@ -93,10 +107,12 @@ xfs_acl_to_disk(struct xfs_acl *aclp, const struct posix_acl *acl)
 		ace->ae_tag = cpu_to_be32(acl_e->e_tag);
 		switch (acl_e->e_tag) {
 		case ACL_USER:
-			ace->ae_id = cpu_to_be32(xfs_kuid_to_uid(acl_e->e_uid));
+			ace->ae_id = cpu_to_be32(
+					from_kuid(&init_user_ns, acl_e->e_uid));
 			break;
 		case ACL_GROUP:
-			ace->ae_id = cpu_to_be32(xfs_kgid_to_gid(acl_e->e_gid));
+			ace->ae_id = cpu_to_be32(
+					from_kgid(&init_user_ns, acl_e->e_gid));
 			break;
 		default:
 			ace->ae_id = cpu_to_be32(ACL_UNDEFINED_ID);
@@ -110,99 +126,86 @@ xfs_acl_to_disk(struct xfs_acl *aclp, const struct posix_acl *acl)
 struct posix_acl *
 xfs_get_acl(struct inode *inode, int type)
 {
-	struct xfs_inode *ip = XFS_I(inode);
-	struct posix_acl *acl = NULL;
-	struct xfs_acl *xfs_acl = NULL;
-	unsigned char *ea_name;
-	int error;
-	int len;
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
+	struct posix_acl	*acl = NULL;
+	struct xfs_da_args	args = {
+		.dp		= ip,
+		.attr_filter	= XFS_ATTR_ROOT,
+		.valuelen	= XFS_ACL_MAX_SIZE(mp),
+	};
+	int			error;
 
 	trace_xfs_get_acl(ip);
 
 	switch (type) {
 	case ACL_TYPE_ACCESS:
-		ea_name = SGI_ACL_FILE;
+		args.name = SGI_ACL_FILE;
 		break;
 	case ACL_TYPE_DEFAULT:
-		ea_name = SGI_ACL_DEFAULT;
+		args.name = SGI_ACL_DEFAULT;
 		break;
 	default:
 		BUG();
 	}
+	args.namelen = strlen(args.name);
 
 	/*
-	 * If we have a cached ACLs value just return it, not need to
-	 * go out to the disk.
+	 * If the attribute doesn't exist make sure we have a negative cache
+	 * entry, for any other error assume it is transient.
 	 */
-	len = XFS_ACL_MAX_SIZE(ip->i_mount);
-	error = xfs_attr_get(ip, ea_name, (unsigned char **)&xfs_acl, &len,
-				ATTR_ALLOC | ATTR_ROOT);
-	if (error) {
-		/*
-		 * If the attribute doesn't exist make sure we have a negative
-		 * cache entry, for any other error assume it is transient.
-		 */
-		if (error != -ENOATTR)
-			acl = ERR_PTR(error);
-	} else  {
-		acl = xfs_acl_from_disk(xfs_acl, len,
-					XFS_ACL_MAX_ENTRIES(ip->i_mount));
-		kmem_free(xfs_acl);
+	error = xfs_attr_get(&args);
+	if (!error) {
+		acl = xfs_acl_from_disk(mp, args.value, args.valuelen,
+					XFS_ACL_MAX_ENTRIES(mp));
+	} else if (error != -ENOATTR) {
+		acl = ERR_PTR(error);
 	}
+
+	kmem_free(args.value);
 	return acl;
 }
 
 int
 __xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 {
-	struct xfs_inode *ip = XFS_I(inode);
-	unsigned char *ea_name;
-	int error;
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_da_args	args = {
+		.dp		= ip,
+		.attr_filter	= XFS_ATTR_ROOT,
+	};
+	int			error;
 
 	switch (type) {
 	case ACL_TYPE_ACCESS:
-		ea_name = SGI_ACL_FILE;
+		args.name = SGI_ACL_FILE;
 		break;
 	case ACL_TYPE_DEFAULT:
 		if (!S_ISDIR(inode->i_mode))
 			return acl ? -EACCES : 0;
-		ea_name = SGI_ACL_DEFAULT;
+		args.name = SGI_ACL_DEFAULT;
 		break;
 	default:
 		return -EINVAL;
 	}
+	args.namelen = strlen(args.name);
 
 	if (acl) {
-		struct xfs_acl *xfs_acl;
-		int len = XFS_ACL_MAX_SIZE(ip->i_mount);
-
-		xfs_acl = kmem_zalloc_large(len, 0);
-		if (!xfs_acl)
+		args.valuelen = XFS_ACL_SIZE(acl->a_count);
+		args.value = kmem_zalloc_large(args.valuelen, 0);
+		if (!args.value)
 			return -ENOMEM;
-
-		xfs_acl_to_disk(xfs_acl, acl);
-
-		/* subtract away the unused acl entries */
-		len -= sizeof(struct xfs_acl_entry) *
-			 (XFS_ACL_MAX_ENTRIES(ip->i_mount) - acl->a_count);
-
-		error = xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl,
-				len, ATTR_ROOT);
-
-		kmem_free(xfs_acl);
-	} else {
-		/*
-		 * A NULL ACL argument means we want to remove the ACL.
-		 */
-		error = xfs_attr_remove(ip, ea_name, ATTR_ROOT);
-
-		/*
-		 * If the attribute didn't exist to start with that's fine.
-		 */
-		if (error == -ENOATTR)
-			error = 0;
+		xfs_acl_to_disk(args.value, acl);
 	}
 
+	error = xfs_attr_set(&args);
+	kmem_free(args.value);
+
+	/*
+	 * If the attribute didn't exist to start with that's fine.
+	 */
+	if (!acl && error == -ENOATTR)
+		error = 0;
 	if (!error)
 		set_cached_acl(inode, type, acl);
 	return error;
@@ -262,3 +265,19 @@ xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 
 	return error;
 }
+
+/*
+ * Invalidate any cached ACLs if the user has bypassed the ACL interface.
+ * We don't validate the content whatsoever so it is caller responsibility to
+ * provide data in valid format and ensure i_mode is consistent.
+ */
+void
+xfs_forget_acl(
+	struct inode		*inode,
+	const char		*name)
+{
+	if (!strcmp(name, SGI_ACL_FILE))
+		forget_cached_acl(inode, ACL_TYPE_ACCESS);
+	else if (!strcmp(name, SGI_ACL_DEFAULT))
+		forget_cached_acl(inode, ACL_TYPE_DEFAULT);
+}
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 94615e34bc86..c042c0868016 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -13,14 +13,16 @@ struct posix_acl;
 extern struct posix_acl *xfs_get_acl(struct inode *inode, int type);
 extern int xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type);
 extern int __xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type);
+void xfs_forget_acl(struct inode *inode, const char *name);
 #else
 static inline struct posix_acl *xfs_get_acl(struct inode *inode, int type)
 {
 	return NULL;
 }
 # define xfs_set_acl					NULL
+static inline void xfs_forget_acl(struct inode *inode, const char *name)
+{
+}
 #endif /* CONFIG_XFS_POSIX_ACL */
 
-extern void xfs_forget_acl(struct inode *inode, const char *name, int xflags);
-
 #endif	/* __XFS_ACL_H__ */
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index f16d5f196c6b..1fd4fb7a607c 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -18,108 +18,22 @@
 #include "xfs_bmap_util.h"
 #include "xfs_reflink.h"
 
-/*
- * structure owned by writepages passed to individual writepage calls
- */
 struct xfs_writepage_ctx {
-	struct xfs_bmbt_irec    imap;
-	int			fork;
+	struct iomap_writepage_ctx ctx;
 	unsigned int		data_seq;
 	unsigned int		cow_seq;
-	struct xfs_ioend	*ioend;
 };
 
-struct block_device *
-xfs_find_bdev_for_inode(
-	struct inode		*inode)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-
-	if (XFS_IS_REALTIME_INODE(ip))
-		return mp->m_rtdev_targp->bt_bdev;
-	else
-		return mp->m_ddev_targp->bt_bdev;
-}
-
-struct dax_device *
-xfs_find_daxdev_for_inode(
-	struct inode		*inode)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-
-	if (XFS_IS_REALTIME_INODE(ip))
-		return mp->m_rtdev_targp->bt_daxdev;
-	else
-		return mp->m_ddev_targp->bt_daxdev;
-}
-
-static void
-xfs_finish_page_writeback(
-	struct inode		*inode,
-	struct bio_vec	*bvec,
-	int			error)
-{
-	struct iomap_page	*iop = to_iomap_page(bvec->bv_page);
-
-	if (error) {
-		SetPageError(bvec->bv_page);
-		mapping_set_error(inode->i_mapping, -EIO);
-	}
-
-	ASSERT(iop || i_blocksize(inode) == PAGE_SIZE);
-	ASSERT(!iop || atomic_read(&iop->write_count) > 0);
-
-	if (!iop || atomic_dec_and_test(&iop->write_count))
-		end_page_writeback(bvec->bv_page);
-}
-
-/*
- * We're now finished for good with this ioend structure.  Update the page
- * state, release holds on bios, and finally free up memory.  Do not use the
- * ioend after this.
- */
-STATIC void
-xfs_destroy_ioend(
-	struct xfs_ioend	*ioend,
-	int			error)
+static inline struct xfs_writepage_ctx *
+XFS_WPC(struct iomap_writepage_ctx *ctx)
 {
-	struct inode		*inode = ioend->io_inode;
-	struct bio		*bio = &ioend->io_inline_bio;
-	struct bio		*last = ioend->io_bio, *next;
-	u64			start = bio->bi_iter.bi_sector;
-	bool			quiet = bio_flagged(bio, BIO_QUIET);
-
-	for (bio = &ioend->io_inline_bio; bio; bio = next) {
-		struct bio_vec	*bvec;
-		struct bvec_iter_all iter_all;
-
-		/*
-		 * For the last bio, bi_private points to the ioend, so we
-		 * need to explicitly end the iteration here.
-		 */
-		if (bio == last)
-			next = NULL;
-		else
-			next = bio->bi_private;
-
-		/* walk each page on bio, ending page IO on them */
-		bio_for_each_segment_all(bvec, bio, iter_all)
-			xfs_finish_page_writeback(inode, bvec, error);
-		bio_put(bio);
-	}
-
-	if (unlikely(error && !quiet)) {
-		xfs_err_ratelimited(XFS_I(inode)->i_mount,
-			"writeback error on sector %llu", start);
-	}
+	return container_of(ctx, struct xfs_writepage_ctx, ctx);
 }
 
 /*
  * Fast and loose check if this write could update the on-disk inode size.
  */
-static inline bool xfs_ioend_is_append(struct xfs_ioend *ioend)
+static inline bool xfs_ioend_is_append(struct iomap_ioend *ioend)
 {
 	return ioend->io_offset + ioend->io_size >
 		XFS_I(ioend->io_inode)->i_d.di_size;
@@ -127,7 +41,7 @@ static inline bool xfs_ioend_is_append(struct xfs_ioend *ioend)
 
 STATIC int
 xfs_setfilesize_trans_alloc(
-	struct xfs_ioend	*ioend)
+	struct iomap_ioend	*ioend)
 {
 	struct xfs_mount	*mp = XFS_I(ioend->io_inode)->i_mount;
 	struct xfs_trans	*tp;
@@ -137,7 +51,7 @@ xfs_setfilesize_trans_alloc(
 	if (error)
 		return error;
 
-	ioend->io_append_trans = tp;
+	ioend->io_private = tp;
 
 	/*
 	 * We may pass freeze protection with a transaction.  So tell lockdep
@@ -200,11 +114,11 @@ xfs_setfilesize(
 
 STATIC int
 xfs_setfilesize_ioend(
-	struct xfs_ioend	*ioend,
+	struct iomap_ioend	*ioend,
 	int			error)
 {
 	struct xfs_inode	*ip = XFS_I(ioend->io_inode);
-	struct xfs_trans	*tp = ioend->io_append_trans;
+	struct xfs_trans	*tp = ioend->io_private;
 
 	/*
 	 * The transaction may have been allocated in the I/O submission thread,
@@ -228,9 +142,8 @@ xfs_setfilesize_ioend(
  */
 STATIC void
 xfs_end_ioend(
-	struct xfs_ioend	*ioend)
+	struct iomap_ioend	*ioend)
 {
-	struct list_head	ioend_list;
 	struct xfs_inode	*ip = XFS_I(ioend->io_inode);
 	xfs_off_t		offset = ioend->io_offset;
 	size_t			size = ioend->io_size;
@@ -257,7 +170,7 @@ xfs_end_ioend(
 	 */
 	error = blk_status_to_errno(ioend->io_bio->bi_status);
 	if (unlikely(error)) {
-		if (ioend->io_fork == XFS_COW_FORK)
+		if (ioend->io_flags & IOMAP_F_SHARED)
 			xfs_reflink_cancel_cow_range(ip, offset, size, true);
 		goto done;
 	}
@@ -265,154 +178,86 @@ xfs_end_ioend(
 	/*
 	 * Success: commit the COW or unwritten blocks if needed.
 	 */
-	if (ioend->io_fork == XFS_COW_FORK)
+	if (ioend->io_flags & IOMAP_F_SHARED)
 		error = xfs_reflink_end_cow(ip, offset, size);
-	else if (ioend->io_state == XFS_EXT_UNWRITTEN)
+	else if (ioend->io_type == IOMAP_UNWRITTEN)
 		error = xfs_iomap_write_unwritten(ip, offset, size, false);
 	else
-		ASSERT(!xfs_ioend_is_append(ioend) || ioend->io_append_trans);
+		ASSERT(!xfs_ioend_is_append(ioend) || ioend->io_private);
 
 done:
-	if (ioend->io_append_trans)
+	if (ioend->io_private)
 		error = xfs_setfilesize_ioend(ioend, error);
-	list_replace_init(&ioend->io_list, &ioend_list);
-	xfs_destroy_ioend(ioend, error);
-
-	while (!list_empty(&ioend_list)) {
-		ioend = list_first_entry(&ioend_list, struct xfs_ioend,
-				io_list);
-		list_del_init(&ioend->io_list);
-		xfs_destroy_ioend(ioend, error);
-	}
-
+	iomap_finish_ioends(ioend, error);
 	memalloc_nofs_restore(nofs_flag);
 }
 
 /*
- * We can merge two adjacent ioends if they have the same set of work to do.
- */
-static bool
-xfs_ioend_can_merge(
-	struct xfs_ioend	*ioend,
-	struct xfs_ioend	*next)
-{
-	if (ioend->io_bio->bi_status != next->io_bio->bi_status)
-		return false;
-	if ((ioend->io_fork == XFS_COW_FORK) ^ (next->io_fork == XFS_COW_FORK))
-		return false;
-	if ((ioend->io_state == XFS_EXT_UNWRITTEN) ^
-	    (next->io_state == XFS_EXT_UNWRITTEN))
-		return false;
-	if (ioend->io_offset + ioend->io_size != next->io_offset)
-		return false;
-	return true;
-}
-
-/*
  * If the to be merged ioend has a preallocated transaction for file
  * size updates we need to ensure the ioend it is merged into also
  * has one.  If it already has one we can simply cancel the transaction
  * as it is guaranteed to be clean.
  */
 static void
-xfs_ioend_merge_append_transactions(
-	struct xfs_ioend	*ioend,
-	struct xfs_ioend	*next)
+xfs_ioend_merge_private(
+	struct iomap_ioend	*ioend,
+	struct iomap_ioend	*next)
 {
-	if (!ioend->io_append_trans) {
-		ioend->io_append_trans = next->io_append_trans;
-		next->io_append_trans = NULL;
+	if (!ioend->io_private) {
+		ioend->io_private = next->io_private;
+		next->io_private = NULL;
 	} else {
 		xfs_setfilesize_ioend(next, -ECANCELED);
 	}
 }
 
-/* Try to merge adjacent completions. */
-STATIC void
-xfs_ioend_try_merge(
-	struct xfs_ioend	*ioend,
-	struct list_head	*more_ioends)
-{
-	struct xfs_ioend	*next_ioend;
-
-	while (!list_empty(more_ioends)) {
-		next_ioend = list_first_entry(more_ioends, struct xfs_ioend,
-				io_list);
-		if (!xfs_ioend_can_merge(ioend, next_ioend))
-			break;
-		list_move_tail(&next_ioend->io_list, &ioend->io_list);
-		ioend->io_size += next_ioend->io_size;
-		if (next_ioend->io_append_trans)
-			xfs_ioend_merge_append_transactions(ioend, next_ioend);
-	}
-}
-
-/* list_sort compare function for ioends */
-static int
-xfs_ioend_compare(
-	void			*priv,
-	struct list_head	*a,
-	struct list_head	*b)
-{
-	struct xfs_ioend	*ia;
-	struct xfs_ioend	*ib;
-
-	ia = container_of(a, struct xfs_ioend, io_list);
-	ib = container_of(b, struct xfs_ioend, io_list);
-	if (ia->io_offset < ib->io_offset)
-		return -1;
-	else if (ia->io_offset > ib->io_offset)
-		return 1;
-	return 0;
-}
-
 /* Finish all pending io completions. */
 void
 xfs_end_io(
 	struct work_struct	*work)
 {
-	struct xfs_inode	*ip;
-	struct xfs_ioend	*ioend;
-	struct list_head	completion_list;
+	struct xfs_inode	*ip =
+		container_of(work, struct xfs_inode, i_ioend_work);
+	struct iomap_ioend	*ioend;
+	struct list_head	tmp;
 	unsigned long		flags;
 
-	ip = container_of(work, struct xfs_inode, i_ioend_work);
-
 	spin_lock_irqsave(&ip->i_ioend_lock, flags);
-	list_replace_init(&ip->i_ioend_list, &completion_list);
+	list_replace_init(&ip->i_ioend_list, &tmp);
 	spin_unlock_irqrestore(&ip->i_ioend_lock, flags);
 
-	list_sort(NULL, &completion_list, xfs_ioend_compare);
-
-	while (!list_empty(&completion_list)) {
-		ioend = list_first_entry(&completion_list, struct xfs_ioend,
-				io_list);
+	iomap_sort_ioends(&tmp);
+	while ((ioend = list_first_entry_or_null(&tmp, struct iomap_ioend,
+			io_list))) {
 		list_del_init(&ioend->io_list);
-		xfs_ioend_try_merge(ioend, &completion_list);
+		iomap_ioend_try_merge(ioend, &tmp, xfs_ioend_merge_private);
 		xfs_end_ioend(ioend);
 	}
 }
 
+static inline bool xfs_ioend_needs_workqueue(struct iomap_ioend *ioend)
+{
+	return ioend->io_private ||
+		ioend->io_type == IOMAP_UNWRITTEN ||
+		(ioend->io_flags & IOMAP_F_SHARED);
+}
+
 STATIC void
 xfs_end_bio(
 	struct bio		*bio)
 {
-	struct xfs_ioend	*ioend = bio->bi_private;
+	struct iomap_ioend	*ioend = bio->bi_private;
 	struct xfs_inode	*ip = XFS_I(ioend->io_inode);
-	struct xfs_mount	*mp = ip->i_mount;
 	unsigned long		flags;
 
-	if (ioend->io_fork == XFS_COW_FORK ||
-	    ioend->io_state == XFS_EXT_UNWRITTEN ||
-	    ioend->io_append_trans != NULL) {
-		spin_lock_irqsave(&ip->i_ioend_lock, flags);
-		if (list_empty(&ip->i_ioend_list))
-			WARN_ON_ONCE(!queue_work(mp->m_unwritten_workqueue,
-						 &ip->i_ioend_work));
-		list_add_tail(&ioend->io_list, &ip->i_ioend_list);
-		spin_unlock_irqrestore(&ip->i_ioend_lock, flags);
-	} else
-		xfs_destroy_ioend(ioend, blk_status_to_errno(bio->bi_status));
+	ASSERT(xfs_ioend_needs_workqueue(ioend));
+
+	spin_lock_irqsave(&ip->i_ioend_lock, flags);
+	if (list_empty(&ip->i_ioend_list))
+		WARN_ON_ONCE(!queue_work(ip->i_mount->m_unwritten_workqueue,
+					 &ip->i_ioend_work));
+	list_add_tail(&ioend->io_list, &ip->i_ioend_list);
+	spin_unlock_irqrestore(&ip->i_ioend_lock, flags);
 }
 
 /*
@@ -421,19 +266,19 @@ xfs_end_bio(
  */
 static bool
 xfs_imap_valid(
-	struct xfs_writepage_ctx	*wpc,
+	struct iomap_writepage_ctx	*wpc,
 	struct xfs_inode		*ip,
-	xfs_fileoff_t			offset_fsb)
+	loff_t				offset)
 {
-	if (offset_fsb < wpc->imap.br_startoff ||
-	    offset_fsb >= wpc->imap.br_startoff + wpc->imap.br_blockcount)
+	if (offset < wpc->iomap.offset ||
+	    offset >= wpc->iomap.offset + wpc->iomap.length)
 		return false;
 	/*
 	 * If this is a COW mapping, it is sufficient to check that the mapping
 	 * covers the offset. Be careful to check this first because the caller
 	 * can revalidate a COW mapping without updating the data seqno.
 	 */
-	if (wpc->fork == XFS_COW_FORK)
+	if (wpc->iomap.flags & IOMAP_F_SHARED)
 		return true;
 
 	/*
@@ -443,17 +288,17 @@ xfs_imap_valid(
 	 * checked (and found nothing at this offset) could have added
 	 * overlapping blocks.
 	 */
-	if (wpc->data_seq != READ_ONCE(ip->i_df.if_seq))
+	if (XFS_WPC(wpc)->data_seq != READ_ONCE(ip->i_df.if_seq))
 		return false;
 	if (xfs_inode_has_cow_data(ip) &&
-	    wpc->cow_seq != READ_ONCE(ip->i_cowfp->if_seq))
+	    XFS_WPC(wpc)->cow_seq != READ_ONCE(ip->i_cowfp->if_seq))
 		return false;
 	return true;
 }
 
 /*
  * Pass in a dellalloc extent and convert it to real extents, return the real
- * extent that maps offset_fsb in wpc->imap.
+ * extent that maps offset_fsb in wpc->iomap.
  *
  * The current page is held locked so nothing could have removed the block
  * backing offset_fsb, although it could have moved from the COW to the data
@@ -461,32 +306,38 @@ xfs_imap_valid(
  */
 static int
 xfs_convert_blocks(
-	struct xfs_writepage_ctx *wpc,
+	struct iomap_writepage_ctx *wpc,
 	struct xfs_inode	*ip,
-	xfs_fileoff_t		offset_fsb)
+	int			whichfork,
+	loff_t			offset)
 {
 	int			error;
+	unsigned		*seq;
+
+	if (whichfork == XFS_COW_FORK)
+		seq = &XFS_WPC(wpc)->cow_seq;
+	else
+		seq = &XFS_WPC(wpc)->data_seq;
 
 	/*
-	 * Attempt to allocate whatever delalloc extent currently backs
-	 * offset_fsb and put the result into wpc->imap.  Allocate in a loop
-	 * because it may take several attempts to allocate real blocks for a
-	 * contiguous delalloc extent if free space is sufficiently fragmented.
+	 * Attempt to allocate whatever delalloc extent currently backs offset
+	 * and put the result into wpc->iomap.  Allocate in a loop because it
+	 * may take several attempts to allocate real blocks for a contiguous
+	 * delalloc extent if free space is sufficiently fragmented.
 	 */
 	do {
-		error = xfs_bmapi_convert_delalloc(ip, wpc->fork, offset_fsb,
-				&wpc->imap, wpc->fork == XFS_COW_FORK ?
-					&wpc->cow_seq : &wpc->data_seq);
+		error = xfs_bmapi_convert_delalloc(ip, whichfork, offset,
+				&wpc->iomap, seq);
 		if (error)
 			return error;
-	} while (wpc->imap.br_startoff + wpc->imap.br_blockcount <= offset_fsb);
+	} while (wpc->iomap.offset + wpc->iomap.length <= offset);
 
 	return 0;
 }
 
-STATIC int
+static int
 xfs_map_blocks(
-	struct xfs_writepage_ctx *wpc,
+	struct iomap_writepage_ctx *wpc,
 	struct inode		*inode,
 	loff_t			offset)
 {
@@ -496,6 +347,7 @@ xfs_map_blocks(
 	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	xfs_fileoff_t		end_fsb = XFS_B_TO_FSB(mp, offset + count);
 	xfs_fileoff_t		cow_fsb = NULLFILEOFF;
+	int			whichfork = XFS_DATA_FORK;
 	struct xfs_bmbt_irec	imap;
 	struct xfs_iext_cursor	icur;
 	int			retries = 0;
@@ -519,7 +371,7 @@ xfs_map_blocks(
 	 * against concurrent updates and provides a memory barrier on the way
 	 * out that ensures that we always see the current value.
 	 */
-	if (xfs_imap_valid(wpc, ip, offset_fsb))
+	if (xfs_imap_valid(wpc, ip, offset))
 		return 0;
 
 	/*
@@ -541,10 +393,10 @@ retry:
 	    xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &imap))
 		cow_fsb = imap.br_startoff;
 	if (cow_fsb != NULLFILEOFF && cow_fsb <= offset_fsb) {
-		wpc->cow_seq = READ_ONCE(ip->i_cowfp->if_seq);
+		XFS_WPC(wpc)->cow_seq = READ_ONCE(ip->i_cowfp->if_seq);
 		xfs_iunlock(ip, XFS_ILOCK_SHARED);
 
-		wpc->fork = XFS_COW_FORK;
+		whichfork = XFS_COW_FORK;
 		goto allocate_blocks;
 	}
 
@@ -552,7 +404,7 @@ retry:
 	 * No COW extent overlap. Revalidate now that we may have updated
 	 * ->cow_seq. If the data mapping is still valid, we're done.
 	 */
-	if (xfs_imap_valid(wpc, ip, offset_fsb)) {
+	if (xfs_imap_valid(wpc, ip, offset)) {
 		xfs_iunlock(ip, XFS_ILOCK_SHARED);
 		return 0;
 	}
@@ -564,11 +416,9 @@ retry:
 	 */
 	if (!xfs_iext_lookup_extent(ip, &ip->i_df, offset_fsb, &icur, &imap))
 		imap.br_startoff = end_fsb;	/* fake a hole past EOF */
-	wpc->data_seq = READ_ONCE(ip->i_df.if_seq);
+	XFS_WPC(wpc)->data_seq = READ_ONCE(ip->i_df.if_seq);
 	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 
-	wpc->fork = XFS_DATA_FORK;
-
 	/* landed in a hole or beyond EOF? */
 	if (imap.br_startoff > offset_fsb) {
 		imap.br_blockcount = imap.br_startoff - offset_fsb;
@@ -592,11 +442,11 @@ retry:
 	    isnullstartblock(imap.br_startblock))
 		goto allocate_blocks;
 
-	wpc->imap = imap;
-	trace_xfs_map_blocks_found(ip, offset, count, wpc->fork, &imap);
+	xfs_bmbt_to_iomap(ip, &wpc->iomap, &imap, 0);
+	trace_xfs_map_blocks_found(ip, offset, count, whichfork, &imap);
 	return 0;
 allocate_blocks:
-	error = xfs_convert_blocks(wpc, ip, offset_fsb);
+	error = xfs_convert_blocks(wpc, ip, whichfork, offset);
 	if (error) {
 		/*
 		 * If we failed to find the extent in the COW fork we might have
@@ -605,7 +455,7 @@ allocate_blocks:
 		 * the former case, but prevent additional retries to avoid
 		 * looping forever for the latter case.
 		 */
-		if (error == -EAGAIN && wpc->fork == XFS_COW_FORK && !retries++)
+		if (error == -EAGAIN && whichfork == XFS_COW_FORK && !retries++)
 			goto retry;
 		ASSERT(error != -EAGAIN);
 		return error;
@@ -616,34 +466,22 @@ allocate_blocks:
 	 * original delalloc one.  Trim the return extent to the next COW
 	 * boundary again to force a re-lookup.
 	 */
-	if (wpc->fork != XFS_COW_FORK && cow_fsb != NULLFILEOFF &&
-	    cow_fsb < wpc->imap.br_startoff + wpc->imap.br_blockcount)
-		wpc->imap.br_blockcount = cow_fsb - wpc->imap.br_startoff;
+	if (whichfork != XFS_COW_FORK && cow_fsb != NULLFILEOFF) {
+		loff_t		cow_offset = XFS_FSB_TO_B(mp, cow_fsb);
 
-	ASSERT(wpc->imap.br_startoff <= offset_fsb);
-	ASSERT(wpc->imap.br_startoff + wpc->imap.br_blockcount > offset_fsb);
-	trace_xfs_map_blocks_alloc(ip, offset, count, wpc->fork, &imap);
+		if (cow_offset < wpc->iomap.offset + wpc->iomap.length)
+			wpc->iomap.length = cow_offset - wpc->iomap.offset;
+	}
+
+	ASSERT(wpc->iomap.offset <= offset);
+	ASSERT(wpc->iomap.offset + wpc->iomap.length > offset);
+	trace_xfs_map_blocks_alloc(ip, offset, count, whichfork, &imap);
 	return 0;
 }
 
-/*
- * Submit the bio for an ioend. We are passed an ioend with a bio attached to
- * it, and we submit that bio. The ioend may be used for multiple bio
- * submissions, so we only want to allocate an append transaction for the ioend
- * once. In the case of multiple bio submission, each bio will take an IO
- * reference to the ioend to ensure that the ioend completion is only done once
- * all bios have been submitted and the ioend is really done.
- *
- * If @status is non-zero, it means that we have a situation where some part of
- * the submission process has failed after we have marked paged for writeback
- * and unlocked them. In this situation, we need to fail the bio and ioend
- * rather than submit it to IO. This typically only happens on a filesystem
- * shutdown.
- */
-STATIC int
-xfs_submit_ioend(
-	struct writeback_control *wbc,
-	struct xfs_ioend	*ioend,
+static int
+xfs_prepare_ioend(
+	struct iomap_ioend	*ioend,
 	int			status)
 {
 	unsigned int		nofs_flag;
@@ -656,157 +494,24 @@ xfs_submit_ioend(
 	nofs_flag = memalloc_nofs_save();
 
 	/* Convert CoW extents to regular */
-	if (!status && ioend->io_fork == XFS_COW_FORK) {
+	if (!status && (ioend->io_flags & IOMAP_F_SHARED)) {
 		status = xfs_reflink_convert_cow(XFS_I(ioend->io_inode),
 				ioend->io_offset, ioend->io_size);
 	}
 
 	/* Reserve log space if we might write beyond the on-disk inode size. */
 	if (!status &&
-	    (ioend->io_fork == XFS_COW_FORK ||
-	     ioend->io_state != XFS_EXT_UNWRITTEN) &&
+	    ((ioend->io_flags & IOMAP_F_SHARED) ||
+	     ioend->io_type != IOMAP_UNWRITTEN) &&
 	    xfs_ioend_is_append(ioend) &&
-	    !ioend->io_append_trans)
+	    !ioend->io_private)
 		status = xfs_setfilesize_trans_alloc(ioend);
 
 	memalloc_nofs_restore(nofs_flag);
 
-	ioend->io_bio->bi_private = ioend;
-	ioend->io_bio->bi_end_io = xfs_end_bio;
-
-	/*
-	 * If we are failing the IO now, just mark the ioend with an
-	 * error and finish it. This will run IO completion immediately
-	 * as there is only one reference to the ioend at this point in
-	 * time.
-	 */
-	if (status) {
-		ioend->io_bio->bi_status = errno_to_blk_status(status);
-		bio_endio(ioend->io_bio);
-		return status;
-	}
-
-	submit_bio(ioend->io_bio);
-	return 0;
-}
-
-static struct xfs_ioend *
-xfs_alloc_ioend(
-	struct inode		*inode,
-	int			fork,
-	xfs_exntst_t		state,
-	xfs_off_t		offset,
-	struct block_device	*bdev,
-	sector_t		sector,
-	struct writeback_control *wbc)
-{
-	struct xfs_ioend	*ioend;
-	struct bio		*bio;
-
-	bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, &xfs_ioend_bioset);
-	bio_set_dev(bio, bdev);
-	bio->bi_iter.bi_sector = sector;
-	bio->bi_opf = REQ_OP_WRITE | wbc_to_write_flags(wbc);
-	bio->bi_write_hint = inode->i_write_hint;
-	wbc_init_bio(wbc, bio);
-
-	ioend = container_of(bio, struct xfs_ioend, io_inline_bio);
-	INIT_LIST_HEAD(&ioend->io_list);
-	ioend->io_fork = fork;
-	ioend->io_state = state;
-	ioend->io_inode = inode;
-	ioend->io_size = 0;
-	ioend->io_offset = offset;
-	ioend->io_append_trans = NULL;
-	ioend->io_bio = bio;
-	return ioend;
-}
-
-/*
- * Allocate a new bio, and chain the old bio to the new one.
- *
- * Note that we have to do perform the chaining in this unintuitive order
- * so that the bi_private linkage is set up in the right direction for the
- * traversal in xfs_destroy_ioend().
- */
-static struct bio *
-xfs_chain_bio(
-	struct bio		*prev)
-{
-	struct bio *new;
-
-	new = bio_alloc(GFP_NOFS, BIO_MAX_PAGES);
-	bio_copy_dev(new, prev);/* also copies over blkcg information */
-	new->bi_iter.bi_sector = bio_end_sector(prev);
-	new->bi_opf = prev->bi_opf;
-	new->bi_write_hint = prev->bi_write_hint;
-
-	bio_chain(prev, new);
-	bio_get(prev);		/* for xfs_destroy_ioend */
-	submit_bio(prev);
-	return new;
-}
-
-/*
- * Test to see if we have an existing ioend structure that we could append to
- * first, otherwise finish off the current ioend and start another.
- */
-STATIC void
-xfs_add_to_ioend(
-	struct inode		*inode,
-	xfs_off_t		offset,
-	struct page		*page,
-	struct iomap_page	*iop,
-	struct xfs_writepage_ctx *wpc,
-	struct writeback_control *wbc,
-	struct list_head	*iolist)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-	struct block_device	*bdev = xfs_find_bdev_for_inode(inode);
-	unsigned		len = i_blocksize(inode);
-	unsigned		poff = offset & (PAGE_SIZE - 1);
-	bool			merged, same_page = false;
-	sector_t		sector;
-
-	sector = xfs_fsb_to_db(ip, wpc->imap.br_startblock) +
-		((offset - XFS_FSB_TO_B(mp, wpc->imap.br_startoff)) >> 9);
-
-	if (!wpc->ioend ||
-	    wpc->fork != wpc->ioend->io_fork ||
-	    wpc->imap.br_state != wpc->ioend->io_state ||
-	    sector != bio_end_sector(wpc->ioend->io_bio) ||
-	    offset != wpc->ioend->io_offset + wpc->ioend->io_size) {
-		if (wpc->ioend)
-			list_add(&wpc->ioend->io_list, iolist);
-		wpc->ioend = xfs_alloc_ioend(inode, wpc->fork,
-				wpc->imap.br_state, offset, bdev, sector, wbc);
-	}
-
-	merged = __bio_try_merge_page(wpc->ioend->io_bio, page, len, poff,
-			&same_page);
-
-	if (iop && !same_page)
-		atomic_inc(&iop->write_count);
-
-	if (!merged) {
-		if (bio_full(wpc->ioend->io_bio, len))
-			wpc->ioend->io_bio = xfs_chain_bio(wpc->ioend->io_bio);
-		bio_add_page(wpc->ioend->io_bio, page, len, poff);
-	}
-
-	wpc->ioend->io_size += len;
-	wbc_account_cgroup_owner(wbc, page, len);
-}
-
-STATIC void
-xfs_vm_invalidatepage(
-	struct page		*page,
-	unsigned int		offset,
-	unsigned int		length)
-{
-	trace_xfs_invalidatepage(page->mapping->host, page, offset, length);
-	iomap_invalidatepage(page, offset, length);
+	if (xfs_ioend_needs_workqueue(ioend))
+		ioend->io_bio->bi_end_io = xfs_end_bio;
+	return status;
 }
 
 /*
@@ -820,8 +525,8 @@ xfs_vm_invalidatepage(
  * transaction as there is no space left for block reservation (typically why we
  * see a ENOSPC in writeback).
  */
-STATIC void
-xfs_aops_discard_page(
+static void
+xfs_discard_page(
 	struct page		*page)
 {
 	struct inode		*inode = page->mapping->host;
@@ -834,7 +539,7 @@ xfs_aops_discard_page(
 	if (XFS_FORCED_SHUTDOWN(mp))
 		goto out_invalidate;
 
-	xfs_alert(mp,
+	xfs_alert_ratelimited(mp,
 		"page discard on page "PTR_FMT", inode 0x%llx, offset %llu.",
 			page, ip->i_ino, offset);
 
@@ -843,246 +548,14 @@ xfs_aops_discard_page(
 	if (error && !XFS_FORCED_SHUTDOWN(mp))
 		xfs_alert(mp, "page discard unable to remove delalloc mapping.");
 out_invalidate:
-	xfs_vm_invalidatepage(page, 0, PAGE_SIZE);
-}
-
-/*
- * We implement an immediate ioend submission policy here to avoid needing to
- * chain multiple ioends and hence nest mempool allocations which can violate
- * forward progress guarantees we need to provide. The current ioend we are
- * adding blocks to is cached on the writepage context, and if the new block
- * does not append to the cached ioend it will create a new ioend and cache that
- * instead.
- *
- * If a new ioend is created and cached, the old ioend is returned and queued
- * locally for submission once the entire page is processed or an error has been
- * detected.  While ioends are submitted immediately after they are completed,
- * batching optimisations are provided by higher level block plugging.
- *
- * At the end of a writeback pass, there will be a cached ioend remaining on the
- * writepage context that the caller will need to submit.
- */
-static int
-xfs_writepage_map(
-	struct xfs_writepage_ctx *wpc,
-	struct writeback_control *wbc,
-	struct inode		*inode,
-	struct page		*page,
-	uint64_t		end_offset)
-{
-	LIST_HEAD(submit_list);
-	struct iomap_page	*iop = to_iomap_page(page);
-	unsigned		len = i_blocksize(inode);
-	struct xfs_ioend	*ioend, *next;
-	uint64_t		file_offset;	/* file offset of page */
-	int			error = 0, count = 0, i;
-
-	ASSERT(iop || i_blocksize(inode) == PAGE_SIZE);
-	ASSERT(!iop || atomic_read(&iop->write_count) == 0);
-
-	/*
-	 * Walk through the page to find areas to write back. If we run off the
-	 * end of the current map or find the current map invalid, grab a new
-	 * one.
-	 */
-	for (i = 0, file_offset = page_offset(page);
-	     i < (PAGE_SIZE >> inode->i_blkbits) && file_offset < end_offset;
-	     i++, file_offset += len) {
-		if (iop && !test_bit(i, iop->uptodate))
-			continue;
-
-		error = xfs_map_blocks(wpc, inode, file_offset);
-		if (error)
-			break;
-		if (wpc->imap.br_startblock == HOLESTARTBLOCK)
-			continue;
-		xfs_add_to_ioend(inode, file_offset, page, iop, wpc, wbc,
-				 &submit_list);
-		count++;
-	}
-
-	ASSERT(wpc->ioend || list_empty(&submit_list));
-	ASSERT(PageLocked(page));
-	ASSERT(!PageWriteback(page));
-
-	/*
-	 * On error, we have to fail the ioend here because we may have set
-	 * pages under writeback, we have to make sure we run IO completion to
-	 * mark the error state of the IO appropriately, so we can't cancel the
-	 * ioend directly here.  That means we have to mark this page as under
-	 * writeback if we included any blocks from it in the ioend chain so
-	 * that completion treats it correctly.
-	 *
-	 * If we didn't include the page in the ioend, the on error we can
-	 * simply discard and unlock it as there are no other users of the page
-	 * now.  The caller will still need to trigger submission of outstanding
-	 * ioends on the writepage context so they are treated correctly on
-	 * error.
-	 */
-	if (unlikely(error)) {
-		if (!count) {
-			xfs_aops_discard_page(page);
-			ClearPageUptodate(page);
-			unlock_page(page);
-			goto done;
-		}
-
-		/*
-		 * If the page was not fully cleaned, we need to ensure that the
-		 * higher layers come back to it correctly.  That means we need
-		 * to keep the page dirty, and for WB_SYNC_ALL writeback we need
-		 * to ensure the PAGECACHE_TAG_TOWRITE index mark is not removed
-		 * so another attempt to write this page in this writeback sweep
-		 * will be made.
-		 */
-		set_page_writeback_keepwrite(page);
-	} else {
-		clear_page_dirty_for_io(page);
-		set_page_writeback(page);
-	}
-
-	unlock_page(page);
-
-	/*
-	 * Preserve the original error if there was one, otherwise catch
-	 * submission errors here and propagate into subsequent ioend
-	 * submissions.
-	 */
-	list_for_each_entry_safe(ioend, next, &submit_list, io_list) {
-		int error2;
-
-		list_del_init(&ioend->io_list);
-		error2 = xfs_submit_ioend(wbc, ioend, error);
-		if (error2 && !error)
-			error = error2;
-	}
-
-	/*
-	 * We can end up here with no error and nothing to write only if we race
-	 * with a partial page truncate on a sub-page block sized filesystem.
-	 */
-	if (!count)
-		end_page_writeback(page);
-done:
-	mapping_set_error(page->mapping, error);
-	return error;
+	iomap_invalidatepage(page, 0, PAGE_SIZE);
 }
 
-/*
- * Write out a dirty page.
- *
- * For delalloc space on the page we need to allocate space and flush it.
- * For unwritten space on the page we need to start the conversion to
- * regular allocated space.
- */
-STATIC int
-xfs_do_writepage(
-	struct page		*page,
-	struct writeback_control *wbc,
-	void			*data)
-{
-	struct xfs_writepage_ctx *wpc = data;
-	struct inode		*inode = page->mapping->host;
-	loff_t			offset;
-	uint64_t              end_offset;
-	pgoff_t                 end_index;
-
-	trace_xfs_writepage(inode, page, 0, 0);
-
-	/*
-	 * Refuse to write the page out if we are called from reclaim context.
-	 *
-	 * This avoids stack overflows when called from deeply used stacks in
-	 * random callers for direct reclaim or memcg reclaim.  We explicitly
-	 * allow reclaim from kswapd as the stack usage there is relatively low.
-	 *
-	 * This should never happen except in the case of a VM regression so
-	 * warn about it.
-	 */
-	if (WARN_ON_ONCE((current->flags & (PF_MEMALLOC|PF_KSWAPD)) ==
-			PF_MEMALLOC))
-		goto redirty;
-
-	/*
-	 * Given that we do not allow direct reclaim to call us, we should
-	 * never be called while in a filesystem transaction.
-	 */
-	if (WARN_ON_ONCE(current->flags & PF_MEMALLOC_NOFS))
-		goto redirty;
-
-	/*
-	 * Is this page beyond the end of the file?
-	 *
-	 * The page index is less than the end_index, adjust the end_offset
-	 * to the highest offset that this page should represent.
-	 * -----------------------------------------------------
-	 * |			file mapping	       | <EOF> |
-	 * -----------------------------------------------------
-	 * | Page ... | Page N-2 | Page N-1 |  Page N  |       |
-	 * ^--------------------------------^----------|--------
-	 * |     desired writeback range    |      see else    |
-	 * ---------------------------------^------------------|
-	 */
-	offset = i_size_read(inode);
-	end_index = offset >> PAGE_SHIFT;
-	if (page->index < end_index)
-		end_offset = (xfs_off_t)(page->index + 1) << PAGE_SHIFT;
-	else {
-		/*
-		 * Check whether the page to write out is beyond or straddles
-		 * i_size or not.
-		 * -------------------------------------------------------
-		 * |		file mapping		        | <EOF>  |
-		 * -------------------------------------------------------
-		 * | Page ... | Page N-2 | Page N-1 |  Page N   | Beyond |
-		 * ^--------------------------------^-----------|---------
-		 * |				    |      Straddles     |
-		 * ---------------------------------^-----------|--------|
-		 */
-		unsigned offset_into_page = offset & (PAGE_SIZE - 1);
-
-		/*
-		 * Skip the page if it is fully outside i_size, e.g. due to a
-		 * truncate operation that is in progress. We must redirty the
-		 * page so that reclaim stops reclaiming it. Otherwise
-		 * xfs_vm_releasepage() is called on it and gets confused.
-		 *
-		 * Note that the end_index is unsigned long, it would overflow
-		 * if the given offset is greater than 16TB on 32-bit system
-		 * and if we do check the page is fully outside i_size or not
-		 * via "if (page->index >= end_index + 1)" as "end_index + 1"
-		 * will be evaluated to 0.  Hence this page will be redirtied
-		 * and be written out repeatedly which would result in an
-		 * infinite loop, the user program that perform this operation
-		 * will hang.  Instead, we can verify this situation by checking
-		 * if the page to write is totally beyond the i_size or if it's
-		 * offset is just equal to the EOF.
-		 */
-		if (page->index > end_index ||
-		    (page->index == end_index && offset_into_page == 0))
-			goto redirty;
-
-		/*
-		 * The page straddles i_size.  It must be zeroed out on each
-		 * and every writepage invocation because it may be mmapped.
-		 * "A file is mapped in multiples of the page size.  For a file
-		 * that is not a multiple of the page size, the remaining
-		 * memory is zeroed when mapped, and writes to that region are
-		 * not written out to the file."
-		 */
-		zero_user_segment(page, offset_into_page, PAGE_SIZE);
-
-		/* Adjust the end_offset to the end of file */
-		end_offset = offset;
-	}
-
-	return xfs_writepage_map(wpc, wbc, inode, page, end_offset);
-
-redirty:
-	redirty_page_for_writepage(wbc, page);
-	unlock_page(page);
-	return 0;
-}
+static const struct iomap_writeback_ops xfs_writeback_ops = {
+	.map_blocks		= xfs_map_blocks,
+	.prepare_ioend		= xfs_prepare_ioend,
+	.discard_page		= xfs_discard_page,
+};
 
 STATIC int
 xfs_vm_writepage(
@@ -1090,12 +563,8 @@ xfs_vm_writepage(
 	struct writeback_control *wbc)
 {
 	struct xfs_writepage_ctx wpc = { };
-	int			ret;
 
-	ret = xfs_do_writepage(page, wbc, &wpc);
-	if (wpc.ioend)
-		ret = xfs_submit_ioend(wbc, wpc.ioend, ret);
-	return ret;
+	return iomap_writepage(page, wbc, &wpc.ctx, &xfs_writeback_ops);
 }
 
 STATIC int
@@ -1104,13 +573,9 @@ xfs_vm_writepages(
 	struct writeback_control *wbc)
 {
 	struct xfs_writepage_ctx wpc = { };
-	int			ret;
 
 	xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED);
-	ret = write_cache_pages(mapping, wbc, xfs_do_writepage, &wpc);
-	if (wpc.ioend)
-		ret = xfs_submit_ioend(wbc, wpc.ioend, ret);
-	return ret;
+	return iomap_writepages(mapping, wbc, &wpc.ctx, &xfs_writeback_ops);
 }
 
 STATIC int
@@ -1118,18 +583,11 @@ xfs_dax_writepages(
 	struct address_space	*mapping,
 	struct writeback_control *wbc)
 {
-	xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED);
-	return dax_writeback_mapping_range(mapping,
-			xfs_find_bdev_for_inode(mapping->host), wbc);
-}
+	struct xfs_inode	*ip = XFS_I(mapping->host);
 
-STATIC int
-xfs_vm_releasepage(
-	struct page		*page,
-	gfp_t			gfp_mask)
-{
-	trace_xfs_releasepage(page->mapping->host, page, 0, 0);
-	return iomap_releasepage(page, gfp_mask);
+	xfs_iflags_clear(ip, XFS_ITRUNCATED);
+	return dax_writeback_mapping_range(mapping,
+			xfs_inode_buftarg(ip)->bt_daxdev, wbc);
 }
 
 STATIC sector_t
@@ -1152,7 +610,7 @@ xfs_vm_bmap(
 	 */
 	if (xfs_is_cow_inode(ip) || XFS_IS_REALTIME_INODE(ip))
 		return 0;
-	return iomap_bmap(mapping, block, &xfs_iomap_ops);
+	return iomap_bmap(mapping, block, &xfs_read_iomap_ops);
 }
 
 STATIC int
@@ -1160,19 +618,14 @@ xfs_vm_readpage(
 	struct file		*unused,
 	struct page		*page)
 {
-	trace_xfs_vm_readpage(page->mapping->host, 1);
-	return iomap_readpage(page, &xfs_iomap_ops);
+	return iomap_readpage(page, &xfs_read_iomap_ops);
 }
 
-STATIC int
-xfs_vm_readpages(
-	struct file		*unused,
-	struct address_space	*mapping,
-	struct list_head	*pages,
-	unsigned		nr_pages)
+STATIC void
+xfs_vm_readahead(
+	struct readahead_control	*rac)
 {
-	trace_xfs_vm_readpages(mapping->host, nr_pages);
-	return iomap_readpages(mapping, pages, nr_pages, &xfs_iomap_ops);
+	iomap_readahead(rac, &xfs_read_iomap_ops);
 }
 
 static int
@@ -1181,18 +634,19 @@ xfs_iomap_swapfile_activate(
 	struct file			*swap_file,
 	sector_t			*span)
 {
-	sis->bdev = xfs_find_bdev_for_inode(file_inode(swap_file));
-	return iomap_swapfile_activate(sis, swap_file, span, &xfs_iomap_ops);
+	sis->bdev = xfs_inode_buftarg(XFS_I(file_inode(swap_file)))->bt_bdev;
+	return iomap_swapfile_activate(sis, swap_file, span,
+			&xfs_read_iomap_ops);
 }
 
 const struct address_space_operations xfs_address_space_operations = {
 	.readpage		= xfs_vm_readpage,
-	.readpages		= xfs_vm_readpages,
+	.readahead		= xfs_vm_readahead,
 	.writepage		= xfs_vm_writepage,
 	.writepages		= xfs_vm_writepages,
 	.set_page_dirty		= iomap_set_page_dirty,
-	.releasepage		= xfs_vm_releasepage,
-	.invalidatepage		= xfs_vm_invalidatepage,
+	.releasepage		= iomap_releasepage,
+	.invalidatepage		= iomap_invalidatepage,
 	.bmap			= xfs_vm_bmap,
 	.direct_IO		= noop_direct_IO,
 	.migratepage		= iomap_migrate_page,
diff --git a/fs/xfs/xfs_aops.h b/fs/xfs/xfs_aops.h
index 45a1ea240cbb..e0bd68419764 100644
--- a/fs/xfs/xfs_aops.h
+++ b/fs/xfs/xfs_aops.h
@@ -6,29 +6,9 @@
 #ifndef __XFS_AOPS_H__
 #define __XFS_AOPS_H__
 
-extern struct bio_set xfs_ioend_bioset;
-
-/*
- * Structure for buffered I/O completions.
- */
-struct xfs_ioend {
-	struct list_head	io_list;	/* next ioend in chain */
-	int			io_fork;	/* inode fork written back */
-	xfs_exntst_t		io_state;	/* extent state */
-	struct inode		*io_inode;	/* file being written to */
-	size_t			io_size;	/* size of the extent */
-	xfs_off_t		io_offset;	/* offset in the file */
-	struct xfs_trans	*io_append_trans;/* xact. for size update */
-	struct bio		*io_bio;	/* bio being built */
-	struct bio		io_inline_bio;	/* MUST BE LAST! */
-};
-
 extern const struct address_space_operations xfs_address_space_operations;
 extern const struct address_space_operations xfs_dax_aops;
 
 int	xfs_setfilesize(struct xfs_inode *ip, xfs_off_t offset, size_t size);
 
-extern struct block_device *xfs_find_bdev_for_inode(struct inode *);
-extern struct dax_device *xfs_find_daxdev_for_inode(struct inode *);
-
 #endif /* __XFS_AOPS_H__ */
diff --git a/fs/xfs/xfs_attr_inactive.c b/fs/xfs/xfs_attr_inactive.c
index a640a285cc52..c42f90e16b4f 100644
--- a/fs/xfs/xfs_attr_inactive.c
+++ b/fs/xfs/xfs_attr_inactive.c
@@ -22,24 +22,21 @@
 #include "xfs_attr_leaf.h"
 #include "xfs_quota.h"
 #include "xfs_dir2.h"
+#include "xfs_error.h"
 
 /*
- * Look at all the extents for this logical region,
- * invalidate any buffers that are incore/in transactions.
+ * Invalidate any incore buffers associated with this remote attribute value
+ * extent.   We never log remote attribute value buffers, which means that they
+ * won't be attached to a transaction and are therefore safe to mark stale.
+ * The actual bunmapi will be taken care of later.
  */
 STATIC int
-xfs_attr3_leaf_freextent(
-	struct xfs_trans	**trans,
+xfs_attr3_rmt_stale(
 	struct xfs_inode	*dp,
 	xfs_dablk_t		blkno,
 	int			blkcnt)
 {
 	struct xfs_bmbt_irec	map;
-	struct xfs_buf		*bp;
-	xfs_dablk_t		tblkno;
-	xfs_daddr_t		dblkno;
-	int			tblkcnt;
-	int			dblkcnt;
 	int			nmap;
 	int			error;
 
@@ -47,47 +44,29 @@ xfs_attr3_leaf_freextent(
 	 * Roll through the "value", invalidating the attribute value's
 	 * blocks.
 	 */
-	tblkno = blkno;
-	tblkcnt = blkcnt;
-	while (tblkcnt > 0) {
+	while (blkcnt > 0) {
 		/*
 		 * Try to remember where we decided to put the value.
 		 */
 		nmap = 1;
-		error = xfs_bmapi_read(dp, (xfs_fileoff_t)tblkno, tblkcnt,
+		error = xfs_bmapi_read(dp, (xfs_fileoff_t)blkno, blkcnt,
 				       &map, &nmap, XFS_BMAPI_ATTRFORK);
-		if (error) {
+		if (error)
 			return error;
-		}
-		ASSERT(nmap == 1);
-		ASSERT(map.br_startblock != DELAYSTARTBLOCK);
+		if (XFS_IS_CORRUPT(dp->i_mount, nmap != 1))
+			return -EFSCORRUPTED;
 
 		/*
-		 * If it's a hole, these are already unmapped
-		 * so there's nothing to invalidate.
+		 * Mark any incore buffers for the remote value as stale.  We
+		 * never log remote attr value buffers, so the buffer should be
+		 * easy to kill.
 		 */
-		if (map.br_startblock != HOLESTARTBLOCK) {
-
-			dblkno = XFS_FSB_TO_DADDR(dp->i_mount,
-						  map.br_startblock);
-			dblkcnt = XFS_FSB_TO_BB(dp->i_mount,
-						map.br_blockcount);
-			bp = xfs_trans_get_buf(*trans,
-					dp->i_mount->m_ddev_targp,
-					dblkno, dblkcnt, 0);
-			if (!bp)
-				return -ENOMEM;
-			xfs_trans_binval(*trans, bp);
-			/*
-			 * Roll to next transaction.
-			 */
-			error = xfs_trans_roll_inode(trans, dp);
-			if (error)
-				return error;
-		}
+		error = xfs_attr_rmtval_stale(dp, &map, 0);
+		if (error)
+			return error;
 
-		tblkno += map.br_blockcount;
-		tblkcnt -= map.br_blockcount;
+		blkno += map.br_blockcount;
+		blkcnt -= map.br_blockcount;
 	}
 
 	return 0;
@@ -101,86 +80,45 @@ xfs_attr3_leaf_freextent(
  */
 STATIC int
 xfs_attr3_leaf_inactive(
-	struct xfs_trans	**trans,
-	struct xfs_inode	*dp,
-	struct xfs_buf		*bp)
+	struct xfs_trans		**trans,
+	struct xfs_inode		*dp,
+	struct xfs_buf			*bp)
 {
-	struct xfs_attr_leafblock *leaf;
-	struct xfs_attr3_icleaf_hdr ichdr;
-	struct xfs_attr_leaf_entry *entry;
+	struct xfs_attr3_icleaf_hdr	ichdr;
+	struct xfs_mount		*mp = bp->b_mount;
+	struct xfs_attr_leafblock	*leaf = bp->b_addr;
+	struct xfs_attr_leaf_entry	*entry;
 	struct xfs_attr_leaf_name_remote *name_rmt;
-	struct xfs_attr_inactive_list *list;
-	struct xfs_attr_inactive_list *lp;
-	int			error;
-	int			count;
-	int			size;
-	int			tmp;
-	int			i;
-	struct xfs_mount	*mp = bp->b_mount;
+	int				error = 0;
+	int				i;
 
-	leaf = bp->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &ichdr, leaf);
 
 	/*
-	 * Count the number of "remote" value extents.
+	 * Find the remote value extents for this leaf and invalidate their
+	 * incore buffers.
 	 */
-	count = 0;
 	entry = xfs_attr3_leaf_entryp(leaf);
 	for (i = 0; i < ichdr.count; entry++, i++) {
-		if (be16_to_cpu(entry->nameidx) &&
-		    ((entry->flags & XFS_ATTR_LOCAL) == 0)) {
-			name_rmt = xfs_attr3_leaf_name_remote(leaf, i);
-			if (name_rmt->valueblk)
-				count++;
-		}
-	}
-
-	/*
-	 * If there are no "remote" values, we're done.
-	 */
-	if (count == 0) {
-		xfs_trans_brelse(*trans, bp);
-		return 0;
-	}
-
-	/*
-	 * Allocate storage for a list of all the "remote" value extents.
-	 */
-	size = count * sizeof(xfs_attr_inactive_list_t);
-	list = kmem_alloc(size, 0);
+		int		blkcnt;
 
-	/*
-	 * Identify each of the "remote" value extents.
-	 */
-	lp = list;
-	entry = xfs_attr3_leaf_entryp(leaf);
-	for (i = 0; i < ichdr.count; entry++, i++) {
-		if (be16_to_cpu(entry->nameidx) &&
-		    ((entry->flags & XFS_ATTR_LOCAL) == 0)) {
-			name_rmt = xfs_attr3_leaf_name_remote(leaf, i);
-			if (name_rmt->valueblk) {
-				lp->valueblk = be32_to_cpu(name_rmt->valueblk);
-				lp->valuelen = xfs_attr3_rmt_blocks(dp->i_mount,
-						    be32_to_cpu(name_rmt->valuelen));
-				lp++;
-			}
-		}
-	}
-	xfs_trans_brelse(*trans, bp);	/* unlock for trans. in freextent() */
+		if (!entry->nameidx || (entry->flags & XFS_ATTR_LOCAL))
+			continue;
 
-	/*
-	 * Invalidate each of the "remote" value extents.
-	 */
-	error = 0;
-	for (lp = list, i = 0; i < count; i++, lp++) {
-		tmp = xfs_attr3_leaf_freextent(trans, dp,
-				lp->valueblk, lp->valuelen);
+		name_rmt = xfs_attr3_leaf_name_remote(leaf, i);
+		if (!name_rmt->valueblk)
+			continue;
 
-		if (error == 0)
-			error = tmp;	/* save only the 1st errno */
+		blkcnt = xfs_attr3_rmt_blocks(dp->i_mount,
+				be32_to_cpu(name_rmt->valuelen));
+		error = xfs_attr3_rmt_stale(dp,
+				be32_to_cpu(name_rmt->valueblk), blkcnt);
+		if (error)
+			goto err;
 	}
 
-	kmem_free(list);
+	xfs_trans_brelse(*trans, bp);
+err:
 	return error;
 }
 
@@ -190,37 +128,35 @@ xfs_attr3_leaf_inactive(
  */
 STATIC int
 xfs_attr3_node_inactive(
-	struct xfs_trans **trans,
-	struct xfs_inode *dp,
-	struct xfs_buf	*bp,
-	int		level)
+	struct xfs_trans	**trans,
+	struct xfs_inode	*dp,
+	struct xfs_buf		*bp,
+	int			level)
 {
-	xfs_da_blkinfo_t *info;
-	xfs_da_intnode_t *node;
-	xfs_dablk_t child_fsb;
-	xfs_daddr_t parent_blkno, child_blkno;
-	int error, i;
-	struct xfs_buf *child_bp;
-	struct xfs_da_node_entry *btree;
+	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_da_blkinfo	*info;
+	xfs_dablk_t		child_fsb;
+	xfs_daddr_t		parent_blkno, child_blkno;
+	struct xfs_buf		*child_bp;
 	struct xfs_da3_icnode_hdr ichdr;
+	int			error, i;
 
 	/*
 	 * Since this code is recursive (gasp!) we must protect ourselves.
 	 */
 	if (level > XFS_DA_NODE_MAXDEPTH) {
+		xfs_buf_mark_corrupt(bp);
 		xfs_trans_brelse(*trans, bp);	/* no locks for later trans */
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
-	node = bp->b_addr;
-	dp->d_ops->node_hdr_from_disk(&ichdr, node);
+	xfs_da3_node_hdr_from_disk(dp->i_mount, &ichdr, bp->b_addr);
 	parent_blkno = bp->b_bn;
 	if (!ichdr.count) {
 		xfs_trans_brelse(*trans, bp);
 		return 0;
 	}
-	btree = dp->d_ops->node_tree_p(node);
-	child_fsb = be32_to_cpu(btree[0].before);
+	child_fsb = be32_to_cpu(ichdr.btree[0].before);
 	xfs_trans_brelse(*trans, bp);	/* no locks for later trans */
 
 	/*
@@ -235,7 +171,7 @@ xfs_attr3_node_inactive(
 		 * traversal of the tree so we may deal with many blocks
 		 * before we come back to this one.
 		 */
-		error = xfs_da3_node_read(*trans, dp, child_fsb, -1, &child_bp,
+		error = xfs_da3_node_read(*trans, dp, child_fsb, &child_bp,
 					  XFS_ATTR_FORK);
 		if (error)
 			return error;
@@ -258,8 +194,9 @@ xfs_attr3_node_inactive(
 			error = xfs_attr3_leaf_inactive(trans, dp, child_bp);
 			break;
 		default:
-			error = -EIO;
+			xfs_buf_mark_corrupt(child_bp);
 			xfs_trans_brelse(*trans, child_bp);
+			error = -EFSCORRUPTED;
 			break;
 		}
 		if (error)
@@ -268,10 +205,17 @@ xfs_attr3_node_inactive(
 		/*
 		 * Remove the subsidiary block from the cache and from the log.
 		 */
-		error = xfs_da_get_buf(*trans, dp, 0, child_blkno, &child_bp,
-				       XFS_ATTR_FORK);
+		error = xfs_trans_get_buf(*trans, mp->m_ddev_targp,
+				child_blkno,
+				XFS_FSB_TO_BB(mp, mp->m_attr_geo->fsbcount), 0,
+				&child_bp);
 		if (error)
 			return error;
+		error = bp->b_error;
+		if (error) {
+			xfs_trans_brelse(*trans, child_bp);
+			return error;
+		}
 		xfs_trans_binval(*trans, child_bp);
 
 		/*
@@ -279,13 +223,15 @@ xfs_attr3_node_inactive(
 		 * child block number.
 		 */
 		if (i + 1 < ichdr.count) {
-			error = xfs_da3_node_read(*trans, dp, 0, parent_blkno,
-						 &bp, XFS_ATTR_FORK);
+			struct xfs_da3_icnode_hdr phdr;
+
+			error = xfs_da3_node_read_mapped(*trans, dp,
+					parent_blkno, &bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
-			node = bp->b_addr;
-			btree = dp->d_ops->node_tree_p(node);
-			child_fsb = be32_to_cpu(btree[i + 1].before);
+			xfs_da3_node_hdr_from_disk(dp->i_mount, &phdr,
+						  bp->b_addr);
+			child_fsb = be32_to_cpu(phdr.btree[i + 1].before);
 			xfs_trans_brelse(*trans, bp);
 		}
 		/*
@@ -310,6 +256,7 @@ xfs_attr3_root_inactive(
 	struct xfs_trans	**trans,
 	struct xfs_inode	*dp)
 {
+	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_da_blkinfo	*info;
 	struct xfs_buf		*bp;
 	xfs_daddr_t		blkno;
@@ -321,7 +268,7 @@ xfs_attr3_root_inactive(
 	 * the extents in reverse order the extent containing
 	 * block 0 must still be there.
 	 */
-	error = xfs_da3_node_read(*trans, dp, 0, -1, &bp, XFS_ATTR_FORK);
+	error = xfs_da3_node_read(*trans, dp, 0, &bp, XFS_ATTR_FORK);
 	if (error)
 		return error;
 	blkno = bp->b_bn;
@@ -341,7 +288,8 @@ xfs_attr3_root_inactive(
 		error = xfs_attr3_leaf_inactive(trans, dp, bp);
 		break;
 	default:
-		error = -EIO;
+		error = -EFSCORRUPTED;
+		xfs_buf_mark_corrupt(bp);
 		xfs_trans_brelse(*trans, bp);
 		break;
 	}
@@ -351,9 +299,15 @@ xfs_attr3_root_inactive(
 	/*
 	 * Invalidate the incore copy of the root block.
 	 */
-	error = xfs_da_get_buf(*trans, dp, 0, blkno, &bp, XFS_ATTR_FORK);
+	error = xfs_trans_get_buf(*trans, mp->m_ddev_targp, blkno,
+			XFS_FSB_TO_BB(mp, mp->m_attr_geo->fsbcount), 0, &bp);
 	if (error)
 		return error;
+	error = bp->b_error;
+	if (error) {
+		xfs_trans_brelse(*trans, bp);
+		return error;
+	}
 	xfs_trans_binval(*trans, bp);	/* remove from cache */
 	/*
 	 * Commit the invalidate and start the next transaction.
diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c
index 00758fdc2fec..5ff1d929d3b5 100644
--- a/fs/xfs/xfs_attr_list.c
+++ b/fs/xfs/xfs_attr_list.c
@@ -49,25 +49,22 @@ xfs_attr_shortform_compare(const void *a, const void *b)
  * we can begin returning them to the user.
  */
 static int
-xfs_attr_shortform_list(xfs_attr_list_context_t *context)
+xfs_attr_shortform_list(
+	struct xfs_attr_list_context	*context)
 {
-	attrlist_cursor_kern_t *cursor;
-	xfs_attr_sf_sort_t *sbuf, *sbp;
-	xfs_attr_shortform_t *sf;
-	xfs_attr_sf_entry_t *sfe;
-	xfs_inode_t *dp;
-	int sbsize, nsbuf, count, i;
-
-	ASSERT(context != NULL);
-	dp = context->dp;
-	ASSERT(dp != NULL);
+	struct xfs_attrlist_cursor_kern	*cursor = &context->cursor;
+	struct xfs_inode		*dp = context->dp;
+	struct xfs_attr_sf_sort		*sbuf, *sbp;
+	struct xfs_attr_shortform	*sf;
+	struct xfs_attr_sf_entry	*sfe;
+	int				sbsize, nsbuf, count, i;
+	int				error = 0;
+
 	ASSERT(dp->i_afp != NULL);
 	sf = (xfs_attr_shortform_t *)dp->i_afp->if_u1.if_data;
 	ASSERT(sf != NULL);
 	if (!sf->hdr.count)
 		return 0;
-	cursor = context->cursor;
-	ASSERT(cursor != NULL);
 
 	trace_xfs_attr_list_sf(context);
 
@@ -84,6 +81,10 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	    (XFS_ISRESET_CURSOR(cursor) &&
 	     (dp->i_afp->if_bytes + sf->hdr.count * 16) < context->bufsize)) {
 		for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) {
+			if (XFS_IS_CORRUPT(context->dp->i_mount,
+					   !xfs_attr_namecheck(sfe->nameval,
+							       sfe->namelen)))
+				return -EFSCORRUPTED;
 			context->put_listent(context,
 					     sfe->flags,
 					     sfe->nameval,
@@ -161,10 +162,8 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			break;
 		}
 	}
-	if (i == nsbuf) {
-		kmem_free(sbuf);
-		return 0;
-	}
+	if (i == nsbuf)
+		goto out;
 
 	/*
 	 * Loop putting entries into the user buffer.
@@ -174,6 +173,12 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			cursor->hashval = sbp->hash;
 			cursor->offset = 0;
 		}
+		if (XFS_IS_CORRUPT(context->dp->i_mount,
+				   !xfs_attr_namecheck(sbp->name,
+						       sbp->namelen))) {
+			error = -EFSCORRUPTED;
+			goto out;
+		}
 		context->put_listent(context,
 				     sbp->flags,
 				     sbp->name,
@@ -183,9 +188,9 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			break;
 		cursor->offset++;
 	}
-
+out:
 	kmem_free(sbuf);
-	return 0;
+	return error;
 }
 
 /*
@@ -195,7 +200,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 STATIC int
 xfs_attr_node_list_lookup(
 	struct xfs_attr_list_context	*context,
-	struct attrlist_cursor_kern	*cursor,
+	struct xfs_attrlist_cursor_kern	*cursor,
 	struct xfs_buf			**pbp)
 {
 	struct xfs_da3_icnode_hdr	nodehdr;
@@ -213,7 +218,7 @@ xfs_attr_node_list_lookup(
 	ASSERT(*pbp == NULL);
 	cursor->blkno = 0;
 	for (;;) {
-		error = xfs_da3_node_read(tp, dp, cursor->blkno, -1, &bp,
+		error = xfs_da3_node_read(tp, dp, cursor->blkno, &bp,
 				XFS_ATTR_FORK);
 		if (error)
 			return error;
@@ -229,7 +234,7 @@ xfs_attr_node_list_lookup(
 			goto out_corruptbuf;
 		}
 
-		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+		xfs_da3_node_hdr_from_disk(mp, &nodehdr, node);
 
 		/* Tree taller than we can handle; bail out! */
 		if (nodehdr.level >= XFS_DA_NODE_MAXDEPTH)
@@ -243,7 +248,7 @@ xfs_attr_node_list_lookup(
 		else
 			expected_level--;
 
-		btree = dp->d_ops->node_tree_p(node);
+		btree = nodehdr.btree;
 		for (i = 0; i < nodehdr.count; btree++, i++) {
 			if (cursor->hashval <= be32_to_cpu(btree->hashval)) {
 				cursor->blkno = be32_to_cpu(btree->before);
@@ -258,7 +263,7 @@ xfs_attr_node_list_lookup(
 			return 0;
 
 		/* We can't point back to the root. */
-		if (cursor->blkno == 0)
+		if (XFS_IS_CORRUPT(mp, cursor->blkno == 0))
 			return -EFSCORRUPTED;
 	}
 
@@ -269,6 +274,7 @@ xfs_attr_node_list_lookup(
 	return 0;
 
 out_corruptbuf:
+	xfs_buf_mark_corrupt(bp);
 	xfs_trans_brelse(tp, bp);
 	return -EFSCORRUPTED;
 }
@@ -277,18 +283,17 @@ STATIC int
 xfs_attr_node_list(
 	struct xfs_attr_list_context	*context)
 {
+	struct xfs_attrlist_cursor_kern	*cursor = &context->cursor;
 	struct xfs_attr3_icleaf_hdr	leafhdr;
-	struct attrlist_cursor_kern	*cursor;
 	struct xfs_attr_leafblock	*leaf;
 	struct xfs_da_intnode		*node;
 	struct xfs_buf			*bp;
 	struct xfs_inode		*dp = context->dp;
 	struct xfs_mount		*mp = dp->i_mount;
-	int				error;
+	int				error = 0;
 
 	trace_xfs_attr_node_list(context);
 
-	cursor = context->cursor;
 	cursor->initted = 1;
 
 	/*
@@ -298,8 +303,8 @@ xfs_attr_node_list(
 	 */
 	bp = NULL;
 	if (cursor->blkno > 0) {
-		error = xfs_da3_node_read(context->tp, dp, cursor->blkno, -1,
-					      &bp, XFS_ATTR_FORK);
+		error = xfs_da3_node_read(context->tp, dp, cursor->blkno, &bp,
+				XFS_ATTR_FORK);
 		if ((error != 0) && (error != -EFSCORRUPTED))
 			return error;
 		if (bp) {
@@ -358,29 +363,32 @@ xfs_attr_node_list(
 	 */
 	for (;;) {
 		leaf = bp->b_addr;
-		xfs_attr3_leaf_list_int(bp, context);
+		error = xfs_attr3_leaf_list_int(bp, context);
+		if (error)
+			break;
 		xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &leafhdr, leaf);
 		if (context->seen_enough || leafhdr.forw == 0)
 			break;
 		cursor->blkno = leafhdr.forw;
 		xfs_trans_brelse(context->tp, bp);
-		error = xfs_attr3_leaf_read(context->tp, dp, cursor->blkno, -1, &bp);
+		error = xfs_attr3_leaf_read(context->tp, dp, cursor->blkno,
+					    &bp);
 		if (error)
 			return error;
 	}
 	xfs_trans_brelse(context->tp, bp);
-	return 0;
+	return error;
 }
 
 /*
  * Copy out attribute list entries for attr_list(), for leaf attribute lists.
  */
-void
+int
 xfs_attr3_leaf_list_int(
 	struct xfs_buf			*bp,
 	struct xfs_attr_list_context	*context)
 {
-	struct attrlist_cursor_kern	*cursor;
+	struct xfs_attrlist_cursor_kern	*cursor = &context->cursor;
 	struct xfs_attr_leafblock	*leaf;
 	struct xfs_attr3_icleaf_hdr	ichdr;
 	struct xfs_attr_leaf_entry	*entries;
@@ -394,7 +402,6 @@ xfs_attr3_leaf_list_int(
 	xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &ichdr, leaf);
 	entries = xfs_attr3_leaf_entryp(leaf);
 
-	cursor = context->cursor;
 	cursor->initted = 1;
 
 	/*
@@ -417,7 +424,7 @@ xfs_attr3_leaf_list_int(
 		}
 		if (i == ichdr.count) {
 			trace_xfs_attr_list_notfound(context);
-			return;
+			return 0;
 		}
 	} else {
 		entry = &entries[0];
@@ -438,8 +445,8 @@ xfs_attr3_leaf_list_int(
 		}
 
 		if ((entry->flags & XFS_ATTR_INCOMPLETE) &&
-		    !(context->flags & ATTR_INCOMPLETE))
-			continue;		/* skip incomplete entries */
+		    !context->allow_incomplete)
+			continue;
 
 		if (entry->flags & XFS_ATTR_LOCAL) {
 			xfs_attr_leaf_name_local_t *name_loc;
@@ -457,6 +464,9 @@ xfs_attr3_leaf_list_int(
 			valuelen = be32_to_cpu(name_rmt->valuelen);
 		}
 
+		if (XFS_IS_CORRUPT(context->dp->i_mount,
+				   !xfs_attr_namecheck(name, namelen)))
+			return -EFSCORRUPTED;
 		context->put_listent(context, entry->flags,
 					      name, namelen, valuelen);
 		if (context->seen_enough)
@@ -464,32 +474,33 @@ xfs_attr3_leaf_list_int(
 		cursor->offset++;
 	}
 	trace_xfs_attr_list_leaf_end(context);
-	return;
+	return 0;
 }
 
 /*
  * Copy out attribute entries for attr_list(), for leaf attribute lists.
  */
 STATIC int
-xfs_attr_leaf_list(xfs_attr_list_context_t *context)
+xfs_attr_leaf_list(
+	struct xfs_attr_list_context	*context)
 {
-	int error;
-	struct xfs_buf *bp;
+	struct xfs_buf			*bp;
+	int				error;
 
 	trace_xfs_attr_leaf_list(context);
 
-	context->cursor->blkno = 0;
-	error = xfs_attr3_leaf_read(context->tp, context->dp, 0, -1, &bp);
+	context->cursor.blkno = 0;
+	error = xfs_attr3_leaf_read(context->tp, context->dp, 0, &bp);
 	if (error)
 		return error;
 
-	xfs_attr3_leaf_list_int(bp, context);
+	error = xfs_attr3_leaf_list_int(bp, context);
 	xfs_trans_brelse(context->tp, bp);
-	return 0;
+	return error;
 }
 
 int
-xfs_attr_list_int_ilocked(
+xfs_attr_list_ilocked(
 	struct xfs_attr_list_context	*context)
 {
 	struct xfs_inode		*dp = context->dp;
@@ -509,12 +520,12 @@ xfs_attr_list_int_ilocked(
 }
 
 int
-xfs_attr_list_int(
-	xfs_attr_list_context_t *context)
+xfs_attr_list(
+	struct xfs_attr_list_context	*context)
 {
-	int error;
-	xfs_inode_t *dp = context->dp;
-	uint		lock_mode;
+	struct xfs_inode		*dp = context->dp;
+	uint				lock_mode;
+	int				error;
 
 	XFS_STATS_INC(dp->i_mount, xs_attr_list);
 
@@ -522,130 +533,7 @@ xfs_attr_list_int(
 		return -EIO;
 
 	lock_mode = xfs_ilock_attr_map_shared(dp);
-	error = xfs_attr_list_int_ilocked(context);
+	error = xfs_attr_list_ilocked(context);
 	xfs_iunlock(dp, lock_mode);
 	return error;
 }
-
-#define	ATTR_ENTBASESIZE		/* minimum bytes used by an attr */ \
-	(((struct attrlist_ent *) 0)->a_name - (char *) 0)
-#define	ATTR_ENTSIZE(namelen)		/* actual bytes used by an attr */ \
-	((ATTR_ENTBASESIZE + (namelen) + 1 + sizeof(uint32_t)-1) \
-	 & ~(sizeof(uint32_t)-1))
-
-/*
- * Format an attribute and copy it out to the user's buffer.
- * Take care to check values and protect against them changing later,
- * we may be reading them directly out of a user buffer.
- */
-STATIC void
-xfs_attr_put_listent(
-	xfs_attr_list_context_t *context,
-	int		flags,
-	unsigned char	*name,
-	int		namelen,
-	int		valuelen)
-{
-	struct attrlist *alist = (struct attrlist *)context->alist;
-	attrlist_ent_t *aep;
-	int arraytop;
-
-	ASSERT(!context->seen_enough);
-	ASSERT(!(context->flags & ATTR_KERNOVAL));
-	ASSERT(context->count >= 0);
-	ASSERT(context->count < (ATTR_MAX_VALUELEN/8));
-	ASSERT(context->firstu >= sizeof(*alist));
-	ASSERT(context->firstu <= context->bufsize);
-
-	/*
-	 * Only list entries in the right namespace.
-	 */
-	if (((context->flags & ATTR_SECURE) == 0) !=
-	    ((flags & XFS_ATTR_SECURE) == 0))
-		return;
-	if (((context->flags & ATTR_ROOT) == 0) !=
-	    ((flags & XFS_ATTR_ROOT) == 0))
-		return;
-
-	arraytop = sizeof(*alist) +
-			context->count * sizeof(alist->al_offset[0]);
-	context->firstu -= ATTR_ENTSIZE(namelen);
-	if (context->firstu < arraytop) {
-		trace_xfs_attr_list_full(context);
-		alist->al_more = 1;
-		context->seen_enough = 1;
-		return;
-	}
-
-	aep = (attrlist_ent_t *)&context->alist[context->firstu];
-	aep->a_valuelen = valuelen;
-	memcpy(aep->a_name, name, namelen);
-	aep->a_name[namelen] = 0;
-	alist->al_offset[context->count++] = context->firstu;
-	alist->al_count = context->count;
-	trace_xfs_attr_list_add(context);
-	return;
-}
-
-/*
- * Generate a list of extended attribute names and optionally
- * also value lengths.  Positive return value follows the XFS
- * convention of being an error, zero or negative return code
- * is the length of the buffer returned (negated), indicating
- * success.
- */
-int
-xfs_attr_list(
-	xfs_inode_t	*dp,
-	char		*buffer,
-	int		bufsize,
-	int		flags,
-	attrlist_cursor_kern_t *cursor)
-{
-	xfs_attr_list_context_t context;
-	struct attrlist *alist;
-	int error;
-
-	/*
-	 * Validate the cursor.
-	 */
-	if (cursor->pad1 || cursor->pad2)
-		return -EINVAL;
-	if ((cursor->initted == 0) &&
-	    (cursor->hashval || cursor->blkno || cursor->offset))
-		return -EINVAL;
-
-	/* Only internal consumers can retrieve incomplete attrs. */
-	if (flags & ATTR_INCOMPLETE)
-		return -EINVAL;
-
-	/*
-	 * Check for a properly aligned buffer.
-	 */
-	if (((long)buffer) & (sizeof(int)-1))
-		return -EFAULT;
-	if (flags & ATTR_KERNOVAL)
-		bufsize = 0;
-
-	/*
-	 * Initialize the output buffer.
-	 */
-	memset(&context, 0, sizeof(context));
-	context.dp = dp;
-	context.cursor = cursor;
-	context.resynch = 1;
-	context.flags = flags;
-	context.alist = buffer;
-	context.bufsize = (bufsize & ~(sizeof(int)-1));  /* align */
-	context.firstu = context.bufsize;
-	context.put_listent = xfs_attr_put_listent;
-
-	alist = (struct attrlist *)context.alist;
-	alist->al_count = 0;
-	alist->al_more = 0;
-	alist->al_offset[0] = context.bufsize;
-
-	error = xfs_attr_list_int(&context);
-	ASSERT(error <= 0);
-	return error;
-}
diff --git a/fs/xfs/xfs_bmap_item.c b/fs/xfs/xfs_bmap_item.c
index 83d24e983d4c..ee6f4229cebc 100644
--- a/fs/xfs/xfs_bmap_item.c
+++ b/fs/xfs/xfs_bmap_item.c
@@ -21,7 +21,7 @@
 #include "xfs_icache.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_trans_space.h"
-
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_bui_zone;
 kmem_zone_t	*xfs_bud_zone;
@@ -35,7 +35,7 @@ void
 xfs_bui_item_free(
 	struct xfs_bui_log_item	*buip)
 {
-	kmem_zone_free(xfs_bui_zone, buip);
+	kmem_cache_free(xfs_bui_zone, buip);
 }
 
 /*
@@ -201,7 +201,7 @@ xfs_bud_item_release(
 	struct xfs_bud_log_item	*budp = BUD_ITEM(lip);
 
 	xfs_bui_release(budp->bud_buip);
-	kmem_zone_free(xfs_bud_zone, budp);
+	kmem_cache_free(xfs_bud_zone, budp);
 }
 
 static const struct xfs_item_ops xfs_bud_item_ops = {
@@ -456,7 +456,7 @@ xfs_bui_recover(
 	if (buip->bui_format.bui_nextents != XFS_BUI_MAX_FAST_EXTENTS) {
 		set_bit(XFS_BUI_RECOVERED, &buip->bui_flags);
 		xfs_bui_release(buip);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -490,7 +490,7 @@ xfs_bui_recover(
 		 */
 		set_bit(XFS_BUI_RECOVERED, &buip->bui_flags);
 		xfs_bui_release(buip);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_itruncate,
@@ -525,6 +525,7 @@ xfs_bui_recover(
 		type = bui_type;
 		break;
 	default:
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto err_inode;
 	}
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 4f443703065e..4f800f7fe888 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -53,15 +53,16 @@ xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
  */
 int
 xfs_zero_extent(
-	struct xfs_inode *ip,
-	xfs_fsblock_t	start_fsb,
-	xfs_off_t	count_fsb)
+	struct xfs_inode	*ip,
+	xfs_fsblock_t		start_fsb,
+	xfs_off_t		count_fsb)
 {
-	struct xfs_mount *mp = ip->i_mount;
-	xfs_daddr_t	sector = xfs_fsb_to_db(ip, start_fsb);
-	sector_t	block = XFS_BB_TO_FSBT(mp, sector);
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
+	xfs_daddr_t		sector = xfs_fsb_to_db(ip, start_fsb);
+	sector_t		block = XFS_BB_TO_FSBT(mp, sector);
 
-	return blkdev_issue_zeroout(xfs_find_bdev_for_inode(VFS_I(ip)),
+	return blkdev_issue_zeroout(target->bt_bdev,
 		block << (mp->m_super->s_blocksize_bits - 9),
 		count_fsb << (mp->m_super->s_blocksize_bits - 9),
 		GFP_NOFS, 0);
@@ -164,13 +165,6 @@ xfs_bmap_rtalloc(
 		xfs_trans_mod_dquot_byino(ap->tp, ap->ip,
 			ap->wasdel ? XFS_TRANS_DQ_DELRTBCOUNT :
 					XFS_TRANS_DQ_RTBCOUNT, (long) ralen);
-
-		/* Zero the extent if we were asked to do so */
-		if (ap->datatype & XFS_ALLOC_USERDATA_ZERO) {
-			error = xfs_zero_extent(ap->ip, ap->blkno, ap->length);
-			if (error)
-				return error;
-		}
 	} else {
 		ap->length = 0;
 	}
@@ -179,29 +173,6 @@ xfs_bmap_rtalloc(
 #endif /* CONFIG_XFS_RT */
 
 /*
- * Check if the endoff is outside the last extent. If so the caller will grow
- * the allocation to a stripe unit boundary.  All offsets are considered outside
- * the end of file for an empty fork, so 1 is returned in *eof in that case.
- */
-int
-xfs_bmap_eof(
-	struct xfs_inode	*ip,
-	xfs_fileoff_t		endoff,
-	int			whichfork,
-	int			*eof)
-{
-	struct xfs_bmbt_irec	rec;
-	int			error;
-
-	error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, eof);
-	if (error || *eof)
-		return error;
-
-	*eof = endoff >= rec.br_startoff + rec.br_blockcount;
-	return 0;
-}
-
-/*
  * Extent tree block counting routines.
  */
 
@@ -229,106 +200,6 @@ xfs_bmap_count_leaves(
 }
 
 /*
- * Count leaf blocks given a range of extent records originally
- * in btree format.
- */
-STATIC void
-xfs_bmap_disk_count_leaves(
-	struct xfs_mount	*mp,
-	struct xfs_btree_block	*block,
-	int			numrecs,
-	xfs_filblks_t		*count)
-{
-	int		b;
-	xfs_bmbt_rec_t	*frp;
-
-	for (b = 1; b <= numrecs; b++) {
-		frp = XFS_BMBT_REC_ADDR(mp, block, b);
-		*count += xfs_bmbt_disk_get_blockcount(frp);
-	}
-}
-
-/*
- * Recursively walks each level of a btree
- * to count total fsblocks in use.
- */
-STATIC int
-xfs_bmap_count_tree(
-	struct xfs_mount	*mp,
-	struct xfs_trans	*tp,
-	struct xfs_ifork	*ifp,
-	xfs_fsblock_t		blockno,
-	int			levelin,
-	xfs_extnum_t		*nextents,
-	xfs_filblks_t		*count)
-{
-	int			error;
-	struct xfs_buf		*bp, *nbp;
-	int			level = levelin;
-	__be64			*pp;
-	xfs_fsblock_t           bno = blockno;
-	xfs_fsblock_t		nextbno;
-	struct xfs_btree_block	*block, *nextblock;
-	int			numrecs;
-
-	error = xfs_btree_read_bufl(mp, tp, bno, &bp, XFS_BMAP_BTREE_REF,
-						&xfs_bmbt_buf_ops);
-	if (error)
-		return error;
-	*count += 1;
-	block = XFS_BUF_TO_BLOCK(bp);
-
-	if (--level) {
-		/* Not at node above leaves, count this level of nodes */
-		nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
-		while (nextbno != NULLFSBLOCK) {
-			error = xfs_btree_read_bufl(mp, tp, nextbno, &nbp,
-						XFS_BMAP_BTREE_REF,
-						&xfs_bmbt_buf_ops);
-			if (error)
-				return error;
-			*count += 1;
-			nextblock = XFS_BUF_TO_BLOCK(nbp);
-			nextbno = be64_to_cpu(nextblock->bb_u.l.bb_rightsib);
-			xfs_trans_brelse(tp, nbp);
-		}
-
-		/* Dive to the next level */
-		pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
-		bno = be64_to_cpu(*pp);
-		error = xfs_bmap_count_tree(mp, tp, ifp, bno, level, nextents,
-				count);
-		if (error) {
-			xfs_trans_brelse(tp, bp);
-			XFS_ERROR_REPORT("xfs_bmap_count_tree(1)",
-					 XFS_ERRLEVEL_LOW, mp);
-			return -EFSCORRUPTED;
-		}
-		xfs_trans_brelse(tp, bp);
-	} else {
-		/* count all level 1 nodes and their leaves */
-		for (;;) {
-			nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
-			numrecs = be16_to_cpu(block->bb_numrecs);
-			(*nextents) += numrecs;
-			xfs_bmap_disk_count_leaves(mp, block, numrecs, count);
-			xfs_trans_brelse(tp, bp);
-			if (nextbno == NULLFSBLOCK)
-				break;
-			bno = nextbno;
-			error = xfs_btree_read_bufl(mp, tp, bno, &bp,
-						XFS_BMAP_BTREE_REF,
-						&xfs_bmbt_buf_ops);
-			if (error)
-				return error;
-			*count += 1;
-			block = XFS_BUF_TO_BLOCK(bp);
-		}
-	}
-	return 0;
-}
-
-/*
  * Count fsblocks of the given fork.  Delayed allocation extents are
  * not counted towards the totals.
  */
@@ -340,26 +211,19 @@ xfs_bmap_count_blocks(
 	xfs_extnum_t		*nextents,
 	xfs_filblks_t		*count)
 {
-	struct xfs_mount	*mp;	/* file system mount structure */
-	__be64			*pp;	/* pointer to block address */
-	struct xfs_btree_block	*block;	/* current btree block */
-	struct xfs_ifork	*ifp;	/* fork structure */
-	xfs_fsblock_t		bno;	/* block # of "block" */
-	int			level;	/* btree level, for checking */
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
+	struct xfs_btree_cur	*cur;
+	xfs_extlen_t		btblocks = 0;
 	int			error;
 
-	bno = NULLFSBLOCK;
-	mp = ip->i_mount;
 	*nextents = 0;
 	*count = 0;
-	ifp = XFS_IFORK_PTR(ip, whichfork);
+
 	if (!ifp)
 		return 0;
 
 	switch (XFS_IFORK_FORMAT(ip, whichfork)) {
-	case XFS_DINODE_FMT_EXTENTS:
-		*nextents = xfs_bmap_count_leaves(ifp, count);
-		return 0;
 	case XFS_DINODE_FMT_BTREE:
 		if (!(ifp->if_flags & XFS_IFEXTENTS)) {
 			error = xfs_iread_extents(tp, ip, whichfork);
@@ -367,26 +231,23 @@ xfs_bmap_count_blocks(
 				return error;
 		}
 
+		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
+		error = xfs_btree_count_blocks(cur, &btblocks);
+		xfs_btree_del_cursor(cur, error);
+		if (error)
+			return error;
+
 		/*
-		 * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
+		 * xfs_btree_count_blocks includes the root block contained in
+		 * the inode fork in @btblocks, so subtract one because we're
+		 * only interested in allocated disk blocks.
 		 */
-		block = ifp->if_broot;
-		level = be16_to_cpu(block->bb_level);
-		ASSERT(level > 0);
-		pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
-		bno = be64_to_cpu(*pp);
-		ASSERT(bno != NULLFSBLOCK);
-		ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
-		ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
-
-		error = xfs_bmap_count_tree(mp, tp, ifp, bno, level,
-				nextents, count);
-		if (error) {
-			XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)",
-					XFS_ERRLEVEL_LOW, mp);
-			return -EFSCORRUPTED;
-		}
-		return 0;
+		*count += btblocks - 1;
+
+		/* fall through */
+	case XFS_DINODE_FMT_EXTENTS:
+		*nextents = xfs_bmap_count_leaves(ifp, count);
+		break;
 	}
 
 	return 0;
@@ -964,8 +825,8 @@ xfs_alloc_file_space(
 		xfs_trans_ijoin(tp, ip, 0);
 
 		error = xfs_bmapi_write(tp, ip, startoffset_fsb,
-					allocatesize_fsb, alloc_type, resblks,
-					imapp, &nimaps);
+					allocatesize_fsb, alloc_type, 0, imapp,
+					&nimaps);
 		if (error)
 			goto error0;
 
@@ -1039,6 +900,7 @@ out_trans_cancel:
 	goto out_unlock;
 }
 
+/* Caller must first wait for the completion of any pending DIOs if required. */
 int
 xfs_flush_unmap_range(
 	struct xfs_inode	*ip,
@@ -1050,9 +912,6 @@ xfs_flush_unmap_range(
 	xfs_off_t		rounding, start, end;
 	int			error;
 
-	/* wait for the completion of any pending DIOs */
-	inode_dio_wait(inode);
-
 	rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_SIZE);
 	start = round_down(offset, rounding);
 	end = round_up(offset + len, rounding) - 1;
@@ -1084,10 +943,6 @@ xfs_free_file_space(
 	if (len <= 0)	/* if nothing being freed */
 		return 0;
 
-	error = xfs_flush_unmap_range(ip, offset, len);
-	if (error)
-		return error;
-
 	startoffset_fsb = XFS_B_TO_FSB(mp, offset);
 	endoffset_fsb = XFS_B_TO_FSBT(mp, offset + len);
 
@@ -1113,7 +968,8 @@ xfs_free_file_space(
 		return 0;
 	if (offset + len > XFS_ISIZE(ip))
 		len = XFS_ISIZE(ip) - offset;
-	error = iomap_zero_range(VFS_I(ip), offset, len, NULL, &xfs_iomap_ops);
+	error = iomap_zero_range(VFS_I(ip), offset, len, NULL,
+			&xfs_buffered_write_iomap_ops);
 	if (error)
 		return error;
 
@@ -1131,48 +987,12 @@ xfs_free_file_space(
 	return error;
 }
 
-/*
- * Preallocate and zero a range of a file. This mechanism has the allocation
- * semantics of fallocate and in addition converts data in the range to zeroes.
- */
-int
-xfs_zero_file_space(
-	struct xfs_inode	*ip,
-	xfs_off_t		offset,
-	xfs_off_t		len)
-{
-	struct xfs_mount	*mp = ip->i_mount;
-	uint			blksize;
-	int			error;
-
-	trace_xfs_zero_file_space(ip);
-
-	blksize = 1 << mp->m_sb.sb_blocklog;
-
-	/*
-	 * Punch a hole and prealloc the range. We use hole punch rather than
-	 * unwritten extent conversion for two reasons:
-	 *
-	 * 1.) Hole punch handles partial block zeroing for us.
-	 *
-	 * 2.) If prealloc returns ENOSPC, the file range is still zero-valued
-	 * by virtue of the hole punch.
-	 */
-	error = xfs_free_file_space(ip, offset, len);
-	if (error || xfs_is_always_cow_inode(ip))
-		return error;
-
-	return xfs_alloc_file_space(ip, round_down(offset, blksize),
-				     round_up(offset + len, blksize) -
-				     round_down(offset, blksize),
-				     XFS_BMAPI_PREALLOC);
-}
-
 static int
 xfs_prepare_shift(
 	struct xfs_inode	*ip,
 	loff_t			offset)
 {
+	struct xfs_mount	*mp = ip->i_mount;
 	int			error;
 
 	/*
@@ -1186,6 +1006,17 @@ xfs_prepare_shift(
 	}
 
 	/*
+	 * Shift operations must stabilize the start block offset boundary along
+	 * with the full range of the operation. If we don't, a COW writeback
+	 * completion could race with an insert, front merge with the start
+	 * extent (after split) during the shift and corrupt the file. Start
+	 * with the block just prior to the start to stabilize the boundary.
+	 */
+	offset = round_down(offset, 1 << mp->m_sb.sb_blocklog);
+	if (offset)
+		offset -= (1 << mp->m_sb.sb_blocklog);
+
+	/*
 	 * Writeback and invalidate cache for the remainder of the file as we're
 	 * about to shift down every extent from offset to EOF.
 	 */
@@ -1231,7 +1062,6 @@ xfs_collapse_file_space(
 	int			error;
 	xfs_fileoff_t		next_fsb = XFS_B_TO_FSB(mp, offset + len);
 	xfs_fileoff_t		shift_fsb = XFS_B_TO_FSB(mp, len);
-	uint			resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
 	bool			done = false;
 
 	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
@@ -1247,32 +1077,34 @@ xfs_collapse_file_space(
 	if (error)
 		return error;
 
-	while (!error && !done) {
-		error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, 0,
-					&tp);
-		if (error)
-			break;
+	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, 0, 0, 0, &tp);
+	if (error)
+		return error;
 
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		error = xfs_trans_reserve_quota(tp, mp, ip->i_udquot,
-				ip->i_gdquot, ip->i_pdquot, resblks, 0,
-				XFS_QMOPT_RES_REGBLKS);
-		if (error)
-			goto out_trans_cancel;
-		xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, 0);
 
+	while (!done) {
 		error = xfs_bmap_collapse_extents(tp, ip, &next_fsb, shift_fsb,
 				&done);
 		if (error)
 			goto out_trans_cancel;
+		if (done)
+			break;
 
-		error = xfs_trans_commit(tp);
+		/* finish any deferred frees and roll the transaction */
+		error = xfs_defer_finish(&tp);
+		if (error)
+			goto out_trans_cancel;
 	}
 
+	error = xfs_trans_commit(tp);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return error;
 
 out_trans_cancel:
 	xfs_trans_cancel(tp);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return error;
 }
 
@@ -1315,35 +1147,41 @@ xfs_insert_file_space(
 	if (error)
 		return error;
 
+	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write,
+			XFS_DIOSTRAT_SPACE_RES(mp, 0), 0, 0, &tp);
+	if (error)
+		return error;
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, 0);
+
 	/*
 	 * The extent shifting code works on extent granularity. So, if stop_fsb
 	 * is not the starting block of extent, we need to split the extent at
 	 * stop_fsb.
 	 */
-	error = xfs_bmap_split_extent(ip, stop_fsb);
+	error = xfs_bmap_split_extent(tp, ip, stop_fsb);
 	if (error)
-		return error;
+		goto out_trans_cancel;
 
-	while (!error && !done) {
-		error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, 0, 0, 0,
-					&tp);
+	do {
+		error = xfs_trans_roll_inode(&tp, ip);
 		if (error)
-			break;
+			goto out_trans_cancel;
 
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 		error = xfs_bmap_insert_extents(tp, ip, &next_fsb, shift_fsb,
 				&done, stop_fsb);
 		if (error)
 			goto out_trans_cancel;
+	} while (!done);
 
-		error = xfs_trans_commit(tp);
-	}
-
+	error = xfs_trans_commit(tp);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return error;
 
 out_trans_cancel:
 	xfs_trans_cancel(tp);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return error;
 }
 
@@ -1611,12 +1449,12 @@ xfs_swap_extent_forks(
 	 * event of a crash. Set the owner change log flags now and leave the
 	 * bmbt scan as the last step.
 	 */
-	if (ip->i_d.di_version == 3 &&
-	    ip->i_d.di_format == XFS_DINODE_FMT_BTREE)
-		(*target_log_flags) |= XFS_ILOG_DOWNER;
-	if (tip->i_d.di_version == 3 &&
-	    tip->i_d.di_format == XFS_DINODE_FMT_BTREE)
-		(*src_log_flags) |= XFS_ILOG_DOWNER;
+	if (xfs_sb_version_has_v3inode(&ip->i_mount->m_sb)) {
+		if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE)
+			(*target_log_flags) |= XFS_ILOG_DOWNER;
+		if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE)
+			(*src_log_flags) |= XFS_ILOG_DOWNER;
+	}
 
 	/*
 	 * Swap the data forks of the inodes
@@ -1651,7 +1489,7 @@ xfs_swap_extent_forks(
 		(*src_log_flags) |= XFS_ILOG_DEXT;
 		break;
 	case XFS_DINODE_FMT_BTREE:
-		ASSERT(ip->i_d.di_version < 3 ||
+		ASSERT(!xfs_sb_version_has_v3inode(&ip->i_mount->m_sb) ||
 		       (*src_log_flags & XFS_ILOG_DOWNER));
 		(*src_log_flags) |= XFS_ILOG_DBROOT;
 		break;
@@ -1663,7 +1501,7 @@ xfs_swap_extent_forks(
 		break;
 	case XFS_DINODE_FMT_BTREE:
 		(*target_log_flags) |= XFS_ILOG_DBROOT;
-		ASSERT(tip->i_d.di_version < 3 ||
+		ASSERT(!xfs_sb_version_has_v3inode(&ip->i_mount->m_sb) ||
 		       (*target_log_flags & XFS_ILOG_DOWNER));
 		break;
 	}
@@ -1750,6 +1588,14 @@ xfs_swap_extents(
 		goto out_unlock;
 	}
 
+	error = xfs_qm_dqattach(ip);
+	if (error)
+		goto out_unlock;
+
+	error = xfs_qm_dqattach(tip);
+	if (error)
+		goto out_unlock;
+
 	error = xfs_swap_extent_flush(ip);
 	if (error)
 		goto out_unlock;
diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h
index 7a78229cf1a7..9f993168b55b 100644
--- a/fs/xfs/xfs_bmap_util.h
+++ b/fs/xfs/xfs_bmap_util.h
@@ -30,8 +30,6 @@ xfs_bmap_rtalloc(struct xfs_bmalloca *ap)
 }
 #endif /* CONFIG_XFS_RT */
 
-int	xfs_bmap_eof(struct xfs_inode *ip, xfs_fileoff_t endoff,
-		     int whichfork, int *eof);
 int	xfs_bmap_punch_delalloc_range(struct xfs_inode *ip,
 		xfs_fileoff_t start_fsb, xfs_fileoff_t length);
 
@@ -59,8 +57,6 @@ int	xfs_alloc_file_space(struct xfs_inode *ip, xfs_off_t offset,
 			     xfs_off_t len, int alloc_type);
 int	xfs_free_file_space(struct xfs_inode *ip, xfs_off_t offset,
 			    xfs_off_t len);
-int	xfs_zero_file_space(struct xfs_inode *ip, xfs_off_t offset,
-			    xfs_off_t len);
 int	xfs_collapse_file_space(struct xfs_inode *, xfs_off_t offset,
 				xfs_off_t len);
 int	xfs_insert_file_space(struct xfs_inode *, xfs_off_t offset,
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 0abba171aa89..65538d18e64f 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -198,20 +198,22 @@ xfs_buf_free_maps(
 	}
 }
 
-static struct xfs_buf *
+static int
 _xfs_buf_alloc(
 	struct xfs_buftarg	*target,
 	struct xfs_buf_map	*map,
 	int			nmaps,
-	xfs_buf_flags_t		flags)
+	xfs_buf_flags_t		flags,
+	struct xfs_buf		**bpp)
 {
 	struct xfs_buf		*bp;
 	int			error;
 	int			i;
 
+	*bpp = NULL;
 	bp = kmem_zone_zalloc(xfs_buf_zone, KM_NOFS);
 	if (unlikely(!bp))
-		return NULL;
+		return -ENOMEM;
 
 	/*
 	 * We don't want certain flags to appear in b_flags unless they are
@@ -238,8 +240,8 @@ _xfs_buf_alloc(
 	 */
 	error = xfs_buf_get_maps(bp, nmaps);
 	if (error)  {
-		kmem_zone_free(xfs_buf_zone, bp);
-		return NULL;
+		kmem_cache_free(xfs_buf_zone, bp);
+		return error;
 	}
 
 	bp->b_bn = map[0].bm_bn;
@@ -256,7 +258,8 @@ _xfs_buf_alloc(
 	XFS_STATS_INC(bp->b_mount, xb_create);
 	trace_xfs_buf_init(bp, _RET_IP_);
 
-	return bp;
+	*bpp = bp;
+	return 0;
 }
 
 /*
@@ -304,7 +307,7 @@ _xfs_buf_free_pages(
  * 	The buffer must not be on any hash - use xfs_buf_rele instead for
  * 	hashed and refcounted buffers
  */
-void
+static void
 xfs_buf_free(
 	xfs_buf_t		*bp)
 {
@@ -324,11 +327,14 @@ xfs_buf_free(
 
 			__free_page(page);
 		}
+		if (current->reclaim_state)
+			current->reclaim_state->reclaimed_slab +=
+							bp->b_page_count;
 	} else if (bp->b_flags & _XBF_KMEM)
 		kmem_free(bp->b_addr);
 	_xfs_buf_free_pages(bp);
 	xfs_buf_free_maps(bp);
-	kmem_zone_free(xfs_buf_zone, bp);
+	kmem_cache_free(xfs_buf_zone, bp);
 }
 
 /*
@@ -461,7 +467,7 @@ _xfs_buf_map_pages(
 		unsigned nofs_flag;
 
 		/*
-		 * vm_map_ram() will allocate auxillary structures (e.g.
+		 * vm_map_ram() will allocate auxiliary structures (e.g.
 		 * pagetables) with GFP_KERNEL, yet we are likely to be under
 		 * GFP_NOFS context here. Hence we need to tell memory reclaim
 		 * that we are in such a context via PF_MEMALLOC_NOFS to prevent
@@ -471,7 +477,7 @@ _xfs_buf_map_pages(
 		nofs_flag = memalloc_nofs_save();
 		do {
 			bp->b_addr = vm_map_ram(bp->b_pages, bp->b_page_count,
-						-1, PAGE_KERNEL);
+						-1);
 			if (bp->b_addr)
 				break;
 			vm_unmap_aliases();
@@ -682,53 +688,39 @@ xfs_buf_incore(
  * cache hits, as metadata intensive workloads will see 3 orders of magnitude
  * more hits than misses.
  */
-struct xfs_buf *
+int
 xfs_buf_get_map(
 	struct xfs_buftarg	*target,
 	struct xfs_buf_map	*map,
 	int			nmaps,
-	xfs_buf_flags_t		flags)
+	xfs_buf_flags_t		flags,
+	struct xfs_buf		**bpp)
 {
 	struct xfs_buf		*bp;
 	struct xfs_buf		*new_bp;
 	int			error = 0;
 
+	*bpp = NULL;
 	error = xfs_buf_find(target, map, nmaps, flags, NULL, &bp);
-
-	switch (error) {
-	case 0:
-		/* cache hit */
+	if (!error)
 		goto found;
-	case -EAGAIN:
-		/* cache hit, trylock failure, caller handles failure */
-		ASSERT(flags & XBF_TRYLOCK);
-		return NULL;
-	case -ENOENT:
-		/* cache miss, go for insert */
-		break;
-	case -EFSCORRUPTED:
-	default:
-		/*
-		 * None of the higher layers understand failure types
-		 * yet, so return NULL to signal a fatal lookup error.
-		 */
-		return NULL;
-	}
+	if (error != -ENOENT)
+		return error;
 
-	new_bp = _xfs_buf_alloc(target, map, nmaps, flags);
-	if (unlikely(!new_bp))
-		return NULL;
+	error = _xfs_buf_alloc(target, map, nmaps, flags, &new_bp);
+	if (error)
+		return error;
 
 	error = xfs_buf_allocate_memory(new_bp, flags);
 	if (error) {
 		xfs_buf_free(new_bp);
-		return NULL;
+		return error;
 	}
 
 	error = xfs_buf_find(target, map, nmaps, flags, new_bp, &bp);
 	if (error) {
 		xfs_buf_free(new_bp);
-		return NULL;
+		return error;
 	}
 
 	if (bp != new_bp)
@@ -738,10 +730,11 @@ found:
 	if (!bp->b_addr) {
 		error = _xfs_buf_map_pages(bp, flags);
 		if (unlikely(error)) {
-			xfs_warn(target->bt_mount,
-				"%s: failed to map pagesn", __func__);
+			xfs_warn_ratelimited(target->bt_mount,
+				"%s: failed to map %u pages", __func__,
+				bp->b_page_count);
 			xfs_buf_relse(bp);
-			return NULL;
+			return error;
 		}
 	}
 
@@ -754,7 +747,8 @@ found:
 
 	XFS_STATS_INC(target->bt_mount, xb_get);
 	trace_xfs_buf_get(bp, flags, _RET_IP_);
-	return bp;
+	*bpp = bp;
+	return 0;
 }
 
 STATIC int
@@ -806,46 +800,77 @@ xfs_buf_reverify(
 	return bp->b_error;
 }
 
-xfs_buf_t *
+int
 xfs_buf_read_map(
 	struct xfs_buftarg	*target,
 	struct xfs_buf_map	*map,
 	int			nmaps,
 	xfs_buf_flags_t		flags,
-	const struct xfs_buf_ops *ops)
+	struct xfs_buf		**bpp,
+	const struct xfs_buf_ops *ops,
+	xfs_failaddr_t		fa)
 {
 	struct xfs_buf		*bp;
+	int			error;
 
 	flags |= XBF_READ;
+	*bpp = NULL;
 
-	bp = xfs_buf_get_map(target, map, nmaps, flags);
-	if (!bp)
-		return NULL;
+	error = xfs_buf_get_map(target, map, nmaps, flags, &bp);
+	if (error)
+		return error;
 
 	trace_xfs_buf_read(bp, flags, _RET_IP_);
 
 	if (!(bp->b_flags & XBF_DONE)) {
+		/* Initiate the buffer read and wait. */
 		XFS_STATS_INC(target->bt_mount, xb_get_read);
 		bp->b_ops = ops;
-		_xfs_buf_read(bp, flags);
-		return bp;
+		error = _xfs_buf_read(bp, flags);
+
+		/* Readahead iodone already dropped the buffer, so exit. */
+		if (flags & XBF_ASYNC)
+			return 0;
+	} else {
+		/* Buffer already read; all we need to do is check it. */
+		error = xfs_buf_reverify(bp, ops);
+
+		/* Readahead already finished; drop the buffer and exit. */
+		if (flags & XBF_ASYNC) {
+			xfs_buf_relse(bp);
+			return 0;
+		}
+
+		/* We do not want read in the flags */
+		bp->b_flags &= ~XBF_READ;
+		ASSERT(bp->b_ops != NULL || ops == NULL);
 	}
 
-	xfs_buf_reverify(bp, ops);
+	/*
+	 * If we've had a read error, then the contents of the buffer are
+	 * invalid and should not be used. To ensure that a followup read tries
+	 * to pull the buffer from disk again, we clear the XBF_DONE flag and
+	 * mark the buffer stale. This ensures that anyone who has a current
+	 * reference to the buffer will interpret it's contents correctly and
+	 * future cache lookups will also treat it as an empty, uninitialised
+	 * buffer.
+	 */
+	if (error) {
+		if (!XFS_FORCED_SHUTDOWN(target->bt_mount))
+			xfs_buf_ioerror_alert(bp, fa);
 
-	if (flags & XBF_ASYNC) {
-		/*
-		 * Read ahead call which is already satisfied,
-		 * drop the buffer
-		 */
+		bp->b_flags &= ~XBF_DONE;
+		xfs_buf_stale(bp);
 		xfs_buf_relse(bp);
-		return NULL;
+
+		/* bad CRC means corrupted metadata */
+		if (error == -EFSBADCRC)
+			error = -EFSCORRUPTED;
+		return error;
 	}
 
-	/* We do not want read in the flags */
-	bp->b_flags &= ~XBF_READ;
-	ASSERT(bp->b_ops != NULL || ops == NULL);
-	return bp;
+	*bpp = bp;
+	return 0;
 }
 
 /*
@@ -859,11 +884,14 @@ xfs_buf_readahead_map(
 	int			nmaps,
 	const struct xfs_buf_ops *ops)
 {
+	struct xfs_buf		*bp;
+
 	if (bdi_read_congested(target->bt_bdev->bd_bdi))
 		return;
 
 	xfs_buf_read_map(target, map, nmaps,
-		     XBF_TRYLOCK|XBF_ASYNC|XBF_READ_AHEAD, ops);
+		     XBF_TRYLOCK | XBF_ASYNC | XBF_READ_AHEAD, &bp, ops,
+		     __this_address);
 }
 
 /*
@@ -880,12 +908,13 @@ xfs_buf_read_uncached(
 	const struct xfs_buf_ops *ops)
 {
 	struct xfs_buf		*bp;
+	int			error;
 
 	*bpp = NULL;
 
-	bp = xfs_buf_get_uncached(target, numblks, flags);
-	if (!bp)
-		return -ENOMEM;
+	error = xfs_buf_get_uncached(target, numblks, flags, &bp);
+	if (error)
+		return error;
 
 	/* set up the buffer for a read IO */
 	ASSERT(bp->b_map_count == 1);
@@ -896,7 +925,7 @@ xfs_buf_read_uncached(
 
 	xfs_buf_submit(bp);
 	if (bp->b_error) {
-		int	error = bp->b_error;
+		error = bp->b_error;
 		xfs_buf_relse(bp);
 		return error;
 	}
@@ -905,20 +934,23 @@ xfs_buf_read_uncached(
 	return 0;
 }
 
-xfs_buf_t *
+int
 xfs_buf_get_uncached(
 	struct xfs_buftarg	*target,
 	size_t			numblks,
-	int			flags)
+	int			flags,
+	struct xfs_buf		**bpp)
 {
 	unsigned long		page_count;
 	int			error, i;
 	struct xfs_buf		*bp;
 	DEFINE_SINGLE_BUF_MAP(map, XFS_BUF_DADDR_NULL, numblks);
 
+	*bpp = NULL;
+
 	/* flags might contain irrelevant bits, pass only what we care about */
-	bp = _xfs_buf_alloc(target, &map, 1, flags & XBF_NO_IOACCT);
-	if (unlikely(bp == NULL))
+	error = _xfs_buf_alloc(target, &map, 1, flags & XBF_NO_IOACCT, &bp);
+	if (error)
 		goto fail;
 
 	page_count = PAGE_ALIGN(numblks << BBSHIFT) >> PAGE_SHIFT;
@@ -928,8 +960,10 @@ xfs_buf_get_uncached(
 
 	for (i = 0; i < page_count; i++) {
 		bp->b_pages[i] = alloc_page(xb_to_gfp(flags));
-		if (!bp->b_pages[i])
+		if (!bp->b_pages[i]) {
+			error = -ENOMEM;
 			goto fail_free_mem;
+		}
 	}
 	bp->b_flags |= _XBF_PAGES;
 
@@ -941,7 +975,8 @@ xfs_buf_get_uncached(
 	}
 
 	trace_xfs_buf_get_uncached(bp, _RET_IP_);
-	return bp;
+	*bpp = bp;
+	return 0;
 
  fail_free_mem:
 	while (--i >= 0)
@@ -949,9 +984,9 @@ xfs_buf_get_uncached(
 	_xfs_buf_free_pages(bp);
  fail_free_buf:
 	xfs_buf_free_maps(bp);
-	kmem_zone_free(xfs_buf_zone, bp);
+	kmem_cache_free(xfs_buf_zone, bp);
  fail:
-	return NULL;
+	return error;
 }
 
 /*
@@ -1205,10 +1240,10 @@ __xfs_buf_ioerror(
 void
 xfs_buf_ioerror_alert(
 	struct xfs_buf		*bp,
-	const char		*func)
+	xfs_failaddr_t		func)
 {
-	xfs_alert(bp->b_mount,
-"metadata I/O error in \"%s\" at daddr 0x%llx len %d error %d",
+	xfs_alert_ratelimited(bp->b_mount,
+"metadata I/O error in \"%pS\" at daddr 0x%llx len %d error %d",
 			func, (uint64_t)XFS_BUF_ADDR(bp), bp->b_length,
 			-bp->b_error);
 }
@@ -1261,8 +1296,7 @@ xfs_buf_ioapply_map(
 	int		map,
 	int		*buf_offset,
 	int		*count,
-	int		op,
-	int		op_flags)
+	int		op)
 {
 	int		page_index;
 	int		total_nr_pages = bp->b_page_count;
@@ -1297,7 +1331,7 @@ next_chunk:
 	bio->bi_iter.bi_sector = sector;
 	bio->bi_end_io = xfs_buf_bio_end_io;
 	bio->bi_private = bp;
-	bio_set_op_attrs(bio, op, op_flags);
+	bio->bi_opf = op;
 
 	for (; size && nr_pages; nr_pages--, page_index++) {
 		int	rbytes, nbytes = PAGE_SIZE - offset;
@@ -1342,7 +1376,6 @@ _xfs_buf_ioapply(
 {
 	struct blk_plug	plug;
 	int		op;
-	int		op_flags = 0;
 	int		offset;
 	int		size;
 	int		i;
@@ -1384,15 +1417,14 @@ _xfs_buf_ioapply(
 				dump_stack();
 			}
 		}
-	} else if (bp->b_flags & XBF_READ_AHEAD) {
-		op = REQ_OP_READ;
-		op_flags = REQ_RAHEAD;
 	} else {
 		op = REQ_OP_READ;
+		if (bp->b_flags & XBF_READ_AHEAD)
+			op |= REQ_RAHEAD;
 	}
 
 	/* we only use the buffer cache for meta-data */
-	op_flags |= REQ_META;
+	op |= REQ_META;
 
 	/*
 	 * Walk all the vectors issuing IO on them. Set up the initial offset
@@ -1404,7 +1436,7 @@ _xfs_buf_ioapply(
 	size = BBTOB(bp->b_length);
 	blk_start_plug(&plug);
 	for (i = 0; i < bp->b_map_count; i++) {
-		xfs_buf_ioapply_map(bp, i, &offset, &size, op, op_flags);
+		xfs_buf_ioapply_map(bp, i, &offset, &size, op);
 		if (bp->b_error)
 			break;
 		if (size <= 0)
@@ -1545,6 +1577,28 @@ xfs_buf_zero(
 }
 
 /*
+ * Log a message about and stale a buffer that a caller has decided is corrupt.
+ *
+ * This function should be called for the kinds of metadata corruption that
+ * cannot be detect from a verifier, such as incorrect inter-block relationship
+ * data.  Do /not/ call this function from a verifier function.
+ *
+ * The buffer must be XBF_DONE prior to the call.  Afterwards, the buffer will
+ * be marked stale, but b_error will not be set.  The caller is responsible for
+ * releasing the buffer or fixing it.
+ */
+void
+__xfs_buf_mark_corrupt(
+	struct xfs_buf		*bp,
+	xfs_failaddr_t		fa)
+{
+	ASSERT(bp->b_flags & XBF_DONE);
+
+	xfs_buf_corruption_error(bp, fa);
+	xfs_buf_stale(bp);
+}
+
+/*
  *	Handling of buffer targets (buftargs).
  */
 
@@ -2063,8 +2117,11 @@ xfs_buf_delwri_pushbuf(
 int __init
 xfs_buf_init(void)
 {
-	xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf",
-						KM_ZONE_HWALIGN, NULL);
+	xfs_buf_zone = kmem_cache_create("xfs_buf", sizeof(struct xfs_buf), 0,
+					 SLAB_HWCACHE_ALIGN |
+					 SLAB_RECLAIM_ACCOUNT |
+					 SLAB_MEM_SPREAD,
+					 NULL);
 	if (!xfs_buf_zone)
 		goto out;
 
@@ -2077,7 +2134,7 @@ xfs_buf_init(void)
 void
 xfs_buf_terminate(void)
 {
-	kmem_zone_destroy(xfs_buf_zone);
+	kmem_cache_destroy(xfs_buf_zone);
 }
 
 void xfs_buf_set_ref(struct xfs_buf *bp, int lru_ref)
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index f6ce17d8d848..9a04c53c2488 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -192,37 +192,40 @@ struct xfs_buf *xfs_buf_incore(struct xfs_buftarg *target,
 			   xfs_daddr_t blkno, size_t numblks,
 			   xfs_buf_flags_t flags);
 
-struct xfs_buf *xfs_buf_get_map(struct xfs_buftarg *target,
-			       struct xfs_buf_map *map, int nmaps,
-			       xfs_buf_flags_t flags);
-struct xfs_buf *xfs_buf_read_map(struct xfs_buftarg *target,
-			       struct xfs_buf_map *map, int nmaps,
-			       xfs_buf_flags_t flags,
-			       const struct xfs_buf_ops *ops);
+int xfs_buf_get_map(struct xfs_buftarg *target, struct xfs_buf_map *map,
+		int nmaps, xfs_buf_flags_t flags, struct xfs_buf **bpp);
+int xfs_buf_read_map(struct xfs_buftarg *target, struct xfs_buf_map *map,
+		int nmaps, xfs_buf_flags_t flags, struct xfs_buf **bpp,
+		const struct xfs_buf_ops *ops, xfs_failaddr_t fa);
 void xfs_buf_readahead_map(struct xfs_buftarg *target,
 			       struct xfs_buf_map *map, int nmaps,
 			       const struct xfs_buf_ops *ops);
 
-static inline struct xfs_buf *
+static inline int
 xfs_buf_get(
 	struct xfs_buftarg	*target,
 	xfs_daddr_t		blkno,
-	size_t			numblks)
+	size_t			numblks,
+	struct xfs_buf		**bpp)
 {
 	DEFINE_SINGLE_BUF_MAP(map, blkno, numblks);
-	return xfs_buf_get_map(target, &map, 1, 0);
+
+	return xfs_buf_get_map(target, &map, 1, 0, bpp);
 }
 
-static inline struct xfs_buf *
+static inline int
 xfs_buf_read(
 	struct xfs_buftarg	*target,
 	xfs_daddr_t		blkno,
 	size_t			numblks,
 	xfs_buf_flags_t		flags,
+	struct xfs_buf		**bpp,
 	const struct xfs_buf_ops *ops)
 {
 	DEFINE_SINGLE_BUF_MAP(map, blkno, numblks);
-	return xfs_buf_read_map(target, &map, 1, flags, ops);
+
+	return xfs_buf_read_map(target, &map, 1, flags, bpp, ops,
+			__builtin_return_address(0));
 }
 
 static inline void
@@ -236,15 +239,14 @@ xfs_buf_readahead(
 	return xfs_buf_readahead_map(target, &map, 1, ops);
 }
 
-struct xfs_buf *xfs_buf_get_uncached(struct xfs_buftarg *target, size_t numblks,
-				int flags);
+int xfs_buf_get_uncached(struct xfs_buftarg *target, size_t numblks, int flags,
+		struct xfs_buf **bpp);
 int xfs_buf_read_uncached(struct xfs_buftarg *target, xfs_daddr_t daddr,
 			  size_t numblks, int flags, struct xfs_buf **bpp,
 			  const struct xfs_buf_ops *ops);
 void xfs_buf_hold(struct xfs_buf *bp);
 
 /* Releasing Buffers */
-extern void xfs_buf_free(xfs_buf_t *);
 extern void xfs_buf_rele(xfs_buf_t *);
 
 /* Locking and Unlocking Buffers */
@@ -260,7 +262,7 @@ extern void xfs_buf_ioend(struct xfs_buf *bp);
 extern void __xfs_buf_ioerror(struct xfs_buf *bp, int error,
 		xfs_failaddr_t failaddr);
 #define xfs_buf_ioerror(bp, err) __xfs_buf_ioerror((bp), (err), __this_address)
-extern void xfs_buf_ioerror_alert(struct xfs_buf *, const char *func);
+extern void xfs_buf_ioerror_alert(struct xfs_buf *bp, xfs_failaddr_t fa);
 
 extern int __xfs_buf_submit(struct xfs_buf *bp, bool);
 static inline int xfs_buf_submit(struct xfs_buf *bp)
@@ -270,6 +272,8 @@ static inline int xfs_buf_submit(struct xfs_buf *bp)
 }
 
 void xfs_buf_zero(struct xfs_buf *bp, size_t boff, size_t bsize);
+void __xfs_buf_mark_corrupt(struct xfs_buf *bp, xfs_failaddr_t fa);
+#define xfs_buf_mark_corrupt(bp) __xfs_buf_mark_corrupt((bp), __this_address)
 
 /* Buffer Utility Routines */
 extern void *xfs_buf_offset(struct xfs_buf *, size_t);
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index d74fbd1e9d3e..1545657c3ca0 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -27,6 +27,23 @@ static inline struct xfs_buf_log_item *BUF_ITEM(struct xfs_log_item *lip)
 
 STATIC void	xfs_buf_do_callbacks(struct xfs_buf *bp);
 
+/* Is this log iovec plausibly large enough to contain the buffer log format? */
+bool
+xfs_buf_log_check_iovec(
+	struct xfs_log_iovec		*iovec)
+{
+	struct xfs_buf_log_format	*blfp = iovec->i_addr;
+	char				*bmp_end;
+	char				*item_end;
+
+	if (offsetof(struct xfs_buf_log_format, blf_data_map) > iovec->i_len)
+		return false;
+
+	item_end = (char *)iovec->i_addr + iovec->i_len;
+	bmp_end = (char *)&blfp->blf_data_map[blfp->blf_map_size];
+	return bmp_end <= item_end;
+}
+
 static inline int
 xfs_buf_log_format_size(
 	struct xfs_buf_log_format *blfp)
@@ -328,7 +345,7 @@ xfs_buf_item_format(
 	 * occurs during recovery.
 	 */
 	if (bip->bli_flags & XFS_BLI_INODE_BUF) {
-		if (xfs_sb_version_hascrc(&lip->li_mountp->m_sb) ||
+		if (xfs_sb_version_has_v3inode(&lip->li_mountp->m_sb) ||
 		    !((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
 		      xfs_log_item_in_current_chkpt(lip)))
 			bip->__bli_format.blf_flags |= XFS_BLF_INODE_BUF;
@@ -688,7 +705,7 @@ static const struct xfs_item_ops xfs_buf_item_ops = {
 	.iop_push	= xfs_buf_item_push,
 };
 
-STATIC int
+STATIC void
 xfs_buf_item_get_format(
 	struct xfs_buf_log_item	*bip,
 	int			count)
@@ -698,14 +715,11 @@ xfs_buf_item_get_format(
 
 	if (count == 1) {
 		bip->bli_formats = &bip->__bli_format;
-		return 0;
+		return;
 	}
 
 	bip->bli_formats = kmem_zalloc(count * sizeof(struct xfs_buf_log_format),
 				0);
-	if (!bip->bli_formats)
-		return -ENOMEM;
-	return 0;
 }
 
 STATIC void
@@ -731,7 +745,6 @@ xfs_buf_item_init(
 	struct xfs_buf_log_item	*bip = bp->b_log_item;
 	int			chunks;
 	int			map_size;
-	int			error;
 	int			i;
 
 	/*
@@ -760,19 +773,22 @@ xfs_buf_item_init(
 	 * Discontiguous buffer support follows the layout of the underlying
 	 * buffer. This makes the implementation as simple as possible.
 	 */
-	error = xfs_buf_item_get_format(bip, bp->b_map_count);
-	ASSERT(error == 0);
-	if (error) {	/* to stop gcc throwing set-but-unused warnings */
-		kmem_zone_free(xfs_buf_item_zone, bip);
-		return error;
-	}
-
+	xfs_buf_item_get_format(bip, bp->b_map_count);
 
 	for (i = 0; i < bip->bli_format_count; i++) {
 		chunks = DIV_ROUND_UP(BBTOB(bp->b_maps[i].bm_len),
 				      XFS_BLF_CHUNK);
 		map_size = DIV_ROUND_UP(chunks, NBWORD);
 
+		if (map_size > XFS_BLF_DATAMAP_SIZE) {
+			kmem_cache_free(xfs_buf_item_zone, bip);
+			xfs_err(mp,
+	"buffer item dirty bitmap (%u uints) too small to reflect %u bytes!",
+					map_size,
+					BBTOB(bp->b_maps[i].bm_len));
+			return -EFSCORRUPTED;
+		}
+
 		bip->bli_formats[i].blf_type = XFS_LI_BUF;
 		bip->bli_formats[i].blf_blkno = bp->b_maps[i].bm_bn;
 		bip->bli_formats[i].blf_len = bp->b_maps[i].bm_len;
@@ -805,6 +821,9 @@ xfs_buf_item_log_segment(
 	uint		end_bit;
 	uint		mask;
 
+	ASSERT(first < XFS_BLF_DATAMAP_SIZE * XFS_BLF_CHUNK * NBWORD);
+	ASSERT(last < XFS_BLF_DATAMAP_SIZE * XFS_BLF_CHUNK * NBWORD);
+
 	/*
 	 * Convert byte offsets to bit numbers.
 	 */
@@ -851,7 +870,7 @@ xfs_buf_item_log_segment(
 	 * first_bit and last_bit.
 	 */
 	while ((bits_to_set - bits_set) >= NBWORD) {
-		*wordp |= 0xffffffff;
+		*wordp = 0xffffffff;
 		bits_set += NBWORD;
 		wordp++;
 	}
@@ -939,7 +958,7 @@ xfs_buf_item_free(
 {
 	xfs_buf_item_free_format(bip);
 	kmem_free(bip->bli_item.li_lv_shadow);
-	kmem_zone_free(xfs_buf_item_zone, bip);
+	kmem_cache_free(xfs_buf_item_zone, bip);
 }
 
 /*
@@ -956,7 +975,7 @@ xfs_buf_item_relse(
 	struct xfs_buf_log_item	*bip = bp->b_log_item;
 
 	trace_xfs_buf_item_relse(bp, _RET_IP_);
-	ASSERT(!(bip->bli_item.li_flags & XFS_LI_IN_AIL));
+	ASSERT(!test_bit(XFS_LI_IN_AIL, &bip->bli_item.li_flags));
 
 	bp->b_log_item = NULL;
 	if (list_empty(&bp->b_li_list))
@@ -1094,7 +1113,7 @@ xfs_buf_iodone_callback_error(
 	if (bp->b_target != lasttarg ||
 	    time_after(jiffies, (lasttime + 5*HZ))) {
 		lasttime = jiffies;
-		xfs_buf_ioerror_alert(bp, __func__);
+		xfs_buf_ioerror_alert(bp, __this_address);
 	}
 	lasttarg = bp->b_target;
 
diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h
index 4a054b11011a..30114b510332 100644
--- a/fs/xfs/xfs_buf_item.h
+++ b/fs/xfs/xfs_buf_item.h
@@ -61,6 +61,7 @@ void	xfs_buf_iodone_callbacks(struct xfs_buf *);
 void	xfs_buf_iodone(struct xfs_buf *, struct xfs_log_item *);
 bool	xfs_buf_resubmit_failed_buffers(struct xfs_buf *,
 					struct list_head *);
+bool	xfs_buf_log_check_iovec(struct xfs_log_iovec *iovec);
 
 extern kmem_zone_t	*xfs_buf_item_zone;
 
diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c
index 283df898dd9f..871ec22c9aee 100644
--- a/fs/xfs/xfs_dir2_readdir.c
+++ b/fs/xfs/xfs_dir2_readdir.c
@@ -17,6 +17,7 @@
 #include "xfs_trace.h"
 #include "xfs_bmap.h"
 #include "xfs_trans.h"
+#include "xfs_error.h"
 
 /*
  * Directory file type support functions
@@ -47,6 +48,7 @@ xfs_dir2_sf_getdents(
 {
 	int			i;		/* shortform entry number */
 	struct xfs_inode	*dp = args->dp;	/* incore directory inode */
+	struct xfs_mount	*mp = dp->i_mount;
 	xfs_dir2_dataptr_t	off;		/* current entry's offset */
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
@@ -68,15 +70,15 @@ xfs_dir2_sf_getdents(
 		return 0;
 
 	/*
-	 * Precalculate offsets for . and .. as we will always need them.
-	 *
-	 * XXX(hch): the second argument is sometimes 0 and sometimes
-	 * geo->datablk
+	 * Precalculate offsets for "." and ".." as we will always need them.
+	 * This relies on the fact that directories always start with the
+	 * entries for "." and "..".
 	 */
 	dot_offset = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
-						dp->d_ops->data_dot_offset);
+			geo->data_entry_offset);
 	dotdot_offset = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
-						dp->d_ops->data_dotdot_offset);
+			geo->data_entry_offset +
+			xfs_dir2_data_entsize(mp, sizeof(".") - 1));
 
 	/*
 	 * Put . entry unless we're starting past it.
@@ -91,7 +93,7 @@ xfs_dir2_sf_getdents(
 	 * Put .. entry unless we're starting past it.
 	 */
 	if (ctx->pos <= dotdot_offset) {
-		ino = dp->d_ops->sf_get_parent_ino(sfp);
+		ino = xfs_dir2_sf_get_parent_ino(sfp);
 		ctx->pos = dotdot_offset & 0x7fffffff;
 		if (!dir_emit(ctx, "..", 2, ino, DT_DIR))
 			return 0;
@@ -108,17 +110,21 @@ xfs_dir2_sf_getdents(
 				xfs_dir2_sf_get_offset(sfep));
 
 		if (ctx->pos > off) {
-			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+			sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 			continue;
 		}
 
-		ino = dp->d_ops->sf_get_ino(sfp, sfep);
-		filetype = dp->d_ops->sf_get_ftype(sfep);
+		ino = xfs_dir2_sf_get_ino(mp, sfp, sfep);
+		filetype = xfs_dir2_sf_get_ftype(mp, sfep);
 		ctx->pos = off & 0x7fffffff;
+		if (XFS_IS_CORRUPT(dp->i_mount,
+				   !xfs_dir2_namecheck(sfep->name,
+						       sfep->namelen)))
+			return -EFSCORRUPTED;
 		if (!dir_emit(ctx, (char *)sfep->name, sfep->namelen, ino,
-			    xfs_dir3_get_dtype(dp->i_mount, filetype)))
+			    xfs_dir3_get_dtype(mp, filetype)))
 			return 0;
-		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
+		sfep = xfs_dir2_sf_nextentry(mp, sfp, sfep);
 	}
 
 	ctx->pos = xfs_dir2_db_off_to_dataptr(geo, geo->datablk + 1, 0) &
@@ -135,17 +141,14 @@ xfs_dir2_block_getdents(
 	struct dir_context	*ctx)
 {
 	struct xfs_inode	*dp = args->dp;	/* incore directory inode */
-	xfs_dir2_data_hdr_t	*hdr;		/* block header */
 	struct xfs_buf		*bp;		/* buffer for block */
-	xfs_dir2_data_entry_t	*dep;		/* block data entry */
-	xfs_dir2_data_unused_t	*dup;		/* block unused entry */
-	char			*endptr;	/* end of the data entries */
 	int			error;		/* error return value */
-	char			*ptr;		/* current data entry */
 	int			wantoff;	/* starting block offset */
 	xfs_off_t		cook;
 	struct xfs_da_geometry	*geo = args->geo;
 	int			lock_mode;
+	unsigned int		offset, next_offset;
+	unsigned int		end;
 
 	/*
 	 * If the block number in the offset is out of range, we're done.
@@ -164,56 +167,57 @@ xfs_dir2_block_getdents(
 	 * We'll skip entries before this.
 	 */
 	wantoff = xfs_dir2_dataptr_to_off(geo, ctx->pos);
-	hdr = bp->b_addr;
 	xfs_dir3_data_check(dp, bp);
-	/*
-	 * Set up values for the loop.
-	 */
-	ptr = (char *)dp->d_ops->data_entry_p(hdr);
-	endptr = xfs_dir3_data_endp(geo, hdr);
 
 	/*
 	 * Loop over the data portion of the block.
 	 * Each object is a real entry (dep) or an unused one (dup).
 	 */
-	while (ptr < endptr) {
+	end = xfs_dir3_data_end_offset(geo, bp->b_addr);
+	for (offset = geo->data_entry_offset;
+	     offset < end;
+	     offset = next_offset) {
+		struct xfs_dir2_data_unused	*dup = bp->b_addr + offset;
+		struct xfs_dir2_data_entry	*dep = bp->b_addr + offset;
 		uint8_t filetype;
 
-		dup = (xfs_dir2_data_unused_t *)ptr;
 		/*
 		 * Unused, skip it.
 		 */
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
-			ptr += be16_to_cpu(dup->length);
+			next_offset = offset + be16_to_cpu(dup->length);
 			continue;
 		}
 
-		dep = (xfs_dir2_data_entry_t *)ptr;
-
 		/*
 		 * Bump pointer for the next iteration.
 		 */
-		ptr += dp->d_ops->data_entsize(dep->namelen);
+		next_offset = offset +
+			xfs_dir2_data_entsize(dp->i_mount, dep->namelen);
+
 		/*
 		 * The entry is before the desired starting point, skip it.
 		 */
-		if ((char *)dep - (char *)hdr < wantoff)
+		if (offset < wantoff)
 			continue;
 
-		cook = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
-					    (char *)dep - (char *)hdr);
+		cook = xfs_dir2_db_off_to_dataptr(geo, geo->datablk, offset);
 
 		ctx->pos = cook & 0x7fffffff;
-		filetype = dp->d_ops->data_get_ftype(dep);
+		filetype = xfs_dir2_data_get_ftype(dp->i_mount, dep);
 		/*
 		 * If it didn't fit, set the final offset to here & return.
 		 */
+		if (XFS_IS_CORRUPT(dp->i_mount,
+				   !xfs_dir2_namecheck(dep->name,
+						       dep->namelen))) {
+			error = -EFSCORRUPTED;
+			goto out_rele;
+		}
 		if (!dir_emit(ctx, (char *)dep->name, dep->namelen,
 			    be64_to_cpu(dep->inumber),
-			    xfs_dir3_get_dtype(dp->i_mount, filetype))) {
-			xfs_trans_brelse(args->trans, bp);
-			return 0;
-		}
+			    xfs_dir3_get_dtype(dp->i_mount, filetype)))
+			goto out_rele;
 	}
 
 	/*
@@ -222,8 +226,9 @@ xfs_dir2_block_getdents(
 	 */
 	ctx->pos = xfs_dir2_db_off_to_dataptr(geo, geo->datablk + 1, 0) &
 								0x7fffffff;
+out_rele:
 	xfs_trans_brelse(args->trans, bp);
-	return 0;
+	return error;
 }
 
 /*
@@ -276,7 +281,7 @@ xfs_dir2_leaf_readbuf(
 	new_off = xfs_dir2_da_to_byte(geo, map.br_startoff);
 	if (new_off > *cur_off)
 		*cur_off = new_off;
-	error = xfs_dir3_data_read(args->trans, dp, map.br_startoff, -1, &bp);
+	error = xfs_dir3_data_read(args->trans, dp, map.br_startoff, 0, &bp);
 	if (error)
 		goto out;
 
@@ -311,7 +316,8 @@ xfs_dir2_leaf_readbuf(
 				break;
 			}
 			if (next_ra > *ra_blk) {
-				xfs_dir3_data_readahead(dp, next_ra, -2);
+				xfs_dir3_data_readahead(dp, next_ra,
+							XFS_DABUF_MAP_HOLE_OK);
 				*ra_blk = next_ra;
 			}
 			ra_want -= geo->fsbcount;
@@ -343,17 +349,17 @@ xfs_dir2_leaf_getdents(
 	size_t			bufsize)
 {
 	struct xfs_inode	*dp = args->dp;
+	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_buf		*bp = NULL;	/* data block buffer */
-	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
 	xfs_dir2_data_entry_t	*dep;		/* data entry */
 	xfs_dir2_data_unused_t	*dup;		/* unused entry */
-	char			*ptr = NULL;	/* pointer to current data */
 	struct xfs_da_geometry	*geo = args->geo;
 	xfs_dablk_t		rablk = 0;	/* current readahead block */
 	xfs_dir2_off_t		curoff;		/* current overall offset */
 	int			length;		/* temporary length value */
 	int			byteoff;	/* offset in current block */
 	int			lock_mode;
+	unsigned int		offset = 0;
 	int			error = 0;	/* error return value */
 
 	/*
@@ -380,7 +386,7 @@ xfs_dir2_leaf_getdents(
 		 * If we have no buffer, or we're off the end of the
 		 * current buffer, need to get another one.
 		 */
-		if (!bp || ptr >= (char *)bp->b_addr + geo->blksize) {
+		if (!bp || offset >= geo->blksize) {
 			if (bp) {
 				xfs_trans_brelse(args->trans, bp);
 				bp = NULL;
@@ -393,36 +399,35 @@ xfs_dir2_leaf_getdents(
 			if (error || !bp)
 				break;
 
-			hdr = bp->b_addr;
 			xfs_dir3_data_check(dp, bp);
 			/*
 			 * Find our position in the block.
 			 */
-			ptr = (char *)dp->d_ops->data_entry_p(hdr);
+			offset = geo->data_entry_offset;
 			byteoff = xfs_dir2_byte_to_off(geo, curoff);
 			/*
 			 * Skip past the header.
 			 */
 			if (byteoff == 0)
-				curoff += dp->d_ops->data_entry_offset;
+				curoff += geo->data_entry_offset;
 			/*
 			 * Skip past entries until we reach our offset.
 			 */
 			else {
-				while ((char *)ptr - (char *)hdr < byteoff) {
-					dup = (xfs_dir2_data_unused_t *)ptr;
+				while (offset < byteoff) {
+					dup = bp->b_addr + offset;
 
 					if (be16_to_cpu(dup->freetag)
 						  == XFS_DIR2_DATA_FREE_TAG) {
 
 						length = be16_to_cpu(dup->length);
-						ptr += length;
+						offset += length;
 						continue;
 					}
-					dep = (xfs_dir2_data_entry_t *)ptr;
-					length =
-					   dp->d_ops->data_entsize(dep->namelen);
-					ptr += length;
+					dep = bp->b_addr + offset;
+					length = xfs_dir2_data_entsize(mp,
+							dep->namelen);
+					offset += length;
 				}
 				/*
 				 * Now set our real offset.
@@ -430,32 +435,38 @@ xfs_dir2_leaf_getdents(
 				curoff =
 					xfs_dir2_db_off_to_byte(geo,
 					    xfs_dir2_byte_to_db(geo, curoff),
-					    (char *)ptr - (char *)hdr);
-				if (ptr >= (char *)hdr + geo->blksize) {
+					    offset);
+				if (offset >= geo->blksize)
 					continue;
-				}
 			}
 		}
+
 		/*
-		 * We have a pointer to an entry.
-		 * Is it a live one?
+		 * We have a pointer to an entry.  Is it a live one?
 		 */
-		dup = (xfs_dir2_data_unused_t *)ptr;
+		dup = bp->b_addr + offset;
+
 		/*
 		 * No, it's unused, skip over it.
 		 */
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
 			length = be16_to_cpu(dup->length);
-			ptr += length;
+			offset += length;
 			curoff += length;
 			continue;
 		}
 
-		dep = (xfs_dir2_data_entry_t *)ptr;
-		length = dp->d_ops->data_entsize(dep->namelen);
-		filetype = dp->d_ops->data_get_ftype(dep);
+		dep = bp->b_addr + offset;
+		length = xfs_dir2_data_entsize(mp, dep->namelen);
+		filetype = xfs_dir2_data_get_ftype(mp, dep);
 
 		ctx->pos = xfs_dir2_byte_to_dataptr(curoff) & 0x7fffffff;
+		if (XFS_IS_CORRUPT(dp->i_mount,
+				   !xfs_dir2_namecheck(dep->name,
+						       dep->namelen))) {
+			error = -EFSCORRUPTED;
+			break;
+		}
 		if (!dir_emit(ctx, (char *)dep->name, dep->namelen,
 			    be64_to_cpu(dep->inumber),
 			    xfs_dir3_get_dtype(dp->i_mount, filetype)))
@@ -464,7 +475,7 @@ xfs_dir2_leaf_getdents(
 		/*
 		 * Advance to next entry in the block.
 		 */
-		ptr += length;
+		offset += length;
 		curoff += length;
 		/* bufsize may have just been a guess; don't go negative */
 		bufsize = bufsize > length ? bufsize - length : 0;
diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c
index 8ec7aab89044..f979d0d7e6cd 100644
--- a/fs/xfs/xfs_discard.c
+++ b/fs/xfs/xfs_discard.c
@@ -13,6 +13,7 @@
 #include "xfs_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
+#include "xfs_discard.h"
 #include "xfs_error.h"
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
@@ -30,6 +31,7 @@ xfs_trim_extents(
 	struct block_device	*bdev = mp->m_ddev_targp->bt_bdev;
 	struct xfs_btree_cur	*cur;
 	struct xfs_buf		*agbp;
+	struct xfs_agf		*agf;
 	struct xfs_perag	*pag;
 	int			error;
 	int			i;
@@ -44,16 +46,16 @@ xfs_trim_extents(
 	xfs_log_force(mp, XFS_LOG_SYNC);
 
 	error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp);
-	if (error || !agbp)
+	if (error)
 		goto out_put_perag;
+	agf = agbp->b_addr;
 
 	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
 
 	/*
 	 * Look up the longest btree in the AGF and start with it.
 	 */
-	error = xfs_alloc_lookup_ge(cur, 0,
-			    be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest), &i);
+	error = xfs_alloc_lookup_ge(cur, 0, be32_to_cpu(agf->agf_longest), &i);
 	if (error)
 		goto out_del_cursor;
 
@@ -70,8 +72,11 @@ xfs_trim_extents(
 		error = xfs_alloc_get_rec(cur, &fbno, &flen, &i);
 		if (error)
 			goto out_del_cursor;
-		XFS_WANT_CORRUPTED_GOTO(mp, i == 1, out_del_cursor);
-		ASSERT(flen <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest));
+		if (XFS_IS_CORRUPT(mp, i != 1)) {
+			error = -EFSCORRUPTED;
+			goto out_del_cursor;
+		}
+		ASSERT(flen <= be32_to_cpu(agf->agf_longest));
 
 		/*
 		 * use daddr format for all range/len calculations as that is
diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index aeb95e7391c1..af2c8e5ceea0 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -48,7 +48,7 @@ static struct lock_class_key xfs_dquot_project_class;
  */
 void
 xfs_qm_dqdestroy(
-	xfs_dquot_t	*dqp)
+	struct xfs_dquot	*dqp)
 {
 	ASSERT(list_empty(&dqp->q_lru));
 
@@ -56,7 +56,7 @@ xfs_qm_dqdestroy(
 	mutex_destroy(&dqp->q_qlock);
 
 	XFS_STATS_DEC(dqp->q_mount, xs_qm_dquot);
-	kmem_zone_free(xfs_qm_dqzone, dqp);
+	kmem_cache_free(xfs_qm_dqzone, dqp);
 }
 
 /*
@@ -113,8 +113,8 @@ xfs_qm_adjust_dqlimits(
  */
 void
 xfs_qm_adjust_dqtimers(
-	xfs_mount_t		*mp,
-	xfs_disk_dquot_t	*d)
+	struct xfs_mount	*mp,
+	struct xfs_disk_dquot	*d)
 {
 	ASSERT(d->d_id);
 
@@ -137,7 +137,7 @@ xfs_qm_adjust_dqtimers(
 		    (d->d_blk_hardlimit &&
 		     (be64_to_cpu(d->d_bcount) >
 		      be64_to_cpu(d->d_blk_hardlimit)))) {
-			d->d_btimer = cpu_to_be32(get_seconds() +
+			d->d_btimer = cpu_to_be32(ktime_get_real_seconds() +
 					mp->m_quotainfo->qi_btimelimit);
 		} else {
 			d->d_bwarns = 0;
@@ -160,7 +160,7 @@ xfs_qm_adjust_dqtimers(
 		    (d->d_ino_hardlimit &&
 		     (be64_to_cpu(d->d_icount) >
 		      be64_to_cpu(d->d_ino_hardlimit)))) {
-			d->d_itimer = cpu_to_be32(get_seconds() +
+			d->d_itimer = cpu_to_be32(ktime_get_real_seconds() +
 					mp->m_quotainfo->qi_itimelimit);
 		} else {
 			d->d_iwarns = 0;
@@ -183,7 +183,7 @@ xfs_qm_adjust_dqtimers(
 		    (d->d_rtb_hardlimit &&
 		     (be64_to_cpu(d->d_rtbcount) >
 		      be64_to_cpu(d->d_rtb_hardlimit)))) {
-			d->d_rtbtimer = cpu_to_be32(get_seconds() +
+			d->d_rtbtimer = cpu_to_be32(ktime_get_real_seconds() +
 					mp->m_quotainfo->qi_rtbtimelimit);
 		} else {
 			d->d_rtbwarns = 0;
@@ -305,8 +305,8 @@ xfs_dquot_disk_alloc(
 	/* Create the block mapping. */
 	xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
 	error = xfs_bmapi_write(tp, quotip, dqp->q_fileoffset,
-			XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA,
-			XFS_QM_DQALLOC_SPACE_RES(mp), &map, &nmaps);
+			XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, 0, &map,
+			&nmaps);
 	if (error)
 		return error;
 	ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
@@ -320,10 +320,10 @@ xfs_dquot_disk_alloc(
 	dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
 
 	/* now we can just get the buffer (there's nothing to read yet) */
-	bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, dqp->q_blkno,
-			mp->m_quotainfo->qi_dqchunklen, 0);
-	if (!bp)
-		return -ENOMEM;
+	error = xfs_trans_get_buf(tp, mp->m_ddev_targp, dqp->q_blkno,
+			mp->m_quotainfo->qi_dqchunklen, 0, &bp);
+	if (error)
+		return error;
 	bp->b_ops = &xfs_dquot_buf_ops;
 
 	/*
@@ -497,7 +497,7 @@ xfs_dquot_from_disk(
 	struct xfs_disk_dquot	*ddqp = bp->b_addr + dqp->q_bufoffset;
 
 	/* copy everything from disk dquot to the incore dquot */
-	memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t));
+	memcpy(&dqp->q_core, ddqp, sizeof(struct xfs_disk_dquot));
 
 	/*
 	 * Reservation counters are defined as reservation plus current usage
@@ -829,11 +829,11 @@ xfs_qm_id_for_quotatype(
 {
 	switch (type) {
 	case XFS_DQ_USER:
-		return ip->i_d.di_uid;
+		return i_uid_read(VFS_I(ip));
 	case XFS_DQ_GROUP:
-		return ip->i_d.di_gid;
+		return i_gid_read(VFS_I(ip));
 	case XFS_DQ_PROJ:
-		return xfs_get_projid(ip);
+		return ip->i_d.di_projid;
 	}
 	ASSERT(0);
 	return 0;
@@ -989,7 +989,7 @@ xfs_qm_dqput(
  */
 void
 xfs_qm_dqrele(
-	xfs_dquot_t	*dqp)
+	struct xfs_dquot	*dqp)
 {
 	if (!dqp)
 		return;
@@ -1018,8 +1018,8 @@ xfs_qm_dqflush_done(
 	struct xfs_buf		*bp,
 	struct xfs_log_item	*lip)
 {
-	xfs_dq_logitem_t	*qip = (struct xfs_dq_logitem *)lip;
-	xfs_dquot_t		*dqp = qip->qli_dquot;
+	struct xfs_dq_logitem	*qip = (struct xfs_dq_logitem *)lip;
+	struct xfs_dquot	*dqp = qip->qli_dquot;
 	struct xfs_ail		*ailp = lip->li_ailp;
 
 	/*
@@ -1105,8 +1105,8 @@ xfs_qm_dqflush(
 	 * Get the buffer containing the on-disk dquot
 	 */
 	error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno,
-				   mp->m_quotainfo->qi_dqchunklen, 0, &bp,
-				   &xfs_dquot_buf_ops);
+				   mp->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK,
+				   &bp, &xfs_dquot_buf_ops);
 	if (error)
 		goto out_unlock;
 
@@ -1126,11 +1126,11 @@ xfs_qm_dqflush(
 		xfs_buf_relse(bp);
 		xfs_dqfunlock(dqp);
 		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/* This is the only portion of data that needs to persist */
-	memcpy(ddqp, &dqp->q_core, sizeof(xfs_disk_dquot_t));
+	memcpy(ddqp, &dqp->q_core, sizeof(struct xfs_disk_dquot));
 
 	/*
 	 * Clear the dirty field and remember the flush lsn for later use.
@@ -1177,7 +1177,7 @@ xfs_qm_dqflush(
 
 out_unlock:
 	xfs_dqfunlock(dqp);
-	return -EIO;
+	return error;
 }
 
 /*
@@ -1188,8 +1188,8 @@ out_unlock:
  */
 void
 xfs_dqlock2(
-	xfs_dquot_t	*d1,
-	xfs_dquot_t	*d2)
+	struct xfs_dquot	*d1,
+	struct xfs_dquot	*d2)
 {
 	if (d1 && d2) {
 		ASSERT(d1 != d2);
@@ -1211,20 +1211,22 @@ xfs_dqlock2(
 int __init
 xfs_qm_init(void)
 {
-	xfs_qm_dqzone =
-		kmem_zone_init(sizeof(struct xfs_dquot), "xfs_dquot");
+	xfs_qm_dqzone = kmem_cache_create("xfs_dquot",
+					  sizeof(struct xfs_dquot),
+					  0, 0, NULL);
 	if (!xfs_qm_dqzone)
 		goto out;
 
-	xfs_qm_dqtrxzone =
-		kmem_zone_init(sizeof(struct xfs_dquot_acct), "xfs_dqtrx");
+	xfs_qm_dqtrxzone = kmem_cache_create("xfs_dqtrx",
+					     sizeof(struct xfs_dquot_acct),
+					     0, 0, NULL);
 	if (!xfs_qm_dqtrxzone)
 		goto out_free_dqzone;
 
 	return 0;
 
 out_free_dqzone:
-	kmem_zone_destroy(xfs_qm_dqzone);
+	kmem_cache_destroy(xfs_qm_dqzone);
 out:
 	return -ENOMEM;
 }
@@ -1232,8 +1234,8 @@ out:
 void
 xfs_qm_exit(void)
 {
-	kmem_zone_destroy(xfs_qm_dqtrxzone);
-	kmem_zone_destroy(xfs_qm_dqzone);
+	kmem_cache_destroy(xfs_qm_dqtrxzone);
+	kmem_cache_destroy(xfs_qm_dqzone);
 }
 
 /*
diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h
index 4fe85709d55d..fe3e46df604b 100644
--- a/fs/xfs/xfs_dquot.h
+++ b/fs/xfs/xfs_dquot.h
@@ -30,33 +30,36 @@ enum {
 /*
  * The incore dquot structure
  */
-typedef struct xfs_dquot {
-	uint		 dq_flags;	/* various flags (XFS_DQ_*) */
-	struct list_head q_lru;		/* global free list of dquots */
-	struct xfs_mount*q_mount;	/* filesystem this relates to */
-	uint		 q_nrefs;	/* # active refs from inodes */
-	xfs_daddr_t	 q_blkno;	/* blkno of dquot buffer */
-	int		 q_bufoffset;	/* off of dq in buffer (# dquots) */
-	xfs_fileoff_t	 q_fileoffset;	/* offset in quotas file */
-
-	xfs_disk_dquot_t q_core;	/* actual usage & quotas */
-	xfs_dq_logitem_t q_logitem;	/* dquot log item */
-	xfs_qcnt_t	 q_res_bcount;	/* total regular nblks used+reserved */
-	xfs_qcnt_t	 q_res_icount;	/* total inos allocd+reserved */
-	xfs_qcnt_t	 q_res_rtbcount;/* total realtime blks used+reserved */
-	xfs_qcnt_t	 q_prealloc_lo_wmark;/* prealloc throttle wmark */
-	xfs_qcnt_t	 q_prealloc_hi_wmark;/* prealloc disabled wmark */
-	int64_t		 q_low_space[XFS_QLOWSP_MAX];
-	struct mutex	 q_qlock;	/* quota lock */
-	struct completion q_flush;	/* flush completion queue */
-	atomic_t          q_pincount;	/* dquot pin count */
-	wait_queue_head_t q_pinwait;	/* dquot pinning wait queue */
-} xfs_dquot_t;
+struct xfs_dquot {
+	uint			dq_flags;
+	struct list_head	q_lru;
+	struct xfs_mount	*q_mount;
+	uint			q_nrefs;
+	xfs_daddr_t		q_blkno;
+	int			q_bufoffset;
+	xfs_fileoff_t		q_fileoffset;
+
+	struct xfs_disk_dquot	q_core;
+	struct xfs_dq_logitem	q_logitem;
+	/* total regular nblks used+reserved */
+	xfs_qcnt_t		q_res_bcount;
+	/* total inos allocd+reserved */
+	xfs_qcnt_t		q_res_icount;
+	/* total realtime blks used+reserved */
+	xfs_qcnt_t		q_res_rtbcount;
+	xfs_qcnt_t		q_prealloc_lo_wmark;
+	xfs_qcnt_t		q_prealloc_hi_wmark;
+	int64_t			q_low_space[XFS_QLOWSP_MAX];
+	struct mutex		q_qlock;
+	struct completion	q_flush;
+	atomic_t		q_pincount;
+	struct wait_queue_head	q_pinwait;
+};
 
 /*
  * Lock hierarchy for q_qlock:
  *	XFS_QLOCK_NORMAL is the implicit default,
- * 	XFS_QLOCK_NESTED is the dquot with the higher id in xfs_dqlock2
+ *	XFS_QLOCK_NESTED is the dquot with the higher id in xfs_dqlock2
  */
 enum {
 	XFS_QLOCK_NORMAL = 0,
@@ -64,21 +67,21 @@ enum {
 };
 
 /*
- * Manage the q_flush completion queue embedded in the dquot.  This completion
+ * Manage the q_flush completion queue embedded in the dquot. This completion
  * queue synchronizes processes attempting to flush the in-core dquot back to
  * disk.
  */
-static inline void xfs_dqflock(xfs_dquot_t *dqp)
+static inline void xfs_dqflock(struct xfs_dquot *dqp)
 {
 	wait_for_completion(&dqp->q_flush);
 }
 
-static inline bool xfs_dqflock_nowait(xfs_dquot_t *dqp)
+static inline bool xfs_dqflock_nowait(struct xfs_dquot *dqp)
 {
 	return try_wait_for_completion(&dqp->q_flush);
 }
 
-static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
+static inline void xfs_dqfunlock(struct xfs_dquot *dqp)
 {
 	complete(&dqp->q_flush);
 }
@@ -112,7 +115,7 @@ static inline int xfs_this_quota_on(struct xfs_mount *mp, int type)
 	}
 }
 
-static inline xfs_dquot_t *xfs_inode_dquot(struct xfs_inode *ip, int type)
+static inline struct xfs_dquot *xfs_inode_dquot(struct xfs_inode *ip, int type)
 {
 	switch (type & XFS_DQ_ALLTYPES) {
 	case XFS_DQ_USER:
@@ -147,31 +150,30 @@ static inline bool xfs_dquot_lowsp(struct xfs_dquot *dqp)
 #define XFS_QM_ISPDQ(dqp)	((dqp)->dq_flags & XFS_DQ_PROJ)
 #define XFS_QM_ISGDQ(dqp)	((dqp)->dq_flags & XFS_DQ_GROUP)
 
-extern void		xfs_qm_dqdestroy(xfs_dquot_t *);
-extern int		xfs_qm_dqflush(struct xfs_dquot *, struct xfs_buf **);
-extern void		xfs_qm_dqunpin_wait(xfs_dquot_t *);
-extern void		xfs_qm_adjust_dqtimers(xfs_mount_t *,
-					xfs_disk_dquot_t *);
-extern void		xfs_qm_adjust_dqlimits(struct xfs_mount *,
-					       struct xfs_dquot *);
-extern xfs_dqid_t	xfs_qm_id_for_quotatype(struct xfs_inode *ip,
-					uint type);
-extern int		xfs_qm_dqget(struct xfs_mount *mp, xfs_dqid_t id,
+void		xfs_qm_dqdestroy(struct xfs_dquot *dqp);
+int		xfs_qm_dqflush(struct xfs_dquot *dqp, struct xfs_buf **bpp);
+void		xfs_qm_dqunpin_wait(struct xfs_dquot *dqp);
+void		xfs_qm_adjust_dqtimers(struct xfs_mount *mp,
+						struct xfs_disk_dquot *d);
+void		xfs_qm_adjust_dqlimits(struct xfs_mount *mp,
+						struct xfs_dquot *d);
+xfs_dqid_t	xfs_qm_id_for_quotatype(struct xfs_inode *ip, uint type);
+int		xfs_qm_dqget(struct xfs_mount *mp, xfs_dqid_t id,
 					uint type, bool can_alloc,
 					struct xfs_dquot **dqpp);
-extern int		xfs_qm_dqget_inode(struct xfs_inode *ip, uint type,
-					bool can_alloc,
-					struct xfs_dquot **dqpp);
-extern int		xfs_qm_dqget_next(struct xfs_mount *mp, xfs_dqid_t id,
+int		xfs_qm_dqget_inode(struct xfs_inode *ip, uint type,
+						bool can_alloc,
+						struct xfs_dquot **dqpp);
+int		xfs_qm_dqget_next(struct xfs_mount *mp, xfs_dqid_t id,
 					uint type, struct xfs_dquot **dqpp);
-extern int		xfs_qm_dqget_uncached(struct xfs_mount *mp,
-					xfs_dqid_t id, uint type,
-					struct xfs_dquot **dqpp);
-extern void		xfs_qm_dqput(xfs_dquot_t *);
+int		xfs_qm_dqget_uncached(struct xfs_mount *mp,
+						xfs_dqid_t id, uint type,
+						struct xfs_dquot **dqpp);
+void		xfs_qm_dqput(struct xfs_dquot *dqp);
 
-extern void		xfs_dqlock2(struct xfs_dquot *, struct xfs_dquot *);
+void		xfs_dqlock2(struct xfs_dquot *, struct xfs_dquot *);
 
-extern void		xfs_dquot_set_prealloc_limits(struct xfs_dquot *);
+void		xfs_dquot_set_prealloc_limits(struct xfs_dquot *);
 
 static inline struct xfs_dquot *xfs_qm_dqhold(struct xfs_dquot *dqp)
 {
diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c
index d60647d7197b..baad1748d0d1 100644
--- a/fs/xfs/xfs_dquot_item.c
+++ b/fs/xfs/xfs_dquot_item.c
@@ -189,7 +189,8 @@ xfs_qm_dquot_logitem_push(
 		if (!xfs_buf_delwri_queue(bp, buffer_list))
 			rval = XFS_ITEM_FLUSHING;
 		xfs_buf_relse(bp);
-	}
+	} else if (error == -EAGAIN)
+		rval = XFS_ITEM_LOCKED;
 
 	spin_lock(&lip->li_ailp->ail_lock);
 out_unlock:
@@ -307,36 +308,62 @@ xfs_qm_qoffend_logitem_committed(
 {
 	struct xfs_qoff_logitem	*qfe = QOFF_ITEM(lip);
 	struct xfs_qoff_logitem	*qfs = qfe->qql_start_lip;
-	struct xfs_ail		*ailp = qfs->qql_item.li_ailp;
 
-	/*
-	 * Delete the qoff-start logitem from the AIL.
-	 * xfs_trans_ail_delete() drops the AIL lock.
-	 */
-	spin_lock(&ailp->ail_lock);
-	xfs_trans_ail_delete(ailp, &qfs->qql_item, SHUTDOWN_LOG_IO_ERROR);
+	xfs_qm_qoff_logitem_relse(qfs);
 
-	kmem_free(qfs->qql_item.li_lv_shadow);
 	kmem_free(lip->li_lv_shadow);
-	kmem_free(qfs);
 	kmem_free(qfe);
 	return (xfs_lsn_t)-1;
 }
 
+STATIC void
+xfs_qm_qoff_logitem_release(
+	struct xfs_log_item	*lip)
+{
+	struct xfs_qoff_logitem	*qoff = QOFF_ITEM(lip);
+
+	if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) {
+		if (qoff->qql_start_lip)
+			xfs_qm_qoff_logitem_relse(qoff->qql_start_lip);
+		xfs_qm_qoff_logitem_relse(qoff);
+	}
+}
+
 static const struct xfs_item_ops xfs_qm_qoffend_logitem_ops = {
 	.iop_size	= xfs_qm_qoff_logitem_size,
 	.iop_format	= xfs_qm_qoff_logitem_format,
 	.iop_committed	= xfs_qm_qoffend_logitem_committed,
 	.iop_push	= xfs_qm_qoff_logitem_push,
+	.iop_release	= xfs_qm_qoff_logitem_release,
 };
 
 static const struct xfs_item_ops xfs_qm_qoff_logitem_ops = {
 	.iop_size	= xfs_qm_qoff_logitem_size,
 	.iop_format	= xfs_qm_qoff_logitem_format,
 	.iop_push	= xfs_qm_qoff_logitem_push,
+	.iop_release	= xfs_qm_qoff_logitem_release,
 };
 
 /*
+ * Delete the quotaoff intent from the AIL and free it. On success,
+ * this should only be called for the start item. It can be used for
+ * either on shutdown or abort.
+ */
+void
+xfs_qm_qoff_logitem_relse(
+	struct xfs_qoff_logitem	*qoff)
+{
+	struct xfs_log_item	*lip = &qoff->qql_item;
+
+	ASSERT(test_bit(XFS_LI_IN_AIL, &lip->li_flags) ||
+	       test_bit(XFS_LI_ABORTED, &lip->li_flags) ||
+	       XFS_FORCED_SHUTDOWN(lip->li_mountp));
+	xfs_trans_ail_remove(lip, SHUTDOWN_LOG_IO_ERROR);
+	kmem_free(lip->li_lv_shadow);
+	kmem_free(qoff);
+}
+
+/*
  * Allocate and initialize an quotaoff item of the correct quota type(s).
  */
 struct xfs_qoff_logitem *
diff --git a/fs/xfs/xfs_dquot_item.h b/fs/xfs/xfs_dquot_item.h
index 1aed34ccdabc..2b86a43d7ce2 100644
--- a/fs/xfs/xfs_dquot_item.h
+++ b/fs/xfs/xfs_dquot_item.h
@@ -11,25 +11,28 @@ struct xfs_trans;
 struct xfs_mount;
 struct xfs_qoff_logitem;
 
-typedef struct xfs_dq_logitem {
-	struct xfs_log_item	 qli_item;	   /* common portion */
-	struct xfs_dquot	*qli_dquot;	   /* dquot ptr */
-	xfs_lsn_t		 qli_flush_lsn;	   /* lsn at last flush */
-} xfs_dq_logitem_t;
+struct xfs_dq_logitem {
+	struct xfs_log_item	qli_item;	/* common portion */
+	struct xfs_dquot	*qli_dquot;	/* dquot ptr */
+	xfs_lsn_t		qli_flush_lsn;	/* lsn at last flush */
+};
 
-typedef struct xfs_qoff_logitem {
-	struct xfs_log_item	 qql_item;	/* common portion */
-	struct xfs_qoff_logitem *qql_start_lip; /* qoff-start logitem, if any */
+struct xfs_qoff_logitem {
+	struct xfs_log_item	qql_item;	/* common portion */
+	struct xfs_qoff_logitem *qql_start_lip;	/* qoff-start logitem, if any */
 	unsigned int		qql_flags;
-} xfs_qoff_logitem_t;
+};
 
 
-extern void		   xfs_qm_dquot_logitem_init(struct xfs_dquot *);
-extern xfs_qoff_logitem_t *xfs_qm_qoff_logitem_init(struct xfs_mount *,
-					struct xfs_qoff_logitem *, uint);
-extern xfs_qoff_logitem_t *xfs_trans_get_qoff_item(struct xfs_trans *,
-					struct xfs_qoff_logitem *, uint);
-extern void		   xfs_trans_log_quotaoff_item(struct xfs_trans *,
-					struct xfs_qoff_logitem *);
+void xfs_qm_dquot_logitem_init(struct xfs_dquot *dqp);
+struct xfs_qoff_logitem	*xfs_qm_qoff_logitem_init(struct xfs_mount *mp,
+		struct xfs_qoff_logitem *start,
+		uint flags);
+void xfs_qm_qoff_logitem_relse(struct xfs_qoff_logitem *);
+struct xfs_qoff_logitem	*xfs_trans_get_qoff_item(struct xfs_trans *tp,
+		struct xfs_qoff_logitem *startqoff,
+		uint flags);
+void xfs_trans_log_quotaoff_item(struct xfs_trans *tp,
+		struct xfs_qoff_logitem *qlp);
 
 #endif	/* __XFS_DQUOT_ITEM_H__ */
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 849fd4476950..a21e9cc6516a 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -257,7 +257,7 @@ xfs_errortag_test(
 
 	xfs_warn_ratelimited(mp,
 "Injecting error (%s) at file %s, line %d, on filesystem \"%s\"",
-			expression, file, line, mp->m_fsname);
+			expression, file, line, mp->m_super->s_id);
 	return true;
 }
 
@@ -329,19 +329,43 @@ xfs_corruption_error(
 	const char		*tag,
 	int			level,
 	struct xfs_mount	*mp,
-	void			*buf,
+	const void		*buf,
 	size_t			bufsize,
 	const char		*filename,
 	int			linenum,
 	xfs_failaddr_t		failaddr)
 {
-	if (level <= xfs_error_level)
+	if (buf && level <= xfs_error_level)
 		xfs_hex_dump(buf, bufsize);
 	xfs_error_report(tag, level, mp, filename, linenum, failaddr);
 	xfs_alert(mp, "Corruption detected. Unmount and run xfs_repair");
 }
 
 /*
+ * Complain about the kinds of metadata corruption that we can't detect from a
+ * verifier, such as incorrect inter-block relationship data.  Does not set
+ * bp->b_error.
+ *
+ * Call xfs_buf_mark_corrupt, not this function.
+ */
+void
+xfs_buf_corruption_error(
+	struct xfs_buf		*bp,
+	xfs_failaddr_t		fa)
+{
+	struct xfs_mount	*mp = bp->b_mount;
+
+	xfs_alert_tag(mp, XFS_PTAG_VERIFIER_ERROR,
+		  "Metadata corruption detected at %pS, %s block 0x%llx",
+		  fa, bp->b_ops->name, bp->b_bn);
+
+	xfs_alert(mp, "Unmount and run xfs_repair");
+
+	if (xfs_error_level >= XFS_ERRLEVEL_HIGH)
+		xfs_stack_trace();
+}
+
+/*
  * Warnings specifically for verifier errors.  Differentiate CRC vs. invalid
  * values, and omit the stack trace unless the error level is tuned high.
  */
@@ -350,7 +374,7 @@ xfs_buf_verifier_error(
 	struct xfs_buf		*bp,
 	int			error,
 	const char		*name,
-	void			*buf,
+	const void		*buf,
 	size_t			bufsz,
 	xfs_failaddr_t		failaddr)
 {
@@ -402,7 +426,7 @@ xfs_inode_verifier_error(
 	struct xfs_inode	*ip,
 	int			error,
 	const char		*name,
-	void			*buf,
+	const void		*buf,
 	size_t			bufsz,
 	xfs_failaddr_t		failaddr)
 {
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index 602aa7d62b66..1717b7508356 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -12,16 +12,17 @@ extern void xfs_error_report(const char *tag, int level, struct xfs_mount *mp,
 			const char *filename, int linenum,
 			xfs_failaddr_t failaddr);
 extern void xfs_corruption_error(const char *tag, int level,
-			struct xfs_mount *mp, void *buf, size_t bufsize,
+			struct xfs_mount *mp, const void *buf, size_t bufsize,
 			const char *filename, int linenum,
 			xfs_failaddr_t failaddr);
+void xfs_buf_corruption_error(struct xfs_buf *bp, xfs_failaddr_t fa);
 extern void xfs_buf_verifier_error(struct xfs_buf *bp, int error,
-			const char *name, void *buf, size_t bufsz,
+			const char *name, const void *buf, size_t bufsz,
 			xfs_failaddr_t failaddr);
 extern void xfs_verifier_error(struct xfs_buf *bp, int error,
 			xfs_failaddr_t failaddr);
 extern void xfs_inode_verifier_error(struct xfs_inode *ip, int error,
-			const char *name, void *buf, size_t bufsz,
+			const char *name, const void *buf, size_t bufsz,
 			xfs_failaddr_t failaddr);
 
 #define	XFS_ERROR_REPORT(e, lvl, mp)	\
@@ -37,32 +38,6 @@ extern void xfs_inode_verifier_error(struct xfs_inode *ip, int error,
 /* Dump 128 bytes of any corrupt buffer */
 #define XFS_CORRUPTION_DUMP_LEN		(128)
 
-/*
- * Macros to set EFSCORRUPTED & return/branch.
- */
-#define	XFS_WANT_CORRUPTED_GOTO(mp, x, l)	\
-	{ \
-		int fs_is_ok = (x); \
-		ASSERT(fs_is_ok); \
-		if (unlikely(!fs_is_ok)) { \
-			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_GOTO", \
-					 XFS_ERRLEVEL_LOW, mp); \
-			error = -EFSCORRUPTED; \
-			goto l; \
-		} \
-	}
-
-#define	XFS_WANT_CORRUPTED_RETURN(mp, x)	\
-	{ \
-		int fs_is_ok = (x); \
-		ASSERT(fs_is_ok); \
-		if (unlikely(!fs_is_ok)) { \
-			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_RETURN", \
-					 XFS_ERRLEVEL_LOW, mp); \
-			return -EFSCORRUPTED; \
-		} \
-	}
-
 #ifdef DEBUG
 extern int xfs_errortag_init(struct xfs_mount *mp);
 extern void xfs_errortag_del(struct xfs_mount *mp);
diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c
index f1372f9046e3..5a4b0119143a 100644
--- a/fs/xfs/xfs_export.c
+++ b/fs/xfs/xfs_export.c
@@ -15,7 +15,6 @@
 #include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_icache.h"
-#include "xfs_log.h"
 #include "xfs_pnfs.h"
 
 /*
@@ -221,18 +220,7 @@ STATIC int
 xfs_fs_nfs_commit_metadata(
 	struct inode		*inode)
 {
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_lsn_t		lsn = 0;
-
-	xfs_ilock(ip, XFS_ILOCK_SHARED);
-	if (xfs_ipincount(ip))
-		lsn = ip->i_itemp->ili_last_lsn;
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
-
-	if (!lsn)
-		return 0;
-	return xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
+	return xfs_log_force_inode(XFS_I(inode));
 }
 
 const struct export_operations xfs_export_operations = {
diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c
index 2183d87be4cf..3991e59cfd18 100644
--- a/fs/xfs/xfs_extent_busy.c
+++ b/fs/xfs/xfs_extent_busy.c
@@ -367,7 +367,7 @@ restart:
 		 * If this is a metadata allocation, try to reuse the busy
 		 * extent instead of trimming the allocation.
 		 */
-		if (!xfs_alloc_is_userdata(args->datatype) &&
+		if (!(args->datatype & XFS_ALLOC_USERDATA) &&
 		    !(busyp->flags & XFS_EXTENT_BUSY_DISCARDED)) {
 			if (!xfs_extent_busy_update_extent(args->mp, args->pag,
 							  busyp, fbno, flen,
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index e44efc41a041..6ea847f6e298 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -21,7 +21,7 @@
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
 #include "xfs_trace.h"
-
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_efi_zone;
 kmem_zone_t	*xfs_efd_zone;
@@ -39,7 +39,7 @@ xfs_efi_item_free(
 	if (efip->efi_format.efi_nextents > XFS_EFI_MAX_FAST_EXTENTS)
 		kmem_free(efip);
 	else
-		kmem_zone_free(xfs_efi_zone, efip);
+		kmem_cache_free(xfs_efi_zone, efip);
 }
 
 /*
@@ -228,6 +228,7 @@ xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt)
 		}
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
@@ -243,7 +244,7 @@ xfs_efd_item_free(struct xfs_efd_log_item *efdp)
 	if (efdp->efd_format.efd_nextents > XFS_EFD_MAX_FAST_EXTENTS)
 		kmem_free(efdp);
 	else
-		kmem_zone_free(xfs_efd_zone, efdp);
+		kmem_cache_free(xfs_efd_zone, efdp);
 }
 
 /*
@@ -624,7 +625,7 @@ xfs_efi_recover(
 			 */
 			set_bit(XFS_EFI_RECOVERED, &efip->efi_flags);
 			xfs_efi_release(efip);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 	}
 
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 1ffb179f35d2..4b8bdecc3863 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -80,19 +80,9 @@ xfs_dir_fsync(
 	int			datasync)
 {
 	struct xfs_inode	*ip = XFS_I(file->f_mapping->host);
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_lsn_t		lsn = 0;
 
 	trace_xfs_dir_fsync(ip);
-
-	xfs_ilock(ip, XFS_ILOCK_SHARED);
-	if (xfs_ipincount(ip))
-		lsn = ip->i_itemp->ili_last_lsn;
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
-
-	if (!lsn)
-		return 0;
-	return xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
+	return xfs_log_force_inode(ip);
 }
 
 STATIC int
@@ -187,8 +177,14 @@ xfs_file_dio_aio_read(
 
 	file_accessed(iocb->ki_filp);
 
-	xfs_ilock(ip, XFS_IOLOCK_SHARED);
-	ret = iomap_dio_rw(iocb, to, &xfs_iomap_ops, NULL);
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED))
+			return -EAGAIN;
+	} else {
+		xfs_ilock(ip, XFS_IOLOCK_SHARED);
+	}
+	ret = iomap_dio_rw(iocb, to, &xfs_read_iomap_ops, NULL,
+			is_sync_kiocb(iocb));
 	xfs_iunlock(ip, XFS_IOLOCK_SHARED);
 
 	return ret;
@@ -215,7 +211,7 @@ xfs_file_dax_read(
 		xfs_ilock(ip, XFS_IOLOCK_SHARED);
 	}
 
-	ret = dax_iomap_rw(iocb, to, &xfs_iomap_ops);
+	ret = dax_iomap_rw(iocb, to, &xfs_read_iomap_ops);
 	xfs_iunlock(ip, XFS_IOLOCK_SHARED);
 
 	file_accessed(iocb->ki_filp);
@@ -351,7 +347,7 @@ restart:
 	
 		trace_xfs_zero_eof(ip, isize, iocb->ki_pos - isize);
 		error = iomap_zero_range(inode, isize, iocb->ki_pos - isize,
-				NULL, &xfs_iomap_ops);
+				NULL, &xfs_buffered_write_iomap_ops);
 		if (error)
 			return error;
 	} else
@@ -486,8 +482,7 @@ xfs_file_dio_aio_write(
 	int			unaligned_io = 0;
 	int			iolock;
 	size_t			count = iov_iter_count(from);
-	struct xfs_buftarg      *target = XFS_IS_REALTIME_INODE(ip) ?
-					mp->m_rtdev_targp : mp->m_ddev_targp;
+	struct xfs_buftarg      *target = xfs_inode_buftarg(ip);
 
 	/* DIO must be aligned to device logical sector size */
 	if ((iocb->ki_pos | count) & target->bt_logical_sectormask)
@@ -547,15 +542,13 @@ xfs_file_dio_aio_write(
 	}
 
 	trace_xfs_file_direct_write(ip, count, iocb->ki_pos);
-	ret = iomap_dio_rw(iocb, from, &xfs_iomap_ops, &xfs_dio_write_ops);
-
 	/*
-	 * If unaligned, this is the only IO in-flight. If it has not yet
-	 * completed, wait on it before we release the iolock to prevent
-	 * subsequent overlapping IO.
+	 * If unaligned, this is the only IO in-flight. Wait on it before we
+	 * release the iolock to prevent subsequent overlapping IO.
 	 */
-	if (ret == -EIOCBQUEUED && unaligned_io)
-		inode_dio_wait(inode);
+	ret = iomap_dio_rw(iocb, from, &xfs_direct_write_iomap_ops,
+			   &xfs_dio_write_ops,
+			   is_sync_kiocb(iocb) || unaligned_io);
 out:
 	xfs_iunlock(ip, iolock);
 
@@ -594,7 +587,7 @@ xfs_file_dax_write(
 	count = iov_iter_count(from);
 
 	trace_xfs_file_dax_write(ip, count, pos);
-	ret = dax_iomap_rw(iocb, from, &xfs_iomap_ops);
+	ret = dax_iomap_rw(iocb, from, &xfs_direct_write_iomap_ops);
 	if (ret > 0 && iocb->ki_pos > i_size_read(inode)) {
 		i_size_write(inode, iocb->ki_pos);
 		error = xfs_setfilesize(ip, pos, ret);
@@ -641,7 +634,8 @@ write_retry:
 	current->backing_dev_info = inode_to_bdi(inode);
 
 	trace_xfs_file_buffered_write(ip, iov_iter_count(from), iocb->ki_pos);
-	ret = iomap_file_buffered_write(iocb, from, &xfs_iomap_ops);
+	ret = iomap_file_buffered_write(iocb, from,
+			&xfs_buffered_write_iomap_ops);
 	if (likely(ret >= 0))
 		iocb->ki_pos += ret;
 
@@ -818,6 +812,36 @@ xfs_file_fallocate(
 	if (error)
 		goto out_unlock;
 
+	/*
+	 * Must wait for all AIO to complete before we continue as AIO can
+	 * change the file size on completion without holding any locks we
+	 * currently hold. We must do this first because AIO can update both
+	 * the on disk and in memory inode sizes, and the operations that follow
+	 * require the in-memory size to be fully up-to-date.
+	 */
+	inode_dio_wait(inode);
+
+	/*
+	 * Now AIO and DIO has drained we flush and (if necessary) invalidate
+	 * the cached range over the first operation we are about to run.
+	 *
+	 * We care about zero and collapse here because they both run a hole
+	 * punch over the range first. Because that can zero data, and the range
+	 * of invalidation for the shift operations is much larger, we still do
+	 * the required flush for collapse in xfs_prepare_shift().
+	 *
+	 * Insert has the same range requirements as collapse, and we extend the
+	 * file first which can zero data. Hence insert has the same
+	 * flush/invalidate requirements as collapse and so they are both
+	 * handled at the right time by xfs_prepare_shift().
+	 */
+	if (mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_ZERO_RANGE |
+		    FALLOC_FL_COLLAPSE_RANGE)) {
+		error = xfs_flush_unmap_range(ip, offset, len);
+		if (error)
+			goto out_unlock;
+	}
+
 	if (mode & FALLOC_FL_PUNCH_HOLE) {
 		error = xfs_free_file_space(ip, offset, len);
 		if (error)
@@ -881,16 +905,30 @@ xfs_file_fallocate(
 		}
 
 		if (mode & FALLOC_FL_ZERO_RANGE) {
-			error = xfs_zero_file_space(ip, offset, len);
+			/*
+			 * Punch a hole and prealloc the range.  We use a hole
+			 * punch rather than unwritten extent conversion for two
+			 * reasons:
+			 *
+			 *   1.) Hole punch handles partial block zeroing for us.
+			 *   2.) If prealloc returns ENOSPC, the file range is
+			 *       still zero-valued by virtue of the hole punch.
+			 */
+			unsigned int blksize = i_blocksize(inode);
+
+			trace_xfs_zero_file_space(ip);
+
+			error = xfs_free_file_space(ip, offset, len);
+			if (error)
+				goto out_unlock;
+
+			len = round_up(offset + len, blksize) -
+			      round_down(offset, blksize);
+			offset = round_down(offset, blksize);
 		} else if (mode & FALLOC_FL_UNSHARE_RANGE) {
 			error = xfs_reflink_unshare(ip, offset, len);
 			if (error)
 				goto out_unlock;
-
-			if (!xfs_is_always_cow_inode(ip)) {
-				error = xfs_alloc_file_space(ip, offset, len,
-						XFS_BMAPI_PREALLOC);
-			}
 		} else {
 			/*
 			 * If always_cow mode we can't use preallocations and
@@ -900,12 +938,14 @@ xfs_file_fallocate(
 				error = -EOPNOTSUPP;
 				goto out_unlock;
 			}
+		}
 
+		if (!xfs_is_always_cow_inode(ip)) {
 			error = xfs_alloc_file_space(ip, offset, len,
 						     XFS_BMAPI_PREALLOC);
+			if (error)
+				goto out_unlock;
 		}
-		if (error)
-			goto out_unlock;
 	}
 
 	if (file->f_flags & O_DSYNC)
@@ -1019,7 +1059,11 @@ xfs_file_remap_range(
 
 	ret = xfs_reflink_update_dest(dest, pos_out + len, cowextsize,
 			remap_flags);
+	if (ret)
+		goto out_unlock;
 
+	if (mp->m_flags & XFS_MOUNT_WSYNC)
+		xfs_log_force_inode(dest);
 out_unlock:
 	xfs_reflink_remap_unlock(file_in, file_out);
 	if (ret)
@@ -1059,7 +1103,7 @@ xfs_dir_open(
 	 */
 	mode = xfs_ilock_data_map_shared(ip);
 	if (ip->i_d.di_nextents > 0)
-		error = xfs_dir3_data_readahead(ip, 0, -1);
+		error = xfs_dir3_data_readahead(ip, 0, 0);
 	xfs_iunlock(ip, mode);
 	return error;
 }
@@ -1156,12 +1200,16 @@ __xfs_filemap_fault(
 	if (IS_DAX(inode)) {
 		pfn_t pfn;
 
-		ret = dax_iomap_fault(vmf, pe_size, &pfn, NULL, &xfs_iomap_ops);
+		ret = dax_iomap_fault(vmf, pe_size, &pfn, NULL,
+				(write_fault && !vmf->cow_page) ?
+				 &xfs_direct_write_iomap_ops :
+				 &xfs_read_iomap_ops);
 		if (ret & VM_FAULT_NEEDDSYNC)
 			ret = dax_finish_sync_fault(vmf, pe_size, pfn);
 	} else {
 		if (write_fault)
-			ret = iomap_page_mkwrite(vmf, &xfs_iomap_ops);
+			ret = iomap_page_mkwrite(vmf,
+					&xfs_buffered_write_iomap_ops);
 		else
 			ret = filemap_fault(vmf);
 	}
@@ -1225,22 +1273,22 @@ static const struct vm_operations_struct xfs_file_vm_ops = {
 
 STATIC int
 xfs_file_mmap(
-	struct file	*filp,
-	struct vm_area_struct *vma)
+	struct file		*file,
+	struct vm_area_struct	*vma)
 {
-	struct dax_device 	*dax_dev;
+	struct inode		*inode = file_inode(file);
+	struct xfs_buftarg	*target = xfs_inode_buftarg(XFS_I(inode));
 
-	dax_dev = xfs_find_daxdev_for_inode(file_inode(filp));
 	/*
 	 * We don't support synchronous mappings for non-DAX files and
 	 * for DAX files if underneath dax_device is not synchronous.
 	 */
-	if (!daxdev_mapping_supported(vma, dax_dev))
+	if (!daxdev_mapping_supported(vma, target->bt_daxdev))
 		return -EOPNOTSUPP;
 
-	file_accessed(filp);
+	file_accessed(file);
 	vma->vm_ops = &xfs_file_vm_ops;
-	if (IS_DAX(file_inode(filp)))
+	if (IS_DAX(inode))
 		vma->vm_flags |= VM_HUGEPAGE;
 	return 0;
 }
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index 574a7a8b4736..1a88025e68a3 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -18,6 +18,7 @@
 #include "xfs_trace.h"
 #include "xfs_ag_resv.h"
 #include "xfs_trans.h"
+#include "xfs_filestream.h"
 
 struct xfs_fstrm_item {
 	struct xfs_mru_cache_elem	mru;
@@ -158,16 +159,15 @@ xfs_filestream_pick_ag(
 
 		if (!pag->pagf_init) {
 			err = xfs_alloc_pagf_init(mp, NULL, ag, trylock);
-			if (err && !trylock) {
+			if (err) {
 				xfs_perag_put(pag);
-				return err;
+				if (err != -EAGAIN)
+					return err;
+				/* Couldn't lock the AGF, skip this AG. */
+				continue;
 			}
 		}
 
-		/* Might fail sometimes during the 1st pass with trylock set. */
-		if (!pag->pagf_init)
-			goto next_ag;
-
 		/* Keep track of the AG with the most free blocks. */
 		if (pag->pagf_freeblks > maxfree) {
 			maxfree = pag->pagf_freeblks;
@@ -374,7 +374,7 @@ xfs_filestream_new_ag(
 		startag = (item->ag + 1) % mp->m_sb.sb_agcount;
 	}
 
-	if (xfs_alloc_is_userdata(ap->datatype))
+	if (ap->datatype & XFS_ALLOC_USERDATA)
 		flags |= XFS_PICK_USERDATA;
 	if (ap->tp->t_flags & XFS_TRANS_LOWMODE)
 		flags |= XFS_PICK_LOWSPACE;
diff --git a/fs/xfs/xfs_fsmap.c b/fs/xfs/xfs_fsmap.c
index d082143feb5a..4eebcec4aae6 100644
--- a/fs/xfs/xfs_fsmap.c
+++ b/fs/xfs/xfs_fsmap.c
@@ -146,6 +146,7 @@ xfs_fsmap_owner_from_rmap(
 		dest->fmr_owner = XFS_FMR_OWN_FREE;
 		break;
 	default:
+		ASSERT(0);
 		return -EFSCORRUPTED;
 	}
 	return 0;
@@ -343,7 +344,7 @@ xfs_getfsmap_datadev_helper(
 	xfs_fsblock_t			fsb;
 	xfs_daddr_t			rec_daddr;
 
-	fsb = XFS_AGB_TO_FSB(mp, cur->bc_private.a.agno, rec->rm_startblock);
+	fsb = XFS_AGB_TO_FSB(mp, cur->bc_ag.agno, rec->rm_startblock);
 	rec_daddr = XFS_FSB_TO_DADDR(mp, fsb);
 
 	return xfs_getfsmap_helper(cur->bc_tp, info, rec, rec_daddr);
@@ -361,7 +362,7 @@ xfs_getfsmap_datadev_bnobt_helper(
 	struct xfs_rmap_irec		irec;
 	xfs_daddr_t			rec_daddr;
 
-	rec_daddr = XFS_AGB_TO_DADDR(mp, cur->bc_private.a.agno,
+	rec_daddr = XFS_AGB_TO_DADDR(mp, cur->bc_ag.agno,
 			rec->ar_startblock);
 
 	irec.rm_startblock = rec->ar_startblock;
@@ -895,6 +896,14 @@ xfs_getfsmap(
 	info.format_arg = arg;
 	info.head = head;
 
+	/*
+	 * If fsmap runs concurrently with a scrub, the freeze can be delayed
+	 * indefinitely as we walk the rmapbt and iterate over metadata
+	 * buffers.  Freeze quiesces the log (which waits for the buffer LRU to
+	 * be emptied) and that won't happen while we're reading buffers.
+	 */
+	sb_start_write(mp->m_super);
+
 	/* For each device we support... */
 	for (i = 0; i < XFS_GETFSMAP_DEVS; i++) {
 		/* Is this device within the range the user asked for? */
@@ -934,6 +943,7 @@ xfs_getfsmap(
 
 	if (tp)
 		xfs_trans_cancel(tp);
+	sb_end_write(mp->m_super);
 	head->fmh_oflags = FMH_OF_DEV_T;
 	return error;
 }
diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index 944add5ff8e0..8bf1d15be3f6 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -44,7 +44,7 @@ xfs_inode_alloc(
 	if (!ip)
 		return NULL;
 	if (inode_init_always(mp->m_super, VFS_I(ip))) {
-		kmem_zone_free(xfs_inode_zone, ip);
+		kmem_cache_free(xfs_inode_zone, ip);
 		return NULL;
 	}
 
@@ -104,7 +104,7 @@ xfs_inode_free_callback(
 		ip->i_itemp = NULL;
 	}
 
-	kmem_zone_free(xfs_inode_zone, ip);
+	kmem_cache_free(xfs_inode_zone, ip);
 }
 
 static void
@@ -289,6 +289,8 @@ xfs_reinit_inode(
 	uint64_t	version = inode_peek_iversion(inode);
 	umode_t		mode = inode->i_mode;
 	dev_t		dev = inode->i_rdev;
+	kuid_t		uid = inode->i_uid;
+	kgid_t		gid = inode->i_gid;
 
 	error = inode_init_always(mp->m_super, inode);
 
@@ -297,6 +299,8 @@ xfs_reinit_inode(
 	inode_set_iversion_queried(inode, version);
 	inode->i_mode = mode;
 	inode->i_rdev = dev;
+	inode->i_uid = uid;
+	inode->i_gid = gid;
 	return error;
 }
 
@@ -907,7 +911,12 @@ xfs_eofblocks_worker(
 {
 	struct xfs_mount *mp = container_of(to_delayed_work(work),
 				struct xfs_mount, m_eofblocks_work);
+
+	if (!sb_start_write_trylock(mp->m_super))
+		return;
 	xfs_icache_free_eofblocks(mp, NULL);
+	sb_end_write(mp->m_super);
+
 	xfs_queue_eofblocks(mp);
 }
 
@@ -934,7 +943,12 @@ xfs_cowblocks_worker(
 {
 	struct xfs_mount *mp = container_of(to_delayed_work(work),
 				struct xfs_mount, m_cowblocks_work);
+
+	if (!sb_start_write_trylock(mp->m_super))
+		return;
 	xfs_icache_free_cowblocks(mp, NULL);
+	sb_end_write(mp->m_super);
+
 	xfs_queue_cowblocks(mp);
 }
 
@@ -1419,7 +1433,7 @@ xfs_inode_match_id(
 		return 0;
 
 	if ((eofb->eof_flags & XFS_EOF_FLAGS_PRID) &&
-	    xfs_get_projid(ip) != eofb->eof_prid)
+	    ip->i_d.di_projid != eofb->eof_prid)
 		return 0;
 
 	return 1;
@@ -1443,7 +1457,7 @@ xfs_inode_match_id_union(
 		return 1;
 
 	if ((eofb->eof_flags & XFS_EOF_FLAGS_PRID) &&
-	    xfs_get_projid(ip) == eofb->eof_prid)
+	    ip->i_d.di_projid == eofb->eof_prid)
 		return 1;
 
 	return 0;
diff --git a/fs/xfs/xfs_icreate_item.c b/fs/xfs/xfs_icreate_item.c
index 3ebd1b7f49d8..490fee22b878 100644
--- a/fs/xfs/xfs_icreate_item.c
+++ b/fs/xfs/xfs_icreate_item.c
@@ -55,7 +55,7 @@ STATIC void
 xfs_icreate_item_release(
 	struct xfs_log_item	*lip)
 {
-	kmem_zone_free(xfs_icreate_zone, ICR_ITEM(lip));
+	kmem_cache_free(xfs_icreate_zone, ICR_ITEM(lip));
 }
 
 static const struct xfs_item_ops xfs_icreate_item_ops = {
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 18f4b262e61c..d1772786af29 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -55,6 +55,12 @@ xfs_extlen_t
 xfs_get_extsz_hint(
 	struct xfs_inode	*ip)
 {
+	/*
+	 * No point in aligning allocations if we need to COW to actually
+	 * write to them.
+	 */
+	if (xfs_is_always_cow_inode(ip))
+		return 0;
 	if ((ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE) && ip->i_d.di_extsize)
 		return ip->i_d.di_extsize;
 	if (XFS_IS_REALTIME_INODE(ip))
@@ -795,26 +801,18 @@ xfs_ialloc(
 		return error;
 	ASSERT(ip != NULL);
 	inode = VFS_I(ip);
-
-	/*
-	 * We always convert v1 inodes to v2 now - we only support filesystems
-	 * with >= v2 inode capability, so there is no reason for ever leaving
-	 * an inode in v1 format.
-	 */
-	if (ip->i_d.di_version == 1)
-		ip->i_d.di_version = 2;
-
 	inode->i_mode = mode;
 	set_nlink(inode, nlink);
-	ip->i_d.di_uid = xfs_kuid_to_uid(current_fsuid());
-	ip->i_d.di_gid = xfs_kgid_to_gid(current_fsgid());
+	inode->i_uid = current_fsuid();
 	inode->i_rdev = rdev;
-	xfs_set_projid(ip, prid);
+	ip->i_d.di_projid = prid;
 
 	if (pip && XFS_INHERIT_GID(pip)) {
-		ip->i_d.di_gid = pip->i_d.di_gid;
+		inode->i_gid = VFS_I(pip)->i_gid;
 		if ((VFS_I(pip)->i_mode & S_ISGID) && S_ISDIR(mode))
 			inode->i_mode |= S_ISGID;
+	} else {
+		inode->i_gid = current_fsgid();
 	}
 
 	/*
@@ -822,9 +820,8 @@ xfs_ialloc(
 	 * ID or one of the supplementary group IDs, the S_ISGID bit is cleared
 	 * (and only if the irix_sgid_inherit compatibility variable is set).
 	 */
-	if ((irix_sgid_inherit) &&
-	    (inode->i_mode & S_ISGID) &&
-	    (!in_group_p(xfs_gid_to_kgid(ip->i_d.di_gid))))
+	if (irix_sgid_inherit &&
+	    (inode->i_mode & S_ISGID) && !in_group_p(inode->i_gid))
 		inode->i_mode &= ~S_ISGID;
 
 	ip->i_d.di_size = 0;
@@ -841,15 +838,13 @@ xfs_ialloc(
 	ip->i_d.di_dmstate = 0;
 	ip->i_d.di_flags = 0;
 
-	if (ip->i_d.di_version == 3) {
+	if (xfs_sb_version_has_v3inode(&mp->m_sb)) {
 		inode_set_iversion(inode, 1);
 		ip->i_d.di_flags2 = 0;
 		ip->i_d.di_cowextsize = 0;
-		ip->i_d.di_crtime.t_sec = (int32_t)tv.tv_sec;
-		ip->i_d.di_crtime.t_nsec = (int32_t)tv.tv_nsec;
+		ip->i_d.di_crtime = tv;
 	}
 
-
 	flags = XFS_ILOG_CORE;
 	switch (mode & S_IFMT) {
 	case S_IFIFO:
@@ -902,20 +897,13 @@ xfs_ialloc(
 
 			ip->i_d.di_flags |= di_flags;
 		}
-		if (pip &&
-		    (pip->i_d.di_flags2 & XFS_DIFLAG2_ANY) &&
-		    pip->i_d.di_version == 3 &&
-		    ip->i_d.di_version == 3) {
-			uint64_t	di_flags2 = 0;
-
+		if (pip && (pip->i_d.di_flags2 & XFS_DIFLAG2_ANY)) {
 			if (pip->i_d.di_flags2 & XFS_DIFLAG2_COWEXTSIZE) {
-				di_flags2 |= XFS_DIFLAG2_COWEXTSIZE;
+				ip->i_d.di_flags2 |= XFS_DIFLAG2_COWEXTSIZE;
 				ip->i_d.di_cowextsize = pip->i_d.di_cowextsize;
 			}
 			if (pip->i_d.di_flags2 & XFS_DIFLAG2_DAX)
-				di_flags2 |= XFS_DIFLAG2_DAX;
-
-			ip->i_d.di_flags2 |= di_flags2;
+				ip->i_d.di_flags2 |= XFS_DIFLAG2_DAX;
 		}
 		/* FALLTHROUGH */
 	case S_IFLNK:
@@ -1117,7 +1105,6 @@ xfs_bumplink(
 {
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
 
-	ASSERT(ip->i_d.di_version > 1);
 	inc_nlink(VFS_I(ip));
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 }
@@ -1153,8 +1140,7 @@ xfs_create(
 	/*
 	 * Make sure that we have allocated dquot(s) on disk.
 	 */
-	error = xfs_qm_vop_dqalloc(dp, xfs_kuid_to_uid(current_fsuid()),
-					xfs_kgid_to_gid(current_fsgid()), prid,
+	error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid,
 					XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT,
 					&udqp, &gdqp, &pdqp);
 	if (error)
@@ -1214,8 +1200,7 @@ xfs_create(
 	unlock_dp_on_error = false;
 
 	error = xfs_dir_createname(tp, dp, name, ip->i_ino,
-				   resblks ?
-					resblks - XFS_IALLOC_SPACE_RES(mp) : 0);
+					resblks - XFS_IALLOC_SPACE_RES(mp));
 	if (error) {
 		ASSERT(error != -ENOSPC);
 		goto out_trans_cancel;
@@ -1304,8 +1289,7 @@ xfs_create_tmpfile(
 	/*
 	 * Make sure that we have allocated dquot(s) on disk.
 	 */
-	error = xfs_qm_vop_dqalloc(dp, xfs_kuid_to_uid(current_fsuid()),
-				xfs_kgid_to_gid(current_fsgid()), prid,
+	error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid,
 				XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT,
 				&udqp, &gdqp, &pdqp);
 	if (error)
@@ -1418,7 +1402,7 @@ xfs_link(
 	 * the tree quota mechanism could be circumvented.
 	 */
 	if (unlikely((tdp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
-		     (xfs_get_projid(tdp) != xfs_get_projid(sip)))) {
+		     tdp->i_d.di_projid != sip->i_d.di_projid)) {
 		error = -EXDEV;
 		goto error_return;
 	}
@@ -1513,10 +1497,8 @@ xfs_itruncate_extents_flags(
 	struct xfs_mount	*mp = ip->i_mount;
 	struct xfs_trans	*tp = *tpp;
 	xfs_fileoff_t		first_unmap_block;
-	xfs_fileoff_t		last_block;
 	xfs_filblks_t		unmap_len;
 	int			error = 0;
-	int			done = 0;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	ASSERT(!atomic_read(&VFS_I(ip)->i_count) ||
@@ -1536,21 +1518,22 @@ xfs_itruncate_extents_flags(
 	 * the end of the file (in a crash where the space is allocated
 	 * but the inode size is not yet updated), simply remove any
 	 * blocks which show up between the new EOF and the maximum
-	 * possible file size.  If the first block to be removed is
-	 * beyond the maximum file size (ie it is the same as last_block),
-	 * then there is nothing to do.
+	 * possible file size.
+	 *
+	 * We have to free all the blocks to the bmbt maximum offset, even if
+	 * the page cache can't scale that far.
 	 */
 	first_unmap_block = XFS_B_TO_FSB(mp, (xfs_ufsize_t)new_size);
-	last_block = XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes);
-	if (first_unmap_block == last_block)
+	if (first_unmap_block >= XFS_MAX_FILEOFF) {
+		WARN_ON_ONCE(first_unmap_block > XFS_MAX_FILEOFF);
 		return 0;
+	}
 
-	ASSERT(first_unmap_block < last_block);
-	unmap_len = last_block - first_unmap_block + 1;
-	while (!done) {
+	unmap_len = XFS_MAX_FILEOFF - first_unmap_block + 1;
+	while (unmap_len > 0) {
 		ASSERT(tp->t_firstblock == NULLFSBLOCK);
-		error = xfs_bunmapi(tp, ip, first_unmap_block, unmap_len, flags,
-				    XFS_ITRUNC_MAX_EXTENTS, &done);
+		error = __xfs_bunmapi(tp, ip, first_unmap_block, &unmap_len,
+				flags, XFS_ITRUNC_MAX_EXTENTS);
 		if (error)
 			goto out;
 
@@ -1570,7 +1553,7 @@ xfs_itruncate_extents_flags(
 	if (whichfork == XFS_DATA_FORK) {
 		/* Remove all pending CoW reservations. */
 		error = xfs_reflink_cancel_cow_blocks(ip, &tp,
-				first_unmap_block, last_block, true);
+				first_unmap_block, XFS_MAX_FILEOFF, true);
 		if (error)
 			goto out;
 
@@ -2115,7 +2098,7 @@ xfs_iunlink_update_bucket(
 	unsigned int		bucket_index,
 	xfs_agino_t		new_agino)
 {
-	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agibp);
+	struct xfs_agi		*agi = agibp->b_addr;
 	xfs_agino_t		old_value;
 	int			offset;
 
@@ -2130,8 +2113,10 @@ xfs_iunlink_update_bucket(
 	 * passed in because either we're adding or removing ourselves from the
 	 * head of the list.
 	 */
-	if (old_value == new_agino)
+	if (old_value == new_agino) {
+		xfs_buf_mark_corrupt(agibp);
 		return -EFSCORRUPTED;
+	}
 
 	agi->agi_unlinked[bucket_index] = cpu_to_be32(new_agino);
 	offset = offsetof(struct xfs_agi, agi_unlinked) +
@@ -2194,6 +2179,8 @@ xfs_iunlink_update_inode(
 	/* Make sure the old pointer isn't garbage. */
 	old_value = be32_to_cpu(dip->di_next_unlinked);
 	if (!xfs_verify_agino_or_null(mp, agno, old_value)) {
+		xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__, dip,
+				sizeof(*dip), __this_address);
 		error = -EFSCORRUPTED;
 		goto out;
 	}
@@ -2205,8 +2192,11 @@ xfs_iunlink_update_inode(
 	 */
 	*old_next_agino = old_value;
 	if (old_value == next_agino) {
-		if (next_agino != NULLAGINO)
+		if (next_agino != NULLAGINO) {
+			xfs_inode_verifier_error(ip, -EFSCORRUPTED, __func__,
+					dip, sizeof(*dip), __this_address);
 			error = -EFSCORRUPTED;
+		}
 		goto out;
 	}
 
@@ -2248,7 +2238,7 @@ xfs_iunlink(
 	error = xfs_read_agi(mp, tp, agno, &agibp);
 	if (error)
 		return error;
-	agi = XFS_BUF_TO_AGI(agibp);
+	agi = agibp->b_addr;
 
 	/*
 	 * Get the index into the agi hash table for the list this inode will
@@ -2257,8 +2247,10 @@ xfs_iunlink(
 	 */
 	next_agino = be32_to_cpu(agi->agi_unlinked[bucket_index]);
 	if (next_agino == agino ||
-	    !xfs_verify_agino_or_null(mp, agno, next_agino))
+	    !xfs_verify_agino_or_null(mp, agno, next_agino)) {
+		xfs_buf_mark_corrupt(agibp);
 		return -EFSCORRUPTED;
+	}
 
 	if (next_agino != NULLAGINO) {
 		struct xfs_perag	*pag;
@@ -2430,7 +2422,7 @@ xfs_iunlink_remove(
 	error = xfs_read_agi(mp, tp, agno, &agibp);
 	if (error)
 		return error;
-	agi = XFS_BUF_TO_AGI(agibp);
+	agi = agibp->b_addr;
 
 	/*
 	 * Get the index into the agi hash table for the list this inode will
@@ -2511,6 +2503,88 @@ out:
 }
 
 /*
+ * Look up the inode number specified and mark it stale if it is found. If it is
+ * dirty, return the inode so it can be attached to the cluster buffer so it can
+ * be processed appropriately when the cluster free transaction completes.
+ */
+static struct xfs_inode *
+xfs_ifree_get_one_inode(
+	struct xfs_perag	*pag,
+	struct xfs_inode	*free_ip,
+	xfs_ino_t		inum)
+{
+	struct xfs_mount	*mp = pag->pag_mount;
+	struct xfs_inode	*ip;
+
+retry:
+	rcu_read_lock();
+	ip = radix_tree_lookup(&pag->pag_ici_root, XFS_INO_TO_AGINO(mp, inum));
+
+	/* Inode not in memory, nothing to do */
+	if (!ip)
+		goto out_rcu_unlock;
+
+	/*
+	 * because this is an RCU protected lookup, we could find a recently
+	 * freed or even reallocated inode during the lookup. We need to check
+	 * under the i_flags_lock for a valid inode here. Skip it if it is not
+	 * valid, the wrong inode or stale.
+	 */
+	spin_lock(&ip->i_flags_lock);
+	if (ip->i_ino != inum || __xfs_iflags_test(ip, XFS_ISTALE)) {
+		spin_unlock(&ip->i_flags_lock);
+		goto out_rcu_unlock;
+	}
+	spin_unlock(&ip->i_flags_lock);
+
+	/*
+	 * Don't try to lock/unlock the current inode, but we _cannot_ skip the
+	 * other inodes that we did not find in the list attached to the buffer
+	 * and are not already marked stale. If we can't lock it, back off and
+	 * retry.
+	 */
+	if (ip != free_ip) {
+		if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) {
+			rcu_read_unlock();
+			delay(1);
+			goto retry;
+		}
+
+		/*
+		 * Check the inode number again in case we're racing with
+		 * freeing in xfs_reclaim_inode().  See the comments in that
+		 * function for more information as to why the initial check is
+		 * not sufficient.
+		 */
+		if (ip->i_ino != inum) {
+			xfs_iunlock(ip, XFS_ILOCK_EXCL);
+			goto out_rcu_unlock;
+		}
+	}
+	rcu_read_unlock();
+
+	xfs_iflock(ip);
+	xfs_iflags_set(ip, XFS_ISTALE);
+
+	/*
+	 * We don't need to attach clean inodes or those only with unlogged
+	 * changes (which we throw away, anyway).
+	 */
+	if (!ip->i_itemp || xfs_inode_clean(ip)) {
+		ASSERT(ip != free_ip);
+		xfs_ifunlock(ip);
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
+		goto out_no_inode;
+	}
+	return ip;
+
+out_rcu_unlock:
+	rcu_read_unlock();
+out_no_inode:
+	return NULL;
+}
+
+/*
  * A big issue when freeing the inode cluster is that we _cannot_ skip any
  * inodes that are in memory - they all must be marked stale and attached to
  * the cluster buffer.
@@ -2533,6 +2607,7 @@ xfs_ifree_cluster(
 	struct xfs_perag	*pag;
 	struct xfs_ino_geometry	*igeo = M_IGEO(mp);
 	xfs_ino_t		inum;
+	int			error;
 
 	inum = xic->first_ino;
 	pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, inum));
@@ -2561,12 +2636,11 @@ xfs_ifree_cluster(
 		 * complete before we get a lock on it, and hence we may fail
 		 * to mark all the active inodes on the buffer stale.
 		 */
-		bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, blkno,
-					mp->m_bsize * igeo->blocks_per_cluster,
-					XBF_UNMAPPED);
-
-		if (!bp)
-			return -ENOMEM;
+		error = xfs_trans_get_buf(tp, mp->m_ddev_targp, blkno,
+				mp->m_bsize * igeo->blocks_per_cluster,
+				XBF_UNMAPPED, &bp);
+		if (error)
+			return error;
 
 		/*
 		 * This buffer may not have been correctly initialised as we
@@ -2610,77 +2684,11 @@ xfs_ifree_cluster(
 		 * even trying to lock them.
 		 */
 		for (i = 0; i < igeo->inodes_per_cluster; i++) {
-retry:
-			rcu_read_lock();
-			ip = radix_tree_lookup(&pag->pag_ici_root,
-					XFS_INO_TO_AGINO(mp, (inum + i)));
-
-			/* Inode not in memory, nothing to do */
-			if (!ip) {
-				rcu_read_unlock();
-				continue;
-			}
-
-			/*
-			 * because this is an RCU protected lookup, we could
-			 * find a recently freed or even reallocated inode
-			 * during the lookup. We need to check under the
-			 * i_flags_lock for a valid inode here. Skip it if it
-			 * is not valid, the wrong inode or stale.
-			 */
-			spin_lock(&ip->i_flags_lock);
-			if (ip->i_ino != inum + i ||
-			    __xfs_iflags_test(ip, XFS_ISTALE)) {
-				spin_unlock(&ip->i_flags_lock);
-				rcu_read_unlock();
+			ip = xfs_ifree_get_one_inode(pag, free_ip, inum + i);
+			if (!ip)
 				continue;
-			}
-			spin_unlock(&ip->i_flags_lock);
-
-			/*
-			 * Don't try to lock/unlock the current inode, but we
-			 * _cannot_ skip the other inodes that we did not find
-			 * in the list attached to the buffer and are not
-			 * already marked stale. If we can't lock it, back off
-			 * and retry.
-			 */
-			if (ip != free_ip) {
-				if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) {
-					rcu_read_unlock();
-					delay(1);
-					goto retry;
-				}
 
-				/*
-				 * Check the inode number again in case we're
-				 * racing with freeing in xfs_reclaim_inode().
-				 * See the comments in that function for more
-				 * information as to why the initial check is
-				 * not sufficient.
-				 */
-				if (ip->i_ino != inum + i) {
-					xfs_iunlock(ip, XFS_ILOCK_EXCL);
-					rcu_read_unlock();
-					continue;
-				}
-			}
-			rcu_read_unlock();
-
-			xfs_iflock(ip);
-			xfs_iflags_set(ip, XFS_ISTALE);
-
-			/*
-			 * we don't need to attach clean inodes or those only
-			 * with unlogged changes (which we throw away, anyway).
-			 */
 			iip = ip->i_itemp;
-			if (!iip || xfs_inode_clean(ip)) {
-				ASSERT(ip != free_ip);
-				xfs_ifunlock(ip);
-				xfs_iunlock(ip, XFS_ILOCK_EXCL);
-				continue;
-			}
-
 			iip->ili_last_fields = iip->ili_fields;
 			iip->ili_fields = 0;
 			iip->ili_fsync_fields = 0;
@@ -3196,6 +3204,7 @@ xfs_rename(
 	struct xfs_trans	*tp;
 	struct xfs_inode	*wip = NULL;		/* whiteout inode */
 	struct xfs_inode	*inodes[__XFS_SORT_INODES];
+	struct xfs_buf		*agibp;
 	int			num_inodes = __XFS_SORT_INODES;
 	bool			new_parent = (src_dp != target_dp);
 	bool			src_is_directory = S_ISDIR(VFS_I(src_ip)->i_mode);
@@ -3270,7 +3279,7 @@ xfs_rename(
 	 * tree quota mechanism would be circumvented.
 	 */
 	if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
-		     (xfs_get_projid(target_dp) != xfs_get_projid(src_ip)))) {
+		     target_dp->i_d.di_projid != src_ip->i_d.di_projid)) {
 		error = -EXDEV;
 		goto out_trans_cancel;
 	}
@@ -3327,7 +3336,6 @@ xfs_rename(
 			goto out_trans_cancel;
 
 		xfs_bumplink(tp, wip);
-		xfs_trans_log_inode(tp, wip, XFS_ILOG_CORE);
 		VFS_I(wip)->i_state &= ~I_LINKABLE;
 	}
 
@@ -3361,6 +3369,22 @@ xfs_rename(
 		 * In case there is already an entry with the same
 		 * name at the destination directory, remove it first.
 		 */
+
+		/*
+		 * Check whether the replace operation will need to allocate
+		 * blocks.  This happens when the shortform directory lacks
+		 * space and we have to convert it to a block format directory.
+		 * When more blocks are necessary, we must lock the AGI first
+		 * to preserve locking order (AGI -> AGF).
+		 */
+		if (xfs_dir2_sf_replace_needblock(target_dp, src_ip->i_ino)) {
+			error = xfs_read_agi(mp, tp,
+					XFS_INO_TO_AGNO(mp, target_ip->i_ino),
+					&agibp);
+			if (error)
+				goto out_trans_cancel;
+		}
+
 		error = xfs_dir_replace(tp, target_dp, target_name,
 					src_ip->i_ino, spaceres);
 		if (error)
@@ -3778,7 +3802,6 @@ xfs_iflush_int(
 	ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
 	       ip->i_d.di_nextents > XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK));
 	ASSERT(iip != NULL && iip->ili_fields != 0);
-	ASSERT(ip->i_d.di_version > 1);
 
 	/* set *dip = inode's place in the buffer */
 	dip = xfs_buf_offset(bp, ip->i_imap.im_boffset);
@@ -3839,7 +3862,7 @@ xfs_iflush_int(
 	 * backwards compatibility with old kernels that predate logging all
 	 * inode changes.
 	 */
-	if (ip->i_d.di_version < 3)
+	if (!xfs_sb_version_has_v3inode(&mp->m_sb))
 		ip->i_d.di_flushiter++;
 
 	/* Check the inline fork data before we write out. */
@@ -3922,3 +3945,22 @@ xfs_irele(
 	trace_xfs_irele(ip, _RET_IP_);
 	iput(VFS_I(ip));
 }
+
+/*
+ * Ensure all commited transactions touching the inode are written to the log.
+ */
+int
+xfs_log_force_inode(
+	struct xfs_inode	*ip)
+{
+	xfs_lsn_t		lsn = 0;
+
+	xfs_ilock(ip, XFS_ILOCK_SHARED);
+	if (xfs_ipincount(ip))
+		lsn = ip->i_itemp->ili_last_lsn;
+	xfs_iunlock(ip, XFS_ILOCK_SHARED);
+
+	if (!lsn)
+		return 0;
+	return xfs_log_force_lsn(ip->i_mount, lsn, XFS_LOG_SYNC, NULL);
+}
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 558173f95a03..c6a63f6764a6 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -37,9 +37,6 @@ typedef struct xfs_inode {
 	struct xfs_ifork	*i_cowfp;	/* copy on write extents */
 	struct xfs_ifork	i_df;		/* data fork */
 
-	/* operations vectors */
-	const struct xfs_dir_ops *d_ops;		/* directory ops vector */
-
 	/* Transaction and locking information. */
 	struct xfs_inode_log_item *i_itemp;	/* logging information */
 	mrlock_t		i_lock;		/* inode lock */
@@ -177,30 +174,11 @@ xfs_iflags_test_and_set(xfs_inode_t *ip, unsigned short flags)
 	return ret;
 }
 
-/*
- * Project quota id helpers (previously projid was 16bit only
- * and using two 16bit values to hold new 32bit projid was chosen
- * to retain compatibility with "old" filesystems).
- */
-static inline prid_t
-xfs_get_projid(struct xfs_inode *ip)
-{
-	return (prid_t)ip->i_d.di_projid_hi << 16 | ip->i_d.di_projid_lo;
-}
-
-static inline void
-xfs_set_projid(struct xfs_inode *ip,
-		prid_t projid)
-{
-	ip->i_d.di_projid_hi = (uint16_t) (projid >> 16);
-	ip->i_d.di_projid_lo = (uint16_t) (projid & 0xffff);
-}
-
 static inline prid_t
 xfs_get_initial_prid(struct xfs_inode *dp)
 {
 	if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
-		return xfs_get_projid(dp);
+		return dp->i_d.di_projid;
 
 	return XFS_PROJID_DEFAULT;
 }
@@ -220,6 +198,13 @@ static inline bool xfs_inode_has_cow_data(struct xfs_inode *ip)
 }
 
 /*
+ * Return the buftarg used for data allocations on a given inode.
+ */
+#define xfs_inode_buftarg(ip) \
+	(XFS_IS_REALTIME_INODE(ip) ? \
+		(ip)->i_mount->m_rtdev_targp : (ip)->i_mount->m_ddev_targp)
+
+/*
  * In-core inode flags.
  */
 #define XFS_IRECLAIM		(1 << 0) /* started reclaiming this inode */
@@ -441,6 +426,7 @@ int		xfs_itruncate_extents_flags(struct xfs_trans **,
 				struct xfs_inode *, int, xfs_fsize_t, int);
 void		xfs_iext_realloc(xfs_inode_t *, int, int);
 
+int		xfs_log_force_inode(struct xfs_inode *ip);
 void		xfs_iunpin_wait(xfs_inode_t *);
 #define xfs_ipincount(ip)	((unsigned int) atomic_read(&ip->i_pincount))
 
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index bb8f076805b9..f779cca2346f 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -17,6 +17,7 @@
 #include "xfs_trans_priv.h"
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
+#include "xfs_error.h"
 
 #include <linux/iversion.h>
 
@@ -124,7 +125,7 @@ xfs_inode_item_size(
 
 	*nvecs += 2;
 	*nbytes += sizeof(struct xfs_inode_log_format) +
-		   xfs_log_dinode_size(ip->i_d.di_version);
+		   xfs_log_dinode_size(ip->i_mount);
 
 	xfs_inode_item_data_fork_size(iip, nvecs, nbytes);
 	if (XFS_IFORK_Q(ip))
@@ -304,13 +305,11 @@ xfs_inode_to_log_dinode(
 	struct inode		*inode = VFS_I(ip);
 
 	to->di_magic = XFS_DINODE_MAGIC;
-
-	to->di_version = from->di_version;
 	to->di_format = from->di_format;
-	to->di_uid = from->di_uid;
-	to->di_gid = from->di_gid;
-	to->di_projid_lo = from->di_projid_lo;
-	to->di_projid_hi = from->di_projid_hi;
+	to->di_uid = i_uid_read(inode);
+	to->di_gid = i_gid_read(inode);
+	to->di_projid_lo = from->di_projid & 0xffff;
+	to->di_projid_hi = from->di_projid >> 16;
 
 	memset(to->di_pad, 0, sizeof(to->di_pad));
 	memset(to->di_pad3, 0, sizeof(to->di_pad3));
@@ -338,10 +337,11 @@ xfs_inode_to_log_dinode(
 	/* log a dummy value to ensure log structure is fully initialised */
 	to->di_next_unlinked = NULLAGINO;
 
-	if (from->di_version == 3) {
+	if (xfs_sb_version_has_v3inode(&ip->i_mount->m_sb)) {
+		to->di_version = 3;
 		to->di_changecount = inode_peek_iversion(inode);
-		to->di_crtime.t_sec = from->di_crtime.t_sec;
-		to->di_crtime.t_nsec = from->di_crtime.t_nsec;
+		to->di_crtime.t_sec = from->di_crtime.tv_sec;
+		to->di_crtime.t_nsec = from->di_crtime.tv_nsec;
 		to->di_flags2 = from->di_flags2;
 		to->di_cowextsize = from->di_cowextsize;
 		to->di_ino = ip->i_ino;
@@ -350,6 +350,7 @@ xfs_inode_to_log_dinode(
 		uuid_copy(&to->di_uuid, &ip->i_mount->m_sb.sb_meta_uuid);
 		to->di_flushiter = 0;
 	} else {
+		to->di_version = 2;
 		to->di_flushiter = from->di_flushiter;
 	}
 }
@@ -369,7 +370,7 @@ xfs_inode_item_format_core(
 
 	dic = xlog_prepare_iovec(lv, vecp, XLOG_REG_TYPE_ICORE);
 	xfs_inode_to_log_dinode(ip, dic, ip->i_itemp->ili_item.li_lsn);
-	xlog_finish_iovec(lv, *vecp, xfs_log_dinode_size(ip->i_d.di_version));
+	xlog_finish_iovec(lv, *vecp, xfs_log_dinode_size(ip->i_mount));
 }
 
 /*
@@ -394,8 +395,6 @@ xfs_inode_item_format(
 	struct xfs_log_iovec	*vecp = NULL;
 	struct xfs_inode_log_format *ilf;
 
-	ASSERT(ip->i_d.di_version > 1);
-
 	ilf = xlog_prepare_iovec(lv, &vecp, XLOG_REG_TYPE_IFORMAT);
 	ilf->ilf_type = XFS_LI_INODE;
 	ilf->ilf_ino = ip->i_ino;
@@ -553,7 +552,8 @@ xfs_inode_item_push(
 		if (!xfs_buf_delwri_queue(bp, buffer_list))
 			rval = XFS_ITEM_FLUSHING;
 		xfs_buf_relse(bp);
-	}
+	} else if (error == -EAGAIN)
+		rval = XFS_ITEM_LOCKED;
 
 	spin_lock(&lip->li_ailp->ail_lock);
 out_unlock:
@@ -666,7 +666,7 @@ xfs_inode_item_destroy(
 	xfs_inode_t	*ip)
 {
 	kmem_free(ip->i_itemp->ili_item.li_lv_shadow);
-	kmem_zone_free(xfs_ili_zone, ip->i_itemp);
+	kmem_cache_free(xfs_ili_zone, ip->i_itemp);
 }
 
 
@@ -731,29 +731,27 @@ xfs_iflush_done(
 	 * holding the lock before removing the inode from the AIL.
 	 */
 	if (need_ail) {
-		bool			mlip_changed = false;
+		xfs_lsn_t	tail_lsn = 0;
 
 		/* this is an opencoded batch version of xfs_trans_ail_delete */
 		spin_lock(&ailp->ail_lock);
 		list_for_each_entry(blip, &tmp, li_bio_list) {
 			if (INODE_ITEM(blip)->ili_logged &&
-			    blip->li_lsn == INODE_ITEM(blip)->ili_flush_lsn)
-				mlip_changed |= xfs_ail_delete_one(ailp, blip);
-			else {
+			    blip->li_lsn == INODE_ITEM(blip)->ili_flush_lsn) {
+				/*
+				 * xfs_ail_update_finish() only cares about the
+				 * lsn of the first tail item removed, any
+				 * others will be at the same or higher lsn so
+				 * we just ignore them.
+				 */
+				xfs_lsn_t lsn = xfs_ail_delete_one(ailp, blip);
+				if (!tail_lsn && lsn)
+					tail_lsn = lsn;
+			} else {
 				xfs_clear_li_failed(blip);
 			}
 		}
-
-		if (mlip_changed) {
-			if (!XFS_FORCED_SHUTDOWN(ailp->ail_mount))
-				xlog_assign_tail_lsn_locked(ailp->ail_mount);
-			if (list_empty(&ailp->ail_head))
-				wake_up_all(&ailp->ail_empty);
-		}
-		spin_unlock(&ailp->ail_lock);
-
-		if (mlip_changed)
-			xfs_log_space_wake(ailp->ail_mount);
+		xfs_ail_update_finish(ailp, tail_lsn);
 	}
 
 	/*
@@ -828,8 +826,10 @@ xfs_inode_item_format_convert(
 {
 	struct xfs_inode_log_format_32	*in_f32 = buf->i_addr;
 
-	if (buf->i_len != sizeof(*in_f32))
+	if (buf->i_len != sizeof(*in_f32)) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 		return -EFSCORRUPTED;
+	}
 
 	in_f->ilf_type = in_f32->ilf_type;
 	in_f->ilf_size = in_f32->ilf_size;
diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index d58f0d6a699e..309958186d33 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -33,6 +33,10 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_health.h"
+#include "xfs_reflink.h"
+#include "xfs_ioctl.h"
+#include "xfs_da_format.h"
+#include "xfs_da_btree.h"
 
 #include <linux/mount.h>
 #include <linux/namei.h>
@@ -290,138 +294,173 @@ xfs_readlink_by_handle(
 	return error;
 }
 
-int
-xfs_set_dmattrs(
-	xfs_inode_t     *ip,
-	uint		evmask,
-	uint16_t	state)
+/*
+ * Format an attribute and copy it out to the user's buffer.
+ * Take care to check values and protect against them changing later,
+ * we may be reading them directly out of a user buffer.
+ */
+static void
+xfs_ioc_attr_put_listent(
+	struct xfs_attr_list_context *context,
+	int			flags,
+	unsigned char		*name,
+	int			namelen,
+	int			valuelen)
 {
-	xfs_mount_t	*mp = ip->i_mount;
-	xfs_trans_t	*tp;
-	int		error;
+	struct xfs_attrlist	*alist = context->buffer;
+	struct xfs_attrlist_ent	*aep;
+	int			arraytop;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return -EPERM;
+	ASSERT(!context->seen_enough);
+	ASSERT(context->count >= 0);
+	ASSERT(context->count < (ATTR_MAX_VALUELEN/8));
+	ASSERT(context->firstu >= sizeof(*alist));
+	ASSERT(context->firstu <= context->bufsize);
 
-	if (XFS_FORCED_SHUTDOWN(mp))
-		return -EIO;
-
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ichange, 0, 0, 0, &tp);
-	if (error)
-		return error;
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-
-	ip->i_d.di_dmevmask = evmask;
-	ip->i_d.di_dmstate  = state;
-
-	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-	error = xfs_trans_commit(tp);
+	/*
+	 * Only list entries in the right namespace.
+	 */
+	if (context->attr_filter != (flags & XFS_ATTR_NSP_ONDISK_MASK))
+		return;
+
+	arraytop = sizeof(*alist) +
+			context->count * sizeof(alist->al_offset[0]);
+
+	/* decrement by the actual bytes used by the attr */
+	context->firstu -= round_up(offsetof(struct xfs_attrlist_ent, a_name) +
+			namelen + 1, sizeof(uint32_t));
+	if (context->firstu < arraytop) {
+		trace_xfs_attr_list_full(context);
+		alist->al_more = 1;
+		context->seen_enough = 1;
+		return;
+	}
+
+	aep = context->buffer + context->firstu;
+	aep->a_valuelen = valuelen;
+	memcpy(aep->a_name, name, namelen);
+	aep->a_name[namelen] = 0;
+	alist->al_offset[context->count++] = context->firstu;
+	alist->al_count = context->count;
+	trace_xfs_attr_list_add(context);
+}
 
-	return error;
+static unsigned int
+xfs_attr_filter(
+	u32			ioc_flags)
+{
+	if (ioc_flags & XFS_IOC_ATTR_ROOT)
+		return XFS_ATTR_ROOT;
+	if (ioc_flags & XFS_IOC_ATTR_SECURE)
+		return XFS_ATTR_SECURE;
+	return 0;
 }
 
-STATIC int
-xfs_fssetdm_by_handle(
-	struct file		*parfilp,
-	void			__user *arg)
+static unsigned int
+xfs_attr_flags(
+	u32			ioc_flags)
 {
-	int			error;
-	struct fsdmidata	fsd;
-	xfs_fsop_setdm_handlereq_t dmhreq;
-	struct dentry		*dentry;
+	if (ioc_flags & XFS_IOC_ATTR_CREATE)
+		return XATTR_CREATE;
+	if (ioc_flags & XFS_IOC_ATTR_REPLACE)
+		return XATTR_REPLACE;
+	return 0;
+}
 
-	if (!capable(CAP_MKNOD))
-		return -EPERM;
-	if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t)))
-		return -EFAULT;
+int
+xfs_ioc_attr_list(
+	struct xfs_inode		*dp,
+	void __user			*ubuf,
+	int				bufsize,
+	int				flags,
+	struct xfs_attrlist_cursor __user *ucursor)
+{
+	struct xfs_attr_list_context	context = { };
+	struct xfs_attrlist		*alist;
+	void				*buffer;
+	int				error;
 
-	error = mnt_want_write_file(parfilp);
-	if (error)
-		return error;
+	if (bufsize < sizeof(struct xfs_attrlist) ||
+	    bufsize > XFS_XATTR_LIST_MAX)
+		return -EINVAL;
 
-	dentry = xfs_handlereq_to_dentry(parfilp, &dmhreq.hreq);
-	if (IS_ERR(dentry)) {
-		mnt_drop_write_file(parfilp);
-		return PTR_ERR(dentry);
-	}
+	/*
+	 * Reject flags, only allow namespaces.
+	 */
+	if (flags & ~(XFS_IOC_ATTR_ROOT | XFS_IOC_ATTR_SECURE))
+		return -EINVAL;
+	if (flags == (XFS_IOC_ATTR_ROOT | XFS_IOC_ATTR_SECURE))
+		return -EINVAL;
 
-	if (IS_IMMUTABLE(d_inode(dentry)) || IS_APPEND(d_inode(dentry))) {
-		error = -EPERM;
-		goto out;
-	}
+	/*
+	 * Validate the cursor.
+	 */
+	if (copy_from_user(&context.cursor, ucursor, sizeof(context.cursor)))
+		return -EFAULT;
+	if (context.cursor.pad1 || context.cursor.pad2)
+		return -EINVAL;
+	if (!context.cursor.initted &&
+	    (context.cursor.hashval || context.cursor.blkno ||
+	     context.cursor.offset))
+		return -EINVAL;
 
-	if (copy_from_user(&fsd, dmhreq.data, sizeof(fsd))) {
-		error = -EFAULT;
-		goto out;
-	}
+	buffer = kmem_zalloc_large(bufsize, 0);
+	if (!buffer)
+		return -ENOMEM;
 
-	error = xfs_set_dmattrs(XFS_I(d_inode(dentry)), fsd.fsd_dmevmask,
-				 fsd.fsd_dmstate);
+	/*
+	 * Initialize the output buffer.
+	 */
+	context.dp = dp;
+	context.resynch = 1;
+	context.attr_filter = xfs_attr_filter(flags);
+	context.buffer = buffer;
+	context.bufsize = round_down(bufsize, sizeof(uint32_t));
+	context.firstu = context.bufsize;
+	context.put_listent = xfs_ioc_attr_put_listent;
+
+	alist = context.buffer;
+	alist->al_count = 0;
+	alist->al_more = 0;
+	alist->al_offset[0] = context.bufsize;
+
+	error = xfs_attr_list(&context);
+	if (error)
+		goto out_free;
 
- out:
-	mnt_drop_write_file(parfilp);
-	dput(dentry);
+	if (copy_to_user(ubuf, buffer, bufsize) ||
+	    copy_to_user(ucursor, &context.cursor, sizeof(context.cursor)))
+		error = -EFAULT;
+out_free:
+	kmem_free(buffer);
 	return error;
 }
 
 STATIC int
 xfs_attrlist_by_handle(
 	struct file		*parfilp,
-	void			__user *arg)
+	struct xfs_fsop_attrlist_handlereq __user *p)
 {
-	int			error = -ENOMEM;
-	attrlist_cursor_kern_t	*cursor;
-	struct xfs_fsop_attrlist_handlereq __user	*p = arg;
-	xfs_fsop_attrlist_handlereq_t al_hreq;
+	struct xfs_fsop_attrlist_handlereq al_hreq;
 	struct dentry		*dentry;
-	char			*kbuf;
+	int			error = -ENOMEM;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
-	if (copy_from_user(&al_hreq, arg, sizeof(xfs_fsop_attrlist_handlereq_t)))
+	if (copy_from_user(&al_hreq, p, sizeof(al_hreq)))
 		return -EFAULT;
-	if (al_hreq.buflen < sizeof(struct attrlist) ||
-	    al_hreq.buflen > XFS_XATTR_LIST_MAX)
-		return -EINVAL;
-
-	/*
-	 * Reject flags, only allow namespaces.
-	 */
-	if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE))
-		return -EINVAL;
 
 	dentry = xfs_handlereq_to_dentry(parfilp, &al_hreq.hreq);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
-	kbuf = kmem_zalloc_large(al_hreq.buflen, 0);
-	if (!kbuf)
-		goto out_dput;
-
-	cursor = (attrlist_cursor_kern_t *)&al_hreq.pos;
-	error = xfs_attr_list(XFS_I(d_inode(dentry)), kbuf, al_hreq.buflen,
-					al_hreq.flags, cursor);
-	if (error)
-		goto out_kfree;
-
-	if (copy_to_user(&p->pos, cursor, sizeof(attrlist_cursor_kern_t))) {
-		error = -EFAULT;
-		goto out_kfree;
-	}
-
-	if (copy_to_user(al_hreq.buffer, kbuf, al_hreq.buflen))
-		error = -EFAULT;
-
-out_kfree:
-	kmem_free(kbuf);
-out_dput:
+	error = xfs_ioc_attr_list(XFS_I(d_inode(dentry)), al_hreq.buffer,
+				  al_hreq.buflen, al_hreq.flags, &p->pos);
 	dput(dentry);
 	return error;
 }
 
-int
+static int
 xfs_attrmulti_attr_get(
 	struct inode		*inode,
 	unsigned char		*name,
@@ -429,28 +468,33 @@ xfs_attrmulti_attr_get(
 	uint32_t		*len,
 	uint32_t		flags)
 {
-	unsigned char		*kbuf;
-	int			error = -EFAULT;
+	struct xfs_da_args	args = {
+		.dp		= XFS_I(inode),
+		.attr_filter	= xfs_attr_filter(flags),
+		.attr_flags	= xfs_attr_flags(flags),
+		.name		= name,
+		.namelen	= strlen(name),
+		.valuelen	= *len,
+	};
+	int			error;
 
 	if (*len > XFS_XATTR_SIZE_MAX)
 		return -EINVAL;
-	kbuf = kmem_zalloc_large(*len, 0);
-	if (!kbuf)
-		return -ENOMEM;
 
-	error = xfs_attr_get(XFS_I(inode), name, &kbuf, (int *)len, flags);
+	error = xfs_attr_get(&args);
 	if (error)
 		goto out_kfree;
 
-	if (copy_to_user(ubuf, kbuf, *len))
+	*len = args.valuelen;
+	if (copy_to_user(ubuf, args.value, args.valuelen))
 		error = -EFAULT;
 
 out_kfree:
-	kmem_free(kbuf);
+	kmem_free(args.value);
 	return error;
 }
 
-int
+static int
 xfs_attrmulti_attr_set(
 	struct inode		*inode,
 	unsigned char		*name,
@@ -458,38 +502,75 @@ xfs_attrmulti_attr_set(
 	uint32_t		len,
 	uint32_t		flags)
 {
-	unsigned char		*kbuf;
+	struct xfs_da_args	args = {
+		.dp		= XFS_I(inode),
+		.attr_filter	= xfs_attr_filter(flags),
+		.attr_flags	= xfs_attr_flags(flags),
+		.name		= name,
+		.namelen	= strlen(name),
+	};
 	int			error;
 
 	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
 		return -EPERM;
-	if (len > XFS_XATTR_SIZE_MAX)
-		return -EINVAL;
 
-	kbuf = memdup_user(ubuf, len);
-	if (IS_ERR(kbuf))
-		return PTR_ERR(kbuf);
+	if (ubuf) {
+		if (len > XFS_XATTR_SIZE_MAX)
+			return -EINVAL;
+		args.value = memdup_user(ubuf, len);
+		if (IS_ERR(args.value))
+			return PTR_ERR(args.value);
+		args.valuelen = len;
+	}
 
-	error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags);
-	if (!error)
-		xfs_forget_acl(inode, name, flags);
-	kfree(kbuf);
+	error = xfs_attr_set(&args);
+	if (!error && (flags & XFS_IOC_ATTR_ROOT))
+		xfs_forget_acl(inode, name);
+	kfree(args.value);
 	return error;
 }
 
 int
-xfs_attrmulti_attr_remove(
+xfs_ioc_attrmulti_one(
+	struct file		*parfilp,
 	struct inode		*inode,
-	unsigned char		*name,
+	uint32_t		opcode,
+	void __user		*uname,
+	void __user		*value,
+	uint32_t		*len,
 	uint32_t		flags)
 {
+	unsigned char		*name;
 	int			error;
 
-	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		return -EPERM;
-	error = xfs_attr_remove(XFS_I(inode), name, flags);
-	if (!error)
-		xfs_forget_acl(inode, name, flags);
+	if ((flags & XFS_IOC_ATTR_ROOT) && (flags & XFS_IOC_ATTR_SECURE))
+		return -EINVAL;
+
+	name = strndup_user(uname, MAXNAMELEN);
+	if (IS_ERR(name))
+		return PTR_ERR(name);
+
+	switch (opcode) {
+	case ATTR_OP_GET:
+		error = xfs_attrmulti_attr_get(inode, name, value, len, flags);
+		break;
+	case ATTR_OP_REMOVE:
+		value = NULL;
+		*len = 0;
+		/* fall through */
+	case ATTR_OP_SET:
+		error = mnt_want_write_file(parfilp);
+		if (error)
+			break;
+		error = xfs_attrmulti_attr_set(inode, name, value, *len, flags);
+		mnt_drop_write_file(parfilp);
+		break;
+	default:
+		error = -EINVAL;
+		break;
+	}
+
+	kfree(name);
 	return error;
 }
 
@@ -503,7 +584,6 @@ xfs_attrmulti_by_handle(
 	xfs_fsop_attrmulti_handlereq_t am_hreq;
 	struct dentry		*dentry;
 	unsigned int		i, size;
-	unsigned char		*attr_name;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -529,56 +609,17 @@ xfs_attrmulti_by_handle(
 		goto out_dput;
 	}
 
-	error = -ENOMEM;
-	attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
-	if (!attr_name)
-		goto out_kfree_ops;
-
 	error = 0;
 	for (i = 0; i < am_hreq.opcount; i++) {
-		ops[i].am_error = strncpy_from_user((char *)attr_name,
-				ops[i].am_attrname, MAXNAMELEN);
-		if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
-			error = -ERANGE;
-		if (ops[i].am_error < 0)
-			break;
-
-		switch (ops[i].am_opcode) {
-		case ATTR_OP_GET:
-			ops[i].am_error = xfs_attrmulti_attr_get(
-					d_inode(dentry), attr_name,
-					ops[i].am_attrvalue, &ops[i].am_length,
-					ops[i].am_flags);
-			break;
-		case ATTR_OP_SET:
-			ops[i].am_error = mnt_want_write_file(parfilp);
-			if (ops[i].am_error)
-				break;
-			ops[i].am_error = xfs_attrmulti_attr_set(
-					d_inode(dentry), attr_name,
-					ops[i].am_attrvalue, ops[i].am_length,
-					ops[i].am_flags);
-			mnt_drop_write_file(parfilp);
-			break;
-		case ATTR_OP_REMOVE:
-			ops[i].am_error = mnt_want_write_file(parfilp);
-			if (ops[i].am_error)
-				break;
-			ops[i].am_error = xfs_attrmulti_attr_remove(
-					d_inode(dentry), attr_name,
-					ops[i].am_flags);
-			mnt_drop_write_file(parfilp);
-			break;
-		default:
-			ops[i].am_error = -EINVAL;
-		}
+		ops[i].am_error = xfs_ioc_attrmulti_one(parfilp,
+				d_inode(dentry), ops[i].am_opcode,
+				ops[i].am_attrname, ops[i].am_attrvalue,
+				&ops[i].am_length, ops[i].am_flags);
 	}
 
 	if (copy_to_user(am_hreq.ops, ops, size))
 		error = -EFAULT;
 
-	kfree(attr_name);
- out_kfree_ops:
 	kfree(ops);
  out_dput:
 	dput(dentry);
@@ -588,13 +629,12 @@ xfs_attrmulti_by_handle(
 int
 xfs_ioc_space(
 	struct file		*filp,
-	unsigned int		cmd,
 	xfs_flock64_t		*bf)
 {
 	struct inode		*inode = file_inode(filp);
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct iattr		iattr;
-	enum xfs_prealloc_flags	flags = 0;
+	enum xfs_prealloc_flags	flags = XFS_PREALLOC_CLEAR;
 	uint			iolock = XFS_IOLOCK_EXCL | XFS_MMAPLOCK_EXCL;
 	int			error;
 
@@ -607,6 +647,9 @@ xfs_ioc_space(
 	if (!S_ISREG(inode->i_mode))
 		return -EINVAL;
 
+	if (xfs_is_always_cow_inode(ip))
+		return -EOPNOTSUPP;
+
 	if (filp->f_flags & O_DSYNC)
 		flags |= XFS_PREALLOC_SYNC;
 	if (filp->f_mode & FMODE_NOCMTIME)
@@ -620,6 +663,7 @@ xfs_ioc_space(
 	error = xfs_break_layouts(inode, &iolock, BREAK_UNMAP);
 	if (error)
 		goto out_unlock;
+	inode_dio_wait(inode);
 
 	switch (bf->l_whence) {
 	case 0: /*SEEK_SET*/
@@ -635,73 +679,21 @@ xfs_ioc_space(
 		goto out_unlock;
 	}
 
-	/*
-	 * length of <= 0 for resv/unresv/zero is invalid.  length for
-	 * alloc/free is ignored completely and we have no idea what userspace
-	 * might have set it to, so set it to zero to allow range
-	 * checks to pass.
-	 */
-	switch (cmd) {
-	case XFS_IOC_ZERO_RANGE:
-	case XFS_IOC_RESVSP:
-	case XFS_IOC_RESVSP64:
-	case XFS_IOC_UNRESVSP:
-	case XFS_IOC_UNRESVSP64:
-		if (bf->l_len <= 0) {
-			error = -EINVAL;
-			goto out_unlock;
-		}
-		break;
-	default:
-		bf->l_len = 0;
-		break;
-	}
-
-	if (bf->l_start < 0 ||
-	    bf->l_start > inode->i_sb->s_maxbytes ||
-	    bf->l_start + bf->l_len < 0 ||
-	    bf->l_start + bf->l_len >= inode->i_sb->s_maxbytes) {
+	if (bf->l_start < 0 || bf->l_start > inode->i_sb->s_maxbytes) {
 		error = -EINVAL;
 		goto out_unlock;
 	}
 
-	switch (cmd) {
-	case XFS_IOC_ZERO_RANGE:
-		flags |= XFS_PREALLOC_SET;
-		error = xfs_zero_file_space(ip, bf->l_start, bf->l_len);
-		break;
-	case XFS_IOC_RESVSP:
-	case XFS_IOC_RESVSP64:
-		flags |= XFS_PREALLOC_SET;
-		error = xfs_alloc_file_space(ip, bf->l_start, bf->l_len,
-						XFS_BMAPI_PREALLOC);
-		break;
-	case XFS_IOC_UNRESVSP:
-	case XFS_IOC_UNRESVSP64:
-		error = xfs_free_file_space(ip, bf->l_start, bf->l_len);
-		break;
-	case XFS_IOC_ALLOCSP:
-	case XFS_IOC_ALLOCSP64:
-	case XFS_IOC_FREESP:
-	case XFS_IOC_FREESP64:
-		flags |= XFS_PREALLOC_CLEAR;
-		if (bf->l_start > XFS_ISIZE(ip)) {
-			error = xfs_alloc_file_space(ip, XFS_ISIZE(ip),
-					bf->l_start - XFS_ISIZE(ip), 0);
-			if (error)
-				goto out_unlock;
-		}
-
-		iattr.ia_valid = ATTR_SIZE;
-		iattr.ia_size = bf->l_start;
-
-		error = xfs_vn_setattr_size(file_dentry(filp), &iattr);
-		break;
-	default:
-		ASSERT(0);
-		error = -EINVAL;
+	if (bf->l_start > XFS_ISIZE(ip)) {
+		error = xfs_alloc_file_space(ip, XFS_ISIZE(ip),
+				bf->l_start - XFS_ISIZE(ip), 0);
+		if (error)
+			goto out_unlock;
 	}
 
+	iattr.ia_valid = ATTR_SIZE;
+	iattr.ia_size = bf->l_start;
+	error = xfs_vn_setattr_size(file_dentry(filp), &iattr);
 	if (error)
 		goto out_unlock;
 
@@ -1116,7 +1108,7 @@ xfs_fill_fsxattr(
 	fa->fsx_extsize = ip->i_d.di_extsize << ip->i_mount->m_sb.sb_blocklog;
 	fa->fsx_cowextsize = ip->i_d.di_cowextsize <<
 			ip->i_mount->m_sb.sb_blocklog;
-	fa->fsx_projid = xfs_get_projid(ip);
+	fa->fsx_projid = ip->i_d.di_projid;
 
 	if (attr) {
 		if (ip->i_afp) {
@@ -1271,7 +1263,7 @@ xfs_ioctl_setattr_xflags(
 
 	/* diflags2 only valid for v3 inodes. */
 	di_flags2 = xfs_flags2diflags2(ip, fa->fsx_xflags);
-	if (di_flags2 && ip->i_d.di_version < 3)
+	if (di_flags2 && !xfs_sb_version_has_v3inode(&mp->m_sb))
 		return -EINVAL;
 
 	ip->i_d.di_flags = xfs_flags2diflags(ip, fa->fsx_xflags);
@@ -1311,10 +1303,9 @@ xfs_ioctl_setattr_dax_invalidate(
 	 * have to check the device for dax support or flush pagecache.
 	 */
 	if (fa->fsx_xflags & FS_XFLAG_DAX) {
-		if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)))
-			return -EINVAL;
-		if (!bdev_dax_supported(xfs_find_bdev_for_inode(VFS_I(ip)),
-				sb->s_blocksize))
+		struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
+
+		if (!bdev_dax_supported(target->bt_bdev, sb->s_blocksize))
 			return -EINVAL;
 	}
 
@@ -1482,8 +1473,7 @@ xfs_ioctl_setattr_check_cowextsize(
 	if (!(fa->fsx_xflags & FS_XFLAG_COWEXTSIZE))
 		return 0;
 
-	if (!xfs_sb_version_hasreflink(&ip->i_mount->m_sb) ||
-	    ip->i_d.di_version != 3)
+	if (!xfs_sb_version_hasreflink(&ip->i_mount->m_sb))
 		return -EINVAL;
 
 	if (fa->fsx_cowextsize == 0)
@@ -1544,9 +1534,9 @@ xfs_ioctl_setattr(
 	 * because the i_*dquot fields will get updated anyway.
 	 */
 	if (XFS_IS_QUOTA_ON(mp)) {
-		code = xfs_qm_vop_dqalloc(ip, ip->i_d.di_uid,
-					 ip->i_d.di_gid, fa->fsx_projid,
-					 XFS_QMOPT_PQUOTA, &udqp, NULL, &pdqp);
+		code = xfs_qm_vop_dqalloc(ip, VFS_I(ip)->i_uid,
+				VFS_I(ip)->i_gid, fa->fsx_projid,
+				XFS_QMOPT_PQUOTA, &udqp, NULL, &pdqp);
 		if (code)
 			return code;
 	}
@@ -1569,7 +1559,7 @@ xfs_ioctl_setattr(
 	}
 
 	if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp) &&
-	    xfs_get_projid(ip) != fa->fsx_projid) {
+	    ip->i_d.di_projid != fa->fsx_projid) {
 		code = xfs_qm_vop_chown_reserve(tp, ip, udqp, NULL, pdqp,
 				capable(CAP_FOWNER) ?  XFS_QMOPT_FORCE_RES : 0);
 		if (code)	/* out of quota */
@@ -1606,13 +1596,12 @@ xfs_ioctl_setattr(
 		VFS_I(ip)->i_mode &= ~(S_ISUID|S_ISGID);
 
 	/* Change the ownerships and register project quota modifications */
-	if (xfs_get_projid(ip) != fa->fsx_projid) {
+	if (ip->i_d.di_projid != fa->fsx_projid) {
 		if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) {
 			olddquot = xfs_qm_vop_chown(tp, ip,
 						&ip->i_pdquot, pdqp);
 		}
-		ASSERT(ip->i_d.di_version > 1);
-		xfs_set_projid(ip, fa->fsx_projid);
+		ip->i_d.di_projid = fa->fsx_projid;
 	}
 
 	/*
@@ -1624,7 +1613,7 @@ xfs_ioctl_setattr(
 		ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog;
 	else
 		ip->i_d.di_extsize = 0;
-	if (ip->i_d.di_version == 3 &&
+	if (xfs_sb_version_has_v3inode(&mp->m_sb) &&
 	    (ip->i_d.di_flags2 & XFS_DIFLAG2_COWEXTSIZE))
 		ip->i_d.di_cowextsize = fa->fsx_cowextsize >>
 				mp->m_sb.sb_blocklog;
@@ -2122,24 +2111,17 @@ xfs_file_ioctl(
 		return xfs_ioc_setlabel(filp, mp, arg);
 	case XFS_IOC_ALLOCSP:
 	case XFS_IOC_FREESP:
-	case XFS_IOC_RESVSP:
-	case XFS_IOC_UNRESVSP:
 	case XFS_IOC_ALLOCSP64:
-	case XFS_IOC_FREESP64:
-	case XFS_IOC_RESVSP64:
-	case XFS_IOC_UNRESVSP64:
-	case XFS_IOC_ZERO_RANGE: {
+	case XFS_IOC_FREESP64: {
 		xfs_flock64_t		bf;
 
 		if (copy_from_user(&bf, arg, sizeof(bf)))
 			return -EFAULT;
-		return xfs_ioc_space(filp, cmd, &bf);
+		return xfs_ioc_space(filp, &bf);
 	}
 	case XFS_IOC_DIOINFO: {
-		struct dioattr	da;
-		xfs_buftarg_t	*target =
-			XFS_IS_REALTIME_INODE(ip) ?
-			mp->m_rtdev_targp : mp->m_ddev_targp;
+		struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
+		struct dioattr		da;
 
 		da.d_mem =  da.d_miniosz = target->bt_logical_sectorsize;
 		da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1);
@@ -2183,22 +2165,6 @@ xfs_file_ioctl(
 	case XFS_IOC_SETXFLAGS:
 		return xfs_ioc_setxflags(ip, filp, arg);
 
-	case XFS_IOC_FSSETDM: {
-		struct fsdmidata	dmi;
-
-		if (copy_from_user(&dmi, arg, sizeof(dmi)))
-			return -EFAULT;
-
-		error = mnt_want_write_file(filp);
-		if (error)
-			return error;
-
-		error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask,
-				dmi.fsd_dmstate);
-		mnt_drop_write_file(filp);
-		return error;
-	}
-
 	case XFS_IOC_GETBMAP:
 	case XFS_IOC_GETBMAPA:
 	case XFS_IOC_GETBMAPX:
@@ -2226,8 +2192,6 @@ xfs_file_ioctl(
 			return -EFAULT;
 		return xfs_open_by_handle(filp, &hreq);
 	}
-	case XFS_IOC_FSSETDM_BY_HANDLE:
-		return xfs_fssetdm_by_handle(filp, arg);
 
 	case XFS_IOC_READLINK_BY_HANDLE: {
 		xfs_fsop_handlereq_t	hreq;
@@ -2399,7 +2363,10 @@ xfs_file_ioctl(
 		if (error)
 			return error;
 
-		return xfs_icache_free_eofblocks(mp, &keofb);
+		sb_start_write(mp->m_super);
+		error = xfs_icache_free_eofblocks(mp, &keofb);
+		sb_end_write(mp->m_super);
+		return error;
 	}
 
 	default:
diff --git a/fs/xfs/xfs_ioctl.h b/fs/xfs/xfs_ioctl.h
index 654c0bb1bcf8..bab6a5a92407 100644
--- a/fs/xfs/xfs_ioctl.h
+++ b/fs/xfs/xfs_ioctl.h
@@ -6,10 +6,14 @@
 #ifndef __XFS_IOCTL_H__
 #define __XFS_IOCTL_H__
 
+struct xfs_bstat;
+struct xfs_ibulk;
+struct xfs_inogrp;
+
+
 extern int
 xfs_ioc_space(
 	struct file		*filp,
-	unsigned int		cmd,
 	xfs_flock64_t		*bf);
 
 int
@@ -31,27 +35,11 @@ xfs_readlink_by_handle(
 	struct file		*parfilp,
 	xfs_fsop_handlereq_t	*hreq);
 
-extern int
-xfs_attrmulti_attr_get(
-	struct inode		*inode,
-	unsigned char		*name,
-	unsigned char		__user *ubuf,
-	uint32_t		*len,
-	uint32_t		flags);
-
-extern int
-xfs_attrmulti_attr_set(
-	struct inode		*inode,
-	unsigned char		*name,
-	const unsigned char	__user *ubuf,
-	uint32_t		len,
-	uint32_t		flags);
-
-extern int
-xfs_attrmulti_attr_remove(
-	struct inode		*inode,
-	unsigned char		*name,
-	uint32_t		flags);
+int xfs_ioc_attrmulti_one(struct file *parfilp, struct inode *inode,
+		uint32_t opcode, void __user *uname, void __user *value,
+		uint32_t *len, uint32_t flags);
+int xfs_ioc_attr_list(struct xfs_inode *dp, void __user *ubuf, int bufsize,
+	int flags, struct xfs_attrlist_cursor __user *ucursor);
 
 extern struct dentry *
 xfs_handle_to_dentry(
@@ -71,16 +59,6 @@ xfs_file_compat_ioctl(
 	unsigned int		cmd,
 	unsigned long		arg);
 
-extern int
-xfs_set_dmattrs(
-	struct xfs_inode	*ip,
-	uint			evmask,
-	uint16_t		state);
-
-struct xfs_ibulk;
-struct xfs_bstat;
-struct xfs_inogrp;
-
 int xfs_fsbulkstat_one_fmt(struct xfs_ibulk *breq,
 			   const struct xfs_bulkstat *bstat);
 int xfs_fsinumbers_fmt(struct xfs_ibulk *breq, const struct xfs_inumbers *igrp);
diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c
index 1e08bf79b478..c1771e728117 100644
--- a/fs/xfs/xfs_ioctl32.c
+++ b/fs/xfs/xfs_ioctl32.c
@@ -107,7 +107,7 @@ xfs_ioctl32_bstime_copyin(
 	xfs_bstime_t		*bstime,
 	compat_xfs_bstime_t	__user *bstime32)
 {
-	compat_time_t		sec32;	/* tv_sec differs on 64 vs. 32 */
+	old_time32_t		sec32;	/* tv_sec differs on 64 vs. 32 */
 
 	if (get_user(sec32,		&bstime32->tv_sec)	||
 	    get_user(bstime->tv_nsec,	&bstime32->tv_nsec))
@@ -352,56 +352,24 @@ xfs_compat_handlereq_to_dentry(
 STATIC int
 xfs_compat_attrlist_by_handle(
 	struct file		*parfilp,
-	void			__user *arg)
+	compat_xfs_fsop_attrlist_handlereq_t __user *p)
 {
-	int			error;
-	attrlist_cursor_kern_t	*cursor;
-	compat_xfs_fsop_attrlist_handlereq_t __user *p = arg;
 	compat_xfs_fsop_attrlist_handlereq_t al_hreq;
 	struct dentry		*dentry;
-	char			*kbuf;
+	int			error;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
-	if (copy_from_user(&al_hreq, arg,
-			   sizeof(compat_xfs_fsop_attrlist_handlereq_t)))
+	if (copy_from_user(&al_hreq, p, sizeof(al_hreq)))
 		return -EFAULT;
-	if (al_hreq.buflen < sizeof(struct attrlist) ||
-	    al_hreq.buflen > XFS_XATTR_LIST_MAX)
-		return -EINVAL;
-
-	/*
-	 * Reject flags, only allow namespaces.
-	 */
-	if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE))
-		return -EINVAL;
 
 	dentry = xfs_compat_handlereq_to_dentry(parfilp, &al_hreq.hreq);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
-	error = -ENOMEM;
-	kbuf = kmem_zalloc_large(al_hreq.buflen, 0);
-	if (!kbuf)
-		goto out_dput;
-
-	cursor = (attrlist_cursor_kern_t *)&al_hreq.pos;
-	error = xfs_attr_list(XFS_I(d_inode(dentry)), kbuf, al_hreq.buflen,
-					al_hreq.flags, cursor);
-	if (error)
-		goto out_kfree;
-
-	if (copy_to_user(&p->pos, cursor, sizeof(attrlist_cursor_kern_t))) {
-		error = -EFAULT;
-		goto out_kfree;
-	}
-
-	if (copy_to_user(compat_ptr(al_hreq.buffer), kbuf, al_hreq.buflen))
-		error = -EFAULT;
-
-out_kfree:
-	kmem_free(kbuf);
-out_dput:
+	error = xfs_ioc_attr_list(XFS_I(d_inode(dentry)),
+			compat_ptr(al_hreq.buffer), al_hreq.buflen,
+			al_hreq.flags, &p->pos);
 	dput(dentry);
 	return error;
 }
@@ -416,7 +384,6 @@ xfs_compat_attrmulti_by_handle(
 	compat_xfs_fsop_attrmulti_handlereq_t	am_hreq;
 	struct dentry				*dentry;
 	unsigned int				i, size;
-	unsigned char				*attr_name;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -443,101 +410,24 @@ xfs_compat_attrmulti_by_handle(
 		goto out_dput;
 	}
 
-	error = -ENOMEM;
-	attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
-	if (!attr_name)
-		goto out_kfree_ops;
-
 	error = 0;
 	for (i = 0; i < am_hreq.opcount; i++) {
-		ops[i].am_error = strncpy_from_user((char *)attr_name,
+		ops[i].am_error = xfs_ioc_attrmulti_one(parfilp,
+				d_inode(dentry), ops[i].am_opcode,
 				compat_ptr(ops[i].am_attrname),
-				MAXNAMELEN);
-		if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
-			error = -ERANGE;
-		if (ops[i].am_error < 0)
-			break;
-
-		switch (ops[i].am_opcode) {
-		case ATTR_OP_GET:
-			ops[i].am_error = xfs_attrmulti_attr_get(
-					d_inode(dentry), attr_name,
-					compat_ptr(ops[i].am_attrvalue),
-					&ops[i].am_length, ops[i].am_flags);
-			break;
-		case ATTR_OP_SET:
-			ops[i].am_error = mnt_want_write_file(parfilp);
-			if (ops[i].am_error)
-				break;
-			ops[i].am_error = xfs_attrmulti_attr_set(
-					d_inode(dentry), attr_name,
-					compat_ptr(ops[i].am_attrvalue),
-					ops[i].am_length, ops[i].am_flags);
-			mnt_drop_write_file(parfilp);
-			break;
-		case ATTR_OP_REMOVE:
-			ops[i].am_error = mnt_want_write_file(parfilp);
-			if (ops[i].am_error)
-				break;
-			ops[i].am_error = xfs_attrmulti_attr_remove(
-					d_inode(dentry), attr_name,
-					ops[i].am_flags);
-			mnt_drop_write_file(parfilp);
-			break;
-		default:
-			ops[i].am_error = -EINVAL;
-		}
+				compat_ptr(ops[i].am_attrvalue),
+				&ops[i].am_length, ops[i].am_flags);
 	}
 
 	if (copy_to_user(compat_ptr(am_hreq.ops), ops, size))
 		error = -EFAULT;
 
-	kfree(attr_name);
- out_kfree_ops:
 	kfree(ops);
  out_dput:
 	dput(dentry);
 	return error;
 }
 
-STATIC int
-xfs_compat_fssetdm_by_handle(
-	struct file		*parfilp,
-	void			__user *arg)
-{
-	int			error;
-	struct fsdmidata	fsd;
-	compat_xfs_fsop_setdm_handlereq_t dmhreq;
-	struct dentry		*dentry;
-
-	if (!capable(CAP_MKNOD))
-		return -EPERM;
-	if (copy_from_user(&dmhreq, arg,
-			   sizeof(compat_xfs_fsop_setdm_handlereq_t)))
-		return -EFAULT;
-
-	dentry = xfs_compat_handlereq_to_dentry(parfilp, &dmhreq.hreq);
-	if (IS_ERR(dentry))
-		return PTR_ERR(dentry);
-
-	if (IS_IMMUTABLE(d_inode(dentry)) || IS_APPEND(d_inode(dentry))) {
-		error = -EPERM;
-		goto out;
-	}
-
-	if (copy_from_user(&fsd, compat_ptr(dmhreq.data), sizeof(fsd))) {
-		error = -EFAULT;
-		goto out;
-	}
-
-	error = xfs_set_dmattrs(XFS_I(d_inode(dentry)), fsd.fsd_dmevmask,
-				 fsd.fsd_dmstate);
-
-out:
-	dput(dentry);
-	return error;
-}
-
 long
 xfs_file_compat_ioctl(
 	struct file		*filp,
@@ -557,18 +447,13 @@ xfs_file_compat_ioctl(
 	case XFS_IOC_ALLOCSP_32:
 	case XFS_IOC_FREESP_32:
 	case XFS_IOC_ALLOCSP64_32:
-	case XFS_IOC_FREESP64_32:
-	case XFS_IOC_RESVSP_32:
-	case XFS_IOC_UNRESVSP_32:
-	case XFS_IOC_RESVSP64_32:
-	case XFS_IOC_UNRESVSP64_32:
-	case XFS_IOC_ZERO_RANGE_32: {
+	case XFS_IOC_FREESP64_32: {
 		struct xfs_flock64	bf;
 
 		if (xfs_compat_flock64_copyin(&bf, arg))
 			return -EFAULT;
 		cmd = _NATIVE_IOC(cmd, struct xfs_flock64);
-		return xfs_ioc_space(filp, cmd, &bf);
+		return xfs_ioc_space(filp, &bf);
 	}
 	case XFS_IOC_FSGEOMETRY_V1_32:
 		return xfs_compat_ioc_fsgeometry_v1(mp, arg);
@@ -651,8 +536,6 @@ xfs_file_compat_ioctl(
 		return xfs_compat_attrlist_by_handle(filp, arg);
 	case XFS_IOC_ATTRMULTI_BY_HANDLE_32:
 		return xfs_compat_attrmulti_by_handle(filp, arg);
-	case XFS_IOC_FSSETDM_BY_HANDLE_32:
-		return xfs_compat_fssetdm_by_handle(filp, arg);
 	default:
 		/* try the native version */
 		return xfs_file_ioctl(filp, cmd, (unsigned long)arg);
diff --git a/fs/xfs/xfs_ioctl32.h b/fs/xfs/xfs_ioctl32.h
index 7985344d3aa6..053de7d894cd 100644
--- a/fs/xfs/xfs_ioctl32.h
+++ b/fs/xfs/xfs_ioctl32.h
@@ -32,7 +32,7 @@
 #endif
 
 typedef struct compat_xfs_bstime {
-	compat_time_t	tv_sec;		/* seconds		*/
+	old_time32_t	tv_sec;		/* seconds		*/
 	__s32		tv_nsec;	/* and nanoseconds	*/
 } compat_xfs_bstime_t;
 
@@ -99,7 +99,7 @@ typedef struct compat_xfs_fsop_handlereq {
 	_IOWR('X', 108, struct compat_xfs_fsop_handlereq)
 
 /* The bstat field in the swapext struct needs translation */
-typedef struct compat_xfs_swapext {
+struct compat_xfs_swapext {
 	int64_t			sx_version;	/* version */
 	int64_t			sx_fdtarget;	/* fd of target file */
 	int64_t			sx_fdtmp;	/* fd of tmp file */
@@ -107,7 +107,7 @@ typedef struct compat_xfs_swapext {
 	xfs_off_t		sx_length;	/* leng from offset */
 	char			sx_pad[16];	/* pad space, unused */
 	struct compat_xfs_bstat	sx_stat;	/* stat of target b4 copy */
-} __compat_packed compat_xfs_swapext_t;
+} __compat_packed;
 
 #define XFS_IOC_SWAPEXT_32	_IOWR('X', 109, struct compat_xfs_swapext)
 
@@ -143,15 +143,6 @@ typedef struct compat_xfs_fsop_attrmulti_handlereq {
 #define XFS_IOC_ATTRMULTI_BY_HANDLE_32 \
 	_IOW('X', 123, struct compat_xfs_fsop_attrmulti_handlereq)
 
-typedef struct compat_xfs_fsop_setdm_handlereq {
-	struct compat_xfs_fsop_handlereq hreq;	/* handle information   */
-	/* ptr to struct fsdmidata */
-	compat_uptr_t			data;	/* DMAPI data   */
-} compat_xfs_fsop_setdm_handlereq_t;
-
-#define XFS_IOC_FSSETDM_BY_HANDLE_32 \
-	_IOW('X', 121, struct compat_xfs_fsop_setdm_handlereq)
-
 #ifdef BROKEN_X86_ALIGNMENT
 /* on ia32 l_start is on a 32-bit boundary */
 typedef struct compat_xfs_flock64 {
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index f780e223b118..bb590a267a7f 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -29,8 +29,8 @@
 #include "xfs_reflink.h"
 
 
-#define XFS_WRITEIO_ALIGN(mp,off)	(((off) >> mp->m_writeio_log) \
-						<< mp->m_writeio_log)
+#define XFS_ALLOC_ALIGN(mp, off) \
+	(((off) >> mp->m_allocsize_log) << mp->m_allocsize_log)
 
 static int
 xfs_alert_fsblock_zero(
@@ -54,9 +54,10 @@ xfs_bmbt_to_iomap(
 	struct xfs_inode	*ip,
 	struct iomap		*iomap,
 	struct xfs_bmbt_irec	*imap,
-	bool			shared)
+	u16			flags)
 {
 	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
 
 	if (unlikely(!xfs_valid_startblock(ip, imap->br_startblock)))
 		return xfs_alert_fsblock_zero(ip, imap);
@@ -77,14 +78,13 @@ xfs_bmbt_to_iomap(
 	}
 	iomap->offset = XFS_FSB_TO_B(mp, imap->br_startoff);
 	iomap->length = XFS_FSB_TO_B(mp, imap->br_blockcount);
-	iomap->bdev = xfs_find_bdev_for_inode(VFS_I(ip));
-	iomap->dax_dev = xfs_find_daxdev_for_inode(VFS_I(ip));
+	iomap->bdev = target->bt_bdev;
+	iomap->dax_dev = target->bt_daxdev;
+	iomap->flags = flags;
 
 	if (xfs_ipincount(ip) &&
 	    (ip->i_itemp->ili_fsync_fields & ~XFS_ILOG_TIMESTAMP))
 		iomap->flags |= IOMAP_F_DIRTY;
-	if (shared)
-		iomap->flags |= IOMAP_F_SHARED;
 	return 0;
 }
 
@@ -95,18 +95,30 @@ xfs_hole_to_iomap(
 	xfs_fileoff_t		offset_fsb,
 	xfs_fileoff_t		end_fsb)
 {
+	struct xfs_buftarg	*target = xfs_inode_buftarg(ip);
+
 	iomap->addr = IOMAP_NULL_ADDR;
 	iomap->type = IOMAP_HOLE;
 	iomap->offset = XFS_FSB_TO_B(ip->i_mount, offset_fsb);
 	iomap->length = XFS_FSB_TO_B(ip->i_mount, end_fsb - offset_fsb);
-	iomap->bdev = xfs_find_bdev_for_inode(VFS_I(ip));
-	iomap->dax_dev = xfs_find_daxdev_for_inode(VFS_I(ip));
+	iomap->bdev = target->bt_bdev;
+	iomap->dax_dev = target->bt_daxdev;
+}
+
+static inline xfs_fileoff_t
+xfs_iomap_end_fsb(
+	struct xfs_mount	*mp,
+	loff_t			offset,
+	loff_t			count)
+{
+	ASSERT(offset <= mp->m_super->s_maxbytes);
+	return min(XFS_B_TO_FSB(mp, offset + count),
+		   XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes));
 }
 
-xfs_extlen_t
+static xfs_extlen_t
 xfs_eof_alignment(
-	struct xfs_inode	*ip,
-	xfs_extlen_t		extsize)
+	struct xfs_inode	*ip)
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	xfs_extlen_t		align = 0;
@@ -129,111 +141,80 @@ xfs_eof_alignment(
 			align = 0;
 	}
 
-	/*
-	 * Always round up the allocation request to an extent boundary
-	 * (when file on a real-time subvolume or has di_extsize hint).
-	 */
-	if (extsize) {
-		if (align)
-			align = roundup_64(align, extsize);
-		else
-			align = extsize;
-	}
-
 	return align;
 }
 
-STATIC int
+/*
+ * Check if last_fsb is outside the last extent, and if so grow it to the next
+ * stripe unit boundary.
+ */
+xfs_fileoff_t
 xfs_iomap_eof_align_last_fsb(
 	struct xfs_inode	*ip,
-	xfs_extlen_t		extsize,
-	xfs_fileoff_t		*last_fsb)
+	xfs_fileoff_t		end_fsb)
 {
-	xfs_extlen_t		align = xfs_eof_alignment(ip, extsize);
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+	xfs_extlen_t		extsz = xfs_get_extsz_hint(ip);
+	xfs_extlen_t		align = xfs_eof_alignment(ip);
+	struct xfs_bmbt_irec	irec;
+	struct xfs_iext_cursor	icur;
+
+	ASSERT(ifp->if_flags & XFS_IFEXTENTS);
+
+	/*
+	 * Always round up the allocation request to the extent hint boundary.
+	 */
+	if (extsz) {
+		if (align)
+			align = roundup_64(align, extsz);
+		else
+			align = extsz;
+	}
 
 	if (align) {
-		xfs_fileoff_t	new_last_fsb = roundup_64(*last_fsb, align);
-		int		eof, error;
+		xfs_fileoff_t	aligned_end_fsb = roundup_64(end_fsb, align);
 
-		error = xfs_bmap_eof(ip, new_last_fsb, XFS_DATA_FORK, &eof);
-		if (error)
-			return error;
-		if (eof)
-			*last_fsb = new_last_fsb;
+		xfs_iext_last(ifp, &icur);
+		if (!xfs_iext_get_extent(ifp, &icur, &irec) ||
+		    aligned_end_fsb >= irec.br_startoff + irec.br_blockcount)
+			return aligned_end_fsb;
 	}
-	return 0;
+
+	return end_fsb;
 }
 
 int
 xfs_iomap_write_direct(
-	xfs_inode_t	*ip,
-	xfs_off_t	offset,
-	size_t		count,
-	xfs_bmbt_irec_t *imap,
-	int		nmaps)
+	struct xfs_inode	*ip,
+	xfs_fileoff_t		offset_fsb,
+	xfs_fileoff_t		count_fsb,
+	struct xfs_bmbt_irec	*imap)
 {
-	xfs_mount_t	*mp = ip->i_mount;
-	xfs_fileoff_t	offset_fsb;
-	xfs_fileoff_t	last_fsb;
-	xfs_filblks_t	count_fsb, resaligned;
-	xfs_extlen_t	extsz;
-	int		nimaps;
-	int		quota_flag;
-	int		rt;
-	xfs_trans_t	*tp;
-	uint		qblocks, resblks, resrtextents;
-	int		error;
-	int		lockmode;
-	int		bmapi_flags = XFS_BMAPI_PREALLOC;
-	uint		tflags = 0;
-
-	rt = XFS_IS_REALTIME_INODE(ip);
-	extsz = xfs_get_extsz_hint(ip);
-	lockmode = XFS_ILOCK_SHARED;	/* locked by caller */
-
-	ASSERT(xfs_isilocked(ip, lockmode));
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_trans	*tp;
+	xfs_filblks_t		resaligned;
+	int			nimaps;
+	int			quota_flag;
+	uint			qblocks, resblks;
+	unsigned int		resrtextents = 0;
+	int			error;
+	int			bmapi_flags = XFS_BMAPI_PREALLOC;
+	uint			tflags = 0;
 
-	offset_fsb = XFS_B_TO_FSBT(mp, offset);
-	last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
-	if ((offset + count) > XFS_ISIZE(ip)) {
-		/*
-		 * Assert that the in-core extent list is present since this can
-		 * call xfs_iread_extents() and we only have the ilock shared.
-		 * This should be safe because the lock was held around a bmapi
-		 * call in the caller and we only need it to access the in-core
-		 * list.
-		 */
-		ASSERT(XFS_IFORK_PTR(ip, XFS_DATA_FORK)->if_flags &
-								XFS_IFEXTENTS);
-		error = xfs_iomap_eof_align_last_fsb(ip, extsz, &last_fsb);
-		if (error)
-			goto out_unlock;
-	} else {
-		if (nmaps && (imap->br_startblock == HOLESTARTBLOCK))
-			last_fsb = min(last_fsb, (xfs_fileoff_t)
-					imap->br_blockcount +
-					imap->br_startoff);
-	}
-	count_fsb = last_fsb - offset_fsb;
 	ASSERT(count_fsb > 0);
-	resaligned = xfs_aligned_fsb_count(offset_fsb, count_fsb, extsz);
 
-	if (unlikely(rt)) {
+	resaligned = xfs_aligned_fsb_count(offset_fsb, count_fsb,
+					   xfs_get_extsz_hint(ip));
+	if (unlikely(XFS_IS_REALTIME_INODE(ip))) {
 		resrtextents = qblocks = resaligned;
 		resrtextents /= mp->m_sb.sb_rextsize;
 		resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
 		quota_flag = XFS_QMOPT_RES_RTBLKS;
 	} else {
-		resrtextents = 0;
 		resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned);
 		quota_flag = XFS_QMOPT_RES_REGBLKS;
 	}
 
-	/*
-	 * Drop the shared lock acquired by the caller, attach the dquot if
-	 * necessary and move on to transaction setup.
-	 */
-	xfs_iunlock(ip, lockmode);
 	error = xfs_qm_dqattach(ip);
 	if (error)
 		return error;
@@ -263,8 +244,7 @@ xfs_iomap_write_direct(
 	if (error)
 		return error;
 
-	lockmode = XFS_ILOCK_EXCL;
-	xfs_ilock(ip, lockmode);
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
 
 	error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
 	if (error)
@@ -277,8 +257,8 @@ xfs_iomap_write_direct(
 	 * caller gave to us.
 	 */
 	nimaps = 1;
-	error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb,
-				bmapi_flags, resblks, imap, &nimaps);
+	error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb, bmapi_flags, 0,
+				imap, &nimaps);
 	if (error)
 		goto out_res_cancel;
 
@@ -301,7 +281,7 @@ xfs_iomap_write_direct(
 		error = xfs_alert_fsblock_zero(ip, imap);
 
 out_unlock:
-	xfs_iunlock(ip, lockmode);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return error;
 
 out_res_cancel:
@@ -410,19 +390,19 @@ xfs_iomap_prealloc_size(
 	if (offset + count <= XFS_ISIZE(ip))
 		return 0;
 
-	if (!(mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) &&
-	    (XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_writeio_blocks)))
+	if (!(mp->m_flags & XFS_MOUNT_ALLOCSIZE) &&
+	    (XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_allocsize_blocks)))
 		return 0;
 
 	/*
 	 * If an explicit allocsize is set, the file is small, or we
 	 * are writing behind a hole, then use the minimum prealloc:
 	 */
-	if ((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ||
+	if ((mp->m_flags & XFS_MOUNT_ALLOCSIZE) ||
 	    XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_dalign) ||
 	    !xfs_iext_peek_prev_extent(ifp, icur, &prev) ||
 	    prev.br_startoff + prev.br_blockcount < offset_fsb)
-		return mp->m_writeio_blocks;
+		return mp->m_allocsize_blocks;
 
 	/*
 	 * Determine the initial size of the preallocation. We are beyond the
@@ -515,219 +495,13 @@ xfs_iomap_prealloc_size(
 	while (alloc_blocks && alloc_blocks >= freesp)
 		alloc_blocks >>= 4;
 check_writeio:
-	if (alloc_blocks < mp->m_writeio_blocks)
-		alloc_blocks = mp->m_writeio_blocks;
+	if (alloc_blocks < mp->m_allocsize_blocks)
+		alloc_blocks = mp->m_allocsize_blocks;
 	trace_xfs_iomap_prealloc_size(ip, alloc_blocks, shift,
-				      mp->m_writeio_blocks);
+				      mp->m_allocsize_blocks);
 	return alloc_blocks;
 }
 
-static int
-xfs_file_iomap_begin_delay(
-	struct inode		*inode,
-	loff_t			offset,
-	loff_t			count,
-	unsigned		flags,
-	struct iomap		*iomap)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
-	xfs_fileoff_t		maxbytes_fsb =
-		XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes);
-	xfs_fileoff_t		end_fsb;
-	struct xfs_bmbt_irec	imap, cmap;
-	struct xfs_iext_cursor	icur, ccur;
-	xfs_fsblock_t		prealloc_blocks = 0;
-	bool			eof = false, cow_eof = false, shared = false;
-	int			whichfork = XFS_DATA_FORK;
-	int			error = 0;
-
-	ASSERT(!XFS_IS_REALTIME_INODE(ip));
-	ASSERT(!xfs_get_extsz_hint(ip));
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-
-	if (unlikely(XFS_TEST_ERROR(
-	    (XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_EXTENTS &&
-	     XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE),
-	     mp, XFS_ERRTAG_BMAPIFORMAT))) {
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		error = -EFSCORRUPTED;
-		goto out_unlock;
-	}
-
-	XFS_STATS_INC(mp, xs_blk_mapw);
-
-	if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) {
-		error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
-		if (error)
-			goto out_unlock;
-	}
-
-	end_fsb = min(XFS_B_TO_FSB(mp, offset + count), maxbytes_fsb);
-
-	/*
-	 * Search the data fork fork first to look up our source mapping.  We
-	 * always need the data fork map, as we have to return it to the
-	 * iomap code so that the higher level write code can read data in to
-	 * perform read-modify-write cycles for unaligned writes.
-	 */
-	eof = !xfs_iext_lookup_extent(ip, &ip->i_df, offset_fsb, &icur, &imap);
-	if (eof)
-		imap.br_startoff = end_fsb; /* fake hole until the end */
-
-	/* We never need to allocate blocks for zeroing a hole. */
-	if ((flags & IOMAP_ZERO) && imap.br_startoff > offset_fsb) {
-		xfs_hole_to_iomap(ip, iomap, offset_fsb, imap.br_startoff);
-		goto out_unlock;
-	}
-
-	/*
-	 * Search the COW fork extent list even if we did not find a data fork
-	 * extent.  This serves two purposes: first this implements the
-	 * speculative preallocation using cowextsize, so that we also unshare
-	 * block adjacent to shared blocks instead of just the shared blocks
-	 * themselves.  Second the lookup in the extent list is generally faster
-	 * than going out to the shared extent tree.
-	 */
-	if (xfs_is_cow_inode(ip)) {
-		if (!ip->i_cowfp) {
-			ASSERT(!xfs_is_reflink_inode(ip));
-			xfs_ifork_init_cow(ip);
-		}
-		cow_eof = !xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb,
-				&ccur, &cmap);
-		if (!cow_eof && cmap.br_startoff <= offset_fsb) {
-			trace_xfs_reflink_cow_found(ip, &cmap);
-			whichfork = XFS_COW_FORK;
-			goto done;
-		}
-	}
-
-	if (imap.br_startoff <= offset_fsb) {
-		/*
-		 * For reflink files we may need a delalloc reservation when
-		 * overwriting shared extents.   This includes zeroing of
-		 * existing extents that contain data.
-		 */
-		if (!xfs_is_cow_inode(ip) ||
-		    ((flags & IOMAP_ZERO) && imap.br_state != XFS_EXT_NORM)) {
-			trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
-					&imap);
-			goto done;
-		}
-
-		xfs_trim_extent(&imap, offset_fsb, end_fsb - offset_fsb);
-
-		/* Trim the mapping to the nearest shared extent boundary. */
-		error = xfs_inode_need_cow(ip, &imap, &shared);
-		if (error)
-			goto out_unlock;
-
-		/* Not shared?  Just report the (potentially capped) extent. */
-		if (!shared) {
-			trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
-					&imap);
-			goto done;
-		}
-
-		/*
-		 * Fork all the shared blocks from our write offset until the
-		 * end of the extent.
-		 */
-		whichfork = XFS_COW_FORK;
-		end_fsb = imap.br_startoff + imap.br_blockcount;
-	} else {
-		/*
-		 * We cap the maximum length we map here to MAX_WRITEBACK_PAGES
-		 * pages to keep the chunks of work done where somewhat
-		 * symmetric with the work writeback does.  This is a completely
-		 * arbitrary number pulled out of thin air.
-		 *
-		 * Note that the values needs to be less than 32-bits wide until
-		 * the lower level functions are updated.
-		 */
-		count = min_t(loff_t, count, 1024 * PAGE_SIZE);
-		end_fsb = min(XFS_B_TO_FSB(mp, offset + count), maxbytes_fsb);
-
-		if (xfs_is_always_cow_inode(ip))
-			whichfork = XFS_COW_FORK;
-	}
-
-	error = xfs_qm_dqattach_locked(ip, false);
-	if (error)
-		goto out_unlock;
-
-	if (eof) {
-		prealloc_blocks = xfs_iomap_prealloc_size(ip, whichfork, offset,
-				count, &icur);
-		if (prealloc_blocks) {
-			xfs_extlen_t	align;
-			xfs_off_t	end_offset;
-			xfs_fileoff_t	p_end_fsb;
-
-			end_offset = XFS_WRITEIO_ALIGN(mp, offset + count - 1);
-			p_end_fsb = XFS_B_TO_FSBT(mp, end_offset) +
-					prealloc_blocks;
-
-			align = xfs_eof_alignment(ip, 0);
-			if (align)
-				p_end_fsb = roundup_64(p_end_fsb, align);
-
-			p_end_fsb = min(p_end_fsb, maxbytes_fsb);
-			ASSERT(p_end_fsb > offset_fsb);
-			prealloc_blocks = p_end_fsb - end_fsb;
-		}
-	}
-
-retry:
-	error = xfs_bmapi_reserve_delalloc(ip, whichfork, offset_fsb,
-			end_fsb - offset_fsb, prealloc_blocks,
-			whichfork == XFS_DATA_FORK ? &imap : &cmap,
-			whichfork == XFS_DATA_FORK ? &icur : &ccur,
-			whichfork == XFS_DATA_FORK ? eof : cow_eof);
-	switch (error) {
-	case 0:
-		break;
-	case -ENOSPC:
-	case -EDQUOT:
-		/* retry without any preallocation */
-		trace_xfs_delalloc_enospc(ip, offset, count);
-		if (prealloc_blocks) {
-			prealloc_blocks = 0;
-			goto retry;
-		}
-		/*FALLTHRU*/
-	default:
-		goto out_unlock;
-	}
-
-	/*
-	 * Flag newly allocated delalloc blocks with IOMAP_F_NEW so we punch
-	 * them out if the write happens to fail.
-	 */
-	iomap->flags |= IOMAP_F_NEW;
-	trace_xfs_iomap_alloc(ip, offset, count, whichfork,
-			whichfork == XFS_DATA_FORK ? &imap : &cmap);
-done:
-	if (whichfork == XFS_COW_FORK) {
-		if (imap.br_startoff > offset_fsb) {
-			xfs_trim_extent(&cmap, offset_fsb,
-					imap.br_startoff - offset_fsb);
-			error = xfs_bmbt_to_iomap(ip, iomap, &cmap, true);
-			goto out_unlock;
-		}
-		/* ensure we only report blocks we have a reservation for */
-		xfs_trim_extent(&imap, cmap.br_startoff, cmap.br_blockcount);
-		shared = true;
-	}
-	error = xfs_bmbt_to_iomap(ip, iomap, &imap, shared);
-out_unlock:
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	return error;
-}
-
 int
 xfs_iomap_write_unwritten(
 	xfs_inode_t	*ip,
@@ -765,6 +539,11 @@ xfs_iomap_write_unwritten(
 	 */
 	resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
 
+	/* Attach dquots so that bmbt splits are accounted correctly. */
+	error = xfs_qm_dqattach(ip);
+	if (error)
+		return error;
+
 	do {
 		/*
 		 * Set up a transaction to convert the range of extents
@@ -783,6 +562,11 @@ xfs_iomap_write_unwritten(
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
 		xfs_trans_ijoin(tp, ip, 0);
 
+		error = xfs_trans_reserve_quota_nblks(tp, ip, resblks, 0,
+				XFS_QMOPT_RES_REGBLKS);
+		if (error)
+			goto error_on_bmapi_transaction;
+
 		/*
 		 * Modify the unwritten extent state of the buffer.
 		 */
@@ -840,23 +624,42 @@ error_on_bmapi_transaction:
 static inline bool
 imap_needs_alloc(
 	struct inode		*inode,
+	unsigned		flags,
 	struct xfs_bmbt_irec	*imap,
 	int			nimaps)
 {
-	return !nimaps ||
-		imap->br_startblock == HOLESTARTBLOCK ||
-		imap->br_startblock == DELAYSTARTBLOCK ||
-		(IS_DAX(inode) && imap->br_state == XFS_EXT_UNWRITTEN);
+	/* don't allocate blocks when just zeroing */
+	if (flags & IOMAP_ZERO)
+		return false;
+	if (!nimaps ||
+	    imap->br_startblock == HOLESTARTBLOCK ||
+	    imap->br_startblock == DELAYSTARTBLOCK)
+		return true;
+	/* we convert unwritten extents before copying the data for DAX */
+	if (IS_DAX(inode) && imap->br_state == XFS_EXT_UNWRITTEN)
+		return true;
+	return false;
 }
 
 static inline bool
-needs_cow_for_zeroing(
+imap_needs_cow(
+	struct xfs_inode	*ip,
+	unsigned int		flags,
 	struct xfs_bmbt_irec	*imap,
 	int			nimaps)
 {
-	return nimaps &&
-		imap->br_startblock != HOLESTARTBLOCK &&
-		imap->br_state != XFS_EXT_UNWRITTEN;
+	if (!xfs_is_cow_inode(ip))
+		return false;
+
+	/* when zeroing we don't have to COW holes or unwritten extents */
+	if (flags & IOMAP_ZERO) {
+		if (!nimaps ||
+		    imap->br_startblock == HOLESTARTBLOCK ||
+		    imap->br_state == XFS_EXT_UNWRITTEN)
+			return false;
+	}
+
+	return true;
 }
 
 static int
@@ -872,15 +675,8 @@ xfs_ilock_for_iomap(
 	 * COW writes may allocate delalloc space or convert unwritten COW
 	 * extents, so we need to make sure to take the lock exclusively here.
 	 */
-	if (xfs_is_cow_inode(ip) && is_write) {
-		/*
-		 * FIXME: It could still overwrite on unshared extents and not
-		 * need allocation.
-		 */
-		if (flags & IOMAP_NOWAIT)
-			return -EAGAIN;
+	if (xfs_is_cow_inode(ip) && is_write)
 		mode = XFS_ILOCK_EXCL;
-	}
 
 	/*
 	 * Extents not yet cached requires exclusive access, don't block.  This
@@ -917,111 +713,73 @@ relock:
 }
 
 static int
-xfs_file_iomap_begin(
+xfs_direct_write_iomap_begin(
 	struct inode		*inode,
 	loff_t			offset,
 	loff_t			length,
 	unsigned		flags,
-	struct iomap		*iomap)
+	struct iomap		*iomap,
+	struct iomap		*srcmap)
 {
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
-	struct xfs_bmbt_irec	imap;
-	xfs_fileoff_t		offset_fsb, end_fsb;
+	struct xfs_bmbt_irec	imap, cmap;
+	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
+	xfs_fileoff_t		end_fsb = xfs_iomap_end_fsb(mp, offset, length);
 	int			nimaps = 1, error = 0;
 	bool			shared = false;
+	u16			iomap_flags = 0;
 	unsigned		lockmode;
 
+	ASSERT(flags & (IOMAP_WRITE | IOMAP_ZERO));
+
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
 
-	if ((flags & (IOMAP_WRITE | IOMAP_ZERO)) && !(flags & IOMAP_DIRECT) &&
-			!IS_DAX(inode) && !xfs_get_extsz_hint(ip)) {
-		/* Reserve delalloc blocks for regular writeback. */
-		return xfs_file_iomap_begin_delay(inode, offset, length, flags,
-				iomap);
-	}
-
 	/*
-	 * Lock the inode in the manner required for the specified operation and
-	 * check for as many conditions that would result in blocking as
-	 * possible. This removes most of the non-blocking checks from the
-	 * mapping code below.
+	 * Writes that span EOF might trigger an IO size update on completion,
+	 * so consider them to be dirty for the purposes of O_DSYNC even if
+	 * there is no other metadata changes pending or have been made here.
 	 */
+	if (offset + length > i_size_read(inode))
+		iomap_flags |= IOMAP_F_DIRTY;
+
 	error = xfs_ilock_for_iomap(ip, flags, &lockmode);
 	if (error)
 		return error;
 
-	ASSERT(offset <= mp->m_super->s_maxbytes);
-	if (offset > mp->m_super->s_maxbytes - length)
-		length = mp->m_super->s_maxbytes - offset;
-	offset_fsb = XFS_B_TO_FSBT(mp, offset);
-	end_fsb = XFS_B_TO_FSB(mp, offset + length);
-
 	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap,
 			       &nimaps, 0);
 	if (error)
 		goto out_unlock;
 
-	if (flags & IOMAP_REPORT) {
-		/* Trim the mapping to the nearest shared extent boundary. */
-		error = xfs_reflink_trim_around_shared(ip, &imap, &shared);
-		if (error)
+	if (imap_needs_cow(ip, flags, &imap, nimaps)) {
+		error = -EAGAIN;
+		if (flags & IOMAP_NOWAIT)
 			goto out_unlock;
-	}
-
-	/* Non-modifying mapping requested, so we are done */
-	if (!(flags & (IOMAP_WRITE | IOMAP_ZERO)))
-		goto out_found;
-
-	/*
-	 * Break shared extents if necessary. Checks for non-blocking IO have
-	 * been done up front, so we don't need to do them here.
-	 */
-	if (xfs_is_cow_inode(ip)) {
-		struct xfs_bmbt_irec	cmap;
-		bool			directio = (flags & IOMAP_DIRECT);
-
-		/* if zeroing doesn't need COW allocation, then we are done. */
-		if ((flags & IOMAP_ZERO) &&
-		    !needs_cow_for_zeroing(&imap, nimaps))
-			goto out_found;
 
 		/* may drop and re-acquire the ilock */
-		cmap = imap;
-		error = xfs_reflink_allocate_cow(ip, &cmap, &shared, &lockmode,
-				directio);
+		error = xfs_reflink_allocate_cow(ip, &imap, &cmap, &shared,
+				&lockmode, flags & IOMAP_DIRECT);
 		if (error)
 			goto out_unlock;
-
-		/*
-		 * For buffered writes we need to report the address of the
-		 * previous block (if there was any) so that the higher level
-		 * write code can perform read-modify-write operations; we
-		 * won't need the CoW fork mapping until writeback.  For direct
-		 * I/O, which must be block aligned, we need to report the
-		 * newly allocated address.  If the data fork has a hole, copy
-		 * the COW fork mapping to avoid allocating to the data fork.
-		 */
-		if (directio || imap.br_startblock == HOLESTARTBLOCK)
-			imap = cmap;
-
+		if (shared)
+			goto out_found_cow;
 		end_fsb = imap.br_startoff + imap.br_blockcount;
 		length = XFS_FSB_TO_B(mp, end_fsb) - offset;
 	}
 
-	/* Don't need to allocate over holes when doing zeroing operations. */
-	if (flags & IOMAP_ZERO)
-		goto out_found;
+	if (imap_needs_alloc(inode, flags, &imap, nimaps))
+		goto allocate_blocks;
 
-	if (!imap_needs_alloc(inode, &imap, nimaps))
-		goto out_found;
+	xfs_iunlock(ip, lockmode);
+	trace_xfs_iomap_found(ip, offset, length, XFS_DATA_FORK, &imap);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, iomap_flags);
 
-	/* If nowait is set bail since we are going to make allocations. */
-	if (flags & IOMAP_NOWAIT) {
-		error = -EAGAIN;
+allocate_blocks:
+	error = -EAGAIN;
+	if (flags & IOMAP_NOWAIT)
 		goto out_unlock;
-	}
 
 	/*
 	 * We cap the maximum length we map to a sane size  to keep the chunks
@@ -1033,48 +791,273 @@ xfs_file_iomap_begin(
 	 * lower level functions are updated.
 	 */
 	length = min_t(loff_t, length, 1024 * PAGE_SIZE);
+	end_fsb = xfs_iomap_end_fsb(mp, offset, length);
 
-	/*
-	 * xfs_iomap_write_direct() expects the shared lock. It is unlocked on
-	 * return.
-	 */
-	if (lockmode == XFS_ILOCK_EXCL)
-		xfs_ilock_demote(ip, lockmode);
-	error = xfs_iomap_write_direct(ip, offset, length, &imap,
-			nimaps);
+	if (offset + length > XFS_ISIZE(ip))
+		end_fsb = xfs_iomap_eof_align_last_fsb(ip, end_fsb);
+	else if (nimaps && imap.br_startblock == HOLESTARTBLOCK)
+		end_fsb = min(end_fsb, imap.br_startoff + imap.br_blockcount);
+	xfs_iunlock(ip, lockmode);
+
+	error = xfs_iomap_write_direct(ip, offset_fsb, end_fsb - offset_fsb,
+			&imap);
 	if (error)
 		return error;
 
-	iomap->flags |= IOMAP_F_NEW;
 	trace_xfs_iomap_alloc(ip, offset, length, XFS_DATA_FORK, &imap);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, iomap_flags | IOMAP_F_NEW);
 
-out_finish:
-	return xfs_bmbt_to_iomap(ip, iomap, &imap, shared);
-
-out_found:
-	ASSERT(nimaps);
+out_found_cow:
 	xfs_iunlock(ip, lockmode);
-	trace_xfs_iomap_found(ip, offset, length, XFS_DATA_FORK, &imap);
-	goto out_finish;
+	length = XFS_FSB_TO_B(mp, cmap.br_startoff + cmap.br_blockcount);
+	trace_xfs_iomap_found(ip, offset, length - offset, XFS_COW_FORK, &cmap);
+	if (imap.br_startblock != HOLESTARTBLOCK) {
+		error = xfs_bmbt_to_iomap(ip, srcmap, &imap, 0);
+		if (error)
+			return error;
+	}
+	return xfs_bmbt_to_iomap(ip, iomap, &cmap, IOMAP_F_SHARED);
 
 out_unlock:
 	xfs_iunlock(ip, lockmode);
 	return error;
 }
 
+const struct iomap_ops xfs_direct_write_iomap_ops = {
+	.iomap_begin		= xfs_direct_write_iomap_begin,
+};
+
 static int
-xfs_file_iomap_end_delalloc(
-	struct xfs_inode	*ip,
+xfs_buffered_write_iomap_begin(
+	struct inode		*inode,
+	loff_t			offset,
+	loff_t			count,
+	unsigned		flags,
+	struct iomap		*iomap,
+	struct iomap		*srcmap)
+{
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
+	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
+	xfs_fileoff_t		end_fsb = xfs_iomap_end_fsb(mp, offset, count);
+	struct xfs_bmbt_irec	imap, cmap;
+	struct xfs_iext_cursor	icur, ccur;
+	xfs_fsblock_t		prealloc_blocks = 0;
+	bool			eof = false, cow_eof = false, shared = false;
+	int			allocfork = XFS_DATA_FORK;
+	int			error = 0;
+
+	/* we can't use delayed allocations when using extent size hints */
+	if (xfs_get_extsz_hint(ip))
+		return xfs_direct_write_iomap_begin(inode, offset, count,
+				flags, iomap, srcmap);
+
+	ASSERT(!XFS_IS_REALTIME_INODE(ip));
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+
+	if (XFS_IS_CORRUPT(mp, !xfs_ifork_has_extents(ip, XFS_DATA_FORK)) ||
+	    XFS_TEST_ERROR(false, mp, XFS_ERRTAG_BMAPIFORMAT)) {
+		error = -EFSCORRUPTED;
+		goto out_unlock;
+	}
+
+	XFS_STATS_INC(mp, xs_blk_mapw);
+
+	if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) {
+		error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
+		if (error)
+			goto out_unlock;
+	}
+
+	/*
+	 * Search the data fork fork first to look up our source mapping.  We
+	 * always need the data fork map, as we have to return it to the
+	 * iomap code so that the higher level write code can read data in to
+	 * perform read-modify-write cycles for unaligned writes.
+	 */
+	eof = !xfs_iext_lookup_extent(ip, &ip->i_df, offset_fsb, &icur, &imap);
+	if (eof)
+		imap.br_startoff = end_fsb; /* fake hole until the end */
+
+	/* We never need to allocate blocks for zeroing a hole. */
+	if ((flags & IOMAP_ZERO) && imap.br_startoff > offset_fsb) {
+		xfs_hole_to_iomap(ip, iomap, offset_fsb, imap.br_startoff);
+		goto out_unlock;
+	}
+
+	/*
+	 * Search the COW fork extent list even if we did not find a data fork
+	 * extent.  This serves two purposes: first this implements the
+	 * speculative preallocation using cowextsize, so that we also unshare
+	 * block adjacent to shared blocks instead of just the shared blocks
+	 * themselves.  Second the lookup in the extent list is generally faster
+	 * than going out to the shared extent tree.
+	 */
+	if (xfs_is_cow_inode(ip)) {
+		if (!ip->i_cowfp) {
+			ASSERT(!xfs_is_reflink_inode(ip));
+			xfs_ifork_init_cow(ip);
+		}
+		cow_eof = !xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb,
+				&ccur, &cmap);
+		if (!cow_eof && cmap.br_startoff <= offset_fsb) {
+			trace_xfs_reflink_cow_found(ip, &cmap);
+			goto found_cow;
+		}
+	}
+
+	if (imap.br_startoff <= offset_fsb) {
+		/*
+		 * For reflink files we may need a delalloc reservation when
+		 * overwriting shared extents.   This includes zeroing of
+		 * existing extents that contain data.
+		 */
+		if (!xfs_is_cow_inode(ip) ||
+		    ((flags & IOMAP_ZERO) && imap.br_state != XFS_EXT_NORM)) {
+			trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
+					&imap);
+			goto found_imap;
+		}
+
+		xfs_trim_extent(&imap, offset_fsb, end_fsb - offset_fsb);
+
+		/* Trim the mapping to the nearest shared extent boundary. */
+		error = xfs_bmap_trim_cow(ip, &imap, &shared);
+		if (error)
+			goto out_unlock;
+
+		/* Not shared?  Just report the (potentially capped) extent. */
+		if (!shared) {
+			trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
+					&imap);
+			goto found_imap;
+		}
+
+		/*
+		 * Fork all the shared blocks from our write offset until the
+		 * end of the extent.
+		 */
+		allocfork = XFS_COW_FORK;
+		end_fsb = imap.br_startoff + imap.br_blockcount;
+	} else {
+		/*
+		 * We cap the maximum length we map here to MAX_WRITEBACK_PAGES
+		 * pages to keep the chunks of work done where somewhat
+		 * symmetric with the work writeback does.  This is a completely
+		 * arbitrary number pulled out of thin air.
+		 *
+		 * Note that the values needs to be less than 32-bits wide until
+		 * the lower level functions are updated.
+		 */
+		count = min_t(loff_t, count, 1024 * PAGE_SIZE);
+		end_fsb = xfs_iomap_end_fsb(mp, offset, count);
+
+		if (xfs_is_always_cow_inode(ip))
+			allocfork = XFS_COW_FORK;
+	}
+
+	error = xfs_qm_dqattach_locked(ip, false);
+	if (error)
+		goto out_unlock;
+
+	if (eof) {
+		prealloc_blocks = xfs_iomap_prealloc_size(ip, allocfork, offset,
+				count, &icur);
+		if (prealloc_blocks) {
+			xfs_extlen_t	align;
+			xfs_off_t	end_offset;
+			xfs_fileoff_t	p_end_fsb;
+
+			end_offset = XFS_ALLOC_ALIGN(mp, offset + count - 1);
+			p_end_fsb = XFS_B_TO_FSBT(mp, end_offset) +
+					prealloc_blocks;
+
+			align = xfs_eof_alignment(ip);
+			if (align)
+				p_end_fsb = roundup_64(p_end_fsb, align);
+
+			p_end_fsb = min(p_end_fsb,
+				XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes));
+			ASSERT(p_end_fsb > offset_fsb);
+			prealloc_blocks = p_end_fsb - end_fsb;
+		}
+	}
+
+retry:
+	error = xfs_bmapi_reserve_delalloc(ip, allocfork, offset_fsb,
+			end_fsb - offset_fsb, prealloc_blocks,
+			allocfork == XFS_DATA_FORK ? &imap : &cmap,
+			allocfork == XFS_DATA_FORK ? &icur : &ccur,
+			allocfork == XFS_DATA_FORK ? eof : cow_eof);
+	switch (error) {
+	case 0:
+		break;
+	case -ENOSPC:
+	case -EDQUOT:
+		/* retry without any preallocation */
+		trace_xfs_delalloc_enospc(ip, offset, count);
+		if (prealloc_blocks) {
+			prealloc_blocks = 0;
+			goto retry;
+		}
+		/*FALLTHRU*/
+	default:
+		goto out_unlock;
+	}
+
+	if (allocfork == XFS_COW_FORK) {
+		trace_xfs_iomap_alloc(ip, offset, count, allocfork, &cmap);
+		goto found_cow;
+	}
+
+	/*
+	 * Flag newly allocated delalloc blocks with IOMAP_F_NEW so we punch
+	 * them out if the write happens to fail.
+	 */
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	trace_xfs_iomap_alloc(ip, offset, count, allocfork, &imap);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, IOMAP_F_NEW);
+
+found_imap:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, 0);
+
+found_cow:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	if (imap.br_startoff <= offset_fsb) {
+		error = xfs_bmbt_to_iomap(ip, srcmap, &imap, 0);
+		if (error)
+			return error;
+	} else {
+		xfs_trim_extent(&cmap, offset_fsb,
+				imap.br_startoff - offset_fsb);
+	}
+	return xfs_bmbt_to_iomap(ip, iomap, &cmap, IOMAP_F_SHARED);
+
+out_unlock:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return error;
+}
+
+static int
+xfs_buffered_write_iomap_end(
+	struct inode		*inode,
 	loff_t			offset,
 	loff_t			length,
 	ssize_t			written,
+	unsigned		flags,
 	struct iomap		*iomap)
 {
+	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
 	xfs_fileoff_t		start_fsb;
 	xfs_fileoff_t		end_fsb;
 	int			error = 0;
 
+	if (iomap->type != IOMAP_DELALLOC)
+		return 0;
+
 	/*
 	 * Behave as if the write failed if drop writes is enabled. Set the NEW
 	 * flag to force delalloc cleanup.
@@ -1119,24 +1102,51 @@ xfs_file_iomap_end_delalloc(
 	return 0;
 }
 
+const struct iomap_ops xfs_buffered_write_iomap_ops = {
+	.iomap_begin		= xfs_buffered_write_iomap_begin,
+	.iomap_end		= xfs_buffered_write_iomap_end,
+};
+
 static int
-xfs_file_iomap_end(
+xfs_read_iomap_begin(
 	struct inode		*inode,
 	loff_t			offset,
 	loff_t			length,
-	ssize_t			written,
 	unsigned		flags,
-	struct iomap		*iomap)
+	struct iomap		*iomap,
+	struct iomap		*srcmap)
 {
-	if ((flags & IOMAP_WRITE) && iomap->type == IOMAP_DELALLOC)
-		return xfs_file_iomap_end_delalloc(XFS_I(inode), offset,
-				length, written, iomap);
-	return 0;
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_bmbt_irec	imap;
+	xfs_fileoff_t		offset_fsb = XFS_B_TO_FSBT(mp, offset);
+	xfs_fileoff_t		end_fsb = xfs_iomap_end_fsb(mp, offset, length);
+	int			nimaps = 1, error = 0;
+	bool			shared = false;
+	unsigned		lockmode;
+
+	ASSERT(!(flags & (IOMAP_WRITE | IOMAP_ZERO)));
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
+	error = xfs_ilock_for_iomap(ip, flags, &lockmode);
+	if (error)
+		return error;
+	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap,
+			       &nimaps, 0);
+	if (!error && (flags & IOMAP_REPORT))
+		error = xfs_reflink_trim_around_shared(ip, &imap, &shared);
+	xfs_iunlock(ip, lockmode);
+
+	if (error)
+		return error;
+	trace_xfs_iomap_found(ip, offset, length, XFS_DATA_FORK, &imap);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, shared ? IOMAP_F_SHARED : 0);
 }
 
-const struct iomap_ops xfs_iomap_ops = {
-	.iomap_begin		= xfs_file_iomap_begin,
-	.iomap_end		= xfs_file_iomap_end,
+const struct iomap_ops xfs_read_iomap_ops = {
+	.iomap_begin		= xfs_read_iomap_begin,
 };
 
 static int
@@ -1145,7 +1155,8 @@ xfs_seek_iomap_begin(
 	loff_t			offset,
 	loff_t			length,
 	unsigned		flags,
-	struct iomap		*iomap)
+	struct iomap		*iomap,
+	struct iomap		*srcmap)
 {
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
@@ -1178,8 +1189,7 @@ xfs_seek_iomap_begin(
 		/*
 		 * Fake a hole until the end of the file.
 		 */
-		data_fsb = min(XFS_B_TO_FSB(mp, offset + length),
-			       XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes));
+		data_fsb = xfs_iomap_end_fsb(mp, offset, length);
 	}
 
 	/*
@@ -1193,7 +1203,7 @@ xfs_seek_iomap_begin(
 		if (data_fsb < cow_fsb + cmap.br_blockcount)
 			end_fsb = min(end_fsb, data_fsb);
 		xfs_trim_extent(&cmap, offset_fsb, end_fsb);
-		error = xfs_bmbt_to_iomap(ip, iomap, &cmap, true);
+		error = xfs_bmbt_to_iomap(ip, iomap, &cmap, IOMAP_F_SHARED);
 		/*
 		 * This is a COW extent, so we must probe the page cache
 		 * because there could be dirty page cache being backed
@@ -1215,7 +1225,7 @@ xfs_seek_iomap_begin(
 	imap.br_state = XFS_EXT_NORM;
 done:
 	xfs_trim_extent(&imap, offset_fsb, end_fsb);
-	error = xfs_bmbt_to_iomap(ip, iomap, &imap, false);
+	error = xfs_bmbt_to_iomap(ip, iomap, &imap, 0);
 out_unlock:
 	xfs_iunlock(ip, lockmode);
 	return error;
@@ -1231,7 +1241,8 @@ xfs_xattr_iomap_begin(
 	loff_t			offset,
 	loff_t			length,
 	unsigned		flags,
-	struct iomap		*iomap)
+	struct iomap		*iomap,
+	struct iomap		*srcmap)
 {
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
@@ -1261,7 +1272,7 @@ out_unlock:
 	if (error)
 		return error;
 	ASSERT(nimaps);
-	return xfs_bmbt_to_iomap(ip, iomap, &imap, false);
+	return xfs_bmbt_to_iomap(ip, iomap, &imap, 0);
 }
 
 const struct iomap_ops xfs_xattr_iomap_ops = {
diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h
index 5c2f6aa6d78f..7d3703556d0e 100644
--- a/fs/xfs/xfs_iomap.h
+++ b/fs/xfs/xfs_iomap.h
@@ -11,13 +11,14 @@
 struct xfs_inode;
 struct xfs_bmbt_irec;
 
-int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
-			struct xfs_bmbt_irec *, int);
+int xfs_iomap_write_direct(struct xfs_inode *ip, xfs_fileoff_t offset_fsb,
+		xfs_fileoff_t count_fsb, struct xfs_bmbt_irec *imap);
 int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, xfs_off_t, bool);
+xfs_fileoff_t xfs_iomap_eof_align_last_fsb(struct xfs_inode *ip,
+		xfs_fileoff_t end_fsb);
 
 int xfs_bmbt_to_iomap(struct xfs_inode *, struct iomap *,
-		struct xfs_bmbt_irec *, bool shared);
-xfs_extlen_t xfs_eof_alignment(struct xfs_inode *ip, xfs_extlen_t extsize);
+		struct xfs_bmbt_irec *, u16);
 
 static inline xfs_filblks_t
 xfs_aligned_fsb_count(
@@ -39,7 +40,9 @@ xfs_aligned_fsb_count(
 	return count_fsb;
 }
 
-extern const struct iomap_ops xfs_iomap_ops;
+extern const struct iomap_ops xfs_buffered_write_iomap_ops;
+extern const struct iomap_ops xfs_direct_write_iomap_ops;
+extern const struct iomap_ops xfs_read_iomap_ops;
 extern const struct iomap_ops xfs_seek_iomap_ops;
 extern const struct iomap_ops xfs_xattr_iomap_ops;
 
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index fe285d123d69..f7a99b3bbcf7 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -20,8 +20,8 @@
 #include "xfs_symlink.h"
 #include "xfs_dir2.h"
 #include "xfs_iomap.h"
+#include "xfs_error.h"
 
-#include <linux/xattr.h>
 #include <linux/posix_acl.h>
 #include <linux/security.h>
 #include <linux/iversion.h>
@@ -49,8 +49,15 @@ xfs_initxattrs(
 	int			error = 0;
 
 	for (xattr = xattr_array; xattr->name != NULL; xattr++) {
-		error = xfs_attr_set(ip, xattr->name, xattr->value,
-				      xattr->value_len, ATTR_SECURE);
+		struct xfs_da_args	args = {
+			.dp		= ip,
+			.attr_filter	= XFS_ATTR_SECURE,
+			.name		= xattr->name,
+			.namelen	= strlen(xattr->name),
+			.value		= xattr->value,
+			.valuelen	= xattr->value_len,
+		};
+		error = xfs_attr_set(&args);
 		if (error < 0)
 			break;
 	}
@@ -470,20 +477,57 @@ xfs_vn_get_link_inline(
 	struct inode		*inode,
 	struct delayed_call	*done)
 {
+	struct xfs_inode	*ip = XFS_I(inode);
 	char			*link;
 
-	ASSERT(XFS_I(inode)->i_df.if_flags & XFS_IFINLINE);
+	ASSERT(ip->i_df.if_flags & XFS_IFINLINE);
 
 	/*
 	 * The VFS crashes on a NULL pointer, so return -EFSCORRUPTED if
 	 * if_data is junk.
 	 */
-	link = XFS_I(inode)->i_df.if_u1.if_data;
-	if (!link)
+	link = ip->i_df.if_u1.if_data;
+	if (XFS_IS_CORRUPT(ip->i_mount, !link))
 		return ERR_PTR(-EFSCORRUPTED);
 	return link;
 }
 
+static uint32_t
+xfs_stat_blksize(
+	struct xfs_inode	*ip)
+{
+	struct xfs_mount	*mp = ip->i_mount;
+
+	/*
+	 * If the file blocks are being allocated from a realtime volume, then
+	 * always return the realtime extent size.
+	 */
+	if (XFS_IS_REALTIME_INODE(ip))
+		return xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog;
+
+	/*
+	 * Allow large block sizes to be reported to userspace programs if the
+	 * "largeio" mount option is used.
+	 *
+	 * If compatibility mode is specified, simply return the basic unit of
+	 * caching so that we don't get inefficient read/modify/write I/O from
+	 * user apps. Otherwise....
+	 *
+	 * If the underlying volume is a stripe, then return the stripe width in
+	 * bytes as the recommended I/O size. It is not a stripe and we've set a
+	 * default buffered I/O size, return that, otherwise return the compat
+	 * default.
+	 */
+	if (mp->m_flags & XFS_MOUNT_LARGEIO) {
+		if (mp->m_swidth)
+			return mp->m_swidth << mp->m_sb.sb_blocklog;
+		if (mp->m_flags & XFS_MOUNT_ALLOCSIZE)
+			return 1U << mp->m_allocsize_log;
+	}
+
+	return PAGE_SIZE;
+}
+
 STATIC int
 xfs_vn_getattr(
 	const struct path	*path,
@@ -513,11 +557,10 @@ xfs_vn_getattr(
 	stat->blocks =
 		XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks);
 
-	if (ip->i_d.di_version == 3) {
+	if (xfs_sb_version_has_v3inode(&mp->m_sb)) {
 		if (request_mask & STATX_BTIME) {
 			stat->result_mask |= STATX_BTIME;
-			stat->btime.tv_sec = ip->i_d.di_crtime.t_sec;
-			stat->btime.tv_nsec = ip->i_d.di_crtime.t_nsec;
+			stat->btime = ip->i_d.di_crtime;
 		}
 	}
 
@@ -543,16 +586,7 @@ xfs_vn_getattr(
 		stat->rdev = inode->i_rdev;
 		break;
 	default:
-		if (XFS_IS_REALTIME_INODE(ip)) {
-			/*
-			 * If the file blocks are being allocated from a
-			 * realtime volume, then return the inode's realtime
-			 * extent size or the realtime volume's extent size.
-			 */
-			stat->blksize =
-				xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog;
-		} else
-			stat->blksize = xfs_preferred_iosize(mp);
+		stat->blksize = xfs_stat_blksize(ip);
 		stat->rdev = 0;
 		break;
 	}
@@ -662,9 +696,7 @@ xfs_setattr_nonsize(
 		 */
 		ASSERT(udqp == NULL);
 		ASSERT(gdqp == NULL);
-		error = xfs_qm_vop_dqalloc(ip, xfs_kuid_to_uid(uid),
-					   xfs_kgid_to_gid(gid),
-					   xfs_get_projid(ip),
+		error = xfs_qm_vop_dqalloc(ip, uid, gid, ip->i_d.di_projid,
 					   qflags, &udqp, &gdqp, NULL);
 		if (error)
 			return error;
@@ -733,7 +765,6 @@ xfs_setattr_nonsize(
 				olddquot1 = xfs_qm_vop_chown(tp, ip,
 							&ip->i_udquot, udqp);
 			}
-			ip->i_d.di_uid = xfs_kuid_to_uid(uid);
 			inode->i_uid = uid;
 		}
 		if (!gid_eq(igid, gid)) {
@@ -745,7 +776,6 @@ xfs_setattr_nonsize(
 				olddquot2 = xfs_qm_vop_chown(tp, ip,
 							&ip->i_gdquot, gdqp);
 			}
-			ip->i_d.di_gid = xfs_kgid_to_gid(gid);
 			inode->i_gid = gid;
 		}
 	}
@@ -883,10 +913,10 @@ xfs_setattr_size(
 	if (newsize > oldsize) {
 		trace_xfs_zero_eof(ip, oldsize, newsize - oldsize);
 		error = iomap_zero_range(inode, oldsize, newsize - oldsize,
-				&did_zeroing, &xfs_iomap_ops);
+				&did_zeroing, &xfs_buffered_write_iomap_ops);
 	} else {
 		error = iomap_truncate_page(inode, newsize, &did_zeroing,
-				&xfs_iomap_ops);
+				&xfs_buffered_write_iomap_ops);
 	}
 
 	if (error)
@@ -1114,7 +1144,7 @@ xfs_vn_fiemap(
 				&xfs_xattr_iomap_ops);
 	} else {
 		error = iomap_fiemap(inode, fieinfo, start, length,
-				&xfs_iomap_ops);
+				&xfs_read_iomap_ops);
 	}
 	xfs_iunlock(XFS_I(inode), XFS_IOLOCK_SHARED);
 
@@ -1227,7 +1257,7 @@ xfs_inode_supports_dax(
 		return false;
 
 	/* Device has to support DAX too. */
-	return xfs_find_daxdev_for_inode(VFS_I(ip)) != NULL;
+	return xfs_inode_buftarg(ip)->bt_daxdev != NULL;
 }
 
 STATIC void
@@ -1274,9 +1304,6 @@ xfs_setup_inode(
 	/* make the inode look hashed for the writeback code */
 	inode_fake_hash(inode);
 
-	inode->i_uid    = xfs_uid_to_kuid(ip->i_d.di_uid);
-	inode->i_gid    = xfs_gid_to_kgid(ip->i_d.di_gid);
-
 	i_size_write(inode, ip->i_d.di_size);
 	xfs_diflags_to_iflags(inode, ip);
 
@@ -1290,9 +1317,7 @@ xfs_setup_inode(
 		lockdep_set_class(&inode->i_rwsem,
 				  &inode->i_sb->s_type->i_mutex_dir_key);
 		lockdep_set_class(&ip->i_lock.mr_lock, &xfs_dir_ilock_class);
-		ip->d_ops = ip->i_mount->m_dir_inode_ops;
 	} else {
-		ip->d_ops = ip->i_mount->m_nondir_inode_ops;
 		lockdep_set_class(&ip->i_lock.mr_lock, &xfs_nondir_ilock_class);
 	}
 
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index 884950adbd16..ff2da28fed90 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -84,10 +84,10 @@ xfs_bulkstat_one_int(
 	/* xfs_iget returns the following without needing
 	 * further change.
 	 */
-	buf->bs_projectid = xfs_get_projid(ip);
+	buf->bs_projectid = ip->i_d.di_projid;
 	buf->bs_ino = ino;
-	buf->bs_uid = dic->di_uid;
-	buf->bs_gid = dic->di_gid;
+	buf->bs_uid = i_uid_read(inode);
+	buf->bs_gid = i_gid_read(inode);
 	buf->bs_size = dic->di_size;
 
 	buf->bs_nlink = inode->i_nlink;
@@ -97,8 +97,8 @@ xfs_bulkstat_one_int(
 	buf->bs_mtime_nsec = inode->i_mtime.tv_nsec;
 	buf->bs_ctime = inode->i_ctime.tv_sec;
 	buf->bs_ctime_nsec = inode->i_ctime.tv_nsec;
-	buf->bs_btime = dic->di_crtime.t_sec;
-	buf->bs_btime_nsec = dic->di_crtime.t_nsec;
+	buf->bs_btime = dic->di_crtime.tv_sec;
+	buf->bs_btime_nsec = dic->di_crtime.tv_nsec;
 	buf->bs_gen = inode->i_generation;
 	buf->bs_mode = inode->i_mode;
 
@@ -110,7 +110,7 @@ xfs_bulkstat_one_int(
 	buf->bs_forkoff = XFS_IFORK_BOFF(ip);
 	buf->bs_version = XFS_BULKSTAT_VERSION_V5;
 
-	if (dic->di_version == 3) {
+	if (xfs_sb_version_has_v3inode(&mp->m_sb)) {
 		if (dic->di_flags2 & XFS_DIFLAG2_COWEXTSIZE)
 			buf->bs_cowextsize_blks = dic->di_cowextsize;
 	}
diff --git a/fs/xfs/xfs_iwalk.c b/fs/xfs/xfs_iwalk.c
index aa375cf53021..233dcc8784db 100644
--- a/fs/xfs/xfs_iwalk.c
+++ b/fs/xfs/xfs_iwalk.c
@@ -298,7 +298,8 @@ xfs_iwalk_ag_start(
 	error = xfs_inobt_get_rec(*curpp, irec, has_more);
 	if (error)
 		return error;
-	XFS_WANT_CORRUPTED_RETURN(mp, *has_more == 1);
+	if (XFS_IS_CORRUPT(mp, *has_more != 1))
+		return -EFSCORRUPTED;
 
 	/*
 	 * If the LE lookup yielded an inobt record before the cursor position,
diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h
index ca15105681ca..9f70d2f68e05 100644
--- a/fs/xfs/xfs_linux.h
+++ b/fs/xfs/xfs_linux.h
@@ -60,6 +60,7 @@ typedef __u32			xfs_nlink_t;
 #include <linux/list_sort.h>
 #include <linux/ratelimit.h>
 #include <linux/rhashtable.h>
+#include <linux/xattr.h>
 
 #include <asm/page.h>
 #include <asm/div64.h>
@@ -163,32 +164,6 @@ struct xstats {
 
 extern struct xstats xfsstats;
 
-/* Kernel uid/gid conversion. These are used to convert to/from the on disk
- * uid_t/gid_t types to the kuid_t/kgid_t types that the kernel uses internally.
- * The conversion here is type only, the value will remain the same since we
- * are converting to the init_user_ns. The uid is later mapped to a particular
- * user namespace value when crossing the kernel/user boundary.
- */
-static inline uint32_t xfs_kuid_to_uid(kuid_t uid)
-{
-	return from_kuid(&init_user_ns, uid);
-}
-
-static inline kuid_t xfs_uid_to_kuid(uint32_t uid)
-{
-	return make_kuid(&init_user_ns, uid);
-}
-
-static inline uint32_t xfs_kgid_to_gid(kgid_t gid)
-{
-	return from_kgid(&init_user_ns, gid);
-}
-
-static inline kgid_t xfs_gid_to_kgid(uint32_t gid)
-{
-	return make_kgid(&init_user_ns, gid);
-}
-
 static inline dev_t xfs_to_linux_dev_t(xfs_dev_t dev)
 {
 	return MKDEV(sysv_major(dev) & 0x1ff, sysv_minor(dev));
@@ -223,26 +198,32 @@ int xfs_rw_bdev(struct block_device *bdev, sector_t sector, unsigned int count,
 		char *data, unsigned int op);
 
 #define ASSERT_ALWAYS(expr)	\
-	(likely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__))
+	(likely(expr) ? (void)0 : assfail(NULL, #expr, __FILE__, __LINE__))
 
 #ifdef DEBUG
 #define ASSERT(expr)	\
-	(likely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__))
+	(likely(expr) ? (void)0 : assfail(NULL, #expr, __FILE__, __LINE__))
 
 #else	/* !DEBUG */
 
 #ifdef XFS_WARN
 
 #define ASSERT(expr)	\
-	(likely(expr) ? (void)0 : asswarn(#expr, __FILE__, __LINE__))
+	(likely(expr) ? (void)0 : asswarn(NULL, #expr, __FILE__, __LINE__))
 
 #else	/* !DEBUG && !XFS_WARN */
 
-#define ASSERT(expr)	((void)0)
+#define ASSERT(expr)		((void)0)
 
 #endif /* XFS_WARN */
 #endif /* DEBUG */
 
+#define XFS_IS_CORRUPT(mp, expr)	\
+	(unlikely(expr) ? xfs_corruption_error(#expr, XFS_ERRLEVEL_LOW, (mp), \
+					       NULL, 0, __FILE__, __LINE__, \
+					       __this_address), \
+			  true : false)
+
 #define STATIC static noinline
 
 #ifdef CONFIG_XFS_RT
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 641d07f30a27..00fda2e8e738 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -24,13 +24,6 @@
 kmem_zone_t	*xfs_log_ticket_zone;
 
 /* Local miscellaneous function prototypes */
-STATIC int
-xlog_commit_record(
-	struct xlog		*log,
-	struct xlog_ticket	*ticket,
-	struct xlog_in_core	**iclog,
-	xfs_lsn_t		*commitlsnp);
-
 STATIC struct xlog *
 xlog_alloc_log(
 	struct xfs_mount	*mp,
@@ -47,8 +40,7 @@ xlog_dealloc_log(
 
 /* local state machine functions */
 STATIC void xlog_state_done_syncing(
-	struct xlog_in_core	*iclog,
-	bool			aborted);
+	struct xlog_in_core	*iclog);
 STATIC int
 xlog_state_get_iclog_space(
 	struct xlog		*log,
@@ -57,33 +49,19 @@ xlog_state_get_iclog_space(
 	struct xlog_ticket	*ticket,
 	int			*continued_write,
 	int			*logoffsetp);
-STATIC int
-xlog_state_release_iclog(
-	struct xlog		*log,
-	struct xlog_in_core	*iclog);
 STATIC void
 xlog_state_switch_iclogs(
 	struct xlog		*log,
 	struct xlog_in_core	*iclog,
 	int			eventual_size);
 STATIC void
-xlog_state_want_sync(
-	struct xlog		*log,
-	struct xlog_in_core	*iclog);
-
-STATIC void
 xlog_grant_push_ail(
 	struct xlog		*log,
 	int			need_bytes);
 STATIC void
-xlog_regrant_reserve_log_space(
-	struct xlog		*log,
-	struct xlog_ticket	*ticket);
-STATIC void
-xlog_ungrant_log_space(
+xlog_sync(
 	struct xlog		*log,
-	struct xlog_ticket	*ticket);
-
+	struct xlog_in_core	*iclog);
 #if defined(DEBUG)
 STATIC void
 xlog_verify_dest_ptr(
@@ -485,84 +463,67 @@ out_error:
 	return error;
 }
 
+static bool
+__xlog_state_release_iclog(
+	struct xlog		*log,
+	struct xlog_in_core	*iclog)
+{
+	lockdep_assert_held(&log->l_icloglock);
 
-/*
- * NOTES:
- *
- *	1. currblock field gets updated at startup and after in-core logs
- *		marked as with WANT_SYNC.
- */
+	if (iclog->ic_state == XLOG_STATE_WANT_SYNC) {
+		/* update tail before writing to iclog */
+		xfs_lsn_t tail_lsn = xlog_assign_tail_lsn(log->l_mp);
 
-/*
- * This routine is called when a user of a log manager ticket is done with
- * the reservation.  If the ticket was ever used, then a commit record for
- * the associated transaction is written out as a log operation header with
- * no data.  The flag XLOG_TIC_INITED is set when the first write occurs with
- * a given ticket.  If the ticket was one with a permanent reservation, then
- * a few operations are done differently.  Permanent reservation tickets by
- * default don't release the reservation.  They just commit the current
- * transaction with the belief that the reservation is still needed.  A flag
- * must be passed in before permanent reservations are actually released.
- * When these type of tickets are not released, they need to be set into
- * the inited state again.  By doing this, a start record will be written
- * out when the next write occurs.
- */
-xfs_lsn_t
-xfs_log_done(
-	struct xfs_mount	*mp,
-	struct xlog_ticket	*ticket,
-	struct xlog_in_core	**iclog,
-	bool			regrant)
-{
-	struct xlog		*log = mp->m_log;
-	xfs_lsn_t		lsn = 0;
-
-	if (XLOG_FORCED_SHUTDOWN(log) ||
-	    /*
-	     * If nothing was ever written, don't write out commit record.
-	     * If we get an error, just continue and give back the log ticket.
-	     */
-	    (((ticket->t_flags & XLOG_TIC_INITED) == 0) &&
-	     (xlog_commit_record(log, ticket, iclog, &lsn)))) {
-		lsn = (xfs_lsn_t) -1;
-		regrant = false;
+		iclog->ic_state = XLOG_STATE_SYNCING;
+		iclog->ic_header.h_tail_lsn = cpu_to_be64(tail_lsn);
+		xlog_verify_tail_lsn(log, iclog, tail_lsn);
+		/* cycle incremented when incrementing curr_block */
+		return true;
 	}
 
+	ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE);
+	return false;
+}
 
-	if (!regrant) {
-		trace_xfs_log_done_nonperm(log, ticket);
+/*
+ * Flush iclog to disk if this is the last reference to the given iclog and the
+ * it is in the WANT_SYNC state.
+ */
+static int
+xlog_state_release_iclog(
+	struct xlog		*log,
+	struct xlog_in_core	*iclog)
+{
+	lockdep_assert_held(&log->l_icloglock);
 
-		/*
-		 * Release ticket if not permanent reservation or a specific
-		 * request has been made to release a permanent reservation.
-		 */
-		xlog_ungrant_log_space(log, ticket);
-	} else {
-		trace_xfs_log_done_perm(log, ticket);
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
+		return -EIO;
 
-		xlog_regrant_reserve_log_space(log, ticket);
-		/* If this ticket was a permanent reservation and we aren't
-		 * trying to release it, reset the inited flags; so next time
-		 * we write, a start record will be written out.
-		 */
-		ticket->t_flags |= XLOG_TIC_INITED;
+	if (atomic_dec_and_test(&iclog->ic_refcnt) &&
+	    __xlog_state_release_iclog(log, iclog)) {
+		spin_unlock(&log->l_icloglock);
+		xlog_sync(log, iclog);
+		spin_lock(&log->l_icloglock);
 	}
 
-	xfs_log_ticket_put(ticket);
-	return lsn;
+	return 0;
 }
 
-int
+void
 xfs_log_release_iclog(
-	struct xfs_mount	*mp,
 	struct xlog_in_core	*iclog)
 {
-	if (xlog_state_release_iclog(mp->m_log, iclog)) {
-		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
-		return -EIO;
+	struct xlog		*log = iclog->ic_log;
+	bool			sync = false;
+
+	if (atomic_dec_and_lock(&iclog->ic_refcnt, &log->l_icloglock)) {
+		if (iclog->ic_state != XLOG_STATE_IOERROR)
+			sync = __xlog_state_release_iclog(log, iclog);
+		spin_unlock(&log->l_icloglock);
 	}
 
-	return 0;
+	if (sync)
+		xlog_sync(log, iclog);
 }
 
 /*
@@ -801,32 +762,69 @@ xfs_log_mount_cancel(
 }
 
 /*
- * Final log writes as part of unmount.
- *
- * Mark the filesystem clean as unmount happens.  Note that during relocation
- * this routine needs to be executed as part of source-bag while the
- * deallocation must not be done until source-end.
+ * Wait for the iclog to be written disk, or return an error if the log has been
+ * shut down.
  */
+static int
+xlog_wait_on_iclog(
+	struct xlog_in_core	*iclog)
+		__releases(iclog->ic_log->l_icloglock)
+{
+	struct xlog		*log = iclog->ic_log;
 
-/* Actually write the unmount record to disk. */
-static void
-xfs_log_write_unmount_record(
-	struct xfs_mount	*mp)
+	if (!XLOG_FORCED_SHUTDOWN(log) &&
+	    iclog->ic_state != XLOG_STATE_ACTIVE &&
+	    iclog->ic_state != XLOG_STATE_DIRTY) {
+		XFS_STATS_INC(log->l_mp, xs_log_force_sleep);
+		xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
+	} else {
+		spin_unlock(&log->l_icloglock);
+	}
+
+	if (XLOG_FORCED_SHUTDOWN(log))
+		return -EIO;
+	return 0;
+}
+
+/*
+ * Write out an unmount record using the ticket provided. We have to account for
+ * the data space used in the unmount ticket as this write is not done from a
+ * transaction context that has already done the accounting for us.
+ */
+static int
+xlog_write_unmount_record(
+	struct xlog		*log,
+	struct xlog_ticket	*ticket,
+	xfs_lsn_t		*lsn,
+	uint			flags)
 {
-	/* the data section must be 32 bit size aligned */
-	struct xfs_unmount_log_format magic = {
+	struct xfs_unmount_log_format ulf = {
 		.magic = XLOG_UNMOUNT_TYPE,
 	};
 	struct xfs_log_iovec reg = {
-		.i_addr = &magic,
-		.i_len = sizeof(magic),
+		.i_addr = &ulf,
+		.i_len = sizeof(ulf),
 		.i_type = XLOG_REG_TYPE_UNMOUNT,
 	};
 	struct xfs_log_vec vec = {
 		.lv_niovecs = 1,
 		.lv_iovecp = &reg,
 	};
-	struct xlog		*log = mp->m_log;
+
+	/* account for space used by record data */
+	ticket->t_curr_res -= sizeof(ulf);
+	return xlog_write(log, &vec, ticket, lsn, NULL, flags, false);
+}
+
+/*
+ * Mark the filesystem clean by writing an unmount record to the head of the
+ * log.
+ */
+static void
+xlog_unmount_write(
+	struct xlog		*log)
+{
+	struct xfs_mount	*mp = log->l_mp;
 	struct xlog_in_core	*iclog;
 	struct xlog_ticket	*tic = NULL;
 	xfs_lsn_t		lsn;
@@ -837,23 +835,7 @@ xfs_log_write_unmount_record(
 	if (error)
 		goto out_err;
 
-	/*
-	 * If we think the summary counters are bad, clear the unmount header
-	 * flag in the unmount record so that the summary counters will be
-	 * recalculated during log recovery at next mount.  Refer to
-	 * xlog_check_unmount_rec for more details.
-	 */
-	if (XFS_TEST_ERROR(xfs_fs_has_sickness(mp, XFS_SICK_FS_COUNTERS), mp,
-			XFS_ERRTAG_FORCE_SUMMARY_RECALC)) {
-		xfs_alert(mp, "%s: will fix summary counters at next mount",
-				__func__);
-		flags &= ~XLOG_UNMOUNT_TRANS;
-	}
-
-	/* remove inited flag, and account for space used */
-	tic->t_flags = 0;
-	tic->t_curr_res -= sizeof(magic);
-	error = xlog_write(log, &vec, tic, &lsn, NULL, flags);
+	error = xlog_write_unmount_record(log, tic, &lsn, flags);
 	/*
 	 * At this point, we're umounting anyway, so there's no point in
 	 * transitioning log state to IOERROR. Just continue...
@@ -865,31 +847,32 @@ out_err:
 	spin_lock(&log->l_icloglock);
 	iclog = log->l_iclog;
 	atomic_inc(&iclog->ic_refcnt);
-	xlog_state_want_sync(log, iclog);
-	spin_unlock(&log->l_icloglock);
+	if (iclog->ic_state == XLOG_STATE_ACTIVE)
+		xlog_state_switch_iclogs(log, iclog, 0);
+	else
+		ASSERT(iclog->ic_state == XLOG_STATE_WANT_SYNC ||
+		       iclog->ic_state == XLOG_STATE_IOERROR);
 	error = xlog_state_release_iclog(log, iclog);
-
-	spin_lock(&log->l_icloglock);
-	switch (iclog->ic_state) {
-	default:
-		if (!XLOG_FORCED_SHUTDOWN(log)) {
-			xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
-			break;
-		}
-		/* fall through */
-	case XLOG_STATE_ACTIVE:
-	case XLOG_STATE_DIRTY:
-		spin_unlock(&log->l_icloglock);
-		break;
-	}
+	xlog_wait_on_iclog(iclog);
 
 	if (tic) {
 		trace_xfs_log_umount_write(log, tic);
-		xlog_ungrant_log_space(log, tic);
-		xfs_log_ticket_put(tic);
+		xfs_log_ticket_ungrant(log, tic);
 	}
 }
 
+static void
+xfs_log_unmount_verify_iclog(
+	struct xlog		*log)
+{
+	struct xlog_in_core	*iclog = log->l_iclog;
+
+	do {
+		ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE);
+		ASSERT(iclog->ic_offset == 0);
+	} while ((iclog = iclog->ic_next) != log->l_iclog);
+}
+
 /*
  * Unmount record used to have a string "Unmount filesystem--" in the
  * data section where the "Un" was really a magic number (XLOG_UNMOUNT_TYPE).
@@ -897,16 +880,11 @@ out_err:
  * currently architecture converted and "Unmount" is a bit foo.
  * As far as I know, there weren't any dependencies on the old behaviour.
  */
-
-static int
-xfs_log_unmount_write(xfs_mount_t *mp)
+static void
+xfs_log_unmount_write(
+	struct xfs_mount	*mp)
 {
-	struct xlog	 *log = mp->m_log;
-	xlog_in_core_t	 *iclog;
-#ifdef DEBUG
-	xlog_in_core_t	 *first_iclog;
-#endif
-	int		 error;
+	struct xlog		*log = mp->m_log;
 
 	/*
 	 * Don't write out unmount record on norecovery mounts or ro devices.
@@ -915,61 +893,30 @@ xfs_log_unmount_write(xfs_mount_t *mp)
 	if (mp->m_flags & XFS_MOUNT_NORECOVERY ||
 	    xfs_readonly_buftarg(log->l_targ)) {
 		ASSERT(mp->m_flags & XFS_MOUNT_RDONLY);
-		return 0;
+		return;
 	}
 
-	error = xfs_log_force(mp, XFS_LOG_SYNC);
-	ASSERT(error || !(XLOG_FORCED_SHUTDOWN(log)));
-
-#ifdef DEBUG
-	first_iclog = iclog = log->l_iclog;
-	do {
-		if (!(iclog->ic_state & XLOG_STATE_IOERROR)) {
-			ASSERT(iclog->ic_state & XLOG_STATE_ACTIVE);
-			ASSERT(iclog->ic_offset == 0);
-		}
-		iclog = iclog->ic_next;
-	} while (iclog != first_iclog);
-#endif
-	if (! (XLOG_FORCED_SHUTDOWN(log))) {
-		xfs_log_write_unmount_record(mp);
-	} else {
-		/*
-		 * We're already in forced_shutdown mode, couldn't
-		 * even attempt to write out the unmount transaction.
-		 *
-		 * Go through the motions of sync'ing and releasing
-		 * the iclog, even though no I/O will actually happen,
-		 * we need to wait for other log I/Os that may already
-		 * be in progress.  Do this as a separate section of
-		 * code so we'll know if we ever get stuck here that
-		 * we're in this odd situation of trying to unmount
-		 * a file system that went into forced_shutdown as
-		 * the result of an unmount..
-		 */
-		spin_lock(&log->l_icloglock);
-		iclog = log->l_iclog;
-		atomic_inc(&iclog->ic_refcnt);
-
-		xlog_state_want_sync(log, iclog);
-		spin_unlock(&log->l_icloglock);
-		error =  xlog_state_release_iclog(log, iclog);
-
-		spin_lock(&log->l_icloglock);
+	xfs_log_force(mp, XFS_LOG_SYNC);
 
-		if ( ! (   iclog->ic_state == XLOG_STATE_ACTIVE
-			|| iclog->ic_state == XLOG_STATE_DIRTY
-			|| iclog->ic_state == XLOG_STATE_IOERROR) ) {
+	if (XLOG_FORCED_SHUTDOWN(log))
+		return;
 
-				xlog_wait(&iclog->ic_force_wait,
-							&log->l_icloglock);
-		} else {
-			spin_unlock(&log->l_icloglock);
-		}
+	/*
+	 * If we think the summary counters are bad, avoid writing the unmount
+	 * record to force log recovery at next mount, after which the summary
+	 * counters will be recalculated.  Refer to xlog_check_unmount_rec for
+	 * more details.
+	 */
+	if (XFS_TEST_ERROR(xfs_fs_has_sickness(mp, XFS_SICK_FS_COUNTERS), mp,
+			XFS_ERRTAG_FORCE_SUMMARY_RECALC)) {
+		xfs_alert(mp, "%s: will fix summary counters at next mount",
+				__func__);
+		return;
 	}
 
-	return error;
-}	/* xfs_log_unmount_write */
+	xfs_log_unmount_verify_iclog(log);
+	xlog_unmount_write(log);
+}
 
 /*
  * Empty the log for unmount/freeze.
@@ -1232,7 +1179,6 @@ xlog_ioend_work(
 	struct xlog_in_core     *iclog =
 		container_of(work, struct xlog_in_core, ic_end_io_work);
 	struct xlog		*log = iclog->ic_log;
-	bool			aborted = false;
 	int			error;
 
 	error = blk_status_to_errno(iclog->ic_bio.bi_status);
@@ -1248,17 +1194,9 @@ xlog_ioend_work(
 	if (XFS_TEST_ERROR(error, log->l_mp, XFS_ERRTAG_IODONE_IOERR)) {
 		xfs_alert(log->l_mp, "log I/O error %d", error);
 		xfs_force_shutdown(log->l_mp, SHUTDOWN_LOG_IO_ERROR);
-		/*
-		 * This flag will be propagated to the trans-committed
-		 * callback routines to let them know that the log-commit
-		 * didn't succeed.
-		 */
-		aborted = true;
-	} else if (iclog->ic_state & XLOG_STATE_IOERROR) {
-		aborted = true;
 	}
 
-	xlog_state_done_syncing(iclog, aborted);
+	xlog_state_done_syncing(iclog);
 	bio_uninit(&iclog->ic_bio);
 
 	/*
@@ -1479,7 +1417,7 @@ xlog_alloc_log(
 
 	log->l_ioend_workqueue = alloc_workqueue("xfs-log/%s",
 			WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_HIGHPRI, 0,
-			mp->m_fsname);
+			mp->m_super->s_id);
 	if (!log->l_ioend_workqueue)
 		goto out_free_iclog;
 
@@ -1495,6 +1433,8 @@ out_free_iclog:
 		prev_iclog = iclog->ic_next;
 		kmem_free(iclog->ic_data);
 		kmem_free(iclog);
+		if (prev_iclog == log->l_iclog)
+			break;
 	}
 out_free_log:
 	kmem_free(log);
@@ -1502,20 +1442,17 @@ out:
 	return ERR_PTR(error);
 }	/* xlog_alloc_log */
 
-
 /*
  * Write out the commit record of a transaction associated with the given
- * ticket.  Return the lsn of the commit record.
+ * ticket to close off a running log write. Return the lsn of the commit record.
  */
-STATIC int
+int
 xlog_commit_record(
 	struct xlog		*log,
 	struct xlog_ticket	*ticket,
 	struct xlog_in_core	**iclog,
-	xfs_lsn_t		*commitlsnp)
+	xfs_lsn_t		*lsn)
 {
-	struct xfs_mount *mp = log->l_mp;
-	int	error;
 	struct xfs_log_iovec reg = {
 		.i_addr = NULL,
 		.i_len = 0,
@@ -1525,12 +1462,15 @@ xlog_commit_record(
 		.lv_niovecs = 1,
 		.lv_iovecp = &reg,
 	};
+	int	error;
+
+	if (XLOG_FORCED_SHUTDOWN(log))
+		return -EIO;
 
-	ASSERT_ALWAYS(iclog);
-	error = xlog_write(log, &vec, ticket, commitlsnp, iclog,
-					XLOG_COMMIT_TRANS);
+	error = xlog_write(log, &vec, ticket, lsn, iclog, XLOG_COMMIT_TRANS,
+			   false);
 	if (error)
-		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
+		xfs_force_shutdown(log->l_mp, SHUTDOWN_LOG_IO_ERROR);
 	return error;
 }
 
@@ -1690,7 +1630,7 @@ xlog_bio_end_io(
 		   &iclog->ic_end_io_work);
 }
 
-static void
+static int
 xlog_map_iclog_data(
 	struct bio		*bio,
 	void			*data,
@@ -1701,11 +1641,14 @@ xlog_map_iclog_data(
 		unsigned int	off = offset_in_page(data);
 		size_t		len = min_t(size_t, count, PAGE_SIZE - off);
 
-		WARN_ON_ONCE(bio_add_page(bio, page, len, off) != len);
+		if (bio_add_page(bio, page, len, off) != len)
+			return -EIO;
 
 		data += len;
 		count -= len;
 	} while (count);
+
+	return 0;
 }
 
 STATIC void
@@ -1727,7 +1670,7 @@ xlog_write_iclog(
 	 * across the log IO to archieve that.
 	 */
 	down(&iclog->ic_sema);
-	if (unlikely(iclog->ic_state & XLOG_STATE_IOERROR)) {
+	if (unlikely(iclog->ic_state == XLOG_STATE_IOERROR)) {
 		/*
 		 * It would seem logical to return EIO here, but we rely on
 		 * the log state machine to propagate I/O errors instead of
@@ -1735,25 +1678,34 @@ xlog_write_iclog(
 		 * the buffer manually, the code needs to be kept in sync
 		 * with the I/O completion path.
 		 */
-		xlog_state_done_syncing(iclog, XFS_LI_ABORTED);
+		xlog_state_done_syncing(iclog);
 		up(&iclog->ic_sema);
 		return;
 	}
 
-	iclog->ic_io_size = count;
-
 	bio_init(&iclog->ic_bio, iclog->ic_bvec, howmany(count, PAGE_SIZE));
 	bio_set_dev(&iclog->ic_bio, log->l_targ->bt_bdev);
 	iclog->ic_bio.bi_iter.bi_sector = log->l_logBBstart + bno;
 	iclog->ic_bio.bi_end_io = xlog_bio_end_io;
 	iclog->ic_bio.bi_private = iclog;
-	iclog->ic_bio.bi_opf = REQ_OP_WRITE | REQ_META | REQ_SYNC | REQ_FUA;
+
+	/*
+	 * We use REQ_SYNC | REQ_IDLE here to tell the block layer the are more
+	 * IOs coming immediately after this one. This prevents the block layer
+	 * writeback throttle from throttling log writes behind background
+	 * metadata writeback and causing priority inversions.
+	 */
+	iclog->ic_bio.bi_opf = REQ_OP_WRITE | REQ_META | REQ_SYNC |
+				REQ_IDLE | REQ_FUA;
 	if (need_flush)
 		iclog->ic_bio.bi_opf |= REQ_PREFLUSH;
 
-	xlog_map_iclog_data(&iclog->ic_bio, iclog->ic_data, iclog->ic_io_size);
+	if (xlog_map_iclog_data(&iclog->ic_bio, iclog->ic_data, count)) {
+		xfs_force_shutdown(log->l_mp, SHUTDOWN_LOG_IO_ERROR);
+		return;
+	}
 	if (is_vmalloc_addr(iclog->ic_data))
-		flush_kernel_vmap_range(iclog->ic_data, iclog->ic_io_size);
+		flush_kernel_vmap_range(iclog->ic_data, count);
 
 	/*
 	 * If this log buffer would straddle the end of the log we will have
@@ -1964,12 +1916,11 @@ xlog_dealloc_log(
 	log->l_mp->m_log = NULL;
 	destroy_workqueue(log->l_ioend_workqueue);
 	kmem_free(log);
-}	/* xlog_dealloc_log */
+}
 
 /*
  * Update counters atomically now that memcpy is done.
  */
-/* ARGSUSED */
 static inline void
 xlog_state_finish_copy(
 	struct xlog		*log,
@@ -1977,16 +1928,11 @@ xlog_state_finish_copy(
 	int			record_cnt,
 	int			copy_bytes)
 {
-	spin_lock(&log->l_icloglock);
+	lockdep_assert_held(&log->l_icloglock);
 
 	be32_add_cpu(&iclog->ic_header.h_num_logops, record_cnt);
 	iclog->ic_offset += copy_bytes;
-
-	spin_unlock(&log->l_icloglock);
-}	/* xlog_state_finish_copy */
-
-
-
+}
 
 /*
  * print out info relating to regions written which consume
@@ -2107,23 +2053,21 @@ xlog_print_trans(
 }
 
 /*
- * Calculate the potential space needed by the log vector.  Each region gets
- * its own xlog_op_header_t and may need to be double word aligned.
+ * Calculate the potential space needed by the log vector.  We may need a start
+ * record, and each region gets its own struct xlog_op_header and may need to be
+ * double word aligned.
  */
 static int
 xlog_write_calc_vec_length(
 	struct xlog_ticket	*ticket,
-	struct xfs_log_vec	*log_vector)
+	struct xfs_log_vec	*log_vector,
+	bool			need_start_rec)
 {
 	struct xfs_log_vec	*lv;
-	int			headers = 0;
+	int			headers = need_start_rec ? 1 : 0;
 	int			len = 0;
 	int			i;
 
-	/* acct for start rec of xact */
-	if (ticket->t_flags & XLOG_TIC_INITED)
-		headers++;
-
 	for (lv = log_vector; lv; lv = lv->lv_next) {
 		/* we don't write ordered log vectors */
 		if (lv->lv_buf_len == XFS_LOG_VEC_ORDERED)
@@ -2145,27 +2089,16 @@ xlog_write_calc_vec_length(
 	return len;
 }
 
-/*
- * If first write for transaction, insert start record  We can't be trying to
- * commit if we are inited.  We can't have any "partial_copy" if we are inited.
- */
-static int
+static void
 xlog_write_start_rec(
 	struct xlog_op_header	*ophdr,
 	struct xlog_ticket	*ticket)
 {
-	if (!(ticket->t_flags & XLOG_TIC_INITED))
-		return 0;
-
 	ophdr->oh_tid	= cpu_to_be32(ticket->t_tid);
 	ophdr->oh_clientid = ticket->t_clientid;
 	ophdr->oh_len = 0;
 	ophdr->oh_flags = XLOG_START_TRANS;
 	ophdr->oh_res2 = 0;
-
-	ticket->t_flags &= ~XLOG_TIC_INITED;
-
-	return sizeof(struct xlog_op_header);
 }
 
 static xlog_op_header_t *
@@ -2263,15 +2196,18 @@ xlog_write_copy_finish(
 	int			log_offset,
 	struct xlog_in_core	**commit_iclog)
 {
+	int			error;
+
 	if (*partial_copy) {
 		/*
 		 * This iclog has already been marked WANT_SYNC by
 		 * xlog_state_get_iclog_space.
 		 */
+		spin_lock(&log->l_icloglock);
 		xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
 		*record_cnt = 0;
 		*data_cnt = 0;
-		return xlog_state_release_iclog(log, iclog);
+		goto release_iclog;
 	}
 
 	*partial_copy = 0;
@@ -2279,21 +2215,29 @@ xlog_write_copy_finish(
 
 	if (iclog->ic_size - log_offset <= sizeof(xlog_op_header_t)) {
 		/* no more space in this iclog - push it. */
+		spin_lock(&log->l_icloglock);
 		xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
 		*record_cnt = 0;
 		*data_cnt = 0;
 
-		spin_lock(&log->l_icloglock);
-		xlog_state_want_sync(log, iclog);
-		spin_unlock(&log->l_icloglock);
-
+		if (iclog->ic_state == XLOG_STATE_ACTIVE)
+			xlog_state_switch_iclogs(log, iclog, 0);
+		else
+			ASSERT(iclog->ic_state == XLOG_STATE_WANT_SYNC ||
+			       iclog->ic_state == XLOG_STATE_IOERROR);
 		if (!commit_iclog)
-			return xlog_state_release_iclog(log, iclog);
+			goto release_iclog;
+		spin_unlock(&log->l_icloglock);
 		ASSERT(flags & XLOG_COMMIT_TRANS);
 		*commit_iclog = iclog;
 	}
 
 	return 0;
+
+release_iclog:
+	error = xlog_state_release_iclog(log, iclog);
+	spin_unlock(&log->l_icloglock);
+	return error;
 }
 
 /*
@@ -2343,39 +2287,28 @@ xlog_write(
 	struct xlog_ticket	*ticket,
 	xfs_lsn_t		*start_lsn,
 	struct xlog_in_core	**commit_iclog,
-	uint			flags)
+	uint			flags,
+	bool			need_start_rec)
 {
 	struct xlog_in_core	*iclog = NULL;
-	struct xfs_log_iovec	*vecp;
-	struct xfs_log_vec	*lv;
+	struct xfs_log_vec	*lv = log_vector;
+	struct xfs_log_iovec	*vecp = lv->lv_iovecp;
+	int			index = 0;
 	int			len;
-	int			index;
 	int			partial_copy = 0;
 	int			partial_copy_len = 0;
 	int			contwr = 0;
 	int			record_cnt = 0;
 	int			data_cnt = 0;
-	int			error;
-
-	*start_lsn = 0;
-
-	len = xlog_write_calc_vec_length(ticket, log_vector);
-
-	/*
-	 * Region headers and bytes are already accounted for.
-	 * We only need to take into account start records and
-	 * split regions in this function.
-	 */
-	if (ticket->t_flags & XLOG_TIC_INITED)
-		ticket->t_curr_res -= sizeof(xlog_op_header_t);
+	int			error = 0;
 
 	/*
-	 * Commit record headers need to be accounted for. These
-	 * come in as separate writes so are easy to detect.
+	 * If this is a commit or unmount transaction, we don't need a start
+	 * record to be written.  We do, however, have to account for the
+	 * commit or unmount header that gets written. Hence we always have
+	 * to account for an extra xlog_op_header here.
 	 */
-	if (flags & (XLOG_COMMIT_TRANS | XLOG_UNMOUNT_TRANS))
-		ticket->t_curr_res -= sizeof(xlog_op_header_t);
-
+	ticket->t_curr_res -= sizeof(struct xlog_op_header);
 	if (ticket->t_curr_res < 0) {
 		xfs_alert_tag(log->l_mp, XFS_PTAG_LOGRES,
 		     "ctx ticket reservation ran out. Need to up reservation");
@@ -2383,9 +2316,8 @@ xlog_write(
 		xfs_force_shutdown(log->l_mp, SHUTDOWN_LOG_IO_ERROR);
 	}
 
-	index = 0;
-	lv = log_vector;
-	vecp = lv->lv_iovecp;
+	len = xlog_write_calc_vec_length(ticket, log_vector, need_start_rec);
+	*start_lsn = 0;
 	while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {
 		void		*ptr;
 		int		log_offset;
@@ -2409,7 +2341,6 @@ xlog_write(
 		while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {
 			struct xfs_log_iovec	*reg;
 			struct xlog_op_header	*ophdr;
-			int			start_rec_copy;
 			int			copy_len;
 			int			copy_off;
 			bool			ordered = false;
@@ -2425,11 +2356,15 @@ xlog_write(
 			ASSERT(reg->i_len % sizeof(int32_t) == 0);
 			ASSERT((unsigned long)ptr % sizeof(int32_t) == 0);
 
-			start_rec_copy = xlog_write_start_rec(ptr, ticket);
-			if (start_rec_copy) {
-				record_cnt++;
+			/*
+			 * Before we start formatting log vectors, we need to
+			 * write a start record. Only do this for the first
+			 * iclog we write to.
+			 */
+			if (need_start_rec) {
+				xlog_write_start_rec(ptr, ticket);
 				xlog_write_adv_cnt(&ptr, &len, &log_offset,
-						   start_rec_copy);
+						sizeof(struct xlog_op_header));
 			}
 
 			ophdr = xlog_write_setup_ophdr(log, ptr, ticket, flags);
@@ -2461,8 +2396,13 @@ xlog_write(
 				xlog_write_adv_cnt(&ptr, &len, &log_offset,
 						   copy_len);
 			}
-			copy_len += start_rec_copy + sizeof(xlog_op_header_t);
+			copy_len += sizeof(struct xlog_op_header);
 			record_cnt++;
+			if (need_start_rec) {
+				copy_len += sizeof(struct xlog_op_header);
+				record_cnt++;
+				need_start_rec = false;
+			}
 			data_cnt += contwr ? copy_len : 0;
 
 			error = xlog_write_copy_finish(log, iclog, flags,
@@ -2506,128 +2446,119 @@ next_lv:
 
 	ASSERT(len == 0);
 
+	spin_lock(&log->l_icloglock);
 	xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
-	if (!commit_iclog)
-		return xlog_state_release_iclog(log, iclog);
+	if (commit_iclog) {
+		ASSERT(flags & XLOG_COMMIT_TRANS);
+		*commit_iclog = iclog;
+	} else {
+		error = xlog_state_release_iclog(log, iclog);
+	}
+	spin_unlock(&log->l_icloglock);
 
-	ASSERT(flags & XLOG_COMMIT_TRANS);
-	*commit_iclog = iclog;
-	return 0;
+	return error;
 }
 
+static void
+xlog_state_activate_iclog(
+	struct xlog_in_core	*iclog,
+	int			*iclogs_changed)
+{
+	ASSERT(list_empty_careful(&iclog->ic_callbacks));
 
-/*****************************************************************************
- *
- *		State Machine functions
- *
- *****************************************************************************
- */
+	/*
+	 * If the number of ops in this iclog indicate it just contains the
+	 * dummy transaction, we can change state into IDLE (the second time
+	 * around). Otherwise we should change the state into NEED a dummy.
+	 * We don't need to cover the dummy.
+	 */
+	if (*iclogs_changed == 0 &&
+	    iclog->ic_header.h_num_logops == cpu_to_be32(XLOG_COVER_OPS)) {
+		*iclogs_changed = 1;
+	} else {
+		/*
+		 * We have two dirty iclogs so start over.  This could also be
+		 * num of ops indicating this is not the dummy going out.
+		 */
+		*iclogs_changed = 2;
+	}
+
+	iclog->ic_state	= XLOG_STATE_ACTIVE;
+	iclog->ic_offset = 0;
+	iclog->ic_header.h_num_logops = 0;
+	memset(iclog->ic_header.h_cycle_data, 0,
+		sizeof(iclog->ic_header.h_cycle_data));
+	iclog->ic_header.h_lsn = 0;
+}
 
 /*
- * An iclog has just finished IO completion processing, so we need to update
- * the iclog state and propagate that up into the overall log state. Hence we
- * prepare the iclog for cleaning, and then clean all the pending dirty iclogs
- * starting from the head, and then wake up any threads that are waiting for the
- * iclog to be marked clean.
- *
- * The ordering of marking iclogs ACTIVE must be maintained, so an iclog
- * doesn't become ACTIVE beyond one that is SYNCING.  This is also required to
- * maintain the notion that we use a ordered wait queue to hold off would be
- * writers to the log when every iclog is trying to sync to disk.
- *
- * Caller must hold the icloglock before calling us.
- *
- * State Change: !IOERROR -> DIRTY -> ACTIVE
+ * Loop through all iclogs and mark all iclogs currently marked DIRTY as
+ * ACTIVE after iclog I/O has completed.
  */
-STATIC void
-xlog_state_clean_iclog(
+static void
+xlog_state_activate_iclogs(
 	struct xlog		*log,
-	struct xlog_in_core	*dirty_iclog)
+	int			*iclogs_changed)
 {
-	struct xlog_in_core	*iclog;
-	int			changed = 0;
-
-	/* Prepare the completed iclog. */
-	if (!(dirty_iclog->ic_state & XLOG_STATE_IOERROR))
-		dirty_iclog->ic_state = XLOG_STATE_DIRTY;
+	struct xlog_in_core	*iclog = log->l_iclog;
 
-	/* Walk all the iclogs to update the ordered active state. */
-	iclog = log->l_iclog;
 	do {
-		if (iclog->ic_state == XLOG_STATE_DIRTY) {
-			iclog->ic_state	= XLOG_STATE_ACTIVE;
-			iclog->ic_offset       = 0;
-			ASSERT(list_empty_careful(&iclog->ic_callbacks));
-			/*
-			 * If the number of ops in this iclog indicate it just
-			 * contains the dummy transaction, we can
-			 * change state into IDLE (the second time around).
-			 * Otherwise we should change the state into
-			 * NEED a dummy.
-			 * We don't need to cover the dummy.
-			 */
-			if (!changed &&
-			   (be32_to_cpu(iclog->ic_header.h_num_logops) ==
-			   		XLOG_COVER_OPS)) {
-				changed = 1;
-			} else {
-				/*
-				 * We have two dirty iclogs so start over
-				 * This could also be num of ops indicates
-				 * this is not the dummy going out.
-				 */
-				changed = 2;
-			}
-			iclog->ic_header.h_num_logops = 0;
-			memset(iclog->ic_header.h_cycle_data, 0,
-			      sizeof(iclog->ic_header.h_cycle_data));
-			iclog->ic_header.h_lsn = 0;
-		} else if (iclog->ic_state == XLOG_STATE_ACTIVE)
-			/* do nothing */;
-		else
-			break;	/* stop cleaning */
-		iclog = iclog->ic_next;
-	} while (iclog != log->l_iclog);
-
+		if (iclog->ic_state == XLOG_STATE_DIRTY)
+			xlog_state_activate_iclog(iclog, iclogs_changed);
+		/*
+		 * The ordering of marking iclogs ACTIVE must be maintained, so
+		 * an iclog doesn't become ACTIVE beyond one that is SYNCING.
+		 */
+		else if (iclog->ic_state != XLOG_STATE_ACTIVE)
+			break;
+	} while ((iclog = iclog->ic_next) != log->l_iclog);
+}
 
+static int
+xlog_covered_state(
+	int			prev_state,
+	int			iclogs_changed)
+{
 	/*
-	 * Wake up threads waiting in xfs_log_force() for the dirty iclog
-	 * to be cleaned.
+	 * We usually go to NEED. But we go to NEED2 if the changed indicates we
+	 * are done writing the dummy record.  If we are done with the second
+	 * dummy recored (DONE2), then we go to IDLE.
 	 */
-	wake_up_all(&dirty_iclog->ic_force_wait);
+	switch (prev_state) {
+	case XLOG_STATE_COVER_IDLE:
+	case XLOG_STATE_COVER_NEED:
+	case XLOG_STATE_COVER_NEED2:
+		break;
+	case XLOG_STATE_COVER_DONE:
+		if (iclogs_changed == 1)
+			return XLOG_STATE_COVER_NEED2;
+		break;
+	case XLOG_STATE_COVER_DONE2:
+		if (iclogs_changed == 1)
+			return XLOG_STATE_COVER_IDLE;
+		break;
+	default:
+		ASSERT(0);
+	}
 
-	/*
-	 * Change state for the dummy log recording.
-	 * We usually go to NEED. But we go to NEED2 if the changed indicates
-	 * we are done writing the dummy record.
-	 * If we are done with the second dummy recored (DONE2), then
-	 * we go to IDLE.
-	 */
-	if (changed) {
-		switch (log->l_covered_state) {
-		case XLOG_STATE_COVER_IDLE:
-		case XLOG_STATE_COVER_NEED:
-		case XLOG_STATE_COVER_NEED2:
-			log->l_covered_state = XLOG_STATE_COVER_NEED;
-			break;
+	return XLOG_STATE_COVER_NEED;
+}
 
-		case XLOG_STATE_COVER_DONE:
-			if (changed == 1)
-				log->l_covered_state = XLOG_STATE_COVER_NEED2;
-			else
-				log->l_covered_state = XLOG_STATE_COVER_NEED;
-			break;
+STATIC void
+xlog_state_clean_iclog(
+	struct xlog		*log,
+	struct xlog_in_core	*dirty_iclog)
+{
+	int			iclogs_changed = 0;
 
-		case XLOG_STATE_COVER_DONE2:
-			if (changed == 1)
-				log->l_covered_state = XLOG_STATE_COVER_IDLE;
-			else
-				log->l_covered_state = XLOG_STATE_COVER_NEED;
-			break;
+	dirty_iclog->ic_state = XLOG_STATE_DIRTY;
 
-		default:
-			ASSERT(0);
-		}
+	xlog_state_activate_iclogs(log, &iclogs_changed);
+	wake_up_all(&dirty_iclog->ic_force_wait);
+
+	if (iclogs_changed) {
+		log->l_covered_state = xlog_covered_state(log->l_covered_state,
+				iclogs_changed);
 	}
 }
 
@@ -2639,7 +2570,8 @@ xlog_get_lowest_lsn(
 	xfs_lsn_t		lowest_lsn = 0, lsn;
 
 	do {
-		if (iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))
+		if (iclog->ic_state == XLOG_STATE_ACTIVE ||
+		    iclog->ic_state == XLOG_STATE_DIRTY)
 			continue;
 
 		lsn = be64_to_cpu(iclog->ic_header.h_lsn);
@@ -2699,61 +2631,48 @@ static bool
 xlog_state_iodone_process_iclog(
 	struct xlog		*log,
 	struct xlog_in_core	*iclog,
-	struct xlog_in_core	*completed_iclog,
 	bool			*ioerror)
 {
 	xfs_lsn_t		lowest_lsn;
 	xfs_lsn_t		header_lsn;
 
-	/* Skip all iclogs in the ACTIVE & DIRTY states */
-	if (iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))
+	switch (iclog->ic_state) {
+	case XLOG_STATE_ACTIVE:
+	case XLOG_STATE_DIRTY:
+		/*
+		 * Skip all iclogs in the ACTIVE & DIRTY states:
+		 */
 		return false;
-
-	/*
-	 * Between marking a filesystem SHUTDOWN and stopping the log, we do
-	 * flush all iclogs to disk (if there wasn't a log I/O error). So, we do
-	 * want things to go smoothly in case of just a SHUTDOWN  w/o a
-	 * LOG_IO_ERROR.
-	 */
-	if (iclog->ic_state & XLOG_STATE_IOERROR) {
+	case XLOG_STATE_IOERROR:
+		/*
+		 * Between marking a filesystem SHUTDOWN and stopping the log,
+		 * we do flush all iclogs to disk (if there wasn't a log I/O
+		 * error). So, we do want things to go smoothly in case of just
+		 * a SHUTDOWN w/o a LOG_IO_ERROR.
+		 */
 		*ioerror = true;
 		return false;
-	}
-
-	/*
-	 * Can only perform callbacks in order.  Since this iclog is not in the
-	 * DONE_SYNC/ DO_CALLBACK state, we skip the rest and just try to clean
-	 * up.  If we set our iclog to DO_CALLBACK, we will not process it when
-	 * we retry since a previous iclog is in the CALLBACK and the state
-	 * cannot change since we are holding the l_icloglock.
-	 */
-	if (!(iclog->ic_state &
-			(XLOG_STATE_DONE_SYNC | XLOG_STATE_DO_CALLBACK))) {
-		if (completed_iclog &&
-		    (completed_iclog->ic_state == XLOG_STATE_DONE_SYNC)) {
-			completed_iclog->ic_state = XLOG_STATE_DO_CALLBACK;
-		}
+	case XLOG_STATE_DONE_SYNC:
+		/*
+		 * Now that we have an iclog that is in the DONE_SYNC state, do
+		 * one more check here to see if we have chased our tail around.
+		 * If this is not the lowest lsn iclog, then we will leave it
+		 * for another completion to process.
+		 */
+		header_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
+		lowest_lsn = xlog_get_lowest_lsn(log);
+		if (lowest_lsn && XFS_LSN_CMP(lowest_lsn, header_lsn) < 0)
+			return false;
+		xlog_state_set_callback(log, iclog, header_lsn);
+		return false;
+	default:
+		/*
+		 * Can only perform callbacks in order.  Since this iclog is not
+		 * in the DONE_SYNC state, we skip the rest and just try to
+		 * clean up.
+		 */
 		return true;
 	}
-
-	/*
-	 * We now have an iclog that is in either the DO_CALLBACK or DONE_SYNC
-	 * states. The other states (WANT_SYNC, SYNCING, or CALLBACK were caught
-	 * by the above if and are going to clean (i.e. we aren't doing their
-	 * callbacks) see the above if.
-	 *
-	 * We will do one more check here to see if we have chased our tail
-	 * around. If this is not the lowest lsn iclog, then we will leave it
-	 * for another completion to process.
-	 */
-	header_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
-	lowest_lsn = xlog_get_lowest_lsn(log);
-	if (lowest_lsn && XFS_LSN_CMP(lowest_lsn, header_lsn) < 0)
-		return false;
-
-	xlog_state_set_callback(log, iclog, header_lsn);
-	return false;
-
 }
 
 /*
@@ -2768,8 +2687,9 @@ xlog_state_iodone_process_iclog(
 static void
 xlog_state_do_iclog_callbacks(
 	struct xlog		*log,
-	struct xlog_in_core	*iclog,
-	bool			aborted)
+	struct xlog_in_core	*iclog)
+		__releases(&log->l_icloglock)
+		__acquires(&log->l_icloglock)
 {
 	spin_unlock(&log->l_icloglock);
 	spin_lock(&iclog->ic_callback_lock);
@@ -2779,7 +2699,7 @@ xlog_state_do_iclog_callbacks(
 		list_splice_init(&iclog->ic_callbacks, &tmp);
 
 		spin_unlock(&iclog->ic_callback_lock);
-		xlog_cil_process_committed(&tmp, aborted);
+		xlog_cil_process_committed(&tmp);
 		spin_lock(&iclog->ic_callback_lock);
 	}
 
@@ -2792,57 +2712,12 @@ xlog_state_do_iclog_callbacks(
 	spin_unlock(&iclog->ic_callback_lock);
 }
 
-#ifdef DEBUG
-/*
- * Make one last gasp attempt to see if iclogs are being left in limbo.  If the
- * above loop finds an iclog earlier than the current iclog and in one of the
- * syncing states, the current iclog is put into DO_CALLBACK and the callbacks
- * are deferred to the completion of the earlier iclog. Walk the iclogs in order
- * and make sure that no iclog is in DO_CALLBACK unless an earlier iclog is in
- * one of the syncing states.
- *
- * Note that SYNCING|IOERROR is a valid state so we cannot just check for
- * ic_state == SYNCING.
- */
-static void
-xlog_state_callback_check_state(
-	struct xlog		*log)
-{
-	struct xlog_in_core	*first_iclog = log->l_iclog;
-	struct xlog_in_core	*iclog = first_iclog;
-
-	do {
-		ASSERT(iclog->ic_state != XLOG_STATE_DO_CALLBACK);
-		/*
-		 * Terminate the loop if iclogs are found in states
-		 * which will cause other threads to clean up iclogs.
-		 *
-		 * SYNCING - i/o completion will go through logs
-		 * DONE_SYNC - interrupt thread should be waiting for
-		 *              l_icloglock
-		 * IOERROR - give up hope all ye who enter here
-		 */
-		if (iclog->ic_state == XLOG_STATE_WANT_SYNC ||
-		    iclog->ic_state & XLOG_STATE_SYNCING ||
-		    iclog->ic_state == XLOG_STATE_DONE_SYNC ||
-		    iclog->ic_state == XLOG_STATE_IOERROR )
-			break;
-		iclog = iclog->ic_next;
-	} while (first_iclog != iclog);
-}
-#else
-#define xlog_state_callback_check_state(l)	((void)0)
-#endif
-
 STATIC void
 xlog_state_do_callback(
-	struct xlog		*log,
-	bool			aborted,
-	struct xlog_in_core	*ciclog)
+	struct xlog		*log)
 {
 	struct xlog_in_core	*iclog;
 	struct xlog_in_core	*first_iclog;
-	bool			did_callbacks = false;
 	bool			cycled_icloglock;
 	bool			ioerror;
 	int			flushcnt = 0;
@@ -2866,11 +2741,11 @@ xlog_state_do_callback(
 
 		do {
 			if (xlog_state_iodone_process_iclog(log, iclog,
-							ciclog, &ioerror))
+							&ioerror))
 				break;
 
-			if (!(iclog->ic_state &
-			      (XLOG_STATE_CALLBACK | XLOG_STATE_IOERROR))) {
+			if (iclog->ic_state != XLOG_STATE_CALLBACK &&
+			    iclog->ic_state != XLOG_STATE_IOERROR) {
 				iclog = iclog->ic_next;
 				continue;
 			}
@@ -2880,14 +2755,14 @@ xlog_state_do_callback(
 			 * we'll have to run at least one more complete loop.
 			 */
 			cycled_icloglock = true;
-			xlog_state_do_iclog_callbacks(log, iclog, aborted);
-
-			xlog_state_clean_iclog(log, iclog);
+			xlog_state_do_iclog_callbacks(log, iclog);
+			if (XLOG_FORCED_SHUTDOWN(log))
+				wake_up_all(&iclog->ic_force_wait);
+			else
+				xlog_state_clean_iclog(log, iclog);
 			iclog = iclog->ic_next;
 		} while (first_iclog != iclog);
 
-		did_callbacks |= cycled_icloglock;
-
 		if (repeats > 5000) {
 			flushcnt += repeats;
 			repeats = 0;
@@ -2897,10 +2772,8 @@ xlog_state_do_callback(
 		}
 	} while (!ioerror && cycled_icloglock);
 
-	if (did_callbacks)
-		xlog_state_callback_check_state(log);
-
-	if (log->l_iclog->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_IOERROR))
+	if (log->l_iclog->ic_state == XLOG_STATE_ACTIVE ||
+	    log->l_iclog->ic_state == XLOG_STATE_IOERROR)
 		wake_up_all(&log->l_flush_wait);
 
 	spin_unlock(&log->l_icloglock);
@@ -2922,25 +2795,22 @@ xlog_state_do_callback(
  */
 STATIC void
 xlog_state_done_syncing(
-	struct xlog_in_core	*iclog,
-	bool			aborted)
+	struct xlog_in_core	*iclog)
 {
 	struct xlog		*log = iclog->ic_log;
 
 	spin_lock(&log->l_icloglock);
-
-	ASSERT(iclog->ic_state == XLOG_STATE_SYNCING ||
-	       iclog->ic_state == XLOG_STATE_IOERROR);
 	ASSERT(atomic_read(&iclog->ic_refcnt) == 0);
 
 	/*
 	 * If we got an error, either on the first buffer, or in the case of
-	 * split log writes, on the second, we mark ALL iclogs STATE_IOERROR,
-	 * and none should ever be attempted to be written to disk
-	 * again.
+	 * split log writes, on the second, we shut down the file system and
+	 * no iclogs should ever be attempted to be written to disk again.
 	 */
-	if (iclog->ic_state != XLOG_STATE_IOERROR)
+	if (!XLOG_FORCED_SHUTDOWN(log)) {
+		ASSERT(iclog->ic_state == XLOG_STATE_SYNCING);
 		iclog->ic_state = XLOG_STATE_DONE_SYNC;
+	}
 
 	/*
 	 * Someone could be sleeping prior to writing out the next
@@ -2949,9 +2819,8 @@ xlog_state_done_syncing(
 	 */
 	wake_up_all(&iclog->ic_write_wait);
 	spin_unlock(&log->l_icloglock);
-	xlog_state_do_callback(log, aborted, iclog);	/* also cleans log */
-}	/* xlog_state_done_syncing */
-
+	xlog_state_do_callback(log);
+}
 
 /*
  * If the head of the in-core log ring is not (ACTIVE or DIRTY), then we must
@@ -2983,7 +2852,6 @@ xlog_state_get_iclog_space(
 	int		  log_offset;
 	xlog_rec_header_t *head;
 	xlog_in_core_t	  *iclog;
-	int		  error;
 
 restart:
 	spin_lock(&log->l_icloglock);
@@ -3032,24 +2900,22 @@ restart:
 	 * can fit into remaining data section.
 	 */
 	if (iclog->ic_size - iclog->ic_offset < 2*sizeof(xlog_op_header_t)) {
+		int		error = 0;
+
 		xlog_state_switch_iclogs(log, iclog, iclog->ic_size);
 
 		/*
-		 * If I'm the only one writing to this iclog, sync it to disk.
-		 * We need to do an atomic compare and decrement here to avoid
-		 * racing with concurrent atomic_dec_and_lock() calls in
+		 * If we are the only one writing to this iclog, sync it to
+		 * disk.  We need to do an atomic compare and decrement here to
+		 * avoid racing with concurrent atomic_dec_and_lock() calls in
 		 * xlog_state_release_iclog() when there is more than one
 		 * reference to the iclog.
 		 */
-		if (!atomic_add_unless(&iclog->ic_refcnt, -1, 1)) {
-			/* we are the only one */
-			spin_unlock(&log->l_icloglock);
+		if (!atomic_add_unless(&iclog->ic_refcnt, -1, 1))
 			error = xlog_state_release_iclog(log, iclog);
-			if (error)
-				return error;
-		} else {
-			spin_unlock(&log->l_icloglock);
-		}
+		spin_unlock(&log->l_icloglock);
+		if (error)
+			return error;
 		goto restart;
 	}
 
@@ -3073,21 +2939,21 @@ restart:
 
 	*logoffsetp = log_offset;
 	return 0;
-}	/* xlog_state_get_iclog_space */
-
-/* The first cnt-1 times through here we don't need to
- * move the grant write head because the permanent
- * reservation has reserved cnt times the unit amount.
- * Release part of current permanent unit reservation and
- * reset current reservation to be one units worth.  Also
- * move grant reservation head forward.
+}
+
+/*
+ * The first cnt-1 times a ticket goes through here we don't need to move the
+ * grant write head because the permanent reservation has reserved cnt times the
+ * unit amount.  Release part of current permanent unit reservation and reset
+ * current reservation to be one units worth.  Also move grant reservation head
+ * forward.
  */
-STATIC void
-xlog_regrant_reserve_log_space(
+void
+xfs_log_ticket_regrant(
 	struct xlog		*log,
 	struct xlog_ticket	*ticket)
 {
-	trace_xfs_log_regrant_reserve_enter(log, ticket);
+	trace_xfs_log_ticket_regrant(log, ticket);
 
 	if (ticket->t_cnt > 0)
 		ticket->t_cnt--;
@@ -3099,21 +2965,20 @@ xlog_regrant_reserve_log_space(
 	ticket->t_curr_res = ticket->t_unit_res;
 	xlog_tic_reset_res(ticket);
 
-	trace_xfs_log_regrant_reserve_sub(log, ticket);
+	trace_xfs_log_ticket_regrant_sub(log, ticket);
 
 	/* just return if we still have some of the pre-reserved space */
-	if (ticket->t_cnt > 0)
-		return;
+	if (!ticket->t_cnt) {
+		xlog_grant_add_space(log, &log->l_reserve_head.grant,
+				     ticket->t_unit_res);
+		trace_xfs_log_ticket_regrant_exit(log, ticket);
 
-	xlog_grant_add_space(log, &log->l_reserve_head.grant,
-					ticket->t_unit_res);
-
-	trace_xfs_log_regrant_reserve_exit(log, ticket);
-
-	ticket->t_curr_res = ticket->t_unit_res;
-	xlog_tic_reset_res(ticket);
-}	/* xlog_regrant_reserve_log_space */
+		ticket->t_curr_res = ticket->t_unit_res;
+		xlog_tic_reset_res(ticket);
+	}
 
+	xfs_log_ticket_put(ticket);
+}
 
 /*
  * Give back the space left from a reservation.
@@ -3129,18 +2994,19 @@ xlog_regrant_reserve_log_space(
  * space, the count will stay at zero and the only space remaining will be
  * in the current reservation field.
  */
-STATIC void
-xlog_ungrant_log_space(
+void
+xfs_log_ticket_ungrant(
 	struct xlog		*log,
 	struct xlog_ticket	*ticket)
 {
-	int	bytes;
+	int			bytes;
+
+	trace_xfs_log_ticket_ungrant(log, ticket);
 
 	if (ticket->t_cnt > 0)
 		ticket->t_cnt--;
 
-	trace_xfs_log_ungrant_enter(log, ticket);
-	trace_xfs_log_ungrant_sub(log, ticket);
+	trace_xfs_log_ticket_ungrant_sub(log, ticket);
 
 	/*
 	 * If this is a permanent reservation ticket, we may be able to free
@@ -3155,71 +3021,15 @@ xlog_ungrant_log_space(
 	xlog_grant_sub_space(log, &log->l_reserve_head.grant, bytes);
 	xlog_grant_sub_space(log, &log->l_write_head.grant, bytes);
 
-	trace_xfs_log_ungrant_exit(log, ticket);
+	trace_xfs_log_ticket_ungrant_exit(log, ticket);
 
 	xfs_log_space_wake(log->l_mp);
+	xfs_log_ticket_put(ticket);
 }
 
 /*
- * Flush iclog to disk if this is the last reference to the given iclog and
- * the WANT_SYNC bit is set.
- *
- * When this function is entered, the iclog is not necessarily in the
- * WANT_SYNC state.  It may be sitting around waiting to get filled.
- *
- *
- */
-STATIC int
-xlog_state_release_iclog(
-	struct xlog		*log,
-	struct xlog_in_core	*iclog)
-{
-	int		sync = 0;	/* do we sync? */
-
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
-		return -EIO;
-
-	ASSERT(atomic_read(&iclog->ic_refcnt) > 0);
-	if (!atomic_dec_and_lock(&iclog->ic_refcnt, &log->l_icloglock))
-		return 0;
-
-	if (iclog->ic_state & XLOG_STATE_IOERROR) {
-		spin_unlock(&log->l_icloglock);
-		return -EIO;
-	}
-	ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE ||
-	       iclog->ic_state == XLOG_STATE_WANT_SYNC);
-
-	if (iclog->ic_state == XLOG_STATE_WANT_SYNC) {
-		/* update tail before writing to iclog */
-		xfs_lsn_t tail_lsn = xlog_assign_tail_lsn(log->l_mp);
-		sync++;
-		iclog->ic_state = XLOG_STATE_SYNCING;
-		iclog->ic_header.h_tail_lsn = cpu_to_be64(tail_lsn);
-		xlog_verify_tail_lsn(log, iclog, tail_lsn);
-		/* cycle incremented when incrementing curr_block */
-	}
-	spin_unlock(&log->l_icloglock);
-
-	/*
-	 * We let the log lock go, so it's possible that we hit a log I/O
-	 * error or some other SHUTDOWN condition that marks the iclog
-	 * as XLOG_STATE_IOERROR before the bwrite. However, we know that
-	 * this iclog has consistent data, so we ignore IOERROR
-	 * flags after this point.
-	 */
-	if (sync)
-		xlog_sync(log, iclog);
-	return 0;
-}	/* xlog_state_release_iclog */
-
-
-/*
- * This routine will mark the current iclog in the ring as WANT_SYNC
- * and move the current iclog pointer to the next iclog in the ring.
- * When this routine is called from xlog_state_get_iclog_space(), the
- * exact size of the iclog has not yet been determined.  All we know is
- * that every data block.  We have run out of space in this log record.
+ * This routine will mark the current iclog in the ring as WANT_SYNC and move
+ * the current iclog pointer to the next iclog in the ring.
  */
 STATIC void
 xlog_state_switch_iclogs(
@@ -3228,6 +3038,8 @@ xlog_state_switch_iclogs(
 	int			eventual_size)
 {
 	ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE);
+	assert_spin_locked(&log->l_icloglock);
+
 	if (!eventual_size)
 		eventual_size = iclog->ic_offset;
 	iclog->ic_state = XLOG_STATE_WANT_SYNC;
@@ -3262,7 +3074,7 @@ xlog_state_switch_iclogs(
 	}
 	ASSERT(iclog == log->l_iclog);
 	log->l_iclog = iclog->ic_next;
-}	/* xlog_state_switch_iclogs */
+}
 
 /*
  * Write out all data in the in-core log as of this exact moment in time.
@@ -3307,7 +3119,7 @@ xfs_log_force(
 
 	spin_lock(&log->l_icloglock);
 	iclog = log->l_iclog;
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
 		goto out_error;
 
 	if (iclog->ic_state == XLOG_STATE_DIRTY ||
@@ -3322,9 +3134,6 @@ xfs_log_force(
 		 * previous iclog and go to sleep.
 		 */
 		iclog = iclog->ic_prev;
-		if (iclog->ic_state == XLOG_STATE_ACTIVE ||
-		    iclog->ic_state == XLOG_STATE_DIRTY)
-			goto out_unlock;
 	} else if (iclog->ic_state == XLOG_STATE_ACTIVE) {
 		if (atomic_read(&iclog->ic_refcnt) == 0) {
 			/*
@@ -3337,14 +3146,10 @@ xfs_log_force(
 			atomic_inc(&iclog->ic_refcnt);
 			lsn = be64_to_cpu(iclog->ic_header.h_lsn);
 			xlog_state_switch_iclogs(log, iclog, 0);
-			spin_unlock(&log->l_icloglock);
-
 			if (xlog_state_release_iclog(log, iclog))
-				return -EIO;
+				goto out_error;
 
-			spin_lock(&log->l_icloglock);
-			if (be64_to_cpu(iclog->ic_header.h_lsn) != lsn ||
-			    iclog->ic_state == XLOG_STATE_DIRTY)
+			if (be64_to_cpu(iclog->ic_header.h_lsn) != lsn)
 				goto out_unlock;
 		} else {
 			/*
@@ -3364,17 +3169,8 @@ xfs_log_force(
 		;
 	}
 
-	if (!(flags & XFS_LOG_SYNC))
-		goto out_unlock;
-
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
-		goto out_error;
-	XFS_STATS_INC(mp, xs_log_force_sleep);
-	xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
-		return -EIO;
-	return 0;
-
+	if (flags & XFS_LOG_SYNC)
+		return xlog_wait_on_iclog(iclog);
 out_unlock:
 	spin_unlock(&log->l_icloglock);
 	return 0;
@@ -3396,7 +3192,7 @@ __xfs_log_force_lsn(
 
 	spin_lock(&log->l_icloglock);
 	iclog = log->l_iclog;
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
+	if (iclog->ic_state == XLOG_STATE_IOERROR)
 		goto out_error;
 
 	while (be64_to_cpu(iclog->ic_header.h_lsn) != lsn) {
@@ -3405,9 +3201,6 @@ __xfs_log_force_lsn(
 			goto out_unlock;
 	}
 
-	if (iclog->ic_state == XLOG_STATE_DIRTY)
-		goto out_unlock;
-
 	if (iclog->ic_state == XLOG_STATE_ACTIVE) {
 		/*
 		 * We sleep here if we haven't already slept (e.g. this is the
@@ -3425,10 +3218,8 @@ __xfs_log_force_lsn(
 		 * will go out then.
 		 */
 		if (!already_slept &&
-		    (iclog->ic_prev->ic_state &
-		     (XLOG_STATE_WANT_SYNC | XLOG_STATE_SYNCING))) {
-			ASSERT(!(iclog->ic_state & XLOG_STATE_IOERROR));
-
+		    (iclog->ic_prev->ic_state == XLOG_STATE_WANT_SYNC ||
+		     iclog->ic_prev->ic_state == XLOG_STATE_SYNCING)) {
 			XFS_STATS_INC(mp, xs_log_force_sleep);
 
 			xlog_wait(&iclog->ic_prev->ic_write_wait,
@@ -3437,27 +3228,14 @@ __xfs_log_force_lsn(
 		}
 		atomic_inc(&iclog->ic_refcnt);
 		xlog_state_switch_iclogs(log, iclog, 0);
-		spin_unlock(&log->l_icloglock);
 		if (xlog_state_release_iclog(log, iclog))
-			return -EIO;
+			goto out_error;
 		if (log_flushed)
 			*log_flushed = 1;
-		spin_lock(&log->l_icloglock);
 	}
 
-	if (!(flags & XFS_LOG_SYNC) ||
-	    (iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY)))
-		goto out_unlock;
-
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
-		goto out_error;
-
-	XFS_STATS_INC(mp, xs_log_force_sleep);
-	xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
-	if (iclog->ic_state & XLOG_STATE_IOERROR)
-		return -EIO;
-	return 0;
-
+	if (flags & XFS_LOG_SYNC)
+		return xlog_wait_on_iclog(iclog);
 out_unlock:
 	spin_unlock(&log->l_icloglock);
 	return 0;
@@ -3504,33 +3282,6 @@ xfs_log_force_lsn(
 }
 
 /*
- * Called when we want to mark the current iclog as being ready to sync to
- * disk.
- */
-STATIC void
-xlog_state_want_sync(
-	struct xlog		*log,
-	struct xlog_in_core	*iclog)
-{
-	assert_spin_locked(&log->l_icloglock);
-
-	if (iclog->ic_state == XLOG_STATE_ACTIVE) {
-		xlog_state_switch_iclogs(log, iclog, 0);
-	} else {
-		ASSERT(iclog->ic_state &
-			(XLOG_STATE_WANT_SYNC|XLOG_STATE_IOERROR));
-	}
-}
-
-
-/*****************************************************************************
- *
- *		TICKET functions
- *
- *****************************************************************************
- */
-
-/*
  * Free a used ticket when its refcount falls to zero.
  */
 void
@@ -3539,7 +3290,7 @@ xfs_log_ticket_put(
 {
 	ASSERT(atomic_read(&ticket->t_ref) > 0);
 	if (atomic_dec_and_test(&ticket->t_ref))
-		kmem_zone_free(xfs_log_ticket_zone, ticket);
+		kmem_cache_free(xfs_log_ticket_zone, ticket);
 }
 
 xlog_ticket_t *
@@ -3678,7 +3429,6 @@ xlog_ticket_alloc(
 	tic->t_ocnt		= cnt;
 	tic->t_tid		= prandom_u32();
 	tic->t_clientid		= client;
-	tic->t_flags		= XLOG_TIC_INITED;
 	if (permanent)
 		tic->t_flags |= XLOG_TIC_PERM_RESERV;
 
@@ -3687,13 +3437,6 @@ xlog_ticket_alloc(
 	return tic;
 }
 
-
-/******************************************************************************
- *
- *		Log debug routines
- *
- ******************************************************************************
- */
 #if defined(DEBUG)
 /*
  * Make sure that the destination ptr is within the valid data region of
@@ -3779,7 +3522,7 @@ xlog_verify_tail_lsn(
 	if (blocks < BTOBB(iclog->ic_offset) + 1)
 		xfs_emerg(log->l_mp, "%s: ran out of log space", __func__);
     }
-}	/* xlog_verify_tail_lsn */
+}
 
 /*
  * Perform a number of checks on the iclog before writing to disk.
@@ -3882,7 +3625,7 @@ xlog_verify_iclog(
 		}
 		ptr += sizeof(xlog_op_header_t) + op_len;
 	}
-}	/* xlog_verify_iclog */
+}
 #endif
 
 /*
@@ -3895,7 +3638,7 @@ xlog_state_ioerror(
 	xlog_in_core_t	*iclog, *ic;
 
 	iclog = log->l_iclog;
-	if (! (iclog->ic_state & XLOG_STATE_IOERROR)) {
+	if (iclog->ic_state != XLOG_STATE_IOERROR) {
 		/*
 		 * Mark all the incore logs IOERROR.
 		 * From now on, no log flushes will result.
@@ -3955,7 +3698,7 @@ xfs_log_force_umount(
 	 * Somebody could've already done the hard work for us.
 	 * No need to get locks for this.
 	 */
-	if (logerror && log->l_iclog->ic_state & XLOG_STATE_IOERROR) {
+	if (logerror && log->l_iclog->ic_state == XLOG_STATE_IOERROR) {
 		ASSERT(XLOG_FORCED_SHUTDOWN(log));
 		return 1;
 	}
@@ -4006,21 +3749,8 @@ xfs_log_force_umount(
 	spin_lock(&log->l_cilp->xc_push_lock);
 	wake_up_all(&log->l_cilp->xc_commit_wait);
 	spin_unlock(&log->l_cilp->xc_push_lock);
-	xlog_state_do_callback(log, true, NULL);
+	xlog_state_do_callback(log);
 
-#ifdef XFSERRORDEBUG
-	{
-		xlog_in_core_t	*iclog;
-
-		spin_lock(&log->l_icloglock);
-		iclog = log->l_iclog;
-		do {
-			ASSERT(iclog->ic_callback == 0);
-			iclog = iclog->ic_next;
-		} while (iclog != log->l_iclog);
-		spin_unlock(&log->l_icloglock);
-	}
-#endif
 	/* return non-zero if log IOERROR transition had already happened */
 	return retval;
 }
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 84e06805160f..1412d6993f1e 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -105,10 +105,6 @@ struct xfs_log_item;
 struct xfs_item_ops;
 struct xfs_trans;
 
-xfs_lsn_t xfs_log_done(struct xfs_mount *mp,
-		       struct xlog_ticket *ticket,
-		       struct xlog_in_core **iclog,
-		       bool regrant);
 int	  xfs_log_force(struct xfs_mount *mp, uint flags);
 int	  xfs_log_force_lsn(struct xfs_mount *mp, xfs_lsn_t lsn, uint flags,
 		int *log_forced);
@@ -121,8 +117,7 @@ void	xfs_log_mount_cancel(struct xfs_mount *);
 xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp);
 xfs_lsn_t xlog_assign_tail_lsn_locked(struct xfs_mount *mp);
 void	  xfs_log_space_wake(struct xfs_mount *mp);
-int	  xfs_log_release_iclog(struct xfs_mount *mp,
-			 struct xlog_in_core	 *iclog);
+void	  xfs_log_release_iclog(struct xlog_in_core *iclog);
 int	  xfs_log_reserve(struct xfs_mount *mp,
 			  int		   length,
 			  int		   count,
@@ -138,7 +133,7 @@ void	  xfs_log_ticket_put(struct xlog_ticket *ticket);
 
 void	xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
 				xfs_lsn_t *commit_lsn, bool regrant);
-void	xlog_cil_process_committed(struct list_head *list, bool aborted);
+void	xlog_cil_process_committed(struct list_head *list);
 bool	xfs_log_item_in_current_chkpt(struct xfs_log_item *lip);
 
 void	xfs_log_work_queue(struct xfs_mount *mp);
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index ef652abd112c..b43f0e8f43f2 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -179,7 +179,7 @@ xlog_cil_alloc_shadow_bufs(
 
 			/*
 			 * We free and allocate here as a realloc would copy
-			 * unecessary data. We don't use kmem_zalloc() for the
+			 * unnecessary data. We don't use kmem_zalloc() for the
 			 * same reason - we don't need to zero the data area in
 			 * the buffer, only the log vector header and the iovec
 			 * storage.
@@ -574,10 +574,10 @@ xlog_discard_busy_extents(
  */
 static void
 xlog_cil_committed(
-	struct xfs_cil_ctx	*ctx,
-	bool			abort)
+	struct xfs_cil_ctx	*ctx)
 {
 	struct xfs_mount	*mp = ctx->cil->xc_log->l_mp;
+	bool			abort = XLOG_FORCED_SHUTDOWN(ctx->cil->xc_log);
 
 	/*
 	 * If the I/O failed, we're aborting the commit and already shutdown.
@@ -613,37 +613,38 @@ xlog_cil_committed(
 
 void
 xlog_cil_process_committed(
-	struct list_head	*list,
-	bool			aborted)
+	struct list_head	*list)
 {
 	struct xfs_cil_ctx	*ctx;
 
 	while ((ctx = list_first_entry_or_null(list,
 			struct xfs_cil_ctx, iclog_entry))) {
 		list_del(&ctx->iclog_entry);
-		xlog_cil_committed(ctx, aborted);
+		xlog_cil_committed(ctx);
 	}
 }
 
 /*
- * Push the Committed Item List to the log. If @push_seq flag is zero, then it
- * is a background flush and so we can chose to ignore it. Otherwise, if the
- * current sequence is the same as @push_seq we need to do a flush. If
- * @push_seq is less than the current sequence, then it has already been
+ * Push the Committed Item List to the log.
+ *
+ * If the current sequence is the same as xc_push_seq we need to do a flush. If
+ * xc_push_seq is less than the current sequence, then it has already been
  * flushed and we don't need to do anything - the caller will wait for it to
  * complete if necessary.
  *
- * @push_seq is a value rather than a flag because that allows us to do an
- * unlocked check of the sequence number for a match. Hence we can allows log
- * forces to run racily and not issue pushes for the same sequence twice. If we
- * get a race between multiple pushes for the same sequence they will block on
- * the first one and then abort, hence avoiding needless pushes.
+ * xc_push_seq is checked unlocked against the sequence number for a match.
+ * Hence we can allow log forces to run racily and not issue pushes for the
+ * same sequence twice.  If we get a race between multiple pushes for the same
+ * sequence they will block on the first one and then abort, hence avoiding
+ * needless pushes.
  */
-STATIC int
-xlog_cil_push(
-	struct xlog		*log)
+static void
+xlog_cil_push_work(
+	struct work_struct	*work)
 {
-	struct xfs_cil		*cil = log->l_cilp;
+	struct xfs_cil		*cil =
+		container_of(work, struct xfs_cil, xc_push_work);
+	struct xlog		*log = cil->xc_log;
 	struct xfs_log_vec	*lv;
 	struct xfs_cil_ctx	*ctx;
 	struct xfs_cil_ctx	*new_ctx;
@@ -657,9 +658,6 @@ xlog_cil_push(
 	xfs_lsn_t		commit_lsn;
 	xfs_lsn_t		push_seq;
 
-	if (!cil)
-		return 0;
-
 	new_ctx = kmem_zalloc(sizeof(*new_ctx), KM_NOFS);
 	new_ctx->ticket = xlog_cil_ticket_alloc(log);
 
@@ -671,6 +669,11 @@ xlog_cil_push(
 	ASSERT(push_seq <= ctx->sequence);
 
 	/*
+	 * Wake up any background push waiters now this context is being pushed.
+	 */
+	wake_up_all(&ctx->push_wait);
+
+	/*
 	 * Check if we've anything to push. If there is nothing, then we don't
 	 * move on to a new sequence number and so we have to be able to push
 	 * this sequence again later.
@@ -682,7 +685,7 @@ xlog_cil_push(
 	}
 
 
-	/* check for a previously pushed seqeunce */
+	/* check for a previously pushed sequence */
 	if (push_seq < cil->xc_ctx->sequence) {
 		spin_unlock(&cil->xc_push_lock);
 		goto out_skip;
@@ -746,6 +749,7 @@ xlog_cil_push(
 	 */
 	INIT_LIST_HEAD(&new_ctx->committing);
 	INIT_LIST_HEAD(&new_ctx->busy_extents);
+	init_waitqueue_head(&new_ctx->push_wait);
 	new_ctx->sequence = ctx->sequence + 1;
 	new_ctx->cil = cil;
 	cil->xc_ctx = new_ctx;
@@ -803,7 +807,7 @@ xlog_cil_push(
 	lvhdr.lv_iovecp = &lhdr;
 	lvhdr.lv_next = ctx->lv_chain;
 
-	error = xlog_write(log, &lvhdr, tic, &ctx->start_lsn, NULL, 0);
+	error = xlog_write(log, &lvhdr, tic, &ctx->start_lsn, NULL, 0, true);
 	if (error)
 		goto out_abort_free_ticket;
 
@@ -841,13 +845,14 @@ restart:
 	}
 	spin_unlock(&cil->xc_push_lock);
 
-	/* xfs_log_done always frees the ticket on error. */
-	commit_lsn = xfs_log_done(log->l_mp, tic, &commit_iclog, false);
-	if (commit_lsn == -1)
-		goto out_abort;
+	error = xlog_commit_record(log, tic, &commit_iclog, &commit_lsn);
+	if (error)
+		goto out_abort_free_ticket;
+
+	xfs_log_ticket_ungrant(log, tic);
 
 	spin_lock(&commit_iclog->ic_callback_lock);
-	if (commit_iclog->ic_state & XLOG_STATE_IOERROR) {
+	if (commit_iclog->ic_state == XLOG_STATE_IOERROR) {
 		spin_unlock(&commit_iclog->ic_callback_lock);
 		goto out_abort;
 	}
@@ -867,28 +872,20 @@ restart:
 	spin_unlock(&cil->xc_push_lock);
 
 	/* release the hounds! */
-	return xfs_log_release_iclog(log->l_mp, commit_iclog);
+	xfs_log_release_iclog(commit_iclog);
+	return;
 
 out_skip:
 	up_write(&cil->xc_ctx_lock);
 	xfs_log_ticket_put(new_ctx->ticket);
 	kmem_free(new_ctx);
-	return 0;
+	return;
 
 out_abort_free_ticket:
-	xfs_log_ticket_put(tic);
+	xfs_log_ticket_ungrant(log, tic);
 out_abort:
-	xlog_cil_committed(ctx, true);
-	return -EIO;
-}
-
-static void
-xlog_cil_push_work(
-	struct work_struct	*work)
-{
-	struct xfs_cil		*cil = container_of(work, struct xfs_cil,
-							xc_push_work);
-	xlog_cil_push(cil->xc_log);
+	ASSERT(XLOG_FORCED_SHUTDOWN(log));
+	xlog_cil_committed(ctx);
 }
 
 /*
@@ -900,7 +897,7 @@ xlog_cil_push_work(
  */
 static void
 xlog_cil_push_background(
-	struct xlog	*log)
+	struct xlog	*log) __releases(cil->xc_ctx_lock)
 {
 	struct xfs_cil	*cil = log->l_cilp;
 
@@ -914,14 +911,36 @@ xlog_cil_push_background(
 	 * don't do a background push if we haven't used up all the
 	 * space available yet.
 	 */
-	if (cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log))
+	if (cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log)) {
+		up_read(&cil->xc_ctx_lock);
 		return;
+	}
 
 	spin_lock(&cil->xc_push_lock);
 	if (cil->xc_push_seq < cil->xc_current_sequence) {
 		cil->xc_push_seq = cil->xc_current_sequence;
 		queue_work(log->l_mp->m_cil_workqueue, &cil->xc_push_work);
 	}
+
+	/*
+	 * Drop the context lock now, we can't hold that if we need to sleep
+	 * because we are over the blocking threshold. The push_lock is still
+	 * held, so blocking threshold sleep/wakeup is still correctly
+	 * serialised here.
+	 */
+	up_read(&cil->xc_ctx_lock);
+
+	/*
+	 * If we are well over the space limit, throttle the work that is being
+	 * done until the push work on this context has begun.
+	 */
+	if (cil->xc_ctx->space_used >= XLOG_CIL_BLOCKING_SPACE_LIMIT(log)) {
+		trace_xfs_log_cil_wait(log, cil->xc_ctx->ticket);
+		ASSERT(cil->xc_ctx->space_used < log->l_logsize);
+		xlog_wait(&cil->xc_ctx->push_wait, &cil->xc_push_lock);
+		return;
+	}
+
 	spin_unlock(&cil->xc_push_lock);
 
 }
@@ -1017,7 +1036,10 @@ xfs_log_commit_cil(
 	if (commit_lsn)
 		*commit_lsn = xc_commit_lsn;
 
-	xfs_log_done(mp, tp->t_ticket, NULL, regrant);
+	if (regrant && !XLOG_FORCED_SHUTDOWN(log))
+		xfs_log_ticket_regrant(log, tp->t_ticket);
+	else
+		xfs_log_ticket_ungrant(log, tp->t_ticket);
 	tp->t_ticket = NULL;
 	xfs_trans_unreserve_and_mod_sb(tp);
 
@@ -1038,9 +1060,9 @@ xfs_log_commit_cil(
 		if (lip->li_ops->iop_committing)
 			lip->li_ops->iop_committing(lip, xc_commit_lsn);
 	}
-	xlog_cil_push_background(log);
 
-	up_read(&cil->xc_ctx_lock);
+	/* xlog_cil_push_background() releases cil->xc_ctx_lock */
+	xlog_cil_push_background(log);
 }
 
 /*
@@ -1199,6 +1221,7 @@ xlog_cil_init(
 
 	INIT_LIST_HEAD(&ctx->committing);
 	INIT_LIST_HEAD(&ctx->busy_extents);
+	init_waitqueue_head(&ctx->push_wait);
 	ctx->sequence = 1;
 	ctx->cil = cil;
 	cil->xc_ctx = ctx;
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index b880c23cb6e4..ec22c7a3867f 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -40,26 +40,22 @@ static inline uint xlog_get_client_id(__be32 i)
 /*
  * In core log state
  */
-#define XLOG_STATE_ACTIVE    0x0001 /* Current IC log being written to */
-#define XLOG_STATE_WANT_SYNC 0x0002 /* Want to sync this iclog; no more writes */
-#define XLOG_STATE_SYNCING   0x0004 /* This IC log is syncing */
-#define XLOG_STATE_DONE_SYNC 0x0008 /* Done syncing to disk */
-#define XLOG_STATE_DO_CALLBACK \
-			     0x0010 /* Process callback functions */
-#define XLOG_STATE_CALLBACK  0x0020 /* Callback functions now */
-#define XLOG_STATE_DIRTY     0x0040 /* Dirty IC log, not ready for ACTIVE status*/
-#define XLOG_STATE_IOERROR   0x0080 /* IO error happened in sync'ing log */
-#define XLOG_STATE_ALL	     0x7FFF /* All possible valid flags */
-#define XLOG_STATE_NOTUSED   0x8000 /* This IC log not being used */
+enum xlog_iclog_state {
+	XLOG_STATE_ACTIVE,	/* Current IC log being written to */
+	XLOG_STATE_WANT_SYNC,	/* Want to sync this iclog; no more writes */
+	XLOG_STATE_SYNCING,	/* This IC log is syncing */
+	XLOG_STATE_DONE_SYNC,	/* Done syncing to disk */
+	XLOG_STATE_CALLBACK,	/* Callback functions now */
+	XLOG_STATE_DIRTY,	/* Dirty IC log, not ready for ACTIVE status */
+	XLOG_STATE_IOERROR,	/* IO error happened in sync'ing log */
+};
 
 /*
- * Flags to log ticket
+ * Log ticket flags
  */
-#define XLOG_TIC_INITED		0x1	/* has been initialized */
-#define XLOG_TIC_PERM_RESERV	0x2	/* permanent reservation */
+#define XLOG_TIC_PERM_RESERV	0x1	/* permanent reservation */
 
 #define XLOG_TIC_FLAGS \
-	{ XLOG_TIC_INITED,	"XLOG_TIC_INITED" }, \
 	{ XLOG_TIC_PERM_RESERV,	"XLOG_TIC_PERM_RESERV" }
 
 /*
@@ -179,8 +175,6 @@ typedef struct xlog_ticket {
  * - ic_next is the pointer to the next iclog in the ring.
  * - ic_log is a pointer back to the global log structure.
  * - ic_size is the full size of the log buffer, minus the cycle headers.
- * - ic_io_size is the size of the currently pending log buffer write, which
- *	might be smaller than ic_size
  * - ic_offset is the current number of bytes written to in this iclog.
  * - ic_refcnt is bumped when someone is writing to the log.
  * - ic_state is the state of the iclog.
@@ -205,9 +199,8 @@ typedef struct xlog_in_core {
 	struct xlog_in_core	*ic_prev;
 	struct xlog		*ic_log;
 	u32			ic_size;
-	u32			ic_io_size;
 	u32			ic_offset;
-	unsigned short		ic_state;
+	enum xlog_iclog_state	ic_state;
 	char			*ic_datap;	/* pointer to iclog data */
 
 	/* Callback structures need their own cacheline */
@@ -247,6 +240,7 @@ struct xfs_cil_ctx {
 	struct xfs_log_vec	*lv_chain;	/* logvecs being pushed */
 	struct list_head	iclog_entry;
 	struct list_head	committing;	/* ctx committing list */
+	wait_queue_head_t	push_wait;	/* background push throttle */
 	struct work_struct	discard_endio_work;
 };
 
@@ -323,13 +317,53 @@ struct xfs_cil {
  * tries to keep 25% of the log free, so we need to keep below that limit or we
  * risk running out of free log space to start any new transactions.
  *
- * In order to keep background CIL push efficient, we will set a lower
- * threshold at which background pushing is attempted without blocking current
- * transaction commits.  A separate, higher bound defines when CIL pushes are
- * enforced to ensure we stay within our maximum checkpoint size bounds.
- * threshold, yet give us plenty of space for aggregation on large logs.
+ * In order to keep background CIL push efficient, we only need to ensure the
+ * CIL is large enough to maintain sufficient in-memory relogging to avoid
+ * repeated physical writes of frequently modified metadata. If we allow the CIL
+ * to grow to a substantial fraction of the log, then we may be pinning hundreds
+ * of megabytes of metadata in memory until the CIL flushes. This can cause
+ * issues when we are running low on memory - pinned memory cannot be reclaimed,
+ * and the CIL consumes a lot of memory. Hence we need to set an upper physical
+ * size limit for the CIL that limits the maximum amount of memory pinned by the
+ * CIL but does not limit performance by reducing relogging efficiency
+ * significantly.
+ *
+ * As such, the CIL push threshold ends up being the smaller of two thresholds:
+ * - a threshold large enough that it allows CIL to be pushed and progress to be
+ *   made without excessive blocking of incoming transaction commits. This is
+ *   defined to be 12.5% of the log space - half the 25% push threshold of the
+ *   AIL.
+ * - small enough that it doesn't pin excessive amounts of memory but maintains
+ *   close to peak relogging efficiency. This is defined to be 16x the iclog
+ *   buffer window (32MB) as measurements have shown this to be roughly the
+ *   point of diminishing performance increases under highly concurrent
+ *   modification workloads.
+ *
+ * To prevent the CIL from overflowing upper commit size bounds, we introduce a
+ * new threshold at which we block committing transactions until the background
+ * CIL commit commences and switches to a new context. While this is not a hard
+ * limit, it forces the process committing a transaction to the CIL to block and
+ * yeild the CPU, giving the CIL push work a chance to be scheduled and start
+ * work. This prevents a process running lots of transactions from overfilling
+ * the CIL because it is not yielding the CPU. We set the blocking limit at
+ * twice the background push space threshold so we keep in line with the AIL
+ * push thresholds.
+ *
+ * Note: this is not a -hard- limit as blocking is applied after the transaction
+ * is inserted into the CIL and the push has been triggered. It is largely a
+ * throttling mechanism that allows the CIL push to be scheduled and run. A hard
+ * limit will be difficult to implement without introducing global serialisation
+ * in the CIL commit fast path, and it's not at all clear that we actually need
+ * such hard limits given the ~7 years we've run without a hard limit before
+ * finding the first situation where a checkpoint size overflow actually
+ * occurred. Hence the simple throttle, and an ASSERT check to tell us that
+ * we've overrun the max size.
  */
-#define XLOG_CIL_SPACE_LIMIT(log)	(log->l_logsize >> 3)
+#define XLOG_CIL_SPACE_LIMIT(log)	\
+	min_t(int, (log)->l_logsize >> 3, BBTOB(XLOG_TOTAL_REC_SHIFT(log)) << 4)
+
+#define XLOG_CIL_BLOCKING_SPACE_LIMIT(log)	\
+	(XLOG_CIL_SPACE_LIMIT(log) * 2)
 
 /*
  * ticket grant locks, queues and accounting have their own cachlines
@@ -399,8 +433,6 @@ struct xlog {
 	/* The following field are used for debugging; need to hold icloglock */
 #ifdef DEBUG
 	void			*l_iclog_bak[XLOG_MAX_ICLOGS];
-	/* log record crc error injection factor */
-	uint32_t		l_badcrc_factor;
 #endif
 	/* log recovery lsn tracking (for buffer submission */
 	xfs_lsn_t		l_recovery_lsn;
@@ -409,7 +441,8 @@ struct xlog {
 #define XLOG_BUF_CANCEL_BUCKET(log, blkno) \
 	((log)->l_buf_cancel_table + ((uint64_t)blkno % XLOG_BC_TABLE_SIZE))
 
-#define XLOG_FORCED_SHUTDOWN(log)	((log)->l_flags & XLOG_IO_ERROR)
+#define XLOG_FORCED_SHUTDOWN(log) \
+	(unlikely((log)->l_flags & XLOG_IO_ERROR))
 
 /* common routines */
 extern int
@@ -445,14 +478,14 @@ xlog_write_adv_cnt(void **ptr, int *len, int *off, size_t bytes)
 
 void	xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);
 void	xlog_print_trans(struct xfs_trans *);
-int
-xlog_write(
-	struct xlog		*log,
-	struct xfs_log_vec	*log_vector,
-	struct xlog_ticket	*tic,
-	xfs_lsn_t		*start_lsn,
-	struct xlog_in_core	**commit_iclog,
-	uint			flags);
+int	xlog_write(struct xlog *log, struct xfs_log_vec *log_vector,
+		struct xlog_ticket *tic, xfs_lsn_t *start_lsn,
+		struct xlog_in_core **commit_iclog, uint flags,
+		bool need_start_rec);
+int	xlog_commit_record(struct xlog *log, struct xlog_ticket *ticket,
+		struct xlog_in_core **iclog, xfs_lsn_t *lsn);
+void	xfs_log_ticket_ungrant(struct xlog *log, struct xlog_ticket *ticket);
+void	xfs_log_ticket_regrant(struct xlog *log, struct xlog_ticket *ticket);
 
 /*
  * When we crack an atomic LSN, we sample it first so that the value will not
@@ -532,17 +565,15 @@ xlog_cil_force(struct xlog *log)
 }
 
 /*
- * Unmount record type is used as a pseudo transaction type for the ticket.
- * It's value must be outside the range of XFS_TRANS_* values.
- */
-#define XLOG_UNMOUNT_REC_TYPE	(-1U)
-
-/*
  * Wrapper function for waiting on a wait queue serialised against wakeups
  * by a spinlock. This matches the semantics of all the wait queues used in the
  * log code.
  */
-static inline void xlog_wait(wait_queue_head_t *wq, spinlock_t *lock)
+static inline void
+xlog_wait(
+	struct wait_queue_head	*wq,
+	struct spinlock		*lock)
+		__releases(lock)
 {
 	DECLARE_WAITQUEUE(wait, current);
 
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index c1a514ffff55..11c3502b07b1 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -103,10 +103,9 @@ xlog_alloc_buffer(
 	 * Pass log block 0 since we don't have an addr yet, buffer will be
 	 * verified on read.
 	 */
-	if (!xlog_verify_bno(log, 0, nbblks)) {
+	if (XFS_IS_CORRUPT(log->l_mp, !xlog_verify_bno(log, 0, nbblks))) {
 		xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer",
 			nbblks);
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
 		return NULL;
 	}
 
@@ -152,11 +151,10 @@ xlog_do_io(
 {
 	int			error;
 
-	if (!xlog_verify_bno(log, blk_no, nbblks)) {
+	if (XFS_IS_CORRUPT(log->l_mp, !xlog_verify_bno(log, blk_no, nbblks))) {
 		xfs_warn(log->l_mp,
 			 "Invalid log block/length (0x%llx, 0x%x) for buffer",
 			 blk_no, nbblks);
-		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
 		return -EFSCORRUPTED;
 	}
 
@@ -244,19 +242,17 @@ xlog_header_check_recover(
 	 * (XLOG_FMT_UNKNOWN). This stops us from trying to recover
 	 * a dirty log created in IRIX.
 	 */
-	if (unlikely(head->h_fmt != cpu_to_be32(XLOG_FMT))) {
+	if (XFS_IS_CORRUPT(mp, head->h_fmt != cpu_to_be32(XLOG_FMT))) {
 		xfs_warn(mp,
 	"dirty log written in incompatible format - can't recover");
 		xlog_header_check_dump(mp, head);
-		XFS_ERROR_REPORT("xlog_header_check_recover(1)",
-				 XFS_ERRLEVEL_HIGH, mp);
 		return -EFSCORRUPTED;
-	} else if (unlikely(!uuid_equal(&mp->m_sb.sb_uuid, &head->h_fs_uuid))) {
+	}
+	if (XFS_IS_CORRUPT(mp, !uuid_equal(&mp->m_sb.sb_uuid,
+					   &head->h_fs_uuid))) {
 		xfs_warn(mp,
 	"dirty log entry has mismatched uuid - can't recover");
 		xlog_header_check_dump(mp, head);
-		XFS_ERROR_REPORT("xlog_header_check_recover(2)",
-				 XFS_ERRLEVEL_HIGH, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
@@ -279,11 +275,10 @@ xlog_header_check_mount(
 		 * by IRIX and continue.
 		 */
 		xfs_warn(mp, "null uuid in log - IRIX style log");
-	} else if (unlikely(!uuid_equal(&mp->m_sb.sb_uuid, &head->h_fs_uuid))) {
+	} else if (XFS_IS_CORRUPT(mp, !uuid_equal(&mp->m_sb.sb_uuid,
+						  &head->h_fs_uuid))) {
 		xfs_warn(mp, "log has mismatched uuid - can't recover");
 		xlog_header_check_dump(mp, head);
-		XFS_ERROR_REPORT("xlog_header_check_mount",
-				 XFS_ERRLEVEL_HIGH, mp);
 		return -EFSCORRUPTED;
 	}
 	return 0;
@@ -299,7 +294,7 @@ xlog_recover_iodone(
 		 * this during recovery. One strike!
 		 */
 		if (!XFS_FORCED_SHUTDOWN(bp->b_mount)) {
-			xfs_buf_ioerror_alert(bp, __func__);
+			xfs_buf_ioerror_alert(bp, __this_address);
 			xfs_force_shutdown(bp->b_mount, SHUTDOWN_META_IO_ERROR);
 		}
 	}
@@ -471,7 +466,7 @@ xlog_find_verify_log_record(
 			xfs_warn(log->l_mp,
 		"Log inconsistent (didn't find previous header)");
 			ASSERT(0);
-			error = -EIO;
+			error = -EFSCORRUPTED;
 			goto out;
 		}
 
@@ -1347,10 +1342,11 @@ xlog_find_tail(
 	error = xlog_rseek_logrec_hdr(log, *head_blk, *head_blk, 1, buffer,
 				      &rhead_blk, &rhead, &wrapped);
 	if (error < 0)
-		return error;
+		goto done;
 	if (!error) {
 		xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
-		return -EIO;
+		error = -EFSCORRUPTED;
+		goto done;
 	}
 	*tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn));
 
@@ -1699,11 +1695,10 @@ xlog_clear_stale_blocks(
 		 * the distance from the beginning of the log to the
 		 * tail.
 		 */
-		if (unlikely(head_block < tail_block || head_block >= log->l_logBBsize)) {
-			XFS_ERROR_REPORT("xlog_clear_stale_blocks(1)",
-					 XFS_ERRLEVEL_LOW, log->l_mp);
+		if (XFS_IS_CORRUPT(log->l_mp,
+				   head_block < tail_block ||
+				   head_block >= log->l_logBBsize))
 			return -EFSCORRUPTED;
-		}
 		tail_distance = tail_block + (log->l_logBBsize - head_block);
 	} else {
 		/*
@@ -1711,11 +1706,10 @@ xlog_clear_stale_blocks(
 		 * so the distance from the head to the tail is just
 		 * the tail block minus the head block.
 		 */
-		if (unlikely(head_block >= tail_block || head_cycle != (tail_cycle + 1))){
-			XFS_ERROR_REPORT("xlog_clear_stale_blocks(2)",
-					 XFS_ERRLEVEL_LOW, log->l_mp);
+		if (XFS_IS_CORRUPT(log->l_mp,
+				   head_block >= tail_block ||
+				   head_cycle != tail_cycle + 1))
 			return -EFSCORRUPTED;
-		}
 		tail_distance = tail_block - head_block;
 	}
 
@@ -1940,6 +1934,12 @@ xlog_recover_buffer_pass1(
 	struct list_head	*bucket;
 	struct xfs_buf_cancel	*bcp;
 
+	if (!xfs_buf_log_check_iovec(&item->ri_buf[0])) {
+		xfs_err(log->l_mp, "bad buffer log item size (%d)",
+				item->ri_buf[0].i_len);
+		return -EFSCORRUPTED;
+	}
+
 	/*
 	 * If this isn't a cancel buffer item, then just return.
 	 */
@@ -2135,13 +2135,11 @@ xlog_recover_do_inode_buffer(
 		 */
 		logged_nextp = item->ri_buf[item_index].i_addr +
 				next_unlinked_offset - reg_buf_offset;
-		if (unlikely(*logged_nextp == 0)) {
+		if (XFS_IS_CORRUPT(mp, *logged_nextp == 0)) {
 			xfs_alert(mp,
 		"Bad inode buffer log record (ptr = "PTR_FMT", bp = "PTR_FMT"). "
 		"Trying to replay bad (0) inode di_next_unlinked field.",
 				item, bp);
-			XFS_ERROR_REPORT("xlog_recover_do_inode_buf",
-					 XFS_ERRLEVEL_LOW, mp);
 			return -EFSCORRUPTED;
 		}
 
@@ -2576,6 +2574,7 @@ xlog_recover_do_reg_buffer(
 	int			bit;
 	int			nbits;
 	xfs_failaddr_t		fa;
+	const size_t		size_disk_dquot = sizeof(struct xfs_disk_dquot);
 
 	trace_xfs_log_recover_buf_reg_buf(mp->m_log, buf_f);
 
@@ -2618,7 +2617,7 @@ xlog_recover_do_reg_buffer(
 					"XFS: NULL dquot in %s.", __func__);
 				goto next;
 			}
-			if (item->ri_buf[i].i_len < sizeof(xfs_disk_dquot_t)) {
+			if (item->ri_buf[i].i_len < size_disk_dquot) {
 				xfs_alert(mp,
 					"XFS: dquot too small (%d) in %s.",
 					item->ri_buf[i].i_len, __func__);
@@ -2746,15 +2745,10 @@ xlog_recover_buffer_pass2(
 	if (buf_f->blf_flags & XFS_BLF_INODE_BUF)
 		buf_flags |= XBF_UNMAPPED;
 
-	bp = xfs_buf_read(mp->m_ddev_targp, buf_f->blf_blkno, buf_f->blf_len,
-			  buf_flags, NULL);
-	if (!bp)
-		return -ENOMEM;
-	error = bp->b_error;
-	if (error) {
-		xfs_buf_ioerror_alert(bp, "xlog_recover_do..(read#1)");
-		goto out_release;
-	}
+	error = xfs_buf_read(mp->m_ddev_targp, buf_f->blf_blkno, buf_f->blf_len,
+			  buf_flags, &bp, NULL);
+	if (error)
+		return error;
 
 	/*
 	 * Recover the buffer only if we get an LSN from it and it's less than
@@ -2875,8 +2869,8 @@ xfs_recover_inode_owner_change(
 		return -ENOMEM;
 
 	/* instantiate the inode */
+	ASSERT(dip->di_version >= 3);
 	xfs_inode_from_disk(ip, dip);
-	ASSERT(ip->i_d.di_version >= 3);
 
 	error = xfs_iformat_fork(ip, dip);
 	if (error)
@@ -2951,17 +2945,10 @@ xlog_recover_inode_pass2(
 	}
 	trace_xfs_log_recover_inode_recover(log, in_f);
 
-	bp = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len, 0,
-			  &xfs_inode_buf_ops);
-	if (!bp) {
-		error = -ENOMEM;
+	error = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len,
+			0, &bp, &xfs_inode_buf_ops);
+	if (error)
 		goto error;
-	}
-	error = bp->b_error;
-	if (error) {
-		xfs_buf_ioerror_alert(bp, "xlog_recover_do..(read#2)");
-		goto out_release;
-	}
 	ASSERT(in_f->ilf_fields & XFS_ILOG_CORE);
 	dip = xfs_buf_offset(bp, in_f->ilf_boffset);
 
@@ -2969,22 +2956,18 @@ xlog_recover_inode_pass2(
 	 * Make sure the place we're flushing out to really looks
 	 * like an inode!
 	 */
-	if (unlikely(!xfs_verify_magic16(bp, dip->di_magic))) {
+	if (XFS_IS_CORRUPT(mp, !xfs_verify_magic16(bp, dip->di_magic))) {
 		xfs_alert(mp,
 	"%s: Bad inode magic number, dip = "PTR_FMT", dino bp = "PTR_FMT", ino = %Ld",
 			__func__, dip, bp, in_f->ilf_ino);
-		XFS_ERROR_REPORT("xlog_recover_inode_pass2(1)",
-				 XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 	ldip = item->ri_buf[1].i_addr;
-	if (unlikely(ldip->di_magic != XFS_DINODE_MAGIC)) {
+	if (XFS_IS_CORRUPT(mp, ldip->di_magic != XFS_DINODE_MAGIC)) {
 		xfs_alert(mp,
 			"%s: Bad inode log record, rec ptr "PTR_FMT", ino %Ld",
 			__func__, item, in_f->ilf_ino);
-		XFS_ERROR_REPORT("xlog_recover_inode_pass2(2)",
-				 XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
@@ -3014,7 +2997,7 @@ xlog_recover_inode_pass2(
 	 * superblock flag to determine whether we need to look at di_flushiter
 	 * to skip replay when the on disk inode is newer than the log one
 	 */
-	if (!xfs_sb_version_hascrc(&mp->m_sb) &&
+	if (!xfs_sb_version_has_v3inode(&mp->m_sb) &&
 	    ldip->di_flushiter < be16_to_cpu(dip->di_flushiter)) {
 		/*
 		 * Deal with the wrap case, DI_MAX_FLUSH is less
@@ -3085,7 +3068,7 @@ xlog_recover_inode_pass2(
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
-	isize = xfs_log_dinode_size(ldip->di_version);
+	isize = xfs_log_dinode_size(mp);
 	if (unlikely(item->ri_buf[1].i_len > isize)) {
 		XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(7)",
 				     XFS_ERRLEVEL_LOW, mp, ldip,
@@ -3166,7 +3149,7 @@ xlog_recover_inode_pass2(
 		default:
 			xfs_warn(log->l_mp, "%s: Invalid flag", __func__);
 			ASSERT(0);
-			error = -EIO;
+			error = -EFSCORRUPTED;
 			goto out_release;
 		}
 	}
@@ -3247,12 +3230,12 @@ xlog_recover_dquot_pass2(
 	recddq = item->ri_buf[1].i_addr;
 	if (recddq == NULL) {
 		xfs_alert(log->l_mp, "NULL dquot in %s.", __func__);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
-	if (item->ri_buf[1].i_len < sizeof(xfs_disk_dquot_t)) {
+	if (item->ri_buf[1].i_len < sizeof(struct xfs_disk_dquot)) {
 		xfs_alert(log->l_mp, "dquot too small (%d) in %s.",
 			item->ri_buf[1].i_len, __func__);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -3279,7 +3262,7 @@ xlog_recover_dquot_pass2(
 	if (fa) {
 		xfs_alert(mp, "corrupt dquot ID 0x%x in log at %pS",
 				dq_f->qlf_id, fa);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 	ASSERT(dq_f->qlf_len == 1);
 
@@ -3537,6 +3520,7 @@ xfs_cui_copy_format(
 		memcpy(dst_cui_fmt, src_cui_fmt, len);
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
@@ -3601,8 +3585,10 @@ xlog_recover_cud_pass2(
 	struct xfs_ail			*ailp = log->l_ailp;
 
 	cud_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_cud_log_format))
+	if (item->ri_buf[0].i_len != sizeof(struct xfs_cud_log_format)) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
+	}
 	cui_id = cud_formatp->cud_cui_id;
 
 	/*
@@ -3654,6 +3640,7 @@ xfs_bui_copy_format(
 		memcpy(dst_bui_fmt, src_bui_fmt, len);
 		return 0;
 	}
+	XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 	return -EFSCORRUPTED;
 }
 
@@ -3677,8 +3664,10 @@ xlog_recover_bui_pass2(
 
 	bui_formatp = item->ri_buf[0].i_addr;
 
-	if (bui_formatp->bui_nextents != XFS_BUI_MAX_FAST_EXTENTS)
+	if (bui_formatp->bui_nextents != XFS_BUI_MAX_FAST_EXTENTS) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
+	}
 	buip = xfs_bui_init(mp);
 	error = xfs_bui_copy_format(&item->ri_buf[0], &buip->bui_format);
 	if (error) {
@@ -3720,8 +3709,10 @@ xlog_recover_bud_pass2(
 	struct xfs_ail			*ailp = log->l_ailp;
 
 	bud_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_bud_log_format))
+	if (item->ri_buf[0].i_len != sizeof(struct xfs_bud_log_format)) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
+	}
 	bui_id = bud_formatp->bud_bui_id;
 
 	/*
@@ -4018,7 +4009,7 @@ xlog_recover_commit_pass1(
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 }
 
@@ -4066,7 +4057,7 @@ xlog_recover_commit_pass2(
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 }
 
@@ -4187,7 +4178,7 @@ xlog_recover_add_to_cont_trans(
 		ASSERT(len <= sizeof(struct xfs_trans_header));
 		if (len > sizeof(struct xfs_trans_header)) {
 			xfs_warn(log->l_mp, "%s: bad header length", __func__);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		xlog_recover_add_item(&trans->r_itemq);
@@ -4243,13 +4234,13 @@ xlog_recover_add_to_trans(
 			xfs_warn(log->l_mp, "%s: bad header magic number",
 				__func__);
 			ASSERT(0);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		if (len > sizeof(struct xfs_trans_header)) {
 			xfs_warn(log->l_mp, "%s: bad header length", __func__);
 			ASSERT(0);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		/*
@@ -4285,7 +4276,7 @@ xlog_recover_add_to_trans(
 				  in_f->ilf_size);
 			ASSERT(0);
 			kmem_free(ptr);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 
 		item->ri_total = in_f->ilf_size;
@@ -4293,7 +4284,16 @@ xlog_recover_add_to_trans(
 			kmem_zalloc(item->ri_total * sizeof(xfs_log_iovec_t),
 				    0);
 	}
-	ASSERT(item->ri_total > item->ri_cnt);
+
+	if (item->ri_total <= item->ri_cnt) {
+		xfs_warn(log->l_mp,
+	"log item region count (%d) overflowed size (%d)",
+				item->ri_cnt, item->ri_total);
+		ASSERT(0);
+		kmem_free(ptr);
+		return -EFSCORRUPTED;
+	}
+
 	/* Description region is ri_buf[0] */
 	item->ri_buf[item->ri_cnt].i_addr = ptr;
 	item->ri_buf[item->ri_cnt].i_len  = len;
@@ -4380,7 +4380,7 @@ xlog_recovery_process_trans(
 	default:
 		xfs_warn(log->l_mp, "%s: bad flag 0x%x", __func__, flags);
 		ASSERT(0);
-		error = -EIO;
+		error = -EFSCORRUPTED;
 		break;
 	}
 	if (error || freeit)
@@ -4460,7 +4460,7 @@ xlog_recover_process_ophdr(
 		xfs_warn(log->l_mp, "%s: bad clientid 0x%x",
 			__func__, ohead->oh_clientid);
 		ASSERT(0);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -4470,7 +4470,7 @@ xlog_recover_process_ophdr(
 	if (dp + len > end) {
 		xfs_warn(log->l_mp, "%s: bad length 0x%x", __func__, len);
 		WARN_ON(1);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	trans = xlog_recover_ophdr_to_trans(rhash, rhead, ohead);
@@ -4947,7 +4947,7 @@ xlog_recover_clear_agi_bucket(
 	if (error)
 		goto out_abort;
 
-	agi = XFS_BUF_TO_AGI(agibp);
+	agi = agibp->b_addr;
 	agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO);
 	offset = offsetof(xfs_agi_t, agi_unlinked) +
 		 (sizeof(xfs_agino_t) * bucket);
@@ -5083,7 +5083,7 @@ xlog_recover_process_iunlinks(
 		 * buffer reference though, so that it stays pinned in memory
 		 * while we need the buffer.
 		 */
-		agi = XFS_BUF_TO_AGI(agibp);
+		agi = agibp->b_addr;
 		xfs_buf_unlock(agibp);
 
 		for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) {
@@ -5172,8 +5172,10 @@ xlog_recover_process(
 		 * If the filesystem is CRC enabled, this mismatch becomes a
 		 * fatal log corruption failure.
 		 */
-		if (xfs_sb_version_hascrc(&log->l_mp->m_sb))
+		if (xfs_sb_version_hascrc(&log->l_mp->m_sb)) {
+			XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 			return -EFSCORRUPTED;
+		}
 	}
 
 	xlog_unpack_data(rhead, dp, log);
@@ -5190,31 +5192,25 @@ xlog_valid_rec_header(
 {
 	int			hlen;
 
-	if (unlikely(rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM))) {
-		XFS_ERROR_REPORT("xlog_valid_rec_header(1)",
-				XFS_ERRLEVEL_LOW, log->l_mp);
+	if (XFS_IS_CORRUPT(log->l_mp,
+			   rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM)))
 		return -EFSCORRUPTED;
-	}
-	if (unlikely(
-	    (!rhead->h_version ||
-	    (be32_to_cpu(rhead->h_version) & (~XLOG_VERSION_OKBITS))))) {
+	if (XFS_IS_CORRUPT(log->l_mp,
+			   (!rhead->h_version ||
+			   (be32_to_cpu(rhead->h_version) &
+			    (~XLOG_VERSION_OKBITS))))) {
 		xfs_warn(log->l_mp, "%s: unrecognised log version (%d).",
 			__func__, be32_to_cpu(rhead->h_version));
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 
 	/* LR body must have data or it wouldn't have been written */
 	hlen = be32_to_cpu(rhead->h_len);
-	if (unlikely( hlen <= 0 || hlen > INT_MAX )) {
-		XFS_ERROR_REPORT("xlog_valid_rec_header(2)",
-				XFS_ERRLEVEL_LOW, log->l_mp);
+	if (XFS_IS_CORRUPT(log->l_mp, hlen <= 0 || hlen > INT_MAX))
 		return -EFSCORRUPTED;
-	}
-	if (unlikely( blkno > log->l_logBBsize || blkno > INT_MAX )) {
-		XFS_ERROR_REPORT("xlog_valid_rec_header(3)",
-				XFS_ERRLEVEL_LOW, log->l_mp);
+	if (XFS_IS_CORRUPT(log->l_mp,
+			   blkno > log->l_logBBsize || blkno > INT_MAX))
 		return -EFSCORRUPTED;
-	}
 	return 0;
 }
 
@@ -5296,8 +5292,12 @@ xlog_do_recovery_pass(
 		"invalid iclog size (%d bytes), using lsunit (%d bytes)",
 					 h_size, log->l_mp->m_logbsize);
 				h_size = log->l_mp->m_logbsize;
-			} else
-				return -EFSCORRUPTED;
+			} else {
+				XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW,
+						log->l_mp);
+				error = -EFSCORRUPTED;
+				goto bread_err1;
+			}
 		}
 
 		if ((be32_to_cpu(rhead->h_version) & XLOG_VERSION_2) &&
@@ -5627,7 +5627,7 @@ xlog_do_recover(
 	error = xfs_buf_submit(bp);
 	if (error) {
 		if (!XFS_FORCED_SHUTDOWN(mp)) {
-			xfs_buf_ioerror_alert(bp, __func__);
+			xfs_buf_ioerror_alert(bp, __this_address);
 			ASSERT(0);
 		}
 		xfs_buf_relse(bp);
@@ -5636,7 +5636,7 @@ xlog_do_recover(
 
 	/* Convert superblock from on-disk format */
 	sbp = &mp->m_sb;
-	xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp));
+	xfs_sb_from_disk(sbp, bp->b_addr);
 	xfs_buf_relse(bp);
 
 	/* re-initialise in-core superblock and geometry structures */
@@ -5809,7 +5809,6 @@ xlog_recover_check_summary(
 	struct xlog	*log)
 {
 	xfs_mount_t	*mp;
-	xfs_agf_t	*agfp;
 	xfs_buf_t	*agfbp;
 	xfs_buf_t	*agibp;
 	xfs_agnumber_t	agno;
@@ -5829,7 +5828,8 @@ xlog_recover_check_summary(
 			xfs_alert(mp, "%s agf read failed agno %d error %d",
 						__func__, agno, error);
 		} else {
-			agfp = XFS_BUF_TO_AGF(agfbp);
+			struct xfs_agf	*agfp = agfbp->b_addr;
+
 			freeblks += be32_to_cpu(agfp->agf_freeblks) +
 				    be32_to_cpu(agfp->agf_flcount);
 			xfs_buf_relse(agfbp);
@@ -5840,7 +5840,7 @@ xlog_recover_check_summary(
 			xfs_alert(mp, "%s agi read failed agno %d error %d",
 						__func__, agno, error);
 		} else {
-			struct xfs_agi	*agi = XFS_BUF_TO_AGI(agibp);
+			struct xfs_agi	*agi = agibp->b_addr;
 
 			itotal += be32_to_cpu(agi->agi_count);
 			ifree += be32_to_cpu(agi->agi_freecount);
diff --git a/fs/xfs/xfs_message.c b/fs/xfs/xfs_message.c
index 9804efe525a9..e0f9d3b6abe9 100644
--- a/fs/xfs/xfs_message.c
+++ b/fs/xfs/xfs_message.c
@@ -20,8 +20,8 @@ __xfs_printk(
 	const struct xfs_mount	*mp,
 	struct va_format	*vaf)
 {
-	if (mp && mp->m_fsname) {
-		printk("%sXFS (%s): %pV\n", level, mp->m_fsname, vaf);
+	if (mp && mp->m_super) {
+		printk("%sXFS (%s): %pV\n", level, mp->m_super->s_id, vaf);
 		return;
 	}
 	printk("%sXFS: %pV\n", level, vaf);
@@ -86,17 +86,25 @@ xfs_alert_tag(
 }
 
 void
-asswarn(char *expr, char *file, int line)
+asswarn(
+	struct xfs_mount	*mp,
+	char			*expr,
+	char			*file,
+	int			line)
 {
-	xfs_warn(NULL, "Assertion failed: %s, file: %s, line: %d",
+	xfs_warn(mp, "Assertion failed: %s, file: %s, line: %d",
 		expr, file, line);
 	WARN_ON(1);
 }
 
 void
-assfail(char *expr, char *file, int line)
+assfail(
+	struct xfs_mount	*mp,
+	char			*expr,
+	char			*file,
+	int			line)
 {
-	xfs_emerg(NULL, "Assertion failed: %s, file: %s, line: %d",
+	xfs_emerg(mp, "Assertion failed: %s, file: %s, line: %d",
 		expr, file, line);
 	if (xfs_globals.bug_on_assert)
 		BUG();
@@ -105,7 +113,7 @@ assfail(char *expr, char *file, int line)
 }
 
 void
-xfs_hex_dump(void *p, int length)
+xfs_hex_dump(const void *p, int length)
 {
 	print_hex_dump(KERN_ALERT, "", DUMP_PREFIX_OFFSET, 16, 1, p, length, 1);
 }
diff --git a/fs/xfs/xfs_message.h b/fs/xfs/xfs_message.h
index 34447dca97d1..0b05e10995a0 100644
--- a/fs/xfs/xfs_message.h
+++ b/fs/xfs/xfs_message.h
@@ -57,9 +57,9 @@ do {									\
 #define xfs_debug_ratelimited(dev, fmt, ...)				\
 	xfs_printk_ratelimited(xfs_debug, dev, fmt, ##__VA_ARGS__)
 
-extern void assfail(char *expr, char *f, int l);
-extern void asswarn(char *expr, char *f, int l);
+void assfail(struct xfs_mount *mp, char *expr, char *f, int l);
+void asswarn(struct xfs_mount *mp, char *expr, char *f, int l);
 
-extern void xfs_hex_dump(void *p, int length);
+extern void xfs_hex_dump(const void *p, int length);
 
 #endif	/* __XFS_MESSAGE_H */
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index ba5b6f3b2b88..c5513e5a226a 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -31,7 +31,7 @@
 #include "xfs_reflink.h"
 #include "xfs_extent_busy.h"
 #include "xfs_health.h"
-
+#include "xfs_trace.h"
 
 static DEFINE_MUTEX(xfs_uuid_table_mutex);
 static int xfs_uuid_table_size;
@@ -310,7 +310,7 @@ reread:
 	/*
 	 * Initialize the mount structure from the superblock.
 	 */
-	xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp));
+	xfs_sb_from_disk(sbp, bp->b_addr);
 
 	/*
 	 * If we haven't validated the superblock, do so now before we try
@@ -360,108 +360,122 @@ release_buf:
 }
 
 /*
- * Update alignment values based on mount options and sb values
+ * If the sunit/swidth change would move the precomputed root inode value, we
+ * must reject the ondisk change because repair will stumble over that.
+ * However, we allow the mount to proceed because we never rejected this
+ * combination before.  Returns true to update the sb, false otherwise.
+ */
+static inline int
+xfs_check_new_dalign(
+	struct xfs_mount	*mp,
+	int			new_dalign,
+	bool			*update_sb)
+{
+	struct xfs_sb		*sbp = &mp->m_sb;
+	xfs_ino_t		calc_ino;
+
+	calc_ino = xfs_ialloc_calc_rootino(mp, new_dalign);
+	trace_xfs_check_new_dalign(mp, new_dalign, calc_ino);
+
+	if (sbp->sb_rootino == calc_ino) {
+		*update_sb = true;
+		return 0;
+	}
+
+	xfs_warn(mp,
+"Cannot change stripe alignment; would require moving root inode.");
+
+	/*
+	 * XXX: Next time we add a new incompat feature, this should start
+	 * returning -EINVAL to fail the mount.  Until then, spit out a warning
+	 * that we're ignoring the administrator's instructions.
+	 */
+	xfs_warn(mp, "Skipping superblock stripe alignment update.");
+	*update_sb = false;
+	return 0;
+}
+
+/*
+ * If we were provided with new sunit/swidth values as mount options, make sure
+ * that they pass basic alignment and superblock feature checks, and convert
+ * them into the same units (FSB) that everything else expects.  This step
+ * /must/ be done before computing the inode geometry.
  */
 STATIC int
-xfs_update_alignment(xfs_mount_t *mp)
+xfs_validate_new_dalign(
+	struct xfs_mount	*mp)
 {
-	xfs_sb_t	*sbp = &(mp->m_sb);
+	if (mp->m_dalign == 0)
+		return 0;
 
-	if (mp->m_dalign) {
+	/*
+	 * If stripe unit and stripe width are not multiples
+	 * of the fs blocksize turn off alignment.
+	 */
+	if ((BBTOB(mp->m_dalign) & mp->m_blockmask) ||
+	    (BBTOB(mp->m_swidth) & mp->m_blockmask)) {
+		xfs_warn(mp,
+	"alignment check failed: sunit/swidth vs. blocksize(%d)",
+			mp->m_sb.sb_blocksize);
+		return -EINVAL;
+	} else {
 		/*
-		 * If stripe unit and stripe width are not multiples
-		 * of the fs blocksize turn off alignment.
+		 * Convert the stripe unit and width to FSBs.
 		 */
-		if ((BBTOB(mp->m_dalign) & mp->m_blockmask) ||
-		    (BBTOB(mp->m_swidth) & mp->m_blockmask)) {
+		mp->m_dalign = XFS_BB_TO_FSBT(mp, mp->m_dalign);
+		if (mp->m_dalign && (mp->m_sb.sb_agblocks % mp->m_dalign)) {
 			xfs_warn(mp,
-		"alignment check failed: sunit/swidth vs. blocksize(%d)",
-				sbp->sb_blocksize);
+		"alignment check failed: sunit/swidth vs. agsize(%d)",
+				 mp->m_sb.sb_agblocks);
 			return -EINVAL;
-		} else {
-			/*
-			 * Convert the stripe unit and width to FSBs.
-			 */
-			mp->m_dalign = XFS_BB_TO_FSBT(mp, mp->m_dalign);
-			if (mp->m_dalign && (sbp->sb_agblocks % mp->m_dalign)) {
-				xfs_warn(mp,
-			"alignment check failed: sunit/swidth vs. agsize(%d)",
-					 sbp->sb_agblocks);
-				return -EINVAL;
-			} else if (mp->m_dalign) {
-				mp->m_swidth = XFS_BB_TO_FSBT(mp, mp->m_swidth);
-			} else {
-				xfs_warn(mp,
-			"alignment check failed: sunit(%d) less than bsize(%d)",
-					 mp->m_dalign, sbp->sb_blocksize);
-				return -EINVAL;
-			}
-		}
-
-		/*
-		 * Update superblock with new values
-		 * and log changes
-		 */
-		if (xfs_sb_version_hasdalign(sbp)) {
-			if (sbp->sb_unit != mp->m_dalign) {
-				sbp->sb_unit = mp->m_dalign;
-				mp->m_update_sb = true;
-			}
-			if (sbp->sb_width != mp->m_swidth) {
-				sbp->sb_width = mp->m_swidth;
-				mp->m_update_sb = true;
-			}
+		} else if (mp->m_dalign) {
+			mp->m_swidth = XFS_BB_TO_FSBT(mp, mp->m_swidth);
 		} else {
 			xfs_warn(mp,
-	"cannot change alignment: superblock does not support data alignment");
+		"alignment check failed: sunit(%d) less than bsize(%d)",
+				 mp->m_dalign, mp->m_sb.sb_blocksize);
 			return -EINVAL;
 		}
-	} else if ((mp->m_flags & XFS_MOUNT_NOALIGN) != XFS_MOUNT_NOALIGN &&
-		    xfs_sb_version_hasdalign(&mp->m_sb)) {
-			mp->m_dalign = sbp->sb_unit;
-			mp->m_swidth = sbp->sb_width;
+	}
+
+	if (!xfs_sb_version_hasdalign(&mp->m_sb)) {
+		xfs_warn(mp,
+"cannot change alignment: superblock does not support data alignment");
+		return -EINVAL;
 	}
 
 	return 0;
 }
 
-/*
- * Set the default minimum read and write sizes unless
- * already specified in a mount option.
- * We use smaller I/O sizes when the file system
- * is being used for NFS service (wsync mount option).
- */
-STATIC void
-xfs_set_rw_sizes(xfs_mount_t *mp)
+/* Update alignment values based on mount options and sb values. */
+STATIC int
+xfs_update_alignment(
+	struct xfs_mount	*mp)
 {
-	xfs_sb_t	*sbp = &(mp->m_sb);
-	int		readio_log, writeio_log;
+	struct xfs_sb		*sbp = &mp->m_sb;
 
-	if (!(mp->m_flags & XFS_MOUNT_DFLT_IOSIZE)) {
-		if (mp->m_flags & XFS_MOUNT_WSYNC) {
-			readio_log = XFS_WSYNC_READIO_LOG;
-			writeio_log = XFS_WSYNC_WRITEIO_LOG;
-		} else {
-			readio_log = XFS_READIO_LOG_LARGE;
-			writeio_log = XFS_WRITEIO_LOG_LARGE;
-		}
-	} else {
-		readio_log = mp->m_readio_log;
-		writeio_log = mp->m_writeio_log;
-	}
+	if (mp->m_dalign) {
+		bool		update_sb;
+		int		error;
 
-	if (sbp->sb_blocklog > readio_log) {
-		mp->m_readio_log = sbp->sb_blocklog;
-	} else {
-		mp->m_readio_log = readio_log;
-	}
-	mp->m_readio_blocks = 1 << (mp->m_readio_log - sbp->sb_blocklog);
-	if (sbp->sb_blocklog > writeio_log) {
-		mp->m_writeio_log = sbp->sb_blocklog;
-	} else {
-		mp->m_writeio_log = writeio_log;
+		if (sbp->sb_unit == mp->m_dalign &&
+		    sbp->sb_width == mp->m_swidth)
+			return 0;
+
+		error = xfs_check_new_dalign(mp, mp->m_dalign, &update_sb);
+		if (error || !update_sb)
+			return error;
+
+		sbp->sb_unit = mp->m_dalign;
+		sbp->sb_width = mp->m_swidth;
+		mp->m_update_sb = true;
+	} else if ((mp->m_flags & XFS_MOUNT_NOALIGN) != XFS_MOUNT_NOALIGN &&
+		    xfs_sb_version_hasdalign(&mp->m_sb)) {
+		mp->m_dalign = sbp->sb_unit;
+		mp->m_swidth = sbp->sb_width;
 	}
-	mp->m_writeio_blocks = 1 << (mp->m_writeio_log - sbp->sb_blocklog);
+
+	return 0;
 }
 
 /*
@@ -687,12 +701,12 @@ xfs_mountfs(
 	}
 
 	/*
-	 * Check if sb_agblocks is aligned at stripe boundary
-	 * If sb_agblocks is NOT aligned turn off m_dalign since
-	 * allocator alignment is within an ag, therefore ag has
-	 * to be aligned at stripe boundary.
+	 * If we were given new sunit/swidth options, do some basic validation
+	 * checks and convert the incore dalign and swidth values to the
+	 * same units (FSB) that everything else uses.  This /must/ happen
+	 * before computing the inode geometry.
 	 */
-	error = xfs_update_alignment(mp);
+	error = xfs_validate_new_dalign(mp);
 	if (error)
 		goto out;
 
@@ -703,10 +717,22 @@ xfs_mountfs(
 	xfs_rmapbt_compute_maxlevels(mp);
 	xfs_refcountbt_compute_maxlevels(mp);
 
+	/*
+	 * Check if sb_agblocks is aligned at stripe boundary.  If sb_agblocks
+	 * is NOT aligned turn off m_dalign since allocator alignment is within
+	 * an ag, therefore ag has to be aligned at stripe boundary.  Note that
+	 * we must compute the free space and rmap btree geometry before doing
+	 * this.
+	 */
+	error = xfs_update_alignment(mp);
+	if (error)
+		goto out;
+
 	/* enable fail_at_unmount as default */
 	mp->m_fail_unmount = true;
 
-	error = xfs_sysfs_init(&mp->m_kobj, &xfs_mp_ktype, NULL, mp->m_fsname);
+	error = xfs_sysfs_init(&mp->m_kobj, &xfs_mp_ktype,
+			       NULL, mp->m_super->s_id);
 	if (error)
 		goto out;
 
@@ -728,9 +754,12 @@ xfs_mountfs(
 		goto out_remove_errortag;
 
 	/*
-	 * Set the minimum read and write sizes
+	 * Update the preferred write size based on the information from the
+	 * on-disk superblock.
 	 */
-	xfs_set_rw_sizes(mp);
+	mp->m_allocsize_log =
+		max_t(uint32_t, sbp->sb_blocklog, mp->m_allocsize_log);
+	mp->m_allocsize_blocks = 1U << (mp->m_allocsize_log - sbp->sb_blocklog);
 
 	/* set the low space thresholds for dynamic preallocation */
 	xfs_set_low_space_thresholds(mp);
@@ -796,9 +825,8 @@ xfs_mountfs(
 		goto out_free_dir;
 	}
 
-	if (!sbp->sb_logblocks) {
+	if (XFS_IS_CORRUPT(mp, !sbp->sb_logblocks)) {
 		xfs_warn(mp, "no log defined");
-		XFS_ERROR_REPORT("xfs_mountfs", XFS_ERRLEVEL_LOW, mp);
 		error = -EFSCORRUPTED;
 		goto out_free_perag;
 	}
@@ -836,12 +864,10 @@ xfs_mountfs(
 
 	ASSERT(rip != NULL);
 
-	if (unlikely(!S_ISDIR(VFS_I(rip)->i_mode))) {
+	if (XFS_IS_CORRUPT(mp, !S_ISDIR(VFS_I(rip)->i_mode))) {
 		xfs_warn(mp, "corrupted root inode %llu: not a directory",
 			(unsigned long long)rip->i_ino);
 		xfs_iunlock(rip, XFS_ILOCK_EXCL);
-		XFS_ERROR_REPORT("xfs_mountfs_int(2)", XFS_ERRLEVEL_LOW,
-				 mp);
 		error = -EFSCORRUPTED;
 		goto out_rele_rip;
 	}
@@ -1277,7 +1303,7 @@ xfs_mod_fdblocks(
 	printk_once(KERN_WARNING
 		"Filesystem \"%s\": reserve blocks depleted! "
 		"Consider increasing reserve pool size.",
-		mp->m_fsname);
+		mp->m_super->s_id);
 fdblocks_enospc:
 	spin_unlock(&mp->m_sb_lock);
 	return -ENOSPC;
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index fdb60e09a9c5..b2e4598fdf7d 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -9,10 +9,8 @@
 struct xlog;
 struct xfs_inode;
 struct xfs_mru_cache;
-struct xfs_nameops;
 struct xfs_ail;
 struct xfs_quotainfo;
-struct xfs_dir_ops;
 struct xfs_da_geometry;
 
 /* dynamic preallocation free space thresholds, 5% down to 1% */
@@ -59,7 +57,6 @@ struct xfs_error_cfg {
 
 typedef struct xfs_mount {
 	struct super_block	*m_super;
-	xfs_tid_t		m_tid;		/* next unused tid for fs */
 
 	/*
 	 * Bitsets of per-fs metadata that have been checked and/or are sick.
@@ -89,8 +86,6 @@ typedef struct xfs_mount {
 	struct percpu_counter	m_delalloc_blks;
 
 	struct xfs_buf		*m_sb_bp;	/* buffer for superblock */
-	char			*m_fsname;	/* filesystem name */
-	int			m_fsname_len;	/* strlen of fs name */
 	char			*m_rtname;	/* realtime device name */
 	char			*m_logname;	/* external log device name */
 	int			m_bsize;	/* fs logical block size */
@@ -98,10 +93,8 @@ typedef struct xfs_mount {
 	xfs_agnumber_t		m_agirotor;	/* last ag dir inode alloced */
 	spinlock_t		m_agirotor_lock;/* .. and lock protecting it */
 	xfs_agnumber_t		m_maxagi;	/* highest inode alloc group */
-	uint			m_readio_log;	/* min read size log bytes */
-	uint			m_readio_blocks; /* min read size blocks */
-	uint			m_writeio_log;	/* min write size log bytes */
-	uint			m_writeio_blocks; /* min write size blocks */
+	uint			m_allocsize_log;/* min write size log bytes */
+	uint			m_allocsize_blocks; /* min write size blocks */
 	struct xfs_da_geometry	*m_dir_geo;	/* directory block geometry */
 	struct xfs_da_geometry	*m_attr_geo;	/* attribute block geometry */
 	struct xlog		*m_log;		/* log specific stuff */
@@ -159,10 +152,6 @@ typedef struct xfs_mount {
 	int			m_dalign;	/* stripe unit */
 	int			m_swidth;	/* stripe width */
 	uint8_t			m_sectbb_log;	/* sectlog - BBSHIFT */
-	const struct xfs_nameops *m_dirnameops;	/* vector of dir name ops */
-	const struct xfs_dir_ops *m_dir_inode_ops; /* vector of dir inode ops */
-	const struct xfs_dir_ops *m_nondir_inode_ops; /* !dir inode ops */
-	uint			m_chsize;	/* size of next field */
 	atomic_t		m_active_trans;	/* number trans frozen */
 	struct xfs_mru_cache	*m_filestream;  /* per-mount filestream data */
 	struct delayed_work	m_reclaim_work;	/* background inode reclaim */
@@ -179,6 +168,11 @@ typedef struct xfs_mount {
 	struct xfs_error_cfg	m_error_cfg[XFS_ERR_CLASS_MAX][XFS_ERR_ERRNO_MAX];
 	struct xstats		m_stats;	/* per-fs stats */
 
+	/*
+	 * Workqueue item so that we can coalesce multiple inode flush attempts
+	 * into a single flush.
+	 */
+	struct work_struct	m_flush_inodes_work;
 	struct workqueue_struct *m_buf_workqueue;
 	struct workqueue_struct	*m_unwritten_workqueue;
 	struct workqueue_struct	*m_cil_workqueue;
@@ -229,7 +223,7 @@ typedef struct xfs_mount {
 #define XFS_MOUNT_ATTR2		(1ULL << 8)	/* allow use of attr2 format */
 #define XFS_MOUNT_GRPID		(1ULL << 9)	/* group-ID assigned from directory */
 #define XFS_MOUNT_NORECOVERY	(1ULL << 10)	/* no recovery - dirty fs */
-#define XFS_MOUNT_DFLT_IOSIZE	(1ULL << 12)	/* set default i/o size */
+#define XFS_MOUNT_ALLOCSIZE	(1ULL << 12)	/* specified allocation size */
 #define XFS_MOUNT_SMALL_INUMS	(1ULL << 14)	/* user wants 32bit inodes */
 #define XFS_MOUNT_32BITINODES	(1ULL << 15)	/* inode32 allocator active */
 #define XFS_MOUNT_NOUUID	(1ULL << 16)	/* ignore uuid during mount */
@@ -238,7 +232,7 @@ typedef struct xfs_mount {
 						 * allocation */
 #define XFS_MOUNT_RDONLY	(1ULL << 20)	/* read-only fs */
 #define XFS_MOUNT_DIRSYNC	(1ULL << 21)	/* synchronous directory ops */
-#define XFS_MOUNT_COMPAT_IOSIZE	(1ULL << 22)	/* don't report large preferred
+#define XFS_MOUNT_LARGEIO	(1ULL << 22)	/* report large preferred
 						 * I/O size in stat() */
 #define XFS_MOUNT_FILESTREAMS	(1ULL << 24)	/* enable the filestreams
 						   allocator */
@@ -246,13 +240,6 @@ typedef struct xfs_mount {
 
 #define XFS_MOUNT_DAX		(1ULL << 62)	/* TEST ONLY! */
 
-
-/*
- * Default minimum read and write sizes.
- */
-#define XFS_READIO_LOG_LARGE	16
-#define XFS_WRITEIO_LOG_LARGE	16
-
 /*
  * Max and min values for mount-option defined I/O
  * preallocation sizes.
@@ -260,37 +247,6 @@ typedef struct xfs_mount {
 #define XFS_MAX_IO_LOG		30	/* 1G */
 #define XFS_MIN_IO_LOG		PAGE_SHIFT
 
-/*
- * Synchronous read and write sizes.  This should be
- * better for NFSv2 wsync filesystems.
- */
-#define	XFS_WSYNC_READIO_LOG	15	/* 32k */
-#define	XFS_WSYNC_WRITEIO_LOG	14	/* 16k */
-
-/*
- * Allow large block sizes to be reported to userspace programs if the
- * "largeio" mount option is used.
- *
- * If compatibility mode is specified, simply return the basic unit of caching
- * so that we don't get inefficient read/modify/write I/O from user apps.
- * Otherwise....
- *
- * If the underlying volume is a stripe, then return the stripe width in bytes
- * as the recommended I/O size. It is not a stripe and we've set a default
- * buffered I/O size, return that, otherwise return the compat default.
- */
-static inline unsigned long
-xfs_preferred_iosize(xfs_mount_t *mp)
-{
-	if (mp->m_flags & XFS_MOUNT_COMPAT_IOSIZE)
-		return PAGE_SIZE;
-	return (mp->m_swidth ?
-		(mp->m_swidth << mp->m_sb.sb_blocklog) :
-		((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ?
-			(1 << (int)max(mp->m_readio_log, mp->m_writeio_log)) :
-			PAGE_SIZE));
-}
-
 #define XFS_LAST_UNMOUNT_WAS_CLEAN(mp)	\
 				((mp)->m_flags & XFS_MOUNT_WAS_CLEAN)
 #define XFS_FORCED_SHUTDOWN(mp)	((mp)->m_flags & XFS_MOUNT_FS_SHUTDOWN)
diff --git a/fs/xfs/xfs_ondisk.h b/fs/xfs/xfs_ondisk.h
index b6701b4f59a9..5f04d8a5ab2a 100644
--- a/fs/xfs/xfs_ondisk.h
+++ b/fs/xfs/xfs_ondisk.h
@@ -111,6 +111,7 @@ xfs_check_ondisk_structs(void)
 	XFS_CHECK_STRUCT_SIZE(xfs_dir2_sf_hdr_t,		10);
 
 	/* log structures */
+	XFS_CHECK_STRUCT_SIZE(struct xfs_buf_log_format,	88);
 	XFS_CHECK_STRUCT_SIZE(struct xfs_dq_logformat,		24);
 	XFS_CHECK_STRUCT_SIZE(struct xfs_efd_log_format_32,	28);
 	XFS_CHECK_STRUCT_SIZE(struct xfs_efd_log_format_64,	32);
diff --git a/fs/xfs/xfs_pnfs.c b/fs/xfs/xfs_pnfs.c
index a339bd5fa260..bb3008d390aa 100644
--- a/fs/xfs/xfs_pnfs.c
+++ b/fs/xfs/xfs_pnfs.c
@@ -12,6 +12,7 @@
 #include "xfs_trans.h"
 #include "xfs_bmap.h"
 #include "xfs_iomap.h"
+#include "xfs_pnfs.h"
 
 /*
  * Ensure that we do not have any outstanding pNFS layouts that can be used by
@@ -59,7 +60,7 @@ xfs_fs_get_uuid(
 
 	printk_once(KERN_NOTICE
 "XFS (%s): using experimental pNFS feature, use at your own risk!\n",
-		mp->m_fsname);
+		mp->m_super->s_id);
 
 	if (*len < sizeof(uuid_t))
 		return -EINVAL;
@@ -142,43 +143,38 @@ xfs_fs_map_blocks(
 	lock_flags = xfs_ilock_data_map_shared(ip);
 	error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb,
 				&imap, &nimaps, bmapi_flags);
-	xfs_iunlock(ip, lock_flags);
 
-	if (error)
-		goto out_unlock;
+	ASSERT(!nimaps || imap.br_startblock != DELAYSTARTBLOCK);
+
+	if (!error && write &&
+	    (!nimaps || imap.br_startblock == HOLESTARTBLOCK)) {
+		if (offset + length > XFS_ISIZE(ip))
+			end_fsb = xfs_iomap_eof_align_last_fsb(ip, end_fsb);
+		else if (nimaps && imap.br_startblock == HOLESTARTBLOCK)
+			end_fsb = min(end_fsb, imap.br_startoff +
+					       imap.br_blockcount);
+		xfs_iunlock(ip, lock_flags);
+
+		error = xfs_iomap_write_direct(ip, offset_fsb,
+				end_fsb - offset_fsb, &imap);
+		if (error)
+			goto out_unlock;
 
-	if (write) {
-		enum xfs_prealloc_flags	flags = 0;
-
-		ASSERT(imap.br_startblock != DELAYSTARTBLOCK);
-
-		if (!nimaps || imap.br_startblock == HOLESTARTBLOCK) {
-			/*
-			 * xfs_iomap_write_direct() expects to take ownership of
-			 * the shared ilock.
-			 */
-			xfs_ilock(ip, XFS_ILOCK_SHARED);
-			error = xfs_iomap_write_direct(ip, offset, length,
-						       &imap, nimaps);
-			if (error)
-				goto out_unlock;
-
-			/*
-			 * Ensure the next transaction is committed
-			 * synchronously so that the blocks allocated and
-			 * handed out to the client are guaranteed to be
-			 * present even after a server crash.
-			 */
-			flags |= XFS_PREALLOC_SET | XFS_PREALLOC_SYNC;
-		}
-
-		error = xfs_update_prealloc_flags(ip, flags);
+		/*
+		 * Ensure the next transaction is committed synchronously so
+		 * that the blocks allocated and handed out to the client are
+		 * guaranteed to be present even after a server crash.
+		 */
+		error = xfs_update_prealloc_flags(ip,
+				XFS_PREALLOC_SET | XFS_PREALLOC_SYNC);
 		if (error)
 			goto out_unlock;
+	} else {
+		xfs_iunlock(ip, lock_flags);
 	}
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 
-	error = xfs_bmbt_to_iomap(ip, iomap, &imap, false);
+	error = xfs_bmbt_to_iomap(ip, iomap, &imap, 0);
 	*device_generation = mp->m_generation;
 	return error;
 out_unlock:
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index ecd8ce152ab1..c225691fad15 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -22,6 +22,7 @@
 #include "xfs_qm.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
+#include "xfs_error.h"
 
 /*
  * The global quota manager. There is only one of these for the entire
@@ -29,10 +30,10 @@
  * quota functionality, including maintaining the freelist and hash
  * tables of dquots.
  */
-STATIC int	xfs_qm_init_quotainos(xfs_mount_t *);
-STATIC int	xfs_qm_init_quotainfo(xfs_mount_t *);
+STATIC int	xfs_qm_init_quotainos(struct xfs_mount *mp);
+STATIC int	xfs_qm_init_quotainfo(struct xfs_mount *mp);
 
-STATIC void	xfs_qm_destroy_quotainos(xfs_quotainfo_t *qi);
+STATIC void	xfs_qm_destroy_quotainos(struct xfs_quotainfo *qi);
 STATIC void	xfs_qm_dqfree_one(struct xfs_dquot *dqp);
 /*
  * We use the batch lookup interface to iterate over the dquots as it
@@ -120,12 +121,11 @@ xfs_qm_dqpurge(
 {
 	struct xfs_mount	*mp = dqp->q_mount;
 	struct xfs_quotainfo	*qi = mp->m_quotainfo;
+	int			error = -EAGAIN;
 
 	xfs_dqlock(dqp);
-	if ((dqp->dq_flags & XFS_DQ_FREEING) || dqp->q_nrefs != 0) {
-		xfs_dqunlock(dqp);
-		return -EAGAIN;
-	}
+	if ((dqp->dq_flags & XFS_DQ_FREEING) || dqp->q_nrefs != 0)
+		goto out_unlock;
 
 	dqp->dq_flags |= XFS_DQ_FREEING;
 
@@ -138,7 +138,6 @@ xfs_qm_dqpurge(
 	 */
 	if (XFS_DQ_IS_DIRTY(dqp)) {
 		struct xfs_buf	*bp = NULL;
-		int		error;
 
 		/*
 		 * We don't care about getting disk errors here. We need
@@ -148,6 +147,8 @@ xfs_qm_dqpurge(
 		if (!error) {
 			error = xfs_bwrite(bp);
 			xfs_buf_relse(bp);
+		} else if (error == -EAGAIN) {
+			goto out_unlock;
 		}
 		xfs_dqflock(dqp);
 	}
@@ -173,6 +174,10 @@ xfs_qm_dqpurge(
 
 	xfs_qm_dqdestroy(dqp);
 	return 0;
+
+out_unlock:
+	xfs_dqunlock(dqp);
+	return error;
 }
 
 /*
@@ -243,14 +248,14 @@ xfs_qm_unmount_quotas(
 
 STATIC int
 xfs_qm_dqattach_one(
-	xfs_inode_t	*ip,
-	xfs_dqid_t	id,
-	uint		type,
-	bool		doalloc,
-	xfs_dquot_t	**IO_idqpp)
+	struct xfs_inode	*ip,
+	xfs_dqid_t		id,
+	uint			type,
+	bool			doalloc,
+	struct xfs_dquot	**IO_idqpp)
 {
-	xfs_dquot_t	*dqp;
-	int		error;
+	struct xfs_dquot	*dqp;
+	int			error;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	error = 0;
@@ -325,23 +330,23 @@ xfs_qm_dqattach_locked(
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	if (XFS_IS_UQUOTA_ON(mp) && !ip->i_udquot) {
-		error = xfs_qm_dqattach_one(ip, ip->i_d.di_uid, XFS_DQ_USER,
-				doalloc, &ip->i_udquot);
+		error = xfs_qm_dqattach_one(ip, i_uid_read(VFS_I(ip)),
+				XFS_DQ_USER, doalloc, &ip->i_udquot);
 		if (error)
 			goto done;
 		ASSERT(ip->i_udquot);
 	}
 
 	if (XFS_IS_GQUOTA_ON(mp) && !ip->i_gdquot) {
-		error = xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP,
-				doalloc, &ip->i_gdquot);
+		error = xfs_qm_dqattach_one(ip, i_gid_read(VFS_I(ip)),
+				XFS_DQ_GROUP, doalloc, &ip->i_gdquot);
 		if (error)
 			goto done;
 		ASSERT(ip->i_gdquot);
 	}
 
 	if (XFS_IS_PQUOTA_ON(mp) && !ip->i_pdquot) {
-		error = xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ,
+		error = xfs_qm_dqattach_one(ip, ip->i_d.di_projid, XFS_DQ_PROJ,
 				doalloc, &ip->i_pdquot);
 		if (error)
 			goto done;
@@ -539,12 +544,12 @@ xfs_qm_shrink_count(
 
 STATIC void
 xfs_qm_set_defquota(
-	xfs_mount_t	*mp,
-	uint		type,
-	xfs_quotainfo_t	*qinf)
+	struct xfs_mount	*mp,
+	uint			type,
+	struct xfs_quotainfo	*qinf)
 {
-	xfs_dquot_t		*dqp;
-	struct xfs_def_quota    *defq;
+	struct xfs_dquot	*dqp;
+	struct xfs_def_quota	*defq;
 	struct xfs_disk_dquot	*ddqp;
 	int			error;
 
@@ -642,7 +647,7 @@ xfs_qm_init_quotainfo(
 
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 
-	qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), 0);
+	qinf = mp->m_quotainfo = kmem_zalloc(sizeof(struct xfs_quotainfo), 0);
 
 	error = list_lru_init(&qinf->qi_lru);
 	if (error)
@@ -709,9 +714,9 @@ out_free_qinf:
  */
 void
 xfs_qm_destroy_quotainfo(
-	xfs_mount_t	*mp)
+	struct xfs_mount	*mp)
 {
-	xfs_quotainfo_t *qi;
+	struct xfs_quotainfo	*qi;
 
 	qi = mp->m_quotainfo;
 	ASSERT(qi != NULL);
@@ -754,11 +759,15 @@ xfs_qm_qino_alloc(
 		if ((flags & XFS_QMOPT_PQUOTA) &&
 			     (mp->m_sb.sb_gquotino != NULLFSINO)) {
 			ino = mp->m_sb.sb_gquotino;
-			ASSERT(mp->m_sb.sb_pquotino == NULLFSINO);
+			if (XFS_IS_CORRUPT(mp,
+					   mp->m_sb.sb_pquotino != NULLFSINO))
+				return -EFSCORRUPTED;
 		} else if ((flags & XFS_QMOPT_GQUOTA) &&
 			     (mp->m_sb.sb_pquotino != NULLFSINO)) {
 			ino = mp->m_sb.sb_pquotino;
-			ASSERT(mp->m_sb.sb_gquotino == NULLFSINO);
+			if (XFS_IS_CORRUPT(mp,
+					   mp->m_sb.sb_gquotino != NULLFSINO))
+				return -EFSCORRUPTED;
 		}
 		if (ino != NULLFSINO) {
 			error = xfs_iget(mp, NULL, ino, 0, 0, ip);
@@ -866,12 +875,20 @@ xfs_qm_reset_dqcounts(
 		ddq->d_bcount = 0;
 		ddq->d_icount = 0;
 		ddq->d_rtbcount = 0;
-		ddq->d_btimer = 0;
-		ddq->d_itimer = 0;
-		ddq->d_rtbtimer = 0;
-		ddq->d_bwarns = 0;
-		ddq->d_iwarns = 0;
-		ddq->d_rtbwarns = 0;
+
+		/*
+		 * dquot id 0 stores the default grace period and the maximum
+		 * warning limit that were set by the administrator, so we
+		 * should not reset them.
+		 */
+		if (ddq->d_id != 0) {
+			ddq->d_btimer = 0;
+			ddq->d_itimer = 0;
+			ddq->d_rtbtimer = 0;
+			ddq->d_bwarns = 0;
+			ddq->d_iwarns = 0;
+			ddq->d_rtbwarns = 0;
+		}
 
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {
 			xfs_update_cksum((char *)&dqb[j],
@@ -1559,7 +1576,7 @@ error_rele:
 
 STATIC void
 xfs_qm_destroy_quotainos(
-	xfs_quotainfo_t	*qi)
+	struct xfs_quotainfo	*qi)
 {
 	if (qi->qi_uquotaip) {
 		xfs_irele(qi->qi_uquotaip);
@@ -1608,8 +1625,8 @@ xfs_qm_dqfree_one(
 int
 xfs_qm_vop_dqalloc(
 	struct xfs_inode	*ip,
-	xfs_dqid_t		uid,
-	xfs_dqid_t		gid,
+	kuid_t			uid,
+	kgid_t			gid,
 	prid_t			prid,
 	uint			flags,
 	struct xfs_dquot	**O_udqpp,
@@ -1617,6 +1634,8 @@ xfs_qm_vop_dqalloc(
 	struct xfs_dquot	**O_pdqpp)
 {
 	struct xfs_mount	*mp = ip->i_mount;
+	struct inode		*inode = VFS_I(ip);
+	struct user_namespace	*user_ns = inode->i_sb->s_user_ns;
 	struct xfs_dquot	*uq = NULL;
 	struct xfs_dquot	*gq = NULL;
 	struct xfs_dquot	*pq = NULL;
@@ -1630,7 +1649,7 @@ xfs_qm_vop_dqalloc(
 	xfs_ilock(ip, lockflags);
 
 	if ((flags & XFS_QMOPT_INHERIT) && XFS_INHERIT_GID(ip))
-		gid = ip->i_d.di_gid;
+		gid = inode->i_gid;
 
 	/*
 	 * Attach the dquot(s) to this inode, doing a dquot allocation
@@ -1645,7 +1664,7 @@ xfs_qm_vop_dqalloc(
 	}
 
 	if ((flags & XFS_QMOPT_UQUOTA) && XFS_IS_UQUOTA_ON(mp)) {
-		if (ip->i_d.di_uid != uid) {
+		if (!uid_eq(inode->i_uid, uid)) {
 			/*
 			 * What we need is the dquot that has this uid, and
 			 * if we send the inode to dqget, the uid of the inode
@@ -1656,7 +1675,8 @@ xfs_qm_vop_dqalloc(
 			 * holding ilock.
 			 */
 			xfs_iunlock(ip, lockflags);
-			error = xfs_qm_dqget(mp, uid, XFS_DQ_USER, true, &uq);
+			error = xfs_qm_dqget(mp, from_kuid(user_ns, uid),
+					XFS_DQ_USER, true, &uq);
 			if (error) {
 				ASSERT(error != -ENOENT);
 				return error;
@@ -1677,9 +1697,10 @@ xfs_qm_vop_dqalloc(
 		}
 	}
 	if ((flags & XFS_QMOPT_GQUOTA) && XFS_IS_GQUOTA_ON(mp)) {
-		if (ip->i_d.di_gid != gid) {
+		if (!gid_eq(inode->i_gid, gid)) {
 			xfs_iunlock(ip, lockflags);
-			error = xfs_qm_dqget(mp, gid, XFS_DQ_GROUP, true, &gq);
+			error = xfs_qm_dqget(mp, from_kgid(user_ns, gid),
+					XFS_DQ_GROUP, true, &gq);
 			if (error) {
 				ASSERT(error != -ENOENT);
 				goto error_rele;
@@ -1693,7 +1714,7 @@ xfs_qm_vop_dqalloc(
 		}
 	}
 	if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) {
-		if (xfs_get_projid(ip) != prid) {
+		if (ip->i_d.di_projid != prid) {
 			xfs_iunlock(ip, lockflags);
 			error = xfs_qm_dqget(mp, (xfs_dqid_t)prid, XFS_DQ_PROJ,
 					true, &pq);
@@ -1737,14 +1758,14 @@ error_rele:
  * Actually transfer ownership, and do dquot modifications.
  * These were already reserved.
  */
-xfs_dquot_t *
+struct xfs_dquot *
 xfs_qm_vop_chown(
-	xfs_trans_t	*tp,
-	xfs_inode_t	*ip,
-	xfs_dquot_t	**IO_olddq,
-	xfs_dquot_t	*newdq)
+	struct xfs_trans	*tp,
+	struct xfs_inode	*ip,
+	struct xfs_dquot	**IO_olddq,
+	struct xfs_dquot	*newdq)
 {
-	xfs_dquot_t	*prevdq;
+	struct xfs_dquot	*prevdq;
 	uint		bfield = XFS_IS_REALTIME_INODE(ip) ?
 				 XFS_TRANS_DQ_RTBCOUNT : XFS_TRANS_DQ_BCOUNT;
 
@@ -1805,7 +1826,7 @@ xfs_qm_vop_chown_reserve(
 			XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS;
 
 	if (XFS_IS_UQUOTA_ON(mp) && udqp &&
-	    ip->i_d.di_uid != be32_to_cpu(udqp->q_core.d_id)) {
+	    i_uid_read(VFS_I(ip)) != be32_to_cpu(udqp->q_core.d_id)) {
 		udq_delblks = udqp;
 		/*
 		 * If there are delayed allocation blocks, then we have to
@@ -1818,7 +1839,7 @@ xfs_qm_vop_chown_reserve(
 		}
 	}
 	if (XFS_IS_GQUOTA_ON(ip->i_mount) && gdqp &&
-	    ip->i_d.di_gid != be32_to_cpu(gdqp->q_core.d_id)) {
+	    i_gid_read(VFS_I(ip)) != be32_to_cpu(gdqp->q_core.d_id)) {
 		gdq_delblks = gdqp;
 		if (delblks) {
 			ASSERT(ip->i_gdquot);
@@ -1827,7 +1848,7 @@ xfs_qm_vop_chown_reserve(
 	}
 
 	if (XFS_IS_PQUOTA_ON(ip->i_mount) && pdqp &&
-	    xfs_get_projid(ip) != be32_to_cpu(pdqp->q_core.d_id)) {
+	    ip->i_d.di_projid != be32_to_cpu(pdqp->q_core.d_id)) {
 		prjflags = XFS_QMOPT_ENOSPC;
 		pdq_delblks = pdqp;
 		if (delblks) {
@@ -1915,20 +1936,21 @@ xfs_qm_vop_create_dqattach(
 
 	if (udqp && XFS_IS_UQUOTA_ON(mp)) {
 		ASSERT(ip->i_udquot == NULL);
-		ASSERT(ip->i_d.di_uid == be32_to_cpu(udqp->q_core.d_id));
+		ASSERT(i_uid_read(VFS_I(ip)) == be32_to_cpu(udqp->q_core.d_id));
 
 		ip->i_udquot = xfs_qm_dqhold(udqp);
 		xfs_trans_mod_dquot(tp, udqp, XFS_TRANS_DQ_ICOUNT, 1);
 	}
 	if (gdqp && XFS_IS_GQUOTA_ON(mp)) {
 		ASSERT(ip->i_gdquot == NULL);
-		ASSERT(ip->i_d.di_gid == be32_to_cpu(gdqp->q_core.d_id));
+		ASSERT(i_gid_read(VFS_I(ip)) == be32_to_cpu(gdqp->q_core.d_id));
+
 		ip->i_gdquot = xfs_qm_dqhold(gdqp);
 		xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1);
 	}
 	if (pdqp && XFS_IS_PQUOTA_ON(mp)) {
 		ASSERT(ip->i_pdquot == NULL);
-		ASSERT(xfs_get_projid(ip) == be32_to_cpu(pdqp->q_core.d_id));
+		ASSERT(ip->i_d.di_projid == be32_to_cpu(pdqp->q_core.d_id));
 
 		ip->i_pdquot = xfs_qm_dqhold(pdqp);
 		xfs_trans_mod_dquot(tp, pdqp, XFS_TRANS_DQ_ICOUNT, 1);
diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h
index b41b75089548..4e57edca8bce 100644
--- a/fs/xfs/xfs_qm.h
+++ b/fs/xfs/xfs_qm.h
@@ -54,7 +54,7 @@ struct xfs_def_quota {
  * Various quota information for individual filesystems.
  * The mount structure keeps a pointer to this.
  */
-typedef struct xfs_quotainfo {
+struct xfs_quotainfo {
 	struct radix_tree_root qi_uquota_tree;
 	struct radix_tree_root qi_gquota_tree;
 	struct radix_tree_root qi_pquota_tree;
@@ -64,9 +64,9 @@ typedef struct xfs_quotainfo {
 	struct xfs_inode	*qi_pquotaip;	/* project quota inode */
 	struct list_lru	 qi_lru;
 	int		 qi_dquots;
-	time_t		 qi_btimelimit;	 /* limit for blks timer */
-	time_t		 qi_itimelimit;	 /* limit for inodes timer */
-	time_t		 qi_rtbtimelimit;/* limit for rt blks timer */
+	time64_t	 qi_btimelimit;	 /* limit for blks timer */
+	time64_t	 qi_itimelimit;	 /* limit for inodes timer */
+	time64_t	 qi_rtbtimelimit;/* limit for rt blks timer */
 	xfs_qwarncnt_t	 qi_bwarnlimit;	 /* limit for blks warnings */
 	xfs_qwarncnt_t	 qi_iwarnlimit;	 /* limit for inodes warnings */
 	xfs_qwarncnt_t	 qi_rtbwarnlimit;/* limit for rt blks warnings */
@@ -76,8 +76,8 @@ typedef struct xfs_quotainfo {
 	struct xfs_def_quota	qi_usr_default;
 	struct xfs_def_quota	qi_grp_default;
 	struct xfs_def_quota	qi_prj_default;
-	struct shrinker  qi_shrinker;
-} xfs_quotainfo_t;
+	struct shrinker	qi_shrinker;
+};
 
 static inline struct radix_tree_root *
 xfs_dquot_tree(
diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c
index 5d72e88598b4..fc2fa418919f 100644
--- a/fs/xfs/xfs_qm_bhv.c
+++ b/fs/xfs/xfs_qm_bhv.c
@@ -54,13 +54,13 @@ xfs_fill_statvfs_from_dquot(
  */
 void
 xfs_qm_statvfs(
-	xfs_inode_t		*ip,
+	struct xfs_inode	*ip,
 	struct kstatfs		*statp)
 {
-	xfs_mount_t		*mp = ip->i_mount;
-	xfs_dquot_t		*dqp;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_dquot	*dqp;
 
-	if (!xfs_qm_dqget(mp, xfs_get_projid(ip), XFS_DQ_PROJ, false, &dqp)) {
+	if (!xfs_qm_dqget(mp, ip->i_d.di_projid, XFS_DQ_PROJ, false, &dqp)) {
 		xfs_fill_statvfs_from_dquot(statp, dqp);
 		xfs_qm_dqput(dqp);
 	}
diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c
index da7ad0383037..5d5ac65aa1cc 100644
--- a/fs/xfs/xfs_qm_syscalls.c
+++ b/fs/xfs/xfs_qm_syscalls.c
@@ -19,9 +19,71 @@
 #include "xfs_qm.h"
 #include "xfs_icache.h"
 
-STATIC int	xfs_qm_log_quotaoff(xfs_mount_t *, xfs_qoff_logitem_t **, uint);
-STATIC int	xfs_qm_log_quotaoff_end(xfs_mount_t *, xfs_qoff_logitem_t *,
-					uint);
+STATIC int
+xfs_qm_log_quotaoff(
+	struct xfs_mount	*mp,
+	struct xfs_qoff_logitem	**qoffstartp,
+	uint			flags)
+{
+	struct xfs_trans	*tp;
+	int			error;
+	struct xfs_qoff_logitem	*qoffi;
+
+	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_quotaoff, 0, 0, 0, &tp);
+	if (error)
+		goto out;
+
+	qoffi = xfs_trans_get_qoff_item(tp, NULL, flags & XFS_ALL_QUOTA_ACCT);
+	xfs_trans_log_quotaoff_item(tp, qoffi);
+
+	spin_lock(&mp->m_sb_lock);
+	mp->m_sb.sb_qflags = (mp->m_qflags & ~(flags)) & XFS_MOUNT_QUOTA_ALL;
+	spin_unlock(&mp->m_sb_lock);
+
+	xfs_log_sb(tp);
+
+	/*
+	 * We have to make sure that the transaction is secure on disk before we
+	 * return and actually stop quota accounting. So, make it synchronous.
+	 * We don't care about quotoff's performance.
+	 */
+	xfs_trans_set_sync(tp);
+	error = xfs_trans_commit(tp);
+	if (error)
+		goto out;
+
+	*qoffstartp = qoffi;
+out:
+	return error;
+}
+
+STATIC int
+xfs_qm_log_quotaoff_end(
+	struct xfs_mount	*mp,
+	struct xfs_qoff_logitem	**startqoff,
+	uint			flags)
+{
+	struct xfs_trans	*tp;
+	int			error;
+	struct xfs_qoff_logitem	*qoffi;
+
+	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_equotaoff, 0, 0, 0, &tp);
+	if (error)
+		return error;
+
+	qoffi = xfs_trans_get_qoff_item(tp, *startqoff,
+					flags & XFS_ALL_QUOTA_ACCT);
+	xfs_trans_log_quotaoff_item(tp, qoffi);
+	*startqoff = NULL;
+
+	/*
+	 * We have to make sure that the transaction is secure on disk before we
+	 * return and actually stop quota accounting. So, make it synchronous.
+	 * We don't care about quotoff's performance.
+	 */
+	xfs_trans_set_sync(tp);
+	return xfs_trans_commit(tp);
+}
 
 /*
  * Turn off quota accounting and/or enforcement for all udquots and/or
@@ -40,7 +102,7 @@ xfs_qm_scall_quotaoff(
 	uint			dqtype;
 	int			error;
 	uint			inactivate_flags;
-	xfs_qoff_logitem_t	*qoffstart;
+	struct xfs_qoff_logitem	*qoffstart = NULL;
 
 	/*
 	 * No file system can have quotas enabled on disk but not in core.
@@ -165,7 +227,7 @@ xfs_qm_scall_quotaoff(
 	 * So, we have QUOTAOFF start and end logitems; the start
 	 * logitem won't get overwritten until the end logitem appears...
 	 */
-	error = xfs_qm_log_quotaoff_end(mp, qoffstart, flags);
+	error = xfs_qm_log_quotaoff_end(mp, &qoffstart, flags);
 	if (error) {
 		/* We're screwed now. Shutdown is the only option. */
 		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
@@ -198,6 +260,8 @@ xfs_qm_scall_quotaoff(
 	}
 
 out_unlock:
+	if (error && qoffstart)
+		xfs_qm_qoff_logitem_relse(qoffstart);
 	mutex_unlock(&q->qi_quotaofflock);
 	return error;
 }
@@ -538,74 +602,6 @@ out_unlock:
 	return error;
 }
 
-STATIC int
-xfs_qm_log_quotaoff_end(
-	xfs_mount_t		*mp,
-	xfs_qoff_logitem_t	*startqoff,
-	uint			flags)
-{
-	xfs_trans_t		*tp;
-	int			error;
-	xfs_qoff_logitem_t	*qoffi;
-
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_equotaoff, 0, 0, 0, &tp);
-	if (error)
-		return error;
-
-	qoffi = xfs_trans_get_qoff_item(tp, startqoff,
-					flags & XFS_ALL_QUOTA_ACCT);
-	xfs_trans_log_quotaoff_item(tp, qoffi);
-
-	/*
-	 * We have to make sure that the transaction is secure on disk before we
-	 * return and actually stop quota accounting. So, make it synchronous.
-	 * We don't care about quotoff's performance.
-	 */
-	xfs_trans_set_sync(tp);
-	return xfs_trans_commit(tp);
-}
-
-
-STATIC int
-xfs_qm_log_quotaoff(
-	xfs_mount_t	       *mp,
-	xfs_qoff_logitem_t     **qoffstartp,
-	uint		       flags)
-{
-	xfs_trans_t	       *tp;
-	int			error;
-	xfs_qoff_logitem_t     *qoffi;
-
-	*qoffstartp = NULL;
-
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_quotaoff, 0, 0, 0, &tp);
-	if (error)
-		goto out;
-
-	qoffi = xfs_trans_get_qoff_item(tp, NULL, flags & XFS_ALL_QUOTA_ACCT);
-	xfs_trans_log_quotaoff_item(tp, qoffi);
-
-	spin_lock(&mp->m_sb_lock);
-	mp->m_sb.sb_qflags = (mp->m_qflags & ~(flags)) & XFS_MOUNT_QUOTA_ALL;
-	spin_unlock(&mp->m_sb_lock);
-
-	xfs_log_sb(tp);
-
-	/*
-	 * We have to make sure that the transaction is secure on disk before we
-	 * return and actually stop quota accounting. So, make it synchronous.
-	 * We don't care about quotoff's performance.
-	 */
-	xfs_trans_set_sync(tp);
-	error = xfs_trans_commit(tp);
-	if (error)
-		goto out;
-
-	*qoffstartp = qoffi;
-out:
-	return error;
-}
-
 /* Fill out the quota context. */
 static void
 xfs_qm_scall_getquota_fill_qc(
diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h
index efe42ae7a2f3..aa8fc1f55fbd 100644
--- a/fs/xfs/xfs_quota.h
+++ b/fs/xfs/xfs_quota.h
@@ -86,7 +86,7 @@ extern int xfs_trans_reserve_quota_bydquots(struct xfs_trans *,
 		struct xfs_mount *, struct xfs_dquot *,
 		struct xfs_dquot *, struct xfs_dquot *, int64_t, long, uint);
 
-extern int xfs_qm_vop_dqalloc(struct xfs_inode *, xfs_dqid_t, xfs_dqid_t,
+extern int xfs_qm_vop_dqalloc(struct xfs_inode *, kuid_t, kgid_t,
 		prid_t, uint, struct xfs_dquot **, struct xfs_dquot **,
 		struct xfs_dquot **);
 extern void xfs_qm_vop_create_dqattach(struct xfs_trans *, struct xfs_inode *,
@@ -109,7 +109,7 @@ extern void xfs_qm_unmount_quotas(struct xfs_mount *);
 
 #else
 static inline int
-xfs_qm_vop_dqalloc(struct xfs_inode *ip, xfs_dqid_t uid, xfs_dqid_t gid,
+xfs_qm_vop_dqalloc(struct xfs_inode *ip, kuid_t kuid, kgid_t kgid,
 		prid_t prid, uint flags, struct xfs_dquot **udqp,
 		struct xfs_dquot **gdqp, struct xfs_dquot **pdqp)
 {
diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c
index cd6c7210a373..38669e827206 100644
--- a/fs/xfs/xfs_quotaops.c
+++ b/fs/xfs/xfs_quotaops.c
@@ -37,9 +37,9 @@ xfs_qm_fill_state(
 	tstate->flags |= QCI_SYSFILE;
 	tstate->blocks = ip->i_d.di_nblocks;
 	tstate->nextents = ip->i_d.di_nextents;
-	tstate->spc_timelimit = q->qi_btimelimit;
-	tstate->ino_timelimit = q->qi_itimelimit;
-	tstate->rt_spc_timelimit = q->qi_rtbtimelimit;
+	tstate->spc_timelimit = (u32)q->qi_btimelimit;
+	tstate->ino_timelimit = (u32)q->qi_itimelimit;
+	tstate->rt_spc_timelimit = (u32)q->qi_rtbtimelimit;
 	tstate->spc_warnlimit = q->qi_bwarnlimit;
 	tstate->ino_warnlimit = q->qi_iwarnlimit;
 	tstate->rt_spc_warnlimit = q->qi_rtbwarnlimit;
@@ -201,6 +201,9 @@ xfs_fs_rm_xquota(
 	if (XFS_IS_QUOTA_ON(mp))
 		return -EINVAL;
 
+	if (uflags & ~(FS_USER_QUOTA | FS_GROUP_QUOTA | FS_PROJ_QUOTA))
+		return -EINVAL;
+
 	if (uflags & FS_USER_QUOTA)
 		flags |= XFS_DQ_USER;
 	if (uflags & FS_GROUP_QUOTA)
diff --git a/fs/xfs/xfs_refcount_item.c b/fs/xfs/xfs_refcount_item.c
index 2328268e6245..8eeed73928cd 100644
--- a/fs/xfs/xfs_refcount_item.c
+++ b/fs/xfs/xfs_refcount_item.c
@@ -17,7 +17,7 @@
 #include "xfs_refcount_item.h"
 #include "xfs_log.h"
 #include "xfs_refcount.h"
-
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_cui_zone;
 kmem_zone_t	*xfs_cud_zone;
@@ -34,7 +34,7 @@ xfs_cui_item_free(
 	if (cuip->cui_format.cui_nextents > XFS_CUI_MAX_FAST_EXTENTS)
 		kmem_free(cuip);
 	else
-		kmem_zone_free(xfs_cui_zone, cuip);
+		kmem_cache_free(xfs_cui_zone, cuip);
 }
 
 /*
@@ -206,7 +206,7 @@ xfs_cud_item_release(
 	struct xfs_cud_log_item	*cudp = CUD_ITEM(lip);
 
 	xfs_cui_release(cudp->cud_cuip);
-	kmem_zone_free(xfs_cud_zone, cudp);
+	kmem_cache_free(xfs_cud_zone, cudp);
 }
 
 static const struct xfs_item_ops xfs_cud_item_ops = {
@@ -497,7 +497,7 @@ xfs_cui_recover(
 			 */
 			set_bit(XFS_CUI_RECOVERED, &cuip->cui_flags);
 			xfs_cui_release(cuip);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 	}
 
@@ -536,6 +536,7 @@ xfs_cui_recover(
 			type = refc_type;
 			break;
 		default:
+			XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
 			error = -EFSCORRUPTED;
 			goto abort_error;
 		}
diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c
index 0f08153b4994..107bf2a2f344 100644
--- a/fs/xfs/xfs_reflink.c
+++ b/fs/xfs/xfs_reflink.c
@@ -143,8 +143,6 @@ xfs_reflink_find_shared(
 	error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
 	if (error)
 		return error;
-	if (!agbp)
-		return -ENOMEM;
 
 	cur = xfs_refcountbt_init_cursor(mp, tp, agbp, agno);
 
@@ -223,8 +221,8 @@ xfs_reflink_trim_around_shared(
 	}
 }
 
-bool
-xfs_inode_need_cow(
+int
+xfs_bmap_trim_cow(
 	struct xfs_inode	*ip,
 	struct xfs_bmbt_irec	*imap,
 	bool			*shared)
@@ -308,13 +306,13 @@ static int
 xfs_find_trim_cow_extent(
 	struct xfs_inode	*ip,
 	struct xfs_bmbt_irec	*imap,
+	struct xfs_bmbt_irec	*cmap,
 	bool			*shared,
 	bool			*found)
 {
 	xfs_fileoff_t		offset_fsb = imap->br_startoff;
 	xfs_filblks_t		count_fsb = imap->br_blockcount;
 	struct xfs_iext_cursor	icur;
-	struct xfs_bmbt_irec	got;
 
 	*found = false;
 
@@ -322,23 +320,22 @@ xfs_find_trim_cow_extent(
 	 * If we don't find an overlapping extent, trim the range we need to
 	 * allocate to fit the hole we found.
 	 */
-	if (!xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &got))
-		got.br_startoff = offset_fsb + count_fsb;
-	if (got.br_startoff > offset_fsb) {
+	if (!xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, cmap))
+		cmap->br_startoff = offset_fsb + count_fsb;
+	if (cmap->br_startoff > offset_fsb) {
 		xfs_trim_extent(imap, imap->br_startoff,
-				got.br_startoff - imap->br_startoff);
-		return xfs_inode_need_cow(ip, imap, shared);
+				cmap->br_startoff - imap->br_startoff);
+		return xfs_bmap_trim_cow(ip, imap, shared);
 	}
 
 	*shared = true;
-	if (isnullstartblock(got.br_startblock)) {
-		xfs_trim_extent(imap, got.br_startoff, got.br_blockcount);
+	if (isnullstartblock(cmap->br_startblock)) {
+		xfs_trim_extent(imap, cmap->br_startoff, cmap->br_blockcount);
 		return 0;
 	}
 
 	/* real extent found - no need to allocate */
-	xfs_trim_extent(&got, offset_fsb, count_fsb);
-	*imap = got;
+	xfs_trim_extent(cmap, offset_fsb, count_fsb);
 	*found = true;
 	return 0;
 }
@@ -348,6 +345,7 @@ int
 xfs_reflink_allocate_cow(
 	struct xfs_inode	*ip,
 	struct xfs_bmbt_irec	*imap,
+	struct xfs_bmbt_irec	*cmap,
 	bool			*shared,
 	uint			*lockmode,
 	bool			convert_now)
@@ -367,7 +365,7 @@ xfs_reflink_allocate_cow(
 		xfs_ifork_init_cow(ip);
 	}
 
-	error = xfs_find_trim_cow_extent(ip, imap, shared, &found);
+	error = xfs_find_trim_cow_extent(ip, imap, cmap, shared, &found);
 	if (error || !*shared)
 		return error;
 	if (found)
@@ -392,7 +390,7 @@ xfs_reflink_allocate_cow(
 	/*
 	 * Check for an overlapping extent again now that we dropped the ilock.
 	 */
-	error = xfs_find_trim_cow_extent(ip, imap, shared, &found);
+	error = xfs_find_trim_cow_extent(ip, imap, cmap, shared, &found);
 	if (error || !*shared)
 		goto out_trans_cancel;
 	if (found) {
@@ -410,8 +408,8 @@ xfs_reflink_allocate_cow(
 	/* Allocate the entire reservation as unwritten blocks. */
 	nimaps = 1;
 	error = xfs_bmapi_write(tp, ip, imap->br_startoff, imap->br_blockcount,
-			XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC,
-			resblks, imap, &nimaps);
+			XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC, 0, cmap,
+			&nimaps);
 	if (error)
 		goto out_unreserve;
 
@@ -427,15 +425,15 @@ xfs_reflink_allocate_cow(
 	if (nimaps == 0)
 		return -ENOSPC;
 convert:
-	xfs_trim_extent(imap, offset_fsb, count_fsb);
+	xfs_trim_extent(cmap, offset_fsb, count_fsb);
 	/*
 	 * COW fork extents are supposed to remain unwritten until we're ready
 	 * to initiate a disk write.  For direct I/O we are going to write the
 	 * data and need the conversion, but for buffered writes we're done.
 	 */
-	if (!convert_now || imap->br_state == XFS_EXT_NORM)
+	if (!convert_now || cmap->br_state == XFS_EXT_NORM)
 		return 0;
-	trace_xfs_reflink_convert_cow(ip, imap);
+	trace_xfs_reflink_convert_cow(ip, cmap);
 	return xfs_reflink_convert_cow_locked(ip, offset_fsb, count_fsb);
 
 out_unreserve:
@@ -1053,6 +1051,7 @@ xfs_reflink_remap_extent(
 		uirec.br_startblock = irec->br_startblock + rlen;
 		uirec.br_startoff = irec->br_startoff + rlen;
 		uirec.br_blockcount = unmap_len - rlen;
+		uirec.br_state = irec->br_state;
 		unmap_len = rlen;
 
 		/* If this isn't a real mapping, we're done. */
@@ -1270,7 +1269,7 @@ xfs_reflink_zero_posteof(
 
 	trace_xfs_zero_eof(ip, isize, pos - isize);
 	return iomap_zero_range(VFS_I(ip), isize, pos - isize, NULL,
-			&xfs_iomap_ops);
+			&xfs_buffered_write_iomap_ops);
 }
 
 /*
@@ -1381,85 +1380,6 @@ out_unlock:
 	return ret;
 }
 
-/*
- * The user wants to preemptively CoW all shared blocks in this file,
- * which enables us to turn off the reflink flag.  Iterate all
- * extents which are not prealloc/delalloc to see which ranges are
- * mentioned in the refcount tree, then read those blocks into the
- * pagecache, dirty them, fsync them back out, and then we can update
- * the inode flag.  What happens if we run out of memory? :)
- */
-STATIC int
-xfs_reflink_dirty_extents(
-	struct xfs_inode	*ip,
-	xfs_fileoff_t		fbno,
-	xfs_filblks_t		end,
-	xfs_off_t		isize)
-{
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_agnumber_t		agno;
-	xfs_agblock_t		agbno;
-	xfs_extlen_t		aglen;
-	xfs_agblock_t		rbno;
-	xfs_extlen_t		rlen;
-	xfs_off_t		fpos;
-	xfs_off_t		flen;
-	struct xfs_bmbt_irec	map[2];
-	int			nmaps;
-	int			error = 0;
-
-	while (end - fbno > 0) {
-		nmaps = 1;
-		/*
-		 * Look for extents in the file.  Skip holes, delalloc, or
-		 * unwritten extents; they can't be reflinked.
-		 */
-		error = xfs_bmapi_read(ip, fbno, end - fbno, map, &nmaps, 0);
-		if (error)
-			goto out;
-		if (nmaps == 0)
-			break;
-		if (!xfs_bmap_is_real_extent(&map[0]))
-			goto next;
-
-		map[1] = map[0];
-		while (map[1].br_blockcount) {
-			agno = XFS_FSB_TO_AGNO(mp, map[1].br_startblock);
-			agbno = XFS_FSB_TO_AGBNO(mp, map[1].br_startblock);
-			aglen = map[1].br_blockcount;
-
-			error = xfs_reflink_find_shared(mp, NULL, agno, agbno,
-					aglen, &rbno, &rlen, true);
-			if (error)
-				goto out;
-			if (rbno == NULLAGBLOCK)
-				break;
-
-			/* Dirty the pages */
-			xfs_iunlock(ip, XFS_ILOCK_EXCL);
-			fpos = XFS_FSB_TO_B(mp, map[1].br_startoff +
-					(rbno - agbno));
-			flen = XFS_FSB_TO_B(mp, rlen);
-			if (fpos + flen > isize)
-				flen = isize - fpos;
-			error = iomap_file_dirty(VFS_I(ip), fpos, flen,
-					&xfs_iomap_ops);
-			xfs_ilock(ip, XFS_ILOCK_EXCL);
-			if (error)
-				goto out;
-
-			map[1].br_blockcount -= (rbno - agbno + rlen);
-			map[1].br_startoff += (rbno - agbno + rlen);
-			map[1].br_startblock += (rbno - agbno + rlen);
-		}
-
-next:
-		fbno = map[0].br_startoff + map[0].br_blockcount;
-	}
-out:
-	return error;
-}
-
 /* Does this inode need the reflink flag? */
 int
 xfs_reflink_inode_has_shared_extents(
@@ -1536,7 +1456,8 @@ xfs_reflink_clear_inode_flag(
 	 * We didn't find any shared blocks so turn off the reflink flag.
 	 * First, get rid of any leftover CoW mappings.
 	 */
-	error = xfs_reflink_cancel_cow_blocks(ip, tpp, 0, NULLFILEOFF, true);
+	error = xfs_reflink_cancel_cow_blocks(ip, tpp, 0, XFS_MAX_FILEOFF,
+			true);
 	if (error)
 		return error;
 
@@ -1596,10 +1517,7 @@ xfs_reflink_unshare(
 	xfs_off_t		offset,
 	xfs_off_t		len)
 {
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_fileoff_t		fbno;
-	xfs_filblks_t		end;
-	xfs_off_t		isize;
+	struct inode		*inode = VFS_I(ip);
 	int			error;
 
 	if (!xfs_is_reflink_inode(ip))
@@ -1607,20 +1525,13 @@ xfs_reflink_unshare(
 
 	trace_xfs_reflink_unshare(ip, offset, len);
 
-	inode_dio_wait(VFS_I(ip));
+	inode_dio_wait(inode);
 
-	/* Try to CoW the selected ranges */
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	fbno = XFS_B_TO_FSBT(mp, offset);
-	isize = i_size_read(VFS_I(ip));
-	end = XFS_B_TO_FSB(mp, offset + len);
-	error = xfs_reflink_dirty_extents(ip, fbno, end, isize);
+	error = iomap_file_unshare(inode, offset, len,
+			&xfs_buffered_write_iomap_ops);
 	if (error)
-		goto out_unlock;
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
-	/* Wait for the IO to finish */
-	error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
+		goto out;
+	error = filemap_write_and_wait(inode->i_mapping);
 	if (error)
 		goto out;
 
@@ -1628,11 +1539,8 @@ xfs_reflink_unshare(
 	error = xfs_reflink_try_clear_inode_flag(ip);
 	if (error)
 		goto out;
-
 	return 0;
 
-out_unlock:
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 out:
 	trace_xfs_reflink_unshare_error(ip, error, _RET_IP_);
 	return error;
diff --git a/fs/xfs/xfs_reflink.h b/fs/xfs/xfs_reflink.h
index 28a43b7f581d..3e4fd46373ab 100644
--- a/fs/xfs/xfs_reflink.h
+++ b/fs/xfs/xfs_reflink.h
@@ -22,11 +22,11 @@ extern int xfs_reflink_find_shared(struct xfs_mount *mp, struct xfs_trans *tp,
 		xfs_agblock_t *fbno, xfs_extlen_t *flen, bool find_maximal);
 extern int xfs_reflink_trim_around_shared(struct xfs_inode *ip,
 		struct xfs_bmbt_irec *irec, bool *shared);
-bool xfs_inode_need_cow(struct xfs_inode *ip, struct xfs_bmbt_irec *imap,
+int xfs_bmap_trim_cow(struct xfs_inode *ip, struct xfs_bmbt_irec *imap,
 		bool *shared);
 
-extern int xfs_reflink_allocate_cow(struct xfs_inode *ip,
-		struct xfs_bmbt_irec *imap, bool *shared, uint *lockmode,
+int xfs_reflink_allocate_cow(struct xfs_inode *ip, struct xfs_bmbt_irec *imap,
+		struct xfs_bmbt_irec *cmap, bool *shared, uint *lockmode,
 		bool convert_now);
 extern int xfs_reflink_convert_cow(struct xfs_inode *ip, xfs_off_t offset,
 		xfs_off_t count);
diff --git a/fs/xfs/xfs_rmap_item.c b/fs/xfs/xfs_rmap_item.c
index 8939e0ea09cd..4911b68f95dd 100644
--- a/fs/xfs/xfs_rmap_item.c
+++ b/fs/xfs/xfs_rmap_item.c
@@ -17,7 +17,7 @@
 #include "xfs_rmap_item.h"
 #include "xfs_log.h"
 #include "xfs_rmap.h"
-
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_rui_zone;
 kmem_zone_t	*xfs_rud_zone;
@@ -34,7 +34,7 @@ xfs_rui_item_free(
 	if (ruip->rui_format.rui_nextents > XFS_RUI_MAX_FAST_EXTENTS)
 		kmem_free(ruip);
 	else
-		kmem_zone_free(xfs_rui_zone, ruip);
+		kmem_cache_free(xfs_rui_zone, ruip);
 }
 
 /*
@@ -171,8 +171,10 @@ xfs_rui_copy_format(
 	src_rui_fmt = buf->i_addr;
 	len = xfs_rui_log_format_sizeof(src_rui_fmt->rui_nextents);
 
-	if (buf->i_len != len)
+	if (buf->i_len != len) {
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 		return -EFSCORRUPTED;
+	}
 
 	memcpy(dst_rui_fmt, src_rui_fmt, len);
 	return 0;
@@ -227,7 +229,7 @@ xfs_rud_item_release(
 	struct xfs_rud_log_item	*rudp = RUD_ITEM(lip);
 
 	xfs_rui_release(rudp->rud_ruip);
-	kmem_zone_free(xfs_rud_zone, rudp);
+	kmem_cache_free(xfs_rud_zone, rudp);
 }
 
 static const struct xfs_item_ops xfs_rud_item_ops = {
@@ -539,7 +541,7 @@ xfs_rui_recover(
 			 */
 			set_bit(XFS_RUI_RECOVERED, &ruip->rui_flags);
 			xfs_rui_release(ruip);
-			return -EIO;
+			return -EFSCORRUPTED;
 		}
 	}
 
@@ -581,6 +583,7 @@ xfs_rui_recover(
 			type = XFS_RMAP_FREE;
 			break;
 		default:
+			XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 			error = -EFSCORRUPTED;
 			goto abort_error;
 		}
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 4a48a8c75b4f..6209e7b6b895 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -792,8 +792,7 @@ xfs_growfs_rt_alloc(
 		 */
 		nmap = 1;
 		error = xfs_bmapi_write(tp, ip, oblocks, nblocks - oblocks,
-					XFS_BMAPI_METADATA, resblks, &map,
-					&nmap);
+					XFS_BMAPI_METADATA, 0, &map, &nmap);
 		if (!error && nmap < 1)
 			error = -ENOSPC;
 		if (error)
@@ -827,12 +826,10 @@ xfs_growfs_rt_alloc(
 			 * Get a buffer for the block.
 			 */
 			d = XFS_FSB_TO_DADDR(mp, fsbno);
-			bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
-				mp->m_bsize, 0);
-			if (bp == NULL) {
-				error = -EIO;
+			error = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
+					mp->m_bsize, 0, &bp);
+			if (error)
 				goto out_trans_cancel;
-			}
 			memset(bp->b_addr, 0, mp->m_sb.sb_blocksize);
 			xfs_trans_log_buf(tp, bp, 0, mp->m_sb.sb_blocksize - 1);
 			/*
diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c
index 113883c4f202..f70f1255220b 100644
--- a/fs/xfs/xfs_stats.c
+++ b/fs/xfs/xfs_stats.c
@@ -57,13 +57,13 @@ int xfs_stats_format(struct xfsstats __percpu *stats, char *buf)
 	/* Loop over all stats groups */
 
 	for (i = j = 0; i < ARRAY_SIZE(xstats); i++) {
-		len += snprintf(buf + len, PATH_MAX - len, "%s",
+		len += scnprintf(buf + len, PATH_MAX - len, "%s",
 				xstats[i].desc);
 		/* inner loop does each group */
 		for (; j < xstats[i].endpoint; j++)
-			len += snprintf(buf + len, PATH_MAX - len, " %u",
+			len += scnprintf(buf + len, PATH_MAX - len, " %u",
 					counter_val(stats, j));
-		len += snprintf(buf + len, PATH_MAX - len, "\n");
+		len += scnprintf(buf + len, PATH_MAX - len, "\n");
 	}
 	/* extra precision counters */
 	for_each_possible_cpu(i) {
@@ -72,9 +72,9 @@ int xfs_stats_format(struct xfsstats __percpu *stats, char *buf)
 		xs_read_bytes += per_cpu_ptr(stats, i)->s.xs_read_bytes;
 	}
 
-	len += snprintf(buf + len, PATH_MAX-len, "xpc %Lu %Lu %Lu\n",
+	len += scnprintf(buf + len, PATH_MAX-len, "xpc %Lu %Lu %Lu\n",
 			xs_xstrat_bytes, xs_write_bytes, xs_read_bytes);
-	len += snprintf(buf + len, PATH_MAX-len, "debug %u\n",
+	len += scnprintf(buf + len, PATH_MAX-len, "debug %u\n",
 #if defined(DEBUG)
 		1);
 #else
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 8d1df9f8be07..a123cd8267d9 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -37,10 +37,10 @@
 #include "xfs_reflink.h"
 
 #include <linux/magic.h>
-#include <linux/parser.h>
+#include <linux/fs_context.h>
+#include <linux/fs_parser.h>
 
 static const struct super_operations xfs_super_operations;
-struct bio_set xfs_ioend_bioset;
 
 static struct kset *xfs_kset;		/* top-level xfs sysfs dir */
 #ifdef DEBUG
@@ -51,7 +51,7 @@ static struct xfs_kobj xfs_dbg_kobj;	/* global debug sysfs attrs */
  * Table driven mount option parser.
  */
 enum {
-	Opt_logbufs, Opt_logbsize, Opt_logdev, Opt_rtdev, Opt_biosize,
+	Opt_logbufs, Opt_logbsize, Opt_logdev, Opt_rtdev,
 	Opt_wsync, Opt_noalign, Opt_swalloc, Opt_sunit, Opt_swidth, Opt_nouuid,
 	Opt_grpid, Opt_nogrpid, Opt_bsdgroups, Opt_sysvgroups,
 	Opt_allocsize, Opt_norecovery, Opt_inode64, Opt_inode32, Opt_ikeep,
@@ -59,382 +59,62 @@ enum {
 	Opt_filestreams, Opt_quota, Opt_noquota, Opt_usrquota, Opt_grpquota,
 	Opt_prjquota, Opt_uquota, Opt_gquota, Opt_pquota,
 	Opt_uqnoenforce, Opt_gqnoenforce, Opt_pqnoenforce, Opt_qnoenforce,
-	Opt_discard, Opt_nodiscard, Opt_dax, Opt_err,
+	Opt_discard, Opt_nodiscard, Opt_dax,
 };
 
-static const match_table_t tokens = {
-	{Opt_logbufs,	"logbufs=%u"},	/* number of XFS log buffers */
-	{Opt_logbsize,	"logbsize=%s"},	/* size of XFS log buffers */
-	{Opt_logdev,	"logdev=%s"},	/* log device */
-	{Opt_rtdev,	"rtdev=%s"},	/* realtime I/O device */
-	{Opt_biosize,	"biosize=%u"},	/* log2 of preferred buffered io size */
-	{Opt_wsync,	"wsync"},	/* safe-mode nfs compatible mount */
-	{Opt_noalign,	"noalign"},	/* turn off stripe alignment */
-	{Opt_swalloc,	"swalloc"},	/* turn on stripe width allocation */
-	{Opt_sunit,	"sunit=%u"},	/* data volume stripe unit */
-	{Opt_swidth,	"swidth=%u"},	/* data volume stripe width */
-	{Opt_nouuid,	"nouuid"},	/* ignore filesystem UUID */
-	{Opt_grpid,	"grpid"},	/* group-ID from parent directory */
-	{Opt_nogrpid,	"nogrpid"},	/* group-ID from current process */
-	{Opt_bsdgroups,	"bsdgroups"},	/* group-ID from parent directory */
-	{Opt_sysvgroups,"sysvgroups"},	/* group-ID from current process */
-	{Opt_allocsize,	"allocsize=%s"},/* preferred allocation size */
-	{Opt_norecovery,"norecovery"},	/* don't run XFS recovery */
-	{Opt_inode64,	"inode64"},	/* inodes can be allocated anywhere */
-	{Opt_inode32,   "inode32"},	/* inode allocation limited to
-					 * XFS_MAXINUMBER_32 */
-	{Opt_ikeep,	"ikeep"},	/* do not free empty inode clusters */
-	{Opt_noikeep,	"noikeep"},	/* free empty inode clusters */
-	{Opt_largeio,	"largeio"},	/* report large I/O sizes in stat() */
-	{Opt_nolargeio,	"nolargeio"},	/* do not report large I/O sizes
-					 * in stat(). */
-	{Opt_attr2,	"attr2"},	/* do use attr2 attribute format */
-	{Opt_noattr2,	"noattr2"},	/* do not use attr2 attribute format */
-	{Opt_filestreams,"filestreams"},/* use filestreams allocator */
-	{Opt_quota,	"quota"},	/* disk quotas (user) */
-	{Opt_noquota,	"noquota"},	/* no quotas */
-	{Opt_usrquota,	"usrquota"},	/* user quota enabled */
-	{Opt_grpquota,	"grpquota"},	/* group quota enabled */
-	{Opt_prjquota,	"prjquota"},	/* project quota enabled */
-	{Opt_uquota,	"uquota"},	/* user quota (IRIX variant) */
-	{Opt_gquota,	"gquota"},	/* group quota (IRIX variant) */
-	{Opt_pquota,	"pquota"},	/* project quota (IRIX variant) */
-	{Opt_uqnoenforce,"uqnoenforce"},/* user quota limit enforcement */
-	{Opt_gqnoenforce,"gqnoenforce"},/* group quota limit enforcement */
-	{Opt_pqnoenforce,"pqnoenforce"},/* project quota limit enforcement */
-	{Opt_qnoenforce, "qnoenforce"},	/* same as uqnoenforce */
-	{Opt_discard,	"discard"},	/* Discard unused blocks */
-	{Opt_nodiscard,	"nodiscard"},	/* Do not discard unused blocks */
-	{Opt_dax,	"dax"},		/* Enable direct access to bdev pages */
-	{Opt_err,	NULL},
+static const struct fs_parameter_spec xfs_fs_parameters[] = {
+	fsparam_u32("logbufs",		Opt_logbufs),
+	fsparam_string("logbsize",	Opt_logbsize),
+	fsparam_string("logdev",	Opt_logdev),
+	fsparam_string("rtdev",		Opt_rtdev),
+	fsparam_flag("wsync",		Opt_wsync),
+	fsparam_flag("noalign",		Opt_noalign),
+	fsparam_flag("swalloc",		Opt_swalloc),
+	fsparam_u32("sunit",		Opt_sunit),
+	fsparam_u32("swidth",		Opt_swidth),
+	fsparam_flag("nouuid",		Opt_nouuid),
+	fsparam_flag("grpid",		Opt_grpid),
+	fsparam_flag("nogrpid",		Opt_nogrpid),
+	fsparam_flag("bsdgroups",	Opt_bsdgroups),
+	fsparam_flag("sysvgroups",	Opt_sysvgroups),
+	fsparam_string("allocsize",	Opt_allocsize),
+	fsparam_flag("norecovery",	Opt_norecovery),
+	fsparam_flag("inode64",		Opt_inode64),
+	fsparam_flag("inode32",		Opt_inode32),
+	fsparam_flag("ikeep",		Opt_ikeep),
+	fsparam_flag("noikeep",		Opt_noikeep),
+	fsparam_flag("largeio",		Opt_largeio),
+	fsparam_flag("nolargeio",	Opt_nolargeio),
+	fsparam_flag("attr2",		Opt_attr2),
+	fsparam_flag("noattr2",		Opt_noattr2),
+	fsparam_flag("filestreams",	Opt_filestreams),
+	fsparam_flag("quota",		Opt_quota),
+	fsparam_flag("noquota",		Opt_noquota),
+	fsparam_flag("usrquota",	Opt_usrquota),
+	fsparam_flag("grpquota",	Opt_grpquota),
+	fsparam_flag("prjquota",	Opt_prjquota),
+	fsparam_flag("uquota",		Opt_uquota),
+	fsparam_flag("gquota",		Opt_gquota),
+	fsparam_flag("pquota",		Opt_pquota),
+	fsparam_flag("uqnoenforce",	Opt_uqnoenforce),
+	fsparam_flag("gqnoenforce",	Opt_gqnoenforce),
+	fsparam_flag("pqnoenforce",	Opt_pqnoenforce),
+	fsparam_flag("qnoenforce",	Opt_qnoenforce),
+	fsparam_flag("discard",		Opt_discard),
+	fsparam_flag("nodiscard",	Opt_nodiscard),
+	fsparam_flag("dax",		Opt_dax),
+	{}
 };
 
-
-STATIC int
-suffix_kstrtoint(const substring_t *s, unsigned int base, int *res)
-{
-	int	last, shift_left_factor = 0, _res;
-	char	*value;
-	int	ret = 0;
-
-	value = match_strdup(s);
-	if (!value)
-		return -ENOMEM;
-
-	last = strlen(value) - 1;
-	if (value[last] == 'K' || value[last] == 'k') {
-		shift_left_factor = 10;
-		value[last] = '\0';
-	}
-	if (value[last] == 'M' || value[last] == 'm') {
-		shift_left_factor = 20;
-		value[last] = '\0';
-	}
-	if (value[last] == 'G' || value[last] == 'g') {
-		shift_left_factor = 30;
-		value[last] = '\0';
-	}
-
-	if (kstrtoint(value, base, &_res))
-		ret = -EINVAL;
-	kfree(value);
-	*res = _res << shift_left_factor;
-	return ret;
-}
-
-/*
- * This function fills in xfs_mount_t fields based on mount args.
- * Note: the superblock has _not_ yet been read in.
- *
- * Note that this function leaks the various device name allocations on
- * failure.  The caller takes care of them.
- *
- * *sb is const because this is also used to test options on the remount
- * path, and we don't want this to have any side effects at remount time.
- * Today this function does not change *sb, but just to future-proof...
- */
-STATIC int
-xfs_parseargs(
-	struct xfs_mount	*mp,
-	char			*options)
-{
-	const struct super_block *sb = mp->m_super;
-	char			*p;
-	substring_t		args[MAX_OPT_ARGS];
-	int			dsunit = 0;
-	int			dswidth = 0;
-	int			iosize = 0;
-	uint8_t			iosizelog = 0;
-
-	/*
-	 * set up the mount name first so all the errors will refer to the
-	 * correct device.
-	 */
-	mp->m_fsname = kstrndup(sb->s_id, MAXNAMELEN, GFP_KERNEL);
-	if (!mp->m_fsname)
-		return -ENOMEM;
-	mp->m_fsname_len = strlen(mp->m_fsname) + 1;
-
-	/*
-	 * Copy binary VFS mount flags we are interested in.
-	 */
-	if (sb_rdonly(sb))
-		mp->m_flags |= XFS_MOUNT_RDONLY;
-	if (sb->s_flags & SB_DIRSYNC)
-		mp->m_flags |= XFS_MOUNT_DIRSYNC;
-	if (sb->s_flags & SB_SYNCHRONOUS)
-		mp->m_flags |= XFS_MOUNT_WSYNC;
-
-	/*
-	 * Set some default flags that could be cleared by the mount option
-	 * parsing.
-	 */
-	mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-
-	/*
-	 * These can be overridden by the mount option parsing.
-	 */
-	mp->m_logbufs = -1;
-	mp->m_logbsize = -1;
-
-	if (!options)
-		goto done;
-
-	while ((p = strsep(&options, ",")) != NULL) {
-		int		token;
-
-		if (!*p)
-			continue;
-
-		token = match_token(p, tokens, args);
-		switch (token) {
-		case Opt_logbufs:
-			if (match_int(args, &mp->m_logbufs))
-				return -EINVAL;
-			break;
-		case Opt_logbsize:
-			if (suffix_kstrtoint(args, 10, &mp->m_logbsize))
-				return -EINVAL;
-			break;
-		case Opt_logdev:
-			kfree(mp->m_logname);
-			mp->m_logname = match_strdup(args);
-			if (!mp->m_logname)
-				return -ENOMEM;
-			break;
-		case Opt_rtdev:
-			kfree(mp->m_rtname);
-			mp->m_rtname = match_strdup(args);
-			if (!mp->m_rtname)
-				return -ENOMEM;
-			break;
-		case Opt_allocsize:
-		case Opt_biosize:
-			if (suffix_kstrtoint(args, 10, &iosize))
-				return -EINVAL;
-			iosizelog = ffs(iosize) - 1;
-			break;
-		case Opt_grpid:
-		case Opt_bsdgroups:
-			mp->m_flags |= XFS_MOUNT_GRPID;
-			break;
-		case Opt_nogrpid:
-		case Opt_sysvgroups:
-			mp->m_flags &= ~XFS_MOUNT_GRPID;
-			break;
-		case Opt_wsync:
-			mp->m_flags |= XFS_MOUNT_WSYNC;
-			break;
-		case Opt_norecovery:
-			mp->m_flags |= XFS_MOUNT_NORECOVERY;
-			break;
-		case Opt_noalign:
-			mp->m_flags |= XFS_MOUNT_NOALIGN;
-			break;
-		case Opt_swalloc:
-			mp->m_flags |= XFS_MOUNT_SWALLOC;
-			break;
-		case Opt_sunit:
-			if (match_int(args, &dsunit))
-				return -EINVAL;
-			break;
-		case Opt_swidth:
-			if (match_int(args, &dswidth))
-				return -EINVAL;
-			break;
-		case Opt_inode32:
-			mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
-			break;
-		case Opt_inode64:
-			mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
-			break;
-		case Opt_nouuid:
-			mp->m_flags |= XFS_MOUNT_NOUUID;
-			break;
-		case Opt_ikeep:
-			mp->m_flags |= XFS_MOUNT_IKEEP;
-			break;
-		case Opt_noikeep:
-			mp->m_flags &= ~XFS_MOUNT_IKEEP;
-			break;
-		case Opt_largeio:
-			mp->m_flags &= ~XFS_MOUNT_COMPAT_IOSIZE;
-			break;
-		case Opt_nolargeio:
-			mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-			break;
-		case Opt_attr2:
-			mp->m_flags |= XFS_MOUNT_ATTR2;
-			break;
-		case Opt_noattr2:
-			mp->m_flags &= ~XFS_MOUNT_ATTR2;
-			mp->m_flags |= XFS_MOUNT_NOATTR2;
-			break;
-		case Opt_filestreams:
-			mp->m_flags |= XFS_MOUNT_FILESTREAMS;
-			break;
-		case Opt_noquota:
-			mp->m_qflags &= ~XFS_ALL_QUOTA_ACCT;
-			mp->m_qflags &= ~XFS_ALL_QUOTA_ENFD;
-			mp->m_qflags &= ~XFS_ALL_QUOTA_ACTIVE;
-			break;
-		case Opt_quota:
-		case Opt_uquota:
-		case Opt_usrquota:
-			mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE |
-					 XFS_UQUOTA_ENFD);
-			break;
-		case Opt_qnoenforce:
-		case Opt_uqnoenforce:
-			mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
-			mp->m_qflags &= ~XFS_UQUOTA_ENFD;
-			break;
-		case Opt_pquota:
-		case Opt_prjquota:
-			mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE |
-					 XFS_PQUOTA_ENFD);
-			break;
-		case Opt_pqnoenforce:
-			mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
-			mp->m_qflags &= ~XFS_PQUOTA_ENFD;
-			break;
-		case Opt_gquota:
-		case Opt_grpquota:
-			mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE |
-					 XFS_GQUOTA_ENFD);
-			break;
-		case Opt_gqnoenforce:
-			mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
-			mp->m_qflags &= ~XFS_GQUOTA_ENFD;
-			break;
-		case Opt_discard:
-			mp->m_flags |= XFS_MOUNT_DISCARD;
-			break;
-		case Opt_nodiscard:
-			mp->m_flags &= ~XFS_MOUNT_DISCARD;
-			break;
-#ifdef CONFIG_FS_DAX
-		case Opt_dax:
-			mp->m_flags |= XFS_MOUNT_DAX;
-			break;
-#endif
-		default:
-			xfs_warn(mp, "unknown mount option [%s].", p);
-			return -EINVAL;
-		}
-	}
-
-	/*
-	 * no recovery flag requires a read-only mount
-	 */
-	if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
-	    !(mp->m_flags & XFS_MOUNT_RDONLY)) {
-		xfs_warn(mp, "no-recovery mounts must be read-only.");
-		return -EINVAL;
-	}
-
-	if ((mp->m_flags & XFS_MOUNT_NOALIGN) && (dsunit || dswidth)) {
-		xfs_warn(mp,
-	"sunit and swidth options incompatible with the noalign option");
-		return -EINVAL;
-	}
-
-#ifndef CONFIG_XFS_QUOTA
-	if (XFS_IS_QUOTA_RUNNING(mp)) {
-		xfs_warn(mp, "quota support not available in this kernel.");
-		return -EINVAL;
-	}
-#endif
-
-	if ((dsunit && !dswidth) || (!dsunit && dswidth)) {
-		xfs_warn(mp, "sunit and swidth must be specified together");
-		return -EINVAL;
-	}
-
-	if (dsunit && (dswidth % dsunit != 0)) {
-		xfs_warn(mp,
-	"stripe width (%d) must be a multiple of the stripe unit (%d)",
-			dswidth, dsunit);
-		return -EINVAL;
-	}
-
-done:
-	if (dsunit && !(mp->m_flags & XFS_MOUNT_NOALIGN)) {
-		/*
-		 * At this point the superblock has not been read
-		 * in, therefore we do not know the block size.
-		 * Before the mount call ends we will convert
-		 * these to FSBs.
-		 */
-		mp->m_dalign = dsunit;
-		mp->m_swidth = dswidth;
-	}
-
-	if (mp->m_logbufs != -1 &&
-	    mp->m_logbufs != 0 &&
-	    (mp->m_logbufs < XLOG_MIN_ICLOGS ||
-	     mp->m_logbufs > XLOG_MAX_ICLOGS)) {
-		xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]",
-			mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
-		return -EINVAL;
-	}
-	if (mp->m_logbsize != -1 &&
-	    mp->m_logbsize !=  0 &&
-	    (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE ||
-	     mp->m_logbsize > XLOG_MAX_RECORD_BSIZE ||
-	     !is_power_of_2(mp->m_logbsize))) {
-		xfs_warn(mp,
-			"invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
-			mp->m_logbsize);
-		return -EINVAL;
-	}
-
-	if (iosizelog) {
-		if (iosizelog > XFS_MAX_IO_LOG ||
-		    iosizelog < XFS_MIN_IO_LOG) {
-			xfs_warn(mp, "invalid log iosize: %d [not %d-%d]",
-				iosizelog, XFS_MIN_IO_LOG,
-				XFS_MAX_IO_LOG);
-			return -EINVAL;
-		}
-
-		mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE;
-		mp->m_readio_log = iosizelog;
-		mp->m_writeio_log = iosizelog;
-	}
-
-	return 0;
-}
-
 struct proc_xfs_info {
 	uint64_t	flag;
 	char		*str;
 };
 
-STATIC void
-xfs_showargs(
-	struct xfs_mount	*mp,
-	struct seq_file		*m)
+static int
+xfs_fs_show_options(
+	struct seq_file		*m,
+	struct dentry		*root)
 {
 	static struct proc_xfs_info xfs_info_set[] = {
 		/* the few simple ones we can get from the mount struct */
@@ -448,30 +128,24 @@ xfs_showargs(
 		{ XFS_MOUNT_FILESTREAMS,	",filestreams" },
 		{ XFS_MOUNT_GRPID,		",grpid" },
 		{ XFS_MOUNT_DISCARD,		",discard" },
-		{ XFS_MOUNT_SMALL_INUMS,	",inode32" },
+		{ XFS_MOUNT_LARGEIO,		",largeio" },
 		{ XFS_MOUNT_DAX,		",dax" },
 		{ 0, NULL }
 	};
-	static struct proc_xfs_info xfs_info_unset[] = {
-		/* the few simple ones we can get from the mount struct */
-		{ XFS_MOUNT_COMPAT_IOSIZE,	",largeio" },
-		{ XFS_MOUNT_SMALL_INUMS,	",inode64" },
-		{ 0, NULL }
-	};
+	struct xfs_mount	*mp = XFS_M(root->d_sb);
 	struct proc_xfs_info	*xfs_infop;
 
 	for (xfs_infop = xfs_info_set; xfs_infop->flag; xfs_infop++) {
 		if (mp->m_flags & xfs_infop->flag)
 			seq_puts(m, xfs_infop->str);
 	}
-	for (xfs_infop = xfs_info_unset; xfs_infop->flag; xfs_infop++) {
-		if (!(mp->m_flags & xfs_infop->flag))
-			seq_puts(m, xfs_infop->str);
-	}
 
-	if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE)
+	seq_printf(m, ",inode%d",
+		(mp->m_flags & XFS_MOUNT_SMALL_INUMS) ? 32 : 64);
+
+	if (mp->m_flags & XFS_MOUNT_ALLOCSIZE)
 		seq_printf(m, ",allocsize=%dk",
-				(int)(1 << mp->m_writeio_log) >> 10);
+			   (1 << mp->m_allocsize_log) >> 10);
 
 	if (mp->m_logbufs > 0)
 		seq_printf(m, ",logbufs=%d", mp->m_logbufs);
@@ -510,32 +184,8 @@ xfs_showargs(
 
 	if (!(mp->m_qflags & XFS_ALL_QUOTA_ACCT))
 		seq_puts(m, ",noquota");
-}
 
-static uint64_t
-xfs_max_file_offset(
-	unsigned int		blockshift)
-{
-	unsigned int		pagefactor = 1;
-	unsigned int		bitshift = BITS_PER_LONG - 1;
-
-	/* Figure out maximum filesize, on Linux this can depend on
-	 * the filesystem blocksize (on 32 bit platforms).
-	 * __block_write_begin does this in an [unsigned] long long...
-	 *      page->index << (PAGE_SHIFT - bbits)
-	 * So, for page sized blocks (4K on 32 bit platforms),
-	 * this wraps at around 8Tb (hence MAX_LFS_FILESIZE which is
-	 *      (((u64)PAGE_SIZE << (BITS_PER_LONG-1))-1)
-	 * but for smaller blocksizes it is less (bbits = log2 bsize).
-	 */
-
-#if BITS_PER_LONG == 32
-	ASSERT(sizeof(sector_t) == 8);
-	pagefactor = PAGE_SIZE;
-	bitshift = BITS_PER_LONG;
-#endif
-
-	return (((uint64_t)pagefactor) << bitshift) - 1;
+	return 0;
 }
 
 /*
@@ -655,7 +305,7 @@ void
 xfs_blkdev_issue_flush(
 	xfs_buftarg_t		*buftarg)
 {
-	blkdev_issue_flush(buftarg->bt_bdev, GFP_NOFS, NULL);
+	blkdev_issue_flush(buftarg->bt_bdev, GFP_NOFS);
 }
 
 STATIC void
@@ -808,33 +458,33 @@ xfs_init_mount_workqueues(
 	struct xfs_mount	*mp)
 {
 	mp->m_buf_workqueue = alloc_workqueue("xfs-buf/%s",
-			WQ_MEM_RECLAIM|WQ_FREEZABLE, 1, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 1, mp->m_super->s_id);
 	if (!mp->m_buf_workqueue)
 		goto out;
 
 	mp->m_unwritten_workqueue = alloc_workqueue("xfs-conv/%s",
-			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id);
 	if (!mp->m_unwritten_workqueue)
 		goto out_destroy_buf;
 
 	mp->m_cil_workqueue = alloc_workqueue("xfs-cil/%s",
 			WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_UNBOUND,
-			0, mp->m_fsname);
+			0, mp->m_super->s_id);
 	if (!mp->m_cil_workqueue)
 		goto out_destroy_unwritten;
 
 	mp->m_reclaim_workqueue = alloc_workqueue("xfs-reclaim/%s",
-			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id);
 	if (!mp->m_reclaim_workqueue)
 		goto out_destroy_cil;
 
 	mp->m_eofblocks_workqueue = alloc_workqueue("xfs-eofblocks/%s",
-			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id);
 	if (!mp->m_eofblocks_workqueue)
 		goto out_destroy_reclaim;
 
 	mp->m_sync_workqueue = alloc_workqueue("xfs-sync/%s", WQ_FREEZABLE, 0,
-					       mp->m_fsname);
+					       mp->m_super->s_id);
 	if (!mp->m_sync_workqueue)
 		goto out_destroy_eofb;
 
@@ -866,6 +516,20 @@ xfs_destroy_mount_workqueues(
 	destroy_workqueue(mp->m_buf_workqueue);
 }
 
+static void
+xfs_flush_inodes_worker(
+	struct work_struct	*work)
+{
+	struct xfs_mount	*mp = container_of(work, struct xfs_mount,
+						   m_flush_inodes_work);
+	struct super_block	*sb = mp->m_super;
+
+	if (down_read_trylock(&sb->s_umount)) {
+		sync_inodes_sb(sb);
+		up_read(&sb->s_umount);
+	}
+}
+
 /*
  * Flush all dirty data to disk. Must not be called while holding an XFS_ILOCK
  * or a page lock. We use sync_inodes_sb() here to ensure we block while waiting
@@ -876,12 +540,15 @@ void
 xfs_flush_inodes(
 	struct xfs_mount	*mp)
 {
-	struct super_block	*sb = mp->m_super;
+	/*
+	 * If flush_work() returns true then that means we waited for a flush
+	 * which was already in progress.  Don't bother running another scan.
+	 */
+	if (flush_work(&mp->m_flush_inodes_work))
+		return;
 
-	if (down_read_trylock(&sb->s_umount)) {
-		sync_inodes_sb(sb);
-		up_read(&sb->s_umount);
-	}
+	queue_work(mp->m_sync_workqueue, &mp->m_flush_inodes_work);
+	flush_work(&mp->m_flush_inodes_work);
 }
 
 /* Catch misguided souls that try to use this interface on XFS */
@@ -1038,13 +705,13 @@ xfs_fs_drop_inode(
 	return generic_drop_inode(inode) || (ip->i_flags & XFS_IDONTCACHE);
 }
 
-STATIC void
-xfs_free_fsname(
+static void
+xfs_mount_free(
 	struct xfs_mount	*mp)
 {
-	kfree(mp->m_fsname);
 	kfree(mp->m_rtname);
 	kfree(mp->m_logname);
+	kmem_free(mp);
 }
 
 STATIC int
@@ -1205,181 +872,6 @@ xfs_quiesce_attr(
 	xfs_log_quiesce(mp);
 }
 
-STATIC int
-xfs_test_remount_options(
-	struct super_block	*sb,
-	char			*options)
-{
-	int			error = 0;
-	struct xfs_mount	*tmp_mp;
-
-	tmp_mp = kmem_zalloc(sizeof(*tmp_mp), KM_MAYFAIL);
-	if (!tmp_mp)
-		return -ENOMEM;
-
-	tmp_mp->m_super = sb;
-	error = xfs_parseargs(tmp_mp, options);
-	xfs_free_fsname(tmp_mp);
-	kmem_free(tmp_mp);
-
-	return error;
-}
-
-STATIC int
-xfs_fs_remount(
-	struct super_block	*sb,
-	int			*flags,
-	char			*options)
-{
-	struct xfs_mount	*mp = XFS_M(sb);
-	xfs_sb_t		*sbp = &mp->m_sb;
-	substring_t		args[MAX_OPT_ARGS];
-	char			*p;
-	int			error;
-
-	/* First, check for complete junk; i.e. invalid options */
-	error = xfs_test_remount_options(sb, options);
-	if (error)
-		return error;
-
-	sync_filesystem(sb);
-	while ((p = strsep(&options, ",")) != NULL) {
-		int token;
-
-		if (!*p)
-			continue;
-
-		token = match_token(p, tokens, args);
-		switch (token) {
-		case Opt_inode64:
-			mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
-			mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
-			break;
-		case Opt_inode32:
-			mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
-			mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
-			break;
-		default:
-			/*
-			 * Logically we would return an error here to prevent
-			 * users from believing they might have changed
-			 * mount options using remount which can't be changed.
-			 *
-			 * But unfortunately mount(8) adds all options from
-			 * mtab and fstab to the mount arguments in some cases
-			 * so we can't blindly reject options, but have to
-			 * check for each specified option if it actually
-			 * differs from the currently set option and only
-			 * reject it if that's the case.
-			 *
-			 * Until that is implemented we return success for
-			 * every remount request, and silently ignore all
-			 * options that we can't actually change.
-			 */
-#if 0
-			xfs_info(mp,
-		"mount option \"%s\" not supported for remount", p);
-			return -EINVAL;
-#else
-			break;
-#endif
-		}
-	}
-
-	/* ro -> rw */
-	if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & SB_RDONLY)) {
-		if (mp->m_flags & XFS_MOUNT_NORECOVERY) {
-			xfs_warn(mp,
-		"ro->rw transition prohibited on norecovery mount");
-			return -EINVAL;
-		}
-
-		if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 &&
-		    xfs_sb_has_ro_compat_feature(sbp,
-					XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) {
-			xfs_warn(mp,
-"ro->rw transition prohibited on unknown (0x%x) ro-compat filesystem",
-				(sbp->sb_features_ro_compat &
-					XFS_SB_FEAT_RO_COMPAT_UNKNOWN));
-			return -EINVAL;
-		}
-
-		mp->m_flags &= ~XFS_MOUNT_RDONLY;
-
-		/*
-		 * If this is the first remount to writeable state we
-		 * might have some superblock changes to update.
-		 */
-		if (mp->m_update_sb) {
-			error = xfs_sync_sb(mp, false);
-			if (error) {
-				xfs_warn(mp, "failed to write sb changes");
-				return error;
-			}
-			mp->m_update_sb = false;
-		}
-
-		/*
-		 * Fill out the reserve pool if it is empty. Use the stashed
-		 * value if it is non-zero, otherwise go with the default.
-		 */
-		xfs_restore_resvblks(mp);
-		xfs_log_work_queue(mp);
-
-		/* Recover any CoW blocks that never got remapped. */
-		error = xfs_reflink_recover_cow(mp);
-		if (error) {
-			xfs_err(mp,
-	"Error %d recovering leftover CoW allocations.", error);
-			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-			return error;
-		}
-		xfs_start_block_reaping(mp);
-
-		/* Create the per-AG metadata reservation pool .*/
-		error = xfs_fs_reserve_ag_blocks(mp);
-		if (error && error != -ENOSPC)
-			return error;
-	}
-
-	/* rw -> ro */
-	if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & SB_RDONLY)) {
-		/*
-		 * Cancel background eofb scanning so it cannot race with the
-		 * final log force+buftarg wait and deadlock the remount.
-		 */
-		xfs_stop_block_reaping(mp);
-
-		/* Get rid of any leftover CoW reservations... */
-		error = xfs_icache_free_cowblocks(mp, NULL);
-		if (error) {
-			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-			return error;
-		}
-
-		/* Free the per-AG metadata reservation pool. */
-		error = xfs_fs_unreserve_ag_blocks(mp);
-		if (error) {
-			xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-			return error;
-		}
-
-		/*
-		 * Before we sync the metadata, we need to free up the reserve
-		 * block pool so that the used block count in the superblock on
-		 * disk is correct at the end of the remount. Stash the current
-		 * reserve pool size so that if we get remounted rw, we can
-		 * return it to the same size.
-		 */
-		xfs_save_resvblks(mp);
-
-		xfs_quiesce_attr(mp);
-		mp->m_flags |= XFS_MOUNT_RDONLY;
-	}
-
-	return 0;
-}
-
 /*
  * Second stage of a freeze. The data is already frozen so we only
  * need to take care of the metadata. Once that's done sync the superblock
@@ -1410,15 +902,6 @@ xfs_fs_unfreeze(
 	return 0;
 }
 
-STATIC int
-xfs_fs_show_options(
-	struct seq_file		*m,
-	struct dentry		*root)
-{
-	xfs_showargs(XFS_M(root->d_sb), m);
-	return 0;
-}
-
 /*
  * This function fills in xfs_mount_t fields based on mount args.
  * Note: the superblock _has_ now been read in.
@@ -1541,60 +1024,337 @@ xfs_destroy_percpu_counters(
 	percpu_counter_destroy(&mp->m_delalloc_blks);
 }
 
-static struct xfs_mount *
-xfs_mount_alloc(
+static void
+xfs_fs_put_super(
 	struct super_block	*sb)
 {
-	struct xfs_mount	*mp;
+	struct xfs_mount	*mp = XFS_M(sb);
 
-	mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL);
-	if (!mp)
-		return NULL;
+	/* if ->fill_super failed, we have no mount to tear down */
+	if (!sb->s_fs_info)
+		return;
 
-	mp->m_super = sb;
-	spin_lock_init(&mp->m_sb_lock);
-	spin_lock_init(&mp->m_agirotor_lock);
-	INIT_RADIX_TREE(&mp->m_perag_tree, GFP_ATOMIC);
-	spin_lock_init(&mp->m_perag_lock);
-	mutex_init(&mp->m_growlock);
-	atomic_set(&mp->m_active_trans, 0);
-	INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker);
-	INIT_DELAYED_WORK(&mp->m_eofblocks_work, xfs_eofblocks_worker);
-	INIT_DELAYED_WORK(&mp->m_cowblocks_work, xfs_cowblocks_worker);
-	mp->m_kobj.kobject.kset = xfs_kset;
-	/*
-	 * We don't create the finobt per-ag space reservation until after log
-	 * recovery, so we must set this to true so that an ifree transaction
-	 * started during log recovery will not depend on space reservations
-	 * for finobt expansion.
-	 */
-	mp->m_finobt_nores = true;
-	return mp;
+	xfs_notice(mp, "Unmounting Filesystem");
+	xfs_filestream_unmount(mp);
+	xfs_unmountfs(mp);
+
+	xfs_freesb(mp);
+	free_percpu(mp->m_stats.xs_stats);
+	xfs_destroy_percpu_counters(mp);
+	xfs_destroy_mount_workqueues(mp);
+	xfs_close_devices(mp);
+
+	sb->s_fs_info = NULL;
+	xfs_mount_free(mp);
 }
 
+static long
+xfs_fs_nr_cached_objects(
+	struct super_block	*sb,
+	struct shrink_control	*sc)
+{
+	/* Paranoia: catch incorrect calls during mount setup or teardown */
+	if (WARN_ON_ONCE(!sb->s_fs_info))
+		return 0;
+	return xfs_reclaim_inodes_count(XFS_M(sb));
+}
 
-STATIC int
-xfs_fs_fill_super(
+static long
+xfs_fs_free_cached_objects(
 	struct super_block	*sb,
-	void			*data,
-	int			silent)
+	struct shrink_control	*sc)
 {
-	struct inode		*root;
-	struct xfs_mount	*mp = NULL;
-	int			flags = 0, error = -ENOMEM;
+	return xfs_reclaim_inodes_nr(XFS_M(sb), sc->nr_to_scan);
+}
+
+static const struct super_operations xfs_super_operations = {
+	.alloc_inode		= xfs_fs_alloc_inode,
+	.destroy_inode		= xfs_fs_destroy_inode,
+	.dirty_inode		= xfs_fs_dirty_inode,
+	.drop_inode		= xfs_fs_drop_inode,
+	.put_super		= xfs_fs_put_super,
+	.sync_fs		= xfs_fs_sync_fs,
+	.freeze_fs		= xfs_fs_freeze,
+	.unfreeze_fs		= xfs_fs_unfreeze,
+	.statfs			= xfs_fs_statfs,
+	.show_options		= xfs_fs_show_options,
+	.nr_cached_objects	= xfs_fs_nr_cached_objects,
+	.free_cached_objects	= xfs_fs_free_cached_objects,
+};
+
+static int
+suffix_kstrtoint(
+	const char	*s,
+	unsigned int	base,
+	int		*res)
+{
+	int		last, shift_left_factor = 0, _res;
+	char		*value;
+	int		ret = 0;
 
+	value = kstrdup(s, GFP_KERNEL);
+	if (!value)
+		return -ENOMEM;
+
+	last = strlen(value) - 1;
+	if (value[last] == 'K' || value[last] == 'k') {
+		shift_left_factor = 10;
+		value[last] = '\0';
+	}
+	if (value[last] == 'M' || value[last] == 'm') {
+		shift_left_factor = 20;
+		value[last] = '\0';
+	}
+	if (value[last] == 'G' || value[last] == 'g') {
+		shift_left_factor = 30;
+		value[last] = '\0';
+	}
+
+	if (kstrtoint(value, base, &_res))
+		ret = -EINVAL;
+	kfree(value);
+	*res = _res << shift_left_factor;
+	return ret;
+}
+
+/*
+ * Set mount state from a mount option.
+ *
+ * NOTE: mp->m_super is NULL here!
+ */
+static int
+xfs_fc_parse_param(
+	struct fs_context	*fc,
+	struct fs_parameter	*param)
+{
+	struct xfs_mount	*mp = fc->s_fs_info;
+	struct fs_parse_result	result;
+	int			size = 0;
+	int			opt;
+
+	opt = fs_parse(fc, xfs_fs_parameters, param, &result);
+	if (opt < 0)
+		return opt;
+
+	switch (opt) {
+	case Opt_logbufs:
+		mp->m_logbufs = result.uint_32;
+		return 0;
+	case Opt_logbsize:
+		if (suffix_kstrtoint(param->string, 10, &mp->m_logbsize))
+			return -EINVAL;
+		return 0;
+	case Opt_logdev:
+		kfree(mp->m_logname);
+		mp->m_logname = kstrdup(param->string, GFP_KERNEL);
+		if (!mp->m_logname)
+			return -ENOMEM;
+		return 0;
+	case Opt_rtdev:
+		kfree(mp->m_rtname);
+		mp->m_rtname = kstrdup(param->string, GFP_KERNEL);
+		if (!mp->m_rtname)
+			return -ENOMEM;
+		return 0;
+	case Opt_allocsize:
+		if (suffix_kstrtoint(param->string, 10, &size))
+			return -EINVAL;
+		mp->m_allocsize_log = ffs(size) - 1;
+		mp->m_flags |= XFS_MOUNT_ALLOCSIZE;
+		return 0;
+	case Opt_grpid:
+	case Opt_bsdgroups:
+		mp->m_flags |= XFS_MOUNT_GRPID;
+		return 0;
+	case Opt_nogrpid:
+	case Opt_sysvgroups:
+		mp->m_flags &= ~XFS_MOUNT_GRPID;
+		return 0;
+	case Opt_wsync:
+		mp->m_flags |= XFS_MOUNT_WSYNC;
+		return 0;
+	case Opt_norecovery:
+		mp->m_flags |= XFS_MOUNT_NORECOVERY;
+		return 0;
+	case Opt_noalign:
+		mp->m_flags |= XFS_MOUNT_NOALIGN;
+		return 0;
+	case Opt_swalloc:
+		mp->m_flags |= XFS_MOUNT_SWALLOC;
+		return 0;
+	case Opt_sunit:
+		mp->m_dalign = result.uint_32;
+		return 0;
+	case Opt_swidth:
+		mp->m_swidth = result.uint_32;
+		return 0;
+	case Opt_inode32:
+		mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
+		return 0;
+	case Opt_inode64:
+		mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
+		return 0;
+	case Opt_nouuid:
+		mp->m_flags |= XFS_MOUNT_NOUUID;
+		return 0;
+	case Opt_ikeep:
+		mp->m_flags |= XFS_MOUNT_IKEEP;
+		return 0;
+	case Opt_noikeep:
+		mp->m_flags &= ~XFS_MOUNT_IKEEP;
+		return 0;
+	case Opt_largeio:
+		mp->m_flags |= XFS_MOUNT_LARGEIO;
+		return 0;
+	case Opt_nolargeio:
+		mp->m_flags &= ~XFS_MOUNT_LARGEIO;
+		return 0;
+	case Opt_attr2:
+		mp->m_flags |= XFS_MOUNT_ATTR2;
+		return 0;
+	case Opt_noattr2:
+		mp->m_flags &= ~XFS_MOUNT_ATTR2;
+		mp->m_flags |= XFS_MOUNT_NOATTR2;
+		return 0;
+	case Opt_filestreams:
+		mp->m_flags |= XFS_MOUNT_FILESTREAMS;
+		return 0;
+	case Opt_noquota:
+		mp->m_qflags &= ~XFS_ALL_QUOTA_ACCT;
+		mp->m_qflags &= ~XFS_ALL_QUOTA_ENFD;
+		mp->m_qflags &= ~XFS_ALL_QUOTA_ACTIVE;
+		return 0;
+	case Opt_quota:
+	case Opt_uquota:
+	case Opt_usrquota:
+		mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE |
+				 XFS_UQUOTA_ENFD);
+		return 0;
+	case Opt_qnoenforce:
+	case Opt_uqnoenforce:
+		mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
+		mp->m_qflags &= ~XFS_UQUOTA_ENFD;
+		return 0;
+	case Opt_pquota:
+	case Opt_prjquota:
+		mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE |
+				 XFS_PQUOTA_ENFD);
+		return 0;
+	case Opt_pqnoenforce:
+		mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
+		mp->m_qflags &= ~XFS_PQUOTA_ENFD;
+		return 0;
+	case Opt_gquota:
+	case Opt_grpquota:
+		mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE |
+				 XFS_GQUOTA_ENFD);
+		return 0;
+	case Opt_gqnoenforce:
+		mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
+		mp->m_qflags &= ~XFS_GQUOTA_ENFD;
+		return 0;
+	case Opt_discard:
+		mp->m_flags |= XFS_MOUNT_DISCARD;
+		return 0;
+	case Opt_nodiscard:
+		mp->m_flags &= ~XFS_MOUNT_DISCARD;
+		return 0;
+#ifdef CONFIG_FS_DAX
+	case Opt_dax:
+		mp->m_flags |= XFS_MOUNT_DAX;
+		return 0;
+#endif
+	default:
+		xfs_warn(mp, "unknown mount option [%s].", param->key);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+xfs_fc_validate_params(
+	struct xfs_mount	*mp)
+{
 	/*
-	 * allocate mp and do all low-level struct initializations before we
-	 * attach it to the super
+	 * no recovery flag requires a read-only mount
 	 */
-	mp = xfs_mount_alloc(sb);
-	if (!mp)
-		goto out;
-	sb->s_fs_info = mp;
+	if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
+	    !(mp->m_flags & XFS_MOUNT_RDONLY)) {
+		xfs_warn(mp, "no-recovery mounts must be read-only.");
+		return -EINVAL;
+	}
+
+	if ((mp->m_flags & XFS_MOUNT_NOALIGN) &&
+	    (mp->m_dalign || mp->m_swidth)) {
+		xfs_warn(mp,
+	"sunit and swidth options incompatible with the noalign option");
+		return -EINVAL;
+	}
+
+	if (!IS_ENABLED(CONFIG_XFS_QUOTA) && mp->m_qflags != 0) {
+		xfs_warn(mp, "quota support not available in this kernel.");
+		return -EINVAL;
+	}
+
+	if ((mp->m_dalign && !mp->m_swidth) ||
+	    (!mp->m_dalign && mp->m_swidth)) {
+		xfs_warn(mp, "sunit and swidth must be specified together");
+		return -EINVAL;
+	}
+
+	if (mp->m_dalign && (mp->m_swidth % mp->m_dalign != 0)) {
+		xfs_warn(mp,
+	"stripe width (%d) must be a multiple of the stripe unit (%d)",
+			mp->m_swidth, mp->m_dalign);
+		return -EINVAL;
+	}
+
+	if (mp->m_logbufs != -1 &&
+	    mp->m_logbufs != 0 &&
+	    (mp->m_logbufs < XLOG_MIN_ICLOGS ||
+	     mp->m_logbufs > XLOG_MAX_ICLOGS)) {
+		xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]",
+			mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
+		return -EINVAL;
+	}
+
+	if (mp->m_logbsize != -1 &&
+	    mp->m_logbsize !=  0 &&
+	    (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE ||
+	     mp->m_logbsize > XLOG_MAX_RECORD_BSIZE ||
+	     !is_power_of_2(mp->m_logbsize))) {
+		xfs_warn(mp,
+			"invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
+			mp->m_logbsize);
+		return -EINVAL;
+	}
+
+	if ((mp->m_flags & XFS_MOUNT_ALLOCSIZE) &&
+	    (mp->m_allocsize_log > XFS_MAX_IO_LOG ||
+	     mp->m_allocsize_log < XFS_MIN_IO_LOG)) {
+		xfs_warn(mp, "invalid log iosize: %d [not %d-%d]",
+			mp->m_allocsize_log, XFS_MIN_IO_LOG, XFS_MAX_IO_LOG);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+xfs_fc_fill_super(
+	struct super_block	*sb,
+	struct fs_context	*fc)
+{
+	struct xfs_mount	*mp = sb->s_fs_info;
+	struct inode		*root;
+	int			flags = 0, error;
 
-	error = xfs_parseargs(mp, (char *)data);
+	mp->m_super = sb;
+
+	error = xfs_fc_validate_params(mp);
 	if (error)
-		goto out_free_fsname;
+		goto out_free_names;
 
 	sb_min_blocksize(sb, BBSIZE);
 	sb->s_xattr = xfs_xattr_handlers;
@@ -1616,12 +1376,12 @@ xfs_fs_fill_super(
 		msleep(xfs_globals.mount_delay * 1000);
 	}
 
-	if (silent)
+	if (fc->sb_flags & SB_SILENT)
 		flags |= XFS_MFSI_QUIET;
 
 	error = xfs_open_devices(mp);
 	if (error)
-		goto out_free_fsname;
+		goto out_free_names;
 
 	error = xfs_init_mount_workqueues(mp);
 	if (error)
@@ -1650,6 +1410,26 @@ xfs_fs_fill_super(
 	if (error)
 		goto out_free_sb;
 
+	/*
+	 * XFS block mappings use 54 bits to store the logical block offset.
+	 * This should suffice to handle the maximum file size that the VFS
+	 * supports (currently 2^63 bytes on 64-bit and ULONG_MAX << PAGE_SHIFT
+	 * bytes on 32-bit), but as XFS and VFS have gotten the s_maxbytes
+	 * calculation wrong on 32-bit kernels in the past, we'll add a WARN_ON
+	 * to check this assertion.
+	 *
+	 * Avoid integer overflow by comparing the maximum bmbt offset to the
+	 * maximum pagecache offset in units of fs blocks.
+	 */
+	if (XFS_B_TO_FSBT(mp, MAX_LFS_FILESIZE) > XFS_MAX_FILEOFF) {
+		xfs_warn(mp,
+"MAX_LFS_FILESIZE block offset (%llu) exceeds extent map maximum (%llu)!",
+			 XFS_B_TO_FSBT(mp, MAX_LFS_FILESIZE),
+			 XFS_MAX_FILEOFF);
+		error = -EINVAL;
+		goto out_free_sb;
+	}
+
 	error = xfs_filestream_mount(mp);
 	if (error)
 		goto out_free_sb;
@@ -1661,7 +1441,7 @@ xfs_fs_fill_super(
 	sb->s_magic = XFS_SUPER_MAGIC;
 	sb->s_blocksize = mp->m_sb.sb_blocksize;
 	sb->s_blocksize_bits = ffs(sb->s_blocksize) - 1;
-	sb->s_maxbytes = xfs_max_file_offset(sb->s_blocksize_bits);
+	sb->s_maxbytes = MAX_LFS_FILESIZE;
 	sb->s_max_links = XFS_MAXLINK;
 	sb->s_time_gran = 1;
 	sb->s_time_min = S32_MIN;
@@ -1758,11 +1538,9 @@ xfs_fs_fill_super(
 	xfs_destroy_mount_workqueues(mp);
  out_close_devices:
 	xfs_close_devices(mp);
- out_free_fsname:
+ out_free_names:
 	sb->s_fs_info = NULL;
-	xfs_free_fsname(mp);
-	kfree(mp);
- out:
+	xfs_mount_free(mp);
 	return error;
 
  out_unmount:
@@ -1771,80 +1549,253 @@ xfs_fs_fill_super(
 	goto out_free_sb;
 }
 
-STATIC void
-xfs_fs_put_super(
-	struct super_block	*sb)
+static int
+xfs_fc_get_tree(
+	struct fs_context	*fc)
 {
-	struct xfs_mount	*mp = XFS_M(sb);
+	return get_tree_bdev(fc, xfs_fc_fill_super);
+}
 
-	/* if ->fill_super failed, we have no mount to tear down */
-	if (!sb->s_fs_info)
-		return;
+static int
+xfs_remount_rw(
+	struct xfs_mount	*mp)
+{
+	struct xfs_sb		*sbp = &mp->m_sb;
+	int error;
 
-	xfs_notice(mp, "Unmounting Filesystem");
-	xfs_filestream_unmount(mp);
-	xfs_unmountfs(mp);
+	if (mp->m_flags & XFS_MOUNT_NORECOVERY) {
+		xfs_warn(mp,
+			"ro->rw transition prohibited on norecovery mount");
+		return -EINVAL;
+	}
 
-	xfs_freesb(mp);
-	free_percpu(mp->m_stats.xs_stats);
-	xfs_destroy_percpu_counters(mp);
-	xfs_destroy_mount_workqueues(mp);
-	xfs_close_devices(mp);
+	if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 &&
+	    xfs_sb_has_ro_compat_feature(sbp, XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) {
+		xfs_warn(mp,
+	"ro->rw transition prohibited on unknown (0x%x) ro-compat filesystem",
+			(sbp->sb_features_ro_compat &
+				XFS_SB_FEAT_RO_COMPAT_UNKNOWN));
+		return -EINVAL;
+	}
 
-	sb->s_fs_info = NULL;
-	xfs_free_fsname(mp);
-	kfree(mp);
+	mp->m_flags &= ~XFS_MOUNT_RDONLY;
+
+	/*
+	 * If this is the first remount to writeable state we might have some
+	 * superblock changes to update.
+	 */
+	if (mp->m_update_sb) {
+		error = xfs_sync_sb(mp, false);
+		if (error) {
+			xfs_warn(mp, "failed to write sb changes");
+			return error;
+		}
+		mp->m_update_sb = false;
+	}
+
+	/*
+	 * Fill out the reserve pool if it is empty. Use the stashed value if
+	 * it is non-zero, otherwise go with the default.
+	 */
+	xfs_restore_resvblks(mp);
+	xfs_log_work_queue(mp);
+
+	/* Recover any CoW blocks that never got remapped. */
+	error = xfs_reflink_recover_cow(mp);
+	if (error) {
+		xfs_err(mp,
+			"Error %d recovering leftover CoW allocations.", error);
+		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+		return error;
+	}
+	xfs_start_block_reaping(mp);
+
+	/* Create the per-AG metadata reservation pool .*/
+	error = xfs_fs_reserve_ag_blocks(mp);
+	if (error && error != -ENOSPC)
+		return error;
+
+	return 0;
 }
 
-STATIC struct dentry *
-xfs_fs_mount(
-	struct file_system_type	*fs_type,
-	int			flags,
-	const char		*dev_name,
-	void			*data)
+static int
+xfs_remount_ro(
+	struct xfs_mount	*mp)
 {
-	return mount_bdev(fs_type, flags, dev_name, data, xfs_fs_fill_super);
+	int error;
+
+	/*
+	 * Cancel background eofb scanning so it cannot race with the final
+	 * log force+buftarg wait and deadlock the remount.
+	 */
+	xfs_stop_block_reaping(mp);
+
+	/* Get rid of any leftover CoW reservations... */
+	error = xfs_icache_free_cowblocks(mp, NULL);
+	if (error) {
+		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+		return error;
+	}
+
+	/* Free the per-AG metadata reservation pool. */
+	error = xfs_fs_unreserve_ag_blocks(mp);
+	if (error) {
+		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+		return error;
+	}
+
+	/*
+	 * Before we sync the metadata, we need to free up the reserve block
+	 * pool so that the used block count in the superblock on disk is
+	 * correct at the end of the remount. Stash the current* reserve pool
+	 * size so that if we get remounted rw, we can return it to the same
+	 * size.
+	 */
+	xfs_save_resvblks(mp);
+
+	xfs_quiesce_attr(mp);
+	mp->m_flags |= XFS_MOUNT_RDONLY;
+
+	return 0;
 }
 
-static long
-xfs_fs_nr_cached_objects(
-	struct super_block	*sb,
-	struct shrink_control	*sc)
+/*
+ * Logically we would return an error here to prevent users from believing
+ * they might have changed mount options using remount which can't be changed.
+ *
+ * But unfortunately mount(8) adds all options from mtab and fstab to the mount
+ * arguments in some cases so we can't blindly reject options, but have to
+ * check for each specified option if it actually differs from the currently
+ * set option and only reject it if that's the case.
+ *
+ * Until that is implemented we return success for every remount request, and
+ * silently ignore all options that we can't actually change.
+ */
+static int
+xfs_fc_reconfigure(
+	struct fs_context *fc)
 {
-	/* Paranoia: catch incorrect calls during mount setup or teardown */
-	if (WARN_ON_ONCE(!sb->s_fs_info))
-		return 0;
-	return xfs_reclaim_inodes_count(XFS_M(sb));
+	struct xfs_mount	*mp = XFS_M(fc->root->d_sb);
+	struct xfs_mount        *new_mp = fc->s_fs_info;
+	xfs_sb_t		*sbp = &mp->m_sb;
+	int			flags = fc->sb_flags;
+	int			error;
+
+	error = xfs_fc_validate_params(new_mp);
+	if (error)
+		return error;
+
+	sync_filesystem(mp->m_super);
+
+	/* inode32 -> inode64 */
+	if ((mp->m_flags & XFS_MOUNT_SMALL_INUMS) &&
+	    !(new_mp->m_flags & XFS_MOUNT_SMALL_INUMS)) {
+		mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
+		mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
+	}
+
+	/* inode64 -> inode32 */
+	if (!(mp->m_flags & XFS_MOUNT_SMALL_INUMS) &&
+	    (new_mp->m_flags & XFS_MOUNT_SMALL_INUMS)) {
+		mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
+		mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
+	}
+
+	/* ro -> rw */
+	if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(flags & SB_RDONLY)) {
+		error = xfs_remount_rw(mp);
+		if (error)
+			return error;
+	}
+
+	/* rw -> ro */
+	if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (flags & SB_RDONLY)) {
+		error = xfs_remount_ro(mp);
+		if (error)
+			return error;
+	}
+
+	return 0;
 }
 
-static long
-xfs_fs_free_cached_objects(
-	struct super_block	*sb,
-	struct shrink_control	*sc)
+static void xfs_fc_free(
+	struct fs_context	*fc)
 {
-	return xfs_reclaim_inodes_nr(XFS_M(sb), sc->nr_to_scan);
+	struct xfs_mount	*mp = fc->s_fs_info;
+
+	/*
+	 * mp is stored in the fs_context when it is initialized.
+	 * mp is transferred to the superblock on a successful mount,
+	 * but if an error occurs before the transfer we have to free
+	 * it here.
+	 */
+	if (mp)
+		xfs_mount_free(mp);
 }
 
-static const struct super_operations xfs_super_operations = {
-	.alloc_inode		= xfs_fs_alloc_inode,
-	.destroy_inode		= xfs_fs_destroy_inode,
-	.dirty_inode		= xfs_fs_dirty_inode,
-	.drop_inode		= xfs_fs_drop_inode,
-	.put_super		= xfs_fs_put_super,
-	.sync_fs		= xfs_fs_sync_fs,
-	.freeze_fs		= xfs_fs_freeze,
-	.unfreeze_fs		= xfs_fs_unfreeze,
-	.statfs			= xfs_fs_statfs,
-	.remount_fs		= xfs_fs_remount,
-	.show_options		= xfs_fs_show_options,
-	.nr_cached_objects	= xfs_fs_nr_cached_objects,
-	.free_cached_objects	= xfs_fs_free_cached_objects,
+static const struct fs_context_operations xfs_context_ops = {
+	.parse_param = xfs_fc_parse_param,
+	.get_tree    = xfs_fc_get_tree,
+	.reconfigure = xfs_fc_reconfigure,
+	.free        = xfs_fc_free,
 };
 
+static int xfs_init_fs_context(
+	struct fs_context	*fc)
+{
+	struct xfs_mount	*mp;
+
+	mp = kmem_alloc(sizeof(struct xfs_mount), KM_ZERO);
+	if (!mp)
+		return -ENOMEM;
+
+	spin_lock_init(&mp->m_sb_lock);
+	spin_lock_init(&mp->m_agirotor_lock);
+	INIT_RADIX_TREE(&mp->m_perag_tree, GFP_ATOMIC);
+	spin_lock_init(&mp->m_perag_lock);
+	mutex_init(&mp->m_growlock);
+	atomic_set(&mp->m_active_trans, 0);
+	INIT_WORK(&mp->m_flush_inodes_work, xfs_flush_inodes_worker);
+	INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker);
+	INIT_DELAYED_WORK(&mp->m_eofblocks_work, xfs_eofblocks_worker);
+	INIT_DELAYED_WORK(&mp->m_cowblocks_work, xfs_cowblocks_worker);
+	mp->m_kobj.kobject.kset = xfs_kset;
+	/*
+	 * We don't create the finobt per-ag space reservation until after log
+	 * recovery, so we must set this to true so that an ifree transaction
+	 * started during log recovery will not depend on space reservations
+	 * for finobt expansion.
+	 */
+	mp->m_finobt_nores = true;
+
+	/*
+	 * These can be overridden by the mount option parsing.
+	 */
+	mp->m_logbufs = -1;
+	mp->m_logbsize = -1;
+	mp->m_allocsize_log = 16; /* 64k */
+
+	/*
+	 * Copy binary VFS mount flags we are interested in.
+	 */
+	if (fc->sb_flags & SB_RDONLY)
+		mp->m_flags |= XFS_MOUNT_RDONLY;
+	if (fc->sb_flags & SB_DIRSYNC)
+		mp->m_flags |= XFS_MOUNT_DIRSYNC;
+	if (fc->sb_flags & SB_SYNCHRONOUS)
+		mp->m_flags |= XFS_MOUNT_WSYNC;
+
+	fc->s_fs_info = mp;
+	fc->ops = &xfs_context_ops;
+
+	return 0;
+}
+
 static struct file_system_type xfs_fs_type = {
 	.owner			= THIS_MODULE,
 	.name			= "xfs",
-	.mount			= xfs_fs_mount,
+	.init_fs_context	= xfs_init_fs_context,
+	.parameters		= xfs_fs_parameters,
 	.kill_sb		= kill_block_super,
 	.fs_flags		= FS_REQUIRES_DEV,
 };
@@ -1853,37 +1804,39 @@ MODULE_ALIAS_FS("xfs");
 STATIC int __init
 xfs_init_zones(void)
 {
-	if (bioset_init(&xfs_ioend_bioset, 4 * (PAGE_SIZE / SECTOR_SIZE),
-			offsetof(struct xfs_ioend, io_inline_bio),
-			BIOSET_NEED_BVECS))
-		goto out;
-
-	xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t),
-						"xfs_log_ticket");
+	xfs_log_ticket_zone = kmem_cache_create("xfs_log_ticket",
+						sizeof(struct xlog_ticket),
+						0, 0, NULL);
 	if (!xfs_log_ticket_zone)
-		goto out_free_ioend_bioset;
+		goto out;
 
-	xfs_bmap_free_item_zone = kmem_zone_init(
-			sizeof(struct xfs_extent_free_item),
-			"xfs_bmap_free_item");
+	xfs_bmap_free_item_zone = kmem_cache_create("xfs_bmap_free_item",
+					sizeof(struct xfs_extent_free_item),
+					0, 0, NULL);
 	if (!xfs_bmap_free_item_zone)
 		goto out_destroy_log_ticket_zone;
 
-	xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t),
-						"xfs_btree_cur");
+	xfs_btree_cur_zone = kmem_cache_create("xfs_btree_cur",
+					       sizeof(struct xfs_btree_cur),
+					       0, 0, NULL);
 	if (!xfs_btree_cur_zone)
 		goto out_destroy_bmap_free_item_zone;
 
-	xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t),
-						"xfs_da_state");
+	xfs_da_state_zone = kmem_cache_create("xfs_da_state",
+					      sizeof(struct xfs_da_state),
+					      0, 0, NULL);
 	if (!xfs_da_state_zone)
 		goto out_destroy_btree_cur_zone;
 
-	xfs_ifork_zone = kmem_zone_init(sizeof(struct xfs_ifork), "xfs_ifork");
+	xfs_ifork_zone = kmem_cache_create("xfs_ifork",
+					   sizeof(struct xfs_ifork),
+					   0, 0, NULL);
 	if (!xfs_ifork_zone)
 		goto out_destroy_da_state_zone;
 
-	xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans");
+	xfs_trans_zone = kmem_cache_create("xf_trans",
+					   sizeof(struct xfs_trans),
+					   0, 0, NULL);
 	if (!xfs_trans_zone)
 		goto out_destroy_ifork_zone;
 
@@ -1893,111 +1846,122 @@ xfs_init_zones(void)
 	 * size possible under XFS.  This wastes a little bit of memory,
 	 * but it is much faster.
 	 */
-	xfs_buf_item_zone = kmem_zone_init(sizeof(struct xfs_buf_log_item),
-					   "xfs_buf_item");
+	xfs_buf_item_zone = kmem_cache_create("xfs_buf_item",
+					      sizeof(struct xfs_buf_log_item),
+					      0, 0, NULL);
 	if (!xfs_buf_item_zone)
 		goto out_destroy_trans_zone;
 
-	xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) +
-			((XFS_EFD_MAX_FAST_EXTENTS - 1) *
-				 sizeof(xfs_extent_t))), "xfs_efd_item");
+	xfs_efd_zone = kmem_cache_create("xfs_efd_item",
+					(sizeof(struct xfs_efd_log_item) +
+					(XFS_EFD_MAX_FAST_EXTENTS - 1) *
+					sizeof(struct xfs_extent)),
+					0, 0, NULL);
 	if (!xfs_efd_zone)
 		goto out_destroy_buf_item_zone;
 
-	xfs_efi_zone = kmem_zone_init((sizeof(xfs_efi_log_item_t) +
-			((XFS_EFI_MAX_FAST_EXTENTS - 1) *
-				sizeof(xfs_extent_t))), "xfs_efi_item");
+	xfs_efi_zone = kmem_cache_create("xfs_efi_item",
+					 (sizeof(struct xfs_efi_log_item) +
+					 (XFS_EFI_MAX_FAST_EXTENTS - 1) *
+					 sizeof(struct xfs_extent)),
+					 0, 0, NULL);
 	if (!xfs_efi_zone)
 		goto out_destroy_efd_zone;
 
-	xfs_inode_zone =
-		kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode",
-			KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD |
-			KM_ZONE_ACCOUNT, xfs_fs_inode_init_once);
+	xfs_inode_zone = kmem_cache_create("xfs_inode",
+					   sizeof(struct xfs_inode), 0,
+					   (SLAB_HWCACHE_ALIGN |
+					    SLAB_RECLAIM_ACCOUNT |
+					    SLAB_MEM_SPREAD | SLAB_ACCOUNT),
+					   xfs_fs_inode_init_once);
 	if (!xfs_inode_zone)
 		goto out_destroy_efi_zone;
 
-	xfs_ili_zone =
-		kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili",
-					KM_ZONE_SPREAD, NULL);
+	xfs_ili_zone = kmem_cache_create("xfs_ili",
+					 sizeof(struct xfs_inode_log_item), 0,
+					 SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
+					 NULL);
 	if (!xfs_ili_zone)
 		goto out_destroy_inode_zone;
-	xfs_icreate_zone = kmem_zone_init(sizeof(struct xfs_icreate_item),
-					"xfs_icr");
+
+	xfs_icreate_zone = kmem_cache_create("xfs_icr",
+					     sizeof(struct xfs_icreate_item),
+					     0, 0, NULL);
 	if (!xfs_icreate_zone)
 		goto out_destroy_ili_zone;
 
-	xfs_rud_zone = kmem_zone_init(sizeof(struct xfs_rud_log_item),
-			"xfs_rud_item");
+	xfs_rud_zone = kmem_cache_create("xfs_rud_item",
+					 sizeof(struct xfs_rud_log_item),
+					 0, 0, NULL);
 	if (!xfs_rud_zone)
 		goto out_destroy_icreate_zone;
 
-	xfs_rui_zone = kmem_zone_init(
+	xfs_rui_zone = kmem_cache_create("xfs_rui_item",
 			xfs_rui_log_item_sizeof(XFS_RUI_MAX_FAST_EXTENTS),
-			"xfs_rui_item");
+			0, 0, NULL);
 	if (!xfs_rui_zone)
 		goto out_destroy_rud_zone;
 
-	xfs_cud_zone = kmem_zone_init(sizeof(struct xfs_cud_log_item),
-			"xfs_cud_item");
+	xfs_cud_zone = kmem_cache_create("xfs_cud_item",
+					 sizeof(struct xfs_cud_log_item),
+					 0, 0, NULL);
 	if (!xfs_cud_zone)
 		goto out_destroy_rui_zone;
 
-	xfs_cui_zone = kmem_zone_init(
+	xfs_cui_zone = kmem_cache_create("xfs_cui_item",
 			xfs_cui_log_item_sizeof(XFS_CUI_MAX_FAST_EXTENTS),
-			"xfs_cui_item");
+			0, 0, NULL);
 	if (!xfs_cui_zone)
 		goto out_destroy_cud_zone;
 
-	xfs_bud_zone = kmem_zone_init(sizeof(struct xfs_bud_log_item),
-			"xfs_bud_item");
+	xfs_bud_zone = kmem_cache_create("xfs_bud_item",
+					 sizeof(struct xfs_bud_log_item),
+					 0, 0, NULL);
 	if (!xfs_bud_zone)
 		goto out_destroy_cui_zone;
 
-	xfs_bui_zone = kmem_zone_init(
+	xfs_bui_zone = kmem_cache_create("xfs_bui_item",
 			xfs_bui_log_item_sizeof(XFS_BUI_MAX_FAST_EXTENTS),
-			"xfs_bui_item");
+			0, 0, NULL);
 	if (!xfs_bui_zone)
 		goto out_destroy_bud_zone;
 
 	return 0;
 
  out_destroy_bud_zone:
-	kmem_zone_destroy(xfs_bud_zone);
+	kmem_cache_destroy(xfs_bud_zone);
  out_destroy_cui_zone:
-	kmem_zone_destroy(xfs_cui_zone);
+	kmem_cache_destroy(xfs_cui_zone);
  out_destroy_cud_zone:
-	kmem_zone_destroy(xfs_cud_zone);
+	kmem_cache_destroy(xfs_cud_zone);
  out_destroy_rui_zone:
-	kmem_zone_destroy(xfs_rui_zone);
+	kmem_cache_destroy(xfs_rui_zone);
  out_destroy_rud_zone:
-	kmem_zone_destroy(xfs_rud_zone);
+	kmem_cache_destroy(xfs_rud_zone);
  out_destroy_icreate_zone:
-	kmem_zone_destroy(xfs_icreate_zone);
+	kmem_cache_destroy(xfs_icreate_zone);
  out_destroy_ili_zone:
-	kmem_zone_destroy(xfs_ili_zone);
+	kmem_cache_destroy(xfs_ili_zone);
  out_destroy_inode_zone:
-	kmem_zone_destroy(xfs_inode_zone);
+	kmem_cache_destroy(xfs_inode_zone);
  out_destroy_efi_zone:
-	kmem_zone_destroy(xfs_efi_zone);
+	kmem_cache_destroy(xfs_efi_zone);
  out_destroy_efd_zone:
-	kmem_zone_destroy(xfs_efd_zone);
+	kmem_cache_destroy(xfs_efd_zone);
  out_destroy_buf_item_zone:
-	kmem_zone_destroy(xfs_buf_item_zone);
+	kmem_cache_destroy(xfs_buf_item_zone);
  out_destroy_trans_zone:
-	kmem_zone_destroy(xfs_trans_zone);
+	kmem_cache_destroy(xfs_trans_zone);
  out_destroy_ifork_zone:
-	kmem_zone_destroy(xfs_ifork_zone);
+	kmem_cache_destroy(xfs_ifork_zone);
  out_destroy_da_state_zone:
-	kmem_zone_destroy(xfs_da_state_zone);
+	kmem_cache_destroy(xfs_da_state_zone);
  out_destroy_btree_cur_zone:
-	kmem_zone_destroy(xfs_btree_cur_zone);
+	kmem_cache_destroy(xfs_btree_cur_zone);
  out_destroy_bmap_free_item_zone:
-	kmem_zone_destroy(xfs_bmap_free_item_zone);
+	kmem_cache_destroy(xfs_bmap_free_item_zone);
  out_destroy_log_ticket_zone:
-	kmem_zone_destroy(xfs_log_ticket_zone);
- out_free_ioend_bioset:
-	bioset_exit(&xfs_ioend_bioset);
+	kmem_cache_destroy(xfs_log_ticket_zone);
  out:
 	return -ENOMEM;
 }
@@ -2010,25 +1974,24 @@ xfs_destroy_zones(void)
 	 * destroy caches.
 	 */
 	rcu_barrier();
-	kmem_zone_destroy(xfs_bui_zone);
-	kmem_zone_destroy(xfs_bud_zone);
-	kmem_zone_destroy(xfs_cui_zone);
-	kmem_zone_destroy(xfs_cud_zone);
-	kmem_zone_destroy(xfs_rui_zone);
-	kmem_zone_destroy(xfs_rud_zone);
-	kmem_zone_destroy(xfs_icreate_zone);
-	kmem_zone_destroy(xfs_ili_zone);
-	kmem_zone_destroy(xfs_inode_zone);
-	kmem_zone_destroy(xfs_efi_zone);
-	kmem_zone_destroy(xfs_efd_zone);
-	kmem_zone_destroy(xfs_buf_item_zone);
-	kmem_zone_destroy(xfs_trans_zone);
-	kmem_zone_destroy(xfs_ifork_zone);
-	kmem_zone_destroy(xfs_da_state_zone);
-	kmem_zone_destroy(xfs_btree_cur_zone);
-	kmem_zone_destroy(xfs_bmap_free_item_zone);
-	kmem_zone_destroy(xfs_log_ticket_zone);
-	bioset_exit(&xfs_ioend_bioset);
+	kmem_cache_destroy(xfs_bui_zone);
+	kmem_cache_destroy(xfs_bud_zone);
+	kmem_cache_destroy(xfs_cui_zone);
+	kmem_cache_destroy(xfs_cud_zone);
+	kmem_cache_destroy(xfs_rui_zone);
+	kmem_cache_destroy(xfs_rud_zone);
+	kmem_cache_destroy(xfs_icreate_zone);
+	kmem_cache_destroy(xfs_ili_zone);
+	kmem_cache_destroy(xfs_inode_zone);
+	kmem_cache_destroy(xfs_efi_zone);
+	kmem_cache_destroy(xfs_efd_zone);
+	kmem_cache_destroy(xfs_buf_item_zone);
+	kmem_cache_destroy(xfs_trans_zone);
+	kmem_cache_destroy(xfs_ifork_zone);
+	kmem_cache_destroy(xfs_da_state_zone);
+	kmem_cache_destroy(xfs_btree_cur_zone);
+	kmem_cache_destroy(xfs_bmap_free_item_zone);
+	kmem_cache_destroy(xfs_log_ticket_zone);
 }
 
 STATIC int __init
diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h
index 763e43d22dee..b552cf6d3379 100644
--- a/fs/xfs/xfs_super.h
+++ b/fs/xfs/xfs_super.h
@@ -11,9 +11,11 @@
 #ifdef CONFIG_XFS_QUOTA
 extern int xfs_qm_init(void);
 extern void xfs_qm_exit(void);
+# define XFS_QUOTA_STRING	"quota, "
 #else
 # define xfs_qm_init()	(0)
 # define xfs_qm_exit()	do { } while (0)
+# define XFS_QUOTA_STRING
 #endif
 
 #ifdef CONFIG_XFS_POSIX_ACL
@@ -50,6 +52,12 @@ extern void xfs_qm_exit(void);
 # define XFS_WARN_STRING
 #endif
 
+#ifdef CONFIG_XFS_ASSERT_FATAL
+# define XFS_ASSERT_FATAL_STRING	"fatal assert, "
+#else
+# define XFS_ASSERT_FATAL_STRING
+#endif
+
 #ifdef DEBUG
 # define XFS_DBG_STRING		"debug"
 #else
@@ -63,6 +71,8 @@ extern void xfs_qm_exit(void);
 				XFS_SCRUB_STRING \
 				XFS_REPAIR_STRING \
 				XFS_WARN_STRING \
+				XFS_QUOTA_STRING \
+				XFS_ASSERT_FATAL_STRING \
 				XFS_DBG_STRING /* DBG must be last */
 
 struct xfs_inode;
diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c
index ed66fd2de327..13fb4b919648 100644
--- a/fs/xfs/xfs_symlink.c
+++ b/fs/xfs/xfs_symlink.c
@@ -17,6 +17,7 @@
 #include "xfs_bmap.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_quota.h"
+#include "xfs_symlink.h"
 #include "xfs_trans_space.h"
 #include "xfs_trace.h"
 #include "xfs_trans.h"
@@ -52,20 +53,10 @@ xfs_readlink_bmap_ilocked(
 		d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock);
 		byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount);
 
-		bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0,
-				  &xfs_symlink_buf_ops);
-		if (!bp)
-			return -ENOMEM;
-		error = bp->b_error;
-		if (error) {
-			xfs_buf_ioerror_alert(bp, __func__);
-			xfs_buf_relse(bp);
-
-			/* bad CRC means corrupted metadata */
-			if (error == -EFSBADCRC)
-				error = -EFSCORRUPTED;
-			goto out;
-		}
+		error = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0,
+				&bp, &xfs_symlink_buf_ops);
+		if (error)
+			return error;
 		byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt);
 		if (pathlen < byte_cnt)
 			byte_cnt = pathlen;
@@ -185,15 +176,12 @@ xfs_symlink(
 		return -ENAMETOOLONG;
 	ASSERT(pathlen > 0);
 
-	udqp = gdqp = NULL;
 	prid = xfs_get_initial_prid(dp);
 
 	/*
 	 * Make sure that we have allocated dquot(s) on disk.
 	 */
-	error = xfs_qm_vop_dqalloc(dp,
-			xfs_kuid_to_uid(current_fsuid()),
-			xfs_kgid_to_gid(current_fsgid()), prid,
+	error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid,
 			XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT,
 			&udqp, &gdqp, &pdqp);
 	if (error)
@@ -203,7 +191,7 @@ xfs_symlink(
 	 * The symlink will fit into the inode data fork?
 	 * There can't be any attributes so we get the whole variable part.
 	 */
-	if (pathlen <= XFS_LITINO(mp, dp->i_d.di_version))
+	if (pathlen <= XFS_LITINO(mp))
 		fs_blocks = 0;
 	else
 		fs_blocks = xfs_symlink_blocks(mp, pathlen);
@@ -289,12 +277,10 @@ xfs_symlink(
 
 			d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock);
 			byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount);
-			bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
-					       BTOBB(byte_cnt), 0);
-			if (!bp) {
-				error = -ENOMEM;
+			error = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
+					       BTOBB(byte_cnt), 0, &bp);
+			if (error)
 				goto out_trans_cancel;
-			}
 			bp->b_ops = &xfs_symlink_buf_ops;
 
 			byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt);
@@ -432,13 +418,12 @@ xfs_inactive_symlink_rmt(
 	 * Invalidate the block(s). No validation is done.
 	 */
 	for (i = 0; i < nmaps; i++) {
-		bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
-			XFS_FSB_TO_DADDR(mp, mval[i].br_startblock),
-			XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0);
-		if (!bp) {
-			error = -ENOMEM;
+		error = xfs_trans_get_buf(tp, mp->m_ddev_targp,
+				XFS_FSB_TO_DADDR(mp, mval[i].br_startblock),
+				XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0,
+				&bp);
+		if (error)
 			goto error_trans_cancel;
-		}
 		xfs_trans_binval(tp, bp);
 	}
 	/*
diff --git a/fs/xfs/xfs_symlink.h b/fs/xfs/xfs_symlink.h
index 9743d8c9394b..b1fa091427e6 100644
--- a/fs/xfs/xfs_symlink.h
+++ b/fs/xfs/xfs_symlink.h
@@ -5,7 +5,7 @@
 #ifndef __XFS_SYMLINK_H
 #define __XFS_SYMLINK_H 1
 
-/* Kernel only symlink defintions */
+/* Kernel only symlink definitions */
 
 int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name,
 		const char *target_path, umode_t mode, struct xfs_inode **ipp);
diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
index bc85b89f88ca..120398a37c2a 100644
--- a/fs/xfs/xfs_trace.c
+++ b/fs/xfs/xfs_trace.c
@@ -6,6 +6,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_shared.h"
+#include "xfs_bit.h"
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
@@ -27,6 +28,7 @@
 #include "xfs_log_recover.h"
 #include "xfs_filestream.h"
 #include "xfs_fsmap.h"
+#include "xfs_btree_staging.h"
 
 /*
  * We include this last to have the helpers above available for the trace
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index eaae275ed430..a4323a63438d 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -35,6 +35,12 @@ struct xfs_icreate_log;
 struct xfs_owner_info;
 struct xfs_trans_res;
 struct xfs_inobt_rec_incore;
+union xfs_btree_ptr;
+
+#define XFS_ATTR_FILTER_FLAGS \
+	{ XFS_ATTR_ROOT,	"ROOT" }, \
+	{ XFS_ATTR_SECURE,	"SECURE" }, \
+	{ XFS_ATTR_INCOMPLETE,	"INCOMPLETE" }
 
 DECLARE_EVENT_CLASS(xfs_attr_list_class,
 	TP_PROTO(struct xfs_attr_list_context *ctx),
@@ -45,39 +51,39 @@ DECLARE_EVENT_CLASS(xfs_attr_list_class,
 		__field(u32, hashval)
 		__field(u32, blkno)
 		__field(u32, offset)
-		__field(void *, alist)
+		__field(void *, buffer)
 		__field(int, bufsize)
 		__field(int, count)
 		__field(int, firstu)
 		__field(int, dupcnt)
-		__field(int, flags)
+		__field(unsigned int, attr_filter)
 	),
 	TP_fast_assign(
 		__entry->dev = VFS_I(ctx->dp)->i_sb->s_dev;
 		__entry->ino = ctx->dp->i_ino;
-		__entry->hashval = ctx->cursor->hashval;
-		__entry->blkno = ctx->cursor->blkno;
-		__entry->offset = ctx->cursor->offset;
-		__entry->alist = ctx->alist;
+		__entry->hashval = ctx->cursor.hashval;
+		__entry->blkno = ctx->cursor.blkno;
+		__entry->offset = ctx->cursor.offset;
+		__entry->buffer = ctx->buffer;
 		__entry->bufsize = ctx->bufsize;
 		__entry->count = ctx->count;
 		__entry->firstu = ctx->firstu;
-		__entry->flags = ctx->flags;
+		__entry->attr_filter = ctx->attr_filter;
 	),
 	TP_printk("dev %d:%d ino 0x%llx cursor h/b/o 0x%x/0x%x/%u dupcnt %u "
-		  "alist %p size %u count %u firstu %u flags %d %s",
+		  "buffer %p size %u count %u firstu %u filter %s",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		   __entry->ino,
 		   __entry->hashval,
 		   __entry->blkno,
 		   __entry->offset,
 		   __entry->dupcnt,
-		   __entry->alist,
+		   __entry->buffer,
 		   __entry->bufsize,
 		   __entry->count,
 		   __entry->firstu,
-		   __entry->flags,
-		   __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS)
+		   __print_flags(__entry->attr_filter, "|",
+				 XFS_ATTR_FILTER_FLAGS)
 	)
 )
 
@@ -169,31 +175,31 @@ TRACE_EVENT(xfs_attr_list_node_descend,
 		__field(u32, hashval)
 		__field(u32, blkno)
 		__field(u32, offset)
-		__field(void *, alist)
+		__field(void *, buffer)
 		__field(int, bufsize)
 		__field(int, count)
 		__field(int, firstu)
 		__field(int, dupcnt)
-		__field(int, flags)
+		__field(unsigned int, attr_filter)
 		__field(u32, bt_hashval)
 		__field(u32, bt_before)
 	),
 	TP_fast_assign(
 		__entry->dev = VFS_I(ctx->dp)->i_sb->s_dev;
 		__entry->ino = ctx->dp->i_ino;
-		__entry->hashval = ctx->cursor->hashval;
-		__entry->blkno = ctx->cursor->blkno;
-		__entry->offset = ctx->cursor->offset;
-		__entry->alist = ctx->alist;
+		__entry->hashval = ctx->cursor.hashval;
+		__entry->blkno = ctx->cursor.blkno;
+		__entry->offset = ctx->cursor.offset;
+		__entry->buffer = ctx->buffer;
 		__entry->bufsize = ctx->bufsize;
 		__entry->count = ctx->count;
 		__entry->firstu = ctx->firstu;
-		__entry->flags = ctx->flags;
+		__entry->attr_filter = ctx->attr_filter;
 		__entry->bt_hashval = be32_to_cpu(btree->hashval);
 		__entry->bt_before = be32_to_cpu(btree->before);
 	),
 	TP_printk("dev %d:%d ino 0x%llx cursor h/b/o 0x%x/0x%x/%u dupcnt %u "
-		  "alist %p size %u count %u firstu %u flags %d %s "
+		  "buffer %p size %u count %u firstu %u filter %s "
 		  "node hashval %u, node before %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		   __entry->ino,
@@ -201,12 +207,12 @@ TRACE_EVENT(xfs_attr_list_node_descend,
 		   __entry->blkno,
 		   __entry->offset,
 		   __entry->dupcnt,
-		   __entry->alist,
+		   __entry->buffer,
 		   __entry->bufsize,
 		   __entry->count,
 		   __entry->firstu,
-		   __entry->flags,
-		   __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS),
+		   __print_flags(__entry->attr_filter, "|",
+				 XFS_ATTR_FILTER_FLAGS),
 		   __entry->bt_hashval,
 		   __entry->bt_before)
 );
@@ -218,8 +224,8 @@ DECLARE_EVENT_CLASS(xfs_bmap_class,
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
 		__field(xfs_ino_t, ino)
-		__field(void *, leaf);
-		__field(int, pos);
+		__field(void *, leaf)
+		__field(int, pos)
 		__field(xfs_fileoff_t, startoff)
 		__field(xfs_fsblock_t, startblock)
 		__field(xfs_filblks_t, blockcount)
@@ -725,7 +731,7 @@ TRACE_EVENT(xfs_iomap_prealloc_size,
 		__entry->writeio_blocks = writeio_blocks;
 	),
 	TP_printk("dev %d:%d ino 0x%llx prealloc blocks %llu shift %d "
-		  "m_writeio_blocks %u",
+		  "m_allocsize_blocks %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino,
 		  __entry->blocks, __entry->shift, __entry->writeio_blocks)
 )
@@ -995,8 +1001,6 @@ DECLARE_EVENT_CLASS(xfs_loggrant_class,
 DEFINE_EVENT(xfs_loggrant_class, name, \
 	TP_PROTO(struct xlog *log, struct xlog_ticket *tic), \
 	TP_ARGS(log, tic))
-DEFINE_LOGGRANT_EVENT(xfs_log_done_nonperm);
-DEFINE_LOGGRANT_EVENT(xfs_log_done_perm);
 DEFINE_LOGGRANT_EVENT(xfs_log_umount_write);
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep);
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake);
@@ -1005,12 +1009,13 @@ DEFINE_LOGGRANT_EVENT(xfs_log_reserve);
 DEFINE_LOGGRANT_EVENT(xfs_log_reserve_exit);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_exit);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_enter);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_exit);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_sub);
-DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_enter);
-DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_exit);
-DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_sub);
+DEFINE_LOGGRANT_EVENT(xfs_log_ticket_regrant);
+DEFINE_LOGGRANT_EVENT(xfs_log_ticket_regrant_exit);
+DEFINE_LOGGRANT_EVENT(xfs_log_ticket_regrant_sub);
+DEFINE_LOGGRANT_EVENT(xfs_log_ticket_ungrant);
+DEFINE_LOGGRANT_EVENT(xfs_log_ticket_ungrant_sub);
+DEFINE_LOGGRANT_EVENT(xfs_log_ticket_ungrant_exit);
+DEFINE_LOGGRANT_EVENT(xfs_log_cil_wait);
 
 DECLARE_EVENT_CLASS(xfs_log_item_class,
 	TP_PROTO(struct xfs_log_item *lip),
@@ -1158,71 +1163,6 @@ DEFINE_RW_EVENT(xfs_file_buffered_write);
 DEFINE_RW_EVENT(xfs_file_direct_write);
 DEFINE_RW_EVENT(xfs_file_dax_write);
 
-DECLARE_EVENT_CLASS(xfs_page_class,
-	TP_PROTO(struct inode *inode, struct page *page, unsigned long off,
-		 unsigned int len),
-	TP_ARGS(inode, page, off, len),
-	TP_STRUCT__entry(
-		__field(dev_t, dev)
-		__field(xfs_ino_t, ino)
-		__field(pgoff_t, pgoff)
-		__field(loff_t, size)
-		__field(unsigned long, offset)
-		__field(unsigned int, length)
-	),
-	TP_fast_assign(
-		__entry->dev = inode->i_sb->s_dev;
-		__entry->ino = XFS_I(inode)->i_ino;
-		__entry->pgoff = page_offset(page);
-		__entry->size = i_size_read(inode);
-		__entry->offset = off;
-		__entry->length = len;
-	),
-	TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx "
-		  "length %x",
-		  MAJOR(__entry->dev), MINOR(__entry->dev),
-		  __entry->ino,
-		  __entry->pgoff,
-		  __entry->size,
-		  __entry->offset,
-		  __entry->length)
-)
-
-#define DEFINE_PAGE_EVENT(name)		\
-DEFINE_EVENT(xfs_page_class, name,	\
-	TP_PROTO(struct inode *inode, struct page *page, unsigned long off, \
-		 unsigned int len),	\
-	TP_ARGS(inode, page, off, len))
-DEFINE_PAGE_EVENT(xfs_writepage);
-DEFINE_PAGE_EVENT(xfs_releasepage);
-DEFINE_PAGE_EVENT(xfs_invalidatepage);
-
-DECLARE_EVENT_CLASS(xfs_readpage_class,
-	TP_PROTO(struct inode *inode, int nr_pages),
-	TP_ARGS(inode, nr_pages),
-	TP_STRUCT__entry(
-		__field(dev_t, dev)
-		__field(xfs_ino_t, ino)
-		__field(int, nr_pages)
-	),
-	TP_fast_assign(
-		__entry->dev = inode->i_sb->s_dev;
-		__entry->ino = inode->i_ino;
-		__entry->nr_pages = nr_pages;
-	),
-	TP_printk("dev %d:%d ino 0x%llx nr_pages %d",
-		  MAJOR(__entry->dev), MINOR(__entry->dev),
-		  __entry->ino,
-		  __entry->nr_pages)
-)
-
-#define DEFINE_READPAGE_EVENT(name)		\
-DEFINE_EVENT(xfs_readpage_class, name,	\
-	TP_PROTO(struct inode *inode, int nr_pages), \
-	TP_ARGS(inode, nr_pages))
-DEFINE_READPAGE_EVENT(xfs_vm_readpage);
-DEFINE_READPAGE_EVENT(xfs_vm_readpages);
-
 DECLARE_EVENT_CLASS(xfs_imap_class,
 	TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count,
 		 int whichfork, struct xfs_bmbt_irec *irec),
@@ -1642,8 +1582,11 @@ DEFINE_ALLOC_EVENT(xfs_alloc_exact_notfound);
 DEFINE_ALLOC_EVENT(xfs_alloc_exact_error);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_nominleft);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_first);
-DEFINE_ALLOC_EVENT(xfs_alloc_near_greater);
-DEFINE_ALLOC_EVENT(xfs_alloc_near_lesser);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur_right);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur_left);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur_lookup);
+DEFINE_ALLOC_EVENT(xfs_alloc_cur_lookup_done);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_error);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_noentry);
 DEFINE_ALLOC_EVENT(xfs_alloc_near_busy);
@@ -1663,6 +1606,32 @@ DEFINE_ALLOC_EVENT(xfs_alloc_vextent_noagbp);
 DEFINE_ALLOC_EVENT(xfs_alloc_vextent_loopfailed);
 DEFINE_ALLOC_EVENT(xfs_alloc_vextent_allfailed);
 
+TRACE_EVENT(xfs_alloc_cur_check,
+	TP_PROTO(struct xfs_mount *mp, xfs_btnum_t btnum, xfs_agblock_t bno,
+		 xfs_extlen_t len, xfs_extlen_t diff, bool new),
+	TP_ARGS(mp, btnum, bno, len, diff, new),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_btnum_t, btnum)
+		__field(xfs_agblock_t, bno)
+		__field(xfs_extlen_t, len)
+		__field(xfs_extlen_t, diff)
+		__field(bool, new)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->btnum = btnum;
+		__entry->bno = bno;
+		__entry->len = len;
+		__entry->diff = diff;
+		__entry->new = new;
+	),
+	TP_printk("dev %d:%d btree %s bno 0x%x len 0x%x diff 0x%x new %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
+		  __entry->bno, __entry->len, __entry->diff, __entry->new)
+)
+
 DECLARE_EVENT_CLASS(xfs_da_class,
 	TP_PROTO(struct xfs_da_args *args),
 	TP_ARGS(args),
@@ -1737,7 +1706,8 @@ DECLARE_EVENT_CLASS(xfs_attr_class,
 		__field(int, namelen)
 		__field(int, valuelen)
 		__field(xfs_dahash_t, hashval)
-		__field(int, flags)
+		__field(unsigned int, attr_filter)
+		__field(unsigned int, attr_flags)
 		__field(int, op_flags)
 	),
 	TP_fast_assign(
@@ -1748,11 +1718,12 @@ DECLARE_EVENT_CLASS(xfs_attr_class,
 		__entry->namelen = args->namelen;
 		__entry->valuelen = args->valuelen;
 		__entry->hashval = args->hashval;
-		__entry->flags = args->flags;
+		__entry->attr_filter = args->attr_filter;
+		__entry->attr_flags = args->attr_flags;
 		__entry->op_flags = args->op_flags;
 	),
 	TP_printk("dev %d:%d ino 0x%llx name %.*s namelen %d valuelen %d "
-		  "hashval 0x%x flags %s op_flags %s",
+		  "hashval 0x%x filter %s flags %s op_flags %s",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->ino,
 		  __entry->namelen,
@@ -1760,7 +1731,11 @@ DECLARE_EVENT_CLASS(xfs_attr_class,
 		  __entry->namelen,
 		  __entry->valuelen,
 		  __entry->hashval,
-		  __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS),
+		  __print_flags(__entry->attr_filter, "|",
+				XFS_ATTR_FILTER_FLAGS),
+		   __print_flags(__entry->attr_flags, "|",
+				{ XATTR_CREATE,		"CREATE" },
+				{ XATTR_REPLACE,	"REPLACE" }),
 		  __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS))
 )
 
@@ -3609,6 +3584,172 @@ DEFINE_KMEM_EVENT(kmem_alloc_large);
 DEFINE_KMEM_EVENT(kmem_realloc);
 DEFINE_KMEM_EVENT(kmem_zone_alloc);
 
+TRACE_EVENT(xfs_check_new_dalign,
+	TP_PROTO(struct xfs_mount *mp, int new_dalign, xfs_ino_t calc_rootino),
+	TP_ARGS(mp, new_dalign, calc_rootino),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(int, new_dalign)
+		__field(xfs_ino_t, sb_rootino)
+		__field(xfs_ino_t, calc_rootino)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->new_dalign = new_dalign;
+		__entry->sb_rootino = mp->m_sb.sb_rootino;
+		__entry->calc_rootino = calc_rootino;
+	),
+	TP_printk("dev %d:%d new_dalign %d sb_rootino %llu calc_rootino %llu",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->new_dalign, __entry->sb_rootino,
+		  __entry->calc_rootino)
+)
+
+TRACE_EVENT(xfs_btree_commit_afakeroot,
+	TP_PROTO(struct xfs_btree_cur *cur),
+	TP_ARGS(cur),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_btnum_t, btnum)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, agbno)
+		__field(unsigned int, levels)
+		__field(unsigned int, blocks)
+	),
+	TP_fast_assign(
+		__entry->dev = cur->bc_mp->m_super->s_dev;
+		__entry->btnum = cur->bc_btnum;
+		__entry->agno = cur->bc_ag.agno;
+		__entry->agbno = cur->bc_ag.afake->af_root;
+		__entry->levels = cur->bc_ag.afake->af_levels;
+		__entry->blocks = cur->bc_ag.afake->af_blocks;
+	),
+	TP_printk("dev %d:%d btree %s ag %u levels %u blocks %u root %u",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
+		  __entry->agno,
+		  __entry->levels,
+		  __entry->blocks,
+		  __entry->agbno)
+)
+
+TRACE_EVENT(xfs_btree_commit_ifakeroot,
+	TP_PROTO(struct xfs_btree_cur *cur),
+	TP_ARGS(cur),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_btnum_t, btnum)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agino_t, agino)
+		__field(unsigned int, levels)
+		__field(unsigned int, blocks)
+		__field(int, whichfork)
+	),
+	TP_fast_assign(
+		__entry->dev = cur->bc_mp->m_super->s_dev;
+		__entry->btnum = cur->bc_btnum;
+		__entry->agno = XFS_INO_TO_AGNO(cur->bc_mp,
+					cur->bc_ino.ip->i_ino);
+		__entry->agino = XFS_INO_TO_AGINO(cur->bc_mp,
+					cur->bc_ino.ip->i_ino);
+		__entry->levels = cur->bc_ino.ifake->if_levels;
+		__entry->blocks = cur->bc_ino.ifake->if_blocks;
+		__entry->whichfork = cur->bc_ino.whichfork;
+	),
+	TP_printk("dev %d:%d btree %s ag %u agino %u whichfork %s levels %u blocks %u",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
+		  __entry->agno,
+		  __entry->agino,
+		  __entry->whichfork == XFS_ATTR_FORK ? "attr" : "data",
+		  __entry->levels,
+		  __entry->blocks)
+)
+
+TRACE_EVENT(xfs_btree_bload_level_geometry,
+	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
+		 uint64_t nr_this_level, unsigned int nr_per_block,
+		 unsigned int desired_npb, uint64_t blocks,
+		 uint64_t blocks_with_extra),
+	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
+		blocks_with_extra),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_btnum_t, btnum)
+		__field(unsigned int, level)
+		__field(unsigned int, nlevels)
+		__field(uint64_t, nr_this_level)
+		__field(unsigned int, nr_per_block)
+		__field(unsigned int, desired_npb)
+		__field(unsigned long long, blocks)
+		__field(unsigned long long, blocks_with_extra)
+	),
+	TP_fast_assign(
+		__entry->dev = cur->bc_mp->m_super->s_dev;
+		__entry->btnum = cur->bc_btnum;
+		__entry->level = level;
+		__entry->nlevels = cur->bc_nlevels;
+		__entry->nr_this_level = nr_this_level;
+		__entry->nr_per_block = nr_per_block;
+		__entry->desired_npb = desired_npb;
+		__entry->blocks = blocks;
+		__entry->blocks_with_extra = blocks_with_extra;
+	),
+	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
+		  __entry->level,
+		  __entry->nlevels,
+		  __entry->nr_this_level,
+		  __entry->nr_per_block,
+		  __entry->desired_npb,
+		  __entry->blocks,
+		  __entry->blocks_with_extra)
+)
+
+TRACE_EVENT(xfs_btree_bload_block,
+	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
+		 uint64_t block_idx, uint64_t nr_blocks,
+		 union xfs_btree_ptr *ptr, unsigned int nr_records),
+	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_btnum_t, btnum)
+		__field(unsigned int, level)
+		__field(unsigned long long, block_idx)
+		__field(unsigned long long, nr_blocks)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, agbno)
+		__field(unsigned int, nr_records)
+	),
+	TP_fast_assign(
+		__entry->dev = cur->bc_mp->m_super->s_dev;
+		__entry->btnum = cur->bc_btnum;
+		__entry->level = level;
+		__entry->block_idx = block_idx;
+		__entry->nr_blocks = nr_blocks;
+		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
+			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
+
+			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
+			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
+		} else {
+			__entry->agno = cur->bc_ag.agno;
+			__entry->agbno = be32_to_cpu(ptr->s);
+		}
+		__entry->nr_records = nr_records;
+	),
+	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
+		  __entry->level,
+		  __entry->block_idx,
+		  __entry->nr_blocks,
+		  __entry->agno,
+		  __entry->agbno,
+		  __entry->nr_records)
+)
+
 #endif /* _TRACE_XFS_H */
 
 #undef TRACE_INCLUDE_PATH
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index f4795fdb7389..28b983ff8b11 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -9,6 +9,7 @@
 #include "xfs_shared.h"
 #include "xfs_format.h"
 #include "xfs_log_format.h"
+#include "xfs_log_priv.h"
 #include "xfs_trans_resv.h"
 #include "xfs_mount.h"
 #include "xfs_extent_busy.h"
@@ -71,7 +72,7 @@ xfs_trans_free(
 	if (!(tp->t_flags & XFS_TRANS_NO_WRITECOUNT))
 		sb_end_intwrite(tp->t_mountp->m_super);
 	xfs_trans_free_dqinfo(tp);
-	kmem_zone_free(xfs_trans_zone, tp);
+	kmem_cache_free(xfs_trans_zone, tp);
 }
 
 /*
@@ -150,8 +151,9 @@ xfs_trans_reserve(
 	uint			blocks,
 	uint			rtextents)
 {
-	int		error = 0;
-	bool		rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0;
+	struct xfs_mount	*mp = tp->t_mountp;
+	int			error = 0;
+	bool			rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0;
 
 	/* Mark this thread as being in a transaction */
 	current_set_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS);
@@ -162,7 +164,7 @@ xfs_trans_reserve(
 	 * fail if the count would go below zero.
 	 */
 	if (blocks > 0) {
-		error = xfs_mod_fdblocks(tp->t_mountp, -((int64_t)blocks), rsvd);
+		error = xfs_mod_fdblocks(mp, -((int64_t)blocks), rsvd);
 		if (error != 0) {
 			current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS);
 			return -ENOSPC;
@@ -191,9 +193,9 @@ xfs_trans_reserve(
 
 		if (tp->t_ticket != NULL) {
 			ASSERT(resp->tr_logflags & XFS_TRANS_PERM_LOG_RES);
-			error = xfs_log_regrant(tp->t_mountp, tp->t_ticket);
+			error = xfs_log_regrant(mp, tp->t_ticket);
 		} else {
-			error = xfs_log_reserve(tp->t_mountp,
+			error = xfs_log_reserve(mp,
 						resp->tr_logres,
 						resp->tr_logcount,
 						&tp->t_ticket, XFS_TRANSACTION,
@@ -213,7 +215,7 @@ xfs_trans_reserve(
 	 * fail if the count would go below zero.
 	 */
 	if (rtextents > 0) {
-		error = xfs_mod_frextents(tp->t_mountp, -((int64_t)rtextents));
+		error = xfs_mod_frextents(mp, -((int64_t)rtextents));
 		if (error) {
 			error = -ENOSPC;
 			goto undo_log;
@@ -229,7 +231,7 @@ xfs_trans_reserve(
 	 */
 undo_log:
 	if (resp->tr_logres > 0) {
-		xfs_log_done(tp->t_mountp, tp->t_ticket, NULL, false);
+		xfs_log_ticket_ungrant(mp->m_log, tp->t_ticket);
 		tp->t_ticket = NULL;
 		tp->t_log_res = 0;
 		tp->t_flags &= ~XFS_TRANS_PERM_LOG_RES;
@@ -237,7 +239,7 @@ undo_log:
 
 undo_blocks:
 	if (blocks > 0) {
-		xfs_mod_fdblocks(tp->t_mountp, (int64_t)blocks, rsvd);
+		xfs_mod_fdblocks(mp, (int64_t)blocks, rsvd);
 		tp->t_blk_res = 0;
 	}
 
@@ -306,6 +308,11 @@ xfs_trans_alloc(
  *
  * Note the zero-length reservation; this transaction MUST be cancelled
  * without any dirty data.
+ *
+ * Callers should obtain freeze protection to avoid two conflicts with fs
+ * freezing: (1) having active transactions trip the m_active_trans ASSERTs;
+ * and (2) grabbing buffers at the same time that freeze is trying to drain
+ * the buffer LRU list.
  */
 int
 xfs_trans_alloc_empty(
@@ -450,7 +457,7 @@ xfs_trans_apply_sb_deltas(
 	int		whole = 0;
 
 	bp = xfs_trans_getsb(tp, tp->t_mountp);
-	sbp = XFS_BUF_TO_SBP(bp);
+	sbp = bp->b_addr;
 
 	/*
 	 * Check that superblock mods match the mods made to AGF counters.
@@ -999,9 +1006,10 @@ out_unreserve:
 	 */
 	xfs_trans_unreserve_and_mod_dquots(tp);
 	if (tp->t_ticket) {
-		commit_lsn = xfs_log_done(mp, tp->t_ticket, NULL, regrant);
-		if (commit_lsn == -1 && !error)
-			error = -EIO;
+		if (regrant && !XLOG_FORCED_SHUTDOWN(mp->m_log))
+			xfs_log_ticket_regrant(mp->m_log, tp->t_ticket);
+		else
+			xfs_log_ticket_ungrant(mp->m_log, tp->t_ticket);
 		tp->t_ticket = NULL;
 	}
 	current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS);
@@ -1060,7 +1068,7 @@ xfs_trans_cancel(
 	xfs_trans_unreserve_and_mod_dquots(tp);
 
 	if (tp->t_ticket) {
-		xfs_log_done(mp, tp->t_ticket, NULL, false);
+		xfs_log_ticket_ungrant(mp->m_log, tp->t_ticket);
 		tp->t_ticket = NULL;
 	}
 
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index 64d7f171ebd3..752c7fef9de7 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -169,21 +169,21 @@ int		xfs_trans_alloc_empty(struct xfs_mount *mp,
 			struct xfs_trans **tpp);
 void		xfs_trans_mod_sb(xfs_trans_t *, uint, int64_t);
 
-struct xfs_buf	*xfs_trans_get_buf_map(struct xfs_trans *tp,
-				       struct xfs_buftarg *target,
-				       struct xfs_buf_map *map, int nmaps,
-				       uint flags);
+int xfs_trans_get_buf_map(struct xfs_trans *tp, struct xfs_buftarg *target,
+		struct xfs_buf_map *map, int nmaps, xfs_buf_flags_t flags,
+		struct xfs_buf **bpp);
 
-static inline struct xfs_buf *
+static inline int
 xfs_trans_get_buf(
 	struct xfs_trans	*tp,
 	struct xfs_buftarg	*target,
 	xfs_daddr_t		blkno,
 	int			numblks,
-	uint			flags)
+	uint			flags,
+	struct xfs_buf		**bpp)
 {
 	DEFINE_SINGLE_BUF_MAP(map, blkno, numblks);
-	return xfs_trans_get_buf_map(tp, target, &map, 1, flags);
+	return xfs_trans_get_buf_map(tp, target, &map, 1, flags, bpp);
 }
 
 int		xfs_trans_read_buf_map(struct xfs_mount *mp,
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index 6ccfd75d3c24..564253550b75 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -32,6 +32,7 @@ STATIC void
 xfs_ail_check(
 	struct xfs_ail		*ailp,
 	struct xfs_log_item	*lip)
+	__must_hold(&ailp->ail_lock)
 {
 	struct xfs_log_item	*prev_lip;
 	struct xfs_log_item	*next_lip;
@@ -108,17 +109,25 @@ xfs_ail_next(
  * We need the AIL lock in order to get a coherent read of the lsn of the last
  * item in the AIL.
  */
+static xfs_lsn_t
+__xfs_ail_min_lsn(
+	struct xfs_ail		*ailp)
+{
+	struct xfs_log_item	*lip = xfs_ail_min(ailp);
+
+	if (lip)
+		return lip->li_lsn;
+	return 0;
+}
+
 xfs_lsn_t
 xfs_ail_min_lsn(
 	struct xfs_ail		*ailp)
 {
-	xfs_lsn_t		lsn = 0;
-	struct xfs_log_item	*lip;
+	xfs_lsn_t		lsn;
 
 	spin_lock(&ailp->ail_lock);
-	lip = xfs_ail_min(ailp);
-	if (lip)
-		lsn = lip->li_lsn;
+	lsn = __xfs_ail_min_lsn(ailp);
 	spin_unlock(&ailp->ail_lock);
 
 	return lsn;
@@ -427,15 +436,15 @@ xfsaild_push(
 
 		case XFS_ITEM_FLUSHING:
 			/*
-			 * The item or its backing buffer is already beeing
+			 * The item or its backing buffer is already being
 			 * flushed.  The typical reason for that is that an
 			 * inode buffer is locked because we already pushed the
 			 * updates to it as part of inode clustering.
 			 *
 			 * We do not want to to stop flushing just because lots
-			 * of items are already beeing flushed, but we need to
+			 * of items are already being flushed, but we need to
 			 * re-try the flushing relatively soon if most of the
-			 * AIL is beeing flushed.
+			 * AIL is being flushed.
 			 */
 			XFS_STATS_INC(mp, xs_push_ail_flushing);
 			trace_xfs_ail_flushing(lip);
@@ -529,8 +538,9 @@ xfsaild(
 {
 	struct xfs_ail	*ailp = data;
 	long		tout = 0;	/* milliseconds */
+	unsigned int	noreclaim_flag;
 
-	current->flags |= PF_MEMALLOC;
+	noreclaim_flag = memalloc_noreclaim_save();
 	set_freezable();
 
 	while (1) {
@@ -601,6 +611,7 @@ xfsaild(
 		tout = xfsaild_push(ailp);
 	}
 
+	memalloc_noreclaim_restore(noreclaim_flag);
 	return 0;
 }
 
@@ -612,7 +623,7 @@ xfsaild(
  * The push is run asynchronously in a workqueue, which means the caller needs
  * to handle waiting on the async flush for space to become available.
  * We don't want to interrupt any push that is in progress, hence we only queue
- * work if we set the pushing bit approriately.
+ * work if we set the pushing bit appropriately.
  *
  * We do this unlocked - we only need to know whether there is anything in the
  * AIL at the time we are called. We don't need to access the contents of
@@ -678,6 +689,28 @@ xfs_ail_push_all_sync(
 	finish_wait(&ailp->ail_empty, &wait);
 }
 
+void
+xfs_ail_update_finish(
+	struct xfs_ail		*ailp,
+	xfs_lsn_t		old_lsn) __releases(ailp->ail_lock)
+{
+	struct xfs_mount	*mp = ailp->ail_mount;
+
+	/* if the tail lsn hasn't changed, don't do updates or wakeups. */
+	if (!old_lsn || old_lsn == __xfs_ail_min_lsn(ailp)) {
+		spin_unlock(&ailp->ail_lock);
+		return;
+	}
+
+	if (!XFS_FORCED_SHUTDOWN(mp))
+		xlog_assign_tail_lsn_locked(mp);
+
+	if (list_empty(&ailp->ail_head))
+		wake_up_all(&ailp->ail_empty);
+	spin_unlock(&ailp->ail_lock);
+	xfs_log_space_wake(mp);
+}
+
 /*
  * xfs_trans_ail_update - bulk AIL insertion operation.
  *
@@ -709,7 +742,7 @@ xfs_trans_ail_update_bulk(
 	xfs_lsn_t		lsn) __releases(ailp->ail_lock)
 {
 	struct xfs_log_item	*mlip;
-	int			mlip_changed = 0;
+	xfs_lsn_t		tail_lsn = 0;
 	int			i;
 	LIST_HEAD(tmp);
 
@@ -724,9 +757,10 @@ xfs_trans_ail_update_bulk(
 				continue;
 
 			trace_xfs_ail_move(lip, lip->li_lsn, lsn);
+			if (mlip == lip && !tail_lsn)
+				tail_lsn = lip->li_lsn;
+
 			xfs_ail_delete(ailp, lip);
-			if (mlip == lip)
-				mlip_changed = 1;
 		} else {
 			trace_xfs_ail_insert(lip, 0, lsn);
 		}
@@ -737,23 +771,23 @@ xfs_trans_ail_update_bulk(
 	if (!list_empty(&tmp))
 		xfs_ail_splice(ailp, cur, &tmp, lsn);
 
-	if (mlip_changed) {
-		if (!XFS_FORCED_SHUTDOWN(ailp->ail_mount))
-			xlog_assign_tail_lsn_locked(ailp->ail_mount);
-		spin_unlock(&ailp->ail_lock);
-
-		xfs_log_space_wake(ailp->ail_mount);
-	} else {
-		spin_unlock(&ailp->ail_lock);
-	}
+	xfs_ail_update_finish(ailp, tail_lsn);
 }
 
-bool
+/*
+ * Delete one log item from the AIL.
+ *
+ * If this item was at the tail of the AIL, return the LSN of the log item so
+ * that we can use it to check if the LSN of the tail of the log has moved
+ * when finishing up the AIL delete process in xfs_ail_update_finish().
+ */
+xfs_lsn_t
 xfs_ail_delete_one(
 	struct xfs_ail		*ailp,
 	struct xfs_log_item	*lip)
 {
 	struct xfs_log_item	*mlip = xfs_ail_min(ailp);
+	xfs_lsn_t		lsn = lip->li_lsn;
 
 	trace_xfs_ail_delete(lip, mlip->li_lsn, lip->li_lsn);
 	xfs_ail_delete(ailp, lip);
@@ -761,7 +795,9 @@ xfs_ail_delete_one(
 	clear_bit(XFS_LI_IN_AIL, &lip->li_flags);
 	lip->li_lsn = 0;
 
-	return mlip == lip;
+	if (mlip == lip)
+		return lsn;
+	return 0;
 }
 
 /**
@@ -789,10 +825,10 @@ void
 xfs_trans_ail_delete(
 	struct xfs_ail		*ailp,
 	struct xfs_log_item	*lip,
-	int			shutdown_type) __releases(ailp->ail_lock)
+	int			shutdown_type)
 {
 	struct xfs_mount	*mp = ailp->ail_mount;
-	bool			mlip_changed;
+	xfs_lsn_t		tail_lsn;
 
 	if (!test_bit(XFS_LI_IN_AIL, &lip->li_flags)) {
 		spin_unlock(&ailp->ail_lock);
@@ -805,17 +841,8 @@ xfs_trans_ail_delete(
 		return;
 	}
 
-	mlip_changed = xfs_ail_delete_one(ailp, lip);
-	if (mlip_changed) {
-		if (!XFS_FORCED_SHUTDOWN(mp))
-			xlog_assign_tail_lsn_locked(mp);
-		if (list_empty(&ailp->ail_head))
-			wake_up_all(&ailp->ail_empty);
-	}
-
-	spin_unlock(&ailp->ail_lock);
-	if (mlip_changed)
-		xfs_log_space_wake(ailp->ail_mount);
+	tail_lsn = xfs_ail_delete_one(ailp, lip);
+	xfs_ail_update_finish(ailp, tail_lsn);
 }
 
 int
@@ -836,7 +863,7 @@ xfs_trans_ail_init(
 	init_waitqueue_head(&ailp->ail_empty);
 
 	ailp->ail_task = kthread_run(xfsaild, ailp, "xfsaild/%s",
-			ailp->ail_mount->m_fsname);
+			ailp->ail_mount->m_super->s_id);
 	if (IS_ERR(ailp->ail_task))
 		goto out_free_ailp;
 
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index b5b3a78ef31c..08174ffa2118 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -112,19 +112,22 @@ xfs_trans_bjoin(
  * If the transaction pointer is NULL, make this just a normal
  * get_buf() call.
  */
-struct xfs_buf *
+int
 xfs_trans_get_buf_map(
 	struct xfs_trans	*tp,
 	struct xfs_buftarg	*target,
 	struct xfs_buf_map	*map,
 	int			nmaps,
-	xfs_buf_flags_t		flags)
+	xfs_buf_flags_t		flags,
+	struct xfs_buf		**bpp)
 {
 	xfs_buf_t		*bp;
 	struct xfs_buf_log_item	*bip;
+	int			error;
 
+	*bpp = NULL;
 	if (!tp)
-		return xfs_buf_get_map(target, map, nmaps, flags);
+		return xfs_buf_get_map(target, map, nmaps, flags, bpp);
 
 	/*
 	 * If we find the buffer in the cache with this transaction
@@ -146,19 +149,20 @@ xfs_trans_get_buf_map(
 		ASSERT(atomic_read(&bip->bli_refcount) > 0);
 		bip->bli_recur++;
 		trace_xfs_trans_get_buf_recur(bip);
-		return bp;
+		*bpp = bp;
+		return 0;
 	}
 
-	bp = xfs_buf_get_map(target, map, nmaps, flags);
-	if (bp == NULL) {
-		return NULL;
-	}
+	error = xfs_buf_get_map(target, map, nmaps, flags, &bp);
+	if (error)
+		return error;
 
 	ASSERT(!bp->b_error);
 
 	_xfs_trans_bjoin(tp, bp, 1);
 	trace_xfs_trans_get_buf(bp->b_log_item);
-	return bp;
+	*bpp = bp;
+	return 0;
 }
 
 /*
@@ -276,7 +280,7 @@ xfs_trans_read_buf_map(
 		ASSERT(bp->b_ops != NULL);
 		error = xfs_buf_reverify(bp, ops);
 		if (error) {
-			xfs_buf_ioerror_alert(bp, __func__);
+			xfs_buf_ioerror_alert(bp, __return_address);
 
 			if (tp->t_flags & XFS_TRANS_DIRTY)
 				xfs_force_shutdown(tp->t_mountp,
@@ -298,36 +302,17 @@ xfs_trans_read_buf_map(
 		return 0;
 	}
 
-	bp = xfs_buf_read_map(target, map, nmaps, flags, ops);
-	if (!bp) {
-		if (!(flags & XBF_TRYLOCK))
-			return -ENOMEM;
-		return tp ? 0 : -EAGAIN;
-	}
-
-	/*
-	 * If we've had a read error, then the contents of the buffer are
-	 * invalid and should not be used. To ensure that a followup read tries
-	 * to pull the buffer from disk again, we clear the XBF_DONE flag and
-	 * mark the buffer stale. This ensures that anyone who has a current
-	 * reference to the buffer will interpret it's contents correctly and
-	 * future cache lookups will also treat it as an empty, uninitialised
-	 * buffer.
-	 */
-	if (bp->b_error) {
-		error = bp->b_error;
-		if (!XFS_FORCED_SHUTDOWN(mp))
-			xfs_buf_ioerror_alert(bp, __func__);
-		bp->b_flags &= ~XBF_DONE;
-		xfs_buf_stale(bp);
-
+	error = xfs_buf_read_map(target, map, nmaps, flags, &bp, ops,
+			__return_address);
+	switch (error) {
+	case 0:
+		break;
+	default:
 		if (tp && (tp->t_flags & XFS_TRANS_DIRTY))
 			xfs_force_shutdown(tp->t_mountp, SHUTDOWN_META_IO_ERROR);
-		xfs_buf_relse(bp);
-
-		/* bad CRC means corrupted metadata */
-		if (error == -EFSBADCRC)
-			error = -EFSCORRUPTED;
+		/* fall through */
+	case -ENOMEM:
+	case -EAGAIN:
 		return error;
 	}
 
diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c
index 16457465833b..d1b9869bc5fa 100644
--- a/fs/xfs/xfs_trans_dquot.c
+++ b/fs/xfs/xfs_trans_dquot.c
@@ -25,8 +25,8 @@ STATIC void	xfs_trans_alloc_dqinfo(xfs_trans_t *);
  */
 void
 xfs_trans_dqjoin(
-	xfs_trans_t	*tp,
-	xfs_dquot_t	*dqp)
+	struct xfs_trans	*tp,
+	struct xfs_dquot	*dqp)
 {
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
 	ASSERT(dqp->q_logitem.qli_dquot == dqp);
@@ -49,8 +49,8 @@ xfs_trans_dqjoin(
  */
 void
 xfs_trans_log_dquot(
-	xfs_trans_t	*tp,
-	xfs_dquot_t	*dqp)
+	struct xfs_trans	*tp,
+	struct xfs_dquot	*dqp)
 {
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
 
@@ -486,12 +486,12 @@ xfs_trans_apply_dquot_deltas(
  */
 void
 xfs_trans_unreserve_and_mod_dquots(
-	xfs_trans_t		*tp)
+	struct xfs_trans	*tp)
 {
 	int			i, j;
-	xfs_dquot_t		*dqp;
+	struct xfs_dquot	*dqp;
 	struct xfs_dqtrx	*qtrx, *qa;
-	bool                    locked;
+	bool			locked;
 
 	if (!tp->t_dqinfo || !(tp->t_flags & XFS_TRANS_DQ_DIRTY))
 		return;
@@ -571,21 +571,21 @@ xfs_quota_warn(
  */
 STATIC int
 xfs_trans_dqresv(
-	xfs_trans_t	*tp,
-	xfs_mount_t	*mp,
-	xfs_dquot_t	*dqp,
-	int64_t		nblks,
-	long		ninos,
-	uint		flags)
+	struct xfs_trans	*tp,
+	struct xfs_mount	*mp,
+	struct xfs_dquot	*dqp,
+	int64_t			nblks,
+	long			ninos,
+	uint			flags)
 {
-	xfs_qcnt_t	hardlimit;
-	xfs_qcnt_t	softlimit;
-	time_t		timer;
-	xfs_qwarncnt_t	warns;
-	xfs_qwarncnt_t	warnlimit;
-	xfs_qcnt_t	total_count;
-	xfs_qcnt_t	*resbcountp;
-	xfs_quotainfo_t	*q = mp->m_quotainfo;
+	xfs_qcnt_t		hardlimit;
+	xfs_qcnt_t		softlimit;
+	time64_t		timer;
+	xfs_qwarncnt_t		warns;
+	xfs_qwarncnt_t		warnlimit;
+	xfs_qcnt_t		total_count;
+	xfs_qcnt_t		*resbcountp;
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
 	struct xfs_def_quota	*defq;
 
 
@@ -635,7 +635,8 @@ xfs_trans_dqresv(
 				goto error_return;
 			}
 			if (softlimit && total_count > softlimit) {
-				if ((timer != 0 && get_seconds() > timer) ||
+				if ((timer != 0 &&
+				     ktime_get_real_seconds() > timer) ||
 				    (warns != 0 && warns >= warnlimit)) {
 					xfs_quota_warn(mp, dqp,
 						       QUOTA_NL_BSOFTLONGWARN);
@@ -662,7 +663,8 @@ xfs_trans_dqresv(
 				goto error_return;
 			}
 			if (softlimit && total_count > softlimit) {
-				if  ((timer != 0 && get_seconds() > timer) ||
+				if  ((timer != 0 &&
+				      ktime_get_real_seconds() > timer) ||
 				     (warns != 0 && warns >= warnlimit)) {
 					xfs_quota_warn(mp, dqp,
 						       QUOTA_NL_ISOFTLONGWARN);
@@ -824,13 +826,13 @@ xfs_trans_reserve_quota_nblks(
 /*
  * This routine is called to allocate a quotaoff log item.
  */
-xfs_qoff_logitem_t *
+struct xfs_qoff_logitem *
 xfs_trans_get_qoff_item(
-	xfs_trans_t		*tp,
-	xfs_qoff_logitem_t	*startqoff,
+	struct xfs_trans	*tp,
+	struct xfs_qoff_logitem	*startqoff,
 	uint			flags)
 {
-	xfs_qoff_logitem_t	*q;
+	struct xfs_qoff_logitem	*q;
 
 	ASSERT(tp != NULL);
 
@@ -852,8 +854,8 @@ xfs_trans_get_qoff_item(
  */
 void
 xfs_trans_log_quotaoff_item(
-	xfs_trans_t		*tp,
-	xfs_qoff_logitem_t	*qlp)
+	struct xfs_trans	*tp,
+	struct xfs_qoff_logitem	*qlp)
 {
 	tp->t_flags |= XFS_TRANS_DIRTY;
 	set_bit(XFS_LI_DIRTY, &qlp->qql_item.li_flags);
@@ -872,6 +874,6 @@ xfs_trans_free_dqinfo(
 {
 	if (!tp->t_dqinfo)
 		return;
-	kmem_zone_free(xfs_qm_dqtrxzone, tp->t_dqinfo);
+	kmem_cache_free(xfs_qm_dqtrxzone, tp->t_dqinfo);
 	tp->t_dqinfo = NULL;
 }
diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h
index 2e073c1c4614..35655eac01a6 100644
--- a/fs/xfs/xfs_trans_priv.h
+++ b/fs/xfs/xfs_trans_priv.h
@@ -91,9 +91,11 @@ xfs_trans_ail_update(
 	xfs_trans_ail_update_bulk(ailp, NULL, &lip, 1, lsn);
 }
 
-bool xfs_ail_delete_one(struct xfs_ail *ailp, struct xfs_log_item *lip);
+xfs_lsn_t xfs_ail_delete_one(struct xfs_ail *ailp, struct xfs_log_item *lip);
+void xfs_ail_update_finish(struct xfs_ail *ailp, xfs_lsn_t old_lsn)
+			__releases(ailp->ail_lock);
 void xfs_trans_ail_delete(struct xfs_ail *ailp, struct xfs_log_item *lip,
-		int shutdown_type) __releases(ailp->ail_lock);
+		int shutdown_type);
 
 static inline void
 xfs_trans_ail_remove(
diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c
index cb895b1df5e4..fc5d7276026e 100644
--- a/fs/xfs/xfs_xattr.c
+++ b/fs/xfs/xfs_xattr.c
@@ -11,51 +11,31 @@
 #include "xfs_da_format.h"
 #include "xfs_inode.h"
 #include "xfs_attr.h"
+#include "xfs_acl.h"
+#include "xfs_da_format.h"
+#include "xfs_da_btree.h"
 
 #include <linux/posix_acl_xattr.h>
-#include <linux/xattr.h>
 
 
 static int
 xfs_xattr_get(const struct xattr_handler *handler, struct dentry *unused,
 		struct inode *inode, const char *name, void *value, size_t size)
 {
-	int xflags = handler->flags;
-	struct xfs_inode *ip = XFS_I(inode);
-	int error, asize = size;
-
-	/* Convert Linux syscall to XFS internal ATTR flags */
-	if (!size) {
-		xflags |= ATTR_KERNOVAL;
-		value = NULL;
-	}
+	struct xfs_da_args	args = {
+		.dp		= XFS_I(inode),
+		.attr_filter	= handler->flags,
+		.name		= name,
+		.namelen	= strlen(name),
+		.value		= value,
+		.valuelen	= size,
+	};
+	int			error;
 
-	error = xfs_attr_get(ip, name, (unsigned char **)&value, &asize, xflags);
+	error = xfs_attr_get(&args);
 	if (error)
 		return error;
-	return asize;
-}
-
-void
-xfs_forget_acl(
-	struct inode		*inode,
-	const char		*name,
-	int			xflags)
-{
-	/*
-	 * Invalidate any cached ACLs if the user has bypassed the ACL
-	 * interface. We don't validate the content whatsoever so it is caller
-	 * responsibility to provide data in valid format and ensure i_mode is
-	 * consistent.
-	 */
-	if (xflags & ATTR_ROOT) {
-#ifdef CONFIG_XFS_POSIX_ACL
-		if (!strcmp(name, SGI_ACL_FILE))
-			forget_cached_acl(inode, ACL_TYPE_ACCESS);
-		else if (!strcmp(name, SGI_ACL_DEFAULT))
-			forget_cached_acl(inode, ACL_TYPE_DEFAULT);
-#endif
-	}
+	return args.valuelen;
 }
 
 static int
@@ -63,23 +43,20 @@ xfs_xattr_set(const struct xattr_handler *handler, struct dentry *unused,
 		struct inode *inode, const char *name, const void *value,
 		size_t size, int flags)
 {
-	int			xflags = handler->flags;
-	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_da_args	args = {
+		.dp		= XFS_I(inode),
+		.attr_filter	= handler->flags,
+		.attr_flags	= flags,
+		.name		= name,
+		.namelen	= strlen(name),
+		.value		= (void *)value,
+		.valuelen	= size,
+	};
 	int			error;
 
-	/* Convert Linux syscall to XFS internal ATTR flags */
-	if (flags & XATTR_CREATE)
-		xflags |= ATTR_CREATE;
-	if (flags & XATTR_REPLACE)
-		xflags |= ATTR_REPLACE;
-
-	if (!value)
-		return xfs_attr_remove(ip, (unsigned char *)name, xflags);
-	error = xfs_attr_set(ip, (unsigned char *)name,
-				(void *)value, size, xflags);
-	if (!error)
-		xfs_forget_acl(inode, name, xflags);
-
+	error = xfs_attr_set(&args);
+	if (!error && (handler->flags & XFS_ATTR_ROOT))
+		xfs_forget_acl(inode, name);
 	return error;
 }
 
@@ -92,14 +69,14 @@ static const struct xattr_handler xfs_xattr_user_handler = {
 
 static const struct xattr_handler xfs_xattr_trusted_handler = {
 	.prefix	= XATTR_TRUSTED_PREFIX,
-	.flags	= ATTR_ROOT,
+	.flags	= XFS_ATTR_ROOT,
 	.get	= xfs_xattr_get,
 	.set	= xfs_xattr_set,
 };
 
 static const struct xattr_handler xfs_xattr_security_handler = {
 	.prefix	= XATTR_SECURITY_PREFIX,
-	.flags	= ATTR_SECURE,
+	.flags	= XFS_ATTR_SECURE,
 	.get	= xfs_xattr_get,
 	.set	= xfs_xattr_set,
 };
@@ -129,7 +106,7 @@ __xfs_xattr_put_listent(
 	if (context->count < 0 || context->seen_enough)
 		return;
 
-	if (!context->alist)
+	if (!context->buffer)
 		goto compute_size;
 
 	arraytop = context->count + prefix_len + namelen + 1;
@@ -138,7 +115,7 @@ __xfs_xattr_put_listent(
 		context->seen_enough = 1;
 		return;
 	}
-	offset = (char *)context->alist + context->count;
+	offset = context->buffer + context->count;
 	strncpy(offset, prefix, prefix_len);
 	offset += prefix_len;
 	strncpy(offset, (char *)name, namelen);			/* real name */
@@ -213,7 +190,6 @@ xfs_vn_listxattr(
 	size_t		size)
 {
 	struct xfs_attr_list_context context;
-	struct attrlist_cursor_kern cursor = { 0 };
 	struct inode	*inode = d_inode(dentry);
 	int		error;
 
@@ -222,14 +198,13 @@ xfs_vn_listxattr(
 	 */
 	memset(&context, 0, sizeof(context));
 	context.dp = XFS_I(inode);
-	context.cursor = &cursor;
 	context.resynch = 1;
-	context.alist = size ? data : NULL;
+	context.buffer = size ? data : NULL;
 	context.bufsize = size;
 	context.firstu = context.bufsize;
 	context.put_listent = xfs_xattr_put_listent;
 
-	error = xfs_attr_list_int(&context);
+	error = xfs_attr_list(&context);
 	if (error)
 		return error;
 	if (context.count < 0)
diff --git a/fs/zonefs/Kconfig b/fs/zonefs/Kconfig
new file mode 100644
index 000000000000..ef2697b78820
--- /dev/null
+++ b/fs/zonefs/Kconfig
@@ -0,0 +1,10 @@
+config ZONEFS_FS
+	tristate "zonefs filesystem support"
+	depends on BLOCK
+	depends on BLK_DEV_ZONED
+	select FS_IOMAP
+	help
+	  zonefs is a simple file system which exposes zones of a zoned block
+	  device (e.g. host-managed or host-aware SMR disk drives) as files.
+
+	  If unsure, say N.
diff --git a/fs/zonefs/Makefile b/fs/zonefs/Makefile
new file mode 100644
index 000000000000..75a380aa1ae1
--- /dev/null
+++ b/fs/zonefs/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_ZONEFS_FS) += zonefs.o
+
+zonefs-y	:= super.o
diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
new file mode 100644
index 000000000000..d79b821ed1c7
--- /dev/null
+++ b/fs/zonefs/super.c
@@ -0,0 +1,1516 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Simple file system for zoned block devices exposing zones as files.
+ *
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ */
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/magic.h>
+#include <linux/iomap.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/blkdev.h>
+#include <linux/statfs.h>
+#include <linux/writeback.h>
+#include <linux/quotaops.h>
+#include <linux/seq_file.h>
+#include <linux/parser.h>
+#include <linux/uio.h>
+#include <linux/mman.h>
+#include <linux/sched/mm.h>
+#include <linux/crc32.h>
+#include <linux/task_io_accounting_ops.h>
+
+#include "zonefs.h"
+
+static int zonefs_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
+			      unsigned int flags, struct iomap *iomap,
+			      struct iomap *srcmap)
+{
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	struct super_block *sb = inode->i_sb;
+	loff_t isize;
+
+	/* All I/Os should always be within the file maximum size */
+	if (WARN_ON_ONCE(offset + length > zi->i_max_size))
+		return -EIO;
+
+	/*
+	 * Sequential zones can only accept direct writes. This is already
+	 * checked when writes are issued, so warn if we see a page writeback
+	 * operation.
+	 */
+	if (WARN_ON_ONCE(zi->i_ztype == ZONEFS_ZTYPE_SEQ &&
+			 (flags & IOMAP_WRITE) && !(flags & IOMAP_DIRECT)))
+		return -EIO;
+
+	/*
+	 * For conventional zones, all blocks are always mapped. For sequential
+	 * zones, all blocks after always mapped below the inode size (zone
+	 * write pointer) and unwriten beyond.
+	 */
+	mutex_lock(&zi->i_truncate_mutex);
+	isize = i_size_read(inode);
+	if (offset >= isize)
+		iomap->type = IOMAP_UNWRITTEN;
+	else
+		iomap->type = IOMAP_MAPPED;
+	if (flags & IOMAP_WRITE)
+		length = zi->i_max_size - offset;
+	else
+		length = min(length, isize - offset);
+	mutex_unlock(&zi->i_truncate_mutex);
+
+	iomap->offset = ALIGN_DOWN(offset, sb->s_blocksize);
+	iomap->length = ALIGN(offset + length, sb->s_blocksize) - iomap->offset;
+	iomap->bdev = inode->i_sb->s_bdev;
+	iomap->addr = (zi->i_zsector << SECTOR_SHIFT) + iomap->offset;
+
+	return 0;
+}
+
+static const struct iomap_ops zonefs_iomap_ops = {
+	.iomap_begin	= zonefs_iomap_begin,
+};
+
+static int zonefs_readpage(struct file *unused, struct page *page)
+{
+	return iomap_readpage(page, &zonefs_iomap_ops);
+}
+
+static void zonefs_readahead(struct readahead_control *rac)
+{
+	iomap_readahead(rac, &zonefs_iomap_ops);
+}
+
+/*
+ * Map blocks for page writeback. This is used only on conventional zone files,
+ * which implies that the page range can only be within the fixed inode size.
+ */
+static int zonefs_map_blocks(struct iomap_writepage_ctx *wpc,
+			     struct inode *inode, loff_t offset)
+{
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+
+	if (WARN_ON_ONCE(zi->i_ztype != ZONEFS_ZTYPE_CNV))
+		return -EIO;
+	if (WARN_ON_ONCE(offset >= i_size_read(inode)))
+		return -EIO;
+
+	/* If the mapping is already OK, nothing needs to be done */
+	if (offset >= wpc->iomap.offset &&
+	    offset < wpc->iomap.offset + wpc->iomap.length)
+		return 0;
+
+	return zonefs_iomap_begin(inode, offset, zi->i_max_size - offset,
+				  IOMAP_WRITE, &wpc->iomap, NULL);
+}
+
+static const struct iomap_writeback_ops zonefs_writeback_ops = {
+	.map_blocks		= zonefs_map_blocks,
+};
+
+static int zonefs_writepage(struct page *page, struct writeback_control *wbc)
+{
+	struct iomap_writepage_ctx wpc = { };
+
+	return iomap_writepage(page, wbc, &wpc, &zonefs_writeback_ops);
+}
+
+static int zonefs_writepages(struct address_space *mapping,
+			     struct writeback_control *wbc)
+{
+	struct iomap_writepage_ctx wpc = { };
+
+	return iomap_writepages(mapping, wbc, &wpc, &zonefs_writeback_ops);
+}
+
+static const struct address_space_operations zonefs_file_aops = {
+	.readpage		= zonefs_readpage,
+	.readahead		= zonefs_readahead,
+	.writepage		= zonefs_writepage,
+	.writepages		= zonefs_writepages,
+	.set_page_dirty		= iomap_set_page_dirty,
+	.releasepage		= iomap_releasepage,
+	.invalidatepage		= iomap_invalidatepage,
+	.migratepage		= iomap_migrate_page,
+	.is_partially_uptodate	= iomap_is_partially_uptodate,
+	.error_remove_page	= generic_error_remove_page,
+	.direct_IO		= noop_direct_IO,
+};
+
+static void zonefs_update_stats(struct inode *inode, loff_t new_isize)
+{
+	struct super_block *sb = inode->i_sb;
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+	loff_t old_isize = i_size_read(inode);
+	loff_t nr_blocks;
+
+	if (new_isize == old_isize)
+		return;
+
+	spin_lock(&sbi->s_lock);
+
+	/*
+	 * This may be called for an update after an IO error.
+	 * So beware of the values seen.
+	 */
+	if (new_isize < old_isize) {
+		nr_blocks = (old_isize - new_isize) >> sb->s_blocksize_bits;
+		if (sbi->s_used_blocks > nr_blocks)
+			sbi->s_used_blocks -= nr_blocks;
+		else
+			sbi->s_used_blocks = 0;
+	} else {
+		sbi->s_used_blocks +=
+			(new_isize - old_isize) >> sb->s_blocksize_bits;
+		if (sbi->s_used_blocks > sbi->s_blocks)
+			sbi->s_used_blocks = sbi->s_blocks;
+	}
+
+	spin_unlock(&sbi->s_lock);
+}
+
+/*
+ * Check a zone condition and adjust its file inode access permissions for
+ * offline and readonly zones. Return the inode size corresponding to the
+ * amount of readable data in the zone.
+ */
+static loff_t zonefs_check_zone_condition(struct inode *inode,
+					  struct blk_zone *zone, bool warn,
+					  bool mount)
+{
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+
+	switch (zone->cond) {
+	case BLK_ZONE_COND_OFFLINE:
+		/*
+		 * Dead zone: make the inode immutable, disable all accesses
+		 * and set the file size to 0 (zone wp set to zone start).
+		 */
+		if (warn)
+			zonefs_warn(inode->i_sb, "inode %lu: offline zone\n",
+				    inode->i_ino);
+		inode->i_flags |= S_IMMUTABLE;
+		inode->i_mode &= ~0777;
+		zone->wp = zone->start;
+		return 0;
+	case BLK_ZONE_COND_READONLY:
+		/*
+		 * The write pointer of read-only zones is invalid. If such a
+		 * zone is found during mount, the file size cannot be retrieved
+		 * so we treat the zone as offline (mount == true case).
+		 * Otherwise, keep the file size as it was when last updated
+		 * so that the user can recover data. In both cases, writes are
+		 * always disabled for the zone.
+		 */
+		if (warn)
+			zonefs_warn(inode->i_sb, "inode %lu: read-only zone\n",
+				    inode->i_ino);
+		inode->i_flags |= S_IMMUTABLE;
+		if (mount) {
+			zone->cond = BLK_ZONE_COND_OFFLINE;
+			inode->i_mode &= ~0777;
+			zone->wp = zone->start;
+			return 0;
+		}
+		inode->i_mode &= ~0222;
+		return i_size_read(inode);
+	default:
+		if (zi->i_ztype == ZONEFS_ZTYPE_CNV)
+			return zi->i_max_size;
+		return (zone->wp - zone->start) << SECTOR_SHIFT;
+	}
+}
+
+struct zonefs_ioerr_data {
+	struct inode	*inode;
+	bool		write;
+};
+
+static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx,
+			      void *data)
+{
+	struct zonefs_ioerr_data *err = data;
+	struct inode *inode = err->inode;
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	struct super_block *sb = inode->i_sb;
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+	loff_t isize, data_size;
+
+	/*
+	 * Check the zone condition: if the zone is not "bad" (offline or
+	 * read-only), read errors are simply signaled to the IO issuer as long
+	 * as there is no inconsistency between the inode size and the amount of
+	 * data writen in the zone (data_size).
+	 */
+	data_size = zonefs_check_zone_condition(inode, zone, true, false);
+	isize = i_size_read(inode);
+	if (zone->cond != BLK_ZONE_COND_OFFLINE &&
+	    zone->cond != BLK_ZONE_COND_READONLY &&
+	    !err->write && isize == data_size)
+		return 0;
+
+	/*
+	 * At this point, we detected either a bad zone or an inconsistency
+	 * between the inode size and the amount of data written in the zone.
+	 * For the latter case, the cause may be a write IO error or an external
+	 * action on the device. Two error patterns exist:
+	 * 1) The inode size is lower than the amount of data in the zone:
+	 *    a write operation partially failed and data was writen at the end
+	 *    of the file. This can happen in the case of a large direct IO
+	 *    needing several BIOs and/or write requests to be processed.
+	 * 2) The inode size is larger than the amount of data in the zone:
+	 *    this can happen with a deferred write error with the use of the
+	 *    device side write cache after getting successful write IO
+	 *    completions. Other possibilities are (a) an external corruption,
+	 *    e.g. an application reset the zone directly, or (b) the device
+	 *    has a serious problem (e.g. firmware bug).
+	 *
+	 * In all cases, warn about inode size inconsistency and handle the
+	 * IO error according to the zone condition and to the mount options.
+	 */
+	if (zi->i_ztype == ZONEFS_ZTYPE_SEQ && isize != data_size)
+		zonefs_warn(sb, "inode %lu: invalid size %lld (should be %lld)\n",
+			    inode->i_ino, isize, data_size);
+
+	/*
+	 * First handle bad zones signaled by hardware. The mount options
+	 * errors=zone-ro and errors=zone-offline result in changing the
+	 * zone condition to read-only and offline respectively, as if the
+	 * condition was signaled by the hardware.
+	 */
+	if (zone->cond == BLK_ZONE_COND_OFFLINE ||
+	    sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZOL) {
+		zonefs_warn(sb, "inode %lu: read/write access disabled\n",
+			    inode->i_ino);
+		if (zone->cond != BLK_ZONE_COND_OFFLINE) {
+			zone->cond = BLK_ZONE_COND_OFFLINE;
+			data_size = zonefs_check_zone_condition(inode, zone,
+								false, false);
+		}
+	} else if (zone->cond == BLK_ZONE_COND_READONLY ||
+		   sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZRO) {
+		zonefs_warn(sb, "inode %lu: write access disabled\n",
+			    inode->i_ino);
+		if (zone->cond != BLK_ZONE_COND_READONLY) {
+			zone->cond = BLK_ZONE_COND_READONLY;
+			data_size = zonefs_check_zone_condition(inode, zone,
+								false, false);
+		}
+	}
+
+	/*
+	 * If error=remount-ro was specified, any error result in remounting
+	 * the volume as read-only.
+	 */
+	if ((sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_RO) && !sb_rdonly(sb)) {
+		zonefs_warn(sb, "remounting filesystem read-only\n");
+		sb->s_flags |= SB_RDONLY;
+	}
+
+	/*
+	 * Update block usage stats and the inode size  to prevent access to
+	 * invalid data.
+	 */
+	zonefs_update_stats(inode, data_size);
+	i_size_write(inode, data_size);
+	zi->i_wpoffset = data_size;
+
+	return 0;
+}
+
+/*
+ * When an file IO error occurs, check the file zone to see if there is a change
+ * in the zone condition (e.g. offline or read-only). For a failed write to a
+ * sequential zone, the zone write pointer position must also be checked to
+ * eventually correct the file size and zonefs inode write pointer offset
+ * (which can be out of sync with the drive due to partial write failures).
+ */
+static void zonefs_io_error(struct inode *inode, bool write)
+{
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	struct super_block *sb = inode->i_sb;
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+	unsigned int noio_flag;
+	unsigned int nr_zones =
+		zi->i_max_size >> (sbi->s_zone_sectors_shift + SECTOR_SHIFT);
+	struct zonefs_ioerr_data err = {
+		.inode = inode,
+		.write = write,
+	};
+	int ret;
+
+	mutex_lock(&zi->i_truncate_mutex);
+
+	/*
+	 * Memory allocations in blkdev_report_zones() can trigger a memory
+	 * reclaim which may in turn cause a recursion into zonefs as well as
+	 * struct request allocations for the same device. The former case may
+	 * end up in a deadlock on the inode truncate mutex, while the latter
+	 * may prevent IO forward progress. Executing the report zones under
+	 * the GFP_NOIO context avoids both problems.
+	 */
+	noio_flag = memalloc_noio_save();
+	ret = blkdev_report_zones(sb->s_bdev, zi->i_zsector, nr_zones,
+				  zonefs_io_error_cb, &err);
+	if (ret != nr_zones)
+		zonefs_err(sb, "Get inode %lu zone information failed %d\n",
+			   inode->i_ino, ret);
+	memalloc_noio_restore(noio_flag);
+
+	mutex_unlock(&zi->i_truncate_mutex);
+}
+
+static int zonefs_file_truncate(struct inode *inode, loff_t isize)
+{
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	loff_t old_isize;
+	enum req_opf op;
+	int ret = 0;
+
+	/*
+	 * Only sequential zone files can be truncated and truncation is allowed
+	 * only down to a 0 size, which is equivalent to a zone reset, and to
+	 * the maximum file size, which is equivalent to a zone finish.
+	 */
+	if (zi->i_ztype != ZONEFS_ZTYPE_SEQ)
+		return -EPERM;
+
+	if (!isize)
+		op = REQ_OP_ZONE_RESET;
+	else if (isize == zi->i_max_size)
+		op = REQ_OP_ZONE_FINISH;
+	else
+		return -EPERM;
+
+	inode_dio_wait(inode);
+
+	/* Serialize against page faults */
+	down_write(&zi->i_mmap_sem);
+
+	/* Serialize against zonefs_iomap_begin() */
+	mutex_lock(&zi->i_truncate_mutex);
+
+	old_isize = i_size_read(inode);
+	if (isize == old_isize)
+		goto unlock;
+
+	ret = blkdev_zone_mgmt(inode->i_sb->s_bdev, op, zi->i_zsector,
+			       zi->i_max_size >> SECTOR_SHIFT, GFP_NOFS);
+	if (ret) {
+		zonefs_err(inode->i_sb,
+			   "Zone management operation at %llu failed %d",
+			   zi->i_zsector, ret);
+		goto unlock;
+	}
+
+	zonefs_update_stats(inode, isize);
+	truncate_setsize(inode, isize);
+	zi->i_wpoffset = isize;
+
+unlock:
+	mutex_unlock(&zi->i_truncate_mutex);
+	up_write(&zi->i_mmap_sem);
+
+	return ret;
+}
+
+static int zonefs_inode_setattr(struct dentry *dentry, struct iattr *iattr)
+{
+	struct inode *inode = d_inode(dentry);
+	int ret;
+
+	if (unlikely(IS_IMMUTABLE(inode)))
+		return -EPERM;
+
+	ret = setattr_prepare(dentry, iattr);
+	if (ret)
+		return ret;
+
+	/*
+	 * Since files and directories cannot be created nor deleted, do not
+	 * allow setting any write attributes on the sub-directories grouping
+	 * files by zone type.
+	 */
+	if ((iattr->ia_valid & ATTR_MODE) && S_ISDIR(inode->i_mode) &&
+	    (iattr->ia_mode & 0222))
+		return -EPERM;
+
+	if (((iattr->ia_valid & ATTR_UID) &&
+	     !uid_eq(iattr->ia_uid, inode->i_uid)) ||
+	    ((iattr->ia_valid & ATTR_GID) &&
+	     !gid_eq(iattr->ia_gid, inode->i_gid))) {
+		ret = dquot_transfer(inode, iattr);
+		if (ret)
+			return ret;
+	}
+
+	if (iattr->ia_valid & ATTR_SIZE) {
+		ret = zonefs_file_truncate(inode, iattr->ia_size);
+		if (ret)
+			return ret;
+	}
+
+	setattr_copy(inode, iattr);
+
+	return 0;
+}
+
+static const struct inode_operations zonefs_file_inode_operations = {
+	.setattr	= zonefs_inode_setattr,
+};
+
+static int zonefs_file_fsync(struct file *file, loff_t start, loff_t end,
+			     int datasync)
+{
+	struct inode *inode = file_inode(file);
+	int ret = 0;
+
+	if (unlikely(IS_IMMUTABLE(inode)))
+		return -EPERM;
+
+	/*
+	 * Since only direct writes are allowed in sequential files, page cache
+	 * flush is needed only for conventional zone files.
+	 */
+	if (ZONEFS_I(inode)->i_ztype == ZONEFS_ZTYPE_CNV)
+		ret = file_write_and_wait_range(file, start, end);
+	if (!ret)
+		ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL);
+
+	if (ret)
+		zonefs_io_error(inode, true);
+
+	return ret;
+}
+
+static vm_fault_t zonefs_filemap_fault(struct vm_fault *vmf)
+{
+	struct zonefs_inode_info *zi = ZONEFS_I(file_inode(vmf->vma->vm_file));
+	vm_fault_t ret;
+
+	down_read(&zi->i_mmap_sem);
+	ret = filemap_fault(vmf);
+	up_read(&zi->i_mmap_sem);
+
+	return ret;
+}
+
+static vm_fault_t zonefs_filemap_page_mkwrite(struct vm_fault *vmf)
+{
+	struct inode *inode = file_inode(vmf->vma->vm_file);
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	vm_fault_t ret;
+
+	if (unlikely(IS_IMMUTABLE(inode)))
+		return VM_FAULT_SIGBUS;
+
+	/*
+	 * Sanity check: only conventional zone files can have shared
+	 * writeable mappings.
+	 */
+	if (WARN_ON_ONCE(zi->i_ztype != ZONEFS_ZTYPE_CNV))
+		return VM_FAULT_NOPAGE;
+
+	sb_start_pagefault(inode->i_sb);
+	file_update_time(vmf->vma->vm_file);
+
+	/* Serialize against truncates */
+	down_read(&zi->i_mmap_sem);
+	ret = iomap_page_mkwrite(vmf, &zonefs_iomap_ops);
+	up_read(&zi->i_mmap_sem);
+
+	sb_end_pagefault(inode->i_sb);
+	return ret;
+}
+
+static const struct vm_operations_struct zonefs_file_vm_ops = {
+	.fault		= zonefs_filemap_fault,
+	.map_pages	= filemap_map_pages,
+	.page_mkwrite	= zonefs_filemap_page_mkwrite,
+};
+
+static int zonefs_file_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	/*
+	 * Conventional zones accept random writes, so their files can support
+	 * shared writable mappings. For sequential zone files, only read
+	 * mappings are possible since there are no guarantees for write
+	 * ordering between msync() and page cache writeback.
+	 */
+	if (ZONEFS_I(file_inode(file))->i_ztype == ZONEFS_ZTYPE_SEQ &&
+	    (vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE))
+		return -EINVAL;
+
+	file_accessed(file);
+	vma->vm_ops = &zonefs_file_vm_ops;
+
+	return 0;
+}
+
+static loff_t zonefs_file_llseek(struct file *file, loff_t offset, int whence)
+{
+	loff_t isize = i_size_read(file_inode(file));
+
+	/*
+	 * Seeks are limited to below the zone size for conventional zones
+	 * and below the zone write pointer for sequential zones. In both
+	 * cases, this limit is the inode size.
+	 */
+	return generic_file_llseek_size(file, offset, whence, isize, isize);
+}
+
+static int zonefs_file_write_dio_end_io(struct kiocb *iocb, ssize_t size,
+					int error, unsigned int flags)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+
+	if (error) {
+		zonefs_io_error(inode, true);
+		return error;
+	}
+
+	if (size && zi->i_ztype != ZONEFS_ZTYPE_CNV) {
+		/*
+		 * Note that we may be seeing completions out of order,
+		 * but that is not a problem since a write completed
+		 * successfully necessarily means that all preceding writes
+		 * were also successful. So we can safely increase the inode
+		 * size to the write end location.
+		 */
+		mutex_lock(&zi->i_truncate_mutex);
+		if (i_size_read(inode) < iocb->ki_pos + size) {
+			zonefs_update_stats(inode, iocb->ki_pos + size);
+			i_size_write(inode, iocb->ki_pos + size);
+		}
+		mutex_unlock(&zi->i_truncate_mutex);
+	}
+
+	return 0;
+}
+
+static const struct iomap_dio_ops zonefs_write_dio_ops = {
+	.end_io			= zonefs_file_write_dio_end_io,
+};
+
+static ssize_t zonefs_file_dio_append(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	struct block_device *bdev = inode->i_sb->s_bdev;
+	unsigned int max;
+	struct bio *bio;
+	ssize_t size;
+	int nr_pages;
+	ssize_t ret;
+
+	nr_pages = iov_iter_npages(from, BIO_MAX_PAGES);
+	if (!nr_pages)
+		return 0;
+
+	max = queue_max_zone_append_sectors(bdev_get_queue(bdev));
+	max = ALIGN_DOWN(max << SECTOR_SHIFT, inode->i_sb->s_blocksize);
+	iov_iter_truncate(from, max);
+
+	bio = bio_alloc_bioset(GFP_NOFS, nr_pages, &fs_bio_set);
+	if (!bio)
+		return -ENOMEM;
+
+	bio_set_dev(bio, bdev);
+	bio->bi_iter.bi_sector = zi->i_zsector;
+	bio->bi_write_hint = iocb->ki_hint;
+	bio->bi_ioprio = iocb->ki_ioprio;
+	bio->bi_opf = REQ_OP_ZONE_APPEND | REQ_SYNC | REQ_IDLE;
+	if (iocb->ki_flags & IOCB_DSYNC)
+		bio->bi_opf |= REQ_FUA;
+
+	ret = bio_iov_iter_get_pages(bio, from);
+	if (unlikely(ret)) {
+		bio_io_error(bio);
+		return ret;
+	}
+	size = bio->bi_iter.bi_size;
+	task_io_account_write(ret);
+
+	if (iocb->ki_flags & IOCB_HIPRI)
+		bio_set_polled(bio, iocb);
+
+	ret = submit_bio_wait(bio);
+
+	bio_put(bio);
+
+	zonefs_file_write_dio_end_io(iocb, size, ret, 0);
+	if (ret >= 0) {
+		iocb->ki_pos += size;
+		return size;
+	}
+
+	return ret;
+}
+
+/*
+ * Handle direct writes. For sequential zone files, this is the only possible
+ * write path. For these files, check that the user is issuing writes
+ * sequentially from the end of the file. This code assumes that the block layer
+ * delivers write requests to the device in sequential order. This is always the
+ * case if a block IO scheduler implementing the ELEVATOR_F_ZBD_SEQ_WRITE
+ * elevator feature is being used (e.g. mq-deadline). The block layer always
+ * automatically select such an elevator for zoned block devices during the
+ * device initialization.
+ */
+static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	struct super_block *sb = inode->i_sb;
+	bool sync = is_sync_kiocb(iocb);
+	bool append = false;
+	size_t count;
+	ssize_t ret;
+
+	/*
+	 * For async direct IOs to sequential zone files, refuse IOCB_NOWAIT
+	 * as this can cause write reordering (e.g. the first aio gets EAGAIN
+	 * on the inode lock but the second goes through but is now unaligned).
+	 */
+	if (zi->i_ztype == ZONEFS_ZTYPE_SEQ && !sync &&
+	    (iocb->ki_flags & IOCB_NOWAIT))
+		return -EOPNOTSUPP;
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock(inode))
+			return -EAGAIN;
+	} else {
+		inode_lock(inode);
+	}
+
+	ret = generic_write_checks(iocb, from);
+	if (ret <= 0)
+		goto inode_unlock;
+
+	iov_iter_truncate(from, zi->i_max_size - iocb->ki_pos);
+	count = iov_iter_count(from);
+
+	if ((iocb->ki_pos | count) & (sb->s_blocksize - 1)) {
+		ret = -EINVAL;
+		goto inode_unlock;
+	}
+
+	/* Enforce sequential writes (append only) in sequential zones */
+	if (zi->i_ztype == ZONEFS_ZTYPE_SEQ) {
+		mutex_lock(&zi->i_truncate_mutex);
+		if (iocb->ki_pos != zi->i_wpoffset) {
+			mutex_unlock(&zi->i_truncate_mutex);
+			ret = -EINVAL;
+			goto inode_unlock;
+		}
+		mutex_unlock(&zi->i_truncate_mutex);
+		append = sync;
+	}
+
+	if (append)
+		ret = zonefs_file_dio_append(iocb, from);
+	else
+		ret = iomap_dio_rw(iocb, from, &zonefs_iomap_ops,
+				   &zonefs_write_dio_ops, sync);
+	if (zi->i_ztype == ZONEFS_ZTYPE_SEQ &&
+	    (ret > 0 || ret == -EIOCBQUEUED)) {
+		if (ret > 0)
+			count = ret;
+		mutex_lock(&zi->i_truncate_mutex);
+		zi->i_wpoffset += count;
+		mutex_unlock(&zi->i_truncate_mutex);
+	}
+
+inode_unlock:
+	inode_unlock(inode);
+
+	return ret;
+}
+
+static ssize_t zonefs_file_buffered_write(struct kiocb *iocb,
+					  struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	ssize_t ret;
+
+	/*
+	 * Direct IO writes are mandatory for sequential zone files so that the
+	 * write IO issuing order is preserved.
+	 */
+	if (zi->i_ztype != ZONEFS_ZTYPE_CNV)
+		return -EIO;
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock(inode))
+			return -EAGAIN;
+	} else {
+		inode_lock(inode);
+	}
+
+	ret = generic_write_checks(iocb, from);
+	if (ret <= 0)
+		goto inode_unlock;
+
+	iov_iter_truncate(from, zi->i_max_size - iocb->ki_pos);
+
+	ret = iomap_file_buffered_write(iocb, from, &zonefs_iomap_ops);
+	if (ret > 0)
+		iocb->ki_pos += ret;
+	else if (ret == -EIO)
+		zonefs_io_error(inode, true);
+
+inode_unlock:
+	inode_unlock(inode);
+	if (ret > 0)
+		ret = generic_write_sync(iocb, ret);
+
+	return ret;
+}
+
+static ssize_t zonefs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	if (unlikely(IS_IMMUTABLE(inode)))
+		return -EPERM;
+
+	if (sb_rdonly(inode->i_sb))
+		return -EROFS;
+
+	/* Write operations beyond the zone size are not allowed */
+	if (iocb->ki_pos >= ZONEFS_I(inode)->i_max_size)
+		return -EFBIG;
+
+	if (iocb->ki_flags & IOCB_DIRECT)
+		return zonefs_file_dio_write(iocb, from);
+
+	return zonefs_file_buffered_write(iocb, from);
+}
+
+static int zonefs_file_read_dio_end_io(struct kiocb *iocb, ssize_t size,
+				       int error, unsigned int flags)
+{
+	if (error) {
+		zonefs_io_error(file_inode(iocb->ki_filp), false);
+		return error;
+	}
+
+	return 0;
+}
+
+static const struct iomap_dio_ops zonefs_read_dio_ops = {
+	.end_io			= zonefs_file_read_dio_end_io,
+};
+
+static ssize_t zonefs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+	struct super_block *sb = inode->i_sb;
+	loff_t isize;
+	ssize_t ret;
+
+	/* Offline zones cannot be read */
+	if (unlikely(IS_IMMUTABLE(inode) && !(inode->i_mode & 0777)))
+		return -EPERM;
+
+	if (iocb->ki_pos >= zi->i_max_size)
+		return 0;
+
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock_shared(inode))
+			return -EAGAIN;
+	} else {
+		inode_lock_shared(inode);
+	}
+
+	/* Limit read operations to written data */
+	mutex_lock(&zi->i_truncate_mutex);
+	isize = i_size_read(inode);
+	if (iocb->ki_pos >= isize) {
+		mutex_unlock(&zi->i_truncate_mutex);
+		ret = 0;
+		goto inode_unlock;
+	}
+	iov_iter_truncate(to, isize - iocb->ki_pos);
+	mutex_unlock(&zi->i_truncate_mutex);
+
+	if (iocb->ki_flags & IOCB_DIRECT) {
+		size_t count = iov_iter_count(to);
+
+		if ((iocb->ki_pos | count) & (sb->s_blocksize - 1)) {
+			ret = -EINVAL;
+			goto inode_unlock;
+		}
+		file_accessed(iocb->ki_filp);
+		ret = iomap_dio_rw(iocb, to, &zonefs_iomap_ops,
+				   &zonefs_read_dio_ops, is_sync_kiocb(iocb));
+	} else {
+		ret = generic_file_read_iter(iocb, to);
+		if (ret == -EIO)
+			zonefs_io_error(inode, false);
+	}
+
+inode_unlock:
+	inode_unlock_shared(inode);
+
+	return ret;
+}
+
+static const struct file_operations zonefs_file_operations = {
+	.open		= generic_file_open,
+	.fsync		= zonefs_file_fsync,
+	.mmap		= zonefs_file_mmap,
+	.llseek		= zonefs_file_llseek,
+	.read_iter	= zonefs_file_read_iter,
+	.write_iter	= zonefs_file_write_iter,
+	.splice_read	= generic_file_splice_read,
+	.splice_write	= iter_file_splice_write,
+	.iopoll		= iomap_dio_iopoll,
+};
+
+static struct kmem_cache *zonefs_inode_cachep;
+
+static struct inode *zonefs_alloc_inode(struct super_block *sb)
+{
+	struct zonefs_inode_info *zi;
+
+	zi = kmem_cache_alloc(zonefs_inode_cachep, GFP_KERNEL);
+	if (!zi)
+		return NULL;
+
+	inode_init_once(&zi->i_vnode);
+	mutex_init(&zi->i_truncate_mutex);
+	init_rwsem(&zi->i_mmap_sem);
+
+	return &zi->i_vnode;
+}
+
+static void zonefs_free_inode(struct inode *inode)
+{
+	kmem_cache_free(zonefs_inode_cachep, ZONEFS_I(inode));
+}
+
+/*
+ * File system stat.
+ */
+static int zonefs_statfs(struct dentry *dentry, struct kstatfs *buf)
+{
+	struct super_block *sb = dentry->d_sb;
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+	enum zonefs_ztype t;
+	u64 fsid;
+
+	buf->f_type = ZONEFS_MAGIC;
+	buf->f_bsize = sb->s_blocksize;
+	buf->f_namelen = ZONEFS_NAME_MAX;
+
+	spin_lock(&sbi->s_lock);
+
+	buf->f_blocks = sbi->s_blocks;
+	if (WARN_ON(sbi->s_used_blocks > sbi->s_blocks))
+		buf->f_bfree = 0;
+	else
+		buf->f_bfree = buf->f_blocks - sbi->s_used_blocks;
+	buf->f_bavail = buf->f_bfree;
+
+	for (t = 0; t < ZONEFS_ZTYPE_MAX; t++) {
+		if (sbi->s_nr_files[t])
+			buf->f_files += sbi->s_nr_files[t] + 1;
+	}
+	buf->f_ffree = 0;
+
+	spin_unlock(&sbi->s_lock);
+
+	fsid = le64_to_cpup((void *)sbi->s_uuid.b) ^
+		le64_to_cpup((void *)sbi->s_uuid.b + sizeof(u64));
+	buf->f_fsid.val[0] = (u32)fsid;
+	buf->f_fsid.val[1] = (u32)(fsid >> 32);
+
+	return 0;
+}
+
+enum {
+	Opt_errors_ro, Opt_errors_zro, Opt_errors_zol, Opt_errors_repair,
+	Opt_err,
+};
+
+static const match_table_t tokens = {
+	{ Opt_errors_ro,	"errors=remount-ro"},
+	{ Opt_errors_zro,	"errors=zone-ro"},
+	{ Opt_errors_zol,	"errors=zone-offline"},
+	{ Opt_errors_repair,	"errors=repair"},
+	{ Opt_err,		NULL}
+};
+
+static int zonefs_parse_options(struct super_block *sb, char *options)
+{
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+	substring_t args[MAX_OPT_ARGS];
+	char *p;
+
+	if (!options)
+		return 0;
+
+	while ((p = strsep(&options, ",")) != NULL) {
+		int token;
+
+		if (!*p)
+			continue;
+
+		token = match_token(p, tokens, args);
+		switch (token) {
+		case Opt_errors_ro:
+			sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
+			sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_RO;
+			break;
+		case Opt_errors_zro:
+			sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
+			sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_ZRO;
+			break;
+		case Opt_errors_zol:
+			sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
+			sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_ZOL;
+			break;
+		case Opt_errors_repair:
+			sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
+			sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_REPAIR;
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int zonefs_show_options(struct seq_file *seq, struct dentry *root)
+{
+	struct zonefs_sb_info *sbi = ZONEFS_SB(root->d_sb);
+
+	if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_RO)
+		seq_puts(seq, ",errors=remount-ro");
+	if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZRO)
+		seq_puts(seq, ",errors=zone-ro");
+	if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZOL)
+		seq_puts(seq, ",errors=zone-offline");
+	if (sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_REPAIR)
+		seq_puts(seq, ",errors=repair");
+
+	return 0;
+}
+
+static int zonefs_remount(struct super_block *sb, int *flags, char *data)
+{
+	sync_filesystem(sb);
+
+	return zonefs_parse_options(sb, data);
+}
+
+static const struct super_operations zonefs_sops = {
+	.alloc_inode	= zonefs_alloc_inode,
+	.free_inode	= zonefs_free_inode,
+	.statfs		= zonefs_statfs,
+	.remount_fs	= zonefs_remount,
+	.show_options	= zonefs_show_options,
+};
+
+static const struct inode_operations zonefs_dir_inode_operations = {
+	.lookup		= simple_lookup,
+	.setattr	= zonefs_inode_setattr,
+};
+
+static void zonefs_init_dir_inode(struct inode *parent, struct inode *inode,
+				  enum zonefs_ztype type)
+{
+	struct super_block *sb = parent->i_sb;
+
+	inode->i_ino = blkdev_nr_zones(sb->s_bdev->bd_disk) + type + 1;
+	inode_init_owner(inode, parent, S_IFDIR | 0555);
+	inode->i_op = &zonefs_dir_inode_operations;
+	inode->i_fop = &simple_dir_operations;
+	set_nlink(inode, 2);
+	inc_nlink(parent);
+}
+
+static void zonefs_init_file_inode(struct inode *inode, struct blk_zone *zone,
+				   enum zonefs_ztype type)
+{
+	struct super_block *sb = inode->i_sb;
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+	struct zonefs_inode_info *zi = ZONEFS_I(inode);
+
+	inode->i_ino = zone->start >> sbi->s_zone_sectors_shift;
+	inode->i_mode = S_IFREG | sbi->s_perm;
+
+	zi->i_ztype = type;
+	zi->i_zsector = zone->start;
+	zi->i_max_size = min_t(loff_t, MAX_LFS_FILESIZE,
+			       zone->len << SECTOR_SHIFT);
+	zi->i_wpoffset = zonefs_check_zone_condition(inode, zone, true, true);
+
+	inode->i_uid = sbi->s_uid;
+	inode->i_gid = sbi->s_gid;
+	inode->i_size = zi->i_wpoffset;
+	inode->i_blocks = zone->len;
+
+	inode->i_op = &zonefs_file_inode_operations;
+	inode->i_fop = &zonefs_file_operations;
+	inode->i_mapping->a_ops = &zonefs_file_aops;
+
+	sb->s_maxbytes = max(zi->i_max_size, sb->s_maxbytes);
+	sbi->s_blocks += zi->i_max_size >> sb->s_blocksize_bits;
+	sbi->s_used_blocks += zi->i_wpoffset >> sb->s_blocksize_bits;
+}
+
+static struct dentry *zonefs_create_inode(struct dentry *parent,
+					const char *name, struct blk_zone *zone,
+					enum zonefs_ztype type)
+{
+	struct inode *dir = d_inode(parent);
+	struct dentry *dentry;
+	struct inode *inode;
+
+	dentry = d_alloc_name(parent, name);
+	if (!dentry)
+		return NULL;
+
+	inode = new_inode(parent->d_sb);
+	if (!inode)
+		goto dput;
+
+	inode->i_ctime = inode->i_mtime = inode->i_atime = dir->i_ctime;
+	if (zone)
+		zonefs_init_file_inode(inode, zone, type);
+	else
+		zonefs_init_dir_inode(dir, inode, type);
+	d_add(dentry, inode);
+	dir->i_size++;
+
+	return dentry;
+
+dput:
+	dput(dentry);
+
+	return NULL;
+}
+
+struct zonefs_zone_data {
+	struct super_block	*sb;
+	unsigned int		nr_zones[ZONEFS_ZTYPE_MAX];
+	struct blk_zone		*zones;
+};
+
+/*
+ * Create a zone group and populate it with zone files.
+ */
+static int zonefs_create_zgroup(struct zonefs_zone_data *zd,
+				enum zonefs_ztype type)
+{
+	struct super_block *sb = zd->sb;
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+	struct blk_zone *zone, *next, *end;
+	const char *zgroup_name;
+	char *file_name;
+	struct dentry *dir;
+	unsigned int n = 0;
+	int ret = -ENOMEM;
+
+	/* If the group is empty, there is nothing to do */
+	if (!zd->nr_zones[type])
+		return 0;
+
+	file_name = kmalloc(ZONEFS_NAME_MAX, GFP_KERNEL);
+	if (!file_name)
+		return -ENOMEM;
+
+	if (type == ZONEFS_ZTYPE_CNV)
+		zgroup_name = "cnv";
+	else
+		zgroup_name = "seq";
+
+	dir = zonefs_create_inode(sb->s_root, zgroup_name, NULL, type);
+	if (!dir)
+		goto free;
+
+	/*
+	 * The first zone contains the super block: skip it.
+	 */
+	end = zd->zones + blkdev_nr_zones(sb->s_bdev->bd_disk);
+	for (zone = &zd->zones[1]; zone < end; zone = next) {
+
+		next = zone + 1;
+		if (zonefs_zone_type(zone) != type)
+			continue;
+
+		/*
+		 * For conventional zones, contiguous zones can be aggregated
+		 * together to form larger files. Note that this overwrites the
+		 * length of the first zone of the set of contiguous zones
+		 * aggregated together. If one offline or read-only zone is
+		 * found, assume that all zones aggregated have the same
+		 * condition.
+		 */
+		if (type == ZONEFS_ZTYPE_CNV &&
+		    (sbi->s_features & ZONEFS_F_AGGRCNV)) {
+			for (; next < end; next++) {
+				if (zonefs_zone_type(next) != type)
+					break;
+				zone->len += next->len;
+				if (next->cond == BLK_ZONE_COND_READONLY &&
+				    zone->cond != BLK_ZONE_COND_OFFLINE)
+					zone->cond = BLK_ZONE_COND_READONLY;
+				else if (next->cond == BLK_ZONE_COND_OFFLINE)
+					zone->cond = BLK_ZONE_COND_OFFLINE;
+			}
+		}
+
+		/*
+		 * Use the file number within its group as file name.
+		 */
+		snprintf(file_name, ZONEFS_NAME_MAX - 1, "%u", n);
+		if (!zonefs_create_inode(dir, file_name, zone, type))
+			goto free;
+
+		n++;
+	}
+
+	zonefs_info(sb, "Zone group \"%s\" has %u file%s\n",
+		    zgroup_name, n, n > 1 ? "s" : "");
+
+	sbi->s_nr_files[type] = n;
+	ret = 0;
+
+free:
+	kfree(file_name);
+
+	return ret;
+}
+
+static int zonefs_get_zone_info_cb(struct blk_zone *zone, unsigned int idx,
+				   void *data)
+{
+	struct zonefs_zone_data *zd = data;
+
+	/*
+	 * Count the number of usable zones: the first zone at index 0 contains
+	 * the super block and is ignored.
+	 */
+	switch (zone->type) {
+	case BLK_ZONE_TYPE_CONVENTIONAL:
+		zone->wp = zone->start + zone->len;
+		if (idx)
+			zd->nr_zones[ZONEFS_ZTYPE_CNV]++;
+		break;
+	case BLK_ZONE_TYPE_SEQWRITE_REQ:
+	case BLK_ZONE_TYPE_SEQWRITE_PREF:
+		if (idx)
+			zd->nr_zones[ZONEFS_ZTYPE_SEQ]++;
+		break;
+	default:
+		zonefs_err(zd->sb, "Unsupported zone type 0x%x\n",
+			   zone->type);
+		return -EIO;
+	}
+
+	memcpy(&zd->zones[idx], zone, sizeof(struct blk_zone));
+
+	return 0;
+}
+
+static int zonefs_get_zone_info(struct zonefs_zone_data *zd)
+{
+	struct block_device *bdev = zd->sb->s_bdev;
+	int ret;
+
+	zd->zones = kvcalloc(blkdev_nr_zones(bdev->bd_disk),
+			     sizeof(struct blk_zone), GFP_KERNEL);
+	if (!zd->zones)
+		return -ENOMEM;
+
+	/* Get zones information from the device */
+	ret = blkdev_report_zones(bdev, 0, BLK_ALL_ZONES,
+				  zonefs_get_zone_info_cb, zd);
+	if (ret < 0) {
+		zonefs_err(zd->sb, "Zone report failed %d\n", ret);
+		return ret;
+	}
+
+	if (ret != blkdev_nr_zones(bdev->bd_disk)) {
+		zonefs_err(zd->sb, "Invalid zone report (%d/%u zones)\n",
+			   ret, blkdev_nr_zones(bdev->bd_disk));
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static inline void zonefs_cleanup_zone_info(struct zonefs_zone_data *zd)
+{
+	kvfree(zd->zones);
+}
+
+/*
+ * Read super block information from the device.
+ */
+static int zonefs_read_super(struct super_block *sb)
+{
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+	struct zonefs_super *super;
+	u32 crc, stored_crc;
+	struct page *page;
+	struct bio_vec bio_vec;
+	struct bio bio;
+	int ret;
+
+	page = alloc_page(GFP_KERNEL);
+	if (!page)
+		return -ENOMEM;
+
+	bio_init(&bio, &bio_vec, 1);
+	bio.bi_iter.bi_sector = 0;
+	bio.bi_opf = REQ_OP_READ;
+	bio_set_dev(&bio, sb->s_bdev);
+	bio_add_page(&bio, page, PAGE_SIZE, 0);
+
+	ret = submit_bio_wait(&bio);
+	if (ret)
+		goto free_page;
+
+	super = kmap(page);
+
+	ret = -EINVAL;
+	if (le32_to_cpu(super->s_magic) != ZONEFS_MAGIC)
+		goto unmap;
+
+	stored_crc = le32_to_cpu(super->s_crc);
+	super->s_crc = 0;
+	crc = crc32(~0U, (unsigned char *)super, sizeof(struct zonefs_super));
+	if (crc != stored_crc) {
+		zonefs_err(sb, "Invalid checksum (Expected 0x%08x, got 0x%08x)",
+			   crc, stored_crc);
+		goto unmap;
+	}
+
+	sbi->s_features = le64_to_cpu(super->s_features);
+	if (sbi->s_features & ~ZONEFS_F_DEFINED_FEATURES) {
+		zonefs_err(sb, "Unknown features set 0x%llx\n",
+			   sbi->s_features);
+		goto unmap;
+	}
+
+	if (sbi->s_features & ZONEFS_F_UID) {
+		sbi->s_uid = make_kuid(current_user_ns(),
+				       le32_to_cpu(super->s_uid));
+		if (!uid_valid(sbi->s_uid)) {
+			zonefs_err(sb, "Invalid UID feature\n");
+			goto unmap;
+		}
+	}
+
+	if (sbi->s_features & ZONEFS_F_GID) {
+		sbi->s_gid = make_kgid(current_user_ns(),
+				       le32_to_cpu(super->s_gid));
+		if (!gid_valid(sbi->s_gid)) {
+			zonefs_err(sb, "Invalid GID feature\n");
+			goto unmap;
+		}
+	}
+
+	if (sbi->s_features & ZONEFS_F_PERM)
+		sbi->s_perm = le32_to_cpu(super->s_perm);
+
+	if (memchr_inv(super->s_reserved, 0, sizeof(super->s_reserved))) {
+		zonefs_err(sb, "Reserved area is being used\n");
+		goto unmap;
+	}
+
+	uuid_copy(&sbi->s_uuid, (uuid_t *)super->s_uuid);
+	ret = 0;
+
+unmap:
+	kunmap(page);
+free_page:
+	__free_page(page);
+
+	return ret;
+}
+
+/*
+ * Check that the device is zoned. If it is, get the list of zones and create
+ * sub-directories and files according to the device zone configuration and
+ * format options.
+ */
+static int zonefs_fill_super(struct super_block *sb, void *data, int silent)
+{
+	struct zonefs_zone_data zd;
+	struct zonefs_sb_info *sbi;
+	struct inode *inode;
+	enum zonefs_ztype t;
+	int ret;
+
+	if (!bdev_is_zoned(sb->s_bdev)) {
+		zonefs_err(sb, "Not a zoned block device\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * Initialize super block information: the maximum file size is updated
+	 * when the zone files are created so that the format option
+	 * ZONEFS_F_AGGRCNV which increases the maximum file size of a file
+	 * beyond the zone size is taken into account.
+	 */
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
+	if (!sbi)
+		return -ENOMEM;
+
+	spin_lock_init(&sbi->s_lock);
+	sb->s_fs_info = sbi;
+	sb->s_magic = ZONEFS_MAGIC;
+	sb->s_maxbytes = 0;
+	sb->s_op = &zonefs_sops;
+	sb->s_time_gran	= 1;
+
+	/*
+	 * The block size is set to the device physical sector size to ensure
+	 * that write operations on 512e devices (512B logical block and 4KB
+	 * physical block) are always aligned to the device physical blocks,
+	 * as mandated by the ZBC/ZAC specifications.
+	 */
+	sb_set_blocksize(sb, bdev_physical_block_size(sb->s_bdev));
+	sbi->s_zone_sectors_shift = ilog2(bdev_zone_sectors(sb->s_bdev));
+	sbi->s_uid = GLOBAL_ROOT_UID;
+	sbi->s_gid = GLOBAL_ROOT_GID;
+	sbi->s_perm = 0640;
+	sbi->s_mount_opts = ZONEFS_MNTOPT_ERRORS_RO;
+
+	ret = zonefs_read_super(sb);
+	if (ret)
+		return ret;
+
+	ret = zonefs_parse_options(sb, data);
+	if (ret)
+		return ret;
+
+	memset(&zd, 0, sizeof(struct zonefs_zone_data));
+	zd.sb = sb;
+	ret = zonefs_get_zone_info(&zd);
+	if (ret)
+		goto cleanup;
+
+	zonefs_info(sb, "Mounting %u zones",
+		    blkdev_nr_zones(sb->s_bdev->bd_disk));
+
+	/* Create root directory inode */
+	ret = -ENOMEM;
+	inode = new_inode(sb);
+	if (!inode)
+		goto cleanup;
+
+	inode->i_ino = blkdev_nr_zones(sb->s_bdev->bd_disk);
+	inode->i_mode = S_IFDIR | 0555;
+	inode->i_ctime = inode->i_mtime = inode->i_atime = current_time(inode);
+	inode->i_op = &zonefs_dir_inode_operations;
+	inode->i_fop = &simple_dir_operations;
+	set_nlink(inode, 2);
+
+	sb->s_root = d_make_root(inode);
+	if (!sb->s_root)
+		goto cleanup;
+
+	/* Create and populate files in zone groups directories */
+	for (t = 0; t < ZONEFS_ZTYPE_MAX; t++) {
+		ret = zonefs_create_zgroup(&zd, t);
+		if (ret)
+			break;
+	}
+
+cleanup:
+	zonefs_cleanup_zone_info(&zd);
+
+	return ret;
+}
+
+static struct dentry *zonefs_mount(struct file_system_type *fs_type,
+				   int flags, const char *dev_name, void *data)
+{
+	return mount_bdev(fs_type, flags, dev_name, data, zonefs_fill_super);
+}
+
+static void zonefs_kill_super(struct super_block *sb)
+{
+	struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
+
+	if (sb->s_root)
+		d_genocide(sb->s_root);
+	kill_block_super(sb);
+	kfree(sbi);
+}
+
+/*
+ * File system definition and registration.
+ */
+static struct file_system_type zonefs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "zonefs",
+	.mount		= zonefs_mount,
+	.kill_sb	= zonefs_kill_super,
+	.fs_flags	= FS_REQUIRES_DEV,
+};
+
+static int __init zonefs_init_inodecache(void)
+{
+	zonefs_inode_cachep = kmem_cache_create("zonefs_inode_cache",
+			sizeof(struct zonefs_inode_info), 0,
+			(SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD | SLAB_ACCOUNT),
+			NULL);
+	if (zonefs_inode_cachep == NULL)
+		return -ENOMEM;
+	return 0;
+}
+
+static void zonefs_destroy_inodecache(void)
+{
+	/*
+	 * Make sure all delayed rcu free inodes are flushed before we
+	 * destroy the inode cache.
+	 */
+	rcu_barrier();
+	kmem_cache_destroy(zonefs_inode_cachep);
+}
+
+static int __init zonefs_init(void)
+{
+	int ret;
+
+	BUILD_BUG_ON(sizeof(struct zonefs_super) != ZONEFS_SUPER_SIZE);
+
+	ret = zonefs_init_inodecache();
+	if (ret)
+		return ret;
+
+	ret = register_filesystem(&zonefs_type);
+	if (ret) {
+		zonefs_destroy_inodecache();
+		return ret;
+	}
+
+	return 0;
+}
+
+static void __exit zonefs_exit(void)
+{
+	zonefs_destroy_inodecache();
+	unregister_filesystem(&zonefs_type);
+}
+
+MODULE_AUTHOR("Damien Le Moal");
+MODULE_DESCRIPTION("Zone file system for zoned block devices");
+MODULE_LICENSE("GPL");
+module_init(zonefs_init);
+module_exit(zonefs_exit);
diff --git a/fs/zonefs/zonefs.h b/fs/zonefs/zonefs.h
new file mode 100644
index 000000000000..ad17fef7ce91
--- /dev/null
+++ b/fs/zonefs/zonefs.h
@@ -0,0 +1,189 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Simple zone file system for zoned block devices.
+ *
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ */
+#ifndef __ZONEFS_H__
+#define __ZONEFS_H__
+
+#include <linux/fs.h>
+#include <linux/magic.h>
+#include <linux/uuid.h>
+#include <linux/mutex.h>
+#include <linux/rwsem.h>
+
+/*
+ * Maximum length of file names: this only needs to be large enough to fit
+ * the zone group directory names and a decimal zone number for file names.
+ * 16 characters is plenty.
+ */
+#define ZONEFS_NAME_MAX		16
+
+/*
+ * Zone types: ZONEFS_ZTYPE_SEQ is used for all sequential zone types
+ * defined in linux/blkzoned.h, that is, BLK_ZONE_TYPE_SEQWRITE_REQ and
+ * BLK_ZONE_TYPE_SEQWRITE_PREF.
+ */
+enum zonefs_ztype {
+	ZONEFS_ZTYPE_CNV,
+	ZONEFS_ZTYPE_SEQ,
+	ZONEFS_ZTYPE_MAX,
+};
+
+static inline enum zonefs_ztype zonefs_zone_type(struct blk_zone *zone)
+{
+	if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+		return ZONEFS_ZTYPE_CNV;
+	return ZONEFS_ZTYPE_SEQ;
+}
+
+/*
+ * In-memory inode data.
+ */
+struct zonefs_inode_info {
+	struct inode		i_vnode;
+
+	/* File zone type */
+	enum zonefs_ztype	i_ztype;
+
+	/* File zone start sector (512B unit) */
+	sector_t		i_zsector;
+
+	/* File zone write pointer position (sequential zones only) */
+	loff_t			i_wpoffset;
+
+	/* File maximum size */
+	loff_t			i_max_size;
+
+	/*
+	 * To serialise fully against both syscall and mmap based IO and
+	 * sequential file truncation, two locks are used. For serializing
+	 * zonefs_seq_file_truncate() against zonefs_iomap_begin(), that is,
+	 * file truncate operations against block mapping, i_truncate_mutex is
+	 * used. i_truncate_mutex also protects against concurrent accesses
+	 * and changes to the inode private data, and in particular changes to
+	 * a sequential file size on completion of direct IO writes.
+	 * Serialization of mmap read IOs with truncate and syscall IO
+	 * operations is done with i_mmap_sem in addition to i_truncate_mutex.
+	 * Only zonefs_seq_file_truncate() takes both lock (i_mmap_sem first,
+	 * i_truncate_mutex second).
+	 */
+	struct mutex		i_truncate_mutex;
+	struct rw_semaphore	i_mmap_sem;
+};
+
+static inline struct zonefs_inode_info *ZONEFS_I(struct inode *inode)
+{
+	return container_of(inode, struct zonefs_inode_info, i_vnode);
+}
+
+/*
+ * On-disk super block (block 0).
+ */
+#define ZONEFS_LABEL_LEN	64
+#define ZONEFS_UUID_SIZE	16
+#define ZONEFS_SUPER_SIZE	4096
+
+struct zonefs_super {
+
+	/* Magic number */
+	__le32		s_magic;
+
+	/* Checksum */
+	__le32		s_crc;
+
+	/* Volume label */
+	char		s_label[ZONEFS_LABEL_LEN];
+
+	/* 128-bit uuid */
+	__u8		s_uuid[ZONEFS_UUID_SIZE];
+
+	/* Features */
+	__le64		s_features;
+
+	/* UID/GID to use for files */
+	__le32		s_uid;
+	__le32		s_gid;
+
+	/* File permissions */
+	__le32		s_perm;
+
+	/* Padding to ZONEFS_SUPER_SIZE bytes */
+	__u8		s_reserved[3988];
+
+} __packed;
+
+/*
+ * Feature flags: specified in the s_features field of the on-disk super
+ * block struct zonefs_super and in-memory in the s_feartures field of
+ * struct zonefs_sb_info.
+ */
+enum zonefs_features {
+	/*
+	 * Aggregate contiguous conventional zones into a single file.
+	 */
+	ZONEFS_F_AGGRCNV = 1ULL << 0,
+	/*
+	 * Use super block specified UID for files instead of default 0.
+	 */
+	ZONEFS_F_UID = 1ULL << 1,
+	/*
+	 * Use super block specified GID for files instead of default 0.
+	 */
+	ZONEFS_F_GID = 1ULL << 2,
+	/*
+	 * Use super block specified file permissions instead of default 640.
+	 */
+	ZONEFS_F_PERM = 1ULL << 3,
+};
+
+#define ZONEFS_F_DEFINED_FEATURES \
+	(ZONEFS_F_AGGRCNV | ZONEFS_F_UID | ZONEFS_F_GID | ZONEFS_F_PERM)
+
+/*
+ * Mount options for zone write pointer error handling.
+ */
+#define ZONEFS_MNTOPT_ERRORS_RO		(1 << 0) /* Make zone file readonly */
+#define ZONEFS_MNTOPT_ERRORS_ZRO	(1 << 1) /* Make zone file offline */
+#define ZONEFS_MNTOPT_ERRORS_ZOL	(1 << 2) /* Make zone file offline */
+#define ZONEFS_MNTOPT_ERRORS_REPAIR	(1 << 3) /* Remount read-only */
+#define ZONEFS_MNTOPT_ERRORS_MASK	\
+	(ZONEFS_MNTOPT_ERRORS_RO | ZONEFS_MNTOPT_ERRORS_ZRO | \
+	 ZONEFS_MNTOPT_ERRORS_ZOL | ZONEFS_MNTOPT_ERRORS_REPAIR)
+
+/*
+ * In-memory Super block information.
+ */
+struct zonefs_sb_info {
+
+	unsigned long		s_mount_opts;
+
+	spinlock_t		s_lock;
+
+	unsigned long long	s_features;
+	kuid_t			s_uid;
+	kgid_t			s_gid;
+	umode_t			s_perm;
+	uuid_t			s_uuid;
+	unsigned int		s_zone_sectors_shift;
+
+	unsigned int		s_nr_files[ZONEFS_ZTYPE_MAX];
+
+	loff_t			s_blocks;
+	loff_t			s_used_blocks;
+};
+
+static inline struct zonefs_sb_info *ZONEFS_SB(struct super_block *sb)
+{
+	return sb->s_fs_info;
+}
+
+#define zonefs_info(sb, format, args...)	\
+	pr_info("zonefs (%s): " format, sb->s_id, ## args)
+#define zonefs_err(sb, format, args...)		\
+	pr_err("zonefs (%s) ERROR: " format, sb->s_id, ## args)
+#define zonefs_warn(sb, format, args...)	\
+	pr_warn("zonefs (%s) WARNING: " format, sb->s_id, ## args)
+
+#endif