1 files changed, 885 insertions, 347 deletions

diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 72c3866a7320..b2d23c98c996 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -1,14 +1,12 @@
+// SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
  * Copyright (C) 2004-2006 Red Hat, Inc.  All rights reserved.
- *
- * This copyrighted material is made available to anyone wishing to use,
- * modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License version 2.
  */
 
 #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>
@@ -17,19 +15,23 @@
 #include <linux/mm.h>
 #include <linux/mount.h>
 #include <linux/fs.h>
+#include <linux/filelock.h>
 #include <linux/gfs2_ondisk.h>
 #include <linux/falloc.h>
 #include <linux/swap.h>
 #include <linux/crc32.h>
 #include <linux/writeback.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 #include <linux/dlm.h>
 #include <linux/dlm_plock.h>
-#include <linux/aio.h>
+#include <linux/delay.h>
+#include <linux/backing-dev.h>
+#include <linux/fileattr.h>
 
 #include "gfs2.h"
 #include "incore.h"
 #include "bmap.h"
+#include "aops.h"
 #include "dir.h"
 #include "glock.h"
 #include "glops.h"
@@ -60,9 +62,7 @@ static loff_t gfs2_llseek(struct file *file, loff_t offset, int whence)
 	loff_t error;
 
 	switch (whence) {
-	case SEEK_END: /* These reference inode->i_size */
-	case SEEK_DATA:
-	case SEEK_HOLE:
+	case SEEK_END:
 		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY,
 					   &i_gh);
 		if (!error) {
@@ -70,8 +70,21 @@ static loff_t gfs2_llseek(struct file *file, loff_t offset, int whence)
 			gfs2_glock_dq_uninit(&i_gh);
 		}
 		break;
+
+	case SEEK_DATA:
+		error = gfs2_seek_data(file, offset);
+		break;
+
+	case SEEK_HOLE:
+		error = gfs2_seek_hole(file, offset);
+		break;
+
 	case SEEK_CUR:
 	case SEEK_SET:
+		/*
+		 * These don't reference inode->i_size and don't depend on the
+		 * block mapping, so we don't need the glock.
+		 */
 		error = generic_file_llseek(file, offset, whence);
 		break;
 	default:
@@ -107,68 +120,63 @@ static int gfs2_readdir(struct file *file, struct dir_context *ctx)
 	return error;
 }
 
-/**
- * fsflags_cvt
- * @table: A table of 32 u32 flags
- * @val: a 32 bit value to convert
- *
- * This function can be used to convert between fsflags values and
- * GFS2's own flags values.
+/*
+ * struct fsflag_gfs2flag
  *
- * Returns: the converted flags
+ * The FS_JOURNAL_DATA_FL flag maps to GFS2_DIF_INHERIT_JDATA for directories,
+ * and to GFS2_DIF_JDATA for non-directories.
  */
-static u32 fsflags_cvt(const u32 *table, u32 val)
+static struct {
+	u32 fsflag;
+	u32 gfsflag;
+} fsflag_gfs2flag[] = {
+	{FS_SYNC_FL, GFS2_DIF_SYNC},
+	{FS_IMMUTABLE_FL, GFS2_DIF_IMMUTABLE},
+	{FS_APPEND_FL, GFS2_DIF_APPENDONLY},
+	{FS_NOATIME_FL, GFS2_DIF_NOATIME},
+	{FS_INDEX_FL, GFS2_DIF_EXHASH},
+	{FS_TOPDIR_FL, GFS2_DIF_TOPDIR},
+	{FS_JOURNAL_DATA_FL, GFS2_DIF_JDATA | GFS2_DIF_INHERIT_JDATA},
+};
+
+static inline u32 gfs2_gfsflags_to_fsflags(struct inode *inode, u32 gfsflags)
 {
-	u32 res = 0;
-	while(val) {
-		if (val & 1)
-			res |= *table;
-		table++;
-		val >>= 1;
-	}
-	return res;
-}
+	int i;
+	u32 fsflags = 0;
 
-static const u32 fsflags_to_gfs2[32] = {
-	[3] = GFS2_DIF_SYNC,
-	[4] = GFS2_DIF_IMMUTABLE,
-	[5] = GFS2_DIF_APPENDONLY,
-	[7] = GFS2_DIF_NOATIME,
-	[12] = GFS2_DIF_EXHASH,
-	[14] = GFS2_DIF_INHERIT_JDATA,
-	[17] = GFS2_DIF_TOPDIR,
-};
+	if (S_ISDIR(inode->i_mode))
+		gfsflags &= ~GFS2_DIF_JDATA;
+	else
+		gfsflags &= ~GFS2_DIF_INHERIT_JDATA;
 
-static const u32 gfs2_to_fsflags[32] = {
-	[gfs2fl_Sync] = FS_SYNC_FL,
-	[gfs2fl_Immutable] = FS_IMMUTABLE_FL,
-	[gfs2fl_AppendOnly] = FS_APPEND_FL,
-	[gfs2fl_NoAtime] = FS_NOATIME_FL,
-	[gfs2fl_ExHash] = FS_INDEX_FL,
-	[gfs2fl_TopLevel] = FS_TOPDIR_FL,
-	[gfs2fl_InheritJdata] = FS_JOURNAL_DATA_FL,
-};
+	for (i = 0; i < ARRAY_SIZE(fsflag_gfs2flag); i++)
+		if (gfsflags & fsflag_gfs2flag[i].gfsflag)
+			fsflags |= fsflag_gfs2flag[i].fsflag;
+	return fsflags;
+}
 
-static int gfs2_get_flags(struct file *filp, u32 __user *ptr)
+int gfs2_fileattr_get(struct dentry *dentry, struct file_kattr *fa)
 {
-	struct inode *inode = file_inode(filp);
+	struct inode *inode = d_inode(dentry);
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_holder gh;
 	int error;
 	u32 fsflags;
 
+	if (d_is_special(dentry))
+		return -ENOTTY;
+
 	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
 	error = gfs2_glock_nq(&gh);
 	if (error)
-		return error;
+		goto out_uninit;
 
-	fsflags = fsflags_cvt(gfs2_to_fsflags, ip->i_diskflags);
-	if (!S_ISDIR(inode->i_mode) && ip->i_diskflags & GFS2_DIF_JDATA)
-		fsflags |= FS_JOURNAL_DATA_FL;
-	if (put_user(fsflags, ptr))
-		error = -EFAULT;
+	fsflags = gfs2_gfsflags_to_fsflags(inode, ip->i_diskflags);
+
+	fileattr_fill_flags(fa, fsflags);
 
 	gfs2_glock_dq(&gh);
+out_uninit:
 	gfs2_holder_uninit(&gh);
 	return error;
 }
@@ -180,7 +188,7 @@ void gfs2_set_inode_flags(struct inode *inode)
 
 	flags &= ~(S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC|S_NOSEC);
 	if ((ip->i_eattr == 0) && !is_sxid(inode->i_mode))
-		inode->i_flags |= S_NOSEC;
+		flags |= S_NOSEC;
 	if (ip->i_diskflags & GFS2_DIF_IMMUTABLE)
 		flags |= S_IMMUTABLE;
 	if (ip->i_diskflags & GFS2_DIF_APPENDONLY)
@@ -198,20 +206,18 @@ void gfs2_set_inode_flags(struct inode *inode)
 			     GFS2_DIF_APPENDONLY|		\
 			     GFS2_DIF_NOATIME|			\
 			     GFS2_DIF_SYNC|			\
-			     GFS2_DIF_SYSTEM|			\
 			     GFS2_DIF_TOPDIR|			\
 			     GFS2_DIF_INHERIT_JDATA)
 
 /**
- * gfs2_set_flags - set flags on an inode
+ * do_gfs2_set_flags - set flags on an inode
  * @inode: The inode
- * @flags: The flags to set
+ * @reqflags: The flags to set
  * @mask: Indicates which flags are valid
  *
  */
-static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask)
+static int do_gfs2_set_flags(struct inode *inode, u32 reqflags, u32 mask)
 {
-	struct inode *inode = file_inode(filp);
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct buffer_head *bh;
@@ -219,17 +225,9 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask)
 	int error;
 	u32 new_flags, flags;
 
-	error = mnt_want_write_file(filp);
-	if (error)
-		return error;
-
 	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
 	if (error)
-		goto out_drop_write;
-
-	error = -EACCES;
-	if (!inode_owner_or_capable(inode))
-		goto out;
+		return error;
 
 	error = 0;
 	flags = ip->i_diskflags;
@@ -237,32 +235,25 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask)
 	if ((new_flags ^ flags) == 0)
 		goto out;
 
-	error = -EINVAL;
-	if ((new_flags ^ flags) & ~GFS2_FLAGS_USER_SET)
-		goto out;
-
-	error = -EPERM;
-	if (IS_IMMUTABLE(inode) && (new_flags & GFS2_DIF_IMMUTABLE))
-		goto out;
-	if (IS_APPEND(inode) && (new_flags & GFS2_DIF_APPENDONLY))
-		goto out;
-	if (((new_flags ^ flags) & GFS2_DIF_IMMUTABLE) &&
-	    !capable(CAP_LINUX_IMMUTABLE))
-		goto out;
 	if (!IS_IMMUTABLE(inode)) {
-		error = gfs2_permission(inode, MAY_WRITE);
+		error = gfs2_permission(&nop_mnt_idmap, inode, MAY_WRITE);
 		if (error)
 			goto out;
 	}
 	if ((flags ^ new_flags) & GFS2_DIF_JDATA) {
-		if (flags & GFS2_DIF_JDATA)
-			gfs2_log_flush(sdp, ip->i_gl);
+		if (new_flags & GFS2_DIF_JDATA)
+			gfs2_log_flush(sdp, ip->i_gl,
+				       GFS2_LOG_HEAD_FLUSH_NORMAL |
+				       GFS2_LFC_SET_FLAGS);
 		error = filemap_fdatawrite(inode->i_mapping);
 		if (error)
 			goto out;
 		error = filemap_fdatawait(inode->i_mapping);
 		if (error)
 			goto out;
+		truncate_inode_pages(inode->i_mapping, 0);
+		if (new_flags & GFS2_DIF_JDATA)
+			gfs2_ordered_del_inode(ip);
 	}
 	error = gfs2_trans_begin(sdp, RES_DINODE, 0);
 	if (error)
@@ -270,6 +261,7 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask)
 	error = gfs2_meta_inode_buffer(ip, &bh);
 	if (error)
 		goto out_trans_end;
+	inode_set_ctime_current(inode);
 	gfs2_trans_add_meta(ip->i_gl, bh);
 	ip->i_diskflags = new_flags;
 	gfs2_dinode_out(ip, bh->b_data);
@@ -280,45 +272,89 @@ out_trans_end:
 	gfs2_trans_end(sdp);
 out:
 	gfs2_glock_dq_uninit(&gh);
-out_drop_write:
-	mnt_drop_write_file(filp);
 	return error;
 }
 
-static int gfs2_set_flags(struct file *filp, u32 __user *ptr)
+int gfs2_fileattr_set(struct mnt_idmap *idmap,
+		      struct dentry *dentry, struct file_kattr *fa)
+{
+	struct inode *inode = d_inode(dentry);
+	u32 fsflags = fa->flags, gfsflags = 0;
+	u32 mask;
+	int i;
+
+	if (d_is_special(dentry))
+		return -ENOTTY;
+
+	if (fileattr_has_fsx(fa))
+		return -EOPNOTSUPP;
+
+	for (i = 0; i < ARRAY_SIZE(fsflag_gfs2flag); i++) {
+		if (fsflags & fsflag_gfs2flag[i].fsflag) {
+			fsflags &= ~fsflag_gfs2flag[i].fsflag;
+			gfsflags |= fsflag_gfs2flag[i].gfsflag;
+		}
+	}
+	if (fsflags || gfsflags & ~GFS2_FLAGS_USER_SET)
+		return -EINVAL;
+
+	mask = GFS2_FLAGS_USER_SET;
+	if (S_ISDIR(inode->i_mode)) {
+		mask &= ~GFS2_DIF_JDATA;
+	} else {
+		/* The GFS2_DIF_TOPDIR flag is only valid for directories. */
+		if (gfsflags & GFS2_DIF_TOPDIR)
+			return -EINVAL;
+		mask &= ~(GFS2_DIF_TOPDIR | GFS2_DIF_INHERIT_JDATA);
+	}
+
+	return do_gfs2_set_flags(inode, gfsflags, mask);
+}
+
+static int gfs2_getlabel(struct file *filp, char __user *label)
 {
 	struct inode *inode = file_inode(filp);
-	u32 fsflags, gfsflags;
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
 
-	if (get_user(fsflags, ptr))
+	if (copy_to_user(label, sdp->sd_sb.sb_locktable, GFS2_LOCKNAME_LEN))
 		return -EFAULT;
 
-	gfsflags = fsflags_cvt(fsflags_to_gfs2, fsflags);
-	if (!S_ISDIR(inode->i_mode)) {
-		gfsflags &= ~GFS2_DIF_TOPDIR;
-		if (gfsflags & GFS2_DIF_INHERIT_JDATA)
-			gfsflags ^= (GFS2_DIF_JDATA | GFS2_DIF_INHERIT_JDATA);
-		return do_gfs2_set_flags(filp, gfsflags, ~0);
-	}
-	return do_gfs2_set_flags(filp, gfsflags, ~GFS2_DIF_JDATA);
+	return 0;
 }
 
 static long gfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	switch(cmd) {
-	case FS_IOC_GETFLAGS:
-		return gfs2_get_flags(filp, (u32 __user *)arg);
-	case FS_IOC_SETFLAGS:
-		return gfs2_set_flags(filp, (u32 __user *)arg);
 	case FITRIM:
 		return gfs2_fitrim(filp, (void __user *)arg);
+	case FS_IOC_GETFSLABEL:
+		return gfs2_getlabel(filp, (char __user *)arg);
 	}
+
 	return -ENOTTY;
 }
 
+#ifdef CONFIG_COMPAT
+static long gfs2_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
+{
+	switch(cmd) {
+	/* 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
- * @file: The struct file
+ * @filep: The struct file
  * @offset: The file offset of the write
  * @size: The length of the write
  *
@@ -336,103 +372,126 @@ static void gfs2_size_hint(struct file *filep, loff_t offset, size_t size)
 	size_t blks = (size + sdp->sd_sb.sb_bsize - 1) >> sdp->sd_sb.sb_bsize_shift;
 	int hint = min_t(size_t, INT_MAX, blks);
 
-	atomic_set(&ip->i_res->rs_sizehint, hint);
+	if (hint > atomic_read(&ip->i_sizehint))
+		atomic_set(&ip->i_sizehint, hint);
 }
 
 /**
- * gfs2_allocate_page_backing - Use bmap to allocate blocks
- * @page: The (locked) page to allocate backing for
+ * gfs2_allocate_folio_backing - Allocate blocks for a write fault
+ * @folio: The (locked) folio 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, thats ok too.
+ * We try to allocate all the blocks required for the folio in one go.  This
+ * might fail for various reasons, so we keep trying until all the blocks to
+ * back this folio 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_folio_backing(struct folio *folio, size_t length)
 {
-	struct inode *inode = page->mapping->host;
-	struct buffer_head bh;
-	unsigned long size = PAGE_CACHE_SIZE;
-	u64 lblock = page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits);
+	u64 pos = folio_pos(folio);
 
 	do {
-		bh.b_state = 0;
-		bh.b_size = size;
-		gfs2_block_map(inode, lblock, &bh, 1);
-		if (!buffer_mapped(&bh))
+		struct iomap iomap = { };
+
+		if (gfs2_iomap_alloc(folio->mapping->host, pos, length, &iomap))
 			return -EIO;
-		size -= bh.b_size;
-		lblock += (bh.b_size >> inode->i_blkbits);
-	} while(size > 0);
+
+		if (length < iomap.length)
+			iomap.length = length;
+		length -= iomap.length;
+		pos += iomap.length;
+	} while (length > 0);
+
 	return 0;
 }
 
 /**
  * gfs2_page_mkwrite - Make a shared, mmap()ed, page writable
- * @vma: The virtual memory area
- * @page: The page which is about to become writable
+ * @vmf: The virtual memory fault containing the page to become writable
  *
  * When the page becomes writable, we need to ensure that we have
  * blocks allocated on disk to back that page.
  */
 
-static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
+static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf)
 {
-	struct page *page = vmf->page;
-	struct inode *inode = file_inode(vma->vm_file);
+	struct folio *folio = page_folio(vmf->page);
+	struct inode *inode = file_inode(vmf->vma->vm_file);
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
-	unsigned long last_index;
-	u64 pos = page->index << PAGE_CACHE_SHIFT;
+	struct gfs2_alloc_parms ap = {};
+	u64 pos = folio_pos(folio);
 	unsigned int data_blocks, ind_blocks, rblocks;
+	vm_fault_t ret = VM_FAULT_LOCKED;
 	struct gfs2_holder gh;
+	size_t length;
 	loff_t size;
-	int ret;
+	int err;
 
 	sb_start_pagefault(inode->i_sb);
 
-	/* Update file times before taking page lock */
-	file_update_time(vma->vm_file);
+	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+	err = gfs2_glock_nq(&gh);
+	if (err) {
+		ret = vmf_fs_error(err);
+		goto out_uninit;
+	}
 
-	ret = get_write_access(inode);
-	if (ret)
-		goto out;
+	/* Check folio index against inode size */
+	size = i_size_read(inode);
+	if (pos >= size) {
+		ret = VM_FAULT_SIGBUS;
+		goto out_unlock;
+	}
 
-	ret = gfs2_rs_alloc(ip);
-	if (ret)
-		goto out_write_access;
+	/* Update file times before taking folio lock */
+	file_update_time(vmf->vma->vm_file);
 
-	gfs2_size_hint(vma->vm_file, pos, PAGE_CACHE_SIZE);
+	/* folio is wholly or partially inside EOF */
+	if (size - pos < folio_size(folio))
+		length = size - pos;
+	else
+		length = folio_size(folio);
 
-	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
-	ret = gfs2_glock_nq(&gh);
-	if (ret)
-		goto out_uninit;
+	gfs2_size_hint(vmf->vma->vm_file, pos, 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_CACHE_SIZE)) {
-		lock_page(page);
-		if (!PageUptodate(page) || page->mapping != inode->i_mapping) {
-			ret = -EAGAIN;
-			unlock_page(page);
+	/*
+	 * iomap_writepage / iomap_writepages currently don't support inline
+	 * files, so always unstuff here.
+	 */
+
+	if (!gfs2_is_stuffed(ip) &&
+	    !gfs2_write_alloc_required(ip, pos, length)) {
+		folio_lock(folio);
+		if (!folio_test_uptodate(folio) ||
+		    folio->mapping != inode->i_mapping) {
+			ret = VM_FAULT_NOPAGE;
+			folio_unlock(folio);
 		}
 		goto out_unlock;
 	}
 
-	ret = gfs2_rindex_update(sdp);
-	if (ret)
+	err = gfs2_rindex_update(sdp);
+	if (err) {
+		ret = vmf_fs_error(err);
 		goto out_unlock;
+	}
 
-	ret = gfs2_quota_lock_check(ip);
-	if (ret)
+	gfs2_write_calc_reserv(ip, length, &data_blocks, &ind_blocks);
+	ap.target = data_blocks + ind_blocks;
+	err = gfs2_quota_lock_check(ip, &ap);
+	if (err) {
+		ret = vmf_fs_error(err);
 		goto out_unlock;
-	gfs2_write_calc_reserv(ip, PAGE_CACHE_SIZE, &data_blocks, &ind_blocks);
-	ret = gfs2_inplace_reserve(ip, data_blocks + ind_blocks, 0);
-	if (ret)
+	}
+	err = gfs2_inplace_reserve(ip, &ap);
+	if (err) {
+		ret = vmf_fs_error(err);
 		goto out_quota_unlock;
+	}
 
 	rblocks = RES_DINODE + ind_blocks;
 	if (gfs2_is_jdata(ip))
@@ -441,35 +500,38 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 		rblocks += RES_STATFS + RES_QUOTA;
 		rblocks += gfs2_rg_blocks(ip, data_blocks + ind_blocks);
 	}
-	ret = gfs2_trans_begin(sdp, rblocks, 0);
-	if (ret)
+	err = gfs2_trans_begin(sdp, rblocks, 0);
+	if (err) {
+		ret = vmf_fs_error(err);
 		goto out_trans_fail;
+	}
 
-	lock_page(page);
-	ret = -EINVAL;
-	size = i_size_read(inode);
-	last_index = (size - 1) >> PAGE_CACHE_SHIFT;
-	/* Check page index against inode size */
-	if (size == 0 || (page->index > last_index))
-		goto out_trans_end;
+	/* Unstuff, if required, and allocate backing blocks for folio */
+	if (gfs2_is_stuffed(ip)) {
+		err = gfs2_unstuff_dinode(ip);
+		if (err) {
+			ret = vmf_fs_error(err);
+			goto out_trans_end;
+		}
+	}
 
-	ret = -EAGAIN;
+	folio_lock(folio);
 	/* If truncated, we must retry the operation, we may have raced
 	 * with the glock demotion code.
 	 */
-	if (!PageUptodate(page) || page->mapping != inode->i_mapping)
-		goto out_trans_end;
+	if (!folio_test_uptodate(folio) || folio->mapping != inode->i_mapping) {
+		ret = VM_FAULT_NOPAGE;
+		goto out_page_locked;
+	}
 
-	/* Unstuff, if required, and allocate backing blocks for page */
-	ret = 0;
-	if (gfs2_is_stuffed(ip))
-		ret = gfs2_unstuff_dinode(ip, page);
-	if (ret == 0)
-		ret = gfs2_allocate_page_backing(page);
+	err = gfs2_allocate_folio_backing(folio, length);
+	if (err)
+		ret = vmf_fs_error(err);
 
+out_page_locked:
+	if (ret != VM_FAULT_LOCKED)
+		folio_unlock(folio);
 out_trans_end:
-	if (ret)
-		unlock_page(page);
 	gfs2_trans_end(sdp);
 out_trans_fail:
 	gfs2_inplace_release(ip);
@@ -479,25 +541,43 @@ out_unlock:
 	gfs2_glock_dq(&gh);
 out_uninit:
 	gfs2_holder_uninit(&gh);
-	if (ret == 0) {
-		set_page_dirty(page);
-		wait_for_stable_page(page);
+	if (ret == VM_FAULT_LOCKED) {
+		folio_mark_dirty(folio);
+		folio_wait_stable(folio);
 	}
-out_write_access:
-	put_write_access(inode);
-out:
 	sb_end_pagefault(inode->i_sb);
-	return block_page_mkwrite_return(ret);
+	return ret;
+}
+
+static vm_fault_t gfs2_fault(struct vm_fault *vmf)
+{
+	struct inode *inode = file_inode(vmf->vma->vm_file);
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	vm_fault_t ret;
+	int err;
+
+	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
+	err = gfs2_glock_nq(&gh);
+	if (err) {
+		ret = vmf_fs_error(err);
+		goto out_uninit;
+	}
+	ret = filemap_fault(vmf);
+	gfs2_glock_dq(&gh);
+out_uninit:
+	gfs2_holder_uninit(&gh);
+	return ret;
 }
 
 static const struct vm_operations_struct gfs2_vm_ops = {
-	.fault = filemap_fault,
+	.fault = gfs2_fault,
+	.map_pages = filemap_map_pages,
 	.page_mkwrite = gfs2_page_mkwrite,
-	.remap_pages = generic_file_remap_pages,
 };
 
 /**
- * gfs2_mmap -
+ * gfs2_mmap
  * @file: The file to map
  * @vma: The VMA which described the mapping
  *
@@ -552,6 +632,9 @@ int gfs2_open_common(struct inode *inode, struct file *file)
 		ret = generic_file_open(inode, file);
 		if (ret)
 			return ret;
+
+		if (!gfs2_is_jdata(GFS2_I(inode)))
+			file->f_mode |= FMODE_CAN_ODIRECT;
 	}
 
 	fp = kzalloc(sizeof(struct gfs2_file), GFP_NOFS);
@@ -562,7 +645,17 @@ 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;
 }
 
 /**
@@ -617,10 +710,11 @@ 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_rs_delete(ip);
+	if (file->f_mode & FMODE_WRITE) {
+		if (gfs2_rs_active(&ip->i_res))
+			gfs2_rs_delete(ip);
+		gfs2_qa_put(ip);
+	}
 	return 0;
 }
 
@@ -650,7 +744,7 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end,
 {
 	struct address_space *mapping = file->f_mapping;
 	struct inode *inode = mapping->host;
-	int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC);
+	int sync_state = inode_state_read_once(inode) & I_DIRTY;
 	struct gfs2_inode *ip = GFS2_I(inode);
 	int ret = 0, ret1 = 0;
 
@@ -660,6 +754,8 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end,
 			return ret1;
 	}
 
+	if (!gfs2_is_jdata(ip))
+		sync_state &= ~I_DIRTY_PAGES;
 	if (datasync)
 		sync_state &= ~I_DIRTY_SYNC;
 
@@ -668,22 +764,331 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end,
 		if (ret)
 			return ret;
 		if (gfs2_is_jdata(ip))
-			filemap_write_and_wait(mapping);
+			ret = file_write_and_wait(file);
+		if (ret)
+			return ret;
 		gfs2_ail_flush(ip->i_gl, 1);
 	}
 
 	if (mapping->nrpages)
-		ret = filemap_fdatawait_range(mapping, start, end);
+		ret = file_fdatawait_range(file, start, end);
 
 	return ret ? ret : ret1;
 }
 
+static inline bool should_fault_in_pages(struct iov_iter *i,
+					 struct kiocb *iocb,
+					 size_t *prev_count,
+					 size_t *window_size)
+{
+	size_t count = iov_iter_count(i);
+	size_t size, offs;
+
+	if (!count)
+		return false;
+	if (!user_backed_iter(i))
+		return false;
+
+	/*
+	 * Try to fault in multiple pages initially.  When that doesn't result
+	 * in any progress, fall back to a single page.
+	 */
+	size = PAGE_SIZE;
+	offs = offset_in_page(iocb->ki_pos);
+	if (*prev_count != count) {
+		size_t nr_dirtied;
+
+		nr_dirtied = max(current->nr_dirtied_pause -
+				 current->nr_dirtied, 8);
+		size = min_t(size_t, SZ_1M, nr_dirtied << PAGE_SHIFT);
+	}
+
+	*prev_count = count;
+	*window_size = size - offs;
+	return true;
+}
+
+static ssize_t gfs2_file_direct_read(struct kiocb *iocb, struct iov_iter *to,
+				     struct gfs2_holder *gh)
+{
+	struct file *file = iocb->ki_filp;
+	struct gfs2_inode *ip = GFS2_I(file->f_mapping->host);
+	size_t prev_count = 0, window_size = 0;
+	size_t read = 0;
+	ssize_t ret;
+
+	/*
+	 * In this function, we disable page faults when we're holding the
+	 * inode glock while doing I/O.  If a page fault occurs, we indicate
+	 * that the inode glock should be dropped, fault in the pages manually,
+	 * and retry.
+	 *
+	 * Unlike generic_file_read_iter, for reads, iomap_dio_rw can trigger
+	 * physical as well as manual page faults, and we need to disable both
+	 * kinds.
+	 *
+	 * For direct I/O, gfs2 takes the inode glock in deferred mode.  This
+	 * locking mode is compatible with other deferred holders, so multiple
+	 * processes and nodes can do direct I/O to a file at the same time.
+	 * There's no guarantee that reads or writes will be atomic.  Any
+	 * coordination among readers and writers needs to happen externally.
+	 */
+
+	if (!iov_iter_count(to))
+		return 0; /* skip atime */
+
+	gfs2_holder_init(ip->i_gl, LM_ST_DEFERRED, 0, gh);
+retry:
+	ret = gfs2_glock_nq(gh);
+	if (ret)
+		goto out_uninit;
+	pagefault_disable();
+	to->nofault = true;
+	ret = iomap_dio_rw(iocb, to, &gfs2_iomap_ops, NULL,
+			   IOMAP_DIO_PARTIAL, NULL, read);
+	to->nofault = false;
+	pagefault_enable();
+	if (ret <= 0 && ret != -EFAULT)
+		goto out_unlock;
+	/* No increment (+=) because iomap_dio_rw returns a cumulative value. */
+	if (ret > 0)
+		read = ret;
+
+	if (should_fault_in_pages(to, iocb, &prev_count, &window_size)) {
+		gfs2_glock_dq(gh);
+		window_size -= fault_in_iov_iter_writeable(to, window_size);
+		if (window_size)
+			goto retry;
+	}
+out_unlock:
+	if (gfs2_holder_queued(gh))
+		gfs2_glock_dq(gh);
+out_uninit:
+	gfs2_holder_uninit(gh);
+	/* User space doesn't expect partial success. */
+	if (ret < 0)
+		return ret;
+	return read;
+}
+
+static ssize_t gfs2_file_direct_write(struct kiocb *iocb, struct iov_iter *from,
+				      struct gfs2_holder *gh)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file->f_mapping->host;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	size_t prev_count = 0, window_size = 0;
+	size_t written = 0;
+	bool enough_retries;
+	ssize_t ret;
+
+	/*
+	 * In this function, we disable page faults when we're holding the
+	 * inode glock while doing I/O.  If a page fault occurs, we indicate
+	 * that the inode glock should be dropped, fault in the pages manually,
+	 * and retry.
+	 *
+	 * For writes, iomap_dio_rw only triggers manual page faults, so we
+	 * don't need to disable physical ones.
+	 */
+
+	/*
+	 * Deferred lock, even if its a write, since we do no allocation on
+	 * this path. All we need to change is the atime, and this lock mode
+	 * ensures that other nodes have flushed their buffered read caches
+	 * (i.e. their page cache entries for this inode). We do not,
+	 * unfortunately, have the option of only flushing a range like the
+	 * VFS does.
+	 */
+	gfs2_holder_init(ip->i_gl, LM_ST_DEFERRED, 0, gh);
+retry:
+	ret = gfs2_glock_nq(gh);
+	if (ret)
+		goto out_uninit;
+	/* Silently fall back to buffered I/O when writing beyond EOF */
+	if (iocb->ki_pos + iov_iter_count(from) > i_size_read(&ip->i_inode))
+		goto out_unlock;
+
+	from->nofault = true;
+	ret = iomap_dio_rw(iocb, from, &gfs2_iomap_ops, NULL,
+			   IOMAP_DIO_PARTIAL, NULL, written);
+	from->nofault = false;
+	if (ret <= 0) {
+		if (ret == -ENOTBLK)
+			ret = 0;
+		if (ret != -EFAULT)
+			goto out_unlock;
+	}
+	/* No increment (+=) because iomap_dio_rw returns a cumulative value. */
+	if (ret > 0)
+		written = ret;
+
+	enough_retries = prev_count == iov_iter_count(from) &&
+			 window_size <= PAGE_SIZE;
+	if (should_fault_in_pages(from, iocb, &prev_count, &window_size)) {
+		gfs2_glock_dq(gh);
+		window_size -= fault_in_iov_iter_readable(from, window_size);
+		if (window_size) {
+			if (!enough_retries)
+				goto retry;
+			/* fall back to buffered I/O */
+			ret = 0;
+		}
+	}
+out_unlock:
+	if (gfs2_holder_queued(gh))
+		gfs2_glock_dq(gh);
+out_uninit:
+	gfs2_holder_uninit(gh);
+	/* User space doesn't expect partial success. */
+	if (ret < 0)
+		return ret;
+	return written;
+}
+
+static ssize_t gfs2_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct gfs2_inode *ip;
+	struct gfs2_holder gh;
+	size_t prev_count = 0, window_size = 0;
+	size_t read = 0;
+	ssize_t ret;
+
+	/*
+	 * In this function, we disable page faults when we're holding the
+	 * inode glock while doing I/O.  If a page fault occurs, we indicate
+	 * that the inode glock should be dropped, fault in the pages manually,
+	 * and retry.
+	 */
+
+	if (iocb->ki_flags & IOCB_DIRECT)
+		return gfs2_file_direct_read(iocb, to, &gh);
+
+	pagefault_disable();
+	iocb->ki_flags |= IOCB_NOIO;
+	ret = generic_file_read_iter(iocb, to);
+	iocb->ki_flags &= ~IOCB_NOIO;
+	pagefault_enable();
+	if (ret >= 0) {
+		if (!iov_iter_count(to))
+			return ret;
+		read = ret;
+	} else if (ret != -EFAULT) {
+		if (ret != -EAGAIN)
+			return ret;
+		if (iocb->ki_flags & IOCB_NOWAIT)
+			return ret;
+	}
+	ip = GFS2_I(iocb->ki_filp->f_mapping->host);
+	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
+retry:
+	ret = gfs2_glock_nq(&gh);
+	if (ret)
+		goto out_uninit;
+	pagefault_disable();
+	ret = generic_file_read_iter(iocb, to);
+	pagefault_enable();
+	if (ret <= 0 && ret != -EFAULT)
+		goto out_unlock;
+	if (ret > 0)
+		read += ret;
+
+	if (should_fault_in_pages(to, iocb, &prev_count, &window_size)) {
+		gfs2_glock_dq(&gh);
+		window_size -= fault_in_iov_iter_writeable(to, window_size);
+		if (window_size)
+			goto retry;
+	}
+out_unlock:
+	if (gfs2_holder_queued(&gh))
+		gfs2_glock_dq(&gh);
+out_uninit:
+	gfs2_holder_uninit(&gh);
+	return read ? read : ret;
+}
+
+static ssize_t gfs2_file_buffered_write(struct kiocb *iocb,
+					struct iov_iter *from,
+					struct gfs2_holder *gh)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
+	struct gfs2_holder *statfs_gh = NULL;
+	size_t prev_count = 0, window_size = 0;
+	size_t orig_count = iov_iter_count(from);
+	size_t written = 0;
+	ssize_t ret;
+
+	/*
+	 * In this function, we disable page faults when we're holding the
+	 * inode glock while doing I/O.  If a page fault occurs, we indicate
+	 * that the inode glock should be dropped, fault in the pages manually,
+	 * and retry.
+	 */
+
+	if (inode == sdp->sd_rindex) {
+		statfs_gh = kmalloc(sizeof(*statfs_gh), GFP_NOFS);
+		if (!statfs_gh)
+			return -ENOMEM;
+	}
+
+	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, gh);
+	if (should_fault_in_pages(from, iocb, &prev_count, &window_size)) {
+retry:
+		window_size -= fault_in_iov_iter_readable(from, window_size);
+		if (!window_size) {
+			ret = -EFAULT;
+			goto out_uninit;
+		}
+		from->count = min(from->count, window_size);
+	}
+	ret = gfs2_glock_nq(gh);
+	if (ret)
+		goto out_uninit;
+
+	if (inode == sdp->sd_rindex) {
+		struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
+
+		ret = gfs2_glock_nq_init(m_ip->i_gl, LM_ST_EXCLUSIVE,
+					 GL_NOCACHE, statfs_gh);
+		if (ret)
+			goto out_unlock;
+	}
+
+	pagefault_disable();
+	ret = iomap_file_buffered_write(iocb, from, &gfs2_iomap_ops,
+			&gfs2_iomap_write_ops, NULL);
+	pagefault_enable();
+	if (ret > 0)
+		written += ret;
+
+	if (inode == sdp->sd_rindex)
+		gfs2_glock_dq_uninit(statfs_gh);
+
+	if (ret <= 0 && ret != -EFAULT)
+		goto out_unlock;
+
+	from->count = orig_count - written;
+	if (should_fault_in_pages(from, iocb, &prev_count, &window_size)) {
+		gfs2_glock_dq(gh);
+		goto retry;
+	}
+out_unlock:
+	if (gfs2_holder_queued(gh))
+		gfs2_glock_dq(gh);
+out_uninit:
+	gfs2_holder_uninit(gh);
+	kfree(statfs_gh);
+	from->count = orig_count - written;
+	return written ? written : ret;
+}
+
 /**
- * gfs2_file_aio_write - Perform a write to a file
+ * gfs2_file_write_iter - Perform a write to a file
  * @iocb: The io context
- * @iov: The data to write
- * @nr_segs: Number of @iov segments
- * @pos: The file position
+ * @from: The data to write
  *
  * We have to do a lock/unlock here to refresh the inode size for
  * O_APPEND writes, otherwise we can land up writing at the wrong
@@ -692,41 +1097,90 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end,
  *
  */
 
-static ssize_t gfs2_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
-				   unsigned long nr_segs, loff_t pos)
+static ssize_t gfs2_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct file *file = iocb->ki_filp;
-	size_t writesize = iov_length(iov, nr_segs);
-	struct gfs2_inode *ip = GFS2_I(file_inode(file));
-	int ret;
-
-	ret = gfs2_rs_alloc(ip);
-	if (ret)
-		return ret;
-
-	gfs2_size_hint(file, pos, writesize);
+	struct inode *inode = file_inode(file);
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	ssize_t ret;
 
-	if (file->f_flags & O_APPEND) {
-		struct gfs2_holder gh;
+	gfs2_size_hint(file, iocb->ki_pos, iov_iter_count(from));
 
+	if (iocb->ki_flags & IOCB_APPEND) {
 		ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
 		if (ret)
 			return ret;
 		gfs2_glock_dq_uninit(&gh);
 	}
 
-	return generic_file_aio_write(iocb, iov, nr_segs, pos);
+	inode_lock(inode);
+	ret = generic_write_checks(iocb, from);
+	if (ret <= 0)
+		goto out_unlock;
+
+	ret = file_remove_privs(file);
+	if (ret)
+		goto out_unlock;
+
+	if (iocb->ki_flags & IOCB_DIRECT) {
+		struct address_space *mapping = file->f_mapping;
+		ssize_t buffered, ret2;
+
+		/*
+		 * Note that under direct I/O, we don't allow and inode
+		 * timestamp updates, so we're not calling file_update_time()
+		 * here.
+		 */
+
+		ret = gfs2_file_direct_write(iocb, from, &gh);
+		if (ret < 0 || !iov_iter_count(from))
+			goto out_unlock;
+
+		iocb->ki_flags |= IOCB_DSYNC;
+		buffered = gfs2_file_buffered_write(iocb, from, &gh);
+		if (unlikely(buffered <= 0)) {
+			if (!ret)
+				ret = buffered;
+			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.  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.
+		 */
+		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 {
+		ret = file_update_time(file);
+		if (ret)
+			goto out_unlock;
+
+		ret = gfs2_file_buffered_write(iocb, from, &gh);
+		if (likely(ret > 0))
+			ret = generic_write_sync(iocb, ret);
+	}
+
+out_unlock:
+	inode_unlock(inode);
+	return ret;
 }
 
 static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len,
 			   int mode)
 {
+	struct super_block *sb = inode->i_sb;
 	struct gfs2_inode *ip = GFS2_I(inode);
+	loff_t end = offset + len;
 	struct buffer_head *dibh;
 	int error;
-	loff_t size = len;
-	unsigned int nr_blks;
-	sector_t lblock = offset >> inode->i_blkbits;
 
 	error = gfs2_meta_inode_buffer(ip, &dibh);
 	if (unlikely(error))
@@ -735,54 +1189,57 @@ static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len,
 	gfs2_trans_add_meta(ip->i_gl, dibh);
 
 	if (gfs2_is_stuffed(ip)) {
-		error = gfs2_unstuff_dinode(ip, NULL);
+		error = gfs2_unstuff_dinode(ip);
 		if (unlikely(error))
 			goto out;
 	}
 
-	while (len) {
-		struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 };
-		bh_map.b_size = len;
-		set_buffer_zeronew(&bh_map);
+	while (offset < end) {
+		struct iomap iomap = { };
 
-		error = gfs2_block_map(inode, lblock, &bh_map, 1);
-		if (unlikely(error))
+		error = gfs2_iomap_alloc(inode, offset, end - offset, &iomap);
+		if (error)
 			goto out;
-		len -= bh_map.b_size;
-		nr_blks = bh_map.b_size >> inode->i_blkbits;
-		lblock += nr_blks;
-		if (!buffer_new(&bh_map))
+		offset = iomap.offset + iomap.length;
+		if (!(iomap.flags & IOMAP_F_NEW))
 			continue;
-		if (unlikely(!buffer_zeronew(&bh_map))) {
-			error = -EIO;
+		error = sb_issue_zeroout(sb, iomap.addr >> inode->i_blkbits,
+					 iomap.length >> inode->i_blkbits,
+					 GFP_NOFS);
+		if (error) {
+			fs_err(GFS2_SB(inode), "Failed to zero data buffers\n");
 			goto out;
 		}
 	}
-	if (offset + size > inode->i_size && !(mode & FALLOC_FL_KEEP_SIZE))
-		i_size_write(inode, offset + size);
-
-	mark_inode_dirty(inode);
-
 out:
 	brelse(dibh);
 	return error;
 }
 
-static void calc_max_reserv(struct gfs2_inode *ip, loff_t max, loff_t *len,
-			    unsigned int *data_blocks, unsigned int *ind_blocks)
+/**
+ * calc_max_reserv() - Reverse of write_calc_reserv. Given a number of
+ *                     blocks, determine how many bytes can be written.
+ * @ip:          The inode in question.
+ * @len:         Max cap of bytes. What we return in *len must be <= this.
+ * @data_blocks: Compute and return the number of data blocks needed
+ * @ind_blocks:  Compute and return the number of indirect blocks needed
+ * @max_blocks:  The total blocks available to work with.
+ *
+ * Returns: void, but @len, @data_blocks and @ind_blocks are filled in.
+ */
+static void calc_max_reserv(struct gfs2_inode *ip, loff_t *len,
+			    unsigned int *data_blocks, unsigned int *ind_blocks,
+			    unsigned int max_blocks)
 {
+	loff_t max = *len;
 	const struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
-	unsigned int max_blocks = ip->i_rgd->rd_free_clone;
 	unsigned int tmp, max_data = max_blocks - 3 * (sdp->sd_max_height - 1);
 
 	for (tmp = max_data; tmp > sdp->sd_diptrs;) {
 		tmp = DIV_ROUND_UP(tmp, sdp->sd_inptrs);
 		max_data -= tmp;
 	}
-	/* This calculation isn't the exact reverse of gfs2_write_calc_reserve,
-	   so it might end up with fewer data blocks */
-	if (max_data <= *data_blocks)
-		return;
+
 	*data_blocks = max_data;
 	*ind_blocks = max_blocks - max_data;
 	*len = ((loff_t)max_data - 3) << sdp->sd_sb.sb_bsize_shift;
@@ -792,25 +1249,22 @@ static void calc_max_reserv(struct gfs2_inode *ip, loff_t max, loff_t *len,
 	}
 }
 
-static long gfs2_fallocate(struct file *file, int mode, loff_t offset,
-			   loff_t len)
+static long __gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 {
 	struct inode *inode = file_inode(file);
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_alloc_parms ap = {};
 	unsigned int data_blocks = 0, ind_blocks = 0, rblocks;
-	loff_t bytes, max_bytes;
+	loff_t bytes, max_bytes, max_blks;
 	int error;
 	const loff_t pos = offset;
 	const loff_t count = len;
 	loff_t bsize_mask = ~((loff_t)sdp->sd_sb.sb_bsize - 1);
 	loff_t next = (offset + len - 1) >> sdp->sd_sb.sb_bsize_shift;
 	loff_t max_chunk_size = UINT_MAX & bsize_mask;
-	next = (next + 1) << sdp->sd_sb.sb_bsize_shift;
 
-	/* We only support the FALLOC_FL_KEEP_SIZE mode */
-	if (mode & ~FALLOC_FL_KEEP_SIZE)
-		return -EOPNOTSUPP;
+	next = (next + 1) << sdp->sd_sb.sb_bsize_shift;
 
 	offset &= bsize_mask;
 
@@ -822,17 +1276,11 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset,
 	if (bytes == 0)
 		bytes = sdp->sd_sb.sb_bsize;
 
-	error = gfs2_rs_alloc(ip);
-	if (error)
-		return error;
-
-	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh);
-	error = gfs2_glock_nq(&ip->i_gh);
-	if (unlikely(error))
-		goto out_uninit;
-
 	gfs2_size_hint(file, offset, len);
 
+	gfs2_write_calc_reserv(ip, PAGE_SIZE, &data_blocks, &ind_blocks);
+	ap.min_target = data_blocks + ind_blocks;
+
 	while (len > 0) {
 		if (len < bytes)
 			bytes = len;
@@ -841,27 +1289,42 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset,
 			offset += bytes;
 			continue;
 		}
-		error = gfs2_quota_lock_check(ip);
-		if (error)
-			goto out_unlock;
 
-retry:
+		/* We need to determine how many bytes we can actually
+		 * fallocate without exceeding quota or going over the
+		 * end of the fs. We start off optimistically by assuming
+		 * we can write max_bytes */
+		max_bytes = (len > max_chunk_size) ? max_chunk_size : len;
+
+		/* Since max_bytes is most likely a theoretical max, we
+		 * calculate a more realistic 'bytes' to serve as a good
+		 * starting point for the number of bytes we may be able
+		 * to write */
 		gfs2_write_calc_reserv(ip, bytes, &data_blocks, &ind_blocks);
+		ap.target = data_blocks + ind_blocks;
 
-		error = gfs2_inplace_reserve(ip, data_blocks + ind_blocks, 0);
-		if (error) {
-			if (error == -ENOSPC && bytes > sdp->sd_sb.sb_bsize) {
-				bytes >>= 1;
-				bytes &= bsize_mask;
-				if (bytes == 0)
-					bytes = sdp->sd_sb.sb_bsize;
-				goto retry;
-			}
+		error = gfs2_quota_lock_check(ip, &ap);
+		if (error)
+			return error;
+		/* ap.allowed tells us how many blocks quota will allow
+		 * us to write. Check if this reduces max_blks */
+		max_blks = UINT_MAX;
+		if (ap.allowed)
+			max_blks = ap.allowed;
+
+		error = gfs2_inplace_reserve(ip, &ap);
+		if (error)
 			goto out_qunlock;
-		}
-		max_bytes = bytes;
-		calc_max_reserv(ip, (len > max_chunk_size)? max_chunk_size: len,
-				&max_bytes, &data_blocks, &ind_blocks);
+
+		/* check if the selected rgrp limits our max_blks further */
+		if (ip->i_res.rs_reserved < max_blks)
+			max_blks = ip->i_res.rs_reserved;
+
+		/* Almost done. Calculate bytes that can be written using
+		 * max_blks. We also recompute max_bytes, data_blocks and
+		 * ind_blocks */
+		calc_max_reserv(ip, &max_bytes, &data_blocks,
+				&ind_blocks, max_blks);
 
 		rblocks = RES_DINODE + ind_blocks + RES_STATFS + RES_QUOTA +
 			  RES_RG_HDR + gfs2_rg_blocks(ip, data_blocks + ind_blocks);
@@ -869,7 +1332,7 @@ retry:
 			rblocks += data_blocks ? data_blocks : 1;
 
 		error = gfs2_trans_begin(sdp, rblocks,
-					 PAGE_CACHE_SIZE/sdp->sd_sb.sb_bsize);
+					 PAGE_SIZE >> inode->i_blkbits);
 		if (error)
 			goto out_trans_fail;
 
@@ -885,43 +1348,86 @@ retry:
 		gfs2_quota_unlock(ip);
 	}
 
-	if (error == 0)
-		error = generic_write_sync(file, pos, count);
-	goto out_unlock;
+	if (!(mode & FALLOC_FL_KEEP_SIZE) && (pos + count) > inode->i_size)
+		i_size_write(inode, pos + count);
+	file_update_time(file);
+	mark_inode_dirty(inode);
+
+	if ((file->f_flags & O_DSYNC) || IS_SYNC(file->f_mapping->host))
+		return vfs_fsync_range(file, pos, pos + count - 1,
+			       (file->f_flags & __O_SYNC) ? 0 : 1);
+	return 0;
 
 out_trans_fail:
 	gfs2_inplace_release(ip);
 out_qunlock:
 	gfs2_quota_unlock(ip);
-out_unlock:
-	gfs2_glock_dq(&ip->i_gh);
-out_uninit:
-	gfs2_holder_uninit(&ip->i_gh);
 	return error;
 }
 
-#ifdef CONFIG_GFS2_FS_LOCKING_DLM
+static long gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
+{
+	struct inode *inode = file_inode(file);
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	int ret;
 
-/**
- * gfs2_setlease - acquire/release a file lease
- * @file: the file pointer
- * @arg: lease type
- * @fl: file lock
- *
- * We don't currently have a way to enforce a lease across the whole
- * cluster; until we do, disable leases (by just returning -EINVAL),
- * unless the administrator has requested purely local locking.
- *
- * Locking: called under i_lock
- *
- * Returns: errno
- */
+	if (mode & ~(FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE))
+		return -EOPNOTSUPP;
+	/* fallocate is needed by gfs2_grow to reserve space in the rindex */
+	if (gfs2_is_jdata(ip) && inode != sdp->sd_rindex)
+		return -EOPNOTSUPP;
 
-static int gfs2_setlease(struct file *file, long arg, struct file_lock **fl)
+	inode_lock(inode);
+
+	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+	ret = gfs2_glock_nq(&gh);
+	if (ret)
+		goto out_uninit;
+
+	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
+	    (offset + len) > inode->i_size) {
+		ret = inode_newsize_ok(inode, offset + len);
+		if (ret)
+			goto out_unlock;
+	}
+
+	ret = get_write_access(inode);
+	if (ret)
+		goto out_unlock;
+
+	if (mode & FALLOC_FL_PUNCH_HOLE) {
+		ret = __gfs2_punch_hole(file, offset, len);
+	} else {
+		ret = __gfs2_fallocate(file, mode, offset, len);
+		if (ret)
+			gfs2_rs_deltree(&ip->i_res);
+	}
+
+	put_write_access(inode);
+out_unlock:
+	gfs2_glock_dq(&gh);
+out_uninit:
+	gfs2_holder_uninit(&gh);
+	inode_unlock(inode);
+	return ret;
+}
+
+static ssize_t gfs2_file_splice_write(struct pipe_inode_info *pipe,
+				      struct file *out, loff_t *ppos,
+				      size_t len, unsigned int flags)
 {
-	return -EINVAL;
+	ssize_t ret;
+
+	gfs2_size_hint(out, *ppos, len);
+
+	ret = iter_file_splice_write(pipe, out, ppos, len, flags);
+	return ret;
 }
 
+#ifdef CONFIG_GFS2_FS_LOCKING_DLM
+
 /**
  * gfs2_lock - acquire/release a posix lock on a file
  * @file: the file pointer
@@ -936,28 +1442,44 @@ static int gfs2_lock(struct file *file, int cmd, struct file_lock *fl)
 	struct gfs2_inode *ip = GFS2_I(file->f_mapping->host);
 	struct gfs2_sbd *sdp = GFS2_SB(file->f_mapping->host);
 	struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+	int ret;
 
-	if (!(fl->fl_flags & FL_POSIX))
-		return -ENOLCK;
-	if (__mandatory_lock(&ip->i_inode) && fl->fl_type != F_UNLCK)
+	if (!(fl->c.flc_flags & FL_POSIX))
 		return -ENOLCK;
-
-	if (cmd == F_CANCELLK) {
-		/* Hack: */
-		cmd = F_SETLK;
-		fl->fl_type = F_UNLCK;
-	}
-	if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) {
-		if (fl->fl_type == F_UNLCK)
-			posix_lock_file_wait(file, fl);
+	if (gfs2_withdrawn(sdp)) {
+		if (lock_is_unlock(fl))
+			locks_lock_file_wait(file, fl);
 		return -EIO;
 	}
-	if (IS_GETLK(cmd))
-		return dlm_posix_get(ls->ls_dlm, ip->i_no_addr, file, fl);
-	else if (fl->fl_type == F_UNLCK)
-		return dlm_posix_unlock(ls->ls_dlm, ip->i_no_addr, file, fl);
-	else
-		return dlm_posix_lock(ls->ls_dlm, ip->i_no_addr, file, cmd, fl);
+	down_read(&ls->ls_sem);
+	ret = -ENODEV;
+	if (likely(ls->ls_dlm != NULL)) {
+		if (cmd == F_CANCELLK)
+			ret = dlm_posix_cancel(ls->ls_dlm, ip->i_no_addr, file, fl);
+		else if (IS_GETLK(cmd))
+			ret = dlm_posix_get(ls->ls_dlm, ip->i_no_addr, file, fl);
+		else if (lock_is_unlock(fl))
+			ret = dlm_posix_unlock(ls->ls_dlm, ip->i_no_addr, file, fl);
+		else
+			ret = dlm_posix_lock(ls->ls_dlm, ip->i_no_addr, file, cmd, fl);
+	}
+	up_read(&ls->ls_sem);
+	return ret;
+}
+
+static void __flock_holder_uninit(struct file *file, struct gfs2_holder *fl_gh)
+{
+	struct gfs2_glock *gl = gfs2_glock_hold(fl_gh->gh_gl);
+
+	/*
+	 * Make sure gfs2_glock_put() won't sleep under the file->f_lock
+	 * spinlock.
+	 */
+
+	spin_lock(&file->f_lock);
+	gfs2_holder_uninit(fl_gh);
+	spin_unlock(&file->f_lock);
+	gfs2_glock_put(gl);
 }
 
 static int do_flock(struct file *file, int cmd, struct file_lock *fl)
@@ -967,37 +1489,51 @@ static int do_flock(struct file *file, int cmd, struct file_lock *fl)
 	struct gfs2_inode *ip = GFS2_I(file_inode(file));
 	struct gfs2_glock *gl;
 	unsigned int state;
-	int flags;
+	u16 flags;
 	int error = 0;
+	int sleeptime;
 
-	state = (fl->fl_type == F_WRLCK) ? LM_ST_EXCLUSIVE : LM_ST_SHARED;
-	flags = (IS_SETLKW(cmd) ? 0 : LM_FLAG_TRY) | GL_EXACT | GL_NOCACHE;
+	state = lock_is_write(fl) ? LM_ST_EXCLUSIVE : LM_ST_SHARED;
+	flags = GL_EXACT | GL_NOPID;
+	if (!IS_SETLKW(cmd))
+		flags |= LM_FLAG_TRY_1CB;
 
 	mutex_lock(&fp->f_fl_mutex);
 
-	gl = fl_gh->gh_gl;
-	if (gl) {
+	if (gfs2_holder_initialized(fl_gh)) {
+		struct file_lock request;
 		if (fl_gh->gh_state == state)
 			goto out;
-		flock_lock_file_wait(file,
-				     &(struct file_lock){.fl_type = F_UNLCK});
-		gfs2_glock_dq_wait(fl_gh);
+		locks_init_lock(&request);
+		request.c.flc_type = F_UNLCK;
+		request.c.flc_flags = FL_FLOCK;
+		locks_lock_file_wait(file, &request);
+		gfs2_glock_dq(fl_gh);
 		gfs2_holder_reinit(state, flags, fl_gh);
 	} else {
 		error = gfs2_glock_get(GFS2_SB(&ip->i_inode), ip->i_no_addr,
 				       &gfs2_flock_glops, CREATE, &gl);
 		if (error)
 			goto out;
+		spin_lock(&file->f_lock);
 		gfs2_holder_init(gl, state, flags, fl_gh);
+		spin_unlock(&file->f_lock);
 		gfs2_glock_put(gl);
 	}
-	error = gfs2_glock_nq(fl_gh);
+	for (sleeptime = 1; sleeptime <= 4; sleeptime <<= 1) {
+		error = gfs2_glock_nq(fl_gh);
+		if (error != GLR_TRYFAILED)
+			break;
+		fl_gh->gh_flags &= ~LM_FLAG_TRY_1CB;
+		fl_gh->gh_flags |= LM_FLAG_TRY;
+		msleep(sleeptime);
+	}
 	if (error) {
-		gfs2_holder_uninit(fl_gh);
+		__flock_holder_uninit(file, fl_gh);
 		if (error == GLR_TRYFAILED)
 			error = -EAGAIN;
 	} else {
-		error = flock_lock_file_wait(file, fl);
+		error = locks_lock_file_wait(file, fl);
 		gfs2_assert_warn(GFS2_SB(&ip->i_inode), !error);
 	}
 
@@ -1012,10 +1548,10 @@ static void do_unflock(struct file *file, struct file_lock *fl)
 	struct gfs2_holder *fl_gh = &fp->f_fl_gh;
 
 	mutex_lock(&fp->f_fl_mutex);
-	flock_lock_file_wait(file, fl);
-	if (fl_gh->gh_gl) {
-		gfs2_glock_dq_wait(fl_gh);
-		gfs2_holder_uninit(fl_gh);
+	locks_lock_file_wait(file, fl);
+	if (gfs2_holder_initialized(fl_gh)) {
+		gfs2_glock_dq(fl_gh);
+		__flock_holder_uninit(file, fl_gh);
 	}
 	mutex_unlock(&fp->f_fl_mutex);
 }
@@ -1031,12 +1567,10 @@ static void do_unflock(struct file *file, struct file_lock *fl)
 
 static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl)
 {
-	if (!(fl->fl_flags & FL_FLOCK))
+	if (!(fl->c.flc_flags & FL_FLOCK))
 		return -ENOLCK;
-	if (fl->fl_type & LOCK_MAND)
-		return -EOPNOTSUPP;
 
-	if (fl->fl_type == F_UNLCK) {
+	if (lock_is_unlock(fl)) {
 		do_unflock(file, fl);
 		return 0;
 	} else {
@@ -1046,56 +1580,60 @@ static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl)
 
 const struct file_operations gfs2_file_fops = {
 	.llseek		= gfs2_llseek,
-	.read		= do_sync_read,
-	.aio_read	= generic_file_aio_read,
-	.write		= do_sync_write,
-	.aio_write	= gfs2_file_aio_write,
+	.read_iter	= gfs2_file_read_iter,
+	.write_iter	= gfs2_file_write_iter,
+	.iopoll		= iocb_bio_iopoll,
 	.unlocked_ioctl	= gfs2_ioctl,
+	.compat_ioctl	= gfs2_compat_ioctl,
 	.mmap		= gfs2_mmap,
 	.open		= gfs2_open,
 	.release	= gfs2_release,
 	.fsync		= gfs2_fsync,
 	.lock		= gfs2_lock,
 	.flock		= gfs2_flock,
-	.splice_read	= generic_file_splice_read,
-	.splice_write	= generic_file_splice_write,
-	.setlease	= gfs2_setlease,
+	.splice_read	= copy_splice_read,
+	.splice_write	= gfs2_file_splice_write,
+	.setlease	= simple_nosetlease,
 	.fallocate	= gfs2_fallocate,
+	.fop_flags	= FOP_ASYNC_LOCK,
 };
 
 const struct file_operations gfs2_dir_fops = {
-	.iterate	= gfs2_readdir,
+	.iterate_shared	= gfs2_readdir,
 	.unlocked_ioctl	= gfs2_ioctl,
+	.compat_ioctl	= gfs2_compat_ioctl,
 	.open		= gfs2_open,
 	.release	= gfs2_release,
 	.fsync		= gfs2_fsync,
 	.lock		= gfs2_lock,
 	.flock		= gfs2_flock,
 	.llseek		= default_llseek,
+	.fop_flags	= FOP_ASYNC_LOCK,
 };
 
 #endif /* CONFIG_GFS2_FS_LOCKING_DLM */
 
 const struct file_operations gfs2_file_fops_nolock = {
 	.llseek		= gfs2_llseek,
-	.read		= do_sync_read,
-	.aio_read	= generic_file_aio_read,
-	.write		= do_sync_write,
-	.aio_write	= gfs2_file_aio_write,
+	.read_iter	= gfs2_file_read_iter,
+	.write_iter	= gfs2_file_write_iter,
+	.iopoll		= iocb_bio_iopoll,
 	.unlocked_ioctl	= gfs2_ioctl,
+	.compat_ioctl	= gfs2_compat_ioctl,
 	.mmap		= gfs2_mmap,
 	.open		= gfs2_open,
 	.release	= gfs2_release,
 	.fsync		= gfs2_fsync,
-	.splice_read	= generic_file_splice_read,
-	.splice_write	= generic_file_splice_write,
+	.splice_read	= copy_splice_read,
+	.splice_write	= gfs2_file_splice_write,
 	.setlease	= generic_setlease,
 	.fallocate	= gfs2_fallocate,
 };
 
 const struct file_operations gfs2_dir_fops_nolock = {
-	.iterate	= gfs2_readdir,
+	.iterate_shared	= gfs2_readdir,
 	.unlocked_ioctl	= gfs2_ioctl,
+	.compat_ioctl	= gfs2_compat_ioctl,
 	.open		= gfs2_open,
 	.release	= gfs2_release,
 	.fsync		= gfs2_fsync,