diff options
Diffstat (limited to 'fs/btrfs/ioctl.c')
| -rw-r--r-- | fs/btrfs/ioctl.c | 5806 |
1 files changed, 3479 insertions, 2327 deletions
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 238a05545ee2..acb484546b1d 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1,79 +1,143 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/kernel.h> #include <linux/bio.h> -#include <linux/buffer_head.h> #include <linux/file.h> #include <linux/fs.h> #include <linux/fsnotify.h> #include <linux/pagemap.h> #include <linux/highmem.h> #include <linux/time.h> -#include <linux/init.h> #include <linux/string.h> #include <linux/backing-dev.h> #include <linux/mount.h> -#include <linux/mpage.h> #include <linux/namei.h> -#include <linux/swap.h> #include <linux/writeback.h> -#include <linux/statfs.h> #include <linux/compat.h> -#include <linux/bit_spinlock.h> #include <linux/security.h> #include <linux/xattr.h> -#include <linux/vmalloc.h> +#include <linux/mm.h> #include <linux/slab.h> #include <linux/blkdev.h> #include <linux/uuid.h> #include <linux/btrfs.h> -#include "compat.h" +#include <linux/uaccess.h> +#include <linux/iversion.h> +#include <linux/fileattr.h> +#include <linux/fsverity.h> +#include <linux/sched/xacct.h> +#include <linux/io_uring/cmd.h> #include "ctree.h" #include "disk-io.h" +#include "export.h" #include "transaction.h" #include "btrfs_inode.h" -#include "print-tree.h" #include "volumes.h" #include "locking.h" -#include "inode-map.h" #include "backref.h" -#include "rcu-string.h" #include "send.h" #include "dev-replace.h" +#include "props.h" +#include "sysfs.h" +#include "qgroup.h" +#include "tree-log.h" +#include "compression.h" +#include "space-info.h" +#include "block-group.h" +#include "fs.h" +#include "accessors.h" +#include "extent-tree.h" +#include "root-tree.h" +#include "defrag.h" +#include "dir-item.h" +#include "uuid-tree.h" +#include "ioctl.h" +#include "file.h" +#include "scrub.h" +#include "super.h" + +#ifdef CONFIG_64BIT +/* If we have a 32-bit userspace and 64-bit kernel, then the UAPI + * structures are incorrect, as the timespec structure from userspace + * is 4 bytes too small. We define these alternatives here to teach + * the kernel about the 32-bit struct packing. + */ +struct btrfs_ioctl_timespec_32 { + __u64 sec; + __u32 nsec; +} __attribute__ ((__packed__)); + +struct btrfs_ioctl_received_subvol_args_32 { + char uuid[BTRFS_UUID_SIZE]; /* in */ + __u64 stransid; /* in */ + __u64 rtransid; /* out */ + struct btrfs_ioctl_timespec_32 stime; /* in */ + struct btrfs_ioctl_timespec_32 rtime; /* out */ + __u64 flags; /* in */ + __u64 reserved[16]; /* in */ +} __attribute__ ((__packed__)); + +#define BTRFS_IOC_SET_RECEIVED_SUBVOL_32 _IOWR(BTRFS_IOCTL_MAGIC, 37, \ + struct btrfs_ioctl_received_subvol_args_32) +#endif + +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) +struct btrfs_ioctl_send_args_32 { + __s64 send_fd; /* in */ + __u64 clone_sources_count; /* in */ + compat_uptr_t clone_sources; /* in */ + __u64 parent_root; /* in */ + __u64 flags; /* in */ + __u32 version; /* in */ + __u8 reserved[28]; /* in */ +} __attribute__ ((__packed__)); + +#define BTRFS_IOC_SEND_32 _IOW(BTRFS_IOCTL_MAGIC, 38, \ + struct btrfs_ioctl_send_args_32) + +struct btrfs_ioctl_encoded_io_args_32 { + compat_uptr_t iov; + compat_ulong_t iovcnt; + __s64 offset; + __u64 flags; + __u64 len; + __u64 unencoded_len; + __u64 unencoded_offset; + __u32 compression; + __u32 encryption; + __u8 reserved[64]; +}; + +#define BTRFS_IOC_ENCODED_READ_32 _IOR(BTRFS_IOCTL_MAGIC, 64, \ + struct btrfs_ioctl_encoded_io_args_32) +#define BTRFS_IOC_ENCODED_WRITE_32 _IOW(BTRFS_IOCTL_MAGIC, 64, \ + struct btrfs_ioctl_encoded_io_args_32) +#endif /* Mask out flags that are inappropriate for the given type of inode. */ -static inline __u32 btrfs_mask_flags(umode_t mode, __u32 flags) +static unsigned int btrfs_mask_fsflags_for_type(const struct inode *inode, + unsigned int flags) { - if (S_ISDIR(mode)) + if (S_ISDIR(inode->i_mode)) return flags; - else if (S_ISREG(mode)) + else if (S_ISREG(inode->i_mode)) return flags & ~FS_DIRSYNC_FL; else return flags & (FS_NODUMP_FL | FS_NOATIME_FL); } /* - * Export inode flags to the format expected by the FS_IOC_GETFLAGS ioctl. + * Export internal inode flags to the format expected by the FS_IOC_GETFLAGS + * ioctl. */ -static unsigned int btrfs_flags_to_ioctl(unsigned int flags) +static unsigned int btrfs_inode_flags_to_fsflags(const struct btrfs_inode *inode) { unsigned int iflags = 0; + u32 flags = inode->flags; + u32 ro_flags = inode->ro_flags; if (flags & BTRFS_INODE_SYNC) iflags |= FS_SYNC_FL; @@ -89,11 +153,13 @@ static unsigned int btrfs_flags_to_ioctl(unsigned int flags) iflags |= FS_DIRSYNC_FL; if (flags & BTRFS_INODE_NODATACOW) iflags |= FS_NOCOW_FL; + if (ro_flags & BTRFS_INODE_RO_VERITY) + iflags |= FS_VERITY_FL; - if ((flags & BTRFS_INODE_COMPRESS) && !(flags & BTRFS_INODE_NOCOMPRESS)) - iflags |= FS_COMPR_FL; - else if (flags & BTRFS_INODE_NOCOMPRESS) + if (flags & BTRFS_INODE_NOCOMPRESS) iflags |= FS_NOCOMP_FL; + else if (flags & BTRFS_INODE_COMPRESS) + iflags |= FS_COMPR_FL; return iflags; } @@ -101,171 +167,178 @@ static unsigned int btrfs_flags_to_ioctl(unsigned int flags) /* * Update inode->i_flags based on the btrfs internal flags. */ -void btrfs_update_iflags(struct inode *inode) +void btrfs_sync_inode_flags_to_i_flags(struct btrfs_inode *inode) { - struct btrfs_inode *ip = BTRFS_I(inode); - - inode->i_flags &= ~(S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC); - - if (ip->flags & BTRFS_INODE_SYNC) - inode->i_flags |= S_SYNC; - if (ip->flags & BTRFS_INODE_IMMUTABLE) - inode->i_flags |= S_IMMUTABLE; - if (ip->flags & BTRFS_INODE_APPEND) - inode->i_flags |= S_APPEND; - if (ip->flags & BTRFS_INODE_NOATIME) - inode->i_flags |= S_NOATIME; - if (ip->flags & BTRFS_INODE_DIRSYNC) - inode->i_flags |= S_DIRSYNC; + unsigned int new_fl = 0; + + if (inode->flags & BTRFS_INODE_SYNC) + new_fl |= S_SYNC; + if (inode->flags & BTRFS_INODE_IMMUTABLE) + new_fl |= S_IMMUTABLE; + if (inode->flags & BTRFS_INODE_APPEND) + new_fl |= S_APPEND; + if (inode->flags & BTRFS_INODE_NOATIME) + new_fl |= S_NOATIME; + if (inode->flags & BTRFS_INODE_DIRSYNC) + new_fl |= S_DIRSYNC; + if (inode->ro_flags & BTRFS_INODE_RO_VERITY) + new_fl |= S_VERITY; + + set_mask_bits(&inode->vfs_inode.i_flags, + S_SYNC | S_APPEND | S_IMMUTABLE | S_NOATIME | S_DIRSYNC | + S_VERITY, new_fl); } /* - * Inherit flags from the parent inode. - * - * Currently only the compression flags and the cow flags are inherited. + * Check if @flags are a supported and valid set of FS_*_FL flags and that + * the old and new flags are not conflicting */ -void btrfs_inherit_iflags(struct inode *inode, struct inode *dir) +static int check_fsflags(unsigned int old_flags, unsigned int flags) { - unsigned int flags; - - if (!dir) - return; + if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \ + FS_NOATIME_FL | FS_NODUMP_FL | \ + FS_SYNC_FL | FS_DIRSYNC_FL | \ + FS_NOCOMP_FL | FS_COMPR_FL | + FS_NOCOW_FL)) + return -EOPNOTSUPP; - flags = BTRFS_I(dir)->flags; + /* COMPR and NOCOMP on new/old are valid */ + if ((flags & FS_NOCOMP_FL) && (flags & FS_COMPR_FL)) + return -EINVAL; - if (flags & BTRFS_INODE_NOCOMPRESS) { - BTRFS_I(inode)->flags &= ~BTRFS_INODE_COMPRESS; - BTRFS_I(inode)->flags |= BTRFS_INODE_NOCOMPRESS; - } else if (flags & BTRFS_INODE_COMPRESS) { - BTRFS_I(inode)->flags &= ~BTRFS_INODE_NOCOMPRESS; - BTRFS_I(inode)->flags |= BTRFS_INODE_COMPRESS; - } + if ((flags & FS_COMPR_FL) && (flags & FS_NOCOW_FL)) + return -EINVAL; - if (flags & BTRFS_INODE_NODATACOW) { - BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW; - if (S_ISREG(inode->i_mode)) - BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM; - } + /* NOCOW and compression options are mutually exclusive */ + if ((old_flags & FS_NOCOW_FL) && (flags & (FS_COMPR_FL | FS_NOCOMP_FL))) + return -EINVAL; + if ((flags & FS_NOCOW_FL) && (old_flags & (FS_COMPR_FL | FS_NOCOMP_FL))) + return -EINVAL; - btrfs_update_iflags(inode); + return 0; } -static int btrfs_ioctl_getflags(struct file *file, void __user *arg) +static int check_fsflags_compatible(const struct btrfs_fs_info *fs_info, + unsigned int flags) { - struct btrfs_inode *ip = BTRFS_I(file_inode(file)); - unsigned int flags = btrfs_flags_to_ioctl(ip->flags); + if (btrfs_is_zoned(fs_info) && (flags & FS_NOCOW_FL)) + return -EPERM; - if (copy_to_user(arg, &flags, sizeof(flags))) - return -EFAULT; return 0; } -static int check_flags(unsigned int flags) +int btrfs_check_ioctl_vol_args_path(const struct btrfs_ioctl_vol_args *vol_args) { - if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \ - FS_NOATIME_FL | FS_NODUMP_FL | \ - FS_SYNC_FL | FS_DIRSYNC_FL | \ - FS_NOCOMP_FL | FS_COMPR_FL | - FS_NOCOW_FL)) - return -EOPNOTSUPP; + if (memchr(vol_args->name, 0, sizeof(vol_args->name)) == NULL) + return -ENAMETOOLONG; + return 0; +} - if ((flags & FS_NOCOMP_FL) && (flags & FS_COMPR_FL)) - return -EINVAL; +static int btrfs_check_ioctl_vol_args2_subvol_name(const struct btrfs_ioctl_vol_args_v2 *vol_args2) +{ + if (memchr(vol_args2->name, 0, sizeof(vol_args2->name)) == NULL) + return -ENAMETOOLONG; + return 0; +} +/* + * Set flags/xflags from the internal inode flags. The remaining items of + * fsxattr are zeroed. + */ +int btrfs_fileattr_get(struct dentry *dentry, struct file_kattr *fa) +{ + const struct btrfs_inode *inode = BTRFS_I(d_inode(dentry)); + + fileattr_fill_flags(fa, btrfs_inode_flags_to_fsflags(inode)); return 0; } -static int btrfs_ioctl_setflags(struct file *file, void __user *arg) +int btrfs_fileattr_set(struct mnt_idmap *idmap, + struct dentry *dentry, struct file_kattr *fa) { - struct inode *inode = file_inode(file); - struct btrfs_inode *ip = BTRFS_I(inode); - struct btrfs_root *root = ip->root; + struct btrfs_inode *inode = BTRFS_I(d_inode(dentry)); + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_trans_handle *trans; - unsigned int flags, oldflags; + unsigned int fsflags, old_fsflags; int ret; - u64 ip_oldflags; - unsigned int i_oldflags; - umode_t mode; + const char *comp = NULL; + u32 inode_flags; if (btrfs_root_readonly(root)) return -EROFS; - if (copy_from_user(&flags, arg, sizeof(flags))) - return -EFAULT; + if (fileattr_has_fsx(fa)) + return -EOPNOTSUPP; - ret = check_flags(flags); + fsflags = btrfs_mask_fsflags_for_type(&inode->vfs_inode, fa->flags); + old_fsflags = btrfs_inode_flags_to_fsflags(inode); + ret = check_fsflags(old_fsflags, fsflags); if (ret) return ret; - if (!inode_owner_or_capable(inode)) - return -EACCES; - - ret = mnt_want_write_file(file); + ret = check_fsflags_compatible(fs_info, fsflags); if (ret) return ret; - mutex_lock(&inode->i_mutex); - - ip_oldflags = ip->flags; - i_oldflags = inode->i_flags; - mode = inode->i_mode; - - flags = btrfs_mask_flags(inode->i_mode, flags); - oldflags = btrfs_flags_to_ioctl(ip->flags); - if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) { - if (!capable(CAP_LINUX_IMMUTABLE)) { - ret = -EPERM; - goto out_unlock; - } - } - - if (flags & FS_SYNC_FL) - ip->flags |= BTRFS_INODE_SYNC; + inode_flags = inode->flags; + if (fsflags & FS_SYNC_FL) + inode_flags |= BTRFS_INODE_SYNC; else - ip->flags &= ~BTRFS_INODE_SYNC; - if (flags & FS_IMMUTABLE_FL) - ip->flags |= BTRFS_INODE_IMMUTABLE; + inode_flags &= ~BTRFS_INODE_SYNC; + if (fsflags & FS_IMMUTABLE_FL) + inode_flags |= BTRFS_INODE_IMMUTABLE; else - ip->flags &= ~BTRFS_INODE_IMMUTABLE; - if (flags & FS_APPEND_FL) - ip->flags |= BTRFS_INODE_APPEND; + inode_flags &= ~BTRFS_INODE_IMMUTABLE; + if (fsflags & FS_APPEND_FL) + inode_flags |= BTRFS_INODE_APPEND; else - ip->flags &= ~BTRFS_INODE_APPEND; - if (flags & FS_NODUMP_FL) - ip->flags |= BTRFS_INODE_NODUMP; + inode_flags &= ~BTRFS_INODE_APPEND; + if (fsflags & FS_NODUMP_FL) + inode_flags |= BTRFS_INODE_NODUMP; else - ip->flags &= ~BTRFS_INODE_NODUMP; - if (flags & FS_NOATIME_FL) - ip->flags |= BTRFS_INODE_NOATIME; + inode_flags &= ~BTRFS_INODE_NODUMP; + if (fsflags & FS_NOATIME_FL) + inode_flags |= BTRFS_INODE_NOATIME; else - ip->flags &= ~BTRFS_INODE_NOATIME; - if (flags & FS_DIRSYNC_FL) - ip->flags |= BTRFS_INODE_DIRSYNC; + inode_flags &= ~BTRFS_INODE_NOATIME; + + /* If coming from FS_IOC_FSSETXATTR then skip unconverted flags */ + if (!fa->flags_valid) { + /* 1 item for the inode */ + trans = btrfs_start_transaction(root, 1); + if (IS_ERR(trans)) + return PTR_ERR(trans); + goto update_flags; + } + + if (fsflags & FS_DIRSYNC_FL) + inode_flags |= BTRFS_INODE_DIRSYNC; else - ip->flags &= ~BTRFS_INODE_DIRSYNC; - if (flags & FS_NOCOW_FL) { - if (S_ISREG(mode)) { + inode_flags &= ~BTRFS_INODE_DIRSYNC; + if (fsflags & FS_NOCOW_FL) { + if (S_ISREG(inode->vfs_inode.i_mode)) { /* * It's safe to turn csums off here, no extents exist. * Otherwise we want the flag to reflect the real COW * status of the file and will not set it. */ - if (inode->i_size == 0) - ip->flags |= BTRFS_INODE_NODATACOW - | BTRFS_INODE_NODATASUM; + if (inode->vfs_inode.i_size == 0) + inode_flags |= BTRFS_INODE_NODATACOW | + BTRFS_INODE_NODATASUM; } else { - ip->flags |= BTRFS_INODE_NODATACOW; + inode_flags |= BTRFS_INODE_NODATACOW; } } else { /* - * Revert back under same assuptions as above + * Revert back under same assumptions as above */ - if (S_ISREG(mode)) { - if (inode->i_size == 0) - ip->flags &= ~(BTRFS_INODE_NODATACOW - | BTRFS_INODE_NODATASUM); + if (S_ISREG(inode->vfs_inode.i_mode)) { + if (inode->vfs_inode.i_size == 0) + inode_flags &= ~(BTRFS_INODE_NODATACOW | + BTRFS_INODE_NODATASUM); } else { - ip->flags &= ~BTRFS_INODE_NODATACOW; + inode_flags &= ~BTRFS_INODE_NODATACOW; } } @@ -274,72 +347,103 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) * flag may be changed automatically if compression code won't make * things smaller. */ - if (flags & FS_NOCOMP_FL) { - ip->flags &= ~BTRFS_INODE_COMPRESS; - ip->flags |= BTRFS_INODE_NOCOMPRESS; - } else if (flags & FS_COMPR_FL) { - ip->flags |= BTRFS_INODE_COMPRESS; - ip->flags &= ~BTRFS_INODE_NOCOMPRESS; - } else { - ip->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS); - } + if (fsflags & FS_NOCOMP_FL) { + inode_flags &= ~BTRFS_INODE_COMPRESS; + inode_flags |= BTRFS_INODE_NOCOMPRESS; + } else if (fsflags & FS_COMPR_FL) { - trans = btrfs_start_transaction(root, 1); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - goto out_drop; + if (IS_SWAPFILE(&inode->vfs_inode)) + return -ETXTBSY; + + inode_flags |= BTRFS_INODE_COMPRESS; + inode_flags &= ~BTRFS_INODE_NOCOMPRESS; + + comp = btrfs_compress_type2str(fs_info->compress_type); + if (!comp || comp[0] == 0) + comp = btrfs_compress_type2str(BTRFS_COMPRESS_ZLIB); + } else { + inode_flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS); } - btrfs_update_iflags(inode); - inode_inc_iversion(inode); - inode->i_ctime = CURRENT_TIME; - ret = btrfs_update_inode(trans, root, inode); + /* + * 1 for inode item + * 2 for properties + */ + trans = btrfs_start_transaction(root, 3); + if (IS_ERR(trans)) + return PTR_ERR(trans); - btrfs_end_transaction(trans, root); - out_drop: - if (ret) { - ip->flags = ip_oldflags; - inode->i_flags = i_oldflags; + if (comp) { + ret = btrfs_set_prop(trans, inode, "btrfs.compression", + comp, strlen(comp), 0); + if (unlikely(ret)) { + btrfs_abort_transaction(trans, ret); + goto out_end_trans; + } + } else { + ret = btrfs_set_prop(trans, inode, "btrfs.compression", NULL, 0, 0); + if (unlikely(ret && ret != -ENODATA)) { + btrfs_abort_transaction(trans, ret); + goto out_end_trans; + } } - out_unlock: - mutex_unlock(&inode->i_mutex); - mnt_drop_write_file(file); +update_flags: + inode->flags = inode_flags; + btrfs_update_inode_mapping_flags(inode); + btrfs_sync_inode_flags_to_i_flags(inode); + inode_inc_iversion(&inode->vfs_inode); + inode_set_ctime_current(&inode->vfs_inode); + ret = btrfs_update_inode(trans, inode); + + out_end_trans: + btrfs_end_transaction(trans); return ret; } -static int btrfs_ioctl_getversion(struct file *file, int __user *arg) +static int btrfs_ioctl_getversion(const struct inode *inode, int __user *arg) { - struct inode *inode = file_inode(file); - 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 btrfs_fs_info *fs_info = btrfs_sb(fdentry(file)->d_sb); struct btrfs_device *device; - struct request_queue *q; struct fstrim_range range; u64 minlen = ULLONG_MAX; u64 num_devices = 0; - u64 total_bytes = btrfs_super_total_bytes(fs_info->super_copy); int ret; if (!capable(CAP_SYS_ADMIN)) return -EPERM; + /* + * btrfs_trim_block_group() depends on space cache, which is not + * available in zoned filesystem. So, disallow fitrim on a zoned + * filesystem for now. + */ + if (btrfs_is_zoned(fs_info)) + return -EOPNOTSUPP; + + /* + * If the fs is mounted with nologreplay, which requires it to be + * mounted in RO mode as well, we can not allow discard on free space + * inside block groups, because log trees refer to extents that are not + * pinned in a block group's free space cache (pinning the extents is + * precisely the first phase of replaying a log tree). + */ + if (btrfs_test_opt(fs_info, NOLOGREPLAY)) + return -EROFS; + rcu_read_lock(); list_for_each_entry_rcu(device, &fs_info->fs_devices->devices, dev_list) { - if (!device->bdev) + if (!device->bdev || !bdev_max_discard_sectors(device->bdev)) continue; - q = bdev_get_queue(device->bdev); - if (blk_queue_discard(q)) { - num_devices++; - minlen = min((u64)q->limits.discard_granularity, - minlen); - } + num_devices++; + minlen = min_t(u64, bdev_discard_granularity(device->bdev), + minlen); } rcu_read_unlock(); @@ -347,243 +451,314 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg) return -EOPNOTSUPP; if (copy_from_user(&range, arg, sizeof(range))) return -EFAULT; - if (range.start > total_bytes || - range.len < fs_info->sb->s_blocksize) + + /* + * NOTE: Don't truncate the range using super->total_bytes. Bytenr of + * block group is in the logical address space, which can be any + * sectorsize aligned bytenr in the range [0, U64_MAX]. + */ + if (range.len < fs_info->sectorsize) return -EINVAL; - range.len = min(range.len, total_bytes - range.start); range.minlen = max(range.minlen, minlen); - ret = btrfs_trim_fs(fs_info->tree_root, &range); - if (ret < 0) - return ret; + ret = btrfs_trim_fs(fs_info, &range); if (copy_to_user(arg, &range, sizeof(range))) return -EFAULT; - return 0; + return ret; +} + +/* + * Calculate the number of transaction items to reserve for creating a subvolume + * or snapshot, not including the inode, directory entries, or parent directory. + */ +static unsigned int create_subvol_num_items(const struct btrfs_qgroup_inherit *inherit) +{ + /* + * 1 to add root block + * 1 to add root item + * 1 to add root ref + * 1 to add root backref + * 1 to add UUID item + * 1 to add qgroup info + * 1 to add qgroup limit + * + * Ideally the last two would only be accounted if qgroups are enabled, + * but that can change between now and the time we would insert them. + */ + unsigned int num_items = 7; + + if (inherit) { + /* 2 to add qgroup relations for each inherited qgroup */ + num_items += 2 * inherit->num_qgroups; + } + return num_items; } -static noinline int create_subvol(struct inode *dir, - struct dentry *dentry, - char *name, int namelen, - u64 *async_transid, +static noinline int create_subvol(struct mnt_idmap *idmap, + struct inode *dir, struct dentry *dentry, struct btrfs_qgroup_inherit *inherit) { + struct btrfs_fs_info *fs_info = inode_to_fs_info(dir); struct btrfs_trans_handle *trans; struct btrfs_key key; - struct btrfs_root_item root_item; + struct btrfs_root_item AUTO_KFREE(root_item); struct btrfs_inode_item *inode_item; struct extent_buffer *leaf; struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_root *new_root; struct btrfs_block_rsv block_rsv; - struct timespec cur_time = CURRENT_TIME; + struct timespec64 cur_time = current_time(dir); + struct btrfs_new_inode_args new_inode_args = { + .dir = dir, + .dentry = dentry, + .subvol = true, + }; + unsigned int trans_num_items; int ret; - int err; + dev_t anon_dev; u64 objectid; - u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID; - u64 index = 0; - u64 qgroup_reserved; - uuid_le new_uuid; + u64 qgroup_reserved = 0; + + root_item = kzalloc(sizeof(*root_item), GFP_KERNEL); + if (!root_item) + return -ENOMEM; - ret = btrfs_find_free_objectid(root->fs_info->tree_root, &objectid); + ret = btrfs_get_free_objectid(fs_info->tree_root, &objectid); if (ret) return ret; - btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP); /* - * The same as the snapshot creation, please see the comment - * of create_snapshot(). + * Don't create subvolume whose level is not zero. Or qgroup will be + * screwed up since it assumes subvolume qgroup's level to be 0. */ + if (btrfs_qgroup_level(objectid)) + return -ENOSPC; + + ret = get_anon_bdev(&anon_dev); + if (ret < 0) + return ret; + + new_inode_args.inode = btrfs_new_subvol_inode(idmap, dir); + if (!new_inode_args.inode) { + ret = -ENOMEM; + goto out_anon_dev; + } + ret = btrfs_new_inode_prepare(&new_inode_args, &trans_num_items); + if (ret) + goto out_inode; + trans_num_items += create_subvol_num_items(inherit); + + btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP); ret = btrfs_subvolume_reserve_metadata(root, &block_rsv, - 7, &qgroup_reserved); + trans_num_items, false); if (ret) - return ret; + goto out_new_inode_args; + qgroup_reserved = block_rsv.qgroup_rsv_reserved; trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { ret = PTR_ERR(trans); - goto out; + goto out_release_rsv; } + btrfs_qgroup_convert_reserved_meta(root, qgroup_reserved); + qgroup_reserved = 0; trans->block_rsv = &block_rsv; trans->bytes_reserved = block_rsv.size; - ret = btrfs_qgroup_inherit(trans, root->fs_info, 0, objectid, inherit); + ret = btrfs_qgroup_inherit(trans, 0, objectid, btrfs_root_id(root), inherit); if (ret) - goto fail; + goto out; - leaf = btrfs_alloc_free_block(trans, root, root->leafsize, - 0, objectid, NULL, 0, 0, 0); + leaf = btrfs_alloc_tree_block(trans, root, 0, objectid, NULL, 0, 0, 0, + 0, BTRFS_NESTING_NORMAL); if (IS_ERR(leaf)) { ret = PTR_ERR(leaf); - goto fail; + goto out; } - memset_extent_buffer(leaf, 0, 0, sizeof(struct btrfs_header)); - btrfs_set_header_bytenr(leaf, leaf->start); - btrfs_set_header_generation(leaf, trans->transid); - btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV); - btrfs_set_header_owner(leaf, objectid); - - write_extent_buffer(leaf, root->fs_info->fsid, - (unsigned long)btrfs_header_fsid(leaf), - BTRFS_FSID_SIZE); - write_extent_buffer(leaf, root->fs_info->chunk_tree_uuid, - (unsigned long)btrfs_header_chunk_tree_uuid(leaf), - BTRFS_UUID_SIZE); - btrfs_mark_buffer_dirty(leaf); - - memset(&root_item, 0, sizeof(root_item)); - - inode_item = &root_item.inode; - inode_item->generation = cpu_to_le64(1); - inode_item->size = cpu_to_le64(3); - inode_item->nlink = cpu_to_le32(1); - inode_item->nbytes = cpu_to_le64(root->leafsize); - inode_item->mode = cpu_to_le32(S_IFDIR | 0755); - - root_item.flags = 0; - root_item.byte_limit = 0; - inode_item->flags = cpu_to_le64(BTRFS_INODE_ROOT_ITEM_INIT); - - btrfs_set_root_bytenr(&root_item, leaf->start); - btrfs_set_root_generation(&root_item, trans->transid); - btrfs_set_root_level(&root_item, 0); - btrfs_set_root_refs(&root_item, 1); - btrfs_set_root_used(&root_item, leaf->len); - btrfs_set_root_last_snapshot(&root_item, 0); - - 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); - root_item.otime.sec = cpu_to_le64(cur_time.tv_sec); - root_item.otime.nsec = cpu_to_le32(cur_time.tv_nsec); - root_item.ctime = root_item.otime; - btrfs_set_root_ctransid(&root_item, trans->transid); - btrfs_set_root_otransid(&root_item, trans->transid); + btrfs_mark_buffer_dirty(trans, leaf); + + inode_item = &root_item->inode; + btrfs_set_stack_inode_generation(inode_item, 1); + btrfs_set_stack_inode_size(inode_item, 3); + btrfs_set_stack_inode_nlink(inode_item, 1); + btrfs_set_stack_inode_nbytes(inode_item, + fs_info->nodesize); + btrfs_set_stack_inode_mode(inode_item, S_IFDIR | 0755); + + btrfs_set_root_flags(root_item, 0); + btrfs_set_root_limit(root_item, 0); + btrfs_set_stack_inode_flags(inode_item, BTRFS_INODE_ROOT_ITEM_INIT); + + btrfs_set_root_bytenr(root_item, leaf->start); + btrfs_set_root_generation(root_item, trans->transid); + btrfs_set_root_level(root_item, 0); + btrfs_set_root_refs(root_item, 1); + btrfs_set_root_used(root_item, leaf->len); + btrfs_set_root_last_snapshot(root_item, 0); + + btrfs_set_root_generation_v2(root_item, + btrfs_root_generation(root_item)); + 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; + btrfs_set_root_ctransid(root_item, trans->transid); + btrfs_set_root_otransid(root_item, trans->transid); btrfs_tree_unlock(leaf); - free_extent_buffer(leaf); - leaf = NULL; - btrfs_set_root_dirid(&root_item, new_dirid); + btrfs_set_root_dirid(root_item, BTRFS_FIRST_FREE_OBJECTID); key.objectid = objectid; + key.type = BTRFS_ROOT_ITEM_KEY; key.offset = 0; - btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY); - ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key, - &root_item); - if (ret) - goto fail; + ret = btrfs_insert_root(trans, fs_info->tree_root, &key, + root_item); + if (ret) { + int ret2; - key.offset = (u64)-1; - new_root = btrfs_read_fs_root_no_name(root->fs_info, &key); - if (IS_ERR(new_root)) { - btrfs_abort_transaction(trans, root, PTR_ERR(new_root)); - ret = PTR_ERR(new_root); - goto fail; + /* + * Since we don't abort the transaction in this case, free the + * tree block so that we don't leak space and leave the + * filesystem in an inconsistent state (an extent item in the + * extent tree with a backreference for a root that does not + * exists). + */ + btrfs_tree_lock(leaf); + btrfs_clear_buffer_dirty(trans, leaf); + btrfs_tree_unlock(leaf); + ret2 = btrfs_free_tree_block(trans, objectid, leaf, 0, 1); + if (unlikely(ret2 < 0)) + btrfs_abort_transaction(trans, ret2); + free_extent_buffer(leaf); + goto out; } - btrfs_record_root_in_trans(trans, new_root); + free_extent_buffer(leaf); + leaf = NULL; - ret = btrfs_create_subvol_root(trans, new_root, new_dirid); - if (ret) { - /* We potentially lose an unused inode item here */ - btrfs_abort_transaction(trans, root, ret); - goto fail; + new_root = btrfs_get_new_fs_root(fs_info, objectid, &anon_dev); + if (IS_ERR(new_root)) { + ret = PTR_ERR(new_root); + btrfs_abort_transaction(trans, ret); + goto out; } + /* anon_dev is owned by new_root now. */ + anon_dev = 0; + BTRFS_I(new_inode_args.inode)->root = new_root; + /* ... and new_root is owned by new_inode_args.inode now. */ - /* - * insert the directory item - */ - ret = btrfs_set_inode_index(dir, &index); - if (ret) { - btrfs_abort_transaction(trans, root, ret); - goto fail; + ret = btrfs_record_root_in_trans(trans, new_root); + if (unlikely(ret)) { + btrfs_abort_transaction(trans, ret); + goto out; } - ret = btrfs_insert_dir_item(trans, root, - name, namelen, dir, &key, - BTRFS_FT_DIR, index); - if (ret) { - btrfs_abort_transaction(trans, root, ret); - goto fail; + ret = btrfs_uuid_tree_add(trans, root_item->uuid, + BTRFS_UUID_KEY_SUBVOL, objectid); + if (unlikely(ret)) { + btrfs_abort_transaction(trans, ret); + goto out; } - btrfs_i_size_write(dir, dir->i_size + namelen * 2); - ret = btrfs_update_inode(trans, root, dir); - BUG_ON(ret); + btrfs_record_new_subvolume(trans, BTRFS_I(dir)); - ret = btrfs_add_root_ref(trans, root->fs_info->tree_root, - objectid, root->root_key.objectid, - btrfs_ino(dir), index, name, namelen); + ret = btrfs_create_new_inode(trans, &new_inode_args); + if (unlikely(ret)) { + btrfs_abort_transaction(trans, ret); + goto out; + } - BUG_ON(ret); + d_instantiate_new(dentry, new_inode_args.inode); + new_inode_args.inode = NULL; -fail: +out: trans->block_rsv = NULL; trans->bytes_reserved = 0; - if (async_transid) { - *async_transid = trans->transid; - err = btrfs_commit_transaction_async(trans, root, 1); - if (err) - err = btrfs_commit_transaction(trans, root); - } else { - err = btrfs_commit_transaction(trans, root); - } - if (err && !ret) - ret = err; + btrfs_end_transaction(trans); +out_release_rsv: + btrfs_block_rsv_release(fs_info, &block_rsv, (u64)-1, NULL); + if (qgroup_reserved) + btrfs_qgroup_free_meta_prealloc(root, qgroup_reserved); +out_new_inode_args: + btrfs_new_inode_args_destroy(&new_inode_args); +out_inode: + iput(new_inode_args.inode); +out_anon_dev: + if (anon_dev) + free_anon_bdev(anon_dev); - if (!ret) - d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry)); -out: - btrfs_subvolume_release_metadata(root, &block_rsv, qgroup_reserved); return ret; } static int create_snapshot(struct btrfs_root *root, struct inode *dir, - struct dentry *dentry, char *name, int namelen, - u64 *async_transid, bool readonly, + struct dentry *dentry, bool readonly, struct btrfs_qgroup_inherit *inherit) { + struct btrfs_fs_info *fs_info = inode_to_fs_info(dir); struct inode *inode; struct btrfs_pending_snapshot *pending_snapshot; + unsigned int trans_num_items; struct btrfs_trans_handle *trans; + struct btrfs_block_rsv *block_rsv; + u64 qgroup_reserved = 0; int ret; - if (!root->ref_cows) - return -EINVAL; + /* We do not support snapshotting right now. */ + if (btrfs_fs_incompat(fs_info, EXTENT_TREE_V2)) { + btrfs_warn(fs_info, + "extent tree v2 doesn't support snapshotting yet"); + return -EOPNOTSUPP; + } - ret = btrfs_start_delalloc_inodes(root, 0); - if (ret) - return ret; + if (btrfs_root_refs(&root->root_item) == 0) + return -ENOENT; - btrfs_wait_ordered_extents(root, 0); + if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) + return -EINVAL; - pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS); + if (atomic_read(&root->nr_swapfiles)) { + btrfs_warn(fs_info, + "cannot snapshot subvolume with active swapfile"); + return -ETXTBSY; + } + + pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_KERNEL); if (!pending_snapshot) return -ENOMEM; - btrfs_init_block_rsv(&pending_snapshot->block_rsv, - BTRFS_BLOCK_RSV_TEMP); + ret = get_anon_bdev(&pending_snapshot->anon_dev); + if (ret < 0) + goto free_pending; + pending_snapshot->root_item = kzalloc(sizeof(struct btrfs_root_item), + GFP_KERNEL); + pending_snapshot->path = btrfs_alloc_path(); + if (!pending_snapshot->root_item || !pending_snapshot->path) { + ret = -ENOMEM; + goto free_pending; + } + + block_rsv = &pending_snapshot->block_rsv; + btrfs_init_block_rsv(block_rsv, BTRFS_BLOCK_RSV_TEMP); /* - * 1 - parent dir inode - * 2 - dir entries - * 1 - root item - * 2 - root ref/backref - * 1 - root of snapshot + * 1 to add dir item + * 1 to add dir index + * 1 to update parent inode item */ - ret = btrfs_subvolume_reserve_metadata(BTRFS_I(dir)->root, - &pending_snapshot->block_rsv, 7, - &pending_snapshot->qgroup_reserved); + trans_num_items = create_subvol_num_items(inherit) + 3; + ret = btrfs_subvolume_reserve_metadata(BTRFS_I(dir)->root, block_rsv, + trans_num_items, false); if (ret) - goto out; + goto free_pending; + qgroup_reserved = block_rsv->qgroup_rsv_reserved; pending_snapshot->dentry = dentry; pending_snapshot->root = root; pending_snapshot->readonly = readonly; - pending_snapshot->dir = dir; + pending_snapshot->dir = BTRFS_I(dir); pending_snapshot->inherit = inherit; trans = btrfs_start_transaction(root, 0); @@ -591,21 +766,17 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, ret = PTR_ERR(trans); goto fail; } - - spin_lock(&root->fs_info->trans_lock); - list_add(&pending_snapshot->list, - &trans->transaction->pending_snapshots); - spin_unlock(&root->fs_info->trans_lock); - if (async_transid) { - *async_transid = trans->transid; - ret = btrfs_commit_transaction_async(trans, - root->fs_info->extent_root, 1); - if (ret) - ret = btrfs_commit_transaction(trans, root); - } else { - ret = btrfs_commit_transaction(trans, - root->fs_info->extent_root); + ret = btrfs_record_root_in_trans(trans, BTRFS_I(dir)->root); + if (ret) { + btrfs_end_transaction(trans); + goto fail; } + btrfs_qgroup_convert_reserved_meta(root, qgroup_reserved); + qgroup_reserved = 0; + + trans->pending_snapshot = pending_snapshot; + + ret = btrfs_commit_transaction(trans); if (ret) goto fail; @@ -617,38 +788,31 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, if (ret) goto fail; - inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry); + inode = btrfs_lookup_dentry(d_inode(dentry->d_parent), dentry); if (IS_ERR(inode)) { ret = PTR_ERR(inode); goto fail; } - BUG_ON(!inode); + d_instantiate(dentry, inode); ret = 0; + pending_snapshot->anon_dev = 0; fail: - btrfs_subvolume_release_metadata(BTRFS_I(dir)->root, - &pending_snapshot->block_rsv, - pending_snapshot->qgroup_reserved); -out: + /* Prevent double freeing of anon_dev */ + if (ret && pending_snapshot->snap) + pending_snapshot->snap->anon_dev = 0; + btrfs_put_root(pending_snapshot->snap); + btrfs_block_rsv_release(fs_info, block_rsv, (u64)-1, NULL); + if (qgroup_reserved) + btrfs_qgroup_free_meta_prealloc(root, qgroup_reserved); +free_pending: + if (pending_snapshot->anon_dev) + free_anon_bdev(pending_snapshot->anon_dev); + kfree(pending_snapshot->root_item); + btrfs_free_path(pending_snapshot->path); kfree(pending_snapshot); - return ret; -} -/* copy of check_sticky in fs/namei.c() -* It's inline, so penalty for filesystems that don't use sticky bit is -* minimal. -*/ -static inline int btrfs_check_sticky(struct inode *dir, struct inode *inode) -{ - kuid_t fsuid = current_fsuid(); - - if (!(dir->i_mode & S_ISVTX)) - return 0; - if (uid_eq(inode->i_uid, fsuid)) - return 0; - if (uid_eq(dir->i_uid, fsuid)) - return 0; - return !capable(CAP_FOWNER); + return ret; } /* copy of may_delete in fs/namei.c() @@ -662,7 +826,7 @@ static inline int btrfs_check_sticky(struct inode *dir, struct inode *inode) * a. be owner of dir, or * b. be owner of victim, or * c. have CAP_FOWNER capability - * 6. If the victim is append-only or immutable we can't do antyhing with + * 6. If the victim is append-only or immutable we can't do anything with * links pointing to it. * 7. If we were asked to remove a directory and victim isn't one - ENOTDIR. * 8. If we were asked to remove a non-directory and victim isn't one - EISDIR. @@ -671,31 +835,34 @@ static inline int btrfs_check_sticky(struct inode *dir, struct inode *inode) * nfs_async_unlink(). */ -static int btrfs_may_delete(struct inode *dir,struct dentry *victim,int isdir) +static int btrfs_may_delete(struct mnt_idmap *idmap, + struct inode *dir, struct dentry *victim, int isdir) { - int error; + int ret; - if (!victim->d_inode) + if (d_really_is_negative(victim)) return -ENOENT; - BUG_ON(victim->d_parent->d_inode != dir); + /* The @victim is not inside @dir. */ + if (d_inode(victim->d_parent) != dir) + return -EINVAL; audit_inode_child(dir, victim, AUDIT_TYPE_CHILD_DELETE); - error = inode_permission(dir, MAY_WRITE | MAY_EXEC); - if (error) - return error; + ret = inode_permission(idmap, dir, MAY_WRITE | MAY_EXEC); + if (ret) + return ret; if (IS_APPEND(dir)) return -EPERM; - if (btrfs_check_sticky(dir, victim->d_inode)|| - IS_APPEND(victim->d_inode)|| - IS_IMMUTABLE(victim->d_inode) || IS_SWAPFILE(victim->d_inode)) + if (check_sticky(idmap, dir, d_inode(victim)) || + IS_APPEND(d_inode(victim)) || IS_IMMUTABLE(d_inode(victim)) || + IS_SWAPFILE(d_inode(victim))) return -EPERM; if (isdir) { - if (!S_ISDIR(victim->d_inode->i_mode)) + if (!d_is_dir(victim)) return -ENOTDIR; if (IS_ROOT(victim)) return -EBUSY; - } else if (S_ISDIR(victim->d_inode->i_mode)) + } else if (d_is_dir(victim)) return -EISDIR; if (IS_DEADDIR(dir)) return -ENOENT; @@ -705,13 +872,16 @@ static int btrfs_may_delete(struct inode *dir,struct dentry *victim,int isdir) } /* copy of may_create in fs/namei.c() */ -static inline int btrfs_may_create(struct inode *dir, struct dentry *child) +static inline int btrfs_may_create(struct mnt_idmap *idmap, + struct inode *dir, const struct dentry *child) { - if (child->d_inode) + if (d_really_is_positive(child)) return -EEXIST; if (IS_DEADDIR(dir)) return -ENOENT; - return inode_permission(dir, MAY_WRITE | MAY_EXEC); + if (!fsuidgid_has_mapping(dir->i_sb, idmap)) + return -EOVERFLOW; + return inode_permission(idmap, dir, MAY_WRITE | MAY_EXEC); } /* @@ -719,654 +889,148 @@ static inline int btrfs_may_create(struct inode *dir, struct dentry *child) * sys_mkdirat and vfs_mkdir, but we only do a single component lookup * inside this filesystem so it's quite a bit simpler. */ -static noinline int btrfs_mksubvol(struct path *parent, - char *name, int namelen, - struct btrfs_root *snap_src, - u64 *async_transid, bool readonly, +static noinline int btrfs_mksubvol(struct dentry *parent, + struct mnt_idmap *idmap, + struct qstr *qname, struct btrfs_root *snap_src, + bool readonly, struct btrfs_qgroup_inherit *inherit) { - struct inode *dir = parent->dentry->d_inode; + struct inode *dir = d_inode(parent); + struct btrfs_fs_info *fs_info = inode_to_fs_info(dir); struct dentry *dentry; - int error; - - error = mutex_lock_killable_nested(&dir->i_mutex, I_MUTEX_PARENT); - if (error == -EINTR) - return error; + struct fscrypt_str name_str = FSTR_INIT((char *)qname->name, qname->len); + int ret; - dentry = lookup_one_len(name, parent->dentry, namelen); - error = PTR_ERR(dentry); + dentry = start_creating_killable(idmap, parent, qname); if (IS_ERR(dentry)) - goto out_unlock; - - error = -EEXIST; - if (dentry->d_inode) - goto out_dput; + return PTR_ERR(dentry); - error = btrfs_may_create(dir, dentry); - if (error) + ret = btrfs_may_create(idmap, dir, dentry); + if (ret) goto out_dput; /* * even if this name doesn't exist, we may get hash collisions. * check for them now when we can safely fail */ - error = btrfs_check_dir_item_collision(BTRFS_I(dir)->root, - dir->i_ino, name, - namelen); - if (error) + ret = btrfs_check_dir_item_collision(BTRFS_I(dir)->root, dir->i_ino, &name_str); + if (ret) goto out_dput; - down_read(&BTRFS_I(dir)->root->fs_info->subvol_sem); + down_read(&fs_info->subvol_sem); 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, name, namelen, - async_transid, readonly, inherit); - } else { - error = create_subvol(dir, dentry, name, namelen, - async_transid, inherit); - } - if (!error) + if (snap_src) + ret = create_snapshot(snap_src, dir, dentry, readonly, inherit); + else + ret = create_subvol(idmap, dir, dentry, inherit); + + if (!ret) fsnotify_mkdir(dir, dentry); out_up_read: - up_read(&BTRFS_I(dir)->root->fs_info->subvol_sem); + up_read(&fs_info->subvol_sem); out_dput: - dput(dentry); -out_unlock: - mutex_unlock(&dir->i_mutex); - return error; -} - -/* - * When we're defragging a range, we don't want to kick it off again - * if it is really just waiting for delalloc to send it down. - * If we find a nice big extent or delalloc range for the bytes in the - * file you want to defrag, we return 0 to let you know to skip this - * part of the file - */ -static int check_defrag_in_cache(struct inode *inode, u64 offset, int thresh) -{ - struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; - struct extent_map *em = NULL; - struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; - u64 end; - - read_lock(&em_tree->lock); - em = lookup_extent_mapping(em_tree, offset, PAGE_CACHE_SIZE); - read_unlock(&em_tree->lock); - - if (em) { - end = extent_map_end(em); - free_extent_map(em); - if (end - offset > thresh) - return 0; - } - /* if we already have a nice delalloc here, just stop */ - thresh /= 2; - end = count_range_bits(io_tree, &offset, offset + thresh, - thresh, EXTENT_DELALLOC, 1); - if (end >= thresh) - return 0; - return 1; + end_creating(dentry); + return ret; } -/* - * helper function to walk through a file and find extents - * newer than a specific transid, and smaller than thresh. - * - * This is used by the defragging code to find new and small - * extents - */ -static int find_new_extents(struct btrfs_root *root, - struct inode *inode, u64 newer_than, - u64 *off, int thresh) +static noinline int btrfs_mksnapshot(struct dentry *parent, + struct mnt_idmap *idmap, + struct qstr *qname, + struct btrfs_root *root, + bool readonly, + struct btrfs_qgroup_inherit *inherit) { - struct btrfs_path *path; - struct btrfs_key min_key; - struct btrfs_key max_key; - struct extent_buffer *leaf; - struct btrfs_file_extent_item *extent; - int type; int ret; - u64 ino = btrfs_ino(inode); - - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - - min_key.objectid = ino; - min_key.type = BTRFS_EXTENT_DATA_KEY; - min_key.offset = *off; - - max_key.objectid = ino; - max_key.type = (u8)-1; - max_key.offset = (u64)-1; - - path->keep_locks = 1; - - while(1) { - ret = btrfs_search_forward(root, &min_key, &max_key, - path, newer_than); - if (ret != 0) - goto none; - if (min_key.objectid != ino) - goto none; - if (min_key.type != BTRFS_EXTENT_DATA_KEY) - goto none; - - leaf = path->nodes[0]; - extent = btrfs_item_ptr(leaf, path->slots[0], - struct btrfs_file_extent_item); - - type = btrfs_file_extent_type(leaf, extent); - if (type == BTRFS_FILE_EXTENT_REG && - btrfs_file_extent_num_bytes(leaf, extent) < thresh && - check_defrag_in_cache(inode, min_key.offset, thresh)) { - *off = min_key.offset; - btrfs_free_path(path); - return 0; - } - - if (min_key.offset == (u64)-1) - goto none; - - min_key.offset++; - btrfs_release_path(path); - } -none: - btrfs_free_path(path); - return -ENOENT; -} - -static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start) -{ - struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; - struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; - struct extent_map *em; - u64 len = PAGE_CACHE_SIZE; /* - * hopefully we have this extent in the tree already, try without - * the full extent lock + * Force new buffered writes to reserve space even when NOCOW is + * possible. This is to avoid later writeback (running delalloc) to + * fallback to COW mode and unexpectedly fail with ENOSPC. */ - read_lock(&em_tree->lock); - em = lookup_extent_mapping(em_tree, start, len); - read_unlock(&em_tree->lock); - - if (!em) { - /* get the big lock and read metadata off disk */ - lock_extent(io_tree, start, start + len - 1); - em = btrfs_get_extent(inode, NULL, 0, start, len, 0); - unlock_extent(io_tree, start, start + len - 1); - - if (IS_ERR(em)) - return NULL; - } - - return em; -} + btrfs_drew_read_lock(&root->snapshot_lock); -static bool defrag_check_next_extent(struct inode *inode, struct extent_map *em) -{ - struct extent_map *next; - bool ret = true; - - /* this is the last extent */ - if (em->start + em->len >= i_size_read(inode)) - return false; - - next = defrag_lookup_extent(inode, em->start + em->len); - if (!next || next->block_start >= EXTENT_MAP_LAST_BYTE) - ret = false; - - free_extent_map(next); - return ret; -} - -static int should_defrag_range(struct inode *inode, u64 start, int thresh, - u64 *last_len, u64 *skip, u64 *defrag_end, - int compress) -{ - struct extent_map *em; - int ret = 1; - bool next_mergeable = true; + ret = btrfs_start_delalloc_snapshot(root, false); + if (ret) + goto out; /* - * make sure that once we start defragging an extent, we keep on - * defragging it + * All previous writes have started writeback in NOCOW mode, so now + * we force future writes to fallback to COW mode during snapshot + * creation. */ - if (start < *defrag_end) - return 1; - - *skip = 0; + atomic_inc(&root->snapshot_force_cow); - em = defrag_lookup_extent(inode, start); - if (!em) - return 0; - - /* this will cover holes, and inline extents */ - if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - ret = 0; - goto out; - } + btrfs_wait_ordered_extents(root, U64_MAX, NULL); - next_mergeable = defrag_check_next_extent(inode, em); + ret = btrfs_mksubvol(parent, idmap, qname, root, readonly, inherit); - /* - * we hit a real extent, if it is big or the next extent is not a - * real extent, don't bother defragging it - */ - if (!compress && (*last_len == 0 || *last_len >= thresh) && - (em->len >= thresh || !next_mergeable)) - ret = 0; + atomic_dec(&root->snapshot_force_cow); out: - /* - * last_len ends up being a counter of how many bytes we've defragged. - * every time we choose not to defrag an extent, we reset *last_len - * so that the next tiny extent will force a defrag. - * - * The end result of this is that tiny extents before a single big - * extent will force at least part of that big extent to be defragged. - */ - if (ret) { - *defrag_end = extent_map_end(em); - } else { - *last_len = 0; - *skip = extent_map_end(em); - *defrag_end = 0; - } - - free_extent_map(em); + btrfs_drew_read_unlock(&root->snapshot_lock); return ret; } /* - * it doesn't do much good to defrag one or two pages - * at a time. This pulls in a nice chunk of pages - * to COW and defrag. + * Try to start exclusive operation @type or cancel it if it's running. * - * It also makes sure the delalloc code has enough - * dirty data to avoid making new small extents as part - * of the defrag - * - * It's a good idea to start RA on this range - * before calling this. + * Return: + * 0 - normal mode, newly claimed op started + * >0 - normal mode, something else is running, + * return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS to user space + * ECANCELED - cancel mode, successful cancel + * ENOTCONN - cancel mode, operation not running anymore */ -static int cluster_pages_for_defrag(struct inode *inode, - struct page **pages, - unsigned long start_index, - int num_pages) -{ - unsigned long file_end; - u64 isize = i_size_read(inode); - u64 page_start; - u64 page_end; - u64 page_cnt; - int ret; - int i; - int i_done; - struct btrfs_ordered_extent *ordered; - struct extent_state *cached_state = NULL; - struct extent_io_tree *tree; - gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping); - - file_end = (isize - 1) >> PAGE_CACHE_SHIFT; - if (!isize || start_index > file_end) - return 0; - - page_cnt = min_t(u64, (u64)num_pages, (u64)file_end - start_index + 1); - - ret = btrfs_delalloc_reserve_space(inode, - page_cnt << PAGE_CACHE_SHIFT); - if (ret) - return ret; - i_done = 0; - tree = &BTRFS_I(inode)->io_tree; - - /* step one, lock all the pages */ - for (i = 0; i < page_cnt; i++) { - struct page *page; -again: - page = find_or_create_page(inode->i_mapping, - start_index + i, mask); - if (!page) - break; - - page_start = page_offset(page); - page_end = page_start + PAGE_CACHE_SIZE - 1; - while (1) { - lock_extent(tree, page_start, page_end); - ordered = btrfs_lookup_ordered_extent(inode, - page_start); - unlock_extent(tree, page_start, page_end); - if (!ordered) - break; - - unlock_page(page); - btrfs_start_ordered_extent(inode, ordered, 1); - btrfs_put_ordered_extent(ordered); - lock_page(page); - /* - * we unlocked the page above, so we need check if - * it was released or not. - */ - if (page->mapping != inode->i_mapping) { - unlock_page(page); - page_cache_release(page); - goto again; - } - } - - if (!PageUptodate(page)) { - btrfs_readpage(NULL, page); - lock_page(page); - if (!PageUptodate(page)) { - unlock_page(page); - page_cache_release(page); - ret = -EIO; - break; - } - } - - if (page->mapping != inode->i_mapping) { - unlock_page(page); - page_cache_release(page); - goto again; - } - - pages[i] = page; - i_done++; - } - if (!i_done || ret) - goto out; - - if (!(inode->i_sb->s_flags & MS_ACTIVE)) - goto out; - - /* - * so now we have a nice long stream of locked - * and up to date pages, lets wait on them - */ - for (i = 0; i < i_done; i++) - wait_on_page_writeback(pages[i]); - - page_start = page_offset(pages[0]); - page_end = page_offset(pages[i_done - 1]) + PAGE_CACHE_SIZE; - - lock_extent_bits(&BTRFS_I(inode)->io_tree, - page_start, page_end - 1, 0, &cached_state); - clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, - page_end - 1, EXTENT_DIRTY | EXTENT_DELALLOC | - EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0, - &cached_state, GFP_NOFS); - - if (i_done != page_cnt) { - spin_lock(&BTRFS_I(inode)->lock); - BTRFS_I(inode)->outstanding_extents++; - spin_unlock(&BTRFS_I(inode)->lock); - btrfs_delalloc_release_space(inode, - (page_cnt - i_done) << PAGE_CACHE_SHIFT); - } - - - set_extent_defrag(&BTRFS_I(inode)->io_tree, page_start, page_end - 1, - &cached_state, GFP_NOFS); - - unlock_extent_cached(&BTRFS_I(inode)->io_tree, - page_start, page_end - 1, &cached_state, - GFP_NOFS); - - for (i = 0; i < i_done; i++) { - clear_page_dirty_for_io(pages[i]); - ClearPageChecked(pages[i]); - set_page_extent_mapped(pages[i]); - set_page_dirty(pages[i]); - unlock_page(pages[i]); - page_cache_release(pages[i]); - } - return i_done; -out: - for (i = 0; i < i_done; i++) { - unlock_page(pages[i]); - page_cache_release(pages[i]); - } - btrfs_delalloc_release_space(inode, page_cnt << PAGE_CACHE_SHIFT); - return ret; - -} - -int btrfs_defrag_file(struct inode *inode, struct file *file, - struct btrfs_ioctl_defrag_range_args *range, - u64 newer_than, unsigned long max_to_defrag) +static int exclop_start_or_cancel_reloc(struct btrfs_fs_info *fs_info, + enum btrfs_exclusive_operation type, bool cancel) { - struct btrfs_root *root = BTRFS_I(inode)->root; - struct file_ra_state *ra = NULL; - unsigned long last_index; - u64 isize = i_size_read(inode); - u64 last_len = 0; - u64 skip = 0; - u64 defrag_end = 0; - u64 newer_off = range->start; - unsigned long i; - unsigned long ra_index = 0; - int ret; - int defrag_count = 0; - int compress_type = BTRFS_COMPRESS_ZLIB; - int extent_thresh = range->extent_thresh; - int max_cluster = (256 * 1024) >> PAGE_CACHE_SHIFT; - int cluster = max_cluster; - u64 new_align = ~((u64)128 * 1024 - 1); - struct page **pages = NULL; - - if (isize == 0) + if (!cancel) { + /* Start normal op */ + if (!btrfs_exclop_start(fs_info, type)) + return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS; + /* Exclusive operation is now claimed */ return 0; - - if (range->start >= isize) - return -EINVAL; - - if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS) { - if (range->compress_type > BTRFS_COMPRESS_TYPES) - return -EINVAL; - if (range->compress_type) - compress_type = range->compress_type; - } - - if (extent_thresh == 0) - extent_thresh = 256 * 1024; - - /* - * if we were not given a file, allocate a readahead - * context - */ - if (!file) { - ra = kzalloc(sizeof(*ra), GFP_NOFS); - if (!ra) - return -ENOMEM; - file_ra_state_init(ra, inode->i_mapping); - } else { - ra = &file->f_ra; - } - - pages = kmalloc(sizeof(struct page *) * max_cluster, - GFP_NOFS); - if (!pages) { - ret = -ENOMEM; - goto out_ra; - } - - /* find the last page to defrag */ - if (range->start + range->len > range->start) { - last_index = min_t(u64, isize - 1, - range->start + range->len - 1) >> PAGE_CACHE_SHIFT; - } else { - last_index = (isize - 1) >> PAGE_CACHE_SHIFT; } - if (newer_than) { - ret = find_new_extents(root, inode, newer_than, - &newer_off, 64 * 1024); - if (!ret) { - range->start = newer_off; - /* - * we always align our defrag to help keep - * the extents in the file evenly spaced - */ - i = (newer_off & new_align) >> PAGE_CACHE_SHIFT; - } else - goto out_ra; - } else { - i = range->start >> PAGE_CACHE_SHIFT; - } - if (!max_to_defrag) - max_to_defrag = last_index + 1; - - /* - * make writeback starts from i, so the defrag range can be - * written sequentially. - */ - if (i < inode->i_mapping->writeback_index) - inode->i_mapping->writeback_index = i; - - while (i <= last_index && defrag_count < max_to_defrag && - (i < (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> - PAGE_CACHE_SHIFT)) { + /* Cancel running op */ + if (btrfs_exclop_start_try_lock(fs_info, type)) { /* - * make sure we stop running if someone unmounts - * the FS - */ - if (!(inode->i_sb->s_flags & MS_ACTIVE)) - break; - - if (btrfs_defrag_cancelled(root->fs_info)) { - printk(KERN_DEBUG "btrfs: defrag_file cancelled\n"); - ret = -EAGAIN; - break; - } - - if (!should_defrag_range(inode, (u64)i << PAGE_CACHE_SHIFT, - extent_thresh, &last_len, &skip, - &defrag_end, range->flags & - BTRFS_DEFRAG_RANGE_COMPRESS)) { - unsigned long next; - /* - * the should_defrag function tells us how much to skip - * bump our counter by the suggested amount - */ - next = (skip + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - i = max(i + 1, next); - continue; - } - - if (!newer_than) { - cluster = (PAGE_CACHE_ALIGN(defrag_end) >> - PAGE_CACHE_SHIFT) - i; - cluster = min(cluster, max_cluster); - } else { - cluster = max_cluster; - } - - if (range->flags & BTRFS_DEFRAG_RANGE_COMPRESS) - BTRFS_I(inode)->force_compress = compress_type; - - if (i + cluster > ra_index) { - ra_index = max(i, ra_index); - btrfs_force_ra(inode->i_mapping, ra, file, ra_index, - cluster); - ra_index += max_cluster; - } - - mutex_lock(&inode->i_mutex); - ret = cluster_pages_for_defrag(inode, pages, i, cluster); - if (ret < 0) { - mutex_unlock(&inode->i_mutex); - goto out_ra; - } - - defrag_count += ret; - balance_dirty_pages_ratelimited(inode->i_mapping); - mutex_unlock(&inode->i_mutex); - - if (newer_than) { - if (newer_off == (u64)-1) - break; - - if (ret > 0) - i += ret; - - newer_off = max(newer_off + 1, - (u64)i << PAGE_CACHE_SHIFT); - - ret = find_new_extents(root, inode, - newer_than, &newer_off, - 64 * 1024); - if (!ret) { - range->start = newer_off; - i = (newer_off & new_align) >> PAGE_CACHE_SHIFT; - } else { - break; - } - } else { - if (ret > 0) { - i += ret; - last_len += ret << PAGE_CACHE_SHIFT; - } else { - i++; - last_len = 0; - } - } - } - - if ((range->flags & BTRFS_DEFRAG_RANGE_START_IO)) - filemap_flush(inode->i_mapping); - - if ((range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)) { - /* the filemap_flush will queue IO into the worker threads, but - * we have to make sure the IO is actually started and that - * ordered extents get created before we return + * This blocks any exclop finish from setting it to NONE, so we + * request cancellation. Either it runs and we will wait for it, + * or it has finished and no waiting will happen. */ - atomic_inc(&root->fs_info->async_submit_draining); - while (atomic_read(&root->fs_info->nr_async_submits) || - atomic_read(&root->fs_info->async_delalloc_pages)) { - wait_event(root->fs_info->async_submit_wait, - (atomic_read(&root->fs_info->nr_async_submits) == 0 && - atomic_read(&root->fs_info->async_delalloc_pages) == 0)); - } - atomic_dec(&root->fs_info->async_submit_draining); + atomic_inc(&fs_info->reloc_cancel_req); + btrfs_exclop_start_unlock(fs_info); - mutex_lock(&inode->i_mutex); - BTRFS_I(inode)->force_compress = BTRFS_COMPRESS_NONE; - mutex_unlock(&inode->i_mutex); - } + if (test_bit(BTRFS_FS_RELOC_RUNNING, &fs_info->flags)) + wait_on_bit(&fs_info->flags, BTRFS_FS_RELOC_RUNNING, + TASK_INTERRUPTIBLE); - if (range->compress_type == BTRFS_COMPRESS_LZO) { - btrfs_set_fs_incompat(root->fs_info, COMPRESS_LZO); + return -ECANCELED; } - ret = defrag_count; - -out_ra: - if (!file) - kfree(ra); - kfree(pages); - return ret; + /* Something else is running or none */ + return -ENOTCONN; } static noinline int btrfs_ioctl_resize(struct file *file, void __user *arg) { + BTRFS_DEV_LOOKUP_ARGS(args); + struct btrfs_root *root = BTRFS_I(file_inode(file))->root; + struct btrfs_fs_info *fs_info = root->fs_info; u64 new_size; u64 old_size; u64 devid = 1; - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; struct btrfs_ioctl_vol_args *vol_args; - struct btrfs_trans_handle *trans; struct btrfs_device *device = NULL; char *sizestr; char *devstr = NULL; int ret = 0; int mod = 0; + bool cancel; if (!capable(CAP_SYS_ADMIN)) return -EPERM; @@ -1375,57 +1039,63 @@ static noinline int btrfs_ioctl_resize(struct file *file, if (ret) return ret; - if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running, - 1)) { - pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n"); - mnt_drop_write_file(file); - return -EINVAL; - } - - mutex_lock(&root->fs_info->volume_mutex); + /* + * Read the arguments before checking exclusivity to be able to + * distinguish regular resize and cancel + */ vol_args = memdup_user(arg, sizeof(*vol_args)); if (IS_ERR(vol_args)) { ret = PTR_ERR(vol_args); - goto out; + goto out_drop; } - - vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; + ret = btrfs_check_ioctl_vol_args_path(vol_args); + if (ret < 0) + goto out_free; sizestr = vol_args->name; + cancel = (strcmp("cancel", sizestr) == 0); + ret = exclop_start_or_cancel_reloc(fs_info, BTRFS_EXCLOP_RESIZE, cancel); + if (ret) + goto out_free; + /* Exclusive operation is now claimed */ + devstr = strchr(sizestr, ':'); if (devstr) { - char *end; sizestr = devstr + 1; *devstr = '\0'; devstr = vol_args->name; - devid = simple_strtoull(devstr, &end, 10); + ret = kstrtoull(devstr, 10, &devid); + if (ret) + goto out_finish; if (!devid) { ret = -EINVAL; - goto out_free; + goto out_finish; } - printk(KERN_INFO "btrfs: resizing devid %llu\n", - (unsigned long long)devid); + btrfs_info(fs_info, "resizing devid %llu", devid); } - device = btrfs_find_device(root->fs_info, devid, NULL, NULL); + args.devid = devid; + device = btrfs_find_device(fs_info->fs_devices, &args); if (!device) { - printk(KERN_INFO "btrfs: resizer unable to find device %llu\n", - (unsigned long long)devid); + btrfs_info(fs_info, "resizer unable to find device %llu", + devid); ret = -ENODEV; - goto out_free; + goto out_finish; } - if (!device->writeable) { - printk(KERN_INFO "btrfs: resizer unable to apply on " - "readonly device %llu\n", - (unsigned long long)devid); + if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) { + btrfs_info(fs_info, + "resizer unable to apply on readonly device %llu", + devid); ret = -EPERM; - goto out_free; + goto out_finish; } if (!strcmp(sizestr, "max")) - new_size = device->bdev->bd_inode->i_size; + new_size = bdev_nr_bytes(device->bdev); else { + char *retptr; + if (sizestr[0] == '-') { mod = -1; sizestr++; @@ -1433,113 +1103,136 @@ static noinline int btrfs_ioctl_resize(struct file *file, mod = 1; sizestr++; } - new_size = memparse(sizestr, NULL); - if (new_size == 0) { + new_size = memparse(sizestr, &retptr); + if (*retptr != '\0' || new_size == 0) { ret = -EINVAL; - goto out_free; + goto out_finish; } } - if (device->is_tgtdev_for_dev_replace) { + if (test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) { ret = -EPERM; - goto out_free; + goto out_finish; } - old_size = device->total_bytes; + old_size = btrfs_device_get_total_bytes(device); if (mod < 0) { if (new_size > old_size) { ret = -EINVAL; - goto out_free; + goto out_finish; } new_size = old_size - new_size; } else if (mod > 0) { + if (new_size > ULLONG_MAX - old_size) { + ret = -ERANGE; + goto out_finish; + } new_size = old_size + new_size; } - if (new_size < 256 * 1024 * 1024) { + if (new_size < SZ_256M) { ret = -EINVAL; - goto out_free; + goto out_finish; } - if (new_size > device->bdev->bd_inode->i_size) { + if (new_size > bdev_nr_bytes(device->bdev)) { ret = -EFBIG; - goto out_free; + goto out_finish; } - do_div(new_size, root->sectorsize); - new_size *= root->sectorsize; - - printk_in_rcu(KERN_INFO "btrfs: new size for %s is %llu\n", - rcu_str_deref(device->name), - (unsigned long long)new_size); + new_size = round_down(new_size, fs_info->sectorsize); if (new_size > old_size) { + struct btrfs_trans_handle *trans; + trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { ret = PTR_ERR(trans); - goto out_free; + goto out_finish; } ret = btrfs_grow_device(trans, device, new_size); - btrfs_commit_transaction(trans, root); + btrfs_commit_transaction(trans); } else if (new_size < old_size) { ret = btrfs_shrink_device(device, new_size); } /* equal, nothing need to do */ + if (ret == 0 && new_size != old_size) + btrfs_info(fs_info, + "resize device %s (devid %llu) from %llu to %llu", + btrfs_dev_name(device), device->devid, + old_size, new_size); +out_finish: + btrfs_exclop_finish(fs_info); out_free: kfree(vol_args); -out: - mutex_unlock(&root->fs_info->volume_mutex); - atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0); +out_drop: mnt_drop_write_file(file); return ret; } -static noinline int btrfs_ioctl_snap_create_transid(struct file *file, - char *name, unsigned long fd, int subvol, - u64 *transid, bool readonly, +static noinline int __btrfs_ioctl_snap_create(struct file *file, + struct mnt_idmap *idmap, + const char *name, unsigned long fd, bool subvol, + bool readonly, struct btrfs_qgroup_inherit *inherit) { - int namelen; int ret = 0; + struct qstr qname = QSTR_INIT(name, strlen(name)); + + if (!S_ISDIR(file_inode(file)->i_mode)) + return -ENOTDIR; ret = mnt_want_write_file(file); if (ret) goto out; - namelen = strlen(name); if (strchr(name, '/')) { ret = -EINVAL; goto out_drop_write; } - if (name[0] == '.' && - (namelen == 1 || (name[1] == '.' && namelen == 2))) { + if (qname.name[0] == '.' && + (qname.len == 1 || (qname.name[1] == '.' && qname.len == 2))) { ret = -EEXIST; goto out_drop_write; } if (subvol) { - ret = btrfs_mksubvol(&file->f_path, name, namelen, - NULL, transid, readonly, inherit); + ret = btrfs_mksubvol(file_dentry(file), idmap, &qname, NULL, + readonly, inherit); } else { - struct fd src = fdget(fd); + CLASS(fd, src)(fd); struct inode *src_inode; - if (!src.file) { + if (fd_empty(src)) { ret = -EINVAL; goto out_drop_write; } - src_inode = file_inode(src.file); + src_inode = file_inode(fd_file(src)); if (src_inode->i_sb != file_inode(file)->i_sb) { - printk(KERN_INFO "btrfs: Snapshot src from " - "another FS\n"); + btrfs_info(BTRFS_I(file_inode(file))->root->fs_info, + "Snapshot src from another FS"); + ret = -EXDEV; + } else if (!inode_owner_or_capable(idmap, src_inode)) { + /* + * Subvolume creation is not restricted, but snapshots + * are limited to own subvolumes only + */ + ret = -EPERM; + } else if (btrfs_ino(BTRFS_I(src_inode)) != BTRFS_FIRST_FREE_OBJECTID) { + /* + * Snapshots must be made with the src_inode referring + * to the subvolume inode, otherwise the permission + * checking above is useless because we may have + * permission on a lower directory but not the subvol + * itself. + */ ret = -EINVAL; } else { - ret = btrfs_mksubvol(&file->f_path, name, namelen, - BTRFS_I(src_inode)->root, - transid, readonly, inherit); + ret = btrfs_mksnapshot(file_dentry(file), idmap, &qname, + BTRFS_I(src_inode)->root, + readonly, inherit); } - fdput(src); } out_drop_write: mnt_drop_write_file(file); @@ -1548,92 +1241,101 @@ out: } static noinline int btrfs_ioctl_snap_create(struct file *file, - void __user *arg, int subvol) + void __user *arg, bool subvol) { struct btrfs_ioctl_vol_args *vol_args; int ret; + if (!S_ISDIR(file_inode(file)->i_mode)) + return -ENOTDIR; + 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'; + ret = btrfs_check_ioctl_vol_args_path(vol_args); + if (ret < 0) + goto out; - ret = btrfs_ioctl_snap_create_transid(file, vol_args->name, - vol_args->fd, subvol, - NULL, false, NULL); + ret = __btrfs_ioctl_snap_create(file, file_mnt_idmap(file), + vol_args->name, vol_args->fd, subvol, + false, NULL); +out: kfree(vol_args); return ret; } static noinline int btrfs_ioctl_snap_create_v2(struct file *file, - void __user *arg, int subvol) + void __user *arg, bool subvol) { 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; + if (!S_ISDIR(file_inode(file)->i_mode)) + return -ENOTDIR; + vol_args = memdup_user(arg, sizeof(*vol_args)); if (IS_ERR(vol_args)) return PTR_ERR(vol_args); - vol_args->name[BTRFS_SUBVOL_NAME_MAX] = '\0'; + ret = btrfs_check_ioctl_vol_args2_subvol_name(vol_args); + if (ret < 0) + goto free_args; - 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 out; + goto free_args; } - if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC) - ptr = &transid; if (vol_args->flags & BTRFS_SUBVOL_RDONLY) readonly = true; if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) { - if (vol_args->size > PAGE_CACHE_SIZE) { + struct btrfs_fs_info *fs_info = inode_to_fs_info(file_inode(file)); + + if (vol_args->size < sizeof(*inherit) || + vol_args->size > PAGE_SIZE) { ret = -EINVAL; - goto out; + goto free_args; } inherit = memdup_user(vol_args->qgroup_inherit, vol_args->size); if (IS_ERR(inherit)) { ret = PTR_ERR(inherit); - goto out; + goto free_args; } - } - ret = btrfs_ioctl_snap_create_transid(file, vol_args->name, - vol_args->fd, subvol, ptr, - readonly, inherit); + ret = btrfs_qgroup_check_inherit(fs_info, inherit, vol_args->size); + if (ret < 0) + goto free_inherit; + } - if (ret == 0 && ptr && - copy_to_user(arg + - offsetof(struct btrfs_ioctl_vol_args_v2, - transid), ptr, sizeof(*ptr))) - ret = -EFAULT; -out: - kfree(vol_args); + ret = __btrfs_ioctl_snap_create(file, file_mnt_idmap(file), + vol_args->name, vol_args->fd, subvol, + readonly, inherit); + if (ret) + goto free_inherit; +free_inherit: kfree(inherit); +free_args: + kfree(vol_args); return ret; } -static noinline int btrfs_ioctl_subvol_getflags(struct file *file, +static noinline int btrfs_ioctl_subvol_getflags(struct btrfs_inode *inode, void __user *arg) { - struct inode *inode = file_inode(file); - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; int ret = 0; u64 flags = 0; if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) return -EINVAL; - down_read(&root->fs_info->subvol_sem); + down_read(&fs_info->subvol_sem); if (btrfs_root_readonly(root)) flags |= BTRFS_SUBVOL_RDONLY; - up_read(&root->fs_info->subvol_sem); + up_read(&fs_info->subvol_sem); if (copy_to_user(arg, &flags, sizeof(flags))) ret = -EFAULT; @@ -1645,17 +1347,21 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file, void __user *arg) { struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; u64 root_flags; u64 flags; int ret = 0; + if (!inode_owner_or_capable(file_mnt_idmap(file), inode)) + return -EPERM; + ret = mnt_want_write_file(file); if (ret) goto out; - if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) { + if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) { ret = -EINVAL; goto out_drop_write; } @@ -1665,34 +1371,40 @@ 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; } - if (!inode_owner_or_capable(inode)) { - ret = -EACCES; - goto out_drop_write; - } - - down_write(&root->fs_info->subvol_sem); + down_write(&fs_info->subvol_sem); /* nothing to do */ if (!!(flags & BTRFS_SUBVOL_RDONLY) == btrfs_root_readonly(root)) goto out_drop_sem; root_flags = btrfs_root_flags(&root->root_item); - if (flags & BTRFS_SUBVOL_RDONLY) + if (flags & BTRFS_SUBVOL_RDONLY) { btrfs_set_root_flags(&root->root_item, root_flags | BTRFS_ROOT_SUBVOL_RDONLY); - else - btrfs_set_root_flags(&root->root_item, + } else { + /* + * Block RO -> RW transition if this subvolume is involved in + * send + */ + spin_lock(&root->root_item_lock); + if (root->send_in_progress == 0) { + btrfs_set_root_flags(&root->root_item, root_flags & ~BTRFS_ROOT_SUBVOL_RDONLY); + spin_unlock(&root->root_item_lock); + } else { + spin_unlock(&root->root_item_lock); + btrfs_warn(fs_info, + "Attempt to set subvolume %llu read-write during send", + btrfs_root_id(root)); + ret = -EPERM; + goto out_drop_sem; + } + } trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { @@ -1700,59 +1412,28 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file, goto out_reset; } - ret = btrfs_update_root(trans, root->fs_info->tree_root, + ret = btrfs_update_root(trans, fs_info->tree_root, &root->root_key, &root->root_item); + if (ret < 0) { + btrfs_end_transaction(trans); + goto out_reset; + } + + ret = btrfs_commit_transaction(trans); - btrfs_commit_transaction(trans, root); out_reset: if (ret) btrfs_set_root_flags(&root->root_item, root_flags); out_drop_sem: - up_write(&root->fs_info->subvol_sem); + up_write(&fs_info->subvol_sem); out_drop_write: mnt_drop_write_file(file); out: return ret; } -/* - * helper to check if the subvolume references other subvolumes - */ -static noinline int may_destroy_subvol(struct btrfs_root *root) -{ - struct btrfs_path *path; - struct btrfs_key key; - int ret; - - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - - key.objectid = root->root_key.objectid; - key.type = BTRFS_ROOT_REF_KEY; - key.offset = (u64)-1; - - ret = btrfs_search_slot(NULL, root->fs_info->tree_root, - &key, path, 0, 0); - if (ret < 0) - goto out; - BUG_ON(ret == 0); - - ret = 0; - if (path->slots[0] > 0) { - path->slots[0]--; - btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); - if (key.objectid == root->root_key.objectid && - key.type == BTRFS_ROOT_REF_KEY) - ret = -ENOTEMPTY; - } -out: - btrfs_free_path(path); - return ret; -} - -static noinline int key_in_sk(struct btrfs_key *key, - struct btrfs_ioctl_search_key *sk) +static noinline bool key_in_sk(const struct btrfs_key *key, + const struct btrfs_ioctl_search_key *sk) { struct btrfs_key test; int ret; @@ -1763,7 +1444,7 @@ static noinline int key_in_sk(struct btrfs_key *key, ret = btrfs_comp_cpu_keys(key, &test); if (ret < 0) - return 0; + return false; test.objectid = sk->max_objectid; test.type = sk->max_type; @@ -1771,21 +1452,22 @@ static noinline int key_in_sk(struct btrfs_key *key, ret = btrfs_comp_cpu_keys(key, &test); if (ret > 0) - return 0; - return 1; + return false; + return true; } -static noinline int copy_to_sk(struct btrfs_root *root, - struct btrfs_path *path, +static noinline int copy_to_sk(struct btrfs_path *path, struct btrfs_key *key, - struct btrfs_ioctl_search_key *sk, - char *buf, + const struct btrfs_ioctl_search_key *sk, + u64 *buf_size, + char __user *ubuf, unsigned long *sk_offset, int *num_found) { u64 found_transid; struct extent_buffer *leaf; struct btrfs_ioctl_search_header sh; + struct btrfs_key test; unsigned long item_off; unsigned long item_len; int nritems; @@ -1805,144 +1487,233 @@ static noinline int copy_to_sk(struct btrfs_root *root, for (i = slot; i < nritems; i++) { item_off = btrfs_item_ptr_offset(leaf, i); - item_len = btrfs_item_size_nr(leaf, i); + item_len = btrfs_item_size(leaf, i); btrfs_item_key_to_cpu(leaf, key, i); if (!key_in_sk(key, sk)) continue; - if (sizeof(sh) + item_len > BTRFS_SEARCH_ARGS_BUFSIZE) + if (sizeof(sh) + item_len > *buf_size) { + if (*num_found) { + ret = 1; + goto out; + } + + /* + * return one empty item back for v1, which does not + * handle -EOVERFLOW + */ + + *buf_size = sizeof(sh) + item_len; item_len = 0; + ret = -EOVERFLOW; + } - if (sizeof(sh) + item_len + *sk_offset > - BTRFS_SEARCH_ARGS_BUFSIZE) { + if (sizeof(sh) + item_len + *sk_offset > *buf_size) { ret = 1; - goto overflow; + goto out; } sh.objectid = key->objectid; - sh.offset = key->offset; sh.type = key->type; + sh.offset = key->offset; sh.len = item_len; sh.transid = found_transid; - /* copy search result header */ - memcpy(buf + *sk_offset, &sh, sizeof(sh)); + /* + * Copy search result header. If we fault then loop again so we + * can fault in the pages and -EFAULT there if there's a + * problem. Otherwise we'll fault and then copy the buffer in + * properly this next time through + */ + if (copy_to_user_nofault(ubuf + *sk_offset, &sh, sizeof(sh))) { + ret = 0; + goto out; + } + *sk_offset += sizeof(sh); if (item_len) { - char *p = buf + *sk_offset; - /* copy the item */ - read_extent_buffer(leaf, p, - item_off, item_len); + char __user *up = ubuf + *sk_offset; + /* + * Copy the item, same behavior as above, but reset the + * * sk_offset so we copy the full thing again. + */ + if (read_extent_buffer_to_user_nofault(leaf, up, + item_off, item_len)) { + ret = 0; + *sk_offset -= sizeof(sh); + goto out; + } + *sk_offset += item_len; } (*num_found)++; - if (*num_found >= sk->nr_items) - break; + if (ret) /* -EOVERFLOW from above */ + goto out; + + if (*num_found >= sk->nr_items) { + ret = 1; + goto out; + } } advance_key: ret = 0; - if (key->offset < (u64)-1 && key->offset < sk->max_offset) + test.objectid = sk->max_objectid; + test.type = sk->max_type; + test.offset = sk->max_offset; + if (btrfs_comp_cpu_keys(key, &test) >= 0) + ret = 1; + else if (key->offset < (u64)-1) key->offset++; - else if (key->type < (u8)-1 && key->type < sk->max_type) { + else if (key->type < (u8)-1) { key->offset = 0; key->type++; - } else if (key->objectid < (u64)-1 && key->objectid < sk->max_objectid) { + } else if (key->objectid < (u64)-1) { key->offset = 0; key->type = 0; key->objectid++; } else ret = 1; -overflow: +out: + /* + * 0: all items from this leaf copied, continue with next + * 1: * more items can be copied, but unused buffer is too small + * * all items were found + * Either way, it will stops the loop which iterates to the next + * leaf + * -EOVERFLOW: item was to large for buffer + * -EFAULT: could not copy extent buffer back to userspace + */ return ret; } -static noinline int search_ioctl(struct inode *inode, - struct btrfs_ioctl_search_args *args) +static noinline int search_ioctl(struct btrfs_root *root, + struct btrfs_ioctl_search_key *sk, + u64 *buf_size, + char __user *ubuf) { - struct btrfs_root *root; + struct btrfs_fs_info *info = root->fs_info; struct btrfs_key key; - struct btrfs_key max_key; - struct btrfs_path *path; - struct btrfs_ioctl_search_key *sk = &args->key; - struct btrfs_fs_info *info = BTRFS_I(inode)->root->fs_info; + BTRFS_PATH_AUTO_FREE(path); int ret; int num_found = 0; unsigned long sk_offset = 0; + if (*buf_size < sizeof(struct btrfs_ioctl_search_header)) { + *buf_size = sizeof(struct btrfs_ioctl_search_header); + return -EOVERFLOW; + } + path = btrfs_alloc_path(); if (!path) return -ENOMEM; if (sk->tree_id == 0) { - /* search the root of the inode that was passed */ - root = BTRFS_I(inode)->root; + /* Search the root that we got passed. */ + root = btrfs_grab_root(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); - if (IS_ERR(root)) { - printk(KERN_ERR "could not find root %llu\n", - sk->tree_id); - btrfs_free_path(path); - return -ENOENT; - } + /* Look up the root from the arguments. */ + root = btrfs_get_fs_root(info, sk->tree_id, true); + if (IS_ERR(root)) + return PTR_ERR(root); } key.objectid = sk->min_objectid; key.type = sk->min_type; key.offset = sk->min_offset; - max_key.objectid = sk->max_objectid; - max_key.type = sk->max_type; - max_key.offset = sk->max_offset; + while (1) { + /* + * Ensure that the whole user buffer is faulted in at sub-page + * granularity, otherwise the loop may live-lock. + */ + if (fault_in_subpage_writeable(ubuf + sk_offset, *buf_size - sk_offset)) { + ret = -EFAULT; + break; + } - path->keep_locks = 1; + ret = btrfs_search_forward(root, &key, path, sk->min_transid); + if (ret) + break; - while(1) { - ret = btrfs_search_forward(root, &key, &max_key, path, - sk->min_transid); - if (ret != 0) { - if (ret > 0) - ret = 0; - goto err; - } - ret = copy_to_sk(root, path, &key, sk, args->buf, + ret = copy_to_sk(path, &key, sk, buf_size, ubuf, &sk_offset, &num_found); btrfs_release_path(path); - if (ret || num_found >= sk->nr_items) + if (ret) break; } - ret = 0; -err: + /* Normalize return values from btrfs_search_forward() and copy_to_sk(). */ + if (ret > 0) + ret = 0; + sk->nr_items = num_found; - btrfs_free_path(path); + btrfs_put_root(root); return ret; } -static noinline int btrfs_ioctl_tree_search(struct file *file, - void __user *argp) +static noinline int btrfs_ioctl_tree_search(struct btrfs_root *root, + void __user *argp) { - struct btrfs_ioctl_search_args *args; - struct inode *inode; - int ret; + struct btrfs_ioctl_search_args __user *uargs = argp; + struct btrfs_ioctl_search_key sk; + int ret; + u64 buf_size; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - args = memdup_user(argp, sizeof(*args)); - if (IS_ERR(args)) - return PTR_ERR(args); + if (copy_from_user(&sk, &uargs->key, sizeof(sk))) + return -EFAULT; - inode = file_inode(file); - ret = search_ioctl(inode, args); - if (ret == 0 && copy_to_user(argp, args, sizeof(*args))) + buf_size = sizeof(uargs->buf); + + ret = search_ioctl(root, &sk, &buf_size, uargs->buf); + + /* + * In the origin implementation an overflow is handled by returning a + * search header with a len of zero, so reset ret. + */ + if (ret == -EOVERFLOW) + ret = 0; + + if (ret == 0 && copy_to_user(&uargs->key, &sk, sizeof(sk))) ret = -EFAULT; - kfree(args); + return ret; +} + +static noinline int btrfs_ioctl_tree_search_v2(struct btrfs_root *root, + void __user *argp) +{ + struct btrfs_ioctl_search_args_v2 __user *uarg = argp; + struct btrfs_ioctl_search_args_v2 args; + int ret; + u64 buf_size; + const u64 buf_limit = SZ_16M; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + /* copy search header and buffer size */ + if (copy_from_user(&args, uarg, sizeof(args))) + return -EFAULT; + + buf_size = args.buf_size; + + /* limit result size to 16MB */ + if (buf_size > buf_limit) + buf_size = buf_limit; + + ret = search_ioctl(root, &args.key, &buf_size, + (char __user *)(&uarg->buf[0])); + if (ret == 0 && copy_to_user(&uarg->key, &args.key, sizeof(args.key))) + ret = -EFAULT; + else if (ret == -EOVERFLOW && + copy_to_user(&uarg->buf_size, &buf_size, sizeof(buf_size))) + ret = -EFAULT; + return ret; } @@ -1962,7 +1733,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info, int total_len = 0; struct btrfs_inode_ref *iref; struct extent_buffer *l; - struct btrfs_path *path; + BTRFS_PATH_AUTO_FREE(path); if (dirid == BTRFS_FIRST_FREE_OBJECTID) { name[0]='\0'; @@ -1973,15 +1744,12 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info, if (!path) return -ENOMEM; - ptr = &name[BTRFS_INO_LOOKUP_PATH_MAX]; + ptr = &name[BTRFS_INO_LOOKUP_PATH_MAX - 1]; - 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, tree_id, true); if (IS_ERR(root)) { - printk(KERN_ERR "could not find root %llu\n", tree_id); - ret = -ENOENT; + ret = PTR_ERR(root); + root = NULL; goto out; } @@ -1989,32 +1757,29 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info, key.type = BTRFS_INODE_REF_KEY; key.offset = (u64)-1; - while(1) { - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + while (1) { + ret = btrfs_search_backwards(root, &key, path); if (ret < 0) goto out; - - l = path->nodes[0]; - slot = path->slots[0]; - if (ret > 0 && slot > 0) - slot--; - btrfs_item_key_to_cpu(l, &key, slot); - - if (ret > 0 && (key.objectid != dirid || - key.type != BTRFS_INODE_REF_KEY)) { + else if (ret > 0) { ret = -ENOENT; goto out; } + l = path->nodes[0]; + slot = path->slots[0]; + iref = btrfs_item_ptr(l, slot, struct btrfs_inode_ref); len = btrfs_inode_ref_name_len(l, iref); ptr -= len + 1; total_len += len + 1; - if (ptr < name) + if (ptr < name) { + ret = -ENAMETOOLONG; goto out; + } *(ptr + len) = '/'; - read_extent_buffer(l, ptr,(unsigned long)(iref + 1), len); + read_extent_buffer(l, ptr, (unsigned long)(iref + 1), len); if (key.offset == BTRFS_FIRST_FREE_OBJECTID) break; @@ -2024,39 +1789,227 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info, key.offset = (u64)-1; dirid = key.objectid; } - if (ptr < name) - goto out; memmove(name, ptr, total_len); - name[total_len]='\0'; + name[total_len] = '\0'; ret = 0; out: - btrfs_free_path(path); + btrfs_put_root(root); return ret; } -static noinline int btrfs_ioctl_ino_lookup(struct file *file, - void __user *argp) +static int btrfs_search_path_in_tree_user(struct mnt_idmap *idmap, + struct inode *inode, + struct btrfs_ioctl_ino_lookup_user_args *args) { - struct btrfs_ioctl_ino_lookup_args *args; - struct inode *inode; - int ret; + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + u64 upper_limit = btrfs_ino(BTRFS_I(inode)); + u64 treeid = btrfs_root_id(BTRFS_I(inode)->root); + u64 dirid = args->dirid; + unsigned long item_off; + unsigned long item_len; + struct btrfs_inode_ref *iref; + struct btrfs_root_ref *rref; + struct btrfs_root *root = NULL; + BTRFS_PATH_AUTO_FREE(path); + struct btrfs_key key; + struct extent_buffer *leaf; + char *ptr; + int slot; + int len; + int total_len = 0; + int ret; - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + /* + * If the bottom subvolume does not exist directly under upper_limit, + * construct the path in from the bottom up. + */ + if (dirid != upper_limit) { + ptr = &args->path[BTRFS_INO_LOOKUP_USER_PATH_MAX - 1]; + + root = btrfs_get_fs_root(fs_info, treeid, true); + if (IS_ERR(root)) + return PTR_ERR(root); + + key.objectid = dirid; + key.type = BTRFS_INODE_REF_KEY; + key.offset = (u64)-1; + while (1) { + struct btrfs_inode *temp_inode; + + ret = btrfs_search_backwards(root, &key, path); + if (ret < 0) + goto out_put; + else if (ret > 0) { + ret = -ENOENT; + goto out_put; + } + + leaf = path->nodes[0]; + slot = path->slots[0]; + + iref = btrfs_item_ptr(leaf, slot, struct btrfs_inode_ref); + len = btrfs_inode_ref_name_len(leaf, iref); + ptr -= len + 1; + total_len += len + 1; + if (ptr < args->path) { + ret = -ENAMETOOLONG; + goto out_put; + } + + *(ptr + len) = '/'; + read_extent_buffer(leaf, ptr, + (unsigned long)(iref + 1), len); + + /* + * We don't need the path anymore, so release it and + * avoid deadlocks and lockdep warnings in case + * btrfs_iget() needs to lookup the inode from its root + * btree and lock the same leaf. + */ + btrfs_release_path(path); + temp_inode = btrfs_iget(key.offset, root); + if (IS_ERR(temp_inode)) { + ret = PTR_ERR(temp_inode); + goto out_put; + } + /* Check the read+exec permission of this directory. */ + ret = inode_permission(idmap, &temp_inode->vfs_inode, + MAY_READ | MAY_EXEC); + iput(&temp_inode->vfs_inode); + if (ret) + goto out_put; + + if (key.offset == upper_limit) + break; + if (key.objectid == BTRFS_FIRST_FREE_OBJECTID) { + ret = -EACCES; + goto out_put; + } + + key.objectid = key.offset; + key.offset = (u64)-1; + dirid = key.objectid; + } + + 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 */ + key.objectid = treeid; + key.type = BTRFS_ROOT_REF_KEY; + key.offset = args->treeid; + ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0); + if (ret < 0) + return ret; + else if (ret > 0) + return -ENOENT; + + leaf = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(leaf, &key, slot); + + item_off = btrfs_item_ptr_offset(leaf, slot); + item_len = btrfs_item_size(leaf, slot); + /* Check if dirid in ROOT_REF corresponds to passed dirid */ + rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref); + if (args->dirid != btrfs_root_ref_dirid(leaf, rref)) + return -EINVAL; + + /* Copy subvolume's name */ + item_off += sizeof(struct btrfs_root_ref); + item_len -= sizeof(struct btrfs_root_ref); + read_extent_buffer(leaf, args->name, item_off, item_len); + args->name[item_len] = 0; + +out_put: + btrfs_put_root(root); + + return ret; +} + +static noinline int btrfs_ioctl_ino_lookup(struct btrfs_root *root, + void __user *argp) +{ + struct btrfs_ioctl_ino_lookup_args *args; + int ret = 0; args = memdup_user(argp, sizeof(*args)); if (IS_ERR(args)) return PTR_ERR(args); - inode = file_inode(file); - + /* + * Unprivileged query to obtain the containing subvolume root id. The + * path is reset so it's consistent with btrfs_search_path_in_tree. + */ if (args->treeid == 0) - args->treeid = BTRFS_I(inode)->root->root_key.objectid; + args->treeid = btrfs_root_id(root); - ret = btrfs_search_path_in_tree(BTRFS_I(inode)->root->fs_info, + if (args->objectid == BTRFS_FIRST_FREE_OBJECTID) { + args->name[0] = 0; + goto out; + } + + if (!capable(CAP_SYS_ADMIN)) { + ret = -EPERM; + goto out; + } + + ret = btrfs_search_path_in_tree(root->fs_info, args->treeid, args->objectid, args->name); +out: + if (ret == 0 && copy_to_user(argp, args, sizeof(*args))) + ret = -EFAULT; + + kfree(args); + return ret; +} + +/* + * Version of ino_lookup ioctl (unprivileged) + * + * The main differences from ino_lookup ioctl are: + * + * 1. Read + Exec permission will be checked using inode_permission() during + * path construction. -EACCES will be returned in case of failure. + * 2. Path construction will be stopped at the inode number which corresponds + * to the fd with which this ioctl is called. If constructed path does not + * exist under fd's inode, -EACCES will be returned. + * 3. The name of bottom subvolume is also searched and filled. + */ +static int btrfs_ioctl_ino_lookup_user(struct file *file, void __user *argp) +{ + struct btrfs_ioctl_ino_lookup_user_args *args; + struct inode *inode; + int ret; + + args = memdup_user(argp, sizeof(*args)); + if (IS_ERR(args)) + return PTR_ERR(args); + + inode = file_inode(file); + + if (args->dirid == BTRFS_FIRST_FREE_OBJECTID && + btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) { + /* + * The subvolume does not exist under fd with which this is + * called + */ + kfree(args); + return -EACCES; + } + + ret = btrfs_search_path_in_tree_user(file_mnt_idmap(file), inode, args); + if (ret == 0 && copy_to_user(argp, args, sizeof(*args))) ret = -EFAULT; @@ -2064,56 +2017,375 @@ static noinline int btrfs_ioctl_ino_lookup(struct file *file, return ret; } +/* Get the subvolume information in BTRFS_ROOT_ITEM and BTRFS_ROOT_BACKREF */ +static int btrfs_ioctl_get_subvol_info(struct inode *inode, void __user *argp) +{ + struct btrfs_ioctl_get_subvol_info_args *subvol_info; + struct btrfs_fs_info *fs_info; + struct btrfs_root *root; + struct btrfs_path *path; + struct btrfs_key key; + struct btrfs_root_item *root_item; + struct btrfs_root_ref *rref; + struct extent_buffer *leaf; + unsigned long item_off; + unsigned long item_len; + int slot; + int ret = 0; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + subvol_info = kzalloc(sizeof(*subvol_info), GFP_KERNEL); + if (!subvol_info) { + btrfs_free_path(path); + return -ENOMEM; + } + + fs_info = BTRFS_I(inode)->root->fs_info; + + /* Get root_item of inode's subvolume */ + key.objectid = btrfs_root_id(BTRFS_I(inode)->root); + root = btrfs_get_fs_root(fs_info, key.objectid, true); + if (IS_ERR(root)) { + ret = PTR_ERR(root); + goto out_free; + } + root_item = &root->root_item; + + subvol_info->treeid = key.objectid; + + subvol_info->generation = btrfs_root_generation(root_item); + subvol_info->flags = btrfs_root_flags(root_item); + + memcpy(subvol_info->uuid, root_item->uuid, BTRFS_UUID_SIZE); + memcpy(subvol_info->parent_uuid, root_item->parent_uuid, + BTRFS_UUID_SIZE); + memcpy(subvol_info->received_uuid, root_item->received_uuid, + BTRFS_UUID_SIZE); + + subvol_info->ctransid = btrfs_root_ctransid(root_item); + subvol_info->ctime.sec = btrfs_stack_timespec_sec(&root_item->ctime); + subvol_info->ctime.nsec = btrfs_stack_timespec_nsec(&root_item->ctime); + + subvol_info->otransid = btrfs_root_otransid(root_item); + subvol_info->otime.sec = btrfs_stack_timespec_sec(&root_item->otime); + subvol_info->otime.nsec = btrfs_stack_timespec_nsec(&root_item->otime); + + subvol_info->stransid = btrfs_root_stransid(root_item); + subvol_info->stime.sec = btrfs_stack_timespec_sec(&root_item->stime); + subvol_info->stime.nsec = btrfs_stack_timespec_nsec(&root_item->stime); + + subvol_info->rtransid = btrfs_root_rtransid(root_item); + subvol_info->rtime.sec = btrfs_stack_timespec_sec(&root_item->rtime); + subvol_info->rtime.nsec = btrfs_stack_timespec_nsec(&root_item->rtime); + + if (key.objectid != BTRFS_FS_TREE_OBJECTID) { + /* Search root tree for ROOT_BACKREF of this subvolume */ + key.type = BTRFS_ROOT_BACKREF_KEY; + key.offset = 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(fs_info->tree_root, path); + if (ret < 0) { + goto out; + } else if (unlikely(ret > 0)) { + ret = -EUCLEAN; + goto out; + } + } + + leaf = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(leaf, &key, slot); + if (key.objectid == subvol_info->treeid && + key.type == BTRFS_ROOT_BACKREF_KEY) { + subvol_info->parent_id = key.offset; + + rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref); + subvol_info->dirid = btrfs_root_ref_dirid(leaf, rref); + + item_off = btrfs_item_ptr_offset(leaf, slot) + + sizeof(struct btrfs_root_ref); + item_len = btrfs_item_size(leaf, slot) + - sizeof(struct btrfs_root_ref); + read_extent_buffer(leaf, subvol_info->name, + item_off, item_len); + } else { + ret = -ENOENT; + goto out; + } + } + + btrfs_free_path(path); + path = NULL; + if (copy_to_user(argp, subvol_info, sizeof(*subvol_info))) + ret = -EFAULT; + +out: + btrfs_put_root(root); +out_free: + btrfs_free_path(path); + kfree(subvol_info); + return ret; +} + +/* + * Return ROOT_REF information of the subvolume containing this inode + * except the subvolume name. + */ +static int btrfs_ioctl_get_subvol_rootref(struct btrfs_root *root, + void __user *argp) +{ + struct btrfs_ioctl_get_subvol_rootref_args *rootrefs; + struct btrfs_root_ref *rref; + struct btrfs_path *path; + struct btrfs_key key; + struct extent_buffer *leaf; + u64 objectid; + int slot; + int ret; + u8 found; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + rootrefs = memdup_user(argp, sizeof(*rootrefs)); + if (IS_ERR(rootrefs)) { + btrfs_free_path(path); + return PTR_ERR(rootrefs); + } + + objectid = btrfs_root_id(root); + key.objectid = objectid; + key.type = BTRFS_ROOT_REF_KEY; + key.offset = rootrefs->min_treeid; + found = 0; + + root = root->fs_info->tree_root; + ret = btrfs_search_slot(NULL, 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); + if (ret < 0) { + goto out; + } else if (unlikely(ret > 0)) { + ret = -EUCLEAN; + goto out; + } + } + while (1) { + leaf = path->nodes[0]; + slot = path->slots[0]; + + btrfs_item_key_to_cpu(leaf, &key, slot); + if (key.objectid != objectid || key.type != BTRFS_ROOT_REF_KEY) { + ret = 0; + goto out; + } + + if (found == BTRFS_MAX_ROOTREF_BUFFER_NUM) { + ret = -EOVERFLOW; + goto out; + } + + rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref); + rootrefs->rootref[found].treeid = key.offset; + rootrefs->rootref[found].dirid = + btrfs_root_ref_dirid(leaf, rref); + found++; + + ret = btrfs_next_item(root, path); + if (ret < 0) { + goto out; + } else if (unlikely(ret > 0)) { + ret = -EUCLEAN; + goto out; + } + } + +out: + btrfs_free_path(path); + + if (!ret || ret == -EOVERFLOW) { + rootrefs->num_items = found; + /* update min_treeid for next search */ + if (found) + rootrefs->min_treeid = + rootrefs->rootref[found - 1].treeid + 1; + if (copy_to_user(argp, rootrefs, sizeof(*rootrefs))) + ret = -EFAULT; + } + + kfree(rootrefs); + + return ret; +} + static noinline int btrfs_ioctl_snap_destroy(struct file *file, - void __user *arg) + void __user *arg, + bool destroy_v2) { - struct dentry *parent = fdentry(file); + struct dentry *parent = file->f_path.dentry; struct dentry *dentry; - struct inode *dir = parent->d_inode; + struct inode *dir = d_inode(parent); + struct btrfs_fs_info *fs_info = inode_to_fs_info(dir); struct inode *inode; struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_root *dest = NULL; - struct btrfs_ioctl_vol_args *vol_args; - struct btrfs_trans_handle *trans; - struct btrfs_block_rsv block_rsv; - u64 qgroup_reserved; - int namelen; - int ret; - int err = 0; + struct btrfs_ioctl_vol_args *vol_args = NULL; + struct btrfs_ioctl_vol_args_v2 *vol_args2 = NULL; + struct mnt_idmap *idmap = file_mnt_idmap(file); + char *subvol_name, *subvol_name_ptr = NULL; + int ret = 0; + bool destroy_parent = false; - vol_args = memdup_user(arg, sizeof(*vol_args)); - if (IS_ERR(vol_args)) - return PTR_ERR(vol_args); + /* We don't support snapshots with extent tree v2 yet. */ + if (btrfs_fs_incompat(fs_info, EXTENT_TREE_V2)) { + btrfs_err(fs_info, + "extent tree v2 doesn't support snapshot deletion yet"); + return -EOPNOTSUPP; + } - 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 (destroy_v2) { + vol_args2 = memdup_user(arg, sizeof(*vol_args2)); + if (IS_ERR(vol_args2)) + return PTR_ERR(vol_args2); + + if (vol_args2->flags & ~BTRFS_SUBVOL_DELETE_ARGS_MASK) { + ret = -EOPNOTSUPP; + 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)) { + ret = btrfs_check_ioctl_vol_args2_subvol_name(vol_args2); + if (ret < 0) + goto out; + subvol_name = vol_args2->name; + + ret = mnt_want_write_file(file); + if (ret) + goto out; + } else { + struct inode *old_dir; + + if (vol_args2->subvolid < BTRFS_FIRST_FREE_OBJECTID) { + ret = -EINVAL; + goto out; + } + + ret = mnt_want_write_file(file); + if (ret) + goto out; + + dentry = btrfs_get_dentry(fs_info->sb, + BTRFS_FIRST_FREE_OBJECTID, + vol_args2->subvolid, 0); + if (IS_ERR(dentry)) { + ret = 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)) { + ret = PTR_ERR(parent); + goto out_drop_write; + } + old_dir = dir; + 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; + + /* + * On idmapped mounts, deletion via subvolid is + * restricted to subvolumes that are immediate + * ancestors of the inode referenced by the file + * descriptor in the ioctl. Otherwise the idmapping + * could potentially be abused to delete subvolumes + * anywhere in the filesystem the user wouldn't be able + * to delete without an idmapped mount. + */ + if (old_dir != dir && idmap != &nop_mnt_idmap) { + ret = -EOPNOTSUPP; + goto free_parent; + } + + subvol_name_ptr = btrfs_get_subvol_name_from_objectid( + fs_info, vol_args2->subvolid); + if (IS_ERR(subvol_name_ptr)) { + ret = PTR_ERR(subvol_name_ptr); + goto free_parent; + } + /* subvol_name_ptr is already nul terminated */ + 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); + + ret = btrfs_check_ioctl_vol_args_path(vol_args); + if (ret < 0) + goto out; + + subvol_name = vol_args->name; + + ret = mnt_want_write_file(file); + if (ret) + goto out; } - err = mnt_want_write_file(file); - if (err) - goto out; + if (strchr(subvol_name, '/') || + strcmp(subvol_name, "..") == 0) { + ret = -EINVAL; + goto free_subvol_name; + } - err = mutex_lock_killable_nested(&dir->i_mutex, I_MUTEX_PARENT); - if (err == -EINTR) - goto out; - dentry = lookup_one_len(vol_args->name, parent, namelen); - if (IS_ERR(dentry)) { - err = PTR_ERR(dentry); - goto out_unlock_dir; + if (!S_ISDIR(dir->i_mode)) { + ret = -ENOTDIR; + goto free_subvol_name; } - if (!dentry->d_inode) { - err = -ENOENT; - goto out_dput; + dentry = start_removing_killable(idmap, parent, &QSTR(subvol_name)); + if (IS_ERR(dentry)) { + ret = PTR_ERR(dentry); + goto out_end_removing; } - inode = dentry->d_inode; + inode = d_inode(dentry); dest = BTRFS_I(inode)->root; - if (!capable(CAP_SYS_ADMIN)){ + if (!capable(CAP_SYS_ADMIN)) { /* * Regular user. Only allow this with a special mount * option, when the user has write+exec access to the @@ -2127,9 +2399,9 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, * Users who want to delete empty subvols should try * rmdir(2). */ - err = -EPERM; - if (!btrfs_test_opt(root, USER_SUBVOL_RM_ALLOWED)) - goto out_dput; + ret = -EPERM; + if (!btrfs_test_opt(fs_info, USER_SUBVOL_RM_ALLOWED)) + goto out_end_removing; /* * Do not allow deletion if the parent dir is the same @@ -2138,120 +2410,51 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, * of the subvol, not a random directory contained * within it. */ - err = -EINVAL; + ret = -EINVAL; if (root == dest) - goto out_dput; + goto out_end_removing; - err = inode_permission(inode, MAY_WRITE | MAY_EXEC); - if (err) - goto out_dput; + ret = inode_permission(idmap, inode, MAY_WRITE | MAY_EXEC); + if (ret) + goto out_end_removing; } /* check if subvolume may be deleted by a user */ - err = btrfs_may_delete(dir, dentry, 1); - if (err) - goto out_dput; - - if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) { - err = -EINVAL; - goto out_dput; - } - - mutex_lock(&inode->i_mutex); - err = d_invalidate(dentry); - if (err) - goto out_unlock; - - down_write(&root->fs_info->subvol_sem); - - err = may_destroy_subvol(dest); - if (err) - goto out_up_write; - - btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP); - /* - * One for dir inode, two for dir entries, two for root - * ref/backref. - */ - err = btrfs_subvolume_reserve_metadata(root, &block_rsv, - 5, &qgroup_reserved); - if (err) - goto out_up_write; - - trans = btrfs_start_transaction(root, 0); - if (IS_ERR(trans)) { - err = PTR_ERR(trans); - goto out_release; - } - trans->block_rsv = &block_rsv; - trans->bytes_reserved = block_rsv.size; + ret = btrfs_may_delete(idmap, dir, dentry, 1); + if (ret) + goto out_end_removing; - ret = btrfs_unlink_subvol(trans, root, dir, - dest->root_key.objectid, - dentry->d_name.name, - dentry->d_name.len); - if (ret) { - err = ret; - btrfs_abort_transaction(trans, root, ret); - goto out_end_trans; + if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) { + ret = -EINVAL; + goto out_end_removing; } - btrfs_record_root_in_trans(trans, dest); - - memset(&dest->root_item.drop_progress, 0, - sizeof(dest->root_item.drop_progress)); - dest->root_item.drop_level = 0; - btrfs_set_root_refs(&dest->root_item, 0); - - if (!xchg(&dest->orphan_item_inserted, 1)) { - ret = btrfs_insert_orphan_item(trans, - root->fs_info->tree_root, - dest->root_key.objectid); - if (ret) { - btrfs_abort_transaction(trans, root, ret); - err = ret; - goto out_end_trans; - } - } -out_end_trans: - trans->block_rsv = NULL; - trans->bytes_reserved = 0; - ret = btrfs_end_transaction(trans, root); - if (ret && !err) - err = ret; - inode->i_flags |= S_DEAD; -out_release: - btrfs_subvolume_release_metadata(root, &block_rsv, qgroup_reserved); -out_up_write: - up_write(&root->fs_info->subvol_sem); -out_unlock: - mutex_unlock(&inode->i_mutex); - if (!err) { - shrink_dcache_sb(root->fs_info->sb); - btrfs_invalidate_inodes(dest); - d_delete(dentry); - - /* the last ref */ - if (dest->cache_inode) { - iput(dest->cache_inode); - dest->cache_inode = NULL; - } - } -out_dput: - dput(dentry); -out_unlock_dir: - mutex_unlock(&dir->i_mutex); + btrfs_inode_lock(BTRFS_I(inode), 0); + ret = btrfs_delete_subvolume(BTRFS_I(dir), dentry); + btrfs_inode_unlock(BTRFS_I(inode), 0); + if (!ret) + d_delete_notify(dir, dentry); + +out_end_removing: + end_removing(dentry); +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; + return ret; } static int btrfs_ioctl_defrag(struct file *file, void __user *argp) { struct inode *inode = file_inode(file); struct btrfs_root *root = BTRFS_I(inode)->root; - struct btrfs_ioctl_defrag_range_args *range; + struct btrfs_ioctl_defrag_range_args range = {0}; int ret; ret = mnt_want_write_file(file); @@ -2270,43 +2473,56 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp) goto out; } ret = btrfs_defrag_root(root); - if (ret) - goto out; - ret = btrfs_defrag_root(root->fs_info->extent_root); break; case S_IFREG: - if (!(file->f_mode & FMODE_WRITE)) { - ret = -EINVAL; + /* + * Note that this does not check the file descriptor for write + * access. This prevents defragmenting executables that are + * running and allows defrag on files open in read-only mode. + */ + if (!capable(CAP_SYS_ADMIN) && + inode_permission(&nop_mnt_idmap, inode, MAY_WRITE)) { + ret = -EPERM; goto out; } - range = kzalloc(sizeof(*range), GFP_KERNEL); - if (!range) { - ret = -ENOMEM; + /* + * Don't allow defrag on pre-content watched files, as it could + * populate the page cache with 0's via readahead. + */ + if (unlikely(FMODE_FSNOTIFY_HSM(file->f_mode))) { + ret = -EINVAL; goto out; } if (argp) { - if (copy_from_user(range, argp, - sizeof(*range))) { + if (copy_from_user(&range, argp, sizeof(range))) { ret = -EFAULT; - kfree(range); goto out; } - /* compression requires us to start the IO */ - if ((range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)) { - range->flags |= BTRFS_DEFRAG_RANGE_START_IO; - range->extent_thresh = (u32)-1; + if (range.flags & ~BTRFS_DEFRAG_RANGE_FLAGS_SUPP) { + ret = -EOPNOTSUPP; + goto out; + } + if ((range.flags & BTRFS_DEFRAG_RANGE_COMPRESS) && + (range.flags & BTRFS_DEFRAG_RANGE_NOCOMPRESS)) { + ret = -EINVAL; + goto out; + } + /* Compression or no-compression require to start the IO. */ + if ((range.flags & BTRFS_DEFRAG_RANGE_COMPRESS) || + (range.flags & BTRFS_DEFRAG_RANGE_NOCOMPRESS)) { + range.flags |= BTRFS_DEFRAG_RANGE_START_IO; + range.extent_thresh = (u32)-1; } } else { /* the rest are all set to zero by kzalloc */ - range->len = (u64)-1; + range.len = (u64)-1; } - ret = btrfs_defrag_file(file_inode(file), file, - range, 0, 0); + ret = btrfs_defrag_file(BTRFS_I(file_inode(file)), &file->f_ra, + &range, BTRFS_OLDEST_GENERATION, 0); if (ret > 0) ret = 0; - kfree(range); break; default: ret = -EINVAL; @@ -2316,99 +2532,231 @@ out: return ret; } -static long btrfs_ioctl_add_dev(struct btrfs_root *root, void __user *arg) +static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg) { struct btrfs_ioctl_vol_args *vol_args; + bool restore_op = false; int ret; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running, - 1)) { - pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n"); + if (btrfs_fs_incompat(fs_info, EXTENT_TREE_V2)) { + btrfs_err(fs_info, "device add not supported on extent tree v2 yet"); + return -EINVAL; + } + + if (fs_info->fs_devices->temp_fsid) { + btrfs_err(fs_info, + "device add not supported on cloned temp-fsid mount"); return -EINVAL; } - mutex_lock(&root->fs_info->volume_mutex); + if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_ADD)) { + if (!btrfs_exclop_start_try_lock(fs_info, BTRFS_EXCLOP_DEV_ADD)) + return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS; + + /* + * We can do the device add because we have a paused balanced, + * change the exclusive op type and remember we should bring + * back the paused balance + */ + fs_info->exclusive_operation = BTRFS_EXCLOP_DEV_ADD; + btrfs_exclop_start_unlock(fs_info); + restore_op = true; + } + vol_args = memdup_user(arg, sizeof(*vol_args)); if (IS_ERR(vol_args)) { ret = PTR_ERR(vol_args); goto out; } - vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; - ret = btrfs_init_new_device(root, vol_args->name); + ret = btrfs_check_ioctl_vol_args_path(vol_args); + if (ret < 0) + goto out_free; + + ret = btrfs_init_new_device(fs_info, vol_args->name); + if (!ret) + btrfs_info(fs_info, "disk added %s", vol_args->name); + +out_free: kfree(vol_args); out: - mutex_unlock(&root->fs_info->volume_mutex); - atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0); + if (restore_op) + btrfs_exclop_balance(fs_info, BTRFS_EXCLOP_BALANCE_PAUSED); + else + btrfs_exclop_finish(fs_info); return ret; } -static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg) +static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; - struct btrfs_ioctl_vol_args *vol_args; + BTRFS_DEV_LOOKUP_ARGS(args); + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); + struct btrfs_ioctl_vol_args_v2 *vol_args; + struct file *bdev_file = NULL; int ret; + bool cancel = false; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - ret = mnt_want_write_file(file); - if (ret) - return ret; - vol_args = memdup_user(arg, sizeof(*vol_args)); - if (IS_ERR(vol_args)) { - ret = PTR_ERR(vol_args); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); + + if (vol_args->flags & ~BTRFS_DEVICE_REMOVE_ARGS_MASK) { + ret = -EOPNOTSUPP; goto out; } - vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; - - if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running, - 1)) { - ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS; + ret = btrfs_check_ioctl_vol_args2_subvol_name(vol_args); + if (ret < 0) goto out; + + if (vol_args->flags & BTRFS_DEVICE_SPEC_BY_ID) { + args.devid = vol_args->devid; + } else if (!strcmp("cancel", vol_args->name)) { + cancel = true; + } else { + ret = btrfs_get_dev_args_from_path(fs_info, &args, vol_args->name); + if (ret) + goto out; } - mutex_lock(&root->fs_info->volume_mutex); - ret = btrfs_rm_device(root, vol_args->name); - mutex_unlock(&root->fs_info->volume_mutex); - atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0); + ret = mnt_want_write_file(file); + if (ret) + goto out; + + ret = exclop_start_or_cancel_reloc(fs_info, BTRFS_EXCLOP_DEV_REMOVE, + cancel); + if (ret) + goto err_drop; + /* Exclusive operation is now claimed */ + ret = btrfs_rm_device(fs_info, &args, &bdev_file); + + btrfs_exclop_finish(fs_info); + + if (!ret) { + if (vol_args->flags & BTRFS_DEVICE_SPEC_BY_ID) + btrfs_info(fs_info, "device deleted: id %llu", + vol_args->devid); + else + btrfs_info(fs_info, "device deleted: %s", + vol_args->name); + } +err_drop: + mnt_drop_write_file(file); + if (bdev_file) + bdev_fput(bdev_file); out: + btrfs_put_dev_args_from_path(&args); kfree(vol_args); + return ret; +} + +static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg) +{ + BTRFS_DEV_LOOKUP_ARGS(args); + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); + struct btrfs_ioctl_vol_args *vol_args; + struct file *bdev_file = NULL; + int ret; + bool cancel = false; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); + + ret = btrfs_check_ioctl_vol_args_path(vol_args); + if (ret < 0) + goto out_free; + + if (!strcmp("cancel", vol_args->name)) { + cancel = true; + } else { + ret = btrfs_get_dev_args_from_path(fs_info, &args, vol_args->name); + if (ret) + goto out; + } + + ret = mnt_want_write_file(file); + if (ret) + goto out; + + ret = exclop_start_or_cancel_reloc(fs_info, BTRFS_EXCLOP_DEV_REMOVE, + cancel); + if (ret == 0) { + ret = btrfs_rm_device(fs_info, &args, &bdev_file); + if (!ret) + btrfs_info(fs_info, "disk deleted %s", vol_args->name); + btrfs_exclop_finish(fs_info); + } + mnt_drop_write_file(file); + if (bdev_file) + bdev_fput(bdev_file); +out: + btrfs_put_dev_args_from_path(&args); +out_free: + kfree(vol_args); return ret; } -static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg) +static long btrfs_ioctl_fs_info(const struct btrfs_fs_info *fs_info, + void __user *arg) { struct btrfs_ioctl_fs_info_args *fi_args; struct btrfs_device *device; - struct btrfs_device *next; - struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices; + struct btrfs_fs_devices *fs_devices = fs_info->fs_devices; + u64 flags_in; int ret = 0; - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; + fi_args = memdup_user(arg, sizeof(*fi_args)); + if (IS_ERR(fi_args)) + return PTR_ERR(fi_args); - fi_args = kzalloc(sizeof(*fi_args), GFP_KERNEL); - if (!fi_args) - return -ENOMEM; + flags_in = fi_args->flags; + memset(fi_args, 0, sizeof(*fi_args)); + rcu_read_lock(); fi_args->num_devices = fs_devices->num_devices; - memcpy(&fi_args->fsid, root->fs_info->fsid, sizeof(fi_args->fsid)); - mutex_lock(&fs_devices->device_list_mutex); - list_for_each_entry_safe(device, next, &fs_devices->devices, dev_list) { + list_for_each_entry_rcu(device, &fs_devices->devices, dev_list) { if (device->devid > fi_args->max_id) fi_args->max_id = device->devid; } - mutex_unlock(&fs_devices->device_list_mutex); + rcu_read_unlock(); + + memcpy(&fi_args->fsid, fs_devices->fsid, sizeof(fi_args->fsid)); + fi_args->nodesize = fs_info->nodesize; + fi_args->sectorsize = fs_info->sectorsize; + fi_args->clone_alignment = fs_info->sectorsize; + + if (flags_in & BTRFS_FS_INFO_FLAG_CSUM_INFO) { + fi_args->csum_type = btrfs_super_csum_type(fs_info->super_copy); + fi_args->csum_size = btrfs_super_csum_size(fs_info->super_copy); + fi_args->flags |= BTRFS_FS_INFO_FLAG_CSUM_INFO; + } + + if (flags_in & BTRFS_FS_INFO_FLAG_GENERATION) { + fi_args->generation = btrfs_get_fs_generation(fs_info); + fi_args->flags |= BTRFS_FS_INFO_FLAG_GENERATION; + } + + if (flags_in & BTRFS_FS_INFO_FLAG_METADATA_UUID) { + memcpy(&fi_args->metadata_uuid, fs_devices->metadata_uuid, + sizeof(fi_args->metadata_uuid)); + fi_args->flags |= BTRFS_FS_INFO_FLAG_METADATA_UUID; + } if (copy_to_user(arg, fi_args, sizeof(*fi_args))) ret = -EFAULT; @@ -2417,51 +2765,41 @@ static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg) return ret; } -static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg) +static long btrfs_ioctl_dev_info(const struct btrfs_fs_info *fs_info, + void __user *arg) { + BTRFS_DEV_LOOKUP_ARGS(args); struct btrfs_ioctl_dev_info_args *di_args; struct btrfs_device *dev; - struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices; int ret = 0; - char *s_uuid = NULL; - char empty_uuid[BTRFS_UUID_SIZE] = {0}; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; di_args = memdup_user(arg, sizeof(*di_args)); if (IS_ERR(di_args)) return PTR_ERR(di_args); - if (memcmp(empty_uuid, di_args->uuid, BTRFS_UUID_SIZE) != 0) - s_uuid = di_args->uuid; - - mutex_lock(&fs_devices->device_list_mutex); - dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, NULL); + args.devid = di_args->devid; + if (!btrfs_is_empty_uuid(di_args->uuid)) + args.uuid = di_args->uuid; + rcu_read_lock(); + dev = btrfs_find_device(fs_info->fs_devices, &args); if (!dev) { ret = -ENODEV; goto out; } di_args->devid = dev->devid; - di_args->bytes_used = dev->bytes_used; - di_args->total_bytes = dev->total_bytes; + di_args->bytes_used = btrfs_device_get_bytes_used(dev); + di_args->total_bytes = btrfs_device_get_total_bytes(dev); memcpy(di_args->uuid, dev->uuid, sizeof(di_args->uuid)); - if (dev->name) { - struct rcu_string *name; - - rcu_read_lock(); - name = rcu_dereference(dev->name); - strncpy(di_args->path, name->str, sizeof(di_args->path)); - rcu_read_unlock(); - di_args->path[sizeof(di_args->path) - 1] = 0; - } else { + memcpy(di_args->fsid, dev->fs_devices->fsid, BTRFS_UUID_SIZE); + if (dev->name) + strscpy(di_args->path, btrfs_dev_name(dev), sizeof(di_args->path)); + else di_args->path[0] = '\0'; - } out: - mutex_unlock(&fs_devices->device_list_mutex); + rcu_read_unlock(); if (ret == 0 && copy_to_user(arg, di_args, sizeof(*di_args))) ret = -EFAULT; @@ -2469,477 +2807,17 @@ out: return ret; } -static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, - u64 off, u64 olen, u64 destoff) -{ - struct inode *inode = file_inode(file); - struct btrfs_root *root = BTRFS_I(inode)->root; - struct fd src_file; - struct inode *src; - struct btrfs_trans_handle *trans; - struct btrfs_path *path; - struct extent_buffer *leaf; - char *buf; - struct btrfs_key key; - u32 nritems; - int slot; - int ret; - u64 len = olen; - u64 bs = root->fs_info->sb->s_blocksize; - int same_inode = 0; - - /* - * 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. - * - allow ranges within the same file to be cloned (provided - * they don't overlap)? - */ - - /* the destination must be opened for writing */ - if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND)) - return -EINVAL; - - if (btrfs_root_readonly(root)) - return -EROFS; - - ret = mnt_want_write_file(file); - if (ret) - return ret; - - src_file = fdget(srcfd); - if (!src_file.file) { - ret = -EBADF; - goto out_drop_write; - } - - ret = -EXDEV; - if (src_file.file->f_path.mnt != file->f_path.mnt) - goto out_fput; - - src = file_inode(src_file.file); - - ret = -EINVAL; - if (src == inode) - same_inode = 1; - - /* the src must be open for reading */ - if (!(src_file.file->f_mode & FMODE_READ)) - goto out_fput; - - /* don't make the dst file partly checksummed */ - if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) != - (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) - goto out_fput; - - ret = -EISDIR; - if (S_ISDIR(src->i_mode) || S_ISDIR(inode->i_mode)) - goto out_fput; - - ret = -EXDEV; - if (src->i_sb != inode->i_sb) - goto out_fput; - - ret = -ENOMEM; - buf = vmalloc(btrfs_level_size(root, 0)); - if (!buf) - goto out_fput; - - path = btrfs_alloc_path(); - if (!path) { - vfree(buf); - goto out_fput; - } - path->reada = 2; - - if (!same_inode) { - if (inode < src) { - mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&src->i_mutex, I_MUTEX_CHILD); - } else { - mutex_lock_nested(&src->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD); - } - } else { - mutex_lock(&src->i_mutex); - } - - /* determine range to clone */ - ret = -EINVAL; - if (off + len > src->i_size || off + len < off) - goto out_unlock; - if (len == 0) - olen = len = src->i_size - off; - /* if we extend to eof, continue to block boundary */ - if (off + len == src->i_size) - len = ALIGN(src->i_size, bs) - off; - - /* verify the end result is block aligned */ - if (!IS_ALIGNED(off, bs) || !IS_ALIGNED(off + len, bs) || - !IS_ALIGNED(destoff, bs)) - goto out_unlock; - - /* verify if ranges are overlapped within the same file */ - if (same_inode) { - if (destoff + len > off && destoff < off + len) - goto out_unlock; - } - - if (destoff > inode->i_size) { - ret = btrfs_cont_expand(inode, inode->i_size, destoff); - if (ret) - goto out_unlock; - } - - /* truncate page cache pages from target inode range */ - truncate_inode_pages_range(&inode->i_data, destoff, - PAGE_CACHE_ALIGN(destoff + len) - 1); - - /* do any pending delalloc/csum calc on src, one way or - another, and lock file content */ - while (1) { - struct btrfs_ordered_extent *ordered; - lock_extent(&BTRFS_I(src)->io_tree, off, off + len - 1); - ordered = btrfs_lookup_first_ordered_extent(src, off + len - 1); - if (!ordered && - !test_range_bit(&BTRFS_I(src)->io_tree, off, off + len - 1, - EXTENT_DELALLOC, 0, NULL)) - break; - unlock_extent(&BTRFS_I(src)->io_tree, off, off + len - 1); - if (ordered) - btrfs_put_ordered_extent(ordered); - btrfs_wait_ordered_range(src, off, len); - } - - /* clone data */ - key.objectid = btrfs_ino(src); - key.type = BTRFS_EXTENT_DATA_KEY; - key.offset = 0; - - while (1) { - /* - * note the key will change type as we walk through the - * tree. - */ - ret = btrfs_search_slot(NULL, BTRFS_I(src)->root, &key, path, - 0, 0); - if (ret < 0) - goto out; - - nritems = btrfs_header_nritems(path->nodes[0]); - 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 (btrfs_key_type(&key) > BTRFS_EXTENT_DATA_KEY || - key.objectid != btrfs_ino(src)) - break; - - if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY) { - 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 endoff; - - size = btrfs_item_size_nr(leaf, slot); - read_extent_buffer(leaf, buf, - btrfs_item_ptr_offset(leaf, slot), - size); - - 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); - } - btrfs_release_path(path); - - if (key.offset + datal <= off || - key.offset >= off + len - 1) - goto next; - - memcpy(&new_key, &key, sizeof(new_key)); - new_key.objectid = btrfs_ino(inode); - if (off <= key.offset) - new_key.offset = key.offset + destoff - off; - else - new_key.offset = destoff; - - /* - * 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; - } - - if (type == BTRFS_FILE_EXTENT_REG || - type == BTRFS_FILE_EXTENT_PREALLOC) { - /* - * a | --- range to clone ---| b - * | ------------- extent ------------- | - */ - - /* substract range b */ - if (key.offset + datal > off + len) - datal = off + len - key.offset; - - /* substract range a */ - if (off > key.offset) { - datao += off - key.offset; - datal -= off - key.offset; - } - - ret = btrfs_drop_extents(trans, root, inode, - new_key.offset, - new_key.offset + datal, - 1); - if (ret) { - btrfs_abort_transaction(trans, root, - ret); - btrfs_end_transaction(trans, root); - goto out; - } - - ret = btrfs_insert_empty_item(trans, root, path, - &new_key, size); - if (ret) { - btrfs_abort_transaction(trans, root, - ret); - btrfs_end_transaction(trans, root); - goto out; - } - - leaf = path->nodes[0]; - slot = path->slots[0]; - write_extent_buffer(leaf, buf, - btrfs_item_ptr_offset(leaf, slot), - size); - - extent = btrfs_item_ptr(leaf, slot, - struct btrfs_file_extent_item); - - /* disko == 0 means it's a hole */ - if (!disko) - datao = 0; - - btrfs_set_file_extent_offset(leaf, extent, - datao); - btrfs_set_file_extent_num_bytes(leaf, extent, - datal); - if (disko) { - inode_add_bytes(inode, datal); - ret = btrfs_inc_extent_ref(trans, root, - disko, diskl, 0, - root->root_key.objectid, - btrfs_ino(inode), - new_key.offset - datao, - 0); - if (ret) { - btrfs_abort_transaction(trans, - root, - ret); - btrfs_end_transaction(trans, - root); - 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; - btrfs_end_transaction(trans, root); - goto out; - } - size -= skip + trim; - datal -= skip + trim; - - ret = btrfs_drop_extents(trans, root, inode, - new_key.offset, - new_key.offset + datal, - 1); - if (ret) { - btrfs_abort_transaction(trans, root, - ret); - btrfs_end_transaction(trans, root); - goto out; - } - - ret = btrfs_insert_empty_item(trans, root, path, - &new_key, size); - if (ret) { - btrfs_abort_transaction(trans, root, - ret); - btrfs_end_transaction(trans, root); - goto out; - } - - if (skip) { - u32 start = - btrfs_file_extent_calc_inline_size(0); - memmove(buf+start, buf+start+skip, - datal); - } - - leaf = path->nodes[0]; - slot = path->slots[0]; - write_extent_buffer(leaf, buf, - btrfs_item_ptr_offset(leaf, slot), - size); - inode_add_bytes(inode, datal); - } - - btrfs_mark_buffer_dirty(leaf); - btrfs_release_path(path); - - inode_inc_iversion(inode); - inode->i_mtime = inode->i_ctime = CURRENT_TIME; - - /* - * we round up to the block size at eof when - * determining which extents to clone above, - * but shouldn't round up the file size - */ - endoff = new_key.offset + datal; - if (endoff > destoff+olen) - endoff = destoff+olen; - if (endoff > inode->i_size) - btrfs_i_size_write(inode, endoff); - - ret = btrfs_update_inode(trans, root, inode); - if (ret) { - btrfs_abort_transaction(trans, root, ret); - btrfs_end_transaction(trans, root); - goto out; - } - ret = btrfs_end_transaction(trans, root); - } -next: - btrfs_release_path(path); - key.offset++; - } - ret = 0; -out: - btrfs_release_path(path); - unlock_extent(&BTRFS_I(src)->io_tree, off, off + len - 1); -out_unlock: - mutex_unlock(&src->i_mutex); - if (!same_inode) - mutex_unlock(&inode->i_mutex); - vfree(buf); - btrfs_free_path(path); -out_fput: - fdput(src_file); -out_drop_write: - mnt_drop_write_file(file); - return ret; -} - -static long btrfs_ioctl_clone_range(struct file *file, void __user *argp) -{ - struct btrfs_ioctl_clone_range_args args; - - if (copy_from_user(&args, argp, sizeof(args))) - return -EFAULT; - return btrfs_ioctl_clone(file, args.src_fd, args.src_offset, - args.src_length, args.dest_offset); -} - -/* - * there are many ways the trans_start and trans_end ioctls can lead - * to deadlocks. They should only be used by applications that - * basically own the machine, and have a very in depth understanding - * of all the possible deadlocks and enospc problems. - */ -static long btrfs_ioctl_trans_start(struct file *file) -{ - struct inode *inode = file_inode(file); - struct btrfs_root *root = BTRFS_I(inode)->root; - struct btrfs_trans_handle *trans; - int ret; - - ret = -EPERM; - if (!capable(CAP_SYS_ADMIN)) - goto out; - - ret = -EINPROGRESS; - if (file->private_data) - goto out; - - ret = -EROFS; - if (btrfs_root_readonly(root)) - goto out; - - ret = mnt_want_write_file(file); - if (ret) - goto out; - - atomic_inc(&root->fs_info->open_ioctl_trans); - - ret = -ENOMEM; - trans = btrfs_start_ioctl_transaction(root); - if (IS_ERR(trans)) - goto out_drop; - - file->private_data = trans; - return 0; - -out_drop: - atomic_dec(&root->fs_info->open_ioctl_trans); - mnt_drop_write_file(file); -out: - return ret; -} - static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) { struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_root *new_root; struct btrfs_dir_item *di; struct btrfs_trans_handle *trans; - struct btrfs_path *path; - struct btrfs_key location; + struct btrfs_path *path = NULL; struct btrfs_disk_key disk_key; + struct fscrypt_str name = FSTR_INIT("default", 7); u64 objectid = 0; u64 dir_id; int ret; @@ -2957,84 +2835,86 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) } if (!objectid) - objectid = root->root_key.objectid; + objectid = BTRFS_FS_TREE_OBJECTID; - location.objectid = objectid; - location.type = BTRFS_ROOT_ITEM_KEY; - location.offset = (u64)-1; - - new_root = btrfs_read_fs_root_no_name(root->fs_info, &location); + new_root = btrfs_get_fs_root(fs_info, objectid, true); if (IS_ERR(new_root)) { ret = PTR_ERR(new_root); goto out; } + if (!btrfs_is_fstree(btrfs_root_id(new_root))) { + ret = -ENOENT; + 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(root->fs_info->super_copy); - di = btrfs_lookup_dir_item(trans, root->fs_info->tree_root, path, - dir_id, "default", 7, 1); + dir_id = btrfs_super_root_dir(fs_info->super_copy); + di = btrfs_lookup_dir_item(trans, fs_info->tree_root, path, + dir_id, &name, 1); if (IS_ERR_OR_NULL(di)) { - btrfs_free_path(path); - btrfs_end_transaction(trans, root); - printk(KERN_ERR "Umm, you don't have the default dir item, " - "this isn't going to work\n"); + 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(root->fs_info, DEFAULT_SUBVOL); - btrfs_end_transaction(trans, root); + 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; } -void btrfs_get_block_group_info(struct list_head *groups_list, - struct btrfs_ioctl_space_info *space) +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; } } -static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) +static long btrfs_ioctl_space_info(struct btrfs_fs_info *fs_info, + void __user *arg) { - struct btrfs_ioctl_space_args space_args; + struct btrfs_ioctl_space_args space_args = { 0 }; struct btrfs_ioctl_space_info space; struct btrfs_ioctl_space_info *dest; - struct btrfs_ioctl_space_info *dest_orig; + struct btrfs_ioctl_space_info AUTO_KFREE(dest_orig); struct btrfs_ioctl_space_info __user *user_dest; struct btrfs_space_info *info; - u64 types[] = {BTRFS_BLOCK_GROUP_DATA, - BTRFS_BLOCK_GROUP_SYSTEM, - BTRFS_BLOCK_GROUP_METADATA, - BTRFS_BLOCK_GROUP_DATA | BTRFS_BLOCK_GROUP_METADATA}; + static const u64 types[] = { + BTRFS_BLOCK_GROUP_DATA, + BTRFS_BLOCK_GROUP_SYSTEM, + BTRFS_BLOCK_GROUP_METADATA, + BTRFS_BLOCK_GROUP_DATA | BTRFS_BLOCK_GROUP_METADATA + }; int num_types = 4; int alloc_size; int ret = 0; @@ -3050,15 +2930,12 @@ static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) struct btrfs_space_info *tmp; info = NULL; - rcu_read_lock(); - list_for_each_entry_rcu(tmp, &root->fs_info->space_info, - list) { + list_for_each_entry(tmp, &fs_info->space_info, list) { if (tmp->flags == types[i]) { info = tmp; break; } } - rcu_read_unlock(); if (!info) continue; @@ -3071,6 +2948,11 @@ static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) up_read(&info->groups_sem); } + /* + * Global block reserve, exported as a space_info + */ + slot_count++; + /* space_slots == 0 means they are asking for a count */ if (space_args.space_slots == 0) { space_args.total_spaces = slot_count; @@ -3084,11 +2966,11 @@ static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) /* we generally have at most 6 or so space infos, one for each raid * level. So, a whole page should be more than enough for everyone */ - if (alloc_size > PAGE_CACHE_SIZE) + if (alloc_size > PAGE_SIZE) return -ENOMEM; space_args.total_spaces = 0; - dest = kmalloc(alloc_size, GFP_NOFS); + dest = kmalloc(alloc_size, GFP_KERNEL); if (!dest) return -ENOMEM; dest_orig = dest; @@ -3101,23 +2983,20 @@ static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) break; info = NULL; - rcu_read_lock(); - list_for_each_entry_rcu(tmp, &root->fs_info->space_info, - list) { + list_for_each_entry(tmp, &fs_info->space_info, list) { if (tmp->flags == types[i]) { info = tmp; break; } } - rcu_read_unlock(); if (!info) continue; down_read(&info->groups_sem); for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) { if (!list_empty(&info->block_groups[c])) { - btrfs_get_block_group_info( - &info->block_groups[c], &space); + get_block_group_info(&info->block_groups[c], + &space); memcpy(dest, &space, sizeof(space)); dest++; space_args.total_spaces++; @@ -3129,13 +3008,27 @@ static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) up_read(&info->groups_sem); } + /* + * Add global block reserve + */ + if (slot_count) { + struct btrfs_block_rsv *block_rsv = &fs_info->global_block_rsv; + + spin_lock(&block_rsv->lock); + space.total_bytes = block_rsv->size; + space.used_bytes = block_rsv->size - block_rsv->reserved; + spin_unlock(&block_rsv->lock); + space.flags = BTRFS_SPACE_INFO_GLOBAL_RSV; + memcpy(dest, &space, sizeof(space)); + space_args.total_spaces++; + } + user_dest = (struct btrfs_ioctl_space_info __user *) (arg + sizeof(struct btrfs_ioctl_space_args)); if (copy_to_user(user_dest, dest_orig, alloc_size)) - ret = -EFAULT; + return -EFAULT; - kfree(dest_orig); out: if (ret == 0 && copy_to_user(arg, &space_args, sizeof(space_args))) ret = -EFAULT; @@ -3143,37 +3036,18 @@ out: return ret; } -/* - * there are many ways the trans_start and trans_end ioctls can lead - * to deadlocks. They should only be used by applications that - * basically own the machine, and have a very in depth understanding - * of all the possible deadlocks and enospc problems. - */ -long btrfs_ioctl_trans_end(struct file *file) -{ - struct inode *inode = file_inode(file); - struct btrfs_root *root = BTRFS_I(inode)->root; - struct btrfs_trans_handle *trans; - - trans = file->private_data; - if (!trans) - return -EINVAL; - file->private_data = NULL; - - btrfs_end_transaction(trans, root); - - atomic_dec(&root->fs_info->open_ioctl_trans); - - mnt_drop_write_file(file); - return 0; -} - static noinline long btrfs_ioctl_start_sync(struct btrfs_root *root, void __user *argp) { struct btrfs_trans_handle *trans; u64 transid; - int ret; + + /* + * Start orphan cleanup here for the given root in case it hasn't been + * started already by other means. Errors are handled in the other + * functions during transaction commit. + */ + btrfs_orphan_cleanup(root); trans = btrfs_attach_transaction_barrier(root); if (IS_ERR(trans)) { @@ -3181,15 +3055,11 @@ static noinline long btrfs_ioctl_start_sync(struct btrfs_root *root, return PTR_ERR(trans); /* No running transaction, don't bother */ - transid = root->fs_info->last_trans_committed; + transid = btrfs_get_last_trans_committed(root->fs_info); goto out; } transid = trans->transid; - ret = btrfs_commit_transaction_async(trans, root, 0); - if (ret) { - btrfs_end_transaction(trans, root); - return ret; - } + btrfs_commit_transaction_async(trans); out: if (argp) if (copy_to_user(argp, &transid, sizeof(transid))) @@ -3197,43 +3067,64 @@ out: return 0; } -static noinline long btrfs_ioctl_wait_sync(struct btrfs_root *root, +static noinline long btrfs_ioctl_wait_sync(struct btrfs_fs_info *fs_info, void __user *argp) { - u64 transid; + /* By default wait for the current transaction. */ + u64 transid = 0; - if (argp) { + if (argp) if (copy_from_user(&transid, argp, sizeof(transid))) return -EFAULT; - } else { - transid = 0; /* current trans */ - } - return btrfs_wait_for_commit(root, transid); + + return btrfs_wait_for_commit(fs_info, transid); } static long btrfs_ioctl_scrub(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; + struct btrfs_fs_info *fs_info = inode_to_fs_info(file_inode(file)); struct btrfs_ioctl_scrub_args *sa; int ret; if (!capable(CAP_SYS_ADMIN)) return -EPERM; + if (btrfs_fs_incompat(fs_info, EXTENT_TREE_V2)) { + btrfs_err(fs_info, "scrub: extent tree v2 not yet supported"); + return -EINVAL; + } + sa = memdup_user(arg, sizeof(*sa)); if (IS_ERR(sa)) return PTR_ERR(sa); + if (sa->flags & ~BTRFS_SCRUB_SUPPORTED_FLAGS) { + ret = -EOPNOTSUPP; + goto out; + } + if (!(sa->flags & BTRFS_SCRUB_READONLY)) { ret = mnt_want_write_file(file); if (ret) goto out; } - ret = btrfs_scrub_dev(root->fs_info, sa->devid, sa->start, sa->end, + ret = btrfs_scrub_dev(fs_info, sa->devid, sa->start, sa->end, &sa->progress, sa->flags & BTRFS_SCRUB_READONLY, 0); + /* + * 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; @@ -3244,15 +3135,15 @@ out: return ret; } -static long btrfs_ioctl_scrub_cancel(struct btrfs_root *root, void __user *arg) +static long btrfs_ioctl_scrub_cancel(struct btrfs_fs_info *fs_info) { if (!capable(CAP_SYS_ADMIN)) return -EPERM; - return btrfs_scrub_cancel(root->fs_info); + return btrfs_scrub_cancel(fs_info); } -static long btrfs_ioctl_scrub_progress(struct btrfs_root *root, +static long btrfs_ioctl_scrub_progress(struct btrfs_fs_info *fs_info, void __user *arg) { struct btrfs_ioctl_scrub_args *sa; @@ -3265,16 +3156,16 @@ static long btrfs_ioctl_scrub_progress(struct btrfs_root *root, if (IS_ERR(sa)) return PTR_ERR(sa); - ret = btrfs_scrub_progress(root, sa->devid, &sa->progress); + ret = btrfs_scrub_progress(fs_info, sa->devid, &sa->progress); - if (copy_to_user(arg, sa, sizeof(*sa))) + if (ret == 0 && copy_to_user(arg, sa, sizeof(*sa))) ret = -EFAULT; kfree(sa); return ret; } -static long btrfs_ioctl_get_dev_stats(struct btrfs_root *root, +static long btrfs_ioctl_get_dev_stats(struct btrfs_fs_info *fs_info, void __user *arg) { struct btrfs_ioctl_get_dev_stats *sa; @@ -3289,16 +3180,17 @@ static long btrfs_ioctl_get_dev_stats(struct btrfs_root *root, return -EPERM; } - ret = btrfs_get_dev_stats(root, sa); + ret = btrfs_get_dev_stats(fs_info, sa); - if (copy_to_user(arg, sa, sizeof(*sa))) + if (ret == 0 && copy_to_user(arg, sa, sizeof(*sa))) ret = -EFAULT; kfree(sa); return ret; } -static long btrfs_ioctl_dev_replace(struct btrfs_root *root, void __user *arg) +static long btrfs_ioctl_dev_replace(struct btrfs_fs_info *fs_info, + void __user *arg) { struct btrfs_ioctl_dev_replace_args *p; int ret; @@ -3306,39 +3198,44 @@ static long btrfs_ioctl_dev_replace(struct btrfs_root *root, void __user *arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; + if (btrfs_fs_incompat(fs_info, EXTENT_TREE_V2)) { + btrfs_err(fs_info, "device replace not supported on extent tree v2 yet"); + return -EINVAL; + } + p = memdup_user(arg, sizeof(*p)); if (IS_ERR(p)) return PTR_ERR(p); switch (p->cmd) { case BTRFS_IOCTL_DEV_REPLACE_CMD_START: - if (atomic_xchg( - &root->fs_info->mutually_exclusive_operation_running, - 1)) { - pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n"); - ret = -EINPROGRESS; + if (sb_rdonly(fs_info->sb)) { + ret = -EROFS; + goto out; + } + if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_REPLACE)) { + ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS; } else { - ret = btrfs_dev_replace_start(root, p); - atomic_set( - &root->fs_info->mutually_exclusive_operation_running, - 0); + ret = btrfs_dev_replace_by_ioctl(fs_info, p); + btrfs_exclop_finish(fs_info); } break; case BTRFS_IOCTL_DEV_REPLACE_CMD_STATUS: - btrfs_dev_replace_status(root->fs_info, p); + btrfs_dev_replace_status(fs_info, p); ret = 0; break; case BTRFS_IOCTL_DEV_REPLACE_CMD_CANCEL: - ret = btrfs_dev_replace_cancel(root->fs_info, p); + p->result = btrfs_dev_replace_cancel(fs_info); + ret = 0; break; default: ret = -EINVAL; break; } - if (copy_to_user(arg, p, sizeof(*p))) + if ((ret == 0 || ret == -ECANCELED) && copy_to_user(arg, p, sizeof(*p))) ret = -EFAULT; - +out: kfree(p); return ret; } @@ -3350,7 +3247,7 @@ static long btrfs_ioctl_ino_to_path(struct btrfs_root *root, void __user *arg) u64 rel_ptr; int size; struct btrfs_ioctl_ino_path_args *ipa = NULL; - struct inode_fs_paths *ipath = NULL; + struct inode_fs_paths *ipath __free(inode_fs_paths) = NULL; struct btrfs_path *path; if (!capable(CAP_DAC_READ_SEARCH)) @@ -3387,8 +3284,10 @@ static long btrfs_ioctl_ino_to_path(struct btrfs_root *root, void __user *arg) ipath->fspath->val[i] = rel_ptr; } - ret = copy_to_user((void *)(unsigned long)ipa->fspath, - (void *)(unsigned long)ipath->fspath, size); + btrfs_free_path(path); + path = NULL; + ret = copy_to_user((void __user *)(unsigned long)ipa->fspath, + ipath->fspath, size); if (ret) { ret = -EFAULT; goto out; @@ -3396,93 +3295,78 @@ static long btrfs_ioctl_ino_to_path(struct btrfs_root *root, void __user *arg) out: btrfs_free_path(path); - free_ipath(ipath); kfree(ipa); return ret; } -static int build_ino_list(u64 inum, u64 offset, u64 root, void *ctx) -{ - struct btrfs_data_container *inodes = ctx; - const size_t c = 3 * sizeof(u64); - - if (inodes->bytes_left >= c) { - inodes->bytes_left -= c; - inodes->val[inodes->elem_cnt] = inum; - inodes->val[inodes->elem_cnt + 1] = offset; - inodes->val[inodes->elem_cnt + 2] = root; - inodes->elem_cnt += 3; - } else { - inodes->bytes_missing += c - inodes->bytes_left; - inodes->bytes_left = 0; - inodes->elem_missed += 3; - } - - return 0; -} - -static long btrfs_ioctl_logical_to_ino(struct btrfs_root *root, - void __user *arg) +static long btrfs_ioctl_logical_to_ino(struct btrfs_fs_info *fs_info, + void __user *arg, int version) { int ret = 0; int size; struct btrfs_ioctl_logical_ino_args *loi; struct btrfs_data_container *inodes = NULL; - struct btrfs_path *path = NULL; + bool ignore_offset; if (!capable(CAP_SYS_ADMIN)) return -EPERM; loi = memdup_user(arg, sizeof(*loi)); - if (IS_ERR(loi)) { - ret = PTR_ERR(loi); - loi = NULL; - goto out; - } + if (IS_ERR(loi)) + return PTR_ERR(loi); - path = btrfs_alloc_path(); - if (!path) { - ret = -ENOMEM; - goto out; + if (version == 1) { + ignore_offset = false; + size = min_t(u32, loi->size, SZ_64K); + } else { + /* All reserved bits must be 0 for now */ + if (memchr_inv(loi->reserved, 0, sizeof(loi->reserved))) { + ret = -EINVAL; + goto out_loi; + } + /* Only accept flags we have defined so far */ + if (loi->flags & ~(BTRFS_LOGICAL_INO_ARGS_IGNORE_OFFSET)) { + ret = -EINVAL; + goto out_loi; + } + ignore_offset = loi->flags & BTRFS_LOGICAL_INO_ARGS_IGNORE_OFFSET; + size = min_t(u32, loi->size, SZ_16M); } - size = min_t(u32, loi->size, 64 * 1024); inodes = init_data_container(size); if (IS_ERR(inodes)) { ret = PTR_ERR(inodes); - inodes = NULL; - goto out; + goto out_loi; } - ret = iterate_inodes_from_logical(loi->logical, root->fs_info, path, - build_ino_list, inodes); + ret = iterate_inodes_from_logical(loi->logical, fs_info, inodes, ignore_offset); if (ret == -EINVAL) ret = -ENOENT; if (ret < 0) goto out; - ret = copy_to_user((void *)(unsigned long)loi->inodes, - (void *)(unsigned long)inodes, size); + ret = copy_to_user((void __user *)(unsigned long)loi->inodes, inodes, + size); if (ret) ret = -EFAULT; out: - btrfs_free_path(path); - vfree(inodes); + kvfree(inodes); +out_loi: kfree(loi); return ret; } -void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock, +void btrfs_update_ioctl_balance_args(struct btrfs_fs_info *fs_info, struct btrfs_ioctl_balance_args *bargs) { struct btrfs_balance_control *bctl = fs_info->balance_ctl; bargs->flags = bctl->flags; - if (atomic_read(&fs_info->balance_running)) + if (test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) bargs->state |= BTRFS_BALANCE_STATE_RUNNING; if (atomic_read(&fs_info->balance_pause_req)) bargs->state |= BTRFS_BALANCE_STATE_PAUSE_REQ; @@ -3493,13 +3377,75 @@ void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock, memcpy(&bargs->meta, &bctl->meta, sizeof(bargs->meta)); memcpy(&bargs->sys, &bctl->sys, sizeof(bargs->sys)); - if (lock) { - spin_lock(&fs_info->balance_lock); - memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat)); - spin_unlock(&fs_info->balance_lock); - } else { - memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat)); + spin_lock(&fs_info->balance_lock); + memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat)); + spin_unlock(&fs_info->balance_lock); +} + +/* + * Try to acquire fs_info::balance_mutex as well as set BTRFS_EXLCOP_BALANCE as + * required. + * + * @fs_info: the filesystem + * @excl_acquired: ptr to boolean value which is set to false in case balance + * is being resumed + * + * Return 0 on success in which case both fs_info::balance is acquired as well + * as exclusive ops are blocked. In case of failure return an error code. + */ +static int btrfs_try_lock_balance(struct btrfs_fs_info *fs_info, bool *excl_acquired) +{ + int ret; + + /* + * Exclusive operation is locked. Three possibilities: + * (1) some other op is running + * (2) balance is running + * (3) balance is paused -- special case (think resume) + */ + while (1) { + if (btrfs_exclop_start(fs_info, BTRFS_EXCLOP_BALANCE)) { + *excl_acquired = true; + mutex_lock(&fs_info->balance_mutex); + return 0; + } + + mutex_lock(&fs_info->balance_mutex); + if (fs_info->balance_ctl) { + /* This is either (2) or (3) */ + if (test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) { + /* This is (2) */ + ret = -EINPROGRESS; + goto out_failure; + + } else { + mutex_unlock(&fs_info->balance_mutex); + /* + * Lock released to allow other waiters to + * continue, we'll reexamine the status again. + */ + mutex_lock(&fs_info->balance_mutex); + + if (fs_info->balance_ctl && + !test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) { + /* This is (3) */ + *excl_acquired = false; + return 0; + } + } + } else { + /* This is (1) */ + ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS; + goto out_failure; + } + + mutex_unlock(&fs_info->balance_mutex); } + +out_failure: + mutex_unlock(&fs_info->balance_mutex); + *excl_acquired = false; + return ret; } static long btrfs_ioctl_balance(struct file *file, void __user *arg) @@ -3508,7 +3454,7 @@ static long btrfs_ioctl_balance(struct file *file, void __user *arg) struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_ioctl_balance_args *bargs; struct btrfs_balance_control *bctl; - bool need_unlock; /* for mut. excl. ops lock */ + bool need_unlock = true; int ret; if (!capable(CAP_SYS_ADMIN)) @@ -3518,152 +3464,102 @@ static long btrfs_ioctl_balance(struct file *file, void __user *arg) if (ret) return ret; -again: - if (!atomic_xchg(&fs_info->mutually_exclusive_operation_running, 1)) { - mutex_lock(&fs_info->volume_mutex); - mutex_lock(&fs_info->balance_mutex); - need_unlock = true; - goto locked; + bargs = memdup_user(arg, sizeof(*bargs)); + if (IS_ERR(bargs)) { + ret = PTR_ERR(bargs); + bargs = NULL; + goto out; } - /* - * mut. excl. ops lock is locked. Three possibilites: - * (1) some other op is running - * (2) balance is running - * (3) balance is paused -- special case (think resume) - */ - mutex_lock(&fs_info->balance_mutex); - if (fs_info->balance_ctl) { - /* this is either (2) or (3) */ - if (!atomic_read(&fs_info->balance_running)) { - mutex_unlock(&fs_info->balance_mutex); - if (!mutex_trylock(&fs_info->volume_mutex)) - goto again; - mutex_lock(&fs_info->balance_mutex); - - if (fs_info->balance_ctl && - !atomic_read(&fs_info->balance_running)) { - /* this is (3) */ - need_unlock = false; - goto locked; - } - - mutex_unlock(&fs_info->balance_mutex); - mutex_unlock(&fs_info->volume_mutex); - goto again; - } else { - /* this is (2) */ - mutex_unlock(&fs_info->balance_mutex); - ret = -EINPROGRESS; - goto out; - } - } else { - /* this is (1) */ - mutex_unlock(&fs_info->balance_mutex); - pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n"); - ret = -EINVAL; + ret = btrfs_try_lock_balance(fs_info, &need_unlock); + if (ret) goto out; - } -locked: - BUG_ON(!atomic_read(&fs_info->mutually_exclusive_operation_running)); + lockdep_assert_held(&fs_info->balance_mutex); - if (arg) { - bargs = memdup_user(arg, sizeof(*bargs)); - if (IS_ERR(bargs)) { - ret = PTR_ERR(bargs); + if (bargs->flags & BTRFS_BALANCE_RESUME) { + if (!fs_info->balance_ctl) { + ret = -ENOTCONN; goto out_unlock; } - if (bargs->flags & BTRFS_BALANCE_RESUME) { - if (!fs_info->balance_ctl) { - ret = -ENOTCONN; - goto out_bargs; - } + bctl = fs_info->balance_ctl; + spin_lock(&fs_info->balance_lock); + bctl->flags |= BTRFS_BALANCE_RESUME; + spin_unlock(&fs_info->balance_lock); + btrfs_exclop_balance(fs_info, BTRFS_EXCLOP_BALANCE); - bctl = fs_info->balance_ctl; - spin_lock(&fs_info->balance_lock); - bctl->flags |= BTRFS_BALANCE_RESUME; - spin_unlock(&fs_info->balance_lock); + goto do_balance; + } - goto do_balance; - } - } else { - bargs = NULL; + if (bargs->flags & ~(BTRFS_BALANCE_ARGS_MASK | BTRFS_BALANCE_TYPE_MASK)) { + ret = -EINVAL; + goto out_unlock; } if (fs_info->balance_ctl) { ret = -EINPROGRESS; - goto out_bargs; + goto out_unlock; } - bctl = kzalloc(sizeof(*bctl), GFP_NOFS); + bctl = kzalloc(sizeof(*bctl), GFP_KERNEL); if (!bctl) { ret = -ENOMEM; - goto out_bargs; + goto out_unlock; } - bctl->fs_info = fs_info; - if (arg) { - memcpy(&bctl->data, &bargs->data, sizeof(bctl->data)); - memcpy(&bctl->meta, &bargs->meta, sizeof(bctl->meta)); - memcpy(&bctl->sys, &bargs->sys, sizeof(bctl->sys)); - - bctl->flags = bargs->flags; - } else { - /* balance everything - no filters */ - bctl->flags |= BTRFS_BALANCE_TYPE_MASK; - } + memcpy(&bctl->data, &bargs->data, sizeof(bctl->data)); + memcpy(&bctl->meta, &bargs->meta, sizeof(bctl->meta)); + memcpy(&bctl->sys, &bargs->sys, sizeof(bctl->sys)); + bctl->flags = bargs->flags; do_balance: /* - * Ownership of bctl and mutually_exclusive_operation_running - * goes to to btrfs_balance. bctl is freed in __cancel_balance, - * or, if restriper was paused all the way until unmount, in - * free_fs_info. mutually_exclusive_operation_running is - * cleared in __cancel_balance. + * Ownership of bctl and exclusive operation goes to btrfs_balance. + * bctl is freed in reset_balance_state, or, if restriper was paused + * all the way until unmount, in free_fs_info. The flag should be + * cleared after reset_balance_state. */ need_unlock = false; - ret = btrfs_balance(bctl, bargs); + ret = btrfs_balance(fs_info, bctl, bargs); + bctl = NULL; - if (arg) { + if (ret == 0 || ret == -ECANCELED) { if (copy_to_user(arg, bargs, sizeof(*bargs))) ret = -EFAULT; } -out_bargs: - kfree(bargs); + kfree(bctl); out_unlock: mutex_unlock(&fs_info->balance_mutex); - mutex_unlock(&fs_info->volume_mutex); if (need_unlock) - atomic_set(&fs_info->mutually_exclusive_operation_running, 0); + btrfs_exclop_finish(fs_info); out: mnt_drop_write_file(file); + kfree(bargs); return ret; } -static long btrfs_ioctl_balance_ctl(struct btrfs_root *root, int cmd) +static long btrfs_ioctl_balance_ctl(struct btrfs_fs_info *fs_info, int cmd) { if (!capable(CAP_SYS_ADMIN)) return -EPERM; switch (cmd) { case BTRFS_BALANCE_CTL_PAUSE: - return btrfs_pause_balance(root->fs_info); + return btrfs_pause_balance(fs_info); case BTRFS_BALANCE_CTL_CANCEL: - return btrfs_cancel_balance(root->fs_info); + return btrfs_cancel_balance(fs_info); } return -EINVAL; } -static long btrfs_ioctl_balance_progress(struct btrfs_root *root, +static long btrfs_ioctl_balance_progress(struct btrfs_fs_info *fs_info, void __user *arg) { - struct btrfs_fs_info *fs_info = root->fs_info; - struct btrfs_ioctl_balance_args *bargs; + struct btrfs_ioctl_balance_args AUTO_KFREE(bargs); int ret = 0; if (!capable(CAP_SYS_ADMIN)) @@ -3675,18 +3571,16 @@ static long btrfs_ioctl_balance_progress(struct btrfs_root *root, goto out; } - bargs = kzalloc(sizeof(*bargs), GFP_NOFS); + bargs = kzalloc(sizeof(*bargs), GFP_KERNEL); if (!bargs) { ret = -ENOMEM; goto out; } - update_ioctl_balance_args(fs_info, 1, bargs); + btrfs_update_ioctl_balance_args(fs_info, bargs); if (copy_to_user(arg, bargs, sizeof(*bargs))) ret = -EFAULT; - - kfree(bargs); out: mutex_unlock(&fs_info->balance_mutex); return ret; @@ -3694,11 +3588,10 @@ out: static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); struct btrfs_ioctl_quota_ctl_args *sa; - struct btrfs_trans_handle *trans = NULL; int ret; - int err; if (!capable(CAP_SYS_ADMIN)) return -EPERM; @@ -3713,31 +3606,50 @@ static long btrfs_ioctl_quota_ctl(struct file *file, void __user *arg) goto drop_write; } - down_write(&root->fs_info->subvol_sem); - trans = btrfs_start_transaction(root->fs_info->tree_root, 2); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - goto out; - } - switch (sa->cmd) { case BTRFS_QUOTA_CTL_ENABLE: - ret = btrfs_quota_enable(trans, root->fs_info); + case BTRFS_QUOTA_CTL_ENABLE_SIMPLE_QUOTA: + down_write(&fs_info->subvol_sem); + ret = btrfs_quota_enable(fs_info, sa); + up_write(&fs_info->subvol_sem); break; case BTRFS_QUOTA_CTL_DISABLE: - ret = btrfs_quota_disable(trans, root->fs_info); + /* + * Lock the cleaner mutex to prevent races with concurrent + * relocation, because relocation may be building backrefs for + * blocks of the quota root while we are deleting the root. This + * is like dropping fs roots of deleted snapshots/subvolumes, we + * need the same protection. + * + * This also prevents races between concurrent tasks trying to + * disable quotas, because we will unlock and relock + * qgroup_ioctl_lock across BTRFS_FS_QUOTA_ENABLED changes. + * + * We take this here because we have the dependency of + * + * inode_lock -> subvol_sem + * + * because of rename. With relocation we can prealloc extents, + * so that makes the dependency chain + * + * cleaner_mutex -> inode_lock -> subvol_sem + * + * so we must take the cleaner_mutex here before we take the + * subvol_sem. The deadlock can't actually happen, but this + * quiets lockdep. + */ + mutex_lock(&fs_info->cleaner_mutex); + down_write(&fs_info->subvol_sem); + ret = btrfs_quota_disable(fs_info); + up_write(&fs_info->subvol_sem); + mutex_unlock(&fs_info->cleaner_mutex); break; default: ret = -EINVAL; break; } - err = btrfs_commit_transaction(trans, root->fs_info->tree_root); - if (err && !ret) - ret = err; -out: kfree(sa); - up_write(&root->fs_info->subvol_sem); drop_write: mnt_drop_write_file(file); return ret; @@ -3745,8 +3657,11 @@ drop_write: static long btrfs_ioctl_qgroup_assign(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); + struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_ioctl_qgroup_assign_args *sa; + struct btrfs_qgroup_list *prealloc = NULL; struct btrfs_trans_handle *trans; int ret; int err; @@ -3754,6 +3669,9 @@ static long btrfs_ioctl_qgroup_assign(struct file *file, void __user *arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; + if (!btrfs_qgroup_enabled(fs_info)) + return -ENOTCONN; + ret = mnt_want_write_file(file); if (ret) return ret; @@ -3764,26 +3682,45 @@ static long btrfs_ioctl_qgroup_assign(struct file *file, void __user *arg) goto drop_write; } + if (sa->assign) { + prealloc = kzalloc(sizeof(*prealloc), GFP_KERNEL); + if (!prealloc) { + ret = -ENOMEM; + goto out; + } + } + trans = btrfs_join_transaction(root); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out; } - /* FIXME: check if the IDs really exist */ + /* + * Prealloc ownership is moved to the relation handler, there it's used + * or freed on error. + */ if (sa->assign) { - ret = btrfs_add_qgroup_relation(trans, root->fs_info, - sa->src, sa->dst); + ret = btrfs_add_qgroup_relation(trans, sa->src, sa->dst, prealloc); + prealloc = NULL; } else { - ret = btrfs_del_qgroup_relation(trans, root->fs_info, - sa->src, sa->dst); - } - - err = btrfs_end_transaction(trans, root); + ret = btrfs_del_qgroup_relation(trans, sa->src, sa->dst); + } + + /* update qgroup status and info */ + mutex_lock(&fs_info->qgroup_ioctl_lock); + err = btrfs_run_qgroups(trans); + mutex_unlock(&fs_info->qgroup_ioctl_lock); + if (err < 0) + btrfs_warn(fs_info, + "qgroup status update failed after %s relation, marked as inconsistent", + sa->assign ? "adding" : "deleting"); + err = btrfs_end_transaction(trans); if (err && !ret) ret = err; out: + kfree(prealloc); kfree(sa); drop_write: mnt_drop_write_file(file); @@ -3792,7 +3729,8 @@ drop_write: static long btrfs_ioctl_qgroup_create(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; + struct inode *inode = file_inode(file); + struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_ioctl_qgroup_create_args *sa; struct btrfs_trans_handle *trans; int ret; @@ -3801,6 +3739,9 @@ static long btrfs_ioctl_qgroup_create(struct file *file, void __user *arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; + if (!btrfs_qgroup_enabled(root->fs_info)) + return -ENOTCONN; + ret = mnt_want_write_file(file); if (ret) return ret; @@ -3816,21 +3757,24 @@ static long btrfs_ioctl_qgroup_create(struct file *file, void __user *arg) goto out; } + if (sa->create && btrfs_is_fstree(sa->qgroupid)) { + ret = -EINVAL; + goto out; + } + trans = btrfs_join_transaction(root); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out; } - /* FIXME: check if the IDs really exist */ if (sa->create) { - ret = btrfs_create_qgroup(trans, root->fs_info, sa->qgroupid, - NULL); + ret = btrfs_create_qgroup(trans, sa->qgroupid); } else { - ret = btrfs_remove_qgroup(trans, root->fs_info, sa->qgroupid); + ret = btrfs_remove_qgroup(trans, sa->qgroupid); } - err = btrfs_end_transaction(trans, root); + err = btrfs_end_transaction(trans); if (err && !ret) ret = err; @@ -3843,7 +3787,8 @@ drop_write: static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; + struct inode *inode = file_inode(file); + struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_ioctl_qgroup_limit_args *sa; struct btrfs_trans_handle *trans; int ret; @@ -3853,6 +3798,9 @@ static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; + if (!btrfs_qgroup_enabled(root->fs_info)) + return -ENOTCONN; + ret = mnt_want_write_file(file); if (ret) return ret; @@ -3872,13 +3820,12 @@ static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg) qgroupid = sa->qgroupid; if (!qgroupid) { /* take the current subvol as qgroup */ - qgroupid = root->root_key.objectid; + qgroupid = btrfs_root_id(root); } - /* FIXME: check if the IDs really exist */ - ret = btrfs_limit_qgroup(trans, root->fs_info, qgroupid, &sa->lim); + ret = btrfs_limit_qgroup(trans, qgroupid, &sa->lim); - err = btrfs_end_transaction(trans, root); + err = btrfs_end_transaction(trans); if (err && !ret) ret = err; @@ -3891,13 +3838,17 @@ drop_write: static long btrfs_ioctl_quota_rescan(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); struct btrfs_ioctl_quota_rescan_args *qsa; int ret; if (!capable(CAP_SYS_ADMIN)) return -EPERM; + if (!btrfs_qgroup_enabled(fs_info)) + return -ENOTCONN; + ret = mnt_want_write_file(file); if (ret) return ret; @@ -3913,7 +3864,7 @@ static long btrfs_ioctl_quota_rescan(struct file *file, void __user *arg) goto out; } - ret = btrfs_qgroup_rescan(root->fs_info); + ret = btrfs_qgroup_rescan(fs_info); out: kfree(qsa); @@ -3922,59 +3873,56 @@ 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 btrfs_root *root = BTRFS_I(file_inode(file))->root; - struct btrfs_ioctl_quota_rescan_args *qsa; - int ret = 0; + struct btrfs_ioctl_quota_rescan_args qsa = {0}; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - qsa = kzalloc(sizeof(*qsa), GFP_NOFS); - if (!qsa) - return -ENOMEM; - - if (root->fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) { - qsa->flags = 1; - qsa->progress = root->fs_info->qgroup_rescan_progress.objectid; + if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) { + qsa.flags = 1; + qsa.progress = fs_info->qgroup_rescan_progress.objectid; } - if (copy_to_user(arg, qsa, sizeof(*qsa))) - ret = -EFAULT; + if (copy_to_user(arg, &qsa, sizeof(qsa))) + return -EFAULT; - kfree(qsa); - return ret; + return 0; } -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) { - struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root; - if (!capable(CAP_SYS_ADMIN)) return -EPERM; - return btrfs_qgroup_wait_for_completion(root->fs_info); + return btrfs_qgroup_wait_for_completion(fs_info, true); } -static long btrfs_ioctl_set_received_subvol(struct file *file, - void __user *arg) +static long _btrfs_ioctl_set_received_subvol(struct file *file, + struct mnt_idmap *idmap, + struct btrfs_ioctl_received_subvol_args *sa) { - struct btrfs_ioctl_received_subvol_args *sa = NULL; struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_root_item *root_item = &root->root_item; struct btrfs_trans_handle *trans; - struct timespec ct = CURRENT_TIME; + struct timespec64 ct = current_time(inode); int ret = 0; + int received_uuid_changed; + + if (!inode_owner_or_capable(idmap, inode)) + return -EPERM; ret = mnt_want_write_file(file); if (ret < 0) return ret; - down_write(&root->fs_info->subvol_sem); + down_write(&fs_info->subvol_sem); - if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) { + if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FIRST_FREE_OBJECTID) { ret = -EINVAL; goto out; } @@ -3984,19 +3932,11 @@ static long btrfs_ioctl_set_received_subvol(struct file *file, goto out; } - if (!inode_owner_or_capable(inode)) { - ret = -EACCES; - goto out; - } - - sa = memdup_user(arg, sizeof(*sa)); - if (IS_ERR(sa)) { - ret = PTR_ERR(sa); - sa = NULL; - goto out; - } - - trans = btrfs_start_transaction(root, 1); + /* + * 1 - root item + * 2 - uuid items (received uuid + subvol uuid) + */ + trans = btrfs_start_transaction(root, 3); if (IS_ERR(trans)) { ret = PTR_ERR(trans); trans = NULL; @@ -4007,60 +3947,155 @@ static long btrfs_ioctl_set_received_subvol(struct file *file, sa->rtime.sec = ct.tv_sec; sa->rtime.nsec = ct.tv_nsec; + received_uuid_changed = memcmp(root_item->received_uuid, sa->uuid, + BTRFS_UUID_SIZE); + if (received_uuid_changed && + !btrfs_is_empty_uuid(root_item->received_uuid)) { + ret = btrfs_uuid_tree_remove(trans, root_item->received_uuid, + BTRFS_UUID_KEY_RECEIVED_SUBVOL, + btrfs_root_id(root)); + if (unlikely(ret && ret != -ENOENT)) { + btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); + goto out; + } + } memcpy(root_item->received_uuid, sa->uuid, BTRFS_UUID_SIZE); btrfs_set_root_stransid(root_item, sa->stransid); btrfs_set_root_rtransid(root_item, sa->rtransid); - root_item->stime.sec = cpu_to_le64(sa->stime.sec); - root_item->stime.nsec = cpu_to_le32(sa->stime.nsec); - root_item->rtime.sec = cpu_to_le64(sa->rtime.sec); - root_item->rtime.nsec = cpu_to_le32(sa->rtime.nsec); + btrfs_set_stack_timespec_sec(&root_item->stime, sa->stime.sec); + btrfs_set_stack_timespec_nsec(&root_item->stime, sa->stime.nsec); + btrfs_set_stack_timespec_sec(&root_item->rtime, sa->rtime.sec); + btrfs_set_stack_timespec_nsec(&root_item->rtime, sa->rtime.nsec); - ret = btrfs_update_root(trans, root->fs_info->tree_root, + ret = btrfs_update_root(trans, fs_info->tree_root, &root->root_key, &root->root_item); if (ret < 0) { - btrfs_end_transaction(trans, root); - trans = NULL; + btrfs_end_transaction(trans); goto out; - } else { - ret = btrfs_commit_transaction(trans, root); - if (ret < 0) + } + if (received_uuid_changed && !btrfs_is_empty_uuid(sa->uuid)) { + ret = btrfs_uuid_tree_add(trans, sa->uuid, + BTRFS_UUID_KEY_RECEIVED_SUBVOL, + btrfs_root_id(root)); + if (unlikely(ret < 0 && ret != -EEXIST)) { + btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); goto out; + } + } + ret = btrfs_commit_transaction(trans); +out: + up_write(&fs_info->subvol_sem); + mnt_drop_write_file(file); + return ret; +} + +#ifdef CONFIG_64BIT +static long btrfs_ioctl_set_received_subvol_32(struct file *file, + void __user *arg) +{ + struct btrfs_ioctl_received_subvol_args_32 *args32 = NULL; + struct btrfs_ioctl_received_subvol_args *args64 = NULL; + int ret = 0; + + args32 = memdup_user(arg, sizeof(*args32)); + if (IS_ERR(args32)) + return PTR_ERR(args32); + + args64 = kmalloc(sizeof(*args64), GFP_KERNEL); + if (!args64) { + ret = -ENOMEM; + goto out; } + memcpy(args64->uuid, args32->uuid, BTRFS_UUID_SIZE); + args64->stransid = args32->stransid; + args64->rtransid = args32->rtransid; + args64->stime.sec = args32->stime.sec; + args64->stime.nsec = args32->stime.nsec; + args64->rtime.sec = args32->rtime.sec; + args64->rtime.nsec = args32->rtime.nsec; + args64->flags = args32->flags; + + ret = _btrfs_ioctl_set_received_subvol(file, file_mnt_idmap(file), args64); + if (ret) + goto out; + + memcpy(args32->uuid, args64->uuid, BTRFS_UUID_SIZE); + args32->stransid = args64->stransid; + args32->rtransid = args64->rtransid; + args32->stime.sec = args64->stime.sec; + args32->stime.nsec = args64->stime.nsec; + args32->rtime.sec = args64->rtime.sec; + args32->rtime.nsec = args64->rtime.nsec; + args32->flags = args64->flags; + + ret = copy_to_user(arg, args32, sizeof(*args32)); + if (ret) + ret = -EFAULT; + +out: + kfree(args32); + kfree(args64); + return ret; +} +#endif + +static long btrfs_ioctl_set_received_subvol(struct file *file, + void __user *arg) +{ + struct btrfs_ioctl_received_subvol_args *sa = NULL; + int ret = 0; + + sa = memdup_user(arg, sizeof(*sa)); + if (IS_ERR(sa)) + return PTR_ERR(sa); + + ret = _btrfs_ioctl_set_received_subvol(file, file_mnt_idmap(file), sa); + + if (ret) + goto out; + ret = copy_to_user(arg, sa, sizeof(*sa)); if (ret) ret = -EFAULT; out: kfree(sa); - up_write(&root->fs_info->subvol_sem); - mnt_drop_write_file(file); 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 btrfs_root *root = BTRFS_I(file_inode(file))->root; - const char *label = root->fs_info->super_copy->label; - size_t len = strnlen(label, BTRFS_LABEL_SIZE); + size_t len; int ret; + char label[BTRFS_LABEL_SIZE]; + + spin_lock(&fs_info->super_lock); + memcpy(label, fs_info->super_copy->label, BTRFS_LABEL_SIZE); + spin_unlock(&fs_info->super_lock); + + len = strnlen(label, BTRFS_LABEL_SIZE); if (len == BTRFS_LABEL_SIZE) { - pr_warn("btrfs: label is too long, return the first %zu bytes\n", - --len); + btrfs_warn(fs_info, + "label is too long, return the first %zu bytes", + --len); } - mutex_lock(&root->fs_info->volume_mutex); ret = copy_to_user(arg, label, len); - mutex_unlock(&root->fs_info->volume_mutex); return ret ? -EFAULT : 0; } static int btrfs_ioctl_set_fslabel(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; - struct btrfs_super_block *super_block = root->fs_info->super_copy; + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_super_block *super_block = fs_info->super_copy; struct btrfs_trans_handle *trans; char label[BTRFS_LABEL_SIZE]; int ret; @@ -4072,8 +4107,9 @@ static int btrfs_ioctl_set_fslabel(struct file *file, void __user *arg) return -EFAULT; if (strnlen(label, BTRFS_LABEL_SIZE) == BTRFS_LABEL_SIZE) { - pr_err("btrfs: unable to set label with more than %d bytes\n", - BTRFS_LABEL_SIZE - 1); + btrfs_err(fs_info, + "unable to set label with more than %d bytes", + BTRFS_LABEL_SIZE - 1); return -EINVAL; } @@ -4081,49 +4117,1103 @@ static int btrfs_ioctl_set_fslabel(struct file *file, void __user *arg) if (ret) return ret; - mutex_lock(&root->fs_info->volume_mutex); trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out_unlock; } - strcpy(super_block->label, label); - ret = btrfs_end_transaction(trans, root); + spin_lock(&fs_info->super_lock); + strscpy(super_block->label, label); + spin_unlock(&fs_info->super_lock); + ret = btrfs_commit_transaction(trans); out_unlock: - mutex_unlock(&root->fs_info->volume_mutex); mnt_drop_write_file(file); return ret; } +#define INIT_FEATURE_FLAGS(suffix) \ + { .compat_flags = BTRFS_FEATURE_COMPAT_##suffix, \ + .compat_ro_flags = BTRFS_FEATURE_COMPAT_RO_##suffix, \ + .incompat_flags = BTRFS_FEATURE_INCOMPAT_##suffix } + +int btrfs_ioctl_get_supported_features(void __user *arg) +{ + static const struct btrfs_ioctl_feature_flags features[3] = { + INIT_FEATURE_FLAGS(SUPP), + INIT_FEATURE_FLAGS(SAFE_SET), + INIT_FEATURE_FLAGS(SAFE_CLEAR) + }; + + if (copy_to_user(arg, &features, sizeof(features))) + return -EFAULT; + + return 0; +} + +static int btrfs_ioctl_get_features(struct btrfs_fs_info *fs_info, + void __user *arg) +{ + struct btrfs_super_block *super_block = fs_info->super_copy; + struct btrfs_ioctl_feature_flags features; + + features.compat_flags = btrfs_super_compat_flags(super_block); + features.compat_ro_flags = btrfs_super_compat_ro_flags(super_block); + features.incompat_flags = btrfs_super_incompat_flags(super_block); + + if (copy_to_user(arg, &features, sizeof(features))) + return -EFAULT; + + return 0; +} + +static int check_feature_bits(const struct btrfs_fs_info *fs_info, + enum btrfs_feature_set set, + u64 change_mask, u64 flags, u64 supported_flags, + u64 safe_set, u64 safe_clear) +{ + const char *type = btrfs_feature_set_name(set); + const char AUTO_KFREE(names); + u64 disallowed, unsupported; + u64 set_mask = flags & change_mask; + u64 clear_mask = ~flags & change_mask; + + unsupported = set_mask & ~supported_flags; + if (unsupported) { + names = btrfs_printable_features(set, unsupported); + if (names) + btrfs_warn(fs_info, + "this kernel does not support the %s feature bit%s", + names, strchr(names, ',') ? "s" : ""); + else + btrfs_warn(fs_info, + "this kernel does not support %s bits 0x%llx", + type, unsupported); + return -EOPNOTSUPP; + } + + disallowed = set_mask & ~safe_set; + if (disallowed) { + names = btrfs_printable_features(set, disallowed); + if (names) + btrfs_warn(fs_info, + "can't set the %s feature bit%s while mounted", + names, strchr(names, ',') ? "s" : ""); + else + btrfs_warn(fs_info, + "can't set %s bits 0x%llx while mounted", + type, disallowed); + return -EPERM; + } + + disallowed = clear_mask & ~safe_clear; + if (disallowed) { + names = btrfs_printable_features(set, disallowed); + if (names) + btrfs_warn(fs_info, + "can't clear the %s feature bit%s while mounted", + names, strchr(names, ',') ? "s" : ""); + else + btrfs_warn(fs_info, + "can't clear %s bits 0x%llx while mounted", + type, disallowed); + return -EPERM; + } + + return 0; +} + +#define check_feature(fs_info, change_mask, flags, mask_base) \ +check_feature_bits(fs_info, FEAT_##mask_base, change_mask, flags, \ + BTRFS_FEATURE_ ## mask_base ## _SUPP, \ + BTRFS_FEATURE_ ## mask_base ## _SAFE_SET, \ + BTRFS_FEATURE_ ## mask_base ## _SAFE_CLEAR) + +static int btrfs_ioctl_set_features(struct file *file, void __user *arg) +{ + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_super_block *super_block = fs_info->super_copy; + struct btrfs_ioctl_feature_flags flags[2]; + struct btrfs_trans_handle *trans; + u64 newflags; + int ret; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (copy_from_user(flags, arg, sizeof(flags))) + return -EFAULT; + + /* Nothing to do */ + if (!flags[0].compat_flags && !flags[0].compat_ro_flags && + !flags[0].incompat_flags) + return 0; + + ret = check_feature(fs_info, flags[0].compat_flags, + flags[1].compat_flags, COMPAT); + if (ret) + return ret; + + ret = check_feature(fs_info, flags[0].compat_ro_flags, + flags[1].compat_ro_flags, COMPAT_RO); + if (ret) + return ret; + + ret = check_feature(fs_info, flags[0].incompat_flags, + flags[1].incompat_flags, INCOMPAT); + if (ret) + return ret; + + ret = mnt_want_write_file(file); + if (ret) + return ret; + + trans = btrfs_start_transaction(root, 0); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + goto out_drop_write; + } + + spin_lock(&fs_info->super_lock); + newflags = btrfs_super_compat_flags(super_block); + newflags |= flags[0].compat_flags & flags[1].compat_flags; + newflags &= ~(flags[0].compat_flags & ~flags[1].compat_flags); + btrfs_set_super_compat_flags(super_block, newflags); + + newflags = btrfs_super_compat_ro_flags(super_block); + newflags |= flags[0].compat_ro_flags & flags[1].compat_ro_flags; + newflags &= ~(flags[0].compat_ro_flags & ~flags[1].compat_ro_flags); + btrfs_set_super_compat_ro_flags(super_block, newflags); + + newflags = btrfs_super_incompat_flags(super_block); + newflags |= flags[0].incompat_flags & flags[1].incompat_flags; + newflags &= ~(flags[0].incompat_flags & ~flags[1].incompat_flags); + btrfs_set_super_incompat_flags(super_block, newflags); + spin_unlock(&fs_info->super_lock); + + ret = btrfs_commit_transaction(trans); +out_drop_write: + mnt_drop_write_file(file); + + return ret; +} + +static int _btrfs_ioctl_send(struct btrfs_root *root, void __user *argp, bool compat) +{ + struct btrfs_ioctl_send_args *arg; + int ret; + + if (compat) { +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + struct btrfs_ioctl_send_args_32 args32 = { 0 }; + + ret = copy_from_user(&args32, argp, sizeof(args32)); + if (ret) + return -EFAULT; + arg = kzalloc(sizeof(*arg), GFP_KERNEL); + if (!arg) + return -ENOMEM; + arg->send_fd = args32.send_fd; + arg->clone_sources_count = args32.clone_sources_count; + arg->clone_sources = compat_ptr(args32.clone_sources); + arg->parent_root = args32.parent_root; + arg->flags = args32.flags; + arg->version = args32.version; + memcpy(arg->reserved, args32.reserved, + sizeof(args32.reserved)); +#else + return -ENOTTY; +#endif + } else { + arg = memdup_user(argp, sizeof(*arg)); + if (IS_ERR(arg)) + return PTR_ERR(arg); + } + ret = btrfs_ioctl_send(root, arg); + kfree(arg); + return ret; +} + +static int btrfs_ioctl_encoded_read(struct file *file, void __user *argp, + bool compat) +{ + struct btrfs_ioctl_encoded_io_args args = { 0 }; + size_t copy_end_kernel = offsetofend(struct btrfs_ioctl_encoded_io_args, + flags); + size_t copy_end; + struct btrfs_inode *inode = BTRFS_I(file_inode(file)); + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct extent_io_tree *io_tree = &inode->io_tree; + struct iovec iovstack[UIO_FASTIOV]; + struct iovec *iov = iovstack; + struct iov_iter iter; + loff_t pos; + struct kiocb kiocb; + ssize_t ret; + u64 disk_bytenr, disk_io_size; + struct extent_state *cached_state = NULL; + + if (!capable(CAP_SYS_ADMIN)) { + ret = -EPERM; + goto out_acct; + } + + if (compat) { +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + struct btrfs_ioctl_encoded_io_args_32 args32; + + copy_end = offsetofend(struct btrfs_ioctl_encoded_io_args_32, + flags); + if (copy_from_user(&args32, argp, copy_end)) { + ret = -EFAULT; + goto out_acct; + } + args.iov = compat_ptr(args32.iov); + args.iovcnt = args32.iovcnt; + args.offset = args32.offset; + args.flags = args32.flags; +#else + return -ENOTTY; +#endif + } else { + copy_end = copy_end_kernel; + if (copy_from_user(&args, argp, copy_end)) { + ret = -EFAULT; + goto out_acct; + } + } + if (args.flags != 0) { + ret = -EINVAL; + goto out_acct; + } + + ret = import_iovec(ITER_DEST, args.iov, args.iovcnt, ARRAY_SIZE(iovstack), + &iov, &iter); + if (ret < 0) + goto out_acct; + + if (iov_iter_count(&iter) == 0) { + ret = 0; + goto out_iov; + } + pos = args.offset; + ret = rw_verify_area(READ, file, &pos, args.len); + if (ret < 0) + goto out_iov; + + init_sync_kiocb(&kiocb, file); + kiocb.ki_pos = pos; + + ret = btrfs_encoded_read(&kiocb, &iter, &args, &cached_state, + &disk_bytenr, &disk_io_size); + + if (ret == -EIOCBQUEUED) { + bool unlocked = false; + u64 start, lockend, count; + + start = ALIGN_DOWN(kiocb.ki_pos, fs_info->sectorsize); + lockend = start + BTRFS_MAX_UNCOMPRESSED - 1; + + if (args.compression) + count = disk_io_size; + else + count = args.len; + + ret = btrfs_encoded_read_regular(&kiocb, &iter, start, lockend, + &cached_state, disk_bytenr, + disk_io_size, count, + args.compression, &unlocked); + + if (!unlocked) { + btrfs_unlock_extent(io_tree, start, lockend, &cached_state); + btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); + } + } + + if (ret >= 0) { + fsnotify_access(file); + if (copy_to_user(argp + copy_end, + (char *)&args + copy_end_kernel, + sizeof(args) - copy_end_kernel)) + ret = -EFAULT; + } + +out_iov: + kfree(iov); +out_acct: + if (ret > 0) + add_rchar(current, ret); + inc_syscr(current); + return ret; +} + +static int btrfs_ioctl_encoded_write(struct file *file, void __user *argp, bool compat) +{ + struct btrfs_ioctl_encoded_io_args args; + struct iovec iovstack[UIO_FASTIOV]; + struct iovec *iov = iovstack; + struct iov_iter iter; + loff_t pos; + struct kiocb kiocb; + ssize_t ret; + + if (!capable(CAP_SYS_ADMIN)) { + ret = -EPERM; + goto out_acct; + } + + if (!(file->f_mode & FMODE_WRITE)) { + ret = -EBADF; + goto out_acct; + } + + if (compat) { +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + struct btrfs_ioctl_encoded_io_args_32 args32; + + if (copy_from_user(&args32, argp, sizeof(args32))) { + ret = -EFAULT; + goto out_acct; + } + args.iov = compat_ptr(args32.iov); + args.iovcnt = args32.iovcnt; + args.offset = args32.offset; + args.flags = args32.flags; + args.len = args32.len; + args.unencoded_len = args32.unencoded_len; + args.unencoded_offset = args32.unencoded_offset; + args.compression = args32.compression; + args.encryption = args32.encryption; + memcpy(args.reserved, args32.reserved, sizeof(args.reserved)); +#else + return -ENOTTY; +#endif + } else { + if (copy_from_user(&args, argp, sizeof(args))) { + ret = -EFAULT; + goto out_acct; + } + } + + ret = -EINVAL; + if (args.flags != 0) + goto out_acct; + if (memchr_inv(args.reserved, 0, sizeof(args.reserved))) + goto out_acct; + if (args.compression == BTRFS_ENCODED_IO_COMPRESSION_NONE && + args.encryption == BTRFS_ENCODED_IO_ENCRYPTION_NONE) + goto out_acct; + if (args.compression >= BTRFS_ENCODED_IO_COMPRESSION_TYPES || + args.encryption >= BTRFS_ENCODED_IO_ENCRYPTION_TYPES) + goto out_acct; + if (args.unencoded_offset > args.unencoded_len) + goto out_acct; + if (args.len > args.unencoded_len - args.unencoded_offset) + goto out_acct; + + ret = import_iovec(ITER_SOURCE, args.iov, args.iovcnt, ARRAY_SIZE(iovstack), + &iov, &iter); + if (ret < 0) + goto out_acct; + + if (iov_iter_count(&iter) == 0) { + ret = 0; + goto out_iov; + } + pos = args.offset; + ret = rw_verify_area(WRITE, file, &pos, args.len); + if (ret < 0) + goto out_iov; + + init_sync_kiocb(&kiocb, file); + ret = kiocb_set_rw_flags(&kiocb, 0, WRITE); + if (ret) + goto out_iov; + kiocb.ki_pos = pos; + + file_start_write(file); + + ret = btrfs_do_write_iter(&kiocb, &iter, &args); + if (ret > 0) + fsnotify_modify(file); + + file_end_write(file); +out_iov: + kfree(iov); +out_acct: + if (ret > 0) + add_wchar(current, ret); + inc_syscw(current); + return ret; +} + +struct btrfs_uring_encoded_data { + struct btrfs_ioctl_encoded_io_args args; + struct iovec iovstack[UIO_FASTIOV]; + struct iovec *iov; + struct iov_iter iter; +}; + +/* + * Context that's attached to an encoded read io_uring command, in cmd->pdu. It + * contains the fields in btrfs_uring_read_extent that are necessary to finish + * off and cleanup the I/O in btrfs_uring_read_finished. + */ +struct btrfs_uring_priv { + struct io_uring_cmd *cmd; + struct page **pages; + unsigned long nr_pages; + struct kiocb iocb; + struct iovec *iov; + struct iov_iter iter; + struct extent_state *cached_state; + u64 count; + u64 start; + u64 lockend; + int err; + bool compressed; +}; + +struct io_btrfs_cmd { + struct btrfs_uring_encoded_data *data; + struct btrfs_uring_priv *priv; +}; + +static void btrfs_uring_read_finished(struct io_tw_req tw_req, io_tw_token_t tw) +{ + struct io_uring_cmd *cmd = io_uring_cmd_from_tw(tw_req); + struct io_btrfs_cmd *bc = io_uring_cmd_to_pdu(cmd, struct io_btrfs_cmd); + struct btrfs_uring_priv *priv = bc->priv; + struct btrfs_inode *inode = BTRFS_I(file_inode(priv->iocb.ki_filp)); + struct extent_io_tree *io_tree = &inode->io_tree; + pgoff_t index; + u64 cur; + size_t page_offset; + ssize_t ret; + + /* The inode lock has already been acquired in btrfs_uring_read_extent. */ + btrfs_lockdep_inode_acquire(inode, i_rwsem); + + if (priv->err) { + ret = priv->err; + goto out; + } + + if (priv->compressed) { + index = 0; + page_offset = 0; + } else { + index = (priv->iocb.ki_pos - priv->start) >> PAGE_SHIFT; + page_offset = offset_in_page(priv->iocb.ki_pos - priv->start); + } + cur = 0; + while (cur < priv->count) { + size_t bytes = min_t(size_t, priv->count - cur, PAGE_SIZE - page_offset); + + if (copy_page_to_iter(priv->pages[index], page_offset, bytes, + &priv->iter) != bytes) { + ret = -EFAULT; + goto out; + } + + index++; + cur += bytes; + page_offset = 0; + } + ret = priv->count; + +out: + btrfs_unlock_extent(io_tree, priv->start, priv->lockend, &priv->cached_state); + btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); + + io_uring_cmd_done(cmd, ret, IO_URING_CMD_TASK_WORK_ISSUE_FLAGS); + add_rchar(current, ret); + + for (index = 0; index < priv->nr_pages; index++) + __free_page(priv->pages[index]); + + kfree(priv->pages); + kfree(priv->iov); + kfree(priv); + kfree(bc->data); +} + +void btrfs_uring_read_extent_endio(void *ctx, int err) +{ + struct btrfs_uring_priv *priv = ctx; + struct io_btrfs_cmd *bc = io_uring_cmd_to_pdu(priv->cmd, struct io_btrfs_cmd); + + priv->err = err; + bc->priv = priv; + + io_uring_cmd_complete_in_task(priv->cmd, btrfs_uring_read_finished); +} + +static int btrfs_uring_read_extent(struct kiocb *iocb, struct iov_iter *iter, + u64 start, u64 lockend, + struct extent_state *cached_state, + u64 disk_bytenr, u64 disk_io_size, + size_t count, bool compressed, + struct iovec *iov, struct io_uring_cmd *cmd) +{ + struct btrfs_inode *inode = BTRFS_I(file_inode(iocb->ki_filp)); + struct extent_io_tree *io_tree = &inode->io_tree; + struct page **pages; + struct btrfs_uring_priv *priv = NULL; + unsigned long nr_pages; + int ret; + + nr_pages = DIV_ROUND_UP(disk_io_size, PAGE_SIZE); + pages = kcalloc(nr_pages, sizeof(struct page *), GFP_NOFS); + if (!pages) + return -ENOMEM; + ret = btrfs_alloc_page_array(nr_pages, pages, 0); + if (ret) { + ret = -ENOMEM; + goto out_fail; + } + + priv = kmalloc(sizeof(*priv), GFP_NOFS); + if (!priv) { + ret = -ENOMEM; + goto out_fail; + } + + priv->iocb = *iocb; + priv->iov = iov; + priv->iter = *iter; + priv->count = count; + priv->cmd = cmd; + priv->cached_state = cached_state; + priv->compressed = compressed; + priv->nr_pages = nr_pages; + priv->pages = pages; + priv->start = start; + priv->lockend = lockend; + priv->err = 0; + + ret = btrfs_encoded_read_regular_fill_pages(inode, disk_bytenr, + disk_io_size, pages, priv); + if (ret && ret != -EIOCBQUEUED) + goto out_fail; + + /* + * If we return -EIOCBQUEUED, we're deferring the cleanup to + * btrfs_uring_read_finished(), which will handle unlocking the extent + * and inode and freeing the allocations. + */ + + /* + * We're returning to userspace with the inode lock held, and that's + * okay - it'll get unlocked in a worker thread. Call + * btrfs_lockdep_inode_release() to avoid confusing lockdep. + */ + btrfs_lockdep_inode_release(inode, i_rwsem); + + return -EIOCBQUEUED; + +out_fail: + btrfs_unlock_extent(io_tree, start, lockend, &cached_state); + btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); + kfree(priv); + return ret; +} + +static int btrfs_uring_encoded_read(struct io_uring_cmd *cmd, unsigned int issue_flags) +{ + struct file *file = cmd->file; + struct btrfs_inode *inode = BTRFS_I(file->f_inode); + struct extent_io_tree *io_tree = &inode->io_tree; + struct btrfs_fs_info *fs_info = inode->root->fs_info; + size_t copy_end_kernel = offsetofend(struct btrfs_ioctl_encoded_io_args, flags); + size_t copy_end; + int ret; + u64 disk_bytenr, disk_io_size; + loff_t pos; + struct kiocb kiocb; + struct extent_state *cached_state = NULL; + u64 start, lockend; + void __user *sqe_addr; + struct io_btrfs_cmd *bc = io_uring_cmd_to_pdu(cmd, struct io_btrfs_cmd); + struct btrfs_uring_encoded_data *data = NULL; + + if (cmd->flags & IORING_URING_CMD_REISSUE) + data = bc->data; + + if (!capable(CAP_SYS_ADMIN)) { + ret = -EPERM; + goto out_acct; + } + sqe_addr = u64_to_user_ptr(READ_ONCE(cmd->sqe->addr)); + + if (issue_flags & IO_URING_F_COMPAT) { +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + copy_end = offsetofend(struct btrfs_ioctl_encoded_io_args_32, flags); +#else + ret = -ENOTTY; + goto out_acct; +#endif + } else { + copy_end = copy_end_kernel; + } + + if (!data) { + data = kzalloc(sizeof(*data), GFP_NOFS); + if (!data) { + ret = -ENOMEM; + goto out_acct; + } + + bc->data = data; + + if (issue_flags & IO_URING_F_COMPAT) { +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + struct btrfs_ioctl_encoded_io_args_32 args32; + + if (copy_from_user(&args32, sqe_addr, copy_end)) { + ret = -EFAULT; + goto out_acct; + } + + data->args.iov = compat_ptr(args32.iov); + data->args.iovcnt = args32.iovcnt; + data->args.offset = args32.offset; + data->args.flags = args32.flags; +#endif + } else { + if (copy_from_user(&data->args, sqe_addr, copy_end)) { + ret = -EFAULT; + goto out_acct; + } + } + + if (data->args.flags != 0) { + ret = -EINVAL; + goto out_acct; + } + + data->iov = data->iovstack; + ret = import_iovec(ITER_DEST, data->args.iov, data->args.iovcnt, + ARRAY_SIZE(data->iovstack), &data->iov, + &data->iter); + if (ret < 0) + goto out_acct; + + if (iov_iter_count(&data->iter) == 0) { + ret = 0; + goto out_free; + } + } + + pos = data->args.offset; + ret = rw_verify_area(READ, file, &pos, data->args.len); + if (ret < 0) + goto out_free; + + init_sync_kiocb(&kiocb, file); + kiocb.ki_pos = pos; + + if (issue_flags & IO_URING_F_NONBLOCK) + kiocb.ki_flags |= IOCB_NOWAIT; + + start = ALIGN_DOWN(pos, fs_info->sectorsize); + lockend = start + BTRFS_MAX_UNCOMPRESSED - 1; + + ret = btrfs_encoded_read(&kiocb, &data->iter, &data->args, &cached_state, + &disk_bytenr, &disk_io_size); + if (ret == -EAGAIN) + goto out_acct; + if (ret < 0 && ret != -EIOCBQUEUED) + goto out_free; + + file_accessed(file); + + if (copy_to_user(sqe_addr + copy_end, + (const char *)&data->args + copy_end_kernel, + sizeof(data->args) - copy_end_kernel)) { + if (ret == -EIOCBQUEUED) { + btrfs_unlock_extent(io_tree, start, lockend, &cached_state); + btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); + } + ret = -EFAULT; + goto out_free; + } + + if (ret == -EIOCBQUEUED) { + u64 count = min_t(u64, iov_iter_count(&data->iter), disk_io_size); + + /* Match ioctl by not returning past EOF if uncompressed. */ + if (!data->args.compression) + count = min_t(u64, count, data->args.len); + + ret = btrfs_uring_read_extent(&kiocb, &data->iter, start, lockend, + cached_state, disk_bytenr, disk_io_size, + count, data->args.compression, + data->iov, cmd); + + goto out_acct; + } + +out_free: + kfree(data->iov); + +out_acct: + if (ret > 0) + add_rchar(current, ret); + inc_syscr(current); + + if (ret != -EIOCBQUEUED && ret != -EAGAIN) + kfree(data); + + return ret; +} + +static int btrfs_uring_encoded_write(struct io_uring_cmd *cmd, unsigned int issue_flags) +{ + struct file *file = cmd->file; + loff_t pos; + struct kiocb kiocb; + ssize_t ret; + void __user *sqe_addr; + struct io_btrfs_cmd *bc = io_uring_cmd_to_pdu(cmd, struct io_btrfs_cmd); + struct btrfs_uring_encoded_data *data = NULL; + + if (cmd->flags & IORING_URING_CMD_REISSUE) + data = bc->data; + + if (!capable(CAP_SYS_ADMIN)) { + ret = -EPERM; + goto out_acct; + } + sqe_addr = u64_to_user_ptr(READ_ONCE(cmd->sqe->addr)); + + if (!(file->f_mode & FMODE_WRITE)) { + ret = -EBADF; + goto out_acct; + } + + if (!data) { + data = kzalloc(sizeof(*data), GFP_NOFS); + if (!data) { + ret = -ENOMEM; + goto out_acct; + } + + bc->data = data; + + if (issue_flags & IO_URING_F_COMPAT) { +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + struct btrfs_ioctl_encoded_io_args_32 args32; + + if (copy_from_user(&args32, sqe_addr, sizeof(args32))) { + ret = -EFAULT; + goto out_acct; + } + data->args.iov = compat_ptr(args32.iov); + data->args.iovcnt = args32.iovcnt; + data->args.offset = args32.offset; + data->args.flags = args32.flags; + data->args.len = args32.len; + data->args.unencoded_len = args32.unencoded_len; + data->args.unencoded_offset = args32.unencoded_offset; + data->args.compression = args32.compression; + data->args.encryption = args32.encryption; + memcpy(data->args.reserved, args32.reserved, + sizeof(data->args.reserved)); +#else + ret = -ENOTTY; + goto out_acct; +#endif + } else { + if (copy_from_user(&data->args, sqe_addr, sizeof(data->args))) { + ret = -EFAULT; + goto out_acct; + } + } + + ret = -EINVAL; + if (data->args.flags != 0) + goto out_acct; + if (memchr_inv(data->args.reserved, 0, sizeof(data->args.reserved))) + goto out_acct; + if (data->args.compression == BTRFS_ENCODED_IO_COMPRESSION_NONE && + data->args.encryption == BTRFS_ENCODED_IO_ENCRYPTION_NONE) + goto out_acct; + if (data->args.compression >= BTRFS_ENCODED_IO_COMPRESSION_TYPES || + data->args.encryption >= BTRFS_ENCODED_IO_ENCRYPTION_TYPES) + goto out_acct; + if (data->args.unencoded_offset > data->args.unencoded_len) + goto out_acct; + if (data->args.len > data->args.unencoded_len - data->args.unencoded_offset) + goto out_acct; + + data->iov = data->iovstack; + ret = import_iovec(ITER_SOURCE, data->args.iov, data->args.iovcnt, + ARRAY_SIZE(data->iovstack), &data->iov, + &data->iter); + if (ret < 0) + goto out_acct; + + if (iov_iter_count(&data->iter) == 0) { + ret = 0; + goto out_iov; + } + } + + if (issue_flags & IO_URING_F_NONBLOCK) { + ret = -EAGAIN; + goto out_acct; + } + + pos = data->args.offset; + ret = rw_verify_area(WRITE, file, &pos, data->args.len); + if (ret < 0) + goto out_iov; + + init_sync_kiocb(&kiocb, file); + ret = kiocb_set_rw_flags(&kiocb, 0, WRITE); + if (ret) + goto out_iov; + kiocb.ki_pos = pos; + + file_start_write(file); + + ret = btrfs_do_write_iter(&kiocb, &data->iter, &data->args); + if (ret > 0) + fsnotify_modify(file); + + file_end_write(file); +out_iov: + kfree(data->iov); +out_acct: + if (ret > 0) + add_wchar(current, ret); + inc_syscw(current); + + if (ret != -EAGAIN) + kfree(data); + return ret; +} + +int btrfs_uring_cmd(struct io_uring_cmd *cmd, unsigned int issue_flags) +{ + if (unlikely(btrfs_is_shutdown(inode_to_fs_info(file_inode(cmd->file))))) + return -EIO; + + switch (cmd->cmd_op) { + case BTRFS_IOC_ENCODED_READ: +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + case BTRFS_IOC_ENCODED_READ_32: +#endif + return btrfs_uring_encoded_read(cmd, issue_flags); + + case BTRFS_IOC_ENCODED_WRITE: +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + case BTRFS_IOC_ENCODED_WRITE_32: +#endif + return btrfs_uring_encoded_write(cmd, issue_flags); + } + + return -EINVAL; +} + +static int btrfs_ioctl_subvol_sync(struct btrfs_fs_info *fs_info, void __user *argp) +{ + struct btrfs_root *root; + struct btrfs_ioctl_subvol_wait args = { 0 }; + signed long sched_ret; + int refs; + u64 root_flags; + bool wait_for_deletion = false; + bool found = false; + + if (copy_from_user(&args, argp, sizeof(args))) + return -EFAULT; + + switch (args.mode) { + case BTRFS_SUBVOL_SYNC_WAIT_FOR_QUEUED: + /* + * Wait for the first one deleted that waits until all previous + * are cleaned. + */ + spin_lock(&fs_info->trans_lock); + if (!list_empty(&fs_info->dead_roots)) { + root = list_last_entry(&fs_info->dead_roots, + struct btrfs_root, root_list); + args.subvolid = btrfs_root_id(root); + found = true; + } + spin_unlock(&fs_info->trans_lock); + if (!found) + return -ENOENT; + + fallthrough; + case BTRFS_SUBVOL_SYNC_WAIT_FOR_ONE: + if ((0 < args.subvolid && args.subvolid < BTRFS_FIRST_FREE_OBJECTID) || + BTRFS_LAST_FREE_OBJECTID < args.subvolid) + return -EINVAL; + break; + case BTRFS_SUBVOL_SYNC_COUNT: + spin_lock(&fs_info->trans_lock); + args.count = list_count_nodes(&fs_info->dead_roots); + spin_unlock(&fs_info->trans_lock); + if (copy_to_user(argp, &args, sizeof(args))) + return -EFAULT; + return 0; + case BTRFS_SUBVOL_SYNC_PEEK_FIRST: + spin_lock(&fs_info->trans_lock); + /* Last in the list was deleted first. */ + if (!list_empty(&fs_info->dead_roots)) { + root = list_last_entry(&fs_info->dead_roots, + struct btrfs_root, root_list); + args.subvolid = btrfs_root_id(root); + } else { + args.subvolid = 0; + } + spin_unlock(&fs_info->trans_lock); + if (copy_to_user(argp, &args, sizeof(args))) + return -EFAULT; + return 0; + case BTRFS_SUBVOL_SYNC_PEEK_LAST: + spin_lock(&fs_info->trans_lock); + /* First in the list was deleted last. */ + if (!list_empty(&fs_info->dead_roots)) { + root = list_first_entry(&fs_info->dead_roots, + struct btrfs_root, root_list); + args.subvolid = btrfs_root_id(root); + } else { + args.subvolid = 0; + } + spin_unlock(&fs_info->trans_lock); + if (copy_to_user(argp, &args, sizeof(args))) + return -EFAULT; + return 0; + default: + return -EINVAL; + } + + /* 32bit limitation: fs_roots_radix key is not wide enough. */ + if (sizeof(unsigned long) != sizeof(u64) && args.subvolid > U32_MAX) + return -EOVERFLOW; + + while (1) { + /* Wait for the specific one. */ + if (down_read_interruptible(&fs_info->subvol_sem) == -EINTR) + return -EINTR; + refs = -1; + spin_lock(&fs_info->fs_roots_radix_lock); + root = radix_tree_lookup(&fs_info->fs_roots_radix, + (unsigned long)args.subvolid); + if (root) { + spin_lock(&root->root_item_lock); + refs = btrfs_root_refs(&root->root_item); + root_flags = btrfs_root_flags(&root->root_item); + spin_unlock(&root->root_item_lock); + } + spin_unlock(&fs_info->fs_roots_radix_lock); + up_read(&fs_info->subvol_sem); + + /* Subvolume does not exist. */ + if (!root) + return -ENOENT; + + /* Subvolume not deleted at all. */ + if (refs > 0) + return -EEXIST; + /* We've waited and now the subvolume is gone. */ + if (wait_for_deletion && refs == -1) { + /* Return the one we waited for as the last one. */ + if (copy_to_user(argp, &args, sizeof(args))) + return -EFAULT; + return 0; + } + + /* Subvolume not found on the first try (deleted or never existed). */ + if (refs == -1) + return -ENOENT; + + wait_for_deletion = true; + ASSERT(root_flags & BTRFS_ROOT_SUBVOL_DEAD); + sched_ret = schedule_timeout_interruptible(HZ); + /* Early wake up or error. */ + if (sched_ret != 0) + return -EINTR; + } + + return 0; +} + +#ifdef CONFIG_BTRFS_EXPERIMENTAL +static int btrfs_ioctl_shutdown(struct btrfs_fs_info *fs_info, unsigned long arg) +{ + int ret = 0; + u32 flags; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (get_user(flags, (u32 __user *)arg)) + return -EFAULT; + + if (flags >= BTRFS_SHUTDOWN_FLAGS_LAST) + return -EINVAL; + + if (btrfs_is_shutdown(fs_info)) + return 0; + + switch (flags) { + case BTRFS_SHUTDOWN_FLAGS_LOGFLUSH: + case BTRFS_SHUTDOWN_FLAGS_DEFAULT: + ret = freeze_super(fs_info->sb, FREEZE_HOLDER_KERNEL, NULL); + if (ret) + return ret; + btrfs_force_shutdown(fs_info); + ret = thaw_super(fs_info->sb, FREEZE_HOLDER_KERNEL, NULL); + if (ret) + return ret; + break; + case BTRFS_SHUTDOWN_FLAGS_NOLOGFLUSH: + btrfs_force_shutdown(fs_info); + break; + } + return ret; +} +#endif + long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { - struct btrfs_root *root = BTRFS_I(file_inode(file))->root; + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); + struct btrfs_root *root = BTRFS_I(inode)->root; void __user *argp = (void __user *)arg; switch (cmd) { - case FS_IOC_GETFLAGS: - return btrfs_ioctl_getflags(file, argp); - case FS_IOC_SETFLAGS: - return btrfs_ioctl_setflags(file, argp); case FS_IOC_GETVERSION: - return btrfs_ioctl_getversion(file, argp); + return btrfs_ioctl_getversion(inode, argp); + case FS_IOC_GETFSLABEL: + 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); + return btrfs_ioctl_snap_create(file, argp, false); case BTRFS_IOC_SNAP_CREATE_V2: - return btrfs_ioctl_snap_create_v2(file, argp, 0); + return btrfs_ioctl_snap_create_v2(file, argp, false); case BTRFS_IOC_SUBVOL_CREATE: - return btrfs_ioctl_snap_create(file, argp, 1); + return btrfs_ioctl_snap_create(file, argp, true); case BTRFS_IOC_SUBVOL_CREATE_V2: - return btrfs_ioctl_snap_create_v2(file, argp, 1); + return btrfs_ioctl_snap_create_v2(file, argp, true); 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); + return btrfs_ioctl_subvol_getflags(BTRFS_I(inode), argp); case BTRFS_IOC_SUBVOL_SETFLAGS: return btrfs_ioctl_subvol_setflags(file, argp); case BTRFS_IOC_DEFAULT_SUBVOL: @@ -4135,58 +5225,73 @@ long btrfs_ioctl(struct file *file, unsigned int case BTRFS_IOC_RESIZE: return btrfs_ioctl_resize(file, argp); case BTRFS_IOC_ADD_DEV: - return btrfs_ioctl_add_dev(root, argp); + return btrfs_ioctl_add_dev(fs_info, argp); case BTRFS_IOC_RM_DEV: return btrfs_ioctl_rm_dev(file, argp); + case BTRFS_IOC_RM_DEV_V2: + return btrfs_ioctl_rm_dev_v2(file, argp); case BTRFS_IOC_FS_INFO: - return btrfs_ioctl_fs_info(root, argp); + return btrfs_ioctl_fs_info(fs_info, argp); case BTRFS_IOC_DEV_INFO: - return btrfs_ioctl_dev_info(root, argp); - case BTRFS_IOC_BALANCE: - return btrfs_ioctl_balance(file, NULL); - case BTRFS_IOC_CLONE: - return btrfs_ioctl_clone(file, arg, 0, 0, 0); - case BTRFS_IOC_CLONE_RANGE: - return btrfs_ioctl_clone_range(file, argp); - case BTRFS_IOC_TRANS_START: - return btrfs_ioctl_trans_start(file); - case BTRFS_IOC_TRANS_END: - return btrfs_ioctl_trans_end(file); + return btrfs_ioctl_dev_info(fs_info, argp); case BTRFS_IOC_TREE_SEARCH: - return btrfs_ioctl_tree_search(file, argp); + return btrfs_ioctl_tree_search(root, argp); + case BTRFS_IOC_TREE_SEARCH_V2: + return btrfs_ioctl_tree_search_v2(root, argp); case BTRFS_IOC_INO_LOOKUP: - return btrfs_ioctl_ino_lookup(file, argp); + return btrfs_ioctl_ino_lookup(root, argp); case BTRFS_IOC_INO_PATHS: return btrfs_ioctl_ino_to_path(root, argp); case BTRFS_IOC_LOGICAL_INO: - return btrfs_ioctl_logical_to_ino(root, argp); + return btrfs_ioctl_logical_to_ino(fs_info, argp, 1); + case BTRFS_IOC_LOGICAL_INO_V2: + return btrfs_ioctl_logical_to_ino(fs_info, argp, 2); case BTRFS_IOC_SPACE_INFO: - return btrfs_ioctl_space_info(root, argp); - case BTRFS_IOC_SYNC: - btrfs_sync_fs(file->f_dentry->d_sb, 1); - return 0; + return btrfs_ioctl_space_info(fs_info, argp); + case BTRFS_IOC_SYNC: { + int ret; + + ret = btrfs_start_delalloc_roots(fs_info, LONG_MAX, false); + if (ret) + return ret; + ret = btrfs_sync_fs(inode->i_sb, 1); + /* + * There may be work for the cleaner kthread to do (subvolume + * deletion, delayed iputs, defrag inodes, etc), so wake it up. + */ + wake_up_process(fs_info->cleaner_kthread); + return ret; + } case BTRFS_IOC_START_SYNC: return btrfs_ioctl_start_sync(root, argp); case BTRFS_IOC_WAIT_SYNC: - return btrfs_ioctl_wait_sync(root, argp); + return btrfs_ioctl_wait_sync(fs_info, argp); case BTRFS_IOC_SCRUB: return btrfs_ioctl_scrub(file, argp); case BTRFS_IOC_SCRUB_CANCEL: - return btrfs_ioctl_scrub_cancel(root, argp); + return btrfs_ioctl_scrub_cancel(fs_info); case BTRFS_IOC_SCRUB_PROGRESS: - return btrfs_ioctl_scrub_progress(root, argp); + return btrfs_ioctl_scrub_progress(fs_info, argp); case BTRFS_IOC_BALANCE_V2: return btrfs_ioctl_balance(file, argp); case BTRFS_IOC_BALANCE_CTL: - return btrfs_ioctl_balance_ctl(root, arg); + return btrfs_ioctl_balance_ctl(fs_info, arg); case BTRFS_IOC_BALANCE_PROGRESS: - return btrfs_ioctl_balance_progress(root, argp); + return btrfs_ioctl_balance_progress(fs_info, argp); case BTRFS_IOC_SET_RECEIVED_SUBVOL: return btrfs_ioctl_set_received_subvol(file, argp); +#ifdef CONFIG_64BIT + case BTRFS_IOC_SET_RECEIVED_SUBVOL_32: + return btrfs_ioctl_set_received_subvol_32(file, argp); +#endif case BTRFS_IOC_SEND: - return btrfs_ioctl_send(file, argp); + return _btrfs_ioctl_send(root, argp, false); +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + case BTRFS_IOC_SEND_32: + return _btrfs_ioctl_send(root, argp, true); +#endif case BTRFS_IOC_GET_DEV_STATS: - return btrfs_ioctl_get_dev_stats(root, argp); + return btrfs_ioctl_get_dev_stats(fs_info, argp); case BTRFS_IOC_QUOTA_CTL: return btrfs_ioctl_quota_ctl(file, argp); case BTRFS_IOC_QGROUP_ASSIGN: @@ -4198,16 +5303,63 @@ 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); case BTRFS_IOC_DEV_REPLACE: - return btrfs_ioctl_dev_replace(root, argp); - case BTRFS_IOC_GET_FSLABEL: - return btrfs_ioctl_get_fslabel(file, argp); - case BTRFS_IOC_SET_FSLABEL: - return btrfs_ioctl_set_fslabel(file, argp); + 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(fs_info, argp); + case BTRFS_IOC_SET_FEATURES: + return btrfs_ioctl_set_features(file, argp); + case BTRFS_IOC_GET_SUBVOL_INFO: + return btrfs_ioctl_get_subvol_info(inode, argp); + case BTRFS_IOC_GET_SUBVOL_ROOTREF: + return btrfs_ioctl_get_subvol_rootref(root, argp); + case BTRFS_IOC_INO_LOOKUP_USER: + return btrfs_ioctl_ino_lookup_user(file, argp); + case FS_IOC_ENABLE_VERITY: + return fsverity_ioctl_enable(file, (const void __user *)argp); + case FS_IOC_MEASURE_VERITY: + return fsverity_ioctl_measure(file, argp); + case FS_IOC_READ_VERITY_METADATA: + return fsverity_ioctl_read_metadata(file, argp); + case BTRFS_IOC_ENCODED_READ: + return btrfs_ioctl_encoded_read(file, argp, false); + case BTRFS_IOC_ENCODED_WRITE: + return btrfs_ioctl_encoded_write(file, argp, false); +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + case BTRFS_IOC_ENCODED_READ_32: + return btrfs_ioctl_encoded_read(file, argp, true); + case BTRFS_IOC_ENCODED_WRITE_32: + return btrfs_ioctl_encoded_write(file, argp, true); +#endif + case BTRFS_IOC_SUBVOL_SYNC_WAIT: + return btrfs_ioctl_subvol_sync(fs_info, argp); +#ifdef CONFIG_BTRFS_EXPERIMENTAL + case BTRFS_IOC_SHUTDOWN: + return btrfs_ioctl_shutdown(fs_info, arg); +#endif } return -ENOTTY; } + +#ifdef CONFIG_COMPAT +long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + /* + * These all access 32-bit values anyway so no further + * handling is necessary. + */ + switch (cmd) { + case FS_IOC32_GETVERSION: + cmd = FS_IOC_GETVERSION; + break; + } + + return btrfs_ioctl(file, cmd, (unsigned long) compat_ptr(arg)); +} +#endif |