From 8ecebf4d767e2307a946c8905278d6358eda35c3 Mon Sep 17 00:00:00 2001 From: Robbie Ko Date: Mon, 6 Aug 2018 10:30:30 +0800 Subject: Btrfs: fix unexpected failure of nocow buffered writes after snapshotting when low on space Commit e9894fd3e3b3 ("Btrfs: fix snapshot vs nocow writting") forced nocow writes to fallback to COW, during writeback, when a snapshot is created. This resulted in writes made before creating the snapshot to unexpectedly fail with ENOSPC during writeback when success (0) was returned to user space through the write system call. The steps leading to this problem are: 1. When it's not possible to allocate data space for a write, the buffered write path checks if a NOCOW write is possible. If it is, it will not reserve space and success (0) is returned to user space. 2. Then when a snapshot is created, the root's will_be_snapshotted atomic is incremented and writeback is triggered for all inode's that belong to the root being snapshotted. Incrementing that atomic forces all previous writes to fallback to COW during writeback (running delalloc). 3. This results in the writeback for the inodes to fail and therefore setting the ENOSPC error in their mappings, so that a subsequent fsync on them will report the error to user space. So it's not a completely silent data loss (since fsync will report ENOSPC) but it's a very unexpected and undesirable behaviour, because if a clean shutdown/unmount of the filesystem happens without previous calls to fsync, it is expected to have the data present in the files after mounting the filesystem again. So fix this by adding a new atomic named snapshot_force_cow to the root structure which prevents this behaviour and works the following way: 1. It is incremented when we start to create a snapshot after triggering writeback and before waiting for writeback to finish. 2. This new atomic is now what is used by writeback (running delalloc) to decide whether we need to fallback to COW or not. Because we incremented this new atomic after triggering writeback in the snapshot creation ioctl, we ensure that all buffered writes that happened before snapshot creation will succeed and not fallback to COW (which would make them fail with ENOSPC). 3. The existing atomic, will_be_snapshotted, is kept because it is used to force new buffered writes, that start after we started snapshotting, to reserve data space even when NOCOW is possible. This makes these writes fail early with ENOSPC when there's no available space to allocate, preventing the unexpected behaviour of writeback later failing with ENOSPC due to a fallback to COW mode. Fixes: e9894fd3e3b3 ("Btrfs: fix snapshot vs nocow writting") Signed-off-by: Robbie Ko Reviewed-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 1 + fs/btrfs/disk-io.c | 1 + fs/btrfs/inode.c | 25 ++++--------------------- fs/btrfs/ioctl.c | 16 ++++++++++++++++ 4 files changed, 22 insertions(+), 21 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 318be7864072..a67cc190a84b 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1280,6 +1280,7 @@ struct btrfs_root { int send_in_progress; struct btrfs_subvolume_writers *subv_writers; atomic_t will_be_snapshotted; + atomic_t snapshot_force_cow; /* For qgroup metadata reserved space */ spinlock_t qgroup_meta_rsv_lock; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 5124c15705ce..05dc3c17cb62 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1187,6 +1187,7 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, atomic_set(&root->log_batch, 0); refcount_set(&root->refs, 1); atomic_set(&root->will_be_snapshotted, 0); + atomic_set(&root->snapshot_force_cow, 0); root->log_transid = 0; root->log_transid_committed = -1; root->last_log_commit = 0; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 3f51ddc18f98..c6d8c5d19ff0 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1271,7 +1271,7 @@ static noinline int run_delalloc_nocow(struct inode *inode, u64 disk_num_bytes; u64 ram_bytes; int extent_type; - int ret, err; + int ret; int type; int nocow; int check_prev = 1; @@ -1403,11 +1403,8 @@ next_slot: * if there are pending snapshots for this root, * we fall into common COW way. */ - if (!nolock) { - err = btrfs_start_write_no_snapshotting(root); - if (!err) - goto out_check; - } + if (!nolock && atomic_read(&root->snapshot_force_cow)) + goto out_check; /* * force cow if csum exists in the range. * this ensure that csum for a given extent are @@ -1416,9 +1413,6 @@ next_slot: ret = csum_exist_in_range(fs_info, disk_bytenr, num_bytes); if (ret) { - if (!nolock) - btrfs_end_write_no_snapshotting(root); - /* * ret could be -EIO if the above fails to read * metadata. @@ -1431,11 +1425,8 @@ next_slot: WARN_ON_ONCE(nolock); goto out_check; } - if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr)) { - if (!nolock) - btrfs_end_write_no_snapshotting(root); + if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr)) goto out_check; - } nocow = 1; } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { extent_end = found_key.offset + @@ -1448,8 +1439,6 @@ next_slot: out_check: if (extent_end <= start) { path->slots[0]++; - if (!nolock && nocow) - btrfs_end_write_no_snapshotting(root); if (nocow) btrfs_dec_nocow_writers(fs_info, disk_bytenr); goto next_slot; @@ -1471,8 +1460,6 @@ out_check: end, page_started, nr_written, 1, NULL); if (ret) { - if (!nolock && nocow) - btrfs_end_write_no_snapshotting(root); if (nocow) btrfs_dec_nocow_writers(fs_info, disk_bytenr); @@ -1492,8 +1479,6 @@ out_check: ram_bytes, BTRFS_COMPRESS_NONE, BTRFS_ORDERED_PREALLOC); if (IS_ERR(em)) { - if (!nolock && nocow) - btrfs_end_write_no_snapshotting(root); if (nocow) btrfs_dec_nocow_writers(fs_info, disk_bytenr); @@ -1532,8 +1517,6 @@ out_check: EXTENT_CLEAR_DATA_RESV, PAGE_UNLOCK | PAGE_SET_PRIVATE2); - if (!nolock && nocow) - btrfs_end_write_no_snapshotting(root); cur_offset = extent_end; /* diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index d3a5d2a41e5f..85c4284bb2cf 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -747,6 +747,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, struct btrfs_pending_snapshot *pending_snapshot; struct btrfs_trans_handle *trans; int ret; + bool snapshot_force_cow = false; if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state)) return -EINVAL; @@ -763,6 +764,11 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, goto free_pending; } + /* + * Force new buffered writes to reserve space even when NOCOW is + * possible. This is to avoid later writeback (running dealloc) to + * fallback to COW mode and unexpectedly fail with ENOSPC. + */ atomic_inc(&root->will_be_snapshotted); smp_mb__after_atomic(); /* wait for no snapshot writes */ @@ -773,6 +779,14 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, if (ret) goto dec_and_free; + /* + * All previous writes have started writeback in NOCOW mode, so now + * we force future writes to fallback to COW mode during snapshot + * creation. + */ + atomic_inc(&root->snapshot_force_cow); + snapshot_force_cow = true; + btrfs_wait_ordered_extents(root, U64_MAX, 0, (u64)-1); btrfs_init_block_rsv(&pending_snapshot->block_rsv, @@ -837,6 +851,8 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, fail: btrfs_subvolume_release_metadata(fs_info, &pending_snapshot->block_rsv); dec_and_free: + if (snapshot_force_cow) + atomic_dec(&root->snapshot_force_cow); if (atomic_dec_and_test(&root->will_be_snapshotted)) wake_up_var(&root->will_be_snapshotted); free_pending: -- cgit v1.2.3 From d4682ba03ef618b6ef4be7cedc7aacaf505d3a58 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 11 Jun 2018 19:24:28 +0100 Subject: Btrfs: sync log after logging new name When we add a new name for an inode which was logged in the current transaction, we update the inode in the log so that its new name and ancestors are added to the log. However when we do this we do not persist the log, so the changes remain in memory only, and as a consequence, any ancestors that were created in the current transaction are updated such that future calls to btrfs_inode_in_log() return true. This leads to a subsequent fsync against such new ancestor directories returning immediately, without persisting the log, therefore after a power failure the new ancestor directories do not exist, despite fsync being called against them explicitly. Example: $ mkfs.btrfs -f /dev/sdb $ mount /dev/sdb /mnt $ mkdir /mnt/A $ mkdir /mnt/B $ mkdir /mnt/A/C $ touch /mnt/B/foo $ xfs_io -c "fsync" /mnt/B/foo $ ln /mnt/B/foo /mnt/A/C/foo $ xfs_io -c "fsync" /mnt/A After the power failure, directory "A" does not exist, despite the explicit fsync on it. Instead of fixing this by changing the behaviour of the explicit fsync on directory "A" to persist the log instead of doing nothing, make the logging of the new file name (which happens when creating a hard link or renaming) persist the log. This approach not only is simpler, not requiring addition of new fields to the inode in memory structure, but also gives us the same behaviour as ext4, xfs and f2fs (possibly other filesystems too). A test case for fstests follows soon. Fixes: 12fcfd22fe5b ("Btrfs: tree logging unlink/rename fixes") Reported-by: Vijay Chidambaram Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/inode.c | 92 ++++++++++++++++++++++++++++++++++++++++++++++------- fs/btrfs/tree-log.c | 48 ++++++++++++++++++++++++---- fs/btrfs/tree-log.h | 10 +++++- 3 files changed, 131 insertions(+), 19 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index c6d8c5d19ff0..cb09873ad270 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6634,6 +6634,8 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, drop_inode = 1; } else { struct dentry *parent = dentry->d_parent; + int ret; + err = btrfs_update_inode(trans, root, inode); if (err) goto fail; @@ -6647,7 +6649,12 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, goto fail; } d_instantiate(dentry, inode); - btrfs_log_new_name(trans, BTRFS_I(inode), NULL, parent); + ret = btrfs_log_new_name(trans, BTRFS_I(inode), NULL, parent, + true, NULL); + if (ret == BTRFS_NEED_TRANS_COMMIT) { + err = btrfs_commit_transaction(trans); + trans = NULL; + } } fail: @@ -9386,14 +9393,21 @@ static int btrfs_rename_exchange(struct inode *old_dir, u64 new_idx = 0; u64 root_objectid; int ret; - int ret2; bool root_log_pinned = false; bool dest_log_pinned = false; + struct btrfs_log_ctx ctx_root; + struct btrfs_log_ctx ctx_dest; + bool sync_log_root = false; + bool sync_log_dest = false; + bool commit_transaction = false; /* we only allow rename subvolume link between subvolumes */ if (old_ino != BTRFS_FIRST_FREE_OBJECTID && root != dest) return -EXDEV; + btrfs_init_log_ctx(&ctx_root, old_inode); + btrfs_init_log_ctx(&ctx_dest, new_inode); + /* close the race window with snapshot create/destroy ioctl */ if (old_ino == BTRFS_FIRST_FREE_OBJECTID) down_read(&fs_info->subvol_sem); @@ -9540,15 +9554,29 @@ static int btrfs_rename_exchange(struct inode *old_dir, if (root_log_pinned) { parent = new_dentry->d_parent; - btrfs_log_new_name(trans, BTRFS_I(old_inode), BTRFS_I(old_dir), - parent); + ret = btrfs_log_new_name(trans, BTRFS_I(old_inode), + BTRFS_I(old_dir), parent, + false, &ctx_root); + if (ret == BTRFS_NEED_LOG_SYNC) + sync_log_root = true; + else if (ret == BTRFS_NEED_TRANS_COMMIT) + commit_transaction = true; + ret = 0; btrfs_end_log_trans(root); root_log_pinned = false; } if (dest_log_pinned) { - parent = old_dentry->d_parent; - btrfs_log_new_name(trans, BTRFS_I(new_inode), BTRFS_I(new_dir), - parent); + if (!commit_transaction) { + parent = old_dentry->d_parent; + ret = btrfs_log_new_name(trans, BTRFS_I(new_inode), + BTRFS_I(new_dir), parent, + false, &ctx_dest); + if (ret == BTRFS_NEED_LOG_SYNC) + sync_log_dest = true; + else if (ret == BTRFS_NEED_TRANS_COMMIT) + commit_transaction = true; + ret = 0; + } btrfs_end_log_trans(dest); dest_log_pinned = false; } @@ -9581,8 +9609,26 @@ out_fail: dest_log_pinned = false; } } - ret2 = btrfs_end_transaction(trans); - ret = ret ? ret : ret2; + if (!ret && sync_log_root && !commit_transaction) { + ret = btrfs_sync_log(trans, BTRFS_I(old_inode)->root, + &ctx_root); + if (ret) + commit_transaction = true; + } + if (!ret && sync_log_dest && !commit_transaction) { + ret = btrfs_sync_log(trans, BTRFS_I(new_inode)->root, + &ctx_dest); + if (ret) + commit_transaction = true; + } + if (commit_transaction) { + ret = btrfs_commit_transaction(trans); + } else { + int ret2; + + ret2 = btrfs_end_transaction(trans); + ret = ret ? ret : ret2; + } out_notrans: if (new_ino == BTRFS_FIRST_FREE_OBJECTID) up_read(&fs_info->subvol_sem); @@ -9659,6 +9705,9 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, int ret; u64 old_ino = btrfs_ino(BTRFS_I(old_inode)); bool log_pinned = false; + struct btrfs_log_ctx ctx; + bool sync_log = false; + bool commit_transaction = false; if (btrfs_ino(BTRFS_I(new_dir)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID) return -EPERM; @@ -9816,8 +9865,15 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (log_pinned) { struct dentry *parent = new_dentry->d_parent; - btrfs_log_new_name(trans, BTRFS_I(old_inode), BTRFS_I(old_dir), - parent); + btrfs_init_log_ctx(&ctx, old_inode); + ret = btrfs_log_new_name(trans, BTRFS_I(old_inode), + BTRFS_I(old_dir), parent, + false, &ctx); + if (ret == BTRFS_NEED_LOG_SYNC) + sync_log = true; + else if (ret == BTRFS_NEED_TRANS_COMMIT) + commit_transaction = true; + ret = 0; btrfs_end_log_trans(root); log_pinned = false; } @@ -9854,7 +9910,19 @@ out_fail: btrfs_end_log_trans(root); log_pinned = false; } - btrfs_end_transaction(trans); + if (!ret && sync_log) { + ret = btrfs_sync_log(trans, BTRFS_I(old_inode)->root, &ctx); + if (ret) + commit_transaction = true; + } + if (commit_transaction) { + ret = btrfs_commit_transaction(trans); + } else { + int ret2; + + ret2 = btrfs_end_transaction(trans); + ret = ret ? ret : ret2; + } out_notrans: if (old_ino == BTRFS_FIRST_FREE_OBJECTID) up_read(&fs_info->subvol_sem); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 1650dc44a5e3..3c2ae0e4f25a 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -6025,14 +6025,25 @@ void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans, * Call this after adding a new name for a file and it will properly * update the log to reflect the new name. * - * It will return zero if all goes well, and it will return 1 if a - * full transaction commit is required. + * @ctx can not be NULL when @sync_log is false, and should be NULL when it's + * true (because it's not used). + * + * Return value depends on whether @sync_log is true or false. + * When true: returns BTRFS_NEED_TRANS_COMMIT if the transaction needs to be + * committed by the caller, and BTRFS_DONT_NEED_TRANS_COMMIT + * otherwise. + * When false: returns BTRFS_DONT_NEED_LOG_SYNC if the caller does not need to + * to sync the log, BTRFS_NEED_LOG_SYNC if it needs to sync the log, + * or BTRFS_NEED_TRANS_COMMIT if the transaction needs to be + * committed (without attempting to sync the log). */ int btrfs_log_new_name(struct btrfs_trans_handle *trans, struct btrfs_inode *inode, struct btrfs_inode *old_dir, - struct dentry *parent) + struct dentry *parent, + bool sync_log, struct btrfs_log_ctx *ctx) { struct btrfs_fs_info *fs_info = trans->fs_info; + int ret; /* * this will force the logging code to walk the dentry chain @@ -6047,9 +6058,34 @@ int btrfs_log_new_name(struct btrfs_trans_handle *trans, */ if (inode->logged_trans <= fs_info->last_trans_committed && (!old_dir || old_dir->logged_trans <= fs_info->last_trans_committed)) - return 0; + return sync_log ? BTRFS_DONT_NEED_TRANS_COMMIT : + BTRFS_DONT_NEED_LOG_SYNC; + + if (sync_log) { + struct btrfs_log_ctx ctx2; + + btrfs_init_log_ctx(&ctx2, &inode->vfs_inode); + ret = btrfs_log_inode_parent(trans, inode, parent, 0, LLONG_MAX, + LOG_INODE_EXISTS, &ctx2); + if (ret == BTRFS_NO_LOG_SYNC) + return BTRFS_DONT_NEED_TRANS_COMMIT; + else if (ret) + return BTRFS_NEED_TRANS_COMMIT; + + ret = btrfs_sync_log(trans, inode->root, &ctx2); + if (ret) + return BTRFS_NEED_TRANS_COMMIT; + return BTRFS_DONT_NEED_TRANS_COMMIT; + } + + ASSERT(ctx); + ret = btrfs_log_inode_parent(trans, inode, parent, 0, LLONG_MAX, + LOG_INODE_EXISTS, ctx); + if (ret == BTRFS_NO_LOG_SYNC) + return BTRFS_DONT_NEED_LOG_SYNC; + else if (ret) + return BTRFS_NEED_TRANS_COMMIT; - return btrfs_log_inode_parent(trans, inode, parent, 0, LLONG_MAX, - LOG_INODE_EXISTS, NULL); + return BTRFS_NEED_LOG_SYNC; } diff --git a/fs/btrfs/tree-log.h b/fs/btrfs/tree-log.h index 122e68b89a5a..7ab9bb88a639 100644 --- a/fs/btrfs/tree-log.h +++ b/fs/btrfs/tree-log.h @@ -71,8 +71,16 @@ void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans, int for_rename); void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans, struct btrfs_inode *dir); +/* Return values for btrfs_log_new_name() */ +enum { + BTRFS_DONT_NEED_TRANS_COMMIT, + BTRFS_NEED_TRANS_COMMIT, + BTRFS_DONT_NEED_LOG_SYNC, + BTRFS_NEED_LOG_SYNC, +}; int btrfs_log_new_name(struct btrfs_trans_handle *trans, struct btrfs_inode *inode, struct btrfs_inode *old_dir, - struct dentry *parent); + struct dentry *parent, + bool sync_log, struct btrfs_log_ctx *ctx); #endif -- cgit v1.2.3 From de02b9f6bb65a6a1848f346f7a3617b7a9b930c0 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 17 Aug 2018 09:38:59 +0100 Subject: Btrfs: fix data corruption when deduplicating between different files If we deduplicate extents between two different files we can end up corrupting data if the source range ends at the size of the source file, the source file's size is not aligned to the filesystem's block size and the destination range does not go past the size of the destination file size. Example: $ mkfs.btrfs -f /dev/sdb $ mount /dev/sdb /mnt $ xfs_io -f -c "pwrite -S 0x6b 0 2518890" /mnt/foo # The first byte with a value of 0xae starts at an offset (2518890) # which is not a multiple of the sector size. $ xfs_io -c "pwrite -S 0xae 2518890 102398" /mnt/foo # Confirm the file content is full of bytes with values 0x6b and 0xae. $ od -t x1 /mnt/foo 0000000 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b * 11467540 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b ae ae ae ae ae ae 11467560 ae ae ae ae ae ae ae ae ae ae ae ae ae ae ae ae * 11777540 ae ae ae ae ae ae ae ae 11777550 # Create a second file with a length not aligned to the sector size, # whose bytes all have the value 0x6b, so that its extent(s) can be # deduplicated with the first file. $ xfs_io -f -c "pwrite -S 0x6b 0 557771" /mnt/bar # Now deduplicate the entire second file into a range of the first file # that also has all bytes with the value 0x6b. The destination range's # end offset must not be aligned to the sector size and must be less # then the offset of the first byte with the value 0xae (byte at offset # 2518890). $ xfs_io -c "dedupe /mnt/bar 0 1957888 557771" /mnt/foo # The bytes in the range starting at offset 2515659 (end of the # deduplication range) and ending at offset 2519040 (start offset # rounded up to the block size) must all have the value 0xae (and not # replaced with 0x00 values). In other words, we should have exactly # the same data we had before we asked for deduplication. $ od -t x1 /mnt/foo 0000000 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b * 11467540 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b ae ae ae ae ae ae 11467560 ae ae ae ae ae ae ae ae ae ae ae ae ae ae ae ae * 11777540 ae ae ae ae ae ae ae ae 11777550 # Unmount the filesystem and mount it again. This guarantees any file # data in the page cache is dropped. $ umount /dev/sdb $ mount /dev/sdb /mnt $ od -t x1 /mnt/foo 0000000 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b * 11461300 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 00 00 00 00 00 11461320 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 * 11470000 ae ae ae ae ae ae ae ae ae ae ae ae ae ae ae ae * 11777540 ae ae ae ae ae ae ae ae 11777550 # The bytes in range 2515659 to 2519040 have a value of 0x00 and not a # value of 0xae, data corruption happened due to the deduplication # operation. So fix this by rounding down, to the sector size, the length used for the deduplication when the following conditions are met: 1) Source file's range ends at its i_size; 2) Source file's i_size is not aligned to the sector size; 3) Destination range does not cross the i_size of the destination file. Fixes: e1d227a42ea2 ("btrfs: Handle unaligned length in extent_same") CC: stable@vger.kernel.org # 4.2+ Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/ioctl.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) (limited to 'fs') diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 85c4284bb2cf..011ddfcc96e2 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -3469,6 +3469,25 @@ static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 olen, same_lock_start = min_t(u64, loff, dst_loff); same_lock_len = max_t(u64, loff, dst_loff) + len - same_lock_start; + } else { + /* + * If the source and destination inodes are different, the + * source's range end offset matches the source's i_size, that + * i_size is not a multiple of the sector size, and the + * destination range does not go past the destination's i_size, + * we must round down the length to the nearest sector size + * multiple. If we don't do this adjustment we end replacing + * with zeroes the bytes in the range that starts at the + * deduplication range's end offset and ends at the next sector + * size multiple. + */ + if (loff + olen == i_size_read(src) && + dst_loff + len < i_size_read(dst)) { + const u64 sz = BTRFS_I(src)->root->fs_info->sectorsize; + + len = round_down(i_size_read(src), sz) - loff; + olen = len; + } } again: -- cgit v1.2.3 From a5b7f4295eeae8b05ca91f6d145cd8773b08de9e Mon Sep 17 00:00:00 2001 From: Lu Fengqi Date: Thu, 9 Aug 2018 09:46:04 +0800 Subject: btrfs: fix qgroup_free wrong num_bytes in btrfs_subvolume_reserve_metadata After btrfs_qgroup_reserve_meta_prealloc(), num_bytes will be assigned again by btrfs_calc_trans_metadata_size(). Once block_rsv fails, we can't properly free the num_bytes of the previous qgroup_reserve. Use a separate variable to store the num_bytes of the qgroup_reserve. Delete the comment for the qgroup_reserved that does not exist and add a comment about use_global_rsv. Fixes: c4c129db5da8 ("btrfs: drop unused parameter qgroup_reserved") CC: stable@vger.kernel.org # 4.18+ Signed-off-by: Lu Fengqi Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-tree.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index de6f75f5547b..2d9074295d7f 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -5800,7 +5800,7 @@ void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans) * root: the root of the parent directory * rsv: block reservation * items: the number of items that we need do reservation - * qgroup_reserved: used to return the reserved size in qgroup + * use_global_rsv: allow fallback to the global block reservation * * This function is used to reserve the space for snapshot/subvolume * creation and deletion. Those operations are different with the @@ -5810,10 +5810,10 @@ void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans) * the space reservation mechanism in start_transaction(). */ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, - struct btrfs_block_rsv *rsv, - int items, + struct btrfs_block_rsv *rsv, int items, bool use_global_rsv) { + u64 qgroup_num_bytes = 0; u64 num_bytes; int ret; struct btrfs_fs_info *fs_info = root->fs_info; @@ -5821,12 +5821,11 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags)) { /* One for parent inode, two for dir entries */ - num_bytes = 3 * fs_info->nodesize; - ret = btrfs_qgroup_reserve_meta_prealloc(root, num_bytes, true); + qgroup_num_bytes = 3 * fs_info->nodesize; + ret = btrfs_qgroup_reserve_meta_prealloc(root, + qgroup_num_bytes, true); if (ret) return ret; - } else { - num_bytes = 0; } num_bytes = btrfs_calc_trans_metadata_size(fs_info, items); @@ -5838,8 +5837,8 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, if (ret == -ENOSPC && use_global_rsv) ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes, 1); - if (ret && num_bytes) - btrfs_qgroup_free_meta_prealloc(root, num_bytes); + if (ret && qgroup_num_bytes) + btrfs_qgroup_free_meta_prealloc(root, qgroup_num_bytes); return ret; } -- cgit v1.2.3 From 801660b040d132f67fac6a95910ad307c5929b49 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Mon, 6 Aug 2018 18:12:37 +0800 Subject: btrfs: btrfs_shrink_device should call commit transaction at the end Test case btrfs/164 reports use-after-free: [ 6712.084324] general protection fault: 0000 [#1] PREEMPT SMP .. [ 6712.195423] btrfs_update_commit_device_size+0x75/0xf0 [btrfs] [ 6712.201424] btrfs_commit_transaction+0x57d/0xa90 [btrfs] [ 6712.206999] btrfs_rm_device+0x627/0x850 [btrfs] [ 6712.211800] btrfs_ioctl+0x2b03/0x3120 [btrfs] Reason for this is that btrfs_shrink_device adds the resized device to the fs_devices::resized_devices after it has called the last commit transaction. So the list fs_devices::resized_devices is not empty when btrfs_shrink_device returns. Now the parent function btrfs_rm_device calls: btrfs_close_bdev(device); call_rcu(&device->rcu, free_device_rcu); and then does the transactio ncommit. It goes through the fs_devices::resized_devices in btrfs_update_commit_device_size and leads to use-after-free. Fix this by making sure btrfs_shrink_device calls the last needed btrfs_commit_transaction before the return. This is consistent with what the grow counterpart does and this makes sure the on-disk state is persistent when the function returns. Reported-by: Lu Fengqi Tested-by: Lu Fengqi Signed-off-by: Anand Jain Reviewed-by: David Sterba [ update changelog ] Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index da86706123ff..f4405e430da6 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -4491,7 +4491,12 @@ again: /* Now btrfs_update_device() will change the on-disk size. */ ret = btrfs_update_device(trans, device); - btrfs_end_transaction(trans); + if (ret < 0) { + btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); + } else { + ret = btrfs_commit_transaction(trans); + } done: btrfs_free_path(path); if (ret) { -- cgit v1.2.3 From b9b8a41adeff5666b402996020b698504c927353 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 20 Aug 2018 11:25:33 +0300 Subject: btrfs: use after free in btrfs_quota_enable The issue here is that btrfs_commit_transaction() frees "trans" on both the error and the success path. So the problem would be if btrfs_commit_transaction() succeeds, and then qgroup_rescan_init() fails. That means that "ret" is non-zero and "trans" is non-NULL and it leads to a use after free inside the btrfs_end_transaction() macro. Fixes: 340f1aa27f36 ("btrfs: qgroups: Move transaction management inside btrfs_quota_enable/disable") Signed-off-by: Dan Carpenter Reviewed-by: Nikolay Borisov Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 4353bb69bb86..d4917c0cddf5 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1019,10 +1019,9 @@ out_add_root: spin_unlock(&fs_info->qgroup_lock); ret = btrfs_commit_transaction(trans); - if (ret) { - trans = NULL; + trans = NULL; + if (ret) goto out_free_path; - } ret = qgroup_rescan_init(fs_info, 0, 1); if (!ret) { -- cgit v1.2.3 From b6fdfbff078975c53383fc146a2a54985eab6b6d Mon Sep 17 00:00:00 2001 From: Misono Tomohiro Date: Fri, 24 Aug 2018 11:35:28 +0900 Subject: btrfs: Fix suspicious RCU usage warning in btrfs_debug_in_rcu Commit 672d599041c8 ("btrfs: Use wrapper macro for rcu string to remove duplicate code") replaces some open coded RCU string handling with macro. It turns out that btrfs_debug_in_rcu() is used for the first time and the macro lacks lock/unlock of RCU string for non-debug case (i.e. when the message is not printed), leading to suspicious RCU usage warning when CONFIG_PROVE_RCU is on. Fix this by adding a wrapper to call lock/unlock for the non-debug case too. Fixes: 672d599041c8 ("btrfs: Use wrapper macro for rcu string to remove duplicate code") Reported-by: David Howells Tested-by: David Howells Signed-off-by: Misono Tomohiro Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index a67cc190a84b..0b856da2fd3b 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -3390,9 +3390,9 @@ do { \ #define btrfs_debug(fs_info, fmt, args...) \ btrfs_no_printk(fs_info, KERN_DEBUG fmt, ##args) #define btrfs_debug_in_rcu(fs_info, fmt, args...) \ - btrfs_no_printk(fs_info, KERN_DEBUG fmt, ##args) + btrfs_no_printk_in_rcu(fs_info, KERN_DEBUG fmt, ##args) #define btrfs_debug_rl_in_rcu(fs_info, fmt, args...) \ - btrfs_no_printk(fs_info, KERN_DEBUG fmt, ##args) + btrfs_no_printk_in_rcu(fs_info, KERN_DEBUG fmt, ##args) #define btrfs_debug_rl(fs_info, fmt, args...) \ btrfs_no_printk(fs_info, KERN_DEBUG fmt, ##args) #endif @@ -3404,6 +3404,13 @@ do { \ rcu_read_unlock(); \ } while (0) +#define btrfs_no_printk_in_rcu(fs_info, fmt, args...) \ +do { \ + rcu_read_lock(); \ + btrfs_no_printk(fs_info, fmt, ##args); \ + rcu_read_unlock(); \ +} while (0) + #define btrfs_printk_ratelimited(fs_info, fmt, args...) \ do { \ static DEFINE_RATELIMIT_STATE(_rs, \ -- cgit v1.2.3