From cffc3374e567ef42954f3c7070b3fa83f20f9684 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 9 Jul 2015 13:13:44 +0100 Subject: Btrfs: fix order by which delayed references are run When we have an extent that got N references removed and N new references added in the same transaction, we must run the insertion of the references first because otherwise the last removed reference will remove the extent item from the extent tree, resulting in a failure for the insertions. This is a regression introduced in the 4.2-rc1 release and this fix just brings back the behaviour of selecting reference additions before any reference removals. The following test case for fstests reproduces the issue: seq=`basename $0` seqres=$RESULT_DIR/$seq echo "QA output created by $seq" tmp=/tmp/$$ status=1 # failure is the default! trap "_cleanup; exit \$status" 0 1 2 3 15 _cleanup() { _cleanup_flakey rm -f $tmp.* } # get standard environment, filters and checks . ./common/rc . ./common/filter . ./common/dmflakey # real QA test starts here _need_to_be_root _supported_fs btrfs _supported_os Linux _require_scratch _require_dm_flakey _require_cloner _require_metadata_journaling $SCRATCH_DEV rm -f $seqres.full _scratch_mkfs >>$seqres.full 2>&1 _init_flakey _mount_flakey # Create prealloc extent covering range [160K, 620K[ $XFS_IO_PROG -f -c "falloc 160K 460K" $SCRATCH_MNT/foo # Now write to the last 80K of the prealloc extent plus 40K to the unallocated # space that immediately follows it. This creates a new extent of 40K that spans # the range [620K, 660K[. $XFS_IO_PROG -c "pwrite -S 0xaa 540K 120K" $SCRATCH_MNT/foo | _filter_xfs_io # At this point, there are now 2 back references to the prealloc extent in our # extent tree. Both are for our file offset 160K and one relates to a file # extent item with a data offset of 0 and a length of 380K, while the other # relates to a file extent item with a data offset of 380K and a length of 80K. # Make sure everything done so far is durably persisted (all back references are # in the extent tree, etc). sync # Now clone all extents of our file that cover the offset 160K up to its eof # (660K at this point) into itself at offset 2M. This leaves a hole in the file # covering the range [660K, 2M[. The prealloc extent will now be referenced by # the file twice, once for offset 160K and once for offset 2M. The 40K extent # that follows the prealloc extent will also be referenced twice by our file, # once for offset 620K and once for offset 2M + 460K. $CLONER_PROG -s $((160 * 1024)) -d $((2 * 1024 * 1024)) -l 0 $SCRATCH_MNT/foo \ $SCRATCH_MNT/foo # Now create one new extent in our file with a size of 100Kb. It will span the # range [3M, 3M + 100K[. It also will cause creation of a hole spanning the # range [2M + 460K, 3M[. Our new file size is 3M + 100K. $XFS_IO_PROG -c "pwrite -S 0xbb 3M 100K" $SCRATCH_MNT/foo | _filter_xfs_io # At this point, there are now (in memory) 4 back references to the prealloc # extent. # # Two of them are for file offset 160K, related to file extent items # matching the file offsets 160K and 540K respectively, with data offsets of # 0 and 380K respectively, and with lengths of 380K and 80K respectively. # # The other two references are for file offset 2M, related to file extent items # matching the file offsets 2M and 2M + 380K respectively, with data offsets of # 0 and 380K respectively, and with lengths of 389K and 80K respectively. # # The 40K extent has 2 back references, one for file offset 620K and the other # for file offset 2M + 460K. # # The 100K extent has a single back reference and it relates to file offset 3M. # Now clone our 100K extent into offset 600K. That offset covers the last 20K # of the prealloc extent, the whole 40K extent and 40K of the hole starting at # offset 660K. $CLONER_PROG -s $((3 * 1024 * 1024)) -d $((600 * 1024)) -l $((100 * 1024)) \ $SCRATCH_MNT/foo $SCRATCH_MNT/foo # At this point there's only one reference to the 40K extent, at file offset # 2M + 460K, we have 4 references for the prealloc extent (2 for file offset # 160K and 2 for file offset 2M) and 2 references for the 100K extent (1 for # file offset 3M and a new one for file offset 600K). # Now fsync our file to make all its new data and metadata updates are durably # persisted and present if a power failure/crash happens after a successful # fsync and before the next transaction commit. $XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo echo "File digest before power failure:" md5sum $SCRATCH_MNT/foo | _filter_scratch # Silently drop all writes and ummount to simulate a crash/power failure. _load_flakey_table $FLAKEY_DROP_WRITES _unmount_flakey # Allow writes again, mount to trigger log replay and validate file contents. # During log replay, the btrfs delayed references implementation used to run the # deletion of back references before the addition of new back references, which # made the addition fail as it didn't find the key in the extent tree that it # was looking for. The failure triggered by this test was related to the 40K # extent, which got 1 reference dropped and 1 reference added during the fsync # log replay - when running the delayed references at transaction commit time, # btrfs was applying the deletion before the insertion, resulting in a failure # of the insertion that ended up turning the fs into read-only mode. _load_flakey_table $FLAKEY_ALLOW_WRITES _mount_flakey echo "File digest after log replay:" md5sum $SCRATCH_MNT/foo | _filter_scratch _unmount_flakey status=0 exit This issue turned the filesystem into read-only mode (current transaction aborted) and produced the following traces: [ 8247.578385] ------------[ cut here ]------------ [ 8247.579947] WARNING: CPU: 0 PID: 11341 at fs/btrfs/extent-tree.c:1547 lookup_inline_extent_backref+0x17d/0x45d [btrfs]() (...) [ 8247.601697] Call Trace: [ 8247.602222] [] dump_stack+0x4f/0x7b [ 8247.604320] [] warn_slowpath_common+0xa1/0xbb [ 8247.605488] [] ? lookup_inline_extent_backref+0x17d/0x45d [btrfs] [ 8247.608226] [] lookup_inline_extent_backref+0x17d/0x45d [btrfs] [ 8247.617061] [] insert_inline_extent_backref+0x41/0xb2 [btrfs] [ 8247.621856] [] __btrfs_inc_extent_ref+0x8c/0x20a [btrfs] [ 8247.624366] [] __btrfs_run_delayed_refs+0xb0c/0xd49 [btrfs] [ 8247.626176] [] btrfs_run_delayed_refs+0x6d/0x1d4 [btrfs] [ 8247.627435] [] ? __cache_free+0x4a7/0x4b6 [ 8247.628531] [] btrfs_commit_transaction+0x4c/0xa20 [btrfs] (...) [ 8247.648430] ---[ end trace 2461e55f92c2ac2d ]--- [ 8247.727263] WARNING: CPU: 3 PID: 11341 at fs/btrfs/extent-tree.c:2771 btrfs_run_delayed_refs+0xa4/0x1d4 [btrfs]() [ 8247.728954] BTRFS: Transaction aborted (error -5) (...) [ 8247.760866] Call Trace: [ 8247.761534] [] dump_stack+0x4f/0x7b [ 8247.764271] [] warn_slowpath_common+0xa1/0xbb [ 8247.767582] [] ? btrfs_run_delayed_refs+0xa4/0x1d4 [btrfs] [ 8247.769373] [] warn_slowpath_fmt+0x46/0x48 [ 8247.770836] [] btrfs_run_delayed_refs+0xa4/0x1d4 [btrfs] [ 8247.772532] [] ? __cache_free+0x4a7/0x4b6 [ 8247.773664] [] btrfs_commit_transaction+0x4c/0xa20 [btrfs] [ 8247.775047] [] ? trace_hardirqs_on+0xd/0xf [ 8247.776176] [] ? kmem_cache_free+0x12b/0x189 [ 8247.777427] [] btrfs_recover_log_trees+0x2da/0x33d [btrfs] [ 8247.778575] [] ? replay_one_extent+0x4fc/0x4fc [btrfs] [ 8247.779838] [] open_ctree+0x1cc0/0x201a [btrfs] [ 8247.781020] [] ? register_shrinker+0x56/0x81 [ 8247.782285] [] btrfs_mount+0x5f0/0x734 [btrfs] (...) [ 8247.793394] ---[ end trace 2461e55f92c2ac2e ]--- [ 8247.794276] BTRFS: error (device dm-0) in btrfs_run_delayed_refs:2771: errno=-5 IO failure [ 8247.797335] BTRFS: error (device dm-0) in btrfs_replay_log:2375: errno=-5 IO failure (Failed to recover log tree) Fixes: c6fc24549960 ("btrfs: delayed-ref: Use list to replace the ref_root in ref_head.") Signed-off-by: Filipe Manana Acked-by: Qu Wenruo --- fs/btrfs/extent-tree.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 1c2bd1723e40..171312d51799 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -2296,9 +2296,22 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans, static inline struct btrfs_delayed_ref_node * select_delayed_ref(struct btrfs_delayed_ref_head *head) { + struct btrfs_delayed_ref_node *ref; + if (list_empty(&head->ref_list)) return NULL; + /* + * Select a delayed ref of type BTRFS_ADD_DELAYED_REF first. + * This is to prevent a ref count from going down to zero, which deletes + * the extent item from the extent tree, when there still are references + * to add, which would fail because they would not find the extent item. + */ + list_for_each_entry(ref, &head->ref_list, list) { + if (ref->action == BTRFS_ADD_DELAYED_REF) + return ref; + } + return list_entry(head->ref_list.next, struct btrfs_delayed_ref_node, list); } -- cgit v1.2.3 From 00d80e342c0f4f1990ab69f594ee1e2348e51dd9 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 20 Jul 2015 14:56:20 +0100 Subject: Btrfs: fix quick exhaustion of the system array in the superblock Omar reported that after commit 4fbcdf669454 ("Btrfs: fix -ENOSPC when finishing block group creation"), introduced in 4.2-rc1, the following test was failing due to exhaustion of the system array in the superblock: #!/bin/bash truncate -s 100T big.img mkfs.btrfs big.img mount -o loop big.img /mnt/loop num=5 sz=10T for ((i = 0; i < $num; i++)); do echo fallocate $i $sz fallocate -l $sz /mnt/loop/testfile$i done btrfs filesystem sync /mnt/loop for ((i = 0; i < $num; i++)); do echo rm $i rm /mnt/loop/testfile$i btrfs filesystem sync /mnt/loop done umount /mnt/loop This made btrfs_add_system_chunk() fail with -EFBIG due to excessive allocation of system block groups. This happened because the test creates a large number of data block groups per transaction and when committing the transaction we start the writeout of the block group caches for all the new new (dirty) block groups, which results in pre-allocating space for each block group's free space cache using the same transaction handle. That in turn often leads to creation of more block groups, and all get attached to the new_bgs list of the same transaction handle to the point of getting a list with over 1500 elements, and creation of new block groups leads to the need of reserving space in the chunk block reserve and often creating a new system block group too. So that made us quickly exhaust the chunk block reserve/system space info, because as of the commit mentioned before, we do reserve space for each new block group in the chunk block reserve, unlike before where we would not and would at most allocate one new system block group and therefore would only ensure that there was enough space in the system space info to allocate 1 new block group even if we ended up allocating thousands of new block groups using the same transaction handle. That worked most of the time because the computed required space at check_system_chunk() is very pessimistic (assumes a chunk tree height of BTRFS_MAX_LEVEL/8 and that all nodes/leafs in a path will be COWed and split) and since the updates to the chunk tree all happen at btrfs_create_pending_block_groups it is unlikely that a path needs to be COWed more than once (unless writepages() for the btree inode is called by mm in between) and that compensated for the need of creating any new nodes/leads in the chunk tree. So fix this by ensuring we don't accumulate a too large list of new block groups in a transaction's handles new_bgs list, inserting/updating the chunk tree for all accumulated new block groups and releasing the unused space from the chunk block reserve whenever the list becomes sufficiently large. This is a generic solution even though the problem currently can only happen when starting the writeout of the free space caches for all dirty block groups (btrfs_start_dirty_block_groups()). Reported-by: Omar Sandoval Signed-off-by: Filipe Manana Tested-by: Omar Sandoval Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 171312d51799..07204bf601ed 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4227,6 +4227,24 @@ out: space_info->chunk_alloc = 0; spin_unlock(&space_info->lock); mutex_unlock(&fs_info->chunk_mutex); + /* + * When we allocate a new chunk we reserve space in the chunk block + * reserve to make sure we can COW nodes/leafs in the chunk tree or + * add new nodes/leafs to it if we end up needing to do it when + * inserting the chunk item and updating device items as part of the + * second phase of chunk allocation, performed by + * btrfs_finish_chunk_alloc(). So make sure we don't accumulate a + * large number of new block groups to create in our transaction + * handle's new_bgs list to avoid exhausting the chunk block reserve + * in extreme cases - like having a single transaction create many new + * block groups when starting to write out the free space caches of all + * the block groups that were made dirty during the lifetime of the + * transaction. + */ + if (trans->chunk_bytes_reserved >= (2 * 1024 * 1024ull)) { + btrfs_create_pending_block_groups(trans, trans->root); + btrfs_trans_release_chunk_metadata(trans); + } return ret; } -- cgit v1.2.3 From d04c6b88320debb403bff8d8b634a1efa48b8d3d Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Mon, 15 Jun 2015 09:41:14 -0400 Subject: btrfs: make btrfs_issue_discard return bytes discarded Initially this will just be the length argument passed to it, but the following patches will adjust that to reflect re-alignment and skipped blocks. Signed-off-by: Jeff Mahoney Reviewed-by: Filipe Manana Tested-by: Filipe Manana Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 07204bf601ed..16655bb5f293 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1883,10 +1883,17 @@ static int remove_extent_backref(struct btrfs_trans_handle *trans, return ret; } -static int btrfs_issue_discard(struct block_device *bdev, - u64 start, u64 len) +static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len, + u64 *discarded_bytes) { - return blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_NOFS, 0); + int ret = 0; + + *discarded_bytes = 0; + ret = blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_NOFS, 0); + if (!ret) + *discarded_bytes = len; + + return ret; } int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr, @@ -1907,14 +1914,16 @@ int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr, for (i = 0; i < bbio->num_stripes; i++, stripe++) { + u64 bytes; if (!stripe->dev->can_discard) continue; ret = btrfs_issue_discard(stripe->dev->bdev, stripe->physical, - stripe->length); + stripe->length, + &bytes); if (!ret) - discarded_bytes += stripe->length; + discarded_bytes += bytes; else if (ret != -EOPNOTSUPP) break; /* Logic errors or -ENOMEM, or -EIO but I don't know how that could happen JDM */ -- cgit v1.2.3 From 4d89d377bbb0e34cd1571b57a984c2326cab69b5 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Mon, 15 Jun 2015 09:41:15 -0400 Subject: btrfs: btrfs_issue_discard ensure offset/length are aligned to sector boundaries It's possible, though unexpected, to pass unaligned offsets and lengths to btrfs_issue_discard. We then shift the offset/length values to sector units. If an unaligned offset has been passed, it will result in the entire sector being discarded, possibly losing data. An unaligned length is safe but we'll end up returning an inaccurate number of discarded bytes. This patch aligns the offset to the 512B boundary, adjusts the length, and warns, since we shouldn't be discarding on an offset that isn't aligned with our sector size. Signed-off-by: Jeff Mahoney Reviewed-by: Filipe Manana Tested-by: Filipe Manana Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 16655bb5f293..7aa6ad1f014d 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1887,12 +1887,21 @@ static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len, u64 *discarded_bytes) { int ret = 0; + u64 aligned_start = ALIGN(start, 1 << 9); - *discarded_bytes = 0; - ret = blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_NOFS, 0); - if (!ret) - *discarded_bytes = len; + if (WARN_ON(start != aligned_start)) { + len -= aligned_start - start; + len = round_down(len, 1 << 9); + start = aligned_start; + } + *discarded_bytes = 0; + if (len) { + ret = blkdev_issue_discard(bdev, start >> 9, len >> 9, + GFP_NOFS, 0); + if (!ret) + *discarded_bytes = len; + } return ret; } -- cgit v1.2.3 From 86557861dfe4f8defde0df40620b97cc60285aa4 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Mon, 15 Jun 2015 09:41:16 -0400 Subject: btrfs: skip superblocks during discard Btrfs doesn't track superblocks with extent records so there is nothing persistent on-disk to indicate that those blocks are in use. We track the superblocks in memory to ensure they don't get used by removing them from the free space cache when we load a block group from disk. Prior to 47ab2a6c6a (Btrfs: remove empty block groups automatically), that was fine since the block group would never be reclaimed so the superblock was always safe. Once we started removing the empty block groups, we were protected by the fact that discards weren't being properly issued for unused space either via FITRIM or -odiscard. The block groups were still being released, but the blocks remained on disk. In order to properly discard unused block groups, we need to filter out the superblocks from the discard range. Superblocks are located at fixed locations on each device, so it makes sense to filter them out in btrfs_issue_discard, which is used by both -odiscard and FITRIM. Signed-off-by: Jeff Mahoney Reviewed-by: Filipe Manana Tested-by: Filipe Manana Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 59 ++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 55 insertions(+), 4 deletions(-) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 7aa6ad1f014d..d763457b3cce 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1883,10 +1883,12 @@ static int remove_extent_backref(struct btrfs_trans_handle *trans, return ret; } +#define in_range(b, first, len) ((b) >= (first) && (b) < (first) + (len)) static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len, u64 *discarded_bytes) { - int ret = 0; + int j, ret = 0; + u64 bytes_left, end; u64 aligned_start = ALIGN(start, 1 << 9); if (WARN_ON(start != aligned_start)) { @@ -1896,11 +1898,60 @@ static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len, } *discarded_bytes = 0; - if (len) { - ret = blkdev_issue_discard(bdev, start >> 9, len >> 9, + + if (!len) + return 0; + + end = start + len; + bytes_left = len; + + /* Skip any superblocks on this device. */ + for (j = 0; j < BTRFS_SUPER_MIRROR_MAX; j++) { + u64 sb_start = btrfs_sb_offset(j); + u64 sb_end = sb_start + BTRFS_SUPER_INFO_SIZE; + u64 size = sb_start - start; + + if (!in_range(sb_start, start, bytes_left) && + !in_range(sb_end, start, bytes_left) && + !in_range(start, sb_start, BTRFS_SUPER_INFO_SIZE)) + continue; + + /* + * Superblock spans beginning of range. Adjust start and + * try again. + */ + if (sb_start <= start) { + start += sb_end - start; + if (start > end) { + bytes_left = 0; + break; + } + bytes_left = end - start; + continue; + } + + if (size) { + ret = blkdev_issue_discard(bdev, start >> 9, size >> 9, + GFP_NOFS, 0); + if (!ret) + *discarded_bytes += size; + else if (ret != -EOPNOTSUPP) + return ret; + } + + start = sb_end; + if (start > end) { + bytes_left = 0; + break; + } + bytes_left = end - start; + } + + if (bytes_left) { + ret = blkdev_issue_discard(bdev, start >> 9, bytes_left >> 9, GFP_NOFS, 0); if (!ret) - *discarded_bytes = len; + *discarded_bytes += bytes_left; } return ret; } -- cgit v1.2.3 From 499f377f49f085ee4aa214c738e948e88626f39b Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Mon, 15 Jun 2015 09:41:17 -0400 Subject: btrfs: iterate over unused chunk space in FITRIM Since we now clean up block groups automatically as they become empty, iterating over block groups is no longer sufficient to discard unused space. This patch iterates over the unused chunk space and discards any regions that are unallocated, regardless of whether they were ever used. This is a change for btrfs but is consistent with other file systems. We do this in a transactionless manner since the discard process can take a substantial amount of time and a transaction would need to be started before the acquisition of the device list lock. That would mean a transaction would be held open across /all/ of the discards collectively. In order to prevent other threads from allocating or freeing chunks, we hold the chunks lock across the search and discard calls. We release it between searches to allow the file system to perform more-or-less normally. Since the running transaction can commit and disappear while we're using the transaction pointer, we take a reference to it and release it after the search. This is safe since it would happen normally at the end of the transaction commit after any locks are released anyway. We also take the commit_root_sem to protect against a transaction starting and committing while we're running. Signed-off-by: Jeff Mahoney Reviewed-by: Filipe Manana Tested-by: Filipe Manana Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 101 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 101 insertions(+) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index d763457b3cce..15411aefbfa0 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -10135,10 +10135,99 @@ int btrfs_error_unpin_extent_range(struct btrfs_root *root, u64 start, u64 end) return unpin_extent_range(root, start, end, false); } +/* + * It used to be that old block groups would be left around forever. + * Iterating over them would be enough to trim unused space. Since we + * now automatically remove them, we also need to iterate over unallocated + * space. + * + * We don't want a transaction for this since the discard may take a + * substantial amount of time. We don't require that a transaction be + * running, but we do need to take a running transaction into account + * to ensure that we're not discarding chunks that were released in + * the current transaction. + * + * Holding the chunks lock will prevent other threads from allocating + * or releasing chunks, but it won't prevent a running transaction + * from committing and releasing the memory that the pending chunks + * list head uses. For that, we need to take a reference to the + * transaction. + */ +static int btrfs_trim_free_extents(struct btrfs_device *device, + u64 minlen, u64 *trimmed) +{ + u64 start = 0, len = 0; + int ret; + + *trimmed = 0; + + /* Not writeable = nothing to do. */ + if (!device->writeable) + return 0; + + /* No free space = nothing to do. */ + if (device->total_bytes <= device->bytes_used) + return 0; + + ret = 0; + + while (1) { + struct btrfs_fs_info *fs_info = device->dev_root->fs_info; + struct btrfs_transaction *trans; + u64 bytes; + + ret = mutex_lock_interruptible(&fs_info->chunk_mutex); + if (ret) + return ret; + + down_read(&fs_info->commit_root_sem); + + spin_lock(&fs_info->trans_lock); + trans = fs_info->running_transaction; + if (trans) + atomic_inc(&trans->use_count); + spin_unlock(&fs_info->trans_lock); + + ret = find_free_dev_extent_start(trans, device, minlen, start, + &start, &len); + if (trans) + btrfs_put_transaction(trans); + + if (ret) { + up_read(&fs_info->commit_root_sem); + mutex_unlock(&fs_info->chunk_mutex); + if (ret == -ENOSPC) + ret = 0; + break; + } + + ret = btrfs_issue_discard(device->bdev, start, len, &bytes); + up_read(&fs_info->commit_root_sem); + mutex_unlock(&fs_info->chunk_mutex); + + if (ret) + break; + + start += len; + *trimmed += bytes; + + if (fatal_signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + + cond_resched(); + } + + return ret; +} + int btrfs_trim_fs(struct btrfs_root *root, struct fstrim_range *range) { struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_block_group_cache *cache = NULL; + struct btrfs_device *device; + struct list_head *devices; u64 group_trimmed; u64 start; u64 end; @@ -10193,6 +10282,18 @@ int btrfs_trim_fs(struct btrfs_root *root, struct fstrim_range *range) cache = next_block_group(fs_info->tree_root, cache); } + mutex_lock(&root->fs_info->fs_devices->device_list_mutex); + devices = &root->fs_info->fs_devices->alloc_list; + list_for_each_entry(device, devices, dev_alloc_list) { + ret = btrfs_trim_free_extents(device, range->minlen, + &group_trimmed); + if (ret) + break; + + trimmed += group_trimmed; + } + mutex_unlock(&root->fs_info->fs_devices->device_list_mutex); + range->len = trimmed; return ret; } -- cgit v1.2.3 From e33e17ee1098d8d751552ac11c111e1c1a3db014 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Mon, 15 Jun 2015 09:41:19 -0400 Subject: btrfs: add missing discards when unpinning extents with -o discard When we clear the dirty bits in btrfs_delete_unused_bgs for extents in the empty block group, it results in btrfs_finish_extent_commit being unable to discard the freed extents. The block group removal patch added an alternate path to forget extents other than btrfs_finish_extent_commit. As a result, any extents that would be freed when the block group is removed aren't discarded. In my test run, with a large copy of mixed sized files followed by removal, it left nearly 2/3 of extents undiscarded. To clean up the block groups, we add the removed block group onto a list that will be discarded after transaction commit. Signed-off-by: Jeff Mahoney Reviewed-by: Filipe Manana Tested-by: Filipe Manana Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 68 +++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 64 insertions(+), 4 deletions(-) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 15411aefbfa0..6b791f394698 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -6131,20 +6131,19 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans, struct btrfs_root *root) { struct btrfs_fs_info *fs_info = root->fs_info; + struct btrfs_block_group_cache *block_group, *tmp; + struct list_head *deleted_bgs; struct extent_io_tree *unpin; u64 start; u64 end; int ret; - if (trans->aborted) - return 0; - if (fs_info->pinned_extents == &fs_info->freed_extents[0]) unpin = &fs_info->freed_extents[1]; else unpin = &fs_info->freed_extents[0]; - while (1) { + while (!trans->aborted) { mutex_lock(&fs_info->unused_bg_unpin_mutex); ret = find_first_extent_bit(unpin, 0, &start, &end, EXTENT_DIRTY, NULL); @@ -6163,6 +6162,34 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans, cond_resched(); } + /* + * Transaction is finished. We don't need the lock anymore. We + * do need to clean up the block groups in case of a transaction + * abort. + */ + deleted_bgs = &trans->transaction->deleted_bgs; + list_for_each_entry_safe(block_group, tmp, deleted_bgs, bg_list) { + u64 trimmed = 0; + + ret = -EROFS; + if (!trans->aborted) + ret = btrfs_discard_extent(root, + block_group->key.objectid, + block_group->key.offset, + &trimmed); + + list_del_init(&block_group->bg_list); + btrfs_put_block_group_trimming(block_group); + btrfs_put_block_group(block_group); + + if (ret) { + const char *errstr = btrfs_decode_error(ret); + btrfs_warn(fs_info, + "Discard failed while removing blockgroup: errno=%d %s\n", + ret, errstr); + } + } + return 0; } @@ -9903,6 +9930,11 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, * currently running transaction might finish and a new one start, * allowing for new block groups to be created that can reuse the same * physical device locations unless we take this special care. + * + * There may also be an implicit trim operation if the file system + * is mounted with -odiscard. The same protections must remain + * in place until the extents have been discarded completely when + * the transaction commit has completed. */ remove_em = (atomic_read(&block_group->trimming) == 0); /* @@ -9977,6 +10009,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) spin_lock(&fs_info->unused_bgs_lock); while (!list_empty(&fs_info->unused_bgs)) { u64 start, end; + int trimming; block_group = list_first_entry(&fs_info->unused_bgs, struct btrfs_block_group_cache, @@ -10076,12 +10109,39 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) spin_unlock(&block_group->lock); spin_unlock(&space_info->lock); + /* DISCARD can flip during remount */ + trimming = btrfs_test_opt(root, DISCARD); + + /* Implicit trim during transaction commit. */ + if (trimming) + btrfs_get_block_group_trimming(block_group); + /* * Btrfs_remove_chunk will abort the transaction if things go * horribly wrong. */ ret = btrfs_remove_chunk(trans, root, block_group->key.objectid); + + if (ret) { + if (trimming) + btrfs_put_block_group_trimming(block_group); + goto end_trans; + } + + /* + * If we're not mounted with -odiscard, we can just forget + * about this block group. Otherwise we'll need to wait + * until transaction commit to do the actual discard. + */ + if (trimming) { + WARN_ON(!list_empty(&block_group->bg_list)); + spin_lock(&trans->transaction->deleted_bgs_lock); + list_move(&block_group->bg_list, + &trans->transaction->deleted_bgs); + spin_unlock(&trans->transaction->deleted_bgs_lock); + btrfs_get_block_group(block_group); + } end_trans: btrfs_end_transaction(trans, root); next: -- cgit v1.2.3 From 868f401ae38acb439005626c04d575e64c5ae760 Mon Sep 17 00:00:00 2001 From: Zhaolei Date: Wed, 5 Aug 2015 16:43:27 +0800 Subject: btrfs: Use ref_cnt for set_block_group_ro() More than one code call set_block_group_ro() and restore rw in fail. Old code use bool bit to save blockgroup's ro state, it can not support parallel case(it is confirmd exist in my debug log). This patch use ref count to store ro state, and rename set_block_group_ro/set_block_group_rw to inc_block_group_ro/dec_block_group_ro. Signed-off-by: Zhao Lei Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 42 +++++++++++++++++++++--------------------- 1 file changed, 21 insertions(+), 21 deletions(-) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 07204bf601ed..5cefa02b40a9 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -8723,14 +8723,13 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags) return flags; } -static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force) +static int inc_block_group_ro(struct btrfs_block_group_cache *cache, int force) { struct btrfs_space_info *sinfo = cache->space_info; u64 num_bytes; u64 min_allocable_bytes; int ret = -ENOSPC; - /* * We need some metadata space and system metadata space for * allocating chunks in some corner cases until we force to set @@ -8747,6 +8746,7 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force) spin_lock(&cache->lock); if (cache->ro) { + cache->ro++; ret = 0; goto out; } @@ -8758,7 +8758,7 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force) sinfo->bytes_may_use + sinfo->bytes_readonly + num_bytes + min_allocable_bytes <= sinfo->total_bytes) { sinfo->bytes_readonly += num_bytes; - cache->ro = 1; + cache->ro++; list_add_tail(&cache->ro_list, &sinfo->ro_bgs); ret = 0; } @@ -8768,7 +8768,7 @@ out: return ret; } -int btrfs_set_block_group_ro(struct btrfs_root *root, +int btrfs_inc_block_group_ro(struct btrfs_root *root, struct btrfs_block_group_cache *cache) { @@ -8776,8 +8776,6 @@ int btrfs_set_block_group_ro(struct btrfs_root *root, u64 alloc_flags; int ret; - BUG_ON(cache->ro); - again: trans = btrfs_join_transaction(root); if (IS_ERR(trans)) @@ -8820,7 +8818,7 @@ again: goto out; } - ret = set_block_group_ro(cache, 0); + ret = inc_block_group_ro(cache, 0); if (!ret) goto out; alloc_flags = get_alloc_profile(root, cache->space_info->flags); @@ -8828,7 +8826,7 @@ again: CHUNK_ALLOC_FORCE); if (ret < 0) goto out; - ret = set_block_group_ro(cache, 0); + ret = inc_block_group_ro(cache, 0); out: if (cache->flags & BTRFS_BLOCK_GROUP_SYSTEM) { alloc_flags = update_block_group_flags(root, cache->flags); @@ -8891,7 +8889,7 @@ u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo) return free_bytes; } -void btrfs_set_block_group_rw(struct btrfs_root *root, +void btrfs_dec_block_group_ro(struct btrfs_root *root, struct btrfs_block_group_cache *cache) { struct btrfs_space_info *sinfo = cache->space_info; @@ -8901,11 +8899,13 @@ void btrfs_set_block_group_rw(struct btrfs_root *root, spin_lock(&sinfo->lock); spin_lock(&cache->lock); - num_bytes = cache->key.offset - cache->reserved - cache->pinned - - cache->bytes_super - btrfs_block_group_used(&cache->item); - sinfo->bytes_readonly -= num_bytes; - cache->ro = 0; - list_del_init(&cache->ro_list); + if (!--cache->ro) { + num_bytes = cache->key.offset - cache->reserved - + cache->pinned - cache->bytes_super - + btrfs_block_group_used(&cache->item); + sinfo->bytes_readonly -= num_bytes; + list_del_init(&cache->ro_list); + } spin_unlock(&cache->lock); spin_unlock(&sinfo->lock); } @@ -9421,7 +9421,7 @@ int btrfs_read_block_groups(struct btrfs_root *root) set_avail_alloc_bits(root->fs_info, cache->flags); if (btrfs_chunk_readonly(root, cache->key.objectid)) { - set_block_group_ro(cache, 1); + inc_block_group_ro(cache, 1); } else if (btrfs_block_group_used(&cache->item) == 0) { spin_lock(&info->unused_bgs_lock); /* Should always be true but just in case. */ @@ -9449,11 +9449,11 @@ int btrfs_read_block_groups(struct btrfs_root *root) list_for_each_entry(cache, &space_info->block_groups[BTRFS_RAID_RAID0], list) - set_block_group_ro(cache, 1); + inc_block_group_ro(cache, 1); list_for_each_entry(cache, &space_info->block_groups[BTRFS_RAID_SINGLE], list) - set_block_group_ro(cache, 1); + inc_block_group_ro(cache, 1); } init_global_block_rsv(info); @@ -9941,7 +9941,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) spin_unlock(&block_group->lock); /* We don't want to force the issue, only flip if it's ok. */ - ret = set_block_group_ro(block_group, 0); + ret = inc_block_group_ro(block_group, 0); up_write(&space_info->groups_sem); if (ret < 0) { ret = 0; @@ -9955,7 +9955,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) /* 1 for btrfs_orphan_reserve_metadata() */ trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { - btrfs_set_block_group_rw(root, block_group); + btrfs_dec_block_group_ro(root, block_group); ret = PTR_ERR(trans); goto next; } @@ -9982,14 +9982,14 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) EXTENT_DIRTY, GFP_NOFS); if (ret) { mutex_unlock(&fs_info->unused_bg_unpin_mutex); - btrfs_set_block_group_rw(root, block_group); + btrfs_dec_block_group_ro(root, block_group); goto end_trans; } ret = clear_extent_bits(&fs_info->freed_extents[1], start, end, EXTENT_DIRTY, GFP_NOFS); if (ret) { mutex_unlock(&fs_info->unused_bg_unpin_mutex); - btrfs_set_block_group_rw(root, block_group); + btrfs_dec_block_group_ro(root, block_group); goto end_trans; } mutex_unlock(&fs_info->unused_bg_unpin_mutex); -- cgit v1.2.3 From d02207512df120a37a8f1e50a5f90c555266282c Mon Sep 17 00:00:00 2001 From: Zhaolei Date: Thu, 6 Aug 2015 22:16:23 +0800 Subject: btrfs: Fix wrong comment of btrfs_alloc_tree_block() These wrong comment was copyed from another function(expired) from init, this patch fixed them. Signed-off-by: Zhao Lei Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 5cefa02b40a9..e2ad72b241ad 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -7567,9 +7567,6 @@ static void unuse_block_rsv(struct btrfs_fs_info *fs_info, /* * finds a free extent and does all the dirty work required for allocation - * returns the key for the extent through ins, and a tree buffer for - * the first block of the extent through buf. - * * returns the tree buffer or an ERR_PTR on error. */ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, -- cgit v1.2.3 From 9ed0dea09fc5e8cb1926c15fc986a46a2db43da6 Mon Sep 17 00:00:00 2001 From: Zhaolei Date: Thu, 6 Aug 2015 22:16:24 +0800 Subject: btrfs: Remove root argument in extent_data_ref_count() Because it is never used. Signed-off-by: Zhao Lei Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs/btrfs/extent-tree.c') diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index e2ad72b241ad..59d59d98bca1 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1316,8 +1316,7 @@ static noinline int remove_extent_data_ref(struct btrfs_trans_handle *trans, return ret; } -static noinline u32 extent_data_ref_count(struct btrfs_root *root, - struct btrfs_path *path, +static noinline u32 extent_data_ref_count(struct btrfs_path *path, struct btrfs_extent_inline_ref *iref) { struct btrfs_key key; @@ -6349,7 +6348,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, } else { if (found_extent) { BUG_ON(is_data && refs_to_drop != - extent_data_ref_count(root, path, iref)); + extent_data_ref_count(path, iref)); if (iref) { BUG_ON(path->slots[0] != extent_slot); } else { -- cgit v1.2.3