diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2023-02-20 12:54:27 -0800 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2023-02-20 12:54:27 -0800 |
| commit | 885ce48739189fac6645ff42d736ee0de0b5917d (patch) | |
| tree | d69300909ff5b4a0291a7311cd25cf90801a650a /fs/btrfs/tree-log.c | |
| parent | 274978f173276c5720a3cd8d0b6047d2c0d3a684 (diff) | |
| parent | 964a54e5e1a0d70cd80bd5a0885a1938463625b1 (diff) | |
| download | lwn-885ce48739189fac6645ff42d736ee0de0b5917d.tar.gz lwn-885ce48739189fac6645ff42d736ee0de0b5917d.zip | |
Merge tag 'for-6.3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs updates from David Sterba:
"The usual mix of performance improvements and new features.
The core change is reworking how checksums are processed, with
followup cleanups and simplifications. There are two minor changes in
block layer and iomap code.
Features:
- block group allocation class heuristics:
- pack files by size (up to 128k, up to 8M, more) to avoid
fragmentation in block groups, assuming that file size and life
time is correlated, in particular this may help during balance
- with tracepoints and extensible in the future
Performance:
- send: cache directory utimes and only emit the command when
necessary
- speedup up to 10x
- smaller final stream produced (no redundant utimes commands
issued)
- compatibility not affected
- fiemap: skip backref checks for shared leaves
- speedup 3x on sample filesystem with all leaves shared (e.g. on
snapshots)
- micro optimized b-tree key lookup, speedup in metadata operations
(sample benchmark: fs_mark +10% of files/sec)
Core changes:
- change where checksumming is done in the io path:
- checksum and read repair does verification at lower layer
- cascaded cleanups and simplifications
- raid56 refactoring and cleanups
Fixes:
- sysfs: make sure that a run-time change of a feature is correctly
tracked by the feature files
- scrub: better reporting of tree block errors
Other:
- locally enable -Wmaybe-uninitialized after fixing all warnings
- misc cleanups, spelling fixes
Other code:
- block: export bio_split_rw
- iomap: remove IOMAP_F_ZONE_APPEND"
* tag 'for-6.3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (109 commits)
btrfs: make kobj_type structures constant
btrfs: remove the bdev argument to btrfs_rmap_block
btrfs: don't rely on unchanging ->bi_bdev for zone append remaps
btrfs: never return true for reads in btrfs_use_zone_append
btrfs: pass a btrfs_bio to btrfs_use_append
btrfs: set bbio->file_offset in alloc_new_bio
btrfs: use file_offset to limit bios size in calc_bio_boundaries
btrfs: do unsigned integer division in the extent buffer binary search loop
btrfs: eliminate extra call when doing binary search on extent buffer
btrfs: raid56: handle endio in scrub_rbio
btrfs: raid56: handle endio in recover_rbio
btrfs: raid56: handle endio in rmw_rbio
btrfs: raid56: submit the read bios from scrub_assemble_read_bios
btrfs: raid56: fold rmw_read_wait_recover into rmw_read_bios
btrfs: raid56: fold recover_assemble_read_bios into recover_rbio
btrfs: raid56: add a bio_list_put helper
btrfs: raid56: wait for I/O completion in submit_read_bios
btrfs: raid56: simplify code flow in rmw_rbio
btrfs: raid56: simplify error handling and code flow in raid56_parity_write
btrfs: replace btrfs_wait_tree_block_writeback by wait_on_extent_buffer_writeback
...
Diffstat (limited to 'fs/btrfs/tree-log.c')
| -rw-r--r-- | fs/btrfs/tree-log.c | 87 |
1 files changed, 36 insertions, 51 deletions
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 58599189bd18..200cea6e49e5 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -279,12 +279,6 @@ void btrfs_end_log_trans(struct btrfs_root *root) } } -static void btrfs_wait_tree_block_writeback(struct extent_buffer *buf) -{ - filemap_fdatawait_range(buf->pages[0]->mapping, - buf->start, buf->start + buf->len - 1); -} - /* * the walk control struct is used to pass state down the chain when * processing the log tree. The stage field tells us which part @@ -2623,11 +2617,12 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, return ret; } + btrfs_tree_lock(next); + btrfs_clear_buffer_dirty(trans, next); + wait_on_extent_buffer_writeback(next); + btrfs_tree_unlock(next); + if (trans) { - btrfs_tree_lock(next); - btrfs_clean_tree_block(next); - btrfs_wait_tree_block_writeback(next); - btrfs_tree_unlock(next); ret = btrfs_pin_reserved_extent(trans, bytenr, blocksize); if (ret) { @@ -2637,8 +2632,6 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, btrfs_redirty_list_add( trans->transaction, next); } else { - if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) - clear_extent_buffer_dirty(next); unaccount_log_buffer(fs_info, bytenr); } } @@ -2693,11 +2686,12 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, next = path->nodes[*level]; + btrfs_tree_lock(next); + btrfs_clear_buffer_dirty(trans, next); + wait_on_extent_buffer_writeback(next); + btrfs_tree_unlock(next); + if (trans) { - btrfs_tree_lock(next); - btrfs_clean_tree_block(next); - btrfs_wait_tree_block_writeback(next); - btrfs_tree_unlock(next); ret = btrfs_pin_reserved_extent(trans, path->nodes[*level]->start, path->nodes[*level]->len); @@ -2706,9 +2700,6 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, btrfs_redirty_list_add(trans->transaction, next); } else { - if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) - clear_extent_buffer_dirty(next); - unaccount_log_buffer(fs_info, path->nodes[*level]->start); } @@ -2776,19 +2767,18 @@ static int walk_log_tree(struct btrfs_trans_handle *trans, next = path->nodes[orig_level]; + btrfs_tree_lock(next); + btrfs_clear_buffer_dirty(trans, next); + wait_on_extent_buffer_writeback(next); + btrfs_tree_unlock(next); + if (trans) { - btrfs_tree_lock(next); - btrfs_clean_tree_block(next); - btrfs_wait_tree_block_writeback(next); - btrfs_tree_unlock(next); ret = btrfs_pin_reserved_extent(trans, next->start, next->len); if (ret) goto out; btrfs_redirty_list_add(trans->transaction, next); } else { - if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) - clear_extent_buffer_dirty(next); unaccount_log_buffer(fs_info, next->start); } } @@ -3652,11 +3642,10 @@ static int flush_dir_items_batch(struct btrfs_trans_handle *trans, /* * If for some unexpected reason the last item's index is not greater - * than the last index we logged, warn and return an error to fallback - * to a transaction commit. + * than the last index we logged, warn and force a transaction commit. */ if (WARN_ON(last_index <= inode->last_dir_index_offset)) - ret = -EUCLEAN; + ret = BTRFS_LOG_FORCE_COMMIT; else inode->last_dir_index_offset = last_index; out: @@ -3794,7 +3783,6 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans, struct btrfs_key min_key; struct btrfs_root *root = inode->root; struct btrfs_root *log = root->log_root; - int err = 0; int ret; u64 last_old_dentry_offset = min_offset - 1; u64 last_offset = (u64)-1; @@ -3835,8 +3823,8 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans, path->slots[0]); if (tmp.type == BTRFS_DIR_INDEX_KEY) last_old_dentry_offset = tmp.offset; - } else if (ret < 0) { - err = ret; + } else if (ret > 0) { + ret = 0; } goto done; @@ -3859,7 +3847,6 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans, if (tmp.type == BTRFS_DIR_INDEX_KEY) last_old_dentry_offset = tmp.offset; } else if (ret < 0) { - err = ret; goto done; } @@ -3881,12 +3868,15 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans, */ search: ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0); - if (ret > 0) + if (ret > 0) { ret = btrfs_next_item(root, path); + if (ret > 0) { + /* There are no more keys in the inode's root. */ + ret = 0; + goto done; + } + } if (ret < 0) - err = ret; - /* If ret is 1, there are no more keys in the inode's root. */ - if (ret != 0) goto done; /* @@ -3897,8 +3887,8 @@ search: ret = process_dir_items_leaf(trans, inode, path, dst_path, ctx, &last_old_dentry_offset); if (ret != 0) { - if (ret < 0) - err = ret; + if (ret > 0) + ret = 0; goto done; } path->slots[0] = btrfs_header_nritems(path->nodes[0]); @@ -3909,10 +3899,10 @@ search: */ ret = btrfs_next_leaf(root, path); if (ret) { - if (ret == 1) + if (ret == 1) { last_offset = (u64)-1; - else - err = ret; + ret = 0; + } goto done; } btrfs_item_key_to_cpu(path->nodes[0], &min_key, path->slots[0]); @@ -3943,7 +3933,7 @@ done: btrfs_release_path(path); btrfs_release_path(dst_path); - if (err == 0) { + if (ret == 0) { *last_offset_ret = last_offset; /* * In case the leaf was changed in the current transaction but @@ -3954,15 +3944,13 @@ done: * a range, last_old_dentry_offset is == to last_offset. */ ASSERT(last_old_dentry_offset <= last_offset); - if (last_old_dentry_offset < last_offset) { + if (last_old_dentry_offset < last_offset) ret = insert_dir_log_key(trans, log, path, ino, last_old_dentry_offset + 1, last_offset); - if (ret) - err = ret; - } } - return err; + + return ret; } /* @@ -5604,10 +5592,8 @@ static int add_conflicting_inode(struct btrfs_trans_handle *trans, * LOG_INODE_EXISTS mode) and slow down other fsyncs or transaction * commits. */ - if (ctx->num_conflict_inodes >= MAX_CONFLICT_INODES) { - btrfs_set_log_full_commit(trans); + if (ctx->num_conflict_inodes >= MAX_CONFLICT_INODES) return BTRFS_LOG_FORCE_COMMIT; - } inode = btrfs_iget(root->fs_info->sb, ino, root); /* @@ -6466,7 +6452,6 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, * result in losing the file after a log replay. */ if (full_dir_logging && inode->last_unlink_trans >= trans->transid) { - btrfs_set_log_full_commit(trans); ret = BTRFS_LOG_FORCE_COMMIT; goto out_unlock; } |