From c0247d289e73e18f6ddb0895de30c09770fbed95 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Tue, 13 Aug 2024 12:53:15 +0200 Subject: [PATCH 1/5] btrfs: send: annotate struct name_cache_entry with __counted_by() Add the __counted_by compiler attribute to the flexible array member name to improve access bounds-checking via CONFIG_UBSAN_BOUNDS and CONFIG_FORTIFY_SOURCE. Signed-off-by: Thorsten Blum Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/send.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 7fc692fc76e1..619fa0b8b3f6 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -347,7 +347,7 @@ struct name_cache_entry { int ret; int need_later_update; int name_len; - char name[]; + char name[] __counted_by(name_len); }; /* See the comment at lru_cache.h about struct btrfs_lru_cache_entry. */ From 3bc2ac2f8f0b78a13140fc72022771efe0c9b778 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 12 Aug 2024 12:30:52 -0400 Subject: [PATCH 2/5] btrfs: update target inode's ctime on unlink Unlink changes the link count on the target inode. POSIX mandates that the ctime must also change when this occurs. According to https://pubs.opengroup.org/onlinepubs/9699919799/functions/unlink.html: "Upon successful completion, unlink() shall mark for update the last data modification and last file status change timestamps of the parent directory. Also, if the file's link count is not 0, the last file status change timestamp of the file shall be marked for update." Signed-off-by: Jeff Layton Reviewed-by: David Sterba [ add link to the opengroup docs ] Signed-off-by: David Sterba --- fs/btrfs/inode.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 333b0e8587a2..b1b6564ab68f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4195,6 +4195,7 @@ err: btrfs_i_size_write(dir, dir->vfs_inode.i_size - name->len * 2); inode_inc_iversion(&inode->vfs_inode); + inode_set_ctime_current(&inode->vfs_inode); inode_inc_iversion(&dir->vfs_inode); inode_set_mtime_to_ts(&dir->vfs_inode, inode_set_ctime_current(&dir->vfs_inode)); ret = btrfs_update_inode(trans, dir); From 008e2512dc5696ab2dc5bf264e98a9fe9ceb830e Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Sun, 11 Aug 2024 15:00:22 +0930 Subject: [PATCH 3/5] btrfs: tree-checker: add dev extent item checks [REPORT] There is a corruption report that btrfs refused to mount a fs that has overlapping dev extents: BTRFS error (device sdc): dev extent devid 4 physical offset 14263979671552 overlap with previous dev extent end 14263980982272 BTRFS error (device sdc): failed to verify dev extents against chunks: -117 BTRFS error (device sdc): open_ctree failed [CAUSE] The direct cause is very obvious, there is a bad dev extent item with incorrect length. With btrfs check reporting two overlapping extents, the second one shows some clue on the cause: ERROR: dev extent devid 4 offset 14263979671552 len 6488064 overlap with previous dev extent end 14263980982272 ERROR: dev extent devid 13 offset 2257707008000 len 6488064 overlap with previous dev extent end 2257707270144 ERROR: errors found in extent allocation tree or chunk allocation The second one looks like a bitflip happened during new chunk allocation: hex(2257707008000) = 0x20da9d30000 hex(2257707270144) = 0x20da9d70000 diff = 0x00000040000 So it looks like a bitflip happened during new dev extent allocation, resulting the second overlap. Currently we only do the dev-extent verification at mount time, but if the corruption is caused by memory bitflip, we really want to catch it before writing the corruption to the storage. Furthermore the dev extent items has the following key definition: ( DEV_EXTENT ) Thus we can not just rely on the generic key order check to make sure there is no overlapping. [ENHANCEMENT] Introduce dedicated dev extent checks, including: - Fixed member checks * chunk_tree should always be BTRFS_CHUNK_TREE_OBJECTID (3) * chunk_objectid should always be BTRFS_FIRST_CHUNK_CHUNK_TREE_OBJECTID (256) - Alignment checks * chunk_offset should be aligned to sectorsize * length should be aligned to sectorsize * key.offset should be aligned to sectorsize - Overlap checks If the previous key is also a dev-extent item, with the same device id, make sure we do not overlap with the previous dev extent. Reported: Stefan N Link: https://lore.kernel.org/linux-btrfs/CA+W5K0rSO3koYTo=nzxxTm1-Pdu1HYgVxEpgJ=aGc7d=E8mGEg@mail.gmail.com/ CC: stable@vger.kernel.org # 5.10+ Reviewed-by: Anand Jain Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/tree-checker.c | 69 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 69 insertions(+) diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index 6f1e2f2215d9..634d69964fe4 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -1764,6 +1764,72 @@ static int check_raid_stripe_extent(const struct extent_buffer *leaf, return 0; } +static int check_dev_extent_item(const struct extent_buffer *leaf, + const struct btrfs_key *key, + int slot, + struct btrfs_key *prev_key) +{ + struct btrfs_dev_extent *de; + const u32 sectorsize = leaf->fs_info->sectorsize; + + de = btrfs_item_ptr(leaf, slot, struct btrfs_dev_extent); + /* Basic fixed member checks. */ + if (unlikely(btrfs_dev_extent_chunk_tree(leaf, de) != + BTRFS_CHUNK_TREE_OBJECTID)) { + generic_err(leaf, slot, + "invalid dev extent chunk tree id, has %llu expect %llu", + btrfs_dev_extent_chunk_tree(leaf, de), + BTRFS_CHUNK_TREE_OBJECTID); + return -EUCLEAN; + } + if (unlikely(btrfs_dev_extent_chunk_objectid(leaf, de) != + BTRFS_FIRST_CHUNK_TREE_OBJECTID)) { + generic_err(leaf, slot, + "invalid dev extent chunk objectid, has %llu expect %llu", + btrfs_dev_extent_chunk_objectid(leaf, de), + BTRFS_FIRST_CHUNK_TREE_OBJECTID); + return -EUCLEAN; + } + /* Alignment check. */ + if (unlikely(!IS_ALIGNED(key->offset, sectorsize))) { + generic_err(leaf, slot, + "invalid dev extent key.offset, has %llu not aligned to %u", + key->offset, sectorsize); + return -EUCLEAN; + } + if (unlikely(!IS_ALIGNED(btrfs_dev_extent_chunk_offset(leaf, de), + sectorsize))) { + generic_err(leaf, slot, + "invalid dev extent chunk offset, has %llu not aligned to %u", + btrfs_dev_extent_chunk_objectid(leaf, de), + sectorsize); + return -EUCLEAN; + } + if (unlikely(!IS_ALIGNED(btrfs_dev_extent_length(leaf, de), + sectorsize))) { + generic_err(leaf, slot, + "invalid dev extent length, has %llu not aligned to %u", + btrfs_dev_extent_length(leaf, de), sectorsize); + return -EUCLEAN; + } + /* Overlap check with previous dev extent. */ + if (slot && prev_key->objectid == key->objectid && + prev_key->type == key->type) { + struct btrfs_dev_extent *prev_de; + u64 prev_len; + + prev_de = btrfs_item_ptr(leaf, slot - 1, struct btrfs_dev_extent); + prev_len = btrfs_dev_extent_length(leaf, prev_de); + if (unlikely(prev_key->offset + prev_len > key->offset)) { + generic_err(leaf, slot, + "dev extent overlap, prev offset %llu len %llu current offset %llu", + prev_key->objectid, prev_len, key->offset); + return -EUCLEAN; + } + } + return 0; +} + /* * Common point to switch the item-specific validation. */ @@ -1800,6 +1866,9 @@ static enum btrfs_tree_block_status check_leaf_item(struct extent_buffer *leaf, case BTRFS_DEV_ITEM_KEY: ret = check_dev_item(leaf, key, slot); break; + case BTRFS_DEV_EXTENT_KEY: + ret = check_dev_extent_item(leaf, key, slot, prev_key); + break; case BTRFS_INODE_ITEM_KEY: ret = check_inode_item(leaf, key, slot); break; From e30729d4bd4001881be4d1ad4332a5d4985398f8 Mon Sep 17 00:00:00 2001 From: Naohiro Aota Date: Thu, 1 Aug 2024 16:47:52 +0900 Subject: [PATCH 4/5] btrfs: zoned: properly take lock to read/update block group's zoned variables __btrfs_add_free_space_zoned() references and modifies bg's alloc_offset, ro, and zone_unusable, but without taking the lock. It is mostly safe because they monotonically increase (at least for now) and this function is mostly called by a transaction commit, which is serialized by itself. Still, taking the lock is a safer and correct option and I'm going to add a change to reset zone_unusable while a block group is still alive. So, add locking around the operations. Fixes: 169e0da91a21 ("btrfs: zoned: track unusable bytes for zones") CC: stable@vger.kernel.org # 5.15+ Reviewed-by: Johannes Thumshirn Signed-off-by: Naohiro Aota Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/free-space-cache.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index f5996a43db24..eaa1dbd31352 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -2697,15 +2697,16 @@ static int __btrfs_add_free_space_zoned(struct btrfs_block_group *block_group, u64 offset = bytenr - block_group->start; u64 to_free, to_unusable; int bg_reclaim_threshold = 0; - bool initial = ((size == block_group->length) && (block_group->alloc_offset == 0)); + bool initial; u64 reclaimable_unusable; - WARN_ON(!initial && offset + size > block_group->zone_capacity); + spin_lock(&block_group->lock); + initial = ((size == block_group->length) && (block_group->alloc_offset == 0)); + WARN_ON(!initial && offset + size > block_group->zone_capacity); if (!initial) bg_reclaim_threshold = READ_ONCE(sinfo->bg_reclaim_threshold); - spin_lock(&ctl->tree_lock); if (!used) to_free = size; else if (initial) @@ -2718,7 +2719,9 @@ static int __btrfs_add_free_space_zoned(struct btrfs_block_group *block_group, to_free = offset + size - block_group->alloc_offset; to_unusable = size - to_free; + spin_lock(&ctl->tree_lock); ctl->free_space += to_free; + spin_unlock(&ctl->tree_lock); /* * If the block group is read-only, we should account freed space into * bytes_readonly. @@ -2727,11 +2730,8 @@ static int __btrfs_add_free_space_zoned(struct btrfs_block_group *block_group, block_group->zone_unusable += to_unusable; WARN_ON(block_group->zone_unusable > block_group->length); } - spin_unlock(&ctl->tree_lock); if (!used) { - spin_lock(&block_group->lock); block_group->alloc_offset -= size; - spin_unlock(&block_group->lock); } reclaimable_unusable = block_group->zone_unusable - @@ -2745,6 +2745,8 @@ static int __btrfs_add_free_space_zoned(struct btrfs_block_group *block_group, btrfs_mark_bg_to_reclaim(block_group); } + spin_unlock(&block_group->lock); + return 0; } From 534f7eff9239c1b0af852fc33f5af2b62c00eddf Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Fri, 16 Aug 2024 10:40:38 +0930 Subject: [PATCH 5/5] btrfs: only enable extent map shrinker for DEBUG builds Although there are several patches improving the extent map shrinker, there are still reports of too frequent shrinker behavior, taking too much CPU for the kswapd process. So let's only enable extent shrinker for now, until we got more comprehensive understanding and a better solution. Link: https://lore.kernel.org/linux-btrfs/3df4acd616a07ef4d2dc6bad668701504b412ffc.camel@intelfx.name/ Link: https://lore.kernel.org/linux-btrfs/c30fd6b3-ca7a-4759-8a53-d42878bf84f7@gmail.com/ Fixes: 956a17d9d050 ("btrfs: add a shrinker for extent maps") CC: stable@vger.kernel.org # 6.10+ Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/super.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 11044e9e2cb1..98fa0f382480 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -2402,7 +2402,13 @@ static long btrfs_nr_cached_objects(struct super_block *sb, struct shrink_contro trace_btrfs_extent_map_shrinker_count(fs_info, nr); - return nr; + /* + * Only report the real number for DEBUG builds, as there are reports of + * serious performance degradation caused by too frequent shrinks. + */ + if (IS_ENABLED(CONFIG_BTRFS_DEBUG)) + return nr; + return 0; } static long btrfs_free_cached_objects(struct super_block *sb, struct shrink_control *sc)