diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2024-03-12 12:28:34 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2024-03-12 12:28:34 -0700 |
commit | 43a7548e28a6df12a6170421d9d016c576010baa (patch) | |
tree | f956b1fc70acceca74f4c69502e592fcbad7ec77 /fs/btrfs/raid56.c | |
parent | 35d4aeea10558d12022d752b20be371aced557da (diff) | |
parent | 1cab1375ba6d5337a25acb346996106c12bb2dd0 (diff) |
Merge tag 'for-6.9-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs updates from David Sterba:
"Mostly stabilization, refactoring and cleanup changes. There rest are
minor performance optimizations due to caching or lock contention
reduction and a few notable fixes.
Performance improvements:
- minor speedup in logging when repeatedly allocated structure is
preallocated only once, improves latency and decreases lock
contention
- minor throughput increase (+6%), reduced lock contention after
clearing delayed allocation bits, applies to several common
workload types
- skip full quota rescan if a new relation is added in the same
transaction
Fixes:
- zstd fix for inline compressed file in subpage mode, updated
version from the 6.8 time
- proper qgroup inheritance ioctl parameter validation
- more fiemap followup fixes after reduced locking done in 6.8:
- fix race when detecting delalloc ranges
Core changes:
- more debugging code:
- added assertions for a very rare crash in raid56 calculation
- tree-checker dumps page state to give more insights into
possible reference counting issues
- add checksum calculation offloading sysfs knob, for now enabled
under DEBUG only to determine a good heuristic for deciding the
offload or synchronous, depends on various factors (block group
profile, device speed) and is not as clear as initially thought
(checksum type)
- error handling improvements, added assertions
- more page to folio conversion (defrag, truncate), cached size and
shift
- preparation for more fine grained locking of sectors in subpage
mode
- cleanups and refactoring:
- include cleanups, forward declarations
- pointer-to-structure helpers
- redundant argument removals
- removed unused code
- slab cache updates, last use of SLAB_MEM_SPREAD removed"
* tag 'for-6.9-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (114 commits)
btrfs: reuse cloned extent buffer during fiemap to avoid re-allocations
btrfs: fix race when detecting delalloc ranges during fiemap
btrfs: fix off-by-one chunk length calculation at contains_pending_extent()
btrfs: qgroup: allow quick inherit if snapshot is created and added to the same parent
btrfs: qgroup: validate btrfs_qgroup_inherit parameter
btrfs: include device major and minor numbers in the device scan notice
btrfs: mark btrfs_put_caching_control() static
btrfs: remove SLAB_MEM_SPREAD flag use
btrfs: qgroup: always free reserved space for extent records
btrfs: tree-checker: dump the page status if hit something wrong
btrfs: compression: remove dead comments in btrfs_compress_heuristic()
btrfs: subpage: make writer lock utilize bitmap
btrfs: subpage: make reader lock utilize bitmap
btrfs: unexport btrfs_subpage_start_writer() and btrfs_subpage_end_and_test_writer()
btrfs: pass a valid extent map cache pointer to __get_extent_map()
btrfs: merge btrfs_del_delalloc_inode() helpers
btrfs: pass btrfs_device to btrfs_scratch_superblocks()
btrfs: handle transaction commit errors in flush_reservations()
btrfs: use KMEM_CACHE() to create btrfs_free_space cache
btrfs: use KMEM_CACHE() to create delayed ref caches
...
Diffstat (limited to 'fs/btrfs/raid56.c')
-rw-r--r-- | fs/btrfs/raid56.c | 31 |
1 files changed, 30 insertions, 1 deletions
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index 792c8e17c31d..6f4a9cfeea44 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -14,7 +14,6 @@ #include <linux/raid/xor.h> #include <linux/mm.h> #include "messages.h" -#include "misc.h" #include "ctree.h" #include "disk-io.h" #include "volumes.h" @@ -918,6 +917,13 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info, */ ASSERT(stripe_nsectors <= BITS_PER_LONG); + /* + * Real stripes must be between 2 (2 disks RAID5, aka RAID1) and 256 + * (limited by u8). + */ + ASSERT(real_stripes >= 2); + ASSERT(real_stripes <= U8_MAX); + rbio = kzalloc(sizeof(*rbio), GFP_NOFS); if (!rbio) return ERR_PTR(-ENOMEM); @@ -955,6 +961,7 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info, ASSERT(btrfs_nr_parity_stripes(bioc->map_type)); rbio->nr_data = real_stripes - btrfs_nr_parity_stripes(bioc->map_type); + ASSERT(rbio->nr_data > 0); return rbio; } @@ -1181,6 +1188,26 @@ static inline void bio_list_put(struct bio_list *bio_list) bio_put(bio); } +static void assert_rbio(struct btrfs_raid_bio *rbio) +{ + if (!IS_ENABLED(CONFIG_BTRFS_DEBUG) || + !IS_ENABLED(CONFIG_BTRFS_ASSERT)) + return; + + /* + * At least two stripes (2 disks RAID5), and since real_stripes is U8, + * we won't go beyond 256 disks anyway. + */ + ASSERT(rbio->real_stripes >= 2); + ASSERT(rbio->nr_data > 0); + + /* + * This is another check to make sure nr data stripes is smaller + * than total stripes. + */ + ASSERT(rbio->nr_data < rbio->real_stripes); +} + /* Generate PQ for one vertical stripe. */ static void generate_pq_vertical(struct btrfs_raid_bio *rbio, int sectornr) { @@ -1212,6 +1239,7 @@ static void generate_pq_vertical(struct btrfs_raid_bio *rbio, int sectornr) pointers[stripe++] = kmap_local_page(sector->page) + sector->pgoff; + assert_rbio(rbio); raid6_call.gen_syndrome(rbio->real_stripes, sectorsize, pointers); } else { @@ -2473,6 +2501,7 @@ static int finish_parity_scrub(struct btrfs_raid_bio *rbio) } if (has_qstripe) { + assert_rbio(rbio); /* RAID6, call the library function to fill in our P/Q */ raid6_call.gen_syndrome(rbio->real_stripes, sectorsize, pointers); |