Changes in compression.c

6f9a8ab7 - btrfs: compression: adjust cb->compressed_folios allocation type

Sat, 26 Apr 2025 06:23:29 +0000

btrfs: compression: adjust cb->compressed_folios allocation typeIn preparation for making the kmalloc() family of allocators type aware,we need to make sure that the returned type from the allocation matchesthe type of the variable being assigned. (Before, the allocator wouldalways return "void *", which can be implicitly cast to any pointer type.)The assigned type is "struct folio **" but the returned type will be"struct page **". These are the same allocation size (pointer size), butthe types don't match. Adjust the allocation type to match the assignment.Reviewed-by: Qu Wenruo Signed-off-by: Kees Cook Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

fc5c0c58 - btrfs: defrag: extend ioctl to accept compression levels

Thu, 06 Mar 2025 13:15:35 +0000

btrfs: defrag: extend ioctl to accept compression levelsThe zstd and zlib compression types support setting compression levels.Enhance the defrag interface to specify the levels as well. For zstd thenegative (realtime) levels are also accepted.Reviewed-by: Qu Wenruo Signed-off-by: Daniel Vacek Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

da798fa5 - btrfs: zstd: enable negative compression levels mount option

Thu, 30 Jan 2025 17:58:19 +0000

btrfs: zstd: enable negative compression levels mount optionAllow using the fast modes (negative compression levels) of zstd as amount option.As per the results, the compression ratio is (expectedly) lower:for level in {-15..-1} 1 2 3; \do printf "level %3d\n" $level; \ mount -o compress=zstd:$level /dev/sdb /mnt/test/; \ grep sdb /proc/mounts; \ cp -r /usr/bin /mnt/test/; sync; compsize /mnt/test/bin; \ cp -r /usr/share/doc /mnt/test/; sync; compsize /mnt/test/doc; \ cp enwik9 /mnt/test/; sync; compsize /mnt/test/enwik9; \ cp linux-6.13.tar /mnt/test/; sync; compsize /mnt/test/linux-6.13.tar; \ rm -r /mnt/test/{bin,doc,enwik9,linux-6.13.tar}; \ umount /mnt/test/; \done |& tee results | \awk '/^level/{print}/^TOTAL/{print$3"\t"$2" |"}' | paste - - - - - 266M bin | 45M doc | 953M wiki | 1.4G source=============================+===============+===============+===============+level -15 180M 67% | 30M 68% | 694M 72% | 598M 40% |level -14 180M 67% | 30M 67% | 683M 71% | 581M 39% |level -13 177M 66% | 29M 66% | 671M 70% | 566M 38% |level -12 174M 65% | 29M 65% | 658M 69% | 548M 37% |level -11 174M 65% | 28M 64% | 645M 67% | 530M 35% |level -10 171M 64% | 28M 62% | 631M 66% | 512M 34% |level -9 165M 62% | 27M 61% | 615M 64% | 493M 33% |level -8 161M 60% | 27M 59% | 598M 62% | 475M 32% |level -7 155M 58% | 26M 58% | 582M 61% | 457M 30% |level -6 151M 56% | 25M 56% | 565M 59% | 437M 29% |level -5 145M 54% | 24M 55% | 545M 57% | 417M 28% |level -4 139M 52% | 23M 52% | 520M 54% | 391M 26% |level -3 135M 50% | 22M 50% | 495M 51% | 369M 24% |level -2 127M 47% | 22M 48% | 470M 49% | 349M 23% |level -1 120M 45% | 21M 47% | 452M 47% | 332M 22% |level 1 110M 41% | 17M 39% | 362M 38% | 290M 19% |level 2 106M 40% | 17M 38% | 349M 36% | 288M 19% |level 3 104M 39% | 16M 37% | 340M 35% | 276M 18% |The samples represent some data sets that can be commonly found and showapproximate compressibility. The fast levels trade off speed for ratioand are best suitable for highly compressible data.As can be seen above, comparing the results to the current default zstdlevel 3, the negative levels are roughly 2x worse at -15 and theratio increases almost linearly with each level.Signed-off-by: Daniel Vacek [ update changelog ]Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

d07eaa99 - btrfs: use filemap_get_folio() helper

Mon, 14 Oct 2024 15:11:38 +0000

btrfs: use filemap_get_folio() helperWhen fgp_flags and gfp_flags are zero, use filemap_get_folio(A, B)instead of __filemap_get_folio(A, B, 0, 0)—no need for the extraarguments 0, 0.Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

0f712026 - btrfs: rename btrfs_folio_(set|start|end)_writer_lock()

Wed, 09 Oct 2024 05:51:07 +0000

btrfs: rename btrfs_folio_(set|start|end)_writer_lock()Since there is no user of reader locks, rename the writer locks into amore generic name, by removing the "_writer" part from the name.And also rename btrfs_subpage::writer into btrfs_subpage::locked.Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

336e69f3 - btrfs: unify to use writer locks for subpage locking

Wed, 09 Oct 2024 05:51:06 +0000

btrfs: unify to use writer locks for subpage lockingSince commit d7172f52e993 ("btrfs: use per-buffer locking forextent_buffer reading"), metadata read no longer relies on the subpagereader locking.This means we do not need to maintain a different metadata/data splitfor locking, so we can convert the existing reader lock users by:- add_ra_bio_pages() Convert to btrfs_folio_set_writer_lock()- end_folio_read() Convert to btrfs_folio_end_writer_lock()- begin_folio_read() Convert to btrfs_folio_set_writer_lock()- folio_range_has_eb() Remove the subpage->readers checks, since it is always 0.- Remove btrfs_subpage_start_reader() and btrfs_subpage_end_reader()Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

a9c50c97 - btrfs: drop unused parameter level from alloc_heuristic_ws()

Wed, 09 Oct 2024 14:32:25 +0000

btrfs: drop unused parameter level from alloc_heuristic_ws()The compression heuristic pass does not need a level, so we can drop theparameter.Reviewed-by: Anand Jain Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

3f4b1bc1 - btrfs: lzo: drop unused paramter level from lzo_alloc_workspace()

Wed, 09 Oct 2024 14:31:25 +0000

btrfs: lzo: drop unused paramter level from lzo_alloc_workspace()The LZO compression has only one level, we don't need to pass theparameter.Reviewed-by: Anand Jain Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

dd5e2762 - btrfs: compression: add an ASSERT() to ensure the read-in length is sane

Fri, 06 Sep 2024 04:57:56 +0000

btrfs: compression: add an ASSERT() to ensure the read-in length is saneThere are already two bugs (one in zlib, one in zstd) that involvedcompression path is not handling sector size < page size cases well.So it makes more sense to make sure that btrfs_compress_folios() returnsSince we already have two bugs (one in zlib, one in zstd) in thecompression path resulting the @total_in be to larger than theto-be-compressed range length, there is enough reason to add an ASSERT()to make sure the total read-in length doesn't exceed the input length.Signed-off-by: Qu Wenruo Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

aeb6d881 - btrfs: convert btrfs_decompress() to take a folio

Wed, 28 Aug 2024 18:29:07 +0000

btrfs: convert btrfs_decompress() to take a folioThe old page API is being gradually replaced and converted to use folioto improve code readability and avoid repeated conversion between pageand folio. Based on the previous patch, the compression path can bedirectly used in folio without converting to page.Signed-off-by: Li Zetao Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

b70f3a45 - btrfs: convert zstd_decompress() to take a folio

Wed, 28 Aug 2024 18:29:06 +0000

btrfs: convert zstd_decompress() to take a folioThe old page API is being gradually replaced and converted to use folioto improve code readability and avoid repeated conversion between pageand folio. And memcpy_to_page() can be replaced with memcpy_to_folio().But there is no memzero_folio(), but it can be replaced equivalently byfolio_zero_range().Signed-off-by: Li Zetao Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

9f9a4e43 - btrfs: convert lzo_decompress() to take a folio

Wed, 28 Aug 2024 18:29:05 +0000

btrfs: convert lzo_decompress() to take a folioThe old page API is being gradually replaced and converted to use folioto improve code readability and avoid repeated conversion between pageand folio. And memcpy_to_page() can be replaced with memcpy_to_folio().But there is no memzero_folio(), but it can be replaced equivalently byfolio_zero_range().Signed-off-by: Li Zetao Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

54c78d49 - btrfs: convert zlib_decompress() to take a folio

Wed, 28 Aug 2024 18:29:04 +0000

btrfs: convert zlib_decompress() to take a folioThe old page API is being gradually replaced and converted to use folioto improve code readability and avoid repeated conversion between pageand folio. And memcpy_to_page() can be replaced with memcpy_to_folio().But there is no memzero_folio(), but it can be replaced equivalently byfolio_zero_range().Signed-off-by: Li Zetao Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

ac325fc2 - btrfs: do not hold the extent lock for entire read

Fri, 16 Aug 2024 19:16:24 +0000

btrfs: do not hold the extent lock for entire readHistorically we've held the extent lock throughout the entire read.There's been a few reasons for this, but it's mostly just caused usproblems. For example, this prevents us from allowing page faultsduring direct io reads, because we could deadlock. This has forced usto only allow 4k reads at a time for io_uring NOWAIT requests because wehave no idea if we'll be forced to page fault and thus have to do awhole lot of work.On the buffered side we are protected by the page lock, as long as we'rereading things like buffered writes, punch hole, and even direct IO to acertain degree will get hung up on the page lock while the page is inflight.On the direct side we have the dio extent lock, which acts much like theway the extent lock worked previously to this patch, however just fordirect reads. This protects direct reads from concurrent direct writes,while we're protected from buffered writes via the inode lock.Now that we're protected in all cases, narrow the extent lock to thepart where we're getting the extent map to submit the reads, no longerholding the extent lock for the entire read operation. Push the extentlock down into do_readpage() so that we're only grabbing it when lookingup the extent map. This portion was contributed by Goldwyn.Co-developed-by: Goldwyn Rodrigues Reviewed-by: Goldwyn Rodrigues Signed-off-by: Josef Bacik Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

792e86ef - btrfs: rename btrfs_submit_bio() to btrfs_submit_bbio()

Tue, 27 Aug 2024 01:40:11 +0000

btrfs: rename btrfs_submit_bio() to btrfs_submit_bbio()The function name is a bit misleading as it submits the btrfs_bio(bbio), rename it so we can use btrfs_submit_bio() when an actual bio issubmitted.Reviewed-by: Qu Wenruo Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

c808c1dc - btrfs: convert add_ra_bio_pages() to use only folios

Wed, 24 Jul 2024 19:16:40 +0000

btrfs: convert add_ra_bio_pages() to use only foliosWilly is going to get rid of page->index, and add_ra_bio_pages usespage->index. Make his life easier by converting add_ra_bio_pages to usefolios so that we are no longer using page->index.Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

8e786054 - btrfs: fix extent map use-after-free when adding pages to compressed bio

Thu, 04 Jul 2024 15:11:20 +0000

btrfs: fix extent map use-after-free when adding pages to compressed bioAt add_ra_bio_pages() we are accessing the extent map to calculate'add_size' after we dropped our reference on the extent map, resultingin a use-after-free. Fix this by computing 'add_size' before dropping ourextent map reference.Reported-by: syzbot+853d80cba98ce1157ae6@syzkaller.appspotmail.comLink: https://lore.kernel.org/linux-btrfs/000000000000038144061c6d18f2@google.com/Fixes: 6a4049102055 ("btrfs: subpage: make add_ra_bio_pages() compatible")CC: stable@vger.kernel.org # 6.1+Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

fea91134 - btrfs: remove the extra_gfp parameter from btrfs_alloc_folio_array()

Thu, 27 Jun 2024 03:45:55 +0000

btrfs: remove the extra_gfp parameter from btrfs_alloc_folio_array()The function btrfs_alloc_folio_array() is only utilized inbtrfs_submit_compressed_read() and no other location, and the onlycaller is not utilizing the @extra_gfp parameter.Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

e2877c2a - btrfs: pass a btrfs_inode to btrfs_compress_heuristic()

Mon, 31 Jan 2022 18:16:41 +0000

btrfs: pass a btrfs_inode to btrfs_compress_heuristic()Pass a struct btrfs_inode to btrfs_compress_heuristic() as it's aninternal interface, allowing to remove some use of BTRFS_I.Reviewed-by: Boris Burkov Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c

a1f4e3d7 - btrfs: switch btrfs_ordered_extent::inode to struct btrfs_inode

Mon, 31 Jan 2022 17:57:43 +0000

btrfs: switch btrfs_ordered_extent::inode to struct btrfs_inodeThe structure is internal so we should use struct btrfs_inode for that,allowing to remove some use of BTRFS_I.Reviewed-by: Boris Burkov Signed-off-by: David Sterba List of files: /linux-6.15/fs/btrfs/compression.c