no-copy bvec
Status from this message: Applied, thanks.
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v3 Latest7 patches 8 repliesApplied
- v27 patches 7 repliesSuperseded
- v16 patches 31 repliesSuperseded
Changes in v3 from the cover letter
- add a comment in 1/7 (Christoph)
- add a note about 0-len bvecs in biovecs.rst (Matthew)
Patches in v3 Applied
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/7 | splice: don't generate zero-len segement bvecsfs/ | +6 | 1 reply | |
| 2/7 | bvec/iter: disallow zero-length segment bvecsDocumentation/ Documentation/ lib/ | +9 | 1 reply | |
| 3/7 | block/psi: remove PSI annotations from direct IOblock/ fs/ | +8 | 1 reply | |
| 4/7 | target/file: allocate the bvec array as part of struct target_core_file_cmddrivers/ | +6 | 1 reply | |
| 5/7 | iov_iter: optimise bvec iov_iter_advance()lib/ | +19 | 1 reply | |
| 6/7 | bio: add a helper calculating nr segments to allocfs/ fs/ include/ | +18 | 1 reply | |
| 7/7 | bio: don't copy bvec for direct IODocumentation/ block/ include/ | +42 | 1 reply |
Reviews trailers per patch
All 7 patches have a Reviewed-by or Acked-by.
| Patch | Christoph Hellwig | Ming Lei |
|---|---|---|
| 1/7 | R-b | R-b |
| 2/7 | R-b | R-b |
| 3/7 | R-b | R-b |
| 4/7 | R-b | |
| 5/7 | R-b | R-b |
| 6/7 | R-b | R-b |
| 7/7 | R-b | R-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH v3 0/7] no-copy bvec
Currently, when iomap and block direct IO gets a bvec based iterator the bvec will be copied, with all other accounting that takes much CPU time and causes additional allocation for larger bvecs. The patchset makes it to reuse the passed in iter bvec.
[1,2] are forbidding zero-length bvec segments to not pile special cases, [3] skip/fix PSI tracking to not iterate over bvecs extra time.
nullblk completion_nsec=0 submit_queues=NR_CORES, no merges, no stats fio/t/io_uring /dev/nullb0 -d 128 -s 32 -c 32 -p 0 -B 1 -F 1 -b BLOCK_SIZE
BLOCK_SIZE 512 4K 8K 16K 32K 64K
===================================================================
old (KIOPS) 1208 1208 1131 1039 863 699 new (KIOPS) 1222 1222 1170 1137 1083 982
Previously, Jens got before 10% difference for polling real HW and small block sizes, but that was for an older version that had one iov_iter_advance() less
since RFC:
- add target_core_file patch by Christoph
- make no-copy default behaviour, remove iter flag
- iter_advance() instead of hacks to revert to work
- add bvec iter_advance() optimisation patch
- remove PSI annotations from direct IO (iomap, block and fs/direct)
- note in d/f/porting
since v1:
- don't allow zero-length bvec segments (Ming)
- don't add a BIO_WORKINGSET-less version of bio_add_page(), just clear the flag at the end and leave it for further cleanups (Christoph)
- commit message and comments rewording (Dave)
- other nits by Christoph
since v2:
- add a comment in 1/7 (Christoph)
- add a note about 0-len bvecs in biovecs.rst (Matthew)
Christoph Hellwig (1):
target/file: allocate the bvec array as part of struct
target_core_file_cmd
Pavel Begunkov (6):
splice: don't generate zero-len segement bvecs bvec/iter: disallow zero-length segment bvecs block/psi: remove PSI annotations from direct IO
iov_iter: optimise bvec iov_iter_advance()
bio: add a helper calculating nr segments to alloc
bio: don't copy bvec for direct IO
Documentation/block/biovecs.rst | 2 + Documentation/filesystems/porting.rst | 16 ++++++ block/bio.c | 71 +++++++++++++-------------- drivers/target/target_core_file.c | 20 +++----- fs/block_dev.c | 7 +-- fs/direct-io.c | 2 + fs/iomap/direct-io.c | 9 ++-- fs/splice.c | 9 ++-- include/linux/bio.h | 13 +++++ lib/iov_iter.c | 21 +++++++- 10 files changed, 106 insertions(+), 64 deletions(-)
Signature · 1 line
2.24.0