Skip to content

optimise registered buffer/file updates

Pavel Begunkov io-uring lkml 2 versions – Applied

Status from this message: Applied, thanks!

Get series with b4
b4 am [email protected]

Versions newest first · linked by author and title

  1. v2 Latest13 patches 2 repliesApplied
  2. v1 RFC11 patches 11 repliesSuperseded

Patches in v2 Applied

#Subject and filesSizeRepliesApplied
1/13io_uring/rsrc: use non-pcpu refcounts for nodesio_uring/rsrc.c io_uring/rsrc.h+9 -120 replies
2/13io_uring/rsrc: keep cached refs per nodeinclude/linux/io_uring_types.h io_uring/rsrc.c io_uring/rsrc.h+18 -140 replies
3/13io_uring: don't put nodes under spinlocksio_uring/io_uring.c io_uring/rsrc.h+8 -50 replies
4/13io_uring: io_free_req() via twio_uring/io_uring.c+7 -10 replies
5/13io_uring/rsrc: protect node refs with uring_lockio_uring/io_uring.c io_uring/rsrc.c io_uring/rsrc.h+21 -560 replies
6/13io_uring/rsrc: kill rsrc_ref_lockinclude/linux/io_uring_types.h io_uring/io_uring.c io_uring/rsrc.c+1 -70 replies
7/13io_uring/rsrc: rename rsrc_listio_uring/rsrc.c io_uring/rsrc.h+10 -40 replies
8/13io_uring/rsrc: optimise io_rsrc_put allocationio_uring/rsrc.c io_uring/rsrc.h+36 -170 replies
9/13io_uring/rsrc: don't offload node freeinclude/linux/io_uring_types.h io_uring/io_uring.c io_uring/rsrc.c+4 -640 replies
10/13io_uring/rsrc: cache struct io_rsrc_nodeinclude/linux/io_uring_types.h io_uring/io_uring.c io_uring/rsrc.c +1 more+32 -120 replies
11/13io_uring/rsrc: add lockdep sanity checksio_uring/io_uring.c io_uring/rsrc.c io_uring/rsrc.h+7 -50 replies
12/13io_uring/rsrc: optimise io_rsrc_data refcountingio_uring/rsrc.c io_uring/rsrc.h+19 -130 replies
13/13io_uring/rsrc: add custom limit for node cachinginclude/linux/io_uring_types.h io_uring/alloc_cache.h io_uring/io_uring.c +1 more+13 -50 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v2 yet.

Read the v2 thread

Cover letter [PATCH v2 00/13] optimise registered buffer/file updates

The patchset optimises registered files and buffers updates / removals, The rsrc-update-bench test showes 11x improvement (1040K -> 11468K updates / sec). It also improves latency by eliminating rcu grace period waiting and bouncing it to another worker, and reduces memory footprint by removing percpu refs.

That's quite important for apps updating files/buffers with medium or higher frequency as updates are slow and expensive, and it currently takes quite a number of IO requests per update to make using fixed files/buffers worthwhile.

Another upside is that it makes it simpler, patch 9 removes very convoluted synchronisation via flush_delayed_work() from the quiesce path.

v2: rebase, add patches 12 and 13 to remove the last pair atomics out
of the path and to limit caching.

Pavel Begunkov (13):
io_uring/rsrc: use non-pcpu refcounts for nodes io_uring/rsrc: keep cached refs per node
io_uring: don't put nodes under spinlocks
io_uring: io_free_req() via tw io_uring/rsrc: protect node refs with uring_lock io_uring/rsrc: kill rsrc_ref_lock io_uring/rsrc: rename rsrc_list io_uring/rsrc: optimise io_rsrc_put allocation io_uring/rsrc: don't offload node free io_uring/rsrc: cache struct io_rsrc_node io_uring/rsrc: add lockdep sanity checks io_uring/rsrc: optimise io_rsrc_data refcounting io_uring/rsrc: add custom limit for node caching

 include/linux/io_uring_types.h |   8 +-
 io_uring/alloc_cache.h         |   6 +-
 io_uring/io_uring.c            |  54 ++++++----
 io_uring/rsrc.c                | 176 ++++++++++++---------------------
 io_uring/rsrc.h                |  58 +++++------
 5 files changed, 136 insertions(+), 166 deletions(-)
Signature · 1 line
2.39.1