public inbox for [email protected]
 help / color / mirror / Atom feed
* [PATCH for-next 0/4] singly linked list for chains
@ 2020-10-15 21:11 Pavel Begunkov
  2020-10-15 21:11 ` [PATCH 1/4] io_uring: inline io_fail_links() Pavel Begunkov
                   ` (4 more replies)
  0 siblings, 5 replies; 6+ messages in thread
From: Pavel Begunkov @ 2020-10-15 21:11 UTC (permalink / raw)
  To: Jens Axboe, io-uring

see [3/4] and [4/4] for motivation, the other two are just preps.
I wasn't expecting to find performance difference, but my naive nop
test yeilds 5030 vs 5160 KIOPS, before and after [3/4] correspondingly.

The test is submitting 32 linked nops and waits for them to complete.
The testing is tuned for consistentcy, and the results are consistent
across reboots.

Pavel Begunkov (4):
  io_uring: inline io_fail_links()
  io_uring: simplify linked timeout cancel
  io_uring: link requests with singly linked list
  io_uring: toss io_kiocb fields for better caching
 
 fs/io_uring.c | 202 +++++++++++++++++++++-----------------------------
 1 file changed, 83 insertions(+), 119 deletions(-)

-- 
2.24.0


^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2020-10-15 21:18 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-10-15 21:11 [PATCH for-next 0/4] singly linked list for chains Pavel Begunkov
2020-10-15 21:11 ` [PATCH 1/4] io_uring: inline io_fail_links() Pavel Begunkov
2020-10-15 21:11 ` [PATCH 2/4] io_uring: simplify linked timeout cancel Pavel Begunkov
2020-10-15 21:11 ` [PATCH 3/4] io_uring: link requests with singly linked list Pavel Begunkov
2020-10-15 21:11 ` [PATCH 4/4] io_uring: toss io_kiocb fields for better caching Pavel Begunkov
2020-10-15 21:15 ` [PATCH for-next 0/4] singly linked list for chains Pavel Begunkov

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox