Skip to content

squeeze more performance

Pavel Begunkov io-uring 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest16 patches 2 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/16io_uring: optimise kiocb layoutfs/io_uring.c+5 -20 replies
2/16io_uring: add more likely/unlikely() annotationsfs/io_uring.c+3 -30 replies
3/16io_uring: delay req queueing into compl-batch listfs/io_uring.c+16 -100 replies
4/16io_uring: optimise request allocationfs/io_uring.c+20 -80 replies
5/16io_uring: optimise INIT_WQ_LISTfs/io-wq.h+0 -10 replies
6/16io_uring: don't wake sqpoll in io_cqring_ev_postedfs/io_uring.c+0 -20 replies
7/16io_uring: merge CQ and poll waitqueuesfs/io_uring.c+1 -70 replies
8/16io_uring: optimise ctx referencing by requestsfs/io_uring.c+14 -120 replies
9/16io_uring: mark cold functionsfs/io_uring.c+64 -590 replies
10/16io_uring: optimise io_free_batch_list()fs/io_uring.c+4 -20 replies
11/16io_uring: control ->async_data with a REQ_F flagfs/io_uring.c+46 -260 replies
12/16io_uring: remove struct io_completionfs/io_uring.c+7 -170 replies
13/16io_uring: inline io_req_needs_clean()fs/io_uring.c+1 -60 replies
14/16io_uring: inline io_poll_completefs/io_uring.c+2 -110 replies
15/16io_uring: correct fill events helpers typesfs/io_uring.c+12 -120 replies
16/16io_uring: mark hot functionsfs/io_uring.c+18 -160 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [PATCH 00/16] squeeze more performance

fio/t/io_uring -s32 -d32 -c32 -N1

          | baseline  | 0-15      | 0-16        | diff
setup 1:  | 34 MIOPS  | 42 MIOPS  | 42.2  MIOPS | 25 %
setup 2:  | 31 MIOPS  | 31 MIOPS  | 32    MIOPS | ~3 $

Setup 1 gets 25% performance improvement, which is unexpected and a share of it should be accounted as compiler/HW magic. Setup 2 is just 3%, but the catch is that some of the patches _very_ unexpectedly sink performance, so it's more like 31 MIOPS -> 29 -> 30 -> 29 -> 31 -> 32

I'd suggest to leave 16/16 aside, maybe for future consideration and refinement. The end result is not very clear, I'd expect probably around 3-5% with a more stable setup for nops32, and a better win for io_cqring_ev_posted() intensive cases like BPF.

Pavel Begunkov (16):
io_uring: optimise kiocb layout
io_uring: add more likely/unlikely() annotations
io_uring: delay req queueing into compl-batch list
io_uring: optimise request allocation
io_uring: optimise INIT_WQ_LIST
io_uring: don't wake sqpoll in io_cqring_ev_posted
io_uring: merge CQ and poll waitqueues
io_uring: optimise ctx referencing by requests
io_uring: mark cold functions
io_uring: optimise io_free_batch_list()
io_uring: control ->async_data with a REQ_F flag
io_uring: remove struct io_completion
io_uring: inline io_req_needs_clean()
io_uring: inline io_poll_complete
io_uring: correct fill events helpers types
io_uring: mark hot functions

 fs/io-wq.h    |   1 -
 fs/io_uring.c | 390 ++++++++++++++++++++++++++------------------------
 2 files changed, 205 insertions(+), 186 deletions(-)
Signature · 1 line
2.33.0