From: Jens Axboe <axboe@kernel.dk>
To: Pavel Begunkov <asml.silence@gmail.com>, io-uring@vger.kernel.org
Subject: Re: [PATCH 00/16] squeeze more performance
Date: Mon, 4 Oct 2021 14:33:10 -0600 [thread overview]
Message-ID: <af5f6b50-f1e7-7647-4477-57e2942018d5@kernel.dk> (raw)
In-Reply-To: <6c6d9fde-5073-792d-312e-a57ee2a09598@kernel.dk>
On 10/4/21 2:19 PM, Jens Axboe wrote:
> On 10/4/21 1:02 PM, Pavel Begunkov wrote:
>> 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.
>
> Looks and tests good for me. I've skipped 16/16 for now, we can
> evaluate that one later.
For reference, running this on just the faster box:
Setup/Test | Peak-1-thread Peak-2-threads NOPS Diff
------------------------------------------------------------------
Setup 2 pre | 5.07M 5.74M 71.1M
Setup 2 post | 5.23M 5.84M 73.9M
which is a pretty substantial win.
--
Jens Axboe
prev parent reply other threads:[~2021-10-04 20:33 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-04 19:02 [PATCH 00/16] squeeze more performance Pavel Begunkov
2021-10-04 19:02 ` [PATCH 01/16] io_uring: optimise kiocb layout Pavel Begunkov
2021-10-04 19:02 ` [PATCH 02/16] io_uring: add more likely/unlikely() annotations Pavel Begunkov
2021-10-04 19:02 ` [PATCH 03/16] io_uring: delay req queueing into compl-batch list Pavel Begunkov
2021-10-04 19:02 ` [PATCH 04/16] io_uring: optimise request allocation Pavel Begunkov
2021-10-04 19:02 ` [PATCH 05/16] io_uring: optimise INIT_WQ_LIST Pavel Begunkov
2021-10-04 19:02 ` [PATCH 06/16] io_uring: don't wake sqpoll in io_cqring_ev_posted Pavel Begunkov
2021-10-04 19:02 ` [PATCH 07/16] io_uring: merge CQ and poll waitqueues Pavel Begunkov
2021-10-04 19:02 ` [PATCH 08/16] io_uring: optimise ctx referencing by requests Pavel Begunkov
2021-10-04 19:02 ` [PATCH 09/16] io_uring: mark cold functions Pavel Begunkov
2021-10-04 19:02 ` [PATCH 10/16] io_uring: optimise io_free_batch_list() Pavel Begunkov
2021-10-04 19:02 ` [PATCH 11/16] io_uring: control ->async_data with a REQ_F flag Pavel Begunkov
2021-10-04 19:02 ` [PATCH 12/16] io_uring: remove struct io_completion Pavel Begunkov
2021-10-04 19:02 ` [PATCH 13/16] io_uring: inline io_req_needs_clean() Pavel Begunkov
2021-10-04 19:02 ` [PATCH 14/16] io_uring: inline io_poll_complete Pavel Begunkov
2021-10-04 19:03 ` [PATCH 15/16] io_uring: correct fill events helpers types Pavel Begunkov
2021-10-04 19:03 ` [PATCH 16/16] io_uring: mark hot functions Pavel Begunkov
2021-10-04 20:19 ` [PATCH 00/16] squeeze more performance Jens Axboe
2021-10-04 20:33 ` Jens Axboe [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=af5f6b50-f1e7-7647-4477-57e2942018d5@kernel.dk \
--to=axboe@kernel.dk \
--cc=asml.silence@gmail.com \
--cc=io-uring@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).