From: Jens Axboe <[email protected]>
To: Pavel Begunkov <[email protected]>, [email protected]
Subject: Re: [RFC 0/4] completion locking optimisation feature
Date: Fri, 18 Mar 2022 08:52:05 -0600 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
On 3/18/22 8:42 AM, Pavel Begunkov wrote:
> On 3/18/22 13:52, Pavel Begunkov wrote:
>> A WIP feature optimising out CQEs posting spinlocking for some use cases.
>> For a more detailed description see 4/4.
>>
>> Quick benchmarking with fio/t/io_uring nops gives extra 4% to throughput for
>> QD=1, and ~+2.5% for QD=4.
>
> Non-io_uring overhead (syscalls + userspace) takes ~60% of all execution
> time, so the percentage should quite depend on the CPU and the kernel config.
> Likely to be more than 4% for a faster setup.
>
> fwiw, was also usingIORING_ENTER_REGISTERED_RING, if it's not yet included
> in the upstream version of the tool.
But that seems to be exclusive of using PRIVATE_CQ?
> Also, want to play after to see if we can also avoid taking uring_lock.
That would certainly also be interesting!
--
Jens Axboe
next prev parent reply other threads:[~2022-03-18 14:52 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-18 13:52 [RFC 0/4] completion locking optimisation feature Pavel Begunkov
2022-03-18 13:52 ` [PATCH 1/4] io_uring: get rid of raw fill cqe in kill_timeout Pavel Begunkov
2022-03-18 13:52 ` [PATCH 2/4] io_uring: get rid of raw fill_cqe in io_fail_links Pavel Begunkov
2022-03-18 13:52 ` [PATCH 3/4] io_uring: remove raw fill_cqe from linked timeout Pavel Begunkov
2022-03-18 13:52 ` [PATCH 4/4] io_uring: optimise compl locking for non-shared rings Pavel Begunkov
2022-03-18 14:54 ` Jens Axboe
2022-03-18 15:13 ` Pavel Begunkov
2022-03-18 15:21 ` Jens Axboe
2022-03-18 15:32 ` Pavel Begunkov
2022-03-18 16:06 ` Jens Axboe
2022-03-18 14:42 ` [RFC 0/4] completion locking optimisation feature Pavel Begunkov
2022-03-18 14:52 ` Jens Axboe [this message]
2022-03-18 15:00 ` Pavel Begunkov
2022-03-18 15:22 ` Jens Axboe
2022-03-18 15:34 ` Pavel Begunkov
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 \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/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