public inbox for [email protected]
 help / color / mirror / Atom feed
From: Pavel Begunkov <[email protected]>
To: Jens Axboe <[email protected]>, [email protected]
Subject: Re: [RFC v2 09/13] io_uring: separate wq for ring polling
Date: Wed, 4 Jan 2023 20:45:31 +0000	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>

On 1/4/23 20:34, Jens Axboe wrote:
> On 1/4/23 1:28?PM, Pavel Begunkov wrote:
>> On 1/4/23 18:08, Jens Axboe wrote:
>>> On 1/2/23 8:04?PM, Pavel Begunkov wrote:
>>>> Don't use ->cq_wait for ring polling but add a separate wait queue for
>>>> it. We need it for following patches.
>>>>
>>>> Signed-off-by: Pavel Begunkov <[email protected]>
>>>> ---
>>>>    include/linux/io_uring_types.h | 1 +
>>>>    io_uring/io_uring.c            | 3 ++-
>>>>    io_uring/io_uring.h            | 9 +++++++++
>>>>    3 files changed, 12 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h
>>>> index dcd8a563ab52..cbcd3aaddd9d 100644
>>>> --- a/include/linux/io_uring_types.h
>>>> +++ b/include/linux/io_uring_types.h
>>>> @@ -286,6 +286,7 @@ struct io_ring_ctx {
>>>>            unsigned        cq_entries;
>>>>            struct io_ev_fd    __rcu    *io_ev_fd;
>>>>            struct wait_queue_head    cq_wait;
>>>> +        struct wait_queue_head    poll_wq;
>>>>            unsigned        cq_extra;
>>>>        } ____cacheline_aligned_in_smp;
>>>>    
>>>
>>> Should we move poll_wq somewhere else, more out of the way?
>>
>> If we care about polling perf and cache collisions with
>> cq_wait, yeah we can. In any case it's a good idea to at
>> least move it after cq_extra.
>>
>>> Would need to gate the check a flag or something.
>>
>> Not sure I follow
> 
> I guess I could've been a bit more verbose... If we consider poll on the
> io_uring rather uncommon, then moving the poll_wq outside of the hotter
> cq_wait cacheline(s) would make sense. Each wait_queue_head is more than
> a cacheline.

Looks it's 24B, and wait_queue_entry is uncomfortable 40B.

> Then we could have a flag in a spot that's hot anyway
> whether to check it or not, eg in that same section as cq_wait.
> Looking at the layout right now, we're at 116 bytes for that section, or
> two cachelines with 12 bytes to spare. If we add poll_wq, then we'll be
> at 196 bytes, which is 4 bytes over the next cacheline. So it'd
> essentially double the size of that section. If we moved it outside of
> the aligned sections, then it'd pack better.

Than it's not about hotness and caches but rather memory
consumption due to padding, which is still a good argument.

-- 
Pavel Begunkov

  reply	other threads:[~2023-01-04 20:46 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-03  3:03 [RFC v2 00/13] CQ waiting and wake up optimisations Pavel Begunkov
2023-01-03  3:03 ` [RFC v2 01/13] io_uring: rearrange defer list checks Pavel Begunkov
2023-01-03  3:03 ` [RFC v2 02/13] io_uring: don't iterate cq wait fast path Pavel Begunkov
2023-01-03  3:03 ` [RFC v2 03/13] io_uring: kill io_run_task_work_ctx Pavel Begunkov
2023-01-03  3:03 ` [RFC v2 04/13] io_uring: move defer tw task checks Pavel Begunkov
2023-01-03  3:03 ` [RFC v2 05/13] io_uring: parse check_cq out of wq waiting Pavel Begunkov
2023-01-03  3:03 ` [RFC v2 06/13] io_uring: mimimise io_cqring_wait_schedule Pavel Begunkov
2023-01-03  3:03 ` [RFC v2 07/13] io_uring: simplify io_has_work Pavel Begunkov
2023-01-03  3:03 ` [RFC v2 08/13] io_uring: set TASK_RUNNING right after schedule Pavel Begunkov
2023-01-03  3:04 ` [RFC v2 09/13] io_uring: separate wq for ring polling Pavel Begunkov
2023-01-04 18:08   ` Jens Axboe
2023-01-04 20:28     ` Pavel Begunkov
2023-01-04 20:34       ` Jens Axboe
2023-01-04 20:45         ` Pavel Begunkov [this message]
2023-01-04 20:53           ` Jens Axboe
2023-01-04 20:52         ` Jens Axboe
2023-01-03  3:04 ` [RFC v2 10/13] io_uring: add lazy poll_wq activation Pavel Begunkov
2023-01-03  3:04 ` [RFC v2 11/13] io_uring: wake up optimisations Pavel Begunkov
2023-01-03  3:04 ` [RFC v2 12/13] io_uring: waitqueue-less cq waiting Pavel Begunkov
2023-01-03  3:04 ` [RFC v2 13/13] io_uring: add io_req_local_work_add wake fast path Pavel Begunkov
2023-01-04 18:05 ` (subset) [RFC v2 00/13] CQ waiting and wake up optimisations Jens Axboe
2023-01-04 20:25 ` 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