public inbox for [email protected]
 help / color / mirror / Atom feed
From: Pavel Begunkov <[email protected]>
To: Jens Axboe <[email protected]>,
	[email protected], [email protected]
Subject: Re: [PATCH 1/4] io_uring: fix hanging iopoll in case of -EAGAIN
Date: Tue, 23 Jun 2020 14:57:38 +0300	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>

On 23/06/2020 05:18, Jens Axboe wrote:
> On 6/22/20 8:07 PM, Jens Axboe wrote:
>> On 6/22/20 4:16 PM, Pavel Begunkov wrote:
>>> io_do_iopoll() won't do anything with a request unless
>>> req->iopoll_completed is set. So io_complete_rw_iopoll() has to set
>>> it, otherwise io_do_iopoll() will poll a file again and again even
>>> though the request of interest was completed long ago.
>>
>> I need to look at this again, because with this change, I previously
>> got various use-after-free. I haven't seen any issues with it, but
>> I agree, from a quick look that I'm not quite sure how it's currently
>> not causing hangs. Yet I haven't seen any, with targeted -EAGAIN
>> testing.

Can io_complete_rw_iopoll() get -EAGAIN after being successfully enqueued
(i.e. EIOCBQUEUED)? It's reliably fails for me, because my hacked nullblk
_can_ (i.e. probabilistically returns BLK_STS_AGAIN from ->iopoll()).

> 
> Ah I think I know what it is - if we run into:
> 
> if (req->result == -EAGAIN)
> 	return -EAGAIN
> 
> in io_issue_sqe() and race with it, we'll reissue twice potentially.
> So the above isn't quite enough, we'll need something a bit broader.

I see, I'll deal with it.


-- 
Pavel Begunkov

  reply	other threads:[~2020-06-23 11:59 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-22 22:16 [PATCH 0/4] iopoll fixes + cleanups Pavel Begunkov
2020-06-22 22:16 ` [PATCH 1/4] io_uring: fix hanging iopoll in case of -EAGAIN Pavel Begunkov
2020-06-23  2:07   ` Jens Axboe
2020-06-23  2:18     ` Jens Axboe
2020-06-23 11:57       ` Pavel Begunkov [this message]
2020-06-23 19:01         ` Jens Axboe
2020-06-24 16:56           ` Pavel Begunkov
2020-06-22 22:16 ` [PATCH 2/4] io_uring: handle EAGAIN iopoll Pavel Begunkov
2020-06-30  4:01   ` kernel test robot
2020-06-22 22:16 ` [PATCH 3/4] io-wq: compact io-wq flags numbers Pavel Begunkov
2020-06-22 22:16 ` [PATCH 4/4] io-wq: return next work from ->do_work() directly 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] \
    [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