All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stefan Hajnoczi <stefanha@redhat.com>
To: Zhenyu Ye <yezhenyu2@huawei.com>
Cc: fam@euphon.net, Kevin Wolf <kwolf@redhat.com>,
	qemu-block@nongnu.org, qemu-devel@nongnu.org,
	xiexiangyou@huawei.com, armbru@redhat.com, pbonzini@redhat.com,
	mreitz@redhat.com
Subject: Re: [PATCH v1 0/2] Add timeout mechanism to qmp actions
Date: Tue, 13 Oct 2020 11:00:33 +0100	[thread overview]
Message-ID: <20201013100033.GB164611@stefanha-x1.localdomain> (raw)
In-Reply-To: <20200921111435.GA2524022@dev>

[-- Attachment #1: Type: text/plain, Size: 1724 bytes --]

On Mon, Sep 21, 2020 at 11:14:35AM +0000, Fam Zheng wrote:
> On 2020-09-19 10:22, Zhenyu Ye wrote:
> > On 2020/9/18 22:06, Fam Zheng wrote:
> > > 
> > > I can see how blocking in a slow io_submit can cause trouble for main
> > > thread. I think one way to fix it (until it's made truly async in new
> > > kernels) is moving the io_submit call to thread pool, and wrapped in a
> > > coroutine, perhaps.
> > >
> > 
> > I'm not sure if any other operation will block the main thread, other
> > than io_submit().
> 
> Then that's a problem with io_submit which should be fixed. Or more
> precisely, that is a long held lock that we should avoid in QEMU's event
> loops.
> 
> > 
> > > I'm not sure qmp timeout is a complete solution because we would still
> > > suffer from a blocked state for a period, in this exact situation before
> > > the timeout.
> > 
> > Anyway, the qmp timeout may be the last measure to prevent the VM
> > soft lockup. 
> 
> Maybe, but I don't think baking such a workaround into the QMP API is a
> good idea. No QMP command should be synchronously long running, so
> having a timeout parameter is just a wrong design.

Sorry, I lost track of this on-going email thread.

Thanks for the backtrace. It shows the io_submit call is done while the
AioContext lock is held. The monitor thread is waiting for the
IOThread's AioContext lock. vcpus threads can get stuck waiting on the
big QEMU lock (BQL) that is held by the monitor in the meantime.

Please collect the kernel backtrace for io_submit so we can understand
why multi-second io_submit latencies happen.

I also suggest trying aio=io_uring to check if Linux io_uring avoids the
latency problem.

Stefan

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

  reply	other threads:[~2020-10-13 10:02 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-10 14:52 [PATCH v1 0/2] Add timeout mechanism to qmp actions Zhenyu Ye
2020-08-10 14:52 ` [PATCH v1 1/2] util: introduce aio_context_acquire_timeout Zhenyu Ye
2020-08-10 14:52 ` [PATCH v1 2/2] qmp: use aio_context_acquire_timeout replace aio_context_acquire Zhenyu Ye
2020-08-10 15:38 ` [PATCH v1 0/2] Add timeout mechanism to qmp actions Kevin Wolf
2020-08-11 13:54   ` Zhenyu Ye
2020-08-21 12:52     ` Stefan Hajnoczi
2020-09-14 13:27     ` Stefan Hajnoczi
2020-09-17  7:36       ` Zhenyu Ye
2020-09-17 10:10         ` Fam Zheng
2020-09-17 15:44         ` Stefan Hajnoczi
2020-09-17 16:01           ` Fam Zheng
2020-09-18 11:23             ` Zhenyu Ye
2020-09-18 14:06               ` Fam Zheng
2020-09-19  2:22                 ` Zhenyu Ye
2020-09-21 11:14                   ` Fam Zheng
2020-10-13 10:00                     ` Stefan Hajnoczi [this message]
2020-10-19 12:40                       ` Zhenyu Ye
2020-10-19 13:25                         ` Paolo Bonzini
2020-10-20  1:34                           ` Zhenyu Ye
2020-10-22 16:29                             ` Fam Zheng
2020-12-08 13:10                               ` Stefan Hajnoczi
2020-12-08 13:47                                 ` Glauber Costa
2020-12-14 16:33                                   ` Stefan Hajnoczi
2020-12-21 11:30                                     ` Zhenyu Ye
2020-09-14 14:42     ` Daniel P. Berrangé
2020-09-17  8:12       ` Zhenyu Ye
2020-08-12 13:51 ` Stefan Hajnoczi
2020-08-13  1:51   ` Zhenyu Ye

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=20201013100033.GB164611@stefanha-x1.localdomain \
    --to=stefanha@redhat.com \
    --cc=armbru@redhat.com \
    --cc=fam@euphon.net \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=xiexiangyou@huawei.com \
    --cc=yezhenyu2@huawei.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.