virtualization.lists.linux-foundation.org archive mirror
 help / color / mirror / Atom feed
From: Heng Qi <hengqi@linux.alibaba.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: netdev@vger.kernel.org, virtualization@lists.linux.dev,
	"Jason Wang" <jasowang@redhat.com>,
	"Xuan Zhuo" <xuanzhuo@linux.alibaba.com>,
	"Eugenio Pérez" <eperezma@redhat.com>,
	"David S. Miller" <davem@davemloft.net>,
	"Eric Dumazet" <edumazet@google.com>,
	"Jakub Kicinski" <kuba@kernel.org>,
	"Paolo Abeni" <pabeni@redhat.com>
Subject: Re: [PATCH net-next v4 0/5] virtio_net: enable the irq for ctrlq
Date: Thu, 20 Jun 2024 15:16:20 +0800	[thread overview]
Message-ID: <1718867780.8623598-4-hengqi@linux.alibaba.com> (raw)
In-Reply-To: <20240619171535-mutt-send-email-mst@kernel.org>

On Wed, 19 Jun 2024 17:16:57 -0400, "Michael S. Tsirkin" <mst@redhat.com> wrote:
> On Thu, Jun 20, 2024 at 12:19:03AM +0800, Heng Qi wrote:
> > Ctrlq in polling mode may cause the virtual machine to hang and
> > occupy additional CPU resources. Enabling the irq for ctrlq
> > alleviates this problem and allows commands to be requested
> > concurrently.
> 
> Any patch that is supposed to be a performance improvement
> has to come with actual before/after testing restults, not
> vague "may cause".

1. If the device does not respond in time, the CPU usage for ctrlq in the polling mode is
   ~100%, and in irq mode is ~0%;
2. If there are concurrent requests, the situation in 1 will be even worse;
3. On 64 queues with dim on, use nginx + wrk with the number of connections is 500:
   a. If ctrlq is in polling mode and concurrent requests are not supported:
      seeing that the dim worker occupies 20%+ CPU usage. Because of the
      large number of queues, dim requests cannot be concurrent, and the
      performance is unstable;
   b. If ctrlq is in irq mode and concurrent requests are supported: the overhead
      of the dim worker is not visible, and the pps increases by ~13%
      compared to a.

Thanks.

> 
> 
> 
> > Changelog
> > =========
> > v3->v4:
> >   - Turn off the switch before flush the get_cvq work.
> >   - Add interrupt suppression.
> > 
> > v2->v3:
> >   - Use the completion for dim cmds.
> > 
> > v1->v2:
> >   - Refactor the patch 1 and rephase the commit log.
> > 
> > Heng Qi (5):
> >   virtio_net: passing control_buf explicitly
> >   virtio_net: enable irq for the control vq
> >   virtio_net: change the command token to completion
> >   virtio_net: refactor command sending and response handling
> >   virtio_net: improve dim command request efficiency
> > 
> >  drivers/net/virtio_net.c | 309 ++++++++++++++++++++++++++++++++-------
> >  1 file changed, 260 insertions(+), 49 deletions(-)
> > 
> > -- 
> > 2.32.0.3.g01195cf9f
> 

      reply	other threads:[~2024-06-20  7:29 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-19 16:19 [PATCH net-next v4 0/5] virtio_net: enable the irq for ctrlq Heng Qi
2024-06-19 16:19 ` [PATCH net-next v4 1/5] virtio_net: passing control_buf explicitly Heng Qi
2024-06-19 16:19 ` [PATCH net-next v4 2/5] virtio_net: enable irq for the control vq Heng Qi
2024-06-19 21:19   ` Michael S. Tsirkin
2024-06-20  7:29     ` Heng Qi
2024-06-20  8:21       ` Jason Wang
2024-06-20  8:26         ` Jason Wang
2024-06-20  9:53           ` Heng Qi
2024-06-20 10:10             ` Michael S. Tsirkin
2024-06-20 10:11               ` Michael S. Tsirkin
2024-06-20 10:31                 ` Heng Qi
2024-06-26  7:52                   ` Jiri Pirko
2024-06-26  8:08                     ` Michael S. Tsirkin
2024-06-26  8:43                       ` Jiri Pirko
2024-06-26  9:58                         ` Michael S. Tsirkin
2024-06-26 11:51                           ` Jiri Pirko
2024-07-08 11:40                             ` Jiri Pirko
2024-07-08 12:19                               ` Heng Qi
2024-06-21  7:41                 ` Xuan Zhuo
2024-06-21 11:46                   ` Michael S. Tsirkin
2024-06-25  1:27                 ` Jason Wang
2024-06-25  7:14                   ` Michael S. Tsirkin
2024-06-20  8:32       ` Michael S. Tsirkin
2024-06-20  8:37         ` Jason Wang
2024-06-20  9:38         ` Heng Qi
2024-06-20 10:07           ` Michael S. Tsirkin
2024-06-24 11:30   ` Michael S. Tsirkin
2024-06-19 16:19 ` [PATCH net-next v4 3/5] virtio_net: change the command token to completion Heng Qi
2024-06-19 16:19 ` [PATCH net-next v4 4/5] virtio_net: refactor command sending and response handling Heng Qi
2024-06-19 16:19 ` [PATCH net-next v4 5/5] virtio_net: improve dim command request efficiency Heng Qi
2024-06-20  6:40   ` kernel test robot
2024-06-19 21:16 ` [PATCH net-next v4 0/5] virtio_net: enable the irq for ctrlq Michael S. Tsirkin
2024-06-20  7:16   ` Heng Qi [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=1718867780.8623598-4-hengqi@linux.alibaba.com \
    --to=hengqi@linux.alibaba.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=eperezma@redhat.com \
    --cc=jasowang@redhat.com \
    --cc=kuba@kernel.org \
    --cc=mst@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=virtualization@lists.linux.dev \
    --cc=xuanzhuo@linux.alibaba.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 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).