qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Jun Piao <piaojun@huawei.com>
To: <kraxel@redhat.com>
Cc: qemu-devel@nongnu.org, jianjay.zhou@huawei.com
Subject: Re: [QUESTION] How to improve virtio-gpu performance
Date: Tue, 26 May 2020 09:02:17 +0800	[thread overview]
Message-ID: <674b5f0b-c926-ec62-2442-c70b5f476876@huawei.com> (raw)
In-Reply-To: <65461f86-c44b-6009-51f3-333b79fb612a@huawei.com>

ping?

在 2020/5/22 10:39, Jun Piao 写道:
> Hi Gerd,
> 
> I'm using glmark2 to test virtio-gpu on AMDGPU, but the result is
> unsatisfactory.
> 
> GPU: AMD Radeon Pro WX 5100
> Host OS: CentOS Linux release 7.7.1908 (AltArch)
>   --QEMU: 4.0.0
>   --linux kernel 4.18
>   --mesa 18.3
>   --virglrenderer
>   --spice server 0.14.2
>   --spice-gtk3-0.35
>   --libepoxy 1.5.2
> Guest OS: CentOS Linux release 7.7.1908 (AltArch)
>   --linux kernel 4.18
>   --mesa 18.3
>   --xorg server 1.20
> TestSuit: glmark2
> 
> Host result:
> glmark2 Score: 10043
> 
> Single Guest result:
> glmark2 Score: 2209
> 
> Multi Guest result:
> Guest1: glmark2 Score: 1136
> Guest2: glmark2 Score: 1231
> Guest3: glmark2 Score: 1282
> Guest4: glmark2 Score: 1175
> Guest5: glmark2 Score: 1398
> 
> It seems the whole performance of multi-guests is better than single
> guest, but each one has lower performance. I change 'ctrl_vq' from 256
> to 1024, but has little effect. I wonder the result is normal or
> abnormal? And which is the bottleneck? I'm appreciated for you help.
> 
> thanks,
> Jun
> 


      reply	other threads:[~2020-05-26  1:03 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-22  2:39 [QUESTION] How to improve virtio-gpu performance Jun Piao
2020-05-26  1:02 ` Jun Piao [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=674b5f0b-c926-ec62-2442-c70b5f476876@huawei.com \
    --to=piaojun@huawei.com \
    --cc=jianjay.zhou@huawei.com \
    --cc=kraxel@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /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).