qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Daniel P. Berrangé" <berrange@redhat.com>
To: Peter Lieven <pl@kamp.de>
Cc: "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Subject: Re: Limiting per vCPU Usage
Date: Mon, 21 Sep 2020 15:09:45 +0100	[thread overview]
Message-ID: <20200921140945.GO1942072@redhat.com> (raw)
In-Reply-To: <ddc410b9-2c15-4398-86a9-df3441e62138@kamp.de>

On Mon, Sep 21, 2020 at 03:47:56PM +0200, Peter Lieven wrote:
> Am 21.09.20 um 10:29 schrieb Daniel P. Berrangé:
> > On Sun, Sep 20, 2020 at 10:24:41PM +0200, Peter Lieven wrote:
> > > Hi Qemu folks,
> > > 
> > > 
> > > is there a BCP to limit just the maximum usage of a virtual (KVM) cpu?
> > > 
> > > I know that there are many approaches, but as far as I know they all limit the complete qemu process which is far more
> > > 
> > > than just the virtual CPUs.
> > > 
> > > Is it possible to limit just the vCPU threads and leave threads that
> > > handle the monitor, vnc, qmp, iothreads, storage backend etc.
> > Libvirt uses cgroups to achieve what you describe. A cgroup for the
> > QEMU process as a whole, then create child cgroups, one for each
> > vCPUs, and a further one for non-vCPU threads. CPU limits are then
> > applied on the child cgroups.
> 
> 
> Hi Daniel,
> 
> 
> I came up with the same solution. Is there a nice way to get the thread ids of the vCPUs?
> 
> Iterating over all thread IDs and check the names works, but the thread names are not considered
> 
> a stable API (text from original commit message).

The QMP  query-cpus  command tells you thread IDs. There is similar for
I/O threads too.  This is what libvirt uses to determine the mappings.

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|



      reply	other threads:[~2020-09-21 14:12 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-20 20:24 Limiting per vCPU Usage Peter Lieven
2020-09-21  8:29 ` Daniel P. Berrangé
2020-09-21 13:47   ` Peter Lieven
2020-09-21 14:09     ` Daniel P. Berrangé [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=20200921140945.GO1942072@redhat.com \
    --to=berrange@redhat.com \
    --cc=pl@kamp.de \
    --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).