public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Marcelo Tosatti <mtosatti@redhat.com>
To: Glauber Costa <glommer@gmail.com>
Cc: Anthony Liguori <aliguori@us.ibm.com>,
	Glauber Costa <gcosta@redhat.com>,
	kvm-devel <kvm@vger.kernel.org>,
	kraxel@redhat.com, chrisw@redhat.com
Subject: Re: kvm guest loops_per_jiffy miscalibration under host load
Date: Mon, 7 Jul 2008 15:48:43 -0300	[thread overview]
Message-ID: <20080707184843.GA14634@dmt.cnet> (raw)
In-Reply-To: <5d6222a80807071127s1ed157eaje2f1fcf511a34813@mail.gmail.com>

On Mon, Jul 07, 2008 at 03:27:16PM -0300, Glauber Costa wrote:
> > I agree.  A paravirt solution solves the problem.
> 
> Please, look at the patch I've attached.
> 
> It does  __delay with host help. This may have the nice effect of not
> busy waiting for long-enough delays, and may well.
> 
> It is _completely_ PoC, just to show the idea. It's ugly, broken,
> obviously have to go through pv-ops, etc.
> 
> Also, I intend to add a lpj field in the kvm clock memory area. We
> could do just this later, do both, etc.
> 
> If we agree this is a viable solution, I'll start working on a patch

This stops interrupts from being processed during the delay. And also 
there are cases like this recently introduced break:

                /* Allow RT tasks to run */
                preempt_enable();
                rep_nop();
                preempt_disable();

I think it would be better to just pass the lpj value via paravirt and
let the guest busy-loop as usual.


  reply	other threads:[~2008-07-07 18:50 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-07-02 16:40 kvm guest loops_per_jiffy miscalibration under host load Marcelo Tosatti
2008-07-03 13:17 ` Glauber Costa
2008-07-04 22:51   ` Marcelo Tosatti
2008-07-07  1:56   ` Anthony Liguori
2008-07-07 18:27     ` Glauber Costa
2008-07-07 18:48       ` Marcelo Tosatti [this message]
2008-07-07 19:21         ` Anthony Liguori
2008-07-07 19:32           ` Glauber Costa
2008-07-07 21:35             ` Glauber Costa
2008-07-11 21:18               ` David S. Ahern
2008-07-12 14:10                 ` Marcelo Tosatti
2008-07-12 19:28                   ` David S. Ahern
2008-07-07 18:17 ` Daniel P. Berrange
  -- strict thread matches above, loose matches on Subject: below --
2008-07-22  3:25 Marcelo Tosatti
2008-07-22  8:22 ` Jan Kiszka
2008-07-22 12:49   ` Marcelo Tosatti
2008-07-22 15:54     ` Jan Kiszka
2008-07-22 22:00     ` Dor Laor
2008-07-22 19:56 ` David S. Ahern
2008-07-23  2:57   ` David S. Ahern
2008-07-29 14:58   ` Marcelo Tosatti

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=20080707184843.GA14634@dmt.cnet \
    --to=mtosatti@redhat.com \
    --cc=aliguori@us.ibm.com \
    --cc=chrisw@redhat.com \
    --cc=gcosta@redhat.com \
    --cc=glommer@gmail.com \
    --cc=kraxel@redhat.com \
    --cc=kvm@vger.kernel.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