kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Erik Jacobson <erikj@sgi.com>
To: Avi Kivity <avi@redhat.com>
Cc: Erik Jacobson <erikj@sgi.com>, kvm@vger.kernel.org
Subject: Re: slow guest performance with build load, looking for ideas
Date: Sun, 28 Jun 2009 14:05:00 -0500	[thread overview]
Message-ID: <20090628190500.GA7347@sgi.com> (raw)
In-Reply-To: <4A477B8C.2010502@redhat.com>

>>   * In some of the timing runs on this system, the "real time" reported by
>>     the time command was off by 10 to 11 times.  Issues were found in
>>     the messages file that seemed to relate to this including HUGE time
>>     adjustments by NTP and kernel hrtimer 'interrupt too slow' messages.
>>     This specific problem seems to be intermittent.
> This is on the host? It can easily ruin your day.

This was in the guest.

>> System:
>> SGI XE270, 8-core, Xeon X5570 (Nehalem), Hyperthreading turned off
> Shoot, was about to blame hyperthreading.

I'll keep it off for the next attempted run.

> kvm guests should have an accurate clock without ntp in the guest  
> (/sys/.../current_clocksource should say 'kvmclock').

OK thanks.

> Can you post kvm_stat output during the run?

Sure, I'll try to get time on the system again next week and post in
to the thread again.  We'll still have the issue with the non-sequential
nodes and incorrect representation of memory for this two-socket
Nehalem system.  I don't think that patch has made it in to the kernel.

Thanks for replying back.

If you have any other things you'd suggest trying, I'm game to give it a
whirl.  Someone suggested trying to export a whole PCI device to the guest.
I won't be able to do that on this machine, maybe later when I have physical
access to the system.  Besides, that exercise might not poke at what I'm
interested in anyway.

Others suggested some potential settings EPT (Extended Page Table) and
VPID (Virtual Path Identifier?) but I don't see where these settings are
exposed (they aren't, for example, in this system's BIOS).

More to come then.  Thanks.

Erik

  reply	other threads:[~2009-06-28 19:04 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-06-12 21:04 slow guest performance with build load, looking for ideas Erik Jacobson
2009-06-14  9:33 ` Avi Kivity
2009-06-15 14:15   ` Erik Jacobson
2009-06-15 14:24     ` Avi Kivity
2009-06-15 15:25       ` Michael Tokarev
2009-06-15 15:27         ` Avi Kivity
2009-06-16  7:03           ` Michael Tokarev
2009-06-16  8:07             ` Avi Kivity
2009-06-18 23:07   ` Erik Jacobson
2009-06-28 14:17     ` Avi Kivity
2009-06-28 19:05       ` Erik Jacobson [this message]
2009-06-28 21:28         ` Avi Kivity
2009-07-01 21:41           ` Erik Jacobson
2009-07-02  5:48             ` Avi Kivity
2009-07-02  9:41               ` Avi Kivity
2009-07-03 15:43                 ` Mark McLoughlin
2009-07-03 16:28                   ` Erik Jacobson
2009-07-09  2:36                   ` Erik Jacobson
2009-07-09  7:48                     ` Avi Kivity
2009-07-09 18:01                       ` Erik Jacobson
2009-07-12  8:38                         ` Avi Kivity
2009-07-16 12:20                           ` Jes Sorensen
2009-07-25 14:33                             ` Avi Kivity
2009-07-03 10:41   ` Matty
2009-07-05  8:07     ` Avi Kivity

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=20090628190500.GA7347@sgi.com \
    --to=erikj@sgi.com \
    --cc=avi@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;
as well as URLs for NNTP newsgroup(s).