qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Andre Przywara <andre.przywara@amd.com>
To: Samuel Thibault <samuel.thibault@gnu.org>
Cc: qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: Re: [Qemu-devel] [RFC] allow multi-core guests: introduce cores= option to -cpu
Date: Sat, 4 Jul 2009 01:28:43 +0200	[thread overview]
Message-ID: <4A4E942B.3070207@amd.com> (raw)
In-Reply-To: <20090703145229.GU32062@const.bordeaux.inria.fr>

Samuel Thibault wrote:
> Andre Przywara, le Fri 03 Jul 2009 16:41:56 +0200, a écrit :
>> -smp 16 -cpu host,cores=8
> 
> That means 8 cores with 2 threads each, thus 16 threads?
No, that meant: 16 vCPUs total with 8 cores per physical packages. I 
don't have any notion for threads in the current code, although I agree 
that this should be introduced.
AFAIK Windows XP Home is limited to one socket, where XP Pro can use up 
to two sockets. Using -smp 4 on a XP Pro (64bit) caused an early BSOD 
stating "MULTIPROCESSOR_CONFIGURATION_NOT_SUPPORTED" (because it saw 
four sockets instead of four cores).
I think WindowsXP Home should even be worse (though I don't have an 
image here to test it), because it would not boot any SMP guest.

> Ok, that can be
> later generalized into for instance
> 
> -smp 16 -cpu host,nodes=2,sockets=2,cores=2
> 
> to define 2 NUMA nodes of 2 sockets of 2 cores, each core thus having 16
> threads.
Well, generalizing is a good idea, but as Brian wrote this shouldn't be 
done in the -cpu option. Putting the NUMA nodes into here seems quite 
logical, but would also requires to
a) describe the NUMA topology separated from this option    or
b) pull all the NUMA topology description into this one.
Maybe one could describe cores, threads, sockets and nodes in -smp and 
declare the memory topology only in -numa.

Regards,
Andre.

-- 
Andre Przywara
AMD-Operating System Research Center (OSRC), Dresden, Germany
Tel: +49 351 488-3567-12

  reply	other threads:[~2009-07-03 23:28 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-07-03 14:41 [Qemu-devel] [RFC] allow multi-core guests: introduce cores= option to -cpu Andre Przywara
2009-07-03 14:52 ` Samuel Thibault
2009-07-03 23:28   ` Andre Przywara [this message]
2009-07-03 23:53     ` Samuel Thibault
2009-07-03 15:16 ` [Qemu-devel] " Brian Jackson
2009-07-03 22:52   ` Andre Przywara
2009-07-04  0:04     ` Jamie Lokier
2009-07-04  7:18     ` Gleb Natapov
2009-07-03 15:46 ` [Qemu-devel] " Paul Brook
2009-07-03 23:45   ` Andre Przywara
2009-07-04  5:58     ` Paul Brook
2009-07-04 15:25 ` [Qemu-devel] " Avi Kivity
2009-07-05 13:23   ` Alexander Graf
2009-07-05 14:53     ` Avi Kivity
2009-07-05 15:04       ` Gleb Natapov
2009-07-05 15:11         ` Avi Kivity
2009-07-05 15:11           ` Alexander Graf
2009-07-05 15:17             ` 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=4A4E942B.3070207@amd.com \
    --to=andre.przywara@amd.com \
    --cc=kvm@vger.kernel.org \
    --cc=qemu-devel@nongnu.org \
    --cc=samuel.thibault@gnu.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).