qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Alexey Kardashevskiy <aik@ozlabs.ru>
To: Alexander Graf <agraf@suse.de>, Mike Day <ncmike@ncultra.org>
Cc: Paul Mackerras <paulus@samba.org>,
	"qemu-ppc@nongnu.org" <qemu-ppc@nongnu.org>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] [RFC PATCH v2] PPC: smp: autodetect numbers of threads per core
Date: Sat, 11 Jan 2014 01:35:29 +1100	[thread overview]
Message-ID: <52D00531.8030105@ozlabs.ru> (raw)
In-Reply-To: <B7FAAA2E-85B6-4F11-9422-78B93EC35F3A@suse.de>

On 01/11/2014 01:25 AM, Alexander Graf wrote:
> 
> On 10.01.2014, at 15:21, Mike Day <ncmike@ncultra.org> wrote:
> 
>> On Fri, Jan 10, 2014 at 9:13 AM, Alexey Kardashevskiy <aik@ozlabs.ru>
>> wrote:
>>> On 01/11/2014 01:00 AM, Alexander Graf wrote:
>> 
>>>> Can't we determine the number of "default threads" at a common
>>>> place, preferably derived from cpu type?
>>> 
>>> We can do anything. I asked how exactly as I really (really) do not 
>>> understand the details.
>>> 
>> 
>> Are you suggesting we create a dictionary with all the cpu type 
>> information stored in it (stepping, cores, threads, memory channels,
>> caches) that we need to keep updated?
> 
> We can always talk in extremes :). Today we have a dictionary of core
> types in QEMU. If a certain core type comes with a specific number of
> threads, that's a property of the core, no?


No, the max smt number comes from cap_ppc_smt, not from CPU type. PR KVM
may return something different (what does it return?) and TCG definitely
will not allow anything but 1. Up to the accelerator, not a CPU class.


-- 
Alexey

  parent reply	other threads:[~2014-01-10 14:35 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-09  5:34 [Qemu-devel] [RFC PATCH v2] PPC: smp: autodetect numbers of threads per core Alexey Kardashevskiy
2014-01-09 21:00 ` Mike Day
2014-01-09 22:12   ` Alexey Kardashevskiy
2014-01-09 23:40     ` Alexander Graf
2014-01-10  1:39       ` Alexey Kardashevskiy
2014-01-10 13:03         ` Mike Day
2014-01-10 13:28           ` Alexander Graf
2014-01-10 13:42             ` Alexey Kardashevskiy
2014-01-10 14:00               ` Alexander Graf
2014-01-10 14:13                 ` Alexey Kardashevskiy
2014-01-10 14:20                   ` Alexander Graf
2014-01-10 14:21                   ` Mike Day
2014-01-10 14:25                     ` Alexander Graf
2014-01-10 14:29                       ` Mike Day
2014-01-10 14:35                       ` Alexey Kardashevskiy [this message]
2014-01-10 14:12             ` Mike Day
2014-01-09 22:50 ` Scott Wood

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=52D00531.8030105@ozlabs.ru \
    --to=aik@ozlabs.ru \
    --cc=agraf@suse.de \
    --cc=ncmike@ncultra.org \
    --cc=paulus@samba.org \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@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).