All of lore.kernel.org
 help / color / mirror / Atom feed
From: Avi Kivity <avi@redhat.com>
To: Joerg Roedel <joro@8bytes.org>
Cc: Alexander Graf <agraf@suse.de>,
	Joerg Roedel <joerg.roedel@amd.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: Re: [Qemu-devel] [PATCH 2/2] qemu-kvm: Add svm cpuid features
Date: Sun, 12 Sep 2010 16:36:58 +0200	[thread overview]
Message-ID: <4C8CE58A.80806@redhat.com> (raw)
In-Reply-To: <20100912143049.GE680@8bytes.org>

  On 09/12/2010 04:30 PM, Joerg Roedel wrote:
>
>>> Either way, I don't think we need a phenom2 type. The features
>>> additional are minor enough to not really matter and all use cases I
>>> can come up with require either -cpu host (local virt) or -cpu phenom
>>> (migration).
>> I'm fine with this (or with adding phenom2).  But don't make phenom
>> contain flags that real phenoms don't have.
> How about features that are not supported by the hardware but can be
> supported in emulation? The VMCBCLEAN feature is one of those which
> makes a lot of sense to reduce the emulated world-switch times. I guess
> its ok to enable those with -cpu host?
>
>

It's a good question.  We do that with x2apic, so yes.

Userspace can do four things with KVM_GET_SUPPORTED_CPUID:
- feed it right back to KVM_SET_CPUID2, getting maximum features and 
hopefully performance
- use it to mask the real cpuid which it fetches directly, getting 
something as similar as possible to the host
- use it to mask a predefined cpu type, getting something as similar as 
possible to that cpu type
- use it to verify that a predefined cpu type is supported, getting 
somthing exactly the same as that cpu type

-cpu host is the first option.  If the need comes for the second option, 
we can provide it.

I'll note that KVM_GET_SUPPORTED_CPUID can return values not present in 
the host cpuid in the documentation.

-- 
error compiling committee.c: too many arguments to function


WARNING: multiple messages have this Message-ID (diff)
From: Avi Kivity <avi@redhat.com>
To: Joerg Roedel <joro@8bytes.org>
Cc: Joerg Roedel <joerg.roedel@amd.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	Alexander Graf <agraf@suse.de>,
	kvm@vger.kernel.org, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 2/2] qemu-kvm: Add svm cpuid features
Date: Sun, 12 Sep 2010 16:36:58 +0200	[thread overview]
Message-ID: <4C8CE58A.80806@redhat.com> (raw)
In-Reply-To: <20100912143049.GE680@8bytes.org>

  On 09/12/2010 04:30 PM, Joerg Roedel wrote:
>
>>> Either way, I don't think we need a phenom2 type. The features
>>> additional are minor enough to not really matter and all use cases I
>>> can come up with require either -cpu host (local virt) or -cpu phenom
>>> (migration).
>> I'm fine with this (or with adding phenom2).  But don't make phenom
>> contain flags that real phenoms don't have.
> How about features that are not supported by the hardware but can be
> supported in emulation? The VMCBCLEAN feature is one of those which
> makes a lot of sense to reduce the emulated world-switch times. I guess
> its ok to enable those with -cpu host?
>
>

It's a good question.  We do that with x2apic, so yes.

Userspace can do four things with KVM_GET_SUPPORTED_CPUID:
- feed it right back to KVM_SET_CPUID2, getting maximum features and 
hopefully performance
- use it to mask the real cpuid which it fetches directly, getting 
something as similar as possible to the host
- use it to mask a predefined cpu type, getting something as similar as 
possible to that cpu type
- use it to verify that a predefined cpu type is supported, getting 
somthing exactly the same as that cpu type

-cpu host is the first option.  If the need comes for the second option, 
we can provide it.

I'll note that KVM_GET_SUPPORTED_CPUID can return values not present in 
the host cpuid in the documentation.

-- 
error compiling committee.c: too many arguments to function

  reply	other threads:[~2010-09-12 14:37 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-09-10 15:38 [PATCH 0/2] Add SVM feature flags to qemu Joerg Roedel
2010-09-10 15:38 ` [Qemu-devel] " Joerg Roedel
2010-09-10 15:38 ` [PATCH 1/2] qemu-kvm: Set cpuid definition to 0 before initializing it Joerg Roedel
2010-09-10 15:38   ` [Qemu-devel] " Joerg Roedel
2010-09-10 15:38 ` [PATCH 2/2] qemu-kvm: Add svm cpuid features Joerg Roedel
2010-09-10 15:38   ` [Qemu-devel] " Joerg Roedel
2010-09-11 13:43   ` Alexander Graf
2010-09-11 13:43     ` Alexander Graf
2010-09-11 14:20     ` Joerg Roedel
2010-09-11 14:20       ` Joerg Roedel
2010-09-11 14:29       ` Alexander Graf
2010-09-11 14:29         ` Alexander Graf
2010-09-11 14:36         ` Joerg Roedel
2010-09-11 14:36           ` Joerg Roedel
2010-09-11 14:38           ` Alexander Graf
2010-09-11 14:38             ` Alexander Graf
2010-09-11 14:42             ` Joerg Roedel
2010-09-11 14:42               ` Joerg Roedel
2010-09-11 14:45               ` Alexander Graf
2010-09-11 14:45                 ` Alexander Graf
2010-09-12  6:05       ` Avi Kivity
2010-09-12  6:05         ` Avi Kivity
2010-09-12  7:16         ` Alexander Graf
2010-09-12  7:16           ` Alexander Graf
2010-09-12  8:01           ` Avi Kivity
2010-09-12  8:01             ` Avi Kivity
2010-09-12 10:06             ` Alexander Graf
2010-09-12 10:06               ` Alexander Graf
2010-09-12 10:22               ` Avi Kivity
2010-09-12 10:22                 ` Avi Kivity
2010-09-12 11:14                 ` Alexander Graf
2010-09-12 11:14                   ` Alexander Graf
2010-09-12 12:02                   ` Avi Kivity
2010-09-12 12:02                     ` Avi Kivity
2010-09-12 14:30             ` Joerg Roedel
2010-09-12 14:30               ` Joerg Roedel
2010-09-12 14:36               ` Avi Kivity [this message]
2010-09-12 14:36                 ` Avi Kivity
  -- strict thread matches above, loose matches on Subject: below --
2010-09-18 22:16 garymberger

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=4C8CE58A.80806@redhat.com \
    --to=avi@redhat.com \
    --cc=agraf@suse.de \
    --cc=joerg.roedel@amd.com \
    --cc=joro@8bytes.org \
    --cc=kvm@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.