From: Avi Kivity <avi@qumranet.com>
To: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Chris Wright <chrisw@redhat.com>,
Glauber Costa <gcosta@redhat.com>,
Anthony Liguori <aliguori@us.ibm.com>,
kvm@vger.kernel.org
Subject: Re: [patch 00/12] fake ACPI C2 emulation v2
Date: Thu, 05 Jun 2008 10:56:21 +0300 [thread overview]
Message-ID: <48479C25.6080009@qumranet.com> (raw)
In-Reply-To: <20080605031228.GA13046@dmt.cnet>
Marcelo Tosatti wrote:
> On Wed, Jun 04, 2008 at 01:49:41PM +0300, Avi Kivity wrote:
>
>>> Test is 1 million gettimeofday calls, Xeon 1.60GHz with 4MB L2.
>>>
>>> guest (qemu emulation):
>>> cycles:1189759332
>>>
>>> guest (in-kernel emulation):
>>> cycles:628046412
>>>
>>> guest (direct pmtimer):
>>> cycles:230372934
>>>
>>> host (TSC):
>>> cycles:14862774
>>>
>>>
>> Ratio is 1:15:80
>>
>> Looks like direct pmtimer is still quite slow. Are there any exits with
>> direct pmtimer, or is it all due to the ioport latency?
>>
>
> host (pmtimer):
> cycles:225768390
>
> So its getting close-to-native performance. As you mentioned earlier,
> acpi_pm can't benefit from vsyscalls.
>
>
Yes, but the host will use tsc (on a modern host with a stable tsc).
--
error compiling committee.c: too many arguments to function
prev parent reply other threads:[~2008-06-05 7:56 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-29 22:22 [patch 00/12] fake ACPI C2 emulation v2 Marcelo Tosatti
2008-05-29 22:22 ` [patch 01/12] expose ACPI pmtimer to userspace (/dev/pmtimer) Marcelo Tosatti
2008-06-01 16:34 ` Thomas Gleixner
2008-06-01 16:56 ` Anthony Liguori
2008-06-04 9:53 ` Avi Kivity
2008-06-04 10:01 ` Thomas Gleixner
2008-06-04 10:35 ` Avi Kivity
2008-06-01 17:56 ` Marcelo Tosatti
2008-06-01 18:17 ` Thomas Gleixner
2008-06-02 16:43 ` John Stultz
2008-06-03 4:09 ` Marcelo Tosatti
2008-05-29 22:22 ` [patch 02/12] KVM: allow multiple IO bitmap pages, provide userspace interface Marcelo Tosatti
2008-05-29 22:22 ` [patch 03/12] KVM: allow userspace to open access to ACPI pmtimer Marcelo Tosatti
2008-05-29 22:22 ` [patch 04/12] KVM: move muldiv64 to x86.c, export Marcelo Tosatti
2008-05-29 22:22 ` [patch 05/12] KVM: in-kernel ACPI timer emulation Marcelo Tosatti
2008-05-29 22:22 ` [patch 06/12] QEMU/KVM: self-disabling C2 emulation Marcelo Tosatti
2008-05-29 22:22 ` [patch 07/12] libkvm: interface to KVM_SET_OPEN_IOPORT Marcelo Tosatti
2008-05-29 22:22 ` [patch 08/12] QEMU/KVM: non-virtualized ACPI PMTimer support Marcelo Tosatti
2008-05-29 22:22 ` [patch 09/12] libkvm: in-kernel ACPI pmtimer interface Marcelo Tosatti
2008-05-29 22:22 ` [patch 10/12] QEMU/KVM: add option to disable in-kernel pmtimer emulation Marcelo Tosatti
2008-05-29 22:23 ` [patch 11/12] libkvm: interface for pmtimer save/restore Marcelo Tosatti
2008-05-29 22:23 ` [patch 12/12] QEMU/KVM: in-kernel pmtimer save/restore support Marcelo Tosatti
2008-06-01 9:21 ` [patch 00/12] fake ACPI C2 emulation v2 Avi Kivity
2008-06-02 16:08 ` Marcelo Tosatti
2008-06-04 10:49 ` Avi Kivity
2008-06-05 3:12 ` Marcelo Tosatti
2008-06-05 7:56 ` Avi Kivity [this message]
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=48479C25.6080009@qumranet.com \
--to=avi@qumranet.com \
--cc=aliguori@us.ibm.com \
--cc=chrisw@redhat.com \
--cc=gcosta@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=mtosatti@redhat.com \
/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.