From: Avi Kivity <avi@redhat.com>
To: Alexander Graf <agraf-l3A5Bk7waGM@public.gmane.org>
Cc: "kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Subject: Re: [PATCH 00/18] KVM: PPC: Virtualize Gekko guests
Date: Wed, 17 Feb 2010 16:03:24 +0000 [thread overview]
Message-ID: <4B7C134C.9040009@redhat.com> (raw)
In-Reply-To: <80F0B53A-6F83-4166-8F85-5D9B07526158-l3A5Bk7waGM@public.gmane.org>
On 02/17/2010 04:56 PM, Alexander Graf wrote:
>
> So I changed to code according to your input by making all FPU calls explicit, getting rid of all binary patching.
>
> On the PowerStation again I'm running this code (simplified to the important instructions) using kvmctl:
>
> li r2, 0x1234
> std r2, 0(r1)
> lfd f3, 0(r1)
> lfd f4, 0(r1)
> do_mul:
> fmul f0, f3, f4
> b do_mul
>
>
> With the following kvm_stat output:
>
> dec 2236 53
> exits 60797802 1171403
> ext_intr 379 4
> halt_wakeup 0 0
> inst_emu 60795247 1171344
> ld 60795132 1171348
>
> So I'm getting 1171403 fmul operations per second. And that's even with non-optimized instruction fetching. Not bad.
>
It's a large number, but won't real hardware be three orders of
magnitude faster?
--
error compiling committee.c: too many arguments to function
WARNING: multiple messages have this Message-ID (diff)
From: Avi Kivity <avi-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Alexander Graf <agraf-l3A5Bk7waGM@public.gmane.org>
Cc: "kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Subject: Re: [PATCH 00/18] KVM: PPC: Virtualize Gekko guests
Date: Wed, 17 Feb 2010 18:03:24 +0200 [thread overview]
Message-ID: <4B7C134C.9040009@redhat.com> (raw)
In-Reply-To: <80F0B53A-6F83-4166-8F85-5D9B07526158-l3A5Bk7waGM@public.gmane.org>
On 02/17/2010 04:56 PM, Alexander Graf wrote:
>
> So I changed to code according to your input by making all FPU calls explicit, getting rid of all binary patching.
>
> On the PowerStation again I'm running this code (simplified to the important instructions) using kvmctl:
>
> li r2, 0x1234
> std r2, 0(r1)
> lfd f3, 0(r1)
> lfd f4, 0(r1)
> do_mul:
> fmul f0, f3, f4
> b do_mul
>
>
> With the following kvm_stat output:
>
> dec 2236 53
> exits 60797802 1171403
> ext_intr 379 4
> halt_wakeup 0 0
> inst_emu 60795247 1171344
> ld 60795132 1171348
>
> So I'm getting 1171403 fmul operations per second. And that's even with non-optimized instruction fetching. Not bad.
>
It's a large number, but won't real hardware be three orders of
magnitude faster?
--
error compiling committee.c: too many arguments to function
next prev parent reply other threads:[~2010-02-17 16:03 UTC|newest]
Thread overview: 106+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-04 15:55 [PATCH 00/18] KVM: PPC: Virtualize Gekko guests Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 01/18] KVM: PPC: Add QPR registers Alexander Graf
2010-02-04 15:55 ` Alexander Graf
[not found] ` <1265298925-31954-1-git-send-email-agraf-l3A5Bk7waGM@public.gmane.org>
2010-02-04 15:55 ` [PATCH 02/18] KVM: PPC: Enable MMIO to do 64 bits, fprs and qprs Alexander Graf
2010-02-04 15:55 ` Alexander Graf
[not found] ` <1265298925-31954-3-git-send-email-agraf-l3A5Bk7waGM@public.gmane.org>
2010-02-07 12:29 ` Avi Kivity
2010-02-07 12:29 ` Avi Kivity
[not found] ` <4B6EB229.8090502-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-07 15:51 ` Alexander Graf
2010-02-07 15:51 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 03/18] KVM: PPC: Teach MMIO Signedness Alexander Graf
2010-02-04 15:55 ` Alexander Graf
[not found] ` <1265298925-31954-4-git-send-email-agraf-l3A5Bk7waGM@public.gmane.org>
2010-02-07 12:32 ` Avi Kivity
2010-02-07 12:32 ` Avi Kivity
[not found] ` <4B6EB2D7.1030500-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-07 15:51 ` Alexander Graf
2010-02-07 15:51 ` Alexander Graf
[not found] ` <3CEF000F-1751-4E65-A08A-C71B2CE8DAEE-l3A5Bk7waGM@public.gmane.org>
2010-02-07 16:15 ` Avi Kivity
2010-02-07 16:15 ` Avi Kivity
2010-02-07 16:27 ` Anthony Liguori
2010-02-07 16:27 ` Anthony Liguori
2010-02-07 21:35 ` Alexander Graf
2010-02-07 21:35 ` Alexander Graf
[not found] ` <1CA08386-21CA-4B4F-A1E6-56C4DE584BA6-l3A5Bk7waGM@public.gmane.org>
2010-02-07 22:13 ` Anthony Liguori
2010-02-07 22:13 ` Anthony Liguori
2010-02-04 15:55 ` [PATCH 04/18] KVM: PPC: Add AGAIN type for emulation return Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 05/18] KVM: PPC: Add hidden flag for paired singles Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 06/18] KVM: PPC: Add Gekko SPRs Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 12/18] KVM: PPC: Make ext giveup non-static Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 14/18] KVM: PPC: Fix error in BAT assignment Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 15/18] KVM: PPC: Add helpers to modify ppc fields Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 16/18] KVM: PPC: Enable program interrupt to do MMIO Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-07 12:54 ` [PATCH 00/18] KVM: PPC: Virtualize Gekko guests Avi Kivity
2010-02-07 12:54 ` Avi Kivity
[not found] ` <4B6EB7F6.10304-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-07 15:49 ` Alexander Graf
2010-02-07 15:49 ` Alexander Graf
2010-02-07 16:22 ` Avi Kivity
2010-02-07 16:22 ` Avi Kivity
2010-02-07 22:02 ` Alexander Graf
2010-02-07 22:02 ` Alexander Graf
2010-02-08 8:53 ` Avi Kivity
2010-02-08 8:53 ` Avi Kivity
[not found] ` <4B6FD118.2090207-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-08 10:58 ` Alexander Graf
2010-02-08 10:58 ` Alexander Graf
[not found] ` <87CEECB5-107A-46EB-89F5-1E1F92AC22AA-l3A5Bk7waGM@public.gmane.org>
2010-02-08 11:09 ` Avi Kivity
2010-02-08 11:09 ` Avi Kivity
[not found] ` <4B6FF0E6.6060309-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-08 11:30 ` Alexander Graf
2010-02-08 11:30 ` Alexander Graf
[not found] ` <A5BC5A7E-D45B-4BAF-804A-B364810F50DA-l3A5Bk7waGM@public.gmane.org>
2010-02-08 12:03 ` Avi Kivity
2010-02-08 12:03 ` Avi Kivity
[not found] ` <4B6FFD85.6090100-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-08 12:05 ` Alexander Graf
2010-02-08 12:05 ` Alexander Graf
[not found] ` <939C8633-1B2C-4888-B1C1-357DF1C56CE6-l3A5Bk7waGM@public.gmane.org>
2010-02-08 12:15 ` Avi Kivity
2010-02-08 12:15 ` Avi Kivity
2010-02-08 12:31 ` Alexander Graf
2010-02-08 12:31 ` Alexander Graf
2010-02-09 11:00 ` Alexander Graf
2010-02-09 11:00 ` Alexander Graf
[not found] ` <4B714049.7010201-l3A5Bk7waGM@public.gmane.org>
2010-02-09 11:06 ` Avi Kivity
2010-02-09 11:06 ` Avi Kivity
2010-02-09 11:13 ` Alexander Graf
2010-02-09 11:13 ` Alexander Graf
[not found] ` <4B71435E.7010103-l3A5Bk7waGM@public.gmane.org>
2010-02-09 12:27 ` Avi Kivity
2010-02-09 12:27 ` Avi Kivity
[not found] ` <4B7154A6.6050809-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-17 14:56 ` Alexander Graf
2010-02-17 14:56 ` Alexander Graf
[not found] ` <80F0B53A-6F83-4166-8F85-5D9B07526158-l3A5Bk7waGM@public.gmane.org>
2010-02-17 16:03 ` Avi Kivity [this message]
2010-02-17 16:03 ` Avi Kivity
[not found] ` <4B7C134C.9040009-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-17 16:23 ` Alexander Graf
2010-02-17 16:23 ` Alexander Graf
2010-02-17 16:34 ` Avi Kivity
2010-02-17 16:34 ` Avi Kivity
[not found] ` <4B7C1A91.8060205-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-17 18:07 ` Alexander Graf
2010-02-17 18:07 ` Alexander Graf
[not found] ` <D3C1F39A-E9C8-46B8-9C4A-12E56D34B5AC-l3A5Bk7waGM@public.gmane.org>
2010-02-18 7:40 ` Avi Kivity
2010-02-18 7:40 ` Avi Kivity
[not found] ` <4B7CEEFF.9000801-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2010-02-18 8:04 ` Avi Kivity
2010-02-18 8:04 ` Avi Kivity
2010-02-04 15:55 ` [PATCH 07/18] KVM: PPC: Combine extension interrupt handlers Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 08/18] KVM: PPC: Preload FPU when possible Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 09/18] KVM: PPC: Fix typo in book3s_32 debug code Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 10/18] KVM: PPC: Implement mtsr instruction emulation Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 11/18] KVM: PPC: Make software load/store return eaddr Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 13/18] KVM: PPC: Add helpers to call FPU instructions Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 17/18] KVM: PPC: Reserve a chunk of memory for opcodes Alexander Graf
2010-02-04 15:55 ` Alexander Graf
2010-02-04 15:55 ` [PATCH 18/18] KVM: PPC: Implement Paired Single emulation Alexander Graf
2010-02-04 15:55 ` Alexander Graf
[not found] ` <1265298925-31954-19-git-send-email-agraf-l3A5Bk7waGM@public.gmane.org>
2010-02-07 12:50 ` Avi Kivity
2010-02-07 12:50 ` Avi Kivity
2010-02-07 15:57 ` Alexander Graf
2010-02-07 15:57 ` Alexander Graf
2010-02-07 16:18 ` Avi Kivity
2010-02-07 16:18 ` 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=4B7C134C.9040009@redhat.com \
--to=avi@redhat.com \
--cc=agraf-l3A5Bk7waGM@public.gmane.org \
--cc=kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.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.