public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: David Ahern <daahern@cisco.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Avi Kivity <avi@redhat.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	kvm@vger.kernel.org, Ingo Molnar <mingo@elte.hu>,
	Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Subject: Re: [PATCH v2 00/11] KVM in-guest performance monitoring
Date: Thu, 16 Jun 2011 09:18:14 -0600	[thread overview]
Message-ID: <4DFA1EB6.6050708@cisco.com> (raw)
In-Reply-To: <1308236915.13240.66.camel@twins>



On 06/16/2011 09:08 AM, Peter Zijlstra wrote:
> On Thu, 2011-06-16 at 08:08 -0600, David Ahern wrote:
>> Command:
>>   perf stat -e instructions  openssl speed aes
>>
>> Guest:
>>    135,522,189,056 instructions              #    0.00  insns per cycle
>>
>>
>> Host:
>>    346,082,922,185 instructions              #    0.00  insns per cycle 
> 
> How does: perf stat -e instructions:u openssl speed aes, compare?

I think the problem is that perf stat in the guest introduces
significant overhead. I ran perf-record in the host on the VM pid while
running 'perf stat openssl speed aes' in the guest.

perf-report on that data shows:

    18.06%       9226  [k] arch_local_irq_enable
            |
            |--99.77%-- kvm_arch_vcpu_ioctl_run
            |          kvm_vcpu_ioctl
            |          do_vfs_ioctl
            |          sys_ioctl
            |          system_call_fastpath
            |          __GI___ioctl
            |          0x10100000002
             --0.23%-- [...]

and then perf-annotate on kvm_arch_vcpu_ioctl_run  shows
         :                vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
   21.47 :           1613a:       48 8b 3b                mov    (%rbx),%rdi

David

  reply	other threads:[~2011-06-16 15:18 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-13 13:34 [PATCH v2 00/11] KVM in-guest performance monitoring Avi Kivity
2011-06-13 13:34 ` [PATCH v2 01/11] perf: add context field to perf_event Avi Kivity
2011-06-13 13:34 ` [PATCH v2 02/11] x86, perf: add constraints for architectural PMU v1 Avi Kivity
2011-06-13 13:34 ` [PATCH v2 03/11] perf: export perf_event_refresh() to modules Avi Kivity
2011-06-13 13:34 ` [PATCH v2 04/11] KVM: Expose kvm_lapic_local_deliver() Avi Kivity
2011-06-13 13:35 ` [PATCH v2 05/11] KVM: Expose a version 1 architectural PMU to guests Avi Kivity
2011-06-13 18:10   ` David Ahern
2011-06-14  8:33     ` Avi Kivity
2011-06-27 14:10   ` lidong chen
2011-06-27 14:13     ` Avi Kivity
2011-06-28  2:54       ` lidong chen
2011-06-28  8:15         ` Avi Kivity
2011-06-13 13:35 ` [PATCH v2 06/11] KVM: Add generic RDPMC support Avi Kivity
2011-06-13 13:35 ` [PATCH v2 07/11] KVM: SVM: Intercept RDPMC Avi Kivity
2011-06-13 13:35 ` [PATCH v2 08/11] KVM: VMX: " Avi Kivity
2011-06-13 16:17   ` David Ahern
2011-06-13 16:29     ` Avi Kivity
2011-06-13 16:59       ` David Ahern
2011-06-13 17:38         ` Ingo Molnar
2011-06-13 18:09           ` David Ahern
2011-06-14  8:34           ` Avi Kivity
2011-06-14  9:27             ` Ingo Molnar
2011-06-13 13:35 ` [PATCH v2 09/11] KVM: Expose the architectural performance monitoring CPUID leaf Avi Kivity
2011-06-13 13:35 ` [PATCH v2 10/11] KVM: x86 emulator: fix RDPMC privilege check Avi Kivity
2011-06-13 13:35 ` [PATCH v2 11/11] KVM: x86 emulator: implement RDPMC (0F 33) Avi Kivity
2011-06-13 19:55 ` [PATCH v2 00/11] KVM in-guest performance monitoring David Ahern
2011-06-14  8:36   ` Avi Kivity
2011-06-14 17:15     ` David Ahern
2011-06-14 17:24       ` Jan Kiszka
2011-06-14 17:33         ` David Ahern
2011-06-14 17:48           ` David Ahern
2011-06-14 18:11             ` David Ahern
2011-06-15  8:57               ` Avi Kivity
2011-06-15 12:40                 ` David Ahern
2011-06-15 13:22                   ` Avi Kivity
2011-06-15 16:08                     ` David Ahern
2011-06-15 16:27                       ` Avi Kivity
2011-06-15 16:51                         ` David Ahern
2011-06-16 13:53                           ` Avi Kivity
2011-06-16 14:08                             ` David Ahern
2011-06-16 14:19                               ` David Ahern
2011-06-16 14:20                                 ` Avi Kivity
2011-06-16 14:32                                   ` David Ahern
2011-06-16 14:36                                     ` Avi Kivity
2011-06-16 15:04                                       ` David Ahern
2011-06-16 15:08                               ` Peter Zijlstra
2011-06-16 15:18                                 ` David Ahern [this message]
2011-06-16 15:19                                 ` David Ahern
2011-06-16 15:27                                   ` Peter Zijlstra
2011-06-16 15:34                                     ` David Ahern
2011-06-16 15:59                                       ` Avi Kivity
2011-06-16 16:04                                         ` David Ahern
2011-06-16 16:31                                           ` Avi Kivity
2011-06-16 17:16                                             ` David Ahern
2011-06-29  7:52 ` Avi Kivity
2011-06-29  8:38   ` Peter Zijlstra
2011-06-29  9:26     ` 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=4DFA1EB6.6050708@cisco.com \
    --to=daahern@cisco.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=acme@ghostprotocols.net \
    --cc=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=mingo@elte.hu \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox