From: Gleb Natapov <gleb@redhat.com>
To: David Ahern <dsahern@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
Avi Kivity <avi@redhat.com>, LKML <linux-kernel@vger.kernel.org>
Subject: Re: perf with precise attribute kills all KVM based VMs
Date: Thu, 12 Jul 2012 19:06:35 +0300 [thread overview]
Message-ID: <20120712160634.GI7298@redhat.com> (raw)
In-Reply-To: <4FFEEB55.5070009@gmail.com>
On Thu, Jul 12, 2012 at 09:20:53AM -0600, David Ahern wrote:
> On 7/11/12 10:29 PM, Gleb Natapov wrote:
> >On Wed, Jul 11, 2012 at 10:11:57PM -0600, David Ahern wrote:
> >>On 7/11/12 3:53 AM, Gleb Natapov wrote:
> >>>On Wed, Jul 11, 2012 at 11:49:47AM +0200, Peter Zijlstra wrote:
> >>>>On Wed, 2012-07-11 at 10:10 +0300, Gleb Natapov wrote:
> >>>>
> >>>>>Looks like Avi is right about the overshoot. Can you test something like this?
> >>>>>
> >>>>>diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
> >>>>>index 166546e..5fb371a 100644
> >>>>>--- a/arch/x86/kernel/cpu/perf_event_intel.c
> >>>>>+++ b/arch/x86/kernel/cpu/perf_event_intel.c
> >>>>>@@ -1374,8 +1374,11 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr)
> >>>>> arr[0].msr = MSR_CORE_PERF_GLOBAL_CTRL;
> >>>>> arr[0].host = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask;
> >>>>> arr[0].guest = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_host_mask;
> >>>>>+ arr[1].msr = MSR_IA32_PEBS_ENABLE;
> >>>>>+ arr[1].host = cpuc->pebs_enabled;
> >>>>>+ arr[1].guest = 0;
> >>>>>+ *nr = 2;
> >>>>>
> >>>>>- *nr = 1;
> >>>>> return arr;
> >>>>> }
> >>>>
> >>
> >>So far the 64-bit Fedora 10 VM with both a Fedora 10 stock kernel
> >>and a 2.6.38 kernel have not faired well - and that's the only VM I
> >>have tried at the moment. Using -e cycles:pp I have been able to
> >>lock up the VM 3 times out of 3 series of tests with perf-kvm that
> >>includes network traffic (e.g., netperf), disk I/O (dd based to
> >>create a file with dsync flag) and pure userspace cpu bound (openssl
> >>speed). May or may not be related.
> >>
> >OK that's may be BTSes. What about -e cycles:p? BTW are you using your
> >patch to set exclude_guest parameter? If not use -e cycles:Hp.
>
> I started with cycles:pp; should not really matter - they all need
> to work without blowing up VMs (cycles:p, cycles:pH, cycles:pG,
> cycles:pp, cycles:ppH, cycles:ppG).
cycles:ppG and cycles:pG should be illegal. Peter's patch takes care of
this. Others should set exclude guest and have to work without blowing
up VMs.
>
> For grins I ran a quick test while reading emails this morning. This
> time a fedora 17 VM with 3.4.0-1.fc17.x86_64 kernel. It too locks up
> pretty quickly - just a couple of iterations of perf:
>
> perf kvm --guestmount=/tmp/guest-mount record -fo /tmp/perf.data -a
> -v -e cycles:pH -- sleep 60
Do not run perf kvm. It does not set exclude_guest and :p and :pp is not
compatible with guest profiling and should be disallowed. Again Peter's
patch takes care of this.
Run perf top -e cycles:pH or similar.
>
> Note the :pH this time.
I am not sure what perf kvm does with :pH modifier, but H modifier does
not make sense with perf kvm and should be reported as an error by perf tool.
>
> I did not have netserver installed in the VM so used a ping flood
> for network traffic.
>
> >
> >>Also, I noted that 'perf kvm --guest record -e cycles:pp' does not
> >>generate a whole lot of samples -- like < 100 in a 20-second sample
> >>-- despite the fact that the guest is rather busy.
> >>
> >Host events do not suppose to generate events while guest is running.
>
> My server has 16 cpus and the VM has only 2 vcpus; with the -a I
> would expect some host sampling. Note: in the above case :pp resets
> the exclude-host modifier set by the perf-kvm part, so hosts samples
> are not excluded. See parse_events_modifier().
Isn't this a bug? Why anything at all resets exclude-host set by
perf-kvm?
>
> So, is the idea of your patch to not enable the PEBS in guest mode?
>
Yes, the idea of my patch is to disable PEBS and the guest entry.
--
Gleb.
next prev parent reply other threads:[~2012-07-12 16:06 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-09 14:12 perf with precise attribute kills all KVM based VMs David Ahern
2012-07-09 14:19 ` Gleb Natapov
2012-07-09 14:24 ` Peter Zijlstra
2012-07-09 14:36 ` Gleb Natapov
2012-07-09 14:39 ` Avi Kivity
2012-07-09 14:47 ` David Ahern
2012-07-09 14:49 ` Peter Zijlstra
2012-07-09 14:51 ` Avi Kivity
2012-07-09 14:54 ` Gleb Natapov
2012-07-09 14:57 ` Gleb Natapov
2012-07-09 14:59 ` Peter Zijlstra
2012-07-10 23:38 ` David Ahern
2012-07-11 7:10 ` Gleb Natapov
2012-07-11 9:49 ` Peter Zijlstra
2012-07-11 9:53 ` Gleb Natapov
2012-07-11 13:34 ` David Ahern
2012-07-12 4:11 ` David Ahern
2012-07-12 4:29 ` Gleb Natapov
2012-07-12 15:20 ` David Ahern
2012-07-12 16:06 ` Gleb Natapov [this message]
2012-07-12 16:13 ` Gleb Natapov
2012-07-12 16:58 ` Peter Zijlstra
2012-07-16 1:51 ` David Ahern
2012-07-15 8:07 ` Avi Kivity
2012-07-15 13:00 ` David Ahern
2012-07-15 13:03 ` Avi Kivity
2012-07-16 1:52 ` David Ahern
2012-07-16 2:19 ` David Ahern
2012-07-20 23:34 ` David Ahern
2012-07-22 9:52 ` Avi Kivity
2012-07-09 14:52 ` David Ahern
2012-07-09 14:58 ` David Ahern
2012-07-09 15:18 ` David Ahern
2012-07-09 14:47 ` Peter Zijlstra
2012-07-20 23:35 ` David Ahern
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=20120712160634.GI7298@redhat.com \
--to=gleb@redhat.com \
--cc=avi@redhat.com \
--cc=dsahern@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.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).