From: stephane eranian <eranian@googlemail.com>
To: David Miller <davem@davemloft.net>
Cc: paulus@samba.org, a.p.zijlstra@chello.nl,
linux-kernel@vger.kernel.org, mingo@elte.hu,
perfmon2-devel@lists.sf.net
Subject: Re: [PATCH 2/2] perf_events: add event constraints support for Intel processors
Date: Wed, 7 Oct 2009 23:30:18 +0200 [thread overview]
Message-ID: <7c86c4470910071430q6c6b9bf0k36458f46d1231420@mail.gmail.com> (raw)
In-Reply-To: <20091007.134626.238756485.davem@davemloft.net>
On Wed, Oct 7, 2009 at 10:46 PM, David Miller <davem@davemloft.net> wrote:
> From: stephane eranian <eranian@googlemail.com>
> Date: Wed, 7 Oct 2009 14:31:58 +0200
>
>> What PPC does is probably the only way to do this given the interface between
>> generic and machine-specific code. The one advantage I see is that it works
>> inside an event group but also across event groups because that code does not
>> look at group boundary, it only looks at the events and the number of available
>> registers. The downside is that you duplicate state.
>>
>> Did I get this right, Paul?
>
> That's basically how his code works, yes. I intend on duplicating
> it to some extent on sparc64 since I'm operating in a similar
> problem space.
>
> So if at least some of this engine went to a generic place, there'd
> be at least a 3rd user :-)
Based on my experience, the assignment problem is best handled
in the architecture or model specific code. On some PMU models,
it is way more complicated than just two events competing for the
same counter. That's the case on Itanium, for instance. And that
is also the case with AMD64 Northbridge events.
Something I forgot to mention earlier is that when you re-run the
assignment code for the new event, no already assigned event
can be kicked out in favor of the new event. An existing event
can be moved to another counter at worst. Otherwise, you will
evict an event which, the generic layer thinks, is currently running.
>
next prev parent reply other threads:[~2009-10-07 21:30 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-06 14:42 [PATCH 0/2] perf_events: correct event assignments on " Stephane Eranian
2009-10-06 14:42 ` [PATCH 1/2] perf_events: check for filters on fixed counter events Stephane Eranian
2009-10-06 14:42 ` [PATCH 2/2] perf_events: add event constraints support for Intel processors Stephane Eranian
2009-10-06 16:29 ` Peter Zijlstra
2009-10-06 17:26 ` stephane eranian
2009-10-06 18:57 ` [perfmon2] " Vince Weaver
2009-10-07 10:31 ` Peter Zijlstra
2009-10-07 11:15 ` Paul Mackerras
2009-10-07 12:31 ` stephane eranian
2009-10-07 20:46 ` David Miller
2009-10-07 21:30 ` stephane eranian [this message]
2009-10-08 20:08 ` Ingo Molnar
2009-10-08 20:28 ` stephane eranian
2009-10-12 9:05 ` Ingo Molnar
2009-10-13 7:17 ` stephane eranian
2009-10-13 7:29 ` Ingo Molnar
2009-10-08 23:18 ` Paul Mackerras
2009-10-09 14:22 ` [tip:perf/core] perf, x86: Add simple group validation tip-bot for Peter Zijlstra
2009-10-09 13:55 ` [PATCH 2/2] perf_events: add event constraints support for Intel processors Ingo Molnar
2009-10-09 14:22 ` [tip:perf/core] perf_events: Add " tip-bot for Stephane Eranian
2009-10-09 14:22 ` [tip:perf/core] perf_events: Check for filters on fixed counter events tip-bot for Stephane Eranian
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=7c86c4470910071430q6c6b9bf0k36458f46d1231420@mail.gmail.com \
--to=eranian@googlemail.com \
--cc=a.p.zijlstra@chello.nl \
--cc=davem@davemloft.net \
--cc=eranian@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.org \
--cc=perfmon2-devel@lists.sf.net \
/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
all inboxes | Powered by JetHome®