From: Paul Mackerras <paulus@samba.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: Vince Weaver <vince@deater.net>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
linux-kernel@vger.kernel.org
Subject: Re: performance counter ~0.4% error finding retired instruction count
Date: Sat, 27 Jun 2009 16:48:32 +1000 [thread overview]
Message-ID: <19013.49344.860564.948905@cargo.ozlabs.ibm.com> (raw)
In-Reply-To: <20090627060432.GB16200@elte.hu>
Ingo Molnar writes:
> I measured 2000, but generally a few thousand cycles per invocation
> sounds about right.
We could actually do a bit better than we do, fairly easily. We could
attach the counters to the child after the fork instead of the parent
before the fork, using a couple of pipes for synchronization. And
there's probably a way to get the dynamic linker to resolve the execvp
call early in the child so we avoid that overhead. I think we should
be able to get the overhead down to tens of userspace instructions
without doing anything unnatural.
Paul.
next prev parent reply other threads:[~2009-06-27 6:48 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-06-24 13:59 performance counter 20% " Vince Weaver
2009-06-24 15:10 ` Ingo Molnar
2009-06-25 2:12 ` Vince Weaver
2009-06-25 6:50 ` Peter Zijlstra
2009-06-25 9:13 ` Ingo Molnar
2009-06-26 18:22 ` Vince Weaver
2009-06-26 19:12 ` Peter Zijlstra
2009-06-27 5:32 ` Ingo Molnar
2009-06-26 19:23 ` Vince Weaver
2009-06-27 6:04 ` performance counter ~0.4% " Ingo Molnar
2009-06-27 6:44 ` [numbers] perfmon/pfmon overhead of 17%-94% Ingo Molnar
2009-06-29 18:25 ` Vince Weaver
2009-06-29 21:02 ` Ingo Molnar
2009-07-02 21:07 ` Vince Weaver
2009-07-03 7:58 ` Ingo Molnar
2009-07-03 21:43 ` Vince Weaver
2009-07-03 18:31 ` Andi Kleen
2009-07-03 21:25 ` Vince Weaver
2009-07-03 23:40 ` Andi Kleen
2009-06-29 23:46 ` [patch] perf_counter: Add enable-on-exec attribute Ingo Molnar
2009-06-29 23:55 ` [numbers] perfmon/pfmon overhead of 17%-94% Ingo Molnar
2009-06-30 0:05 ` Ingo Molnar
2009-06-27 6:48 ` Paul Mackerras [this message]
2009-06-27 17:28 ` performance counter ~0.4% error finding retired instruction count Ingo Molnar
2009-06-29 2:12 ` Paul Mackerras
2009-06-29 2:13 ` Paul Mackerras
2009-06-29 3:48 ` Ingo Molnar
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=19013.49344.860564.948905@cargo.ozlabs.ibm.com \
--to=paulus@samba.org \
--cc=a.p.zijlstra@chello.nl \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=vince@deater.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®