From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
To: "Li, Tong N" <tong.n.li@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>,
Andrew Morton <akpm@linux-foundation.org>,
Alexey Dobriyan <adobriyan@gmail.com>,
linux-kernel@vger.kernel.org
Subject: Re: [patch 10/10] Scheduler profiling - Use immediate values
Date: Fri, 6 Jul 2007 17:03:51 -0400 [thread overview]
Message-ID: <20070706210351.GA13491@Krystal> (raw)
In-Reply-To: <5FD5754DDBA0B1499B5A0B4BB5419485015CC9E1@fmsmsx411.amr.corp.intel.com>
* Li, Tong N (tong.n.li@intel.com) wrote:
> > I found that memory latency is difficult to measure in modern x86
> > CPUs because they have very clever prefetchers that can often
> > outwit benchmarks.
>
> A pointer-chasing program that accesses a random sequence of addresses
> usually can produce a good estimate on memory latency. Also, prefetching
> can be turned off in BIOS or by modifying the MSRs.
>
> > Another trap on P4 is that RDTSC is actually quite slow and
> synchronizes
> > the CPU; that can add large measurement errors.
> >
> > -Andi
>
> The cost can be amortized if the portion of memory accesses is long
> enough.
>
> tong
>
That's what I am currently doing.. the results are coming in a few
moments... :)
--
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
next prev parent reply other threads:[~2007-07-06 21:04 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-07-03 16:40 [patch 00/10] Immediate Values Mathieu Desnoyers
2007-07-03 16:40 ` [patch 01/10] Immediate values - Global modules list and module mutex Mathieu Desnoyers
2007-07-03 16:40 ` [patch 02/10] Immediate Value - Architecture Independent Code Mathieu Desnoyers
2007-07-03 16:40 ` [patch 03/10] Immediate Values - Non Optimized Architectures Mathieu Desnoyers
2007-07-03 16:40 ` [patch 04/10] Immediate Value - Add kconfig menus Mathieu Desnoyers
2007-07-03 16:40 ` [patch 05/10] Immediate Values - kprobe header fix Mathieu Desnoyers
2007-07-03 16:40 ` [patch 06/10] Immediate Value - i386 Optimization Mathieu Desnoyers
2007-07-03 18:45 ` H. Peter Anvin
2007-07-03 19:16 ` Mathieu Desnoyers
2007-07-03 20:18 ` H. Peter Anvin
2007-07-03 20:37 ` Chuck Ebbert
2007-07-03 21:30 ` H. Peter Anvin
2007-07-03 23:10 ` Jeremy Fitzhardinge
2007-07-03 20:43 ` Mathieu Desnoyers
2007-07-03 21:30 ` H. Peter Anvin
2007-07-03 16:40 ` [patch 07/10] Immediate Value - PowerPC Optimization Mathieu Desnoyers
2007-07-03 16:40 ` [patch 08/10] Immediate Value - Documentation Mathieu Desnoyers
2007-07-03 16:40 ` [patch 09/10] F00F bug fixup for i386 - use immediate values Mathieu Desnoyers
2007-07-04 20:43 ` Alexey Dobriyan
2007-07-03 16:40 ` [patch 10/10] Scheduler profiling - Use " Mathieu Desnoyers
2007-07-03 18:11 ` Alexey Dobriyan
2007-07-03 18:57 ` Mathieu Desnoyers
2007-07-04 14:23 ` Adrian Bunk
2007-07-04 20:31 ` Alexey Dobriyan
2007-07-05 20:21 ` Andrew Morton
2007-07-05 20:29 ` Andrew Morton
2007-07-05 20:41 ` Mathieu Desnoyers
2007-07-06 11:44 ` Andi Kleen
2007-07-06 17:50 ` Li, Tong N
2007-07-06 20:03 ` Andi Kleen
2007-07-06 20:57 ` Li, Tong N
2007-07-06 21:03 ` Mathieu Desnoyers [this message]
2007-07-07 1:50 ` [patch 10/10] *Tests* " Mathieu Desnoyers
2007-07-07 6:08 ` Li, Tong N
2007-07-11 5:02 ` Mathieu Desnoyers
2007-07-06 22:14 ` [patch 10/10] " Chuck Ebbert
2007-07-06 23:28 ` Adrian Bunk
2007-07-06 23:38 ` Dave Jones
2007-07-07 0:10 ` Adrian Bunk
2007-07-07 15:45 ` Frank Ch. Eigler
2007-07-07 17:01 ` Adrian Bunk
2007-07-07 17:20 ` Willy Tarreau
2007-07-07 17:59 ` Adrian Bunk
2007-07-07 17:55 ` Frank Ch. Eigler
2007-07-06 23:43 ` Mathieu Desnoyers
2007-07-07 2:25 ` Adrian Bunk
2007-07-07 2:35 ` Mathieu Desnoyers
2007-07-07 4:03 ` Adrian Bunk
2007-07-07 5:02 ` Willy Tarreau
2007-07-04 20:35 ` Alexey Dobriyan
2007-07-04 22:41 ` Andi Kleen
[not found] <8CTJM-4Z1-27@gated-at.bofh.it>
[not found] ` <8CU38-5E1-23@gated-at.bofh.it>
[not found] ` <8DjNU-4bL-47@gated-at.bofh.it>
[not found] ` <8DkTC-5Vy-11@gated-at.bofh.it>
2007-07-05 15:23 ` Bodo Eggert
2007-07-05 15:46 ` Mathieu Desnoyers
2007-07-06 21:08 ` Adrian Bunk
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=20070706210351.GA13491@Krystal \
--to=mathieu.desnoyers@polymtl.ca \
--cc=adobriyan@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tong.n.li@intel.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