From: Avi Kivity <avi@argo.co.il>
To: Dave Jones <davej@redhat.com>
Cc: Andrew Morton <akpm@osdl.org>, Andi Kleen <ak@suse.de>,
linux-kernel <linux-kernel@vger.kernel.org>,
Lee Revell <rlrevell@joe-job.com>,
cpufreq <cpufreq@www.linux.org.uk>
Subject: Re: [PATCH] CPU frequency display in /proc/cpuinfo
Date: Mon, 05 Dec 2005 18:29:02 +0200 [thread overview]
Message-ID: <43946ACE.9040405@argo.co.il> (raw)
In-Reply-To: <20051205011611.GA12664@redhat.com>
Dave Jones wrote:
>I can't think of a single valid reason why a program would want
>to know the MHz rating of a CPU. Given that it's a) approximate,
>b) subject to change due to power management, c) completely nonsensical
>across CPU vendors, and d) only one of many variables regarding CPU
>performance, any program that bases any decision on the values found
>by parsing that field of /proc/cpuinfo is utterly broken beyond belief.
>
>
Sometimes you need extremely low overhead time measurements, which need
not be too accurate. One way to do this is to dump rdtsc measurements
into some array, and later scale it using the cpu frequency.
I've done exactly this. The processes were pinned to their processors,
and there was no frequency scaling in effect. It worked very well.
WARNING: multiple messages have this Message-ID (diff)
From: Avi Kivity <avi@argo.co.il>
To: Dave Jones <davej@redhat.com>
Cc: Lee Revell <rlrevell@joe-job.com>, Andi Kleen <ak@suse.de>,
Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>,
Andrew Morton <akpm@osdl.org>, cpufreq <cpufreq@www.linux.org.uk>,
linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] CPU frequency display in /proc/cpuinfo
Date: Mon, 05 Dec 2005 18:29:02 +0200 [thread overview]
Message-ID: <43946ACE.9040405@argo.co.il> (raw)
In-Reply-To: <20051205011611.GA12664@redhat.com>
Dave Jones wrote:
>I can't think of a single valid reason why a program would want
>to know the MHz rating of a CPU. Given that it's a) approximate,
>b) subject to change due to power management, c) completely nonsensical
>across CPU vendors, and d) only one of many variables regarding CPU
>performance, any program that bases any decision on the values found
>by parsing that field of /proc/cpuinfo is utterly broken beyond belief.
>
>
Sometimes you need extremely low overhead time measurements, which need
not be too accurate. One way to do this is to dump rdtsc measurements
into some array, and later scale it using the cpu frequency.
I've done exactly this. The processes were pinned to their processors,
and there was no frequency scaling in effect. It worked very well.
next prev parent reply other threads:[~2005-12-05 16:29 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-02 18:13 [PATCH] CPU frequency display in /proc/cpuinfo Venkatesh Pallipadi
2005-12-02 18:19 ` Andi Kleen
2005-12-02 18:43 ` Venkatesh Pallipadi
2005-12-04 16:43 ` Dominik Brodowski
2005-12-04 18:32 ` Andi Kleen
2005-12-04 19:49 ` Lee Revell
2005-12-04 19:49 ` Lee Revell
2005-12-04 20:13 ` Andi Kleen
2005-12-04 20:13 ` Andi Kleen
2005-12-04 21:01 ` Horst von Brand
2005-12-05 1:16 ` Dave Jones
2005-12-05 1:16 ` Dave Jones
2005-12-05 13:02 ` Erik Mouw
2005-12-05 17:25 ` Dave Jones
2005-12-05 17:25 ` Dave Jones
2005-12-05 17:27 ` Lee Revell
2005-12-05 17:27 ` Lee Revell
2005-12-06 11:13 ` Erik Mouw
2005-12-06 16:56 ` Dave Jones
2005-12-06 16:56 ` Dave Jones
2005-12-06 17:35 ` Erik Mouw
2005-12-05 15:32 ` Lee Revell
2005-12-05 15:32 ` Lee Revell
2005-12-05 18:36 ` Andi Kleen
2005-12-05 15:59 ` Mark Lord
2005-12-05 17:26 ` Dave Jones
2005-12-05 17:26 ` Dave Jones
2005-12-05 16:29 ` Avi Kivity [this message]
2005-12-05 16:29 ` Avi Kivity
2005-12-05 16:46 ` linux-os (Dick Johnson)
2005-12-05 16:46 ` linux-os (Dick Johnson)
2005-12-05 17:27 ` Dave Jones
2005-12-05 17:27 ` Dave Jones
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=43946ACE.9040405@argo.co.il \
--to=avi@argo.co.il \
--cc=ak@suse.de \
--cc=akpm@osdl.org \
--cc=cpufreq@www.linux.org.uk \
--cc=davej@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=rlrevell@joe-job.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.