From: Andi Kleen <ak@suse.de>
To: "Joachim Deguara" <joachim.deguara@amd.com>
Cc: "Mark Langsdorf" <mark.langsdorf@amd.com>,
discuss@x86-64.org, linux-kernel@vger.kernel.org,
cpufreq@lists.linux.org.uk
Subject: Re: [discuss] Re: [PATCH] Allow all Opteron processors to change pstate at same time
Date: Tue, 11 Jul 2006 15:07:39 +0200 [thread overview]
Message-ID: <200607111507.39079.ak@suse.de> (raw)
In-Reply-To: <1152622554.4489.32.camel@lapdog.site>
>
> Jul 11 21:23:35 gradient kernel: CPU 2: Syncing TSC to CPU 0.
> Jul 11 21:23:35 gradient kernel: CPU 2: synchronized TSC with CPU 0
> (last diff -91 cycles, maxerr 621 cycles)
> Jul 11 21:23:35 gradient kernel: CPU 3: Syncing TSC to CPU 0.
> Jul 11 21:23:35 gradient kernel: CPU 3: synchronized TSC with CPU 0
> (last diff -122 cycles, maxerr 1129 cycles)
This means the CPUs diverged between 500 and 1100 cycles in the night.
This can already cause severe timing problems with the clock going
backwards if a task switches CPUs - and there are many programs that
don't like that. If the system is up longer it will be worse.
The only way to possibly make the concept work would be regular TSC resyncs
during runtime, but I think I would prefer using per CPU TSC offsets
using RDTSCP instead because they should be able to tolerate arbitary
shifts.
-Andi
next prev parent reply other threads:[~2006-07-11 13:07 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-07-06 20:31 [PATCH] Allow all Opteron processors to change pstate at same time Mark Langsdorf
2006-07-07 12:10 ` Andi Kleen
2006-07-07 17:36 ` [discuss] " Langsdorf, Mark
2006-07-10 12:45 ` Joachim Deguara
2006-07-10 13:02 ` Joachim Deguara
2006-07-11 12:55 ` Joachim Deguara
2006-07-11 13:07 ` Andi Kleen [this message]
2006-07-11 13:14 ` Arjan van de Ven
2006-07-11 16:15 ` Alan Cox
2006-07-11 16:01 ` Arjan van de Ven
2006-07-11 16:04 ` Andi Kleen
2006-07-11 13:31 ` Langsdorf, Mark
2006-07-11 13:34 ` Arjan van de Ven
2006-07-11 13:51 ` Andi Kleen
2006-07-11 13:14 ` [OT] Evolution use (was: Re: [discuss] Re: [PATCH] Allow all Opteron processors to change pstate at same time) Xavier Bestel
2006-07-12 14:06 ` [discuss] Re: [PATCH] Allow all Opteron processors to change pstate at same time Joachim Deguara
2006-07-12 14:54 ` Andi Kleen
2006-07-25 21:47 ` Langsdorf, Mark
2006-07-26 11:31 ` Joachim Deguara
2006-07-26 16:42 ` Langsdorf, Mark
2006-07-26 16:54 ` Andi Kleen
2006-07-26 18:34 ` Langsdorf, Mark
2006-07-26 18:53 ` Andi Kleen
-- strict thread matches above, loose matches on Subject: below --
2006-07-12 16:11 shin, jacob
2006-07-12 16:14 ` Langsdorf, Mark
2006-07-13 13:06 ` Pavel Machek
2006-07-13 14:32 ` Joachim Deguara
2006-07-16 1:56 ` Pavel Machek
2006-07-17 7:37 ` Joachim Deguara
2006-07-20 15:59 ` Pavel Machek
2006-07-13 13:37 Bhavana Nagendra
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=200607111507.39079.ak@suse.de \
--to=ak@suse.de \
--cc=cpufreq@lists.linux.org.uk \
--cc=discuss@x86-64.org \
--cc=joachim.deguara@amd.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mark.langsdorf@amd.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