From: Ingo Molnar <mingo@elte.hu>
To: Jie Chen <chen@jlab.org>
Cc: Simon Holm Th??gersen <odie@cs.aau.dk>,
Eric Dumazet <dada1@cosmosbay.com>,
linux-kernel@vger.kernel.org,
Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: Possible bug from kernel 2.6.22 and above, 2.6.24-rc4
Date: Wed, 5 Dec 2007 21:03:43 +0100 [thread overview]
Message-ID: <20071205200343.GA14570@elte.hu> (raw)
In-Reply-To: <4756E44E.8080607@jlab.org>
* Jie Chen <chen@jlab.org> wrote:
> Since I am using affinity flag to bind each thread to a different
> core, the synchronization overhead should increases as the number of
> cores/threads increases. But what we observed in the new kernel is the
> opposite. The barrier overhead of two threads is 8.93 micro seconds vs
> 1.86 microseconds for 8 threads (the old kernel is 0.49 vs 1.86). This
> will confuse most of people who study the
> synchronization/communication scalability. I know my test code is not
> real-world computation which usually use up all cores. I hope I have
> explained myself clearly. Thank you very much.
btw., could you try to not use the affinity mask and let the scheduler
manage the spreading of tasks? It generally has a better knowledge about
how tasks interrelate.
Ingo
next prev parent reply other threads:[~2007-12-05 20:04 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-11-21 20:34 Possible bug from kernel 2.6.22 and above Jie Chen
2007-11-21 22:14 ` Eric Dumazet
2007-11-22 1:52 ` Jie Chen
2007-11-22 2:32 ` Simon Holm Thøgersen
2007-11-22 2:58 ` Jie Chen
2007-11-22 20:19 ` Matt Mackall
2007-12-04 13:17 ` Possible bug from kernel 2.6.22 and above, 2.6.24-rc4 Ingo Molnar
2007-12-04 15:41 ` Jie Chen
2007-12-05 15:29 ` Jie Chen
2007-12-05 15:40 ` Ingo Molnar
2007-12-05 16:16 ` Eric Dumazet
2007-12-05 16:25 ` Ingo Molnar
2007-12-05 16:29 ` Eric Dumazet
2007-12-05 16:22 ` Jie Chen
2007-12-05 16:47 ` Ingo Molnar
2007-12-05 17:47 ` Jie Chen
2007-12-05 20:03 ` Ingo Molnar [this message]
2007-12-05 20:23 ` Jie Chen
2007-12-05 20:46 ` Ingo Molnar
2007-12-05 20:52 ` Jie Chen
2007-12-05 21:02 ` Ingo Molnar
2007-12-05 22:16 ` Jie Chen
2007-12-06 10:43 ` Ingo Molnar
2007-12-06 16:29 ` Jie Chen
2007-12-10 10:59 ` Ingo Molnar
2007-12-10 20:04 ` Jie Chen
2007-12-11 10:51 ` Ingo Molnar
2007-12-11 15:28 ` Jie Chen
2007-12-11 15:52 ` Ingo Molnar
2007-12-11 16:39 ` Jie Chen
2007-12-11 21:23 ` Ingo Molnar
2007-12-11 22:11 ` Jie Chen
2007-12-12 12:49 ` Peter Zijlstra
2007-12-05 20:36 ` Possible bug from kernel 2.6.22 and above Peter Zijlstra
2007-12-05 20:53 ` Jie Chen
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=20071205200343.GA14570@elte.hu \
--to=mingo@elte.hu \
--cc=a.p.zijlstra@chello.nl \
--cc=chen@jlab.org \
--cc=dada1@cosmosbay.com \
--cc=linux-kernel@vger.kernel.org \
--cc=odie@cs.aau.dk \
/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.