All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jakob Oestergaard <jakob@unthought.net>
To: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Phy Prabab <phyprabab@yahoo.com>,
	William Lee Irwin III <wli@holomorphy.com>,
	Andrew Morton <akpm@osdl.org>,
	linux-kernel@vger.kernel.org
Subject: Re: Help understanding slow down
Date: Mon, 24 May 2004 12:43:32 +0200	[thread overview]
Message-ID: <20040524104332.GH30687@unthought.net> (raw)
In-Reply-To: <40B1AB68.40209@yahoo.com.au>

On Mon, May 24, 2004 at 05:59:36PM +1000, Nick Piggin wrote:
> Phy Prabab wrote:
> >NO HT, disabled in bios and did not enable in kernel:
> >cat /proc/cpuinfo|grep processor|wc -l
> >  2
> >grep SMT .config (2.6.7-rc1)
> ># CONFIG_SCHED_SMT is not set
> >
> >On 2.4.21 I also include "append=noht"
> >
> 
> OK good, that makes things simpler.
> 
> I'm out of ideas though. The kernel just doesn't seem to be
> the problem here. Can you put together a testcase that causes
> the problem and that we can download and reproduce it?

You could try running your program with 'strace -T' to see which system
calls are causing the slowdown.

If you make a lot of system calls, this can be a little tedious to go
thru though...  But it should reveal to you where your program is
waiting.

Note; if you use gprof and compile with the '-pg' switch, the profiler
will show you a profile of CPU time, not wall-clock time.  Your
performance problem is wall-clock time, and if your program actually
spends a lot of time waiting in some system call, this will *NOT* show
up in a normal gprof profile.

 / jakob


  reply	other threads:[~2004-05-24 10:43 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-05-24  0:32 Help understanding slow down Phy Prabab
2004-05-24  0:57 ` Phy Prabab
2004-05-24  1:04   ` William Lee Irwin III
2004-05-24  1:25     ` Jakob Oestergaard
2004-05-24  1:28       ` William Lee Irwin III
2004-05-24  1:51         ` Phy Prabab
2004-05-24  2:42           ` Nick Piggin
2004-05-24  2:43           ` Andrew Morton
2004-05-24  6:23             ` Phy Prabab
2004-05-24  6:27               ` William Lee Irwin III
2004-05-24  6:39                 ` Phy Prabab
2004-05-24  7:21                   ` Phy Prabab
2004-05-24  7:44                     ` Nick Piggin
2004-05-24  7:50                       ` Phy Prabab
2004-05-24  7:59                         ` Nick Piggin
2004-05-24 10:43                           ` Jakob Oestergaard [this message]
2004-05-24 14:13                     ` Martin J. Bligh
2004-05-24  7:53                   ` Andrew Morton
2004-05-24  8:00                     ` Phy Prabab
2004-05-25 10:32                     ` Ingo Molnar
2004-05-25  9:29                       ` Andrew Morton
2004-05-25 11:42                         ` Ingo Molnar
2004-05-25  9:58                           ` Andrew Morton
2004-05-25 12:25                             ` Ingo Molnar
2004-05-24  6:33   ` Paul Rolland
2004-05-24 17:28     ` Roger Luethi
  -- strict thread matches above, loose matches on Subject: below --
2004-05-24  1:37 Robert M. Stockmann
2004-05-24  1:50 ` Phy Prabab
2004-05-24  3:31 ` Robert M. Stockmann
2004-05-24  6:14   ` Phy Prabab
2004-05-24 18:36     ` Robert M. Stockmann

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=20040524104332.GH30687@unthought.net \
    --to=jakob@unthought.net \
    --cc=akpm@osdl.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nickpiggin@yahoo.com.au \
    --cc=phyprabab@yahoo.com \
    --cc=wli@holomorphy.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.