From: Andrew Morton <akpm@zip.com.au>
To: Rik van Riel <riel@conectiva.com.br>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [RFC][PATCH] iowait statistics
Date: Mon, 13 May 2002 19:18:34 -0700 [thread overview]
Message-ID: <3CE073FA.57DAC578@zip.com.au> (raw)
In-Reply-To: <Pine.LNX.4.44L.0205132214480.32261-100000@imladris.surriel.com>
Rik van Riel wrote:
>
> Hi,
>
> the following patch implements iowait statistics in a simple way:
>
> 1) if we go to sleep while waiting on a page or buffer, we
> increment nr_iowait_tasks, note that this is done only in
> the slow path so overhead shouldn't even be measurable
>
> 2) if no process is running, the timer interrupt adds a jiffy
> to the iowait time
>
> 3) iowait time is counted separately from user/system/idle and
> can overlap with either system or idle (when no process is
> running the system can still be busy processing interrupts)
>
> 4) on SMP systems the iowait time can be overestimated, no big
> deal IMHO but cheap suggestions for improvement are welcome
I suspect that a number of these statistical accounting mechanisms
are going to break. The new irq-affinity code works awfully well.
The kernel profiler in 2.5 doesn't work very well at present.
When investigating this, I ran a busy-wait process. It attached
itself to CPU #3 and that CPU received precisely zero interrupts
across a five minute period. So the profiler cunningly avoids profiling
busy CPUs, which is rather counter-productive. Fortunate that oprofile
uses NMI.
> ...
> ===== fs/buffer.c 1.64 vs edited =====
> --- 1.64/fs/buffer.c Mon May 13 19:04:59 2002
> +++ edited/fs/buffer.c Mon May 13 19:16:57 2002
> @@ -156,8 +156,10 @@
> get_bh(bh);
> add_wait_queue(&bh->b_wait, &wait);
> do {
> + atomic_inc(&nr_iowait_tasks);
> run_task_queue(&tq_disk);
> set_task_state(tsk, TASK_UNINTERRUPTIBLE);
> + atomic_dec(&nr_iowait_tasks);
> if (!buffer_locked(bh))
> break;
> schedule();
Shouldn't the atomic_inc cover the schedule()?
-
next prev parent reply other threads:[~2002-05-14 2:14 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-05-14 1:19 [RFC][PATCH] iowait statistics Rik van Riel
2002-05-14 2:18 ` Andrew Morton [this message]
2002-05-14 12:30 ` Rik van Riel
2002-05-15 17:02 ` Denis Vlasenko
2002-05-16 7:41 ` Andrew Morton
2002-05-16 14:04 ` Denis Vlasenko
2002-05-14 15:39 ` William Lee Irwin III
2002-05-14 16:36 ` Rik van Riel
2002-05-14 16:54 ` William Lee Irwin III
2002-05-15 17:17 ` Denis Vlasenko
2002-05-15 14:03 ` Rik van Riel
2002-05-15 20:17 ` Denis Vlasenko
2002-05-15 16:13 ` Rik van Riel
2002-05-15 16:21 ` William Lee Irwin III
2002-05-15 17:00 ` William Lee Irwin III
2002-05-15 18:16 ` Bill Davidsen
2002-05-15 18:30 ` William Lee Irwin III
2002-05-15 18:33 ` Rik van Riel
2002-05-15 18:46 ` William Lee Irwin III
2002-05-15 19:00 ` Rik van Riel
2002-05-16 11:42 ` Denis Vlasenko
2002-05-16 11:14 ` Denis Vlasenko
2002-05-15 15:15 ` Bill Davidsen
2002-05-16 10:58 ` Denis Vlasenko
2002-05-14 18:19 ` Martin J. Bligh
2002-05-15 1:31 ` Bill Davidsen
2002-05-15 1:41 ` William Lee Irwin III
2002-05-15 14:39 ` Bill Davidsen
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=3CE073FA.57DAC578@zip.com.au \
--to=akpm@zip.com.au \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=riel@conectiva.com.br \
/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;
as well as URLs for NNTP newsgroup(s).