From: Matt Mackall <mpm@selenic.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: Rusty Russell <rusty@rustcorp.com.au>,
akpm@linux-foundation.org, Linux-kernel@vger.kernel.org
Subject: Re: Interesting interaction between lguest and CFS
Date: Tue, 5 Jun 2007 09:03:42 -0500 [thread overview]
Message-ID: <20070605140342.GR11115@waste.org> (raw)
In-Reply-To: <20070605071904.GB25163@elte.hu>
On Tue, Jun 05, 2007 at 09:19:04AM +0200, Ingo Molnar wrote:
>
> * Matt Mackall <mpm@selenic.com> wrote:
>
> > sleep_max : 57476665627
> > block_max : 18014060106626075
>
> hm, this block_max looks a bit suspect, it's 003fffb1359e341b. Does your
> box make any use of cpufreq? (what CPU is it?)
Yes, I use the conservative governor. As I posted yesterday, setting
the governor to performance doesn't help: the CPU hog drives the CPU
to full speed.
> > 24936 13:39:44.469677 rt_sigprocmask(SIG_UNBLOCK, [USR1], NULL, 8) = 0
> > 24936 13:39:44.469765 read(7, 0xbfe6c010, 8) = -1 EINTR (Interrupted system call)
> > 24936 13:39:48.699490 --- SIGUSR1 (User defined signal 1) @ 0 (0) ---
>
> hm, this indeed implicates some wakeup problem. lguest task 24936's
> relevant stats are:
>
> block_start : 0
> sleep_max : 21492554
> block_max : 27044576
> exec_max : 4008057
> wait_max : 1253670288
>
> the wait_max means it was delayed on the runqueue for 1.2 seconds. Could
> you try to get a /proc/sched_debug snapshot done exactly during the
> 'delay' window - it's 4 seconds so you should in theory be able to
> trigger it by doing something like:
>
> sleep 3; cat /proc/sched_debug > sched_debug.txt
>
> Click into the lguest window and trigger the delay.
I did:
while true; do sleep 1; cat /proc/sched_debug > sched_debug.txt; done
and got this, hopefully inside the window:
Sched Debug Version: v0.02
now at 257428593818894 nsecs
cpu: 0
.nr_running : 3
.raw_weighted_load : 2063
.nr_switches : 242830075
.nr_load_updates : 30172063
.nr_uninterruptible : 0
.jiffies : 64282148
.next_balance : 0
.curr->pid : 27182
.clock : 125650217819008823
.prev_clock_raw : 257428516403535
.clock_warps : 9
.clock_unstable_events : 41133344
.clock_max_delta : 3954619
.fair_clock : 125631632368075562
.prev_fair_clock : 0
.exec_clock : 125649753800946713
.prev_exec_clock : 0
.wait_runtime : -12865352643
.wait_runtime_overruns : 41121996
.wait_runtime_underruns: 4107250
.cpu_load[0] : 1039
.cpu_load[1] : 1039
.cpu_load[2] : 1039
.cpu_load[3] : 1041
.cpu_load[4] : 1053
.wait_runtime_rq_sum : -48091562
runnable tasks:
task PID tree-key delta waiting switches prio sum-exec sum-wait sum-sleep wait-overrun wait-underrun
------------------------------------------------------------------------------------------------------------------------------------------------------------------
R cat 27182 125631632367998747 -76815 -91562 6 120 250189 -91562 75163 0 0
python 26973 125631632404141566 36066004 -24000000 25486 120 93635486661 -1940103390 140903 49 2069
lguest 26986 125631633015454796 647379234 -24000000 1481 139 7638721 37069331905 90071079196 1461 2
--
Mathematics is the supreme nostalgia of our time.
next prev parent reply other threads:[~2007-06-05 14:04 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-06-04 17:37 Interesting interaction between lguest and CFS Matt Mackall
2007-06-04 17:54 ` Ingo Molnar
2007-06-04 18:41 ` Matt Mackall
2007-06-05 7:19 ` Ingo Molnar
2007-06-05 14:03 ` Matt Mackall [this message]
2007-06-05 15:14 ` Dmitry Adamushko
2007-06-05 15:41 ` Dmitry Adamushko
2007-06-05 19:50 ` Ingo Molnar
2007-06-06 20:23 ` Matt Mackall
2007-06-08 9:34 ` Ingo Molnar
2007-06-08 10:18 ` Ingo Molnar
2007-06-04 18:11 ` Ingo Molnar
2007-06-04 18:42 ` Matt Mackall
2007-06-04 19:12 ` Ingo Molnar
2007-06-04 19:27 ` Matt Mackall
2007-06-04 19:40 ` Ingo Molnar
2007-06-04 19:59 ` Matt Mackall
2007-06-04 20:13 ` Ingo Molnar
2007-06-04 20:20 ` Ingo Molnar
2007-06-04 20:15 ` James Morris
2007-06-04 20:24 ` Ingo Molnar
2007-06-04 22:58 ` Olof Johansson
2007-06-05 1:02 ` Rusty Russell
2007-06-05 1:24 ` Stephen Rothwell
2007-06-05 2:23 ` Olof Johansson
2007-06-05 0:18 ` Rusty Russell
2007-06-05 2:31 ` Rusty Russell
2007-06-05 4:18 ` Matt Mackall
2007-06-05 4:37 ` Rusty Russell
2007-06-05 13:49 ` Matt Mackall
2007-06-05 19:38 ` Ingo Molnar
2007-06-05 10:43 ` Ingo Molnar
2007-06-05 19:08 ` Ingo Molnar
2007-06-06 20:19 ` Matt Mackall
2007-06-05 2:34 ` Matt Mackall
2007-06-05 7:10 ` Ingo Molnar
2007-06-05 9:23 ` Ingo Molnar
2007-06-05 14:06 ` Matt Mackall
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=20070605140342.GR11115@waste.org \
--to=mpm@selenic.com \
--cc=Linux-kernel@vger.kernel.org \
--cc=akpm@linux-foundation.org \
--cc=mingo@elte.hu \
--cc=rusty@rustcorp.com.au \
/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