linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: "Jörg Sommer" <joerg@alea.gnuu.de>
To: linuxppc-dev@ozlabs.org
Subject: Re: time accounting problem (powerpc only?)
Date: Fri, 23 Nov 2007 14:08:10 +0000 (UTC)	[thread overview]
Message-ID: <slrnfkdnma.681.joerg@alea.gnuu.de> (raw)
In-Reply-To: 1195814816.4149.94.camel@johannes.berg

Hallo Johannes,

Johannes Berg <johannes@sipsolutions.net> wrote:
> Doing some scripting to read /proc/stat every half second and print the
> differences, I get output like this on a mostly idle system:
>
> # for reference:
> # [user, nice, system, idle, iowait, irq, softirq, steal, guest]
>
> [4, 0, 3, 46, 0, 264, 0, 0, 0]
>
> As you can see, the "irq" time is very high so that "user" (mostly my
> CPU burning program) only accounts for maybe 15%. This leads to
> powernowd thinking that all is fine and no switching is required. Also,
> the whole stuff leads to top(1) displaying about 80-90% "hi" (hard irq?
> calculated as irq - softirq?) time.

I've the same problem and I see these interrupts in /proc/interrups:

% diff /proc/interrupts <(sleep 2; cat /proc/interrupts)
--- /proc/interrupts    2007-11-23 15:04:06.004846901 +0100
+++ /proc/self/fd/11    2007-11-23 15:04:05.952841422 +0100
@@ -1,15 +1,15 @@
            CPU0       
  21:          5   MPIC 1    Edge      PMac Output
  24:         68   MPIC 1    Level     ide1
- 25:   18063968   MPIC 1    Level     VIA-PMU
+ 25:   18064241   MPIC 1    Level     VIA-PMU
  26:       2426   MPIC 1    Level     keywest i2c
  29:          1   MPIC 1    Level     ohci_hcd:usb2
  30:          1   MPIC 1    Edge      PMac Input
  39:     843910   MPIC 1    Level     ide0
  41:     990592   MPIC 1    Level     eth0
- 42:    1415066   MPIC 1    Level     keywest i2c
- 47:    2075159   MPIC 1    Level     GPIO1 ADB
- 48:    6686659   MPIC 1    Level     radeon@pci:0000:00:10.0
+ 42:    1415084   MPIC 1    Level     keywest i2c
+ 47:    2075193   MPIC 1    Level     GPIO1 ADB
+ 48:    6686778   MPIC 1    Level     radeon@pci:0000:00:10.0
  61:          0   MPIC 1    Edge      Sound Headphone Detection
  63:      94238   MPIC 1    Level     ehci_hcd:usb1, ohci_hcd:usb3, ohci_hcd:usb4
 BAD:         19

I don't know where they come from, but that's the cause of the high IRQ
time.

Bye, Jörg.
-- 
Je planmäßiger ein Mensch vorgeht,
desto stärker mag ihn der Zufall treffen.
		    Erich Krunau ‚Die Physiker‘

  reply	other threads:[~2007-11-23 15:12 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-23 10:46 time accounting problem (powerpc only?) Johannes Berg
2007-11-23 14:08 ` Jörg Sommer [this message]
2007-11-23 20:37   ` Johannes Berg
2007-11-23 20:48     ` Johannes Berg
2007-11-25 10:31     ` Jörg Sommer
2007-11-25 17:45       ` Johannes Berg
2007-11-26 16:23 ` Johannes Berg
2007-11-27  5:00   ` Tony Breeds
2007-11-27 13:42     ` Johannes Berg
2007-11-27  9:57   ` Paul Mackerras
2007-11-27 13:06     ` Johannes Berg
2007-11-26 23:55 ` Johannes Berg

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=slrnfkdnma.681.joerg@alea.gnuu.de \
    --to=joerg@alea.gnuu.de \
    --cc=linuxppc-dev@ozlabs.org \
    /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).