mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Roeland Th. Jansen" <roel@grobbebol.xs4all.nl>
To: linux-kernel@vger.kernel.org
Subject: IO APIC (smp) / crashes ?
Date: Tue, 30 Oct 2001 12:40:37 +0000	[thread overview]
Message-ID: <20011030124037.A26140@grobbebol.xs4all.nl> (raw)

using 2.4.13 I still experience crashes when a very specific set of
programs are used.

as I have a Abit BP6 and it's known to be not the best SMP board, I was
wondering if it could be possible that a high rate of interrupts
(specific IRQ's) could cause the 'crashes' I experience.

The setup is really weird. I use freeamp to play streams; I have an
ethernet card connected to a 512kbits adsl router; I use an SB16 under
opensound.com and use X with an AGP card.

if the stuff 'crashes, it in fact doesn't crash but starts to become
_very_ slow. if you look at /proc/interrupts you would see timer ticks
increase one by one but it woud take ages. like one tich in hald an hour
or so. this basically means -- system unuseable. it does respond to
pings though.

my wild guess is that the combination and rate of interrupts cause the
well known 

Oct 30 07:29:37 grobbebol kernel: APIC error on CPU1: 02(08)
Oct 30 07:29:37 grobbebol kernel: APIC error on CPU0: 02(04)
Oct 30 08:30:43 grobbebol kernel: APIC error on CPU0: 04(04)
Oct 30 08:30:43 grobbebol kernel: APIC error on CPU1: 08(08)
[....]

entries that finally after some time hit a combination that causes the
system to become very slow. would that be a possibility or am I just (as
usual :-) wrong ?


           CPU0       CPU1
  0:   13213246   13178957    IO-APIC-edge  timer
  1:      60279      59829    IO-APIC-edge  keyboard
  2:          0          0          XT-PIC  cascade
  3:      44383      44113    IO-APIC-edge  serial
  4:          3          5    IO-APIC-edge  serial
  5:      75105      73827    IO-APIC-edge  soundblaster
  8:          0          1    IO-APIC-edge  rtc
 14:     218635     212466    IO-APIC-edge  ide0
 15:       6903       6858    IO-APIC-edge  ide1
 18:      30878      30618   IO-APIC-level  BusLogic BT-930
 19:    5695922    5718641   IO-APIC-level  eth0
NMI:          0          0
LOC:   26392880   26392845
ERR:         82
MIS:         70


anyways, it basically only happens when I use X, when I use sound, when
I use xmms _and_ it comes from eth0. it happens not directly but
sometimes, after, say an hour, sometimes after 4 hours.

(fwiw -- it also happens under the opensound drivers in the kernel but
less frequent)

-- 
Grobbebol's Home                      |  Don't give in to spammers.   -o)
http://www.xs4all.nl/~bengel          | Use your real e-mail address   /\
Linux 2.4.13 (apic) SMP 466MHz/768 MB |        on Usenet.             _\_v  

             reply	other threads:[~2001-10-30 13:06 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-10-30 12:40 Roeland Th. Jansen [this message]
2001-10-31 14:49 ` Ken Brownfield
2001-11-02 15:56   ` Roeland Th. Jansen
2001-11-04 23:11   ` Roeland Th. Jansen

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=20011030124037.A26140@grobbebol.xs4all.nl \
    --to=roel@grobbebol.xs4all.nl \
    --cc=linux-kernel@vger.kernel.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

all inboxes | Powered by JetHome®