mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Palfrader <peter@palfrader.org>
To: linux-kernel@vger.kernel.org, openipmi-developer@lists.sourceforge.net
Subject: Re: [Openipmi-developer] BUG: soft lockup detected on CPU#1, ipmi_si
Date: Mon, 19 Jun 2006 11:38:51 +0200	[thread overview]
Message-ID: <20060619093851.GL27377@asteria.noreply.org> (raw)
In-Reply-To: <44962116.70302@acm.org>

On Sun, 18 Jun 2006, Corey Minyard wrote:

> The IPMI driver spawns a low-priority thread that will poll the driver
> when it finds there is something to do.  It's possible that the hardware
> is not setting things properly and is always telling the driver it has
> to do something.  It's possible that the new version of the firmware
> enabled interrupts; I think there's a problem with the driver here; it
> should not really enable the kernel thread if interrupts are working. 
> The driver should also probably call schedule() instead of udelay() in
> the kernel thread when a short timeout is requested by the state machine.
> 
> In either situation, the kernel thread will sit there and spin, and if
> nothing else is scheduled for 10 seconds on that CPU you will get that
> warning.  Can you check a few things for me?
> 
> cat /proc/ipmi/0/si_stats and send me the output.

After running for about 35 minutes (and one instance of the soft lockup
warning):

| interrupts_enabled:    0
| short_timeouts:        8835
| long_timeouts:         263709
| timeout_restarts:      0
| idles:                 793108
| interrupts:            0
| attentions:            0
| flag_fetches:          2137
| hosed_count:           0
| complete_transactions: 3516
| events:                0
| watchdog_pretimeouts:  0
| incoming_messages:     0


> If you do "top", is the kipmi0 always running?

Yes, running since the system started around 11:00:

| root      1331  0.8  0.0     0    0 ?        SN   10:59   0:17  \_ [kipmi0]

> Is your IPMI interface KCS or SMIC?  The IPMI driver should report this
> in the system log at startup.

It's KCS:

| laura:~# dmesg  | grep -i ipmi
| [   85.110244] ipmi message handler version 39.0
| [   85.111491] ipmi device interface
| [   85.127866] IPMI System Interface driver.
| [   85.127929] ipmi_si: Trying SMBIOS-specified KCS state machine at I/O address 0xca2, slave address 0x20, irq 0
| [   85.274699] ipmi: Found new BMC (man_id: 0x000f85,  prod_id: 0x0000, dev_id: 0x00)
| [   85.274852]  IPMI KCS interface initialized
| [   85.284710] IPMI Watchdog: driver initialized

Cheers,
Peter
-- 
                           |  .''`.  ** Debian GNU/Linux **
      Peter Palfrader      | : :' :      The  universal
 http://www.palfrader.org/ | `. `'      Operating System
                           |   `-    http://www.debian.org/

  reply	other threads:[~2006-06-19  9:38 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-06-13 23:35 Peter Palfrader
2006-06-19  3:59 ` [Openipmi-developer] " Corey Minyard
2006-06-19  9:38   ` Peter Palfrader [this message]
2006-06-22 14:03     ` Corey Minyard
2006-06-23  2:56       ` Matt Domsch
2006-06-23 14:55         ` Corey Minyard
2006-06-23 23:44         ` Corey Minyard
2006-06-24 19:37           ` Matt Domsch
2006-06-25  1:00       ` Peter Palfrader

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=20060619093851.GL27377@asteria.noreply.org \
    --to=peter@palfrader.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=openipmi-developer@lists.sourceforge.net \
    /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®