Re: panic: APIC: Previous IPI is stuck

From: Kris Kennaway <kris_at_obsecurity.org>
Date: Sat, 17 Jul 2004 12:56:32 -0700
On Sat, Jul 17, 2004 at 02:42:40PM -0400, Brian Fundakowski Feldman wrote:
> On Sat, Jul 17, 2004 at 11:10:26AM -0700, Kris Kennaway wrote:
> > On Sat, Jul 17, 2004 at 06:10:45PM +0200, Lars Erik Gullerud wrote:
> > > On Sat, 17 Jul 2004, Andre Guibert de Bruet wrote:
> > > 
> > > > pmap_rltacziynftirx(:a 9sfp4u4nc 8f4o,r0 ,56040f0b0f0c0108
> > > > ,a9f44c80,8) at 0x606f31c5
> > > [snip]
> > > 
> > > > I am a little worried by the line that starts with pmap_rltac. It appears
> > > > to be the output of two interleaved messages, which I have yet to make
> > > > sense of. At this point the machine is deadlocked and nothing short of a
> > > > reset will get it to budge.
> > > 
> > > FYI, the two interleaved lines actually seem to be:
> > > 
> > > rtcintr(a9f44c84,0,64fbfc18,a9f44c80,8) at 0x606f31c5
> > > pmap_lazyfix: spun for 50000000
> > > 
> > > I seem to remember Kris Kennaway also having a couple of panics where
> > > pmap_lazyfix errors got interleaved with the output.
> > 
> > Yeah, I've had to disable SMP because this panic happens too often.
> 
> Does it only happen with a constant $NCPU+ load average?  I have never
> seen it here.

Yeah, it's usually a pretty busy machine.  It does a lot of
buildworlds, nfs, concurrent www serving and ssh dispatching.

Kris

Received on Sat Jul 17 2004 - 17:56:36 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:02 UTC