Weird K8 output

[Stefan Reinauer]

Hi,

on a 4cpu K8 system I get the following output before the system hangs..

Is the machine drunk or are CPU1-3 just running away?

LinuxBIOS-1.1.7.0-fallback Tue Nov  2 20:34:14 CET 2004 starting...
setting up resource map....done.
Enabling routing table for node 00 done.
Enabling SMP settings
setup_remote_node: done
Renaming current temporary node to 01 done.
anabling routing table for node 01 dConlee.
 tRiennamgin g mcutrrrenrt
  emporary node to 02 done.
Enabling routing table for node 02 dConle.e
mRernaiminngg c urmretntr tre
 porary node to 03 done.
Enabling routing table for node 03 dColnee.a
iz04n ngod esm itnirtiral
  ed.
coherent_ht_finalize
done
PCI_DEV=(0,0x18,0) 0xb4=00030000
bus=00 id =74501022
bus=00 id =74601022
PCI_DEV=(0,0x19,0) 0xd4=000a0400

Stefan

Stefan Reinauer  writes:
> Hi,
>
> on a 4cpu K8 system I get the following output before the system hangs..

Hmm.  It looks like you have non-bootstrap cpus running through the code.

Usually it is this hunk of code that traps errors.
    if (!boot_cpu()) {
    	stop_this_cpu();

Although I can see something very weird happening if you have
BIST failure, on your secondary cpus. But that is quite 
unlikely.
> Is the machine drunk or are CPU1-3 just running away?

It looks like the cpus are just running away :(  Usually I would
expect to interlaced characters form a message I recognize.  But I
don't recognize see anything I recognize here.

Eric
> LinuxBIOS-1.1.7.0-fallback Tue Nov  2 20:34:14 CET 2004 starting...
> setting up resource map....done.
> Enabling routing table for node 00 done.
> Enabling SMP settings
> setup_remote_node: done
> Renaming current temporary node to 01 done.
> anabling routing table for node 01 dConlee.
>  tRiennamgin g mcutrrrenrt
>   emporary node to 02 done.
> Enabling routing table for node 02 dConle.e
> mRernaiminngg c urmretntr tre
>  porary node to 03 done.
> Enabling routing table for node 03 dColnee.a
> iz04n ngod esm itnirtiral
>   ed.
> coherent_ht_finalize
> done
> PCI_DEV=(0,0x18,0) 0xb4=00030000
> bus=00 id =74501022
> bus=00 id =74601022
> PCI_DEV=(0,0x19,0) 0xd4=000a0400

[Stefan Reinauer]

* Eric W. Biederman  [041102 22:05]:
> Hmm.  It looks like you have non-bootstrap cpus running through the code.
>
> Usually it is this hunk of code that traps errors.
>    if (!boot_cpu()) {
>    	stop_this_cpu();
> Although I can see something very weird happening if you have
> BIST failure, on your secondary cpus. But that is quite 
> unlikely.

But that is something that would be reported, somehow, before setting
the resource map, no?
> > Is the machine drunk or are CPU1-3 just running away?
>
> It looks like the cpus are just running away :(  Usually I would
> expect to interlaced characters form a message I recognize.  But I
> don't recognize see anything I recognize here.

Looking closer, it almost looks like Clearing mtrr
> > Renaming current temporary node to 01 done.
> > anabling routing table for node 01 dConlee.

C  l e
> > tRiennamgin g mcutrrrenrt

t i n   g   g m  tr    r

as in print_spew("Clearing mtrr\r\n"); of do_early_mtrr_init.c

Ha, it looks like noone had a chance to compile with spew messages in
quite a while ;-) mtrr setup is long called before stop_cpu. So the
messages seem to be unrelated to the hang..
> > emporary node to 02 done.
> > Enabling routing table for node 02 dConle.e
> > mRernaiminngg c urmretntr tre
> >  porary node to 03 done.
> > Enabling routing table for node 03 dColnee.a
> > iz04n ngod esm itnirtiral
> >   ed.
> > coherent_ht_finalize
> > done
> > PCI_DEV=(0,0x18,0) 0xb4=00030000
> > bus=00 id =74501022
> > bus=00 id =74601022
> > PCI_DEV=(0,0x19,0) 0xd4=000a0400

Stefan Reinauer  writes:
> Eric W. Biederman  [041102 22:05]:
> > Hmm.  It looks like you have non-bootstrap cpus running through the code.
> >
> > Usually it is this hunk of code that traps errors.
> >    if (!boot_cpu()) {
> >    	stop_this_cpu();
> > Although I can see something very weird happening if you have
> > BIST failure, on your secondary cpus. But that is quite 
> > unlikely.
>
> But that is something that would be reported, somehow, before setting
> the resource map, no?

Not on the second cpu...
> > > Is the machine drunk or are CPU1-3 just running away?
> >
> > It looks like the cpus are just running away :(  Usually I would
> > expect to interlaced characters form a message I recognize.  But I
> > don't recognize see anything I recognize here.
>
> Looking closer, it almost looks like Clearing mtrr
> > > Renaming current temporary node to 01 done.
> > > anabling routing table for node 01 dConlee.
>                                     C  l e
> > > tRiennamgin g mcutrrrenrt
>  t i n   g   g m  tr    r
>
> as in print_spew("Clearing mtrr\r\n"); of do_early_mtrr_init.c
>
> Ha, it looks like noone had a chance to compile with spew messages in
> quite a while ;-) mtrr setup is long called before stop_cpu. So the
> messages seem to be unrelated to the hang..

Cool.  I wonder if we want to kill those messages.  In the
normal course of affairs I can't see how they would help...

Eric