On Thu, Mar 31, 2005 at 02:49:29PM +0200, Joel CARNAT wrote:
> On the last episode (Thu, 31 Mar 2005 12:22:36 +0200), Martin Husemann <martin@duskware.de> said:
> 
> > On Thu, Mar 31, 2005 at 12:09:16PM +0200, Joel CARNAT wrote:
> > >  this one tells me everything's all right (cf attached).
> > >  except I have 512Mo of RAM and memtester only use 128.
> > >  on PROM's memory check, 512 are seen (same as is dmesg).
> > 
> > That's probably your resource limit for data size - try to unlimit
> > it (depends on your shell, look at ulimit/limit/unlimit).
> > 
> > But the bonnie crash is more interesting.
> > 
> 
> woops... since the freeze during "hbench-OS", the U10 won't boot anymore...
> ****************************************************************
> NetBSD 3.0_BETA (GENERIC) #2: Tue Mar 29 14:19:20 CEST 2005
>         jcarnat@beezdi:/usr/obj/sys/arch/sparc64/compile/GENERIC
> total memory = 512 MB
> avail memory = 490 MB
> bootpath: /pci@1f,0/pci@1,0/scsi@3,0/disk@1,0
> mainbus0 (root): SUNW,Ultra-5_10: hostid 80c66894
> cpu0 at mainbus0: SUNW,UltraSPARC-IIi @ 440 MHz, version 0 FPU
> cpu0: 32K instruction (32 b/l), 16K data (32 b/l), 2048K external (64 b/l)
> psycho0 at mainbus0 addr 0xfffc4000
> SUNW,sabre: impl 0, version 0: ign 7c0 bus range 0 to 2; PCI bus 0
> DVMA map: c0000000 to e0000000
> IOTSB: 864000 to 8e4000
> pci0 at psycho0
> panic: psycho0: PCI bus A error AFAR 1fff1400060 AFSR 100000000
> kdb breakpoint at 135d72c
> Stopped in pid 0.1 (swapper) at netbsd:cpu_Debugger+0x4:        nop
> db> 
> ****************************************************************
> 
> looks like she's upset ;)
> any useful gdb command to run in this state (before I re-install from CD) ?

This is quite strange.  You got an interrupt indicating a PCI error,
but there are no syndrome bits set in the AFSR so it doesn't look like
this is a valid error.

Eduardo