> > > ENABLING IO-APIC IRQs
> > > ..... changing IO-APIC physical APIC ID to 16
> > > kernel panic: could not set ID
> > > In swapper task - not syncing
> >
> > Here's my "me too" on the subject. The system affected is also a HP
> > Vectra XU with dual PPros. The interesting thing here is that two
> > slightly older Vectra XUs run 2.2.14/SMP just fine.
>
> I have a workaround for this for 2.2.16pre. It seems to be a bug in the BIOS
> tables. We now spot stupid table entries and pray the bios setup APIC value
> was ok
I grabbed 2.2.16-pre2 and tried it on the problem machine. It now gets
past all the APIC init code, complete with the new message, but then it
hangs in the IDE detection.
Normally the detection looks like this:
CMD646: IDE controller on PCI bus 00 dev 08
CMD646: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xfc00-0xfc07, BIOS settings: hda:pio, hdb:pio
ide1: BM-DMA at 0xfc08-0xfc0f, BIOS settings: hdc:pio, hdd:pio
hdc: HITACHI CDR-8130, ATAPI CDROM drive
ide1 at 0x170-0x177,0x376 on irq 15
hdc: ATAPI 16X CD-ROM drive, 128kB Cache
.. whereas with 2.2.16-pre2, it hangs right before the last line shown.
It's a solid lock, with both magic sysrq and the console scrollback dead.
Using noapic avoids this hang, but seems to be the easy way out.
Guinea pig services are available to anyone with ideas / code to try.
I'll start sprinkling some printks around the source to see where
it's getting stuck. My guess is that it's not making it into
ide_cdrom_probe_capabilities, but the real question is why.
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/dmentre/smp-howto/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
