Interrupt routing issue on Compaq nx6325

Ariff Abdullah ariff at FreeBSD.org
Thu Dec 14 10:18:55 PST 2006


On Thu, 14 Dec 2006 17:29:47 +0100
Joerg Wunsch <j at uriah.heep.sax.de> wrote:
> As John Baldwin wrote:
> 
> > > My current workaround is (for RELENG_6)
> > > 
> > > http://people.freebsd.org/~ariff/misc/idlecpu_apic_4.diff
> > > 
> > > (hw.apic.idle_cpu=1 or 2)
> 
> > This doesn't sound like his problem.  Joerg, can you get a linux
> > dmesg and a freebsd dmesg from the box so I can compare them?
> 
> Well, at least Ariff's workaround helped my system as well.  My
> only/major concern with it is about the additional power consumption
> when running on batteries, but for the time being, I can live with
> two different boot modes for this: one that uses HLT and only 1 CPU
> (without the APIC) for minimal battery power consumption, and
> another one using the APIC and both CPU cores, with HLT disabled.
> 
> After reading through the lengthy thread in the PR, yes, most of the
> symptoms mentioned there apply to my nx6325 as well.  In particular,
> booting into single-user mode, bouncing caps lock a number of times
> (so the prompt appears), and starting a background job to eat up one
> entire CPU did make the machine work as well (but with certainly
> worse impact on power consumption than Ariff's workaround).
>

Just ignore large part of it. Most of the workarounds (disabling
hlt, selective hlt) there does not applicable anymore with my above
patch which:
1) HLT enabled, 2) APIC enabled 3) Piggybacking APIC
timer through RTC interrupt (hw.apic.idle_cpu=1)

Workaround is still a workaround, though.

> I've put the boot messages up at
> 
> http://www.sax.de/~joerg/dmesg-nx6325.tar.bz2
> 
> What in particular caught my eye in the Linux messages is:
> 
> Dec 12 23:33:26 remi kernel: ENABLING IO-APIC IRQs
> Dec 12 23:33:26 remi kernel: ..TIMER: vector=0x31 apic1=0 pin1=2
> apic2=-1 pin2=-1 Dec 12 23:33:26 remi kernel: ..MP-BIOS bug: 8254
> timer not connected to IO-APIC Dec 12 23:33:26 remi kernel:
> ...trying to set up timer (IRQ0) through the 8259A ...  failed. Dec
> 12 23:33:26 remi kernel: ...trying to set up timer as Virtual Wire
> IRQ... works.
> 
> I didn't quite follow the referred to discussion of the Linux kernel
> hackers, is the above their workaround, or is that something
> entirely different?
> 
> What's also strange to me is that on my boss'es nx6325 (BIOS
> revision F.02 instead of the F.03 in my machine), the 6.2-BETA3 CD
> boots without troubles.  That's not SMP (just the GENERIC kernel),
> but for me, even in uniprocessor mode I cannot get it to run unless
> I disable the APIC.  I tried using the F.02 revision ACPI AML code,
> but to no avail (but in retrospect, I might have forgotten to
> re-check that with a GENERIC kernel).
> 
> I've also got the F.02 dmesg output somewhere, but unfortunately, it
> appears I've lost the USB stick they are supposed to be on. :-(  I
> can refetch them from that other machine though, but not before next
> week.
> 


--
Ariff Abdullah
FreeBSD

... Recording in stereo is obviously too advanced
    and confusing for us idiot ***** users :P ........
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-acpi/attachments/20061214/6117e5fb/attachment.pgp


More information about the freebsd-acpi mailing list