Re: Acer Aspire AS5672 laptop - acpi problems

On Tuesday 18 July 2006 09:20, Bruno Ducrot wrote:
> On Mon, Jul 17, 2006 at 11:38:04AM -0400, John Baldwin wrote:
> > On Monday 17 July 2006 10:55, Bruno Ducrot wrote:
> > > On Mon, Jul 17, 2006 at 04:27:09PM +0200, Torfinn Ingolfsen wrote:
> > > > On Mon, 17 Jul 2006 14:59:27 +0200
> > > > Bruno Ducrot <[email protected]> wrote:
> > > > 
> > > > > Maybe you can try without ACPI and without APIC?
> > > > 
> > > > I've already tried without acpi. I will dig out the incantation for
> > > > disabling the apic and try that.
> > > > Is there any use in testing both together (ie. no acip + no apic)?
> > > > Or is that just nonsense?
> > > 
> > > With ACPI, the interrupt for the bge is 17, and without it's 18.
> > > Maybe it should be really 17 instead of 18?
> > > 
> > > Therefore trying with atpic instead of apic may solve at least
> > > a bad interrupt routing for this device if apic is used.
> > 
> > You can force the IRQ to 18 in the non-ACPI case via a tunable in the 
loader:
> > 
> > hw.pci4.0.INTA.irq=18
> > 
> > Odd, I only see dmesg's with ACPI enabled (which have IRQ 18), I've yet to 
> > find one with ACPI disabled.  Ah found it an earlier thread which has 17 
for 
> > the !ACPI case.  So try it with ACPI disabled and the above tunable.  
Also, 
> > it should work with both ACPI and APIC disabled.
> 
> I don't see how this tunable (irq 18) can work if APIC is disabled,
> shouldn't this tunable lesser than 16 in that case?

Yes, I was intending it for the case where ACPI is disabled but APIC is 
enabled.  The last sentence is confusing I guess.  The system should just 
work without the tunable with both ACPI and APIC disabled is what I meant.

-- 
John Baldwin
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-mobile
To unsubscribe, send any mail to "[email protected]"