Re: Switch APIC (+nmi, +oprofile) to driver model

Mikael Pettersson (mikpe@csd.uu.se)
Fri, 14 Feb 2003 11:32:52 +0100


Maciej W. Rozycki writes:
> On Thu, 13 Feb 2003, Mikael Pettersson wrote:
>
> > +static int __init init_local_apic_devicefs(void)
> > {
> > - if (apic_pm_state.active)
> > - pm_register(PM_SYS_DEV, 0, apic_pm_callback);
> > + if (!cpu_has_apic)
>
> This looks broken -- what if an external local APIC is present?

My goal was to not change any behaviour from our current code, and from
what I can tell, our current code does not support PM suspend and resume
for old external-local-APIC machines. (They're mostly 486 MPs, right?)

The suspend/resume procedures only work on P6/K7 and up. There's a
bug there in that we may try to run the suspend on a UP P5 with enabled
local APIC, which won't work. So far, no one seems to have noticed :->

(Admittedly, the P5 errata sheets say "BIOS should disable the P5 local
APIC on UP" so I don't think this case is very likely.)

/Mikael
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/