Re: [PATCH] linux-2.5.69_subarch-fix_A0.patch

[john stultz <johnstul@us.ibm.com>]

On Wed, 2003-05-14 at 23:51, Andi Kleen wrote:
> On Wed, May 14, 2003 at 07:37:09PM -0700, john stultz wrote:
> > All,
> > 	This patch fixes a circular dependency (a function in mach_apic.h
> > requires hard_smp_processor_id() and hard_smp_processor_id() requires
> > macros from mach_apic.h) that has been in the subarch code for a bit,
> > but was hacked around with some #ifdefs. 
> > 
> > With the inclusion of the generic-subarch the hack was dropped and
> > bigsmp and summit promptly broke. So this makes things compile again. 
> 
> What broke exactly? I thought worked around this problem for the generic
> subarchitecture.


GET_APIC_ID was moved into mach_apic.h (it used to be wrapped in
CONFIG_X86_SUMMIT in apicdef.h). However, init_apic_ldr()  in
mach_apic.h uses hard_smp_processor_id() (in smp.h), which to complete
the circle, requires GET_APIC_ID from mach_apic.h

You did get around it in the generic subarch (which I love, by the way,
thanks so much for doing that work), but in a roundabout way. (via
#ifdef APIC_DEFINITION trickery). 


> Accessing the APIC directly this way just to work around a bad include
> looks quite ugly to me.

Yea, ok. I guess we can move GET_APIC_ID out into its own mach specific
.h file. That will eliminate the circularity as well. I'll respin the
patch later today.

thanks
-john