RE: 2.6.12 with dual way dual core ck804 MB

[YhLu]

andi,

some hints for you.
when using Opteron 875, the problem was more easily catched. 

in smp_callin in smpboot.c
if change
Dprintk("CPU#%d (phys ID: %d) waiting for CALLOUT\n", cpuid, phys_id);
to 
printk("CPU#%d (phys ID: %d) waiting for CALLOUT\n", cpuid, phys_id);
will have 50% to pass through.

if change
Dprintk("CALLIN, before setup_local_APIC().\n");
to 
printk("CALLIN, before setup_local_APIC().\n");

will pass through every time.

YH



> -----Original Message-----
> From: YhLu 
> Sent: Friday, June 24, 2005 4:44 PM
> To: 'Andi Kleen'
> Cc: Peter Buckingham; linux-kernel@xxxxxxxxxxxxxxx
> Subject: RE: 2.6.12 with dual way dual core ck804 MB
> 
> start from 2.6.12.rc5.
> 
> YH 
> 
> > -----Original Message-----
> > From: Andi Kleen [mailto:ak@xxxxxxx]
> > Sent: Friday, June 24, 2005 4:42 PM
> > To: YhLu
> > Cc: Andi Kleen; Peter Buckingham; linux-kernel@xxxxxxxxxxxxxxx
> > Subject: Re: 2.6.12 with dual way dual core ck804 MB
> > 
> > On Fri, Jun 24, 2005 at 04:42:58PM -0700, YhLu wrote:
> > > Andi,
> > > 
> > > the timing problem come out from 2.6.12.rc5....
> > 
> > What do you mean? It's gone or it started there?
> > -Andi
> > 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/