RE: [Xen-devel] please revert c/s 17686

["Wei, Gang" <gang.wei@xxxxxxxxx>]

On Friday, June 13, 2008 4:25 PM, Keir Fraser wrote:
> On 13/6/08 09:12, "Jan Beulich" <jbeulich@xxxxxxxxxx> wrote:
> 
>>>> I missed that warning printk. It does indeed look odd.
>>> 
>>> As to this warning printk, we can simply replace it with an assert.
>> 
>> That would make things worse, not better - the condition simply must
be
>> allowed (as said before, otherwise you won't be able to bring all
CPUs
>> at once into C3).
> 
> I think that C2/C3 are entered with IRQs disabled, but IRQ pending
will
> kick the CPU out of C2/C3 nonetheless. That CPU will then execute
> hpet_broadcast_exit() before local_irq_enable() and hence the warning
> printk will never actually fire. So it would be correct as a BUG_ON().
> 
> Is this correct, Wei?

Absolutely right. And we will look into whether it is better to move
hpet_broadcast_exit() after local_irq_enable().

Jimmy

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel