On 19.11.2012, at 18:20, Scott Wood wrote:
> On 11/19/2012 06:36:42 AM, Alexander Graf wrote:
>> On 14.11.2012, at 05:32, Paul Mackerras wrote:
>> > +/*
>> > + * Note modification of an HPTE; set the HPTE modified bit
>> > + * if it wasn't modified before and anyone is interested.
>> > + */
>> > +static inline void note_hpte_modification(struct kvm *kvm,
>> > + struct revmap_entry *rev)
>> > +{
>> > + if (!(rev->guest_rpte & HPTE_GR_MODIFIED) &&
>> > + atomic_read(&kvm->arch.hpte_mod_interest))
>> Aren't atomic reads expensive?
>
> No, it's a simple load instruction (done with inline asm so the compiler can't break it up into smaller loads).
Oh, you're right. Fun. I would've assumed that some cache line locking needs to be done to guarantee atomicity :). But this of course makes things even easier.
Alex
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
