2015-02-03 16:39+0100, Paolo Bonzini:
>
>
> On 03/02/2015 16:18, Radim Krčmář wrote:
> > (I see the same code in handle_ept_violation(), but wasn't that needed
> > just because of a hardware error?)
>
> That was how I read it initially, but actually that means: "this
> statement could be broken if the processor has that erratum".
Thanks, that was a nice ruse for the original bug :)
> >> +static void vmx_slot_enable_log_dirty(struct kvm *kvm,
> >> + struct kvm_memory_slot *slot)
> >> +{
> >> + kvm_mmu_slot_leaf_clear_dirty(kvm, slot);
> >
> > (New slot contains dirty pages?)
>
> New slots contain clean pages as far as the KVM dirty log is concerned.
>
> In the case of PML, note that D=1 does not mean the page is dirty. It
> only means that writes will not be logged by PML. The page may thus
> also have logging disabled.
Yeah, it would be a problem if we had dirty pages at the beginning, but
I don't think it is possible as was too lazy to check.
(It's not important and I wanted to do this review today :)
--
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
