Re: [PATCH v3 1/7] KVM: MMU: fix release noslot pfn

[Avi Kivity <avi@redhat.com>]

On 09/24/2012 02:32 PM, Xiao Guangrong wrote:

> 3 places after the whole patchset (There are some cleanups after this patch).
> 
>> and one by even stronger is_error_pfn(). 
> 
> This one is:
> 
> |	if (!is_error_pfn(pfn)) {
> |                kvm_release_pfn_clean(pfn);
> |                return true;
> |	}
> |
> |	return false;
> 
> We can change it to:
> 
> | if (is_error_pfn(pfn))
> |	return false;
> |
> | kvm_release_pfn_clean(pfn);
> | return true;
> 
>> I guess when/if other architectures will add MMIO MMU
>> caching they will need to guard kvm_release_pfn_clean() by is_noslot_pfn()
>> too in most cases. I am not insisting, but as this patch shows it is
>> easy to miss the check before calling the function.
> 
> Sounds reasonable. I will consider it if Avi/Marcelo have no object on
> it.

I think it's a good idea.

Looks like we traded the unscalable error pages for these branches, I
think it's a reasonable tradeoff.

-- 
error compiling committee.c: too many arguments to function