Re: [PATCH v5 07/14] nEPT: Add EPT tables support to paging_tmpl.h

[Gleb Natapov <gleb@redhat.com>]

On Thu, Aug 01, 2013 at 01:05:47PM +0200, Paolo Bonzini wrote:
>  On Aug 01 2013, Xiao Guangrong wrote:
> > On 08/01/2013 03:42 PM, Gleb Natapov wrote:
> > > On Thu, Aug 01, 2013 at 03:31:01PM +0800, Xiao Guangrong wrote:
> > >> On 08/01/2013 03:18 PM, Xiao Guangrong wrote:
> > >>  +#endif
> > >>>>>
> > >>>>> Hmm, why not use shadow_x_mask, shadow_user_mask instead? PT_WRITABLE_MASK
> > >>>>> is also suitable for ept, i guess we can remove the "#if/#else/#endif" after
> > >>>>> that.
> > >>>>>
> > >>>> shadow_x_mask and shadow_user_mask do not depend on guest paging mode,
> > >>>> so cannot be used here. Since we have to use ifdefs anyway relying on
> > >>>> VMX_EPT_WRITABLE_MASK == PT_WRITABLE_MASK is not necessary. Makes code
> > >>>> easier to read.
> > >>>
> > >>> Oh, yes, you are right.
> > >>>
> > >>> Reviewed-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
> > >>
> > >> BTW, i notice the code in mmu.c uses PT64_NX_MASK to check the permission,
> > >> i.e:
> > >>
> > >> static bool need_remote_flush(u64 old, u64 new)
> > >> {
> > >> 	if (!is_shadow_present_pte(old))
> > >> 		return false;
> > >> 	if (!is_shadow_present_pte(new))
> > >> 		return true;
> > >> 	if ((old ^ new) & PT64_BASE_ADDR_MASK)
> > >> 		return true;
> > >> 	old ^= PT64_NX_MASK;
> > >> 	new ^= PT64_NX_MASK;
> > >> 	return (old & ~new & PT64_PERM_MASK) != 0;
> > >> }
> > >>
> > >> It checks shadow page table and the mask is wrong one nest ept spte.
> > > So shadow_x_mask need to be used here, correct?
> > 
> > Yes. The code checks shadow page table which does not depend on guest mode. :)
> 
> The XOR should be with shadow_nx_mask, no?  And PT64_PERM_MASK
> should include both shadow_x_mask and shadow_nx_mask, I think.
> 
Yes :) That what I did eventually.

--
			Gleb.