>>> Sergey Dyasli <sergey.dya...@citrix.com> 08/03/17 4:20 PM >>>
>On Tue, 2017-08-01 at 09:55 +0200, Egger, Christoph wrote:
>> On 18.07.17 12:34, Sergey Dyasli wrote:
>> > @@ -1773,6 +1780,27 @@ p2m_flush_nestedp2m(struct domain *d)
>> > p2m_flush_table(d->arch.nested_p2m[i]);
>> > }
>> >
>> > +void np2m_flush_eptp(struct vcpu *v, unsigned long eptp)
>> > +{
>> > + struct domain *d = v->domain;
>> > + struct p2m_domain *p2m;
>> > + unsigned int i;
>> > +
>> > + eptp &= ~(0xfffull);
>> > +
>> > + nestedp2m_lock(d);
>> > + for ( i = 0; i < MAX_NESTEDP2M; i++ )
>> > + {
>> > + p2m = d->arch.nested_p2m[i];
>> > + p2m_lock(p2m);
>> > + if ( p2m->np2m_base == eptp )
>> > + p2m_flush_table_locked(p2m);
>> > + else
>> > + p2m_unlock(p2m);
>> > + }
>> > + nestedp2m_unlock(d);
>> > +}
>> > +
>>
>> What exactly is eptp specific in this function ?
>
>Yes, good point. I seem to be too focused on Intel. The correct parameter
>name should be np2m_base, of course.
And (at the risk of stating the obvious) the function name shouldn't include
"ept" then either.
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
