Hello Sergej,
On 05/07/16 09:45, Sergej Proskurin wrote:
+struct p2m_domain *p2m_get_altp2m(struct vcpu *v)
+{
+ unsigned int index = vcpu_altp2m(v).p2midx;
+
+ if ( index == INVALID_ALTP2M )
+ return NULL;
+
+ BUG_ON(index >= MAX_ALTP2M);
+
+ return v->domain->arch.altp2m_p2m[index];
+}
+
+static void p2m_init_altp2m_helper(struct domain *d, unsigned int i)
+{
+ struct p2m_domain *p2m = d->arch.altp2m_p2m[i];
+ struct vttbr_data *vttbr = &p2m->vttbr;
+
+ p2m->lowest_mapped_gfn = INVALID_GFN;
+ p2m->max_mapped_gfn = 0;
Would not it be easier to reallocate p2m from scratch everytime you
enable it?
Do you mean instead of dynamically allocating memory for all altp2m_p2m
entries at once in p2m_init_altp2m by means of p2m_init_one? If yes,
then I agree. Thankyou.
I mean that the altp2m memory should only be allocated when you use it.
I.e in p2m_init_next_altp2m.
Regards,
--
Julien Grall
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
