KVM: Don't bother the mmu if cr3 load doesn't change cr3
If the guest requests just a tlb flush, don't take the vm lock and
drop the mmu context pointlessly.
Signed-off-by: Avi Kivity <avi@qumranet.com>
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index ac09f38..15e1203 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -166,6 +166,26 @@
return ret;
}
+static bool pdptrs_changed(struct kvm_vcpu *vcpu)
+{
+ u64 pdpte[ARRAY_SIZE(vcpu->pdptrs)];
+ bool changed = true;
+ int r;
+
+ if (is_long_mode(vcpu) || !is_pae(vcpu))
+ return false;
+
+ mutex_lock(&vcpu->kvm->lock);
+ r = kvm_read_guest(vcpu->kvm, vcpu->cr3 & ~31u, pdpte, sizeof(pdpte));
+ if (r < 0)
+ goto out;
+ changed = memcmp(pdpte, vcpu->pdptrs, sizeof(pdpte)) != 0;
+out:
+ mutex_unlock(&vcpu->kvm->lock);
+
+ return changed;
+}
+
void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
{
if (cr0 & CR0_RESERVED_BITS) {
@@ -271,6 +291,11 @@
void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
{
+ if (cr3 == vcpu->cr3 && !pdptrs_changed(vcpu)) {
+ kvm_mmu_flush_tlb(vcpu);
+ return;
+ }
+
if (is_long_mode(vcpu)) {
if (cr3 & CR3_L_MODE_RESERVED_BITS) {
printk(KERN_DEBUG "set_cr3: #GP, reserved bits\n");