KVM: s390: Make facility bits future-proof
Christian Borntraeger [Fri, 29 Aug 2008 11:29:45 +0000 (13:29 +0200)]
Heiko Carstens pointed out, that its safer to activate working facilities
instead of disabling problematic facilities. The new code uses the host
facility bits and masks it with known good ones.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

arch/s390/kvm/priv.c

index d1faf5c..cce40ff 100644 (file)
@@ -157,8 +157,8 @@ static int handle_stfl(struct kvm_vcpu *vcpu)
        int rc;
 
        vcpu->stat.instruction_stfl++;
-       facility_list &= ~(1UL<<24); /* no stfle */
-       facility_list &= ~(1UL<<23); /* no large pages */
+       /* only pass the facility bits, which we can handle */
+       facility_list &= 0xfe00fff3;
 
        rc = copy_to_guest(vcpu, offsetof(struct _lowcore, stfl_fac_list),
                           &facility_list, sizeof(facility_list));