KVM: s390: vsie: speed up VCPU external calls

Whenever a SIGP external call is injected via the SIGP external call
interpretation facility, the VCPU is not kicked. When a VCPU is currently
in the VSIE, the external call might not be processed immediately.

Therefore we have to provoke partial execution exceptions, which leads to a
kick of the VCPU and therefore also kick out of VSIE. This is done by
simulating the WAIT state. This bit has no other side effects.

Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c
index c8c8763e78229bbf27ef524b9061b1cd198a2e3f..90781ba528030541d90d2d12487fd8819f7130e8 100644 (file)
--- a/arch/s390/kvm/vsie.c
+++ b/arch/s390/kvm/vsie.c
@@ -844,6 +844,11 @@ static void register_shadow_scb(struct kvm_vcpu *vcpu,
                                struct vsie_page *vsie_page)
 {
        WRITE_ONCE(vcpu->arch.vsie_block, &vsie_page->scb_s);
+       /*
+        * External calls have to lead to a kick of the vcpu and
+        * therefore the vsie -> Simulate Wait state.
+        */
+       atomic_or(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags);
 }
 
 /*
@@ -851,6 +856,7 @@ static void register_shadow_scb(struct kvm_vcpu *vcpu,
  */
 static void unregister_shadow_scb(struct kvm_vcpu *vcpu)
 {
+       atomic_andnot(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags);
        WRITE_ONCE(vcpu->arch.vsie_block, NULL);
 }