KVM: s390/diag: fix racy access of physical cpu number in diag 9c handler

We do check for target CPU == -1, but this might change at the time we
are going to use it. Hold the physical target CPU in a local variable to
avoid out-of-bound accesses to the cpu arrays.

Cc: Pierre Morel <[e-mail]>
Fixes: 87e28a15c42c ("KVM: s390: diag9c (directed yield) forwarding")
Reported-by: Marc Hartmayer <[e-mail]>
Reviewed-by: Nico Boehr <[e-mail]>
Reviewed-by: Pierre Morel <[e-mail]>
Signed-off-by: Christian Borntraeger <[e-mail]>
Signed-off-by: Janosch Frank <[e-mail]>

arch/s390/kvm/diag.c

···
 static int __diag_time_slice_end_directed(struct kvm_vcpu *vcpu)
 {
 	struct kvm_vcpu *tcpu;
+	int tcpu_cpu;
 	int tid;
 
 	tid = vcpu->run->s.regs.gprs[(vcpu->arch.sie_block->ipa & 0xf0) >> 4];
···
 		goto no_yield;
 
 	/* target guest VCPU already running */
-	if (READ_ONCE(tcpu->cpu) >= 0) {
+	tcpu_cpu = READ_ONCE(tcpu->cpu);
+	if (tcpu_cpu >= 0) {
 		if (!diag9c_forwarding_hz || diag9c_forwarding_overrun())
 			goto no_yield;
 
 		/* target host CPU already running */
-		if (!vcpu_is_preempted(tcpu->cpu))
+		if (!vcpu_is_preempted(tcpu_cpu))
 			goto no_yield;
-		smp_yield_cpu(tcpu->cpu);
+		smp_yield_cpu(tcpu_cpu);
 		VCPU_EVENT(vcpu, 5,
 			   "diag time slice end directed to %d: yield forwarded",
 			   tid);