KVM: fix polling for guest halt continued even if disable it

If there is already some polling ongoing, it's impossible to disable the
polling, since as soon as somebody sets halt_poll_ns to 0, polling will
never stop, as grow and shrink are only handled if halt_poll_ns is != 0.

This patch fix it by reset vcpu->halt_poll_ns in order to stop polling
when polling is disabled.

Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

virt/kvm/kvm_main.c

@@ -2043,7 +2043,8 @@ out:
 		else if (vcpu->halt_poll_ns < halt_poll_ns &&
 			block_ns < halt_poll_ns)
 			grow_halt_poll_ns(vcpu);
-	}
+	} else
+		vcpu->halt_poll_ns = 0;
 
 	trace_kvm_vcpu_wakeup(block_ns, waited);
 }