From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marcelo Tosatti <mtosatti@redhat.com>
Subject: [patch 1/2] KVM: consolidate check for pending vcpu requests
Date: Fri, 06 Jun 2008 16:37:35 -0300
Message-ID: <20080606194010.897342530@localhost.localdomain>
References: <20080606193734.495417169@localhost.localdomain>
Cc: kvm@vger.kernel.org, Marcelo Tosatti <mtosatti@redhat.com>
To: Avi Kivity <avi@qumranet.com>, Chris Wright <chrisw@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([66.187.233.31]:53153 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1755087AbYFFTl7 (ORCPT <rfc822;kvm@vger.kernel.org>);
	Fri, 6 Jun 2008 15:41:59 -0400
Content-Disposition: inline; filename=better-vcpu-run-request-handling
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

A guest vcpu instance can be scheduled to a different physical CPU
between the test for KVM_REQ_MIGRATE_TIMER and local_irq_disable().

If that happens, the timer will only be migrated to the current pCPU on
the next exit, meaning that guest LAPIC timer event can be delayed until
a host interrupt is triggered.

Fix it by cancelling guest entry if any vcpu request is pending.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Index: kvm/arch/x86/kvm/x86.c
===================================================================
--- kvm.orig/arch/x86/kvm/x86.c
+++ kvm/arch/x86/kvm/x86.c
@@ -2798,6 +2798,8 @@ again:
 	if (vcpu->requests) {
 		if (test_and_clear_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests))
 			__kvm_migrate_timers(vcpu);
+		if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
+			kvm_x86_ops->tlb_flush(vcpu);
 		if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS,
 				       &vcpu->requests)) {
 			kvm_run->exit_reason = KVM_EXIT_TPR_ACCESS;
@@ -2820,21 +2822,13 @@ again:
 
 	local_irq_disable();
 
-	if (need_resched()) {
+	if (vcpu->requests || need_resched()) {
 		local_irq_enable();
 		preempt_enable();
 		r = 1;
 		goto out;
 	}
 
-	if (vcpu->requests)
-		if (test_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests)) {
-			local_irq_enable();
-			preempt_enable();
-			r = 1;
-			goto out;
-		}
-
 	if (signal_pending(current)) {
 		local_irq_enable();
 		preempt_enable();
@@ -2864,9 +2858,6 @@ again:
 
 	kvm_guest_enter();
 
-	if (vcpu->requests)
-		if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
-			kvm_x86_ops->tlb_flush(vcpu);
 
 	KVMTRACE_0D(VMENTRY, vcpu, entryexit);
 	kvm_x86_ops->run(vcpu, kvm_run);

--