kvm: i8254: Deactivate APICv when using in-kernel PIT re-injection mode.

AMD SVM AVIC accelerates EOI write and does not trap. This causes
in-kernel PIT re-injection mode to fail since it relies on irq-ack
notifier mechanism. So, APICv is activated only when in-kernel PIT
is in discard mode e.g. w/ qemu option:

  -global kvm-pit.lost_tick_policy=discard

Also, introduce APICV_INHIBIT_REASON_PIT_REINJ bit to be used for this
reason.
Suggested-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NSuravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

arch/x86/include/asm/kvm_host.h

@@ -879,6 +879,7 @@ enum kvm_irqchip_mode {
 #define APICV_INHIBIT_REASON_HYPERV     1
 #define APICV_INHIBIT_REASON_NESTED     2
 #define APICV_INHIBIT_REASON_IRQWIN     3
+#define APICV_INHIBIT_REASON_PIT_REINJ  4
 
 struct kvm_arch {
 	unsigned long n_used_mmu_pages;

arch/x86/kvm/i8254.c

@@ -295,12 +295,24 @@ void kvm_pit_set_reinject(struct kvm_pit *pit, bool reinject)
 	if (atomic_read(&ps->reinject) == reinject)
 		return;
 
+	/*
+	 * AMD SVM AVIC accelerates EOI write and does not trap.
+	 * This cause in-kernel PIT re-inject mode to fail
+	 * since it checks ps->irq_ack before kvm_set_irq()
+	 * and relies on the ack notifier to timely queue
+	 * the pt->worker work iterm and reinject the missed tick.
+	 * So, deactivate APICv when PIT is in reinject mode.
+	 */
 	if (reinject) {
+		kvm_request_apicv_update(kvm, false,
+					 APICV_INHIBIT_REASON_PIT_REINJ);
 		/* The initial state is preserved while ps->reinject == 0. */
 		kvm_pit_reset_reinject(pit);
 		kvm_register_irq_ack_notifier(kvm, &ps->irq_ack_notifier);
 		kvm_register_irq_mask_notifier(kvm, 0, &pit->mask_notifier);
 	} else {
+		kvm_request_apicv_update(kvm, true,
+					 APICV_INHIBIT_REASON_PIT_REINJ);
 		kvm_unregister_irq_ack_notifier(kvm, &ps->irq_ack_notifier);
 		kvm_unregister_irq_mask_notifier(kvm, 0, &pit->mask_notifier);
 	}

arch/x86/kvm/svm.c

@@ -1739,7 +1739,13 @@ static int avic_update_access_page(struct kvm *kvm, bool activate)
 	int ret = 0;
 
 	mutex_lock(&kvm->slots_lock);
-	if (kvm->arch.apic_access_page_done == activate)
+	/*
+	 * During kvm_destroy_vm(), kvm_pit_set_reinject() could trigger
+	 * APICv mode change, which update APIC_ACCESS_PAGE_PRIVATE_MEMSLOT
+	 * memory region. So, we need to ensure that kvm->mm == current->mm.
+	 */
+	if ((kvm->arch.apic_access_page_done == activate) ||
+	    (kvm->mm != current->mm))
 		goto out;
 
 	ret = __x86_set_memory_region(kvm,
@@ -7353,7 +7359,8 @@ static bool svm_check_apicv_inhibit_reasons(ulong bit)
 	ulong supported = BIT(APICV_INHIBIT_REASON_DISABLE) |
 			  BIT(APICV_INHIBIT_REASON_HYPERV) |
 			  BIT(APICV_INHIBIT_REASON_NESTED) |
-			  BIT(APICV_INHIBIT_REASON_IRQWIN);
+			  BIT(APICV_INHIBIT_REASON_IRQWIN) |
+			  BIT(APICV_INHIBIT_REASON_PIT_REINJ);
 
 	return supported & BIT(bit);
 }