[PATCH v3 07/12] KVM: x86: APICv: fix race in kvm_request_apicv_update on SVM

[Maxim Levitsky]

Currently on SVM, the kvm_request_apicv_update toggles the APICv
memslot without doing any synchronization.

If there is a mismatch between that memslot state and the AVIC state,
on one of the vCPUs, an APIC mmio access can be lost:

For example:

VCPU0: enable the APIC_ACCESS_PAGE_PRIVATE_MEMSLOT
VCPU1: access an APIC mmio register.

Since AVIC is still disabled on VCPU1, the access will not be intercepted
by it, and neither will it cause MMIO fault, but rather it will just be
read/written from/to the dummy page mapped into the
APIC_ACCESS_PAGE_PRIVATE_MEMSLOT.

Fix that by adding a lock guarding the AVIC state changes, and carefully
order the operations of kvm_request_apicv_update to avoid this race:

1. Take the lock
2. Send KVM_REQ_APICV_UPDATE
3. Update the apic inhibit reason
4. Release the lock

This ensures that at (2) all vCPUs are kicked out of the guest mode,
but don't yet see the new avic state.
Then only after (4) all other vCPUs can update their AVIC state and resume.

Signed-off-by: Maxim Levitsky <mlevitsk@xxxxxxxxxx>
---
 arch/x86/include/asm/kvm_host.h |  3 +++
 arch/x86/kvm/x86.c              | 36 +++++++++++++++++++--------------
 2 files changed, 24 insertions(+), 15 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 430447ce19c0..80323b5fb20a 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1054,6 +1054,9 @@ struct kvm_arch {
 	struct kvm_apic_map __rcu *apic_map;
 	atomic_t apic_map_dirty;
 
+	/* Protects apic_access_memslot_enabled and apicv_inhibit_reasons */
+	struct mutex apicv_update_lock;
+
 	bool apic_access_memslot_enabled;
 	unsigned long apicv_inhibit_reasons;
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 1de7d97341dd..9d2e4594c4eb 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -8579,6 +8579,8 @@ EXPORT_SYMBOL_GPL(kvm_apicv_activated);
 
 static void kvm_apicv_init(struct kvm *kvm)
 {
+	mutex_init(&kvm->arch.apicv_update_lock);
+
 	if (enable_apicv)
 		clear_bit(APICV_INHIBIT_REASON_DISABLE,
 			  &kvm->arch.apicv_inhibit_reasons);
@@ -9240,6 +9242,8 @@ void kvm_vcpu_update_apicv(struct kvm_vcpu *vcpu)
 	if (!lapic_in_kernel(vcpu))
 		return;
 
+	mutex_lock(&vcpu->kvm->arch.apicv_update_lock);
+
 	vcpu->arch.apicv_active = kvm_apicv_activated(vcpu->kvm);
 	kvm_apic_update_apicv(vcpu);
 	static_call(kvm_x86_refresh_apicv_exec_ctrl)(vcpu);
@@ -9252,41 +9256,43 @@ void kvm_vcpu_update_apicv(struct kvm_vcpu *vcpu)
 	 */
 	if (!vcpu->arch.apicv_active)
 		kvm_make_request(KVM_REQ_EVENT, vcpu);
+
+	mutex_unlock(&vcpu->kvm->arch.apicv_update_lock);
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_update_apicv);
 
 void kvm_request_apicv_update(struct kvm *kvm, bool activate, ulong bit)
 {
-	unsigned long old, new, expected;
+	unsigned long old, new;
 
 	if (!kvm_x86_ops.check_apicv_inhibit_reasons ||
 	    !static_call(kvm_x86_check_apicv_inhibit_reasons)(bit))
 		return;
 
-	old = READ_ONCE(kvm->arch.apicv_inhibit_reasons);
-	do {
-		expected = new = old;
-		if (activate)
-			__clear_bit(bit, &new);
-		else
-			__set_bit(bit, &new);
-		if (new == old)
-			break;
-		old = cmpxchg(&kvm->arch.apicv_inhibit_reasons, expected, new);
-	} while (old != expected);
+	mutex_lock(&kvm->arch.apicv_update_lock);
+
+	old = new = kvm->arch.apicv_inhibit_reasons;
+	if (activate)
+		__clear_bit(bit, &new);
+	else
+		__set_bit(bit, &new);
+
+	kvm->arch.apicv_inhibit_reasons = new;
 
 	if (!!old == !!new)
-		return;
+		goto out;
 
 	trace_kvm_apicv_update_request(activate, bit);
 
+	kvm_make_all_cpus_request(kvm, KVM_REQ_APICV_UPDATE);
+
 	if (!activate)
 		kvm_zap_gfn_range(kvm, gpa_to_gfn(APIC_DEFAULT_PHYS_BASE),
 				  gpa_to_gfn(APIC_DEFAULT_PHYS_BASE + PAGE_SIZE));
 
 	kvm->arch.apic_access_memslot_enabled = activate;
-
-	kvm_make_all_cpus_request(kvm, KVM_REQ_APICV_UPDATE);
+out:
+	mutex_unlock(&kvm->arch.apicv_update_lock);
 }
 EXPORT_SYMBOL_GPL(kvm_request_apicv_update);
 
-- 
2.26.3