KVM: hyperv: fix locking of struct kvm_hv fields

Introduce a new mutex to avoid an AB-BA deadlock between kvm->lock and
vcpu->mutex.  Protect accesses in kvm_hv_setup_tsc_page too, as suggested
by Roman.

Reported-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Roman Kagan <rkagan@virtuozzo.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Documentation/virtual/kvm/locking.txt

@@ -13,8 +13,12 @@ The acquisition orders for mutexes are as follows:
 - kvm->slots_lock is taken outside kvm->irq_lock, though acquiring
   them together is quite rare.
 
-For spinlocks, kvm_lock is taken outside kvm->mmu_lock.  Everything
-else is a leaf: no other lock is taken inside the critical sections.
+On x86, vcpu->mutex is taken outside kvm->arch.hyperv.hv_lock.
+
+For spinlocks, kvm_lock is taken outside kvm->mmu_lock.
+
+Everything else is a leaf: no other lock is taken inside the critical
+sections.
 
 2: Exception
 ------------

arch/x86/include/asm/kvm_host.h

@@ -704,6 +704,7 @@ struct kvm_apic_map {
 
 /* Hyper-V emulation context */
 struct kvm_hv {
+	struct mutex hv_lock;
 	u64 hv_guest_os_id;
 	u64 hv_hypercall;
 	u64 hv_tsc_page;

arch/x86/kvm/hyperv.c

@@ -852,14 +852,18 @@ void kvm_hv_setup_tsc_page(struct kvm *kvm,
 	if (!(hv->hv_tsc_page & HV_X64_MSR_TSC_REFERENCE_ENABLE))
 		return;
 
+	mutex_lock(&kvm->arch.hyperv.hv_lock);
+	if (!(hv->hv_tsc_page & HV_X64_MSR_TSC_REFERENCE_ENABLE))
+		goto out_unlock;
+
 	gfn = hv->hv_tsc_page >> HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT;
 	/*
 	 * Because the TSC parameters only vary when there is a
 	 * change in the master clock, do not bother with caching.
 	 */
 	if (unlikely(kvm_read_guest(kvm, gfn_to_gpa(gfn),
 				    &tsc_seq, sizeof(tsc_seq))))
-		return;
+		goto out_unlock;
 
 	/*
 	 * While we're computing and writing the parameters, force the
@@ -868,15 +872,15 @@ void kvm_hv_setup_tsc_page(struct kvm *kvm,
 	hv->tsc_ref.tsc_sequence = 0;
 	if (kvm_write_guest(kvm, gfn_to_gpa(gfn),
 			    &hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence)))
-		return;
+		goto out_unlock;
 
 	if (!compute_tsc_page_parameters(hv_clock, &hv->tsc_ref))
-		return;
+		goto out_unlock;
 
 	/* Ensure sequence is zero before writing the rest of the struct.  */
 	smp_wmb();
 	if (kvm_write_guest(kvm, gfn_to_gpa(gfn), &hv->tsc_ref, sizeof(hv->tsc_ref)))
-		return;
+		goto out_unlock;
 
 	/*
 	 * Now switch to the TSC page mechanism by writing the sequence.
@@ -891,6 +895,8 @@ void kvm_hv_setup_tsc_page(struct kvm *kvm,
 	hv->tsc_ref.tsc_sequence = tsc_seq;
 	kvm_write_guest(kvm, gfn_to_gpa(gfn),
 			&hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence));
+out_unlock:
+	mutex_unlock(&kvm->arch.hyperv.hv_lock);
 }
 
 static int kvm_hv_set_msr_pw(struct kvm_vcpu *vcpu, u32 msr, u64 data,
@@ -1142,9 +1148,9 @@ int kvm_hv_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data, bool host)
 	if (kvm_hv_msr_partition_wide(msr)) {
 		int r;
 
-		mutex_lock(&vcpu->kvm->lock);
+		mutex_lock(&vcpu->kvm->arch.hyperv.hv_lock);
 		r = kvm_hv_set_msr_pw(vcpu, msr, data, host);
-		mutex_unlock(&vcpu->kvm->lock);
+		mutex_unlock(&vcpu->kvm->arch.hyperv.hv_lock);
 		return r;
 	} else
 		return kvm_hv_set_msr(vcpu, msr, data, host);
@@ -1155,17 +1161,17 @@ int kvm_hv_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
 	if (kvm_hv_msr_partition_wide(msr)) {
 		int r;
 
-		mutex_lock(&vcpu->kvm->lock);
+		mutex_lock(&vcpu->kvm->arch.hyperv.hv_lock);
 		r = kvm_hv_get_msr_pw(vcpu, msr, pdata);
-		mutex_unlock(&vcpu->kvm->lock);
+		mutex_unlock(&vcpu->kvm->arch.hyperv.hv_lock);
 		return r;
 	} else
 		return kvm_hv_get_msr(vcpu, msr, pdata);
 }
 
 bool kvm_hv_hypercall_enabled(struct kvm *kvm)
 {
-	return kvm->arch.hyperv.hv_hypercall & HV_X64_MSR_HYPERCALL_ENABLE;
+	return READ_ONCE(kvm->arch.hyperv.hv_hypercall) & HV_X64_MSR_HYPERCALL_ENABLE;
 }
 
 static void kvm_hv_hypercall_set_result(struct kvm_vcpu *vcpu, u64 result)

arch/x86/kvm/x86.c

@@ -7881,6 +7881,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 
 	raw_spin_lock_init(&kvm->arch.tsc_write_lock);
 	mutex_init(&kvm->arch.apic_map_lock);
+	mutex_init(&kvm->arch.hyperv.hv_lock);
 	spin_lock_init(&kvm->arch.pvclock_gtod_sync_lock);
 
 	kvm->arch.kvmclock_offset = -ktime_get_boot_ns();