KVM: arm64: Avoid vcpu->mutex v. kvm->lock inversion in CPU_ON

KVM/arm64 had the lock ordering backwards on vcpu->mutex and kvm->lock
from the very beginning. One such example is the way vCPU resets are
handled: the kvm->lock is acquired while handling a guest CPU_ON PSCI
call.

Add a dedicated lock to serialize writes to kvm_vcpu_arch::{mp_state,
reset_state}. Promote all accessors of mp_state to {READ,WRITE}_ONCE()
as readers do not acquire the mp_state_lock. While at it, plug yet
another race by taking the mp_state_lock in the KVM_SET_MP_STATE ioctl
handler.

As changes to MP state are now guarded with a dedicated lock, drop the
kvm->lock acquisition from the PSCI CPU_ON path. Similarly, move the
reader of reset_state outside of the kvm->lock and instead protect it
with the mp_state_lock. Note that writes to reset_state::reset have been
demoted to regular stores as both readers and writers acquire the
mp_state_lock.

While the kvm->lock inversion still exists in kvm_reset_vcpu(), at least
now PSCI CPU_ON no longer depends on it for serializing vCPU reset.

Cc: stable@vger.kernel.org
Tested-by: Jeremy Linton <jeremy.linton@arm.com>
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20230327164747.2466958-2-oliver.upton@linux.dev

1 files changed, 16 insertions, 12 deletions

diff --git a/arch/arm64/kvm/psci.c b/arch/arm64/kvm/psci.c
index 7fbc4c1b9df0..5767e6baa61a 100644
--- a/arch/arm64/kvm/psci.c
+++ b/arch/arm64/kvm/psci.c
@@ -62,6 +62,7 @@ static unsigned long kvm_psci_vcpu_on(struct kvm_vcpu *source_vcpu)
 	struct vcpu_reset_state *reset_state;
 	struct kvm *kvm = source_vcpu->kvm;
 	struct kvm_vcpu *vcpu = NULL;
+	int ret = PSCI_RET_SUCCESS;
 	unsigned long cpu_id;
 
 	cpu_id = smccc_get_arg1(source_vcpu);
@@ -76,11 +77,15 @@ static unsigned long kvm_psci_vcpu_on(struct kvm_vcpu *source_vcpu)
 	 */
 	if (!vcpu)
 		return PSCI_RET_INVALID_PARAMS;
+
+	spin_lock(&vcpu->arch.mp_state_lock);
 	if (!kvm_arm_vcpu_stopped(vcpu)) {
 		if (kvm_psci_version(source_vcpu) != KVM_ARM_PSCI_0_1)
-			return PSCI_RET_ALREADY_ON;
+			ret = PSCI_RET_ALREADY_ON;
 		else
-			return PSCI_RET_INVALID_PARAMS;
+			ret = PSCI_RET_INVALID_PARAMS;
+
+		goto out_unlock;
 	}
 
 	reset_state = &vcpu->arch.reset_state;
@@ -96,7 +101,7 @@ static unsigned long kvm_psci_vcpu_on(struct kvm_vcpu *source_vcpu)
 	 */
 	reset_state->r0 = smccc_get_arg3(source_vcpu);
 
-	WRITE_ONCE(reset_state->reset, true);
+	reset_state->reset = true;
 	kvm_make_request(KVM_REQ_VCPU_RESET, vcpu);
 
 	/*
@@ -108,7 +113,9 @@ static unsigned long kvm_psci_vcpu_on(struct kvm_vcpu *source_vcpu)
 	vcpu->arch.mp_state.mp_state = KVM_MP_STATE_RUNNABLE;
 	kvm_vcpu_wake_up(vcpu);
 
-	return PSCI_RET_SUCCESS;
+out_unlock:
+	spin_unlock(&vcpu->arch.mp_state_lock);
+	return ret;
 }
 
 static unsigned long kvm_psci_vcpu_affinity_info(struct kvm_vcpu *vcpu)
@@ -168,8 +175,11 @@ static void kvm_prepare_system_event(struct kvm_vcpu *vcpu, u32 type, u64 flags)
 	 * after this call is handled and before the VCPUs have been
 	 * re-initialized.
 	 */
-	kvm_for_each_vcpu(i, tmp, vcpu->kvm)
-		tmp->arch.mp_state.mp_state = KVM_MP_STATE_STOPPED;
+	kvm_for_each_vcpu(i, tmp, vcpu->kvm) {
+		spin_lock(&tmp->arch.mp_state_lock);
+		WRITE_ONCE(tmp->arch.mp_state.mp_state, KVM_MP_STATE_STOPPED);
+		spin_unlock(&tmp->arch.mp_state_lock);
+	}
 	kvm_make_all_cpus_request(vcpu->kvm, KVM_REQ_SLEEP);
 
 	memset(&vcpu->run->system_event, 0, sizeof(vcpu->run->system_event));
@@ -229,7 +239,6 @@ static unsigned long kvm_psci_check_allowed_function(struct kvm_vcpu *vcpu, u32
 
 static int kvm_psci_0_2_call(struct kvm_vcpu *vcpu)
 {
-	struct kvm *kvm = vcpu->kvm;
 	u32 psci_fn = smccc_get_function(vcpu);
 	unsigned long val;
 	int ret = 1;
@@ -254,9 +263,7 @@ static int kvm_psci_0_2_call(struct kvm_vcpu *vcpu)
 		kvm_psci_narrow_to_32bit(vcpu);
 		fallthrough;
 	case PSCI_0_2_FN64_CPU_ON:
-		mutex_lock(&kvm->lock);
 		val = kvm_psci_vcpu_on(vcpu);
-		mutex_unlock(&kvm->lock);
 		break;
 	case PSCI_0_2_FN_AFFINITY_INFO:
 		kvm_psci_narrow_to_32bit(vcpu);
@@ -395,7 +402,6 @@ static int kvm_psci_1_x_call(struct kvm_vcpu *vcpu, u32 minor)
 
 static int kvm_psci_0_1_call(struct kvm_vcpu *vcpu)
 {
-	struct kvm *kvm = vcpu->kvm;
 	u32 psci_fn = smccc_get_function(vcpu);
 	unsigned long val;
 
@@ -405,9 +411,7 @@ static int kvm_psci_0_1_call(struct kvm_vcpu *vcpu)
 		val = PSCI_RET_SUCCESS;
 		break;
 	case KVM_PSCI_FN_CPU_ON:
-		mutex_lock(&kvm->lock);
 		val = kvm_psci_vcpu_on(vcpu);
-		mutex_unlock(&kvm->lock);
 		break;
 	default:
 		val = PSCI_RET_NOT_SUPPORTED;