1 files changed, 83 insertions, 36 deletions

diff --git a/arch/arm64/kvm/vgic/vgic.c b/arch/arm64/kvm/vgic/vgic.c
index 4ec93587c8cd..8f8096d48925 100644
--- a/arch/arm64/kvm/vgic/vgic.c
+++ b/arch/arm64/kvm/vgic/vgic.c
@@ -29,14 +29,18 @@ struct vgic_global kvm_vgic_global_state __ro_after_init = {
  *       its->cmd_lock (mutex)
  *         its->its_lock (mutex)
  *           vgic_cpu->ap_list_lock		must be taken with IRQs disabled
- *             kvm->lpi_list_lock		must be taken with IRQs disabled
- *               vgic_dist->lpi_xa.xa_lock	must be taken with IRQs disabled
- *                 vgic_irq->irq_lock		must be taken with IRQs disabled
+ *             vgic_dist->lpi_xa.xa_lock	must be taken with IRQs disabled
+ *               vgic_irq->irq_lock		must be taken with IRQs disabled
  *
  * As the ap_list_lock might be taken from the timer interrupt handler,
  * we have to disable IRQs before taking this lock and everything lower
  * than it.
  *
+ * The config_lock has additional ordering requirements:
+ * kvm->slots_lock
+ *   kvm->srcu
+ *     kvm->arch.config_lock
+ *
  * If you need to take multiple locks, always take the upper lock first,
  * then the lower ones, e.g. first take the its_lock, then the irq_lock.
  * If you are already holding a lock and need to take a higher one, you
@@ -80,17 +84,11 @@ static struct vgic_irq *vgic_get_lpi(struct kvm *kvm, u32 intid)
  * struct vgic_irq. It also increases the refcount, so any caller is expected
  * to call vgic_put_irq() once it's finished with this IRQ.
  */
-struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
-			      u32 intid)
+struct vgic_irq *vgic_get_irq(struct kvm *kvm, u32 intid)
 {
-	/* SGIs and PPIs */
-	if (intid <= VGIC_MAX_PRIVATE) {
-		intid = array_index_nospec(intid, VGIC_MAX_PRIVATE + 1);
-		return &vcpu->arch.vgic_cpu.private_irqs[intid];
-	}
-
 	/* SPIs */
-	if (intid < (kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS)) {
+	if (intid >= VGIC_NR_PRIVATE_IRQS &&
+	    intid < (kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS)) {
 		intid = array_index_nospec(intid, kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS);
 		return &kvm->arch.vgic.spis[intid - VGIC_NR_PRIVATE_IRQS];
 	}
@@ -102,6 +100,20 @@ struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 	return NULL;
 }
 
+struct vgic_irq *vgic_get_vcpu_irq(struct kvm_vcpu *vcpu, u32 intid)
+{
+	if (WARN_ON(!vcpu))
+		return NULL;
+
+	/* SGIs and PPIs */
+	if (intid < VGIC_NR_PRIVATE_IRQS) {
+		intid = array_index_nospec(intid, VGIC_NR_PRIVATE_IRQS);
+		return &vcpu->arch.vgic_cpu.private_irqs[intid];
+	}
+
+	return vgic_get_irq(vcpu->kvm, intid);
+}
+
 /*
  * We can't do anything in here, because we lack the kvm pointer to
  * lock and remove the item from the lpi_list. So we keep this function
@@ -126,7 +138,6 @@ void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq)
 	__xa_erase(&dist->lpi_xa, irq->intid);
 	xa_unlock_irqrestore(&dist->lpi_xa, flags);
 
-	atomic_dec(&dist->lpi_count);
 	kfree_rcu(irq, rcu);
 }
 
@@ -315,7 +326,7 @@ static bool vgic_validate_injection(struct vgic_irq *irq, bool level, void *owne
  * with all locks dropped.
  */
 bool vgic_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq,
-			   unsigned long flags)
+			   unsigned long flags) __releases(&irq->irq_lock)
 {
 	struct kvm_vcpu *vcpu;
 
@@ -434,7 +445,10 @@ int kvm_vgic_inject_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 
 	trace_vgic_update_irq_pending(vcpu ? vcpu->vcpu_idx : 0, intid, level);
 
-	irq = vgic_get_irq(kvm, vcpu, intid);
+	if (intid < VGIC_NR_PRIVATE_IRQS)
+		irq = vgic_get_vcpu_irq(vcpu, intid);
+	else
+		irq = vgic_get_irq(kvm, intid);
 	if (!irq)
 		return -EINVAL;
 
@@ -496,7 +510,7 @@ static inline void kvm_vgic_unmap_irq(struct vgic_irq *irq)
 int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
 			  u32 vintid, struct irq_ops *ops)
 {
-	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
+	struct vgic_irq *irq = vgic_get_vcpu_irq(vcpu, vintid);
 	unsigned long flags;
 	int ret;
 
@@ -521,7 +535,7 @@ int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
  */
 void kvm_vgic_reset_mapped_irq(struct kvm_vcpu *vcpu, u32 vintid)
 {
-	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
+	struct vgic_irq *irq = vgic_get_vcpu_irq(vcpu, vintid);
 	unsigned long flags;
 
 	if (!irq->hw)
@@ -544,7 +558,7 @@ int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int vintid)
 	if (!vgic_initialized(vcpu->kvm))
 		return -EAGAIN;
 
-	irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
+	irq = vgic_get_vcpu_irq(vcpu, vintid);
 	BUG_ON(!irq);
 
 	raw_spin_lock_irqsave(&irq->irq_lock, flags);
@@ -557,7 +571,7 @@ int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int vintid)
 
 int kvm_vgic_get_map(struct kvm_vcpu *vcpu, unsigned int vintid)
 {
-	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
+	struct vgic_irq *irq = vgic_get_vcpu_irq(vcpu, vintid);
 	unsigned long flags;
 	int ret = -1;
 
@@ -593,7 +607,7 @@ int kvm_vgic_set_owner(struct kvm_vcpu *vcpu, unsigned int intid, void *owner)
 	if (!irq_is_ppi(intid) && !vgic_valid_spi(vcpu->kvm, intid))
 		return -EINVAL;
 
-	irq = vgic_get_irq(vcpu->kvm, vcpu, intid);
+	irq = vgic_get_vcpu_irq(vcpu, intid);
 	raw_spin_lock_irqsave(&irq->irq_lock, flags);
 	if (irq->owner && irq->owner != owner)
 		ret = -EEXIST;
@@ -858,6 +872,15 @@ void kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu)
 {
 	int used_lrs;
 
+	/* If nesting, emulate the HW effect from L0 to L1 */
+	if (vgic_state_is_nested(vcpu)) {
+		vgic_v3_sync_nested(vcpu);
+		return;
+	}
+
+	if (vcpu_has_nv(vcpu))
+		vgic_v3_nested_update_mi(vcpu);
+
 	/* An empty ap_list_head implies used_lrs == 0 */
 	if (list_empty(&vcpu->arch.vgic_cpu.ap_list_head))
 		return;
@@ -887,6 +910,35 @@ static inline void vgic_restore_state(struct kvm_vcpu *vcpu)
 void kvm_vgic_flush_hwstate(struct kvm_vcpu *vcpu)
 {
 	/*
+	 * If in a nested state, we must return early. Two possibilities:
+	 *
+	 * - If we have any pending IRQ for the guest and the guest
+	 *   expects IRQs to be handled in its virtual EL2 mode (the
+	 *   virtual IMO bit is set) and it is not already running in
+	 *   virtual EL2 mode, then we have to emulate an IRQ
+	 *   exception to virtual EL2.
+	 *
+	 *   We do that by placing a request to ourselves which will
+	 *   abort the entry procedure and inject the exception at the
+	 *   beginning of the run loop.
+	 *
+	 * - Otherwise, do exactly *NOTHING*. The guest state is
+	 *   already loaded, and we can carry on with running it.
+	 *
+	 * If we have NV, but are not in a nested state, compute the
+	 * maintenance interrupt state, as it may fire.
+	 */
+	if (vgic_state_is_nested(vcpu)) {
+		if (kvm_vgic_vcpu_pending_irq(vcpu))
+			kvm_make_request(KVM_REQ_GUEST_HYP_IRQ_PENDING, vcpu);
+
+		return;
+	}
+
+	if (vcpu_has_nv(vcpu))
+		vgic_v3_nested_update_mi(vcpu);
+
+	/*
 	 * If there are no virtual interrupts active or pending for this
 	 * VCPU, then there is no work to do and we can bail out without
 	 * taking any lock.  There is a potential race with someone injecting
@@ -919,10 +971,13 @@ void kvm_vgic_flush_hwstate(struct kvm_vcpu *vcpu)
 
 void kvm_vgic_load(struct kvm_vcpu *vcpu)
 {
-	if (unlikely(!vgic_initialized(vcpu->kvm)))
+	if (unlikely(!irqchip_in_kernel(vcpu->kvm) || !vgic_initialized(vcpu->kvm))) {
+		if (has_vhe() && static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
+			__vgic_v3_activate_traps(&vcpu->arch.vgic_cpu.vgic_v3);
 		return;
+	}
 
-	if (kvm_vgic_global_state.type == VGIC_V2)
+	if (!static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
 		vgic_v2_load(vcpu);
 	else
 		vgic_v3_load(vcpu);
@@ -930,26 +985,18 @@ void kvm_vgic_load(struct kvm_vcpu *vcpu)
 
 void kvm_vgic_put(struct kvm_vcpu *vcpu)
 {
-	if (unlikely(!vgic_initialized(vcpu->kvm)))
+	if (unlikely(!irqchip_in_kernel(vcpu->kvm) || !vgic_initialized(vcpu->kvm))) {
+		if (has_vhe() && static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
+			__vgic_v3_deactivate_traps(&vcpu->arch.vgic_cpu.vgic_v3);
 		return;
+	}
 
-	if (kvm_vgic_global_state.type == VGIC_V2)
+	if (!static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
 		vgic_v2_put(vcpu);
 	else
 		vgic_v3_put(vcpu);
 }
 
-void kvm_vgic_vmcr_sync(struct kvm_vcpu *vcpu)
-{
-	if (unlikely(!irqchip_in_kernel(vcpu->kvm)))
-		return;
-
-	if (kvm_vgic_global_state.type == VGIC_V2)
-		vgic_v2_vmcr_sync(vcpu);
-	else
-		vgic_v3_vmcr_sync(vcpu);
-}
-
 int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu)
 {
 	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
@@ -1010,7 +1057,7 @@ bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, unsigned int vintid)
 	if (!vgic_initialized(vcpu->kvm))
 		return false;
 
-	irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
+	irq = vgic_get_vcpu_irq(vcpu, vintid);
 	raw_spin_lock_irqsave(&irq->irq_lock, flags);
 	map_is_active = irq->hw && irq->active;
 	raw_spin_unlock_irqrestore(&irq->irq_lock, flags);