KVM: Add resampling irqfds for level triggered interrupts

To emulate level triggered interrupts, add a resample option to
KVM_IRQFD.  When specified, a new resamplefd is provided that notifies
the user when the irqchip has been resampled by the VM.  This may, for
instance, indicate an EOI.  Also in this mode, posting of an interrupt
through an irqfd only asserts the interrupt.  On resampling, the
interrupt is automatically de-asserted prior to user notification.
This enables level triggered interrupts to be posted and re-enabled
from vfio with no userspace intervention.

All resampling irqfds can make use of a single irq source ID, so we
reserve a new one for this interface.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>

Documentation/virtual/kvm/api.txt

@@ -1950,6 +1950,19 @@ the guest using the specified gsi pin.  The irqfd is removed using
 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd
 and kvm_irqfd.gsi.
 
+With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify
+mechanism allowing emulation of level-triggered, irqfd-based
+interrupts.  When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an
+additional eventfd in the kvm_irqfd.resamplefd field.  When operating
+in resample mode, posting of an interrupt through kvm_irq.fd asserts
+the specified gsi in the irqchip.  When the irqchip is resampled, such
+as from an EOI, the gsi is de-asserted and the user is notifed via
+kvm_irqfd.resamplefd.  It is the user's responsibility to re-queue
+the interrupt if the device making use of it still requires service.
+Note that closing the resamplefd is not sufficient to disable the
+irqfd.  The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment
+and need not be specified with KVM_IRQFD_FLAG_DEASSIGN.
+
 4.76 KVM_PPC_ALLOCATE_HTAB
 
 Capability: KVM_CAP_PPC_ALLOC_HTAB

arch/x86/kvm/x86.c

@@ -2176,6 +2176,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_PCI_2_3:
 	case KVM_CAP_KVMCLOCK_CTRL:
 	case KVM_CAP_READONLY_MEM:
+	case KVM_CAP_IRQFD_RESAMPLE:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
@@ -6268,6 +6269,9 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 
 	/* Reserve bit 0 of irq_sources_bitmap for userspace irq source */
 	set_bit(KVM_USERSPACE_IRQ_SOURCE_ID, &kvm->arch.irq_sources_bitmap);
+	/* Reserve bit 1 of irq_sources_bitmap for irqfd-resampler */
+	set_bit(KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID,
+		&kvm->arch.irq_sources_bitmap);
 
 	raw_spin_lock_init(&kvm->arch.tsc_write_lock);
 	mutex_init(&kvm->arch.apic_map_lock);

include/linux/kvm.h

@@ -625,6 +625,7 @@ struct kvm_ppc_smmu_info {
 #ifdef __KVM_HAVE_READONLY_MEM
 #define KVM_CAP_READONLY_MEM 81
 #endif
+#define KVM_CAP_IRQFD_RESAMPLE 82
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
@@ -690,12 +691,21 @@ struct kvm_xen_hvm_config {
 #endif
 
 #define KVM_IRQFD_FLAG_DEASSIGN (1 << 0)
+/*
+ * Available with KVM_CAP_IRQFD_RESAMPLE
+ *
+ * KVM_IRQFD_FLAG_RESAMPLE indicates resamplefd is valid and specifies
+ * the irqfd to operate in resampling mode for level triggered interrupt
+ * emlation.  See Documentation/virtual/kvm/api.txt.
+ */
+#define KVM_IRQFD_FLAG_RESAMPLE (1 << 1)
 
 struct kvm_irqfd {
 	__u32 fd;
 	__u32 gsi;
 	__u32 flags;
-	__u8  pad[20];
+	__u32 resamplefd;
+	__u8  pad[16];
 };
 
 struct kvm_clock_data {

include/linux/kvm_host.h

@@ -119,7 +119,8 @@ static inline bool is_error_page(struct page *page)
 #define KVM_REQ_PMU               16
 #define KVM_REQ_PMI               17
 
-#define KVM_USERSPACE_IRQ_SOURCE_ID	0
+#define KVM_USERSPACE_IRQ_SOURCE_ID		0
+#define KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID	1
 
 struct kvm;
 struct kvm_vcpu;
@@ -343,6 +344,8 @@ struct kvm {
 	struct {
 		spinlock_t        lock;
 		struct list_head  items;
+		struct list_head  resampler_list;
+		struct mutex      resampler_lock;
 	} irqfds;
 	struct list_head ioeventfds;
 #endif

virt/kvm/eventfd.c

@@ -43,6 +43,31 @@
  * --------------------------------------------------------------------
  */
 
+/*
+ * Resampling irqfds are a special variety of irqfds used to emulate
+ * level triggered interrupts.  The interrupt is asserted on eventfd
+ * trigger.  On acknowledgement through the irq ack notifier, the
+ * interrupt is de-asserted and userspace is notified through the
+ * resamplefd.  All resamplers on the same gsi are de-asserted
+ * together, so we don't need to track the state of each individual
+ * user.  We can also therefore share the same irq source ID.
+ */
+struct _irqfd_resampler {
+	struct kvm *kvm;
+	/*
+	 * List of resampling struct _irqfd objects sharing this gsi.
+	 * RCU list modified under kvm->irqfds.resampler_lock
+	 */
+	struct list_head list;
+	struct kvm_irq_ack_notifier notifier;
+	/*
+	 * Entry in list of kvm->irqfd.resampler_list.  Use for sharing
+	 * resamplers among irqfds on the same gsi.
+	 * Accessed and modified under kvm->irqfds.resampler_lock
+	 */
+	struct list_head link;
+};
+
 struct _irqfd {
 	/* Used for MSI fast-path */
 	struct kvm *kvm;
@@ -52,6 +77,12 @@ struct _irqfd {
 	/* Used for level IRQ fast-path */
 	int gsi;
 	struct work_struct inject;
+	/* The resampler used by this irqfd (resampler-only) */
+	struct _irqfd_resampler *resampler;
+	/* Eventfd notified on resample (resampler-only) */
+	struct eventfd_ctx *resamplefd;
+	/* Entry in list of irqfds for a resampler (resampler-only) */
+	struct list_head resampler_link;
 	/* Used for setup/shutdown */
 	struct eventfd_ctx *eventfd;
 	struct list_head list;
@@ -67,8 +98,58 @@ irqfd_inject(struct work_struct *work)
 	struct _irqfd *irqfd = container_of(work, struct _irqfd, inject);
 	struct kvm *kvm = irqfd->kvm;
 
-	kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID, irqfd->gsi, 1);
-	kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID, irqfd->gsi, 0);
+	if (!irqfd->resampler) {
+		kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID, irqfd->gsi, 1);
+		kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID, irqfd->gsi, 0);
+	} else
+		kvm_set_irq(kvm, KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID,
+			    irqfd->gsi, 1);
+}
+
+/*
+ * Since resampler irqfds share an IRQ source ID, we de-assert once
+ * then notify all of the resampler irqfds using this GSI.  We can't
+ * do multiple de-asserts or we risk racing with incoming re-asserts.
+ */
+static void
+irqfd_resampler_ack(struct kvm_irq_ack_notifier *kian)
+{
+	struct _irqfd_resampler *resampler;
+	struct _irqfd *irqfd;
+
+	resampler = container_of(kian, struct _irqfd_resampler, notifier);
+
+	kvm_set_irq(resampler->kvm, KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID,
+		    resampler->notifier.gsi, 0);
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(irqfd, &resampler->list, resampler_link)
+		eventfd_signal(irqfd->resamplefd, 1);
+
+	rcu_read_unlock();
+}
+
+static void
+irqfd_resampler_shutdown(struct _irqfd *irqfd)
+{
+	struct _irqfd_resampler *resampler = irqfd->resampler;
+	struct kvm *kvm = resampler->kvm;
+
+	mutex_lock(&kvm->irqfds.resampler_lock);
+
+	list_del_rcu(&irqfd->resampler_link);
+	synchronize_rcu();
+
+	if (list_empty(&resampler->list)) {
+		list_del(&resampler->link);
+		kvm_unregister_irq_ack_notifier(kvm, &resampler->notifier);
+		kvm_set_irq(kvm, KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID,
+			    resampler->notifier.gsi, 0);
+		kfree(resampler);
+	}
+
+	mutex_unlock(&kvm->irqfds.resampler_lock);
 }
 
 /*
@@ -92,6 +173,11 @@ irqfd_shutdown(struct work_struct *work)
 	 */
 	flush_work_sync(&irqfd->inject);
 
+	if (irqfd->resampler) {
+		irqfd_resampler_shutdown(irqfd);
+		eventfd_ctx_put(irqfd->resamplefd);
+	}
+
 	/*
 	 * It is now safe to release the object's resources
 	 */
@@ -203,7 +289,7 @@ kvm_irqfd_assign(struct kvm *kvm, struct kvm_irqfd *args)
 	struct kvm_irq_routing_table *irq_rt;
 	struct _irqfd *irqfd, *tmp;
 	struct file *file = NULL;
-	struct eventfd_ctx *eventfd = NULL;
+	struct eventfd_ctx *eventfd = NULL, *resamplefd = NULL;
 	int ret;
 	unsigned int events;
 
@@ -231,6 +317,54 @@ kvm_irqfd_assign(struct kvm *kvm, struct kvm_irqfd *args)
 
 	irqfd->eventfd = eventfd;
 
+	if (args->flags & KVM_IRQFD_FLAG_RESAMPLE) {
+		struct _irqfd_resampler *resampler;
+
+		resamplefd = eventfd_ctx_fdget(args->resamplefd);
+		if (IS_ERR(resamplefd)) {
+			ret = PTR_ERR(resamplefd);
+			goto fail;
+		}
+
+		irqfd->resamplefd = resamplefd;
+		INIT_LIST_HEAD(&irqfd->resampler_link);
+
+		mutex_lock(&kvm->irqfds.resampler_lock);
+
+		list_for_each_entry(resampler,
+				    &kvm->irqfds.resampler_list, list) {
+			if (resampler->notifier.gsi == irqfd->gsi) {
+				irqfd->resampler = resampler;
+				break;
+			}
+		}
+
+		if (!irqfd->resampler) {
+			resampler = kzalloc(sizeof(*resampler), GFP_KERNEL);
+			if (!resampler) {
+				ret = -ENOMEM;
+				mutex_unlock(&kvm->irqfds.resampler_lock);
+				goto fail;
+			}
+
+			resampler->kvm = kvm;
+			INIT_LIST_HEAD(&resampler->list);
+			resampler->notifier.gsi = irqfd->gsi;
+			resampler->notifier.irq_acked = irqfd_resampler_ack;
+			INIT_LIST_HEAD(&resampler->link);
+
+			list_add(&resampler->link, &kvm->irqfds.resampler_list);
+			kvm_register_irq_ack_notifier(kvm,
+						      &resampler->notifier);
+			irqfd->resampler = resampler;
+		}
+
+		list_add_rcu(&irqfd->resampler_link, &irqfd->resampler->list);
+		synchronize_rcu();
+
+		mutex_unlock(&kvm->irqfds.resampler_lock);
+	}
+
 	/*
 	 * Install our own custom wake-up handling so we are notified via
 	 * a callback whenever someone signals the underlying eventfd
@@ -276,6 +410,12 @@ kvm_irqfd_assign(struct kvm *kvm, struct kvm_irqfd *args)
 	return 0;
 
 fail:
+	if (irqfd->resampler)
+		irqfd_resampler_shutdown(irqfd);
+
+	if (resamplefd && !IS_ERR(resamplefd))
+		eventfd_ctx_put(resamplefd);
+
 	if (eventfd && !IS_ERR(eventfd))
 		eventfd_ctx_put(eventfd);
 
@@ -291,6 +431,8 @@ kvm_eventfd_init(struct kvm *kvm)
 {
 	spin_lock_init(&kvm->irqfds.lock);
 	INIT_LIST_HEAD(&kvm->irqfds.items);
+	INIT_LIST_HEAD(&kvm->irqfds.resampler_list);
+	mutex_init(&kvm->irqfds.resampler_lock);
 	INIT_LIST_HEAD(&kvm->ioeventfds);
 }
 
@@ -340,7 +482,7 @@ kvm_irqfd_deassign(struct kvm *kvm, struct kvm_irqfd *args)
 int
 kvm_irqfd(struct kvm *kvm, struct kvm_irqfd *args)
 {
-	if (args->flags & ~KVM_IRQFD_FLAG_DEASSIGN)
+	if (args->flags & ~(KVM_IRQFD_FLAG_DEASSIGN | KVM_IRQFD_FLAG_RESAMPLE))
 		return -EINVAL;
 
 	if (args->flags & KVM_IRQFD_FLAG_DEASSIGN)

virt/kvm/irq_comm.c

@@ -228,6 +228,9 @@ int kvm_request_irq_source_id(struct kvm *kvm)
 	}
 
 	ASSERT(irq_source_id != KVM_USERSPACE_IRQ_SOURCE_ID);
+#ifdef CONFIG_X86
+	ASSERT(irq_source_id != KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID);
+#endif
 	set_bit(irq_source_id, bitmap);
 unlock:
 	mutex_unlock(&kvm->irq_lock);
@@ -238,6 +241,9 @@ int kvm_request_irq_source_id(struct kvm *kvm)
 void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id)
 {
 	ASSERT(irq_source_id != KVM_USERSPACE_IRQ_SOURCE_ID);
+#ifdef CONFIG_X86
+	ASSERT(irq_source_id != KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID);
+#endif
 
 	mutex_lock(&kvm->irq_lock);
 	if (irq_source_id < 0 ||