KVM: irqfd

KVM provides a complete virtual system environment for guests, including
support for injecting interrupts modeled after the real exception/interrupt
facilities present on the native platform (such as the IDT on x86).
Virtual interrupts can come from a variety of sources (emulated devices,
pass-through devices, etc) but all must be injected to the guest via
the KVM infrastructure.  This patch adds a new mechanism to inject a specific
interrupt to a guest using a decoupled eventfd mechnanism:  Any legal signal
on the irqfd (using eventfd semantics from either userspace or kernel) will
translate into an injected interrupt in the guest at the next available
interrupt window.

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Signed-off-by: Avi Kivity <avi@redhat.com>

arch/x86/kvm/Kconfig

@@ -26,6 +26,7 @@ config KVM
 	select MMU_NOTIFIER
 	select ANON_INODES
 	select HAVE_KVM_IRQCHIP
+	select HAVE_KVM_EVENTFD
 	---help---
 	  Support hosting fully virtualized guest machines using hardware
 	  virtualization extensions.  You will need a fairly recent

arch/x86/kvm/Makefile

@@ -2,7 +2,7 @@
 EXTRA_CFLAGS += -Ivirt/kvm -Iarch/x86/kvm
 
 kvm-y			+= $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \
-				coalesced_mmio.o irq_comm.o)
+				coalesced_mmio.o irq_comm.o eventfd.o)
 kvm-$(CONFIG_KVM_TRACE)	+= $(addprefix ../../../virt/kvm/, kvm_trace.o)
 kvm-$(CONFIG_IOMMU_API)	+= $(addprefix ../../../virt/kvm/, iommu.o)
 

arch/x86/kvm/x86.c

@@ -1126,6 +1126,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_REINJECT_CONTROL:
 	case KVM_CAP_IRQ_INJECT_STATUS:
 	case KVM_CAP_ASSIGN_DEV_IRQ:
+	case KVM_CAP_IRQFD:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:

include/linux/kvm.h

@@ -418,6 +418,7 @@ struct kvm_trace_rec {
 #ifdef __KVM_HAVE_MCE
 #define KVM_CAP_MCE 31
 #endif
+#define KVM_CAP_IRQFD 32
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
@@ -470,6 +471,15 @@ struct kvm_x86_mce {
 };
 #endif
 
+#define KVM_IRQFD_FLAG_DEASSIGN (1 << 0)
+
+struct kvm_irqfd {
+	__u32 fd;
+	__u32 gsi;
+	__u32 flags;
+	__u8  pad[20];
+};
+
 /*
  * ioctls for VM fds
  */
@@ -514,6 +524,7 @@ struct kvm_x86_mce {
 #define KVM_ASSIGN_SET_MSIX_ENTRY \
 			_IOW(KVMIO, 0x74, struct kvm_assigned_msix_entry)
 #define KVM_DEASSIGN_DEV_IRQ       _IOW(KVMIO, 0x75, struct kvm_assigned_irq)
+#define KVM_IRQFD                  _IOW(KVMIO, 0x76, struct kvm_irqfd)
 
 /*
  * ioctls for vcpu fds

include/linux/kvm_host.h

@@ -136,6 +136,12 @@ struct kvm {
 	struct list_head vm_list;
 	struct kvm_io_bus mmio_bus;
 	struct kvm_io_bus pio_bus;
+#ifdef CONFIG_HAVE_KVM_EVENTFD
+	struct {
+		spinlock_t        lock;
+		struct list_head  items;
+	} irqfds;
+#endif
 	struct kvm_vm_stat stat;
 	struct kvm_arch arch;
 	atomic_t users_count;
@@ -525,4 +531,22 @@ static inline void kvm_free_irq_routing(struct kvm *kvm) {}
 
 #endif
 
+#ifdef CONFIG_HAVE_KVM_EVENTFD
+
+void kvm_irqfd_init(struct kvm *kvm);
+int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags);
+void kvm_irqfd_release(struct kvm *kvm);
+
+#else
+
+static inline void kvm_irqfd_init(struct kvm *kvm) {}
+static inline int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags)
+{
+	return -EINVAL;
+}
+
+static inline void kvm_irqfd_release(struct kvm *kvm) {}
+
+#endif /* CONFIG_HAVE_KVM_EVENTFD */
+
 #endif

virt/kvm/Kconfig

@@ -5,3 +5,7 @@ config HAVE_KVM
 
 config HAVE_KVM_IRQCHIP
        bool
+
+config HAVE_KVM_EVENTFD
+       bool
+       select EVENTFD