In Hyper-V, the fast guest->host notification mechanism is the
SIGNAL_EVENT hypercall, with a single parameter of the connection ID to
signal.
Currently this hypercall incurs a user exit and requires the userspace
to decode the parameters and trigger the notification of the potentially
different I/O context.
To avoid the costly user exit, process this hypercall and signal the
corresponding eventfd in KVM, similar to ioeventfd. The association
between the connection id and the eventfd is established via the newly
introduced KVM_HYPERV_EVENTFD ioctl, and maintained in an
(srcu-protected) IDR.
Signed-off-by: Roman Kagan <rka...@virtuozzo.com>
Reviewed-by: David Hildenbrand <da...@redhat.com>
[asm/hyperv.h changes approved by KY Srinivasan. - Radim]
Signed-off-by: Radim Krčmář <rkrc...@redhat.com>
(cherry picked from commit faeb7833eee0d6afe0ecb6bdfa6042556c2c352e)
[rkagan: adjusted for vzkernel]
---
Documentation/virtual/kvm/api.txt | 32 +++++++++++
arch/x86/include/asm/kvm_host.h | 2 +
arch/x86/include/uapi/asm/hyperv.h | 2 +
arch/x86/kvm/hyperv.h | 1 +
include/uapi/linux/kvm.h | 14 +++++
arch/x86/kvm/hyperv.c | 110 ++++++++++++++++++++++++++++++++++++-
arch/x86/kvm/x86.c | 10 ++++
7 files changed, 170 insertions(+), 1 deletion(-)
diff --git a/Documentation/virtual/kvm/api.txt
b/Documentation/virtual/kvm/api.txt
index 3b146b1fea9a..453e6fd68b24 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -2690,6 +2690,38 @@ an implementation for these despite the in kernel
acceleration.
This capability is always enabled.
+4.113 KVM_HYPERV_EVENTFD
+
+Capability: KVM_CAP_HYPERV_EVENTFD
+Architectures: x86
+Type: vm ioctl
+Parameters: struct kvm_hyperv_eventfd (in)
+
+This ioctl (un)registers an eventfd to receive notifications from the guest on
+the specified Hyper-V connection id through the SIGNAL_EVENT hypercall, without
+causing a user exit. SIGNAL_EVENT hypercall with non-zero event flag number
+(bits 24-31) still triggers a KVM_EXIT_HYPERV_HCALL user exit.
+
+struct kvm_hyperv_eventfd {
+ __u32 conn_id;
+ __s32 fd;
+ __u32 flags;
+ __u32 padding[3];
+};
+
+The conn_id field should fit within 24 bits:
+
+#define KVM_HYPERV_CONN_ID_MASK 0x00ffffff
+
+The acceptable values for the flags field are:
+
+#define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0)
+
+Returns: 0 on success,
+ -EINVAL if conn_id or flags is outside the allowed range
+ -ENOENT on deassign if the conn_id isn't registered
+ -EEXIST on assign if the conn_id is already registered
+
5. The kvm_run structure
------------------------
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index bcf17448dcab..838996db6973 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -678,6 +678,8 @@ struct kvm_hv {
u64 hv_crash_ctl;
HV_REFERENCE_TSC_PAGE tsc_ref;
+
+ struct idr conn_to_evt;
};
struct kvm_arch {
diff --git a/arch/x86/include/uapi/asm/hyperv.h
b/arch/x86/include/uapi/asm/hyperv.h
index b142c019514a..d801a6c7996c 100644
--- a/arch/x86/include/uapi/asm/hyperv.h
+++ b/arch/x86/include/uapi/asm/hyperv.h
@@ -263,7 +263,9 @@
#define HV_STATUS_INVALID_HYPERCALL_CODE 2
#define HV_STATUS_INVALID_HYPERCALL_INPUT 3
#define HV_STATUS_INVALID_ALIGNMENT 4
+#define HV_STATUS_INVALID_PARAMETER 5
#define HV_STATUS_INSUFFICIENT_MEMORY 11
+#define HV_STATUS_INVALID_PORT_ID 17
#define HV_STATUS_INVALID_CONNECTION_ID 18
#define HV_STATUS_INSUFFICIENT_BUFFERS 19
diff --git a/arch/x86/kvm/hyperv.h b/arch/x86/kvm/hyperv.h
index cc2468244ca2..837465d69c6d 100644
--- a/arch/x86/kvm/hyperv.h
+++ b/arch/x86/kvm/hyperv.h
@@ -90,5 +90,6 @@ void kvm_hv_setup_tsc_page(struct kvm *kvm,
void kvm_hv_init_vm(struct kvm *kvm);
void kvm_hv_destroy_vm(struct kvm *kvm);
+int kvm_vm_ioctl_hv_eventfd(struct kvm *kvm, struct kvm_hyperv_eventfd *args);
#endif
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index ca4c81cc7a6f..890c2cb7d0fa 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -729,6 +729,7 @@ struct kvm_ppc_smmu_info {
#define KVM_CAP_X2APIC_API 129
#define KVM_CAP_HYPERV_SYNIC2 148
#define KVM_CAP_HYPERV_VP_INDEX 149
+#define KVM_CAP_HYPERV_EVENTFD 154
#ifdef KVM_CAP_IRQ_ROUTING
@@ -1117,6 +1118,9 @@ struct kvm_s390_ucas_mapping {
/* Available with KVM_CAP_X86_SMM */
#define KVM_SMI _IO(KVMIO, 0xb7)
+/* Available with KVM_CAP_HYPERV_EVENTFD */
+#define KVM_HYPERV_EVENTFD _IOW(KVMIO, 0xbd, struct kvm_hyperv_eventfd)
+
#define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0)
#define KVM_DEV_ASSIGN_PCI_2_3 (1 << 1)
#define KVM_DEV_ASSIGN_MASK_INTX (1 << 2)
@@ -1170,4 +1174,14 @@ struct kvm_assigned_msix_entry {
#define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0)
#define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1)
+struct kvm_hyperv_eventfd {
+ __u32 conn_id;
+ __s32 fd;
+ __u32 flags;
+ __u32 padding[3];
+};
+
+#define KVM_HYPERV_CONN_ID_MASK 0x00ffffff
+#define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0)
+
#endif /* __LINUX_KVM_H */
diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
index 692ae19d2d72..5eed6593a005 100644
--- a/arch/x86/kvm/hyperv.c
+++ b/arch/x86/kvm/hyperv.c
@@ -28,6 +28,8 @@
#include <linux/kvm_host.h>
#include <linux/highmem.h>
+#include <linux/eventfd.h>
+
#include <asm/apicdef.h>
#include <trace/events/kvm.h>
@@ -1219,6 +1221,43 @@ static int kvm_hv_hypercall_complete_userspace(struct
kvm_vcpu *vcpu)
return 1;
}
+static u16 kvm_hvcall_signal_event(struct kvm_vcpu *vcpu, bool fast, u64 param)
+{
+ struct eventfd_ctx *eventfd;
+
+ if (unlikely(!fast)) {
+ int ret;
+ gpa_t gpa = param;
+
+ if ((gpa & (__alignof__(param) - 1)) ||
+ offset_in_page(gpa) + sizeof(param) > PAGE_SIZE)
+ return HV_STATUS_INVALID_ALIGNMENT;
+
+ ret = kvm_vcpu_read_guest(vcpu, gpa, ¶m, sizeof(param));
+ if (ret < 0)
+ return HV_STATUS_INVALID_ALIGNMENT;
+ }
+
+ /*
+ * Per spec, bits 32-47 contain the extra "flag number". However, we
+ * have no use for it, and in all known usecases it is zero, so just
+ * report lookup failure if it isn't.
+ */
+ if (param & 0xffff00000000ULL)
+ return HV_STATUS_INVALID_PORT_ID;
+ /* remaining bits are reserved-zero */
+ if (param & ~KVM_HYPERV_CONN_ID_MASK)
+ return HV_STATUS_INVALID_HYPERCALL_INPUT;
+
+ /* conn_to_evt is protected by vcpu->kvm->srcu */
+ eventfd = idr_find(&vcpu->kvm->arch.hyperv.conn_to_evt, param);
+ if (!eventfd)
+ return HV_STATUS_INVALID_PORT_ID;
+
+ eventfd_signal(eventfd, 1);
+ return HV_STATUS_SUCCESS;
+}
+
int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
{
u64 param, ingpa, outgpa, ret;
@@ -1269,8 +1308,17 @@ int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
case HVCALL_NOTIFY_LONG_SPIN_WAIT:
kvm_vcpu_on_spin(vcpu);
break;
- case HVCALL_POST_MESSAGE:
case HVCALL_SIGNAL_EVENT:
+ res = kvm_hvcall_signal_event(vcpu, fast, ingpa);
+ if (res != HV_STATUS_INVALID_PORT_ID)
+ break;
+ /* maybe userspace knows this conn_id: fall through */
+ case HVCALL_POST_MESSAGE:
+ /* don't bother userspace if it has no way to handle it */
+ if (!vcpu_to_synic(vcpu)->active) {
+ res = HV_STATUS_INVALID_HYPERCALL_CODE;
+ break;
+ }
vcpu->run->exit_reason = KVM_EXIT_HYPERV;
vcpu->run->hyperv.type = KVM_EXIT_HYPERV_HCALL;
vcpu->run->hyperv.u.hcall.input = param;
@@ -1293,8 +1341,68 @@ set_result:
void kvm_hv_init_vm(struct kvm *kvm)
{
mutex_init(&kvm->arch.hyperv.hv_lock);
+ idr_init(&kvm->arch.hyperv.conn_to_evt);
}
void kvm_hv_destroy_vm(struct kvm *kvm)
{
+ struct eventfd_ctx *eventfd;
+ int i;
+
+ idr_for_each_entry(&kvm->arch.hyperv.conn_to_evt, eventfd, i)
+ eventfd_ctx_put(eventfd);
+ idr_destroy(&kvm->arch.hyperv.conn_to_evt);
+}
+
+static int kvm_hv_eventfd_assign(struct kvm *kvm, u32 conn_id, int fd)
+{
+ struct kvm_hv *hv = &kvm->arch.hyperv;
+ struct eventfd_ctx *eventfd;
+ int ret;
+
+ eventfd = eventfd_ctx_fdget(fd);
+ if (IS_ERR(eventfd))
+ return PTR_ERR(eventfd);
+
+ mutex_lock(&hv->hv_lock);
+ ret = idr_alloc(&hv->conn_to_evt, eventfd, conn_id, conn_id + 1,
+ GFP_KERNEL);
+ mutex_unlock(&hv->hv_lock);
+
+ if (ret >= 0)
+ return 0;
+
+ if (ret == -ENOSPC)
+ ret = -EEXIST;
+ eventfd_ctx_put(eventfd);
+ return ret;
+}
+
+static int kvm_hv_eventfd_deassign(struct kvm *kvm, u32 conn_id)
+{
+ struct kvm_hv *hv = &kvm->arch.hyperv;
+ struct eventfd_ctx *eventfd;
+
+ mutex_lock(&hv->hv_lock);
+ eventfd = idr_find(&hv->conn_to_evt, conn_id);
+ idr_remove(&hv->conn_to_evt, conn_id);
+ mutex_unlock(&hv->hv_lock);
+
+ if (!eventfd)
+ return -ENOENT;
+
+ synchronize_srcu(&kvm->srcu);
+ eventfd_ctx_put(eventfd);
+ return 0;
+}
+
+int kvm_vm_ioctl_hv_eventfd(struct kvm *kvm, struct kvm_hyperv_eventfd *args)
+{
+ if ((args->flags & ~KVM_HYPERV_EVENTFD_DEASSIGN) ||
+ (args->conn_id & ~KVM_HYPERV_CONN_ID_MASK))
+ return -EINVAL;
+
+ if (args->flags == KVM_HYPERV_EVENTFD_DEASSIGN)
+ return kvm_hv_eventfd_deassign(kvm, args->conn_id);
+ return kvm_hv_eventfd_assign(kvm, args->conn_id, args->fd);
}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 31d84499a3da..24e84a4809af 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2615,6 +2615,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long
ext)
case KVM_CAP_HYPERV_SYNIC:
case KVM_CAP_HYPERV_SYNIC2:
case KVM_CAP_HYPERV_VP_INDEX:
+ case KVM_CAP_HYPERV_EVENTFD:
case KVM_CAP_PCI_SEGMENT:
case KVM_CAP_DEBUGREGS:
case KVM_CAP_X86_ROBUST_SINGLESTEP:
@@ -4123,6 +4124,15 @@ long kvm_arch_vm_ioctl(struct file *filp,
r = kvm_vm_ioctl_enable_cap(kvm, &cap);
break;
}
+ case KVM_HYPERV_EVENTFD: {
+ struct kvm_hyperv_eventfd hvevfd;
+
+ r = -EFAULT;
+ if (copy_from_user(&hvevfd, argp, sizeof(hvevfd)))
+ goto out;
+ r = kvm_vm_ioctl_hv_eventfd(kvm, &hvevfd);
+ break;
+ }
default:
r = kvm_vm_ioctl_assigned_device(kvm, ioctl, arg);
}
--
2.14.3
_______________________________________________
Devel mailing list
Devel@openvz.org
https://lists.openvz.org/mailman/listinfo/devel
