From 8f624361f108bf676a36a96efe9a7b35e1ce2e66 Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Wed, 9 Dec 2009 01:54:10 -0200
Subject: [PATCH] KVM: x86: Add KVM_GET/SET_VCPU_EVENTS

RH-Author: Marcelo Tosatti <mtosatti@redhat.com>
Message-id: <20091209015410.GA11364@amt.cnet>
Patchwork-id: 3879
O-Subject: [RHEL 5.5 PATCH] KVM: x86: Add KVM_GET/SET_VCPU_EVENTS
Bugzilla: 541084
RH-Acked-by: Glauber Costa <glommer@redhat.com>
RH-Acked-by: Juan Quintela <quintela@redhat.com>
RH-Acked-by: Gleb Natapov <gleb@redhat.com>

commit 74e2733179bd1c32a48830c5a9a74344b26bb9ae
Author: Jan Kiszka <jan.kiszka@web.de>
Date:   Thu Nov 12 01:04:25 2009 +0100

KVM: x86: Add KVM_GET/SET_VCPU_EVENTS

This new IOCTL exports all yet user-invisible states related to
exceptions, interrupts, and NMIs. Together with appropriate user space
changes, this fixes sporadic problems of vmsave/restore, live migration
and system reset.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Bugzilla: 541084

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
 arch/x86/include/asm/kvm.h      |   27 ++++++++++++++
 arch/x86/include/asm/kvm_host.h |    2 +
 arch/x86/kvm/svm.c              |   23 +++++++++++-
 arch/x86/kvm/vmx.c              |   30 +++++++++++++++
 arch/x86/kvm/x86.c              |   76 +++++++++++++++++++++++++++++++++++++++
 include/linux/kvm.h             |    5 +++
 6 files changed, 162 insertions(+), 1 deletions(-)

diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h
index cf24d4b..8fa0064 100644
--- a/arch/x86/include/asm/kvm.h
+++ b/arch/x86/include/asm/kvm.h
@@ -213,4 +213,31 @@ struct kvm_reinject_control {
 	__u8 pit_reinject;
 	__u8 reserved[31];
 };
+
+/* for KVM_GET/SET_VCPU_EVENTS */
+struct kvm_vcpu_events {
+	struct {
+		__u8 injected;
+		__u8 nr;
+		__u8 has_error_code;
+		__u8 pad;
+		__u32 error_code;
+	} exception;
+	struct {
+		__u8 injected;
+		__u8 nr;
+		__u8 soft;
+		__u8 pad;
+	} interrupt;
+	struct {
+		__u8 injected;
+		__u8 pending;
+		__u8 masked;
+		__u8 pad;
+	} nmi;
+	__u32 sipi_vector;
+	__u32 flags;
+	__u32 reserved[10];
+};
+
 #endif /* _ASM_X86_KVM_H */
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 35dd13b..7b58e15 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -495,6 +495,8 @@ struct kvm_x86_ops {
 	void (*inject_pending_vectors)(struct kvm_vcpu *vcpu,
 				       struct kvm_run *run);
 	int (*interrupt_allowed)(struct kvm_vcpu *vcpu);
+	bool (*get_nmi_mask)(struct kvm_vcpu *vcpu);
+	void (*set_nmi_mask)(struct kvm_vcpu *vcpu, bool masked);
 	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
 	int (*get_tdp_level)(void);
 	int (*get_mt_mask_shift)(void);
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 3489142..6364a60 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -1647,6 +1647,26 @@ static void update_cr8_intercept(struct kvm_vcpu *vcpu)
 		vmcb->control.intercept_cr_write |= INTERCEPT_CR8_MASK;
 }
 
+static bool svm_get_nmi_mask(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	return !!(svm->vcpu.arch.hflags & HF_NMI_MASK);
+}
+
+static void svm_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	if (masked) {
+		svm->vcpu.arch.hflags |= HF_NMI_MASK;
+		svm->vmcb->control.intercept |= (1UL << INTERCEPT_IRET);
+	} else {
+		svm->vcpu.arch.hflags &= ~HF_NMI_MASK;
+		svm->vmcb->control.intercept &= ~(1UL << INTERCEPT_IRET);
+	}
+}
+
 static int svm_interrupt_allowed(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
@@ -2090,7 +2110,8 @@ static struct kvm_x86_ops svm_x86_ops = {
 	.inject_pending_irq = svm_intr_assist,
 	.inject_pending_vectors = do_interrupt_requests,
 	.interrupt_allowed = svm_interrupt_allowed,
-
+	.get_nmi_mask = svm_get_nmi_mask,
+	.set_nmi_mask = svm_set_nmi_mask,
 	.set_tss_addr = svm_set_tss_addr,
 	.get_tdp_level = get_npt_level,
 	.get_mt_mask_shift = svm_get_mt_mask_shift,
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 8c45cdb..b1832e4 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -2523,6 +2523,34 @@ static int vmx_interrupt_allowed(struct kvm_vcpu *vcpu)
 	return vcpu->arch.interrupt_window_open;
 }
 
+static bool vmx_get_nmi_mask(struct kvm_vcpu *vcpu)
+{
+	if (!cpu_has_virtual_nmis())
+		return to_vmx(vcpu)->soft_vnmi_blocked;
+	else
+		return !!(vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) &
+			  GUEST_INTR_STATE_NMI);
+}
+
+static void vmx_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (!cpu_has_virtual_nmis()) {
+		if (vmx->soft_vnmi_blocked != masked) {
+			vmx->soft_vnmi_blocked = masked;
+			vmx->vnmi_blocked_time = 0;
+		}
+	} else {
+		if (masked)
+			vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO,
+				      GUEST_INTR_STATE_NMI);
+		else
+			vmcs_clear_bits(GUEST_INTERRUPTIBILITY_INFO,
+					GUEST_INTR_STATE_NMI);
+	}
+}
+
 static void do_interrupt_requests(struct kvm_vcpu *vcpu,
 				       struct kvm_run *kvm_run)
 {
@@ -3672,6 +3700,8 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.inject_pending_irq = vmx_intr_assist,
 	.inject_pending_vectors = do_interrupt_requests,
 	.interrupt_allowed = vmx_interrupt_allowed,
+	.get_nmi_mask = vmx_get_nmi_mask,
+	.set_nmi_mask = vmx_set_nmi_mask,
 	.set_tss_addr = vmx_set_tss_addr,
 	.get_tdp_level = get_ept_level,
 	.get_mt_mask_shift = vmx_get_mt_mask_shift,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 6f5e7e8..3ba88a2 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1069,6 +1069,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_IRQ_INJECT_STATUS:
 	case KVM_CAP_ASSIGN_DEV_IRQ:
  	case KVM_CAP_ADJUST_CLOCK:
+	case KVM_CAP_VCPU_EVENTS:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
@@ -1498,6 +1499,60 @@ static int vcpu_ioctl_tpr_access_reporting(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
+static void kvm_vcpu_ioctl_x86_get_vcpu_events(struct kvm_vcpu *vcpu,
+					       struct kvm_vcpu_events *events)
+{
+	vcpu_load(vcpu);
+
+	events->exception.injected = vcpu->arch.exception.pending;
+	events->exception.nr = vcpu->arch.exception.nr;
+	events->exception.has_error_code = vcpu->arch.exception.has_error_code;
+	events->exception.error_code = vcpu->arch.exception.error_code;
+
+	events->interrupt.injected = vcpu->arch.interrupt.pending;
+	events->interrupt.nr = vcpu->arch.interrupt.nr;
+	events->interrupt.soft = false;
+
+	events->nmi.injected = vcpu->arch.nmi_injected;
+	events->nmi.pending = vcpu->arch.nmi_pending;
+	events->nmi.masked = kvm_x86_ops->get_nmi_mask(vcpu);
+
+	events->sipi_vector = vcpu->arch.sipi_vector;
+
+	events->flags = 0;
+
+	vcpu_put(vcpu);
+}
+
+static int kvm_vcpu_ioctl_x86_set_vcpu_events(struct kvm_vcpu *vcpu,
+					      struct kvm_vcpu_events *events)
+{
+	if (events->flags)
+		return -EINVAL;
+
+	vcpu_load(vcpu);
+
+	vcpu->arch.exception.pending = events->exception.injected;
+	vcpu->arch.exception.nr = events->exception.nr;
+	vcpu->arch.exception.has_error_code = events->exception.has_error_code;
+	vcpu->arch.exception.error_code = events->exception.error_code;
+
+	vcpu->arch.interrupt.pending = events->interrupt.injected;
+	vcpu->arch.interrupt.nr = events->interrupt.nr;
+	if (vcpu->arch.interrupt.pending && irqchip_in_kernel(vcpu->kvm))
+		kvm_pic_clear_isr_ack(vcpu->kvm);
+
+	vcpu->arch.nmi_injected = events->nmi.injected;
+	vcpu->arch.nmi_pending = events->nmi.pending;
+	kvm_x86_ops->set_nmi_mask(vcpu, events->nmi.masked);
+
+	vcpu->arch.sipi_vector = events->sipi_vector;
+
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
 long kvm_arch_vcpu_ioctl(struct file *filp,
 			 unsigned int ioctl, unsigned long arg)
 {
@@ -1631,6 +1686,27 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 		kvm_lapic_set_vapic_addr(vcpu, va.vapic_addr);
 		break;
 	}
+	case KVM_GET_VCPU_EVENTS: {
+		struct kvm_vcpu_events events;
+
+		kvm_vcpu_ioctl_x86_get_vcpu_events(vcpu, &events);
+
+		r = -EFAULT;
+		if (copy_to_user(argp, &events, sizeof(struct kvm_vcpu_events)))
+			break;
+		r = 0;
+		break;
+	}
+	case KVM_SET_VCPU_EVENTS: {
+		struct kvm_vcpu_events events;
+
+		r = -EFAULT;
+		if (copy_from_user(&events, argp, sizeof(struct kvm_vcpu_events)))
+			break;
+
+		r = kvm_vcpu_ioctl_x86_set_vcpu_events(vcpu, &events);
+		break;
+	}
 	default:
 		r = -EINVAL;
 	}
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 56ef826..8f0def3 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -416,6 +416,7 @@ struct kvm_trace_rec {
 /* Another bug in KVM_SET_USER_MEMORY_REGION fixed: */
 #define KVM_CAP_JOIN_MEMORY_REGIONS_WORKS 30
 #define KVM_CAP_ADJUST_CLOCK 39
+#define KVM_CAP_VCPU_EVENTS 41
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
@@ -549,6 +550,10 @@ struct kvm_clock_data {
 #define KVM_SET_MP_STATE          _IOW(KVMIO,  0x99, struct kvm_mp_state)
 /* Available with KVM_CAP_NMI */
 #define KVM_NMI                   _IO(KVMIO,  0x9a)
+/* Available with KVM_CAP_VCPU_EVENTS */
+#define KVM_GET_VCPU_EVENTS       _IOR(KVMIO,  0x9f, struct kvm_vcpu_events)
+#define KVM_SET_VCPU_EVENTS       _IOW(KVMIO,  0xa0, struct kvm_vcpu_events)
+
 
 #define KVM_TRC_INJ_VIRQ         (KVM_TRC_HANDLER + 0x02)
 #define KVM_TRC_REDELIVER_EVT    (KVM_TRC_HANDLER + 0x03)
-- 
1.6.3.rc4.29.g8146