[PATCH -v3] KVM: Add VT-x machine check support

[Huang Ying]

From: Andi Kleen <ak@xxxxxxxxxxxxxxx>

VT-x needs an explicit MC vector intercept to handle machine checks in the 
hyper visor.

It also has a special option to catch machine checks that happen
during VT entry.

Do these interceptions and forward them to the Linux machine check
handler. Make it always look like user space is interrupted because
the machine check handler treats kernel/user space differently.

Thanks to Jiang Yunhong for help and testing.

v3: Rebased on kvm.git

v2: Handle machine checks still in interrupt off context
to avoid problems on preemptible kernels.

Signed-off-by: Andi Kleen <ak@xxxxxxxxxxxxxxx>
Signed-off-by: Huang Ying <ying.huang@xxxxxxxxx>

---
 arch/x86/include/asm/kvm_host.h     |    2 +
 arch/x86/include/asm/vmx.h          |    1 
 arch/x86/kernel/cpu/mcheck/mce_64.c |    1 
 arch/x86/kvm/vmx.c                  |   41 +++++++++++++++++++++++++++++++++---
 4 files changed, 42 insertions(+), 3 deletions(-)

--- a/arch/x86/include/asm/vmx.h
+++ b/arch/x86/include/asm/vmx.h
@@ -247,6 +247,7 @@ enum vmcs_field {
 #define EXIT_REASON_MSR_READ            31
 #define EXIT_REASON_MSR_WRITE           32
 #define EXIT_REASON_MWAIT_INSTRUCTION   36
+#define EXIT_REASON_MCE_DURING_VMENTRY	 41
 #define EXIT_REASON_TPR_BELOW_THRESHOLD 43
 #define EXIT_REASON_APIC_ACCESS         44
 #define EXIT_REASON_EPT_VIOLATION       48
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -32,6 +32,7 @@
 #include <asm/desc.h>
 #include <asm/vmx.h>
 #include <asm/virtext.h>
+#include <asm/mce.h>
 
 #define __ex(x) __kvm_handle_fault_on_reboot(x)
 
@@ -487,7 +488,7 @@ static void update_exception_bitmap(stru
 {
 	u32 eb;
 
-	eb = (1u << PF_VECTOR) | (1u << UD_VECTOR);
+	eb = (1u << PF_VECTOR) | (1u << UD_VECTOR) | (1u << MC_VECTOR);
 	if (!vcpu->fpu_active)
 		eb |= 1u << NM_VECTOR;
 	if (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) {
@@ -2605,6 +2606,29 @@ static int handle_rmode_exception(struct
 	return 0;
 }
 
+/*
+ * Trigger machine check on the host. We assume all the MSRs are already set up
+ * by the CPU and that we still run on the same CPU as the MCE occurred on.
+ * We pass a fake environment to the machine check handler because we want
+ * the guest to be always treated like user space, no matter what context
+ * it used internally.
+ */
+static void kvm_machine_check(void)
+{
+	struct pt_regs regs = {
+		.cs = 3, /* Fake ring 3 no matter what the guest ran on */
+		.flags = X86_EFLAGS_IF,
+	};
+
+	do_machine_check(&regs, 0);
+}
+
+static int handle_machine_check(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	/* already handled by vcpu_run */
+	return 1;
+}
+
 static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
@@ -2616,6 +2640,10 @@ static int handle_exception(struct kvm_v
 	vect_info = vmx->idt_vectoring_info;
 	intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
 
+	ex_no = intr_info & INTR_INFO_VECTOR_MASK;
+	if (ex_no == MC_VECTOR)
+		return handle_machine_check(vcpu, kvm_run);
+
 	if ((vect_info & VECTORING_INFO_VALID_MASK) &&
 						!is_page_fault(intr_info))
 		printk(KERN_ERR "%s: unexpected, vectoring info 0x%x "
@@ -2662,7 +2690,6 @@ static int handle_exception(struct kvm_v
 		return 1;
 	}
 
-	ex_no = intr_info & INTR_INFO_VECTOR_MASK;
 	switch (ex_no) {
 	case DB_VECTOR:
 		dr6 = vmcs_readl(EXIT_QUALIFICATION);
@@ -3189,6 +3216,7 @@ static int (*kvm_vmx_exit_handlers[])(st
 	[EXIT_REASON_WBINVD]                  = handle_wbinvd,
 	[EXIT_REASON_TASK_SWITCH]             = handle_task_switch,
 	[EXIT_REASON_EPT_VIOLATION]	      = handle_ept_violation,
+	[EXIT_REASON_MCE_DURING_VMENTRY]      = handle_machine_check,
 };
 
 static const int kvm_vmx_max_exit_handlers =
@@ -3200,7 +3228,7 @@ static const int kvm_vmx_max_exit_handle
  */
 static int vmx_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 {
-	u32 exit_reason = vmcs_read32(VM_EXIT_REASON);
+	u32 exit_reason = vcpu->arch.exit_reason;
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	u32 vectoring_info = vmx->idt_vectoring_info;
 
@@ -3284,6 +3312,13 @@ static void vmx_complete_interrupts(stru
 
 	exit_intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
 
+	vmx->vcpu.arch.exit_reason = vmcs_read32(VM_EXIT_REASON);
+
+	/* Handle machine checks before interrupts are enabled */
+	if ((vmx->vcpu.arch.exit_reason == EXIT_REASON_MCE_DURING_VMENTRY) ||
+	    (exit_intr_info & INTR_INFO_VECTOR_MASK) == MC_VECTOR)
+		kvm_machine_check();
+
 	/* We need to handle NMIs before interrupts are enabled */
 	if ((exit_intr_info & INTR_INFO_INTR_TYPE_MASK) == INTR_TYPE_NMI_INTR &&
 	    (exit_intr_info & INTR_INFO_VALID_MASK)) {
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -382,6 +382,8 @@ struct kvm_vcpu_arch {
 	u64 mcg_status;
 	u64 mcg_ctl;
 	u64 *mce_banks;
+
+	u32 exit_reason;
 };
 
 struct kvm_mem_alias {
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -420,6 +420,7 @@ void do_machine_check(struct pt_regs * r
  out2:
 	atomic_dec(&mce_entry);
 }
+EXPORT_SYMBOL_GPL(do_machine_check);
 
 #ifdef CONFIG_X86_MCE_INTEL
 /***


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/