diff mbox series

[v7,1/2] i386: kvm: extend kvm_{get, put}_vcpu_events to support pending triple fault

Message ID 20220923073333.23381-2-chenyi.qiang@intel.com (mailing list archive)
State New, archived
Headers show
Series Enable notify VM exit | expand

Commit Message

Chenyi Qiang Sept. 23, 2022, 7:33 a.m. UTC 
For the direct triple faults, i.e. hardware detected and KVM morphed
to VM-Exit, KVM will never lose them. But for triple faults sythesized
by KVM, e.g. the RSM path, if KVM exits to userspace before the request
is serviced, userspace could migrate the VM and lose the triple fault.

A new flag KVM_VCPUEVENT_VALID_TRIPLE_FAULT is defined to signal that
the event.triple_fault_pending field contains a valid state if the
KVM_CAP_X86_TRIPLE_FAULT_EVENT capability is enabled.

Acked-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Chenyi Qiang <chenyi.qiang@intel.com>
---
 target/i386/cpu.c     |  1 +
 target/i386/cpu.h     |  1 +
 target/i386/kvm/kvm.c | 20 ++++++++++++++++++++
 3 files changed, 22 insertions(+)

Comments

Paolo Bonzini Sept. 27, 2022, 1:14 p.m. UTC | #1 
On 9/23/22 09:33, Chenyi Qiang wrote:
> For the direct triple faults, i.e. hardware detected and KVM morphed
> to VM-Exit, KVM will never lose them. But for triple faults sythesized
> by KVM, e.g. the RSM path, if KVM exits to userspace before the request
> is serviced, userspace could migrate the VM and lose the triple fault.
> 
> A new flag KVM_VCPUEVENT_VALID_TRIPLE_FAULT is defined to signal that
> the event.triple_fault_pending field contains a valid state if the
> KVM_CAP_X86_TRIPLE_FAULT_EVENT capability is enabled.

Note that you are not transmitting the field on migration.  You need
this on top:

diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index b97d182e28..d4124973ce 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1739,7 +1739,7 @@ typedef struct CPUArchState {
      uint8_t has_error_code;
      uint8_t exception_has_payload;
      uint64_t exception_payload;
-    bool triple_fault_pending;
+    uint8_t triple_fault_pending;
      uint32_t ins_len;
      uint32_t sipi_vector;
      bool tsc_valid;
diff --git a/target/i386/machine.c b/target/i386/machine.c
index cecd476e98..310b125235 100644
--- a/target/i386/machine.c
+++ b/target/i386/machine.c
@@ -1562,6 +1562,25 @@ static const VMStateDescription vmstate_arch_lbr = {
      }
  };
  
+static bool triple_fault_needed(void *opaque)
+{
+    X86CPU *cpu = opaque;
+    CPUX86State *env = &cpu->env;
+
+    return env->triple_fault_pending;
+}
+
+static const VMStateDescription vmstate_triple_fault = {
+    .name = "cpu/triple_fault",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .needed = triple_fault_needed,
+    .fields = (VMStateField[]) {
+        VMSTATE_UINT8(env.triple_fault_pending, X86CPU),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
  const VMStateDescription vmstate_x86_cpu = {
      .name = "cpu",
      .version_id = 12,
@@ -1706,6 +1725,7 @@ const VMStateDescription vmstate_x86_cpu = {
          &vmstate_amx_xtile,
  #endif
          &vmstate_arch_lbr,
+        &vmstate_triple_fault,
          NULL
      }
  };
Chenyi Qiang Sept. 28, 2022, 1:06 a.m. UTC | #2 
On 9/27/2022 9:14 PM, Paolo Bonzini wrote:
> On 9/23/22 09:33, Chenyi Qiang wrote:
>> For the direct triple faults, i.e. hardware detected and KVM morphed
>> to VM-Exit, KVM will never lose them. But for triple faults sythesized
>> by KVM, e.g. the RSM path, if KVM exits to userspace before the request
>> is serviced, userspace could migrate the VM and lose the triple fault.
>>
>> A new flag KVM_VCPUEVENT_VALID_TRIPLE_FAULT is defined to signal that
>> the event.triple_fault_pending field contains a valid state if the
>> KVM_CAP_X86_TRIPLE_FAULT_EVENT capability is enabled.
> 
> Note that you are not transmitting the field on migration.  You need
> this on top:
> 
> diff --git a/target/i386/cpu.h b/target/i386/cpu.h
> index b97d182e28..d4124973ce 100644
> --- a/target/i386/cpu.h
> +++ b/target/i386/cpu.h
> @@ -1739,7 +1739,7 @@ typedef struct CPUArchState {
>       uint8_t has_error_code;
>       uint8_t exception_has_payload;
>       uint64_t exception_payload;
> -    bool triple_fault_pending;
> +    uint8_t triple_fault_pending;
>       uint32_t ins_len;
>       uint32_t sipi_vector;
>       bool tsc_valid;
> diff --git a/target/i386/machine.c b/target/i386/machine.c
> index cecd476e98..310b125235 100644
> --- a/target/i386/machine.c
> +++ b/target/i386/machine.c
> @@ -1562,6 +1562,25 @@ static const VMStateDescription vmstate_arch_lbr = {
>       }
>   };
> 
> +static bool triple_fault_needed(void *opaque)
> +{
> +    X86CPU *cpu = opaque;
> +    CPUX86State *env = &cpu->env;
> +
> +    return env->triple_fault_pending;
> +}
> +
> +static const VMStateDescription vmstate_triple_fault = {
> +    .name = "cpu/triple_fault",
> +    .version_id = 1,
> +    .minimum_version_id = 1,
> +    .needed = triple_fault_needed,
> +    .fields = (VMStateField[]) {
> +        VMSTATE_UINT8(env.triple_fault_pending, X86CPU),
> +        VMSTATE_END_OF_LIST()
> +    }
> +};
> +
>   const VMStateDescription vmstate_x86_cpu = {
>       .name = "cpu",
>       .version_id = 12,
> @@ -1706,6 +1725,7 @@ const VMStateDescription vmstate_x86_cpu = {
>           &vmstate_amx_xtile,
>   #endif
>           &vmstate_arch_lbr,
> +        &vmstate_triple_fault,
>           NULL
>       }
>   };

Thanks Paolo for your review!

I'll add this in next version.

>
diff mbox series

Patch

diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 1db1278a59..6e107466b3 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -6017,6 +6017,7 @@  static void x86_cpu_reset(DeviceState *dev)
     env->exception_has_payload = false;
     env->exception_payload = 0;
     env->nmi_injected = false;
+    env->triple_fault_pending = false;
 #if !defined(CONFIG_USER_ONLY)
     /* We hard-wire the BSP to the first CPU. */
     apic_designate_bsp(cpu->apic_state, s->cpu_index == 0);
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 82004b65b9..b97d182e28 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1739,6 +1739,7 @@  typedef struct CPUArchState {
     uint8_t has_error_code;
     uint8_t exception_has_payload;
     uint64_t exception_payload;
+    bool triple_fault_pending;
     uint32_t ins_len;
     uint32_t sipi_vector;
     bool tsc_valid;
diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
index a1fd1f5379..3838827134 100644
--- a/target/i386/kvm/kvm.c
+++ b/target/i386/kvm/kvm.c
@@ -132,6 +132,7 @@  static int has_xcrs;
 static int has_pit_state2;
 static int has_sregs2;
 static int has_exception_payload;
+static int has_triple_fault_event;
 
 static bool has_msr_mcg_ext_ctl;
 
@@ -2483,6 +2484,16 @@  int kvm_arch_init(MachineState *ms, KVMState *s)
         }
     }
 
+    has_triple_fault_event = kvm_check_extension(s, KVM_CAP_X86_TRIPLE_FAULT_EVENT);
+    if (has_triple_fault_event) {
+        ret = kvm_vm_enable_cap(s, KVM_CAP_X86_TRIPLE_FAULT_EVENT, 0, true);
+        if (ret < 0) {
+            error_report("kvm: Failed to enable triple fault event cap: %s",
+                         strerror(-ret));
+            return ret;
+        }
+    }
+
     ret = kvm_get_supported_msrs(s);
     if (ret < 0) {
         return ret;
@@ -4299,6 +4310,11 @@  static int kvm_put_vcpu_events(X86CPU *cpu, int level)
         }
     }
 
+    if (has_triple_fault_event) {
+        events.flags |= KVM_VCPUEVENT_VALID_TRIPLE_FAULT;
+        events.triple_fault.pending = env->triple_fault_pending;
+    }
+
     return kvm_vcpu_ioctl(CPU(cpu), KVM_SET_VCPU_EVENTS, &events);
 }
 
@@ -4368,6 +4384,10 @@  static int kvm_get_vcpu_events(X86CPU *cpu)
         }
     }
 
+    if (events.flags & KVM_VCPUEVENT_VALID_TRIPLE_FAULT) {
+        env->triple_fault_pending = events.triple_fault.pending;
+    }
+
     env->sipi_vector = events.sipi_vector;
 
     return 0;