[Qemu-devel] [PATCH] x86: Work around SMI migration breakages

Dr. David Alan Gilbert (git) posted 1 patch 7 years, 1 month ago
Patches applied successfully (tree, apply log)
git fetch https://github.com/patchew-project/qemu tags/patchew/20170223133441.16010-1-dgilbert@redhat.com
Test checkpatch passed
Test docker passed
Test s390x passed
include/hw/i386/pc.h | 4 ++++
target/i386/cpu.c    | 2 ++
target/i386/cpu.h    | 3 +++
target/i386/kvm.c    | 7 ++++++-
4 files changed, 15 insertions(+), 1 deletion(-)
[Qemu-devel] [PATCH] x86: Work around SMI migration breakages
Posted by Dr. David Alan Gilbert (git) 7 years, 1 month ago
From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>

Migration from a 2.3.0 qemu results in a reboot on the receiving QEMU
due to a disagreement about SM (System management) interrupts.

2.3.0 didn't have much SMI support, but it did set CPU_INTERRUPT_SMI
and this gets into the migration stream, but on 2.3.0 it
never got delivered.

~2.4.0 SMI interrupt support was added but was broken - so
that when a 2.3.0 stream was received it cleared the CPU_INTERRUPT_SMI
but never actually caused an interrupt.

The SMI delivery was recently fixed by 68c6efe07a, but the
effect now is that an incoming 2.3.0 stream takes the interrupt it
had flagged but it's bios can't actually handle it(I think
partly due to the original interrupt not being taken during boot?).
The consequence is a triple(?) fault and a reboot.

Tested from:
  2.3.1 -M 2.3.0
  2.7.0 -M 2.3.0
  2.8.0 -M 2.3.0
  2.8.0 -M 2.8.0

This corresponds to RH bugzilla entry 1420679.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
---
 include/hw/i386/pc.h | 4 ++++
 target/i386/cpu.c    | 2 ++
 target/i386/cpu.h    | 3 +++
 target/i386/kvm.c    | 7 ++++++-
 4 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 079e8d9..cb49826 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -624,6 +624,10 @@ bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
         .driver   = "Broadwell-noTSX" "-" TYPE_X86_CPU,\
         .property = "xlevel",\
         .value    = stringify(0x8000000a),\
+    },{\
+        .driver = TYPE_X86_CPU,\
+        .property = "kvm-no-smi-migration",\
+        .value    = "on",\
     },
 
 #define PC_COMPAT_2_2 \
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index fd7add2..b2058cd 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -3708,6 +3708,8 @@ static Property x86_cpu_properties[] = {
     DEFINE_PROP_BOOL("cpuid-0xb", X86CPU, enable_cpuid_0xb, true),
     DEFINE_PROP_BOOL("lmce", X86CPU, enable_lmce, false),
     DEFINE_PROP_BOOL("l3-cache", X86CPU, enable_l3_cache, true),
+    DEFINE_PROP_BOOL("kvm-no-smi-migration", X86CPU, kvm_no_smi_migration,
+                     false),
     DEFINE_PROP_BOOL("vmware-cpuid-freq", X86CPU, vmware_cpuid_freq, true),
     DEFINE_PROP_END_OF_LIST()
 };
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 8df124f..c5dfef8 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1255,6 +1255,9 @@ struct X86CPU {
     /* if true override the phys_bits value with a value read from the host */
     bool host_phys_bits;
 
+    /* Stop SMI delivery for migration compatibility with old machines */
+    bool kvm_no_smi_migration;
+
     /* Number of physical address bits supported */
     uint32_t phys_bits;
 
diff --git a/target/i386/kvm.c b/target/i386/kvm.c
index 27fd050..5bfe633 100644
--- a/target/i386/kvm.c
+++ b/target/i386/kvm.c
@@ -2531,7 +2531,12 @@ static int kvm_put_vcpu_events(X86CPU *cpu, int level)
             events.smi.pending = 0;
             events.smi.latched_init = 0;
         }
-        events.flags |= KVM_VCPUEVENT_VALID_SMM;
+        /* Stop SMI delivery on old machine types to avoid a reboot
+         * on an inward migration of an old VM.
+         */
+        if (!cpu->kvm_no_smi_migration) {
+            events.flags |= KVM_VCPUEVENT_VALID_SMM;
+        }
     }
 
     if (level >= KVM_PUT_RESET_STATE) {
-- 
2.9.3


Re: [Qemu-devel] [PATCH] x86: Work around SMI migration breakages
Posted by Paolo Bonzini 7 years, 1 month ago

On 23/02/2017 14:34, Dr. David Alan Gilbert (git) wrote:
> From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> 
> Migration from a 2.3.0 qemu results in a reboot on the receiving QEMU
> due to a disagreement about SM (System management) interrupts.
> 
> 2.3.0 didn't have much SMI support, but it did set CPU_INTERRUPT_SMI
> and this gets into the migration stream, but on 2.3.0 it
> never got delivered.
> 
> ~2.4.0 SMI interrupt support was added but was broken - so
> that when a 2.3.0 stream was received it cleared the CPU_INTERRUPT_SMI
> but never actually caused an interrupt.
> 
> The SMI delivery was recently fixed by 68c6efe07a, but the
> effect now is that an incoming 2.3.0 stream takes the interrupt it
> had flagged but it's bios can't actually handle it(I think
> partly due to the original interrupt not being taken during boot?).
> The consequence is a triple(?) fault and a reboot.
> 
> Tested from:
>   2.3.1 -M 2.3.0
>   2.7.0 -M 2.3.0
>   2.8.0 -M 2.3.0
>   2.8.0 -M 2.8.0
> 
> This corresponds to RH bugzilla entry 1420679.
> 
> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> ---
>  include/hw/i386/pc.h | 4 ++++
>  target/i386/cpu.c    | 2 ++
>  target/i386/cpu.h    | 3 +++
>  target/i386/kvm.c    | 7 ++++++-
>  4 files changed, 15 insertions(+), 1 deletion(-)
> 
> diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
> index 079e8d9..cb49826 100644
> --- a/include/hw/i386/pc.h
> +++ b/include/hw/i386/pc.h
> @@ -624,6 +624,10 @@ bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
>          .driver   = "Broadwell-noTSX" "-" TYPE_X86_CPU,\
>          .property = "xlevel",\
>          .value    = stringify(0x8000000a),\
> +    },{\
> +        .driver = TYPE_X86_CPU,\
> +        .property = "kvm-no-smi-migration",\
> +        .value    = "on",\
>      },
>  
>  #define PC_COMPAT_2_2 \
> diff --git a/target/i386/cpu.c b/target/i386/cpu.c
> index fd7add2..b2058cd 100644
> --- a/target/i386/cpu.c
> +++ b/target/i386/cpu.c
> @@ -3708,6 +3708,8 @@ static Property x86_cpu_properties[] = {
>      DEFINE_PROP_BOOL("cpuid-0xb", X86CPU, enable_cpuid_0xb, true),
>      DEFINE_PROP_BOOL("lmce", X86CPU, enable_lmce, false),
>      DEFINE_PROP_BOOL("l3-cache", X86CPU, enable_l3_cache, true),
> +    DEFINE_PROP_BOOL("kvm-no-smi-migration", X86CPU, kvm_no_smi_migration,
> +                     false),
>      DEFINE_PROP_BOOL("vmware-cpuid-freq", X86CPU, vmware_cpuid_freq, true),
>      DEFINE_PROP_END_OF_LIST()
>  };
> diff --git a/target/i386/cpu.h b/target/i386/cpu.h
> index 8df124f..c5dfef8 100644
> --- a/target/i386/cpu.h
> +++ b/target/i386/cpu.h
> @@ -1255,6 +1255,9 @@ struct X86CPU {
>      /* if true override the phys_bits value with a value read from the host */
>      bool host_phys_bits;
>  
> +    /* Stop SMI delivery for migration compatibility with old machines */
> +    bool kvm_no_smi_migration;
> +
>      /* Number of physical address bits supported */
>      uint32_t phys_bits;
>  
> diff --git a/target/i386/kvm.c b/target/i386/kvm.c
> index 27fd050..5bfe633 100644
> --- a/target/i386/kvm.c
> +++ b/target/i386/kvm.c
> @@ -2531,7 +2531,12 @@ static int kvm_put_vcpu_events(X86CPU *cpu, int level)
>              events.smi.pending = 0;
>              events.smi.latched_init = 0;
>          }
> -        events.flags |= KVM_VCPUEVENT_VALID_SMM;
> +        /* Stop SMI delivery on old machine types to avoid a reboot
> +         * on an inward migration of an old VM.
> +         */
> +        if (!cpu->kvm_no_smi_migration) {
> +            events.flags |= KVM_VCPUEVENT_VALID_SMM;
> +        }
>      }
>  
>      if (level >= KVM_PUT_RESET_STATE) {
> 

Queued, thanks.

Paolo