On some x86 processors, CPUID leaf 0xA provides information
on Architectural Performance Monitoring features. It
advertises a PMU version which Qemu uses to determine the
availability of additional MSRs to manage the PMCs.
Upon receiving a KVM_GET_SUPPORTED_CPUID ioctl request for
the same, the kernel constructs return values based on the
x86_pmu_capability irrespective of the vendor.
This leaf and the additional MSRs are not supported on AMD
processors. If PerfMonV2 is detected, the PMU version is
set to 2 and guest startup breaks because of an attempt to
access a non-existent MSR. Return zeros to avoid this.
Fixes: a6c06ed1a60a ("KVM: Expose the architectural performance monitoring CPUID leaf")
Reported-by: Vasant Hegde <[email protected]>
Signed-off-by: Sandipan Das <[email protected]>
---
arch/x86/kvm/cpuid.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 4b62d80bb22f..6bd65cad75ef 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -872,6 +872,11 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
union cpuid10_eax eax;
union cpuid10_edx edx;
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
+ entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
+ break;
+ }
+
perf_get_x86_pmu_capability(&cap);
/*
--
2.32.0
On Thu, Apr 21, 2022 at 11:16:59AM +0530, Sandipan Das wrote:
> On some x86 processors, CPUID leaf 0xA provides information
> on Architectural Performance Monitoring features. It
> advertises a PMU version which Qemu uses to determine the
> availability of additional MSRs to manage the PMCs.
>
> Upon receiving a KVM_GET_SUPPORTED_CPUID ioctl request for
> the same, the kernel constructs return values based on the
> x86_pmu_capability irrespective of the vendor.
>
> This leaf and the additional MSRs are not supported on AMD
> processors. If PerfMonV2 is detected, the PMU version is
> set to 2 and guest startup breaks because of an attempt to
> access a non-existent MSR. Return zeros to avoid this.
>
> Fixes: a6c06ed1a60a ("KVM: Expose the architectural performance monitoring CPUID leaf")
> Reported-by: Vasant Hegde <[email protected]>
> Signed-off-by: Sandipan Das <[email protected]>
> ---
> arch/x86/kvm/cpuid.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
> index 4b62d80bb22f..6bd65cad75ef 100644
> --- a/arch/x86/kvm/cpuid.c
> +++ b/arch/x86/kvm/cpuid.c
> @@ -872,6 +872,11 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
> union cpuid10_eax eax;
> union cpuid10_edx edx;
>
> + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
This is the obligatory question about HYGON; should they be included
here?
x86 is getting a number of me-too patches from both sides, where
behaviour has diverged for no raisin and then needs to be fixed up
again.
> + entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
> + break;
> + }
> +
> perf_get_x86_pmu_capability(&cap);
>
> /*
> --
> 2.32.0
>
Hi Peter,
On 4/26/2022 1:45 PM, Peter Zijlstra wrote:
> On Thu, Apr 21, 2022 at 11:16:59AM +0530, Sandipan Das wrote:
>> On some x86 processors, CPUID leaf 0xA provides information
>> on Architectural Performance Monitoring features. It
>> advertises a PMU version which Qemu uses to determine the
>> availability of additional MSRs to manage the PMCs.
>>
>> Upon receiving a KVM_GET_SUPPORTED_CPUID ioctl request for
>> the same, the kernel constructs return values based on the
>> x86_pmu_capability irrespective of the vendor.
>>
>> This leaf and the additional MSRs are not supported on AMD
>> processors. If PerfMonV2 is detected, the PMU version is
>> set to 2 and guest startup breaks because of an attempt to
>> access a non-existent MSR. Return zeros to avoid this.
>>
>> Fixes: a6c06ed1a60a ("KVM: Expose the architectural performance monitoring CPUID leaf")
>> Reported-by: Vasant Hegde <[email protected]>
>> Signed-off-by: Sandipan Das <[email protected]>
>> ---
>> arch/x86/kvm/cpuid.c | 5 +++++
>> 1 file changed, 5 insertions(+)
>>
>> diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
>> index 4b62d80bb22f..6bd65cad75ef 100644
>> --- a/arch/x86/kvm/cpuid.c
>> +++ b/arch/x86/kvm/cpuid.c
>> @@ -872,6 +872,11 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
>> union cpuid10_eax eax;
>> union cpuid10_edx edx;
>>
>> + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
>
> This is the obligatory question about HYGON; should they be included
> here?
>
> x86 is getting a number of me-too patches from both sides, where
> behaviour has diverged for no raisin and then needs to be fixed up
> again.
>
From what I understand, Hygon processors also do not support this leaf.
Copying Pu Wen for confirmation.
- Sandipan
>> + entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
>> + break;
>> + }
>> +
>> perf_get_x86_pmu_capability(&cap);
>>
>> /*
>> --
>> 2.32.0
>>
On Tue, Apr 26, 2022 at 11:45:32AM +0000, Wen Pu wrote:
> I have confirmed that Hygon processors do not support CPUID leaf 0xA.
> Sandipan, could you please also add X86_VENDOR_HYGON in the if
> conditional?
Will do, thanks!
On 2022/4/26 16:40, Sandipan Das wrote:
> Hi Peter,
>
> On 4/26/2022 1:45 PM, Peter Zijlstra wrote:
>> On Thu, Apr 21, 2022 at 11:16:59AM +0530, Sandipan Das wrote:
>>> On some x86 processors, CPUID leaf 0xA provides information
>>> on Architectural Performance Monitoring features. It
>>> advertises a PMU version which Qemu uses to determine the
>>> availability of additional MSRs to manage the PMCs.
>>>
>>> Upon receiving a KVM_GET_SUPPORTED_CPUID ioctl request for
>>> the same, the kernel constructs return values based on the
>>> x86_pmu_capability irrespective of the vendor.
>>>
>>> This leaf and the additional MSRs are not supported on AMD
>>> processors. If PerfMonV2 is detected, the PMU version is
>>> set to 2 and guest startup breaks because of an attempt to
>>> access a non-existent MSR. Return zeros to avoid this.
>>>
>>> Fixes: a6c06ed1a60a ("KVM: Expose the architectural performance monitoring CPUID leaf")
>>> Reported-by: Vasant Hegde <[email protected]>
>>> Signed-off-by: Sandipan Das <[email protected]>
>>> ---
>>> arch/x86/kvm/cpuid.c | 5 +++++
>>> 1 file changed, 5 insertions(+)
>>>
>>> diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
>>> index 4b62d80bb22f..6bd65cad75ef 100644
>>> --- a/arch/x86/kvm/cpuid.c
>>> +++ b/arch/x86/kvm/cpuid.c
>>> @@ -872,6 +872,11 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
>>> union cpuid10_eax eax;
>>> union cpuid10_edx edx;
>>>
>>> + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
>>
>> This is the obligatory question about HYGON; should they be included
>> here?
>>
>> x86 is getting a number of me-too patches from both sides, where
>> behaviour has diverged for no raisin and then needs to be fixed up
>> again.
>>
>
> From what I understand, Hygon processors also do not support this leaf.
> Copying Pu Wen for confirmation.
Hi Peter, Sandipan：
I have confirmed that Hygon processors do not support CPUID leaf 0xA.
Sandipan, could you please also add X86_VENDOR_HYGON in the if
conditional?
--
Regards,
Pu Wen
> - Sandipan
>
>>> + entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
>>> + break;
>>> + }
>>> +
>>> perf_get_x86_pmu_capability(&cap);
>>>
>>> /*
>>> --
>>> 2.32.0
>>>