Re: [PATCH v2] KVM: x86/pmu: Support full width counting

["Xu, Like" <like.xu@intel.com>]

Hi Paolo,

Thanks for your detailed comments.

On 2020/5/7 15:57, Paolo Bonzini wrote:
> On 07/05/20 04:14, Like Xu wrote:
>> +static inline u64 vmx_get_perf_capabilities(void)
>> +{
>> +	u64 perf_cap = 0;
>> +
>> +	if (boot_cpu_has(X86_FEATURE_PDCM))
>> +		rdmsrl(MSR_IA32_PERF_CAPABILITIES, perf_cap);
>> +
>> +	/* Currently, KVM only supports Full-Width Writes. */
>> +	perf_cap &= PMU_CAP_FW_WRITES;
>> +
>> +	return perf_cap;
>> +}
>> +
> Since counters are virtualized, it seems to me that you can support
> PMU_CAP_FW_WRITES unconditionally, even if the host lacks it.  So just
> return PMU_CAP_FW_WRITES from this function.
Sure, let's export PMU_CAP_FW_WRITES unconditionally.
>
>> +	case MSR_IA32_PERF_CAPABILITIES:
>> +		return 1; /* RO MSR */
>>   	default:
> You need to allow writes from the host if (data &
> ~vmx_get_perf_capabilities()) == 0.
Yes, it makes sense after I understand the intention of host_initiated.
>
>> -			if (!msr_info->host_initiated)
>> +		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
>> +			(pmc = get_gp_pmc(pmu, msr, MSR_IA32_PMC0))) {
>> +			if (data & ~pmu->counter_bitmask[KVM_PMC_GP])
>> +				return 1;
>> +			if (!fw_writes_is_enabled(pmu))
>>   				data = (s64)(s32)data;
>
> You are dropping the test on msr_info->host_initiated here, you should
> keep it otherwise you allow full-width write to MSR_IA32_PERFCTR0 as
> well.  So:
>
> #define MSR_PMC_FULL_WIDTH_BIT      (MSR_IA32_PMC0 - MSR_IA32_PERFCTR0)
>
> 	if (!msr_info->host_initiated && !(msr & MSR_PMC_FULL_WIDTH_BIT))
> 		data = (s64)(s32)data;
Thanks, it looks good to me and I'll apply it.
>
>> +	case MSR_IA32_PERF_CAPABILITIES:
>> +		if (!nested)
>> +			return 1;
>> +		msr->data = vmx_get_perf_capabilities();
>> +		return 0;
> The !nested check is wrong.
You're absolutely right about this.
>
>> +++ b/arch/x86/kvm/x86.c
>> @@ -1220,6 +1220,13 @@ static const u32 msrs_to_save_all[] = {
>>   	MSR_ARCH_PERFMON_EVENTSEL0 + 12, MSR_ARCH_PERFMON_EVENTSEL0 + 13,
>>   	MSR_ARCH_PERFMON_EVENTSEL0 + 14, MSR_ARCH_PERFMON_EVENTSEL0 + 15,
>>   	MSR_ARCH_PERFMON_EVENTSEL0 + 16, MSR_ARCH_PERFMON_EVENTSEL0 + 17,
>> +
>> +	MSR_IA32_PMC0, MSR_IA32_PMC0 + 1, MSR_IA32_PMC0 + 2,
>> +	MSR_IA32_PMC0 + 3, MSR_IA32_PMC0 + 4, MSR_IA32_PMC0 + 5,
>> +	MSR_IA32_PMC0 + 6, MSR_IA32_PMC0 + 7, MSR_IA32_PMC0 + 8,
>> +	MSR_IA32_PMC0 + 9, MSR_IA32_PMC0 + 10, MSR_IA32_PMC0 + 11,
>> +	MSR_IA32_PMC0 + 12, MSR_IA32_PMC0 + 13, MSR_IA32_PMC0 + 14,
>> +	MSR_IA32_PMC0 + 15, MSR_IA32_PMC0 + 16, MSR_IA32_PMC0 + 17,
>>   };
> This is not needed because the full-width content is already accessible
> from the host via MSR_IA32_PERFCTRn.
That's true because they're just alias registers for MSR_IA32_PERFCTRn.
>
> Given the bugs, it is clear that you should also modify the pmu.c
> testcase for kvm-unit-tests to cover full-width writes (and especially
> the non-full-width write behavior of MSR_IA32_PERFCTRn).  Even before
> the QEMU side is begin worked on, you can test it with "-cpu
> host,migratable=off".
Sure, I added some testcases in pmu.c to cover this feature.

Please review the v3 patch 
https://lore.kernel.org/kvm/20200508083218.120559-1-like.xu@linux.intel.com/
as well as the kvm-unit-tests testcase.

Thanks,
Like Xu
>
> Thanks,
>
> Paolo
>