linux-kernel - Re: [PATCH v2 3/3] KVM: x86/cpuid: Add AMD CPUID ExtPerfMonAndDbg leaf 0x80000022

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 10 Nov 2022 17:26:55 +0800
From:   Like Xu <like.xu.linux@...il.com>
To:     Sean Christopherson <seanjc@...gle.com>
Cc:     Paolo Bonzini <pbonzini@...hat.com>,
        Jim Mattson <jmattson@...gle.com>, kvm@...r.kernel.org,
        linux-kernel@...r.kernel.org, Sandipan Das <sandipan.das@....com>
Subject: Re: [PATCH v2 3/3] KVM: x86/cpuid: Add AMD CPUID ExtPerfMonAndDbg
 leaf 0x80000022

On 28/10/2022 6:37 am, Sean Christopherson wrote:
> On Mon, Sep 19, 2022, Like Xu wrote:
>> From: Sandipan Das <sandipan.das@....com>
>>
>> From: Sandipan Das <sandipan.das@....com>
> 
> Duplicate "From:"s.
> 
>> CPUID leaf 0x80000022 i.e. ExtPerfMonAndDbg advertises some
>> new performance monitoring features for AMD processors.
> 
> Wrap changelogs closer to ~75 chars.
> 
>> Bit 0 of EAX indicates support for Performance Monitoring
>> Version 2 (PerfMonV2) features. If found to be set during
>> PMU initialization, the EBX bits of the same CPUID function
>> can be used to determine the number of available PMCs for
>> different PMU types.
>>
>> Expose the relevant bits via KVM_GET_SUPPORTED_CPUID so
>> that guests can make use of the PerfMonV2 features.
>>
>> Co-developed-by: Like Xu <likexu@...cent.com>
>> Signed-off-by: Like Xu <likexu@...cent.com>
>> Signed-off-by: Sandipan Das <sandipan.das@....com>
>> ---
>>   arch/x86/include/asm/perf_event.h |  8 ++++++++
>>   arch/x86/kvm/cpuid.c              | 32 ++++++++++++++++++++++++++++++-
>>   2 files changed, 39 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h
>> index f6fc8dd51ef4..c848f504e467 100644
>> --- a/arch/x86/include/asm/perf_event.h
>> +++ b/arch/x86/include/asm/perf_event.h
>> @@ -214,6 +214,14 @@ union cpuid_0x80000022_ebx {
>>   	unsigned int		full;
>>   };
>>   
>> +union cpuid_0x80000022_eax {
>> +	struct {
>> +		/* Performance Monitoring Version 2 Supported */
>> +		unsigned int	perfmon_v2:1;
>> +	} split;
>> +	unsigned int		full;
>> +};
> 
> I'm not a fan of perf's unions, but I at least understand the value added for
> CPUID entries that are a bunch of multi-bit values.  However, this leaf appears
> to be a pure features leaf.  In which case a union just makes life painful.
> 
> Please add a CPUID_8000_0022_EAX kvm_only_cpuid_leafs entry (details in link[*]
> below) so that KVM can write sane code like
> 
> 	guest_cpuid_has(X86_FEATURE_AMD_PMU_V2)
> 
> and cpuid_entry_override() instead of manually filling in information.
> 
> where appropriate.
> 
> [*] https://lore.kernel.org/all/Y1AQX3RfM+awULlE@google.com

When someone is selling syntactic sugar in the kernel space, extra attention
needs to be paid to runtime performance (union) and memory footprint 
(reverse_cpuid).

Applied for this case, while the cpuid_0x80000022_eax will be used again
in the perf core since the other new AMD PMU features are pacing at the door.

> 
>>   struct x86_pmu_capability {
>>   	int		version;
>>   	int		num_counters_gp;
>> diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
>> index 75dcf7a72605..34ba845c91b7 100644
>> --- a/arch/x86/kvm/cpuid.c
>> +++ b/arch/x86/kvm/cpuid.c
>> @@ -1094,7 +1094,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
>>   		entry->edx = 0;
>>   		break;
>>   	case 0x80000000:
>> -		entry->eax = min(entry->eax, 0x80000021);
>> +		entry->eax = min(entry->eax, 0x80000022);
>>   		/*
>>   		 * Serializing LFENCE is reported in a multitude of ways, and
>>   		 * NullSegClearsBase is not reported in CPUID on Zen2; help
>> @@ -1203,6 +1203,36 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
>>   		if (!static_cpu_has_bug(X86_BUG_NULL_SEG))
>>   			entry->eax |= BIT(6);
>>   		break;
>> +	/* AMD Extended Performance Monitoring and Debug */
>> +	case 0x80000022: {
>> +		union cpuid_0x80000022_eax eax;
>> +		union cpuid_0x80000022_ebx ebx;
>> +
>> +		entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
>> +		if (!enable_pmu)
> 
> Shouldn't
> 
> 	case 0xa: { /* Architectural Performance Monitoring */
> 
> also check enable_pmu instead of X86_FEATURE_ARCH_PERFMON?

Applied as a separate patch, though KVM will have zero-padded kvm_pmu_cap to do
subsequent assignments when !enable_pmu but that doesn't hurt.

> 
>> +			break;
>> +
>> +		if (kvm_pmu_cap.version > 1) {
>> +			/* AMD PerfMon is only supported up to V2 in the KVM. */
>> +			eax.split.perfmon_v2 = 1;
> 
> With a proper CPUID_8000_0022_EAX, this becomes:
> 
> 		entry->ecx = entry->edx = 0;
> 		if (!enable_pmu || !kvm_cpu_cap_has(X86_FEATURE_AMD_PMU_V2)) {
> 			entry->eax = entry->ebx;
> 			break;
> 		}
> 
> 		cpuid_entry_override(entry, CPUID_8000_0022_EAX);
> 
> 		...

Then in this code block, we will have:

	/* AMD PerfMon is only supported up to V2 in the KVM. */
	entry->eax |= BIT(0);

to cover AMD Perfmon V3+, any better move ?

> 
> 		entry->ebx = ebx.full;