linux-kernel - Re: [PATCH v3 1/2] x86: KVM: Advertise FSRS and FSRC on AMD to userspace

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] [day] [month] [year] [list]

Message-ID: <51541100-d98a-482a-96aa-20790b23df28@amd.com>
Date: Mon, 9 Dec 2024 13:22:28 -0600
From: "Moger, Babu" <babu.moger@....com>
To: Maksim Davydov <davydov-max@...dex-team.ru>, "Moger, Babu"
 <bmoger@....com>, Jim Mattson <jmattson@...gle.com>
Cc: kvm@...r.kernel.org, linux-kernel@...r.kernel.org, x86@...nel.org,
 seanjc@...gle.com, mingo@...hat.com, bp@...en8.de, tglx@...utronix.de,
 dave.hansen@...ux.intel.com, hpa@...or.com, pbonzini@...hat.com
Subject: Re: [PATCH v3 1/2] x86: KVM: Advertise FSRS and FSRC on AMD to
 userspace



On 12/9/24 06:11, Maksim Davydov wrote:
> 
> 
> On 12/6/24 21:11, Moger, Babu wrote:
>>
>> On 12/4/2024 10:57 AM, Jim Mattson wrote:
>>> On Wed, Dec 4, 2024 at 5:43 AM Maksim Davydov
>>> <davydov-max@...dex-team.ru>  wrote:
>>>> Fast short REP STOSB and fast short CMPSB support on AMD processors are
>>>> provided in other CPUID function in comparison with Intel processors:
>>>> * FSRS: 10 bit in 0x80000021_EAX
>>>> * FSRC: 11 bit in 0x80000021_EAX
>>> I have to wonder why these bits aren't documented in the APM. I assume
>>> you pulled them out of some PPR? I would be hesitant to include CPUID
>>> bit definitions that may be microarchitecture-specific rather than
>>> architectural.
>>>
>>> Perhaps someone from AMD should at least ACK this change?
>>
>> APM updates are in progress right now, but haven’t been able to get an ETA.
>>
>> Will confirm once APM is released.
>>
> 
> Thanks a lot!
> It means that this series should be sent as 2 independent parts:
> 1. FSRS and FSRC will wait for updated APM

Yes. We can wait on this. APM is being updated right now. We can add FSRS
and FSRC support when APM is publicly available.

> 2. Speculation control bits will be sent as a separate patch

Sure.
> 
>>>> AMD bit numbers differ from existing definition of FSRC and
>>>> FSRS. So, the new appropriate values have to be added with new names.
>>>>
>>>> It's safe to advertise these features to userspace because they are a
>>>> part
>>>> of CPU model definition and they can't be disabled (as existing Intel
>>>> features).
>>>>
>>>> Fixes: 2a4209d6a9cb ("KVM: x86: Advertise fast REP string features
>>>> inherent to the CPU")
>>>> Signed-off-by: Maksim Davydov<davydov-max@...dex-team.ru>
>>>> ---
>>>>   arch/x86/include/asm/cpufeatures.h | 2 ++
>>>>   arch/x86/kvm/cpuid.c               | 4 ++--
>>>>   2 files changed, 4 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/arch/x86/include/asm/cpufeatures.h
>>>> b/arch/x86/include/asm/cpufeatures.h
>>>> index 17b6590748c0..45f87a026bba 100644
>>>> --- a/arch/x86/include/asm/cpufeatures.h
>>>> +++ b/arch/x86/include/asm/cpufeatures.h
>>>> @@ -460,6 +460,8 @@
>>>>   #define X86_FEATURE_NULL_SEL_CLR_BASE  (20*32+ 6) /* Null Selector
>>>> Clears Base */
>>>>   #define X86_FEATURE_AUTOIBRS           (20*32+ 8) /* Automatic IBRS */
>>>>   #define X86_FEATURE_NO_SMM_CTL_MSR     (20*32+ 9) /* SMM_CTL MSR is
>>>> not present */
>>>> +#define X86_FEATURE_AMD_FSRS           (20*32+10) /* AMD Fast short
>>>> REP STOSB supported */
>>>> +#define X86_FEATURE_AMD_FSRC           (20*32+11) /* AMD Fast short
>>>> REP CMPSB supported */
>>>>
>>>>   #define X86_FEATURE_SBPB               (20*32+27) /* Selective
>>>> Branch Prediction Barrier */
>>>>   #define X86_FEATURE_IBPB_BRTYPE                (20*32+28) /*
>>>> MSR_PRED_CMD[IBPB] flushes all branch type predictions */
>>>> diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
>>>> index 097bdc022d0f..7bc095add8ee 100644
>>>> --- a/arch/x86/kvm/cpuid.c
>>>> +++ b/arch/x86/kvm/cpuid.c
>>>> @@ -799,8 +799,8 @@ void kvm_set_cpu_caps(void)
>>>>
>>>>          kvm_cpu_cap_mask(CPUID_8000_0021_EAX,
>>>>                  F(NO_NESTED_DATA_BP) | F(LFENCE_RDTSC) | 0 /*
>>>> SmmPgCfgLock */ |
>>>> -               F(NULL_SEL_CLR_BASE) | F(AUTOIBRS) | 0 /*
>>>> PrefetchCtlMsr */ |
>>>> -               F(WRMSR_XX_BASE_NS)
>>>> +               F(NULL_SEL_CLR_BASE) | F(AUTOIBRS) | F(AMD_FSRS) |
>>>> +               F(AMD_FSRC) | 0 /* PrefetchCtlMsr */ |
>>>> F(WRMSR_XX_BASE_NS)
>>>>          );
>>>>
>>>>          kvm_cpu_cap_check_and_set(X86_FEATURE_SBPB);
>>>> -- 
>>>> 2.34.1
>>>>
> 

-- 
Thanks
Babu Moger