linux-kernel - Re: [PATCH v6 00/44] KVM: x86: Add support for mediated vPMUs

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]

Message-ID: <f6408786-8b40-4ded-8a20-307fe0133e98@amd.com>
Date: Thu, 18 Dec 2025 14:49:34 +0530
From: Manali Shukla <manali.shukla@....com>
To: Peter Zijlstra <peterz@...radead.org>, Sean Christopherson
	<seanjc@...gle.com>
CC: Marc Zyngier <maz@...nel.org>, Oliver Upton <oupton@...nel.org>, "Tianrui
 Zhao" <zhaotianrui@...ngson.cn>, Bibo Mao <maobibo@...ngson.cn>, Huacai Chen
	<chenhuacai@...nel.org>, Anup Patel <anup@...infault.org>, Paul Walmsley
	<pjw@...nel.org>, Palmer Dabbelt <palmer@...belt.com>, Albert Ou
	<aou@...s.berkeley.edu>, Xin Li <xin@...or.com>, "H. Peter Anvin"
	<hpa@...or.com>, Andy Lutomirski <luto@...nel.org>, Ingo Molnar
	<mingo@...hat.com>, Arnaldo Carvalho de Melo <acme@...nel.org>, Namhyung Kim
	<namhyung@...nel.org>, Paolo Bonzini <pbonzini@...hat.com>,
	<linux-arm-kernel@...ts.infradead.org>, <kvmarm@...ts.linux.dev>,
	<kvm@...r.kernel.org>, <loongarch@...ts.linux.dev>,
	<kvm-riscv@...ts.infradead.org>, <linux-riscv@...ts.infradead.org>,
	<linux-kernel@...r.kernel.org>, <linux-perf-users@...r.kernel.org>, "Mingwei
 Zhang" <mizhang@...gle.com>, Xudong Hao <xudong.hao@...el.com>, Sandipan Das
	<sandipan.das@....com>, Dapeng Mi <dapeng1.mi@...ux.intel.com>, Xiong Zhang
	<xiong.y.zhang@...ux.intel.com>, Jim Mattson <jmattson@...gle.com>
Subject: Re: [PATCH v6 00/44] KVM: x86: Add support for mediated vPMUs

On 12/8/2025 9:07 PM, Peter Zijlstra wrote:
> On Fri, Dec 05, 2025 at 04:16:36PM -0800, Sean Christopherson wrote:
>> My hope/plan is that the perf changes will go through the tip tree with a
>> stable tag/branch, and the KVM changes will go the kvm-x86 tree.
> 
>> Kan Liang (7):
>>   perf: Skip pmu_ctx based on event_type
>>   perf: Add generic exclude_guest support
>>   perf: Add APIs to create/release mediated guest vPMUs
>>   perf: Clean up perf ctx time
>>   perf: Add a EVENT_GUEST flag
>>   perf: Add APIs to load/put guest mediated PMU context
>>   perf/x86/intel: Support PERF_PMU_CAP_MEDIATED_VPMU
>>
>> Mingwei Zhang (3):
>>   perf/x86/core: Plumb mediated PMU capability from x86_pmu to
>>     x86_pmu_cap
>>
>> Sandipan Das (3):
>>   perf/x86/core: Do not set bit width for unavailable counters
>>   perf/x86/amd: Support PERF_PMU_CAP_MEDIATED_VPMU for AMD host
>>
>> Sean Christopherson (19):
>>   perf: Move security_perf_event_free() call to __free_event()
>>   perf/x86/core: Register a new vector for handling mediated guest PMIs
>>   perf/x86/core: Add APIs to switch to/from mediated PMI vector (for
>>     KVM)
> 
> That all looks to be in decent shape; lets go get this merged. There is
> the nit on patch 4 and I think a number of the exports want to be
> EXPORT_SYMBOL_FOR_KVM() instead, but we can do that on top.
> 
> I'll queue these patches after rc1.

I have tested this series on AMD EPYC 9745 (Turin, 128-Core Processor)
with 12 hours of perf_fuzzer runs across multiple configurations. No new
failures were observed.

Test Platform:
--------------
Hardware: AMD EPYC 9745 128-Core Processor (Turin)
Test Duration: 12 hours of perf_fuzzer per configuration

Configuration 1: enable_mediated_pmu=Y
---------------------------------------
SVM guests:
  -cpu host
  -cpu host,-perfmon-v2
  -cpu host,-perfctr-core

SEV-ES guest:
  -cpu host

SEV-SNP guest:
  -cpu host

Configuration 2: enable_mediated_pmu=N
---------------------------------------
SVM guest:
  -cpu EPYC-Genoa-v2,+pmu
  -cpu EPYC-Genoa-v2,+pmu,-perfmon-v2
  -cpu EPYC-Genoa-v2,+pmu,-perfctr-core

SEV-ES guest:
  -cpu EPYC-Genoa-v2,+pmu

SEV-SNP guest:
  -cpu EPYC-Genoa-v2,+pmu

KVM Unit Tests (KUT):
---------------------
Tested all combinations of:
  - enable_mediated_pmu={0,1}
  - force_emulation_prefix={0,1}

CPU configurations tested:
  -cpu host
  -cpu host,-perfmon-v2
  -cpu host,-perfctr-core

All KUT tests passed as expected across all combinations.

Tested-by: Manali Shukla <manali.shukla@....com>

-Manali