linux-kernel - Re: [PATCH V2 0/3] Support auto counter reload

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 for Android: free password hash cracker in your pocket

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

Message-ID: <108c7d02-257c-4131-a7f0-f9cb34ed3f5c@linux.intel.com>
Date: Fri, 14 Mar 2025 09:06:07 -0400
From: "Liang, Kan" <kan.liang@...ux.intel.com>
To: Ingo Molnar <mingo@...nel.org>
Cc: peterz@...radead.org, acme@...nel.org, namhyung@...nel.org,
 irogers@...gle.com, adrian.hunter@...el.com, ak@...ux.intel.com,
 linux-kernel@...r.kernel.org, eranian@...gle.com, thomas.falcon@...el.com
Subject: Re: [PATCH V2 0/3] Support auto counter reload



On 2025-03-14 5:51 a.m., Ingo Molnar wrote:
> 
> * kan.liang@...ux.intel.com <kan.liang@...ux.intel.com> wrote:
> 
>> From: Kan Liang <kan.liang@...ux.intel.com>
>>
>> Changes since V1:
>> - Add a check to the reload value which cannot exceeds the max period
>> - Avoid invoking intel_pmu_enable_acr() for the perf metrics event.
>> - Update comments explain to case which the event->attr.config2 exceeds
>>   the group size
> 
>> The 2498 samples are all from the branch-misses events for the Loop 2.
>>
>> The number of samples and overhead is significantly reduced without
>> losing any information.
> 
> Ok, that looks like a pretty sweet PMU feature.
> 

Thanks for the review.

> What is the hardware support range of this auto count reload feature, 
> how recent CPU does one have to have?

The feature was first introduced into the Sierra Forest server, which
was launched last year.
https://en.wikipedia.org/wiki/Sierra_Forest

All the future platforms should have it support as well.

> 
> The series has aged a bit though, while a variant of patch #1 has been 
> merged already under:
> 
>   47a973fd7563 perf/x86/intel: Fix ARCH_PERFMON_NUM_COUNTER_LEAF
> 
> ... but #2 and #3 don't apply cleanly anymore.
> 
> Mind sending a refreshed series perhaps?

No problem.

I saw Peter just gives several feedback. I will also address the
concerns in the new series.

Thanks,
Kan