Re: [PATCH v2 09/12] perf vendor events intel: Add novalake v1.00 events

[Ian Rogers]

On Tue, Jul 14, 2026 at 1:40 PM Chun-Tse Shao <ctshao@xxxxxxxxxx> wrote:
[snip]
> diff --git a/tools/perf/pmu-events/arch/x86/mapfile.csv b/tools/perf/pmu-events/arch/x86/mapfile.csv
> index a15cd3280993..7bbb8b237b35 100644
> --- a/tools/perf/pmu-events/arch/x86/mapfile.csv
> +++ b/tools/perf/pmu-events/arch/x86/mapfile.csv
> @@ -39,6 +39,7 @@ GenuineIntel-6-8[CD],v1.19,tigerlake,core
>  GenuineIntel-6-2C,v5,westmereep-dp,core
>  GenuineIntel-6-25,v4,westmereep-sp,core
>  GenuineIntel-6-2F,v4,westmereex,core
> +GenuineIntel-18-[13],v1.00,novalake,core

I think we should avoid editing the mapfile.csv and just use what
comes out of create_perf_json.py:
https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py
To avoid the strings differing we can modify get_cpuid_str:
https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/tree/tools/perf/arch/x86/util/header.c#n70
so that "%s-%u-%X-%X$" is "%s-%u-%02X-%X$" (the values are vendor,
family, model and stepping with the stepping being optional).
This ensures the model number is always 2 characters wide, zero-padded
if necessary. I don't believe this impacts any other x86 cpuid
strings.

Thanks,
Ian

>  AuthenticAMD-23-([12][0-9A-F]|[0-9A-F]),v2,amdzen1,core
>  AuthenticAMD-23-[[:xdigit:]]+,v1,amdzen2,core
>  AuthenticAMD-25-([245][[:xdigit:]]|[[:xdigit:]]),v1,amdzen3,core

> +        "Unit": "cpu_atom"
> +    }
> +]
> --
> 2.55.0.141.g00534a21ce-goog
>