Re: [PATCH v8 0/3] ACPI / cpufreq: CPPC: Add ospm_nominal_perf support

[Christian Loehle]

On 11/6/26 16:08, Rafael J. Wysocki (Intel) wrote:
> Hi Sumit,
> 
> On Fri, Oct 30, 2026 at 6:28 PM Sumit Gupta <sumitg@xxxxxxxxxx> wrote:
>>
>> This series adds support for the OSPM Nominal Performance register
>> (ACPI 6.6, Section 8.4.6.1.2.6) to the CPPC ACPI and cpufreq drivers.
>>
>> Unlike the read-only Nominal Performance register, OSPM Nominal
>> Performance is writable and lets OSPM request a nominal level below
>> the platform-reported nominal. The platform treats performance above
>> this level as boost and below as throttle.
>>
>>  - Patch 1: adds ACPI helpers to write the register and to check that
>>    the platform supports it.
>>
>>  - Patch 2: adds the per-policy ospm_nominal_freq attribute in kHz. The
>>    value is kept across CPU hotplug and suspend/resume. The register
>>    cannot be read back, so a read returns the last value written.
>>
>>  - Patch 3: uses the OSPM nominal as the maximum frequency while boost
>>    is off. Lowering the nominal creates a boost range on a platform
>>    that has none, and boost can then be enabled.
>>
>> This series applies on top of v7 of:
>>   cpufreq: CPPC: Preserve OSPM-set registers across hotplug and unload
>>   https://lore.kernel.org/lkml/20261030092416.1993889-1-sumitg@xxxxxxxxxx/
>>
>> v7 -> v8:
>>  - Rebase on Christian's CPPC fixes series.
>>  - Patch 1:
>>    - use cpc_is_writable() from that series and drop cpc_reg_writable().
>>    - add the register to the list of controls the driver writes.
>>    - take cpc_desc_lock in cppc_ospm_nominal_perf_supported(), so the
>>      descriptor cannot be removed during the check.
>>  - Patch 2:
>>    - make ospm_nominal_freq read-write. A read returns the last value
>>      written, or "<unset>" before the first write. (Christian, Mario)
>>    - keep saving the values in the store handler, not at init(), so
>>      the driver writes the register only after the OS sets it. (Pierre)
>>  - Patch 3:
>>    - document in the ABI file how the OSPM nominal sets cpuinfo_max_freq
>>      and caps scaling_max_freq.
>>
>> Sumit Gupta (3):
>>   ACPI: CPPC: Add ospm_nominal_perf support
>>   cpufreq: CPPC: Add ospm_nominal_freq attribute
>>   cpufreq: CPPC: Reflect the OSPM nominal in boost and limits
>>
>>  .../ABI/testing/sysfs-devices-system-cpu      |  31 ++++
>>  drivers/acpi/cppc_acpi.c                      |  44 +++++
>>  drivers/cpufreq/cppc_cpufreq.c                | 173 +++++++++++++++++-
>>  include/acpi/cppc_acpi.h                      |  10 +
>>  4 files changed, 248 insertions(+), 10 deletions(-)
>>
>> v7: https://lore.kernel.org/lkml/20260807214837.863209-1-sumitg@xxxxxxxxxx/
>> v6: https://lore.kernel.org/lkml/20260717215330.2215058-1-sumitg@xxxxxxxxxx/
>> v5: https://lore.kernel.org/lkml/20260615185934.2383514-1-sumitg@xxxxxxxxxx/
>> v4: https://lore.kernel.org/lkml/20260527194626.185286-1-sumitg@xxxxxxxxxx/
>> v3: https://lore.kernel.org/lkml/20260514194822.1841748-1-sumitg@xxxxxxxxxx/
>> v2: https://lore.kernel.org/lkml/20260430142430.755437-1-sumitg@xxxxxxxxxx/
>> v1: https://lore.kernel.org/lkml/20260427051823.280419-1-sumitg@xxxxxxxxxx/
>>
>> --
> 
> This would be good to go except that it doesn't apply for me on top of 7.4-rc1.
> 
> Can you please rebase it and resend, with Christian's tags included?
> 
> Thanks!

It's based on the hotplug/resume series, I guess Sumit could post them as a combined patchset