Re: [PATCH v3 5/6] platform/x86: bitland-mifs-wmi: Add per-machine ops table

Next message: Bradley Morgan: "Re: [PATCH v2 2/4] sysctl: Reject uint arrays before calling the general proc_vec"
Previous message: pr-tracker-bot: "Re: [GIT PULL] OpenRISC fix for 7.2"
In reply to: Mingyou Chen: "[PATCH v3 5/6] platform/x86: bitland-mifs-wmi: Add per-machine ops table"
Next in thread: Mingyou Chen: "[PATCH v3 6/6] platform/x86: bitland-mifs-wmi: Add Redmi laptop support"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

From: Rsplwe

Date: Sat Aug 15 2026 - 12:01:46 EST

Hi,

Regarding the handling of `has_full_speed`: when a machine does not have a Full Speed mode, `PLATFORM_PROFILE_PERFORMANCE` should still be exposed if the machine supports the Performance mode.

In other words, the absence of Full Speed should not make `BALANCED_PERFORMANCE` the highest available profile. For machines without Full Speed, the expected profile choices should be: `low-power` / `balanced` / `performance`

where `PLATFORM_PROFILE_PERFORMANCE` maps to the machine's highest available performance mode.

Therefore, `has_full_speed` should probably distinguish whether the machine has a separate Full Speed mode, while the profile choice logic should ensure that `PLATFORM_PROFILE_PERFORMANCE` remains available when Performance is the highest supported mode.