cpufreq: intel_pstate: Avoid missing HWP max limit updates with powersave governor
Get series with b4
b4 am 2183878.gTFULuzKx9@kreacherVersions newest first · linked by author and title
- v2 Latest4 patches 11 repliesIn review
- v12 patches 3 repliesSuperseded
Patches in v2 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/4 | cpufreq: Introduce CPUFREQ_NEED_UPDATE_LIMITS driver flagdrivers/ include/ | +11 | 1 reply | |
| 2/4 | cpufreq: intel_pstate: Avoid missing HWP max updates in passive modedrivers/ | +6 | 2 replies | |
| 3/4 | cpufreq: Introduce cpufreq_driver_test_flags()drivers/ include/ | +13 | 0 replies | |
| 4/4 | cpufreq: schedutil: Always call drvier if need_freq_update is setkernel/ | +6 | 3 replies | |
| 4/4 | cpufreq: schedutil: Always call driver if need_freq_update is setkernel/ | +6 | 3 replies | |
| 4/4 | cpufreq: schedutil: Always call driver if CPUFREQ_NEED_UPDATE_LIMITS is setkernel/ | +4 | 2 replies |
Reviews trailers per patch
5 of 6 patches have a Reviewed-by or Acked-by.
| Patch | Viresh Kumar | Zhang Rui |
|---|---|---|
| 1/4 | A-b | |
| 2/4 | A-b | T-b |
| 3/4 | ||
| 4/4 | A-b | T-b T-b T-b |
| 4/4 | A-b | T-b T-b T-b |
| 4/4 | A-b | T-b T-b T-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH v2 0/4] cpufreq: intel_pstate: Avoid missing HWP max limit updates with powersave governor
Hi All,
There is a problem in intel_pstate that if it works in the passive mode with HWP enabled, changing the policy max frequency may not cause the HWP max limit to be updated accordingly which is quite confusing and may be incorrect.
That happens because of two checks, one in the cpufreq core and one in the driver itself, that are there to avoid unnecessary HW/FW updates when the current frequency doesn't change. Of course, that is the case when the policy max limit changes under the "powersave" governor (which sets the current frequency to the policy min limit) and in that particular case, the checks turn out to be harmful.
There is also an analogous problem in the schedutil governor that prevents driver callbacks from being invoked if the target frequency doesn't change and which also affects intel_pstate in the passive mode with HWP enabled (see the changelog of patch [4/4]).
The v2 addresses some review comments from Viresh and adds patches [3-4/4] to address the schedutil issue.
The cleanup posted as the [2/2] previously will be applied independently and it is not included in the v2.
Thanks!