diff options
| author | Aaron Tian <atian@nvidia.com> | 2021-01-08 03:10:53 -0500 |
|---|---|---|
| committer | mobile promotions <svcmobile_promotions@nvidia.com> | 2021-04-14 13:39:55 -0400 |
| commit | 2caa44ddd85627ab3c5d1d86a396b373221bf4e4 (patch) | |
| tree | 79f90879ac0a0c691427dbabd98c3702f3446ee7 /include/linux | |
| parent | 6ea5dc903a246194665d65026b4f4c46181937db (diff) | |
cpufreq: t194: set req. to Fmin when CPU offline
Since offline CPU should not join cluster frequency
judgement, set cpufreq request to Fmin during cpu_down
to let cluster frequency judgement logic won't take
offline CPUs into account.
During cpu_up the cpufreq request value need to restored
as latest requested level (from cpufreq governor).
Add an per-CPU array 'latest_freq_req' to memorize
latest request level.
Register CPUHP_AP_TEGRA_CPUFREQ cpuhp callback
to do related ndiv operations due to following reasons
- The ndiv of a CPU can only be accessed by the same CPU.
- ndiv restore need to be handled as soon as possible
since cluster PLL frequency = Fmin during cpu_up() of
1st core in a cluster until ndiv value restored.
Bug 200610512
Change-Id: I60fe3f3aca58f51357b1e2a10a639427049b8b67
Signed-off-by: Aaron Tian <atian@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2467766
(cherry picked from commit e2cb988601c600d3f5c4fdbe19821bc1392dfbce)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2513653
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Diffstat (limited to 'include/linux')
0 files changed, 0 insertions, 0 deletions
