Due to commit 1b600da51073 ("PM: EM: Optimize em_cpu_energy() and remove division"), the logic for energy consumption calculation has been modified. The actual calculation of cost is 10 * power * max_frequency / frequency instead of power * max_frequency / frequency. Therefore, the comment for cost has been updated to reflect the correct content.
Showing the changes to include/linux (1 of 1 file). Show all
Unified · Split
| @@ -18,7 +18,7 @@ · 17 unmodified lines | |||
| 18 | * @power: The power consumed at this level (by 1 CPU or by a registered | 18 | * @power: The power consumed at this level (by 1 CPU or by a registered |
| 19 | * device). It can be a total power: static and dynamic. | 19 | * device). It can be a total power: static and dynamic. |
| 20 | * @cost: The cost coefficient associated with this level, used during | 20 | * @cost: The cost coefficient associated with this level, used during |
| 21 | * energy calculation. Equal to: power * max_frequency / frequency | 21 | * energy calculation. Equal to: 10 * power * max_frequency / frequency |
| 22 | * @flags: see "em_perf_state flags" description below. | 22 | * @flags: see "em_perf_state flags" description below. |
| 23 | */ | 23 | */ |
| 24 | struct em_perf_state { | 24 | struct em_perf_state { |