Re: [PATCH v2] cpufreq: CPPC: Correct desired_perf calculation

2016-10-13 Thread Hoan Tran
On Thu, Oct 13, 2016 at 2:12 PM, Rafael J. Wysocki wrote: > On Thu, Oct 13, 2016 at 9:37 PM, Prakash, Prashanth > wrote: >> On 10/13/2016 11:33 AM, Hoan Tran wrote: >>> The desired_perf is an abstract performance number. Its value should >>> be in the range of [lowest perf, highest perf] of CPPC.

Re: [PATCH v2] cpufreq: CPPC: Correct desired_perf calculation

2016-10-13 Thread Rafael J. Wysocki
On Thu, Oct 13, 2016 at 9:37 PM, Prakash, Prashanth wrote: > On 10/13/2016 11:33 AM, Hoan Tran wrote: >> The desired_perf is an abstract performance number. Its value should >> be in the range of [lowest perf, highest perf] of CPPC. >> The correct calculation is >> desired_perf = freq * cppc_hig

Re: [PATCH v2] cpufreq: CPPC: Correct desired_perf calculation

2016-10-13 Thread Prakash, Prashanth
On 10/13/2016 11:33 AM, Hoan Tran wrote: > The desired_perf is an abstract performance number. Its value should > be in the range of [lowest perf, highest perf] of CPPC. > The correct calculation is > desired_perf = freq * cppc_highest_perf / cppc_dmi_max_khz > > And cppc_cpufreq_set_target() ret

[PATCH v2] cpufreq: CPPC: Correct desired_perf calculation

2016-10-13 Thread Hoan Tran
The desired_perf is an abstract performance number. Its value should be in the range of [lowest perf, highest perf] of CPPC. The correct calculation is desired_perf = freq * cppc_highest_perf / cppc_dmi_max_khz And cppc_cpufreq_set_target() returns if desired_perf is exactly the same with the ol