diff options
| author | Nicola Mazzucato <nicola.mazzucato@arm.com> | 2020-06-17 12:43:32 +0300 |
|---|---|---|
| committer | Sudeep Holla <sudeep.holla@arm.com> | 2020-06-30 16:07:43 +0300 |
| commit | fb3571276b970cbe7b45ecdc762e92f3f305ad6d (patch) | |
| tree | 3d7067f3e3c9123b9541c7f5a55f74e6af77af37 /drivers/cpufreq/scmi-cpufreq.c | |
| parent | 1909872ff20fc378ec6a44ea1a2b2966d834e504 (diff) | |
| download | linux-fb3571276b970cbe7b45ecdc762e92f3f305ad6d.tar.xz | |
cpufreq: arm_scmi: Set fast_switch_possible conditionally
Currently the fast_switch_possible flag is set unconditionally to true.
Based on this, schedutil does not create a thread for frequency
switching and would always use the fast switch path.
However, if the platform does not support SCMI fast channel, we use
polling mode for SCMI message transfer. This may be possible only if
there is dedicated channel for DVFS and all operations are in polling
mode.
Update this by retrieving the fast_switch capability based on the
presence of fast channels in SCMI platform firmware.
Link: https://lore.kernel.org/r/20200617094332.8391-2-nicola.mazzucato@arm.com
Suggested-by: Lukasz Luba <lukasz.luba@arm.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Nicola Mazzucato <nicola.mazzucato@arm.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Diffstat (limited to 'drivers/cpufreq/scmi-cpufreq.c')
| -rw-r--r-- | drivers/cpufreq/scmi-cpufreq.c | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/drivers/cpufreq/scmi-cpufreq.c b/drivers/cpufreq/scmi-cpufreq.c index 61623e2ff149..1cf688fcb56b 100644 --- a/drivers/cpufreq/scmi-cpufreq.c +++ b/drivers/cpufreq/scmi-cpufreq.c @@ -198,7 +198,8 @@ static int scmi_cpufreq_init(struct cpufreq_policy *policy) policy->cpuinfo.transition_latency = latency; - policy->fast_switch_possible = true; + policy->fast_switch_possible = + handle->perf_ops->fast_switch_possible(handle, cpu_dev); em_register_perf_domain(policy->cpus, nr_opp, &em_cb); |