cpufreq: governor: Simplify performance and powersave governors

The performance and powersave cpufreq governors handle the
CPUFREQ_GOV_START event in the same way as CPUFREQ_GOV_LIMITS.
However, the cpufreq core always invokes cpufreq_governor() with the
event argument equal to CPUFREQ_GOV_LIMITS right after invoking it with
event equal to CPUFREQ_GOV_START.  As a result, for both the governors
in question, __cpufreq_driver_target() is executed twice in a row
with the same arguments which is not useful.

For this reason, simplify the performance and powersave governors
to handle the CPUFREQ_GOV_LIMITS event only as that's going to be
sufficient for the governor start too.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>

drivers/cpufreq/cpufreq_performance.c

@@ -20,10 +20,8 @@ static int cpufreq_governor_performance(struct cpufreq_policy *policy,
 					unsigned int event)
 {
 	switch (event) {
-	case CPUFREQ_GOV_START:
 	case CPUFREQ_GOV_LIMITS:
-		pr_debug("setting to %u kHz because of event %u\n",
-						policy->max, event);
+		pr_debug("setting to %u kHz\n", policy->max);
 		__cpufreq_driver_target(policy, policy->max,
 						CPUFREQ_RELATION_H);
 		break;

drivers/cpufreq/cpufreq_powersave.c

@@ -20,10 +20,8 @@ static int cpufreq_governor_powersave(struct cpufreq_policy *policy,
 					unsigned int event)
 {
 	switch (event) {
-	case CPUFREQ_GOV_START:
 	case CPUFREQ_GOV_LIMITS:
-		pr_debug("setting to %u kHz because of event %u\n",
-							policy->min, event);
+		pr_debug("setting to %u kHz\n", policy->min);
 		__cpufreq_driver_target(policy, policy->min,
 						CPUFREQ_RELATION_L);
 		break;