From: Rafael J. Wysocki <rafael.j.wyso...@intel.com> None of the cpufreq governors currently in the tree will ever fail an invocation of the ->governor() callback with the event argument equal to CPUFREQ_GOV_LIMITS (unless invoked with incorrect arguments which doesn't matter anyway) and had it ever failed, the result of it wouldn't have been very clean.
For this reason, rearrange the code in the core to ignore the return value of cpufreq_governor() when called with event equal to CPUFREQ_GOV_LIMITS. Signed-off-by: Rafael J. Wysocki <rafael.j.wyso...@intel.com> --- drivers/cpufreq/cpufreq.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) Index: linux-pm/drivers/cpufreq/cpufreq.c =================================================================== --- linux-pm.orig/drivers/cpufreq/cpufreq.c +++ linux-pm/drivers/cpufreq/cpufreq.c @@ -2054,7 +2054,11 @@ static int cpufreq_start_governor(struct cpufreq_update_current_freq(policy); ret = cpufreq_governor(policy, CPUFREQ_GOV_START); - return ret ? ret : cpufreq_governor(policy, CPUFREQ_GOV_LIMITS); + if (ret) + return ret; + + cpufreq_governor(policy, CPUFREQ_GOV_LIMITS); + return 0; } int cpufreq_register_governor(struct cpufreq_governor *governor) @@ -2195,7 +2199,8 @@ static int cpufreq_set_policy(struct cpu if (new_policy->governor == policy->governor) { pr_debug("cpufreq: governor limits update\n"); - return cpufreq_governor(policy, CPUFREQ_GOV_LIMITS); + cpufreq_governor(policy, CPUFREQ_GOV_LIMITS); + return 0; } pr_debug("governor switch\n");