On 01/27/2015 12:36 AM, Viresh Kumar wrote:
Earlier we used to find the 'policy' belonging to a cpu with the help of a per-cpu variable. And if 'cpu' passed to cpufreq_cpu_get() is bigger than 'nr_cpu_ids', it would have caused unpredictable issues as the per-cpu variable wouldn't have covered that value of 'cpu'. And so we had this check.
We traverse active-policy list to find policy for a cpu now. Even if 'cpu' passed to cpufreq_cpu_get() is an invalid number (i.e. greater than nr_cpu_ids), we will be able to manage it without any unpredictable behavior.
And so this check isn't required anymore. Get rid of it.
Signed-off-by: Viresh Kumar viresh.kumar@linaro.org
drivers/cpufreq/cpufreq.c | 3 --- 1 file changed, 3 deletions(-)
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 7f947287ba46..d9528046f651 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -172,9 +172,6 @@ struct cpufreq_policy *cpufreq_cpu_get(unsigned int cpu) struct cpufreq_policy *policy = NULL; unsigned long flags;
- if (cpu >= nr_cpu_ids)
return NULL;- if (!down_read_trylock(&cpufreq_rwsem)) return NULL;
You are getting rid of this check because you no longer use the per-cpu variable. So, just squash it with the previous patch that removes the per-cpu variable?
-Saravana