On 18-02-16, 02:28, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <rafael.j.wyso...@intel.com>
>
> The ondemand_powersave_bias_init() function used for resetting data
> fields related to the powersave bias tunable of the ondemand governor
> works by walking all of the online CPUs in the system and updating the
> od_cpu_dbs_info_s structures for all of them.
>
> However, if governor tunables are per policy, the update should not
> touch the CPUs that are not associated with the given dbs_data.
>
> Moreover, since the data fields in question are only ever used for
> policy->cpu in each policy governed by ondemand, the update can be
> limited to those specific CPUs.
Now that I am looking at struct **_cpu_dbs_info_s definitions, I think
we need to change the design of the structures a bit. Only cdbs is
something that is per-cpu and everything else is per-policy. And so we
will be able to move the per-cpu structures present in individual
governors into cpufreq_governor.c and save some memory.
> Rework the code to take the above observations into account.
>
> Signed-off-by: Rafael J. Wysocki <rafael.j.wyso...@intel.com>
> ---
> drivers/cpufreq/cpufreq_ondemand.c | 30 +++++++++++++-----------------
> 1 file changed, 13 insertions(+), 17 deletions(-)
>
> Index: linux-pm/drivers/cpufreq/cpufreq_ondemand.c
> ===================================================================
> --- linux-pm.orig/drivers/cpufreq/cpufreq_ondemand.c
> +++ linux-pm/drivers/cpufreq/cpufreq_ondemand.c
> @@ -34,14 +34,6 @@ static struct od_ops od_ops;
>
> static unsigned int default_powersave_bias;
>
> -static void ondemand_powersave_bias_init_cpu(int cpu)
> -{
> - struct od_cpu_dbs_info_s *dbs_info = &per_cpu(od_cpu_dbs_info, cpu);
> -
> - dbs_info->freq_table = cpufreq_frequency_get_table(cpu);
> - dbs_info->freq_lo = 0;
> -}
> -
> /*
> * Not all CPUs want IO time to be accounted as busy; this depends on how
> * efficient idling at a higher frequency/voltage is.
> @@ -120,12 +112,13 @@ static unsigned int generic_powersave_bi
> return freq_hi;
> }
>
> -static void ondemand_powersave_bias_init(void)
> +static void ondemand_powersave_bias_init(struct cpufreq_policy *policy)
> {
> - int i;
> - for_each_online_cpu(i) {
> - ondemand_powersave_bias_init_cpu(i);
> - }
> + unsigned int cpu = policy->cpu;
> + struct od_cpu_dbs_info_s *dbs_info = &per_cpu(od_cpu_dbs_info, cpu);
> +
> + dbs_info->freq_table = cpufreq_frequency_get_table(cpu);
> + dbs_info->freq_lo = 0;
> }
>
> static void dbs_freq_increase(struct cpufreq_policy *policy, unsigned int
> freq)
> @@ -306,6 +299,7 @@ static ssize_t store_powersave_bias(stru
> size_t count)
> {
> struct od_dbs_tuners *od_tuners = dbs_data->tuners;
> + struct policy_dbs_info *policy_dbs;
> unsigned int input;
> int ret;
> ret = sscanf(buf, "%u", &input);
> @@ -317,7 +311,10 @@ static ssize_t store_powersave_bias(stru
> input = 1000;
>
> od_tuners->powersave_bias = input;
> - ondemand_powersave_bias_init();
> +
> + list_for_each_entry(policy_dbs, &dbs_data->policy_dbs_list, list)
> + ondemand_powersave_bias_init(policy_dbs->policy);
> +
> return count;
> }
>
> @@ -398,11 +395,10 @@ static void od_exit(struct dbs_data *dbs
>
> static void od_start(struct cpufreq_policy *policy)
> {
> - unsigned int cpu = policy->cpu;
> - struct od_cpu_dbs_info_s *dbs_info = &per_cpu(od_cpu_dbs_info, cpu);
> + struct od_cpu_dbs_info_s *dbs_info = &per_cpu(od_cpu_dbs_info,
> policy->cpu);
>
> dbs_info->sample_type = OD_NORMAL_SAMPLE;
> - ondemand_powersave_bias_init_cpu(cpu);
> + ondemand_powersave_bias_init(policy);
> }
>
> define_get_cpu_dbs_routines(od_cpu_dbs_info);
I agree with the patch, but because we have all these fields per-cpu,
I feel a bit scared while depending on policy->cpu for everything.
Acked-by: Viresh Kumar <viresh.ku...@linaro.org>
--
viresh
