@@ -216,6 +216,7 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
struct cpufreq_policy *policy = sg_policy->policy;
unsigned long util, max;
unsigned int next_f;
+ bool rt_mode;
bool busy;
/* Skip updates generated by sugov kthreads */
@@ -230,7 +231,15 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
busy = sugov_cpu_is_busy(sg_cpu);
- if (flags & SCHED_CPUFREQ_RT_DL) {
+ /*
+ * While RT/DL tasks are running we do not want FAIR tasks to
+ * overvrite this CPU's flags, still we can update utilization and
+ * frequency (if required/possible) to be fair with these tasks.
+ */
+ rt_mode = task_has_dl_policy(current) ||
+ task_has_rt_policy(current) ||
+ (flags & SCHED_CPUFREQ_RT_DL);
+ if (rt_mode) {
next_f = policy->cpuinfo.max_freq;
} else {
sugov_get_util(&util, &max);
@@ -293,6 +302,7 @@ static void sugov_update_shared(struct update_util_data *hook, u64 time,
struct sugov_policy *sg_policy = sg_cpu->sg_policy;
unsigned long util, max;
unsigned int next_f;
+ bool rt_mode;
/* Skip updates generated by sugov kthreads */
if (unlikely(current == sg_policy->thread))
@@ -310,17 +320,27 @@ static void sugov_update_shared(struct update_util_data *hook, u64 time,
sg_cpu->flags = 0;
goto done;
}
- sg_cpu->flags = flags;
+
+ /*
+ * While RT/DL tasks are running we do not want FAIR tasks to
+ * overwrite this CPU's flags, still we can update utilization and
+ * frequency (if required/possible) to be fair with these tasks.
+ */
+ rt_mode = task_has_dl_policy(current) ||
+ task_has_rt_policy(current) ||
+ (flags & SCHED_CPUFREQ_RT_DL);
+ if (rt_mode)
+ sg_cpu->flags |= flags;
+ else
+ sg_cpu->flags = flags;
sugov_set_iowait_boost(sg_cpu, time, flags);
sg_cpu->last_update = time;
if (sugov_should_update_freq(sg_policy, time)) {
- if (flags & SCHED_CPUFREQ_RT_DL)
- next_f = sg_policy->policy->cpuinfo.max_freq;
- else
- next_f = sugov_next_freq_shared(sg_cpu, time);
-
+ next_f = rt_mode
+ ? sg_policy->policy->cpuinfo.max_freq
+ : sugov_next_freq_shared(sg_cpu, time);
sugov_update_commit(sg_policy, time, next_f);
}
The policy in use for RT/DL tasks sets the maximum frequency when a task in these classes calls for a cpufreq_update_this_cpu(). However, the current implementation might cause a frequency drop while a RT/DL task is still running, just because for example a FAIR task wakes up and it's enqueued in the same CPU. This issue is due to the sg_cpu's flags being overwritten at each call of sugov_update_*. Thus, the wakeup of a FAIR task resets the flags and can trigger a frequency update thus affecting the currently running RT/DL task. This can be fixed, in shared frequency domains, by ORing (instead of overwriting) the new flag before triggering a frequency update. This grants to stay at least at the frequency requested by the RT/DL class, which is the maximum one for the time being. This patch does the flags aggregation in the schedutil governor, where it's easy to verify if we currently have RT/DL workload on a CPU. This approach is aligned with the current schedutil API design where the core scheduler does not interact directly with schedutil, while instead are the scheduling classes which call directly into the policy via cpufreq_update_{util,this_cpu}. Thus, it makes more sense to have flags aggregation in the schedutil code instead of the core scheduler. Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Viresh Kumar <viresh.kumar@linaro.org> Cc: Steve Muckle <smuckle.linux@gmail.com> Cc: linux-kernel@vger.kernel.org Cc: linux-pm@vger.kernel.org --- Changes from v1: - use "current" to check for RT/DL tasks (PeterZ) --- kernel/sched/cpufreq_schedutil.c | 34 +++++++++++++++++++++++++++------- 1 file changed, 27 insertions(+), 7 deletions(-)