On 11-Nov 18:08, Peter Zijlstra wrote:
> On Mon, Oct 29, 2018 at 06:33:01PM +0000, Patrick Bellasi wrote:
> > When a util_max clamped task sleeps, its clamp constraints are removed
> > from the CPU. However, the blocked utilization on that CPU can still be
> > higher than the max clamp value enforced while that task was running.
> >
> > The release of a util_max clamp when a CPU is going to be idle could
> > thus allow unwanted CPU frequency increases while tasks are not
> > running. This can happen, for example, when a frequency update is
> > triggered from another CPU of the same frequency domain.
> > In this case, when we aggregate the utilization of all the CPUs in a
> > shared frequency domain, schedutil can still see the full not clamped
> > blocked utilization of all the CPUs and thus, eventually, increase the
> > frequency.
>
> > @@ -810,6 +811,28 @@ static inline void uclamp_cpu_update(struct rq *rq,
> > unsigned int clamp_id)
> > if (max_value >= SCHED_CAPACITY_SCALE)
> > break;
> > }
> > +
> > + /*
> > + * Just for the UCLAMP_MAX value, in case there are no RUNNABLE
> > + * task, we want to keep the CPU clamped to the last task's clamp
> > + * value. This is to avoid frequency spikes to MAX when one CPU, with
> > + * an high blocked utilization, sleeps and another CPU, in the same
> > + * frequency domain, do not see anymore the clamp on the first CPU.
> > + *
> > + * The UCLAMP_FLAG_IDLE is set whenever we detect, from the above
> > + * loop, that there are no more RUNNABLE taks on that CPU.
> > + * In this case we enforce the CPU util_max to that of the last
> > + * dequeued task.
> > + */
> > + if (max_value < 0) {
> > + if (clamp_id == UCLAMP_MAX) {
> > + rq->uclamp.flags |= UCLAMP_FLAG_IDLE;
> > + max_value = last_clamp_value;
> > + } else {
> > + max_value = uclamp_none(UCLAMP_MIN);
> > + }
> > + }
> > +
> > rq->uclamp.value[clamp_id] = max_value;
> > }
>
> *groan*, so it could be jet-lag, but I find the comment really hard to
> understand.
>
> Would not something like:
>
> /*
> * Avoid blocked utilization pushing up the frequency when we go
> * idle (which drops the max-clamp) by retaining the last known
> * max-clamp.
> */
>
> Be more clear?
It works: short and effective... will update in v6.
Thanks.
--
#include <best/regards.h>
Patrick Bellasi
