On 01/19/21 15:35, Quentin Perret wrote: > On Tuesday 19 Jan 2021 at 12:07:55 (+0000), Qais Yousef wrote: > > If the task is pinned to a cpu, setting the misfit status means that > > we'll unnecessarily continuously attempt to migrate the task but fail. > > > > This continuous failure will cause the balance_interval to increase to > > a high value, and eventually cause unnecessary significant delays in > > balancing the system when real imbalance happens. > > > > Caught while testing uclamp where rt-app calibration loop was pinned to > > cpu 0, shortly after which we spawn another task with high util_clamp > > value. The task was failing to migrate after over 40ms of runtime due to > > balance_interval unnecessary expanded to a very high value from the > > calibration loop. > > > > Not done here, but it could be useful to extend the check for pinning to > > verify that the affinity of the task has a cpu that fits. We could end > > up in a similar situation otherwise. > > Do you mean failing the sched_setaffinity syscall if e.g. the task > has a min clamp that is higher than the capacity of the CPUs to which it > will be pinned? If so, I'm not sure if we really want that.
No. In Android for instance, I'm worried a background task affined to little cores that has a utilization > capacity_of(little) will trigger the same problem. It'll be affined to more than just 1 cpu, but none of the little cpus will actually fit. Makes sense? > But this patch makes sense on its own for sure, so: > > Reviewed-by: Quentin Perret <qper...@google.com> Thanks -- Qais Yousef