PELT initial task load and wake_up_new_task()

[Steve Muckle]

In init_entity_runnable_average() the last_update_time is initialized to
zero. The task is given max load and utilization as a pessimistic
initial estimate.

But if in wake_up_new_task() the task is placed on a CPU other than
where it was created, __update_load_avg() will be called via
set_task_cpu() -> migrate_task_rq_fair() -> remove_entity_load_avg().

Since last_update_time is zero the delta will be huge and the task's
load will be entirely decayed away before it is enqueued at the
destination CPU.

If last_update_time is initialized to cfs_rq_clock_task() the load will
not go away, but it will also then be subtracted from the original CPU
in remove_entity_load_avg() if the task is placed on a different CPU,
which is bad since it was never added there before.

Thinking about this more it seemed questionable to treat the assignment
of a task to a new CPU in wake_up_new_task() as a migration given that
the task has never executed previously. Would it make sense to call
__set_task_cpu() there instead of set_task_cpu()?

thanks,
Steve
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/