Re: Slowdown due to threads bouncing between HT cores

[Andi Kleen]

"Steinar H. Gunderson" <sgunderson@xxxxxxxxxxx> writes:
>
> So, benchmark:
>
>  - Default: 13266 kN/sec
>  - Change from ondemand to performance on all cores: 14600 kN/sec
>  - taskset -c 0-19 (locking affinity to only one set of hyperthreads):
>    17512 kN/sec

iirc the main reasons for unbound threads migrating away are:

- something else gets scheduled on these logical CPUs, so
the scheduler tries to balance to run queue lengths

You could check that with perf timechart or perf sched record/map
or kernelshark.

- there is some IO or communication which causes wakeup affinity.

You could try disabling WAKEUP_PREEMPTION or NEXT_BUDDY in 
/sys/kernel/debug/sched_features

-Andi
-- 
ak@xxxxxxxxxxxxxxx -- Speaking for myself only
--
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/