All of lore.kernel.org
 help / color / mirror / Atom feed
* [LTP] numa01
@ 2014-03-31  7:59 Stanislav Kholmanskikh
  2014-03-31 12:31 ` chrubis
  0 siblings, 1 reply; 2+ messages in thread
From: Stanislav Kholmanskikh @ 2014-03-31  7:59 UTC (permalink / raw)
  To: ltp-list; +Cc: Vasily Isaenko

Hi!

numa01.sh does something like:
  * numastat > /tmp/before
  * starts a binary with the specified memory policy using numactl (or a 
like):
    numactl --interleave=all get_some_memory_with_malloc_and_write_it
  * `sleep` for few seconds
  * numastat > /tmp/after
  * compares /tmp/before and /tmp/after to check that the numa policy 
was applied the right way

But the problem is that on a host with many NUMA nodes (8) the process 
of updating that numastats statistics takes some time. Even 10 seconds 
may be not enough. Therefore the test fails.

Does anybody know if there is a direct or indirect way to force the 
kernel to update the NUMA statistics?

Thank you.

------------------------------------------------------------------------------
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [LTP] numa01
  2014-03-31  7:59 [LTP] numa01 Stanislav Kholmanskikh
@ 2014-03-31 12:31 ` chrubis
  0 siblings, 0 replies; 2+ messages in thread
From: chrubis @ 2014-03-31 12:31 UTC (permalink / raw)
  To: Stanislav Kholmanskikh; +Cc: Vasily Isaenko, ltp-list

Hi!
> numa01.sh does something like:
>   * numastat > /tmp/before
>   * starts a binary with the specified memory policy using numactl (or a 
> like):
>     numactl --interleave=all get_some_memory_with_malloc_and_write_it
>   * `sleep` for few seconds
>   * numastat > /tmp/after
>   * compares /tmp/before and /tmp/after to check that the numa policy 
> was applied the right way
> 
> But the problem is that on a host with many NUMA nodes (8) the process 
> of updating that numastats statistics takes some time. Even 10 seconds 
> may be not enough. Therefore the test fails.
> 
> Does anybody know if there is a direct or indirect way to force the 
> kernel to update the NUMA statistics?

I don't think you can force the statistics to update. I guess that all
we can do is to poll the statistics and wait until it changes the way we
expect (with some timeout).

And the best soulution would be to ask kernel memory management
developers, they should know better (feel free to CC LTP ML).

-- 
Cyril Hrubis
chrubis@suse.cz

------------------------------------------------------------------------------
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2014-03-31 12:32 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-03-31  7:59 [LTP] numa01 Stanislav Kholmanskikh
2014-03-31 12:31 ` chrubis

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.