Re: [PATCH] Preemption Latency Measurement Tool

From: Robert Love (rml@tech9.net)
Date: Thu Sep 20 2001 - 16:09:17 EST


On Thu, 2001-09-20 at 03:57, Dieter Nützel wrote:
> You've forgotten a one liner.
>
> #include <linux/locks.h>
> +#include <linux/compiler.h>
>
> But this is not enough. Even with reniced artsd (-20).
> Some shorter hiccups (0.5~1 sec).

Note (I am repeated myself from an email I just sent) that the
conditional schedule won't show better results if
current->need_reschedule is unset, since preemption won't be enabled. I
need to add explicit support to the preemption-test patch for this.

So you may see some better results, but just one time the condition
schedule does not occur, you will see the worst result in
/proc/latencytimes -- remembers its the 20 worst (perhaps we need
average or total latency, too?)

Now, with all that said, you should _see_ an improvement with this
patch. You say short hiccups. Some? All? How much better is it?

-- 
Robert M. Love
rml at ufl.edu
rml at tech9.net

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



This archive was generated by hypermail 2b29 : Sun Sep 23 2001 - 21:00:40 EST