lkml.org 
[lkml]   [2000]   [Oct]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: this_slice in kernel/sched.c?
On Tue, 24 Oct 2000, Rasmus Andersen wrote:

> Reading through kernel/sched.c I came across this block (on line 597):
>
> {
> cycles_t t, this_slice;
>
> t = get_cycles();
> this_slice = t - sched_data->last_schedule;
> sched_data->last_schedule = t;
> }
>
> It seems to me that this_slice is somewhat unused in its scope and
> therefore I propose that the block is replaced by

> The fact that the scheduler probably have been examined many
> times by people far more knowledgeable that me makes the
> probability that this is wrong pretty big. In that case I would
> like to know what use this_slice has, if somebody would care to
> share.

this_slice was used in reschedule_idle(), to determine
whether it was worth it to interrupt the currently running
task or if it is better to wait a while until the currently
running task voluntarily reschedules.

The performance report by the Fujitsu guy (that was posted
here today) suggests that we really want to use the per-task
average time slice again to prevent cache thrashing...

regards,

Rik
--
"What you're running that piece of shit Gnome?!?!"
-- Miguel de Icaza, UKUUG 2000

http://www.conectiva.com/ http://www.surriel.com/

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

\
 
 \ /
  Last update: 2005-03-22 12:41    [W:0.076 / U:2.668 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site