lkml.org 
[lkml]   [1998]   [Dec]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Increase number of tasks
On Tue, 1 Dec 1998, MOLNAR Ingo wrote:
> On Tue, 1 Dec 1998, Rik van Riel wrote:
>
> > help since the CPU spent on the standard scheduler is
> > dependant on NR_TASKS [...]
>
> i find this comment disturbingly inaccurate. first, it's not
> dependent on NR_TASKS, but nr_tasks, which is different. (even if we
> are talking about many processes, the difference should be
> mentioned)

OK, this certainly is quite a difference for normal systems.

> but more importantly, the dependency is very small, and only scales
> with the number of timeslices expired. (ie. constant and independent
> of the number of schedules per second!

Yes, this is different from my scheduler bigpatch. In my
patch the overhead is (somewhat) dependant on the number
of processes that enter and leave the runqueue per second.

The overhead from timeslice expiration can be ignored
for all purposes since it is almost constant in my patch.

> Third, it's a feature which makes sleeping tasks gain timeslices
> logaritmically with time.

Which is too slow when there are more than about 5 running
tasks in the system. But that's another story :)

cheers,

Rik -- now completely used to dvorak kbd layout...
+-------------------------------------------------------------------+
| Linux memory management tour guide. H.H.vanRiel@phys.uu.nl |
| Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ |
+-------------------------------------------------------------------+


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

\
 
 \ /
  Last update: 2005-03-22 13:45    [W:0.076 / U:0.156 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site