Re: [patch] "HT scheduler", sched-2.5.63-B3

Martin Waitz (tali@admingilde.org)
Thu, 6 Mar 2003 23:03:07 +0100


This is a MIME-formatted message. If you see this text it means that your
E-mail software does not support MIME-formatted messages.

--=_courier-3578-1046988258-0001-2
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

hi :)

On Wed, Mar 05, 2003 at 07:20:34PM -0800, Linus Torvalds wrote:
> How about something more like this (yeah, untested, but you get the idea)=
:=20
> the person who wakes up an interactive task gets the interactivity bonus=
=20
> if the interactive task is already maxed out. I dunno how well this plays=
=20
> with the X server, but assuming most clients use UNIX domain sockets, the=
=20
> wake-ups _should_ be synchronous, so it should work well to say "waker=20
> gets bonus".

i used a similar method to correctly account resource usage
(cpu,energy,..) of processes in my diploma thesis:
work done by a sever (e.g. X) is accounted to the current client,
giving more resources to the server
http://admingilde.org/~martin/papers/

implementation is working but far from being mergeable...

RE: the patch, i think using sleep_avg is a wrong metric from the
beginning.

in addition, timeslices should be shortest for high priority processes
(depending on dynamic prio, not static)

but these are of course just simple statements and i don't have
a patch that makes a really good scheduler :(

--=20
CU, / Friedrich-Alexander University Erlangen, Germany
Martin Waitz // [Tali on IRCnet] [tali.home.pages.de] _________
______________/// - - - - - - - - - - - - - - - - - - - - ///
dies ist eine manuell generierte mail, sie beinhaltet //
tippfehler und ist auch ohne grossbuchstaben gueltig. /
-
Wer bereit ist, grundlegende Freiheiten aufzugeben, um sich=20
kurzfristige Sicherheit zu verschaffen, der hat weder Freiheit=20
noch Sicherheit verdient. Benjamin Franklin (1706 - 1790)

--=_courier-3578-1046988258-0001-2
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+Z8Wbj/Eaxd/oD7IRAhUhAJ45pIxUq6hKfZabbckI09Js9npo5gCeN2n1
nu0YBh8gsp5arD1uNrl0/uI=
=v87b
-----END PGP SIGNATURE-----

--=_courier-3578-1046988258-0001-2--