Re: [PATCH] O2int 0307041440 for 2.5.74-mm1

Wiktor Wodecki (wodecki@gmx.de)
Sun, 6 Jul 2003 14:36:35 +0200


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-16993-1057496119-0001-2
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Fri, Jul 04, 2003 at 02:59:08PM +1000, Con Kolivas wrote:
Content-Description: clearsigned data
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>=20
> Here is a patch against the current O1int patch in 2.5.74-mm1.
> Since the O1int didn't mean anything I thought I'd call this O2int.
>=20
> This one wont blow you away but tames those corner cases.
>=20
> Changes:
> The child penalty is set on 80% which means that tasks that wait on their=
=20
> children have children forking just on the edge of the interactive delta =
so=20
> they shouldn't starve their own children.
>=20
> The non linear sleep avg boost is scaled down slightly to prevent this=20
> particular boost from being capable of making a task highly interactive. =
This=20
> makes very new tasks less likely to have a little spurt of too high prior=
ity.
>=20
> Idle tasks now get their static priority over the full time they've been=
=20
> running rather than starting again at 1 second. This makes it harder for =
idle=20
> tasks to suddenly become highly interactive and _then_ fork an interactiv=
e=20
> bomb. Not sure on this one yet.
>=20
> The sched_exit penalty to parents of cpu hungry children is scaled accord=
ingly=20
> (was missed on the original conversion so works better now).
>=20
> Hysteresis on interactive buffer removed (was unecessary).
>=20
> Minor cleanup.
>=20
> Known issue remaining:
> Mozilla acts just like X in that it is mostly interactive but has bursts =
of=20
> heavy cpu activity so it gets the same bonus as X. However it makes X jer=
ky=20
> during it's heavy cpu activity, and might in some circumstances make audi=
o=20
> skip. Fixing this kills X smoothness as they seem very similar to the=20
> estimator. Still haven't sorted a workaround for this one but I'm working=
on=20
> it. Ingo's original timeslice granularity patch helps a little and may be=
=20
> worth resuscitating (and the desktop only people can change the granulari=
ty=20
> down to 10ms to satisfy their needs).
>=20
> Con

this one behaves worse than the previou one. When loading a big page
(such as www.heise.de or www.ebay.de) and moving the mouse in circles=20
it gets very jumpy.

--=20
Regards,

Wiktor Wodecki

--=_courier-16993-1057496119-0001-2
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

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

iD8DBQE/CBfT6SNaNRgsl4MRArVsAJ41mY0atWNFNAGMUx6aBbs/yz2YCQCfSzY6
6Pf5vKHV1YvCp5ZZbbyIXXM=
=HByo
-----END PGP SIGNATURE-----

--=_courier-16993-1057496119-0001-2--