Re: Real Time scheduler?

Kurt Garloff (K.Garloff@ping.de)
Thu, 11 Feb 1999 00:55:05 +0100


--a8Wt8u1KmwUX3Y2C
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable

On Wed, Feb 10, 1999 at 05:39:43AM -0700, Victor Yodaiken wrote:
> > The solution is PRIORITY INVERSION: If a high prio process is waiting f=
or
> > something a low prio porcess provides, the low prio process should
> > temporarily get the high prio.
>=20
> You mean "PRIORITY INHERITANCE"=20

Yes, of course. I'm sure, I wrote it down correctly, but to clear things, I
typed them in capitals. Obviously wrong. Arghh !

> ...
>=20
> And then consider how a kernel that depends on the operation of many daem=
on
> processes will continue to operate when users can introduce arbitrarily m=
any
> "RT" processes that can block daemons indefinitely.

RT processes make absolutely no sense, when there are many of them.
Maybe, only one or two or NUM_CPU should be allowed at the same time.

--=20
Kurt Garloff <kurt@garloff.de> [Dortmund, FRG]
Plasma physics, high perf. computing [Linux-ix86,-axp, DUX]
PGP key: see mailheader [Linux SCSI driver: DC390]

--a8Wt8u1KmwUX3Y2C
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: 2.6.3in

iQCVAwUBNsIcWRaQN/7O/JIVAQHsUwQAkMmB3AOhgheQvR+CO57t/l8MHG0VGAWO
mgN13dy7nyIc7fwB5Prx8PYC4fNGIkDNDmCK//PJVvzRwtEMo3GqrC3/0uo/htNd
uP0BXfqkqsiBc7rYXFpJMP+9e8pY2NTjK2R5mUycz/QwHiM96MxXDL1v/+soOep8
xMkL5EwwBv4=
=0PXq
-----END PGP SIGNATURE-----

--a8Wt8u1KmwUX3Y2C--

-
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/