Re: RasterMan on linux and threads

Rik van Riel (riel@nl.linux.org)
Sat, 18 Dec 1999 20:39:33 +0100 (CET)


On Sat, 18 Dec 1999 michael_schulz@public.uni-hamburg.de wrote:

> But i could think of a situation in a boss/worker setup (keeping
> workers in a pool), where some workers got scheduled on different cpus
> and then stay assigned to their cpus during several requests.

This is exactly what happens on Linux, except
that the switching can also happen mid-request
(but it usually won't happen very often).

There still is the issue of cache flushing
though. Keeping the threads on their own
CPU does not alleviate the program of pushing
the data from one thread back to the other
(or from one CPU to the other).

regards,

Rik

--
The Internet is not a network of computers. It is a network
of people. That is its real strength.

- 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/