I also coded a simple method for locking sched_realtime processes on a
single processor.. It was trivial to add and it worked, but my interface
didn't allow me to pick which cpu it got locked on.. :)
On Thu, 9 Jul 1998, Perry Harrington wrote:
> Adding processor affinity should be straightforward. I believe that Linux
> only has 1 process queue right now. The most performance oriented method to
> do this would be to move the process to a processor specific queue. Reality
> dictates that you will incurr more contention on the process queue with the
> more processors you have.
>
> --Perry
>
> >
> > : Does Mr McVoy by any chance happen to have an "is L2 cache shared" detector
> > : lurking in the lmbench and other magic ?
> >
> > Hmm. Not as such. I could probably make one. It would be a lot easier
> > if Linux had a pin() function which pinned a process to a processor.
> > SGI's interface looks like:
> >
> > sysmp(MP_MUSTRUN, cpuNumber);
> >
> > If I had that sucker, then I could pin two processes to two different
> > processors, timing how long it took to touch powers of two size arrays
> > with one process, then do the same thing with two processes. It shouldn't
> > make any difference until you get out to >L2$ sizes if the caches are
> > unshared.
> >
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to majordomo@vger.rutgers.edu
> >
>
>
> --
> Perry Harrington Linux rules all OSes. APSoft ()
> email: perry@apsoft.com Think Blue. /\
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu