Philip Blundell wrote:
>
> >Yes, it does, but only if you run the process with SCHED_RR or
> >SCHED_FIFO. And this is dangerous. If the process hangs (maybe because
> >of a buggy endless loop), wo can't even kill it, because the process
> >will use up all of your CPU, and you cannot type something (unless you
> >have a shell running with SCHED_RR and a higher static priority than the
> >hanging process).
>
> Well, you could introduce a new scheduling policy that has the realtime
> behaviour for nanosleep but behaves like SCHED_OTHER in all other respects.
> That seems like a better solution than adding wacky stuff to ppdev.
This would be the way I'd prefer. But will Linus do this? Will someone
ask him?
> Or, run as SCHED_OTHER most of the time but promote yourself to SCHED_RR when
> you are actually doing port I/O. That should remove most of the risk from
> infinite loops caused by bugs.
Another possibility. As lcd4linux does port I/O every 100 msec, I don't
know how this would affect the performance of the whole system and
especially the scheduler if I chenge scheduling mode all the time.
bye, Michael
-- netWorks Vox: +43 316 692396 Michael Reinelt Fax: +43 316 692343 Geisslergasse 4 GSM: +43 676 3079941 A-8045 Graz, Austria e-mail: reinelt@eunet.at-- To unsubscribe, send mail to: linux-parport-request@torque.net -- -- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2b29 : Mon Mar 19 2001 - 02:54:21 EST