> On Tue, Mar 16, 1999 at 04:54:43PM +1100, Richard Gooch wrote:
> > yodaiken@chelm.cs.nmt.edu writes:
> > > The POSIX "minimal" standard is actually very close to what RTL offers
> > > and is what some other "POSIX" rt systems like LynxOs seem to follow.
> > > The "multi" standard includes all the problematical stuff. How useful
> > > would it be to have "minimal" on its own?
> >
> > So what exactly are the "minimal" and "multi" standards?
>
> "minimal" is also called "single process". The system is required to
> provide threads, timers, and rt signals, plus semaphores.
> "multi" is what people think of: sched_setsched, mlock, +basic nonrt
> posix. It seems like LynxOS and VxWorks people insistet on a minimal
> standard so that they could claim POSIX conformance without adding all
> that junk.
Posix stuff is interesting if it improves inter-communications among realtime
tasks & linux processes ( shared mem, mbx, sem, sig, ....). And so trying to
keep Realtime behavior between both Worlds ( no priority inversion, multi
thread safe, fair context switch,...) .
This stuff could be used to develop ( and debug ) applications first on
plain linux and then depending on Architectural choices and realtime
constraints to migrate towards RTLinux.
It could also help to port legacy applications from OS to OS.
Hope this Help!
Patrick
>
>
> If you want, I can scan in an email you the key pages.
>
> -
> 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/
-
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/