> It's the thin edge of the wedge. Soon you'll be getting people to say
> "we need read(2) to keep RT priority".
I was just getting to that... ;-)
It would be nice to have:
{
enter_critical_section();
read(something);
process(something);
write(something);
leave_critical_section();
}
Do not confuse *_critical_section() things with Win32 ones... :-)
There all between enter_critical_section() and leave_critical_section()
would be executed NOW. No pre-emption allowed between those points.
SCHED_FIFO keeps the control when we are in process(), but things can go
wrong with read or write. One solution would be to implement SCHED_FIFO such
way that only sched_yield() would be pre-emption point.
I get better performance with GNU Pth because there I can have some control
over "pre-emption" points. Interthread only of course.
I was banging my head to wall last week with WinNT, because making system
call was like saying "we are not busy now, go do what ever you like". When
you make five system calls one after another is like saying "we wan't to
sleep here forever" to OS... I just hope that I don't have to see that
behaviour in Linux ever.
- Jussi Laako
-- PGP key fingerprint: 161D 6FED 6A92 39E2 EB5B 39DD A4DE 63EB C216 1E4B Available at: ldap://certserver.pgp.com, http://keys.pgp.com:11371- 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/
This archive was generated by hypermail 2b29 : Fri Jul 07 2000 - 21:00:10 EST