> Anyway i feel quite well that it basically isn't my problem but a
> serious OS problem. They're smart, but not smart enuf
> to understand that i'm running a parallel program where it's very
> important that seemingly idle processes get a big deal of systemtime,
> instead of waiting for a second before getting it.
You can drop hints in a posix.4 compliant system using sched_yield() which
basically says "I can't go forward, run someone else, but I want to get running
again too straight afterwards". Netscape uses buzz locks that spin briefly
then yield. The code for that is in the Mozilla source.
To do it right on a multiprocessor machine you want a scheduler that is aware
of such issues. They exist, but I dont know a lot about them
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
