[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
GCC 2.5.8
>main reason -- "Geneva uses ccoperative multitasking. This is the
>method used by MAC's system 7 and MSDOS Windows. .... MultiTOS is a
>pre-emtive multitasking. It severly controls each application's access
>to the CPU ......" [more nonsense deleted].
Now I've heard everything :-)
The overhead in any preemptive multitasking system is of course going
to be greater than in a cooperative system. MiNT's process scheduling
isn't so hairy that the overhead really becomes an issue.
>maybe multiTOS is a dog, i dunno, never used it. but MiNT itself, as we
>all know is very acceptable (i use it with a shell usually,
>and/or with toswin sometimes).
I've never used MultiTOS either...I have seen Geneva used, and for
what it does it seems really nice. It does have some serious problems
with some software (caused by the cooperative multitasking praised so
highly in that article :-) ...for example the friend that showed it to
me cannot use his favorite raytracing software with geneva: I assume
it changes context only on kernel entry, and this software doesn't do
any traps while calculating the display... Of course MiNT would have
similar troubles with a program that decided to compute a huge
fibonacci sequence in Supervisor mode :-)
--
entropy -- it's not just a good idea, it's the second law.
Personal mail: entropy@gnu.ai.mit.edu
MiNT library mail: entropy@terminator.rs.itd.umich.edu
"what do you have against octal?" -jrb