David Whysong <dwhysong@physics.ucsb.edu> said:
[...]
> Alan, does it make sense, in ANY situation, to disable overcommit? Because
> AFAIK, a system without overcommit would always have to kill processes
> before an equivalent overcommitting system would.
In an non-overcommiting system you can be sure your process won't get
SIGSEGV or SIGBUS if it doesn't do anything wrong, and (as long as it is
careful to check and ask for the resources) won't get killed because
memory gets low.
OTOH, as Alan said, the cat might decide to play with the keyboard, or the
overworked sysadmin trip over the power cord, a disk might die or you get
an area-wide blackout. In my humble experience, these "accidents" are way
more probable than a correctely speced machine running totally out of
virtual memory.
> It seems to me that this whole overcommit debate is just avoiding the real
> problem, which is how to decide which tasks to kill.
That is important in any case. But it is much more fun to troll the list,
and to aim ye olde firebreather at everybody in sight than to code it up.
-- Horst von Brand vonbrand@sleipnir.valparaiso.cl Casilla 9G, Viņa del Mar, Chile +56 32 672616- 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 Mar 31 2000 - 21:00:14 EST