On Fri, 04 Aug 2000, Jes Sorensen wrote:
> Compiling gigs of source code doesn't probe a thing.
No, showing otherwise just takes a single example. This counter proof is
still missing.
> I am beginning to wonder if you really know what you are doing. Do you
> actually know what -O6 does? Automatically unrolling loops and
> inlining functions behind your back is not necessarily something you
> want on a modern CPU.
That is exactly what I want, particularly do I want C++ crap inlined. I
know this bloats my code and eats up my memory, but it's the only way to
get reasonable performance out of some projects that I have; apart from
the fact that -O6 won't have much to inline when your object files are
small.
I certainly would not use -O6 for debugging, and I saw -O6 break code
that -O2 did not break, that would be the reason to debug gcc. If my
kernels (all 2.95.2 compiles) started to misbehave, I'd certainly see
that or get reports from the production machines, alas, there are none.
Instead, random processes (mostly cron) start to misbehave after an
extended amount of time, and that cron has NOT been compiled with gcc
2.95.2.
> I use 2.96 every day and I frequently report bugs in it, on the ia64!
2.96 is not a release version yet, so I don't care. It's nice however
that somebody is using it to find the bugs and get them squished so I
can have a stable gcc release some time in the future :-)
-- Matthias Andree- 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 : Mon Aug 07 2000 - 21:00:13 EST