Re: gcc -O2 on alpha miscompiles ifupdown?
On Fri, Jan 05, 2001 at 11:24:15PM -0800, Ron Farrer wrote:
>
> I haven't looked at this (yet?), but I would guess it to be more bugs
> in gcc 2.95 we are using. If you look at the gcc 2.96 changelog you'll
> see there are quite a few bug fixes, a lot of them on Alpha. We've
> been plauged with bad gcc optimization on Alpha for about as long as I
> can remember.
Truer words were never spoken :-P
> If no one knows for sure, I may have time in the next couple of days
> to play with it and see. Chris Chimelis (chris@debian.org) is the one
> to talk to about gcc bugs on Alpha. I would say either don't use -O2
> (duh) or use a different gcc (probably not gonna happen anytime soon).
I have assembly dumps of all levels of optimisation for ifupdown already,
but haven't had time to really compare them. On a slightly promising
note, though, I just ran today's gcc snapshot and the testsuite looks
VERY good (best results I've seen yet out of gcc on Alpha), but I
need to run a few "real world" tests on it. If that goes well, I'll
ask Matthias to package it up for experimental so we can start playing
with it more. The true test is Qt, of course, on a C++ front and I have
some benchmarking code here to test it with for C. Good F77 code for
testing is always welcome (I forgot more about Fortran at this point
than I ever knew to begin with).
C
Reply to: