Re: gcc bug?

[Thorsten Glaser <tg@mirbsd.de>]

Finn Thain dixit:

>Well, I didn't find anything messed up. So I built the gcc-4.5-20101021 
>snapshot and, sure enough, same problem.

Ouch, okay. Thanks for all the assistance, nevertheless. I’m feeling a
bit bad sometimes for not coding, only compiling… but getting the pak-
kages into the right order and hunting, in some cases, dependencies is
also quite… fulfilling ☺ plus it appears I’m now a linux-2.6 contribu-
tor ☻

>The gcc bugzilla seems to be having issues at the moment, so I've uploaded 

For me, reloading (Lynx: z x) until the page comes usually works, the
entire gcc.gnu.org webserver has issues…

>BTW, this can be worked around by by disabling optimisation (in case 
>util-linux is holding up things).

Beat me, hit me, etc. but I’m putting this as /usr/bin/gcc-4.4 and,
with s/gcc/g++/ as g++-4.4, in a clean cowbuilder shell:

#!/bin/sh
gcc-4.4-real "$@" >/tmp/g1.$$ 2>/tmp/g2.$$ || {
        echo >&2 gcc-4.4: retrying with -O0
        exec gcc-4.4-real "$@" -O0
        exit 255
}
cat /tmp/g1.$$
cat /tmp/g2.$$ >&2
rm -f /tmp/g1.$$ /tmp/g2.$$
exit 0

The original binaries are /usr/bin/g{cc,++}-4.4-real obviously.
Ok now that I’ve been beaten, let’s see whether this works…

bye,
//mira“creative”bilos
-- 
I believe no one can invent an algorithm. One just happens to hit upon it
when God enlightens him. Or only God invents algorithms, we merely copy them.
If you don't believe in God, just consider God as Nature if you won't deny
existence.		-- Coywolf Qi Hunt