[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gentoo-user
Subject:    [gentoo-user] bootstrapping v1.3b
From:       Laurent <lsinitam () noos ! fr>
Date:       2002-07-31 11:15:22
[Download RAW message or body]

Trying to bootstrap gentoo 1.3b on a K6-2 350Mhz with 2 SCSI disk 9Go each and
384Mo of RAM.
Well first, Gcc 3.1 doesn't like -march=k6-2 : switched to -march=k6
then it was -m3dnow : suppressed it
then it was -mmmx : suppressed it
and finally it appeared that just the -march=k6 was a problem
GRRRR  :-/
Ok, I've tried again with -march=i586 -mmmx
Well, had to suppress the -mmmx again... :(

Well, at least gcc was OK with -fomit-frame-pointer :p

And at last it worked until I got into sys-apps/texinfo-4.2-r5
/usr/include/bits/string2.h: In function `__strcspn_c3':
/usr/include/bits/string2.h:926: internal error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
make[3]: *** [terminal.o] Error 1

Well, gcc bug again... As always, I tried to emerge just the texinfo package 
just to confirm that my settings were wrong. And this time I was shocked to 
see that
# emerge /usr/portage/sys-apps/texinfo/texinfo-4.2-r5.ebuild
compiled thing FLAWLESSLY even though the gcc options I used were the same 
that during the bootstrap.
I've repeated it, many time. I can confirm, that gcc will take only 
-march=i586 (no mmx, k6, 3dnow, etc). And I can confirm that with that 
setting, gcc always segfaults when compiling texinfo during the bootstrap and 
always works OK when compiling texinfo from a regular emerge command.

Does anyone knows what's going on here ???

-- 
Laurent Sinitambirivoutin
lsinitam@noos.fr

_______________________________________________
gentoo-user mailing list
gentoo-user@gentoo.org
http://lists.gentoo.org/mailman/listinfo/gentoo-user
[prev in list] [next in list] [prev in thread] [next in thread]