Re: [gentoo-alt] Segmentation fault when trying to emerge gcc-4.3.0

[Fabian Groffen]

On 31-03-2008 19:23:39 +0200, Rabbe Fogelholm wrote:
> Today I attempted a bootstrapping from scratch, on an x86 Linux host, and
> applying the fix suggested in
> <http://bugs.gentoo.org/show_bug.cgi?id=215304>.
>
> I was fairly convinced that everything should go well, having done a very
> similar bootstrapping in the early morning of 2008-03-30.

I suspect masking gcc-4.3.0 will solve more than it breaks for the
moment.


> The problem occurred when emerging gcc, as part of the first `emerge -e
> system'. Here are some lines from the console log:
>
>    ----------
>
> if [ x"-fpic" != x ]; then \
> 	  /local/scratch/portage/sys-devel/gcc-4.3.0/work/build/./prev-gcc/xgcc
> -B/local/scratch/portage/sys-devel/gcc-4.3.0/work/build/./prev-gcc/
> -B/local/scratch/nightly/2008-03-31/usr/i686-pc-linux-gnu/bin/ -c
> -DHAVE_CONFIG_H -fprofile-use -I.
> -I/local/scratch/portage/sys-devel/gcc-4.3.0/work/gcc-4.3.0/libiberty/../include
>  -W -Wall -Wwrite-strings -Wc++-compat -Wstrict-prototypes -pedantic  -fpic
> /local/scratch/portage/sys-devel/gcc-4.3.0/work/gcc-4.3.0/libiberty/hashtab.c
> -o pic/hashtab.o; \
> 	else true; fi
> /local/scratch/portage/sys-devel/gcc-4.3.0/work/gcc-4.3.0/libiberty/hashtab.c:
> In function 'htab_expand':
> /local/scratch/portage/sys-devel/gcc-4.3.0/work/gcc-4.3.0/libiberty/hashtab.c:554:
> internal compiler error: Segmentation fault
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <http://bugs.gentoo.org/> for instructions.
> make[3]: *** [hashtab.o] Error 1
> make[3]: Leaving directory
> `/local/scratch/portage/sys-devel/gcc-4.3.0/work/build/libiberty'
> make[2]: *** [all-stagefeedback-libiberty] Error 2
> make[2]: Leaving directory `/local/scratch/portage/sys-devel/gcc-4.3.0/work/build'
> make[1]: *** [stagefeedback-bubble] Error 2
> make[1]: Leaving directory `/local/scratch/portage/sys-devel/gcc-4.3.0/work/build'
> make: *** [profiledbootstrap] Error 2
>  * ERROR: sys-devel/gcc-4.3.0 failed:
>  *   emake failed with profiledbootstrap
>  *
>  * Call stack:
>  *               ebuild.sh:  49: <call src_compile>
>  *             environment:4690: <call gcc_src_compile>
>  *             environment:2950: <call gcc_do_make>
>  *             environment:2780:     emake LDFLAGS="${LDFLAGS}"
> STAGE1_CFLAGS="${STAGE1_CFLAGS}" LIBPATH="${LIBPATH}"
> BOOT_CFLAGS="${BOOT_CFLAGS}" ${GCC_MAKE_TARGET} || die "emake failed with
> ${GCC_MAKE_TARGET}";
>  *
>  * If you need support, post the topmost build error, and the call stack if relevant.
>  * build log: '/local/scratch/portage/sys-devel/gcc-4.3.0/temp/build.log'
>  * ebuild environment: '/local/scratch/portage/sys-devel/gcc-4.3.0/temp/environment'
>  * S: '/local/scratch/portage/sys-devel/gcc-4.3.0/work/build'
>
>    -------------
>
> The problem is reproducible for me. When it happened the first time I
> suspected that my MAKEOPTS=-j2 might cause the problem (it is a dual core
> host), so I removed it and re-ran, with precisely the same result.
>
> My successful bootstrapping some day ago took a different path as regards
> gcc: `emerge -e system' got me gcc-4.2.3, and a subsequent `emerge --sync
> && emerge -vuND world' (several hours after bootstrapping) got me
> gcc-4.3.0.
>
> The host gcc is 4.1.2/SUSE. Perhaps the climb from 4.1.2 to 4.3.0 is just
> too steep, but doable if using 4.2.3 as a stepping stone?

-- 
Fabian Groffen
Gentoo on a different level
-- 
gentoo-alt@l.g.o mailing list