Created attachment 390104 [details] emerge --info Emerging sys-devel/gcc-4.8.3 (from 4.6.3) fails to compile with the following message: /bin/sh: line 1: 27287 Bus error build/genautomata /var/tmp/portage/sys-devel/gcc-4.8.3/wor$ Makefile:3589: recipe for target 's-automata' failed make[3]: *** [s-automata] Error 135 make[3]: *** Waiting for unfinished jobs.... /bin/sh /var/tmp/portage/sys-devel/gcc-4.8.3/work/gcc-4.8.3/gcc/../move-if-change tmp-emit.c insn-emit.c echo timestamp > s-emit rm gcc.pod make[3]: Leaving directory '/var/tmp/portage/sys-devel/gcc-4.8.3/work/build/gcc' Makefile:4152: recipe for target 'all-stage1-gcc' failed make[2]: *** [all-stage1-gcc] Error 2 make[2]: Leaving directory '/var/tmp/portage/sys-devel/gcc-4.8.3/work/build' Makefile:16150: recipe for target 'stage1-bubble' failed make[1]: *** [stage1-bubble] Error 2 make[1]: Leaving directory '/var/tmp/portage/sys-devel/gcc-4.8.3/work/build' Makefile:16465: recipe for target 'bootstrap-lean' failed make: *** [bootstrap-lean] Error 2
Adding the build log via "Add an attachment" failed with no reason: Pushing submit didn't have any effect...
Created attachment 390106 [details] build log
Wierd. You usually get a SIGBUS if you have unaligned memory trapped (this is off on x86 by default). Or when a page in fails. I'd guess that you're running out of memory, but it looks like you have enough. Does dmesg show an OOM? Is the error reproduceable?
Several repeated compilings lead to the same result. Additional changes were: more swap, more hdd-memory, setting "-multilib", but same result.
(In reply to mike from comment #4) > Several repeated compilings lead to the same result. Additional changes > were: more swap, more hdd-memory, setting "-multilib", but same result. I've tried to reproduce this but cannot. The only thing I didn't try and may try today, is to use -mno-avx. If I've understood theose registers correctly, they extend the SIMD registers from 128 to 256 bits wide. I can't help but suspect that this might not be causing some alignment error. Can I understadn why you added -mn-avx?
This configuration was used on a Thinkpad x200s which let me compile prior gcc-versions there (there have been some problems with the sandy bridge). I've switched to an HP Elitebook2530p. Since then I've left the config as it was. Perhaps I should mention that the system is installed on an USB-stick, but I see no reason why this should trigger this bug (I'm using a second drive for /var/tmp/portage and there is enough space left on the stick).
Finally solved, Solution: switched to hardened profile "hardened/linux/x86" and recompiled gcc 4.8.3. I don't want this profile but seems to be the only way to get it working.
I was never able to get a handle on this one. I wouldn't say its fixed. We really don't know what was wrong in the first place.
(In reply to Anthony Basile from comment #8) > I was never able to get a handle on this one. I wouldn't say its fixed. We > really don't know what was wrong in the first place. ok alright.
