Date: 16 Aug 2002 13:20:13 -0500 From: Craig Boston <craig@meoqu.gank.org> To: freebsd-stable@freebsd.org Subject: Odd buildworld problem Message-ID: <1029522013.64529.20.camel@owen1492.it.oot>
next in thread | raw e-mail | index | archive | help
I'm seeing a strange problem during buildworld on one of my boxes. It dies at this point every time: [snip] cc -fpic -DPIC -O -pipe -m486 -I. -I/usr/compile/src/lib/libncurses -I/usr/compile/src/lib/libncurses/../../contrib/ncurses/ncurses -I/usr/compile/src/lib/libncurses/../../contrib/ncurses/include -Wall -DFREEBSD_NATIVE -DNDEBUG -DHAVE_CONFIG_H -DTERMIOS -c /usr/compile/src/lib/libncurses/../../contrib/ncurses/ncurses/tinfo/access.c -o access.So cc -fpic -DPIC -O -pipe -m486 -I. -I/usr/compile/src/lib/libncurses -I/usr/compile/src/lib/libncurses/../../contrib/ncurses/ncurses -I/usr/compile/src/lib/libncurses/../../contrib/ncurses/include -Wall -DFREEBSD_NATIVE -DNDEBUG -DHAVE_CONFIG_H -DTERMIOS -c /usr/compile/src/lib/libncurses/../../contrib/ncurses/ncurses/tinfo/add_tries.c -o add_tries.So cc -fpic -DPIC -O -pipe -m486 -I. -I/usr/compile/src/lib/libncurses -I/usr/compile/src/lib/libncurses/../../contrib/ncurses/ncurses -I/usr/compile/src/lib/libncurses/../../contrib/ncurses/include -Wall -DFREEBSD_NATIVE -DNDEBUG -DHAVE_CONFIG_H -DTERMIOS -c /usr/compile/src/lib/libncurses/../../contrib/ncurses/ncurses/tinfo/alloc_entry.c -o alloc_entry.So cc: Internal compiler error: program cc1 got fatal signal 11 *** Error code 1 Stop in /usr/compile/src/lib/libncurses. This happens consistently in the same spot every time. I seem to remember 4.5 building fine, so I think this is relatively new. For reference, I'm tracking RELENG_4_6 on this box. None of the other machines I am tracking that branch on exhibit this problem. I would suspect bad hardware as it's an El Cheapo Cyrix chip (hence the 486 cputype :), except that it's *always* alloc_entry.c. Indeed, if I cd to /usr/obj/usr/compile/src/lib/libncurses and run that command, it gets a signal 11 every time. I've managed to work around the problem by compiling this file manually without the "-fpic" and building again with -DNOCLEAN. I haven't noticed any problems with the resulting libncurses. The rest of the buildworld goes fine once it gets past this point. Anyway, just wondering if anyone had any wild theories about this one. It's not too critical since it's only on this one box and I have a work-around. It's puzzling though, because IIRC isn't -fpic essentially a no-op on i386? I'm not sure how I would go about building a gcc with debug symbols to try and track down where it dies. Thanks, Craig To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1029522013.64529.20.camel>