On trying to emerge OpenOffice 1.1.2, I encountered a number of crashes. A couple of times, it would crash with a cc1 (GCC) compile error. However, the error doesn't seem to appear at any fixed place and the exact nature of the failure is not consistant between emerges. I updated the version of GCC and rebuilt X, Gnome and Java, to see if the glitch could be in related code and only impacting OpenOffice. I also did a bit of stress-testing. As far as I can tell, the problem genuinely is in OpenOffice. The error occurs towards the end of the compilation. Given that OpenOffice 1.1.2 binaries exist, I'm guessing that it may be a CFLAGS filter problem, or some other building tweak. I'm also going to hazard a guess that all other OpenOffice 1.1.2 sourcecode problems are all part of the same problem. (Again, we know it compiles from the original sources, and there aren't that many bug reports, so it seems to be a problem with the way those systems are set up in relation to OpenOffice.) Reproducible: Always Steps to Reproduce: (Starting from a new install with Gnome and KDE installed and whatever packages come on the current Live CD. No additional USES flags. No modifications to CFLAGS.) 1. emerge gnome 2. emerge openoffice Actual Results: Gnome is installed correctly. Some components, such as Bonobo, don't seem to have been fetched in. I'm assuming that the dependencies are set so that if I need those components, they'll be included. OpenOffice brings in various dependencies, including the Blackdown Java compiler. They compile and install just fine. Adding --deep to subsequent builds shows there to be no "hidden" dependencies - at least, not that are listed. OpenOffice starts compiling. It gets through a lot of Java code, core stuff, dialog boxes, etc. The problem occurs towards the end of compiling OpenOffice. It either produces a fatal compilation error OR an internal GCC error. Expected Results: It would be expected that the compilation be completed and the package installed. Pentium 4, Linux-only system, Reiserfs, X w/ KDE running. 20 Gb HD, 128 Mb RAM. 256 Mb swapspace. Portage 2.0.50-r11 (default-x86-2004.2, gcc-3.3.4, glibc-2.3.3.20040420-r1, 2.4.26-gentoo-r6) ================================================================= System uname: 2.4.26-gentoo-r6 i686 Intel(R) Pentium(R) 4 CPU 2.53GHz Gentoo Base System version 1.4.16 Autoconf: sys-devel/autoconf-2.59-r3 Automake: sys-devel/automake-1.8.3 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-O2 -mcpu=i686 -fomit-frame-pointer" CHOST="i386-pc-linux-gnu" COMPILER="" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3.3/env /usr/kde/3.3/share/config /usr/kde/3.3/shutdown /usr/kde/3/share/config /usr/lib/mozilla/defaults/pref /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/bind /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-O2 -mcpu=i686 -fomit-frame-pointer" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache sandbox" GENTOO_MIRRORS="http://gentoo.osuosl.org http://distro.ibiblio.org/pub/Linux/distributions/gentoo" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="" SYNC="rsync://rsync.gentoo.org/gentoo-portage"
"Random" errors are normally a sign of either faulty hardware or too less diskspace. How much free space did you have before starting the compile? OOo needs up to 4-5 GB
I've 70 gigabytes of free disk space, so disk space isn't a problem. (Depending on the largest component, it's possible - though very unlikely - I'm running into a virtual memory problem.) Likewise, I've done fairly exhaustive testing on the system, and I'm not seeing any errors. (Though it should be remembered that a tool can't find an error it isn't programmed to detect.) Over the weekend, I'm going to try the vanilla sources. I'm also going to try to add some flags for debugging GCC activity, see if I can get a better mechanistic description of what's going on.
Still: Random errors are not the fault of the ebuild, so not a lot we can do here, closing
