After bootstrapping the system using the "2004.0" live cd and using the "stage1-athlon-xp-2004.0-20040210.tar.bz2" tarball, going into the "emerge system" stage, I receive this error when compiling "groff" "make[2]: *** [examples/letters.ps] Error 139 make[2]: Leaving directory '/var/tmp/portage/groff-1.18.1-r4/work/groff-1.18.1/contrib/mom' make[1]: *** [contrib/mom] Error 2 make[1]: Leaving directory '/var/tmp/portage/groff-1.18.1-r4/work/groff-1.18.1' make: *** [all] Error 2 !!! ERROR: sys-apps/groff-1.18.1-r4 failed. !!! Function src_compile, Line 85, Exitcode 2 !!! (no error message)" I've tried using previous versions of groff, but I get exactly the same error no matter what I do. I figured this was a bug because upon re-formatting and using an older stage1 tarball, I received no errors, installation went smooth Reproducible: Always Steps to Reproduce: 1.Download stage1-athlon-xp-2004.0-20040210.tar.bz2 tarball 2.Follow installation instructions till you finish the bootstrap process 3.Do an "Emerge System" Actual Results: "Groff-1.18.1" stopped compiling with this error "make[2]: *** [examples/letters.ps] Error 139 make[2]: Leaving directory '/var/tmp/portage/groff-1.18.1-r4/work/groff- 1.18.1/contrib/mom' make[1]: *** [contrib/mom] Error 2 make[1]: Leaving directory '/var/tmp/portage/groff-1.18.1-r4/work/groff-1.18.1' make: *** [all] Error 2 !!! ERROR: sys-apps/groff-1.18.1-r4 failed. !!! Function src_compile, Line 85, Exitcode 2 !!! (no error message)" Every single time Expected Results: Compiled groff and carried on emerging system. portage-2.0.50-r1 (hardened-x86-2004.0, gcc-3.3.2, glibc-2.3.2-r9, 2.4.24-xfs- r0) ========================================= System uname: 2.4.24-xfs-r0 i686 AMD Athlon(tm) XP 2400+ Gentoo Base System version 1.4.3.13 Autoconf: Automake: ACCEPT_Keywords="~x86" AUTOCLEAN="yes" CFLAGS="-march=athlon-xp -O2 -pipe -fomit-frame-pointer" CHOST="i686-pc-linux-gnu" COMPILER="gcc3" CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config /usr/share/ config /var/qmail/control" CXXFLAGS="-march=athlon-xp -O2 -pipe -fomit-frame-pointer" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache sandbox sfperms strict" GENTOO_MIRRORS="http://gentoo.oregonstate.edu 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" USE="3dnow alsa berkdb crypt kde mmx nls pam pic qt readline ssl tcpd x86 xmms zlib" Running 2004.0 live cd
Same thing here, using 2004.0 stage1 image (not using a livecd), 3 different times. Going to try the 1.4 stage1.
Same here, "clean" install of gentoo 2004.0, no specific options turned on. Groff bails out with the error above.
Seems to be a problem with binutils(ld says "cannot find entry symbol _start; defaulting to ...."
Want to post the actual error and not just the makefile failing out? Might be a ways back.
this is a hardened stage, not a vanilla x86 stage
kurt, see here for what I was talking about wrt the confusion about hardened and regular living together
Created attachment 26399 [details] Output of "emerge groff" I have the same problem, clean install from stage1-x86-pie-ssp-2004.0.tar.bz2 USE="-gtk -gnome qt kde dvd alsa cdr" CHOST="i686-pc-linux-gnu" CFLAGS="-mcpu=athlon-xp -O3 -pipe -fomit-frame-pointer" Problem occurs on "emerge system" and also on "emerge groff"
please do etc-update and hcc -a if that does not help, unmerge hardened-gcc if that does not help, gcc -dumpspecs > $(gcc-config -L)/specs if that does not help, write to me or meet me on gentoo-hardened, we can provide you some fresh specs if you need one sorry for inconveniences, Alex
Thanks Alexander. That solution solved the problem for me (etc-update && hcc -a). I used the hardened stage; and did emerge system again, and groff emerge without issue. Just thought you would like to know. -Bryan
close the bug close the bug close the bug close the bug close it please close it yeah yeah yeah close it yeah wow, that was close ;-) (just a little pun here, no offense intented)
Reopening and changing title at request of pappy.
"Steps to Reproduce: 1.Download stage1-athlon-xp-2004.0-20040210.tar.bz2 tarball" that does show it being a pie-ssp stage, our stages are suppose to be labeled with pie-ssp, i've just looked and the url is http://gentoo.oregonstate.edu/experimental/x86/stages/athlon-xp/hardened/ which isn't clear enough i guess that this is not a plain stage. I'm cc'ing zhen to take care of the naming conventions, perhaps we also need to add a README to the hardened directory Zhen: in addition to the naming scheme this needs to be fixed so that this error doesn't happen to hardened users [19:23] <pappy-> updating hardened-gcc replaces config script [19:23] <pappy-> which does not run etc-update during update [19:23] <pappy-> which then cannot run hcc -a with the right config when gcc made it through
as talked to rac on the -dev channel, we will have to do something about this step 1) mask all versions of hardened-gcc, remove hardened-gcc from CVS step 2) mask sys-devel/hardened-gcc in all profiles, forcing blocker removal step 3) mask sys-devel/hardened as block in recent gcc ebuilds to force removal before gcc can be updated the last step will be introducing the new "hardened" USE flag (IUSE and USE.description need to be updated for this) again, thanks to rac for being so pervasive about this hot issue, i was not aware that the problem was so wide-spread. i think we can solve this issue tonight and prevent users from falling down this way in the future.
hardened-gcc is now masked for all users, and the hardened stages are moved to a less confusing location. Looks like we are past this particular bump in the road.
