emerge gcc 3.2.3-r1 fails Reproducible: Always Steps to Reproduce: 1. ACCEPT_KEYWORDS="~sparc" emerge gcc 2. 3. Actual Results: /usr/include/bits/sigset.h: At top level: /var/tmp/portage/gcc-3.2.3-r1/work/gcc-3.2.3/gcc/protector.c:82: warning: `debuginsn' defined but not used make[2]: *** [protector.o] Error 1 make[2]: *** Waiting for unfinished jobs.... make[2]: Leaving directory `/var/tmp/portage/gcc-3.2.3-r1/work/build/gcc' make[1]: *** [stage1_build] Error 2 make[1]: Leaving directory `/var/tmp/portage/gcc-3.2.3-r1/work/build/gcc' make: *** [bootstrap-lean] Error 2 !!! ERROR: sys-devel/gcc-3.2.3-r1 failed. !!! Function src_compile, Line 345, Exitcode 2 !!! (no error message) Expected Results: should emerge sucessfully # emerge info Portage 2.0.48-r1 (default-sparc64-1.4, gcc-3.2.2, glibc-2.3.1-r4) ================================================================= System uname: 2.4.20-sparc-r8 sparc64 sun4u GENTOO_MIRRORS="ftp://ftp.ussg.iu.edu/pub/linux/gentoo ftp://cs.ubishops.ca/pub/gentoo ftp://ftp.gtlib.cc.gatech.edu/pub/gentoo" CONFIG_PROTECT="/etc /var/qmail/control /usr/share/config /usr/kde/2/share/config /usr/kde/3/share/config" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" PORTDIR="/usr/portage" DISTDIR="/usr/portage/distfiles" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR_OVERLAY="" USE="sparc arts apm avi encode esd gif gtk imlib jpeg libwww mikmod motif mpeg oggvorbis oss spell xv xmms zlib gdbm readline pam -X -kde -java -opengl -pdflib -qt -truetype -gnome -nls berkdb crypt cups dga directfb doc fbcon kerberos mysql ncurses perl png python samba ssl sdl slang tcpd tiff xml xml2" COMPILER="gcc3" CHOST="sparc-unknown-linux-gnu" CFLAGS="-mcpu=ultrasparc -mtune=ultrasparc -O3 -pipe" CXXFLAGS="-mcpu=ultrasparc -mtune=ultrasparc -O3 -pipe" ACCEPT_KEYWORDS="sparc" MAKEOPTS="-j4" AUTOCLEAN="yes" SYNC="rsync://rsync.us.gentoo.org/gentoo-portage/" FEATURES="sandbox ccache"
I encountered the same problem on an i586 system. After increasing the swap by around 200M, the bug was gone. I guess the sparc might need a bit more swap than the intel architecture, so you might try that solution
# free total used free shared buffers cached Mem: 885896 870080 15816 0 135856 544008 -/+ buffers/cache: 190216 695680 Swap: 391400 0 391400 Ram is 900MB, and Swap is 384MB... :( I don't *think* that should be an issue.
I have just hit the same thing on my new dual xeon 2.66ghz at work. Running ~x86. 1Gb Registered ECC RAM, 3gb Swap, so I don't think that is an issue. However when I run the emerge again (nothing else done in the meantime on the system) after a little while it worked 100%. Possibly this is either an intermittant bug with something common to all systems or a weird place that gcc hits really hard on hardware.
Changed MAKEOPTS="-j3" and problem solved. This is an SMP system, and I noticed emerge info makes no indication of SMP. per Keith M Wesolowski: <wesolows> a good possibility on that gcc fail is parallel make b0rkage <BladeMan> ah, so crank back the j4? <wesolows> can't hurt to try # emerge info Portage 2.0.48-r1 (default-sparc64-1.4, gcc-3.2.3, glibc-2.3.1-r4) ================================================================= System uname: 2.4.21 sparc64 sun4u # uname -a Linux ultra.dummy.net 2.4.21 #2 SMP Thu Jun 19 13:52:23 EDT 2003 sparc64 sun4u TI UltraSparc II (BlackBird) GNU/Linux
Thanks, that is worthwhile noting. Could the gcc people please see about fixing their makefiles so that parallel makes work properly?
Closing