fails during a test for vfork Reproducible: Always Steps to Reproduce: 1. setup chroot and bootstrap 2. 3. Actual Results: ------- Compiling vs uClibc: gcc -Wall -Wstrict-prototypes -O0 -g -c vfork.c -o vfork.o gcc -g -Wl,-warn-common vfork.o -o vfork true -Since_we_are_debugging -x -R .note -R .comment vfork ./vfork Hi. I'm the child process... Hello. I'm the parent process. make[1]: *** [vfork] Killed make[1]: *** Deleting file `vfork' make[1]: Leaving directory `/var/tmp/portage/uclibc-0.9.27/work/uClibc-0.9.27/test/unistd' make: *** [_dir_unistd] Error 2 Portage 2.0.51.22-r1 (uclibc/x86, gcc-3.4.3-20050110, uclibc-0.9.27-r0, 2.6.11-gentoo-r7 i686) ================================================================= System uname: 2.6.11-gentoo-r7 i686 AMD Athlon(TM) XP2200+ Gentoo Base System version 1.4.16 ccache version 2.4 [enabled] dev-lang/python: [Not Present] sys-apps/sandbox: 1.2.8 sys-devel/autoconf: [Not Present] sys-devel/automake: [Not Present] sys-devel/binutils: 2.15.92.0.2-r9 sys-devel/libtool: [Not Present] virtual/os-headers: 2.6.11 ACCEPT_KEYWORDS="x86 ~x86" AUTOCLEAN="yes" CBUILD="i686-pc-linux-uclibc" CFLAGS="-O2 -march=athlon-xp -fomit-frame-pointer -ftracer -fprefetch-loop-arrays -pipe" CHOST="i686-pc-linux-uclibc" CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.4/env /usr/kde/3.4/share/config /usr/kde/3.4/shutdown /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /var/qmail/alias /var/qmail/control /var/vpopmail/domains /var/vpopmail/etc" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-O2 -march=athlon-xp -fomit-frame-pointer -ftracer -fprefetch-loop-arrays -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs autoconfig candy ccache cvs distlocks nodoc noinfo noman sandbox sfperms strict" GENTOO_MIRRORS="http://open-systems.ufl.edu/mirrors/gentoo http://mirror.datapipe.net/gentoo" LANG="en_US.UTF8" LC_ALL="en_US.UTF8" LDFLAGS="-Wl,-O1 -Wl,--sort-common" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="x86 berkdb ncurses readline uclibc zlib userland_GNU kernel_linux elibc_uclibc" Unset: ASFLAGS, CTARGET, LINGUAS, MAKEOPTS, PORTDIR_OVERLAY
and if you try it with normal CFLAGS ?
Even with normal CFLAGS it still fails same exact position.
Can confirm this bug. It happened to me right after upgrading gcc form gcc-3.3.5-r1 to gcc-3.3.5.20050130-r2. With gcc-3.3.5-r1 I was able to emerge uclibc-0.9.27 just fine. Also i have to mention that this is with my modified gcc and uclibc ebuilds with soft vfp (see Bug #75585). But the behavior is just the same as the one exposed in Comment #1. My CFLAGS="-Os -march=armv4 -pipe -falign-functions=4 -falign-labels=4 -falign-loops=4 -falign-jumps=4" Yuri.
Confirming that this also happens here with gcc version 3.4.4 (Gentoo Hardened 3.4.4, ssp-3.4.4-1.0, pie-8.7.8). Funny thing is that if I cd to the work dir and run make && ./vfork from command line, it seems to work. alpine-testing:/var/tmp/portage/uclibc-0.9.27/work/uClibc-0.9.27/test/unistd # ./vfork Hi. I'm the child process... Hello. I'm the parent process. total 196 drwxr-xr-x 2 root root 4096 Jun 3 07:15 ./ drwxr-xr-x 21 root root 4096 Jan 12 07:59 ../ -rw-r--r-- 1 root root 71 Jan 12 07:59 .cvsignore -rw-r--r-- 1 root root 3045 Jan 12 07:59 Makefile -rwxr-xr-x 1 root root 10208 Jun 3 07:04 fork* -rw-r--r-- 1 root root 2335 Jan 12 07:59 fork.c -rw-r--r-- 1 root root 5756 Jun 3 07:04 fork.o -rwxr-xr-x 1 root root 10208 Jun 3 07:04 fork_glibc* -rw-r--r-- 1 root root 5756 Jun 3 07:04 fork_glibc.o -rwxr-xr-x 1 root root 8375 Jun 3 07:15 getcwd* -rw-r--r-- 1 root root 1619 Jan 12 07:59 getcwd.c -rw-r--r-- 1 root root 3288 Jun 3 07:15 getcwd.o -rwxr-xr-x 1 root root 8698 Jun 3 07:15 getopt* -rw-r--r-- 1 root root 1104 Jan 12 07:59 getopt.c -rw-r--r-- 1 root root 4460 Jun 3 07:15 getopt.o -rwxr-xr-x 1 root root 8986 Jun 3 07:15 getopt_long* -rw-r--r-- 1 root root 1600 Jan 12 07:59 getopt_long.c -rw-r--r-- 1 root root 4632 Jun 3 07:15 getopt_long.o -rwxr-xr-x 1 root root 15042 Jun 3 07:15 preadwrite* -rw-r--r-- 1 root root 4004 Jan 12 07:59 preadwrite.c -rw-r--r-- 1 root root 8248 Jun 3 07:15 preadwrite.o -rwxr-xr-x 1 root root 8500 Jun 3 07:15 vfork* -rw-r--r-- 1 root root 1572 Jan 12 07:59 vfork.c -rw-r--r-- 1 root root 3216 Jun 3 07:15 vfork.o -rwxr-xr-x 1 root root 8500 Jun 3 07:15 vfork_glibc* -rw-r--r-- 1 root root 3216 Jun 3 07:15 vfork_glibc.o Child process exited. Goodbye.
just for those of you who haven't found it yet, FEATURES="-sandbox" is a workaround until this gets fixed
The work around that Brian suggest is *only* good if you are updating an already running system and/or if you do not pass build to emerge ... build however is automatically passed when bootstrap.sh is called.
seems to be fixed as of late .. if it should re-occur later I will re-open
