Summary: | sys-devel/gcc-4.9.3 fails to build on multilib | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Fatih <fthtmn+gentoo> |
Component: | Current packages | Assignee: | Gentoo Toolchain Maintainers <toolchain> |
Status: | RESOLVED OBSOLETE | ||
Severity: | normal | CC: | fthtmn+gentoo |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | AMD64 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
Failed emerge build log for sys-devel:gcc-4.9.3
Failed emerge build log for sys-devel:gcc-4.8.5 |
Description
Fatih
2015-09-23 15:48:28 UTC
Created attachment 412662 [details]
Failed emerge build log for sys-devel:gcc-4.9.3
Created attachment 412664 [details]
Failed emerge build log for sys-devel:gcc-4.8.5
...while initially I'd go on OOM angle, the fact that it goes there for both 4.9.3 *and* 4.8.5 makes it a bit doubtful. Register exhaustion would be a good idea too...if we weren't talking about a 64bit cpu. The important part seems to be that it happens without an error message...so: 1. go into /var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/build/x86_64-pc-linux-gnu/32/libgomp 2. run '/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/build/./gcc/xgcc -B/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/build/./gcc/ -B/usr/x86_64-pc-linux-gnu/bin/ -B/usr/x86_64-pc-linux-gnu/lib/ -isystem /usr/x86_64-pc-linux-gnu/include -isystem /usr/x86_64-pc-linux-gnu/sys-include -DHAVE_CONFIG_H -I. -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp/config/linux/x86 -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp/config/linux -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp/config/posix -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp -Wall -ftls-model=initial-exec -march=i486 -pthread -mtune=generic -g -O2 -pipe -march=corei7-avx -m32 -MT barrier.lo -MD -MP -MF .deps/barrier.Tpo -c /var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp/barrier.c -o barrier.o' 3. hope for a proper error from the compiler On not quite related note: '-sandbox' is almost never a good idea. Dear Rafel, thank you for your quick suggestion. I had to reproduce the build dir with -fail-clean. It failed as usual. But here is surprise. I ran the below command which was failing during emerge: /var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/build/./gcc/xgcc -B/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/build/./gcc/ -B/usr/x86_64-pc-linux-gnu/bin/ -B/usr/x86_64-pc-linux-gnu/lib/ -isystem /usr/x86_64-pc-linux-gnu/include -isystem /usr/x86_64-pc-linux-gnu/sys-include -DHAVE_CONFIG_H -I. -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp/config/linux/x86 -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp/config/linux -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp/config/posix -I/var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp -Wall -ftls-model=initial-exec -march=i486 -pthread -mtune=generic -g -O2 -pipe -march=corei7-avx -m32 -MT barrier.lo -MD -MP -MF .deps/barrier.Tpo -c /var/tmp/notmpfs/portage/sys-devel/gcc-4.8.5/work/gcc-4.8.5/libgomp/barrier.c -o barrier.o and got no errors! # file barrier.o barrier.o: ELF 32-bit LSB relocatable, Intel 80386, version 1 (SYSV), not stripped Where to go with this? bump Okay, I narrowed it down a bit. I hope it helps. In the Chrooted the latest stage3 and gcc-4.8.5 emerges gcc-4.9.3 as well as glibc just fine with the same CFLAGS # grep CFLAGS /etc/portage/make.conf CFLAGS="-O2 -pipe -march=corei7-avx -fomit-frame-pointer" CXXFLAGS="${CFLAGS} (march=native also did not help on host) Using this binary package to build gcc-4.9.3 on the host gives the same error. Does this give you any hint on the issue? This version is long gone. If you still have problems, please file a new bug. |