View previous topic :: View next topic |
Author |
Message |
DawgG l33t
Joined: 17 Sep 2003 Posts: 872
|
Posted: Wed Aug 21, 2024 2:05 pm Post subject: portage cannot use GCC anymore; fails with IA32_EMULATION |
|
|
since the latest update to glibc-2.40 and GCC 14.2 portage on my ~amd64-system seems unable to use GCC on most ebuilds; they fail with
Code: | * Checking if the system can execute 32-bit binaries ...
/var/tmp/portage/sys-libs/glibc-2.40/temp/environment: line 3315: 172 Segmentation fault "${T}/check-ia32-emulation.elf32"
[ !! ]
* Ensure that CONFIG_IA32_EMULATION is enabled in the kernel. |
which IS enabled
Code: | zgrep CONFIG_IA32_EMULATION /proc/config.gz
CONFIG_IA32_EMULATION=y |
in the case of glibc
or
Code: | checking whether we are cross compiling... configure: error: in `/var/tmp/portage/media-libs/xvid-1.3.7-r2/work/xvidcore/build/generic-abi_x86_32.x86':
configure: error: cannot run C compiled programs.
If you meant to cross compile, use `--host'.
| (eg media-libs/xvid-1.3.7-r2; but also many others)
dmesg shows a segfault in conftest in all cases.
i first suspected the reason is that i usually compile for this (older) amd-system in a (newer) intel-chroot, but it started happening at the same time and natively and in the chroot.
GCC itself can be compiled with the ebuild and kernels also.
emerge --info (native system)
Code: | Portage 3.0.65 (python 3.12.5-final-0, default/linux/amd64/23.0/desktop, gcc-14, glibc-2.40, 6.10.6-gentoo-GCC x86_64)
=================================================================
System uname: Linux-6.10.6-gentoo-GCC-x86_64-AMD_Ryzen_7_1700_Eight-Core_Processor-with-glibc2.40
KiB Mem: 16318572 total, 7854092 free
KiB Swap: 0 total, 0 free
Timestamp of repository gentoo: Wed, 21 Aug 2024 12:45:00 +0000
Head commit of repository gentoo: 8d65001c80f026fc7306aaa81f4aa63d49d7cf4f
Head commit of repository librewolf: c61babe7383f9ef2b8195f52dba8cfdbc6060673
sh bash 5.2_p32
ld GNU ld (Gentoo 2.43 p1) 2.43.0
app-misc/pax-utils: 1.3.7::gentoo
app-shells/bash: 5.2_p32::gentoo
dev-build/autoconf: 2.13-r8::gentoo, 2.72-r1::gentoo
dev-build/automake: 1.16.5-r2::gentoo, 1.17-r1::gentoo
dev-build/cmake: 3.30.2::gentoo
dev-build/libtool: 2.4.7-r4::gentoo
dev-build/make: 4.4.1-r1::gentoo
dev-build/meson: 1.5.1::gentoo
dev-java/java-config: 2.3.4::gentoo
dev-lang/perl: 5.40.0::gentoo
dev-lang/python: 3.11.9-r1::gentoo, 3.12.5::gentoo, 3.13.0_rc1_p2::gentoo
dev-lang/rust: 1.79.0::gentoo
dev-lang/rust-bin: 1.80.1::gentoo
sys-apps/baselayout: 2.15::gentoo
sys-apps/openrc: 0.54.2::gentoo
sys-apps/sandbox: 2.39::gentoo
sys-devel/binutils: 2.42-r2::gentoo, 2.43::gentoo
sys-devel/binutils-config: 5.5.2::gentoo
sys-devel/clang: 16.0.6::gentoo, 17.0.6::gentoo, 18.1.8::gentoo
sys-devel/gcc: 14.2.1_p20240817::gentoo
sys-devel/gcc-config: 2.11::gentoo
sys-devel/lld: 16.0.6::gentoo, 17.0.6::gentoo, 18.1.8::gentoo
sys-devel/llvm: 16.0.6::gentoo, 17.0.6-r2::gentoo, 18.1.8-r1::gentoo
sys-kernel/linux-headers: 6.10::gentoo (virtual/os-headers)
sys-libs/glibc: 2.40::gentoo
(...)
ACCEPT_KEYWORDS="amd64 ~amd64"
ACCEPT_LICENSE="*"
CBUILD="x86_64-pc-linux-gnu"
CFLAGS="-march=znver1 -O2 -pipe"
CHOST="x86_64-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/lib64/libreoffice/program/sofficerc /usr/share/config /usr/share/gnupg/qualified.txt /var/spool/munin-async/.ssh"
CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/dconf /etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release /etc/revdep-rebuild /etc/sandbox.d"
CXXFLAGS="-march=znver1 -O2 -pipe"
DISTDIR="/usr/portage/distfiles"
EMERGE_DEFAULT_OPTS="--keep-going"
ENV_UNSET="CARGO_HOME DBUS_SESSION_BUS_ADDRESS DISPLAY GDK_PIXBUF_MODULE_FILE GOBIN GOPATH PERL5LIB PERL5OPT PERLPREFIX PERL_CORE PERL_MB_OPT PERL_MM_OPT XAUTHORITY XDG_CACHE_HOME XDG_CONFIG_HOME XDG_DATA_HOME XDG_RUNTIME_DIR XDG_STATE_HOME"
FCFLAGS="-O2 -pipe"
FEATURES="assume-digests binpkg-docompress binpkg-dostrip binpkg-logs binpkg-multi-instance buildpkg buildpkg-live clean-logs config-protect-if-modified distlocks ebuild-locks fixlafiles ipc-sandbox merge-sync merge-wait metadata-transfer multilib-strict network-sandbox news parallel-fetch pid-sandbox pkgdir-index-trusted preserve-libs protect-owned qa-unresolved-soname-deps sandbox sfperms strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync"
FFLAGS="-O2 -pipe"
GENTOO_MIRRORS="https://linux.rz.ruhr-uni-bochum.de/download/gentoo-mirror/ https://ftp.fau.de/gentoo"
LANG="de_DE.utf8"
LC_ALL=""
LDFLAGS="-Wl,-O1 -Wl,--as-needed -Wl,-z,pack-relative-relocs"
LEX="flex" |
the chroot is configured identically except CFLAGS="-march=x86-64 -mtune=znver1" CXXFLAGS="${CFLAGS}" _________________ DUMM KLICKT GUT. |
|
Back to top |
|
|
myga Tux's lil' helper
Joined: 12 Jun 2023 Posts: 121
|
Posted: Thu Aug 22, 2024 2:05 am Post subject: |
|
|
Not related: Not 100% sure, but could this have something to do with it?! (we need the full fledged BBcode support to properly format posts & amends). _________________ [Never break the law to save people. Nope, there aint no heroes here. Unless you are 'Satoshi Nakamoto', that's a true anonymous.]
Last edited by myga on Thu Aug 22, 2024 2:36 am; edited 1 time in total |
|
Back to top |
|
|
Ionen Developer
Joined: 06 Dec 2018 Posts: 2852
|
Posted: Thu Aug 22, 2024 2:12 am Post subject: |
|
|
Sounds like all 32bit applications are segfaulting for some reason?. About the IA32 message, it just says that because it doesn't know why it failed and gives you the "typical" reason, which is not a segfault -- so can ignore.
Try a 32bit hello world? e.g. Code: | gcc -m32 -march=x86-64 -mtune=znver1 -O2 -pipe -o /tmp/hello -x c - <<EOF
#include <stdio.h>
int main(void) {
printf("hello world\n");
return 0;
}
EOF | Then run /tmp/hello
If that works, then maybe it's specific to when portage runs it under sandbox.
myga wrote: | Not 100% sure, but could this have something to do with it?! | No, this has no relation to an amd64 system. |
|
Back to top |
|
|
DawgG l33t
Joined: 17 Sep 2003 Posts: 872
|
Posted: Thu Aug 22, 2024 10:07 am Post subject: portage issue [solved] |
|
|
THX very much for your input!
@myga: no, it's got nothing to do with itanium, they're just off-the-rack standard cpus, one younger core-i7 that hosts the chroots and an older (2017) ryzen-7 that runs the software.
@Ionen: the "hello world" compiled and runs inside the chroot.
you pointed me in the right direction. it MUST be a portage issue, since with
Code: | FEATURES="-sandbox -usersandbox -userpriv" emerge ... | everything compiles.
it's strange, though, because the host-system and all the chroots use the same FEATURES and this happens only ind ONE of them:
all in make.conf
Code: | host system:
/etc/portage/make.conf:FEATURES="-collision-protect -ccache parallel-fetch sandbox buildpkg userpriv unmerge-orphans fixlafiles -candy split-log "
chroot DOES NOT work:
/tmp/rice/etc/portage/make.conf:FEATURES="sandbox buildpkg distlocks metadata-transfer parallel-fetch userfetch userpriv clean-logs"
chroot works:
/mnt/1/GST/etc/portage/make.conf:FEATURES="sandbox buildpkg distlocks metadata-transfer parallel-fetch userfetch userpriv clean-logs binpkg-multi-instance" |
from emerge --info
Code: | host system:
FEATURES="assume-digests binpkg-docompress binpkg-dostrip binpkg-logs binpkg-multi-instance buildpkg buildpkg-live config-protect-if-modified distlocks ebuild-locks fixlafiles ipc-sandbox merge-sync merge-wait multilib-strict network-sandbox news parallel-fetch pid-sandbox pkgdir-index-trusted preserve-libs protect-owned qa-unresolved-soname-deps sandbox sfperms split-log strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync"
chroot DOES NOT work:
FEATURES="assume-digests binpkg-docompress binpkg-dostrip binpkg-logs binpkg-multi-instance buildpkg buildpkg-live clean-logs config-protect-if-modified distlocks ebuild-locks fixlafiles ipc-sandbox merge-sync merge-wait metadata-transfer multilib-strict network-sandbox news parallel-fetch pid-sandbox pkgdir-index-trusted preserve-libs protect-owned qa-unresolved-soname-deps sandbox sfperms strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync"
chroot works:
FEATURES="assume-digests binpkg-docompress binpkg-dostrip binpkg-logs binpkg-multi-instance buildpkg buildpkg-live clean-logs config-protect-if-modified distlocks ebuild-locks fixlafiles ipc-sandbox merge-sync merge-wait metadata-transfer multilib-strict network-sandbox news parallel-fetch pid-sandbox pkgdir-index-trusted preserve-libs protect-owned qa-unresolved-soname-deps sandbox sfperms strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync"
|
also, just for testing, i (sucessfully) recompiled gcc on the old amd target-system and (the testing-pkgs) glibc and xvid can only be emerged with Code: | FEATURES="-usersandbox" |
(except for the host-system., these are all ancient installations that somehow do not age and do not need (m)any changes, but maybe i should read up on EMERGE_FEATURES a little bit...) _________________ DUMM KLICKT GUT. |
|
Back to top |
|
|
Hu Administrator
Joined: 06 Mar 2007 Posts: 22648
|
Posted: Thu Aug 22, 2024 11:48 am Post subject: |
|
|
That suggests to me that your x86 sandbox is broken. If you run FEATURES='-usersandbox' emerge --ask --verbose --oneshot sys-apps/sandbox, does that fix it so that xvid can be recompiled without a FEATURES override? |
|
Back to top |
|
|
DawgG l33t
Joined: 17 Sep 2003 Posts: 872
|
Posted: Thu Aug 22, 2024 12:36 pm Post subject: |
|
|
Quote: | run FEATURES='-usersandbox' emerge --ask --verbose --oneshot sys-apps/sandbox, does that fix it so that xvid can be recompiled without a FEATURES override? |
just did that, but no, the error remains the same. _________________ DUMM KLICKT GUT. |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|