[Debian GNUstep maintainers] Bug#918366: FTBFS for armhf on arm64, fails some tests

Yavor Doganov yavor at gnu.org
Sat Jan 19 22:53:53 GMT 2019


On Mon, 07 Jan 2019 21:10:31 +0200,
Steve McIntyre wrote:
> base/NSInvocation/general.m:
> Failed test:     general.m:208 ... Can send/return NSRect

I tried to reproduce this on gcc117 (an AMD Opteron 1100 machine from
the GCC Compile Farm, which should be able to run 32-bit executables
according to one of your previous messages on the subject) with a
manually built cross-toolchain (pristine upstream binutils/2.31,
gcc/8.2.0, glibc/2.28 and libffi/3.2.1) but unfortunately the test
passes.

Am I missing something?  Or doing something wrong?

yavor at gcc117:~$ dpkg --print-architecture
arm64
yavor at gcc117:~$ dpkg --print-foreign-architectures
yavor at gcc117:~$ cat /etc/debian_version 
9.5
yavor at gcc117:~$ uname -a
Linux gcc117 4.9.0-5-arm64 #1 SMP Debian 4.9.65-3+deb9u2 (2018-01-04) aarch64 GNU/Linux
yavor at gcc117:~$ lscpu
Architecture:          aarch64
Byte Order:            Little Endian
CPU(s):                8
On-line CPU(s) list:   0-7
Thread(s) per core:    1
Core(s) per socket:    2
Socket(s):             4
Model:                 2
BogoMIPS:              500.00
Flags:                 fp asimd evtstrm aes pmull sha1 sha2 crc32
yavor at gcc117:~$ arm-linux-gnueabihf-gcc -v
Using built-in specs.
COLLECT_GCC=arm-linux-gnueabihf-gcc
COLLECT_LTO_WRAPPER=/home/yavor/libexec/gcc/arm-linux-gnueabihf/8.2.0/lto-wrapper
Target: arm-linux-gnueabihf
Configured with: ../configure --prefix=/home/yavor --disable-nls --disable-multilib --enable-languages=c,c++,objc --target=arm-linux-gnueabihf
Thread model: posix
gcc version 8.2.0 (GCC)
yavor at gcc117:~$ ~/arm-linux-gnueabihf/lib/libc.so.6
GNU C Library (GNU libc) stable release version 2.28.
Copyright (C) 2018 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.
Compiled by GNU CC version 8.2.0.
libc ABIs: UNIQUE ABSOLUTE
For bug reporting instructions, please see:
<http://www.gnu.org/software/libc/bugs.html>.
yavor at gcc117:~$ file ~/arm-linux-gnueabihf/lib/libc-2.28.so 
/home/yavor/arm-linux-gnueabihf/lib/libc-2.28.so: ELF 32-bit LSB shared object, ARM, EABI5 version 1 (SYSV), dynamically linked, interpreter /home/yavor/arm-linux-gnueabihf/lib/ld-linux.so.3, for GNU/Linux 3.2.0, not stripped

yavor at gcc117:~/src/gnustep-base-1.26.0$ file Source/obj/libgnustep-base.so.1.26.0 
Source/obj/libgnustep-base.so.1.26.0: ELF 32-bit LSB shared object, ARM, EABI5 version 1 (SYSV), dynamically linked, not stripped
yavor at gcc117:~/src/gnustep-base-1.26.0$ grep NSRect Tests/tests.log
Passed test:     general.m:208 ... Can send/return NSRect

I have no other idea except to retry the test with all Debian patches
to the toolchain packages applied.  If I'm still unable to reproduce,
I'd need some aid from people who have access to a machine that
exhibits the problem.

As a last resort we can disable the test, like Ubuntu.  I really don't
like this approach as it's just sweeping the problem under the rug.
There's a bug somewhere which ideally should be hunted down and fixed.



More information about the pkg-GNUstep-maintainers mailing list