I am trying to install PnetCDF from source. I used the 1.12.2 version for installation. MPICH, Zlib, Szip, HDF5 and NetCDF-4 have been installed from source successfully (HDF5 and NetCDF-4 have been built with parallel I/O support).
MPICH is installed as follows (v3.4.2):
./configure --prefix=${DIR}/mpich --with-device=ch4:ofi CFLAGS="-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64" --enable-fortran
make
make check
make install
export PATH=${DIR}/mpich/bin:${PATH}
Where DIR is the root directory all the libraries are installed to (with their own directory).
Zlib (v1.2.11):
./configure --prefix=${DIR}/grib2
make
make check
make install
export CPPFLAGS="-I${DIR}/grib2/include"
export LDFLAGS="-L${DIR}/grib2/lib"
export LD_LIBRARY_PATH="${DIR}/grib2/lib:$LD_LIBRARY_PATH"
Szip (v2.1.1):
./configure --prefix=${DIR}/szip
CC=mpicc
make
make check
make install
export LD_LIBRARY_PATH="${DIR}/szip/lib:$LD_LIBRARY_PATH"
export CPPFLAGS="-I${DIR}/szip/include ${CPPFLAGS}"
export LDFLAGS="-L${DIR}/szip/lib ${LDFLAGS}"
HDF5 (v1.8.22):
export HDF5=${DIR}/hdf5
./configure --prefix=${HDF5} --with-szlib=${DIR}/szip --with-zlib=${DIR}/grib2 --enable-parallel CC=mpicc --enable-fortran FC=mpifort
make
make check
make install
export LD_LIBRARY_PATH="${HDF5}/lib:$LD_LIBRARY_PATH"
export CPPFLAGS="-I${HDF5}/include ${CPPFLAGS}"
export LDFLAGS="-L${HDF5}/lib ${LDFLAGS}"
export PATH="${HDF5}/bin:${PATH}"
NetCDF4-C (v4.8.0):
./configure --prefix=${DIR}/netcdf --disable-dap --enable-parallel-tests --enable-large-file-tests --disable-extreme-numbers CC=mpicc
make
make check
make install
export PATH="${NETCDF}/bin:${PATH}"
export LD_LIBRARY_PATH="${NETCDF}/lib:${LD_LIBRARY_PATH}"
export CPPFLAGS="-I${NETCDF}/include ${CPPFLAGS}"
export LDFLAGS="-L${NETCDF}/lib ${LDFLAGS}"
export LIBS="$(nc-config --libs) ${LIBS}"
NetCDF4-Fortran (v4.5.3):
./configure --prefix=${NETCDF} --enable-large-file-tests --enable-parallel-tests CC=${DIR}/mpich/bin/mpicc FC=mpifort F77=mpif77
make
make check
make install
The HDF5 configuration shown by h5pcc -showconfig is given below (only the part I think relevant is included):
Features:
---------
Parallel HDF5: yes
High-level library: yes
Build HDF5 Tests: yes
Build HDF5 Tools: yes
Threadsafety: no
Default API mapping: v18
With deprecated public symbols: yes
I/O filters (external): deflate(zlib),szip(encoder)
MPE:
Direct VFD: no
(Read-Only) S3 VFD: no
(Read-Only) HDFS VFD: no
dmalloc: no
Clear file buffers before write: yes
Using memory checker: no
Function stack tracing: no
Strict file format checks: no
Optimization instrumentation: no
And the output of nc-config --all:
--cc -> /media/case/work/hwrf/libraries/mpich/bin/mpicc
--cflags -> -I/media/case/work/hwrf/libraries/netcdf/include -I/media/case/work/hwrf/libraries/hdf5/include -I/media/case/work/hwrf/libraries/szip/include -I/media/case/work/hwrf/libraries/grib2/include
--libs -> -L/media/case/work/hwrf/libraries/netcdf/lib -lnetcdf
--static -> -lhdf5_hl -lhdf5 -lm -ldl -lz -lcurl
--has-c++ -> no
--cxx ->
--has-c++4 -> no
--cxx4 ->
--has-fortran -> yes
--fc -> mpifort
--fflags -> -I/media/case/work/hwrf/libraries/netcdf/include -I/media/case/work/hwrf/libraries/netcdf/include
--flibs -> -L/media/case/work/hwrf/libraries/netcdf/lib -lnetcdff
--has-f90 ->
--has-f03 -> yes
--has-dap -> no
--has-dap2 -> no
--has-dap4 -> no
--has-nc2 -> yes
--has-nc4 -> yes
--has-hdf5 -> yes
--has-hdf4 -> no
--has-logging -> no
--has-pnetcdf -> no
--has-szlib -> yes
--has-cdf5 -> yes
--has-parallel4 -> yes
--has-parallel -> yes
--has-nczarr -> no
--prefix -> /media/case/work/hwrf/libraries/netcdf
--includedir -> /media/case/work/hwrf/libraries/netcdf/include
--libdir -> /media/case/work/hwrf/libraries/netcdf/lib
--version -> netCDF 4.8.0
So far, so good. Everything was installed without any failure. All make checks have been passed.
The trouble arises when I try to compile PnetCDF from source. I tried PnetCDF version 1.11.2 for installation (in fact, I triedd several older versions as well, more on that later). The installation is attempted as following:
./configure --prefix=${DIR}/pnetcdf --enable-large-single-req --enable-netcdf4 --enable-shared=yes --with-netcdf4="${DIR}/netcdf" MPICC=mpicc MPIF77=mpif77 MPIF90=mpif90 MPICC=mpicc MPIF77=mpif77 MPIF90=mpif90
make
make tests
make check
make ptest
make ptests
But, it fails at make ptest. The failed test is tst_rec_vars.c and the error message is:
Error at line 78 of tst_rec_vars.c: expect dim X len 4 but got 3
Error at line 78 of tst_rec_vars.c: expect dim X len 4 but got 3 fail with 2 mismatches
make[2]: *** [Makefile:1354: ptest4] Error 1
make[2]: Leaving directory '/media/case/work/hwrf/libraries/source/pnetcdf-1.12.2/test/nc4'
make[1]: *** [Makefile:780: ptests] Error 1
make[1]: Leaving directory '/media/case/work/hwrf/libraries/source/pnetcdf-1.12.2/test'
make: *** [Makefile:965: ptests] Error 1
Heres the test that is failing (GitHub link): tst_rec_vars.c
It seems that the test is checking a NetCDF file for the lenth of dimension in it. I assume the NetCDF file it is checking is tst_rec_vars.nc. The output of ncdump -h on this file is given below:
netcdf tst_rec_vars {
dimensions:
X = UNLIMITED ; // (4 currently)
variables:
int U(X) ;
int V(X) ;
}
That is, the NetCDF file's X dimension indeed has a length 4, although the test program beleives it is only 3. But, if I were to run the test manually as ./tst_rec_vars, it runs correctly.
*** TESTING C tst_rec_vars for record variables to NetCDF4 file ------ pass
Could it be due to parallel processing/threading? The test in question is run under a section titled test/nc4: Parallel testing on 4 MPI processes. I assume the NetCDF file was tested before it was formed completely, which could explain why I got a length 1 X dimension in the NetCDF file from ncdump at one point.
>>>ncdump -h tst_rec_vars.nc
netcdf tst_rec_vars {
dimensions:
X = UNLIMITED ; // (1 currently)
variables:
int U(X) ;
int V(X) ;
}
What exactly is going on here? How do I fix it?
Edit:
I'm trying to install it on CentOS 7, since I was encountering the above mentioned issue, I tried to install it on a Debian 11 system. But I am getting the very exact error in both systems.
Related
The goal is to compile the latest stable GSL to web assembly and make it available as a Node.js module.
I tried the following procedure inspired by this section of the emscripten manual:
git clone git://git.savannah.gnu.org/gsl.git
cd gsl
git checkout tags/release-2-6
autoreconf -i
emconfigure ./configure
emmake make
Unfortunately, I get multiple wasm-ld: error: duplicate symbol.
But compiling the GSL (make) works perfectly fine.
I am using emsdk version 2.0.16 on Ubuntu 18.04.
Does anyone know, how to fix this problem?
Help will be much appreciated.
Finally found a solution inspired by this gist. The problem had to do with dynamic linking and shared libraries.
Anyway, the following code successfully compiles the GSL to a Node.js module:
git clone git://git.savannah.gnu.org/gsl.git
cd gsl
git checkout tags/release-2-6
autoreconf -i
emconfigure ./configure
# Note the flag indicating STATIC linking:
# -------------------------======---------
emmake make LDFLAGS=-all-static
emcc -g -O2 -o .libs/gsl.js -s MODULARIZE -s EXPORTED_RUNTIME_METHODS=\[ccall\] -s LINKABLE=1 -s EXPORT_ALL=1 ./.libs/libgsl.a -lm
The above creates the node module ./.libs/gsl.js which can be used as shown in the following example script:
// test_gsl.js
var factory = require('./.libs/gsl.js');
factory().then((instance) => {
// Compute the value of the Bessel function for x = 5.0:
var besselRes = instance._gsl_sf_bessel_J0(5.0);
// Calculate the hypergeometric cumulative probability distribution for:
// 4: Number of successes (white balls among the taken)
// 7: Number of white balls in the urn
// 19: Number of black balls in the urn
// 13: Number of balls taken
var hg = instance._gsl_cdf_hypergeometric_P(4, 7, 19, 13);
// Do the same using `ccall`
var hg_ccal = instance.ccall("gsl_cdf_hypergeometric_P",
"number",
["number", "number", "number", "number" ],
[4, 7, 19, 13]);
console.log(`besselRes is: ${besselRes}`);
console.log(`gsl_cdf_hypergeometric_P(4,7,19,13): ${hg}`);
console.log(`ccall("gsl_cdf_hypergeometric_P",4,7,19,13): ${hg_ccal}`);
});
Please note that all GSL functions are prefixed with an underscore (_) in the generated Node.js module.
The above script (node ./test_gsl.js) generates this output:
besselRes is: -0.17759677131433826
gsl_cdf_hypergeometric_P(4,7,19,13): 0.8108695652173905
ccall("gsl_cdf_hypergeometric_P",4,7,19,13): 0.8108695652173905
I sincerely hope, this post helps any one.
Have a good one and Cheers!
I'm trying to build an external tool with AOSP. My OS is Linux, Distribution ArchLinux (i3wm), but to compile AOSP I use Ubuntu in Docker (https://android.googlesource.com/platform/build/+/master/tools/docker)
First step:
# init repo
repo init -u https://android.googlesource.com/platform/manifest -b android-8.0.0_r36 --depth=1
repo sync
. build/envsetup.sh # set up environment
lunch aosp_arm-eng # select target to build
Second step: select tool and build
cd external/selinux
mma -j48
Output:
ninja: error: unknown target 'MODULES-IN-'
15:41:55 ninja failed with: exit status 1
make: *** [run_soong_ui] Error 1
make: Leaving directory `/home/user/aosp'
#### make failed to build some targets (6 seconds) ###
Another tool
cd external/wpa_supplicant_8
mma -j48
Output:
ninja: error: unknown target 'MODULES-IN-external-wpa_supplicant_8'
15:41:55 ninja failed with: exit status 1
make: *** [run_soong_ui] Error 1
make: Leaving directory `/home/user/aosp'
#### make failed to build some targets (2 seconds) ###
This happens with any aosp generic target:
Lunch menu... pick a combo:
1. aosp_arm-eng # fails
2. aosp_arm64-eng # fails
3. aosp_mips-eng # fails
4. aosp_mips64-eng # fails
5. aosp_x86-eng # fails
6. aosp_x86_64-eng # fails
7. full_fugu-userdebug # works
8. aosp_fugu-userdebug # works
9. car_emu_arm64-userdebug # fails
10. car_emu_arm-userdebug # fails
11. car_emu_x86_64-userdebug # fails
12. car_emu_x86-userdebug # fails
13. mini_emulator_arm64-userdebug # fails
14. m_e_arm-userdebug # fails
15. m_e_mips64-eng # fails
16. m_e_mips-userdebug # fails
17. mini_emulator_x86_64-userdebug # fails
18. mini_emulator_x86-userdebug # fails
19. aosp_dragon-userdebug # works
20. aosp_dragon-eng # works
21. aosp_marlin-userdebug # works
22. aosp_marlin_svelte-userdebug # works
23. aosp_sailfish-userdebug # works
24. aosp_angler-userdebug # works
25. aosp_bullhead-userdebug # works
26. aosp_bullhead_svelte-userdebug # works
27. hikey-userdebug # works
I want to compile some binary tools in all arch: arm, arm64, x86 and x86_64.
Why aosp_arm-eng does not work? Or how can I change the CPU architecture from non generic aosp target?
Using tapas does not work.
You probably need to make a full AOSP build before trying to use mm... shortcuts - looks like some build script files were not generated yet.
clear all the binaries from /out dir with make clean command & then try to make a full build
Actually there is no need to do a full build. Just open your Android.mk or Android.bp of the module you want to build, and look for variable MODULE_NAME.
Then from the root of the project make
mm $MODULE_NAME
It should build all the dependencies you need first.
I suggest that use the mmm command for building. Also make sure that the directory you are pointing to contains a Android.bp or Android.mk
mmm /external/selinux
Also I suggest to clean the outputs by remove the out directory or simply run:
make clean
If still you have that problem, remove the --depth=1 argument on repo init and sync it again. that argument limits the commit fetching from remote branch.
Use below commands to compile module from root dir
Goto root dir
make clean
source build/envsetup.sh
lunch "select option"
make "module name" -j8
My app uses library: boost, ffmpeg (avcodec, avformat,...). I want to pack my app to run on Ubuntu 14 or higher. I don't want to use static build.
I have checked on some computers which install Ubuntu. Here is results:
Ubuntu 14.04:
ls -l libavcodec*
libavcodec.a
libavcodec.so -> libavcodec.so.54.35.1
libavcodec.so.54.35.1
ls -l libboost_thread*
libboost_thread.a
libboost_thread.so -> libboost_thread.so.1.54.0
libboost_thread.so.1.54.0
Ubuntu 16.04:
ls -l libavcodec*
libavcodec.a -> libavcodec-ffmpeg.a
libavcodec-ffmpeg.a
libavcodec-ffmpeg.so -> libavcodec-ffmpeg.so.56.60.100
libavcodec-ffmpeg.so.56 -> libavcodec-ffmpeg.so.56.60.100
libavcodec-ffmpeg.so.56.60.100
libavcodec.so -> libavcodec-ffmpeg.so
ls -l libboost_thread*
libboost_thread.a
libboost_thread.so -> libboost_thread.so.1.58.0
libboost_thread.so.1.58.0
My app
objdump -p ControlStation | grep NEEDED
NEEDED libavcodec-ffmpeg.so.56
NEEDED libavformat-ffmpeg.so.56
NEEDED libavutil-ffmpeg.so.54
NEEDED libswscale-ffmpeg.so.3
NEEDED libboost_system.so.1.58.0
NEEDED libboost_thread.so.1.58.0
....
I compile on Ubuntu 16, so my app uses file libavcodec-ffmpeg.so.56 and libboost_thread.so.1.58.0. Therefore, it couldn't run on other computers.
Normally, a shared library should be
libName.so ---> libName.so.MAJOR
libName.so.MAJOR --> real library file
But ffmpeg has many versions, boost library doesn't have symbolic file libboost_thread.so.1
Could anyone help me about this problem ? Thank you !
I've been compiling OpenSSL (and thus Perl >5.10, as it is a dependency) on multiple platforms. I've managed to get 1.1.0b compiled on every single platform except AIX, which I can't even compile Perl. I've tried several versions and looked at the documentation Perl provides online. From what I can tell, it suggests version 5.12.2.
When I attempt to compile version 5.12.2,
I take the following from the documentation, and fill in a few local system variables, such as using XLC rv7.
export OBJECT_MODE=64
./Configure \
-d \
-Dcc=/usr/vac/bin/xlc_r7 \
-Duseshrplib \
-Duse64bitall \
-Dprefix=`pwd`/../PERL
Then I attempt to make as prompted, and get the following error:
/usr/vac/bin/xlc_r7 -q64 -o miniperl -brtl -bdynamic -L/usr/local/lib -b64 gv.o toke.o perly.o pad.o regcomp.o dump.o util.o mg.o reentr.o mro.o hv.o av.o run.o pp_hot.o sv.o pp.o scope.o pp_ctl.o pp_sys.o doop.o doio.o regexec.o utf8.o taint.o deb.o universal.o globals.o perlio.o perlapi.o numeric.o mathoms.o locale.o pp_pack.o pp_sort.o miniperlmain.o opmini.o perlmini.o -lbind -lnsl -ldl -lld -lm -lcrypt -lc
LIBPATH=.../perl-5.12.2 ./miniperl -w -Ilib -MExporter -e '<?>' || make minitest
LIBPATH=.../perl-5.12.2 ./miniperl -Ilib autodoc.pl
/usr/bin/ln -s perl5122delta.pod pod/perldelta.pod
LIBPATH=.../perl-5.12.2 ./miniperl -Ilib -Icpan/Cwd -Icpan/Cwd/lib pod/perlmodlib.PL -q
readdir(./../../../../..): Bad file number at lib/FindBin.pm line 116
stat(/pod/): No such file or directory at lib/FindBin.pm line 197
stat(/pod/): No such file or directory at lib/FindBin.pm line 200
Use of chdir('') or chdir(undef) as chdir() is deprecated at pod/perlmodlib.PL line 9.
No such file or directory at pod/perlmodlib.PL line 19.
make: The error code from the last command is 2.
Taking a look at pod/perlmodlib.PL, we see the following:
# MANIFEST itself is Unix style filenames, so we have to assume that Unix style
# filenames will work.
open (MANIFEST, "../MANIFEST") or die $!;
In my desperation I tried to hack it up and avoid writing to the manifest, but then I get this issue:
Creating Makefile.PL in cpan/Archive-Extract for Archive::Extract
Running Makefile.PL in cpan/Archive-Extract
../../miniperl Makefile.PL INSTALLDIRS=perl INSTALLMAN1DIR=none INSTALLMAN3DIR=none PERL_CORE=1 LIBPERL_A=libperl.a
readdir(./../../../../../../..): No such file or directory at ../../lib/File/Find.pm line 610
Use of chdir('') or chdir(undef) as chdir() is deprecated at ../../lib/File/Find.pm line 773.
readdir(./../../..): No such file or directory at ../../cpan/Cwd/lib/File/Spec/Unix.pm line 483
Could not open 'lib/Archive/Extract.pm': No such file or directory at ../../cpan/ExtUtils-MakeMaker/lib/ExtUtils/MM_Unix.pm line 2588.
512 from cpan/Archive-Extract's Makefile.PL at make_ext.pl line 390.
Warning: No Makefile!
make: Cannot find a rule to create target config from dependencies.
Stop.
make config PERL_CORE=1 LIBPERL_A=libperl.a failed, continuing anyway...
Making all in cpan/Archive-Extract
make all PERL_CORE=1 LIBPERL_A=libperl.a
make: Cannot find a rule to create target all from dependencies.
Stop.
Unsuccessful make(cpan/Archive-Extract): code=512 at make_ext.pl line 449.
make: The error code from the last command is 2.
The frequency of errors makes me feel like perhaps I don't have something configured right...
My next thought was to try the most updated version, which at the time of writing is 5.24. Using the same configuration and attempting to make I get the following issue:
Can't locate strict.pm in #INC (you may need to install the strict module) (#INC contains: /cpan/AutoLoader/lib /dist/Carp/lib /dist/PathTools /dist/PathTools/lib /cpan/ExtUtils-Install/lib /cpan/ExtUtils-MakeMaker/lib /cpan/ExtUtils-Manifest/lib /cpan/File-Path/lib /ext/re /dist/Term-ReadLine/lib /dist/Exporter/lib /ext/File-Find/lib /cpan/Text-Tabs/lib /dist/constant/lib /cpan/version/lib /lib .) at autodoc.pl line 25.
BEGIN failed--compilation aborted at autodoc.pl line 25.
make: The error code from the last command is 2.
Which I know from other compilations means I need to edit the PERL5LIB variable.
If I keep adding all the modules to the path:
export PERL5LIB=`pwd`/dist/Exporter/lib:$PERL5LIB
export PERL5LIB=`pwd`/cpan/Text-Tabs/lib:$PERL5LIB
export PERL5LIB=`pwd`/ext/re:$PERL5LIB
export PERL5LIB=`pwd`/dist/constant/lib:$PERL5LIB
export PERL5LIB=`pwd`/cpan/ExtUtils-MakeMaker/lib:$PERL5LIB
export PERL5LIB=`pwd`/dist/Carp/lib:$PERL5LIB
export PERL5LIB=`pwd`/cpan/File-Path/lib:$PERL5LIB
export PERL5LIB=`pwd`/dist/PathTools:$PERL5LIB
export PERL5LIB=`pwd`/dist/PathTools/lib:$PERL5LIB
export PERL5LIB=`pwd`/ext/File-Find/lib:$PERL5LIB
Even so, I'll still get an error for a module that isn't even present in the 5.24.0 source!
Can't locate ExtUtils/MakeMaker/version/vpp.pm in #INC (you may need to install the ExtUtils::MakeMaker::version::vpp module) (#INC contains: <removed from post>) at (eval 2) line 2.
BEGIN failed--compilation aborted at (eval 2) line 2.
Compilation failed in require at .../perl-5.24.0/cpan/ExtUtils-MakeMaker/lib/ExtUtils/MakeMaker.pm line 10.
BEGIN failed--compilation aborted at .../perl-5.24.0/cpan/ExtUtils-MakeMaker/lib/ExtUtils/MakeMaker.pm line 10.
Compilation failed in require at Makefile.PL line 7.
BEGIN failed--compilation aborted at Makefile.PL line 7.
Unsuccessful Makefile.PL(cpan/Archive-Tar): code=512 at make_ext.pl line 517.
make: The error code from the last command is 2.
I have MakeMaker, but the 5.24 version does not contain vpp.pm! Again, in desperation I attempted to put it in there from a different source, and I get this error:
LIBPATH=.../perl-5.24.0 ./miniperl -Ilib make_ext.pl cpan/Archive-Tar/pm_to_blib MAKE="make" LIBPERL_A=libperl.a
readdir(./../../../../../../..): No such file or directory at .../perl-5.24.0/ext/File-Find/lib/File/Find.pm line 142.
Can't cd to : No such file or directory
Unsuccessful Makefile.PL(cpan/Archive-Tar): code=512 at make_ext.pl line 517.
make: The error code from the last command is 2.
All of this makes me feel like perhaps I'm not configuring something right... Can anyone with some experience help me out with some cut and dry installation instructions for Perl on AIX? I'd be super grateful. Thanks!
Below is some information about my system:
> prtconf
System Model: IBM,8231-E1D
Machine Serial Number: Not Available
Processor Type: Not Available
Processor Implementation Mode: POWER 7
Processor Version: PV_7_Compat
Number Of Processors: 0
Processor Clock Speed: Not Available
CPU Type: 64-bit
Kernel Type: 64-bit
Memory Size: 10240 MB
Good Memory Size: Not Available
Platform Firmware level: Not Available
Firmware Version: IBM,AL770_092
Console Login: disable
Auto Restart: true
Full Core: false
I pull down and compile openssh and openssl frequently. I use gcc and no real magic that I can recall.
Here is my script to build it. I don't know which perl is native to AIX but it seems to work for me.
https://github.com/pedz/aix-build-scripts/blob/master/build-scripts/build-openssl
I have ELDK-3.1 installed in a Ubuntu box working perfectly.
In another machine, running 64 bits OpenSuse 12.1, I cloned the ELDK installation and, for some time it worked very well.
Now when I try to configure my projects I see:
configure: error: C compiler cannot create executables
See `config.log' for more details
And the log shows:
configure:3411: ppc-linux-gcc conftest.c >&5
/opt/ELDK-3.1/usr/bin/../lib/gcc-lib/ppc-linux/3.3.3/../../../../ppc-linux/bin/ld: warning: ld.so.1, needed by /opt/ELDK-3.1//usr/../ppc_8xx/lib/libc.so.6, not found (try using -rpath or -rpath-link)
/opt/ELDK-3.1//usr/../ppc_8xx/lib/libc.so.6: undefined reference to `_dl_lookup_versioned_symbol_skip#GLIBC_PRIVATE'
...
The file ld.so.1 is in the same directory as libc.so.6.
s -l /opt/ELDK-3.1//usr/../ppc_8xx/lib/ld.so.1
lrwxrwxrwx 1 root root 11 Jan 31 11:43 /opt/ELDK-3.1//usr/../ppc_8xx/lib/ld.so.1 -> ld-2.3.1.so
As far as I can see, I am correctly defining all the needed environment and trying to build using exactly the same build system as in the Ubuntu box (the project is "automaked").
So I wrote a script trying to mimic everything my automaked configure does:
#!/bin/bash
if [ ! -f confdefs.c ]; then
cat > confdefs.c << EOF
/* confdefs.h */
#define PACKAGE_NAME "xyz"
#define PACKAGE_TARNAME "xyz"
#define PACKAGE_VERSION "1.00"
#define PACKAGE_STRING "xyz 1.00"
#define PACKAGE_BUGREPORT "<contact#company>"
#define PACKAGE_URL ""
#define PACKAGE "xyz"
#define VERSION "1.00"
/* end confdefs.h. */
int
main ()
{
;
return 0;
}
EOF
fi
ARCH=powerpc
export CROSS_COMPILE=ppc_8xx
TOOLCHAIN=ppc-linux-
TOOLCHAIN_ROOT=/opt/ELDK
LD=`which ${TOOLCHAIN}ld`
CC=`which ${TOOLCHAIN}gcc`
GCC=$CC
export CFLAGS="-Wall -g -I${TOOLCHAIN_ROOT}/ppc_8xx/usr/include/"
export CPPFLAGS=$CFLAGS
# export LDFLAGS="-shared"
$CC $CFLAGS $LDFLAGS confdefs.c -o confdefs
This gives me exactly the same error as configure.
If I uncomment the line export LDFLAGS="-shared", on the other hand, it builds perfectly!.
> ls -l confdefs
-rwxr-xr-x 1 myself users 16136 Fev 1 09:52 confdefs
> file confdefs
confdefs: ELF 32-bit MSB shared object, PowerPC or cisco 4500, version 1 (SYSV), dynamically linked, not stripped
Could anybody here please give me any clue of what I am missing so that my projects work finely on one box and not in the other?
Thanks!
I'm not 100% sure that it solves all problems, but it works for us.
We discovered that symlink "ld.so.1 -> ../../../ppc_8xx/lib/ld.so.1" to eldk-3.1/usr/ppc-linux/lib solves linking error.
I suspect something changed with environment between F15 and F16. Same for OpenSUSE (11->12).
Also bug was submitted against Fedora https://bugzilla.redhat.com/show_bug.cgi?id=754695 which was terminated as intentional ABI changes.