[Numpy-discussion] ANN: Scipy 0.13.0 beta 1 release
Neal Becker
ndbecker2 at gmail.com
Wed Sep 4 08:59:51 EDT 2013
David Cournapeau wrote:
> On Wed, Sep 4, 2013 at 1:00 PM, Neal Becker <ndbecker2 at gmail.com> wrote:
>
>> Failed building on fedora 19 x86_64 using atlas:
>>
>> creating build/temp.linux-x86_64-2.7/numpy/linalg
>> creating build/temp.linux-x86_64-2.7/numpy/linalg/lapack_lite
>> compile options: '-DATLAS_INFO="\"3.8.4\"" -I/usr/include
>> -Inumpy/core/include -
>> Ibuild/src.linux-x86_64-2.7/numpy/core/include/numpy
>> -Inumpy/core/src/private -
>> Inumpy/core/src -Inumpy/core -Inumpy/core/src/npymath -
>> Inumpy/core/src/multiarray -Inumpy/core/src/umath -Inumpy/core/src/npysort
>> -
>> Inumpy/core/include -I/usr/include/python2.7 -c'
>> gcc: numpy/linalg/lapack_litemodule.c
>> gcc: numpy/linalg/lapack_lite/python_xerbla.c
>> /usr/bin/gfortran -Wall -Wl,-rpath=/opt/intel/mkl/lib/intel64
>> build/temp.linux-
>> x86_64-2.7/numpy/linalg/lapack_litemodule.o build/temp.linux-
>> x86_64-2.7/numpy/linalg/lapack_lite/python_xerbla.o -L/usr/lib64/atlas -
>> L/usr/lib64 -Lbuild/temp.linux-x86_64-2.7 -llapack -lptf77blas -lptcblas
>> -latlas
>> -lpython2.7 -lgfortran -o
>> build/lib.linux-x86_64-2.7/numpy/linalg/lapack_lite.so
>> /usr/lib/gcc/x86_64-redhat-linux/4.8.1/../../../../lib64/crt1.o: In
>> function
>> `_start':
>> (.text+0x20): undefined reference to `main'
>> collect2: error: ld returned 1 exit status
>> /usr/lib/gcc/x86_64-redhat-linux/4.8.1/../../../../lib64/crt1.o: In
>> function
>> `_start':
>> (.text+0x20): undefined reference to `main'
>> collect2: error: ld returned 1 exit status
>>
>> Build command was:
>> env ATLAS=/usr/lib64 FFTW=/usr/lib64 BLAS=/usr/lib64 LAPACK=/usr/lib64
>> CFLAGS="-mtune=native -march=native -O3" LDFLAGS="-Wl,-
>> rpath=/opt/intel/mkl/lib/intel64" python setup.py build
>>
>
> This command never worked: you need to add the -shared flag to LDFLAGS (and
> you may want to remove rpath to MKL if you use ATLAS).
>
> David
>
OK, building with -shared (and removing rpath) works. numpy.test('full')
reports no unexpected failures.
More information about the NumPy-Discussion
mailing list