Bug#682410: viennacl: FTBFS in a minimal sid amd64 chroot

[Andreas Beckmann]

Package: viennacl
Version: 1.2.0-1
Severity: serious
Justification: fails to build from source

Hi,

your package fails to build in a minimal sid chroot on amd64:

make[3]: Entering directory `/tmp/buildd/viennacl-1.2.0/obj-x86_64-linux-gnu'
/usr/bin/cmake -E cmake_progress_report /tmp/buildd/viennacl-1.2.0/obj-x86_64-linux-gnu/CMakeFiles 88
[ 85%] Building CXX object examples/parameters/CMakeFiles/matrixparams.dir/matrix.cpp.o
cd /tmp/buildd/viennacl-1.2.0/obj-x86_64-linux-gnu/examples/parameters && /usr/bin/c++    -Wall -pedantic -I/tmp/buildd/viennacl-1.2.0 -I/tmp/buildd/viennacl-1.2.0/obj-x86_64-linux-gnu -I/tmp/bu
ildd/viennacl-1.2.0/external    -o CMakeFiles/matrixparams.dir/matrix.cpp.o -c /tmp/buildd/viennacl-1.2.0/examples/parameters/matrix.cpp
In file included from /tmp/buildd/viennacl-1.2.0/viennacl/matrix.hpp:28:0,
                 from /tmp/buildd/viennacl-1.2.0/examples/parameters/matrix.cpp:20:
/tmp/buildd/viennacl-1.2.0/viennacl/linalg/matrix_operations.hpp: In instantiation of 'void viennacl::linalg::prod_impl(const viennacl::matrix_expression<const viennacl::matrix<SCALARTYPE, F, AL
IGNMENT>, const viennacl::matrix<SCALARTYPE, F, ALIGNMENT>, viennacl::op_trans>&, const viennacl::vector<SCALARTYPE, VECTOR_ALIGNMENT>&, viennacl::vector<SCALARTYPE, VECTOR_ALIGNMENT>&) [with SC
ALARTYPE = float; F = viennacl::row_major; unsigned int ALIGNMENT = 1u; unsigned int VECTOR_ALIGNMENT = 1u]':
/tmp/buildd/viennacl-1.2.0/viennacl/linalg/matrix_operations.hpp:1023:9:   required from 'viennacl::vector<SCALARTYPE, ALIGNMENT>& viennacl::vector<SCALARTYPE, ALIGNMENT>::operator=(const vienna
cl::vector_expression<const viennacl::matrix_expression<const viennacl::matrix<SCALARTYPE, F, MAT_ALIGNMENT>, const viennacl::matrix<SCALARTYPE, F, MAT_ALIGNMENT>, viennacl::op_trans>, const vie
nnacl::vector<SCALARTYPE, ALIGNMENT>, viennacl::op_prod>&) [with F = viennacl::row_major; unsigned int MAT_ALIGNMENT = 1u; SCALARTYPE = float; unsigned int ALIGNMENT = 1u]'
/tmp/buildd/viennacl-1.2.0/examples/parameters/matrix_functors.hpp:90:3:   required from 'void matrix_trans_vec_mul(TestData&) [with TestData = test_data<viennacl::scalar<float>, viennacl::vecto
r<float, 1u>, viennacl::matrix<float, viennacl::row_major, 1u> >]'
/tmp/buildd/viennacl-1.2.0/examples/parameters/matrix.cpp:154:3:   required from 'int run_matrix_benchmark(test_config&, viennacl::io::parameter_database&) [with ScalarType = float]'
/tmp/buildd/viennacl-1.2.0/examples/parameters/matrix.cpp:237:42:   required from here
/tmp/buildd/viennacl-1.2.0/viennacl/linalg/matrix_operations.hpp:369:7: error: 'trans_prod_impl' was not declared in this scope, and no declarations were found by argument-dependent lookup at the point of instantiation [-fpermissive]
/tmp/buildd/viennacl-1.2.0/viennacl/linalg/matrix_operations.hpp:381:10: note: 'template<class SCALARTYPE, class F, unsigned int ALIGNMENT, unsigned int VECTOR_ALIGNMENT> void viennacl::linalg::trans_prod_impl(const viennacl::matrix<SCALARTYPE, F, ALIGNMENT>&, const viennacl::vector<SCALARTYPE, VECTOR_ALIGNMENT>&, viennacl::vector<SCALARTYPE, VECTOR_ALIGNMENT>&)' declared here, later in the translation unit
/tmp/buildd/viennacl-1.2.0/viennacl/linalg/matrix_operations.hpp: In instantiation of 'void viennacl::linalg::prod_impl(const viennacl::matrix_expression<const viennacl::matrix<SCALARTYPE, F, ALIGNMENT>, const viennacl::matrix<SCALARTYPE, F, ALIGNMENT>, viennacl::op_trans>&, const viennacl::vector<SCALARTYPE, VECTOR_ALIGNMENT>&, viennacl::vector<SCALARTYPE, VECTOR_ALIGNMENT>&) [with SCALARTYPE = double; F = viennacl::row_major; unsigned int ALIGNMENT = 1u; unsigned int VECTOR_ALIGNMENT = 1u]':
/tmp/buildd/viennacl-1.2.0/viennacl/linalg/matrix_operations.hpp:1023:9:   required from 'viennacl::vector<SCALARTYPE, ALIGNMENT>& viennacl::vector<SCALARTYPE, ALIGNMENT>::operator=(const viennacl::vector_expression<const viennacl::matrix_expression<const viennacl::matrix<SCALARTYPE, F, MAT_ALIGNMENT>, const viennacl::matrix<SCALARTYPE, F, MAT_ALIGNMENT>, viennacl::op_trans>, const viennacl::vector<SCALARTYPE, ALIGNMENT>, viennacl::op_prod>&) [with F = viennacl::row_major; unsigned int MAT_ALIGNMENT = 1u; SCALARTYPE = double; unsigned int ALIGNMENT = 1u]'
/tmp/buildd/viennacl-1.2.0/examples/parameters/matrix_functors.hpp:90:3:   required from 'void matrix_trans_vec_mul(TestData&) [with TestData = test_data<viennacl::scalar<double>, viennacl::vector<double, 1u>, viennacl::matrix<double, viennacl::row_major, 1u> >]'
/tmp/buildd/viennacl-1.2.0/examples/parameters/matrix.cpp:154:3:   required from 'int run_matrix_benchmark(test_config&, viennacl::io::parameter_database&) [with ScalarType = double]'
/tmp/buildd/viennacl-1.2.0/examples/parameters/matrix.cpp:253:45:   required from here
/tmp/buildd/viennacl-1.2.0/viennacl/linalg/matrix_operations.hpp:369:7: error: 'trans_prod_impl' was not declared in this scope, and no declarations were found by argument-dependent lookup at the point of instantiation [-fpermissive]
/tmp/buildd/viennacl-1.2.0/viennacl/linalg/matrix_operations.hpp:381:10: note: 'template<class SCALARTYPE, class F, unsigned int ALIGNMENT, unsigned int VECTOR_ALIGNMENT> void viennacl::linalg::trans_prod_impl(const viennacl::matrix<SCALARTYPE, F, ALIGNMENT>&, const viennacl::vector<SCALARTYPE, VECTOR_ALIGNMENT>&, viennacl::vector<SCALARTYPE, VECTOR_ALIGNMENT>&)' declared here, later in the translation unit
make[3]: *** [examples/parameters/CMakeFiles/matrixparams.dir/matrix.cpp.o] Error 1

Full log attached.

Andreas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: viennacl_0.log.gz
Type: application/x-gzip
Size: 11897 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/debian-science-maintainers/attachments/20120722/a0d34a7c/attachment-0001.bin>