x86_64 Dockerfile Fails to Build Arm Compute Library

[Hamptonjc]

...

src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
 });
  ^
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’
src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.cpp:71:2: error: converting to ‘const std::tuple<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<unsigned int, std::allocator<unsigned int> > >’ from initializer list would use explicit constructor ‘constexpr std::tuple<_T1, _T2>::tuple(const _T1&, const _T2&) [with _T1 = std::__cxx11::basic_string<char>; _T2 = std::vector<unsigned int>]’

...

scons: *** [build/src/gpu/cl/kernels/ClGemmMatrixMultiplyNativeKernel.o] Error 1

[james-conroy-arm]

Hi @Hamptonjc ,

Thanks for reporting this build issue. We are currently working on solutions to improve the maintainability and usability of this Dockerfile.

I have pushed a patch to our Gerrit code review system, please have a look to see if it resolves your issue: https://review.mlplatform.org/c/ml/armnn/+/6737

May I ask what you are using the Dockerfile for? It may help us with the improvements we are making.

Cheers, James

[Hamptonjc]

@james-conroy-arm ,

The patch on Gerrit builds successfully. I am using the Dockerfile for cross compilation.

Thanks!

[james-conroy-arm]

Great to hear, no bother at all.

James