tensorflow: macOS build failed

macOS 10.13 CUDA9.0 CUDNN7.0 bazel 0.7.0

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (4) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=float, IntType=tensorflow::int32, useSmem=true]" 
(231): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=float, IntType=tensorflow::int32]" 
(251): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (4) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=float, IntType=tensorflow::int32, useSmem=false]" 
(236): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=float, IntType=tensorflow::int32]" 
(251): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (4) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=float, IntType=tensorflow::int64, useSmem=true]" 
(231): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=float, IntType=tensorflow::int64]" 
(251): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (4) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=float, IntType=tensorflow::int64, useSmem=false]" 
(236): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=float, IntType=tensorflow::int64]" 
(251): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (8) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=double, IntType=tensorflow::int32, useSmem=true]" 
(231): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=double, IntType=tensorflow::int32]" 
(251): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (8) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=double, IntType=tensorflow::int32, useSmem=false]" 
(236): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=double, IntType=tensorflow::int32]" 
(251): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (8) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=double, IntType=tensorflow::int64, useSmem=true]" 
(231): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=double, IntType=tensorflow::int64]" 
(251): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (8) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=double, IntType=tensorflow::int64, useSmem=false]" 
(236): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=double, IntType=tensorflow::int64]" 
(251): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (8) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=tensorflow::complex64, IntType=tensorflow::int32, useSmem=true]" 
(231): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=tensorflow::complex64, IntType=tensorflow::int32]" 
(252): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (8) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=tensorflow::complex64, IntType=tensorflow::int32, useSmem=false]" 
(236): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=tensorflow::complex64, IntType=tensorflow::int32]" 
(252): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (8) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=tensorflow::complex64, IntType=tensorflow::int64, useSmem=true]" 
(231): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=tensorflow::complex64, IntType=tensorflow::int64]" 
(252): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (8) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=tensorflow::complex64, IntType=tensorflow::int64, useSmem=false]" 
(236): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=tensorflow::complex64, IntType=tensorflow::int64]" 
(252): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (16) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=tensorflow::complex128, IntType=tensorflow::int32, useSmem=true]" 
(231): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=tensorflow::complex128, IntType=tensorflow::int32]" 
(253): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (16) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=tensorflow::complex128, IntType=tensorflow::int32, useSmem=false]" 
(236): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=tensorflow::complex128, IntType=tensorflow::int32]" 
(253): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (16) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=tensorflow::complex128, IntType=tensorflow::int64, useSmem=true]" 
(231): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=tensorflow::complex128, IntType=tensorflow::int64]" 
(253): here

tensorflow/core/kernels/split_lib_gpu.cu.cc(122): error: specified alignment (16) is different from alignment (2) specified on a previous declaration
          detected during:
            instantiation of "void tensorflow::split_v_kernel<T,IntType,useSmem>(const T *, tensorflow::CudaDeviceArrayStruct<IntType, 8>, IntType, IntType, tensorflow::CudaDeviceArrayStruct<T *, 8>) [with T=tensorflow::complex128, IntType=tensorflow::int64, useSmem=false]" 
(236): here
            instantiation of "void tensorflow::SplitVOpGPULaunch<T, IntType>::Run(const Eigen::GpuDevice &, __nv_bool, const T *, int, int, const tensorflow::CudaDeviceArrayStruct<IntType, 8> &, const tensorflow::CudaDeviceArrayStruct<T *, 8> &) [with T=tensorflow::complex128, IntType=tensorflow::int64]" 
(253): here

16 errors detected in the compilation of "/var/folders/2f/891g9y691gzcy23blt644z6m0000gn/T//tmpxft_00002e92_00000000-6_split_lib_gpu.cu.cpp1.ii".
ERROR: /Users/odin/local/tensorflow/tensorflow/core/kernels/BUILD:387:1: output 'tensorflow/core/kernels/_objs/split_lib_gpu/tensorflow/core/kernels/split_lib_gpu.cu.pic.o' was not created.
ERROR: /Users/odin/local/tensorflow/tensorflow/core/kernels/BUILD:387:1: not all outputs were created or valid.
Target //tensorflow/tools/pip_package:build_pip_package failed to build
Use --verbose_failures to see the command lines of failed build steps.

About this issue

Original URL
State: closed
Created 7 years ago
Comments: 28 (8 by maintainers)

Most upvoted comments

I have listed all the steps to build tensorflow 1.4 with CUDA 9 at here https://gist.github.com/smitshilu/53cf9ff0fd6cdb64cca69a7e2827ed0f

smitshilu on Nov 7, 2017

Build success! Thank you @chaoswith

smitshilu on Nov 6, 2017

I just checked, and they’re definitely not there. (If they were there, then we wouldn’t have this issue.)

For both of them I do see:

Load command 26
          cmd LC_RPATH
      cmdsize 48
         path ../local_config_cuda/cuda/lib (offset 12)
Load command 27
          cmd LC_RPATH
      cmdsize 56
         path ../local_config_cuda/cuda/extras/CUPTI/lib (offset 12)

So it looks like an install tool directive that is improperly replacing a variable.

elbamos on Jan 5, 2019