visual studio gives undeclared indentifier for threadIdx, blockIdx, blockDim et al

Accelerated Computing CUDA CUDA Programming and Performance
1

New to this; modeled my application on the Nvidia examples, especially reduction.
Using VS 9. Using rules that came with reduction example.
I have all 6 kernels in file: MSI_kernel.cu
I have wrappers that call kernel files in file: MSI_Cuda.cpp
Cuda.cpp contains the line:
#include “MSI_kernel.cu”

near the beginning.

i have my application code in file: msi.cpp

now when I compile and build the cuda code passes first stage of compilation:

1>“C:\CUDA\bin\nvcc.exe” -ccbin “C:\Program Files\Microsoft Visual Studio 9.0\VC\bin” -I"C:\CUDA\include" -I"./" -I"…/…/common/inc" -I"…/…/…/shared/inc" -Xcompiler "/EHsc /W3 /nologo /O2 /Zi /MT " -maxrregcount=32 -gencode=arch=compute_10,code="sm_10,compute_10" -gencode=arch=compute_13,code="sm_13,compute_13" --compile -o “Debug\MSI_kernel.cu.obj” “c:\MSI\MSI_C\MSI_C\MSI_kernel.cu”
1>MSI_kernel.cu
1>tmpxft_00000e78_00000000-6_MSI_kernel.compute_10.cudafe1.gpu
1>tmpxft_00000e78_00000000-10_MSI_kernel.compute_10.cudafe2.gpu
1>MSI_kernel.cu
1>tmpxft_00000e78_00000000-3_MSI_kernel.compute_13.cudafe1.gpu
1>tmpxft_00000e78_00000000-14_MSI_kernel.compute_13.cudafe2.gpu
1>tmpxft_00000e78_00000000-6_MSI_kernel.compute_10.cudafe1.cpp
1>tmpxft_00000e78_00000000-19_MSI_kernel.compute_10.ii
1>Compiling…
1>MSI_Sub.cpp
1>MSI_CUDA.cpp

then wehn compiling the wrapper [host] code I get error messages for ALL C language extensions: dim3, threadIdx, global etc. of “undeclared identifier”

I assume i have something botched in the Project file and/or rules file? can anybody tell me where to look?

ps – the reduction example that i downloaded from nvidia compiles fine, no problems.

thanks

2

New to this; modeled my application on the Nvidia examples, especially reduction.
Using VS 9. Using rules that came with reduction example.
I have all 6 kernels in file: MSI_kernel.cu
I have wrappers that call kernel files in file: MSI_Cuda.cpp
Cuda.cpp contains the line:
#include “MSI_kernel.cu”

near the beginning.

i have my application code in file: msi.cpp

now when I compile and build the cuda code passes first stage of compilation:

1>“C:\CUDA\bin\nvcc.exe” -ccbin “C:\Program Files\Microsoft Visual Studio 9.0\VC\bin” -I"C:\CUDA\include" -I"./" -I"…/…/common/inc" -I"…/…/…/shared/inc" -Xcompiler "/EHsc /W3 /nologo /O2 /Zi /MT " -maxrregcount=32 -gencode=arch=compute_10,code="sm_10,compute_10" -gencode=arch=compute_13,code="sm_13,compute_13" --compile -o “Debug\MSI_kernel.cu.obj” “c:\MSI\MSI_C\MSI_C\MSI_kernel.cu”
1>MSI_kernel.cu
1>tmpxft_00000e78_00000000-6_MSI_kernel.compute_10.cudafe1.gpu
1>tmpxft_00000e78_00000000-10_MSI_kernel.compute_10.cudafe2.gpu
1>MSI_kernel.cu
1>tmpxft_00000e78_00000000-3_MSI_kernel.compute_13.cudafe1.gpu
1>tmpxft_00000e78_00000000-14_MSI_kernel.compute_13.cudafe2.gpu
1>tmpxft_00000e78_00000000-6_MSI_kernel.compute_10.cudafe1.cpp
1>tmpxft_00000e78_00000000-19_MSI_kernel.compute_10.ii
1>Compiling…
1>MSI_Sub.cpp
1>MSI_CUDA.cpp

then wehn compiling the wrapper [host] code I get error messages for ALL C language extensions: dim3, threadIdx, global etc. of “undeclared identifier”

I assume i have something botched in the Project file and/or rules file? can anybody tell me where to look?

ps – the reduction example that i downloaded from nvidia compiles fine, no problems.

thanks