Using 3.2 RC: nvcc fatal - 'Don't know what to do with

Accelerated Computing CUDA CUDA Programming and Performance
1

So I’ve jumped to 3.2 RC, because those CURAND functions look tasty.

I’ve not implemented them yet though. I just installed 3.2RC toolkit and SDK, and driver. For reference, I’m using VS2008 in XP SP3 32-bit, dual E5520 with a Quadro FX580. Have access to a Linux based tesla cluster.

So I changed the Build Rules from v3.0.14 to CUDA Runtime Api Build Rule v3.2 - the exact same one used in the SDK examples.

On compilation, I get the following error:

nvcc fatal   : Don't know what to do with 'sample.cu '

So I went through the SDK, and matched each one of my options to the same ones they had in the project properties, and that didn’t work. It only worked when I pasted my code into the SDK file and removed the fluff from that project.

I’ve looked through the nvcc build command, but I’m stumped. I’d just like to change the build rules on my projects, rather than have to cannibalise the SDK examples each time I have to start a new one. For reference, here it the current command line in release mode:

"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v3.2\bin\nvcc.exe"	-gencode=arch=compute_10,code=\"sm_10,compute_10\" -gencode=arch=compute_13,code=\"sm_13,compute_13\" -gencode=arch=compute_20,code=\"sm_20,compute_20\" --machine 32 -ccbin "C:\Program Files\Microsoft Visual Studio 9.0\VC\bin"	 -Xcompiler "/EHsc /W3 /nologo /O2 /Zi   /MT  " -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v3.2\" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v3.2\include" -I"C:\Documents and Settings\All Users\Application Data\NVIDIA Corporation\NVIDIA GPU Computing SDK 3.2\C\common\inc" -I"C:\Documents and Settings\All Users\Application Data\NVIDIA Corporation\NVIDIA GPU Computing SDK 3.2\C\common\shared" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v3.2\include" -maxrregcount=32  --compile -o "Cuda 2.0/20100826 CUDA Open RW.vcproj.obj" "d:\IJC\WORK\C++ Projects0100826 CUDA Open RW0100826 CUDA Open RW0100826 CUDA Open RW.vcproj"

If I need to provide more info etc, if anyone would like to help debug this problem, please let me know.

Many thanks to anyone that can help or direct me to help.

2

So I’ve jumped to 3.2 RC, because those CURAND functions look tasty.

I’ve not implemented them yet though. I just installed 3.2RC toolkit and SDK, and driver. For reference, I’m using VS2008 in XP SP3 32-bit, dual E5520 with a Quadro FX580. Have access to a Linux based tesla cluster.

So I changed the Build Rules from v3.0.14 to CUDA Runtime Api Build Rule v3.2 - the exact same one used in the SDK examples.

On compilation, I get the following error:

nvcc fatal   : Don't know what to do with 'sample.cu '

So I went through the SDK, and matched each one of my options to the same ones they had in the project properties, and that didn’t work. It only worked when I pasted my code into the SDK file and removed the fluff from that project.

I’ve looked through the nvcc build command, but I’m stumped. I’d just like to change the build rules on my projects, rather than have to cannibalise the SDK examples each time I have to start a new one. For reference, here it the current command line in release mode:

"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v3.2\bin\nvcc.exe"	-gencode=arch=compute_10,code=\"sm_10,compute_10\" -gencode=arch=compute_13,code=\"sm_13,compute_13\" -gencode=arch=compute_20,code=\"sm_20,compute_20\" --machine 32 -ccbin "C:\Program Files\Microsoft Visual Studio 9.0\VC\bin"	 -Xcompiler "/EHsc /W3 /nologo /O2 /Zi   /MT  " -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v3.2\" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v3.2\include" -I"C:\Documents and Settings\All Users\Application Data\NVIDIA Corporation\NVIDIA GPU Computing SDK 3.2\C\common\inc" -I"C:\Documents and Settings\All Users\Application Data\NVIDIA Corporation\NVIDIA GPU Computing SDK 3.2\C\common\shared" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v3.2\include" -maxrregcount=32  --compile -o "Cuda 2.0/20100826 CUDA Open RW.vcproj.obj" "d:\IJC\WORK\C++ Projects0100826 CUDA Open RW0100826 CUDA Open RW0100826 CUDA Open RW.vcproj"

If I need to provide more info etc, if anyone would like to help debug this problem, please let me know.

Many thanks to anyone that can help or direct me to help.

3

The same error. Any help?

4

Solved.
I replaced in Project\CUDA Build Rule\Additional Include Directories:

c:\myinclude\myinclude\

with

c:/myinclude/myinclude/

5

sir,hello.also i had this problem,the compiler tells me:1>nvcc : fatal error : Don’t know what to do with ‘E:/jk2/Myprojects/SURF_Qt_DemoV2.0/fasthessian.cuh’
i have spend too much time about this problem in website,but i can not slve it,could u give me some suggestion?