Could somebody help me with this problem?
I tried to run the GPU Computing SDK C compiled samples but it seems that there is a problem with the CUDA Runtime API. For example, when I run the deviceQuery.exe sample I get the following error:
The same happens with bandwidthTest.exe. The other samples just terminate as soon as they are executed. However, the *Drv.exe samples that use the Driver API run without problems. I can also compile the sample projects without problems using Visual Studio 2008.
I also tried the OpenCL, DirectCompute samples and they run correctly without problems. I even tried the MATLAB Parallel Computing Toolbox GPU functions and they also work without problems. Apparently, the problem is only with the Runtime API.
I installed each software package following the CUDA Getting Started Guide (Windows) instructions, and I tried reinstalling each component without success.
Could somebody provide me with a possible solution or at least some insight about the nature of this problem? I’ve been battling with this issue for 3 days :S
The Dell workstation that I’m working on has the following specifications:
2 Intel Xeon E5620 processors.
Intel Server Board S5520HC.
12GB RAM DDR3.
NVIDIA Tesla C2050.
NVIDIA Quadro 2000.
Windows 7 Professional 64 bit (Japanese language version).
Microsoft Visual Studio 2008 (Japanese language version).
Tesla C2050 device driver 270.61 (Windows 7 64 bit International version).
Quadro 2000 device driver 270.73 (Windows 7 64 bit International version).
CUDA Toolkit 4.0.17 (final) 64-bit.
CUDA Tools 4.0.17 64-bit.
CUDA Computing SDK code samples 4.0.17 64-bit.
If you need to know more information about the hardware and software of my system please let me know and I’ll post it.
Thank you very much in advance.