i have a problem to which i want a solution and fast....any help is appreciated.
I have a laptop with ubuntu10.04 and Nvidia 8400M GS graphics card. I recently started some work on OpenCL and wanted to 开发者_运维百科run it on my laptop's GPU...but i ran into many problems the day I started.
I downloaded the latest drivers and sdk's needed for running OpenCL on an Nvidia GPU. I have installed 1. NVIDIA-Linux-x86-260.19.36 driver
2. devdriver-3.2-linux-32-260.19.26 driver 3. cudatoolkit_3.2.16_linux_32_ubuntu10.04 CUDA Toolkit 4. gpucomputingsdk-3.2.16-linux GPU SDKI set the PATH and the LD_LIBRARY_PATH variables as was indicated after the installation process of the cuda toolkit. After that i went to the *NVIDIA_GPU_Computing_SDK/OpenCL* directory and typed sudo make (as was given here). But i get the error
‘glXGetCurrentContext’ was not declared in this scope
I can't understand what the problem is. Have I missed something to install or is there some problem with the MakeFile and I have to change some options in there?
Also after this step, how am I supposed to link run an OpenCL program code using this?
Thanks in advance.
Have you installed libgl1-mesa-dev
?
These packages provide this file: http://packages.ubuntu.com/search?searchon=contents&keywords=libgl.so&mode=exactfilename&suite=lucid&arch=any
Perhaps kludgy, but I just static link in my build files to the libraries and includes. There are now at least 3 (Intel, AMD, NVIDIA) OpenCL implementations to juggle.
精彩评论