Forum Discussion

Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
8 years ago

Help with porting a CUDA code to OpenCL

Hi everyone. I am porting a CUDA kernel to OpenCL kernel to execute it on a FPGA (Stratix V). The code looks very similar so it shouldn't be a problem except that this FPGA board supports only O...
Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
8 years ago

Ok so I have managed to run the same code on both CPU and GPU without having problems or crashes. But if I want to run it on GPU all clReleaseMemObject calls have to be changed to clRetainMemObject otherwise it would sometime output wrong values and take much more time to finish run. On CPU is the opposite, I need to switch all clRetainMemObject to clReleaseMemObject to make it work otherwise the program throws CL_OUT_OF_RESOURCES when creating some momentary buffers during execution.

So what is going on? What is the difference between the two? Thanks for the support.