I have two kernels to run in sequence. After the first is finished, should I use cudaThreadExit() and then start over with the other kernel?
If I don’t use cudaThreadExit, I have this error message:
Cuda error: cudaGetDeviceProperties: setting the device when a process is active is not allowed.
How to check if a process is still active?
If I use cudaThreadExit, my screen keep flashing at each call.