I wrote a program to benchmark fftw and clfft, but I cannot get same results with same inputs when transform complex data to real data.
I attach my code. Could u help me to check where is wrong? Thanks very much.
As nowadays there is not much activity in AMD Compute Libraries forum, I'm moving this thread to OpenCL forum where you might get a quicker response from the community.
Also, you've been white-listed now.
Thanks so much. Looking forward to getting a quicker response.
Retrieving data ...