from Hacker News

How to isolate an algorithm with CUDA

by zeryx on 9/16/15, 1:35 AM with 2 comments

• by wyldfire on 9/16/15, 1:57 AM

  if(commandQueue[itr].first().def== typeHidden ... else if(commandQueue[itr].first().def == typeMemGateIn)

  I don't know if it's still the case but in the past CUDA/OCL kernels would do all of the execution work for each path in the CFG and only write the results for the actual path to global memory.