Stack-Less SIMT Reconvergence at Low Cost

Download Now
Provided by: CNRS
Topic: Hardware
Format: PDF
Parallel architectures following the SIMT model such as GPUs benefit from application regularity by issuing concurrent threads running in lockstep on SIMD units. As threads take different paths across the control-flow graph, lockstep execution is partially lost, and must be regained whenever possible in order to maximize the occupancy of SIMD units. In this paper, the authors propose a technique to handle SIMT control divergence that operates in constant space and handles indirect jumps and recursion.
Download Now

Find By Topic