Compiling Data Dependent Control Flow on SIMD GPUs

Current Graphic Processing Units (GPUs) (circa. 2003/2004) have programmable vertex and fragment units. Often these units are implemented as SIMD processors employing parallel pipelines. Data dependent conditional execution on SIMD architectures implemented using processor idling is inefficien...

Full description

Bibliographic Details
Main Author: Popa, Tiberiu
Language:en
Published: University of Waterloo 2006
Subjects:
GPU
Online Access:http://hdl.handle.net/10012/1186