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...
Main Author: | |
---|---|
Language: | en |
Published: |
University of Waterloo
2006
|
Subjects: | |
Online Access: | http://hdl.handle.net/10012/1186 |