Compiling Data Dependent Control Flow on SIMD GPUs

Loading...
Thumbnail Image

Date

2004

Authors

Popa, Tiberiu

Advisor

Journal Title

Journal ISSN

Volume Title

Publisher

University of Waterloo

Abstract

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 inefficient. I propose a multi-pass approach based on conditional streams which allows dynamic load balancing of the fragment units of the GPU and better theoretical performance on programs using data dependent conditionals and loops. The proposed system can be used to turn the fragment unit of a SIMD GPU into a stream processor with data dependent control flow.

Description

Keywords

Computer Science, GPU, stream processing, computer graphics

LC Subject Headings

Citation