Compiler Techniques for Rendering

This course summarizes five cutting-edge projects that apply compiler technology to improve the performance and functionality of renderers and shading systems. Several of the projects use LLVM, so the course begins with a gentle introduction to LLVM architecture and concepts. Topics include: customizing shading languages for global illumination and other advanced rendering, analysis of shaders so that renderers may perform physically based light transport in correct units, automatic differentiation, and use of LLVM and dynamic code generation for improved shader performance.

COURSE SCHEDULE

9 am
Introduction to LLVM, and Native RSL Shader Compilation
Leone

9:45 am
Open Shading Language
Gritz

10:15 am
Break

10:30 am
AnySL: Efficient and Portable Multi-Language Shading
Slusallek

11 am
Automatic Bounding of Shaders for Efficient Global Illumination
Walter

11:30 am
Compilation for GPU Accelerated Ray Tracing in OptiX

12:05 pm
Questions and Answers

Level

Intermediate

Prerequisites

Basic understanding of how renderers are constructed and some cursory knowledge of the stages of operation of compilers (parsing, code generation, optimization, etc.).

Intended Audience

Researchers, engineers, and practitioners who are interested in rendering systems, compilers, and the intersections among them.

Instructor(s)

Larry Gritz
Sony Pictures Imageworks

Mark Leone
Weta Digital Ltd

Steven Parker
NVIDIA Corporation

Philipp Slusallek
German Research Center for Artificial Intelligence (DFKI)

Bruce Walter
Cornell University