Graphics has been one of the best success stories of parallel processing. Using a unique combination of specialized hardware and a specialized programming model, game developers routinely write high performance code using millions of threads. Each generation of graphic processors (GPU's) delivers higher performance and is more programmable then the last. Unlike CPU's, these processors are designed from the beginning to run highly parallel programs and fit a different programming model. We would claim that a GPU is much more like a traditional supercomputer then a desktop processor. The programming model for graphics, called the graphics pipeline, is quite different from any of the CPU models. We will include a quick overview of this model. We believe that the Graphics programming model for utilizing the GPU will continue to be radically different from the CPU programming model for some time to come. The key feature that makes games perform well (at least to compiler writers) is that games are shipped as byte code and JIT compiled each time the game is started.

The main focus of this talk is the AMD provided JIT compiler. We will describe some of the design decisions made in the JIT compiler and give some statistics on how well they worked. For instance we will explain what the compiler does with 256 thousand registers and why it would like more. Finally we will highlight some of the open research problems raised by graphics hardware.

Finally we will provide some insight into how a commercial compiler for graphics is tested before production hardware is available.

Biography:

Dr Rubin is a fellow at AMD where he is the main architect for the AMD/ATI graphics compiler. He has built commercial compilers for processors ranging from embedded (ARM), desktop (HP, ALPHA) and supercomputer (KSR). He has published numerous papers on compiler design. Norm holds a PhD from the Courant Institute of NYU. Besides his work in compilers, he is well known for his work in compiler related parts of the tool chain, binary translators and dynamic optimizers.

Outside Olsson Hall 009 at 10:30 a.m.