NyuziRaster: Optimizing rasterizer performance and energy in the Nyuzi open source GPU

The Intel Larrabee GPU was not a traditional GPU. Designers reasoned that many functions typically implemented in dedicated hardware could instead be done efficiently in software using a wide vector ALU. Among these functions was rasterization. To compensate for the absence of a dedicated rasteriser, Intel engineers designed a recursive rasterizer algorithm that efficiently utilizes Larrabee's compute resources. However, researchers recently have shown software rasterization to have substantial overhead for some workloads.

[1]  David Defour,et al.  Barra: A Parallel Functional Simulator for GPGPU , 2010, 2010 IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems.

[2]  Marc Olano,et al.  Triangle scan conversion using 2D homogeneous coordinates , 1997, HWWS '97.

[3]  Abdullah Al-Dujaili,et al.  Guppy: A GPU-like soft-core processor , 2012, 2012 International Conference on Field-Programmable Technology.

[4]  David R. Kaeli,et al.  Multi2Sim: A simulation framework for CPU-GPU computing , 2012, 2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT).

[5]  Edward T. Grochowski,et al.  Larrabee: A many-Core x86 architecture for visual computing , 2008, 2008 IEEE Hot Chips 20 Symposium (HCS).

[6]  Bryan D. Ackland,et al.  The edge flag algorithm — A fill method for raster scan displays , 1981, IEEE Transactions on Computers.

[7]  Juan Pineda,et al.  A parallel algorithm for polygon rasterization , 1988, SIGGRAPH.

[8]  Carlos González,et al.  ATTILA: a cycle-level execution-driven simulator for modern GPU architectures , 2006, 2006 IEEE International Symposium on Performance Analysis of Systems and Software.

[9]  Karthikeyan Sankaralingam,et al.  MIAOW - An open source RTL implementation of a GPGPU , 2015, 2015 IEEE Symposium in Low-Power and High-Speed Chips (COOL CHIPS XVIII).

[10]  Aaron Carpenter,et al.  Nyami: a synthesizable GPU architectural model for general-purpose and graphics-specific workloads , 2015, 2015 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS).