Naturally, I have to cite anything (or at least anything scholarly) that
even attempts to develop an open-architecture GPU.  I'm coming up just
short of bone dry, but here are some random things I've found so far that
might be slightly interesting:

I knew that someone was trying something with the LEON core, and here's a
2012 article on it:
http://www.comp.nus.edu.sg/~wongwf/papers/FPT12.pdf
But LEON is a SPARC processor.  This isn't a GPU.  It's GPU-like, but in a
very weak sort of way.  This reminds me of BarraSim that tries to make a
GPU simulator out of PowerPC cores.

This paper HINTS a the idea of making a synthesizable GPU:
http://casmail.ee.ncku.edu.tw/research/publications/CASLab_2010_CNF_01.pdf

This one does some geometry calculations in FPGA:
https://www.scss.tcd.ie/Michael.Manzke/fyp2004-2005/DanielMcKeon.pdf

This appears to purport to do ray-tracing in hardware:
http://opencores.org/project,theia_gpu
I looked at the code, and it's a mess, so I can't be sure what it does, but
it isn't architected like a GPU, as far as I can tell.

This one ALMOST looks interesting, until you look at the code and realize
it's just bits and pieces of some course assignment, missing vital
components:
http://code.google.com/p/verilogpipelinedgpu/source/checkout

Random discussion started from 2009 by someone wanting to design their own
GPU:
http://beyond3d.com/showthread.php?t=55644

There are tons of papers that use GPUs to accelerate all sorts of things,
including simulation, synthesis, and functional verification, which is
nothing like what we're doing.  This may be drowning out my searchers.
Examples:
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6105404
http://developer.download.nvidia.com/GTC/PDF/GTC2012/PresentationPDF/S0050-GTC2012-High-Logic-Simulation.pdf



-- 
Timothy Normand Miller, PhD
Assistant Professor of Computer Science, Binghamton University
http://www.cs.binghamton.edu/~millerti/
Open Graphics Project
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to