[Mesa-dev] llvm-pipe : how fast on this new chip?

[not42 at gmx.com]

How well would llvm-pipe run on the"Paralella" chip? 
 See this kickstarter project below. Looks like it might actually get funded.
http://www.kickstarter.com/projects/adapteva/parallella-a-supercomputer-for-everyone/ 
 Ignoring for a moment there is no LLVM backend for this CPU yet, how well does the llvm-pipe architecture map this chip's capability?
 It's sort of like a "mobile phone" class low-cost 32-bit RISC version of the Intel MIC.
 In short:
 - 2 ARM host cores.
 - 16 compute cores @ 800Mhz (64 on better model)
 - Small amount of embedded SRAM per compute core.
 - Each core can issue both one Integer and one FP instruction per cycle. (Including FMA)
 - Flat 64 entry register file.
 - All cores on 2D mesh fabric and share one address space.
 - No SIMD, sadly.
 Currently they only have GCC running for this chip, but are looking at LLVM.
 Just based on the profile of the typicle Gnome 3 style GUI workload, how well might llvm-pipe run on this chip?
 Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/mesa-dev/attachments/20121026/b8716e55/attachment.html>