TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

VeriGPU: GPU in Verilog loosely based on RISC-V ISA

101 pointsby btdmasterabout 3 years ago

9 comments

mrintellectualabout 3 years ago
Even as an ML-focused graphics-less GPU, this is great. If this can be prototyped on an FPGA, it would be even better. Using block RAM for shared memory and built-in PCIe and DDR IP blocks should help speed things up considerably.<p>It unfortunately wouldn&#x27;t be very cost-effective for training ML models, but it would take things a step closer to actual tape-out (if some organization has the $$$ for it).
tux3about 3 years ago
Worth noting this is targeting ML applications, so I don&#x27;t think you&#x27;ll be able to display even a text console with it for the foreseeable future.<p>But I love that this is even in the realm of possibilities! There&#x27;s no reason we couldn&#x27;t, in principle, have a small open-source GPU taping out on the free Skywater shuttle, and I am here for it!
评论 #31137421 未加载
评论 #31140181 未加载
gnufxabout 3 years ago
Perhaps also see the (OpenPOWER-based) Libre-SOC effort <a href="https:&#x2F;&#x2F;libre-soc.org&#x2F;" rel="nofollow">https:&#x2F;&#x2F;libre-soc.org&#x2F;</a>
yjftsjthsd-habout 3 years ago
&gt; Internal GPU Core ISA loosely compliant with RISC-V ISA. Where RISC-V conflicts with designing for a GPU setting, we break with RISC-V.<p>Very amateur question: I thought RISC-V added vector extensions so you could use it directly for GPU&#x2F;TPU chips without having to fragment the ecosystem?
评论 #31139193 未加载
Sirenedabout 3 years ago
Neat! Looks like it&#x27;s very much in its early stages (no concurrent execution&#x2F;threads yet) but it&#x27;s so great to see FOSS digital design work in an industry dominated by huge players
avianesabout 3 years ago
I&#x27;m not really optimistic about the hardware and the tape-out goal. The author seems to have a very basic knowledge about it.<p>For instance, the int multiplier design is overengineered, very naive and far from state of the art (no pipeline, no adder compressor). I would suggest the author to check wallace tree multiplier.<p>But at this stage it would be preferable to use the native verilog multiply or a DSP macro to target FPGA for prototyping, and to focus the SIMT architecture and the pipelining. Arithmetic unit design is a science in its own.<p>However it&#x27;s a beautiful project!
terafoabout 3 years ago
I&#x27;m interested whether it would be a good idea to implement a Vulkan driver for such GPU via emulating TMUs and ROPs in software, and it might not even matter that much since modern rendering pipelines are more and more compute reliant anyways(UE5 Nanite barely uses hardware rasterizers, latest idTech uses software rasterization as well). The only problem I see lies is with raytracing since it is pretty reliant on fixed function units.
评论 #31144765 未加载
评论 #31143182 未加载
atq2119about 3 years ago
From the planning document:<p>&gt; Branching: Done<p>&gt; Single instruction multiple thread (SIMT): Planned<p>I guess we should be supportive, and it is impressive how far they got on the software side, but boy, is the author in for a surprise.
lizardactivistabout 3 years ago
Very cool, but can someone remind me what &quot;GPU&quot; means again?
评论 #31138357 未加载
评论 #31138415 未加载