Garnet
Next generation CGRA generator
Install / Use
/learn @StanfordAHA/GarnetREADME
This repo lets you investigate and experiment with implementing our CGRA using new generator infrastructure. Here you will find: original Genesis2 source for top level modules, functional models, and testing infrastructure. Also, you will find common generator patterns abstracted away to make designing, testing, and programming the CGRA faster.
Usage
Once garnet is installed, you can build e.g. a 2x2 CGRA simply by doing
$ python garnet.py --help
$ python garnet.py --width 4 --height 2
For installation instructions, read on.
Install and Build (also see issue https://github.com/StanfordAHA/garnet/issues/1037)
We use a docker environment to build the chip. Here's how:
Use AHA repo to boot up a docker image and container
$ git clone https://github.com/StanfordAHA/aha aha
$ cd aha; git submodule init update --recursive
$ docker build . -t aha_image # May need sudo depending on your setup
$ docker exec -it aha_container bash # May need sudo depending on your setup
(Inside docker now) build a 4x2 CGRA garnet.v
$ source /aha/bin/activate
$ cd /aha/garnet
$ python garnet.py --width 4 --height 2 --verilog
Verify functionality
We can verify that everything is setup properly by running the test suite using pytest.
$ cd /aha; ./garnet/.github/scripts/run_pytest.sh
(FYI last time I tried this it did not work until I first did cd /aha/garnet; git checkout master ...?)
Style guide
Please read and follow the style guide in the AHA wiki.
Related Skills
node-connect
349.9kDiagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps
frontend-design
109.8kCreate distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
openai-whisper-api
349.9kTranscribe audio via OpenAI Audio Transcriptions API (Whisper).
qqbot-media
349.9kQQBot 富媒体收发能力。使用 <qqmedia> 标签,系统根据文件扩展名自动识别类型(图片/语音/视频/文件)。
