search

google / CFU-Playground

Want a faster ML processor? Do it yourself! -- A framework for playing with custom opcodes to accelerate TensorFlow Lite for Microcontrollers (TFLM). . . . . . Online tutorial: https://google.github.io/CFU-Playground/ For reference docs, see the link below.
http://cfu-playground.rtfd.io/
Apache License 2.0
470 stars 120 forks source link

Can we run other cpu-types in CFU Playground (initially w/o CFU)? #699

Open tcal-x opened 2 years ago

tcal-x commented 2 years ago

There are some (many?) places in CFU Playground that assume the CPU is VexRiscv. What if any work would need to be done to allow other Litex-supported CPUs (SERV, PicoRV32, corev*, etc) to be dropped into the SoC and run the CFU Playground software on them?

This could be useful for benchmarking initially, and pave the way for work of adding compatible CFU interfaces to these other CPUs.

FYI @ShvetankPrakash

ShvetankPrakash commented 2 years ago

Yes, I think this is very interesting for other directions and would be cool to see if it works!