How to accelerate `make prog` run time

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

453 stars 116 forks source link

How to accelerate `make prog` run time #789

Open Siris-Limx opened 1 year ago

Siris-Limx commented 1 year ago

When first run make prog, it will take a long time (about 6~7 mins) to finish synthesis, place&route, etc. I find out that during this process, only one CPU core is used, leaving other 7 cores idle. My question is can we add some parameters to let all CPUs run during make prog? My working environment is VM Ubuntu 18.04, 10G RAM, 8 CPU cores.

tcal-x commented 1 year ago

Hi @limingxuan-pku , unfortunately there is no easy way to parallelize that I know of. To the best of my knowledge, the F4PGA / Symbiflow tools are all single-threaded. @mithro , is that correct?