Native commands to run code on RISC-V soft CPU on FPGA

0

Is there a tutorial that uses AWS commands showing how to:

  1. run Linux on a chipyard RISC-V SoC loaded on to the FPGA ?
  2. run riscv elf on a chipyard RISC-V SoC loaded on to the FPGA?

Something like this (atamaran/ARIFIC) tutorial

https://github.com/rsnikhil/Tutorial_at_HPCA-29/blob/main/Slides.pdf

which uses their own custom commands:

  • ./UART_Console for uart connection
  • ./exe_Host_AWSF1 to run
    • assert_reset
    • load_elf ../C_Examples/hello/hello.RV64.bare.elf
    • deassert_reset
  • ./exe_Host_AWSF1. --elf ./Elfs/Linux.elf --blockdev ./disk.img --tundev tap0
asked a month ago66 views
2 Answers
0
Accepted Answer

I believe I have worked something out. In case it helps anyone else, I'll update this with a link when it is made public.

answered a month ago
-1

I apologize, but I don't have information about specific native commands or tutorials for running Linux or RISC-V ELF files directly on AWS FPGA instances in the way you've described.

AWS FPGA instances (F1 and F2) are designed to run custom hardware accelerations, not to directly boot operating systems or execute ELF files as you might on a typical processor. The process for using FPGAs on AWS is quite different from the custom commands you've mentioned.

With AWS FPGA instances, you typically:

  1. Develop your custom logic using hardware description languages or high-level synthesis tools.
  2. Use the AWS FPGA Development Kit to create an Amazon FPGA Image (AFI).
  3. Deploy this AFI to an F1 or F2 instance.
  4. Interact with the FPGA using your application code running on the instance's CPU.

If you want to implement a RISC-V core or Linux support on an AWS FPGA, you would need to design this as part of your custom logic, then create and deploy an AFI containing this design. The interaction would then happen through your application code, not through direct commands to the FPGA.

For getting started with AWS FPGA instances, you would use the AWS FPGA Development Kit and the FPGA Developer AMI. These provide the necessary tools for developing, simulating, and deploying your FPGA designs on AWS. However, the workflow is quite different from the custom commands you've referenced from the ARIFIC tutorial.
Sources
F2
Amazon EC2 examples using AWS CLI - AWS SDK Code Examples
Amazon EC2 F1 Instances
Now Available – Second-Generation FPGA-Powered Amazon EC2 instances (F2) | AWS News Blog

profile picture
answered a month ago
  • I have clarified the question to make clear that a ChipYard RISC-V SoC has been loaded onto the FPGA.

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions