HI ,
I have an RK3588 board running Debian, and I would like to use it with an Axelera M.2 AI card. However, I cannot find any related installation or porting documentation.
Does anyone have experience with this?
Thank you.
Solved
Howto install driver into RK3588 board?
Best answer by Victor Labian
Hi
We are currently working on enabling more OS flexibility, so stay tuned!
This said, you can work with Debian by using Docker, which will not affect your performance or memory usage.
Please find the documentation on how to use Voyager SDK with Docker here: https://support.axelera.ai/hc/en-us/articles/25953148201362-Install-Voyager-SDK-in-a-Docker-Container
In that guide it explains how to get the driver and install it. It will work on Debian ;)
May I ask which RK3588 host you are using? I have indeed worked on supporting RK3588 hosts like the OrangePi5 or the NanoPC-T6. I created a guide for those two hosts (it can help you as reference). The main think to notice is that you will probably need to modify the device tree. But don’t worry, as that is explained in both guides:
- https://support.axelera.ai/hc/en-us/articles/31859388491794-Bring-up-Voyager-SDK-in-NanoPC-T6
- https://support.axelera.ai/hc/en-us/articles/27059519168146-Bring-up-Voyager-SDK-in-Orange-Pi-5-Plus
I hope you find this info helpful.
Let us know if you have any more questions.
Best,
Victor
+3- Axelera Team
Hi there
Ah, sounds cool - there’s been some experimentation going on around Debian, from what I’ve seen.
These are both RK3588 boards - maybe these can shed some light on things, and if metis would play nice with your host?
HI
Thank you. I also noticed that Axelera has developed on the RK3588 (ARM) platform and is even selling boards with the same CPU. For example, the Metis M.2 Eval System with Aetina RK3588 Industrial Motherboard | Axelera AI Store. So I assume the driver should already be available. However, I am not sure whether it is designed for Yocto or Debian.
I just cannot find the official download location. That is why I am asking on the forum to see if anyone has relevant experience to share.
However, I also found that the RK official support seems to be Debian only, while the documents you provided are all based on Ubuntu. I would like to ask whether Axelera can provide the driver source code so that we can port it to the Debian file system.
Metis M.2 Eval System with Aetina RK3588 Industrial Motherboard | Axelera AI Store
+3- Axelera Team
Indeed, yeah, we don’t have official support for Debian (yet!) although maybe something like this could be of assistance in building it?
I know a couple of people have managed it, but I couldn’t say exactly what they needed to do.
And I think the driver source code is included when you install the Voyager SDK. It gets installed via DKMS, so after installation it’d be in /usr/src/metis-{version}/driver/, which you can potentially use to rebuild for your kernel.
Let me know if this helps, and how it goes!
Hi
We are currently working on enabling more OS flexibility, so stay tuned!
This said, you can work with Debian by using Docker, which will not affect your performance or memory usage.
Please find the documentation on how to use Voyager SDK with Docker here: https://support.axelera.ai/hc/en-us/articles/25953148201362-Install-Voyager-SDK-in-a-Docker-Container
In that guide it explains how to get the driver and install it. It will work on Debian ;)
May I ask which RK3588 host you are using? I have indeed worked on supporting RK3588 hosts like the OrangePi5 or the NanoPC-T6. I created a guide for those two hosts (it can help you as reference). The main think to notice is that you will probably need to modify the device tree. But don’t worry, as that is explained in both guides:
- https://support.axelera.ai/hc/en-us/articles/31859388491794-Bring-up-Voyager-SDK-in-NanoPC-T6
- https://support.axelera.ai/hc/en-us/articles/27059519168146-Bring-up-Voyager-SDK-in-Orange-Pi-5-Plus
I hope you find this info helpful.
Let us know if you have any more questions.
Best,
Victor
Sign up
Already have an account? Login
Log in, or create an Axelera AI account
Log In or Register HereEnter your E-mail address. We'll send you an e-mail with instructions to reset your password.