Re: RPI 4/5 u-boot new port, is there any interest?
- Reply: Mark Millard : "Re: RPI 4/5 u-boot new port, is there any interest? [16 GiByte RPi5B (D0) basics operational via the EDK2 variant for such]"
- Reply: Klaus_Küchemann : "Re: RPI 4/5 u-boot new port, is there any interest?"
- In reply to: Klaus_Küchemann : "Re: RPI 4/5 u-boot new port, is there any interest?"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Thu, 05 Jun 2025 05:15:57 UTC
On Wed, 4 Jun 2025 at 10:52, Klaus Küchemann <maciphone2@googlemail.com>
wrote:
> Yes, Mark,
> you've been around for so long ;-), you know how things work here...
> The discussion is as old as the Pi itself...
> No agreement on which bootloader, patches, or upstreams to use,
> so I can stop working on it for now.
> Maybe not the worst news for me personally :-), hehe
>
Hi,
working solutions beat ideal solutions.
If y'all have a patched chain-loading u-boot that can then itself chainload
our loader and have that chainload freebsd, then please:
* create a git tree somewhere;
* get an upstream copy of u-boot 2025.04 into it;
* add whatever diffs you need to get freebsd's loader chainloading from it;
* add documentation on how it works;
* let's get it working.
The RP1 silicon looks cool; I wish I had time to go write stuff for that
but my plate is full.
Don't worry about the "correct freebsd way", we already have a bazillion
u-boot 2025.04 forked ports ("pkg search u-boot" on your local freebsd).
I know people prefer a UEFI interface, and if someone has a maintained one,
great! create a port and ship it!
But otherwise, I really just recommend the "smash the keyboard to
demonstrate it working, and let's get it into a bootloader port" path.
I'd hate to see work go to waste. Let's get it into a github repo forked
off of 2025.04 to match the rest of our current bootloader ports
and churn on whatever's needed to get it done and shipping.
-adrian