Re: Nanopi R5S and U-boot from Ports

Mark Millard <[email protected]>
Newsgroups gmane.os.freebsd.devel.arm
Message-ID <[email protected]>
On Feb 12, 2025, at 09:59, Florian Smeets <[email protected]> wrote:

> On 11.02.25 17:20, Søren Schmidt wrote:
>> I use several of those boards and they work OK, granted with at modified src tree…
> 
> Can we get what's needed for these boards committed to the tree?

My understanding is that FreeBSD does not maintain any *.dts* sources
or *.dtb binaries of itself. It uses what it considers upstream: For
all but the RPi*'s: mainline Linux the source directory tree(s) for
the *.dts* files are copied over. Those supported are built as part
of FreeBSD building for the platform involved. For the RPi*'s binary
files are grabbed via the sysutils/rpi-firmware port/package and
those are used. The port Makefile controls which vintage it gets and
normally it is an older vintage (rarely updated).

The mainline Linux tracking does not really allow for using different
vintages of mainline for different small arm boards.

Also, any FreeBSD kernel update to track Device Tree changes is a
separate effort that only happens at some point after discovering
any incompatibilities with the pre-existing kernel code. Similar
status for something new, like the RPi5 variations: the kernel
has no expect support and any *.dtb is far from sufficient on its
own.

U-Boot is a separate issue but is also involved. To my knowledge,
FreeBSD does not claim any support of a small arm board via the
likes of EDK2, neither UEFI/ACPI nor UEFI/DeviceTree style. (Device
Tree style need not track mainline Linux.) But there is a EDK2
port with flavors for a few contexts, even if FreeBSD makes no use
of any of the flavors.

> I'm also using an OrangePi5, I was able to get a stable/14 working with patches gathered from https://personalbsd.org/images/download/.

If the patches aren not based on tracking mainline Linux on the
schedule FreeBSD ends up using for updates, it seems unlikely
that FreeBSD will go that direction: no signficant, sustained
deviation from the upstream context seems likely.

===
Mark Millard
marklmi at yahoo.com
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.