On Wed, Jan 27, 2021 at 2:10 AM Carvel Baus <
cbaus@xxxxxxxxxx> wrote:
My current interests lie in 4 primary areas: Arm, Rust, Kernel, Apple M1. Where two or more of these intersect would be very interesting to me. In no particular order:
- Fedora on Apple Silicon
That's a long way out, the first of the enablement patches have just
been posted for review and that doesn't even have storage to boot to a
usable system, it basically just enough to start CPUs and output to a
console. From experience of watching other platform bootstraps this is
going to be a multi year approach to get to something we can actively
support in Fedora.
- Rust on Arm
The rust support on arm is pretty good, it's now an upstream Tier 1
platform, and overall it's on equal footing as far as I can tell in
Fedora. There's a pretty active rust SIG:
https://fedoraproject.org/wiki/SIGs/Rust- Fedora Kernel (C)
That's a pretty broad topic.
and down the road: Rust for kernel modules
Ultimately, I’d like to code - that could be C/C++, Rust, Scala (JVM)…probably not Python.
I don’t know if there are any plans for Fedora on Apple Silicon but this seems like an obvious target if there are. I am open to helping however I can but writing code is a primary objective. I am certainly a team player so open to discussing the pressing needs and see where there is a good fit.
The answer is two fold, the first is most certainly, the second it
when it's ready. While there's a means of using it from an enthusiast
PoV it's no where near close to anything we can actively support in
Fedora. As described above kernel upstreaming is only just starting to
begin, and GPU reverse engineering is starting to make baby steps [1],
based on experience from other SoCs, such as Raspberry Pi, and other
GPUs such as MALI this is going to be a mult year process to get
something in main Fedora that's supportable by the average user with a
reasonable experience, I'm sure well before then there will be
terrible hacked up versions of Fedora that blitt via a framebuffer for
un-accelerated graphics which will be great for people that are
developing on them or enjoy terrible user experiences just to say they
can do it, but that's unsupportable in main Fedora because it ends up
creating a lot of support work for those that support the Fedora Arm
initiative.
Now Fedora Arm in a VM on an Apple M1 Mac is definitely something that
I'd love to support, and while my time to actively hack on it of late
has been limited, it's certainly something that I think is achievable
in the short term as a step on the route to full bare metal
enablement.
