On 6/30/26 00:21, Tom Rini wrote:
On Fri, Jun 26, 2026 at 08:07:16PM +0200, Quentin Schulz wrote:
[snip]
I'll finish that my gut feeling (shared with some of my current and former
colleagues) is that U-Boot is disappointingly unstable. Unlike the Linux
kernel, U-Boot suffers from fewer eyes looking at code, fewer people
developing it, fewer people testing master or even any recent release (the
number of reports we have for bugs on releases from years ago...), and lack
of proper and extensive CI test infrastructure (and if we have some, it's
heavily centralized into a handful of people's offices). I loathe every time
I need to update U-Boot because I don't know what I'll have to debug for
days or if I really do test everything there's to test (is it my fault for
not keeping up with master, yes, but I shouldn't feel this way).
So I really think we could have much more confidence in code being merged if
we had more testing, possibly on real hardware, possibly by companies
offering to do some proper CI (e.g. like Intel is doing with Yocto, spending
days testing release candidates before the project tags a release). Maybe
with distros slowly adopting Aarch64 and U-Boot with it we'll have more
coverage. I know you've worked on labgrid support for U-Boot and you have
some merge request(s) still open there but it seems you're hitting the same
wall there you're hitting here.
Thanks for the feedback here. I've been staying out of this general
thread as I believe my thoughts are well known already.
For the specific topic here, part of it comes down I believe to a very
uneven level of testing. Both AMD/Xilinx and TI (and in turn, a number
of their vendors) have labs and run tests with a good deal of frequency.
I suspect NXP has some testing going on too, and this isn't an intended
as an exhaustive list. But others really are very much volunteer lead
and getting a reliable and remote lab setup really is a lot of work, so
no one has done it. Rockchip is an unfortunate example here that I know
you run in to. I have an rk3399 I've been trying to find time and space
to get up in the sage lab, for example, but it's still just not there.
This is a topic which we should talk about more.
It is clear that we can't expect that internal labs will join our project
because it is very difficult to provide an access through infrastructure.
Also you can't wait for companies to finish their testing before release.
But on the other hand developers have enough HW on their desks which shouldn't
be that difficult to share. That's why I think we should provide much easier
way how to share these HW and wire it up with u-boot CI. That could also allow
to connect custom qemu models instead of just upstream version.
Thanks,
Michal
