On Wed Jul 30, 2025 at 9:50 PM CEST, Dragan Simic wrote:

I've spent some time digging through the LED subsystem, which I'm
already somewhat familiar with, and I think I've narrowed down the
root cause of this deadlock.

I'll send a preliminary patch soon, after I make sure that the root
cause is identified correctly, and I hope Diederik will be willing
to test the patch.  If so, and if the patch checks out to be the
cure, I'll prepare and submit a proper patch, of course.

Sure, but AFAIC you can send it to the ML(s) directly as well.
Please do CC me as I'm only subscribed to the linux-rockchip ML.

Cheers,
  Diederik