Partially revert "rgbmatrix: Avoid leaving an incompletely configured display bus" #9892
Conversation
|
I loaded the matrix S3 artifact and it appears to have resolved the safe mode issue. Thanks! |
|
Well... This artifact doesn't have Dan's USB fix so maybe that's contributing to this latest behavior, but using this artifact, when I don't have the USB cable connected the matrix doesn't initialize, and if I disconnect the USB cable while the RGB matrix is active the board appears to crash (it's hard to tell exactly what's happening since the USB port isn't working for terminal or CIRCUITPY use without Dan's fix). I've recently replaced my computer and perhaps this is an excuse to test a CircuitPy build. I think I need to build firmware with both Dan's fix and this reversion..... |
jepler
force-pushed
the
revert-9865-rgbmatrix-cleanup-failure
branch
from
57169fc to
62f8923
Compare
jepler
changed the title
|
I updated the branch to leave the itsy config changes & merge latest main. |
|
@RetiredWizard My USB fix #9889 was just merged, but that is not in this PR (oh maybe it was just merged in??) |
|
I cloned CP after Scott merged Dan's fix and then built using Jeff's changes here and everything seems to be working great. The issue I was seeing when unplugging the USB looks like it was on my end. Either one of my power bricks died or I pulled loose one of the cables.... (The bad power brick just happened to be the one powering the micro controller so it worked fine if the USB was connected and supplying power). |
There was a problem hiding this comment.
@RetiredWizard thanks for testing. I'll merge if/when the builds finish successfully.
Reverts #9865
According to @RetiredWizard this led to the problems reported in #9891.
Since the fix was only for failed matrix initialization, it seems we're better off just reverting it until it can be solved again properly.