The first patch in this series fixes dereferencing a NULL mode in the error path. The second one extends the resets to not only reset RESX but also TP_RSTN since otherwise the display will stay completely blank. I didn't spot that before initial submission since the reset line was bound to the touch controller and although that failed to probe it came far enough to deassert the reset line (at an arbitrary point in time during boot) and hence we got a picture. Since touch and panel are on the same IC they're not completely independent and i might have to turn the whole thing into an MFD at some point but this series gets the panel to a reliably working state on boot and on fb blank/unblank.
Since the reset-gpios are active low we can deassert in prepare and assert in unprepare simplifying the code making sure lines are kept low when the panel is off.
The binding were not part of a stable kernel so I hope it's okay to not worry about backward compatibility.
Changes from v1: - As per review comments by Fabio Estevam https://lore.kernel.org/dri-devel/CAOMZO5B5ECcConvKej=RcaF8wvOxgq7nUzKJ-ad0a... - Fix typo in commit messages - As per review comments by Rob Herring https://lore.kernel.org/dri-devel/20200929174624.GA832332@bogus/ - Don't use an array of reset lines
Guido Günther (3): drm/panel: mantix: Don't dereference NULL mode drm/panel: mantix: Fix panel reset dt-binding: display: Require two resets on mantix panel
.../display/panel/mantix,mlaf057we51-x.yaml | 4 +++ .../gpu/drm/panel/panel-mantix-mlaf057we51.c | 25 +++++++++++++------ 2 files changed, 21 insertions(+), 8 deletions(-)