This patch addresses pre-existing issues that came up during the review process of Sankeerth's series trying to add eDP for Qualcomm SoCs [1].
It's really sorta two series but jammed into one. The first two patches fix a problem with ps8640 when the panel doesn't finish probing right away. The rest of the patches attempt to improve how eDP panel drivers deal with the HPD signal. NOTE: if everyone hates the "generic driver" that I added in the first patch, I have a different version that just adds uses the Linux auxiliary bus stright in ps8640. I'm happy to switch back to that, but it seemed like a buncha copy-pasta that I was hoping to avoid.
I haven't done a crazy amount of testing with this, but it seems to work and I wanted to get something out there. I'll try to do some more testing next week. This is why I added the tag "RFC". It's entirely possibled that I've actually caught all the bugs and this is great, but I just wanted to be sure.
This _doesn't_ attempt to fix the Analogix driver. If this works out, ideally someone can post a patch up to do that.
[1] https://lore.kernel.org/r/1648656179-10347-2-git-send-email-quic_sbillaka@qu...
Douglas Anderson (6): drm/dp: Helpers to make it easier for drivers to use DP AUX bus properly drm/bridge: parade-ps8640: Break probe in two to handle DP AUX better drm/dp: Add is_hpd_asserted() callback to struct drm_dp_aux drm/panel-edp: Take advantage of is_hpd_asserted() in struct drm_dp_aux drm/panel: atna33xc20: Take advantage of is_hpd_asserted() in struct drm_dp_aux drm/bridge: parade-ps8640: Provide is_hpd_asserted() in struct drm_dp_aux
drivers/gpu/drm/bridge/parade-ps8640.c | 87 +++++---- drivers/gpu/drm/dp/drm_dp_aux_bus.c | 165 +++++++++++++++++- drivers/gpu/drm/panel/panel-edp.c | 37 +++- .../gpu/drm/panel/panel-samsung-atna33xc20.c | 35 +++- include/drm/dp/drm_dp_aux_bus.h | 58 ++++++ include/drm/dp/drm_dp_helper.h | 14 ++ 6 files changed, 353 insertions(+), 43 deletions(-)