On Mon Aug 31, 2026 at 4:58 PM CEST, Doug Anderson wrote: > Hi, > > On Mon, Aug 31, 2026 at 2:01 AM Luca Ceresoli <[email protected]> > wrote: >> >> Hi, >> >> On Thu Aug 27, 2026 at 6:23 PM CEST, Doug Anderson wrote: >> > Hi, >> > >> > On Tue, Aug 25, 2026 at 3:11 AM Luca Ceresoli <[email protected]> >> > wrote: >> >> >> >> This driver uses the drm_panel_bridge API and not the drm_panel API so it >> >> should select DRM_PANEL_BRIDGE and not DRM_PANEL. >> > >> > Which exact "drm_panel_bridge" API does it use? I don't see any... >> >> It calls devm_drm_of_get_bridge() [0], which is in bridge/panel.c [1], >> which is enabled by CONFIG_DRM_PANEL_BRIDGE [2]. >> >> [0] >> https://elixir.bootlin.com/linux/v7.2.2/source/drivers/gpu/drm/bridge/parade-ps8640.c#L636 >> [1] >> https://elixir.bootlin.com/linux/v7.2.2/source/drivers/gpu/drm/bridge/panel.c#L483-L515 >> [2] >> https://elixir.bootlin.com/linux/v7.2.2/source/drivers/gpu/drm/Makefile#L159 > > Ah, indeed! I missed that since it didn't have "panel" in the name. > FWIW, then I think for this patch: > > Reviewed-by: Douglas Anderson <[email protected]> > > ...but, ironically, that means that for the other patch I looked at > (ti-sn65dsi86.c) we're missing the "select DRM_PANEL_BRIDGE" since it > uses the same API call, doesn't it?
Ouch, indeed! Thanks for noticing. I will send v3 with that fixed, can I keep your R-by if there's just that change to the sn65dsi86 patch? I am also going to double check all patches for this mistake. Luca -- Luca Ceresoli, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
