Hi, On Wed, Aug 7, 2024 at 5:39 AM Michael Nazzareno Trimarchi <mich...@amarulasolutions.com> wrote: > > Hi Doug > > +cc Doug > > I have seen that you have done some re-working and investigation on > drm stack, do you have some > suggestion on this case? > > On Mon, Jun 24, 2024 at 8:53 PM Michael Trimarchi > <mich...@amarulasolutions.com> wrote: > > > > The shutdown function can be called when the display is already > > unprepared. For example during reboot this trigger a kernel > > backlog. Calling the drm_panel_unprepare, allow us to avoid > > to trigger the kernel warning > > > > Signed-off-by: Michael Trimarchi <mich...@amarulasolutions.com> > > --- > > > > It's not obviovus if shutdown can be dropped or this problem depends > > on the display stack as it is implmented. More feedback is required > > here
In general the shutdown should be dropped and it should be up to the display driver to do the shutdown. If your panel needs to be used with a DRM Modeset driver that doesn't properly call shutdown then the ideal solution would be to fix the DRM Modeset driver. If this is somehow impossible, I suspect folks would (begrudgingly) accept some other solution. >From a super quick look, I see: * This panel seems to be used upstream by "imx8mn-bsh-smm-s2-display.dtsi" * In "imx8mn.dtsi" I see "lcdif" is "fsl,imx6sx-lcdif". * "fsl,imx6sx-lcdif" seems to be handled by "drivers/gpu/drm/mxsfb/mxsfb_drv.c" * Previously I determined that "mxsfb-drm" was indeed calling drm_atomic_helper_shutdown() properly [1] ...so it seems like just dropping the shutdown handler in this panel is correct. [1] https://lore.kernel.org/r/20240611074846.1.Ieb287c2c3ee3f6d3b0d5f49b29f746b93621749c@changeid -Doug