[RESEND PATCH v11 13/18] drm: exynos: dsi: Add Exynos based host irq hooks

[Jagan Teki]

On Wed, Jan 25, 2023 at 10:57 PM Marek Vasut <marex at denx.de> wrote:
>
> On 1/25/23 18:12, Jagan Teki wrote:
> > On Wed, Jan 25, 2023 at 10:16 PM Marek Vasut <marex at denx.de> wrote:
> >>
> >> On 1/25/23 15:04, Jagan Teki wrote:
> >>> On Wed, Jan 25, 2023 at 7:23 PM Marek Vasut <marex at denx.de> wrote:
> >>>>
> >>>> On 1/25/23 07:54, Jagan Teki wrote:
> >>>>> On Wed, Jan 25, 2023 at 2:54 AM Jagan Teki <jagan at amarulasolutions.com> wrote:
> >>>>>>
> >>>>>> On Wed, Jan 25, 2023 at 2:54 AM Jagan Teki <jagan at amarulasolutions.com> wrote:
> >>>>>>>
> >>>>>>> On Wed, Jan 25, 2023 at 2:42 AM Marek Vasut <marex at denx.de> wrote:
> >>>>>>>>
> >>>>>>>> On 1/24/23 22:01, Jagan Teki wrote:
> >>>>>>>>> On Wed, Jan 25, 2023 at 2:18 AM Marek Vasut <marex at denx.de> wrote:
> >>>>>>>>>>
> >>>>>>>>>> On 1/23/23 16:12, Jagan Teki wrote:
> >>>>>>>>>>> Enable and disable of te_gpio's are Exynos platform specific
> >>>>>>>>>>> irq handling, so add the exynos based irq operations and hook
> >>>>>>>>>>> them for exynos plat_data.
> >>>>>>>>>>
> >>>>>>>>>> If this is just an optional generic GPIO IRQ, why not keep it in the
> >>>>>>>>>> core code ? TE (tearing enable?) should be available on MX8M too.
> >>>>>>>>>
> >>>>>>>>> So far the discussion (since from initial versions) with Marek
> >>>>>>>>> Szyprowski, seems to be available in Exynos. So, I keep it separate
> >>>>>>>>> from the DSIM core.
> >>>>>>>>
> >>>>>>>> Isn't TE a generic GPIO IRQ ? If so, it is available also on i.MX8M .
> >>>>>>
> >>>>>> I will check this.
> >>>>>
> >>>>> In order to use TE_GPIO we need te handler implementation, right now
> >>>>> Exynos CRTC DRM drivers have implementation for this. That is the main
> >>>>> reason to keep the TE_GPIO handling in exynos, maybe if we handle that
> >>>>> generically then it is a viable option to move TE_GPIO to the DSIM
> >>>>> core.
> >>>>
> >>>> I think you can do this exactly the same way exynos does it -- check
> >>>> whether te_handler() callback is implemented by the glue code (the one
> >>>> you already have for various exynos and imx8mm/imx8mm SoCs) and if so,
> >>>> call it. If it is not implemented, do not call anything in the TE IRQ
> >>>> handler.
> >>>
> >>> I need to understand how i.MX8MM handles this on TE IRQ in the DSIM
> >>> host side, Can I do this in future patch set as it might involve
> >>> bindings changes as well if it's part of DSIM?
> >>
> >> Why not leave an empty te_handler implementation on MX8M for now ?
> >> You can fill that implementation in future patchset, but the generic
> >> part of the code would be in place .
> >
> > Look like we have one issue to move regster te_irq into samsung-dsim.
> >
> > exynos_dsi_register_te_irq is done after the bridge attach is done in
> > Exynos, here bridge attach is triggered in the component ops bind
> > call, since samsung-dsim is a pure bridge w/o any component ops.
> > https://github.com/openedev/kernel/blob/imx8mm-dsi-v12/drivers/gpu/drm/bridge/samsung-dsim.c#L1527
> > https://github.com/openedev/kernel/blob/imx8mm-dsi-v12/drivers/gpu/drm/exynos/exynos_drm_dsi.c#L112
> >
> > Any suggestion on how to handle this?
>
> Why isn't the generic code calling drm_bridge_attach() in
> samsung_dsim_host_attach(), like the exynos one ?

Exynos drm drivers follow component ops and generic dsim is a pure drm
bridge whose downstream bridge will attach in bridge ops attach and
the component-based drivers require an initial bridge attach (whose
previous is NULL) call in the component bind hook for establishing the
bridge chain.

Jagan.