[linux-sunxi] [PATCH 3/6] drm/sun4i: dsi: Add bridge support

Jagan Teki jagan at amarulasolutions.com
Wed May 22 12:01:10 UTC 2019


Hi Paul and Maxime,

On Fri, Mar 15, 2019 at 7:03 PM Paul Kocialkowski
<paul.kocialkowski at bootlin.com> wrote:
>
> Hi,
>
> On Fri, 2019-03-15 at 18:38 +0530, Jagan Teki wrote:
> > Some display panels would come up with a non-DSI output which
> > can have an option to connect DSI interface by means of bridge
> > convertor.
> >
> > This DSI to non-DSI bridge convertor would require a bridge
> > driver that would communicate the DSI controller for bridge
> > functionalities.
> >
> > So, add support for bridge functionalities in Allwinner DSI
> > controller.
>
> See a few comments below.
>
> > Signed-off-by: Jagan Teki <jagan at amarulasolutions.com>
> > ---
> >  drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c | 65 +++++++++++++++++++-------
> >  drivers/gpu/drm/sun4i/sun6i_mipi_dsi.h |  1 +
> >  2 files changed, 49 insertions(+), 17 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
> > index 0960b96b62cc..64d74313b842 100644
> > --- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
> > +++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
> > @@ -781,6 +781,9 @@ static void sun6i_dsi_encoder_enable(struct drm_encoder *encoder)
> >       if (!IS_ERR(dsi->panel))
> >               drm_panel_prepare(dsi->panel);
> >
> > +     if (!IS_ERR(dsi->bridge))
> > +             drm_bridge_pre_enable(dsi->bridge);
> > +
> >       /*
> >        * FIXME: This should be moved after the switch to HS mode.
> >        *
> > @@ -796,6 +799,9 @@ static void sun6i_dsi_encoder_enable(struct drm_encoder *encoder)
> >       if (!IS_ERR(dsi->panel))
> >               drm_panel_enable(dsi->panel);
> >
> > +     if (!IS_ERR(dsi->bridge))
> > +             drm_bridge_enable(dsi->bridge);
> > +
> >       sun6i_dsi_start(dsi, DSI_START_HSC);
> >
> >       udelay(1000);
> > @@ -812,6 +818,9 @@ static void sun6i_dsi_encoder_disable(struct drm_encoder *encoder)
> >       if (!IS_ERR(dsi->panel)) {
> >               drm_panel_disable(dsi->panel);
> >               drm_panel_unprepare(dsi->panel);
> > +     } else if (!IS_ERR(dsi->bridge)) {
> > +             drm_bridge_disable(dsi->bridge);
> > +             drm_bridge_post_disable(dsi->bridge);
> >       }
> >
> >       phy_power_off(dsi->dphy);
> > @@ -973,11 +982,16 @@ static int sun6i_dsi_attach(struct mipi_dsi_host *host,
> >       struct sun6i_dsi *dsi = host_to_sun6i_dsi(host);
> >
> >       dsi->device = device;
> > -     dsi->panel = of_drm_find_panel(device->dev.of_node);
> > -     if (IS_ERR(dsi->panel))
> > -             return PTR_ERR(dsi->panel);
> >
> > -     dev_info(host->dev, "Attached device %s\n", device->name);
> > +     dsi->bridge = of_drm_find_bridge(device->dev.of_node);
> > +     if (!dsi->bridge) {
>
> You are using IS_ERR to check that the bridge is alive in the changes
> above, but switch to checking that it's non-NULL at this point.
>
> Are both guaranteed to be interchangeable?
>
> > +             dsi->panel = of_drm_find_panel(device->dev.of_node);
> > +             if (IS_ERR(dsi->panel))
> > +                     return PTR_ERR(dsi->panel);
> > +     }
>
> You should probably use drm_of_find_panel_or_bridge instead of
> duplicating the logic here.

True, In-fact I did try this API. but pipeline were unable to bound.
Usually the panel and bridge were attached first and then the pipeline
bound would start from front-end (in A33) But in my below cases I have
seen only panel or bridge attached but no pipeline bound at all.

And I'm using drm_of_find_panel_or_bridge(host->dev->of_node, 1, 0,
&dsi->panel, &dsi->bridge); in dsi attach API.

Case-1, panel:

&dsi {
    vcc-dsi-supply = <&reg_dcdc1>;        /* VCC3V3-DSI */
    status = "okay";

    ports {
        dsi_out: port at 1 {
            reg = <1>;

            dsi_out_panel: endpoint {
                remote-endpoint = <&panel_out_dsi>;
            };
        };
    };

    panel at 0 {
        compatible = "bananapi,s070wv20-ct16-icn6211";
        reg = <0>;
        enable-gpios = <&pio 1 7 GPIO_ACTIVE_HIGH>; /* LCD-PWR-EN: PB7 */
        reset-gpios = <&r_pio 0 5 GPIO_ACTIVE_HIGH>; /* LCD-RST: PL5 */
        backlight = <&backlight>;

        port {
            panel_out_dsi: endpoint {
                remote-endpoint = <&dsi_out_panel>;
            };
        };
    };
};

Case-2, bridge:

    panel {
        compatible = "bananapi,s070wv20-ct16", "simple-panel";
        enable-gpios = <&pio 1 7 GPIO_ACTIVE_HIGH>; /* LCD-PWR-EN: PB7 */
        backlight = <&backlight>;

        port {

            panel_out_bridge: endpoint {
                remote-endpoint = <&bridge_out_panel>;
            };
        };
    };

&dsi {
    vcc-dsi-supply = <&reg_dcdc1>;        /* VCC-DSI */
    status = "okay";

    ports {
        dsi_out: port at 1 {
            reg = <1>;

            dsi_out_bridge: endpoint {
                remote-endpoint = <&bridge_out_dsi>;
            };
        };
    };

    bridge at 0 {
        reg = <0>;
        compatible = "bananapi,icn6211", "chipone,icn6211";
        reset-gpios = <&r_pio 0 5 GPIO_ACTIVE_HIGH>; /* LCD-RST: PL5 */
        #address-cells = <1>;
        #size-cells = <0>;

        ports {
            #address-cells = <1>;
            #size-cells = <0>;

            bridge_in: port at 0 {
                reg = <0>;

                bridge_out_dsi: endpoint {
                    remote-endpoint = <&dsi_out_bridge>;
                };
            };

            bridge_out: port at 1 {
                reg = <1>;

                bridge_out_panel: endpoint {
                    remote-endpoint = <&panel_out_bridge>;
                };
            };
        };
    };
};

I think, I'm sure about the pipeline connections as per my
understanding. but something loosely missed here or in the code.
Please do let me know for any suggestions.

Jagan.


More information about the dri-devel mailing list