Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <CAO9szn18KsR0c+U8EEY1=xnmsKMxy6SEArMUic0z=aYJDVwWCQ@mail.gmail.com>
 <023f6cf9-0f08-f27e-d203-5ff78faf110f@linaro.org> <CAO9szn1EsbuPSRrOW8CLqhp+QUcL=9NE93FAwsg2n3htd_aJTw@mail.gmail.com>
 <CAMty3ZCAP6CRsJWMUZZ6+hd5igX3NgyNfhdEv2FwuDtqj4iaaA@mail.gmail.com> <HB0TUR.1IPEEV2C5LMB1@gmail.com>
In-Reply-To: <HB0TUR.1IPEEV2C5LMB1@gmail.com>
From:   Jagan Teki <jagan@amarulasolutions.com>
Date:   Wed, 17 May 2023 22:02:47 +0530
Message-ID: <CAMty3ZCSUx-81m4SQcJZvUq3NyhzZbe8ow+LiV7iyUmVLCmNYQ@mail.gmail.com>
Subject: Re: [PATCH v2] dt-bindings: display: panel: add panel-mipi-dsi-bringup
To:     Paulo <pavacic.p@gmail.com>
Cc:     Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
        neil.armstrong@linaro.org, conor+dt@kernel.org,
        devicetree@vger.kernel.org, sam@ravnborg.org,
        linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
        robh+dt@kernel.org, krzysztof.kozlowski+dt@linaro.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi,

Please don't post, use inline replies.

On Wed, May 17, 2023 at 6:34=E2=80=AFPM Paulo <pavacic.p@gmail.com> wrote:
>
> On Wed, May 17 2023 at 05:50:22 PM +0530, Jagan Teki
> <jagan@amarulasolutions.com> wrote:
>  > Just to add a few pieces of information for you to understand better
>  > on the context of dsi panels. DSI panels can be part of
> panel-simple.c
>  > or panel-<vendor-part>.c DSI panels whose init and exit sequence is
>  > generic are suitable to add it in panel-simple and have bindings on
>  > panel-simple.yml.
>
> This panel doesn't fit that well into panel-simple.c since it has
> initialization sequence. For that reason it would fit more into
> panel-sortofsimple.c which didn't exist so I have created new driver
> and called it panel-mipi-dsi-bringup.c.
>
>  > Some DSI panels have specific init and exit
>  > sequences in terms of power, reset and DCS then those have separate
>  > drivers to handle and whose driver name must be panel-<vendor-part>.c
>  > or similar and bindings also follow a similar naming convention.
>
> I have made a driver exactly for that purpose. Driver that allows
> adding new panels which have specific init sequences (and of course
> timings and other stuff). fannal,c3004 can be seen as a working example.
>
> Here is code snippet from the driver:
> ```
> static const struct brup_panel_info brup_fannal_c3004_panel_info =3D {
>  .display_mode =3D &brup_fannal_c3004_display_mode,
>  .num_of_dsi_lanes =3D 2, //how many wires are connected to the panel
>  .video_mode =3D BRUP_VIDEO_MODES[BRUP_SYNC_PULSE],
>  .mipi_dsi_format =3D MIPI_DSI_FMT_RGB888,
>  .mipi_dsi_mode_flags =3D
>   MIPI_DSI_CLOCK_NON_CONTINUOUS | MIPI_DSI_MODE_VSYNC_FLUSH |
>   MIPI_DSI_MODE_VIDEO_HSE | MIPI_DSI_MODE_NO_EOT_PACKET,
>  .bus_flags =3D DRM_BUS_FLAG_DE_LOW | DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE=
,
>  .panel_enable_function =3D &brup_panel_fannal_c3004_enable_function
> };
> ```
> where enable function is function with init sequence and other values
> are values that might be different for different displays.
>
> All the inputs are appreciated as this is my first time submitting
> patch. If you see anything that is odd to you please reach out to me.
> All in all I believe I now understand how should device tree look and
> the reasons/ideology behind it.

So, the driver has to be panel-fannal-c3004.c and binding to be
fannal,c3004.yaml.

Thanks,
Jagan.