Hi,
On Wed, 2018-05-09 at 09:12 +0200, Maxime Ripard wrote:
> On Tue, May 08, 2018 at 12:04:11AM +0200, Paul Kocialkowski wrote:
> > This adds timings for the RGB666 variant of the Innolux AT070TN90 panel,
> > as found on the Ainol AW1 tablet.
> >
> > The panel also supports RGB888 output. When RGB666 mode is used instead,
> > the two extra lanes per component are grounded.
> >
> > In the future, it might become necessary to introduce a dedicated
> > device-tree property to specify the bus format to use instead of the
> > default one for the panel. This will allow supporting different bus
> > formats for the same panel modes.
> >
> > Signed-off-by: Paul Kocialkowski <contact@xxxxxxxx>
> > ---
> > drivers/gpu/drm/panel/panel-simple.c | 26 ++++++++++++++++++++++++++
> > 1 file changed, 26 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
> > index cbf1ab404ee7..32e30d5a8f08 100644
> > --- a/drivers/gpu/drm/panel/panel-simple.c
> > +++ b/drivers/gpu/drm/panel/panel-simple.c
> > @@ -1063,6 +1063,29 @@ static const struct panel_desc innolux_at043tn24 = {
> > .bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_POSEDGE,
> > };
> >
> > +static const struct drm_display_mode innolux_at070tn90_mode = {
> > + .clock = 40000,
> > + .hdisplay = 800,
> > + .hsync_start = 800 + 112,
> > + .hsync_end = 800 + 112 + 1,
> > + .htotal = 800 + 112 + 1 + 87,
> > + .vdisplay = 480,
> > + .vsync_start = 480 + 141,
> > + .vsync_end = 480 + 141 + 1,
> > + .vtotal = 480 + 141 + 1 + 38,
> > + .vrefresh = 60,
> > +};
> > +
> > +static const struct panel_desc innolux_at070tn90 = {
> > + .modes = &innolux_at070tn90_mode,
> > + .num_modes = 1,
> > + .size = {
> > + .width = 154,
> > + .height = 86,
> > + },
> > + .bus_format = MEDIA_BUS_FMT_RGB666_1X18,
> > +};
> > +
>
> I'm not really convinced this is the right approach. You said it
> yourself, the panel is using a 24-bits interface, and you just happen
> to have a tablet that routed it using a 18-bits interface instead.
>
> That doesn't belong in the driver, especially associated to the
> compatible, but where the routing is described: in the device
> tree. And given that the panel interface is a 24 bits panel, if we
> were to have a default, we should have this one, and not the one
> fitting your use case.
I fully agree, this is why I suggested introducing a dedicated dt
property for selecting the bus format (in the commit message). I still
proposed this patch as a temporary solution, but I'm definitely willing
to craft a proper solution as well.
Here is an initial proposition:
1. Making bus_format an array in struct panel_desc and listing all the
relevant bus formats that the panel can support there;
2. Introducing an optional "bus-format" dt property that indicates which
bus format to use, and using the first index of the bus formats array if
the property is not present;
3. Checking that the bus format requested from dt is supported by the
panel.
What do you think?
Paul
--
Developer of free digital technology and hardware support.
Website: https://www.paulk.fr/
Coding blog: https://code.paulk.fr/
Git repositories: https://git.paulk.fr/ https://git.code.paulk.fr/
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
