Re: [PATCH v2 3/7] drm/panel: simple: Add waiken WKS101WX001-WCT support
Liu Ying <[email protected]>
| Newsgroups | dev.linux.lists.imx,org.freedesktop.lists.dri-devel,org.infradead.lists.linux-arm-kernel,org.kernel.vger.linux-devicetree,org.kernel.vger.linux-kernel |
|---|---|
| Message-ID | <amLibmRsBY8lgj2K@raspi> |
On Thu, Jul 23, 2026 at 10:16:18PM -0500, Frank Li wrote: > On Fri, Jul 24, 2026 at 10:27:13AM +0800, Liu Ying wrote: > > On Thu, Jul 23, 2026 at 04:36:39PM -0500, Frank Li wrote: > > > On Thu, Jul 23, 2026 at 10:22:10AM +0800, Liu Ying wrote: > > > > On Wed, Jul 22, 2026 at 05:12:47PM -0500, Frank Li wrote: > > > > > On Wed, Jul 22, 2026 at 11:57:09AM +0800, Liu Ying wrote: > > > > > > On Wed, Jul 22, 2026 at 11:44:09AM +0800, Liu Ying wrote: > > > > > > > On Tue, Jul 21, 2026 at 07:05:03PM -0500, Frank Li wrote: > > > > > > > > On Fri, Jul 17, 2026 at 01:48:31PM +0800, Liu Ying wrote: > > > > > > > > > On Wed, Jul 15, 2026 at 02:34:48PM -0400, [email protected] wrote: > > > > > > > > > > From: Frank Li <[email protected]> > > > > > > > > > > > > > > > > > > > > Add support for WKS101WX001-WCT panel, which is an industrial-grade > > > > > > > > > > 10.1-inch IPS TFT LCD capacitive touch screen display module. > > > > > > > > > > > > > > > > > > > > Signed-off-by: Frank Li <[email protected]> > > > > > > > > > > --- > > > > > > > > > > change in v2 > > > > > > > > > > - update height to 135 from 136 (sashiko) > > > > > > > > > > --- > > > > > > > > > > drivers/gpu/drm/panel/panel-simple.c | 31 +++++++++++++++++++++++++++++++ > > > > > > > > > > 1 file changed, 31 insertions(+) > > > > > > > > > > > > > > [...] > > > > > > > > > > > > > > > > > + .connector_type = DRM_MODE_CONNECTOR_DPI, > > > > > > > > > > > > > > > > > > Seems that the panel also supports LVDS input according to the specification, > > > > > > > > > since it mentions 'RGB or LVDS Interface'? > > > > > > > > > > > > > > > > Need hardware rework to use LVDS. Need remove banch of resistors. Should it > > > > > > > > be treat as difference panel by different compatible string? hardware > > > > > > > > actually changed. > > > > > > > > > > > > > > Not sure how to handle that. > > > > > > > > > > > > Wait, you probably may drop this patch, just document the compatible > > > > > > string in panel-simple.yaml and panel-lvds.yaml and use add a fallback > > > > > > panel-simple.yaml don't allow fallback to panel-dpi. Need seperated yaml > > > for it. > > > > Really? How does it disallow that? > > > > I can see newhaven,nhd-4.3-480272ef-atxl and osddisplays,osd070t1718-19ts > > are documented in panel-simple.yaml and used with the "panel-dpi" fallback > > compatible string. > > > > Documentation/devicetree/bindings/display/panel/panel-simple.yaml:244: - newhaven,nhd-4.3-480272ef-atxl > > arch/arm/boot/dts/ti/omap/am437x-sk-evm.dts:127: compatible = "newhaven,nhd-4.3-480272ef-atxl", "panel-dpi"; > > > > Documentation/devicetree/bindings/display/panel/panel-simple.yaml:268: - osddisplays,osd070t1718-19ts > > arch/arm/boot/dts/ti/omap/am437x-gp-evm.dts:89: compatible = "osddisplays,osd070t1718-19ts", "panel-dpi"; > > arch/arm/boot/dts/ti/omap/am43x-epos-evm.dts:45: compatible = "osddisplays,osd070t1718-19ts", "panel-dpi"; > > arch/arm/boot/dts/ti/omap/am57xx-evm.dtso:62: compatible = "osddisplays,osd070t1718-19ts", "panel-dpi"; > > It can't pass DTB_CHECK. it may add at earily time. panel-simple.yaml > have not fallback compatible string. I tried it. DTB_CHECK will complain > too long compatible string if add "panel-dpi". Yeah, I missed testing panel-simple.yaml with DTB_CHECK. So, just don't document the panel compatible string in panel-simple.yaml and only document it in panel-lvds.yaml(not today, but until someone really uses it). Then, I'd assume that the "fallback compatible" solution still works. > > > > > > > > > > > > compatible string "panel-dpi" or "panel-lvds" in device tree plus a > > > > > > panel-timing node. > > > > > > previous dt-binding discuss, don't allow failback to two compatible string > > > sorry, I can't remember exactly thread. > > > > > > like > > > items: > > > - const: ws101w001-wct > > > - enum: > > > - panel-dpi > > > - panel-lvds > > > > I meant just to document the panel in panel-simple.yaml and panel-lvds.yaml. > > No need to add a dedicated DT binding file with such a compatible property > > for this panel. > > panel-lvds.yaml allow fallback to panel-lvds, but panel-simple.yaml not. > all binding which allow fallback to panel-dpi, there are seperate yaml file > for it. > > of course TI or old platform, especially arm32 platform wrongly use it > regardless DTB check warning. But for new platform, we can't do that as it. > > Frank > > > > > > > > > > > > > > > Look like recently prefer add panel-timing informaiton into c code instead > > > > > of put it into dts file. But I am not sure about it. Just check recent's > > > > > log history. > > > > > > > > If display timing is specified in panel-simple.c, then you have to specify > > > > the connector_type there too, right? The problem here is that you don't > > > > know the connector_type, as it could be either LVDS or DPI, and I'd assume > > > > that we don't want two compatible strings - one for LVDS and the other for > > > > DPI. > > > > > > If supported both type, generally need customer property and new panel > > > driver to switch between dpi / lvds > > > > > > or use > > > ports { > > > > > > port@0: //connect dpi > > > port@1: //connect lvds > > > }; > > > > I'd say a new panel driver for this panel is a bit overkill. > > > > > > > > > > > I think we can defer this lvds work until it is actually used because > > > - need a big hardware rework to support lvds, most likely it will not be > > > > Per panel spec, just remove a few resistors, which doesn't sound like a > > big rework to me. But, perhaps, this doesn't impact SW solution a lot though. > > > > > appeared at real productions. If it mass production, vendor most likely use > > > difference module number for it. > > > - it may be end of life if there are not much users. > > > > I'd prefer a mature solution which handles both DPI and LVDS interfaces > > correctly in the first place. > > > > Anyway, just my two cents here. "DRM PANEL DRIVERS" maintainers or DRM > > subsystem maintainers may override me. > > > > > > > > Frank > > > > > > > Although display timing in panel-simple.c avoids duplicated panel-timing > > > > nodes across various potential DT files, I think you have the "good" reason > > > > here to use the panel-timing node + fallback compatible string with one single > > > > panel specific compatible string. > > > > > > > > > > > > > > Add Neil Armstrong for comments. > > > > > > > > > > Frank > > > > > > > > > > > > > > > > > -- > > > > > > Regards, > > > > > > Liu Ying > > > > > > > > -- > > > > Regards, > > > > Liu Ying > > > > -- > > Regards, > > Liu Ying -- Regards, Liu Ying