Message ID | 20230930145951.23433-1-jacopo.mondi@ideasonboard.com |
---|---|
Headers | show |
Series | media: bindings: Fix handling of video-interface-device | expand |
On Sat, Sep 30, 2023 at 04:59:46PM +0200, Jacopo Mondi wrote: > Only properties explicitly listed in the schema are accepted as > endpoint properties. > > Make sure this is actually enforced by setting 'additionalProperties' > to false and explicitly allow 'remote-endpoint' in the list of > endpoint properties. > > Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com> > --- > Documentation/devicetree/bindings/media/i2c/hynix,hi846.yaml | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/Documentation/devicetree/bindings/media/i2c/hynix,hi846.yaml b/Documentation/devicetree/bindings/media/i2c/hynix,hi846.yaml > index 60f19e1152b3..f2ca86501d3c 100644 > --- a/Documentation/devicetree/bindings/media/i2c/hynix,hi846.yaml > +++ b/Documentation/devicetree/bindings/media/i2c/hynix,hi846.yaml > @@ -58,7 +58,7 @@ properties: > properties: > endpoint: > $ref: /schemas/media/video-interfaces.yaml# > - unevaluatedProperties: false > + additionalProperties: false If anything 'additionalProperties' is the exception and 'unevaluatedProperties' is the rule. Just grep refs to video-interfaces.yaml. Why did you change this for just this 1 device? I'm more worried about undocumented properties than a documented property which makes no sense for the h/w being present. So I think as-is was fine. > > properties: > data-lanes: > @@ -73,6 +73,7 @@ properties: > - const: 2 > > link-frequencies: true I suppose we could remove this as it has no effect on the schema validation, but it's probably worthwhile to keep for documentation purposes. Rob > + remote-endpoint: true > > required: > - data-lanes > -- > 2.42.0 >
On Sat, 30 Sep 2023 16:59:47 +0200, Jacopo Mondi wrote: > Fix handling of properties from video-interface-device.yaml for > Omnivision OV02A10 sensor. > > There is no reason to restrict the allowed rotation degrees to 0 and 180, > as the sensor can be mounted with any rotation. > > Also, as all the properties described by video-interface-device.yaml are > allowed for the image sensor, make them accepted by changing > "additionalProperties: false" to "unevaluatedProperties: false" at the > schema top-level. > > Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com> > --- > .../devicetree/bindings/media/i2c/ovti,ov02a10.yaml | 8 +------- > 1 file changed, 1 insertion(+), 7 deletions(-) > Reviewed-by: Rob Herring <robh@kernel.org>
Hi Jacopo, On Sat, Sep 30, 2023 at 04:59:45PM +0200, Jacopo Mondi wrote: > Allow properties from video-interface-device.yaml for the SK Hynix Hi-846 > sensor. > > All properties specified in video-interface-device.yaml schema are > valid, so make them accepted by changing "additionalProperties: false" > to "unevaluatedProperties: false" at the schema top-level. The patch seems fine to me, but I wonder if we should change the title of video-interface-devices.yaml (it's plural) to something that refers to camera sensors, and possibly split it. It's currently not relevant for other types of devices.
On Tue, Oct 03, 2023 at 08:27:30AM +0000, Sakari Ailus wrote: > Hi Jacopo, > > On Sat, Sep 30, 2023 at 04:59:45PM +0200, Jacopo Mondi wrote: > > Allow properties from video-interface-device.yaml for the SK Hynix Hi-846 > > sensor. > > > > All properties specified in video-interface-device.yaml schema are > > valid, so make them accepted by changing "additionalProperties: false" > > to "unevaluatedProperties: false" at the schema top-level. > > The patch seems fine to me, but I wonder if we should change the title of > video-interface-devices.yaml (it's plural) to something that refers to > camera sensors, and possibly split it. It's currently not relevant for > other types of devices. I was thinking exactly the same yesterday.