[PATCH 1/2] dt-bindings: adv7180: Introduce adv,adv7182-force-bt656-4

[PATCH 1/2] dt-bindings: adv7180: Introduce adv,adv7182-force-bt656-4

[Fabio Estevam]

Captured NTSC video would be out of sync when using the adv7280 with
the BT.656-4 protocol. Certain registers (0x04, 0x31, 0xE6) need to
be configured properly to ensure BT.656-4 compatibility.

Introduce the 'adv,adv7182-force-bt656-4' property to allow the ADV7280
behavior to be consistent with the ADV7180, where BT.656-4 timing
is used.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
---
 Documentation/devicetree/bindings/media/i2c/adv7180.yaml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/media/i2c/adv7180.yaml b/Documentation/devicetree/bindings/media/i2c/adv7180.yaml
index bcfd93739b4f..83e47b0c9fa2 100644
--- a/Documentation/devicetree/bindings/media/i2c/adv7180.yaml
+++ b/Documentation/devicetree/bindings/media/i2c/adv7180.yaml
@@ -35,6 +35,11 @@ properties:
   powerdown-gpios:
     maxItems: 1
 
+  adv,adv7182-force-bt656-4:
+    description:
+      Indicates that ADV7182 should output a BT.656-4 compatible stream.
+    type: boolean
+
   port:
     $ref: /schemas/graph.yaml#/properties/port
 
-- 
2.25.1

[PATCH 2/2] media: i2c: adv7180: fix adv7280 BT.656-4 compatibility

[Fabio Estevam]

From: Matthew Michilot <matthew.michilot@gmail.com>

Captured video would be out of sync when using the adv7280 with
the BT.656-4 protocol. Certain registers (0x04, 0x31, 0xE6) had to
be configured properly to ensure BT.656-4 compatibility.

An error in the adv7280 reference manual suggested that EAV/SAV mode
was enabled by default, however upon inspecting register 0x31, it was
determined to be disabled by default.

Signed-off-by: Matthew Michilot <matthew.michilot@gmail.com>
Reviewed-by: Tim Harvey <tharvey@gateworks.com>
[fabio: Introduce "adv,adv7182-force-bt656-4"]
Signed-off-by: Fabio Estevam <festevam@gmail.com>
---
Changes since v1:
- Only force BT656-4 when the DT property is present.
 drivers/media/i2c/adv7180.c | 30 ++++++++++++++++++++++++++----
 1 file changed, 26 insertions(+), 4 deletions(-)

diff --git a/drivers/media/i2c/adv7180.c b/drivers/media/i2c/adv7180.c
index 44bb6fe85644..a532df813d74 100644
--- a/drivers/media/i2c/adv7180.c
+++ b/drivers/media/i2c/adv7180.c
@@ -94,6 +94,7 @@
 #define ADV7180_REG_SHAP_FILTER_CTL_1	0x0017
 #define ADV7180_REG_CTRL_2		0x001d
 #define ADV7180_REG_VSYNC_FIELD_CTL_1	0x0031
+#define ADV7180_VSYNC_FIELD_CTL_1_NEWAV 0x12
 #define ADV7180_REG_MANUAL_WIN_CTL_1	0x003d
 #define ADV7180_REG_MANUAL_WIN_CTL_2	0x003e
 #define ADV7180_REG_MANUAL_WIN_CTL_3	0x003f
@@ -216,6 +217,7 @@ struct adv7180_state {
 	struct i2c_client	*vpp_client;
 	const struct adv7180_chip_info *chip_info;
 	enum v4l2_field		field;
+	bool			force_bt656_4;
 };
 #define to_adv7180_sd(_ctrl) (&container_of(_ctrl->handler,		\
 					    struct adv7180_state,	\
@@ -963,10 +965,26 @@ static int adv7182_init(struct adv7180_state *state)
 		adv7180_write(state, ADV7180_REG_EXTENDED_OUTPUT_CONTROL, 0x57);
 		adv7180_write(state, ADV7180_REG_CTRL_2, 0xc0);
 	} else {
-		if (state->chip_info->flags & ADV7180_FLAG_V2)
-			adv7180_write(state,
-				      ADV7180_REG_EXTENDED_OUTPUT_CONTROL,
-				      0x17);
+		if (state->chip_info->flags & ADV7180_FLAG_V2) {
+			if (state->force_bt656_4) {
+				/* ITU-R BT.656-4 compatible */
+				adv7180_write(state,
+					      ADV7180_REG_EXTENDED_OUTPUT_CONTROL,
+					      ADV7180_EXTENDED_OUTPUT_CONTROL_NTSCDIS);
+				/* Manually set NEWAVMODE */
+				adv7180_write(state,
+					      ADV7180_REG_VSYNC_FIELD_CTL_1,
+					      ADV7180_VSYNC_FIELD_CTL_1_NEWAV);
+				/* Manually set V bit end position in NTSC mode */
+				adv7180_write(state,
+					     ADV7180_REG_NTSC_V_BIT_END,
+					     ADV7180_NTSC_V_BIT_END_MANUAL_NVEND);
+			} else {
+				adv7180_write(state,
+					      ADV7180_REG_EXTENDED_OUTPUT_CONTROL,
+					      0x17);
+			}
+		}
 		else
 			adv7180_write(state,
 				      ADV7180_REG_EXTENDED_OUTPUT_CONTROL,
@@ -1314,6 +1332,7 @@ static int init_device(struct adv7180_state *state)
 static int adv7180_probe(struct i2c_client *client,
 			 const struct i2c_device_id *id)
 {
+	struct device_node *np = client->dev.of_node;
 	struct adv7180_state *state;
 	struct v4l2_subdev *sd;
 	int ret;
@@ -1338,6 +1357,9 @@ static int adv7180_probe(struct i2c_client *client,
 		return ret;
 	}
 
+	if (of_property_read_bool(np, "adv,adv7182-force-bt656-4"))
+		state->force_bt656_4 = true;
+
 	if (state->chip_info->flags & ADV7180_FLAG_MIPI_CSI2) {
 		state->csi_client = i2c_new_dummy_device(client->adapter,
 				ADV7180_DEFAULT_CSI_I2C_ADDR);
-- 
2.25.1

Re: [PATCH 1/2] dt-bindings: adv7180: Introduce adv,adv7182-force-bt656-4

[Rob Herring]

On Thu, May 27, 2021 at 02:32:08PM -0300, Fabio Estevam wrote:
> Captured NTSC video would be out of sync when using the adv7280 with
> the BT.656-4 protocol. Certain registers (0x04, 0x31, 0xE6) need to
> be configured properly to ensure BT.656-4 compatibility.
> 
> Introduce the 'adv,adv7182-force-bt656-4' property to allow the ADV7280
> behavior to be consistent with the ADV7180, where BT.656-4 timing
> is used.

I'm confused by the part numbers. 7182 or 7280? In any case, I'd just 
drop the part number from the property name.

> 
> Signed-off-by: Fabio Estevam <festevam@gmail.com>
> ---
>  Documentation/devicetree/bindings/media/i2c/adv7180.yaml | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/media/i2c/adv7180.yaml b/Documentation/devicetree/bindings/media/i2c/adv7180.yaml
> index bcfd93739b4f..83e47b0c9fa2 100644
> --- a/Documentation/devicetree/bindings/media/i2c/adv7180.yaml
> +++ b/Documentation/devicetree/bindings/media/i2c/adv7180.yaml
> @@ -35,6 +35,11 @@ properties:
>    powerdown-gpios:
>      maxItems: 1
>  
> +  adv,adv7182-force-bt656-4:
> +    description:
> +      Indicates that ADV7182 should output a BT.656-4 compatible stream.
> +    type: boolean
> +
>    port:
>      $ref: /schemas/graph.yaml#/properties/port
>  
> -- 
> 2.25.1

Re: [PATCH 1/2] dt-bindings: adv7180: Introduce adv,adv7182-force-bt656-4

[Fabio Estevam]

Hi Rob,

On Fri, Jun 4, 2021 at 5:54 PM Rob Herring <robh@kernel.org> wrote:

> I'm confused by the part numbers. 7182 or 7280? In any case, I'd just
> drop the part number from the property name.

Yes, dropping the part number from the property name is less confusing.

Will do as suggested in v2.

Thanks