2023-04-28 12:17:40

by Valentin Caron

[permalink] [raw]
Subject: [PATCH 4/7] dt-bindings: spi: stm32: add bindings regarding stm32h7 spi slave

From: Alain Volmat <[email protected]>

Update the spi-stm32 binding yaml regarding to the SPI slave support.

Signed-off-by: Alain Volmat <[email protected]>
Signed-off-by: Valentin Caron <[email protected]>
---
Documentation/devicetree/bindings/spi/st,stm32-spi.yaml | 8 ++++++++
1 file changed, 8 insertions(+)

diff --git a/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml b/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
index c599eb359d56..1d26fa2658c5 100644
--- a/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
+++ b/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
@@ -27,6 +27,7 @@ allOf:
then:
properties:
st,spi-midi-ns: false
+ spi-slave: false

properties:
"#address-cells": true
@@ -62,6 +63,13 @@ properties:
- const: rx
- const: tx

+ cs-gpios:
+ description:
+ In case of spi-slave not defined, cs-gpios behave as defined in
+ spi-controller.yaml.
+ In case of spi-slave defined, if <0>, indicate that SS should be
+ detected via the dedicated HW pin
+
patternProperties:
"^[a-zA-Z][a-zA-Z0-9,+\\-._]{0,63}@[0-9a-f]+$":
type: object
--
2.25.1


2023-04-28 21:52:22

by Rob Herring (Arm)

[permalink] [raw]
Subject: Re: [PATCH 4/7] dt-bindings: spi: stm32: add bindings regarding stm32h7 spi slave

On Fri, Apr 28, 2023 at 02:15:21PM +0200, Valentin Caron wrote:
> From: Alain Volmat <[email protected]>
>
> Update the spi-stm32 binding yaml regarding to the SPI slave support.

Why? What problem are you trying to solve.

>
> Signed-off-by: Alain Volmat <[email protected]>
> Signed-off-by: Valentin Caron <[email protected]>
> ---
> Documentation/devicetree/bindings/spi/st,stm32-spi.yaml | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml b/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
> index c599eb359d56..1d26fa2658c5 100644
> --- a/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
> +++ b/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
> @@ -27,6 +27,7 @@ allOf:
> then:
> properties:
> st,spi-midi-ns: false
> + spi-slave: false
>
> properties:
> "#address-cells": true
> @@ -62,6 +63,13 @@ properties:
> - const: rx
> - const: tx
>
> + cs-gpios:
> + description:
> + In case of spi-slave not defined, cs-gpios behave as defined in
> + spi-controller.yaml.
> + In case of spi-slave defined, if <0>, indicate that SS should be
> + detected via the dedicated HW pin

I don't understand. I though cs-gpios was for master mode. You want to
define 'cs-gpios = <0>;'? How would that be different than just omitting
cs-gpios?

Rob

2023-05-04 17:19:12

by Valentin Caron

[permalink] [raw]
Subject: Re: [PATCH 4/7] dt-bindings: spi: stm32: add bindings regarding stm32h7 spi slave


On 4/28/23 23:46, Rob Herring wrote:
> On Fri, Apr 28, 2023 at 02:15:21PM +0200, Valentin Caron wrote:
>> From: Alain Volmat <[email protected]>
>>
>> Update the spi-stm32 binding yaml regarding to the SPI slave support.
> Why? What problem are you trying to solve.
To add some documentation about cs-gpios and don't allow to use
spi-slave property with st,stm32f4-spi compatible.

I will specify this in next patchset.

>
>> Signed-off-by: Alain Volmat <[email protected]>
>> Signed-off-by: Valentin Caron <[email protected]>
>> ---
>> Documentation/devicetree/bindings/spi/st,stm32-spi.yaml | 8 ++++++++
>> 1 file changed, 8 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml b/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
>> index c599eb359d56..1d26fa2658c5 100644
>> --- a/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
>> +++ b/Documentation/devicetree/bindings/spi/st,stm32-spi.yaml
>> @@ -27,6 +27,7 @@ allOf:
>> then:
>> properties:
>> st,spi-midi-ns: false
>> + spi-slave: false
>>
>> properties:
>> "#address-cells": true
>> @@ -62,6 +63,13 @@ properties:
>> - const: rx
>> - const: tx
>>
>> + cs-gpios:
>> + description:
>> + In case of spi-slave not defined, cs-gpios behave as defined in
>> + spi-controller.yaml.
>> + In case of spi-slave defined, if <0>, indicate that SS should be
>> + detected via the dedicated HW pin
> I don't understand. I though cs-gpios was for master mode. You want to
> define 'cs-gpios = <0>;'? How would that be different than just omitting
> cs-gpios?
>
> Rob
We try to keep consistency with spi master, but, as you said, it is useless.
I will remove it in next patchset.

Thanks,
Valentin