tcan4x5x requires an external clock called cclk, add it here.

Signed-off-by: Sean Nyekjaer <sean@geanix.com>
---
Changes in v2:
- added clock-names as a required property
- Link to v1: https://lore.kernel.org/r/20241127-tcancclk-v1-1-5493d3f03db1@geanix.com
---
 Documentation/devicetree/bindings/net/can/ti,tcan4x5x.yaml | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/can/ti,tcan4x5x.yaml b/Documentation/devicetree/bindings/net/can/ti,tcan4x5x.yaml
index ff18cf7393550d1b7107b1233d8302203026579d..384e15da27136c0bad39c4d9f6cc0456fb0d5b19 100644
--- a/Documentation/devicetree/bindings/net/can/ti,tcan4x5x.yaml
+++ b/Documentation/devicetree/bindings/net/can/ti,tcan4x5x.yaml
@@ -29,6 +29,10 @@ properties:
   clocks:
     maxItems: 1
 
+  clock-names:
+    items:
+      - const: cclk
+
   reset-gpios:
     description: Hardwired output GPIO. If not defined then software reset.
     maxItems: 1
@@ -138,6 +142,7 @@ required:
   - reg
   - interrupts
   - clocks
+  - clock-names
   - bosch,mram-cfg
 
 unevaluatedProperties: false
@@ -155,6 +160,7 @@ examples:
             compatible = "ti,tcan4x5x";
             reg = <0>;
             clocks = <&can0_osc>;
+            clock-names = "cclk";
             pinctrl-names = "default";
             pinctrl-0 = <&can0_pins>;
             spi-max-frequency = <10000000>;
@@ -180,6 +186,7 @@ examples:
             compatible = "ti,tcan4552", "ti,tcan4x5x";
             reg = <0>;
             clocks = <&can0_osc>;
+            clock-names = "cclk";
             pinctrl-names = "default";
             pinctrl-0 = <&can0_pins>;
             spi-max-frequency = <10000000>;

---
base-commit: e0b741bc53c94f9ae25d4140202557a0aa51b5a0
change-id: 20241127-tcancclk-c149c0b3b050

Best regards,
-- 
Sean Nyekjaer <sean@geanix.com>

On 28/11/2024 08:09, Sean Nyekjaer wrote:
> tcan4x5x requires an external clock called cclk, add it here.

Nothing improved. Device already has this clock, so your rationale is
not correct.

Also, offending patch is not in the next for some reason, so it should
be squashed there. Conversion which leads to incorrect binding is not a
correct conversion.

Best regards,
Krzysztof

Hi Krzysztof,

On Thu, Nov 28, 2024 at 08:31:41AM +0100, Krzysztof Kozlowski wrote:
> On 28/11/2024 08:09, Sean Nyekjaer wrote:
> > tcan4x5x requires an external clock called cclk, add it here.
> 
> Nothing improved. Device already has this clock, so your rationale is
> not correct.

Please explain, yes the device has this clock, but not the name. The
driver depends on the name.

> 
> Also, offending patch is not in the next for some reason, so it should
> be squashed there. Conversion which leads to incorrect binding is not a
> correct conversion.

"offending patch":
https://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next.git/commit/?h=testing&id=77400284f54b9a1f6b6127c08cb935fc05e5c3d2

"dt-bindings: can: convert tcan4x5x.txt to DT schema" did the
conversion no more or less.
The original txt file fails to mention the clock required,
therefore IMHO this patch should be as a seperate patch :)

/Sean

On 28/11/2024 08:49, Sean Nyekjaer wrote:
> Hi Krzysztof,
> 
> On Thu, Nov 28, 2024 at 08:31:41AM +0100, Krzysztof Kozlowski wrote:
>> On 28/11/2024 08:09, Sean Nyekjaer wrote:
>>> tcan4x5x requires an external clock called cclk, add it here.
>>
>> Nothing improved. Device already has this clock, so your rationale is
>> not correct.
> 
> Please explain, yes the device has this clock, but not the name. The
> driver depends on the name.

You say that the device - tcan4x5x - requires an external clock. This is
already satisfied.

> 
>>
>> Also, offending patch is not in the next for some reason, so it should
>> be squashed there. Conversion which leads to incorrect binding is not a
>> correct conversion.
> 
> "offending patch":
> https://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next.git/commit/?h=testing&id=77400284f54b9a1f6b6127c08cb935fc05e5c3d2
> 
> "dt-bindings: can: convert tcan4x5x.txt to DT schema" did the
> conversion no more or less.
> The original txt file fails to mention the clock required,
> therefore IMHO this patch should be as a seperate patch :)

Stop repeating the same as last time.

No, I said it already multiple times in multiple threads, including
guideline I gave some time ago:
https://social.kernel.org/notice/Ai9hYRUKo8suzX3zNY

You conversion is supposed to be complete and pass checks on existing
DTS, which also implies matching driver.


Best regards,
Krzysztof