Remove unused compatibles from i.MXRT1050.dtsi.
Change GPT clock-names to match documentation.

Cc: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Jesse Taube <Mr.Bossman075@gmail.com>
---
 arch/arm/boot/dts/imxrt1050.dtsi | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/arch/arm/boot/dts/imxrt1050.dtsi b/arch/arm/boot/dts/imxrt1050.dtsi
index 03e6a858a7be..8d79de239046 100644
--- a/arch/arm/boot/dts/imxrt1050.dtsi
+++ b/arch/arm/boot/dts/imxrt1050.dtsi
@@ -29,7 +29,7 @@ osc3M: osc3M {
 
 	soc {
 		lpuart1: serial@40184000 {
-			compatible = "fsl,imxrt1050-lpuart", "fsl,imx7ulp-lpuart";
+			compatible = "fsl,imxrt1050-lpuart";
 			reg = <0x40184000 0x4000>;
 			interrupts = <20>;
 			clocks = <&clks IMXRT1050_CLK_LPUART1>;
@@ -40,7 +40,6 @@ lpuart1: serial@40184000 {
 		iomuxc: pinctrl@401f8000 {
 			compatible = "fsl,imxrt1050-iomuxc";
 			reg = <0x401f8000 0x4000>;
-			fsl,mux_mask = <0x7>;
 		};
 
 		anatop: anatop@400d8000 {
@@ -83,7 +82,7 @@ edma1: dma-controller@400e8000 {
 		};
 
 		usdhc1: mmc@402c0000 {
-			compatible = "fsl,imxrt1050-usdhc", "fsl,imx6sl-usdhc";
+			compatible = "fsl,imxrt1050-usdhc";
 			reg = <0x402c0000 0x4000>;
 			interrupts = <110>;
 			clocks = <&clks IMXRT1050_CLK_IPG_PDOF>,
@@ -150,11 +149,11 @@ gpio5: gpio@400c0000 {
 		};
 
 		gpt: timer@401ec000 {
-			compatible = "fsl,imxrt1050-gpt", "fsl,imx6dl-gpt", "fsl,imx6sl-gpt";
+			compatible = "fsl,imxrt1050-gpt", "fsl,imx6dl-gpt";
 			reg = <0x401ec000 0x4000>;
 			interrupts = <100>;
-			clocks = <&osc3M>;
-			clock-names = "per";
+			clocks = <&osc3M>, <&osc3M>;
+			clock-names = "ipg", "per";
 		};
 	};
 };
-- 
2.37.2

On Mon, Nov 7, 2022 at 8:15 AM Jesse Taube <mr.bossman075@gmail.com> wrote:

> Remove unused compatibles from i.MXRT1050.dtsi.
> Change GPT clock-names to match documentation.
>
> Cc: Giulio Benetti <giulio.benetti@benettiengineering.com>
> Signed-off-by: Jesse Taube <Mr.Bossman075@gmail.com>

I applied patches 1-6 to the pinctrl tree, this one seems independent
and possibly should not even be applied as I understand it.

Yours,
Linus Walleij

On Mon, Nov 7, 2022, at 08:15, Jesse Taube wrote:
> Remove unused compatibles from i.MXRT1050.dtsi.
> Change GPT clock-names to match documentation.
>
> Cc: Giulio Benetti <giulio.benetti@benettiengineering.com>
> Signed-off-by: Jesse Taube <Mr.Bossman075@gmail.com>

Can you make sure your changelog texts explain why you do this?
Are they fundamentally different from the devices you had
claimed to be compatible with that need a different driver,
or are there drivers in the field that bind to the wrong
string first?

       Arnd

On 11/7/22 02:44, Arnd Bergmann wrote:
> On Mon, Nov 7, 2022, at 08:15, Jesse Taube wrote:
>> Remove unused compatibles from i.MXRT1050.dtsi.
>> Change GPT clock-names to match documentation.
>>
>> Cc: Giulio Benetti <giulio.benetti@benettiengineering.com>
>> Signed-off-by: Jesse Taube <Mr.Bossman075@gmail.com>
> 
> Can you make sure your changelog texts explain why you do this?
Yes, sorry I wasn't clear.
> Are they fundamentally different from the devices you had
> claimed to be compatible with that need a different driver,
UART and SDHC had drivers added which are better fit.
The GPT binds to imx6dl which is also the same as imx6sl.
> or are there drivers in the field that bind to the wrong
> string first?
I don't understand?

Thanks,
Jesse Taube
> 
>        Arnd

On Mon, Nov 7, 2022, at 16:09, Jesse Taube wrote:
> On 11/7/22 02:44, Arnd Bergmann wrote:
>> On Mon, Nov 7, 2022, at 08:15, Jesse Taube wrote:
>>> Remove unused compatibles from i.MXRT1050.dtsi.
>>> Change GPT clock-names to match documentation.
>>>
>>> Cc: Giulio Benetti <giulio.benetti@benettiengineering.com>
>>> Signed-off-by: Jesse Taube <Mr.Bossman075@gmail.com>
>> 
>> Can you make sure your changelog texts explain why you do this?
> Yes, sorry I wasn't clear.
>
>> Are they fundamentally different from the devices you had
>> claimed to be compatible with that need a different driver,
>
> UART and SDHC had drivers added which are better fit.
> The GPT binds to imx6dl which is also the same as imx6sl.

Where are those drivers added? Looking at linux-6.1-rc2
and linux-next, I still see them use the same drivers as
the original ones, and listing both strings would be the
preferred method.

>> or are there drivers in the field that bind to the wrong
>> string first?
> I don't understand?

I mean if you had run into the case where you have
a driver that misbehaves when the fallback string is
present in addition to the most specific one.

     Arnd