linux-next: manual merge of the watchdog tree with the samsung-krzk tree

Posted by Stephen Rothwell 2 weeks, 2 days ago

Hi all,

Today's linux-next merge of the watchdog tree got a conflict in:

  arch/arm64/boot/dts/exynos/exynosautov920.dtsi

between commit:

  ef1c2a54cbc7 ("arm64: dts: exynosautov920: add peric1, misc and hsi0/1 clock DT nodes")

from the samsung-krzk tree and commit:

  3595a523d043 ("arm64: dts: exynosautov920: add watchdog DT node")

from the watchdog tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc arch/arm64/boot/dts/exynos/exynosautov920.dtsi
index c759134c909e,2b3e8debda3d..000000000000
--- a/arch/arm64/boot/dts/exynos/exynosautov920.dtsi
+++ b/arch/arm64/boot/dts/exynos/exynosautov920.dtsi
@@@ -172,17 -172,26 +172,37 @@@
  			reg = <0x10000000 0x24>;
  		};
  
 +		cmu_misc: clock-controller@10020000 {
 +			compatible = "samsung,exynosautov920-cmu-misc";
 +			reg = <0x10020000 0x8000>;
 +			#clock-cells = <1>;
 +
 +			clocks = <&xtcxo>,
 +				 <&cmu_top DOUT_CLKCMU_MISC_NOC>;
 +			clock-names = "oscclk",
 +				      "noc";
 +		};
 +
+ 		watchdog_cl0: watchdog@10060000 {
+ 			compatible = "samsung,exynosautov920-wdt";
+ 			reg = <0x10060000 0x100>;
+ 			interrupts = <GIC_SPI 953 IRQ_TYPE_LEVEL_HIGH>;
+ 			clocks = <&xtcxo>, <&xtcxo>;
+ 			clock-names = "watchdog", "watchdog_src";
+ 			samsung,syscon-phandle = <&pmu_system_controller>;
+ 			samsung,cluster-index = <0>;
+ 		};
+ 
+ 		watchdog_cl1: watchdog@10070000 {
+ 			compatible = "samsung,exynosautov920-wdt";
+ 			reg = <0x10070000 0x100>;
+ 			interrupts = <GIC_SPI 952 IRQ_TYPE_LEVEL_HIGH>;
+ 			clocks = <&xtcxo>, <&xtcxo>;
+ 			clock-names = "watchdog", "watchdog_src";
+ 			samsung,syscon-phandle = <&pmu_system_controller>;
+ 			samsung,cluster-index = <1>;
+ 		};
+ 
  		gic: interrupt-controller@10400000 {
  			compatible = "arm,gic-v3";
  			#interrupt-cells = <3>;

Re: linux-next: manual merge of the watchdog tree with the samsung-krzk tree

Posted by Krzysztof Kozlowski 2 weeks, 2 days ago

On 07/11/2024 06:59, Stephen Rothwell wrote:
> Hi all,
> 
> Today's linux-next merge of the watchdog tree got a conflict in:
> 
>   arch/arm64/boot/dts/exynos/exynosautov920.dtsi
> 
> between commit:
> 
>   ef1c2a54cbc7 ("arm64: dts: exynosautov920: add peric1, misc and hsi0/1 clock DT nodes")
> 
> from the samsung-krzk tree and commit:
> 
>   3595a523d043 ("arm64: dts: exynosautov920: add watchdog DT node")

The main problem is above patch should have never been taken to watchdog
tree. I never agreed on that. I never acked it. It is against SoC
policies which are always requesting entire DTS to go through SoC tree.

Please drop the patch from watchdog. Or revert it.

Best regards,
Krzysztof

Re: linux-next: manual merge of the watchdog tree with the samsung-krzk tree

Posted by Wim Van Sebroeck 2 weeks, 2 days ago

Hi Krzysztof,

> On 07/11/2024 06:59, Stephen Rothwell wrote:
> > Hi all,
> > 
> > Today's linux-next merge of the watchdog tree got a conflict in:
> > 
> >   arch/arm64/boot/dts/exynos/exynosautov920.dtsi
> > 
> > between commit:
> > 
> >   ef1c2a54cbc7 ("arm64: dts: exynosautov920: add peric1, misc and hsi0/1 clock DT nodes")
> > 
> > from the samsung-krzk tree and commit:
> > 
> >   3595a523d043 ("arm64: dts: exynosautov920: add watchdog DT node")
> 
> The main problem is above patch should have never been taken to watchdog
> tree. I never agreed on that. I never acked it. It is against SoC
> policies which are always requesting entire DTS to go through SoC tree.
> 
> Please drop the patch from watchdog. Or revert it.
> 
> Best regards,
> Krzysztof
> 

See my other e-mail. Since the 3 patches were about adding a new watchdog driver, I indeed took them in.
This was reverted and I can only presume that you will take the 3 patches and do the necessary via the SoC tree.

Kind regards,
Wim.

Re: linux-next: manual merge of the watchdog tree with the samsung-krzk tree

Posted by Guenter Roeck 2 weeks, 2 days ago

On 11/7/24 02:37, Wim Van Sebroeck wrote:
> Hi Krzysztof,
> 
>> On 07/11/2024 06:59, Stephen Rothwell wrote:
>>> Hi all,
>>>
>>> Today's linux-next merge of the watchdog tree got a conflict in:
>>>
>>>    arch/arm64/boot/dts/exynos/exynosautov920.dtsi
>>>
>>> between commit:
>>>
>>>    ef1c2a54cbc7 ("arm64: dts: exynosautov920: add peric1, misc and hsi0/1 clock DT nodes")
>>>
>>> from the samsung-krzk tree and commit:
>>>
>>>    3595a523d043 ("arm64: dts: exynosautov920: add watchdog DT node")
>>
>> The main problem is above patch should have never been taken to watchdog
>> tree. I never agreed on that. I never acked it. It is against SoC
>> policies which are always requesting entire DTS to go through SoC tree.
>>
>> Please drop the patch from watchdog. Or revert it.
>>
>> Best regards,
>> Krzysztof
>>
> 
> See my other e-mail. Since the 3 patches were about adding a new watchdog driver, I indeed took them in.
> This was reverted and I can only presume that you will take the 3 patches and do the necessary via the SoC tree.
> 

I think the idea was that the watchdog tree would take the driver and
its devicetree property description, and the SoC tree would take the
actual devicetree changes. At least that is what I do in hwmon.

Guenter

Re: linux-next: manual merge of the watchdog tree with the samsung-krzk tree

Posted by Wim Van Sebroeck 2 weeks, 2 days ago

Hi Guenter,

> On 11/7/24 02:37, Wim Van Sebroeck wrote:
> >Hi Krzysztof,
> >
> >>On 07/11/2024 06:59, Stephen Rothwell wrote:
> >>>Hi all,
> >>>
> >>>Today's linux-next merge of the watchdog tree got a conflict in:
> >>>
> >>>   arch/arm64/boot/dts/exynos/exynosautov920.dtsi
> >>>
> >>>between commit:
> >>>
> >>>   ef1c2a54cbc7 ("arm64: dts: exynosautov920: add peric1, misc and hsi0/1 clock DT nodes")
> >>>
> >>>from the samsung-krzk tree and commit:
> >>>
> >>>   3595a523d043 ("arm64: dts: exynosautov920: add watchdog DT node")
> >>
> >>The main problem is above patch should have never been taken to watchdog
> >>tree. I never agreed on that. I never acked it. It is against SoC
> >>policies which are always requesting entire DTS to go through SoC tree.
> >>
> >>Please drop the patch from watchdog. Or revert it.
> >>
> >>Best regards,
> >>Krzysztof
> >>
> >
> >See my other e-mail. Since the 3 patches were about adding a new watchdog driver, I indeed took them in.
> >This was reverted and I can only presume that you will take the 3 patches and do the necessary via the SoC tree.
> >
> 
> I think the idea was that the watchdog tree would take the driver and
> its devicetree property description, and the SoC tree would take the
> actual devicetree changes. At least that is what I do in hwmon.

That's how it is now.

Kind regards,
Wim.