linux-next: manual merge of the net-next tree with the spacemit tree

Mark Brown posted 1 patch 2 weeks ago
linux-next: manual merge of the net-next tree with the spacemit tree
Posted by Mark Brown 2 weeks ago
Hi all,

Today's linux-next merge of the net-next tree got conflicts in:

  arch/riscv/boot/dts/spacemit/k1-bananapi-f3.dts
  arch/riscv/boot/dts/spacemit/k1-milkv-jupiter.dts

between commit:

  0e28eab0ca512 ("riscv: dts: spacemit: Enable PDMA on Banana Pi F3 and Milkv Jupiter")

from the spacemit tree and commits:

  3c247a6366d58 ("riscv: dts: spacemit: Add Ethernet support for BPI-F3")
  e32dc7a936b11 ("riscv: dts: spacemit: Add Ethernet support for Jupiter")

from the net-next 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.

diff --cc arch/riscv/boot/dts/spacemit/k1-bananapi-f3.dts
index 6013be2585428,33e223cefd4bd..0000000000000
--- a/arch/riscv/boot/dts/spacemit/k1-bananapi-f3.dts
+++ b/arch/riscv/boot/dts/spacemit/k1-bananapi-f3.dts
@@@ -40,10 -42,52 +42,56 @@@
  	status = "okay";
  };
  
+ &eth0 {
+ 	phy-handle = <&rgmii0>;
+ 	phy-mode = "rgmii-id";
+ 	pinctrl-names = "default";
+ 	pinctrl-0 = <&gmac0_cfg>;
+ 	rx-internal-delay-ps = <0>;
+ 	tx-internal-delay-ps = <0>;
+ 	status = "okay";
+ 
+ 	mdio-bus {
+ 		#address-cells = <0x1>;
+ 		#size-cells = <0x0>;
+ 
+ 		reset-gpios = <&gpio K1_GPIO(110) GPIO_ACTIVE_LOW>;
+ 		reset-delay-us = <10000>;
+ 		reset-post-delay-us = <100000>;
+ 
+ 		rgmii0: phy@1 {
+ 			reg = <0x1>;
+ 		};
+ 	};
+ };
+ 
+ &eth1 {
+ 	phy-handle = <&rgmii1>;
+ 	phy-mode = "rgmii-id";
+ 	pinctrl-names = "default";
+ 	pinctrl-0 = <&gmac1_cfg>;
+ 	rx-internal-delay-ps = <0>;
+ 	tx-internal-delay-ps = <250>;
+ 	status = "okay";
+ 
+ 	mdio-bus {
+ 		#address-cells = <0x1>;
+ 		#size-cells = <0x0>;
+ 
+ 		reset-gpios = <&gpio K1_GPIO(115) GPIO_ACTIVE_LOW>;
+ 		reset-delay-us = <10000>;
+ 		reset-post-delay-us = <100000>;
+ 
+ 		rgmii1: phy@1 {
+ 			reg = <0x1>;
+ 		};
+ 	};
+ };
+ 
 +&pdma {
 +	status = "okay";
 +};
 +
  &uart0 {
  	pinctrl-names = "default";
  	pinctrl-0 = <&uart0_2_cfg>;
diff --cc arch/riscv/boot/dts/spacemit/k1-milkv-jupiter.dts
index c615fcadbd333,89f4132778931..0000000000000
--- a/arch/riscv/boot/dts/spacemit/k1-milkv-jupiter.dts
+++ b/arch/riscv/boot/dts/spacemit/k1-milkv-jupiter.dts
@@@ -20,10 -22,52 +22,56 @@@
  	};
  };
  
+ &eth0 {
+ 	phy-handle = <&rgmii0>;
+ 	phy-mode = "rgmii-id";
+ 	pinctrl-names = "default";
+ 	pinctrl-0 = <&gmac0_cfg>;
+ 	rx-internal-delay-ps = <0>;
+ 	tx-internal-delay-ps = <0>;
+ 	status = "okay";
+ 
+ 	mdio-bus {
+ 		#address-cells = <0x1>;
+ 		#size-cells = <0x0>;
+ 
+ 		reset-gpios = <&gpio K1_GPIO(110) GPIO_ACTIVE_LOW>;
+ 		reset-delay-us = <10000>;
+ 		reset-post-delay-us = <100000>;
+ 
+ 		rgmii0: phy@1 {
+ 			reg = <0x1>;
+ 		};
+ 	};
+ };
+ 
+ &eth1 {
+ 	phy-handle = <&rgmii1>;
+ 	phy-mode = "rgmii-id";
+ 	pinctrl-names = "default";
+ 	pinctrl-0 = <&gmac1_cfg>;
+ 	rx-internal-delay-ps = <0>;
+ 	tx-internal-delay-ps = <250>;
+ 	status = "okay";
+ 
+ 	mdio-bus {
+ 		#address-cells = <0x1>;
+ 		#size-cells = <0x0>;
+ 
+ 		reset-gpios = <&gpio K1_GPIO(115) GPIO_ACTIVE_LOW>;
+ 		reset-delay-us = <10000>;
+ 		reset-post-delay-us = <100000>;
+ 
+ 		rgmii1: phy@1 {
+ 			reg = <0x1>;
+ 		};
+ 	};
+ };
+ 
 +&pdma {
 +	status = "okay";
 +};
 +
  &uart0 {
  	pinctrl-names = "default";
  	pinctrl-0 = <&uart0_2_cfg>;
Re: linux-next: manual merge of the net-next tree with the spacemit tree
Posted by Vivian Wang 2 weeks ago
Hi all,

On 9/17/25 19:30, Mark Brown wrote:
> [...]
>
>   3c247a6366d58 ("riscv: dts: spacemit: Add Ethernet support for BPI-F3")
>   e32dc7a936b11 ("riscv: dts: spacemit: Add Ethernet support for Jupiter")
>
> from the net-next tree.

I originally submitted these net-next patches [1]. AFAICT, this is the
correct conflict resolution. Thank you.

Just FYI, Yixun has proposed for net-next to back out of the DTS changes
and taking them up through the spacemit tree instead [1], resolving the
conflicts in the spacemit tree. This would certainly mean less headaches
while managing pull requests, as well as allowing Yixun to take care of
code style concerns like node order. However, I do not know what the
norms here are.

Vivian "dramforever" Wang

[1]: https://lore.kernel.org/spacemit/20250914-net-k1-emac-v12-0-65b31b398f44@iscas.ac.cn/
[2]: https://lore.kernel.org/spacemit/20250916122026-GYB1255161@gentoo.org/
Re: linux-next: manual merge of the net-next tree with the spacemit tree
Posted by Mark Brown 2 weeks ago
On Wed, Sep 17, 2025 at 07:48:34PM +0800, Vivian Wang wrote:

> Just FYI, Yixun has proposed for net-next to back out of the DTS changes
> and taking them up through the spacemit tree instead [1], resolving the
> conflicts in the spacemit tree. This would certainly mean less headaches
> while managing pull requests, as well as allowing Yixun to take care of
> code style concerns like node order. However, I do not know what the
> norms here are.

Thanks.  They're pretty trivial conflicts so I'm not sure it's critical,
though like you say node order might easily end up the wrong way round
depending on how the conflict resolution gets done.
Re: linux-next: manual merge of the net-next tree with the spacemit tree
Posted by Yixun Lan 2 weeks ago
Hi Mark,

On 13:03 Wed 17 Sep     , Mark Brown wrote:
> On Wed, Sep 17, 2025 at 07:48:34PM +0800, Vivian Wang wrote:
> 
> > Just FYI, Yixun has proposed for net-next to back out of the DTS changes
> > and taking them up through the spacemit tree instead [1], resolving the
> > conflicts in the spacemit tree. This would certainly mean less headaches
> > while managing pull requests, as well as allowing Yixun to take care of
> > code style concerns like node order. However, I do not know what the
> > norms here are.
> 
> Thanks.  They're pretty trivial conflicts so I'm not sure it's critical,
> though like you say node order might easily end up the wrong way round
> depending on how the conflict resolution gets done.

Thanks for the help and fixing this, but ..

If it's possible to revert the DT patch 3-5, then I'd be happy to take,
but if this is too much job, e.g. the net-next's main branch is imuutable
and reverting it will cause too much trouble, then I'm fine with current
solution - carry the fix via net-next tree..

But please use commit: 0f084b221e2c5ba16eca85b3d2497f9486bd0329 of
https://github.com/spacemit-com/linux/tree/k1/dt-for-next as the merge
parent, which I'm about to send to Arnd (the SoC tree)

BTW, The 'for-next' branch is a merged branch contains clock and DT patches
for SpacemiT SoC tree's which isn't immutable..

Let me know what I should proceed, thank you

-- 
Yixun Lan (dlan)
Re: linux-next: manual merge of the net-next tree with the spacemit tree
Posted by Yixun Lan 2 weeks ago
Hi Mark,

On 20:30 Wed 17 Sep     , Yixun Lan wrote:
> Hi Mark,
> 
> On 13:03 Wed 17 Sep     , Mark Brown wrote:
> > On Wed, Sep 17, 2025 at 07:48:34PM +0800, Vivian Wang wrote:
> > 
> > > Just FYI, Yixun has proposed for net-next to back out of the DTS changes
> > > and taking them up through the spacemit tree instead [1], resolving the
> > > conflicts in the spacemit tree. This would certainly mean less headaches
> > > while managing pull requests, as well as allowing Yixun to take care of
> > > code style concerns like node order. However, I do not know what the
> > > norms here are.
> > 
> > Thanks.  They're pretty trivial conflicts so I'm not sure it's critical,
> > though like you say node order might easily end up the wrong way round
> > depending on how the conflict resolution gets done.
> 
> Thanks for the help and fixing this, but ..
> 
> If it's possible to revert the DT patch 3-5, then I'd be happy to take,
> but if this is too much job, e.g. the net-next's main branch is imuutable
> and reverting it will cause too much trouble, then I'm fine with current
> solution - carry the fix via net-next tree..
> 
> But please use commit: 0f084b221e2c5ba16eca85b3d2497f9486bd0329 of
> https://github.com/spacemit-com/linux/tree/k1/dt-for-next as the merge
> parent, which I'm about to send to Arnd (the SoC tree)
> 
No matter which way choose to go, I've created an immutable tag here,

https://github.com/spacemit-com/linux/ spacemit-dt-for-6.18-1

> BTW, The 'for-next' branch is a merged branch contains clock and DT patches
> for SpacemiT SoC tree's which isn't immutable..
> 
> Let me know what I should proceed, thank you
> 

-- 
Yixun Lan (dlan)
Re: linux-next: manual merge of the net-next tree with the spacemit tree
Posted by Yixun Lan 1 week, 6 days ago
Hi Paolo, Mark, Arnd

I'd like to have your attentions, see below

On 20:59 Wed 17 Sep     , Yixun Lan wrote:
> Hi Mark,
> 
> On 20:30 Wed 17 Sep     , Yixun Lan wrote:
> > Hi Mark,
> > 
> > On 13:03 Wed 17 Sep     , Mark Brown wrote:
> > > On Wed, Sep 17, 2025 at 07:48:34PM +0800, Vivian Wang wrote:
> > > 
> > > > Just FYI, Yixun has proposed for net-next to back out of the DTS changes
> > > > and taking them up through the spacemit tree instead [1], resolving the
> > > > conflicts in the spacemit tree. This would certainly mean less headaches
> > > > while managing pull requests, as well as allowing Yixun to take care of
> > > > code style concerns like node order. However, I do not know what the
> > > > norms here are.
> > > 
> > > Thanks.  They're pretty trivial conflicts so I'm not sure it's critical,
> > > though like you say node order might easily end up the wrong way round
> > > depending on how the conflict resolution gets done.
> > 
> > Thanks for the help and fixing this, but ..
> > 
> > If it's possible to revert the DT patch 3-5, then I'd be happy to take,
> > but if this is too much job, e.g. the net-next's main branch is imuutable
> > and reverting it will cause too much trouble, then I'm fine with current
> > solution - carry the fix via net-next tree..
> > 
> > But please use commit: 0f084b221e2c5ba16eca85b3d2497f9486bd0329 of
> > https://github.com/spacemit-com/linux/tree/k1/dt-for-next as the merge
> > parent, which I'm about to send to Arnd (the SoC tree)
> > 
> No matter which way choose to go, I've created an immutable tag here,
> 
> https://github.com/spacemit-com/linux/ spacemit-dt-for-6.18-1
> 

I've sent out the PR of DT changes to SoC tree for inclusion, see 
https://lore.kernel.org/all/20250919055525-GYC5766558@gentoo.org/

There is a potential conflict with commit from net-next:
 e32dc7a936b11e437298bcc4601476befcbcb88f ("riscv: dts: spacemit: Add Ethernet support for Jupiter")

the conflict itself is quite trivial, and should be easy to fix, and I'm also
personally fine to have it solved in net-next tree if Arnd has no objection

But if need assistance from my side, just let me know - I can handle it
- if the ethernet DT patches can be reverted from net-next
- I can apply them at SpacemiT SoC tree
- send a incremental v2 PR to the SoC tree

> > BTW, The 'for-next' branch is a merged branch contains clock and DT patches
> > for SpacemiT SoC tree's which isn't immutable..
> > 
> > Let me know what I should proceed, thank you
> > 
> 

-- 
Yixun Lan (dlan)