On Thu, Oct 18, 2018 at 10:08 AM A.s. Dong [off-list ref] wrote:

quoted

-----Original Message-----
From: Rob Herring [mailto:robh at kernel.org]
Sent: Thursday, October 18, 2018 10:34 PM
To: A.s. Dong <aisheng.dong@nxp.com>

[...]

quoted

On Wed, Oct 17, 2018 at 07:24:40AM +0000, A.s. Dong wrote:

quoted

As the power domain API might change in the future for new SoCs,
although in a very low possibility, it's still better to make the
compatible string more SoC specific to avoid the possible version change for

new SoCs.

quoted

Due to there're still no users in kernel, it's safe to update it
without breaking anything.

Cc: Shawn Guo <shawnguo@kernel.org>
Cc: Sascha Hauer <kernel@pengutronix.de>
Cc: Fabio Estevam <redacted>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: devicetree at vger.kernel.org
Suggested-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Dong Aisheng <aisheng.dong@nxp.com>

checkpatch says:

WARNING: Missing Signed-off-by: line by nominal patch author 'A.s. Dong
[off-list ref]'

Do you think if we can ignore this warning?

Probably, but better if you can fix it.

It seems checkpatch checks the email sender name as nominal patch author
which is abbreviated as 'A.S. Dong'. But I always use the full name in
upstreaming patches.

The patch contents should start with a 'From:' line if your sending
email doesn't match the patch author (such as when you sent another
author's patch). Maybe it did match and your email server is
overwriting it. I would imagine that can be fixed.

quoted

quoted

---
ChangeLog:
v5->v6:
 * new patch
---
 Documentation/devicetree/bindings/arm/freescale/fsl,scu.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git
a/Documentation/devicetree/bindings/arm/freescale/fsl,scu.txt
b/Documentation/devicetree/bindings/arm/freescale/fsl,scu.txt
index 46d0af1..87fc4b4 100644

--- a/Documentation/devicetree/bindings/arm/freescale/fsl,scu.txt
+++ b/Documentation/devicetree/bindings/arm/freescale/fsl,scu.txt

@@ -58,7 +58,7 @@ This binding for the SCU power domain providers uses

the generic power  domain binding[2].

 Required properties:
-- compatible:              Should be "fsl,scu-pd".
+- compatible:              Should be "fsl,imx8qxp-scu-pd".

You can keep both if you think future SoCs will be compatible.

They're likely to be compatible AFAIK.

Just to be clear, you mean keep both of them?

Yes, but...

e.g.
compatible: Should be "fsl,imx8qxp-scu-pd" or "fsl,scu-pd"

Not exactly, "fsl,scu-pd" should be a fallback and the driver can
match on that (until you need to distinguish). So like this:

Should be one of:
  "fsl,imx8qxp-scu-pd"
followed by "fsl,scu-pd"

Can you please help clarify a bit more on why it's better to do that as I'm not quite
Understand?

And for later when mx8qm is supported, should we add it again as follows?
compatible: Should be "fsl,imx8qm-scu-pd", "fsl,imx8qxp-scu-pd" or "fsl,scu-pd"

If formatted as above, then it is a one line change to add:

Should be one of:
  "fsl,imx8qm-scu-pd"
  "fsl,imx8qxp-scu-pd"
followed by "fsl,scu-pd"

Rob