On Tue 2021-03-16 16:34:50, Geert Uytterhoeven wrote:
> Hi Tianling,
> 
> CC Jacek, Pavel
> 
> On Tue, Mar 16, 2021 at 4:00 PM Tianling Shen <cnsztl@gmail.com> wrote:
> > On 2021-03-16 02:23 Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> > > Personally, I'm not so fond of the <foo>-%u node names, and prefer
> > > <foo>-<function>.  With the former, it's way too easy to have a silent
> > > override in your .dts(i) stack.
> > > Cfr. commit 45f5d5a9e34d3fe4 ("arm64: dts: renesas: r8a77995: draak:
> > > Fix backlight regulator name")
> >
> > How about using `lan-led`, `sys-led` and `wan-led` here?
> 
> Documentation/devicetree/bindings/leds/leds-gpio.yaml says "led-%u"
> is the preferred form, but that anything containing "led" as a substring
> is accepted.  So I'd go for "led-lan" etc.
> 
> BTW, you can validate your DTB against the leds-gpio DT bindings
> by running:
> 
>     make dtbs_check
> DT_SCHEMA_FILES=Documentation/devicetree/bindings/leds/leds-gpio.yaml
> 
> Background info for CCed parties:
>
https://lore.kernel.org/linux-arm-kernel/20210316150033.15987-1-cnsztl@gmail.com/

I don't care much either way, lan-0 is okay as is lan-led.

but...

+			label = "nanopi-r4s:green:lan";
+			label = "nanopi-r4s:red:sys";
+			label = "nanopi-r4s:green:wan";


It would be good to have common labels, that means LED_FUNCTION_LAN,
LED_FUNCTION_WAN, and figuring out something better than "sys",
possibly LED_FUNCTION_FAULT?

Thanks,
								Pavel

-- 
http://www.livejournal.com/~pavelmachek