net: phy: mxl-gpy: Add mode for 2 leds

[PATCH net-next 0/2] net: phy: mxl-gpy: Add mode for 2 leds

Posted by Marcus Carlberg 3 years, 6 months ago

GPY211 phy default to using all four led pins.
Hardwares using only two leds where led0 is used as the high
network speed led and led1 the low network speed led will not
get the correct behaviour since 1Gbit and 2.5Gbit will not be
represented at all in the existing leds.


Marcus Carlberg (2):
  dt-bindings: net: Add mxl,gpy
  net: phy: mxl-gpy: Add mode for 2 leds

 .../devicetree/bindings/net/mxl,gpy.yaml      | 39 ++++++++++++++++
 .../devicetree/bindings/vendor-prefixes.yaml  |  2 +
 MAINTAINERS                                   |  1 +
 drivers/net/phy/mxl-gpy.c                     | 45 +++++++++++++++++++
 4 files changed, 87 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/mxl,gpy.yaml

--
2.20.1

Re: [PATCH net-next 0/2] net: phy: mxl-gpy: Add mode for 2 leds

Posted by Michael Walle 3 years, 6 months ago

Hi,

> GPY211 phy default to using all four led pins.
> Hardwares using only two leds where led0 is used as the high
> network speed led and led1 the low network speed led will not
> get the correct behaviour since 1Gbit and 2.5Gbit will not be
> represented at all in the existing leds.

I might be wrong, but PHY LED bindings should be integrated with/using
the LED subsystem. Although I didn't see any development regarding this
for a long time.

That being said, it seems you are adding a new (DT) property which
just matches your particular hardware design, no?

-michael

Re: [PATCH net-next 0/2] net: phy: mxl-gpy: Add mode for 2 leds

Posted by Andrew Lunn 3 years, 6 months ago

On Thu, Sep 22, 2022 at 10:05:29AM +0200, Michael Walle wrote:
> Hi,
> 
> > GPY211 phy default to using all four led pins.
> > Hardwares using only two leds where led0 is used as the high
> > network speed led and led1 the low network speed led will not
> > get the correct behaviour since 1Gbit and 2.5Gbit will not be
> > represented at all in the existing leds.
> 
> I might be wrong, but PHY LED bindings should be integrated with/using
> the LED subsystem. Although I didn't see any development regarding this
> for a long time.
> 
> That being said, it seems you are adding a new (DT) property which
> just matches your particular hardware design, no?

Thanks Michael

Replying to this was on my TODO list. We have NACKed patches like this
for a few years now, wanting the Linux LED subsystem be used to
configure PHY LEDs. The patches to implement the core of that
sometimes makes some progress, then goes dormant for a while. The last
effort was getting close to being ready. Maybe the needs for this
driver can help get them finished?

       Andrew