Re: [RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Jacek Anaszewski]

Hi Rob,

Any thoughts on my analysis? Do you think that LED function naming
standardization would really make sense having the abundance
of LED functions categories present in the mainline dts files?

The list of standard LED function names would have to be
continuously extended as people would be adding new boards.
Limiting users to only existing set of LED functions wouldn't
be practical IMHO.

I'd propose only to modify 'label' property description in LED common
bindings, so that it would explicitly state that it should contain only
"colour:functon" segments. It would be driver's responsibility to add
"devicename:" prefix to the label and use the whole string for a LED
class device name. Some drivers do that already. All DT bindings, dts
files and LED class drivers that don't adhere to this rule would have
to be updated accordingly.

Regarding colour - the "devicename:colour:function" LED device naming
convention defined in Documentation/leds/leds-class.txt is around for
a long time. Since LED colour can vary from board two board and is
independent of a driver, we have to have a means for defining it in
DT. Would you see providing separate 'colour' DT property as a solution?

Best regards,
Jacek Anaszewski

On 09/23/2017 11:12 PM, Jacek Anaszewski wrote:
> On 09/22/2017 11:07 PM, Jacek Anaszewski wrote:
>> On 09/20/2017 10:53 PM, Rob Herring wrote:
>>> On Tue, Sep 19, 2017 at 11:01:02PM +0200, Jacek Anaszewski wrote:
 Hi Pavel,

 On 09/18/2017 10:54 PM, Pavel Machek wrote:
> On Mon 2017-09-18 17:49:23, Sakari Ailus wrote:
>> Hi Pavel,
>>
>> On Mon, Sep 18, 2017 at 12:56:55PM +0200, Pavel Machek wrote:
>>> Hi!
>>>
 Specify the exact label used if the label property is omitted in DT, as
 well as use label in the example that conforms to LED device naming.

 @@ -69,11 +73,11 @@ Example
flash-max-microamp = <32>;
led-max-microamp = <6>;
ams,input-max-microamp = <175>;
 -  label = "as3645a:flash";
 +  label = "as3645a:white:flash";
};
indicator@1 {
reg = <0x1>;
led-max-microamp = <1>;
 -  label = "as3645a:indicator";
 +  label = "as3645a:red:indicator";
};
};
>>>
>>> Ok, but userspace still has no chance to determine if this is flash
>>> from main camera or flash for front camera; todays smartphones have
>>> flashes on both cameras.
>>>
>>> So.. Can I suggset as3645a:white:main_camera_flash or main_flash or
>>> ?
>>
>> If there's just a single one in the device, could you use that?
>>
>> Even if we name this so for N9 (and N900), the application still would 
>> only
>> work with the two devices.
>
> Well, I'd plan to name it on other devices, too.
>
>> My suggestion would be to look for a flash LED, and perhaps the maximum
>> current as well. That should generally work better than assumptions on 
>> the
>> label.
>
> If you just look for flash LED, you don't know if it is front one or
> back one. Its true that if you have just one flash it is usually on
> the back camera, but you can't know if maybe driver is not available
> for the main flash.
>
> Lets get this right, please "main_camera_flash" is 12 bytes more than
> "flash", and it saves application logic.. more than 12 bytes, I'm sure. 

 What you are trying to introduce is yet another level of LED class
 device naming standard, one level below devicename:colour:function.
 It seems you want also to come up with the set of standarized LED
 function names. This would certainly have to be covered for consistency.
>>>
>>> I really dislike how this naming convention is used for label. label is 
>>> supposed to be the phyically identifiable name. Having the devicename 
>>> defeats that. Perhaps color, too. We'd be better off with a color 
>>> property. It seems we're overloading the naming with too many things. 
>>> Now we're adding device association.
>>
>> Regarding devicename - there is indeed inconsistency in the way how LED
>> DT bindings use label, as some of them use it for defining full LED
>> class device name, and the rest fill only colour and function, leaving
>> addition of a devicename to the driver.
>>
>> The problem is also in current definition of label in LED common
>> bindings documentation, which says:
>>
>> "It has to uniquely identify a device, i.e. no other LED class device
>> can be assigned the same label."
>>
>> In view of your above words this is not true, and we probably should
>> remove this sentence (it doesn't have DT maintainer ack btw).
>>
>>> I do want to 

Re: [RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Jacek Anaszewski]

On 09/22/2017 11:07 PM, Jacek Anaszewski wrote:
> On 09/20/2017 10:53 PM, Rob Herring wrote:
>> On Tue, Sep 19, 2017 at 11:01:02PM +0200, Jacek Anaszewski wrote:
>>> Hi Pavel,
>>>
>>> On 09/18/2017 10:54 PM, Pavel Machek wrote:
 On Mon 2017-09-18 17:49:23, Sakari Ailus wrote:
> Hi Pavel,
>
> On Mon, Sep 18, 2017 at 12:56:55PM +0200, Pavel Machek wrote:
>> Hi!
>>
>>> Specify the exact label used if the label property is omitted in DT, as
>>> well as use label in the example that conforms to LED device naming.
>>>
>>> @@ -69,11 +73,11 @@ Example
>>> flash-max-microamp = <32>;
>>> led-max-microamp = <6>;
>>> ams,input-max-microamp = <175>;
>>> -   label = "as3645a:flash";
>>> +   label = "as3645a:white:flash";
>>> };
>>> indicator@1 {
>>> reg = <0x1>;
>>> led-max-microamp = <1>;
>>> -   label = "as3645a:indicator";
>>> +   label = "as3645a:red:indicator";
>>> };
>>> };
>>
>> Ok, but userspace still has no chance to determine if this is flash
>> from main camera or flash for front camera; todays smartphones have
>> flashes on both cameras.
>>
>> So.. Can I suggset as3645a:white:main_camera_flash or main_flash or
>> ?
>
> If there's just a single one in the device, could you use that?
>
> Even if we name this so for N9 (and N900), the application still would 
> only
> work with the two devices.

 Well, I'd plan to name it on other devices, too.

> My suggestion would be to look for a flash LED, and perhaps the maximum
> current as well. That should generally work better than assumptions on the
> label.

 If you just look for flash LED, you don't know if it is front one or
 back one. Its true that if you have just one flash it is usually on
 the back camera, but you can't know if maybe driver is not available
 for the main flash.

 Lets get this right, please "main_camera_flash" is 12 bytes more than
 "flash", and it saves application logic.. more than 12 bytes, I'm sure. 
>>>
>>> What you are trying to introduce is yet another level of LED class
>>> device naming standard, one level below devicename:colour:function.
>>> It seems you want also to come up with the set of standarized LED
>>> function names. This would certainly have to be covered for consistency.
>>
>> I really dislike how this naming convention is used for label. label is 
>> supposed to be the phyically identifiable name. Having the devicename 
>> defeats that. Perhaps color, too. We'd be better off with a color 
>> property. It seems we're overloading the naming with too many things. 
>> Now we're adding device association.
> 
> Regarding devicename - there is indeed inconsistency in the way how LED
> DT bindings use label, as some of them use it for defining full LED
> class device name, and the rest fill only colour and function, leaving
> addition of a devicename to the driver.
> 
> The problem is also in current definition of label in LED common
> bindings documentation, which says:
> 
> "It has to uniquely identify a device, i.e. no other LED class device
> can be assigned the same label."
> 
> In view of your above words this is not true, and we probably should
> remove this sentence (it doesn't have DT maintainer ack btw).
> 
>> I do want to see standard names though. On 96boards for example, there 
>> are defined LEDs and locations. The function on some are defined (e.g. 
>> WiFi/BT) and somewhat undefined on others (user{1-4}). I'd like to see 
>> the same label across all boards.
> 
> Currently we have following LED functions (obtained with
> grep label Documentation/devicetree/bindings/leds/* | sed
> s'/^.*label/label/g' | awk -F"=" '{print $2}' | sed '/^$/d' | sed
> s'/.*:\(.*\)";/\1/' | sed '/^\s\{1,\}/d' | sort -u)
> 
> 0
> 1
> 2
> 2g
> 3
> 4
> 5
> 6
> 7
> adsl
> alarm
> alive
> aux
> broadband
> chrg
> dsl
> flash
> green
> indicator
> inet
> keypad
> phone
> power
> red
> sata
> sata0
> sata1
> tel
> tv
> upgrading
> usb
> usr0
> usr1
> usr35
> wan
> white
> wireless
> wps
> yellow
> 
> By extracting numerical pattern names and replacing numbers with N
> we're getting something like this:
> 
> N
> Ng
> colour
> adsl
> alarm
> alive
> aux
> broadband
> chrg
> dsl
> flash
> indicator
> inet
> keypad
> phone
> power
> sataN
> tel
> tv
> upgrading
> usb
> usrN
> wan
> wireless
> wps
> 
> Is this list something you'd like to see as a base of standard LED
> functions? It seems that this list would have to be continuously
> supplemented with new positions.
> 

Even better option is to grep through all *.dts* files in the arch
directory. A slightly modified command 

Re: [RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Jacek Anaszewski]

On 09/20/2017 10:53 PM, Rob Herring wrote:
> On Tue, Sep 19, 2017 at 11:01:02PM +0200, Jacek Anaszewski wrote:
>> Hi Pavel,
>>
>> On 09/18/2017 10:54 PM, Pavel Machek wrote:
>>> On Mon 2017-09-18 17:49:23, Sakari Ailus wrote:
 Hi Pavel,

 On Mon, Sep 18, 2017 at 12:56:55PM +0200, Pavel Machek wrote:
> Hi!
>
>> Specify the exact label used if the label property is omitted in DT, as
>> well as use label in the example that conforms to LED device naming.
>>
>> @@ -69,11 +73,11 @@ Example
>>  flash-max-microamp = <32>;
>>  led-max-microamp = <6>;
>>  ams,input-max-microamp = <175>;
>> -label = "as3645a:flash";
>> +label = "as3645a:white:flash";
>>  };
>>  indicator@1 {
>>  reg = <0x1>;
>>  led-max-microamp = <1>;
>> -label = "as3645a:indicator";
>> +label = "as3645a:red:indicator";
>>  };
>>  };
>
> Ok, but userspace still has no chance to determine if this is flash
> from main camera or flash for front camera; todays smartphones have
> flashes on both cameras.
>
> So.. Can I suggset as3645a:white:main_camera_flash or main_flash or
> ?

 If there's just a single one in the device, could you use that?

 Even if we name this so for N9 (and N900), the application still would only
 work with the two devices.
>>>
>>> Well, I'd plan to name it on other devices, too.
>>>
 My suggestion would be to look for a flash LED, and perhaps the maximum
 current as well. That should generally work better than assumptions on the
 label.
>>>
>>> If you just look for flash LED, you don't know if it is front one or
>>> back one. Its true that if you have just one flash it is usually on
>>> the back camera, but you can't know if maybe driver is not available
>>> for the main flash.
>>>
>>> Lets get this right, please "main_camera_flash" is 12 bytes more than
>>> "flash", and it saves application logic.. more than 12 bytes, I'm sure. 
>>
>> What you are trying to introduce is yet another level of LED class
>> device naming standard, one level below devicename:colour:function.
>> It seems you want also to come up with the set of standarized LED
>> function names. This would certainly have to be covered for consistency.
> 
> I really dislike how this naming convention is used for label. label is 
> supposed to be the phyically identifiable name. Having the devicename 
> defeats that. Perhaps color, too. We'd be better off with a color 
> property. It seems we're overloading the naming with too many things. 
> Now we're adding device association.

Regarding devicename - there is indeed inconsistency in the way how LED
DT bindings use label, as some of them use it for defining full LED
class device name, and the rest fill only colour and function, leaving
addition of a devicename to the driver.

The problem is also in current definition of label in LED common
bindings documentation, which says:

"It has to uniquely identify a device, i.e. no other LED class device
can be assigned the same label."

In view of your above words this is not true, and we probably should
remove this sentence (it doesn't have DT maintainer ack btw).

> I do want to see standard names though. On 96boards for example, there 
> are defined LEDs and locations. The function on some are defined (e.g. 
> WiFi/BT) and somewhat undefined on others (user{1-4}). I'd like to see 
> the same label across all boards.

Currently we have following LED functions (obtained with
grep label Documentation/devicetree/bindings/leds/* | sed
s'/^.*label/label/g' | awk -F"=" '{print $2}' | sed '/^$/d' | sed
s'/.*:\(.*\)";/\1/' | sed '/^\s\{1,\}/d' | sort -u)

0
1
2
2g
3
4
5
6
7
adsl
alarm
alive
aux
broadband
chrg
dsl
flash
green
indicator
inet
keypad
phone
power
red
sata
sata0
sata1
tel
tv
upgrading
usb
usr0
usr1
usr35
wan
white
wireless
wps
yellow

By extracting numerical pattern names and replacing numbers with N
we're getting something like this:

N
Ng
colour
adsl
alarm
alive
aux
broadband
chrg
dsl
flash
indicator
inet
keypad
phone
power
sataN
tel
tv
upgrading
usb
usrN
wan
wireless
wps

Is this list something you'd like to see as a base of standard LED
functions? It seems that this list would have to be continuously
supplemented with new positions.

-- 
Best regards,
Jacek Anaszewski

Re: [RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Rob Herring]

On Tue, Sep 19, 2017 at 11:01:02PM +0200, Jacek Anaszewski wrote:
> Hi Pavel,
> 
> On 09/18/2017 10:54 PM, Pavel Machek wrote:
> > On Mon 2017-09-18 17:49:23, Sakari Ailus wrote:
> >> Hi Pavel,
> >>
> >> On Mon, Sep 18, 2017 at 12:56:55PM +0200, Pavel Machek wrote:
> >>> Hi!
> >>>
>  Specify the exact label used if the label property is omitted in DT, as
>  well as use label in the example that conforms to LED device naming.
> 
>  @@ -69,11 +73,11 @@ Example
>   flash-max-microamp = <32>;
>   led-max-microamp = <6>;
>   ams,input-max-microamp = <175>;
>  -label = "as3645a:flash";
>  +label = "as3645a:white:flash";
>   };
>   indicator@1 {
>   reg = <0x1>;
>   led-max-microamp = <1>;
>  -label = "as3645a:indicator";
>  +label = "as3645a:red:indicator";
>   };
>   };
> >>>
> >>> Ok, but userspace still has no chance to determine if this is flash
> >>> from main camera or flash for front camera; todays smartphones have
> >>> flashes on both cameras.
> >>>
> >>> So.. Can I suggset as3645a:white:main_camera_flash or main_flash or
> >>> ?
> >>
> >> If there's just a single one in the device, could you use that?
> >>
> >> Even if we name this so for N9 (and N900), the application still would only
> >> work with the two devices.
> > 
> > Well, I'd plan to name it on other devices, too.
> > 
> >> My suggestion would be to look for a flash LED, and perhaps the maximum
> >> current as well. That should generally work better than assumptions on the
> >> label.
> > 
> > If you just look for flash LED, you don't know if it is front one or
> > back one. Its true that if you have just one flash it is usually on
> > the back camera, but you can't know if maybe driver is not available
> > for the main flash.
> > 
> > Lets get this right, please "main_camera_flash" is 12 bytes more than
> > "flash", and it saves application logic.. more than 12 bytes, I'm sure. 
> 
> What you are trying to introduce is yet another level of LED class
> device naming standard, one level below devicename:colour:function.
> It seems you want also to come up with the set of standarized LED
> function names. This would certainly have to be covered for consistency.

I really dislike how this naming convention is used for label. label is 
supposed to be the phyically identifiable name. Having the devicename 
defeats that. Perhaps color, too. We'd be better off with a color 
property. It seems we're overloading the naming with too many things. 
Now we're adding device association.

I do want to see standard names though. On 96boards for example, there 
are defined LEDs and locations. The function on some are defined (e.g. 
WiFi/BT) and somewhat undefined on others (user{1-4}). I'd like to see 
the same label across all boards.

Rob

Re: [RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Jacek Anaszewski]

Hi Pavel,

On 09/18/2017 10:54 PM, Pavel Machek wrote:
> On Mon 2017-09-18 17:49:23, Sakari Ailus wrote:
>> Hi Pavel,
>>
>> On Mon, Sep 18, 2017 at 12:56:55PM +0200, Pavel Machek wrote:
>>> Hi!
>>>
 Specify the exact label used if the label property is omitted in DT, as
 well as use label in the example that conforms to LED device naming.

 @@ -69,11 +73,11 @@ Example
flash-max-microamp = <32>;
led-max-microamp = <6>;
ams,input-max-microamp = <175>;
 -  label = "as3645a:flash";
 +  label = "as3645a:white:flash";
};
indicator@1 {
reg = <0x1>;
led-max-microamp = <1>;
 -  label = "as3645a:indicator";
 +  label = "as3645a:red:indicator";
};
};
>>>
>>> Ok, but userspace still has no chance to determine if this is flash
>>> from main camera or flash for front camera; todays smartphones have
>>> flashes on both cameras.
>>>
>>> So.. Can I suggset as3645a:white:main_camera_flash or main_flash or
>>> ?
>>
>> If there's just a single one in the device, could you use that?
>>
>> Even if we name this so for N9 (and N900), the application still would only
>> work with the two devices.
> 
> Well, I'd plan to name it on other devices, too.
> 
>> My suggestion would be to look for a flash LED, and perhaps the maximum
>> current as well. That should generally work better than assumptions on the
>> label.
> 
> If you just look for flash LED, you don't know if it is front one or
> back one. Its true that if you have just one flash it is usually on
> the back camera, but you can't know if maybe driver is not available
> for the main flash.
> 
> Lets get this right, please "main_camera_flash" is 12 bytes more than
> "flash", and it saves application logic.. more than 12 bytes, I'm sure. 

What you are trying to introduce is yet another level of LED class
device naming standard, one level below devicename:colour:function.
It seems you want also to come up with the set of standarized LED
function names. This would certainly have to be covered for consistency.

-- 
Best regards,
Jacek Anaszewski

Re: [RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Pavel Machek]

On Mon 2017-09-18 17:49:23, Sakari Ailus wrote:
> Hi Pavel,
> 
> On Mon, Sep 18, 2017 at 12:56:55PM +0200, Pavel Machek wrote:
> > Hi!
> > 
> > > Specify the exact label used if the label property is omitted in DT, as
> > > well as use label in the example that conforms to LED device naming.
> > > 
> > > @@ -69,11 +73,11 @@ Example
> > >   flash-max-microamp = <32>;
> > >   led-max-microamp = <6>;
> > >   ams,input-max-microamp = <175>;
> > > - label = "as3645a:flash";
> > > + label = "as3645a:white:flash";
> > >   };
> > >   indicator@1 {
> > >   reg = <0x1>;
> > >   led-max-microamp = <1>;
> > > - label = "as3645a:indicator";
> > > + label = "as3645a:red:indicator";
> > >   };
> > >   };
> > 
> > Ok, but userspace still has no chance to determine if this is flash
> > from main camera or flash for front camera; todays smartphones have
> > flashes on both cameras.
> > 
> > So.. Can I suggset as3645a:white:main_camera_flash or main_flash or
> > ?
> 
> If there's just a single one in the device, could you use that?
> 
> Even if we name this so for N9 (and N900), the application still would only
> work with the two devices.

Well, I'd plan to name it on other devices, too.

> My suggestion would be to look for a flash LED, and perhaps the maximum
> current as well. That should generally work better than assumptions on the
> label.

If you just look for flash LED, you don't know if it is front one or
back one. Its true that if you have just one flash it is usually on
the back camera, but you can't know if maybe driver is not available
for the main flash.

Lets get this right, please "main_camera_flash" is 12 bytes more than
"flash", and it saves application logic.. more than 12 bytes, I'm sure. 

> For association with a particular camera --- in the long run I'd propose to
> use Media controller / property API for that in the long run. We don't have
> that yet though.

We don't have that yet. Plus simple applications may not want to talk
v4l2 ioctls

Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) 
http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html


signature.asc
Description: Digital signature

Re: [RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Sakari Ailus]

Hi Pavel,

On Mon, Sep 18, 2017 at 12:56:55PM +0200, Pavel Machek wrote:
> Hi!
> 
> > Specify the exact label used if the label property is omitted in DT, as
> > well as use label in the example that conforms to LED device naming.
> > 
> > @@ -69,11 +73,11 @@ Example
> > flash-max-microamp = <32>;
> > led-max-microamp = <6>;
> > ams,input-max-microamp = <175>;
> > -   label = "as3645a:flash";
> > +   label = "as3645a:white:flash";
> > };
> > indicator@1 {
> > reg = <0x1>;
> > led-max-microamp = <1>;
> > -   label = "as3645a:indicator";
> > +   label = "as3645a:red:indicator";
> > };
> > };
> 
> Ok, but userspace still has no chance to determine if this is flash
> from main camera or flash for front camera; todays smartphones have
> flashes on both cameras.
> 
> So.. Can I suggset as3645a:white:main_camera_flash or main_flash or
> ?

If there's just a single one in the device, could you use that?

Even if we name this so for N9 (and N900), the application still would only
work with the two devices.

My suggestion would be to look for a flash LED, and perhaps the maximum
current as well. That should generally work better than assumptions on the
label.

For association with a particular camera --- in the long run I'd propose to
use Media controller / property API for that in the long run. We don't have
that yet though.

Cc Jacek, too.

-- 
Sakari Ailus
e-mail: sakari.ai...@iki.fi

Re: [RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Pavel Machek]

Hi!

> Specify the exact label used if the label property is omitted in DT, as
> well as use label in the example that conforms to LED device naming.
> 
> @@ -69,11 +73,11 @@ Example
>   flash-max-microamp = <32>;
>   led-max-microamp = <6>;
>   ams,input-max-microamp = <175>;
> - label = "as3645a:flash";
> + label = "as3645a:white:flash";
>   };
>   indicator@1 {
>   reg = <0x1>;
>   led-max-microamp = <1>;
> - label = "as3645a:indicator";
> + label = "as3645a:red:indicator";
>   };
>   };

Ok, but userspace still has no chance to determine if this is flash
from main camera or flash for front camera; todays smartphones have
flashes on both cameras.

So.. Can I suggset as3645a:white:main_camera_flash or main_flash or
?

Thanks,
Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) 
http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html


signature.asc
Description: Digital signature

[RESEND PATCH v2 4/6] dt: bindings: as3645a: Improve label documentation, DT example

[Sakari Ailus]

Specify the exact label used if the label property is omitted in DT, as
well as use label in the example that conforms to LED device naming.

Signed-off-by: Sakari Ailus 
---
 Documentation/devicetree/bindings/leds/ams,as3645a.txt | 12 
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/Documentation/devicetree/bindings/leds/ams,as3645a.txt 
b/Documentation/devicetree/bindings/leds/ams,as3645a.txt
index fdc40e354a64..9adba41e74b3 100644
--- a/Documentation/devicetree/bindings/leds/ams,as3645a.txt
+++ b/Documentation/devicetree/bindings/leds/ams,as3645a.txt
@@ -39,7 +39,9 @@ ams,input-max-microamp: Maximum flash controller input 
current. The
 Optional properties of the flash child node
 ===
 
-label  : The label of the flash LED.
+label  : The label of the flash LED. The label is otherwise assumed to
+ be the device node name concatenated with ":white:" and the
+ name of the flash LED node is assumed if omitted.
 
 
 Required properties of the indicator child node (1)
@@ -52,7 +54,9 @@ led-max-microamp: Maximum indicator current. The allowed 
values are
 Optional properties of the indicator child node
 ===
 
-label  : The label of the indicator LED.
+label  : The label of the indicator LED. The label is otherwise assumed
+ to be the device node name concatenated with ":red:" and the
+ name of the indicator LED node is assumed if omitted.
 
 
 Example
@@ -69,11 +73,11 @@ Example
flash-max-microamp = <32>;
led-max-microamp = <6>;
ams,input-max-microamp = <175>;
-   label = "as3645a:flash";
+   label = "as3645a:white:flash";
};
indicator@1 {
reg = <0x1>;
led-max-microamp = <1>;
-   label = "as3645a:indicator";
+   label = "as3645a:red:indicator";
};
};
-- 
2.11.0