Hi Pavel, On 11/11/2018 01:02 PM, Pavel Machek wrote: > Hi! > >> Add public led_compose_name() API for composing LED class device >> name basing on fwnode_handle data. The function composes device name >> according to either a new <color:function> pattern or the legacy >> <devicename:color:function> pattern. The decision on using the >> particular pattern is made basing on whether fwnode contains new >> "function" and "color" properties, or the legacy "label" proeprty. >> >> Backwards compatibility with in-driver hard-coded LED class device >> names is assured thanks to the default_desc argument. >> >> In case none of the aformentioned properties was found, then, for OF >> nodes, the node name is adopted for LED class device name. >> > >> diff --git a/Documentation/leds/leds-class.txt b/Documentation/leds/leds-class.txt >> index 836cb16..e9009c4 100644 >> --- a/Documentation/leds/leds-class.txt >> +++ b/Documentation/leds/leds-class.txt >> @@ -43,7 +43,7 @@ LED Device Naming >> >> Is currently of the form: >> >> -"devicename:colour:function" >> +"colour:function" >> > > I don't think we want to do it in all cases. > > So, on my cellphone seeing lp5523:green:led is indeed not useful. > > But on notebook with usb keyboard attached, you need to keep the > devicename to be able to distinguish capslock on internal keyboard and > capslock on first USB keyboard and capslock on second USB keyboard. > > Taking look at the list of functions, here's stuff like "hdd" and > "hdderr". I assume the first means hdd activity... If we can do it, it > would be nicest to have "sda:green:activity" and maybe > "sda:red:error". For a raid array with 8 drives... > > For example I have a router running linux. It has 4 lan ports, with > correspondings LED, and an wan led. > > Having "green:lan1" to "green:lan4" and "green:wan" plus > "red:wanerror" would work, but I'd really preffer > "eth0:green:link"... "adsl0:green:link", "adsl0:red:error". > > There are now phones with flashes on both main and selfie > cameras. Again, knowing which device is which is important. As is > knowing which display is controlled by particular backlight. It's overcomplicating the naming again. In every case you can tweak the function name to eth0_link, eth1_link etc. -- Best regards, Jacek Anaszewski