Hi!

> Also, function is totally unclear. The whole reason we want to use
> Linux LEDs is triggers, and it is the selected trigger which
> determines the function.

Usually, yes. But "function" is what _manufacturer_ wanted LED to
mean, and "trigger" is how user is using it. I have currently disk
activity LED mapped on scrollock.

There are some mini desktops which have skull LED, probably meant to
be user defined LED. In such cases, user will have to select trigger.

> Colour is also an issue. The IGC Ethernet controller has no idea what
> colour the LEDs are in the RG-45 socket. And this is generic to
> Ethernet MAC and PHY chips. The data sheets never mention colour.

Maybe datasheet does not mention color, but the LED _has_ color.

> might know the colour in DT (and maybe ACPI) systems where you have
> specific information about the board. But in for PCIe card, USB
> dongles, etc, colour is unknown.

Not.. really. You don't know from chip specificiations, but you should
know the color as soon as you have USB IDs (because they should tell
you exact hardware). And I believe same is true for PCIe cards. It may
be more tricky if no actual PCIe card exists and it is just a chip
built into generic notebook. (But then, you should have notebook's DMI
id etc...?)

Anyway, you can leave the color field empty if you don't know.

Best regards,
								Pavel
-- 
http://www.livejournal.com/~pavelmachek