On Wed 2020-10-14 10:05:42, Udo van den Heuvel wrote:
> On 14-10-2020 09:58, Pavel Machek wrote:
> > Contrary to his claims, Udo very probably has LEDs in his systems...
>
> We have a visible power LED.
> WE have a HDD LED.
> The board has LEDs, yes, but the SilverStone Fortress FT02 hides them
> fairly well.
> I did not ask for LEDs nor need them this way.
> It's a computer, not a disco-light or anything like that.
And you probably have numlock LED.
> Whether the code is big or not does not matter, it is a matter of being
> able to select what one needs without getting bothered with other code
> that will do nothing.
No.
Additional config options have costs, too, and we don't want to
support gazillion config options. LED core should be small enough that
it does not matter. Sound was inventing its own "tiny LED core"
before.
> So please consider.
I did. Answer is no. Please accept it.
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
On 14-10-2020 10:11, Pavel Machek wrote:
>> It's a computer, not a disco-light or anything like that.
>
> And you probably have numlock LED.
On the case? no way.
It is on the keyboard, a separate device, and already has a function.
We also have a caps lock LED and scroll lock LED there, with separate
functions.
I do not need 'extra' functionality for those.
> Additional config options have costs, too, and we don't want to
> support gazillion config options.
That is not the issue.
One should have thought about stuff beforehand.
One changes an existing situation and one should have understood that
new stuff should never be forced, no matter what the size or functionality.
The non-selectability is not my fault.
The perceived 'complexity' in the patch is not my fault.
My points of view w.r.t. the situation are fairly reasonable and common.
The fact that this change made it to the kernel means that people did
not look into it that well.
Kind regards,
Udo