Hi Pavel.
On Fri, Jun 29, 2018 at 9:29 AM Pavel Machek <pavel@xxxxxx> wrote:
> On Fri 2018-06-29 09:21:24, Geert Uytterhoeven wrote:
> > On Fri, Jun 29, 2018 at 9:17 AM Pavel Machek <pavel@xxxxxx> wrote:
> > > > > Yeah, well... I don't think decoding sentences in morse code is going
> > > > > to be much fun.
> > > > >
> > > > > And we don't really want encoder in kernel. Just do encoding in
> > > > > userspace, and use pattern trigger to display it.
> > > > >
> > > > > For many uses, morse code is "too geeky", and other patterns will be
> > > > > used.
> > > > >
> > > > > Like " X " for error 1, " X X " for error 2, " .xX .xX " for
> > > > > charging, " .xXx. " for everything okay...
> > > > >
> > > >
> > > > The hardware i'm using is not able to adjust brightness. It can just switch the
> > > > LED on or off. That's it.
> > > >
> > > > If anybody is interested i can submit a version 2 of the morse trigger with the
> > > > improvements suggested by Greg and Geert. Please let me know.
> > >
> > > I'd prefer more general pattern trigger.
> >
> > There are two parts here:
> > 1. The general pattern trigger (which is not yet upstream),
> > 2. A way to supply a string, convert it into morse code, and feed it to
> > the LED subsystem.
> >
> > Given 1 is not yet upstream, I think Andreas solution is fine.
> > Once 1 is upstream, 2 can be modified to feed into 1, if available.
>
> There's no reason 2. should be in kernel ("mechanism, not policy").
>
> Given 1. is not really more complicated than the morse
> trigger... please just help with pattern trigger if you want this to
> happen.
Ah, you want to do the conversion to morse code in userspace?
Yes, that makes sense.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
