They could be; I've had code to do it. It never stays in
very long, because for a pure userspace interace it can't
be helpful unless userspace makes the labels: the labels
displayed otherwise just reuse the same constant.
For a "kernel cooperates with userspace" model that might
be more useful. And I've certainly found the debugfs info
hard to interpret without such labels!! But Trent has
said he doesn't see anyone modifying enough code to adopt
such a "cooperates" model. Which is why this particular
side-argument seems like a waste of time to me: even he
has not proposed the key change his argument depends on.
The original control file syntax I thought about was
export NN mode label
where it would gpio_request(NN, "label") and then set the
direction (and maybe value) according to "mode". But that
was just a PITA to cope with ... so it's now been trimmed
down to just "NN" (and "-NN").
- Dave
--
