Andy Green wrote: > Maybe Werner has some ideas about moving the rfkill out into machine > specific code so this "doesn't delay ar6000 sumbission to upstream" any > longer than the other brazillion problems with the code for upstream will.
hif2.c is actually about the only thing that almost passes general guidelines ;-) Henrique promised a while ago that he'd add some infrastructure for this to the general rfkill code. I didn't see it in the week or so following this, but I have to check if he's posted it since. - Werner
