[PATCH] libertas: implement new scanning logic

[Holger Schurig]

> So I tested this one, not the one from libertas-dev (which was posted
> a day earlier).  And this one seems fine.  Ack for commit to
> libertas-2.6.

Hmm, for the patch you just "acked" inofficially is already older.
I know at least two changes that I made since then:

* add a comment to the section where I change the basic rates by
  setting the high bit (your idea)

* adapt to the patch "ibertas: separate mesh connectivity from that
  of the main interface" that appeared in wireless-2.6/everything

Please note that my current patch doesn't apply anymore to 
wireless-2.6/everything (a.k.a wl26e). It might still apply to 
libertas-2.6, because this tree lacked (four hours ago) at least 4 
patches that are in wl26e.

From the 40 mails about git, commits, patches, rebases etc I was not 
able to draw a real conclusion. I'm tending towards basing my patches 
against wl26e, so that John, Jeff etc can push the patches easily up. 
And if a patch still applies, I can commit/push the patch once you 
acked them. If the patch doesn't apply there, I'll post about this face 
in libertas-dev. I hope that helps all sides the most.


> Would you like me to confirm on SDIO?

Yes, please. I think the patch is hardware agnistic, but you can
never know because SDIO firmware != CF firmware != USB firmware.