On Thu, 2011-01-13 at 15:15 -0500, Shawn J. Goff wrote: > I'd love to work on a rewrite, possibly making it work with the existing > WiMAX stack (from Intel). I've never done a driver, and probably can't > do it on my own at this point, but that's probably not going to stop me > from trying. First, though, I'm trying to get the existing one working. The current wimax userspace bits from intel (with the exception of wimax-tools, which Inaky wrote from scratch) are ugly and badly written. They don't need to be even 1/10th of the size they currently are. While converting the Beceem driver over to using the Intel wimaxd would certainly be the path of least resistance, the wimaxd code is a mess and you'll spend some time tearing your hair out trying to figure out how to modify that code. But you can certainly do it. The long-term path should probably be to write a device-agnostic wimaxd that supports multiple vendor's devices, since we now have at least two drivers. The Intel cards have like 10 or 20 different actual commands (ie connect, scan, hwinfo, etc) and that certainly doesn't warrant the current wimaxd's spaghetti. Having a cleaner-written, smaller wimaxd would also make it easier to fix bugs and add support for new devices. Dan > > On 01/13/2011 03:09 PM, dormando wrote: > >> then it sits around for a few seconds, and: > >> ERROR: RSA private key encryption timeout > >> ... which is from the beceem engine stuff, and at the same time the auth > >> log is spitting out what I noted before. > >> > >> It's also probably worth saying that no matter what I put in for > >> UserIdentity or TLSDevicePrivateKeyPassword I get the same timeout in the > >> same place. > > and of course, it was something stupid. I finally noticed that the > > "sprint4gdeveloperpack" thing had another firmware file in it, which has a > > different md5 than the one I pulled from my mac (the mac probably being > > newer). I used the one from the same source as the driver and it suddenly > > magically started working. > > > > But my other open question still stands; this beceem stuff is a mess and > > the driver doesn't seem very efficient. Is anyone working on it, or have > > any opinions on this wimaxd/wimac + beceemAPI thing either way? > > > > -Dormando > > _______________________________________________ > > wimax mailing list > > wimax at linuxwimax.org > > http://lists.linuxwimax.org/listinfo/wimax > _______________________________________________ > wimax mailing list > wimax at linuxwimax.org > http://lists.linuxwimax.org/listinfo/wimax
