On Thu, 2009-04-30 at 17:37 +0300, Paulius Zaleckas wrote: > Inaky Perez-Gonzalez wrote: > > On Wednesday 22 April 2009, Darius Augulis wrote: > >> Hi Dirk, > >> > >> Dirk Brandewie wrote: > >>> Hi Darius, > >>> > >>> On Wed, 2009-04-22 at 06:46 -0700, Darius Augulis wrote: > >>>> Hi All, > >>>> > >>>> What is the state of this project? What are the main ideas of it? > >>>> I read old posts from Dirk Brandewie and now I'm looking for more info. > >>> The main idea of the project is to provide a standard interface to WiMax > >>> devices to the Connection/Network managers in the system. The current > >>> definition of the interface is available via git at: > >>> git://linuxwimax.org/dbus_wimaxapi.git > >> ok, that's exactly what I imagined. > >> > >>> The state of the project is I am trying to free up the resources/time to > >>> get the implementation done, my team has been consumed getting a release > >>> of another product over the last couple of months. > >>> > >>> The current state of the interface definition is the result of a lot of > >>> work and input from others on the list, I just happen to be holding the > >>> the pen ATM. I would be interested in your comments on the interface > >>> and how it maps to your adapter. > >> I will review your work and try to test in my project. > > > > In addition to this, we are working on a kernel level API that would work > > at > > WiMAX NAP level -- the proposal is here: > > > > http://linuxwimax.org/pipermail/wimax/2008-December/000300.html > > > > The idea is that that API is provided by the kernel (accessible as commands > > / > > signals marshalled over generic netlink) and the driver implements the > > backend of them, the WiMAX stack doing as much of the common heavy lifting > > as > > possible. > > > > Devices that expose a basestation-level or Network Access Provider level > > interface can use that. I haven't heard yet of any devices that work at the > > NSP level without host aid. > > > > Would this interface work for your device? > > I am working together with Darius on this Fujitsu WiMAX project. > Darius - connection manager and me - driver. > > I gave this interface quick review and it looks like it should work perfectly. > I plan to implement network discovery during next week on this interface. > Will let you know if it fits as expected.
Great; the largest problem we have right now with WiMAX drivers is that Intel is the only one with an open-source driver, so we only know what level their hardware and driver plug together. Getting at least one more vendor's hardware supported and figuring out how the current kernel WiMAX API needs to be adjusted to support that new hardware will help prove (and improve) the current API. Dan _______________________________________________ wimax mailing list [email protected] http://www.linuxwimax.org/mailman/listinfo/wimax
