On Thu, Aug 30, 2018 at 03:34:01PM -0700, Joe Perches wrote: > On Wed, 2018-08-29 at 22:55 +0100, John Whitmore wrote: > > On Wed, Aug 29, 2018 at 04:21:54PM -0500, Larry Finger wrote: > > > On 08/29/2018 04:14 PM, Joe Perches wrote: > [] > > > > Perhaps support for the chipset could be converted to use > > > > mac80211 and moved into the directory with the other realtek > > > > drivers in drivers/net/wireless/realtek/rtl8xxxu/... > [] > > > All the Realtek USB devices should be added to rtl8xxxu, not merely moved > > > into that directory. Jes Sorensen created a well-designed driver the is > > > structured to permit addition of different initialization routines, etc. > > > That said, the conversion will not be easy. In addition, it will require > > > having your hands on a real device - a requirement that I cannot meet for > > > the RTL8192U. > [] > > > Oops... it probably doesn't look like it, at the moment, but 'actual code > > changes to this driver' was where I'm hoping to get to. There's a lot of > > stuff in the header files, including member variables, which are not used > > in the code. I was hoping to strip these out incrementally to minimise > > the amount of source code that has to be understood. I'm not familiar > > with the overall network, or 80211, subsystem architecture, but it's hard > > to see why this driver has it's own private ieee80211.h file. > > That's true and the reason why this code should > eventually be deleted if possible and its functionality > integrated into the existing realtek/rtl8xxxxu code. > > > I must try and get my hands on the device in question. > > That'd be good, though likely this device is obsolete. > Good luck finding one. > Maybe if the whole driver was deleted nobody would notice ;) Thanks for your help. I'm afraid I carried on in the same vain cleaning stuff up but I do have to look at realtek/rtl8xxxxu, thanks for pointing out where the goalposts are, miles away ;) > > Thank ye for your comments, and sorry for the "somewhat useful" changes > > to date, but doing these changes is letting me get the hang of this code, > > which I refuse to comment on because there are probably guidelines on using > > naughty language on here. ;) > > Feel free. > > The one look I took at any existing multi-platform realtek > code with all the #ifdefs made me not want look at it again. > > https://github.com/hadess/rtl8723bs/ > > I said a naughty under my breath too. _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
