Hey there. Sorry to nag, but it seems we came to the conclusion that there's indeed a problem with the current code, and then collectively decided that we're done. Can I do something to expedite this? Are there any issues with my proposed patch? If there are, what can I do to fix those? Alternatively, perhaps it would be easier/faster if a bona fide kernel developer could fix problems described in this thread (well, actually, I believe we're down to one somewhat pressing problem, which could be fixed just by moving some code around) On Thu, 17 Jan 2019 at 06:53, Gopal, Saranya <saranya.gopal@xxxxxxxxx> wrote: > > > The result is this: > > Before f13912d3f014a, if an RNDIS device has non-vendor-specific > > configurations after the first one, that one would be selected. > > After ff2a8c532c14, the first configuration would always be selected > > for RNDIS devices. Besides, there are several redundant checks in this > > case, which is, if nothing else, confusing. > > > > Hopefully I've explained my point clearly enough. > > Agreed. Thanks for pointing it out. I somehow missed this :( > > Thanks, > Saranya
