hi! Ok, so it's because in theory the firmware could be buggy and need patching via the ath3kfw tool (which gosh I really should just import into -head already) before it starts up.
can you clone https://github.com/erikarn/ath3k and try to load in what it thinks the right patch set / config file is? See if it comes up ok? Now that we have the hotplug work from warner and others I bet we could import ath3k and just have it autorun via devd in a non-terrible fashion. -adrian On Wed, 25 Mar 2020 at 23:09, Chris <bsd-li...@bsdforge.com> wrote: > > On Wed, 25 Mar 2020 19:05:40 -0700 (PDT) jason-fbsd-blueto...@shalott.net said > > > >> Hello. I am trying to get an ath3k-based USB bluetooth adapter > > >> working. I previously had this adapter working under FreeBSD, several > > >> years ago[.] After loading the firmware, it is not detected by ng_ubt. > > > > > I tracked this down: > > > > > > https://svnweb.freebsd.org/base?view=revision&revision=249178 > > > > > > Can someone explain why these devices were blacklisted from the ng_ubt > > > driver? It seems like the devices will fail to work if the firmware is > > > not loaded to the device before ng_ubt is loaded into the kernel; but it > > > seems like the failure mode is just that those devices don't work in > > > that case. So blacklisting them from the driver seems a lot worse... > > > > Pinging again on this. > > > > Any chance we can revert the above commit? > > > > After reverting the above commit on my box, I am able to fully use my > > bluetooth adapter (pair HID devices, play audio through my headset with > > virtual_oss, etc). I don't want to have to maintain a custom kernel in > > perpetuity to maintain that capability. > > > > Am I missing something about the current situation? As far as I can tell, > > with all of those devices blacklisted in the ng_ubt driver, there is no > > way to use those devices on FreeBSD. But if those devices are re-instated > > in ng_ubt, the only downside is that they _might_ not work, because it's > > left as an exercise to the user to make sure that the device firmware is > > pushed to the device before ng_ubt is loaded into the kernel. So my > > understanding is: current situation, those devices are impossible to use; > > reverting the above commit, those devices might be usable if the user > > knows what they're doing. Is that the wrong understanding? And if it's > > correct, is there any reason to not re-instate them? > > > > > > If some other change is needed or wanted instead of just reverting the > > above commit -- extra checks, warning log messages, etc -- I would be > > happy to take a run at it if someone could describe what's needed. I > > don't know anything about Bluetooth and don't have any kernel-hacking > > experience, but I am an experienced C programmer. > It might be worth opening a pr (https://bugs.freebsd.org) for this. Doing > so might give it higher visibility, and I *think* that's also the > preferred direction. After all. As far as your concerned, this seems like > a bug. No? :) > > --Chris > > > > Thanks. > > > > > > -Jason > > > > _______________________________________________ > > freebsd-bluetooth@freebsd.org mailing list > > https://lists.freebsd.org/mailman/listinfo/freebsd-bluetooth > > To unsubscribe, send any mail to "freebsd-bluetooth-unsubscr...@freebsd.org" > > _______________________________________________ freebsd-bluetooth@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-bluetooth To unsubscribe, send any mail to "freebsd-bluetooth-unsubscr...@freebsd.org"
