On Tue, Jul 2, 2019 at 8:42 PM Jes Sorensen <jes.sorensen@xxxxxxxxx> wrote: > We definitely don't want to bring over the vendor code, since it's a > pile of spaghetti, but we probably need to get something sorted. This > went down the drain when the bluetooth driver was added without taking > it into account - long after this driver was merged. Yeah, I didn't mean bring over quite so literally.. Chris is studying it and figuring out the neatest way to reimplement the required bits. As for the relationship with bluetooth.. actually the bug that Chris is working on here is that the rtl8xxxu wifi signal is totally unusable *until* the bluetooth driver is loaded. Once the bluetooth driver is loaded, at the point of bluetooth firmware upload, the rtl8xxxu signal magiaclly strength becomes good. I think this is consistent with other rtl8xxxu problem reports that we saw lying around, although they had not been diagnosed in so much detail. The rtl8723bu vendor driver does not suffer this problem, it works fine with or without the bluetooth driver in place. Daniel