Barry Day <briselec@xxxxxxxxx> writes: > On Mon, Nov 21, 2016 at 11:59:47AM -0500, Jes Sorensen wrote: >> Barry Day <briselec@xxxxxxxxx> writes: >> > Testing is simple. Connect to an AP in the usual >> > manner..disconnect..reconnect. The 8192eu will fail to reconnect and >> > John Heenan reported the 8723bu also fails to reconnect. Even though >> > he was directly stopping and restarting wpa_supplicant, it's the same >> > thing to the driver - connect..disconnect..reconnect. >> >> Thanks for the details - I'll have a look shortly. >> >> > With using a usb_device pointer, each message starts with the usb bus >> > address. Plug it into a different port and that address could >> > change. By using a pointer to the device associated with the wiphy >> > each message will begin with the driver name. Just makes it easier for >> > the average user to report what's in the log because he can just grep >> > for "rtl8xxxu". >> >> I see - that would be problematic for me as I quite often have 3-4 of >> these things plugged in at the same time. Not knowing which port spits >> out the message would make it a lot harder to track. In fact my primary >> devel box for this (Lenovo Yoga 13) has an rtl8723au soldered on the >> motherboard, so the moment I plug in any other dongle I'll have two. >> >> The alternative would be to add a prefer to the individual messages. >> >> Cheers, >> Jes > > I should have mentioned it also places the usb address after the driver name. If you're willing to cook up a patch, I'll have a look at it - I want to see how it behaves though before deciding whether to approve it. Cheers, Jes
