Barry Day <brise...@gmail.com> writes: > On Sat, Nov 19, 2016 at 06:53:42PM -0500, Jes Sorensen wrote: >> Barry Day <brise...@gmail.com> writes: >> > On Fri, Nov 18, 2016 at 09:00:10PM -0500, Jes Sorensen wrote: >> >> Barry Day <brise...@gmail.com> writes: >> >> > On Fri, Nov 18, 2016 at 04:44:21PM -0500, jes.soren...@redhat.com wrote: >> >> >> From: Jes Sorensen <jes.soren...@redhat.com> >> >> >> >> >> >> Kalle, >> >> >> >> >> >> Please find attached a number of patches for the rtl8xxxu >> >> >> driver. >> >> >> >> >> >> The issues reported with wpa_supplicant on 8723bu still needs further >> >> >> investigation. >> >> >> >> >> > >> >> > The patch I posted that you want tested more will also fix the >> >> > wpa_supplicant issue. Currently I'm looking at why the tx rate is not >> >> > what it should be. I feel fixing that first will be beneficial for >> >> > fixing any other issues. >> >> >> >> Interesting, I was thinking that might be the case. I do want to dig >> >> into this further to understand it better. If we use your solution I >> >> will want to make sure we cover both gen1 and gen2 parts. >> >> >> >> > The recent merge has made my local branch of rtl8xxxu-devel 14 >> >> > commits ahead. >> >> > Do I need to do a reset and submit a new patch for the DWA-131 dongle? >> >> >> >> In general you need to use 'git pull --rebase' on my tree. I rebase it >> >> to stay in sync with Kalle's tree. >> >> >> >> The DWA-131 is the 8192eu? Sorry a bit behind and my mind is losing >> >> bits. If it's the patch you posted earlier I can dig it out and play >> >> with it - I am still catching up though, so please be patient. >> > >> > yes it's an 8192eu. >> >> Gotcha - how do you do your testing to reproduce the problem btw? Most >> of my testing is using the 8723au as a primary device and the next >> device as a follow-on, so I may not see as long a run with the device >> active as you see. >> > > 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
