Re: svn commit: r306007 - head/sys/dev/iwm

2016-09-21 Thread Adrian Chadd
Heh, I don't think we have to do that right now. My laptop is the same; I just got lucky that antenna A is fine but antenna B isn't. The firmware and NVM says both A and B are available, and I /physically installed this NIC/ into the laptop so I know that bit is right. My guess it it's something t

Re: svn commit: r306007 - head/sys/dev/iwm

2016-09-21 Thread Larry Rosenman
On 2016-09-21 20:36, Adrian Chadd wrote: larry and I did some private debugging. The long and short of it is that his setup doesn't receive frames on antenna A, only on antenna B. Antenna A has like 50dB or more less signal: iwm0: energy In A -114 B -51 C -256 , and max -51 So my guess is it's

Re: svn commit: r306007 - head/sys/dev/iwm

2016-09-21 Thread Adrian Chadd
larry and I did some private debugging. The long and short of it is that his setup doesn't receive frames on antenna A, only on antenna B. Antenna A has like 50dB or more less signal: iwm0: energy In A -114 B -51 C -256 , and max -51 So my guess is it's either not connected internally, or (like

Re: svn commit: r306007 - head/sys/dev/iwm

2016-09-21 Thread Adrian Chadd
Hi, Please update to the latest code, then try this diff: https://reviews.freebsd.org/D7994 It fixes up the initial frame exchange sequence for 802.1x negotiation to use the management rate, rather than the rate control rate. I'm still not convinced that the iwm antenna setup / selection is rig