On Tue, Jan 22, 2019 at 06:32:05PM +0100, Jeroen Roovers wrote: > > then triggered a kernel panic. Sadly I couldn't capture it in time but > > I did spot that more phyN (up to phy4) devices had been added. > > Yes, even when some rt2x00usb_vendor_request starts to fail but keeps > on trying, the WLAN modules remain unrecoverable except by removing > power, so the 10 retries are usually not reached and the device is > never removed. Could it be that some operations do succeed, perhaps > because the MCU was reset and is now capable of responding to some > "vendor requests" but not others? > > Checking for num_proto_errs > 1 would then make as much sense as > num_proto_errs > 10 when running an AP. Maybe it's different for an > STA? Does it make sense to do num_proto_err > 3 check. Would that be helpful on your problem with rt2800usb device? I don't want make num_proto_err > 1 since this will remove device just after 2 errors. Also have you tested tx_power patch with lowering txpower ? Did it improve things ? Stanislaw
