On Tue, Jul 28, 2015 at 5:47 AM, Adrian Chadd <adrian@xxxxxxxxxxx> wrote: > Those are some high values (eg from noise/spur somewhere) and the > driver is deciding to program in -110. So the radio thinks its deaf. That made me think... I have a second ath9k device, this time with an AR9300 (note, this is 3x3 instead of 2x2). I have set it up with similar settings as the AR9227 (= AP mode) and this is the result: # cat /sys/kernel/debug/ieee80211/phy0/ath9k/dump_nfcal Channel Noise Floor : -95 Chain | privNF | # Readings | NF Readings 0 -122 5 -122 -122 -121 -122 -121 1 -119 5 -119 -119 -119 -119 -119 2 -121 5 -121 -121 -120 -120 -121 The values on that AR9300 device seem to be much better. While I was at it, here is the output from a third device (TP-Link TL-WR1043ND), which uses an Atheros AR9100 MAC/BB Rev:7 AR2133 RF Rev:a2 Please note that this device is "STA" mode, whereas all other devices are in "AP" mode. # cat /sys/kernel/debug/ieee80211/phy0/ath9k/dump_nfcal Channel Noise Floor : -95 Chain | privNF | # Readings | NF Readings 0 -89 5 -89 -89 -88 -89 -89 1 -88 5 -88 -88 -88 -88 -88 2 -81 5 -81 -81 -80 -82 -82 > Hm, I think that explains a few other NICs behaving badly too.... I have found an OpenWrt ticket where users are describing similar problems, but on an TP-Link TL-WR1043ND (AR9100 + AR2133) ([0] Now here comes the interesting part: Only my device with the AR9227 shows bad performance. All other devices (including the AR9100 + AR2133 based one, which also ready "bad values") are performing fine - at least for me. Regards, Martin [0] https://dev.openwrt.org/ticket/17582 -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
