Bob, there are some new data points to help narrowing down the cause of the bug. 1) Bruno's ANI patches are *not* to blame. I can reproduce the bug with or without the ANI patches applied. 2) Your patches from Bug 14342 do *not* solve the problem. I can reproduce the error with or without your patches applied. 3) I could reproduce the bug in the following usage patterns: a) iperf/TCP server running on ath5k/hostapd access point b) bi-directional iperf/UDP server running on ath5k/hostapd access point c) bi-directional iperf/UDP client running on ath5k/hostapd access point 4) I could *not* reproduce the problem in the following situations: a) uni-directional iperf/UDP server running on ath5k/hostapd access point a) uni-directional iperf/UDP client running on ath5k/hostapd access point
>From 3) and 4) I would conclude the bug is more likely to trigger when the card is both, sending *AND* receiving. Does this narrow the problem down? BTW: Why does ath5k_tasklet_calibrate call ath5k_reset_wake? It calls ieee80211_stop_queues and at the end unconditionally calls ieee80211_wake_queues. -- Regards Joerg __________________________________________________ Do You Yahoo!? Sie sind Spam leid? Yahoo! Mail verfügt über einen herausragenden Schutz gegen Massenmails. http://mail.yahoo.com _______________________________________________ ath5k-devel mailing list ath5k-devel@lists.ath5k.org https://lists.ath5k.org/mailman/listinfo/ath5k-devel
