Wojciech Dubowik wrote: > I agree it's more of a workaround because there can be situations > where measured value is far from calibrated. In my setup it > helps but I haven't yet tested what happens in the noisy environment. > > Do you know a better way to fix this? I am not so deep into calibration > tasks but I have a working setup so I could spend more time on that.
Well, like I mentioned before, I am not sure if this needs to be fixed. The calibration history is cleared only once on association, anyway. If this happens when running background scans when associated, we can try to fix it. But, I have not seen such behavior, so it would be good to have logs that show abnormal rssi during scan. Sujith _______________________________________________ ath9k-devel mailing list [email protected] https://lists.ath9k.org/mailman/listinfo/ath9k-devel
