> so far i see no regressions with 9984 with that patch > > except that 9984 has a rekeying problem at all. with wds ap -> wds sta > mode rekeying will fail and it will reauthenticate at each interval. (it > disconnects and reconnects) > but this is a long term issue qca never fixed for years. 988x doesnt > suffer from that issue Thanks for testing, sounds promising. If anyone is interested how it looks in my test environment I've uploaded two sample captures to https://www.awhome.eu/index.php/s/abxgp9pfi2ssCNy, showing how the unpatched and patched mac80211 are reacting to the rekey. The WPA Password is Induction and the AP rekeys all 30s. The AP is running lede 17.01.4, so it's way off from the current kernel/mac80211. The client is a HTC 10 phone running Lineageos. (The phone also has a WLAN card which has problems when rekeying.) There are quite many interesting things visible here, not the least one that ath9k leaks unencrypted frames for both patched and unpatched mac80211 which at least for my patched variant probably allow to calculate the TK key and encrypt all frames. I'm now experimenting now with KEY_FLAG_TAINTED, but it's not as straight forward as I expected.
