On Tue, 2023-02-21 at 10:57 +0100, Johannes Berg wrote: > Hi, > > > > > Since there's no any keep alive packet being received by host > > > > stack, leads to mac80211 destrory BA sesion. > > > > > > > > > > More specifically, the BA session relies on client side's Tx data > > > to > > > > Typo... I mean *our side*. Something like this > > Sorry. I'm just totally confused - I thought the initiator only set > the > timeout, but I see now that it's negotiated and the actual value used > is > from the client. > > Which explains basically everything. > > Yup ... after accepting the AddBA Response we activated a timer, *resetting it after each frame that we send* - sta_tx_agg_session_timer_expired(). The .net_fill_forward_path() offloads tx path to HW, so it can only rely on other way to reset as mac80211 isn't aware of that. Ryder
