On Thu, 2022-08-25 at 15:26 -0700, James Prestwood wrote: > > > > But I guess we could leave setting > > NL80211_EXT_FEATURE_AUTH_TX_RANDOM_TA > > to the driver if we think it'd be less risky that way? > > (I assume you meant this new POWERED_ADDR_CHANGE feature) Err, yes, copy/pasted that wrong. > Anyways, is this something all mac80211 drivers _should_ be able to do > (I thought I remember you indicating this was the case)? I'd hate to > introduce (expose) a bug for some driver I couldn't test... but also > the WARN_ON would be very easy to pinpoint what happened. Yeah, true. > IMO if mac80211-based drivers should be able to handle this, I think we > should keep it in mac80211. Any issues that arise would be a driver bug > that was exposed, and needs to be fixed anyways. > Agree. And you're setting it early in alloc_hw() so drivers that (think they) have problems with it can unset it. johannes
