Re: Bisected 3.9 regression for iwl4965 connection problem to1672c0e3

From: Johannes Berg
Date: Fri May 24 2013 - 16:28:52 EST


On Wed, 2013-05-22 at 13:59 +0200, Stanislaw Gruszka wrote:

> > AFICT, we wake queues only if beacon arrives or mac80211 call drv_config
> > with BSS_CHANGED_IDLE. I'm not sure if the latter prevent stuck.
>
> It should prevent stuck. When we fail to auth, drv_config() with BSS_CHANGED_IDLE
> is called via:
>
> ieee80211_destroy_auth_data ->
> ieee80211_vif_release_channel ->
> __ieee80211_vif_release_channel ->
> ieee80211_unassign_vif_chanctx ->
> ieee80211_bss_info_change_notify
>
> But there is need to have ->vif.chanctx_conf valid in
> __ieee80211_vif_release_channel(), where is below condition:
>
> conf = rcu_dereference_protected(sdata->vif.chanctx_conf,
> lockdep_is_held(&local->chanctx_mtx));
> if (!conf)
> return;
>
> I'm not sure if that always happen. Perhaps would be better to change
> BSS_CHANGED_IDLE to BSS_CHANGED_BSSID, which is called directly from
> ieee80211_destroy_auth_data() ?

I don't think the "!conf" can hit in this case, since to even try to
associate you have to have a channel context assigned.

johannes

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/