From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mail.toke.dk ([52.28.52.200]:42901 "EHLO mail.toke.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726008AbeIEQKg (ORCPT ); Wed, 5 Sep 2018 12:10:36 -0400 From: Toke =?utf-8?Q?H=C3=B8iland-J=C3=B8rgensen?= To: Johannes Berg , linux-wireless@vger.kernel.org Cc: Felix Fietkau Subject: Re: [PATCH] mac80211: TDLS: fix skb queue/priority assignment In-Reply-To: <1536147453.3528.11.camel@sipsolutions.net> References: <20180905113402.7665-1-johannes@sipsolutions.net> <1536147453.3528.11.camel@sipsolutions.net> Date: Wed, 05 Sep 2018 13:40:43 +0200 Message-ID: <87k1o05g04.fsf@toke.dk> (sfid-20180905_134050_407059_0CDDF5C2) MIME-Version: 1.0 Content-Type: text/plain Sender: linux-wireless-owner@vger.kernel.org List-ID: Johannes Berg writes: > On Wed, 2018-09-05 at 13:34 +0200, Johannes Berg wrote: >> From: Johannes Berg >> >> If the TDLS setup happens over a connection to an AP that >> doesn't have QoS, we nevertheless assign a non-zero TID >> (skb->priority) and queue mapping, which may confuse us or >> drivers later. >> >> Fix it by just assigning the special skb->priority and then >> using ieee80211_select_queue() just like other data frames >> would go through. >> >> Signed-off-by: Johannes Berg >> --- >> OK, this addresses the case I was worried about - perhaps >> better take this than the txq patch? (This is just you talking to yourself, right? :)) > I really only found this. All the other cases aren't doing real data > frames, only null-data and management. So I guess for those we're OK, > and this was the only special case after all. SGTM. Guess we'll have to deal with everything else if we ever move management frames onto the TXQ path as well... -Toke