Return-path: Received: from s3.sipsolutions.net ([5.9.151.49]:48619 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932555AbcA0Odv (ORCPT ); Wed, 27 Jan 2016 09:33:51 -0500 Message-ID: <1453905229.2351.8.camel@sipsolutions.net> (sfid-20160127_153356_502496_03234AF8) Subject: Re: [RFC] mac80211: introduce netdev queue control for txqs From: Johannes Berg To: Michal Kazior , linux-wireless@vger.kernel.org Date: Wed, 27 Jan 2016 15:33:49 +0100 In-Reply-To: <1453904917-25458-1-git-send-email-michal.kazior@tieto.com> (sfid-20160127_152732_241232_E83365C7) References: <1453904917-25458-1-git-send-email-michal.kazior@tieto.com> (sfid-20160127_152732_241232_E83365C7) Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Wed, 2016-01-27 at 15:28 +0100, Michal Kazior wrote: > Until now all stations (notably in AP mode) would > share 4 AC queues exposed via netdev. This meant > that whenever a single station's AC became full > all other stations' AC were considered full and > stopped as well. > > This was suboptimal and could lead to per station > traffic starvation and/or suboptimal Tx > aggregation performance. I really think you need to Cc netdev on this :) johannes