Return-path: Received: from s3.sipsolutions.net ([144.76.63.242]:34176 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933949AbeF2H6D (ORCPT ); Fri, 29 Jun 2018 03:58:03 -0400 Message-ID: <1530259080.3481.7.camel@sipsolutions.net> (sfid-20180629_095807_076424_FD033236) Subject: Re: [PATCH] mac80211: add stop/start logic for software TXQs From: Johannes Berg To: Manikanta Pubbisetty , Toke =?ISO-8859-1?Q?H=F8iland-J=F8rgensen?= Cc: linux-wireless@vger.kernel.org Date: Fri, 29 Jun 2018 09:58:00 +0200 In-Reply-To: <5adc5faf-8af0-7af9-8261-784a7ec0e4f7@codeaurora.org> References: <1529997415-20551-1-git-send-email-mpubbise@codeaurora.org> <878t71n611.fsf@toke.dk> <5adc5faf-8af0-7af9-8261-784a7ec0e4f7@codeaurora.org> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Tue, 2018-06-26 at 18:01 +0530, Manikanta Pubbisetty wrote: > > Valid point, txqs_paused is set in ieee80211_tx_dequeue; I was thinking > a case where tx_dequeue is called without rcu_read_lock but seems that > is not the case. All drivers using wake_tx_queue seems are using > rcu_read_lock before tx_dequeue; can there be a case where tx_dequeue is > called without rcu_read_lock? FWIW, the rcu_read_lock() isn't relevant at all - multiple threads can well (and will for sure!) enter it. johannes