Return-path: Received: from mail-qy0-f196.google.com ([209.85.221.196]:35500 "EHLO mail-qy0-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751095AbZG3QIZ (ORCPT ); Thu, 30 Jul 2009 12:08:25 -0400 Received: by qyk34 with SMTP id 34so892564qyk.33 for ; Thu, 30 Jul 2009 09:08:25 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <200907301752.23855.mb@bu3sch.de> References: <200907301631.50780.mb@bu3sch.de> <1248968229.16129.9.camel@johannes.local> <200907301752.23855.mb@bu3sch.de> Date: Thu, 30 Jul 2009 12:08:25 -0400 Message-ID: Subject: Re: Ath5k/mac80211 triggers WARN_ON in kernel/softirq.c From: Bob Copeland To: Michael Buesch Cc: Johannes Berg , jirislaby@gmail.com, mickflemm@gmail.com, lrodriguez@atheros.com, wireless , John Linville Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Thu, Jul 30, 2009 at 11:52 AM, Michael Buesch wrote: >> John, can you revert the BH patch ("mac80211: do not use irq locks where >> not necessary") and I'll send a new one with those changes removed? > > I didn't look at the code, but can't we simply request the beacon from > mac80211 before locking? Or isn't the spin_lock_irqsave() done in the driver? > We could drop the lock around it, but it'd be nice to keep the option open here. For example, some people want to move the beacon tasklet into the hard IRQ - in that case the driver would have to jump through a lot of hoops to refresh the beacon. -- Bob Copeland %% www.bobcopeland.com