From owner-freebsd-wireless@FreeBSD.ORG Thu Sep 8 21:19:02 2011 Return-Path: Delivered-To: freebsd-wireless@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 82778106566C; Thu, 8 Sep 2011 21:19:02 +0000 (UTC) (envelope-from joel@vnode.se) Received: from mail.vnode.se (mail.vnode.se [62.119.52.80]) by mx1.freebsd.org (Postfix) with ESMTP id 30F4E8FC08; Thu, 8 Sep 2011 21:19:02 +0000 (UTC) Received: from mail.vnode.se (localhost [127.0.0.1]) by mail.vnode.se (Postfix) with ESMTP id 22474E3F07A; Thu, 8 Sep 2011 23:19:00 +0200 (CEST) X-Virus-Scanned: amavisd-new at vnode.se Received: from mail.vnode.se ([127.0.0.1]) by mail.vnode.se (mail.vnode.se [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id X9IXA1kSnDvm; Thu, 8 Sep 2011 23:18:57 +0200 (CEST) Received: from goofy03.vnode.local (wg.benders.se [212.247.52.4]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mail.vnode.se (Postfix) with ESMTPSA id 57279E3F079; Thu, 8 Sep 2011 23:18:57 +0200 (CEST) Date: Thu, 8 Sep 2011 23:18:55 +0200 From: Joel Dahl To: Adrian Chadd Message-ID: <20110908211855.GS52426@goofy03.vnode.local> References: <09C13664-4FC0-41F3-8849-CE875B3A6CC0@vnode.se> <20110905062453.GM52426@goofy03.vnode.local> <20110906204242.GP52426@goofy03.vnode.local> <20110907105325.GQ52426@goofy03.vnode.local> <20110907174755.GR52426@goofy03.vnode.local> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Cc: freebsd-wireless@freebsd.org, weongyo@freebsd.org, Bernhard Schmidt Subject: Re: BETA2 panic X-BeenThere: freebsd-wireless@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "Discussions of 802.11 stack, tools device driver development." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Sep 2011 21:19:02 -0000 On 08-09-2011 12:35, Adrian Chadd wrote: > Hi, > > Please try this patch. > > I bet what's happening is: > > * transition is going from RUN -> ! RUN; > * the state is changed before the swbmiss timer is cancelled; > * the callout gets called in another process context; > * bang! > > This patch: > > * changes the order of things so the callout is cancelled -before- the > state is changed; > * grabs the ic lock in the swbmiss task just to make sure nothing else > is fiddling with it. I've been using this patch for 9 hours and I haven't seen any panics so far. Do you still want me to try without the ieee80211_swbmiss locks (as discussed on irc)? -- Joel