Return-path: Received: from mtiwmhc13.worldnet.att.net ([204.127.131.117]:37726 "EHLO mtiwmhc13.worldnet.att.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758983AbYETNrE (ORCPT ); Tue, 20 May 2008 09:47:04 -0400 Message-ID: <4832D661.60902@lwfinger.net> (sfid-20080520_154711_759299_7409BB75) Date: Tue, 20 May 2008 08:47:13 -0500 From: Larry Finger MIME-Version: 1.0 To: Johannes Berg CC: Tomas Winkler , Helmut Schaa , John Linville , linux-wireless@vger.kernel.org, Andrew Morton Subject: Re: [PATCHv5] mac80211: fix NULL pointer dereference in ieee80211_compatible_rates References: <20080520095637.2cq5p5ohhc8440o4@imap.suse.de> <1ba2fa240805200554w9354d14v9abc70f676540b9b@mail.gmail.com> <1211288251.6252.86.camel@johannes.berg> <1ba2fa240805200611o7c221c86na2c627242a2ce67@mail.gmail.com> <1211289769.6252.90.camel@johannes.berg> <1ba2fa240805200633y6730a1e3ufbc0adfedd3f8243@mail.gmail.com> (sfid-20080520_153405_359637_5B9C0929) <1211290714.6252.93.camel@johannes.berg> In-Reply-To: <1211290714.6252.93.camel@johannes.berg> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: Johannes Berg wrote: >>> Thing is, I'm not totally convinced it is wrong to the code while it may >>> or may not be wrong... >> Doesn't should be bss pinned int he bss list if you are associating to >> it. If it's not there you don't have access to it's info It looks very >> wrong to me. > > Well, yes, it is a bit odd. > >>> I think this patch should go in first as it >>> actually fixes the oops, and then we can discuss the merits of adding a >>> warning there separately. Maybe after we look a bit at the code and try >>> to figure out whether it can still happen after that patch from >>> Abhijeet. >> I'm not sure if this patch is complete without this warning. What is >> in the else statement is a hack and it should be obvious. > > Considering that the message won't help us at all, why bother? We know > it's triggering, we know this might be a problem, and we know we can > only solve it by auditing the code. So why add a message that will get > us countless emails/complaints from people we cannot do anything about > anyway without doing the audit? This argument could go on endlessly; however, it is clear that we need to settle on a patch and get it upstream ASAP! Now that mainline is broken, the urgency is _MUCH_ greater. Larry