Return-path: Received: from g4t0015.houston.hp.com ([15.201.24.18]:7550 "EHLO g4t0015.houston.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755191AbYIOVUN (ORCPT ); Mon, 15 Sep 2008 17:20:13 -0400 Date: Mon, 15 Sep 2008 14:17:15 -0700 To: Pavel Roskin Cc: Dan Williams , kilroyd@googlemail.com, linux-wireless@vger.kernel.org, orinoco-devel@lists.sourceforge.net, David Kilroy Subject: Re: [PATCH 01/19] orinoco: Add ESSID specific scanning for Agere fw Message-ID: <20080915211715.GA4633@bougret.hpl.hp.com> (sfid-20080915_232017_961535_B7B172A8) Reply-To: jt@hpl.hp.com References: <1217672073-7094-1-git-send-email-kilroyd@gmail.com> <1217672073-7094-2-git-send-email-kilroyd@gmail.com> <1217825304.10989.20.camel@dv> <1217864060.3139.10.camel@localhost.localdomain> <20080805162238.GA6294@bougret.hpl.hp.com> <20080902230621.GA28885@bougret.hpl.hp.com> <1220878110.2369.3.camel@dv> <20080908183203.GA7090@bougret.hpl.hp.com> <1221279475.14059.29.camel@dv> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1221279475.14059.29.camel@dv> From: Jean Tourrilhes Sender: linux-wireless-owner@vger.kernel.org List-ID: On Sat, Sep 13, 2008 at 12:17:55AM -0400, Pavel Roskin wrote: > On Mon, 2008-09-08 at 11:32 -0700, Jean Tourrilhes wrote: > > > Try the attached patch. > > It's working on x86_64! Thank you! Great, so it's not only me ;-) > The same problem exists in airo and atmel, but they don't use that > argument. Several drivers, including mac80211 use union iwreq_data, > which is fragile. Fortunately, most drivers don't use that argument or > use the "data" field, which is correct. > > The only incorrect use of union iwreq_data is in rndis_wlan. It would > ignore ESSID on x86_64. > > I think the use of union iwreq_data is dangerous and should be avoided. Note, as soon as you are using the data for real, as opposed to only the flags, you would have noticed that something was amiss. So, in essence it should be self correcting. The way I found the cause is that I wanted to dump the ESSID in the driver for debugging purpose. You can't access the ESSID through a iw_param ;-) Note that as you mentionned, the other faulty drivers are not using the arguments, and all drivers using the argument seems to have the right bits. > Regards, > Pavel Roskin Thanks, and have fun... Jean