Return-path: Received: from mail-wi0-f174.google.com ([209.85.212.174]:46664 "EHLO mail-wi0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751428Ab2AHGBQ convert rfc822-to-8bit (ORCPT ); Sun, 8 Jan 2012 01:01:16 -0500 Received: by wibhm6 with SMTP id hm6so1897850wib.19 for ; Sat, 07 Jan 2012 22:01:15 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <20120107214724.GA7746@tuxdriver.com> References: <201201071127.06717.toralf.foerster@gmx.de> <20120107214724.GA7746@tuxdriver.com> Date: Sun, 8 Jan 2012 08:01:15 +0200 Message-ID: (sfid-20120108_070128_968829_6C74A463) Subject: Re: suggestion for a reasonable value for /sys/module/iwlwifi/parameters/debug From: Emmanuel Grumbach To: "John W. Linville" Cc: =?ISO-8859-1?Q?Toralf_F=F6rster?= , linux-wireless@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Sat, Jan 7, 2012 at 23:47, John W. Linville wrote: > On Sat, Jan 07, 2012 at 10:55:30PM +0200, Emmanuel Grumbach wrote: >> > If I choosed 0x43ff my syslog message files exploded - and I'd like to trouble >> > shoot a problem with my wlan at a ThinkPad T400 (iwlwifi, 5100 AGN controler). >> > The symptoms are that 1-2 times in a hour the network stops working without >> > any message in /var/log/messages (kernel 3.0.x, 3.1.x, 3.2.0 tested). >> > >> > Thanks for any advice >> > -- >> >> For this kind of issues I added the TX_QUEUES debug flags, but I just >> checked and it is not in 3.2. It is in John's wireless tree though. It >> is bit 31: debug=0x80000000 >> >> BTW - John, I have tried many times to understand all this, but stupid >> me, couldn't ever make it. How come a patch is in wireless which in my >> understanding (probably wrong) is supposed to feed the current rc >> cycle (meaning the 3.2-rcX which just ended) and not be in 3.2 stable >> release ? Sorry for the ignorance... > > We are betwen the release and the next -rc1. ?Right now everything > in wireless is the stuff that was in wireless-next last week. > > Hth! > yup, it does. Thanks Toralf, back to you. There are quite a few fixes that will be included in the next stable release (3.3) in about 3 months. I really believe these fixes can help you. You can also try to disable 11n to see if it helps. To do so, add 11n_disable as a module parameter.