From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mx1.redhat.com ([209.132.183.28]:42462 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752985Ab3JYKIK (ORCPT ); Fri, 25 Oct 2013 06:08:10 -0400 Date: Fri, 25 Oct 2013 12:04:40 +0200 From: Stanislaw Gruszka To: Dan Williams Cc: Dietmar Rudolph , linux-wireless@vger.kernel.org Subject: Re: iwl3945 filling kern.log when HW switch is off Message-ID: <20131025100436.GA16384@redhat.com> (sfid-20131025_120815_227787_7E9CE8D6) References: <526795E1.90701@crlf.de> <20131024112750.GC1677@redhat.com> <1382626081.4328.3.camel@dcbw.foobar.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1382626081.4328.3.camel@dcbw.foobar.com> Sender: linux-wireless-owner@vger.kernel.org List-ID: On Thu, Oct 24, 2013 at 09:48:01AM -0500, Dan Williams wrote: > On Thu, 2013-10-24 at 13:27 +0200, Stanislaw Gruszka wrote: > > On Wed, Oct 23, 2013 at 11:24:49AM +0200, Dietmar Rudolph wrote: > > > Hello, I hope this is the correct place to put a bug report in. I do > > > not subscribe to this mailing list, but feel free to send PM in case > > > you need additional info. You shoudn't :-) > > > > > > After installing Ubuntu 13.10, I noticed kern.log growing faster > > > than you could look at. Within a few working hours, kern.log was > > > >6GB! > > > > > > The reason is that I intentionally disabled WiFi on this laptop by > > > turning the HW switch off. However, this makes iwl3945 write a log > > > entry into kern.log every few microseconds! > > > > > > The workaround is to turn the HW switch on. The bug fix would be to > > > write just a single line to kern.log. > > > > This message is printed when someone try to UP device when RF kill > > switch is on. Either you did not compile kernel with CONFIG_RFKILL or > > your user space network management software does not handle rfkill . > > Though maybe it shouldn't actually print a dmesg error when that > happens? Lots of software (scripts included) might try to open the > device but not really know much about rfkill... Is there a way to print > it once-per-rfkill instead maybe? I'll just remove this message. We can return ERFKILL error instead. That will allow user to identify problem without message. Thanks Stanislaw