From mboxrd@z Thu Jan 1 00:00:00 1970 From: chas williams - CONTRACTOR Subject: Re: scheduling while atomic: br2684ctl (atm) Date: Wed, 22 Sep 2010 14:43:29 -0400 Message-ID: <20100922144329.0cb3cf86@thirdoffive.cmf.nrl.navy.mil> References: <647401.39574.qm@web58406.mail.re3.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Karl Hiramoto , "David S. Miller" , netdev@vger.kernel.org, linux-kernel@vger.kernel.org To: Mikko Vinni Return-path: Received: from hedwig.cmf.nrl.navy.mil ([134.207.12.162]:38031 "EHLO hedwig.cmf.nrl.navy.mil" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754937Ab0IVTBN (ORCPT ); Wed, 22 Sep 2010 15:01:13 -0400 In-Reply-To: <647401.39574.qm@web58406.mail.re3.yahoo.com> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, 22 Sep 2010 00:42:29 -0700 (PDT) Mikko Vinni wrote: > I get the following message apparently every time (on 2.6.36-rc4+ at > least) when killing any old br2684ctl processes. > "unregister_atmdevice_notifier" was added by > 7313bb8f ("atm: propagate signal changes via notifier") so perhaps > the unregister > part was not tested properly for rcu interaction? it looks like atomic_notifier_chain_unregister() cannot be called in an atomic context. the only reason there is a lock is to the list of br2684_devs atomically. however, i think the event handler could be made to just register/unregister on module load/unload instead. > So far I haven't noticed any ill effects from the message. Let me > know if I should test something. Thanks. it isnt good. it should be fixed. unless someone has a better idea i will go ahead with the above.