From: Takashi Iwai Subject: Re: 2.6.17-rc4-mm3-lockdep BUG: possible deadlock detected! Date: Tue, 30 May 2006 16:14:35 +0200 Message-ID: References: <6bffcb0e0605291132u701cd69tb855cf60fa317994@mail.gmail.com> <20060529190707.GB24445@elte.hu> Mime-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from mx2.suse.de (ns2.suse.de [195.135.220.15]) by alsa.jcu.cz (ALSA's E-mail Delivery System) with ESMTP id 8BECD138 for ; Tue, 30 May 2006 16:14:36 +0200 (MEST) In-Reply-To: <20060529190707.GB24445@elte.hu> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: alsa-devel-bounces@lists.sourceforge.net Errors-To: alsa-devel-bounces@lists.sourceforge.net To: Ingo Molnar Cc: Andrew Morton , alsa-devel@alsa-project.org, Michal Piotrowski , perex@suse.cz, LKML List-Id: alsa-devel@alsa-project.org At Mon, 29 May 2006 21:07:07 +0200, Ingo Molnar wrote: > > > * Michal Piotrowski wrote: > > > I get this with Ingo's lockdep patch from > > http://people.redhat.com/mingo/generic-irq-subsystem/ > > sigh, that patchset is not released yet ... it showed up in the genirq > directory accidentally. (will release it later today) > > > ==================================== > > [ BUG: possible deadlock detected! ] > > ------------------------------------ > > at first sight this looks like a rare case of nested locking not yet > covered by the lock validator. Could you try the patch below, to > correctly express this locking construct to the lock validator? > > Btw., beyond this false positive, i dont see how the lock ordering > between ports is guaranteed - maybe there's some implicit rule that > enforces it. As mentioned in another post, different locks are used depending whether it's source or destination. Thus the confliction doesn't occur in the reverse order. > And the whole grp->list_lock and grp->list_mutex lock use > seems quite fragile - using list_lock in atomic contexts and list_mutex > in schedulable contexts? Yes, exactly. read_lock(grp->list_lock) is used in seq_clientmgr.c for the atomic contexts to follow the linked list. Takashi