Return-Path: Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Mac OS X Mail 7.1 \(1827\)) Subject: Re: [PATCH v3 0/4] Regression fixes for rfcomm/tty.c From: Marcel Holtmann In-Reply-To: <20140128083154.GA29060@sottospazio.it> Date: Tue, 28 Jan 2014 04:08:58 -0800 Cc: Alexander Holler , "Gustavo F. Padovan" , peter@hurleysoftware.com, "linux-bluetooth@vger.kernel.org development" , Greg KH , jslaby@suse.cz, stable@vger.kernel.org Message-Id: References: <1389039834-24491-1-git-send-email-gianluca@sottospazio.it> <53C8DDF7-94EA-4E01-A2A9-C366F9F85530@holtmann.org> <52DCDF84.1040902@ahsoftware.de> <20140128083154.GA29060@sottospazio.it> To: Gianluca Anzolin Sender: linux-bluetooth-owner@vger.kernel.org List-ID: Hi Gianluca, >>>> all 4 patches have been applied to bluetooth-next tree. >>> >>> Maybe a bit late, but I've just seen they miss a Cc: stable@vger.kernel.org to automatically end up in 3.12 and 3.13 too. >> >> we can always promote them to stable. On Purpose I wanted them to cycle through bluetooth-next for a while to make sure they do not cause any other regressions. >> > > Unfortunately it seems I overlooked the fact that rfcomm_dev_activate() is > called with the port->mutex held. So patches 2/3/4 cause a regression I missed > because I didn't turn on the appropriate debug options (circular locking > dependency, a bug report already appeared on this list). > > I'm afraid this all stems from my partial knowledge of the tty_port code and > unfortunately I don't know how to solve the problem right now. > > I think it's better to revert those patches for the moment. can we get this fixed instead. If I revert them, I have a different issue. So I am trading one bug for another one. Regards Marcel