Return-Path: MIME-Version: 1.0 In-Reply-To: <1314114448.4095.40.camel@THOR> References: <1313596884-8733-1-git-send-email-doronkeren@ti.com> <1313617329.3373.188.camel@aeonflux> <13872098A06B02418CF379A158C0F1460162C51B1A@dnce02.ent.ti.com> <1314114448.4095.40.camel@THOR> Date: Wed, 24 Aug 2011 21:27:43 +0200 Message-ID: Subject: Re: [PATCH] Bluetooth-next: Add incremental indexing in sysfs HCI connection name. From: David Herrmann To: Peter Hurley Cc: "Keren, Doron" , Marcel Holtmann , "linux-bluetooth@vger.kernel.org" , "Ilia, Kolominsky" , "Hadar, Amir" Content-Type: text/plain; charset=ISO-8859-1 List-ID: Hi Peter On Tue, Aug 23, 2011 at 5:47 PM, Peter Hurley wrote: > David, > I know you already posted a patch for this (which looks ok for fixing > the oops specifically), but it'd still be helpful to see a debug kernel > log that led up to that. There's only a couple of reasons -- all bad -- > why the hci connection could not be found in the connection list while > the ctrl & intr sockets are connected (which are tested just prior to > hidp_add_connection). Sorry for the delay. I currently have no other machine to run a debug kernel and I really don't like provoking oopses on my working machine. However, the next days I will setup a machine and try to send a full trace. Are there any useful config symbols for bluetooth debug? Despite the standard kernel-wide debug symbols. What are possible reasons why an l2cap connection can be available without an underlying ACL connection? My fix eliminates the oops but I still think that it is not fixing the real problem (anyway, I'd really like to see it upstream). I will try to send a debug log if I can reproduce it. > Regards, > Peter Hurley Cheers David